Placenta and Trophoblast - Methods and Protocols (Vol 1) - M. Soares, J. Hunt (Humana, 2006) WW PDF

M E T H O D S I N M O L E C U L A R M E D I C I N E TM
Placenta
and Trophoblast
Methods and Protocols
Volume I
Edited by
Michael J. Soares
Joan S. Hunt
Placenta and Trophoblast
M E T H O D S I N M O L E C U L A R M E D I C I N E™
John M. Walker, SERIES EDITOR

125. Myeloid Leukemia: Methods and Protocols, 104. Stroke Genomics: Methods and Reviews,
edited by Harry Iland, Mark Hertzberg, edited by Simon J. Read and David Virley,
and Paula Marlton, 2006 2004
124. Magnetic Resonance Imaging: Methods and 103. Pancreatic Cancer: Methods and Protocols,
Biological Applications, edited by Pottumarthi edited by Gloria H. Su, 2004
V. Prasad, 2006 102. Autoimmunity: Methods and Protocols,
123. Marijuana and Cannabinoid Research: edited by Andras Perl, 2004
Methods and Protocols, edited by Emmanuel S. 101. Cartilage and Osteoarthritis: Volume 2,
Onaivi, 2006 Structure and In Vivo Analysis, edited by
122. Placenta Research Methods and Protocols: Frédéric De Ceuninck, Massimo Sabatini, and
Volume 2, edited by Michael J. Soares and Philippe Pastoureau, 2004
Joan S. Hunt, 2006 100. Cartilage and Osteoarthritis: Volume 1,
121. Placenta Research Methods and Protocols: Cellular and Molecular Tools, edited by
Volume 1, edited by Michael J. Soares and Massimo Sabatini, Philippe Pastoureau, and
Joan S. Hunt, 2006 Frédéric De Ceuninck, 2004
120. Breast Cancer Research Protocols, edited by 99. Pain Research: Methods and Protocols,
Susan A. Brooks and Adrian Harris, 2006 edited by David Z. Luo, 2004
119. Human Papillomaviruses: Methods and 98. Tumor Necrosis Factor: Methods and
Protocols, edited by Clare Davy and John Protocols, edited by Angelo Corti and Pietro
Doorbar, 2005 Ghezzi, 2004
118. Antifungal Agents: Methods and Protocols, 97. Molecular Diagnosis of Cancer: Methods and
edited by Erika J. Ernst and P. David Rogers, Protocols, Second Edition, edited by Joseph E.
2005 Roulston and John M. S. Bartlett, 2004
117. Fibrosis Research: Methods and Protocols, 96. Hepatitis B and D Protocols: Volume 2,
edited by John Varga, David A. Brenner, Immunology, Model Systems, and Clinical
and Sem H. Phan, 2005 Studies, edited by Robert K. Hamatake and
116. Inteferon Methods and Protocols, edited by Johnson Y. N. Lau, 2004
Daniel J. J. Carr, 2005 95. Hepatitis B and D Protocols: Volume 1,
115. Lymphoma: Methods and Protocols, edited by Detection, Genotypes, and Characterization,
Timothy Illidge and Peter W. M. Johnson, 2005 edited by Robert K. Hamatake and Johnson Y.
114. Microarrays in Clinical Diagnostics, edited by N. Lau, 2004
Thomas O. Joos and Paolo Fortina, 2005 94. Molecular Diagnosis of Infectious Diseases,
113. Multiple Myeloma: Methods and Protocols, Second Edition, edited by Jochen Decker and
edited by Ross D. Brown and P. Joy Ho, 2005 Udo Reischl, 2004
112. Molecular Cardiology: Methods and Protocols, 93. Anticoagulants, Antiplatelets, and
edited by Zhongjie Sun, 2005 Thrombolytics, edited by Shaker A. Mousa,
111. Chemosensitivity: Volume 2, In Vivo Mod- 2004
els, Imaging, and Molecular Regulators, edited 92. Molecular Diagnosis of Genetic Diseases,
by Rosalyn D. Blumethal, 2005 Second Edition, edited by Rob Elles and
110. Chemosensitivity: Volume 1, In Vitro Assays, Roger Mountford, 2004
edited by Rosalyn D. Blumethal, 2005 91. Pediatric Hematology: Methods and
109. Adoptive Immunotherapy: Methods and Protocols, edited by Nicholas J. Goulden
Protocols, edited by Burkhard Ludewig and and Colin G. Steward, 2003
Matthias W. Hoffman, 2005 90. Suicide Gene Therapy: Methods and Reviews,
108. Hypertension: Methods and Protocols, edited by Caroline J. Springer, 2004
edited by Jérôme P. Fennell and Andrew 89. The Blood–Brain Barrier: Biology and
H. Baker, 2005 Research Protocols, edited by Sukriti Nag, 2003
107. Human Cell Culture Protocols, Second 88. Cancer Cell Culture: Methods and Protocols,
Edition, edited by Joanna Picot, 2005 edited by Simon P. Langdon, 2003
106. Antisense Therapeutics, Second Edition, 87. Vaccine Protocols, Second Edition, edited by
edited by M. Ian Phillips, 2005 Andrew Robinson, Michael J. Hudson, and
105. Developmental Hematopoiesis: Methods Martin P. Cranage, 2003
and Protocols, edited by Margaret H. Baron, 86. Renal Disease: Techniques and Protocols,
2005 edited by Michael S. Goligorsky, 2003
M E T H O D S I N M O L E C U L A R M E D I C I N E™
Placenta and
Trophoblast
Volume 1
Edited by
Michael J. Soares
Institute of Maternal–Fetal Biology
Division of Cancer and Developmental Biology
Department of Pathology and Laboratory Medicine
University of Kansas Medical Center, Kansas City, KS
and
Joan S. Hunt
University Distinguished Professor, Vice Chancellor for Research
Department of Anatomy and Cell Biology
© 2006 Humana Press Inc.
999 Riverview Drive, Suite 208
Totowa, New Jersey 07512
www.humanapress.com
All rights reserved. No part of this book may be reproduced, stored in a retrieval system, or transmitted in
any form or by any means, electronic, mechanical, photocopying, microfilming, recording, or otherwise
without written permission from the Publisher. Methods in Molecular MedicineTM is a trademark of The
Humana Press Inc.
All papers, comments, opinions, conclusions, or recommendations are those of the author(s), and do not
necessarily reflect the views of the publisher.
This publication is printed on acid-free paper. ∞

ANSI Z39.48-1984 (American Standards Institute)
Permanence of Paper for Printed Library Materials.
Cover illustration: Background: Figure 5 from Chapter 11 (Volume 1), “Mouse Trophoblast Stem Cells”
by J. Quinn et al. Foreground: Figure 4 from Chapter 26 (Volume 1), “Vascular Corrosion Casting of the
Uteroplacental and Fetoplacental Vasculature in Mice” by K. J. Whiteley et al.
Cover design by Patricia F. Cleary.

For additional copies, pricing for bulk purchases, and/or information about other Humana titles, contact
Humana at the above address or at any of the following numbers: Tel.: 973-256-1699; Fax: 973-256-8341;
E-mail: orders@humanapr.com; or visit our Website: www.humanapress.com
Photocopy Authorization Policy:
Authorization to photocopy items for internal or personal use, or the internal or personal use of specific
clients, is granted by Humana Press Inc., provided that the base fee of US $30.00 per copy is paid directly
to the Copyright Clearance Center at 222 Rosewood Drive, Danvers, MA 01923. For those organizations
that have been granted a photocopy license from the CCC, a separate system of payment has been arranged
and is acceptable to Humana Press Inc. The fee code for users of the Transactional Reporting Service is:
[1-58829-404-8/06 $30.00 ].
eISBN 1-59259-983-4
Printed in the United States of America. 10 9 8 7 6 5 4 3 2 1
Library of Congress Cataloging in Publication Data
Placenta and trophoblast: methods and protocols / edited by Michael J. Soares and Joan S. Hunt.
p. ; cm. — (Methods in molecular medicine ; 121-122)
Includes bibliographical references and index.
ISBN 1-58829-404-8 (alk. paper) — ISBN 1-58829-608-3 (alk. paper)
1. Placenta. 2. Molecular biology.
[DNLM: 1. Placenta. 2. Molecular Biology. WQ 212 P6974 2005] I.
Soares, Michael J. II. Hunt, Joan S. III. Series.
QP281.P5435 2005
612.6’3—dc22 2005006428
Preface
The aim of the two-volume set of Placenta and Trophoblast: Methods
and Protocols is to offer contemporary approaches for studying the biology of
the placenta. The chapters contained herein also address critical features of the
female organ within which the embryo is housed, the uterus, and some aspects
of the embryo–fetus itself, particularly those of common experimental animal
models. In keeping with the organization used effectively in other volumes in
this series, each chapter has a brief introduction followed by a list of required
items, protocols, and notes designed to help the reader perform the experi-
ments without difficulty. In both volumes, sources of supplies are given and
illustrations highlight particular techniques as well as expected outcomes. A
key aspect of these volumes is that the contributors are at the forefronts of their
disciplines, thus ensuring the accuracy and usefulness of the chapters.
Placenta research has progressed rapidly over the past several decades
by taking advantage of the technical advances made in other fields. For example,
the reader will note that many techniques, such as reverse transcriptase poly-
merase chain reaction, northern and western blotting, microarray analyses and in
situ hybridization experiments, are routinely used for dissecting a wide range of
experimental questions. Protein analysis and functional experiments on tissues
and cells that comprise the maternal–fetal interface benefit from studies in endo-
crinology, immunology, and developmental biology. These volumes also
present new ideas on investigating gene imprinting and gene transfer via viral
vectors.
In developing these volumes we encountered the problem of how to
organize the contents so as to be reader-friendly. Our decision was to subdi-
vide in large part by the chronology of pregnancy so that in vivo aspects of
implantation come first, followed by in vitro systems of investigation, then
protocols for phenotypic analyses of placentas of several species. Special tech-
niques mentioned above conclude Volume I. Volume II continues with proto-
cols for studying trophoblast invasion, followed by dissection of how invading
trophoblast cells might be received by uterine immune cells. Returning to the pla-
centa itself, methods for researching trophoblast endocrine and transport functions
are followed by a final series of chapters on how placentas adapt to disease. In this
latter group, two chapters offer help to investigators interested in animal models of
human placental disorders and two address working with the oxygen switches
that program gene expression in early pregnancy, a concept entirely unexplored
v
vi Preface
less than a decade ago. The reader is referred to the Introductions in each of the
two volumes for a more detailed description of the contents.
This project would not have been possible without the contributions of
many individuals. We wish to express our gratitude to the contributing authors
for their time, effort, creativity, and their willingness to share their knowledge
and expertise. Our deep appreciation and gratefulness also goes to Stacy
McClure for her dedicated efforts in maintaining the organization of the manu-
scripts and the correspondence between the editors and the authors. During
this process the publisher has provided us with helpful guidance and instruc-
tion essential for the completion of this effort.
Finally, we hope that these volumes are useful and provide a valuable
resource for both trainees and established scientists striving to advance our
understanding of this unique, entirely essential organ of reproduction.
Michael J. Soares
Joan S. Hunt
Contents
Preface .............................................................................................................. v
Contributors .....................................................................................................xi
Companion Table of Contents for Volume II .................................................. xv
Companion CD-ROM .................................................................................... xix
PART I. INTRODUCTION
1 Placenta and Trophoblast: Methods and Protocols: Overview I
Michael J. Soares and Joan S. Hunt ...................................................... 3
PART II. METHODS FOR STUDYING EMBRYO IMPLANTATION

AND UTERINE BIOLOGY
2 Methodologies to Study Implantation in Mice
Kaushik Deb, Jeff Reese, and Bibhash C. Paria .................................... 9
3 Blastocyst Culture
D. Randall Armant .............................................................................. 35
4 Isolation of Hormone Responsive Uterine Stromal Cells: An In Vitro
Model for Stromal Cell Proliferation and Differentiation
Virginia Rider ...................................................................................... 57
5 Rat Decidual Cell Cultures
Yan Gu and Geula Gibori ................................................................... 69
6 The Immortalization of Human Endometrial Cells
Graciela Krikun, Gil Mor, and Charles Lockwood ............................. 79
7 Sheep Uterine Gland Knockout (UGKO) Model
Thomas E. Spencer and C. Allison Gray ............................................. 85
8 A Baboon Model for Inducing Endometriosis
Asgerally T. Fazleabas ......................................................................... 95
9 A Baboon Model for Simulating Pregnancy
Asgerally T. Fazleabas ....................................................................... 101
10 The Common Marmoset Monkey as a Model for Implantation
and Early Pregnancy Research
Almuth Einspanier, Kai Lieder, Ralf Einspanier,
and Bettina Husen ........................................................................ 111
vii
viii Contents
PART III. IN VITRO TROPHOBLAST AND PLACENTAL MODEL SYSTEMS

11 Mouse Trophoblast Stem Cells
Jennifer Quinn, Tilo Kunath, and Janet Rossant ............................... 125
12 Connexins and Trophoblast Cell Lineage Development
Mark Kibschull and Elke Winterhager .............................................. 149
13 Rcho-1 Trophoblast Stem Cells: A Model System for Studying
Trophoblast Cell Differentiation
Namita Sahgal, Lindsey N. Canham, Brent Canham,
and Michael J. Soares ................................................................... 159
14 Bovine Trophoblast Cell Culture Systems: A Technique to Culture
Bovine Trophoblast Cells Without Feeder Cells
Kazuyoshi Hashizume, Arata Shimada, Haruo Nakano,
and Toru Takahashi ...................................................................... 179
15 In Vitro Induction of Trophoblast from Human Embryonic
Stem Cells
Ren-He Xu ......................................................................................... 189
16 Isolation and Culture of Term Human Trophoblast Cells
Margaret G. Petroff, Teresa A. Phillips, Hakhyun Ka,
Judith L. Pace, and Joan S. Hunt .................................................. 203
17 Production of Human Trophoblast Cell Lines
Guy St J. Whitley .............................................................................. 219
18 Culture and Transfection of Human Choriocarcinoma Cells
Michael W. Wolfe ............................................................................. 229
19 In Vitro Methods for Studying Vascularization of the Murine
Allantois and Allantoic Union with the Chorion
Karen M. Downs .............................................................................. 241
PART IV. PHENOTYPIC ANALYSIS OF THE PLACENTA

20 Phenotypic Analysis of the Mouse Placenta
David R. C. Natale, Maja Starovic, and James C. Cross ................... 275
21 Phenotypic Analysis of the Rat Placenta
Rupasri Ain, Toshihiro Konno, Lindsey N. Canham,
and Michael J. Soares ................................................................... 295
22 Analysis of the Structure of the Ruminant Placenta:
Methods of Fixation, Embedding, and Antibody Localization
at Light and Electron Microscope Levels
F. B. P. Wooding ............................................................................... 315
Contents ix
23 Characterization of the Bovine Placenta by Cytoskeleton,

Integrin Receptors, and Extracellular Matrix
Christiane D. Pfarrer ......................................................................... 323
24 Molecular Markers for Human Placental Investigation
Berthold Huppertz ............................................................................ 337
25 Correlative Microscopy of Ultrathin Cryosections
in Placental Research
Toshihiro Takizawa and John M. Robinson ...................................... 351
26 Vascular Corrosion Casting of the Uteroplacental
and Fetoplacental Vasculature in Mice
Kathie J. Whiteley, Christiane D. Pfarrer, and S. Lee Adamson ....... 371
27 Analysis of Fetal and Maternal Microvasculature in Ruminant
Placentomes by Corrosion Casting
Rudolf Leiser and Christiane D. Pfarrer ........................................... 393
PART V. MOLECULAR ANALYSIS AND GENE TRANSFER TECHNIQUES

28 Microarray Analysis of Trophoblast Cells
Vikram Budhraja and Yoel Sadovsky ................................................ 411
29 Gene Expression Microarray Data Analysis of Decidual
and Placental Cell Differentiation
Sue Kong, Bruce J. Aronow, and Stuart Handwerger ....................... 425
30 Assays to Determine Allelic Usage of Gene Expression
in the Placenta
Paul B. Vrana .................................................................................... 439
31 Adenoviral-Mediated Gene Delivery to Trophoblast Cells
Bing Jiang and Carole R. Mendelson ................................................ 451
Index ............................................................................................................ 463
Contributors
S. LEE ADAMSON • Samuel Lunenfeld Research Institute, Mount Sinai Hospital,
University of Toronto, Toronto, Ontario, Canada
RUPASRI AIN • Institute of Maternal–Fetal Biology, Division of Cancer &
Developmental Biology, Department of Pathology & Laboratory Medicine,
D. RANDALL ARMANT • Departments of Obstetrics & Gynecology, Anatomy
& Cell Biology, Wayne State University School of Medicine, Detroit, MI
BRUCE J. ARONOW • Division of Biomedical Informatics and Developmental
Biology, Cincinnati Children’s Hospital Medical Center, Cincinnati, OH
VIKRAM BUDHRAJA • Departments of Obstetrics & Gynecology, Cell Biology
& Physiology, Washington University School of Medicine, St. Louis, MO
BRENT CANHAM • Institute of Maternal–Fetal Biology, Division of Cancer &
Developmental Biology, Department of Pathology & Laboratory Medicine,
LINDSEY N. CANHAM • Institute of Maternal–Fetal Biology, Division of Cancer
& Developmental Biology, Department of Pathology & Laboratory
Medicine, University of Kansas Medical Center, Kansas City, KS
JAMES C. CROSS • Genes & Development Research Group, Department of
Biochemistry & Molecular Biology, University of Calgary, Calgary,
Alberta, Canada
KAUSHIK DEB • Division of Reproductive and Developmental Biology, Department
of Pediatrics, Vanderbilt University Medical Center, Nashville, TN
KAREN M. DOWNS • Department of Anatomy, University of Wisconsin-
Madison Medical School, Madison, WI
ALMUTH EINSPANIER • Institute of Physiological Chemistry,
Facility of Veterinary Medicine, University of Leipzig, Leipzig, Germany
and Department of Reproductive Biology, German Primate Centre,
Göttingen, Germany
RALF EINSPANIER • Department of Veterinary Biochemistry, Free University
of Berlin, Berlin, Germany
ASGERALLY T. FAZLEABAS • Department of Obstetrics & Gynecology, Center
for Women’s Health & Reproduction, University of Illinois, Chicago, IL
GEULA GIBORI • Department of Physiology and Biophysics, University of
Illinois, Chicago, IL
C. ALLISON GRAY • Department of Animal Science, Center for Animal Biotech-
nology and Genomics, Texas A&M University, College Station, TX
xi
xii Contributors
YAN GU • Center for Food Safety and Applied Nutrition, Food and Drug
Administration, College Park, MD
STUART HANDWERGER • Division of Endocrinology, Cincinnati Children’s
Hospital Medical Center, Cincinnati, OH
KAZUYOSHI HASHIZUME • Department of Veterinary Medicine, Laboratory of
Veterinary Physiology, Iwate University, Morioka City, Iwate, Japan
JOAN S. HUNT • Department of Anatomy and Cell Biology, University of
Kansas Medical Center, Kansas City, KS
BERTHOLD HUPPERTZ • Department of Anatomy II, University Hospital RWTH
Aachen, Aachen, Germany
BETTINA HUSEN • Department of Reproductive Biology, German Primate
Centre, Göttingen, Germany
BING JIANG • Department of Obstetrics & Gynecology, University of Texas
Southwestern Medical Center, Dallas, TX
HAKHYUN KA • Department of Biological Resources & Technology, Yonsei
University, Wonju, Kangwon-Do, South Korea
MARK KIBSCHULL • Institute of Anatomy, University Hospital Essen, University
of Essen-Duisburg, Essen, Germany
SUE KONG • Division of Biomedical Informatics, Cincinnati Children’s
Hospital Medical Center, Cincinnati, OH
TOSHIHIRO KONNO • Institute of Maternal–Fetal Biology, Division of Cancer
GRACIELA KRIKUN • Department of Obstetrics and Gynecology,
Yale University, New Haven, CT
TILO KUNATH • Samuel Lunenfeld Research Institute, Mount Sinai Hospital,
RUDOLF LEISER • Department of Veterinary Anatomy, Histology
and Embryology, Justus-Liebig-University Giessen, Giessen, Germany
KAI LIEDER • Institute of Physiological Chemistry, Facility of Veterinary
Medicine, University of Leipzig, Leipzig, Germany and Department of
Reproductive Biology, German Primate Centre, Göttingen, Germany
CHARLES LOCKWOOD • Department of Obstetrics and Gynecology, Yale
University, New Haven, CT
CAROLE R. MENDELSON • Departments of Biochemistry and Obstetrics and
Gynecology, University of Texas Southwestern Medical Center, Dallas,
TX
GIL MOR • Department of Obstetrics and Gynecology, Reproductive Immunology
Unit, Yale University School of Medicine, New Haven, CT
Contributors xiii
HARUO NAKANO • Reproductive Biology and Technology Laboratory,

Developmental Biology Department, National Institute of Agrobiological
Sciences, Tsukuba, Japan
DAVID R. C. NATALE • Genes & Development Research Group, Department
of Biochemistry & Molecular Biology, University of Calgary, Calgary,
Alberta, Canada
JUDITH L. PACE • Department of Molecular and Integrative Physiology,
BIBHASH C. PARIA • Division of Reproductive and Developmental Biology,
Department of Pediatrics, Vanderbilt University Medical Center,
Nashville, TN
MARGARET G. PETROFF • Department of Anatomy and Cell Biology, University
of Kansas Medical Center, Kansas City, KS
CHRISTIANE D. PFARRER • Department of Obstetrics and Gynecology, Justus-
Liebig-University Giessen, Giessen, Germany
TERESA A. PHILLIPS • Department of Internal Medicine, University of Kansas
Medical Center, Kansas City, KS
JENNIFER QUINN • Samuel Lunenfeld Research Institute, Mount Sinai Hospital,
JEFF REESE • Division of Reproductive and Developmental Biology, Department
of Pediatrics, Vanderbilt University Medical Center, Nashville, TN
VIRGINIA RIDER • Department of Biology, Pittsburg State University,
Pittsburg, KS
JOHN M. ROBINSON • Department of Physiology and Cell Biology, Ohio State
University, Columbus, OH
JANET ROSSANT • Samuel Lunenfeld Research Institute, Mount Sinai Hospital,
YOEL SADOVSKY • Departments of Obstetrics & Gynecology, Cell Biology and
Physiology, Washington University School of Medicine, St. Louis, MO
NAMITA SAHGAL • Institute of Maternal–Fetal Biology, University of Kansas
Medical Center, Kansas City, KS
ARATA SHIMADA • Reproductive Biology and Technology Laboratory,
MICHAEL J. SOARES • Institute of Maternal–Fetal Biology, Division of Cancer
THOMAS E. SPENCER • Department of Animal Science, Center for Animal
Biotechnology and Genomics, Texas A&M University, College Station, TX
xiv Contributors
MAJA STAROVIC • Genes and Development Research Group, Department of

Biochemistry & Molecular Biology, University of Calgary, Calgary,
Alberta, Canada
TORU TAKAHASHI • Reproductive Biology and Technology Laboratory,
TOSHIHIRO TAKIZAWA • Department of Anatomy, Nippon Medical School,
Bunkyo-ku, Tokyo, Japan
PAUL B. VRANA • Department of Biological Chemistry, College of Medicine,
University of California Irvine, Irvine, CA
KATHIE J. WHITELEY • Samuel Lunenfeld Research Institute, Mount Sinai
Hospital, Toronto, Ontario, Canada
GUY ST J. WHITLEY • Department of Basic Medical Sciences, Biochemistry
and Immunology, St George’s Hospital Medical School, University of
London, London, United Kingdom
ELKE WINTERHAGER • Institute of Anatomy, University Hospital Essen,
University of Essen-Duisburg, Essen, Germany
MICHAEL W. WOLFE • Department of Molecular and Integrative Physiology,
F. B. P. WOODING • Department of Physiology, Cambridge, Cambridge,
United Kingdom
REN-HE XU • WiCell Research Institute, Madison, WI
Contents of Volume 2
Preface
Contributors
Companion Table of Contents for Volume I
PART I. INTRODUCTION
1 Placenta and Trophoblast: Methods and Protocols: Overview II
Michael J. Soares and Joan S. Hunt
PART II. ANALYSIS OF TROPHOBLAST INVASION

2 In Vivo Analysis of Trophoblast Cell Invasion in the Human
Robert Pijnenborg, Elizabeth Ball, Judith N. Bulmer,
Myriam Hanssens, Stephen C. Robson,
and Lisbeth Vercruysse
3 In Vitro Analysis of Trophoblast Invasion
John D. Aplin
4 An In Vitro Model of Trophoblast Invasion of Spiral Arteries
Judith E. Cartwright and Mark Wareing
PART III. ANALYSIS OF UTEROPLACENTAL IMMUNE

CELLS AND THEIR FUNCTIONS
5 In Vivo Models for Studying Homing and Function of the Murine
Uterine Natural Killer Cells
B. Anne Croy and Xuemei Xie
6 Immune and Trophoblast Cells at the Rhesus Monkey
Maternal–Fetal Interface
Thaddeus G. Golos, Gennadiy I. Bondarenko, Edith E. Breburda,
Svetlana V. Dambaeva, Maureen Durning, and Igor I. Slukvin
7 Methods for Isolation of Cells from the Human
Fetal–Maternal Interface
Anita Trundley, Lucy Gardner, Jacquie Northfield,
Chiwen Chang, and Ashley Moffett
xvi Contents for Volume 2
8 In Vitro Models for Studying Human Uterine

and Placental Macrophages
Ramsey H. McIntire, Margaret G. Petroff, Teresa A. Phillips,
and Joan S. Hunt
9 Macrophage–Trophoblast Interactions
Gil Mor, Shawn L. Straszewski-Chavez, and Vikki M. Abrahams
10 Methods for Evaluating Histocompatibility Antigen Gene
Expression in the Baboon
Daudi K. Langat, Asgerally T. Fazleabas, and Joan S. Hunt
11 Analysis of the Soluble Isoforms of HLA-G: mRNAs and Proteins
Judith L. Pace, Pedro J. Morales, Teresa A. Phillips,
and Joan S. Hunt
PART IV. ANALYSIS OF PLACENTA FUNCTION:

TRANSPORT AND ENDOCRINOLOGY
12 In Vivo Techniques for Studying Placental Nutrient Uptake,
Metabolism, and Transport
Timothy R. H. Regnault and William W. Hay, Jr.
13 In Vitro Models for Studying Trophoblast Transcellular Transport
Claudia J. Bode, Hong Jin, Erik Rytting, Peter S. Silverstein,
Amber M. Young, and Kenneth L. Audus
14 In Vitro Methods for Studying Human Placental
Amino Acid Transport: Placental Plasma Membrane Vesicles
Jocelyn D. Glazier and Colin P. Sibley
15 In Vitro Methods for Studying Human Placental
Amino Acid Transport: Placental Villous Fragments
Susan L. Greenwood and Colin P. Sibley
16 Methods for Investigating Placental Fatty Acid Transport
Yan Xu, Thomas J. Cook, and Gregory T. Knipp
17 Heterologous Expression Systems
for Studying Placental Transporters
Vadivel Ganapathy, You-Jun Fei, and Puttur D. Prasad
18 Analysis of Trophoblast Giant Cell Steroidogenesis
in Primary Cultures
Noa Sher and Joseph Orly
19 Establishment of an ELISA for the Detection of Native Bovine
Pregnancy-Associated Glycoproteins Secreted by Trophoblast
Binucleate Cells
Jonathan A. Green and R. Michael Roberts
Contents for Volume 2 xvii
20 Alkaline Phosphatase Fusion Proteins as Tags for Identifying

Targets for Placental Ligands
Heiner Müller and Michael J. Soares
21 Bacterial Expression of Prolactin Family Proteins
Arieh Gertler
22 Analysis of Placental Regulation of Hematopoiesis
Beiyan Zhou and Daniel I. H. Linzer
23 Methods for Studying Interferon Tau Stimulated Genes
Fuller W. Bazer and Thomas E. Spencer
PART V. ANALYSIS OF PLACENTA ADAPTATION TO DISEASE

24 Reduced Uterine Perfusion Pressure (RUPP) Model
for Studying Cardiovascular–Renal Dysfunction
in Response to Placental Ischemia
Joey P. Granger, B. Babbette D. LaMarca, Kathy Cockrell,
Mona Sedeek, Charles Balzi, Derrick Chandler,
and William Bennett
25 In Vivo Rat Model of Preeclampsia
S. Ananth Karumanchi and Isaac E. Stillman
26 A Novel Mouse Model for Preeclampsia by Transferring
Activated Th1 Cells into Normal Pregnant Mice
Ana Claudia Zenclussen
27 Working with Oxygen and Oxidative Stress In Vitro
Graham J. Burton, D. Stephen Charnock-Jones,
and Eric Jauniaux
28 Hypobaric Hypoxia as a Tool to Study Pregnancy-Dependent
Responses at the Maternal–Fetal Interface
Jennifer K. Ho-Chen, Rupasri Ain, Adam R. Alt, John G. Wood,
Norberto C. Gonzalez, and Michael J. Soares
29 Infection with Listeria monocytogenes as a Probe
for Placental Immunological Function
Ellen M. Barber, Indira Guleria, and Jeffrey W. Pollard
Index
Companion CD-ROM
This book is accompanied by a CD-ROM that contains all the color
illustrations.
xix
Overview 3
1
Overview I
Michael J. Soares and Joan S. Hunt
1. Introduction
The placenta is a specialized pregnancy-specific structure that develops con-
currently with development of the embryo and fetus. From an evolutionary
perspective, the placenta was the essential factor in permitting viviparity, a
reproductive strategy in which fetal development proceeds within the female
reproductive tract. Viviparous species are able to provide greater protection
from environmental risks and can more precisely control the development of
their progeny while they reside in utero. The placenta is comprised of numer-
ous cell types. Among the cell types are specialized epithelioid cells, called
trophoblast, that possess several important functions enabling viviparous devel-
opment (1,2). Trophoblast cells play key roles in protecting the embryo/fetus
from noxious substances, programming maternal support, and preventing mater-
nal immune rejection while at the same time ensuring appropriate bidirectional
nutrient/waste flow required for growth and maturation of the embryo. Although
placenta functions are highly conserved, species-specific elements of placenta
organization and activity are evident. Consequently, placental research has
benefited and will continue to benefit from a comparative approach. Each spe-
cies presents experimentally valuable attributes that can be exploited to better
understand the biology of the placenta and viviparity.
Research on placentas involves not only work on the placenta itself, but also
on the placenta’s maternal home, the uterus. Such studies include incorpora-
tion of experimental strategies beginning with the preparation of the uterus for
embryo implantation and techniques directed at understanding embryo–uter-
ine interactions. Cells involved in this initial interaction are trophoblast arising
from the blastocyst and those associated with the uterine epithelium (3). Tro-
phoblast cells expand in number and organize in species-specific patterns. In
From: Methods in Molecular Medicine, Vol. 121: Placenta and Trophoblast: Methods and Protocols, Vol. 1
Edited by: M. J. Soares and J. S. Hunt © Humana Press Inc., Totowa, NJ
01_Soares_1_6_F 3 8/29/05, 11:14 AM

4 Soares and Hunt
some species, trophoblast cells penetrate into the uterine compartment, estab-
lishing intimate relationships with the maternal vasculature, a process referred
to as hemochorial placentation and most commonly found in primates and
rodents (1,4). This type of placentation is also associated with a unique
specialization of the uterine stromal compartment, which is termed
decidualization. Other species exhibit minimal trophoblast invasion resulting
in a segregation of maternal and trophoblast tissues. This type of placentation
is referred to as epitheliochorial or synepitheliochorial and is seen in domesti-
cated animals, including the pig and ruminants (1). A key feature of the uterine
environment in animals possessing this superficial type of placentation is an
extensive development of uterine glands. These structures provide vital nutri-
tive support for the developing embryo and fetus throughout pregnancy in
domesticated species, and also during early stages of gestation in primates.
Thus, placentation is fundamental to creating the milieu in which the embryo
and fetus develop. The quality of the embryonic and fetal environment has last-
ing effects, influencing postnatal health and disease (5,6).
In this volume, the reader is guided through the major experimental strate-
gies and steps required for placenta research on multiple species, focusing on
modifications in the uterus, implantation, the nature of the trophoblast cell lin-
eage, vascular development and genetic analysis.
2. Methods for Studying Embryo Implantation and Uterine Biology

Part II of this volume focuses on protocols for learning more about the ini-
tial steps of pregnancy, including uterine preparation for pregnancy, blastocyst
development, and the process of implantation. Detailed in vivo procedures for
a comprehensive analysis of murine implantation are described in Chapter 2.
These techniques range from animal husbandry to preparation of an implanta-
tion-receptive uterus and embryo transfer. Protocols for assessing in vitro blas-
tocyst development and function are presented in Chapter 3, and include the
utilization of methods for imaging blastocyst–extracellular matrix interactions.
Chapters 4 through 6 outline strategies for isolating proliferative and differen-
tiated uterine stromal cells. The in vitro analyses include the utilization of both
rodent and human endometrial cells and permit the mechanistic examination of
factors controlling endometrial cell growth and differentiation. In vivo experi-
mental models are presented in Chapters 7 through 10. An in vivo technique
for studying the role of uterine glands and their secretions in the establishment
and maintenance of pregnancy is described in Chapter 7, whereas Chapter 8
provides an experimental in vivo strategy utilizing the baboon for studying
endometriosis, a prominent uterine disorder causing infertility. This section
01_Soares_1_6_F 4 8/29/05, 11:14 AM

Overview 5
concludes with presentations of nonhuman primate models for investigating

early pregnancy in vivo (Chapters 9 and 10).
3. In Vitro Trophoblast and Placental Model Systems

In this section, emphasis shifts to the placenta and its unique cell lineage,
the trophoblast. Key strategies for elucidating mechanisms controlling tropho-
blast cell growth, differentiation, and placental morphogenesis are presented.
The experimental approaches include trophoblast cell and placental organ cul-
ture systems, and the chapters describe techniques for isolating and culturing
rodent-, ruminant-, and primate-derived trophoblast cells. Chapters 11 and 12
provide detailed protocols for the establishment and manipulation of mouse
trophoblast stem cells from blastocyst and extraembryonic ectoderm. In Chap-
ter 13, the Rcho-1 rat trophoblast stem cell culture model, which is a well-
characterized system for studying rodent trophoblast giant cell differentiation, is
described. An in vitro procedure for the establishment and characterization of
bovine trophoblast stem cells is presented in Chapter 14. Each of the above
trophoblast cell systems can be expanded or induced to differentiate.
Chapters 15–18 provide four different model systems for studying human
trophoblast cells in vitro. They include the use of human embryonic stem cells,
the isolation and culture of term trophoblast cells, the production of tropho-
blast cell lines, and the culture of choriocarcinoma cell lines. The merits and
limitations of each system are described. Organ culture systems for analyzing
the primordial placenta and allantois, their vascularization and interconnec-
tions are documented in Chapter 19.
4. Phenotypic Analysis of the Placenta

In Part IV of this volume, protocols are provided for examination of placen-
tal development from rodents, ruminants, and primates. Chapters 20 and 21
present a comprehensive series of strategies for studying the development of
the mouse and rat placenta. Techniques described include approaches for tis-
sue dissection, trophoblast cell isolation, and structural and molecular analyses
of the rodent placenta. Morphological approaches for investigating ruminant
and primate placentas are described in Chapters 22 to 25. Immunocytochemi-
cal and in situ hybridization techniques and the appropriate immunological
and molecular probes for studying rodent, ruminant, and primate placental
development are provided in each chapter.
Strategies for investigating the maternal and fetal vasculature associated
with mouse and bovine placentas are described in Chapters 26 and 27. These
methods provide precise structural information on uteroplacental blood vessel
01_Soares_1_6_F 5 8/29/05, 11:14 AM

6 Soares and Hunt
development. Uteroplacental vascular beds dictate nutrient delivery and are

particularly susceptible to environmental insults and disease.
5. Molecular Analysis and Gene Transfer Techniques

The concluding portions of Volume I contain methods relevant to gene dis-
covery and analyses. For example, Chapters 28 and 29 describe techniques for
DNA microarray analysis, a powerful gene discovery tool; Chapter 28 focuses
on the design and execution of DNA microarray experiments; and Chapter 29
concentrates on data analysis. A characteristic of many placenta regulatory
genes is their allele-specific expression, a process termed imprinting. A thor-
ough guide to investigating gene imprinting is presented in Chapter 30. Analy-
sis of gene function using in vitro models requires the establishment of reliable
methods for gene manipulation. Chapter 31 describes the use of adenoviral-
mediated gene delivery systems that can be effectively used in primary tropho-
blast cells and trophoblast-derived cell lines.
References
1. Wooding, F. B. P. and Flint, A. P. F. (1994) Placentation. In: Marshall’s Physiol-
ogy of Reproduction, Fourth Edition, Vol. 3 (Lamming, G. E., Ed.), Chapman &
Hall, London: pp. 233–460.
2. Rossant, J. and Cross, J. C. (2002) Extraembryonic lineages. In: Mouse Develop-
ment (Rossant, J., and Tam, P. P. L., Eds.), Academic, San Diego: pp. 155–190.
3. Paria, B. C., Reese, J., Das, S. K., and Dey, S. K. (2002) Deciphering the cross-
talk of implantation: advances and challenges. Science 296, 2185–2188.
4. Georgiades, P., Ferguson-Smith, A. C., and Burton, G. J. (2002) Comparative
developmental anatomy of the murine and human definitive placentae. Placenta
23, 3–19
5. Bateson, P., Barker, D., Clutton-Brock, T., et al. (2004) Developmental plasticity
and human health. Nature 430, 419–421.
6. Gluckman, P. D. and Hanson, M. A. (2004) Living with the past: evolution, de-
velopment, and patterns of disease. Science 305, 1733–1736.
01_Soares_1_6_F 6 8/29/05, 11:14 AM

