Basic Cell Culture Protocols

METHODS IN MOLECULAR BIOLOGY™
Series Editor
John M. Walker
School of Life Sciences
University of Hertfordshire
Hatfield, Hertfordshire, AL10 9AB, UK
For further volumes:

http://www.springer.com/series/7651
Basic Cell Culture Protocols
Fourth Edition
Edited by
Cheryl D. Helgason
Experimental Therapeutics, BC Cancer Agency, Vancouver, BC, Canada
Cindy L. Miller
STEMCELL Technologies Inc., Vancouver, BC, Canada
Editors
Cheryl D. Helgason Cindy L. Miller
Experimental Therapeutics STEMCELL Technologies Inc.,
BC Cancer Agency Vancouver, BC
Vancouver, BC Canada
Canada
ISSN 1064-3745 ISSN 1940-6029 (electronic)

ISBN 978-1-62703-127-1 ISBN 978-1-62703-128-8 (eBook)
DOI 10.1007/978-1-62703-128-8
Springer New York Heidelberg Dordrecht London
Library of Congress Control Number: 2012950590
© Springer Science+Business Media, LLC 2013

This work is subject to copyright. All rights are reserved by the Publisher, whether the whole or part of the material is
concerned, specifically the rights of translation, reprinting, reuse of illustrations, recitation, broadcasting, reproduction
on microfilms or in any other physical way, and transmission or information storage and retrieval, electronic adaptation,
computer software, or by similar or dissimilar methodology now known or hereafter developed. Exempted from this
legal reservation are brief excerpts in connection with reviews or scholarly analysis or material supplied specifically for
the purpose of being entered and executed on a computer system, for exclusive use by the purchaser of the work.
Duplication of this publication or parts thereof is permitted only under the provisions of the Copyright Law of the
Publisher’s location, in its current version, and permission for use must always be obtained from Springer. Permissions
for use may be obtained through RightsLink at the Copyright Clearance Center. Violations are liable to prosecution
under the respective Copyright Law.
The use of general descriptive names, registered names, trademarks, service marks, etc. in this publication does not
imply, even in the absence of a specific statement, that such names are exempt from the relevant protective laws and
regulations and therefore free for general use.
While the advice and information in this book are believed to be true and accurate at the date of publication, neither
the authors nor the editors nor the publisher can accept any legal responsibility for any errors or omissions that may be
made. The publisher makes no warranty, express or implied, with respect to the material contained herein.
Printed on acid-free paper
Humana Press is a brand of Springer

Springer is part of Springer Science+Business Media (www.springer.com)
Preface
Tissue culture techniques were first developed at the beginning of the twentieth century
and have undergone dramatic changes and improvements since that time. They are invalu-
able tools for the exploration of numerous biological questions related both to cellular
processes and to the signaling mechanisms that regulate them. At some point in their
careers, virtually every scientist and technician, as well as many medical professionals,
regardless of their area of specialization has a need to utilize cell culture systems.
Our objective in preparing this book was to provide the novice cell culturist with
sufficient information to perform the basic techniques, to ensure the health and identity of
their cell lines, and to be able to isolate and culture specialized primary cell types. It was our
intent to generate a valuable resource book containing clear methodologies pertinent to
current areas of investigation, rather than attempting to educate cell culturists on specific
cell types or organ systems. We have thus included updates on several of the chapters from
the previous edition but have also added a significant number of new chapters to this
volume.
It is anticipated that many readers will have a solid background in the fundamentals of
anatomy, histology, and biochemistry but little or no experience in cell culture. We antici-
pate that this book will be a useful resource for technicians, graduate students, and postdoc-
toral fellows, as well as for the research leaders (both basic scientists and clinicians) and cell
culture experts moving toward the use of new model systems.
The chapters that follow provide step-by-step instructions for the isolation and growth
of various primary cell types. In addition, they illustrate the techniques required for defining
the properties of various types of cells as well as for cell differentiation.
We wish to extend our sincerest appreciation to all the contributors who willingly took
the time to share their expertise and knowledge and to the many individuals who assisted
us in the preparation of this book. Special thanks go to Dr. Carrie Peters and Jessica Ata
who assisted with the editing process and were instrumental in keeping everything orga-
nized and on track.
Vancouver, BC, Canada Cheryl D. Helgason

Vancouver, BC, Canada Cindy L. Miller
v
Contents
Preface. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . v
Contributors. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . xi
1 Detection of Mycoplasma Contaminations . . . . . . . . . . . . . . . . . . . . . . . . . . . 1

Cord C. Uphoff and Hans G. Drexler
2 Eradication of Mycoplasma Contaminations . . . . . . . . . . . . . . . . . . . . . . . . . . 15
3 STR DNA Typing of Human Cell Lines: Detection of Intra-
and Interspecies Cross-Contamination . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27
Wilhelm G. Dirks and Hans G. Drexler
4 Classical and Molecular Cytogenetic Analysis. . . . . . . . . . . . . . . . . . . . . . . . . . 39
Roderick A.F. MacLeod and Hans G. Drexler
5 Fluorescent In Situ Hybridization of DNA Probes in the Interphase
and Metaphase Stages of the Cell Cycle. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 61
Linda A. Cannizzaro
6 The Development of T Lymphocytes in Fetal Thymus Organ Culture . . . . . . . 85
Takeshi Nitta, Izumi Ohigashi, and Yousuke Takahama
7 Generation, Isolation, and Engraftment of In Vitro-Derived
Human T Cell Progenitors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 103
Génève Awong and Juan Carlos Zúñiga-Pflücker
8 In Vitro Generation of Human T Regulatory Cells: Generation, Culture,
and Analysis of FOXP3-Transduced T Cells. . . . . . . . . . . . . . . . . . . . . . . . . . . 115
Alicia N. McMurchy and Megan K. Levings
9 Simultaneous Cloning and Selection of Hybridomas and Transfected Cell
Lines in Semisolid Media. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 133
Bert Wognum and Tracy Lee
10 Isolation and Characterization of Mouse Side Population Cells . . . . . . . . . . . . 151
Aysegul V. Ergen, Mira Jeong, Kuanyin K. Lin, Grant A. Challen,
and Margaret A. Goodell
11 Stem Cell Identification by DyeCycle Violet Side Population Analysis . . . . . . . 163
William G. Telford
12 Isolation and Characterization of Cancer Stem Cells In Vitro . . . . . . . . . . . . . 181
Craig Gedye and Laurie Ailles
13 Ex Vivo Differentiation of Cord Blood Stem Cells into Megakaryocytes
and Platelets . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 205
Nicolas Pineault, Amélie Robert, Valérie Cortin, and Lucie Boyer
vii
viii Contents
14 Generation and Characterization of Murine Alternatively

Activated Macrophages . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 225
Shelley B. Weisser, Keith W. McLarren, Etsushi Kuroda, and Laura M. Sly
15 Human Long-Term Culture Initiating Cell Assay . . . . . . . . . . . . . . . . . . . . . . 241
Min Liu, Cindy L. Miller, and Connie J. Eaves
16 Long-Term Culture-Initiating Cell Assay for Mouse Cells . . . . . . . . . . . . . . . . 257
Stefan Woehrer, Cindy L. Miller, and Connie J. Eaves
17 Colony Forming Cell Assays for Human Hematopoietic Progenitor Cells . . . . 267
Bert Wognum, Ning Yuan, Becky Lai, and Cindy L. Miller
18 Studying Leukocyte Recruitment Under Flow Conditions. . . . . . . . . . . . . . . . 285
Sean A. Parsons, Christophe Jurzinsky, Susan L. Cuvelier,
and Kamala D. Patel
19 Generation and Establishment of Murine Adherent Cell Lines. . . . . . . . . . . . . 301
Rouzanna Istvanffy and Robert A.J. Oostendorp
20 Isolation, Enumeration, and Expansion of Human Mesenchymal
Stem Cells in Culture . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 315
Ravenska Wagey and Brenton Short
21 Isolation and Culture of Mesenchymal Stem Cells from Mouse
Compact Bone . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 335
Brenton Short and Ravenska Wagey
22 Generation of a Pool of Human Platelet Lysate and Efficient
Use in Cell Culture . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 349
Katharina Schallmoser and Dirk Strunk
23 In Vitro Methods to Culture Primary Human Breast Epithelial Cells. . . . . . . . 363
Afshin Raouf and Yu Jia Sun
24 Human Prostate Epithelial Cell Cultures . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 383
Johng S. Rhim
25 Enzymatic Dissociation, Flow Cytometric Analysis, and Culture
of Normal Mouse Mammary Tissue . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 395
Michael Prater, Mona Shehata, Christine J. Watson, and John Stingl
26 Isolation and Characterization of Human Hair Follicle Epithelial Cells . . . . . . 411
Keita Inoue and Kotaro Yoshimura
27 Cocultivation of Human Oral Keratinocytes and Human
Osteoblast-Like Cells . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 423
Ricarda Glaum and Margit Wiedmann-Al-Ahmad
28 Isolation and Culture of Skeletal Muscle Myofibers as a Means to
Analyze Satellite Cells . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 431
Paul Keire, Andrew Shearer, Gabi Shefer, and Zipora Yablonka-Reuveni
29 Hepatic Differentiation of Embryonic Stem Cells by Murine Fetal Liver
Mesenchymal Cells . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 469
Takamichi Ishii, Kentaro Yasuchika, and Iwao Ikai
30 Methods to Culture, Differentiate, and Characterize Neural Stem Cells
from the Adult and Embryonic Mouse Central Nervous System . . . . . . . . . . . 479
Sharon A. Louis, Carmen K.H. Mak, and Brent A. Reynolds
Contents ix
31 Feeder-Independent Culture Systems for Human Pluripotent Stem Cells. . . . . 507

Jennifer Moody
32 Formation of Embryoid Bodies from Human Pluripotent Stem Cells
Using AggreWell™ Plates . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 523
Jennifer Antonchuk
33 Techniques in Embryoid Body Formation from Human Pluripotent
Stem Cells. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 535
Nirupama K. Shevde and Amber A. Mael
Index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 547
Contributors
LAURIE AILLES • Department of Medical Biophysics, Ontario Cancer Institute,

University of Toronto, Toronto, ON, Canada
JENNIFER ANTONCHUK • STEMCELL Technologies Inc, Vancouver, BC, Canada
GÉNÈVE AWONG • Department of Immunology, University of Toronto, Toronto,
ON, Canada
LUCIE BOYER • Département de Recherche et Développement, Héma-Québec,
Québec City, QC, Canada
LINDA A. CANNIZZARO • Department of Pathology, Montefiore Medical Center,
Albert Einstein College of Medicine, Bronx, NY, USA
GRANT A. CHALLEN • The Center for Cell and Gene Therapy, Baylor College of Medicine,
Houston, TX, USA
VALÉRIE CORTIN • Département de Recherche et Développement, Héma-Québec,
SUSAN L. CUVELIER • Calvin, Phoebe and Joan Snyder Institute for Infection,
Immunity and Inflammation, University of Calgary, Calgary, AB, Canada
WILHELM G. DIRKS • Department of Human and Animal Cell Lines, DSMZ, German
Collection of Microorganisms and Cell Cultures and German Biological Resource
Center, Braunschweig, Germany
HANS G. DREXLER • Department of Human and Animal Cell Lines, DSMZ, German
Collection of Microorganisms and Cell Cultures and German Biological Resource
Center, Braunschweig, Germany
CONNIE J. EAVES • Terry Fox Laboratory, British Columbia Cancer Agency,
Vancouver, BC, Canada
AYSEGUL V. ERGEN • The Center for Cell and Gene Therapy, Baylor College of Medicine,
Houston, TX, USA
CRAIG GEDYE • Ontario Cancer Institute, Toronto, ON, Canada
RICARDA GLAUM • Department of Oral and Maxillofacial Surgery,
Albert Ludwigs University, Freiburg, Germany
MARGARET A. GOODELL • Stem Cells and Regenerative Medicine Center, Baylor College
of Medicine, Houston, TX, USA
IWAO IKAI • Department of Surgery, Graduate School of Medicine, Kyoto University;
Department of Surgery, Kyoto Medical Center, National Hospital Organization,
Kyoto, Japan
KEITA INOUE • Department of Plastic and Reconstructive Surgery, Shizuoka Cancer
Center, Shizuoka, Japan
TAKAMICHI ISHII • Department of Surgery, Graduate School of Medicine,
Kyoto University, Kyoto, Japan
ROUZANNA ISTVANFFY • The Stem Cell Physiology Laboratory, Medizinische Klinik,
Technische Universität München, Munich, Germany
xi
xii Contributors
MIRA JEONG • The Center for Cell and Gene Therapy, Baylor College of Medicine,
Houston, TX, USA; University of Science and Technology, Daejeon, South Korea
CHRISTOPHE JURZINSKY • Calvin, Phoebe and Joan Snyder Institute for Infection,
PAUL KEIRE • Department of Biological Structure, School of Medicine,
University of Washington, Seattle, WA, USA
ETSUSHI KURODA • Division of Gastroenterology, Department of Pediatrics,
BC Children’s Hospital and The Child and Family Research Institute,
University of British Columbia, Vancouver, BC, Canada
BECKY LAI • STEMCELL Technologies, Vancouver, BC, Canada
TRACY LEE • STEMCELL Technologies, Vancouver, BC, Canada
MEGAN K. LEVINGS • Department of Surgery, University of British Columbia,
KUANYIN K. LIN • The Center for Cell and Gene Therapy, Baylor College of Medicine,
Houston, TX, USA
MIN LIU • Terry Fox Laboratory, Vancouver, BC, Canada; Private Chinese Culture
University, Taipei, Taiwan
SHARON A. LOUIS • STEMCELL Technologies, Vancouver, BC, Canada
RODERICK A.F. MACLEOD • Department of Human and Animal Cell Cultures,
DSMZ-German Collection of Microorganisms and Cell Cultures, Braunschweig,
Germany
AMBER A. MAEL • WiCell Research Institute, Madison, WI, USA
CARMEN K.H. MAK • STEMCELL Technologies, Vancouver, BC, Canada
KEITH W. MCLARREN • Division of Gastroenterology, Department of Pediatrics,
BC Children’s Hospital and The Child and Family Research Institute,
University of British Columbia, Vancouver, BC, Canada
ALICIA N. MCMURCHY • Department of Surgery, University of British Columbia,
CINDY L. MILLER • STEMCELL Technologies, Vancouver, BC, Canada
JENNIFER MOODY • STEMCELL Technologies, Vancouver, BC, Canada
TAKESHI NITTA • Division of Experimental Immunology, Institute for Genome Research,
University of Tokushima, Tokushima, Japan
IZUMI OHIGASHI • Division of Experimental Immunology, Institute for Genome Research,
University of Tokushima, Tokushima, Japan
ROBERT A.J. OOSTENDORP • The Stem Cell Physiology Laboratory, Medizinische Klinik,
Technische Universität München, Munich, Germany
SEAN A. PARSONS • Calvin, Phoebe and Joan Snyder Institute for Infection,
KAMALA D. PATEL • Calvin, Phoebe and Joan Snyder Institute for Infection,
NICOLAS PINEAULT • Département de Recherche et Développement, Héma-Québec,
Université Laval, Québec City, QC, Canada
MICHAEL PRATER • Cancer Research, UK Cambridge Research Institute, Cambridge,
UK
Contributors xiii
AFSHIN RAOUF • Department of Immunology, Faculty of Medicine, University of

Manitoba and Manitoba Institute of Cell Biology, CancerCare Manitoba,
Winnipeg, MB, Canada
BRENT A. REYNOLDS • Department of Neurosurgery, University of Florida, Gainesville,
FL, USA
JOHNG S. RHIM • Department of Defense, Center for Prostate Disease Research,
Rockville, MD, USA
AMÉLIE ROBERT • Département de Recherche et Développement, Héma-Québec,
KATHARINA SCHALLMOSER • Stem Cell Research Unit Graz and Clinic for Blood Group
Serology and Transfusion Medicine, Medical University of Graz, Graz, Austria
ANDREW SHEARER • Department of Biological Structure, School of Medicine,
GABI SHEFER • Department of Biological Structure, Faculty of Medicine,
MONA SHEHATA • Cancer Research UK, Cambridge Research Institute, Cambridge,
UK
NIRUPAMA K. SHEVDE • WiCell Research Institute, Madison, WI, USA
BRENTON SHORT • STEMCELL Technologies, Vancouver, BC, Canada
LAURA M. SLY • Division of Gastroenterology, Department of Pediatrics, BC Children’s
Hospital and The Child and Family Research Institute, University of British Columbia,
JOHN STINGL • Cancer Research UK, Cambridge Research Institute, Cambridge, UK
DIRK STRUNK • Stem Cell Research Unit and Department of Hematology,
Medical University of Graz, Graz, Austria
YU JIA SUN • Department of Immunology, Faculty of Medicine, University of Manitoba
and Manitoba Institute of Cell Biology, CancerCare Manitoba, Winnipeg, MB,
Canada
YOUSUKE TAKAHAMA • Division of Experimental Immunology, Institute for Genome
Research, The University of Tokushima, Tokushima, Japan
WILLIAM G. TELFORD • Experimental Transplantation and Immunology Branch,
National Institutes of Health, Bethesda, MD, USA
CORD C. UPHOFF • Department of Human and Animal Cell Lines, DSMZ,
German Collection of Microorganisms and Cell Cultures, Braunschweig, Germany
RAVENSKA WAGEY • STEMCELL Technologies, VancouverBC, Canada
CHRISTINE J. WATSON • Department of Pathology, University of Cambridge,
Cambridge, UK
SHELLEY B. WEISSER • Division of Gastroenterology, Department of Pediatrics,
BC Children’s Hospital and The Child and Family Research Institute, University of
British Columbia, Vancouver, BC, Canada
MARGIT WIEDMANN-AL-AHMAD • Department of Oral and Maxillofacial Surgery,
Albert Ludwigs University, Freiburg, Germany
BERT WOGNUM • STEMCELL Technologies, Vancouver, BC, Canada
STEFAN WOEHRER • Terry Fox Laboratory, BC Cancer Agency, Vancouver, BC, Canada;
Medical University of Vienna, Vienna, Austria
xiv Contributors
ZIPORA YABLONKA-REUVENI • Department of Biological Structure, University of

Washington, Seattle, WA, USA
KENTARO YASUCHIKA • Department of Surgery, Graduate School of Medicine,
Kyoto University, Kyoto, Japan
KOTARO YOSHIMURA • Department of Plastic Surgery, University of Tokyo, Tokyo, Japan
NING YUAN • STEMCELL Technologies, Vancouver, BC, Canada
JUAN CARLOS ZUÑIGA-PFLÜCKER • Department of Immunology, University of Toronto,
Toronto, ON, Canada
Chapter 1
Detection of Mycoplasma Contaminations

Abstract
Mycoplasma contamination of cell lines is one of the major problems in cell culture technology. The
specific, sensitive, and reliable detection of mycoplasma contamination is an important part of mycoplasma
control and should be an established method in every cell culture laboratory. New cell lines as well as cell
lines in continuous culture must be tested in regular intervals. The polymerase chain reaction (PCR) meth-
odology offers a fast and sensitive technique to monitor all cultures in a laboratory and can also be used to
determine the contaminating species.
The described assay can be performed in less than 3 hours, including sample preparation, DNA extrac-
tion, PCR run, and analysis of the PCR products. Special emphasis is given to steps taken to avoid false-
negative results due to the presence of inhibitors of the Taq polymerase in the crude samples and the
interpretation of the results. The technique can also be adapted to the requirements of the pharmacopoeia.
Key words: Bacteria, Cell lines, Contamination, Mycoplasma, PCR
1. Introduction
1.1. Mycoplasma Acute contaminations of cell lines are frequently observed in

Contaminations routine cell culture and can often be attributed to improper
of Cell Lines handling of the growing culture. These contaminations can usually be
detected by the turbidity evolving after a short incubation time or
by routine observation of the cell culture under the inverted
microscope. In addition to these obvious contaminations, other
hidden infections can occur consisting of mycoplasmas, viruses, or
cross-contaminations with other cell lines. Although known for
many years and despite the multitude of publications dealing with
mycoplasma infections of cell cultures, a high proportion of scien-
tists are not aware of the potential contamination of cell cultures
with mycoplasmas. As seen in our cell repository, more than 20%
of the incoming cell lines are infected with mycoplasmas, and in
Cheryl D. Helgason and Cindy L. Miller (eds.), Basic Cell Culture Protocols, Methods in Molecular Biology, vol. 946,
DOI 10.1007/978-1-62703-128-8_1, © Springer Science+Business Media, LLC 2013
1
2 C.C. Uphoff and H.G. Drexler
most cases, the depositor was not aware of this. Whereas in the
early years of cell culture, bovine serum was one of the major
sources of infections, nowadays mycoplasmas seem to be mainly
transferred from one infected culture to another by using labora-
tory equipment, media, or reagents that came into contact with
infected cultures. This culture hopping is concordant with the
occurrence of cell cross-contaminations with a proved incidence
of 15% plus an estimated number of unknown cases (see STR
DNA Typing of Human Cell Lines: Detection of Intra- and
Interspecies Cross-contamination) and is substantiated by the
finding that either all or none of the cell cultures of a given labora-
tory are infected with the same mycoplasma strain. Thus, methods
for the detection, elimination (see Chapter 2), and prevention (1)
of mycoplasma contaminations should belong to the basic
panel of cell culture techniques applied.
The term “Mycoplasma” is commonly used as synonym for the
class of Mollicutes that represents a large group of highly special-
ized bacteria and which are all characterized by their lack of a rigid
cell wall. Mycoplasma is the largest genus within this class. Due to
their small size and flexibility, some of these bacteria are able to
pass through conventional microbiological filters (0.2 mm). Their
reduced metabolic abilities cause a relatively long generation time
which is often in the range of that of cell lines, and they do nor-
mally not overgrow or kill the eukaryotic cells. The extent of infec-
tion is highly diverse and depends on the mycoplasma species, cell
type, and culture conditions. Their influence on the biological
characteristics of the eukaryotic cells is manifold and almost every
experimental or production setting can be influenced.
The identification of infecting mycoplasmas shows that only a
limited number of about seven Mycoplasma and Acholeplasma spe-
cies from human, swine, and bovine natural hosts occur predomi-
nantly in cell cultures, and no species specificity can be observed.
Additionally, a few mycoplasma species were shown to enter the
eukaryotic cells actively and to exist intracytoplasmically (2).
M. fermentans is one of those penetrating mycoplasma species.
Hence, sensitive methods need to be established and frequently
employed in every cell culture laboratory to detect mycoplasma
contaminations.
1.2. Mycoplasma The biological diversity of mycoplasmas and their close adaptation
Detection to the cell cultures render it very difficult to detect all contamina-
tions in one general assay. A wide spectrum of approaches have
been proposed to detect mycoplasmas in cell cultures. Most of
these methods are lengthy, complex, and not applicable in routine
cell culture (e.g., electron microscopy, biochemical and radioactive
incorporation assays, etc.), or are restricted to specific groups of
mycoplasmas. Molecular biological methods were the first to be
able to detect all the different mycoplasma types in cell cultures,
1 Detection of Mycoplasma Contaminations 3
regardless of their biological properties, with relatively low effort

in terms of time and labor (3).
Polymerase chain reaction (PCR) provides a very sensitive and
specific option for the direct detection of mycoplasmas in cell cul-
tures. PCR combines many of the features previously offered by
different assays: sensitivity, specificity, low expenditure of labor,
time, and cost, simplicity, objectivity of interpretation, reproduc-
ibility, and documentation of the results. On the other hand, a
number of indispensable control reactions have to be included in
the PCR assay to avoid false-negative or false-positive results.
A comparison of the PCR method with other well-established
assays (DNA/RNA hybridization, microbiological culture) showed
that the PCR assay is a very robust, efficient, and reliable method
for the detection of mycoplasmas (4). Another advantage is the
possibility to adapt the method to the requirements of the pharma-
copoeia, which demand a detection limit of 10 cfu/mL for each of
the seven different mycoplasma species (5). Each laboratory must
individually validate the assay.
The choice of the primer sequences is one of the most crucial
decisions. Several primer sequences are published for single or
nested PCR (see Note 1) and with narrow or broad specificity for
mycoplasma or eubacteria species. In most cases, the 16S rDNA
sequences are used as target sequences, because this gene contains
regions with more or less conserved sequences. This gene also
offers the opportunity to perform a PCR with the 16 S rDNA or a
reverse transcriptase-PCR (RT-PCR) with the cDNA of the 16 S
rRNA (see Note 2) (6). Here, we describe the use of a mixture of
oligonucleotides for the specific detection of mycoplasmas. This
approach reduces significantly the generation of false positive
results due to contaminations of the solutions for the sample prep-
aration and the PCR run and of other materials with airborne bac-
teria. Nevertheless, major emphasis should be placed on the
preparation of the template DNA, the amplification of positive and
negative control reactions, and the observance of general rules for
the preparation of PCR reactions. One of the main problems con-
cerning PCR reactions with samples from cell cultures is the inhibi-
tion of the Taq polymerase by unspecified substances. To eliminate
those inhibitors, we strongly recommend that the sample DNA be
extracted and purified by conventional phenol–chloroform extrac-
tion or by the more convenient column- or matrix-binding extrac-
tion methods.
We compared the conventional hot start PCR method with a
real-time PCR employing SYBR green DNA-binding dye and
found no significant advantage of the real-time PCR over the con-
ventional PCR method. In contrast, due to the short amplification
cycles of the conventional PCR and the necessity to perform a
denaturation curve in the real-time PCR assay, the PCR method
presented here is faster and similarly convenient.
To confirm the error-free preparation of the sample and PCR

run, appropriate control reactions have to be included in the PCR.
These comprise internal control DNA for every sample reaction
and, in parallel, positive and negative as well as water control reac-
tions. The internal control consists of a DNA fragment with the
same primer sequences for amplification, but is of a different size
than the amplicon of mycoplasma-contaminated samples. This
control DNA is added to the PCR mixture in a previously deter-
mined limiting dilution to demonstrate the sensitivity of the PCR
reaction. In this chapter, detailed protocols are provided to estab-
lish the PCR method for the monitoring of mycoplasma contami-
nations in any laboratory.
2. Materials
1. Phosphate-buffered saline (PBS): 140 mM NaCl, 2.7 mM

KCl, 7.2 mM Na2HPO4 × 12 H2O, 1.47 mM KH2PO4, pH
7.2. Autoclave for 20 min at 121°C to sterilize the solution.
2. 50× Tris–acetic acid–EDTA (TAE): 2 M Tris base, 5.71% gla-
cial acetic acetate (v/v), 100 mM EDTA. Adjust to pH of
approx. 8.5.
3. DNA extraction and purification system (e.g., phenol/chloro-
form extraction and ethanol precipitation, or DNA extraction
kits applying DNA binding matrices).
4. GeneAmp 9700 thermal cycler (Applied Biosystems,
Weiterstadt, Germany).
5. Platinum Taq DNA polymerase, hot start, 5 U/mL (Invitrogen,
Karlsruhe, Germany), including 10× PCR buffer provided by
the manufacturer.
6. 50 mM MgCl2.
7. 6× loading buffer: 0.09% (w/v) bromophenol blue, 0.09%
(w/v) xylene cyanol FF, 60% glycerol (v/v), 60 mM EDTA.
8. Primers (any supplier) (see Note 3):
5¢ primers (Myco-5¢):
cgc ctg agt agt acg twc gc
tgc ctg rgt agt aca ttc gc
crc ctg agt agt atg ctc gc
cgc ctg ggt agt aca ttc gc
3¢ primers (Myco-3¢):
gcg gtg tgt aca ara ccc ga
gcg gtg tgt aca aac ccc ga
(r = mixture of g and a; w = mixture of t and a)

Stock solutions: 100 mM in dH2O, stored frozen at −20°C.
Working solutions: Mixture of forward primers at 5 mM each
(Myco-5¢) and mixture of reverse primers at 5 mM each (Myco-
3¢) in dH2O, aliquoted in small amounts (i.e., 25–50 mL ali-
quots), and stored frozen at −20°C.
9. Mycoplasma PCR internal control DNA (4): Plasmid contain-
ing an insert producing a longer PCR product than the myco-
plasma wild-type PCR product. The plasmid can be obtained
from the DSMZ (German Collection of Microorganisms and
Cell Cultures, Braunschweig, Germany). Before use, a limiting
dilution should be determined experimentally by performing a
PCR with a dilution series of the internal control DNA (see
Note 4).
10. Mycoplasma-positive control DNA: A tenfold dilution of any
mycoplasma-positive sample prepared as described in
Subheading 3.1, or obtained from the DSMZ.
11. 5 mM deoxy-nucleotide triphosphate mixture (dNTP-mix) is
prepared from the nucleotides of the 100 mM dNTP set
(Invitrogen, Karlsruhe, Germany) by diluting each of deoxy-
adenosine triphosphate (dATP), deoxycytidine triphosphate
(dCTP), deoxyguanosine triphosphate (dGTP), and deoxythy-
midine triphosphate (dTTP) 1:5 with dH2O and combining
the same volumes (e.g., 100 mL each). The resulting working
solution with each nucleotide at a concentration of 5 mM is
then aliquoted in small amounts (i.e., 25–50 mL aliquots) and
stored at −20°C.
12. 1.3% Agarose-TAE gel (7).
13. Optional: Restriction enzymes AspI, HaeIII, HpaII, XbaI
including the 10× restriction enzyme buffers recommended by
the manufacturer.
3. Methods
The following subsections describe the sample collection, extrac-

tion of the DNA, setting up and performing the PCR reaction,
interpretation of the results, and, in addition, the identification of
the mycoplasma species. These techniques can also be used to
detect mycoplasma contamination in culture media or other sup-
plements (see Note 5).
Every incoming cell culture should be kept in quarantine until
mycoplasma detection assays are completed and the infection status
is clearly determined. Positive cultures should either be discarded
and replaced by clean cultures or cured with specific antibiotics

(see Chapter 2). Only definitely clean cultures should be used for
research experiments and for the production of biologically active
pharmaceuticals. Additionally, stringent rules for the prevention of
further mycoplasma contamination of cell cultures should be
strictly followed (1).
3.1. Sample Collection 1. Prior to collecting the samples, the cell line to be tested for
and Preparation mycoplasma contamination should be in continuous culture
of DNA for several days and without any antibiotics (even penicillin and
streptomycin) or for at least 2 weeks after thawing. This should
assure that the titer of the mycoplasmas in the supernatant is
above the detection level of the PCR assay.
2. Take one milliliter of the supernatant of adherently growing
cells or of cultures with settled suspension cells (by placing the
cell culture vessel in an upright position for approximately
30–60 min) for the analysis. By collecting the samples in this
way, some viable and/or dead eukaryotic cells are always present
in the samples. This is intended as some mycoplasma strains pre-
dominantly adhere to the eukaryotic cells or even invade them.
On the other hand, only a limited number of eukaryotic cells is
desired to prevent an excess of eukaryotic DNA. Thus, it is also
not necessary to centrifuge the sample to eliminate the eukary-
otic cells. The crude cell culture supernatants can be stored at
4°C for a few days or frozen at −20°C for several weeks. After
thawing, the samples should be further processed immediately.
3. Centrifuge the cell culture suspension at 13,000 × g for 5 min.
Resuspend the pellet in 1 mL PBS by vortexing.
4. Centrifuge the suspension again and wash one more time with
PBS as described in step 3.
5. After centrifugation, resusupend the pellet in 100 mL PBS by
vortexing and heat to 95°C for 15 min.
6. Immediately after lysing the cells, extract and purify the DNA
by standard phenol/chloroform extraction and ethanol pre-
cipitation (7) or other DNA isolation methods (see Note 6).
3.2. PCR The amplification procedure and the parameters described here are
optimized for the application in thin-walled 0.2-mL reaction tubes
in an Applied Biosystems GeneAmp 9700 thermal cycler. An
adjustment to any other equipment might be necessary (see Note
7). Amplified positive samples contain high amounts of target
DNA. Thus, established rules to avoid DNA carryover should be
strictly followed: (1) the places where the DNA is extracted, the
PCR reaction is set up, and the gel is run after the PCR should be
separated from each other; (2) all reagents should be stored in
small aliquots to provide a constant source of uncontaminated
reagents; (3) avoid reamplifications; (4) reserve pipettes, tips, and

tubes for their use in the PCR only and irradiate the pipettes fre-
quently by ultraviolet (UV) light; (5) the succession of the PCR
setup described below should be followed strictly; (6) wear gloves
during the whole sample preparation and PCR setup; and (7)
include the appropriate control reactions, such as internal, positive,
negative, and the water control reaction.
1. For each sample to be tested, set up one reaction with the sam-
ple only and another parallel reaction also containing the inter-
nal control DNA (to confirm the accuracy of the PCR run):
Sample plus internal
Sample only control DNA
dNTPs 1 mL 1 mL
Myco-5¢ 0.5 mL 0.5 mL
Myco-3¢ 0.5 mL 0.5 mL
10× PCR buffer 2.5 mL 2.5 mL
MgCl2 1 mL 1 mL
dH2O 18.3 mL 17.3 mL
Platinum Taq polymerase 0.2 mL 0.2 mL
Internal control DNA – 1 mL
When testing several samples, pre-master mixtures can be

prepared. For the reaction without internal control DNA two
reactions should be added (for the positive and the water con-
trol reactions) and for the reactions with the internal control
DNA two reactions should be added (for the positive and the
negative control reaction) (see Notes 4 and 8). For both pre-
master mixtures, also add the amounts for one additional reac-
tion to have a surplus for pipetting variations.
2. Transfer 24 mL of each of the pre-master mixtures to 0.2 mL
PCR reaction tubes and add 1 mL dH2O to the water control
reaction and the internal control only reaction.
3. Set aside all reagents used for the preparation of the master
mixtures and take out the samples of DNA to be tested and the
positive control DNA. Do not handle the reagents and samples
simultaneously. For each DNA sample, and for the positive
control DNA, add 1 mL of the DNA preparation to a reaction
tube that contains no internal control DNA and another 1 mL
to a tube containing the internal control DNA.
4. Transfer the reaction tubes to the thermal cycler and start a
one-step cycle at 96°C for 2 min to activate the hot start Taq
polymerase.
5. After this initial cycle, perform 35 thermal cycles with the

following parameters:
Cycle step 1: 4 s at 95°C.
Cycle step 2: 8 s at 65°C.
Cycle step 3: 16 s at 72°C plus 1 s of extension time during
each cycle.
6. Finish the reaction with a final amplification step at 72°C for
7 min, and then cool the samples down to room temperature.
7. Prepare a 1.3% agarose-TAE gel containing either 0.3 mg of
ethidium bromide per mL or with another nucleic acid gel
stain. Submerge the gel in 1× TAE and add 12 mL of the
amplification product (10 mL reaction mixtures plus 2 mL of 6×
loading buffer) to each well. Run the gel at 10 V/cm.
8. Visualize the specific products on a suitable UV screen and
document the result.
3.3. Interpretation Figure 1 shows a representative ethidium bromide-stained gel with

of Results some samples that produce the following results:
– Ideally, all samples containing the internal control DNA show
a band at 986 bp. This band might be more or less bright, but
the band should be visible even if no other bands are amplified
(see Note 9). If the band is not visible, the reaction may have
been contaminated with Taq polymerase inhibitors from the
cell culture that were not removed by the DNA extraction
method. For example, melanin is a potent Taq polymerase
inhibitor that occurs in melanoma cell cultures and cannot be
eliminated by any DNA extraction method. In such a case,
bovine serum albumin (BSA) can be added to the sample at
high concentrations to overcome the inhibition. However,
usually it is sufficient to repeat the PCR run with the same
DNA solution as used previously. It is not necessary to collect
a new sample from the cell culture. If the second run also does
not show a band for the sample and internal control, the whole
procedure should be repeated.
– Mycoplasma-positive samples show a band at 502–520 bp,
depending on the mycoplasma species. In the case of a
Acholeplasma laidlawii contamination, and when the DSMZ
internal control DNA is applied, a third band might be visible
between the internal control band and the mycoplasma-specific
band. This is formed by cross-hybridization of the comple-
mentary sequences of the single-stranded long internal control
DNA and the shorter, single-stranded wild-type mycoplasma
DNA form.
– Contamination of reagents with mycoplasma-specific DNA or
PCR products is revealed by a band in the water control and/
or in the negative control sample.
Fig. 1. PCR analysis of mycoplasma status in cell lines. Shown is an ethidium bromide-stained gel containing the reaction
products following PCR amplification with the primer mix listed in Subheading 2. Products of about 510 bp are obtained;
the differences in length reflect the sequence variation between different mycoplasma species. Shown are various exam-
ples of mycoplasma-negative and -positive cell lines. Two paired PCR reactions were performed: one PCR reaction con-
tained an aliquot of the sample only (a) and the second reaction contained the sample under study plus the control DNA as
internal standard (b). Cell cultures A, C, and E are mycoplasma positive; cell culture B is mycoplasma negative. The result
of cell culture D is not evaluable because the internal control was not amplified and no other mycoplasma-specific band
appeared in the gel. In this case, the analysis needs to be repeated. Cell line C 2 weeks post antibiotic treatment shows a
weak but distinctive band in the reaction without internal control. This band results from residual DNA in the medium,
because after a further 2 weeks of culture no contamination was detected.
– Weak mycoplasma-specific bands can occur after treatment of

infected cell cultures with anti-mycoplasma reagents for the
elimination of mycoplasma or when other antibiotics are used
routinely. In these cases the positive reaction might be due to
residual DNA derived either from dead mycoplasma cells in
the culture medium or from viable mycoplasma cells which are
present at a very low titer. Therefore, special caution should be
taken when cell cultures are tested that were treated with anti-
biotics. Prior to PCR testing, cell cultures should be cultured
for at least 2–3 weeks without antibiotics or retested at fre-
quent intervals to demonstrate either a decrease or an increase
of mycoplasma.
3.4. Identification of Although the method described is sufficient to detect mycoplasma

Mycoplasma Species contaminations, it might be of advantage to know the infecting
mycoplasma species, e.g., in efforts to determine the source of a
contamination. This PCR allows the identification of the myco-
plasma species most commonly infecting cell cultures by modified
Xba I
M. arginini – M. hominis – M. hyorhinis – M. orale A. laidlawii – M. bovis – M. fermentans

~265 bp / ~253 bp
Hpa II Hpa II
M. arginini – M. hominis – M. hyorhinis M. orale A. laidlawii – M. bovis M. fermentans

288 bp / 230 bp 357 bp / 111 bp / 48 bp
Asp I Asp I
M. hyorhinis M. arginini – M. hominis M. bovis A. laidlawii

303 bp / 213 bp 303 bp / 213 bp
Hae III
M. hominis M. arginini digested

336 bp / 180 bp
not digested
Fig. 2. Flowchart for the identification of the mycoplasma species. Digesting aliquots of the amplified PCR product with the
indicated restriction enzymes will result in undigested (solid lines ) or digested (dashed lines ) fragments of the sizes men-
tioned below the species names.
restriction fragment length polymorphism analysis. In case of a

contamination detected by PCR, the PCR reaction is repeated in a
50 mL volume without the internal control DNA to amplify only
the wild-type mycoplasma-specific PCR fragment. Per reaction,
8 mL of the amplified DNA is taken directly from the PCR reaction
and is digested in parallel reactions with the restriction endonu-
cleases Asp I, Hae III, Hpa II, and Xba I by the addition of 1 mL
of the appropriate 10× restriction enzyme buffer and 1 mL of the
restriction enzyme. The mycoplasma species can be determined
directly by the restriction pattern (see Fig. 2). This analysis only
identifies those mycoplasma species which most often (>98%) occur
in cell cultures, and is not suitable for the global identification of
all types of mycoplasma species. Cell culture infections are com-
monly restricted to the mycoplasma species listed in Fig. 2.
4. Notes
1. Originally, the described method was designed as nested PCR

(8). Here, the second round of PCR was omitted, because
in standard applications no significant differences in the
results were observed between only one round of PCR and
nested PCR. Mycoplasma-positive cell cultures were detected

as positive in the first round of PCR and negative samples
were consistently negative employing nested PCR.
Furthermore, applying a nested PCR increases the risk of
transmission of first-round PCR products to the reagents
used in the second amplification and possibly to those shared
with the first round.
2. In this protocol, genomic DNA is applied for the PCR reac-
tion. As the primers hybridize to the 16S rRNA, an RT-PCR
can also be performed after extracting RNA and preparation of
cDNA. Theoretically, RT-PCR should increase the sensitivity
of the assay, because the number of rRNA molecules per organ-
ism is much higher than the number of the coding genes.
Nevertheless, we found no significant increase in sensitivity
employing cDNA, and thus the additional labor, time, and
costs required for RT-PCR are unwarranted.
3. The primers can be prepared using a degenerated code to
incorporate two different nucleotides at one site to form a
mixture of two primers. When the forward or reverse oligo-
nucleotides are mixed and aliquoted for use in the PCR reac-
tion, it must be taken into account that the molarities of the
oligonucleotides with mixed bases are reduced by 50%. The
primer solutions should be aliquoted into small portions
(i.e., 25 mL aliquots) and stored frozen at −20°C to avoid
multiple freeze-thawing cycles and to minimize contamina-
tion risks.
4. The limiting dilution of the internal control DNA can be used
for 2 or 3 months at most when stored at 4°C. After this time,
the amplification of the internal control DNA might fail even
when no inhibitors are present in the reaction, because the
DNA concentration may be reduced due to degradation or
adherence to the plastic tube.
5. To use this PCR method to test cell culture media or supple-
ments expected to have low titers of mycoplasmas (e.g., fetal
bovine serum [FBS]), a larger quantity of the sample can be
centrifuged. The pellet is further processed as described in this
protocol without adjusting the reagent amounts to the higher
initial sample size.
6. We do not recommend using the crude lysate of a sample for
the PCR reaction as described in some publications, because it
often contains inhibitors of Taq polymerase and may lead to
false negative results. For convenience and speed of the assay
we use commercially available DNA extraction/purification
kits based on binding of the DNA to matrices and subsequent
elution of the DNA. We tested normal phenol/chloroform
extraction and ethanol precipitation, the High Pure PCR
Template Preparation Kit from Roche (Mannheim, Germany),

the Invisorb DNA/RNA Virus Kit from Invitek (Berlin,
Germany), the Wizard DNA Clean-Up System (for plasmid
purification) from Promega (Mannheim, Germany), and the
Epicentre MasterPure Complete DNA & RNA Purification Kit
(Biozym, Hessisch Oldendorf, Germany). Following the rec-
ommendations of the manufacturers, the amplifications of the
mycoplasma sequences were all similar when the same amounts
were used for the elution or resuspension. For screening many
samples, the Wizard system works very well with the vacuum
manifold.
7. The use of thermal cyclers other than the GeneAmp 9700
might require some modifications in the amplification param-
eters (e.g., duration of the cycling steps, which are short in
comparison to other applications). Magnesium, primer, or
dNTP concentrations might also need to be altered. The same
may be true if another Taq polymerase is used, either poly-
merases from different suppliers or different kinds of Taq poly-
merase. Although the described PCR method is very robust,
we found that the type of Taq polymerase and the appropriate
buffer system have a major impact on the performance and
sensitivity of the PCR.
8. Using the internal control DNA, the described PCR method
is competitive only for the group of mycoplasma species that
carry primer sequences identical to the one from which the
internal control DNA was prepared. Other primer sequences
are not used up in the PCR reaction because of mismatches.
Usually, one reaction per sample is sufficient to detect myco-
plasma in long-term infected cell cultures. But to avoid the
risk of performing a competitive reaction and of decreasing
the sensitivity of the PCR reaction (e.g., after anti-myco-
plasma treatment or for the testing of cell culture reagents),
two separate reactions are performed: (1) without internal
control DNA to make all reagents available for the amplification
of the specific product and (2) including the internal control
DNA to demonstrate the integrity of the PCR reaction (see
Fig. 1).
9. Heavily infected cell cultures might show the mycoplasma-
specific band, whereas the internal control is not visible. In this
case the mycoplasma target DNA suppresses the internal con-
trol which is present in the reaction mixture at much lower
concentrations. The reaction is classified as mycoplasma posi-
tive (see Fig. 1).
References
1. Uphoff CC, Drexler HG (2001) Prevention of 5. European Pharmacopoeia (2008) General

mycoplasma contamination in leukemia-lym- chapter 2.6.7., Mycoplasmas, suppl. 5.6.
phoma cell lines. Hum Cell 14:244–247 Council of Europe, Strasbourg, France
2. Drexler HG, Uphoff CC (2002) Mycoplasma 6. Uphoff CC, Drexler HG (1999) Detection of
contamination of cell cultures: incidence, mycoplasma contamination in cell cultures by
sources, effects, detection, elimination, preven- PCR analysis. Hum Cell 12:229–236
tion. Cytotechnology 39:23–38 7. Sambrook J, Fritsch EF, Maniatis T (eds)
3. Uphoff CC, Drexler HG (2010) Contamination (1989) Molecular cloning, a laboratory man-
of cell cultures, mycoplasma. In: Flickinger M ual, 2nd edn. Cold Spring Harbor Laboratory
(ed) The encyclopedia of industrial biotechnol- Press, Cold Spring Harbor, NY
ogy. Wiley, New York 8. Hopert A, Uphoff CC, Wirth M, Hauser H,
4. Uphoff CC, Drexler HG (2002) Comparative Drexler HG (1993) Specificity and sensitivity
PCR analysis for detection of mycoplasma of polymerase chain reaction (PCR) in com-
infections in continuous cell lines. In Vitro Cell parison with other methods for the detection
Dev Biol Animal 38:79–85 of mycoplasma contamination in cell lines.
J Immunol Methods 164:91–100
Chapter 2
Eradication of Mycoplasma Contaminations

Abstract
Mycoplasma contaminations have a multitude of effects on the cultured cell lines that may influence the
results of experiments or pollute bioactive substances used in human medicine. The elimination of myco-
plasma contaminations of cell cultures has become a practical alternative to discarding and reestablishing
important or irreplaceable cell lines. Different quinolones, tetracyclines, and pleuromutilins shown to have
strong antimycoplasma properties are employed for the decontamination. We provide detailed protocols
to assure eradication of mycoplasma, to prevent formation of resistant mycoplasma strains, and to cure
heavily contaminated and damaged cells. To date, we have not detected any consistent and permanent
alterations to eukaryotic cells either during or after the treatment.
Key words: Antibiotic elimination, Cell lines, Mycoplasma
1. Introduction
The use of human and animal cell lines for the examination of
biological functions and for the production of bioactive substances
requires rigorous quality control to exclude contamination with
organisms (i.e., other eukaryotic cells, bacteria, and viruses). In
this respect, mycoplasmas play an important but undesirable role,
because a high portion (more than 20%) of the cell cultures arriv-
ing at our cell lines collection are contaminated with these wall-less
bacteria. Mycoplasmas can have a multitude of effects on eukary-
otic cells and can alter almost every cellular parameter, from prolif-
eration via signaling pathways to virus susceptibility and production.
Most striking are the effects resulting from the competition for
nutrients that leads to the depletion of a number of essential nutri-
ents. Consequentially, many downstream effects, such as altered
15
levels of protein, DNA, and RNA synthesis, and alterations of

cellular metabolism and cell morphology, can be detected.
Mycoplasmas do not gain energy by oxidative phosphorylation,
but from the fermentative metabolism of diverse nutrients. This
can lead to an alteration of the pH in the medium and to the
production of metabolites that are toxic to the eukaryotic cells
(e.g., NH3). The dependence of many mycoplasmas on choles-
terols, sterols, and lipids can result in an alteration of the mem-
brane composition. Other activation and suppression processes
have also been described (e.g., lymphocyte activation, cytokine
expression, induction of chromosomal aberrations, etc.). It has
been noted that many experimentally analyzed parameters that
were at first attributed to the eukaryotic cells were later ascribed to
the contaminating mycoplasmas or were influenced by them. For
example, mycoplasmas carry a uridine phosphorylase that can inac-
tivate the artificial deoxynucleotide bromodeoxyuridine (BrdU).
Cells with a thymidine kinase defect are commonly used for cell
fusions and selected by the addition of BrdU. If mycoplasmas inac-
tivate BrdU, the growing eukaryotic cells might appear to carry
the enzyme deficiency and are misleadingly selected for cell fusions.
Cell lines for virus propagation are also often affected by mycoplasma
infections, leading to higher or lower titers of viruses (1).
When an infected cell culture is detected, it should be auto-
claved and discarded immediately and replaced by a mycoplasma-
free culture. However, some cell lines are not replaceable because
of unique characteristics of the cells or due to all the work that has
been invested to manipulate those particular cells.
A number of methods have been suggested to eradicate
mycoplasmas from cell cultures. They comprise physical, chemical,
immunological, and chemotherapeutic treatment. Some treat-
ments are restricted to surfaces only (e.g., exposure to detergents),
to eukaryotic cell-free solutions such as fetal bovine serum (FBS)
(e.g., filtration through microfilters), and to specific mycoplasma
species (e.g., culture with antimycoplasma antisera), are not practi-
cable for a standard cell culture laboratory (e.g., in vivo passage of
continuous cell lines through nude mice cell cloning), or are inef-
fective in eliminating the mycoplasmas quantitatively (e.g., heat
treatment, exposure to complement) (2). That some mycoplasma
species have the ability to penetrate the eukaryotic cell should also
be considered. Mycoplasma fermentans is one of the main infecting
mycoplasma species that can also enter eukaryotic cells. Thus,
eliminating agents must also be active intracytoplasmically.
Chemotherapeutic treatment can be efficiently employed using
specific antibiotics. Because mycoplasmas possess no rigid cell walls
and have a highly reduced metabolism, many of the commonly
used antibiotics exhibit no effect on the viability of the mycoplas-
mas. They are naturally resistant to antibiotics targeting cell wall
biosynthesis (e.g., penicillins), have an acquired resistance against
2 Eradication of Mycoplasma Contaminations 17
other antibiotics that are often prophylactically used in cell culture

(e.g., streptomycin), or the antibiotics are effective only at concen-
trations which are detrimental to the eukaryotic cells as well.
Hence, the general use of antibiotics in cell culture is not recom-
mended, except under special circumstances and then only for
short durations. General use of antibiotics could lead to selection
of drug-resistant organisms, to lapses in aseptic technique, and to
delayed detection of low-level infection with either mycoplasmas
or other bacteria (3).
Three classes of antibiotics have been shown to be highly
effective against mycoplasmas, both in human/veterinary medi-
cine and in cell culture: tetracyclines, pleuromutilins, and quino-
lones. These antibiotics can be used at relatively low concentrations,
with a negligible likelihood of resistance development, and, finally,
with low or no effects on the eukaryotic cells. Tetracyclines and
pleuromutilins inhibit protein synthesis by binding to the 30S and
50S ribosomal subunits, respectively (4). Quinolones inhibit the
bacterial DNA gyrase which is essential for mycoplasma DNA rep-
lication. The risk of development of resistant clones is minimized
by the application of antibiotics with different mechanisms of
action, by sufficient treatment durations, and by constant concen-
trations of the antibiotics in the medium (5). Here, we describe
the use of several antibiotics for the treatment of mycoplasma-
contaminated cells, the rescue of heavily infected cultures, the
salvage treatment of resistant cultures, and some pitfalls during
and after the treatment.
2. Materials
(See Note 1)
1. BM-Cyclin (Roche, Mannheim, Germany) contains the pleu-
romutilin tiamulin (BM-Cyclin 1) and the tetracycline minocy-
cline (BM-Cyclin 2), both in lyophilized states. Dissolve the
antibiotics in 10 mL sterile distilled water (dH2O), aliquot in
1-mL fractions, and store at −20°C. These stock solutions have
concentrations of 2.5 mg/mL and 1.25 mg/mL, respectively.
Repeated freezing and thawing of the solutions are not detri-
mental to the activity of the antibiotics. The dissolved solu-
tions can be used at 1:250 dilutions in cell culture (at 10 mg/
mL and 5 mg/mL final concentration, respectively).
2. Plasmocin (InvivoGen, San Diego, CA) contains two antibiot-
ics; one is active against protein synthesis of the bacteria, and
one inhibits the DNA replication (gyrase inhibitor) (specific
types of reagents not disclosed). The mixture is a ready-to-use
solution and applied 1:1,000 in the cell culture (at 25 mg/mL
final concentration).
3. Ciprobay 100 (Bayer, Leverkusen, Germany) is a ready-to-use

solution containing 2 mg/mL ciprofloxacin. It can be used
1:200 in cell culture (at 10 mg/mL final concentration). One-
milliliter aliquots should be taken sterilely from the bottle and
stored at 4°C. Crystals form at 4°C and can be redissolved at
room temperature.
4. Baytril (Bayer) contains 100 mg/mL of enrofloxacin and is
diluted 1:100 with RPMI 1640 medium immediately prior to
the treatment. The dilution should be prepared freshly for every
antimycoplasma treatment. This solution is used as 1:40 final
dilution in cell culture (at 25 mg/mL final concentration).
5. Mycoplasma Removal Agent (MRA, ICN, Eschwege, Germany)
is a ready-to-use dilution containing 50 mg/mL of a 4-oxo-
quinolone-3-carboxylic acid derivative (specific type of reagent
not disclosed). It is used in the treatment of cell cultures in
1:100 dilutions (at 0.5 mg/mL final concentration).
6. MycoZap (Lonza, Verviers, Belgium) is a combination of an
antimicrobial peptide (MycoZap reagent 1) and an antibiotic
(MycoZap reagent 2) (specific types of reagents not dis-
closed) that are employed consecutively. The solutions are
ready-to-use.
7. Phosphate-buffered saline (PBS): 140 mM NaCl, 2.7 mM
KCl, 7.2 mM Na2HPO4 × 12 H2O, 1.47 mM KH2PO4. Adjust
to pH 7.2 and autoclave for 20 min at 121°C.
8. Cell culture media and supplements as appropriate and recom-
mended for the particular cell lines.
3. Methods
3.1. Pretreatment 1. If no frozen reserve ampoules of the cell line are available,
Procedures aliquots of the contaminated cell line should be stored frozen
before treatment. Whenever possible, the ampoules should be
kept isolated from noninfected cultures, either at −80°C for
short time (over the complete curation time of 1–2 months)
or, preferably, in liquid nitrogen in separate tanks (see Note 2).
The ampoules should be marked properly as “mycoplasma-
positive” to prevent a mix-up of ampoules containing cured or
infected cells. After successful cure, these mycoplasma-positive
ampoules should be removed and the cells destroyed by
autoclaving.
2. Prepare the antibiotic working solutions freshly for every treat-
ment and add the solution directly to the cell culture, not to
the stored medium.
3. The FBS concentration should be increased to 20% before,

during, and for at least 2 weeks after the treatment to ensure
optimal growth conditions, even if the cells grow well at lower
concentrations.
3.2. Antibiotic Mycoplasma infection often impairs the growth and viability of
Treatment eukaryotic cells. After addition of the antibiotic, heavily infected
cells might recover immediately and the viability of the culture
might increase rapidly. However, in other cases, the delicate health
of the cells is further aggravated by the exposure to the antibiotics.
One reason might be the partial inhibition of mitochondrial respi-
ration by the antibiotic(s). Even though the optimal concentra-
tions of the antibiotics were determined in many trials, different
cell types or cells under different infection conditions might behave
differently upon treatment. Thus, in some instances, the cultures
might be killed by the treatment (5). In these events, the treatment
must be repeated with an aliquot that was stored frozen prior to
the treatment. Even when no antibiotics are added to the medium,
the cells might reach a crisis and die. To counteract the treatment-
associated harm, a few general suggestions should be followed to
improve the culture conditions and to reduce the stress of infection
and treatment on the eukaryotic cells (these rules are suitable for
most cell lines, but some cell lines require special care which must
be determined by the user):
● Keep the concentration of the antibiotic constant during the
treatment period; degradation of the antibiotic can be avoided
by frequent complete exchange of the medium noting the
following caveats.
● Culture the cells at a medium or higher cell density and keep
this density almost constant during the treatment and for a few
weeks after; a higher density of the cells demands a more fre-
quent change of medium, which is commonly preferable to a
relatively low cell density and long intervals between medium
changes. However, some cell lines reportedly produce their
own growth factors and, therefore, the medium should not be
fully exchanged, depending on the cell line (see Note 3).
● Observe the culture daily under the inverted microscope to
recognize quickly any alteration in general appearance, growth,
or morphology, decrease in cell viability, detachment of cells,
formation of granules, vacuoles, and so forth.
● In the case of deterioration of the cell culture, interrupt the
treatment for a few days and let the cells recover (but this should
only be the last resort); culture conditions should be changed
immediately after recognition of the alterations, because if the
cells are already beyond a certain degree of damage, it is usually
difficult to reverse the progression of apoptosis.
● If possible, frequently detach slowly growing adherent cells in

order to facilitate the exposure of all mycoplasmas to the anti-
biotic; the contaminants should not have the opportunity to
survive in sanctuaries such as cell membrane pockets. It is simi-
larly helpful to break up clumps of suspension cells by vigorous
pipetting or using other reagents (e.g., trypsin, TrypLE Express
(Invitrogen, Darmstadt, Germany], or Accutase (Sigma,
Munich, Germany]).
● As antibiotics are light sensitive; protect cultures from the
light, as much as possible.
Generally, there are three different approaches to the treatment of
infected cell cultures: (1) the use of a single antibiotic compound
(e.g., the quinolones), with basically the same procedure employed
for each antibiotic of that group; (2) the simultaneous application
of two different antibiotics (e.g., in the case of Plasmocin); and (3)
the use of a combination therapy applying two antimycoplasma
agents subsequently (e.g., MycoZap) or in alternating cycles
(e.g., BM-Cyclin) (4) (see Fig. 1 and Note 4). The latter method
is more time-consuming, but also highly effective, and avoids the
possible interference of two antibiotics. For example, the action of
the bactericidal quinolones depends on the proliferation of the
cells, which is compromised by bacteriostatic agents, such as tetra-
cyclines. We recommend using two of the three types of treatment
in parallel or subsequently, if one method fails.
A schematic overview of the procedure is given in Fig. 1; an
exemplary representation of the treatment with BM-Cyclin is
shown in Fig. 2.
3.2.1. Treatment 1. Prepare a cell suspension (detach adherent cells, break up

with BM-Cyclin clumps by pipetting or using other methods) (see Note 5);
determine the cell density and viability by trypan blue exclu-
sion staining. Seed the cells at a medium density (see Note 6)
in a 25 cm2 flask or one well of a 6- or 24-well culture plate
with the appropriate fresh and rich culture medium (10 mL for
the flask, and 4 mL and 2 mL for the wells, respectively). Add
4 mL of a 2.5 mg/mL solution BM-Cyclin 1 (tiamulin) per
milliliter of medium. Incubate the cell culture for 2 days.
2. Remove all cell culture medium in flasks or wells containing
adherent cells or after centrifugation of suspension cells. If
applicable, dilute the cell cultures to a medium cell density.
Add fresh medium and the same concentration of BM-Cyclin
1 as used in step 1. Incubate for another day. This procedure
will keep the concentration of the antibiotic approximately
constant over the 3-day application of tiamulin.
3. Remove the medium and wash the cells once with PBS to
remove the residual antibiotic agent completely from the cells
Mycoplasma-infected Cell Line 1
2 Freeze Aliquots of the

Infected Cell Line as Back-up
Treat Cell Line with
Quinolone Plasmocin MycoZap BM-Cyclin
Mycoplasma PCR Testing Mycoplasma-negative
Mycoplasma-positive Expand Cells, Freeze Master

Stock, Store in Liquid
Nitrogen, Discard Myco-
plasma-positive Back-ups
Quinolone- / Plasmocin- / MycoZap-resistant BM-Cyclin-resistant
(1) BM-Cyclin Treatment

(1) Repeat Treatment of Untreated
(2) Treatment of Untreated Aliquot with
Aliquot with BM-Cyclin
another Quinolone / Plasmocin /
(2) Treatment of Untreated Aliquot
MycoZap
with a Quinolone / Plasmocin
(3) Other Elimination Method
Fig. 1. Scheme for mycoplasma eradication. Different antibiotics can be used to treat mycoplasma-contaminated cell lines
with a high rate of expected success. We recommend (1) cryopreservation of original mycoplasma-positive cells as back-
ups and (2) splitting of the growing cells into different aliquots. These aliquots should be exposed singly to the various
antibiotics. Posttreatment mycoplasma analysis and routine monitoring with a sensitive and reliable method (for example
by PCR) are of utmost importance.
and loosely attached mycoplasmas. Seed the cells at the appro-

priate density (as described in step 1; see Note 6) and add 4 mL
of the 1.25 mg/mL solution BM-Cyclin 2 (minocycline) per
milliliter of medium. Incubate the culture for 2 days.
4. Remove the culture medium and substitute with fresh medium.
Add the same concentration of BM-Cyclin 2 as used in step 3.
Washing with PBS is not necessary at this step. Incubate the
cell culture for 2 days to complete the 4-day minocycline
treatment.
Post-
treatment Mycoplasma-
Treatment Testing
Passaging
W W W W W W
0 4 6 8 10 12 14 16 18 20 22 24 26 28 30 32 34 36 38 40 Days
Antibiotic-free Medium
1 12 2 1 1 2 2 1 1 2 2
Medium with
10 µg/mL BM-Cyclin 1 or
5 µg/mL BM-Cyclin 2
Fig. 2. Treatment protocol for BM-Cyclin. Antibiotics are given on the days indicated by arrows. Cells are washed (indicated
by W) with PBS prior to the cyclical change of antibiotics to avoid formation of resistant mycoplasmas due to low concen-
trations of the antibiotics. At the end of the decontamination period, cells are washed with PBS and suspended in antibiotic-
free medium. After a minimum of 2 weeks of posttreatment, the mycoplasma status of the cells is examined with sensitive
and robust methods (for example by PCR).
5. After washing the cells with PBS, repeat steps 1–4 twice
(three cycles of BM-Cyclin 1 and BM-Cyclin 2 altogether).
Proceed with Subheading 3.3.
3.2.2. Treatment with 1. Prepare a cell suspension (detach adherent cells, break up
Quinolones and Plasmocin clumps by pipetting or using other methods) (see Note 5);
sion staining. Seed the cells at a medium density (see Note 6)
in a 25 cm2 flask or one well of a 6- or 24-well culture plate
with the appropriate fresh and rich culture medium (10 mL for
the flask, and 4 mL and 2 mL for the wells, respectively). Add
one of the following antibiotics to the cell culture and incubate
for 2 days:
(a) 25 mL of a 1 mg/mL solution of enrofloxacin (Baytril)
per milliliter of medium.
(b) 10 mL of a 50 mg/mL solution of MRA per milliliter of
medium.
(c) 5 mL of a 2 mg/mL solution of ciprofloxacin (Ciprobay)
per milliliter of medium.
(d) 1 mL of a 25 mg/mL solution of Plasmocin per milliliter
of medium.
2. Remove all cell culture medium in flasks or wells containing
adherent cells or after centrifugation of suspension cells.
If applicable, dilute the cell cultures to a medium cell density.
Add fresh medium and the same concentration of the respec-
tive antibiotic as used in step 1. Incubate for another 2 days.
3. If using enrofloxacin or MRA, repeat step 2 another two times

(altogether an 8-day treatment). If using ciprofloxacin or
Plasmocin, repeat step 2 five times (altogether a 14-day treat-
ment). Proceed with Subheading 3.3.
3.2.3. Treatment 1. The activity of the antimicrobial peptide is influenced by the

with MycoZap concentration of the FBS. Thus, the FBS concentration of the
cell culture medium should not exceed 5% during the treat-
ment with MycoZap reagent 1. Add 500 mL MycoZap reagent
1 to 4.5 mL cell culture medium supplemented with maxi-
mally 5% FBS.
2. Prepare a cell suspension (detach adherent cells, break up
clumps by pipetting or using other methods) (see Note 5);
sion staining. Seed 5 × 105 cells in 4.5 mL cell culture medium
supplemented with maximally 5% FBS in a 25 cm2 flask. Add
5 mL medium containing MycoZap reagent 1 prepared in
step 1 and incubate the cells until the culture reaches a medium
density, but at least for 2 days.
3. Remove the complete cell culture medium in the flask contain-
ing adherent cells or after centrifugation of suspension cells. If
applicable, dilute the cell cultures to the medium cell density.
Add 9.5 mL fresh medium (containing the FBS concentration
appropriate for the cell culture) and 0.5 mL of MycoZap
reagent 2. Incubate for 2 days.
4. Repeat step 3 another two times (altogether a 6-day treatment)
(see Note 6). Proceed with Subheading 3.3.
3.3. Culture and 1. After completion of the treatment, remove the antibiotics by
Testing Post Treatment washing the cells with PBS. Culture the cells in the same
manner (enriched medium, higher cell concentration, etc.) as
during the treatment period, but do not add any antibiotics.
Even penicillin and streptomycin should not be added to the
medium. Culture the cells for at least another 2 weeks. Even
if initially the cells appear to be in good health after the treat-
ment, the cells might go into a crisis after the treatment,
especially following treatment with BM-Cyclin. The reason
for this posttreatment crisis is not clear, but it might be a
result of reduced activity of the mitochondria. Thus, the cell
status should be frequently examined under the inverted
microscope.
2. After passaging, test the cultures for mycoplasma contamina-
tion. If the cells are clean, freeze and store aliquots in liquid
nitrogen. The cells in active culture have to be retested peri-
odically to ensure continued freedom from mycoplasma con-
tamination (see Note 7).
3. After complete decontamination, expand the cells and freeze

master stocks of the mycoplasma-free cell line and store them
in liquid nitrogen to provide a continuous supply of clean
cells. Discard the ampoules of mycoplasma-infected cells
(see Fig. 1).
4. Notes
1. Store the antibiotics at the recommended concentrations, tem-

peratures, and usually in the dark, and do not use them after
the expiration date. Upon formation of precipitates, completely
dissolve the crystals at room temperature in the dark before
use. As the antibiotics are light sensitive, protect both the stock
and working solutions from light.
2. Storage in liquid nitrogen might be a source of cell culture
contamination with mycoplasmas. Indeed, mycoplasmas were
shown to survive in liquid nitrogen even without cryopreser-
vation. Once introduced into the nitrogen, mycoplasmas may
persist in the tank for an indefinite time, not proliferating, but
are able to contaminate cell cultures in leaky ampoules stored
in the liquid phase of the nitrogen. Thus, storing the ampoules
in the gaseous phase of the nitrogen is recommended to pre-
vent contamination. Additionally, contaminated cell cultures
and those of unknown status should be stored separately from
noninfected cells, preferably in separate tanks to prevent a
mix-up of contaminated and mycoplasma-free cell cultures. If
this is not possible, store the ampoules at different locations
within a tank.
3. Some cell lines are sensitive to a complete exchange of the
medium. If the medium can only be exchanged partially, 50%
of the antibiotic concentration should be added to the remain-
ing conditioned medium that already contains the antibiotic,
whereas 100% of the antibiotic concentration is added to the
fresh medium.
4. It is advantageous to employ two types of treatments
(BM-Cyclin and one of the quinolones or Plasmocin or
MycoZap) in parallel, as usually at least one of the treatments
is successful. In the rare event of resistance, cells of the
untreated frozen backup aliquots can be thawed and treated
again with another antibiotic. As MRA, ciprofloxacin, and
enrofloxacin all belong to the group of quinolones, it is likely
that the use of an alternative compound from the same group
will produce the same end result (cure, resistance, or culture

death). In the case of loss of the culture during or after the treat-
ment, aliquots can be treated with quinolones, as these are
usually better tolerated by the eukaryotic cells. We recom-
mend using MycoZap which shows almost no effect on the
growth parameters during the treatment procedure. This
treatment or the use of MRA is also recommended when the
cells are already in very bad condition prior to treatment and
the number of available cells would suffice only for a single
type of treatment. Sometimes, the cells recover rapidly after
starting the treatment due to the immediate reduction of the
mycoplasmas.
5. Adherent cells are detached by methods appropriate for the
cell line being treated. It is important to break up all clumps
and clusters and to detach cells from the surface of the culture
vessels. Although the antibiotics are in solution and should be
accessible to all parts of the cells, the membranes might be bar-
riers that cannot be passed by the antibiotics. Mycoplasmas
trapped within clumps of eukaryotic cells or even in cavities
formed by the cell membrane of a single cell might be pro-
tected from the antibiotic. This is also the reason for the advice
to keep the concentration of the antibiotic constantly high by
frequently exchanging the medium. Some mycoplasma species
were shown to penetrate the eukaryotic cells, which may be a
source of resistance if the eukaryotic cell membrane is a barrier
for the antibiotics. On the other hand, it was shown that
specific antibiotics (e.g., ciprofloxacin) accumulate in the
eukaryotic cells so that the concentration is higher inside the
cells than in the extracellular environment.
6. Depending on the growth rate of the cell line, which might be
severely altered by the antibiotic, the cell density should be
reduced, kept constant, or even increased. If no data are avail-
able at all for a given cell culture or if the cell culture is in very
poor condition, the cell density, growth rate, and viability
should be recorded frequently and adjusted to improve the
condition of the culture.
7. Applying the overly sensitive PCR for the detection of myco-
plasma, we found that the treated cell cultures might show a
weak false positive signal even after 2 weeks of posttreatment
passaging. This is not necessarily the result of a resistance of
the mycoplasmas and their regrowth, but might be caused by
residual DNA in the culture medium. These cell cultures
should not be discarded after being found positive, but should
be retested after further culturing (see Chapter 1).
References
1. Barile MF, Rottem S (1993) Mycoplasmas in 4. Schmidt J, Erfle V (1984) Elimination of

cell culture. In: Kahane I, Adoni A (eds) Rapid mycoplasmas from cell cultures and establishment
diagnosis of mycoplasmas. Plenum, New York of mycoplasma-free cell lines. Exp Cell Res 152:
2. Uphoff CC, Drexler HG (2010) Mycoplasma 565–570
contamination of cell cultures. In: Flickinger 5. Uphoff CC, Drexler HG (2002) Comparative
M (ed) The encyclopedia of industrial bio- antibiotic eradication of mycoplasma infections
technology, vol 5. Wiley, New York from continuous cell lines. In Vitro Cell Dev
3. Uphoff CC, Drexler HG (2001) Prevention of Biol Animal 38:86–89
mycoplasma contamination in leukemia-lym-
phoma cell lines. Hum Cell 14:244–247
Chapter 3
STR DNA Typing of Human Cell Lines: Detection of Intra-

and Interspecies Cross-Contamination
Wilhelm G. Dirks and Hans G. Drexler
Abstract
Inter- and intraspecies cross-contaminations (CCs) of human and animal cells represent a chronic problem
in cell cultures leading to false data. Microsatellite loci in the human genome harboring short tandem
repeat (STR) DNA markers allow individualization of cell lines at the DNA level. Thus, fluorescence poly-
merase chain reaction amplification of STR loci D5S818, D13S317, D7S820, D16S539, vWA, TH01,
TPOX, CSF1PO, and Amelogenin for gender determination is the gold standard for authentication of
human cell lines and represents an international reference technique. The major cell banks of the USA,
Germany, and Japan (ATCC, DSMZ, JCRB, and RIKEN, respectively) have built compatible STR data-
bases to ensure the availability of STR reference profiles. Upon determination of an STR profile of a
human cell line, the suspected identity can be proven by online verification of customer-made STR data
sets on the homepage of the DSMZ institute. Furthermore, an additional tetraplex PCR has been estab-
lished to detect mitochondrial DNA sequences of rodent cells within a human cell culture population.
Since authentic cell lines are the main prerequisite for rational research and biotechnology, the next sec-
tions describe a rapid and reliable method available to students, technicians, and scientists for certifying
identity and purity of human cell lines of interest.
Key words: Authentication, Cross-contamination, DNA STR typing, Human cell lines, mtDNA typ-
ing, Misidentification, Quality control
1. Introduction
Most facilities culturing cells use multiple cell lines simultaneously.

Because of the complexity of experimental designs today and the
increasing use of cell lines in science and biotechnology, the pos-
sibility of inadvertent mixing of cell lines during the course of day-
to-day cell culture is always present. Cross-contaminations (CCs)
27
28 W.G. Dirks and H.G. Drexler
may be intraspecies, when two genetically different cell lines of

human origin are cultured in one population, or interspecies, when
animal cells invade human cell cultures. Interspecies and intraspe-
cies CCs represent a widespread problem and alerts on misidentified
or cross-contaminated cell lines are most often simply ignored by
the scientific community (1).
Several factors contribute to the increasing problem of cross-
contaminated cell cultures worldwide. The high incidence (~15%)
of CC among cell lines obtained directly from original investiga-
tors or from secondary sources implies that the majority of false cell
lines arise in the originators’ own laboratories (2). In addition to
failure to correctly identify cells during the establishment of new
cell lines, neglect of quality control procedures and disregard for
good cell culture practices are the key causes of CC in new cell
lines. In addition, less attendance to rules at the bench side is
another driving force behind the widespread CC since many
supervisors are not meeting their obligations and allow Ph.D. stu-
dents and technicians who are only briefly and poorly introduced
to adequate cell culture techniques to handle multiple cell lines.
Furthermore, information on cell lines exchanged between
scientific collaborators is normally simply assumed to be cor-
rect. A plethora of reports unmasking bogus cancer cell lines,
including members of the NCI-60 panel used to generate refer-
ence baseline transcriptional drug responses, has triggered calls for
remedial action (3, 4).
While intraspecies CCs of human cells have been widely
reported (5, 6), the problem of contaminating animal cells within
a human culture has been neglected. The scientific authentication
service of DSMZ demonstrates that both types of CCs are perva-
sive. Evaluation of the identities of over 500 cell lines from external
customers shows that the incidence of intraspecies CC is ~10%
while the incidence of interspecies CC is ~6%. These results high-
light the need for a standard authentication procedure for cell line
identity and purity testing.
Researchers using transgenic animal technology utilize vari-
ous rodent cell lines. The most relevant cell lines for the biotech-
nology industry, BHK-21 and CHO, were derived from Syrian
and Chinese hamsters, respectively. To detect contamination of
human cell cultures by a broad spectrum of rodent lines, we
developed a PCR-based method for detecting mitochondrial
DNA sequences (mtDNA) from mouse, rat, and Syrian and
Chinese hamster cells. A test to both generate an STR profile of
human cells and detect animal cells represents a major and novel
advance in detecting CCs and thereby decreasing the use of false
cell lines.
3 STR DNA Typing of Human Cell Lines… 29
2. Materials
All solutions should be prepared in water that has a resistivity of

18.2 MΩ-cm and total organic content of less than five parts per
billion. This standard is referred to as “distilled water” in this text.
2.1. Preparation 1. Phosphate-buffered saline (PBS): 140 mM NaCl, 27 mM KCl,

of High-Molecular- 7.2 mM Na2HPO4 × 12 H2O, 14.7 mM KH2PO4, pH 7.2,
Weight DNA autoclaved.
2. Absolute isopropanol.
3. Absolute ethanol.
4. TE 10/1: 10 mM Tris, 1 mM EDTA, pH 8.0, prewarmed to
50°C.
5. High Pure PCR template preparation kit (Roche).
6. Water bath prewarmed to 72°C.
7. Standard tabletop microcentrifuge capable of 13,000 × g cen-
trifugal force.
8. Standard spectral photometer for determination of DNA
concentration.
2.2. Multiplex STR Fluorescent labeling of the PCR primers permits the multiplexing
Typing of STR loci, even when alleles fall into the same size range, by
labeling the overlapping loci with differently colored fluorescent
dyes. Group I primers are labeled with one specific dye, which
should be different from the dye used for group II primers or for
the size standard. The choice of specific dyes depends on the capil-
lary electrophoresis unit available. See Tables 1 and 2 for a list of
primers. Only one primer of a pair of an STR locus should be
labeled at the 5¢-end, regardless if it is the forward or reverse primer
(see Note 1).
1. Thermal cycler (any supplier).
2. Standard tabletop microcentrifuge or centrifuge capable of
spinning 96-well plates.
3. 0.2 mL reaction tubes.
4. Fluorescently labeled primer pairs for human DNA typing (see
Table 1 for a list of primer sequences) or for animal mtDNA
detection (see Table 2 for a list of primer sequences) concen-
trated at 100 mM in TE (10/1) as a stock solution and stored
at −20°C.
5. Working solutions of labeled primers, aliquoted at 10 mM in
small amounts (~ 25–50 mL aliquots) and stored frozen at
−20°C (see Note 2).
Table 1
STR primer sequences for human DNA typing
STR primer group I (dye I) for human DNA typing
STR Locus Sequence

D16S539 16q22-24 Forward: 5¢-GGG GGT CTA AGA GCT TGT AAA AAG;
Reverse: 5¢-GTT TGT GTG TGC ATC TGT AAG CAT GTA TC
D13S317 13q22-q31 Forward: 5¢-ACA GAA GTC TGG GAT GTG GAG GA;
Reverse: 5¢-GGC AGC CCA AAA AGA CAG A
D5S818 5q21-q31 Forward: 5¢-GGT GAT TTT CCT CTT TGG TAT CC;
Reverse: 5¢-AGC CAC AGT TTA CAA CAT TTG TAT CT
D7S820 7q11.21-22 Forward: 5¢-ATG TTG GTC AGG CTG ACT ATG;
Reverse: 5¢-GAT TCC ACA TTT ATC CTC ATT GAC
STR primer group II (dye II) for human DNA typing
STR Locus Sequence

CSF1 5q33.3-34 Forward: 5¢-AAC CTG AGT CTG CCA AGG ACT AGC;
Reverse: 5¢-TTC CAC ACA CCA CTG GCC ATC TTC
TPOX 2p23-2pter Forward: 5¢-ACT GGC ACA GAA CAG GCA CTT AGG;
Reverse: 5¢-GGA GGA ACT GGG AAC CAC ACA GGT TA
TH01 11p15-15.5 Forward: 5¢-ATT CAA AGG GTA TCT GGG CTC TGG;
Reverse: 5¢-GTG GGC TGA AAA GCT CCC GAT TAT
vWA 12p12-pter Forward: 5¢-CTA GTG GAT GAT AAG AAT AAT CAG TAT GTG;
Reverse: 5¢-GGA CAG ATG ATA AAT ACA TAG GAT GGA TGG
Amel Xp22.1-22.3 Forward: 5¢-ACC TCA TCC TGGG CAC CCT GGT T;
Yp11.2 Reverse: 5¢-AGG CTT GAG GCC AAC CAT CAG
Alternate primer sequences for amplification of human STRs, as well as information on the PCR products and allele
sizes, are available at http://www.cstl.nist.gov/div831/strbase
Table 2
STR primer sequences for animal mtDNA detection
STR primer group III (dye III) for animal mtDNA detection
Species PCR fragment Primer sequences

Mouse 300 bp mtDNA Forward: 5¢-AGG ATT CCC AAT CGT CGT AGC
Reverse: 5¢-TGT AAT TAC GGC TCC AGC TCA
Rat 500 bp mtDNA Forward: 5¢-CAA TCC ACC AAG CAC AAG TG
Reverse: 5¢-CCCCAACCGAAATTTGGTAGTTC
C. Hamster 605 bp mtDNA Forward: 5¢-CCG GCG TAA AAC GTG TTA TAG ACT
Reverse: 5¢-GTA TTA GGT ATA ATA TCG GCA GTC
S. Hamster 245 bp mtDNA Forward: 5¢-GAC CTC TTA GGT GTA TTC CTA C
Reverse: 5¢-GTA TTA GGT ATA ATA TCG GCA GTC
Information on the PCR products and allele sizes is available at http://www.cstl.nist.gov/div831/strbase
2.2.1. Hot Start Nonaplex In addition to the equipment and reagents listed in Subheading 2.2,
PCR for Human DNA Typing the following reagents are required:
1. Master mix consisting of 25 mL per reaction for each sample,
plus one additional reaction for every ten samples. For a single
reaction, the components are as follows:
(a) 10 pmol of STR primer for human DNA typing (1 mL of
a 10 mM working solution).
(b) 2.5 mL 10× Hot start PCR buffer (any supplier).
(c) 1 mL dNTP (5 mM stock solution).
(d) 0.2 mL (1 unit) hot start Taq polymerase (any supplier).
(e) 19.5 mL distilled water.
2. Positive control DNA, e.g., HeLa DNA.
3. Negative control: Distilled water.
4. Test DNA.
2.2.2. Hot Start Tetraplex In addition to the equipment and reagents listed in Subheading 2.2,
PCR for Animal mtDNA the following reagents are required:
Detection
1. Master mix containing 25 mL per reaction for each sample,
plus one additional reaction for every ten samples. For a single
reaction, the components are as follows:
(a) 10 pmol of mtDNA primer for animal mtDNA typing
(1 mL of a 10 mM working solution).
(b) 2.5 mL 10× Hot start PCR buffer (any supplier).
(c) 1 mL dNTP (5 mM stock solution).
(d) 0.2 mL (1 unit) hot start Taq polymerase (any supplier).
(e) 19.5 mL distilled water.
2. Positive control DNA, e.g., rodent DNA mixed with HeLa
DNA template.
3. Negative control DNA, e.g., H2O mixed with HeLa DNA
template.
4. Test DNA.
2.3. Capillary 1. Capillary electrophoresis unit (any supplier).

Electrophoresis 2. Internal size standard kit 400 (Beckman-Coulter).
for DNA Fragment
3. Microtiter plates.
Detection
4. Sample loading solution; 0.3 mM EDTA in deionized
formamide.
5. Test samples (PCR amplification products).
3. Methods
Multiplex PCR targets multiple locations throughout the genome

and is an ideal technique for DNA typing because the probability
of identical alleles in two individuals decreases with an increase in
the number of polymorphic loci examined (7). The ability to
fluorescently label PCR primers permits the multiplexing of STR
loci which may have alleles that fall into the same size range, by
labeling the overlapping loci with differently colored fluorescent
dyes that can then be resolved spectrally.
General rules to avoid DNA carryover contamination should be
strictly followed. DNA extraction should be carried out using
equipment (pipettes, microcentrifuge, etc.) which is independent
from the PCR setup. Optimally, this laboratory is separated from
those rooms where the PCR reaction is set up or the PCR products
are analyzed. All reagents should be stored in small aliquots to
provide a constant source of uncontaminated reagents. New ali-
quot batches should be tested and compared for quality prior to
any use. Re-amplifications should never be carried out. If possible,
PCR reactions should be set up in a designated working station or
a hood capable of irradiating used pipettes, tips, and tubes by UV
light. It is highly recommended that gloves are worn during the
whole procedure.
3.1. Preparation Briefly, cells are lysed during a short incubation time with protei-
of High-Molecular- nase K in the presence of a chaotropic salt (guanidinium-
Weight DNA hydrochloride), which immediately inactivates all nucleases. Nucleic
acids bind selectively to glass fibers prepacked in the filter tube.
Bound genomic DNA is purified in a series of rapid washing and
spinning steps to remove inhibiting cellular components. Finally,
low-salt elution releases the DNA from the glass fiber cushion.
1. Centrifuge cell culture suspensions containing 3–5 × 106 dip-
loid cells in an Eppendorf tube at 1,000 × g for 4 min in a
14 mL tube (see Notes 3 and 4). Remove the supernatant with
a disposable pipette and discard. Carefully suspend the remain-
ing pellet in 5 mL PBS using a pipette. Repeat centrifugation.
2. Suspend the cell pellet in 200 mL PBS by vortexing. Make sure
that even tiny clumps of cells are carefully re-suspended.
3. Pre-warm the water bath to 72°C.
4. Using the commercially available DNA extraction kit from
Roche, add 200 mL of well-mixed solution I (guanidinium-
hydrochloride) to the sample solution and mix by careful
pipetting.
5. Immediately add 40 mL proteinase K, mix well using a vortex,
and incubate at 72°C for at least 10 min.
6. Add 100 mL of isopropanol to the sample, mix well, and apply

the whole mixture to a filter tube. Centrifuge for 1 min at
6,000 × g.
7. Discard the flow through, add 500 mL of inhibitor removal
buffer, and centrifuge again for 1 min at 6,000 × g.
8. Discard the flow through, add 500 mL of wash buffer, and
centrifuge again for 1 min at 6,000 × g.
9. Repeat step 8.
10. Add 200 mL of elution buffer preheated to 72°C and centri-
fuge for 1 min at 6,000 × g. For maximum yield the elution
step should be repeated using 100 mL elution buffer. Resuspend
pellet in TE 10/1 (10 mM Tris, 1 mM EDTA, pH 8.0).
11. Using a standard spectral photometer, adjust the concentration of
genomic DNA to 10 ng/mL per sample for a diploid cell line.
12. Store genomic DNA at 4°C (see Note 5).
3.2. Hot Start Nonaplex In the first step, a multiplexed STR PCR amplification of eight
PCR of Genomic DNA prominent and highly polymorphic STR loci and one additional
and Tetraplex PCR of locus for gender determination will be carried out. The STRs of the
Mitochondrial DNA loci D5S818, D13S317, D7S820, D16S539, vWA, TH01, TPOX,
and CSF1PO consist of exclusive tetrameric repeats, which are inher-
ited in a Mendelian way. The combination of 8 STRs increases the
exclusion rate sufficiently to allow the discrimination of one human
cell line from another at the level of 108. Amelogenin (AMEL) is the
most suitable gene for gender determination of samples of human
origin (8). Using specific primers in PCR applications, the sequence
of the X-chromosomal version (AMELX, Xp22.1-Xp22.3) yields a
209 bp amplicon, while the Y-chromosomal gene (AMELY, Yp11.2)
yields a 215 bp DNA fragment, which are easily separated by differ-
ent electrophoretical techniques (9). Hence, samples from male
sources will show two bands (209/215 bp), while female-derived
cell lines will show only one band (209 bp, see Fig. 1). In a second
step, an independent tetraplex PCR for detection of mtDNA
sequences derived from rodent cell lines is carried out.
3.2.1. Hot Start Nonaplex The multiplex PCR described here identifies eight different STR
PCR for Human DNA Typing loci and, in addition, determines gender. The amplification proce-
dure and the parameters are optimized for 0.2 mL reaction tubes
in an i-Cycler thermal cycler (Bio-Rad). It is essential to incorpo-
rate the appropriate positive and negative controls (e.g., HeLa
DNA template and H2O, respectively).
When using commercial multiplex STR kits, the specific manu-
als of Promega Corporation and Applied Biosystems should be
strictly followed. If not using commercial kits, it is important to
optimize the general amplification parameters. A “hot start” PCR
should be always performed in order to activate the Taq DNA
Fig. 1. Electropherogram of an interspecies CC of cervical carcinoma line HELA. CC of HELA by mouse, rat, and hamster
cells. 10 ng of genomic DNA from the cell line HELA (DSMZ ACC 057) was analyzed using the nonaplex PCR for human STR
fragment amplification (green and black peaks ) and the tetraplex mtDNA amplification of the rodent lines of mouse, rat,
and Chinese/Syrian hamster as indicated.
Polymerase and to prevent the formation of nonspecific amplification

products. The number of cycles depends on the amount of DNA,
so with the protocol as outlined 30 cycles are recommended for all
samples.
1. Prepare a master mix (see Subheading 2.2.1 for components)
containing 25 mL per reaction for each sample, plus one addi-
tional reaction for every ten samples. Include enough for both
a positive and a negative control (water only).
2. Aliquot 25 mL of the master mix into reaction tubes or 96-well
reaction plates, one for each sample.
3. Adjust genomic DNA to a concentration between 0.2 and
1 ng/mL.
4. Add 1 mL of genomic DNA to each tube or well.
5. Centrifuge the tubes/plates for 4 min at 600 × g.
6. Program the thermal cycler as follows:
Cycle 1 (×1): 94°C for 3 min

Cycle 2 (×10): 94°C for 30 s, 60°C for 15 s, 70°C for 45 s
Cycle 5: 4°C end
3.2.2. Hot Start Tetraplex Detection of animal sequences is carried out using a separate
PCR for Animal mtDNA tetraplex PCR reaction using specific primers for mtDNA sequences
Detection of mouse, rat, and Syrian and Chinese hamster. The main advan-
tage of using mitochondrial versus genomic DNA sequences is the
presence of high amounts of mitochondrial genomes (up to 104 in
liver) compared to a diploid nuclear genome. While equimolar
ratios of DNA sequences allow a detection limit of ~10%, the pres-
ence of a single rodent cell can be detected in 1 out of 104 to 105
human cells. It is essential to incorporate the appropriate positive
and negative controls (e.g., each rodent DNA mixed with HeLa
DNA template and H2O mixed with HeLa DNA template,
respectively).
1. Prepare a master mix (see Subheading 2.2.2 for components)
containing 25 mL per reaction for each sample, plus one addi-
tional reaction for every ten samples. Include enough for both
a positive and a negative control.
2. Aliquot 25 mL of the master mix into reaction tubes or 96-well
reaction plates, one for each sample.
3. Adjust genomic DNA to a concentration between 0.2 and
1 ng/mL.
4. Add 1 mL of genomic DNA to each tube or well.
5. Centrifuge the tubes/plates for 4 min at 600 × g.
6. Program the thermal cycler as follows:

Cycle 4: 4°C end
3.3. Capillary Since DNA possesses a constant mass-to-charge ratio, some form
Electrophoresis of separation matrix is needed to separate different sizes of DNA
for DNA Fragment fragments by their molecular weight. In traditional gel electropho-
Detection and Allelic resis, the requirement for a sieving matrix is met with polyacrylam-
STR Lists ide or agarose gels. The movement of larger DNA fragments is
impeded relative to that of the smaller DNA fragments as the
molecules migrate through the gel under the influence of an elec-
tric field. Polyacrylamide gels are no longer the only slab gel
systems available for resolving STR alleles. A recent publication
demonstrated that small agarose gels have sufficient resolving
power to type tetranucleotide repeats. Even dinucleotide repeats
could be resolved with MetaPhor agarose and detected with SYBR
Green staining (9).
Various automated fluorescence detection systems have been
used for separation, detection, and typing of STR alleles. Full auto-
mation of the electrophoresis process with no need to pour the gel

or manually pipet the samples onto the gel is one of the big advan-
tages of capillary electrophoresis (CE). With the higher sensitivity
of laser-induced fluorescence, sample preparation is no longer nec-
essary. Samples are diluted in water or formamide and can be easily
detected. Separation of STR alleles may be performed in a matter
of minutes rather than hours.
1. Aliquot 1 mL of each PCR amplification product into a well in
a microtiter plate.
2. Add 0.25 mL of the internal size standard.
3. Add 29 mL of sample loading solution for a total volume of
30 mL.
4. Run the Capillary Electrophoresis as per the manufacturer’s
instructions.
Using Capillary Electrophoresis, samples are automatically loaded
and analyzed using fragment analysis parameters, which are pro-
vided by the supplier of the Capillary Electrophoresis unit (Applied
Biosystems, Beckman-Coulter). In general, fragment analysis
software of any supplier enables the precise determination of
detected alleles resulting in an allelic genotype list. Allelic num-
bers of STR DNA fragments may be precisely determined by
other techniques or by allelic ladders of the ABI system.
5. Once the sizes are known, determine the respective allele num-
bers from Table 3 (see Note 6).
3.4. Online Evaluation The main Biological Resource Centers (BRCs) ATCC, DSMZ,
of Suspicious STR JCRB, and RIKEN have generated large databases of STR cell line
Profiles profiles for identity control. In cooperation with the Japanese
BRCs, DSMZ has piloted the generation of the most comprehen-
sive international reference database which is linked to a simple
search engine for interrogating STR cell line profiles (10). A simple
search engine for interrogating STR cell line profiles has been
made available on the Website of DSMZ. Once the problem of
false negatives due to discrepant representation of single STR
alleles—e.g., by losses of heterozygosity and bottlenecking
selection—has been tackled and unambiguous search results are
produced, human cell lines should be consistent with consensus
STR reference data sets. STR profiles of all human cell lines distrib-
uted by DSMZ, JCRB, and RIKEN and one-third of the cell lines
distributed by ATCC are now publicly accessible at http://www.
dsmz.de/services/services-human-and-animal-cell-lines/online-str-
analysis.html using an interactive database where match-criteria
have been arbitrarily set to 60%. Registered users simply login at
the online-STR-analysis-site on the DSMZ homepage and are
guided through the procedure. Aided by simple prompts, users can
input their own cell line STR data to retrieve best matches with
Table 3
Allele organization and sizes of amplified human STR loci
Allele D5S818 D13S317 D7S820 D16S539 vWA TH01 TPOX CSF1PO Amelogenin
3 169
4 173 209 = X
5 164 212 266 177 220 287 215 = Y
6 114 168 216 270 181 224 291
7 118 172 220 274 185 228 295
8 122 176 224 278 189 232 299
9 126 180 228 282 193 236 303
9.3 196
10 130 184 232 286 118 197 240 307
11 134 188 236 290 122 201 244 311
12 138 192 240 294 126 205 248 315
13 142 196 244 298 130 252 319
14 146 200 248 302 134 256 323
15 150 204 252 306 138 327
16 154 142 331
17 158 146
18 150
19 154
20 158
21 162
22 166
23 170
Nucleotide range and the number of known alleles of each STR loci are summarized. Further information is available at
http://www.cstl.nist.gov/div831/strbase. Regular fragment sizes in base pairs of alleles are printed in bold, variant
alleles are printed in italics
authenticated cell lines listed on the database. Inevitably, reference

profiles remain subject to revision until all commonly held cell lines
have been STR typed across participating repositories. At present,
about 2,342 such cell lines have been STR typed and are repre-
sented as reference sets on the database. Armed with this tool,
online verification of cell line identity should prove a vital weapon
to combat the havoc of cell line cross-contamination which has
dogged cancer research since inception.
4. Notes
1. If an individual constellation of genomic loci is of interest, the

dye-labeled primer pairs should be tested in single reactions
and the primer mixture adjusted for the generation of similar
peak heights of the measured loci.
2. Working solutions of primers are stable for 1 month at 4°C and
should be kept in lightproof reaction tubes.
3. Prior to isolation of genomic DNA, cell viability of samples
should be analyzed using trypan blue exclusion assay. With the
exception of factor-dependent hematopoietic cell lines, viability
of a cultured cell line should not be below 85% at sampling.
4. To avoid retention of PCR inhibitors in biological samples, it
is very important to carefully and completely suspend the cell
pellet in PBS prior to DNA isolation.
5. Quality control of isolated genomic DNA should be carried
out by using 1% agarose gel electrophoresis. 300 ng of high-
molecular-weight DNA should result in a single band, while
DNA of apoptotic cells would show specific DNA laddering.
6. Degradation of primers will result in nonspecific bands. Two
bands of a diploid cell line should be detected within the sizes
of each genomic locus as presented in Tables 1 and 2.
References
1. Buehring GC, Eby EA, Eby MJ (2004) Cell line profiling analysis of 40 human thyroid cancer
cross-contamination: how aware are mammalian cell lines reveals cross-contamination resulting
cell culturists of the problem and how to monitor in cell line redundancy and misidentification. J
it? In Vitro Cell Dev Biol Anim 40:211–215 Clin Endocrinol Metab 93:4331–4341
2. MacLeod RAF, Dirks WG, Kaufmann M, 7. Masters JR, Thompson JA, Daly-Burns B, Reid
Matsuo Y, Milch H, Drexler HG (1999) YA, Dirks WG, Packer P, Toji LH, Ohno T,
Widespread intra-species cross-contamination Tanabe H, Arlett CF, Kelland LR, Harrison M,
of human tumor cell line arising at source. Int Virmani A, Ward TH, Ayres KL, Debenham PG
J Cancer 83:555–563 (2001) Short tandem repeat profiling provides
3. Liscovitch M, Ravid D (2007) A case study in an international reference standard for human
misidentification of cancer cell lines: MCF-7/ cell lines. Proc Natl Acad Sci 98:8012–8017
AdrR cells (re-designated NCI/ADR-RES) are 8. Sullivan KM, Mannucci A, Kimpton CP, Gill P
derived from OVCAR-8 human ovarian carci- (1993) A rapid and quantitative DNA sex test:
noma cells. Cancer Lett 245:350–352 fluorescence-based PCR analysis of X-Y homol-
4. Nardone RM (2007) Eradication of cross-con- ogous gene amelogenin. Biotechniques
taminated cell lines: a call for action. Cell Biol 15:636–641
Toxicol 23:367–372 9. White HW, Kusukawa N (1997) Agarose-based
5. Drexler HG, Dirks WG, Matsuo Y, MacLeod system for separation of short tandem repeat
RAF (2003) False leukemia-lymphoma cell loci. Biotechniques 22:976–980
lines: an update on over 500 cell lines. Leukemia 10. Dirks WG, MacLeod RAF, Nakamura Y,
17:416–426 Kohara A, Reid Y, Milch H, Drexler HG,
6. Schweppe RE, Klopper JP, Korch C, Mizusawa H (2010) Cell line cross-contamina-
Pugazhenthi U, Benezra M, Knauf JA, Fagin tion initiative: an interactive reference database
JA, Marlow LA, Copland JA, Smallridge RC, of STR profiles covering common cancer cell
Haugen BR (2008) Deoxyribonucleic acid lines. Int J Cancer 126:302–304
Chapter 4
Classical and Molecular Cytogenetic Analysis

Roderick A.F. MacLeod and Hans G. Drexler
Abstract
Cytogenetic analysis is performed on cell cultures for several reasons, notably, to perform identity checks
by verifying species of origin or the retention of key chromosome rearrangements in cell lines described
previously. De novo chromosome analysis is usually performed when characterizing cancer cell lines for the
presence of neoplastic rearrangements associated with specific tumors. This usually involves fluorescence in
situ hybridization (FISH) using clones covering gene loci near recurrent chromosome breakpoints.
Chromosome breakage is an important endpoint in radiation biology and mutagenesis, enabling cell lines
to be used for measuring genotoxic dosage and repair. Finally, cytogenetic analysis may be performed to
monitor stability in culture. Unlike most preparative techniques, chromosome preparation resists standard-
ization. Hence, procedures must be optimized for each cell line. Thus, evidence-based protocols are
described for hypotonic harvesting, rapid G-banding, FISH, and Spectral Karyotyping (SKY) analysis of
cell cultures to allow troubleshooting and fine-tuning to suit the requirements of individual cell lines.
Key words: Chromosomes, Cytogenetics, FISH, G-banding, Hypotonic treatment, SKY
1. Introduction
1.1. Background: Cell lines are karyotyped for various reasons. In human cancer cell
Why Perform lines, arguably the most important single cell line resource, cytoge-
Cytogenetic Analysis? netics is used to investigate chromosome rearrangements which
may betray oncogenomic changes targeting specific cancer genes.
We now know that such alterations switch cancer genes on or off
inappropriately, or even fuse coding regions to create new proteins.
Because oncogenomic alterations tend to occur nonrandomly—
many associated with specific tumors (1)—chromosome analysis
informs tumor cell line classification side-by-side with transcrip-
tional profiling (2). Bearing in mind that cell lines are often estab-
lished from mixed samples or taken from samples remote from
39
40 R.A.F. MacLeod and H.G. Drexler
original tumors, e.g., ascites or metastases, oncogenomics forms a

key step upon which adequate cell line characterization critically
depends. Oncogenomic characterization highlights which cell lines
are eligible to model specific entities, e.g., for use in targeted drug
development. Accordingly, the tyrosine kinase inhibitor Imatinib
(3) used to treat both leukemias and solid tumors was developed
using cell lines (sourced from this institute) established from
patients with chronic myeloid leukemia (CML). Cell lines estab-
lished from CML patients with t(9;22)(q34;q11) causing fusion of
BCR with ABL—the primary oncogenomic change in CML—
invariably retain this rearrangement in vitro (4). The usefulness of
karyotype analysis for characterizing cancer cell lines is not limited
to leukemia, however (5); both mesenchymal and epithelial (1) as
well as neuronal (6) tumors also carry recurrent chromosome
translocations.
Cytogenetic analysis, like other microscopic methods, involves
observations performed at the single-cell level, encouraging detec-
tion of differences between, as well as within, cells. It is sometimes
opined that “cell lines are genetically unstable” (whatever that means);
cytogenetics, unlike biochemical methods, permits detection of dis-
tinct subclones and the monitoring of stability therein. In the authors’
experience, with the exception of symmetrical doublings in their
modal chromosome numbers, cell lines appear to be rather more
stable than is commonly supposed (6–9). Indeed the most intense
phase of chromosomal rearrangement occurs in vivo, namely, physi-
ological receptor gene rearrangement in lymphocytes (10, 11).
Cells respond to ionizing radiation (or analogous treatments
with radiomimetic drugs) by undergoing chromosome breakage in
a dose-related fashion. If examined soon after insult, i.e., prior to
apoptotic loss, DNA repair, or conversion into other types of rear-
rangement, enumeration of chromosome breakage allows cells to
be used as biological dosimeters.
Chromosome breakage may also occur spontaneously, notably
in those predisposed to cancer. Chromosome fragility occurs non-
randomly, clustered at so-called “fragile sites,” which may be over-
expressed in those bearing certain heritable diseases. To visualize
these sites it is often necessary to culture primary cells in specific
conditions conducive to their expression—procedures facilitated in
cell lines. Our own data suggest a connection between chromo-
some instability and loci bearing micro-RNA genes (9).
A key role for cytogenetic data is in the fight against false cell
lines. On karyotyping, we found that about one-in-six supposedly
new cancer cell lines as supplied by their originators were nothing of
the kind. The first clue was that their karyotypes bore uncanny
resemblances to older “classic” cell lines, which tend to be those
most widely circulated. Cell line cross-contamination (CL-CC) was
confirmed by DNA profiling (12, 13) as described in another chap-
ter in this volume. We now know that spurious cell lines are mainly
4 Classical and Molecular Cytogenetic Analysis 41
generated by poor technique, e.g., use of shared reagent bottles or

mislabeling of culture flasks. CL-CC, first publicized over 30 years
ago (14) but neglected of late (15), poses an insidious threat, which
only now with the advent of forensic DNA profiling may be starting
to abate (16). Unfortunately, vanishingly few first publications of
new cell lines include DNA profiling data; and those with matching
patient profiles even less. In the case of cancer cell lines, however,
many original publications include karyotypes showing chromosome
rearrangements which are to all intents unique. To those equipped
with cytogenetic facilities, descriptive karyotypes provide useful
identification data for retrospective matching. In addition, “quick
and dirty” cytogenetic testing provides a rapid way to check the
species-of-origin most commonly encountered among cell lines,
such as mouse, rat, and hamster, while experienced operators may be
able to identify more exotic species. To guide those unsure about the
status of cell lines they are using (how many students or postdocs
have wasted formative years working on false cell lines?), we and oth-
ers have just compiled a list of false cell lines together with their true
identities (17).
1.2. Cytogenetic Cytogenetics owes its key position in the cell culturist’s armamen-
Methodology tarium to a happy chain of technical and informational advances. In
the early 1970s it first became possible to distinguish each of the
24 different human chromosomes (referred to as numbers 1–22,
X, and Y) by using a variety of staining procedures to reveal their
unique patterns of latent striation, termed “chromosome band-
ing.” The original standard banding methods were Q(uinacrine)-
banding (18), and G(iemsa)-banding (19). A further modification,
trypsin G-banding (20), has gained the widest currency since its
introduction in 1973 because of its speed and robustness. Banding
techniques were instrumental in the identification of the
“Philadelphia chromosome” (Ph) marker and its origin via a recip-
rocal translocation, t(9;22)(q34;q11) (21), a mechanism not
guessed when the Ph marker was first observed as an insignificant
dot-like chromosome present in unbanded bone marrow chromo-
some preparations of CML patients more than a decade earlier
(22). This observation ushered in the realization that cancer is
caused by somatic gene alterations.
While G-banding enabled chromosomes to be readily identified
in normal cells, the often complex rearrangements (principally
translocations, deletions, amplifications and inversions) in cancer
cells often thwarted their analysis by traditional photographic
methods. The development of computer-based methods in the
1990s, which allowed real-time analysis on-screen, brought about
improvements in speed, sensitivity and accuracy. Image analysis
enables evaluation of complex tumor karyotypes. Nevertheless, full
resolution of tumor karyotypes was still hampered by complex
marker chromosomes yielding abnormal banding patterns.
A major technical advance was the advent of fluorescence in

situ hybridization (FISH) during the 1990s (23, 24). FISH exploits
the stability and specificity of DNA–DNA hybrids formed after
exposure of chromosomes to homologous DNA under renaturat-
ing conditions. FISH required non-isotopically labeled deoxynu-
cleotides and a straightforward method for their efficient
incorporation into DNA by nick translation, which prompted the
commercial development of chromosome library (“painting”)
probes specific for each of the 24 different human chromosomes.
Pairwise /threeway combinations of painting probes combined
with counterstaining the remaining chromosomes may be used to
resolve chromosome translocations.
By definition, however, the components of complex “marker”
karyotypes are unknown, demanding extra time and resources.
These can be tackled by using multicolor probe mixtures enabling
each chromosome to be distinguished (reviewed in ref. 25). Analysis
requires short pass chromatic visualization systems, either filter-
based m(ultiplex)-FISH or spectrophotometric-based Spectral
Karyotyping (SKY). Accurately mapped and sequenced bacterial
artificial chromosome (BAC), fosmid, cosmid or plasmid clones
allow suitably equipped investigators to locate chromosome break-
points in cancer cell lines with a level of precision of ca. 20 Kbp or
better, thereby highlighting potentially relevant genomic features,
such as cancer genes or regulatory regions. In this way, FISH bridges
the gap between classical cytogenetics and molecular biology.
In this chapter, we describe basic cytogenetic procedures that
have been adapted in our laboratory for use with cell cultures. For
those planning de novo cytogenetic analysis of tumor cell lines, it
is convenient to split the task into the following steps: harvesting
(see Subheadings 2.1 and 3.1), G-banding (see Subheadings 2.2,
2.3, and 3.2), and FISH (see Subheadings 2.3, 2.4, and 3.3).
2. Materials
Unless otherwise indicated, reagents may be stored up to 4 weeks

at 4°C.
2.1. Harvesting 1. Cell culture(s) maintained in logarithmic phase.

2. N-Deacetyl-N-methylcolchicine (colcemid) 100× solution
(Invitrogen): 4 mg/mL stock solution; store refrigerated for
up to 1 year.
3. FUDR/uridine100× stock solution. Mix 1 part 5-fluoro-2¢-
deoxyuridine (FUDR) (Sigma) (25 mg/mL) and 3 parts
1-b-d-Ribofuranosyluracil (uridine) (Sigma; 1 mg/mL); store
refrigerated for up to 1 year.
4. Thymidine 100× stock solution: 1-(2-deoxy-b-d-ribofuranosyl)-

5-methyluracil (thymidine) (Sigma). Dissolve 50 mg in 100 mL
autoclaved TE buffer (10 mM Tris–HCl pH 7.5; 1 mM
EDTA). Filter-sterilize through 0.22-mm filter.
5. Trypsin 0.5 g/L–EDTA 0.2 g/L (Invitrogen) for removal and
dispersal of adherent cells; store at (−20°C) for up to
6 months.
6. Stock hypotonic solutions: KCl 5.59 g/L; or Na–citrate
9.0 g/L. Working hypotonic solutions: mix KCl and Na–citrate
(e.g., 20:1, 10:1, 1:1, 1:10, 1:20, etc.) shortly before use,
allowing time to reach desired temperature.
7. Fixative. Mix absolute methanol and glacial acetic acid at 3:1.
Use fresh but can be stored up to 4 h at 4°C.
2.2. G-Banding Only 1. Slides (frosted ends for annotation). Wash mechanically
overnight in warm ion-free detergent; rinse twice in deionized
water; oven-dry, and leave overnight in ethanol (70%). Slides
should then be polished using a lint-free cloth (or non-shredding
tissue) and stored for several months wrapped in aluminum foil
at (−20°C) until use.
2. Phosphate-buffered saline (PBS): adjusted to pH 6.8 (Trypsin
solution) or pH 7.2 (Giemsa stain).
3. Trypsin stock solution (140×): dissolve 17.5 mg trypsin 1:250
(Difco) in PBS (pH 6.8). Store 500 mL aliquots at (−20°C) for
up to 6 months.
4. Giemsa stain (1.09204.0500 Merck). Dissolve 5 mL in 100 mL
PBS (pH 7.2) and filter before use.
5. Routine microscope with phase-contrast (PC) illuminator and
the following objectives: 10× (phase contrast), 40× (phase
contrast), and 50× (brightfield–dry) for slide evaluation and
preliminary analysis.
2.3. G-Banding 1. Image analysis system for G-banding and FISH (see Note 1).
and FISH 2. Laboratory oven for slide aging (G-banding) or slide drying
(FISH).
3. Coplin jars, 100 mL (glass), for staining and washing.
4. 4× SSC: 35.1 g NaCl, 17.7 g Na–citrate made up to 1 L. Adjust
to pH 7.2.
5. 0.5× SSC, 2× SSC, and so forth: dilute from 4× SSC stock but
monitor pH.
2.4. FISH Only 1. Ethanol: absolute, 90%, 70%. Can be used twice, thereafter
discarded.
2. Pepsin stock solution: Dissolve 250 mg pepsin (Sigma P7012)
in 12.5 mL deionized H2O. Freeze 500 mL aliquots (−20°C)
and store for up to 6 months.
3. Pepsin working solution: Dilute 500 mL stock solution in

100 mL deionized H2O containing 1 mL of 1 N HCl; store at
(−20°C) for up to 6 months.
4. Formaldehyde solution: 1% formaldehyde in PBS (pH 7.2)
containing 50 mM MgCl2.
5. Acetone, for use in mild pretreatment.
6. Hybridization buffer (“hybrisol”): Mix 5 ml deionized forma-
mide (GenomeLab sample loading solution, Beckman-Coulter,
Fullerton, CA), 1 ml 40% dextran sulfate and 4 ml 250 mM
Na2HPO4 in 5× SSC. (In our experience, commercial hybrid-
ization buffers are not entirely reliable because they may con-
tain impure formamides capable of spoiling fluorescent probe
signals.). Store at room temperature (contains formamide).
7. Cold competitor DNA for prehybridization with probes con-
taining repeat sequences: Cot-1 DNA, 1 mg/mL (Roche); store
indefinitely at (−20°C).
8. Nail varnish (clear).
9. Rubber cement.
10. Hybridization chamber: Sealed container with an internal shelf
to separate slides (above) from humidifier (e.g., water-impreg-
nated towels [below]). Lidded stainless steel instrument steril-
ization trays make admirable hybridization chambers, being
readily sterilizable and both rustproof and heat resistant.
11. Hybridization bed: Prewarmed freezer block kept in incubator
at 37°C; use during application of probes to slides.
12. Wash solution: 4× SSC with 0.1% Tween-20, molecular biol-
ogy grade (Sigma). Slides can be dipped into wash solution
between any steps to prevent drying out.
13. Plastic cover slips for probe detection (Qbiogene).
14. Mounting medium: Dissolve 50 ng/mL 4¢, 6-diamidino-2-
phenylindole dihydrochloride (DAPI) in Vectashield antifade
mounting medium (Alexis).
15. Cover slips: glass, grade 0.22 × 60 mm.
16. Chromosome painting probes: Store at (−20°C) unless other-
wise stated (see Note 2).
17. Research microscope with the following objectives with as large
numerical apertures as budgets permit: 10× (phase contrast for
evaluating unstained preparations), 50× Epiplan, brightfield dry
(for evaluating Giemsa-stained preparations), 63× Planapochromat
(oil), or equivalents from other manufacturers. We can specially
recommend the 100× Zeiss Apochromat (with 1.46 numerical
aperture) oil objective; this is equally useful for both brightfield
and FISH work. Ideally, a cytogenetics research microscope
should be equipped with a motorized filter wheel and configured
to an appropriate FISH imaging system (see Note 1).
3. Methods
3.1. Harvesting Mitotic metaphase, the only cell cycle stage when chromosomes are
and Slide Preparation clearly visible, lasts a mere 0.5–1 h in cells in continuous culture.
(see Notes 3 and 4) This limitation severely reduces numbers of cells available for chro-
mosome analysis. Fractions of dividing cells may be enriched by
exposure of growing cultures to colcemid or some other mitotic
blocking agent for a few hours. Culture conditions inimical to loga-
rithmic growth must be avoided by maintaining an adequate supply
of fresh nutrients and growth supplements. It is difficult to over-
state just how crucial initial harvesting and slide preparation are to
success with both G-banding (see Subheading 3.2) and FISH (see
Subheading 3.3). Handbooks almost invariably list standardized
harvesting protocols limited to incubation in 0.075 M KCl at ambi-
ent temperature for 7 min with little discussion of possible options.
While a “one size fits all” approach may well suffice when con-
fronted by the limited range of cell types encountered in genetic
diagnosis, cancer cells and nonhuman cells represent a far wider
range of developmental stages demanding an equally wide range of
hypotonic treatments. In our experience, choice of hypotonic treat-
ment is the main key to successfully harvesting cancer cell lines for
cytogenetic analysis (26). The images in Fig. 1a, b illustrate how
subtle changes in hypotonic treatment (outlined in Table 1)
influence success in G-banding. Hypotonic treatments which con-
sistently yield good preparations with one cell line are often unsuit-
able for another of similar origin. It is therefore necessary to
determine empirically which harvesting procedures are optimal.
This may be achieved by harvesting, in parallel, cell aliquots exposed
to a range of hypotonic conditions (namely, with a variety of differ-
ent buffers and incubation times and, if need be, incubation tem-
peratures, etc.).
Hypotonic treatment is halted by fixation in chilled acid-alco-
hol, a process which does permit standardization. Although some
deterioration is inevitable, fixed cells can be stored several years at
(−20°C) until required. Immediately prior to slide-making, cell
suspensions should be washed in fresh fixative. Slide-making is per-
formed by dropping the cell suspension onto ice-cold, precleaned
slides held at a slight angle (about 5–10% slope) atop a prefrozen
(−20°C) freezer cold block. Two drops aimed at the slide region
immediately under the frosted zone and at the lower middle,
respectively, should result in figure-of-eight spreading patterns
suitable for both G-banding (enabling two timezones) and FISH
(enabling two probe mixtures). Once made, slides can be variously
stored for a few years at (−80°C), for short intervals at room tem-
perature for FISH, or baked overnight at 60°C for G-banding the
following day.
Fig. 1. Classical cytogenetic analysis. Images (a, b): Analysis of primary human endothelial cells showing substandard and
satisfactory metaphases, respectively. (c) G-banded karyotype of cells depicted in (b). Metaphase cells were prepared as
described in Subheading 3.1 using hypotonic treatment specified in Table 1, and slide preparations aged overnight at 60°C
for G-banding performed as described in Subheading 3.2. The consensus karyotype was found to be: 46, XY with no con-
sistent abnormality present. (d, e) G-banding of a cell line, OCI-Ly-19 (DSM ACC 528), established from a patient with dif-
fuse large B-cell lymphoma (DLBCL). Chromosomes present in the metaphase image (d) were assigned using an image
analysis system and the rearrangements resolved and confirmed by FISH to yield the following consensus karyotype:
48(46-52)<2n>X, −X,t(4;8)(q32;q24),+6,+6,del(6)(q15)x2, +8,r(8), t(14;18)(q32;q21), add(18)(q23). The t(14;18) rear-
rangement (arrows) juxtaposes the BCL2 oncogene (at 18q21) with the immunoglobulin heavy chain gene IGH (at 14q32).
Note the presence of a large ring chromosome 8 (arrowhead) which serves to increase copy number of another oncogene
CMYC (at 8q24), and deletions of the long arm of chromosome 6 (twin arrowheads).
Day 1
1. Add colcemid (final concentration 40 ng/mL) to growing
cultures for 2–4 h.
2. As an alternative to colcemid treatment, incubate cells overnight
with FUDR to improve chromosome morphology (see Note 4).
Day 2
3. Suspension cell cultures: aliquot cells (e.g., four times in 10 mL
tubes), centrifuge (5 min at 400 × g), and discard supernatant.
4. Adherent cell cultures: Shake vigorously to remove mitoses
and retain supernatant in centrifuge tube (50 mL). Meanwhile,
rinse remaining adherent cells with serum-free medium or PBS
Table 1
Harvesting record sheet for primary human umbilical venous endothelial cells (HUVEC)
Hypotonic treatment Results Slides & suspensions
Harvest Colcemid Na 60°C −80°C −20°C

Date no. time Tube KCl Citrate Other Temp Min Q Spr Morph Total Untr GTG FISH Susp
1a 2h 10 mL + – – RT 7 AB A B – – – – Reserve
only
4
1b – + – 7 AB AA C – – – – Discard
1c + – – 37°C 7 B C C – – – – Discard
1d – + – 7 C B C – – – – Discard
G-banding: inadequate Repeat: yes Action: discard; try KCl:NaCit 20:1 and 1:1
2a 2h 10 mL 10 1 – RT 7 A AA B− – Discard
2b 1 1 – 7 A A B+ 16 1 8 7 Pool
2c 10 1 – 1 A A B+ 2 ml
2d 1 1 – 1 AB AB B – discard
G-banding: yes, was OK Repeat: no Action: mix tubes 2b and 2c discard rest
Abbreviations: Q(uantity) of metaphases is defined as follows “A,” ³ 1 metaphase per low power (~100×) microscope field; “B,” ³ 1 metaphase per 10 low power fields; “C,” £
1 metaphase per row. Spr(eading) is defined as: “A,” optimal with all or most chromosomes separately visible; “AA” (possibly usable for FISH), as “A,” but mostly broken; “B”
(usable), with most metaphases showing crossed-over chromosomes; and “C” (unusable), with no chromosomes separately visible. Morph(ology): “A” (good), with parallel,
solid, clearly separated chromatids; “B” (average); and “C” (poor) with amorphous or refractile chromatids when viewed under phase-contrast. Intermediate quantities and quali-
ties are defined by “AB,” “BC,” etc. Other abbreviations: GTG, G-banding; temp(erature); susp, cell suspension, untr(eated). In the case of HUVEC, although the first harvest
was discarded, it provided information to direct the choice of hypotonic buffers in second harvest towards a more satisfactory conclusion
Classical and Molecular Cytogenetic Analysis
47
and discard wash. Add sufficient trypsin/EDTA to cover the

cells and incubate briefly (5–15 min) with intermittent light
agitation. When cells are ready (i.e., “rounded up”), shake vig-
orously and remove by rinsing with supernatant from the cen-
trifuge tube. Then, centrifuge aliquots as with suspension
cultures. (The serum present in the culture medium will act to
inactivate residual trypsin activity).
5. Suspend cell pellets gently by manual agitation. Add 5–20 vol
from various working hypotonic solutions (20:1, 1:1, etc.).
Incubate paired aliquots at (initially) room temperature for
1 min and 7 min, respectively. (See Table 1 for example).
6. Centrifuge and discard supernatant. Suspend cells gently and
carefully add ice-cold fixative, at first dropwise, and then faster,
until the tube is full.
7. Store refrigerated for 1–2 h.
8. Equilibrate to room temperature to minimize clumping and
then centrifuge (5 min at 400 × g). Repeat.
9. Store fixed cells overnight at 4°C.
Day 3
10. Equilibrate to ambient temperature and then centrifuge (5 min
at 400 × g). Repeat twice.
11. Resuspend cells in sufficient fixative to yield a lightly opaque
suspension. Typical cell concentrations range from 2 million to
8 million cells per milliliter.
12. Remove four precleaned slides (one per harvest tube) from
storage (at −20°C) and place on a plastic-covered freezer block
held at a slight incline away from the operator by insertion of a
pipet.
13. Locally humidify by breathing heavily on the slides.
14. Holding the pipet approx 30 cm above the slides, place two
drops of cell suspension onto each slide—the first immediately
below the frosted zone and the second about two-thirds along
the slide. Do not flood.
15. Lift slides in pairs for speed. Breathe on them again to maxi-
mize spreading.
16. (Optional) To improve spreading, gently ignite residual fixative
(with a camping stove or Bunsen burner). Do not allow slide
to get hot, as this could spoil subsequent G-banding and
FISH.
17. Label and air-dry. Stand slides vertically until dry.
18. Examine slides by phase-contrast microscopy and assess each
hypotonic treatment individually (see Note 3). Differences in
chromosome quality, as shown by the cells depicted in Fig. 1a,
b, should be manifestly obvious at this stage. The superior

metaphase (Fig. 1b) yielded satisfactory G-banding (Fig. 1c).
19. Prepare slides from successful treatments, mixing cell suspen-
sions if more than one is deemed adequate. Label.
20. Store unused cell suspensions at −20°C in sealed 2 mL
microfuge tubes filled to the brim to exclude air. Under such
conditions, suspensions remain stable for several years; we have
performed FISH successfully using 5-year-old suspensions.
Suspensions cryopreserved in this way must be thoroughly
washed in fresh fixative prior to slide preparation. After sam-
pling, suspensions should be refilled to the brim, marking the
original level to control dilution.
3.2. Trypsin G-Banding Although several banding methods are in use, the standard proce-
(see Notes 5 and 6) dure involves G-banding by trypsin pretreatment (20). G-Banding
selectively depletes the chromatin of certain proteins to produce
strong lateral bands after staining with Giemsa (see Fig. 1c–e).
Analysis of chromosomes harvested using the above-described
technique should typically reveal some 300 bands, the absolute
minimum required for detecting non-cryptic rearrangements.
However, with stretched or submaximally condensed (prometa-
phase) chromosome preparations up to 1,000 bands may be
distinguished.
1. Fresh slides are unsuitable for immediate G-banding. Slides
must be aged first. This is best achieved by baking overnight at
60°C in a dry oven. About six to eight slides containing an
adequate supply of well-spread metaphases with good chromo-
some morphology should be prepared for each cell line.
2. First prepare three Coplin jars, one each for 500 mL trypsin in
70 mL PBS (pH 7.2), ice-cold PBS (pH 6.8) to stop enzy-
matic activity, and 5% Giemsa in PBS (pH 6.8).
3. The Coplin jar containing trypsin in PBS should be placed in a
water bath at 37°C and equilibrated to 37°C before use.
4. The second Coplin jar containing PBS alone should be
placed on ice nearby and allowed to equilibrate to ca. 4°C
before use.
5. To estimate optimal trypsin incubation times, dip the first slide
halfway into the trypsin for 10 s and, thereafter, the whole slide
for another 10 s to test, in this case, for 10 s and 20 s trypsini-
zation times, respectively.
6. Immediately stop trypsin activity by immersion in ice-cold PBS
for a few seconds.
7. Stain in Giemsa solution for 15 min.
8. Rinse briefly in deionized H2O and carefully blot-dry using
paper towels.
9. Examine microscopically (see Note 5). Scan for likely meta-

phases at low power. Examine those selected in more detail at
higher power using the Epiplan dry objective. From the chro-
mosome banding quality, decide whether the suitable trypsin
time lies within the 10- to 20 s range spanned by the test slide.
If satisfactory, repeat steps 1–8. If unsatisfactory, repeat steps
1–9 using longer (e.g., 30–45 s) or shorter (e.g., 3–6 s) trypsin
test times, as appropriate until the optimal incubation time
becomes apparent.
3.3. FISH (see Notes 7 FISH is a versatile methodology which may be used to examine a
and 8) variety of genetic material by using probes of varying size: whole
genomes of ~3,000 Mbp using M-FISH or SKY, individual chro-
mosomes of ~50–200 Mbp using chromosome painting; ~100–
200 Kbp using bacterial artificial chromosome (BAC) clones ,or
~40 Kbp using fosmids. Those equipped with the most sensitive
cameras can also utilize FISH plasmids of ~2–10 Mbp, but results
are variable depending on locus “visibility” amid remaining chro-
matin. Painting probes may be used singly or in color-contrasted
mixtures—the latter maximizing the informational possibilities,
(e.g., for confirming a translocation inferred from G-banding).
Hybridization with painting probes is shown in Fig. 2a, illustrating
normal and rearranged chromosomes in cancer cells. Regardless of
the probe combination chosen, counterstaining is usually essential.
The standard chromosomal counterstain is 6-diamidino-2-phe-
nylindole dihydrochloride (DAPI), which yields deep blue color,
most intense at the centromeres, notably those of chromosomes 1,
9, and 16, and in the terminal long-arm region of the Y chromo-
some. DAPI generates negative G-bands which image analysis pro-
grams can converted into G-bands. Although painting probes are
normally sourced commercially, these may be self-produced by
selective PCR amplification of human chromosomal material
retained by monochromosomal human/rodent hybrid cell lines. It
is also possible to amplify human DNA selectively by exploiting
human-specific repeat sequences (e.g., Alu) as primer targets.
Laboratories equipped with a standard fluorescence microscope
and suitable filters and image analysis software can now evaluate
chromosome alterations at the gene level. Tilepath BAC and fos-
mid probes have become the current gold standard for this type of
analysis, because these provide direct links between chromosome
rearrangements and genes. Clones of interest are identified on
genome browsers, such as that hosted by the University of California
at Santa Cruz (http://genome.ucsc.edu/). Probes are produced
by labeling large-insert clones obtained from BACPAC Resources
(http://bacpac.chori.org/). Our current labeling protocol is given
elsewhere (26). Alternately, labeled probes for common cancer
gene loci are available commercially for a variety of neoplastic loci.
Fig. 2. FISH and SKY. Images (a, b) show FISH analysis of a human neuroblastoma cell line CHP-134 (DSM ACC 653). Image
(a) chromosome painting with library probes for chromosomes 3 (Spectrum Orange) and 7 (Spectrum Green). Note normal
configuration of chromosome 3 contrasting with that of chromosome 7 which is solely represented by rearrangements
(arrows). Image (b): FISH using a single locus fosmid probe (G248P89427C3 labelled green) covering the MYCN locus. Note
genomic amplification on three marker chromosomes corresponding to chromosome 7 derivative marker chromosomes as
shown in Fig. 2a. MYCN amplification is pathognomonic for advanced neuroblastoma and may, therefore, be used to confirm
the status of candidate cell lines. Image (c) FISH data from another DLBCL cell line, SU-DHL-16 (DSM ACC 577), depicting
rearrangement of the BCL6 locus as found in 40% patients diagnosed with this lymphoma. Three contrastingly labelled
RP11-library BAC clones (BACPAC Resources, Oakland, Ca., USA) were co-hybridized (inset). The central RP11-211 G3 clone
(yellow fluor ) straddles the BCL6 locus. Note translocation of part of RP11-211 G3 containing BCL6 to 12p11 (arrow) where
it is juxtaposed with ITPR2 (27). Image (d) SKY analysis of an acute promyelocytic leukemia cell line, AP-1060 (DSM ACC
593). The image shows paired raw and processed chromosomal images; those on the right of each pair have been enhanced
and pseudocolored to assist on-screen representation and discrimination. This cell line carries multiple rearrangements,
including t(15;17)(q22;q21) (red arrow) which causes fusion of two genes, PML (at15q22) and RARA (at 17q21), resulting in
the formation of a hybrid gene and chimeric PML-RARA protein. This protein blocks granulocyte differentiation, a step on the
road to leukemic transformation, and is a key therapeutic target, e.g., by treatment with all-trans retinoic acid (ATRA), a
derivative of vitamin A, which reverses the differentiation blockade by PML-RARA protein. G-banding, FISH and SKY images
were captured using image analysis systems (Applied Spectral Imaging, Edingen, Germany, or Smart Capture 2, Genetix,
Newcastle, UK) configured to Axioimager or Axioplan 2 photomicroscopes with x63/100 Planapochromat objectives, respec-
tively (Zeiss, Göttingen, Germany).
FISH using custom single-locus BAC/fosmid clones covering the

MYCN and BCL6 loci are depicted in Fig. 2b, c, respectively.
BAC clones used to prepare such probes contain repeat
sequences that require suppression by prehybridization with Cot-1
DNA. The posthybridization stringency wash, which can be performed
at either low temperatures including formamide, which lowers the

stability of the DNA double helix, or at higher temperatures using
low SSC concentrations alone, is critical to success. Stringency
washing allows the operator to control the balance of probe signal
intensity against background. The stability of DNA–DNA hybrids
on FISH slides allows repeated cycles of stringency washing. For
those starting with untested FISH probes, it is feasible to start off
using a less stringent wash, which, if yielding unacceptable back-
ground levels, can be repeated at higher stringencies (i.e., at lower
salt concentrations), the highest stringencies imposed by washes
are performed in water alone.
The protocol described below is applicable to a wide variety of
probes and, therefore, useful for those intending to combine probes
from different sources. Indirectly labeled probes (e.g., with digoxi-
genin or biotin) require additional detection steps that can be plugged
into the following protocol. In our hands, the same protocol also
works for complex probe mixes, such as those used for m-FISH or
SKY. The accompanying SKY images (Fig. 2d) were produced using
the same protocol. In our hands the “official” SKY protocol supplied
by the manufacturer took longer and yielded inferior results.
Day 1
1. Use either fresh (1–7 day old) or archival slides stored at
(−80°C).
2. Extraneous background signal, if present, can be reduced by pre-
incubation in pepsin solution for 2 min at 37°C. (See Note 7.)
3. Slide dehydration: pass slides sequentially through an alcohol
series for 2 min in 70% (two times), 90% (two times), and 100%
ethanol in Coplin jars.
4. Desiccate slides overnight at 42°C in a dry oven.
Day 2
5. De-proteinize in acetone for 10 min (to minimize background
autofluorescence).
6. Slide denaturation: place slides for 2 min at 72°C in 30 mL of 2×
SSC plus 70 mL formamide. The temperature of this step is
critical. Therefore, avoid denaturing too many slides simultane-
ously. If a high throughput is desired, slides should be pre-
warmed. Quench in prechilled (−20°C) 70% ethanol for 2 min.
7. Repeat step 3 (the alcohol series).
8. Varnish slide label (to prevent subsequent eradication).
9. Place slide on prewarmed block at 37°C.
10. Remove probe from the freezer noting the concentration of
labeled DNA. Add excess Cot-1 DNA (20–50× probe).
11. Probe denaturation: place desired volume of probe into
microfuge tube (sterile) and incubate in a “floater” for 5 min
at 72°C in a water bath. (Important: If recommended by man-

ufacturer, omit probe denaturation.)
12. Probe prehybridization: collect probe by brief centrifugation,
then incubate for 15–60 min at 37°C in a second water bath.
13. Probe application: using shortened micropipet tips (sterile),
carefully drop 8–12 mL of probe (making up the volume with
hybrisol, if necessary) onto each slide half. Thus, two hybrid-
izations may be performed on each slide (separated by a drop
of hybrisol, to inhibit mixing). Cover slides carefully with glass
cover slips, tapping out any bubbles, and seal with rubber
cement.
14. Hybridization: place slides carefully in a moistened and sealed
hybridization chamber. Leave overnight (or up to 72 h) in
incubator (preferably humidified) at 37°C.
Day 3
15. After hybridization, carefully remove rubber cement and cover
slips in 2× SSC using tweezers.
16. Stringency washing: wash slides for 5 min at 72°C in 0.5× SSC.
17. (Optional) For use with digoxigenin labeled probes: briefly
prewash in wash solution at ambient temperature and shake to
remove excess liquid. Important: Do not allow slides to dry
out until dehydration (step 18). To each slide, apply 40 mL
anti-digoxigenin antibody labeled with FITC (Qbiogene) and
cover with plastic cover slip. Incubate for 15–30 min at 37°C
in hybridization chamber. Wash for 5 min (three times) in wash
solution at room temperature in subdued light.
18. Dehydration (alcohol series): dehydrate slides as described in
step 3, but performed in subdued light.
19. Mounting and sealing: using shortened micropipet tips to ensure
even bubblefree coverage carefully place three 30 mL drops of
DAPI/Vectashield mountant along the slide. Apply cover slip
and tap out any large bubbles using the blunt end of a pencil or
equivalent. Seal with nail varnish. Allow varnish to dry.
20. Visualization: visualize slides at high power under oil immer-
sion with a 63× objective with a high numerical aperture. (see
Note 8).
21. Analysis and interpretation: see Notes 8 and 9.
4. Notes
1. Imaging: electronic imaging systems have completely replaced

film cameras for analysis and documentation. For further infor-
mation, consult the Web site of Applied Spectral Imaging
(http://www.spectral-imaging.com/), Genetix (http://www.

genetix.com/en/home/index.html), Metasystems (http://
www.metasystems.de/), or Zeiss (http://www.zeiss.com/
micro) which supply a variety of systems. Image analysis con-
fers significant benefits, notably real-time imaging, amplification
of weak signals, merging of differently colored signals, contrast
enhancement, background reduction, generation of G-bands
from DAPI counterstain, and rapid documentation and print-
ing. While SKY, supplied by Applied Spectral Imaging, uses
spectrophotometric separation to distinguish fluorescence
excitation and emission spectra, the remaining systems use
filters. Having compared all three systems, for multicolor FISH
we have obtained the best results with SKY (see Fig. 2d).
2. FISH probes: complex rearrangements discovered by
G-banding can only be resolved by chromosome painting
probes which are now available for a variety of mammalian spe-
cies. However, commercial painting and satellite DNA probes
all-too-often yield unsatisfactory results. Thus, it is necessary
first to calibrate new DNA probes using normal chromosomes.
This effort is usually well invested. Some probes generate
unnecessarily bright signals. Knowing this beforehand allows
such probes to be “stretched” by dilution with hybridization
buffer. All too often, probes arrive that yield inadequate or
inappropriate signals. Troubleshooting minimizes the risk of
following false trails inspired by spurious data.
3. Slide-making: for analysis, slides should meet three criteria:
plentiful metaphases, satisfactory chromosome spreading, and
superior morphology (i.e., large but undistended chromatids
lying in parallel). To document progress in harvesting proce-
dures and aid evidence-based searches for their improvement,
we use a standard data sheet that records progress toward these
ideals. An actual example is shown in Table 1, which presents
harvesting data for primary human endothelial cells as depicted
in Fig. 1a–c. In this case, reasonable preparations were only
obtained on the second attempt using the standard protocol
(Subheading 3.1, step 3). Although, on the second attempt, all
four hypotonic combinations yielded adequate numbers of
metaphases, only tubes -b and -c yielded satisfactory spreading
and morphology and were mixed for subsequent slide prepara-
tion. A total of 16 slides were prepared: 7 for G-banding, 1 for
Giemsa staining alone (to check for the presence of smaller
chromosomal elements that G-banding sometimes renders
invisible, such as so-called double minute chromosomes which
may harbor oncogenes), and 8 for FISH. The remaining cell
suspension in fixative was stored (−20°C) for future use.
Slides with sparse yields of metaphases are unsuitable for
FISH, where probe costs are often critical. For slowly dividing
cell lines (doubling times > 48 h), colcemid times can be

increased first to 6 h, then to 17 h (overnight), simultaneously
reducing colcemid concentrations by a half to minimize toxic-
ity. However, paucity of metaphases is usually the result of
depletion by overly harsh hypotonic treatments. Paradoxically,
we find that reducing hypotonic exposures to 1 min and, if
necessary, performing this step in microfuge tubes to facilitate
speedy centrifugation to reduce total hypotonic times may
improve spreading and yield by enabling survival of fragile cells
which might otherwise be lost. Tight metaphases with an
excess of overlapping chromosomes might be useful for FISH
but are unsuitable for G-banding. In such cases, spreading can
sometimes be improved by harsher hypotonic treatment,
whether by increasing the proportion of KCl to 100%, by
increasing the hypotonic time up to 15 min, or by performing
the latter at 37°C instead of at room temperature. Gentle
flaming often assists spreading and, contrary to received wis-
dom, has little or no deleterious effect on G-banding or FISH.
Dropping suspensions from suicidal heights seldom improves
spreading, although heavy breathing, performed both immedi-
ately before and after dropping, is beneficial, by increasing
local humidity levels. Excessive spreading, on the other hand,
is often cured by reducing the proportion of KCl, by reducing
hypotonic treatment times, or by retaining more of the origi-
nal medium from the first centrifugation (Subheading 3.1,
steps 3 and 4).
4. Harvesting with FUDR: As a general rule, the best morpholo-
gies are produced by hypotonic solutions containing 50% or
less Na–citrate. Excessive amounts of the latter tend to yield
fuzzy irregular morphologies that produce disappointing
results with G-banding and FISH alike. Some types of cells and
derived cell lines consistently yield short stubby chromosomes
that appear refractory to all attempts at improvement. In such
cases, it may be helpful to try FUDR pretreatment. Accordingly,
treat cultures overnight with FUDR/uridine. The next morn-
ing, resuspend in fresh medium with added thymidine to
reverse the blockade and harvest 7–9 h later.
5. G-Banding: Good chromosomes usually yield good G-banding
(see Fig. 1b, c for an example). Exceptions include “young”
chromosomes (puffed up or faint banding) or “has-beens”
(poor contrast or dark banding). Artificial aging by baking
overnight at 60°C not only speeds up results but reduces varia-
tions in trypsin times due to variations in temperature or
humidity. For those requiring a same-day result, aging times
could be shortened to 60–90 min by increasing the hot plate/
oven temperature to 90°C. Trypsin G-banding is a robust
technique and problems unconnected with poor chromosome
morphology are rare. Those used to working with one species

should note, however, that chromosomes of other species
could exhibit higher/lower sensitivities to trypsin. Losses in
tryptic activity occur after about 6 month among aliquots
stored at (−80°C), which should then be discarded in favor of
fresh stocks.
6. Karyotyping: Although SKY and m-FISH may be used to dis-
tinguish all 24 chromosomes and to detect rearrangements,
classical G-Banding remains at the core of cytogenetic analysis
as it is quicker, cheaper, and offers higher resolution. The abil-
ity to recognize each of the 24 normal human chromosome
homologs necessarily precedes analysis of rearrangements.
Because the majority of human cancer cell lines carry chromo-
some rearrangements, the choice of cell lines for learning pur-
poses is critical. Learning should be performed using either
primary cultures of normal unaffected individuals (e.g., lym-
phocyte cultures) or B-lymphoblastoid cell lines known to have
retained their diploid character. Those intent on acquiring the
ability to perform karyotyping are strongly advised to spend
some time in a laboratory where such skills are practiced daily
(e.g., a routine diagnostic laboratory). An image analysis sys-
tem was used to convert the G-banded metaphase (Fig. 1d)
into a karyotype (Fig. 1e).
7. FISH signals and Noise: Excessively high background signals
often plague FISH using BACs, which may cover 200 Kbp and
may include DNA sequences which cross-hybridize to other
loci. Pretested commercial probes are usually, but not always,
free of this problem. Increasing the wash stringency (Subheading
3.3, step 16) by reducing the SSC concentration to 0.1× might
help. Alternatively, adding Cot-1 DNA to the hybridization
mix should minimize hybridization noise. When using non-
commercial probes, excessive noise can often be cured by
reducing the probe concentration. Normal DNA concentra-
tions for single-locus probes should range from 2–6 ng/mL to
10–20 ng/mL for painting probes. Assuming that it is not the
result of “dirty” slides, nonspecific noise may be caused by
either autofluorescence or protein–protein binding after anti-
body staining, which can often be reduced by additional slide
pretreatment in pepsin solution (Subheading 3.3, step 2).
Incubate slides for 2 min in acidified pepsin solution at 37°C.
Rinse in PBS (pH 7.2) for 3 min at room temperature. Postfix
slides, held flat, in 1% formaldehyde solution for 10 min at
room temperature using plastic cover slips. Rinse in PBS (pH
7.2) for 3 min at room temperature. Continue with step 3 of
Subheading 3.3. Weak FISH signal intensity might arise
because the probe itself is inherently weak, the wash too
stringent, or the chromosomes insufficiently denatured. To

test for these alternatives, repeat the stringency wash
(Subheading 3.3, step 16) but with either 2× or 1× SSC in the
wash buffer. In parallel, repeat the slide denaturation
(Subheading 3.3, step 6), increasing the denaturation time to
4 min. If neither modification brings any improvement, nor
the probe is new and untested or old and infrequently used, it
is likely that the probe is inherently weak. (Even large-insert
clones sometimes deliver puzzlingly weak signals, usually
attributed to the inaccessibility of their chromosomal targets.)
For those equipped with advanced imaging systems incorpo-
rating a camera of high sensitivity, it is often possible to capture
images from probe signals invisible to the naked eye. In the
case of new commercial probes, the supplier should be con-
tacted. Probes with larger targets often cross-hybridize to simi-
lar DNA sequences present on other chromosomes. It is
important first to identify patterns of cross-hybridization by
FISH onto normal chromosomes to avoid misinterpreting the
latter as rearrangements.
8. FISH: FISH helps resolve rearrangements of interest which
resist analysis by G-banding. This inevitably requires both
intuition and luck. Clearly, the need for the latter is reduced
where G-banding is optimized. Although chromosome paint-
ing probes (such as those shown in Fig. 2a) became widely
available in the mid-1990s, the advent of mapped BAC clones
enabling do-it-yourself construction of panels to detect specific
oncogene alterations (such as that for MYCN shown in Fig. 2b,
or BCL6 in Fig. 2c) came nearly a decade later in the wake of
the Human Genome Project data and resources.
In humans FISH was originally used to map genes. Now
mapping is restricted to chromosome rearrangements. The
most difficult rearrangements to resolve are unbalanced ones
involving multiple chromosomes. Sometimes, originally recip-
rocal translocations appear unbalanced because of loss or addi-
tional rearrangement of one partner. In such cases, the identity
of the “missing partner” might be often guessed at from among
those chromosomes where one or more homologs appear to
be missing. Having identified the chromosomal constituents of
cryptic rearrangements, the next task is to reconcile FISH with
G-banding data, enabling breakpoint identification. In cases
where chromosome segments are short or their banding pat-
terns nondescript, this aim might be frustrated. The
International System for Chromosome Nomenclature (ISCN)
allows almost all rearrangements to be described with minimal
ambiguity in most cases (28). This system was updated in 1991
for cancer cells (29), and in 1995, 2005, and 2009 for FISH
(30). The most efficient way to detect and analyze multiple

chromosome rearrangements is to combine G-banding with
SKY (Fig. 2d) which in our hands has proved more reliable
than m-FISH.
9. Data usage: Having successfully completed cytogenetic analy-
sis of primary cells or an established cell line to the point of
ISCN karyotyping, questions of what to do with the data arise.
The first question concerns identity: Has the cell line in ques-
tion been karyotyped previously and, if so, does the observed
karyotype correspond with that previously reported; or, in the
case of animal cells, is the karyotype compatible with the sup-
posed species-of-origin? In our experience, complete corre-
spondence between historical and actual cell line karyotypes is
rare, even where their identity has been confirmed by DNA
fingerprinting. Among complex karyotypes, complete resolu-
tion might be unnecessary and is, indeed, rarely achieved, leav-
ing significant scope for uncertainty and differences in
interpretation. Wherever possible, consult the original journal
or reprint, as photocopies seldom permit reproduction of
intermediate tones, which are the “devil in the detail” of
G-banding. Those wishing to compare their karyotypes with
those derived at the DSMZ can consult either “catalogues rai-
sonnés” of human leukemia lymphoma cell lines (31, 32), or
the DSMZ Web site which features an interactive searchable
database of all types of (mainly) human cancer cell lines
(http://www.dsmz.de/). In the case of species confirmation,
certain scientific journals specialize in karyotypic studies per-
formed on nonhuman animals, notably Caryologia,
Chromosoma, Chromosome Research, Chromosome Science,
Cytogenetics and Genome Research (formerly Cytogenetics and
Cell Genetics), and Cytologia. For those denied access to some
of these journals, alternative sources are available on the Web
where karyotype and metaphase images of common (and some
exotic) species are accessible. Again, in the case of cells har-
vested from noncancerous tissue, chromosome number is a
useful identifier. Diploid chromosome numbers (2 N) of some
common animal species are given in Table 2 which includes
data on a range of species detailed elsewhere (en.wikipedia.
org/wiki/List_of_organisms_by_chromosome_count).
Acknowledgments
The authors wish to thank colleagues at the DSMZ for their useful
comments and suggestions. A special mention goes to Maren
Kaufmann, many of whose ideas are silently incorporated in the
foregoing protocols.
Table 2
Diploid chromosome numbers (2N) of some common animal
and insect species
Organism Scientific name 2N

Cat Felis catus 38
Chicken Gallus gallus domesticus 78
Chimpanzee Pan troglodytes 48
Cow Bos primigenius 60
Dingo Canis lupus dingo 78
Dog Canis lupus familiaris 78
Fruit fly Drosophila melanogaster 8
Giraffe Giraffa camelopardalis 62
Golden jackal Canis aureus 78
Gorilla Gorilla gorilla 48
Gray fox Urocyon cinereoargenteus 66
Horse Equus ferus caballus 64
Human Homo sapiens 46
Lion Panthera leo 38
Mosquito Aedes aegypti 6
Mouse Mus spp. 40
Pig Sus scrofus 38
Rabbit Oryctolagus cuniculus 44
Rat Rattus spp. 40
Red deer Cervus elaphus 68
Red fox Vulpes vulpes 34
Rhesus monkey Macaca mulatta 48
Sheep Ovis spp. 54
Wolf Canis lupus 78
References
1. Mitelman F, Johansson B, Mertens F (2007) 3. Deininger M, Buchdunger E, Druker BJ
The impact of translocations and gene fusions (2005) The development of imatinib as a ther-
on cancer causation. Nat Rev Cancer apeutic agent for chronic myeloid leukemia.
7:233–245 Blood 105:2640–2653
2. MacLeod RAF, Nagel S, Scherr M, Schneider 4. Drexler HG, MacLeod RAF, Uphoff CC
B, Dirks WG, Uphoff CC et al (2008) Human (1999) Leukemia cell lines: in vitro models for
leukemia and lymphoma cell lines as models the study of Philadelphia chromosome-positive
and resources. Curr Med Chem 15:339–359 leukemia. Leuk Res 23:207–215
5. Drexler HG, MacLeod RAF, Borkhardt A, 18. Caspersson T, Zech L, Johansson C (1970)
Janssen JWG (1995) Recurrent chromosomal Differential binding of alkylating fluorochromes
translocations and fusion genes in leukemia– in human chromosomes. Exp Cell Res
lymphoma cell lines. Leukemia 9:480–500 60:315–319
6. Marini P, MacLeod RAF, Treuner C, Bruchelt 19. Sumner AT, Evans HJ, Buckland RA (1971)
G, Böhm W, Wolburg H et al (1999) SiMa, a New technique for distinguishing between
new neuroblastoma cell line combining poor human chromosomes. Nat New Biol 232:31–32
prognostic cytogenetic markers with high 20. Seabright M (1973) Improvement of trypsin
adrenergic differentiation. Cancer Genet method for banding chromosomes. Lancet
Cytogenet 112:161–164 1:1249–1250
7. Drexler HG, Matsuo Y, MacLeod RAF (2000) 21. Rowley JD (1973) A new consistent chromo-
Continuous hematopoietic cell lines as model somal abnormality in chronic myelogenous
systems for leukemia–lymphoma research. Leuk leukaemia identified by quinacrine fluorescence
Res 24:881–911 and Giemsa staining. Nature 243:290–293
8. Tosi S, Giudici G, Rambaldi A, Scherer SW, 22. Nowell PC, Hungerford DA (1960) A minute
Bray-Ward P, Dirscherl L et al (1999) chromosome in human granulocytic leukemia.
Characterization of the human myeloid leuke- Science 132:1497
mia-derived cell line GF-D8 by multiplex
23. Cremer T, Lichter P, Borden J, Ward DC,
fluorescence in situ hybridization, subtelomeric
Manuelidis L (1988) Detection of chromo-
probes, and comparative genomic hybridization.
some aberrations in metaphase and interphase
Genes Chromosomes Cancer 24:213–221
tumor cells by in situ hybridization using
9. Schneider B, Nagel S, Kaufmann M, chromosome-specific library probes. Hum
Winkelmann S, Drexler HG, MacLeod RAF Genet 80:235–246
(2008) t(3;7)(q27;q32) fuses BCL6 to a non-
coding region at FRA7H near miR-29. 24. Lichter P, Cremer T, Borden J, Manuelidis L,
Leukemia 22:1262–1266 Ward DC (1988) Delineation of individual
human chromosomes in metaphase and inter-
10. Vanasse GJ, Concannon P, Willerford DM phase cells by in situ suppression hybridization
(1999) Regulated genomic instability and neo- using recombinant DNA libraries. Hum Genet
plasia in the lymphoid lineage. Blood 80:224–234
94:3997–4010
25. Lichter P (1997) Multicolor FISHing: what’s
11. Küppers R, Dalla-Favera R (2001) Mechanisms
the catch? Trends Genet 12:475–479
of chromosomal translocations in B cell lym-
phomas. Oncogene 20:5580–5594 26. MacLeod RAF, Kaufmann M, Drexler HG
(2007) Cytogenetic harvesting of commonly
12. MacLeod RAF, Dirks WG, Matsuo Y,
used tumor cell lines. Nat Protoc 2:372–382
Kaufmann M, Milch H, Drexler HG (1999)
Widespread intraspecies cross-contamination 27. Schneider B, Nagel S, Ehrentraut S,
of human tumor cell lines arising at source. Int Kaufmann M, Meyer C, Geffers R, et al.
J Cancer 83:555–563 (2012) Neoplastic MiR-17~92 deregulation
13. Drexler HG, Dirks WG, MacLeod RAF (1999) at a DNA fragility motif (SIDD). Genes
False human hematopoietic cell lines: cross- Chromosomes Cancer 51:219–228
contaminations and misinterpretations. 28. ISCN (1985) An International system for
Leukemia 13:1601–1607 human cytogenetic nomenclature: report of
14. Nelson-Rees WA, Daniels DW, Flandermeyer the standing committee on human cytogenetic
RR (1981) Cross-contamination of cells in cul- nomenclature. Karger, Basel
ture. Science 212:446–452 29. ISCN (1991) Guidelines for cancer cytogenet-
15. MacLeod RAF, Dirks WG, Drexler HG (2002) ics: supplement to an international system for
Persistent use of misidentified cell lines and its human cytogenetic nomenclature. Karger, Basel
prevention. Genes Chromosomes Cancer 30. ISCN 2009: an international system for human
33:103–105 cytogenetic nomenclature: recommendations
16. American Type Culture Collection Standards of the international standing committee on
Development Organization Workgroup ASN- human cytogenetic nomenclature. Published in
0002 (2010) Cell line misidentification: the collaboration with ‘Cytogenetic and Genome
beginning of the end. Nat Rev Cancer. Research.’ Karger, Basel
doi:10.1038/nrc2852 31. Drexler HG (2001) The leukemia-lymphoma
17. Capes-Davis A, Theodosopoulos G, Atkin I, cell line factsbook. Academic, San Diego
Drexler HG, Kohara A, MacLeod RAF et al 32. Drexler HG (2010) Guide to leukemia-lym-
(2010) Check your cultures! A list of cross- phoma cell lines, 2nd edn. eBook (available
contaminated or misidentified cell lines. Int J from the author), DSMZ, Braunschweig,
Cancer 127:1–8 Germany
Chapter 5
Fluorescent In Situ Hybridization of DNA Probes

in the Interphase and Metaphase Stages of the Cell Cycle
Linda A. Cannizzaro
Abstract
In the past decade, fluorescent in situ hybridization (FISH) has been used routinely in detecting molecular
abnormalities in the interphase and metaphase stages of the cell cycle. Many of the molecular anomalies
which are detected in this manner are diagnostic of a prenatal, postnatal, or neoplastic genetic disorder.
With the continuous isolation of commercially available DNA probes specific to a particular chromosome
region, FISH analysis has become standardized in its ability to detect characteristic chromosomal anoma-
lies in association with genetic and neoplastic diseases. In recent years, FISH has also become automated
to accommodate the increased volume of slide preparations necessary for the number of DNA probes
needed to detect characteristic molecular anomalies in cancer tissues and bone marrow samples. FISH
technology provides essential information to the physician regarding the diagnosis, response to treatment,
and ultimately the prognosis of their patients’ disorder. It has become an important source of information
routinely used in conjunction with chromosome analyses, and presently to confirm molecular alterations
detected by array comparative genomic hybridization (aCGH) analyses. In this chapter we describe the
methods for performing FISH analyses in order to determine the presence or the absence of genetic abnor-
malities which define whether the patient has either a genetic syndrome or malignant disease.
Key words: In situ hybridization, Fluorescent analyses, DNA probes, Cancer, Genetic disorders
1. Introduction
Fluorescent in situ hybridization (FISH) is a sensitive yet powerful

method for mapping and positioning DNA sequences in mamma-
lian genome systems (1–3). DNA sequences ranging in size
from < 1 kb to several megabases can be localized to a specific chro-
mosome site. The DNA is inserted into a variety of vectors, i.e.,
cosmids or BACS, and then labeled with a nonradioactive
immunofluorescent compound such as biotin-11-dUTP or digox-
61
62 L.A. Cannizzaro
igenin-11-dUTP (3). Once labeled, the DNA probe is hybridized

to cell or chromosome preparations, including tissue samples fixed
in paraffin. The signal(s) is detected under ultraviolet light using
filters of wavelengths specific to the fluorescent compound.
In the early years after its introduction, FISH was primarily used
to construct a physical map of chromosome regions consistently
involved in genetic and malignant disorders (4–6). Determining the
orientation of genes resulting from alterations such as translocations
or inversions would enable investigators to understand the conse-
quences of such genetic alterations usually leading to an altered or
nonfunctional form of a gene product (6). Such alterations have a
domino effect in the individuals’ genome ultimately causing mani-
festation of a genetic or a malignant disorder.
DNA probe hybridizations to nondividing cells in the inter-
phase stage of the cell cycle enables detection of molecular altera-
tions without the need to obtain metaphase chromosomes,
especially in patients undergoing chemo- or radiation therapy. The
trauma caused by such therapeutic regimens, is often reflected by
the limited number of metaphases available for analysis in the
patients’ marrow or peripheral blood sample. FISH analysis in the
interphase stage in such patients not only can detect the presence
or absence of alterations characteristic of the disorder but also can
detect low levels of residual abnormal clones (4, 5). High resolu-
tion detection of such alterations is a quantitative measurement
which can be used to determine the effectiveness of a specific ther-
apeutic regimen, by determining whether the patient responds or
shows improvement to the drug or radiation treatment.
Frequent follow-up evaluations of the patients’ condition at
each stage of the progression of the disease can be performed using
routine FISH analyses (7). Hybridizations of multiple DNA probes
to a patient sample can be carried out in a matter of hours. This
enables the physician to maintain or to modify the treatment, or to
provide more aggressive therapy to the patient, based on the results
from these follow-up studies. In many hospitals and clinics it is not
unusual for the physician to request multiple FISH follow-up stud-
ies for a patient with cancer as their disorder moves from one phase
of treatment or stage of development, to the next.
Clinical applications of FISH technology continue to evolve as
more consistent chromosome and molecular alterations are uncov-
ered in association with genetic and malignant disorders. Initially,
most DNA probes were isolated in an effort to determine the
placement and orientation of genes along the chromosome itself
(8–16). “Painting” probes are used to detect a signal from one
specific chromosome or chromosome segment. These probes are
known as whole chromosome painting (WCP) probes, and most
WCP probes were isolated from flow sorting individual human
chromosomes or by PCR amplification of DNA from somatic cell
hybrids retaining only one human chromosome or segment of one
5 FISH Analysis in Interphase and Metaphase 63
Table 1
FISH analysis for microdeletion syndromes
Abnormality
Syndrome DNA probes Chromosome location detected
1p36 microdeletion p58/LSI 1q25 1p36/1q25 1p36 deletion
Wolf Hirschhorn WHS/CEP4 4p16.3/cen 4 4p16.3 deletion
Cri-du-Chat D5S23/D5S721 5p15.2 5p15.2 deletion
Williams ELN/D7S486 7q11.23/7q31 7q11.23 deletion
Prader-Willi SNRPN/GARB3/D15S10 15q11-13 15q11.2 deletion
Angelman SNRPN/GARB3/D15S10 15q11-13 15q11.2 deletion
Miller Dieker LIS1/RARA 17p13.3/17q21.1 17p13.3 deletion
Smith Magenis SMS/RARA 17p11.2/17p21.1 17p11.2 deletion
DiGeorge/Velocard- HIRA/TUPLE1/ARSA/ 22q11.2/22q13 22q11.2 deletion
iofacial (VCF) N25
Kallmann KAL/DXZ1 Xp22.3/cen X Xp22.3 deletion
X-linked ichthyosis STS/DXZ1 Xp22.3/cen X Xp22.3 deletion
Sex reversal/Ambig SRY Yp11.3 Yp11.3 deletion
genitalia
chromosome. A positive hybridization signal of the WCP probes

can be detected in either the metaphase or interphase stage of the
cell cycle. The limitation of these probes is to detect rearrange-
ments of whole chromosome segments only. However, WCP
probes are still extremely valuable in detecting the chromosomal
origin of cryptic or subtle chromosome alterations.
At the same time, a number of unique sequence DNA probes
were designed that can detect microdeletions, such as that specific
for Williams syndrome on chromosome region 7ql1.2 (Table 1).
This deletion is not visible by high resolution banded chromo-
somes, yet is readily detected with a DNA probe specific for the
elastin gene (ELN) locus within this chromosome region. A num-
ber of DNA probes are now available which can resolve deletions
of specific loci diagnostic of a microdeletion syndrome (Table 1).
Since these probes are now commercially available and have been
standardized for clinical diagnosis, it is easy to detect microdele-
tions prenatally as well as postnatally. Communication of this infor-
mation in nontechnical language to the expecting family is an
essential component of prenatal counseling in those situations
where the deletion is detected in the fetus.
64 L.A. Cannizzaro
Significant advances have been made in the last decade for

FISH analysis of samples from patients with hematopoietic and
solid tumor disorders. A substantial number of DNA probes are
now available for diagnosis of specific translocations and other
alterations consistently found in patients with different types of
cancer (7, 15, 17–21, 27, 32–35). Improvement in the detection
and diagnosis of all types of cancers has helped to uncover a con-
tingent of genetic loci which were isolated as markers to diagnose,
and define specific malignancies and stages of disease progression
in that particular cancer type (15, 19, 20).
For detecting chromosome translocations in cancer, DNA
probes have been isolated which are designed in different ways to
detect the presence of the translocation (Tables 2 and 3). For
instance, DNA probes which detect the bcr:abl fusion from the 9;22
chromosome translocation typical of chronic myelogenous leukemia
(CML), are now available as a dual fusion probe, i.e., one which
detects the fusion of the bcr:abl loci as a fused yellow signal when the
red and green signals from the two different chromosome regions
are joined with one another (4–6). Other DNA probes are also avail-
able known as breakapart probes, in which the yellow fusion signal
detects the normal chromosome arrangement, and when the trans-
location is present, it is detected as separate red and green signals in
interphase or metaphase, and hence broken apart!
Another routine use of FISH is in detecting minimal residual
disease to determine if the treatment regimen is working or not (17).
And in some cases, amplification of oncogene loci, such as MLL, by
FISH (Fig. 1a, b) is another indicator as to whether the patient is
responding well to treatment or may indicate that they are resistant
to the prescribed treatment regimen (22–24). Such information
obtained in a very short time is extremely useful to the physician,
who can then modify their patients’ treatment accordingly.
Use of a combination/multiple DNA probes specifically tar-
geting molecular alterations consistently present in leukemias or
lymphomas (Table 2) (32, 34) can be technically problematic. The
number of probes needed for many hematopoietic disorders results
in the need to use a greater volume of slides to complete each
FISH experiment. Since the amount of patient sample is limited,
especially from bone marrow and processed tissue samples, it is
essential that each FISH experiment works the first time. To ensure
that the hybridization conditions are comparable for each slide,
many labs now use a Thermobrite (Vysis), which standardizes the
denaturation and hybridization of multiple FISH probes on mul-
tiple slides. The use of this device can significantly improve turn-
around time for the increased volume of patient samples.
Not only has the hybridization process become more standard-
ized and automated, but so has the screening of the slides under the
fluorescent microscope. Many of the computer imaging systems
now used to capture and document the results of FISH experiments,
Table 2
FISH panels in leukemias and lymphomas
Panel DNA probes Chromosome location Abnormality detected
Myelodysplastic D5S323,D5S721/EGR1 5p15.2/5q31 Monosomy 5 and

syndrome (MDS) deletion 5q31
D7Z1/D7S486 Centromere 7/7q31 Monosomy 7 and
deletion 7q31
D8Z2 Centromere 8 Trisomy 8
D20S108 20q12 20q12 deletion
P53 17p13.1 17p13.1 deletion
Acute myeloid EGR1 5q31 Monosomy 5 and
leukemia (AML) deletion 5q31
D7Z1/D7S486 Centromere 7/7q31 Monosomy 7 and
deletion 7q31
RUNX1T1/RUNX1 8q22/21q22 t(8;21)(q22;q22)
MLL 11q23 11q23 rearrangement
CBFB 16q22 Inversion 16
CBFB/MYH11 inv(16) fusion gene
RARA 17q21.1 17q21.1 (APL)
rearrangement
BCR/ABL1 22q11.2/9q34 t(9;22)(q34;q11.2)
Chronic myeloid BCR/ABL1 D.F. 22q11.2/9q34 t(9;22)(q34;q11.2)
leukemia (CML) D8Z2 Centromere 8 + 8( Blast crisis)
RARA 17q21 i(17)(q10)
E2A 19p13 +19
Chronic lymphocytic MYB 6q23.3 6q23.3 deletion
leukemia (CLL) CCND1/IGH D.F. t(11;14)(q13;q32) t(11;14)(q13;q32)
ATM 11q22.3 11q22.3 deletion
D13S319 13q14.3 13q14.3 deletion
TP53 17p13.1 17p13.1 deletion
Acute lymphocytic MYC 8q24 8q24 rearrangement
leukemia (ALL)
ETV6 B.A. 12p13 t(12;21)(p13;q22)
p16 9q21 9q21 deletion
BCR/ABL D.F. 22q11.2/9q34 t(9;22)(q34;q11.2)
IGH B.A. 14q32 14q32 rearrangement
E2A 19p13 t(1;19)(q23;p13)
CEN4, CEN10, CEN17 4,10,17 centromeres aneuploidy
(continued)
66 L.A. Cannizzaro
Table 2
(Continued)
Panel DNA probes Chromosome location Abnormality detected

Myeloproliferative D5S323,D5S721/CSFIR 5p15.2/5q33-34 Monosomy 5 and
disease (MPD) deletion 5q33-34
D7Z1 Centromere 7 Loss or gain of 7
BCR/ABL1 DF 9q34;22q11.2 t(9;22)(q34;q11.2)
D13S319 13q14.3 13q14.3 deletion
Multiple myeloma MLL 11q23 11q23 rearrangement
(MM) RB1 13q14.3 13q14.3 deletion
IGH 14q32 14q32 rearrangement
TP53 17p13.1 17p13.1 deletion
Non-Hodgkins ALK 2p23 2p23 rearrangement
lymphoma (NHL) BCL6 3q27 3q27 rearrangement
MYC 8q24 8q24 rearrangement
IGH 14q32 Rearrangement
BCL2 18q21 Rearrangement
CCND1 11q13 Rearrangement
Mantle cell IGH/CCND1 14q32/11q13 t(11;14)(q13;q32)
lymphoma
Follicular IGH/BCL2 14q32/18q21 t(14;18)(q32;q21)
lymphoma
Mucosa-associated MALT1 14q32/18q21 t(14;18)(q32;q21)
lymphoma
Burkitt lymphoma MYC/IGH 8q24/14q32 t(8;14)(q24;q32)
Diffuse large BCL6 3q27/14q32 t(3;14)(q27;q32)
B-cell lymphoma
(DLBCL)
Bone marrow DXZ1/DYZ3 X and Y centromere Monitoring sex-mis-
transplant (BMT) matched BMT
Chronic D5S323,D5S721/CSFIR 5p15.2/5q33-34 Monosomy 5 and
myelomonocytic deletion 5q33-34
leukemia D7S486 7q31/Centromere 7 Loss or gain of 7;
(CMML) Deletion of 7q31
ETV6 12p13 12p13 rearrangement
D13S319 13q14.3 13q14.3 deletion
PDGFRB B.A. 5q33.1/12p13 t(5;12)(q33;p13)
Abbreviations: D.F. dual fusion DNA probe, B.A. break apart DNA probe
have automated screening mechanisms attached to the fluorescent

microscope which can be set up to automatically count signals in
interphase nuclei or count metaphase spreads. However, because of
possible variation of the results generated from such systems, some
hands-on human manipulation is necessary to complete the screen-
ing process. In using automated systems, one has to exercise care in
the final interpretation of the data despite the time saved in process-
ing and screening the FISH experiment.
Due to the complex nature of chromosome rearrangements
found in progressive malignancies, another type of FISH method-
ology known as M-FISH/SKY (Multiplex-FISH, Spectral
Karyotyping) has been used successfully in patients with more pro-
gressive cancer. These techniques are somewhat similar to one
another and basically involve labeling the DNA probes with several
fluorescent compounds to ultimately produce a different fluorescent
color for each chromosome (25–33), thus enabling more accurate
identification of complex chromosome alterations. If several differ-
ent chromosomes become rearranged to produce a large marker
chromosome, the origin of each chromosome is easily identified by
the different colors along the length of the marker chromosome.
The disadvantage of this technique is that because of the longer
hybridization times, it is not a useful technique for routine clinical
diagnosis.
Today, FISH is multifunctional and continues to be a valuable
technical resource for clinical diagnosis of genetic and malignant
disease. Panel FISH analyses using multiple DNA probes on
patients with a specific malignancy are routine clinical diagnostic
procedures. More importantly, FISH is also now used in conjunc-
tion with array comparative genomic hybridization analyses
(aCGH). Array CGH is a high resolution molecular technique
which detects copy number variations in DNA from all types of
patient samples (36–39). The array itself consists of thousands of
genes located on a slide/disc to which the patients’ DNA is
hybridized.
Clinically aCGH can identify microdeletions or duplications of
chromosome regions that cannot be detected by high resolution
cytogenetics. But like most techniques, aCGH has its limitations,
and cannot detect chromosome translocations or inversions. In
addition, this technology has not yet been standardized in deter-
mining cutoff points for determining whether the results obtained
are significant or not. In those situations where extremely small
deletions or duplications are detected, usually less than 0.5 MB, it
is not always possible to determine whether the alteration is real,
and that presents a dilemma to the physician in interpreting the
information to the patient.
Thus, FISH analyses must be used simultaneously to confirm
the presence of such alterations in the same patient sample.
68
Table 3
FISH analysis for soft tissue and solid tumors
Tumor type DNA probes Chromosome location Abnormality detected

L.A. Cannizzaro
Breast cancer (FDA D17Z1/HER-2/neu Centromere 17 and 17q11.2-12 HER-2/neu amplification

approved)
Bladder cancer and D3Z1/D7Z1/D17Z1/p16 Centromere 3, 7, 17 and 9p21 Loss of centromere 3, 7, 17
Cholangiocarcinoma and p16 loci
(FDA approved)
Lung cancer D5S23/D5S721/D6Z1/EGFR/ 5p15.2/Centromere 6/7p12/8q24/2p23 Loss or gain of 5p15.2/2p23
C-MYC/ALK Centromere 6/7p12/8q24
Retinoblastoma RB1 13q14 Deletion of 13q14
Neuroblastoma NMYC 2p24.1 NMYC amplification
Burkitt lymphoma C-MYC 8q24 8q24 rearrangement
Ewing sarcoma EWSR1 22q12 22q12/t(11;22)(q24;q12)
rearrangement
Oligodendroglioma LSI 1p36 (TP73)/ 1p25(ABL2) 1p36/1p25/19q13/19p13 1p36 and 19q13 deletions
LSI 19q13(ZNF)/19p13(CRX)
Gliomas PTEN 10q23 10q23 deletion
Synovial sarcoma SYT 18q11.2 18q11.2 rearrangement
Myxoid/round cell liposarcoma CHOP 12q13 12q13 rearrangement
Alveolar rhabdomyosarcoma FKHR 13q14 13q14 rearrangement
Fig. 1. (a) Karyotype of patient with AML showing 42–44,XX,del(5)(q22q34),+8,-8,-11,-15,der(16) hsr(11;16)(q23;p13.3)

x1–3,-17,der(17)t(15;17)(q10;q10),+1–2mar(cp). (b) Amplification of MLL to homogeneously staining region (hsr) on chro-
mosome 11 attached to chromosome 16 (See karyotype above) in patient with AML.
In order for an aCGH result to be validated for clinical use, it is

necessary in NY State to confirm the sequence found altered by
aCGH using either FISH hybridization of probes to the same
altered sequence or high resolution chromosome analyses to detect
the possible presence of translocations and inversions. At the pres-
ent time, it is possible to apply FISH with aCGH in postnatal
genetic disorders, such as autism or developmental delay (Fig. 2a,
b). To date, only a few laboratories in NY State have been approved
to offer aCGH for clinical diagnosis of these postnatal genetic dis-
orders. However, clinical diagnosis of hematopoietic and solid
tumor disorders is still limited to the combinations of chromosome
and FISH panels. The use of these techniques in combination with
aCGH has been performed on an investigational basis only and not
for clinical diagnosis and have resulted in some new information
for each malignancy. How the data of such studies is related to the
complex process of malignant development is not yet fully under-
stood and additional research is necessary before aCGH is used
routinely for clinical diagnosis of malignancies.
Interpretation of data produced by aCGH and by FISH analy-
ses must be validated with respect to accuracy. This requires screen-
ing a diverse variety of patient samples with the same DNA probe
or in the case of aCGH, a specific set of probes, and getting results
which are essentially similar in repeat experiments.
The methodology below encompasses the variety of FISH
analyses available today. The protocols are designed to optimize
generating successful results with different types of DNA probes.
In conjunction with this, it is essential for each laboratory to vali-
date those DNA probes used for clinical diagnosis which are not
FDA approved, in order to establish cutoffs for defining the
specificity and sensitivity of the probes in different patient samples.
70 L.A. Cannizzaro
Fig. 2. (a) FISH of BAC clone, RP11-279 M12-R specific to the 16p11.2 region to confirm the deletion found in patient by
aCGH analysis. Green signal is a normal control DNA probe for chromosome 16, and red signal is the BAC clone which
spans the deleted region at 16p11.2. (b) Reverse G banded metaphase showing identity of all chromosomes including the
normal and deleted chromosomes 16.
Such quantitation is essential in establishing the accuracy of each

DNA probe and its effectiveness in detecting an abnormality asso-
ciated with genetic and malignant diseases.
2. Materials
2.1. Chromosome: Cell 1. RPMI medium supplemented with 15% heat-inactivated fetal
Preparations bovine serum (Gibco).
2. 200 mM Glutamine (100×).
3. 5,000 mg/ml solution of Penicillin or streptomycin.
4. 10 μg/ml Colcemid.
5. 0.075 M KCl.
6. 3:1 Methanol–glacial acetic acid.
2.2. Fluorescent In Situ 1. 100% Ethanol (EtOH).