Mouse Implantation Methodologies 7
II
METHODS FOR STUDYING EMBRYO IMPLANTATION
AND UTERINE BIOLOGY
02_Deb_7_34_F 7 8/29/05, 11:15 AM

8 Deb, Reese, and Paria
02_Deb_7_34_F 8 8/29/05, 11:15 AM

2
Methodologies to Study Implantation in Mice
Kaushik Deb, Jeff Reese, and Bibhash C. Paria
Summary
Pregnancy begins with fertilization of the ovulated oocyte by the sperm. After fertilization,
the egg undergoes time-dependent mitotic division while trying to reach the blastocyst stage
and the uterus for implantation. Uterine preparation for implantation is regulated by coordi-
nated secretions and functions of ovarian sex steroids. The first sign of contact between the
blastocyst and the uterus can be detected experimentally by an intravenous blue dye injection
as early as the end of day 4 or the beginning of day 5 of pregnancy. This blastocyst–uterine
attachment reaction leads to stromal decidual reaction only at sites of implantation. The pro-
cess of implantation can be postponed and reinstated experimentally by manipulating ovarian
estrogen secretion. Stromal decidualization can also be induced experimentally in the hormon-
ally prepared uterus in response to stimuli other than the embryo. Fundamental biological ques-
tions surrounding these essential features of early pregnancy can be addressed through the
application of various techniques and manipulation of this period of early pregnancy. This
chapter describes the routine laboratory methodologies to study the events of early pregnancy,
with special emphasis on the implantation process in mice.
Key Words: Mouse; blastocyst; implantation; ovariectomy; vasectomy; delayed implanta-
tion; embryo transfer.
1. Introduction
The mouse, as one of the most common laboratory animals, is widely used
in basic biological research and could provide useful information that is rel-
evant to human biology. This chapter focuses on some of the procedures for
studying events of early pregnancy in mice. Following mating and fertiliza-
tion, the embryo develops to the blastocyst stage. Attachment of the blastocyst
into the uterine wall is an absolute requirement for further growth and collec-
tion of nutrients from the maternal vasculature. Hence, the implantation pro-
cess is a critical event in the embryo’s life and a central step to the establishment
of placentation and pregnancy.
02_Deb_7_34_F 9 8/29/05, 11:15 AM

The following description and methodologies are intended for investigators

who wish to pursue research in various aspects of early pregnancy, with spe-
cial reference to implantation-related processes. Studies on early pregnancy
start with breeding to generate pregnant females. Ideally, experimental mice
are maintained and bred in an institutional animal facility governed by the insti-
tutional animal care and use committee guidelines with the help of veterinarians
who supervise the health and well being of animals. Natural breeding is rou-
tinely used for performing research on pregnancy.
1.1. The Reproductive Cycle in Mice
The reproductive cycle in mice is known as the estrous cycle. The estrous
cycle is the time period from the onset of estrus until the onset of the next
estrus. The length of the estrous cycle varies depending on the animal species.
The average length of the estrous cycle is 4–5 d in mice, but it is highly vari-
able. The estrus stage signifies a period when females show signs of mating
behavior. Mice spontaneously ovulate during each estrous cycle. Females
become cyclic when they reach puberty by 4 wk of age. The different phases
of the estrous cycle in adult females are regulated by a functional hypothalamo–
pituitary–ovarian axis. Sexual maturity is coincident with pulsatile release of
gonadotropin releasing hormone (GnRH) from the hypothalamus with rising
levels of circulating gonadotropins, follicle-stimulating hormone (FSH), and
luteinizing hormone (LH) from the pituitary. While rising levels of FSH trig-
ger follicular growth and maturation, ovulation occurs under the influence of in-
creasing levels of LH. These changes are reflected in ovarian steroid production
during each cycle.
The stages of the estrous cycle are estrus, metaestrus, diestrus, and proestrus
(1). These stages occur in each cycle and in a sequential manner. The day of
estrus is usually designated as day 1 of the cycle. The stages of the estrous
cycle are best determined by the cell types observed in the vaginal smear. Nor-
mally, vaginal smears should be examined in the morning (0800 to 0900 h).
The estrous cycle is also divided into two ovarian phases: follicular phase and
luteal phase. Follicular phase is the period of ovarian follicle development, and
consists of proestrus and estrus. The luteal phase is the period of corpus luteum
formation and function, and comprises metaestrus and diestrus. The uterus also
undergoes hormonal changes during the estrous cycle. The uterus is distended
during proestrus and estrus as a result of an increase in uterine vascular perme-
ability and accumulation of fluid due to a higher level of circulating estrogen.
The distention starts to decline in late estrus and it is no longer observed at
diestrus. The wet and dry uterine weights are lowest at diestrus and heaviest at
proestrus.
02_Deb_7_34_F 10 8/29/05, 11:15 AM

1.2. Breeding of Mice

Sexual maturity in females occurs earlier than in males. Whereas females
are normally used for breeding at 45–50 d of age, males are not ready until 60 d
of age. Sexual maturity may be delayed a week or so in both males and females
depending on the strain of mice. Typically, males can breed successfully longer
than females because spermatogenesis continues throughout life. Many male
mice tend to become overweight with age, which may negatively affect their
ability to breed successfully. It is often recommended to retire old males (9 to
10 mo of age) and set up cages with new males.
1.2.1. Effect of Light Cycle on the Reproductive Behavior of Mice
The light cycle controls the reproductive performance of females and males
(2). Breeding conditions such as light cycle and timing must be carefully con-
trolled and regulated. The artificial light–dark cycle of an animal facility is
critical to the synchronous development of eggs. If female mice are purchased
from a commercial supplier, they should be allowed approx 1 wk to adjust to
the institutional animal room’s light–dark cycle. The release of LH, a pituitary
hormone that induces ovulation, is regulated by the light–dark cycle. The ani-
mal rooms are usually maintained at 12 h light:12 h dark or 14 h light:10 h dark
cycles.
1.2.2. Copulatory Plug Formation
The presence of a vaginal plug in the morning following copulation with a
male indicates successful mating. The ejaculate from the male’s accessory sex
glands forms a short-lived, whitish-looking or cream-yellow-colored plug in
the vagina of a female. The presence of a vaginal plug only indicates success-
ful mating, but does not always mean that a pregnancy will occur from this
mating. It should also be noted that sometimes plugs fall out of the vagina; this
may result in pregnancy that initially remains unnoticed, especially if checked
late in the morning.
1.3. Early Pregnancy
After ovulation, eggs released by the ovaries enter the associated oviduct.
Mouse ovaries are covered with the bursa (a thin membrane) and no egg can
escape into the abdominal cavity. Fertilization of the egg occurs in the ampulla
(ovarian end of the oviduct) after a successful mating. The egg completes its
first maturation division by the time ovulation occurs. If the egg has not com-
pleted the first maturation division, it does so very quickly after ovulation.
Female mice normally ovulate 8–10 ova in each cycle. Freshly ovulated eggs
are surrounded by a mass of cumulus cells. The uterus in the mouse consists of
02_Deb_7_34_F 11 8/29/05, 11:15 AM

two horns (duplex). After mating, sperm travel through both uterine horns to
reach the site of fertilization. They penetrate the cumulus cells to fertilize eggs.
Usually, more than one sperm enters the perivitelline space. However, only
one sperm penetrates and fertilizes the egg. After fertilization, the zygote
divides mitotically to eventually reach the blastocyst stage. After mating, the
mating stimulus triggers prolactin release from the pituitary, which leads to
the formation of a functional corpus luteum in the ovary, blocking further
ovulation and cyclicity to continue pregnancy.
1.4. Experimental Delay in Implantation
In many animals, implantation is delayed for an extended period, during
which the blastocyst remains in a quiescent state called embryonic diapause
(3,4). Delayed implantation in these animals seems to be a strategic plan for
regulating the time of birth coincident with favorable environmental condi-
tions. In some species, delayed implantation occurs under specific conditions.
Mice show postpartum estrus immediately after parturition. If conception takes
place immediately after parturition, embryos develop into blastocysts, but
remain in a dormant state until the lactational stimuli from suckling pups are
removed. In mice, implantation can be experimentally delayed by removing
the ovarian source of steroids (5). The timing of normal blastocyst implanta-
tion is tightly controlled in mice. Normally, initiation of implantation occurs at
night (2200–2300 h) of day 4 (6). Ovarian steroid hormones are necessary to
prepare the endometrium for the process of implantation. In mice, both ovarian
progesterone and estrogen are required for implantation. The ovary secretes a
small amount of estrogen in addition to progesterone in the morning of day 4
of pregnancy in mice. This preimplantation estrogen secretion is an absolute
requirement for blastocyst activation, preparation of the uterus, and initiation
of the implantation process. Surgical removal of both ovaries in mice before
the preimplantation ovarian estrogen secretion occurs on day 4 leads to delayed
implantation.
1.5. Artificial Decidualization
During normal pregnancy uterine stromal cells first proliferate and then dif-
ferentiate to decidual cells in response to an implanting blastocyst (4). This
process is known as decidualization. Decidualization starts following initia-
tion of blastocyst implantation in mice. The decidua enlarges as the embryo
grows. Decidual cells are characterized by the presence of polyploid nuclei,
and glycogen and lipid in their cytoplasm. As the deciduum grows, it occupies
the uterine lumen at the mesometrial side (dorsal to the embryo). The
antimesometrial decidua is divided into two zones. A thin and dense cellular
zone that immediately surrounds the blastocyst is known as the primary decidual
02_Deb_7_34_F 12 8/29/05, 11:15 AM

zone (PDZ). This is an avascular zone of the endometrium. The secondary

decidual zone (SDZ) surrounds the PDZ. The SDZ is a broad, well-vascular-
ized edematous zone.
The decidual response can also be induced experimentally without the pres-
ence of an embryo (4). However, a proper hormonally prepared uterus is needed
for this purpose. The uterus of a pseudopregnant female is a preferred choice.
The uterus of a progesterone-primed (at least 48 h), ovariectomized mouse can
also be used for this purpose.
Methods pertinent to the investigation of early pregnancy in the mouse are
described in this chapter.
2. Materials
2.1. Monitoring the Estrous Cycle
2.1.1. Collection of Vaginal Smears
1. Sexually mature female mice (45–50 d old).
2. Clean mouse cage with a wire-top cage cover.
3. Plastic dropper (Fisher Scientific, Hanover Park, IL, cat. no. 13-711-10).
4. Saline (0.9% Sodium chloride solution, Baxter Healthcare Corporation,
Deerfield, IL, cat. no. 281324).
5. Glass slides (Fisher Scientific, cat. no. 12-518-104).
2.1.2. Identification of the Stage of Estrous Cycle

1. Glass slides (Fisher Scientific, cat. no. 12-518-104).
2. A compound microscope with a 10× and 40× objectives.
2.2. Breeding and Plug Checking

2.2.1. Natural Breeding
2. Sexually mature male mice (60 d old).
3. Mouse cages.
2.2.2. Checking Copulatory Plug

1. Female mice mated with fertile males.
2. A pair of curved forceps (Fine Science Tools, Inc., Foster City, CA, cat. no.
11152-10).
2.3. Early Pregnancy Determination

2.3.1. Noninvasive Method
Vaginal plug-positive mice.
02_Deb_7_34_F 13 8/29/05, 11:15 AM

2.3.2. Invasive Method

2.3.2.1. EMBRYO COLLECTION FROM THE OVIDUCT AND UTERUS
1.Vaginal plug-positive mice.
2.70% ethanol (Aaper Alcohol & Chemical Co., Shelbyville, KY).
3.Paper towels.
4.Forceps (Fine Science Tools, Inc., cat. nos. 11151-10, 11153-10, and 11150-10).
5.Scissors (Fine Science Tools, Inc., cat. nos. 14558-11 and 15000-10).
6.Falcon Petri dish (Fisher Scientific, cat. no. 08-757-100B).
7.Whitten’s culture medium (per/100 mL): 514 mg NaCl, 36 mg KCl, 16 mg
KH2PO4, 29 mg MgSO4, 190 mg NaHCO3, 53 mg calcium lactate, 100 mg glu-
cose, 2 mg penicillin, 2 mg streptomycin sulfate, 0.5 mg phenol red, 3.5 mg pyru-
vic acid (sodium salt), 0.37 mL lactic acid (sodium salt), and 300 mg bovine
serum albumin.
8. BD brand disposable 3-mL syringe (Fisher Scientific, cat. no. 14-829-14B).
9. Hamilton 31-gauge steel needle (Fisher Scientific, cat. no. 14-815-619).
10. 27-gauge BD PrecisionGlide sterile disposable needle (Fisher Scientific, cat. no.
14-826-48).
2.3.2.2. DETECTION OF EARLY IMPLANTATION SITES BY INTRAVENOUS DYE INJECTION

1. Anesthesia (Avertin).
2. Paper towels.
3. BD PrecisionGlide 1-mL syringe with 27-gauge, one-half-inch needle (Becton
Dickinson & Co., Franklin Lakes, NJ; cat. no. 309623)
4. Blue dye (Chicago Blue B or Evans blue, or pontamine blue from Sigma Chemi-
cal Co.).
5. Saline (0.9% sodium chloride, Baxter Healthcare Corporation, cat. no. 281324).
6. Warm water.

2.4.1. Ovariectomy
1. Seminal plug-positive female mice (day 4 of pregnancy).
3. Animal clipper (Fisher Scientific, cat. no. 01-305-10).
4. 70% Ethanol (Aaper Alcohol & Chemical Co.).
5. Povidone-Iodine solution (Aplicare, Inc., Branford, CT).
6. Forceps (Fine Science Tools, Inc., cat. nos. 11153-10 and 11150-10).
7. Scissors (Fine Science Tools, Inc., cat. no. 14558-11).
8. BD Autoclip wound clips (Fisher Scientific, cat. no. 01-804-5) and applier (Fisher
Scientific, cat. no. 01-804).
2.4.2. Experimentally Delayed Implantation

1. Progesterone (Sigma Chemical Co., cat. no. P-1030)
2. Estradiol-17β (Sigma Chemical Co., cat. no. E-8875)
02_Deb_7_34_F 14 8/29/05, 11:15 AM


Dickinson & Co., cat. no. 309623).

2.5.1. Induction of Pseudopregnancy in Females Using Vasectomized
Males
2.5.1.1. PREPARATION OF VASECTOMIZED MALES
1. Sexually mature male (60 d old).
5. Povidone-Iodine solution (Aplicare, Inc.).
8. Ethicon nonabsorbable surgical suture (size 4.0; Ethicon, Inc., Somerville, NJ).
9. Slide Warmer (Lab-line Instruments, Inc., Melrose Park, IL; Model No. 26020).
2.5.1.2. INDUCTION OF PSEUDOPREGNANCY

1. Vasectomized male mice.
3. A pair of curved forceps for checking vaginal plugs (Fine Science Tools, Inc.,
cat. no. 11152-10).
2.5.1.3. HORMONAL PRIMING OF OVARIECTOMIZED MICE

1. Ovariectomized females rested for 10–15 d.
2. Sesame seed oil (Sigma Chemical Co., cat. no. S-3547).
4. Progesterone (Sigma Chemical Co., cat. no. P-1030).
5. Estradiol-17β (Sigma Chemical Co., cat. no. E-8875).
6. Hotplate (Fisher Scientific, cat. no. 11-497-6A).
2.5.1.4. INDUCTION OF DECIDUALIZATION BY ARTIFICIAL MEANS

1. Day 4 pseudopregnant female or ovariectomized progesterone-treated female
mice.
02_Deb_7_34_F 15 8/29/05, 11:15 AM


9. Sesame seed oil (Sigma Chemical Co., cat. no. S-3547).
2.6. Intrauterine Blastocyst Transfer

1. Day 4 pseudopregnant or progesterone-treated ovariectomized mice.
3. Paper towels.
8. Scissors (Fine Science Tools, Inc., cat. nos. 14558-11).
9. 23-gauge BD PrecisionGlide needle (Fisher Scientific, cat. no. 14-826A).
11. Slide Warmer (Lab-line Instruments, Inc., Model No. 26020).
13. Serrefine clip (Fine Science Tools, Inc., cat. no. 18050-35).
14. 1-mL Hamilton pipet controller syringe (Hamilton Company, Reno, NA; cat. no.
84001).
15. 6-in thin capillary (1 mm diameter) glass pipet (World Precision Instruments,
Inc., Sarasota, FL., cat. no. TW 100-6).
16. Popper 16-gauge steel needle (Fisher Scientific, cat. no. 14-825-16J).
2.7. Commonly Used Anesthetics

1. Injectable anesthetic, Avertin: (components: Avertin [2,2,2-tribromoethanol,
Sigma Aldrich Chemie GmbH, Steinheim, Germany; cat. no. T4,840-2] and tert-
amyl alcohol, Fisher Scientific, cat. no. A730-1]).
2. Short-lasting inhalant anesthetic, Isoflurane (Minrad, Inc., Buffalo, NY).
2.8. Common Injection Techniques

1. Mice.
3. Clean cage with a cage top.
2.9. Euthanasia
2.9.1. Cervical Dislocation
1. Mice.
2. Clean cage with a cage top.
02_Deb_7_34_F 16 8/29/05, 11:15 AM

2.9.2 Inhalants
2.9.2.1. CARBON DIOXIDE
1. Carbon dioxide cylinder (local gas supplier).
2. A cage specifically designed for killing mice.
2.9.2.2. ISOFLURANE
1. Isoflurane (Minrad, Inc.).
2. Cotton wool (Absorbent Cotton Co., Inc., Valley Park, MO) or gauze (Kendall
Healthcare Products Co., Mansfield, MA).
3. Bell jar or a scew cap glass container (Fisher Scientific).
3. Methods
3.1. Monitoring the Mouse Estrous Cycle
3.1.1. Collection of Vaginal Smears
1. Grasp the tail of a mouse with the thumb and forefinger of one hand.
2. Place the mouse on the top of the cage cover (wire top). As the mouse attempts to
move forward, quickly grasp the loose skin at the back of the neck using the
thumb and forefinger of the other hand. The head of the mouse will be immobi-
lized, if the skin is held properly.
3. Lift the mouse in your hand and secure the tail between the small finger and the
palm of the same hand.
4. Keep the face of the mouse up and locate the vagina.
5. Fill a plastic dropper with a small amount of saline (0.05 to 0.1 mL 0.9% NaCl)
and insert the tip superficially, but not deeply, into the vagina.
6. Gently squeeze the bulb of the dropper to release saline inside the vagina. Slowly
release the pressure on the bulb to aspirate the vaginal lavage inside the dropper.
3.1.2. Identification of the Stage of the Estrous Cycle
1. Place a drop of aspirated vaginal fluid on a glass slide.
2. Unstained material is observed under a light microscope with 10× and 40× objec-
tive lenses to identify the stage of the estrous cycle.
3. The following criteria are used to identify the specific stage of the cycle:
a. The estrus stage vaginal smear contains anucleated cornified cells (irregular
shaped cells).
b. The metaestrus stage vaginal smear is composed of a mixture of cornified
epithelial cells and leukocytes.
c. At diestrus, the smear contains predominantly leukocytes.
d. At proestrus, the smear shows a predominance of nucleated epithelial cells.
3.2. Breeding and Plug Checking

3.2.1. Natural Breeding
1. In general, 2–3 mature females irrespective of their estrous cycle are placed into
a male’s cage for breeding in the afternoon (1600–1800 h).
02_Deb_7_34_F 17 8/29/05, 11:15 AM

2. If not bred, the same females should be used for breeding 3–4 d in a row because
the pairing with males helps to synchronize the cycle in females. Males copulate
with females at proestrus at around the midpoint of the dark cycle.
3. After a successful mating, the male should be given a rest of 2–3 d.
3.2.2. Checking Copulatory Plugs

1. A method for holding the mouse has already been described in the section on
collection of vaginal smear (see Subheading 3.1.1.).
2. Hold the mouse in one hand with its face up.
3. Locate the vagina and use a small pair of curved forceps to spread the lips of the
vulva to identify the plug.
4. When gently touched with a pair of forceps, the plug feels solid and blocks the
vagina. It is a common practice to check the plug early in the morning, before
0900 h. The presence of a plug in the vagina is usually considered day 1 of preg-
nancy (see Note 1).
3.3. Detection of Early Pregnancy

3.3.1. Noninvasive Method
1. The presence of a plug in the vagina is defined as day 1 of pregnancy (see Note 2).
2. Implantation sites that look like a string of pearls in both uterine horns can be
detected by palpating the abdomen from day 8 onward (see Note 3).
3.3.2. Invasive Method

Dating of early pregnancy starts from the time of fertile mating to the time
of implantation occurring late on day 4 or early on day 5 (5). Timing can be
determined by assessing the developmental stage of the preimplantation embryo.
Developmental stages of preimplantation embryo and their location in the repro-
ductive tract are described in Table 1. Mice are normally sacrificed to collect
preimplantation embryos.
3.3.2.1. PREPARATION OF MICE FOR EXCISION OF OVIDUCTS AND UTERINE HORNS

1. Place a euthanized mouse on its back (supine position) on a paper towel.
2. Swab the belly with 70% ethanol.
3. Holding the lower half of the abdominal skin with a pair of forceps, make a small
lateral incision on the skin just below the forceps with a pair of scissors.
4. Holding the skin above the incision with your thumb and forefinger of one hand
and below the cut with other hand, pull the skin towards head and tail to expose
abdominal muscle.
5. Lift the abdominal wall muscle up with a pair of forceps and cut the abdominal
muscle from the midline laterally on both sides to expose internal organs.
6. Push aside the intestine to visualize uteri, oviducts, and ovaries.
02_Deb_7_34_F 18 8/29/05, 11:15 AM

Table 1
Dating of Early Pregnancy Depending on the Developmental Stage of Preim-
plantation Embryos*
Day of pregnancy Developmental stage of embryos Reproductive tract
1 One-cell zygote Oviduct

2 Two-cell Oviduct
3 (0100–0400 h) Four-cell Oviduct
3 (0500–1400 h) Eight-cell Oviduct
3 (1500–2300 h) Morula Oviduct
4 Blastocyst Uterus
*The time of embryonic development may be slow or fast, depending on the strain of mice
and light-dark cycle of an animal facility.
3.3.2.2. EXCISION AND FLUSHING OF OVIDUCTS

1. Using a pair of curved forceps, grasp the uterine horn just below the utero-tubal
junction and cut the horn just below the pair of forceps (Fig. 1).
2. Then lift the uterine horn and clean the underlying fat pad and mesentery that are
attached to the oviduct using a pair of iris scissors.
3. Separate the oviduct from the ovary using the same pair of scissors and place the
oviduct in a Petri dish with a drop of Whitten’s media for flushing to recover
embryos.
4. Repeat this procedure for the contralateral oviduct.
5. Place a fresh Petri dish on the stage of a stereo dissecting microscope.
6. Attach a 31-gauge needle with a blunt end that has been bent in the middle to
form a 120° angle to a 1-mL or a 3-mL plastic syringe filled with Whitten’s
medium.
7. Using a pair of forceps, place an oviduct on the dish under the microscope and
manipulate the oviduct to locate the fimbriated end of the oviduct known as the
infundibulum.
8. Hold the infundibulum loosely between the forceps and insert the needle inside
the oviduct (Fig. 2).
9. Holding both the needle and the oviduct together, gently squeeze the syringe to
pass the Whitten’s medium through the oviduct. This procedure should inflate
the oviduct and flush the embryos from the oviduct into the dish.
10. Examine the flushing under the microscope to identify developmental stage of
embryos.
3.3.2.3. EXCISION AND FLUSHING OF UTERI
1. Using a pair of forceps, grasp one uterine horn just above the cervical bifurcation
and cut below the point of holding with a pair of scissors (Fig. 3).
2. Pull up the uterine horn and trim the horn free of fat and mesentery with a pair of
scissors (Fig. 3).
02_Deb_7_34_F 19 8/29/05, 11:15 AM

Fig. 1. Schematic representation to show excision of the mouse oviduct. The method
of excision of mouse oviduct is described under Subheading 3.3.2.2.
Fig. 2. Schematic representation to show flushing of the mouse oviduct. The method
of flushing oviducts is described under Subheading 3.3.2.2. The oviduct is flushed to
recover preimplantation embryos.
3. Cut the other end of the horn just below the utero–tubal junction and keep the
uterine horn in a clean moistened tissue paper to absorb blood.
4. Hold one of the uterine horns at the utero–tubal junction and insert the tip of a 27-
gauge needle, with a 3-mL plastic syringe filled with Whitten’s medium attached,
inside the uterine lumen (Fig. 4).
02_Deb_7_34_F 20 8/29/05, 11:15 AM

Fig. 3. Schematic representation to show excision of uterine horns. The method of

excision of mouse uterus is described under Subheading 3.3.2.3.
Fig. 4. Schematic representation to show flushing of the mouse uterine horn to

recover blastocysts. The method of flushing uterine horns is described under Sub-
heading 3.3.2.3.
5. Holding the needle and the uterine horn together, push the plunger of the syringe
to flush the uterine luminal contents into a Petri dish. It is important to flush
gently (Fig. 4).
6. Repeat this procedure to excise and flush the other uterine horn.
7. Check uterine flushings under a stereomicroscope for the presence of appropriate
stages of developing embryos.
02_Deb_7_34_F 21 8/29/05, 11:15 AM

3.3.2.4. DETERMINATION OF EARLY IMPLANTATION SITES BY INTRAVENOUS DYE INJECTION

Implantation sites in mice can be detected as early as late at night on day 4
(2200–2300 h) and onward, considering the presence of a copulatory plug as
day 1 of pregnancy. This is achieved by intravenous injection of a macromo-
lecular blue dye solution, normally via a tail vein (7).
1. Fill a 1-mL syringe attached to a 27-gauge needle with 1% blue dye solution
(Chicago Blue B, Evans blue, or pontamine blue) avoiding any air bubbles inside
the syringe.
2. After a mouse is anesthetized, dilate the tail veins by the application of a paper
towel soaked in warm water.
3. Locate one of the two lateral veins in the tail (veins are located on both sides of
the central artery) and place the mouse on that side.
4. Hold the tail gently between the thumb and forefinger and keep the tail parallel to
the body of the mouse (Fig. 5).
5. Align the needle (bevel side up) with the plane of the vein. Insert the needle into
the vein and slowly inject the desired amount of dye (0.1 mL/mouse, 0.25 mL/
rat). As a result of increased capillary permeability in the endometrial bed at the
sites of implantation, the dye bound with the serum proteins accumulates in the
interstitial space at the sites of blastocysts, showing distinct blue bands (Fig. 6).
Chicago Blue B dye has been used for many years to identify implantation sites
(see Note 4).
6. Animals are sacrificed 3–5 min after dye injection to identify blue bands in the
uterus. Identification of uterine implantation sites from day 6 onward does not
require blue-dye injection. Visual observation of prominent intermittent swell-
ings in the uterus indicates that blastocyst implantation is in progress.

3.4.1. Ovariectomy
1. Under anesthesia, shave the lumbar dorsum bilaterally and place day 4 pregnant
animals in a prone position (face down).
2. Clean the exposed skin of the back with a 10% povidone-iodine scrub followed
by 70% alcohol for aseptic surgery.
3. Make a midline skin incision near the abdominal cavity (on the back).
4. Turn the animal to one side (left or right). Pull the skin incision laterally away
from the spine and make an incision on the abdominal muscle to locate the para-
ovarian fatty tissue (light whitish-yellow-colored fat) (Fig. 3).
5. Lift out the para-ovarian fatty tissue and excise the ovary by making a sharp cut
between the oviduct and the ovary.
6. Rejoin the fat tissue with the oviduct in the abdominal cavity.
7. Repeat the same procedure on the other side. If the incisions are small, there is no
need to close the incisions in the abdominal muscle.
8. Close the midline skin incision using wound clips.
02_Deb_7_34_F 22 8/29/05, 11:15 AM

Fig. 5 (see companion CD for color version). Detection of implantation sites by

intravenous injection of macromolecular dye into tail vein. The method of intravenous
Chicago Blue B dye injection into tail vein is described under Subheading 3.3.2.4.
Fig. 6 (see companion CD for color version). Implantation sites in uterine horns on
day 5 of pregnancy as detected by intravenous Chicago blue B dye injection. The
method of intravenous Chicago Blue B dye injection into tail vein is described under
Subheading 3.3.2.4.
02_Deb_7_34_F 23 8/29/05, 11:15 AM

3.4.2. Experimentally Delayed Implantation

During experimentally delayed implantation, blastocysts enter a state of
dormancy or diapause and the uterus enters a neutral state (3,4). This condition
can be maintained for several days and sometimes even weeks by continuing
daily progesterone (1 or 2 mg/mouse) injection. However, the number of dor-
mant blastocysts is gradually reduced if this delayed condition is maintained
for a long time. Uterine luminal closure and blastocyst apposition occur during
delayed implantation, but the attachment and invasion of trophoblast cells and
decidual transformation of uterine stromal cells do not occur. Activation of the
dormant blastocyst and implantation can be achieved by giving an injection
(subcutaneous) of estradiol-17β (3–25 ng/0.1 mL of sesame seed oil). The implan-
tation sites in the uterus can be detected by tail-vein injection of blue dye (see
Subheading 3.3.2.4.) 18–24 h after estradiol-17β injection (8) (see Note 5).
3.5.1. Induction of Pseudopregnancy in Females Using Vasectomized
Males
Regular cyclic females are mated with vasectomized infertile males to induce
pseudopregnancy.
3.5.1.1. VASECTOMY
Vasectomy is performed on fertile male mice at 6–8 wk of age. Surgical
resection of the vas deferens eliminates sperm from the ejaculate.
1. Place an anesthetized male on the table in supine position (face up).
2. Shave the lower half of the abdomen (anterior to the genital) to remove hair,
swab, and make a midline ventral incision (one-half of an inch) in the skin ante-
rior to the genitalia.
3. Using a pair of curved forceps and scissors, pull the abdominal muscle up and cut
the muscle in the midline (a white line in the muscle). Wipe the incision site with
a lint-free tissue dampened with saline to remove excess cut hair.
4. Push the left scrotum with fingers so that the testis moves up into the abdominal
cavity. The testis will be out together with the vas deferens.
5. Grasp the left vas deferens gently with forceps and lift a section clear of the
incision. Tuck the curved forceps underneath the vas deferens and allow them to
spread. Maintaining this position, use the suture to make two firm knots in the
vas deferens, about 4–5 mm apart, tying both knots firmly.
6. Cut out a section of vas deferens between the knots.
7. Place all tissues back inside the abdominal cavity and push the testis back into
the scrotum. Repeat the procedure on the right side of the body.
8. When both sides have been done, sew up the incision in the body wall with sepa-
rate sutures.
02_Deb_7_34_F 24 8/29/05, 11:15 AM

9. Close up the skin with two to three auto clips. The mouse should be wrapped in a
tissue to keep it warm (loss of body heat is common in abdominal surgery) or,
alternatively, placed on a heating pad, and allowed to recover. Animals that are
placed under anesthetic should always be supervised and monitored until fully
awake.
10. Following the operation, the mice are allowed to recover for 2 wk before being
test-bred to confirm their sterility. One or two female mice are placed with the
vasectomized male and are checked for plugs the following morning. Plug-posi-
tive females are sacrificed on day 2, and their oviducts are flushed with saline.
The eggs should be at unfertilized, one-cell stage. The presence of two-cell
embryos would indicate incomplete vasectomy.
3.5.1.2. INDUCTION OF PSEUDOPREGNANCY

The uterine environment of most laboratory animals becomes receptive to
implantation only after mating. Vaginal–cervical stimulation during mating
results in ovarian hormonal changes that alter the estrous cycle in preparation
for a possible pregnancy. Pseudopregnancy can be achieved in one of two ways:
(1) by mating a female to a sterile (vasectomized) male or (2) by mechanical
stimulation of the vagina and cervix with a rod or vibrating tool. Currently,
natural mating of a mature female with a vasectomized male (1:1) is a pre-
ferred choice of producing pseudopregnant females. Detection of a copulatory
plug in the vagina of a female mated with a vasectomized male is defined as
day 1 of pseudopregnancy. Procedures for breeding and checking plugs are as
described previously (see also Note 6).
3.5.1.3. HORMONAL PRIMING OF OVARIECTOMIZED MICE

The method of ovariectomy is described earlier (see Subheading 3.4.1.). In
general, cyclic young females are ovariectomized and rested for a couple of
weeks to eliminate ovarian steroids from the circulation. After recovery, ova-
rian steroid hormones are administered to prepare the uterus for a chemical or
physical stimulus. Either progesterone alone or a regimen of progesterone and
estrogen-17β supplementation are used to sensitize the uterus for the decidual
cell reaction (9,10).
1. Treatment schedule of progesterone alone: ovariectomized mice receive proges-
terone (1 mg/d/mouse) injection (subcutaneous) for 3 d (days 1–3). The induc-
tion of deciduoma is initiated on day 4 by infusing oil inside the uterine lumen.
Daily progesterone injection is continued after the stimulus to maintain the
decidualization response (9).
2. Treatment schedule of estradiol-17β and progesterone: ovariectomized mice re-
ceive injections of estradiol-17β (100 ng/d/mouse) for 3 d (days 1–3), no treat-
ment on days 4 and 5, progesterone (1 mg/d/mouse) plus estradiol-17β (10 ng/d/
02_Deb_7_34_F 25 8/29/05, 11:15 AM

mouse) on days 6–8. The induction of deciduoma is initiated on day 8 by infusing

oil inside the uterine lumen (10). Daily progesterone (1 mg/d/mouse) injection is
continued after the stimulus to maintain the decidualization response.
3.5.1.4. INDUCTION OF DECIDUALIZATION BY ARTIFICIAL MEANS

Artificial decidualization is usually induced by injecting sesame seed (or
corn) oil or by placing a silk thread inside the uterus. Injection of sesame seed
oil inside the uterus is a less invasive procedure and a preferred method to
induce decidualization.
1. Fill a plastic 1-mL syringe attached to a 27-gauge needle with sesame seed oil.
2. Prepare the female and expose one uterine horn as previously described (see Sub-
heading 3.4.1.).
3. Hold the uterine horn with a pair of forceps very close to the tip (slightly below
the utero–tubal junction) and slowly inject about 20 µL of oil inside the uterine
lumen. Because the luminal fluid volume is very low, the uterine horn will tem-
porarily swell during the course of oil injection (see Note 7).
3.6. Intrauterine Blastocyst Transfer

3.6.1. Induction of Pseudopregnancy in Females
The method of inducing pseudopregnancy in females is described in an ear-
lier section (see Subheading 3.5.1.2.). The pseudopregnant mouse contributes
a womb for the transfer embryos. Commercially available outbred albino CD-1
females are a great choice for recipient mice.
3.6.2. Hormonal Priming of Ovariectomized Mouse for Embryo Transfer
These females are often used to study the role of the steroid hormones,
progesterone and estrogen, for preparation of the uterus for implantation. The
method of performing ovariectomy and the preparation of progesterone-treated
ovariectomized mice are described above (see Subheadings 3.4.1. and
3.5.1.3.).
1. Ovariectomized females start to receive daily subcutaneous injection of progest-
erone (1 or 2 mg/mouse/0.1 mL of sesame seed oil) for two consecutive days
before they are ready to receive a blastocyst. Progesterone injections make the
uterus achieve the prereceptive state.
2. This progesterone-primed prereceptive uterus achieves receptivity in response to
a single subcutaneous injection of estradiol-17β (25 ng/mouse/0.1 mL sesame
seed oil) (10). It has been established that a minimum of 48 h of uterine exposure
to progesterone is necessary before an injection of estradiol-17β is provided in
order to attain uterine receptivity in mice.
3. The blastocyst transfer is usually performed on the third day at the time of the
third progesterone injection and an estradiol-17β injection.
02_Deb_7_34_F 26 8/29/05, 11:15 AM

4. Implantation of the blastocyst can be checked by blue-dye methods 24 h after the

blastocyst transfer and estrogen injection.
3.6.3. Preparation of a Blastocyst Transfer Pipet

1. Take a 6-inch thin capillary (1 mm diameter) glass pipet and rotate it in a fine
flame approx 1 inch from one end. As soon as the glass becomes soft, withdraw
the glass from the heat and quickly pull both ends apart.
2. Check the diameter of the pulled pipet under a microscope and break the end at a
place where the diameter will be greater (approx 200 µm) than the size of a blasto-
cyst. One can use an oilstone to mark the breaking point of the glass for an even tip.
3. Fire-polish the tip of the pipet by quickly touching the flame (an uneven tip may
damage the blastocyst and the uterus).
4. Bend the tube (at a 120–130° angle) over a flame about one-half of an inch from
the unpulled end.
5. Place the unpulled end of the pipet inside a 16-gauge steel needle and seal it
(make it air-tight) with “super” glue. The leakage of solution can be tested by
passing water through the needle.
6. This pipet-and-needle assembly is then fitted to special 1-mL Hamilton pipet-
controller syringe. This special syringe has a plunger assembly with a thumb-
wheel cap inside of a glass barrel. Pull approx 0.2 mL of sterile water into the pipet
barrel by pulling the pipet-controller plunger assembly. Fill the inside plunger with
water by turning the plunger thumb-wheel counterclockwise (all the way to the
end). Avoid drawing any bubbles inside the syringe. Connect the luer tip of the
pipet-controller syringe barrel and the embryo transfer pipet. Turn the plunger
thumb-wheel clockwise to push the water into the transfer pipet. Fix the water
level in the middle of the transfer pipet. The transfer pipet is now ready for load-
ing blastocysts.
3.6.4. The Blastocyst Transfer Technique