Hybridization 2. Purified water (distilled or deionized).
Reagents
3. 12 N HCl (for adjusting pH of wash solutions).
4. 1 N NaOH (for adjusting pH of wash solutions).
5. 20× SSC (500 g).
6. NP-40.
7. DAPI I counterstain.
8. DAPI II counterstain.
9. Formamide, ultrapure grade.
10. Methanol (for slide preparation).

11. Purified water.
12. Water baths (37°C and 72°C).
2.3. Preparation 1. 2× SSC/0.1% NP-40: Add 100 ml 20× SSC (pH 5.3) to
of Saline, Buffers, 850 ml purified H2O. Add 1.0 ml NP-40. Adjust pH to 7.0–
Alcohol for 7.5 with NaOH. Add H2O to bring final volume of the solu-
Hybridizations tion to 1 L. Store up to 6 months at room temperature.
and Washes 2. 0.4× SSC/0.3% NP-40 wash solution: Mix thoroughly 20 ml
of 20× SSC with 950 ml purified H2O. Add 3 ml NP-40. Mix
thoroughly until NP-40 is dissolved. Adjust pH to 7.0–7.5
with NaOH. Add purified H2O to bring final volume to 1 L.
Store at ambient temperature. Discard the stock solution after
6 months, or if solution appears cloudy or contaminated.
3. 20× SSC, pH 5.3: Add 132 g 20× SSC to 400 ml H2O and mix
thoroughly. Adjust pH at room temperature with a pH meter
to 5.3 using concentrated HCl and adjust to final volume of
500 ml. Filter through a 0.45 micron pore filtration unit. Store
up to 6 months at room temperature.
4. Denaturing solution: Add 49 ml formamide, 7 ml 20× SSC
(pH 5.3), and 14 ml purified H2O to a glass coplin jar and mix
thoroughly. Measure pH at room temperature to verify pH is
between 7.0 and 8.0. Use each batch of denaturant for 7 days
and then discard. Between periods of use, store at 4°C.
5. Ethanol wash solutions: For final concentrations of 70%, 85%,
and 100%. Prepare v/v dilutions of 100% ethanol with H2O.
Use dilutions for up to 7 days and then discard. If solution
evaporates or becomes diluted, replace with fresh solution.
Between periods of use, store at room temperature.
2.4. HER2/neu 1. Paraffin Pretreatment Kit I (32-801200).

Pretreatment 2. FDA approved HER2 DNA Probe Kit: 35-161060 (50
Reagents Assays).
3. Denaturation solution (70% Formamide/2×SSC): 49 ml forma-
mide, 7 ml 20×SSC and 14 ml purified water in a glass coplin jar,
adjust to pH 7.0–8.0, store at 2–8°C. Discard after 7 days.
4. Post-hybridization washing solution (0.4×SSC/0.3% NP-40
solution).
5. 20× SSC Solution.
2.5. Nick Translation 1. Bovine serum albumin (BSA).

and Labeling 2. dNTP mix: 100 mM dATP, dCTP, and dGTP 5 μl of each f.c.
Fluorochrome for (0.5 mM), 100 mM dTTP 1 μl f.c. (0.1 mM), Sterile water
aCGH Confirmation 984 μl for a total of 1,000 μl. Final concentrations equal
0.5 mM each of dATP, dCTP, and dGTP, and 0.05 mM dTTP.
Aliquot and store at −20°C.
72 L.A. Cannizzaro
3. DNase I stock solution, 1 mg/ml: DNase I 10 mg, NaCl, 1 M

1.5 ml f.c. (0.15 M), Glycerol 5 ml f.c. (50%), and sterile water
to bring up the volume to 10 ml. Aliquot and store at −20°C.
4. dUTP (conjugated to hapten or fluorochrome of choice).
5. EDTA, 0.5 M.
6. Glycerol.
7. Lambda HindIII DNA marker.
8. Magnesium chloride (MgCl2), 0.5 M.
9. 0.1 M ß-Mercaptoethanol: 99% solution (14.4 M) 34.7 μl plus
sterile water to bring up the volume to 5 ml. Aliquot and store
at −20°C.
10. Polymerase (Kornberg).
11. NaCl, 1 M.
12. Tris–HCl, 1 M, pH 8.0.
13. Water, sterile.
14. 10× NT-Buffer: Tris–HCl, 1 M, pH 8.0 500 μl f.c. (0.5 M),
MgCl2, 0.5 M 100 μl f.c. (50 mM), BSA, 10 mg/ml 50 μl f.c.
(0.5 mg.ml), and sterile water 350 μl for a total volume of
1,000 μl. Aliquot and store at −20°C.
3. Methods
3.1. Chromosome 1. Chromosome preparations for FISH analysis (see Note 1) can
Preparations be used from any type of tissue, with modifications dependent
on the growth characteristics of the cell type. Cells grown in
suspension, such as peripheral blood lymphocytes, bone mar-
row, and lymphoblasts, require minimal pretreatment and
incubation with colcemid, whereas preparations from solid tis-
sues, such as tumor biopsy specimens and fibroblast/epithelial
cell lines, require longer incubation times in colcemid.
2. Optimal growth and division of cells for suspension cultures is
obtained if the cells are split and fed 24 h prior to harvesting.
In the case where cells are attached to the flask, optimal chro-
mosome preparations are obtained after cells are trypsinized,
split, and fed 48 h before harvesting for chromosomes.
3.2. Cell Harvesting 1. Preincubate cells with colcemid at a final concentration of

0.02 mg/ml for 20 min to several hours depending on the cell
type.
2. At the end of the incubation, centrifuge cells at 1,000 × g for
10 min. Discard supernatant, then add approximately 5 ml of
0.75 M KCI. Leave at room temperature for 15 min, then add
1 ml of 3:1 methanol–glacial acetic acid fixative. Centrifuge

again at the same speed for 10 min.
3. Remove supernatant, then add 5–10 ml of the fixative solu-
tion. Leave at room temperature for 20 min to 1 h. Spin down
in the same manner, wash with about 5 ml fixative at least one
to two more times, then centrifuge. Remove supernatant and
place final cell suspension in a small volume (from 0.5 to 1 ml)
of 3:1 fixative.
4. Place several drops of the cell suspension on cold slides pre-
wetted using distilled water. Air-dry or leave on a warm plate
at 56°C until dry.
3.3. Fluorescent In Situ This procedure has been standardized for all types of DNA probes,
Hybridization including those used to detect microdeletions (unique), whole
chromosome painting (WCP) analyses, and FISH panel analyses
for detecting alterations in neoplastic tissues. DNA probes are
placed on slides containing metaphases and interphase cells from
all tissue types, which include: peripheral blood, bone marrow,
solid tissues, amniotic fluid, chorionic villi, lymph node and solid
tumors. WCP probes are used to detect chromosome abnormali-
ties only in the metaphase stage in these tissues.
Control loci (internal or external) are used for each FISH probe
analysis. When normal chromosome targets are expected to be
present within a sample, an internal control for the target should
be used during each hybridization, i.e., another sample that is
known to be normal for the probe target is run in parallel with the
patient sample.
3.3.1. Slide Pretreatment 1. Allow slide(s) to completely dry at room temperature.

2. Immerse slide(s) in 2× SSC for at least 30 min at 37 ± 1°C.
3. Dehydrate slide(s) by immersing in 70% ethanol solution at
room temperature for 1 min. Repeat with 85% ethanol for
1 min, followed by 100% ethanol for 1 min.
4. Dry the slides at room temperature (RT).
3.3.2. Probe Preparation 1. Probe mixture: mix 7 μL of hybridization buffer, 1 μL DNA

probe, and 2 μL purified H2O at room temperature. Centrifuge
for 1–3 s at maximum speed in a microfuge, vortex and then
centrifuge again.
2. Denature the probe in a 73°C water bath for 5 min. Place on
slide warmer set to 45–50°C.
3.3.3. Slide Preparation 1. Mark hybridization area with a diamond tipped scribe.
2. Denature slides by preparing one coplin jar with denaturant
solution (70% formamide/2× SSC) and place into the 73 ± 1°C
water bath for at least 30 min. Immerse the slides in the
74 L.A. Cannizzaro
73 ± 1°C denaturant solution (70% formamide/2× SSC) for

5 min. No more than four slides at one time (to prevent
significant temperature change).
3. Dehydrate slides in 70%, 85% and 100% cold EtOH for 1 min
each.
4. Dry slides and place on a 45–50°C slide warmer for 2 min.
3.3.4. Hybridization 1. Apply 10 μL of denatured probe mixture to each slide. Apply

coverslip immediately upon placing the probe on the slide, and
seal coverslip with rubber cement.
2. Place slide in a pre-warmed humidified box and allow hybrid-
ization to proceed overnight for 12–16 h in a 37°C incubator
(see Note 2).
3.3.5. Post-hybridization 1. Prepare one coplin jar with 0.4× SSC/0.3%NP-40 wash and
Wash place into the 73 ± 1°C water bath for at least 30 min. Prepare
a second coplin jar of 2× SSC/0.1% NP-40 at room
temperature.
2. Remove rubber cement seal and the coverslip and immediately
place slide into wash coplin jar (0.4× SSC/0.3%NP-40),
agitating the slide for 1–3 s. Repeat to a maximum of four
slides, and then leave all slides in the coplin jar 73 ± 1°C for 2 min.
Do not remove the coverslips from slides before placing any of
the slides in the wash bath. Begin timing the incubation when
the last slide has been added to the wash bath (see Note 3).
3. Wash slide in 2× SSC/0.1% NP-40 at room temperature for 5 s
to 1 min, agitating for 1–3 s as the slide is placed in the bath.
4. Allow slide to air-dry in darkness.
5. Apply 10 μL DAPI II counterstain to the target area of slide and
add coverslip. Slides are viewed using a suitable filter set. Store
slides at −20°C in the dark (they can be stored this way for a few
days if needed, but not for an indefinite period of time).
3.3.6. FISH Analysis/ For interphase studies, 200–500 cells are analyzed depending on
Interpretation which probe is used; when probes are hybridized to chromosome
preparations, a minimum of 20 metaphases are evaluated. At least
two electronic images are captured to document the hybridization
signal results of each probe (see Note 4) and stored in the Image
Analysis System (Metasystem Inc., USA).
3.4. HER2/neu FISH 1. Breast tumor specimens must be fixed in formalin a minimum
Analyses for Breast of 6 h and a maximum 48 h before histologic processing to
Tumors insure accurate determination of Her2/neu protein and/or
DNA levels;
3.4.1. Tumor Tissue
Fixation (Applied to All 2. The dates and times of placement of tissue into formalin and
Tumors and Solid Tissues) into the tissue processor must be included in the gross descrip-
tion of the tumor tissue sample.
3. To facilitate analysis, Hematoxylin and Eosin (H & E) stained

slide sections are included with the sample to ensure the pres-
ence of tumor cells on the slides.
4. At least five slides of paraffin sections from breast tissue, pre-
pared by the histology laboratory in advance, are baked at
56°C overnight.
3.4.2. Deparaffinizing This procedure is used for all slides with paraffin irrespective of the
Slides DNA probes used for analyses (see Notes 5 and 6):
1. Immerse Slides in Hemo-De (derived from d-Limonene it is a
superior solvent and cleaning agent) for 10 min at ambient
temperature.
2. Repeat twice using new Hemo-De each time.
3. Dehydrate slides in 100% EtOH for 5 min at ambient tempera-
ture in a coplin jar.
4. Repeat step 3.
5. Air-dry slides or place slides on a 45–50°C slide warmer for
2–5 min.
3.4.3. Pretreating Slides:
1. Immerse slides in 0.2 N HCl for 20 min.
Follow Paraffin
Pretreatment Kit 2. Immerse slides in purified water for 3 min.
3. Immerse slides in Wash Buffer for 3 min.
4. Immerse slides in Pretreatment Solution at 80°C for 30 min.
5. Immerse slides in purified water for 1 min.
6. Immerse slides in Wash Buffer for 5 min.
7. Repeat using the second jar of Wash Buffer.
3.4.4. Treating Slides 1. Remove slides from the second jar of wash buffer and
with Protease remove excess buffer by blotting the edges of the slides on
a paper towel.
2. Immerse slides in Protease solution at 37°C for 10 min.
3. Immerse slides in Wash Buffer for 5 min. Repeat using the
second jar of Wash Buffer.
4. Dry slides on a 45–50°C slide warmer for 2–5 min.
3.4.5. Fixing the Specimen 1. Immerse the slides in 10% buffered formalin at ambient tem-
perature for 10 min.
2. Immerse the slides in Wash Buffer for 5 min. Repeat using the
second jar of Wash Buffer.
3. Dry slides on a 45–50°C slide warmer for 2–5 min. Proceed
with the previous probe protocol for hybridization/washing
(Subheadings 3.3.4 and 3.3.5).
76 L.A. Cannizzaro
3.4.6. HER2/neu 1. Review corresponding H & E stained slide to localize the inva-
Interpretation sive cancer areas.
2. Review corresponding control slides.
3. Nuclei of one color or with no hybridization signal are not
scored.
4. Score only those nuclei with one or more signals of each color.
5. Count at least 20 nonoverlapping cells in two separate areas of
invasive cancer.
6. Determine HER2/CEP17 signal ratio using the following
criteria:
Positive: HER2/CEP17 signal ratio > 2.2
Negative: HER2/CEP17 signal ratio < 1.8
Equivocal: HER2/CEP17 signal ratio between 1.8 and 2.2
3.4.7. HER2/neuTEST 1. Sample with only limited number of invasive cancer cells.
Failure Is Determined 2. Tissue fixed in fixatives other than buffered formalin.
by Following Criteria
3. Control with unexpected results.
4. FISH signals not uniformly detected in cancer cells.
5. Background obscures signals (>10% of signals over cytoplasm).
6. Not optimal enzymatic digestion (poor nuclear resolution).
3.5. Biliary Brush 1. Brushing samples taken from biliary strictures are placed in a
Sample Collection container containing 15 ml of Saccomanno fixative, (Cytology
and Transport Solution: Reagent Alcohol 50%, Polyethylene glycol 2%). The
(Cholangiocarcinoma fixed sample should be sent to the lab within 24 h of collection
FISH) without exposure to extreme temperatures (low and high
temperatures).
3.5.1. Biliary Brush Sample 1. Remove the brush from the container, place in a petri dish with
Processing some of the same solution from the container, and scrape the
brush with a scalpel blade. Rinse the petri dish with some of
the solution from the container, and return everything to the
original container.
2. Pour the entire contents of the container into a 50 ml test tube
and centrifuge at 1,200 × g for 8 min at room temperature
(15–30°C). Remove supernatant, leaving 1–2 ml of solution
with pellet.
3. Suspend the pellet in the remaining 1–2 ml of supernatant and
transfer the contents to a 15 ml conical centrifuge tube. Rinse the
50 ml tube with 10 ml of 1× PBS and add it to the 15 ml tube.
4. Centrifuge sample(s) at 1,200 × g for 8 min at room
temperature.
5. Remove the supernatant to within approximately 0.5 ml of the
cell pellet.
6. Suspend pellet in the remaining 0.5 ml of supernatant. Slowly

add 5 ml of fresh fixative (3:1, methanol–acetic acid), dropwise
at first, with frequent agitation. Centrifuge the sample at
600 × g for 10 min. Repeat this step depending on the size of
the pellet.
7. Remove most of the fix and drop onto slides. Check density of
cells under phase contrast microscope. Optimal cell density for
FISH analysis is when cells are not overlapping, but numerous.
8. Hybridization probes used for this analysis are the same as
those for the urovysion FISH (UFISH) procedure for bladder
cancer (see below and Table 3).
9. Interpretation of results is based on criteria established by the
Manufacturer.
3.6. Urovysion FISH 1. Centrifuge urine in a 50 ml centrifuge tube at 600 × g for

Procedure for Bladder 10 min at room temperature (15–30°C).
Cancer 2. Remove the supernatant to within approximately 1–2 ml of the
3.6.1. Urine Sample cell pellet, being careful not to disturb the pellet.
Processing 3. Suspend the pellet in the remaining 1–2 ml of supernatant and
transfer the contents to a 15 ml conical centrifuge tube. Rinse
the 50 ml tube with 10 ml of 1× PBS and transfer the contents
to a 15 ml tube.
4. Centrifuge sample(s) at 600 × g for 10 min at room
temperature.
5. Remove the supernatant to within approximately 0.5 ml of the
cell pellet.
6. Suspend the pellet in the remaining 0.5 ml of supernatant.
Slowly add 1–5 ml of fresh fixative (3:1 Methanol–Acetic Acid),
dropwise at first, with frequent agitation.
7. Let fixed specimens stand at −20°C for a minimum of 30 min.
8. Centrifuge sample(s) at 600 × g for 5 min at room temperature.
Carefully remove the supernatant.
9. Wash pellet by suspending in 1–5 ml fixative.
10. Centrifuge sample(s) at 600 × g for 5 min at room temperature.
Repeat steps 8 and 9 twice.
11. After centrifugation of the cell suspension in fixative, if the cell
pellet is very small and hardly visible, carefully remove as much
fixative as possible, leaving approximately 100 μL solution. If
the cell pellet is easily visible, remove as much fixative as pos-
sible and add 0.5–1 ml fresh fixative to the cell pellet.
12. Proceed immediately with the slide pretreatment protocol.
13. Interpretation of results is based on criteria established by
manufacturer.
78 L.A. Cannizzaro
3.7. FISH with BAC DNA sequences/loci found altered by aCGH analyses are inserted
Clone to Confirm into BAC clones. The insert is sequenced to verify the identity of
Alteration Detected the DNA derived from the BAC. After the sequence is verified, the
by aCGH BAC clones are used to confirm by FISH, the alteration found by
aCGH analyses in the patient DNA sample (Fig. 3a, b).
3.7.1. Nick Translation The BAC probes are labeled with appropriate fluorophores (see
of BAC Clone Note 7):
1. For each DNA sample, add to an eppendorf tube:
2 μg DNA
10 μl 10× NT-Buffer
10 μl dNTP
10 μl 0.1 M ß-Mercaptoethanol
4 μl BIO-16-dUTP or 4 μl DIG-11-dUTP (1 mM)
X μl sterile water (The total volume including reagents added
in step 3 should be 100 μl).
2. Vortex, centrifuge, and place tubes on ice.
3. Add 2 μl Polymerase (Kornberg) first, and then 3–8 μl DNAse
(see #8 below) (1 mg/ml) diluted 1:1,000.
4. Flick tube to mix.
5. Incubate at 15°C for 1.5–2 h.
6. Prepare gel electrophoresis.
7. Run 5 μl of each sample with loading buffer and the Lambda
HindIII DNA marker; Ideally the length of the DNA should
be 500–900 bp for chromosome paint probes or 300–600 bp
for gene specific probes after nick translation.
8. If the DNA is too large, add more DNAse and incubate at
15°C for 10–30 min.
9. Stop the nick translation with 1 μl of 0.5 M EDTA and incu-
bate at 65°C for 10 min.
10. Store DNA at −20°C or precipitate the same day for hybridization.
3.8. Interpretation The BAC FISH probes are used on metaphase chromosomes using
of aCGH confirmation the procedure described in Subheading 3.3.4 above (Fig. 3a, b).
by FISH Expected hybridization pattern:
1. In a normal cell, there will be two green signals for the control
and two orange signals for the BAC FISH probe.
2. In an abnormal cell where there is a deletion, there will be two
green signals for the control and one orange signal for the BAC
FISH probe.
3. In an abnormal cell, where there is a duplication, there will be
two green signals and three or more orange signals.
Fig. 3. (a) UroVysion FISH for bladder cancer (UroVysion kit (Vysis Inc.)) CEP3-Red/CEP7-Green/CEP17-Aqua/LSIp16-
Gold(9p21) showing normal hybridization pattern or disomy for chromosomes 3, 7, 17, and no deletion of the 9p21 (p16)
region. (b) UroVysion FISH for bladder CEP3-Red/CEP7-Green/CEP17-Aqua/LSIp16-Gold(9p21) showing an abnormal hybrid-
ization pattern with trisomy or tetrasomy of chromosomes 3, 7, 17 and homozygous deletion of the 9p21 (p16) region.
Scoring criteria:
A minimum of 20 metaphases and/or 200 interphase nuclei
(for duplications) are analyzed per BAC Probe using the Axioplan
2 Imaging Microscope (Zeiss, Germany). At least five electronic
images are captured and archived using the Isis FISH analysis
software (Metasystem Inc., Germany) for each normal and/or
abnormal signal pattern.
Normal nomenclature (40):
ish 10q23.1(RP11-910C22x2),(CEP10x2)
Deletion:
ish del(10)(q23.1q23.1)(RP11-910 C22-),(CEP10x2)
Duplication:
ish dup(10)(q23.1q23.1)(RP11-910 C22++)(CEP10x2)
3.9. Validation of Fish Cells fixed in a methanol–acetic acid solution from peripheral blood
Probes of five normal males are pooled prior to chromosome slide prepa-
ration. Chromosome slides are prepared in a controlled room tem-
3.9.1. Preparation of Slides
perature (70–75°F) and relative humidity (40–45%).
3.9.2. Validation of Probe A minimum of five metaphases are scored to verify that each probe
Localization hybridizes to the expected chromosome target and not to any
other chromosome region. To determine chromosomal localiza-
tion, inverted DAPI bands are cross-checked by metaphase cap-
ture. Captured color metaphase images of the involved probes are
reversed using the Isis software (Metasystem Inc., Germany) to
80 L.A. Cannizzaro
produce black and white inverted DAPI metaphase images, while

the site of the probe retains its respective color. These inverted
DAPI images produce the standard G-banding pattern for normal
human chromosomes and facilitates probe localization on the
chromosome. Chromosomes on the metaphase are labeled with
their respective name and saved in a data base. Control probes,
such as centromere probes and/or specific loci probes serve as
internal controls that are used to determine the expected probe
localization pattern.
3.9.3. Analytical Validation Comparable analytic sensitivity and specificity is established for
of Sensitivity and each new test or new probe lot. Sensitivity is defined as the percent-
Specificity age of metaphases with the expected signal pattern at the correct chro-
mosomal location. Specificity is defined by the percentage of signals
that hybridize to the correct locus (41). Analytical specificity is deter-
mined by calculating the proportion of probe bound to the target
versus the proportion bound to other chromosome regions.
Analytical sensitivity and specificity is established by analysis of the
hybridization of the probe to chromosomes representing at least
200 distinct genomic targets. A target sequence for which the
hybridization signals from each of the chromatid are separable
would require analysis of 50 cells (4 targets per metaphase). If the
target sequence is at or near the centromere such that the hybrid-
ization signals are not clearly separable, the analysis would require
100 cells (2 targets per metaphase).
Sensitivity (%): Analyze 50 metaphases (for sex chromosomes and
DNA probes very close to the centromere, analyze 100 meta-
phases) using the Zeiss Axioplan 2 microscope.
Calculate the percentage of metaphases with the expected signals
at the correct chromosomal locations. Sensitivity should be more
than 98%.
Specificity (%): Calculate the proportions of probes bound to the
targets versus proportions bound to other chromosome regions.
Specificity should be in less than 2% of cells.
4. Notes
1. A key to consistent success with the FISH technique is to make

sure the chromosome or cell preparations are of the highest
quality. If chromosome preparations contain a significant
amount of cytoplasmic material, it will be difficult to resolve
the signal and the slide will contain significantly more back-
ground making it difficult to resolve the target signal. If there
is a lot of background signal, then this can be resolved by wash-
ing cells several additional times in 3:1 fixative to improve cell

resolution and remove excess cytoplasmic material.
2. Excess background signal is the most frequently encountered
problem with FISH. If the chromosome preparations are of
optimum quality then there are other ways to resolve the excess
background. First, try reducing the concentration of the DNA
in the hybridization mixture. One can also increase the strin-
gency of the post-hybridization washes in 50% formamide-2×
SSC by either adding several more washes or by increasing the
amount of time in each wash. The stringency can be enhanced
by increasing the amount of formamide in the solution as well.
3. If the signal is not detected after the final amplification of
fluorescent label, this can be an indication that the DNA con-
centration may not have been correctly calculated. This can be
easily resolved by reducing the stringency of the post-hybrid-
ization washes in 50% formamide-2× SSC by either reducing
the amount of time each slide is washed in the 50% forma-
mide-2× SSC, or by reducing the number of washes.
4. Panel FISH analyses require running multiple slide prepara-
tions with multiple DNA probes, sometimes with more than
one probe per slide. In this case, it is essential to accurately
label each slide with the name of the DNA probe or probes
which are being placed on that slide to avoid mix-ups and
ensure results are properly documented from that slide
preparation.
5. The most important aspect of a successful HER2/neu hybrid-
ization experiment is to make sure that all of the tissue prepara-
tions are fixed for the optimum time span (between 6 and
48 h). Otherwise, the probe will not hybridize effectively to
the tissue.
6. Slides with paraffin embedded tissue can be somewhat prob-
lematic. Due to the different tissue fixation times, as well as
varied thickness of tissue embedded in the paraffin, it becomes
necessary to vary the protease digestion times on the slides. To
ensure efficient hybridization, several slides should be run
simultaneously to determine the most effective protease diges-
tion time, one slide with more than 10 min, another slide with
less. Once the optimum digestion time is found, then the DNA
probe will hybridize more effectively and successfully generate
results.
7. Insufficient incorporation of label is a problem not only attrib-
uted to imprecise determination of the DNA concentration
but may result from an impure DNA preparation. If there is an
abundance of RNA still present in the DNA preparation, RNA
must be eliminated by appropriate purification procedures
before the DNA can be nick translated effectively.
82 L.A. Cannizzaro
References
1. Lichter P, Cremer T, Tang C-JC, Watkins PC, and permits ordering ofcosmids spanning 250
Manuelidis L (1988) Rapid detection of human kilobase pairs. Genomics 5:710–717
chromosome 21 aberrations by in situ hybridiza- 12. Lawrence JB, Singer RH, McNeil JA (1990)
tion. Proc Natl Acad Sci USA 85:9664–9668 Interphase and metaphase resolution of differ-
2. Singer RH, Lawrence JB, Villnave C (1986) ent distances within the human dystrophin
Optimization of in situ hybridization using iso- gene. Science 249:928–932
topic and non-isotopic detection methods. 13. Brandriff B, Gordon L, Trask B (1991) A new
Biotechniques 4:230 system for high-resolution DNA sequence
3. Pinkel D, Straume T, Gray JW (1986) mapping in interphase pronuclei. Genomics
Cytogenetic analysis using quantitative, high- 10:75–82
sensitivity, fluorescence hybridization. Proc 14. Lawrence JB, Singer RH (1991) Spatial orga-
Natl Acad Sci USA 83:2934–2938 nization of nucleic acid sequences within cells.
4. Tkachuk DC, Westbrook CA, Andreeff M, Sem Cell Biol 2:82–101
Donlon TA, Cleary ML, Suryanarayan K, 15. Cannizzaro LA (1991) Gene mapping in can-
Homge M, Redner A, Gray J, Pinkel D (1990) cer. Cancer Genet Cytogenet 55:139–147
Detection of bcr-abl fusion in chronic myelo- 16. Shi G, Cannizzaro LA (1996) Mapping of 29
geneous leukemia by in situ hybridization. YAC clones and identification of 3 YACs span-
Science 250:559–562 ning the translocation t(3;8)(p14.2;q24.1)
5. Cannizzaro LA (1997) Special techniques in breakpoint at 8q24.1 in hereditary renal cell
cytogenetics. In: Wolman SR, Sell S (eds) carcinoma. Cytogenet Cell Genet 75:180–185
Cytogenetic markers of human cancer. 17. Engel H, Drach J, Keyhani A, Jiang S, Van NT,
Humana, NY pp 461–477 Kimmel M, Sanchez-Williams G, Coodacre A,
6. Cannizzaro LA, Shi G (1997) Fluorescent in Andreef M (1999) Quantitation of minimal
situ hybridization. In: Walker J, Pollard J (eds) residual disease in acute myelogenous leukemia
Animal cell culture. Methods in molecular biol- and myelodysplastic syndromes in complete
ogy, vol 25. Humana, NY pp 313–322 remission by molecular cytogenetics of pro-
7. Poddighe EJ, Moesker O, Smeets D, Awwad grenitor cells. Leukemia 13(4):568–577
BH, Ramackers FCS, Hopman AHN (1993) 18. Tanaka K, Arif M, Eguchi M, Shintani T,
Interphase cytogenetics of hematological can- Kumaravel TS, Asaoku H, Kyo T, Dohy H,
cer: comparison of classical karyotyping and in Kamada N (1999) Interphase fluorescence in
situ hybridization using a panel of eleven chro- situ hybridization overcomes pitfalls of
mosome specific DNA probes. Cancer Res G-banding analysis with special reference to
51:1959–1967 underestimation of chromosome aberration
8. Xing Y, Johnson CV, Dobner ER, Lawrence JB rates. Cancer Genet Cytogenet 115(1):32–38
(1993) Higher level organization of individual 19. Rigolin GM, Bigoni R, Milani R, Cavazzini F,
gene transcription and RNA splicing. Science Roberti MG, Bardi A, Agostini P, Della Porta
259:1326–1335 M, Teighi A, Piva N, Cuneo A, Castoldi G
9. Rupa DS, Hasegawa L, Eastmond DA (1995) (2001) Clinical importance of interphase cyto-
Detection of chromosomal breakage in the genetics detecting occult chromosome lesions
lcen-lql2 region of interphase human lympho- in myelodysplastic syndromes with normal
cytes using multicolor fluorescence in situ karyotype. Leukemia 15(12):1841–1847
hybridization with tandem DNA probes. 20. Ishikawa M, Yagasaki F, Okamura D, Maeda T,
Cancer Res 55:640–645 Sugahara Y, Jinnai I, Bessho M (2007) A novel
10. Ariyama T, Inazawa J, Ezaki T, Nakamura Y, gene, ANKRD28 on 3p25, is fused with
Horii A, Abe T (1995) High-resolution cyto- NUP98 on 11p15 in a cryptic 3-way transloca-
genetic mapping of the short arm of chromo- tion of t(3;5;11)(p25;q35;p15) in an adult
some 1 with newly isolated 41 l cosmid markers patient with myelodysplastic syndrome/acute
by fluorescence in situ hybridization: the pre- myelogenous leukemia. Int J Hematol
cise order of 18 markers on lp36.1 on prophase 86(3):238–245
chromosomes and “stretched” DNAs. 21. Barouk-Simonet E, Soenen-Cornu V, Roumier
Genomics 25:114–123 C, Cosson A, Lai JL, Fenaux P, Preudhomme
11. Trask B, Pinkel D, Van Den Engh G (1989) C (2005) Role of multiplex FISH in identify-
The proximity of DNA sequences in interphase ing chromosome involvement in myelodysplas-
cell nuclei is correlated to genomic distance tic syndromes and acute myeloid leukemias
with complex karyotypes: a report of 28 cases. in hematologic malignancies. Cancer Genet

Cancer Genet Cytogenet 157(2):118–126 Cytogenet 189(1):24–28
22. Papenhausen PR, Griffin S, Tepperberg J 31. Maitta RW, Cannizzaro LA, Ramesh KH
(2005) Oncogene amplification in transform- (2009) Association of MLL amplification with
ing myelodysplasia. Exp Mol Pathol poor outcome in acute myeloid leukemia.
79(2):168–175 Cancer Genet Cytogenet 192:40–43
23. Phan CL, Megat Baharuddin PJ, Chin LP, 32. Cox MC, Panetta P, Venditti A, Del Poeta G,
Zakaria Z, Yegappan S, Sathar J, Tan SM, Franchi A, Buccisano F, Tamburini A, Maurillo
Purushothaman V, Chang KM (2008) L, Amadori S (2003) Comparison between
Amplification of BCR-ABL and t(3;21) in a conventional banding analysis and FISH
patient with blast crisis of chronic myelogenous screening with an AML-specific set of probes in
leukemia. Cancer Genet Cytogenet 260 patients. Hematol J 4(4):263–270
180(1):60–64 33. Italiano A, Attias R, Aurias A, Perot G, Burel-
24. Lugthart S, van Drunen E, van Norden Y, van Vandenbos F, Otto J, Venissac N, Pedeutour F
Hoven A, Erpelinch CA, Valk PJ, Beverloo (2006) Molecular cytogenetic characterization
HB, Lowenberg B, Delwel R (2008) High of a metastatic lung sarcomatoid carcinoma:
EVI1 levels predict adverse outcome in acute 9p23 neocentromere and 9p23-p24
myeloid leukemia: prevalence of EVI1 overex- amplification including JAK2 and JMJD2C.
pression and chromosome 3q26 abnormalities Cancer Genet Cytogenet 167(2):122–130
underestimated. Blood 111(8):4329–4337 34. Fenaux P (2001) Chromosome and molecular
25. Mrozek K (2008) Cytogenetic, molecular abnormalities in myelodysplastic syndromes.
genetic, and clinical characteristics of acute Int J Hematol 73(4):429–437
myeloid leukemia with a complex karyotype. 35. Cheng L, Ramesh KH, Wei D, Ratech H,
Simin Oncol 35(4):365–377 Radel E, Cannizzaro LA (2001) t(11;19)
26. Schoch C, Haferlach T, Bursch S, Gerstner D, (q23;p13.3) rearrangement in a patient with
Schnittger S, Dugas M, Kern W, Loffler H, therapy related acute myeloid leukemia. Cancer
Hiddemann W (2002) Loss of genetic material Genet Cytogenet 129:17–22
is more common than gain in acute myeloid 36. Castuma MV, Rao PH, Acevedo SH, Larripa
leukemia with complex aberrant karyotype: a IB (2000) Comparative genomic hybridization
detailed analysis of 125 cases using conven- study of de novo myeloid neoplasia. Acta
tional chromosome analysis and fluorescence in Haematol 104(1):25–30
situ hybridization including 24-color FISH.
Genes Chromosomes Cancer 35(1):20–29 37. Baloglu H, Cannizzaro LA, Jones J, Koss LG
(2001) Atypyical endometrial hyperplasia
27. Frohling S, Kayser S, Mayer C, Miller S,
shares genomic abnormalities with endometri-
Wieland C, Skelin S, Schlenk RF, Dohner H,
oid carcinoma by comparative genomic hybrid-
Dohner K (2005) Diagnostic value of
ization. Hum Pathol 32:615–622
fluorescence in situ hybridization for the detec-
tion of genomic aberrations in older patients 38. Suela J, Alvarez S, Cigudosa JC (2007) DNA
with acute myeloid leukemia. Haematologica profiling by arrayCGH in acute myeloid leuke-
90(2):194–199 mia and myelodysplastic syndromes. Cytogenet
Genome Res 118(2–4):304–309
28. Ohsaka A, Otsubo K, Yokota H, Hisa T, Saito
H, Kozaki T (2008) Spectral karyotyping and 39. Evers C, Beier M, Poelitz A, Hildebrandt B,
fluorescence in situ hybridization analyses Servan K, Drechsler M, Germing U, Royer HD,
identified a novel three-way translocation Royer-Pokora B (2007) Molecular definition of
involving inversion 16 in therapy-related acute chromosome arm 5q deletion end points and
myeloid leukemia M4eo. Cancer Genet detection of hidden aberrations in patients with
Cytogenet 184(2):113–118 myelodysplastic syndromes and isolated del(5q)
29. Seller MJ, Bint S, Kavalier F, Brown RN, using oligonucleotide array CGH. Genes
Ogilvie CM (2006) Multicolor banding detects Chromosomes Cancer 46(12):1119–1128
a complex three chromosome, seven break- 40. Shaffer LG, Tommerup N (eds) (2005) ISCN
point unbalanced rearrangement in an ICSI- 2005: an international system for human cyto-
derived fetus with multiple abnormalities. Am J genetic nomenclature. S. Karger, Basel
Med Genet A 140(10):1102–1107 41. American College of Medical Genetics,
30. Moosavi SA, Sanchez J, Adeyinka A (2009) Standards and Guidelines for Clinical Genetics
Marker chromosomes are a significant mecha- Laboratories. 2006 Edition. 9. E. http://www.
nism of high-level RUNX1 gene amplification acmg.net/
Chapter 6
The Development of T Lymphocytes in Fetal Thymus

Organ Culture
Takeshi Nitta, Izumi Ohigashi, and Yousuke Takahama
Abstract
Fetal thymus organ culture (FTOC) is a unique and powerful culture system that allows intrathymic
T-lymphocyte development in vitro. T-cell development in FTOC well represents fetal thymocyte develop-
ment in vivo. Here we describe the basic method for FTOC as well as several related techniques, including
reconstitution of thymus lobes with T-lymphoid progenitor cells, high-oxygen submersion culture,
reaggregation thymus organ culture, retrovirus-mediated gene transfer to developing thymocytes in
FTOC, and coculture of progenitor cells with OP9-DL1 cells.
Key words: T lymphocyte, Fetal thymus, Organ culture, Retrovirus, Flow cytometry
1. Introduction
Among the various lineages of hematopoietic cells, T lymphocytes

are the only cells whose development requires the environment of
the thymus in addition to the bone marrow or the fetal liver. Recent
studies have identified several molecules that take part in specifying
multiple thymic microenvironments. Those molecules include
IL-7, DLL-4, β5t-containing thymoproteasome, Aire, CCL21,
and class I/class II MHC molecules. Despite their identification,
however, it is still unclear whether any combination of the known
molecules is sufficient to represent the thymic microenvironments
that support T-cell development and repertoire selection. Thus,
the use of isolated thymic stromal cells in organ culture provides
the most reliable and reproducible condition that supports the
development and selection of T cells in the thymus.
85
86 T. Nitta et al.
The analysis of T-cell development in organ culture of mouse

fetal thymus was first established by Owen (1, 2) and Mandel
(3, 4) and later refined largely by Owen’s group (5, 6). The fetal
thymus organ culture (FTOC) technique offers a unique in vitro
cell culture system in that functional T-cells are differentiated from
immature progenitor cells. As such, T-cell development in FTOC
closely reflects T-cell development during fetal ontogeny, even
with respect to the time course of differentiation (7, 8). FTOC
allows the addition of various reagents, such as chemicals, antibodies,
and viruses, to examine their effects on T-cell development.
This chapter describes a basic method for FTOC
(Subheadings 3.1–3.4) and several related techniques, including
the reconstitution of thymus lobes with T-lymphoid progenitor
cells (Subheading 3.5), high-oxygen submersion culture
(Subheading 3.6), reaggregation thymus organ culture
(Subheading 3.7), retrovirus-mediated gene transfer to developing
thymocytes in FTOC (Subheading 3.8), and coculture with
OP9-DL1 cells (Subheading 3.9).
2. Materials
2.1. Isolation 1. Timed pregnant C57BL/6 mice: Mice should be mated in an

of Fetuses from animal facility according to institutional guidelines. We usually
Pregnant Mice place two female mice and one male mouse in a cage in the
evening (7–8 p.m.), and separate them in the morning
(8–9 a.m.). Gestational age is tentatively designated by assign-
ing the day when mice are separated as E0.5, and is confirmed
on the day of experiment according to the sizes and develop-
mental features of fetuses (see Note 1 and refs. 9–11).
2. Regular dissecting forceps and scissors. At least one set for
non-sterile dissection of skin and two to three autoclaved sets
for sterile use should be prepared.
2.2. Preparation 1. Sterile Gelfoam gelatin sponges (Pfizer). Cut into small pieces
of Culture Wells (e.g., 1-cm square) and store dry at room temperature.
2. Polycarbonate (PC) filter membranes (Whatman, cat. no.
110409, 13 mm diameter). Autoclave to sterilize and store dry
at room temperature.
3. 24-Well plates (16 mm diameter, sterile).
4. Culture medium: RPMI1640 supplemented with 10% fetal calf
serum (FCS), 50 μM 2-mercaptoethanol, 10 mM HEPES,
2 mM L-glutamine, 1× nonessential amino acids, 1 mM sodium
pyruvate, 100 U/ml penicillin, and 100 μg/ml streptomycin.
6 T-cell Development in FTOC 87
All medium components except 2-mercaptoethanol are

purchased from Invitrogen. 2-Mercaptoethanol is purchased
from Sigma. FCS is pretreated for 30 min at 56°C and stored
frozen in 50-ml aliquots. Screening of FCS is essential (see
Note 2).
2.3. Isolation 1. Fetuses from timed pregnant mice (refer to Subheading 2.1).
and Organ Culture 2. #7 Forceps, biology grade (e.g., Dumont, Switzerland). Stored
of Fetal Thymus Lobes sterile in 70% ethanol.
3. Dissecting microscope with zoom, e.g., 7× to 42× magnification,
preferably equipped with fiber-optic light. The microscope
should be placed in a clean hood.
4. Gauze sponges (e.g., Johnson and Johnson, 2 × 2-in. square,
6–8 ply, sterile).
5. 100-mm Sterile plastic dishes.
2.4. Isolation 1. Suspension buffer: PBS, pH 7.2 supplemented with 0.2% BSA
of Single-Cell and 0.1% NaN3.
Suspensions from 2. 1-ml Syringes.
Fetal Thymus Organ
3. 26-Gauge needles.
Culture
4. 30-mm Plastic dishes.
5. Nylon mesh (approximately 300 meshes/sq. in.). Cut into
small pieces of approximately 5-mm square.
2.5. Optional 1. 2-Deoxyguanosine (dGuo, e.g., Sigma, cat. no. D7145).

Technique: Hanging- Aliquots of a stock solution at 13.5 mM in PBS are stored fro-
Drop Reconstitution zen at −20°C and can be thawed at 37°C.
of Deoxyguanosine- 2. Terasaki 60-well plates (sterile).
Treated Thymus Lobes
with T-Lymphoid
Progenitor Cells
2.6. Optional 1. 96-Well round-bottom plates (sterile).

Technique: High- 2. Plastic bags (3–5 L) and a heat-sealer.
Oxygen Submersion
3. Gas consisting of 70% O2, 25% N2, and 5% CO2.
Culture of Fetal
Thymus Lobes
2.7. Optional 1. Trypsin (0.5%)/EDTA (5.3 mM) solution (e.g., Invitrogen).

Technique: 2. Ca2+-free Mg2+-free PBS.
Reaggregation
3. 1-ml Syringes (sterile).
Thymus Organ Culture
4. 26-Gauge needles (sterile).
88 T. Nitta et al.
2.8. Optional 1. 10-ml Syringes (sterile).

Technique: Retroviral 2. Syringe-driven filter (0.22 μm pore size, 16 mm diameter,
Gene Transfer sterile).
into Developing
3. Parafilm.
Thymocytes for Fetal
Thymus Organ Culture 4. Plat-E cells (12) and a retrovirus vector pMRX-IRES-EGFP
(13). The culture medium for Plat-E cells is DMEM supple-
mented with 10% FCS, 100 U/ml penicillin, 100 μg/ml strep-
tomycin, 1 μg/ml puromycin, and 10 μg/ml blasticidin S.
For transfection experiments, use the medium without puro-
mycin and blasticidin S.
5. Polybrene (hexadimethrine bromide) (e.g., Sigma).
2.9. Optional 1. Fetuses from timed pregnant mice (refer to Subheading 2.1).
Technique: T-Cell 2. Antibodies: Biotinylated antibodies specific for hematopoietic
Development in lineages (Lin) (Thy1 for T cells, B220 for B cells, CD11b
Coculture with for macrophages, NK1.1 for NK cells, and TER119 for
OP9-DL1 Cells erythrocytes). Phycoerythrin-conjugated antibody specific
2.9.1. Isolation of Lin-Kit+
for c-Kit.
Fetal Liver Cells 3. Streptavidin-conjugated microbeads (Miltenyi Biotec).
4. MACS buffer: PBS supplemented with 0.5% BSA and 2 mM
EDTA.
5. MACS LS column (Miltenyi Biotec).
6. MACS Separation unit (Miltenyi Biotec).
7. MACS Multistand (Miltenyi Biotec).
8. FACS buffer: PBS supplemented with 0.2% BSA and 0.1% NaN3.
9. 100- and 48-μm nylon mesh.
10. Cell sorter.
2.9.2. OP9-DL1 Cell 1. OP9-DL1 cells: OP9 cells retrovirally transduced to express
Culture the Notch ligand DLL-1 (14). The culture medium for
OP9-DL1 cells is α-MEM (e.g., Sigma) supplemented
with 20% FCS, 100 U/ml penicillin, and 100 μg/ml
streptomycin.
2. 100-mm Tissue culture dish.
3. 24-Well tissue culture plate.
4. Culture medium (see Subheading 2.2, item 4).
5. Flt3 ligand (e.g., PeproTech).
6. IL-7 (e.g., PeproTech).
3. Methods
3.1. Isolation 1. All procedures should be performed in sterile conditions in a

of Fetuses cell culture hood.
from Pregnant Mice 2. Prepare 100-mm sterile dishes, each containing 20–30 ml of
culture medium (a minimum of 3 dishes).
3. Sacrifice timed pregnant mice (usually at gestational age
E14.5 or E15.5) by CO2 asphyxiation according to institu-
tional guidelines.
4. Wipe abdomens with 70% ethanol and make an abdominal
incision using a non-sterile set of scissors and forceps.
5. Remove fetus-filled uteri with a sterile set of scissors and
forceps.
6. Transfer uteri to an empty 100-mm plastic dish.
7. Using a sterile set of sharp scissors and forceps, remove fetuses
from the uteri and transfer them to a new dish containing
culture medium.
8. Ascertain the gestational age of fetuses (see Note 3).
9. Wash out the blood by transferring fetuses to a new dish
containing fresh culture medium.
10. Repeat washing 2–3 times to remove the blood. Gentle swirl-
ing of the dishes helps to remove the blood and other debris.
11. Count the number of fetuses and plan the experiment. For
flow cytometry analysis, 4–6 fetal thymuses are usually used for
one group of experiments. Fetuses may be temporarily stored
in a refrigerator or on ice while preparing culture wells as
below.
3.2. Preparation 1. Cut a gelatin sponge into approximately 1 cm2 pieces using a
of Culture Wells clean set of sterile scissors and forceps.
2. Place one piece of the sponge in a culture well of a 24-well
plate.
3. Fill the culture well with 1 ml of culture medium.
4. Press the sponge with forceps so that it is wet by the culture
medium and air bubbles are completely released.
5. Place a piece of sterile PC membrane on each sponge. Flip the
membrane with forceps so that both sides are completely wet
with the culture medium.
6. Gently remove 0.5 ml of the culture medium from each well
using a 1-ml pipette. The final volume of the culture medium
is 0.5 ml per well.
90 T. Nitta et al.
3.3. Isolation 1. Place a dissecting microscope in the cell culture hood.

and Organ Culture 2. Prepare a surgery dish by wetting a 2 × 2 in. gauge sponge in a
of Fetal Thymus Lobes 100-mm dish with approximately 5 ml of culture medium.
3. Wash two sterile #7 forceps with culture medium to remove all
traces of ethanol as fetal thymocytes tend to die when exposed
to ethanol.
4. The following procedures (steps 5–9) are performed using #7
forceps under the microscope.
5. Place a fetus in the supine position in the surgery dish under
the microscope (Fig. 1a, b).
Fig. 1. Isolation of thymus lobes from fetal mice. (a) A fetus at gestational age E14.5 from
a C57BL/6 mouse is placed under a dissecting microscope. (b) The fetus is set in the
supine position so that its abdomen is facing up. (c) The neck is raised up to expose the
chest. (d) The chest is opened to expose two thymus lobes as shown by arrows. (e) High
magnification of (d). Arrows indicate two thymus lobes in the chest. (f) Isolated thymus
lobes. (g) Diagram of culture well for FTOC. Scale bar = 1 mm.
6. Raise the head (Fig. 1c).

7. Gently open the chest and locate the two lobes of the thymus
(Fig. 1d, e).
8. The thymus lobes are removed from the body by raising them
with forceps so that the whole lobes are lifted. The isolated
lobes are placed on a piece of gauze pre-wetted with culture
medium to remove blood (Fig. 1f; see Note 4).
9. Place the thymus lobes onto the filter membrane in a culture
well. Usually, 4–6 lobes are placed on each membrane (Fig. 1g).
Try to randomize the way the lobes are placed. For example,
two lobes from one fetus should be divided into different
groups when there are multiple experimental groups.
10. Ascertain that the lobes are placed at the interface between the
filter membrane and air. The lobes should not be soaked in
culture medium (see Note 5 for an alternate method describing
the addition of reagents to the cultures).
11. Add 1–2 ml of fresh culture medium to each empty well of the
24-well plate to minimize evaporation from the culture wells.
12. Place the culture plate in a 37°C, 5% CO2 incubator.
3.4. Isolation 1. Apply a drop of the suspension buffer (100 μl) at the center of
of Single-Cell the reverse side of the lid of a 30-mm dish.
Suspensions 2. Transfer thymus lobes into the drop with #7 forceps. Count
from Fetal Thymus the number of lobes.
Organ Culture
3. Place a small (approximately 5 mm × 5 mm) piece of nylon
mesh on the drop.
4. Attach 26-gauge needles to 1-ml syringes. Using forceps, bend
the tip (top 5 mm, 90° angle) of needles. Two needle/syringe
sets are needed per group.
5. Gently tease the lobes under a small piece of nylon mesh
(approximately 5-mm square) by softly pressing them with the
needles to release the thymocytes. If needed, use a dissecting
microscope.
6. Transfer the cell suspension to a plastic tube and determine the
cell number. Use the cell suspension for further examination of
T-cell development, e.g., immunofluorescence and flow cytom-
etry analysis (Fig. 2; see Notes 6–9).
3.5. Optional The hanging-drop reconstitution technique is useful for testing the
Technique: Hanging- developmental potential of T-precursor cells in fetal thymus lobes.
Drop Reconstitution T-precursor cells from a given genetic background and/or with a
of Deoxyguanosine- given gene modification can be used for the reconstitution.
Treated Thymus Lobes 1. Thymus lobes from fetal mice at gestational age E14.5 or
with T-Precursor Cells E15.5 are cultured as in Subheading 2.3 in the presence of
1.35 mM deoxyguanosine (dGuo) for 5–7 days (see Note 5).
92 T. Nitta et al.
Fig. 2. T-lymphocyte differentiation in FTOC. Contour histograms indicate CD4/CD8 two-color immunofluorescence
profiles of thymocytes generated in FTOC. E14.5 fetal thymus lobes from C57BL/6 mice were organ-cultured for the
indicated number of days. Number within a box indicates frequency of cells in that box. Cell number recovered per thymus
lobe is indicated in parentheses. The profile of cells isolated from an adult thymus is also shown.
In a typical experiment, 10–20 thymus lobes are treated with

dGuo (see Note 10).
2. Fill a 30-mm sterile dish with 3–4 ml of culture medium.
Detach individual thymus lobes from the filter membrane and
transfer them into the medium using sterile forceps and a
micropipette. Swirl the thymus lobes in the culture medium.
3. Transfer the lobes to fresh culture medium using a micropipette.
4. Diffuse away dGuo in a 37°C, 5% CO2 incubator for approxi-
mately 20 min with two additional transfers into fresh culture
medium as indicated above.
5. Transfer 15 μl of culture medium containing one dGuo-treated
thymus lobe per well of a Terasaki plate.
6. Add 20 μl of culture medium containing T-precursor cells,
e.g., 100–1,000 fetal thymocytes or 1,000–10,000 fetal liver
cells.
7. Place the lid on the plate and gently invert.
8. Ascertain that the thymus lobes are located at the bottom of
the drop. If not, gently pipette the well.
9. Culture in a 37°C, 5% CO2 incubator for 1 day.
10. Transfer the thymus lobes to a freshly prepared filter/sponge
for culture in conventional thymus organ culture conditions
(Subheading 2.3). The thymus lobes may be rinsed with fresh
culture medium in order to remove the cells that merely attach
to the surface and do not enter the thymus organ.
11. Culture in a 37°C, 5% CO2 incubator. Cultures can be evaluated

in various ways, including the determination of cell number
and the flow cytometric analysis of T-cell development. Typical
results of T-cell development in this culture method can be
found in refs. 6, 15.
3.6. Optional T-cell development in fetal thymus lobes may occur in a submer-
Technique: High- sion culture under high-oxygen pressure. The high-oxygen sub-
Oxygen Submersion mersion culture is useful for the reconstitution of the thymus lobes
Culture of Fetal using a limited number of T-precursor cells (see Note 11).
Thymus Lobes 1. Fetal thymus lobes are placed in round-bottom wells of a
96-well plate (1 lobe/well). For the reconstitution of dGuo-
treated thymus lobes, cells for the reconstitution are also
included in the culture (see Note 11).
2. Spin the plate at 150 × g for 30 s to allow the thymus lobes to
settle at the bottom of the well.
3. Place the culture plates in a plastic bag (3–5 L), fill the bag
with a gas consisting of 70% O2, 25% N2, and 5% CO2, and
heat-seal the bag.
4. Place the bag in a 37°C, 5% CO2 incubator. Cultures can be
evaluated in various ways, including the determination of cell
number and the flow cytometric analysis of T-cell develop-
ment (16).
3.7. Optional Reaggregation thymus organ culture (RTOC) provides a model in

Technique: which cellular interactions required for T-lymphocyte development
Reaggregation can be studied under controlled in vitro conditions (17). In this
Thymus Organ Culture model, the thymus lobes are depleted of endogenous T-cell pro-
genitors by treatment with dGuo (see Subheading 3.5). Surviving
stromal cells are then enzymatically dissociated to generate single-
cell suspensions. The cell slurry generated by centrifugation of a
mixture of thymocytes and stromal cells reforms a structure resem-
bling a thymus lobe (Fig. 3).
3.7.1. Preparation 1. Culture E15.5 fetal thymus lobes in the presence of 1.35 mM
of Thymic Stromal Cells dGuo for 5–7 days to deplete lymphoid elements
(Subheading 3.5; see Note 12).
2. Fill a 30-mm sterile dish with 5 ml of culture medium. Transfer
the dGuo-treated thymus lobes from the filter membrane to
the culture medium using sterile forceps and a micropipette.
3. Transfer the lobes to Ca2+-free Mg2+-free PBS with a
micropipette.
4. Diffuse away dGuo at 37°C, 5% CO2 for 20 min.
5. Harvest the thymus lobes to a 1.5-ml Eppendorf tube or a
24-well plastic well and remove the supernatant.
94 T. Nitta et al.
Fig. 3. Schematic diagram of reaggregation thymus organ culture.
6. Dissociate the thymus lobes by adding 1 ml of 0.125% trypsin–

EDTA solution in Ca2+-free Mg2+-free PBS for 30 min at 37°C,
5% CO2.
7. Stop trypsinization by adding 1 ml of FCS-containing culture
medium.
8. Disperse the stromal cells by vigorous pipetting and mechanical
agitation with a 1-ml syringe and a 26-gauge needle.
9. Pass the dispersed stromal cell suspension through a 100-μm
nylon mesh to remove the clumps.
10. Spin down the cells at 300 × g and discard the supernatant.
11. Suspend the cells in 200 μl of FCS-containing culture medium
and determine the cell number (see Note 12). If needed, the
cells can be stained with fluorescence-labeled antibodies and
sorted by flow cytometry (see Note 13).
3.7.2. Formation 1. Mix thymocyte populations of interest (see Note 14) with the
of Reaggregates dispersed stromal cells at a ratio of 1:1 to 10:1 in a sterile
1.5-ml Eppendorf tube. Typically 3–5 × 105 thymocytes mixed
with an equal number of thymic stromal cells are used.
2. Spin down the cells at 300 × g for 5 min to form a pellet.
3. Gently remove the supernatant.

4. Disperse the cell pellet into a slurry by careful mixing with a
micropipette and draw the slurry into a tip (or mix with a
vortex mixer and draw into a fine, mouth-controlled glass
capillary pipette).
5. Transfer and expel the slurry as a discrete standing drop on the
surface of a PC filter prepared for conventional FTOC condi-
tions (see Subheading 3.2). The cell “slurry” reaggregates will
reform a thymus-lobe-like structure within 12 h. Maintain the
RTOC in a 37°C, 5% CO2 incubator (see Note 15).
3.8. Optional Retroviral gene transfer into developing thymocytes in FTOC

Technique: Retroviral provides a quick and economical method (vs. germline transgen-
Gene Transfer esis) to explore gene function during T-cell development. Immature
into Developing thymocytes can be efficiently and rapidly infected with a retrovirus
Thymocytes in Fetal using the spin-fection method. Gene-transferred cells can be readily
Thymus Organ Culture detected and sorted using flow cytometry, by the co-expression of
marker proteins, such as the green fluorescent protein (GFP)
(Fig. 4). Retrovirus vectors expressing GFP along with a gene of
interest using the internal ribosomal entry site (IRES) sequence
are widely used. High-titer retrovirus can be produced by the tran-
sient transfection of packaging cells with a retroviral plasmid. Plat-E
packaging cells (12), combined with the pMRX-IRES-EGFP plasmid
vector (13), are excellent tools for the production of high-titer
retroviruses. Other virus constructs, such as pMSCV-IRES-EGFP
(18, 19), may also be used.
3.8.1. Preparation 1. Set up the Plat-E cell culture. In a 10-cm dish, 2.5 × 106 cells
of Retroviral Supernatant are seeded in 10 ml of culture medium without puromycin or
blasticidin S. Cells are cultured in a 37°C, 5% CO2 incubator
for 18–24 h.
2. Transfect Plat-E cells with retroviral plasmid DNA. To a 10-cm
dish of Plat-E cells, 30 μg of DNA is introduced by the
conventional calcium phosphate precipitation method (see
Note 16). Twelve hours after the transfection, remove the
supernatant containing precipitates, gently wash the cells with
PBS, and add 10 ml of fresh culture medium.
3. Thirty-six hours after the transfection, collect culture superna-
tant containing retroviruses. The supernatant should be filtered
through 0.2-μm syringe filters and may be stored at −80°C or
used immediately. After collecting the supernatant, the cells
can be used for further retroviral production. To do so, gently
add 10 ml of fresh culture medium to the plate and continue
culture in a 37°C, 5% CO2 incubator. Retroviral supernatants
can be collected every 12 h between 36 and 72 h after transfec-
tion (see Note 17).
96 T. Nitta et al.
Fig. 4. In vitro reconstitution of the thymus by retrovirus-infected thymocytes. (a) E14.5 fetal thymocytes were infected
with the pMRX-IRES-EGFP retrovirus and cultured in a dGuo-treated fetal thymus for the indicated number of days.
Dot plots indicate CD4/CD8 immunofluorescence profiles. (b) Total thymocytes from neonatal mice were infected with the
pMRX-IRES-EGFP retrovirus and reaggregated with thymic stromal cells. RTOC was conducted for the indicated number of
days. (c) Neonatal thymocytes in panel (b) were cultured in vitro for 24 h after infection. Histograms indicate GFP expres-
sion. The CD4/CD8 expression profiles of the GFP− and GFP+ fractions are also shown.
3.8.2. Retroviral Infection 1. For gene transfer into CD4−CD8− thymocytes, prepare a
of Thymocytes single-cell suspension of E14.5 or E15.5 mouse fetal thymo-
cytes (see Subheading 3.4). For CD4+CD8+ thymocytes, pre-
pare total thymocytes from neonatal mice (0–14 days old).
Add 500 μl of retroviral supernatant (see Note 18) and 1.2 μl
of 10 mg/ml polybrene (final concentration 20 μg/ml) into
each well of a 24-well plate containing the thymocyte suspen-
sion (1–10 × 105 cells/100 μl) in the culture medium (see
Subheading 2.2).
2. Seal the plate with parafilm and spin at 1,000 × g for 1.5 h at
30°C or at room temperature.
3. Transfer the cells into a sterile 1.5-ml microtube, spin at 400 × g
for 5 min, remove the supernatant, and suspend the cells in an
appropriate volume (e.g., 100 μl) of fresh culture medium.
4. The developmental fate of retrovirus-infected thymocytes is

assessed by transferring to FTOC (see Note 19).
5. Alternatively, infected cells can be cultured in a 37°C, 5% CO2
incubator (see Note 20).
3.9. Optimal It has been shown that the bone marrow stromal cell line OP9
Technique: T-cell ectopically expressing the Notch ligand DLL-1 loses its ability to
Development in support B-cell lymphopoiesis but acquires the capacity to induce
Coculture with the differentiation of hematopoietic progenitors into T lineage
OP9-DL1 Cells cells at least up to the CD4+CD8+ double-positive stage (14).
It was later shown that DLL-4 rather than DLL-1 expressed in
thymic epithelial cells is essential for the T-cell inducing activity of
the thymus (20, 21). Nevertheless, the OP9-DL1 coculture is a
highly useful method for the molecular and genetic analyses of
in vitro T-cell development (22–24) (Fig. 5).
3.9.1. Isolation of Lin−Kit+ 1. Prepare a cell suspension of liver from E14.5 fetal mice.
Fetal Liver Cells 2. Pass the cell suspension through a 100-μm nylon mesh to
remove clumps and determine cell number.
3. Spin down the cells at 400 × g for 5 min and remove the
supernatant.
4. Stain cells with biotinylated antibodies specific for Thy1, B220,
TER119, CD11b, and NK1.1 (10 μl each of 10 μg/ml anti-
bodies per 107 cells) for 30 min on ice.
5. Add 1 ml MACS buffer, spin down the cells at 400 × g for
5 min, and remove the supernatant (two times).
Fig. 5. T-cell development in coculture with OP9-DL1 cells. Lin-Kit+ fetal liver cells from
E14.5 C57BL/6 embryos were cocultured with OP9-DL1 cells for the indicated number of
days. Number within a box indicates frequency of cells in that box.
98 T. Nitta et al.
6. Suspend the cell pellet in the proportion of 90 μl of MACS

buffer per 1 × 107 total cells and add 10 μl of streptavidin-
coated microbeads. Incubate for 30 min at 4°C.
7. Add 1 ml MACS buffer, spin down the cells at 400 × g for
5 min, and remove the supernatant (two times).
8. Suspend cell pellet in 1 ml of MACS buffer.
9. Place the MACS separation unit on the MACS multistand.
Place the LS column in the magnetic field.
10. Rinse the column with 3 ml of MACS buffer.
11. Apply the cell suspension. After the cells have entered the col-
umn add 1 ml of MACS buffer onto the column. Collect the
unlabeled cells that pass through the column.
12. Spin down the cells at 400 × g for 5 min and remove the
supernatant.
13. Stain cells with phycoerythrin-conjugated antibody specific for
c-Kit (10 μl of 10 μg/ml antibody per 107 cells) for 30 min on
ice (as in step 4).
14. Add 1 ml FACS buffer, spin down the cells at 400 × g for 5 min,
remove the supernatant, and suspend in FACS buffer. Pass the
cell suspension through a 48-μm nylon mesh.
15. Sort c-Kit+ cells with a cell sorter.
3.9.2. OP9-DL1 Cell 1. Culture OP9-DL1 cells in a 100-mm culture dish with α-MEM
Culture culture medium. The cells should be kept in the pre-confluent
(up to 80%) condition, by the passage of one-fifth cells every
2 days.
2. One to two days prior to the seeding of Lin−Kit+ cells, culture
1.5 × 104 OP9-DL1 cells in a 24-well culture plate until approx-
imately 75% confluence.
3. Seed Lin−Kit+ cells onto the OP9-DL1 cells in cell culture
medium (see Subheading 2.2, item 4). Add Flt-3 ligand and
IL-7 solutions to a final concentration of 5 ng/ml each.
Incubate the culture at 5% CO2 and 37°C for 7 days.
4. Refresh the cocultures by transferring onto freshly prepared
OP9-DL1 cells every 4–7 days. Cells can be removed by gentle
pipetting and collected by centrifugation at 400 × g for 5 min.
4. Notes
1. Timed pregnant mice may be purchased from various mouse

suppliers. Generally, eight fetuses are expected from a pregnant
C57BL/6 mouse. As the number of fetuses may differ, it is
necessary to check the number of fetuses in each mouse strain.
If FTOC is an unfamiliar technique, preliminary organ cultures

of E15.5 fetal thymus lobes for 4–5 days are recommended.
Fetuses and fetal thymuses are easiest to handle at gestational
age E15.5.
2. It is important to screen the FCS for FTOC. We usually pre-
screen 10–20 independent lots of FCS by an overnight suspen-
sion culture of adult thymocytes followed by the determination
of cell number recovered the following morning. Five or six
FCS lots that allow close to 100% cell recovery are selected for
further screening in an actual test of T-cell development in
FTOC. Progression along the CD4/CD8 developmental
pathway yielding profiles and cell numbers as shown in Fig. 2
would be a good indication of the expected T-cell develop-
ment in culture and thus an acceptable FCS lot.
3. Fetuses with abnormal developmental features as judged by
size and other developmental signs, such as the formation of
hair follicles and crests in the limbs should be eliminated (see
refs. 9–11). Such abnormalities in the developmental stage of
the fetuses will dramatically affect the stages of T-cell develop-
ment in the thymus (see Fig. 2).
4. This technique may be difficult for beginners. Adept handling
of forceps under the microscope requires practice.
5. When reagents are added, first remove 50 μl of the culture
medium. Then, add 50 μl (1:10 volume) of 10× concentrated
reagents slowly and directly onto the lobes.
6. In order to examine T-cell development in FTOC, we gener-
ally use flow cytometry. The two-color profiles of CD4/CD8
and CD25/CD44 are commonly used.
7. The advantages of FTOC for analyzing T-cell development
include reproducibility and convenience of in vitro cultures.
The disadvantages include cell number limitation and necrotic
cell death in the middle of the thymus lobe, which is not
observed in the physiological thymus in vivo (Fig. 2, ref. 7).
8. Neonatal thymus organ culture (NTOC) is used for the analysis
of positive selection signals that induce the generation of
mature “single-positive” thymocytes (25, 26). NTOC of
0-day-old newborn thymus lobes is useful for in vitro stimula-
tion of in vivo generated CD4+CD8+ thymocytes. However, it
should be noted that unlike FTOC, total cell numbers decrease
during 4–5-day cultures in the NTOC condition (7), which
may complicate the interpretation of the results.
9. As regards dGuo treatment (27), fetal thymus lobes should be
cultured with dGuo for at least 5 days. Otherwise, residual
T-cell precursors retain their developmental potential and
undergo T-cell development. Thymus lobes cultured for
100 T. Nitta et al.
7–8 days with dGuo are still capable of supporting the T-cell
development of reconstituted precursor cells.
10. The high-oxygen submersion culture of FTOC (16) is useful
for reconstitution using a limited number of progenitor cells,
as the thymus lobes can be continuously cultured at the
bottom of round or V-shaped culture wells and the entry of
progenitor cells may occur efficiently during the culture with
the help of gravity. However, it should be noted that T-cell
development in this high-oxygen condition seems to occur
more rapidly than T-cell development in vivo or in regular
FTOC conditions.
11. To prepare thymic stromal cells for RTOC, dGuo-treated
E14.5–E15.5 fetal thymus lobes may be used. Approximately
2 × 104 thymic stromal cells can be isolated from one dGuo-
treated E15.5 thymus lobe. The cell number obtained from
one dGuo-treated E15.5 thymus lobe is approximately 1.5–2-
fold larger than the cell number from one dGuo-treated E14.5
thymus lobe.
12. The thymic stroma is made up of a number of stromal cell
types. To study the interactions between thymocytes and a
defined thymic stromal cell population, such as MHC class II+
thymic epithelial cells or MHC class II− mesenchymal cells,
thymic stromal cells isolated from dGuo-treated fetal thymus
lobes can be stained with anti-MHC II and anti-CD45 anti-
bodies and purified by flow cytometry or magnetic cell sorting.
Anti-CD45 antibody staining is used to deplete CD45+ thymo-
cytes and dendritic cells that survive even after the dGuo
treatment.
13. Thymocytes for RTOC may be CD4−CD8− double negative
(DN) thymocytes, CD4+CD8+ double positive (DP) thymo-
cytes, or even mature CD4+CD8−/CD4−CD8+ single positive
(SP) thymocytes, depending on the purpose of the experiment.
Thymocyte populations may be prepared from adult thymuses,
newborn thymuses, or fetal thymuses. Cells from different
species may also be used. Cell sorting or magnetic cell sorting
may be employed to purify thymocyte populations.
14. To form a reaggregated lobe on the filter membrane (28), it is
important to keep the surface of the filter membrane dry and
the volume of the transferred cell slurry low, usually at 2–4 μl.
15. Mix 60 μl of 2 M CaCl2, 30 μl of DNA solution (1 μg/μl), and
360 μl of distilled water in a sterile 1.5-ml microtube. Add this
solution quickly into 450 μl of 2× HBS (HEPES-buffered
saline; 140 mM NaCl, 1.5 mM Na2HPO4, 50 mM HEPES,
pH 7.05) in a 1.5-ml microtube and mix by pipetting. Gently
add this solution containing calcium phosphate-DNA co-
precipitates onto pre-cultured Plat-E cells. Thirty minutes
later, check the formation of precipitates under the microscope.

X-tremeGENE HP (Roche Applied Science), instead of cal-
cium phosphate co-precipitation, may also work for the trans-
fection of Plat-E cells.
16. The efficiency of the transfection should be monitored after the
collection of retroviruses. Transfected Plat-E cells can be
trypsinized and analyzed for GFP expression by flow cytometry.
In general, the transfection efficiency ranges from 50 to 90%.
17. Frozen retroviral suspensions should be quickly thawed in a
37°C water bath immediately before use.
18. DN thymocytes can be transferred to dGuo-treated fetal thymus
lobes by the hanging-drop method (see Subheading 3.5;
Fig. 4a). DP and SP thymocytes should be reaggregated with
dGuo-treated thymic stromal cells (see Subheading 3.8;
Fig. 4b). Retrovirus-infected cells present after FTOC can be
detected by GFP expression using flow cytometry (see Note 6).
19. After 18–24 h of culture, retroviral infection can be evaluated
by GFP expression (see Fig. 4c). It should be noted that GFP
expression is not detectable immediately after the infection,
and is generally detected after 18–24 h of culture. To maintain
the developmental potential and the survival of immature
thymocytes, IL-7 (Sigma, final concentration 1–5 ng/ml) may
be added to the culture. GFP+ cells can be purified by cell
sorting and then be transferred to FTOC.
20. Video instructions for the fetal thymus preparation, the fetal
thymus organ culture, and the reaggregation thymus organ
culture are available online (23, 29, 30).
References
1. Owen JJT, Ritter MA (1969) Tissue interac- thymus, producing phenotypically distinct
tion in the development of thymus lympho- T-cell populations. Nature 317:811–813
cytes. J Exp Med 129:431–442 7. Takahama Y (2000) Differentiation of mouse
2. Owen JJT (1974) Ontogeny of the immune thymocytes in fetal thymus organ culture.
system. Prog Immunol 2:163–173 Methods Mol Biol 134:37–46
3. Mandel T, Russel PJ (1971) Differentiation of 8. Takahama Y, Hasegawa T, Itohara S, Ball EL,
foetal mouse thymus. Ultrastructure of organ Sheard MA, Hashimoto Y (1994) Entry of
cultures and of subcapsular grafts. Immunology CD4-CD8- immature thymocytes into the
21:659–674 CD4/CD8 developmental pathway is con-
4. Mandel TE, Kennedy MM (1978) The differ- trolled by tyrosine kinase signals that can be
entiation of murine thymocytes in vivo and provided through T cell receptor components.
in vitro. Immunology 35:317–331 Int Immunol 6:1505–1514
5. Jenkinson EJ, van Ewijk W, Owen JJT (1981) 9. Theiler K (1989) The house mouse. Springer-
Major histocompatibility complex antigen Verlag, New York, NY
expression on the epithelium of the developing 10. Kaufman MH (1992) The atlas of mouse
thymus in normal and nude mice. J Exp Med development. Academic, San Diego, CA
153:280–292 11. Butler H, Juurlink BH (1987) An atlas for
6. Kingston R, Jenkinson EJ, Owen JJT (1985) staging mammalian and chick embryos. CRC,
A single stem cell can recolonize an embryonic Boca Raton, FL
102 T. Nitta et al.
12. Morita S, Kojima T, Kitamura T (2000) Plat-E: Delta-like 4 is the essential, nonredundant
an efficient and stable system for transient ligand for Notch1 during thymic T cell lineage
packaging of retroviruses. Gene Ther 7: commitment. J Exp Med 205:2515–2523
1063–1066 22. Taghon TN, David ES, Zúñiga-Pflücker JC,
13. Saitoh T, Nakano H, Yamamoto N, Yamaoka S Rothenberg EV (2005) Delayed, asynchro-
(2002) Lymphotoxin-β receptor mediates nous, and reversible T-lineage specification
NEMO-independent NF-kB activation. FEBS induced by Notch/Delta signaling. Genes Dev
Lett 532:45–51 19:965–978
14. Schmitt MT, Zuniga-Pflucker JC (2002) 23. Ramsdell F, Zúñiga-Pflücker JC, Takahama Y
Induction of T cell development from (2006) In vitro systems for the study of T cell
hematopoietic progenitor cells by Delta-like-1 development: fetal thymus organ culture and
in vitro. Immunity 17:749–756 OP9-DL1 cell coculture. Curr Protoc Immunol
15. Tsuda S, Rieke S, Hashimoto Y, Nakauchi H, Chapter 3, Unit 3.18
Takahama Y (1996) IL-7 supports D-J but not 24. de Pooter R, Zúñiga-Pflücker JC (2007) T-cell
V-DJ rearrangement of TCR-β gene in fetal potential and development in vitro: the
liver progenitor cells. J Immunol 156: OP9-DL1 approach. Curr Opin Immunol
3233–3242 19:163–168
16. Watanabe Y, Katsura Y (1993) Development 25. Takahama Y, Suzuki H, Katz KS, Grusby MJ,
of T cell receptor αβ bearing T cells in the sub- Singer A (1994) Positive selection of CD4+ T
mersion organ culture of murine fetal thymus cells by TCR ligation without aggregation
at high oxygen concentration. Eur J Immunol even in the absence of MHC. Nature 371:
23:200–205 67–70
17. Jenkinson EJ, Anderson G, Owen JJT (1992) 26. Takahama Y, Nakauchi H (1996) Phorbol ester
Studies on T cell maturation on defined thymic and calcium ionophore can replace TCR signals
stromal cell populations in vitro. J Exp Med that induce positive selection of CD4 T cells.
176:845–853 J Immunol 157:1508–1513
18. Hawley RG, Fong AZC, Burns BF, Hawley TS 27. Jenkinson EJ, Franchi LL, Kingston R, Owen
(1992) Transplantable myeloproliferative JJT (1982) Effect of deoxyguanosine on
disease induced in mice by interleukin 6 retro- lymphopoiesis in the developing thymus rudi-
virus. J Exp Med 176:1149–1163 ment in vitro: application in the production of
19. Nitta T, Nasreen M, Seike T, Goji A, Ohigashi chimeric thymus rudiments. Eur J Immunol
I, Miyazaki T, Ohta T, Kanno M, Takahama Y 12:583–587
(2006) IAN family critically regulates survival 28. Anderson G, Jenkinson EJ, Moore NC, Owen
and development of T lymphocytes. PLoS Biol JJT (1993) MHC class II-positive epithelium
4:e103 and mesenchyme cells are both required for
20. Hozumi K, Mailhos C, Negishi N, Hirano K, T-cell development in the thymus. Nature
Yahata T, Ando K, Zuklys S, Holländer GA, 362:70–73
Shima DT, Habu S (2008) Delta-like 4 is indis- 29. Jenkinson W, Jenkinson E, Anderson G (2008)
pensable in thymic environment specific for Preparation of 2-dGuo-treated thymus organ
T cell development. J Exp Med 205:2507–2513 cultures. J Vis Exp pii:906. doi:10.3791/906
21. Koch U, Fiorini E, Benedito R, Besseyrias V, 30. White A, Jenkinson E, Anderson G (2008)
Schuster-Gossler K, Pierres M, Manley NR, Reaggregate thymus cultures. J Vis Exp pii:905.
Duarte A, Macdonald HR, Radtke F (2008) doi:10.3791/905
Chapter 7
Generation, Isolation, and Engraftment of In Vitro-Derived

Human T Cell Progenitors
Génève Awong and Juan Carlos Zúñiga-Pflücker
Abstract
T cells typically differentiate via a series of coordinated steps within the highly specialized microenvironment
of the thymus. Traditionally, human T-lymphopoiesis in vitro has been studied using the hybrid human/
mouse fetal thymic organ culture system. Pioneering work by McCune et al. devised a method to examine
human T cell development in vivo in relation to HIV-1 using the SCID/hu (thy/liv) model. This was
followed by models that better reflected the ability of human hematopoietic cells to home and differentiate
within the mouse host without human fetal tissues; however, human T cell development in these animals
was poor. In this chapter, we outline a procedure to generate human progenitor T (proT) cells in vitro
from umbilical cord blood-derived hematopoietic stem cells using the OP9-DL1 cell system; in addition,
we describe the method used to examine the engraftment of in vitro-derived proT cells into immunodeficient
mouse strains.
Key words: T cell development, T cell reconstitution, Thymus engraftment, NOD/SCID/γcnull,

RAG2−/−γc−/−, Immunodeficient, Umbilical cord blood, Notch, Delta-like 1, Hematopoietic stem
cells, IL-7
1. Introduction
T cells undergo development within the thymus from bone

marrow-derived hematopoietic progenitors, and follow a series of
differentiation stages that have been broadly characterized by the
developmentally coordinated expression of CD4 and CD8. Human
T cell development has been traditionally assayed in vitro using the
hybrid human/mouse fetal thymic organ culture (FTOC) approach
(1). Using this system, human progenitors differentiate within the
mouse thymic microenvironment and their developmental stages
are assessed at various days after introduction into the thymic lobes.
103
104 G. Awong and J.C. Zúñiga-Pflücker
Our laboratory undertook an examination of Notch receptor

ligands, which were shown to be involved in T-lymphopoiesis
(2, 3). When OP9 bone marrow stromal cells were made to ectopi-
cally express the Notch ligand Delta-like 1 (OP9-DL1) the OP9
stromal cell line ( 4) was converted from one that normally
supports B-lymphopoiesis (5, 6) into one that preferentially
supports T cell development from multiple sources of hematopoi-
etic progenitors (3, 7, 8), thus, providing a straightforward assay
to achieve mouse and human T-lymphopoiesis in vitro (9).
The study of human hematopoiesis employing mouse models
first arose in the late 1980s following the discovery of the scid
mutation in the C.B-17 mouse strain that lacked mature T and B
cells (10). Although the SCID mouse proved groundbreaking for
human cell engraftment, levels of human chimerism were low and
human thymopoiesis was lacking (11–13). Human cell engraft-
ment, and more importantly T cell development, was further
improved with the generation of the NOD/SCID mouse (11).
This led to the generation of two mouse models that exhibit a
superior capacity to accept xenogeneic grafts—the BALB/c
Rag2−/−γc−/− and NOD/SCID/γcnull mice (14), which lack the
common cytokine receptor γ chain, a critical subunit for the IL-2,
IL-4, IL-7, IL-9, IL-15, and IL-21 cytokines, rendering these
cytokines nonfunctional on their target cells. Most importantly,
NK cells do not develop in these mouse strains as IL-15 is critical
for their development. These recent mouse models were shown to
support human T-lymphopoiesis (14–17).
The capacity of the OP9-DL1 cell system to generate potential
proT cell candidates in combination with an in vivo model having
a superior capacity to accept xenogeneic grafts provides a powerful
tool to gain insight into human T-lymphopoiesis and for the study
of T cell reconstitution (18). Here we outline the procedures
for (1) the maintenance of OP9-DL1 cells, (2) isolation of cells
enriched for hematopoietic stem cells (HSCs) from umbilical cord
blood (UCB) samples, (3) generation of proT cells in OP9-DL1
coculture, (4) injection of human proT cells into immunodeficient
neonates, and (5) analysis of recipient mice.
2. Materials
It is important to note that the first step in successful cell culture

and the handling of immune-deficient mice is to practice sterile
technique. All reagents should be prepared and maintained under
sterile conditions. Extreme care should be taken with both the
cells and the reagents needed to culture them. This protocol
describes the use of T cell progenitors generated in vitro from
7 Human proT Cell Engraftment into Immunodeficient Mice 105
UCB HSCs and their engraftment into immunodeficient mice.

Both the UCB-derived HSC and mice should be handled in accor-
dance to local Institutional Ethical Review Board policies.
2.1. Maintenance 1. OP9-DL1 cells: OP9 cells (Riken Bioresource Center, http://
of Cellular Components www2.brc.riken.jp/lab/cell/detail.cgi?cell_no=RCB1124&
and Coculture type=1) retrovirally transduced to express the gene Delta-like 1
(Dll-1), as previously reported (3).
2. α-Modified Eagle’s Medium (αMEM) (Gibco 12561-056).
Store at 4°C.
3. Fetal bovine serum (FBS) (see Note 1). Heat-inactivate (hi) at
56°C for 30 min. Store at 4°C.
4. Penicillin/streptomycin: 100× or 10,000 U/mL penicillin and
10,000 U/mL streptomycin (Hyclone SV30010). Use at 1×.
Store at 4°C once opened.
5. Phosphate-buffered saline (PBS) 1× without Ca2+/Mg2+ (Gibco
14190-144).
6. Trypsin 2.5% (Gibco 15090). Dilute with PBS to 0.25%
solution. Store at 4°C.
7. OP9 medium: αMEM supplemented with 20% hiFBS and
1× penicillin/streptomycin.
8. 40 μm cell strainers (BD Falcon 352340).
9. 70 μm nylon mesh filters (N70R; Biodesign Inc.).
10. Human IL-7 (Peprotech 200-07). Reconstitute at 5 μg/mL
(1,000×) in OP9 media. Aliquot and store at −80°C.
11. Human Flt-3L (R&D 308-FK). Reconstitute at 5 μg/mL
(1,000×) in OP9 media. Aliquot and store at −80°C.
12. Human SCF (Peprotech 300-07). Reconstitute at 30 μg/mL
(1,000×) in OP9 media. Aliquot and store at −80° C.
13. Freezing media: 90% hiFBS, 10% dimethyl sulfoxide (DMSO).
Sterile filter through a 0.22 μM filter.
14. Hank’s Balanced Salt Solution (HBSS) 1× without phenol red,
Ca2+ or Mg2+ (Hyclone SH30268.01).
15. Sorting buffer: HBSS, 1% Bovine Serum Albumin (BSA)
Fraction V (OmniPur 2890).
16. Fluorescent-labeled mAbs to human CD7 (clone M-T701),
CD34 (clone 581), and CD38 (clone HIT2) (BD-Biosciences).
17. Tissue culture ware (10 cm dishes, 6-well plates, cryovials),
tissue culture treated (suggested: Sarstedt or Falcon).
2.2. Mononuclear 1. Human umbilical cord blood (UCB): Obtained in accordance

Cell Isolation with Institutional Ethical Review Board approval and upon
and Enrichment parental consent following healthy deliveries in blood collection
to Obtain HSCs bags containing anticoagulant (Baxter 4R3610nm) (see Note 2).
Alternatively, UCB CD34+ cells can be purchased from

StemCell Technologies (CB007F).
2. Ficoll-Paque PLUS (GE healthcare 17-1440-03).
3. Lysing buffer 10× (red blood cell lysis) (BD biosciences
555899).
4. Hank’s Balanced Salt Solution (HBSS) 1× without phenol red,
Ca2+ or Mg2+ (Hyclone SH30268.01).
5. Bovine Serum Albumin (BSA) Fraction V (OmniPur 2890).
6. Ethylenediaminetetraacetic acid (EDTA) (Sigma E1644).
7. DNAse I (StemCell Technologies 07900).
8. StemSep negative selection human progenitor enrichment kit
(StemCell Technologies 14056A) or StemSep CD34+ cell
selection kit (StemCell Technologies 14756A).
9. AutoMACS Running buffer: HBSS 1×, 2 mM EDTA, 0.5%
BSA. Filter sterilized through a 0.22 μM filter.
10. AutoMACS Rinsing buffer: HBSS 1×, 2 mM EDTA (keep
sterile).
11. Propidium iodide (Sigma P4170). Use at a final concentration
of 0.2 μg/mL.
2.3. Injection of Mice 1. 6–8-week-old NOD.cg-PrkdcscidIL2rgtm/Wjl/Sz (NOD/

and Postinjection SCID/γc ) mice purchased from the Jackson Laboratory
null
Analysis (JAX, Bar harbor, ME, USA).

2. BALB/c Rag2−/−γc−/− mice obtained from the Amsterdam
Medical Centre (AMC, Amsterdam, the Netherlands) (19).
3. rhIL-7 and M25 (anti-IL7 mAb) obtained from Charles Surh
(Scripps Res. Inst.).
4. BD Ultra-Fine II Insulin syringe; 30 gauge needle; 1 cc (BD
Consumer Healthcare).
5. FACS buffer: HBSS containing 1% BSA and 0.1% sodium azide.
6. 4¢-6-Diamidino-2-phenylindol (DAPI) (Invitrogen D3571).
3. Methods
3.1. OP9-DL1 Stromal All incubations are performed in a standard, humidified, cell culture
Cells incubator at 37°C in 5% CO2. In addition, cells are centrifuged
at 450 × g for 5 min at room temperature, unless otherwise
indicated.
1. Thaw a vial of frozen OP9-DL1 cells in a 37°C water bath
using a gentle swirling motion and then transfer slowly by
adding drop-wise using a 1 mL pipette into a 15 mL conical

tube containing OP9 medium.
2. Centrifuge the cells to obtain a pellet. Suspend the cells in
9–10 mL of fresh OP9 medium to seed the cells in a 10 cm dish.
3. Change the medium the following day. Always ensure to
passage the cells when the cultures are 80–90% confluent.
Appropriate confluence can be generally maintained by split-
ting 1:4 every 2 days (see Note 3).
4. To passage OP9-DL1 cells from a 10 cm plate, remove the
medium and then add 5 mL PBS to wash off the remaining
medium. Remove PBS, replace with 5 mL 0.25% trypsin and
incubate for 5 min at 37°C.
5. Following the trypsinization period, vigorously pipette the
cells to remove them from the surface of the plate and transfer
to a 15 mL conical tube containing 5 mL of OP9 media.
Rinse the plate with 5 mL of PBS and add to the contents of
the first collection. Centrifuge the cells and suspend in OP9
media. Divide the cells among 10 cm and/or 6-well plates
(see Note 4). Gently rock the plate back and forth for even cell
distribution.
3.2. Isolation 1. Dilute whole cord-blood with an equal volume of HBSS con-
of Mononuclear taining 2 mM EDTA. Carefully and gently layer 30 mL of the
Cells from Umbilical diluted blood using a 25 mL pipette into a 50 mL conical tube
Cord Blood already containing 15 mL Ficoll-paque solution (an approxi-
(NB: Subheadings 3.2 mate 2:1 ratio is used). Avoid mixing the blood-Ficoll layer.
and 3.3 Can Be 2. Centrifuge at 750 × g for 30 min at 18°C with the centrifuge
Omitted If UCB CD34+ brake “off”. After centrifugation, carefully remove the mono-
Cells Are Purchased) nuclear cell fraction present at the “cloudy” plasma/Ficoll
interface using a 10 mL pipette.
3. Transfer the mononuclear cell fraction to a 50 mL conical tube,
suspend in HBSS with four to five times the volume of inter-
face collected and centrifuge at 515–585 × g for 5 min. Carefully
remove the supernatant and discard.
4. Lyse any contaminating red blood cells by suspending the
pellet in 25 mL of 1× lysis buffer for 8–10 min at room
temperature (RT). Wash the cells by adding 25 mL of HBSS,
centrifuge at 515–585 × g for 5 min, and carefully remove the
supernatant.
5. Suspend the cells in PBS to a final volume suitable to obtain
the desired cell concentration. Set aside an aliquot for cell
count determination.
6. Centrifuge the cells once more at 515–585 × g for 5 min. At
this point that the cells can either be frozen down using freezing
media or be used for lineage depletion (Subheading 3.3).
If cells are to be frozen, suspend the mononuclear cells in

1.5 mL ice cold freezing media and aliquot into 2 mL cryovial(s)
at 50 × 106 cells/vial. Transfer the vials from ice to a −80°C
freezer for overnight storage and then to a liquid nitrogen tank
the next day for long-term storage. These vials can be thawed
at a later time to proceed with lineage depletion.
3.3. Enrichment Before isolating human hematopoietic progenitor cells, decide

of Human which subfraction of the stem cell compartment (e.g.,
Hematopoietic CD133+CD34−; CD34+CD38−) is desired as this may affect which
Progenitors by Either enrichment protocol is performed or purchased. The following
Positive or Negative steps assume the use of StemCell Technologies progenitor enrich-
Selection ment kit for either CD34+ cell enrichment or lineage depletion
using magnetic labeling and separation on the autoMACS system
(see Note 5). Instructions are provided with the manufacturers’
product inserts and should be followed according to their guide-
lines. However, it will be described in brief here.
1. Cells from Subheading 3.2, step 6 (freshly isolated or thawed—
see Note 6) should be suspended in filter sterilized running
buffer at the specified cell concentration suggested by the
manufacturer. Save a small aliquot (1–3 × 104 cells) for enrich-
ment determination (see Note 7).
2. Add the appropriate volume of selection cocktail depending on
enrichment protocol, mix well using a pipette and incubate for
the specified time at 4°C.
3. Add magnetic colloid at 60 μL/mL, mix well and incubate for
10 min at 4°C. Wash the cells by adding sterile autoMACS
running buffer, centrifuge, remove the supernatant, and
suspend the pellet in 1 mL running buffer.
4. Perform positive or negative selection on the autoMACS
following manufacturer’s instructions. Save a small aliquot to
determine enrichment between pre- and post-magnetic separa-
tion (see Note 8).
3.4. Generation The autoMACS fraction enriched for human HSCs can be further
and Isolation purified and sorted by flow cytometry based on the surface expres-
of Progenitor T Cells sion of stem cell markers. Cell sorter purified Lin− CD34+CD38−/low
from Cocultures HSC/ from UCB is seeded onto OP9-DL1 cells in the presence of IL-7,
OP9-DL1 Cocultures Flt-3L, and SCF. Human HSC/OP9-DL1 cocultures are main-
tained in 3 mL of OP9 medium in a 6-well plate for 10–12 days.
At this time, cocultures are disaggregated and sorted for
CD34+CD7++ proT cells.
1. Sorted cells (~3–5 × 104 Lin− CD34+CD38−/low) isolated from
the AutoMacs-enriched CD34+ or Lin− fraction and suspended
in 3 mL of OP9 medium are seeded into one well of a 6-well
plate containing OP9-DL1 cells at 80% confluency. The human
cytokines Flt3-L, IL-7, and SCF are added to the wells from a
1,000× stock solution (1× final concentration).
2. After 5 days, remove the medium, which will contain cells, and
pass the cells through a 70 μm sterile nylon mesh or a 40 μm
cell strainer into a 50 mL conical tube. Add 5 mL of PBS to
disaggregate the coculture by vigorous pipetting (5 mL pipette)
and pass through the same cell strainer. Add 5 mL of PBS to
obtain the remaining cells from the 6-well plate, pass through
the same cell strainer.
3. Centrifuge the cells at 515–585 × g for 5 min, remove the
supernatant and suspend the cells in 1 mL of OP9 medium.
Transfer the cells into a new 6-well plate with OP9-DL1 cells
already containing 2 mL of OP9 medum, and continue the
coculture with the addition of cytokines.
4. At day 10, repeat step 2. Centrifuge the cells at 515–585 × g for
5 min and suspend in 1 mL of PBS. Remove 10 μL for cell
count determination and viability assessment using trypan blue.
5. Centrifuge the cells at 515–585 × g for 5 min and suspend in an
appropriate volume of sorting buffer for staining and cell sort-
ing (do not exceed 50 × 106 cells/mL). Incubate the cells with
appropriately titered anti-CD34 and anti-CD7 antibodies for
30 min on ice. Keep dark.
6. Add sorting buffer (~3–4× the staining volume), centrifuge
the cells at 515–585 × g for 5 min, and suspend the cells in
sorting buffer containing a viability dye, such as propidium
iodide at a final concentration of 0.2 μg/mL or 4¢-6-Diamidino-
2-phenylindole (DAPI) at a final concentration of 0.1 μg/mL.
The volume needed will vary depending on the number of cells
to be sorted but cell density should not exceed 30 × 106 cells/
mL for sorting on the FACS-ARIA or 20 × 106 cells/mL for
the FACS–DiVa. Sort for the CD34+ CD7++ cell population
(see Fig. 1 and Note 9).
3.5. Intrahepatic In vitro-generated, sorted proT cells can be injected into 2–5 day
Injection of Progenitor old neonatal mice at 2 × 105 cells/mouse in a 30 μL volume con-
T Cells into NOD/ taining pre-complexed rhIL-7 (0.5 μg/mouse) and M25 (2.5 μg/
SCID/gcnull or BALB/c mouse). As intrahepatic injection can only be performed within the
Rag2−/−gc−/− Recipients first few days of birth, it is critical that HSC/OP9-DL1 cocultures
are initiated at least 1 week before the litter is due. In order to best
ascertain the due date of the litter, male and female mice should be
set up for timed-matings and checked for plugs. Alternatively, if
females were not checked for plugs, pregnancy may be visually
apparent as bulges at the sides of the abdomen beyond gestational
days 12–13.
1. Prepare a solution containing rhIL-7 (0.5 μg) and anti-IL7
mAb, M25 (2.5 μg), in 30 μL of PBS (see Note 10).
Day 0 Day 10
50.8
rhIL-7
rhFlt3-L
rhSCF
CD38
CD7
CD34 OP9-DL1 CD34
Fig. 1. Overview of the protocol for generating human progenitor T cells in vitro. AutoMACS-enriched UCB cells are sorted
as CD34+CD38−/low (indicated by inner box) and cocultured with OP9-DL1 cells in the presence of rhSCF (30 ng/mL), rhIL-7
(5 ng/mL), and rhFlt-3L (5 ng/mL). After 10 days, CD34+CD7++ cells (indicated by inner box) are sorted and injected into
NOD/SCID/γcnull or BALB/c Rag2−/−γc−/− recipient mice.
2. Centrifuge sorted cells and suspend the cells in 30 μL of the

rhIL-7/M25 cocktail.
3. Draw cells into a syringe (30 gauge needle), making sure to
remove air bubbles.
4. Carefully pick up the pup (2–5 day old) to be injected by lightly
scruffing the skin on the back and neck and turn pup to visu-
alize the abdomen. Locate the liver, which can be found to
the left and above the milk spot as a large reddish area (see
Note 11).
5. Inject 30 μL containing the cells and IL7/M25 cocktail into
the liver (see Note 12).
6. Inject mice with 30 μL of the IL7/M25 cocktail into the
peritoneal cavity every 4–5 days until analysis (see Note 13).
The thymuses of mice analyzed at approximately 4 weeks will
contain mostly CD4+ CD8+ double-positive T cells. CD4+
CD8− single-positive or CD4− CD8+ single-positive T cells can
be found within the thymus of reconstituted mice between
week 6 and week 9 after injection of proT cells.
3.6. Isolation and Flow 1. Isolate the thymus and place it in a 6-well plate containing
Cytometric Analysis 3 mL of OP9 medium or PBS.
of Recipient Thymus 2. Place the thymus on a 40 μM cell strainer, sitting on top of a
50 mL Falcon tube, and harvest the cells by using the plunger
of a 1-cm3 syringe as a pestle to crush the thymus. Add 20 mL
PBS to flush and wash any remaining cells through the cell
strainer.
3. Centrifuge the cells and suspend in 500 μL FACS buffer.
a b
CD4
CD3
SSC
CD45 CD8 CD45
Fig. 2. Flow cytometric analysis of thymocytes obtained from a recipient mouse injected with in vitro-derived human
progenitor T cells. CD34+CD7++ cells (as in Fig. 1) were sorted from a day 10 HSC/OP9-DL1 coculture and injected at
2 × 105 cells/mouse in a 30 μL volume containing rhIL-7 (0.5 μg/mouse) and anti-IL7 mAb, M25 (2.5 μg/mouse).
The thymus was harvested at 3 weeks after injection, (a) the expression of human CD45 was examined; and (b) the cell
surface markers CD4, CD8, and CD3 were examined on human CD45+-gated cells.
4. Remove 10 μL for cell count and viability determinations in

trypan blue.
5. Use the remaining cells for flow cytometry analysis (see Fig. 2
and Note 14).
4. Notes
1. New lots of FBS serum should be batched tested against a stan-

dard lot of FBS known to support human T-lymphopoiesis.
2. Umbilical cord blood can also be collected in heparinized
vacutainers. However, blood collection bags aid in the ease of
both collection and processing of the blood. In addition, larger
volumes can be obtained when collected in bags.
3. To preserve early passage stocks of OP9 stromal cells: Allow
OP9 cells to grow to 80–90% confluence in a 10 cm dish. Split
the 10 cm dish into four more dishes and continue the subcul-
turing procedure until at least 16 plates are 80% confluent.
Freeze one confluent plate per cryovial in freezing media.
4. As mentioned earlier, one 10 cm dish can be split to obtain
four 10 cm dishes that will be confluent in 2 days. In addition,
one 10 cm dish is equivalent to one 6-well plate so that four
6-well plates can be made from one 10 cm dish.
5. If an autoMACS is not available, a benchtop magnet and
columns can be used for enrichment.
6. Previously frozen mononuclear cells should be incubated with

DNAse I (1 mg/mL) following the manufacturer’s instruc-
tions, to prevent clumping of cryopreserved cells and ultimately
cell loss.
7. Enrichment of hematopoietic progenitors can be assessed by
saving small aliquots of cells pre- and post-selection on the
autoMACS. This can be done with antibody staining for CD34
and CD38 and analysis with a flow cytometer.
8. Following autoMACS enrichment, progenitor cells can be
frozen at this step (if not previously frozen at step 5,
Subheading 3.2). Cells should only be frozen once.
9. The expected yield of proT cells following sorting will vary
based on the number of input cells. However the approximate
yield of progenitors for 1 × 105 Lin− CD34+CD38−/low UCB
cells is 1 × 106 cells.
10. The IL-7:anti-IL7 mAb complex solution should be prepared
ahead of time as a master mix when more than one mouse is to
be injected.
11. The milk spot will appear by 24 h after birth if pups are being
fed and will disappear by postnatal day 6. Mice without a clear
milk spot should not be used.
12. If more than 30 μL is injected, the excess volume will spill out
of the liver at the site of injection. If only a portion of the litter
has been injected, mark the back of the pup using an ethanol
resistant permanent felt tip marker to identify which mouse has
received cells. Ears will be more developed after postnatal day
4 and snips to the ear using small scissors can be made at this
time to indicate treatment. Ethanol resistant markers are less
resistant to fading from the female grooming/licking her pups
and will last up to 48 h before remarking needs to be done.
13. The addition of IL7/M25 cocktail increases the cellularity
found within the thymus and the frequency of mice that engraft
with proT cells.
14. Engraftment of in vitro-derived proT cells into immunodeficient
mice can be assessed by phenotypic characterization of cells
within the thymus of the recipient mouse. If more than one
stain is required split the cell suspension equally into separate
tubes before adding antibodies. Useful markers for analysis
include CD45, CD3/TCR, CD8, CD4, CD5, CD7, and
CD1a. Flow cytometry analysis, including viability discrimina-
tion (PI or DAPI), using an LSRII for multicolor flow cytom-
etry, is recommended. When in vitro-derived proT cells are
injected into these mice ~100% of the cells within the thymus
correspond to T-lineage cells. However, one can assess for the
absence/presence of lineages other than T cells by adding the
appropriate lineage-specific antibodies.
References
1. Fisher AG, Larsson L, Goff LK, Restall DE, 11. Bhatia M, Bonnet D, Murdoch B, Gan OI,
Happerfield L, Merkenschlager M (1990) Dick JE (1998) A newly discovered class of
Human thymocyte development in mouse human hematopoietic cells with SCID-
organ cultures. Int Immunol 2:571–578 repopulating activity. Nat Med 4:1038–1045
2. Jaleco AC, Neves H, Hooijberg E, Gameiro P, 12. Lapidot T, Pflumio F, Doedens M, Murdoch B,
Clode N, Haury M, Henrique D, Parreira L Williams DE, Dick JE (1992) Cytokine stimu-
(2001) Differential effects of Notch ligands lation of multilineage hematopoiesis from
Delta-1 and Jagged-1 in human lymphoid immature human cells engrafted in SCID mice.
differentiation. J Exp Med 194:991–1002 Science 255:1137–1141
3. Schmitt TM, Zuniga-Pflucker JC (2002) 13. McCune JM, Namikawa R, Kaneshima H,
Induction of T cell development from Shultz LD, Lieberman M, Weissman IL (1988)
hematopoietic progenitor cells by delta-like-1 The SCID-hu mouse: murine model for the
in vitro. Immunity 17:749–756 analysis of human hematolymphoid differentia-
4. Kodama H, Nose M, Niida S, Nishikawa S tion and function. Science 241:1632–1639
(1994) Involvement of the c-kit receptor in the 14. Legrand N, Weijer K, Spits H (2006)
adhesion of hematopoietic stem cells to stromal Experimental models to study development
cells. Exp Hematol 22:979–984 and function of the human immune system
5. Cho SK, Webber TD, Carlyle JR, Nakano T, in vivo. J Immunol 176:2053–2058
Lewis SM, Zuniga-Pflucker JC (1999) 15. Shultz LD, Ishikawa F, Greiner DL (2007)
Functional characterization of B lympho- Humanized mice in translational biomedical
cytes generated in vitro from embryonic research. Nat Rev Immunol 7:118–130
stem cells. Proc Natl Acad Sci USA 96: 16. Traggiai E, Chicha L, Mazzucchelli L, Bronz
9797–9802 L, Piffaretti JC, Lanzavecchia A, Manz MG
6. Nakano T, Kodama H, Honjo T (1994) (2004) Development of a human adaptive
Generation of lymphohematopoietic cells from immune system in cord blood cell-transplanted
embryonic stem cells in culture. Science mice. Science 304:104–107
265:1098–1101 17. Goldman JP, Blundell MP, Lopes L, Kinnon C,
7. Schmitt TM, de Pooter RF, Gronski MA, Cho Di Santo JP, Thrasher AJ (1998) Enhanced
SK, Ohashi PS, Zuniga-Pflucker JC (2004) human cell engraftment in mice deficient in
Induction of T cell development and establish- RAG2 and the common cytokine receptor
ment of T cell competence from embryonic gamma chain. Br J Haematol 103:335–342
stem cells differentiated in vitro. Nat Immunol 18. Awong G, Herer E, Surh CD, Dick JE, La
5:410–417 Motte-Mohs RN, Zuniga-Pflucker JC (2009)
8. La Motte-Mohs RN, Herer E, Zuniga-Pflucker Characterization in vitro and engraftment
JC (2005) Induction of T-cell development potential in vivo of human progenitor T cells
from human cord blood hematopoietic stem generated from hematopoietic stem cells.
cells by Delta-like 1 in vitro. Blood 105: Blood 114:972–982
1431–1439 19. Legrand N, Cupedo T, van Lent AU, Ebeli MJ,
9. Zuniga-Pflucker JC (2004) T-cell development Weijer K, Hanke T, Spits H (2006) Transient
made simple. Nat Rev Immunol 4:67–72 accumulation of human mature thymocytes
10. Bosma GC, Custer RP, Bosma MJ (1983) A and regulatory T cells with CD28 superagonist
severe combined immunodeficiency mutation in “human immune system” Rag2(−/−)gam-
in the mouse. Nature 301:527–530 mac(−/−) mice. Blood 108:238–245
Chapter 8
In Vitro Generation of Human T Regulatory Cells: Generation,

Culture, and Analysis of FOXP3-Transduced T Cells
Alicia N. McMurchy and Megan K. Levings
Abstract
T regulatory cells (Tregs) suppress immune responses and therefore have potential to be used in the clinic
as a cellular therapy for autoimmune disease and to prevent rejection of transplanted organs. Obtaining
sufficient numbers of these cells for therapeutic use is a challenge, however, since there are currently no
Treg cell-specific markers, and they have a poor in vitro expansion potential. Tregs express high levels of
FOXP3, a transcription factor that is critical for their function. We have shown that lentivirus-based
overexpression of FOXP3 can reprogram naïve or memory human CD4+ T cells into cells which possess a
phenotype and function similar to ex vivo Tregs. Here we will review the methodology involved in
generating, expanding, and testing FOXP3-transduced cells and their ex vivo Treg counterparts.
Key words: T regulatory cells, Lentivirus, FOXP3, Cellular therapy, Tolerance
1. Introduction
T regulatory cells are immunosuppressive cells which regulate

immune responses by inhibiting various cell types including antigen-
presenting cells, B cells and T cells (1–6). Although there are many
different types of Tregs, the best characterized are those which are
CD4+ and constitutively express the IL-2 receptor alpha chain
(CD25) (7) and the transcription factor FOXP3 (8–10), hereafter
called “Tregs.” Experiments in mouse models have shown that
adoptive transfer of Tregs, as an innovative cellular therapy, can
suppress autoimmune disease (8, 11, 12), graft rejection (13, 14),
and graft-versus-host disease (15, 16) and establish long-term and
antigen-specific tolerance. Thus, there is much excitement about
the potential use of Tregs as a cellular therapy in humans.
115
116 A.N. McMurchy and M.K. Levings
The major barrier to translating the use of Tregs in mice to

humans is the lack of suitable methods to generate a large and
homogenous population of cells in vitro since Tregs represent only
~3% of circulating CD4+ T cells and typically expand poorly in vitro.
Moreover, there are no methods currently available to isolate pure
populations of cells as no Treg-specific cell surface molecules have
been identified. Consequently, during in vitro expansion, small
numbers of contaminating conventional T cells rapidly outgrow
the Tregs, decreasing the suppressive capacity and hence the thera-
peutic potential of the population (17, 18). Furthermore, increasing
evidence indicates that human FOXP3+ cells are a highly hetero-
geneous population, with a significant fraction containing pro-
inflammatory IL-17-producing cells (19–21). Therefore, better
methods to generate and expand Tregs in vitro are required for the
clinical translation of Treg cellular therapy.
We showed that when human CD4+ T cells are stably trans-
duced with the transcription factor FOXP3, they acquire the
phenotype and in vitro function of Tregs (22). Thus, a large popu-
lation of stably suppressive cells can be generated by transducing
readily abundant naïve T cells with a lentivirus encoding FOXP3.
A key feature of the lentivirus vector is the bidirectional promoter
which allows for coordinate and constitutive expression of FOXP3
and a truncated form of the low affinity nerve growth factor
receptor (ΔLNGFR) as a cell surface marker of transduced cells.
Transduced cells are purified based on ΔLNGFR expression and
can be expanded and tested as necessary. A series of in vitro assays
must be performed to confirm the Treg-like phenotype of FOXP3-
transduced cells and should include testing of ex vivo expanded
Tregs in parallel as a positive control. Key assays to confirm the
expected phenotype of FOXP3-transduced T cells include analysis
of expression of Treg-associated cell surface molecules, capacity to
produce cytokines, and the ability to suppress the proliferation of
CD4+ and/or CD8+ T cells.
2. Materials
Clones of antibodies used are listed in Table 1.
2.1. Isolation 1. Phosphate-buffered saline (PBS).

of Human Peripheral 2. Ficoll-Paque. Store at room temperature or 4°C.
Blood Mononuclear
3. Ammonium chloride solution (0.8% ammonium chloride,
Cells
0.1 mM EDTA). Store aliquots at −20°C. Store thawed aliquots
at 4°C for up to 2 weeks.
8 In Vitro Generation of Human T Regulatory Cells 117
Table 1
Antibodies and clones
Antibody Clone
CD25 for purity check 4E3

CD45RO for purity check UCHL1
CD4 for purity check RPA-T4
CD25 M-A251
CD4 RPA-T4
LNGFR (CD271) ME20.4-1.H4 or C40-1457
FOXP3 236A/E7
IL-2 MQ1-17H12
IFN-γ 4S.B3
CD8 HIT8a
CD3 OKT3
2.2. Isolation 1. Easy Sep Buffer: 2% fetal bovine serum in PBS.

of Human 2. Easy Sep CD4+ Negative Selection Kit (StemCell Technologies,
CD4+CD25−CD45RO− T Vancouver, Canada).
Cells and Human
3. Easy Sep Magnet (StemCell Technologies, Vancouver, Canada).
Antigen Presenting
Cells from PBMCs 4. MACS buffer: PBS plus 2 mM EDTA and 0.5% fetal bovine
serum, degassed. Store at 4°C.
2.2.1. CD4+CD25−
5. CD25 MicroBeads II (Miltenyi Biotec, Auburn, USA).
CD45RO− Cells
6. LS, MS, and LD columns (Miltenyi Biotec, Auburn, USA).
7. Midi MACS magnet and MACS stand (Miltenyi Biotec,
Auburn, USA).
8. CD45RO MicroBeads (Miltenyi Biotec, Auburn, USA).
9. Antibodies: anti-CD25, anti-CD45RO, anti-CD4. Ensure
anti-CD25 and anti-CD45RO antibodies recognize a different
epitope than the antibody conjugated to the CD25 and
CD45RO microbeads, respectively.
10. X-VIVO 15 supplemented with 5% AB human serum, 1× peni-
cillin (100 U/mL), streptomycin (100 μg/mL), and 1×
Glutamax (2 mM). Store up to 1 month at 4°C. Add recombi-
nant human rhIL-2 and rhIL-7 to the medium as required
according to Subheading 3. Do not refreeze IL-2 or IL-7.
2.2.2. Antigen Presenting 1. Easy Sep Buffer as in Subheading 2.2.1.

Cells 2. Easy Sep CD3+ Positive Selection Kit (StemCell Technologies,
Vancouver, Canada).
3. Easy Sep Magnet as above.

4. X-VIVO 15 as in Subheading 2.2.1, supplemented with con-
centrations of rhIL-2 and rhIL-7 indicated in Subheading 3.
2.3. Transduction 1. Anti-CD3 monoclonal antibody (OKT3).

of Human 2. Concentrated and titred lentivirus. Refer to ref. 22 and http://
CD4+CD25−CD45RO− tronolab.epfl.ch for information on lentivirus production.
Cells with pCCL or Always transduce cells with the pCCL empty vector control
pCCL.FP3 Lentivirus and pCCL.FOXP3 virus in parallel. Store virus aliquots at
−80°C. Avoid freeze-thawing, but if necessary, refreeze virus as
soon as possible after use to prevent degradation.
3. X-VIVO 15 as in Subheading 2.2.1, supplemented with con-
centrations of rhIL-2 and rhIL-7 indicated in Subheading 3.
4. Anti-LNGFR antibody.
5. 4% Formaldehyde in PBS.
2.4. Purification, 1. PBE buffer: PBS + 5 mM EDTA.

Culture, and 2. MACSelect LNGFR MicroBeads (Miltenyi Biotec, Auburn,
Restimulation of USA).
Transduced Cell Lines
3. LS columns (Miltenyi Biotec, Auburn, USA).
2.4.1. Purification 4. Midi MACS magnet and MACS stand. (Miltenyi Biotec,
of Transduced Cell Lines Auburn, USA).
5. X-VIVO 15 as in Subheading 2.2.1, supplemented with rhIL-2
as indicated in Subheading 3.
2.4.2. Culture 1. Human peripheral blood mononuclear cells (PBMCs) isolated

and Restimulation as described in Subheading 3.1.
of Transduced Cell Lines 2. Epstein-Barr virus transformed JY cells, or equivalent lympho-
blastoid cell line. Keep aliquots of JY cells frozen in liquid
nitrogen. Keep cells in culture for no longer than 2 months
before thawing a new vial. With each thaw, expand, and freeze
down more cells for future use.
3. Phytohemagluttinin (PHA). Store stock dissolved in sterile
distilled water at 1 mg/mL at −20°C.
4. Anti-LNGFR antibody.
5. X-VIVO 15 as in Subheading 2.2.1.
6. FBS and 10% DMSO for freezing cells.
7. Cryovials for freezing cells.
8. Freezing container (optional).
2.5. Phenotypic 1. PBS.

and Functional Assays 2. FACS buffer. PBS plus 1% FCS. Optional: Add sodium azide to
of Transduced Cells a final concentration of 0.02% as a preservative. Store at 4°C.
2.5.1. CD25 Expression 3. Antibodies including: anti-CD25, anti-CD4, and anti-LNGFR.
2.5.2. FOXP3 Expression 1. PBS and FACS buffer as in Subheading 2.5.1.

2. Antibodies including anti-CD4, anti-LNGFR, and anti-FOXP3
(236A/E7).
3. FOXP3 Fixation/Permeabilization Diluent and Concentrate
(eBioscience, San Diego, USA). Store at 4°C. Prepare fresh for
each use by combining 1 part concentrate and 3 parts diluent.
4. 10× Permeabilization buffer (eBioscience, San Diego, USA).
Store at 4°C. Prepare fresh for each use by diluting to 1× in
distilled water.
2.5.3. Intracellular 1. Phorbol 12-myristate 13-acetate (PMA) dissolved in DMSO to

Cytokine Staining 1 mg/mL. Store in aliquots at −80°C. Do not freeze–thaw.
2. Ionomycin dissolved in DMSO to 5 mg/mL and stored in
aliquots at −80°C. Do not freeze/thaw.
3. Brefeldin A. Store stock at 10 mg/mL in DMSO at −20°C.
Dilute 1 in 10 to 1 mg/mL before use.
4. FACS buffer as in Subheading 2.5.1.
5. FOXP3 Fixation/Permeabilization Diluent and Concentrate
as in Subheading 2.5.2.
6. 10× Permeabilization buffer as in Subheading 2.5.2.
7. Antibodies including anti-CD4, anti-LNGFR, anti-IL-2, anti-
IFN-γ, and anti-FOXP3 (236A/E7).
2.5.4. In Vitro Suppression 1. PBMCs isolated as described in Subheading 3.1.

Assay 2. PBS + 5% fetal bovine serum.
3. Carboxyfluorescein diacetate succinimidyl ester (CFDA-SE).
Store dissolved in DMSO at a concentration of 5 mM in
aliquots at −80°C. Do not freeze–thaw. CFDA-SE is converted
to carboxyfluorescein succinimidyl ester (CFSE) in the cyto-
plasm of the cell.
4. Anti-CD3 monoclonal antibody (OKT3) or anti-CD3/anti-
CD28-coated beads.
5. Anti-CD8 (conjugated to a flurochrome other than FITC or
Alexa 488).
3. Methods
3.1. Isolation 1. To process a full donation of blood (450 mL), centrifuge whole
of Human PBMCs blood at 600 × g for 25 min without brake (see Note 1). PBMCs
from Whole Blood are located at the interface (Buffy coat).
2. Remove the Buffy coat by pipetting carefully at the interface
and transfer to a new tube (see Note 2). Dilute the Buffy coat
1:1 with PBS. Alternatively, for smaller amounts of whole

blood, dilute 1:1 in PBS directly without pre-enrichment for
white blood cells.
3. Aliquot 15 mL of room temperature Ficoll-Paque into 50 mL
Falcon tubes. Tilt the 50 mL Falcon 45° and slowly and care-
fully layer 30 mL of the diluted buffy coat onto the Ficoll (see
Note 3). Try to minimize mixing of the Ficoll and Buffy coat.
Centrifuge tubes at 600 × g for 25 min without brake.
4. After the spin, the PBMCs will be at the Ficoll-plasma inter-
face, with the red blood cells and granulocytes at the bottom
of the tube. Carefully pipette at the interface, removing all
PBMCs (see Note 4).
5. Dilute collected PBMCs 1:1 in 50 mL Falcon tubes with PBS,
and centrifuge at 450 × g for 10 min.
6. Decant the supernatant (see Note 5) and collect the cells into
one 50 mL Falcon tube. Top up the tube to 50 mL with PBS,
and centrifuge at 450 × g for 5 min to wash out any remaining
Ficoll.
7. Decant the supernatant. If there is red blood cell contamina-
tion in the pellet, suspend the pellet in 5 mL room tempera-
ture ammonium chloride solution and incubate at room
temperature for 5 min (see Note 6) to lyse the red blood cells.
Dilute the ammonium chloride with PBS to 50 mL after the
5 min incubation and centrifuge at 450 × g for 5 min.
8. Decant the supernatant and suspend the pellet again in 50 mL
PBS. Centrifuge at 130 × g for 10 min to remove platelets.
9. Decant the supernatant and resuspend in PBS to count. Keep
some PBMCs for isolation of antigen presenting cells (APCs),
and use the rest to isolate CD4+CD25−CD45RO− cells (see
Note 7).
3.2. Isolation 1. Suspend PBMCs in Easy Sep Buffer at 5 × 107 cells/mL.

of Human Transfer cells to a 5 mL polystyrene tube if there are less than
CD4+CD25−CD45RO− 1 × 108 cells or a 14 mL polystyrene tube for up to 4.25 × 108
Cells (see Note 8) cells.
and Human APCs 2. Add enrichment cocktail from StemCell Easy Sep CD4+
from PBMCs Negative Selection Kit at 50 μL/mL cells. Mix with a pipette
and incubate at room temperature for 10 min.
3.2.1. CD4+CD25−
CD45RO− Cells 3. Vortex the magnetic particles from the StemCell kit and add at
100 μL/mL cells. Mix with a pipette and incubate at room
temperature for 5 min.
4. Top up to 2.5 mL for less than 1 × 108 cells or 10 mL for
1–4.25 × 108 cells with Easy Sep Buffer. Mix gently with a
pipette and place the tube without the cap into a small or large
Easy Sep magnet, respectively. Incubate at room temperature
for 5 min.
5. Pour off the supernatant which contains CD4+ cells. Do not

shake or blot off drops remaining at the lip of the tube.
6. Suspend CD4+ T cells in 90 μL cold MACS buffer per 1 × 107
cells in a 15 mL Falcon tube.
7. Add 10 μL of CD25 MicroBeads II per 1 × 107 cells. Incubate
for 15 min at 4°C.
8. Top up to 10 mL with cold MACS buffer and centrifuge at
450 × g for 5 min. While the cells are spinning, prepare an LS
column by placing it in a Midi MACS magnet on an MACS
stand and washing with 3 mL of cold MACS buffer. Discard
the flow-through.
9. Suspend cells well in 3 mL cold MACS buffer. Pass over the
washed LS column. Once the 3 mL have passed through the
column, wash three times with 3 mL cold MACS buffer. Keep
the flow-through as CD25− cells.
10. If a CD25+ Treg line is desired in parallel (see Note 9), elute
the CD25+ cells by removing the column from the magnet and
adding 5 mL of MACS buffer to the column. Immediately
plunge the 5 mL through the column into a clean 15 mL tube
using the plunger provided with the column. Remove the
plunger from the column, add another 5 mL of MACS buffer
to the column, and plunge again for a final volume of 10 mL.
Centrifuge eluted cells at 450 × g for 5 min. While cells are
spinning, place an MS column in the magnet on the MACS
stand and wash with 1 mL MACS buffer. Suspend cells in 1 mL
cold MACS buffer, and add to the prewashed MS column.
After addition of cells, rinse the column three times with 1 mL
cold MACS buffer. Elute the cells by removing the column
from the magnet and placing over a clean 15 mL tube. Add
1 mL of media to the column and plunge through with the
plunger provided with the column. Remove the plunger and
repeat the elution with another 1 mL of media. Keep the
eluent as CD4+ CD25+ Tregs.
11. Count the CD25− cells and suspend in 80 μL cold MACS
buffer per 1 × 107 cells.
12. Add 20 μL CD45RO MicroBeads per 1 × 107 cells. Incubate
for 15 min at 4°C.
13. Top up to 10 mL with cold MACS buffer and centrifuge at
450 × g for 5 min. While the cells are spinning, place an LD
column in the magnet on the MACS stand and prewash with
2 mL MACS buffer. Discard the flow-through.
14. Suspend the cells in 2 mL cold MACS buffer and add the cells
to the column. Wash the column two times with 1 mL and col-
lect the flow-through as CD4+CD25−CD45RO− cells.
15. Check the purity of the isolated cells by flow cytometry with
anti-CD4, anti-CD25, and anti-CD45RO antibodies.
16. Suspend the cells in X-VIVO 15 supplemented with 5% human

serum, penicillin (100 U/mL), streptomycin (100 μg/mL),
and GlutaMAX (2 mM), as well as 100 U/mL rhIL-2 and
10 ng/mL rhIL-7 (see Note 10).
3.2.2. Antigen Presenting 1. Suspend PBMCs at 1 × 108/mL in Easy Sep Buffer in a 5 mL

Cells polystyrene tube (for up to 2 × 108 cells).
2. From a StemCell EasySep CD3+ Positive Selection Kit, add
100 μL of positive selection cocktail per 1 mL of cells. Incubate
at room temperature for 15 min.
3. Add 50 μL of nanoparticles per 1 mL of cells and incubate at
room temperature for 10 min.
4. Add Easy Sep Buffer to a final volume of 2.5 mL. Mix and add
to a small Easy Sep magnet for 10 min at room temperature.
5. Pour off the supernatant and keep this as CD3− cells (APCs).
6. Suspend in the same medium plus cytokines as above in
Subheading 3.2.1 for CD4+CD25−CD45RO− cells.
3.3. Transduction Refer to Fig. 1 for a time line of the procedure from the activation
of Human and transduction of CD4+ cells to the assays for biological activity.
CD4+CD25−CD45RO−
1. Plate 2–3 × 105 CD4+CD25−CD45RO− cells per well in a
Cells with pCCL
24-well plate (or 105 in a 48-well plate). Add APCs at a 5:1
or pCCL.FP3 Lentivirus ratio of APCs:T cells.
2. Add anti-CD3 (OKT3) to a final concentration of 1 μg/mL,
with a total of 1 mL final volume for a 24-well plate or 0.5 mL
final volume for a 48-well plate.
3. Incubate overnight (16–18 h) at 37°C, 5% CO2.
4. Remove half the volume from each well and transfer to an
Eppendorf tube. Add pCCL.FP3 or pCCL control virus (see
Note 11) to the Eppendorf at a multiplicity of infection of 10
(don’t count APCs in the calculation, only T cells). Mix gently
and add media plus virus slowly and carefully back on top of
the cells, placing the tip of the pipette at the edge of the well.
Do not mix with the pipette, but swirl gently.
Fig. 1. Key time points in the generation, culture, and analysis of FOXP3-transduced T cells. The first 20–26 days are
outlined, which includes transduction of naïve T cells, purification of ΔLNGFR+ cells, and restimulation and analysis of
the cell lines.
5. Keep some cells untransduced and transduce some cells with

pCCL control lentivirus as experimental controls.
6. After another 24 h, remove half the medium and replenish to
dilute out the virus. Replace with fresh medium plus cytokines.
7. Monitor cell growth and split as required, keeping cells at
approximately 1 × 106/mL. Provide cells with fresh rhIL-2 and
rhIL-7 every 2–3 days, as part of the splitting procedure or as
a medium change. For a medium change, remove half the
medium and add fresh medium containing 200 U/mL rhIL-2
for a final concentration of 100 U/mL rhIL-2 and 20 ng/mL
rh-IL-7 for a final concentration of 10 ng/mL (assume cytokine
consumption). Cease addition of rhIL-7 after ΔLNGFR
purification (see below).
8. Six days after activation of the cells (5 days after transduction)
the transduction efficiency can be checked by staining with an
anti-LNGFR antibody (see Note 12). Fix the cells with 4%
formaldehyde to a final concentration of 2% formaldehyde after
staining and before reading by flow cytometry as a safety
precaution against the possibility of live virus. An example of
the transduction efficiency analysis is shown in Fig. 2.
Fig. 2. Transduction efficiency and ΔLNGFR expression on purified cell lines. The left
column shows an example of an average transduction efficiency for pCCL control (top)
and pCCL.FP3 (bottom) transduced cells. The right column shows ΔLNGFR expression
after purification.
3.4. Purification, 1. Between days 8 and 10 post-activation, purify the cells based
Culture, and on ΔLNGFR expression with MACSelect LNGFR MicroBeads.
Restimulation of Wash the cells in PBE buffer and suspend in 160 μL cold PBE
Transduced Cell Lines plus 40 μL MACSelect LNGFR MicroBeads in a 15 mL
conical tube for up to 4 × 107 cells.
3.4.1. Purification
of Transduced Cell Lines 2. Incubate on ice for 15 min.
3. Top up to 10 mL with cold PBE and centrifuge at 450 × g for
5 min. While cells are spinning, place an LS column in a Midi
MACS magnet on an MACS stand and prewash with 3 mL
cold PBE buffer. Discard the flow-through.
4. Suspend the cells in 3 mL cold PBE and put over the pre-
washed LS column.
5. Wash the column three times with 3 mL cold PBE buffer
6. Elute LNGFR+ cells in X-VIVO 15 medium containing 100 U/
mL IL-2 by removing the column from the magnet, adding
3 mL of medium, and plunging the medium through the col-
umn using the plunger provided with the column into a clean
15 mL tube. Repeat the elution step by removing the plunger
from the column, adding another 3 mL of medium, and plung-
ing again for a final volume in the tube of 6 mL. Note the
discontinuation of rhIL-7 in the medium since Tregs do not
express the IL-7Rα chain (CD127) and addition of IL-7 favors
outgrowth of contaminating cells (22). Culture as usual at
approximately 1 × 106 cells/mL. An example of ΔLNGFR
expression post-purification is shown in Fig. 2.
3.4.2. Culture 1. Monitor the activation state of the cells by noting cell shape
and Restimulation and clustering. When cells enter the resting phase (become
of Transduced Cell Lines small and round, stop proliferating), restimulate with a T cell
feeder (see below). This will usually occur 10–13 days post-
activation depending on the donor. Avoid restimulating within
48 h of the ΔLNGFR purification because the purification pro-
cess can activate the cells, and restimulating them too soon
after this process can lead to activation-induced cell death.
2. Prepare a 2× T cell feeder mixture according to the following
recipe: 2 × 106/mL irradiated (5,000 rad) allogeneic human
PBMCs, prepared as described above (see Note 13), 2 × 105/
mL irradiated (7,500 rad) JY cells, 2 μg/mL PHA, and 200 U/
mL rhIL-2.
3. Plate cells in 0.5 mL in a 24-well plate with between 1 and
5 × 105 cells per well (2 × 105 per well is optimal). Add 0.5 mL
2× feeder on top.
4. Change medium after 2–3 days, adding fresh medium plus
200 U/mL rhIL-2, or split if necessary. Keep cells at approxi-
mately 1 × 106/mL as normal (see Note 14).
5. Check the phenotype and function of transduced cells

10–13 days after restimulation (see below). To keep cells in
culture, restimulate again with a feeder mixture. Cells can also
be frozen in FBS plus 10% DMSO at this point at densities
between 1 × 105 and 5 × 107 cells/mL. Freeze cells by sus-
pending them in cold FBS plus 10% DMSO and transferring to
a cryovial. Freeze cells slowly by either incubating vials on ice
for 25 min before transferring to −80°C or using a freezing
container. Transfer frozen to cells liquid nitrogen the following
day for maximal viability post-thaw.
6. Routinely check LNGFR expression of transduced cells since
contaminating untransduced cells can outgrow FOXP3-
expressing cells. If necessary (see Note 15), repurify ΔLNGFR+
cells, but always wait until days 8–10 after restimulation and at
least 2 days before the next restimulation and assays to avoid
activation-induced cell death.
3.5. Phenotypic 1. Before performing phenotypic and functional assays the cells
and Functional Assays should be rested overnight. Wash cells once with PBS and
of Transduced Cells replate in medium lacking rhIL-2 at 2–3 × 106/mL the night
(see Note 16) before the assays. In the morning, wash again with PBS and
suspend in medium lacking rhIL-2.
2. Count cells and adjust the concentration to 1 × 106 cells/mL.
Perform assays as described below.
3.5.1. CD25 Expression 1. Suspend 5 × 104–1 × 105 cells in 25–50 μL of FACS buffer and
(see Note 17) stain for CD25, CD4, and LNGFR in a V-bottom 96-well
plate. Incubate at 4°C for 20–30 min.
2. Top up to 200 μL with FACS buffer and centrifuge the plate
at 980 × g for 3 min.
3. Suspend in 200 μL FACS buffer and analyze on a flow cytom-
eter. Expected results are shown in Fig. 3.
3.5.2. FOXP3 Expression 1. Suspend 1–2 × 105 cells in 25–50 μL of FACS buffer and stain
(See Note 18) for CD4 and LNGFR in a V-bottom 96-well plate. Stain for
20–30 min at 4°C (see Notes 19 and 20).
2. Wash cells by topping up to 200 μL and centrifuging at 980 × g
for 3 min. Prepare eBioscience Fixation/Permeabilization buffer
and add 100 μL per well. Incubate at 4°C for 30–60 min. Top up
with PBS to 200 μL and centrifuge at 980 × g for 3 min.
3. At this point, cells can be suspended in FACS buffer and left
overnight at 4°C to continue the next morning, or the proce-
dure can be continued immediately.
4. Suspend cells in 200 μL eBioscience 1× Permeabilization
buffer. Centrifuge at 980 × g for 3 min and wash again with
Permeabilization buffer.
Fig. 3. CD25 is upregulated on pCCL.FP3-transduced cells compared to pCCL control-

transduced cells. CD25 expression is measured when cells are in the resting state,
10–13 days after initial activation or restimulation.
Fig. 4. FOXP3 and ΔLNGFR expression on pCCL.FP3-transduced cells and control cells. FOXP3 and ΔLNGFR expression
when cells are in the resting state 10–13 days after restimulation.
5. Suspend cells in 25 μL Permeabilization buffer and add

anti-FOXP3 antibody. Incubate 30 min at room temperature.
6. Top up to 200 μL with Permeabilization buffer and centrifuge
at 980 × g for 3 min. Wash once more with FACS buffer.
Suspend cells in 200 μL FACS buffer and read results on a flow
cytometer. Expected results are shown in Fig. 4.
3.5.3. Intracellular 1. Remove 2 × 200 μL of 1 × 106/mL cells (or 2 × 100 μL if cell

Cytokine Staining number is limited and adjust the volume to 200 μL with
(See Note 21) medium) for intracellular cytokine staining. Place each 200 μL
aliquot in one well of a 96-well round-bottom plate. One well
will be stimulated with PMA and ionomycin and one well will
be left as an unstimulated control. Also take extra cells from
control transduced cells for flow cytometry compensation
controls.
2. Prepare a mixture of PMA and ionomycin as follows: Dilute

stock PMA (1 mg/mL) 1 in 1,000 in X VIVO 15 media supple-
mented with 5% human AB serum, 100 U/mL penicillin,
100 μg/mL streptomycin, and 2 mM GlutaMAX. Using the
same medium, dilute stock ionomycin (5 mg/mL) 1 in 100. Add
100 μL of diluted PMA to 100 μL of diluted ionomycin and mix.
Add 4 μL of this mixture to one of the two wells for each sample
resulting in final concentrations of 10 ng/mL for PMA and
500 ng/mL for ionomycin. Incubate at 37°C for 2 h.
3. Add 2 μL of 1 mg/mL Brefeldin A to all wells for a final
concentration of 10 μg/mL.
4. Incubate for 4 h. Transfer the cells to a V-bottom plate. Spin
the plate at 980 × g for 3 min and shake out the supernatant.
5. Suspend the cells in 25–50 μL of FACS buffer and stain for
CD4 and LNGFR (see Notes 19 and 20). Incubate for
20–30 min at 4°C.
6. Top up each well to 200 μL and centrifuge at 980 × g for 3 min
to wash the cells. Prepare eBioscience Fixation/Permeabilization
buffer and add 100 μL per well. Incubate at 4°C for 30–60 min.
Top up with PBS to 200 μL and centrifuge at 980 × g for 3 min.
7. At this point, cells can be suspended in FACS buffer and left
overnight at 4°C to continue the next morning, or the proce-
dure can be continued immediately.
8. Suspend cells in 200 μL eBioscience 1× Permeabilization buf-
fer. Centrifuge at 980 × g for 3 min and wash again with
Permeabilization buffer.
9. Suspend cells in Permeabilization buffer and add antibodies:
anti-IL-2, anti-IFN-γ, and anti-FOXP3 (see Note 20). Incubate
30 min at room temperature.
10. Top up to 200 μL with Permeabilization buffer and centrifuge
at 980 × g for 3 min. Wash once more with FACS buffer. Suspend
cells in FACS buffer and acquire data on a flow cytometer.
Stimulated pCCL control transduced cells and untransduced
cells should produce a significant amount of IL-2 and IFN-γ,
while stimulated FOXP3-transduced cells should not produce
much of either cytokine. Expected results are shown in Fig. 5.
3.5.4. In Vitro Suppression 1. Isolate human PBMCs as described in Subheading 3.1 (see
Assay Note 22).
2. Suspend PBMCs in PBS plus 5% FBS to 1 × 106/mL.
3. Label the PBMCs with CFSE by diluting stock CFDA-SE
(5 mM in DMSO) 1 in 100 in PBS plus 5% FBS. For each
1 mL of PBMCs, add 100 μL of diluted CFDA-SE.
4. Incubate for 3.5 min at room temperature and wash with
PBS + 5% FBS.
Fig. 5. Cytokines are downregulated in pCCL.FP3-transduced cells compared to pCCL control-transduced cells. Cells
are activated with PMA and Ionomycin for 6 h, with Brefeldin A added for the last 4 h. Following activation, pCCL.FP3
and expanded ex vivo CD4+CD25+ T cells produce significantly less IL-2 and IFN-γ than pCCL control transduced cells.
Fig. 6. Ex vivo CD4+CD25+ T cells suppress the proliferation of CD8+ responder cells. 1 × 105 human PBMCs are labeled
with CFSE and cocultured with ex vivo CD4+CD25+ cells at the indicated ratios in the presence of 1 μg/mL anti-CD3. Four
days later, cells are stained with anti-CD8 antibody and analyzed by flow cytometry. Analysis is done on gated on CD8+ T
cells to ensure no CD4+ Tregs are included in the gate. The negative control contains PBMCs alone in the absence of anti-
CD3, and the positive control contains PBMCs alone in the presence of anti-CD3.
5. Suspend cells in media lacking rhIL-2 and adjust the concen-

tration to 1 × 106/mL.
6. Plate cells in a 96-well round-bottom plate as described below.
Negative control: 100 μL PBMCs + 150 μL medium
Positive control: 100 μL PBMCs + 100 μL medium
1:1—100 μL PBMCs + 100 μL test cells
1:2—100 μL PBMCs + 50 μL test cells + 50 μL medium
1:4—100 μL PBMCs + 25 μL test cells + 75 μL medium
1:8—100 μL PBMCs + 12.5 μL test cells + 87.5 μL medium
1:16—100 μL PBMCs + 6.3 μL test cells + 93.7 μL medium
7. To all wells EXCEPT the negative control, add 50 μL 5 μg/
mL anti-CD3 for a final concentration in the well of 1 μg/mL
(see Note 23).
8. Incubate at 37°C for 4 days.

9. Stain all samples with anti-CD8 and read by FACS (CFSE is in
the FITC channel.) Expected results for ex vivo CD4+CD25+
T cells are shown in Fig. 6. Using the proliferation platform of
a flow cytometry software package, calculate the average
number of divisions undergone by a cell in the starting popula-
tion (defined as division index or proliferation index, depending
on the software package).
4. Notes
1. Alternatively, whole blood can be left overnight at room

temperature to separate. Collect Buffy coat as normal the
next day.
2. In a 250 mL conical tube, mark 1 cm above and below the
interface. Remove and discard plasma to the top mark and care-
fully pipette at the interface until the bottom mark is reached,
transferring the Buffy coat to a new 250 mL conical tube.
3. Alternatively, 30 mL of diluted PBMCs can be added to the
50 mL falcon tube first, and the Ficoll can be underlayed by
slowly pipetting the Ficoll into the bottom of the tube.
4. Remove the cells from the Ficoll immediately, as Ficoll is toxic
to the cells. Pipetting Ficoll with the cells cannot be avoided,
but try to minimize Ficoll contamination by pipetting just
above the cell layer in the plasma to favor plasma collection
over Ficoll collection.
5. A high degree of Ficoll contamination may hinder pelleting of
cells. The supernatant from this spin can be kept, diluted further
with PBS, and recentrifuged in an attempt to recover more cells.
6. Do not incubate in the ammonium chloride solution for
longer than 5 min, as a greater length of time will affect the
white blood cells in addition to the red blood cells.
7. Expect the yield of APC from PBMCs to be about 15–30% and
the yield of CD4+CD25−CD45RO− T cells from PBMCs to be
about 2–10%. Keep in mind that five times as many APCs as
CD4+CD25−CD45RO− cells are required.
8. CD4+CD25−CD45RO− T cells are naïve cells which are most
readily converted into Tregs. However CD4+CD25− or total CD4+
cells may also be transduced. Depleting CD25+ cells removes
natural T regulatory cells from the transduced population.
9. Depending on their purity, a bead-sorted CD4+CD25+ T cell
line may serve as an adequate Treg cell line control, however
the most reliable source of CD4+ Tregs for expansion can be
obtained by sorting the CD25hiCD45RA+ T cells from CD25-

enriched T cells according to Hoffmann et al. (23).
10. Addition of rhIL-7 is optional, but use it to obtain maximum
transduction efficiency.
11. Consult with your local biosafety office prior to commencing
work with lentivirus to ensure adequate biosafety precautions
are taken. Wear appropriate personal protective equipment and
decontaminate all waste with 10% bleach before discarding.
12. The average transduction efficiency for pCCL control lentivi-
rus is 86 ± 12% (76–97%) and the average for pCCL.FP3 lenti-
virus is 47 ± 18% (28–71%) (22).
13. If possible, it is best to combine PBMCs from two different
donors (1 × 106/mL for each). Using two donors better ensures
a successful stimulation in case one of the donors does not
produce a strong allogeneic response. It is also preferable to
use PBMCs that have been freshly prepared; however, a stock
of PBMCs can be frozen in liquid nitrogen and thawed as
needed for restimulations.
14. Cells will need to be split more often early after the stimulation
(in the first 6–7 days) and less often later on in the cycle. In the
second week after the activation, as cells enter the resting
phase, they should be kept denser (approximately 2 × 106/mL).
If cells appear to be dying in the second week, they can be
washed to remove dead cells and replated at a higher density
with fresh rhIL-2. In an average experiment, starting with
5 × 105 naïve T cells, approximately 1.5 × 107 pCCL.FP3-
transduced T cells should be obtained after 20–26 days in
culture (after the initial ΔLNGFR+ purification and at the end
of the first restimulation) (22).
15. Cells should be repurified if they are less than 85% ΔLNGFR+
at 8–10 days after T cell receptor stimulation.
16. Lack of IL-2 and IFN-γ production and upregulation of CD25
can be observed after the first round of expansion (10–13)
days in culture, but suppressive capacity is not fully realized
until after a second round of expansion. The assays can be
conducted over 2 days, with the intracellular staining and
CD25 expression performed 1 day and the FOXP3 expression
and suppression assay performed another. It is useful to check
FOXP3 expression on the same day that the suppression assay
is set up so that the proportion of FOXP3-expressing cells at
the time of the assay is known.
17. Cell surface markers in addition to CD25 can also be examined
including CTLA4, CCR4, GITR, and CD127, but CD25 is the
most robust and reliable. This assay should be performed when
cells are in the resting state, when CD25 expression should be
high in FOXP3-transduced cells and low in pCCL control and
untransduced cells. If CD25 is high in pCCL control or untrans-

duced cells, they may still be activated, and the assay should be
repeated at a later time point. Avoid giving cells fresh rhIL-2 the
day before the assay to ensure cells are in the resting state.
18. The CD25 and FOXP3 stains can be combined; however, the
permeablization step of FOXP3 intracellular staining can some-
times interfere with CD25 and LNGFR surface staining, so it
is also preferable to do surface stains alone in parallel to surface
plus FOXP3 intracellular staining.
19. Also perform staining for flow cytometry controls including an
unstained control, single-stained controls, and fluorescence
minus one controls. Treat controls the same way as samples
throughout the procedure.
20. The FOXP3 fixation/permeablization buffer can break apart
conjugated monoclonal antibodies, so avoid using these when
staining for FOXP3 and intracellular cytokines.
21. Cytokine production can also be determined by ELISA.
22. Autologous untransduced CD4+ cells cultured in parallel may
also be used as responders instead of freshly isolated PBMCs.
Indeed, use of a cell line cultured in parallel may provide more
accurate results than freshly isolated responders because freshly
isolated T cells have different kinetics of activation compared
with T cell lines. When analyzing data with CD4+ T cells as
responders, gate out CFSElow T cells or label test cells with
another cell proliferation dye to distinguish them from the
responders.
23. Alternatively, stimulate PBMCs with anti-CD3/anti-CD28-
coated beads. Perform a titration of beads to determine the
optimum ratio of beads:cells that will result in at least 20% of
the cells dividing at least once. Usually a ratio of 1:16 or 1:32
beads:cells results in good proliferation.
Acknowledgments
We thank Sarah E. Allan, Rosa Bacchetta, Maria Grazia Roncarolo,

Mario Amendola, and Luigi Naldini for their contributions to the
development of this protocol. Supported by the Roche Organ
Transplant Research Foundation, CIHR (MOP-93793), and
Stemcell Technologies, Inc. Core support for flow cytometry and
virus production was funded by the Immunity and Infection
Research Centre Michael Smith Foundation for Health Research
(MSFHR) Unit. MKL holds a Canada Research Chair in
Transplantation. ANM holds a Canada Vanier Scholarship, a
MSFHR Junior Graduate Studentship, and a CIHR Transplantation
Training Program award.
References
1. Miyara M, Sakaguchi S (2007) Natural regula- 13. Xia G, He J, Zhang Z et al (2006) Targeting
tory T cells: mechanisms of suppression. Trends acute allograft rejection by immunotherapy
Mol Med 13:108–116 with ex vivo-expanded natural CD4+CD25+
2. van Maurik A, Herber M, Wood KJ et al (2002) regulatory T-cells. Transplantation 82:
Cutting edge: CD4+CD25+ alloantigen 1749–1755
specific immunoregulatory cells that can pre- 14. Graca L, Le Moine A, Lin CY et al (2004)
vent CD8+ T cell-mediated graft rejection: Donor-specific transplantation tolerance: the
implications for anti-CD154 immunotherapy. paradoxical behaviour of CD3+CD25+ T-cells.
J Immunol 169:5401–5404 Proc Natl Acad Sci USA 101:10122–10126
3. Lin CY, Graca L, Cobbold SP et al (2002) 15. Taylor PA, Lees CJ, Blazar BR (2002) The
Dominant transplantation tolerance impairs infusion of ex vivo activated and expanded
CD8+ T cell function but not expansion. Nat CD4(+)CD25(+) immune regulatory cells
Immunol 3:1208–1213 inhibits graft-versus-host disease lethality.
4. Lim HW, Hillsamer P, Banham AH et al (2005) Blood 99:3493–3499
Cutting edge: direct suppression of B cells by 16. Hoffmann P, Ermann J, Edinger M et al (2002)
CD4+ CD25+ regulatory T cells. J Immunol Donor-type CD4(+)CD25(+) regulatory
175:4180–4183 T-cells suppress lethal acute graft-versus-host
5. Maloy KJ, Antonelli LR, Lefevre M et al (2005) disease after allogeneic bone marrow transplan-
Cure of innate intestinal immune pathology by tation. J Exp Med 196:389–399
CD4+CD25+ regulatory T cells. Immunol 17. Riley JL, June CH, Blazar BR (2009) Human
Lett 97:189–192 T regulatory cell therapy: take a billion or so
6. Tiemessen MM, Jagger AL, Evans HG et al and call me in the morning. Immunity
(2007) CD4+CD25+Foxp3+ regulatory T 30:656–665
cells induce alternative activation of human 18. Sagoo P, Lombardi G, Lechler RI (2008)
monocytes/macrophages. Proc Natl Acad Sci Regulatory T cells as therapeutic cells. Curr
USA 104:19446–19451 Opin Organ Transplant 13:645–653
7. Sakaguchi S, Sakaguchi N, Asano M et al 19. Miyara M, Yoshioka Y, Kitoh A et al (2009)
(1995) Immunologic self-tolerance maintained Functional delineation and differentiation
by activated T cells expressing IL-2 receptor dynamics of human CD4+ T cells expressing
alpha-chains (CD25). Breakdown of a single the FoxP3 transcription factor. Immunity
mechanism of self-tolerance causes various 30:899–911
autoimmune diseases. J Immunol 155: 20. Ayyoub M, Deknuydt F, Raimbaud I et al
1151–1164 (2009) Human memory FoxP3+ Tregs secrete
8. Fontenot JD, Gavin MA, Rudensky AY (2003) IL-17 ex vivo and constitutively express the
Foxp3 programs the development and function Th17 lineage-specific transcription factor
of CD4+CD25+ regulatory T cells. Nat RORγt. Proc Natl Acad Sci USA 106:
Immunol 4:330–336 8635–8640
9. Hori S, Nomura T, Sakaguchi S (2003) Control 21. Voo KS, Wang KH, Santori FR et al (2009)
of regulatory T cell development by the Identification of IL-17-producing FOXP3+
transcription factor Foxp3. Science 299: regulatory T cells in humans. Proc Natl Acad
1057–1061 Sci USA 106:4793–4798
10. Khattri R, Cox T, Yasayko SA et al (2003) An 22. Allan SE, Alstad AN, Merindol N et al (2008)
essential role for Scurfin in CD4+CD25+ T Generation of potent and stable human CD4+
regulatory cells. Nat Immunol 4:337–342 T regulatory cells by activation-independent
11. Tang Q, Henriksen KJ, Bi M et al (2004) In expression of FOXP3. Mol Ther 16:
vitro-expanded antigen-specific regulatory 194–202
T-cells suppress autoimmune diabetes. J Exp 23. Hoffmann P, Eder R, Boeld TJ et al (2006)
Med 199:1455–1465 Only the CD45RA+ subpopulation of CD4+
12. Mottet C, Uhlig HH, Powrie F (2003) Cutting CD25high T cells gives rise to homogeneous
edge: cure of colitis by CD4+CD25+ regula- regulatory T-cell lines upon in vitro expansion.
tory T-cells. J Immunol 170:3939–3943 Blood 108:4260–4267
Chapter 9
Simultaneous Cloning and Selection of Hybridomas

and Transfected Cell Lines in Semisolid Media
Bert Wognum and Tracy Lee
Abstract
Selection and cloning are essential but often laborious and time-consuming steps during the generation
of hybridomas and genetically modified cell lines that produce monoclonal antibodies or other proteins
with desired properties. Methods for the simultaneous selection and cloning of hybridomas and trans-
fected cell lines (e.g., CHO-S cells) in semisolid methylcellulose-based media have been developed.
By using semisolid selection media, the cells that survive the selection process proliferate and form colonies
of cells that remain physically separated from other colonies. Each colony thus originates from a single
hybridoma or transfected cell and can be isolated and characterized separately. This approach avoids the
isolation of multiple identical clones and the loss of useful clones due to overgrowth by other faster-
growing, but possibly nonproducing clones, which are major problems of conventional procedures in
liquid media. In this chapter, protocols are described for the generation of mouse hybridomas by fusion of
spleen cells from immunized mice with myeloma cells and the subsequent selection and cloning of hybri-
domas in semisolid selection media. Protocol are also described for selection and cloning of transfected cell
lines using semisolid antibiotic-containing selection media, as well as strategies to optimize selection and
cloning in serum-containing, serum-free, and chemically defined selection media.
Key words: Monoclonal antibodies, Methylcellulose, Colonies, Transfection, Chinese Hamster

Ovary Cell Line
1. Introduction
Procedures for hybridoma or transfected cell line generation are

aimed at the development of stable clonal cell lines that produce
high levels of monoclonal antibodies or other proteins with desired
properties. Conventional methods for cell line development involve
the selection, screening and cloning of hundreds or even thousands
of cell cultures to generate stable clonal cell lines. Traditionally
133
134 B. Wognum and T. Lee
the selection and cloning of newly generated cell lines is done by

culturing the cells in multiple parallel cultures (e.g., in individual
wells of 96-well plates), screening each culture for the presence of
the protein of interest and then cloning the selected cell line by at
least one round of culturing under limiting dilution conditions,
and then screening each culture for the presence of cells that
produce the desired antibody or recombinant protein. These
methods require multiple culture and screening steps.
This chapter describes methods for simultaneous cloning and
selection of hybridomas and transfected cell lines using semisolid
media. By plating the cells in a semisolid selection medium after
fusion or transfection the progeny of the selected cells stay
together and form distinct colonies that can be harvested and
screened individually. As each colony is derived from a single
cell, selection of replicate clones is avoided. In addition, smaller,
slow-growing clones remain physically separated from larger
faster-growing clones and can be isolated and screened separately,
thus avoiding loss of the smaller clones due to overgrowth by
larger, faster growing clones and therefore increasing the diver-
sity of clones that can be identified and isolated. This approach
can significantly reduce the time and work needed to generate
stable clonal cell lines that express the protein of interest.
The high viscosity of semisolid media used for selection and
cloning is achieved by the addition of methylcellulose to the
culture medium. Methylcellulose is a relatively inert polymer with
good optical clarity and suitable viscosity at concentrations of
~0.9–1.5%. It is possible to make methylcellulose-based selection
and cloning media in one’s own lab from individual components,
but prescreening of methylcellulose batches, fetal bovine sera,
cytokines, and other components is required to avoid large
batch-to-batch variability in performance. Alternatively commer-
cially available media can be used that have been developed
specifically for hybridoma development or transfected cell line
generation. Suggestions for making and testing semisolid media
from liquid media and culture-tested concentrated methylcellulose
solutions are also provided.
2. Materials
2.1. Laboratory 1. Biosafety cabinet certified for level II handling of biological

Equipment materials.
and Supplies 2. Incubator with humidity and gas control to maintain 37°C,
>95% humidity, and 5% CO2.
3. Inverted microscope.
9 Cloning in Semi-solid Media 135
4. Low speed bench centrifuge.