1. Prepare a pseudopregnant or hormonally treated ovariectomized mouse as described
above (see Subheadings 3.5.1.2. and 3.5.1.3.).
2. Anesthetize the mouse with avertin administered intraperitoneally.
3. Place the anesthetized mouse on a clean piece of tissue paper on a clean table in
a prone position (head down).
4. Shave its lower back and both sides of the abdomen. Swab the shaven area with a
10% povidone-iodine scrub and 70% ethanol.
5. Turn the mouse toward one side. Using a pair of forceps, hold the skin of the
abdomen and make a small cut (approx 1 cm long) with a pair of scissors.
6. Next, hold and slightly pull the abdominal muscle and make a small incision in
the abdominal muscle, avoiding the blood vessels of the muscle. If the cut has
been made in the right place, the ovarian fat pad is easily visible. If not, slightly
lift the edge of the body wall and try to locate the fat pad.
7. Once the fat pad is located, gently lift the fat pad out the body. Ovary, oviducts,
and a part of the uterus will come out with the fat pad (Fig. 7).
02_Deb_7_34_F 27 8/29/05, 11:15 AM

Fig. 7 (see companion CD for color version). Schematic representation to show

exteriorization of utero-tubal junction in the anesthetized mouse. The method of exte-
riorization of utero–tubal junction is described under Subheading 3.6.4.
8. Attach a serrefine clip to the fat pad (do not to clip the ovary). In the absence of
the ovary in the hormone-treated ovariectomized mice, clip the fat and mesentery
near the oviduct. Try not to touch the uterine horns during this procedure.
9. Locate the tip and gently hold the uterine horn with a pair of forceps approx 1 cm
below the utero–tubal junction (Fig. 7).
10. Five to seven blastocysts must be transferred to each horn. Take up a minute
amount of embryo culture medium (Whitten’s media) in the very tip of the trans-
fer pipet by moving the plunger cap counterclockwise. Next, make a small bubble
by taking up a little air. Then take up some more medium—roughly the same
volume as you hope to transfer the blastocysts in. Take up another bubble, the
same size as before. Then take up blastocysts in the smallest possible volume of
medium, lining them up side by side in the transfer pipet (see Note 8).
11. Once the pipet is loaded and the uterine horn positioned, gently grasp the top of
the uterine horn inside a pair of forceps.
12. While still holding the horn with one hand, use the other hand to gently insert a
26-gauge hypodermic needle through the uterine wall (close to the oviduct) and
into the lumen (Fig. 8). Choose an area of the horn that is relatively devoid of
blood vessels because blood will clot in the tip of the pipet and block it. Remove
02_Deb_7_34_F 28 8/29/05, 11:15 AM

Fig. 8 (see companion CD for color version). Insertion of a transfer pipette into the
uterine horn for embryo transfer. The method of blastocyst transfer inside uterine lu-
men is described under Subheading 3.6.4.
the needle and carefully (so as not to lose the site of the hole), without averting
your eyes, pick up the loaded transfer pipet. Gently insert the transfer pipet tip
about 3 mm into the uterine lumen (Fig. 8). Gently release the blastocysts into
the uterus by turning the plunger cap clockwise. Be careful not to allow any air
into the uterus.
13. Once the transfer is complete, quickly rinse the transfer pipet in some Whitten’s
medium and check to see if there were any blastocysts stuck in the transfer pipet.
If there were, transfer these blastocysts again.
14. With the transfer complete, the serrefine clip can now be removed and the uterine
horn gently eased back into the body. Do not touch the uterus, but ease it back by
lifting the edges of the incision in the body wall and allowing the horn to fall
back in, without actually handling it.
15. This procedure is then repeated on the other uterine horn.
16. The incision in the body wall is not sutured. The skin is closed with Michelle
clips—two per incision is usually sufficient. Once surgery is complete, the mouse
is placed in a clean cage and warmed to facilitate recovery (see Note 9).
02_Deb_7_34_F 29 8/29/05, 11:15 AM

3.7. Chemicals Commonly Used to Anesthetize the Mouse

There are three components of anesthesia: analgesia (pain relief), amnesia
(loss of memory), and immobilization (no movement). Because general anes-
thetics affect the central nervous system and anesthesia is required by law to
prevent pain and distress in research animals, it must not be taken lightly. It is
not necessary to withhold food and water from rodents prior to anesthesia.
3.7.1. Avertin (A Commonly Used Injectable Anesthetic)
1. Preparation of an Avertin (components: Avertin [2,2,2-tribromoethanol] and tert-
amyl alcohol) stock solution (1.6 g/mL):
a. Add 15.5 mL T-amyl alcohol to 25 g of Avertin in a dark bottle.
b. Stir on magnetic stirrer until the avertin is dissolved (approx 12 h).
c. Store in the dark at room temperature (see Note 10).
2. Preparation of an Avertin working solution (20 mg/mL):
a. Mix 0.5 mL Avertin stock solution and 39.5 mL normal saline in a beaker.
b. Seal container with parafilm, wrap in foil to exclude light, and stir on mag-
netic stirrer for about 12 h or until dissolved.
c. Avertin working solutions must be kept refrigerated in a dark bottle until used
and should be replaced at least every month.
d. Sterilize through 0.2-µm filter and store at 4°C. Working solutions should be
replaced at least every month.
3. Dosages (0.4–0.6 mg/g body weight). 0.45–0.75 mL/mouse administered intra-
peritoneally (see Note 11).
3.7.2. Commonly Used Short-Lasting Inhalant Anesthetic (Isoflurane)
1. Approximately 1 mL isoflurane is placed on a cotton ball in a bell jar or screw-
top glass jar.
2. The mouse is then inserted inside the jar and removed when it is fully uncon-
scious (unconsciousness can be judged by pinching the toe).
3. Duration of these anesthetics is 30 s to 2 min.
4. A nose cone containing a small amount of anesthetic is placed in front of the nose
to maintain the depth of anesthesia.
3.8. Injection Techniques in the Mouse
Four types of injections are commonly used in mice: subcutaneous (SC),
intraperitoneal (IP), intravenous (IV), and intramuscular (IM). Animals must
be adequately restrained or anesthetized to receive injection.
3.8.1. Subcutaneous Injection (Anesthesia Not Required)
1. Grasp the base of the tail with the thumb and forefinger of one hand.
2. Place the mouse on the top of the cage cover (wire top). Clear the back of the
neck with an alcohol swab.
02_Deb_7_34_F 30 8/29/05, 11:15 AM

3. Hold the syringe attached to a 27-gauge needle parallel to the head.

4. As the mouse attempts to move forward, quickly insert the tip of the needle in the
scruff (loose skin on the back of the neck) at a very shallow angle and lift the skin
with the needle to avoid underlying muscle. Now inject the solution and remove
the needle slowly to avoid leaking. The larger the volume of injected solution,
the greater the likelihood of leakage. It is also advisable to pinch the injection
side with your thumb and forefinger to prevent leaking.
3.8.2. Intraperitoneal Injection (Anesthesia Not Required)

1. The procedures for grabbing and holding the mouse are described earlier.
2. Clean the injection site of one side of the abdomen with an alcohol swab. The
caudal left abdominal quadrant is the preferred place for IP injection in order to
avoid the cecum on the right.
3. Tilt the animal toward its head in order to allow the abdominal contents to fall
away from the injection site.
4. Quickly insert a 27-gauge needle attached to a 1-mL sterile syringe containing
the drug down through the abdominal wall to the peritoneal cavity and inject the
animal. It is not uncommon to inject volumes up to 1 mL by this route.
3.8.3. Intravenous Injection (Anesthesia Required)

This method is similar to the method of injection of blue dye through tail
vein (Fig. 1).
3.8.4. Intramuscular Injection (Anesthesia Not Required)

1. Grasp the mouse as described in the methodology for subcutaneous injection.
2. The quadricep muscle and the posterior thigh are acceptable sites for intramuscu-
lar injections. Clean the injection sites with an alcohol swab.
3. Insert the needle through the skin into the muscle and inject the desired amount.
3.9. Euthanasia
Cervical dislocation is the most common method of killing mice. However,
mice can be killed using inhalants.
3.9.1. Cervical Dislocation

1. Hold the mouse at the base of the tail with the thumb and forefinger of one hand.
2. Keep the mouse on the cage top. As the mouse tries to move forward, quickly
place the thumb and the forefinger of other hand behind the skull and hold
firmly on the cage top.
3. Next, pull the tail in the direction away from the body. This will dislocate the
neck. This should be performed very quickly.
02_Deb_7_34_F 31 8/29/05, 11:15 AM

3.9.2. Inhalants
3.9.2.1. CARBON DIOXIDE
Carbon dioxide inhalation is the most efficient and acceptable method of
euthanasia.
1. A mouse cage with a solid lid is connected to a carbon dioxide gas cylinder.
2. Place the mice in the cage and cover the cage with a lid.
3. Open the carbon dioxide cylinder and fill the cage with gas.
4. The animals will die within 1–2 min.
3.9.2.2. ISOFLURANE
Isoflurane overdose can also be used to kill mice.
1. A cotton wool or gauze soaked with isoflurane is placed inside a bell jar or a
screw-cap glass container.
2. Place the mouse inside the jar. The mice will die within 1 min (see Note 12).
4. Notes
1. The process of detecting copulatory plugs should be performed gently because
stimulation of the vagina may induce pseudopregnancy.
2. Pregnant mice usually do not breed when placed with male mice. There is no
other visual or noninvasive method for definitive identification of early preg-
nancy.
3. Abdominal distention is apparent in most mice by day 8 or later depending on the
litter size and degree of swelling of the implantation sites.
4. This is a relatively simple procedure but requires practice. If the needle is inside
the vein, injection will be smooth. If the syringe plunger does not move smoothly
and resistance is felt while injecting or swelling around the injection site occurs,
withdraw the needle and try again slightly above the first injection site (proximal
to the body). It is always advisable to start injecting from the tip of the tail. After
several attempts, it is advisable to change the needle because the tip becomes
blunt.
5. The purpose of describing delayed implantation is that this model provides a
powerful tool to examine steroid hormone regulation of uterine and embryonic
changes with respect to embryo–uterine interactions during implantation.
6. The pseudopregnant female will display the hormonal profile of a normal preg-
nant female for several days after mating. The hormonal milieu of pseudopreg-
nancy begins to differ from pregnancy after 7 to 8 d as a result of the absence of
a developing embryo inside the uterus.
7. Injection of too much oil inside the lumen may migrate to the other uterine horn
and cause decidualization. Prominent swelling of the uterus will indicate the
extent of stromal cell decidualization in response to the artificial stimulus.
Swelling of the uterus due to decidualization will be visible 48 h after the oil
injection. The intraluminal oil injection to a pseudopregnant mouse uterus on day
02_Deb_7_34_F 32 8/29/05, 11:15 AM

4 afternoons (1300–1400 h) will yield the best results. If ovariectomized mice are
used, animals should be exposed to progesterone (SC injection) for at least 48 h
(daily injection of 2 mg progesterone per 0.1 mL sesame seed oil per day for 2 d)
before the injection of oil inside the uterine lumen. These animals must also be
maintained with daily progesterone injection after the induction of decidualization.
8. Loading blastocysts into the transfer pipet will take some practice. If it is likely
to take more than a few minutes to load the transfer pipet, then do not expose the
uterine horn until the pipet has been loaded. This prevents drying out and further
trauma to the uterine horn. Alternatively, the uterine horn, ovary, and so on may
be moistened repeatedly with a sterile cotton bud and saline.
9. Animal welfare guidelines recommend that all vertebrates undergoing procedures
that might cause more than momentary pain or distress be treated with analge-
sics, unless it can be scientifically justified that the treatment will interfere with
the experimental procedure. Analgesics should be given immediately after the
surgery. A simple skin incision may only require 24 h of analgesic treatment.
Rodents subjected to abdominal surgery or similar procedures normally require
analgesic for the first 12 h. It is not appropriate to wait until signs of pain or
distress are demonstrated before administering analgesics. In rodents, the signs
of pain following surgery are manifested as decreases in food and water con-
sumption. Consult your veterinarian and animal care committee for specific post-
operative care in your institute, because these procedures are based on
institutional rules and regulations.
10. The Avertin stock solution is light-sensitive and hydroscopic. Keep Avertin in a
dark bottle at room temperature. The Avertin stock solution is quite stable at
room temperature
11. It will take about 3–5 min for the mouse to become fully anesthetized (lack of toe-
pinch reflex). An additional 0.1–0.2 mL can be administered if required. The mouse
will remain anesthetized for approx 15–20 min and recover within 30–60 min.
Keep the mouse warm during recovery. The effective dosage is dependent upon
the weight of the mouse.
12. Isoflurane should be used in a fume hood to minimize the risk of exposure to the
gas by the operator.
Acknowledgments
We gratefully acknowledge Dr. S. K. Dey for helpful discussions and expert
advice. This work was supported by National Institutes of Health (NIH) grant
HD 42636, HD 40193, HD 44741 and HD 37394 to B.C.P.
References
1. Allen, E. (1922) The estrous cycle in the mouse. Am. J. Anat. 30, 297–371.
2. Snell, G. D., Fekete, E., Hummel, K. P., and Law, L. W. (1940) The relation of
mating, ovulation and estrous smear in the house mouse to time of day. Anat. Rec.
76, 39–54.
02_Deb_7_34_F 33 8/29/05, 11:15 AM

3. Carson, D. D., Bagchi, I., Dey, S. K., et al. (2000) Embryo implantation. Dev.
Biol. 223, 217–237.
4. Dey, S. K. (1996) Implantation. In: Reproductive Endocrinology, Surgery and
Technology (Adashi, E. Y., Rock, J. A., and Rosenwaks, Z, Eds.), Lippincott-
Raven, New York: pp. 421–434.
5. Yoshinaga, K. and Adams, C. E. (1966) Delayed implantation in spayed, proges-
terone treated adult mouse. J. Reprod. Fert. 12, 593–595.
6. Das, S. K., Wang, X. N., Paria, B. C., et al. (1994) Heparin-binding EGF like
growth factor gene is induced in the mouse uterus temporally by blastocysts solely
at the site of its apposition: a possible ligand for interaction with blastocyst EGF-
receptor in implantation. Development 120, 1071–1083.
7. Psychoyos, A. (1973) Endocrine control of egg implantation. In: Handbook of
Physiology, Williams and Wilkins, Baltimore, MD: pp.187–215.
8. Paria, B. C., Huet-Hudson, Y. M., and Dey, S. K. (1993) Blastocyst’s state of
activity determines the “window” of implantation in the mouse receptive uterus.
Proc. Natl. Acad. Sci. USA 90, 10,159–10,162.
9. Paria, B. C., Tan, J., Lubahn, D. B., Dey, S. K., and Das, S. K. (1999) Uterine
decidual response occurs in estrogen receptor-α-deficient mice. Endocrinology
140, 2704–2710.
10. Wordinger, R. J., Jackson, F. L., and Morrill, A. (1986) Implantation, deciduoma
formation and live births in mast cell-deficient mice (W/Wv). J. Reprod. Fertil.
77, 471–476.
02_Deb_7_34_F 34 8/29/05, 11:15 AM

Blastocyst Culture 35
3
Blastocyst Culture
D. Randall Armant
Summary
Experimental models of blastocyst development based on in vitro culture have played a
prominent role in advancing our understanding of peri-implantation development, a process
that is relatively inaccessible in vivo. Blastocyst culture provides a robust approach for exam-
ining embryonic interactions with the microenvironment under highly controlled conditions.
Major events that occur in utero can be followed in vitro, including blastocyst expansion, hatch-
ing, and adhesion to extracellular matrices. This chapter will describe a method for obtaining
and culturing mouse blastocysts. Morphological changes that occur during blastocyst culture
will be discussed and related to the corresponding development in utero. Finally, quantita-
tive assays will be detailed for monitoring peri-implanatation development of the trophoblast
in vitro.
Key Words: Blastocyst; mouse; trophoblast; embryo culture; hatching; implantation; attach-
ment reaction; adhesion; migration; fibronectin binding assay; outgrowth; microspheres; image
analysis; morphometry; extracellular matrix; fibronectin; Matrigel; collagen gel.
1. Introduction
Preimplantation development in mice takes place between embryonic days
(E) 0.0 and E 3.5, resulting in the formation of a blastocyst. As with all pre-
implantation stages, the blastocyst can complete development outside of the
female reproductive tract under the direction of an endogenous developmen-
tal program. The early stages of peri-implantation development are recapitu-
lated in vitro, including blastocyst expansion, cell proliferation, and hatching
from the zona pellucida. After hatching, interactions with the uterus that occur
in vivo from E 4.0 to E 6.5 can be simulated by following the attachment of the
blastocyst to the culture plate and, if an appropriate extracellular matrix (ECM)
is provided, monitoring the migratory or invasive activity of trophoblast cells.
The maternal milieu plays a central role in augmenting blastocyst develop-
ment. Blastocyst culture provides an experimental system for investigating
35
03_Armant_35_56_F 35 8/29/05, 11:15 AM

36 Armant
interactions between the embryo and individual components of the maternal

environment. Using this approach, it has been possible to establish several
molecular mechanisms that contribute to the embryonic–maternal dialogue.
1.1. Blastocyst Formation and Architecture
Spatial information plays a critical role in blastocyst formation. During pre-
implantation development, the trophectoderm differentiates from cells lining
the outside of the compacted morula (Fig. 1). Developmental cues garnered by
the presence or absence of asymmetrical cell contacts determine cell fate after
compaction (1). Outside cells differentiate into trophectoderm, the first epithe-
lium and precursor of all trophoblast lineages, whereas inside cells become the
pluripotent inner cell mass (ICM). During trophectoderm differentiation, tight
junctions form between the outside blastomeres and they begin to actively
transport fluid into the embryo, generating a blastocoel. As the blastocyst
forms, the trophectoderm is stretched and the cluster of cells that forms the
ICM becomes confined to a sector of the trophectoderm wall (2).
The trophoblast originates from the trophectoderm layer of the blastocyst
during peri-implantation development. Therefore, much can be learned about
trophoblast development by culturing blastocysts in vitro. The blastocoel physi-
cally separates some of the trophectoderm cells from the ICM and, thus, pro-
vides spatial cues for differentiation of the mural trophectoderm. Polar
trophectoderm comprises the outside cells that directly contact the ICM (3).
Under the direction of the ICM, polar trophectoderm cells proliferate rapidly
until displaced into the mural trophectoderm, where cell division slows consid-
erably (4,5). In the mouse, mural trophectoderm cells endoreduplicate their
DNA and differentiate into invasive giant trophoblast cells.
1.2. Modeling Implantation In Vitro
Mural trophectoderm cells convert during implantation to an invasive phe-
notype, the trophoblast, by altering their adhesive interactions, dissociating
from one another, and penetrating the endometrium (6,7). Trophoblast cells
eventually infiltrate the maternal vascular system. The undifferentiated tro-
phectoderm cells possess a nonadhesive apical surface. Uterine edema pro-
motes tight apposition between the hatched blastocyst and uterine epithelium,
leading quickly to attachment between the luminal epithelium of the uterus and
the apical surface of the trophectoderm. The attachment reaction precipitates
the death of luminal epithelial cells directly contacting the blastocyst, which
allows the trophoblast giant cells to adhere to the underlying basal lamina. In
the final stages of implantation, trophoblast cells complete their differentiation
to the invasive phenotype. These developmental events are recapitulated dur-
ing in vitro blastocyst culture, provided certain requirements detailed in this
03_Armant_35_56_F 36 8/29/05, 11:15 AM

Fig. 1. Formation of the mouse blastocyst. At the eight-cell stage, the embryo com-
pacts to produce a morula (A). After the next cell division, the embryo begins to pro-
duce a cavity or blastocoel and forms a blastocyst by the 32-cell stage (B). Two
populations of cells emerge during blastocyst formation. The inside cells of the morula
(white) become the inner cell mass (ICM), a group of pluripotent stem cells eccentri-
cally positioned within the blastocyst. Cells in the outer layer of the morula (gray)
form the first epithelium, the trophectoderm, which differentiates into the trophoblast
lineages of the placenta. The blastocyst expands by accumulating additional fluid
within the blastocoel and hatches free of the zona pellucida (C). In the mouse, polar
trophectoderm cells contacting the ICM proliferate, while the remaining mural tro-
phectoderm cells differentiate into trophoblast giant cells.
chapter are met. Blastocyst differentiation in vitro from gestation day (GD) 4
to GD 7 (see Note 1) is slow, as assessed by activities that are believed to
correspond to developmental events occurring in utero (Table 1). However,
the introduction of growth factors normally provided by the uterus can acceler-
ate in vitro development significantly (8).
1.2.1. Expansion and Hatching
Developing blastocysts accumulate fluid, expanding their circumference as
the blastocoel volume increases (Fig. 1C). Expansion of the blastocoel can be
monitored by morphometric measurement and proceeds linearly from GD 4 to
03_Armant_35_56_F 37 8/29/05, 11:15 AM

38 Armant
Table 1
Comparison of Blastocyst Developmenta
Time in vitro
Developmental events (in vitro counterpart) Time in utero Morning of:
Early blastocyst E 3.5 GD 4

Expanded and hatching E 3.8 GD 5
Attachment reaction (in vitro attachment) E 4.0 GD 6
Adhesion to basal lamina (ECM-binding activityb) E 5.0 GD 7
Invasion of stroma/ECM (trophoblast outgrowth) E 6.2 GD 7-10
aDevelopment of blastocysts in utero from embryonic day 3.5 (E 3.5) to E 6.2 compared with
that of blastocysts collected from the uterus on gestation day (GD) 4 at 0900 h and cultured in
serum-free medium. In vitro, blastocysts attach nonspecifically to plates on GD 6. They adhere
specifically to surfaces or microspheres coated with extracellular matrix (ECM) proteins on GD
7. Invasive activity is assessed from GD 7 onward by monitoring trophoblast outgrowth on ECM.
bDiscussed in this chapter as fibronectin-binding activity
GD6 (9). On GD 7, adhesion-competent blastocysts collapse, perhaps as a result

of the breakdown of trophoblast junctions prior to outgrowth. During blastocyst
expansion, the zona pellucida ruptures as a result of proteolytic activity pro-
duced by the trophectoderm (10). The combined action of proteases and pres-
sure generated by blastocyst expansion produces an opening in the zona
pellucida that the trophectoderm eventually squeezes through in order to hatch
(Fig. 2A). Hatching is usually completed by mid-day on GD 5, leaving behind
an empty zona pellucida. In utero, additional proteases that are present in the
uterine luminal fluid completely dissolve the zona pellucida on E 3.8–4.0.
Hatching is easily observed during blastocyst culture and can be quantified by
counting embryos that are unhatched, hatching, or completely free of the zona
pellucida (11).
1.2.2. Attachment Phase
Attachment between the apical surfaces of the trophectoderm and luminal
epithelium in utero has been documented by ultrastructural analysis (7) and its
molecular basis is under intense investigation (12). There is a clear distinction
between this attachment of opposing apical cell surfaces, a rare event for epi-
thelia, and the adhesion of trophoblast cells to components of the basal lamina
and stromal ECM that occurs later. The attachment reaction begins in mice at
approximately E 4.0 (13). Direct evidence that trophoblast cells become physi-
cally attached to the epithelium is derived primarily from observations made in
vitro. As blastocysts differentiate during co-culture with a monolayer of uter-
ine epithelial cells, they become difficult to dislodge by shaking the culture
03_Armant_35_56_F 38 8/29/05, 11:15 AM

Fig. 2. Differentiation of the blastocyst in vitro. The blastocyst expands and hatches
by squeezing through a tear in the zona pellucida (A). Around the time that the tropho-
blast becomes adhesion competent, the blastocyst collapses (B). An empty zona to the
left clearly shows the opening formed during hatching. The earliest sign of trophoblast
outgrowth is the appearance of spreading cells near the base of the embryo (arrow-
heads; C). During the next 24 to 72 h, the field of migrating trophoblast cells sur-
rounding the embryo increases in area (D–F). The clump of cells at the center of the
outgrowth consists of remnants of the ICM and undifferentiated trophoblast cells.
vessel (14,15). Because blastocysts cultured to GD 6 will attach nonspecifically

to the culture plate as readily as to a cell monolayer, it is questionable whether
their attachment to cultured cells is physiologically significant. However, the
ability of cultured blastocysts to attach in vitro is developmentally regulated,
and may represent cellular events associated with attachment to the luminal
epithelium in utero. There are sweeping changes in surface charge and
proteoglycan expression known to occur coincidentally with the attachment
reaction (16) that could make blastocysts “sticky” in culture. The ability of
blastocysts to attach in medium supplemented only with bovine serum albumin
(BSA) peaks on GD 6 and ceases shortly before they begin to adhere and out-
grow on GD 7 (17,18), further suggesting that transient attachment to culture
plates could represent differentiation related to the attachment reaction in utero.
Therefore, assessment of blastocyst attachment during the course of in vitro
03_Armant_35_56_F 39 8/29/05, 11:15 AM

40 Armant
culture can provide useful information regarding developmental progression,

but it should be recognized as an event separate from trophoblast adhesion to
ECM.
1.2.3. Adhesion/Invasion Phase
Adhesion commences in utero between E 5.0 and 5.2, based on ultrastruc-
tural observation of mural trophoblast cells adhering to the endometrial
basal lamina that is exposed upon sloughing of luminal epithelial cells (19).
Experimental studies using cultured blastocysts have shed important light on
the molecular basis of trophoblast adhesion to ECM. Mouse blastocysts will
adhere and their trophoblast cells will spread and outgrow in serum-free media
on substrata containing highly purified preparations of ECM proteins (20). Tro-
phoblast invasion of three-dimensional ECM has also been investigated using
blastocysts cultured in collagen gels, basement membranes or ECM purified
from the endometrium (21,22). Blastocyst outgrowth culture was developed
using serum supplementation (23), and provided a useful model system of peri-
implantation development (24,25). It can be inferred that the ability of mouse
trophoblast cells to outgrow in vitro correlates with their competence to implant
in utero (26). Trophoblast cells are promiscuous in their choice of adhesion pro-
teins, readily outgrowing on fibronectin, laminin, vitronectin, collagens,
entactin, and other components of the ECM (20,27–30).
The rate of blastocyst differentiation in culture can be determined by assess-
ing embryos for the first sign of trophoblast outgrowth, which presents as a
spreading monolayer of cells around the base of the embryo concomitant with
disappearance of the spherical blastocyst morphology (Fig. 2). A commonly
used strategy for assessing the influence of culture conditions and additives on
blastocyst differentiation is to compare outgrowth rates. In serum-free culture
on ECM-coated surfaces, blastocysts become adhesive and commence tropho-
blast outgrowth between GD 6 and GD 7 (Table 1). Whereas the developmen-
tal program of the blastocyst proceeds in the absence of an adhesive substratum,
trophoblast cells must contact the ECM to undergo morphological transforma-
tion (31). Without the provision of ECM proteins or serum, the late blastocyst
collapses and remains essentially intact until it finally degenerates. If blasto-
cysts are transferred to an ECM after they become adhesion-competent on GD 7,
signs of spreading trophoblast are apparent within 1–2 h. By GD 8, their ability
to adhere diminishes, suggesting that there is a finite window of time for tro-
phoblast adhesion. The trophoblast appears to be unable to recover once the
window is past (31).
We have developed an alternate approach to more precisely gage adhesive
trophoblast differentiation (18). Prior to observable outgrowth, trophoblast
03_Armant_35_56_F 40 8/29/05, 11:15 AM

Fig. 3. Adhesion of fibronectin coated microspheres to mouse blastocysts. Groups

of five embryos are viewed by fluorescence microscopy after decoration with fluores-
cent microspheres coated with FN-110, as detailed in the text. Before assay, blasto-
cysts cultured to gestation day 7 were incubated for 1 h with either bovine serum
albumin (A) or 50 µg/mL FN-110 (B). Arrowhead in B denotes polar trophoblast.
cells must become adhesive at their apical surfaces, dissociate cell–cell junc-
tions, spread, and begin to migrate. The time between acquisition of adhesion
competence and our ability to detect migrating trophoblast cells posed a seri-
ous technical limitation for experimentally assessing blastocyst development.
We were able to monitor early signs of adhesive activity on the apical surface
of trophoblast cells by decorating intact blastocysts with fluorescent
microspheres coated with adhesive ECM proteins (18) (Fig. 3). Coating
microspheres with a proteolytic fragment of fibronectin containing the central
cell adhesion-promoting domain (FN-110), we demonstrated that binding was
specific and dependent on appropriate integrins. The onset of fibronectin-
binding activity correlated with trophoblast outgrowth on plates coated with
fibronectin, demonstrating a physiological relationship between the two experi-
mental approaches.
The fibronectin-binding assay, quantified using fluorescence microscopy
and image analysis, has proven highly useful in experimental designs to test
agents that delay (31) or accelerate (8) the rate of blastocyst development.
Using this approach, we found that adhesion to fibronectin at the apical sur-
face of the trophoblast is upregulated after an initial exposure to fibronectin for
1–3 h (18) (Fig. 3). Furthermore, fibronectin was shown to ligate integrins
localized at the apical surface of the trophectoderm, activating intracellular
03_Armant_35_56_F 41 8/29/05, 11:15 AM

42 Armant
signaling pathways that increase fibronectin-binding activity (32). The traf-

ficking of additional proteins, including the αIIb integrin subunit, into the apical
domain of the trophoblast plasma membrane is induced by fibronectin-mediated
integrin signaling and contributes to the intensified adhesion (33). Used in com-
bination with molecular and cytological approaches to assess the expression
and localization of adhesion complex components, the evaluation of adhesive
activity during trophoblast differentiation has provided new insights into the
developmental mechanisms that guide blastocyst implantation (34).
This chapter details protocols for the production and culture of mouse blas-
tocysts, as well as assays used to monitor the developmental progression of
blastocysts in vitro. The laboratory manual by Hogan and colleagues (35) can
be consulted for additional information regarding embryo culture methods and
specialized laboratory techniques.
2. Materials
1. B6SJLF1/J male mice, 3 to 9 mo old (Jackson Laboratory, Bar Harbor, ME) (see
Note 2).
2. Outbred (e.g., CF1, CD1, Swiss-Webster) female mice, 4–6 wk old (Charles
River Laboratories, Wilmington, MA) (see Note 3).
3. Pregnant mare serum gonadotropin (PMSG; Sigma Chemical Co., St. Louis, MO)
and human chorionic gonadotropin (hCG; Sigma), each in phosphate-buffered saline
(PBS) at 50 IU/mL. Stocks (100X) are prepared in distilled water at 5000 IU/mL and
stored at –70°C.
4. Dissecting scissors (surgical and fine) and forceps (No. 5).
5. 27-gauge hypodermic needle and 1–2 mL syringe.
6. Embryo culture medium: Ham’s F-10 culture medium (Invitrogen, Carlsbad,
CA), supplemented with 4 mg/mL BSA (A3311, Sigma), 75 µg/mL penicillin
(P4687, Sigma) and 75 µg/mL streptomycin (S1277, Sigma) (see Note 4).
7. M2 medium (M5910, Sigma) is prepared with 4 mg/mL BSA and antibiotics, as
for Ham’s F-10 medium (see Note 5).
8. Water-extracted oil: light mineral oil (Aldrich Chemical Co. Inc., Milwaukee,
WI) is shaken with one part sterile distilled water to nine parts oil and allowed to
phase separate.
9. Small Petri dishes (351007 or 351008, BD Biosciences, Bedford, MA).
10. Micropipets prepared from Pasteur pipets heated in a flame and pulled to a diam-
eter of 75–100 µm. Latex tubing with a mouthpiece is attached for mouth opera-
tion, as described in Hogen et al. (35).
11. Dissecting stereomicroscope.
12. A nonhumidified tissue culture incubator set at 37°C and 5% CO2 (see Note 6).
13. Matrigel™, collagen type I and collagen type IV (BD Biosciences).
14. Human plasma fibronectin, mouse laminin-1 from Engelbreth-Holm-Swarm
(EHS) sarcoma (both from Invitrogen), human plasma vitronectin (Calbiochem,
La Jolla, CA) and α1-acid glycoprotein (Sigma). Protein stocks are prepared at
03_Armant_35_56_F 42 8/29/05, 11:15 AM

1 mg/mL in PBS and stored frozen in small aliquots that are slowly thawed on ice
before dilution to the appropriate final concentration.
15. Peptides for coating plates: FN-110 (08-103, Upstate Biotech, Lake Placid, NY),
a 110 kDa proteolytic fragment of fibronectin with adhesive activity, and
GRGDSP (03-34-0035, Calbiochem), a synthetic peptide containing the critical
sequence that mediates adhesion in the cell binding domain of fibronectin. Both
peptides are prepared and stored as in item 14.
16. Inverted microscope with fluorescence detection system (DM-IRB, Leica,
Wetzlar, Germany), interfaced through a B/W CCD digital camera (Orca,
Hamamatsu Photonics K.K., Hamamatsu City, Japan) with computer-based image
analysis software (Simple PCI, Compix Inc., Canberry Township, PA), or a com-
parable system from other manufacturers.
17. Fluorescent polystyrene microspheres (Bangs Laboratories, Fishers, IN) approx
1 µm in diameter (see Note 7).
18. Heparitinase (from Flavobacterium heparinum; E.C. 4.2.2.8; 100704-1, Seikagaku
America, East Falmouth, MA) prepared at 0.1 U/mL in Hank’s balanced salt solu-
tion (Invitrogen).
3. Methods
3.1. Blastocyst Production and Culture
3.1.1. Superovulation and Mating
1. Female mice are superovulated by injection of 5 IU PMSG, followed 44–48 h
later with 5 IU of hCG. Injections of 0.1 mL are given intraperitoneally between
1200 and 1600 h (see Note 8).
2. Immediately after the hCG injection, each female is paired for mating overnight
with a stud male.
3. The females are checked the next morning for a vaginal plug (solidified semen at
the entrance to the vagina) by lifting the tail and probing with a small spatula or
blunt rod. Plugs must be checked early, as they may fall out by the late morning.
4. The pregnant female mice are then separated from males and may be housed in
groups until the desired stage of pregnancy is reached. Nonpregnant animals may
be re-used for superovulation and mating after resting them for 2–3 wk.
5. Embryos can also be obtained without superovulation by monitoring the estrus
cycle of each female to determine the appropriate time for mating (35).
3.1.2. Blastocyst Collection

Blastocysts are collected from the uterus (see Note 9) before they hatch
from the zona pellucida on E 3.5 (Fig. 1B).
1. Before collecting embryos, set up at least two microdrop cultures (see Note 10).
In addition to the microdrop cultures, fill a Petri dish with M2 medium for col-
lecting uterine horns. All plates should be equilibrated in a tissue culture incuba-
tor for at least 30 min before embryo collection begins.
03_Armant_35_56_F 43 8/29/05, 11:15 AM

44 Armant
2. Pregnant mice are euthanized by cervical dislocation, the body is doused with
70% ethanol, and a small ventral incision is made with surgical scissors in the
abdominal skin. The skin is then grasped on either side of the incision and pulled
apart to expose the underlying body wall. The peritoneal cavity is opened using
fine scissors to make a lateral incision through the ventral abdominal wall at a
level approximating the top of the rear legs. After pushing the intestines upward
to expose the reproductive tract, grasp the junction between the oviduct and the
left uterine horn with fine forceps. Free as much fat and membrane from the
uterine horn as possible.
3. Cut the uterotubal junction on the side of the forceps towards the oviduct. A
second cut just above the junction of the two uterine horns will free the left horn.
Place the uterine horn into M2 medium and repeat the procedure with the right
uterine horn.
4. After collection, each uterine horn is flushed with 1 mL of M2 medium using a
syringe needle inserted into the lumen of the horn at the uterotubal junction. The
junction wall is held tightly to the needle with forceps as medium is gently released
from the syringe into the lumen, exiting at the opposite end into an empty Petri
dish. Collection of embryos is made easier by preventing the medium from reach-
ing the wall of the Petri dish.
5. Using a stereomicroscope, the medium is scanned for blastocysts, which are col-
lected by mouth-operated micropipet (see Note 11).
6. Embryos are transferred to microdrop cultures and separated from debris or
contaminating epithelial cells by transfer through at least three more drops of
medium. They can then be transferred to new microdrop cultures or harvested
for other purposes.
3.1.3. Blastocyst Culture

1. Microdrop cultures containing blastocysts are incubated without humidification
in a tissue culture incubator at 37°C and 5% CO2. Embryos may be cultured
singly or in groups, allowing 1–2 µL of medium per embryo.
2. Once the blastocysts have hatched from the zona pellucida, it is better to culture
them singly to prevent their aggregation. Another advantage of culturing blasto-
cysts singly is the ability to monitor the development of each embryo individu-
ally. It is not necessary to change the culture medium unless the ratio of embryos
to culture medium is significantly increased.
3.2. Blastocyst Developmental Assays