5. Liquid nitrogen tank and freezing head; optional.
6. Freezing container (i.e., “Mr. Frosty” Nalgene Catalog #5100);
optional.
7. Culture supplies: 15 and 50 mL conical polypropylene centri-
fuge tubes, 1, 5, and 10 mL sterile pipettes, 10 cm petri dishes,
3 and 12 mL syringes; 16-gauge blunt end hypodermic
needles, 96- and 24-well tissue culture plates, tissue culture
flasks.
8. Automated cell counter or hemocytometer and routine light
microscope, 0.4% trypan blue dye solution for viable cell counts.
2.2. Generation 1. Myeloma Cell Line. The myeloma cells should be mycoplasma
of Mouse Hybridoma free, fuse well and allow the formation of stable hybridomas
that continually secrete specific monoclonal antibodies. Parental
myeloma cells that meet these criteria (such as SP2/0 and
X63Ag8.653) are widely available. Whenever possible, obtain
a parental myeloma cell that has been proven to yield stable
hybridomas.
2. Immunized mouse 1–4 days after final antigen boost.
3. Sterile sets of fine, sharp scissors and forceps for animal dissec-
tion. Sterilize by autoclaving for 40 min at 121°C.
4. ClonaCell-HY monoclonal antibody development kit
(STEMCELL Technologies, Catalog #03800). The kit contains
Medium A—ClonaCell-HY perfusion medium and hybridoma
expansion medium, 500 mL.
Medium B—ClonaCell-HY fusion medium, 500 mL.
Medium C—ClonaCell-HY Hybridoma Recovery medium,
100 mL.
Medium D—ClonaCell-HY Hybridoma Selection medium
containing hypoxanthine, aminopterin, and thymidine
(HAT), 90 mL.
Medium E—ClonaCell-HY Hybridoma growth medium
containing hypoxanthine and thymidine (HT), 500 mL.
Polyethylene Glycol—ClonaCell-HY PEG solution, pretested
for cell fusion, 1.5 mL.
Store according to supplier’s instructions.
5. Additional supplies for detecting hybridoma antigen-specific
antibody production (e.g., using ELISA, immunocytochemis-
try, immunoblotting or flow cytometry).
2.3. Transfection, 1. A cell line capable of growing in suspension under nonadherent

Selection and Cloning conditions, e.g., CHO-S (available from ATCC; www.atcc.org).
of Cell Lines For examples of cell lines that have been found suitable for
selection and cloning in methylcellulose-based media (see

Note 1).
2. A liquid medium suitable for expanding the cell line prior to
transfection, maintaining the cells during the recovery phase
immediately after transfection before plating in semisolid
medium and for expanding cloned cell lines after selection and
cloning, e.g., ClonaCell-CHO CD Liquid (catalog #03817,
STEMCELL), for CHO-S and other CHO sublines that have
been adapted to suspension cultures (see Note 2).
3. An appropriate DNA vector containing the gene of interest
and a selectable marker gene, encoding, e.g., for resistance to
neomycin, hygromycin-B, or other antibiotics.
4. Antibiotic appropriate for the selectable marker gene used,
e.g., G418, Hygromycin B.
5. Reagents and/or equipment for DNA transfection of eukary-
otic cells.
6. ClonaCell-TCS, methylcellulose-based semisolid medium for
transfected cell selection (catalog # 03814, STEMCELL).
7. ClonaCell-CHO-CD chemically defined protein-free medium
(Catalog # 03815, STEMCELL) or ClonaCell-CHO-ACF
Medium (Catalog # 03816, STEMCELL) for selection and
cloning of CHO cells in the absence of serum.
8. A methylcellulose stock solution, e.g., 2.6% methylcellulose in
Iscoves’s MDM (Catalog # 04100, STEMCELL) for prepar-
ing a semisolid selection medium with a liquid medium of
choice (see Notes 2–4).
3. Methods
3.1. Hybridoma An overview of the protocol is shown in Fig. 1.

Development
3.1.1. Prepare Myeloma 1. Thaw the parental myeloma cells and culture in ClonaCell-HY
Cell Line Pre-Fusion Medium (Medium A) for at least 1 week prior to
fusion to ensure that the cells are well adapted to ClonaCell-HY
medium. Seed cells at approximately 5 × 104 cells/mL and
passage every 2 days. The suggested maximum cell density is
approximately 4 × 105 cells/mL, although a cell density of up
to 8 × 105 cells/mL is acceptable.
2. Calculate the cell growth rate at every passage. The day before
the fusion, count the viable cells and split so that at least 2 × 107
parental myeloma cells will be available for fusion. The cells
should be at early-mid log phase growth prior to fusion.
Fig. 1. Procedure for hybridoma selection and cloning in semisolid medium.
The recommended cell density for fusion is 2 × 105 cells/mL.

Only 100 mL of these cells is needed, but 200 mL should be
cultured to ensure sufficient cell numbers for fusion.
3. Harvest the parental myeloma cells by centrifuging in a 50 mL
conical centrifuge tube at room temperature (RT) at 300 × g
for 10 min. Wash three times by using 30 mL of ClonaCell-HY
Fusion Medium (Medium B). Remove the supernatant by
pipette and resuspend the cell pellet in 25 mL of Medium B
(see Note 5).
4. Count live cells using a viability stain (e.g., Trypan Blue). The
viability of parental myeloma cells should be >95%.
5. Calculate the volume of cell suspension that contains 2 × 107
viable cells. Keep cells at RT or 37°C until fusion (up to ~3 h).
3.1.2. Harvest Spleen 1. Sacrifice an immunized mouse according to procedures

and Prepare Spleen Cells recommended by your institution and wash the fur with 95%
ethanol. Clip fur and pull back to expose chest. Remove spleen
and place in a sterile petri dish containing 5 mL of Medium A.
Trim off any large pieces of fatty tissue.
2. Disaggregate the spleen into a single cell suspension. Transfer
the spleen to a nylon or metal screen placed on top of a 50 mL
conical centrifuge tube, and use the plunger of a 3 mL syringe
to grind the cells out of the spleen. Rinse the screen with
Medium B to assist the cells through the screen. Only the
spleen membrane should remain on the screen. Gently pipette
the cells up and down to disrupt clumps.
3. Wash splenocytes three times in 30 mL of Medium B, centri-
fuging at 400 × g (~1,350 rpm) at RT for 10 min each time and
removing the supernatant by pipette. After the final wash resus-
pend the cells in 25 mL Medium B (see Note 5).
4. Count live cells using viability stain (e.g., Trypan Blue).
Calculate the volume of cell suspension that contains 1 × 108
cells. Keep cells at RT or 37°C until fusion (up to ~3 h).
3.1.3. Cell Fusion 1. Prepare PEG and media (Medium A, B, C) for fusion by pre-
warming to 37°C. If using fusion Method A, prepare a 37°C
water bath.
2. Add 2 × 107 parental myeloma cells and 1 × 108 viable spleno-
cytes to a 50 mL conical centrifuge tube and centrifuge for
10 min at 400 × g. Aspirate off supernatant taking care not to
disrupt cell pellet. Complete removal of the supernatant is
essential to avoid dilution of PEG in the next step.
3. Fuse cells using one of the two methods outlined below.
Method A 1. Disrupt the cell pellet by gently tapping the bottom of the
tube. The pellet must be disrupted for optimal fusion. Slowly
add 1 mL of ClonaCell-HY PEG Solution (PEG) to the pellet
dropwise using a 1 mL pipette, over a period of 1 min without
stirring. Continually stir the cells gently, with the pipette tip,
over the next minute.
2. Add 4 mL Medium B to the fusion mixture, continuously
stirring as before, over a period of 4 min.
3. Slowly add 10 mL Medium B to the cells. Incubate for 15 min
in water bath at 37°C.
4. Slowly add 30 mL of Medium A and centrifuge the cells at
400 × g for 7 min.
5. Discard the supernatant and wash cells with 40 mL of Medium
A to ensure that all PEG is removed.
6. Slowly resuspend the cell pellet in 10 mL of ClonaCell-HY

Hybridoma Recovery Medium (Medium C). Transfer the cell
suspension to a T-75 cm2 tissue culture flask containing 20 mL
of Medium C (total culture volume = 30 mL). Incubate for
16–24 h at 37°C in 5% CO2.
Method B 1. Disrupt the cell pellet by gently tapping the bottom of the
tube. Add 0.5 mL of ClonaCell-HY PEG Solution (PEG)
dropwise to the pellet using a 1 mL pipette. Centrifuge
the mixture at 133 × g at RT for 3 min. Aspirate off all PEG
(see Note 6).
2. Carefully add 5 mL of Medium B dropwise to the pellet while
gently swirling the tube to resuspend the cells.
3. Slowly add 5 mL of ClonaCell-HY Hybridoma Recovery
Medium (Medium C) to the solution. Continue to swirl the
tube.
4. Transfer the cell suspension to a T-75 cm2 tissue culture flask
containing 20 mL of Medium C (total culture volume = 30 mL).
Incubate for 16–24 h at 37°C in 5% CO2. There will still be
clumps of cells at this point which will break up overnight. Be
gentle with these cells.
3.1.4. Selection and 1. On the day of the fusion, place ClonaCell-HY Hybridoma
Cloning of Hybridoma Cells Selection Medium (Medium D) at 2–8°C and thaw overnight.
On the day after the fusion, shake the bottle vigorously to mix
contents well and let warm to RT.
2. Transfer fused cell suspension into a 50 mL conical tube and
centrifuge for 10 min at 400 × g at RT. Remove the superna-
tant. Resuspend the cells in Medium C to a total volume of
10 mL. It is critical not to exceed the 10 mL final volume.
If you wish to add any additional cytokines or supplements
to Medium D, include this volume in the total 10 mL.
3. Transfer the 10 mL cell suspension into the 90 mL of Medium
D. Mix thoroughly by gently inverting the bottle several times.
Let sit for 15 min to allow the bubbles to rise to the top.
Using a 12 mL syringe and 16 gauge blunt-end needle,
aseptically plate out 9.5 mL of cell suspension medium into
each of ten 100 mm petri plates (see Note 7). Tilt each plate to
evenly distribute the medium to cover the bottom of the plate.
Avoid the introduction of bubbles during plating.
4. Incubate plates at 37°C in 5% CO2 (see Note 7). Do not dis-
turb plates for 10–14 days.
3.1.5. Screening 1. 10–14 days after cells are plated in Medium D, examine the
and Harvesting of Clones plates for the presence of colonies visible to the naked eye
(Fig. 2) (see Note 8). Remove isolated colonies from the plates
Fig. 2. Colonies of hybridoma cells after 14 days of culture in ClonaCell-HY Medium D.
using a pipettor set to 10 μL and sterile pipette tips. Pipette

each clone into an individual well of a 96-well tissue culture
plate containing 200 μL of ClonaCell-HY Growth Medium
(Medium E). With the pipettor set at 150 μL, pipette the entire
contents of the well up and down several times to resuspend
the cells. Ensure a new sterile tip is used for each colony to
maintain clonality of the colony.
2. Incubate the plates at 37°C in 5% CO2 for 1–4 days without
feeding. By the fourth day, most wells should have a high cell
density and medium that is turning yellow. As the colonies
have different growth rates, media in some wells may turn
yellow sooner than 4 days. It is a good idea to pick clones of
different sizes as slower growing clones (i.e., smaller colonies)
are often very good antibody producers.
3. Transfer 150 μL of supernatant from each hybridoma to a
separate well on a new 96-well plate and analyze by using an
assay system appropriate for the antigen involved (e.g., ELISA,
flow cytometry, Western Blotting, etc.).
4. Add 150 μL of fresh Medium E to every well of the original
hybridoma containing plates and return the plates to the
incubator.
5. Gently resuspend the hybridomas that showed a positive

response in step 3. Transfer 100 μL of cells to each of two wells
of a 24-well plate, containing 1 mL of Medium E.
6. When cells have grown to a suitable density (approximately
4 × 105 cells/mL), freeze the cells from one well and expand
the remaining positive clones in a T-25 cm2 tissue culture flask
containing 5 mL of Medium A and 5 mL of Medium E. This
step adapts the cells to growth in Medium A. In addition, keep
a sample of cells in Medium E, in case the cells don’t adapt well
to the 1:1 mixture. The cryopreserved cells serve as backup in
case the cultured cells are lost, or in case antibody expression is
lost and recloning of the cultured cells is unsuccessful.
7. When cells have grown to a suitable density (approximately
4 × 105 cells/mL), transfer 5–10 mL of cell culture by pipette
into 20 mL of Medium A in a T-75 cm2. Adjust the volume of
cells to ensure the final cell concentration is between 1 × 104
and 5 × 104 cells/mL. Maintain expanded hybridomas in 100%
Medium A at a concentration of 5 × 104–5 × 105 cells/mL.
More aliquots of cells can be frozen at this point in order to
secure the supply of the hybridoma clone.
8. Recloning of the clone of interest may be performed if desired
(see Note 9).
3.2. Transfection, An overview of the protocol is shown in Fig. 3.

Cloning, and Selection The following is a general procedure which may require addi-
of Eukaryotic Cell tional modifications and necessary controls based on the choice of
Lines cell line, vector, transfection method, and other experimental
requirements. Although many nonadherent cell lines and some
adherent cell lines grow well in semisolid medium and form tight
colonies, some do not. Cell lines that have been demonstrated to
grow in semisolid media are listed (see Note 1). For cell lines that
have not been tested before using this procedure the cloning
efficiency and growth properties (specifically, colony size, morphol-
ogy, cell viability after colony harvesting) should be determined
prior to using semisolid medium for selection and cloning after
transfection or for subcloning established cell lines. Since antibi-
otic resistance of a given cell line is dependent on the culture con-
ditions and may vary over time with continued passaging or
subcloning, it may be necessary to assess the minimum lethal anti-
biotic concentration in semisolid medium prior to transfection (see
Note 10 and Table 1).
3.2.1. Transfection Prior to transfection cells should be maintained in an appropriate

growth medium at logarithmic phase of growth. Nucleic acids may
be introduced into eukaryotic cells using various chemical, lipid or
physical methods (1–5).The efficiency of each method will vary
depending on the cell line and vector used, and the transfection
Fig. 3. Procedure for transfected cell line selection and cloning in semisolid medium.
conditions will need to be optimized for each cell line. After

transfection, cells should be incubated in growth medium without
antibiotics for 24–48 h before commencing the selection and
cloning procedure.
3.2.2. Cloning and Transfection efficiency and cell survival depend on a number of
Selection in ClonaCell-TCS factors including the gene transfected, the cells used and the trans-
or ClonaCell-CHO fection method employed. Optimal cell numbers per plate need to
Table 1
Suggested antibiotic concentrations required for selection of transfected cell lines
in ClonaCell-TCS medium
Cell line
Antibiotic BAF3 Molt4 K562 Jurkat Daudi UT-7 FD5 CHO-S
Puromycin (μg/mL) 1 0.5 N.D. 0.2 0.25 0.3 N.D. N.D.

G418 (mg/mL) 1 2 1.2 1 1 0.5 0.5 1.6
Hygromycin (mg/mL) N.D. 0.5 1 0.5 0.15 0.3 N.D. N.D.
be determined for different cell lines and applications. An example

of the conditions and concentrations required for 1 × 107 trans-
fected cells (ten 10 cm plates at 106 cells/plate) is described
below.
1. Thaw ClonaCell-TCS or ClonaCell-CHO medium overnight
at 2–8°C. Warm the bottle to RT before proceeding to next
step.
2. Shake bottle of ClonaCell medium vigorously to mix contents.
Let bubbles rise the top (approximately 10 min).
3. The antibiotic used for selection of stable transfectants will
vary depending on the selectable marker present on the plasmid
used in the transfection. The concentration of the selective
agent required will need to be determined for each cell line and
will need to be performed in the medium used for the selection
of transfectants. Prepare antibiotic solution: Add 10× final
antibiotic concentration required for selection (see Note 10).
Any other compounds, e.g., cytokines (if required) can be
added to the growth medium, but the total volume should not
exceed 10 mL.
4. Using a 12-mL syringe and 16-gauge blunt-end needle, dis-
pense 8 mL of ClonaCell medium to each of ten 14 mL Falcon
sterile tubes. Add 1 mL of the 10× antibiotic and any other
supplements to the 8 mL of ClonaCell-TCS in each tube. Mix
tubes thoroughly by inverting several times (see Note 7).
5. Harvest the transfected cells into a 50 mL conical tube and
centrifuge for 10 min at 400 × g. Resuspend the cells in a total
volume of 10 mL liquid growth medium without antibiotic.
6. Plating at several cell densities is recommended as transfection
efficiency may vary from experiment to experiment. Optionally,
to plate cells at different plating densities (e.g., 2 × 106, 1 × 106,
and 0.5 × 106 cells per plate), resuspend the cells in a total
volume of 5 mL, remove 2 mL, and add 1 mL to each of two

tubes with medium suspension (see step 7 below). To the
remaining 3 mL of cell suspension add 3 mL of liquid growth
medium, mix well, and remove 3 mL, and add the cell suspen-
sion to each of three tubes of semisolid medium (step 7).
To the remaining cell suspension add 2–3 mL of liquid growth
medium, mix well, and dispense 1 mL to each of the remaining
five tubes with culture medium (step 7).
7. Add 1 mL of cell suspension to each tube. Mix tubes thor-
oughly by inverting several times and let sit for 15 min to allow
air to rise to the top. Thorough mixing is essential to ensure
thorough distribution of antibiotics and cells throughout the
viscous methylcellulose-based cloning medium.
8. Using 12-mL syringes and 16-gauge blunt-end needles, asep-
tically plate out 9.5 mL from each tube into separate 10 cm
sterile petri dishes. Tilt the plates gently to level mixture, being
careful to avoid trapping of air (see Note 7).
9. Incubate plates at 37°C in 5% CO2 and >95% humidity. Do not
disturb plates for 7–14 days. The time required to see macro-
scopic colonies varies depending on the cell line used for trans-
fection and the antibiotic used for selection.
3.2.3. Harvest 1. After 7–14 days of culture, depending on the cell type and
and Screening antibiotic used for selection, examine the plates for the pres-
ence of colonies that are visible to the naked eye. Place plates
in a biosafety cabinet and aseptically remove isolated colonies
from the plates using a pipettor set to 10 μL, and sterile pipette
tips. Pipette each clone into an individual well of a 96-well
tissue culture plate containing 200 μL of growth medium
containing the specific antibiotic used in the selection process.
Incubate the plates at 37°C in 5% CO2 for 1–4 days without
feeding. Alternatively, an automated colony harvester may be
used (see Note 8). It is recommended to pick colonies of
different sizes, as slower growing transfectants which produce
smaller colonies may be very good protein producers. Such
slow growing transfectants are often missed in other transfec-
tion screening procedures. Usually by the fourth day, each well
has a high cell density and the medium has begun to turn
yellow.
2. Transfer 150 μL of each cell suspension to a separate well on a
96-well plate and assay for expression of the desired transfected
gene product (e.g., ELISA, Flow Cytometry, Western Blotting,
etc.). Add 150 μL of fresh growth medium containing anti-
biotic to every well of the original plate.
3. Transfer 2 × 100 μL of cell suspension of the positive clones
identified in step 2 to each of two wells of a 24-well plate
containing 1 mL of growth medium and antibiotics.
4. When cells have grown to a suitable density, cryopreserve

the cells from one well and expand the other in increasing
volumes of growth medium. If desired, the concentration of
antibiotics in the culture media may be decreased or even
eliminated after several passages. However, some transfectants
are unstable and need selective pressure to be maintained, or
need to be recloned periodically. The cryopreserved cells serve
as backup in case the cultured cells are lost, or in case protein
expression is lost and subcloning of the cultured cells is
unsuccessful.
4. Notes
1. Selection and cloning of transfected cell lines in semisolid

medium works best with cell lines that can grow as nonadherent
cells in suspension cultures or that can be adapted to growth
under nonadherent conditions. The following nonadherent
cell lines have been found to grow well in semisolid ClonaCell-
TCS medium: BaF/3 (murine pro-B cell), Molt-4 (human T
cell lymphoma), K562 (human myeloid-erythroid cell), Jurkat
(human T lymphoblastoid), Daudi (human B lymphoblastoid),
UT-7 (human T cells), and FD-5 (murine pre-myeloid).
In addition various adherent cell lines have been shown to
adapt well to nonadherent culture conditions and to form
colonies in semisolid medium (i.e., BHK-21, CHO, HEK293).
Although nonadherent cell lines are more likely to grow in
semisolid media, not every nonadherent cell line grows as well.
For example, the TF1, KG1, and M1 cell lines don’t form
distinct colonies after plating in ClonaCell-TCS medium.
If there is no prior knowledge of the ability of a cell line to
grow in semisolid medium or the cell line will be used with a
different semisolid medium than used in earlier experiments, it
is recommended to test the ability of the cell line to grow as
nonadherent colonies in the semisolid medium before using
this cell line in transfection experiments (see Note 11). Non-
tissue culture treated plastic ware should be used.
2. Adaptation of cell lines to growth in the same base medium as
that used in the semisolid medium may be necessary for opti-
mal plating efficiency. Culture media from different suppliers
have different compositions and ability to support clonal
growth of different cell lines. Cells that have been cultured in
a specific medium will over time result in a unique subpopula-
tion with optimal growth in that particular medium. These
cells may adjust poorly to a rapid switch to a semisolid medium
with different media composition, but may grow better if
adapted gradually to the same liquid medium as used in the

semisolid medium. Adaptation prior to plating in semisolid
media is particularly important when serum-free or protein-
free media are used, as cells grown in such media often have
slower growth rate and lower cloning efficiency than cells
grown in serum-containing media. Serum-free and protein-
free media (e.g., ClonaCell-CHO-ACF and ClonaCell-
CHO-CD) are often preferred for cell line development, in
particular for expression of recombinant proteins with poten-
tial therapeutic applications. Cells can be adapted slowly to a
new medium by gradually increasing the relative volume of
new medium during consecutive passages. The growth rate
and cell viability should be closely monitored at each passage.
Some reduction in growth rate may be expected when switch-
ing cells from a serum-containing medium to a serum-free or
protein-free medium. The cloning efficiency may also decrease
as the growth of cell lines in serum-free and protein-free media
may be more dependent on cell density than in serum-containing
media. After transfection the cells may need to be plated at
higher cell densities than cell lines grown in serum-containing
media to ensure adequate colony numbers. However, the
number of colonies is also determined by the efficiency of
the transfection and antibiotics selection. As these variables are
often unknown it is recommended to plate newly transfected
cells at a range of cell densities.
3. The procedures and ClonaCell-HY media for selection and
cloning of hybridoma cell lines described in this chapter have
been optimized for mouse hybridomas. Selection and cloning
in semisolid media may also be useful for generation of mono-
clonal antibodies from other species, specifically rats, rabbits,
and hamsters, but different medium formulations and/or
culture conditions may need to be selected. If established
hybridomas from these species are already available the useful-
ness of ClonaCell-HY for selection and cloning hybridomas
from these species could be tested in a cloning experiment
(see Note 11). The formulation that gives highest recloning
efficiency (preferably >50%, but lower efficiencies may be
acceptable), and that supports the development of distinct and
large colonies with good morphology, consistency (i.e., not
runny or hazy) and high cell viability after plucking can then be
used for selection and cloning of newly generated hybridoma
cell lines.
4. If a suitable liquid culture medium formulation is already used
in one’s lab for selection and cloning of transfected cell lines in
suspension cultures one could prepare a semisolid version of
this medium by combining the liquid medium with a stock
solution of culture-tested methylcellulose, (e.g., MethoCult
H4100 or M3134, STI), It may be useful to prepare several

media, each containing a different methylcellulose concentra-
tion (e.g., between 0.8 and 1.2%), and test these media in a
cloning experiment using nontransfected cells (see Note 11).
The medium formulation that gives highest cloning efficiency,
best colony size and morphology, and highest cell viability can
then be used for selection and cloning of new transfectants.
5. It is important to remove all the serum adhering to the cells,
by washing with serum-free Medium B. If the serum is not
removed, the PEG will not fuse the cell membranes and the
fusion frequency will drop drastically.
6. It is important to completely break up the cell pellet prior to
adding PEG in order to ensure efficient fusion of the cells.
During this procedure, not all cells will form a pellet, as some
will clump in the PEG. Do not aspirate the clumped cells. Work
quickly since cells must not be exposed to PEG for too long or
cell viability will drop.
7. Methylcellulose is a viscous solution and cannot be accurately
dispensed using pipettes due to adherence of the medium to
pipette walls. Syringes with blunt-end needles, rather than reg-
ular hypodermic needles should be used for aliquoting the
media to prevent needle-stick injuries. It is recommended to
put the plates in a separate plastic container together with an
open 100 mm petri dish containing 10 mL sterile distilled
water to maintain moisture content in methylcellulose cultures.
Open and close the incubator door carefully to avoid shaking.
It is important not to disturb the plates for the first 10 days as
this may result in dispersed colonies.
8. Manual picking of colonies from the semisolid medium is the
most time- and labor intensive step of the cloning and selec-
tion procedure. In a typical experiment several hundreds of
colonies can be obtained and it may not be possible to harvest
each colony present in each culture dish. An automated colony
harvester, such as the ClonaCell EasyPick (catalog # 30000,
STI) may be used to reduce the amount of manual manipula-
tion of the cultures and increase the number of colonies that
are harvested. Alternatively, the procedures as described in
Subheadings 3.1.4 and 3.2.2 can be modified to enable
screening of clones prior to plucking and thus reduce the total
number of colonies that need to be plucked. In this modified
procedure the cells are not plated in 10 cm dishes after
resuspending in the semisolid medium as described in
Subheading 3.1.4, step 3 and Subheading 3.2.2, step 7, but
instead are distributed over individual wells of 96-well plates at
60–80 μL per well. After culture, the wells that have colonies
in them, or alternatively all wells, are carefully overlaid with
150 μL of liquid medium (Medium E for hybridomas; an
appropriate liquid medium for transfected cell lines). The cultures

are incubated at 37°C for 2 days to allow antibodies or recom-
binant proteins to diffuse into the liquid medium. A fraction of
the overlaid liquid medium (e.g., 100 μL) is carefully removed
without disturbing the colonies in the semisolid medium and
tested in appropriate screening assays. Colonies in positive
wells are then harvested and expanded following the standard
procedures. A major advantage of this procedure is that only
colonies in positive wells need to be harvested. A disadvantage
is that there may be more than one clone in a single well. If this
is the case, each colony needs to be harvested from the positive
wells, expanded and rescreened to identify which clone is
positive. If the colonies overlap and it is not possible to harvest
them separately from the well, it will be necessary to reclone
the cells to ensure that a monoclonal cell line is obtained that
produces the antibody or recombinant protein of interest.
9. Newly generated hybridomas and transfected cell lines may be
recloned to select for subclones that have better growth rate
and/or higher protein production than the parental clone.
The plating efficiency of different hybridomas and transfected
cell lines may be variable. Therefore, a range of plating densi-
ties should be used, (e.g., 100, 500, 1,000 cells per 10 cm
dish) to ensure that at least one of the dishes will yield enough
individual colonies for harvesting and testing and will not be
overplated.
10. Before selection and cloning of transfected cell lines in semi-
solid media it is important to determine the minimum anti-
biotic concentration that is lethal for non-transfected cells.
The optimal antibiotic concentration will have to be deter-
mined by titration of the antibiotic in a cloning experiment
using non-transfected cells (see Note 9). It is not sufficient to
use information from liquid cultures as antibiotic sensitivity
may be different between liquid and semisolid culture condi-
tions and may also be dependent on the medium formulation.
Nontransfected cell should be plated at approximately the same
density that will be used when selecting for transfectants, or
the antibiotics titration could be done at a range of plating
densities to ensure that useful results will be obtained. Each set
of cultures should be supplemented with a range of different
concentrations of the appropriate antibiotic (e.g., G418 or
hygromycin, dependent on the selectable marker in the vector
chosen for transfection). After culture, the number of colonies
that are visible to the naked eye should be counted and the
minimal antibiotic concentration required to kill all the cells
established. This is the antibiotic concentration that should be
used to select the cells after transfection. An example of typical
antibiotic concentrations used for selection of different cell
lines in ClonaCell-TCS is shown in Table 1.These data are only

meant as indication of the concentration ranges for different
antibiotics. The growth characteristics of cell lines may change
over time, and antibiotics from different sources may have
different activities. Therefore the minimum lethal antibiotic
concentration should be determined in each lab using the cell
line, culture media, and antibiotic preparations selected for use
in the subsequent transfection experiments.
11. To test the growth of a cell line in semisolid medium, non-
transfected cells are plated at a range of cell concentrations,
e.g., 100, 500, and 1,000 cells per 10 cm petri dish. After
culture colonies are counted and the plating efficiency is
calculated according to the following formula: Plating
efficiency = number of colonies counted × 100%/number of
cells plated. If the plating efficiency is high at a range of plating
densities, e.g., >20%, and the colonies are distinct and non-
overlapping, it is likely that the cell line will form colonies after
transfection, assuming that other conditions (specifically, trans-
fection efficiency, antibiotic concentration) are adequate as
well. If a cell line produces no or only very few colonies in the
semisolid medium, even at high plating densities, it may require
a semisolid medium with different composition that better
supports clonal growth of these cells or it may need to be
adapted first to growth in a liquid version of the semisolid
medium to be used in transfection and cloning experiments
(see Notes 2–4).
References
1. Sambrook J, Fritsch EF, Maniatis T (1989) 3. Kreigler M (1990) Gene transfer and expression:
Molecular cloning: a laboratory manual, 2nd a laboratory manual. Stockton Press, New York
edn. Cold Spring Harbor Laboratory Press, 4. Ravid K, Freshney RI (1998) DNA transfer to
New York cultured cells. Wiley, New York
2. Carey M, Smale ST (2000) Transcription reg- 5. Dasso M (2005) Expression and introduction of
ulation in eukaryotes: concepts, strategies, and macromolecules into cells. In: Bonifacino JS,
techniques. Cold Spring Harbor Laboratory Dasso M, Harford JB (eds) Current protocols in
Press, New York cell biology. Wiley, New York, pp 20.0.1–20.0.2
Chapter 10
Isolation and Characterization of Mouse Side

Population Cells
Aysegul V. Ergen, Mira Jeong, Kuanyin K. Lin, Grant A. Challen,
and Margaret A. Goodell
Abstract
The side population (SP) is a subpopulation of mouse bone marrow cells highly enriched for hematopoietic
stem cell activity. The SP is identified using flow cytometry as a minor population that efficiently effluxes
the DNA-binding dye Hoechst 33342 relative to the rest of the bone marrow. Phenotypic and functionally
analysis has established SP cells as highly phenotypically homogeneous and functional active. In this
chapter we describe a detailed protocol for the purification of murine bone marrow SP cells based on
Hoechst dye efflux in combination with the presence of HSC surface markers.
Key words: SP, Hoechst 33342, Dye efflux, Hematopoietic stem cell, Purification, Side population
1. Introduction
1.1. Side Population The side population (SP) was first found in murine hematopoietic
Cells stem cells (HSCs) in bone marrow by their ability to pump out
fluorescent DNA-binding dye Hoechst 33342 (1). Hoechst 33342
binds to the AT-rich region of the DNA and emits primarily in the
blue range (around 450 nm) and also has a weaker red emission
(>675 nm) component. When these two emission wavelengths are
detected and plotted against each other, the “side population” can
be easily resolved (Fig. 1). In a dot plot of emission spectra they
appear on the side of the staining pattern and constitute a discrete
population of cells with an emission profile that differs from that of
the other cells (Fig. 1). This side population consists of highly
enriched HSCs and comprises 0.02–0.15% of the whole mouse
bone marrow cells depending on the age and gender. The frequency
151
152 A.V. Ergen et al.
Fig. 1. SP profile of unenriched murine bone marrow sample. Flow cytometric profile of
SP population is visualized after staining bone marrow cells with 5 μg/ml Hoechst 33342.
Signals are displayed in a Hoechst Blue vs. Hoechast Red dot plot. The PMT voltages are
adjusted until the majority of cells are at the upper right corner, whereas red blood cells
and debris are at the lower left corner. SP cells (~0.02–0.05% of whole bone marrow)
are very distinct and small subset of cells at the left side of the plot. PI positive cells
(dead cells) are much brighter in the Hoechst Red channel.
of SP cells steadily increases with age corresponding with the

increasing proportion of HSCs in the bone marrow over time.
SP cells have been found in the hematopoietic tissues of
various animal species including mice, monkeys, and humans (2),
in cell lines and primary cells from a variety of tissues and tumor
types (3–6). SP cells are characterized by their high expression of
multidrug-resistance ABC transporters such as transporter p-gly-
coprotein (MDR1) and ABCG2, and these transporters are the
major molecular mechanism of efflux activity of the Hoechst 33342
dye and many types of chemotherapy agents (3, 7). SP cells were
shown to be sensitive to the ABC transporter protein inhibitor
verapamil which reverses their phenotype (1, 8). SP displayed
elevated expression of ABCG2, and the ABCG2 knockout mouse
was demonstrated to have a severe reduction in the SP (7, 9).
However, some non-SP cells were also found to express a detect-
able level of ABCG2, indicating that ABCG2 expression is essential
but not sufficient to characterize SP phenotype (7, 9, 10). It is
likely that multiple multidrug-resistance transporters contribute to
the SP phenotype (11).
10 Mouse Side Population Cells 153
1.2. SP Cells and HSC The purification of HSCs has been substantially improved by the
Surface Markers application of flow cytometry with combinatorial SP staining and
surface marker staining. The SP assay is a novel method by which
rare HSC populations in the bone marrow can be identified without
surface markers. Subsequent multiparameter flow cytometric anal-
ysis of mouse bone marrow SP cells showed that approximately
95% expressed HSC surface markers and exhibited the highest
hematopoietic repopulating activity. HSC activity is determined by
quantifying the long-term repopulation of the transplanted cells to
the peripheral blood. First it was found that HSC activity resides in
cells that express c-kit (K) and Sca-1 (S) and do not express any of
several surface markers found on different more mature blood cells
(lineagenegative, L). KSL is the canonical cell surface marker cocktail
that is used to enrich for HSCs for more than a decade. However,
it is still a very heterogeneous population that includes lineage-
primed multipotent progenitors as well as short-term HSCs and
long-term HSCs. Additionally, several alternative or improved
HSC enrichment approaches have been developed. More studies
have identified a number of additional HSC cell surface antigens
including Thy1.1 (12), CD34 (13), Flk-2 (14), the Tie-2 (15),
endoglin (16), Epcr (17), and CD150 (18). Cells within the SP are
very similar in terms of expression of canonical stem cell markers.
SP highly overlaps with HSCs isolated via classical cell surface
marker schemes KSL-Thy1loCD34−Flk2− or EPCR+ CD48−. Unlike
all other markers, CD150 shows a bimodal distribution on the SP
(Fig. 2) (19). While both CD150+ and CD150− cells from the
SP are functional HSC (19), the CD150+ subset has greater long-
term self-renewal and engraftment potential, but a myeloid-biased
lineage differentiation output, and thus may be selected if the
most homogeneous and most potent HSC population is desired.
We typically purify HSCs by the phenotype of SP + KSL + CD150
(called SPKSL CD150+) (Fig. 2).
1.3. Functional Recent studies have identified new HSC subtypes with distinct
Characterization functional properties within previously characterized populations.
of SP Cells Our group showed HSCs from different regions of the SP, desig-
nated as lower SP and upper SP, possess different functional poten-
tials. Lower SP cells predominantly generated myeloid cells with
great self-renewal potential, whereas upper SP cells were much
more effective at generating lymphoid cells (19). Other groups
reported similar findings using different enrichment strategies.
The Eaves group used combinations of CD150, EPCR, CD48,
and CD45, for the enrichment of HSCs, and they showed lymphoid
vs. myeloid patterns associated with the absence or presence of
CD150 (20). They showed that HSC with higher repopulating
activity and strong myeloid bias are enriched in the CD150+ subset
of EPCR+CD48−CD45+ bone marrow cells, whereas those in the
CD150− subset have limited self-renewal activity and a lymphoid
Fig. 2. The SPKSL CD150+ (SP, c-Kit+, Sca-1+, Lin−, CD150+) cells. SP cells are co-stained with Sca-1, c-Kit, CD150 and
lineage marker antibodies in order to exclude low level contamination of progenitor cells. This sample was pre-enriched
with Sca-1 antibody using magnetic sorting.
differentiation bias. Both the Nakauchi and Rossi groups also

detected the same lineage bias associated with CD150 expression
(21, 22). They demonstrated that CD150high subsets of CD34−KSL
exhibit the highest long-term HSC activity correlating with persis-
tent myelopoiesis and CD150high HSCs can give rise to CD150high
as well as to CD150low and CD150neg HSCs, but CD150low and
CD150neg HSCs fail to give rise to CD150high cells, suggesting that
CD150high HSCs reside at the top of the HSC hierarchy.
Usually, only a few thousand HSCs can be obtained from one
mouse, and even this small population seems very heterogeneous
and the cells differ in their functional properties when assayed on
an individual level. Therefore, to ensure successful purification of
the most homogenous HSC based on the SP, it is optimal to

use the SP in combination with conventional cell surface marker
staining methods. Furthermore, because HSC are present at such
a low proportion in the bone marrow, the highest purities are
achieved by first enriching for stem cells using magnetic enrich-
ment for cells expressing a stem cell marker (e.g., Sca1+ or c-Kit+)
or lack of differentiation markers (lineage depletion). In the
following sections, we introduce techniques for the purification
of highly homogeneous long-term HSCs by combining both SP
and HSC surface marker staining methods.
2. Materials
2.1. Isolation of Bone 1. C57Bl/6(B6) mice, 8–10 weeks of age (see Note 1).
Marrow SP Cells 2. DMEM+: Dulbecco’s Modified Eagle’s Medium (DMEM)
with high glucose (Cat. No. 11965-092, Gibco Invitrogen)
supplemented with penicillin/streptomycin (Cat. No. 15140-
122, Gibco Invitrogen), 10 mM HEPES (Cat. No. 15630-
080, Gibco Invitrogen), and 2% fetal bovine serum (FBS).
3. HBSS+: Hank’s balanced salt solution (HBSS, Cat. No. 14170-
112, Gibco Invitrogen) supplemented with 10 mM HEPES
(Cat. No. 15630-080, Gibco Invitrogen) and 2% FBS.
4. Hoechst 33342 powder (Cat. No. B2261, Sigma) is dissolved
in distilled water and filter sterilized at 1 mg/ml concentration
which makes 200× stock and frozen at −20°C. One whole
bottle of powder is used to make ~500 ml of Hoechst stock
solution at once, and frozen in small (~1 ml) aliquots. Thawed
Hoechst powder may be less reliable after re-freezing, possibly
due to acquisition of water.
5. Red blood cell lysis buffer (D-5001, Gentra).
6. Verapamil (Cat. No. V-4629, Sigma) is dissolved in 95% ethanol
as a 5 mM 100× stock. Stored at −20°C in 100 μl aliquots.
7. Dissecting tools, scissors, and forceps.
8. 18-G and 27-G needles.
9. 40 μm Cell strainers (Cat. No 22363547, Fisher).
10. 15 and 50 ml Conical polypropylene centrifuge tubes, sterile
(Falcon).
11. 10-cm Tissue culture dishes.
12. Refrigerated centrifuge.
13. 250 ml Polypropylene tubes (Cat. No 430776, Corning).
14. Circulating water bath at exactly 37°C.
Table 1
Monoclonal antibody list for purification of SPKSLCD150+
Antibody Clone Conjugate Dilution Company
Mac-1 M1/70 PE-Cy5 1:100 eBioscience

Gr-1 RB6-8C5 PE-Cy5 1:100 eBioscience
B220 RA3-6B2 PE-Cy5 1:100 eBioscience
Ter119 TER119 PE-Cy5 1:100 eBioscience
CD4 RM4-5 PE-Cy5 1:100 eBioscience
CD8 53-6.7 PE-Cy5 1:100 eBioscience
Sca-1 E13-161.7 Biotin 1:100 BD Pharmingen
c-Kit 2B8 AlexaFluor-750 1:100 eBioscience
CD150 TC15-12F12.2 PE 1:100 BioLegend
15. Biotinylated Sca-1 antibody (Cat. No 553334, BD

Pharmingen).
16. Anti-biotin magnetic microbeads from Miltenyi Biotech
(Cat. No. 130-090-485).
17. Magnetic separation machine: autoMACS.
18. Monoclonal Antibodies (Table 1).
19. Propidium iodide (Cat. No. P-4170, Sigma) is dissolved at
200 μg/ml in PBS as 100× stock and covered with aluminum
foil and kept in 4°C fridge. Final concentration of PI in HBSS+
should be 2 μg/ml.
20. Flow/sorting equipment with UV laser capable of excitation at
350 nm and detection with 450/20 and 675LP optical filters.
3. Methods
3.1. Harvesting Bone HSCs have the ability to efflux Hoechst dye which appears as the
Marrow Cells side population in FACS (Fig. 1) after staining with Hoechst
33342. Reproducible SP staining is dependent on many parame-
ters such as Hoechst concentration, cell number, staining tempera-
ture, and time. A proper Hoechst staining will yield an SP
population comprising 0.02–0.05% (Fig. 1) of whole bone mar-
row cells from ~8-week-old C57Bl/6 mice (see Note 2). To
increase the yield, a magnetic-based enrichment of progenitor cells
using a canonical cell surface marker (Sca-1 or c-Kit) can be per-
formed prior to FACS. Thus, an enrichment protocol, which
provides tenfold enrichment for bone marrow SP (Fig. 2) is also

provided. It is generally expected that 3,000–5,000 HSCs can be
purified from one 10–12-week-old mouse. Moreover, it is impor-
tant to note that the percentage of bone marrow HSCs increases
with age, and this is also reflected in the percentage of SP cells.
One may expect to obtain up to 10,000 from a mouse ~1-year-old
(however, these HSC are of lower quality in terms of function).
1. Warm-up DMEM+ medium in a 37°C water bath. It is critical
to have the temperature of water bath exactly 37°C.
2. Euthanize C57Bl/6 mice of 8–10 weeks of age. Dissect out
femora and tibiae from mice and remove all the muscle and
connective tissue from the bone using scissors and forceps.
Additional bones can be used as desired, such as the hips which
have additional bone marrow. Place the bones in ice-cold
HBSS+. Keep them on ice throughout the process.
3. Trim the ends of the each bone and flush out the bone marrow
into a sterile tissue culture dish using a syringe (5–10 ml) with
a 27-G needle that is filled with ice-cold HBSS+. Flush from
both ends to ensure all the marrow is removed. Bones should
be very pale after flushing of the bone marrow. Crushing of all
the bones with mortar and pestle can also be used to isolate
bone marrow cells which results in more cells. Spines and other
bones can also be used to collect more bone marrow cells.
4. Change the needle to 18-G and pass bone marrow-media
mixture through 18-G needle several times in order to make a
single cell suspension, while trying to avoid excessive bubble
formation which reduces cell viability.
5. Filter cells through a 70 μm cell strainer into a 50 ml falcon
tube.
6. Count nucleated cells (see Note 3). In order to avoid counting
red blood cells (RBC), an aliquot of the bone marrow cell
suspension can be removed and mixed with RBC lysis buffer
for counting. Take out 5 μl from bone marrow suspension and
mix it with 95 μl of RBC lysis buffer, vortex thoroughly, and
take 10 μl to count cells using a hemacytometer. Do not use
RBC lysis for the whole bone marrow suspension. This proce-
dure generally yields an average of 40–70 million nucleated
cells per C57Bl/6 mouse (2 femur and 2 tibias).
7. Spin down the cells in a refrigerated centrifuge (1,050 × g, for
6 min at 4°C).
8. Remove supernatant. Resuspend cell pellet at 106 cells/ml in
pre-warmed DMEM+. Polypropylene tubes must be used
while staining with Hoechst to avoid retention of cells in tubes.
For large volumes, staining in 250 ml polypropylene tubes is
the most convenient method.
3.2. Hoechst Staining 1. Add Hoechst to a final concentration of 5 μg/ml.

2. Incubate cells in a circulating 37°C water bath for exactly
90 min (see Note 4).
3. Spin down the cells in a centrifuge at 1,050 × g for 6 min at 4°C
and remove the supernatant. Resuspend cells at 108 cells/ml in
ice-cold HBSS+.
4. In order to ensure optimal HSC purification, cells should be
co-stained with antibodies such as Sca-1, c-Kit, CD150, and
lineage markers. Antibodies are added at concentrations deter-
mined by standard antibody titration procedures or as recom-
mended by the manufacturer (e.g., Becton Dickinson/
Pharmingen). All staining and centrifugation should be
performed at 4°C.
5. When samples are ready for fluorescent activated cell sorting
(FACS), resuspend cells in cold HBSS+ with 2 μg/ml propidium
iodide (PI) to distinguish and eliminate dead cells.
3.3. Magnetic Hoechst-stained cells can be enriched for progenitors by using

Enrichment of biotinylated Sca-1 or c-Kit antibodies. This will increase the yield,
Hoechst-Stained Cells increase the purity, and decrease the time required for sorting.
and Antibody Staining 1. For Sca-1 enrichment, add biotinylated Sca-1 antibody to the
cell suspension at 1/100 dilution, and incubate on ice for
15 min. Alternatively, biotinylated c-Kit antibody can also be
used for the enrichment.
2. Wash out unbound antibodies with tenfold volume of ice-cold
HBSS+.
3. Centrifuge the cells for 6 min at 1,050 × g, at 4°C and remove
the supernatant.
4. Resuspend cells at 108 cells/ml in ice-cold HBSS+.
5. Label cells with magnetic beads. We typically use anti-biotin
magnetic microbeads from Miltenyi Biotech, but alternatives
are also effective. Incubate cells with 20% volume of micro-
beads and place at 4°C in the fridge for 15 min. It is recom-
mended to incubate cells in the fridge instead of on ice, because
of the low binding efficiency of microbeads on ice.
6. Wash the cells with a tenfold volume of ice-cold HBSS+.
the supernatant.
8. Resuspend cells at 2 × 108 cells/ml in ice-cold HBSS+.
9. Load the cells into autoMACS (or alternative) column.
10. Take the positive fraction from the autoMACS column and
wash with ice-cold HBSS+.
the supernatant.
12. Resuspend cells at 1 × 108 cells/ml in ice-cold HBSS+.

13. Prepare the stem and lineage marker antibody cocktail. Mix
monoclonal antibodies of anti-mouse c-Kit, anti-mouse
CD150, lineage markers of anti-mouse CD4, anti-mouse
CD8a, anti-mouse B220, anti-mouse Gr-1, anti-mouse Mac-1,
and anti-mouse Ter-119 at 1/100 dilution. Use streptavidin
conjugated fluorescently labeled antibody to verify magnetic
enrichment.
3.4. FACS Analysis Analysis of SP cells has been performed on a variety of instruments,
for Hoechst SP Cells but we have had the most experience with cytometers from either
BD (Aria) or Cytomation (MoFlo). In order to view the SP popu-
lation, an ultraviolet laser is needed to excite the Hoechst 33342
dye and PI. A violet laser has also been used with good results (23).
Excitation of the Hoechst dye occurs at 350 nm and the emission
of Hoechst dye is measured with Hoechst Blue and Hoechst Red
detectors. Ideally, lasers with 100 mW of power give the best
results, but lasers with lower power have been used successfully.
Hoechst Blue is measured with a 450/20 band pass (BP) filter and
red is measured with a 675 edge filter long pass (EFLP; Omega
Optical, Brattleboro VT) filter. Emission wavelengths are separated
with a 610 dichroic mirror short pass (DMSP). Fluorescence of PI
is also measured with the 675EFLP filter, when excited with
350 nm. Although other filter sets similar to these ones works fine,
these give better resolution of the SP.
1. Samples stained with Hoechst are placed on the cytometer and
kept cold by a chilling apparatus if possible.
2. First, Hoechst fluorescence is displayed with Hoechst Blue
(450BP filter) on the vertical axis vs. Hoechst Red on the hori-
zontal axis, both in linear mode. Voltage adjustments are made
so that red blood cells can be viewed in the lower left corner
(they have no nuclei so uptake of the DNA-binding Hoechst
dye is minimal) and dead cells which are stained brightly with
PI are seen against on the far right in a vertical line. The majority
of the cells can be viewed in the center or in the upper right
quarter (Fig. 1). A major GO-G1 population with S-G2M cells
going toward the upper right corner can also be detected.
3. In order to obtain an SP profile similar to the one shown in
Fig. 1, a sample gate is drawn to exclude red blood cells and
dead cells. 50,000–100,000 events should be collected within
this sample gate for an unenriched bone marrow sample.
The SP region should be similar to that shown in Fig. 1.
The SP prevalence is around 0.01–0.05% of an unenriched
whole bone marrow in the mouse (see Note 5).
4. SP cells are highly enriched for HSCs in mouse bone marrow.
With a proper Hoechst staining, 60–80% of them are lineage
negative and Sca-1 and c-Kit positive. These cells can be

further separated for CD150 expression which marks for
myeloid-biased HSC population. In young mice, 25–40% of
SPKSL cells express CD150, whereas in the old mice this
percentage is increased to 60–85%. In order to confirm proper
SP staining, verapamil can be included in a separate control
sample (the bulk of the SP should be eliminated when Hoechst
staining in the presence of verapamil) (see Note 6).
4. Notes
1. This protocol is first developed for mouse HSC from bone

marrow of C57Bl/6 mice, thus optimization may be required
for other strains and tissues. We recommend using C57Bl/6
bone marrow first to establish the protocol before establishing
Hoechst staining on other species or tissues.
2. Staining conditions are critical. Thus, the staining protocol
should be followed precisely, otherwise it will result in low-
quality of Hoechst stain, hence a decrease in the purity of
HSCs. The concentration of Hoechst dye, numbers of the
cells, staining temperature, and time are all parameters which
have an effect on the SP profile. It is also very important to
keep cells at 4°C after Hoechst staining to prevent further
Hoechst efflux. Ficoll or other extended higher temperature
procedures should not be performed after the Hoechst staining.
The increased temperature will allow other bone marrow cells
to efflux Hoechst dye, resulting in an artificially increased
percentage of SP which is contaminated with non-stem cells.
3. A precise cell count of the bone marrow suspension is critical
for a successful Hoechst staining, so counting correctly is
important. As mentioned in Note 2, incorrect counting may
result in low-quality staining and low purity.
4. Temperature and time are crucial for Hoechst staining, so
DMEM+ should be warmed up and tubes should be fully
immersed in the water bath in order to maintain the tempera-
ture of cells at 37°C. Tubes should be mixed periodically
during incubation to ensure equal exposure of the cells to
the dye.
5. A very high proportion of SP cells (>0.05%) in normal mouse
bone marrow may indicate poor staining and contamination of
the population with non-stem cells. This problem can also be
identified by identifying the presence of a large number of cells
within the SP population that are Lineage + or do not express
canonical HSC cell surface markers Sca-1 and c-Kit. In a proper
Hoechst staining, 75–95% of the SP cells will have the canonical

HSC cell surface markers (Sca-1 and c-Kit positive, lineage
negative, CD34-negative/low, Flk2-negative). Sorting on at
least two of these surface markers in addition to Hoechst
staining will ensure the highest purity. We typically sort on SP
and Kit+, Sca+, and Lineage-neg.
6. Confirmation of SP cells can be made by including a Verapamil-
treated control which blocks the SP phenotype. Verapamil
(Sigma, 100× stock made in 95% EtOH) is used at 50 μM final
concentration and is added throughout the entire Hoechst
staining. Verapamil treatment will result in the absence of SP
which confirms SP identity.
Acknowledgments
The authors are supported by grants from the NIH, the Ellison
Foundation, and the American Heart Association. M.J. was
supported by University of Science and Technology though
UST Post-Doc Research Program. G.A.C. is a scholar of the
American Society of Hematology.
References
1. Goodell MA, Brose K, Paradis G, Conner AS, 6. Dean M (2009) ABC transporters, drug resis-
Mulligan RC (1996) Isolation and functional tance, and cancer stem cells. J Mammary Gland
properties of murine hematopoietic stem Biol Neoplasia 14(1):3–9
cells that are replicating in vivo. J Exp Med 7. Zhou S, Schuetz JD, Bunting KD, Colapietro
183(4):1797–1806 AM, Sampath J, Morris JJ et al (2001) The
2. Goodell MA, Rosenzweig M, Kim H, Marks ABC transporter Bcrp1/ABCG2 is expressed
DF, DeMaria M, Paradis G et al (1997) Dye in a wide variety of stem cells and is a molecu-
efflux studies suggest that hematopoietic stem lar determinant of the side-population pheno-
cells expressing low or undetectable levels of type. Nat Med 7(9):1028–1034
CD34 antigen exist in multiple species. Nat 8. Moserle L, Indraccolo S, Ghisi M, Frasson C,
Med 3(12):1337–1345 Fortunato E, Canevari S et al (2008) The side
3. Hirschmann-Jax C, Foster AE, Wulf GG, population of ovarian cancer cells is a primary
Nuchtern JG, Jax TW, Gobel U et al (2004) A target of IFN-alpha antitumor effects. Cancer
distinct “side population” of cells with high Res 68(14):5658–5668
drug efflux capacity in human tumor cells. 9. Zhou S, Morris JJ, Barnes Y, Lan L, Schuetz JD,
Proc Natl Acad Sci USA 101(39): Sorrentino BP (2002) Bcrp1 gene expression
14228–14233 is required for normal numbers of side popula-
4. Challen GA, Little MH (2006) A side order of tion stem cells in mice, and confers relative
stem cells: the SP phenotype. Stem Cells protection to mitoxantrone in hematopoietic
24(1):3–12 cells in vivo. Proc Natl Acad Sci USA
5. Patrawala L, Calhoun T, Schneider-Broussard 99(19):12339–12344
R, Zhou J, Claypool K, Tang DG (2005) Side 10. Scharenberg CW, Harkey MA, Torok-Storb B
population is enriched in tumorigenic, stem- (2002) The ABCG2 transporter is an efficient
like cancer cells, whereas ABCG2+ and Hoechst 33342 efflux pump and is preferen-
ABCG2− cancer cells are similarly tumorigenic. tially expressed by immature human hematopoi-
Cancer Res 65(14):6207–6219 etic progenitors. Blood 99(2):507–512
11. Venezia TA, Merchant AA, Ramos CA, 18. Kiel MJ, Yilmaz OH, Iwashita T, Terhorst C,
Whitehouse NL, Young AS, Shaw CA et al Morrison SJ (2005) SLAM family receptors
(2004) Molecular signatures of proliferation distinguish hematopoietic stem and progenitor
and quiescence in hematopoietic stem cells. cells and reveal endothelial niches for stem
PLoS Biol 2(10):e301 cells. Cell 121(7):1109–1121
12. Morrison SJ, Weissman IL (1994) The long- 19. Challen GA, Boles NC, Chambers SM, Goodell
term repopulating subset of hematopoietic MA (2010) Distinct hematopoietic stem cell
stem cells is deterministic and isolatable by subtypes are differentially regulated by TGF-
phenotype. Immunity 1(8):661–673 beta1. Cell Stem Cell 6(3):265–278
13. Osawa M, Hanada K, Hamada H, Nakauchi H 20. Kent DG, Copley MR, Benz C, Wohrer S,
(1996) Long-term lymphohematopoietic Dykstra BJ, Ma E et al (2009) Prospective
reconstitution by a single CD34-low/negative isolation and molecular characterization of
hematopoietic stem cell. Science 273(5272): hematopoietic stem cells with durable self-
242–245 renewal potential. Blood 113(25):
14. Christensen JL, Weissman IL (2001) Flk-2 is a 6342–6350
marker in hematopoietic stem cell differentiation: 21. Morita Y, Ema H, Nakauchi H (2010)
a simple method to isolate long-term stem cells. Heterogeneity and hierarchy within the most
Proc Natl Acad Sci USA 98(25):14541–14546 primitive hematopoietic stem cell compart-
15. Hirao A, Arai F, Suda T (2004) Regulation of ment. J Exp Med 207(6):1173–1182
cell cycle in hematopoietic stem cells by the 22. Beerman I, Bhattacharya D, Zandi S,
niche. Cell Cycle 3(12):1481–1483 Sigvardsson M, Weissman IL, Bryder D et al
16. Chen CZ, Li M, de Graaf D, Monti S, Gottgens (2010) Functionally distinct hematopoietic
B, Sanchez MJ et al (2002) Identification of stem cells modulate hematopoietic lineage
endoglin as a functional marker that defines potential during aging by a mechanism of
long-term repopulating hematopoietic stem clonal expansion. Proc Natl Acad Sci USA
cells. Proc Natl Acad Sci USA 99(24): 107(12):5465–5470
15468–15473 23. Simpson C, Pearce DJ, Bonnet D, Davies D
17. Balazs AB, Fabian AJ, Esmon CT, Mulligan (2006) Out of the blue: a comparison of
RC (2006) Endothelial protein C receptor Hoechst side population (SP) analysis of
(CD201) explicitly identifies hematopoietic murine bone marrow using 325, 363 and
stem cells in murine bone marrow. Blood 407 nm excitation sources. J Immunol
107(6):2317–2321 Methods 310(1–2):171–181
Chapter 11
Stem Cell Identification by DyeCycle Violet

Side Population Analysis
William G. Telford
Abstract
Hoechst side population (SP) analysis remains a critical technique for identifying stem cell and progenitor
populations in hematopoietic and non-hematopoietic tissues, as well as potential cancer stem cells. More
recently, DyeCycle Violet (DCV), a DNA binding dye structurally similar to Hoechst 33342 but with an
excitation spectrum shifted toward the violet range, has also been used for SP analysis on flow cytometers
equipped with violet laser diodes. In this chapter, we briefly review the history of this method and provide
a detailed procedure. Critical parameters for good labeling, details on integrating simultaneous immuno-
labeling with DCV SP analysis, and proper data acquisition and analysis techniques are covered in detail.
Key words: Stem cell, Progenitor, Flow cytometry, Side population, Hoechst 33342, DyeCycle Violet
1. Introduction
Since its original development nearly 15 years ago, Hoechst side

population (SP) analysis remains an important technique for identi-
fying stem cells and early progenitors in both hematopoietic and
non-hematopoietic tissues (1, 2). The Hoechst side population
phenomenon was originally discovered during cell cycle analysis of
hematopoietic cells using the cell-permeable DNA binding dye
Hoechst 33342 (1). When murine hematopoietic cells were labeled
with Hoechst dye and analyzed on a fluorescence-activated cell
sorter equipped with an ultraviolet laser, most nucleated cells pro-
duced a familiar cell cycle distribution. However, a small number of
cells showed considerably less Hoechst dye fluorescence, having
rapidly effluxed the dye via a transmembrane pump. When analyzed
through both a traditional blue ~450 nm Hoechst dye filter and a
163
164 W.G. Telford
mouse bone marrow

256K
fluorescence (450/50 nm
192K
Hoechst 33342 blue
128K
64K
SP
64K 128K 192K 256K

Hoechst 33342 red
fluorescence (675/20 nm
Fig. 1. Hoechst 33342 side population (SP) analysis of unpurified mouse bone marrow. Cells were analyzed on a BD
Biosciences LSR II using a Cobolt Zouk 355 nm laser emitting at 10 mW.
red ~675 nm filter and displayed in a bivariate plot, this population

of cells appeared as a “tail” projecting from the normal G0/G1 pop-
ulation (Fig. 1). When these “side population” (SP) cells were
sorted and transferred into irradiated immune-deficient mice, 100
or fewer of these cells could reconstitute all hematopoietic lineages
(1, 2). These SP cells also expressed cell surface markers of stem
cells and progenitors, including Sca-1 and c-kit and were negative
for mature lineage markers (3). Collectively this evidence suggested
that the SP cell population represented stem cells or progenitors.
Further work has demonstrated that the SP phenomenon,
while by no means universal for the stem cell phenotype, is wide-
spread in both tissues and mammalian species. SP cells are present
in rodent, nonhuman primate and human hematopoietic and non-
hematopoietic tissues (2). Hoechst side population (SP) analysis
has proven to be a valuable technique for isolating candidate stem
cells from a variety of non-hematopoietic tissues (4–6). SP cells are
also present in tumors and may be identifying potential cancer stem
cell populations in some tumor types (7–11). As known markers
for tissue-specific and cancer stem cells are few, SP analysis has pro-
vided an important starting point for identifying and purifying
these populations (7).
The ABCG2 transporter pump (Breast cancer resistance pro-
tein or Bcrp-1 in mice), a member of the ABC transporter family,
11 DCV SP Analysis 165
is primarily responsible for dye efflux in a variety or normal and

drug resistant cells. ABCG2 knockout mice do not exhibit the SP
phenomenon (12–15) and stem cells and hematopoietic tissues in
these animals show normal development. A Hoechst SP phenom-
enon is therefore present, but not essential, for normal stem cell
development. Its purpose in stem cell physiology therefore remains
a mystery. The SP phenotype is not completely universal for stem
cells; a number of studies have shown non-SP cell populations to
contain both normal and cancer stem cells, and stem cell popula-
tions do not necessarily possess SP activity (16–19).
While Hoechst SP has been applied to a variety of stem cell
studies, the need for specialized cytometry equipment has limited
its use. Good excitation of Hoechst 33342 requires an ultraviolet
laser. Due to its high cost and complexity, ultraviolet lasers have
not been widely employed in flow cytometers. The original ultra-
violet laser sources for flow cytometry were argon- and krypton-
ion water-cooled lasers, which are large, maintenance intensive,
and difficult to integrate into all but the largest cell sorters. More
recently, a new generation of solid state diode, Nd:YAG and
Nd:YVO4 ultraviolet lasers have been developed that are smaller,
easier to maintain and can be more readily integrated into bench
top analyzers and small cell sorting systems. However, they remain
expensive and are infrequent additions to flow cytometers (20).
Advances in solid-state laser technology have led to the wide-
spread introduction of inexpensive violet laser diodes into more
advanced flow cytometers (21). This has permitted the exploita-
tion of a variety of violet-excited fluorescent probes for flow cyto-
metric analysis. A number of attempts have been made to perform
Hoechst SP analysis using violet laser excitation, since Hoechst
33342 is somewhat excited at the violet laser wavelength. While
several laboratories reported some success in using violet excitation
for SP analysis, most studies have shown a significant loss of side
population resolution (22–24). This is understandable when the
excitation spectrum for Hoechst 33342 is examined: the 400–
410 nm emission range of violet laser diodes should give only 2–5%
excitation of Hoechst 33342 compared to a more optimal UV
excitation source (Fig. 2). More information on lasers applicable
for SP analysis can be found in Note 1.
Since violet laser diodes are now common fixtures on flow
cytometers, a search was then made to find a replacement for
Hoechst 33342 that demonstrated the same ABCG2 pump
specificity but with more optimal excitation in the violet range.
A cell-permeable DNA binding dye with the trade name DyeCycle
Violet (Invitrogen Life Technologies) was subsequently identified
that possessed similar cell permeability characteristics to Hoechst
33342, but demonstrated an excitation spectra that was shifted
approximately 20 nm toward to the violet (25) (Fig. 2). When
loaded into mouse hematopoietic cells and analyzed on instruments
166 W.G. Telford
Violet laser
EX diode
EM
Relative fluorescence (%)
Hoechst
33342
Hoechst
33342
DCV
DCV
300 350 400 450 400 500 600

Wavelength (nm) Wavelength (nm)
Fig. 2. Excitation and emission spectra of DyeCycle Violet (DCV), in comparison with Hoechst 33342.
Fig. 3. DCV SP analysis of mouse bone marrow, human bone marrow, and human cord blood using either an ultraviolet or
violet laser source (top and bottom rows, respectively). Forward versus side scatter dot plot at left is mouse bone marrow.
with either UV or violet lasers, DyeCycle Violet (DCV) also dem-

onstrated a side population, differing somewhat in form but simi-
larly corresponding to the Sca-1+ c-kit+ Lin− phenotype (Fig. 3).
DCV also possessed the same ABC pump specificity as Hoechst
33342 (25). DCV has subsequently been employed for SP analysis
on instruments without ultraviolet lasers but with violet laser diodes
(26, 27). The following procedure describes how to load cells with
DCV, equip a cytometer for DCV SP analysis, and analyze the col-
lected data. While the techniques for Hoechst and DCV SP cell
preparation are very similar, the distribution of DCV SP differs from
Hoechst SP and analysis should be carried out carefully. Unlike

Hoechst 33342, DCV is also somewhat excited at 488 nm, with
small but noticeable emission in the fluorescein and phycoerythrin
bandwidths. Precautions should therefore be followed in setting up
SP experiments with simultaneous immunolabeling.
2. Materials
The following reagents and equipment should be obtained and

prepared in advance.
2.1. Reagents 1. DyeCycle Violet (DCV). Available from Invitrogen Life

Technologies at a stock concentration of 5 mM in distilled
water. DCV should be stored at 4°C and should not be frozen.
DCV can generally be stored for approximately 6 months with-
out loss of activity. During this time the yellow color of the
solution may darken slightly, and a small precipitate can form.
These changes do not significantly reduce the efficacy of the
reagent, although this is an indicator of degradation and it
should be replaced.
For SP labeling, DCV should be prepared at 5–10 μM in
the DMEM+ buffer described below. Dye solutions should be
used the same day and the remainder discarded, as dye precipi-
tation can occur in phosphate buffers. DCV should not be
stored in diluted form.
2. Efflux inhibitors. An ABCG2 pump inhibitor should be added
to a control sample to allow accurate gating of the SP popula-
tion. Effective inhibitors include verapamil and fumitremorgin
C. Verapamil is a broad-spectrum inhibitor of organic anion
membrane transporters; fumitremorgin C is more selective for
specific members of the ABC transporter family.
(a) Verapamil: Prepare a 10 mM stock solution in DMSO, for
final dilution to 50 μM. Verapamil solutions can be stored
at 4°C for extended periods.
(b) Fumitremorgin C: Prepare a stock at 2 mM in distilled
water, for final dilution to 10 μM. Inhibitors can be stored
in frozen aliquots at −20°C.
3. Wash buffer (HBSS+). HBSS (without phenol red), supple-
mented with 2% fetal bovine serum, and 2 mM HEPES. Store
at 4°C prior to use.
4. SP buffer (DMEM+). DMEM low glucose (no phenol red),
with 2% fetal bovine serum, and 2 mM HEPES. See Note 2 for
additional information on buffers. Store at 4°C prior to use.
168 W.G. Telford
5. Propidium iodide (PI). This is used as a viability label to exclude

dead cells from the analysis. Prepare a 1 mg/ml stock in PBS,
for final dilution at 2 μg/ml. Unlike PI stocks used for cell
cycle analysis, no RNase should be added to this preparation.
The stock can be stored for up to 6 months at 4°C.
2.2. Equipment and 1. A 37°C water bath, centrifuge and source of ice for 4°C
Instrumentation storage.
2. A flow cytometer capable of detecting DCV SP. The instrument
must be equipped with either an ultraviolet or violet laser
source, plus two detectors aligned to this laser. Examples of
appropriate instruments include the BD Biosciences
FACSCanto series (I and II), LSR II, LSR Fortessa, FACSAria
series (I, II, and III) (http://www.bdbiosciences.com), the
Beckman-Coulter CyAn and Gallios (http://www.coulterflow.
com), the Stratedigm S-series (http://www.stratedigm.com),
the Partec CyFlow instruments (http://www.partec.de), and
the Sony iCyt instruments (http://www.i-cyt.com) that are
frequently equipped with violet and sometimes ultraviolet
lasers. Before preparing the cells for analysis, make certain your
instrument has the correct laser, and that it is equipped with
the necessary detectors (two required) and filters. Lasers are
described in more detail in Note 1.
SP cells frequently need to be separated using a fluorescence-
activated cell sorter. While conditions for analysis-only and
sorting are the same, special conditions may be required for
sorting of stem cells. Cell sorting of SP cells is described in
Note 3.
The violet-aligned detectors should have blue and red narrow
bandpass filters inserted in the correct positions prior to analy-
sis. Any Pacific Blue or DAPI bandpass filter (i.e., 450/40 nm,
440/10 nm, etc.) will work for the Hoechst or DCV blue
signal, and any APC or Cy5 filter (675/20 nm, 660/20 nm)
will work for the red. Depending on the instrument design, a
short-pass or long-pass dichroic mirror ranging from 560 to
610 nm will work to split the signals. A 580 nm long-pass
mirror is typically used on a BD Biosciences LSR II, FACSCanto
II, or FACSAria II. A ~600 nm short-pass filter will be used on
a Beckman-Coulter Gallios. Sample filter configurations are
shown in Fig. 4.
3. Alignment verification microspheres. Poor instrument alignment
can cause poor SP resolution, particularly if the UV or violet
laser is not properly aligned. Verify instrument alignment using
UV or violet excited alignment verification microspheres.
Examples are shown in Fig. 5. These include Spherotech
Rainbow Ultra microspheres (Spherotech, Libertyville, IL),
which are well-excited by both UV and violet lasers and can be
BD LSR II
BD LSR Fortessa
Beckman-Coulter Gallios
BD FACSAria I and II
BD FACSCanto II
DCV
DCV blue
red 45
0/5
0
5 /20
67 600SP
P
0L
58
0
5/2
67
450/
50 DCV
red
DCV
blue
Fig. 4. Detection filter configurations for BD Biosciences and Beckman-Coulter flow cytometers.
Fig. 5. Alignment verification microspheres analyzed with ultraviolet or violet lasers (top and bottom rows, respectively), includ-
ing Spherotech Rainbow Ultra (Spherotech, Libertyville, IL) and InSpeck Blue (Invitrogen Life Technologies, Carlsbad, CA).
170 W.G. Telford
used for daily quality control of your instrument. They emit far
brighter in the blue range than the red, so the blue detector
should be used for alignment verification. InSpeck Blue micro-
sphere arrays (Invitrogen Life Technologies, Carlsbad, CA) are
a mixture of seven bright-to-dim microspheres that also excite
with UV or violet lasers and emit in the blue range; they are even
more useful for identifying minor degradation in instrument
alignment. InSpeck Blue spheres will be somewhat better excited
by UV than violet lasers. Remember that many alignment
verification particles are not well-excited by UV or violet lasers.
3. Methods
Several excellent review articles and methods chapters cover

Hoechst side population (SP) labeling (28–31). The original paper
by Margaret Goodell is also still very relevant (1). The DCV SP
technique is discussed in refs. 25, 32.
Like Hoechst SP, DyeCycle Violet (DCV) SP labeling is
very sensitive to labeling conditions, and these instructions
should therefore be closely followed. DCV labeling is very
similar to Hoechst 33342 labeling, with the exception of the
dye concentration and spectral properties.
3.1. DCV Labeling 1. Prepare the cells to be labeled as a suspension in SP buffer

(DMEM+), at concentrations up to 5 × 106 cells/ml. Lower
3.1.1. Cell Preparation
concentrations can be used if necessary. Use washing buffer
(HBSS+) for cell washing steps, but suspend in SP buffer
(DMEM+) for DCV labeling.
2. Cell suspensions should be prepared according to protocols
specific to the tissue. Freshly isolated mouse bone marrow is
removed from both femurs and tibias from euthanized mice
using a 20 ml syringe containing HBSS+ and a 30 gauge nee-
dle. Cells in HBSS+ are carefully passed through an 18 gauge
needle to break up clumps. Cells are then filtered through
40 μm mesh, centrifuged at 400 × g for 7 min, and suspended
in DMEM+ at no more than 5 × 106 cells/ml.
3. For frozen cell suspensions, considerable clumping can occur
after thawing. DNase should be added to the wash medium
following thawing and should be included in the SP buffer. See
Note 4 for more information on the use of DNAse to reduce
clumping.
3.1.2. DCV Labeling 1. Divide the cell sample into two halves, one with no inhibitor
and the other as an inhibitor control. Warm the samples to
37°C in a water bath prior to labeling. As a rule, an efflux
inhibitor sample should be included for all samples.
2. Add inhibitor (verapamil or fumitremorgin C) to the control

sample. At the concentrations given above (10 and 2 mM,
respectively), this is a 1:200 dilution for both inhibitors (5 μl
per 1 ml). Incubate at 37°C for 15 min.
3. After pre-incubation with inhibitor, quickly add 2 μl of DCV
(5 mM stock solution) per 1 ml of cells, and mix gently (no
vortexing), for a final concentration of 10 μM. Incubate for
90 min at 37°C with gentle mixing at 30 min intervals. Keep
the cells suspended as much as possible during the incubation.
4. Centrifuge the sample tubes at 400 × g for 7 min, decant and
suspend in cold DMEM+ in the same volume as the labeling.
The tubes can be stored for up to 4 h on ice, but should be
analyzed as quickly as possible. Add propidium iodide at 2 μg/
ml final concentration (2 μl of a 1 mg/ml stock solution per
1 ml cell suspension) a few minutes before analysis. More infor-
mation on viability analysis can be found in Note 5. If immu-
nolabeling is to be performed after DCV labeling, do not add
PI until after antibody labeling.
3.1.3. DCV SP Labeling As with Hoechst 33342, DCV SP labeling is compatible with
with Simultaneous simultaneous immunolabeling for stem cell surface markers. Unlike
Immunophenotyping Hoechst 33342, however, DCV is somewhat excited by the 488 nm
laser, causing minor emission in the fluorescein and PE range in
DCV-labeled cells. While this emission is small, it can cause a
significant amount of background fluorescence in the fluorescein
and PE channels. These fluorochromes should therefore not be
used with DCV-labeled cells, although the PE channel can be used
for PI fluorescence. However, PE-Cy5, PE-Cy5.5, PE-Cy7, APC,
APC-Cy5.5, and APC-Cy7 are all spectrally compatible with DCV
labeling (although PE-Cy5 is sometimes reserved for PI viability).
Stem cell antibodies are now available as direct conjugates for all of
these fluorochromes. Simultaneous immunophenotyping should
be carried out after DCV labeling, once the cells are on ice.
If several overlapping surface markers are to be used, an “unla-
beled” and “single” color controls should also be prepared to set
instrument compensation. The “unlabeled” sample will have DCV,
and the “single” controls should include a single surface marker
and DCV labeling. When calculating compensation, the DCV
fluorescence is ignored and the control will be treated as being
labeled with a single fluorochrome.
1. Following incubation with DCV and washing by centrifuga-
tion as described above, suspend the cells in 200–500 μl HBSS+
and place on ice.
2. Add pre-titered antibody and incubate at 4°C for 15–30 min.
Make sure that the necessary single color controls are included
at this step. Add 3 ml of cold HBSS+ buffer and centrifuge at
400 × g for 5–7 min.
172 W.G. Telford
3. Repeat the above steps for additional antibodies if necessary.

For human cells, antibody labeling can usually be done in one
step. For mouse cell labeling, several steps separated by centri-
fuge washes may be necessary.
4. After the cells are labeled, return the suspension to its original
volume with DMEM+ and store at 4°C until analysis. As above,
add PI 5 min prior to analysis.
3.2. DCV SP 1. Set up the instrument for analysis. For DCV, either a UV or
Acquisition violet laser must be used, with two aligned detectors (often set
and Analysis up for Pacific Blue and Pacific Orange on many commercial
instruments). Specific examples are listed in Subheading 2.2,
3.2.1. DCV SP Analysis and filter configurations are shown in Fig. 4.
Without Simultaneous
Immunolabeling
2. Verify instrument alignment using an alignment verification
microsphere preparation (Fig. 5).
3. Set both DCV blue and red detectors for linear acquisition
(they may normally be set to log scaling).
4. Run the cells on the flow cytometer and display them in a for-
ward versus side scatter two-parameter dot plot. This is shown
in Fig. 6a for mouse bone marrow. Bone marrow can have a
complex multi-population forward versus side scatter profile,
so ensure all populations of interest are on scale. Gate on the
scatter populations of interest. For mouse bone marrow, the
SP cells usually fall between the smaller and large primary clus-
ters. However, it is advisable to save ALL cells during data
acquisition.
5. Then create a side scatter versus PI fluorescence dot plot, and
draw a gate for the PI-negative cells. The viable cell back-
ground fluorescence in the PI detector will be somewhat higher
with DCV than that normally observed for Hoechst 33342,
since DCV is somewhat excited at 488 nm. However, it should
still be possible to distinguish PI-negative viable cells from
PI-positive apoptotic and necrotic cells (Fig. 6b).
6. Finally, display a DCV red (X-axis) versus DCV blue (Y-axis)
dot plot, gated for scatter and PI viability. Adjust the voltages
on the blue and red detectors to place the dominant G1 popu-
lation roughly in the center of the dot plot. For DCV, the SP
population should arch up along the Y-axis, and eventually
curve back down toward the minimum points of both axes.
This separation will be much more pronounced than that nor-
mally observed for Hoechst 33342 (Fig. 6c). It is good prac-
tice to include the entire G1/G2 cell cycle of the sample on
the dot plot, although doing this may undesirably compress
the SP population into the lower left corner of the plot. If this
appears to be happening, the voltages can be increased and the
G2/M portion of the cell cycle allowed off-scale.
Fig. 6. Analysis of DCV SP in mouse bone marrow. Data was first visualized in a forward versus side scatter dot plot
(a), followed by forward scatter versus PI fluorescence (b), then DCV red versus blue fluorescence (c).
7. In most hematopoietic tissues, the SP population will be rare,

usually less than 0.05% of the viable nucleated cells. Collect at
least 500,000 events per sample as a minimum, several million
if possible. Some cytometers may have an upper limit on
acquired events.
8. Use the inhibitor control to set the upper limit for the SP. With
DCV, this cutoff may be lower on the SP arch than with
Hoechst 33342. In many cases, verapamil and fumitremorgin
C may not completely inhibit dye loss in the SP population.
3.2.2. DCV Analysis
Analysis of DCV SP with simultaneous immunolabeling is essen-
with Simultaneous
tially the same as without, except that the detection of surface
Immunolabeling
markers must be confirmed and set appropriately, and compensa-
tion between them should be carried out. DCV fluorescence should
be ignored when setting compensation; if using an automated
compensation routine, this requires a special setup described in
Subheading 3.2.3.
1. Prior to labeling, remember that fluorescein and PE cannot be
used with DCV, due to spectral overlap of DCV into these
detection bandwidths with 488 nm excitation. PI can be
detected in the PE channel, however.
2. Check all individual phenotypic markers on the cytometer and
set their detector voltages appropriately. When setting com-
pensation, ignore the DCV blue and red signals; very little
spectral overlap should occur if you are not using fluorescein or
PE. When using automated spillover routines, exclude DCV
blue and red signals from the calculation, since the software
will attempt to compensate DCV blue and red overlap, giving
an altered labeling pattern.
174 W.G. Telford
3.2.3. Automated It is usually necessary to calculate compensation matrices (spill-

Compensation with DCV over) automatically when doing complex multicolor experiments,
SP Labeling rather than attempt to determine compensation values manually.
Most cytometer acquisition software packages (BD DiVa, Beckman-
Coulter Kaluza, etc.) can calculate spectral overlap, as do most
third party cytometry software packages (i.e., Tree Star FlowJo, De
Novo Software FCS Express, Verity WinList). DCV SP analysis dif-
fers from normal multicolor compensation, since we do not want
to compensate between the DCV blue and red values or between
DCV and other fluorophores. However, DCV must be physically
present in the analysis. The procedure below describes how to
carry out software-based compensation analysis for simultaneous
cell surface phenotyping and SP on the BD DiVa systems, leaving
the SP parameters uncompensated. The procedure is illustrated in
Fig. 5. This approach can be adapted to other analysis packages.
As described previously, “unlabeled” and “single” color controls
will be needed to calculate compensation. The “unlabeled” sample
should actually be labeled with DCV, since these dyes will contrib-
ute slightly to the overall fluorescence of the other fluorophores and
needs to be taken into account. The “single” color controls should
be labeled with each surface marker and DCV. In essence, we will
pretend it is not there for the purposes of compensation calculation.
Compensation beads (such as BD Biosciences CompBeads) will not
work as controls, since they cannot bind DCV or PI.
1. Create a sample protocol. For a hypothetical experiment (with
PI for viability, PE-Cy7, APC and APC-Cy7 for surface label-
ing, and DCV), the samples would be
(a) “unlabeled”—labeled with DCV only
(b) “PI single color”—labeled with PI and DCV
(c) “PE-Cy7 single color”—labeled with PE-Cy7 and DCV
(d) “APC single color”—labeled with APC and DCV
(e) “APC-Cy7 single color”—labeled with APC-Cy7 and DCV
2. Start the DiVa software, and create a New Experiment. Go to
the Instrument Control Panel under the Parameters tab, and
delete all parameters except the ones you will need for your
experiment. Include DCV blue or red (or the equivalent in
your system), and PE or PE-Cy5 for propidium iodide
fluorescence (Fig. 7a). Remember to avoid fluorescein and PE
for immunlolabeling as 488 nm excited DCV fluorescence will
overlap into these channels. PE-Cy5, PE-Cy5.5, PE-Cy7,
APC, APC-Cy5.5, and APC-Cy7 can be used.
3. Change the DCV blue and red parameters to linear scaling.
4. In the Workspace, create a forward versus side scatter dot plot,
a forward scatter versus PI dot plot, and a DCV red (X-axis)
versus blue (Y-axis) dot plot (as described previously).
Fig. 7. Procedure for automated compensation of multicolor experiments including DCV SP analysis.
5. Analyze a DCV SP sample without saving. Adjust the DCV

blue and red voltage controls until the SP pattern is correct.
These settings must be made prior to saving the compensation
samples, since they cannot be changed once the controls are
acquired.
6. Go to the Instrument menu, then Instrument Setup, then
Create Compensation controls. The Create Compensation
controls panel will appear, showing all of the selected parame-
ters (Fig. 7b). Then delete the DCV blue and red parameters
from the Create Compensation Controls panel (Fig. 7c). They
will still be present on the Instrument Parameters panel and
will still be saved (Fig. 7d). Press OK.
7. Analyze the “unlabeled” sample (actually containing DCV)
and set the detector gains. Then run the “single” color control
samples as usual, with no DCV single color controls (although
the unlabeled and single controls must have DCV present).
Propidium iodide should also be run as a control. Moving the
scatter gate into the nonviable cell region will allow PI-positive
cells to be displayed for calculation purposes. Once the “unla-
beled” sample is analyzed, detector gains cannot be changed.
176 W.G. Telford
8. Once the controls are run, go to Instrument menu, Instrument

Setup, then Calculate Compensation. The spillover matrix will
be calculated. The DCV blue and red values will be included in
the matrix, yet will be set at 0%. The multicolor experiment can
now be analyzed.
4. Notes
1. Lasers. An ultraviolet or violet laser is essential for this tech-

nique. Ultraviolet lasers for flow cytometers include (1) water-
cooled argon-ion and krypton-ion sources, emitting at a
relatively high 50–200 mW in the 351–365 nm range; (2)
Nd:YVO4 frequency tripled solid state UV lasers, emitting at
355 nm, and (3) near-UV laser diodes, emitting at 370–
395 nm. Water-cooled gas lasers are mainly found on large-
frame cell sorters such as the BD Biosciences FACSVantage
DiVa, Coulter Altra and Beckman-Coulter (formerly Dako
Cytomation) MoFlo, and have been largely superseded by solid
state 355 nm sources. These solid state UV lasers are smaller in
size and can be integrated into high-end multicolor cytome-
ters. However, they remain expensive. Near-UV laser diodes
are smaller and less expensive, but have a somewhat longer UV
emission at ~375 nm. They work for Hoechst and DCV SP but
are less useful for other UV flow cytometry applications such
indo-1 calcium measurements (33). As a result, they are less
common on commercial flow cytometers. The BD Biosciences
FACSAria cell sorter series offers a near-UV laser diode as an
option for SP analysis.
Violet lasers include water-cooled krypton-ion gas lasers
(407 and 415 nm) and the much more common violet laser
diode. As with UV sources, krypton-ion lasers are large and
can only be installed on large cell sorters. They have also been
largely superseded by violet laser diodes, which are small, inex-
pensive and are now included on many high-end cytometers.
Their power levels now exceed 100 mW, making them equiva-
lent to older gas lasers.
2. Buffers. Buffers are a critical parameter for side population
analysis, both Hoechst 33342 and DCV. Modifications to the
DMEM+ buffer (SP buffer) can cause significant alterations in
the appearance and pattern of the SP population. If immuno-
labeling is to be carried out after DCV labeling, the DMEM+
SP buffer should be used for this purpose as well.
3. Cell sorting. Fluorescence-activated cell sorting for purification
of stem cells by Hoechst or DCV SP is at least as common as non-
sorting analysis. Most cell sorters employ jet-in-air technology,

the technology originally used to detect the SP. However, jet-
in-air systems show reduced sensitivity and resolution of the SP
region due to their less efficient optics. These systems include
the BD Biosciences FACSVantage series (most recently the
DiVa system), the Beckman-Coulter MoFlo series, the iCyt
Reflection, the BD Biosciences (formerly Cytopeia) InFlux sys-
tem, the Bay Biosciences JSAN, and the Beckman-Coulter
Astrios. An ultraviolet or violet laser of sufficient power is
highly recommended. The BD Biosciences FACSAria series (I,
II, and III) use a hybrid cuvette-based flow cell for initial cell
interrogation; this system is more similar to standard cuvette
cytometers, and lower power lasers can be used.
Sheath pressure is an important consideration when sort-
ing stem cells. Modern cell sorters can achieve sheath pressures
of nearly 100 psi. The sorted cells must therefore be able to
withstand a change from normal air pressure to the pressure of
the sorter, and the sudden reduction in pressure that occurs
when cells leave the nozzle. Some cell types, including lym-
phocytes and tumor cells, show good viability following sort-
ing under these conditions and can be sorted at high pressures.
However, stem cells frequently show less tolerance for this dif-
ferential, with loss of viability and often complete physical
destruction during and after the sorting process. Sheath pres-
sures of 25 psi or less are often required for successful stem cell
and progenitor sorting.
Since both Hoechst 33342 and DCV are DNA binding
dyes, it can be anticipated that labeling followed by ultraviolet
exposure can cause DNA damage, particularly single-strand
DNA breaks. DNA structural relaxation and single-strand
breaks following Hoechst 33342 labeling and UV exposure
have in fact been reported in murine bone marrow (34). This
potentially mutagenic effect will not affect analysis-only appli-
cations, but could be of critical importance for sorting
applications.
4. DNase. Cell freezing causes a certain amount of cell damage
and death, resulting in the release of free DNA into the media.
Upon thawing, this free DNA can cause undesirable cell clump-
ing, often excluding a large proportion of cells from the sus-
pension. To avoid this, DNase can be included in the freezing
media, the thawing media, and subsequent wash buffers. It
effectively prevents DNA-mediated cell aggregation. Obtain
powdered DNAse and prepare a 1 mg/ml stock in PBS con-
taining 0.5 M MgCl2 (a 100× stock). Add 10 μl of this stock
per ml to all cell buffers to avoid clumping.
5. Viability assessment. A viability probe (such as propidium
iodide) should always be included in any flow cytometric assay,
particularly Hoechst 33342 or DCV SP. Dead cells and debris
178 W.G. Telford
can also show low levels of DCV incorporation and can be eas-
ily mistaken for SP cells, particularly in non-hematopoietic tis-
sues where large amounts of debris are present. While it is
sometimes possible to distinguish dead cells and debris by for-
ward versus side scatter alone, this is very difficult in complex
cell mixtures like bone marrow and cord blood, and even more
difficult in dissociated tissue samples. It is certainly possible to
use other viability probes in place of PI, if this parameter needs
to be moved to another instrument detector. Amine-reactive
viability probes are available in a wide variety of fluorochromes
and can place the viability assessment in the APC or APC-Cy7
channel, for example.
References
1. Goodell MA, Brose K, Paradis G, Connor AS, 10. Haraguchi N, Utsunomiya T, Inoue H, Tanaka
Mulligan RC (1996) Isolation and functional F, Mimori K, Barnard GF, Mori M (2006)
properties of murine hematopoetic stem cells Characterization of a side population of cancer
that are replicating in vivo. J Exp Med cells from human gastrointestinal system. Stem
183:1797–1806 Cells 24:506–513
2. Challen GA, Little MH (2006) A side order of 11. Ho MM, Ng AV, Lam S, Hung JY (2007) Side
stem cells: the SP phenotype. Stem Cells population in human lung cancer cell lines and
24:3–12 tumors is enriched with stem-like cancer cells.
3. Weksberg DC, Chambers SM, Boles NC, Cancer Res 67:4827–4833
Goodell MA (2008) CD150− side population 12. Kim M, Turnquist H, Jackson J, Sgagias M,
cells represent a functionally distinct popula- Yan Y, Gong M, Dean M, Sharp JG, Cowan K
tion of long-term hematopoietic stem cells. (2002) The multidrug resistance transporter
Blood 111:2444–2451 ABCG2 (breast cancer resistance protein 1)
4. Dor FJ, Ramirez ML, Parmar K, Altman EL, effluxes Hoechst 33342 and is overexpressed in
Huang CA, Down JD, Cooper DK (2006) hematopoietic stem cells. Clin Cancer Res
Primitive hematopoietic cell populations reside 8:22–28
in the spleen: studies in the pig, baboon, and 13. Zhou S, Morris JJ, Barnes Y, Lan L, Schuetz
human. Exp Hematol 34:1573–1582 JD, Sorrentino BP (2002) Bcrp1 gene expression is
5. Kato K, Yoshimoto M, Kato K, Adachi S, required for normal numbers of side popula-
Yamayoshi A et al (2007) Characterization of tion stem cells in mice, and confers relative
side-population cells in human normal endo- protection to mitoxantrone in hematopoietic
metrium. Hum Reprod 22:1214–1223 cells in vivo. Proc Natl Acad Sci USA
6. Summer R, Kotton DN, Sun X, Ma B, 99:12339–12344
Fitzsimmons K, Fine A (2003) Side population 14. Zhou S, Zong Y, Lu T, Sorrentino BP (2003)
cells and Bcrp1 expression in lung. Am J Physiol Hematopoietic cells from mice that are deficient
Lung Cell Mol Physiol 285:L97–L104 in both Bcrp1/Abcg2 and Mdr1a/1b develop
7. Hadnagy A, Gaboury L, Beaulieu R, Balicki D normally but are sensitized to mitoxantrone.
(2006) SP analysis may be used to identify can- Biotechniques 35:1248–1252
cer stem cell populations. Exp Cell Res 15. Jonker JW, Freeman J, Bolscher E, Musters S,
312:3701–3710 Alvi AJ, Titley I, Schinkel AH, Dale TC (2005)
8. Brown MD, Gilmore PE, Hart CA, Samuel Contribution of the ABC transporters Bcrp1
JD, Ramani VA, George NJ, Clarke NW (2007) and Mdr1a/1b to the side population pheno-
Characterization of benign and malignant type in mammary gland and bone marrow of
prostate epithelial Hoechst 33342 side popula- mice. Stem Cells 23:1059–1065
tions. Prostate 67:1384–1396 16. Morita Y (2006) Non-side-population
9. Fukuda K, Saikawa Y, Ohashi M, Kumagai K, hematopoietic stem cells in mouse bone mar-
Kitajima M, Okano H, Matsuzaki Y, Kitagawa row. Blood 108:2850–2856
Y (2009) Tumor initiating potential of side 17. Alt R, Wilhelm F, Pelz-Ackermann O, Egger
population cells in human gastric cancer. Int J D, Niederwieser D, Cross M (2009) ABCG2
Oncol 34:1201–1207 expression is correlated neither to side popula-
tion nor to hematopoietic progenitor function from bladder cancer cells by DyeCycle Violet
in human umbilical cord blood. Exp Hematol staining. Cancer Biol Ther 7: 1663–1668
37:294–301 27. Mathew G, Timm EA Jr, Sotomayor P, Godoy
18. Addla SK, Brown MD, Hart CA, Ramani VA, A, Montecinos VP, Smith GJ, Huss WJ (2009)
Clarke NW (2008) Characterization of the ABCG2-mediated DyeCycle Violet efflux
Hoechst 33342 side population from normal defined side population in benign and malig-
and malignant human renal epithelial cells. Am nant prostate. Cell Cycle 8:1053–1061
J Physiol Renal Physiol 295:F680–F687 28. Montanaro F, Liadaki K, Schienda J, Flint A,
19. Triel C, Vestergaard ME, Bolund L, Jensen Gussoni E, Kunkel LM (2004) Demystifying
TG, Jensen UB (2004) Side population cells in SP cell purification: viability, yield, and pheno-
human and mouse epidermis lack stem cell type are defined by isolation parameters. Exp
characteristics. Exp Cell Res 295:79–90 Cell Res 298:144–154
20. Cabana R, Frolova EG, Thomas RA, Krishan A, 29. Goodell MA (2005) Stem cell identification and
Telford WG (2006) The minimal instrumenta- sorting using the Hoechst 33342 side popula-
tion requirements for Hoechst side population tion (SP). In: Robinson JP, Darzynkiewicz Z,
analysis: stem cell analysis on low-cost flow Dobrucki J, Hoffman RA, Nolan JP, Orfao A,
cytometry platforms. Stem Cells 24:2573–2581 Rabinovitch PS (eds) Current protocols in
21. Shapiro HM, Perlmutter NG (2001) Violet cytometry. Wiley, New York, pp 9.18.1–9.18.11
laser diodes as light sources for cytometry. 30. Lin KK, Goodell MA (2006) Purification of
Cytometry 44:133–136 hematopoietic stem cells using the side popula-
22. Telford WG, Frolova EG (2004) Discrimination tion. In: Lanza R and Klimanskaya I (eds)
of Hoechst side population in mouse bone Methods in enzymology, vol 420. Elsevier,
marrow with violet and near-UV laser diodes. Inc., New York, pp 255–264
Cytometry 57A:45–52 31. Petriz J (2007) Flow cytometry of the side pop-
23. Eaker SS, Hawley TS, Ramezani A, Hawley RG ulation (SP). In: Robinson JP, Darzynkiewicz Z,
(2004) Detection and enrichment of hemato- Dobrucki J, Hoffman RA, Nolan JP, Orfao, A.,
poetic stem cells by side population phenotype. Rabinovitch, P.S., eds. Current protocols in
In: Hawley TS, Hawley RG (eds) Methods cytometry. Wiley, New York, pp 9.23.1–9.23.14
molecular biology, vol 263, 2nd edn, Flow 32. Telford WG (2010) Stem cell side population
cytometry protocols. Humana, Totawa, NJ, side population analysis and sorting using
pp 161–180 DyeCycle Violet. In: Robinson JP,
24. Simpson C, Pearce DJ, Bonnet D, Davies D Darzynkiewicz Z, Dobrucki J, Hoffman RA,
(2006) Out of the blue: a comparison of Hoechst Nolan JP, Orfao A, Rabinovitch PS (eds)
side population (SP) analysis of murine bone Current protocols in cytometry, chap 9, Unit
marrow using 325, 363 and 407 nm excitation 9.30. Wiley, New York
sources. J Immunol Methods 310:171–181 33. Telford WG (2004) Analysis of UV-excited
25. Telford WG, Bradford J, Godfrey W, Robey fluorochromes by flow cytometry using a near-
RW, Bates SE (2007) Side population analysis UV laser diode. Cytometry 61A:9–17
using a violet-excited cell permeable DNA 34. Zhao H, Traganos F, Dobrucki J, Wlodkowic
binding dye. Stem Cells 25:1029–1036 D, Darzynkiewicz Z (2009) Induction of DNA
26. She JJ, Zhang PG, Wang ZM, Gan WM, Che damage response by the supravital probes of
XM (2008) Identification of side population cells nucleic acids. Cytometry A 75: 510–519
Chapter 12
Isolation and Characterization of Cancer Stem Cells In Vitro

Craig Gedye and Laurie Ailles
Abstract
The cancer stem cell hypothesis is an appealing concept to account for intratumoral heterogeneity and the
observation that systemic metastasis and treatment failure are often associated with the survival of a small
number of cancer cells. Whilst in vivo evidence forms the foundation of this concept, in vitro methods and
reagents are attractive as they offer opportunities to perform experiments that are not possible in an animal
model. While there is abundant evidence that existing cancer cell lines are not reliable models of tumor
heterogeneity, recent advances based on well validated novel cancer cell lines established de novo in defined
serum-free media are encouraging, particularly in the study of glioblastoma multiforme. In this chapter we
wish to broadly outline the process of establishing, characterizing, and managing novel cancer cell lines in
defined serum-free media, and discuss the limitations and potential opportunities that may arise from these
model systems.
Key words: Cancer stem cell, Defined serum-free media, Tumor-initiating cell, Model fidelity
1. Introduction
Manfred Eigen is quoted as saying, “A theory has only the alterna-

tive of being right or wrong. A model has a third possibility: it may
be right, but irrelevant” (1). The cancer stem cell (CSC) hypoth-
esis is a contentious yet intriguing theory proposed to account for
the intratumoral heterogeneity seen in many cancers (2). While it
is clear that the acquisition of driver genetic mutations propels the
development of most clinically detected patient tumors (3), these
are in many cases genetically monoclonal (4), and thus, clonal evo-
lution alone cannot account for the functional, morphological and
antigenic heterogeneity observed within the malignant cell com-
partment at the time of surgical excision. The CSC hypothesis
competes with other epigenetic theories (e.g., epithelial–mesen-
chymal transition (5)) to account for this heterogeneity. CSC are
181
182 C. Gedye and L. Ailles
functionally defined by a priori identification as a subpopulation

of malignant cells within a tumor that are highly enriched for
tumorigenic potential upon xenograft implantation into an immu-
nocompromised mouse recipient; they are thus better defined as
tumor-initiating cells (TIC). There is growing experimental support
for the CSC hypothesis but the ultimate relevance of the theory
will be determined if these insights are effectively translated into
improvements in cancer patient outcomes. The CSC hypothesis
does not imply that “only stem cells are the cell-of-origin for can-
cer” and there is no evidence to suggest that clonal evolution and
epigenetic heterogeneity are mutually exclusive processes.
As TIC are functionally defined by their in vivo behavior, from
one perspective it is not possible to culture “true” CSC or TIC
in vitro. Indeed there is substantial evidence that many traditional
cancer cell lines are poor models for human cancer, let alone for CSC
(6). Traditional cancer cell lines, commonly grown in media supple-
mented with animal serum, undergo extensive genetic change such
that they develop genotypes and phenotypes that are distinct from
the human tumors from which they were derived (7). Moreover, we
commonly subject these lines to cross-contamination (8) and infec-
tion with Mycoplasma (9), further diluting their utility. For example
the NCI60 panel of human cancer cell lines contains three lines that
are contaminated replicates of other NCI60 lines (10) and nine lines
are phenotypically unlikely to be from the presumed tissue of origin
(11); for example neither of the “prostate” cancer cell lines PC3 and
DU145 appear to have arisen from prostate cancer.
Despite these limitations cancer cell lines remain appealing
tools for research as they allow experiments that are impractical or
not possible in vivo to be performed on large numbers of cells
which can be more easily manipulated and assayed. Fortunately
there is increasing evidence across many tumor types that relevant
cell line models of human cancer can be generated and propagated
in vitro. Perhaps the most important point we wish to make in this
chapter is that it is critical to establish the relevance and fidelity of
any cancer cell line model by validating in vitro findings in primary
xenograft and patient samples. This has been best exemplified by
several publications within the glioblastoma multiforme (GBM)
research field.
Work from the Fine lab demonstrated that establishment of
GBM cell lines in bovine serum rapidly generates cells with a
homogeneous, differentiated phenotype that form non-physiolog-
ical tumors, and have gene expression signatures indistinguishable
from GBM cell lines that have been in culture for decades (12). In
contrast matched patient samples cultured in a defined serum-free
media formulation originally employed to grow neural stem cells (13)
generated cell lines with much richer morphological heterogeneity,
which expressed markers of a more primitive stem-like phenotype,
and formed tumors that were diffusely invasive as is typically seen
12 Cancer Stem Cells Isolation and Characterization 183
in GBM. Most importantly these cell lines in serum-free defined

media closely preserved the phenotype and genotype of the original
patient’s cancer. This landmark paper was associated with a practice
change within the GBM research field; for example these defined
cell lines have been employed to demonstrate the efficacy of targeting
Notch (14), Shh (15), and TGF-β (16, 17) signalling pathways,
and the TIC niche (18). These models have been recently extended
by work that demonstrates that cells grown in similar media but
with the support of a laminin substrate to encourage adherent
growth further improves the viability and phenotype of these novel
GBM cell lines (19).
These culture conditions have been applied in many tumor
types and detailed instructions for establishing novel cell lines from
a variety of tumors such as glioblastoma multiforme (20), prostate
(21, 22), and colon (23, 24) cancers have been published else-
where. In this chapter we wish to focus on optimizing this process
more generally while discussing some of the challenges we continue
to face in attempting to translate the stringent validation that has
been demonstrated in GBM into other cancer models. We also
highlight opportunities that are available in the isolation and char-
acterisation of “cancer stem cells” in vitro.
2. Materials
2.1. Sample Dissociation 1. Instruments: number 25 (straight-edge, 45° angle) sterile

scalpel blades (Swann-Morton, Sheffield, UK), No. 4 scalpel
blade handles, sterile forceps.
2. Disposable labware: 100 mm diameter × 20 mm deep plastic
dishes, 50 mL plastic tubes, 70 μm cell strainer, 5 mL syringe.
3. Frozen primary specimen: Cryomolds (TissueTek II, Sakura
Finetek), O.C.T. Compound (Sakura Finetek), 2-methyl-butane,
dry ice, insulated bath.
4. Digestion enzymes: collagenase/hyaluronidase 10× (Stem Cell
Technologies), DNase I (Invitrogen), ammonium chloride red
blood cell lysis buffer (Gibco Invitrogen) (see Note 1).
2.2. Cell Culture 5. Typical cell culture laboratory equipment, e.g., laminar flow
biosafety cabinet (BSC), pipet-aid, micropipettes (10 μL,
100 μL, and 1,000 μL), filtered serological pipettes and filtered
micropipette tips, centrifuge, hemocytometer, inverted micro-
scope, 37°C humidified cell culture air/CO2 incubator, pref-
erably nitrogen fed hypoxic incubator (see Note 2), various
sized tissue culture flasks (see Note 3).
6. Defined serum-free culture medium (D-SFCM): DMEM/F12
1:1 media, B27 serum-free supplement (50×), penicillin
(10,000 IU/mL), and streptomycin (10,000 μg/mL) 100×

stock, recombinant human fibroblast growth factor, recombi-
nant human epidermal growth factor (all Invitrogen), HEPES
1 M solution, cell culture tested heparin (both Sigma-Aldrich),
500 mL 22 μm sterile filtration systems (Stericup, Millipore)
(see Note 4).
7. Cell culture flask surface coatings: laminin (L2020), collagen
type I (C3867; both Sigma-Aldrich).
8. Cell line passaging: trypsin 0.25% solution (Invitrogen), 2 mM
EDTA in PBS pH 7.4, dimethyl sulfoxide (sterile filtered, fro-
zen in 1 mL aliquots), cryopreservation vials, 1°C/min freez-
ing container, isopropanol.
2.3. Cell Line Model 9. Mycoplasma testing: MycoAlert Mycoplasma detection kit
alidation (Lonza), luminescence plate reader (e.g., SpectraMax
microplate reader, Molecular Devices) (see Note 5).
10. Cell line identity by short-tandem repeat (STR) profiling: Kit
for DNA extraction and purification (e.g., DNeasy Mini Kit,
QIAgen).
11. Flow cytometry: Hanks’ buffered saline solution, heat inactivated
fetal bovine serum (many suppliers), BD CompBeads (Becton
Dickinson), purified mouse, rat, goat, etc. IgG from pooled
normal serum (Cedarlane or Sigma-Aldrich), DAPI (4¢,6-diami-
dino-2-phenylindole dihydrochloride; Sigma-Aldrich).
2.4. Identification 12. Limiting dilution analysis: D-SFCM, 96-well and 6-well flat
of Clonogenic and bottom tissue culture treated plates.
Tumor-Initiating Cells 13. Tumorigenicity: NOD/SCID or NOD/SCID/IL2γR−/−
(NSG) immunocompromised mice, basement membrane
matrix solution (Matrigel, standard growth factor (see Note 6),
BD Biosciences or BME Cultrex, Trevigen), 96-well round
bottom non-treated microplates, 29G 300 μL insulin syringes
(Becton Dickinson).
3. Methods
3.1. Patient Specimen Studying intratumoral heterogeneity requires reliable access to

Acquisition freshly excised human cancer specimens. As such the success of the
project relies on generous patient donation of tissue in excess of
that required for pathological diagnosis, and close collaboration
with the appropriate surgical, oncology, and pathology team mem-
bers. Our experience has been that all parties are enthusiastic, and
that with appropriate consultation and institutional human ethics
review board approval we can routinely collect adequate tissue. At
the time of asking consent for excess cancer tissues, we would nor-
mally request a blood sample from the patient to collect peripheral
blood mononuclear cells (which can be stored directly or used to
generate a lymphoblastoid B-cell line (25)). This provides a source
of normal genomic DNA from the patient, providing a normal
control for genomic studies, and a gold-standard for identification
of derived cell lines by short-tandem repeat (STR) profiling.
Collection of the cancer sample should occur as soon as possible
after removal from the operating room, and in direct consultation
with the responsible pathologist. Saving directly adjacent samples
for immediate RNA/DNA extraction and immunohistochemistry
for later comparison is crucial, as many cancers can have variable
histology in different parts of the same tumor. The sample is trans-
ported in an aliquot of defined serum-free culture media on ice.
For some cancers we have found that samples are stable at room
temperature, or can be left for processing overnight if refrigerated.
This may allow for collection of samples from consenting patients
at geographically distant sites.
3.2. Sample 1. All procedures should take place within an appropriate BSC.
Dissociation Treat all specimens as potentially infected carriers of blood-
borne pathogens and use Universal Precautions. Save a frag-
ment of the donated tumor for later histological characterisation
by freezing in optimal cutting temperature (O.C.T.) solution.
Place a thin layer of O.C.T. into a pre-labelled cryomold, place
the piece of tissue into the cryomold, and cover completely
with more O.C.T. In a fume hood, place the cryomold into a
bath of 2-methyl-butane cooled by dry ice, being careful not
to allow the liquid to come over the top of the cryomold. Once
the O.C.T. is solid white, store the cryomold in a −80°C
freezer.
2. Place remaining tissue into a 100 mm × 20 mm deep plastic
dish with sterile forceps, and using the No. 25 scalpel blades
cut the tissue into small pieces, in a “crossed-blades,” shearing
fashion (see Note 7).
3. Continue to gently mince tumor into a slurry until fragments
are small enough to pass through the tip of a 5 mL pipette.
4. Add the D-SFCM used to transport the sample (9 mL; which
may contain cells in suspension), collagenase/hyaluronidase
(1 mL of 10×, final concentration 1×) and DNase (100 μL,
final concentration 125 U/mL) and incubate at 37°C in a 5%
CO2 incubator.
5. Every 10–15 min, return digesting tumor fragments to the
BSC and pipette up and down with a 5 mL, then a 1,000 μL
pipette until the tumor is well dissociated (determined by ease
of pipetting, and microscopic evaluation of presence of single
cells; see Note 8). The specimen should not be left in the
solution for longer than necessary to achieve cellular dissociation.

Depending on the tumor, this should take anywhere from
30 min to 2 h.
6. Pass the suspension through a sterile 70 μm filter into a 50 mL
tube, and gently break up any remaining fragments by squeez-
ing against the top of the filter with the rubber end of a sterile
5 mL syringe plunger; rinse filter thoroughly with PBS.
7. Centrifuge and resuspend the pellet in a small volume of cold
ammonium chloride red blood cell lysis buffer. Incubate on ice
for 5 min, then wash with a 10× volume of PBS; centrifuge and
wash again in PBS.
8. Resuspend cells in D-SFCM and perform a cell count with a
hemocytometer, then use cells for culture, cryopreservation,
clonogenic or xenotransplantation assays as appropriate. For
immediate ex vivo clonogenicity and tumorigenicity assays
ensure a single cell suspension by passing through a 40 μm cell
filter before preparation of dilutions.
3.3. Cell Culture Preparation of defined serum-free culture medium (D-SFCM)

Medium
1. To a 500 mL bottle of DMEM/F12, add antibiotics, HEPES,
heparin, and the B27 supplement. Thaw and add FGF and
EGF aliquots (see Table 1).
Adjust pH to 7.4 with 1 M NaOH, then sterile filter and
refrigerate.
2. Aliquots (40 mL in a 50 mL tube) may be frozen and thawed
once for later use. Stock solutions can be prepared and frozen
for later use. EGF, FGF: Reconstitute 50 μg in 500 μL PBS,
sterile filter and freeze as 5 μg/50μL aliquots. Heparin: prepare
a 50 mg/mL stock solution in PBS and sterile filter. Store at
4°C for up to 2 years.
3.4. Cell Culture 1. Plate the primary cell suspension at a density of at least 10,000
Work fl ow viable cells per cm2 in standard tissue culture flasks some of
which may be coated with various substrates (see Note 9). We
typically use smaller T25 flasks or multiwell plates depending
on how many cells are available. Keep a stock of refrigerated
pre-coated flasks or plates on hand.
2. Culture cells in a 37°C humidified incubator with 5% CO2, and
if available, in a hypoxic incubator (O2 tension of 2–5%) (see
Note 10).
3. Inspect daily by inverted microscope to monitor growth and
confluency. Cells may require feeding with a half volume media
change at intervals (e.g., weekly) if slow-growing.
4. Passage flasks when cells are 70–90% confluent. Collect culture
supernatant and centrifuge at 450 × g for 5 min to collect
Table 1
Basic formulation for defined serum-free medium (D-SFCM)
Reagent Stock concentration Final concentration Dilution Volume/amount
DMEM/F12 – – – 500 mL
B-27 50× 1× 1:50 10 mL
Heparin 50 mg/mL 4 μg/mL 1:12,500 40 μL
HEPES 1M 10 mM 1:100 5 mL
FGF 100 μg/mL 10 ng/mL 1:10,000 50 μL
EGF 100 μg/mL 10 ng/mL 1:10,000 50 μL
Pen/strep 10,000 U/mL + 10,000 μg/mL 100 U/mL + 100 μg/mL 1:100 5 mL
Total 520 mL
non-adherent viable cells (see Note 11). Aspirate and save some
conditioned media at this point.
5. Wash any non-adherent cell pellet with PBS, centrifuge again
at 450 × g for 5 min, aspirate and discard the supernatant, and
resuspend in an appropriate volume of 0.05% trypsin in 2 mM
EDTA (0.25% trypsin (Gibco), diluted 1:5 with 2 mM EDTA
in PBS) to dissociate spheres or aggregates and incubate at
37°C for 5–10 min.
6. In parallel, wash flask with PBS, then add an appropriate
volume of 0.05% trypsin in 2 mM EDTA and incubate at 37°C
for 5–10 min until all adherent cells detach.
7. When all cells are detached, inactivate the trypsin in both the
flask and non-adherent cell pellet with an equal volume of the
saved conditioned media. Combine and wash the flask twice
with PBS to collect all detached cells.
8. Centrifuge at 450 × g for 5 min, resuspend in fresh D-SFCM
and perform a cell count.
9. Replate cells at a minimum density of 10,000 cells/cm2. Choose
a flask that allows this to be approximately a 1:2–1:4 split.
10. Continue growing in culture for up to three passages, or when
at least 10 million cells are available.
11. Cryopreserve 2–5 million cells per mL in 1 mL cryovials using
1°C/min freezing container. Freezing media consists of 45%
saved conditioned media, 45% fresh media and 10% DMSO.
12. Ensure that cryopreserved cells can be successfully revived
from frozen stocks before identifying a successful cell line
establishment. Pellet and freeze a separate aliquot of cells at
the time of initial cryopreservation for RNA and DNA extrac-

tion for subsequent cell line verification and characterization.
3.5. Cell Line Model Mycoplasma infection in cancer cell lines remains a common prob-
Validation lem despite the relative simplicity of its detection and eradication.
Spread mostly due to poor laboratory technique, infections gener-
3.5.1. Mycoplasma
ally remain superficially occult but have wide-ranging effects on cell
Detection and Eradication
biology and behavior. Mycoplasma infection is almost always spread
from existing infected cell lines by laboratory workers during cell
culturing (double-dipping pipettes, using the same suction pipette
twice, generating aerosols in an over-filled BSC). Infection from the
patient sample or laboratory worker themselves is very rare.
1. Detection of Mycoplasma infection: Many methods are available
for effective detection and surveillance of infection (26), but we
recommend the Lonza MycoAlert luminescence kit, with a
complementary PCR assay to confirm positive samples (27).
Though perhaps more expensive than PCR detection, the
MycoAlert test is rapid, sensitive and specific in our hands.
False positive results may occur if absolute luminescence readings
are low; use well conditioned media, centrifuge samples to remove
debris and ensure that the luminometer settings (e.g., sensitivity,
number and duration of reads) are optimized to minimize this
possibility. 1 mL samples of centrifuged conditioned media may
be stored at 4°C for up to 1 month, thus facilitating routine
surveillance; we have noted samples that consistently return
positive readings after over 18 months at 4°C.
2. Eradication of Mycoplasma infection: With good laboratory
practice infection of novel cell lines ought not to occur, but if
needed Mycoplasma can be simply and reliably eradicated. Many
methods have been described by leaders in the field (9), but we
favor BM Cyclin (Roche) (28). This regimen is more time-con-
suming but we have encountered quinolone-resistant
Mycoplasma species where ciprofloxacin, enrofloxacin and
Plasmocin all failed to eradicate the infection. We hypothesize
that this strain had become resistant after a past ineffective treat-
ment with Mycoplasma Removal Agent (29). Use of antibiotics
for “maintenance or prophylaxis” is unnecessary and indeed
harmful; rather one should focus on inculcating good labora-
tory technique and effective surveillance to prevent infection.
3.5.2. Cell Line Identification Cell line cross-contamination remains as much of a problem as
Mycoplasma infection, whether cells are adherent or non-adherent.
Though commonly and frequently described this problem has
received increasing attention as major scientific journals seek to
hold researchers more accountable (8). Best practice for cell line
management is the use of a cell bank (such as the Johns Hopkins
CellCenter http://cellcenter.grcf.jhmi.edu/) but in the absence
of a centralized service one can still maintain identity validation of

one’s cell lines by serial short-tandem repeat (STR) profiling (30)
on a routine basis. This testing requires a small amount of DNA
from the original patient (either their tumor or peripheral blood
mononuclear cells) to act as a reference sample. Testing can be
requested at many academic (e.g., Johns Hopkins, SickKids
Toronto) and commercial vendors (e.g., ATCC, Cell Bank
Australia, DSMZ). Cross-contamination with commercial cell lines
can also be disproven by routine comparison against the Online
STR Analysis database available at http://www.dsmz.de/human_
and_animal_cell_lines/main.php?contentleft_id=101.
3.5.3. Validation of DSFM A critical initial criterion of isolating and characterizing “cancer
Cell Line Versus Primary stem cells” in vitro is to validate that the novel cell line is actually a
Patient Tumor reasonable model of the patient cancer from which it was derived.
A number of modalities are appropriate including assays at the pro-
tein, RNA and DNA levels.
1. Flow Cytometry: Flow cytometry represents a powerful tech-
nique to rapidly interrogate the phenotype of single cells in
suspension (whether ex vivo, ex xenograft or from an estab-
lished cell line). For example flow cytometry can be used to
establish the phenotype and cellular identity of a cell popula-
tion (e.g., EpCAM (CD326) or MUC1 (CD227) positive epi-
thelial cells) or to investigate if subpopulations of cells (e.g.,
CD44+ cells in HNSCC) exist within the novel cancer cell line
(31). This information can then be applied prospectively to
de novo cell lines and ex vivo patient samples, e.g., for the
identification of lineage markers that allow discrimination of
tumor versus stroma, or for interrogating putative TIC sub-
populations (see below).
(a) Staining cells for flow cytometry analysis: Prepare cells as a
single cell suspension from patient tumor, xenograft, or cell
line. Centrifuge and resuspend cells in FACS buffer (Hanks’
balanced salt solution (HBSS) with 2% heat-inactivated fetal
bovine serum) at 105–106 cells per 100 μL. To further block
nonspecific binding of antibodies, and depending on the spe-
cies in which your antibodies of interest are generated, add
mouse, rat, goat etc. IgG at a final concentration of 20 μg/mL
and incubate on ice for 5 min. Do not wash. Ensure that ade-
quate control samples are set aside (see Note 12). To 100 μL
aliquots of blocked cells add 100 μL of buffer with a 2× con-
centration of desired antibodies (can be prepared while incu-
bating cells with blocking IgG) to give a final volume of 200 μL
with 1× antibody concentrations. The optimal antibody con-
centration should be determined empirically by performing
titrations in preliminary experiments (see Note 13) (32).
Incubate on ice for 15 min. Wash with 10× volume of FACS
buffer, centrifuge, resuspend in 100 μL FACS buffer contain-

ing any secondary reagents (e.g., fluorophore-labelled second-
ary antibodies or streptavidin) at the appropriate concentration.
Wash again and resuspend in FACS buffer with DAPI (final
concentration 0.1 μg/mL), propidium iodide or 7-aminoac-
tinomycin-D to identify nonviable cells. We favor DAPI due to
lower spectral overlap with other fluorophores in common
practice; however, this requires access to a flow cytometer with
the appropriate laser (Violet or UV). Record fluorescence data
on a suitable flow cytometer, and analyze data using FlowJo
software or Cytobank (https://www.cytobank.org/cytobank/
experiments) according to published guidelines (32).
(b) Immunohistochemistry and immunofluorescence: While flow
cytometry provides a great deal of phenotypic information,
immunohistochemistry (IHC) and immunofluorescence (IF)
provides valuable additional information, such as the intracel-
lular location of markers of interest, the location of marker
positive cells within established patient tumors or xenografts
(e.g., CD44+ cells adjacent to tumor and xenograft fibroblast
stromal cells (33)) or the co-expression of functional markers
(e.g., CD44+/BMI1+ in HNSCC CSCs). Cultured monolay-
ers on suitable matrices (chamber slides), cell pellets embedded
in thrombin/plasma clots as cell blocks (34), cell line derived
xenografts and non-adherent clusters or spheroids (35)
can all be examined by IHC and IF and compared to tissue
sections cut from the piece of primary patient tissue that was
saved in O.C.T. (see above). IF on paraffin-embedded samples
is also now more easily accomplished (36) and samples pre-
pared in this way may better preserve cellular morphology and
antigen location.
(c) Transcriptome, methylome and epigenome analysis: Global
transcriptome analysis by cDNA microarray is now widely
available, and represents a convenient and cost-effective
method for rapidly phenotyping bulk or separated cell popula-
tions. While comparison of TIC marker-positive and marker-
negative subpopulations is an obvious application, we would
suggest first validating the phenotype of novel serum-free cancer
cell line versus the patient’s tumor. Again, many studies have
demonstrated that traditional serum cell lines poorly replicate
the transcriptomic phenotype of the tumors of origin (6, 37).
Various platforms are available, and we have found the Illumina
BeadChip technology gives reliable results at a reasonable cost.
A criticism of studies investigating the cancer stem cell hypoth-
esis has been a lack of detail on the mechanisms controlling the
postulated irreversible epigenetic hierarchy. Some recent stud-
ies are beginning to address epigenetic regulation in cancer
stem cell biology (38), and we would suggest that routine
interrogation of epigenetic mechanisms such as methylation,

microRNA expression, and acetylation within de novo serum-
free cancer cell lines as compared to the original patient’s tumor
would add to a robust understanding of the validity and limita-
tions of this cancer model. This is particularly important as
there is evolving evidence that the cell culture process itself
causes significant changes in methylation (39), in particular in
developmentally regulated genes (40).
(d) Genotype analysis: Cancer cells in culture on average appear to
carry a more extensive complement of mutations than the
tumors from which they are derived (41–44). There is increas-
ing evidence that the acquisition of these mutations may be a
function of the culture process itself (7), while there is compel-
ling evidence in GBM that this evolution away from the geno-
type of the original patient’s tumor can be prevented for some
time by the use of DSFM culture conditions. To examine the
genotypic fidelity in your novel cell cultures, collect genomic
DNA from a patient blood sample (requested at the time of
surgery), from an adjacent fragment of the original tumor spec-
imen, from cells cultured for multiple passages in vitro and
xenografts established from these cell lines. A variety of whole
genome methods are available (array CGH, SNP arrays or
sequencing) but given the quantity and quality of genomic
DNA that can be expected to be collected from patient samples,
we favor SNP arrays such as the Affymetrix® Genome-Wide
Human SNP Array 6.0. This platform gives a balance of density
versus cost and with sufficient resolution for genotypic validity.
3.5.4. Clonogenicity Clonogenic frequency may be a useful surrogate for tumorigenic

frequency, but while useful for rapid screening of subpopulations,
any marker or relevant pathway defined by in vitro techniques must
be promptly investigated in vivo. Clonogenic frequency can be
estimated in adherent or non-adherent conditions, using limiting
dilution analysis.
1. Limiting dilution assay: To assess clonogenic frequency by lim-
iting dilution, multiple replicates in a 96-well format is a con-
venient and reproducible technique. With a final volume of
200 μL per well, and using 16 wells per concentration, a pre-
plating volume of 4 mL gives 20 replicates and enough for
losses. This also gives six cell concentrations per plate, allowing
a useful range of dilutions. Pre-plating dilutions can be per-
formed in 6-well plates; add media first, then add cells, and mix
gently. A manual eight-channel multi-pipette (with only four
tips attached) can then be used four times to aliquot 4 × 200 μL
of cells into 4 × 4 = 16 wells of a 96-well flat bottom plate. Thus
six doses can be tested per plate. We suggest “range-finding”
of cell concentrations initially (e.g., 10× dilutions ranging from
1 to 10,000 cells per well) and once an approximate range

has been identified, more closely spaced (e.g., if 10× dilu-
tions indicated a clonogenic frequency of approximately 1 in
1,000, then perform 2× dilutions ranging from 200 to
6,400 cells per well). To assess clonogenicity under non-adher-
ent conditions, polyHEMA-coated (45) or commercially avail-
able ultra-low attachment plates can be used. If adherent
colony formation is of interest, tissue culture treated plates, or
plates coated with laminin, collagen or other substrates can be
used. After an appropriate interval in culture (2–4 weeks),
count wells that contain colonies and plot log10 (percentage of
wells without colonies) versus number of input cells/well.
Limiting dilution theory states that the mean number of stem/
progenitor cells per well can be calculated from the formula:
u = −lnF0, where u represents the mean number of clonogenic
cells per well and F0 is the fraction of negative wells (46). On
the line of best fit by linear regression analysis, the value at
which this line intercepts 37% estimates the minimum clono-
genic frequency (47–49). Analysis can also be performed using
L-Calc Software (StemCell Technologies Inc.), or online using
the extreme limiting dilution analysis (ELDA (50)) website;
http://bioinf.wehi.edu.au/software/elda/index.html.
3.5.5. Tumorigenicity A defining feature of “cancer stem cells” is their ability to form
tumors as xenografts in immunocompromised mice. This property
must therefore be rigorously tested to prove that the chosen cul-
ture conditions maintain TIC in vitro.
1. Absolute Tumorigenicity: Initial experiments should be per-
formed at high doses (e.g., 106) to establish if the cultured cell
lines are in fact tumorigenic. Resulting xenograft tumors
should then be compared histologically to the original patient
tumor (see above). Passaged, washed and 40 μm filtered single
cell suspensions should be counted and resuspended in the
desired volume of PBS (ranging from 10 to 50 μL); the vol-
ume will depend upon the site of injection, with sites such as
the renal capsule or the brain requiring small volumes, while
sites such as subcutaneous or mammary fat pad are able to
accommodate higher, more manageable volumes. Allowing
excess cells for losses, mix equivalent volumes of cells and
Matrigel in a round-bottomed 96 well plate that is resting on
ice. Aspirate 20–100 μL per injection into cooled insulin
syringes for injection and keep filled syringes on ice until injec-
tion into mice. The appropriate strain of mouse (e.g., NOD/
SCID, NSG, or Rag2γDKO) should be selected and xenograft
techniques should be optimized carefully, as should the most
relevant injection site. For many tumor types an orthotopic site
is obvious (e.g., GBM into brain, hepatocellular carcinoma
into liver), but the best determinant of the appropriate niche

for xenotransplantation is reproducibility of patient tumor
phenotype. In some cases additional manipulations may be
necessary, such as preconditioning the mice or implanting
estrogen pellets subcutaneously (51). Once xenografts are
established, at a minimum, it should be verified that they his-
tologically resemble the primary patient sample by hematoxy-
lin and eosin staining, ideally with the help of a qualified
pathologist who is expert in the particular type of cancer being
studied. If other markers are routinely in clinical use (e.g., ER
and PR staining in breast cancer), then similar markers should
be investigated on the xenograft.
2. TIC frequency by limiting dilution analysis: The frequency of
TIC in the newly established cell line should be determined by
limiting dilution analysis (LDA) in immunocompromised
mice. Cells should be serially diluted in cold PBS and mixed
1:1 with Matrigel as above. Initial experiments should be based
on a broad range of cell doses (e.g., 102–106) to find the most
appropriate range of tumorigenicity as this is known to vary
from tumor to tumor and from cell line to cell line in the same
histological subtype (33). A minimum of five mice should be
injected with each cell dose. Once an approximate median
tumorigenic frequency is established then tighter dilutions can
be used to define the minimum tumorigenic dose. A compari-
son of in vivo TIC frequency and clonogenicity can then be
performed to determine whether clonogenicity correlates with
tumorigenicity.
3. Does clonogenicity reflect tumorigenicity? Wells from LDA
experiments in vitro that were plated at the minimum dose and
contain a single colony should be further assessed for their
“stem cell” characteristics. First, they should be replated to
determine whether secondary colonies can be formed, demon-
strating self-renewal; and second, they should be expanded and
injected into mice to demonstrate that a single colony-forming
cell can ultimately generate a tumor in vivo. Again, histological
verification, in consultation with a pathologist, should be done
by comparing the xenograft tumor histology to that of the pri-
mary patient sample.
3.5.6. Identification of TIC 1. Selection of candidate TIC markers: Once your novel defined
in Defined Serum-Free serum-free cell line is validated as a useful model of the patient’s
Cancer Cell Lines cancer in vitro, it can then be investigated for the presence of an
intercellular heterogeneity and hierarchy. Candidate TIC marker
selection is informed by a number of criteria. For example, pub-
lished CSC markers can be informative (51), as can markers
that have functional significance (52) or prognostic relevance
(53) within the tumor type, or markers that are expressed in
presumptive CSC niches within tumors (33). In the first

instance novel cell lines should be profiled by IHC and/or flow
cytometry to validate that the marker is present, and that it is
also expressed on a subset of primary patient cells before
embarking upon sorting experiments. Cell surface markers are
often employed, and some functional markers (e.g.,
ALDEFLUOR®) are showing promising results in some tumor
systems (53, 54). Our experience and literature (55, 56) reports
of the “side-population” method suggests that results obtained
with this modality must be rigorously validated, i.e., just because
a cell population contains a side population does not mean that
these are “cancer stem cells.” Detailed methods for flow cytom-
etry analysis have been discussed above and elsewhere (32).
2. TIC assays of purified cell subsets: Once candidate TIC mark-
ers have been selected, their functional significance can be
assayed by purifying subsets from the novel cell line. For cell
surface markers, stain samples as described above for flow
cytometry. Cell sorting of positive and negative populations
can be done on any available fluorescence activated cell sorting
machine at your institute, such as a FACSAria (BD), MoFlo
(Coulter), or Influx (Cytopia). The number of cells recovered
from the sort should be verified using a hemocytometer, and
clonogenic or TIC assays set up as described above. Flow
cytometry cell sorting is a reliable and reproducible method,
but where this is unavailable magnetic bead cell sorting (e.g.,
MACS from Miltenyi, EasySep or StemSep from STEMCELL
Technologies) can generate cell populations with high purity
(57). Separated cell populations can be assayed initially using
in vitro clonogenicity assays, and if a functional difference in
colony formation is noted, this will provide useful preliminary
evidence for exploring this marker in vivo, firstly using the
novel cell line itself, and then validating back on fresh ex vivo
patient cancer samples. Once a TIC marker has been validated
on several novel serum-free cell lines and in several ex vivo
patient samples, separated cell populations can be explored by
bioinformatic (58, 59) and functional assays to investigate their
mechanisms of self-renewal (15, 16), differentiation (17, 60),
microenvironmental niche interaction (18, 61), and treatment
resistance (62, 63).
3.5.7. Data Interpretation While the validity of the cancer stem cell hypothesis can only be
conclusively determined in a particular tumor type by in vivo stud-
ies, there are many in vitro studies that are attractive as they may
corroborate in vivo data or generate novel mechanistic hypotheses
and therapeutic candidates. For example GBM cell lines generated
de novo in serum-free media and grown on laminin coated plates
have been employed to identify that targeting serotonin signalling
may be relevant in this disease (19). Novel defined serum-free cancer
cell lines can also be applied in genetic screens; for example in a
recent publication that identified TRRAP as a regulator of TIC in

GBM (64). Finally if a novel serum-free cell line is established as a
valid model of a patient’s cancer in vitro, these cultures can be
labelled with fluorescent proteins that allow clonal tracking in vitro
and in vivo, which in turn could provide further evidence for (65),
or against (66) the cancer stem cell hypothesis. George E.P. Box
cautioned us to “remember that all models are wrong; the practical
question is how wrong do they have to be to not be useful” (67).
The key point we wish to emphasize when isolating and character-
izing “cancer stem cells” in vitro is to ruthlessly validate the novel
cell line and one’s findings against real human cancers, both in vivo
and ex vivo. If we can honestly reflect on “how wrong” our novel
defined serum-free cancer cell lines are as models of cancer, then
we can more confidently employ them to interrogate tumor het-
erogeneity and therapeutic hypotheses, which will more rapidly
lead us to more useful outcomes for patients with cancer.
4. Notes
1. The combination of enzymes for tissue dissociation should be

optimized on a tissue-specific basis. For example some tumors
are so friable (e.g., melanoma) that enzyme digestion is hardly
necessary, whereas others are tough and fibrous and require
significant digestion (e.g., head and neck squamous carci-
noma). A useful online tissue dissociation guide has been pro-
duced by one manufacturer of digestion enzymes (http://
www.worthington-biochem.com/tissuedissociation/default.
html). There will also be variability between different patients’
tumors within the same tumor type, necessitating careful
observation of the tumor digestion in every case.
2. Many commercial hypoxic incubators are available, some of
which use premixed gas at the appropriate concentrations (e.g.,
93%N2/5%CO2/2%O2). A large airtight silicone-sealed plastic
food container modified with ports to admit premixed gas and
incubated in a standard incubator may be a useful alternative to
pilot these conditions.
3. There is evolving evidence that laboratory plasticware leaches a
combination of soluble contaminants that can have significant
biological effects (summarized by Nature News, 26 April 2010,
doi:10.1038/news.2010.200). We are not aware of any specific
manufacturer that is free of these problems, but we would
greatly appreciate if such materials became available.
4. A variety of media formulations have been applied in different
tumor types; a summary of several of these formulations is pre-
sented in Table 2. The evolution of early formulations employed
in the neural stem cell field is also presented.
5. If one does not have access to a luminescence plate reader,

PCR-based detection methods for Mycoplasma as described by
Uphoff and Drexler (27) (and Chapters 1 and 2 in this edi-
tion) are highly sensitive and reliable.
6. We have noted considerable batch-to-batch variability with
Matrigel, which has made it challenging to use. When thawed
on ice at 4°C overnight, the Matrigel should be liquid and
barely viscous. With mixing and warming it will thicken and
irreversibly gel rapidly.
7. At this point you may also wish to set aside small (1–3 mm
square) fragments of tumor for direct implantation into mice
for establishment of primary tumor xenografts.
8. Whilst enzymatically digesting the mechanically dissociated
tumor slurry, gentle agitation may be helpful. For example one
can use a sterile stir-bar and magnetic stir-plate, a rocking
mixer, or a tube rotator that can be used in a 37°C incubator.
9. In our experience many de novo cell lines established in defined
serum-free culture media are adherent and grow well in stan-
dard tissue culture treated flasks. Some lines have grown poorly
under such conditions, and we are exploring the use of addi-
tional substrates that may perhaps offer a better microenviron-
mental stimulus (19). Collagen and laminin have shown
promising early results. These are prepared as solutions and
coated onto plates according to the manufacturer’s instruc-
tions (Sigma-Aldrich). Fibronectin, Matrigel, poly-l-lysine, or
gelatin are other possible substrates to consider.
10. Culture under hypoxic conditions has been very successful in
several tumor types in our hands. There is a well established
literature on the benefits of culturing stem cell populations
under hypoxic conditions (88, 89), and we have noted that
proliferation is generally equivalent if not faster, and some cell
lines that failed to grow when cultured in ambient air have
grown robustly and reproducibly at 2% O2.
11. A repeated theme in the literature is that placing an immortal-
ized cancer cell line into serum-free media and non-adherent
conditions leads to formation of spheres so they must be cancer
stem cells. “(Insert name of your cancer here)-spheres” are not
cancer stem cells. The use of ultra-low attachment plates,
reduced serum conditions, or roller bottles can generate these
non-adherent clusters in the presence of any culture media
including serum (90). While there is clearly a difference in gene
expression and phenotype upon this microenvironmental alter-
ation, this does not mean that these “tumor-spheres” replicate
the phenotype or hierarchy that existed in the original patient’s
cancer. Indeed at concentrations of cells above 10 cells per μL
clustering is a common occurrence and observed “spheres” are
unlikely to have come from a clonal event (91).
Table 2
Selected serum-free media formulations reported in the establishment of de novo cancer cell lines
Tumor References Year Basal media Supplement EGFa bFGF Other
Transitional carcinoma of Bentivegna (68) 2010 DMEM – 20 20 –

bladder
Ewing’s sarcoma Suva (69) 2009 DMEM/F12 Knockout SRb 10 10 –
Acute lymphoblastic leukemia Nijmeijer (70) 2009 IMDM Insulin, albumin, – – Cholesterol, β-ME
transferrin
Endometrial carcinoma Rutella (71) 2009 DMEM/F12 BSA, ITSc 20 10 Progesterone, putrescine
12
Prostate carcinoma Guzmán-Ramírez (72) 2009 CnT-52 BPE – – –

Head and neck squamous Chen (73) 2009 DMEM/F12 N2 10 10 –
carcinoma
Pituitary adenoma Xu (74) 2009 DMEM/F12 B27 20 20 –
Breast carcinoma Grimshaw (75) 2008 DMEM/F12 B27 20 – Insulin, hydrocortisone
Prostate carcinoma Vander Griend (76) 2008 PrEGM – – – –
Colorectal Dylla (77) 2008 DMEM/F12 B27, ITS 20 20 LIF, heparin, hydrocortisone
Colorectal Ricci-Vitiani, Todaro, 2008 DMEM BSA, ITS 20 10 Progesterone, heparin
Cammareri et al. (24, 78–81)
Non-small-cell lung carcinoma Eramo (82) 2007 DMEM/F12 BSA, ITS 20 10 Progesterone, putrescine
Prostate carcinoma Collins (83) 2005 K-SFM BPE, EGF – – LIF, SCF, cholera toxin
Glioblastoma multiforme Beier (58) 2007 DMEM/F12 B27 20 20 LIF
Glioblastoma multiforme Pellegatta (84) 2006 DMEM/F12 B27 20 20 –
Cancer Stem Cells Isolation and Characterization
Glioblastoma multiforme Lee (12) 2006 Neurobasald B27:N2 50 50 –
(continued)
197
Table 2
198
(continued)
Tumor References Year Basal media Supplement EGFa bFGF Other
Breast carcinoma Dontu, Charafe-Jauffret, 2003 MEGM B27 20 20 heparin

Ginestier et al. (52, 53, 85)
Glioblastoma multiforme Singh (47), Lenkiewicz (20) 2003 DMEM/F12 NSF 20 20 LIF
Rat embryonal hippocampal Brewer (13) 1993 Neurobasal B27 – – –
neurons
C. Gedye and L. Ailles
Murine neural stem cells Reynolds (86) 1992 DMEM/F12 “N2” 20 – –

e
Murine striatal neurons Weiss (87) 1986 DMEM/F12 “N2” – – –
a
bFGF and EGF in ng/mL
b
Summarized in the patent http://www.freepatentsonline.com/WO1998030679A1.pdf
c
ITS: insulin, transferrin, selenium; from various suppliers
d
Summarized in the patent http://www.patentstorm.us/patents/6180404/description.html
e
Original report of the basic formulation that became the N2 supplement
12. Controls for flow cytometry should include compensation

controls, unstained controls and controls for background stain-
ing despite compensation, so as to allow for appropriate gat-
ing. While isotype controls can be used, we prefer
“fluorescence-minus-one” (FMO) controls, where control
samples containing antibodies labelled with fluorophores for
all channels bar the channel of interest are included. For exam-
ple in an experiment using FITC, PE, APC, and PECy7 labelled
antibodies, the FITC FMO control contains PE, APC, PECy7
antibodies, etc. For compensation controls we use BD
CompBeads (Becton Dickinson) in the appropriate species, as
this spares the use of potentially limited number of cells from
patient tissues or xenografts.
13. Titrating antibodies for flow cytometry is essential prepara-
tion, particularly when attempting to detect small subpop-
ulations, which may often be penumbra rather a distinct
binary population. A number of laboratories have kindly
shared their protocols for this process online including the
Altman lab at Emory (http://www.microbiology.emory.
edu/altman/f_protocols/f_flowCytometry/p_titering_Abs.
htm), the Herzenberg lab at Stanford (http://herzenberg.
stanford.edu/Protocols/default.htm), and the University of
Chicago Flow Cytometry Core Facility (http://ucflow.blog-
spot.com/2009/06/antibody-titrations.html).
14. Cell culture conditions will need to be optimized for each and
every cancer studied. The microenvironment of in vitro con-
ditions is obviously markedly different from where the tumor
was removed. In some cases feeder cells such as mouse
embryonic fibroblasts, cancer associated fibroblasts or
endothelial cells may be necessary to adequately establish
cell lines in the absence of serum. Alternatively, cell lines may
be difficult to establish directly from human specimens but
may be more easily generated from passaged xenografts. The
same caveats apply however, and the validity of the cell line
as a model must be based upon comparison to the original
patient’s tumor.
Acknowledgments
This research was supported by the Ontario Institute for Cancer

Research and the Ontario Ministry of Health and Long Term
Care. The views expressed do not necessarily reflect those of the
OMOHLTC. C.G. is supported by a Royal Australasian College
of Physicians CSL Fellowship and a National Health and Medical
Research Council Postdoctoral Training Fellowship. L.E.A. is
supported by a New Investigator Award from the Ontario Institute

for Cancer Research.
References
1. Eigen M (1973) The origin of biological 12. Lee J, Kotliarova S, Kotliarov Y, Li A, Su Q,

information. In: Mehra J (ed) The physicist’s Donin NM, Pastorino S, Purow BW,
conception of nature. Reidel, Dordecht, Holland Christopher N, Zhang W, Park JK, Fine HA
2. Ailles LE, Weissman IL (2007) Cancer stem (2006) Tumor stem cells derived from glio-
cells in solid tumors. Curr Opin Biotechnol blastomas cultured in bFGF and EGF more
18:460–466 closely mirror the phenotype and genotype of
3. Nowell PC (1976) The clonal evolution of primary tumors than do serum-cultured cell
tumor cell populations. Science 19:23–28 lines. Cancer Cell 9:391–403
4. Liu W, Laitinen S, Khan S, Vihinen M, 13. Brewer GJ, Torricelli JR, Evege EK, Price PJ
Kowalski J, Yu G, Chen L, Ewing CM, (1993) Optimized survival of hippocampal
Eisenberger MA, Carducci MA, Nelson WG, neurons in B27-supplemented Neurobasal, a
Yegnasubramanian S, Luo J, Wang Y, Xu J, new serum-free medium combination. J
Isaacs WB, Visakorpi T, Bova GS (2009) Copy Neurosci Res 35:567–576
number analysis indicates monoclonal origin 14. Fan X, Khaki L, Zhu TS, Soules ME, Talsma
of lethal metastatic prostate cancer. Nat Med CE, Gul N, Koh C, Zhang J, Li YM, Maciaczyk
15:559–565 J, Nikkhah G, Dimeco F, Piccirillo S, Vescovi
5. Turner C, Kohandel M (2010) Investigating AL, Eberhart CG (2010) NOTCH pathway
the link between epithelial-mesenchymal tran- blockade depletes CD133-positive glioblas-
sition and the cancer stem cell phenotype: a toma cells and inhibits growth of tumor neu-
mathematical approach. J Theor Biol rospheres and xenografts. Stem Cells 28:5–16
265:329–335 15. Clement V, Sanchez P, de Tribolet N,
6. van Staveren WC, Solis DY, Hebrant A, Radovanovic I, Ruiz i Altaba A (2007)
Detours V, Dumont JE, Maenhaut C (2009) HEDGEHOG-GLI1 signaling regulates human
Human cancer cell lines: experimental models glioma growth, cancer stem cell self-renewal,
for cancer cells in situ? For cancer stem cells? and tumorigenicity. Curr Biol 17:165–172
Biochim Biophys Acta 1795:92–103 16. Peñuelas S, Anido J, Prieto-Sánchez RM, Folch
7. Daniel VC, Marchionni L, Hierman JS, Rhodes G, Barba I, Cuartas I, García-Dorado D, Poca
JT, Devereux WL, Rudin CM, Yung R, MA, Sahuquillo J, Baselga J, Seoane J (2009)
Parmigiani G, Dorsch M, Peacock CD, Watkins TGF-beta increases glioma-initiating cell self-
DN (2009) A primary xenograft model of renewal through the induction of LIF in human
small-cell lung cancer reveals irreversible glioblastoma. Cancer Cell 15:315–327
changes in gene expression imposed by culture 17. Piccirillo SG, Reynolds BA, Zanetti N, Lamorte
in vitro. Cancer Res 69:3364–3373 G, Binda E, Broggi G, Brem H, Olivi A,
8. American Type Culture Collection Standards Dimeco F, Vescovi AL (2006) Bone morpho-
Development Organization Workgroup (2010) genetic proteins inhibit the tumorigenic poten-
Cell line misidentification: the beginning of tial of human brain tumour-initiating cells.
the end. Nat Rev Cancer 10:441–448 Nature 444:761–765
9. Drexler H, Uphoff C (2000) Contamination of 18. Calabrese C, Poppleton H, Kocak M, Hogg
cell culture, mycoplasma. In: Spier E (ed) TL, Fuller C, Hamner B, Oh EY, Gaber MW,
Encyclopedia of cell technology. Wiley, New York Finklestein D, Allen M, Frank A, Bayazitov IT,
Zakharenko SS, Gajjar A, Davidoff A,
10. Lorenzi PL, Reinhold WC, Varma S, Gilbertson RJ (2007) A perivascular niche for
Hutchinson AA, Pommier Y, Chanock SJ, brain tumor stem cells. Cancer Cell 11:69–82
Weinstein JN (2009) DNA fingerprinting of
the NCI-60 cell line panel. Mol Cancer Ther 19. Pollard SM, Yoshikawa K, Clarke ID, Danovi
8:713–724 D, Stricker S, Russell R, Bayani J, Head R, Lee
M, Bernstein M, Squire JA, Smith A, Dirks P
11. Wang H, Huang S, Shou J, Su EW, Onyia JE, (2009) Glioma stem cell lines expanded in
Liao B, Li S (2006) Comparative analysis and adherent culture have tumor-specific pheno-
integrative classification of NCI60 cell lines types and are suitable for chemical and genetic
and primary tumors using gene expression screens. Cell Stem Cell 4:568–580
profiling data. BMC Genomics 7:166
20. Lenkiewicz M, Li N, Singh SK (2009) Culture MF, Ailles LE (2007) Identification of a sub-
and isolation of brain tumor initiating cells. population of cells with cancer stem cell proper-
Curr Protoc Stem Cell Biol Chapter 3:Unit3.3 ties in head and neck squamous cell carcinoma.
21. Yao M, Taylor RA, Richards MG, Sved P, Proc Natl Acad Sci USA 104:973–978
Wong J, Eisinger D, Xie C, Salomon R, 34. Nigro K, Tynski Z, Wasman J, Abdul-Karim F,
Risbridger GP, Dong Q (2010) Prostate- Wang N (2007) Comparison of cell block
regenerating capacity of cultured human adult preparation methods for nongynecologic
prostate epithelial cells. Cells Tissues Organs ThinPrep specimens. Diagn Cytopathol
191:203–212 35:640–643
22. Wang S (2009) Anchorage-independent 35. Weiswald L-B, Guinebretiere J-M, Richon S,
growth of prostate cancer stem cells. Methods Bellet D, Saubamea B, Dangles-Marie V
Mol Biol 568:151–160 (2010) In situ protein expression in tumour
23. Kreso A, O’Brien CA (2008) Colon cancer spheres: development of an immunostaining
stem cells. Curr Protoc Stem Cell Biol Chapter protocol for confocal microscopy. BMC Cancer
3:Unit 3.1 10:106
24. Cammareri P, Lombardo Y, Francipane MG, 36. Robertson D, Savage K, Reis-Filho JS, Isacke
Bonventre S, Todaro M, Stassi G (2008) CM (2008) Multiple immunofluorescence
Isolation and culture of colon cancer stem labelling of formalin-fixed paraffin-embedded
cells. Methods Cell Biol 86:311–324 (FFPE) tissue. BMC Cell Biol 9:13
25. Oh H-M, Oh J-M, Choi S-C, Kim S-W, Han 37. Li A, Walling J, Kotliarov Y, Center A, Steed
W-C, Kim T-H, Park D-S, Jun C-D (2003) An ME, Ahn SJ, Rosenblum M, Mikkelsen T,
efficient method for the rapid establishment of Zenklusen JC, Fine HA (2008) Genomic
Epstein-Barr virus immortalization of human changes and gene expression profiles reveal
B lymphocytes. Cell Prolif 36:191–197 that established glioma cell lines are poorly
26. Garner CM, Hubbold LM, Chakraborti PR representative of primary human gliomas. Mol
(2000) Mycoplasma detection in cell cultures: Cancer Res 6:21–30
a comparison of four methods. Br J Biomed 38. Tabu K, Sasai K, Kimura T, Wang L, Aoyanagi
Sci 57:295–301 E, Kohsaka S, Tanino M, Nishihara H, Tanaka
27. Uphoff CC, Drexler HG (2005) Detection of S (2008) Promoter hypomethylation regulates
mycoplasma contaminations. Methods Mol CD133 expression in human gliomas. Cell Res
Biol 290:13–23 18:1037–1046
28. Uphoff CC, Drexler HG (2005) Eradication 39. Saferali A, Grundberg E, Berlivet S, Beauchemin
of mycoplasma contaminations. Methods Mol H, Morcos L, Polychronakos C, Pastinen T,
Biol 290:25–34 Graham J, McNeney B, Naumova AK (2010)
Cell culture-induced aberrant methylation of
29. Gignac SM, Uphoff CC, MacLeod RA, Steube the imprinted IG DMR in human lymphoblas-
K, Voges M, Drexler HG (1992) Treatment of toid cell lines. Epigenetics 5:50–60
mycoplasma-contaminated continuous cell
lines with mycoplasma removal agent (MRA). 40. Meissner A, Mikkelsen TS, Gu H, Wernig M,
Leuk Res 16:815–822 Hanna J, Sivachenko A, Zhang X, Bernstein
BE, Nusbaum C, Jaffe DB, Gnirke A, Jaenisch
30. Masters JR, Thomson JA, Daly-Burns B, Reid R, Lander ES (2008) Genome-scale DNA
YA, Dirks WG, Packer P, Toji LH, Ohno T, methylation maps of pluripotent and differen-
Tanabe H, Arlett CF, Kelland LR, Harrison tiated cells. Nature 454:766–770
M, Virmani A, Ward TH, Ayres KL, Debenham
PG (2001) Short tandem repeat profiling pro- 41. Kuo K-T, Guan B, Feng Y, Mao T-L, Chen X,
vides an international reference standard for Jinawath N, Wang Y, Kurman RJ, Shih I-M,
human cell lines. Proc Natl Acad Sci USA Wang T-L (2009) Analysis of DNA copy num-
98:8012–8017 ber alterations in ovarian serous tumors
identifies new molecular genetic changes in
31. Tarnok A, Ulrich H, Bocsi J (2010) Phenotypes low-grade and high-grade carcinomas. Cancer
of stem cells from diverse origin. Cytometry A Res 69:4036–4042
77:6–10
42. Pollack JR, Sorlie T, Perou CM, Rees CA,
32. Alexander CM, Puchalski J, Klos KS, Badders Jeffrey SS, Lonning PE, Tibshirani R,
N, Ailles L, Kim CF, Dirks P, Smalley MJ Botstein D, Borresen-Dale AL, Brown PO
(2009) Separating stem cells by flow cytome- (2002) Microarray analysis reveals a major
try: reducing variability for solid tissues. Cell direct role of DNA copy number alteration
Stem Cell 5:579–583 in the transcriptional program of human
33. Prince ME, Sivanandan R, Kaczorowski A, Wolf breast tumors. Proc Natl Acad Sci USA
GT, Kaplan MJ, Dalerba P, Weissman IL, Clarke 99:12963–12968
43. Beroukhim R, Brunet J-P, Di Napoli A, Mertz mediate metastasis and poor clinical outcome
KD, Seeley A, Pires MM, Linhart D, Worrell in inflammatory breast cancer. Clin Cancer Res
RA, Moch H, Rubin MA, Sellers WR, 16:45–55
Meyerson M, Linehan WM, Kaelin WG Jr, 54. Jiang F, Qiu Q, Khanna A, Todd NW, Deepak
Signoretti S (2009) Patterns of gene expres- J, Xing L, Wang H, Li Z, Su Y, Stass SA, Katz
sion and copy-number alterations in von-hip- RL (2009) Aldehyde dehydrogenase 1 is a
pel lindau disease-associated and sporadic clear tumor stem cell-associated marker in lung can-
cell carcinoma of the kidney. Cancer Res cer. Mol Cancer Res 7:330–338
69:4674–4681 55. Sales-Pardo I, Avendano A, Barquinero J,
44. Virtanen C, Ishikawa Y, Honjoh D, Kimura Domingo JC, Marin P, Petriz J, Camargo FD,
M, Shimane M, Miyoshi T, Nomura H, Jones Goodell MA (2006) The Hoechst low-
MH (2002) Integrated classification of lung fluorescent profile of the side population: clo-
tumors and cell lines by expression profiling. nogenicity versus dye retention. Blood
Proc Natl Acad Sci USA 99:12357–12362 108:1774–1775
45. Srebrow A, Friedmann Y, Ravanpay A, Daniel 56. Burkert J, Otto WR, Wright NA (2008) Side
CW, Bissell MJ (1998) Expression of Hoxa-1 populations of gastrointestinal cancers are not
and Hoxb-7 is regulated by extracellular enriched in stem cells. J Pathol 214:564–573
matrix-dependent signals in mammary epithe- 57. Gedye C, Quirk J, Browning J, Svobodova S,
lial cells. J Cell Biochem 69:377–391 John T, Sluka P, Dunbar PR, Corbeil D, Cebon
46. Lefkovits I, Waldmann H (1984) Limiting J, Davis ID (2009) Cancer/testis antigens can
dilution analysis of the cells of immune system be immunological targets in clonogenic
I. The clonal basis of the immune response. CD133+ melanoma cells. Cancer Immunol
Immunol Today 5:265–268 Immunother 58:1635–1646
47. Singh SK, Clarke ID, Terasaki M, Bonn VE, 58. Beier D, Hau P, Proescholdt M, Lohmeier A,
Hawkins C, Squire J, Dirks PB (2003) Wischhusen J, Oefner PJ, Aigner L, Brawanski
Identification of a cancer stem cell in human A, Bogdahn U, Beier CP (2007) CD133(+)
brain tumors. Cancer Res 63:5821–5828 and CD133(−) glioblastoma-derived cancer
48. Bellows CG, Aubin JE (1989) Determination stem cells show differential growth characteris-
of numbers of osteoprogenitors present in iso- tics and molecular profiles. Cancer Res
lated fetal rat calvaria cells in vitro. Dev Biol 67:4010–4015
133:8–13 59. Lottaz C, Beier D, Meyer K, Kumar P,
49. Tropepe V, Sibilia M, Ciruna BG, Rossant J, Hermann A, Schwarz J, Junker M, Oefner PJ,
Wagner EF, van der Kooy D (1999) Distinct Bogdahn U, Wischhusen J, Spang R, Storch A,
neural stem cells proliferate in response to EGF Beier CP (2010) Transcriptional profiles of
and FGF in the developing mouse telencepha- CD133+ and CD133− glioblastoma-derived
lon. Dev Biol 208:166–188 cancer stem cell lines suggest different cells of
50. Hu Y, Smyth GK (2009) ELDA: extreme lim- origin. Cancer Res 70:2030–2040
iting dilution analysis for comparing depleted 60. Silber J, Lim DA, Petritsch C, Persson AI,
and enriched populations in stem cell and Maunakea AK, Yu M, Vandenberg SR,
other assays. J Immunol Methods 347:70–78 Ginzinger DG, James CD, Costello JF,
51. Al-Hajj M, Wicha MS, Benito-Hernandez A, Bergers G, Weiss WA, Alvarez-Buylla A,
Morrison SJ, Clarke MF (2003) Prospective Hodgson JG (2008) miR-124 and miR-137
identification of tumorigenic breast cancer cells. inhibit proliferation of GBM cells and induce
Proc Natl Acad Sci USA 100:3983–3988 differentiation of brain tumor stem cells.
52. Ginestier C, Hur MH, Charafe-Jauffret E, BMC Med 6:14
Monville F, Dutcher J, Brown M, Jacquemier 61. Bao S, Wu Q, Sathornsumetee S, Hao Y, Li Z,
J, Viens P, Kleer CG, Liu S, Schott A, Hayes Hjelmeland AB, Shi Q, McLendon RE, Bigner
D, Birnbaum D, Wicha MS, Dontu G (2007) DD, Rich JN (2006) Stem cell-like glioma
ALDH1 is a marker of normal and malignant cells promote tumor angiogenesis through
human mammary stem cells and a predictor of vascular endothelial growth factor. Cancer Res
poor clinical outcome. Cell Stem Cell 66:7843–7848
1:555–567 62. Bao S, Wu Q, McLendon RE, Hao Y, Shi Q,
53. Charafe-Jauffret E, Ginestier C, Iovino F, Hjelmeland AB, Dewhirst MW, Bigner DD,
Tarpin C, Diebel M, Esterni B, Houvenaeghel Rich JN (2006) Glioma stem cells promote
G, Extra JM, Bertucci F, Jacquemier J, Xerri L, radioresistance by preferential activation of the
Dontu G, Stassi G, Xiao Y, Barsky SH, DNA damage response. Nature 444:756–760
Birnbaum D, Viens P, Wicha MS (2009) Aldehyde 63. Campos B, Wan F, Farhadi M, Ernst A,
dehydrogenase 1-positive cancer stem cells Zeppernick F, Tagscherer KE, Ahmadi R, Lohr
J, Dictus C, Gdynia G, Combs SE, Goidts V, 72. Guzmán-Ramírez N, Völler M, Wetterwald A,

Helmke BM, Eckstein V, Roth W, Beckhove P, Germann M, Cross NA, Rentsch CA, Schalken
Lichter P, Unterberg A, Radlwimmer B, J, Thalmann GN, Cecchini MG (2009) In vitro
Herold-Mende C (2010) Differentiation ther- propagation and characterization of neoplastic
apy exerts antitumor effects on stem-like stem/progenitor-like cells from human pros-
glioma cells. Clin Cancer Res 16:2715–2728 tate cancer tissue. Prostate 69:1683–1693
64. Wurdak H, Zhu S, Romero A, Lorger M, 73. Chen YC, Chen YW, Hsu HS, Tseng LM,
Watson J, C-y C, Zhang J, Natu V, Lairson Huang PI, Lu KH, Chen DT, Tai LK, Yung
LL, Walker JR, Trussell CM, Harsh GR, Vogel MC, Chang SC, Ku HH, Chiou SH, Lo WL
H, Felding-Habermann B, Orth AP, Miraglia (2009) Aldehyde dehydrogenase 1 is a puta-
LJ, Rines DR, Skirboll SL, Schultz PG (2010) tive marker for cancer stem cells in head and
An RNAi screen identifies TRRAP as a regula- neck squamous cancer. Biochem Biophys Res
tor of brain tumor-initiating cell differentia- Commun 385:307–313
tion. Cell Stem Cell 6:37–47 74. Xu Q, Yuan X, Tunici P, Liu G, Fan X, Xu M,
65. Schatton T, Murphy GF, Frank NY, Yamaura Hu J, Hwang JY, Farkas DL, Black KL, Yu JS
K, Waaga-Gasser AM, Gasser M, Zhan Q, (2009) Isolation of tumour stem-like cells
Jordan S, Duncan LM, Weishaupt C, from benign tumours. Br J Cancer
Fuhlbrigge RC, Kupper TS, Sayegh MH, 101:303–311
Frank MH (2008) Identification of cells initi- 75. Grimshaw MJ, Cooper L, Papazisis K, Coleman
ating human melanomas. Nature JA, Bohnenkamp HR, Chiapero-Stanke L,
451:345–349 Taylor-Papadimitriou J, Burchell JM (2008)
66. Roesch A, Fukunaga-Kalabis M, Schmidt EC, Mammosphere culture of metastatic breast
Zabierowski SE, Brafford PA, Vultur A, Basu cancer cells enriches for tumorigenic breast
D, Gimotty P, Vogt T, Herlyn M (2010) A cancer cells. Breast Cancer Res 10:R52
temporarily distinct subpopulation of slow- 76. Vander Griend DJ, Karthaus WL, Dalrymple
cycling melanoma cells is required for continu- S, Meeker A, DeMarzo AM, Isaacs JT (2008)
ous tumor growth. Cell 141:583–594 The role of CD133 in normal human prostate
67. Box GEP, Draper NR (1987) Empirical model- stem cells and malignant cancer-initiating cells.
building and response surfaces. Wiley, New Cancer Res 68:9703–9711
York 77. Dylla SJ, Beviglia L, Park IK, Chartier C, Raval
68. Bentivegna A, Conconi D, Panzeri E, Sala E, J, Ngan L, Pickell K, Aguilar J, Lazetic S,
Bovo G, Viganò P, Brunelli S, Bossi M, Tredici Smith-Berdan S, Clarke MF, Hoey T, Lewicki
G, Strada G, Dalprà L (2009) Biological het- J, Gurney AL (2008) Colorectal cancer stem
erogeneity of putative bladder cancer stem-like cells are enriched in xenogeneic tumors follow-
cell populations from human bladder transi- ing chemotherapy. PLoS One 3:e2428
tional cell carcinoma samples. Cancer Sci 78. Ricci-Vitiani L, Lombardi DG, Pilozzi E,
101:416–424 Biffoni M, Todaro M, Peschle C, De Maria R
69. Suva ML, Riggi N, Stehle JC, Baumer K, (2007) Identification and expansion of human
Tercier S, Joseph JM, Suva D, Clement V, colon-cancer-initiating cells. Nature
Provero P, Cironi L, Osterheld MC, Guillou 445:111–115
L, Stamenkovic I (2009) Identification of can- 79. Todaro M, Alea MP, Di Stefano AB, Cammareri
cer stem cells in Ewing’s sarcoma. Cancer Res P, Vermeulen L, Iovino F, Tripodo C, Russo
69:1776–1781 A, Gulotta G, Medema JP, Stassi G (2007)
70. Nijmeijer BA, Szuhai K, Goselink HM, van Colon cancer stem cells dictate tumor growth
Schie MLJ, van der Burg M, de Jong D, Marijt and resist cell death by production of interleu-
EW, Ottmann OG, Willemze R, Falkenburg kin-4. Cell Stem Cell 1:389–402
JHF (2009) Long-term culture of primary 80. Vermeulen L, Todaro M, de Sousa MF, Sprick
human lymphoblastic leukemia cells in the MR, Kemper K, Perez Alea M, Richel DJ,
absence of serum or hematopoietic growth Stassi G, Medema JP (2008) Single-cell clon-
factors. Exp Hematol 37:376–385 ing of colon cancer stem cells reveals a multi-
71. Rutella S, Bonanno G, Procoli A, Mariotti A, lineage differentiation capacity. Proc Natl Acad
Corallo M, Prisco MG, Eramo A, Napoletano Sci 105:13427–13432
C, Gallo D, Perillo A, Nuti M, Pierelli L, Testa 81. Todaro M, D’Asaro M, Caccamo N, Iovino F,
U, Scambia G, Ferrandina G (2009) Cells Francipane MG, Meraviglia S, Orlando V, La
with characteristics of cancer stem/progenitor Mendola C, Gulotta G, Salerno A, Dieli F,
cells express the CD133 antigen in human Stassi G (2009) Efficient killing of human
endometrial tumors. Clin Cancer Res 15: colon cancer stem cells by gammadelta T lym-
4299–4311 phocytes. J Immunol 182:7287–7296
82. Eramo A, Lotti F, Sette G, Pilozzi E, Biffoni of the adult mammalian central nervous
M, Di Virgilio A, Conticello C, Ruco L, Peschle system. Science 255:1707–1710
C, De Maria R (2008) Identification and 87. Weiss S, Pin JP, Sebben M, Kemp DE, Sladeczek
expansion of the tumorigenic lung cancer stem F, Gabrion J, Bockaert J (1986) Synaptogenesis
cell population. Cell Death Differ 15:504–514 of cultured striatal neurons in serum-free
83. Collins AT, Berry PA, Hyde C, Stower MJ, medium: a morphological and biochemical
Maitland NJ (2005) Prospective identification study. Proc Natl Acad Sci USA 83:2238–2242
of tumorigenic prostate cancer stem cells. 88. Clarke L, van der Kooy D (2009) Low oxygen
Cancer Res 65:10946–10951 enhances primitive and definitive neural stem
84. Pellegatta S, Poliani PL, Corno D, Menghi F, cell colony formation by inhibiting distinct cell
Ghielmetti F, Suarez-Merino B, Caldera V, death pathways. Stem Cells 27:1879–1886
Nava S, Ravanini M, Facchetti F, Bruzzone 89. Csete M (2005) Oxygen in the cultivation of
MG, Finocchiaro G (2006) Neurospheres stem cells. Ann N Y Acad Sci 1049:1–8
enriched in cancer stem-like cells are highly 90. Ghosh S, Spagnoli GC, Martin I, Ploegert S,
effective in eliciting a dendritic cell-mediated Demougin P, Heberer M, Reschner A (2005)
immune response against malignant gliomas. Three-dimensional culture of melanoma cells
Cancer Res 66:10247–10252 profoundly affects gene expression profile: a
85. Dontu G, Abdallah WM, Foley JM, Jackson high density oligonucleotide array study. J Cell
KW, Clarke MF, Kawamura MJ, Wicha MS Physiol 204:522–531
(2003) In vitro propagation and transcrip- 91. Coles-Takabe BL, Brain I, Purpura KA,
tional profiling of human mammary stem/ Karpowicz P, Zandstra PW, Morshead CM, van
progenitor cells. Genes Dev 17:1253–1270 der Kooy D (2008) Don’t look: growing clonal
86. Reynolds BA, Weiss S (1992) Generation of versus non-clonal neural stem cell colonies.
neurons and astrocytes from isolated cells Stem Cells. doi:10.1634/stemcells.2008-0558
Chapter 13
Ex Vivo Differentiation of Cord Blood Stem Cells

into Megakaryocytes and Platelets
Nicolas Pineault, Amélie Robert, Valérie Cortin, and Lucie Boyer
Abstract
Megakaryocytes (MK) are hematopoietic cells present in the bone marrow that are responsible for the
production and release of platelets in the circulation. Given their very low frequency (<1%), human MK
often need to be derived in culture to study their development or to generate sufficient material for bio-
logical studies. This chapter describes a simplified 14-day culture protocol that efficiently leads to the
production of MK and platelets from cord blood enriched progenitor cells. A serum-free medium is sug-
gested for the growth of the CB cells together with an optimized cytokine cocktail developed specifically
for MK differentiation, expansion, and maturation. Methodologies for flow cytometry analysis, MK and
platelets estimation, and MK progenitor assay are also presented.
Key words: Megakaryocytes, Platelets, Hematopoietic stem cells, Cord blood, Flow cytometry,
Ploidy analysis
1. Introduction
Megakaryocytes (MK) are derived in the marrow following the dif-

ferentiation of uncommitted progenitor cells along the MK lineage
(recently reviewed in ref. 1). The cytokine thrombopoietin (TPO)
is the principal regulator of megakaryopoiesis; it can be used as sole
cytokine in culture to favor specific MK differentiation and for the
maturation of MK into platelet producing cells. However, the yields
of MK and platelets obtained in TPO cultures are very low. Hence,
pleiotropic cytokines such as Stem Cell Factor/Kit ligand (SCF,
KL) (2, 3), Interleukin-3 (IL-3) (2, 4, 5), IL-6 (2, 6, 7), IL-9 (3,
8), IL-11 (9–11), and FLT-3 (FMS-like tyrosine kinase 3) ligand
(FL) (9–13) can be used with TPO in various combinations to
205
206 N. Pineault et al.
increase MK and/or platelet yields (6, 14). However, the use of

such cytokine reduces the purity of the MK in culture (3, 6, 14). In
this regard, erythroid cells are the major contaminant in MK cul-
tures derived from cord blood (CB) CD34+ cell cultures (3, 14).
Numerous culture protocols have been developed over the
years to derive MK and/or platelets ex vivo from human CB,
mobilized and bone marrow CD34+ cells as well as embryonic stem
cells (2–4, 6, 8, 11, 14–23). Most rely on the use of serum-free
medium supplemented with cytokine cocktails and often consist of
two or three different culture phases, as this strategy often leads to
improved yields (recently reviewed in ref. 24).
Importantly, most key events occurring during MK maturation
(recently reviewed in refs. 25, 26) appear to be well replicated ex
vivo, except for the regulatory activities normally provided by the
bone marrow niches (27). In brief, three major MK maturation
events occur: (1) polyploidization through endomitosis (28), (2)
cytoplasmic maturation leading to the formation of the membrane
demarcation system (29), and (3) proplatelet formation (30).
Platelets are eventually released from proplatelet filaments. This
process has recently been shown to occur in the marrow-intravas-
cular sinusoidal space (27).
MK express a large panel of cell surface receptors that have key
functions in platelets physiology. MK are commonly identified in unfrac-
tionated bone marrow cells or in heterogeneous cultures using mono-
clonal antibodies (mAb) against the fibrinogen receptor (composed of
CD41a (GPIIb) and CD61 (GPIIIa)) and/or the von Willebrand fac-
tor receptor (composed of CD42a (GPIX), CD42b (GPIba), CD42c
(GPIbb), and CD42d (GPV), at a stoichiometry of 2:2:2:1 respec-
tively). In general, the sequence of differentiation of uncommitted cells
toward mature MK is as follows: CD34+CD41−CD61−CD42b− (pro-
genitor cells), CD34+CD41+CD61+CD42b− (committed MK progeni-
tors), CD34−CD41+CD61+CD42b− (immature MK), and finally
CD34−CD41+CD61+CD42b+ (mature MK). Cytometry analysis of
cells in culture with double or triple staining with fluorochrome conju-
gated mAb is a simple and reliable technique to follow the kinetics of
MK differentiation and maturation occurring in culture. Cytometry is
also commonly used to evaluate the ploidy of the MK in culture.
Morphology of the MK can be examined by Wright-Giemsa stain-
ing of cells.
1.1. Chapter Overview In this chapter, a simple 14-day culture protocol is described to
produce MK and platelets in culture from CB CD34+ enriched
cells. Efficient MK differentiation of adult CD34+ cells can also be
achieved with this protocol, though MK and platelet yields are
substantially reduced. The cytokine cocktail proposed known as
BS1 is composed of SCF (1 ng/mL), TPO (30 ng/mL), IL-6
(7.5 ng/mL), and IL-9 (13.5 ng/mL). This cocktail was devel-
oped for the maturation of CB-derived MK by statistical design of
13 Megakaryocyte and Platelet Production in Culture 207
experiments, which included the screening of 13 cytokines and

optimization of the cytokine concentrations by a response surface
methodology (3). The same cocktail was subsequently shown to
induce efficient differentiation of CD34+ into MK (14), and also to
induce excellent MK progenitor expansion (Pineault N. et al., sub-
mitted manuscript). Readers are invited to consult the recent work
of Robert et al. to obtain information on how to purify culture-
derived platelets and assess their function and quality (31).
2. Materials
2.1. CD34+ Cell 1. Ficoll-Hypaque isotonic solution (Ficoll-Paque™ PLUS, GE

Preparation Healthcare, CAT# 17-1440-03).
2. EasySep™ separation magnet (EasySep™, StemCell
Technologies, Vancouver, BC, Canada).
3. The Human Progenitor Enrichment Cocktail kit with CD41
(StemCell Technologies, CAT # 19057). Lineage depletion
antibody mixtures (anti-CD2, -CD3, -CD11b, -CD11c,
-CD14, -CD16, -CD19, -CD24, -CD41, -CD56, -CD66b,
and -glycophorin A).
4. Cryoprotective medium:
(a) 40% Fetal Bovine Serum (FBS CAT # 26140, qualified,
Invitrogen).
(b) 10% Dimethylsulfoxide (DMSO; Sigma-Aldrich, CAT #
D2650).
(c) Mixed in Iscove’s modified Dulbecco medium (IMDM;
Invitrogen, CAT # 12440).
Equipment and cell culture supplies
5. Centrifuge (Beckman GS-6; Beckman Coulter, Montréal, QC,
Canada).
6. 15- and 50-mL polypropylene centrifuge tubes (Falcon,
Beckton Dickinson Labware, Franklin Lakes, NJ, USA).
7. 2-mL propylene cryotubes (VWR International, Mississauga,
ON, Canada).
2.2. Cell Cultures 1. Enriched CD34+ cells (see Note 1).

2. Iscove’s Modified Dulbecco’s Medium (IMDM, Gibco, CAT
# 12440, Burlington, ON, Canada).
3. 1 M PO4. For 1 L:
(a) 13.6 g Na2HPO4 (Fluka, CAT# 71636).
(b) 27.2 g KH2PO4 (Fluka, CAT# 60219).

(c) Adjust to pH 7.4.
(d) Complete to 1 L with H2O.
4. Phosphate buffered saline (PBS). For 1 L:
(a) 10 mL 1 M PO4.
(b) 8.5 g NaCl (Fisher CAT# S271).
(c) Complete to 1 L with H2O.
5. 50 mM 2-Mercaptoethanol (2-ME) containing 50 mM EDTA.
For 10 mL:
(a) 35 ml of 14.3 M 2-ME (Sigma-Aldrich, CAT # M7522).
(b) 9 mL of sterile-filtered (0.22-mm) PBS.
(c) 2.5 mL EDTA 200 mM (Sigma, CAT # E4884).
(d) Adjust to pH 6.0 with HCL 0.1 N.
(e) Complete to 10 mL with PBS before filtration (0.22-mm
filter).
(f) Keep sterile at 4°C up to 2 weeks (see Note 2).
6. Fetal Bovine Serum (FBS) (Gibco, CAT # 26140).
7. DNase (Sigma, CAT # D4513) resuspended at 1 mg/mL.
8. PBS-1% BSA. For 100 mL:
(a) 1 g of fraction V bovine serum albumin (BSA; Fisher,
CAT # BP1605).
(b) Complete to 100 mL with PBS.
(c) Filter (0.22-mm).
(d) Keep sterile at 4°C.
9. 10 mg/mL stock human recombinant cytokines (SCF, TPO,
IL-9, and IL-6; Peprotech (Rocky Hill, NJ)). Reconstitute
cytokine with PBS-1%BSA or following manufacturer’s instruc-
tions. Mix and dissolve well before aliquoting. Store at −35°C
(see Note 3).
10. Serum-free medium (SFM) (6) (see Note 4): IMDM asepti-
cally supplemented with 20% Bovine serum albumin/insulin/
transferin (BIT, serum substitute solution (Stem Cell
Technologies, CAT# 09500)), 20 mg/mL Low density lipo-
protein (LDL; Sigma-Aldrich, CAT # L7914), 100 mM 2-ME.
Keep sterile at 4°C. To prepare 25 mL:
(a) 19.9 mL IMDM.
(b) 5 mL BIT.
(c) 50 ml 5 × 10−2 M 2-ME.
(d) 100 ml LDL (see Note 4).
11. MK culture medium: SFM (above) supplemented with the

required volumes of cytokines. We recommend the use of the
cytokine cocktail BS1 (final concentration of 1 ng/mL SCF,
30 ng/mL TPO, 13.5 ng/mL IL-9, and 7.5 ng/mL IL-6)
(see Note 5).
12. GM6001(Calbiochem, CAT # 364205): Dilute stock solution at
40 mM with DMSO. Keep in small aliquots frozen at −20°C.
13. PBS-glucose solution: 2 g/L Glucose (Sigma-Aldrich, CAT #
G5146) and 0.5% Phenol red (Sigma-Aldrich, CAT # P0290)
in PBS. Sterilized by filtration (0.22 mm).
14. 0.4% Trypan blue solution (Invitrogen, CAT # 15250).
15. MegaCult™ kit (StemCell Technologies, Vancouver, Canada).
Equipment and cell culture supplies
16. Hemocytometer (Hausser Scientific, Horsham, PA, USA).
17. Microculture plates: 6-wells and 24-wells cell culture (Costar,
Corning Incorporated, Corning, or Beckton Dickinson
Labware).
18. Cell culture flasks: 25 cm2, 75 cm2 (Corning, NY).
19. 100 × 20 mm tissue culture dishes (Beckton Dickinson
Labware).
20. 15- and 50-mL polypropylene centrifuge tubes (Falcon,
Beckton Dickinson Labware).
21. Filter units: Millex-GV, 0.22 mm, PVDF, 33 mm filters
(Millipore, CAT# SLGV033RS).
2.3. Flow Cytometry 1. 10× MK buffer: 20 mM Theophylline and 272 mM Na-citrate

in PBS dissolved in a water bath at 37°C, sterilized by filtration
(0.22-mm). Keep sterile at 4°C. Stable for 3 months at 4°C (see
Note 6). To prepare 200 mL:
(a) 360 mg Theophylline (Sigma-Aldrich, CAT # T1633).
(b) 7,643 g Na-citrate (Fisher Scientific, CAT # S4641).
(c) Complete to 200 mL with PBS.
2. 1× MK buffer. Keep sterile at 4°C. For 200 mL:

(a) 20 mL 10× MK buffer.
(b) 2 g BSA (Fisher, CAT # BP1605).
(c) Complete to 200 mL with PBS (see Note 6).
3. 1% Formaldehyde. For 100 mL:
(a) 2.7 mL formaldehyde 37% (Sigma-Aldrich, CAT #
252549).
(b) Complete to 100 mL with PBS.
(c) Store at 4°C protected from light.
4. Fluorochrome-conjugated antibodies and their respective iso-

typic murine controls:
(a) Phycoerythrin (PE) coupled anti-human (h) CD34
(Immunotech, Beckman Coulter Co., Marseille, France).
(b) PE coupled anti-hCD62 (Immunotech).
(c) Allophycocyanin (APC)-coupled anti-hCD41a (Becton
Dickinson).
(d) Fluorescein isothiocyanate (FITC)-coupled anti-hCD42b
(Becton Dickinson).
(e) Isotypic controls; mouse IgG1 isotype control conjugated
to -APC, -PE, and -FITC.
5. Propidium iodide (PI) solution (1 mg/mL): For 25 mL:
(a) 25 mg PI (Sigma-Aldrich, CAT # P4170).
(b) Dissolve in 25 mL deionized water.
(c) Prepare 1-mL aliquots and store at −20°C.
6. PI/Triton solution (prepare fresh): For each sample, prepare
the following:
(a) 8 mL 10% Triton X-100 (Sigma, CAT # T9284).
(b) 16 mL PI 1 mg/mL.
(c) 40 mL DNase-free RNase A 2 mg/mL (Sigma-Aldrich,
CAT # R6513).
(d) 736 mL PBS.
7. Centrifuge microtubes made of homopolymer, 1.5 mL (Axygen
Scientific, Union City, CA, USA).
8. Centrifuge (Biofuge Pico, Heraeus Instruments).
9. Flow cytometer (FACS-Calibur, Becton Dickinson
Immunocytometry Systems, San Jose, CA, USA).
10. 5-mL Polystyrene FACS tubes (Bio-Rad Laboratories).
3. Methods
3.1. CD34+ Cell CB mononuclear cells (MNC) from umbilical CB units are isolated
Purification on a Ficoll-Hypaque density gradient and cryopreserved (see Note
1). MNC from three to six CB units are then thawed and mixed
prior to CD34 enrichment in order to reduce the inter-donor vari-
ability and to allow production of larger CD34 cell lots. CD34+
cells can also be enriched by fluorescent activated cell sorter
(FACS). It’s recommended to carefully review the manufacturer’s
instruction for the purification procedure.
3.1.1. mnc Preparation 1. MNC are isolated on a Ficoll-Hypaque density gradient fol-
lowing the manufacturer’s instructions.
2. MNC are then cryopreserved (Subheading 3.1.2), or CD34+
cells can be purified immediately (Subheading 3.1.4).
3.1.2. MNC 1. Centrifuge the MNC at 514 × g for 10 min.

Cryopreservation 2. Resuspend the cell pellet in cold (4°C) cryoprotective medium
at a density less than 300 × 106 cells/mL. Prepare 1 mL ali-
quots of this suspension into cryotubes.
3. Place the cryotubes at −80°C for 24 h and then transfer to
liquid nitrogen.
3.1.3. MNC Thawing 1. Thaw cryopreserved MNC in a 37°C water bath, without
shaking.
2. Transfer cells to a 15-mL tube. Complete to 14 mL with cold
(4°C) IMDM containing 20% FBS.
3. Centrifuge at 228 × g for 10 min at room temperature.
4. Remove supernatant and resuspend pellet with 10 mL of
IMDM containing 20% FBS and DNase (100 mg/mL) pre-
warmed at room temperature.
5. Incubate for 15 min at room temperature.
6. Take a sample to count cells and assess viability (see Note 7).
7. Centrifuge at room temperature (228 × g for 10 min).
8. Decant supernatant, resuspend cells to a density of
2–8 × 107 cells/mL in PBS-glucose containing 2% FBS and
1 mM EDTA for magnetic labeling (see Subheading 3.1.4).
3.1.4. Magnetic Labeling CD34+ cells can be purified by positive or negative selection.
and CD34+ Cell Separation Several commercial kits are available for this purpose. The 14-day
MK and Platelets culture protocol described in this chapter was
developed with CD34+ cells immuno-selected negatively with a kit
offered by StemCell Technologies. Follow manufacturer’s instruc-
tion for the purification procedure. CD34+ enriched cells
(purity ³ 75%) are then aliquoted and cryopreserved.
3.2. Ex Vivo Expansion The culture protocol usually lasts about 14 days. MK and platelet
and Maturation of MK productions usually peak at day-14. However, it is recommended
Starting from CB to follow MK and platelet-like particles (PLPs) production kinetics
CD34+ Cells (e.g., day-12, -14, -16) to find the optimal production time, since
it can vary between laboratories due to technical differences.
Mature MK can be observed as larger cells in culture using a
phase contrast inverted microscope, and proplatelet filaments can
be observed by day-10. Great care must be taken to observe these
structures as they are fragile. Ploidy analyses (Subheading 3.4)
are typically done by cytometry starting at day-10. Cell counts and
flow cytometry analyses (Subheading 3.3.1) should be performed

on the same day to accurately follow the expansion and differentia-
tion kinetics of the CB CD34+ cells into MK and platelets.
1. Gently thaw the cryopreserved CD34+ cells in a 37°C water
bath, without shaking.
2. Transfer the cells to a 15-mL tube. Complete with PBS-glucose
solution at 37°C by adding the first 5 mL very slowly.
3. Centrifuge at 228 × g for 10 min and discard the supernatant.
4. Resuspend the cells in MK culture medium to a density of
3–6 × 105 cells/mL.
5. Measure the cell density and cell viability (see Note 7).
6. Initiate CD34+ cell culture (40,000 cells/mL) in MK culture
medium. Cultures are usually performed in 24-well tissue cul-
ture plates (0.8–1 mL). Use the 8-central wells for cell cultures
and fill the external wells with PBS to reduce evaporation.
Greater cell density can be used if analyses of the cultures are
performed within the first few days of culture (see Note 8).
Cultures can done in smaller or larger culture dishes or flasks as
preferred.
7. Incubate cultures at 39°C, 10% CO2, in a humidified incubator
(see Note 9).
8. At day-4, mix gently the cell cultures, remove half of the cell
suspension volume and add an equivalent volume of fresh MK
culture medium. For large-scale platelet production, just add
an equivalent volume of fresh MK culture medium.
9. At day-7, mix gently the cell cultures, dilute the cultures with
fresh MK culture medium to reach a cell density of
2 × 105–3 × 105 cells/mL, and transfer the cultures at 37°C.
10. At day-10 or 11:
(a) Mix gently the cell cultures.
(b) Measure cell density.
(c) Transfer the cell culture to 15- or 50 mL tube(s).
(d) Centrifuge at 228 × g for 10 min and discard the
supernatant.
(e) Resuspend the cells in fresh MK culture medium to reach
3 × 105 cells/mL. For large-scale culture, greater platelet
yield can be reached if the culture is transferred to 100 mm
dishes instead of flask. In this case, do not exceed 10 mL
per dishes.
11. At day-12, add the metalloproteinase inhibitor GM6001 (final
concentration of 25 mM) to all culture dishes (see Note 10).
12. At day-14 or at the determined optimal PLPs production time,
mix gently the cultures by pipetting up and down (5–10 times)
the cultures with a p1000 micropipette. For larger cultures,

place a p1000 tip at the end of a serological (5- or 10-mL)
pipette and mix the culture by pipetting up and down
10–15 times.
3.3. Analysis of the MK For a quantitative monitoring of megakaryopoiesis (see Subheading

and Platelet Cultures “Analysis of MK Expansion and Platelet Production Kinetics”),
cells should be counted and characterized by flow cytometry (see
Subheading 3.3.1, Fig. 1). Ploidy analysis of the MK can also be
done by cytometry (see Subheading 3.4), while MK progenitor
frequency is determined using a colony-based assay (see
Subheading 3.5). For morphology analyses, cells can be deposited
on glass slide (cytospin) and stained with Wright-Giemsa staining,
or if necessary, visualized by electron microscopy (31).
3.3.1. Flow Cytometry Phenotypic analysis by flow cytometry is carried out using
Analysis of the MK and fluorochrome-conjugated antibodies against CD41a (or CD61)
Platelet Cultures and CD42b. Other markers can also be used, such as CD62 which
is used to determine the activation status of platelet (see Note 11).
Note that MK and platelets do not express CD45. MK-progenitors
can also be estimated by flow cytometry as the CD34+CD41a+
population, though we recommend using a more robust assay such
as MegaCultTM (see Subheading 3.5). The first step is to establish a
proper cell/platelet analysis acquisition template for the acquisi-
tion software used with the cytometer (such as CellQuest™).
Flow Cytometry Figure 1 presents an overview of the recommended flow cytometry

Acquisition Setup methodology used to analyze normal blood platelets (Fig. 1a–c),
MK (Fig. 1d–f) and culture-derived platelets (Fig. 1d, g, h). Recent
results from our laboratory demonstrated that functional platelets
produced in culture co-express CD41a and CD42b, and that those
that fail to maintain CD42b expression are for most part metaboli-
cally inactive (31). Hence, culture-derived platelets (referred to as
PLPs) are enumerated as PI-negative CD41a+CD42b+ events with
scatter (FSC, SSC) properties similar to fresh platelets. A hematopoi-
etic cell line and control platelets prepared from fresh blood (see
Subheading “Preparation of Control Platelets for Validation of the
Platelet Region”) should be used to facilitate and validate the
cytometry acquisition setup. Follow the instructions provided
below to establish the acquisition setup on the cytometer.
1. Cell region R1: In a dot-plot acquisition window, with forward
scatter (FSC) and side scatter (SSC) on the x- and y-axis respec-
tively, draw the R1 region around the nucleated cells, which
have FSC and SSC properties significantly greater than plate-
lets and cell debris (Fig. 1d). This region is used as a gate to
specifically select cell events.
2. Propidium iodide negative (PI) cell region R3: In a dot-plot
acquisition window acquiring events from the R1 region only,
Fig. 1. Flow cytometry methodology used to analyze MK and platelets. (a–c) Analysis of fresh platelets from platelet-rich
plasma (PRP). (d–i) Analysis of ex vivo produced MK and platelets. In more details; (a) Normal platelets derived from a PRP
are first used to set the analytical platelet region “R2” in a 2D dot-plot acquisition window showing the forward-side scatter
(FSC) and side scatter (SSC) properties on the x- and the y-axis respectively. (b) On an independent dot-plot window,
platelet events from the R2 region are analyzed for their retention of Propidium iodide (PI), which is null for normal platelets.
(c) PI negative platelet events (R2*R4 region) are then analyzed for the expression of CD41a and CD42b, on the y- and
x-axis respectively. The use of fresh PRP platelets also ensures proper adjustments of the various settings of the flow
cytometer. (d) FSC and SSC analysis window of cell cultures. Platelet events appear in the previously drawn R2 region,
whereas MNC appear as distinctive events of greater size (FSC) and greater granularity (SSC). The cell region “R1” is drawn
around that second population. (e, g) The cell and platelet events from the regions R1 and R2 are then analyzed for their
retention of PI. The PI negative selection regions for the cells (R3) and platelets (R4) events are shown in (e) and (g), respec-
tively. (f and h) The PI negative cell and platelet events (from R1*R3 and R2*R4 regions, respectively) are analyzed for
CD41a and CD42b (or other markers of interest) expression in (f) and (h), respectively. Reproduced in part from Cortin et al.
(32) with permission from Humana Press.
select FSC and PI (read on FL3 channel) on the x- and y-axis

respectively, draw the R3 region around the PI negative cell events
(Fig. 1e). This region is used as a gate to select viable cells.
3. Platelet region R2: Draw the platelet region R2 in the same
dot-plot acquisition window (Fig. 1d) in which the cell region
was drawn (step 1 above). Platelet events have reduced FSC
and SSC properties. The platelet region is used as a gate to
select platelet events and it must be set and validated using
fresh platelets (see Subheading “Preparation of Control Platelets
for Validation of the Platelet Region”).
4. PI negative platelet region R4: In a dot-plot acquisition win-
dow acquiring events from the R2 region only, select FSC and
PI (on FL3 channel) on the x- and y-axis respectively, draw the
R4 region around the PI negative platelet events (Fig. 1g). This
region is used as a gate to select platelets events. Normal plate-
lets are negative for PI staining (Fig. 1b). PI+ platelets events
seen with culture samples are most likely cellular debris.
5. In dot-plot acquisition window, select CD41a (if APC anti-
body, FL4 channel) and CD42b (if FITC antibody, FL1 chan-
nel) on the y- and x-axis respectively to analyze MK (gating
R1*R3 events only), or platelets (gating R2*R4 events only) as
shown in Fig. 1f and h, respectively.
6. Run unstained and isotopic control stain(s) to adjust the
cytometer settings and to measure background levels.
7. Run stained samples.
8. Once the flow cytometry acquisition template and flow cytom-
eter settings are validated with culture samples and normal
platelets they should be saved for future use. In the latter case,
new samples can be analyzed rapidly without the requirement
of preparing fresh normal platelets.
Preparation of Control 1. Centrifuge 1–10 mL of citrated whole adult blood at 330 × g

Platelets for Validation for 10 min, without brake.
of the Platelet Region 2. Collect with a pipette the upper phase of the centrifuged sam-
ple which represents the platelet-rich plasma (PRP) fraction.
3. Centrifuge the PRP (1,428 × g for 15 min).
4. Discard the supernatant and resuspend platelets in MK buffer.
5. Sample an aliquot for platelet count; this can be done on a
hematologic analyzer. Alternatively, counts can be estimated
using a hemocytometer. If the latter method is used, dilute
sample in 3% acetic acid to lyse red cells; platelets will appear as
small dots since they are 10–25× smaller than nucleated cells.
6. Stain a small fraction of PRP samples: dilute ~1 × 106 platelets
in MK buffer pre-warmed to room temperature, add antibod-
ies against platelet antigens such as CD41a, CD42b, and CD62
and incubate for 15–30 min at room temperature (see Note

12). CD62 is optional (see Note 11). Each specific stain should
be accompanied by an isotopic control stain done in parallel at
the same antibody dilution.
7. Add directly to the 0.2-mL stained aliquots, 0.5-mL MK buf-
fer pre-warmed to room temperature containing 7 mg/mL PI
in order to reach a final PI concentration of 5 mg/mL (see
Note 13).
8. Analyze by flow cytometry as indicated in Subheading “Flow
Cytometry Acquisition Setup” as soon as possible since plate-
lets are sensitive to spontaneous activation. Acquire a mini-
mum of 10,000 events to identify the platelet region R2
(Fig. 1a–c) with specific FSC and SSC characteristics.
Preparation of the This procedure is designed for the simultaneous analysis of cells and
Culture Samples for Flow platelets produced in culture by flow cytometry. This can be per-
Cytometry Analysis of Cells formed any time during the culture (e.g., day-0, 7, 10, and 14).
and Platelets
1. Mix gently the cell cultures.
2. Measure cell density using an hemocytometer and trypan blue
(see Note 7).
3. Transfer 0.2-mL aliquots of cell suspension (100,000–
500,000 cells), into two 5-mL FACS tube.
4. Add antibodies (i.e., x-CD41a, x-CD42b, etc.) to each tube
(see Note 12) and incubate for 15–30 min in the dark (on ice
for cell analysis only, or room temperature otherwise). Each
specific stain should be accompanied by an isotopic control
stain done in parallel at the same antibody dilution.
5. Option (1) For cell acquisition only.
(a) Wash the cell–antibody mixtures by adding 1–2 mL of
MK buffer (room temperature)
(b) Centrifuge at 1,000 × g for 5 min.
(c) Discard supernatants and resuspend pellets in 0.3–0.5-mL
MK buffer containing 5 mg/mL PI.
Option (2) for cell and platelet acquisition:
(a) Add 0.5-mL of MK buffer (pre-warmed to room tempera-
ture) containing 7 mg/mL PI to the 0.2-mL stained ali-
quots (see Note 13).
6. Proceed to flow cytometry analysis as indicated in Subheading
“Flow Cytometry Acquisition Setup” as soon as possible since
samples are not fixed and platelets are sensitive to spontaneous
activation. Acquire a minimum of 10,000 PI-negative cell
events (R1*R3 events).
Analysis of MK Expansion For each time point of the culture analysis, MK and platelet yields
and Platelet Production can be calculated. Typically, PLPs estimation is only done after
Kinetics 10 days of culture. Calculation of MK yields:
1. The actual total number of MK produced at time t:
Total MKs = DVC(t ) × CV(t ) × %CD41a +
Total mature MKs = DVC(t) × CV(t) × %CD41a + CD42b +
DVC(t): density of viable cells at time t (see Note 7)
CV(t): volume of culture at time t
%CD41+ and %CD41+CD42b+ cells: derived from the cytom-
etry analysis at time t
2. The theoric cumulaqive ( „ ) number of MK produced per
seeded cells at time t. That is, if all cells had been maintained
in culture from day-0 to time “t”:
∑ total MK = TNC(t ) × %CD41a +
∑ mature MK = TNC(t ) × %CD41a +

CD42b +
TNC(t): cumulative total cell expansion (TNC) per seeded cells

TNC(t) = DVC(t) × CDF(t)/DVC(day 0) (see Note 14)
CDF(t): cumulative dilution factor (CDF, see Note 15)
3. Estimation of PLPs production:
There are several ways to calculate the platelet yields. Our lab-
oratory uses the cell density and the platelet to cell ratio mea-
surements to estimate the number of PLPs produced.
The actual number of PLPs (Nb PLP) produced in culture
at time “t”:
Nb PLT(t ) = DVC(t ) × CV(t ) × %CD41+ CD42+ PLPs × R( P / C )t
%CD41+CD42+ PLPs: percentage of platelets that are

CD41+CD42b+ (Fig. 1h, upper right quadrant)
R(P/C)t: ratio at time “t” of the number of platelet events in
R4 (Fig. 1g) divided by the number of cell events in R3
(Fig. 1e). These numbers are presented in the statistical analy-
sis of the dot-plots.
4. The theoric cumulative number of PLPs produced in culture at
time “t” per seeded cells % Nb PLPt):
∑ Nb PLPt = TNC(t ) × %CD41a +
CD42b + PLPs × R( P / C )t
To get the theoretical
5. Total number of PLPs produced, mulqiply Nb PLPt by the
actual number of cells seeded at day-0.
Fig. 2. Ploidy analysis of MK produced in culture. Overview of the flow cytometry acquisition setup and ploidy analysis of
MK. Ploidy analysis can be done on whole cells or in the CD41a+ and CD42b+ MK subpopulations
3.4. Ploidy Analysis This protocol is based in part on a technique previously described
of MK by Cytometry by Darzynkiewick et al. (33) (see Note 16). For further details and
troubleshooting, refer to original work. An overview of the cytom-
etry analysis of ploidy is presented in Fig. 2. Polyploid MK are rela-
tively rare when derived from CB CD34+ cells compared to adult
CD34+-derived cells (34, 35). Ploidy analysis can be done on total
cells, total MK (CD41+) or on CD42b+ cells. The latter is recom-
mended as the frequency of polyploid MK is greater in the CD42b
bright cell population (Fig. 2).
3.4.1. Preparation 1. Gently mix the cultures and transfer 5 × 105 to 1 × 106 cells into
of Samples two 5-mL FACS tube.
2. Centrifuge cells at 200 × g for 5 min.
3. Resuspend in 200 mL of PBS-1%BSA.
4. In tube #1, add the x-CD41a and x-CD42b antibodies. In
tube #2, add the isotopic control antibodies. Incubate for
20 min in the dark.
5. Wash the cells: Add 2 mL of PBS-1%BSA, centrifuge cells
(200 × g for 5 min), discard supernatants, and resuspend pellets
in 0.4 mL of formaldehyde 1% solution.
6. Incubate for 15 min at room temperature. Samples can also be
stored for longer period at 4°C.
7. Wash the cells: Add 2 mL of PBS, centrifuge cells (200 × g for

5 min), and discard supernatants.
8. Repeat step 7 and resuspend cell pellets 0.8 mL of PI-Triton
solution. Incubate for 30 min in the dark at room
temperature.
9. The samples are then ready for cytometry analysis.
3.4.2. Cytometry Analysis Ploidy analysis of CD41+ or CD42b+ cells is usually done using
of the Prepared Samples histograms (Fig. 2c), though it can also be done on dot plots, by
plotting CD41 (or CD42b, or FCS) on the x-axis, and PI (FL2-A
channel) on the y-axis.
1. Establish the cytometry acquisition setup as shown in Fig. 2.
2. Run samples with the cytometer. It’s recommended to acquire
as many events as possible to improve the quality and esthetic
of the analysis.
3. Once acquisition is finished.
4. Analysis of the cytometry data can be done with the use of his-
tograms as shown in Fig. 2. Ploidy analysis can be done on the
whole cells, on MK only (CD41+ cells), on mature MK (CD42b+
cells) and finally on CD42b bright cells. Polyploid MK are
enriched in the CD42b bright fraction. Draw regions over the
<2 N, 2 N, 4 N, 8 N, and >8 N populations on the histograms
(Fig. 2). If polyploid cells are too low in frequency, it may be
difficult to accurately draw the 8 N population. In this event, we
recommend to draw a ³ 8 N region as shown in Fig. 2.
3.5. Assay for MK Culture favoring MK expansion and differentiation also support
Progenitors the expansion of MK-progenitors. The BS1 cocktail provides good
MK-progenitor expansion ranging from 5- to 75-fold for culture
maintain for 3–10 days at 39°C. We recommend the use of the
collagen-based kit known as MegaCult™ to measure MK progeni-
tors, which are referred to as Colony-Forming Unit-Megakaryocyte
(CFU-MK). The frequency of CFU-MK in culture decreases as a
function of culture time, we suggest to plate two different cell
doses (1× to 2×). We recommend to plate 2,250 cells between day
0 and day 5 of culture, and 3,500 between day 6 and day 10. The
MegaCult™ assay uses an anti-GPIIb/IIIa immunostaining to
ensure specific enumeration of MK-progenitors (MK colonies
appearing in pink). MK colonies are scored based on their size;
small colonies (3–20 cells), medium colonies (21–49 cells) and
large colonies (>50 cells). Large and small colonies are formed
from immature and mature MK progenitors, respectively, while
the medium colonies are derived from semi-mature progenitors.
Background information, material and methods are all supplied in
the MegaCult™ kit (StemCell Technologies, Vancouver, Canada).
To assess the expansion of MK-progenitors in culture,
MK-progenitor frequency (f) should be determined at day 0, and

at any other time of interest (day X). The net MK-progenitor
expansion is then calculated as follows: the total cell expansion is
multiplied by f on day X and divided by f on day 0.
4. Notes
1. CB were collected after informed consent from the mother

and approved protocols from Héma-Québec and Québec City
Saint-François d’Assise hospital ethical committees. MNC
were isolated within 12 h following their collection in aseptic
conditions.
2. 2-ME is added in culture to prevent oxidation of LDL. Pre-
dilution of 2-ME is prepared following Sigma-Aldrich recom-
mendations which specify that antioxidant properties of the
compound is better preserved at pH 6 in presence of EDTA.
3. Cytokine aliquots should not be submitted to more than one
freeze/thaw cycle.
4. Low yields of cell expansion (<100 TNC), MK differentiation
(<75% CD41a+) and maturation (<60% CD41a+CD42b+) and/
or low platelet yields (<20 per CD34+ cells) by day-14 are usu-
ally due to a culture media-related problem. For this reason,
we recommend to prepare fresh basal medium on the day that
the cell culture is fed or diluted past day 7. Our MK-culture
protocol was developed with presolubilized LDL solution.The
type and source of LDL have an important impact on the
platelet yields. Some LDL lots fail to support efficient platelet
production for reasons curently unknown. LDL from StemCell
Technologies was found as efficient as that from Sigma.
5. The cytokine cocktail BS1 (3) can be used for the entire length
of culture, which gives similar MK purity (> 90% CD41a+ cells
on day-14), but increases MK and platelet yields (~5-fold)
when compared to TPO (10–100 ng/mL) alone.
6. It is important to keep the MK buffer (10× and 1×) sterile even
though cytometry analysis does not require aseptic manipula-
tions. Indeed, bacterial contaminated MK buffer results in
numerous false platelet events (R2 region, Fig. 1) that will per-
turb flow cytometry analysis. Moreover, MK buffer concen-
trated more than 10× will crystallize.
7. For cell counts and viability assessment, a 20 ml cell suspension
sample is mixed with 20 ml of 0.4% trypan blue solution.
Further dilution with trypan blue is made in order to count no
more than 200 events in the hemocytometer. Dead cells will
absorb the stain and become blue, while live cells do not and
remain uncolored. Only consider the uncolored cells to calcu-

late the density of viable cells (DVC).
8. The culture size depends principally on the quantity of cells
desired or the number of different conditions needed to be
tested. Thus, early large cell requirement will necessitate larger
culture volume or greater seeding cell-density (cell expansion
ratio is around fourfold at day 4), whereas analysis done after
10 days of culture requires fewer starting cells due to increas-
ing cell expansion (expected final cell expansion ratio of 100–
250 and 200–600-fold at day 10 and 14, respectively). For
large scale platelets production, start with 300,000 CD34+
cells to obtain 150–250 mL of culture at day 14 with a PLP
yield between 12 and 54 millions.
9. Héma-Québec Laboratory has previously demonstrated that
CB cells cultured at 39°C experience an accelerated MK dif-
ferentiation kinetics and improved MK and platelet yields (36),
but the culture protocol described herein can also be per-
formed at 37°C. Recent optimization of the culture period
length at 39°C led to the discovery that MK and platelet yields
are even greater if the cultures are switch from 39°C back to
37°C on day 4 or 7 (37).
10. The broad range metalloproteinase inhibitor GM6001 is added
to cultures to avoid ectodomain shedding of key receptors on
newly released platelets, such as GP1ba, GPV and GPVI (23,
38). Though this step is optional, addition of GM6001 is highly
beneficial to the quality of PLPs produced at day-14 (31).
11. Platelet activation can be conveniently monitored by flow
cytometry using an antibody recognizing the P-selectin CD62.
This antigen is present in the alpha granules of platelets, and is
exposed to their surface following activation of the platelets.
12. Routinely, all antibodies listed in Subheadings 2.3 and 3.3 are
diluted 1:40, with the exception of the anti-CD41a which is
used at 1:200. We recommend titrating all antibodies
solutions.
13. For platelet analyses, centrifugation is omitted in order to avoid
platelet activation. In our experience, no significant difference
has been noticed between washed and unwashed stained cells
in regards to non specific staining.
14. The cumulative total cell expansion can be calculated more
accurately by calculating each expansion factor between cell
passages as follows: DVC(t) × CV(t)/Nb cells plated. Then,
TNC(t) is calculated by multiply together all expansion factors
up to time t. This technique is more accurate since it takes in
account changes in culture volume.
15. The cumulative dilution factor (CDF) is calculated by multi-
plying all dilution factors used prior to the time of analysis.
For example, CDF is equal to 12, if the cultures were diluted

1/2 at day 4, 1/3 at day 7 and 1/2 at day 10, the CDF = 1/
(1/2 × 1/3 × 1/2).
16. In this protocol, fixation of the cells and DNA staining are
done using a formaldehyde and PI solutions, respectively.
Other methods for the fixation (e.g., ethanol) and DNA stain-
ing (e.g., DAPI, Hoechst 33342) are also available (28, 39).
Acknowledgments
This work was supported in part by a National Blood Foundation

Grant. A. Robert was the recipient of an Industrial R&D Fellowships
from the Canadian Natural Science and Engineering Research
Council.
References
1. Pang L, Weiss MJ, Poncz M (2005) 7. Williams JL, Pipia GG, Datta NS, Long MW
Megakaryocyte biology and related disorders. (1998) Thrombopoietin requires additional
J Clin Invest 115:3332–3338 megakaryocyte-active cytokines for optimal ex
2. Dolzhanskiy A, Basch RS, Karpatkin S (1997) vivo expansion of megakaryocyte precursor
The development of human megakaryocytes: cells. Blood 91:4118–4126
III. Development of mature megakaryocytes 8. Fujiki H, Kimura T, Minamiguchi H, Harada
from highly purified committed progenitors in S, Wang J, Nakao M, Yokota S, Urata Y, Ueda
synthetic culture media and inhibition of Y, Yamagishi H, Sonoda Y (2002) Role of
thrombopoietin-induced polyploidization by human interleukin-9 as a megakaryocyte
interleukin-3. Blood 89:426–434 potentiator in culture. Exp Hematol
3. Cortin V, Garnier A, Pineault N, Lemieux R, 30:1373–1380
Boyer L, Proulx C (2005) Efficient in vitro 9. De Bruyn C, Delforge A, Martiat P, Bron D
megakaryocyte maturation using cytokine (2005) Ex vivo expansion of megakaryocyte
cocktails optimized by statistical experimental progenitor cells: cord blood versus mobilized
design. Exp Hematol 33:1182–1191 peripheral blood. Stem Cells Dev 14:415–424
4. van den Oudenrijn S, von dem Borne AE, de 10. Feng Y, Zhang L, Xiao ZJ, Li B, Liu B, Fan
Haas M (2000) Differences in megakaryocyte CG, Yuan XF, Han ZC (2005) An effective
expansion potential between CD34(+) stem and simple expansion system for megakaryo-
cells derived from cord blood, peripheral cyte progenitor cells using a combination of
blood, and bone marrow from adults and chil- heparin with thrombopoietin and interleu-
dren. Exp Hematol 28:1054–1061 kin-11. Exp Hematol 33:1537–1543
5. Schattner M, Lefebvre P, Mingolelli SS, 11. Matsunaga T, Tanaka I, Kobune M, Kawano
Goolsby CL, Rademaker A, White JG, Foster Y, Tanaka M, Kuribayashi K, Iyama S, Sato T,
D, Green D, Cohen I (1996) Thrombopoietin- Sato Y, Takimoto R, Takayama T, Kato J,
stimulated ex vivo expansion of human bone Ninomiya T, Hamada H, Niitsu Y (2006) Ex
marrow megakaryocytes. Stem Cells vivo large-scale generation of human platelets
14:207–214 from cord blood CD34+ cells. Stem Cells
6. Proulx C, Boyer L, Hurnanen DR, Lemieux R 24:2877–2887
(2003) Preferential ex vivo expansion of mega- 12. Shaw PH, Gilligan D, Wang XM, Thall PF,
karyocytes from human cord blood CD34+- Corey SJ (2003) Ex vivo expansion of mega-
enriched cells in the presence of thrombopoietin karyocyte precursors from umbilical cord
and limiting amounts of stem cell factor and blood CD34 cells in a closed liquid culture sys-
Flt-3 ligand. J Hematother Stem Cell Res tem. Biol Blood Marrow Transplant
12:179–188 9:151–156
13. Sigurjonsson OE, Gudmundsson KO, regulation improves in vitro generation of

Haraldsdottir V, Rafnar T, Agnarsson BA, efficacious platelets from mouse embryonic
Gudmundsson S (2004) Flt3/Flk-2 ligand in stem cells. J Exp Med 205:1917–1927
combination with thrombopoietin decreases 24. Reems JA, Pineault N, Sun S (2010) In vitro
apoptosis in megakaryocyte development. megakaryocyte production and platelet bio-
Stem Cells Dev 13:183–191 genesis: state of the art. Transfus Med Rev
14. Boyer L, Robert A, Proulx C, Pineault N 24:33–43
(2008) Increased production of megakaryo- 25. Battinelli EM, Hartwig JH, Italiano JE Jr
cytes near purity from cord blood CD34+ cells (2007) Delivering new insight into the biology
using a short two-phase culture system. of megakaryopoiesis and thrombopoiesis. Curr
J Immunol Methods 332:82–91 Opin Hematol 14:419–426
15. Blair A, Baker CL, Pamphilon DH, Judson PA 26. Bluteau D, Lordier L, Di Stefano A, Chang Y,
(2002) Ex vivo expansion of megakaryocyte Raslova H, Debili N, Vainchenker W (2009)
progenitor cells from normal bone marrow Regulation of megakaryocyte maturation and
and peripheral blood and from patients with platelet formation. J Thromb Haemost 7(Suppl
haematological malignancies. Br J Haematol 1):227–234
116:912–919 27. Junt T, Schulze H, Chen Z, Massberg S,
16. Choi ES, Hokom MM, Nichol JL, Hornkohl Goerge T, Krueger A, Wagner DD, Graf T,
A, Hunt P (1995) Functional human platelet Italiano JE Jr, Shivdasani RA, von Andrian UH
generation in vitro and regulation of cytoplas- (2007) Dynamic visualization of thrombopoi-
mic process formation. C R Acad Sci III esis within bone marrow. Science
318:387–393 317:1767–1770
17. Ungerer M, Peluso M, Gillitzer A, Massberg S, 28. Lordier L, Jalil A, Aurade F, Larbret F,
Heinzmann U, Schulz C, Munch G, Gawaz M Larghero J, Debili N, Vainchenker W, Chang Y
(2004) Generation of functional culture- (2008) Megakaryocyte endomitosis is a failure
derived platelets from CD34+ progenitor cells of late cytokinesis related to defects in the con-
to study transgenes in the platelet environ- tractile ring and Rho/Rock signaling. Blood
ment. Circ Res 95:e36–e44 112:3164–3174
18. Lefebvre P, Winter JN, Meng Y, Cohen I 29. Schulze H, Korpal M, Hurov J, Kim SW,
(2000) Ex vivo expansion of early and late Zhang J, Cantley LC, Graf T, Shivdasani RA
megakaryocyte progenitors. J Hematother (2006) Characterization of the megakaryocyte
Stem Cell Res 9:913–921 demarcation membrane system and its role in
19. Sullenbarger B, Bahng JH, Gruner R, Kotov thrombopoiesis. Blood 107:3868–3875
N, Lasky LC (2009) Prolonged continuous 30. Italiano JE Jr, Lecine P, Shivdasani RA,
in vitro human platelet production using three- Hartwig JH (1999) Blood platelets are assem-
dimensional scaffolds. Exp Hematol bled principally at the ends of proplatelet pro-
37:101–110 cesses produced by differentiated
20. Gandhi MJ, Drachman JG, Reems JA, megakaryocytes. J Cell Biol 147:1299–1312
Thorning D, Lannutti BJ (2005) A novel strat- 31. Robert A, Boyer L, Pineault N (2011)
egy for generating platelet-like fragments from Glycoprotein Iba receptor instability is associ-
megakaryocytic cell lines and human progeni- ated with loss of quality in platelets produced
tor cells. Blood Cells Mol Dis 35:70–73 in culture. Stem Cells Dev 20:379–390
21. Case J, Hicks C, Trickett A, Kwan YL, 32. Cortin V, Pineault N, Garnier A (2009) Ex
Manoharan A (2006) The expansion of mega- vivo megakaryocyte expansion and platelet
karyocyte progenitors from CD34+-enriched production from human cord blood stem cells.
mobilized peripheral blood stem cells is inhib- Methods Mol Biol 482:109–126
ited by Flt3-L. J Interferon Cytokine Res 33. Darzynkiewick Z, Huang X (2004) In:
26:76–82 Collingan JE, Kruisbeek AM, Margulies DH,
22. Shaw PH, Olszewski M, Kletzel M (2001) Shevach EM, Strober W (eds) Current proto-
Expansion of megakaryocyte precursors and cols in immunology. Wiley, New York, pp
stem cells from umbilical cord blood CD34+ 5.7.1–5.7.18
cells in collagen and liquid culture media. J 34. Bornstein R, Garcia-Vela J, Gilsanz F, Auray
Hematother Stem Cell Res 10:391–403 C, Cales C (2001) Cord blood megakaryo-
23. Nishikii H, Eto K, Tamura N, Hattori K, cytes do not complete maturation, as indicated
Heissig B, Kanaji T, Sawaguchi A, Goto S, by impaired establishment of endomitosis and
Ware J, Nakauchi H (2008) Metalloproteinase low expression of G1/S cyclins upon throm-
bopoietin-induced differentiation. Br J Characterization of the effects and potential

Haematol 114:458–465 mechanisms leading to increased megakaryo-
35. Mattia G, Vulcano F, Milazzo L, Barca A, cytic differentiation under mild hyperthermia.
Macioce G, Giampaolo A, Hassan HJ (2002) Stem Cells Dev 17:483–494
Different ploidy levels of megakaryocytes gen- 38. Bergmeier W, Burger PC, Piffath CL,
erated from peripheral or cord blood CD34+ Hoffmeister KM, Hartwig JH, Nieswandt B,
cells are correlated with different levels of Wagner DD (2003) Metalloproteinase inhibi-
platelet release. Blood 99:888–897 tors improve the recovery and hemostatic
36. Proulx C, Dupuis N, St-Amour I, Boyer L, function of in vitro-aged or -injured mouse
Lemieux R (2004) Increased megakaryopoie- platelets. Blood 102:4229–4235
sis in cultures of CD34-enriched cord blood 39. Muntean AG, Pang L, Poncz M, Dowdy SF,
cells maintained at 39 degrees C. Biotechnol Blobel GA, Crispino JD (2007) Cyclin D-Cdk4
Bioeng 88:675–680 is regulated by GATA-1 and required for
37. Pineault N, Boucher J-F, Cayer M-P, Palmqvist megakaryocyte growth and polyploidization.
L, Boyer L, Lemieux R, Proulx C (2008) Blood 109:5199–5207
Chapter 14
Generation and Characterization of Murine Alternatively

Activated Macrophages
Shelley B. Weisser, Keith W. McLarren, Etsushi Kuroda,
and Laura M. Sly
Abstract
Macrophages play a key role in the innate immune response and help to direct the acquired immune
response. Early in the innate immune response, they produce reactive oxygen species and pro-inflammatory
cytokines and chemokines to drive inflammation and are referred to as “classically activated” or “killer”
macrophages (M1). During the resolution phase of inflammation, they switch to what is known as an
“alternatively activated” phenotype or “healer” macrophage (M2) and contribute to debris scavenging,
angiogenesis, and wound healing. M1 macrophages are activated by treatment with IFNg or LPS and M2
macrophages are activated by treatment with Th2 cytokines IL-4 or IL-13 and the M2 phenotype switch
can be enhanced by IL-10. Macrophages can also be skewed during differentiation in vitro, and the resultant
phenotype depends upon the cytokine provided to support their differentiation. In murine macrophages,
MCSF promotes differentiation to an M1 phenotype, GM-CSF promotes differentiation to an M2 pheno-
type and IL-3 promotes differentiation into a profoundly M2 skewed phenotype. A defining feature of the
phenotype of murine M1 versus M2 macrophages is how they metabolize L-arginine. In response to an
inflammatory stimulus like LPS, M1 macrophages produce inducible nitric oxide synthase (iNOS) which
uses L-arginine as a substrate to produce nitric oxide (NO). M2 macrophages constitutively produce the
enzyme arginase I (argI), which sequesters L-arginine from iNOS and results in the production of
ornithine and downstream polyamines and L-proline. M1 macrophages also produce relatively higher
levels of pro-inflammatory IL-12 and lower levels of anti-infl ammatory IL-10 relative to M2 mac-
rophages. In this chapter, we describe in vitro derivation of polarized bone marrow macrophages and
methods to analyze the resulting phenotype including Q-PCR, Western blotting, and enzyme assays to
determine argI and iNOS expression and activity, as well as production of IL-12p40 and IL-10 and
determination of IL-12/IL-10 ratios. Production of iNOS, NO, IL-12p40, and IL-10 are measured after
treatment with LPS.
Keywords: Macrophage phenotype, Macrophage polarization, Alternative activation, Bone marrow

derived macrophages, Macrophage colony stimulating factor, Granulocyte macrophage colony stimu-
lating factor, IL-3, IL-4, Inducible nitric oxide synthesis, Nitric oxide, Arginase I, IL-12p40, IL-10
225
226 S.B. Weisser et al.
1. Introduction
Macrophages are critical players in all aspects of the immune

response to foreign pathogens and tumor cells. Resident tissue mac-
rophages are poised to respond to infection or injury and initiate an
inflammatory response to danger or pathogen associated molecular
patterns. Within 24 h of insult, monocytes are recruited from the
circulation and move into a site of injury or infection where they
mature into classically activated macrophages, also called killer or
M1 macrophages (1). These M1 macrophages amplify the innate
immune response producing cytokines and chemokines. They can
also present antigen to initiate the acquired immune response.
When the inflammatory insult has been dealt with, macrophages
remain at the scene and are converted by the local cytokine milieu
to participate in the resolution phase of inflammation. These mac-
rophages participate in debris scavenging, angiogenesis, tissue
remodeling, and wound healing (2). These macrophages are called
alternatively activated, healer, or M2 macrophages (3).
M1 and M2 macrophages represent extremes of macrophage
polarization. Increasingly, we recognize that macrophages are het-
erogeneous both in their phenotype and functional responses to
inflammatory stimuli (4). In complex systems, this may be due to
multiple, simultaneous stimuli acting on individual cells, interme-
diate or transitional phenotypes, as well as the effects of popula-
tions of macrophages. Despite this, it is still critical to define
features of polarized macrophages to enable comparison and cate-
gorization of macrophage phenotype and function to better under-
stand their role in normal and pathophysiologies. M1 macrophages
are activated by IFNg or LPS and produce robust amounts of reac-
tive oxygen species and pro-inflammatory cytokines (IL-12, TNFa,
IL-23) and chemokines and murine M1 macrophages upregulate
inducbile nitric oxide synthase (iNOS) to produce the reactive
nitrogen species, nitric oxide (NO) (1). The canonical M2 mac-
rophages, also referred to as M2a macrophages, are activated by
IL-4 or IL-13 and their activation can be enhanced by co-treat-
ment with IL-10. In response to inflammatory stimuli, these mac-
rophages produce lower amounts of pro-inflammatory cytokines
and higher amounts of anti-inflammatory IL-10 relative to their
M1 counterparts. Additionally, murine M2 macrophages up-regu-
late expression of arginase I (argI), Ym1 (a mammalian chitinase),
and FIZZ1 (also known as RELMa) (5).
A critical switch that defines murine macrophage activation
and polarization is the way in which the cells metabolize L-arginine
(6, 7). Murine M1 macrophages metabolize L-arginine by iNOS
to produce NO. NO is a reactive nitrogen intermediate that can
damage DNA, thereby killing foreign microorganisms or tumor
cells and also causes host tissue damage. M2 macrophages
14 Murine Alternatively Activated Macrophages 227
metabolize L-arginine via argI to produce L-ornithine. L-ornithine

is a precursor for putrescine, spermidine, and spermine produc-
tion, which promote cell proliferation and tissue repair and for
L-proline, which is an essential component of collagen biosynthe-
sis required for tissue repair. In addition, when both argI and iNOS
are expressed, argI sequesters L-arginine from iNOS acting as a
cell intrinsic inhibitor of NO production. Finally, argI induction
has been shown in one system to block transcription of iNOS in
response to inflammatory stimuli providing another level of
negative regulation of pro-inflammatory NO production (8).
It is important to understand the forces driving macrophage
phenotype and the characteristics that define the resultant pheno-
type so that we can better understand the role of macrophages in
normal physiology and in disease. Aberrant macrophage pheno-
type contributes to autoimmune and auto-inflammatory diseases as
well as solid tumor growth, via tumor associated macrophages
which share features with M2 macrophages (9, 10). The potential
to drive select features of macrophage phenotype could provide
novel targets for intervention in these pathologies as well as in
treatment of infectious diseases. Techniques provided here will
allow evaluation of the impact of unique experimental systems on
deriving and characterizing polarized macrophages. Polarization of
mature macrophages to an M2 phenotype has been described pre-
viously (11) and genetic models of polarized macrophage pheno-
type have also been described (12–14). Herein, we describe
approaches to generate bone marrow derived macrophages com-
paring the impact of different cytokines available to the cells in vivo
during differentiation that result in differentially skewed argI/NO
metabolism and cytokine production profiles (15–17).
2. Materials
2.1. Tissue Culture 1. C57BL/6 mouse (8–12 weeks old). Animals housed and
sacrificed according to institutional requirements.
2. Iscove’s Modified Dulbecco’s Medium (IMDM).
3. Fetal bovine serum.
4. Recombinant murine macrophage colony stimulating factor
(MCSF or CSF-1).
5. Recombinant murine granulocyte-macrophage colony stimu-
lating factor (GM-CSF).
6. Recombinant murine interleukin-3 (IL-3).
7. Recombinant murine interleukin-4 (IL-4).
8. Monothioglycerol (MTG).
9. Culture medium: IMDM, 10% FCS, 150 mM MTG. Combine

and filter-sterilize using low protein binding 0.2 micron filter.
Store at 4°C for up to 1 month.
10. Cell Dissociation Buffer (Gibco-BRL, Bethesda, MA).
11. Lipopolysaccharide from Escherichia coli serotype O127:B8 (LPS).
12. Incubator set at 37°C and 5% CO2.
2.2. Quantitative PCR 1. RNAse-free 1.5 ml eppendorf tubes.

2. Sterile and RNAse-free filter tips.
3. TRIzol® (Invitrogen).
4. Isopropanol.
5. 75% (v/v) ethanol in diethylpyrocarbonate (DEPC)-treated
water.
6. Oligo p(dT)20-40 (5 U per 750 ml of DEPC-treated water).
7. dNTPs mix (10 mM each in DEPC-treated water).
8. MMLV reverse transcriptase.
9. MMLV reverse transcriptase reaction buffer.
10. iQ SYBR green Supermix Q-PCR master mix (2×) (Bio-Rad
Laboratories, Hercules, CA).
11. Primers:
argI forward: 5¢-TTGCGAGACGTAGACCCTGG-3¢
argI reverse: 5¢-CAAAGCTCAGGTGAATCGGC-3¢
iNOS forward: 5¢-GCCACCAACAATGGCAACA-3¢
iNOS reverse: 5¢-CGTACCGGATGAGCTGTGAATT-3¢
GUS forward: 5¢-ACGTTAGCCGGGCTGCACTC-3¢
GUS reverse: 5¢-TCGGTTTGCGGTCGCGAGTG-3¢
12. NanoDrop Spectrophotometer (Thermo Scientific, Ottawa,
ON, Canada).
2.3. SDS-PAGE 1. Laemmli’s digestion mix (LDM): 75 mM Tris–HCl pH 6.8,

7.5% (w/v) glycerol, 200 mM b-mercaptoethanol, 1.5% (w/v)
bromophenol blue.
2. Separating gel mix (4×): 1.5 M Tris–Cl pH 8.8, 0.4% SDS.
3. Stacking gel mix (4×): 0.3 M Tris–HCl pH 6.8, 0.4% SDS.
4. 40% Acrylamide/bisacrylamide solution (37.5:1 with 2.6% C).
5. N,N,N¢,N¢-Tetramethyl-ethylenediamine (TEMED).
6. 10% (w/v) Ammonium persulfate (APS). Freeze aliquots at
−20°C for up to 3 months, thawing an aliquot for use and stor-
ing at 4°C for no more than 7 days.
7. PageRuler prestained protein ladder (Fermentas Life Sciences,
Thermofisher).
8. Running buffer (10×): 0.25 M Trizma base, 1.9 M glycine, 1%

(w/v) sodium dodecyl sulfate (SDS).
9. Glass plates (Fischer Scientific Co., Pittsburgh, PA).
10. Bio-Rad Protean II xi Cell (Bio-Rad Laboratories Inc, Hercules, CA).
2.4. Western Blotting 1. Transfer buffer (10×): 0.25 M Trizma base, 1.92 M glycine,
0.5% (w/v) SDS.
2. Methanol.
3. Immobilon-P membrane (0.45 mm pore polyvinyldifluoride
(PVDF)) (Bio-Rad Laboratories, Hercules, CA).
4. Whatman filter paper.
5. Tris-buffered saline with Tween-20 (TBST): 137 mM NaCl,
2.7 mM KCl, 25 mM Tris–Cl pH 7.4, 0.1% (v/v) Tween-20.
6. Blocking buffer: 5% (w/v) bovine serum albumin fraction V
(BSA), 0.02% NaN3 in TBST.
7. Primary antibody buffer: 2% (w/v) BSA, 0.008% NaN3 in
TBST.
8. Primary antibodies: anti-arginase I (murine) (BD Biosciences,
Mississauga, ON, Canada), anti-Ym1 (rabbit) (STEMCELL
Technologies Inc, Vancouver, BC, Canada), GAPDH (murine)
(Fitzgerald Industries, Acton, MA, USA). Each primary anti-
body is used at 1 in 1,000 (v/v) in primary antibody buffer.
9. Secondary antibodies: anti-mouse horse-radish peroxidase (HRP),
anti-rabbit HRP. Secondary antibodies are used at 1 in 10,000
(v/v) in TBST (Bio-Rad Laboratories, Hercules, CA, USA).
10. Bio-Rad Trans-Blot Cell (Bio-Rad Laboratories Inc, Hercules,
CA, USA).
11. Enhanced chemiluminescent reagent “Western Lightning”
(PerkinElmer, Waltham, MA, USA).
12. Kodak X-OMAT blue film.
2.5. Arginase Assay 1. Locking eppendorf tubes.

2. Urea standard: 50 mM in ddH2O.
3. Arginase lysis buffer: 0.1% (v/v) Triton X-100 in 25 mM
Tris–Cl pH 8.0.
4. 10 mM MnCl2.
5. 0.5 M L-arginine in ddH2O, pH 9.7.
6. Acid mixture: 1:3:7 (v/v/v) H3PO4 (44.6 N):H2SO4
(36 N):ddH2O.
7. Colorimetric reagent: 9% (w/v) a-isonitrosopropiophenone
(aISPP) in absolute ethanol.
8. Bio-Rad protein quantification assay (Bio-Rad, Hercules,
CA, USA).
2.6. NO Assay 1. 100 mM NaNO2: 0.69 mg NaNO2 in IMDM, 10% (v/v) FBS.
2. Sulfanilamide (H2NC6SO2NH2) solution: 1% (w/v) in 2.5%
(v/v) phosphoric acid (H3PO4).
3. Naphthylethylenediamine dihyrochloride (C 10H 7NHCH 2
CH2NH2-2HCl-MeOH) solution: 0.1% (w/v) in 2.5% (v/v)
H3PO4.
2.7. Enzyme-Linked 1. ELISA kits for IL-12p40 and IL-10 (BD Biosciences).
Immunosorbent 2. Coating buffer: 0.2 M sodium phosphate pH 6.8.
Assays
3. Assay diluent: 10% (v/v) heat-inactivated FBS (56°C for
30 min) in Dulbecco’s PBS pH 7.4.
4. ELISA color detection substrate: reagent A and reagent B (BD
OptEIA TMB Substrate Reagent Set; BD Biosciences).
5. Stop solution: 0.2 N H2SO4.
3. Methods
3.1. Tissue Culture 1. Harvest bone marrow aspirates from femurs and tibias of an
8–12 week old C57BL/6 mouse using a 5 ml syringe and a 26
gauge needle to flush the marrow out with IMDM, 10% FCS.
2. Dilute bone marrow aspirates to 40 ml in IMDM, 10% FCS,
150 mM MTG and place cells in a 75 cm2 tissue culture flask to
adhere at 37°C, 5% CO2.
3. After 4 h, remove culture supernatant to a 50 ml conical Falcon
tube and spin down non-adherent cells (300 × g for 5 min).
Count nucleated cells (see Notes 1 and 2).
4. Resuspend cells at 0.5 × 106 cells/ml (i.e., about 160 ml) (see
Note 3) in bone marrow macrophage base medium (IMDM,
10% FCS, 150 mM MTG, no additional growth factors).
5. Divide equally into four 75 cm2 filter top tissue culture flasks
(about 40 ml per flask), and add 10 ng/ml of recombinant
growth factors. Add MCSF to 2 flasks, GM-CSF to 1 flask, and
IL-3 to the final flask (see Notes 4 and 5).
6. Replace medium at day 4, spinning down non-adherent cells
and returning them to the flask and at day 7, discarding non-
adherent cells.
7. At day 7, add IL-4 (10 ng/ml) to 1 of the flasks derived in MCSF
alone. Incubate cells for 3 more days (see Notes 6 and 7).
8. At day 10, adherent cells are lifted and replated for stimula-
tions and analyses. Cells are lifted off the tissue culture flask
using Cell Dissociation Buffer. Place 5 ml of buffer on cells for
2 min and then bang the side of the flask with the heel of your
palm firmly several times. Ensure that cells have lifted off of the
flask by examining the flask under the microscope. Remove
resuspended cells into a 15 ml conical Falcon tube and wash
the flask with an additional 10 ml of IMDM, 10% FBS. Pool
and spin down the cells at 300 × g for 5 min. Resuspend cells in
a small volume and count viable cells using a hemocytometer.
9. Cells can be harvested into the appropriate buffer for assays
(Q-PCR, SDS-PAGE, and Western blotting, or arginase) or
replated at a concentration of 0.5 × 106 cells/ml in IMDM,
10% FCS, 150 mM MTG + growth factor used for their deriva-
tion (MCSF, GMCSF, IL-3) or treatment (MCSF + IL-4) dur-
ing growth for stimulations (for NO assays or ELISAs).
10. To stimulate cells, replate in 6 wells (1 ml in a 6-well plate).
Add 10 ng/ml LPS to 3 wells and incubate at 37°C, 5% CO2
for 24 h.
11. Harvest cell supernatants to an eppendorf tube and remove con-
taminating cells by microfuging at 13,000 × g for 5 min. Divide
clarified supernatants into two fresh eppendorf tubes and store
at −20°C until ready to assay supernatants (see Note 8).
3.2. Quantitative PCR 1. For RNA isolation, solubilize 105 cells in 100 ml of TRIzol and
(see Note 9) incubate at 23°C for 5 min (see Note 10).
2. Add 20 ml of chloroform, caps tubes, and shake each sample
vigorously by hand and incubate at 23°C for 2 min.
3. Centrifuge at 12,000 × g for 15 min at 4°C and carefully remove
upper aqueous phase to a fresh tube (approximately 60 ml).
4. Add 30 ml of isopropanol and incubate at 23°C for 10 min.
5. Centrifuge at 12,000 × g for 15 min at 4°C and remove the
isopropanol. Wash one time with 100 ml of 75% ethanol by
vortexing and centrifuging at 12,000 × g for 15 min at 4°C.
Remove ethanol carefully but thoroughly and allow samples to
air-dry at 23°C for 5 min. Do not over dry or dry under vac-
uum because this will dramatically reduce the solubility of the
RNA pellet.
6. Resuspend RNA in 20 ml of DEPC-treated water by gently
pipetting up and down. If RNA is difficult to resuspend, heat
at 65°C for 5 min and pipet up and down. Quantitate RNA
using a NanoDrop Spectrophotometer (see Note 11).
7. For reverse transcription, combine 0.1 mg of RNA for each
sample with 1 ml of oligodT and increase volume to 12.5 ml
with DEPC-treated H2O. Incubate tubes at 65°C for 5 min
and plunge into ice (see Note 12).
8. Prepare a master mix for reverse transcription combining 2.5 ml
10× reaction buffer, 0.625 ml 10 mM (each) dNTPs, 0.5 ml
MMLV-RT, and 8.875 ml of DEPC-treated water per reaction.
Prepare enough master mix for the number of reactions

required +10% extra.
9. Add 12.5 ml of master mix to the side of each reaction tube.
Quick-spin to combine contents at the bottom of tube. Incubate
at 37°C for 2 min. Incubate at 40°C for 50 min. Incubate at
70°C for 15 min. Store cDNA at −20°C for use in Q-PCR.
10. For quantitative PCR (Q-PCR), dilute cDNA samples 1 in 4
and prepare three serial twofold dilutions for each sample (see
Note 13).
11. Prepare a Q-PCR master mix for each primer pair used. For
12.5 ml reactions, combine 0.1875 ml of forward primer
(20 mM), 0.1875 ml reverse primer (20 mM), 6.25 ml SYBR
green master mix, and 4.875 ml of ddH2O. Prepare enough
master mix to do all of the samples in triplicate, a blank sample
(no cDNA template) and an additional 10%.
12. Aliquot 11.5 ml of master mix into each well and add 1 ml of
template into the reaction mix at the bottom of each well
changing tips every time.
13. Cover the plate with the plate sealer (single use transparent
film specific to Q-PCR machine).
14. Q-PCR requires melting of the template and activation of the
polymerase followed by a two step reaction that alternates
between annealing/extension and melting so is programmed
as follows: 95°C for 10 min, 40 cycles of: 60°C for 30 s (anneal-
ing/extension), followed by 95°C melting for 15 s. Fold dif-
ferences in gene expression are determined using the software
accompanying the light cycler (SDS 2.1) and are compared
between samples relative to a housekeeping gene within the
sample, as illustrated in Fig. 1.
3.3. SDS-PAGE 1. For SDS-PAGE, 1 × 106 cells can be lysed in 200 ml of 1× LDM.
Shear DNA in samples by passing five times through a 26
gauge needle attached to a 1 ml tuberculin syringe (see Note
14). Boil 1 min. Store samples in the freezer at −20°C until
ready to load on SDS-PAGE.
2. SDS-PAGE instructions provided here are for preparation of a
40 ml separating gel (1.5 mm thick × 5 cm wide × 16 cm long)
to be used with the Bio-Rad Protean II xi Cell. Clean glass
plates thoroughly, rinse with water and then rinse with 95%
ethanol and air-dry immediately before use. Clean spacers and
combs with 95% ethanol, air-dry, and assemble the apparatus as
per manufacturers’ instructions (see Note 15).
3. To prepare a 10% separating gel, combine 20 ml distilled water,
10 ml 4× separating gel buffer and 10 ml acrylamide stock
solution (wearing gloves) in a 50 ml Falcon tube and mixing
by inversion. Degas for 2 min using a vacuum pump or 10 min
using a house vacuum (see Note 16).
Fig. 1. Q-PCR analysis of gene expression of argI and iNOS. Bone marrow aspirates were differentiated into macrophages
in the presence of 10 ng/ml of MCSF, GMCSF, or IL-3 for 10 days or in the presence of MCSF for 7 days followed by addition
of 10 ng/ml of IL-4 for an additional 3 days. After 10 days, macrophages were treated with 10 ng/ml of LPS for 24 h and
cells were harvested into TRIzol for RNA extraction. cDNA was prepared by reverse transcription and used for Q-PCR analy-
sis. Expression of argI and iNOS were evaluated for each sample relative to b-glucuronidase (GUS) as an internal control.
Data shown are means ± SD for three independent experiments performed in triplicate.
4. Add APS (80 mL) and 20 mL TEMED and mix gently but
thoroughly by rocking the Falcon tube to avoid introducing air
bubbles. Pour the entire 40 ml solution between the glass plates.
Using a Pasteur pipet, gently add 5 ml of H2O-saturated butanol
to overlay the top of the gel. Be careful not to cause mixing with
the denser gel solution. Allow gel to polymerize about 30 min.
5. Pour off the alcohol overlay. Rinse the gel top with distilled
water and drain water well (see Note 17).
6. To make a 4% stacking gel, combine 9.75 ml water, 3.75 ml 4×
stacking gel buffer, and 1.5 ml acrylamide stock solution. Add
75 ml APS and 15 ml TEMED in a 50 ml Falcon tube. Mix by
inversion and pipet onto the top of the separating gel. Place
the comb into the top of the gel. Avoid trapping air bubbles
below or on the side of the comb during insertion. Allow to
polymerize for 60 min before removing comb.
7. Using gel loading tips, add 100 ml of the sample (one-half) to
bottom of the wells.
8. Prepare 1.4 L of running buffer by diluting 140 ml of 10× run-
ning buffer stock solution to 1.4 L with dH2O. Gently add
running buffer to top up the wells with a Pasteur pipet and
then fill the upper buffer chamber with running buffer. Pour
the remaining running buffer into the bottom buffer reservoir
of the gel apparatus ensuring that it covers the bottom of the
gel and glass plates.
9. Fill the inner chamber of the gel apparatus with cold water and
run gel overnight (16 h) at 65 V.
3.4. Western Blotting 1. Instructions provided are for use with the Bio-Rad Trans-Blot
Cell. Cut one piece of PVDF membrane and two pieces of
Whatman filter paper to 5 cm × 16 cm.
2. Wet PVDF membrane in methanol and the hydrate the mem-
brane by adding 50 ml dH2O. Agitate at room temperature for
about 15 min until the water no longer beads or streaks off of

the membrane.
3. Disassemble gel apparatus, cut the stacking gel off and discard
and soak the separating gel in transfer buffer along with the
PVDF membrane.
4. To prepare transfer buffer, combine 400 ml of 10× transfer buffer
and 3.2 L dH2O. Finally, add 400 ml methanol (see Note 18).
5. Assemble the gel sandwich on the clear side of the transfer
tank holder wetting each piece generously in transfer buffer
as you assemble. The gel sandwich is assembled in the follow-
ing order: 1 Scotch-Brite pad, 1 piece of Whatman filter
paper, PVDF membrane, gel (from left to right, note the ori-
entation), 1 piece of filter paper, 1 Scotch-Brite pad. Firmly
roll out the gel sandwich with a 10 ml pipet applying down-
ward pressure to thoroughly remove air bubbles trapped
between the layers.
6. Secure the gel sandwich in its holding apparatus and move it
into the transfer tank. Fill the transfer tank with transfer buffer.
Run cold water through the transfer tank constantly during
transfer. Transfer gels for 4 h at 0.6 amps. Ensure that the buf-
fer tank does not overheat during transfer. If the transfer appa-
ratus feels too warm, place the entire assembly into a secondary
container and pack ice around it.
7. Remove and disassemble the gel sandwich. Peel the membrane
back from the gel, and place in a container suitable for prob-
ing. Mark the molecular weight markers on the membrane
with an indelible pen. Add 50 ml of blocking solution and
incubate for 2 h at 23°C on an orbital shaker.
8. Incubate the blocked membrane in primary antibody over-
night at 4°C on an orbital shaker (see Notes 19 and 20).
9. Wash the membrane 3 × 10 min in TBST at 23°C on an orbital
shaker.
10. Incubate with secondary antibody (anti-mouse-HRP for argI
and GAPDH; anti-rabbit-HRP for Ym1) for 45 min at 23°C
on an orbital shaker.
11. Wash the membrane 3 × 10 min in TBST at 23°C on an orbital
shaker.
12. For ECL detection, combine 7.5 ml of ECL reagent A and
7.5 ml of reagent B together and pipet onto membrane to
cover the entire surface. Gently agitate by hand for 1 min.
13. Drain excess fluid from the membrane and place it between two
layers of saran wrap and expose to film in a dark room. Exposure
times for these antibodies are very short, typically in the range
of 5–30 s for Ym1 and GAPDH and 30–60 s for arginase I.
M GM IL-3 IL-4
Ym1
ArgI
GAPDH
Fig. 2. Western blot analysis of M2 macrophage marker protein expression. Bone marrow
aspirates were differentiated into macrophages in the presence of 10 ng/ml of MCSF,
GMCSF, or IL-3 for 10 days or in the presence of MCSF for 7 days followed by addition of
10 ng/ml of IL-4 for an additional 3 days. Macrophage cell lysates were separated on a
10% SDS-PAGE, transferred onto PVDF and probed for M2 macrophage markers, Ym1 and
argI, as well as GAPDH, as a loading control.
Develop film in a film processor (see Note 21). An example of

the results produced by this technique is shown in Fig. 2.
3.5. Arginase Assay 1. Lyse 0.25 × 106 macrophages in 50 mL arginase lysis buffer.
2. Determine protein concentration in cell lysates using Bio-Rad
protein quantification assay
3. Pipet 5 mg of protein lysate into an eppendorf tube and top up
the volume to 100 mL with arginase lysis buffer.
4. Add 10 mL of 10 mM MnCl2 and incubate the samples at 55°C
in a water bath for 10 min.
5. Add 100 mL 0.5 M L-arginine into each sample and incubate
at 37°C for 1 h (see Note 22).
6. Add 800 mL acid mixture to each sample. Add 40 mL of ISPF
solution into each reaction and pipet to mix.
7. To prepare a standard curve, make twofold serial dilutions of
urea stock solution in dH2O using dH2O as a blank. Add
100 ml of each to an eppendorf tube. Add 400 mL acid solution
and then add 25 ml ISPF to each tube.
8. Boil samples and standards for 30 min in locking eppendorf
tubes. Let samples cool to room temperature 23°C in the dark
(10 min) (see Note 23).
9. Read absorbance at 550 nm within 30 min. Arginase activity
detected ± SD for three independent assays performed in trip-
licate is shown in Fig. 3a.
3.6. NO Assay 1. To prepare a standard curve, prepare twofold serial dilutions of

NaNO2 stock in IMDM, 10% FBS using IMDM, 10% FBS as
a blank.
2. Pipet 50 mL of standard, blank or clarified culture superna-
tant into a flat bottom polystyrene non-tissue-culture-treated
96-well plate.
3. Add 50 mL of sulfanilamide solution into each well.

4. Add 50 mL of naphthylethylenediamine dihyrochloride solu-
tion into each well.
5. Incubate the plates for 10 min in the dark. Read the absorbance
at 550 nm within 30 min. Nitrite detected ± SD for three inde-
pendent assays performed in triplicate is shown in Fig. 3b.
3.7. Enzyme-Linked 1. ELISA kits for IL-12p40 and IL-10 were purchased from
Immunosorbent BD Biosciences and assays were performed as per manufac-
Assays turer’s instructions. Cytokine production ± SD from four
independent experiments assayed in duplicate are shown in
Fig. 4.
Fig. 3. Analysis of enzymatic activity of argI and iNOS. Bone marrow aspirates were differentiated into macrophages in the
presence of 10 ng/ml of MCSF, GMCSF, or IL-3 for 10 days or in the presence of MCSF for 7 days followed by addition of
10 ng/ml of IL-4 for an additional 3 days. After 10 days, macrophages were harvested into arginase lysis buffer for arginase
activity assays or left untreated or treated with 10 ng/ml of LPS for 24 h. After LPS treatment, cell supernatants were
harvested, clarified and subjected to the Griess assay to measure NO2−, downstream of NO production. Data shown are
means ± SD for three independent experiments performed in triplicate.
Fig. 4. Measurement of pro-inflammatory (IL-12 p40) and anti-inflammatory (IL-10) cytokine production and IL-12/IL-10
ratios. Bone marrow aspirates were differentiated into macrophages in the presence of 10 ng/ml of MCSF, GMCSF, or IL-3
for 10 days or in the presence of MCSF for 7 days followed by addition of 10 ng/ml of IL-4 for an additional 3 days.
Resulting macrophages were treated with LPS for 24 h and cell supernatants were harvested, clarified and assayed by
ELISA for IL-12p40 and IL-10 and the ratio of IL-12/IL-10 produced in response to LPS was calculated. Data shown are
means ± SD for four independent experiments assayed in duplicate.
4. Notes
1. Nucleated cell counts can be performed by diluting cell

suspensions 1 in 20 in 3% acetic acid. This procedure lyses all
cells including red blood cells and the remaining nuclei can be
counted on a hemocytometer.
2. A thorough bone marrow flush results from two femurs and
two tibias gives up to 80 × 106 cells and approximately 90% of
cells remain in suspension after 4 h of adherence depletion.
The number of cells that become adherent does vary in some
genetically modified animals and so this is a very important
step if macrophages from genetically modified mice are being
compared to wild type mice.
3. It is much easier to resuspend cell pellets in a small volume of
medium (5 ml) and then to dilute it up to a larger volume.
This ensures a homogeneous cell suspension.
4. We have assayed several different sources of conditioned media
and the amount of growth factor that they provide varies
dramatically between source and batch. Because the procedure
described here aims to compare alternative activation strategies
based on growth factors used during differentiation, we rec-
ommend that recombinant sources of growth factor are used
to obtain similar results.
5. Cell concentration during derivation is important because
macrophage skewing during differentiation requires cell intrin-
sic and cell extrinsic factors. Cell extrinsic factors are affected
by cell concentration.
6. Macrophages derived in this way are consistently more than 95%
positive for Mac-1 and F4/80 by flow cytometric analysis.
7. Recombinant IL-13 also skews MCSF derived macrophages to
an alternatively activate phenotype, although it is less potent
than IL-4 when compared directly. IL-10 (10 ng/ml) enhances
IL-4 or IL-13 induced alternative activation of macrophages,
but it does not mediate skewing macrophages to an M2a phe-
notype on its own.
8. Cell supernatants harvested for ELISAs should be stored in ali-
quots because some cytokines are sensitive to freeze–thaw cycles.
9. RNA is extremely sensitive to degradation by ubiquitous
RNAses. All sample handling must be done with gloves, all
plasticware used should be RNAse free, and all water should be
treated with DEPC.
10. TRIzol containing samples should be handled in a fume hood.
11. An A260/A280 ratio of 1.6–2.0 reflects pure and well-solubilized

RNA and an A260 of 1.0 abs unit = 40 mg/ml of RNA. Store
RNA for up to 6 months at −80°C.
12. Plunging RNA into ice after melting prevents formation of sec-
ondary structures that can interfere with reverse transcription.
13. Q-PCR should always be performed in triplicate for each sam-
ple and analysis of transcripts is done relative to an unaffected
control gene (b-glucuronidase or GUS), which should also be
performed in triplicate for each sample.
14. If cell suspensions are too viscous, a larger bore needle can be
use to begin to shear the DNA and then decreased until the
sample passes easily through a 26 gauge needle.
15. Before pouring your running gel mix into the SDS-PAG appa-
ratus, fill the assembled gel apparatus with dH2O to ensure
that it is not leaking.
16. Use a trap between the solution being degassed and the vac-
uum assembly to avoid contamination with acrylamide, which
is a neurotoxin.
17. Do not leave the alcohol on top of the gel for too long, as it
can cause the gel to dehydrate.
18. Methanol will cause the salts to precipitate out of the 10×
transfer buffer so should be added last to the pre-diluted trans-
fer buffer. However, if this is done in the wrong order, simply
add a stir bar and place the slurry onto a stir plate and the pre-
cipitate will go back into solution.
19. Antibodies can be “multiplexed” if you are confident that each
antibody does not have a cross-reactive band that will affect
detection by the other antibodies. Another way to multiplex
detection is to cut your membrane horizontally ensuring that
you do not cut through a band of interest. For the detection
described here, we routinely cut our membrane horizontally
between the 55 and 40 kDa molecular weight markers and probe
the upper half of the membrane with anti-Ym1 (Mw 55 kDa)
and the lower half of the membrane with anti-argI (Mw 36.5 and
38 kDa dimer) and anti-GAPDH (Mw 35 kDa) simultaneously.
20. Primary antibodies incubation can be at room temperature for
2 h, but our best experience to minimize background is to
incubate overnight (16 h) at 4°C.
21. Alternatively activated macrophages are larger than classically
activated macrophages so when comparing the same number
of macrophages, the protein loading (GAPDH) will increase.
For that reason, we routinely harvest enough cells to run our
gels twice, the first time comparing equal cell numbers and for
a second run, we will adjust our loading according to the load-
ing control to load equal amounts of protein. Our lab does not
typically assay for protein prior to loading because some of the

additional proteins that we are interested in are extremely sen-
sitive to degradation upon cell lysis and we avoid that problem
by resuspending immediately in LDM, shearing and boiling
our samples.
22. Increasing this incubation time up to 2 h can increase the sen-
sitivity of this assay if arginase activity is low.
23. ISPF will form a precipitate in the reaction mixture. Read
absorbance of clear supernatants.
References
1. Martinez FO, Sica A, Mantovani A, Locati M 11. Ho VW, Sly LM (2009) Derivation and charac-
(2008) Macrophage activation and polarization. terization of murine alternatively activated (M2)
Front Biosci 13:453–461 macrophages. Methods Mol Biol 531:173–185
2. Adamson R (2009) Role of macrophages in 12. Brombacher F, Arendse B, Peterson R, Holscher
normal wound healing: an overview. J Wound A, Holscher C (2009) Analyzing classical and
Care 18:349–351 alternative macrophage activation in mac-
3. Gordon S (2003) Alternative activation of rophage/neutrophil-specific IL-4 receptor-alpha-
macrophages. Nat Rev Immunol 3:23–35 deficient mice. Methods Mol Biol 531:225–252
4. Gordon S (2007) Macrophage heterogeneity 13. Rauh MJ, Ho V, Pereira C, Sham A, Sly LM,
and tissue lipids. J Clin Invest 117:89–93 Lam V, Huxham L, Minchinton AI, Mui A,
Krystal G (2005) SHIP represses the genera-
5. Nair MG, Gallagher IJ, Taylor MD, Loke P,
tion of alternatively activated macrophages.
Coulson PS, Wilson RA, Maizels RM, Allen JE
Immunity 23:361–374
(2005) Chitinase and Fizz family members are a
generalized feature of nematode infection with 14. Sinha P, Clements VK, Ostrand-Rosenberg S
selective upregulation of Ym1 and Fizz1 by anti- (2005) Reduction of myeloid-derived suppres-
gen-presenting cells. Infect Immun 73:385–394 sor cells and induction of M1 macrophages
facilitate the rejection of established metastatic
6. Munder M (2009) Arginase: an emerging key disease. J Immunol 174:636–645
player in the mammalian immune system. Br J
Pharmacol 158:638–651 15. Kuroda E, Ho V, Ruschmann J, Antignano F,
Hamilton M, Rauh MJ, Antov A, Flavell RA,
7. Yeramian A, Martin L, Arpa L, Bertran J, Soler Sly LM, Krystal G (2009) SHIP represses the
C, McLeod C, Modolell M, Palacin M, generation of IL-3-induced M2 macrophages
Lloberas J, Celada A (2006) Macrophages by inhibiting IL-4 production from basophils.
require distinct arginine catabolism and trans- J Immunol 183:3652–3660
port systems for proliferation and for activa-
tion. Eur J Immunol 36:1516–1526 16. Fleetwood AJ, Lawrence T, Hamilton JA, Cook
AD (2007) Granulocyte-macrophage colony-
8. Lee J, Ryu H, Ferrante RJ, Morris SM Jr, stimulating factor (CSF) and macrophage CSF-
Ratan RR (2003) Translational control of dependent macrophage phenotypes display
inducible nitric oxide synthase expression by differences in cytokine profiles and transcription
arginine can explain the arginine paradox. Proc factor activities: implications for CSF blockade
Natl Acad Sci USA 100:4843–4848 in inflammation. J Immunol 178:5245–5252
9. Heinsbroek SE, Gordon S (2009) The role of 17. Kuroda E, Noguchi J, Doi T, Uematsu S, Akira
macrophages in inflammatory bowel diseases. S, Yamashita U (2007) IL-3 is an important
Expert Rev Mol Med 11:e14 differentiation factor for the development of
10. Mantovani A, Allavena P, Sica A, Balkwill F prostaglandin E2-producing macrophages
(2008) Cancer-related inflammation. Nature between C57BL/6 and BALB/c mice. Eur J
454:436–444 Immunol 37:2185–2195
Chapter 15
Human Long-Term Culture Initiating Cell Assay

Min Liu, Cindy L. Miller, and Connie J. Eaves
Abstract
The long-term culture initiating cell (LTC-IC) assay, founded on the bone marrow long-term culture
(LTC) system, measures primitive hematopoietic stem cells (termed LTC-IC) based on their capacity to
produce myeloid progeny for at least 5 weeks. Adaptations of the LTC system including the use of stromal
cell lines, application of limiting dilution analysis, and estimation of average hematopoietic progenitor
output per LTC-IC under defined conditions have made it possible to accurately determine LTC-IC con-
tent in minimally separated and highly purified cell populations from human hematopoietic tissue sources
such as bone marrow, peripheral blood, cord blood, fetal liver as well as cord blood and mobilized periph-
eral blood. Methodologies for measuring human LTC-IC using bulk cultures, limiting dilution analysis,
and single cell cultures are described.
Key words: Long-term initiating cell assay, LTC-IC, Stromal cells, Hematopoiesis
1. Introduction
Hematopoiesis is a continuum process of blood cell production in

all lineages, which lasts the lifespan of the individual. Hematopoietic
stem cells (HSC) are defined as having the ability to generate
daughter cells that are still competent to produce all blood cell
lineages (self-renewal) and/or to differentiate irreversibly generat-
ing progeny in all hematopoietic lineages (pluripotency).
The first concept that a normal adult has a population of primi-
tive hematopoietic cells that have the ability to produce multiple
types of blood cell progeny and gradually become more lineage
restricted as they divide was inferred from the observation in 1951
that patients with myeloproliferative disorders have an increased
number of precursor cells of all lineages but only one type of mature
blood cells is elevated in the circulation (1). Since then, several
241
242 M. Liu et al.
assays have been developed to detect and quantify different types

of primitive hematopoietic cells, precursors, and lineage restricted
progenitors based on their displayed physiological growth and dif-
ferentiation properties when they are stimulated in vitro or in vivo.
Surrogate assays using various in vitro systems and in vivo xeno-
transplant models have been developed. However, the animal facilities
required to conduct in vivo assays of human cells in xenograft sys-
tems is not easily accessible and not yet routine for many laborato-
ries; reproducible in vitro assays have been widely adopted for this
purpose. In vitro systems have some advantages including the abil-
ity to manipulate culture conditions and evaluate cellular responses to
different extrinsic factors. It is also possible to adapt in vitro assay
systems—to not only allow the detection and quantification of
primitive hematopoietic cells but also selectively distinguish cells at
discrete stages of early hematopoietic cell differentiation.
Over three decades ago, Dexter and colleagues first discovered
that adherent layers generated from murine bone marrow cells
supports the continuous generation of granulocytes and mac-
rophages for many weeks under horse serum and corticosteroids-
containing culture conditions without the addition of exogenous
growth factors (2–5). Later studies showed that serum is required
to establish a competent feeder layer of stromal cells and to sup-
port the generation and proliferation of primitive hematopoietic
cells (6). Corticosteroids such as hydrocortisone also appear to be
important and need to be added during the duration of the long-
term culture (4, 7). These findings have led to development of
long-term culture (LTC) systems, in which a continuous genera-
tion of myeloid as well as lymphoid cells from primitive hematopoi-
etic cells in vitro is allowed (2, 3, 7–11) and cells thus identified
have been called LTC-initiating cells (LTC-IC).
The original LTC-IC assays for human cells are initiated with
irradiated primary bone marrow adherent feeder layers followed by
the addition of testing cell populations. Several studies have dem-
onstrated that pre-adipocyte fibroblast cells as well as stromal cell
lines, such as M2-10B4 (12), MS-5 (13–15), AFT204 (16, 17),
and S17 (18, 19), are also capable of supporting human hematopoi-
etic progenitors in LTC system. Generation of genetically engi-
neered stromal cell lines and modification of existing culture system
have led to the development of derivative LTC-IC assays for the
quantitation of primitive cells with myeloid and/or limited lym-
phoid potential. Several studies have reported that LTC-ICs that
have myeloid and limited (NK- and/or B-lineage) differentiation
potential (20–23) can be detected in cocultures that contain certain
stromal cell feeders. Similarly, cells with T-cell potential also been
detected in other types of stromal cocultures (24). Finally, murine
fibroblasts genetically engineered to produce human growth
factors such as interleukin-3 (IL-3), granulocyte-macrophage
15 Human LTC-IC Assay 243
colony-stimulating factor (GM-CSF), and steel factor (SF, or also

known as stem cell factor [SCF]) have been shown to greatly
enhance the sensitivity of LTC-IC detection for the long-term cul-
ture system (25). However, an in vitro system that allows the co-
generation of a single human cell with full myeloid and lymphoid
differentiation potential (myeloid, T-, B-, and NK-cells) has not
been previously described and identification of such multi-lineage
LTC-ICs is therefore precluded.
To carry out a LTC-IC assay, minimally manipulated cells,
density-gradient separated cells, or purified primitive hematopoi-
etic cells are first seeded onto irradiated marrow feeders, or a
stromal cell line with equivalent supportive activity. Committed
progenitors initially present in the testing cell population will rap-
idly mature and disappear during the initial 3–5 weeks of culture
due to their limited proliferative potential. On the contrary, more
primitive cells such as LTC-IC will be maintained throughout the
duration of culture and generate a new cohort of committed pro-
genitors (e.g., colony-forming cells, CFC), which can be later
detected and enumerated at the end of the assay. The number of
LTC-IC present in the initial test cell suspension can then be cal-
culated by dividing the total number of CFC detected in the cul-
ture by the average number of CFC produced per LTC-IC for the
standard conditions used. Alternatively, the number and frequency
of LTC-IC in a given test cell population can be determined using
limiting dilution analysis (LDA) based on Poisson statistics and the
method of maximum likelihood (26) if the number of CFC pro-
duced per LTC-IC is unknown for a given cell sample. This strat-
egy is based on the fact that the frequency of LTC-IC is the
reciprocal of the concentration of test cells that yields 37% negative
cultures, meaning no CFC is detected at the end of assay, by the
application of methodologies described above. This value can also
be used to derive the average CFC output per LTC-IC in that
population. However, this analysis is only valid when the require-
ment for a linear relationship between CFC output and number of
cells seeded is met. It is important to emphasize that the prolifera-
tive potential exhibited by individual LTC-IC and the mean CFC
generated per LTC-IC can vary significantly in a given cell popula-
tion and according to the ontological source of cell population
tested (25, 27–30). The type of feeder layer used and other culture
conditions also greatly affect the outcomes of the assay (14, 25,
31). Therefore, it is advisable that investigator should proceed with
caution and use LDA for LTC-IC quantitation when average
CFC/LTC-IC has not yet been predetermined for a given cell
population under validated culture conditions.
In this chapter, we describe methodologies for quantitation of
human LTC-IC using M2-10B4, or engineered M2-10B4 and Sl/
Sl stromal cell lines by bulk LTC, LDA, and single cell assay.
244 M. Liu et al.
2.. Materials
2.1. Chemicals 1. Hydrocortisone 21-hemisuccinate sodium salt: (e.g.,

and Media STEMCELL Technologies, Vancouver, BC or Sigma, St Louis,
MO) Powder should be stored at −20°C. Hydrocortisone (HC)
solution can be made in culture media such as Alpha Minimal
Essential Medium (Alpha MEM) to a final concentration of
10−4 M. The solution should be sterilized using 0.22 μm syringe
filter and is stable at 4°C for a week. Fresh hydrocortisone solu-
tion should be prepared weekly due to a relative short half-life.
2. G418 sulfate: This antibiotic is used to select neomycine resis-
tance gene transfected M2-10B4 and Sl/Sl-engineered cell
lines. Powder or ready-to-use sterile solution can be purchased
from qualified vendors and should be stored at 4°C. Handle
according to manufacturer’s instructions.
3. Hygromycin B: This antibiotic is used to select hygromycin
resistance gene transfected M2-10B4 and Sl/Sl-engineered
cell lines. Powder or ready-to-use sterile solution can be pur-
chased from qualified vendors and should be stored at 4°C.
Handle according to manufacturer’s instructions.
4. Human long-term culture medium (hLTCM): (STEMCELL,
MyeloCult™ H5100, Catalog # 05100).The medium is stable
for 1 year when stored at −20°C and 1 month when stored
at 4°C.
5. hLTCM/10−6 HC. Prepare for use within 1–2 days by adding
1 mL of 10−4 hydrocortisone stock solution to 99 mL hLTCM.
Store at 4°C.
6. Fetal bovine serum (FBS) (e.g., STEMCELL). FBS should be
pretested for its ability to support the proliferation of cell
lines.
7. RPMI-1640/10% FBS: Prepare RPMI-1640 (e.g., Gibco
BRL, Hyclone, Sigma, and STEMCELL) with 10% of fetal
bovine serum (FBS) for culture of M2-10B4 cell lines and
store at 4°C for up to 1 month.
8. DMEM/10% FBS: Prepare Dulbecco’s modified Eagle’s
medium (DMEM) (e.g., Gibco BRL, Hyclone, Sigma, and
STEMCELL) with 10% FBS for culture of engineered Sl/Sl
cell lines and store at 4°C for up to 1 month.
9. IMDM/2% FBS: Prepare Iscove’s modified Dulbecco’s
medium (e.g., Gibco BRL, Hyclone, Sigma, and STEMCELL)
with 2% FBS. Store at 4°C for up to 1 month.
10. HBSS: Calcium and magnesium free Hank’s balanced salt
solution (e.g., Gibco BRL, Hyclone, Sigma, and STEMCELL).
It is advisable that 10 mM of HEPES (4-(2-hydroxyethyl)-1-
piperazineethanesulfonic acid) is supplemented to the

solution to avoid pH fluctuation when is used to handle cells
outside the CO2 incubator for a long period of time. The solu-
tion should be stored at 4°C.
11. PBS: Phosphate-buffered saline (pH 7.2–7.6) may be stored at
4°C or room temperature (RT).
12. Bovine Collagen solution 3 mg/mL (STEMCELL, Catalog #
04902). This solution is used to coat dishes or culture plates.
Store at 4°C.
13. CFC Medium—Methocult™GF+ H4435 (STEMCELL,
Catalog # 04435)
Methocult™GF+ H4435 methylcellulose-based medium contain-
ing recombinant cytokines (IL-3, IL-6, GM-CSF, G-CSF, SCF,
Erythropoietin) is used to detect colony-forming-unit-erythroid
(CFU-E), burst-forming unit-erythroid (BFU-E), CFU-
granulocyte, macrophage (CFU-GM) and CFU-granulocyte,
erythoroid, megakaryocyte, macrophage (CFU-GEMM) Store at
at −20°C and for up to 1 month at 2–8°C. Handle according to
manufacturer’s instructions (www.stemcell.com).
14. 0.25% trypsin-citrate or trypsin-ethylenediaminetetraaceticacid
(EDTA)(STEMCELL).
15. 3% acetic acid.
16. Trypan blue solution: (e.g., Gibco BRL, Hyclone, Sigma,
STEMCELL). This solution is used for trypan blue exclusion
assay to assess the number of viable cells present in samples.
2.2. Equipments/ 1. Incubator (set at 37°C, 5% CO2, >95% humidity).

Instruments/Software 2. Inverted microscope (with 10× or 12.5× eyepiece objective,
and 2×, 4×, and 10× planar objectives).
3. Standard light microscope for cell counting.
4. Pipette aid and micropipettors (single channel or multichannel).
5. Neubauer hemacytomer (for cell counting).
6. Sterile cultureware: T25 cm2 and T75 cm2 flasks; 35 and
60 mm tissue-culture-treated dishes; 96-well flat-bottom tis-
sue culture plates; U-bottom culture plates; polypropylene
tubes (5 mL 12 × 75 mm, 15 mL 17 × 100 mm, 15 and 50 mL
conical).
7. Syringes: 1 mL tuberculin syringe, 3 mL luer lock.
8. For CFC assays: 35 mm petri dishes (STEMCELL, Catalog #)
and 16-gauge blunt-end needles (STEMCELL, Catalog #
28110).
9. L-Calc™. L-Calc limiting dilution analysis software for calcu-
lating LTC-IC frequencies can be downloaded from
STEMCELL Web site (www.stemcell.com).
246 M. Liu et al.
2.3. Cell Lines 1. M2-10B4 murine fibroblast cell line can be obtained from
American Tissue Culture Company (ATCC: Catalog #
CRL1972).
2. Engineered M2-10B4 and Sl/Sl murine fibroblast cell lines
(25) can be accessed by contacting STEMCELL Technologies
(www.stemcell.com).
3. Methods
Perform all procedures in a certified Level II biosafety cabinet with

personal protective equipment. Use sterile techniques at all times
and handle all materials with appropriate biohazard caution.
3.1. Maintenance of M2-10B4 is maintained in RPMI 1640/10% FBS. Engineered

M2-10B4 and Sl/Sl M2-10B4 maintained in RPMI 1640/10% FBS should be supple-
Fibroblast Cell Lines mented with 0.4 mg/mL G418 and 0.06 mg/mL Hygromycin B
every second passage. Engineered Sl/Sl maintained in DMEM/10%
FBS should be supplemented with 0.8 mg/mL G418 and 0.15 mg/
mL Hygromycin B every second passage (see Notes 1 and 2) Cultures
should be subcultured when they are near confluence (~90–100%)
1. Harvest cell lines by decanting or suctioning off medium, and
rinse once with 2–5 mL PBS or HBSS to remove remaining
media and loosely adherent cells.
2. Remove PBS or HBSS. Add 0.25% trypsin solution (2 mL for
25 cm2 flask and 5 mL for 75 cm2 flasks). Incubate for 2–10 min
at 37°C until adherent cells start to detach from the surface of
the flask.
3. Add 0.2 mL (for 25 cm2 flask) or 0.5 mL (for 75 cm2 flask)
FBS to neutralize trypsin and mix with pipette to obtain a sin-
gle-cell suspension.
4. Wash cells twice with culture media (M2-10B4: RPMI
1640/10% FBS and Sl/Sl: DMEM/10% FBS).
5. Cells are seeded at 1:50 to 1:100 ratio into a new flask (10 mL
in 25 cm2 flask and 20 mL in 75 cm2 flask) and cultured in
humidified incubator set at 37°C and 5% CO2.
3.2. Collagen Coating 1. Add 1 mL or 2 mL of collagen solution per 35 mm or 60 mm

of Tissue Culture tissue culture dish. Multi-well culture plates can also be coated
Dishes for Bulk LTC-IC with collagen.
Assay 2. Ensure that solution is spread uniformly over the surface of the
dish and the remove excess collagen solution. This collagen
solution can then be reused to coat remaining dishes.
3. Allow the dishes to air-dry at RT for a minimum of 1 h within
a biosafety cabinet with the tissue culture lids partially or com-
pletely removed.
4. Rinse coated tissue culture dishes once with sterile PBS or

HLTM to neutralize the acidity prior to use.
5. Dishes can be used immediately or tightly wrapped coated
dishes may be stored at 4°C for up to 2 weeks.
3.3. Preparation of 1. Obtain fibroblast cell lines by trypsinization from flasks as

Irradiated Feeder described in Subheading 3.1.
Layers 2. Wash cells twice with culture media (RPMI 1640/10% FBS or
DMEM/10% FBS) and perform a nucleated cell counts using
trypan blue exclusion assay (see Note 3).
3. Resuspend cells at a concentration of 106 to 108 cells/mL in
HLTM/10−6 M HC and irradiate cells with 8,000 cGy from an
X-ray or gamma-irradiation source (e.g., 37Cesium). A prelimi-
nary experiment should be done to confirm this irradiation
dose allows feeder cells to support LTC-IC but is sufficient to
inhibit cell proliferation.
4. Dilute irradiated M2-10B4 to 1.5 × 105/mL or combine irra-
diated engineered cell lines in a 1:1 ratio to achieve 1.5 × 105/
mL (7.5 × 104/mL M2-10B4 and 7.5 × 104/mL Sl/Sl)
5. Seed the appropriate number of irradiated M2-10B4, or engi-
neered M2-10B4 and Sl/Sl cells in freshly prepared
HLTM/10−6 M HC in collagen-coated tissue culture dishes or
in multi-well flat-bottom culture plates for limiting dilution or
single cell analysis (see Note 4).
6. Maintain the cultures at 37°C in a humidified incubator (>95%)
with 5% CO2
7. Test cells may be added immediately to the irradiated cells
lines. However, it is recommended that irradiated feeder cul-
tures be incubated overnight before use. Irradiated feeder cul-
tures can be incubated for up to 10 days before addition of test
cells (medium change with freshly prepared HLTM/10−6 M
HC should be done on day 7)
3.4. Preparation of Human LTC-IC are present in various hematopoietic sources

Test Cell Suspensions such as bone marrow, mobilized peripheral blood, cord blood,
and in fetal liver. The LTC-IC frequency unprocessed cell suspen-
sions is usually too low for accurate quantitation and enrichment
protocols are recommended (see Note 5). Enrichment techniques
include:
● Light density fractionation (e.g., Ficoll-Pague, GE Healthcare)
● Isolation of CD34+ cells and lineage (Lin−) depleted cells (e.g.,
EasySep™, RosetteSep™ STEMCELL)
● Fluorescence activated cell sorting (FACS) (e.g., CD34+,
CD34+CD38−, ALDHhiCD133+Lin−, etc.)
248 M. Liu et al.
Table 1
Recommended cell number for initiation of LTC-IC assay on
M2-10B4 or engineered M210-B4 and Sl/Sl feeder layers
Cell population Cells per LTC-IC assaya,b
Ammonium chloride-treated bone marrow 4 × 105 to 2 × 106

Light-density bone marrow 2 × 105 to 1 × 106
Light-density cord blood 1 × 105 to 5 × 105
Lin−CD34+c 1,000 to 5,000
Lin−CD34+CD38−c 100 to 500
a
The number is based on assay carried out on a 35 mm culture dish
b
The number of LTC-IC present in different ontological sources can vary significantly;
therefore, it is advisable to initiate assay at two or more different cell doses
c
For cells collected from bone marrow, G-CSF mobilized peripheral blood, fetal liver
Table 2
Number of CFC generated per LTC-IC from various
ontological sources
Cell population Average number of CFC generated per LTC-ICa
Adult bone marrowb 18 ± 6

G-CSF mobilized blood 25 ± 5
Cord blood 28 ± 2
Fetal liver 72 ± 18
a
The average number of CFC per LTC-IC present here is based on 6-week long-term
culture assay carried out using a mixture of M2-10B4 and Sl/Sl engineered feeder layers
(25, 28)
b
The average number of CFC per LTC-IC per human bone marrow LTC-IC cultured
for 5 weeks on irradiated primary marrow or M2-10B4 cells is 7 ± 3 (25)
3.5. LTC-IC Bulk Assay LTC-IC numbers can be determined using “bulk LTC” carried
out in 35-mm tissue-culture-treated dishes if the average number
of CFC per LTC-IC is known. The cultures should be initiated
with at least 10–20 LTC-IC to ensure statistically representative
sampling (see Subheading 1, Tables 1 and 2). It is recommended
that replicate cultures be performed at two or more doses for each
test sample.
1. Prepare test cells at the appropriate concentration (see Table 1)
in 2 mL HLTCM/10−6 M HC.
2. Carefully remove medium from dishes containing irradiated
feeder cells and replace with HLTCM/10−6 M HC containing
appropriate number of test cells without disturbing adherent

feeder layer.
3. Place the culture dishes in a 100 mm petri dish with an addi-
tional uncovered 35 mm dish containing 3 mL sterile distilled
water.
4. Place cultures at 37°C in humidified incubator (>95%) with 5%
CO2 in air for 5–6 weeks.
5. For weekly culture maintenance, gently rotate dish to mix con-
tents and remove one half of the HLTCM/10−6 M HC (~1 mL)
and cells and replace with freshly prepared HLTCM/10−6 M
HC at weekly intervals. Do not disturb the adherent feeder
layer.
6. It is advisable to examine the culture periodically using an
inverted microscope to assess hematopoiesis and to detect any
contamination.
3.6. LTC-IC Limiting LTC-IC LDA are carried out in 96-well flat-bottom culture plates.
Dilution Analysis The LTC-IC mini-cultures (wells) are initiated with various doses
of test cells in several replicates. Set up the assay with 8–24 repli-
cates of three to four cell doses that bracket the expected LTC-IC
frequency of a given hematopoietic cell population.
1. Prepare test cells at the appropriate concentrations in
HLTCM/10−6 M HC.
2. Carefully remove ~90% of HLTCM/10−6 M HC from each
well containing irradiated feeder cells (leaving ~10 μL medium)
taking care not to allow the wells to dry out or disturb the
adherent feeder layer.
3. Add test cells in 0.1 mL HLTCM/10−6 M HC using multi-
channel pipettor with sterile tips and without disturbing the
adherent feeder layer.
4. Incubate culture plates at 37°C in humidified incubator (>95%)
with 5% CO2 for 5–6 weeks. 96-well plates should be placed in
suitable containers with proper gas exchange and open dishes
containing sterile water to reduce evaporation.
5. Cultures are maintained with scheduled weekly one half media
exchanges. Perform media change by removing one half of the
medium (~50 μL per well) and non-adherent cells from each
dish and replacing with HLTCM/10−6 M HC. A multichannel
pipettor can be used to for media changes to manipulate 3–6
wells at a time. To avoid contamination, do not touch tips on
the exterior of the wells and use new sterile tips each time cells
and/or medium are removed from wells. When removing and
replacing media to each well, care must be taken to avoid dis-
turbing the adherent layer
250 M. Liu et al.
3.7. LTC-IC: Single Cell Single cell assays are generally performed when using highly
Analysis purified cell populations (e.g., LTC-IC frequency ³1 in 10).
Preliminary studies to estimate LTC-IC frequency using LDA are
recommended. Reader is also encouraged to consult published lit-
erature to identify suitable purification strategies.
1. Perform enrichment strategies to increase the LTC-IC content
in the test cell population.
2. Aseptically sort single cells into 96-well U-bottom plate con-
taining 0.1 mL of HLTCM/10−6 M HC by FACS. The medium
should be pre-filtered to remove any particulates which may
interfere with visual confirmation of a single cell within indi-
vidual wells.
3. Incubate culture plates with singly sorted cells at 37°C in
humidified incubator (>95%) with 5% CO2 for 2 h to allow
cells to settle down to the bottom of the well.
4. Visually check each well using inverted light microscope to
confirm the presence of a single cell.
5. Carefully remove medium from each well containing irradiated
feeder cells (see Subheading 3.6) and be careful not to allow
the wells to dry (leave ~10 μL of medium in each well).
6. Transfer singly sorted cells in HLTCM/10−6 M HC to the irra-
diated feeder cells. 96-well plates should be placed in suitable
containers with proper gas exchange and open dishes contain-
ing sterile water to reduce evaporation.
7. Incubate culture plates at 37°C in humidified incubator (>95%)
with 5% CO2 for 5–6 weeks.
8. Perform weekly half media exchanges as described in
Subheading 3.6.
3.8. Harvest and Following 5 or 6 weeks of culture, LTC-IC cultures are harvested
LTC-IC Analysis (both adherent and non-adherent cells), and the LTC-IC derived
clonogenic progenitors (CFCs) are quantified.
3.8.1. Bulk LTC-IC Harvest 1. Remove HLTCM/HC medium and non-adherent cells from
the culture dish and transfer into a labeled sterile 17 × 100 mm
(i.e., 14 mL) tube.
2. Rinse the culture dish twice with 1 mL HBSS to remove loosely
attached cells and remaining serum-containing medium.
Transfer to the harvest tube.
3. Add 1.0 mL of Trypsin-Citrate or Trypsin-EDTA solution and
place dishes at 37°C incubator for 3–5 min (up to a maximum
of 10 min). At intervals, swirl culture gently and examine using
an inverted microscope for evidence of detachment of the
adherent layer.
4. Add 0.2 mL FBS to neutralize the trypsin.

5. Using sterile pipette, rinse the Trypsin-Citrate or Trypsin-
EDTA solution over the surface of the dish several times to
ensure that all adherent cells are detached and to make a single
cell suspension.
6. Add all cells and medium to the harvest tube.
7. Rinse the culture dish twice with ~1.5 mL IMDM/2% FBS.
8. Transfer all cells and medium to the harvest tube. Fill tube
with IMDM/2% FBS and centrifuge for 7–10 min at 300 × g.
9. Carefully decant or suction off supernatants without disturb-
ing cell pellets and wash cells twice in IMDM/2%.
10. Resuspend cells in approximately 1 mL of IMDM/2%. Record
volume and perform a nucleated cell count with 3% acetic acid.
To exclude dead cells, dye exclusion stain such as trypan blue
can be used (see Note 6).
11. Dilute cells to the appropriate concentration in IMDM/2%.
CFC numbers are determined using MethoCult™GF+ H4435
(see Subheading 3.8.2 and manufacturer’s instructions). It is
not necessary to assess the entire culture contents after the har-
vest; (a) Plate a predetermined proportion of the bulk LTC-IC
culture according to the number of CFC expected (b) Plate at
two to three different cell concentrations (i.e., 2–5 × 104 per
CFC culture). It is recommended that two to four CFC repli-
cates be done for each condition (see Note 7).
3.8.2. Colony-Forming Cell 1. Dilute harvested cells to 2–5 × 105 cells/mL in IMDM/2%
Assay FBS. Mix well.
2. Add 0.3 mL of cell suspension to 3 mL of MethoCult™ GF+
H4435 (for duplicate assays,) or 0.5 mL of cells to 5 mL of
H4435 (for quadruplicate assays) and vortex. Let mixture
stand for 5 min to allow bubbles to rise (see Note 8).
3. Plate 1.1 mL per 35 mm petri dish each using 3 cc syringe
attached to 16 gauge blunt-end needle (see Note 9).
4. Rotate gently to spread methylcellulose medium over the sur-
face of the dish.
5. Place two plated dishes within a 100 mm petri dish containing
a third uncovered 35 mm dish with 3 mL sterile water.
6. Incubate methylcellulose cultures for 16–18 days at 37°C in
humidified incubator (>95%) with 5% CO2.
7. On days 16–18, enumerate the total number of colonies per
dish (see Note 10).
8. The number of LTC-IC present in the initial test cell suspen-
sion is calculated by dividing the total number of CFC detected
in the culture by the average number of clonogenic progenitors
252 M. Liu et al.
produced per LTC-IC for the standard conditions used (see

Table 2). Alternatively values can be expressed as LTC-IC
derived CFC per number of input test cells.
3.8.3. Harvest LDA and Wells should be harvested in groups of 8–12 wells to avoid excess
Single Cell LTC-IC Assays trypsinization and drying out of wells. A multichannel pipette can
be used to manipulate 3–4 wells at a time.
1. At the end of the 5- or 6-week culture period, remove
HLTCM/HC and non-adherent cells from each well (~0.1 mL)
and place into individual labeled 12 × 75 mm sterile tubes (see
Note 11).
2. Rinse each well once with 0.1 mL HBSS to remove remaining
loosely adherent cells and transfer to corresponding tube.
3. Add 0.1 mL 0.25% Trypsin-Citrate or Trypsin-EDTA to each
well and incubate at 37°C for 3–5 min.
4. Monitor plate using inverted microscope to confirm detach-
ment of adherent cell layer. Incubation with trypsin solution
should continue until adherent cells loosen. Trypsin solution
should not be left in wells for periods longer than approxi-
mately 10 min.
5. Add 10 μL of FBS to neutralize trypsin and pipette up and
down gently to obtain a single cell suspension.
6. Transfer all cells and medium to the appropriate tube.
7. Rinse each well once with IMDM/2% FBS and transfer to
tube.
8. Fill tube with IMDM/2% FBS and centrifuge at 300 × g for
7–10 min.
9. Carefully decant or suction off supernatant, leaving ~0.1 mL
medium. Vortex to resuspend cells.
10. Add 1 mL of MethoCult™ GF+ H4435 to each tube and vor-
tex. Leave mixture for 5 min to allow bubbles to rise.
11. Draw up the entire contents of each tube using 1 mL syringe
without needle attached. Dispense into labeled 35 mm petri
dish avoiding excess bubbles. Rotate dish to ensure that meth-
ylcellulose-based medium is spread evenly.
12. Place two plated dishes into a 100 mm petri dish and include a
third 35 mm dish with 3 mL sterile water (without a lid on).
13. Incubate cultures at 37°C in humidified incubator (>95%) with
5% CO2 for 16–18 days.
14. Enumerate colonies. A well is scored as positive if one or more
colonies (e.g., BFU-E, CFU-GM or CFU-GEMM) are detected.
A well is scored as negative if no colonies are present.
15. The LTC-IC frequency in the test cell population is calculated

using Poisson statistics and the method of maximum likelihood
(26). The interpolation of the frequency is based on the pro-
portion of negative wells (no CFC present) detected in an assay
(see Note 12).
16. Appropriate statistical analysis can be used to determine the
number of CFC per LTC-IC using standard conditions of cul-
ture (27).
3.8.4. Single Cell LTC-IC The LTC-IC frequency in the test cell population is calculated by
dividing positive wells by total wells tested.
4. Notes
1. Avoid keeping cell lines in culture for extended periods without

returning to frozen stock. Generally, all cell lines should be
replaced after ~6 months. Therefore, it is important to estab-
lish a large number of vials with early-passage cells.
2. It is recommended that engineered M2-10B4 and Sl/Sl be
checked periodically to ensure that cells are still producing
expected levels of growth factors, using assays such as cytokine
ELISA available from R&D Systems.
3. Trypan blue exclusion assay is commonly used to enumerate
viable cells in a given cell sample. Cell samples should be mixed
with trypan blue solution at 1/5 to 1/100. Only non-viable
cells will stain with the trypan blue dye and viable cells will
remain unstained. It is recommended that samples should be
examined within 10 min after mixing with trypan blue solution
to avoid cell toxicity effects and result in an inaccurate count.
4. 3 × 105 total cells in 2.0 mL of HLTCM/10−6 HC is used to
prepare irradiated feeder cell layers when 35-mm culture dishes
are used for bulk LTC-IC or 1.5 × 104 total cells in 100 μL of
HLTCM/10−6 HC when 96-well flat-bottom tissue culture
plates are used to perform limiting dilution or single cells
LTC-IC.
5. T and B cells depletion is highly recommended if normal
peripheral blood is the source of test cells due to the possible
outgrowth of Epstein-Barr virus transformants or obscuration
of the CFC assays (32). For human fetal liver, macrophage
precursors must first be removed by depletion of linage posi-
tive (lin+) cells. The LTC-IC content of bone marrow can be
assayed in samples following lysis of red blood cells by ammo-
nium chloride treatment without prior light density cell separa-
tion (32, 33) Methods to obtain cell suspensions enriched for
254 M. Liu et al.
LTC-IC content have been described and the reader is referred

to the literature (12, 27, 33–36)
6. Generally, hematopoietic cells have a round, uniform, and
retractile appearance, compared to the larger murine fibroblast
feeder cells (M2-10B4 and engineered M2-10B4 and Sl/Sl)
when stained with trypan blue. In addition, hematopoietic
cells appear to have more compact nuclei when stained using
3% acetic acid. It is recommended a “murine fibroblast cell
only” control is included in each experiment until the operator
is proficient at distinguishing the differences between the mor-
phology of hematopoietic and fibroblast feeder cells.
7. To obtain accurate quantitation of CFCs for analysis of human
LTC-IC, it is necessary that a linear relationship exists between
the input cell dose and the number of colonies generated at the
end of assay. Overplatting may cause inhibition of progenitor
proliferation and underplatting may lead to statistically inac-
curacy. Therefore sufficient cells should be seeded to yield
~25–150 colonies per 1.1 mL culture in 35 mm dishes.
8. If the investigator chooses to prepare their own media for CFC
assay, batches of fetal bovine serum in combination with other
components should be prescreened and tested for proper sup-
port of CFC growth. The methylcellulose-based medium can
be prepared with final concentrations of 1.0% methylcellulose,
30% fetal bovine serum, 1% bovine serum albumin, 10−4 M
2-mercaptoethanol, 20 ng/mL each of recombinant human
IL-3, GM-CSF, IL-6, G-CSF and 50 ng/mL of SCF, and
3 U/mL EPO.
9. 35 mm culture dish used for CFC assay may be purchased from
certified vendor (e.g., STEMCELL Technologies: Catalog #
27100). These 35 mm dishes are pretested for optimal colony
growth without supporting adherent cells in methylcellulose-
based assays. 16 gauge blunt end needles are used to manipu-
late the viscous methylcellulose-based medium and to prevent
needle-stick injuries.
10. Human CFC:CFU-E (Colony forming unit-erythroid): These
are clonogenic erythroid progenitors that produce only one or
two clusters of hemoglobinized erythroblasts. BFU-E (Burst
forming unit-erythroid): These are primitive erythroid progen-
itors which generate colonies with 3–8 clusters (small), 9–16
clusters (intermediate), >16 clusters (large) of hemoglobinized
erythroblasts. CFU-GM (Colony forming unit-granulocyte,
macrophage): Progenitors that give rise to colonies containing
macrophages and granulocytes. CFU-GEMM (Colony forming
unit-granulocyte, erythrocyte, macrophage, megakaryocyte):
Multi-lineage progenitors that generate colony with granulo-
cyte, erythroid, macrophage, and megakaryocyte lineages.
11. Use a pipettor with sterile tips and harvest 12–16 wells at a
time. Care must be taken to avoid contamination if a multi-
channel pipettor is used, and it is highly recommended to use
a test tube rack allowing uncapped 12 × 75 mm tubes to set
close together. One should avoid touching tips on the exterior
of tubes and new tips must be used for each well.
12. The frequency of LTC-IC present in testing cell population
can be determined statistically by limiting dilution analysis
using L-Calc™ software (STEMCELL: Catalog # 28600). For
free software download, please visit http://www.STEMCELL.
com.
References
1. Dameshek W (1951) Some speculations on 11. Miller JS, Verfaillie C, McGlave P (1992) The
the myeloproliferative syndromes. Blood generation of human natural killer cells from
6(4):372–375 CD34+/DR− primitive progenitors in long-
2. Dexter TM, Allen TD, Lajtha LG (1977) term bone marrow culture. Blood
Conditions controlling the proliferation of 80(9):2182–2187
haemopoietic stem cells in vitro. J Cell Physiol 12. Sutherland HJ et al (1991) Differential regula-
91(3):335–344 tion of primitive human hematopoietic cells in
3. Dexter TM et al (1980) The role of cells and long-term cultures maintained on genetically
their products in the regulation of in vitro engineered murine stromal cells. Blood
stem cell proliferation and granulocyte 78(3):666–672
development. J Supramol Struct 13(4): 13. Issaad C et al (1993) A murine stromal cell line
513–524 allows the proliferation of very primitive
4. Greenberger JS (1978) Sensitivity of corti- human CD34++/CD38− progenitor cells in
costeroid-dependent insulin-resistant lipo- long-term cultures and semisolid assays. Blood
genesis in marrow preadipocytes of 81(11):2916–2924
obese-diabetic (db/db) mice. Nature 14. Croisille L et al (1994) Hydrocortisone differ-
275(5682):752–754 entially affects the ability of murine stromal
5. Greenberg HM et al (1981) Human granulo- cells and human marrow-derived adherent cells
cytes generated in continuous bone marrow to promote the differentiation of CD34++/
culture are physiologically normal. Blood CD38− long-term culture-initiating cells.
58(4):724–732 Blood 84(12):4116–4124
6. Gronthos S, Simmons PJ (1995) The growth 15. Hao QL et al (1995) A functional comparison
factor requirements of STRO-1-positive of CD34+ CD38− cells in cord blood and
human bone marrow stromal precursors under bone marrow. Blood 86(10):3745–3753
serum-deprived conditions in vitro. Blood 16. Thiemann FT et al (1998) The murine stromal
85(4):929–940 cell line AFT024 acts specifically on human
7. Gartner S, Kaplan HS (1980) Long-term cul- CD34+CD38− progenitors to maintain primi-
ture of human bone marrow cells. Proc Natl tive function and immunophenotype in vitro.
Acad Sci USA 77(8):4756–4759 Exp Hematol 26(7):612–619
8. Rawlings DJ et al (1995) Long-term culture 17. Wineman J et al (1996) Functional heterogene-
system for selective growth of human B-cell ity of the hematopoietic microenvironment: rare
progenitors. Proc Natl Acad Sci USA stromal elements maintain long-term repopulat-
92(5):1570–1574 ing stem cells. Blood 87(10):4082–4090
9. Whitlock CA, Witte ON (1982) Long-term 18. Wineman JP, Nishikawa S, Muller-Sieburg CE
culture of B lymphocytes and their precursors (1993) Maintenance of high levels of pluripo-
from murine bone marrow. Proc Natl Acad Sci tent hematopoietic stem cells in vitro: effect of
USA 79(11):3608–3612 stromal cells and c-kit. Blood 81(2):365–372
10. van den Brink MR et al (1990) The generation 19. Collins LS, Dorshkind K (1987) A stromal cell
of natural killer (NK) cells from NK precursor line from myeloid long-term bone marrow
cells in rat long-term bone marrow cultures. cultures can support myelopoiesis and B lym-
J Exp Med 172(1):303–313 phopoiesis. J Immunol 138(4):1082–1087
256 M. Liu et al.
20. Miller JS et al (1999) Single adult human 29. Prosper F, Stroncek D, Verfaillie CM (1996)
CD34(+)/Lin−/CD38(−) progenitors give rise Phenotypic and functional characterization of
to natural killer cells, B-lineage cells, dendritic long-term culture-initiating cells present in
cells, and myeloid cells. Blood 93(1):96–106 peripheral blood progenitor collections of nor-
21. Berardi AC et al (1997) Individual mal donors treated with granulocyte colony-
CD34+CD38lowCD19−CD10− progenitor stimulating factor. Blood 88(6):2033–2042
cells from human cord blood generate B lym- 30. Roy V, Miller JS, Verfaillie CM (1997)
phocytes and granulocytes. Blood 89(10): Phenotypic and functional characterization of
3554–3564 committed and primitive myeloid and lym-
22. Hao QL et al (1998) In vitro identification phoid hematopoietic precursors in human fetal
of single CD34+CD3− cells with both lym- liver. Exp Hematol 25(5):387–394
phoid and myeloid potential. Blood 31. Punzel M et al (1999) The type of stromal
91(11):4145–4151 feeder used in limiting dilution assays influences
23. Punzel M et al (1999) The myeloid-lymphoid frequency and maintenance assessment of
initiating cell (ML-IC) assay assesses the fate of human long-term culture initiating cells.
multipotent human progenitors in vitro. Blood Leukemia 13(1):92–97
93(11):3750–3756 32. Heather JS, Eaves AC, Eaves CJ (1991)
24. Holmes R, Zuniga-Pflucker JC (2009) The Quantitative assays for human hematopoietic
OP9-DL1 system: generation of T-lympho- progenitor cells. In: Gee AP (ed) Bone mar-
cytes from embryonic or hematopoietic stem row processing and purging: A practical guide.
cells in vitro. Cold Spring Harb Protoc CRC Press Inc, Boca Raton, FL, pp 155–171
2009(2):pdb.prot5156 33. Sutherland HJ et al (1989) Characterization
25. Hogge DE et al (1996) Enhanced detection, and partial purification of human marrow cells
maintenance, and differentiation of primitive capable of initiating long-term hematopoiesis
human hematopoietic cells in cultures contain- in vitro. Blood 74(5):1563–1570
ing murine fibroblasts engineered to produce 34. Lansdorp PM, Dragowska W (1992) Long-
human steel factor, interleukin-3, and granulo- term erythropoiesis from constant numbers of
cyte colony-stimulating factor. Blood CD34+ cells in serum-free cultures initiated
88(10):3765–3773 with highly purified progenitor cells from
26. Fazekas de St G (1982) The evaluation of lim- human bone marrow. J Exp Med
iting dilution assays. J Immunol Methods 175(6):1501–1509
49(2):R11–R23 35. Sauvageau G et al (1994) Differential expres-
27. Sutherland HJ et al (1990) Functional charac- sion of homeobox genes in functionally dis-
terization of individual human hematopoietic tinct CD34+ subpopulations of human bone
stem cells cultured at limiting dilution on sup- marrow cells. Proc Natl Acad Sci USA
portive marrow stromal layers. Proc Natl Acad 91(25):12223–12227
Sci USA 87(9):3584–3588 36. Petzer AL et al (1996) Self-renewal of primi-
28. Nicolini FE et al (1999) Unique differentia- tive human hematopoietic cells (long-term-
tion programs of human fetal liver stem cells culture-initiating cells) in vitro and their
shown both in vitro and in vivo in NOD/ expansion in defined medium. Proc Natl Acad
SCID mice. Blood 94(8):2686–2695 Sci USA 93(4):1470–1474
Chapter 16
Long-Term Culture-Initiating Cell Assay for Mouse Cells

Stefan Woehrer, Cindy L. Miller, and Connie J. Eaves
Abstract
The long-term culture-initiating cell (LTC-IC) assay is a well-established in vitro assay used to enumerate
primitive mouse hematopoietic stem cells (HSCs) and relies on the two cardinal functions of HSCs: ability
to self-renew and differentiation capacity. LTC-ICs present in minimally processed and purified cell sus-
pensions and cocultured on a supportive feeder layer are detected by their sustained ability to produce
hematopoietic progenitors (colony forming cells) after ³ 4 weeks in culture. Refinements including the use
of a defined stromal cell line, and extending the in vitro culture to 6 weeks allow detection of LTC-IC at
similar frequencies to transplantable HSCs quantified using in vivo assays.
Key words: Hematopoietic stem cells, Stromal cells, Stem cell niche, LTC-IC, Long-term culture-
initiating cell
1. Introduction
The long-term culture-initiating cell (LTC-IC) assay is an in vitro

test system used to detect and enumerate primitive hematopoietic
stem cells (HSCs) termed LTC-IC. The underlying principles of
this assay is to mimic the HSC niche with bone marrow stromal
cells or stromal cell lines that support the survival, self-renewal, and
differentiation of primitive HSCs (1–5). Basically, this test consists
of two steps: The first step is to coculture test cells on a supportive
feeder layer in a limiting dilution assay for 4–6 weeks to allow the
differentiation of less primitive hematopoietic cells (present in the
input cell suspension) while maintaining or expanding LTC-IC
numbers. The second step is to detect LTC-IC-derived myeloid
hematopoietic progenitors using the colony forming cell (CFC)
assay in methylcellulose-based medium. The frequency of LTC-IC
is determined using Poisson statistics and method of maximum
likelihood (6–9). Additional information is given in Chapter 15.
257
258 S. Woehrer et al.
Fig. 1. Frequency of long-term culture-initiating cells (LTC-ICs) derived from single cells with durable (CD45posEPCRpos
CD48neg CD150pos, solid line) and limited (CD45posEPCRpos CD48neg CD150neg, dashed line) self-renewal capacity after differ-
ent periods of long-term culture (LTC). Data are means +/− SE. EPCR endothelial protein c receptor.
Since it is deemed that only primitive HSCs have the potential

to give rise to LTC-ICs, the number of LTC-ICs directly correlates
with the number of HSCs that are initially present in the test cell
population. If the cells are cultured for shorter periods of time, some
progenitor cells and HSCs with limited self-renewal may read out as
LTC-ICs. Recent advances in discriminating HSCs with durable
self-renewal (DSR: CD45posEPCRposCD48negCD150pos) from those
with limited self-renewal (LSR: CD45posEPCRposCD48negCD150neg),
made it possible to test those very closely related cell populations
separately in the LTC-IC assay. Figure 1 illustrates that a 4-week
culture period is not long enough to discriminate those two cell
populations with statistical confidence. In contrast, a culture period
of 6 weeks not only significantly separates HSCs with DSR from
those with LSR but also reflects almost exactly the competitive
repopulating unit (CRU) frequencies derived from mouse trans-
plantation experiments (9).
Compared to the in vivo CRU assay, the LTC-IC assay has the
advantage of not having the constraints of histocompatibility
requirements and cell homing issues. Besides these biological
advantages, the LTC-IC assay is comparatively fast, inexpensive,
and many samples can be assessed at the same time. One of the
major disadvantages is that lymphoid cells are not readily detected
and assay modifications are required to detect primitive cells with
both myeloid and lymphoid potential (LTC-ICML) (7). The CRU
assay is therefore still considered the gold standard to quantify and
characterize HSCs.
Two variations of the LTC-IC are described in this chapter.
The first is the classical LTC-IC that is based on the long-term
culture with primary bone marrow feeder cells. The second is a
modified LTC-IC that uses a fetal liver stromal cell line AFT024 to
16 Mouse LTC-IC 259
support the LTC-IC survival (10). This cell line has been shown to
specifically support the survival of HSCs and hence LTC-ICs.
AFT024 are commercially available from ATCC, easy to culture,
yields consistent results, and therefore simplifies the LTC-IC
assay.
2. Materials
2.1. Sources of Feeder

1. Mouse bone marrow. Animals are housed and sacrificed using
Layers and Test Cells
protocols approved by the host institution. Use a minimum of
two C57BL/6 mice, 6–12 weeks old. Other laboratory mouse
strains may be used as mitotically inactivated feeder cells but
preliminary experiments are required to confirm their suitabil-
ity (see Note 1).
2. AFT024 fetal liver stromal cell line (ATCC® SCRC-1007,
http://www.atcc.org) cultured in DMEM/10% FBS (see
Subheading 3.2) and incubated at 33°C, 5% CO2, and >95%
humidity.
3. Test Cells: LTC-IC can be measured in hematopoietic cell
samples including unseparated and purified bone marrow, and
purified fetal liver (i.e., see Table 1).
2.2. Cell Culture Media 1. DMEM/10% FBS. Dulbecco’s modified essential medium
and Reagents (STEMCELL catalog # 36250) with 10% fetal bovine serum
(STEMCELL, catalog # 06500) and 5 × 10−5 M 2-mercapto-
ethanol (Sigma-Aldrich catalog # M7522).
2. Hydrocortisone 21-hemisuccinate sodium salt (HC)
(STEMCELL, Sigma-Aldrich). Store powder desiccated at
−20°C. Dissolve hydrocortisone powder in α-MEM
(STEMCELL, Cat# 36450) to a final concentration of 10−4 M,
Table 1
Frequencies of LTC-IC in mouse hematopoietic cell
populations
Cell populations LTC-IC (%) References
Adult BM 0.002–0.005 (1, 6, 7)

Lin−Sca-1+WGA+BM ~2 (7)
− +
Lin Sca-1 day 14.5 fetal liver ~2 (8)
pos pos neg pos
CD45 EPCR CD48 CD150 32–52 (9)
CD45posEPCRposCD48negCD150neg 3.5–8.5 (9)
filter-sterilize using a 0.22 μm filter, and store at 4°C. As

hydrocortisone has a relatively short half-life in solution, it
is necessary to prepare a fresh stock solution within 1 week
of use.
3. mLTCM/10−6 M HC: Mouse long-term culture medium with
10−6 M HC. mLTCM (MyeloCult™ M5300, STEMCELL,
Cat# 05300) is stored at −20°C for up to 1 year or at 4°C for
up to 1 month. Prepare for use within 1–2 days by adding
1 mL of 10−4 M hydrocortisone stock solution to 99 mL
MyeloCult™. The addition of antibiotics is not required if
aseptic techniques are used.
4. IMDM/2% FBS. Iscove’s MDM with 2% FBS. Aseptically add
10 mL FBS (STEMCELL, Cat# 06240) to 500 mL IMDM. If
required, filter-sterilize using 0.2 micron filter. Store at 4°C for
up to 2 weeks.
5. Hanks’ Balanced Salt Solution Modified (HBSS) with 10 mM
Hepes, without phenol red (STEMCELL Cat# 37150). Store
at 4°C.
6. Bovine collagen solution, 3 mg/mL (STEMCELL, Cat#
04902). Store at 4°C (see Note 2).
7. 0.25% trypsin-ethylenediaminetetraacetic acid (EDTA) solu-
tion (STEMCELL Cat# 07910). Store in aliquots at −20°C.
8. CFC Medium. (STEMCELL, MethoCult™ M3434, Cat#
03434). Methylcellulose-based medium containing recombi-
nant mouse IL-3, rhuman (rh) IL-6, rm stem cell factor, and
rh erythropoietin for mouse CFU-GM, BFU-E, and CFU-
GEMM assays. Store in aliquots at −20°C. See manufacturer’s
instructions for handling of MethoCult™ methylcellulose-
based medium (http://www.stemcell.com).
2.3. Equipment and 1. Incubator maintained at 33°C (and 37°C for irradiated AFT024
Cultureware cells), 5% CO2 and >95% humidity.
2. Sterile cultureware: T25 and T75 cm2 flasks (BD), 96-well flat-
bottom culture plates (i.e., Falcon, Costar), 100 mm petri
dishes.
3. Sterile pipettes, multichannel pipettes for accurate dispensing
of 10–100 μL and >100 μL volumes.
4. CFC assay materials: 1-mL syringe (BD Cat# 309602), 3-mL
luer lock syringes (BD, STEMCELL), 35 mm petri dishes
(STEMCELL, Cat# 27114/27116), and 16-gauge blunt-end
needles (STEMCELL Cat# 28110).
5. Limiting dilution software: L-Calc™ software program
(STEMCELL) free download at http://www.stemcell.com.
16 Mouse LTC-IC 261
3. Methods
3.1. Establishing
1. Sacrifice a minimum of two adult C57BL/6 mice, 6–12 weeks old,
Mouse Marrow Feeder
using protocols approved by the host institution.
Layers
2. Remove the femurs and tibias, cut off both ends of each bone
with sterile sharp surgical scissors, and flush the marrow plugs
into 1–2 mL IMDM/2% FBS using a sterile 21- or 22-gauge
needle attached to a 3-mL syringe. Prepare a single-cell suspen-
sion by drawing the media and cells up and down once or twice
using the same syringe and needle. HBSS or DMEM supple-
mented with 2% FBS are also suitable for the isolation of mar-
row cells. It is preferable to isolate cells at a high cell concentration
(³107 cells per mL), and do not wash cells prior to use.
3. Count the number of BM nucleated cells (i.e., using 3% acetic
acid and a Neubauer counting chamber). A yield of ~5 × 107 BM
cells is obtained per four long bones.
4. Dilute BM cells to 2 × 105/mL in mLTCM/10−6 M HC. Place
0.15 mL of the cell suspension into each well of a 96-well flat-
bottom tissue culture plate (3 × 104 cells per well) (see Note 1).
Use a multichannel pipettor and sterile tips to manipulate three
to six wells at a time. 96-well plates should be stored in a larger
covered container containing open 35 mm dishes with sterile
water to minimize culture dehydration.
5. Incubate cultures at 33°C in a humidified incubator with 5%
CO2 and 1 week later feed cultures by doing a ~50% medium
change. Remove 75 μL of non-adherent cells and medium
using a multichannel pipettor and sterile tips, taking care not
to disturb the adherent cell layer. Carefully add 80 μL of freshly
prepared mLTCM/10−6 M HC.
6. Incubate 96-well plates for total of 10–14 days or until the
adherent layer has reached ~90% confluency (with a ~50%
medium change on day 14 if required).
7. Irradiate the 96-well plates (without subculturing) with
1,500 cGy from a γ-irradiation or X-ray source. Incubate cul-
tures for minimum of 24 h prior to addition of test cells.
8. Irradiated mouse BM feeders can be used for up to 14 days,
but if delays of >7 days are anticipated, perform a ~50% medium
change after first 7 days.
3.2. Establishing 1. Harvest flask of AFT024 cells using 0.25 trypsin/EDTA solu-
AFT024 Feeder Layers tion. Completely remove DMEM/10% FBS medium by
decanting or suctioning. Add 2 mL HBSS and tilt gently to
detach loosely adherent cells, and discard.
2. Add 2 mL (T25) or 4 mL (T75) trypsin/EDTA solution.
Incubate for 3–5 min at 37°C or until AFT024 start to detach
from flask.
3. Add 0.2 mL FBS to neutralize trypsin, and mix with pipette to

break up cell clumps and obtain a single cell suspension.
4. Wash cells twice with DMEM/10% FBS and centrifuge at
250 × g for 7 min. Decant or suction medium and resuspend in
2–3 mL of mLTCM/10−6 M HC.
5. Count AFT024 and dilute to 3 × 104/mL in mLTCM/10−6 M
HC. Place 0.15 mL of the cell suspension into each well of a
96-well flat-bottom tissue culture plate (4.5 × 103 cells per well)
(see Note 2). Use a multichannel pipettor and sterile tips to
manipulate three to six wells at a time. 96-well plates should be
stored in a larger covered container containing open 35 mm
dishes with sterile water to minimize culture dehydration.
6. Incubate at 33°C in 5% CO2 for 3–5 days until the AFT024
has reached about 90% confluency.
7. Irradiate the 96-well plates (without subculturing) with
2,000 cGy from a γ-irradiation or X-ray source. Place irradi-
ated AFT024 feeders at 37°C in 5% CO2 humidified
incubator.
8. Incubate irradiated AFT024 feeders for a minimum of 24 h
prior to addition of test cells, or use within 7 days after
irradiation.
3.3. LTC-IC Experiment The frequencies of LTC-IC in unseparated bone marrow and
Design purified cell populations given in Table 1, and from published lit-
erature can be used to estimate appropriate test cell doses for the
LTC-IC limiting dilution assay. Eight to 24 replicates of three to
four cell doses that bracket the estimated LTC-IC frequency is
generally sufficient to provide data with reasonable 95% confidence
limits. For example, if the expected LTC-IC frequency in a purified
cell population is ~1/100 to 1/500, then aliquots of 30, 100, 300,
and 900 test cells per well with replicates of 8–12 wells per dose
should yield an accurate estimation of LTC-IC content. It is also
advisable to set up several replicate experiments compared one
large experiment.
3.4. Preparation Unseparated mouse BM test cells can be isolated using the proce-
of Mouse Test Cells dure described in Subheading 3.1, steps 1–3. Detailed protocols
for LTC-IC for isolating purified cell populations using techniques such as
fluorescent activated cell sorting (FACS) and immunomagnetic
cell separation are beyond the scope of this chapter, and readers
should refer to published literature. LTC-IC assay of unseparated
mouse day 12.5–16.5 fetal liver can result in an inhibitory out-
growth of macrophages. Therefore fetal liver cells require further
cell processing to remove mature cells (e.g., depletion of linage
positive cells (Lin+), Sca+ cell selection).
16 Mouse LTC-IC 263
3.5. Setup of Mouse 1. Prepare fresh mLTCM/10−6 M HC. Remove about 80% of
LTC-IC Assay medium from the wells of the 96-well plate using a multichan-
nel pipettor and sterile tips, taking care not to disturb feeder
layer (irradiated BM feeder or AFT024 cell line).
2. Prepare different dilutions of test cells in mLTCM/10−6 M HC.
3. Carefully add test cells in a 0.15 mL volume to the wells and
place the 96-well plate in a loosely covered container (i.e.,
20 cm square bacterial plates) with two to three uncovered
35-mm dishes containing sterile water. Incubate cultures at
33°C for BM feeders and 37°C for AFT024 feeder cells.
4. Perform weekly one-half media changes with freshly prepared
mLTCM/10−6 M HC. Remove 75 μL of non-adherent cells
and medium using a multichannel pipettor and sterile tips, tak-
ing care not to disturb the adherent cell layer. Carefully add
80 μL of mLTCM/10−6 M HC.
5. Incubate LTC-IC assays for 4 weeks (BM feeder cells) or
6 weeks (AFT024 feeder cells) (see Note 3).
3.6. Harvesting This section describes harvesting of LTC-IC cultures and plating
of LTC-IC Assay of the cells (adherent and non-adherent) from each individual well
into methylcellulose-based medium to detect LTC-IC-derived
mouse CFC (see Note 4).
1. Prepare all reagents, and label tubes. During the LTC-IC har-
vest, all media, reagents, non-adherent cells and adherent cells
from an individual well will be combined into a corresponding
tube. Arrange labeled 12 × 75 mm tubes in a rack that holds 72
tubes closely aligned (e.g., Nalgene, Thermo Fisher Scientific).
It is advisable to harvest the cells in batches of 24 or less to
ensure that wells do not dry out or become overexposed to trypsin.
A multichannel pipettor and sterile tips are used to manipulate
3 wells at a time.
2. Using a multichannel pipettor and sterile 200 μL tips to
manipulate 3 wells at a time, remove medium and non-adher-
ent cells and place in corresponding labeled tubes. Change
tips each time to avoid cross-contamination, and repeat for
12–24 wells.
3. Add 0.1 mL HBSS to each well (to dilute traces of mLTCM),
and then using new sterile tips add HBSS and loosely adherent
cells to the appropriate tube.
4. Add 0.1 mL 0.25 trypsin-EDTA solution to each well and
incubate at 37°C. After 3–5 min, scan the plate using an inverted
microscope to determine if the adherent layer has started to
detach from the surface of the well. If necessary, continue incu-
bating for up to 15 min until the adherent cells are loosened.
Add 10 μL of FBS to each well to neutralize the trypsin.
5. Mix the contents of each well gently to obtain a single-cell

suspension and add the suspended cells to the appropriate
tubes (containing medium, reagents, and non-adherent cell
from corresponding well).
6. Rinse each well once with HBSS/2% FBS and transfer again to
the appropriate tube.
7. Fill the tubes with HBSS/2% FBS and centrifuge at 350 × g for
7–10 min. Carefully decant or suction off the supernatants
without disturbing cell pellets, leaving about 0.1 mL media
behind.
8. Gently vortex the tube to resuspend the cells, add 1 mL mouse
MethoCult™ M3434 methylcellulose-based medium and vor-
tex vigorously.
9. Draw up the entire content of each tube individually using a
1-mL syringe (without needles and then eject into a 35-mm
dish. Rotate dishes to ensure methylcellulose is spread evenly
over surface of the dish. Place two dishes in a 100 mm petri
dish containing a third open 35-mm petri dish with sterile
water.
10. Incubate for 12 days and then record the number of colonies
present in each methylcellulose culture. Record the well as
negative if no CFC is present, and as positive if ³1 CFC (col-
ony of more than 30 cells) is present.
3.7. Analysis of Murine LTC-IC frequencies in the starting cell suspension are determined
LTC-IC by LDA by application of Poisson statistics and the method of maximum
likelihood assuming “single-hit kinetics,” (i.e., each LTC-IC will
produce ³1 CFC (detectable after long-term culture), indepen-
dent of the other cells present in the input test cell suspension
(11). A well is thus scored as negative when no CFC is detected in
it. The LTC-IC frequency is given by the reciprocal of the concen-
tration of test cell that gives 37% negative wells. Interpolation of
this value is best done using a software program to perform the
calculation (i.e., L-Calc™). It is advisable to consult a qualified stat-
istician to confirm the appropriate methodology used to compare
LTC-IC frequencies from different test populations.
4. Notes
1. The supportive effect of different types of feeder cells on the

survival of HSCs is highly variable and hence influences the
LTC-IC output significantly. It is therefore paramount to cali-
brate the LTC-IC assay whenever different feeder cells are
used. Both feeder cells presented here (primary bone marrow
16 Mouse LTC-IC 265
feeder cells and AFT024) have been extensively validated, yield

consistent results, and are readily available.
2. Collagen coating of tissue culture dishes promotes and pro-
longs the adherence to tissue-culture dishes of cell lines, par-
ticularly after they are irradiated. To coat dishes, add 1–2 mL
of sterile collagen solution (about 1 mg/ML of Type 1 from
bovine, rat, or human sources) and spread evenly; then remove
excess collagen and allow the surface to dry in biosafety hood.
Store tightly wrapped at 4°C for up to 1 month. The pre-
coated dishes can be rinsed once with sterile phosphate buff-
ered saline (PBS) or culture medium to neutralize the acidity
of the thin collagen coating prior to use.
3. It is important to note that cobblestone like areas will appear
in the LTC-IC culture. These areas should not be confused
with the cobblestone formation in the cobblestone-area-form-
ing cell (CAFC) assay. In the LTC-IC assay, not all wells that
contain cobblestone areas will give rise to CFC and vice versa.
4. It is also possible to add stimulatory cytokines directly to the
wells at the end of the long-term culture. Wells that contain
cells with colony-forming potential will show a noticeable cell
expansion/colony formation. However, it is sometimes difficult
to clearly discriminate proliferating cells from the background
of pre-existing cells (cobble-stone areas). Furthermore, feeder
cells exert a certain inhibition on proliferating cells that could
mask colony formation and lead to false negative results.
Therefore, we still recommend harvesting the cells before
replating them in methylcellulose.
References
1. Ploemacher RE, van der Sluijs JP, Voerman JS, 5. Gartner S, Kaplan HS (1980) Long-term cul-
Brons NH (1989) An in vitro limiting-dilution ture of human bone marrow cells. Proc Natl
assay of long-term repopulating hematopoietic Acad Sci USA 77(8):4756–4759
stem cells in the mouse. Blood 6. Miller CL, Rebel VI, Lemieux ME, Helgason
74(8):2755–2763 CD, Lansdorp PM, Eaves CJ (1996) Studies
2. Dexter TM, Allen TD, Lajtha LG (1977) of W mutant mice provide evidence for alter-
Conditions controlling the proliferation of nate mechanisms capable of activating
haemopoietic stem cells in vitro. J Cell Physiol hematopoietic stem cells. Exp Hematol
91(3):335–344 24(2):185–194
3. Wineman J, Moore K, Lemischka I, Müller- 7. Lemieux ME, Rebel VI, Lansdorp PM, Eaves
Sieburg C (1996) Functional heterogeneity of CJ (1995) Characterization and purification of
the hematopoietic microenvironment: rare a primitive hematopoietic cell type in adult
stromal elements maintain long-term repopu- mouse marrow capable of lymphomyeloid dif-
lating stem cells. Blood 87(10):4082–4090 ferentiation in long-term marrow “switch”
4. Dexter TM, Spooncer E, Toksoz D, Lajtha LG cultures. Blood 86(4):1339–1347
(1980) The role of cells and their products in 8. Miller CL, Rebel VI, Helgason CD, Lansdorp
the regulation of in vitro stem cell proliferation PM, Eaves CJ (1997) Impaired steel factor
and granulocyte development. J Supramol responsiveness differentially affects the detec-
Struct 13(4):513–524 tion and long-term maintenance of fetal liver
hematopoietic stem cells in vivo. Blood plantable hematopoietic stem cells. Blood
89(4):1214–1223 89(12):4337–4347
9. Kent DG, Copley MR, Benz C, Wöhrer S, 11. Fazekas de St. Groth S (1982) The evaluation
Dykstra BJ, Ma E, Cheyne J, Zhao Y, Bowie of limiting dilution assays. J Immunol Methods
MB, Zhao Y, Gasparetto M, Delaney A, Smith 49:R11
C, Marra M, Eaves CJ (2009) Prospective iso- 12. Ploemacher RE, van der Sluijs JP, van Beurden
lation and molecular characterization of CA, Baert MR, Chan PL (1991) Use of limit-
hematopoietic stem cells with durable self- ing-dilution type long-term marrow cultures
renewal potential. Blood 113(25):6342–6350 in frequency analysis of marrow-repopulating
10. Moore KA, Ema H, Lemischka IR (1997) In and spleen colony-forming hematopoietic stem
vitro maintenance of highly purified, trans- cells in the mouse. Blood 78(10):2527–2533
Chapter 17
Colony Forming Cell Assays for Human Hematopoietic

Progenitor Cells
Bert Wognum, Ning Yuan, Becky Lai, and Cindy L. Miller
Abstract
Hematopoietic stem cells (HSCs) present in small numbers in adult bone marrow (BM), peripheral blood
(PB) and umbilical cord blood (CB) produce a heterogeneous pool of progenitors that can be detected
in vitro using colony forming cell (CFC) assays. Hematopoietic progenitor cells proliferate and differenti-
ate to produce colonies of maturing cells when cultured in a semisolid methylcellulose-based medium that
is supplemented with suitable growth factors and other supplements. The colonies are then classified and
enumerated in situ by light microscopy or an automated imaging instrument. CFC assays are important
tools in basic hematology research but are also used by clinical cell processing laboratories to measure the
progenitor cell content of BM, CB and mobilized PB (MPB) preparations used for cell transplantation.
Standard CFC assays for human progenitor cells require a culture period of at least 14 days to enable opti-
mal outgrowth and differentiation of the maximum number of CFCs in a cell preparation. In this chapter
protocols are described for the detection and enumeration of myeloid multipotential progenitors and com-
mitted progenitors of the erythroid, monocyte, and granulocyte lineages in samples from human PB,
MPB, BM, and CB. In addition protocols are described for a modified version of the CFC-assay that allows
accurate enumeration of total CFC numbers in CB or MPB after a culture period of only 7 days, but with-
out distinction of colony types.
Key words: Hematopoietic progenitors, Bone marrow, Peripheral blood, Umbilical cord blood,
Colony-forming cell assays, CFU-GEMM, BFU-E, CFU-E, CFU-GM
1. Introduction
During fetal development and in the adult bone marrow, a small

number of hematopoietic stem cells (HSCs) undergo self-renewal
cell divisions and proliferate to produce a heterogeneous compart-
ment of hematopoietic progenitors. Progressive proliferation and
differentiation steps result in the production of large numbers of
267
268 B. Wognum et al.
mature blood cells including T and B lymphoid cells, natural killer

cells (NK), dendritic cells, monocyte/macrophages, granulocytes,
red blood cells, and platelets.
Numerous in vitro and in vivo assays have been developed to
characterize and quantify hematopoietic cells at various stages of
differentiation. The most definitive assays to detect HSCs with
extensive potential for self-renewal, proliferation, and multi-lineage
differentiation involve the transplantation of test cells into host
animals and detection of donor-derived hematopoietic cells weeks
to months later. The limiting dilution competitive repopulating
unit (CRU) assay is carried out using xenogeneic, immunocom-
promised recipients to detect human HSCs, and irradiated con-
genic strains to quantify mouse HSCs (1–3). The in vitro long-term
culture-initiating cell (LTC-IC) (4, 5) and cobblestone area form-
ing cell (CAFC) (6, 7) assays quantify primitive cells capable of
continuously producing myeloid cells for a minimum of 4–5 weeks
when cultured on a suitable feeder layer. Clonogenic assays have
been developed to detect hematopoietic progenitors, termed col-
ony-forming cells (CFC), in vitro. Colony forming unit (CFU)-
blast (8, 9), high proliferative potential-CFC (HPP-CFC) (10),
and CFU-granulocyte, erythroid, monocyte/macrophage, mega-
karyocyte (CFU-GEMM) (11) are representative of progenitors
with multi-lineage differentiation potential and limited self-renewal
capacity. More mature hematopoietic CFCs that have no (or mini-
mal) self-renewal capacity and are committed to differentiate into
cells of one or two hematopoietic lineages include CFU-
granulocyte/macrophage (CFU-GM, CFU-M, and CFU-G),
burst-forming-unit erythroid (BFU-E), CFU-erythroid (CFU-E),
CFU-megakaryocyte (CFU-Mk) (12–17), and, in the mouse,
CFU-pre-B (i.e., a progenitor that generates colonies of
B-lymphocytes (18)). It is very important to emphasize that
although CFC assays may detect cells that sustain (short-term)
hematopoiesis in vivo, the majority of CFCs are not HSCs, but
lineage-committed progenitors without in vivo hematopoietic
capacity (19). Nevertheless, many studies have demonstrated posi-
tive correlations between the number of hematopoietic progeni-
tors in a graft, as detected in the CFC-assay, with clinical parameters,
such as time to neutrophil and platelet engraftment and overall
survival of recipients of clinical BM, MPB, or CB grafts (20–25).
Therefore, CFC assays are frequently used by clinical therapy labo-
ratories to measure the hematopoietic cell content of BM, MPB,
and CB cell preparations selected for clinical transplantation and to
evaluate the functional integrity of the hematopoietic cells follow-
ing cell manipulations, such as volume reduction, red blood cell
removal, freezing and thawing. CFC assays are also used in basic
and preclinical research to study the biological properties of
hematopoietic progenitors in health and disease, and to examine
the efficacy and toxicity of new drugs on hematopoietic cells.
17 Human CFC Asays 269
CFC assays are performed by placing hematopoietic cell sus-

pensions into a semisolid matrix that allows individual progenitors
to divide and differentiate during the course of a suitable culture
period to produce a discrete colony containing mature cells. The
types and numbers of CFCs in the original cell preparation are then
determined on the basis of the morphological features of the col-
ony in situ using an inverted microscope or an automated imaging
instrument. The number of colonies obtained is linearly propor-
tional to the CFC content in the input cell suspension provided
that sufficiently low numbers of cells are plated, and the culture
media and culture conditions are optimal.
Methylcellulose (MC), a relatively inert polymer that forms a
stable gel with good optical clarity at a final of concentration of
0.9–1.5%, is the most commonly used agent to achieve high viscos-
ity of culture media in CFC assays. For some applications, e.g., the
detection of megakaryocyte progenitors, other culture matrices,
such as agarose and collagen are also used (16, 17).
Supplementation of CFC-assay media with a combination of
recombinant hematopoietic cytokines is essential to promote pro-
liferation and differentiation of different types of progenitor cells.
For example, Stem cell factor (SCF) (also known as c-kit ligand
and Steel factor), thrombopoietin (TPO), interleukin (IL)-3, IL-6,
GM-CSF, and G-CSF support the proliferation of immature mul-
tipotential progenitors, as well as lineage-committed progenitors
of various lineages. Other cytokines, including erythropoietin
(Epo), IL-7, and M-CSF are more specific for the development of
individual cell lineages, e.g., only erythroid, lymphoid, or mac-
rophage lineages, respectively. The reader is encouraged to consult
the published literature for more information about the actions of
cytokines individually and in synergisms (i.e., 26–29).
The formation of morphologically recognizable colonies of
maturing granulocytes, macrophages red blood cells and other
cells from human hematopoietic progenitors typically requires a
culture period ~14 days. Smaller colonies of less differentiated cells
can be detected much earlier, e.g., after 7 days of culture. The total
number of colonies detected after 7 days correlates with the num-
ber of colonies detected after 14 days, provided that optimal com-
binations of cytokines and other stimulatory agents are selected
and that the performance of the culture medium is optimized by
rigorous prescreening the various medium components, including
the methylcellulose, fetal bovine serum, and bovine serum albu-
min. Such shorter assays are useful to measure the frequency of
total hematopoietic progenitors in cell preparations, e.g., for evalu-
ating CB products in CB banks or clinical therapy labs. However,
the shorter assays cannot completely replace the longer 14-day
assays, as distinction between different progenitor types is not pos-
sible after 7 days of culture.
This chapter describes methods for the preparation of RBC-

depleted and mononuclear cell suspensions from human samples
and assays for detection of human CFU-E, BFU-E, CFU-GM, and
CFU-GEMM in methylcellulose-based media. Methods for the
enumeration of total CFCs from CB in a shorter 7-day CFC-assay
are also presented. Methods for CFC assays on cells from other
species, e.g., mouse, rat, dog, and nonhuman primates, are not
presented in this chapter, but can be found elsewhere (30).
2. Materials
2.1. Culture Media 1. Complete methylcellulose (MC)-based medium for standard

and Reagents 14-day CFC assays (e.g., STEMCELL Technologies Inc
(STEMCELL) MethoCult™ #04434, #04034 and #04435)
of CFU-E, BFU-E, CFU-GM and CFU-GEMM progenitors.
Store at −20°C.
2. MethoCult™ Express, (STEMCELL, # 04437) for 7-day CFC
assays of total CFCs, without distinction of CFC types. Store
at −20°C.
3. Media components for preparation of MC media for specific
applications (see Notes 1 and 2).
(a) 2.6% methylcellulose in Iscove’s Modified Dulbecco’s
Medium (IMDM), 40 mL per bottle (MethoCult™
#04100, STEMCELL) Store at −20°C.
(b) Fetal bovine serum (FBS) for human CFC assays: (#06250,
STEMCELL). Store in aliquots at −20°C.
(c) 10% bovine serum albumin (BSA): (#09300, STEMCELL).
Store in aliquots at −20°C.
(d) 200 mM L-glutamine stock solution: L-glutamine in phos-
phate buffered saline (PBS) (#07100, STEMCELL). Store
in aliquots at −20°C.
(e) 10−2 M 2-mercaptoethanol (2-ME) stock solution: Prepare
a 10−1 M solution by adding 0.1 mL 2-ME (#M7522,
Sigma-Aldrich, www.sigmaaldrich.com) in a total of
14.3 mL PBS. Dilute 1/10 to prepare 10−2 M stock. Store
in aliquots at −20°C for up to 6 months.
(f) Iscove’s MDM (IMDM); available from several suppliers,
e.g., STEMCELL (#36150), Invitrogen (www.invitrogen.
com, and Sigma (www.sigmaaldrich.com). Store at 2–8°C.
(g) Recombinant human (rh) cytokine stock solutions.
Cytokines are available from various suppliers. [i.e.,
STEMCELL, R&D Systems (www.rndsystems.com),
BioSource (www.biosource.com)]. Reconstitute accord-
ing to manufacturer’s instructions. Prepare individual
stock solutions at concentrations of 1–5 μg/mL in IMDM

with 0.1% BSA. Store in working aliquots at −20°C for up
to 6 months.
4. 2% FBS in IMDM (IMDM/2% FBS). (#07700 STEMCELL)
or prepare by adding 2 mL FBS to 98 mL IMDM. Filter-
sterilize using 0.2 μm filter. Store in working aliquots at 2–8°C
for up to 1 month.
5. Ficoll-Paque™ Plus (Ficoll), density 1.077 g/mL GE Healthcare
(www.gelifesciences.com)). Store at room temperature.
6. Ammonium chloride solution: (#07800, STEMCELL). Store
at −20°C and working aliquots for 1 week at 2–8°C.
7. 0.4% Trypan blue dye (#07050, STEMCELL) and 3% acetic
acid (#07060, STEMCELL) for viable and nucleated cell
counts respectively.
8. Human BM, PB, MPB, and CB samples. Cells are collected
using heparin as the anticoagulant following procedures and
handling precautions approved by the institution.
2.2. Equipment and 1. Micropipettors and 20, 200, 1,000 μL sterile tips.
Supplies 2. Culture supplies: 1, 5, 10 mL sterile pipettes; 6, 15, 50 mL
sterile tubes; 100 mm petri dishes or square bacterial dishes,
3 mL luer lock syringes, 16 gauge blunt-end needles (#28110,
STEMCELL).
3. 35 mm low adherence petri culture dishes (#27100/27150,
STEMCELL) (see Note 3).
4. 60 mm gridded dishes (#27120/27121, STEMCELL).
5. Automated cell counter or Neubauer hemocytometer.
6. Biohazard safety cabinet approved for Level II handling of bio-
logical material.
7. Incubator set at 37°C with 5% CO2 in air and >95% humidity
(see Note 4c).
8. Inverted microscope equipped with 10× or 12.5× eyepiece
objectives; 2×, 4×, and 10× planar objectives; and a moveable
stage holder for 60 mm dishes.
3. Methods
The following sections describe methods for: (1) preparation of

human hematopoietic cell samples, (2) setup of CFC assays, and
(3) identification and enumeration of CFCs. General consider-
ations for performing CFC assays (see Note 4) and procedures to
isolate individual colonies or cells from the entire culture for special
applications are discussed (see Note 5). All cell culture procedures
should be performed using sterile technique in a certified biosafety
cabinet, and universal procedures for handling potentially biohaz-
ardous materials should be followed.
Processing of the cell sample is often required to deplete red
blood cells (RBC) that can obscure colonies and make colony
counts inaccurate, deplete accessory cells (i.e., macrophages) that
produce endogenous factors in cultures and can potentially inhibit
or promote CFC growth, and to enrich hematopoietic progenitors
in samples where the CFC frequency is expected to be very low.
Ficoll density separation of PB, MPB, CB, and BM samples enriches
mononuclear cells (and thus CFCs) by removing RBCs and other
non-progenitor cell types. If it is required to use total nucleated
cell suspensions, RBCs can be lysed using ammonium chloride
treatment. This technique works well for BM samples. It can also
be used with PB and CB, but complete RBC removal may be more
difficult to achieve as the ratio of RBCs to white blood cells is
much higher in PB and CB than in BM. In addition CB contains
large numbers of nucleated RBC precursors that are not effectively
removed by this treatment.
Procedures to enrich hematopoietic progenitor cells, on the
basis of the specific cell-surface antigen expression profile of these
cells (e.g., CD34 expression using, e.g., immunomagnetic cell sep-
aration technologies or fluorescent activated cell sorting (FACS)
methodologies) are beyond the scope of this chapter and will not
be presented.
3.1. Human 1. Measure and record the volume of PB, MPB, BM, or CB start
Mononuclear Cell sample.
Isolation 2. Dilute sample with an equal volume of IMDM/2%FBS and mix
well by pipetting up and down four to five times or vortexing.
3. Add 15 mL of Ficoll per 50 mL conical tube or 3 mL per
14 mL tube for smaller sample volumes.
4. Slowly layer 30 mL of the cell suspension per 50 mL tube or
6 mL of the cell suspension per 14 mL tube onto the surface
of the Ficoll by resting the tip of the pipette against the side of
the tube. It is important to avoid mixing the cell suspension
into the Ficoll density medium as poor cell recoveries may
result. Centrifuge at room temperature for 30 min at 400 × g
with the brake “off.”
5. Using a sterile pipette, carefully remove the cells from the
interface between the plasma/medium layer and the Ficoll.
Transfer cells to a 15 mL tube and dilute cell suspension with
a minimum of two volumes (>1:2 ratio) of IMDM/2% FBS.
Mix well by pipetting up and down four to five times, and then
centrifuge for 10 min at 300 × g with brake “on.”
6. Carefully decant off supernatant leaving ~0.5 mL of medium

on the cell pellet. Vortex to resuspend the cells, fill tube with
IMDM/2% FBS and mix well by vortexing or by pipetting up
and down four to five times. Centrifuge tube(s) for 10 min at
300 × g with brake “on.”
7. Carefully decant off supernatant. Add 1–3 mL of IMDM/2%
FBS to the cell pellet and make a single cell suspension by vor-
texing or by pipetting up and down four to five times.
8. Measure and record processed sample volume. Perform a
nucleated cell count using an automated cell counter or manu-
ally using 3% acetic acid and a Neubauer hemocytometer.
A dilution of 1:20 to 1:50 is usually suitable. Calculate and
record the cell concentration. The following yields of mononu-
clear cells from the start samples can be expected; 1–2 × 106 per
mL PB, 1–2.5 × 106 per mL CB and 0.5–1 × 107 per mL BM.
3.2. Ammonium 1. Measure and record volume of start sample to be processed.