In addition to monitoring the rates of blastocyst expansion and hatching, as
described under Subheading 1.2.1., there are several measures of late blasto-
cyst differentiation that can be quantified experimentally. The attachment
reaction requires interactions with the uterine epithelium, but the correspond-
ing stage of blastocyst differentiation occurs during culture after the blastocyst
hatches, causing the embryo to transit from completely free floating to immo-
03_Armant_35_56_F 44 8/29/05, 11:15 AM

bile on the culture plate surface. A fibronectin binding assay can be used to
directly measure the adhesion competence of trophoblast cells, demonstrating
adhesive activity on the apical surface prior to the onset of outgrowth (18).
Blastocyst outgrowth on ECM is representative of the period of trophoblast
invasion and may be quantified either by determining the percentage of blasto-
cysts with migrating trophoblast cells, or by measuring the area occupied by
trophoblast cells as they migrate outward. Later, we provide protocols for assess-
ing these measures of blastocyst differentiation during development in vitro.
Normally, blastocysts are cultured on BSA-coated surfaces until GD 5,
allowing them to hatch from the zona pellucida and approach the attachment
competent stage. Most embryos will hatch independently, but those that fail
should be freed of the zona by repeatedly drawing the blastocyst in and out of
a micropipet drawn to a diameter slightly smaller than that of the embryo.
Hatched or dezonaed blastocysts are best cultured singly in small microdrops
to avoid their aggregation. It is convenient to arrange the microdrops in a cir-
cular pattern on the Petri dish for easy scanning when assessing their progress.
Once embryos are added to culture medium, minimize their exposure to ambi-
ent room conditions by rapidly returning them to the incubator after each
observation.
3.2.1. Blastocyst Attachment
Attachment is determined by swirling or tapping the culture plate while
observing the embryo through a stereomicroscope. Unattached blastocysts
will move relative to debris or marks in the plastic, whereas attached embryos
remain firmly in place. Removing the zona pellucida prematurely does not alter
the timing of blastocyst attachment during culture (24). However, delayed hatch-
ing will obscure detection of attachment and subsequent outgrowth. Therefore,
it is expedient to mechanically remove zonae from any blastocysts that have
not hatched by the end of GD 5.
Whereas attachment is transient in medium containing only BSA, it is con-
tinuous with trophoblast adhesion and outgrowth on surfaces coated with ECM
components or in serum-supplemented medium. The time between the onset of
attachment and the beginning of outgrowth can be as long as 12–24 h, although
supplementation with serum or growth factors can shorten it to less than 1 h.
3.2.2. Trophoblast Outgrowth
To form a blastocyst outgrowth (Fig. 2), trophoblast cells must adhere to an
ECM that can be provided by supplementing the medium with serum, which
contains fibronectin and vitronectin. Alternatively, blastocysts will outgrow in
a defined serum-free medium when cultured either on polystyrene surfaces
coated with individual ECM proteins or on gels composed of purified collagen
or a complex basement membrane (Matrigel).
03_Armant_35_56_F 45 8/29/05, 11:15 AM

46 Armant
3.2.2.1. OUTGROWTH ON GELS

1. Several types of collagen, prepared in acidified solution, can be used to form a
gel by neutralization with culture medium and incubation at 37°C. Gel formation
will not occur if the solution is maintained at 0–4°C.
2. Matrigel is a complex, collagen-rich basement membrane that is provided frozen
in a neutral solution. It is thawed overnight at 4°C to prevent gel formation.
3. A volume of ice-cold Matrigel or neutralized collagen sufficient to generate the
desired gel thickness is removed with a chilled pipet, quickly deposited on a
surface, and warmed to 37°C to promote gel formation.
4. The newly formed gel is rinsed several times with culture medium and equili-
brated in an incubator before introducing embryos.
5. For microdrop culture, this procedure can be carried out on the surface of a Petri
dish using 1–5-µL drops of collagen or Matrigel.
6. After gelation at 37°C in a humidified chamber, a small amount of medium is
applied to wet each gel and the plate is flooded with mineral oil.
7. Medium above the gel is changed and embryos are introduced through the oil
using micropipets, as described under Subheading 3.1.2.
8. Using an inverted microscope, each embryo is observed at regular intervals to
determine if migrating trophoblast cells are present at the base of the blastocyst.
Mouse trophoblast cells outgrow on the surface of Matrigel, but penetrate type I
collagen gels to produce a three-dimensional outgrowth (21). Trophoblast cells
of certain other species with highly invasive implantation, including guinea pigs
(21), nonhuman primates (36), and humans (37), are capable of invading
Matrigel.
3.2.2.2. OUTGROWTH ON COATED PLASTIC

1. Coating the culture surface with individual ECM proteins is best accomplished
using non-tissue culture-treated plastic (e.g., Petri dish, plates made with untreated
polystyrene). In addition to intact proteins (fibronectin, laminin-1, collagen,
vitronectin, entactin), we have used a proteolytic fragment of fibronectin that
contains the central cell binding domain (FN-110), the E8 proteolytic fragment
of laminin-1 or synthetic peptides having the Arg-Gly-Asp-Ser sequence recog-
nized by several integrins (27,38,39).
2. Matrix proteins or peptides are diluted from stock solutions to 50 µg/mL in ster-
ile PBS at room temperature (PBS diluent must contain no additional protein)
and added immediately to the culture surface. Higher protein concentrations are
required to coat glass surfaces.
3. For microdrop cultures, drops of the precoating solution are placed on a Petri
dish, which is flooded with oil and incubated overnight at 4°C–37°C. The
precoating solution is then removed and the surface washed several times
with sterile PBS containing 4 mg/mL BSA. The same solution is used to block
the coated surface for 1 h. Care should be taken not to scrape the treated culture
surface with a pipet. The PBS/BSA solution is then replaced with serum-free
03_Armant_35_56_F 46 8/29/05, 11:15 AM

Fig. 4. Determination of outgrowth area. A blastocyst was cultured to gestational

day (GD) 5, transferred to a plate coated with fibronectin and further cultured until
GD 9. The resulting embryo with outgrowing trophoblast cells appears in A. The
image is assessed for area using image analysis software by tracing the outer edge of
the outgrowth using a computer mouse, as depicted in B. Calibration performed using
an image of a stage micrometer obtained at the same magnification allows the soft-
ware to determine the delineated outgrowth area in µm2.
medium and allowed to equilibrate in a tissue culture incubator before inserting

embryos.
4. Observation with a steromicroscope is adequate for assessing outgrowth, although
an inverted microscope used at higher magnification can be helpful. The blastocoel
collapses shortly before embryos begin to outgrow (Fig. 2B). When scoring out-
growth cultures, blastocysts are assessed for a spreading monolayer of tropho-
blast cells around the base of the embryo concomitant with disappearance of the
spherical blastocyst morphology (Fig. 2C–F). Positive scores are given to embryos
as soon as the first signs of trophoblast outgrowth are visualized (Fig. 2C).
3.2.3. Measurement of Outgrowth Area

Once trophoblast outgrowth is intitiated, the area occupied by the cells can
be measured to determine rates of cell spreading and migration.
1. Images of blastocyst outgrowths are obtained daily from GD 6 to GD 10 using an
inverted microscope (200× magnification) interfaced through a digital camera
with an image analysis system.
2. Using software that includes morphometry capabilities, the perimeter of each
outgrowth is traced with a computer mouse (Fig. 4) and converted to area in µm2,
based on prior calibration of the system with a stage micrometer.
03_Armant_35_56_F 47 8/29/05, 11:15 AM

48 Armant
3. After the progression of outgrowth area expansion has been established in a time
study, subsequent experiments may be conducted by choosing one time point
that is within the linear range of increasing outgrowth area (see Note 12).
3.2.4. Measurement of Fibronectin-Binding Activity
Although the binding assay described here is intended for measuring cell
adhesion to fibronectin, a similar approach can be adapted for other proteins
that mediate trophoblast adhesion. For large proteins, it is best whenever pos-
sible to coat polystyrene microspheres with a small active fragment of the pro-
tein to increase ligand density. We have coated fluorescent microspheres with
proteolytic fragments containing the cell binding domains of fibronectin,
laminin, and entactin to assess the adhesive activity of developing blastocysts.
The fibronectin-binding assay is performed in three steps.
1. It is necessary to remove charged heparan sulfate from the surface of the tropho-
blast to prevent nonspecific binding of microspheres through electrostatic inter-
actions.
2. Fibronectin-binding activity at the apical surface of the blastocyst must be
upregulated through exposure to soluble or immobilized fibronectin for reasons
discussed under Subheading 1.2.3.
3. Blastocysts are incubated with fibronectin-coated microspheres and high affinity
binding is assessed.
Procedures for each of these steps will be detailed, along with protocols for
preparing fibronectin-coated microspheres and quantifying microsphere bind-
ing to blastocysts.
3.2.4.1.PREPARATION OF MICROSPHERES
1. Fluorescent microspheres (1.0-µm diameter) are supplied as 2.5% solutions and
must be washed to remove surfactants.
2. Centrifuge 200 µL of the microspheres suspension in a 0.6-mL tube at 10,000g
for 1 min.
3. Remove the supernatant, add 200 µL sterile PBS and vortex to resuspend.
4. Repeat steps 2 and 3 twice.
5. Centrifuge the suspension at 10,000g for 1 min.
6. Remove the supernatant and add 200 µL sterile PBS containing 144 µg/mL FN-110.
7. Agitate the suspension at room temperature for 24 h on a vortex mixer set to its
lowest speed.
8. Repeat steps 2 and 3 three times.
9. Remove the supernatant and add 200 µL sterile PBS containing 1 mg/mL α1-acid
glycoprotein as a blocking step (see Note 13).
10. Repeat step 7.
11. Repeat steps 2 and 3 three times.
12. Remove the supernatant and add 200 µL PBS containing 10 mg/mL BSA (PBS/
BSA).
03_Armant_35_56_F 48 8/29/05, 11:15 AM

13. Store microspheres at 4°C for up to 2 mo (final concentration is 2.5%).

14. To begin an experiment, remove 10 to 20 µL of the microsphere suspension, and
dilute 1:10 (final concentration of 0.25%) in ice-cold PBS/BSA. For a negative
control, dilute 1:10 with PBS/BSA containing 1 mg/mL of fibronectin.
15. Prepare ice-cold drops of washing solutions and microspheres in microdrop cul-
ture and keep cold in refrigerator until use.
3.2.4.2. LIGAND-MEDIATED UPREGULATION OF BINDING ACTIVITY

Upregulate fibronectin-binding activity by incubating blastocysts for 1 h at
37°C in Ham’s F10 medium containing 50 µg/mL FN-110 (Fig. 3). This can
also be accomplished by culturing blastocysts for 3 h on a Petri dish coated
with fibronectin, as described in the initial reports of this method (18,31). How-
ever, it is recommended that soluble ligand be used to upregulate fibronectin-
binding activity, as some of the blastocysts become damaged or too adherent to
recover from the ligand-coated plate. Blastocysts incubated on non-coated
plates in medium containing only BSA will provide an indication of the extent
of increase in binding activity, which should be three- to sixfold (see Note 14)
3.2.4.3. REMOVAL OF SURFACE HEPARAN SULFATE

After exposure to ligand, transfer blastocysts through three drops of me-
dium. Blastocysts are then incubated in 0.1 U/mL heparitinase for 30 min at
37°C in a tissue culture incubator.
3.2.4.4. INCUBATION WITH MICROSPHERES

1. Wash blastocysts in ice-cold PBS/BSA three times by sequential transfer through
microdrops.
2. Incubate blastocysts with FN-110 coated microspheres (0.25% in PBS/BSA) at
4°C for 30 min (see Note 15).
3. Wash blastocysts four times in PBS/BSA at 4°C by transfer through microdrops.
This will remove microspheres bound nonspecifically from the embryos.
4. Fix the embryos for 1 h at 4°C in a microdrop of PBS containing 3% parafor-
madehyde.
3.2.4.5. QUANTIFICATION OF MICROSPHERE BINDING

The fibronectin-binding activity is determined by quantifying the fluores-
cence intensity of microspheres bound to the blastocysts using microscopy and
image analysis.
1. Embryos labeled with fluorescent microspheres are viewed directly through the
bottom of a microdrop culture dish using a microscope equipped with
epifluorescence illumination and filters that will provide the appropriate excita-
tion and emission frequencies for the fluorescent microspheres.
03_Armant_35_56_F 49 8/29/05, 11:15 AM

50 Armant
Fig. 5. Determination of fibronectin-binding activity. An image of fluorescent

microspheres bound to a blastocyst (A) is obtained by focusing on its outer edge. Two
methods are shown for quantifying the fluorescence intensity of the bound
microspheres using image analysis software. One method (B) is to trace the outer edge
of the image with the computer mouse, beginning at the arrowhead, and then double
back a short distance inward from the edge until reaching the origin. The enclosed
area, consisting of the in-focus region of the image, is then used by the program to
determine the average grey level. An alternate method (C) is to create a fixed shape
that is placed along the in-focus edge of the image with the computer mouse (gray
squares shown) to obtain an average gray level.
2. Locate an embryo in bright field and then switch to fluorescence illumination.

Bring the fluorescent microspheres at the outer edge of the blastocyst into focus,
which will yield the strongest signal.
3. Adhesiveness to fibronectin tends to localize at the abembryonic pole of the
mouse blastocyst (Fig. 3B), opposite the ICM (18). Rotate the blastocyst so the
ICM is positioned to the center of the image and fluorescence is generally con-
tinuous around the perimeter of the embryo (Fig. 5A).
4. A digital fluorescent image of each embryo is captured using a charge-coupled
device (CCD) camera and stored in the computer.
5. Use an image analysis program to determine the grey level of selected regions of
each image (see Note 16). The outer edge of the blastocyst may be selected by
tracing with a mouse (Fig. 5B). Another technique is to use a “square” or “circle”
tool to randomly sample ten to twenty areas along the embryo perimeter to obtain
an average grey level (Fig. 5C).
3.2.5. Data Analysis and Statistics

All experiments should be repeated at least three times to demonstrate
reproducibility. Record observations at the same times during development
in each replicate experiment so data from all experiments can be pooled for
final analysis. A balance must be struck between obtaining too few measure-
ments to construct precise rates of change and removing embryos from the
03_Armant_35_56_F 50 8/29/05, 11:15 AM

incubator for observation too frequently, which could compromise their viabil-
ity and alter development.
The rates of blastocyst expansion, hatching, attachment, and outgrowth are
all determined by repetitive observation as embryos develop in culture, scoring
the percentage of embryos that have achieved the respective endpoint. Choose
a specific time during development to initiate the experiment, designated as
0 h. Examination of the embryos three to five times daily will usually pro-
vide robust data. A total N of 30 to 50 embryos in each experimental group
usually provides sufficient power for statistical analysis. It is preferable to obtain
overall percentages by pooling data from all experiments rather than to calculate
percentages for each individual experiment and then average the percentages.
Probit analysis, available in the SPSS statistical software package (SPSS Inc.,
Chicago, IL), is used to calculate a T50 (the time when half of the embryos are
scored positively) for each experimental group and generate confidence inter-
vals (CIs) to indicate whether there are statistically significant differences
between groups. The T50 is a measure of developmental rate, where faster
progression results in a lower T50 value. If their 95% CIs do not overlap, T50s
are statistically different (p < 0.05).
Outgrowth area and fibronectin-binding activity are measured once or twice
daily. Values are pooled from all experiments and the data for each experimen-
tal group is subjected to analysis of variance. A total N of 15 to 30 embryos
in each experimental group usually provides sufficient power for statistical
analysis.
4. Notes
1. Fertilization and commencement of embryonic development occur within a few
hours of ovulation, around midnight after pairing mice for mating. Therefore, in
vivo development is estimated to begin at midnight, which is designated E 0.0.
Zygotes can be collected from the oviduct the following morning at E 0.5,
whereas blastocysts are collected from the uterus 3 d later at E 3.5. A different
system is used to track in vitro development after embryos are removed from the
reproductive tract. The day after mating is designated GD 1. For example, if
blastocysts are collected from pregnant dams on E 3.5, development in vitro from
that time forth begins with GD 4. The distinction between developmental time in
vivo and in vitro is important because the rate is highly variable in vitro, depend-
ing on culture conditions, and does not necessarily correspond to the stage attained
at that time in vivo (Table 1).
2. Other mouse strains may be used. Males of F1 strains, such as B6SJLF1/J, are
particularly good studs that successfully mate at rates in excess of 80%.
3. Other mouse strains may be used. Outbred female mice are easily induced to
superovulate. Hogan et al. (35) provide a list of inbred and F1 strains that are
high and low ovulators.
03_Armant_35_56_F 51 8/29/05, 11:15 AM

52 Armant
4. We use Ham’s F-10 medium for blastocyst culture, but many other medium for-
mulations work well. The principal requirements for complete blastocyst differ-
entiation in culture are the inclusion of amino acids (23,25,40). Other complex
media (e.g., CMRL 1066, Dulbecco’s modified Eagle’s medium [DMEM], RPMI
1620) are satisfactory. We have no experience in the production of trophoblast
outgrowths with the newer media developed specifically for preimplantation blas-
tocyst culture. Those supplemented with amino acids, such as KSOM+AA (11),
are likely to work well.
5. M2 is useful for embryo collection because it is a HEPES-buffered medium that
will maintain pH at ambient CO2 levels.
6. Microdrop culture (see Note 10) is performed in a nonhumidified incubator. Hu-
midification causes moisture to collect on the surface of the oil, eventually drip-
ping through the oil and diluting the culture medium.
7. Microspheres are available in various fluorescent colors to match most fluores-
cence detection systems. They are also available conjugated to reactive groups
for covalent coupling to proteins. Alternate commercial sources of this product
include Molecular Probes (Eugene, OR) and Polysciences (Warrington, PA).
8. Mice should be housed in an approved animal care facility with a regulated light-
ing cycle. For the protocol described here, lights were set to turn on at 0700 h and
turn off at 2100 h (14 h light/10 h dark). If blastocyst collection is desired at a
time of day other than morning, the light cycle and injection schedule should be
adjusted accordingly.
9. It is also possible to generate blastocysts by initiating in vitro development at
earlier preimplantation stages, collecting embryos from the oviduct (35). Gener-
ally, embryo production diminishes with gestational age, but earlier initiation of
in vitro development delays blastocyst formation. Removing blastocysts from
the uterus is technically less challenging than flushing oviducts. Both approaches
will generate embryos that can be cultured through the peri-implantation stages.
10. Some comments on the microdrop culture, as described by Hogan et al. (35), are
provided. Drops (2–10 µL) of embryo culture medium are arrayed on a Petri
dish. If a tissue culture-treated plate is used, the drops tend to run together. The
plate is then flooded with water-extracted mineral oil to cover the drops and pre-
vent their evaporation. At the outset, plates should be marked on the underside
with a laboratory pen to keep track of individual embryos or groups that will be
cultured on the plate.
11. Micropipets (see p. 134 of ref. 35) are made by heating a glass Pasteur pipet
about half way up the narrow end over a Bunsen burner. Once the glass is fluid,
move the pipet out of the flame and quickly pull a thin strand of glass, which
should remain hollow. Break it to create an opening that has a diameter 25–50%
larger than a blastocyst. Latex hosing with a mouth piece is attached with a cot-
ton filter to the pipet to control suction. While observing embryos through a ste-
reomicroscope, place the pipet in medium or through oil into a microdrop until it
rests next to an embryo. Controlling the suction, draw the embryo just into the
end of the pipet. Removing the pipet from the medium stops the suction. Once it
03_Armant_35_56_F 52 8/29/05, 11:15 AM

is inserted into another microdrop, create pressure to deliver the embryo. Take
care not to draw the embryo in too far or to expel all of the medium in the pipet,
as bubbles will be produced. Turn the dish to position the drop on the side oppo-
site your hand so the pipet can be held at a shallow angle. With a little practice,
this technique can be quickly mastered. Microspheres of 50–100 µm diameter
are useful for learning the technique without using valuable embryos.
12. The area occupied by the blastocyst prior to outgrowth should be taken into con-
sideration in analyzing outgrowth area. For example, the average area of blasto-
cysts before beginning outgrowth could be determined and subtracted from
measurements obtained during outgrowth to calculate the increase in outgrowth
area. This will provide a more accurate basis for comparing experimental groups.
13. The strong negative charge of α1-acid glycoprotein conferred by multiple sialic
acid moieties reduces nonspecific interactions of the microspheres with nega-
tively charged cell surfaces.
14. Adhesion incompetent blastocysts will produce low levels of fibronectin-binding
activity that are unaffected by prior upregulation with fibronectin or by competi-
tion with soluble fibronectin during incubation with microspheres. By calculat-
ing the difference in fibronectin-binding activity between embryos upregulated
with fibronectin and those exposed only to BSA or embryos assayed in the pres-
ence of competing fibronectin (∆ fibronectin-binding activity), blastocysts from
different stages of development are most accurately assessed for adhesion com-
petence.
15. Prior to adding embryos, it may be necessary to resuspend microspheres that
have settled during the chilling process using a micropipette to mix the medium.
The microsphere suspension will obscure your view of the embryos, making it
difficult to locate them after the 30-min incubation. This problem may be re-
duced by placing the embryos in the shallow portion of the drop, near its edge.
Medium drawn from the top of the drop can be blown over the embryos to assist
in their recovery.
16. For an eight-bit imaging program, grey level will range from 0 (black) to 255
(white). By setting the program to determine the average grey level (not the total
grey level), it will not matter if the area you are tracing over each embryo varies.
Acknowledgments
This research was supported by National Institutes of Health grants HD
36764 and AA12057. The author thanks Dr. Zitao Liu, Dr. Jun Wang and Mr.
Po Jen Chiang for preparing images of mouse embryos that were used for illus-
trations.
References
1. Johnson, M. H. and Ziomek, C. A. (1983) Cell interactions influence the fate of
mouse blastomeres undergoing the transition from the 16- to the 32-cell stage.
Dev. Biol. 95, 211–218.
03_Armant_35_56_F 53 8/29/05, 11:15 AM

54 Armant
2. Watson, A. J. and Barcroft, L. C. (2001) Regulation of blastocyst formation. Front.

Biosci. 6, D708–D730.
3. Gardner, R. L., Papaioannou, V. E., and Barton, S. C. (1973) Origin of the ectopla-
cental cone and secondary giant cells in mouse blastocysts reconstituted from iso-
lated trophoblast and inner cell mass. J. Embryol. Exp. Morphol. 30, 561–572.
4. Gardner, R. L. (1983) Origin and differentiation of extraembryonic tissues in the
mouse. Int. Rev. Exp. Pathol. 24, 63–133.
5. Gardner, R. L. (2000) Flow of cells from polar to mural trophectoderm is polar-
ized in the mouse blastocyst. Hum. Reprod. 15, 694–701.
6. Cross, J. C., Werb, Z., and Fisher, S. J. (1994) Implantation and the placenta: key
pieces of the development puzzle. Science 266, 1508–1518.
7. Carson, D. D., Bagchi, I., Dey, S. K., et al. (2000) Embryo Implantation. Dev.
Biol. 223, 217–237.
8. Armant, D. R., Wang, J., and Liu, Z. (2000) Intracellular signaling in the develop-
ing blastocyst as a consequence of the maternal-embryonic dialogue. Semin.
Reprod. Med. 18, 273–287.
9. Stachecki, J. J., Yelian, F. D., Schultz, J. F., Leach, R. E., and Armant, D. R.
(1994) Blastocyst cavitation is accelerated by ethanol- or ionophore- induced
elevation of intracellular calcium. Biol. Reprod. 50, 1–9.
10. O’Sullivan, C. M., Liu, S. Y., Karpinka, J. B., and Rancourt, D. E. (2002) Embry-
onic hatching enzyme strypsin/ISP1 is expressed with ISP2 in endometrial glands
during implantation. Mol. Reprod. Dev. 62, 328–334.
11. Biggers, J. D., McGinnis, L. K., and Raffin, M. (2000) Amino acids and preim-
plantation development of the mouse in protein-free potassium simplex optimized
medium. Biol. Reprod. 63, 281–293.
12. Kimber, S. J. (2000) Molecular interactions at the maternal-embryonic interface
during the early phase of implantation. Semin. Reprod. Med. 18, 237–253.
13. Blankenship, T. N. and Given, R. L. (1995) Loss of laminin and type IV collagen
in uterine luminal epithelial basement membranes during blastocyst implantation
in the mouse. Anat. Rec. 243, 27–36.
14. Salomon, D. S. and Sherman, M. I. (1975) Implantation and invasiveness of mouse
blastocysts on uterine monolayers. Exp. Cell Res. 90, 261–268.
15. Glass, R. H., Spindle, A. I., and Pedersen, R. A. (1979) Mouse embryo attachment
to substratum and interaction of trophoblast with cultured cells. J. Exp. Zool. 208,
327–336.
16. Smith, S. E., French, M. M., Julian, J., Paria, B. C., Dey, S. K., and Carson, D.
D. (1997) Expression of heparan sulfate proteoglycan (perlecan) in the mouse
blastocyst is regulated during normal and delayed implantation. Dev. Biol. 184,
38–47.
17. Sellens, M. H. and Sherman, M. I. (1980) Effects of culture conditions on the
developmental programme of mouse blastocysts. J. Embryol. Exp. Morphol. 56,
1–22.
18. Schultz, J. F. and Armant, D. R. (1995) Beta1- and beta3-class integrins mediate
fibronectin binding activity at the surface of developing mouse peri-implantation
03_Armant_35_56_F 54 8/29/05, 11:15 AM

blastocysts. Regulation by ligand-induced mobilization of stored receptor. J. Biol.

Chem. 270, 11,522–11,531.
19. Blankenship, T. N. and Given, R. L. (1992) Penetration of the uterine epithelial
basement membrane during blastocyst implantation in the mouse. Anat. Rec. 233,
196–204.
20. Armant, D. R., Kaplan, H. A., and Lennarz, W. J. (1986) Fibronectin and laminin
promote in vitro attachment and outgrowth of mouse blastocysts. Dev. Biol. 116,
519–523.
21. Wordinger, R. J., Brun-Zinkernagel, A. M., and Jackson, T. (1991) An ultrastruc-
tural study of in-vitro interaction of guinea-pig and mouse blastocysts with extra-
cellular matrices. J. Reprod. Fert. 93, 585–597.
22. Armant, D. R. and Kameda, S. (1994) Mouse trophoblast cell invasion of extra-
cellular matrix purified from endometrial tissue: a model for peri-implantation
development. J. Exp. Zool. 269, 146–156.
23. Gwatkin, R. B. (1966) Defined media and development of mammalian eggs in
vitro. Ann. N. Y. Acad. Sci. 139, 79–90.
24. Sherman, M. I. and Atienza-Samols, S. B. (1978) In vitro studies on the surface
adhesiveness of mouse blastocysts, in Human Fertilization (Ludwig, H. and
Tauber, P. F., eds.). Georg Thieme, Stuttgart, Germany: pp. 179–183.
25. Spindle, A. I. and Pedersen, R. A. (1973) Hatching, attachment, and outgrowth of
mouse blastocysts in vitro: fixed nitrogen requirements. J. Exp. Zool. 186, 305–318.
26. Enders, A. C., Chavez, D. J., and Schlafke, S. (1981) Comparison of implantation
in utero and in vitro, in Cellular and Molecular Aspects of Implantation (Glasser,
S. R. and Bullock, D. W., eds.). Plenum, New York, NY: pp. 365–382.
27. Armant, D. R., Kaplan, H. A., Mover, H., and Lennarz, W. J. (1986) The effect of
hexapeptides on attachment and outgrowth of mouse blastocysts cultured in vitro:
evidence for the involvement of the cell recognition tripeptide Arg-Gly-Asp. Proc.
Natl. Acad. Sci. USA 83, 6751–6755.
28. Sutherland, A. E., Calarco, P. G., and Damsky, C. H. (1988) Expression and func-
tion of cell surface extracellular matrix receptors in mouse blastocyst attachment
and outgrowth. J. Cell Biol. 106, 1331–1348.
29. Carson, D. D., Tang, J. P., and Gay, S. (1988) Collagens support embryo attach-
ment and outgrowth in vitro: effects of the Arg-Gly-Asp sequence. Dev. Biol.
127, 368–375.
30. Yelian, F. D., Edgeworth, N. A., Dong, L. J., Chung, A. E., and Armant, D. R.
(1993) Recombinant entactin promotes mouse primary trophoblast cell adhesion
and migration through the Arg-Gly-Asp (RGD) recognition sequence. J. Cell Biol.
121, 923–929.
31. Schultz, J. F., Mayernik, L., Rout, U. K., and Armant, D. R. (1997) Integrin traf-
ficking regulates adhesion to fibronectin during differentiation of mouse peri-im-
plantation blastocysts. Dev. Genet. 21, 31–43.
32. Wang, J., Mayernik, L., and Armant, D. R. (2002) Integrin signaling regulates
blastocyst adhesion to fibronectin at implantation: intracellular calcium transients
and vesicle trafficking in primary trophoblast cells. Dev. Biol. 245, 270–279.
03_Armant_35_56_F 55 8/29/05, 11:15 AM

56 Armant
33. Rout, U. K., Wang, J., Paria, B. C., and Armant, D. R. (2004) alpha5beta1,
alphaVbeta3 and the platelet-associated integrin alphaIIbbeta3 coordinately regu-
late adhesion and migration of differentiating mouse trophoblast cells. Dev. Biol.
268, 135–151.
34. Wang, J. and Armant, D. R. (2002) Integrin-mediated adhesion and signaling dur-
ing blastocyst implantation. Cells Tissues Organs 172, 190–201.
35. Hogan, B. L., Beddington, R. S., Constantini, F., and Lacy, E. (1994) Manipulat-
ing the Mouse Embryo. A Laboratory Manual. Cold Spring Harbor Laboratory,
Cold Spring Harbor, NY.
36. Lopata, A., Kohlman, D. J., Bowes, L. G., and Watkins, W. B. (1995) Culture of
marmoset blastocysts on matrigel: a model of differentiation during the implanta-
tion period. Anat. Rec. 241, 469–486.
37. Kliman, H. J. and Feinberg, R. F. (1990) Human trophoblast-extracellular matrix
(ECM) interactions in vitro: ECM thickness modulates morphology and pro-
teolytic activity. Proc. Natl. Acad. Sci. USA 87, 3057–3061.
38. Yelian, F. D., Yang, Y., Hirata, J. D., Schultz, J. F., and Armant, D. R. (1995)
Molecular interactions between fibronectin and integrins during mouse blastocyst
outgrowth. Mol. Reprod. Dev. 41, 435–448.
39. Armant, D. R. (1991) Cell interactions with laminin and its proteolytic fragments
during outgrowth of mouse primary trophoblast cells. Biol. Reprod. 45, 664–672.
40. Martin, P. M., Sutherland, A. E., and Van Winkle, L. J. (2003) Amino acid trans-
port regulates blastocyst implantation. Biol. Reprod. 69, 1101–1108.
03_Armant_35_56_F 56 8/29/05, 11:15 AM

Stromal Cells In Vitro 57
4
Isolation of Hormone Responsive Uterine Stromal Cells
An In Vitro Model for Stromal Cell Proliferation and Differentiation
Virginia Rider
Summary
The female sex hormones estrogen and progesterone stimulate proliferation and differentia-
tion of human and rodent uterine cells. The purpose of this chapter is to provide a method for
isolating hormone-responsive rat uterine stromal cell lines that can be used to study steroid
control of the cell cycle. Uteri from ovariectomized rats are differentially digested with trypsin
to separate epithelial and stromal cells. The stromal cells are cultured in a standard growth
medium containing 10% fetal bovine serum. After several passages, the purity of the stromal
cell lines is determined using immunocytochemistry. Cell proliferation is studied by culturing
the stromal cells in serum-free medium containing sex steroids and other mitogens. Cell cycle
progression is assessed by flow cytometry, 3H-thymidine and BrdU incorporation, whereas
proliferation is monitored using the MTT assay. Cell cycle regulators are visualized by North-
ern and Western blotting whereas cyclin–cyclin-dependent kinase activity is monitored using
immune complex kinase assays. Uterine stromal cell lines isolated using the methods reported
in this chapter provide a suitable model system to investigate the signal transduction events that
stimulate hormone-dependent control of the cell cycle.
Key Words: Uterus; stromal cells; decidua; cell cycle; estrogen; progesterone.
1. Introduction
The growth and function of any tissue is dependent on regulated prolifera-
tion and differentiation of its cellular components. Within endocrine target tis-
sues of the reproductive tract, hormones exert specific temporal, spatial, and
interactive effects. Estrogens are associated generally with cell proliferation in
uterine and breast tissues, whereas progesterone is considered more as the hor-
mone promoting cellular differentiation in these organs. However, progester-
one is a potent mitogen for stromal cells in the uterus and the lobuloalveolar
cells in the mammary gland (1). The preeminence of progesterone in female
57
04_Rider_57_68_F 57 8/29/05, 11:16 AM

58 Rider
reproduction has been highlighted by studies from mice lacking the progester-
one receptor by targeted mutagenesis (2). The progesterone receptor “knock-
out” mouse exhibits abnormalities in all aspects of reproduction including sexual
behavior, mammary gland development, ovulation, and implantation (2).
The rodent uterus is a well-characterized model system for studying the hor-
monal control of uterine cell proliferation and differentiation (3–8). Under the
influence of estrogen at days 2 and 3 post coitum, the luminal and glandular
epithelial cells proliferate (3,8). At day 4 of pregnancy in the rat, proliferation
switches from epithelial to stromal compartments (5). Stromal cells do not
divide without progesterone, and proliferation is blocked by progesterone
antibodies and progesterone receptor antagonists (8,9). During normal preg-
nancy, the uterine stromal cells proliferate and differentiate (decidualize) into
decidual cells (11). The stromal cells located at the antimesometrial region of
the endometrium are the first to show signs of differentiation. These cells
uncouple DNA replication from cytokinesis and become polyploid (11,12).
The differentiation of the uterine stroma spreads from cells located in the
antimesometrial and periluminal regions outwards towards the myometrium
(13). Decidualization of stromal cells is more restricted to the periluminal
region in the rat compared with the mouse (13).
Because the decidual cells located in the antimesometrial region of the uterus
express different proteins than the mesometrial cells, it is possible that these
two cell populations arise from different progenitors. However, stromal cells
isolated from these two regions and placed in culture lose their differentiated
gene expression (14). This suggests that stromal cells arise from a single stem
cell population and differences in gene expression are due to positional effects
within the endometrium rather than inherent genetic differences. This is an
important concept because it suggests that the mesometrial and antimesometrial
stromal cells share the same lineage, but differences in phenotype occur because
localized effects are exerted on stromal cells depending on their position within
the endometrium. If this interpretation is correct, then it should be possible to
stimulate either the mesometrial or antimesometrial differentiation programs
when the appropriate signal transduction cascades involved have been iden-
tified.
Major progress in understanding the control of the cell cycle has come from
proliferative studies of cells in model culture systems (reviewed in refs. 15–
18). It is our view that conceptual advancement about steroid mediated prolif-
eration and differentiation will be generated using cell culture systems that
recapitulate key hormone-dependent control of cell cycle events. Moreover, it
seems essential to clearly define steroid-dependent effects on the proliferative
cycle vs the differentiation pathway. Although these two events are likely
intertwined, there must be important signals that direct stromal cells from a
04_Rider_57_68_F 58 8/29/05, 11:16 AM

proliferative pathway into the differentiation program. The development of

suitable culture systems to study endocrine-dependent regulation in endome-
trial cells have proved problematic because these target cells often lose their
responsiveness to sex hormones in culture (19). Many of the earlier studies on
this topic (19–21) were concerned with steroid mediated control of epithelial
cell proliferation. Epithelial cells present additional and unique problems in
culture because their functions in vivo depend on cell polarity (apical vs basal
surfaces) as well as junctional connections between the cells.
Our laboratory has been interested in uterine stromal cells (22–29) because
these cells proliferate and differentiate in response to sex steroids to form the
maternal part of the placenta. We anticipated that the structural organization of
stromal cells in vivo could obviate some of the problems encountered with the
epithelial cell culture systems. Stromal cells exist as individual cells surrounded
by the extracellular matrix and do not appear to be constrained by polarity in
the same manner as epithelial cells. Moreover, early reports suggested that
progesterone-dependent proliferation was maintained in cultured human (30)
and rat (31) uterine stromal cells. Uterine stromal cells, isolated by the meth-
ods reported in this chapter, provide a suitable model system to investigate the
signal transduction events that stimulate hormone-dependent proliferation and
ultimately to more fully understand the signals required for differentiation.
2. Materials
2.1. Culture Media
1. Standard growth medium. Medium 199 culture medium supplemented with
Earle’s salts and L-glutamine (Fisher Scientific, Hanover Park, IL) containing
100 U/mL penicillin (Sigma), 100 µg/mL streptomycin (Sigma), and 10%
heat-inactivated fetal bovine serum (FBS; Sigma) (see Note 1). Medium is
stored at 4°C.
2. Serum-free medium. Dulbecco’s modified Eagle’s medium, phenol red-free
(Gibco, Grand Island, NY) in a 3:1 mixture with MCDB-105 (Sigma) containing
supplements including 100 U/mL penicillin (Sigma), 100 µg/mL streptomycin
(Sigma), 5 µg/mL bovine insulin (Sigma), 10 µg/mL human transferrin (Sigma),
50 µg/mL ascorbic acid (Sigma), and 1 mg/mL reagent grade bovine serum albu-
min (Sigma). The medium is sterilized by filtration and stored at 4°C.
3. Freezing medium. Medium 199 culture medium supplemented with Earle’s salts
and L-glutamine (Fisher Scientific, Hanover Park, IL) containing 100 U/mL peni-
cillin (Sigma), 100 µg/mL streptomycin (Sigma), 10% dimethlysulfoxide
(DMSO; Sigma) and 20% heat-inactivated FBS (Sigma). The medium is stored
at 4°C.
4. Trypan blue (0.4%). Trypan blue powder (Sigma) is dissolved in phosphate-buff-
ered saline (PBS) and sterilized by filtration. The solution is stored at 22°C. It is
diluted 1:1 with the cell sample to be counted.
04_Rider_57_68_F 59 8/29/05, 11:16 AM