Chloride Treatment Perform a cell count if estimation of cell recovery is required.
Add a 4:1 v/v ratio of ammonium chloride solution (e.g.,
2 mL of heparinized BM or blood and 8 mL ammonium chlo-
ride solution).
2. Mix well by inverting tube three to four times or by vortexing
gently. Place tube on ice for a total of 10 min with mixing as
above after ~5 min of incubation. The majority of RBC should
now be lysed. Fill tube with IMDM/2% FBS and centrifuge
for 10 min at 300 × g with brake “on.”
3. Carefully decant off supernatant leaving ~0.5 mL of medium
on the cell pellet. Vortex to resuspend the cell pellet, fill tube
with IMDM/2% FBS and mix well by vortexing or by pipetting
up and down four to five times. Centrifuge tube for 10 min at
300 × g with brake “on.” Repeat this wash step once more.
4. Carefully decant off supernatant. Add 2 mL of IMDM/2%
FBS and make a single cell suspension by vortexing or by
pipetting up and down four to five times. Measure and record
processed sample volume. Perform a nucleated cell count and
calculate the cell concentration and cell recovery. Percent cell
recovery is calculated using the following formula: (cell con-
centration × volume of processed sample) divided by (cell con-
centration × volume of start sample) × 100. A recovery of
60–80% of the nucleated cells from the start sample of normal
BM can be expected.
3.3. Setup of Human 1. To identify assays, label lids of 35 mm petri dishes at the edge
CFC Assays using a permanent fine felt marker.
2. Thaw aliquots of human MC-medium (see Notes 6 and 7) at
room temperature or under refrigeration. Vortex tubes to
ensure all components are thoroughly mixed.
Table 1
Recommended input cell numbers for Human CFC assays
Recommended input cell concentration

Cell source in 1.1 mL per 35 mm disha
Bone marrow—ammonium chloride treated 5 × 104 (2 × 104 –1 × 105)

Bone marrow—mononuclear cellsb 2 × 104 (1 × 104 – 5 × 104)
Cord blood—unprocessed, minimally processed or 2 × 104 (1 × 104 – 4 × 104)
RBC-depleted
Cord blood—mononuclear cellsb 1 × 104 (5 × 103 – 2 × 104)
Peripheral blood—mononuclear cellsb 2 × 105 (1– 4 × 105)
Lineage-depleted cell suspensions (BM, CB, MPB) 1,000c (500 – 2 × 103)
CD34+ cells (BM, CB, MPB) 400c (150 –1 × 103)
a
The recommended input cell concentration should optimally yield 30–80 colonies per culture using normal samples
from the various tissues. As the progenitor frequency often cannot be estimated (e.g., in samples from leukemic and
drug-treated patients), two or three input cell doses within the range shown in parenthesis should be set up
b
Mononuclear cells isolated using Ficoll-Paque™
c
Dependent on CD34+ cell frequency, generally, one can assume that 10–20% of CD34+ cells form colonies.(see Note 8)
3. Dilute hematopoietic cells to 10× the final concentration required

in IMDM/2% FBS (see Table 1 and Notes 4b and 8). For exam-
ple, to achieve a final concentration of 1 × 105 cells per 35 mm
dish, dilute cells to 1 × 106 cells per mL in IMDM/2% FBS.
4. Add 0.3 mL of cells to 3 mL of complete human MC-medium
for duplicate cultures or 0.4 mL of cells to 4 mL of MC-medium
for triplicate cultures.
5. Vortex tubes and let stand for 2–5 min to allow bubbles to
rise.
6. Using a 3 mL luer-lock syringe and 16 g blunt-end needle,
draw up ~1 mL and expel completely to remove most of air
from syringe. Draw up ~3 mL and carefully dispense 1.1 mL
into each 35 mm dish. Distribute methylcellulose evenly by
gently tilting and rotating each dish. Avoid getting MC on lids
or up the sides and break any large bubbles using a dry sterile
micropipettor tip.
7. Place the two labeled 35 mm petri dishes into a 100 mm dish.
Add a third 35 mm dish (without lid) containing 3–4 mL of
sterile water to help maintain a high humidity over the culture
period. Larger petri dishes or square bacterial culture dishes
can be used as outer dishes when three or more replicate cul-
tures are set up.
8. Place cultures in an incubator maintained at 37°C, 5% CO2 in
air, and >95% humidity (see Note 4c). Cells plated in
MethoCult™ Express should be cultured for at least 7 days;

cells cultured in methylcellulose-media for standard assays
(e.g., MethoCult™H4034) should be cultured for 14–16 days.
If assays cannot be counted immediately after completion of
the culture period, transfer cultures to an incubator maintained
at 33°C, 5% CO2 in air, and >95% humidity and count within
3–4 days (see Note 4d).
3.4. Identifying and 1. To prepare a reusable gridded template, draw a centered “+”
Counting Human CFCs on the bottom of a 60 mm gridded dish, and place a mark on
these lines corresponding to the outer edges of a 35 mm dish
using a fine permanent marker.
2. Keeping cultures as level as possible, center the 35 mm dish
within the gridded 60 mm dish and place on the moveable
stage of an inverted microscope. Scan the entire dish on low
power (2× objective) by moving the stage vertically up and
down and then laterally (helps minimize “motion nausea”).
Note the relative distribution of the colonies (see Note 4e)
3.4.1. Counting Colonies 1. Scan the dish on low power (2× objective, 20× to 25×
in a 7-Day CFC-Assay magnification) to evaluate relative distribution of colonies.
Cultured in MethoCult™ Score colonies with a 4× objective and count all colonies con-
Express taining more than 20 cells. It is important to continuously
refocus to identify colonies that are present in different planes
and at the outer edges of the cultures. As most colonies are
immature, scoring individual colony types is not recommended
after 7 days of culture (see Note 9). For example of colonies,
see Fig. 1.
2. Optionally, the cultures can be returned to the incubator after
scoring and cultured for another 7–9 days. Individual colony
Fig. 1. Photographs of hematopoietic colonies taken after 7 days (a) and 14 days (b) of culture of Human CB cells in
MethoCult™ Express.
Fig. 2. Photographs of human hematopoietic colonies taken after 14 days of culture in MethoCult H4434 (a) CFU-E
containing £200 small hemoglobinized erythroblasts which have a reddish color when viewed microscopically; (b) BFU-E
containing many clusters and >200 cells which appear red when viewed microscopically;(c) CFU-GM containing >200
granulocyte and monocyte lineage cells; (d) CFU-GEMM containing erythroid cells surrounded by 20 or more granulocyte
and monocyte lineage cells
types (BFU-E, CFU-GM, and CFU-GEMM) can then be

identified and scored, as described in Subheading 3.4.2.
Colonies in MethoCult™ Express after 14–16 days will appear
larger than in other MC media, but the morphology of the
colony types will be similar (see Note 9c).
3. Alternatively, an automated imaging instrument can be used to
enumerate CFC colonies (see Note 10).
3.4.2. Counting Colonies in 1. Count CFU-E on the entire plate using high power (4× objec-
a 14–16 Day CFC-Assay tive). BFU-E, CFU-GM, and CFU-GEMM are then counted
using a lower power (2× objective). A higher power (4× or 10×
objective) is used to identify cell types within a colony for the
purpose of confirming colony classification. It is important to
continuously refocus to identify colonies that are present in dif-
ferent planes and at the outer edges of the cultures. Observe
that some CFU-GM colonies have two or more focal points.
For descriptions of human CFC refer to Fig. 2 and Table 2.
2. Alternatively, an automated imaging instrument can be used to
enumerate CFC colonies (see Note 10).
Table 2
Description of human CFCs and cytokine combinations used for CFC assays
CFC class Cytokine(s)a (optional)b Human CFC description
CFU-E Epo Produces one or two clusters of

erythroblasts, £200 cells
BFU-E IL-3 + Epo + SCF Produces one or more clusters of erythro-
blasts and greater than 200 total cells
CFU-G/Mc IL-3 + GM-CSF + SCF (G-CSF, Produces one or more clusters containing
M-CSF, IL-6, IL-5) 40 or more granulocyte (CFU-G),
monocyte (CFU-M), or granulocyte
and monocyte (CFU-GM) cells
CFU-GEMM Cytokines to support each lineage. Produces a minimum of 40 cells (usually
IL-3 + GM-CSF + SCF + Epo larger) and contains erythroid cells as
(G-CSF, M-CSF, IL-6, Tpo) well as granulocytic, monocytic, and
megakaryocytic lineage cells
CFU-Mk Tpo + IL-3 + IL-6 (SCF, IL-11) Produces 3 or more megakaryocytes
a
Minimal cytokine combination used for detection of these progenitors in vitro
b
Cytokines indicated in parenthesis can be added (or substituted in some applications) as desired
c
Total CFU-G/M = CFU-GM + CFU-G + CFU-M
4. Notes
1. Semisolid media can be prepared by adding individual compo-

nents to a base MC-medium. This allows flexibility in modify-
ing the medium formulation, e.g., by substituting or adding
cytokines or other compounds. To prepare 100 mL of human
MC-medium from 2.6% MC stock solution (e.g., MethoCult
H4100, STI) and media components: first thaw a 40 mL bot-
tle of 2.6% MC stock solution at room temperature or in the
refrigerator and then add the individual components directly
to the bottle: FBS (final 10–30%), BSA (0.5–2%), l-glutamine
(final 1–2 mM), 2-ME (final 10−4 M), cytokines, and other
optional compounds.. Adjust the volume to 100 mL using
Iscove’s MDM. The MC concentration will be ~1% in the final
formulation. Mix components thoroughly and aliquot into
tubes for storage at −20°C. If desired the total volume of liq-
uid media components can be adjusted to achieve either a
higher or a lower MC-concentration and obtain more dis-
persed (“loose”) or tighter colonies respectively. The added
components and their final concentrations should be tested in
the MC culture system and selected to provide satisfactory
colony growth.
2. If the researcher chooses to prepare methylcellulose-based
medium using components from suppliers other than those
listed in Section 2.1.3 several factors must be taken into

consideration. There is large variability among raw materials
(i.e., methylcellulose powder, FBS, and BSA) from different
suppliers, and from one batch to another for their ability to
support the growth of CFCs. As such, samples from several
batches of each component should be obtained and compared
for their ability to support growth of the maximal number of
colonies. The selected components should then be combined
and retested. Once components have been selected, sufficient
amounts to last several years should be purchased since com-
ponent screening is very time-consuming and labor-intensive.
3. It is important to use petri culture dishes that have been
screened for low adherence because excessive cell adherence
can inhibit colony growth and make it difficult to distinguish
individual colonies.
4. General considerations for attaining accurate and reproducible
results when performing hematopoietic CFC assays.
(a) Cell preparations. It is advisable to set up assays using
freshly isolated cells. When this is not feasible it is impor-
tant to perform preliminary experiments to establish opti-
mal assay conditions, document all cell processing
information and include appropriate controls in each
experiment. For example, human CFC assays can be done
using cryopreserved cells and using samples stored for
24 h at room temperature.
(b) Input cell concentrations and colony numbers. The optimal
input cell number is dependent on the progenitor cell fre-
quency, cell source, and culture conditions. Refer to
Table 1 for recommended plating densities for different
cell sources. As a guideline, colony numbers between 30
and 80 colonies per 35 mm dish (1.1 mL culture) can be
considered optimal under most conditions. Colony num-
bers below 30 per dish may not yield statistically accurate
data. Colony numbers above 80 per dish may causes inac-
curacies by inhibition of progenitor proliferation resulting
from depletion of essential nutrients and accumulation of
toxic cellular metabolic products, and counting errors due
to difficulty in identifying individual colonies. As a result
colony numbers may not be proportional to input cell
numbers if plating densities are too high and colony counts
may be inaccurate. In particular for 14 day CFC assays on
CB cells, which can result in very large colonies, plating
densities that yield more than 80 colonies can be consid-
ered too high. For 14-day CFC assays on BM and MPB
and also for 7-day CFC assays on CB, which result in small
colonies, plating densities that give more than 80 colonies
(e.g., up to 120 colonies) may be acceptable. It is
recommended to plate cell samples with unknown

progenitor frequencies at two or more different plating
densities to ensure that at least one culture condition yields
colony numbers that are in the useful range.
(c) Culture conditions. It is important to maintain correct incu-
bator incubation conditions. It is important to routinely
monitor the temperature, CO2 and humidity levels, as well
as to regularly clean the incubator. Small chamber incuba-
tors (i.e., approximately 0.3 m3) with a water-pan placed in
the bottom of the chamber give more uniform temperature
and humidity than large chamber water-jacketed incuba-
tors. A small amount of copper sulfate added into the water
pan inhibits bacterial and fungal growth. The temperature
and CO2 levels should be monitored independently from
incubator gauges using in-chamber thermometers and gas
monitors (e.g., Fyrite CO2 device), respectively.
(d) If cultures cannot be counted at the end of the appropriate
time, transferring the cells to an incubator set at a lower
temperature (i.e., 33°C) will maintain colony morphology
for a few days provided that high humidity is maintained.
However, it is recommended to count the colonies as soon
as possible.
(e) Colony enumeration. Practice is required to gain compe-
tence in CFC identification and enumeration. Recounting
the same dishes on consecutive days and comparative
counting with coworkers (same dishes) and with research-
ers at other institutions (CFC assays set up with the same
(cryopreserved) cell suspension and culture conditions) is
recommended. Atlases of human hematopoietic colonies
are available from STEMCELL. In addition technical vid-
eos and tutorials on CFC-assay setup colony scoring can
be found on the Web site www.Stemcell.com.
5. For certain applications such as cytogenic analysis, DNA, RNA,
and protein analyses, and replating experiments to detect pro-
genitor self-renewal capacity, individual colonies or cells from
the entire culture can be isolated. The cultures are usually
incubated for shorter time periods to ensure high viability of
cells within the colonies (~7–12 days). A well-isolated colony
is identified using an inverted microscope (can be placed within
a biosafety cabinet if desired). Individual colonies are “plucked”
in the smallest possible volume using a micropipettor and
200 μL pipette tips. The colony is placed in sterile 0.5 or
1.5 mL microtubes containing the appropriate wash medium.
Entire MC-cultures are harvested by adding 1–2 mL of medium
(i.e., PBS or IMDM/2% FBS) to the dish and then gently mix-
ing using a pipettor and 1,000 μL tips. The rinsing step is
repeated several times and all washes are combined into a
15 mL conical tube for a single dish or into a 50 mL conical

tube when two to three dishes are harvested. Sufficient wash
medium is added to microtubes or tubes to dilute MC by five-
to tenfold. Centrifuge tubes for 10 min at 300 × g and micro-
tubes for 3–5 min at 300 × g. Cells are washed at least once
more before use.
6. Special handling procedures are required when working with
MC-based media. The freezing process causes focal areas where
the MC becomes more concentrated and “lumps” may form if
the media is thawed rapidly (i.e., at 37°C). Due to the unique
properties of MC solutions, lower temperatures are required
to dissolve these lumps. Therefore, MC-based media should
always be thawed at room temperature (this requires approxi-
mately 4 h) or at 2–8°C (i.e., overnight in the refrigerator). If
accidentally thawed at 37°C, place on ice or in refrigerator for
1–2 h to dissolve the lumps.
7. MC solutions are very viscous and syringes and large bore nee-
dles (i.e., 16-gauge) should be used for accurate aliquoting
and dispensing. Use of blunt-end needles is recommended to
prevent needle prick injuries. To aliquot bottles of MC, mix
well by shaking vigorously for 30–60 s and let stand for 2–5 min
to allow air bubbles to rise. Immerse the needle end just below
the surface of the methylcellulose and slowly draw up the
medium. Expel the medium back into the bottle to remove the
air bubbles present in the syringe. Repeat twice more before
drawing up the final volume to be dispensed plus an extra vol-
ume (example: if dispensing 3 mL, then draw up to the 4 mL
mark and dispense from the 4 mL mark to the 1 mL mark for
an accurate 3 mL volume). Dispense the MC-medium, cap the
tubes tightly, and store at −20°C.
8. Appropriate input cell numbers for human CFC assays can be
estimated if the CD34+ cell content of the sample is known
(i.e., by anti-CD34 antibody immunostaining and FACS anal-
ysis) Approximately 10–20% of the CD34+ cell population are
BFU-E, CFU-GM, and CFU-GEMM. For example, for a BM
sample containing 1% CD34+ cells, an input cell number of
5 × 104 cells per 35 mm dish should yield 50–100 colonies.
9. Considerations related to 7-Day CFC assays for measuring
total progenitor numbers.
(a) Colonies in MethoCult™ Express after 7 days of culture
can either be compact, i.e., consisting of a single cluster of
cells, or composed of several clusters of cells (Fig. 1). As
some immature hematopoietic progenitors are mobile,
even in semisolid medium, the individual clusters of one
colony (i.e., derived from a single progenitor) can be as far
apart from one another as the distance equivalent to sev-
eral cluster diameters. This is more apparent in the 7 day
Fig. 3. Correlation between total colony numbers after 7 days of culture in MethoCult
TM
Express and after 14 days of culture in MethoCult TMH4034, Optimum. Results are shown
for 30 CB samples (diamonds) and 5 mobilized PB samples (squares). Correlation
coefficient, R 2 = 0.97.
assay than the 14 day assay because, after 7 days of culture,

the cells within the colony have not yet proliferated
sufficiently to fill the spaces between the clusters. To decide
whether different clusters belong to the same or different
colonies, it is important to examine the context, size and
morphology of the clusters. If clusters in close proximity
to each other are of similar size and morphology, they are
likely to belong to the same colony. If adjacent clusters are
different in size and/or show different morphologies, and
the colony density in the dish is high (>50 colonies per
35 mm dish), the clusters likely belong to different colo-
nies (i.e., are derived from different progenitors).
(b) Essential requirements for 7-day CFC-assay are that col-
ony numbers detected after 7 days of culture give an accu-
rate estimate of the total number of progenitors in the
sample. Total colony numbers counted on day 7 should
correlate strongly and be very similar (at least 90%) of total
colony numbers counted in 14 day CFC assays on the
same cell preparations. These requirements can be met by
selecting MC media that promote rapid progenitor cell
proliferation such that most progenitors already develop
into detectable colonies of 20 cells or more by day 7
(Fig. 3, MethoCultTM Express day 7 vs. H4034 day 14).
Media that have been optimized for 14 day CFC assays,
e.g., MethoCultTM H4034, may not give accurate colony
counts on day 7, as proliferation is slower and most colo-

nies may still be too small to be reliably identified after
7 days of culture.
(c) In 7-day CFC assays it is not possible to identify different
colony types as most cells have not yet sufficiently matured
to be recognizable on the basis of their morphology.
However, cultures can be maintained for a longer period
and different colony types can be scored after ~14 days.
Colonies in a MC medium that has been optimized for 7 day
assays, e.g., MethoCultTM Express, can get very large by day
14 and it may be difficult to accurately distinguish individual
colonies in dishes plated at high cell concentrations. This is
in particular relevant for CFC assays on CB cells, as colonies
cultured from CB are typically larger than colonies cultured
from BM or MPB. It is recommended to plate CFC assays
at different cell concentrations to ensure that at least one
plating density gives optimal colony numbers.
10. STEMvision™ is an instrument designed specifically for imag-
ing and scoring colonies in the hematopoietic colony-forming
cell (CFC) assay using MethoCult™ media and meniscus-
reducing SmartDish™ cultureware. (www.stemcell.com).
References
1. Szilvassy SJ, Humphries RK, Lansdorp PM, Hematopoietic stem cell protocols. Humana,
Eaves AC, Eaves CJ (1990) Quantitative assay Totowa, NJ
for totipotent reconstituting hematopoietic 6. Ploemacher RE, van der Sluijs JP, van Beurden
stem cells by a competitive repopulation strat- CA, Baert MR, Chan PL (1991) Use of a lim-
egy. Proc Natl Acad Sci USA 87:8736–8740 iting-dilution type long-term cultures in fre-
2. Conneally E, Cashman J, Petzer A, Eaves C quency analysis of marrow-repopulating and
(1997) Expansion in vitro of transplantable spleen colony-forming hematopoietic stem
human cord blood stem cells demonstrated cells in the mouse. Blood 78:2527–2533
using a quantitative assay of their lympho- 7. de Haan G, Ploemacher R (2002) The cobble-
myeloid repopulating activity in nonobese area-forming cell assay. In: Klug CA, Jordan
diabetic-scid/scid mice. Proc Natl Acad Sci CT (eds) Hematopoietic stem cell protocols.
USA 94:9836–9841 Humana, Totowa, NJ
3. Szilvassy SJ, Nicolini FE, Eaves CJ, Miller CL 8. Nakahata T, Ogawa M (1982) Identification in
(2002) Quantitation of murine and human culture of a class of hemopoietic colony-form-
hematopoietic stem cells by limiting dilution ing units with extensive capability to self-renew
analysis in competitively repopulated hosts. In: and generate multipotential hemopoietic colo-
Klug CA, Jordan CT (eds) Hematopoietic nies. Proc Natl Acad Sci USA 79:3843–3847
stem cell protocols. Humana, Totowa, NJ 9. Brandt JE, Baird N, Lu L, Srour E, Hoffman
4. Sutherland HJ, Lansdorp PM, Henkelman R (1988) Characterization of a human
DH, Eaves AC, Eaves CJ (1990) Functional hematopoietic progenitor cell capable of
characterization of individual human forming blast cell containing colonies in vitro.
hematopoietic stem cells cultured at limiting J Clin Invest 82:1017–1027
dilution on supportive marrow stromal layers. 10. McNiece IK, Stewart FM, Deacon DM,
Proc Natl Acad Sci USA 87:3584–3588 Temeles DS, Zsebo KM, Clark SC, Quesenberry
5. Miller CL, Eaves CJ (2002) Long-term cul- PJ (1989) Detection of a human CFC with a
ture-initiating cell assays for human and murine high proliferative potential. Blood
cells. In: Klug CA, Jordan CT (eds) 74:609–612
11. Fauser AA, Messner HA (1979) Identification 21. Hogge DE, Lambie K, Sutherland HJ, Benny
of megakaryocytes, macrophages, and eosino- WB, Dalal B, Currie C, Barnett MJ, Eaves AC,
phils in colonies of human bone marrow con- Eaves CJ (2000) Quantitation of primitive and
taining neutrophilic granulocytes and lineage-committed progenitors in mobilized
erythroblasts. Blood 53:1023–1027 peripheral blood for prediction of platelet
12. Ichikawa Y, Pluznik DH, Sachs L (1966) In recovery post autologous transplant. Bone
vitro control of the development of mac- Marrow Transplant 25:589–598
rophage and granulocyte colonies. Proc Natl 22. Iori AP, Cerretti R, De Felice L, Screnci M,
Acad Sci USA 56:488–495 Mengarelli A, Romano A, Caniglia M, Cerilli L,
13. Stephenson JR, Axelrad AA, McLeod DL, Gentile G, Moleti ML, Giona F, Agostini F,
Shreeve MM (1971) Induction of colonies of Pasqua I, Perrone MP, Pinto MR, Grapulin L,
hemoglobin-synthesizing cells by erythropoi- Testi AM, Martino P, De Rossi G, Mandelli F,
etin in vitro. Proc Natl Acad Sci USA Arcese W (2004) Pre-transplant prognostic fac-
68:1542–1546 tors for patients with high-risk leukemia under-
14. Eaves CJ, Eaves AC (1978) Erythropoietin going an unrelated cord blood tranplantation.
(Ep) dose-response curves for three classes of Bone Marrow Transplant 33:1097–1105
erythroid progenitors in normal human mar- 23. Yang H, Acker JP, Cabuhat M, Letcher B,
row and in patients with polycythemia vera. Larratt L, McGann LE (2005) Association of
Blood 52:1196–1210 post-thaw viable CD34+ cells and CFU-GM
15. Nakeff A, Daniels-McQueen S (1976) In vitro with time to hematopoietic engraftment. Bone
colony assay for a new class of megakaryocyte Marrow Transplant 35:881–887
precursor: colony-forming unit megakaryocyte 24. Yoo KH, Lee SH, Kim HJ, Sung KW, Jung
(CFU-M). Proc Soc Exp Biol Med HL, Cho EJ, Park HK, Kim HA, Koo HH
151:587–590 (2007) The impact of post-thaw colony-form-
16. Dobo I, Allegraud A, Navenot JM, Boasson ing units-granulocyte/macrophage on engraft-
M, Bidet JM, Praloran VJ (1995) Collagen ment following unrelated cord blood
matrix: an attractive alternative to agar and transplantation in pediatric recipients. Bone
methylcellulose for the culture of hematopoi- Marrow Transplant 39:515–521
etic progenitors in autologous transplantation 25. Prasad VK, Mendizabal A, Parikh SH, Szabolcs
products. J Hematother 4:281–287 P, Driscoll TA, Page K, Lakshminarayanan S,
17. Hogge D, Fanning S, Bockhold K, Petzer A, Allison J, Wood S, Semmel D, Escolar ML,
Lambie K, Lansdorp P, Eaves A, Eaves C Martin PL, Carter S, Kurtzberg J (2008)
(1997) Quantitation and characterization of Unrelated donor umbilical cord blood transplan-
human megakaryocyte colony-forming cells tation for inherited metabolic disorders in 159
using a standardized serum-free agarose assay. pediatric patients from a single center: influence
Br J Haematol 96:790–800 of cellular composition of the graft on transplan-
tation outcomes. Blood 112:2979–2989
18. Paige CJ, Kincade PW, Shinefeld LA, Sato VL
(1981) Precursors of murine B lymphocytes. 26. Ogawa M (1993) Differentiation and prolif-
Physical and functional characterization and eration of hematopoietic stem cells. Blood
distinctions from myeloid stem cells. J Exp 81:2844–2853
Med 153:154–165 27. Krystal G, Alai M, Cutler RL, Dickeson H,
19. Zijlmans JM, Visser JW, Laterveer L, Kleiverda Mui AL, Wognum AW (1991) Hematopoietic
K, Heemskerk DP, Kluin PM, Willemze R, growth factor receptors. Hematol Pathol
Fibbe WE (1998) The early phase of engraft- 5:141–162
ment after murine blood cell transplantation is 28. Kaushansky K, Drachman JG (2002) The
mediated by hematopoietic stem cells. Proc molecular and cellular biology of thrombopoi-
Natl Acad Sci USA 95:725–729 etin: the primary regulator of platelet produc-
20. Migliaccio AR, Adamson JW, Stevens CE, tion. Oncogene 21:3359–3367
Dobrila NL, Carrier CM, Rubinstein P (2000) 29. Metcalf D (2008) Hematopoietic cytokines.
Cell dose and speed of engraftment in placen- Blood 111:485–491
tal/umbilical cord blood transplantation: graft 30. Pereira C, Clarke E, Damen J (2007)
progenitor cells content is a better predictor Hematopoietic colony-forming cell assays. In:
than nucleated cell quantity. Blood Vemuri MC (ed) Methods in molecular biology
96:2717–2722 407, stem cell assays. Humana, Totowa, NJ
Chapter 18
Studying Leukocyte Recruitment Under Flow Conditions

Sean A. Parsons, Christophe Jurzinsky, Susan L. Cuvelier,
and Kamala D. Patel
Abstract
Leukocyte recruitment from the vasculature occurs under conditions of haemodynamic shear stress. The
parallel plate flow chamber apparatus is an in vitro system that is widely used to study leukocyte recruit-
ment under shear conditions. The flow chamber is a versatile tool for examining adhesive interactions, as
it can be used to study a variety of adhesive substrates, ranging from monolayers of primary cells to isolated
adhesion molecules, and a variety of adhesive particles, ranging from leukocytes in whole blood to anti-
body-coated latex beads. We describe here methods for studying leukocyte recruitment to cytokine-stim-
ulated, transfected or transduced endothelial cells using both whole blood and isolated leukocyte
suspensions. These methods enable multiple parameters to be measured, including the total number of
recruited leukocytes, the percentage of leukocytes that are rolling or firmly adherent, and the percentage
of leukocytes that have transmigrated. Although these methods are described for interactions between
leukocytes and endothelial cells, they are broadly applicable to the study of interactions between many
combinations of adhesive substrates and adhesive particles.
Key words: Parallel plate flow chamber, Recruitment, Tethering, Rolling, Adhesion, Detachment,
Transmigration, Transfection, Transduction, Shear stress, Whole blood, Leukocyte, PBMC, PMN,
Lymphocyte, Monocyte, Neutrophil, Eosinophil
1. Introduction
The recruitment of leukocytes from the vasculature is a process

that is essential for the normal function of the immune system.
When inappropriately regulated, leukocyte recruitment can con-
tribute to inflammatory diseases such as asthma, sepsis, and multi-
ple sclerosis. The importance of leukocyte recruitment to both
physiological and pathological processes has led to extensive
research in this field (1, 2). Critical to this field is the parallel plate
285
286 S.A. Parsons et al.
Fig. 1. A schematic diagram of a circular, 35-mm dish parallel plate flow chamber setup (upper) or an ibidi μ-slide VI flow
chamber setup (lower).
flow chamber apparatus, which is widely used as a tool for studying

leukocyte recruitment in vitro (3–5).
Parallel plate flow chambers are used to generate shear condi-
tions that are similar to those that are found in the vasculature.
Although the design of flow chambers is quite varied, all have a
number of features in common (see Fig. 1). All flow chambers have
two surfaces arranged in parallel: one lower plate onto which an
adhesive substrate is immobilized and one upper plate that serves as
the top of the chamber. Some flow chambers have a device such as
a gasket that maintains a defined distance between the lower and
upper plates, while others are constructed as one solid piece, there-
fore removing the need for a gasket (see Fig. 1). Finally, all flow
18 Paralled Plate Flow Chamber Assay 287
Table 1
Examples of adhesive surfaces and adhesive “particles” that can be used in a
parallel plate flow chamber apparatus
Adhesive surface Adhesive particles
Monolayer of isolated primary cells Whole blood

Monolayer of adherent cell line Isolated primary cells
Lipid bilayer containing isolated adhesion molecule Suspension cell line
Isolated adhesion molecule Ligand-coated latex beads
Fragment of isolated adhesion molecule Antibody-coated latex beads
chambers are equipped with the ability to move liquid over the
adhesive substrate on the lower plate at a defined rate. These ele-
ments of the flow chamber apparatus enable defined shear rates and
shear stresses to be generated in vitro, thereby facilitating the study
of leukocyte recruitment under physiologically relevant conditions.
The two primary requirements for a flow chamber experiment
are an adhesive substrate, which is immobilized on the lower plate
of the flow chamber, and a suspension of adhesive particles, which
is perfused over the adhesive substrate. Many different choices
exist for both the adhesive substrate and the adhesive particles (see
Table 1). Adhesive substrates that can be used in the flow chamber
range from monolayers of primary cells to isolated adhesion mol-
ecules, while adhesive particles range from leukocytes in whole
blood to antibody-coated latex beads. The wide variety of adhesive
substrates and adhesive particles that can be used in the flow cham-
ber system makes it a versatile tool for addressing multiple ques-
tions regarding leukocyte recruitment.
In these protocols, we detail methods for examining leukocyte
recruitment to cytokine-stimulated endothelial cells. Methods for
measuring recruitment from both whole blood and isolated leuko-
cyte suspensions are described. Although a specific adhesive sub-
strate and specific adhesive particles are used here for illustrative
purposes, these protocols should be applicable to any combination
of substrate and particles.
2. Materials
2.1. Endothelial Cell 1. Cultured human endothelial cells in 35-mm culture dishes, or
Stimulation ibidi μ-slide VI flow chambers (ibidi, Munich, Germany).
2. 20% Human Serum Albumin (HSA) (Gemini Bioproducts,
West Sacramento, CA). Supplied sterile. Store at 4°C.
3. Sterile-filtered M199 (Sigma Chemicals, St. Louis, MO). Store

at 4°C. Heat to 37°C before use.
4. Endothelial cell medium (ECM): M199 with 20% human
serum (isolated from donors—see Note 1), and 1% penicillin/
streptomycin/glutamine.
5. Hanks Balanced Salt Solution; with calcium chloride, magne-
sium chloride, and magnesium sulfate; without sodium bicar-
bonate and phenol red (HBSS). Store at 4°C. Heat to 37°C
before use.
6. Recombinant human cytokine(s) (R&D Systems, Minneapolis,
MN). Store in aliquots at −20°C and transfer to storage at 4°C
as needed.
2.2. Endothelial Cell 1. Cultured human endothelial cells in 35-mm culture dishes or
Transduction ibidi μ-slide VI flow chambers.
2. 20% Human Serum Albumin (HSA). Supplied sterile. Store at
4°C.
3. Sterile-filtered M199. Store at 4°C. Heat to 37°C before use.
5. Sterile aerosol pipette tips, RNase-, DNase-, endotoxin-free
(VWR, West Chester, PA).
6. Appropriate adenovirus construct (see Note 2).
2.3. Endothelial Cell 1. Cultured human endothelial cells in 35-mm dishes or ibidi
siRNA Transfection μ-slide VI flow chambers.
2. 20% Human Serum Albumin (HSA).
3. Sterile-filtered M199. Store at 4°C. Heat to 37°C before use.
4. Opti-MEM medium (Invitrogen Life Technologies, Carlsbad,
CA).
6. HiPerFect reagent (Qiagen Inc., Mississauga, Ontario).
7. Appropriate siRNA. (Qiagen Inc., Mississauga, Ontario).
2.4. Whole Blood 1. HBSS (As described in Subheading 2.1).

Preparation and 2. Heparin LEO (10,000 IU/ml), preservative-free (LEO
Leukocyte Isolation Pharma, Ballerup, Copenhagen).
3. Human blood donors.
4. Hemacolor Stain Set (VWR International, West Chester, PA).

Store at 20°C.
2.5. Flow Chamber 1. HBSS (as described in Subheading 2.1).
Setup 2. Phase-contrast microscope with 10×, 20×, and 40× objectives
and equipped with a stage that holds 35-mm dishes (Carl Zeiss,
Inc., Thornwood, NY), or standard microscope slides.
3. CCD camera (Hitachi Denshi, Ltd, Tokyo, Japan).
4. DVD Recorder (Pioneer).
5. Flow chamber and gasket (Glycotech, Rockville, MD), or ibidi
μ-slide VI flow chamber.
6. Infuse/refill syringe pump (Harvard Apparatus, Inc, Holliston,
MA).
7. Pharmed 65, 1/16 in. internal diameter, 1/16 in. external
diameter tubing (Saint-Gobain Performance Plastics Co.,
Akron, OH).
8. For ibidi chambers only: No lock male luer adaptors, 1/16 in.
barb. (www.smallparts.com).
9. Vacuum pump and vacuum flask.
10. 37°C water bath.
2.6. Whole Blood 1. HBSS and Hemacolor Stain Set (as described in Subheadings 2.1
Recruitment and 2.4).
Experiment 2. Stimulated, transfected, or transduced human endothelial cells
(From Subheading 3.1, 3.2, or 3.3).
3. Heparinized (10 IU/ml) whole blood from human donors
(From Subheading 3.4).
4. Flow Chamber Setup (From Subheadings 3.5.1 and 3.5.2).
2.7. Whole Blood 1. DVD recording (From Subheading 3.6).

Recruitment Analysis
2.8. Isolated Leukocyte 1. HBSS (as described in Subheading 2.1).

Recruitment 2. Stimulated, transfected, or transduced human endothelial cells
Experiment (From Subheading 3.1, 3.2, or 3.3).
3. Isolated leukocytes from human donors (From
Subheading 3.4).
4. Flow Chamber Setup (From Subheadings 3.5.1 and 3.5.2).
2.9. Isolated Leukocyte 1. DVD recording (From Subheading 3.8).

Recruitment Analysis
3. Methods
3.1. Endothelial Cell 1. Isolate endothelial cells and grow in 35-mm culture dishes or
Stimulation ibidi μ-slide VI chambers, as described for Human Umbilical
Vein Endothelial Cells (HUVEC) (6, 7), Human Pulmonary
Microvascular Endothelial Cells (HPMEC) (8), and Human
Dermal Microvascular Endothelial Cells (HDMEC) (9–11).
Endothelial cells are also available commercially from compa-
nies such as Cambrex (East Rutherford, NJ) (see Note 3).
Grow to tight confluence (see Note 4).
2. Remove the medium and wash cells once with approximately
2 ml of M199 for 35 mm dishes, or 100 μl for ibidi
chambers.
3. Stimulate cells in 1.5 ml (100 μl for ibidi chambers) of M199
with 0.5% HSA containing cytokine(s) of interest (see Note 5)
at 37°C and 5% CO2. If the stimulation time is 4 h or less, cells
can be stimulated in HBSS with 0.5% HSA instead of M199
with 0.5% HSA.
Transduction ibidi chambers as described for HUVEC (see Subheading 3.1.1
and Note 3). Grow to 80–90% confluence (see Note 4).
2. Remove the growing medium and keep it in a tube labeled
depleted medium. Meanwhile, wash the HUVEC two times
with approximately 1 ml (100 μl for ibidi chambers) of M199
with 0.5% HSA.
3. Add 1 ml (100 μl for ibidi chambers) of M199 with 0.5% HSA
on each 35-mm petri dish and add the appropriate quantity of
adenovirus construct (see Note 2).
4. Incubate HUVEC at 37°C and 5% CO2 for 3 h.
5. Add 1 ml of 37°C depleted medium on each 35-mm dish
(100 μl for ibidi chambers) without removing the M199 and
0.5% HSA with adenoviral construction. HUVEC must be
incubated overnight at 37°C and 5% CO2, and then medium
must be replaced with 2 ml (100 μl for ibidi chambers) of
ECM. HUVEC can be used between 24 and 48 h after adding
the adenoviral construct.
siRNA Transfection ibidi chambers as described for HUVEC (see Subheading 1.1
and Note 3). Grow to 80–90% confluence (see Note 4).
2. Aspirate medium off of HUVEC, and add 500 μl of warmed
Opti-MEM medium for 35 mm dishes, or 50 μl for ibidi
chambers.
3. Prepare siRNA complex by adding 3 μl of 2 μM siRNA and

6 μl of HiPerFect reagent for every 100 μl of Opti-MEM
medium in a 1.5 ml Eppendorf tube. Mix by pipetting up and
down ten times, and incubate at room temperature for
5–10 min.
4. For 35 mm dishes, add 200 μl of the complex drop-wise onto
the cells, and gently shake plate back and forth, and forward
and backward to ensure uniform distribution on the cells. For
ibidi chambers, add 15 μl of the complex, and mix by rocking
the ibidi chamber back and forth ten times.
5. Incubate the cells with the transfection complexes for 4 h at
37°C and 5% CO2.
6. Do not remove the depleted medium. Add 1 ml of fresh ECM
(50 μl for ibidi chambers) and mix by gently swirling, or rock-
ing back and forth. Incubate overnight at 37°C and 5% CO2.
7. The next morning, aspirate the old medium and add fresh
ECM; 2 ml for 35 mm dishes, and 100 μl for ibidi chambers.
Incubate overnight at 37°C and 5% CO2.
8. The cells can be used for experimentation 48 h after transfec-
tion (see Note 6).
3.4. Whole Blood 1. For whole blood recruitment experiments, draw blood from
Preparation and human donors into a syringe containing heparin (10 U hepa-
Leukocyte Isolation rin/ml blood). Gently invert the tube two to three times to
ensure that the blood and heparin are mixed. Prepare periph-
eral blood smears and stain with Wright-Giemsa stain accord-
ing to manufacturer’s instructions, in order to facilitate the
differentiation between the different types of blood cells.
Whole blood should be used within 90 min of blood draw.
2. For isolated leukocyte recruitment experiments, isolate leuko-
cytes as previously described for T lymphocytes (12), B lym-
phocytes (13), monocytes (14), neutrophils (6), and eosinophils
(15). Resuspend isolated leukocytes in 37°C HBSS containing
0.5% human serum albumin. In previous studies, isolated leu-
kocytes have been used at concentrations ranging from 0.5 to
1.5 × 106 cells/ml (12, 16, 17).
3.5. Flow Chamber 1. Connect a phase-contrast microscope to a CCD camera and a

Setup DVD recorder according to the manufacturer’s instructions.
Alternatively, software packages are available to record directly
3.5.1. Glycotech Chamber
to a computer, without the need for a DVD recorder.
(See Note 7 and Fig. 1)
2. Clean the flow chamber and gasket with isopropanol swabs.
Assemble the flow chamber according to the manufacturer’s
instructions, using an empty 35-mm culture dish as the bot-
tom plate. Place the flow chamber in the microscope stage.
3. Pull back on the plunger of a 30-ml syringe to break the seal.

Attach a 2-way stopcock to the syringe. Load the syringe into
an infuse/refill syringe pump. Program the syringe pump for
the desired refill rate (see Note 8) according to the manufac-
turer’s instructions.
4. Connect the 2-way stopcock to the outlet fitting of the flow
chamber using a piece of Pharmed tubing (see Note 9).
Connect a vacuum flask and vacuum pump to the vacuum
fitting of the flow chamber using a second piece of Pharmed
tubing. Attach a third piece of Pharmed tubing to the inlet
fitting of the flow chamber, and place the free end of this piece
into a 50-ml Falcon tube containing approximately 40 ml of
HBSS. Replenish this HBSS as needed during experiments.
Place the Falcon tube containing HBSS into a 37°C water
bath. Place a second 50-ml Falcon tube containing the whole
blood or isolated leukocyte suspension into the same water
bath.
5. Turn on the vacuum pump. Manually pull on the pusher block
of the syringe pump to fill the inlet line and flow chamber with
HBSS. Ensure that there are no air bubbles in the inlet line or
flow chamber. Close the 2-way stopcock and clamp the inlet
line with a hemostat.
6. Remove the empty culture dish and replace it with a culture
dish containing endothelial cells (see Note 10). To do this,
invert the flow chamber, cover the open section of the gasket
with HBSS, and lower the new culture dish onto the flow
chamber. Alternatively, replace the culture dish by lowering the
flow chamber onto the new culture dish.
3.5.2. ibidi Chamber (See 1. Same as step 1 in Subheading 3.5.1.

Note 7 and Fig. 1) 2. Assemble the flow chamber according to the manufacturer’s
instructions. A male luer adaptor is used to connect the
Pharmed tubing to the inlet and outlet wells of the ibidi
chamber.
3. Same as step 3 in Subheading 3.5.1.
4. Connect the 2-way stopcock to the outlet fitting of the flow
chamber using a piece of Pharmed tubing. Attach a second
piece of Pharmed tubing to the inlet fitting of the flow cham-
ber and place the free end of this into a 50 ml Falcon tube
containing approximately 40 ml of HBSS. Replenish this HBSS
as needed during experiments. Place the Falcon tube contain-
ing HBSS into a 37°C water bath. Place a second 50 ml Falcon
tube containing the whole blood or isolated leukocyte suspen-
sion into the same water bath.
5. Turn on the syringe pump, and watch the chamber until no
more air bubbles can be observed to pass through. Close the
2-way stopcock and clamp the inlet line with a hemostat (see
Note 10).
3.6. Whole Blood 1. Visualize the endothelial cell monolayer at 200× magnification
Recruitment using bright-field optics.
Experiment 2. Place the inlet line into the whole blood. Open the 2-way stop-
cock and unclamp the hemostat from the inlet line. Start the
syringe pump and pull blood into the flow chamber for 5 min
(see Note 11).
3. Switch the inlet line into the HBSS (see Note 12). Buffer will
begin to enter the chamber soon after the inlet line is switched
into the HBSS; the time required for this to happen will depend
on the length of the inlet line and the flow rate used.
4. After buffer has entered the chamber and enough blood has
cleared from the field to permit visualization of accumulated cells,
begin recording fields using the CCD camera and DVD recorder
(see Note 13). Record four random fields for 15 s each.
5. For 35 mm dishes: Hold the flow chamber at a 45 to 90° angle
relative to the stage and remove the inlet line from the flow
chamber. Allow air to flow over the monolayer until it has dis-
placed all of the buffer in the flow chamber. Remove the cul-
ture dish from the flow chamber, holding the dish at a 45 to
90° angle relative to the stage to prevent any residual buffer
from flowing back over the monolayer. For ibidi chambers:
Remove the inlet and outlet lines from the chamber. Hold the
chamber at a 45° angle, with the outlet chamber lower than
the inlet. Remove the medium using a pipette.
6. Wright-Giemsa stain the culture dish according to manufac-
turer’s instructions.
3.7. Whole Blood 1. The four fields that were recorded in step 4 of Subheading 3.6
Recruitment Analysis can be used to measure total leukocyte recruitment, the per-
(See Notes 14 and 15) centage of firmly adherent or rolling leukocytes, and the roll-
ing velocities of leukocytes.
2. To measure total leukocyte recruitment, count the total num-
ber of cells in all of the recorded fields. Divide the total by the
number of recorded fields to give the average number of cells
per field. Divide the average number of cells per field by the
area of the field to give the average number of cells/mm2 (see
Note 16).
3. To measure the percentage of leukocytes that are firmly adher-
ent, count the number of firmly adherent cells in all of the four
fields. We define firmly adherent cells as those that moved less
than one cell diameter in a 10 s period. Divide the number of
firmly adherent cells by the total number of cells (measured in
step 2) and multiply by 100%.
4. To measure the percentage of leukocytes that are rolling, count

the number of rolling cells in all of the four fields. We define
rolling cells as those that moved one cell diameter or more in a
10 s period. Divide the number of rolling cells by the total
number of cells (measured in step 2) and multiply by 100%.
The percentage of leukocytes that are firmly adherent (mea-
sured in step 3) and the percentage of leukocytes that are roll-
ing should add up to 100%.
5. To measure the rolling velocity of a leukocyte, select a leuko-
cyte and measure the distance that it traveled in a 10 s period.
Divide this distance by 10 to give the distance traveled per
second. We usually measure all of the cells that are present on
the endothelial cell monolayer and represent the data using a
histogram of rolling velocities.
6. Perform a 200-cell differential on the culture dish that was
stained in step 6 of Subheading 3.6 (see Note 17).
7. Perform a 200-cell whole blood differential on the peripheral
blood smears that were prepared in step 1 of Subheading 3.4.
8. Calculate a recruitment factor (R-factor) for each leukocyte
subclass. The R-factor for a given leukocyte subclass is calcu-
lated by dividing the percentage of leukocytes on the plate that
are of that subclass by the percentage of leukocytes in the whole
blood differential that are of that subclass (see Note 18).
3.8. Isolated Leukocyte 1. Visualize the endothelial cell monolayer at 100× magnification
Recruitment using phase-contrast optics.
Experiment 2. Place the inlet line into the isolated leukocytes. Open the 2-way
stopcock and unclamp the hemostat from the inlet line. Start
the syringe pump.
3. When leukocytes enter the chamber, begin recording a single
field using the CCD camera and DVD recorder. Record this
field for 1 min. Wait for 3 min.
4. Switch the inlet line into the HBSS (see Note 12). Record six
random fields for 10 s each (see Note 19).
5. Switch to 400× magnification and wait for 1 min. Scan the
monolayer to find groups of cells. Record multiple groups of
cells while focusing up and down on the monolayer for a total
of 1 min.
6. Analysis of the data is carried out as described below in
Subheading 3.9.
3.9. Isolated Leukocyte 1. The six fields that were recorded in step 4 of Subheading 3.8
Recruitment Analysis can be used to measure total leukocyte recruitment, the num-
(See Note 14) ber of firmly adherent or rolling leukocytes, and the rolling
velocity of leukocytes. The methods for measuring these

parameters are described in Subheading 3.7.
2. The single field that was recorded in step 3 of Subheading 3.8
can be used to measure leukocyte tethering. Analyze this
recording frame-by-frame for primary tethers, secondary teth-
ers, and leukocyte-leukocyte interactions. We define primary
tethers as direct tethers between a flowing leukocyte and the
endothelial cell monolayer; secondary tethers as tethers in
which a flowing leukocyte makes contact with an adherent leu-
kocyte prior to attaching to the endothelial cell monolayer; and
leukocyte–leukocyte interactions as direct tethers between a
flowing leukocyte and an adherent leukocyte. We express the
data as the number of each of primary tethers, secondary teth-
ers, and leukocyte–leukocyte interactions per mm2 per minute.
3. The fields that were recorded in step 5 of Subheading 3.8 can
be used to measure the percentage of leukocytes that have
transmigrated (see Note 20). Count the total number of each
of transmigrated and non-transmigrated cells in all of the
recorded fields. Divide the number of transmigrated cells by
the total number of cells (transmigrated and non-transmi-
grated) and multiply by 100%.
4. Notes
1. Each 250 ml batch of ECM requires 50 ml of serum pooled

from at least three donors (approximately 17.0 ml of serum
each). Blood is collected into Vacutainer® Serum Separation
(SST) tubes (Becton Dickinson), and each 8.5 ml SST tube
yields about 4 ml of serum, with variation depending on the
donor’s hematocrit. Therefore, to obtain about 17.0 ml of
serum, draw four to five SST tubes. As each SST tube is filled,
gently invert it three times to aid clot formation. Allow the
blood to coagulate for a minimum of 15 min. Centrifuge the
tubes for 15 min at 500 ´ g. After centrifugation, inspect the
tubes to determine if a complete barrier has formed between
the coagulum and serum. In a tissue culture hood, pour off the
serum (17 ml from each of three donors) into 50 ml Falcon
tubes, and freeze at −20°C. Freeze any extra serum as it can be
pooled later with other donors. A similar process has been
described elsewhere (6). Once made, ECM is stable for approx-
imately 2 weeks at 4°C.
2. For our experiments, we produced virus using the ViraPower™
Adenoviral expression system from Invitrogen according to the
manufacturer’s instructions. Appropriate adenoviral construct
quantity for efficient HUVEC transduction depends on the
adenovirus, the gene of interest and the construct. This quan-

tity must be determined by using serial dilution and assessing
the percentage of cells stained positive for the protein of interest
by flow cytometry, as well as assessing the shape and the general
health of the cells by microscopy. The transduction control can
be adenoviral DNA without the gene of interest, and the aden-
ovirus buffer alone (dialysis buffer: 150 mM NaCl, 10 mM
Tris–HCl, 2 mM MgCl2 3% sucrose, adjust pH to 7.4).
3. Although endothelial cells can be purchased from commercial
sources, we isolate our own endothelial cells for use in our
laboratory and recommend freshly isolated endothelial cells
over commercially available endothelial cells, especially when
experiments require primary or first passage cells.
4. Many protocols for culturing endothelial cells recommend
using fetal bovine serum in the culture medium; however, our
laboratory has found that endothelial cells grow better in
medium containing human serum than in medium containing
fetal bovine serum.
5. The cytokines tumor necrosis factor-α (TNF-α), interleu-
kin-1® (IL-1®), and IL-4 are frequently used to stimulate
human endothelial cells for leukocyte recruitment assays using
parallel plate flow chambers. The concentrations and incuba-
tion periods that are used for these cytokines vary greatly. We
present here some examples of concentrations and incubation
periods that have been used for these cytokines. These exam-
ples are by no means exhaustive; we recommend that you con-
sult the relevant literature to determine the concentration and
incubation period that should be used in your experimental
setup. TNF has been used at concentrations between 10 and
20 ng/ml (15, 18, 19) and for incubation periods between 6
and 24 h (15, 18–21). IL-1® has been used at concentrations
between 0.1 and 50 ng/ml (22–24) and for incubation peri-
ods between 4 and 24 h (22, 23, 25). IL-4 has been used at
concentrations between 10 and 20 ng/ml and for incubation
periods between 24 and 48 h (16, 19, 21, 26).
6. The appropriate amount of time required for maximal knock-
down of your protein of interest must be determined empiri-
cally. For our proteins of interest, we determined this to be
48 h post-transfection, but this timeline may be different for
different proteins and different siRNAs.
7. These protocols are written for use with circular flow chambers
(for 35-mm culture dishes), which are commercially available
through GlycoTech (www.glycotech.com), or for ibidi μ-slide
VI flow chambers, which are commercially available through
various distributors listed at www.ibidi.de.
8. The relationship between the wall shear stress in a flow cham-
ber setup and the flow rate through the chamber is given by
the equation τW = μγ = 6μQ/a2b, where τW is the wall shear stress

(dynes/cm2), μ is the viscosity of the medium (P), γ is the shear
rate (s−1), Q is the volumetric flow rate (ml/s), a is the channel
height (gasket thickness) (cm), and b is the channel width (gas-
ket width) (cm). The viscosity of whole blood varies between
donors, thus we use shear rates rather than wall shear stresses
for whole blood recruitment experiments. In previous studies,
leukocyte recruitment from whole blood has been examined at
shear rates ranging from 50 to 400 s−1 (19). For isolated leuko-
cyte experiments, the shear stress can be calculated using 1 cP
(0.01 P) as the viscosity of the cell suspension. In previous
studies, isolated leukocyte recruitment has been examined at
shear stresses ranging from 0.5 to 4 dyn/cm2 (16, 17, 27).
9. It is not necessary to use a specific length of tubing for the
outlet, vacuum, or inlet line; instead, these lengths should be
chosen to facilitate movement of the flow chamber in a par-
ticular experimental setup. The length of tubing used for the
inlet line, however, should be kept constant both within and
between experiments. A fresh inlet line should be used for each
new experiment or donor and the inlet line should be switched
after using blood or a leukocyte suspension that has been
treated with an antibody or inhibitor.
10. It is very important to prevent the introduction of air into the
flow chamber setup at any stage in an experiment, as endothe-
lial cells may become activated or damaged if air is perfused
over them (28). Before using the flow chamber for the first
time, we recommend practicing placing empty culture dishes
onto the flow chamber until you can do this without introduc-
ing air into the system. For ibidi chambers, use an old chamber
to first remove all air out of the system, and then transfer inlet
and outlet lines to the chamber containing the adhesive sub-
strate of interest in the following order: (1) Close 2-way stop-
cock and clamp inlet line with hemostats. (2) Transfer outlet
line from practice chamber onto the outlet well of the new
chamber of interest. (3) Remove inlet line from old chamber,
and hold over the inlet well of new chamber. (4) Unclamp
hemostat and slowly let liquid drip from the inlet line into the
inlet well of the new chamber, until well is just overflowing. (5)
Insert inlet line into the inlet well of the new chamber, and
wipe away excess liquid.
11. The specific perfusion times used in whole blood and isolated
leukocyte recruitment assays vary between laboratories. The
times that are given in these protocols are the times that we use
in our laboratory.
12. When switching the inlet line during an experiment, close the
2-way stopcock to prevent air from being pulled into the inlet
line and work quickly to prevent pressure from building up in

the system. This should take less than 5 s.
13. The amount of time required for the field to clear after the HBSS
has entered the chamber may vary, but usually ranges from 30 s
to 2 min in our system. For consistency, we recommend that
you select an amount of time to wait between the time that the
HBSS enters the chamber and the time that you begin recording
accumulated cells and use this in all experiments.
14. There are a number of different computer programs that can
facilitate the analysis of whole blood recruitment and leukocyte
recruitment assays. We analyze our experiments using NIH
ImageJ and Volocity. NIH ImageJ works on most operating
systems, and is available for download at rsbweb.nih.gov/ij/.
Volocity software is available commercially through Perkin
Elmer Inc.
15. The analysis of whole blood rolling experiments is difficult due
to the presence of red blood cells, which can obscure the field
of view. This analysis can be particularly difficult for experi-
ments in which endothelial cells are used as a substrate. As a
result, you may be unable to perform some of the analysis that
is described here, such as the measurement of rolling velocities,
on whole blood rolling experiments in which endothelial cells
are used as a substrate.
16. An object of known dimension, such as a hemocytometer, can
be used to measure the dimensions of a field.
17. It may be difficult or impossible to differentiate between lym-
phocytes and monocytes that have interacted with endothelial
cells using only Wright-Giemsa staining. We do not attempt to
differentiate between these two leukocyte subclasses in whole
blood recruitment experiments, but instead classify them all as
peripheral blood mononuclear cells (PBMC).
18. For a given leukocyte subclass, an R-factor of 1 indicates that
there is no selective or preferential recruitment of that subclass.
An R-factor of less than 1 indicates that there is selectivity
against that subclass, while an R-factor of greater than 1 indi-
cates that there is selectivity for that subclass.
19. Leukocytes may become activated and transmigrate across the
endothelium, making them difficult to identify at 100×
magnification. If this occurs, the 10-s fields can be recorded at
200× magnification.
20. Transmigrated cells will generally appear as flattened, phase-
dark cells with irregular edges and will be below the focal plane
of the endothelial cell monolayer. When measuring transmigra-
tion for the first time, we recommend that you carefully exam-
ine the interactions between leukocytes and endothelial cells at
800 to 1000× magnification while focusing up and down on the
monolayer so as to learn to differentiate between activated and

transmigrated cells. Photographs and supplementary videos of
transmigrating leukocytes can be found in articles on lympho-
cyte and eosinophil transmigration (15); these images may assist
with the identification of transmigrated leukocytes.
References
1. Kubes P (2002) The complexities of leukocyte activates alpha L beta 2 integrin-mediated adhe-
recruitment. Semin Immunol 14:65–72 sion to ICAM-1 in human T cells. J Immunol
2. Ley K, Laudanna C, Cybulsky MI, Nourshargh 164:746–753
S (2007) Getting to the site of inflammation: 13. Yago T, Tsukuda M, Tajima H, Nishi T,
the leukocyte adhesion cascade updated. Nat Kurata-Miura K, Ohkubo J, Minami M (1997)
Rev Immunol 7:678–689 Analysis of initial attachment of B cells to
3. Lawrence MB, Springer TA (1991) Leukocytes endothelial cells under flow conditions. J
roll on a selectin at physiologic flow rates: dis- Immunol 158:707–714
tinction from and prerequisite for adhesion 14. Luscinskas FW, Kansas GS, Ding H, Pizcueta
through integrins. Cell 65:859–873 P, Schleiffenbaum BE, Tedder TF, Gimbrone
4. Lawrence MB, McIntire LV, Eskin SG (1987) MA Jr (1994) Monocyte rolling, arrest and
Effect of flow on polymorphonuclear leuko- spreading on IL-4-activated vascular endothe-
cyte/endothelial cell adhesion. Blood lium under flow is mediated via sequential
70:1284–1290 action of L-selectin, beta 1-integrins, and beta
5. Forrester JV, Lackie JM (1984) Adhesion of 2-integrins. J Cell Biol 125:1417–1427
neutrophil leucocytes under conditions of flow. 15. Cuvelier SL, Patel KD (2001) Shear-dependent
J Cell Sci 70:93–110 eosinophil transmigration on interleukin
6. Zimmerman GA, McIntyre TM, Prescott SM 4-stimulated endothelial cells: a role for
(1985) Thrombin stimulates the adherence of endothelium-associated eotaxin-3. J Exp Med
neutrophils to human endothelial cells in vitro. 194:1699–1709
J Clin Invest 76:2235–2246 16. Patel KD (1998) Eosinophil tethering to inter-
7. Jaffe EA, Nachman RL, Becker CG, Minick leukin-4-activated endothelial cells requires
CR (1973) Culture of human endothelial cells both P-selectin and vascular cell adhesion mol-
derived from umbilical veins. Identification by ecule-1. Blood 92:3904–3911
morphologic and immunologic criteria. J Clin 17. Ostrovsky L, King AJ, Bond S, Mitchell D,
Invest 52:2745–2756 Lorant DE, Zimmerman GA, Larsen R, Niu
8. Hewett PW, Murray JC (1993) Human lung XF, Kubes P (1998) A juxtacrine mechanism
microvessel endothelial cells: isolation, culture, for neutrophil adhesion on platelets involves
and characterization. Microvasc Res 46:89–102 platelet-activating factor and a selectin-depen-
dent activation process. Blood 91:3028–3036
9. Richard L, Velasco P, Detmar M (1998) A
simple immunomagnetic protocol for the 18. Fitzhugh DJ, Naik S, Caughman SW, Hwang
selective isolation and long-term culture of ST (2000) Cutting edge: C-C chemokine recep-
human dermal microvascular endothelial cells. tor 6 is essential for arrest of a subset of memory
Exp Cell Res 240:1–6 T cells on activated dermal microvascular
endothelial cells under physiologic flow condi-
10. Marks RM, Czerniecki M, Penny R (1985) tions in vitro. J Immunol 165:6677–6681
Human dermal microvascular endothelial cells:
an improved method for tissue culture and a 19. Patel KD (1999) Mechanisms of selective leu-
description of some singular properties in cul- kocyte recruitment from whole blood on
ture. In Vitro Cell Dev Biol 21:627–635 cytokine-activated endothelial cells under flow
conditions. J Immunol 162:6209–6216
11. Gupta K, Ramakrishnan S, Browne PV, Solovey
A, Hebbel RP (1997) A novel technique for 20. Ulfman LH, Joosten DP, van der Linden JA,
culture of human dermal microvascular Lammers JW, Zwaginga JJ, Koenderman L
endothelial cells under either serum-free or (2001) IL-8 induces a transient arrest of roll-
serum-supplemented conditions: isolation by ing eosinophils on human endothelial cells. J
panning and stimulation with vascular endothe- Immunol 166:588–595
lial growth factor. Exp Cell Res 230:244–251 21. Kitayama J, Mackay CR, Ponath PD, Springer
12. Chan JR, Hyduk SJ, Cybulsky MI (2000) TA (1998) The C-C chemokine receptor
Alpha 4 beta 1 integrin/VCAM-1 interaction CCR3 participates in stimulation of eosinophil
arrest on inflammatory endothelium in shear Ballantyne CM (1998) Soluble cell adhesion

flow. J Clin Invest 101:2017–2024 molecules in hypertriglyceridemia and poten-
22. Kukreti S, Konstantopoulos K, Smith CW, tial significance on monocyte adhesion.
McIntire LV (1997) Molecular mechanisms of Arterioscler Thromb Vasc Biol 18:723–731
monocyte adhesion to interleukin-1beta-stim- 26. Cuvelier SL, Paul S, Shariat N, Colarusso P,
ulated endothelial cells under physiologic flow Patel KD (2005) Eosinophil adhesion under
conditions. Blood 89:4104–4111 flow conditions activates mechanosensitive sig-
23. von Hundelshausen P, Weber KS, Huo Y, naling pathways in human endothelial cells. J
Proudfoot AE, Nelson PJ, Ley K, Weber C Exp Med 202:865–876
(2001) RANTES deposition by platelets trig- 27. Phillipson M, Heit B, Parsons SA, Petri B,
gers monocyte arrest on inflamed and athero- Mullaly SC, Colarusso P, Gower RM, Neely G,
sclerotic endothelium. Circulation Simon SI, Kubes P (2009) Vav1 is essential for
103:1772–1777 mechanotactic crawling and migration of neu-
24. Wiese G, Barthel SR, Dimitroff CJ (2009) trophils out of the inflamed microvasculature.
Analysis of physiologic E-selectin-mediated J Immunol 182:6870–6878
leukocyte rolling on microvascular endothe- 28. Cinamon G, Shinder V, Alon R (2001) Shear
lium. J Vis Exp 24:pii–1009 forces promote lymphocyte migration across
25. Abe Y, El-Masri B, Kimball KT, Pownall H, vascular endothelium bearing apical chemok-
Reilly CF, Osmundsen K, Smith CW, ines. Nat Immunol 2:515–522
Chapter 19
Generation and Establishment of Murine Adherent Cell Lines

Rouzanna Istvanffy and Robert A.J. Oostendorp
Abstract
We describe a method to derive cell lines and clones from cells of the murine midgestation aorta-gonads-
mesonephros (AGM) microenvironment. We start from subdissected AGM regions in “explant” or “single
cell suspension” type cultures from embryos transgenic for tsA58, a temperature-sensitive mutant of the
SV40 T antigen gene. The number of cells in such cultures initially expand, but in most cases, this expan-
sion phase is followed by a stable or even decline in cell number. After this so-called crisis phase, cell pro-
liferation is noticeable in more than 90% of the cultures. Stromal cell clones can be isolated from these
cultures, some of which have been cultured for more than 50 population doublings, and functionally char-
acterized using various methods These stromal cell clones are valuable tools for the study of the regulation
of hematopoietic stem and progenitor cells in the midgestation mouse embryo.
Key words: Aorta-gonads-mesonephros, AGM, Hematopoietic stem cells, Stromal cell lines, tsA58
mutants
1. Introduction
The differentiation of progenitor and stem cells of many tissues

depends on their interactions with surrounding cells which form a
cellular unit: the niche, which includes mesenchymal, vascular and
other, tissue-specific cells. Our understanding of the molecular
mechanisms governing development, self-renewal, and differentia-
tion of stem cells has improved over many years through the wide-
spread use of cell line models of the niche, or microenvironment.
Such cells have been isolated from tumors, spontaneous immortal-
ized variants of normal cells or from primary cells transduced with
genes facilitating unlimited growth (immortalizing genes) (1–3).
Central to the use of cell lines in the study of cellular differentiation
and development is the assumption that they are representative of
301
302 R. Istvanffy and R.A.J. Oostendorp
cells that function within the normal cellular physiology of the

organism. However, the methodology required for their isolation
and growth necessitates extended cultivation periods or cultivation
conditions that may alter them (4).
1.1. Immortalizing The most commonly used immortalizing gene to generate cell lines
Genes is that encoding the SV40 large T antigen (TAg). In addition,
investigators have used ectopic expression of the catalytic compo-
nent of telomerase gene (Tert, 5) and Tp53-deficient cells (6, 7) to
generate cell lines. It is important to note that expression of one
immortalizing gene does not suffice to transform cells, but that
additional gene mutations are required (8). The conditionally
active form of the TAg gene, tsA58, produces a thermolabile pro-
tein that is active at 33°C (9). Most often, the TAg or the tsA58
gene has been introduced into cells via retroviral-mediated trans-
duction by the co-cultivation of target cells with virus-producing
feeder layers (10). However, this method of gene transduction
requires that the cells of interest are dividing in order to achieve the
integration of the provirus and immortalizing gene DNA sequences
into the cellular genome, and subsequent gene expression. The
extended cultivation period necessary to allow cell proliferation,
integration and drug selection of the transduced cells may alter or
exclude the physiologically relevant cells. Hence, we and others
have generated transgenic mice expressing the immortalizing genes
(TAg, tsA58, hTERT) or deleted Tp53 to alleviate such problems
by allowing for the immediate expression upon plating cells in vitro,
without requirement for previous proliferation or selection steps.
An additional advantage of using temperature-sensitive mutants
such as tsA58 is that they proliferate at the activating temperature
(33°C) and usually stop proliferation and differentiation at the
non-permissive temperature (37–39°C) (11, 12).
1.2. The Hematopoietic Our own work focuses on the study of the influence of the
Microenvironment hematopoietic niche on the development, expansion and differen-
tiation of hematopoietic stem cells (HSC). This microenvironment
is composed of stromal cells that interact and regulate the hierar-
chy of HSC, progenitors, committed cells and functional circulat-
ing blood cells (13). Stromal cells within the context of the bone
marrow and fetal liver are thought to maintain and support
hematopoiesis throughout adult and fetal stages, respectively (14).
It has also become clear that disturbances of proper microenviron-
mental function can contribute to the development and progres-
sion of hematopoietic diseases (15, 16). Thus, stromal cells may be
an additional target in the therapy of leukemia. Our goal was to
establish stromal cell lines as a model of hematopoietic niche, to
identify which factors are produced by the microenvironment that
maintain normal HSC and how these factors affect malignantly
transformed hematopoietic cells.
19 Stromal Cell Lines Derived from Mouse AGM 303
To facilitate the isolation of cell lines representative of the

in vivo hematopoietic microenvironments present in the midgesta-
tion embryo, as well as to isolate cell lines from other tissues of the
embryo and adult, we generated transgenic mouse lines that express
the thermolabile tsA58 gene in a constitutive and ubiquitous man-
ner (17). In this chapter, we describe methods to derive stromal
cell lines and clones from cells of the murine midgestation aorta-
gonads-mesonephros (AGM) microenvironment.
Using the protocols described, cell lines from wild-type and
transgene-expressing mouse bone marrow, spleen, liver, and thy-
mus tissue, as well as embryonic liver, gastrointestinal tissue, as
well as androgen-responsive Vas Deferens cell lines (18) have been
generated.
Below, we describe not only the techniques we used to gener-
ate cell lines but also some of the techniques we have used to study
HSC maintenance on these lines (19, 20). In addition, we refer the
reader to reports in which the use of these cell lines has been used
to generate natural killer cells (21) as well as HSC from both
murine (22) and human (23) embryonic stem cells.
2. Materials
2.1. Generation of 1. Alpha MEM: Alpha MEM is from Gibco-Invitrogen with

Stromal Cell Lines added Glutamax I to achieve 1× according to Supplier instruc-
tions (Invitrogen # 32571-028). Store medium at 2–8°C.
2. Conditioned medium: Conditioned medium (CM) is prepared
from each passage of the developing cell lines. The CM is
collected in conical tubes and spun at 2500 ´ g for 7 min to
remove debris and contaminating cells. Larger samples (>1 ml)
and CM used for cloning is additionally 0.2 μm-filtered using
a syringe or bottle-top filter (for example Millipore Millex-GV
filters). Store at 2–8°C for up to 2 weeks or for longer periods
3. Freeze medium: 90% FBS and 10% DMSO (Sigma, # D-5879).
Prepare just before use.
4. 0.1% Gelatin. A suspension of 0.4 g gelatin powder (Sigma-
Aldrich # G-9391) in 400 ml distilled water in a 500 ml bottle
(loose cap) is autoclaved. The gelatin is now dissolved and ster-
ile. Store this solution at 4°C or room temperature. If you do
not culture cells often, make smaller aliquots (100 or 200 ml).
5. Glutamax-I (Gibco-Invitrogen, # 35050-038). Store in ali-
quots at −20°C.
6. Hydrocortisone (Sigma-Aldrich, H0888). Store in aliquots at
−20°C.
7. HF2+ buffer contains 2% FBS, 10 mM HEPES (Gibco-

Invitrogen, # 15630049), antibiotics (penicillin and streptomy-
cin; Gibco-Invitrogen, # 15140-122) in HBSS (Gibco-Invitrogen,
# 14065-049). Store at 2–8°C.
8. Lineage separation. We have used two kits with similar results:
The Lineage Cell Depletion Kit (Milteney Biotec, # 130-090-
858) and the Mouse Hematopoietic Progenitor Cell
Enrichment Kit (STEMCELL Technologies, # 13046). Both
kits require additional magnetic columns and magnets for per-
formance of lineage depletion.
9. Fetal bovine serum (FBS) (see Note 1).
10. Long-term culture medium: MyeloCult M5300 (STEMCELL
Technologies, # 05300). Store in aliquots at −20°C.
11. Methylcellulose medium with recombinant cytokines for
mouse cells (STEMCELL Technologies, # 03434).
12. Stroma Medium: The Stroma Medium contains 50% long-term
culture medium, 15% FBS (see Note 1), 35% Alpha MEM, anti-
biotics (penicillin and streptomycin; Gibco-Invitrogen, #
15140-122), and 10 μM β-mercaptoethanol (Sigma-Aldrich, #
M-7522). Filter medium using 0.2 μm-filtered (Millipore,
SCGPT05RE bottle-top filters) to remove debris and other par-
ticles which could stimulate phagocytosis and promote stromal
cell differentiation and senescence. Store in aliquots at −20°C.
13. 0.5% Trypsin. (Gibco-Invitrogen, # 15400054). Store in ali-
quots at −20°C.
14. 0.4% Trypan blue for viable cell counts.
15. Cultureware: (Costar) 94, 60, and 35 mm tissue-culture treated
dishes and, 48- and 24-well tissue-culture treated plates.
3. Methods
The methods below outline the different steps in establishing a

new cell line and its use as microenvironmental models for func-
tional assays in vitro. (1) Choice of mouse strain, (2) isolation of
primary cells, (3) growth of primary cells until growth crisis, (4)
growth of cells after growth crisis and cloning, (5) characterization
of isolated stromal cell clones, and (6) cocultures of lineage nega-
tive murine bone marrow cells on stromal cell lines.
3.1. Choice of Mouse To generate cell lines, one should consider the mouse strain that
Strain will be used. For our experiments, we developed transgenic mouse
strains expressing tsA58 under the control of the β-actin (TAg05)
and phosphoglycerate kinase-1 (TAg11) promoters (17). Cell lines
can also be generated from other mouse strains (see Note 2).
Animals should be housed according to institutional guidelines,
with free access to food and water. Animal procedures should be
carried out in compliance with the Standards for Humane Care
and Use of Laboratory Animals.
3.2. Isolation of AGM, and subdissected tissues were obtained from E10 and E11
Primary Cells embryos as described in detail elsewhere (24).
3.3. Primary Explant Throughout this procedure, cells are cultured on 0.1% gelatin-
and Single Cell coated tissue culture plates. Culture vessels are coated with 0.1%
Cultures gelatin (100 μl/cm2) either at 37°C (for at least 1 h) or at 4°C
(overnight) with similar results. The plates can be stored at 4° for
up to 1 week. Prior to cell seeding, the excess 0.1% gelatin solution
is washed off and the vessel washed once with PBS. Once washed,
these vessels should be used immediately.
Since optimal growth conditions were unknown for AGM
stromal cells (see Note 3), we chose to culture the subdissected tis-
sues on 0.1%-gelatin coated 24-well plates in either long-term cul-
ture medium or Stroma Medium at 33°C (permissive temperature
for tsA58), 5% CO2, and greater than 95% humidity (see Note 4)
using both explant and single cell culture methods. Both the “explant”
and the “single cell suspension” methods yield stromal cell lines.
3.3.1. Explant Cultures In this type of culture, the tissue of interest is cultured as a whole
and the stromal cells are allowed to migrate and grow out of the
intact tissue. Isolated tissues are cultured at the air–medium inter-
face on 24-well plates (one tissue piece per well) with a minimal
amount of Stroma MediumMedium (100 μl/cm2 of culture area).
Thus, the tissue is in contact with the gelatin-coated cultureware.
Tissues will attach to the plastic cultureware surface, and at the
same time, fibroblastoid cells can be seen to migrate out of the tis-
sue (Fig. 1).
Fig. 1. Outgrowth of fibroblastoid cells 4 days after the start of primary cell culture. Cells from midgestation embryonic
tissues were cultured using the “explant” method on 0.1% gelatin-coated culture dishes. Shown are explants of embryonic
liver EL17 (a), aorta-mesenchyme AM20 (b), and urogenital ridges UG26 (c).
3.3.2. Single Cell Cultures Spin the isolated tissues at 400 × g for 5 min and then wash the tis-
sues once in serum-free AlphaMEM. Then, subject tissues to a
15 min incubation with 0.25% trypsin and gently spin at 400 × g for
10 min at room temperature. Resuspend in a small volume of
Stroma Mediumand vigorously pipette to dissociate remaining cell
clumps and obtain a single cell suspension. Count the number of
viable cells prior to plating using the trypan blue exclusion and a
Neubauer cytometer. The cell suspension is cultured on 24-well
plates in 300 μl Stroma Medium at a density of 105 cells per well
or, if less cells are available per tissue, one tissue per well. After a
day, single cells can be observed to be attached to the cultureware.
As an alternative to establishing cell lines from single embryos, tis-
sues from several embryos can be pooled, treated in the same man-
ner and cultured in 6-well plates.
3.3.3. Cell Culture Until Cultures are incubated at 33°C, 5% CO2, and greater than 95%
Growth Crisis humidity (see Note 4)
1. After 1 or 2 days, the first fibroblastoid cells can be seen to
grow out of the explanted tissues (Fig. 1). The explant-like
cultures are now topped of to a total volume of Stroma Medium
of 300 μl/cm2 of cultureware area.
2. After 2 to 3 more days the culture supernatant (conditioned
medium) is collected as described in Subheading 2.1. The
adherent cells (from explant and single cell cultures) are washed
once in AlphaMEM (no serum), and harvested by brief trypsin
exposure (not more than 10 min). Detached cells are collected
in polypropylene 15 ml tubes. The cells are then replated at a
density of 5 × 104 cells/cm2.
3. Since the growth factor requirements of the derived cell line is
often not known (see Note 3), the stromal medium is supple-
mented with 20% 0.2 μm-filtered CM from its own previous
passage as a source of autocrine growth factors for all the sub-
sequent culture steps.
4. In the first few passages, the total cell number will increase.
This is usually followed by a period of passages in which the
number of cells harvested is stable and then begins to be lower
than the number cells initially seeded (growth crisis). During
this phase, the cells are seeded in consecutively smaller culture
vessels (94 mm dish (70 cm2, 10 ml) → 60 mm dish (28 cm2,
4 ml) → 35 mm dish/6-well plate (both 10 cm2, 2 ml) → 24
well plate (2 cm2, 300 μl) → 96 well (0.8 cm2, 100 μl)) to main-
tain the number of cells at around 5 × 104 cells/cm2. This pro-
cedure facilitates cell–cell contact and allows for the sufficiently
high production of autocrine growth factors. Always add the
20% CM obtained from the previous passage. Alternatively, if
sufficient CM is not available, a 0.22 μm-filtered CM from a
semi-confluent cell line from the same tissue can be used as

growth supplement.
5. This procedure is repeated each week (regardless of whether
cell proliferation is observed) until a consistent increase in cells
is notable (see Note 5).
3.4. Culture of Cells The crisis period of cell senescence is usually followed (in 32 of 36
After Growth Crisis: cases in our hands) by outgrowth of cells. As soon as a cell line
Cloning shows consistent growth (see Note 5), cells are cloned at a density
of 1 cell per 300 μl per well in 0.1% gelatin-coated 24- or 48-well
plates. Cultures are incubated at 33°C, 5% CO2, and greater than
95% humidity (see Note 4).
1. Conditioned medium is prepared from the parental cell line.
The clones are grown on 0.1% gelatin-coated wells in Stroma
Medium supplemented with 30% 0.2 μm-filtered CM of the
parental cells. The cloning was more efficient when using 30%
instead of the usual 20% CM.
2. After 3 days, the wells are supplemented with 300 μl Stroma
Medium supplemented with 30% 0.2 μm-filtered CM of the
parental cell line.
3. The clones are maintained for 2–3 weeks with medium changes
every 3 or 4 days.
4. When individual wells are subconfluent (see Note 6), clones
are harvested by trypsin-treatment (first passage) and expanded
in larger culture vessels (100 mm dishes).
5. When these larger vessels are subconfluent (range, i.e.,
50–80%), again the cells are harvested by trypsin treatment
and an aliquot of the clones should be frozen at 3–5 × 105 cells
per vial in freeze medium. It is important to freeze cells at the
earliest stage, to ensure availability of low passage cells for
future use (see Notes 7 and 8).
6. The clones are propagated as 5 × 104 cells/100 mm dish and
passaged once a week, or more often if cells reach subconfluence
more quickly.
7. Clones generated in this manner can usually be cultured for
more than 50 passage doublings without any sign of cellular
senescence (see Fig. 2 and Note 9).
3.5. Functional The final phase in cell line development is to screen isolated cell
Characterization of clones on a functional level. The way to screen cell clones is differ-
Isolated Stromal Cell ent for each researcher and depends on the particular functional
Clones activity the clones were established for. In our case, we sought cell
lines which would support HSC in culture (17, 25). The method-
ologies for this screening have been described in detail elsewhere
(26, 27).
Fig. 2. Growth curves of the aorta-mesenchyme (AM)-derived AM14 and AM30 cell lines
and clones thereof. AM30 (open squares) was derived from a pool of eight embryos of a
TAg11 litter, whereas AM14 (open triangles) was derived from a “control” litter which did
not express the immortalizing tsA58 gene. Please note that the AM14 crisis period lasted
for about 8 weeks, whereas AM30 did not seem to show signs of a proliferation crisis.
AM14 and AM30 were cloned after nine and seven passages, respectively (arrows). Two
of the clones generated were followed for more than 50 population doublings after cloning
(AM14-1C4 (closed triangles) and AM30-3F5 (closed squares)) without any sign of cellular
senescence.
Once the optimal cell lines have been chosen, one can employ sev-
eral different methods to determine the unique characteristics of
the cell line generated. We chose to determine the gene expres-
sional profile of different cell lines (19, 23) and perform confirmatory
studies using routine methods such as flow cytometry, western blot
analysis, and real-time PCR (see also ref. 20). These methods are
described in detail elsewhere. To determine whether the differen-
tially expressed genes or pathways are a relevant component of the
functional activity of the cell line, one could add antagonists or
block with antagonists, or block soluble factors with antibodies.
Alternatively, the gene or pathway can be modified genetically.
To test the relevance of the generated stromal cell lines, or in a
more advanced phase of testing, the genes or signaling pathways,
for the function of interest, additional experiments are required.
The methods described below for investigating hematopoiesis in
culture are similar to the ones described elsewhere, with a few
modifications.
3.5.1. Genetic Modification Genetic modifications of stromal cell lines allow rapid functional
of Stromal Cell Lines analysis of targeted genes. Since studies regarding maintenance of
hematopoietic progenitors and stem cells require culture for at least
2 to 12 weeks, depending on the assay used, stable modifications
have to be introduced. We have a great experience in generation of
stable knockdowns of single genes using lentiviral with shRNAmir
constructs. For this purpose, we have mostly used vectors (pLKO.1)
from commercial sources (Thermo Fischer-OpenBiosystems,
Huntsville, AL, USA). This vector allows antibiotic selection
stromal cells expressing the shRNAmir (20) (see Note 10).
3.5.2. Maintenance of 1. Prepare plastic ware by coating with 0.1% gelatin as described
Hematopoiesis on Stromal in Subheading 3.3.
Cell Clones 2. Plate 3 × 104/cm2 of stromal cells on the gelatinized dishes and
culture them in stromal cell medium and culture to confluence.
Preparation of Stromal Cell
3. Stromal cell proliferation should be mitotically inactivated by
Lines for Cocultures
irradiation with 30 Gy (see Note 11). Replace the supernatant
with fresh Stroma Medium directly after irradiation.
4. Culture cells for 3–4 days prior to adding hematopoietic cells
in order to reduce irradiation stress.
Separation of Lineage In our experiments, we have noticed that cell density of hematopoi-
Negative Cells from etic cells at the start of coculture, may influence the outcome of
Mouse BM this culture (19). Thus, we now routinely start cocultures with
lineage-negative (Lin−) selected cells, or more stringently sorted
Lin− Sca-1+ Kit+ (LSK) cells.
1. Isolate mouse bone marrow cells as described in detail else-
where in this series (27).
2. Count viable cells and proceed with the separation of Lin−
cells. We have isolated Lin-cells using different commercially
available kits (STEMCELL Technologies, Vancouver, Canada
# 13046 or Miltenyi Biotec, # 130-090-858).
3. In all methods, we modified the instructions of the manufactur-
ers to perform labeling the BM cells and subsequent isolation
steps in appropriate amount of HF2+ buffer. Also, to avoid cell
clumping and cellular activation all steps were performed on ice
and isolation was performed in chilled columns and magnets.
4. After the last wash step, count cells in the Lin− fraction and
keep them on ice in HF2+ buffer till further use.
Setup of Cocultures Cocultures can be setup in many different ways. In our procedures, we
chose to minimize the number of Lin− at the start of the coculture.
1. Replace Stroma Medium with 2 ml of long-term culture
medium, supplemented with Glutamax I, hydrocortisone
[10−6 M], and antibiotics.
Fig. 3. Functional testing of the cell lines generated. In a general scheme, the cell lines could be tested by themselves and
through gene expression patterning or other ways to establish phenotype, find out which tissue cell type the cells resem-
ble. Or, and this is shown in (a), co culture experiments with tissue-type stem, progenitor or mature cells could set up in
much the same way as described under Subheading 3.5 and after coculture, the function of these cells evaluated through
different methodologies. In experiments in which we study the hematopoietic system (b), we isolate Lin− cells characterize
those before coculture (shown underneath column) and after coculture for 3 days to 4 weeks. Functional endpoints include
colony assays (see also ref. 26), transplantation into myeloablated recipients (28), as well as molecular endpoints.
2. Seed 5 × 103 of Lin− cells per 3 cm dish (up to 5 × 104 per

100 cm dish) and cultivate at 33°C, 5% CO2 for 2–12 weeks,
depending on the endpoint of the culture.
3. Every week, replace half of the supernatant with fresh medium
with the above supplements (see Note 12).
Choice of Functional In any assay, the choice of endpoint determines which information
Endpoints you can collect from your experiment. In the study of hematopoi-
esis, we routinely determine several endpoints can be chosen.
Alteration of gene expression in stromal cells can result in changes
in cell number (count cells), survival (determine annexin V+ frac-
tion with flow cytometry), or differentiation by flow cytometry of
mature hematopoietic cell subsets and/or methylcellulose-based
colony formation (see Note 13). The important properties of HSC
maintenance and self-renewal can also tested by transplanting
whole cultures into recipients (see Fig. 3).
4. Notes
1. Select an FBS batch that gives good performance of the pri-

mary cells in the assays that you wish to perform. If such a
batch is not available in your laboratory, please try to obtain
such a batch from your colleagues performing similar assays.
This batch will serve as a “positive” control. To obtain your
own batch, it is prudent to test at least ten different batches of
FBS in this same assay. The assay you will use to test FBS
batches is, however, up to you.
2. We found that it was possible to generate cell lines from early
midgestation embryos from “normal” mice (the lacZ trans-
genic BL1b strain) as well as mice expressing tsA58 (20). Thus,
the expression of an immortalizing is not required for cell line
generation. By direct comparison, however, twofold more lines
were isolated from the tsA58 transgenic embryos than from
the control lacZ transgenic embryos. Furthermore, the pres-
ence of the tsA58 gene allowed for a three- to fourfold greater
cloning efficiency compared to the control lacZ marker trans-
genics. Although tsA58 gene had an enhancing effect on the
growth of liver, urogenital ridge and gastrointestinal derived
cell lines, no enhancing effect was observed with the aorta-
mesenchyme derived lines (17).
3. It is important to know under which culture conditions the
primary cells you are interested in will grow. Issues you should
resolve prior to generating cell lines are as follows: (a) which
medium do the primary cells require (with or without serum)?
(b) do you need CM or have growth factor requirements been
established? (c) do the primary cells require anchoring? Using
gelatin, fibronectin, laminin or other coatings can drastically
alter the cell type that will grow out of your culture. The meth-
ods described can be used to generate cell lines from different
types of tissues and to grow different types of adherent cells.
We have not tried to derive non-adherent, suspension-type cell
lines by the method described here.
4. It is very important to regularly check the temperature, CO2 lev-
els, and humidity of the incubator used and make sure the incu-
bator is level. Humidity is checked by weighing a 100 mm dish
and adding exactly 10 ml of water (=10 g, weigh again). One
week later, weigh the dish. Water dissipation should not exceed
10% (1.0 g), 5% or less water loss is optimal. In particular, clon-
ing efficiency depends on optimal levels of CO2 and humidity.
5. The generation of cell lines is a time-consuming and labor-
intensive process. We found that optimal results were obtained
when cells are passaged weekly or prior to reaching confluence.
Do not keep cells unpassaged for more than 1 week. In our

hands, it appeared that failure to passage cells regularly favored
cell senescence. In some cases the crisis period can last for
weeks, sometimes for more than 3 months (17). Thus, it is
important to keep culturing and passaging the cells, even when
no cell proliferation is apparent. In our hands, cell lines eventu-
ally grew out of 30 of 32 (94%) of primary cell cultures of
embryonic tissues.
6. The density of your cultures should be monitored daily. Always
passage your cultures prior to reaching confluence (i.e.,
between 40 and 80%). Especially in the case of contact-inhibi-
tion, a sizeable proportion of cells will cease to be reactivated
once proliferation has stopped by contact inhibition.
7. It is important to freeze samples of newly established cell lines
(pre- and post-cloning) at low passage numbers. This will
ensure that there are low passage cells to go back to in case
certain functional phenotypes are revealed only at these pas-
sages or when disaster strikes (contamination, CO2 failure,
etc.). In addition, once the cell lines have been characterized,
cells with suitable passage numbers can be shared with
collaborators.
8. The cell lines that will be generated will differ in growth char-
acteristics and requirements: some will be contact-inhibited,
and some of the generated cell lines will be growth factor
dependent. Since after thawing no fresh CM will be available,
a mixture of CMs from semi-confluent cells of different tissues
(either per tissue or all tissues together) can be prepared,
filtered (0.2 μm bottletop filter) and stored at 4°C. We stored
this CM not more than 6 months. This CM-mix can then be
used as growth factor supplement for the stromal cells until the
first passage after thawing. After this first passage, cells will pro-
duce their own CM for the next passage which can be collected
as described in Subheading 2, item 6.
9. Cell lines generated from tsA58 transgenic mice showed a sta-
ble functional phenotype up to 50–60 population doublings
after cloning (17). It is known that expression of the immortal-
izing SV40 large T gene is, by itself, not sufficient to immortal-
ize cells. Rather, a secondary event, such as activation of TERT,
is required to produce a stable phenotype of cells for more
than 150 population doublings (8). Thus, it is likely that cul-
turing cell lines beyond 60 population doublings will select for
transformed cells. This should be born in mind when early pas-
sage cells are compared with late passage cells (>60 population
doublings).
10. We did not reclone the selected shRNAmir-expressing stromal
cells, since this would require extensive testing of every single
subclone generated. Instead, we used pools of transduced

stromal cells, which contained at least ten different subclones.
The number of subclones can be estimated from the number
of visible colonies formed during antibiotic selection.
11. The optimal dose of irradiation is determined by cessation of
proliferation and should be de determined for each cell line or
mouse background prior to functional experiments. In our
tsA58-derived cells, we required a dose of at least 30 Gy.
12. Prior to the weekly half-medium change: measure the total
amount of medium present in the culture. If the incubator you
use is opened frequently, the medium may dissipate. In that
case, remove half of the medium and top up with freshly pre-
pared medium with such an amount that after medium change
the total volume of medium is again 2 ml.
13. The protocol for mouse colony forming cell assay can be
obtained from manufacturer’s Web site at www.stemcell.com.
Acknowledgement
We gratefully acknowledge Jessyca Maltman (Terry Fox Laboratory,

BC Cancer Agency, Vancouver, BC, Canada) for her help and
experience in the culture of adherent mouse marrow cells. We also
would like to thank Elaine Dzierzak for the opportunity to estab-
lish cell lines from midgestation embryos.
References
1. Santerre RF, Cook RA, Crisel RM, Sharp JD, 6. Ohsawa K, Imai Y, Nakajima K, Kohsaka S
Schmidt RJ, Williams DC, Wilson CP (1981) (1997) Generation and characterization of a
Insulin synthesis in a clonal cell line of simian microglial cell line, MG5, derived from a p53-
virus 40-transformed hamster pancreatic beta deficient mouse. Glia 21:285–298
cells. Proc Natl Acad Sci USA 78:4339–4343 7. Thompson DL, Lum KD, Nygaard SC,
2. Bayley SA, Stones AJ, Smith CG (1988) Kuestner RE, Kelly KA, Gimble JM, Moore
Immortalization of rat keratinocytes by trans- EE (1998) The derivation and characteriza-
fection with polyomavirus large T gene. Exp tion of stromal cell lines from the bone mar-
Cell Res 177:232–236 row of p53−/− mice: new insights into
3. Jat PS, Sharp PA (1986) Large T antigens of sim- osteoblast and adipocyte differentiation. J
ian virus 40 and polyomavirus efficiently estab- Bone Miner Res 13:195–204
lish primary fibroblasts. J Virol 59:746–750 8. O’Hare MJ, Bond J, Clarke C, Takeuchi Y,
4. Ridley AJ, Paterson HF, Noble M, Land H Atherton AJ, Berry C, Moody J, Silver ARJ,
(1988) Ras-mediated cell cycle arrest is altered Davies DC, Alsopi AE, Neville AM, Jat PS
by nuclear oncogenes to induce Schwann cell (2001) Conditional immortalization of freshly
transformation. EMBO J 7:1635–1645 isolated human mammary fibroblasts and
5. Morales CP, Holt SE, Ouelette M, Kaur KJ, endothelial cells. Proc Natl Acad Sci USA
Yan Y, Wilson KJ, White MA, Wright WE, 98:646–651
Shay JW (1999) Absence of cancer associated 9. Tegtmeyer P (1975) Function of simian virus
changes in human fibroblasts immortalized 40 gene A in transforming infection. J Virol
with telomerase. Nat Genet 21:115–118 15:613–618
10. Jat PS, Cepko CL, Mulligan RC, Sharp PA 20. Renström J, Istvanffy R, Gauthier K, Shimono
(1986) Recombinant retroviruses encoding A, Mages J, Jardon-Alvarez A, Kröger M,
simian virus 40 large T antigen and polyomavi- Schiemann M, Busch DH, Esposito I, Lang R,
rus large and middle T antigens. Mol Cell Biol Peschel C, Oostendorp RAJ (2009) Secreted
6:1204–1217 frizzled-related protein 1 extrinsically regulates
11. Morgan JE, Beauchamp JR, Pagel CN, cycling activity and maintenance of hematopoi-
Peckham M, Ataliotis P, Jat PS, Noble MD, etic stem cells. Cell Stem Cell 5:157–167
Farmer K, Partridge TA (1994) Myogenic cell 21. Grzywacz B, Kataria N, Sikora M, Oostendorp
lines derived from transgenic mice carrying a RAJ, Dzierzak EA, Blazar BR, Miller JS,
thermolabile T antigen: a model system for the Verneris MR (2006) Coordinated acquisition
derivation of tissue-specific and mutation- of inhibitory and activating receptors and func-
specific cell lines. Dev Biol 162:486–498 tional properties by developing human natural
12. Okuyama R, Yanai N, Obinata M (1995) killer cells. Blood 108:3824–3833
Differentiation capacity toward mesenchymal 22. Gordon-Keylock SA, Jackson M, Huang C,
cell lineages of bone marrow stromal cells Samuel K, Axton RA, Oostendorp R, Taylor
established from temperature-sensitive SV40 AH, Wilson J, Forrester L (2010) Induction of
T-antigen gene transgenic mouse. Exp Cell haematopoietic differentiation of mouse
Res 218:424–429 embryonic stem cells by an AGM-derived
13. Renström J, Kröger M, Peschel C, Oostendorp stromal cell line is not further enhanced by
RAJ (2010) How the niche regulates over-expression of HOXB4. Stem Cells Dev
hematopoietic stem cells. Chem Biol Interact 19(11):1687–1698
184:7–15 23. Ledran MH, Krassowska A, Armstrong L,
14. Garrett RW, Emerson SG (2009) Bone and Dimmick I, Renström J, Lang R, Yung S,
blood vessels: the hard and the soft of Santibanez-Coref M, Dzierzak E, Stojkovic M,
hematopoietic stem cell niches. Cell Stem Cell Oostendorp RAJ, Forrester L, Lako M (2008)
4:503–506 Efficient hematopoietic differentiation of
15. Walkley CR, Yuan YD, Chandraratna RA, human embryonic stem cells on stromal cells
McArthur GA (2002) Retinoic acid receptor derived from hematopoietic niches. Cell Stem
antagonism in vivo expands the numbers of Cell 3:85–98
precursor cells during granulopoiesis. Leukemia 24. Dzierzak EA, de Bruijn M (2001) Isolation
16(9):1763–1772 and analysis of hematopoietic stem cells from
16. Raaijmakers MH, Mukherjee S, Guo S, Zhang mouse embryos. In: Klug CA, Jordan CT (eds)
S, Kobayashi T, Schoonmaker JA, Ebert BL, Methods in molecular medicine: hematopoi-
Al-Shahrour F, Hasserjian RP, Scadden EO, etic stem cell protocols. Humana, New Jersey
Aung Z, Matza M, Merkenschlager M, Lin C, 25. Oostendorp RAJ, Harvey KN, Kusadasi N, de
Rommens JM, Scadden DT (2010) Bone pro- Bruijn MF, Saris C, Ploemacher RE, Medvinsky
genitor dysfunction induces myelodysplasia AL, Dzierzak EA (2002) Stromal cell lines
and secondary leukaemia. Nature from mouse aorta-gonads-mesonephros sub-
464(7290):852–857 regions are potent supporters of hematopoietic
17. Oostendorp RAJ, Medvinsky AJ, Kusadasi N, stem cell activity. Blood 99:1183–1189
Nakayama N, Harvey K, Orelio C, Ottersbach 26. Miller CL, Eaves CJ (2001) Long-term cul-
K, Covey T, Ploemacher RE, Saris C, Dzierzak ture-initiating cell assays for human and murine
E (2002) Embryonal subregion-derived cells. In: Klug CA, Jordan CT (eds) Methods
stromal cell lines from novel temperature-sen- in molecular medicine: hematopoietic stem
sitive SV40 T antigen transgenic mice support cell protocols. Humana, New Jersey
hematopoiesis. J Cell Sci 115:2099–2108 27. Miller CL, Lai B (2005) Human and mouse
18. Umar A, Luider TM, Berrevoets CA, colony-forming assays. In: Helgason CD,
Grootegoed JA, Brinkmann AO (2003) Miller CL (eds) Methods in molecular biol-
Proteomic analysis of androgen-regulated pro- ogy: basic cell culture protocols. Humana,
tein expression in a mouse fetal vas deferens New Jersey
cell line. Endocrinology 144:1147–1154 28. Szilvassy SJ, Nicolini FE, Eaves CJ, Miller CL
19. Oostendorp RAJ, Robin C, Steinhoff C, Marz (2001) Quantitation of murine and human
S, Bräuer R, Nuber UA, Dzierzak EA, Peschel hematopoietic stem cells by limiting-dilution
C (2005) Long-term maintenance of analysis in competitively repopulated hosts. In:
hematopoietic stem cells does not require con- Klug CA, Jordan CT (eds) Methods in molec-
tact with embryo-derived stromal cells in ular medicine: hematopoietic stem cell proto-
cocultures. Stem Cells 23:842–851 cols. Humana, New Jersey
Chapter 20
Isolation, Enumeration, and Expansion of Human

Mesenchymal Stem Cells in Culture
Ravenska Wagey and Brenton Short
Abstract
Human bone marrow (BM) contains a population of non-hematopoietic stem cells also termed stromal
cells, mesenchymal cells or multipotent mesenchymal stromal cells (MSCs). These cells have unique stem
cell-like properties including their ability to self-renew, differentiate into multiple tissue types, and modu-
late immune cell responses through paracrine effects. These properties have positioned mesenchymal cells
as biological agents in clinical trials for various diseases since the 1990s. Mesenchymal cells have been iso-
lated from various tissues and cultured using various media and methods resulting in a lack of standardiza-
tion in culture methods for these cells. Consequently, cells cultured in different laboratories exhibit
different characteristics of MSC-like cells. This chapter outlines protocols for optimal isolation, enumera-
tion, and expansion of human MSCs from BM in fetal bovine serum (FBS)-containing medium, as well as
in xeno-free medium.
Key words: FBS-containing medium, Xeno-free medium, Human mesenchymal stem cells, CFU-F,
Expansion, Bone marrow mononuclear cells
1. Introduction
The non-hematopoietic stem cells in the bone marrow (BM), also

termed multipotent marrow stromal cells or mesenchymal stem
cells (MSCs), are a heterogeneous population of plastic-adherent,
fibroblast-like cells, which in culture are able to self-renew and dif-
ferentiate into bone, adipose, and cartilage tissue (1–3). Recently,
cells with characteristics of MSCs have been isolated from various
tissues including adipose tissue, cord blood, skeletal muscle, amni-
otic fluid, fetal liver and lung, synovium, and the circulatory system
(4–10). Accumulating evidence indicates a perivascular location for
these cells, leading to the suggestion that MSCs are pericytes which
315
316 R. Wagey and B. Short
exist in close proximity to endothelial cells in capillaries and

microvessels in multiple organs (11).
Friedenstein and colleagues in the 1960s conducted the initial
studies to characterize single cell suspensions of BM stromal cells.
His work revealed the osteogenic potential of non-hematopoietic
cells in the BM. These plastic-adherent, fibroblast like cells are able
to give rise to colonies of cells termed colony forming unit-
fibroblasts (CFU-F) (12, 13). The CFU-F assay is now widely used
as an in vitro assay to quantify the frequency of MSCs from a
specific donor tissue. There appears to be a strong correlation
between age and proliferative potential, with decreasing progeni-
tor cell proliferation and number associated with increasing age
(14). Since abnormal function of stromal precursor cells has been
implicated in several diseases, the CFU-F assay provides the means
to examine the quality and quantity of MSC progenitors from
specific tissues and donors prior to cell expansion.
Studies have shown that the clonogenic precursor cells initiat-
ing these colonies are capable of differentiating into multiple mes-
enchymal lineages including adipose tissue, cartilage, bone, and
muscle (15–18). More recently, it has been reported that MSCs are
also capable of giving rise to non-mesodermal cell types including
neural cells (19) and cardiac muscle (20).
The multi-differentiation potential of MSCs, as well as their
efficacy in the modulation of other cell types (paracrine effect),
have generated considerable interest in utilizing MSCs for various
cell-based therapies and as vehicles for gene therapy in the field of
tissue engineering and regenerative medicine. Accordingly, MSC
have already been used in clinical trials including the treatment of
childhood Osteogenesis Imperfecta (OI) (21), in the facilitation of
hematopoietic reconstitution following HSC transplantation (22)
and in the prevention and treatment of graft versus host disease
(GvHD) (23). This chapter describes the methods for isolation
and culture of human MSCs in FBS-containing medium.
There are also several publications on culturing MSCs in
serum-free formulations including the use of platelet lysates in the
medium (24–26). The medium formulation, as well as the culture
method for isolation and expansion of MSCs, can determine the
characteristics of the cultured cells (e.g., Multipotent Adult
Progenitor Cells—MAPC, Marrow-isolated adult multilineage
inducible cells—MIAMI, and Mesenchymal Stem Cells—MSCs)
(3, 27, 28). Hence, MSCs cultured in a xeno-free medium can
have different morphology and growth characteristics compared to
cells cultured in serum-containing medium. The protocol for
expansion of MSCs in the xeno-free formulation described here,
called MesenCult™-XF, has been optimized for culturing cells in
this particular medium with the selected substrate and dissociation
enzyme. The xeno-free formulation, MesenCult™-XF contains
components selective for growth of MSCs while suppressing
20 Culture of Human Mesenchymal Stem Cells 317
growth of hematopoietic cells at early passage. Hence, superior

clonogenic growth, higher purity of MSCs at lower passage, and
greater expansion of cells can be obtained in MesenCult™-XF
compared to FBS-containing medium. In this chapter we also
describe the protocol to isolate, enumerate, and expand human
MSCs in a xeno-free medium formulation.
2. Materials
1. 14 mL polypropylene tubes.
2.1. Tissue Culture
2. 50 mL conical tubes.
3. Recommended tissue culture plates/flasks: 6-well plates
(Corning Catalog #3516) or T-25 cm2 Tissue culture flasks
(Falcon Catalog #353109).
4. Parafilm®.
2.2. Reagents 1. Dulbecco PBS without Ca2+ and Mg2+.

2. PBS + 2% FBS.
3. Methanol, ACS Grade.
4. Giemsa Stain Solution.
5. Ficoll-Paque™ PLUS.
6. 3% Acetic Acid with Methylene Blue.
7. Trypan Blue.
8. 0.25% Trypsin-EDTA.
9. Complete MesenCult™ Medium consisting of MesenCult™
MSC Basal Medium (Human; StemCell Catalog #05401) and
Mesenchymal Stem Cell Stimulatory Supplements (Human;
StemCell Catalog #05402). To prepare the medium, thaw the
supplements at room temperature or 2–8°C overnight. Add
the entire contents of the supplements to the basal medium
and mix thoroughly. Complete MesenCult™ Medium (Human)
is stored at 2–8°C and should be prepared in volumes that can
be used within 1 month (see Note 1).
10. L-Glutamine 200 mM.
11. Distilled water (autoclaved).
12. Iscove’s MDM (IMDM).
13. Fresh Bone marrow (25 mL) or frozen cultured BM-MSCs.
14. Human Serum Albumin (HSA; quality cell culture-tested and
verified nontoxic for MSCs) for preparation of 10% HSA in
Iscove’s MDM.
15. 0.5 M EDTA (Ethylenediaminetetraacetic acid).
16. Cryostor.
17. Minimum Essential Medium Eagle (MEM), Alpha Modification
(see Note 2).
18. MesenCult™-ACF Enzymatic Dissociation Solution (StemCell
Cat #05427).
19. MesenCult™-ACF Enzyme Inhibition Solution (StemCell Cat
#05428).
20. MesenCult™-XF Basal Medium (StemCell Catalog #05421).
21. MesenCult™-XF Supplement (5×; StemCell Catalog #05422).
22. MesenCult™-XF Attachment Substrate (StemCell Catalog
#05425).
23. 0.2 μm filter.
3. Methods
3.1. Culture, Isolation and establishment of human MSC cultures should be

Expansion, and performed in Level II Biosafety hoods using sterile techniques.
Enumeration of MSC in
FBS-Containing When working with fresh bone marrow (BM) samples, the cells
Medium need to be processed to remove the red blood cells or to enrich for
the desired cells prior to culture. If the BM is freshly collected from
3.1.1. Isolation of MSCs the posterior iliac crest or sternum of the BM donor, heparin
from Unprocessed Human (5000 U) should be added as an anticoagulant following aspiration
Bone Marrow of the BM. When purchasing fresh BM, heparin is already added.
MSCs are obtained from fresh BM samples by isolating mononu-
clear cells using density gradient separation as detailed below.
1. Purchase/obtain 25 mL of fresh bone marrow.
2. Dilute heparinized unprocessed BM cells at a 1:2 ratio in
PBS + 2% FBS (i.e., 25 mL BM and 50 mL of PBS + 2% FBS for
a total volume of 75 mL)
3. In three 50 mL conical tubes (BD Catalog # 352070), pipette
17 mL Ficoll-Paque™ PLUS (StemCell Catalog #07907/07957)
into each tube. Carefully layer 25 mL diluted BM on top of the
Ficoll-Paque™ PLUS in each tube using a smaller diameter
pipette (e.g., 10 mL instead of 25 mL pipette). Slightly tilt the
50 mL conical tube containing Ficoll-Paque™ and slowly
release (dropwise) BM from a 10 mL pipette on top of the
Ficoll layer.
4. Centrifuge at room temperature (15–25°C) for 30 min at
300 × g in a benchtop centrifuge with the brake off.
5. Remove and discard the upper plasma layer with a pipette
without disturbing the plasma: Ficoll-Paque™ PLUS interface.
Carefully pipette and retain the mononuclear cells located at

the interface layer and place in a new 50 mL conical tube.
Suspend the mononuclear cells in 40 mL cold (2–8°C) PBS + 2%
FBS Buffer. Mix gently by pipetting the cell suspension.
6. Centrifuge the cells at 300 × g (~1,200 rpm) for 10 min at
room temperature in a benchtop centrifuge with the brake on.
Aspirate the supernatant and suspend the cells in 1–2 mL
Complete MesenCult™ Medium (Human)
7. Dilute an aliquot of cells (e.g., 10 μL) 1/50 to 1/100 in 3%
Acetic Acid with Methylene Blue and count the total number
of nucleated cells using a hemacytometer.
3.1.2. Plating Cells for the 1. Primary BM mononuclear cells should be plated at densities
CFU-F Assay between 2.0 and 12 × 104 cells/cm2 in Complete MesenCult™
Medium (Human). For example, dilute the cells (after the total
nucleated cell count) to a stock cell concentration of 3 × 106
cells/mL in Complete MesenCult™ Medium (Human). Plate
three different cell densities by adding 1.0, 0.67, and 0.33 mL
of the cell stock to separate T-25 cm2 culture-treated dishes
filled with Complete MesenCult™ Medium (Human) to a total
volume of 8–10 mL. This will yield final cell concentrations of
3 × 106 cells, 2 × 106 cells and 1 × 106 cells per flask. For 6-well
plates, cells should be plated at densities between 3 × 105 cells/
well and 1 × 106 cells/well.
2. Place the T-25 cm2 tissue culture flasks or 6-well plates into a
37°C humidified incubator with 5% CO2 in air and >95%
humidity for 10–14 days. Maximum colony size and numbers
are typically observed at 14 days (see Note 3).
3.1.3. Staining of 1. Remove the medium from the CFU-F cultures using a 10 mL
CFU-F-Derived Colonies pipette and discard it into the biohazardous waste. The adher-
ent colonies will remain attached to the plate. This staining
procedure can be done on a benchtop since sterility is not
required.
2. Gently wash the culture dishes or flasks twice by adding PBS
(e.g., 2 mL/well of a 6-well plate or 5 mL/T25 cm2 flask) to
the CFU-F culture. The wash is to remove any remaining
medium. Discard the PBS from the two washes into the waste.
3. Add 5 mL of methanol, using a 5 mL pipette, to each culture
dish or flask and incubate for 5 min at room temperature to fix
the cells.
4. Remove the methanol using a 10 mL pipette and discard into
the biohazardous waste. Let the culture dishes or flasks air dry
at room temperature for about 5–10 min.
5. Add 5 mL of Giemsa Staining Solution to each culture dish or
flask and leave for 5–10 min.
6. Remove the Giemsa Staining Solution using a 10 mL pipette,

discard the Giemsa solution and rinse the culture dishes or
flasks under a low running tap. Swirl and discard the water.
A minimum of four rinses are required until water remains
clear. Allow the tissue culture dishes or flasks to dry at room
temperature for about 1–2 h.
3.1.4. Enumeration of CFU-F colonies from human cells in FBS-containing medium are
CFU-F-Derived Colonies typically between 1 and 8 mm in diameter and should be scored
both macroscopically and microscopically (for confirmation of col-
ony numbers). Photographs of representative CFU-F-derived col-
onies are shown in Fig. 1. It is important to note that some colonies
do not take up enough stain to be easily visible macroscopically,
and therefore it is important to verify the number of colonies
counted by scoring colonies microscopically. We recommend tak-
ing a felt-tip pen and marking each CFU-F on the bottom of the
well when counted. This prevents counting colonies more than
once.
Ensure that there is a linear relationship between the cell num-
bers plated and the resulting colony numbers, by confirming that
there are twice as many colonies when 2 × 106 cells are plated as
compared to 1.0 × 106 cells. Likewise, there should be twice as
many colonies when 1.0 × 106 cells are plated as compared to
0.5 × 106 cells. Ideally there should be 10–40 colonies per T-25 cm2
flask. Linearity may not be observed outside of this range as the
cells would have been under- or over-plated (see Note 4).
Fig. 1. Photographs of representative CFU-F-derived colonies cultured in MesenCult™ Proliferation medium. A large colony
(Left ) and a medium size colony (Right ).
Fig. 2. Optimal density (80% confluency) for passaging cells (Left ) and over-confluent density (100% confluency) for pas-
saging (Right )
3.1.5. Expansion and Confluent mesenchymal cell cultures can be produced when cells
Passaging of Cultured from BM are plated at relatively high densities on tissue-culture-
Mesenchymal Cells treated flasks or dishes in Complete MesenCult™ Medium (Human).
Mesenchymal cell numbers can then be expanded by splitting the
cells when they become 70–80% confluent. If cells remain in a highly
confluent state (>90%) for a significant time (days) it may reduce
their longevity and their potential to differentiate. Optimal and over-
confluent densities for passaging are shown in Fig. 2a, b, respec-
tively. Culture-expanded mesenchymal cells can be used for a number
of applications including plasticity studies, assessment of differentia-
tion or expansion potential, and the evaluation of phenotype.
1. Plate primary cells at 1.0–5.0 × 105 cells/cm2 in complete
MesenCult™ Medium (Human) in tissue-culture-treated
dishes or flasks. The recommended cell numbers for a T-25 cm2
flask in 10 mL Complete MesenCult™ Medium (Human) are
2.5 × 106–1.0 × 107 BM mononuclear cells. For frozen marrow
stromal cells (P1) plate between 1.25 × 105 cells to 2.5 × 105
cells/T-25 cm2 flask.
2. Observe mesenchymal cell cultures under a microscope to
ensure that the cells are at an adequate stage for passaging
(~80% confluence). This should take approximately 7–14 days
for primary BM cells but less time (3–6 days) for culture-
expanded cells. If the medium in the flask or dish appears acidic
(more yellow in color than orange/red) prior to reaching 80%
confluence, a half-medium change can be done by removing
one half of the acidic medium and replacing it with fresh
Complete MesenCult™ Medium (Human) pre-warmed to
37°C (see Note 5).
3. If the cells are ready to be passaged, remove the Complete
MesenCult™ Medium (Human) leaving the adherent
mesenchymal cells behind. Wash the cells with PBS (add PBS
to the culture to cover the entire culture, swirl the dish/flask
in PBS to remove residual FBS-containing medium and remove
PBS with a pipette).
4. For cells in a T-25 cm2 flask, add 3–5 mL Trypsin-EDTA to
cover the cells and incubate at 37°C for 3–5 min.
5. Check under the microscope to ensure that the mesenchymal
cells have detached. If a small percentage of cells are still
attached, gently tap the flask on a bench surface to detach the
remaining cells. Add 1 mL FBS or 5 mL of Complete
MesenCult™ Medium (Human) to neutralize the action of the
trypsin.
6. Collect the trypsinized cells into a 14 mL tube and centrifuge
the cells at 300 × g for 8 min at room temperature with the
brake on. Remove the supernatant and suspend the cell pellet
in 1–2 mL of Complete MesenCult™ Medium (Human).
7. Perform a cell count using Trypan Blue dye exclusion by diluting
an aliquot of cells (e.g., 10 μL) 1/3 to 1/10 with Trypan Blue.
Replate the cells at 4–10 × 104 cells/cm2. Alternatively, the cells
can be divided into new tissue-culture-treated flasks at a recom-
mended dilution of 1/4 (e.g., one T-25 cm2 tissue-culture-
treated flask containing 80% confluent mesenchymal cells can be
passaged into four T-25 cm2 tissue-culture-treated flasks).
3.1.6. Freezing Mesenchymal cells can be frozen at any passage. Studies in our
Mesenchymal Cells laboratory have shown that cryopreserved cells from passage num-
bers 2–7 maintain their phenotype and differentiation potential.
Before beginning have all reagents COLD (2–8°C) and label ster-
ile cryovials using an indelible marker.
1. Make up 20% Dimethyl Sulfoxide (DMSO) in Fetal Bovine
Serum and filter-sterilize using a 0.2 μm filter. Keep on ice.
2. Harvest the cells from the tissue culture surface as described in
Subheading 3.1.5, steps 3–5. Centrifuge the cells at 300 × g,
25°C for 7 min. Remove the supernatant and suspend the cells
in FBS to give a maximum concentration of 2 × 106 cells/mL.
Place this cell suspension on ice.
3. Slowly add 20% DMSO in FBS dropwise to the cells. Mix cells
gently with 20% DMSO in FBS at a ratio of 1:1 (the final cell
suspension will be 90% FBS/10% DMSO). Transfer 1 mL of
cells in freezing medium to each cryovial. The final cell con-
centration will be ~1 × 106 cells per vial.
4. Place the cryovials immediately into a thawed 70% isopropanol
freezing container and place the container in a −135°C freezer
overnight. On the next day, remove frozen vials from the freez-
ing container and store at −135°C (or colder) or in liquid
nitrogen (see Note 6).
3.1.7. Thawing 1. Thaw the cells quickly in a 37°C water bath or a beaker of
Mesenchymal Cells warm water in a tissue culture hood. Wipe the cryovial with
70% ethanol.
2. Gently transfer the cells into a 50 mL centrifuge tube by
pipetting the cells from the cryovial into the centrifuge tube
(see Note 7).
3. Slowly add 15 mL IMDM containing 2% FBS drop-wise while
holding the tube and gently swirling. Then fill tube to 50 mL
with IMDM containing 2% FBS. Gently invert the tube to
mix.
4. Centrifuge the cells at 300 × g for 8 min.
5. Discard the supernatant and “flick” the tube gently to suspend
the pellet.
6. Suspend the cells at the desired concentration in Complete
MesenCult™ Medium (Human).
3.2. Culture, Expansion 1. Thaw MesenCult™-XF Supplement (5×; StemCell Catalog

and Enumeration of #05422) overnight at 2–8°C. Add the entire MesenCult™-XF
MSC in Xeno-Free Supplement (100 mL) to one bottle (400 mL) of
Medium MesenCult™-XF Basal Medium (StemCell Catalog #05421).
Add L-Glutamine to a final concentration of 2 mM. Cells cul-
3.2.1. Preparation of
tured in MesenCult™-XF medium will not grow without addi-
Reagents and Coating of
tion of 2 mM L-glutamine. This is now referred to as Complete
Plates
MesenCult™-XF Medium. Complete MesenCult™-XF
Medium can be stored at 2–8°C for no more than 5 days (see
Note 8).
2. Tissue culture flasks need coating with MesenCult™-XF
Attachment Substrate (StemCell Catalog #05425) to support
cell adherence. It is recommended to coat plates 1 day prior to
usage (i.e., coat overnight at 2–8°C), but if time is limited,
they can be coated for 2 h at 15–25°C (room temperature)
prior to use.
3. Coating plates for the CFU-F assay: Dilute MesenCult™-XF
Attachment Substrate 1/40 in sterile PBS without Ca2+ and
Mg2+ (StemCell Catalog #37350). Gently mix by inverting the
tube twice. Prepare an amount slightly more than required, to
account for pipetting variability. For example, to coat one
6-well plate, dilute 167 μL MesenCult™-XF Attachment
Substrate in 4.8 mL PBS and add 800 μL per well.
4. To coat plates for cell expansion from primary tissue: Dilute
MesenCult™-XF Attachment Substrate 1:20 in sterile PBS
without Ca2+ and Mg2+ and gently mix by inverting the tube
twice. Prepare an amount slightly more than required, to
T-75 cm2 flask, dilute 250 μL MesenCult™-XF Attachment
Substrate in 5 mL PBS.
5. To coat plates for cell expansion from previously cultured cells:

Dilute MesenCult™-XF Attachment Substrate 1:28 in sterile
PBS without Ca2+ and Mg2+. Gently mix by inverting the tube
twice. Prepare an amount slightly more than required, to
T-75 cm2 flask, dilute 185 μL MesenCult™-XF Attachment
Substrate in 5 mL PBS. Substrate volumes to use and sug-
gested plates or flasks are indicated in Tables 1.
6. Wrap plates with Parafilm®, sealing the junction between the
base and lid, and incubate at 2–8°C (in the refrigerator) over-
night or for 2 h at 15–25°C (room temperature). For flasks,
seal the vent on the cap with Parafilm® and incubate as
described.
7. If plates/flasks were incubated overnight at 2–8°C, bring to
room temperature (approximately 20 min) prior to washing.
Table 1
Volume of attachment substrate for coating flasks
Size Volume of attachment substrate Suggested plates/flasks
6-Well plate 800 μL/well Corning Catalog #3516

2
T-25 cm flask 2 mL/flask BD Falcon Catalog #353109
2
T-75 cm flask 5 mL/flask BD Falcon™ Catalog #353136
Table 2
Plating densities for expansion of primary BM mononuclear
cells in MesenCult™-XF
Tissue culture Volume of Suggested plating

vessel medium Surface area densities
6-Well plate 2.5 mL/well 9.5 cm2/well 3 × 105 cells/well

4.5 × 105 cells/well
T-25 cm2 10 mL/flask 25 cm2/flask 8.0 × 105 cells/flask
10 × 105 cells/flask
12.5 × 105 cells/flask
Gently pipette off, and discard, remaining MesenCult™-XF

Attachment Substrate without touching the newly coated
surface.
8. Wash plates/flasks once with sterile distilled water by slowly
pipetting water down the side of the well/flask, being careful
not to scrape the newly coated surface. Swirl gently to rinse the
entire surface and then carefully aspirate off water.
9. Allow to dry for at least 15 min at 15–25°C prior to use.
3.2.2. Preparation of a Prior to initiating this protocol prepare 500 mL Isolation Buffer
Mononuclear Cell (PBS + 0.5% HSA + 2 mM EDTA) by adding 25 mL HSA (10%
Suspension from Fresh stock solution in sterile dH2O) and 2 mL EDTA (0.5 M stock
Human Bone Marrow solution) to 473 mL of 1× PBS. Once made, the Isolation Buffer
can be stored at 2–8°C for 1 month. If any of the components are
not sterile (i.e., EDTA), be sure to filter sterilize the individual
components or the complete buffer with a 0.2 μm filter.
1. Count the total number of nucleated cells in the BM sample by
taking 10 μL BM and diluting it 1/40 to 1/100 with 3% Acetic
Acid with Methylene Blue (StemCell Catalog #07060). Count
cells using a hemacytometer (see Note 9).
2. Warm 50 mL Isolation Buffer at room temperature for 20 min
prior to use. Dilute BM approximately 1:3 with room tempera-
ture Isolation Buffer (e.g., dilute 25 mL BM with 50 mL
Isolation Buffer for a total volume of 75 mL).
3. Pipette 17 mL Ficoll-Paque™ PLUS into each of three 50 mL
conical tubes. Carefully layer 25 mL diluted BM on top of the
Ficoll-Paque™ PLUS in each tube.
4. Centrifuge at room temperature (15–25°C) for 30 min at
300 × g in a benchtop centrifuge with the brake off.
5. Remove and discard the upper plasma layer without disturbing
the plasma:Ficoll-Paque™ PLUS interface. Carefully remove
and retain the mononuclear cells located at the interface layer
and place in a new 50 mL conical tube. Suspend the mononu-
clear cells in 40 mL cold (2–8°C) Isolation Buffer. Mix gently
by pipetting.
6. Centrifuge the cells at 300 × g for 10 min at room temperature
in a benchtop centrifuge with the brake on. Remove the super-
natant and suspend cells in 1–2 mL cold Isolation Buffer.
7. Dilute an aliquot of cells (i.e., 10 μL) 1/50 in 3% Acetic Acid
with Methylene Blue and count the total number of nucleated
cells using a hemacytometer.
8. Dilute cells in Complete MesenCult™-XF Medium at a final
concentration of 1 × 106 cells/mL.
3.2.3. Plating, Staining, The cell source for setting up the CFU-F assay can be either mono-
and Enumerating Cells in nuclear cells from a fresh BM sample or culture-expanded mesen-
the CFU-F Assay chymal cells. It is recommended not to use previously frozen BM
mononuclear cells, as freezing mononuclear cells may affect the
viability of mesenchymal progenitor cells which are present at low
frequency in the BM. CFU-F assays must be performed using tissue-
culture-treated plates that have been coated with MesenCult™-XF
Attachment Substrate (StemCell Catalog #05425) as described in
Subheading 3.2.1.
1. Using fresh BM-derived mononuclear cells processed accord-
ing to Subheading 3.2.2 seed cells at three different densities
(between 1.5 and 5 × 104 cells/cm2) in Complete
MesenCult™-XF Medium. For example: In a 6-well plate add
150, 250 and 500 μL cells (stock: 1 × 106 cells/mL) for a con-
centration of 1.5 × 105 cells/well, 2.5 × 105 cells/well and
5.0 × 105 cells/well in 2 mL MesenCult™-XF Medium.
2. When using culture-expanded mesenchymal stem cells, seed
25–250 cells per well of a 6-well plate at five different densities
in Complete MesenCult™-XF Medium (see Note 10)
3. Place the cultures in a 37°C incubator with 5% CO2 in air and
95% humidity for 9–12 days. After day 7, monitor the growth
of colony-forming cells daily, to prevent overgrowth. Cultures
should be stained before adjacent colonies become too large
and merge (see Note 10)
4. Gently remove MesenCult™-XF Medium from CFU-F cul-
tures with a 5 mL or 10 mL pipette and discard. Adherent
CFU-F colonies will remain attached.
5. Gently wash colonies once with 2 mL PBS per well of a 6-well
plate to remove any residual culture medium. Fix cultures with
methanol and stain with Giemsa as described in
Subheading 3.1.3.
6. Refer to Subheading 3.1.4 for details on CFU-F enumeration.
3.2.4. Expansion and 1. When initially plating bone marrow mononuclear cells in
Passage of Cultured MesenCult™-XF Medium for expansion, plate between 3.0
Mesenchymal Stem Cells and 7.0 × 104 cells/cm2 in Complete MesenCult™-XF Medium
into tissue-culture-treated plates/flasks that have been coated
with MesenCult™-XF Attachment Substrate (StemCell Catalog
#05425), as described in Subheading 2.1. Suggested plating
densities are outlined in Table 2 (see Note 11)
2. Place the cultures in a 37°C incubator with 5% CO2 in air and
95% humidity for 9–13 days.
3. Observe primary MSCs under a microscope 7 days post-plat-
ing to determine if they are ready for passaging or if the medium
is acidic and a half-medium change needs to be performed.
Cells should be passaged when cultures are 80% confluence in

MesenCult™-XF medium (see Note 12).
4. If the cells are ready to be passaged, warm the MesenCult™-
ACF Enzymatic Dissociation Solution (Cat #05427) and the
MesenCult™-ACF Enzyme Inhibition Solution (Cat #05428)
to room temperature. Do not incubate at 37°C (see Note 13)
5. To passage cells, slowly remove the medium from the cultures
with a 5 mL or 10 mL pipette. The adherent cells will remain
attached to the culture dish.
6. Wash cells with 2 mL sterile Ca/Mg2+ free PBS per well of a
6-well plate to remove residual culture medium. Remove PBS
with a 5 mL or 10 mL pipette.
7. Add 1 mL MesenCult™-ACF Enzymatic Dissociation Solution
to each well of a 6-well plate, 3 mL to a T-25 cm2 flask, or
6 mL to a T-75 cm2 flask. Incubate at 37°C for 2–5 min.
8. After 2 min, observe cells under the microscope to ensure that
all cells have detached. Gently tap plate/flask to detach remain-
ing cells. If cells remain adherent, return to incubator for 1 min
further and again observe cells microscopically to assess cell
detachment. Do not incubate for longer than 6 min.
9. Add 1 mL MesenCult™-ACF Enzyme Inhibition Solution to
each well of a 6-well plate, 3 mL to a T-25 cm2 flask or 6 mL
to a T-75 cm2 flask.
10. For cells cultured in a 6-well plate pipette the 2 mL cell sus-
pension into a 14 mL polystyrene tube and wash each well
with 3 mL MEM Alpha to recover remaining cells. Add MEM
Alpha to bring the total volume to 8 mL. Duplicate wells
seeded at the same density may be pooled into one tube if
desired. See Note 14 for the protocols to deal with cells cul-
tured in other sizes of culture flasks.
11. Centrifuge cells at 300 × g for 8 min at room temperature with
the brake on.
12. Remove the supernatant and suspend the cell pellet in 0.5–1 mL
Complete MesenCult™-XF Medium.
13. Dilute an aliquot of cells (i.e., 10 μL) 1/3 to 1/10 with Trypan
Blue and perform a viable cell count using a hemacytometer.
14. Suspend cells in Complete MesenCult™-XF Medium for plat-
ing into new tissue-culture-treated plates/flasks that have been
coated with MesenCult™-XF Attachment Substrate, as
described in Subheading 3.2.1. The recommended plating
density for passaged cells is between 1.5 and 4.0 × 103 cells/
cm2. The optimal plating densities for each tissue culture vessel
are indicated in Table 3.
Table 3
Plating densities for expansion of cultured cells in
MesenCult™-XF
Tissue culture Volume of Suggested plating

vessel medium Surface area densities
6-Well plate 2.5 mL/well 9.5 cm2/well 1.5 × 104 cells/well

T-75 cm2 15 mL/flask 75 cm2/flask 15 × 104 cells/flask
25 × 104 cells/flask
15. Culture the cells in a 37°C incubator with 5% CO2 in air and
95% humidity until they reach 80% confluence. When cells
reach 80% confluence and are ready to be passaged, repeat
steps 4–16 of Subheading 3.2.4. A half-medium change is only
necessary if the medium appears acidic (yellowish in color)
prior to reaching 80% confluence (see Note 15).
4. Notes
1. If less than 500 mL will be required in a month, smaller vol-

umes can be prepared. Prepare Complete MesenCult™
Medium (Human) by diluting Mesenchymal Stem Cell
Stimulatory Supplements (Human) 1/10 with MesenCult™
MSC Basal Medium. For example, prepare 100 mL of Complete
MesenCult™ Medium (Human) by adding 10 mL of
Mesenchymal Stem Cell Stimulatory Supplements (Human)
to 90 mL of MesenCult™ MSC Basal Medium (Human).The
complete medium is stable at 2–8°C for 1 month. Repeated
freezing and thawing is not recommended as it may lead to
suboptimal performance (i.e. lower cell expansion).
2. Suggested Basal media for culturing MSCs include MesenCult™
MSC Basal medium-Human-(StemCell Cat. #05401), alpha
MEM, or McCoy 5A medium. The supplement consists of pre-
tested FBS. Each FBS lot should be pretested for optimal for-
mation of CFU-F and long-term expansion of human MSCs.
3. CFU-F assays can also be set up using culture-expanded BM
cells to assess the cloning efficiency of the cells. The plating
density for cultured cells ranges from 10 to 150 cells/cm2.
4. Each bone marrow sample is unique and thus the number of

CFU-F may vary depending on a number of factors including
age, presence of disease and previous treatments given to the
patient. To achieve optimal CFU-F number, size, and mor-
phology it is recommended to prescreen FBS batches. There is
no need to supplement the prescreened FBS with any growth
factors.
5. The proliferative ability of each BM sample is donor-dependent
and can be affected by a number of factors including age, dis-
ease, or whether the sample comes from a transplant recipient.
Therefore not all BM samples may be confluent in a week and
a half-medium change may help cells to proliferate in some
samples. It is important to observe the culture on a regular
basis. Cells cultured at high density (>90%) and at high passage
number (>P6) tend to loose their telomerase length and have
reduced DNA methylation leading to reduced proliferation
and differentiation potentials.
6. In order to ensure high cell viability upon freezing and thaw-
ing do not let cells sit in freezing medium at room tempera-
ture. Keep the cells on ice and transfer within 5 min to the
freezing container.
7. Do not vortex cells at anytime.
8. MesenCult™-XF Supplement can be aliquoted into smaller
working volumes and stored at −20°C until required for use.
Repeated thawing and freezing is not recommended. Complete
MesenCult™-XF Medium should be prepared in volumes that
can be used within 5 days. Prepare an amount suitable for your
needs by diluting MesenCult™-XF Supplement 1/5 (final
dilution) in MesenCult™-XF Basal Medium (i.e., 20 mL
MesenCult™-XF Supplement + 80 mL MesenCult™-XF Basal
Medium).
9. Acetic acid 3% with Methylene Blue will lyse red blood cells
and white blood cell membranes. The remaining white blood
cell nuclei will stain lightly with Methylene Blue.
10. Plating different cell densities will ensure that the resulting
numbers of colonies can be scored. The proliferative potential
of CFU-F from various bone marrow samples is widely vari-
able. If too few cells are plated, CFU-F may be undetectable or
the number of colonies scored may be too low to give a reliable
estimation of CFU-F. If too many cells are plated, the CFU-F
may grow such that individual colonies cannot be determined.
Human bone marrow-derived CFU-F colonies cultured in
MesenCult™-XF medium are generally large enough to see
with the naked eye following staining with Giemsa (see
Fig. 3).
Fig. 3. The circled colonies are easily visible macroscopically. It is important to look at the CFU-F cultures under a micro-
scope for confirmation because some colonies may not take up enough stain and could be missed when scored macro-
scopically. The CFU-F assay was performed in a 6-well plate (note the edges of the well in each image). A Lumenera Infinity
2–3 C camera was used to capture the images using Image Pro 6.2 software.
Fig. 4. The morphology of CFU-F colonies generated when MSCs are cultured in MesenCult™-XF Medium (a) differs from
the morphology of CFU-F generated when MSCs are cultured with MesenCult™ Proliferation Kit (b) containing FBS.
CFU-F cultured in MesenCult™-XF Medium have a slightly

different morphology than CFU-F typically obtained when per-
forming the CFU-F assay with the MesenCult™ Proliferation
Kit (Human) (see Fig. 4a, b). Be sure to monitor CFU-F colony
size. MSCs cultured in MesenCult™-XF Medium proliferate
faster than cells cultured in a traditional serum-based medium.
11. The proliferation potential of cells obtained from different bone
marrow donors is highly variable. To ensure that cultures con-
tain an optimal number of cells for expansion, it is recommended
to seed two to three different cell densities. If too few cells are
Fig. 5. It is important that cells are passaged in MesenCult™-XF medium when they reach 80% confluence. Figures a and
b depict cells at an optimal density for passaging.
plated, cells grow too slowly and reach recommended splitting

density too late (cells may start to detach from the surface). If
too many cells are plated, the cells will reach confluence too fast
and will become senescent and lose pluripotency.
12. The cells are ready to be passaged when they reach 80%
confluence (see Fig. 5a, b). Normally cells reach 70–80%
confluence between 9 and 13 days after initial plating of pri-
mary BM mononuclear cells, but this depends on the donor
and initial plating density.
Monitor the color of the medium after day 7: if the medium
appears acidic (yellowish in color) prior to reaching 80%
confluence, a half-medium change can be performed by remov-
ing 1/2 of the medium and replacing it with fresh Complete
MesenCult™-XF Medium warmed to 37°C.
13. During development of this Xeno-free formulation, we tested
different components of the culture system to promote opti-
mal cell growth. The cells cultured in a serum- and Xeno-free
formulation normally do not adhere easily due to the absence
of FBS. FBS is a rich mixture of various proteins including
extracellular matrix which can promote cell adherence. There
are several critical factors to obtain optimal cell adherence and
growth when culturing MSCs in a xeno-Free medium. They
are as follows:
(a) Culture dish. It is essential to use tissue-culture-treated
dishes with strong adherence properties. There are slight
differences in the strength of cell attachment between tis-
sue-culture-treated dishes from different suppliers. The
ones that promote best cell adherence are from Corning
(Catalog #3516) or BD (Falcon™ Catalog #353109 and
Falcon™ Catalog #353136).
(b) Attachment substrate (coating solution). This is one of the

most important steps to ensure strong cell adherence. The
attachment substrate which is an extracellular matrix
should be mixed gently (inverting the tube). Repeated
mixing with a pipette is not recommended (as this may
interfere with the stability of the substrate). Following the
proper coating procedure is critical to achieve strong cell
adherence.
(c) Gentle dissociation solution. During subculture it is critical
to use a gentle dissociation solution to ensure cell surface
molecules are intact. A strong detachment enzyme can
lead to lack of cell adherence following subculture.
(d) Cell plating concentrations. The cell plating densities sug-
gested in this protocol have been optimized for this par-
ticular culture system (attachment substrate, medium
formulation, and dissociation enzyme). For the CFU-F
assay, plating different cell densities will ensure that the
resulting numbers of colonies can be scored. The prolif-
erative potential of CFU-F from various BM samples is
widely variable. If too few cells are plated, CFU-F may be
undetectable or the number of colonies scored may be too
low to give a reliable estimation of CFU-F. If too many
cells are plated, too many CFU-F may grow such that
individual colonies cannot be determined. For the expan-
sion assay, plating too few or too many cells can result in
poor proliferation. If using other attachment substrates
and dissociation enzymes (not as suggested in this proto-
col) with MesenCult™-XF, further optimization is required
to obtain optimal cell adherence and cell growth.
14. For a T-25 cm2 flask pipette the 6 mL cell suspension into a
14 mL polystyrene tube and wash each flask by adding 5 mL
MEM Alpha using a pipette to recover remaining cells. Add
MEM Alpha with a 2 mL pipette to bring the total volume to
12 mL. For a T-75 cm2 flask, collect the 12 mL cell suspension
with a pipette, transfer to a 50 mL conical tube and wash each
flask with 8 mL MEM Alpha to recover remaining cells. Add
MEM Alpha with a 10 mL pipette to the tube to bring the
total volume to 30 mL. It is important to add additional
medium when washing the cells so the MesenCult™-ACF
Enzymatic Dissociation Solution is sufficiently washed from
the cells. Continue to step 11 under Subheading 3.2.4.
15. BM-derived mesenchymal cells cultured in MesenCult™-XF
medium have less hematopoietic contaminating-cells com-
pared to cells cultured in MesenCult™- proliferation medium
containing FBS (see Fig. 6a, b).
Fig. 6. Passage 0 human bone marrow-derived mesenchymal stem cells show less hematopoietic cell contamination when
cultured in MesenCult™-XF Medium (a) compared to serum-based medium (b).
Acknowledgments
The authors would like to thank Betty Hoac and Jacky Yau for
technical assistance, Bert Wognum and Emer Clarke for technical
advice, Terry Thomas and Allen Eaves for continuous support.
References
1. Bruder SP, Jaiswal N, Haynesworth SE (1997) 7. Tondreau T, Meuleman N, Delforge A et al
Growth kinetics, self-renewal, and osteogenic (2005) Mesenchymal stem cells derived from
potential of purified human mesenchymal stem CD133 positive cells in mobilized peripheral
cells during extensive subcultivation and fol- blood and cord blood: proliferation, Oct4
lowing cryopreservation. J Cell Biochem expression, and plasticity. Stem Cells
64:278–294 23:1105–1112
2. Mackay AM, Beck SC, Murphy JM, Barry FP, 8. Roberts IA, Campagnoli IA, Kumar S et al
Chichester CO, Pittenger MF (1998) (2001) Identification of mesenchymal stem/
Chondrogenic differentiation of cultured progenitor cells in human first trimester fetal
human mesenchymal stem cells from marrow. blood, liver and bone marrow. Blood
Tissue Eng 4:415–428 98:2396–2402
3. Pittenger MF, Mackay AM, Beck SC, Jaiswal 9. Int’l Anker PS, Scherjon SA, Kleiburg-van der
RK, Douglas R, Mosca JD, Moorman MA, Keur C et al (2003) Amniotic fluid as a novel
Simonetti DW, Craig S, Marshak DR (1999) source of mesenchymal stem cells for thera-
Multilineage potential of adult human mesen- peutic transplantation. Blood 102:1548–1549
chymal stem cells. Science 284:143147 10. Zuk PA, Zhu M, Ashjian P et al (2002) Human
4. Erices A, Conget P, Minguell JJ (2000) adipose tissue is a source of multipotent stem
Mesenchymal progenitor cells in human cells. Mol Biol Cell 13:4279–4295
umbilical cord blood. Br J Haematol 11. Crisan M, Chen CW, Corselli M, Andriolo G,
109:235–242 Lazzari L, Péault B (2009) Perivascular multi-
5. Debari C, Dell’Accio F, Tylazanowski P et al potent progenitor cells in human organs. Ann
(2001) Multipotent mesenchymal stem cells N Y Acad Sci 1176:118–123
form adult human synovial membrane. Arthritis 12. Friedenstein AJ, Chailakhjan RK, Lalykina KS
Rheum 44:1928–1942 (1970) The development of fibroblast colonies
6. Kuznetsov SA, Mankani MH, Gronthos S et al in monolayer cultures of guinea-pig bone mar-
(2001) Circulating skeletal stem cells. J Cell row and spleen cells. Cell Tissue Kinet
Biol 153:113–114 3:393–403
13. Friedenstein AJ (1980) Stromal mechanisms effects of bone marrow-derived mesenchymal

of bone marrow: cloning in vitro and trans- cells in children with osteogenesis imperfecta.
plantation in vivo. Haematol Blood Transfus Nat Med 5:309–313
25:19–29 22. Lazarus HM, Koc ON, Devine SM, Curtin P,
14. Clarke E, McCann SR (1989) Age dependent Maziarz RT, Holland HK, Shpall EJ, McCarthy
in vitro stromal growth. Bone Marrow P, Atkinson K, Cooper BW, Gerson SL,
Transplant 4:596–597 Laughlin MJ, Loberiza FR Jr, Moseley AB,
15. Gronthos S, Graves SE, Ohta S, Simmons PJ Bacigalupo A (2005) Cotransplantation of
(1994) The STRO-1+ fraction of adult human HLA-identical sibling culture-expanded mes-
bone marrow contains the osteogenic precur- enchymal stem cells and hematopoietic stem
sors. Blood 84:4164–4173 cells in hematologic malignancy patients. Biol
16. Sekiya I, Vuoristo JT, Larson BL, Prockop DJ Blood Marrow Transplant 11:389–398
(2002) In vitro cartilage formation by human 23. Le Blanc K, Frassoni F, Ball L, Locatelli F,
adult stem cells from bone marrow stroma Roelofs H, Lewis I, Lanino E, Sundberg B,
defines the sequence of cellular and molecular Bernardo ME, Remberger M, Dini G, Egeler
events during chondrogenesis. Proc Natl Acad RM, Bacigalupo A, Fibbe W, Ringdén O, Group
Sci U S A 99:4397–4402 for Blood and Marrow Transplantation (2008)
17. Russell KC, Phinney DG, Lacey MR, Mesenchymal stem cells for treatment of ste-
Barrilleaux BL, Meyertholen KE, O’Connor roid-resistant, severe, acute graft-versus-host
KC (2010) In vitro high-capacity assay to disease: a phase II study. Lancet 371:1579–86
quantify the clonal heterogeneity in trilineage 24. Meuleman N, Tondreau T, Delforge A et al
potential of mesenchymal stem cells reveals a (2006) Human marrow mesenchymal stem
complex hierarchy of lineage commitment. cell culture: serum-free medium allows better
Stem Cells 28:788–798 expansion than classical alpha-MEM medium.
18. Phinney DG, Prockop DJ (2007) Concise Eur J Haematol 76:309–316
review: mesenchymal stem/multipotent 25. Kocaoemer A, Kern S, Kluter H, Bieback K
stromal cells: the state of transdifferentiation (2007) Human AB-serum and thrombin-acti-
and modes of tissue repair—current views. vated platelet-rich plasma are suitable alterna-
Stem Cells 25:2896–2902 tives to fetal calf serum for the expansion of
19. Deng J, Petersen BE, Steindler DA, Jorgensen mesenchymal stem cells from adipose tissue.
ML, Laywell ED (2006) Mesenchymal stem Stem Cells 25:1270–1278
cells spontaneously express neural proteins in 26. Schallmoser K, Bartmann C, Rohde E et al
culture and are neurogenic after transplanta- (2007) Human platelet lysate can replace fetal
tion. Stem Cells 24:1054–1064 bovine serum for clinical-scale expansion of
20. Tsuji H, Miyoshi S, Ikegami Y, Hida N, Asada functional mesenchymal stromal cells.
H, Togashi I, Suzuki J, Satake M, Nakamizo Transfusion 47:1436–1446
H, Tanaka M, Mori T, Segawa K, Nishiyama 27. Reyes M, Lund T, Lenvik T, Aguiar D, Koodie
N, Inoue J, Makino H, Miyado K, Ogawa S, L, Verfaillie CM (2001) Purification and ex vivo
Yoshimura Y, Umezawa A (2010) Xenografted expansion of postnatal human marrow meso-
human amniotic membrane-derived mesen- dermal progenitor cells. Blood 98:2615–2625
chymal stem cells are immunologically toler- 28. D’Ippolito G, Diabira S, Howard GA, Menei
ated and transdifferentiated into P, Roos BA, Schiller PC (2004) Marrow-
cardiomyocytes. Circ Res 106:1613–1623 isolated adult multilineage inducible (MIAMI)
21. Horwitz EM, Prockop DJ, Fitzpatrick LA, cells, a unique population of postnatal young
Koo WWK, Gordon PL, Neel M, Sussman M, and old human cells with extensive expansion
Orchard P, Marx JC, Pyeritz RE, Brenner MK and differentiation potential. J Cell Sci
(1999) Transplantability and therapeutic 117:2971–2981
Chapter 21
Isolation and Culture of Mesenchymal Stem Cells from

Mouse Compact Bone
Brenton Short and Ravenska Wagey
Abstract
The bone marrow (BM) of numerous species, including rodents and man, contains a rare population of
cells termed marrow stromal cells or mesenchymal stem cells (MSC). Given the ability of these cells to
differentiate into cells of the osteogenic, chondrogenic and adipogenic lineages, there is considerable inter-
est in utilizing MSCs in a broad repertoire of cell-based therapies for the treatment of human disease.
Before such potential therapies can be realized, a preclinical animal model in which to test and refine strate-
gies utilizing MSC is required. Here we describe methods for the isolation of a highly enriched population
of MSC from mouse cortical/compact bone (CB), quantitation using the colony forming unit-fibroblast
assay (CFU-F) and in vitro expansion. These cells are both multipotent and capable of extensive in vitro
expansion and thus represent an ideal cellular source to explore both the biological properties of MSC as
well as their potential efficacy in a variety of cellular therapies.
Key words: Mesenchymal Stem Cell, MSC, Colony forming unit-fibroblast, CFU-F, Multipotent,
Compact bone
1. Introduction
In contrast to the defined phenotype of human MSC (1, 2), by

which populations of highly purified MSC may be prospectively
isolated, neither the stem nor progenitor cell populations of the
mouse mesenchymal system have been well characterized either
phenotypically or at the molecular level. This, coupled with their
extremely low incidence (3) and a lack of knowledge of their pre-
cise location within the marrow, has meant that much of our cur-
rent understanding of mouse mesencymal stem cells has arisen
from in vitro assays and tissue culture reliant manipulations.
335
336 B. Short and R. Wagey
Pioneering studies conducted by Friedenstein and colleagues

demonstrated that the explantation of single cell suspensions of bone
marrow (BM) from multiple mammalian species at appropriate den-
sities (104–105 cells/cm2) resulted in the outgrowth of adherent col-
onies of cells morphologically resembling fibroblasts (4). Originally
termed fibroblast colony-forming cells, F-CFC, the clonogenic mes-
enchymal precursor cells initiating these colonies have more recently
been termed colony-forming unit fibroblasts (CFU-F).
Mouse mesenchymal cells are typically obtained by flushing the
long bones of the hind limbs followed by culture selection. More
recently phenotypes allowing for the prospective isolation of mesen-
chymal cells from mouse bone marrow using a combination of
hematopoietic lineage exclusion and positive selection with antibod-
ies to stem cell antigen-1 (Sca-1) (5) and the platelet derived growth
factor receptor (PDGFR) alpha-chain (6) have been identified.
We have shown that the CB rather than the BM is the major
source of MSC in the adult mouse. Similar studies have recently
shown that culturing BM depleted compact bone fragments allows
the egress and establishment of bone derived mesenchymal cell
cultures (7). Enzymatic digestion of bone fragments followed by
depletion of mature hematopoietic cells (Lin−) and subsequent
FACS allows the resolution of a population of cells with the com-
posite phenotype Lin−CD45−CD31−Sca-1+. In this chapter, meth-
ods for the isolation of highly enriched populations of MSC from
mouse CB, their quantitation using the CFU-F assay, and in vitro
cell expansion are described.
2. Materials
2.1. Mice The following protocols were developed from experiments performed
using specific pathogen-free (SPF) C57BL6/J (Ly5.2) mice.
2.2. Equipment 1. 70 mm porcelain Mortar and pestle (VWR Catalog

#89038-144).
2. #22 Scalpel blade and #4 scalpel handle.
3. FACSDiva (Becton Dickinson, San Jose, CA).
4. Inverted microscope and Standard light microscope (for cell
counting).
5. Sterile pipettes: (1, 5, and 10 mL).
6. Micropipette with 20 μL, 200 μL, and 1 mL sterile tips.
7. Parafilm® (Sigma Catalog #P7793).
8. 70 mm nylon cell strainer (BD Falcon, Bedford, MA).
9. Falcon™ 5 mL Polystyrene round-bottom tubes (BD. Catalog
#352058).
21 Mouse Mesenchymal Stem Cells 337
2.3. Reagents 1. Dulbecco phosphate buffered saline (PBS) (STEMCELL

Technologies Catalog #37350).
2. PBS + 2% fetal bovine serum (PBS + 2%FBS) (STEMCELL
Catalog #07905).
3. 0.25% solution of Type 1 Collagenase containing 20% FBS
(STEMCELL Catalog #07902) or 3 mg/mL solution
(Worthington Biochemical Corporation, Franklin Lakes, NJ)
supplemented with 20% FBS. Collagenase should be prepared
in PBS and sterile filtered prior to use and may be prepared
fresh or stored frozen at −20°C. Collagenase should be warmed
to 37°C prior to use.
4. Methanol, ACS Grade (BDH Catalog #ACS531).
5. Giemsa Stain Solution (EMD Chemicals Catalog #R03055).
6. 3% Acetic Acid with Methylene Blue (STEMCELL Catalog
#07060).
7. Trypan Blue (STEMCELL Catalog #07050).
8. Ethylenediaminetetraacetic acid (EDTA); cell culture grade is
available from numerous suppliers.
9. Cell isolation buffer (Buffer): PBS (Ca2+ and Mg2+ free)/2%
FBS/1 mM EDTA. Filter-sterilize using 0.2 μm filter and store
at 2–8°C.
10. Propidium Iodide (Sigma).
2.3.1. Easysep® 1. Mouse Mesenchymal Progenitor Enrichment kit for Compact

Enrichment of Mouse Bone (STEMCELL Catalog #19771).
Mesenchymal Stem Cells 2. Purple Easysep™ Magnet (STEMCELL Catalog #18000).
3. Flurochrome conjugated rat antibodies to mouse CD45 and
Ter-119 (BD Pharmingen, Franklin Lakes, NJ).
Or Alternatively
2.3.2. Enrichment of 1. Purified rat antibodies to mouse CD3, CD4, CD5, CD8,
Mouse Mesenchymal Stem CD11b (Mac-1), Gr-1, B220, and Ter-119 (BD Pharmingen,
Cells by Lineage Depletion Franklin Lakes, NJ) diluted accordingly (see Note 1) into a
and FACS single “lineage cocktail.” Lineage antibody cocktail should be
freshly prepared for each experiment.
2. Sheep-anti-rat IgG Dynabeads (Invitrogen Catalog #110-35).
3. FITC-conjugated rat anti-mouse stem cell antigen-1 (Sca-1),
PE-conjugated rat antibodies to mouse-platelet endothelial cell
molecule-1 (PECAM-1/CD31), and CD45 (BD Pharmingen,
Franklin Lakes, NJ) diluted accordingly (see Note 2).
4. FACS analysis buffer consisting of PBS 2% FBS/1 mM EDTA
containing 1 μg/mL Propidium Iodide (Sigma).
2.4. Tissue Culture 1. 14 mL polypropylene tubes (Falcon Catalog #352001).