60 Rider
2.2. Reagents and Hormones

1. Tissue dissociation buffer. PBS without calcium and magnesium containing
0.25% trypsin, 20 mM HEPES, 50 U/mL penicillin (Sigma), 50 µg/mL strepto-
mycin (Sigma). This solution should be made fresh on the day of use.
2. Cell dissociation medium. A trypsin-ethylenediamine tetraacetic acid (EDTA)
solution containing 0.05% trypsin and 0.02% EDTA in Hanks’s balanced salt
solution (1X trypsin-EDTA, Sigma). Hanks’s medium is stored at 4°C.
3. 6α-Methyl-17α-hydroxy-progesterone acetate (Sigma). 10 mM stock made up in
ethanol. Final concentration is 1 µM. Steroids in ethanol are stored at 22°C for
short periods (<2 mo) or at 4°C for longer-term storage (see Note 2).
4. β-Estradiol (Sigma). 100 µM stock made up in ethanol. Final concentration is
10 nM. Hormones in ethanol are stored at 22°C for short periods (<2 mo) or at
4° for longer-term storage.
2.3. Proliferation Assays

1. [ 3H]thymidine: 20–30 Ci/mmol, Amersham Pharmacia Biotech (Arlington
Heights, IL).
2. Nitrocellulose membrane filters, Whatman BA85 (Fairfield, NJ).
3. 5-Bromo-2'-deoxyuridine (BrdU), BrdU Labeling and Detection Kit, Roche
Molecular Biochemicals (Indianapolis, IN).
4. MTT reagent (3-[4,5-dimethylthiazol-2-yl]2,5-diphenyltetrazoliumbromide;
Thiazolyl blue). Dissolve the MTT reagent in PBS to a final concentration of
5 mg/mL. Filter-sterilize and store at –20°C. The working solution is pro-
tected from light and stored at 4°C.
5. Chamber slides. Eight-well Lab-Tek chamber slides (Nunc, Naperville, IL) are
used to visualize BrdU incorporation.
6. Solubilization/stop solution. The stop solution is prepared by dissolving 100 g of
sodium dodecyl sulfate (SDS) in approximately 400 mL of N,N'-dimethyl-
formamide diluted 1:1 in reagent grade water. When the SDS has dissolved, 1 N
HCL (12.5 mL) and 80% acetic acid (12.5 mL) are added to the mixture. The
final volume is adjusted to 500 mL with the dimethylformamide: water (1:1)
mixture.
3. Methods
3.1. Cell Isolation
1. Sexually mature rats (150–175 g body weight) are ovariectomized and rested for
10 ds prior to stromal cell isolation.
2. Uterine horns are removed under anesthesia and trimmed of fat and mesentery.
The tissue is kept on ice.
3. The uterine horns are slit open longitudinally under a dissecting microscope. The
uterine horns from each rat are placed in a 60-mm culture dish (Fisher) contain-
ing 3.0 mL of tissue dissociation medium.
04_Rider_57_68_F 60 8/29/05, 11:16 AM

4. The tissue is incubated at 37°C for 35 min. The dish is vortexed at low speed for 10 s.
5. The tissue dissociation medium is removed and replaced with 2.0 mL of fresh
dissociation medium.
6. The uterine horns are incubated at 37°C for 60 min. The dish is vortexed at low
speed for 10 s.
7. The dissociation medium containing the cells is transferred to a 15-mL sterile
culture tube containing 0.1 vol of FBS. The uteri are washed with dissociation
medium and the wash is added to the sterile tube.
8. The medium containing the dissociated cells is centrifuged at 500g for 5 min.
9. The supernatant is discarded.
10. The cell pellet is suspended in 3.0 mL of standard growth medium.
11. When all of the uteri have been processed, the cell suspensions are combined and
centrifuged at 500g for 5 min.
12. The supernatant is discarded and the cell pellet is suspended in 10 mL of standard
growth medium. The cells are seeded on two 60-mm dishes, 5 mL each.
13. The cells are cultured in standard growth medium in an atmosphere of 5% CO2
and 95% air in a humidified chamber at 37°C (see Note 3).
14. Purity of the cells should be assessed using immunocytochemical analysis with
vimentin, desmin, and cytokeratin antibodies (32) (see Note 4).
15. Estrogen receptor transcripts can be measured using reverse transcription and
polymerase chain amplification (28). Progesterone (25) and estrogen (Fig. 1)
receptor proteins are detected by Western immunoblotting.
16. The response to mitogenic agents, including sex steroids, can be assessed by
flow cytometry (26,27) and the MTT assay (25). Entry into DNA replication is
assessed by 3H-thymidine incorporation (26) and BrdU incorporation (29).
17. Temporal expression of cell cycle regulators can be monitored by Northern and
Western blotting. Activity of cyclin–cyclin-dependent kinases can be monitored
using immune complex kinase assays (27).
3.2. Routine Passaging

1. Stromal cells are washed once with 1X PBS to remove serum.
2. The cells are incubated in 1X trypsin-EDTA at 37°C for 5 min (see Note 5).
3. The released cells are collected with a pipet and immediately added to a 15-mL
sterile tube containing 0.1 vol of FBS. The flask is washed with 1X PBS and the
wash is added to the 15-mL tube.
4. The cells are pelleted by centrifugation at 500g for 5 min.
5. The supernatant is removed and the pellet is suspended in normal growth
medium (see Note 6).
3.3. Freezing and Thawing Cells

1. The cells are released from the T75 flasks using 1X trypsin-EDTA.
2. The cells are suspended in normal growth medium. An aliquot (25 µL) is removed
and counted (see Note 7).
04_Rider_57_68_F 61 8/29/05, 11:16 AM

62 Rider
Fig. 1. Rat uterine stromal cell lines express estrogen receptor (ER)-α protein. Uter-
ine stromal cells (UIII, passage 22) were cultured in medium 199 containing 10% fetal
bovine serum as detailed elsewhere (26). Cell extracts were prepared (27) and size
fractionated by sodium dodecyl sulfate-polyacrylamide gel (10%) electrophoresis. As
a positive control for ER-αexpression, extract from T47D breast cancer cells (34) was
electrophoresed on the same gel. The proteins were transferred to a nitrocellulose
membrane by standard methods (27). The membrane was reacted with a mouse ER-
αantibody (1:500 dilution, AER611, NeoMarkers, Freemont, CA) and an anti-mouse
secondary antibody (1:25,000 dilution). The blot was incubated with the SuperSignal
West Femto Maximum Sensitivity Substrate kit (Pierce, Rockford, IL) and exposed to
X-ray film for 1 min. The size of the major reactive species shown by the arrow at Mr
65,000 was determined from molecular size standards (BioRad, Hercules, CA). This
protein is consistent with the size for ER-α. In the absence of primary antibody the
major reactive protein was not detected (data not shown). Lane 1: Rat uterine stromal
cell extract. Lane 2: T47D breast cancer cell extract. Arrows on the left indicate the
position of the molecular size standards.
3. The cells are centrifuged at 500g for 5 min and the supernatant is carefully
removed from the resulting cell pellet.
4. The cells are suspended in 1 mL of freezing medium and the suspension is placed
into a cryovial.
5. The cryovial is kept at –80°C for at least 8 h before freezing the cells in liquid
nitrogen (see Note 8).
6. To thaw the cells, the cryovial is removed from the liquid nitrogen and quickly
placed in a water bath at 37°C. The thawed cells are transferred into a T75 flask
containing 10 mL of growth medium with 20% FBS (see Note 9).
7. The day after thawing the cells, the medium containing 20% serum is removed
and replaced by normal growth medium (10% FBS).
04_Rider_57_68_F 62 8/29/05, 11:16 AM

3.4. Proliferation Assays

3.4.1. Preparation of the Cells
1. A sufficient number of cells are propagated from the stock (see Note 10).
2. The cells (1–3 × 104) are seeded into 24-well plates in normal growth medium
(see Note 11).
3. The cells are allowed to attach for 2 h. The growth medium is removed and the
cells are washed twice with 1X PBS.
4. Quiescence is induced in the cells by culturing in serum-free medium for 72 h
(see Note 12).
5. The cells are stimulated to re-enter the cell cycle with fresh serum-free medium
containing mitogenic agents or an equal volume of vehicle (negative control).
3.4.2. [3H]Thymidine Incorporation

1. Proliferation agents are added to quiescent cells and plates are incubated for 20 h.
2. The cells are pulsed for 2 h with 1 µCi [3H] thymidine per well.
3. The cells are washed twice with ice-cold 1X PBS.
4. The plates are incubated at –20°C for 60 min to detach the cells.
5. The cells are suspended in trichloroacetic acid (10%). Incorporated [3H] thymi-
dine is separated from unincorporated by retention on nitrocellulose membrane
filters.
6. Filters are washed three times with ice-cold 5% trichloroacetic acid and counted
using liquid scintillation.
3.4.3. BrdU Incorporation

1. Stromal cells (6–7 × 103) are plated into chamber slides and quiescence is in-
duced by culture for 72 h in serum-free medium.
2. Stimulated cells are pulsed for 30 min with BrdU-containing medium (10 µmol/L).
3. The cells are fixed and reacted with anti-BrdU antibodies according to the
manufacturer’s protocol.
3.4.4. MTT Assay
1. Stimulated cells are cultured in 250 µL of serum-free medium containing mito-
genic agents for 48 h.
2. At the end of culture the MTT reagent (25 µL) is added to each well. The cells are
incubated at 37°C from 30 min to 4 h, depending on the number of cells (see
Note 13).
3. Following the incubation period, the reaction is stopped using 750 µL of solubi-
lization/stop solution.
4. The 24-well tray is placed in a humidified chamber at 37°C overnight.
5. The absorbance is determined using a spectrophotometer with wave-length set-
tings of 570 nm, 700 nm, and 570–700 nm.
6. Treatment effects are compared by one-way analysis of variance with appropri-
ate post hoc tests.
04_Rider_57_68_F 63 8/29/05, 11:16 AM

64 Rider
4. Notes
1. We have used heat-inactivated FBS from a variety of distributors. The cells
respond well in all of the FBS samples we have tested. Once the cell lines are
established, the cells grow rapidly and the medium should be changed as the
color shifts from red to orange. Confluent cells should be split and seeded into
new flasks or frozen and kept in liquid nitrogen. We do not maintain cells in
continuous culture because the hormonal responsiveness is better if they are fro-
zen down between experiments. The cells should not be allowed to overgrow.
2. Stromal cells have been stimulated to proliferate with both progesterone (25) and
medroxy progesterone acetate (27). Because the endpoint of the assays (prolif-
eration) occurs within 48 h after stimulation, there is no obvious difference in the
proliferative response to these two hormones.
3. This is the most critical aspect of the isolation procedure. During the first few
days of culture, there will be substantial cell death. As the cells begin to grow, it
is tempting to split and seed them into culture flasks. However, it is important to
maintain the cells for approx 10 d in the original 60-mm dishes or until they are
confluent. If the cells are split too soon, they will die. The cells should be seeded
into a T25 (25 cm2) flask when first split. If the density of the cells is too low
upon transfer, they will die.
4. After the fourth passage, the stromal cell lines will stabilize and growth will be
rapid. The purity of the cells can be assessed using antibodies that distinguish
stromal from epithelial cells. We have not assessed purity prior to the fourth
passage. The advantage of obtaining stromal cell lines over freshly isolated stro-
mal cells is that the cell lines are more homogeneous. Regardless of the care
taken in the isolation procedure, the initial cell isolates will be contaminated with
nonstromal cells (endothelial cells, immune cells, glandular epithelial cells).
During the first weeks after isolation, and for the first few passages, there is
considerable cell death. The cells that remain express stromal cell markers exclu-
sively (25).
5. Stromal cells adhere tightly to the surface of the culture flasks. It generally requires
5 min for the cells to detach but this should be monitored under the microscope.
Cells left in trypsin too long will incur damage to their membranes and die.
6. It is important to split the cells before overgrowth occurs. In general, there will
be between 3–5 × 106 cells per T75 flask. For routine passaging, we split the cells
1:2 and culture in 10 mL of normal growth medium per flask.
7. For freezing the cells, it is optimal to have between 2–5 × 106 total cells. The
cells are suspended in 1 mL of freezing medium. When the cells are thawed, the
frozen cells are split into two T75 flasks (0.5 mL each) in a total volume of 10 mL.
8. The basic procedure is to always freeze the cells slowly and thaw rapidly. Cells
will maintain viability at –80°C for several days. However, they will not be
viable stored at that temperature for long periods of time. Their viability (>90
%) is excellent when stored in liquid nitrogen.
9. The viability of the frozen cells is better when they are cultured initially in me-
dium containing 20%, rather than 10%, FBS. However, this medium is not opti-
04_Rider_57_68_F 64 8/29/05, 11:16 AM

mal for cell growth and maintenance. The medium containing 20% FBS is replaced
with medium containing 10% FBS 24 h after thawing the cells.
10. For each individual experiment a sufficient number of cells are propagated such
that cells within a given experiment are from the same passage.
11. The consistency in cell number in each well for a given experiment is critical for
measuring significant differences among treatments. Plating density is another
crucial factor. If the cells overgrow in the serum-free medium because the initial
plating density is too high they will undergo apoptosis in response to mitogenic
agents. If too few cells are plated, they will die in the serum-free medium. We
find that once the optimal number of cells is determined, the mitogenic response
is consistent, regardless of passage number.
12. With the rat uterine stromal cell lines, the percentage of cells in G1 phase after
72 h of serum-starvation is approx 70%. When cells are serum starved for 48 h,
only about 60% are at G1 phase. The greater the percentage of cells in G1 phase,
at the start of the experiment, the greater the synchronous response to mitogenic
agents.
13. Cell number is critical for determining the amount of time samples are incubated
with the MTT reagent. The MTT assay relies on the conversion of MTT into a
formazan product by the activity of mitochondrial dehydrogenases (33). The
relationship between cell number and absorbance should be determined in pilot
studies. If the cells are incubated with MTT reagent for too long, the assay is not
valid because linearity between cell number and absorbance is lost. Under the
conditions described here for stromal cell proliferation assays we find the 30–
60 min incubation time is optimal.
Acknowledgments
The author would like to thank Oliver Flieger, Marta Piva, Stephanie Jones,
Bruce Kimler and William Justice for contributing to the methods used to isolate
and characterize endocrine-dependent stromal cell proliferation. This research is
supported by the National Science Foundation (NSF IBN-0091504).
References
1. Clarke, C. L. and Sutherland, R. L. (1990) Progestin regulation of cellular prolif-
eration. Endocr. Rev. 11, 266–301.
2. Lyndon, J. P., DeMayo, F. J., Funk, C. R., et al. (1995) Mice lacking progesterone
receptor exhibit pleiotropic reproductive abnormalities. Genes Dev. 9, 2266–2278.
3. Finn, C. A. and Martin, L. (1967) Patterns of cell division in the mouse uterus
during early pregnancy. J. Endocrinol. 39, 593–597.
4. Quarmby, V. E. and Korach, K. S. (1984) The influence of 17β estradiol on pat-
terns of cell division in the uterus. Endocrinology 114, 694–702.
5. Rider, V. and Psychoyos, A. (1994) Inhibition of progesterone receptor func-
tion results in loss of basic fibroblast growth factor expression and stromal cell
proliferation during uterine remodeling in the pregnant rat. J. Endocrinol. 140,
239–249.
04_Rider_57_68_F 65 8/29/05, 11:16 AM

66 Rider
6. Galassi, L. (1968) Autoradiographic study of the decidual cell reaction in the rat.
Dev. Biol. 17, 75–84.
7. Martin, L. and Finn, C. A. (1968) Hormonal regulation of cell division in epithe-
lial and connective tissues of the mouse uterus. J. Endocrinol. 4, 1363–1371.
8. Cullingford, R. W. and Pollard, J. W. (1988) RU486 completely inhibits the
action of progesterone on cell proliferation in the mouse uterus. J. Reprod. Fert.
83, 909–914.
9. Rider, V., Wang, M-Y., Finn, C. and Heap, R. B. (1986) Antifertility effect of pas-
sive immunization against progesterone is influenced by genotype. J. Endocrinol.
108, 117–121.
10. Finn, C. A. (1971) The biology of decidual cells. Adv. Reprod. Physiol. 5, 1–26.
11. Moulton, B. C. and Koenig, B. B. (1984) Uterine deoxyribonucleic acid synthesis
during preimplantation in precursors of stromal cell differentiation during
decidualization. Endocrinology 115, 1203–1307.
12. McConnel, K. N., Sillar, R. G., Young, B. D., and Green, B. (1982) Ploidy and
progesterone-receptor distribution in flow sorted deciduomal nuclei. Mol. Cell.
Endocrinol. 25, 99–104.
13. Krehbiel, R. H. (1937) Cytological studies of the decidual reaction in the rat
during early pregnancy and in the production of deciduomata. Physiol. Zool. 10,
213–241.
14. Gu, Y. and Gibori, G. (1995) Isolation, culture and characterization of the two
cell subpopulations forming the rat decidua: differential gene expression for
activin, follistatin and decidual-related prolactin protein. Endocrinology 136,
2451–2458.
15. Musgrove, E. A. and Sutherland, R. L. (1994) Cell cycle control by steroid hor-
mones. Cancer Biol. 5, 381–389.
16. Pardee, A. B. (1989) G1 events and regulation of cell proliferation. Science 246,
603–608.
17. Sherr, C. J. (1993) Mammalian G1 cyclins. Cell 73, 1059–1065.
18. Mani, S.K., Julian, J., Lampelo, S., and Glasser, S.R. (1992) Initiation and main-
tenance of in vitro decidualization are independent of hormonal sensitization in
vivo. Biol. Reprod. 47, 785–799.
19. Fukamachi, H. and McLachlan, J. A. (1991) Proliferation and differentiation of
mouse uterine epithelial cells in primary serum-free culture: estradiol-17 beta sup-
presses uterine epithelial proliferation cultured on a basement membrane-like sub-
stratum. In Vitro Cell Dev. Biol. 27A, 907–913.
20. Jacobs, L. L., Sehgal, P. B., Julian, J., and Carson D. D. (1992) Secretion and
hormonal regulation of interleukin-6 production by mouse uterine stromal and
polarized epithelial cells cultured in vitro. Endocrinology 131, 1037–1046.
21. Whitworth, C. M., Mulholland, J., Dunn, R. C., and Glasser S. R. (1994) Growth
factor effects on endometrial epithelial cell differentiation and protein synthesis
in vitro. Fertil. Steril. 61, 91–96.
04_Rider_57_68_F 66 8/29/05, 11:16 AM

22. Rider, V., Piva, M., Cohen, M. E., and Carlone, D. L. (1995) Alternative splicing
and differential targeting of fibroblast growth factor receptor 1 in the pregnant rat
uterus. Endocrinology 136, 3137–3145.
23. Rider, V., Carlone, D. L., and Foster, R.T. (1997) Oestrogen and progesterone
control basic fibroblast growth factor messenger RNA in the rat uterus. J.
24. Rider, V., Carlone, D. L., Witrock, D., Cai, C., and Oliver, N. (1992) Uterine
fibronectin content and localization are modulated during implantation. Dev.
Dynamics 195, 1–14.
25. Piva, M., Flieger, O., and Rider, V. (1996) Growth factor control of cultured rat
uterine stromal cell proliferation is progesterone dependent. Biol. Reprod. 55,
1333–1342.
26. Rider, V., Kimler, B. F., and Justice, W. M. (1998) Progesterone-growth factor
interactions in uterine stromal cells. Biol. Reprod. 59, 464–469.
27. Jones, S. R., Kimler, B. F., Justice, W. M., and Rider, V. (2000) Transit of normal
rat uterine stromal cells through G1 phase of the cell cycle requires progesterone-
growth factor interactions. Endocrinology 141, 637–648.
28. Rider, V. (2002) Progesterone and the control of uterine cell proliferation and
differentiation. Front. Biosci. 7, d1545–d1555.
29. Rider, V., Thomson, E., and Seifert, C. (2003) Transit of rat uterine stromal cells
through G1 phase of the cell cycle requires temporal and cell-specific hormone-
dependent changes on cell cycle regulators. Endocrinology 144, 5450–5458.
30. Irwin, J. C., Utian, W. H., and Eckert, R. L. (1991) Sex steroids and growth
factors differentially regulate the growth and differentiation of cultured human
endometrial stromal cells. Endocrinology 129, 2385–2392.
31. Cohen, H., Pageaux, H.-F., Melinand, C., Fayard, J.-M., and Laugier, C. (1993)
Normal rat uterine stromal cells in continuous culture: characterization and
progestin regulation of growth. Eur. J. Cell Biol. 6, 1116–1125.
32. Glasser, S. K., Lampelo, S., Munir, I. M., and Julian, J. (1987) Expression of
desmin, laminin and fibronectin during in situ differentiation (decidualization) of
rat uterine stromal cells. Differentiation 35, 463–474.
33. Mossman, T. (1983) Rapid colorimetric assay for cellular growth and survival: ap-
plication to proliferation and cytotoxicity assays. J. Immunol. Methods 65, 55–63.
34. Edwards, D. P., Kuhnel, B., Estes, P. A., and Nordeen, S.K. (1989) Human proges-
terone receptor binding to mouse mammary tumor virus deoxyribonucleic acid:
dependence on hormone and nonreceptor nuclear factor(s). Mol. Endocrinol. 3,
381–391.
04_Rider_57_68_F 67 8/29/05, 11:16 AM

04_Rider_57_68_F 68 8/29/05, 11:16 AM
Rat Decidual Cell Cultures 69
5
Rat Decidual Cell Cultures
Yan Gu and Geula Gibori
Summary
Pregnancy requires profound reorganization of the different tissues forming the uterus.
Growth and differentiation of the uterine endometrial cells give rise to the decidual tissue, a
transitory organ, which plays a key role in fetal survival. In this chapter, we describe a tech-
nique for the dispersion and the separation of the two different decidual cell subpopulations
with high yield and viability. We also detail a cell culture method, which allows the mainte-
nance of the function and life span of these highly purified decidual cells when cultured either
separately or in a co-culture system.
Key Words: Rat; pseudopregnancy; decidualization; enzymatic tissue dispersion;
antimesometrial and mesometrial decidual cells; cell culture.
1. Introduction
A marked response to implantation and pregnancy in rodents and primates
is the growth and transformation of the uterine endometrial stromal cells known
as decidualization. In humans, decidualization normally occurs with each men-
strual cycle, and the formation of the decidual tissue depends primarily on lev-
els of progesterone and estradiol in the circulation. However, in other species,
including rodents, decidualization requires, in addition to adequate levels of
these hormones, an exogenous trigger, which may be either the contact of the
blastocyst with the endometrium or artificial stimulation at the luminal surface
of uterine horns. Decidualization of the endometrial stroma, induced by either
the blastocyst in pregnant rats or by artificial stimuli in pseudopregnant rats,
gives rise to at least two major cell populations located in opposite sides of the
uterus. The cells that decidualize in the antimesometrial region (opposite to
where blood vessels gain access to the uterus) become more extensively differ-
entiated than the cells in the mesometrial region, which undergo only limited
differentiation. These two decidual cell populations differ not only in their
morphology, but also by the genes they express and the putative roles they play
69
05_Gu_69_78_F 69 8/29/05, 11:16 AM

70 Gu and Gibori
in pregnancy. Moreover, the selective expression of each gene in a defined cell

population is tightly regulated through a cell-to-cell communication between
these two decidual cell subpopulations (1–3).
The antimesometrial cells that decidualize first form what is known as
antimesometrial decidua in pseudopregnant rats and decidua capsularis in preg-
nant animal. The mesometrial decidua forms the decidua basalis, which is the
site of trophoblast invasion. Because the decidual tissue of either pregnant or
pseudopregnant rats is similar in its formation, regression, and secretory
capacity, the pseudopregnant rat has been extensively used as a model to
study this organ in the absence of contaminating trophoblast cells. Because the
antimesometrial decidua is formed primarily by giant-sized, polyploid, and
closely packed cells whereas the mesometrial decidua is formed by much
smaller, loosely packed cells, it is relatively easy to separate these two sub-
populations by the differences in their size and density.
In this chapter, we describe a method for obtaining highly purified subpopu-
lations of rat antimesometrial and mesometrial decidual cells, to facilitate the
study of their function, gene expression, and cell-to-cell communication. We
illustrate in detail the procedures of the induction of decidualization in
pseudopregnant rats, enzymatic dispersion of decidual tissues, separation of
two decidual cells by elutriation, and decidual cell cultures, either in separated
or in co-culture system. We also discuss some common problems one may
encounter in this method and how they might be overcome.
2. Materials
1. Sprague-Dawley rats (Harlan Sprague-Dawley, Madison, WI).
2. Hank’s balanced salt solution (HBSS) without Ca2+ and Mg2+.
3. Fluorescein diacetate (FDA).
4. Trypan blue.
5. Nylon mesh.
6. Oxygen tank.
7. Peristaltic pump (Pharmacia, Peapack, NJ).
8. In-line surge suppressor (Cole Parmer, Chicago, IL).
9. Water-jacketed Cellstir (Wheaton Scientific, Millville, NJ).
10. Elutriator (JE-6B rotor with a Sanderson chamber).
11. 10- or 20-mL syringe.
12. Bubble trap.
13. Three-way connecters.
14. Connecting rubber tubes.
15. Cell Culture Inserts with Cyclopore membrane (Falcon Plasticware, BD Bio-
sciences, Bedford, MA ).
16. Enzyme Dispersion Solution: RPMI-1640 without glutamine containing collage-
nase type I (50 U/mL), dispase (2.4 U/mL), and deoxyribonuclease (200 U/mL).
05_Gu_69_78_F 70 8/29/05, 11:16 AM

17. HBSS Elutriation Buffer: HBSS with 25 mM HEPES and 0.1% bovine serum
albumin (BSA), pH 7.4.
18. Decidual cell culture medium: RPMI-1640 supplemented with 10% fetal bovine
serum (FBS), 2X antibiotic-antimycotic, 1X glutamine, 1X nonessential amino
acids, 1X sodium pyruvate, and 0.5% D-glucose.
3. Methods
3.1. Animals
3.1.1. Pseudopregnancy
The decidual tissue described here is collected from pseudopregnant female rats.
1. Rats are housed in a controlled environmental temperature (22°C) and kept under
a photoperiod of 14 h of light and 10 h of darkness. Rat chow and water are
provided ad libitum.
2. To generate rats pseudopregnant, young Sprague-Dawley female rats at proestrus
are mated with vasectomized males. The day a vaginal plug is found is desig-
nated as day 1 of pseudopregnancy (see Note 1).
3.1.2. Induction of Decidualization

Decidualization of uterine endometrium is artificially induced by a trau-
matic stimulus on day 5 of pseudopregnancy.
1. Under ether anesthesia, a dorsal insertion is made at the low abdominal of a
pseudopregnant female rat and the uterine horns are exposed.
2. The antimesometrial luminal surface of both uterine horns is gently scratched by
a hooked needle inserted at the end of uterine horn next to the ovary.
3. Decidualization can be checked by abdominal palpation a few days after the sur-
gical procedure (see Notes 2 and 3).
3.2. Separation of Decidual Cells

1. After decidualization of the uterine endometrium, the decidual tissue can be col-
lected at different stages of pseudopregnancy, ideally from pseudopregnant ani-
mals between days 8 and 14.
2. Rats are sacrificed by an overdose of ether; uterine horns are dissected out and
washed with ice-cold phosphate-buffered saline (PBS).
3. The dissected uterine horns are cut open along the longitude axis.
4. Decidual tissue can be easily scrapped off from uterine horns and is kept moist
and cold until enzymatic dispersion (see Note 4).
3.2.1. Enzymatic Dispersion

1. The dissected decidual tissue is washed with ice-cold PBS thoroughly to remove
excess blood, pooled, quickly minced to 2- to 3-mm3 pieces on ice, and then
incubated in Enzyme Dispersion Solution with mild stirring in a water-jacketed
05_Gu_69_78_F 71 8/29/05, 11:16 AM

72 Gu and Gibori
Cellstir at 35°C for 45–60 min. Generally, 3–4 g of pooled decidual tissues per
100- to 150-mL enzyme solution should give a complete dispersion. However, in
case of an incomplete dispersion occurring after the first incubation, as a result of
either the presence of excessive tissue or aged enzymes used or other reasons, the
undissolved tissues are allowed to settle for 1 min.
2. The supernatant containing dispersed cells is collected and kept at 4°C.
3. Fresh enzyme solution is added again and the incubation is repeated as described
above.
4. Such dispersed decidual cells are filtered through a nylon mesh to remove any
undissolved tissue debris and centrifuged at 200g for 5 min at 4°C.
5. Cell pellets are gently resuspended in 10 mL HBSS Elutriation Buffer and are
kept in room temperature until elutriation.
3.2.2. Separation of Different Decidual Cells by Elutriation

Antimesometrial and mesometrial decidual cell subpopulations are separated
by elutriation on the basis of cell size difference. The technique described here
is modified from a similar one previously developed for separating luteal cells
(5,6). The elutriation system includes a JE-6B elutriator, whose rotor is fitted
with a Sanderson chamber, a peristaltic pump, an in-line surge suppressor, a
10- or 20-mL syringe, an air bubble trap, several three-way connecters, and
connecting rubber tubes.
3.2.2.1. PREPARATION OF PERISTALTIC PUMP AND ELUTRIATOR
1. Prior to the elutriation, the HBSS elutriation buffer is gassed with 80% oxygen
for 1 h at 22°C with a mild stirring.
2. Meanwhile, the Sanderson chamber is siliconized by adding 1 mL of
dichlorodimethylsilane to the inside of the chamber and allowed to evaporate in a
closed beaker for a few minutes.
3. After the chamber is attached into the elutriator’s rotor, the elutriator, peristaltic
pump, air bubble trap, and all connecting tubes are washed with 1 L of distilled
water, then 500 mL of 70% ethanol or peroxide as instructed by the manufac-
turer, and finally 500 mL of HBSS elutriation buffer.
4. The elutriation system should be calibrated after the system is sterilized. The
settings of peristaltic pump are calibrated according to the required flow rate
(mL/min) and proper speed of the elutriator’s rotor (rpm). An in-line surge sup-
pressor is added into the system to prevent pulsations caused by the peristaltic
pump and to trap any air bubbles in addition to the air bubble trap. Also, a pres-
sure gauge is attached into the system between the peristaltic pump and elutriator
for monitoring the pressure in the elutriator’s chamber.
3.2.2.2. ELUTRIATION PROCEDURES

1. While the elutriation system is running with the setting for the Fraction 1 collec-
tion, dispersed decidual cells suspended in 10 mL HBSS elutriation buffer are
05_Gu_69_78_F 72 8/29/05, 11:16 AM

Table 1
Settings of Peristaltic Pump Flow Rate and Elutriator Speed
Fractions Revolution Pump flow rate Volume
of elutriation of elutriator (rpm) (mL/min) (mL) Resulting contents
1 1600 11 200 Red blood cells, other

blood cells
2 1600 22 200 Mesometrial decidual
cells
3 1600 27 200 Mixed decidual cells
4 1000 25 200 Antimesometrial
decidual cells
slowly injected into the elutriator through the air bubble trap using a 10- or 20-mL
syringe, as shown in the elutriator’s menu. Sometimes a blockage in the elutriation
system can occur during the injection as a result of either a fast injection or over-
load of excessive cells in the system. The blockage can be seen as a sharp and
constant increase of the chamber pressure and visible cell build-up in the cham-
ber. The pressure and cell build-up in the elutriator’s chamber must be monitored
constantly. If an appropriate amount of cells are slowly injected into the elutriation
system, the chamber pressure showing in the pressure gage should not rise.
2. Elutriation is carried out at the room temperature.
3. Four 200-mL fractions are collected using different pump flow rate and revolu-
tion parameters as shown in Table 1.
4. Each fraction is collected in four 50-mL conical centrifuge tubes and contains
different cell subpopulations.
5. As shown in Table 1, highly purified mesometrial decidual cells are collected in
fraction 2, and antimesometrial decidual cells are primarily collected in fraction 4.
6. Cells from the same fractions are pooled together, washed twice with HBSS
elutriation buffer, and finally resuspended in the Decidual Cell Culture Medium.
7. Cell viability is determined by the trypan blue dye exclusion (7) and/or fluores-
cein diacetate staining method (8), approx 80% in general (see Notes 5–7).
3.3. Decidual Cell Culture

3.3.1. Separated Decidual Cell Culture
1. Elutriated antimesometrial or mesometrial decidual cells can be separately plated
at a density about 3 × 106 viable cells per 3 mL culture medium per 5 cm2 into
culture plates or flasks (see Note 8).
2. Cells are incubated at 37°C under an atmosphere of 5% CO2-95% air for 16–18 h
to allow attachment.
3. The cells are washed at least three times with FBS free culture medium prior to
appropriate treatments or experiments (9,10).
05_Gu_69_78_F 73 8/29/05, 11:16 AM

74 Gu and Gibori
Fig. 1 (see companion CD for color version). Elutriated and cultured decidual cells.
Decidual tissues were obtained from day-9 pseudopregnant rats. Antimesometrial and
mesometrial decidual cells were enzymatically dispersed, separated by elutriation, and
stained with fluorescein diacetate in a suspension (upper panel). Upper left: small
mesometrial decidual cells (~10–15 µm in diameter). Upper right: large antimeso-
metrial decidual cells (~30 µm or greater, depending on their differentiation stage).
Elutriated decidual cells were then seeded for 24 h and their distinct morphology shown
in the lower panel. Lower left: mesometrial cells are small and mostly binucleated,
contain less lipid droplets, and remain undifferentiated, with a fibroblast-like appear-
ance in culture. Lower right: antimesometrial decidual cells are large and
polynucleated, with a syncytial-like appearance, and are rich in lipid droplets (red or
dark black dots).
4. At this point, microscopic observation should be performed to check the cell

attachment. Both antimesometrial and mesometrial decidual cells should have
completed the attachment and spread after 16–18 h incubation. After 24 h, the
distinct morphological appearance of each decidual cell type should be clearly
visible in the culture, as shown in Fig. 1.
3.3.2. Decidual Cell Co-Culture System

1. If a co-culture is desired, then one cell type, e.g., antimesometrial decidual
cells (3 × 10 6 viable cells), is seeded onto individual cell culture inserts (an
05_Gu_69_78_F 74 8/29/05, 11:16 AM

Fig. 2. The co-culture system for antimesometrial and mesometrial decidual cells.
Upper panel: antimesometrial cells are seeded onto a cell culture insert and mesome-
trial cells are seeded into a well of six-well culture plate. Lower panel: a reversed
co-culture arrangement of panel A.
approx 5-cm2 area) and another type, e.g., mesometrial decidual cells (3 × 106
viable cells), is plated into wells of an appropriate culture plate (see Note 9).
2. The inserts are then placed into the wells of the plate as shown in Fig. 2, and the
culture medium (~2.5 mL) is added to the wells to reach and maintain a level
equal to that in the insert.
3. The arrangement of cell types in this co-culture and the viable cell density in
each portion of the co-culture should be determined according to the design of
each experiment.
4. Cells should be incubated at 37°C under an atmosphere of 5% CO2-95% air for
16–18 h to allow attachment.
5. The cells are washed at least three times with FBS-free culture medium prior to
appropriate treatments or experiments (9).
6. Microscopic observation should be performed at this point to check the cell
attachment for those cells seeded in the wells.
7. At the end of a culture, decidual cells can be easily scraped from both the inserts
and wells for extraction of proteins or nucleic acids. If it is desired, the cells can
also be dissolved enzymatically or chemically (see Note 10).
4. Notes
1. Pseudopregnancy can be also induced by physical stimulation of the vagina of
rats on estrus using a glass rod. In ovariectomized female rodents, pseudopreg-
nancy can also be induced by a sequential progesterone and estradiol treatment.
However, the decidualization reaction in such induced pseudopregnant females
generally is not as strong as that seen in females mated with infertile males.
2. The decidualization can also be induced by intrauterine injection of oil. How-
ever, this method generally does not induce a strong decidualization reaction in
uterine endometrium as seen by the surgical scratching procedure.
3. There is a narrow window for the induction of decidualization. On day 5 of
pseudopregnancy and/or pregnancy, the uterine endometrium of the female rat
becomes very sensitive to a physical stimulation. Therefore, the surgical proce-
05_Gu_69_78_F 75 8/29/05, 11:16 AM

76 Gu and Gibori
dure should be performed on day 5 of pseudopregnancy. If the induction is per-

formed before or after this window, the degree of decidualization reaction will be
severely reduced. Thus, the accurate day count of pseudopregnancy is a critical
factor. In case a unilateral decidualized uterine horn is desired, one must treat the
other uterine horn with extreme carefulness and gentleness because a rough touch
sometimes can result in a certain degree of decidualization.
4. Decidual tissue is a transitory organ with constant dynamic changes in its ana-
tomical appearance and physiological function. Decidual tissue sometimes can
be collected as early as on day 7 of pseudopregnancy. The decidual reaction
reaches the peak on days 9–10 of the pseudopregnancy. Then the degeneration of
decidual tissue starts and becomes apparent on day 13. On day 15 or after, most
degenerated decidual tissue is liquefied.
5. In general, this elutriation cell separation method gives a highly purified mesome-
trial decidual cell subpopulation because the basis of separation of two cell types is
dependent on the physical size of cells. On the other hand, sometimes mesome-
trial decidual cells form clumps, either as a result of the incomplete enzymatic
dispersion or adhesion, which may have a comparable size as an antimesometrial
decidual cell. Therefore antimesometrial decidual cell population is relatively
easier to get contaminated by those mesometrial cell clumps. Apparently, com-
plete enzymatic dispersion is critical for obtaining a pure antimesometrial cell
population. An alternative dissecting method (see Note 7) can also reduce such
cross-contamination if time permits.
6. As mentioned under Subheading 3.2.2.2., if the appropriate amounts of dispersed
decidual cells are slowly injected into the elutriation system, the chamber pres-
sure should not increase. Another way to monitor the injection speed and/or
amount of injected cells is to keep watching the cell build-up in the chamber
through the observation window in the elutriator as described in the menu. Some-
times, either a fast injection or, especially, the injection of excessive decidual
cells can cause a cell-build-up inside the chamber and consequently results in a
blockage in the chamber, indicated by the sharply increased pressure and visible
cell build-up, i.e., the cells occupied more than half of the chamber. If the block-
age takes place, the elutriator should be stopped and quickly flushed with
elutriation buffer. After collecting the elutriation buffer during the flushing, the
cells can be pelleted by centrifugation at room temperature, re-suspended in 10–
20 mL HBSS buffer, and re-injected into the elutriation system with a slower
injection speed and/or the proper amount cells as described under Subheading
3.2.2.2.
7. To reduce the cross-contamination of the two different types of decidual cells,
especially reduce the contamination of antimesometrial cells from mesometrial
cells (clumps), the antimesometrial and mesometrial decidual tissues can be first
dissected out from each other by cutting uterine horns at both sides along the
middle lines between the center of antimesometrial and mesometrial decidua, as
described previously (11). Then, process the tissue separately by following the
procedures described under Subheadings 3.2.1. and 3.2.2.
05_Gu_69_78_F 76 8/29/05, 11:16 AM

8. The number of yielded cells per gram of decidual tissue varies, depending on the
pseudopregnancy stage of an animal. In general, at the peak of decidual reaction
(day 9 of pseudopregnancy), one gram of decidual tissue can yield approx 3 × 106
antimesometrial and 6 × 106 mesometrial decidual cells.
9. The cell plating density should be considered according to the experimental
design, especially when a co-culture system is employed. As shown in Fig. 1,
the size of an antimesometrial cell is much larger than that of a mesometrial cell,
but on the other hand, the number of antimesometrial cells yield per gram of
decidual tissue is much fewer than mesometrial cells. Therefore, one must decide
whether a same plating density for both cell types should be used or it is neces-
sary. It is also worthy to note that if both cell subpopulations are plated equally
with a high density in culture, it may result in a confluence stage for
antimesometrial cells even just after the attachment period, but not for mesome-
trial cells.
10. Cultured decidual cells also show dynamic changes in their function as seen in
decidual tissue in vivo. However, it is not clear yet whether these functional
changes observed in vitro completely reflect and/or parallel that in vivo. For
example, the expression of prolactin receptor mRNA in decidual mesometrial
cells from day-9 pseudopregnant rats is abundant during the first 24-h culture. It
starts to decline at 48 h, and completely disappears after 72 h in cultured mesome-
trial cells.
References
1. O’Shea, J. D., Kleinfeld, R. G., and Morrow, H. A. (1983) Ultrastructure of
decidualization in the pseudopregnant rat. Am. J. Anat. 166, 271–298.
2. Gibori, G. (1994) The decidual hormones and their role in pregnancy recognition,
in Endocrinology of Embryo-Endometrium Interactions (Glasser, S.R.,
Mulholland, J., and Psychoyos, A., eds.). Plenum, New York: pp.217–221.
3. Gu, Y. and Gibori, G. (1999) Deciduoma, in Encyclopedia of Reproduction
(Knobil, E. and Neill, J.D., eds). Academic, San Diego: pp. 836–842.
4. Fitz, T. A., Mayan, M. H., Sawyer, H. R., and Niswender, G. D. (1982) Character-
ization of two steroidogenic cell types in the ovine corpus luteum. Biol. Reprod.
27, 703–711.
5. Nelson, S. E., McLean, M. P., Jayatilak, P. G., and Gibori, G. (1992) Isolation,
characterization, and culture of cell subpopulation forming the pregnant rat cor-
pus luteum. Endocrinology 130, 954–966.
6. Nelson, S. E. and Gibori, G. (1993) Dispersion, separation and culture of different
cell population of the rat corpus luteum, in Methods in Reproduction Toxicology
(Chapin, R. E. and Heidel, J., eds.). Academic, New York: pp 340–359.
7. Tennant, J. R. (1964) Evaluation of the trypan blue technique for determination of
cell viability. Transplantation 2, 685–694.
8. Rotman, B. and Papermaster, B. W. (1966) Membrane properties of living mam-
malian cells as studies by enzymatic hydrolysis of flourogenic esters. Proc. Natl.
Acad. Sci. USA 55, 134–141.
05_Gu_69_78_F 77 8/29/05, 11:16 AM

78 Gu and Gibori
9. Gu, Y. and Gibori, G. (1995) Isolation, culture, and characterization of the two
cell subpopulations forming the rat decidua: differential gene expression for
activin, follistatin, and decidual prolactin-related protein. Endocrinology 136,
2451–2458.
10. Gu, Y., Soares, M. J., Srivastava, R. K., and Gibori, G. (1994) Expression of
decidual prolactin-related protein in the rat decidua. Endocrinology 135, 1422–
1427.
11. Martel, D., Monier, M. N., Psychoyos, A., and De Feo, V. J. (1984) Estrogen and
progesterone receptors in the endometrium, myometrium and metrial gland of the
rat during the decidualization process. Endocrinology 114, 1627–1634.
05_Gu_69_78_F 78 8/29/05, 11:16 AM

Immortalization of Human Endometrial Cells 79
6
The Immortalization of Human Endometrial Cells
Graciela Krikun, Gil Mor, and Charles Lockwood
Summary
The loss of replicative potential with each cell division has been attributed to the progres-
sive shortening of telomeres. This “mitotic clock” occurs because most normal human cells are
telomerase-negative. Telomerase is a multicomponent enzyme that prevents loss of telomeric
DNA associated with normal cell division. Transfection of cells with vectors expressing the
catalytic subunit of human telomerase (hTERT) is often sufficient for immortalization. In this
article, we will address this approach in the establishment of immortalized endometrial cells
and its value in facilitating in vitro studies.
Key Words: Uterus; endometrium; endometrial cells; stromal cells; glandular epithelial
cells; endothelial cells; immortalization; telomerase.
1. Introduction
Scarcity of human tissue and the inability to passage and maintain cells in
culture for long periods of time makes immortalization of primary cells an
ideal research tool. Unfortunately, the process of immortalization often results
in abnormal karyotypes and aberrant functional characteristics. To avoid the
latter drawback, several laboratories, including our own, have introduced
telomerase into cultured primary cells. Telomerase is a multicomponent
enzyme that comprises a template RNA plus an essential catalytic protein
subunit (human telomerase [hTERT]) (1). This method results in immortaliza-
tion of many target cells by preventing the normal shortening of telomeres
observed in adult somatic cells during mitosis. Telomeres are specialized DNA/
protein structures, which are located at the ends of eukaryotic chromosomes.
They contain tandemly repeated DNA sequences, which have a role in main-
taining the chromosomes during cell division by serving as the templates for
telomerase (2).
79
06_Krikun_79_84_F 79 8/29/05, 11:17 AM

80 Krikun, Mor, and Lockwood
The function of telomerase is to add TTAGGG repeats to telomeres by

reverse-transcribing the RNA template, and thus compensating for the loss
of telomeric DNA associated with normal cell division (3). Indeed, the average
length of telomeres is shortened by 10 to 200 basepairs (bp) per division, result-
ing in senescence and, ultimately, cell death (4). Whereas most normal adult
human cells are telomerase-negative (5), germ cells and immortal cells express
telomerase and maintain telomere length through countless cell divisions (6).
Although introduction of exogenous hTERT expression is sufficient for
immortalization or life span elongation of certain cells, other cells need addi-
tional steps to render them immortalized. Indeed, the successful immortaliza-
tion of endometrial glandular epithelial cells required not only the introduction
of hTERT, but also the inactivation of the Rb/p16 and the p53 pathway (7). By
contrast, myometrial cells were successfully immortalized by hTERT alone
(1). Both the glandular epithelial and myometrial cells were karyotypically
normal, retained their phenotype and showed no indication of cancer-associ-
ated changes (1,7).
Recently, our laboratory demonstrated the ability to immortalize human en-
dometrial stromal cells by transfection with hTERT (8). These cells were
karyotypically, morphologically, and phenotypically similar to the primary
parent cells and were responsive to estradiol and progestin. In this review, we
describe the procedure undertaken for human endometrial stromal cell (HESC)
immortalization.
2. Materials
1. Antibiotic and antimycotic (ABAM; Gibco , Grand Island, NY).
2. ITS+(tm) Premix (Becton-Dickinson/Collaborative Research, Bedford, MA).
3. Glutamine (Gibco).
4. Hank’s balanced salt solution (HBSS) (Gibco).
5. Stripped calf serum (SCS) (Gemini, Woodland, CA).
6. Type I collagenase (Worthington, Lakewood, NJ).
7. Basal medium (BM): Dulbecco’s modified Eagle’s medium (DMEM) + Nutrient
Mixture F-12 HAM (Sigma, St. Louis, MO) supplemented with NaHCO3 (1.2 g/L),
10 mL ABAM/L, 10 mL/L ITS+, 100 mL SCS, and 50 mL/L glutamine adjusted to
pH 7.3 and filtered through a 0.2-µm sieve (see Note 1).
8. 45 mesh stainless-steel sieve (Newark Wire-Cloth Co.).
9. Human hTERT-expressing cell line: pA317 hTERT plus puromycin resistance
(Geron Corp. Menlo Park, CA).
10. Medium for initial growth of pA317 cell line: DMEM (high-glucose with L-gluta-
mine) +10% fetal bovine serum + ABAM.
11. Polybrene (Sequabrene, Sigma).
12. TRAPeze enzyme-linked immunosorbent assay (ELISA) Detection Kit
(Chemicon International, Inc., Temecula, CA)
06_Krikun_79_84_F 80 8/29/05, 11:17 AM

3. Methods
3.1. Human Endometrial Stromal Cell Isolation and Cell Culture
After obtaining written informed consent and institutional approval, early
secretory endometria from reproductive age women are obtained from hyster-
ectomies for benign conditions (e.g., myomas). The endometrium is collected
and transported in BM to a sterile laminar flow hood and processed as follows:
1. The tissue is washed with HBSS. After most of the blood has been removed, the
tissue is finely minced in 10 mL HBSS and spun for 10 min at 4°C at 500g.
2. Approximately 300 mg of the wet pellet is resuspended in 10 mL of Hams F10 +
10% SCS containing 0.25% type I collagenase for 90 min in a vigorously shaking
water bath at 37°C (see Note 2).
3. In a sterile laminar hood, the digestate is filtered through a 45 mesh stainless
steel sieve to remove the glands. The supernatant is spun as described above and
the pellet is resuspended in DMEM + 10% SCS.
4. Cells are seeded in polystyrene plastic cell culture flasks. After 40 min, the
medium is changed to remove any floating material.
5. The HESCs are grown to confluence in BM+ at 37°C in a standard humidified
95% air/5% CO2 incubator.
3.2. Immortalization Protocol

Immortalization of primary HESCs derived from the mid-secretory phase is
achieved by transfection of telomerase using a retroviral system. This system
employs the pA317 cell line expressing hTERT and puromycin resistance.
3.2.1. Preparation of Supernatants for Transfection
1. pA317 hTERT are seeded in DMEM and grown to log phase but not allowed to
reach 90% confluence. At this point, the medium is changed with a minimal
amount of fresh medium (6 mL for each T75 flask) overnight (see Note 3).
2. The medium is harvested the following morning and replaced with a minimal
amount of fresh medium, which is collected at the end of the day.
3. The collected media is filtered through a 0.45-µm filter to remove pA317 cells or
any cellular debris and add Polybrene (4 µg/mL). This is the medium, which will
be used to transfect the target cells.
3.2.2. Transfection of Target Cells (HESCs) With the Viral Supernatant

1. Check that the target cells (HESCs) are in good log phase growth (they must be
dividing at the time of transfection).
2. Prior to the day of the transfection, add fresh BM to the HESCs.
3. On the day of transfection, remove the medium and replace it with the harvested
viral supernatant (above) containing Polybrene.
4. Place in a standard cell culture incubator at 37°C for eight h.
5. Remove medium, repeat transfection procedure and leave in incubator overnight.
06_Krikun_79_84_F 81 8/29/05, 11:17 AM

82 Krikun, Mor, and Lockwood
6. The next day, discard the transfection medium and add BM.
7. After a 48 h incubation, trypsinize the cells and culture them under selection
media (BM plus puromycin [800 ng/mL]) (see Note 4).
3.3. Telomerase Detection

Telomerase activity is assayed using the TRAPeze ELISA Detection Kit per
manufacturer’s instructions.
1. HESCs are lysed and incubated with a biotinylated substrate oligonucleotide
(included in the kit), which allows the hTERT subunit to add telomeric repeats
for 30 min.
2. The products are amplified by polymerase chain reaction (PCR) with a
biotinylated primer and DNP-labeled dCTP (included in the kit).
3. The resulting tagged PCR products are immobilized onto streptavidin-coated
microtiter plates via biotin–streptavidin interactions, and then detected by anti-
DNP antibody conjugated to horseradish peroxidase.
4. The amount of product is determined following incubation with 3,3',5,5'-
tetramethylbenzidine by determining absorbance at 450 and 595 nm.
5. Telomerase activity is determined as the difference between the two absorbance
readings (Abs450–Abs595).
6. Positive and negative controls are provided in the kit.
3.4. Freezing the Immortalized Cells

1. Cells are washed twice with phosphate-buffered saline (PBS), trypsinized, (for
2 mL of trypsinized cells, 5 mL of complete medium is added to neutralize).
2. The dispersed cells are collected in a conical tube and an additional 5 mL of
media are added to rinse the flask.
3. After centrifuging (800g, 5 min, 4°C), the cell pellet is collected, resuspended in
BM + puromycin containing 10% dimethyl sulfoxide (DMSO) (very slowly),
aliquoted and frozen (–20°C for 4–5 h, then –80°C overnight, then in liquid N2).
3.5. Thawing and Growing the Immortalized Cells

1. Cells are first thawed on ice.
2. When the cells are thawed they are added to a T75 culture flask and approx 12 mL
of media + puromycin added.
3. After 4–5 h the medium is changed.
4. When confluent, the experimental conditions are carried out without puromycin
in the medium of choice.
4. Notes
1. To avoid nonspecific estrogenic effects, DMEM must not contain phenol red.
2. During the 90 min collagenase digestion, the tubes are vortexed vigorously to
break up the clumped fragments.
06_Krikun_79_84_F 82 8/29/05, 11:17 AM

3. Passage 1:10 when the flask reaches 80–90% confluence using Trypsin-ethyl-
enediamine tetraacetic acid (EDTA); do not allow the cells to become over
confluent.
4. Typical doses of puromycin used for selection are between 0.5 and 2.5 µg/mL.
Because each cell has a different sensitivity to puromycin, testing the sensitivity
prior to the transfection process is recommended. For this, treat the cells with
increasing concentrations of puromycin for 24 or 48 h and determine cell viabil-
ity by trypan blue, MTT, or any other cell viability assay.
Acknowledgments
We would like to acknowledge Mizanur Rahman, MD, Rebeca Caze, MS,
Ayesha Alvero, MD., Seth Guller, PhD, Frederick Schatz, PhD, Eva Sapi, PhD,
Paula Aldo, and Mazin Qumsiyeh, MD for their input in this project. This work
was supported in part by grants from the National Institutes of Health: RO1
HD33937-06 (CJL) and RO1 HL70004-01A1 (CJL).
References
1. Condon, J., Yin, S., Mayhew, B., et al. (2002) Telomerase immortalization of
human myometrial cells. Biol. Reprod. 7, 506–514.
2. Emrich, T., Chang, S.-Y., Karl, G., Panzinger, B., and Santini, C. (2002) Quanti-
tative detection of telomerase components by real-time, online RT-PCR analysis
with the LightCycler, in Methods in Molecular Biology, Vol. 191: Telomeres and
Telomerases: Methods and Protocols (Double, J. A. and Thompson, M. J., eds.).
Humana, Totowa, NJ: pp. 99–108.
3. Toouli, C. D., Huschtscha, L. I., Neumann, A. A., et al. (2002) Comparison of
human mammary epithelial cells immortalized by simian virus 40 T-Antigen or
by the telomerase catalytic subunit. Oncogene 21, 128–139.
4. Chiu, C. P. and Harley, C. B. (1997) Replicative senescence and cell immortality:
the role of telomeres and telomerase. Proc. Soc. Exp. Biol. Med. 214, 99–106.
5. Weinrich, S. L., Pruzan, R., Ma, L., et al. (1997). Reconstitution of human
telomerase with the template RNA component hTR and the catalytic protein sub-
unit hTRT. Nat. Genet. 17, 498–502
6. Bibby, M. C. (2002) Introduction to telomeres and telomerase, in Methods in
Molecular Biology, Vol. 191: Telomeres and Telomerases: Methods and Proto-
cols (Double, J. A. and Thompson, M. J., eds.). Humana, Totowa, NJ: pp. 1–12.
7. Kyo, S., Nakamura, M., Kiyono, T., et al. (2003) Successful immortalization of
endometrial glandular cells with normal structural and functional characteristics.
Am. J. Pathol. 163, 2259–2269.
8. Krikun, G., Mor, G., Alvero, A., et al.. (2004) A novel immortalized human en-
dometrial stromal cell line with normal progestational response. Endocrinology
145, 2291–2296.
06_Krikun_79_84_F 83 8/29/05, 11:17 AM

06_Krikun_79_84_F 84 8/29/05, 11:17 AM
Sheep UGKO Model 85
7
Sheep Uterine Gland Knockout (UGKO) Model
Thomas E. Spencer and C. Allison Gray
Summary
Endometrial gland development is a postnatal event in the ovine uterus that can be inhibited
epigenetically by chronic exposure of ewe lambs to a synthetic progestin after birth. The uterus
of neonatally progestinized ewes lack endometrial glands and display a uterine gland knockout
(UGKO) phenotype. Progestin ablation of endometrial gland development is specific, because
it does not affect development of extra-uterine reproductive tract structures or the hypotha-
lamic–pituitary–ovarian axis. The UGKO ewe is a useful model for study of uterine develop-
ment and the role of endometrial glands in uterine function during the estrous cycle and
pregnancy. UGKO ewes exhibit altered estrous cycles due to the inability of the uterus to pro-
duce luteolytic pulses of prostaglandin F2α. UGKO ewes are infertile, and blastocysts hatch
normally but fail to survive or elongate during early pregnancy. This pregnancy defect is pri-
marily due to the absence of endometrial glands and their secretions rather than alterations in
expression of either anti-adhesive or adhesive molecules on the endometrial epithelium.
Genomics and proteomics are being used to identify specific components of histotroph that are
absent or diminished in the UGKO ewe and will serve as markers of endometrial function and
uterine receptivity.
Key Words: Sheep; uterus; endometrium; gland; histotroph; epigenetic; steroid; progester-
one; neonate; pregnancy; implantation; conceptus; defect; embryo loss.
1. Introduction
The bicornuate ovine uterus consists of two uterine horns connected by a
short uterine body. The uterine wall can be divided functionally into the
endometrium and myometrium. The adult endometrium of ruminants
(sheep, cattle, and goats) consists of two epithelial cell types (luminal epithe-
lium [LE] and glandular epithelium [GE]), stratified stromal compartments that
include a densely organized adluminal zone of fibroblastic cells (stratum
compactum) extending into a more loosely organized zone in the deeper or
basal endometrium (stratum spongiosum), blood vessels, and immune cells.
Grossly, the adult ovine endometrium is divided into raised, aglandular
85
07_Spencer_85_94_F 85 8/29/05, 11:17 AM

86 Spencer and Gray
caruncular and intensely glandular intercaruncular areas (1). The caruncular

areas have LE and compact stroma, and are the sites of implantation and pla-
centation (2). Synepitheliochorial placentation in sheep involves the fusion of
placental cotyledons with endometrial caruncles to form placentomes, which
support fetal–maternal gas exchange and placental nutrient transport.
Intercaruncular endometrial areas contain large numbers of uterine glands that
synthesize and secrete a complex array of proteins and related substances,
termed histotroph, into the uterine lumen (3). Epithelial products in this cat-
egory include a wide variety of enzymes, growth factors, cytokines, lymphok-
ines, hormones, transport proteins, and other substances (4,5). Uterine
epithelial secretions are thought to influence conceptus survival and develop-
ment, onset of pregnancy recognition signals, and growth of both the placenta
and fetus in humans, other primates, domesticated animals, and laboratory ani-
mals (3,4,6–12).
Endometrial glands are characteristic features of all mammalian uteri.
Endometrial gland development (adenogenesis) in domesticated and labo-
ratory animals occurs rapidly after birth (9,13). Withdrawal of fetal tissues
from a progesterone-dominated prenatal environment at birth was proposed to
be an endocrine cue for adenogenesis in the neonatal ovine uterus (14). Subse-
quently, Frank F. Bartol and his colleagues at Auburn University, Alabama
(15) demonstrated that exposure of ewe lambs to the synthetic progestin from
birth to postnatal day (PND) 13 inhibited endometrial adenogenesis. Removal
of the progestin block to adenogenesis on PND 13 permitted glands to develop
by PND 26. However, these glands were not well developed and were histo-
logically abnormal. This original observation served as the foundation for the
hypothesis that prolonged exposure of neonatal ewes to a progestin during the
entire critical period of endometrial adenogenesis could be used as a tool to
permanently inhibit endometrial gland differentiation, thereby producing a
uterine gland knockout (UGKO) phenotype in the adult (9,13).
Indeed, chronic exposure of neonatal lambs to progestins for 8, 16, or 32 wk
from birth was shown to prevent uterine adenogenesis and induce a unique,
stable adult endometrial phenotype characterized by the histological absence
of uterine glands (13,16,17). Interestingly, exposure of neonatal ewes to a
progestin does not affect development or function of the brain, hypothalamic–
pituitary–ovarian axis, ovary, or Müllerian duct-derived reproductive tract tis-
sues, including the oviduct, cervix or vagina (18,19). However, the uteri of
UGKO ewes weighs less and has shorter horns (18). As illustrated in Fig. 1,
the UGKO uterine wall is essentially devoid of endometrial glands and lacks
stromal delineation characteristic of intercaruncular endometrium in normal
ewes (18). Consistent with a reduction in uterine size and weight, endometrial
width, area and luminal epithelial length are decreased in the UGKO uterus,
07_Spencer_85_94_F 86 8/29/05, 11:17 AM

Sheep UGKO Model 87
Fig. 1. Histological comparison of day-14 pregnant uteri from control and uterine
gland knockout (UGKO) ewes. The uteri were fixed in paraformaldehyde, sectioned
(5 µm), and stained with hematoxylin and eosin. (A) Control uterus that contained two
normal day 14 conceptuses. (B) UGKO uterus that failed to support conceptus devel-
opment. (C) UGKO uterus that contained only a single tubular conceptus. (D) UGKO
uterus that contained a single fragile filamentous conceptus. The UGKO phenotype
was produced by exposing neonatal ewe lambs to a 19-norprogestin from birth for 8
wk. Legend: L, lumenal epithelium; G, glandular epithelium; S, stroma; M, myo-
metrium (original magnification ×166).
whereas myometrial width and morphology are not different from normal ewes
(18). The specific targeting of only the uterine endometrium by the progestin
exposure makes it an attractive model with which to study mechanisms regu-
lating endometrial organization and gland morphogenesis, also termed
adenogenesis, in the neonate, as well as the functional role of endometrial
glands in the adult.
Recent studies of the UGKO ewe model revealed an essential role for
endometrial glands and their secretions in normal estrous cycles and in peri-
implantation conceptus survival and growth. Mature UGKO ewes are unable
to exhibit normal estrous cycles as a result of insufficient production of
07_Spencer_85_94_F 87 8/29/05, 11:17 AM

88 Spencer and Gray
luteolytic pulses of prostaglandin F2α (PGF2α) by the uterus (17). Luteolytic

pulses of PGF2α are produced by oxytocin, from the posterior pituitary and/or
corpus luteum, binding to oxytocin receptors expressed by the endometrial lu-
minal and glandular epithelia (20). Oxytocin receptor gene expression in the
ovine uterus is predominantly controlled by ovarian steroid hormones, estro-
gen and progesterone, and their receptors in the endometrial epithelium (20).
No differences in expression of receptors for estrogen, progesterone, or oxyto-
cin were observed in the UGKO uterus (17). The inability of the UGKO uterus
to produce luteolytic pulses of PGF2α was hypothesized to result from the lack
of superficial or ductal GE, coupled with an overall reduction in LE surface
area, that reduced the numbers of oxytocin receptors that could respond to oxy-
tocin (17,20). Nonetheless, exogenous PGF2α induces luteolysis in UGKO
ewes, and they display normal estrus mating behavior (17).
Adult UGKO ewes are unable to establish pregnancy, despite repeated
matings to rams of proven fertility (17,18,21,22). Transfer of normal, hatched
blastocysts into the uteri of timed recipient UGKO ewes failed to ameliorate
this defect and to establish pregnancy (21). Morphologically normal blasto-
cysts are present in uterine flushes of bred UGKO ewes on day 6 or 9 post
mating, but not on day 14 (21,22). On day 14, uterine flushes of mated UGKO
ewes contain either no conceptus or a severely growth-retarded tubular con-
ceptus (21,22). The peri-implantation period of pregnancy in sheep is marked
by rapid elongation of the conceptus from a tubular to filamentous form
between days 11 and 16 and production of interferon (IFN)-τ, the signal for
maternal recognition of pregnancy (20). Although the growth-retarded con-
ceptuses recovered from mated UGKO ewes produced little or no IFN-τ, the
endometrium of UGKO ewes responded appropriately to intrauterine infusions
of recombinant ovine IFN-τ with increased expression of IFN-τ-stimulated
genes (22). These results supported the hypothesis that the inability of the
UGKO uterus to support peri-implantation conceptus survival and growth was
primarily due to an absence of histotroph derived from endometrial gland
secretions.
Implantation in ruminants is a highly coordinated process that involves appo-
sition, adhesion, and attachment of the conceptus trophectoderm to luminal epi-
thelium (23). In sheep, the blastocyst enters the uterus on day 4 and hatches
from the zona pellucida on day 7. Apposition of conceptus trophectoderm and
luminal epithelium is initiated between days 10 and 14, followed by adhesion
on day 15 and attachment within days 16 to 18. Elongation of spherical blasto-
cysts to a filamentous form is thought to require transient attachment and adhe-
sion of conceptus trophectoderm to luminal epithelium. Initially, the
nonadhesive property of the luminal epithelium appears to be partially due to
apical expression of mucins, such as mucin glycoprotein one (Muc-1), that
07_Spencer_85_94_F 88 8/29/05, 11:17 AM

Sheep UGKO Model 89
sterically impair interactions between trophectoderm and adhesive glycopro-

teins, such as integrins, as a result of their extensive glycosylation and extended
extracellular structure (24). Immunoreactive Muc-1 expression by luminal epi-
thelium decreases between days 9 and 17 of early pregnancy in normal (24)
and UGKO (22) ewes. Extracellular matrix and integrins are thought to be
responsible for trophectoderm attachment and adhesion to LE (25,26). During
the peri-implantation period of pregnancy in ewes, integrin subunits αv, α4,
α5, β1, β3, and β5 are constitutively expressed on both conceptus trophecto-
derm and the apical surface of luminal epithelium (24). Integrin expression on
endometrial luminal epithelium of UGKO ewes is not different from normal
ewes (22). Furthermore, expression of receptors for estrogen (estrogen recep-
tor [ER]α), progesterone (progesterone receptor [PR]), and oxytocin (oxytocin
receptor [OTR]), as well as several LE-specific genes, does not differ between
the endometrium of UGKO and that of normal ewes (17,22). Thus, by these
measures, the endometrial LE does not appear to be defective in UGKO ewes.
Uterine flushes of UGKO ewes were analyzed for the presence of
osteopontin (OPN) and glycosylated cell adhesion molecule one (GlyCAM-1)
proteins, which are expressed by the endometrial glands of the ovine uterus
and are suggested to play a role in regulation of conceptus implantation (27,28).
Uterine flushes of day-14 bred UGKO ewes contained lower amounts of
GlyCAM-1 and no OPN compared with day-14 pregnant ewes (22). These
results were expected, because GlyCAM-1 is expressed by both the endome-
trial LE and GE (28), whereas OPN is expressed solely by the endometrial
glands (29). A general marker of endometrial gland differentiation in the devel-
oping postnatal and adult uterus is the prolactin receptor (30–33), because it is
exclusively expressed in the endometrial GE, but not in the LE or superficial
ductal GE. During pregnancy, the endometrial glands undergo a program of
hyperplasia and hypertrophy that is accompanied by the onset of OPN and
uterine milk protein (UTMP; also named ovine uterine serpin or OvUS) gene
expression (29,30,34). The expression of the UTMP gene disappears in the
endometrial glands immediately after parturition, suggesting that parturition
terminates the terminal differentiation program of pregnancy (20,35).
Genomics and proteomics are being used to identify specific components of
histotroph that are absent or diminished in the UGKO ewe, thereby causing the
peri-implantation defect in conceptus survival and growth (16,22). A better
understanding of the components of histotroph may lead to the development of
better maturation medium for in vitro production of embryos. In addition, these
important histotroph components will serve as useful markers of endometrial
function and fertility in both domestic animals and humans. Therefore, the
UGKO ewe is an attractive model with which to study mechanisms regulating
endometrial organization and adenogenesis in the neonate, as well as the func-
tional role of endometrial glands in adult ewes.
07_Spencer_85_94_F 89 8/29/05, 11:17 AM

90 Spencer and Gray
2. Materials
2.1. Animals
Newborn female sheep (Ovis aries).
2.2. Hormone Treatments

1. Synchromate B® norgestomet implant that releases approx 6 mg of norgestomet
(Nor; 17α-acetoxy-11β-methyl-19-norpreg-4-ene-3,20 dione), a potent synthetic
19-norprogestin, over a 14-d period (Rhone Merieux, Athens, GA). This product
may not be currently available.
2. Medroxyprogesterone acetate (MPA) biodegradable pellets that release approx
200 mg of MPA over a 60-d period (Innovative Research of America, Boca Raton,
FL). This product will substitute for norgestomet (C. A. Gray and T. E. Spencer,
unpublished results).
3. Estradiol-17β valerate or estradiol-17β benzoate may also be used in the neona-
tal ewe lamb to produce the UGKO phenotype (33). However, the adult pheno-
type of neonatally estrogenized ewes has not been investigated.
3. Methods
1. The skin in the area of implant administration is sheared and disinfected with
betadine followed by an alcohol scrub.
2. A sterile scalpel blade is used to make a small incision through the skin in the
periscapular area.
3. The progestin implant is inserted subcutaneously in the periscapular area of ewe
lambs within 12 h of birth. The progestin inhibits differentiation and develop-
ment of the endometrial glandular epithelial cells from the luminal epithelium in
the uterus of the neonatal ewe. Progestin ablation of endometrial adenogenesis is
permanent and produces the UGKO phenotype. Any delay in progestin implant
administration may not be 100% effective to inhibit endometrial gland differ-
entiation. Unpublished observations indicate that administration of the proges-
tin implants on day 7 after birth does not inhibit endometrial gland development
(C. A. Gray and T. E. Spencer, unpublished observations).
4. If the Synchromate B implant is used, a new implant must be administered to the
ewes on PND 14, 28, and 42 to ensure exposure to the progestin from birth to at
least PND 56 (see Note 1).
5. Implanted ewe lambs are maintained according to standard animal husbandry
practices until they reach puberty.
6. After puberty, the UGKO ewes will exhibit altered estrous cycles of 17 to 43 d in
length. Luteolysis and behavioral estrus can be induced in UGKO ewes using
exogenous PGF2α (Lutalyse®, Kalamazoo, MI).
7. The uterus should be removed and assessed to confirm the absence of endome-
trial glands (see Note 2). A general marker of endometrial glandular epithelium
in the ovine uterus is expression of the long and short forms of the prolactin
07_Spencer_85_94_F 90 8/29/05, 11:17 AM

Sheep UGKO Model 91
receptor gene (30,31). The mRNA for the prolactin receptor is expressed only in
the endometrial glandular epithelium of the uterus of postnatal, cyclic, pregnant,
and postpartum ewes (30,31,35). During early pregnancy, expression of the
progesterone receptor is lost from the endometrial glands between days 11 and
17 post mating (day 0 = mating) (36). Loss of the PR is accompanied by the onset
of OPN and UTMP gene expression in the endometrial glandular epithelium
(29,30). The expression of OPN and UTMP genes are indicative of terminal dif-
ferentiation of the endometrial glands of pregnancy, because their expression is
abrogated after parturition in the postpartum uterus (35). Interestingly, the termi-
nally differentiated endometrial glands of pregnancy are devoid of detectable PR
gene expression (37), but the PR returns in the endometrial glands after parturi-
tion, concomitant with the loss of OPN and UTMP gene expression (35).
4. Notes
1. Exposure of neonatal ewes to a progestin for 8, 16, or 32 weeks prevented endome-
trial adenogenesis and produced the UGKO phenotype in adult ewes (17).
2. Three endometrial phenotypes are consistently observed in norgestomet-treated
ewes: (1) no glands; (2) slight glandular invaginations into the stroma; and (3)
limited numbers of cyst- or glandlike structures in the stroma (17) (see Fig. 1).
Most neonatally progestinized ewes exhibit the first phenotype as adults. The
uterus of individual sheep only exhibit one of the phenotypes, because the pheno-
type is homogenous within a horn. The different endometrial phenotypes do not
appear to result from genetic differences in responsiveness to neonatal progestins,
but rather are due to the timing of progestin implant administration. The implant
must be administered very soon after birth to inhibit the program of endometrial
gland differentiation and development. Another potential cause of the differing
endometrial phenotypes is infection at the site of implant administration, which
decreases systemic delivery of the progestin. No specific differences in cyclicity
have been observed in ewes with one of the endometrial phenotypes compared
with the other phenotypes. However, ewes whose uterus exhibits limited num-
bers of cyst- or glandlike structures may possess the ability to nurture the con-
ceptus to a more advanced stage of development (e.g., fragile filamentous) as
compared to ewes with the other phenotypes (e.g., no conceptus or tubular,
growth-retarded conceptus) (22). Regardless of the endometrial phenotypes, all
three types of UGKO ewes are not capable of sustaining pregnancy much past
day 14 and exhibit a peri-implantation type of pregnancy defect (21,22).
Acknowledgments
The authors would like to thank Frank F. Bartol, Fuller W. Bazer, and Kristin
M. Taylor, as well as former graduate students in the Bazer/Spencer Labora-
tory, for their participation in the development of the sheep UGKO model.
This work was supported by grants from the United States Department of
Agriculture (USDA) (98-35203-6322 & 2001-02259).
07_Spencer_85_94_F 91 8/29/05, 11:17 AM