2. 50 mL conical tubes (BD Catalog #352070).
3. Recommended tissue culture plates/flasks: Corning (Catalog
#3516) or Falcon™ (Catalog #353502).
4. Complete MesenCult™ Medium (Mouse) (STEMCELL
Catalog #05511).
5. 0.25% Trypsin-EDTA (STEMCELL Catalog #07901).
6. Tissue culture incubator with Oxygen sensor for hypoxic cul-
ture (Sanyo Incusafe MCO-18M or Thermo Scientific Forma
CO2-O2 Series II model 3141), or Hypoxic Incubator cham-
ber (STEMCELL Catalog #27310) with gas flow meter
(STEMCELL Catalog #27311).
3. Methods
3.1. Isolation of Cells A cell population derived from the compact bone tissue in the adult
from Compact Bone mouse has been reported to be highly enriched for CFU-F. In the
adult mouse, the CFU-F frequency in the compact bone is
significantly higher than the marrow plug, indicating the site of the
major reservoir of CFU-F is the surrounding bone tissue rather
than the marrow itself. To isolate cells by crushing compact bones
(femur, tibia, and iliac crest), the following procedure is
recommended:
1. Clean a mortar and pestle and dissection instruments (scissors
and forceps) with 70% isopropanol and allow to air-dry in a
sterile biohazard safety cabinet for 30 min. Rinse with sterile
PBS prior to use. Alternatively, all instruments may be pre-
sterilized by autoclaving and used as required.
2. Sacrifice mice by cervical dislocation. Wet the pelt thoroughly
with 70% isopropanol and excise the tibiae, femurs and iliac
crests (see Note 3).
3. Using a #22 scalpel, scrape bones thoroughly to remove mus-
cle, and cut to remove epiphyses. Place cleaned bones in a
50 mL tube containing PBS (Ca2+ and Mg2+ free)/2%
FBS/1 mM EDTA. (henceforth referred to as “Buffer”).
4. Once all bones have been processed, transfer to a 100 mm petri
dish and, using forceps and a fresh scalpel blade, carefully
remove any residual muscle and tendons from bones.
5. Transfer cleaned bones to a mortar containing 10 mL buffer.
Crush bones with pestle, using only enough force to crack
open the bones (see Note 4). Agitate gently to free bone mar-
row (BM) from bone fragments and pipette buffer off. Buffer
containing BM can be filtered (70 μm cell strainer) and used

for other applications.
6. Add 10 mL fresh Buffer and repeat agitation and removal of
BM. Repeat wash step an additional four times (for a total of
six washes) or until the majority of the BM has been removed
(bone fragments turn white in color).
7. Transfer the bone fragments to a 100 mm dish. Add 2 mL of
0.25% Collagenase Type 1 in PBS containing 20% FBS. Ensure
bones are completely covered in solution. Let sit for 3–5 min.
8. Using a #22 scalpel, chop the remaining bone fragments into
fine pieces (1–2 mm fragments). Proper bone fragmentation is
required to release sufficient amount of cells for cell
separation.
9. Transfer the bone fragments to a 50 mL polypropylene tube
and add further 0.25% Collagenase to a final volume of 2 mL
per mouse used, or a minimum of 10 mL.
10. Seal lid with Parafilm® and place tube in a shaking 37°C water
bath at maximum speed for 45 min. If using a bacterial culture
shaker, set speed to ~200 rpm
11. After 45 min, remove the tube from the shaker and add Buffer
to a total volume of 30 mL. Collect supernatant and filter
through a 70 μm cell strainer. Wash bone fragments by mixing
with an additional 10 mL of Buffer and allowing fragments to
settle for 3–4 min. Filter wash through the 70 μm strainer,
combining with the previously collected cells (a final volume of
40 mL).
12. Centrifuge at 300 × g (~1,200 rpm) for 10 min with the brake
on. Remove supernatant and after gently resuspending pellet,
transfer cells into a 5 mL polystyrene tube. Rinse 50 mL tube
with 250 μL Buffer and add to 5 mL tube containing cell sus-
pension (note that small particles and debris may be visible in
the cell suspension).
13. Remove a small aliquot of cells and dilute 1/50 to 1/100 in
3% Acetic Acid with Methylene Blue. Count nucleated cells
using a hemocytometer. Expected cell recovery: 3–5 × 106 cells
per mouse. If the cell yield is less than 5 × 106 cells/mouse, this
is an indication that the marrow was not depleted sufficiently
during processing.
14. Place cells on ice until ready for use.
3.2. Easysep™ This section describes the enrichment of mesenchymal cells using
Enrichment of Cells antibodies designed to remove essentially all hematopoietic cells
Isolated from Mouse whilst leaving the MSCs unlabeled. This procedure is used for pro-
Compact Bone cessing 200–500 μL of sample (up to a maximum of 2.5 × 107 cells)
1. Prepare single nucleated cell suspension at a concentration of

2–5 × 107 cells/mL in Buffer. For samples containing 4 × 106
cells or fewer, resuspend in 200 μL. Cells must be placed in a
5 mL (12 × 75 mm) polystyrene tube to properly fit into the
purple EasySep® Magnet. Falcon™ 5 mL Polystyrene Round-
Bottom Tubes are recommended.
2. Add Mesenchymal Progenitor Cell Enrichment Cocktail con-
taining biotinylated antibodies at 50 μL/mL of cells (e.g. for
0.5 mL of cells, add 25 μL of cocktail). Mix well and incubate
at 2–8°C for 15 min.
3. Add 4 mL buffer and centrifuge cells for 5 min at 300 × g.
Resuspend cells at 2–5 × 107 cells/mL.
4. Add Biotin Selection Cocktail at 250 μL/mL of cells (e.g. for
0.5 mL of cells, add 125 μL of selection cocktail). Mix well and
incubate at 2–8°C for 15 min.
5. Vortex EasySep® Mouse Progenitor Magnetic Microparticles
for 30 s to ensure that the particles are in a uniform suspension
with no visible aggregates.
6. Add the microparticles at 250 μL/mL of cells (e.g. for 0.5 mL
of cells, add 125 μL of microparticles). Mix well and incubate
at 2–8°C for 15 min.
7. Bring the cell suspension to a total volume of 2.5 mL by add-
ing buffer. Mix the cells in the tube by pipetting gently two to
three times. Place the tube (without cap) into the magnet. Set
aside for 5 min.
8. Pick up the EasySep™ magnet, and in one smooth motion
invert the magnet and tube, pouring off the desired fraction
into a new 5 mL polystyrene tube. The magnetically labeled
unwanted cells will remain bound inside the original tube.
Leave the magnet and the tube inverted for 2–3 s, then return
to upright position. Do not shake or blot off any drops that
may remain hanging from the mouth of the tube. The enriched
cells are now ready for further use.
9. Cell purity may be analyzed by FACS using flurochrome con-
jugated antibodies against CD45 and Ter119 (Fig. 1).
3.3. Enrichment of CB 1. Count viable CB cells obtained as described using a viability

Derived MSC by dye such as trypan blue. Typical yields are between 3 and
Immunomagnetic Cell 5 × 106 CB cells per mouse. Centrifuge cells for 5 min at 300 × g,
Separation remove supernatant, and resuspend cells in 200 μL buffer.
2. Prepare the lineage depletion antibody cocktail described in
3.3.1. Lineage Depletion of Subheading 3. Add 50 microlitre of the aforementioned cock-
Compact Bone Derived tail per 5 × 106 viable CB cells and incubate on ice for 25 min.
Cells
3. Wash cells twice with 3 mL ice-cold buffer, aspirate superna-
tant, gently resuspend cell pellet in a minimal volume, and
store on ice.
Fig. 1. EasySep™ Enrichment of Cells isolated from mouse Compact Bone. Compact bone
cells were stained with PE-conjugated anti CD45 and Ter119 antibodies (top panel) to
resolve the target population in the lower left quadrant (1.06 ± 0.5%, n = 10). Following
EasySep™ enrichment (lower panel) these CD45−Ter119− cells are significantly enriched,
comprising 74.5 ± 16% of the total recovered fraction.
4. Aliquot Dynabeads into a 5 mL tube. The total number of

beads used is twice the total number of cells being depleted
(Dynabead solution comprises 4 × 108 beads/mL). As the
beads are added in equal volume in two stages, the initial deple-
tion step thus has a 1:1 bead–cell ratio with the subsequent
step having a large excess of beads to remove cells expressing
low levels of lineage antigens.
5. Wash beads twice using on an EasySep™ magnet with 2 mL
buffer. Resuspend beads in 1 mL.
6. Add half the beads (500 μL) to the CB cell pellet and incubate
on ice with gentle agitation for 5 min.
7. Place cells on the EasySep® magnet for 1 min to facilitate clear-

ance of bead-bound lineage positive cells. In one smooth
motion invert the magnet and tube, pouring off the unbound
cellular fraction into a new 5 mL polystyrene tube.
8. Remove tube from magnet and gently resuspend the bead bound
lineage positive cells in 2 mL Buffer and place the cell suspension
back in the magnet for 1 min. Pour off the non-bound cells as
above and add to the tube containing the lineage negative cells
previously collected. Add remaining 500 μL of beads.
9. Parafilm the lid of the tube containing the beads and depleted
cells and place on a rotator at 4°C for 25 min.
10. Place tube on EasySep™ magnet for 5 min to remove lineage
positive cells. Transfer non-bound cells to a fresh 5 mL tube.
11. Count viable lineage cells using a viability dye such as trypan
blue and store on ice.
3.3.2. Isolation of Compact 1. The lineage depleted CB cells are now prepared for FACS
Bone Derived MSC by FACS analysis. Aliquot an appropriate number of cells (25,000–
30,000) into sterile polystyrene tubes for use as isotype and
compensation controls (see Note 5), retaining the rest of the
cells (the “sort sample”) in the 5 mL tube.
2. Centrifuge all cells and aspirate supernatant leaving approxi-
mately 250 microlitre of Buffer in the tubes.
3. Resuspend cells and add antibodies to appropriate control
tubes (see Note 6) and store on ice under foil or in the dark.
4. Add test antibodies (Sca-1-FITC, CD45-PE, and CD31-PE)
to the sort sample (see Notes 6 and 7) and store on ice under
foil or in the dark.
5. Following a 15 min incubation, wash cells twice with 4 mL
ice-cold Buffer.
6. Resuspend cells in either Buffer alone (unstained control cells
and all fluorochrome-conjugated compensation controls) or
FACS analysis buffer (PI control and sort sample), adding
500 μL to control tubes and 1 mL per 106 cells in the sort
sample, which may need to be divided between multiple tubes.
7. Immediately prior to FACS analysis the sort sample should be
passed through a 70 μm cell strainer to remove any clumps
that may have formed and to prevent blockages.
8. Samples are now analyzed by FACS using a FACSDiva (Becton
Dickinson, San Jose, CA). Unstained cells are used to set forward
and side-scatter parameters (FSC and SSC respectively), whilst
the isotype negative controls are used to quantitate nonspecific
binding of antibodies. Samples containing single color positive
controls (CD45-PE/FITC and Sca-1-FITC) are used in setting
the compensation between flurochrome channels.
Fig. 2. Isolation of Lin−CD45−CD31−Sca-1+ MSC by FACS. Compact bone derived cells

were lineage depleted as described and stained with PE-conjugated anti-CD45 and CD31
and FITC-conjugated anti-Sca-1. 7AAD is used to exclude nonviable cells. (a) A dot-plot
displaying forward scatter versus 7AAD fluorescence is used to select viable cells for
analysis (boxed region). (b) Analysis of PE versus FITC fluorescence resolves a population
of MSC lacking expression of both CD45 and CD31 and expressing high levels of Sca-1
(boxed region).
9. To set gates for sorting, a dot-plot of FSC versus the viability

dye PI is employed as shown in Fig. 2a. The gate should
exclude both dead cells which have incorporated PI and non-
cellular debris (see Note 8).
10. Viable cells gated as described are subsequently displayed on a
dot-plot showing PE versus FITC fluorescence. A gate select-
ing CD45/CD31− Sca-1+ MSC is set (Fig. 2b) and these cells
are collected into a tube containing complete Mouse
MesenCult™ Medium
3.4. CFU-F Assay for 1. Prepare pre-enriched cells at the appropriate cell density.
Compact Bone Cells (a) For setting up CFU-F assay, it is recommended that pre-
enriched cells are plated at 1,000, 5,000, and 10,000 cells
per cm2 in triplicate (e.g. plate 10,000, 50,000, and
100,000 cells/well) of a 6-well plate in 2 mL Complete
Mouse MesenCult™ Medium when culturing cells under
Hypoxic conditions (see Note 9).
(b) When using regular normoxic incubator with
20%O2/5%CO2, plate pre-enriched cells for CFU-F assay
at 10,000, 20,000, and 40,000 cells per cm2 (e.g. plate
100,000, 200,000, and 400,000 cells/well of a 6-well
plate) in 2 mL Complete MesenCult Medium.
(c) When assaying CFU-F from EasySep™ enriched CB, cells
are plated at 50, 100, and 150 cells per cm2 (e.g. plate
500, 1,000, and 1,500 cells/well of a 6-well plate). For
CD45−CD31− Sca-1+ MSC isolated by FACS, cells are
plated at 5, 10, and 25 cells per cm2 (e.g. 50, 100, and 250
cells/well of a 6-well plate).
2. Incubate for 10–12 days.
3.5. Giemsa Staining 1. Remove culture medium from CFU-F cultures and discard.
of CFU-F Colonies Adherent CFU-F colonies will remain attached.
2. Wash colonies once with PBS to remove any residual culture
medium.
3. Fix cells by adding 2 mL methanol, to each well of a 6-well
plate. Incubate for 5 min at room temperature.
4. Remove methanol and discard. Air-dry plates at room tem-
perature (~5 min).
5. Add 2 mL Giemsa Stain Solution to each well of a 6-well plate.
Incubate for 5–10 min at room temperature.
6. Remove Giemsa Stain Solution and rinse with distilled water to
remove unbound stain. Rinse until water remains clear.
7. Discard the distilled water and allow the tissue culture dishes
to dry at room temperature with the lid open.
3.6. Expansion 1. Harvest mouse compact bone cells as described.

Protocol: MSC Isolated 2. Plate unprocessed compact bone cells in a Hypoxic incubator
from Compact Bone or chamber gassed with 5%O2/10%CO2/85%N2 as follows:
(a) 2–5 × 105 cells in 2 mL Complete MesenCult™ Medium
(Mouse) in 1 well of a 6-well plate
(b) 6 × 105–1.2 × 106 cells in 7 mL Complete MesenCult™
Medium (Mouse) in a T-25 cm2 flask.
(c) 2.5–5 × 106 cells in 12 mL Complete MesenCult™ Medium
(Mouse) in a T-75 cm2 flask.
3. When using EasySep™ enriched compact bone cells, plate cells

in a Hypoxia chamber gassed with 5%O2/10%CO2/85%N2 as
follows:
(a) 2.5–5 × 103 cells in 2 mL Complete MesenCult™ Medium
(Mouse) in 1 well of a 6-well plate
(b) 5–10 × 103 cells in 7 mL Complete MesenCult™ Medium
(c) 2.5–5 × 104 cells in 12 mL Complete MesenCult™ Medium
4. Culture cells at 37°C in 5%O2/10%CO2/85%N2 for 10–14 days
until an adherent cell layer has formed. After 8 days, when the
color of the media has turned orange, a half medium change
may be performed.
5. Observe mesenchymal cell cultures microscopically after 7 days
to determine confluency. When the cells have reached 80%
confluency or once individual CFU-F are observed that are
large (5 mm diameter or larger) and locally confluent, they are
ready to passaged. Note that time to confluency/first passage
will depend upon seeding density and oxygen tension in which
cells are plated, with hypoxia cultured cells expanding more
rapidly than normoxic cultures.
6. When passaging mesenchymal cells, remove tissue culture
media with an aspirator or pipette. Wash cultures three times in
PBS and add 1 mL 0.25% Trypsin-EDTA per well of a 6-well
plate, 2.5 mL per T-25 cm2 flask, or 5 mL per T-75 cm2 flask.
Incubate at 37°C for 5 min and observe microscopically to
assess cell detachment. If cells have not fully detached return to
incubator. Cells should be in Trypsin-EDTA for no longer
than 10 min.
7. Quench trypsin by addition of an equal volume of complete
medium. Centrifuge cells for 5 min at 400 × g. Aspirate super-
natant and count viable cells using a hemacytometer and viabil-
ity dye.
8. Replate cells at 3–6 × 104 cells per well of a 6-well plate,
1–2 × 105 cells per T-25 cm2 flask or 3–6 × 105 cells per T-75 cm2
flask. Culture cells at 37°C in 5%O2/10%CO2/85%N2. Examine
cultures microscopically after 3–4 days and passage as described
once cells reach 80% confluence.
4. Notes
1. We routinely use BD Pharmingen antibodies at a final dilution

of 1/250; however, each antibody may be titrated individually
on BM to ensure optimum staining.
2. Different antibody dilutions are required following lineage

depletion as there are fewer cells to be stained. We routinely
use 0.5 μL of fluorochrome-conjugated antibody per 106 lin-
eage negative cells or a minimum of 1 μL if fewer than 106 cells
are obtained
3. When harvesting bones, make an incision in the skin over the
abdomen and carefully peel the skin from the lower half of the
animal. Collect all three bones (tibia, femur, and iliac crest)
together by cutting (with scissors) along the spine up from the
tail. Remove and discard the foot, then cutting through the
knee joint separate the tibia and femur. The iliac crest can be
detached from the femur by holding the distal end of the femur
and using the scalpel to dislocate the iliac crest from the femur.
The iliac crest is difficult to clean; cut in half above the socket
for the head of the femur and discard the lower portion. Tibias
and femurs may be cleaned efficiently by holding one end of
the bone with forceps and scraping along the bone with a scal-
pel blade.
4. When crushing bones in the mortar it is important to use the
least amount of force possible, the goal being to crack the
bones and facilitate removal of the BM rather than to pulverize
the bone. DO NOT use a circular grinding motion; rather
apply downwards force to the pestle to fragment the bones. A
video demonstrating the procedure may be found on the
STEMCELL Technologies website (www.stemcell.com).
5. Several different control tubes are required to ensure correct
setup for FACS. You should have an unstained sample, one
containing cells with the viability dye (PI) only, isotype matched
control antibodies for Sca-1, CD45, and CD31 as well as posi-
tive control antibodies. For positive controls useful in compen-
sation settings, you may wish PE and FITC-conjugated
antibodies to CD, as the majority of the lineage negative cells
obtained are primitive hematopoietic cells expressing CD45
and thus provide a strong signal. A tube containing cells labeled
with Sca-1-FITC may be used to fine tune the compensation
settings as the level of expression of this antigen by MSC is
typically a log brighter than that of CD45.
6. We routinely use 0.5 μL of fluorochrome-conjugated antibody
per 106 lineage negative cells or a minimum of 1 μL if fewer
than 106 cells are obtained. For isotype control and compensa-
tion tubes where few cells are labeled, a 1/500 dilution (0.5 μL
of antibody in 250 μL) is sufficient.
7. Using this protocol you will obtain several distinct cellular frac-
tions. If you wish to sort other CB-derived populations in
addition to MSC, you may wish to use separate fluorochromes
for CD31 and CD45. In this protocol, the two antigens are
labeled with the same fluorochrome as an easy means of exclud-

ing cells expressing these markers. The Lineage−Sca-1+CD31+
fraction represents vascular endothelial cells whilst the
Lineage−Sca-1+CD45+ are hematopoietic progenitor cells.
8. When setting forward scatter, it is important to note that this
protocol generates cell samples containing bone fragments.
FSC settings should be such that these fragments are excluded
as in general the bone debris is significantly smaller than the
cellular fraction.
9. Culture of mouse MSC in low oxygen tension is a crucial fac-
tor in the successful culture of these cells. Comparison of
CFU-F frequencies between CB-derived cells cultured in a
standard atmospheric oxygen tension (20% O2) versus 5% O2
reveals a sevenfold increase in the number of colonies when
grown in low oxygen conditions. If you cannot dedicate an
entire incubator to these conditions, an alternate method is to
culture the cells in humidified, airtight hypoxia chambers
gassed from a cylinder containing premixed hypoxic “triple
mix” gas: 5% O2, 10% CO2, 85% N2.
References
1. Simmons PJ, Torok-Storb B (1991) Identification Kucia M, Ratajczak MZ, Krebsbach PH (2010)
of stromal cell precursors inhuman bone marrow Prospective identification and skeletal localiza-
by a novel monoclonal antibody, STRO-1. Blood tion of cells capable of multilineage differen-
78(1):55–62 tiation in vivo. Stem Cells Dev 19(10):
2. Gronthos S et al (2003) Molecular and cellular 1557–1570
characterisation of highly purified stromal stem 6. Morikawa S, Mabuchi Y, Kubota Y, Nagai Y,
cells derived from human bone marrow. J Cell Niibe K, Hiratsu E, Suzuki S, Miyauchi-Hara C,
Sci 116(Pt 9):1827–1835 Nagoshi N, Sunabori T, Shimmura S, Miyawaki A,
3. Phinney DG et al (1999) Plastic adherent stromal Nagagawa T, Suda T, Okano H, Matsuzaki Y
cells from the bone marrow of commonly used (2009) Prospective identification, isolation, and
strains of inbred mice: variations in yield, growth, systemic transplantation of multipotent mesen-
and differentiation. J Cell Biochem 72(4): chymal stem cells in murine bone marrow. J Exp
570–585 Med 206(11):2483–2496
4. Friedenstein AJ, Gorskaja JF, Kulagina NN 7. Zhu H, Guo ZK, Jiang XX, Li H, Wang XY, Yao
(1976) Fibroblast precursors in normal and HY, Zhang Y, Mao N (2010) A protocol for
irradiated mouse hematopoietic organs. Exp isolation and culture of mesenchymal stem cells
Hematol 4(5):267–274 from mouse compact bone. Nat Protoc
5. Taichman RS, Wang Z, Shiozawa Y, Jung Y, 5:550–560
Song J, Balduino A, Wang J, Patel LR, Havens AM,
Chapter 22
Generation of a Pool of Human Platelet Lysate

and Efficient Use in Cell Culture
Katharina Schallmoser and Dirk Strunk
Abstract
Human platelets represent a promising source of bioactive substances as growth factors not just for in vivo
wound healing and tissue repair, but also for the expansion of human stem and progenitor cells in vitro.
The replacement of fetal bovine serum (FBS) as a standard culture supplement by human platelet-derived
growth factors now allows for the GMP-compliant implementation of various cell therapeutics in the
growing field of regenerative medicine.
For this purpose a protocol for the preparation of human platelet lysate (HPL) by several freeze–thaw
cycles has been developed, resulting in platelet fragmentation and the release of stored growth factors. By
pooling up to 15 U of HPL derived from individual blood donors, a virtually standardized product is
achieved. The depletion of platelet particles and fragments in a final centrifugation step reduces the risk of
alloimmunization against platelet antigens and the formation of aggregates in cell culture.
The successful application of pooled human platelet lysate (pHPL) as a culture medium supplement
for the ex vivo propagation of human mesenchymal stem/progenitor cells (MSPCs) and endothelial col-
ony forming progenitor cells (ECFCs) indicates the feasibility of this animal serum-free source of growth
factors. Further studies will evaluate efficacy and safety of pHPL.
Key words: Platelet-rich plasma, Pooled human platelet lysate, Apheresis, Buffy coat, Platelet concentrate,
Platelet-derived growth factors, Mesenchymal stem/progenitor cells (MSPCs), Endothelial colony-
forming progenitor cells (ECFCs)
1. Introduction
Platelets are essential for blood coagulation, wound healing and tis-
sue repair. In their specific granules they store a plethora of coagula-
tion factors, cytokines, chemokines, and growth factors such as
platelet-derived growth factors (PDGFs), epidermal growth factor
(EGF), basic fibroblast growth factor (bFGF), transforming growth
factor-β (TGF-β), hepatocyte growth factor (HGF), insulin-like
growth factor-1 (IGF-1), and also vascular endothelial growth
349
350 K. Schallmoser and D. Strunk
factor (VEGF) [1]. At sites of tissue or vascular injury, platelets are

attracted and activated leading to the release of these substances into
the immediate environment. First of all, endothelial defects are
efficiently sealed by clot formation. The platelet-derived growth fac-
tors further stimulate fibroblasts, endothelial and smooth muscle cells
to proliferate and repair injured tissue [2, 3]. Therefore, the topical
application of platelet-rich plasma (PRP) preparations in chronic
wounds, ulcera, tendon and bone defects is one focus of regenerative
medicine [4–7]. This stimulatory effect of platelet-derived growth
factors can also be used for cell propagation in vitro. Successful
replacement of fetal bovine serum by human platelet lysate (HPL) in
cell culture has been proven by several reports [8–12]. Notably,
diverse preparation methods for platelet lysates prevent an objective
comparison of the results. Nevertheless, the avoidance of animal
serum components appears to be mandatory for a GMP-compliant
development of cell therapeutics for clinical application [13]. The
standardized production of a large pool of human platelet lysate
(pHPL) can balance variations in the concentration of growth factors
[14, 15]. We have established a protocol for the generation of pHPL
starting from buffy coat-derived or single donor apheresis platelet
concentrates [16]. Alternatively, pHPL can be prepared from plasma-
reduced platelet concentrates with platelet additive or human albu-
min solution [17]. Evidence exists that pHPL may be efficiently
utilized for the ex vivo expansion of human bone marrow-, cord
blood-, and cord-derived mesenchymal stem/progenitor cells
(MSPCs) as well as human cord blood- and peripheral blood-derived
endothelial colony-forming progenitor cells (ECFCs) [18–20].
2. Materials
2.1. Preparation 1. Cell separator suitable for platelet apheresis (e.g., Amicus®
of Single Donor [Fenwal, Inc., Lake Zurich, Illinois, USA] or Trima Accel®
Platelet Concentrates [CaridianBCT, Zaventem, Belgium] or Fresenius AS-TEC 204
by Apheresis [Fresenius AG, Bad Homburg, Germany] or Haemonetics®
MCS®+ [Haemonetics Corporation, Braintree, MA, USA]).
2. Platelet collection set according to the respective cell separator.
3. Platelet additive solution (SSP+, MacoPharma, Tourcoing, France).
Or alternatively
2.2. Whole Blood 1. One unit of a whole blood donation (450 ± 45 mL).
Donation and 2. Standard citrate phosphate dextrose top-and-bottom quadru-
Preparation of Buffy ple bag (MacoPharma).
Coat-Derived Platelet
3. Cooling unit (butane-1,4-diol plate, Hemocare COMPOCOOL
Concentrates
system, Fresenius Kabi, Bad Homburg, Germany).
Pool Human Platelet Lysate (pHPL) Generation 351
4. Component separator (Compomat G4; NPBI, Amsterdam,

The Netherlands).
5. Buffy coat units from four blood donors of blood group O
(see Note 1).
6. One unit fresh plasma of blood group AB or alternatively
250 mL platelet additive solution (SSP+, MacoPharma), or
human albumin solution (50 g/L, e.g., Albunorm®,
Octapharma Pharmazeutika, Vienna, Austria) supplemented
with 10% acid citrate dextrose (ACD, e.g., Fresenius Hemocare
Austria GmbH, Eugendorf, Austria) (see Note 1).
7. Leukocyte filter (AutoStop BC, Pall Medical, Portsmouth,
England).
8. Platelet storage bag (ELX platelet bag, Pall Medical).
2.3. Preparation 1. Sterile connection device (TSCD-II, Terumo Europe N.V.,

of Pooled Human Leuven, Belgium).
Platelet Lysate from 2. Tube sealing system (Composeal, Fresenius Kabi, Bad
Platelet Concentrates Homburg, Germany).
3. Pooling double bag (each 3,500 mL, originally for puncture,
Macopharma).
4. Storage bag (600 mL, Baxter Healthcare Corporation,
Deerfield, IL, USA).
2.4. Preparation 1. Storage vials (for example 50 mL Falcon tubes, Becton

of Platelet Fragment- Dickinson BD, Germany).
Depleted pHPL
2.5. Use of pHPL for 1. Alpha-modified Minimum Essential Medium (α-MEM, Sigma-
the Propagation of Aldrich, St. Louis, MO).
Mesenchymal Stem/ 2. Preservative-free heparin ([2 U/mL] Biochrom AG, Berlin,
Progenitor Cells and Germany) (see Note 2).
Endothelial Colony- 3. l-Glutamine ([2 mM] Sigma-Aldrich).
Forming Progenitor
Cells
4. Penicillin [10,000 U/mL]/Streptomycin [10 mg/mL]
(Sigma-Aldrich).
2.5.1. Components of the 5. pHPL (10% v/v) replacing fetal bovine serum (FBS) (see
Medium for MSPC Culture Notes 3 and 4).
6. Sterile bottle filter (0.22 μm; Millipore Corporate, Billerica, MA).
2.5.2. Components of the Components of the modified supplemented endothelial cell growth
Medium for ECFC Culture medium EGM-2 (Lonza, Walkersville, Inc., if not otherwise stated):
1. Endothelial cell basal medium (EBM-2).
2. Epidermal Growth Factor (EGF [0.5 mL/500 mL]).
3. Vascular Endothelial Growth Factor (VEGF [0.5 mL/500 mL]).
4. Basic Fibroblastic Growth Factor (b-FGF [2 mL/500 mL]).
5. Insulin-like Growth Factor-1 (IGF-1 [0.5 mL/500 mL]).

6. Ascorbic Acid (0.5 mL/500 mL) all reagents (items 2–6)
provided as “SingleQuots” by Lonza.
7. Penicillin [10,000 U/mL]/Streptomycin [10 mg/mL] (Sigma-
Aldrich).
8. l-Glutamine ([2 mM] Sigma-Aldrich).
9. Preservative-free Heparin ([10 U/mL] Biochrom AG) (see
Note 2).
10. pHPL (10% v/v) replacing fetal bovine serum (FBS) (see Notes
3 and 4).
11. Sterile bottle filter (0.22 μm; Millipore Corporate, Billerica, MA).
3. Methods
The starting material of pHPL may originate from single donor

apheresis platelet concentrates (PCs) or from whole blood dona-
tions using buffy coat-derived PCs. Notably, this procedure bears
the advantage of utilizing otherwise discarded blood components.
When pHPL is used for the expansion of human cell therapeutics,
the donors have to fulfill the respective national regulations for
blood donation. In this case the preparation of PCs is performed
according to the international guidelines [21]. In the processing of
PCs to pHPL, a closed bag system is recommended. In Fig. 1a, b,
two different modalities of PC preparation are summarized. In
Fig. 2a–c, the pooling of HPL units and preparation of platelet
fragment-depleted pHPL are shown.
3.1. Preparation 1. After written informed consent and medical accreditation, the
of Single Donor donor undergoes platelet apheresis.
Platelet Concentrates 2. For the preparation of pHPL, donors of blood group O are
by Apheresis preferred (see Note 1)
3. Based on the donor’s initial blood platelet count and his whole
blood volume a PC containing 3–6 × 1011 platelets per unit is
collected (see Note 5).
4. Apheresis technology provides leukocyte-depleted platelets
without any need for filtration. Platelets are automatically
resuspended in an adequate amount of fresh blood group AB
plasma resulting in approximately 200 mL for a single dose PC
(see Note 1). Alternatively, additive solution may be used for
resuspension reducing the plasma portion of the PC to 40%.
5. After taking samples for quality control and sterility testing,
the PC is stored at 20–24°C under agitation until further pro-
cessing (see Note 6).
Or alternatively
Fig. 1. Preparation of platelet concentrates (PCs). (a) PCs may be produced from whole blood donations of healthy blood
donors. After an initial centrifugation step of 4,250 × g the whole blood unit (400 ± 45 mL) is separated into the fractions of
plasma, red blood cells, and the intermediate buffy coat layer. Four buffy coat units are pooled with one plasma unit (or an
alternative suitable solution) and are centrifuged at 340 × g. The supernatant fluid consisting of platelets (plts) suspended
in plasma or additive solution is separated from the pellet, transferred into the storage bag through a leukocyte-depleting
filter and is finally named platelet concentrate (PC). (b) Alternatively, PCs may be prepared by single donor apheresis.
3.2. Whole Blood 1. Whole blood (450 ± 45 mL) is collected from a healthy donor
Donation and after written informed consent into a quadruple bag contain-
Preparation of Buffy ing citrate-phosphate-dextrose (63 mL) as anticoagulant.
Coat-Derived Platelet 2. Until further processing the blood bag is rapidly cooled to
Concentrates 20–24°C using a cooling unit containing butane-1,4-diol and
is kept at 22 ± 2°C for up to 18 h.
3. For separation of blood components, the whole quadruple bag
system is packed and centrifuged for 13 min at 4,250 × g at 22°C
to sediment platelets and leukocytes to the buffy coat (BC) layer.
4. By a component separator (Compomat) the fraction of red
blood cells (RBCs) and the plasma supernatant are separated
automatically from the intermediate BC layer by transfer into
Fig. 2. Preparation of pooled human platelet lysate (pHPL). (a) The platelet concentrates are frozen at −30°C for at least
24 h. During thawing at 37°C the platelets are lysed and stored growth factors are released into the plasma or alternative
solution. (b) Up to 15 HPL units are pooled to one batch of pooled HPL (pHPL) with a final volume of 3–4 L. (c) The pHPL
batch is divided into suitable aliquots of 100–150 mL and these bags are frozen again resulting in a more efficient platelet
fragmentation. In the last step the pHPL aliquots are thawed and centrifuged. The supernatant solution is separated from
the pellet and is now defined as platelet fragment-depleted pHPL. Aliquots are again frozen until use in cell culture.
the respective satellite containers. The RBC and plasma bags

are then disconnected from the BC bag by sealing.
5. After a resting period of 4 h following the first centrifugation
step, the bags of four BC units of blood group O and one fresh
unit of AB plasma are sterilely connected in a line. The content
is pooled by gravity into the lowest bag. Alternatively, the four
BC units may also be pooled with 250 mL platelet additive
solution (e.g., SSP+, Macopharma), or human albumin solu-
tion (50 g/L, e.g., Albunorm®, Octapharma) supplemented
with 10% ACD (e.g., Fresenius) instead of plasma [17].
6. The empty bags are removed and a platelet storage bag is con-
nected to the BC/plasma pool [16].
7. In a further centrifugation step (“soft spin”, 340 × g for 6 min
at +22°C) a supernatant platelet-rich solution is generated.
8. By manually squeezing the supernatant, the platelet rich solu-
tion is transferred through an inline leukocyte depletion filter
to the storage bag and is further defined as platelet concentrate
(PC). The primary bag containing the residual BC is discon-
nected and appropriately discarded.
9. For sterility testing [21] a small satellite container is filled with
approximately 20 mL of the PC and is disconnected by sealing
(see Note 6). In this step the PC is also de-aerated to remove
residual air bubbles in the bag.
10. For random quality control testing [21] a further sample of
approximately 3 mL is sterilely taken (as indicated in Table 1).
3.3. Preparation of 1. Within 24 h after preparation, the PCs are frozen at −30°C in
Pooled Human Platelet the original storage bag without further manipulation (first
Lysate from Platelet freeze step).
Concentrates 2. For further processing, appropriate sterility and donor testing
results need to be available. In one procedure up to 15 frozen
PCs are thawed in a water bath at 37°C (see Note 7). This
rapid increase in temperature leads to a lysis of platelet mem-
branes and to the release of stored growth factors into the
solution. The PC-derived product is further defined as human
platelet lysate (HPL).
3. For pooling the single HPL bags are connected consecutively
to the pooling double bag (MacoPharma) and the lysate is
transferred into these two bags (see Note 8). The empty HPL
bags are disconnected by sealing. By mixing the content of the
double bag, a final volume of 3–4 L of pooled human platelet
lysate (pHPL) is generated. For sterility check of the pooled
product, a bag (Baxter) is connected to take a sample of 20 mL
pHPL. Thereafter this bag is also disconnected by sealing.
4. To get suitable volumes for further processing we recommend
aliquoting the pHPL. For this reason bags (applicable for a
Table 1
Quality control of single donor apheresis or buffy coat (BC)-derived platelet
concentrates (PCs) in accordance to [17]
Parameter to be checked Quality requirement Frequency of control
Volume >40 mL per 60 × 109 platelets All units

Sterility testinga Negative All units
9
Platelet content >200 × 10 /unit 1% of all units with a minimum
Apheresis PCs >240 × 109 per 4 pooled units of 10 units per month
BC-derived PCs
Residual leukocytes after leukocyte <1.0 × 106/PC unit 1% of all units with a minimum
depletion of 10 units per month
pH measured (+22°C) at the end >6.4 1% of all units with a minimum
of the recommended shelf lifeb of 4 units per month
a
Sterility testing includes the culture of potential aerobic and anaerobic microorganisms monitored for up to 7 days
under semiautomatic incubation (e.g., BacT/ALERT, bioMerieux, Marcy l’Etoile, France)
b
The recommended shelf life for clinically applied PCs is 5 days due to the increasing risk of bacterial contamination after
this period
maximum volume of 600 mL, Baxter) are connected to the

pooling double bag and volumes of up to 250 mL pHPL are
transferred to these smaller bags, then disconnected by sealing.
5. The rate of platelet fragmentation and the amount of released
growth factors can be increased by a second freeze–thaw cycle.
Therefore, the bags of pHPL aliquots are again frozen at −30°C.
6. After at least 24 h the pHPL bags are thawed again in a water
bath at 37°C. For further processing the lysate is transferred
into 50 mL vials (Falcon tubes, BD) by cutting the tubing of
the bag using sterile scissors and pouring the contents into the
vials. This step is performed in a laminar flow hood to avoid
bacterial or fungal contamination.
3.4. Preparation 1. Due to lysis, the pHPL solution contains high amounts of plate-
of Platelet Fragment- let particles and membrane fragments. In cell culture this can lead
Depleted pHPL to aggregates and bears the potential risk of alloimmunization
against platelet antigens in vivo. To remove these particles, the
pHPL vials are centrifuged at minimum 4,000 × g for 15 min at
+4°C. In a laminar flow hood the supernatant solution is
transferred into the final storage vials (e.g., 50 mL Falcon
tubes, BD), the platelet pellets are discarded appropriately.
2. Suitable aliquots of 30–50 mL of platelet fragment-depleted
pHPL are frozen again at −30°C and stored until use in cell
culture (see Notes 3 and 4).
3.5. Use of pHPL As pHPL is a promising substitute for the standard culture supple-
for the Propagation ment FBS [9], we tested both medium supplements for the culture
of Mesenchymal of human MSPCs and ECFCs. As shown in Fig. 3a–c in MSPC
Stem/Progenitor Cells cultures proliferation rates and clonogenicity revealed a significantly
and Endothelial higher efficiency of pHPL-supplementation for in vitro cell propa-
Colony-Forming gation. However, this effect was less prominent in ECFC culture
Progenitor Cells possibly caused by supplementation of EGM-2 medium with other
growth factors. In Fig. 4a–e results of pHPL titration to select the
optimal concentration in MSPC culture medium are shown. MSPC
proliferation and colony formation were higher in medium supple-
mented with 10% pHPL; therefore, we recommend this as stan-
dard pHPL concentration in culture (see Note 12).
The following steps describe the preparation of media for
the culture of MSPCs and ECFCs:
3.5.1. Preparation 1. For MSPC culture the basal medium is α-MEM (Sigma-
of MSPC Medium Aldrich)
2. For 500 mL α-MEM an aliquot of 57 mL pHPL (10% v/v) is
thawed at 37°C in a water bath.
3. In a laminar flow hood 500 mL of α-MEM is transferred to the
top of a filter flask (Millipore) and is supplemented with
(a) 226 μL preservative-free Heparin ([2 U/mL] Biochrom)
(see Notes 2 and 9).
(b) 5 mL l-Glutamine ([2 mM] Sigma-Aldrich) and
(c) 10 mL Penicillin/Streptomycin (Sigma-Aldrich) (see Note 10).
4. Finally 57 mL of pHPL is added and the supplemented medium
is sterile-filtered (Millipore) (see Notes 3, 4, and 11).
5. The complete medium is warmed to 37°C before use or can be
stored up to 48 h at 4°C.
3.5.2. Preparation 1. For ECFC culture the basal medium is EBM-2 (Lonza)
of Medium for ECFCs 2. For 500 mL EBM-2 an aliquot of 57 mL of pHPL (10% v/v)
is thawed at 37°C in a water bath.
3. In a laminar flow hood 500 mL of EBM-2 is transferred to the
top of a filter flask (Millipore), the following reagents are recon-
stituted or thawed at 37°C and added to the basal medium:
4. Epidermal Growth Factor (EGF [0.5 mL/500 mL])
5. Vascular Endothelial Growth Factor (VEGF [0.5 mL/500 mL])
6. Basic Fibroblastic Growth Factor (b-FGF [2 mL/500 mL])
7. Insulin-like Growth Factor-1 (IGF-1 [0.5 mL/500 mL])
8. Ascorbic Acid (0.5 mL/500 mL), all reagents (steps 4-8) pro-
vided as “SingleQuots” by Lonza
9. 10 mL Penicillin/Streptomycin (Sigma-Aldrich), (see Note 10).
358 COMPARISON OF FBS AND pHPL IN CELL CULTURE
a b
Cell number Population doubling time [days]
1x108 3.0
* *
2.5
1x106
2.0
4 FBS FBS
1x10
pHPL 1.5 pHPL
1.0
1x102
0.5
0
1x10 0.0
MSCs ECFCs MSCs ECFCs
c
MSCs
500µm 500µm
ECFCs
500µm 500µm
FBS pHPL
Fig. 3. Comparison of fetal bovine serum (FBS) and pooled human platelet lysate (pHPL) as medium supplements for the
culture of human mesenchymal stem/progenitor cells (MSPCs) and endothelial colony-forming progenitor cells (ECFCs).
The proliferation rate of human cord-derived MSPCs and ECFCs was compared in FBS- and pHPL-supplemented media
(10% in MSPC culture, 5% in ECFC culture) (see Note 12 ). Cells were seeded at 100–300/cm2 (MSPCs; n = 5) and 150–
500/cm2 (ECFCs; n = 4) and were harvested by trypsinization when reaching 80–90% confluence after 7–12 days (MSPCs)
and 7–14 days (ECFCs). (a) MSPCs in pHPL-supplemented medium reached significantly higher cell numbers than in FBS-
supplemented medium (8.6 ± 0.4 × 106 vs. 1.3 ± 0.5 × 106; *p < 0.05). This effect of pHPL was less pronounced in ECFC
cultures (7.2 ± 0.7 × 106 in pHPL- vs. 4.6 ± 0.7 × 106 in FBS-supplemented medium; p > 0.05) probably due to the primary
content of growth factors in the basal medium EGM-2. (b) The population doubling time was significantly shorter in MSPCs
in pHPL- than in FBS-supplemented medium (1.2 ± 0.2 vs. 2.2 ± 0.5 days, *p < 0.05) but similar in both culture conditions
of ECFCs (1.7 ± 1.8 and 1.8 ± 0.2 days). Results are shown as mean ± SEM. (c) Representative microphotographs of MSPCs
and ECFCs cultured in FBS- and pHPL-supplemented media are shown, taken at the day of harvest.
Fig. 4. (continued) 2.7 × 106 cells. A reduced proliferation rate was observed for MSPCs in 1% pHPL (1.3 × 106 cells
after 14 days). (b) These differences could also be confirmed by analyzing the time per population doubling (4.5, 2.2, 1.9
and 1.3 days for 1, 2.5, 5 and 10% pHPL-supplementation, respectively). (c) The cloning efficiency (calculated as the
percentage of colony-forming cells out of the total number of seeded cells per plate) was similar for MSPCs cultured
in 2.5, 5 and 10% pHPL- (35%, 36% and 38%, respectively) but was decreased in 1% pHPL-supplementation (23%)
despite a longer culture period of 14 days vs. 9 days. (d) Photograph showing culture plates stained for colony formation
analysis after culture in different pHPL-concentrations, the CFU-F counts are given as mean ± SEM. (e) Representative
microphotographs were taken immediately before harvest. The various cell densities reflect the different propensities for
proliferation depending on the concentrations of pHPL supplementation.
TITRATION OF pHPL IN MSC CULTURE 359
a Cell number
b Population doubling time [days]
5
6x 106
1 % pHPL 4
6 2.5 % pHPL
4 x 10 5 % pHPL 3
10 % pHPL
2
6
2 x 10
1
0 pHPL 1% 2.5 % 5% 10 %
0 2 4 6 8 10 12 14
Days of culture Culture
14 Days 9 Days 7 Days
c Cloning efficiency [%] d Culture 14 Days 9 Days

50
40
30
20
10
pHPL 1% 2.5 % 5% 10%

pHPL 1% 2.5 % 5% 10 %
Culture CFU-F Count
14 Days 9 Days 38.5±5.5 57.5 ±0.5 60.0±5.0 63.0±0.0
[mean ± SEM]
e pHPL 1% 2.5% 5% 10%
Day 7
Day 9
Day 14
Fig. 4. Titration of pHPL to optimize the efficiency in human MSPC culture. To select the optimal concentration of pHPL in
the basal medium α-MEM for the culture of MSPCs, supplementation with 1, 2.5, 5 and 10% pHPL was tested by compar-
ing the proliferation rates (see Note 12). Human umbilical cord-derived MSPCs were seeded at 1,000/cm2 and cells were
harvested after reaching 80–90% confluence (after 7 days with 5 and 10%, after 9 days with 2.5% and 14 days with 1%
pHPL). At confluence cell detachment in single dense colonies can occur. Analysis of clonogenicity was performed by seeding
MSPCs at three cells per cm2 in 55 cm2 plates. The colony forming units of fibroblasts (CFU-F) were stained (for technical
details see ref. 22) and counted after 9 days (2.5–10% pHPL-supplementation) and 14 days (1% pHPL). (a) After 7 days
of culture, MSPCs in 10% pHPL reached the highest cell number of 5.8 × 106 cells compared to 2.0 × 106 cells in 5%
pHPL. As MSPCs in 2.5% pHPL tended to detach after 9 days, cells were harvested before reaching confluence resulting in
10. 5 mL l-Glutamine ([2 mM] Sigma-Aldrich)

11. 1 mL preservative-free Heparin ([10 U/mL] Biochrom AG)
(see Notes 2 and 9)
12. Finally 57 mL of pHPL is added and the supplemented medium
is sterile-filtered (Millipore) (see Notes 3, 4, and 11).
13. The complete medium is warmed to 37°C before use or can be
stored up to 48 h at 4°C.
4. Notes
1. The combination of blood group O platelets in AB plasma

(of a male donor) or in a suitable alternative solution is useful
to avoid the presence of blood group determinants, isoaggluti-
nines and possibly pregnancy-induced antibodies against
human leukocyte antigens in cell culture.
2. Heparin has to be free of preservatives otherwise the cell
proliferation is inhibited remarkably.
3. For sterile filtration of more then 300 mL supplemented
medium it may become necessary to use two filters due to
sticked filter pores.
4. In our experience the storage period of pHPL samples at −30°C
should not exceed 6 months, otherwise a decrease of efficiency
cannot be excluded. pHPL samples stored at −30°C for 2 years
were tested and showed a reduction of approximately 10% in
the population doubling rate of expanded MSPCs.
5. PCs prepared for clinical application with more than 4 × 1011
platelets per unit are routinely split into two therapeutic single
dose PCs.
6. The recommended sterility testing of apheresis PCs and
BC-derived PCs is summarized in Table 22.1.
7. Frozen PCs should be thawed in a protective cover until ice
clots disappear but without warming up.
8. Although the double bag (Macopharma) has a potential vol-
ume of 2 × 3,500 mL, we recommend filling each bag only
with a maximum of 2,000 mL to be able to sufficiently mix the
content of both.
9. It is absolutely mandatory to first add heparin and then to
add pHPL to the basal medium. Otherwise, following the
pHPL addition to the medium, the coagulation factors are
activated due to the Ca2+ content leading to clotting and gel
formation.
10. When using the culture medium for the propagation of clinically
applied cell therapeutics, we avoid the addition of antibiotics
and suggest replacing l-glutamine (Sigma) by the clinically
applicable l-alanyl-l-glutamine (Dipeptiven®, Fresenius).
11. In thawed pHPL samples fibrin clots are regularly observed.
It is recommended to discard these clots as they may hamper
sterile filtration of the supplemented medium due to an occlu-
sion of the filter. Fibrin filaments or aggregates of residual
platelet fragments appearing in the medium during culture
seem not to affect cell proliferation. The problem may be
overcome by slightly increasing the heparin concentration.
12. For further technical details regarding MSPC and ECFC culture
please see also refs. 10, 11, 19 and 18–20, respectively.
Acknowledgments
This work was supported by the Austrian Research Foundation

(FWF, grant N211-NAN; DS) and the Adult Stem Cell Research
Foundation (KS). The authors thank Eva Rohde and Konrad
Rosskopf for critical review, Claudia Url and Marianne Keller for
excellent technical assistance, Tina Schreiner for graphics editing,
and Monica Farrell for editorial assistance.
References
1. Reed GL (2007) Platelets. Elsevier Science, 7. Foster TE, Puskas BL, Mandelbaum BR,
San Diego, California Gerhardt MB, Rodeo SA (2009) Platelet-rich
2. Nurden AT, Nurden P, Sanchez M, Andia I, plasma: from basic science to clinical applica-
Anitua E (2008) Platelets and wound healing. tions. Am J Sports Med 37:2259–2272
Front Biosci 13:3532–3548 8. Doucet C, Ernou I, Zhang YZ, Llense JR, Begot
3. Barrientos S, Stojadinovic O, Golinko MS, L, Holy X et al (2005) Platelet lysates promote
Brem H, Tomic-Canic M (2008) Growth fac- mesenchymal stem cell expansion: a safety sub-
tors and cytokines in wound healing. Wound stitute for animal serum in cell-based therapy
Repair Regen 16:585–601 applications. J Cell Physiol 205:228–236
4. Borzini P, Mazzucco L (2005) Tissue regen- 9. Schallmoser K, Bartmann C, Rohde E, Reinisch
eration and in loco administration of platelet A, Kashofer K, Stadelmeyer E et al (2007)
derivatives: clinical outcome, heterogeneous Human platelet lysate can replace fetal bovine
products, and heterogeneity of the effector serum for clinical-scale expansion of functional
mechanisms. Transfusion 45:1759–1767 mesenchymal stromal cells. Transfusion
5. Anitua E, Sanchez M, Orive G, Andia I (2008) 47:1436–1446
Delivering growth factors for therapeutics. 10. Schallmoser K, Rohde E, Reinisch A, Bartmann
Trends Pharmacol Sci 29:37–41 C, Thaler D, Drexler C et al (2008) Rapid
6. Martinez-Zapata MJ, Marti-Carvajal A, Sola I, large-scale expansion of functional mesenchy-
Bolibar I, Angel Exposito J, Rodriguez L et al mal stem cells from unmanipulated bone mar-
(2009) Efficacy and safety of the use of autolo- row without animal serum. Tissue Eng Part C
gous plasma rich in platelets for tissue regen- Methods 14:185–196
eration: a systematic review. Transfusion 11. Reinisch A, Bartmann C, Rohde E, Schallmoser
49:44–56 K, Bjelic-Radisic V, Lanzer G et al (2007)
Humanized system to propagate cord blood- 17. Strunk D, Schallmoser K, Rohde E (2008)
derived multipotent mesenchymal stromal Plasma-free platelet lysate for use as a supple-
cells for clinical application. Regen Med ment in cell cultures and for the preparation
2:371–382 of cell therapeutics. Patent (WO/2008/
12. Bieback K, Hecker A, Kocaomer A, Lannert 034803)
H, Schallmoser K, Strunk D et al (2009) 18. Reinisch A, Hofmann NA, Obenauf AC,
Human alternatives to fetal bovine serum for Kashofer K, Rohde E, Schallmoser K et al
the expansion of mesenchymal stromal cells (2009) Humanized large-scale expanded
from bone marrow. Stem Cells 27: endothelial colony-forming cells function
2331–2341 in vitro and in vivo. Blood 113:6716–6725
13. Rohde E, Schallmoser K, Bartmann C, Reinisch 19. Reinisch A, Strunk D (2009) Isolation and
A, Strunk D (2008) GMP-compliant propaga- animal serum free expansion of human umbili-
tion of human multipotent mesenchymal cal cord derived mesenchymal stromal cells
stromal cells. Pharmaceutical manufacturing (MSCs) and endothelial colony forming pro-
handbook: regulations and quality. Wiley, genitor cells (ECFCs). J Vis Exp 32:1525
Hoboken, New Jersey 20. Hofmann NA, Reinisch A, Strunk D (2009)
14. Weibrich G, Kleis WK, Hafner G, Hitzler WE Isolation and large scale expansion of adult
(2002) Growth factor levels in platelet-rich human endothelial colony forming progenitor
plasma and correlations with donor age, sex, cells. J Vis Exp 32:1524
and platelet count. J Craniomaxillofac Surg 21. Guide to the preparation, use and quality
30:97–102 assurance of blood components (2008).
15. Frechette JP, Martineau I, Gagnon G (2005) European Directorate for the Quality of
Platelet-rich plasmas: growth factor content Medicine & Healthcare of the Council of
and roles in wound healing. J Dent Res Europe, Strasbourg, France
84:434–439 22. Bartmann C, Rohde E, Schallmoser K, Purstner
16. Schallmoser K, Strunk D (2009) Preparation P, Lanzer G, Linkesch W et al (2007) Two
of pooled human platelet lysate (pHPL) as an steps to functional mesenchymal stromal
efficient supplement for animal serum-free cells for clinical application. Transfusion 47:
human stem cell cultures. J Vis Exp 32:1523 1426–1435
Chapter 23
In Vitro Methods to Culture Primary Human Breast

Epithelial Cells
Afshin Raouf and Yu Jia Sun
Abstract
Current evidence suggests that much like leukemia, breast tumors are maintained by a small subpopulation
of tumor cells that have stem cell properties. These cancer stem cells are envisaged to be responsible for
tumor formation and relapse. Therefore, knowledge about their nature will provide a platform to develop
therapies to eliminate these breast cancer stem cells. This concept highlights the need to understand the
mechanisms that regulate the normal functions of the breast stem cells and their immediate progeny as
alterations to these same mechanisms can cause these primitive cells to act as cancer stem cells. The study
of the primitive cell functions relies on the ability to isolate them from primary sources of breast tissue.
This chapter describes processing of discarded tissue from reduction mammoplasty samples as sources of
normal primary human breast epithelial cells and describes cell culture systems to grow single-cell suspen-
sions prepared from these reduction samples in vitro.
Key words: Reduction mammoplasty samples, Primary breast epithelial cell, In vitro cultures,
Mammospheres, 3D Matrigel cultures, Breast stem and progenitor cells
1. Introduction
The ability to support multiple pregnancies suggests that breast tis-

sue possesses an enormous regenerative capacity. This ability is due
to the unique properties and special functions of a small population
of cells in the breast tissue called stem cells. Our work, as well as
works of others, indicates that human and mouse breast epithelial
cells are organized in a lineage hierarchy where self-renewing breast
stem cells produce undifferentiated bipotential progenitors which
in turn differentiate into lineage-restricted progenitors that can
produce the differentiated luminal and myoepithelial cells that make
363
364 A. Raouf and Y.J. Sun
up the functional elements of the mammary gland (1–4). In this

context, the luminal cells in the alveolar structures (grape-like struc-
tures at the end of mammary ducts) can further differentiate into
milk producing cells upon stimulation by lactation hormones (5–7).
Because of the properties (lifelong self-renewal, proliferation, and
differentiation) uniquely possessed by the mammary epithelial stem
cells and their immediate progeny (progenitors), it has been argued
that these cells are prime cellular targets for accumulating trans-
forming mutations that can confer a breast cancer stem cell pheno-
type on these primitive cells (2, 3, 8–10). Moreover, if the high
proliferative potential of the stem and progenitor cells is dysregu-
lated, it might result in a malignant phenotype. These ideas have
greatly focused recent interest in developing an understanding of
the molecular mechanisms that control the normal functions of
mammary stem cells and progenitors in order to provide a frame-
work for developing more effective ways to diagnose and treat, or
event prevent, breast cancer since these same pathways may be
operational in the breast cancer stem cells.
The study of mechanisms that regulate the normal functions
of the breast stem and progenitors cells requires the ability to iso-
late these primitive cells at high purities from primary sources of
breast tissue. For this purpose, discarded tissue samples from
reduction mammoplasty surgeries are a good source to obtain
stem cells and progenitors. The methods in this chapter will
describe the dissociation of discarded tissue samples from breast
reduction surgeries to produce epithelial-enriched fractions that
can then be cryogenically preserved. The primitive breast epithe-
lial cells that can be obtained from the single-cell suspensions pre-
pared from these fractions can be maintained in vitro using
two-dimensional (2D) tissue culture on a plastic surface, liquid
cultures as mammospheres, and the three-dimensional (3D)
Matrigel culture systems. These culture systems have been shown
to contain breast stem cells and progenitors and therefore, pro-
vide an excellent model to study the biology and functions of
these primitive cells.
Recently, we demonstrated that distinct human breast epithe-
lial progenitors can be detected and quantified from the 2D in vitro
cultures of primary human breast epithelial cells. This primitive cell
isolation strategy utilizes fluorescent activated cell sorting (FACS)
to detect the expression of cell surface markers such as Epithelial
Cell Adhesion Molecule (EpCAM), a6 integrin (CD49f), Mucine-1
(MUC1), Prominin1 (CD133), THY-1 (CD90), and Common
Acute Lymphocytic Leukemia Antigen (CALLA1, CD10) (11).
Once isolated, these progenitor subtypes can be detected and
quantified using colony forming cell (CFC) assays. These methods
are not described in this chapter but have been described previ-
ously (2, 3, 11).
23 Methods to Culture Human Breast Cells 365
Methods to maintain mammary epithelial stem cells and pro-

genitors in vitro will facilitate in depth study of the genes that
regulate the biological function of these primitive cells. The altered
functions of these regulatory genes can contribute to phenotypic
changes that are part of the repertoire of breast cancer stem cells.
Therefore, identifying these genes can help develop new treatment
approaches that are focused on cancer stem cells, which are the
relevant cell population.
The methods discussed in this chapter describe processing of
discarded tissue from reduction mammoplasty surgeries to produce
an organoid-enriched fraction which can be turned into single-cell
suspensions and cultured in vitro on tissue culture plastic, or in
liquid cultures as mammospheres, or in 3D Matrigel cultures.
2. Materials
2.1. Dissociation of the 1. Glass petri dishes 100 mm.

Discarded Reduction 2. Dissociation flasks 250 mL.
Mammoplasty
3. Scalpel Handles.
Samples
4. Scalpel Blades #22.
5. Sterile plastic specimen containers (4½ oz or 110 mL cups).
6. Temperature controlled shaking incubator.
7. Centrifuge.
8. Cryovials.
9. Liquid Nitrogen tank.
10. Handi-freeze freezing tray from Taylor-Wharton or “Mr.
Frosty” freezing container from Nalgene (VWR).
2.2. Transport Media 1. Basic Medium which consists of Dulbecco’s Modified Eagle’s
Medium (DMED) with Ham’s F12 (1:1 ratio) and HEPES
(4-(2-hydroxyethyl)-1-piperazineethanesulfonic acid) at 10 mM
from StemCell Technologies Inc. (STI).
2. Bovine Serum (STI)—use at 5% in the basic medium.
3. Insulin Stock Solution prepared at 5 mg/mL in PBS, used at 1
in 100 (the stock solution should be stored at −20°C).
4. 50× Antibiotics and antifungal Stock solution: prepared by
adding 50% (vol/vol) solution of Penicillin (1 × 104 units/mL)
and Streptomycin (10 mg/mL) mix from STI; G418 (2.4 mg/
mL final concentration) from SIGMA; Fungizone (50 mg/mL
final concentration) from SIGMA (see Note 1). This stock
solution should be stored at −20°C.
2.3. Dissociation 1. Basic Medium.

Media 2. Bovine Serum Albumin powder (BSA, use at 20 mg/mL in
dissociation media) from SIGMA.
3. Hydrocotisone from STI (500 mg/mL) diluted in Dulbecco’s
Modified Eagle Medium (DMEM). Store the stock solution at
−20°C and use at a 1 in 1,000 dilution.
4. Collagenase/Hyaluronidase at 100× Stock (kept at −70°C)
from STI (see Note 2).
5. Insulin Stock Solution prepared at 5 mg/mL in PBS. Use at a
1 in 100 dilution.
6. Solution of Penicillin (1 × 104 units/mL) and Streptomycin
(10 mg/mL) from STI (use at 1:100 dilution) (see Note 1).
2.4. Preparation of 1. Solution of trypsin (0.25%) and Ethylenediamine Tetraacetic

Single Cell Suspension Acid (EDTA, 1 mM) from STI.
2. Dispase solution (5 mg/mL) from STI.
3. DNase stock solution at 1 mg/mL in Phosphate Buffer Saline
(PBS). Store the stock solution in −20°C and use at 1:10
dilution.
4. 40 mm cell strainer from BD Falcon.
5. Hank’s balanced salt solution (HBSS) with 2% Fetal Calf Serum
from STI (2% HBSS).
2.5. 2D Cultures 1. 10 cm tissue culture plates.

of Primary Human 2. Irradiated mouse embryonic fibroblasts (see Note 3).
Mammary Epithelial
3. EpiCult-B growth medium from STI.
Cells
4. Hydrocortisone from SIGMA (500 mg/mL) diluted in
Dulbecco’s Modified Eagle Medium. Store the stock solution
at −20°C and use at 1 in 1,000 dilution.
5. Fetal Calf Serum from STI (5% of volume).
2.6. Mammosphere 1. Ultra-low adhesion 96-well Tissue culture plates from STI.
Cultures of Human 2. MammoCult basal medium from STI.
Mammary Epithelial
3. MammoCult Proliferation Supplements from STI.
Cells
4. Hydrocortisone from STI (500 mg/mL) diluted in Dulbecco’s
Modified Eagle Medium. Store the stock solution in −20°C
and use at 1 in 1,000 dilution.
5. Heparin Sodium Salt solution (0.2% Vol/Vol) from STI.
2.7. 3D Matrigel 1. 24-well tissue culture plates from BD.

Cultures of Human 2. Matrigel from BD Biosciences (with growth factors, Catalog
Mammary Epithelial Number 354234).
Cells
3. Matrigel Basic medium (for this formulation use DMEM

without phenol red indicator and Ham’s F-12 at a 1:1 ratio)
supplemented with 2.6 ng/mL sodium selenite, 100 ng/mL
epidermal growth factor (EGF), 0.5 mg/mL hydrocortisone,
l0 nM triiodothyronine, 100 ng/mL fibronectin, 2 mM glu-
tamine, 25 ng/mL transferrin, 10 nM dibutyryl cyclic AMP
(also known as N6,2¢-O-Dibutyryladenosine 3¢,5¢-cyclic mono-
phosphate sodium salt), 0.1 mM phosphoethanolamine (also
known as 2-Aminoethyl dihydrogen phosphate), 20 ng/mL
fetuin (from calf serum), 10 mg/mL freshly prepared ascorbic
acid, 0.01% bovine serum albumin (fraction V), 1 mg/mL
insulin, 0.1 nM estradiol, and 0.1 mM ethanolamine. All com-
ponents should be cell culture grade and can be acquired from
SIGMA Aldrich. This medium is also referred to as modified
CDM3 media (12, 13) (see Note 4).
3. Methods
3.1. Dissociation Discarded tissues from reduction mammoplasty surgeries represent

of Discarded Tissue a good source of normal nonmalignant primary human breast epi-
from Reduction thelial cells. These precious tissue samples can be obtained through
Mammoplasty informed patient consent and close collaboration with plastic sur-
Surgeries geons and the Pathology Department of the corresponding hospi-
tal where the surgeries take place.
The procedure described in this section will yield organoid-
enriched pellets that are a great source of human breast epithelial
stem and progenitor cells. As well, this procedure will yield a cell
pellet containing both breast epithelial cells and human breast
fibroblasts and a pellet that is enriched for the human breast
fibroblasts.
1. Discarded tissue from breast reduction surgeries are best trans-
ported from the operating room or Pathology cutting room in
sterile specimen cups.
Prepare 200 mL of transport media on the day of the surgery
and place 100 mL per sterile plastic cup with lid (see Note 1
and Fig. 1).
2. Prepare 40 mL of dissociation media for two samples (left and
right reduction samples).
3. Add 0.8 g of BSA to 30 mL of basic medium and filter-sterilize
(0.22 mm syringe filter) into a 50 mL sterile falcon tube. Then
add 4 mL of 10× collagenase–hyaluronidase enzyme mix and
add 40 mL of insulin. Invert the tube several times to mix (do
not vortex). Bring the volume to 40 mL using sterile DMEM
and Hank’s F12 (1:1 mixture) medium.
Fig. 1. Discarded breast reduction samples in transport medium. The breast reduction samples should be removed from a
glandular section of the breast tissue and immediately placed in the transport medium. The sample can be stored in trans-
port media for up to 48 h with minimal loss of viability. Samples that are dense and less fatty may sink to the bottom of the
transport cup.
4. To mince the tissue samples, work in a Biological Safety Cabinet

(BSC). Place the tissue in a glass petri dish and add 3–5 mL of
transport media to keep the tissue moist.
5. Trim off big chunks of fat and skin using the #22 scalpel blades.
It is not necessary to remove all the fat tissue, as it will dissolve
during the dissociation process. It is also not necessary to
mince the tissue completely as this may result in poor organoid
recovery due to over-digestion (Fig. 2).
6. Warm up the dissociation media to 37°C using a water bath
and turn on the shaking incubator to warm up the inside tem-
perature to 37°C.
7. Transfer the minced breast tissue to sterile dissociation flasks
containing warm dissociation media. Cap the flasks with sterile
aluminum foil and seal with Parafilm. Shake the dissociation
flasks overnight (16–18 h) at 37°C at a low setting (Fig. 3a and
see Note 5).
8. The following morning, warm approximately 55 mL of basic
medium to 37°C using a water bath.
9. Remove dissociation flasks from the shaker and examine each
sample. All of the minced tissue should be dissociated at this
Fig. 2. (a) Minced breast tissue in 10 cm glass petri dish. As shown, there is no need to completely mince the tissue and
bigger chunks will be dissociated. (b) Minced tissue in dissociation media, prior to overnight dissociation in a shaking
incubator at 37°C.
Fig. 3. Dissociated breast tissue. (a) Dissociation flasks are shown after 17 h at 37°C. Small organoid structures should be
visible at this point. The loss of organoid structures may indicate over-digestion. Note that fat tissue has melted, and floats
on the top layer, indicated by arrows. Note that colors vary depending on the content of blood cells. (b) The dissociated
breast sample has been transferred to a 50 mL sterile conical tube. Note that there is a 5–7 mL of melted fat floating on
the top of the dissociated sample.
point (if not, continue shaking for an additional 1 h). Organoids

released from the tissue should, however, be visible at this point.
Fibrous materials do not dissociate using this technique.
10. Transfer the suspension from the dissociation flasks to sterile
50 mL conical tubes. Wash the walls of the dissociation flasks with
7 mL of warm basic medium (Fig. 3b). Use a pipette to remove
and discard the fibrotic (long white strings of fiber) tissue.
Fig. 4. Organoid-enriched pellet. A loose, organoid-enriched pellet formed after centrifugation of the dissociated breast
sample at 80 × g for 40 s (indicated by the arrow).
11. Spin the 50 mL conical tubes at 75–80 × g for 40 s. At this low

speed an organoid-enriched pellet will form (Fig. 4a).
12. Discard the floating fat layer and pipette the supernatant into a
sterile 50 mL conical tube labeled epithelial cell pellet.
13. Combine the organoid-enriched pellets (from the left and right
breast reduction samples) and add 10 mL of warm basic
medium and spin at 75–80 × g for 40 s.
14. Add the supernatant to the 50 mL conical tubes labeled “epi-
thelial cell pellet” from step 12. Wash the organoid-enriched
pellet one more time with 5 mL of warm basic medium and
spin at 75–80 × g for 40 s.
15. Remove as much of the supernatant as possible. At this point
the organoid-enriched pellet can be frozen for long-term pres-
ervation (explained in step 24) or can be placed on ice for a
short period of time (no more than 30 min).
16. Spin the falcon tube labeled epithelial cell pellet at 200 × g for
4 min. This spin will produce a pellet that is enriched in dis-
sociated breast epithelial cells as well as fibroblasts (Fig. 5).
Notice that this cell pellet may contain some blood and there-
fore would have a red color.
17. Remove and discard the floating fat layer and transfer the
supernatant to a fresh 50 mL conical tube labeled mammary
fibroblasts.
Fig. 5. Epithelial cell pellet. A firm epithelial pellet is created by centrifuging (200 × g for 4 min) the supernatant extracted
from Organoid-enriched pellet. Notice that this pellet contains red blood cells (red in appearance). These cells will be elimi-
nated upon freeze-thawing of the samples.
18. Wash the pellet with 10 mL of warm basic media and spin
again at 200 × g for 4 min.
19. Add the supernatant to the 50 mL conical tube labeled mam-
mary fibroblasts (from step 17) and wash the epithelial cell
pellet in 5 mL of basic medium and spin one last time at 200 × g
for 4 min.
20. Remove as much of the supernatant as possible and place the
pellet on ice. The cell pellet can be stored on ice for up to
30 min or should be frozen at once using the method described
in step 24.
21. Spin the 50 mL conical tube labeled mammary fibroblasts at
800 × g for 5 min (Fig. 6).
22. Remove and discard the floating fat layer as well as the super-
natant. Wash the pellet with 10 mL warm basic medium and
spin again at 800 × g for 5 min.
23. Discard the supernatant and wash the pellet with 5 mL of warm
basic medium and spin at 800 × g for 5 min. Discard the
supernatant.
24. Freeze all three cell pellets in freshly prepared freezing media
consisting of 6% DMSO, 44% fetal calf serum, and 50% basic
medium. Using a 5 mL pipette resuspend the organoids in the
freezing medium and aliquot into cryovials (1 mL per vial).
Fig. 6. Mammary fibroblast pellet. The mammary fibroblast pellet is created by centrifuging (800 × g for 5 min) the super-
natant from epithelial cell pellet. This pellet may also appear red due to the presence of some red blood cells. These cells
will be eliminated upon freeze-thawing of the samples.
Use 1 mL pipettes to resuspend the epithelial and fibroblast

cell pellets. The volume of the freezing medium will depend
on the size of the organoid or cell pellets. Typically a large
reduction sample that filled 2/3 of a 10 cm plate will produce
two to three organoid vials, two epithelial vials and one to two
fibroblast vials. Freeze the samples right away using the Handi-
freeze freezing tray placed on top of a liquid Nitrogen tank
(use the manufacturer’s protocol and see Note 6). Alternatively,
the cryovials can be placed in “Mr. Frosty” filled with Isopropyl
Alcohol and placed in a freezer cooled to −80°C for 2–4 h.
Subsequently, the cryovials can be stored in liquid Nitrogen
for long-term preservation.
3.2. Single-Cell The organoid-enriched pellets, once turned into single-cell suspen-
Suspension sions, are a great source of primary breast epithelial cells that contain
Preparation breast stem and progenitor cells. To initiate in vitro cultures from
these pellets and to study the differentiation potential of the primi-
tive breast epithelial cells, the organoid suspensions need to be
turned into single-cell suspension. The epithelial cell pellets that are
collected second to the organoid-enriched pellets also contain a fair
number of breast epithelial cells, but they will also contain an appre-
ciable number of fibroblasts and endothelial cells and should be used
with that caveat in mind.
The procedures described in this section will yield single-cell
preparations from organoid-enriched pellets of human breast epi-
thelial cells which can then be used for culturing in vitro or can be
used directly to isolate human breast stem and progenitors-enriched

cell subpopulations using FACS (11).
1. Thaw Trypsin/EDTA, Dispase, and DNase stock solutions in
a 37°C water bath. Take care not to leave these stock solutions
at 37°C for more than 10 min as it will decrease their enzy-
matic activity. Once they are thawed and warmed to above
room temperature remove vials from the water bath and place
inside the BSC unit.
2. Thaw organoid-enriched pellets as needed in a 37°C water
bath (see Note 7). Just as the last trace of ice dissolves transfer
the vials to a BSC unit.
3. Use a 2 mL pipette to transfer the organoid sample into a
50 mL conical tube, add 10 mL of cold 2% HBSS and spin at
200 × g for 5 min.
4. Discard the supernatant and add warm Trypsin/EDTA to the
organoids. Use a 1,000 mL pipette to suspend the organoids
and place in a 37°C water bath for 5 min.
5. Use the mechanical force of repetitive pipetting (15–20 times)
to break the organoid further. Use a 1,000 mL pipette for this
purpose and do not brace the pipette against the bottom of the
tube. At this point the suspension will not be smooth. Add
10 mL of cold 2% HBSS and spin at 200 × g for 5 min.
6. Discard the supernatant and add 1 mL of Dispase–DNase mix-
ture (100 mL DNase/mL of Dispase) for a small pellet or 2 mL
for larger pellets, and mix well by pipetting.
7. Incubate at 37°C for 5 min and use repetitive pipetting as in
step 5 to completely break up the organoids. If some solid bits
remain, place the sample at 37°C for an additional 2 min and
use repetitive pipetting to break up the solid bits (at least 15
times). Any solid bits remaining at this point are fibrotic tissue
and will not dissociate further.
8. Place a 40 mm cell strainer on top of a 50 mL conical tube. Add
5 mL of cold 2% HBSS to the dissociated organoids and pass
through a cell strainer. Wash the strainer with 5 mL of cold 2%
HBSS.
9. Spin the cell suspension at 200 × g for 5 min and discard the
supernatant. Suspend the cell pellet in 1 mL of cold 2% HBSS.
Observe a small aliquot of the cell suspension under a micro-
scope. If more than 10% of cells appear to be in aggregates
repeat the Dispase–DNase treatment and pass through the cell
strainer (steps 6–8).
10. Obtain a cell count if needed.
11. Keep the single-cell suspension on ice until used. Cold tem-
peratures will decrease aggregation of the epithelial cells.
3.3. Culturing Human Human breast epithelial cells can be cultured on the 2D surface of
Breast Epithelial Cells tissue culture plastic dishes but most samples from reduction mam-
in Tissue Culture moplasty samples can only survive 4–5 passages under these cul-
Plastics ture conditions before they exhibit senescence.
However, the 2D cultures can be used for the short-term
in vitro culturing of human breast epithelial cells and therefore, are
very useful in eliminating hematopoietic contaminants from the
dissociated organoid preps. Culturing the breast epithelial cells
under these conditions will produce cuboidal (luminal) and tear-
drop (myoepithelial) shaped cells (Fig. 7a). These 2D cultures are
in fact used to isolate subpopulations of human breast epithelial
cells that are highly enriched for distinct mammary progenitor sub-
types (11).
1. Start with preparing a single-cell suspension from organoid-
enriched pellets as described in Subheading 3.2.
2. Thaw out one vial of EpiCult-B cytokine mix per 100 mL of
basic medium. Add the content of one cytokine vial to 100 mL
of basic medium and then add hydrocortisone (final concen-
tration 0.5 mg/mL, see Note 8). This solution is called
EpiCult-B growth medium.
3. Prepare 10 mL of 5% Fetal Calf Serum (FCS) supplemented
EpiCult-B growth medium.
4. Culture the dissociated organoids at a density of at least 62,000
cells/cm2. This density yields 3.5 × 106 epithelial cells per 10 cm
tissue culture plates. Such high density is required since the
growth of breast epithelial cells is cell-density dependent.
5. Incubate plates in a humidified incubator with 5% CO2 at
37°C. Similar to other cell types, primary human breast epithe-
lial cells should not be cultured to confluence. Therefore once
the culture is 80% confluent, it should be passaged using stan-
dard trypsin/EDTA protocols.
(a) Briefly, wash the cells with warm phosphate buffer saline
(PBS) by adding enough PBS to cover the cells and rock-
ing the cell culture plate several times to dilute out any
residual serum; remove the PBS.
(b) Add enough warm trypsin/EDTA (37°C) to cover the cells
and place in a humidified incubator at 37°C for 5 min.
(c) Remove the tissue culture plate from the incubator and
add an equal volume of 2% Hank’s solution to deactivate
the Trypsin/EDTA
(d) Use a 2 mL pipette to gently dislodge any cells that are still
adherent to the tissue culture plate/flask. It is often useful
to check under the microscope to ensure all cells have lifted
off the plate, as breast epithelial cells can be sticky.
Fig. 7. Culturing primary human breast epithelial cells in vitro. Single-cell suspensions were prepared from organoid-en-
riched pellets that were isolated from discarded mammoplasty tissue samples. The cell suspensions were then cultured
in vitro using three different cell culture systems. When unseparated human breast epithelial (hbe) cells are cultured on a
tissue culture plastic surface (a) cuboidal (luminal) and tear-drop (myoepithelial) shaped cells can be observed. The picture
was taken after 7 days at 160× magnification. When hbe cells are cultured in non-adherent liquid cultures as mammo-
spheres (b), two different types of spheres can be observed, namely, solid and hollow spheres. The picture was taken on
day 7 at 50× magnification. Culturing hbe cells in the three-dimensional (3D) matrigel cultures (c) will allow the formation
of polarized bilayer epithelium that can be organized in rudimentary alveolar structures. The picture was taken after
14 days at 25× magnifications.
(e) Place the cell suspension in an appropriately sized sterile

tube and spin down the cells in a centrifuge at 700–
1,000 × g for 5 min
(f) Remove the supernatant and suspend the cells in an appro-
priate volume of 2% Hank’s
6. Low density cultures are possible; however, in this case the
breast epithelial cells should be supplemented with irradiated
mouse fibroblasts. For this purpose: Culture embryonic mouse
3T3 fibroblast cells in DMEM growth medium supplemented
with 5% FCS. Maintain these cultures at 80–85% confluency as
3T3 cells are prone to contact inhibition. Based on our experi-
ence, maintaining 3T3 cells at higher densities changes their
properties with respect to their influence on human breast epi-
thelial progenitor cell differentiation. X-irradiate (at 50 Gy) to
produce proliferation-incompetent 3T3 cells. Mix single-cell
suspensions prepared from organoid-enriched fractions with
400,000 irradiated 3T3 cells. Then add 10 mL of warm complete
EpiCult-B medium with 5% FCS.
3.4. Mammosphere It has been shown previously that normal and malignant human
Cultures breast epithelial cells can be cultured in vitro in a non-adherent
liquid culture system. Under this culturing condition unseparated
human breast epithelial cells formed spheres and were thus
referred to as mammospheres. These mammospheres as well as
their subsequent passages (up to three times) were shown to con-
tain breast stem cells and progenitors (14, 15). Therefore, mam-
mosphere cultures can be used to study the biology of the
primitive human breast epithelial cells in vitro for at least 21 days.

In our hands, we observe two distinct spheres, namely solid and
hollow spheres (Fig. 7b).
1. Begin by preparing single-cell suspensions from organoid-en-
riched samples and perform cell counts. While performing cell
counts make note of the cell aggregates (two or more cells). If
the cell aggregates make up more than 10% of the total cells,
repeat the single-cell suspension protocol (steps 6–11).
2. Prepare MammoCult medium by adding 50 mL of MammoCult
Proliferation Supplements to 450 mL of MammoCult Basal
Medium. Supplement an aliquot of the MammoCult medium
with 4 mg/mL of Heparin and 4.8 mg/mL of Hydrocortisone
(final concentration of 1 mM).
3. From single-cells suspensions prepared from organoid-enriched
samples, plate cells at a density of between 4 × 103 cells/cm2
and 10 × 105 cells/cm2 using Mammocult Medium into 6-well
ultralow adherent plates in triplicates. The higher densities are
recommended since some human breast reduction samples
form mammospheres at very low efficiency. The optimum den-
sity should be determined for each sample before setting up
large-scale experiments.
4. Incubate plates in a humidified incubator with 5% CO2 at 37°C
for 7 days.
5. If desired, count the number of spheres that have formed (typ-
ically 60 mm or larger). Generally mammosphere cultures con-
sist of hollow and solid spheres. Mammosphere formation
efficiency has been reported to be about 0.5% when using
unseparated normal human breast epithelial cells.
6. To create secondary mammosphere cultures, collect the spheres
in a conical tube and centrifuge at 350 × g for 5 min. At this
speed only spheres would form a pellet.
7. From the mammosphere pellets prepare a single-cell suspen-
sion as described in Subheading 3.2 and determine cell num-
bers. The cell numbers can be used as an index of cell
proliferation or cell loss in the mammosphere cultures.
8. Culture single cells dissociated from the primary mammo-
spheres in MammoCult media at densities between 4 × 103
cells/cm2 and 10 × 104 cells/cm2 to create secondary spheres.
The efficiency at which secondary mammospheres form is vari-
able and therefore, it is advisable to use multiple seeding densi-
ties (see Note 9).
3.5. Three- The culture systems described thus far allow the maintenance of
Dimensional Matrigel human breast epithelial cells in vitro. However, they cannot provide
Cell Cultures the environment that is needed to produce fully differentiated lumi-
nal cells capable of milk production. To this end, culturing human
breast epithelial cells in the laminin-enriched environment of

Matrigel has proven to be very useful. Matrigel is reconstituted
basement membrane that is extracted from a mouse sarcoma tumor
(Engelbreth-Holm-Swarm (EHS)) (16–18). It has been previously
shown that placing human breast organoid or breast epithelial cells
grown as 2D cultures or mammospheres in 3D Matrigel cultures
can lead to the formation of rudimentary breast structures (19, 20).
These structures were shown to consist of bilayer, polarized epithe-
lium with hollow centers (Fig. 7c). More importantly, the luminal
epithelial cells in these cultures further differentiate into milk pro-
ducing cells in the presence of lactogenic hormones (21–23).
The protocol outlined in this section describes Matrigel culture
conditions that would lead to formation of rudimentary mammary
structures from primary uncultured human breast organoids, as well
as 2D or mammosphere cultures of human breast epithelial cells.
3.5.1. Setting Up Matrigel 1. Thaw out Matrigel on ice at 4°C overnight (see Note 10)
Cultures Using Human 2. Coat the bottom of 8-well chamber slides with 50 mL of thawed
Organoids Matrigel to prevent contact of epithelial cells with the plastic
surface.
3. Mix the organoid-enriched fraction (see Note 11) with
1,600 mL of thawed Matrigel and pipette 200 mL per chamber
of the Matrigel-coated 8-well chamber slide (see Note 12) and
incubate in a humidified incubator at 37°C for 1 h.
4. Once the gels are polymerized, add up to 400 mL of Matrigel
growth medium and incubate at 37°C with 5% CO2, up to
14 days with media changes once every 3 days.
5. Rudimentary breast structures will be distinguishable after
7 days.
3.5.2. Setting Up Matrigel 1. Prepare single-cell suspensions from organoid-enriched frac-

Cultures Using Human tions and place in 2D cultures as described in Subheading 3.2.
Breast Cells Grown in 2D 2. After 2 days, prepare a single-cell suspension (as described in
Cultures Subheading 3.2) from the cells grown in 2D cultures. For this
purpose start by removing the growth medium and wash the
cells with PBS. Subsequently add 2 mL of warm Trypsin (at
37°C) per one 10 cm tissue culture plate and place the plates
back in the incubator for 5 min. Deactivate the trypsin with 2%
Hank’s solution and follow the single cells suspension protocol
(Subheading 3.2) from steps 6 to 11.
3. Coat the bottom of 24-well plates with 50 mL of thawed
Matrigel and place the plate at 37°C to polymerize for 30 min.
4. Mix 2.5 × 105 cells per 300 mL of Matrigel for each well of the
Matrigel-coated 24-well plate and place plates at 37°C for 1 h
to allow the gels to polymerize. Add 400 mL of Matrigel growth
medium to each well and culture up to 14 days with media
changes once every 3 days.
5. After 10–14 days rudimentary breast structures should be visible.

3.5.3. Setting Up Matrigel
1. Prepare a single-cell suspension from the organoid-enriched
Cultures Using
fraction and plate them in mammosphere culture conditions as
Mammospheres
described in Subheading 3.3.
2. After 7 days thaw out aliquots of Matrigel on ice overnight and
coat the bottom of 8-well chamber slides with 50 mL of thawed
Matrigel and polymerize by incubating at 37°C for 30 min (see
Note 12).
3. Culture mammospheres in Matrigel by gently mixing 100–900
mammospheres with 300 mL of Matrigel per well of the
Matrigel-coated 24-well plate (see Note 13).
4. Incubate the plates at 37°C for 1 h to polymerize the gels.
5. Add 400 mL of Matrigel growth medium to each well and
incubate in a humidified incubator at 37°C and 5% CO2 with
change of medium every 3 days.
6. After 10–14 days rudimentary breast structures should be
visible.
4. Notes
1. The cocktail of antibiotics and antifungal agents are only nec-

essary if the reduction mammoplasty samples are obtained in
non-sterile conditions (i.e., outside of the operating room
where the breast reduction surgery took place). It should be
noted that the use of the antibiotics and antifungal agents are
only recommended if frequent sample contamination is
observed. Also the antibiotics and antifungal agent mix can be
used at lower (up 50% less) strength to increase cell viability.
2. Collagenase–hyaluronidase enzyme stock solution preparation:
Measure 3,000 units/mL of collagenase and 1,000 units/mL
of hyaluronidase and dissolve in DMEM. Filter-sterilize the
stock solution by passing the solution through 0.8 mm, then
0.45 mm, and finally through 0.2 mm syringe filters. Store the
stock solution in 4 mL aliquots at −20°C.
3. Embryonic 3T3 fibroblasts can be purchased from ATCC and
should be cultured in DMEM growth medium supplemented
with 5% fetal calf serum. The 3T3 fibroblast cultures should
never surpass 85% confluence as this will diminish their ability
to support breast epithelial cell growth and differentiation.
The 3T3 cells can be irradiated at 50 Gy in large numbers and
kept as frozen aliquots in liquid nitrogen for future use.
However, numbers of live cell should be determined before

use using trypan blue exclusion and a hemocytometer.
4. It should be mentioned while CDM3 has widely been used as
a defined media to culture primary human breast cells in
Matrigel, other growth medium have also been used (for
example see ref. 24). In our laboratory, we routinely use
Epicult-B medium (from StemCell Technologies) supple-
mented with 5% Fetal Bovine Serum.
5. Over-digestion of the reduction mammoplasty samples could
lead to poor organoid recovery. To avoid over-digestion of the
samples, the concentrations of the collagenase and hyaluroni-
dase enzymes should be carefully measured and the SIGMA
Aldrich Company should be consulted to identify batches of
enzymes with the lowest trypsin activity. As well, it should be
noted that the concentration of the enzymes vary from batch
to batch. The enzyme stock solution concentrations are calcu-
lated based on the actual concentration of each enzyme as indi-
cated by the manufacturer. The shaking incubator speed setting
should be optimized as it would vary between different instru-
ments. It is recommended to start with a very low/gentle set-
ting as vigorous shaking can lead to over-digestion.
6. The concept behind using the Handi-freeze freezing tray is to
minimize the cellular shock due to sudden temperature
changes. As such the Handi-freeze apparatus, which sits on top
of a liquid nitrogen tank, allows gradual lowering of the sam-
ples through the vapor phase of the liquid nitrogen. Caution
should be taken to use cryogenic vials appropriate to the liquid
nitrogen storage unit that will be used (i.e., storage in liquid
phase vs. vapor phase). As an alternative to the Handi-freeze
apparatus, “Mr. Frosty” containers can be utilized. “Mr. Frosty,
VWR” freezing containers use Isopropyl Alcohol that allows
for a gradual decline in temperature and therefore will enhance
cell-survival during the freezing process.
7. When thawing cryogenic vials from liquid nitrogen storage
tanks extreme caution should be exercised as these vials are
prone to explode when warmed suddenly. To avoid injury each
cryovial should be wrapped in Parafilm before placing in a
37°C water bath. This will prevent injuries due to projectiles in
the unlikely event that the cryovials should explode.
8. Hydrocortisone is unstable in solution at 4°C. Therefore,
complete EpiCult-B should be made in small batches that can
be used up in less than 2 weeks.
9. A small portion of the cells obtained from the primary mam-
moshpere cultures can be plated in CFC assays to detect the
presence of progenitor cells in these cultures (for protocols
consult STI EpiCult-B product literature).
10. Matrigel is a difficult material to work with. Overnight thawing

of Matrigel aliquots at 4°C on ice will ensure homogenous
thawing. Aliquots of Matrigel can also be thawed in a 4°C
fridge for 1–2 h (with no ice). However, extreme care must be
taken to ensure homogenous thawing of the Matrigel by thor-
ough mixing of the Matrigel sample without introducing bub-
bles. This should be done by keeping the Matrigel-containing
vial on ice and gently pipetting it up and down several times.
Once the Matrigel has been liquefied, it must be kept on ice at
all times to prevent untimely polymerization.
11. Organoids cannot be counted and therefore there is no robust
way of adding similar numbers of organoids to the Matrigel
cultures. However, it should be noted that overcrowding of the
Matrigels may not yield distinguishable rudimentary mammary
structures. Therefore, experiments involving organoids should
be optimized based on different organoid preparations.
12. While 8-well chamber slides can be used for growing rudimen-
tary structures in Matrigel, it can prove to be difficult and may
require some optimization. Alternatively, 96-well plates can be
used by scaling down the number of organoids and volume of
Matrigel used. Typically 50 mL of Matrigel per each well of a
96-well plate will be sufficient. The number of organoids would
still have to be optimized for each experiment.
13. There is a great deal of variation in the efficiency at which rudi-
mentary structures grow from mammospheres (it can be as high
as 50%). Since overcrowding of the gels can suppress the forma-
tion of rudimentary structures, the number of mammospheres
that are cultured in Matrigel per well should be optimized. It is
recommended that at the beginning different numbers of mam-
mospheres ranging from 100 to 900 be tested in triplicate.
References
1. Kordon EC, Smith GH (1998) An entire func- 5. Howlin J, McBryan J, Martin F (2006) Pubertal
tional mammary gland may comprise the prog- mammary gland development: insights from
eny from a single cell. Development mouse models. J Mammary Gland Biol
125:1921–1930 Neoplasia 11:283–297
2. Stingl J, Eaves CJ, Kuusk U, Emerman JT 6. Rosen JM (2003) Hormone receptor pattern-
(1998) Phenotypic and functional character- ing plays a critical role in normal lobuloalveolar
ization in vitro of a multipotent epithelial cell development and breast cancer progression.
present in the normal adult human breast. Breast Dis 18:3–9
Differentiation 63:201–213 7. Russo J, Russo IH (2004) Development of the
3. Stingl J, Eaves CJ, Zandieh I, Emerman JT human breast. Maturitas 49:2–15
(2001) Characterization of bipotent mammary 8. Shipitsin M, Campbell LL, Argani P,
epithelial progenitor cells in normal adult Weremowicz S, Bloushtain-Qimron N, Yao J,
human breast tissue. Breast Cancer Res Treat Nikolskaya T, Serebryiskaya T, Beroukhim R,
67:93–109 Hu M, Halushka MK, Sukumar S, Parker LM,
4. Wagner KU, Smith GH (2005) Pregnancy and Anderson KS, Harris LN, Garber JE,
stem cell behavior. J Mammary Gland Biol Richardson AL, Schnitt SJ, Nikolsky Y, Gelman
Neoplasia 10:25–36 RS, Polyak K (2007) Molecular definition of
breast tumor heterogeneity. Cancer Cell nin, and heparan sulfate proteoglycan from the
11:259–273 EHS sarcoma. Biochemistry 21:6188–6193
9. Smith GH (2002) Mammary cancer and epi- 18. Vukicevic S, Kleinman HK, Luyten FP, Roberts
thelial stem cells: a problem or a solution? AB, Roche NS, Reddi AH (1992) Identification
Breast Cancer Res 4:47–50 of multiple active growth factors in basement
10. Stingl J, Raouf A, Emerman JT, Eaves CJ membrane Matrigel suggests caution in inter-
(2005) Epithelial progenitors in the normal pretation of cellular activity related to extracel-
human mammary gland. J Mammary Gland lular matrix components. Exp Cell Res
Biol Neoplasia 10:49–59 202:1–8
11. Raouf A, Zhao Y, To K, Stingl J, Delaney A, 19. Ip MM, Darcy KM (1996) Three-dimensional
Barbara M, Iscove N, Jones S, McKinney S, mammary primary culture model systems. J
Emerman J, Aparicio S, Marra M, Eaves CJ Mammary Gland Biol Neoplasia 1:91–110
(2008) Transcriptome analysis of the normal 20. Weaver VM, Bissell MJ (1999) Functional cul-
human mammary cell commitment and dif- ture models to study mechanisms governing
ferentiation process. Cell Stem Cell 3: apoptosis in normal and malignant mammary
109–118 epithelial cells. J Mammary Gland Biol
12. Petersen OW, van Deurs B (1987) Preservation Neoplasia 4:193–201
of defined phenotypic traits in short-term cul- 21. Bissell MJ, Ram TG (1989) Regulation of func-
tured human breast carcinoma derived epithe- tional cytodifferentiation and histogenesis in
lial cells. Cancer Res 47:856–866 mammary epithelial cells: role of the extracellular
13. Spancake KM, Anderson CB, Weaver VM, matrix. Environ Health Perspect 80:61–70
Matsunami N, Bissell MJ, White RL (1999) 22. Gomm JJ, Coope RC, Browne PJ, Coombes
E7-transduced human breast epithelial cells RC (1997) Separated human breast epithelial
show partial differentiation in three-dimen- and myoepithelial cells have different growth
sional culture. Cancer Res 59:6042–6045 factor requirements in vitro but can reconsti-
14. Dontu G, Abdallah WM, Foley JM, Jackson tute normal breast lobuloalveolar structure. J
KW, Clarke MF, Kawamura MJ, Wicha MS Cell Physiol 171:11–19
(2003) In vitro propagation and transcriptional 23. Petersen OW, Ronnov-Jessen L, Howlett AR,
profiling of human mammary stem/progenitor Bissell MJ (1992) Interaction with basement
cells. Genes Dev 17:1253–1270 membrane serves to rapidly distinguish growth
15. Dontu G, Jackson KW, McNicholas E, and differentiation pattern of normal and
Kawamura MJ, Abdallah WM, Wicha MS malignant human breast epithelial cells. Proc
(2004) Role of Notch signaling in cell-fate Natl Acad Sci U S A 89:9064–9068
determination of human mammary stem/pro- 24. Lim E, Vaillant F, Wu D, Forrest NC, Pal B,
genitor cells. Breast Cancer Res 6:R605–R615 Hart AH, Asselin-Labat ML, Gyorki DE, Ward
16. Kleinman HK, McGarvey ML, Hassell JR, Star T, Partanen A, Feleppa F, Huschtscha LI,
VL, Cannon FB, Laurie GW, Martin GR Thorne HJ, Fox SB, Yan M, French JD, Brown
(1986) Basement membrane complexes with MA, Smyth GK, Visvader JE, Lindeman GJ
biological activity. Biochemistry 25:312–318 (2009) Aberrant luminal progenitors as the
17. Kleinman HK, McGarvey ML, Liotta LA, Robey candidate target population for basal tumor
PG, Tryggvason K, Martin GR (1982) Isolation development in BRCA1 mutation carriers. Nat
and characterization of type IV procollagen, lami- Med 15:907–913
Chapter 24
Human Prostate Epithelial Cell Cultures

Johng S. Rhim
Abstract
Prostate cancer is the most common male cancer in the United States. Research on the mechanisms of
prostate cancer progression has been limited by the lack of suitable in vitro systems. A hurdle in under-
standing the molecular genetic changes in prostate cancer has been the difficulty in establishing premalig-
nant lesions and primary prostate tumors as in vitro cell cultures. Primary prostate epithelial cells grow for
a finite life span and then senesce. Immortalization is defined by continuous growth of otherwise senescing
cells and is believed to represent an early stage in tumor progression. To examine these early stages, we and
others have developed in vitro models of prostate epithelial cell immortalization. Methods are described
for the processing of primary human prostate biopsy samples and the generation of human prostate epi-
thelial (HPE) cells in serum-free conditions. Retrovirus containing human telomerase reverse transcriptase
(hTERT) is used for the immortalization of primary HPE cells, and the methods for the characterization
of HPE cell lines are discussed. These in vitro prostate cell culture models are useful for the study of pros-
tate normal and cancer stem cells, are critical for defining the mechanisms of prostate cancer progression
and for testing preventive and therapeutic regimens.
Key words: Primary human prostate epithelial cells, Immortalized human prostate epithelial cell
lines, Keratinocyte serum-free medium, Prostate normal and cancer stem cells
1. Introduction
Prostate cancer is the most common male cancer in the Western

World and second leading cause of male cancer death in the US (1).
The therapy most widely used against advanced disease is androgen
ablation and, initially, it almost always produces objective clinical
responses. However, most patients eventually relapse with ablation-
resistant prostate cancer and develop metastatic disease; currently,
there is no treatment that will cure progressive hormone-refractory
metastatic prostate cancer. The mechanisms of progression of pros-
383
384 J.S. Rhim
tate cancer have been extensively studied, yet are poorly understood.
One of the concepts that have been evolved is that cancer arises from
the neoplastic transformation of normal prostate epithelial stem cells
or transit amplifying cells. Understanding normal stem cells and can-
cer stem cells (CSCs) may provide insight into the origin of and new
therapeutics for prostate cancer. However, research in this field is
limited by the lack of suitable in vitro systems.
Studies of prostate cancer have been hampered by various fac-
tors including (a) restricted access to tissues, (b) slow in vivo
growth, (c) difficulties in propagating tumor cells as well as normal
cells in vitro, and (d) limited availability of prostate cancer cell lines
or immortalized prostate epithelial cell lines for in vitro studies. In
January 2000, we created a Prostate Cancer Cell Center in the
newly established Center for Prostate Disease Research (CPDR)
laboratory in the Department of Surgery, Uniformed Services
University of the Health Sciences (USUHS). This center has suc-
cessfully generated more than 100 primary prostate epithelial cells
from primary tumors of prostate cancer patients as well as normal
prostate tissue of the same patients. We have for the first time found
that a commercially readily available serum-free medium developed
for human keratinocyte (K-SFM, Gibco, Grand Island, NY) is very
useful in growing and maintaining primary HPE cells and for the
cultivation of short-term cultures of primary HPE cells (2, 3).
Efforts spanning more than half a century, since the pioneering
work of Burrows et al. (4), have produced only a few cell lines
derived from human prostate epithelium. To date, only three read-
ily and well-studied long-term human prostate cancer cell lines
exist (DU-145, PC-3, and LNCaP). All were derived from meta-
static lesions, thus leaving a void in reagents representing primary
localized adenocarcinoma of the prostate. Nevertheless, their use
has greatly contributed to current understanding of human pros-
tate carcinogensis and progression. Better understanding of the
process of malignant transformation, the availability of recombi-
nant DNA technology, and telomerase resulted in the successful
establishment of novel primary nonmalignant and malignant
tumor-derived HPE cell lines during the past decade. However,
despite extensive work on the development of human prostate can-
cer cell lines, the proportion of patients that give rise to immortal-
ized human prostate cancer cell lines is still disappointingly low.
Since the inception of this cell center, we have successfully been
able to establish for the first time a number of novel immortalized
HPE cell lines derived from primary malignant prostate tumor as
well as benign prostate tissues using telomerase, the gene that pre-
vent senescence. Furthermore, we have succeeded in the establish-
ment of HPE cell models for the study of prostate cancer in high
risk populations, one focusing on African American prostate cancer
and one focusing on familial prostate cancer (Table 1). Telomerase
is an enzyme responsible for replicating telomere and is composed
24 Human Prostate Epithelial Cell Cultures 385
Table 1
Phenotypic characteristics of hTERT-immortalized human prostate epithelial
cell lines
Soft agar colony Androgen Tumor formation

Cell line Tissue derivation formation (%) sensitivity in SCID mice
957 E/hTERT Malignanta <0.001 No 0/5
RC-58T/hTERT/SA#4 Malignant 0.120 Yes 5/5
RC92a/hTERT Malignant 0.155 No 2/5
RC-165N/hTERT Benignb <0.001 Yes 0/5
RC-170N/hTERT Benign <0.001 No 0/5
a
Familial prostate cancer patient
b
African American prostate cancer patient
of an RNA subunit containing an integral catalytic subunit, human

telomerase reverse transcriptase (hTERT). Recent findings have
implicated telomerase in the escape from cellular senescence.
Transfection of hTERT into selected human cell type can itself
induce immortalization. Interestingly, telomerse expression in
human somatic cells does not induce changes associated with a
transformed phenotype or an altered genetic phenotype.
Normal prostate epithelial stem cells have been demonstrated
to exist in the basal component. Studies have proposed that andro-
gen-independent stem cells give rise to two types of cells: stem cells
and androgen-independent transit amplifying cells which can divide
rapidly with limited proliferative capacity and can differentiate into
luminal cells through an intermediate phenotype. To date, several
putative stem cell populations have been identified as prostate stem
cells by means of clonal assay, identification of several cell surface
markers, and side population (SP) analysis. The first evidence for
cancer stem cells (CSCs) was shown in hematopoietic tumors, and
this principle has been implicated in other tumors including pros-
tate. Recent studies have postulated the existence of CSCs in the
primary prostate cancer cells, prostate cancer cell lines, and animal
models. However, this hypothetical model for the hierarchical
organization of CSCs remains unproven because of the lack of
appropriate in vitro and in vivo models.
Primary cell cultures derived directly from tissues or tumors
have a number of advantages because it is believed that primary
cells reflect well the characteristics of the original tissues. However,
primary cell cultures do have difficulties because of the limited
access, their finite lifespan and the specific culture techniques. On
the other hand, cell lines are widely used in many aspects of research
as the most common in vitro culture model because they have a big
386 J.S. Rhim
Table 2
Cell culture models and markers/methods for the studying prostate normal
and cancer stem cells
Purpose of the Methods or

study Cell culture models markers References
Normal stem cells HPE cells Type ii colonies Hudson et al. (13)
HPE cells CD44 Liu et al. (14)
HPE cells a2b1-integrin Collins et al. (15)
HPE cells CD133 Richardson et al. (16)
HPE cells SP Bhatt et al. (17)
RWPE-1:HPV-immortalized Clonal Tokar et al. (18)
HPE cells
RC165N/hTERT-immortalized CD133 Miki et al. (10)
benign cells
Cancer stem cells Primary cancer cells CD133 Collins et al. (19)
DU145, LAPC-4 & LAPC-9, CD44 Patrawha et al. (20)
a metastatic prostate cancer
cell line and xenograft
prostate tumor
LAPC-9, a xenograft human SP Patrawha et al. (21)
prostate tumor
RC92a/hTERT-immortalized CD133 Miki et al. (10)
malignant HPE cells
HPET, hTERT-immortalized Clonal assay Gu et al. (22)
malignant HPE cells
HPE cells human prostate epithelial cells
advantage in being easy to handle for their infinite reproducible

quantities. As described, most of the human prostate cancer cell
lines have been established from metastatic lesions or from xeno-
graft tumors. Despite extensive work on the development of many
human prostate cancer cell lines, only a few patient samples gave
rise to immortalized HPE cell lines but none from primary HPE
cells from prostate cancer patients.
Although it has been suggested that the study of CSCs should
be performed using primary cancer cells rather than prostate
cancer cell lines, recent evidence using SP analysis and cell surface
markers, such as CD133 or clonal analysis, shows that long-term
cultured cell lines may retain a hierarchical proliferation or differ-
entiation pattern as a potential of CSCs. We have shown that
nonmalignant- and malignant-tumor-immortalized cell lines
may contain a subpopulation of cells with stem cell properties
(5, 6). Several in vitro culture systems including primary cells,
immortalized benign tissue-derived, and prostate tumor-derived
cell lines (Table 2) (7–16), have been reported as useful in the

study of prostate normal stem and CSCs (17).
CACs theory has now emerged as an innovative theory within
the field of cancer biology, especially regarding solid tumors. It will
be important to show the existence of normal stem cells and CSCs
in prostate tissue. Although cell lines have some disadvantages or
shortcomings compared to primary cells, in vitro culture systems
including primary cells, immortalized normal cell lines, and cancer
cell lines may contain a heterogeneous and hierarchical subpopula-
tion as described. Thus, the development of in vitro culture sys-
tems not only provides a useful novel tool for understanding key
molecular pathways in prostate epithelial differentiation and pros-
tate cancer progression, but also may have important implication
for biological and pharmacological functions. Primary cells, immor-
talized normal and cancer cell lines, as in vitro models for normal
stem and CSCs in the prostate, may provide new insights into the
mechanisms of prostate cancer development.
In this chapter, we describe our experimental protocols used in
our center for the following:
(a) Technique of processing biopsy.
(b) Generation of primary HPE cells.
(c) hTERT-immortalization of primary HPE cells.
(d) Characterization of new hTERT-immortalized primary
HPE cell lines.
2. Materials
2.1. Processing Biopsy collection media (Collection medium).

Prostate Biopsy Tissue Keratinocyte serum-free medium (K-SFM) (Gibco, Introgen,
Corp. Cat. No. 17005-042) plus 2.5 μg/ml epidermal growth fac-
tor (EGF) human recombinant (Cat. No. 10450-013) and 25 μg/
ml bovine pituitary extract (BPE) (Cat. No. 13028-014) with
● 5% heat-inactivated fetal bovine serum (FBS) (Gibco, Introgen,
Corp.).
● 1% PSN (Penicillin 5 mg/Streptomycin 5 mg/ml. Gibco,
Introgen, Corp.).
● 1% fungizon (Amphotericin B 250 mg/ml, Gibco, Introgen,
Corp. Cat.No. 15290-034).
● 10 m M HEPES (Sigma Cat. No. H4034).
The prepared collection medium is sterile-filtered using a 0.2 μ
filter in 500 ml and stored at 4°C for up to 2 weeks.
2.2. Generation
of Primary HPE Cells 1. Growth and Maintenance Medium (KGM)
388 J.S. Rhim
K-SFM (Gibco, Introgen, Corp. Cat. No. 17005-042 with

supplements EGF, BPE, and antibiotics). The growth and
maintenance medium is the same as biopsy collection media,
but does not contain FBS.
2. Type 1 collagen-coated petri dishes (Becton-Dickinson,
Boston, MA).
3. Trypsin/EDTA solution; 0.2% trypsin/002% EDTA in Hanks
Balanced Salt Solution (HBSS).
4. Freezing medium: KGM, growth medium plus 10% dimethyl
sulfoxide (DMSO).
2.3. hTERT- 1. Immortalizing agent:

Immortalization A recombinant retrovirus construct LXSN-hTERT (18) gen-
of Primary HPE Cells erously provided by Vimla Band, Nebraska Cancer Center
containing the hTERT and a neomycin resistance gene.
2. Dulbecco’s Modified Eagle’s Medium (DMEM) with 15%
FBS (DMEM/15% FBS).
3. Polybrene (final concentration 10 μg/ml).
4. Phosphate buffered saline (PBS).
2.4. Equipments 1. Optimal cutting temperature (OCT), compound for fixing

and Culture Supplies specimens prior to cryosectioning.
2. Type-1 collagen-coated petri dishes (Becton-Dickinson,
Boston, MA) 100 cm.
3. Scalpel blades.
4. Glass microscope slides.
5. Sterile petri dishes (100 cm) for dissection steps.
6. Dissecting microscope.
7. Incubator at 37°C and 5% CO2.
3. Methods
3.1. Technique for Biopsies of prostate tumor tissues and nonmalignant tissues for
Processing Biopsy generating primary HPE cells were obtained from prostatectomy
specimens from prostate cancer patients according to Water Reed
Army Medical Center and the USUHS Internal Review Board pro-
tocol (see Notes 1 and 2) The tissues were pathologically diag-
nosed and their results were recorded as well as their clinical data
in CPDR data bank. The frozen tissues were also kept in deep
freezer for future studies at the pathology laboratory.
1. Working under a biosafety cabinet, assemble two to three

collagen-coated petri dishes, one microscope slide, OCT, one
scapel blade, a container of dry ice, and a bottle of collection
medium.
2. Label the petri dishes and slide with date and patient ID
number.
3. Remove the tube containing the biopsy, and spray with etha-
nol, and put it in the hood.
4. Invert the tube several times until you have dislodged the
chunk of biopsied tissue from the bottom of the tube and pour
into one petri dish.
5. If the tissue did not come out into the dish using a 5 ml
Strippette, wash the tube with media until it does.
6. Aspirate off the media using a Pasteur pipette.
7. Open the scalpel blade, and cut the tissue into three pieces.
8. Put a dollop of OCT onto the slide, then a tissue chunk, and
immediately place on dry ice.
9. Allow OCT to freeze, wrap the slide in foil, and store in a
−70 C freezer.
10. Process the remaining tissues for generation of primary HPE cells.
3.2. Generation We have processed more than 150 tissue samples derived from
of Primary HPE Cells nonmalignant and primary prostate cancer tissues of various histo-
from Biopsy Specimen logical grades. Although serum-free conditions have been devel-
oped to propagate HPE cells from primary tumors (19–21), we
have found that our tissue-processing of biopsy specimens and cul-
tivation methods were very effective, reproducible, and useful for
growing primary HPE cells. Keratinocyte-SFM retards fibroblasts
overgrowth, thereby allowing the study of a highly pure popula-
tion of keratinocytes. The low calcium ion concentration (<0.1 nM)
permits greater cell growth and shows cell differentiation (22). We
have successfully generated 134 strains of primary cells using our
protocol and mediums. Attempts were made to generate spontane-
ously immortalized cell lines by serial subcultivation in the KGM
medium, but all failed (see Note 3). About 75% of the primary
HPE cells were senesced within five serial passages (see Fig. 1)
agents. However, cryopreserved early passaged HPE cells can be
used for various future studies.
1. With a sterile blade, make a fine mince of the remaining tissues
as far as you can.
2. Add 10 ml of collection medium, mix well, and divide equally
among two to three petri dishes, making sure that at least one
or two pieces of tissue goes into each collagen-coated dish.
390 J.S. Rhim
Fig. 1. Comparative morphology of primary prostate epithelial cells and their serially passaged cells. Left: Primary prostate
epithelial cells from a fragment of prostate tissue. A typical epithelial morphology is seen. Right: The morphology of the
same primary prostate epithelial cells following five serial subcultivations. No live cells were present. All the cells were
senesced.
3. Add 5 ml of collection media and place dishes into a 37°C 5%

CO2 incubator. Within a week the minced tissue should be com-
pletely attached to the bottom surface of the culture dishes.
4. Do not change the medium before 1 week or the cells may not
have a chance to adhere and grow.
5. After 7 days of culture, aseptically remove the collection
medium and add 10 ml of KGM growth medium (no 5%FBS)
into a petri dish.
6. Incubate the cells at 37° C in humidified air of 5% CO2 until
reaching semiconfluency (about 70–80%), changing the
medium twice weekly.
7. Freeze aliquots of primary cells using 10% DMSO in KGM
until the cells are reestablished in secondary cultures for fur-
ther passages.
8. For serial passages, routine trypsinization using trypsin/EDTA
was employed once a week in the collagen-coated petri dishes
with a split ratio of 1–2.
3.3. hTERT- 1. PA 317 packaging cells producing retrovirus containing

Immortalization hTERT and neomycin-resistant gene (pLXSN hTERT) are
of Primary HPE Cells grown to 70–80% confluence in a 75 cm2 flask and are fed
(See Note 4) 10 ml of fresh DMEM +15FBS. After overnight incubation in
5%CO2 incubator, supernatants were collected and 0.22 μm
HA filtered and stored in aliquots at −70°C. These aliquots are
used as virus stock.
2. One-day-old early passaged HPE cells (about 70–80% confluent
in a 25 cm flask) are infected with 300 μl of virus stock in 3 ml
of fresh KGM containing polybrene at a concentration of
10 μg/ml and incubated overnight.
3. The virus is then removed and the infected cells are washed
with PBS once and replaced with fresh KGM medium.
4. The infected cells are thereafter subcultured weekly for further
serial passages.
5. No G418 selection is necessary because the uninfected cells
senesced a few passages later.
6. HPE cell lines should be tested mycoplasma contamination
and screening performed frequently.
3.4. Characterization Serial subcultivation of primary HPE cells in the KGM medium
of New hTERT- rarely survived beyond 15 passages. Therefore, we considered the
Immortalized Primary 20-passaged primary HPE cell line as an immortalized cell line and
HPE Cell Lines suitable to the cell characterization. Immortalized HPE cell line
can be characterized and their origin authenticated using methods
(2, 3) described in general below. Detailed protocols are beyond
the scope of this chapter.
1. Telomerase activity in transduced cells: To confirm that immortal-
ized primary HPE cells do contain the transduced hTERT, the
telomerase repeat amplification protocol (TRAP) can be used.
2. The epithelial origin of the cells can be verified by examining
the morphology and cytokeratin expression.
3. HPE cell lines can be examined by RT-PCR for the presence of
the prostate specific markers such as androgen-regulated pros-
tate specific homeobox gene NKX3.1, epithelial cell specific
cytokeratin 8, androgen receptor (AR), and prostate specific
antigen (PSA).
4. Western immunoblot analysis is performed to detect the
expression of prostate specific markers such as AR and AMACR
(alpha-methyl-CoA racemase).
5. Colony forming efficiency. A cell suspension (1 × 105 cells/ml)
in 5 ml of 0.35% Nobel agar with K-SFM is overlaid into a
60 mm dish containing a 0.5% agar base. Colonies >0.2 mm in
diameter were counted on day 21.
6. Tumorigenic potential. To determine tumorigenicity,
1 × 107 cells in 0.2 ml of PBS are injected into the mid-dorsal
intracapular region of adult male mice with immunodeficiency
disease (SCID)
7. Androgen sensitivity assay. 1 × 105 cells are plated into six-well
plates. K-SFM is supplemented with 0.1% BSA without dihy-
drotestosterone (DHT) (Sigma-Aldrich, St. Louis, MO) as
control or containing 0, 10, 1.0, 10.0 and 100 nM DHT,
respectively. Plates are placed into 37° C incubator with 5%CO2
for 4 days with changes of media 2 days later. Cells are
trypsinized and counted with a Coulter counter. The number
of cell counts was expressed as the mean value of triplicate
392 J.S. Rhim
observations (n = 3). To examine the cell response to DHT,

differences in mean values of cell counts between control and
treatment groups were compared by one way analysis of vari-
ance with Turkey’s pair-wise comparison test (n = 3). The level
of p < 0.05 was considered statistically significant.
8. Cytogenetic analysis. Cells are examined for their cytogenetic
characteristics by determing of the chromosome counts, ploidy
distribution and Giemsa (G)-banded karyotypes. Fluorescence
in site hybridization can also performed to check for the pres-
ence of chromosome translocations.
4. Notes
1. A tissue culture facility to generate prostate tissue specimens

must be based on coordinated, multidisciplinary effort to
obtain fresh tissues.
2. At the time the specimen is collected, sampling and selection
must be implemented through the histopathological assess-
ment of viable cells in the fresh specimens. Biopsy samples
should be processed for generation of primary cells as soon as
possible.
3. As described, although efficiently successful generation of pri-
mary cells was achieved using serum-free mediums (6–8)
including KGM medium (2, 3), no immortalized cell lines
were derived in the serum-free KGM medium system with
biopsy samples. However, the establishment of a few useful
immortalized human prostate cancer cell lines was achieved
from metastatic legions or xenografts using serum containing
commercially available medium (9).
4. As demonstrated, we were successful in the establishment of
primary HPE cells by using telomerase, the gene that prevents
cellular senescence. However, we have found that the immor-
talizing efficiency of hTERT of primary HPE cells was very
low. Therefore, we have also used the HPV-16 E6E7 gene, a
known efficient immortalizing agent as an additional immor-
talization approach.
Acknowledgments
This work was supported by a Department of Defense Prostate

Cancer Research Program (PCO30694).
References
1. Jemal A, Siegel R, Ward E, Murray T, Xu J, cells and cancer stem cells. Prostate Cancer
Thun MJ (2007) Cancer statistics 2007. CA Prostatic Dis 11:32–39
Cancer J Clin 57:43–66 13. Hudson DL, O’Hare M, Watt F, Masters JR
2. Yasunaga Y, Nakamura K, Ewing CM, Isaacs (2000) Proliferative heterogeneity in the
WB, Hukku B, Rhim JS (2001) A novel human human prostate evidence from epithelial stem
cell culture model for the study of familial pros- cells. Lab Invest 80:1243–1250
tate cancer. Cancer Res 61:5969–5973 14. Liu AY, True LD, LaTray L, Nelson PS, Ellis
3. Gu Y, Li H, Miki J, Kim KH, Furusato B, WJ, Vessella RL et al (1997) Cell-cell interac-
Sesterhenn IA et al (2006) Phenotypic charac- tion in prostate gene regulation and cytodif-
terization of telomerase-immortalized primary ferentiation. Proc Natl Acad Sci USA
non-malignant and malignant tumor-derived 94:10705–10710
human prostate epithelial cell lines. Exp Cell 15. Collins AT, Habib FK, Maitland NJ, Neal DF
Res 312:831–843 (2001) Identification and isolation of human
4. Rhim JS (2000) In vitro human cell models for prostate epithelial stem cells based on alpha(2)
the study of prostate cancer. Prostate Cancer beta(1)-integrin expression. J Cell Sci 114:
Prostatic Dis 3:229–235 3865–3872
5. Ratsch SB, Gao Q, Srinivasan DE, Wazer DE, 16. Richardson GD, Robson CN, Lang SH, Neal
Band V (2001) Multiple genetic changes are DE, Maitland NJ, Collins AT (2004) CD133,
required for efficient immortalization of differ- a novel marker for human prostate epithelial
ent subtypes of normal human mammary epi- stem cells. J Cell Sci 117:3539–3545
thelial cells. Radiat Res 155:143–150 17. Bhatt RI, Brown MD, Hart CA, Gilmore P,
6. Peehl DM (1992) Culture of human prostate epi- Ramani VA, George NJ et al (2003) Novel
thelial cells. In: Freshney RA (ed) Culture of epi- method for the isolation and characterization
thelial cells. Wiley-Liss, New York, pp 159–180 of the putative prostate stem cells. Cytometry
7. Chaproniere DM, McKeehan WI (1986) Serial 54:89–99
culture of single adult human prostate epithe- 18. Toker EJ, Ancrite BB, Cunha GR, Webber
lial cells in serum-free medium containing low MM (2005) Stem/progenitor and interme-
calcium and a new growth factor from bovine diate cell types and the origin of human
brain. Cancer Res 46:819–824 prostate cancer. Differentiation 73:
8. Daley JP, Epstein DA, Hawley-Nelson P (1990) 463–473
Growth of human epidermal keratinocytes in 19. Collins AT, Berry PA, Hyde C, Stower MJ,
kerotinocyte serum-free medium. Focus 12:68 Maitland NJ (2005) Prospective identification
9. Novone NM, von Logothetis CI, Eschenbach of tumorigenic prostate cancer stem cells.
AC, Troncoso P (1999) Model systems of Cancer Res 65:10946–10951
prostate cancer: uses of limitations. Cancer 20. Patrawala I, Calhoon T, Schneider-Broussard
Metastasis Rev 17:361–371 R, Li H, Rhatia B, Tang S et al (2006) Highly
10. Miki J, Furusato B, Li H, Gu Y, Takahashi H, purified CD44(+) prostate cancer cells from
Egawa S et al (2007) Identification of putative xenograft human tumors are enriched in tum-
stem cell markers, CD133 and CXCR4 in origenic and metastatic progenitor cells.
hTERT-immortalized primary nonmalignant Oncogene 25:1696–1708
and malignant tumor-derived human prostate 21. Patrawala I, Calhoon T, Schneider-Broussard
epithelial cell lines and in prostate cancer speci- R, Zou J, Claypool K, Tang DG (2005) Side
mens. Cancer Res 67:3153–3161 population is enriched in tumorigenic, stem-
11. Li H, Zhou J, Miki J, Furusato B, Gu Y, like cancer cells, whereas ABCG2+ and
Srivastava S et al (2008) Telomerase- ABCG2-cancer cells and similarly tumorigenic.
immortalized non-malignant human prostate Cancer Res 65:6207–6219
epithelial cells retain the properties of multipo- 22. Gu G, Yuan J, Willis M, Kasper S (2007)
tent stem cells. Exp Cell Res 314:92–102 Prostate cancer cells with stem cell characteris-
12. Miki J, Rhim JS (2008) Prostate cell culture as tics reconstitute the original human tumor in
in vitro models for the study of normal stem in vivo. Cancer Res 67:4807–4815
Chapter 25
Enzymatic Dissociation, Flow Cytometric Analysis,

and Culture of Normal Mouse Mammary Tissue
Michael Prater*, Mona Shehata*, Christine J. Watson, and John Stingl
Abstract
Evidence is emerging that the mouse mammary epithelium is arranged as a hierarchy that spans from stem
cells to lineage-restricted progenitor cells to differentiated luminal and myoepithelial cells. The use of
fluorescence-activated cell sorting (FACS) in combination with quantitative functional clonal assays
represents a powerful tool for studying the properties of mouse mammary stem and progenitor cells.
This chapter outlines the experimental procedures for generating single viable cell suspensions of mouse
mammary epithelial cells, immunostaining cells for flow cytometry, in vitro assays for the detection and
enumeration of mouse mammary progenitor cells, and in vivo assays for the detection and enumeration of
mouse mammary stem cells.
Key words: Mouse mammary gland, Stem cells, Flow cytometry, Cell culture
1. Introduction
The mouse mammary gland is a compound tubulo-alveolar gland

that is composed of a series of branched ducts that drain milk pro-
duced by alveolar structures during lactation. There are two gen-
eral lineages of epithelial cell in the gland: luminal cells which line
the lumen of the ducts and alveoli, and basal-positioned myoepi-
thelial cells that reside below the luminal cells and adjacent to the
basement membrane (1, 2). Evidence suggests that distinct epithe-
lial cell subtypes exist within the mammary gland and that these
cells are organized as a hierarchy with the mammary stem cell at
the apex of this hierarchy (3–11). A mammary stem cell is defined
*
Michael Prater and Mofna Shehata contributed equally.
395
396 M. Prater et al.
as a cell that can generate both ducts and lobules (complete with
luminal and myoepithelial cells) and can self-renew. It has been
demonstrated that distinct progenitor and differentiated luminal
and basal cell subpopulations exist in the mouse mammary epithe-
lium (3–6). This chapter describes the methods to (1) dissociate
mouse mammary glands, (2) generate a single cell suspension of
mammary cells, (3) separate luminal, basal, progenitor, stem cell-
enriched and stromal cells by fluorescence-activated cell sorting
(FACS)™, (4) detect progenitor cells using a colony-forming assay,
and (5) detect mammary stem cells by their ability to generate
ductal-lobular outgrowths when transplanted serially into epithe-
lium-divested mammary glands of female weanling mice.
2. Materials
2.1. Dissociation 1. Dulbecco’s Modified Eagle’s Medium/Nutrient Mixture F12

of Mouse Mammary Ham supplemented with 2 mM L-glutamine and 15 mM
Glands HEPES (DMEM/F12/H). Store at 4°C.
2. Collagenase and hyaluronidase enzyme dissociation mixture
“Collagenase/Hyaluronidase Gentle™” (10× concentrated stock,
StemCell Technologies) stored as 1 mL aliquots at −20°C.
3. Gentamicin solution (50 mg/mL, Gibco) stored at 4°C. Use
at 50 mg/mL final concentration.
2.2. Single Cell 1. Hanks’ Balanced Salt Solution (HBSS) (1×), liquid with cal-
Preparation and cium chloride and magnesium chloride, supplemented with
Staining 10 mM HEPES and 2% fetal bovine serum (FBS). Referred to
as HF.
2. Ammonium chloride solution (NH4Cl, StemCell Technologies)
stored at −20°C.
3. Trypsin (0.25% porcine trypsin) stored at −20°C.
4. Dispase solution 1 (5 mg/mL, StemCell Technologies) stored
as 1 mL aliquots at −20°C.
5. Deoxyribonuclease one (DNase, Sigma) dissolved at 1 mg/
mL in Dulbecco’s Modified Eagle Medium. Sterile-filter with
a 0.22 mm filter. Store DNase stock (1 mg/mL) in 100 mL
aliquots at −20°C.
6. Normal rat serum stored at −20°C.
7. 15 and 50 mL centrifuge tubes.
8. 5 mL FACS™ tubes Becton Dickinson.
9. 5 mL FACS tubes with 35 mm cell strainer caps (Becton
Dickinson).
25 Mouse Mammary Cell Culture 397
10. 40 mm cell strainer (Becton Dickinson).

11. 1.7 and 2.0 mL eppendorf tubes.
12. Antibodies and reagents for flow cytometry (deluted in HF):
CD45-biotin (clone 30-F11, eBioscience, use at 1 mg/mL)
Ter119-biotin (clone Ter119, eBioscience, use at 1 mg/mL)
CD31-biotin (clone 390, eBioscience, use at 1 mg/mL)
BP1-biotin (clone 6C3 eBioscience, use at 1 mg/mL)
EpCAM-Alexa Fluor (AF) 647 (clone G8.8, Biolegend, use at
1 mg/mL)
CD49f-Alexa Fluor 488 (clone GoH3, Biolegend, use at 1:100
dilution)
Rat IgG2a,k-Alexa Fluor 488 (clone RTK2758, Biolegend,
use at 1 mg/mL)
Rat IgG2a,k-Alexa Fluor 647 (clone RTK2758, Biolegend,
use at 1 mg/mL)
Streptavidin APC-Cy7 (Biolegend, use at 0.4 mg/mL)
13. 4¢,6-diamidino-2-phenylindole (DAPI, Invitrogen). Make up as
a sterile 1 mg/mL stock solution in distilled water and store at
−20°C in 1 mL aliquots. Use at 1 mg/mL final concentration.
2.3. Mouse Mammary 1. 60 mm cell culture dishes.

Colony-Forming Assay 2. Mouse EpiCult-B™ basal media (StemCell Technologies)
stored at 4°C and Mouse-EpiCult-B™ supplements, stored in
5 mL aliquots at −20°C.
3. FBS (StemCell Technologies catalogue number 06100) stored
in 5 mL aliquots at −20°C.
4. Recombinant human epidermal growth factor (EGF, Sigma) is
dissolved at 10 mg/mL in DMEM/F12/H with 1% bovine
serum albumin Fraction V (BSA, Sigma). Sterile-filter with a
0.22 mm filter. Store EGF stock (10 mg/mL) in aliquots at
−20°C.
5. Recombinant human fibroblast growth factor-basic (bFGF,
Peprotech) is dissolved at 10 mg/mL in DMEM/F12/H with
1% BSA. Sterile-filter with a 0.22 mm filter. Store FGF stock
(10 mg/mL) in aliquots at −20°C.
6. Heparin solution (2 mg/mL, StemCell Technologies) stored
at 4°C.
7. Gentamicin solution (50 mg/mL, Gibco) stored at 4°C. Use
at 50 mg/mL final concentration.
8. NIH 3T3 mouse embryonic fibroblasts stored in liquid
nitrogen.
9. Dimethyl sulfoxide (DMSO, Fisher Scientific).
10. Hypoxic (5% O2) incubator.

11. Giemsa stain (Fisher Scientific).
2.4. Transplantation 1. Growth factor reduced Matrigel™ (BD Biosciences). Store at

of Mammary Cells into −20°C. Thaw on ice.
Mouse Fat Pads 2. Trypan blue solution (0.4%, Sigma).
3. 25 mL Hamilton syringe 702LT Luer tip (Fisher).
4. Phosphate buffered saline (PBS).
5. BD Microlance needles, 22 Gauge (BD Biosciences).
6. Carnoy’s fixative (60% ethanol, 30% chloroform, and 10%
glacial acetic acid).
7. Carmine (Sigma).
8. Aluminum potassium sulfate dodecahydrate (Sigma).
9. Permount Mounting Medium (Fisher Scientific).
3. Methods
3.1. Dissociation 1. Prepare enzyme dissociation enzyme by adding 1 mL of 10×

of Mouse Mammary collagenase–hyaluronidase mixture to 9 mL DMEM/F12/H
Tissue and Generation supplemented with 50 mg/mL gentamicin in a 15 mL centri-
of Single Cell fuge tube.
Suspensions 2. Euthanize 1–2 virgin female mice according to institutional
3.1.1. Removal
guidelines and then pin the mouse in a stretched position on a
and Dissociation of Mouse
dissecting board and soak the abdomen with a generous
Mammary Glands
amount of 70% ethanol to flatten the fur so that the mammary
glands are not contaminated upon removal (see Fig. 1a).
3. Make an inverted Y-shaped incision along the midsection (see
Fig. 1b).
4. Pin back the skin (see Fig. 1c) and excise the fourth and third
mammary glands (see Figs. 1d–f).
5. Add glands to the diluted collagenase–hyaluronidase solution
and incubate at 37°C in a cell culture incubator for 14–16 h to
digest the tissue (see Note 1).
3.1.2. Preparation 1. Remove the 15 mL centrifuge tube containing the mammary

of a Single Cell Suspension glands from the 37°C incubator and pipette with a P1000 to
of Mammary Cells break up the glands (approximately 30 s).
2. Spin at 450 × g for 5 min at 4°C.
3. Discard the supernatant and suspend the cells in 1 mL of cold
HF (see Note 2).
4. Add 4 mL of cold NH4Cl (to lyse red blood cells), mix and
spin at 450 × g for 5 min at 4°C.
Fig. 1. Dissection of the number 3 (upper blue arrow in panel c) and 4 mammary glands (lower blue arrow in panel c) from mice.
5. Discard supernatant and add 1 mL of pre-warmed trypsin.