92 Spencer and Gray
References
1. Wimsatt, W. A. (1950) New histological observations on the placenta of the sheep.
Am. J. Anat. 87, 391–436.
2. Amoroso, E. C. (1952) Placentation, in Marshall’s Physiology of Reproduction,
Vol. 2 (Parkes, A. S., ed.). Little Brown, Boston: pp. 127–311.
3. Bazer, F. W. (1975) Uterine protein secretions: relationship to development of the
conceptus. J. Anim. Sci. 41, 1376–1382.
4. Martal, J., Chene, N., Camous, S., et al. (1997) Recent developments and potenti-
alities for reducing embryo mortality in ruminants: the role of IFN-tau and other
cytokines in early pregnancy. Reprod. Fertil. Dev. 9, 355–380.
5. Kane, M. T., Morgan, P. M., and Coonan, C. (1997) Peptide growth factors and
preimplantation development. Hum. Reprod. Update 3, 137–157.
6. Bell, S. C. (1988) Secretory endometrial/decidual proteins and their function in
early pregnancy. J. Reprod. Fertil. Suppl. 36, 109–125.
7. Beier, H. M. (2000) The discovery of uteroglobin and its significance for repro-
ductive biology and endocrinology. Ann. N.Y. Acad. Sci. 923, 9–24.
8. Carson, D. D., Bagchi, I., Dey, S. K., et al. (2000) Embryo implantation. Dev.
Biol. 223, 217–237.
9. Gray, C. A., Bartol, F. F., Tarleton, B. J., et al. (2001) Developmental biology of
uterine glands. Biol. Reprod. 65, 1311–1323.
10. Burton, G. J., Watson, A. L., Hempstock, J., Skepper, J. N., and Jauniaux, E.
(2002) Uterine glands provide histiotrophic nutrition for the human fetus during
the first trimester of pregnancy. J. Clin. Endocrinol. Metabol. 87, 2954–2959.
11. Roberts, R. M. and Bazer, F. W. (1988) The functions of uterine secretions. J. Reprod.
Fertil. 82, 875–892.
12. Fazleabas, A. T., Hild-Petito, S., and Verhage, H. G. (1994) Secretory proteins
and growth factors of the baboon (Papio anubis) uterus: potential roles in preg-
nancy. Cell. Biol. Int. 18, 1145–1153.
13. Bartol, F. F., Wiley, A. A., Floyd, J. G., et al. (1999) Uterine differentiation as a
foundation for subsequent fertility. J Reprod. Fertil. Suppl. 54, 287–302
14. Wiley, A. A., Bartol, F. F., and Barron, D. H. (1987) Histogenesis of the ovine
uterus. J. Anim. Sci. 64, 1262–1269.
15. Bartol, F. F., Wiley, A. A., Coleman, D. A., Wolfe, D. F., and Riddell, M. G.
(1988) Ovine uterine morphogenesis: effects of age and progestin administra-
tion and withdrawal on neonatal endometrial development and DNA synthesis.
J. Anim. Sci. 66, 3000–3009.
16. Spencer, T. E., Stagg, A. G., Joyce, M. M., et al. (1999) Discovery and character-
ization of endometrial epithelial messenger ribonucleic acids using the ovine uter-
ine gland knockout model. Endocrinology 140, 4070–4080.
17. Gray, C., Bartol, F. F., Taylor, K. M., et al. (2000) Endometrial glands are required
for preimplantation conceptus elongation and survival. Biol. Reprod. 62, 448–456.
18. Gray, C. A., Bazer, F. W., and Spencer, T. E. (2001) Effects of neonatal progestin
exposure on female reproductive tract structure and function in the adult ewe.
Biol. Reprod. 64, 797–804.
07_Spencer_85_94_F 92 8/29/05, 11:17 AM

Sheep UGKO Model 93
19. Gray, C. A., Taylor, K. M., Bazer, F. W., and Spencer, T. E. (2000) Mechanisms
regulating norgestomet inhibition of endometrial gland morphogenesis in the neo-
natal ovine uterus. Mol. Reprod. Dev. 57, 67–78.
20. Spencer, T. E., and Bazer, F. W. (2002) Biology of progesterone action during
pregnancy recognition and maintenance of pregnancy. Front. Biosci. 7, d1879–
d1898.
21. Gray, C. A., Taylor, K. M., Ramsey, W. S., et al. (2001) Endometrial glands are
required for preimplantation conceptus elongation and survival. Biol. Reprod. 64,
1608–1613.
22. Gray, C. A., Burghardt, R. C., Johnson, G. A., Bazer, F. W., and Spencer, T. E.
(2002) Evidence that absence of endometrial gland secretions in uterine gland
knockout ewes compromises conceptus survival and elongation. Reproduction
124, 289–300.
23. Guillomot, M. (1995) Cellular interactions during implantation in domestic rumi-
nants. J. Reprod. Fertil. Suppl. 49, 39–51.
24. Johnson, G. A., Bazer, F. W., Jaeger, L. A., et al. (2001) Muc-1, integrin, and
osteopontin expression during the implantation cascade in sheep. Biol. Reprod.
65, 820–828.
25. Burghardt, R. C., Johnson, G. A., Jaeger, L. A., et al. (2002) Integrins and extra-
cellular matrix proteins at the maternal-fetal interface in domestic animals. Cells
Tissues Organs 171, 202–217.
26. Johnson, G. A., Burghardt, R. C., Bazer, F. W., and Spencer, T. E. (2003)
Osteopontin: roles in implantation and placentation. Biol. Reprod. 69, 1458–1471.
27. Johnson, G. A., Burghardt, R. C., Spencer, T. E., Newton, G. R., Ott, T. L.,
and Bazer, F. W. (1999) Ovine osteopontin: II. Osteopontin and
alpha(v)beta(3) integrin expression in the uterus and conceptus during the
periimplantation period. Biol. Reprod. 61, 892–899.
28. Spencer, T. E., Bartol, F. F., Bazer, F. W., Johnson, G. A., and Joyce, M. M. (1999)
Identification and characterization of glycosylation-dependent cell adhesion mol-
ecule 1-like protein expression in the ovine uterus. Biol. Reprod. 60, 241–250.
29. Johnson, G. A., Spencer, T. E., Burghardt, R. C., and Bazer, F. W. (1999) Ovine
osteopontin: I. Cloning and expression of messenger ribonucleic acid in the uterus
during the periimplantation period. Biol. Reprod. 61, 884–891.
30. Stewart, M. D., Johnson, G. A., Gray, C. A., et al. (2000) Prolactin receptor and
uterine milk protein expression in the ovine endometrium during the estrous cycle
and pregnancy. Biol. Reprod. 62, 1779–1789.
31. Taylor, K. M., Gray, C. A., Joyce, M. M., Stewart, M. D., Bazer, F. W., and
Spencer, T. E. (2000) Neonatal ovine uterine development involves alterations in
expression of receptors for estrogen, progesterone, and prolactin. Biol. Reprod.
63, 1192–1204.
32. Carpenter, K. D., Gray, C. A., Noel, S., Gertler, A., Bazer, F. W., and Spencer, T.
E. (2003) Prolactin regulation of neonatal ovine uterine gland morphogenesis.
Endocrinology 144, 110–120.
07_Spencer_85_94_F 93 8/29/05, 11:17 AM

94 Spencer and Gray
33. Carpenter, K. D., Gray, C. A., Bryan, T. M., Welsh, T. H., Jr., and Spencer, T. E.
(2003) Estrogen and antiestrogen effects on neonatal ovine uterine development.
Biol. Reprod. 69, 708–717.
34. Johnson, G. A., Burghardt, R. C., Joyce, M. M., et al. (2003) Osteopontin is syn-
thesized by uterine glands and a 45-kDa cleavage fragment is localized at the
uterine-placental interface throughout ovine pregnancy. Biol. Reprod. 69, 92–98.
35. Gray, C. A., Stewart, M. D., Johnson, G. A., and Spencer, T. E. (2003) Postpar-
tum uterine involution in sheep: histoarchitecture and changes in endometrial gene
expression. Reproduction 125, 185–198.
36. Spencer, T. E. and Bazer, F. W. (1995) Temporal and spatial alterations in uterine
estrogen receptor and progesterone receptor gene expression during the estrous
cycle and early pregnancy in the ewe. Biol. Reprod. 53, 1527–1543.
37. Spencer, T. E., Johnson, G. A., Burghardt, R. C., and Bazer, F. W. (2004) Proges-
terone and placental hormone actions on the uterus: insights from domestic ani-
mals. Biol. Reprod. 71, 2–10.
07_Spencer_85_94_F 94 8/29/05, 11:17 AM

Baboon Model of Endometriosis 95
8
A Baboon Model for Inducing Endometriosis
Asgerally T. Fazleabas
Summary
Endometriosis is a disease that is associated with severe pelvic pain and is a major cause of
infertility in women. It is an enigmatic disease whose etiology and pathophysiology has been
studied to a limited extent. The events associated with the establishment of the disease and
mechanisms associated with infertility are difficult to assess in a systemic manner in women. In
order to understand the early and progressive events associated with the establishment of the
disease, we have developed a baboon model in which the disease can be induced. This induc-
tion manifests itself in a manner that recapitulates the spontaneous disease. The advantage of
the induced model is that the progressive changes in both the ectopic and eutopic endometrium
can be studied in a nonhuman primate model at specific times during the menstrual cycle and as
the disease process continues.
Key Words: Baboon; Papio anubis; endometriosis; uterus; endometrium.
1. Introduction
Endometriosis is defined as the presence of endometrium-like tissue outside
of the uterine cavity. It is one of the most common causes of infertility and
chronic pelvic pain and affects 1 in 10 women in the reproductive-age group
(1). This incidence increases up to 30% in patients with infertility and up to
45% in patients with chronic pelvic pain (2). Endometriosis is an estrogen-
dependent gynecological condition. According to Sampson’s theory (3), frag-
ments of menstrual endometrium are refluxed through the fallopian tubes into
the peritoneal cavity, then attach to and grow on peritoneal surfaces. However,
the fundamental mechanisms by which menstrual endometrium adheres, pro-
liferates, and establishes a functional vasculature in an ectopic site remain to
be elucidated. We propose that endometriosis develops in two distinct phases.
Phase I is invasive and dependent on ovarian steroids. Phase II, which is the
active phase of the disease, is characterized by endogenous estrogen biosyn-
thesis.
95
08_Fazleabas_95_100_F 95 8/29/05, 11:17 AM

96 Fazleabas
The baboon is a valuable and clinically relevant model with which to study
the etiology and consequences to fertility of this enigmatic disease (4). Experi-
mental evidence indicates that intrapelvic injections of menstrual endometrium
can induce endometriosis in this primate, thereby supporting the basic tenets of
the Sampson’s hypothesis (3). Furthermore, because hormonal manipulations
are possible, we can also address the role that ovarian and extra-ovarian hor-
mones play in the establishment and maintenance of this disease.
2. Materials
1. Laproscopic equipment with video and photographic accessories.
2. Unimar Pipelle (Cooper Surgical Inc., Shelton, CT)
3. Methods
3.1. Induction of Disease
The experimentally induced baboon endometriosis model was first estab-
lished by D’Hooghe et al. (5). We have modified the original procedure, which
is described in this chapter and in refs. 6 and 7).
Menstrual endometrium (approx 1 g) is harvested on day 2 of menses using
a Unimar Pipelle just prior to laparoscopy. The Pipelle is inserted through the
external cervical os into the uterine lumen. The cervical os is visualized by
dilating the vagina using a pediatric speculum. Menstrual fluid and tissue is
aspirated into the Pipelle by rotating it gently within the lumen and applying
suction aspiration. Usually, two Pipelles filled with fluid and tissue is suffi-
cient for inoculation and the induction of endometriosis (see Notes 1 and 2).
The peritoneal cavity and reproductive tract is visualized by laparoscopy
and the absence of any lesions or adhesions is documented by video recording
(see Note 3). Under laparoscopic guidance, the menstrual tissue in the Pipelle
is deposited at three sites: the pouch of Douglas, the broad ligament adjacent to
the oviducts, and on the uterus. At the subsequent menses, the animals undergo
a second laparoscopy and endometrial re-seeding at the same ectopic sites.
Following the second seeding, and the third menses, laparoscopy is performed
to evaluate the extent of endometriosis (6). Additional laparoscopies are per-
formed at periodic intervals every 3 mo following the initial inoculation (see
Note 4). Figure 1A shows a blue lesion documented by video recording and
the corresponding histological appearance of the lesions. The presence of glands
and stroma at this ectopic site that was seeded using menstrual endometrium
meets the classical pathological definition of endometriosis (Fig. 1B,C).
Following video documentation, endometriotic lesions and portions of nor-
mal peritoneum can be biopsied using a harmonic scalpel under laparoscopic
guidance. If both the eutopic and ectopic tissue is required, an endometriectomy
08_Fazleabas_95_100_F 96 8/29/05, 11:17 AM

Fig. 1. (A) Video micrograph of a blue lesion on the peritoneum. (B,C) Histologi-
cal sections of endometriotic lesions showing the presence of both glands and stroma.
can be performed on the appropriate day of the menstrual cycle as described in

Chapter 9. These tissues can then be processed for histological, biochemical,
or molecular analyses (7).
3.2. Evaluation of the Lesions
At the time of laparoscopy, the size and shape of the lesion is documented
on a pelvic diagram sheet and the color and anatomical location is recorded.
The stage of the disease can be classified using a modification of the revised
American Society for Reproductive Medicine staging criteria. Details for this
scoring method in the baboon have been published (8).
Endometriosis can take on a variety of appearances. A wide range of lesion
types described in humans are also evident in the baboon model. For example,
red, raised nodules on the superficial peritoneum or reddish-blue proliferative
endometriotic nodules are most commonly encountered in the baboon during
the early stages (see ref. 6 and Fig. 1A). The opaque white peritoneal nodules
that lack hemosiderin are also evident. The classic brownish focal adhesions
are only observed at the later stages following inoculation (6). In the baboon,
08_Fazleabas_95_100_F 97 8/29/05, 11:17 AM

98 Fazleabas
the majority of lesions that result following inoculation of menstrual endometrium

are observed on the peritoneum and uterine surface. There is no evidence of ova-
rian implants, termed “endometriomas,” which are commonly observed in
women. The lesions that are present on the peritoneum and uterine surface at
the early stages (1–3 mo) are primarily raised red lesions or reddish blue
lesions. Larger lesions obtained from the animals at the later stages of the
disease (>10 mo) are more readily comparable with the “chocolate cysts”
described in humans. There is also evidence of significant adhesion of the
peritoneum to the ovary as the disease progresses. Filmy adhesions in the
peritoneal cavity are commonly observed and lesions on the bladder are not
uncommon. Another characteristic of the disease is the presence of perito-
neal fluid, which is not readily seen in animals without disease.
Microscopic evaluation of endometriotic lesions should reveal endometrial
glands and stroma along with evidence of fibrosis and hemorrhage (Fig. 1B,C).
Pathological confirmation of the disease traditionally requires the presence of
at least two of these components. In our baboon model, we established the
criteria that both endometrial glands and stroma must be evident in lesions
classified as endometriosis (6). Based on these criteria, we have been able to
document that baboon lesions following the artificial induction of the disease
do have distinct endometrial glands and stroma (see refs. 6 and 7, and Fig.
1B,C). These procedures indicate that intraperitoneal inoculations of menstrual
endometrium result in endometriotic lesions that substantiate the Sampson hypoth-
esis (3) in a nonhuman primate model.
4. Notes
1. Baboons, unlike rhesus macaques, usually have a straight cervix without exten-
sive folds. Occasionally, there are baboons whose cervixes are difficult to navi-
gate with the Pipelle. It is prudent to evaluate the animals in a mock cycle either
during menses or in the proliferative stage of the menstrual cycle to evaluate the
accessibility of the uterine cavity.
2. To study direct hormonal effects on lesion formation, the ovariectomized baboon
model could be utilized. Using the steroid hormone replacement regimen described
in ref. 9, a progesterone withdrawal bleed is induced following 28 d of estradiol
and progesterone treatment. The baboons usually have a menstrual bleed 48 h
following progesterone withdrawal. Menstrual tissue collected from these ani-
mals can be used for inoculation to induce endometriosis. Steroid hormone or
control treatments could then be initiated in the ovariectomized animals.
3. Baboons are also afflicted with spontaneous endometriosis, and this disease is
progressive throughout the animal’s lifespan (8). Therefore, it is critical that prior
to the induction of endometriosis using the induced model, a thorough
laparoscopic evaluation be performed to rule out spontaneous disease in the study
animals.
08_Fazleabas_95_100_F 98 8/29/05, 11:17 AM

4. For control studies, we inoculate the peritoneal cavity with endometrial tissue
obtained at the mid secretory phrase of the menstrual cycle. In our experience,
this tissue does not attach to the peritoneum and form active endometriotic
lesions, as do tissues from the menstrual effluent (6).
Acknowledgments
The expert assistance of the Veterinary Staff at the Biological Resources
Laboratory and the technical assistance of Ms. Allison Brudney is gratefully
acknowledged. These studies have been supported by National Institutes of
Health (NIH) grant HD40093.
References
1. Eskenazi, B. and Warner, M. L. (1997) Epidemiology of endometriosis. Obstet.
Gynecol. Clin. North Am. 24, 235–258.
2. Gruppo italiano. (1994) Prevalence and anatomical distribution of endometriosis in
women with selected gynaecological conditions: results from a multicentric Italian
study. Gruppo italiano per lo studio dell’endometriosi. Hum. Reprod. 9, 1158–1162.
3. Sampson, J. A. (1927) Peritoneal endometriosis due to menstrual dissemination of
endometrial tissue into the peritoneal cavity. Am. J. Obstet. Gynecol. 14, 422–469.
4. D’Hooghe, T. M. (1997) Clinical relevance of the baboon as a model for the study
of endometriosis. Fertil. Steril. 68, 613–625.
5. D’Hooghe, T. M., Bambra, C. S., Raeymaekers, B. M., De Jonge, I., Lauweryns,
J. M. and Koninckx, P. R. (1995) Intrapelvic injection of menstrual endometrium
causes endometriosis in baboons (Papio cynocephalus and Papio anubis). Am. J.
Obstet. Gynecol. 173, 125–134.
6. Fazleabas, A. T., Brudney, A., Gurates, B., Chai, D., and Bulun, S. E. (2002)
A primate model for endometriosis, in Proceedings of a NIH Workshop on
Endometriosis: Emerging Research and Intervention Strategies (Yoshinaga,
K. and Parrot, E., eds.). Ann. N. Y. Acad. Sci. 955, 308–317.
7. Fazleabas, A. T., Brudney, A., Chai, D., Langoi, D., and Bulun, S. E. (2003) Ste-
roid receptor and aromatase expression in baboon endometriotic lesions. Fertil.
Steril. 80(Suppl 2), 820–827.
8. D’Hooghe, T. M., Bambra, C. S., Raeymaekers, B. M., and Koninckx, P. R. (1996)
Serial laparoscopies over 30 months show that endometriosis in captive baboons
(Papio anubis, Papio cynocephalus) is a progressive disease. Fertil. Steril. 65,
645–649.
9. Fazleabas, A. T., Miller, J. B., and Verhage, H. G. (1988) Synthesis of estrogen
and progesterone dependent proteins by the baboon (Papio anubis) endometrium.
Biol. Reprod. 39, 729–736.
08_Fazleabas_95_100_F 99 8/29/05, 11:17 AM

08_Fazleabas_95_100_F 100 8/29/05, 11:17 AM
Baboon Model of Pregnancy 101
9
A Baboon Model for Simulating Pregnancy
Asgerally T. Fazleabas
Summary
Estrogen and progesterone secreted by the corpus luteum regulate the function of the uterine
endometrium in preparation for pregnancy. Embryonic signals superimposed on this steroid
hormone-primed uterus further modulate the uterine environment to make it conducive to preg-
nancy. Understanding the signaling mechanisms that initiate the embryonic–maternal dialog in
humans is not feasible. In an effort to elucidate the role of chorionic gonadotropin as a mediator
of endometrial function in addition to its luteotrophic role, we have developed a simulated
pregnant model in the baboon. Infusion of chorionic gonadotropin in a manner that mimics
blastocyst transit induces major changes in the morphology and secretory activity of the
endometrium. This model provides a method by which the function of various embryonic
factors on endometrial can be tested in an in vivo model.
Key Words: Baboon; Papio anubis; pregnancy; chorionic gonadotropin; implantation.
1. Introduction
The establishment of pregnancy requires a synchronous interaction between
the embryo and the endometrium. These interactions are necessary to both pro-
long corpus luteum function and modulate the uterine environment to permit a
normal embryo to attach and invade a receptive uterine endometrium (1–4). In
ruminants and pigs, embryo signals act in a paracrine manner on the uterine
endometrium to inhibit the pulsatile release of the luteolytic factor, prostaglan-
din F2α. In contrast, the primate embryonic signal, chorionic gonadotropin
(CG), has a direct luteotrophic effect on the corpus luteum. The presence of
luteinizing hormone (LH)/CG receptors have been documented to the present
in the primate endometrium (5,6), and in vitro stimulation of both epithelial
and stromal endometrial cells activates signal transduction and gene transcrip-
tion (7–11). To determine if CG further modulates the estrogen and progester-
one primed receptive endometrium in vivo, we have developed a nonhuman
primate simulated pregnant model. The baboon (Papio anubis) was used as the
nonhuman primate of choice.
101
09_Fazleabas_101_110_F 101 8/29/05, 11:18 AM

102 Fazleabas
2. Materials
1. V4 Polyvinyl Cannula (BioLab Products, Lake Havasu City, AZ).
2. Dow Corning silastic capsules (Dow Corning, Midland, MI).
3. Silastic medical adhesive Type A (Dow Corning, Midland, MI).
4. Alzet osmotic minipump 2ML1 (Alza Corp, Palo Alto, CA).
5. Recombinant human CG (hCG; Serono Pharmaceuticals).
6. Microsurgical dissection instruments.
7. 4.0 nonabsorbable suture on a tapered needle.
8. Coat-a-Count Estradiol Assay kit (Diagnostic Products Corporation, Los Ange-
les, CA).
3. Methods
Ovulation is monitored in normally cycling female baboons by measuring
serum estradiol levels beginning 7 d following the first day of visible menses.
Day 1 of ovulation is designated to be 48 h following the estradiol surge.
3.1. Placement of Cannula
On day 6 post ovulation, baboons are sedated with Ketamide hydrochloride
(10 mg/kg) and transferred to intravenous administration of Thiopentol. The
animals are maintained on a surgical plane of anesthesia with isofluorane and
oxygen.
Under sterile operating conditions, the oviducts are exteriorized following a
mid-ventral incision. Using a pair of microdissection scissors, an initial cut is
made through the mesosalpinx of the oviduct in a region of the isthmus with no
convolutions. Care must be taken not to cut through the entire mesosalpinx.
The oviductal lumen is visualized using microdissection forceps to ensure that
the lumen is open. The polyvinyl cannula with a medical adhesive bead attached
1–2 cm from the tip is inserted into the oviductal lumen (Fig. 1). Generally, the
cannula will extend approx 1 cm into the oviductal lumen. The cannula is
sutured in place with 4.0 nonabsorbable suture and the cut edges of the
mesosalpinx are pulled over the adhesive bead and sutured in place (Fig. 1).
Once the cannula is secure in the oviduct, a small subcutaneous flank inci-
sion is made on the side of the body wall that the cannula is attached. The body
wall is punctured using blunt dissection with a hemostat and the polyvinyl
cannula is exteriorized. The open end of the cannula is attached to the Alzet
minipump (Fig. 1).
3.2. Preparation of Alzet Minipump
The 2ML1 Alzet pump holds 2 mL of fluid and lasts for 7 d with a flow rate
of 10 µL/h. For hCG infusion, the solution containing 1.25 IU/10 µL is made
up in a 2.5-mL volume. The Alzet minipump is primed in sterile saline over-
09_Fazleabas_101_110_F 102 8/29/05, 11:18 AM

Fig. 1 (see companion CD for color version). A diagrammatic illustration of the

placement of the cannula within the oviduct and the subcutaneous attachment to the
Alzet minipump for the infusion studies.
night at 37°C. For control studies, the hCG is heat-inactivated by boiling the
solution for 30 min. The manufacturer provides detailed information for filling
and preparing the osmotic minipump.
The appropriately primed minipump is inserted into the open end of the
cannula. Prior to insertion, the dead space within the cannula is gently primed
09_Fazleabas_101_110_F 103 8/29/05, 11:18 AM

104 Fazleabas
with part of the balance 0.5 mL of solution. A 21-gauge stub adapter needle
attached to a 1-cc syringe is inserted into the cannula and the solution is gently
infused to fill up the dead space. The cannula is then attached to the metal
prong on the Alzet pump and held in place by 2.0 silk sutures. The pump is
then inserted into a subcutaneous pocket and the infusion is initiated (Fig. 1).
3.3. Harvesting of Endometrial Tissues
Endometrial tissue is harvested on either day 10 or day 14 post ovulation.
Day 10 corresponds to the approximate day of implantation in the baboon and
day 14 corresponds to the earliest time point at which an implantation site can
be readily identified (12). The 24-h infusion rate of 30 IU hCG is equivalent to
the amount of baboon CG secreted by dispersed baboon trophoblast cells cul-
tured in vitro (13). The 4-d infusion period (days 6–10 post ovulation) corre-
sponds to the window of time when the blastocyst is present within the uterine
cavity and is associated with the initial phases of attachment and invasion.
Thus, this simulated pregnant model mimics the intrauterine hormonal milieu
associated with blastocyst transit and attachment in the baboon (see Notes 1
and 2).
The uterus is carefully exteriorized following a mid-ventral incision (Fig. 2B).
The myometrium is gently injected with 2 mL of vasopressin (20 IU/mL) using
a 3-cc syringe and a 25-gauge needle. This constricts the blood vessels in the
myometrium and decreases the bleeding during the myometrial incision. Using
a number 11 scalpel blade, a longitudinal incision is made through the myo-
metrium until the functionalis tissue is exposed (Fig. 2D). The functionalis is
gently peeled away from the basalis using the pointed end of a metal weighing
spatula (Fig. 2E,F). This procedure is termed an endometriectomy. Following
removal of the endometrial tissue, the myometrial incision is closed by con-
tinuous suture using 4.0 Vicryl and a small tapered needle (Fig. 2G). Prior to
closing the myometrial incision, the uterine cavity is flushed extensively with
warm sterile saline to minimize bleeding and adhesions.
If the corpus luteum is required for analysis, a small incision is made at the
site of the ovarian stigma using a number 15 scalpel blade. The corpus luteum
is gently teased out of the ovary using a small curved hemostat. The incision is
closed with 4.0 Vicryl suture.
Following the endometriectomy and lutectomy, the peritoneal cavity is
extensively flushed with warm saline until there is minimal blood in the flush.
This extensive flushing prevents significant adhesion formation and permits
the multiple use of these animals as permitted by the Institutional Animal Care
and Use Committee (IACUC) guidelines.
09_Fazleabas_101_110_F 104 8/29/05, 11:18 AM

09_Fazleabas_101_110_F
105
Baboon Model of Pregnancy
105
8/29/05, 11:18 AM
Fig. 2. Step-by-step illustration of the endometriectomy procedure in the baboon to harvest endometrial tissues.
105
106 Fazleabas
The peritoneum and fascia are closed using 2.0 Vicryl on a tapered needle
using a simple interrupted suture technique. The subcutaneous tissue is closed
with a 3.0 Vicryl simple continuous suture line and the skin is then closed with
3.0 Vicryl with a continuous subcuticular suture (Fig. 2H). This is a subdermal
suture that ensures that the skin suture will not be loosened by the animal.
The Alzet minipump that has been placed subcutaneously is then removed
following a subcutaneous incision. The tip of the cannula attached to the pump
is cut and it is anchored to the body wall using a 0.5-cm piece of silastic tubing
and a 2.0 silk suture. This provides a landmark for locating the cannula for the
next infusion (see Note 3). The skin incision is closed using 3.0 Vicryl and a
subdermal suture. The amount of solution left in the pump is measured by
aspirating the fluid using a 25-gauge needle provided by the manufacturer.
This provides assurance that the expected volume of hormone has been infused
and that the pump has functioned appropriately (see Notes 4 and 5).
3.5. Analysis of Endometrial Tissues
The endometrial tissues obtained by endometriectomy are transported to the
laboratory in ice-cold Ca2+/Mg2+-free Hank’s balanced salt solution (see Note
6). The tissue is carefully dissected under a microscope under sterile condi-
tions so that the luminal surface is exposed. Using a sharp, single-edged blade,
portions of the tissue are fixed in the appropriate fixatives for histological and
immunocytochemical analyses or for in situ hybridization (9). Tissues can also
be snap frozen in liquid nitrogen for RNA extraction (9) or subjected to enzy-
matic digestion for stromal and epithelial cell isolation (14,15). In general,
following CG stimulation, the amount of functionalis tissue harvested is approx
500–700 g. Figure 3 provides a composite example of the response of the
baboon endometrium to CG stimulation during the window of receptivity (9).
None of the changes are evident if the hCG had been heat inactivated prior to
infusion (9).
Fig. 3. (opposite page) Endometrial responses to human chorionic gonadotropin

(hCG) infusion on day 10 post ovulation. A shows the characteristic endometrial
plaque response in the luminal epithelial cells usually seen in response to CG stimula-
tion and early pregnancy (9,14). B shows the induction of α-smooth muscle actin in
the subepithelial stroma that is characteristic of a predecidual response in the baboon
(14,15). C shows secretory proteins labeled with 35S methionine evaluated by two-
dimensional gel electrophoresis following organ culture (18). The boxed area repre-
sents glycodelin, which is a protein that undergoes both transcriptional (reverse
transcriptase-polymerase chain reaction) and posttranslational (Western blotting
[WB]) modification in response to CG and is localized specifically to the glandular
epithelium (D) (9,19).
09_Fazleabas_101_110_F 106 8/29/05, 11:18 AM

107
09_Fazleabas_101_110_F 107 8/29/05, 11:18 AM

108 Fazleabas
4. Notes
1. The cannulated animals could serve as their own controls. Surgeries could be
scheduled on day 10 or 14 post ovulation of a regular menstrual cycle. Endome-
trial tissues can be obtained by endometriectomy in the absence of CG stimula-
tion, which reduces between animal variability with or without treatment. If this
procedure is selected, the cannula is inserted into the oviduct after the
endometriectomy. The open end of the cannula is exteriorized through the flank
incision and sutured to the body wall using a 0.5-cm piece of silastic tubing. The
animals are rested for three consecutive menstrual cycles prior to the next sur-
gery. For CG treatment, a small incision is made above the silastic anchor and the
cannula is gently extruded from under the skin. The cannula is flushed with the
treatment solution and inserted into the metal tip of the Alzet minipump. The
pump is placed in the subcutaneous pocket and the treatment initiated on day 6
post ovulation. These manipulations do not require an incision into the animal’s
body cavity at the initiation of treatment.
2. To determine if the effects are directly on the uterine endometrium, the treatment
paradigm could be done in ovariectomized animals following hormone replace-
ment (16). The cannula is placed into the oviduct at the time of ovariectomy and
anchored to the body wall. CG stimulation can be initiated following the sequen-
tial treatment with estrogen and progesterone implants to mimic hormonal
changes during the menstrual cycle (9,16,17).
3. Over a period of time, the cannula that is placed subcutaneously for easy access
becomes brittle. This appears to be primarily to the result of a granulation reac-
tion. Because it is less pliable, insertion into the metal post on the Alzet osmotic
minipump can be more difficult. This can be overcome by expanding the open
end of the cannula with 21-gauge stub adapter or by using the tip of the dissecting
scissors to expand the top of the tubing. Once inserted into the pump, the cannula
is secured using 2.0 silk ties.
4. The procedure describes infusion of CG to determine the role of the major pri-
mate embryonic signal on uterine receptivity. However, this cannulation proce-
dure can be used for infusing a variety of other hormones or growth factors in the
presence or absence of CG to determine individual or synergistic effects of these
factors on endometrial function during the window of receptivity.
5. Stimulation with CG via the Alzet pump and cannula is only effective upto 14 or
15 days post ovulation. Longer infusions do not provide sufficient luteotrophic
stimulation to maintain the corpus luteum. If long-term treatment with CG (up to
day 18 post ovulation) is required, then the animals can be given hCG injections
twice daily for up to 12 d, beginning on day 6 post ovulation. This regimen mim-
ics hormonal and endometrial changes that are comparable to the initial stages of
pregnancy. Details of this procedure have been previously published (17).
6. In both cycling and CG treated animals the uterine lumen can be flushed at the
time of surgery to obtain uterine flushings. Following exteriorization of the
uterus, the assisting surgeon clamps the cervix and the oviducts with his fingers.
Five to ten milliliters of Ca2+/Mg2+-free Hands Buffered saline is aspirated into a
09_Fazleabas_101_110_F 108 8/29/05, 11:18 AM

10-mL syringe attached to a 21-gauge needle. Another empty 10-cc syringe is

also attached to a 21-gauge needle. With the cervix and oviducts clamped off,
the syringe containing the media is inserted through the myometrium into the
uterine lumen. A small volume (1 mL) of fluid is injected to dilate the lumen. The
empty syringe is then inserted into the lumen and the uterus infused (Fig. 2C).
The uterine flushings are forced into the empty syringe by positive pressure. The
syringes and needles are slowly withdrawn and the intrusion points on the myo-
metrium are cauterized via conductance through the needle as it is being removed.
Acknowledgments
The expert assistance of the Veterinary Staff at the Biological Resources
Laboratory and the technical assistance of Ms. Allison Brudney is gratefully
acknowledged. These studies have been supported by National Institutes of
Health (NIH) grants HD29964 and HD42280.
References
1. Paria, B. C., Reese, J., Das, S. K., and Dey S. K. (2002) Deciphering the cross-talk
of implantation: advances and challenges. Science 296, 2185–2188.
2. Jaeger, L. A., Johnson, G. A., Ka, H., et al. (2001) Functional analysis of autocrine
and paracrine signaling at the uterine-conceptus interface in pigs. Reprod. Suppl.
58, 191–207.
3. Spencer, T. E. and Bazer, F. W. (2002) Biology of progesterone action during
pregnancy recognition and maintenance of pregnancy. Front. Biosci. 7, d1879–
d1898.
4. Fazleabas, A. T. and Strakova, Z. (2002) Endometrial function: cell specific
changes in the uterine environment. Mol. Cell. Endocrinol. 186, 143–147.
5. Rao, C. V. (2001) An overview of the past, present, and future of nongonadal LH/
hCG actions in reproductive biology and medicine. Semin. Reprod. Med. 19, 7–17.
6. Licht, P., von Wolff, M., Berkholz, A., and Wildt, L. (2003) Evidence for cycle-
dependent expression of full-length human chorionic gonadotropin/luteinizing
hormone receptor mRNA in human endometrium and decidua. Fertil. Steril. 79
(Suppl 1), 718–723.
7. Licht, P., Russu, V., Lehmeyer, S., and Wildt, L. (2001) Molecular aspects of
direct LH/hCG effects on human endometrium-lessons from intrauterine
microdialysis in the human female in vivo. Reprod. Biol. 1, 10–19.
8. Licht, P., Russu, V., and Wildt, L. (2001) On the role of human chorionic gona-
dotropin (hCG) in the embryo–endometrial microenvironment: implications for
differentiation and implantation. Semin. Reprod. Med. 19, 37–47.
9. Fazleabas, A. T., Donnelly, K. M., Srinivasan, S., Fortman, J. D., and Miller, J. B.
(1999). Modulation of the baboon (Papio anubis) uterine endometrium by chori-
onic gonadotrophin during the period of uterine receptivity. Proc. Natl. Acad. Sci.
USA 96, 2453–2458.
10. Banaszak, S., Donnelly, K. M., Brudney, A., Chai, D., Chwalisz, K., and
Fazleabas, A. T. (2000) Modulation of the action of chorionic gonadotrophin on
09_Fazleabas_101_110_F 109 8/29/05, 11:18 AM

110 Fazleabas
the baboon endometrium by a progesterone receptor antagonist (ZK 137.316).

Biol. Reprod. 63, 819–823.
11. Srisuparp, S., Strakova, Z., Brudney, A., et al. (2003) Signal transduction path-
ways activated by chorionic gonadotrophin (CG) in the primate endometrial epi-
thelial cells. Biol. Reprod. 68, 457–464.
12. Jones, C. J., Enders, A. C., and Fazleabas, A. T. (2001) Early implantation events
in the baboon (Papio anubis) with special reference to the establishment of an-
choring villi. Placenta 22, 440–456.
13. Bambra, C. S. and Tarara, R. (1990) Immunohistochemical localization of chori-
onic gonadotrophin on baboon placenta, dispersed trophoblast cells and those de-
rived from blastocysts grown in vitro. J. Reprod. Fertil. 88, 9–16.
14. Kim, J. J., Jaffe, R. C., and Fazleabas, A. T. (1998) Comparative studies on the in
vitro decidualization process in baboons (Papio anubis) and humans. Biol.
Reprod. 59, 160–168.
15. Kim, J. J., Jaffe, R. C., and Fazleabas, A. T. (1999) Insulin-like growth binding
protein 1 in baboon endometrial stromal cells: regulation by filamentous actin and
requirement for de novo protein synthesis. Endocrinology 140, 997–1004.
16. Fazleabas, A. T., Miller, J. B., and Verhage, H. G. (1988) Synthesis of estrogen
and progesterone dependent proteins by the baboon (Papio anubis) endometrium.
Biol. Reprod. 39, 729–736.
17. Hild-Petito, S., Donnelly, K. M., Miller, J. B., Verhage, H. G., and Fazleabas, A.
T. (1995) A baboon (Papio anubis) simulated-pregnant model: cell specific ex-
pression of insulin-like growth factor binding protein-1 (IGFBP-1), type I IGF
receptor (IGF-I R) and retinol binding protein (RBP) in the uterus. Endocrine 3,
639–651.
18. Fazleabas, A. T., Donnelly, K. M., Mavrogianis, P. A., and Verhage, H. G. (1993)
Secretory and morphological changes in the baboon (Papio anubis) uterus and
placenta during early pregnancy. Biol. Reprod. 49, 695–704.
19. Hausermann, H. M., Donnelly, K. M., Bell, S. C., Verhage, H. G., and Fazleabas,
A. T. (1998) Regulation of the glycosylated β-lactoglobulin homologue,
glycodelin [placental protein 14 (PP14 )] in the baboon uterus. J. Clin. Endocrinol.
Metab. 83, 1226–1233.
09_Fazleabas_101_110_F 110 8/29/05, 11:18 AM