Pipette gently for 1–2 min.
6. Dilute with 10 mL of cold HF and spin at 450 × g for 5 min at 4°C.
7. Gently remove the supernatant. Add 1 mL of pre-warmed dis-
pase solution and 100 mL of 1 mg/mL DNAse. Pipette gently
for 1 min.
8. Add 10 mL of cold HF and filter through a 40 mm cell
strainer.
10. Cells are now ready to be stained for flow sorting.
3.2. Preparing Cells 1. Pre-block cells by suspending them in 1 mL of HF supple-

for Flow Cytometry mented with 10% normal rat serum and incubating them on
ice for 10 min.
2. During the preblock incubation, remove 10 mL of the cells and
add these to 10 mL trypan blue and 80 mL HF and perform a
trypan blue exclusion cell count using a hemocytometer.
3. After a 10 min preblock incubation on ice, add 6 mL of HF
and aliquot the cell suspension to seven FACS™ tubes at
1 mL/tube (see Note 3).
5. Discard supernatant, being careful to remove all liquid without
disturbing the pellet.
6. Stain cells as indicated below. Do all antibody incubations with
the cells at a cell density no greater than 107 cells/mL. All anti-
body incubations should be done on ice for 10 min followed
by one wash with 3 mL of cold HF and a centrifugation at
450 × g for 5 min at 4°C. On the last wash before addition of
DAPI, filter the diluted cells through a 5 mL FACS™ tube
with a 35 mm cell strainer cap before the centrifugation step
(see Notes 4–6).
Tube 1 = unstained
Tube 2 = DAPI
Tube 3 = CD49f-AF488/wash/DAPI
Tube 4 = EpCAM-AF647/wash/DAPI
Tube 5 = CD45-biotin + CD31-biotin + Ter119-biotin + BP1-
biotin/wash/streptavidin APC-Cy7/wash/DAPI
biotin + isotype control-AF488 + isotype control-AF647/
wash/streptavidin APC-Cy7/wash/DAPI
biotin + CD49f-AF488 + EpCAM-AF647/wash/streptavidin
APC-Cy7/wash/DAPI
7. After the last wash, suspend all cells in cold HF supplemented

with 1 mg/mL DAPI such that the final density of cells in sus-
pension is approximately 2 × 106 cells/mL.
8. Cells are now ready for flow cytometric analysis and/or
sorting.
3.3. Flow Cytometric 1. Place unstained control tube onto the FACS™ machine and
Analysis and Sorting run sample. Adjust voltages such that the background
of Mouse Mammary fluorescence is within the first log decade (see Note 7).
Cells 2. Run single color control tubes adjusting for background spec-
tral overlap and compensate accordingly.
3. To analyze the sample, collect at least 30,000 events. Gate
around all events based on forward (FSC) and side (SSC) scat-
ter, but excluding the events with the highest side scatter (see
Fig. 2a and Note 8). Then exclude doublets by gating the
events in the FSC-height by FSC-area parameters (see Fig. 2b).
Dead and dying cells are then excluded by gating on the DAPI-
negative events and by avoiding debris using the FSC parameter
(see Fig. 2c). Also exclude the events expressing intermediate
Fig. 2. Flow cytometric dot plots illustrating the gating strategy to identify viable mammary luminal and basal cells. The
MRU-enriched subpopulation is highlighted by the red circle.
levels of DAPI as these have low viability. To exclude most non-

epithelial (termed “lineage-negative”) cells, select the events
that do not express CD31, CD45, Ter119, and BP-1 (see
Fig. 2d). Draw another plot with EpCAM on the y-axis and
CD49f on the x-axis (see Fig. 2e), which permits visualization
of the luminal (EpCAMhighCD49fmed), basal
med high - -
(EpCAM CD49f ), and stromal (EpCAM CD49f ) cell
populations (see Note 9). The Mammary Repopulating Unit
(MRU; operational term for a mammary stem cell)-enriched
subpopulation is highlighted by the red circle (see Note 10).
4. Flow sorted subpopulations of cells can be collected into 5 mL
tubes containing 1–2 mL of HF (see Note 11). When collect-
ing lower cell numbers, an eppendorf tube containing 500 mL
of HF can be used as a collection tube. When sorting very low
cell numbers (<200 cells) and when accurate cell counts are
required, the cells can be sorted directly into 100 mL of PBS in
the wells of a 96-well ultralow adherent tissue culture dishes
that have a counting grid scratched into the underside of the
plate. After sorting, the plates can be spun at 450 × g for 5 min
to make the cells settle and the number of cells in the well can
be accurately counted under a phase contrast microscope. Add
cold Matrigel to a final concentration of 25% immediately after
counting the cells and place the plates on ice.
3.4. Mouse Mammary 1. Culture NIH 3T3 cells in high glucose Dulbecco’s Modified
Colony-Forming Eagle Medium (DMEM) supplemented with 5% FBS. When
Assays the cultures become 70–80% sub-confluent (see Note 12),
remove the media and wash once with PBS. Add pre-warmed
0.05% trypsin and incubate at 37°C with occasional agitation
until the cells lift off the flask (about 5 min). Add an equal
volume of DMEM/F12/H supplemented with 2% FBS and
centrifuge at 450 × g for 5 min to pellet the cells. Suspend the
cells at 1 × 106 cells/mL in DMEM/F12 supplemented with
2% FBS in 5 mL tubes. Irradiate the cells at 50 Grays (Gy) of
gamma (g) ionizing irradiation (or treat with mitomycin-C if
there is no access to a gamma irradiator). Centrifuge the cells
at 450 × g for 5 min and discard the supernatant. Suspend the
irradiated cells at a concentration of 2 × 106 viable cells/mL in
freezing media (50% DMEM/F12 + 44% FBS + 6% DMSO).
Cool at −1°C/min to −80°C and store in working aliquots of
2 × 106 cells per cryovial in liquid nitrogen for long-term
storage.
2. Make up the mouse mammary progenitor cell culture medium:
Mouse EpiCult-B™ basal medium supplemented with Mouse
EpiCult-B™ supplements, 5% FBS, 10 ng/mL EGF, 10 ng/
mL bFGF, 4 mg/mL heparin, and 50 mg/mL gentamicin (see
Note 13).
3. Label the bottom of the 60 mm dishes before cell seeding. Do

not label the lids as these can get easily mixed up when using a
large number of dishes.
4. Add a suitable number of dissociated single cells to 8 mL of
complete Mouse EpiCult-B™ (see Notes 14 and 15). Add
50,000 viable irradiated NIH 3T3 cells per mL of complete
Mouse EpiCult-B™ media. Seed the cell suspension across two
duplicate 60 mm cell culture dishes at 4 mL per dish.
5. Incubate the cell culture dishes in a hypoxic incubator (5%
oxygen, 5% carbon dioxide, 37°C) for 7 days (see Note 16).
6. After 7 days, remove the media and gently rinse the dishes with
PBS. Completely remove the PBS and add 1 mL of acetone–meth-
anol (1:1) per 60 mm dish for 30 s. Remove the acetone–methanol
and allow the dish to air-dry. Add 1–2 mL of Giemsa stain (diluted
1:10 with distilled water) for 2–3 min. Remove the Giemsa stain
and rinse the dishes twice with distilled water and air-dry.
7. Count the number of colonies per dish using a scoring grid,
which can be made by taking a 100 mm dish and making verti-
cal lines using a fine-tip pen. The space between the lines
should be one field-of-view under the microscope.
3.5. Characterization Epithelial colonies can be derived from either luminal or basal cells
of Mouse Mammary and typically have a “cobblestone” appearance, although the basal
Epithelial Cell Colonies colonies are usually more dispersed than the compact luminal colo-
nies. Distinguishing luminal and basal colonies is difficult by mor-
phology alone. Immunofluorescence staining for cytokeratins 14
(K14) and 18 (K18) allows basal and luminal cells, respectively, to
be distinguished. Colonies derived from luminal cells express K18
only. Basal cells usually give rise to mixed colonies that contain
both K18+ luminal and K14+ basal cells. Other markers for luminal
cells include MUC1 and K8, whereas basal cells can also be
identified by expression of K5, p63, and smooth muscle actin
(2, 12–14) (see Fig. 3).
Stromal colonies have a dispersed and mesenchymal phenotype
and can be easily distinguished from epithelial colonies (see Fig. 4).
These colonies should be excluded from colony counts if only epi-
thelial progenitors are of interest.
3.6. Detection The frequency of mammary repopulating units (MRUs) within a

of Mouse Mammary mammary cell preparation can be estimated (see Note 17) by trans-
Stem Cells planting the cells at limiting dilutions (e.g., a dose where at least
one negative outcome will be obtained) into cleared mouse mam-
mary fat pads (4–7). MRUs are the operational term for cells that
have the ability to generate ductal-lobular outgrowths when trans-
planted into a cleared mammary fat pad. Limiting dilution analysis
is a low-resolution tool, and as a result the 95% confidence inter-
vals generated with the MRU frequency estimates are large. This
Fig. 3. Colonies derived from a flow sorted luminal cell (a) and a basal cell (b) stained for cytokeratin 18 (red), cytokeratin
14 (green) and DAPI (purple) by immunofluorescence. Bar = 100 mm.
Fig. 4. Stromal colonies (a and b) grown in Mouse EpiCult-B™ media. Bar = 500 mm.
confidence interval can be narrowed by transplanting cells at limit-

ing numbers, that is, at that dose in which an MRU may or may
not be in the transplant inoculum. For example, if the MRU fre-
quency is 1 MRU for every 1,000 total cells, then it would be more
efficient to transplant cells at a dose of 1,000 cells/transplant,
rather than doses that are significantly higher or lower than this.
Initial pilot transplants using a broad range of transplant doses may
need to be performed to determine the approximate MRU fre-
quency. A second round of transplants can then be done at more
refined doses to generate accurate MRU estimates. It is our experi-
ence that different donor mouse strains have different MRU fre-
quencies (e.g., 1 MRU in 100 total mammary cells isolated from
FVB mice, and 1 MRU in 1,300 total mammary cells isolated from
C57Bl6 mice). A suggested protocol to identify MRUs among
total or flow-sorted cell preparations is as follows (see Note 18):
1. Suspend cells in 65% PBS + 25% Growth Factor Reduced

Matrigel + 10% trypan blue solution (0.4%) at an appropriate
concentration such that 10 mL contains the desired cell dose.
Keep the cells on ice.
2. Clear the endogenous mammary gland epithelium from the
inguinal (number 4) mammary glands of a 21 day-old female
recipient mouse (see Note 19).
3. Using a 25 mL Hamilton syringe and a 22 Gauge needle, inject
10 mL of cell suspension into each cleared fat pad (see Note 20).
4. Mate the mice 3 weeks after surgery and remove the fat pads
when the mice are visibly pregnant.
5. Fix the fat pads in 5 mL of 6 parts absolute ethanol, 3 parts
chloroform and 1 part glacial acetic acid (Carnoy’s fixative)
overnight at room temperature.
6. Remove the Carnoy’s fixative and wash in 5 mL of 70% ethanol
for 15 min at room temperature.
7. Slowly add 5 mL of distilled water to the 70% ethanol to grad-
ually reduce the concentration of ethanol and then wash the
glands once in distilled water for 5 min. Remove the distilled
water.
8. Make up carmine stain: 1 g carmine mixed with 2.5 g alumi-
num potassium sulfate in 500 mL distilled water. Boil for
20 min. Filter through Whatman filter paper to remove any
precipitates. Store at 4°C until color begins to fade.
9. Add 5 mL of carmine stain and leave overnight at room
temperature.
10. Remove the carmine stain and wash sequentially in 70, 95, and
100% ethanol for 15 min each. Immerse in two sequential
baths of xylene for 15 min each and mount with Permount
Mounting Medium.
11. A positive engraftment must originate from within the fat pad
(see Note 21) and must contain ducts and lobules. Note how
much of the fat pad has been filled by the engrafted epithelium.
12. To assess the self-renewal capacity of an MRU, the primary
engraftment should be dissociated as described above (see
Subheading 3.1) and cells should be transplanted at limiting
dilutions into cleared fat pads of secondary recipients.
4. Notes
1. It is not necessary to remove the lymph nodes from the mam-

mary gland since these cells will be excluded during cell sort-
ing, nor is it necessary to mince the glands prior to enzymatic
dissociation.
2. Keep all buffers cold and keep cells on ice to minimize cell
clumping.
3. Cells can be evenly distributed among the different staining
FACS™ tubes when doing flow cytometric analysis. However,
when sorting cells rather than just analyzing, it is appropriate
to minimize the number of cells in the control tubes since a
large number of cells are not required to be analyzed for these
controls; most of the cells from the original sample can be
placed into the tube which contains the cells that will be sorted.
This will minimize cell wastage.
4. The use of an anti-EpCAM antibody instead of an anti-CD24
antibody is recommended since the EpCAM/CD49f antibody
combination permits better resolution of the luminal and basal
cell subpopulations for some strains of mice (e.g., C57 Bl6).
5. In some cases it may be desirable to deplete the contaminating
hematopoietic, stromal and endothelial cells using an immuno-
magnetic approach (e.g., Mouse Epithelial Enrichment Kit
from StemCell Technologies catalogue number 19758). This
will free a channel on the flow cytometer and permits an addi-
tional fluorochrome to be used.
6. The inclusion of an anti-BP1 antibody in the lineage depletion
cocktail results in depletion of a large proportion of stromal
cells. If stromal cells are the cells of interest, it may be desirable
to remove the BP1 antibody from the lineage depletion
cocktail.
7. A thorough discussion of the flow cytometric analysis of cells
derived from mouse mammary tissue is presented in ref. 15.
8. Events with very high SSC have high levels of autofluorescence
when analyzed on the channel used to detect Alexa Fluor 647;
as a result it is best to gate out these events.
9. The luminal subpopulation can be further subdivided into pro-
genitor and non-progenitor cell subpopulations by the expres-
sion of CD61 (see ref. 16).
10. Approximately 70–80% of all MRUs are localized in the bright-
est 20% of the EpCAM+ cells within the basal cell population
(e.g., in the red circle outlined in Fig. 2e). Their frequency in
this gate when analyzing C57 Bl6 mice is approximately 1
MRU in every 50 sorted cells.
11. The actual number of cells collected by FACS™ often does not
mirror the number of events the machine states that it has
sorted. It is recommended that a test sort is performed at the
beginning of every sorting session. To do this, collect 100,000
DAPI− events and then count the number of viable cells in this
cell preparation using trypan blue and a hemocytometer to
determine the actual cell yield, and use this to calculate a sort
Fig. 5. Mouse mammary epithelial colonies grown in Mouse-EpiCult-B™ in atmospheric oxygen (a and b) versus 5%
oxygen (c and d). Bar = 500 mm.
correction factor. Adjust all subsequent cell-sorting counts by

this correction factor.
12. Do not allow cultures of NIH 3T3 mouse embryonic fibroblasts
to become confluent or they will not support clonal growth of
mammary epithelial cells. Over time, the NIH 3T3 cells may
lose their ability to support clonogenic growth of the mam-
mary cells. In such cases new stocks of feeder cells should be
generated from early passage master stocks that are preserved
in liquid nitrogen.
13. Add gentamicin to culture media if the cells to be cultured
have been sorted by flow cytometry under non-sterile condi-
tions. Do not use penicillin–streptomycin because this may
influence the number of epithelial colonies generated in vitro.
14. In our experience, the vast majority of progenitors that grow
in Mouse EpiCult-B™ are progenitors that have a luminal-like
phenotype and generate luminal-restricted progeny, although
a small proportion (~1–3%) of progenitors are localized to the
MRU gate and generate mixed lineage (luminal and basal)
colonies in vitro.
15. The number of cells seeded should be titrated the first time
around. The colony forming cell (Ma-CFC) frequency in
unsorted mammary epithelial cells is approximately 3%. Choose
a seeding density that results in 50–100 colonies because at
this density individual colonies can be easily distinguished.
16. Although the numbers of epithelial colonies generated in
hypoxic and normoxic conditions is equivalent, colonies gen-
erated in hypoxic conditions are larger than those generated in
normoxic conditions (see Fig. 5).
17. The MRU frequency can be estimated using the L-Calc com-
puter statistical program. The program can be downloaded for
free from www.stemcell.com. The program is compatible with
PCs only. Alternatively, the more robust ELDA computer sta-
tistical program can be downloaded from http://bioinf.wehi.
edu.au/software/elda and is compatible with both Macs and
PCs.
18. There is a certain cell toxicity associated with antibody expo-
sure and flow sorting among mouse mammary MRUs and
Ma-CFCs. We have estimated this toxicity level to be approxi-
mately 50%.
19. The endogenous mammary gland epithelium will not have
grown past the lymph node in 21-day-old female mice. Remove
all endogenous tissue that is lateral to the lymph node. A detailed
protocol on mammary fat pad clearing can be found in ref. 17.
20. Placing a small piece of Teflon (plumber’s) tape around the
end of the Hamilton syringe creates a better seal with the nee-
dle. Also remove the plunger from the syringe and dip it into
mineral oil. This thin coating of oil permits a better vacuum
seal to be made between the barrel and the syringe.
21. If the endogenous mammary epithelium was not cleared prop-
erly, false positive engraftments can occur. If the engraftment
does not originate from within the fat pad but instead enters
from the edges, then exclude it from the results. To avoid this
problem, genetically tagged (e.g., eGFP) cells could be
transplanted.
Acknowledgments
MP, MS, and JS are funded by Cancer Research UK, The University
of Cambridge, and Hutchison Whampoa Limited. MP and JS are
also funded by the Breast Cancer Campaign. CJW is funded by the
Biotechnology and Biological Sciences Research Council, the
Medical Research Council (UK), and the Breast Cancer Campaign.
References
1. Daniel CW, Silberstein GB (1987) Postnatal 10. Smith GH (1996) Experimental mammary epi-
development of the rodent mammary gland. thelial morphogenesis in an in vivo model: evi-
In: Neville MC, Daniel CW (eds) The mam- dence for distinct cellular progenitors of the
mary gland: development, regulation and func- ductal and lobular phenotype. Breast Cancer
tion. Plenum, New York, pp 3–36 Res Treat 39:21–31
2. Taylor-Papadimitriou J, Lane EB (1987) 11. Kordon EC, Smith GH (1998) An entire func-
Keratin expression in the mammary gland. In: tional mammary gland may comprise the prog-
Neville MC, Daniel CW (eds) The mammary eny from a single cell. Development
gland: development, regulation and function. 125:1921–1930
Plenum, New York, pp 181–215 12. Skalli O, Ropraz P, Trzeciak A, Benzonana G,
3. Smalley MJ, Titley J, O’Hare MO (1998) Gillessen D, Gabbiani G (1986) A monoclonal
Clonal characterization of mouse mammary antibody against a-smooth muscle actin: a new
luminal epithelial and myoepithelial cells sepa- probe for smooth muscle differentiation. J Cell
rated by fluorescence-activated cell sorting. In Biol 103:2787–2796
Vitro Cell Dev Biol Anim 34:711–721 13. Pemberton L, Taylor-Papadimitriou J, Gendler
4. Shackleton M, Vaillant F, Simpson KJ, Stingl J, SJ (1992) Antibodies to the cytoplasmic
Smyth GK, Asselin-Labat ML, Wu L, Lindeman domain of the MUC1 mucin show conserva-
GJ, Visvader JE (2006) Generation of a function throughout mammals. Biochem Biophys
tional mammary gland from a single stem cell. Res Commun 185:167–175
Nature 439:84–88 14. Barbareschi M, Pecciarini L, Cangi MG,
5. Stingl J, Eirew P, Ricketson I, Shackleton M, Macrì E, Rizzo A, Viale G, Doglioni C
Vaillant F, Choi D, Li HI, Eaves CJ (2006) (2001) p63, a p53 homologue, is a selective
Purification and unique properties of mammary nuclear marker of myoepithelial cells of the
epithelial stem cells. Nature 439:993–997 human breast. Am J Surg Pathol 25:
6. Sleeman KE, Kendrick H, Ashworth A, Isacke 1054–1060
CM, Smalley MJ (2006) CD24 staining of 15. Alexander CM, Puchalski J, Klos KS, Badders
mouse mammary gland cells defines luminal N, Ailles L, Kim CF, Dirks P, Smalley MJ
epithelial, myoepithelial/basal and non-epithe- (2009) Separating stem cells by flow cytome-
lial cells. Breast Cancer Res 8:R7 try: reducing variability for solid tissues. Cell
7. Moraes RC, Chang H, Harrington N, Landua Stem Cell 5:579–583
JD, Prigge JT, Lane TF, Wainwright BJ, Hamel 16. Asselin-Labat ML, Sutherland KD, Barker H,
PA, Lewis MT (2009) Ptch1 is required locally Thomas R, Shackleton M, Forrest NC, Hartley
for mammary gland morphogenesis and sys- L, Robb L, Grosveld FG, van der Wees J,
temically for ductal elongation. Development Lindeman GJ, Visvader JE (2007) Gata-3 is an
136:1423–1432 essential regulator of mammary-gland mor-
8. Hoshino K (1964) Regeneration and growth of phogenesis and luminal-cell differentiation.
quantitatively transplanted mammary glands of Nat Cell Biol 9:201–209
normal female mice. Anat Rec 150: 221–235 17. Young LJT (2000) The cleared mammary fat
9. Daniel CW, DeOme KB, Young JT, Blair PB, pad and the transplantation of mammary gland
Faulkin LJ (1968) The in vivo life span of nor- morphological structures and cells. In: Ip MM,
mal and preneoplastic mouse mammary glands: Asch BB (eds) Methods in mammary gland
a serial transplantation study. Proc Natl Acad biology and breast cancer research. Kluwer/
Sci USA 61:52–60 Plenum, New York, pp 67–74
Chapter 26
Isolation and Characterization of Human

Hair Follicle Epithelial Cells
Keita Inoue and Kotaro Yoshimura
Abstract
The mammalian hair follicle epithelial component contains various lineages of keratinocytes as well as their
progenitor/stem cells. To characterize the subpopulations contained within this component and assess
their functional capacity, the development of a feasible method to isolate them is greatly needed. In this
chapter we describe a standard method by which a small subset of human follicular keratinocytes can be
isolated for subsequent analysis. Detailed methods for enzymatic digestion of fresh human scalp, flow
cytometry, cultivation of sorted cells and a colony forming assay are described.
Key words: Keratinocyte, Human, Hair follicle, Flow cytometry, Colony forming, Stem cell
1. Introduction
The mammalian hair follicle has a stem cell niche, called the bulge,
in which a quiescent or slowly cycling minor cell population resides
(1). Once stepping outside this niche, they give rise to all lineages
of keratinocytes, both hair follicle and epidermis, and the seba-
ceous gland as well (2, 3). In addition, recent progress in the analy-
ses of mammalian hair follicle biology has revealed the existence of
smaller subsets of cells inside this niche than have ever been
described before. For instance, it has been shown recently that
there are distinct stem cell populations other than the bulge, which
arise from different progenies (4–6). Since the classical manipula-
tion of the hair follicle using a surgical microscope to obtain small
subsets of follicular cells is not feasible, approaches such as
fluorescence activated cell sorting (FACS) or laser capture micro-
dissection to isolate follicular cells of interest are more relevant
411
412 K. Inoue and K. Yoshimura
now. It has been shown in mice using a gradient of reporter

fluorescence expression that a minor subset of hair follicle cells can be
successfully isolated using FACS for various functional studies
including microarray analysis (7). Likewise, human bulge cells can
be successfully isolated using FACS and/or magnetic activated cell
sorting (MACS) based on a specific combination of antibodies
against surface antigens (8). In the human hair follicle, as well as in
mouse, there is an increasing need for functional studies of the
minor compartments of various lineages. There are two inevitable
requirements to fulfill this purpose. First, each cell population must
be isolated as live cells, allowing subsequent cell culture. Second,
each population should be isolated depending on the features that
specify their biological properties such as gene expression, cell
cycle, or cell morphology.
We describe in this chapter a practical method for isolating the
minor components of human hair follicle epithelial cells, which we
have recently developed (9). The method basically comprises three
steps: (1) confirmation of the anatomical location of the target
subpopulation by immunostaining, (2) microdissection and enzy-
matic digestion of hair follicles to obtain single cells, followed by
their isolation by FACS and cell cycle analyses, and (3) functional
analyses of the isolated cells in culture, including their colony form-
ing ability. We especially put an emphasis on how to obtain single
live cells, as well as sort and analyze them (steps 2 and 3). This
method enables the specific isolation of living cell populations of
interest, allowing downstream analyses which are difficult with
other methods.
2. Materials
2.1. Microdissection 1. Autoclaved dissection tools: spring microscissors, #5 forceps

and Enzymatic (Fine Science Tools).
Digestion of Hair 2. Dissection stereoscopic microscope.
Follicle
3. Dulbecco’s Phosphate-buffered saline without calcium and
magnesium (PBS).
4. Dulbecco’s modified Eagle’s medium (DMEM), high glucose.
5. 100× Penicillin (5,000 unit/ml)-Streptomycin (5,000 μg/ml)
mixture (Penstrep).
6. Fetal Bovine Serum.
7. Dispase I (Sanko Junyaku), purified and sterile Bacillus poly-
myxa-derived protease, which cleaves nonspecific proteins
including collagen I/IV and fibronectin but not cell membrane.
26 Hair Follicle Epithelial Cells 413
Dissolve in high-glucose DMEM supplemented with 10% FBS

and use at the final concentration of 1,000 protease unit (PU)/
ml. Use freshly made solution for every experiment.
An alternative product is available as “Dispase Grade I” from
Roche, although one unit of this equals approximately 416 PU
of Sanko-Junyaku Dispase I.
8. 10× Trypsin (0.5%) + EDTA (0.2%).
9. 40-μm-pore nylon mesh cell strainer.
2.2. Flow Cytometry Equipment: FACS machine (e.g., BD FACS Aria). Any FACS
and Cell Sorting machine can be used if it enables analyses of as many fluorochromes
as you want to use.
Software: FACS software compatible to the FACS machine
(e.g., BD FACS Diva).
Reagents:
1. Fluorescent-conjugated antibodies against target of interest for
flow cytometry use.
You should titer the antibody concentration to optimize
the working dilution before its first use. The following anti-
bodies are typical positive and negative markers for the bulge
cell population.
(a) Positive control: R-phycoerythrin (PE) conjugated anti-
CD200 antibody (clone MRC OX-104, 1:100).
(b) Negative control: Allophycocyanin (APC) conjugated
anti-CD34 antibody (clone MRC 8 G12, 1:100).
2. 7-Aminoactinomycin D, unconjugated (7-AAD, available as a
ready-to-use product from BD Pharmingen). Store at 4°C and
protect from light.
3. Prepare sterile 0.2% bovine serum albumin (BSA) in PBS as
“wash buffer” and 0.5% BSA in PBS as “sample buffer.” Sterile
7.5% BSA solution in PBS is commercially available. If you pre-
pare the wash/sample buffers from powdered BSA, pass the
dissolved solution through a 0.22 μm filter prior to use. The
solution can be stored at 4°C up to 2 weeks.
4. PBS, calcium and phenol red free.
5. Sample tube, 5-ml 40-μm-pore filter top tubes.
6. Collection tube, 15-ml collection tubes.
2.3. Cell Culture 1. Defined Keratinocyte serum-free Medium (Gibco).

and Colony Assay 2. Laminin-coated 6-well multiwell plate (BioCoat, BD).
3. 4% paraformaldehyde (PFA) in PBS.
4. Rhodanile blue: 2% Rhodamine B (Sigma) and 2% Nile blue
(Sigma) diluted in ethanol.
3. Methods
3.1. Microdissection 1. Collect a human hairy scalp sample from surgery following
and Enzymatic institutional guidelines (for example, facelift operations). Store
Digestion of the the scalp sample in wet gauze with PBS at 4°C until it can be
Hair Follicle processed (see Note 1).
2. Cut hair shaft leaving approximately 2 mm on the scalp and
rinse the scalp in 10 ml of PBS containing 2× Penstrep
(100 unit/ml Penicillin and 100 μg/ml Streptomycin) in a
50 ml-conical tube for 5 min. Repeat once. Transfer the scalp
to a new 10-cm culture plate containing 5 ml of PBS without
Penstrep.
3. Trim the subcutaneous fat well using spring microscissors and #5
forceps under a dissection stereoscopic microscope. Cut the scalp
into strips of about 3-mm width (Fig. 1a, b) (see Note 2).
4. Transfer the scalp skin strips into a new 6-cm culture plate
containing 8 ml of Dispase solution (Dispase I, 1,000 PU/ml,
in high-glucose DMEM supplemented with 10% FBS). DMEM
can be with phenol red at this point as phenol red can be
washed away in the downstream steps.
5. Incubate overnight at 4°C .
6. Wash the strips in 10 ml of fresh PBS twice.
7. Transfer the strips into a new 10-cm culture plate containing
5 ml of PBS.
8. Separate the epidermis/follicle and the dermis using #5
forceps under a dissection microscope (see Note 3). Then, cut
the follicles at the level of the infundibulum to separate them
from the epidermis (Fig. 1c, d).
9. Transfer the follicles into a new 3.5-cm plate containing 1 ml
of trypsin (0.05%)/EDTA (0.02%) mixture in PBS.
10. Incubate at 37°C for 30 min. Briefly pipette follicles, using a
1 ml pipettor, in the enzyme solution every 5 min during the
digestion (Fig. 2a, b) (see Note 4).
11. Transfer all of the follicles and solution into a 50-ml conical
tube containing 10 ml of PBS with 10% FBS to stop digestion.
Pipette ten times using a 10-ml pipette.
12. Place a 40-μm-pore nylon mesh cell strainer on top of a new
50-ml conical tube. Pass the resultant solution through the
strainer (Fig. 2c) (see Note 5).
13. Centrifuge the cell suspension at 170 × g for 5 min at RT.
14. Discard supernatant and suspend cells in an appropriate vol-
ume of cooled sample buffer, PBS with 0.5% BSA (see Note 6).
Keep them on ice.
Fig. 1. Preparation and dispase digestion of hair follicles. (a) Human scalp skin strip from facelift surgery. (b) Subcutaneous
fat was removed using microscissors. (c) Each hair follicle was plucked after Dispase digestion. Note every epithelial
component of pilosebaceous unit, including follicular epithelium, sebaceous gland, and epidermis as well, is kept attached
to each other. (d) Follicular epithelial components to be digested by trypsin. Epidermis and infundibulum distal to the
sebaceous gland duct have been cut away.
15. Take 20 μl of the sample solution and stain the cells by add-
ing an equal volume of Trypan blue. Count the number of
viable cells by excluding blue-stained dead cells using a
hemocytometer under microscope. Otherwise, use an auto-
mated cell counter to count both viable and nonviable
nucleated cells.
3.2. Flow Cytometry/ Prepare human scalp sections and perform immunohistochemistry
Cell Sorting on the sections with specific antibodies which can label the target
cells of interest. Determine the anatomical location where your
3.2.1. Determining the
target of interest is expressed (see Note 7). Anti-CD200 antibody
Target Populations and
(clone MRC OX-104, 1:50) and anti-keratin-15 antibody (clone
Their Markers Based on
LHK15, 1:300) can be used as the positive control for the bulge
Immunohistochemistry
region which is located around the insertion of the arrector pili
muscle. Anti-CD34 antibody (clone 8 G12, 1:300) works as the
negative marker for the bulge region.
Fig. 2. Trypsin digestion of hair follicles. (a) Cells start to detach from hair follicle just after applying trypsin. (b) At the end
of digestion (30 min). The cells in the basal layer are more susceptible to the enzyme. Some inner components remained
attached. (c) Various types of cells from the hair follicle are contained in the resultant suspension.
3.2.2. Labeling of Cells 1. Adjust the cell concentration to 106 cells/ml in sample buffer.
with Fluorescent-
2. Select fluorescent-conjugated antibodies which detect the cell sur-
Conjugated Antibodies
face markers expressed in the cell population of interest. Typically
use a combination of fluorochromes such as R-phycoerythrin
(PE), allophycocyanin (APC), and fluoroscein isothiocyanate
(FITC). Label sample tubes for each antibody, including “unstained
control,” “7-AAD,” “isotype IgG control,” “single staining,” and
“multiple staining (for sorting)” (see Note 8).
3. Dispense 100 μl (1 × 105 cells) of the cell suspension into each
sample tube for analysis. For sorting, dispense up to 1,000 μl
(1 × 106 cells) of cells per tube. Multiply the sorting tubes, if
necessary and there are more cells available. Keep them on ice
throughout the following steps
4. Add antibody to each tube (see Note 9). Incubate on ice for
30 min in the dark.
5. Add 4 ml of cooled wash buffer (PBS with 0.2% BSA).
6. Centrifuge at 170 × g for 5 min.
7. Aspirate the supernatant.
8. For analyses, suspend pellets in 300 μl of sample buffer and
keep the cells on ice in the dark until analysis is carried out. For
sorting, suspend pellets in 3,000 μl of sample buffer. Just make
sure every tube has more than 300 μl of solution, because less
volume may cause bubbling in the sheath line.
3.2.3. Cell Sorting 1. Set up the fluidics with a 100 μm nozzle. Do not use a nozzle
with smaller diameter (i.e., 70 μm) since it may cause damage
to the cells by shear stress.
Fig. 3. Example of FACS analyses to exclude debris and dead cells from freshly digested
human scalp derived epithelial cells. (a) Gate to exclude small debris from cells. (b) Gate
to exclude 7-AAD positive dead cells.
2. Turn on the FACS sorter and laser exciter, then computer and
FACS software.
3. Prime the sample sheath and fluidics. Calibrate the cell sorter
in “low pressure” setting. Do not use a higher pressure setting
since it may damage the cells by shear stress.
4. Add 1 μl of 7-AAD to each sample, except the “unstained con-
trol,” and incubate 15 min on ice in the dark. Proceed to the
following steps within 30 min (see Note 10). Pass cells through
a 40 μm filter top just before analysis or sorting.
5. Run the “unstained control,” followed by the “7-AAD” sam-
ple. Adjust detector voltages. Determine gating on FSC and
SSC parameters to exclude cellular debris (small) and large dif-
ferentiated keratinocytes. Use the ratio area/width on FSC for
doublet exclusion. Determine the gating on 7-AAD parameter
to exclude 7-AAD-positive dead cells (Fig. 3).
6. Run the “isotype IgG control” samples and determine the
appropriate gates for each fluorochrome (see Note 11).
7. Run the “single stain” samples and determine the appropriate

compensation values.
8. Run the “multi-stain” samples and define the gate to sort the
cell population of interest based on the standard FSC/SSC
profile, as well as each fluorochrome (e.g., CD200+CD34− human
hair follicle epithelial stem cells around the bulge region).
9. Set up 15-ml collection tubes, each containing 8 ml of Defined
Keratinocyte Serum-free Medium (see Note 12). Cells should
be sorted at a low flow rate (i.e., below 2,000 cells per second).
Record the sorted cell number. Keep tubes on ice until plating
for culture.
3.3. Cell Culture and 1. Centrifuge the sorted cells at 170 × g for 5 min. Suspend the
Colony Forming Assay cell pellet in 1 ml of fresh Defined Keratinocyte Serum-free
Medium.
3.3.1. Culture of Sorted
Keratinocytes 2. For colony forming assay, plate cells at three different concentra-
tions: 103, 3 × 103 and 104 cells, in laminin-coated 6-well plates con-
taining 2 ml of serum free medium. (see Note 13) For serial
subculture plate 5 × 104 cells. Culture at 37°C in 5% CO2 for
1–2 weeks. Change the medium every second day (see Note 14).
3. For serial subculture, trypsinize the primary culture and seed
them on new plates. First aspirate the medium, wash the
adherent cells twice using 4 ml of warm (37°C ) PBS, add
0.4 ml of warm (37°C ) 1× trypsin (0.05%) and EDTA
(0.02%) and incubate the plate at 37°C for exactly 5 min.
Do not digest the cells for a longer time. Then neutralize
the trypsin by adding 4 ml of 10% FBS supplemented media
(Defined keratinocyte serum-free media). Collect the cells
and deposit them into a 15-ml centrifuge tube and spin
down the cells at 170 × g for 5 min. Aspirate the supernatant
using a Pasteur pipette and suspend the pellet in 1 ml of
fresh medium (Defined keratinocyte serum-free media).
Count the cells in the same way as described in the
Subheading 3.1, step 15. Adjust cell number to the concen-
tration of 5 × 104 in 2 ml media per well for subculture. For
purposes other than serial subculture, the cell densities may
vary depending on the downstream experiments (e.g., col-
ony forming assay, cell cycle analysis, 3-D culture to make
in vitro skin equivalent, in vivo transplantation, hair follicle
reconstitution assay and RNA/Protein collection).
3.3.2. Colony Forming 1. Rinse the culture plate with 4 ml PBS for each well and fix the
Assay cells with 2 ml of 4% PFA for 5 min at room temperature. Add
4 ml PBS gently so as not to disturb the adherent cells and
then aspirate the medium. Repeat the wash step twice.
2. Stain the cultures by adding 2 ml Rhodanile blue solution.

Incubate at room temperature for 5 min.
3. Immerse the plate under running tap water for 2–3 min to
wash out the excessive dye. Although the colonies are unlikely
to come off the plate, keep the plate away from the direct flow
of the water. Air-dry with the plate upside down for a day.
4. Scan the wells and measure colony number and the surface
area using image processing software (e.g., Image-J or Adobe
Photoshop).
4. Notes
1. The scalp skin sample should include at least 50 follicles; oth-

erwise, FACS isolation and culture can be difficult. Start the
isolation within a few hours after sample collection. Do not
leave the sample in excess water (PBS or saline) for more
than 2 h.
2. Keep the hair follicle intact when cutting the scalp. The thickness
and width of the strip is a critical factor for successful Dispase
digestion.
3. The epidermis and follicle are supposed to separate from the
dermis without any resistance if the enzyme digestion is
sufficient. If not, in most cases, the scalp is too thick and needs
more trimming.
4. The basal cells of follicular keratinocytes are first digested by
trypsinization. It is sometimes hard to dissociate the suprabasal
cells completely. You had better not try to digest them com-
pletely, since a prolonged incubation will affect the viability of
basal cells.
5. Gently contact the tip of the pipette on the strainer mesh to
give the cell suspension the pressure to pass through.
6. The typical number of cells obtained is around 5–8 × 103 from
one follicle. For the following FACS experiments, you should
prepare a solution of 106 cells/ml. If the initial sample contains
100 hair follicles, digested cells can be suspended in 500 μl of
sample buffer.
7. Prepare human scalp sections both in the longitudinal and
transverse direction, which greatly helps to determine the
precise anatomical location of the cell population. It will be
necessary to optimize the embedding and fixation method if
you want to use unfamiliar antibodies. For CD200 staining,
scalp skin should be embedded in OCT compound and frozen
in liquid nitrogen. For keratin-15 and CD34 staining, either

frozen sections or paraffin embedded sections can be used. For
CD34 staining, the target antigen should be heat-retrieved by
boiling sections in Target Retrieval Solution (Dako, PH6) at
95°C for 20 min. Targets expressed in the basement mem-
brane are ideal to analyze because the outermost layer (basal
layer) of the epithelial component of the hair follicle is easier to
obtain following the enzymatic digestion protocol described in
Subheading 3.1.
8. For example, if you want to analyze and sort cells using
PE-conjugated anti-CD200 and APC-conjugated anti-CD34,
prepare sample tubes for “unstained,” “PE/APC-conjugated
isotype IgG,” “PE-CD200,” “APC-CD34,” and “PE-CD200
plus APC-CD34.”
9. The working dilution of the antibody should be optimized by
individual researchers. Usually it is 1:100 to 1:50.
10. Prolonged incubation times will result in excess staining with
7-AAD. Usually the proportion of dead cells is between 15 and
40% of the total cells. If more dead cells are observed, it is pos-
sible that any step of the isolation procedure has damaged the
cells or that 7-AAD staining has been excessive.
11. Less than 0.1% of the cell fraction can be regarded as negative.
12. It is important that the sorted cells reach directly the surface of
the medium to prevent cells from being lost on the wall of the
collection tube.
13. Keratinocytes, especially of differentiated populations, do
not grow and form colonies very well if plated too sparsely in
serum free media. To assess the colony forming ability, it is
desirable to prepare plates with different cell concentrations at
this point.
14. The culture duration may vary depending on the initial cell
concentration. Usually it is 1–2 weeks to grow to 70%
confluence. Passage cells before they get fully confluent to pre-
vent unwanted differentiation of keratinocytes. For example,
the CD200+CD34− population can be passaged every week up
to six passages.
Acknowledgments
We thank other members of the Laboratory of Plastic

Reconstructive Surgery in the University of Tokyo.
References
1. Cotsarelis G, Sun T-T, Lavker RM (1990) 6. Snippert HJ, Haegebarth A, Kasper M, Jaks V,
Label-retaining cells reside in the bulge of van Es JH, Barker N, van de Wetering M, van
pilosebaceous unit: implications for follicular den Born M, Begthel H, Vries RG, Stange
stem cells, hair cycle, and skin carcinogenesis. DE, Toftgård R, Clevers H (2010) Lgr6 marks
Cell 61:1329–1337 stem cells in the hair follicle that generate all
2. Oshima H, Rochat A, Kedzia C et al (2001) cell lineages of the skin. Science
Morphogenesis and renewal of hair follicles from 327(5971):1385–1389
adult multipotent stem cells. Cell 104:233–245 7. Rendl M, Lewis L, Fuchs E (2005) Molecular
3. Fuchs E (2009) The tortoise and the hair: dissection of mesenchymal-epithelial interac-
slow-cycling cells in the stem cell race. Cell tions in the hair follicle. PLoS Biol 3:e331
137:811–819 8. Ohyama M, Terunuma A, Tock CL et al
4. Nowak JA, Polak L, Pasolli HA, Fuchs E (2006) Characterization and isolation of stem
(2008) Hair follicle stem cells are specified and cell-enriched human hair follicle bulge cells.
function in early skin morphogenesis. Cell J Clin Invest 116:249–260
Stem Cell 3:33–43 9. Inoue K, Aoi N, Sato T, Yamauchi Y, Suga H,
5. Jaks V, Barker N, Kasper M, van Es JH, Eto H, Kato H, Araki J, Yoshimura K (2009)
Snippert HJ, Clevers H, Toftgård R (2008) Differential expression of stem-cell-associated
Lgr5 marks cycling, yet long-lived, hair follicle markers in human hair follicle epithelial cells.
stem cells. Nat Genet 40:1291–1299 Lab Invest 89:844–856
Chapter 27
Cocultivation of Human Oral Keratinocytes

and Human Osteoblast-Like Cells
Ricarda Glaum and Margit Wiedmann-Al-Ahmad
Abstract
Head and neck reconstruction transplants often require a bony structure but also tissue for the intraoral
lining. This is why oral keratinocytes and osteoblast-like cells are essential cell types for combined tissue
engineered transplants for defects in the field of craniomaxillofacial surgery. Therefore, we isolated oral
keratinocytes and osteoblast-like cells from human tissue samples and cocultivated both cell types on the
same carrier. Cell proliferation and morphological analysis showed that the contemporaneous cultivation
of human oral keratinocytes and human osteoblast-like cells is possible.
The successful in vitro cocultivation of hard and soft tissue derived cells on the same carrier will be an important
advancement for developing hard and soft tissue reconstruction therapies especially in the oral cavity.
Key words: Human oral keratinocytes , Human osteoblast-like cells , Cell cocultivation ,
Tissue engineering, Cell proliferation, Cell morphology
1. Introduction
In head and neck reconstruction, combined hard and soft tis-

sue reconstructive therapy is often necessary. Besides the bony
reconstruction which is addressed by different approaches, in
many cases an intraoral lining by oral mucosa is required (1–
5 ). Therefore, ideal transplants for combined hard and soft
tissue reconstruction by the means of tissue engineering
approaches require the combination of different cell types. We
investigated the cocultivation of human oral keratinocytes and
human osteoblast-like cells as common cell types present in
head and neck region. The cells were isolated from bone and
oral mucosa samples, obtained during oral surgical procedures,
by direct explant technique. In contrast to the commonly used
technique of culturing single cell suspensions, the tissue
423
424 R. Glaum and M. Wiedmann-Al-Ahmad
explant technique has been shown to be more effective for oral

keratinocytes as well as for osteoblast-like cells (6–9).
We have shown that the simultaneous cultivation of these two
cell types is possible. Cells can be cultivated in culture dishes as well
as on different membranes. There are many factors that influence
the result of the experiments including contamination of the patient’s
samples with oral microflora, contamination of the keratinocytes
with fibroblasts and as well as different cell cycles times of the kera-
tinocytes and osteoblasts. Furthermore, the proliferation analysis of
cocultivated cells within the same culture shows lower levels of via-
bility than one would expect compared to the separate cultivation of
each cell type (10). It is possible that the different cell types hinder
the proliferation of each other. A barrier between the cells has been
shown to enhance their viability (11). Different carriers vary in their
ability to support cocultivation, and changing culture conditions
(for example by using perfusion culture) does not necessarily result
in the same advantages for cocultivation when compared to cultiva-
tion of only one cell type (12, 13). In this chapter, we describe meth-
ods for the isolation and culture of oral keratinocytes and
osteoblast-like cells isolated from human tissue samples.
2. Materials
1. Dulbecco’s Phosphate Buffered Saline (PBS, PAA Laboratories,

Linz, Austria), store at room temperature (RT).
2.1. Medium for 1. Opti-MEM1 with Glutamax (Opti Minimal Essential Medium,
Cultivation of Gibco Invitrogen Life Technologies, Paisley, UK). Store at 4–8°C.
Osteoblast-Like Cells 2. Fetal calf serum (FCS, PAA Laboratories, Linz, Austria). Inactive
FCS at 56°C for 30 min and store in aliquots at −20°C.
3. 1 M N-2-hydroxyethylpiperazine-N¢-2-ethanesulfonic acid
(HEPES, Roth, Karlsruhe, Germany), store at room
temperature.
4. CaCl2 (Roth, Karlsruhe, Germany), mix the powder with high
grade water, and sterilize by autoclaving. Store at room tem-
perature for up to 1 year.
5. Penicillin/streptomycin solution (PAA Laboratories, Linz,
Austria). Store aliquots at −20°C.
6. Osteoblast-like cell medium: To prepare medium combine
components to final concentrations of:
Opti-MEM1 with Glutamax, 10% FCS, 2% HEPES
(0.02 M HEPES), 1,000 mg/L CaCl2, and 1% penicillin/
streptomycin. Store at 4–8°C for up to 3 weeks.
27 Cocultivation of Human Oral Keratinocytes and Human Osteoblast-Like Cells 425
2.2. Medium for 1. Dulbecco’s Modified Eagle Medium (DMEM) with

Cultivation of Oral 1 g L-Glucose (Bio Whittaker, Cambrex Bio Science Verviers,
Keratinocytes Verviers, Belgium), store at 4–8°C for up to 3 weeks.
2. Nutrient Mixture Ham’s F-12 with Glutamine (Gibco
Invitrogen Life Technologies, Paisley, UK), storage at 4–8°C
for 3 weeks.
3. L-glutamine (PAA Laboratories, Linz, Austria), thaw and store
4. The following reagents are dissolved in DMEM:F12 medium
and filter-sterilized using a 0.2 micron low protein binding
filter. Prepare aliquots sufficient for 1 L of medium and store at
−20°C.
(a) Epidermal growth factor (Sigma, Roedermark, Germany
(10 mg/L)).
(b) Insulin (from bovine pancreas, Sigma, Roedermark,
Germany) (5 mg/L).
(c) Human transferrin (apo-transferrin, Sigma, Roedermark,
Germany) (5 mg/L).
(d) Triiodotyronine (Sigma, Roedermark, Germany) (1.36 mg/L).
(e) Hydrocortisone (Sigma, Roedermark, Germany)
(0.2 mg/L).
(f) Cholera toxin (Sigma, Roedermark, Germany)
(0.0085 mg/L).
(g) Adenine (Sigma, Roedermark, Germany) (0.024 g/L).
(h) Spermine (Sigma, Roedermark, Germany) (10.5 mg/L).
5. To prepare oral keratinocyte medium combine:
750 ml DMEM and 250 ml nutrient mixture HAM’s F12, and
aseptically add 10% FCS, 2% HEPES (0.02 M HEPES), 1%
penicillin/streptomycin, 10 micrograms/L EGF, 5 mg/L
insulin, 5 mg/L transferrin, 0.0085 mg/L cholera toxin,
0.024 g/L adenosine,10.5 mg/L spermine
2.3. Cell Trypsinization 1. 0.5% trypsin in PBS (PAA Laboratories, Linz, Austria), Thaw
overnight in refrigerator, aliquot, and store at −20°C. Dilute in
PBS as needed for different cell types.
2.4. Medium for Cell 1. RPMI 1640 (Gibco Laboratories Life Technologies, NY,
Cocultivation USA), supplemented with10% FCS, 1% penicillin/streptomy-
cin solution, 5% HEPES (final concentration). Store at 4–8°C
for up to 3 weeks.
2.5. Cell Culture 1. 100 mm cell strainer, 50 ml tubes (Falcon, Heidelberg, Germany).
Supplies 2. Culture flasks; 25 cm3 and 75 cm3, (Greiner, Frickenhausen,
Germany).
3. Carriers for cocultivation. Polycarbonate membranes (Millipore,

Minucells and Minutissue, Bad Abbach, Germany, pore size 0.4 m m,
diameter 13 mm) as well as equine collagen membranes
(Tissue Foil E®, Baxter, Resorba, Nürnberg, Germany)
were used.
3. Methods
3.1. Isolation and Samples of bone and oral mucosa are harvested from patients (for
Cultivation of Cells example during oral and maxillofacial interventions) under sterile
conditions and placed into sterile 0.9% NaCl solution (see Note 1).
The different samples are processed separately. For isolating the
cells out of the samples, we use the direct explant technique (6–9).
3.2. Isolation 1. For the cultivation of osteoblast-like cells samples of bone are
and Cultivation rinsed with ethanol (70%) for 10–20 s, then rinsed three to four
of Osteoblast-Like times with PBS and minced into pieces of 1–2 mm diameter.
Cells 2. Culture flasks (25 cm3) are filled with 1.5 ml osteoblast-like
cell medium (see Subheading 2.1). Place three to five pieces of
sample into each culture flask.
3. Incubate explant cultures at 37°C in humidified atmosphere
with 5% CO2
4. After 1 week add 1 ml medium, another 1 ml 3–4 days later
and 3–4 days later 1.5 ml medium is added (~5 ml total
volume per flask).
5. The osteoblast-like cell medium is changed every 3–4 days
until the adherent cells cover the bottom of the culture flask.
6. For the first passage the confluent cell culture (primary cul-
ture) are detached from the culture flask by incubation with
0.5% trypsin in phosphate buffered saline for 8 min at 37°C.
7. The cell solution is filtered through a 100 mm cell strainer in
50 ml tube, centrifuged at 1120 × g, 12 min at 30°C and resus-
pended in 1 ml culture medium.
8. Cells are placed in a 75 cm3 culture flask with cocultivation
medium (see Subheading 2.3) and incubated.
9. After reaching confluence again the cells are detached from the
culture flask by incubation with 0.5% trypsin in phosphate
buffered saline for 8 min at 37°C, centrifuged, resuspended in
1 ml medium and cultivated (second passage).
10. Cells are used for experiments after second passage (see Note 4).
To verify the cell type, a portion of the cells should be cultivated,
e.g., for 1 week separately and used for cell characterization
(see Note 5).
3.3. Isolation and 1. For the cultivation of oral keratinocytes samples of mucosa are
Cultivation of Oral rinsed with ethanol (70%) for 10–20 s and afterwards three to
Keratinocytes four times with PBS, minced into pieces of 1–2 mm diameter.
For pure cultures of keratinocytes it is crucial to separate the
epithelial layer of the sample from the underlying connective
tissue using a sharp blade (see Note 3). Otherwise, there is a
high risk of contamination with fibroblasts.
2. Place three to five pieces of the sample into culture flasks
(75 cm3) filled with 4 ml of medium for cultivation of oral
keratinocytes (see Subheading 2.2; Note 2).
3. The explants are incubated at 37°C in humidified atmosphere
with 5% CO2.
4. The amount of medium is increased step by step to a total vol-
ume of 10 ml. After outgrowth of the cells, 30 ml of medium
is used for medium change.
5. Cells are used for experiments after the first passage (see Note 4).
To verify the cell type a part of the cells should be cultivated
(e.g., for 1 week) separately for cell characterization (see Note 5).
3.4. Cocultivation 1. Separately cultivated cells are trypsinized and seeded into cul-
ture dishes or on different carriers. Another medium is used
for cocultivation to fulfill the demands of both cell types (see
Subheading 2.3). It contains a higher amount of HEPES because
many experiments are carried out outside the incubator.
2. For experiments using membranes as carriers the cells are
placed on the opposite sides of the membranes (diameter of
the membranes: 13 mm). Therefore, one cell type is placed
onto the carrier on 1 day (12). After adherence of the cells the
other cell type is placed 1 day later onto the other side of the
carrier. We use 1 × 105 cells diluted in 50 ml medium of each
cell type (see Note 4). Polycarbonate membranes as well as
equine collagen membranes were used.
3.5. Cell Morphology In scanning electron microscopy as well as by light microscopy the cell
types can be differentiated by their typical morphology and growth-
pattern: Osteoblast-like cells are longish cells arranged like a draught
of fishes in confluent layers. Oral keratinocytes are compact, polygo-
nal, flat and show a cobblestone pattern when confluent (Fig. 1).
4. Notes
1. For all cell culture experiments it is essential to work under

sterile conditions (e.g., using sterile media and solutions, work-
ing under laminar flow).
Fig. 1. The typical morphology and growth pattern of the different cell types is visible in scanning electron micrographs of
osteoblast-like cells (a, b) and oral keratinocytes (c, d) on the opposite sides of a collagen membrane (Tissue Foil E®,
Baxter, Resorba, Nuremberg, Germany) after two weeks of cultivation (13). Magnification ×200 (a and c) and ×1,000
(b and d).
2. Especially the samples for oral keratinocytes should be taken

under sterile conditions, e.g., intraoral disinfection with betadine/
iodine before taking the biopsy.
3. For the isolation of keratinocytes from mucosa it is essential to
separate completely the epithelial layer of the tissue from the
underlying connective tissue to avoid contaminations with
fibroblasts.
4. For cocultivation of both cell types, it is essential to use the
second passage of human osteoblast-like cells and the first pas-
sage of keratinocytes because of differences in the growth
behavior.
5. Before using the cells for experiments the cell type should be
verified by cell characterization like testing for example osteo-
calcin, alkaline phosphatase, collagen type 1 for osteoblast-like
cells and cytokeratins type 13 and 19 for oral keratinocytes.
References
1. Wan DC, Nacamuli RP, Longaker MT (2006) 8. Jonsson KB, Frost A, Nilsson O et al (1999)
Craniofacial bone tissue engineering. Dent Three isolation techniques for primary culture
Clin North Am 50:175–190, vii of human osteoblast-like cells: a comparison.
2. Lendeckel S, Jodicke A, Christophis P et al Acta Orthop Scand 70:365–373
(2004) Autologous stem cells (adipose) and 9. Voegele TJ, Voegele-Kadletz M, Esposito V
fibrin glue used to treat widespread traumatic et al (2000) The effect of different isolation
calvarial defects: case report. J Craniomaxillofac techniques on human osteoblast-like cell
Surg 32:370–373 growth. Anticancer Res 20:3575–3581
3. Schleicher I, Parker A, Leavesley D et al 10. Glaum R (2008) Tissue Engineering von
(2005) Surface modification by complexes of Composite Grafts: Cokultivierung von
vitronectin and growth factors for serum-free Gingivakeratinozyten und Osteoblasten auf
culture of human osteoblasts. Tissue Eng 11: Polycarbonat- und Kollagenmembranen. VVB
1688–1698 Laufersweiler Verlag, Giessen
4. Schmelzeisen R, Schimming R, Sittinger M 11. Sendic M (2005) Wachstumsverhalten und
(2003) Making bone: implant insertion into gegenseitige Beeinflussung von Gingivakeratino-
tissue-engineered bone for maxillary sinus zyten und Osteoblasten auf Kollagen.
floor augmentation – a preliminary report. J Inaugural-Dissertation zur Erlangung des
Craniomaxillofac Surg 31:34–39 Zahnmedizinischen Doktorgrades der
5. Terheyden H, Warnke P, Dunsche A et al Medizinischen Fakultät der Albert-Ludwigs-
(2001) Mandibular reconstruction with pre- Universität Freiburg i Br
fabricated vascularized bone grafts using rec- 12. Glaum R, Wiedmann-Al-Ahmad M, Huebner
combinant human osteogenic protein-1: an U et al (2010) Tissue engineering of composite
experimental study in miniature pigs. Part II: grafts: cocultivation of human oral keratinocytes
transplantation. Int J Oral Maxillofac Surg and human osteoblast-like cells on laminin-
30:469–478 coated polycarbonate membranes and equine
6. Lauer G, Otten JE (1997) Cultivation of gin- collagen membranes under different culture
gival keratinocytes on permeable membranes: conditions. J Biomed Mater Res A 93:704–715
simulation of the function of mouth cavity epi- 13. Otto F (2004) Vergleichende Untersuchungen
thelium. Mund Kiefer Gesichtschir 1:35–38 des Wachstums humaner Osteoblasten nach
7. Kedjarune U, Pongprerachok S, Arpornmaekl- Kultivierung im Brutschrank und in der
ong P et al (1997) Culturing primary human Perfusionskammer. Inaugural-Dissertation zur
gingival epithelial cells: comparison of two Erlangung des Medizinischen Doktorgrades
isolation techniques. J Craniomaxillofac Surg der Medizinischen Fakultät der Albert-
29:224–231 Ludwigs-Universität Freiburg i Br, Freiburg
Chapter 28
Isolation and Culture of Skeletal Muscle Myofibers

as a Means to Analyze Satellite Cells
Paul Keire, Andrew Shearer, Gabi Shefer, and Zipora Yablonka-Reuveni
Abstract
Multinucleated myofibers are the functional contractile units of skeletal muscle. In adult muscle, mononu-
clear satellite cells, located between the basal lamina and the plasmalemma of the myofiber, are the primary
myogenic stem cells. This chapter describes protocols for isolation, culturing, and immunostaining of
myofibers from mouse skeletal muscle. Myofibers are isolated intact and retain their associated satellite
cells. The first protocol discusses myofiber isolation from the flexor digitorum brevis (FDB) muscle. These
short myofibers are cultured in dishes coated with PureCol collagen (formerly known as Vitrogen) using
a serum replacement medium. Employing such culture conditions, satellite cells remain associated with the
myofibers, undergoing proliferation and differentiation on the myofiber surface. The second protocol
discusses the isolation of longer myofibers from the extensor digitorum longus (EDL) muscle. Different
from the FDB preparation, where multiple myofibers are processed together, the longer EDL myofibers
are typically processed and cultured individually in dishes coated with Matrigel using a growth factor rich
medium. Under these conditions, satellite cells initially remain associated with the parent myofiber and
later migrate away, giving rise to proliferating and differentiating progeny. Myofibers from other types of
muscles, such as diaphragm, masseter, and extraocular muscles can also be isolated and analyzed using
protocols described herein. Overall, cultures of isolated myofibers provide essential tools for studying the
interplay between the parent myofiber and its associated satellite cells. The current chapter provides back-
ground, procedural, and reagent updates, and step-by-step images of FDB and EDL muscle isolations, not
included in our 2005 publication in this series.
Key words: Skeletal muscle, Satellite cells, Stem cells, Collagen, Matrigel, Myofiber isolation,
Flexor digitorum brevis, Extensor digitorum longus, Diaphragm, Masseter, Extraocular, Mouse,
Immunostaining, Pax7
1. Introduction
Myofibers are the functional contractile units of skeletal muscle.

While myofibers are established during embryogenesis by fusion of
myoblasts into myotubes, processes involved in their growth and
431
432 P. Keire et al.
Fig. 1. A schematic (a) and EM micrograph (b) of satellite cell location. The myofiber basement and plasma membranes
have been routinely detected by immunostaining with antibodies against laminin and dystrophin, respectively. In panel (a),
myofiber nuclei depicted at the myofiber periphery represent the state of healthy adult myofibers, while regenerating
muscles display centralized myofiber nuclei (not shown). In panel (b), black arrows depict the basal lamina, white arrows
depict apposing satellite cell and myofiber membranes; note the sarcomeric organization within the myofiber.
repair continue throughout life. These processes are supported by

myogenic progenitors known as satellite cells that are located
between the basal lamina and the plasmalemma of the myofiber
(1, 2); for a schematic and electron microscope image see Fig. 1.
In a growing muscle at least some of the satellite cells are proliferat-
ing, and contribute myonuclei to the enlarging myofibers, whereas
in intact adult muscles most satellite cells are quiescent. In response
to a variety of conditions, ranging from increased muscle utiliza-
tion to muscle injury, satellite cells can enter the cell cycle, produc-
ing progeny that fuse into existing myofibers, or form new myofibers
(3, 4). Satellite cells are considered stem cells because in addition
to giving rise to progeny needed for myofiber repair, they can self-
renew (5, 6). It is not known, however, if all satellite cells are iden-
tical with regard to their amplification and renewal potential (6, 7).
Insights into the cascade of cellular and molecular events control-
ling satellite cell myogenesis are essential for understanding the
mechanisms controlling muscle maintenance as well as for devel-
oping strategies to enhance muscle repair after trauma or in myo-
pathic diseases (8–11).
Satellite cells were initially identified using electron microscopy
by their location under the myofiber basal lamina (1, 12, 13)
(Fig. 1). More recently it has become possible to monitor satellite
cells by light microscopy based on the expression of a range of
markers that can be detected by immunostaining (14). In particu-
lar, the specific expression of the paired box transcription factor
Pax7 by satellite cells and the availability of an excellent antibody
for immunodetection of this protein provide a uniform means to
identify satellite cells in their native position in a range of species,
such as mouse, rat, and chicken (15–18). In humans, however,
Pax7 expression may not necessarily identify all satellite cells (14).
Additionally, genetically manipulated reporter mice permit direct
28 Mouse Myofiber Cultures 433
detection of satellite cells based on specific expression of a

fluorescent tag or of beta-galactosidase (7, 16, 19, 20). We dem-
onstrated that transgenic expression of GFP under the control of
nestin regulatory elements (NES-GFP) allows detection of satellite
cells in freshly isolated myofibers. NES-GFP mice also facilitate the
isolation of satellite cells using fluorescent-activated cell sorting
(FACS) and subsequent studies of purified populations (7, 16).
Satellite cell progeny can be distinguished from their quiescent
progenitors based on distinctive gene expression patterns (2, 4, 7).
In particular, the myogenic regulatory factors MyoD and myo-
genin have been used extensively to monitor progeny of satellite
cells (21–24). Proliferating progeny (myoblasts) continue to
express Pax7, but distinctly from their quiescent progenitors, also
express MyoD. A decline in Pax7 along with the induction of the
muscle-specific transcription factor myogenin mark myoblasts that
have entered into the differentiation phase and subsequently fuse
into myotubes. Reemergence of cells that express Pax7, but not
MyoD (reserve cells), define a self-renewing population of satellite
cells (2, 5–7, 22–24).
Two main cell culture approaches have been employed in the
study of satellite cells: (a) primary myogenic cultures prepared
from mononucleated cells dissociated from whole muscle; and (b)
cultures of isolated myofibers (also referred to below as “fibers”)
where the satellite cells remain in their in situ position underneath
the myofiber basal lamina. Protocols for obtaining primary myo-
genic cultures involve releasing satellite cells from their niche. Steps
of mincing, enzymatic digestion and repetitive triturations of the
muscle are required for breaking down both the connective tissue
network and the myofibers in order to release the satellite cells
from the muscle bulk. These steps are followed by procedures for
removing tissue debris and reducing the contribution of non-myo-
genic cells typically present in primary isolates of myogenic cells (6,
16, 22, 25–29). In contrast, protocols for isolating individual mus-
cle fibers result in the release of intact myofibers that retain satellite
cells in their native position underneath the basal lamina (16, 21–
23, 26). These protocols allow the study of satellite cells and their
progeny in their in situ position on the myofiber, and after they
migrate from the parent myofiber.
This chapter describes two protocols used in our laboratory for
isolation and culture of single myofibers from mouse skeletal mus-
cles (22, 30). One protocol, first introduced by Bekoff and Betz
(31) and further developed by Bischoff (32, 33), has been adopted
by us for studies of satellite cells in isolated myofibers from both
rats (21, 26, 34) and mice (22, 35, 36). In this case, single myofibers
are isolated from the flexor digitorum brevis (FDB) muscle of the
hind feet. Because these FDB myofibers are short and do not get
tangled, typically multiple myofibers are processed and cultured
together. A second approach, introduced by Rosenblatt and
434 P. Keire et al.
colleagues (37, 38), allows isolation of longer myofibers from a

variety of muscles, including extensor digitorum longus (EDL),
tibialis anterior (TA) and soleus (5, 20, 37, 38), and has been used
extensively by our laboratory as well (2, 7, 16, 22, 39). These lon-
ger myofibers can get tangled, and therefore, when working with
muscles such as the EDL, the released myofibers are typically pro-
cessed and cultured individually. The EDL single myofiber isola-
tion procedure described here has also been adapted in our
laboratory for diaphragm, masseter and extraocular muscles. Both
the short and long myofibers are cultured in dishes that have been
pre-coated with commercially available matrices that facilitate rapid
and firm adherence of the myofibers to the dish surface, as detailed
below. It is worth noting, however, that in addition to the matrix-
attached myofiber cultures described herein, other laboratories
have introduced approaches where isolated myofibers are cultured
in suspension (23, 40).
The current chapter contains background, procedural and
reagent updates for FDB and EDL myofiber isolation. We also
include figures depicting step-by-step “real live” images of respec-
tive muscle dissection and harvesting, not illustrated in our previ-
ous 2005 report on myofiber isolation and culture (30). In addition,
new to this chapter is a description of myofiber isolation from
diaphragm, masseter and extraocular muscles. Table 1 compares
the two approaches of myofiber isolation from FDB and EDL
muscles and the specific use of each procedure, while representa-
tive micrographs of FDB and EDL myofiber cultures are shown in
Figs. 2 and 3, respectively. Figures 4 and 5 are presented later in
Subheading 3 to assist the investigator in the dissection of the FDB
and the EDL muscles. Protocols for immunocytochemical analysis
of satellite cells in cultures of FDB and EDL myofibers and of
freshly isolated myofibers are also included in the chapter.
Altogether, to achieve the isolation of intact myofibers, it is of
utmost importance to delicately manipulate the muscle of interest.
Following the procedures and protocol notes detailed in this chapter,
investigators can successfully isolate, culture and analyze myofibers
from well studied EDL and FDB fibers, and also use these proto-
cols as a framework for the study of other muscles.
2. Materials
2.1. General Comments 1. As a general rule, only sterile materials and supplies are to be
used. All solutions, unless otherwise noted, are sterilized by
filtering through 0.22-mm filters, all glassware and dissection
tools are sterilized by autoclaving, and all cell culture proce-
dures are performed using sterile techniques.
Table 1
Characteristics of myofiber cultures from FDB and EDL muscles of adult mice
Donor muscle Flexor digitorum brevis (FDB) Extensor digitorum longus (EDL)
Relative myofiber Short Long

length
Number of fibers ~30–50 1
per culture
Typical tissue 35-mm dish 24-well multiwell dish
culture dish
Dish coating PureCol. Thick, gel-like layer of native Thin coating of diluted, growth factor
collagen type I prepared from bovine reduced Matrigel. Matrigel is a
hide (Advanced BioMatrix; formerly basement membrane preparation
known as Vitrogen; see Note 1) isolated from a mouse tumor (BD
Biosciences; see Note 2)
Medium Dulbecco’s modified Eagle’s medium DMEM-based, serum-rich/mitogen-
(DMEM)-based, mitogen-depleted rich; medium can be modified to a
serum; specific exogenous growth serum-poor/mitogen-poor to allow
factors are added to study their effect analysis of satellite cell activation
on satellite cell activation, prolifera- (7, 16, 22, 37, 38, 42)
tion, and differentiation (21, 22, 35,
36, 41)
Satellite cell Satellite cells remain at the surface of Satellite cells emigrate from the parent
profile after the parent myofiber as they prolifer- myofiber and undergo multiple
culturing ate and differentiate. Satellite cells rounds of proliferation, giving rise to
undergo a limited number of an elaborate network of myotubes,
proliferative cycles and rapidly resembling regular primary cultures
differentiate without fusing with the of cells dissociated from whole muscle
parent myofiber
Summary Cultures may model in vivo behavior of Cultures may model events after muscle
satellite cells in intact fibers during trauma where new myofibers are
growth and routine muscle utilization formed
Cultures typically are maintained Cultures typically are maintained
short-term and can be employed for long-term and employed in studies of
studying satellite cell activation and myogenic cells and progeny of
entry into the cell cycle. Steps of satellite cells that emigrate from the
proliferation and differentiation are myofiber to the myofiber surround-
highly synchronous (21, 22, 35, 36) ings (7, 16, 22, 37)
Cultures can be further used to study Cultures can also be used for analysis of
cells emigrating from the myofibers as molecular and cellular events associated
described for the EDL fiber cultures with the first round of satellite cell
Satellite cells can be monitored in proliferation, as in FDB cultures (42)
freshly isolated (Time 0) myofibers Satellite cells can be monitored in
freshly isolated (Time 0) myofibers
(7, 22)
436 P. Keire et al.
Fig. 2. Parallel phase and immunofluorescent micrographs of an isolated FDB myofiber with associated satellite cells
undergoing myogenesis. Myofibers were isolated from a 3-month-old mouse and cultured in 35-mm tissue culture dishes
coated with isotonic Vitrogen collagen in solution (now known as PureCol). Cultures were maintained for 4 days in basal
medium containing fibroblast growth factor 2 (FGF2, 2 ng/mL) and fixed with methanol as described in Subheading 3.3.1.
(a, b) Phase and DAPI stained images (both myofiber nuclei and satellite cell nuclei are labeled with DAPI). (c, d) Myofiber
culture reacted by double immunofluorescence with a monoclonal antibody against myogenin (identifies the nuclei of
differentiated myogenic cells) and a polyclonal antibody immunostaining against ERK1/ERK2 mitogen activated protein
kinases (MAPK) (identifies the cytoplasm of all fiber-associated cells). Reactivity with the monoclonal and polyclonal anti-
bodies was traced with fluorescein- and rhodamine-labeled secondary antibodies, respectively. Arrows in parallel panels
point to the location of the same cell. Additional immunopositive cells present on the myofiber are not shown, as not all
positive nuclei or cells on the fibers are in the same focal plane. All micrographs were taken at 400× magnification.
Additional details regarding the source of the antibodies and the rationale of using these antibodies are provided in our
previous publications (22, 35, 36, 41).
2. Cultures are maintained at 37°C and 5% CO2 in a humidified

tissue culture incubator.
3. All culture media are stored at 4°C and used within 3 weeks
from preparation.
4. Before starting the isolation procedure, tissue culture medium
is pre-warmed to 37°C and then held at room temperature
throughout the procedures. Before transferring solutions/
media into the tissue culture hood, spray the glass/plastic con-
tainers with 70% ethanol.
Fig. 3. Phase micrographs of EDL myofibers depicting the temporal development of myogenic cultures from cells emanat-
ing from individual myofibers. Myofibers were isolated from 3 month-old mice and cultured individually in 24-well multi-
well tissue culture dishes coated with Matrigel. Cultures were maintained in serum-rich/mitogen-rich growth medium and
fixed with paraformaldehyde, as described in Subheading 3.3.2. Satellite cells begin to emigrate from the myofiber within
the first day in culture and continue to emigrate during subsequent days. Progeny of satellite cells that have emigrated
from the myofibers proliferate, differentiate and fuse into myotubes, establishing a dense myogenic culture. (a) Satellite
cells remained attached to the muscle fiber during the first hours after culturing. (b) Nineteen hours after culturing, two to
three cells detached from the fiber but remained in close proximity to the fiber. (c) Four days following culturing more cells
are seen in the vicinity of the myofibers (at least four cells are visible). (d) By day 7, progeny of satellite cells that emigrated
from the myofiber have established a culture containing mostly proliferating myoblasts and some myotubes. Micrographs
in panels (a–c) were taken at 400× magnification to show details of the few cells that emigrated from the myofiber, while
the micrograph in panel (d) was taken at 100× magnification to show the establishment of a dense myogenic culture. See
our published study for additional details about growth of satellite cell progeny in long-term EDL myofiber cultures (22).
5. The quantities of glassware, media and reagents as well as the

time intervals for enzymatic digestion described in this chapter
are appropriate for the isolation of myofibers from one adult
mouse of the age and strain detailed below (see Note 3).
2.2. General Equipment The following facilities are required for the cultures described in
this chapter:
1. Standard humidified tissue culture incubator (37°C, 5% CO2
in air).
2. Tissue-culture hood.
438 P. Keire et al.
Fig. 4. Dissection of FDB muscle from the rear foot of adult mouse. (a) Rear foot before dissection. (b) Cutting of “T” toward
the ankle, left to right; arrowheads identify the circumferential cut at the ankle and arrow shows the direction of cutting.
(c) Peeling the skin back from the ankle exposing the muscles and tendons. (d) Digit tendons of the FDB exposed on the
sole of the foot. (e) Cutting the connective tissue under the FDB toward the heal of the foot. (f) Freeing the FDB from
the underlying connective tissue. (g) Cutting the FDB at the heal origin; arrow indicates direction of cutting. (h) Preparing
the release of the FDB from its tendon insertion points at the digits.
Fig. 5 Dissection of EDL muscle from the hindlimb of adult mouse. (a) Anterior lower hindlimb with skin removed. (b) Facia
covering the anterior lower hindlimb muscles is removed to allow access to tendons. (c) The four foot tendon insertion points
of the EDL are isolated and cut. (d) The common tendon of the EDL is carefully exposed and isolated at the ankle. (e) Once
isolated and foot insertions are cut, the EDL tendons are pulled proximally up from the foot; arrows indicate the direction of
pulling. The tendons should easily slide underneath the connective tissue sheath at the ankle up from the foot. If the tendons
do not easily slide out, then reexamine the foot tendons to ensure that they have been cut. (f) Origin of the EDL is exposed then
cut at the lateral surface of the tibia condyle head. (g) Grasping only the EDL tendon (do not grasp the muscle as it can easily
be damaged), carefully pull distally toward the toes to remove the EDL muscle; arrow indicates the direction of pulling. (h) The
EDL should slide underneath the tibialis anterior muscle and should pull out easily. It is important to pull gently and there
should be little resistance; if the muscle does not slide out easily, one or both tendons at the muscle origin may still be attached
to the bone. In this case identify the attached tendon and cut it.
440 P. Keire et al.
3. Stereo dissecting microscope with transmitted light base

(microscope is either placed inside a tissue culture hood or in
an isolation box/clean area).
4. Bunsen or alcohol burner.
5. Water bath (37°C).
6. pH meter and pH paper strips (e.g., EMD, colorpHast sτrips).
7. Inverted phase contrast microscope for monitoring cell culture.
2.3. Surgical Tools 1. Straight operating scissors: V. Mueller, fine-tipped, Sharp/

Sharp stainless steel, 165-mm (6½″), for delicate cutting and
fine incisions.
2. Dissecting scissors: stainless steel, 140-mm (5½″) length; both
blades blunt, to protect the surrounding tissue from any
unwanted nicks.
3. Dressing forceps: V. Mueller, serrated, stainless steel, rounded
points, and 140-mm (5½″) length.
4. Two, very fine point forceps: extra-fine tips, smooth spring
action, stainless steel. Straight, 110-mm (4½″) length.
5. Microscissors, Vannas type: 8-cm long, straight 5-mm blades,
0.1-mm tips.
6. Scalpel handle and blades: size 3 handle for blade numbers
10–15 and sterile blades (#10).
7. Two straight, 5″ hemostatic forceps.
8. Dissecting board with tissue pins.
2.4. Animals C57BL/6 mice, 2–5 months old, maintained according to institu-
tional animal care regulations. Aged mice (up to 33 months old)
and other mouse strains have also been used in our studies follow-
ing the same myofiber isolation procedures (e.g., (7, 22); see Note
3). When harvesting muscles for fiber preparation, we prefer cervi-
cal dislocation for euthanizing mice as this method is more rapid
and minimizes muscle stiffening that occurs after death. Muscle
stiffening can make the isolation of single fibers more difficult and
decrease overall fiber yield.
2.5. Plastic 1. Standard 9″ glass Pasteur pipettes; fire polish the ends to avoid
and Glassware for damage to myofibers, which are transferred using these pipettes.
Myofiber Isolation As noted above in item 1 in Subheading 2.1, all Pasteur pipettes
and Culture are sterilized by autoclaving before use.
2.5.1. FDB Myofiber
2. Standard 5″ glass Pasteur pipettes. Prepare three gradually nar-
Isolation and Culture
rower-bore pipettes from standard 5″ Pasteur pipettes. Use a
file or a diamond knife to prepare a set of pipettes with bore
diameter of approximately 3, 2, and 1 mm. Shake the pipette
to remove any glass fragments and fire polish the sharp ends.
These pipettes are used to triturate the digested muscle in
order to release single myofibers.
3. Syringe filters, 0.22-mm PVDF low protein binding filters

(Millipore is recommended).
4. 3- or 10-cc disposable plastic syringes.
5. Bottle top filters, 0.22 mm.
6. Polypropylene conical centrifuge tubes, sterile, 15 and 50 mL.
7. Three glass Corex tubes, 15 mL (Sorvall centrifuge tubes; or
alternatively 15 mL bicarbonate Sorvall tubes).
8. Wide-bore 100-ml micropipette tips. Trim 100-ml micropipette
tips approximately 3 mm from the end. Use of these trimmed
micropipettes minimizes myofiber shearing when transferring
or dispensing FDB myofibers.
9. Tissue culture dishes, 35-mm.
10. Two L-shape bent pipette spreaders prepared from standard 9″
Pasteur pipettes. Use flame to first seal the distal end, then
flame about 2 cm from the sealed end until the pipette starts to
bend. The bent pipettes are used to spread the coating solution
on the tissue culture dishes; the length of the bent end is
designed for working with the 35-mm culture dishes for FDB
myofiber cultures. Spreaders should be prepared in advance
and allowed to cool before use.
2.5.2. EDL Myofiber 1. Standard 9″ and 5″ sterile Pasteur pipettes and syringe filters
Isolation and Culture listed and treated as described in items 1 – 4 in
Subheading 2.5.1.
2. Plastic petri dishes 60 × 15 mm and 100 × 15 mm (for muscle
and myofiber rinsing), 35-mm tissue culture dishes.
3. Twenty-four well, multiwell tissue culture dishes (see Note 4).
4. Bottle filters and conical tubes and as in items 5 and 6 in
Subheading 2.5.1.
5. 1-mL serological glass pipettes (used for Matrigel aliquoting,
see Note 2).
6. Cryogenic vials sealed with O-rings (for storing Matrigel ali-
quotes, see Note 2).
2.6. Media, Enzymes, 1. DMEM/high glucose (Dulbecco’s Modified Eagle Medium with
and Cell Culture 4,500 mg/L glucose, 4.0 mM L-glutamine, and 110 mg/L
Reagents sodium pyruvate; readily available from multiple vendors),
supplemented with 100 U/mL penicillin and 100 μg/mL
2.6.1. FDB Myofiber
streptomycin.
Isolation and Culture
2. Horse serum (HS); standard, not heat inactivated (see Note
5). Original bottles are stored at −80°C; once thawed and ali-
quoted, store at −20°C.
3. Controlled Process Serum Replacement (CPSR, Sigma-
Aldrich, stored at −80°C; once thawed and aliquoted, store at
−20°C; see Note 6 for product composition and availability).
442 P. Keire et al.
Alternative serum replacement products (e.g., Sigma-Aldrich,

cat. no. S0638 or S9388 (43)) can also be used depending on
experimental requirements (see Note 6).
4. FDB myofiber culture medium: DMEM/high glucose (sup-
plemented with antibiotics), 20% Controlled Process Serum
Replacement, and 1% HS.
5. PureCol collagen (Advanced BioMatrix, cat. no. 5005-B; this
collagen in solution was formerly known as Vitrogen when
sold by Cohesion Technologies) for coating 35-mm tissue cul-
ture dishes (see Note 1).
6. 7× DMEM made from powder DMEM (Sigma-Aldrich, cat.
no. D5648); used to prepare isotonic PureCol collagen (see
Note 1).
7. Collagenase (type I, Sigma-Aldrich, cat. no. C-0130). The
final working solution is prepared as described in step 3 in
Subheading 3.1.1.
8. 100 mL of DMEM containing 10% HS. HS is freshly filtered
on the day of use through a 0.22-mm filter. This DMEM-10%
HS medium is used for FDB myofiber purification as detailed
in Subheading 3.1.5. Also, all Pasteur pipettes and micropi-
pette tips are pre-flushed with this DMEM-10% HS medium
to prevent sticking of myofibers during manipulation.
2.6.2. EDL Myofiber 1. DMEM and horse serum (HS) as listed and prepared in items
Isolation and Culture 1 and 2 in Subheading 2.6.1.
2. Fetal bovine serum (FBS; standard, not heat inactivated; see
Note 7). Original bottles are stored at −80°C; once thawed
and aliquoted, stored at −20°C.
3. Chicken embryo extract (CEE; see Notes 8 and 9); stored at
−80°C for long term or −20°C when aliquoted.
4. EDL myofiber culture medium: DMEM/high glucose (same
formulation as in item 1 in Subheading 2.6.1 for FDB fibers
and supplemented with 100 U/mL penicillin and 100 mg/mL
streptomycin), 20% fetal bovine serum, 10% HS, and 1% CEE.
5. Matrigel (BD Biosciences; see Note 2) for coating 24-well,
multiwell dishes. We typically dispense Matrigel into aliquots
of 100–200 ml and freeze back at −20°C. See Note 2 for all
handling details.
6. Collagenase, as listed in item 7 in Subheading 2.6.1, working solu-
tion is prepared as in step 1 of Subsection 3.2.1, in Subheading
“Preparation of the Digesting Enzyme Solution and Post-digestion
Rinse Plates”.
7. HS, 10 ml, freshly filtered on day of use with 22-mm syringe
filter. Used to coat petri dishes and pre-flush Pasteur pipettes
to minimize potential sticking of myofibers during isolation
procedure.
2.7. Reagents and 1. Pre-fixation rinse solution: DMEM as in item 1 in

Solutions for Fixing Subheading 2.6.1.
and Immunostaining 2. Fixative: ice-cold 100% methanol (see Note 10).
Myofiber Cultures
3. Rinse solution: Tris-buffered saline (TBS); 0.05 M Tris, 0.15 M
2.7.1. FDB Myofiber NaCl, pH 7.4 (see Note 11).
Cultures 4. Detergent: Tween 20.
5. Detergent solution: TBS containing 0.05% Tween 20 (TBS-
TW20).
6. Blocking reagent: Normal goat serum (standard, e.g.,
Invitrogen, cat. no. 16210-072). Can be stored at −80°C;
once thawed and aliquoted, store at −20°C.
7. Blocking Solution: TBS containing 1% normal goat serum
(TBS-NGS).
8. DAPI solution (4¢,6-diamidino-2-phenylindole, dihydrochlo-
ride); stock concentration 10 mg/mL and a working concen-
tration of 1 mg/mL diluted in TBS-NGS prior to use (see
Note 12).
9. Mounting medium: Vectashield (Vector Laboratories, cat. no.
H-1000); store at 4°C.
10. Cover glass, 22 mm2.
2.7.2. EDL Myofiber 1. Fixative: 4% paraformaldehyde in a sodium phosphate buffer con-

Cultures taining 0.03 M sucrose (see Notes 13 and 14 for specific buffer
details and preparation). Store at 4°C, pre-warm to room tempera-
ture before use. To maintain quality and effectiveness of fixative,
pre-warm only the volume that is required for immediate use.
2. Rinse solution: TBS as in item 3 in Subheading 2.7.1.
3. Detergents: Triton X-100 and Tween 20.
4. Detergent solution: TBS containing 0.5% Triton X-100 (TBS-
TRX100); TBS-TW20 as in item 5 in Subheading 2.7.1.
5. Blocking reagent (NGS) and solution (TBS-NGS), same as
items 6 and 7 in Subheading 2.7.1.
6. DAPI working solution and Vectashield; same as items 8 and 9
in Subheading 2.7.1.
7. Sterile glycerol solution: 25% glycerol in TBS, store at 4°C.
3. Methods
3.1. Isolation of Single The information in this introductory section is provided to assist in
Myofibers from the the identification of the flexor digitorum brevis (FDB) muscles.
Flexor Digitorum The FDB is a superficial, multipennate, broad and thin muscle of
Brevis Muscle the foot and paw (33, 44); it arises from the tendon of the plantaris
444 P. Keire et al.
as three slender muscles converging into long tendons. At the base

of the first phalanx it divides into two, passes around the tendon of
the flexor hallucis longus obliquely across the dorsum of the foot,
and ends as the tendons insert into the second phalanx of the 2nd
through the 5th digits. As the FDB contracts, digits 2–5 are flexed.
For additional details about the anatomy of the FDB muscle (see
Note 15).
For uniformity, we typically use only the hindlimb muscles in
our studies. Figure 4 depicts “real-live” images of steps in FDB
muscle isolation that emphasize the location of the specific tendons
that are handled during the process. It is of utmost importance to
delicately manipulate the muscle of interest only at the tendons
during its excision and further processing.
3.1.1. Initial Steps Prior 1. Add 3 mL of DMEM to six 35-mm tissue culture dishes and
to Harvesting the Muscle place the dishes in the tissue culture incubator until muscle dis-
and Preparation of section begins.
Digestive Enzyme 2. Add 3 mL of DMEM containing 10% HS to three 35-mm tis-
sue culture dishes and place them in the tissue culture incuba-
tor until needed for the isolated single myofibers.
3. Add 6 mg of collagenase type I to 3 mL of DMEM in order to
prepare 0.2% (w/v) collagenase type I solution. Use a 0.22-mm
syringe filter attached to a 3- or 10-cc syringe to filter the col-
lagenase solution into a 35-mm tissue culture dish (see Note 16).
We prepare this solution fresh for each experiment.
3.1.2. Dissection of FDB 1. Euthanize one mouse according to institute regulations.

Muscle (Fig. 4) 2. Spray the hind foot (Fig. 4a) lightly with 70% ethanol.
3. All the following steps, until the muscle is dissected out, are
carried in an enclosure dedicated for this procedure in order to
limit contamination.
4. Secure the mouse, lying on its back, to the dissecting board by
pinning down the forelimb diagonally across from the limb
being dissected.
5. Use a scalpel to carefully cut the skin circumferentially just
above the ankle joint, so that the skin above and below the cut
site are completely separated (after this circular cut, the skin
below resembles a sock).
6. Using scissors, cut the skin in a straight line along the center
of the ventral part of the foot almost all the way to the digits
(the cut as viewed from the front of the foot should resemble
a “T”) (Fig. 4b).
7. Using a hemostat, grasp one of the upper corners of the cut
tissue (at the junction of the circular and longitudinal cuts) and
reflect the skin away from the foot (Fig. 4c).
8. Hold the scalpel with its blade parallel to the longitudinal axis
of the partially exposed muscle and carefully separate the skin
from the connective tissue. Be especially careful not to cut into

the muscle tissue at the back of the leg, as the FDB is the most
superficial muscle of the back of the foot.
9. Clamp a second hemostat to the other corner of the cut tissue
and repeat step 8.
10. When the skin is completely cut away from the foot, the FDB
should be exposed all the way to the tendons reaching the dig-
its (Fig. 4d).
11. Turn the mouse over so that it lies on its stomach, and identify
the FDB. During the next steps of the dissection, be careful
not to injure the small medial plantar artery that supplies blood
to the FDB to limit blood cell contamination of the myofiber
preparation. This artery passes along the medial part of the
sole of the foot and branches into the digits.
12. Carefully run the tip of the scalpel along each side of the FDB
to dissect the connective tissue holding the muscle in place
(Fig. 4e).
13. When the FDB is separated from the surrounding muscles,
carefully lift the FDB by inserting one arm of your smooth for-
ceps or a fine blunt probe underneath the FDB so that the flat
side of the scalpel may be inserted horizontally underneath it.
14. With the blade of the scalpel underneath the muscle, running
horizontal and parallel to the muscle, cut away the underlying
connective tissue (Fig. 4f). It is best to cut towards the heel and
only lift that portion of the muscle directly over the scalpel.
15. Cut underneath the tendon to separate the muscle and a large
portion of its tendon from the heel bone (Fig. 4g).
16. Grasp the freed tendon as far as possible from the muscle tissue
with a hemostat and gently lift the FDB away from the leg.
While lifting the FDB, use the scalpel, running parallel to the
muscle, to cut through the connective tissue while holding the
foot down.
17. Continue cutting through the connective tissue until the ten-
dons that connect the FDB muscle to the digits have been
exposed (Fig. 4h). When about half the length of the three
tendons has been exposed, cut the tendons and release the
entire muscle from the leg. The fourth small lateral tendon
(attached to the 5th digit) and its attached myofibers can be
trimmed off.
18. Retrieve from the incubator three 35-mm tissue culture dishes
containing DMEM and place them close to the dissection area.
19. Place the harvested FDB in one of the 35-mm tissue culture
dishes.
20. For harvesting the FDB from the other hind foot repeat steps
4–18, and place the muscle in a second 35-mm tissue culture
dish.
446 P. Keire et al.
21. Place the 35-mm tissue culture dishes, one at a time, under the
stereo dissecting microscope.
22. Use fine point forceps to pull the connective tissue perpendic-
ular to the line of the muscle and use fine dissection scissors to
cut it off.
23. Once the muscle is clean, shorten the tendons but do not cut
all of them off.
24. Use a wide-bore Pasteur pipette to transfer the cleaned muscle
to another 35-mm tissue culture dish containing DMEM.
25. Repeat steps 21–24 to clean the second FDB muscle.
3.1.3. Enzymatic Digestion 1. Working in the tissue culture hood transfer the two cleaned
FDB muscles to a 35-mm tissue culture dish containing 1.5 ml
of the 0.2% collagenase I solution.
2. Place this 35-mm tissue culture dish inside the tissue culture
incubator for 2.5 h (see Notes 3 and 16). Gently swirl the dish
every 15–20 min during digestion or, if available, one can use
a low speed agitator placed inside the tissue culture incubator.
In the latter case, the speed should be adjusted to the lowest
possible speed for minimal agitation, to avoid damage to the
myofibers.
3. At the end of the digestion period, transfer each muscle to a
35-mm tissue culture dish containing 10% HS.
3.1.4. Separation of the 1. Pre-flush all Pasteur pipettes with 10% HS, prepared as
Three Tendons and described in item 8 in Subheading 2.6.1.
Release of Myofibers 2. Place one muscle at a time under the stereo dissecting
microscope.
3. Identify the two grooves running between the three tendons
separating the middle from the two lateral tendons.
4. Being careful not to touch the muscle, insert the tip of a pair
of forceps into one of the grooves and hold the muscle in place
by securing the connective tissue between the tendons to
the dish.
5. Use another pair of forceps to gently pull away the connective
tissue that holds the tendons and their attached muscle tissue
together.
6. Continue removing the connective tissue until the lateral ten-
dons are separated from the middle tendon and its attached
myofibers.
7. Holding the muscle only at its tendons, transfer the muscle
preparation to a 35-mm dish containing 3 mL of DMEM
containing 10% HS.
8. While grasping one end of the middle tendon with a pair of
forceps, use a second pair of forceps to grip its surrounding
connective tissue sheath and pull gently. If the sheath does not
come off easily, use fine point forceps to pull the connective
tissue perpendicular to the line of the muscle and cut it off.
9. Repeat steps 1–7 with the second FDB muscle until all six ten-
dons and their attached myofibers are in the 35-mm tissue cul-
ture dish containing 10% HS.
10. For one tendon at a time: hold one end of the tendon with a
pair of forceps and with the tip of a second pair gently separate
the myofibers from the tendon. The liberation of the myofibers
from the two lateral tendons should be easy, while the middle
tendon requires patience since the myofibers are attached to it
more firmly.
11. Use a wide-bore Pasteur pipette to gently triturate the clumps
of myofibers until they disengage into single myofibers. The
number of trituration rounds can vary, but it may take at least
five times. Excessive trituation can lead to fiber damage (see
Note 3).
12. Remaining clumps should be transferred to another 35-mm
tissue culture dish containing 10% HS and further triturated
until disengaged into single myofibers.
13. Set the stereo dissecting microscope magnification so that the
small pieces of connective tissue floating around in the suspen-
sion are visible and use fine forceps (or standard narrow-bore
Pasteur pipette, fire-polished) to pick them out. Continue until
the myofiber suspension is clean of any connective tissue
debris.
14. Triturate the myofiber suspension ten more times using a 9″
Pasteur pipette with a fire-polished tip to further separate small
clumps of myofibers.
3.1.5. Further Purification 1. Add 10 mL of DMEM containing 10% HS to each of the three
of FDB Myofibers glass Corex tubes.
2. Using the trimmed 100-ml pipette tip, transfer the myofiber
suspension to the top of the 10% HS column in the first Corex
tube. Allow the myofibers to settle (at 1 × g) through the HS
column for 15 min at room temperature (see Note 17). This
step is important for purifying the myofibers from free mono-
nucleated cells, debris, and occasional damaged myofibers.
3. As soon as the myofibers are settled, aspirate about 11 mL of
the supernatant (leaving about 1–1.5 mL). Triturate the
myofiber suspension gently with a 5″ fire-polished Pasteur
pipette and transfer the suspension to the next Corex tube as
described in step 2.
4. Allow myofibers to settle and transfer the myofiber suspension
to the third Corex tube as in steps 2 and 3.
448 P. Keire et al.
5. Allow myofibers to settle and harvest the final myofiber

suspension. Following the third purification, the residual vol-
ume of medium to be left with the myofiber suspension
depends on the number of culture dishes and the desired
myofiber number per dish. Typically in our studies the volume
of the final myofiber suspension is 300 ml, which is sufficient
for culturing four to six dishes.
3.1.6. Preparation of Isotonic PureCol collagen can be prepared during the settling of
Isotonic PureCol myofibers. The isotonic mixture should be kept on ice. Stock
Collagen PureCol is an acidic solution, and when made isotonic, it gels rap-
idly if not maintained at 4°C (see Note 1).
1. Place the PureCol collagen stock bottle, the 7× DMEM, and
one 15-mL conical tube on ice.
2. On ice: Add 1 volume of 7× DMEM and 6 volumes of PureCol
to the 15-mL conical tube and mix gently. Calculate the vol-
ume of stock PureCol needed for the experiment based on
using 120-ml isotonic PureCol collagen to coat each 35-mm
tissue culture dish. Use pH paper strips to ensure a neutral pH
of the PureCol collagen in DMEM solution. The pH of this
solution rises slightly after coating the culture dish. If the pH
remains acidic after coating a test dish, add one to two drops
of 1 M NaOH to the PureCol collagen in DMEM solution.
3.1.7. Coating Culture 1. On ice: Transfer 120 ml of isotonic PureCol collagen to the
Dishes with Isotonic center of a 35-mm culture dish and immediately use the
PureCol Collagen and L-shape spreader to coat the dish evenly. The coated culture
Myofiber Culturing plates need to be kept on ice until used as detailed below, to
avoid premature matrix gelling.
2. Gently swirl the myofiber suspension (in the 15-mL tube) for
even distribution of myofibers throughout the residual medium.
3. Remove one culture dish at a time from ice to allow rapid
warming to room temperature.
4. Use a wide-bore, 100-ml micropipette tip to dispense about
50 ml of the myofiber suspension per each culture dish.
5. Gently swirl the culture dish to allow even distribution of the
myofibers.
6. Repeat steps 2–5, one dish at a time, for additional culture
dishes.
7. Transfer the culture dishes to the tissue culture incubator for a
minimum of 20–30 min to allow the formation of PureCol
collagen matrix and the adherence of the myofibers to the
matrix.
8. Remove dishes from the incubator. Gently add 1 mL of
myofiber culture medium to each dish without agitating the
myofibers and return dishes to the incubator.
When the effect of growth factors on satellite cell prolifera-

tion/differentiation is investigated, parallel cultures are main-
tained in myofiber culture medium with/without additives,
and the medium is replaced every 24 h to ensure that growth
factors do not become rate limiting. These cultures can be used
for monitoring satellite cells and their progeny in live cultures
and for fixed/immunostained cultures as detailed in Fig. 2 and
in Subheading 3.3.
3.2. Isolation of Single The information in this introductory section is provided to assist in
Myofibers from the the identification of the extensor digitorum longus (EDL) muscles.
Extensor Digitorum The EDL muscle is situated at the ventral-lateral aspect of the
Longus Muscle hindlimb, running from the knee to the ankle, extending to the
2nd-5th digits (44). The EDL actually consists of four combined
muscle bellies and their tendons; the bellies arise from the lateral
condyle of the tibia and the front edge of the fibula (2 tendons at
the origin of the muscle). The tendons lie close to each other and
appear as one glistening white tendon that continues down to the
surface of the ankle. At the ankle joint it separates to four tendons,
each attached to one of the 2nd-5th digits. As the EDL contracts,
the four digits are extended. For additional details about the anat-
omy of the EDL muscle see Note 15.
As detailed in Subheading 3.1, we typically use only the
hindlimb muscles in our studies. Figure 5 depicts “real-live” images
of the steps in EDL muscle isolation with emphasis on the location
of the specific tendons that are handled during the process. It is of
utmost importance to delicately manipulate the muscle of interest
only at the tendons during its excision and further processing.
The EDL single myofiber isolation procedure described here has
also been adapted in our laboratory for the isolation of myofibers from
the diaphragm, masseter and extraocular muscles (see Note 18).
3.2.1. Initial Steps Prior to Matrigel solution preparation and plate coating (steps 1-6) are
Harvesting the Muscle and done on ice.
Preparation of Digestive
1. Thaw the required amount of Matrigel by placing frozen
Enzyme
aliquot(s) on ice for at least 30 min and as much as 1.5 h to
Preparation of Matrigel allow the Matrigel stock to completely liquefy for subsequent
Working Mixture and dilution to the working solution (see Note 2).
Coating 24-Well Tissue 2. Pre-chill a 50-mL conical tube on ice and transfer the thawed
Culture Dishes with Matrigel into the tube. Add ice-cold DMEM to dilute the
Matrigel Matrigel to a final concentration of 1 mg/mL. Gently mix the
Matrigel and DMEM by several repetitive drawings through a
1-mL glass pipette. An optimal Matrigel stock is at ~10 mg/
mL protein concentration, further diluted at 1:10 for the work-
ing Matrigel solution. Stock protein concentration can vary
greatly from lot to lot and should be monitored. Allow the
diluted Matrigel solution to cool on ice for at least 15 min.
450 P. Keire et al.
3. After 15 min, use a chilled 1-mL glass pipette to draw up the

diluted Matrigel solution and coat wells with an appropriate
volume (250–300 ml per well for a 24-well plate). In our expe-
rience, 2 mL of working Matrigel solution can be used to coat
an entire 24-well plate; we typically coat six to eight wells at a
time as detailed next.
4. Per each series of wells, leave the culture plate coated with
the Matrigel working solution on ice for 7 min, then use the
same pipette as before (held cooled in a tube on ice) to
remove the Matrigel solution and place it back in the 50-mL
conical tube that is kept on ice. This will leave a thin coat of
Matrigel at the bottom of the wells.
5. Once all of Matrigel solution has been placed back in the tube,
use the same pipette to coat the next set of wells. Leave the
diluted Matrigel in each well for 7 min.
6. Having coated all the desired wells per 1 tray, tilt the tray and
use a 20-ml pipette tip to carefully remove residual Matrigel
and place it back in the 50-ml conical tube that is kept on ice
(see Note 19).
7. Incubate the Matrigel-coated multiwell dishes in the tissue cul-
ture incubator for at least 1 h.
8. About 10 min before culturing myofibers, take the Matrigel-
coated, 24-well dish out of the incubator to the tissue culture
hood and open the lid. This will allow evaporation of water
that otherwise will condense on the underside of the lid when
moving the dish from the warm incubator to room tempera-
ture. If allowed to form, the condensation will drip into the
well, disturbing the Matrigel coating.
Coating Glassware and This is done to minimize adherence of myofibers to plasticware

Plasticware Dishes with and glassware used during the isolation process.
Horse Serum
1. For each EDL muscle being processed, coat six plastic 100-
mm petri dishes and one to two 60-mm petri dishes with
filtered horse serum (HS). Successively transfer a volume of
HS to each petri dish that is sufficient to cover the bottom of
the plate, then swirl the dish to coat evenly. Allow each dish to
sit with HS for about 2–3 min at room temperature and then
remove HS and apply it to the next dish. After all dishes have
been coated, add 9–12 mL of DMEM to each 100-mm petri
dish and 3–5 mL to each 60-mL petri dish. One may consider
processing a pair of EDLs together (which reduces usage of
materials and supplies), but we typically process each EDL
alone to allow for better separation of fibers with less debris.
2. Incubate the petri dishes in the tissue culture incubator until
needed following muscle digestion.
3. Coat the fire-polished Pasteur pipettes by flushing HS solution

through the pipettes several times. The coated pipettes are
then placed vertically in sterile plastic tubes (e.g., 5 mL Falcon
tubes) to maintain sterility and also for reflushing HS through
the pipettes to refresh the coating.
Preparation of the 1. Prepare 0.4% collagenase type I solution by dissolving 0.012 g

Digesting Enzyme Solution of collagenase in 3 mL of DMEM. Use a 0.22-mm syringe filter
and Post-digestion Rinse attached to a 3- or 10-cc syringe to filter the collagenase solu-
Plates tion into a 35-mm tissue culture dish. We prepare this solution
fresh for each experiment (see Note 16).
2. Fill three 100-mm petri dishes with 9-mL DMEM and place in
tissue culture incubator to warm dishes for later use as rinse
dishes.
3.2.2. Dissection of EDL 1. Euthanize one mouse according to institute regulations.

Muscle (Fig. 5) 2. Spray the hindlimbs with 70% ethanol.
3. Secure the mouse, lying on its back, to the dissecting board by
pinning down the forelimb diagonal to the hindlimb to be
dissected.
4. Use straight rounded-tip scissors to cut through the skin,
opening a small incision above the knee.
5. Holding the skin with fine forceps, insert the rounded-tip scis-
sors beneath the skin and carefully open the scissors to loosen
the skin from the underlying muscles.
6. Extend the incision to a point just in front of the digits.
7. Loosen the skin as you go, being careful not to cut the under-
lying muscles or blood vessels.
8. Cut and remove the skin from the knee to the paw (Fig. 5a)
and cut the fascia (thin connective tissue layer that covers the
muscles) that overlays the EDL and TA muscles (Fig. 5b). This
will facilitate the identification of the tendons.
9. Identify the four tendons in the foot at the insertion of the EDL,
each extending to one of the digits but not the large toe.
10. Use the microscissors to cut all four tendons (Fig. 5c).
11. Using fine forceps, gently isolate and pull the portion of the
tendon before its division (into 4 tendons) at the ankle up from
the paw (Fig. 5d, e); the tendon should slide up and out from
under the connective tissue sheath at the ankle, with the four
divisions trailing behind it. Carefully work the tendon of the
EDL out from underneath the TA tendon and pull the tendon
out of the ankle with the four divisions trailing behind it
(Fig. 5e).
452 P. Keire et al.
12. Identify the two tendons that are located by the knee cap, fac-
ing the lateral part of the leg (i.e., opposite to the midline of
the body).
13. Use microscissors to cut these tendons as far as possible from
the muscle itself (Fig. 5f).
14. Grasp the four tendons and carefully pull distally toward the
toes to remove the EDL muscle.
15. The EDL should slide underneath the TA muscle and should
pull out easily (Fig. 5 g, h). It is important not to apply too
much force. If the muscle does not slide out easily, one or both
tendons at the muscle origin at the knee may still be attached
to the bone. In this case, identify the yet attached tendon and
cut it.
16. The muscle should be handled only by its tendons to prevent
damage to the myofibers. Be careful not to injure the anterior
tibial artery that supplies blood to the EDL, to avoid blood cell
contamination of the myofiber preparation.
3.2.3. Enzymatic Digestion 1. Holding the muscle by its four tendons, place the EDL in a
35-mm tissue culture dish containing warm DMEM to rinse.
Next, transfer the muscle to the 35-mm tissue culture dish
containing 0.4% collagenase I solution. A pair of EDLs can be
digested in the same dish.
2. Place the dish inside the tissue culture incubator for 45–60 min
(see Notes 3 and 16). Gently swirl the dish every 15–20 min
during digestion (alternatively, one can use a low speed agita-
tor placed inside the tissue culture incubator) to facilitate mus-
cle dissociation.
3.2.4. Liberation of Single Use a stereo dissecting microscope throughout the procedure,
Myofibers from Muscle Bulk which involves rinses of the digested muscle bulk and a 3-step
sequence of muscle bulk trituration to release myofibers. All Pasteur
pipettes used in this process should be fire-polished. It is recom-
mended to spend no more than 5–7 min at a time per each tritura-
tion step. When processing multiple EDLs it is a good strategy to
alternate between muscle bulks so that only one EDL is outside of
the incubator at a time in order to minimize muscle cooling.
Additionally, the recommended number of rinses of the digested
muscle and of individual myofibers as detailed in this section should
not be overlooked. The myofiber rinses are essential for minimiz-
ing the contribution of non-myogenic cells that are released from
the muscle bulk during the enzymatic digestion. Unless myofibers
are well rinsed, such non-myogenic cells will be co-isolated with
the myofibers and eventually produce many progeny in the rich
culture conditions.
1. Inspect the muscle under the stereo dissecting microscope to
make sure that the myofibers are loosened from the muscle
bulk; the muscle should look like a loose skein of yarn. If the
myofibers are not loosened, continue enzymatic digestion for
another 10 min and check again.
2. Retrieve from the incubator the three 100-mm petri dishes
containing 9-mL DMEM (rinse plates). Use the widest bore
Pasteur pipette to transfer the muscle bulk from the collagenase
solution to the first DMEM rinse plate to wash away the colla-
genase and debris that might have dissociated from the muscle
during digestion. Transfer the muscle to the second then third
petri dish for further dilution of any possible collagenase that
may remain. These rinses must be performed with great care;
limit the amount of mechanical manipulation of the muscle or
swirling of the dish until the trituration step is reached.
3. Retrieve from the incubator one of the six 100-mm petri dishes
that were pre-coated with HS and filled with DMEM (this will
be the holding dish for the muscle bulk and will be used in
several of the steps described below). Transfer the rinsed mus-
cle bulk to the holding dish. Place the dish in the incubator for
approximately 10 min to allow the tissue to warm up.
4. Retrieve from the incubator a second HS-coated, DMEM con-
taining 100-mm dish. Using the same widest-bore pipette,
transfer the muscle bulk to this dish (1st trituration dish).
Return the holding dish to the incubator to warm.
5. Use another HS-coated Pasteur pipette (tip diameter: approx
3–4 mm) to triturate the muscle along its length. This orienta-
tion of the EDL muscle during triturations is critical to prevent
damage to the myofibers.
6. When single myofibers are liberated from the muscle, its diam-
eter decreases. Therefore, use a narrower bore pipette for sub-
sequent triturations.
7. When 10–15 viable single myofibers are released, retrieve from
the incubator the holding plate, transfer the muscle bulk into
it and place it back in the incubator. Additionally, place the
dish with the single myofibers in the tissue culture incubator to
keep the fibers warm (typically the fibers from this 1st tritura-
tion round are not used, but save the plate in case it is needed).
Allow the holding plate with muscle bulk to warm up for at
least 5–10 min in the incubator before the next round of
trituration.
8. Retrieve from the incubator a third HS-coated, DMEM con-
taining 100-mm dish (2nd trituration dish) and the holding
plate with muscle bulk. Transfer the muscle bulk to the 2nd
trituration dish using the same widest-bore pipette used in the
1st trituration dish. Using a wide bore-pipette with a smaller
diameter, triturate the tissue until 30–50 myofibers are obtained
(but do not triturate the tissue for more than 5–7 min). Transfer
454 P. Keire et al.
the muscle bulk back to the holding dish and place both the
holding dish and the dish with released myofibers back in the
incubator.
9. Follow the pattern of moving the muscle bulk as described in
steps 7 and 8, create a 3rd trituration dish; triturate the muscle
bulk until approximately 100 myofibers have been released.
When the 3rd trituration step is complete, transfer the tissue
back into the holding dish and place both the holding dish and
the dish with released myofibers back into the incubator.
Typically, three rounds of triturations are sufficient to dissoci-
ate the muscle bulk entirely.
10. Using a HS-coated 9″ pipette (standard bore size) begin to
transfer individual fibers from the 2nd and 3rd trituration plates
to the remaining two HS-coated, DMEM containing 100-mm
petri dishes (collection plates). Refresh the HS coating of the
pipette before each fiber transfer so that fibers do not adhere to
the glass. Alternate between (at least) two collection plates to
minimize cooling of the myofibers. As a general scheme:
(a) Transfer ten fibers from the 2nd trituration dish to one of
the collection plates and then move both plates back to
the incubator.
(b) Remove the 3rd trituration dish from the incubator and
transfer ten fibers to a second collection plate. Try to avoid
using the first trituration dish as the fibers from this tritu-
ration are much more fragile and often have more non-
myogenic cells attached to them.
(c) Repeat this process when triturating the second EDL,
alternating with the first EDL throughout the processing.
If using only one EDL, always allow the plates to rest for
10 min in the incubator before repeating the process.
Collect those fibers that are relatively straight and are not
partially contracted.
11. Once a large enough number of fibers has been collected (gen-
erally 20–30 per collection plate) begin selecting fibers that
will be used for analysis. Remove the 100-mm collection plates
one at a time and visually inspect the fibers under the highest
magnification available. Avoid fibers that have visible associ-
ated debris, also avoid those that are kinked or partially con-
tracted (see Note 3). Transfer fibers that pass these criteria to
another HS-coated 60-mm dish (final dish). Try not to place
too many fibers into one single plate as the fibers may become
entangled with each other or associated with debris that may
have been carried over in the transfer (as an approximation, no
more than 3 fibers/1 cm² of dish surface area). Although
including this step of fiber selection and transfer to the final
plate requires extra time, it allows for another wash step to
remove non-myogenic cells that may have been carried over

during trituration, thereby ensuring a more optimal fiber
preparation.
3.2.5. Culturing Single This section describes how to establish and maintain EDL myofiber
Myofibers in 24-Well cultures. We also harvest freshly isolated myofibers for satellite cell
Multiwell Dishes analysis (2, 7, 16, 22) (see Subheading 3.3.3).
1. Transfer a Matrigel-coated, 24-well multiwell dish from the
incubator to the tissue culture hood and open its lid to allow
moisture, generated during the incubation period, to evapo-
rate. Add 500 ml of pre-warmed, culture medium (see item 4
in Subheading 2.6.2) to each well.
2. Bring the 60-mm petri dish containing single fibers (final dish)
to the dissection microscope along with the 24-well plate.
3. Under the dissection microscope, use a fire-polished, HS-coated
9″ Pasteur pipette to select fibers that are free of associated
debris or connective tissue. Transfer one fiber at a time with
minimal residual medium and gently release the myofiber into
the bottom of the well as close to the center as possible. After
myofibers are dispensed to the desired number of wells, check
again under the stereo dissecting microscope to ensure that
indeed there is a myofiber in each well. This step is necessary
since occasionally myofibers adhere to the Pasteur pipette and
are not released into the well or the fiber becomes damaged in
the transfer. Avoid excessive agitation of the fibers.
4. If needed, add a myofiber to empty well(s) or replace with an
intact fiber. Minimize the length of time the final plate and
multiwell dish are held at room temperature; transfer dishes
back to the incubator after 10 min to warm while continuing
to dispense isolated fibers.
5. When the desired number of fibers has been plated, place the
24-well multiwell dish in the tissue culture incubator. Avoid
handling the plate (i.e., to inspect fibers) for a minimum of
18 h (overnight). Myofibers can also be cultured for early time
points (e.g., to analyze satellite cell numbers from freshly iso-
lated fibers; see Subheading 3.3.3), however, extra special care
should be exercised when handling such early time points for
microscopic examination or immunostaining because the fibers
are only loosely adhered and too much manipulation can dam-
age the fibers and cause contraction.
6. After the fibers have been in culture for 3 days, gently add an
additional 500 ml of complete media to the fibers. After
3 more culture days, replace the entire old medium with 500 ml
of fresh growth medium. Continue changing the media every
3 days. We typically maintain myofiber cultures for 10–14 days
without any apparent decline in culture quality. Depending on
456 P. Keire et al.
the goal of the project, we also have maintained fiber cultures

for up to 3 weeks, but Matrigel may be partially degraded by
then, and myotubes may detach from the plate. Moreover, the
medium change schedule may need to be more frequent for
longer culture periods.
3.3. Immunolabeling This section details current protocols used in our laboratory to fix
of FDB and EDL myofiber cultures for immunofluorescent studies of satellite cells and
Myofiber Cultures their progeny. FDB myofiber cultures are typically fixed with ice-cold
methanol (the preferred fixative when working with dishes coated
with PureCol collagen), whereas the EDL myofiber cultures are typi-
cally fixed with paraformaldehyde that is pre-warmed to room tem-
perature. Ideal fixatives for FDB or EDL myofiber cultures are not
necessarily the optimal fixatives for specific antigen detection. Thus,
when analyzing single myofibers via immunofluorescence, fixatives
should be optimized for both preserving the myofibers and the anti-
gens being analyzed. Fixation protocols described in this section are
also appropriate for detecting proliferating satellite cells in single
myofibers by autoradiography following labeling with 3H-thymidine
(32, 34) or when analyzing proliferation using bromodeoxyuridine
(2, 16, 42, 43). All steps are done in a sterile manner. Handling anti-
bodies strictly in the tissue culture hood minimizes possible bacterial
contamination and helps maintain antibody stocks for years.
3.3.1. Fixing and 1. Warm DMEM in a water bath set at 37°C.

Immunofluorescent 2. Rinse cultures with 500 ml warm DMEM three times. Following
Staining of Isolated FDB the final rinse add 1-mL ice-cold 100% methanol to each
Myofiber Cultures 35-mm tissue culture dish and transfer the dishes to 4°C for
10 min.
3. Return dishes to room temperature, aspirate the methanol and
allow the dishes to air-dry for 10–15 min in the tissue culture
hood (see Note 20).
4. Add 1.5 mL of blocking solution (TBS-NGS) to each culture
dish, to block nonspecific antibody binding.
5. Cultures are then kept at 4°C for at least overnight and up to
2 weeks. Bring cultures to room temperature when ready to
start antibody labeling.
6. Dilute the appropriate primary antibody in the NGS-TBS
blocking solution. If not otherwise published, before diluting
your antibody, test a range of dilutions to determine the lowest
concentration of antibody that gives a clear specific signal with-
out nonspecific background.
7. Rinse the cultures three times with 500 ml TBS-TW20.
8. Remove the final TBS-TW20 rinse and add 100 ml of the pri-
mary antibody solution. Incubate for 1 h at room temperature
followed by an overnight incubation at 4°C in a humidified
chamber (see Notes 21 and 22). Primary and secondary anti-

bodies are applied at the center of the dish followed by a light
swirling on a flat surface to ensure optimal spreading of the
antibody across the dish. This approach allows using just 100 ml
antibody solution, which is beneficial for conserving antibody
stocks.
9. Dilute the appropriate secondary antibody in the NGS-TBS
blocking solution. Secondary antibodies are often diluted at
1:1,000 or greater, but the researcher needs to determine the
optimal dilutions for their specific study.
10. Rinse cultures with 500 ml TBS-TW20 three times.
11. Remove the final TBS-TW20 rinse and add 100 ml of the
diluted secondary antibody. Incubate for 1–2 h at room
temperature.
12. Remove the secondary antibody and wash three times with
500 ml TBS-TW20.
For nuclear visualization, add at least 100 ml of DAPI working solu-
tion (1 mg/mL, diluted in TBS-NGS prior to use; see item 8 in
Subheading 2.7.1) and incubate for 30 min at room
temperature.
13. Rinse the cultures twice with 500 ml TBS-TW20 followed by a
final rinse with 500 ml TBS.
14. Remove the TBS and mount in Vectashield mounting medium.
Add one drop at the center of each culture dish and cover with
a cover slip. Cultures should be viewed as soon as possible, but
if not, then stored at 4°C sealed in Parafilm, covered with alu-
minum foil to protect from light, and viewed within a week
after immunostaining to avoid fading.
3.3.2. Fixing and EDL myofiber cultures are fixed by slightly different approaches
Immunostaining when fixing long term cultures (detailed in this section) or when
Long-Term EDL Myofiber fixing freshly isolated (Time 0; T0 fibers; detailed in the follow-
Cultures ing section). Importantly, when fixing T0 cultures and early time
points, use a stereo dissecting microscope throughout the proce-
dure to ensure that the fibers are not lost or become damaged.
All additional wash steps should be performed using a 9″ glass
fire-polished Pasteur pipette. At later time points, when fibers
and emanating cells are adhering strongly to the matrix, one
may not necessarily require the aid of a microscope when fixing
or rinsing the cultures.
1. Warm the needed volume of the 4% paraformaldehyde fixative
solution to room temperature (according to the number of
wells to be fixed, and using about 500 ml per well).
2. While observing each myofiber under the stereo dissecting
microscope, use a Pipetman to gently, without agitating the
458 P. Keire et al.
culture or touching the myofiber, add an equal volume (500 ml)

of the 4% paraformaldehyde fixative solution to the culture
medium in each well in the 24-well dish. Allow 10 min at
room temperature for the fixation, then carefully remove (by
aspiration or using a pipette) the culture medium-paraformal-
dehyde fixative mixture and rinse each well three times with
500 ml TBS.
3. Add 500 ml of TBS-TRX100 for 5 min at room temperature.
Alternatively, Triton X-100 can be omitted (but cultures can
be treated with it later) as some antigens may be more opti-
mally detected if Triton X-100 has not been used.
4. Add 500 ml of blocking solution (TBS-NGS) to each of the 24
wells, to block nonspecific antibody binding.
5. Follow steps 5–13 as described in Subheading 3.3.1. However,
when exposed to antibodies, the 24-well multiwell trays should
be continuously and gently swirled as described in Note 22, as
uneven antibody staining can otherwise occur.
6. Remove the final TBS rinse and add one drop of Vectashield
mounting medium as in step 14, Subheading 3.3.1. We prefer
not to use cover slips when working with 24-well, multiwell
trays. Instead, we add 300 ml of the glycerol mounting solu-
tion (25% glycerol in TBS) following the initial drop of
Vectashield to allow sufficient mounting medium coverage of
individual wells in 24-multiwell trays. Trays should be viewed
as soon as possible. If they cannot be viewed immediately, they
should be stored at 4°C sealed in Parafilm, covered with alumi-
num foil to protect from light, and viewed within a week after
imunostaining to avoid fading.
3.3.3. Fixing and Plate EDL fibers as previously described in Subheading 3.2.5, but
Immunostaining Freshly instead of plating the fiber in a well containing 500-ml medium,
Isolated (T0 ) EDL Fibers transfer the fiber with residual DMEM (~150 ml) into the center of
a Matrigel-coated well that has not received growth medium. The
fiber should be sitting in a droplet of DMEM, on top of the
Matrigel to ensure that it does not dry out. After the desired num-
ber of fibers has been dispensed (1 per well), place the plate back
in the incubator for 3 h to allow the fibers to adhere to the Matrigel.
Minimize the amount of time that the fibers remain outside of the
incubator and do not subject the plate to sudden motion as this
can cause the fibers to contract or lose contact with the plate
substrate.
1. Use a fire-polished Pasteur pipette to slowly add the 4% para-
formaldehyde fixative solution (pre-warmed to room tempera-
ture) until the droplet containing the fiber has approximately
doubled in volume. Allow the fiber to sit in the fixative solu-
tion for 10 min at room temperature.
2. Follow steps 2–5 as described in Subheading 3.3.2.

3. Remove the final TBS rinse and add one drop of Vectashield
plus 300 ml 25% glycerol-TBS.
4. Observe and analyze the fibers under the microscope then seal
the multiwell tray with Parafilm and store at 4°C and in the
dark (e.g., can be stored wrapped with aluminum foil) when
not in use. Typically, we aim to complete analyses within a
week following fiber harvesting.
4. Notes
1. PureCol collagen (formally known as Vitrogen), is a sterile

solution of purified, pepsin-solubilized, bovine hide collagen
(97% Type I, 3% Type III) dissolved in 0.01 N HCl and stored
at 4°C until used (vendor: Advanced BioMatrix). In our stud-
ies, PureCol collagen is made isotonic by mixing 6 volumes of
stock PureCol collagen with 1 volume of 7× DMEM. The iso-
tonic solution is prepared just prior to coating dishes because
it gels rapidly at room temperature. To obtain consistent coat-
ing, the culture dishes should be pre-cooled and coated on ice.
When removed from the ice, these dishes warm up rapidly and
are ready for myofiber addition. Preparations of collagen Type
I from other sources (e.g., Sigma-Aldrich) have been used by
some laboratories as an alternative to PureCol collagen. The
use of alternative sources would require pre-screening to ensure
compatibility; we only have experience with the bovine-derived
product.
2. Matrigel (BD Biosciences) is a solubilized basement membrane
preparation extracted from the Engelbreth-Holm-Swarm
mouse sarcoma, a tumor rich in extracellular matrix proteins.
Its major component is laminin, followed by collagen IV,
entactin, and heparan sulfate proteoglycan (45). Matrigel is
shipped on dry ice and is stored at −20°C until aliquoted.
Matrigel should be thawed on ice; never use at a warmer tem-
perature, as it will prematurely gel. To ensure Matrigel stabil-
ity, we follow the manufacturer’s handling instructions, thawing
the product on ice (overnight in an ice bucket placed at 4°C).
Once liquefied, Matrigel is aliquoted with pre-chilled 1-mL
serological glass pipettes into tubes chilled on ice. Typically, we
aliquot 200 ml each into 2-mL cryogenic vials sealed with
O-rings. These aliquots are stored at −20°C. We have observed
some batch-to-batch variation in the time it takes to thaw the
aliquots for final dish coating, therefore, for consistency, we
typically allow Matrigel aliquots to thaw for 1.5 h. Matrigel can
be purchased in its standard format (BD Biosciences, cat. no.
460 P. Keire et al.
354234) or in its growth factor reduced format (BD Biosciences,

cat. no. 354230). We have typically used the growth factor
reduced format, but more recently have begun using the stan-
dard format for routine studies in rich growth medium.
Invitrogen carries Matrigel-like products that might be useful
as an alternative to Matrigel (e.g., Geltrex; cat. no. A11343);
however, we do not have sufficient experience with the latter
product for detailed recommendations.
3. Adjustments, such as concentration of collagenase, length of
muscle digestion, and extent of muscle trituration for releasing
myofibers, may be needed when isolating myofibers from
younger/older mice, other mouse strains, mutant mice, or
other rodents such as rats. Prolonged digestion and extensive
trituration of the muscle bulk will result in poor yields of intact
myofibers. Myofibers that are damaged in the course of the
isolation can be distinguished from the intact myofibers since
they typically hypercontract. Bent myofibers are also damaged
to some degree and should not be collected when preparing
myofiber cultures.
4. Falcon Primaria 24-well multiwell dishes (BD Biosciences; cat.
no. 353847) were initially used for single myofiber cultures;
however, we find that the standard, less expensive, Falcon
24-well, multiwell dishes (cat. no. 353047) are as good.
5. Horse serum is used for tissue culture medium and for coating
plastic and glassware. HS used for tissue culture media should
be pre-characterized by comparing sera from various suppliers
(e.g., over years of studies, our preferred serum lots came typi-
cally from Invitrogen, HyClone, or Sigma-Aldrich). We select
HS based on its capacity to support proliferation and differen-
tiation of primary chicken myoblasts cultured at standard and
clonal densities (46). One may consider replacing HS with
bovine serum albumin (BSA) for coating plastic and glassware
to further minimize any possible activation of satellite cells
during myofiber isolation. However, attention should be given
to the purity of the BSA as some lots may contain growth-
promoting factors.
6. The Controlled Processed Serum Replacement 2 (CPSR-2;
Sigma-Aldrich) that had been routinely used in our myofiber
culture studies (21, 22, 26, 34–36, 41) has been discontinued.
The source of this discontinued CPSR-2 was dialyzed bovine
plasma. This product was further processed in a manner that
also reduced lipids. Another alternative serum replacement
product, serum replacement 2 (50×) (Sigma-Aldrich; cat.
C9388) contains highly purified bovine serum albumin, insu-
lin, and transferrin, and its use for mouse myofiber cultures has
been previously described (43).
7. Fetal bovine serum (FBS) should be pre-characterized by com-

paring sera from several suppliers (e.g., over years of studies,
our preferred serum lots came from Invitrogen, HyClone, or
JR Scientific). We select FBS based on the capacity of the serum
to support proliferation and differentiation of mouse primary
myoblasts cultured at various cell concentrations. Only sera
able to support growth and differentiation over a wide range of
concentrations, down to a clonal density, are employed in our
studies. Primary myogenic cultures are prepared according to
our published procedures (26, 27, 29).
8. Chicken embryo extract (CEE) is available commercially from
several sources with which we have no experience. We prepare
CEE in our laboratory using 10-day old White Leghorn
embryos (47). The procedure is similar to a previously described
method (48) but uses the entire embryo. We recommend this
approach over purchasing CEE if the investigator can obtain
embryonated chicken eggs, as the quality is thought to be
higher and the cost lower than that of purchased CEE.
9. Preparation of chicken embryo extract:
(a) Embryonated chicken eggs (8 dozen, White Leghorn; from
Charles River) are maintained in a standard egg incubator
(incubation conditions: a dry temperature of 38°C, a wet
temperature of 30°C and relative humidity of 56%). The fol-
lowing egg incubator is well suited for basic research use:
Marsh Automatic Incubator, model # PROFI, cat. no. 910-
028, manufactured by Lyon Technologies, Chula Vista, CA.
(b) After 10 days, batches of 15–30 eggs are removed from
the incubator and transferred into the tissue culture hood.
All steps from here on are performed in a sterile manner.
(c) Place the eggs lengthwise in the rack and spray with 70%
ethanol to sterilize. Wait for several minutes until the etha-
nol evaporates.
(d) Crack open one egg at a time into a 150-mm petri dish.
(e) Remove the embryo from surrounding membranes by pierc-
ing it with fine forceps. Rinse the embryo by transferring it
through three 150-mm petri dishes containing DMEM sup-
plemented with antibiotics (see item 1 in Subheading 2.6.1).
Swirl embryo a few times in each dish for a good rinse.
(f) Empty the egg remains from the initial 150-mm dish
(described in step d) into a waste beaker and repeat steps
d–f until the final rinse dish contains about 30 embryos.
(g) The embryos are transferred with fine forceps into a 60-mL
disposable syringe, forced through with the syringe
plunger, and the suspension is collected into a 500-mL
sterile glass bottle.
462 P. Keire et al.
(h) The extract is diluted with approximately an equal volume of

DMEM (supplemented with antibiotics as detailed in item 1 in
Subheading 2.6.1) and gently agitated for 2 h at room tem-
perature. To ensure good agitation, keep the maximum vol-
ume to one-half bottle capacity and place the bottle at a 45°
angle during the agitation.
(i) The extract is frozen at −80°C for a minimum of 48 h. It is
then thawed, dispensed into 50-mL conical tubes, and centri-
fuged at approximately 500 × g for 10 min to remove residual
tissue.
(j) The supernatant is pooled, divided into 40-mL aliquots and
kept frozen at −80°C for long-term storage. For short-term
storage, we typically prepare aliquots of 2.5 mL that are kept
frozen at −20°C.
(k) Prior to use, the CEE-thawed aliquot should again be centri-
fuged at about 800–1,000 × g for 10 min to remove aggre-
gates. The supernatant is then collected and added to the
DMEM-based medium to prepare the rich growth medium for
EDL myofiber cultures. The growth medium is then passed
through a sterile 0.22-mm filter (to clear remaining particles
and sterilize). All details of supplies for generating the medium
are in Subheading 2.6.2. To ensure optimal cell growth condi-
tions, we typically prepare only 250-mL medium each time,
and use it up within a few weeks.
10. Methanol is a colorless flammable liquid with an alcohol-like
odor. Use nitrile gloves, safety goggles, and a fume hood when
handling. It is important to refer to the MSDS instructions and
institutional regulations for further information regarding
storage, handling and first-aid.
11. Preparation 1 L of 10× Tris-buffered saline (TBS): Weigh
60.5 g of Tris-Base into a beaker and add 700-mL deionized
water. Stir on a magnetic stirrer until the powder has dissolved
and adjust the pH to 7.4. Add deionized water to bring the
volume up to 1 L, mix well, then autoclave or sterilize by pass-
ing it through a 0.22-mm filter, and store at 4°C. To make 1 L
of 1×TBS: Weigh 8.766 g NaCl into a beaker and add 100 mL
of 10× TB. Mix vigorously until the powder has dissolved. Add
deionized water to bring the volume up to 1 L, mix well, then
sterile filter and store at 4°C.
12. DAPI is potentially harmful. Avoid prolonged or repeated
exposure. We typically dissolve the entire powder in its original
container and generate a concentrated stock solution.
Alternatively, a ready-made DAPI reagent is available from
Molecular Probes. It is important to refer to the MSDS instruc-
tions and institutional regulations for further information
regarding storage and handling.
13. Paraformaldehyde is a white powder with a formaldehyde-like odor.

It is a rapid fixative and a potential carcinogen. When handling
paraformaldehyde, wear gloves, a mask, and goggles. It is impor-
tant to refer to the MSDS instructions and institutional regulations
for further information regarding storage, handling and first-aid.
14. Preparation of 100 mL of 4% paraformaldehyde with 0.03 M
sucrose: In a fume hood mix 4 g of paraformaldehyde powder
and 80 mL of deionized water in a glass beaker. Warm the
solution to 60°C with continuous stirring to dissolve the pow-
der. Allow the solution to cool to room temperature. Add one
to four drops of 1 N NaOH, until the opaque color of the
solution clears. Add 10-mL 1 M sodium phosphate. Adjust the
pH to 7.2–7.4 using concentrated HCl and color pH strips.
Add 1.026 g of sucrose. Bring the volume to 100 mL and filter
through a 0.22 mm disposable filter unit (Millipore; cat. no.
SCGPT01RE) into a bottle. Store at 4°C in an aluminum foil-
wrapped bottle for no more than 1 month.
15. For additional details about FDB-muscle anatomy refer to:
http://www.bartleby.com/107/illus443.html.
http://www.bartleby.com/107/131.html.
For additional details about EDL muscle anatomy refer to:
http://www.bartleby.com/107/129.html.
We recommend these links as good resources for anatomi-
cal description and schematic images of the muscles although
they refer to human muscles.
16. Collagenase concentration, as well as the optimal time for enzy-
matic digestion, should be adjusted for younger or older mice and
for other muscle groups. The enzyme sold by Sigma-Aldrich tends
to have consistent specific activity between batches, but attention
should be given to the specific activity with each batch. The vol-
ume needed for the preparation should be evaluated based on the
size of the tissue, so that the tissue will be fully covered by the col-
lagenase solution (e.g., 1.5 mL is sufficient to cover an EDL mus-
cle but more will be needed to cover a tibialis anterior).
17. The time required for the myofiber suspension to settle (at
1 × g) through 10 mL of 10% HS can vary between 5 and
15 min and the investigator should adjust this time accord-
ingly. A prolonged period results in a preparation with more
debris and residual single-cell carryover (not necessarily
myofiber-associated) released from the digested tissue.
Depending on mouse age, the number of rounds of myofiber
settling in the 15-mL glass Corex tubes, as well as the amount
of medium in the tube, may also need to be adjusted.
464 P. Keire et al.
18. Isolation of myofibers from Masseter, Diaphragm, and

Extraocular muscles:
(a) General comment:
Details in this section are provided in brief and focus mainly on
muscle harvesting and dissociation. All reagents are as described
for EDL myofiber isolation and culture in earlier sections of
this chapter. For all muscles, tissue is dissociated with 0.4% col-
lagenase (type I, source as listed item 7, Subheading 2.6.1, and
preparation as in step 1, Subheading “Preparation of the
Digesting Enzyme Solution and Post-digestion Rinse Plates”).
(b) The diaphragm muscle:
The diaphragm muscle consists of two portions; the costal
muscle, radially arrayed from a central non-contractile tendon
(the central tendon), and the crural diaphragm through which
the aorta, thoracic duct, and esophagus are transmitted. When
isolating single myofibers from the diaphragm (49), we have
found that at the risk of contaminating the preparation with a
greater amount of debris and shorter intercostal fibers of the
ribcage, removal of the diaphragm in whole with its immediate
supporting ribcage structure provides the greatest fiber yield.
After removing the diaphragm, rinse the muscle in a dish con-
taining pre-warmed DMEM. Observe the diaphragm under
the dissection scope and without touching the muscle, care-
fully remove any obvious fat or connective tissue, otherwise
this material can foul the prep. Other steps that ensure higher
and purer yield of fibers are the addition a wash step post-en-
zymatic digest and not “over-digesting” the muscle. Digesting
for 45–60 min in 0.4% collagenase is recommended. The mus-
cle is triturated centrally from the position of the central ten-
don and ends of the ribcage using the largest bore, fire-polished
Pasteur pipette. Expect no more that 50 ideal fibers (undam-
aged and without associated debris) per diaphragm.
(c) The masseter muscle:
The masseter muscle of the jaw (like the diaphragm) requires
harvesting some of the supporting skeletal structure along with
the muscle for the highest fiber yields. The masseter muscle is
a multilayered muscle with complex and extensive investment
geometry. We have found that including the origin and invest-
ment surfaces in the enzymatic digest, rather than risking dam-
age to the muscle at the tendon-bone interface ensures more
intact muscle fibers. Once removing the masseter muscles with
the associated bones, rinse and remove debris in a pre-warmed
100-mm petri dish containing DMEM. Digest the masseter
muscle en bulk with jaw and skull bone attachments in 0.4%
collagenase for 45 min followed by careful washes and tritura-
tions. This preparation typically yields 30–100 fibers.
(d) The extraocular muscles (EOM):
EOM are a unique set of muscles that control eye movements.

There are 6 EOMs per eye accessed by first bisecting the skull
from the top of the head and removing the brain to expose the
bone at top of the eye socket. Once exposed, the eye socket is
broken along its suture lines to expose the eye. Following
removal of the lacrimal gland, the entire eye with its associated
muscles is then removed “en bulk” by first cutting the optic
nerve at a point just behind the annulus of Zinn (the point
where 5 of the 6 EOM muscles meet). The isolation is com-
pleted by cutting the remaining soft tissue from around the
eye, freeing the eye from the socket. The eye, with the muscles
still attached, is placed in 0.4% collagenase digest for 90 min at
37°C and then transferred to fresh 0.4% collagenase for an
additional 45–60 min digestion. Gentle swirling of the digest
every 15 min will help increase the yield. The rinsing and tritu-
ration steps that follow are similar to those described for the
EDL, except with less vigor. This procedure yields an array of
short and long, thick and thin EOM myofibers for analysis;
30–100 fibers are typically derived per preparation.
19. The working Matrigel solution can be used to coat additional
trays after completing the first tray coating. Matrigel that has
been used to coat too many dishes, however, is less effective in
supporting myofiber adhesion. We typically limit reuse of
diluted Matrigel to three rounds of coating and work with a
larger volume of diluted Matrigel if coating more than 1 tray.
Also, we only use Matrigel that has been diluted the day of the
fiber isolation to maintain consistency.
20. The tissue culture dishes are dry when the bottom appears
opaque white.
21. For some antibodies the cultures may be blocked for just 2–4 h
at room temperature if overnight blocking is not desired.
22. For even and continuous distribution of the antibodies (both
primary and secondary), it is recommended to place the dishes
on a gyrating platform rotator (e.g., Lab-Line Maxi Rotator,
model no. 4631R) when staining cultures in 24-well, multi-
well dishes; without this agitation, the antibody solution tends
to rapidly accumulate at the well periphery, leading to uneven
staining across the culture.
Acknowledgments
The authors are grateful to the granting agencies that funded this
study. Our current research is supported by grants to Z.Y.R. from
the National Institutes of Health (AG021566; AG035377;
AR057794) and the Muscular Dystrophy Association (135908).
466 P. Keire et al.
The development the FDB myofiber isolation protocol described

in this chapter could not be possible without the valuable contri-
bution of our former lab member, Anthony Rivera, and previous
funding from the Muscular Dystrophy Association, the Cooperative
State Research, Education and Extension Service/US Department
of Agriculture (National Research Initiative), the National Institutes
of Health, and the Nathan Shock Center of Excellence in the Basic
Biology of Aging, University of Washington.
References
1. Mauro A (1961) Satellite cell of skeletal mus- 11. Yablonka-Reuveni Z, Day K (2011) Skeletal
cle fibers. J Biophys Biochem Cytol muscle stem cells in the spotlight: the satellite
9:493–495 cell. In: Cohen I, Gaudette G (eds)
2. Yablonka-Reuveni Z, Day K, Vine A, Shefer G Regenerating the Heart: Stem Cells and the
(2008) Defining the transcriptional signature Cardiovascular System. Springer, Humana
of skeletal muscle stem cells. J Anim Sci Press. pp. 173–200
86:E207–E216 12. Muir AR, Kanji AH, Allbrook D (1965) The
3. Hawke TJ, Garry DJ (2001) Myogenic satel- structure of the satellite cells in skeletal muscle.
lite cells: physiology to molecular biology. J J Anat 99:435–444
Appl Physiol 91:534–551 13. Yablonka-Reuveni Z (1995) Development and
4. Zammit PS, Partridge TA, Yablonka-Reuveni postnatal regulation of adult myoblasts.
Z (2006) The skeletal muscle satellite cell: the Microsc Res Tech 30:366–380
stem cell that came in from the cold. J 14. Boldrin L, Muntoni F, Morgan JE (2010) Are
Histochem Cytochem 54:1177–1191 human and mouse satellite cells really the
5. Collins CA, Olsen I, Zammit PS, Heslop L, same? J Histochem Cytochem 58:941–955
Petrie A, Partridge TA, Morgan JE (2005) 15. Seale P, Sabourin LA, Girgis-Gabardo A,
Stem cell function, self-renewal, and behav- Mansouri A, Gruss P, Rudnicki MA (2000)
ioral heterogeneity of cells from the adult mus- Pax7 is required for the specification of myo-
cle satellite cell niche. Cell 122:289–301 genic satellite cells. Cell 102:777–786
6. Sacco A, Doyonnas R, Kraft P, Vitorovic S, 16. Day K, Shefer G, Richardson JB, Enikolopov
Blau HM (2008) Self-renewal and expansion G, Yablonka-Reuveni Z (2007) Nestin-GFP
of single transplanted muscle stem cells. Nature reporter expression defines the quiescent state
456:502–506 of skeletal muscle satellite cells. Dev Biol
7. Day K, Shefer G, Shearer A, Yablonka-Reuveni 304:246–259
Z (2010) The depletion of skeletal muscle sat- 17. Shefer G, Rauner G, Yablonka-Reuveni Z,
ellite cells with age is concomitant with reduced Benayahu D (2010) Reduced satellite cell
capacity of single progenitors to produce numbers and myogenic capacity in aging can
reserve progeny. Dev Biol 340:330–343 be alleviated by endurance exercise. PLoS One
8. Charge SB, Rudnicki MA (2004) Cellular and 5:e13307
molecular regulation of muscle regeneration. 18. Allouh MZ, Yablonka-Reuveni Z, Rosser BW
Physiol Rev 84:209–238 (2008) Pax7 reveals a greater frequency and
9. Shefer G, Yablonka-Reuveni Z (2008) Ins and concentration of satellite cells at the ends of
outs of satellite cell myogenesis: the role of the growing skeletal muscle fibers. J Histochem
ruling growth factors. In: Schiaffino S, Cytochem 56:77–87
Partridge T (eds) Skeletal muscle repair and 19. Montarras D, Morgan J, Collins C, Relaix F,
regeneration. Springer, Dordrecht, The Zaffran S, Cumano A, Partridge T, Buckingham
Netherlands, pp 107–144 M (2005) Direct isolation of satellite cells for skel-
10. Morgan JE, Zammit PS (2010) Direct effects etal muscle regeneration. Science 309:2064–2067
of the pathogenic mutation on satellite cell 20. Beauchamp JR, Heslop L, Yu DS, Tajbakhsh
function in muscular dystrophy. Exp Cell Res S, Kelly RG, Wernig A, Buckingham ME,
316:3100–3108 Partridge TA, Zammit PS (2000) Expression
of CD34 and Myf5 defines the majority of 33. Bischoff R (1989) Analysis of muscle regenera-
quiescent adult skeletal muscle satellite cells. tion using single myofibers in culture. Med Sci
J Cell Biol 151:1221–1234 Sports Exerc 21:S164–S172
21. Yablonka-Reuveni Z, Rivera AJ (1994) 34. Yablonka-Reuveni Z, Rivera AJ (1997)
Temporal expression of regulatory and struc- Proliferative dynamics and the role of FGF2
tural muscle proteins during myogenesis of during myogenesis of rat satellite cells on iso-
satellite cells on isolated adult rat fibers. Dev lated fibers. Basic Appl Myol 7:189–202
Biol 164:588–603 35. Yablonka-Reuveni Z, Anderson JE (2006)
22. Shefer G, Van de Mark DP, Richardson JB, Satellite cells from dystrophic (mdx) mice dis-
Yablonka-Reuveni Z (2006) Satellite-cell pool play accelerated differentiation in primary cul-
size does matter: defining the myogenic tures and in isolated myofibers. Dev Dyn
potency of aging skeletal muscle. Dev Biol 235:203–212
294:50–66 36. Yablonka-Reuveni Z, Rudnicki MA, Rivera AJ,
23. Zammit PS, Golding JP, Nagata Y, Hudon V, Primig M, Anderson JE, Natanson P (1999)
Partridge TA, Beauchamp JR (2004) Muscle The transition from proliferation to differenti-
satellite cells adopt divergent fates: a mechanism ation is delayed in satellite cells from mice lack-
for self-renewal? J Cell Biol 166:347–357 ing MyoD. Dev Biol 210:440–455
24. Day K, Paterson B, Yablonka-Reuveni Z 37. Rosenblatt JD, Lunt AI, Parry DJ, Partridge
(2009) A distinct profile of myogenic regula- TA (1995) Culturing satellite cells from living
tory factor detection within Pax7+ cells at S single muscle fiber explants. In Vitro Cell Dev
phase supports a unique role of Myf5 during Biol Anim 31:773–779
posthatch chicken myogenesis. Dev Dyn 38. Rosenblatt JD, Parry DJ, Partridge TA (1996)
238:1001–1009 Phenotype of adult mouse muscle myoblasts
25. Yablonka-Reuveni Z, Quinn LS, Nameroff M reflects their fiber type of origin. Differentiation
(1987) Isolation and clonal analysis of satellite 60:39–45
cells from chicken pectoralis muscle. Dev Biol 39. Shefer G, Wleklinski-Lee M, Yablonka-Reuveni
119:252–259 Z (2004) Skeletal muscle satellite cells can
26. Kastner S, Elias MC, Rivera AJ, Yablonka- spontaneously enter an alternative mesenchy-
Reuveni Z (2000) Gene expression patterns of mal pathway. J Cell Sci 117:5393–5404
the fibroblast growth factors and their recep- 40. Kuang S, Kuroda K, Le Grand F, Rudnicki MA
tors during myogenesis of rat satellite cells. J (2007) Asymmetric self-renewal and commit-
Histochem Cytochem 48:1079–1096 ment of satellite stem cells in muscle. Cell
27. Yablonka-Reuveni Z (2004) Isolation and cul- 129:999–1010
ture of myogenic stem cells. In: Lanza R, Blau 41. Yablonka-Reuveni Z, Seger R, Rivera AJ
D, Melton D, Moore M, Thomas ED, Verfaillie (1999) Fibroblast growth factor promotes
C, Weissman I, West M (eds) Handbook of recruitment of skeletal muscle satellite cells in
stem cells—vol 2: adult and fetal stem cells. young and old rats. J Histochem Cytochem
Elsevier, San Diego 47:23–42
28. Ieronimakis N, Balasundaram G, Rainey S, 42. Shefer G, Partridge TA, Heslop L, Gross JG,
Srirangam K, Yablonka-Reuveni Z, Reyes M Oron U, Halevy O (2002) Low-energy laser
(2010) Absence of CD34 on murine skeletal mus- irradiation promotes the survival and cell cycle
cle satellite cells marks a reversible state of activa- entry of skeletal muscle satellite cells. J Cell Sci
tion during acute injury. PLoS One 5:e10920 115:1461–1469
29. Danoviz ME, Yablonka-Reuveni Z (2012) 43. Wozniak AC, Pilipowicz O, Yablonka-Reuveni
Skeletal muscle satellite cells: background and Z, Greenway S, Craven S, Scott E, Anderson
methods for isolation and analysis in a primary JE (2003) C-Met expression and mechanical
culture system. Methods Mol Biol activation of satellite cells on cultured muscle
798:21–52. fibers. J Histochem Cytochem 51:1437–1445
30. Shefer G, Yablonka-Reuveni Z (2005) Isolation 44. Greene EC (1963) Anatomy of the rat. Hafner
and culture of skeletal muscle myofibers as a Publishing Company, New York, NY
means to analyze satellite cells. Methods Mol 45. Kleinman HK, McGarvey ML, Liotta LA,
Biol 290:281–304 Robey PG, Tryggvason K, Martin GR (1982)
31. Bekoff A, Betz W (1977) Properties of isolated Isolation and characterization of type IV pro-
adult rat muscle fibres maintained in tissue cul- collagen, laminin, and heparan sulfate proteo-
ture. J Physiol 271:537–547 glycan from the EHS sarcoma. Biochemistry
32. Bischoff R (1986) Proliferation of muscle sat- 21:6188–6193
ellite cells on intact myofibers in culture. Dev 46. Yablonka-Reuveni Z, Seifert RA (1993)
Biol 115:129–139 Proliferation of chicken myoblasts is regulated
468 P. Keire et al.
by specific isoforms of platelet-derived growth 48. O’Neill MC, Stockdale FE (1972) A kinetic
factor: evidence for differences between analysis of myogenesis in vitro. J Cell Biol
myoblasts from mid and late stages of embryo- 52:52–65
genesis. Dev Biol 156:307–318 49. Stuelsatz P, Keire P, Almuly R, Yablonka-
47. Yablonka-Reuveni Z (1995) Myogenesis in the Reuveni Z (2012) A contemporary atlas of the
chicken: the onset of differentiation of adult mouse diaphragm: myogenicity, vascularity
myoblasts is influenced by tissue factors. Basic and the Pax3 connection. J Histochem
Appl Myol 5:33–42 Cytochem 60:638–657
Chapter 29
Hepatic Differentiation of Embryonic Stem Cells

by Murine Fetal Liver Mesenchymal Cells
Takamichi Ishii, Kentaro Yasuchika, and Iwao Ikai
Abstract
Hepatocytes derived from embryonic stem cells (ESCs) are a potential cell source for regenerative medi-
cine. However, it has been technically difficult to differentiate ESCs into mature hepatocytes because the
definitive growth factors and molecular mechanisms governing hepatocyte differentiation have not yet
been well defined. The CD45−CD49f+/−Thy1+gp38+ mesenchymal cells that reside in murine fetal livers
induce hepatic progenitor cells to differentiate into mature hepatocytes by direct cell–cell contact. Utilizing
these cells, we employ a two-step procedure for hepatic maturation of ESCs: first, ESCs are differentiated
into endodermal cells or hepatic progenitor cells, and second, ESC-derived endodermal cells are matured
into functional hepatocytes by coculture with murine fetal liver mesenchymal cells. The ESC-derived hepa-
tocyte-like cells possess hepatic functions, including ammonia removal activity, albumin secretion ability,
glycogen synthesis and storage, and cytochrome P450 enzymatic activity.
Key words: Embryonic stem cell, Fetal liver, Hepatocyte, Hepatic progenitor cell, Mesenchymal cell,
Thy1, gp38
1. Introduction
Embryonic stem cells (ESCs) are established from inner cell masses
and possess the pluripotent potential to differentiate into all three
germ layers. Hepatocytes derived from ESCs are anticipated as a
cell source for cell transplantation, bio-artificial livers, and drug
discovery support systems. However, there have been difficulties
differentiating ESCs into mature functional hepatocytes because
the molecular mechanisms that underlie hepatic development are
largely unknown.
Our previous study revealed that the hepatic maturation of
fetal hepatic progenitor cells is greatly facilitated by mesenchymal
469
470 T. Ishii et al.
cells that reside in the fetal livers (1). These mesenchymal cells are
fractionized as CD45−CD49f+/−Thy1+gp38+ cells (2). In addition,
our further experiments demonstrated the ability of these cells to
induce maturation of murine and human ESCs into functional
hepatocytes (3, 4). The effect of the CD45−CD49f+/−Thy1+gp38+
mesenchymal cells on hepatic maturation is achieved by direct cell–
cell contact (5). They do not induce hepatic maturation of undif-
ferentiated ESCs, suggesting that mesenchymal cells are effective
in hepatic maturation of immature endodermal cells, but are rela-
tively ineffective at hepatic specification and differentiation of
undifferentiated ESCs (4).
In this chapter, we describe a two-step procedure for the hepatic
maturation of mouse ESCs utilizing the CD45−CD49f+/−Thy1+gp38+
mesenchymal cells based on their biological characteristics. First, undif-
ferentiated ESCs are differentiated into endodermal cells of the hepatic
lineage using several growth factors and extracellular matrix. Second,
the ESC-derived endodermal cells are matured into functional hepato-
cyte-like cells by coculture with CD45−CD49f+/−Thy1+gp38+ mesen-
chymal cells.
2. Materials
2.1. Culture of Mouse 1. A murine ESC line (see Note 1).

ESCs 2. ESC culture medium: Dulbecco’s modified Eagle’s medium
(DMEM) supplemented with 20% fetal bovine serum (FBS,
HyClone, Logan, UT), 0.1 mM 2-mercaptoethanol, nones-
sential amino acids, 1 mM sodium pyruvate, and 1,000 U/ml
leukemia inhibitory factor (LIF, ESGRO, Chemicon
International Inc., Temecula, CA) (see Note 2). A stock solu-
tion of 1 × 107 U/ml LIF is stored at 4°C.
3. A solution of 0.25% trypsin and 1 mM ethylenediaminetetraa-
cetic acid (EDTA).
4. 35 mm plastic culture dishes with a mouse embryonic fibroblast
(MEF) feeder layer treated with 10 μg/ml mitomycin C for
2 h (see Note 3).
2.2. Differentiation of 1. Serum-free endoderm differentiation medium (SFE medium):

ESCs into Endoderm DMEM supplemented with 10% Knockout SR (Gibco, Grand
Island, NY), 2 mM L-glutamine, 1 mM sodium pyruvate, and
penicillin/streptomycin (50 units/ml each).
2. All-trans retinoic acid is dissolved at 10 mM in 99.5% ethanol,
and stored in aliquots at −80°C. LIF is dissolved at 1 × 106 U/
ml in culture medium, and stored at 4°C. Basic fibroblast
growth factor (bFGF) and hepatocyte growth factor (HGF) are
dissolved to a concentration of 20 μM in phosphate-buffered
29 Hepatic Differentiation of ESCs by Murine Liver Mesenchymal Cell 471
saline (PBS, Ca2+-free) supplemented with 0.5% bovine serum

albumin (BSA), and stored in aliquots at −80°C. These growth
factors are added to culture dishes as required (see Note 4).
3. 60 mm culture dishes coated with type I collagen (pre-coated
dishes purchased from BD Biosciences, Franklin Lakes, NJ).
2.3. Primary Culture 1. A stereomicroscope system.

of Murine Fetal 2. A set of sterilized surgical instruments, including scissors,
Liver Cells micro forceps, and a surgical knife.
3. A pair of sterilized surgical gloves.
4. Two pregnant C57/BL6 mice at day 13.5 of gestation (see
Note 5).
5. HBSS-based buffer: Ca2+-free Mg+-free Hank’s balanced salt
solution with phenol red (HBSS (−)) with 10 mM HEPES and
0.5 mM EDTA.
6. Irrigation solution 1 (50 ml): HBSS-based buffer (45 ml) sup-
plemented with 10% FBS (5 ml), and 2 U/ml heparin sodium
solution (0.1 ml). Heparin sodium solution at 1,000 U/ml is
readily purchased from several pharmaceutical companies. This
solution is prepared as required and kept at 4°C (see Note 6).
7. Irrigation solution 2 (50 ml): HBSS-based buffer (45 ml) sup-
plemented with 50 mg/ml DNase I (1 ml of stock solution),
and 2 U/ml heparin sodium (0.1 ml). DNase I is dissolved at
25 mg/ml in distilled water, and stored in single-use aliquots
at −20°C. This solution is prepared as required, and kept at
4°C (see Note 7).
8. Digestion medium (30 ml): 0.5% (w/v) collagenase type II
(Gibco) is dissolved in 30 ml collagenase buffer and 0.1 ml
heparin sodium. This medium is kept at 37°C prior to use.
Collagenase buffer contains 0.2 g MgSO4·7H2O, 0.735 g
CaCl2·2H2O, 2.383 g HEPES, and 0.05 g trypsin inhibitor in
1 L HBSS (−). This buffer can be preserved at 4°C for a
month.
9. Hepatocyte differentiation medium (HD medium): DMEM
with 10% FBS, 1 mM sodium pyruvate, penicillin/streptomy-
cin (50 units/ml each), 10 mM nicotinamide, 2 mM L-ascorbic
acid phosphate, insulin–transferrin–selenium supplement mix-
ture (Gibco), 1 × 10−7 M dexamethasone, 20 ng/ml HGF, and
10 ng/ml oncostatin M. Oncostatin M is dissolved at 10 mg/
ml in 0.5% BSA/PBS, and stored at −80°C. This medium can
be preserved at 4°C for a month, and should be kept at 37°C
before use.
10. Autoclaved nylon meshes with 50 μm and 100 μm pore size.
11. 100 mm and 35 mm Petri dishes (BD Biosciences).
472 T. Ishii et al.
12. 6-well culture plates coated with type I collagen (pre-coated

culture plates purchased from BD Biosciences).
13. A water bath at 37°C.
2.4. Isolation of Murine 1. Solution of 0.25% trypsin–EDTA.

Fetal Liver 2. PBS (30 ml) with 3% FBS (1 ml) (3%FBS/PBS).
Mesenchymal Cells
3. The following antibodies are used:
Using Flow Cytometry
(a) Anti-CD45-PE (clone 30-F11, diluted at 1:100, BD
Biosciences).
(b) Anti-CD49f-PE (clone GoH3, 3:100, BD Biosciences).
(c) Anti-Thy1-FITC (clone 30-H12, 1:100, BD Biosciences).
(d) Anti-gp38 (Podoplanin, 1:100, Medical and Biological
Laboratories, Nagoya, Japan).
The anti-gp38 antibody is conjugated with APC using a
conjugation kit (see Note 8).
4. 5 ml round-bottom tubes with 35 μm nylon meshes.
5. FACSVantage SE (BD Biosciences).
6. HD medium (see Subheading 2.3, item 9).
7. 24-well culture plates coated with type I collagen (pre-coated
plates from BD Biosciences).
2.5. Coculture of 1. A 24-well culture plate coated with type I collagen with murine
ESC-Derived fetal liver mesenchymal cells as a feeder layer.
Endodermal Cells and 2. HD medium (see Subheading 2.3, item 9).
Murine Fetal Liver
Mesenchymal Cells
3. Methods
3.1. Culture of ESC Mouse ESCs are cultured in the undifferentiated state on MEF feeder
layers (6, 7). They are subcultured using 0.25% trypsin/EDTA solu-
tion. MEFs are prepared according to standard protocols.
3.2. Differentiation of 1. Undifferentiated mouse ESCs that are maintained on 60 mm

ESCs into Endodermal culture dishes are washed twice with 4 ml PBS. They are then
Cells incubated with 2 ml 0.25% trypsin/EDTA solution at 37°C
for 2 min, and then 2 ml of ESC medium is added and the cells
are pipetted well to dissociate the cells.
2. Centrifuge the cells at 180 × g for 3 min, remove the superna-
tant, and suspend the cell pellet in 4 ml ESC medium.
3. In order to deplete the MEFs, the dissociated cells are trans-
ferred onto a plastic culture dish in ESC culture medium, and
incubated at 37°C for 15–30 min.
4. The cell suspension is harvested carefully to avoid removal of

the adherent cells and centrifuged at 180 × g for 3 min. The cell
pellet is suspended and plated at a concentration of 1 × 105 cells/
ml on 60 mm culture dishes coated with type I collagen in SFE
medium (see Note 9).
5. 10 μM ATRA and 1,000 U/ml LIF are added to the SFE
medium for the first 2 days (days 0–1), and 20 ng/ml HGF
and 20 ng/ml bFGF are added for the next 5 days (days 2–6).
Culture medium should be changed every other day at least.
6. At day 7, the cultured cells are dissociated using a 0.25%
trypsin/EDTA solution (as in Subheading 3.2, step 1) and
suspended with HD medium for further experiments (see
Subheading 3.5) (see Note 10).
3.3. Primary Culture 1. Two timed-pregnant mice are sacrificed according to institu-
of Murine Fetal tional guidelines. All uteri are removed and placed into a 100 mm
Liver Cells Petri dish with cold irrigation solution 1. Amniotic membranes
and placentae are removed, and fetal mice are transferred to a
new 100 mm Petri dish with cold irrigation solution 1.
2. The liver tissues are dissected under a stereomicroscope
(Fig. 1), and placed into a new 100 mm Petri dish with cold
irrigation solution 1. The harvested livers are then minced into
pieces no larger than 1 mm in diameter with a surgical knife.
3. A nylon mesh (50 μm pore size) is placed on a 50 ml centrifuge
tube. The minced liver tissues are filtered through this mesh. The
mesh is inverted and carefully transferred onto a new 50 ml cen-
trifuge tube (see Note 11). The flow-through can be discarded.
Fig. 1. This photograph shows a mouse fetus after removal of the amniotic membrane and
placenta. The fetal liver is a red organ located in the middle of a fetus. Under a stereomi-
croscope, the liver is dissected using micro forceps. The gallbladder and intestinal tract
should be removed from the liver.
474 T. Ishii et al.
4. Warm digestion medium (30 ml) is added through the inverted

mesh into the 50 ml centrifuge tube, collecting the liver tissues
into the tube together with the digestion medium. The tube is
incubated in a water bath at 37°C for 12–15 min with agitation.
5. Four nylon meshes (100 μm pore size) are placed onto four
15 ml centrifuge tubes. The digested tissues are divided into
four equal aliquots, filtered through meshes using a pipette,
collected in the centrifuge tubes, and then centrifuged at 10 × g
for 5 min (see Note 12).
6. The cell pellet is suspended in 25 ml irrigation solution 2, col-
lected in a 15 ml centrifuge tube, and then centrifuged at 10 × g
for 5 min. This procedure is repeated three times in total.
7. The cell pellet is suspended in HD medium at a density of
5 × 105 cells/ml to 1 × 106 cells/ml, and inoculated onto
35 mm Petri dishes (see Note 13).
8. The dissociated cells are incubated at 37°C, 5%CO2 overnight.
Cell aggregates are collected in a 15 ml tube and subjected to
gravity sedimentation for 10 min (see Note 14).
9. After the supernatant is removed with a pipette, the sedimented
cell aggregates are suspended in new HD media and plated on
6-well culture plates coated with type I collagen (see Note 15).
10. The cell aggregates are cultured at 37°C, 5%CO2 for 1–2 days.
The culture media are changed every day. The cell aggregates
adhere to the culture plates and grow as monolayer colonies.
3.4. Preparation of A key step in this procedure is to obtain a pure and viable popula-
CD45−CD49f+/−- tion of mesenchymal cells from murine fetal livers.
Thy1+gp38+ 1. Following 1–2 days of culture, the adherent cells are washed
Mesenchymal Cells twice with 500 μl PBS and are then incubated with 200 μl
as a Feeder Layer 0.25% trypsin/EDTA at 37°C for 10 min. HD medium is
added to stop trypsin activity, and all of the cell suspension is
collected in a 15 ml tube.
2. The collected cells are centrifuged at 180 × g for 3 min, and
then washed with 5 ml 3% FBS/PBS twice by centrifuging at
180 × g for 3 min.
3. The cell pellet is suspended in 200 μl 3% FBS/PBS, and trans-
ferred into a 1.5 ml tube. The dissociated cells are incubated
with 2 μl CD45-PE (1:100 dilution), 6 μl CD49f-PE (3:100),
2 μl Thy1-FITC (1:100), and 2 μl gp38-APC (1:100) anti-
bodies on ice in the dark for 30 min.
4. The cells are centrifuged at 630 × g for 2 min, and then the
supernatant is discarded.
5. The cells are washed with 500 μl 3% FBS/PBS three times.
6. The cells are resuspended in 2–4 ml 3% FBS/PBS and col-
lected in a 5 ml round-bottom tube through a nylon mesh
(35 μm pore size).
a Cell aggregates from fetal livers b CD45−CD49f±Thy1−mesenchymal cells

104 104
CD45+Thy1− CD45−CD49f±Thy1+gp38+
hematopoietic cells mesenchymal cells
CD45-PE / CD49f-PE
103 103
gp38-APC
CD45−CD49f±Thy1+
mesenchymal cells
102 102 CD45−CD49f±Thy1+gp38−
mesenchymal cells
101 101
CD45−CD49f+dimThy1−
hepatic progenitor cells
100 100
100 101 102 103 104 100 101 102 103 104
Thy1-FITC Thy1-FITC
Fig. 2. Dot plots following flow cytometric analyses. (a) Cell aggregates derived from murine fetal livers are mainly divided
by CD45, CD49f, and Thy1 into three cell fractions. The CD45+Thy1− fraction corresponds to hematopoietic cells, the
CD45−CD49f+dimThy1− cell fraction corresponds to hepatic progenitor cells, and the CD45−CD49f+/−Thy1+ cell fraction is
mesenchymal cells. (b) The CD45−CD49f+/−Thy1+ mesenchymal cells are further fractionated by gp38 into two groups. The
gp38-positive cells account for approximately 16% of the CD45−CD49f+/−Thy1+ mesenchymal cells.
7. The CD45−CD49f+/−Thy1+gp38+ cell fraction is separated

using an FACSVantage SE. Dot plots using CD45, CD49f,
Thy1, and gp38 antibodies are shown in Fig. 2 (see Note 16).
The separated CD45−CD49f+/−Thy1+gp38+ mesenchymal cells
are collected in HD medium.
8. The collected cells are suspended in HD medium and seeded
in 24-well culture plates coated with collagen type I at a den-
sity of 1 × 104 cells per well (see Note 17).
9. The CD45−CD49f+/−Thy1+gp38+ mesenchymal cells are grown
to approximately 80% confluency and treated with 10 μg/ml
mitomycin C for 2 h. After washing twice in PBS, fresh HD
medium is added. The inactivated cells can be used from the
next day.
3.5. Maturation of 1. The dissociated endoderm cells derived from mouse ESCs,
ESC-Derived generated in Subheading 3.2, are inoculated on a feeder layer
Endodermal Cells of CD45−CD49f+/−Thy1+gp38+ mesenchymal cells at a density
of 1 × 104 cells/well (see Note 18).
2. The ESC-derived endoderm cells are cultured in HD medium
on the CD45−CD49f+/−Thy1+gp38+ mesenchymal feeder layer
for 7–14 days. Culture media are changed every day.
3. These ESC-derived mature hepatocyte-like cells can be used
for further analyses including drug metabolism and albumin
secretion.
476 T. Ishii et al.
4. Notes
1. In this protocol, a murine ESC line derived from C57BL6

mice is used. The passage number is less than 50. Our protocol
works with a human ESC line as well (4).
2. Because the ESC culture medium contains LIF, it should be
used within a month.
3. Culture dishes with an MEF-feeder layer can be used for 1 week
after mitomycin C treatment.
4. Unless stated otherwise, growth factors are added from stock
solutions to culture medium as required.
5. The number of pregnant mice can be increased to four mice
per experiment. In this case, the described protocol can be
scaled up. However, it might be technically difficult to handle
more than five mice at one time since it may be more time-
consuming, decreasing the viability of harvested liver cells.
6. Heparin is added in order to prevent clot formation.
7. DNase is added in order to reduce viscosity caused by DNA
that is released from damaged cells.
8. The combination of the fluorescent dyes is actually atypical,
because PE labels both anti-CD45 and anti-CD49f antibodies.
However, as shown in Fig. 2, the CD45-positive cell fraction is
clearly distinguishable from the CD49f-positive cell fractions
based on their fluorescent intensities (1). This may result from
the difference in the expression levels between CD45 and
CD49f antigens.
9. Usually two to four 60 mm collagen type I-coated dishes can
be harvested from undifferentiated ESCs on one confluent
60 mm dish. The concentration of ESCs seeded on culture
plates is important. Under this condition, differentiating ESCs
can be cultured for 7 days without further passages. During this
period, culture medium should be changed every other day.
10. This protocol makes it possible to obtain alpha-fetoprotein
(AFP)-producing endodermal cells at an efficiency of more
than 40% at day 7. However, the optimal condition for endo-
dermal differentiation may vary widely with the type of ESCs.
For example, our recent study revealed that ATRA does not
induce hepatic differentiation in human ESCs, and that
Matrigel (BD Biosciences) is more efficient for hepatic differ-
entiation than Type I Collagen (8). Therefore, culture proto-
cols may require optimization depending on the ESC lines.
11. Because fetal livers act as hematopoietic organs during embry-
onic stages, the harvested liver tissues contain a large number
of hematopoietic cells. This procedure is necessary in order to

eliminate hematopoietic cells. The flow-through contains
hematopoietic cells.
12. This procedure is performed to eliminate undigested liver
tissues.
13. The harvested liver cells obtained from one pregnant mouse
can usually be seeded onto three 35 mm Petri dishes, although
the yield depends on the number of fetuses.
14. The cell aggregates can be formed in a few hours. The cell
aggregates consist of hepatic progenitor cells, hematopoietic
cells, and mesenchymal cells. Forming cell aggregates can
enrich hepatic progenitor cells and mesenchymal cells, and
greatly facilitate the purification of mesenchymal cells using
flow cytometry (9).
15. The cell aggregates cultured in one 35 mm Petri dish can be
usually transferred to one well of a 6-well culture plate.
16. This protocol can be also used to separate hepatic progenitor
cells that reside in fetal murine livers. The CD45−CD49f+/−Thy1−
cell fraction corresponds to hepatic progenitor cells. In this
case, it is best to set a small gate for the CD45−CD49f+/−Thy1−
fraction in order to obtain a pure fraction of hepatic progenitor
cells.
17. Cell viability is remarkably decreased at a lower density of cul-
tured cells. Eventually, 1.5–2.0 × 104
− +/− + +
CD45 CD49f Thy1 gp38 mesenchymal cells can be isolated
from one pregnant mouse.
18. The ESC-derived endodermal cells consist of not only endo-
dermal cells, but also ectodermal and mesodermal cells.
Therefore, experimental circumstances may require purification
of only endodermal cells. In these cases, it may be helpful to
generate transgenic ESCs that express fluorescent proteins
driven by an endoderm-specific gene (e.g., AFP or albumin)
promoter by gene manipulation (3, 10).
Acknowledgments
The authors would like to thank Prof. Norio Nakatsuji (Institute

for Integrated Cell-Material Science, Kyoto University), Dr.
Hirofumi Suemori (Institute for Frontier Medical Sciences, Kyoto
University), and Prof. Shinji Uemoto (Graduate School of Medicine
Kyoto University) for supporting the study. This work was sup-
ported in part by grants from the Scientific Research Fund of
Ministry of Education, Culture, Sports, Science, and Technology
of Japan.
478 T. Ishii et al.
References
1. Hoppo T, Fujii H, Hirose T, Yasuchika K, mouse embryonic stem cells by a coculture
Azuma H, Baba S, Naito M, Machimoto T, method. Tissue Eng Part A 15:3847–3856
Ikai I (2004) Thy1-positive mesenchymal cells 6. Sumi T, Tsuneyoshi N, Nakatsuji N, Suemori
promote the maturation of CD49f-positive H (2007) Apoptosis and differentiation of
hepatic progenitor cells in the mouse fetal liver. human embryonic stem cells induced by sus-
Hepatology 39:1362–1370 tained activation of c-Myc. Oncogene
2. Kamo N, Yasuchika K, Fujii H, Hoppo T, 26:5564–5576
Machimoto T, Ishii T, Fujita N, Tsuruo T, 7. Suemori H, Yasuchika K, Hasegawa K, Fujioka
Yamashita JK, Kubo H, Ikai I (2007) Two T, Tsuneyoshi N, Nakatsuji N (2006) Efficient
populations of Thy1-positive mesenchymal establishment of human embryonic stem cell
cells regulate in vitro maturation of hepatic lines and long-term maintenance with stable
progenitor cells. Am J Physiol Gastrointest karyotype by enzymatic bulk passage. Biochem
Liver Physiol 292:G526–G534 Biophys Res Commun 345:926–932
3. Ishii T, Yasuchika K, Fujii H, Hoppo T, Baba S, 8. Ishii T, Fukumitsu K, Yasuchika K, Adachi K,
Naito M, Machimoto T, Kamo N, Suemori H, Kawase E, Suemori H, Nakatsuji N, Ikai I,
Nakatsuji N, Ikai I (2005) In vitro differentia- Uemoto S (2008) Effects of extracellular
tion and maturation of mouse embryonic stem matrixes and growth factors on the hepatic dif-
cells into hepatocytes. Exp Cell Res 309:68–77 ferentiation of human embryonic stem cells.
4. Ishii T, Yasuchika K, Fukumitsu K, Kawamoto Am J Physiol Gastrointest Liver Physiol
T, Kawamura-Saitoh M, Amagai Y, Ikai I, 295:G313–G321
Uemoto S, Kawase E, Suemori H, Nakatsuji N 9. Yasuchika K, Hirose T, Fujii H, Oe S, Hasegawa
(2010) In vitro hepatic maturation of human K, Fujikawa T, Azuma H, Yamaoka Y (2002)
embryonic stem cells by using a mesenchymal Establishment of a highly efficient gene trans-
cell line derived from murine fetal livers. Cell fer system for mouse fetal hepatic progenitor
Tissue Res 339:505–512 cells. Hepatology 36:1488–1497
5. Fukumitsu K, Ishii T, Yasuchika K, Amagai Y, 10. Ishii T, Yasuchika K, Machimoto T, Kamo N,
Kawamura-Saito M, Kawamoto T, Kawase E, Komori J, Konishi S, Suemori H, Nakatsuji N,
Suemori H, Nakatsuji N, Ikai I, Uemoto S Saito M, Kohno K, Uemoto S, Ikai I (2007)
(2009) Establishment of a cell line derived Transplantation of embryonic stem cell-derived
from a mouse fetal liver that has the character- endodermal cells into mice with induced lethal
istic to promote the hepatic maturation of liver damage. Stem Cells 25:3252–3260
Chapter 30
Methods to Culture, Differentiate, and Characterize Neural

Stem Cells from the Adult and Embryonic Mouse Central
Nervous System
Sharon A. Louis, Carmen K.H. Mak, and Brent A. Reynolds
Abstract
Since the discovery of neural stem cells (NSC) in the embryonic and adult mammalian central nervous
system (CNS), there have been a growing numbers of tissue culture media and protocols to study and
functionally characterize NSCs and its progeny in vitro. One of these culture systems introduced in 1992
is referred to as the Neurosphere Assay, and it has been widely used to isolate, expand, differentiate and
even quantify NSC populations. Several years later because its application as a quantitative in vitro assay for
measuring NSC frequency was limited, a new single-step semisolid based assay, the Neural Colony Forming
Cell (NCFC) assay was developed to accurately measure NSC numbers. The NCFC assay allows the dis-
crimination between NSCs and progenitors by the size of colonies they produce (i.e., their proliferative
potential). The evolution and continued improvements made to these tissue culture tools will facilitate
further advances in the promising application of NSCs for therapeutic use.
Key words: Murine, Embryonic neural stem cells, Subventricular zone, Neurospheres, Differentiation,
CNS, Culture, Stem cells
1. Introduction
In the mammalian central nervous system (CNS), neural stem cells

(NSCs) have been isolated from both embryonic and adult tissues
(1–5). While NSCs can be isolated from multiple regions of the
developing embryo (5), there is clear evidence now that in the
adult brain of both rodents and primates cell proliferation occurs in
two specific regions: within the subventricular zone (SVZ) and the
subgranular zone (SGZ) of the dentate gyrus of the hippocampus
(6, 7). In these restricted regions of the adult brain, it is believed
that NSC continues to generate new neurons and glia to areas such
479
480 S.A. Louis et al.
as the olfactory bulb, the corpus callosum and the hippocampus.

The functional characterization of putative NSC often relies on
in vitro culture systems which provide a retrospective readout of
their function and frequencies. The neurosphere culture system
(1), provided for the first time, a relatively simple and robust means
to characterize NSC and investigate the regulation of NSCs. The
neurosphere culture system can be used to demonstrate the cardi-
nal stem cell properties of: (a) self-renewal over an extended period
of time, (b) generation of a large number of progeny, and (c) multi-
lineage differentiation potential and remains the most frequently
adopted method to enrich, expand, and differentiate NSCs (1–3).
However, since its original discovery, many modifications to the
neurosphere culture media and protocols have been introduced
into the neural stem cell field by individual labs. For example, sev-
eral labs are using adherent culture condition instead of the neuro-
sphere suspension culture system and reported more pure
populations of undifferentiated cells in these cultures (8, 9).
However, the modifications have not been standardized across the
field, and has lead to variations in protocols and media being used
by different labs and thus contributing to the discrepancies in the
results obtained within labs and between labs. As it is crucial that
the in vitro methodologies designed to study NCS be robust, accu-
rate and the user clearly understands what the methodologies are
purported to measure, this chapter attempts to provide complete
validated protocols, assays and improved reagents based on the
original Reynolds and Weiss methodology (10).
To generate neurospheres, tissue from the embryonic (mouse,
rat, human) or adult (mouse, human) CNS is microdissected,
dissociated, and plated in a defined serum-free medium in the pres-
ence of a mitogen (e.g., EGF ± FGF-b) (Fig. 1a). The NSC (as well
as neural progenitors) begins to proliferate after about 24 h in cul-
ture and form small clusters of cells by 2–3 days. The clusters con-
tinue to grow in size and by day 3–5 the majority of the clusters
detaches from the surface and float in suspension. By day 7 the
clusters, called neurospheres, typically measure 100–200 μm in
diameter (Fig. 1b), are composed of approximately 10,000 cells
and are ready to be dissociated and passaged or subcultured.
Neurospheres are individually, or as a population, dissociated into
a single cell suspension and replated under the same conditions as
the primary culture. The neural stem (and even progenitor) cells
begin to proliferate within the first 24 h and again form new clus-
ters that are ready to be dissociated and passaged again 5–7 days
later (Fig. 1d). NSC will self-renew and proliferate resulting in a
relatively consistent arithmetic increase in total cell number. Cells
from neurospheres derived from embryonic mouse CNS tissue
treated in this manner have been passaged for up to 10 weeks with
no loss in their proliferative ability, resulting in a 107-fold increase
in total cell number. Cells within the neurospheres can be induced
30 Growth and Differentiation of Mouse Neural Stem Cells 481
Fig. 1. (a) Cells isolated from the E14 cortex, were grown for 7 days in culture and then
passaged. Two days after passaging small clusters of cells can be identified. (b)Two
spheres from (a) are enlarged showing the appearance of microspikes (arrow heads) that
are commonly seen on young healthy neurospheres. (c) By 4 days in vitro (DIV) neuro-
spheres have grown in size, detached from the substrate and float in suspension. (d) A
floating 7 DIV neurosphere. Magnification: (a), (c) & (d) ×200; (b) ×400.
to differentiate by removal of the mitogens EGF (and FGF) and

plating either as intact neurospheres or dissociated cells on an
adhesive substrate in a low serum-containing medium. After sev-
eral days, virtually all of the stem cell progeny will differentiate into
the three primary cells types found in the CNS: neurons, astrocytes
and oligodendrocytes (Fig. 2).
Fig. 2. (a) Neurospheres were differentiated as detailed in Subheading 3.5. Phase contrast
micrograph shows cells with processes (mostly neurons and oligodendrocytes) sitting on
top of layer of astrocytes. (b) Neurons were identified with a fluorescent label antibody
raised against Beta-tubulin (a neuron specific antigen found in cell bodies and processes).
A large number of positive labeled cells with a neuronal morphology can be identified. (c)
Both protoplasmic and stellate astrocytes are identified with a fluorescent tagged anti-
body against the astrocyte specific protein GFAP. (d) Three oligodendrocytes (arrows)
labeled with an antibody against myelin basic protein (MBP).
While the neurosphere culture system has been the method of

choice for the isolation, expansion and even the enumeration of
neural stem and progenitor cells, several recent publications have
highlighted some of the limitations of the neurosphere culture sys-
tem as a readout for NSC numbers (11–13). To address some of
the shortcomings of the neurosphere assay, a new assay called the

Neural Colony-Forming Cell (NCFC) assay which is able to more
accurately discriminate between neural stem and progenitor cells
compared to the neurosphere assay was developed (14). In the
NCFC assay colony size is an indicator of proliferative potential and
cells from colonies >2 mm in diameter meet all the functional crite-
ria for a NSC which includes the ability to self-renew, generate large
numbers of progeny, and maintain multipotency over an extended
period of time. Cells which form colonies < 2 mm lack self-renewal
ability and are likely progenitor. The NCFC assay therefore pro-
vides a quantitative readout for NSCs that is more accurate than
measurements of neurosphere number or secondary/tertiary neu-
rosphere formation ability that is routinely practiced in the field.
This chapter describes complete protocols for working with neu-
ral stem cells from the embryonic and adult mouse CNS including
methods for the isolation, expansion, differentiation and enumeration
of NSC. With these complete set of accurate protocols and validated
reagents, we attempt to facilitate standardization in this field.
2. Materials
2.1. Dissection 1. Extra fine spring micro scissors (1) (cat. no. 15396-01, Fine
Equipment Science Tools).
2. Bead Sterilizer (cat. no. 250, Fine Science Tools).
3. Dissecting microscope (e.g. Stemi 2000, 1:7 zoom, Zeiss).
4. Phosphate-buffered saline (PBS) (e.g., cat. no. 37350,
STEMCELL Technologies Inc.) containing 2% glucose, cold,
sterile.
5. Plastic Petri dishes, 100 mm, sterile (4–5 plates to hold uteri,
embryos, heads, and brains) (cat. no. 27125, STEMCELL
Technologies Inc.).
6. Plastic Petri dishes, 35 mm, sterile (4–5 to hold dissected brain
regions) (e.g. cat. no. 21700, STEMCELL Technologies Inc.).
7. Tubes, 17 × 100-mm polystyrene test tubes, sterile (e.g., cat.
no. 2057, Falcon).
2.2. General 1. Biological safety cabinet (e.g., Canadian Cabinets) certified for
Equipment Level II.
2. Low-speed centrifuge (e.g., Beckman TJ-6) equipped with
biohazard containers.
3. Incubator with humidity and gas control to maintain 37°C, >95%
humidity and an atmosphere of 5% CO2 (e.g., Forma 3326).
4. Pipette-aid (e.g., Drummond Scientific).
5. Hemacytometer (e.g., Brightline).
6. Trypan blue (e.g., cat. no. 07050, STEMCELL Technologies

Inc.).
7. Light microscope with 5× and 10× objectives for hemacytom-
eter cell counts.
8. Inverted microscope with flatfield objectives and eyepieces to
give object magnification of approx 20–30×, 80×, and 125×
(e.g., Nikon Diaphot TMD).
9. Pasteur glass pipettes, sterile.
2.3. Tissue Culture 1. T-25 cm2 tissue culture flask (cat. no. 156367, Nunc or cat.
Equipment for no. 15708-130, VWR) or T-162 cm2 Flask (cat. no. 3151,
Neurosphere and Corning).
Adherent Cultures 2. Tubes, 17 × 100-mm polystyrene test tubes, sterile (e.g., cat.
no. 2057, Falcon).
3. Tubes, 50 mL , polypropylene, sterile (e.g., cat. no. Falcon).
4. 24-Well culture dishes (e.g., cat. no. 3526, Corning).
5. Round glass coverslips (cat no.633029, Carolina) sterilized by
autoclaving.
2.4. Media, 1. 30% Glucose (cat. no. G-7021, Sigma). Mix 30 g of glucose
Supplements and in 100 mL of distilled water. Filter-sterilize and store at
Associated Reagents 4°C.
for Neurosphere and 2. 7.5% Sodium bicarbonate (NaHCO3) (cat. no. S-5761, Sigma).
Adherent Cultures Mix 7.5 g of NaHCO3 in 100 mL of distilled water. Filter-
sterilize and store at 4°C.
2.4.1. Proliferation
3. 1 M HEPES solution (cat. no. H-0887, Sigma).
4. 10× Stock solution of DMEM/F12. Mix five 1-L packages
each of DMEM powder high glucose with L-glutamine, minus
sodium pyruvate, minus sodium bicarbonate (cat. no. 12100-
046, Invitrogen) and F12 powder contains L-glutamine, no
sodium bicarbonate (cat. no. 21700-075, Invitrogen) in 1 L
water. Filter-sterilize and store at 4°C.
5. 3 mM Sodium selenite (Na2SeO3) (cat. no. S-9133, Sigma).
Mix 1 mg of Na2SeO3 with 1.93 mL of distilled water. Filter-
sterilize and store at −20°C.
6. 2 mM Progesterone (cat. no. P-6149, Sigma). Inject 1.59 mL
of 95% ethanol to a 1-mg stock of progesterone and mix.
Aliquot into sterile tubes and store at −20°C.
7. Bovine Transferrin Iron Poor (APO) (cat. no. 820056-1,
Serologicals). Dissolve 400 mg of Apo transferrin directly to
the 10× hormone mix.
8. Insulin: Dissolve 100 mg insulin in 4 mL of sterile 0.1 N HCl.
Mix in 36 mL of distilled water and add entire volume directly
to the 10× hormone mix.
9. Putrescine (cat. no. P-7505, Sigma). Dissolve 38.6 mg

putrescine in 40 mL of distilled water and add entire volume
directly to the 10× hormone mix.
10. 200 mM L-Glutamine (e.g., cat. no. 07100, STEMCELL
Technologies Inc.).
11. Basal medium. To prepare 450 mL of basal medium, the indi-
vidual components are added in the following order: 375 mL
of ultrapure distilled water, 50 mL of 10× DMEM/F12, 10 mL
of 30% glucose, 7.5 mL of 7.5% sodium bicarbonate, 2.5 mL
of 1 M HEPES, and 5 mL of 20 mM L-glutamine. Mix com-
ponents well and filter-sterilize (see Note 1).
12. 10× Hormone mix. To prepare 10× hormone mix, the individ-
ual components are added in the following order: 300 mL of
ultrapure distilled water, 40 mL of 10× DMEM/F12, 8 mL of
30% glucose, 6 mL of 7.5% sodium bicarbonate, and 2 mL of
1 M HEPES. Mix components well at this point. The following
components are then added to above mixture in the order
listed: 400 mg of Apo transferrin, 40 mL of 2.5 mg/mL insulin
stock, 40 mL of 10 mg/mL putrescine stock, 40 μL of 3 mM
sodium selenite, and 40 μL of 2 mM progesterone. Mix all
components well and filter-sterilize. Aliquot into 10- or 50-mL
volumes in sterile tubes and store at −20°C (see Note 1).
13. Basal-hormone mix media. This media is prepared as follows:
Thaw an aliquot of the 10× hormone mix (see Subheading 2.4,
items 1–12). Add 50 mL of the 10× hormone mix to 450 mL
of basal medium from item 11 to give a 1:10 dilution. The
hormone-supplemented neural culture media should be stored
at 4°C and used within 1 week.
14. Human recombinant epidermal growth factor (rhEGF) (cat.
no. 02633, STEMCELL Technologies Inc.). A stock solution
of 10 μg/mL of rhEGF is made up in 0.1 mL sterile 10 mM
acetic acid containing at least 0.1% bovine serum albumin, then
adding 19.9 mL of the Basal-Hormone Mix Media (see
Subheading 2.4, items 1–13) and stored as 1-mL aliquots at
−20°C until required for use.
15. Human recombinant fibroblast growth factor (rhFGF) (for adult
mouse CNS cells; cat. no. 02634, STEMCELL Technologies
Inc.). A stock solution of 10 μg/mL of rhFGF is made up in the
Basal-Hormone Mix Media (see Subheading 2.4, items 1–13)
and stored as 1-mL aliquots at −20°C until required for use.
16. 0.2% Heparin (for adult mouse CNS cells). Mix 100 mg of hepa-
rin (cat. no. H-3393, Sigma-Aldrich) in 50 mL of distilled water.
Filters sterilize. Store aliquots of 1 mL at 4°C (see Note 2).
17. “Complete” NSC proliferation medium. Add 2 μL of rhEGF to
every 1 mL of the Basal-Hormone Mix Media from item 13 to
give a final concentration of 20 ng/mL of EGF (see Note 1).
18. “Complete” NSC proliferation medium (adult). Add 2 μL of

rhEGF, 1 μL of rhFGF and 1 μL of heparin to every 1 mL of
the Basal-Hormone Mix Media (see Subheading 2.4, items
1–13) to give a final concentration of 20 ng/mL of EGF,
10 ng/mL bFGF and 0.0002% heparin (see Note 1).
19. Poly-D-lysine (cat. no. P7280, Sigma) Dissolve 50 mg of poly-
D-lysine (PDL) in 50 mL of autoclaved water to yield a 100 μg/
mL stock solution. Aliquot diluted PDL solution in polypro-
pylene vials and store at −20°C. Do not perform freeze/thaw
cycles more than twice.
20. Laminin (Cat. No. L2020, Sigma). Thaw laminin at 4°C to
prevent laminin from gelling. Dilute laminin to 15 μg/ mL
with sterile phosphate buffered saline or autoclaved water.
21. Poly-d-lysine/laminin coated glass coverslips. Using a sterile for-
ceps, place an autoclaved round coverslip at the bottom of an
individual well of a 24-well plate. Dispense 1 mL of the 100 μg/
mL PDL solution into each well containing the autoclaved cov-
erslips. Ensure that the coverslips are completely submerged in
the PDL solution as the coverslips tend to float. If it happens, use
a plastic disposable tip to push the coverslip to the bottom of the
well. Incubate the coverslips with the PDL solution for 2 h at
room temperature or overnight at 4°C. At the end of the incuba-
tion, wash each well two times with 1 mL of sterile PBS to remove
PDL solution. Remove as much PBS as possible. Next dispense
600 μL of 15 μg/mL laminin onto the PDL-coated coverslips.
Incubate the coverslips with the laminin solution overnight at
4°C. The next day wash each well with sterile PBS to remove any
residual laminin solution. The PDL/laminin coverslips are ready
for use and should be used within the same day.
22. Poly-d-lysine/laminin coated plastic tissue culture wells or flasks.
Dispense sufficient volume of the 100 μg/mL PDL solution
into each well or flasks to be coated. Generally, 1 mL is required
per 6 well or 5 mL per T25 cm2 flask to ensure even coating.
Ensure that the wells or flasks are completely coated with the
PDL solution and that there is no area which is left uncovered by
liquid. Incubate the wells or flasks with PDL solution for 2 h at
room temperature or overnight at 4°C. At the end of the incuba-
tion, wash each well or flask two times with sufficient volume of
sterile PBS to remove PDL solution. Use 3 mL per well for 6
well plates or 10 mL for T25 cm2 flasks. Remove as much PBS as
possible. Next dispense 1 mL per 6-well plate or 5 mL per
T25 cm2 flasks of 15 μg/mL laminin onto the already PDL-
coated well or flasks. Incubate with the laminin solution over-
night at 4°C. The next day wash each well with sterile PBS to
remove any residual laminin solution. The PDL/laminin wells or
flasks are ready for use and should be used within the same day.
23. Accutase™ (cat. no. 07920, STEMCELL Technologies Inc.).
2.4.2. Differentiation 1. Fetal Bovine Serum (cat. no. 06550 STEMCELL Technologies
Inc).
2. “Complete” NSC differentiation medium. Basal-hormone mix
media. This media is prepared as follows: Thaw an aliquot of
the 10× hormone mix (see Subheading 2.4, items 1–12). Add
50 mL of the 10× hormone mix to 450 mL of basal medium
(see Subheading 2.4, items 1–11) to give a 1:10 dilution. The
hormone-supplemented neural culture media should be stored
at 4°C and used within 1 week. Add the appropriate volume of
FBS to the Basal-Hormone Mix Media (see Subheading 2.4,
items 1–13) to give a final concentration of 1–5% depending
on the specific application. For example add 5 mL of fetal
bovine serum to the 500 mL of Basal-Hormone Mix Media.
The complete NSC differentiation medium should be stored at
4°C and used within 1 week (see Note 1).
3. Poly-L-ornithine (cat. no. P3655 Sigma): Dissolve 0.15 mg in
10 mL of PBS to yield a 15 μg/mL stock solution.
4. Poly-L-ornitine coated glass coverslips: Coverslips are sterilized
by autoclaving. Using a sterile forceps, transfer a single glass
coverslip per well into a 24-well plate. Dispense 1 mL of a
15 μg/mL poly-L-ornithine solution into each well and incu-
bate glass coverslips for a minimum of 3 h at 37°C. After the
incubation, remove the poly-L-ornithine solution from each
well by aspiration and rinse each well three times for 15 min
with sterile PBS. The poly-L-ornitine coated glass coverslips
should be used on the day of preparation.
5. Phosphate-buffered saline (PBS) (cat. no. 37350, STEMCELL
Technologies Inc).
6. Matrigel™ (Cat. no. BD catalog number 354277). Matrigel™
has to be diluted to the appropriate concentration and is calcu-
lated for each lot based on the protein concentration. Prepare
aliquots according to the dilution factor provided on the
Certificate of Analysis from the supplier. Aliquot 23 mL of
Basal-Hormone mix (see Subheading 2.4, items 1–13) into a
50 mL tube. Add 1 mL of the Basal-Hormone mix media to a
2 mg Matrigel™ vial, pipette up and down to thaw Matrigel™.
Transfer the diluted Matrigel™ to the 50 mL conical tube con-
taining 23 mL medium. Mix well by gentle pipetting. Aliquots
may be stored at −70°C for up to 6 months. The volume of the
aliquots is typically between 270 and 350 μL.
7. Matrigel™ coated glass coverslips: Matrigel™ can be used to
coat plastic 24-wells and glass coverslips. For coating cover-
slips, aseptically place sterile glass coverslips in a well of a
24-well plate. Add 1 mL of diluted Matrigel™ (see
Subheading 2.4, items 2–6) per 24-well. Allow to coat at room
temperature for 2 h. When ready to use the plate, pipette to
remove the Matrigel™. Cells can be plated directly on to the

Matrigel™ coated coverslips or plates without the requirement
of a wash step of the coated plates.
2.4.3. Reagents for the NeuroCultTM enzymatic dissociation kit for adult CNS tissue (mouse
Dissociation of Adult CNS and rat) (cat. no. 05715, STEMCELL Technologies Inc). The kit is
Tissue composed of four optimized solutions: Tissue Collection Solution,
Dissociation Solution, Inhibition Solution and Resuspension
Solution.
2.5. Media, Note that while we originally developed the media formulation
Supplements and and procedures for the NCFC assay, the reagents to perform this
Tissue Culture assay are now available through a commercial supplier and there-
Equipment for Neural fore media formulations remain proprietary. The Neural Colony
Colony Forming Cell Forming Assay called The Mouse NeuroCultTM NCFC Assay Kit
Assay (Mouse) is currently supplied as a ready-to-use kit from STEMCELL
Technologies Inc cat. no. 05740.
1. NeuroCultTM NCFC Serum-Free Medium without Cytokines
(cat. no. 05720).
2. NeuroCultTM NSC Proliferation Supplements (Mouse) (cat.
no. 05701).
3. NeuroCultTM NSC Basal Medium (Mouse) (cat. no. 05700).
4. Collagen solution (cat. no. 04902).
5. 35 mm Culture dishes 7 packs (10 dishes/pack) (cat. no.
27100).
6. Gridded Scoring Dishes (cat. no. 27500).
7. 40 μm cell strainer (cat. no. 27305).
8. 245 mm square bioassay dish (cat. no. 27130, STEMCELL
Technologies Inc.).
3. Methods
3.1. Establishment and All culture procedures including dissections of CNS regions
Subculture of Primary should be performed in Level II Biosafety Cabinets using asep-
Neurospheres from tic technique and universal safety precautions. The micro-
Embryonic CNS dissection of the CNS tissue from mouse embryos is extensively
Tissues referred (20, 21) and various videos are available online (e.g.,
http://www.stemcell.com), therefore not described in this
chapter. Once dissections are complete, collect all the tissues
(obtained from 12 to 30 embryos) with a glass pipette and
transfer the tissues and collection media (e.g. PBS + 2% glucose
is commonly used) into 14 mL tube and allow tissues to settle.
Pipette off supernatant.
1. Resuspend tissues in 1 mL of Basal-Hormone Mix Media (see

Subheading 2.4, items 1–13).
2. Using a Pasteur glass pipette or a plastic disposable tip attached
to a P1000 micropipetor set at 1 mL; triturate the tissue for
approximately three times (see Note 3).
3. Resuspend cells in a total volume of 10 mL Basal-Hormone
Mix Media.
4. Centrifuge the cells at 800 rpm (110 × g) for 5 min. Remove
supernatant and resuspend the cells with a brief trituration in
2 mL medium.
5. If undissociated tissue remains, allow the suspension to settle
for 1–2 min and then pipette off the supernatant containing
single cells into a fresh tube.
6. Filter single cell suspension over a 40 μm cell strainer.
7. Measure the precise volume and count cell numbers using a dilu-
tion in trypan blue (1/5 or 1/10 dilution) and hemacytometer.
8. For primary cultures, seed cells at a density of
2 × 106 cells per 10 mL or 80,000 cells/cm2 (T-25 cm2 flask) or
8 × 106 cells in 40 mL media (T-162 cm2 flask), in complete
NSC medium.
3.1.1. Subculturing of The procedures outlined below can be performed repeatedly over
Mouse Neurospheres multiple culture passages for neurospheres derived from embry-
Derived from Embryonic onic cells to generate neurosphere cultures for multiple passages.
Mouse CNS Cells If stem cells are maintained during the subculture, they will self-
renew and generate large numbers of progeny over time. Some
progenitors have also be shown to form new neurospheres for up
to four to five subcultures; however, because of the loss of prolif-
erative potential, these cells eventually do not produce more prog-
eny. Neurospheres can be dissociated into a single cell suspension
using mechanical, chemical and enzymatic dissociation techniques
as described below (see Notes 4 and 5). The resulting suspension
of cells can be subsequently be cultured in different sized tissue
cultureware depending on the researchers experimental needs with
protocols to harvest neurospheres from T25 cm2 flasks and subcul-
ture into the same size flasks described here.
1. Observe the neurosphere cultures under a microscope to determine
if the neurospheres are ready for passaging. If neurospheres are
attached to the culture substrate, tapping the culture flask against the
bench top should detach them (Figs. 1 and 3; see Note 6).
2. Using a disposable pipette, remove medium with suspended
neurosphere from a T25 cm2 flask and place in a 14 mL sterile
tissue culture tube. If some cells remain attached to the sub-
strate, detach them by shooting a stream of media across the
attached cells. Spin at 400 rpm (75 × g) for 5 min.
Fig. 3. A comparison of healthy (d–f) and unhealthy (a–c) neurospheres. (a) At least two different types of neurospheres
can be identified in this micrograph—dark dense spheres and light translucent spheres. The dark spheres are unhealthy
and are composed of more dead cells than the lighter colored spheres. (b) An unhealthy sphere. The inset on the upper
right hand corner (arrow) is a higher magnification of the left side of the sphere. Note the irregular surface and a high
proportion of dead cells on the sphere’s periphery. (c) Often when the spheres are past their prime they begin to clump
together as can be seen in this figure. The right side inset is a magnification of a small section of the clumped spheres.
Note the large number of unhealthy or dead cells on the outer edge of the sphere. (d) The neurospheres in this figure are
healthy and ready to be passaged. If left in culture for 2–3 more days the majority of the spheres would become unhealthy.
(e) While these spheres are relatively healthy they are beginning to reach the end of their prime, they should have been
passed a day ago. The largest sphere (at the bottom) is becoming dark in the centre, a sign that a number of cells are dying.
The inset (arrow) represents a higher magnification of the right side of this sphere. Although the centre of the sphere is
becoming dark the outer portion of the sphere is still light and translucent with light, round healthy cells on the outer edge.
(f) The sphere in the centre of this photograph is light in color and translucent—a sign that it is composed primarily of live
healthy cells. A higher magnification of the lower left corner of this sphere (arrow) reveals the absence of dead cells and
abundance of large, healthy round cells. Magnification: (a)–(c) = ×200; (d) = ×400.
3. Remove all the supernatant and resuspend cells in a maximum

of 200 μL of complete NSC medium.
4. With a plastic disposable pipette tip attached to a P200 pipettor
set at ~180 μL, triturate the neurospheres until single cell sus-
pension is achieved (see Note 7).
6. Set up the cells for the next culture passage in complete NSC
medium at 5 × 105 cells per 10 mL or 20,000 cells/cm2
(T-25 cm2 flask) (see Note 8).
3.2. Establishment and Cells isolated from the subventricular zone (SVZ) of the adult brain
Subculture of Primary can generate neurospheres. However, before the cells from adult
Neurospheres from mouse CNS tissue can be used for experiments, the tissue needs to
SVZ Cells from Adult be pretreated with appropriately buffered solutions containing tis-
Mouse CNS sue digestion enzymes and undergo mechanical dissociation in
order to obtain a cell suspension that is free of cell clumps, debris,
3.2.1. Processing and myelin, dead cells and undissociated tissue. This procedure uses the
Preparation of SVZ Tissue commercially available kit—NeuroCultTM Enzymatic Dissociation
Kit for Adult CNS Tissue. This procedure is optimized for process-
ing SVZ tissue isolated from up to eight adult mice. If more brains
are used, set up more tubes for the additional tissue and thaw addi-
tional tubes containing NeuroCultTM Dissociation Solution.
1. Prepare a 100 mm petri dish by adding 10 mL of NeuroCultTM
Tissue Collection Solution.
2. Thaw at room temperature sufficient NeuroCultTM Dissociation
Solution, previously aliquoted into 3 mL volumes.
3. Transfer dissected tissue pieces to the 100 mm dish containing
NeuroCultTM Tissue Collection Solution.
4. Once the dissections are complete and all the tissue has been
collected in the petri dish, remove all the NeuroCultTM Tissue
Collection Solution and mince tissue by chopping in a quick
rhythm with a sterile scalpel for approximately 1 min. Divide
the tissues into two roughly equal piles in the same 100 mm
dish and mince each pile separately ensuring that the tissue is
minced into the smallest pieces possible and can be pipetted
through disposable plastic tips (P200 or P1000).
5. Dispense 1 mL of NeuroCultTM Dissociation Solution into the
dish containing the minced tissue resuspending the tissue
pieces and transfer the minced tissue into a 15 mL tube. Do
not introduce any air bubbles in the transfer step. Repeat twice,
each time dispensing 1 mL of NeuroCultTM Dissociation
Solution into the dish, resuspending the tissues and transfer-
ring the minced tissues into a 15 mL tube so that the entire
3 mL of NeuroCultTM Dissociation Solution is used.
6. Incubate the minced tissue in NeuroCultTM Dissociation

Solution for 7 min at 37°C, preferably in a water bath.
7. Remove from water bath and add 3 mL of NeuroCultTM
Inhibition Solution and mix the tissue suspension gently avoid-
ing air bubbles.
8. Centrifuge the suspension at 700 rpm (100 × g) for 7 min.
9. Discard the supernatant and resuspend the pellet with 0.15 mL
of NeuroCultTM Resuspension Solution.
10. Mechanically dissociate or “triturate gently” the digested tis-
sue using a plastic disposable tip attached to a P200 micropi-
pettor, which has been set to 0.18 mL, by pipetting up and
down ~5–10 times until a smooth and “creamy” suspension is
achieved. The suspension should be able to pass through the
plastic tip bore without getting lodged in the tip.
11. Once a smooth suspension has been achieved, add 0.1 mL of
NeuroCultTM Resuspension Solution to bring the volume to
approximately 0.3 mL.
12. Pipette the suspension approximately five more times with a
disposable plastic tip attached to a P1000 micropipettor (set to
0.28 mL) to achieve a homogenous cell suspension without
any remaining chunks of tissue.
13. Add NeuroCultTM Resuspension Solution to the tube to a final
volume of 10 mL and mix with the cells.
14. Wash the cell suspension by centrifugation at 700 rpm (100 × g)
for 7 min.
15. Discard the supernatant and resuspend the pellet by adding
0.2 mL of NeuroCultTM Resuspension Solution and pipetting
up and down five times using a plastic disposable tip attached
to a P200 micropipettor.
16. Add NeuroCultTM Resuspension Solution to a final volume of
10 mL, mix the cells and wash the cell suspension by centrifu-
gation at 700 rpm (100 × g) for 7 min.
17. Repeat steps 15 and 16 once more for a total of three washes
with the NeuroCultTM Resuspension Solution.
18. Resuspend the final pellet in 0.5–1 mL (depending on the size
of the pellet and number of brains used) of an appropriate
medium necessary for subsequent experiments.
19. Proceed to Subheading 3.3 for culturing neurospheres from
the SVZ region of adult mouse CNS.
3.3. Culture of Primary 1. Place a sterile 40 μm cell strainer over the mouth of a sterile
Neurospheres from 50 mL conical tube. Filter the resuspended cell suspension from
SVZ Cells from Adult Subheading 3.2 by gently dispensing the cell suspension through
Mouse CNS the 40 μm cell strainer to remove any clumps of cells and remain-
ing chunks of debris. Allow the cell suspension to flow through

the filter by gravity flow without forcing the cells through.
2. Measure the precise volume and count cell numbers using a
dilution in Trypan Blue (1/5 or 1/10 dilution) and hemacy-
tometer to assess cell density and viability. Counting of cells
can be difficult because of the small pieces of debris, damaged
blood vessels and myelin, which may still remain. However,
the yield of viable cells should be approximately 3 × 105 cells
from the subventricular region of one mouse brain.
3. To generate neurospheres, seed primary adult cells at
2 × 104 cells/cm2 (1.9 × 105 total cells) in 6-well tissue culture
dishes (cat. no. 3506, Corning) or 2 × 104 cells/cm2 (5 × 105 total
cells) in a T-25 cm2 flask (cat. no. 156367, Nunc) containing
“Complete” NeuroCultTM NSC Proliferation Medium (Adult).
3.3.1. Subculturing of Similar to neurospheres derived from embryonic mouse CNS cells,
Mouse Neurospheres the procedures outlined below can be performed repeatedly over
Derived from Adult Mouse multiple culture passages for neurospheres derived from adult mouse
SVZ CNS Cells cells to generate long term neurosphere cultures. Neurospheres
derived from adult mouse CNS cells are generally dissociated into a
single cell suspension using mechanical dissociation techniques; how-
ever, dissociation with commercially supplied enzymes such as
Accutase™ has provided good results for cell viability and ability of the
dissociated cells to produce new neurospheres for multiple passages.
1. Observe the neurosphere cultures under a microscope to deter-
mine if the neurospheres are ready for passaging. If neuro-
spheres are attached to the culture substrate, tapping the
culture flask against the bench top should detach them (Fig. 4;
Note 9).
2. Using a disposable pipette, remove medium with suspended
neurosphere from a T25 cm2 flask and place in a 14 mL sterile
tissue culture tube. If some cells remain attached to the sub-
strate, detach them by shooting a stream of media across the
attached cells. Spin at 400 rpm (75 × g) for 5 min.
of 200 μL of complete NSC medium.
4. With a plastic disposable pipette tip attached to a P200 pipettor
set at ~180 μL, triturate the neurospheres until single cell sus-
pension is achieved (see Note 7).
medium at 3.8 × 104 cells per 3 mL per well of a 6-well tissue
culture plate or 1 × 105 cells per 10 mL or 4,000 cells/cm2
(T-25 cm2 flask).
Fig. 4. Neurosphere cultures derived from adult SVZ cells at different days during the culture. (a) Cells isolated from adult
SVZ region cultured in complete proliferation medium for 1 day. Cultures have a slurry and will contain a high amount of
debris and dead cells from the tissue preparation. (b) Cultures of SVZ cells after 4 days still contain debris but small aggre-
gates of cells are observed. These aggregates which are spheroid in shape, tightly packed and have an intact periphery
are quite distinguishable from clumps of dead cells or clumps of debris which are more loosely held together with uneven
peripheries and not spheroid in morphology. (c) Between 5 and 10 days in culture, SVZ cells formed healthy neurospheres
which continue to grow in size and are phase contrast bright. (d) Neurospheres at day 7 of passage one.
3.4. Establishments of E14 and SVZ adult CNS cells can also be cultured in adherent cul-
Adherent Cultures of tures in serum free media in the presence of an extracellular matrix
E14 or SVZ Adult CNS (ECM) such as poly-D-lysine, laminin, poly-D-ornithine or combina-
Cells in the Presence tions of these substrates. The procedure for culturing single cell
of an Attachment suspensions of undifferentiated adult SVZ cells in adherent cultures
Substrate uses the same serum-free media formulation and cytokines (i.e.,
EGF, FGF) used in neurosphere cultures. However, in adherent
cultures an appropriate ECM is also used to coat the tissue culture
wells or flasks, thus promoting strong cells-to-ECM attachment
rather than cell-to-cell attachment and aggregate formation in
suspension. Cell suspensions from E14 or SVZ adult mouse pre-

pared according to Subheadings 3.1 and 3.2.1 above can be used to
set up adherent cultures according to the protocols outlined below.
1. Perform a final cell count based on Trypan blue exclusion to
determine the cell concentration and total cell numbers in the
start sample.
2. Determine if there is sufficient cell numbers to set up adherent
cultures at 800 cells per mm2 for embryonic cells and
158 cells per mm2 for adult cells in pre-coated 6-well tissue
culture plates or pre-coated T25 cm2 flasks.
3. Prepare the appropriate volume of Complete Proliferation
Medium or Complete Proliferation Medium (Adult Mouse)
according the cell samples used. The Complete Proliferation
media contains the Basal-Hormone Media mix with 20 ng/
mL rhEGF for E14 cells or includes 20 ng/mL rhEGF, 10 ng/
mL bFGF and 0.0002% heparin for the adult cells.
4. Dispense the total number of E14 or SVZ cells as calculated
from step 2 above into the appropriate volume of Complete
Proliferation Medium to give the correct final plating cell
density.
5. Plate 3 mL of the cell suspension in the poly-D-lysine/laminin
pre-coated 6-well plates or 10 mL of the cell suspension in
poly-D-lysine/laminin pre-coated T25cm2 flasks.
3.4.1. Subculturing Cells In the presence of an ECM, neural stem and progenitor cells will
from Adherent Cultures of adhere to the tissue culture vessel within 24 h. The attached cells
E14 or SVZ Adult CNS Cells show a flattened morphology and are mostly bipolar (Fig. 5). Cells
which do not attach will float in suspension as single cells; however,
under optimal culture conditions, the presence of single cells will
be minimal. The undifferentiated cells will proliferate and over time
form a monolayer of cells on the bottom of the tissue culture vessel.
The cells should be allowed to grow until there is 60–80% confluency
before subculture. Before harvesting the cells, it is important to
prepare the required number of pre-coated wells or flasks needed
for subculture of the cells. Adherent cultures derived from embry-
onic and adult mouse CNS cells are generally detached from the
tissue culture vessel with enzymes such as Trypsin or other similar
enzymes. The procedure outlined below employs Accutase™ which
provided good results for cell viability and ability of the dissociated
cells to initiate new adherent cultures for multiple passages.
1. Observe the adherent cultures under a microscope to deter-
mine if the cells are 60–80% confluent and are ready for passag-
ing (Fig. 5; Note 10).
2. Using a disposable pipette, remove medium from a 6-well dish
or T25 cm2 flask and discard.
Fig. 5. Adherent cultures derived from adult SVZ cells at different days. (a) Cells isolated from the SVZ region after 1 day in
adherent culture conditions may contain some debris which makes it difficult to see adherent cells. (b) SVZ region cells
after 4 days in adherent cultures may contain some aggregates of cells in suspension; however, attached cells can be
observed. (c) Cells cultured for 10 days have attached to the bottom of the vessel and formed a monolayer of cells which
is about 70–80% confluent. (d) A monolayer of cells at 70% confluence after 7 days culture in passage 1.
3. Add 3 mL of PBS to each 6-well dish or 10 mL of PBS to a

T-25 flask. Swirl the culture plate or flask gently.
4. Remove PBS from a 6-well or T-25 cm2 flask and discard.
5. Add 0.5 mL of Accutase™ per well of the 6-well plate or 1 mL
for a T25 cm2 flask. Incubate the cells with the Accutase™ for
5 min at 37°C.
6. Observe the culture to determine if the cells are starting to
detach and detachment is complete.
7. Add 2 mL per well to 6-well plate or 5 mL per flask for a
T25 cm2 flask of Complete Proliferation Media using a dispos-
able pipette. Using the same pipette, resuspend the detached
cells, pipette up and down the cells-media suspension around
the well or flask two to three times to finally collect all the
detached cells and place in a new sterile 14 mL tube. If cells
remain, add a small volume of Complete Proliferation Medium
again and repeat the procedure to collect the remaining cells.
8. Spin at 850 rpm (110 × g) for 5 min.
of 200 μL of Complete Proliferation medium with a plastic
disposable pipette tip attached to a P200 pipettor set at
~180 μL, pipetting until single cell suspension is achieved.
10. Resuspend cells in an appropriate (approx. ~0.5–1 mL) vol-
ume of Complete Proliferation medium.
medium at 200 cells per mm2 in 3 mL of Complete Proliferation
medium per well for a 6-well tissue culture plate for embryonic
cells or 80 cells per mm2 in 10 mL medium per T-25 cm2 flask
for adult SVZ cells.
3.5. Differentiation of Mouse CNS cells can be cultured in the presence of serum or in
E14 or SVZ Mouse CNS serum-free conditions. In the presence of growth factors such as
Cells in Serum EGF and FGF, neural stem cells and their progeny will form neu-
Containing Media rospheres and are in a relatively undifferentiated state. Upon
in the Presence of removal of the growth factor and addition of a small amount of
Appropriate Substrate serum and appropriate attachment substrate, differentiation of the
NSC progeny into neurons, astrocytes and oligodendrocytes is
induced (see Fig. 2). If E14 mouse CNS cells are cultured directly
in absence of growth factors in a serum-free media which promotes
survival of mature neurons with the appropriate substrate, primary
mature neurons will be maintained and extend axonal projections
while astrocytes and oligodendrocytes will be undetectable.
3.5.1. Differentiation of 1. After 7–8 days in the neurosphere cultures, remove the medium
Cells from E14 Mouse CNS with suspended neurospheres and place in an appropriate sized
Cells in Serum Containing sterile tissue culture tube. If some spheres remain attached to
Media and Poly-D-Lysine/ the substrate detach them by shooting a stream of media across
Laminin Substrate the attached cells. Spin at 400 rpm (75 × g) for 5 min.
2. The cytokine-containing supernatant (EGF) is pipetted off and
discarded.
3. Resuspend the neurospheres in 0.2 mL of complete NSC dif-
ferentiation medium (see Subheading 2.4.2).
4. With a plastic disposable tip attached to a 200 μL pipettor set
at 0.180 mL, triturate the neurosphere suspension until a sin-
gle cell suspension is achieved. If undissociated tissue remains,
follow instruction as in the procedure for subculturing primary
cultures (see Subheading 3.1, step 3). Measure the precise

volume and count cell numbers using a dilution in Trypan blue
(1/5 or 1/10 dilution) and hemacytometer.
5. Based on the cell counts obtained above, use the complete
NSC Differentiation Medium to prepare an appropriate vol-
ume for plating the number of desired wells in a 24-well plate
containing PDL/Laminin coated coverslips with a cell density
of 1 × 105 cells/well.
6. Observe cultures after 6–10 days and up to 14 days if necessary
with an inverted light microscope and determine if cells have
differentiated and are viable (see Note 11).
7. Coverslips containing differentiated neural cells can be removed
and processed immediately for indirect immunofluorescence.
3.5.2. Differentiation of 1. After 7–8 days of culture in neurosphere cultures, remove the
Dissociated Cells from SVZ medium with suspended neurospheres and place in an appro-
Mouse CNS Cells in Serum priate sized sterile tissue culture tube. If neurospheres remain
Containing Media and attached to the substrate detach them by shooting a stream of
Matrigel™ media across the attached cells. Spin at 400 rpm (75 × g) for
5 min.
2. The cytokine-containing supernatant (EGF, FGF and heparin)
is pipetted off and discarded.
3. Resuspend the neurospheres in 0.2 mL of complete NSC
Proliferation Medium. With a 200 μL micropipettor set to
0.18 mL attached to a disposable tip, triturate the neurosphere
suspension until a single cell suspension is achieved. If undis-
sociated tissue remains, follow instruction as in the procedure
for subculturing primary cultures. Measure the precise volume
and count cell numbers using a dilution in Trypan blue (1/5
or 1/10 dilution) and hemacytometer.
4. Based on the cell counts obtained above, use the complete
NSC Proliferation Medium with 20 ng/mL b-FGF to prepare
an appropriate volume for plating the number of desired wells
in a 24-well plate containing Matrigel™ coated coverslips with
a cell density of 1 × 105 cells/well.
5. On day 2, 800 μL of the cytokine-containing supernatant
(FGF) is pipette off and discarded.
6. Add 800 μL of complete NSC Differentiation Medium (with
2% FBS, without cytokines) dropwise to each 24-well.
7. Observe cultures after 6–14 days with an inverted light micro-
scope and determine if cells have differentiated and are viable.
8. Coverslips containing differentiated neural cells can be removed
and processed immediately for indirect immunofluorescence.
3.6. Enumeration of 1. Thaw bottles or aliquots of NeuroCultTM NCFC Serum-Free

Neural Stem Cells and Medium without Cytokines and NeuroCultTM NSC Proliferation
Neural Progenitor Supplements (Mouse) at room temperature or overnight at 4°C.
Cells Using the Neural 2. Place thawed medium and supplements at 37°C and the
Colony Forming Cell Collagen Solution on ice (see Note 12).
Assay 3. Filter the single cell suspension of neural cells (derived from
cultured neurospheres or primary embryonic cells) through a
40 μm cell strainer to remove any undissociated cells or clumps
of cells (see Note 13).
(a) Primary embryonic cells: Dilute primary embryonic cells to
6.5 × 105 cells per mL in Complete Proliferation Medium
which will give a final cell plating density of 7,500 cells per
35 mm culture dish in a 25 μL volume.
(b) Cells derived from neurosphere cultures: Dilute these cells
to 2.2 × 105 cells per mL which will give a final cell plating
density of 2,500 cells per 35 mm culture dish in a 25 μL
volume (see Note 14).
4. Prepare and label the correct number of 35 mm dishes required
for the intended experiment. The volumes of reagents listed
below are designed for duplicate 35 mm plates.
5. Place a sterile 14 mL tube or the appropriate number of tubes
for dispensing the neural colony forming cell (NCFC) assay
reagents and cells for each test condition in a tube rack.
6. To make the semisolid collagen NCFC Medium (allowing
duplicate 35 mm culture dishes) add the following components
in the given order: 1.7 mL of NeuroCultTM NCFC Serum-Free
Medium without Cytokines, 0.33 μL of NeuroCultTM NSC
Proliferation Supplements (Mouse), 6.6 μL of a (rhEGF) stock
solution of 10 μg/mL and then 25 μL of the cell suspension
either at a concentration of 6.5 × 105 cells/mL (for primary
cells) or 2.2 × 105 cells/mL (for cultured cells) (see Note 15).
7. Mix the medium containing cells (~2.1 mL total) by pipetting
with a disposable 2 mL pipet.
8. Using a separate sterile 2 mL pipette, transfer 1.3 mL of cold
Collagen Solution to the tube and mix again by pipetting.
Using the same 2 mL pipette, remove 1.5 mL of the final cul-
ture mixture and dispense this volume into a 35 mm culture
dish. Dispense another 1.5 mL in the same manner into a sec-
ond 35 mm dish. Remove any air bubbles by gently touching
bubble with the end of the pipette (see Note 15).
9. Gently tip each culture dish using a circular motion to allow
the mixture in the dishes to spread evenly over the surface.
10. Place the 35 mm culture dishes in a 100 mm petri dish. This
petri dish must also contain an open 35 mm culture dish filled
with 3 mL of sterile water to maintain optimal humidity during
the prolonged incubation period. Replace the lid of the

100 mm Petri dish (see Note 16).
11. Transfer the plates to an incubator set at 37°C, 5% CO2 and
>95% humidity (see Note 17).
12. Culture cells for 21 days (differences in colony size can be
clearly discerned after 21 days).
13. As cultures are incubated for an extended period of time
(21–28 days), cultures should be replenished with Complete
Replenishment Medium once a week to avoid depletion of
culture media (see Note 18).
14. Gently (so as not to disrupt gel) add 60 μL of Complete
Replenishment Medium into the center of each NCFC dish
once every 7 days during the entire NCFC culture incubation
(21 days).
15. Cultures should be visually assessed regularly for overall colony
growth and morphology using an inverted microscope.
3.6.1. Categorizing NCFC Within 4–7 days of plating in the NCFC Assay, neural stem and
Colonies and Procedure for progenitor cells begin to proliferate forming small colonies (Fig. 6).
Scoring NCFC Colonies By day 14, these small colonies have grown in size and differences
can be discerned between colonies. A number of the colonies
Fig. 6. Colonies from the four size categories in the NCFC assay. (a) Colony < 0.5 mm in diameter. (b) Colony 0.5–1 mm in
diameter. (c) Colony 1–2 mm in diameter. (d) Colonies >2 mm in diameter are derived from cells which met all the func-
tional criteria of a NSC.
appear to stop growing after approximately 10–14 days while other

colonies continue to expand. By day 21–28, colonies can be
classified into four categories: (1) less than 0.5 mm in diameter, (2)
0.5–1 mm in diameter, (3) 1–2 mm in diameter, and (4) 2.0 mm
or > 2 mm in diameter. Refer to Fig. 6 for representative examples
of different colony sizes.
1. Place an individual 35 mm culture dish on a gridded scoring
dish and then place both the culture dish and gridded dish on
the dissecting microscope stage.
2. First scan the entire dish using a low power (2.5–5×) objective
lens, noting the relative proximity of the colonies to each other.
Scoring can then be performed with the same lens. Use a
higher power (10×) objective to examine colonies in greater
detail.
Classify colonies into four categories:
(a) Less than 0.5 mm in diameter (Fig. 6a)
(b) 0.5–1 mm in diameter (Fig. 6b)
(c) 1–2 mm in diameter (Fig. 6c)
(d) 2.0 mm or > 2 mm in diameter (Fig. 6d)
3.6.2. Estimation of NSCS The following criteria are applied for the quantification of NSCs
or Neural Progenitors and progenitor cells from primary embryonic cells or cultured neu-
rospheres derived from embryonic cells:
The original cell that forms a colony 2.0 mm or > 2 mm in
diameter is referred to as a Neural Colony Forming Cell—Neural
Stem Cell (NCFC-NSC) as this cell has high proliferative potential
and multi-lineage potential. Colonies < 2 mm contain cells which
lack self-renewal ability and multipotency and are likely produced
by a progenitor.
The NCFC assay can also be used to measure total colony
numbers which is a similar readout to total sphere numbers read-
out in the neurosphere cultures which detects the cells which have
colony-forming ability or sphere-forming ability.
4. Notes
1. The performance of media prepared in the laboratory is highly

dependent on the quality and purity of the water and raw mate-
rials. If media is prepared in the laboratory, use only tissue-
culture-grade materials and if necessary source various suppliers
to determine the best quality reagents as there is significant
batch to batch variability in some critical reagents. To avoid
variability in media performance, STEMCELL provides
optimized and standardized kits for the proliferation neural.
An optimized basal medium for the culture of neurospheres

from embryonic and adult mouse CNS cells is available,
NeuroCultTM NSC Basal Media (Mouse) (cat. no. 05700,
STEMCELL Technologies Inc). An optimized 10× hormone
mix for the culture of neurospheres from embryonic and adult
mouse CNS cells is available, NeuroCultTM NSC Proliferation
Supplements (cat. no. 05701, STEMCELL Technologies Inc).
To avoid variability in media performance, STEMCELL pro-
vides optimized and standardized kits for the differentiation of
neural stem and progenitor cells from the mouse CNS (cat. no.
05704, STEMCELL Technologies Inc).
2. Both EGF and bFGF have been shown to be mitogens for
CNS stem cells. In general, the number of neurospheres gen-
erated and the rate of expansion are enhanced when the two
mitogens are used simultaneously; however, each growth fac-
tor can act on different populations of stem cells. EGF is rou-
tinely used for embryonic day 14 mouse CNS cultures, while
EGF, FGF and heparin is required for culture of adult mouse
sub-ventricular zone cells, embryonic rat and fetal human CNS
cells. Heparin facilitates high efficient binding of FGF to the
FGF receptor.
3. To triturate, slightly tilt the tip and press it against the bottom
or side of the tube to generate resistance to break up the tissue.
The mechanical dissociation of cells by trituration with a fire-
polished pipette or disposable plastic tip is known to cause cell
death. While earlier published protocols suggested the use of
fire-polished glass Pasteur pipettes, non-fire polished glass
Pasteur pipette and disposable plastic tips work well too.
However, some precautionary steps can be performed during
trituration to diminish the negative effects. For example, avoid
forcing air bubbles into the cell suspensions. Also, it is impor-
tant to wet the pipette with a small amount of media before
sucking the cells into the pipette to reduce the number of cells
sticking to the glass or plastic surface.
4. Trituration must be repeated until cell clumps and intact neu-
rospheres are dissociated. Because clumps of cells are heavier
than single cells, these will settle to the bottom of the tube
when left standing for about 5 min. The clumps can be allowed
to settle, then the single-cell suspension can be removed to a
fresh sterile tube and used for subsequent cultures, leaving the
undissociated clusters at the bottom of the tube. Repeat this
procedure of dissociating the remaining undissociated clusters,
letting the undissociated clumps to settle at the bottom and
collecting the supernatant containing the dissociated cells into
the tube with the single-cell suspension until the majority of
clusters have been dissociated.
5. Other nonmechanical methods to dissociate neurospheres to

produce a homogenous single cell suspension are available. For
example, the NeuroCultTM Chemical Dissociation Kit Catalog
(cat. no. 05707, STEMCELL Technologies Inc.) offers a non-
mechanical, nonenzymatic alternate procedure for dissociating
neurospheres derived from embryonic and adult mouse CNS
cells which yields greater cell viabilities compared to the tritu-
ration method. The commercially available product Accutase™
(cat. no. 07920, STEMCELL Technologies Inc.) can also be
used for dissociating neurospheres derived from embryonic
and adult mouse CNS without adversely affecting stem cell
function.
6. Initially single cells should proliferate to form small clusters or
aggregates of cells that might lightly adhere to the culture
vessel which will eventually lift off from the substratum as the
density of the aggregates increases. Viable neurospheres are
characterized by their semitransparent appearance, with many
of the cells on the outer surface displaying microspikes (see
Fig. 1). Neurospheres should be passaged before they grow
too large (>150–200 μm in diameter) (Fig. 3). The size of
the neurospheres can be estimated by sampling ~ 10 μL from
the culture in a hematocytometer. In the larger neurospheres,
the cells within the core of the neurospheres lack appropriate
gas and nutrient/waste exchange and become necrotic and
appear dark and are also more difficult to dissociate. It is
important that cultures be monitored regularly and with
increased experience, healthy neurospheres cultures will
become recognizable.
7. By setting the volume of the pipettor lower the volume of
the cell suspension avoids expulsion of all the liquid and intro-
duction of air bubbles into the cell suspension.
8. The cell density for replating cells from the striatum, cortex,
ventral mesencephalon, or other regions of the E14 mouse
brain is lower than that for primary culture conditions.
9. In neurosphere cultures initiated from primary adult SVZ cells,
a lot of debris will be commonly observed (Fig. 4). A half
media change can be performed 2–3 days after the cultures are
set up. However, similar to E14 mouse cells, neural precursor
cells initially lightly adhere to the culture vessel and proliferate
to form small clusters which eventually lift off from the sub-
stratum as the density of the aggregates increases. By perform-
ing a media change too earlier after the culture set up, the risk
of removing some neural precurscor cells which have not
attached to the bottom of the tissue culture vessel is greater. At
the end of the culture period of 5–7 days, viable neurospheres
can be observed amongst all the debris by their semitranspar-
ent appearance and defined periphery (see Fig. 4). These
neurospheres can be distinguished from clusters of debris or

dead cells which are normally not spheroid in shape and have a
irregular periphery with cells blebbing off.
10. In adherent cultures initiated from primary adult SVZ cells, a lot
of debris will commonly be observed (Fig. 5). A half media
change can be performed 2–3 days after the cultures are set up.
By performing a media change too earlier after the culture set
up, the risk of removing some neural precurscor cells which have
not attached to the bottom of the tissue culture vessel is greater.
At the end of the culture period of 7–10 days, viable adherent
cells can be observed amongst all the debris (see Fig. 5). In some
cases, particularly with the culture of primary embryonic CNS
cells, the cells may form aggregates of cells or “neurospheres”
instead of adherent cultures. However, after several 5–6 days,
these aggregates tend to attach to the ECM and flatten out and
continue to proliferate as a monolayer of cells.
11. In the differentiation cultures, depending on the number of
cells plated the medium may or may not have to be changed
during the differentiation procedure. Plates should be checked
daily. If the medium becomes acidic it should be changed by
removing approximately 50% of the medium and replacing
with fresh complete NSC differentiation medium.
12. It is important to keep the collagen on ice throughout the
culture set up to prevent the collagen from gelling.
13. The NCFC assay is based on the formation of colonies from a
single or clonal colony forming cell; therefore, it is critical that
the initial single cell suspension is homogenous for this step. In
mixing experiments, the frequency of chimera (non-clonal)
colonies in the NCFC assay was estimated to be <1% compared
to >20% chimera neurospheres observed in the neurosphere
cultures.
14. The cell plating density was determined by performing titra-
tion curves and determining the linearity ranges for primary
cells isolated from normal embryos of pregnant CD1 albino
mice or cells from neurospheres derived from normal E14
CD1 albino mice cortices and/or striata cultured for two pas-
sages. It may be necessary to perform a titration curve within
the range of 5,000–50,000 cells per dish for primary cells or
1,000–5,000 cells per dish cultured cells when different species
and transgenic animals are used. It is possible that the cloning
efficiency is changed for these cells types. Adjust the initial con-
centration of the cells so that 25 μL volume of cells is always
added to the medium mixture described in Subheading 6
below to maintain accurate media concentrations.
15. Do not add the collagen solution yet. If multiple tubes are
being set up, add cells to a single tube then add collagen and
plate cells. Do not let cells sit in NCFC medium for an extended
period of time before plating. The collagen starts to gel within
several minutes following the addition to the cell suspension. If
more than one tube is being set-up, collagen should be added
to the first tube only, and the contents dispensed into dishes
before proceeding to the next tube.
16. If many dishes are used, these dishes can also be placed in a
covered 245 mm square bioassay dish with two or three open
35 mm culture dishes containing sterile water.
17. Gel formation will occur within approximately 1 h. It is impor-
tant not to disturb the cultures during this time.
18. Make up 10 mL of Complete Replenishment Medium by mix-
ing 9 mL of Basal Medium and 1 mL of 10× Hormone Mix.
To this, add 500 μL of the 10 μg/mL stock solution of hEGF.
The Complete Replenishment Medium contains 1: 10 Basal
and 10× Hormone Mix and 0.5 μg/mL of hEGF. Store the
prepared media at 4°C for up to the 3 weeks of replenishing.
Acknowledgments
We would like to acknowledge our collaborators Drs. Brent

Reynolds, Rod L. Rietze, and Angelo L. Vescovi for their contin-
ued technical and scientific help and discussions.
References
1. Reynolds B, Weiss S (1992) Generation of and FFG in the developing mouse telencepha-
neurons and astrocytes from isolated cells of lon. Dev Biol 208:166–188
the adult mammalian central nervous system. 6. Gross CG (2000) Neurogenesis in the adult
Science 255:1701–1710 brain: death of a dogma. Nat Rev Neurosci
2. Reynolds BA, Tetzlaff W, Weiss S (1992) A 1:67–73
multipotent EGF-responsive striatal embry- 7. Ray J, Peterson DA, Schinstine M, Gage FH
onic progenitor cell produces neurons and (1993) Proliferation, differentiation, and
astrocytes. J Neurosci 12:4565–4574 long-term culture of primary hippocampal
3. Reynolds BA, Weiss S (1996) Clonal and pop- neurons. Proc Natl Acad Sci USA 90:
ulation analyses demonstrate that an EGF- 3602–3606
responsive mammalian embryonic CNS 8. Conti L, Pollard SM, Gorba T et al (2005)
precursor is a stem cell. Dev Biol 175:1–13 Niche-independent symmetrical self-renewal
4. Morshead CM, Reynolds BA, Craig CG, of a mammalian tissue stem cell. PLoS Biol
McBurney MW, Staines WA, Morassutti D, 3:e283
Weiss S, van der Kooy D (1994) Neural stem 9. Conti L, Cattaneo E (2010) Neural stem cell
cells in the adult mammalian forebrain: a rela- systems: physiological players or in vitro enti-
tively quiescent subpopulation of subependy- ties? Nat Rev Neurosci 11:176–187
mal cells. Neuron 13:1071–1082 10. O’Connor TJ, Vescovi AL, Reynolds BA
5. Tropepe V, Sibilia M, Ciruna BG, Rossant J, (1998) Isolation and propagation of stem cells
Wagner EF, van der Kooy D (1999) Distinct from various regions of the embryonic mam-
neural stem cells proliferate in response to EGF malian central nervous system. In: Celis JE
(ed) Cell biology: a laboratory handbook, 2nd 13. Singec I, Knoth R, Meyer RP et al (2006)
edn. Academic, New York Defining the actual sensitivity and specificity of
11. Reynolds BA, Rietze RL (2005) Neural stem the neurosphere assay in stem cell biology. Nat
cells and neurospheres – re-evaluating the rela- Methods 3:801–806
tionship. Nat Methods 2:333–336 14. Louis SA, Rietze RL, Deleyrolle L et al (2008)
12. Rietze RL, Reynolds BA (2006) Neural stem Enumeration of neural stem and progenitor
cell isolation and characterization. Methods cells in the neural colony-forming cell assay.
Enzymol 419:3–23 Stem Cells 26(4):988–996
Chapter 31
Feeder-Independent Culture Systems for Human

Pluripotent Stem Cells
Jennifer Moody
Abstract
The continued success of pluripotent stem cell research is ultimately dependent on access to reliable and
defined reagents for the consistent culture and cryopreservation of undifferentiated, pluripotent cells. The
development of defined and feeder-independent culture media has provided a platform for greater repro-
ducibility and standardization in this field. Here we provide detailed protocols for the use of mTeSR™1
and TeSR™2 with various cell culture matrices as well as defined cryopreservation protocols for human
embryonic and human induced pluripotent stem cells.
Key words: Pluripotent stem cell, Embryonic stem cell, Induced pluripotent stem cell, Cell culture,
Cell culture media, Cell culture matrix, Cryopreservation
1. Introduction
Undifferentiated human pluripotent stem cells (hPSC) have the

potential for unlimited expansion and the ability to differentiate
into cells that emanate from all three germ layers: endoderm,
mesoderm, and ectoderm. As such, hPSC represent a valuable tool
for the study of human cellular and developmental systems and
hold promise for a variety of future clinical applications. Human
cells with pluripotent characteristics were initially derived from the
inner cell mass of preimplantation blastocysts and termed human
embryonic stem cells (hESCs) (1, 2). More recently, the discovery
of methods for creating human induced pluripotent stem cells
(hiPSCs) using over-expression of key factors in somatic cell types
has revealed a potentially limitless alternate source of pluripotent
cells (3–6). hiPSCs phenotypically and functionally resemble hESCs
507
508 J. Moody
but can be created from a multitude of cell types and patients offer-
ing a diverse way of studying the cellular mechanisms of disease
and pursuing patient-specific therapies (7–9).
Over the last decade, the basic techniques to culture hPSC
have been continuously evolving towards better definition of cul-
ture components and elimination of animal-derived components.
Many of the most widely used lines were originally established
using a culture system composed of mouse embryonic fibroblast
(MEF) feeder cells and fetal calf serum-containing media, a con-
ventional method developed for mouse ESC culture (1, 2). Both
of these reagents are undefined and have batch to batch variability
making quality control a time consuming challenge and experi-
mental reproducibility between labs next to impossible. Furthermore
the use of feeders and animal-derived components for culturing
these cells raises concerns regarding subsequent clinical applica-
tions due to the presence of immunogenic material or the risk of
transmitting animal virus or prion material. The requirements for
the ultimate culture system (which encompasses both the media
and the culture matrix) vary depending on the desired application.
Basic research applications do not necessarily require the complete
elimination of animal-derived components, but do benefit from
further defining the components from a reproducibility standpoint.
The ultimate culture system for cells destined for clinical applica-
tions would be fully defined, fully recombinant and free from all
animal-sourced material. Achieving this latter goal while maintain-
ing the growth characteristics, genomic stability, and differentia-
tion potential of these cells is currently a work in progress for the
field, but key incremental steps have been made in this direction.
mTeSR™1 is a serum-free, defined, and feeder-independent
medium that was developed and published by Tenneille Ludwig
and colleagues at the WiCell Research Institute (Madison, WI)
(10). It was shown to support self-renewal of multiple hESCs lines
over many passages without requiring feeder cells. The formula-
tion included high levels of bFGF together with transforming
growth factor β (TGFβ), gamma-aminobutyric acid (GABA), pipe-
colic acid, and lithium chloride. As published, it utilized zebrafish
FGF (zFGF) and a bovine albumin source and was developed for
use with growth factor-reduced (GFR) Matrigel™ (BD Biosciences)
as a matrix. Matrigel™ is an extracellular matrix derived from the
Engelbreth Holm Swarm murine tumors. It predominantly con-
tains laminin, fibronectin, entactin, and heparin sulfate. mTeSR™1
was subsequently made commercially available by STEMCELL
Technologies with recombinant human FGF instead of zFGF and
was paired with an ES qualified version of GFR-Matrigel™. The
Ludwig group also published an animal protein-free version of this
medium and as a proof of principle utilized it with a cell support
matrix composed of four human components (collagen IV,
fibronectin, laminin, and vitronectin) (11). This was the first
31 Feeder-Independent hPSC Culture 509
publication that described a feeder-independent culture system

that was free from animal proteins and could support the deriva-
tion and maintenance of hESC. TeSR™2 is the commercially avail-
able formulation that is based on the formulation in the publication
but has been modified to contain recombinant proteins.
This chapter discusses the use of defined and feeder-indepen-
dent culture systems for hPSC that are relevant for various down-
stream applications.
2. Materials
2.1. Cell Culture 1. mTeSR™1 (STEMCELL Technologies, catalog #05850).

Reagents Basal Medium, 400 mL (05851), stored at 2–8°C. 5×
Supplement, 100 mL (05852) stored at −20°C. Thaw 5× sup-
plement at room temperature (RT) or overnight in refrigera-
tor. Combine Basal Medium and Supplement, mix well and
aliquot. Store complete media aliquots at −20°C for up to
6 months and for up to 2 weeks at 2–8°C.
2. TeSR™2 (STEMCELL, catalog #05860). Basal Medium,
400 mL (05862) stored at 2–8°C. 5× Supplement, 100 mL
(05862) and 250× Supplement, 2 mL (05863) stored at
−20°C. Thaw TeSR™2 Supplements at RT or overnight in
refrigerator. Combine the three components, mix well and ali-
quot. Store complete media aliquots at −20°C for up to
6 months and for up to 2 weeks at 2–8°C.
3. Matrigel™ hESC-qualified matrix (BD Biosciences, catalog
#354277). Store in aliquots at −70°C for up to 6 months (see
Note 1).
4. Human recombinant Vitronectin (R&D Systems, catalog
#2308-VN). Reconstitute in phosphate-buffered saline (PBS)
at 250 μg/mL. Store in aliquots at −20°C to −70°C for
3 months.
5. Dispase, 1 mg/mL. (STEMCELL, catalog #07923). Store in
aliquots at −20°C. Store at 2–8°C for up to 2 weeks.
6. DMEM/F12 (STEMCELL, Invitrogen).
7. mFreSR™ (STEMCELL, catalog #05854). Store at −20°C.
Use aliquots immediately upon thawing.
8. Cryostor™CS10 (STEMCELL, catalog #07930). Store at
2–8°C until the expiry date listed on the label.
2.2. Supplies and 1. Conical Tubes, 15 mL and 50 mL (e.g., BD Biosciences).

Equipment 2. Serological pipettes, 2 mL, 5 mL, 10 mL (e.g., BD Biosciences).
3. Cell scrapers (Corning, catalog # 3010).
510 J. Moody
4. Tissue culture treated plates, 96-well, 4-well, 6-well, 10 cm

dishes (e.g., BD Biosciences).
5. Isopropanol freezing container (e.g., Nalgene Mr. Frosty).
6. Incubator with humidity and gas control to maintain 37°C and
>95% humidity in an atmosphere of 5% CO2 in air.
7. Inverted microscope with 2×, 4×, and 10× phase objectives
(e.g., Olympus CKX31) with eyepiece micrometer (e.g.,
Olympus B-L0501).
3. Methods
Cell culture procedures should be performed in a biosafety cabinet

certified for Level II handling of biological materials using aseptic
technique.
3.1. Coating Plates 1. Dispense 25 mL of DMEM/F-12 into a 50 mL tube and keep

on ice.
3.1.1. Coating Plates with
BD Matrigel™ 2. Thaw Matrigel™ on ice until liquid, then add to the cold
DMEM/F-12 and mix well (see Note 1).
3. Immediately use the diluted Matrigel™ to coat cultureware.
For a 6-well plate, use 1 mL per well; for a 100 mm dish, use
8 mL per plate. Swirl the plate to spread the solution evenly
across the surface.
4. Leave coated cultureware at RT (15–25°C) for at least 1 h
before use. Do not remove Matrigel™ solution or allow the
plates to dehydrate until they are ready to be used.
5. If not used immediately, the plates must be sealed to prevent
dehydration (e.g., with Parafilm™) and can be stored at 2–8°C
for up to 7 days before use. The Matrigel™ solution is removed
immediately before use.
3.1.2. Coating Plates with 1. Dilute the required volume of recombinant human vitronectin
Recombinant Human (rhVitronectin) to 5 μg/mL in PBS. Use 1 mL per well of
Vitronectin 6-well plate. To coat other sized cultureware, scale the volume
required to the surface area of the vessel to be coated.
2. Incubate coated plates at RT (15–25°C) for at least 1 h before
use. rhVitronectin coated plates may be stored overnight at
2–8°C if desired and should be brought to room temperature
(15–25°C) before use.
3.2. Morphological There are various ways to measure the quality of an undifferentiated
Assessment of culture of hPSC including FACS and RT-PCR analysis for specific
Undifferentiated hPSC pluripotency-related markers, in vitro differentiation assays, and in vivo
teratoma assays. However, to the experienced eye, much can be
Fig. 1. Morphology of undifferentiated hPSC cultured in the TeSR® system. (a) H9 cells grown in mTeSR®1 on Matrigel™
approximately 6 days after the cells were passaged illustrating the dense bright centers that indicate it is time to passage
the culture. (b) An example of an H1 colony in mTeSR®1 on Matrigel™ approximately 3 days after the cells were passaged.
(c) High magnification illustrates the prominent nucleoli and high nucleus to cytoplasm ratio of the cells within the colony.
gleaned from regular examination of the culture. Feeder-independent

growth of hPSC affords a slight advantage over feeder-based culture
systems in that differentiated cell types are sometimes more obvious to
identify without a background of feeders to blend in with.
Undifferentiated hPSCs in the TeSR™ system grow as compact, mul-
ticellular colonies, as shown in Fig. 1a. When viewed under high
magnification (i.e., 40× objective), individual cells should exhibit a
high nuclear-to-cytoplasmic ratio and prominent nucleoli (Fig. 1b, c).
It is normal for cells at the edge to appear more spread out than cells
more central in the colony. Differentiation within the culture is char-
acterized by loss of border integrity, gross nonuniformity of cell mor-
phology within a colony, and the emergence of obvious alternate cell
types between or within individual colonies (Fig. 2a, b).
The appearance, size, and density of cells within the colonies
changes significantly over a number of days in culture, so it is
important to examine the cultures regularly and become familiar
with hPSC and colony morphology. As colonies grow in mTeSR™1
or TeSR™2, they become more densely packed with cells and
512 J. Moody
Fig. 2. Spontaneous differentiation in the TeSR™ system. H1 cells grown in mTeSR™1 on Matrigel™ showing areas of
differentiation at the edges of two colonies (a) and differentiation developing in the centers of colonies (b).
Fig. 3. Colony morphology changes over time in the TeSR™ system. H9 cells grown in TeSR™2 on Matrigel™ 1 day (a),
3 days (b), and 5 days (c) after passaging.
develop phase bright centers. Development of these bright centers

is key to determining the optimal time of passaging. Figure 3a–c
demonstrates the change in appearance of hPSC colonies that
occurs from immediately after seeding (day 1) until the cells are
ready to passage (in this case day 5).
3.3. Thawing hPSCs cultured using other maintenance protocols (e.g., with
Cryopreserved hPSCs mouse embryonic feeders or their conditioned medium) can be
thawed into mTeSR™1 or TeSR™2 using this protocol. However,
if thawing a new cell line where there is limited numbers of vials
and limited experience with the line, it is recommended to thaw
and culture with the protocol and reagents recommended by the
supplier of the cells. hPSCs should be thawed into either 4- or
6-well plates coated with BD Matrigel™. If there are limited or
unknown numbers of clumps in the vial, a 4-well plate is recom-
mended. Have all tubes, warmed medium and plates ready before
starting the protocol to ensure that the thawing procedure is done
as quickly as possible. If the BD Matrigel™ plates have been stored
at 2–8°C, allow the plates to come to room temperature (15–25°C)
for 30 min before removing the BD Matrigel™ solution.
1. Quickly thaw the hPSCs in a 37°C waterbath by gently shaking
the cryovial continuously until only a small frozen pellet
remains. Remove the cryovial from the waterbath and wipe
with 70% ethanol to sterilize.
2. Use a 2 mL pipette to transfer the contents of the cryovial to a
15 mL conical tube. Use of a 2 mL pipette will minimize break-
age of cell clumps.
3. Add 5–7 mL of warm mTeSR™1 or TeSR™2 dropwise to the
tube, gently mixing as the medium is added.
4. Centrifuge cells at 300 × g for 5 min at room temperature
(15–25°C).
5. Aspirate the medium, leaving the cell pellet intact. Using a
2 mL pipette, gently resuspend the cell pellet in 1-2 mL of
mTeSR™1 or TeSR™2, taking care to maintain the cells as
aggregates.
6. Remove the Matrigel™ from a coated tissue culture plate by
gently tilting the plate and allowing the excess Matrigel™ solu-
tion to collect in that corner.
7. Remove the solution using a serological pipette or by aspira-
tion. Ensure that the tip of the pipette does not scratch the
coated surface.
8. Transfer the appropriate amount of medium containing the
cell aggregates to a BD Matrigel™-coated 4-well or 6-well
plate. Transfer 0.5 mL per well if using a 4-well plate. Transfer
2 mL per well if using a 6-well plate. Ensure that clumps are
evenly distributed between wells.
9. Place the plate into the 37°C incubator and move the plate in
quick side to side, forward to back motions to evenly distribute
the clumps within the wells (see Note 2). Culture the cells at
37°C, with 5% CO2 and 95% humidity.
514 J. Moody
10. Perform daily medium changes (see Note 3). Observe culture
daily to determine the optimal time for passaging (see
Subheading 3.2 and Note 4).
3.4. Passaging hPSCs Volumes given in this section are for 6-well culture dishes; scale
(See Note 5) reagent amounts accordingly for different sized cultureware.
3.4.1. Passaging Using 1. Aliquot sufficient mTeSR™1 or TeSR™2 to passage cells.

Split Ratios Warm aliquoted mTeSR™1 or TeSR™2, Dispase, and
DMEM/F-12 to room temperature (15–25°C).
2. Use a microscope to visually identify regions of differentiation
(see, e.g., Fig. 2a, b). Mark these using a felt tip or lens marker
on the bottom of the plate. This selection should not exceed
20% of the well if the culture is of high quality.
3. Remove regions of differentiation by scraping with a pipette
tip or by aspiration.
4. Aspirate medium from the hPSC culture and rinse with
DMEM/F-12 (2 mL/well).
5. Add 1 mL per well of dispase. Place at 37°C for 5–7 min if
using mTeSR™1 or for 3–4 min if using TeSR™2 (see Note 6).
Observe colonies under the microscope and look for the for-
mation of “sharp” edges that signify that the edges are just
beginning to roll up (see Fig. 4a, b and Note 6). When this is
observed immediately begin step 6.
6. Remove dispase, and gently rinse each well three times with
2 mL of DMEM/F-12 per well to dilute away any remaining
dispase.
Fig. 4. Dispase treatment of TeSR™ cultures on Matrigel™. H1 cells grown in mTeSR™1 on Matrigel™ were treated for
7 min with STEMCELL Technologies’ dispase. (a) Low magnification examination reveals areas of sharp brightness along
the edges of colonies that indicates the edges are beginning to roll back from treatment with the enzyme. (b) At higher
magnification the rolled edge that indicates that the enzyme treatment is sufficient becomes more obvious.
7. Add 2 mL/well of DMEM/F-12 or mTeSR™1/TeSR™2 and

scrape colonies off with a cell scraper.
8. Transfer the detached cell aggregates to a 15 mL conical tube
and rinse the well with an additional 2 mL of DMEM/F-12 or
mTeSR™1/TeSR™2 to collect any remaining aggregates. Add
the rinse to the 15 mL tube.
9. If cells are scraped in mTeSR™1 or TeSR™2, steps 9 and 10 are
not necessary. Ensure aggregates are of suitable size (see Note 6),
adjust the volume of medium for an appropriate split (see Note
7), and proceed to step 11. Centrifuge the 15 mL tube containing
the aggregates at 300 × g for 5 min at RT.
10. Aspirate the supernatant and resuspend the pellet in ~1 mL of
mTeSR™1 or TeSR™2 by gently pipetting up and down with a
P1000 micropipetter (1–2 times). Ensure that cells are main-
tained as aggregates (see Note 7). Adjust the volume of medium
to allow for an appropriate split (see Note 8 or alternatively per-
form a clump count as outlined in Subheading 3.4.2).
11. Remove excess Matrigel™ solution from the plate wells and
plate the hPSC aggregates with mTeSR™1 or TeSR™2 onto a
new plate coated with Matrigel™.
12. Place the plate in a 37°C incubator. On the incubator shelf,
move the plate in several quick, short, back-and-forth and side-
to-side motions to disperse cells evenly across the surface of
the wells.
3.4.2. Passaging Using An alternative to splitting cultures based on volume is to plate a

Clump Counts defined number of clumps according to the size of the well or dish
that is being seeded. This can be a valuable learning tool for those
new to hPSC culture because it aids in defining how much a sus-
pension should be pipetted to achieve optimally sized clumps. An
eyepiece micrometer placed in the microscope eyepiece is required
to enumerate clumps of appropriate size (~60 μm diameter in two
directions) that are likely to attach and grow.
1. Aliquot 40 μL of DMEM/F-12 into each of 2 wells of a
96-well flat-bottom plate.
2. With a fine-tipped marker, draw a “+” centered on the bottom
of these wells to serve as a counting grid.
3. Add 5 μL of a freshly mixed clump suspension to each well.
Count clumps that are approximately 60 μm or greater in
diameter in two directions (using a calibrated eyepiece microm-
eter). Perform duplicate counts, then average the results, and
calculate the total number (x) of clumps, where # of clumps
counted/5 μL =x clumps/total volume of suspension (μL).
4. Consult Table 1 as a guide for appropriate seeding densities
using the clump count method. Calculate the volume of clump
516 J. Moody
Table 1
Plating of clumps for hESC passaging
Plate or well size Target # of clumps/plate or well
100 mm dish 2,400 clumps

60 mm dish 1,000 clumps
Wells in a 6-well plate 350 clumps
suspension (y) required to seed new dishes. For example, to

seed 1 well of a 6-well dish, the volume of clump suspension
required for 350 clumps is calculated as follows: # of clumps
counted/5μL = 350 clumps/y μL.
5. Plate the calculated number of clumps into mTeSR™1 or
TeSR™2 in a coated plate as previously described (see Subheading
3.4.1, steps 11 and 12).
3.5. Alternate Matrix For an animal protein-free alternative to Matrigel™ it is possible to

for Use with Either use rhVitronectin as a culture surface (12). The following steps
mTeSR ®1 or TeSR™2 describe how to transition a culture from growth on Matrigel™ to
growth on rhVitronectin.
1. Coat rhVitronectin coated plates as described in
Subheading 3.1.2.
2. Prepare hPSC clumps as described in Subheading 3.4.1, (see
Note 7).
3. Rinse the rhVitronectin coated wells two times with PBS
immediately prior to plating cells.
4. Immediately plate the hPSC aggregates in mTeSR™1 or
TeSR™2 using a conservative split ratio (1:3) or higher
number of clumps (500 per well of a 6-well dish) than you
would use on Matrigel™.
5. Place the plate in a 37°C incubator. On the incubator shelf,
move the plate in several quick, short, back-and-forth and side-
to-side motions to disperse cells evenly across the surface of
the wells.
6. Perform daily media changes and observe the culture daily to
determine the optimal time for passaging (see Fig. 5 and
Note 9).
7. Subsequent passaging of cultures on rhVitronectin should be
performed without dispase treatment of the hPSC colonies
(see Note 9). Manually scrape the colonies with a cell scraper
and replate the aggregates onto new plates coated with
rhVitronectin.
Fig. 5. Colony morphology in TeSR™2 on rhVitronectin culture matrix. H9 cells grown in

TeSR™2 on rhVitronectin are less spread out than when grown on Matrigel™ and there-
fore appear smaller and more densely packed.
3.6. Cryopreserving The following is based on hPSC cultures in 6-well plates where
hPSCs Using initial clump seeding is adjusted so that wells are 60–70% confluent
mFreSR™ or Cryostor at time of cryopreservation.
CS10
1. Bring required amount of mFreSR™ to room temperature
(15–25°C). If using Cryostor™ CS10, warming to room tem-
perature is not required. 1 mL of cryopreservation medium
should be used for every well of a 6-well plate being frozen.
However, if the wells are at low density (less than 50%
confluent), 1 mL of medium may be used for every 2 wells.
2. In the hPSC culture to be cryopreserved, use a microscope to
visually identify regions of differentiation (see Note 10). Mark
these using a felt tip or lens marker on the bottom of the plate.
Remove regions of differentiation by scraping with a pipette
tip or by aspiration.
3. Aspirate remaining medium from wells.
4. Rinse wells with 2 mL of DMEM/F-12 and aspirate.
5. Add 1 mL per well of dispase at a concentration of 1 mg/mL.
Place at 37°C for 7 min if using mTeSR™1 or for 3–4 min if
using TeSR™2 as previously described.
6. Remove dispase and gently rinse each well two to three times
with 2 mL of DMEM/F12 per well to dilute away any remain-
ing dispase.
7. Add 2 mL/well of DMEM/F12 or mTeSR™1/TeSR™2 and
scrape colonies off using a cell scraper. Take care to keep the
clumps as big as possible.
518 J. Moody
8. Transfer the detached cell aggregates into a 15 mL conical

tube and rinse the wells with additional 2 mL DMEM/F12 or
mTeSR™1/TeSR™2 to collect any remaining aggregates. Add
the rinse to the 15 mL tube containing the cell aggregates.
9. Centrifuge the 15 mL tube containing the aggregates at 300 × g
for 5 min at room temperature (15–25°C). Label the cryo-
tubes while cells are centrifuging.
10. Gently aspirate the supernatant taking care to keep the cell pel-
let intact.
11. Using a 2 mL pipette, gently resuspend the pellet in mFreSR™
or Cryostor™CS10, taking care to leave the clumps larger than
would normally be done for passaging.
12. Gently flick the tube to mix the suspension and mFreSR™ or
Cryostor™CS10. Then transfer 1 mL of clumps into each
labeled cryovial using a 2 mL pipette. Draw up 1 mL at a time
and aliquot 1 mL/tube. Mix gently before taking each aliquot.
This will ensure even distribution of clumps between the vials.
13. Place vials into an isopropanol freezing container and place the
container at −80°C to −150°C overnight. Transfer to a liquid
nitrogen vapor tank or liquid nitrogen the next day.
14. Alternatively, cells can be frozen using an isopropanol freezing
container and a multi-step protocol: −20°C for 2 h, followed
by −80°C for 2 h, followed by storage at LN2 temperature
(−135°C).
4. Notes
1. The appropriate dilution of BD Matrigel™ hESC-qualified

matrix may vary depending on the batch used. Consult the
LOT-specific product insert supplied to determine the appro-
priate aliquot volume. Record volume and LOT# on each
vial.
2. When plating clumps of cells, it is important to evenly distrib-
ute the clumps in the wells and to minimize disturbance of the
plate until they have settled (24 h). Uneven distribution of cell
clumps (i.e., all in the center) will result in increased rate of
spontaneous differentiation of hPSCs.
3. hPSCs require a full daily medium change for optimal growth.
hPSC lines will usually tolerate once a week double feeding
(adding twice the required volume of medium). For instance,
it is possible to perform a double feed on Friday, with the next
medium change on Sunday. However, it is not recommended
to either go longer than 1 day without a medium change or
feed the cultures every other day continuously.
4. hPSCs grown in mTeSR™1 or TeSR™2 are ready to passage

when the colonies are large, beginning to merge, and have cen-
ters that are dense and phase-bright compared to their edges
(see Figs. 1a and 5). If greater than 50% of colonies in the cul-
ture have developed bright centers, the culture should be pas-
saged within 24 h. In both media formulations, if colonies are
passaged too early or too frequently, the cells may not attach
well, yields will be decreased and cells may start to differenti-
ate. If colonies are passaged too late, the culture will begin to
show signs of differentiation (characterized by the emergence
of cell types with different morphologies). Depending on the
size and density of seeded clumps, cultures are usually passaged
5–7 days after seeding in mTeSR®1 and passaged 4–6 days after
seeding in TeSR™2. We generally note that cells grow slightly
faster in TeSR™2 than in mTeSR®1.
5. No adaptation step is required when seeding cells from feeder
or conditioned medium cultures to mTeSR™1 or TeSR™2.
Simply replate hPSC clumps into mTeSR™1 or TeSR™2 on
BD Matrigel™-coated plates at the time of passaging. It is
important that the starting culture is of high quality and is
primarily undifferentiated. The morphologic hallmarks of dif-
ferentiation should appear in less than 20% of colonies in a
healthy culture. Cultures that have large amounts of differenti-
ated cells will continue to yield differentiated cell types after
transition into mTeSR®1 or TeSR™2.
6. Cells cultured in TeSR™2 are more sensitive to dispase than
cells cultured in mTeSR™1. If using TeSR™2, the dispase
incubation period should be limited to 3–4 min. The times
suggested in this protocol are based on the use of dispase from
STEMCELL Technologies. Specific unit activity can differ
between suppliers so the use of dispase from alternate suppliers
may require optimization and more or less time.
7. Preparation of a uniform suspension of suitable sized clumps
for passaging is very important for the successful culture of
hPSCs. If the clumps are too large, an increased rate of differ-
entiation within the colonies may occur. If the clumps are too
small with many single cells present, cell survival will be com-
promised. Following the dispase and washing steps, one or two
gentle draws with a micropipetter and P1000 tip should be
sufficient to generate appropriately sized clumps for passaging
(approximately 60 μm in two directions as measured by an
eyepiece micrometer). If the clumps are the correct size, the
majority will remain in suspension after gently swirling the
tube. If large clumps are present that rapidly sink to the bot-
tom of the tube, perform one or two more gentle draws with
the micropipetter or pipette.
520 J. Moody
8. If the colonies are at an optimal density, defined at approxi-

mately 75% confluent at the time of passage, the cells in
mTeSR™1 and TeSR™2 on Matrigel™ can be split every
4–7 days using 1:6 to 1:10 splits (i.e., the clumps from 1-well
of a 6-well plate can be replated in 6–10 wells of the same
sized plate). If the colonies are too dense or too sparse, adjust
the next split ratio accordingly. Culture density is a critical
aspect of maintaining hPSCs in mTeSR™1 or TeSR™2.
Cultures that are either too sparsely or too densely populated
can lead to differentiation.
9. Colonies grown on rhVitronectin as a culture matrix will gen-
erally be smaller and less spread out than colonies on Matrigel™
(see Fig. 5). Colony centers tend to get dense very quickly and
hence cultures on rhVitronectin may need more frequent pas-
saging than those on Matrigel™ (approximately every 4 days).
Using rhVitronectin as a culture matrix generally affords less
expansion than cultures on Matrigel™. While this may be cell
line specific the recommended split ratios are in the range of
1:3 to 1:5. To sustain the culture through multiple passages on
rhVitronectin coated plates, do not use dispase when passag-
ing. Mechanically scrape the cells, break into suitable sized
aggregates and replate into mTeSR®1 or TeSR™2 medium on
freshly coated plates.
10. Before cryopreservation, hPSCs should be of high quality, that
is, primarily undifferentiated with less than 20% of the culture
showing morphological signs of differentiation (examples in
Fig. 2a, b) Cryopreservation should be done approximately
1 day before the cells are ready to passage. hPSCs will have
improved survival following thawing if cryopreserved as large
clumps.
Acknowledgments
The author would like to acknowledge to contributions of Dr.

Tennielle Ludwig, Dr. Michael O’Connor, Dr. Chris Lannon, and
Debbie King to the development of these protocols.
References
1. Thomson JA, Itskovitz-Eldor J, Shapiro SS, from human blastocysts: somatic differentia-
Waknitz MA, Swiergiel JJ, Marshall VS, tion in vitro. Nat Biotechnol 18:399–404
Jones JM (1998) Embryonic stem cell lines 3. Takahashi K, Tanabe K, Ohnuki M, Narita M,
derived from human blastocysts. Science 282: Ichisaka T, Tomoda K, Yamanaka S (2007)
1145–1147 Induction of pluripotent stem cells from adult
2. Reubinoff BE, Pera MF, Fong CY, Trounson human fibroblasts by defined factors. Cell
A, Bongso A (2000) Embryonic stem cell lines 131:861–872
4. Yu J, Vodyanik MA, Smuga-Otto K, 9. Okita K, Yamanaka S (2010) Induction of

Antosiewicz-Bourget J, Frane JL, Tian S, Nie pluripotency by defined factors. Exp Cell Res
J, Jonsdottir GA, Ruotti V, Stewart R, Slukvin 316(16):2565–2570
II, Thomson JA (2007) Induced pluripotent 10. Ludwig TE, Bergendahl V, Levenstein ME,
stem cell lines derived from human somatic Yu J, Probasco MD, Thomson JA (2006)
cells. Science 318:1917–1920 Feeder-independent culture of human
5. Park IH, Zhao R, West JA, Yabuuchi A, Hu H, embryonic stem cells. Nat Methods
Ince TA, Lerou PH, Lensch MW, Daley GQ 3:637–646
(2008) Reprogramming of human somatic 11. Ludwig TE, Levenstein ME, Jones JM,
cells to pluripotency with defined factors. Berggren WT, Mitchen ER, Frane JL, Crandall
Nature 451:141–146 LJ, Daigh CA, Conard KR, Piekarczyk MS,
6. Sun N, Panetta NJ, Gupta DM, Wilson KD, Llanas RA, Thomson JA (2006) Derivation of
Lee A, Jia F, Hu S, Cherry AM, Robbins RC, human embryonic stem cells in defined condi-
Longaker MT, Wu JC (2009) Feeder-free deri- tions. Nat Biotechnol 24:185–187
vation of induced pluripotent stem cells from 12. Braam SR, Zeinstra L, Litjens S, Ward-van
adult human adipose stem cells. Proc Natl Oostwaard D, van den Brink S, van Laake L,
Acad Sci USA 106:15720–15725 Lebrin F, Kats P, Hochstenbach R, Passier R,
7. Hochedlinger K, Plath K (2009) Epigenetic Sonnenberg A, Mummery CL (2008)
reprogramming and induced pluripotency. Recombinant vitronectin is a functionally
Development 136:509–523 defined substrate that supports human embry-
8. Lengner CJ (2010) iPS cell technology in regen- onic stem cell self-renewal via alphavbeta5
erative medicine. Ann N Y Acad Sci 1192:38–44 integrin. Stem Cells 26:2257–2265
Chapter 32
Formation of Embryoid Bodies from Human Pluripotent Stem

Cells Using AggreWell™ Plates
Jennifer Antonchuk
Abstract
Many human embryonic stem (hES) and induced pluripotent stem (hiPS) cell differentiation protocols
begin with the formation of three-dimensional aggregates of cells called embryoid bodies (EBs). Traditional
EB formation methods result in a heterogeneous population of EB sizes and shapes, which then undergo
heterogeneous differentiation efficiencies. AggreWellTM400 and AggreWellTM800 use the spin-EB method
to force the aggregation of a defined number of cells, thereby controlling EB size and generating a popula-
tion of uniform EBs. Moreover, the dense array of microwells on the bottom surface of AggreWellTM400
provide for the rapid and simple production of thousands of EBs at a time.
Key words: Human embryonic stem cell, Embryoid bodies, Induced pluripotent stem cells
1. Introduction
Pluripotent stem cells (PSCs), including human embryonic stem

(hES) and induced pluripotent stem (hiPS) cells have the capacity
to differentiate to all three germ layers (endoderm, mesoderm, and
ectoderm). This has been shown in vivo by the formation of hES-
or hiPS-derived multi-lineage teratomas, and also in vitro by the
differentiation of hES- or hiPS-derived cells, wherein examples of
mature end cells representing all three germ layers are found.
Most in vitro differentiation protocols begin with the forma-
tion of three-dimensional aggregates of cells called embryoid bod-
ies (EBs). The traditional method of forming human EBs involves
enzymatic dissociation followed by scraping of adherent PSC colo-
nies to release large clumps of cells, and placing the resulting aggre-
gates into suspension culture. Under these conditions, the
523
524 J. Antonchuk
aggregates will over the course of the next several days spontane-
ously form into spherical EBs, which will then develop to contain
cells of the three germ layers (1). However, EBs formed by this
method are heterogeneous in size, and that heterogeneity limits
the development of standardized in vitro differentiation
protocols.
To address this issue, the “spin-EB” protocol was developed
(2), whereby defined numbers of hES/hiPS cells are placed into
close association and allowed to form EBs. By this method, it has
been shown that human EB size directly affects subsequent differ-
entiation efficiencies (3–5). For example, Mohr et al. (4) showed
that cardiomyocyte differentiation was most efficient when the
starting EBs were approximately 100 μm in diameter.
AggreWell™ plates were developed to increase the ease and
throughput of spin-EB formation (6). AggreWell™400 plates con-
tain an array of inverse pyramidal microwells of 400 μm diameter,
whereas AggreWellTM800 plates contain an array of microwells of
800 μm diameter. Both of these sizes of microwells can be used to
aggregate hES or hiPS cells into EBs; the diameter (and volume)
of the microwell determines the maximum size of EBs possible, as
described below.
EB formation is accomplished by adding a well-dispersed sus-
pension of single cells of known density into the plate well, centri-
fuging the plate gently to force cells evenly into the microwells,
and then culturing for 24 h to allow EB formation. By this method,
EBs are generated which are spherical, uniform in size, and robust
to handling.
It is easy to control the size of EBs generated by the AggreWellTM
method, simply by adjusting the density of the cell suspension
added to the overlying plate well. Microwell capacity in
AggreWellTM400 allows for up to 3,000 cells to be aggregated in
each microwell. For even larger EB sizes, AggreWelllTM800 con-
tains microwells of 800 μm diameter, and their larger capacity can
be used to form EBs in the range of 1,000–20,000 cells. Although
this chapter focuses on the use of AggreWellTM400, the method of
use for AggreWellTM800 is very similar, and notes are provided to
assist the reader in using AggreWellTM800 where necessary.
Each of the 8 wells on an AggreWell™400 plate contains
approximately 1,200 microwells, and can therefore be used to gen-
erate approximately 1,200 EBs of defined size. Each well of
AggreWellTM800 can similarly generate approximately 300 EBs of
larger size. This simple, high throughput method for standardized
EB formation can help to improve the efficiency of in vitro differ-
entiation protocols by controlling EB size and also to facilitate
transfer of differentiation protocols between laboratories.
32 Human EBs Using AggreWell 525
2. Materials
2.1. Preparation of 1. AggreWell™400 plate (STEMCELL Technologies, Vancouver

AggreWellTM400 Plates Canada, Catalog #27845). Each plate contains 8 microwell-
containing wells, each with approximately 1,200 microwells of
400 μm diameter.
2. AggreWellTM800 plate (STEMCELL, Catalog #27865). Each
plate contains 8 microwell-containing wells, each with approx-
imately 300 microwells of 800 μm diameter (Optional, see
Note 1).
3. DMEM/F-12 (STEMCELL, Catalog #36254). Store at 4°C.
4. AggreWell Medium (STEMCELL, Catalog #05893). Store at
−20°C until ready to use, then thaw at 4°C overnight and store
at 4°C for up to 4 weeks (see Note 2).
5. Y-27632 ROCK Inhibitor (STEMCELL, Catalog #07171).
Reconstitute lyophilized powder in water to 5 mM. Store
20–100 μL aliquots at −20°C.
6. Pluronic F127 (Sigma Catalog #P2443). 5% (w/v) solution in
water, autoclave to sterilize (Optional, see Note 3).
7. Centrifuge with a swinging bucket rotor fitted with plate
holders.
2.2. Preparation of a 1. hES or hiPS cells (see Note 4).

Single Cell Suspension 2. Accutase™ (STEMCELL, Catalog #07920). Store at −20°C
of Undifferentiated until use, then thaw at 4°C overnight and store at 4°C for up
hES or hiPS Cells to 2 months.
3. DMEM/F-12.
4. AggreWellTM Medium supplemented with 10 μM Y-27632
ROCK Inhibitor.
5. Trypan Blue (STEMCELL, Catalog #07050).
2.3. Generation of 1. AggreWellTM400 plate prepared in Subheading 3.1.

Embryoid Bodies from 2. Single cell suspension of hES or hiPS cells prepared in
hES or hiPS Cells Subheading 3.2.
Using AggreWell™400
2.4. Harvesting Human 1. 40 μm Cell strainer (STEMCELL, Catalog #27305).

Embryoid Bodies from 2. 6-Well ultra low adherence plate (STEMCELL, Catalog
AggreWell™ Plates #27145).
3. AggreWell Medium.
4. Additional growth factors (see Note 5).
5. Large bore tips (e.g., Rainin Catalog #HR-1000 WS). Required
only if harvesting EBs of 3,000 cells or larger (see Note 6).
526 J. Antonchuk
3. Methods
3.1. Preparation of 1. Pre-warm aliquots of AggreWellTM Medium EB formation

AggreWell™ Plates medium (e.g., AggreWell Medium) and DMEM/F-12 to
room temperature. Approximately 2.5 mL of EB formation
medium should be aliquoted per well of AggreWellTM to be
used. Supplement EB formation medium with Y-27632 ROCK
Inhibitor to a final concentration of 10 μM.
2. Aseptically remove the AggreWell™400 plate from the packag-
ing in a biosafety cabinent.
3. Rinse each well to be used with 1 mL of DMEM/F-12 and
aspirate to remove.
4. Add 0.5 mL of EB formation medium to each well of the
AggreWell plate that will be used (see Note 7).
5. Centrifuge the AggreWell™ plate at 2,000 × g (or at maximum
speed) for 5 min in a swinging bucket rotor that is fitted with a
plate holder, to remove any air bubbles from the microwells. Plates
must be well balanced at this speed. It is recommended to make a
balance plate using a standard 24-well plate filled with water to
match the weight and position of the AggreWell™ plate.
6. Observe plate using inverted microscope, to confirm that bub-
bles have been removed from microwells (see Note 8).
7. Set plate aside in tissue culture hood while preparing a single
cell suspension of hES or hiPS cells.
3.2. Preparation of a 1. Pre-warm AggreWell™ Medium, ACCUTASE™, and

Single Cell Suspension DMEM/F-12 to room temperature (15–25°C). Supplement
of Undifferentiated AggreWell™ Medium with ROCK inhibitor Y-27632, to a
hES or hiPS Cells final concentration of 10 μM.
2. Aspirate the maintenance medium from the hES or hiPS cell cul-
ture plate(s), and rinse the cells once with 2 mL of DMEM/F12.
3. Add 2 mL of ACCUTASE™ per 100 mm dish or enough to
cover the cells if other sized tissue cultureware is used.
4. Incubate at 37°C until cells detach easily from the plate with
gentle shaking. This usually takes 5–10 min. Microscopically
inspect the plate to ensure that cells have detached.
5. Gently pipette the cell suspension five to six times with a sero-
logical pipette to ensure any remaining clumps are fully disso-
ciated and to dislodge any cells that are still attached to the
surface of the dish.
6. Transfer the cells to a 15 or 50 mL conical tube.
7. Rinse the plate with at least 5 mL of DMEM/F-12 per 1 mL
of ACCUTASE™ used, and add the rinsing solution to the
tube containing the cell suspension.
8. Optional: Residual clumps of cells can be removed by passing

the cell suspension through a 40 μm cell strainer and retaining
the filtrate.
9. Centrifuge the cells at 300 × g for 5 min at room temperature
(15–25°C). Aspirate the supernatant and resuspend the pellet
in a small volume of AggreWell™ Medium supplemented with
10 μM Y-27632, such that the cell concentration will be
approximately 0.5–1.0 × 107 cells/mL. For example, resuspend
the pellet in 1 mL of medium per 100 mm dish harvested.
10. Count viable cells using Trypan Blue. Dilute a sample of the
resuspended cells 1:9 (1/10 dilution) in trypan blue and mix
gently. Count viable, unstained cells using a hemacytometer.
11. Calculate the viable cell concentration as follows:
Cell concentration = (average count for 16 squares / 0.1 mm3 )
× (Dilution factor = 10) × (104 ).
3.3. Generation of 1. Calculate the number of undifferentiated hES or hiPS cells

Embryoid Bodies from needed per well, according the desired EB size. To calculate
hES or hiPS Cells this, refer to Table 1 or use the following formula:
Using AggreWell™400 R = E×M,
Table 1
Number of PSCs (hESCs or hiPSCs) required to generate various sized EBs using
AggreWell™400 or AggreWellTM800
Required number of cells per well

Desired number AggreWell™400 each well contains AggreWell™800 each well contains
of cells per EB approximately 1,200 microwells approximately 300 microwells
50 6 × 104 cells –
100 1.2 × 105 cells –
5
200 2.4 × 10 cells –
5
500 6 × 10 cells –
1,000 1.2 × 106 cells 3.0 × 105 cells
2,000 2.4 × 106 cells 6.0 × 105 cells
3,000 3.6 × 106 cells 9.0 × 105 cells
4,000 – 1.2 × 106 cells
5,000 – 1.5 × 106 cells
10,000 – 3.0 × 106 cells
15,000 – 4.5 × 106 cells
20,000 – 6.0 × 106 cells
528 J. Antonchuk
Fig. 1. Even distribution of hESCs in microwells of an AggreWell™400 plate. An array of

microwells is shown, each containing approximately 2,000 hES cells, generated by plat-
ing 2.4 × 106 hES cells to a well of an AggreWell™400 plate, followed by gentle centrifu-
gation to capture the cells in the microwells. Photo taken at ×40 magnification.
where R is the required number of cells to add to a well, E is

the desired size of each EB (number of cells per EB), M is the
number of microwells per well. For AggreWellTM400 M = 1,200;
for AggreWellTM800 M = 300.
2. Without removing the EB formation medium added previ-
ously (see Subheading 3.1), add the appropriate volume of
undifferentiated cells (generated according to Subheading 3.2
and calculated in step 1 above) to a well of the AggreWell™400.
After adding cells, immediately pipette gently several times to
distribute the cells evenly throughout the well.
3. Adjust the medium to a final volume of 2 mL per well using
EB formation medium (e.g., AggreWell™ Medium) supple-
mented with 10 μM Y-27632. Pipette gently again to reestab-
lish an even distribution of cells in the well (see Note 9).
4. Centrifuge the AggreWell™ plate at 300 × g for 3 min to cap-
ture the cells in the microwells.
5. Examine the AggreWell™ plate under a microscope to verify
that cells are evenly distributed among the microwells (see
Fig. 1, and Notes 9 and 10)
6. Incubate the cells at 37°C with 5% CO2 and 95% humidity for
~24 h.
3.4. Harvesting Human 1. After 24 h of culture, EBs should be visible inside the microw-
Embryoid Bodies from ells of the AggreWell™400 plate (Fig. 2) (see Note 11).
AggreWell™ Plates 2. Pre-warm EB suspension culture medium (e.g., AggreWellTM
Medium) and DMEM/F-12 to 37°C.
Fig. 2. EBs formed in AggreWell™400 plate, ready to be harvested. Each microwell was
seeded with approximately 2,000 hESCs in AggreWell™ Medium supplemented with
10 μM Y-27632 ROCK Inhibitor. After 24 h, spherical aggregates are clearly visible within
each microwell. Photo taken at ×100 magnification.
3. Harvest EBs from microwells by firmly pipetting medium in

the well up and down two to three times with a micropipettor
outfitted with a 1 mL disposable tip to dislodge most of the
EBs (see Note 6).
4. Pass the EB suspension through an inverted 40 μm cell strainer
on top of a 50 mL conical tube to remove single cells. The aggre-
gates will remain on top of the inverted cell strainer. The unwanted
single cells will flow through into the 50 mL waste tube.
5. Wash the AggreWell™ surface five times with 1 mL each of
DMEM/F-12, pipetting across the entire surface to dislodge
all aggregates. Collect washes and pass over the inverted 40 μm
cell strainer used in step 4.
6. Turn the cell strainer right-side up over a fresh 50 mL conical
tube and collect the aggregates by washing with 2–5 mL of EB
suspension culture medium per well of the AggreWell™400
plate used. Use the same medium that will be used in the sub-
sequent EB differentiation (see Note 5).
7. Check the AggreWell™ plate under the microscope to ensure
that all aggregates have been removed from the wells. Repeat
washing if necessary (steps 5 and 6) (see Note 12).
8. Optional: Count the EBs to determine efficiency of yield com-
pared to the expected yield of approximately 1,200 EBs per
well of an AggreWell™400 plate. Place 50 μL of the evenly
distributed EB suspension into a flat-bottomed 96-well plate.
530 J. Antonchuk
Count at 20–100× magnification and calculate EB yield as

follows
Count (in 50μL) / 50 × total volume(μL) = total # of EBs.
9. Plate the EBs at £1,000 EBs/well in an ultra-low adherence

6-well plate (Catalog #27145) in the same EB suspension cul-
ture medium as used in step 2. Place the plate in the 37°C
incubator and move the plate in quick but gentle side-to-side,
forward-to-back motions to evenly distribute the EBs within
the wells (see Notes 13 and 14).
10. Culture the cells at 37°C with 5% CO2 and 95% humidity, and
continue with differentiation protocol of choice (see Notes 14
and 15).
4. Notes
1. AggreWellTM400 plates can be used to make EBs of 50–3,000

cells each. AggreWellTM800 plates can be used to make EBs of
1,000–20,000 cells each. The main protocol described here is
for AggreWellTM400, however notes will be provided wherever
the AggreWellTM800 protocol is divergent.
2. The type of culture medium to use for EB formation depends
on both the conditions used to maintain the hES or hiPS cells
prior to EB formation and the specific differentiation protocol
to be used following EB formation. For cells grown previously
in mTeSR™1 or TeSR™2, DMEM/F12 with 20% FBS (tradi-
tional “EB media”) is not recommended; serum-free medium
is preferable. EBs can be formed in mTeSR™1 or TeSR™2,
however these maintenance media contain pluripotency factors
which may prevent efficient downstream differentiation.
AggreWell™ Medium contains lower concentrations of pluri-
potency factors and was designed specifically to support the
formation and subsequent survival of AggreWellTM-generated
EBs made from mTeSR™1- or TeSR™2-cultured cells.
3. Some cell types (e.g., mouse ES cells) may adhere slightly to
the microwell surface, making them difficult to retrieve. To
prevent this, pre-coat the microwells with Pluronic F127 (Sigma
Catalog #P2443, 5% (w/v) solution in water, autoclaved), prior
to preparing the plate as outlined in Subheading 3.1. Add 1 mL
of the Pluronic F127 solution per well of an AggreWell™ plate
and centrifuge at 2,000 × g for 5 min. Incubate for 30 min at
room temperature (15–25°C). Then remove the Pluronic solu-
tion and wash with 1 mL DMEM/F-12. Continue as per the
protocol (Subheading 3.1).
4. AggreWell™400 plates can be used to form EBs from hES or

hiPS cells that were maintained via a variety of different meth-
ods, including cells grown on mouse embryonic fibroblast
(MEF) feeders, or those grown in feeder-free conditions such
as in mTeSR™1 (STEMCELL Technologies Cat #05850) or
TeSR™2 (STEMCELL Technologies Cat #05860) media on
hES-qualified BD Matrigel™ Matrix (BD Cat #354277). When
generating EBs, it is essential that a high quality starting popu-
lation of undifferentiated hES or hiPS cells is used; avoid using
cultures with greater than 15% differentiated cells. Refer to
Subheading 3.4 prior to starting, to determine the required
number of hES or hiPS cells needed. Expected yields of hES
cells from cultures using mTeSR™1 or TeSR™2 are approxi-
mately 1 × 106 cells per well of a 6-well plate, 2 × 106 cells per
60 mm dish, or 0.5–1.0 × 107 cells per 100 mm dish. Protocols
for culturing hPSCs using mTeSR®1 or TeSR™2 are presented
in this edition and additional technical information is available
at http://www.stemcell.com.
5. The addition of germ layer-specific or lineage-specific induc-
tion factors is recommended for culture of EBs after formation
in AggreWell™. The specific combinations of factors to be
added to the differentiation media will depend on the differen-
tiation protocol of choice, and is outside the scope of this book
chapter.
6. When harvesting EBs of 3,000 cells or greater, use large bore
tips (e.g., Rainin Catalog #HR-1000 WS) or aseptically cut the
tip of a standard 1 mL tip with sterile scissors to increase the
bore size. However, if using self-cut tips, there may not be
enough force to dislodge the EBs—in that case, use a regular
1 mL tip to dislodge the EBs, and then use the wider (cut) tip
to collect the EBs. Be careful not to pass large EBs though a
regular 1 mL tip, or they may break apart.
7. Ensure the medium is placed in the AggreWell™ wells contain-
ing microwells. AggreWell™ wells are located in wells from B2
to C5 only. Do not use empty wells for EB formation.
8. When medium is added to the AggreWellTM400 or
AggreWellTM800 plate, surface tension will prevent it from
entering the microwells. This can be observed under the micro-
scope as black space in the microwells. If bubbles are not
removed from the plate, cells will not be able to enter the
microwells. Plates must be prepared in advance of adding cells,
by centrifugation of the plate at 2,000 × g (or maximum speed)
in a swinging bucket centrifuge equipped with plate holder
attachments. Maximum speed will depend on the specific cen-
trifuge type, rotor and plate holders used, and is commonly
listed directly on the plate holder attachments.
532 J. Antonchuk
9. To facilitate even distribution of cells, the required volume of

hES/hiPS single cell suspension should be added dropwise, if
possible. Immediately after adding the cell suspension, pipette
several times with a micropipettor and 1 mL pipette tip, to gen-
erate an evenly distributed cell suspension. After topping up
the well with medium, pipette several times again to ensure the
final suspension contains an evenly distributed cell suspension.
Even distribution of the cells at this stage will ensure even filling
of the underlying microwells, and thereby equal EB sizes.
10. Cells should not overflow microwell capacity. AggreWellTM400
plates have an hES/hiPS cell capacity of up to 3,000 cells per
microwell. If using larger cell types, the capacity in cell number
may be reduced. When cells overflow the microwells, con-
glomerate aggregates can form among the cells in multiple
microwells. When making EBs near the maximum size, a
slightly harder centrifuge (300 × g for 5 min) may help to pack
cells tightly into the microwells. Alternatively, move to a larger
microwell size, such as AggreWellTM800, which has up to
20,000 hES/hiPS cell capacity.
11. Note that it is possible not all cells will be incorporated into the
EB. At least 50% of input hES/hiPS cells within each microw-
ell should incorporate into the EB. The level of incorporation
can be affected by the quality of the starting undifferentiated
cell population. Starting cells should be in log growth phase,
and largely free of differentiated cells. If incorporation is con-
sistently less than 50%, consider thawing an earlier passage of
hES cells. The inclusion of Y-27632 ROCK Inhibitor will also
significantly improve survival of single hES cells, and thereby
their inclusion into the EB (7).
12. Right-handed operators often find that they miss the EBs at the
lower right-hand corner of a well, whereas left-handed opera-
tors often miss the EBs at the lower left-hand corner. Ensure
that all areas of the well have been well rinsed in steps 3–6 of
Subheading 3.4. If necessary, rotate the AggreWell™ plate to
harvest EBs from hard-to-reach areas. Check the plate under
the microscope to ensure that all EBs have been removed.
13. After harvesting EBs from AggreWellTM400, they sometimes
have a tendency to adhere to one other, forming “doublets” or
even long strings of EBs. To prevent this, use media that has
been pre-warmed to 37°C, keep EB density low (£1,000 EBs/
well in a 6-well ULA plate), and distribute the EBs evenly in
the dish to minimize contact with each other.
14. Do not add Y27632 Rock Inhibitor to EB suspension culture,
as it may increase cellular adhesion. ROCK Inhibitor is only
required during EB formation and not for subsequent EB
culture.
15. Gentle shaking of the plate 1–2 times per day for the first few
days of suspension culture may help to keep the EBs separated
in the well, and avoid EBs sticking together. Constant shaking
or stirring of the plate can also be used. For example, EBs can
be cultured in ultra-low adherent flasks (e.g., Corning Cat
#3814/3815) on an orbital shaker placed inside the incubator.
Stirring rate will need to be optimized depending on the size
of aggregates and specific differentiation protocol. Avoid cen-
trifugation of EBs, as this may also cause them to stick together.
If a medium change is required, allow the EBs to settle to the
bottom of a 15 mL conical tube by leaving them undisturbed
for 5 min at room temperature. Then carefully remove the
medium and replace with fresh medium. When EBs are cul-
tured in DMEM or Iscove’s Modified Dulbecco’s Medium
(IMDM) supplemented with 10–20% FBS, EB integrity can be
lost. For better maintenance of EB integrity, culture EBs in
AggreWell™ Medium, or other serum-free alternatives. This is
particularly important if the starting hES/hiPS cells were
maintained in serum-free culture media such as mTeSR™1 or
TeSR™2.
References
1. Itskovitz-Eldor J, Schuldiner M, Karsenti D, embryoid body size in order to regulate cardiac

Eden A, Yanuka O, Amit M, Soreq H, Benvisty N differentiation of human embryonic stem cells.
(2000) Differentiation of human embryonic stem Biomaterials 7:1885–1893
cells into embryoid bodies comprising the three 5. Bratt-Leal AM, Carpenedo RL, McDevitt TC
embryonic germ layers. Mol Med 6(2):88–95 (2009) Engineering the embryoid body
2. Ng E, Davis R, Azzola L, Stanley E, Elefanty A microenvironment to direct embryonic stem
(2005) Forced aggregation of defined numbers cell differentiation. Biotechnol Prog 25:43–51
of human embryonic stem cells into embryoid 6. Ungrin MD, Joshi C, Nica A, Bauwens C,
bodies fosters robust, reproducible hematopoi- Zandstra PW (2008) Reproducible, ultra-high-
etic differentiation. Blood 106:1601–1603 throughput formation of multicellular organi-
3. Bauwens CL, Peerani R, Niebruegge S, zation from single cell suspension-derived
Woodhouse KA, Kumacheva E, Husain M, human embryonic stem cell aggregates. PLoS
Zandstra PW (2008) Control of human embry- One 3(2):e1565
onic stem cell colony and aggregate size het- 7. Watanabe K, Ueno M, Kamiya D, Nishiyama
erogeneity influences differentiation trajectories. A, Matsumura M, Wataya T, Takahashi JB,
Stem Cells 9:2300–2310 Nishikawa S, Nishikawa S, Muguruma K, Sasai
4. Mohr JC, Zhang J, Azarin SM, Soerens AG, de Y (2007) A ROCK inhibitor permits survival of
Pablo JJ, Thomson JA, Lyons GE, Palecek SP, dissociated human embryonic stem cells. Nat
Kamp TJ (2010) The microwell control of Biotechnol 25:681–686
Chapter 33
Techniques in Embryoid Body Formation from Human

Pluripotent Stem Cells
Nirupama K. Shevde and Amber A. Mael
Abstract
Embryoid bodies (EBs) can be generated by culturing human pluripotent stem cells in ultra-low attach-
ment culture vessels, under conditions that are adverse to pluripotency and proliferation. EBs generated in
suspension cultures are capable of differentiating into cells of the ectoderm, mesoderm, and endoderm. In
this chapter, we describe techniques for generation of EBs from human pluripotent stem cells. Once
formed, the EBs can then be dissociated using specific enzymes to acquire a single cell population that has
the potential to differentiate into cells of all three germ layers. This population can then be cultured in
specialized conditions to obtain progenitor cells of specific lineages. Pure populations of progenitor cells
generated on a large scale basis can be used for research, drug discovery/development, and cellular trans-
plantation therapy.
Key words: Human pluripotent stem cells, Embryoid bodies, Embryoid body formation,
Differentiation, Suspension cultures, Mesenchymal stem cells, Induced pluripotent stem cells,
Embryonic stem cells, Germ layers
1. Introduction
Human pluripotent stem cells such as embryonic stem (hES) cells

and induced pluripotent stem (iPS) cells have the ability of long-term
self-renewal in vitro. They are also capable of differentiating into cells
all three germ layers—ectoderm, mesoderm, and endoderm. Common
techniques used to induce differentiation of these cells in vitro include
monolayer culture on defined matrices (1), coculture with hetero-
typic cell types (2) and the formation of cell aggregates grown in
suspension termed embryoid bodies (EBs) (3). When human pluri-
potent stem cells are cultured in suspension in the absence of factors/
microenvironment that promote pluripotency and continuous
535
536 N.K. Shevde and A.A. Mael
proliferation, they spontaneously develop into three-dimensional

spheroidal cell aggregates known as embryoid bodies (4, 5).
Many aspects of the lineage-specific differentiation programs
observed within the EBs reflect those found in the embryo, indi-
cating that this model system provides access to normal develop-
ment in early cell populations. Recent studies involving the
differentiation of genetically altered embryonic stem cells highlight
the potential of this in vitro differentiation system for defining the
function of genes in early development (6). Global DNA microar-
ray analysis indicates that EBs temporally express genes in a man-
ner that recapitulates the sequence of normal development from
primitive ectoderm formation, to gastrulation, and eventual early
cell specification prior to organogenesis (7, 8). Since formation of
EBs is a consistent and reproducible technique to induce differen-
tiation, it provides a valuable tool for mechanistic studies of embry-
ological development in vitro, including the examination of the
effects of morphogenic cues on cell fate determination (9).
In this chapter, we delineate the methodology for generation of
EBs from human pluripotent stem cells. This technique allows stem
cell colonies cocultured with mouse embryonic fibroblasts (MEF)
as well as in the feeder-independent system (10, 11) to generate
EBs in various media and ultra-low attachment culture vessels. EBs
in suspension culture undergo a process of organization ranging
from spheroids with dense cores to cystic structures filled with fluid.
The well-organized or cystic EBs are then dissociated using specific
enzymes to acquire single cells. These cells are then cultured in
specialty media and growth factors/cytokines to direct them
towards progenitors of specific lineages. With subsequent passag-
ing of these cells it is possible to obtain a large number of progeni-
tors representing cells of a specific germ layer for experimentation.
In recent years, the EB generation process has been greatly
enhanced using techniques such as bioreactor cultures, hydrogel
embedding systems, and specifically the Aggrewell™ technique (12)
that offers an easy and standardized approach to generate uniform
size EBs on a large scale and in a consistently reproducible manner.
Such technological advancements will enable large-scale pro-
duction of pure populations of progenitor cells that can be used for
research, drug discovery/development, and cellular transplanta-
tion therapy.
2. Materials
2.1. Embryoid Body 1. Dispase (Gibco, Rockville, MD): Weigh and dissolve 1 mg/ml
Formation in Dulbecco’s Modified Eagle’s Medium: Nutrient Mixture
F-12 (DMEM/F-12) (Gibco, Rockville, MD). Dispase solu-
tion should be made fresh prior to each use and pre-warmed to
37°C in a water bath.
33 hPSC Embryoid Body Formation 537
2. Dulbecco’s Modified Eagle’s Medium: Nutrient Mixture F-12

(DMEM/F-12) (Gibco, Rockville, MD). Pre-warm to 37°C
in water bath.
3. EB Formation Medium. Iscove’s Modified Dulbecco’s Medium
(Gibco, Rockville, MD) supplemented with 15% Fetal Bovine
Serum Defined (HyClone, Logan, UT) (IMDM/15% FBS)
(see Note 1). Pre-warm to 37°C in a water bath. Once pre-
pared, complete medium can be used up to 14 days when
stored at 4°C.
4. AggreWell™ Medium (StemCell Technologies, Vancouver,
BC, Canada). Pre-warm to 37°C in a water bath.
5. Ultra Low Attachment (ULA) 25 or 75 cm2 flasks or 6-well
plate (Corning, Corning, NY).
6. Orbital shaker (Boekel Scientific, Feasterville, PA).
2.2. Embryoid Body 1. EB Formation Medium (IMDM/15% FBS) (see Subheading 2.1,
Maintenance item 3) AggreWell™ Medium (see Subheading 2.1, item 4).
2.3. Embryoid Body 1. Dulbecco’s Modified Eagle’s Medium: Nutrient Mixture F-12
Dissociation (DMEM/F-12) (see Subheading 2.1, item 2).
2. EB Formation Medium (IMDM/15% FBS) (see
Subheading 2.1, item 3) Accutase (Millipore, Temecula, CA).
Single use aliquots can be stored at −20°C. Thaw aliquots at
4°C and pre-warmed to room temperature before use. The
aliquots can be stored at 4°C up to 2 months after thawing.
3. Trypsin Inhibitor (Cascade Biologics, Portland, OR). Single
use aliquots can be stored at −20°C. Aliquots should be at
room temperature before use. The aliquots can stored at 4°C
up to 6 months after thawing.
4. Gelatin. Add 0.5 g gelatin to 500 ml endotoxin-free water to
make a 0.1% gelatin solution (see Note 2). Gelatin will not be
soluble at this point. Autoclave for 30 min. Gelatin will solubi-
lize and remain a liquid. Store at room temperature.
2.4. Mesenchymal 1. MesenCult® MSC Basal Medium (Human) (StemCell

Stem Cell Expansion Technologies, Vancouver, BC, Canada) supplemented with
10% Mesenchymal StemCell Stimulatory Supplements
(Human) (StemCell) and 10% Fetal Bovine Serum for Human
Mesenchymal Stem Cells (StemCell). Pre-warm to 37°C in a
water bath. Once prepared, complete medium can be used up
to 14 days when stored at 4°C.
2. 0.05% Trypsin–EDTA (Gibco, Rockville, MD). Pre-warm to
37°C in a water bath.
3. Dulbecco’s phosphate-buffered saline without calcium and
magnesium (D-PBS −/−) (Gibco, Rockville, MD). Store at
room temperature.
3. Methods
The success of EB formation from human pluripotent stem cells

depends on the following criteria.
● Healthy and robust human pluripotent stem cell colonies with
minimal differentiation.
● The presence of differentiated cells in human pluripotent stem
cell cultures will adversely affect the process EB formation.
However, it is normal for human pluripotent stem cell cultures
to exhibit minimal amounts of differentiation, which will not
affect the overall quality of resulting EBs. If the population of
differentiated cells in culture exceeds 20%, it is crucial to remove
the differentiation, subsequently passage cells, and then initiate
EB formation after confirming minimal differentiated.
● Large number of human pluripotent stem cell colonies.
● The optimal number of EBs is directly dependent on the num-
ber of stem cell colonies within the culture plate.
● Colonies in log phase of growth.
● EB formation should be initiated from human pluripotent
stem cell colonies that have been in culture for 3–5 days. This
ensures medium-sized colonies with actively dividing cells (see
Fig. 1). Overgrown colonies that have a tendency to fuse will
not form EBs.
3.1. Embryoid Body When first culturing EBs, start with two 6-well plates of human
Formation pluripotent stem cells, yielding six T25 culture flasks of EBs. With
experience this number may be adjusted and scaled down depend-
ing on specific yield and needs.
Fig. 1. (a) Human embryonic stem cell colonies cultured on mouse embryonic fibroblasts (MEF). Majority of the colonies are
of medium size and exhibit minimal differentiation (×2.5). (b) Human embryonic stem cell colonies cultured in feeder-
independent system (mTeSR™1). Majority of the colonies are of medium size and exhibit minimal differentiation (×2.5).
Fig. 2. (a) Human embryonic stem cell colonies cultured on MEF after 30 min of dispase treatment. Colonies display rolled
edges as an indication of initial detachment. Often times detachment of the MEF layer is observed. The detached MEF are
removed in subsequent washes, and do not interfere with the generation of EBs (×2.5). (b) Human embryonic stem cell
colonies cultured in feeder-independent system (mTeSR™1) after 5 min of dispase treatment. Colonies display rolled
edges as an indication of initial detachment (×2.5).
1. Observe stem cell colonies under the microscope to ensure

they are of optimal size, age, and quality.
2. In a biosafety cabinet, aspirate the media and add 1 ml of 37°C
dispase solution to each well.
3. Place cells in the incubator at 37°C with dispase for 20–30 min.
After 15–20 min, observe the plates under a microscope.
Colonies will begin to partially peel off the plate and show
rolled-up edges after treatment with dispase (see Note 3 and
Fig. 2).
4. With a 5 ml glass pipette, gently wash colonies off the surface
of the plate using the dispase solution in each well. Wash colo-
nies off one well at a time, and transfer the contents of one or
two wells into a 15 ml conical tube. For example, three 15 ml
conical tubes will be needed for one 6-well plate. Do not scrape
colonies off of the plate with the pipette. It is important to
treat the cells very gently in order for the colonies to remain
intact. Only intact colonies have the ability to form EBs (see
Note 4 and Fig. 3).
5. Add 2 ml EB Formation medium to each well to wash. Pipet
gently to rinse each well. Transfer the medium and any remain-
ing colonies to the 15 ml conical tube and gently mix contents
with the glass pipette (see Note 5).
6. Stand the 15 ml tube upright in the biological safety cabinet
for 1–3 min or until colonies have settled to the bottom of the
tube.
7. Gently remove the medium with a glass pipet to prevent the
EBs from being accidentally aspirated.
Fig. 3. Detached colonies as a result of dispase treatment have been washed and trans-
ferred into a ULA flask. Most colonies display a distinct marginal zone that has initiated the
formation of a spherical structure (EB). A few colonies that exhibit jagged or torn edges
lack the ability to form EBs (×5).
8. Wash colonies to remove traces of enzyme by adding 4 ml EB

Formation medium to colonies and pipeting slowly to gently
resuspend colonies (see Note 5).
9. Stand tube upright for 1–3 min to allow colonies to settle to
the bottom.
10. Gently remove the wash medium with a glass pipet to prevent
the EBs from being accidentally aspirated.
11. Add 8 ml EB Formation medium to each tube and very gently
pipet to resuspend the cell colonies/clumps.
12. Transfer the colonies and medium to an ULA flask or 6-well
plate. Label and number each flask or plate.
13. Place flask on an orbital shaker in a 37°C, humidified incubator
(see Note 6). The amount of media used in a T25 culture flask
is not sufficient to support more than approximately 50 EBs. If
more than 50 EBs are observed in a single flask, divide the EBs
into two or more flasks.
3.2. Embryoid Body 1. Observe cells daily under a microscope (see Note 7 and Fig. 4)
Maintenance and exchange EB Formation medium every Monday,
Wednesday, and Friday (see Note 8).
2. While changing medium, place the culture flasks in a vertical
position in the biological safety cabinet for 1–2 min to allow
the EBs to settle to the bottom of the flask.
3. Using a 10 ml glass pipet, gently transfer EBs and medium
from each flask into a separate sterile 15 ml conical tube. Label
the tube to match the number and date on the flask.
Fig. 4. (a) EBs in culture at day 4 show the dense center core which is characteristic of early EB development. These EBs
were generated from hES cells cultured on MEF (×5). (b) EBs in culture at day 2 show the dense center core which is
characteristic of early EB development. These EBs were generated from induced pluripotent stem cells cultured in the
feeder-independent system (mTeSR™1) using the Aggrewell™ technique (×5). (c) EBs in culture at day 10 display an
organized pattern with the presence of a cystic center. Please note the fused EBs that typically result from fused human
embryonic stem cell colonies. These EBs were generated from hES cells cultured on MEF (×10). (d) EB in culture at day 10
displaying an organized pattern with the presence of a cystic center. This EB was generated from induced pluripotent stem
cells cultured in feeder-independent system (mTeSR™1) using the Aggrewell™ technique (×10).
4. Re-cap the culture flasks and set aside.

5. Let the EBs settle by gravity to the bottom of the tube for
2–3 min.
6. Gently remove the spent medium with a glass pipet to prevent
the EBs from being accidentally aspirated.
7. Add 8 ml of fresh, pre-warmed EB Formation medium to each
tube.
8. Gently resuspend each EB pellet and transfer the contents of
each tube to its respective ULA culture flask or plate.
3.3. Embryoid Body EBs can be dissociated into single cells at defined time points,
Dissociation depending on desired germ layer and cell type (see Note 9). This
method may be used for EBs that have been generated using the
protocol outlined here or using the AggrewellTM plate method (see
Chapter 32). Using a 10 ml glass pipet, gently transfer the EBs and
medium from a ULA culture flask or plate to a 15 ml conical tube.
If the flask or plate contains less than 50 EBs, use one 15 ml conical
tube per T25 flask. If the flask or plate contains more than 50 EBs,
divide the EBs into two 15 ml conical tubes. An excessive number
of EBs in a tube will not permit optimal digestion.
1. Allow 2–3 min for the EBs settle to the bottom of the tube.
2. Gently aspirate the supernatant with a 5 or 10 ml pipet, careful
not to disturb the EB pellet. It is important to remove as much
of the serum-containing EB medium as possible to maximize
the effectiveness of the enzyme digestion.
3. Add 2 ml of Accutase to EB pellet and pipet up and down to

resuspend EBs.
4. Place the tube in a 37°C water bath for 15–20 min. After
10 min at 37°C, bring tube back to the biosafety cabinet and
pipet contents with a P1000 micropipette to mix and aid EB
digestion. Repeat pipeting or gently shake the tube every
5 min, until most EBs appear to be dissociated.
5. After incubation at 37°C, pipet the contents of the tube vigor-
ously to further break up cell clumps.
6. Once the dissociation process is successful, neutralize the enzy-
matic activity of the Accutase with an equal amount (2 ml) of
Trypsin Inhibitor and mix.
7. Spin the tube for 3 min at 200 × g.
8. Gently aspirate the supernatant medium, careful not to disrupt
the pellet.
9. Add 5 ml of basal medium and mix to remove any residual
Accutase and Trypsin Inhibitor (see Note 5).
10. Spin the tube for 3 min at 200 × g.
11. Aspirate the supernatant from the cell pellet.
12. Add 2 ml of EB medium, mix, and remove a small aliquot.
Count the total number of cells using a hemacytometer.
Calculations should be adjusted to account for the dilution fac-
tor if trypan blue is used.
13. Add an appropriate volume of EB medium and plate cells in
new culture vessels (see Note 10) at 500,000 to 1 × 106 cells
per well of a standard 6-well plate.
3.4. Mesenchymal 1. Replace EB medium after 1 day with Complete Mesencult®

Stem Cell Culture and Medium (Mesencult® MSC Basal medium with Mesenchymal
Expansion Stem Cell Stimulatory Supplements) (see Fig. 5).
2. Check mesenchymal cells under a microscope to ensure that
the cells are approximately 80–90% confluent (Fig. 6). This
should take approximately 10–14 days. Cells are ready to be
passaged at 80–90% confluency.
3. Aspirate Complete Mesencult Medium and wash each well of a
6-well plate with 1 ml of PBS.
4. Aspirate the PBS and add 1 ml 0.05% Trypsin–EDTA to each
well. Place the plate/s in the incubator for approximately 5 min
until cells have detached from the surface and formed a single
cell suspension.
5. Add 3 ml of Complete Mesencult® Medium to each well and
pool all wells into a 15 ml conical tube.
6. Take small sample of cell suspension for counting, and spin the
remaining cells at 200 × g for 5 min.
Fig. 5. Cells in culture at day 1 following dissociation of EBs (×5).
Fig. 6. A confluent monolayer of mesenchymal stem cells derived from dissociated EBs.
The EBs were generated from human pluripotent stem cells cultured in feeder-independent
system (mTeSR™1) using the Aggrewell™ technique (×10).
7. Resuspend the cells to a final cell concentration of 3–5 × 105 cells

per ml and plate 1 ml of cell suspension to each 75 cm2 flask.
Add an additional 14 ml of Complete Mesencult® Medium to
each flask for a total volume of 15 ml.
8. Place in incubator overnight and observe daily.
9. Cells can be passaged weekly at 80–90% confluency. The rec-
ommended dilution is 1/3 (e.g., one 75 cm2 tissue culture-
treated flask containing 80% confluent mesenchymal cells can
be passaged into three 75 cm2 tissue culture-treated flasks.

Dilution ratio can be adjusted to match the proliferation rate
of the cells).
10. Cells can be frozen and stored under liquid nitrogen using
standard cyropreservation technique.
4. Notes
1. This recipe is for EB formation using generic medium. EB for-

mation media contains formulations and factors that favorably
promote the generation/development of a specific germ layer
and subsequent differentiation into a specific lineage.
Components of the EB formation medium will vary based on
the germ layer and lineage of interest. In general, EB forma-
tion media need the following criteria.
(a) Serum or serum replacer
(b) A rich basal medium
(c) Low concentration of basic FGF
(d) Growth factors/cytokiness that promote lineage of inter-
est can also be added
2. Do not use glass bottles that have been washed with detergent.
Dedicated glassware must be acid stripped and rinsed thor-
oughly with distilled water prior to use. Do not allow water or
gelatin solution to sit unsterilized for any longer than 2 h
before autoclaving. Loosen the caps before autoclaving.
3. If peeling in the form of rolled edges is not obvious in
15–20 min, return plate to incubator for an additional 10 min.
Some small colonies may peel off the plate completely during
the extra incubation period.
4. When cultured in MEF, it is possible that some MEF will be
transferred to the tube along with the cell colonies. Most of
the MEF will be removed in subsequent washes, and the pres-
ence of MEF will not interfere with the generation of EBs.
5. Basal medium may also be used for the washes. For Aggrewell™
formulation, DMEM/F-12 is used as the basal medium.
6. Constant gentle motion is necessary for the EBs to stay in sus-
pension and to prevent their attachment and differentiation.
The orbital shaker should be set at a low speed to provide con-
stant gentle motion. If the shaker is set to a high speed, the
sheer stress will either prevent optimal EB formation or lead to
rupture and disintegration of developing EBs. Shaking at high
speeds may also cause the culture medium to enter the neck of
the flasks and increase the risk of contamination. It is not

recommended to shake ULA 6-well plates.
7. The colonies in suspension cultures they undergo a process of
organization to form EBs. Within several hours of being in
suspension cultures, the flat, “pancake-like” colonies will begin
to fold over in a concave format to form three-dimensional
spherical structures. Over time, these simple EBs organize fur-
ther to reveal a structured distribution of cells representing
germ layers. The organization process ranges for EBs with
dense cores to highly organized EBs with cystic fluid-filled
centers (Fig. 4).
8. When exchanging medium on Friday, add an additional 4 ml
of medium for a total of 12 ml.
9. A suggested guideline is as follows: 7–9 days for ectoderm,
9–11 days for mesoderm, and 21–25 days for endoderm.
10. It is necessary to coat the new culture vessels with gelatin when
using the Accutase digestion method. At least 1 day prior to
plating, coat wells with gelatin solution by placing 1 ml of gela-
tin per well of a six-well plate. Tilt plate in several directions so
that liquid covers the entire surface area. Place plates in incuba-
tor overnight. Plates can remain in the incubator for longer
than 1 day. However, do not use the plate if the gelatin has
dried up. Aspirate the gelatin from the wells immediately prior
to plating dissociated EBs. It is not recommended to plate the
digested cells on to a surface area larger than 1-well of a 6-well
plate because cells do not grow well when plated sparsely. After
one or two passages, these cells may be transferred to 10 cm2
dishes or 75 cm2 flasks. It is no longer necessary to gelatin coat
any additional flasks or plates for the remaining passages.
References
1. Ying QL, Stavridis M, Griffiths D, Li M, Smith comprising the three embryonic germ layers.
A (2003) Conversion of embryonic stem cells Mol Med 6(2):88–95
into neuroectodermal precursors in adherent 5. Kurosawa H (2007) Methods for inducing
monoculture. Nat Biotechnol 21(2):183–186 embryoid body formation: in vitro differentia-
2. Nakano T, Kodama H, Honjo T (1994) tion system of embryonic stem cells. J Biosci
Generation of lymphohematopoietic cells from Bioeng 103(5):389–398
embryonic stem cells in culture. Science 6. Keller M (1995) In vitro differentiation of
265(5175):1098–1101 embryonic stem cells. Curr Opin Cell Biol
3. Doetschman TC, Eistetter H, Katz M, Schmidt 7(6):862–869
W, Kemler R (1985) The in vitro development 7. Dvash T, Mayshar Y, Darr H, McElhaney M,
of blastocyst derived embryonic stem cell lines: Barker D, Yanuka O, Kotkow KJ, Rubin LL,
formation of visceral yolk sac, blood islands Benvenisty N, Eiges R (2004) Temporal gene
and myocardium. J Embryol Exp Morphol expression during differentiation of human
87:27–45 embryonic stem cells and embryoid bodies.
4. Itskovitz-Eldor J, Schuldiner M, Karsenti D, Hum Reprod 19(12):2875–2883
Eden A, Yanuka O, Amit M, Soreq H, 8. Trounson A (2006) The production and
Benvenisty N (2000) Differentiation of human directed differentiation of human embryonic
embryonic stem cells into embryoid bodies stem cells. Endocr Rev 27(2):208–219
9. Bratt-Leal A, Carpenedo L, McDevitt T (2009) LJ, Daigh CA, Conard KR, Piekarczyk MS,
Engineering the embryoid body microenvi- Llanas RA, Thomson JA (2006) Derivation of
ronment to direct embryonic stem cell differ- human embryonic stem cells in defined condi-
entiation. Biotechnol Prog 25(1):43–51 tions. Nat Biotechnol 24(2):185–187
10. Ludwig TE, Bergendahl V, Levenstein ME, Yu 12. Ungrin M, Joshi C, Nica A, Bauwens C,
J, Probasco MD, Thomson JA (2006) Feeder- Zandstra P (2008) Reproducible, ultra high-
independent culture of human embryonic throughput formation of multicellular organi-
stem cells. Nat Methods 3:637–646 zation from single cell suspension-derived
11. Ludwig TE, Levenstein ME, Jones JM, human embryonic stem cell aggregates. PLoS
Berggren WT, Mitchen ER, Frane JL, Crandall One 3(2):e1565
INDEX
A human prostate epithelial (HPE) ...................... 383–392

karyotype ..............................................40, 46, 56, 58, 69
Aggrewell™ plates ................................................... 532–545 mitochondrial DNA (mtDNA) typing ...... 28–31, 33–35
Antibiotics, mycoplasma multiplex PCR ....................................................... 32, 33
BM-cyclin ...................................................................21 NIH 3T3 ............................................397, 402, 403, 407
pleuromutilins ..............................................................17 OP9 ............................................................... 88, 97, 104
quinolones ...................................................................17 OP9-DL1 ...........................................86, 88, 97, 98, 104
tetracyclines .................................................................17 short tandem repeat (STR) DNA typing ......... 2, 27–38,
Aorta-gonads-mesenophros (AGM) ....................... 303, 305 184, 185, 189
Arginase assay.......................................................... 229, 235 spectral karyotyping (SKY ) .................. 42, 50, 51, 56, 58
mouse macrophages ...................................................235 stroma ......................................... 189, 304, 306, 307, 309
Attachment substrate tetraplex PCR .................................................. 31, 33–35
matrigel......................................................................498 transfection .........................134–136, 141–146, 148, 149
mesenchymal cells...............................318, 323–327, 332 Cloning
cell lines .........16, 133–149, 303, 304, 307, 308, 311, 312
B
hybridoma.......................................................... 133–149
BFU-E. See Burst forming unit-erythroid (BFU-E) Colony forming cell assays
Bone marrow CFU-Mk ................................................... 219, 220, 268
human...........................41, 103, 104, 152, 166, 206, 242, 7 day assay ..........................270, 275–276, 278, 280–282
247, 248, 253, 303, 318–319, 325, 329, 333, 350 hematopoieitic ................................................... 267–282
mouse ...................86, 103, 104, 151, 153, 157, 159, 160, LTC-IC analysis ................................................ 250–253
164, 166, 170, 172, 173, 230, 259, 303, 309, 336 Colony forming unit-granulocyte ............................ 254, 268
Burst forming unit-erythroid (BFU-E) ................. 245, 252, Colony forming unit-granulocyte, erythroid, monocyte,
254, 260, 268, 270, 276, 277, 280 megakaryocyte (CFU-GEMM) ............ 254, 268
Compact bone, mouse ............................................. 335–347
C Cord blood
Cancer stem cells (CSCs) ........................164, 165, 181–199, CFC assays ................................................................274
364, 365, 384–387 megakaryocyte and platelet differentiation ........ 205–222
CD34 isolation ........................................................ 108, 247 Cryopreservation .......... 21, 24, 184, 186, 188, 211, 517–518
Cell cocultivation.............................................................425 human PSC ................................513–514, 517–518, 520
oral keratinocyte and osteoblast-like cells .......... 423–428 CSCs. See Cancer stem cells (CSCs)
Cell cycle ....................................... 45, 61–81, 163, 168, 172,
D
12, 418, 424, 432, 435
Cell lines DAPI. See 4’-6-Diamidino-2-phenylindole (DAPI)
authentication ................................................ 28, 36, 391 2-Deoxyquanosine thymus treatment ................................87
biological resource centres (BRC)................................ 36 4’-6-Diamidino-2-phenylindole (DAPI) ................... 44, 50,
chromosome banding ..................................................41 53, 54, 70, 74, 79, 80, 106, 109, 112, 168,
cross-contamination ................................1, 2, 27–38, 40, 184, 190, 222, 397, 400–402, 404, 406,
41, 182, 188, 189 436, 443, 457, 462
cytogenetic analysis Dispase .............................366, 373, 396, 400, 412–415, 419,
FISH (see Fluorescent in situ hybridization (FISH)) 509, 514, 516, 517, 519, 520, 536, 539, 540
G-banding (see G-banding) hPSC passaging ................................................. 514–516
DOI 10.1007/978-1-62703-128-8, © Springer Science+Business Media, LLC 2013
547
BASIC CELL CULTURE PROTOCOLS
548 Index
E reaggregate thymus organ culture (RTOC) .......... 93, 95,

96, 100
ECFCs. See Endothelial colony forming progenitor cells retroviral gene transfer ..................................... 88, 95–97
(ECFCs) single cell isolation ................................................. 87, 91
ELISA. See Enzyme-linked immunosorbent assays thymic stromal cells ....................85, 93–94, 96, 100, 101
(ELISA) thymus lobe culture.............................87, 90–93, 99, 100
Embryoid bodies (EBs), human T-lymphocyte characterization ............................ 95–101
in Aggrewell plates ............................................ 523–533 FISH. See Fluorescent in situ hybridization (FISH)
differentiation to mesenchymal cells .................. 542, 543 Flow cytometry.................................. 89, 91, 94, 95, 99–101,
dissociation of ............................................................542 108, 111, 112, 121, 123, 126, 128, 129, 131, 135,
EB medium ........................................526, 541, 542, 544 140, 144, 153, 165, 176, 184, 189, 190, 194, 199,
mesenchymal stem cell differentiation ....... 537, 542–544 209–210, 212–218, 220, 221, 280, 296, 308, 310,
mouse........................................................... 335–347 346, 397, 400–401, 407, 413, 415–418, 472, 477
spin-EB method ........................................................524 Fluorescent in situ hybridization (FISH)
in suspension cultures ................................................545 array comparative genomic hybridization
Embryonic stem cells (ESCs) (aCGH) ................................... 67, 69–72, 78–79
hepatic differentiation of mouse ES cells........... 469–477 biliary brush sample ...............................................76–77
endoderm differentiation .....................................470 cancer...................................42, 62, 64, 67, 68, 76, 77, 79
hepatic progenitor cell ......................... 469, 475, 477 chromosome ........... 50, 51, 54–57, 61–70, 72–74, 78–81
murine fetal liver mesenchymal cells............ 469–477 colcemid ...............................................45, 46, 55, 70, 72
human DNA probes .................................................... 54, 61–81
cryopreservation of....................................... 517–518 genetic disorders ..........................................................69
defined culture media.................................. 470, 472, her2/neu ...............................................68, 71, 74–76, 81
475, 476, 508 interphase ..............................................................61–81
embryoid bodies............................524, 527, 530, 532 metaphase ..................... 45–47, 49, 50, 54–56, 58, 61–81
feeder-independent culture of multiplex-FISH, spectral karyotyping
(see Feeder-independent culture, ESC) (M-FISH/SKY ) ....................... 50, 52, 56, 58, 67
Endothelial colony forming progenitor cells probe preparation.........................................................73
(ECFCs) ......................... 350–352, 357, 358, 361 probes ....................................... 42, 44, 45, 50–54, 56, 57
culture medium .................................................. 357, 361 in situ hybridization ......................................... 42, 61–81
Enzyme-linked immunosorbent assays slide preparation ........... 45, 46, 49, 54, 71, 73–74, 79, 81
(ELISA)................................. 131, 135, 140, 144, urovysion ............................................................... 77, 79
230, 231, 236, 237, 253 validation of FISH probes .....................................79–80
cytokine detection...............................131, 236, 237, 253 whole chromosome painting probes
Epithelial cell...........................................363–380, 383–392, (WCP)................................................. 62, 63, 73
395, 396, 403, 407, 408, 411–420 FTOC. See Fetal thymus organ cultures (FTOC)
human hair follicle ............................................. 411–420
ESCs. See Embryonic stem cells (ESCs) G
G-banding ............................................41–43, 45–51, 54–58
F
slide preparation ....................................................45–49
FACS analysis. See Flow cytometry
H
Feeder-independent culture, ESC ...................................509
clump count method..................................................515 Hematopoietic progenitor cells ........108, 267–282, 347, 477
matrices .....................................................................510 colony forming cell assay ................................... 267–282
passaging ....................................512, 514–516, 518–520 Hematopoietic stem cells
single cell preparation ................................ 525, 526–527 human......................................... 104, 152, 268, 303, 316
thawing ...............................................509, 513–514, 520 mouse .................................................104, 151, 258, 268
undifferentiated hPSC assessment .................... 510–512 Hoechst 3342 staining ......................151, 152, 156, 159, 222
Fetal thymus organ cultures (FTOC) hTERT. See Human telomerase reverse transcriptase
fetal liver cells ............................................ 88, 92, 97–98 (hTERT)
fetal thymus isolation ....................................... 87, 90–91 Human pluripotent stem cells. See Embryonic stem
hanging drop reconstitution ............................ 87, 91–93 cells (ESCs)
high-oxygen submersion ...........................86, 87, 93, 100 Human telomerase reverse transcriptase
OP9-DL1 ...................................................86, 88, 97–98 (hTERT) ........................302, 385–388, 390–392
Index
549
Hybridoma mammoplasty dissociation (see Mammoplasty
cloning ............................................................... 133–149 dissociation)
development .............................................. 133–141, 146 mammospheres ............................................ 375, 376
myoepithelial progenitors ............................ 374, 375
I organoid suspensions ................................... 372, 373
Induced pluripotent stem cells. See Embryonic stem cells mouse
(ESCs); Pluripotent stem cells human basal cells .............................................................403
Inducible nitric oxide synthesis (iNOS) ................. 225–228, flow cytometry ............................................. 400, 401
233, 236 luminal cells .........................................................403
mammary colony forming assay ................... 402, 403
K mammary gland dissociation ...............................396
mammary repopulating unit (MRU) ........... 403, 404
Keratinocyte
myoepithelial cells........................................ 395, 396
colony forming assay ..................................................418
stem cells ..................................................... 403–405
culture ........................................................................ 418
transplantation of mammary cells ........................398
human hair follicle
Mammoplasty dissociation
enzymatic digestion .............................................412
collagenase and hyaluronidase digestion ............ 365–367
microdissection ....................................................411
generation of single cell suspension ................... 372, 373
progenitor cells ..........................................................205
Matrices, cell culture
stem cells ...................................................................411
for hPSC....................................................................516
Keratinocyte, human oral. See Oral keratinocyte
matrigel.............................................................. 376–377
L vitronectin.................................................. 509, 516, 517
MEF. See Mouse embryonic fibroblast (MEF)
Leukocyte recruitment Megakaryocytes
endothelial cell siRNA transfection ...........................288 CFC assay.................................................. 268, 269, 277
endothelial cell stimulation ................................ 287–288 cord blood differentiation of .............................. 205–222
flow chamber set-up ..................................................289 FACS analysis............................................................210
isolated leukocyte recruitment ...................................289 ploidy analysis .................................................... 213, 218
leukocyte isolation ............................................. 288, 289 Mesenchymal progenitor cells ......................... 326, 337, 340
parallel plate flow chamber ........................ 286, 287, 296 Mesenchymal stem cells, human
shear stress ................................................. 287, 296, 297 adipocyte differentiation .................................... 315, 316
whole blood recruitment ................................... 289, 291, CFU-F assay ..................................................... 319, 320
293, 297, 298 culture and expansion ........................................ 318–323
Long term culture initiating cell assay (LTC-IC) FBS-containing medium ................................... 318–323
human................................................................ 241–255 isolation from bone marrow .......................................325
mouse ................................................................ 257–265 multipotent mesenchymal stromal cells
(MSCs) .................................................. 315–333
M phenotype .......................................................... 321, 322
Macrophages, mouse pHPL medium, use of ............................... 350, 351, 357
alternative activation .......................................... 225–239 xeno-free medium (MesenCult™-XF) .............. 316, 317
characterization of ............................................. 151–161 Mesenchymal stem cells, mouse
culture of............................................................ 395–408 CFU-F assay .............................................................344
M1 ............................................................................. 226 culture and expansion ........................................ 344–345
M2 ..............................................................226, 227, 235 enrichment from bone marrow .......................... 339–340
Magnetic cell enrichment ........................................ 158–159 isolation from compact bone marrow ................ 342–343
Mammary tissue Methylcellulose-based medium
human hematopoietic CFCs...........................245, 254, 257, 260
bipotent progenitors ............................................363 hybridoma cloning .....................................................134
2D cultures .................................................. 366, 374 LTC-IC assays...........................................................252
3D matrigel cultures ........................... 364, 366–367, Monoclonal antibodies ............................118, 119, 131, 133,
376–377 135, 146, 156, 159, 206, 436
human breast epithelial progenitor cells ...... 364, 375 Mouse embryonic fibroblast (MEF) ....................... 199, 366,
human breast epithelial stem cells........................367 397, 407, 470, 472, 476, 508, 531, 536, 538, 539,
luminal progenitors ...................................... 374, 396 541, 544
550 Index
Mycoplasma S
contamination ...................................... 1–11, 15–25, 391
detection Serum-free medium...................................46, 147, 187, 206,
identification of mycoplasma species .................9–11 208, 384, 387, 392, 413, 418, 480, 488, 499, 530
PCR reaction ............................................. 3–7, 9–12 Side population cells
sample collection and preparation of DNA .............6 DyeCycle Violet (DCV) ................................... 163–178
eradication fumitremorgin C........................................ 167, 171, 173
antibiotics (see Antibiotics, mycoplasma) Hoechst 33342 ..................................151, 152, 155, 156,
antibiotic treatment .........................................19–23 159, 163–167, 170–173, 176–178
culture and testing post-treatment ...................23–24 Hoechst side population (SP) .................... 163, 164, 170
species ..................................2, 3, 5, 8–10, 12, 16, 25, 188 laser set-up......................................................... 156, 159
verapamil ...................................................152, 155, 160,
N 161, 167, 171, 173
Skeletal muscle myofiber
Neural colony forming cell (NCFC) ...................... 483, 488,
chick embryo extract .......................................... 460, 461
499, 501
extensor digitorum longus (EDL)
Neural stem cells
fixing and immunostaining .......................... 457–459
adult, mouse....................................................... 479–505
isolation and culture.............................................449
differentiation .................................................... 479–505
isotonic purecol collagen ............................................448
fetal, mouse ........................................................ 479–505
massetor muscle ......................................... 434, 449, 464
immunostaining .........................................................219
matrigel..............................................435, 437, 441, 442,
media for ........................................................... 484–488
449, 450, 455–460, 465
neurosphere culture....................................................484
satellite cells ....................................................... 431–465
primary cultures ................................................. 488–491
Stromal cell lines
sub culturing ..............................................................493
cloning ............................................................... 307, 308
Neurospheres ............ 480–482, 488–494, 497–499, 501–504
explant cultures ..........................................................305
functional characterization ................................ 307–308
O
generation of ...................................................... 303–304
Oral keratinocyte growth crisis ...................................................... 306, 307
co-cultivation ..................................................... 423–428 Stromal cells .................................... 85, 93, 94, 96, 100, 101,
culture medium ..........................................................425 104, 106–107, 111, 190, 242, 257, 302, 305, 309,
isolation and culture of ...................................... 426, 427 310, 312, 315, 316, 321, 396, 406
morphology ...............................................................427 Subventricular zone, mouse ............................. 479, 491–492
Osteoblast-like cells
cocultivation ..............................................................427 T
culture medium .................................................. 425, 426 T-lymphocytes
isolation and culture of ..............................................426 fetal thymus organ culture ................................... 85–101
morphology ....................................................... 427–428 flow cytometry ..............................89, 91, 94, 95, 99–101
human proT cells .......................................................104
P
t cell development ........................................... 85, 86, 88,
Platelet, generation from CB .......................... 205–222, 268, 91, 93, 95, 97–100
316, 336, 349–361 thymus engraftment...................................................104
Platelet lysate, human umbilical cord blood derived proT .................... 103–113
preparation from apheresis......................... 350, 352–353 T regulatory cells (Treg) human
preparation from buffy coat ............................... 350–353 antigen presenting cells.............................. 117–118, 122
Platelet rich plasma (PRP) .............................. 214, 215, 350 CD4+ T cells ......................................116, 121, 122, 131
Pluripotent stem cells human ..................507–520, 523–533, FoxP3................................................................. 115–131
535–545 lentivirus .....................................116, 118, 122, 123, 130
Prostate epithelial cells, human human mononuclear cells .................................. 116–117
cell line characterization .................................... 391–392 phenotype analysis ..................................... 116, 125–129
culture medium ..........................................................388 suppression assay ............................................... 127–129
generation of primary HPE ............................... 387–390 Tumor
hTERT immortalization ....................387, 388, 390–391 cell culture ....................................42, 183, 191, 199, 385
prostate tissue biopsy process .....................................387 Tumor-initiating cell ................ 182, 183, 189, 190, 192–195

Basic Cell Culture Protocols

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Basic Cell Culture Protocols

Uploaded by

Copyright:

Available Formats

METHODS IN MOLECULAR BIOLOGY™

For further volumes:

ISSN 1064-3745 ISSN 1940-6029 (electronic)

Library of Congress Control Number: 2012950590

© Springer Science+Business Media, LLC 2013

Printed on acid-free paper

Humana Press is a brand of Springer

Vancouver, BC, Canada Cheryl D. Helgason

1 Detection of Mycoplasma Contaminations . . . . . . . . . . . . . . . . . . . . . . . . . . . 1

14 Generation and Characterization of Murine Alternatively

31 Feeder-Independent Culture Systems for Human Pluripotent Stem Cells. . . . . 507

LAURIE AILLES • Department of Medical Biophysics, Ontario Cancer Institute,

AFSHIN RAOUF • Department of Immunology, Faculty of Medicine, University of

ZIPORA YABLONKA-REUVENI • Department of Biological Structure, University of

Detection of Mycoplasma Contaminations

Key words: Bacteria, Cell lines, Contamination, Mycoplasma, PCR

1.1. Mycoplasma Acute contaminations of cell lines are frequently observed in

regardless of their biological properties, with relatively low effort

To confirm the error-free preparation of the sample and PCR

1. Phosphate-buffered saline (PBS): 140 mM NaCl, 2.7 mM

(r = mixture of g and a; w = mixture of t and a)

The following subsections describe the sample collection, extrac-

and replaced by clean cultures or cured with specific antibiotics

reagents; (3) avoid reamplifications; (4) reserve pipettes, tips, and

When testing several samples, pre-master mixtures can be

5. After this initial cycle, perform 35 thermal cycles with the

3.3. Interpretation Figure 1 shows a representative ethidium bromide-stained gel with

– Weak mycoplasma-specific bands can occur after treatment of

3.4. Identification of Although the method described is sufficient to detect mycoplasma

M. arginini – M. hominis – M. hyorhinis – M. orale A. laidlawii – M. bovis – M. fermentans

M. arginini – M. hominis – M. hyorhinis M. orale A. laidlawii – M. bovis M. fermentans

M. hyorhinis M. arginini – M. hominis M. bovis A. laidlawii

M. hominis M. arginini digested

restriction fragment length polymorphism analysis. In case of a

1. Originally, the described method was designed as nested PCR

nested PCR. Mycoplasma-positive cell cultures were detected

Template Preparation Kit from Roche (Mannheim, Germany),

1. Uphoff CC, Drexler HG (2001) Prevention of 5. European Pharmacopoeia (2008) General

Eradication of Mycoplasma Contaminations

Key words: Antibiotic elimination, Cell lines, Mycoplasma

levels of protein, DNA, and RNA synthesis, and alterations of

other antibiotics that are often prophylactically used in cell culture

3. Ciprobay 100 (Bayer, Leverkusen, Germany) is a ready-to-use

3. The FBS concentration should be increased to 20% before,

● If possible, frequently detach slowly growing adherent cells in

3.2.1. Treatment 1. Prepare a cell suspension (detach adherent cells, break up

Mycoplasma-infected Cell Line 1

2 Freeze Aliquots of the

Treat Cell Line with

Quinolone Plasmocin MycoZap BM-Cyclin

Mycoplasma PCR Testing Mycoplasma-negative

Mycoplasma-positive Expand Cells, Freeze Master

Quinolone- / Plasmocin- / MycoZap-resistant BM-Cyclin-resistant

(1) BM-Cyclin Treatment

and loosely attached mycoplasmas. Seed the cells at the appro-

3. If using enrofloxacin or MRA, repeat step 2 another two times

3.2.3. Treatment 1. The activity of the antimicrobial peptide is influenced by the

3. After complete decontamination, expand the cells and freeze

1. Store the antibiotics at the recommended concentrations, tem-

will produce the same end result (cure, resistance, or culture

1. Barile MF, Rottem S (1993) Mycoplasmas in 4. Schmidt J, Erfle V (1984) Elimination of

STR DNA Typing of Human Cell Lines: Detection of Intra-

Most facilities culturing cells use multiple cell lines simultaneously.

may be intraspecies, when two genetically different cell lines of