Marmoset Implantation/Early Pregnancy Model 111
10
The Common Marmoset Monkey as a Model
for Implantation and Early Pregnancy Research
Almuth Einspanier, Kai Lieder, Ralf Einspanier, and Bettina Husen
Summary
This chapter describes methods used to investigate implantation in the common marmoset
monkey, Callithrix jacchus. A reverse-transcriptase polymerase chain reaction-strategy with
which to detect transcripts for steroid receptors and enzymes involved in estradiol biosynthesis
is described, and an immunohistochemistry approach for detecting proteins within the implan-
tation site is presented.
Key Words: Marmoset; early pregnancy; steroid receptors; aromatase; 17β-hydroxysteroid
dehydrogenase type 1 and type 7.
1. Introduction
The marmoset monkey (Callithrix jacchus) belongs to the New World Mon-
keys and is widely used as a primate model for reproductive medicine (1,2).
One of its advantages is its small size, which is the reason why it is easy to
handle and breed. The marmoset has a high fecundity and has no restricted
breeding season like other laboratory primates such as rhesus monkeys. Because
its cycle can be controlled by prostaglandin F2α (PGF2α) application (3), exact
prognosis of ovulation and therefore status of pregnancy is possible. The mar-
moset monkey has a placenta hemochorialis like humans and, consequently, it
shows similarities to the human situation in morphology and function (4). Our
knowledge about factors concerning the implantation process in humans is lim-
ited as a result of ethical constraints, which make primate models necessary.
Furthermore, there is an urgent need for more information on mechanisms of
implantation. This is easily understandable in view of current progress in repro-
ductive medicine, especially because pregnancy rates in assisted reproduction
are still unsatisfactory (5).
111
10_Einspanier_111_122_F 111 8/29/05, 11:18 AM

112 Einspanier et al.
Implantation requires an appropriately developed embryo and a receptive

endometrium primed by relevant endocrine as well as paracrine factors. This
permits a remodelling of the endometrium to provide optimal development and
nutrition of the embryo including appropriate embryo-maternal interactions.
Suitable methods for studies on implantation using the marmoset monkey
model will be presented in this paper.
2. Materials
2.1. Reproductive Staging of Animals
1. Adult (>18 mo old) marmoset monkeys. Characteristics: 8- to 10-d follicular phase
and 18- to 20d luteal phase (1); no visible menstruation, nonseasonal primate.
2. PGF2α (0.8 µg per animal, Estrumate, Pitman-Moore, Germany) application can
induce luteolysis when applied between days 10–15 of luteal phase (6).
3. Immunoassays (6):
a. Progesterone for monitoring cycle stage (detection range: 0.1–50 ng/mL).
b. Relaxin (RLX) for early pregnancy detection (detection range: 0.02–5 ng/mL).
4. Ultrasound examination with 7.5-, 10-, and 15-MHz probes (Logiq 400 Pro CL,
General Electrics, Solingen, Germany) of ovarian and uterine activity throughout
the cycle and pregnancy.
2.2. Reverse-Transcription-Polymerase Chain Reaction

1. RNeasy Mini-Kit (Qiagen, Hilden, Germany).
2. SuperScript II reverse transcriptase (Invitrogen, Karlsruhe, Germany).
3. RNase H (Invitrogen, Karlsruhe, Germany).
4. Oligo(dT)12–18 primer (500 µg/mL) (Hermann GbR, Freiburg, Germany).
5. Taq-DNA-Polymerase (Eppendorf GmbH, Hamburg, Germany).
6. Pfu Polymerase (Fermentas GmbH, St. Leon-Rot, Germany).
7. dNTP-set 1 (Roth GmbH, Karlsruhe, Germany).
8. Oligonucleotide primers for marmoset mRNA-transcripts (see Table 1).
9. GeneRuler(tm) 100-basepair (bp) DNA Ladder plus and 6X Loading Dye Solu-
tion (Fermentas, St. Leon-Rot, Germany).
2.3. Immunohistochemistry
1. Fixation solution: 4% paraformaldehyde, pH 7.3.
2. Buffer for microwave pretreatment: 10 mM citrate buffer, pH 6.0 at 120°C.
3. Tris-buffered saline (TBS), poly-L-lysine (0.01%).
4. ABC-method (DAKO Diagnostika, Hamburg, Germany; Vectastain, Vector
Laboratories, Burlingame, CA, USA).
5. Primary antibodies:
a. Monoclonal mouse antibody against estradiol α receptor (ERα; cat. no. B10,
Euromedex; Souffel Weyersheim, France).
10_Einspanier_111_122_F 112 8/29/05, 11:18 AM

10_Einspanier_111_122_F
Table 1
Oligonucleotide Primers Used for Marmoset Transcript Amplification
Annealing
temperature Product
113
Transcript GenBank Forward primer Reverse primer (°C) (bp)
Aromatase (ARO) AY034779 5' ACA ACT CGG CCC 5' AGG AGC TGC AAT CAG 60 498
CTC TTT AT 3' CAT TT 3'
Estrogen receptor α (ERα) X03635 (human) 5' ATG ACC ATG ACC 5' CGG AGA CAC GCT GTT 58 315
CTC CAC AC 3' GAG T 3'
Progesterone receptor (PR) Z86038 5' GTA TTC CAA ATG 5' AAC CAA TTG CCT TGA 60 591
AAA GCC AAG C 3' TGA GC 3'
17-β-Hydroxysteroid AF272013 5' GGC CTG CAC TTG 5' GGC CTG CAG CAT CCG 60 330
Marmoset Implantation/Early Pregnancy Model
dehydrogenase type 1 GCC GTA CG 3' CAC AG 3'

(17HSD1)
17HSD7 AF263468 5' CTG CTG AGG TCA 5' CCA GAT GAG CTG AGA 60 344
CCA TTG TA 3' TGG AT 3'
18S rRNA housekeeping AF176811 5' GAG AAA CGG CTA 5' GAC ACT CAG CTA AGA 64 317
8/29/05, 11:18 AM
gene (18S) CCA CAT CCA A 3' GCA TCG A 3'
113
b. Monoclonal mouse antibody against progesterone receptor (PR; cat. no. AT

4.14, Dianova, Hamburg, Germany).
c. Affinity-purified polyclonal rabbit antisera against 17β-hydroxysteroid dehy-
drogenase type 1 (17HSD1), 17β-hydroxysteroid dehydrogenase type 7
(17HSD7), and aromatase (ARO) (7).
d. Polyclonal rabbit antibodies against cytokeratin (Biotrend, Köln, Germany)
and vimentin (Biotrend, Köln, Germany).
6. 3% H2O2.
7. Biotinylated goat-anti-rabbit immunoglobulin (Ig)G (Vector Laboratories,
Burlingame, CA, USA).
8. Streptavidin–horeseradish peroxidase (HRP) (DAKO; Vector)
9. AEC-Substrate Chromogen (DAKO, Vector Laboratories).
10. Meyer´s hematoxylin (Merck, Darmstadt, Germany) for nuclear counterstaining.
11. Histogel medium (Vector Laboratories) or ImmunoMount mounting medium
(Shandon, Pittsburgh, PA, USA) and Neo-Mount (Merck, Darmstadt, Germany).
12. Axiophot microscope (Zeiss, Oberkochen, Germany).
13. Openlab 3.0. digital image analysis (Improvision, Coventry, UK).
14. Controls: control tissue (positive, negative), IgG, pre-immune sera, peptides.
3. Methods
3.1. Reproductive Staging of Animals (see Notes 1–4)
The analysis of progesterone and RLX content in the peripheral blood (~2
times per week) provide a method of cycle (~28 d) and pregnancy (~144 d)
classification. Application of PGF2α during mid luteal phase (10–15 d luteal
phase) induces luteolysis of the formed corpora lutea and the initiation of new
growing follicles. Eight to ten days later, ovulation of preovulatory follicles
(1–4) occurs, followed by an increase of blood progesterone concentrations
above 10 ng/mL (1). For exact tissue collection, cycle staging is further con-
firmed by transabdominal ultrasound examination (8). This examination is car-
ried out on unshaved as well as unsedated marmoset monkeys for a time period
of approx 10 min. By ultrasound examination, identification of the number of
follicles and corpora lutea is possible, as well as the status of the uterus (preg-
nant vs nonpregnant, see Note 4). RLX content analysis in peripheral blood
allows early pregnancy detection at day 15 of luteal phase (6).
3.2. Tissue Samples
Uteri are collected by hysterectomy from cyclic and pregnant common mar-
moset monkeys (German Primate Centre, Germany) and immediately fixed in
4% buffered formaldehyde up to 8 h or in liquid nitrogen. The following cycle
stages are routinely collected: mid luteal phase (days 8–12 of luteal phase;
nonpregnant) and pregnancy stages (days 17–135 of pregnancy). Samples from
10_Einspanier_111_122_F 114 8/29/05, 11:18 AM

early pregnancy include uterine tissue comprising the implantation site,

whereas samples from days 95–135 of pregnancy consist of separated uterine
and placental tissues (see Notes 2–4).
3.3. RNA Extraction and RT-PCR

1. For reverse-transcription (RT)-polymerase chain reaction (PCR), frozen tissue
(–80°C) is homogenized in RLT lysis buffer containing 0.1% [v/v] β-mer-
captoethanol. Total RNA is extracted using RNeasy Mini Kit following the
manufacturer’s protocol.
2. The yield of total RNA is spectroscopically determined at 260 nm.
3. Quality and quantity of Callithrix RNA is verified after denaturing electrophore-
sis on a 1% (w/v) formaldehyde containing agarose gel followed by ethidium
bromide staining as described elsewhere (9).
4. An amount of 2 µg total RNA, in a 20-µL reaction volume, is reverse transcribed
to obtain cDNA using SuperScript II reverse transcriptase and oligo(dT)12–18
primers. Reaction mixtures and first-strand-synthesis is performed according to
the manufacturer’s protocol.
5. Marmoset-specific PCR fragments encoding for ER, PR, ARO and 18S are
amplified using Taq-polymerase as previously described (9), whereas Pfu-
polymerase is used for 17HSD1 and 17HSD7 (50 pmol of each primer).
6. The PCR is performed for 35 cycles at the optimized annealing temperature of
each primer pair.
7. 10 µL of each PCR product mixed with 2 µL of 6X Loading Dye Solution is run
on 1.5% agarose gels containing 1 µg/mL ethidium bromide. For sizing of PCR
products, the GeneRuler(tm) 100-bp DNA Ladder Plus is used.
8. As a negative control, water is substituted for RNA and used for the RT-PCR.
9. All reactions are performed three times for each RNA preparation.
10. Specificity of RT-PCR products should be checked by subcloning and DNA
sequencing.
11. Amplification of the housekeeping gene 18S rRNA is used to provide internal
standardization and to demonstrate RNA integrity and loading.
12. Results from an RT-PCR analysis is shown in Fig. 1 (see Notes 5–8).
3.4. Immunohistochemistry (see Notes 4 and 9–11)

3.4.1. Pretreatment for Paraffin Sections
1. Uteri are fixed in 4% buffered formaldehyde not longer than 8 h and embedded in
paraffin.
2. Five micron sections are mounted on poly-L-lysine-coated slides (0.01%).
3. Paraffin sections are incubated at 45°C for 1 h, and then dewaxed in xylol two
times for 15 min each, followed by descending ethanol concentrations (2 × 5 min
100% and each 3 min in 90%, 80%, 70%, 50% ethanol) and washed for 5 min in
purified water.
10_Einspanier_111_122_F 115 8/29/05, 11:18 AM

Fig. 1. Uterine expression of estradiol receptor α (ERα), progesterone (PR),

aromatase (ARO), 17β-hydroxysteroid dehydrogenase type 1 (17HSD1), and 17β-
hydroxysteroid dehydrogenase type 7 (17HSD7) at days 8 and 9 of the secretory phase
(nonconceptive), days 29, 35, 98, and 135 of pregnancy in the common marmoset
monkey. Samples did not contain placental tissue. 18S was monitored as housekeep-
ing gene to verify similar RNA amounts within each sample.
4. For steroid receptor detection, the paraffin sections are subjected to an antigen
retrieval protocol, incubating them for 10 min in 10 mM citrate buffer, pH 6.0 at
120°C, then allowing them to cool over 30 min.
3.4.2. Pretreatment for Cryostat Sections

1. Seven micron sections are mounted on poly-L-lysine-coated slides (0.01%).
2. Sections are air-dried under the lamina flow for 2 h followed by 10 min fixation
in 4 % buffered formaldehyde or spray-fixation with buffered formaldehyde.
3. Followed by a quick washing step in purified water.
3.4.3. Immunohistochemistry Protocol

1. Deparaffinized sections are washed two times for 5 min in TBS and incubated in
3% H2O2 for 30–45 min at room temperature in a moist chamber.
2. The tissue sections are then blocked with 10% normal goat or mouse serum for
45 min.
3. After removing the solution from the slide, sections are then incubated with the
primary antiserum overnight at 4°C in a moist chamber, using dilutions of 1:5000
(ARO), 1:1000 (17HSD1 and 17HSD7), 1:500 (ERα and PR) and 1:100
(vimentin and cytokeratin), respectively.
4. The tissue sections are washed twice for 5 min with TBS.
5. Following the washing step, the tissue sections are incubated with biotinylated
goat-anti-rabbit IgG (5 µg/mL in TBS) for 30 min at room temperature and rins-
ing 2 times each 5 min with TBS again.
6. Tissue sections are washed and then incubated with streptavidin-HRP for 30 min
at 37°C in a moist chamber. Followed by rinsing two times for 5 min with TBS.
10_Einspanier_111_122_F 116 8/29/05, 11:18 AM

7. Visualization of the immune reaction is achieved by applying either AEC-Sub-

strate-Chromogen for 20–30 min or DAB-Chromogen for 15 min, followed by
washing with tap water.
8. For nuclear staining, sections can be optional counterstained with Meyer´s hema-
toxylin.
9. Sections are then mounted with Histogel medium or ImmunoMount mounting
medium. After 2 h, drying cover slips are sealed with Neo-Mount.
10. To control for specificity of the antibodies, immunohistochemistry is performed
separately with either immune sera or pre-immune sera at the same dilutions.
Alternatively, when available the primary antiserum can be blocked by 1 h of
preincubation at room temperature with an excess concentration of the peptide
used for immunization. Also positive as well as negative tissue should be used to
test specificity of reaction.
11. Results are analyzed and documented with Axiophot microscope using Openlab
3.0. digital image analysis.
12. Results from immunohistochemical analyses of marmoset uterine tissues are
shown in Fig. 2 (see Notes 9–11).
4. Notes
1. Experiments with marmosets should be approved by the local ethics commission
on animal welfare.
2. In the wild, marmoset monkeys live in social groups of 8–15 members; therefore,
a social structure is recommended for their welfare in captivity. Moreover, single
housing of females often results in variation of cycle length. For reproductive
studies, at least pair housing (one adult male and female) is required. Also, if
collection of tissue from nonpregnant females is necessary, it is recommended
that females be kept with vasectomized or castrated males.
3. The use of trained monkey is advantageous for blood collection, injections, or
ultrasound examination. Training greatly decreases stress. Stress can induce ir-
regular cycles or abortion, resulting from high glucocorticoid levels.
4. Collection of appropriately staged tissue is dependent on accurate endocrinologi-
cal monitoring. The marmoset is a particularly good model as compared with Old
World monkeys because of its sensitivity to PGF2α. PGF2α induces luteolysis and
the initiation of a new wave of follicle development. As ovulation approaches,
more frequent blood collections facilitate accurate determination of ovulation
and the initiation of pregnancy. This requires the availability of rapid assay sys-
tems within 24 h. Hormonal profiles for progesterone, estradiol, chorionic gona-
dotropin, and RLX provide the requisite information for determining cycle and
pregnancy staging. Further confirmation of cycle or pregnancy stage is given by
using ultrasound examination. Detection of pregnancy is possible on day 15 of
luteal phase by the appearance of a double endometrial echo indicating fluid ac-
cumulation in the uterus (8). Taken together, all of these methods enable exact
staging and therefore collection of the marmoset tissues for experimentation on
early pregnancy.
10_Einspanier_111_122_F 117 8/29/05, 11:18 AM

Fig. 2 (see companion CD for color version). Localization of immunoreactive

estradiol receptor α (ERα), progesterone (PR), aromatase (ARO), 17β-hydroxysteroid
dehydrogenase type 1 (17HSD1), and 17β-hydroxysteroid dehydrogenase type 7
(17HSD7), at day 25 of pregnancy. Bars indicate 50 µm. Steroid receptor expression
is located within the cell nucleus, whereas the other immunoreactions are within the
cytoplasm.
10_Einspanier_111_122_F 118 8/29/05, 11:18 AM

5. Marmoset organs and implantation sites in general are of limited size. Accurate
manual dissection under the stereomicroscope is a prerequisite for valid results.
When drawing conclusions from RT-PCR experiments, the origin of RNA
extracted must be clearly defined. Possible “contamination” by surrounding
tissue with a presumed differing gene expression profile has to be considered. It
is recommended to verify RT-PCR data by immunohistochemistry and vice versa.
6. Two alternative methods for disrupting and homogenizing frozen tissues in RLT
buffer containing β-mercaptoethanol (β-Me/RLT) are used. For tissues smaller
than 4 mm3, a conventional rotor-stator homogenizer is used to lyse the tissue
directly in the appropriate volume of buffer β-Me/RLT. For larger tissue samples
or whole organs, a cryostat microtome is used for sectioning into 30-µm slides at
–30°C, which are then lysed in buffer β-Me/RLT. Quick lysis of cryopreserved
samples is important during homogenization in order to inactivate endogenous
RNases.
7. Be cautious of contamination with RNase in processes like RNA isolation and
reverse transcription. The risk of contamination can be decreased by using a dedi-
cated set of automatic pipettors or by using disposal tips with aerosol barrier
filters certified to be free of RNase. Preparation of all solutions and buffers with
RNase-free glassware, diethylpyrocarbonate (DEPC)-treated water, and chemi-
cals reserved for work with RNA is also recommended. Additionally, use
microfuge tubes certified to be free of RNase.
8. Homologous Callithrix sequences are not available; therefore the HUSAR pro-
gram package has been used to deduce PCR primers with heterologous cDNAs
(e.g., from human or mouse). Annealing temperature as well as cycle number is
crucial for the reliability of the resulting transcript determination. More than 40
cycles should never been used to estimate relative mRNA concentrations. All
PCR products generated with heterologous primer pairs should be sequenced.
Newly generated sequences should then be used to design new homologous
primer pairs. For quantification purposes real-time RT-PCR is recommended.
9. The literature describing immunohistological detection of steroid receptors is not
entirely consistent. One contributing factor is the use of different fixation times.
We have analyzed different fixation periods for steroid receptor detection in mar-
moset tissues. We observe a massive decline in steroid receptor expression in
tissues 1 cm3 with fixation durations longer than 24 h, optimal fixation duration
is up to 8 h. Shrinkage or distortions of these fragile tissues can also occur with
too long fixation period.
10. An example of our use of these techniques is described as follows. First, the
temporal dynamics of steroid receptor expression has been analyzed in total RNA
preparations from whole marmoset uteri throughout nonconceptive as well as
conceptive cycles. Transcripts for both, ERα and PR, are expressed throughout
mid luteal phase and entire pregnancy (Fig. 1). Secondly, using immunohistol-
ogy, ERα is strongly expressed within both maternal and fetal compartments
during early pregnancy (day 25), whereas PR is mainly expressed within the
maternal compartment. The expression levels (gene and protein) for both steroid
10_Einspanier_111_122_F 119 8/29/05, 11:18 AM

receptors are relatively constant as pregnancy progresses (Figs. 1 and 2). Examina-
tion of local expression and distribution patterns provides insights into embryo–
uterine interactions during implantation. The absence of apparent differences in
steroid receptor expression between nonconceptive and conceptive uteri suggest
other regulatory factors are involved.
Similarly to ERα, the three enzymes catalyzing the last steps in the synthesis
of estradiol (ARO, 17HSD1, and 17HSD7) are detectable in nonconceptive and
conceptive marmoset uteri by RT-PCR (Fig. 1). 17HSD1 mRNA expression is
minimal and restricted to early pregnancy, whereas 17HSD7 is expressed later in
pregnancy (day 95) and is undetectable by the end of pregnancy. ARO is detected
at day 35 of pregnancy and through the remainder of pregnancy. Transcripts for
enzymes involved in estradiol synthesis are present in the uterus at the time of
implantation (7). Again, the immunohistochemical results for ARO, 17HSD1,
and 17HSD7 are consistent with the RT-PCR analyses (see Fig. 1). There is a
local up regulation of 17HSD1 and 17HSD7 mainly in the fetal compartments
with weak expression in the maternal compartment during early pregnancy (day
25 of pregnancy). ARO is weakly expressed in the fetal as well as in the maternal
compartments during early pregnancy (Fig. 2).
These complementary approaches for monitoring gene and protein expression
provide insights into the etiology of pregnancy failure and potential therapeutic
strategies. The common marmoset is an excellent model to obtain appropriately
staged tissues, which closely mimics important features of human reproduction (4).
11. Fetal vs maternal compartments can be distinguished using vimentin and
cytokeratin immunostaining. Trophoblasts are visualized by using cytokeratin
staining, whereas decidual cells and stromal cells at the implantation side show
positive staining for vimentin.
Acknowledgments
The authors would like to thank the German Primate Centre for the marmo-
set tissue. This work was supported by grants from the Deutsche
Forschungsgemeinschaft [DFG Ei 333/6-3 and Ei333/11-4 (A.E.) and Ei 296/
10-2 (R.E.)].
References
1. Hearn, J. P. (1983) The common marmoset (Callithrix jacchus), in Reproduction
in New World Primates. New Models in Medical Science (Hearn, J. P., ed.). MTP,
Lancaster, UK: pp. 181–215.
2. Einspanier, A. and Gore, M. (2005) Definition of primate model: female fertility,
in The Laboratory Primate: Reproduction Part 1 (Wolfe-Coate, S., ed.). Elsevier.
3. Summers, P. M., Wennink, C. J., and Hodges, J. K. (1985) Cloprostenol-induced
luteolysis in the marmoset monkey (Callithrix jacchus). J. Reprod. Fertil. 73,
133–138.
10_Einspanier_111_122_F 120 8/29/05, 11:18 AM

4. Enders, A. C. and Lopata, A. (1999) Implantation in the marmoset monkey: ex-

pansion of the early implantation side. Anat. Rec. 256, 279–299.
5. ASRM Report (1999) Assisted reproductive technology in the united states: 1996
results generation from the American Society for Reproductive Medicine/Society
for assisted reproductive technology registry. Fertil. Steril. 71, 798–806.
6. Einspanier, A., Nubbemeyer, R., Schlote, S., et al. (1999) Relaxin in the marmo-
set monkey: secretion pattern in the ovarian cycle and early pregnancy. Biol.
Reprod. 61, 512–520.
7. Husen, B., Adamski, J., Brüns, A., et al. (2003). Characterization of 17β-
hydroxysteroid dehydrogenase type 7 in reproductive tissues of the marmoset
monkey. Biol. Reprod. 68, 2092–2099.
8. Oerke, A. K., Einspanier, A., and Hodges, J. K. (1995) Detection of pregnancy
and monitoring patterns of uterine and fetal growth in the marmoset monkey
(Callithrix jacchus) by real time ultrasonography. Am. J. Primatol. 36, 1–13.
9. Gabler, C., Plath-Gabler, A., Einspanier, A., and Einspanier, R.(1998) Insulin-
like growth factors and their receptors are differentially expressed in the oviduct
of the common Marmoset Monkey (Callithrix jacchus) during the ovulatory cycle.
Biol. Reprod. 58, 1451–1457.
10_Einspanier_111_122_F 121 8/29/05, 11:18 AM

10_Einspanier_111_122_F 122 8/29/05, 11:18 AM
Mouse Trophoblasts 123
III
IN VITRO TROPHOBLAST AND PLACENTAL MODEL
SYSTEMS
124 Quinn, Kunath, and Rossant
11
Mouse Trophoblast Stem Cells
Jennifer Quinn, Tilo Kunath, and Janet Rossant
Summary
The trophectoderm is one of the earliest cell types to differentiate in the forming mamma-
lian embryo. It is responsible for the initial implantation and the formation of the trophoblast
components of the placenta, an organ essential for nutrient and waste exchange between the
fetus and its mother. The trophoblast can be modeled in vitro using trophoblast stem cells.
Trophoblast stem cells require fibroblast growth factor (FGF)4, heparin, and contact with
embryonic fibroblasts, or fibroblast-conditioned medium. They grow as tight epithelial colo-
nies, which express markers of the early trophectoderm and have been shown to contribute to
all of the components of the placenta through chimera studies. These cells can be passaged
indefinitely and can be differentiated by removal of FGF4 and fibroblasts and will express
genetic markers of later placental cell types. This chapter will discuss the initial derivation of
trophoblast stem cells from the blastocyst stage, maintenance, differentiation, flow cytometry
and transfection techniques that can be used with these cells.
Key Words: Trophoblast stem cell; trophectoderm; derivation; culture maintenance; differ-
entiation; flow cytometry; transfection.
1. Introduction
In a mouse blastocyst at embryonic day (E) 3.5, the specification of the
trophectoderm and the inner cell mass is the first differentiation to occur. By
E 4.5, there are three cell types: the primitive endoderm, which will form the
visceral and parietal endoderm; the primitive ectoderm, which will form the
embryo proper; and the trophectoderm, which will produce all the trophoblast
tissues (1,2). The trophoblast is essential for survival of the mammalian con-
ceptus because it mediates implantation and ultimately creates the placenta,
which allows nutrient and waste exchange between the fetus and its mother (3).
The outer cells of the blastocyst—the trophectoderm—can be divided into
two distinct components: polar and mural (4). The mural trophectoderm is com-
prised of the cells that are most distal to the inner cell mass. These cells will
differentiate into primary trophoblast giant cells. Giant cells undergo
125
endoreduplication, which results in large polyploid cells (5). The mural tro-
phectoderm and its resulting primary giant cells are important for the initial
implantation of the blastocyst (3). This differentiation continues laterally toward
the border of the inner cell mass (4). The polar trophectoderm is located in direct
contact with the inner cell mass (4). These cells remain diploid and continue to
divide, giving rise to the trophoblast lineage. This includes the extra-embry-
onic ectoderm and the ectoplacental cone and, eventually, the components of
the mature chorioallantoic placenta—the spongiotrophoblast, labyrinth and
giant cell layer (3). This chapter discusses trophoblast stem cells as an in vitro
model of the trophoblast cell lineage.
Embryonic stem cells from the primitive ectoderm are a well-established in
vitro model (6–8). These cells can be genetically manipulated and have pro-
vided insight into the development of the embryo and essential genes involved
in this process (9). Trophoblast stem (TS) cells may be used in a similar fash-
ion to elucidate the mechanism of differentiation and the role of genes and cell
types in the development of the placenta.
Trophoblast stem cells are diploid, permanent, and self-renewing when they
are maintained in stem cell conditions. They express markers of the trophecto-
derm, extra-embryonic ectoderm, and ectoplacental cone (see Note 1). TS cells
can be derived from E 3.5 blastocysts, the extra-embryonic ectoderm from E 6.5
conceptuses, and the chorionic ectoderm from E 7.5 to E 10 embryos (10–12).
Specific mutant TS cell lines can be developed if the gene in question is not
required for stem cell initiation or maintenance. TS cells require: fibroblast
growth factor (FGF)4 , heparin, and embryonic fibroblasts (EMFIs) or embry-
onic fibroblast-conditioned medium (FCM) to maintain their stem cell mor-
phology of tight adherent epithelial colonies. These cells have been shown to
contribute to all trophectoderm derivates through chimera experiments and can
be maintained in culture indefinitely (10).
When stem cell factors are removed, TS cells differentiate and show an in-
crease in expression of genetic markers for the spongiotrophoblast, labyrinth,
and giant cells and decrease in expression of genes from the blastocyst, extra-
embryonic ectoderm, and ectoplacental cone. Ultimately, these cells become ter-
minally differentiated giant cells with large cytoplasm and high ploidy (10).
This chapter describes methods for TS cell derivation from blastocysts,
maintenance, differentiation, fluorescence-activated cell sorting (FACS), and
transfection.
2. Materials
2.1. Embryonic Fibroblasts
1. Dulbecco’s modified Eagle’s medium (DMEM) (Sigma, St. Louis, MO, Cat. Nos.
D2650 and D5879).
2. Trypsin/ethylenediamine tetraacetic acid (EDTA) (Invitrogen Life Technologies,

Carlsbad, CA, cat. no. 25200-056).
3. DMEM + 10% fetal bovine serum (FBS).
4. Mitomycin C (Sigma, cat. no. M0503).
5. 0.45-µm filter (Corning, Acton, MA, cat. no. 430945).
6. TS cell medium: to 500 mL RPMI 1640 + antibiotics (penicillin/streptomycin at
50 µg/mL each final concentration; Invitrogen, cat nos. 61870 or 11875), add the
following:
a. 130 mL FBS, final concentration 20% (Invitrogen).
b. 6.5 mL, 100 mM sodium pyruvate (final concentration 1 mM) (Invitrogen).
c. 6.5 mL, 10 mM β-mercaptoethanol (final concentration 100 µM) (Sigma).
d. 6.5 mL, 200 mM L-glutamine (final concentration 2 mM) (Invitrogen).
7. 2X Freezing medium. 50% FBS, 30% TS medium, 20% dimethylsulfoxide
(DMSO); cool to 4°C.
8. 1000X FGF4 human recombinant FGF4 (Sigma, cat. no. F2278), 25 µg
a. Resuspend lyophilized FGF4 in its vial with 1.0 mL of phosphate-buffered
saline (PBS)/0.1% w/v fraction V bovine serum albumin (BSA).
b. Mix well with P200 and make 10 aliquots of 100 µM into 1.5-mL microfuge
tubes and freeze at –70°C.
c. Thaw each aliquot as needed and store at 4°C; do not re-freeze. 10 mL PBS/
0.1% (w/v) BSA is prepared by dissolving BSA (Sigma, cat. no. A3311) in
PBS without Ca 2+/Mg 2+; filter through a 0.45-µL syringe filter, and make
1-mL aliquots in microfuge tubes; store at –70°C and thaw one tube when
a vial of FGF4 must be reconstituted.
9. 1000X Heparin (Sigma, cat. no. H3149 10,000 U):
a. Resuspend heparin in PBS to a final concentration of 1.0 mg/mL (1000X).
b. Make nine aliquots of 1.1 mL in 1.5 microfuge tubes.
c. Store at –70°C.
d. Thaw aliquots as needed and store at 4°C.
e. Heparin can also be prepared as a 10,000X (10 mg/mL) stock in PBS without
Ca2+/Mg2+ and stored at –70°C. This can be used multiple times to make
batches of 1000X heparin.
10. 70% FCM + F4H. 70% FCM, 30% TS medium, 1/1000 FGF4, 1/1000 heparin.
11. TS Medium + F4H. TS medium, 1/1000 FGF4, 1/1000 heparin.
12. PBS without calcium and magnesium.
13. Tissue culture equipment.
14. Hemocytometer to count cells.
15. Isopropanol freezing container.
2.2. TS Cell Derivation

1. M2 Medium (Specialty Media, Phillipsburg, NJ, MR-015-D).
2. KSOM Medium (Specialty Media, MR-121-D).
3. Pulled glass pipet (12-in.).
4. Mouth pipetting equipment (tubing [one-thirty-second of an inch wall, VWR

Scientific Products, West Chester, PA, cat. no. 62996-350], p1000 pipet tip, cot-
ton plug, aspirator mouth piece).
5. TS cell culture medium.
8. TS Medium + F4H. TS medium, 1/1000 FGF4, 1/1000 heparin..
2.3. TS Cell Maintenance

7. 2X Freezing medium: 50% FBS, 30% TS medium, 20% DMSO, cool to 4°C.
8. Isopropanol freezing container.
2.4. TS Cell Differentiation

2.5. Flow Cytometry

1. 70% Ethanol.
2. Propidium iodide (PI; Molecular Probes, Eugene, OR), 1 mg/mL in water.
3. Triton-X 100, 0.1% final concentration (Sigma).
4. RNase A.
5. PI/Triton X-100 staining solution with RNase A: to 10 mL of 0.1% Triton X-100
in PBS, add 2 mg DNase-free RNase A and 200 µL of 1 mg/mL PI.
6. Polypropylene or polystyrene tubes (5 mL).
7. Hoechst 33342, 1 mg/mL in water.
8. 70-µm Cell Strainer (BD Biosciences, Bedford, MA, Falcon cat. no. 352350).
9. Flow cytometer with 488-nm argon laser.
2.6. Transfection
1. LipofectAMINE PLUS (Gibco-BRL, Gaithersburg, MD, cat. no. 10964-013):
RPMI culture medium, PLUS reagent, lipofectamine.
2. GenePulser cuvet (BioRad, Hercules, CA, cat. no.165-2088).

3. Linearized DNA.
4. PBS.
5. Drugs for selection, e.g., G418 (200 µg/mL), puromycin (1 µg/mL), hygromycin
(150 µg/mL).
3. Methods
The methods detailed in this chapter describe the (1) the isolation and cul-
turing of mouse embryo fibroblasts, (2) derivation of TS cells from blastocysts,
(3) maintenance, and (4) differentiation of TS cells, as well as (5) protocols to
perform flow cytometry to sort cells and analyze DNA content and (6) to per-
form DNA transfection.
3.1. Embryonic Fibroblasts

The protocol for isolating and culturing mouse embryo fibroblasts is based
on the procedure previously described (9).
3.1.1. Isolation and Expansion of EMFI Cultures
1. Dissect one litter of E 15.5 to E 16.5 embryos into PBS.
2. Remove embryos’ limbs, brains, and internal organs.
3. Place the carcasses into 50-mL Falcon tubes with PBS.
4. Rinse with DMEM three times.
5. Aspirate the medium.
6. Mince the embryos into small pieces.
7. Add 10 mL Trypsin/EDTA to minced pieces in 50-mL tube.
8. Add 5 mL of sterile glass beads and a stir bar.
9. Incubate at 37°C for 30 min while stirring.
10. Repeat steps 8–10.
11. Split cell suspension into two 50-mL Falcon tubes, each containing 3 mL FBS.
12. Wash the original Falcon tube twice with DMEM + 10% FBS and add to tubes
from step 11.
13. Centrifuge at 200g for 5 min.
14. Aspirate supernatant.
15. Resuspend pellet in 50 mL of DMEM + 10% FBS.
16. Use trypan blue to quantify viable nucleated cells.
17. Plate 5 × 106 cells per 15-cm tissue culture dish containing DMEM + 10% FBS.
18. Change the medium the next day.
19. Continue feeding every 2 d until cells reach confluency.
20. Split cells 1:6.
21. Continue to feed every 2 d until cells reach confluency.
22. Freeze in chilled 2X Freezing medium (follow steps outlined under Subhead-
ing 3.3.).
3.1.2. EMFI Feeders

1. Thaw one vial of EMFIs in five 15-cm tissue culture dishes containing DMEM +
10% FBS.
2. When cells are confluent (approx 3 d), treat with 100 µL mitomycin C (1 mg/mL)
in DMEM + 10% FBS.
3. Use as feeders.
4. Mitomycin treated EMFIs can be frozen for later use.
3.1.3. Fibroblast-Conditioned Medium

FCM is used to culture TS cells in the absence of EMFIs.
1. Plate mitomycin-treated primary EMFIs in 100-mm dishes (2 × 106 cells; 2 × 105
cells/mL).
2. Culture in approx 11 mL TS medium.
3. Incubate for 72 h.
4. Lift TS media into 14-mL Falcon tubes.
5. Spin at 200g for 4 min to remove floating cells and debris.
6. Filter through 0.45-µm filter.
7. Aliquot.
8. Store at –20°C.
9. Thaw each aliquot as needed and store at 4°C; do not re-freeze.
10. Follow steps 2–8 to prepare two more batches of FCM.
11. Discard the cells. EMFIs are only used up to 10 d after the mitomycin treatment.
3.2. TS Cell Derivation

The initial derivation of trophoblast stem cells is described under Subhead-
ings 3.2.1.–3.2.5. These sections cover the methods for obtaining, isolating,
and culturing blastocysts as well as the early culturing techniques required for
TS cell derivation and maintenance. The time lines indicated are generalized
and can differ between genotypes and even between blastocysts from the same
litter. It is common to observe the cells every 2 d throughout the process of
derivation and maintenance.
3.2.1. Obtaining Blastocysts
TS cells arise from the trophectoderm layer of the blastocyst. Thus, the first
step involved in deriving TS cells is to obtain blastocysts to culture (see Notes
2 and 3).
1. Day 0. Place female mice in estrus with preteased males mid-afternoon. By 1000 h
the following morning, check for a plug to ensure that the mice have mated. This is
E 0.5. It is assumed that the mating took place at 2400 h. Continue to house the
pregnant female with adequate food and water on a 12-h light/dark cycle for
three more days.
2. E 3, AM. In a four-well tissue culture dish, plate 0.5 mL mitomycin-treated EMFIs

at a density of 5 × 104 cells/mL in TS medium + F4H in each well. Place in a
standard incubator at 37°C/5% CO2. A mouth pipet is needed to collect and trans-
fer the blastocysts once they are removed from the mother. To prepare a mouth
pipet, insert mouthpiece into one end of 15 cm of tubing (diameter one-thirty-
second inch), and at the opposite end insert a p1000 pipet tip, filled with cotton
(9) (see Note 4)
3. E 3, PM. Prepare a pipet with a suitable diameter to maneuver the blastocysts. A
12-in, glass pipet must be heated at the junction where the glass thickens until it
becomes slightly pliable. At this point, pull the pipet in one fluid motion until it is
about 50 cm long. Break off so that about 10 cm of the thin diameter glass
remains (9). Store pipet with thin end up in a clean location. The pipet can be
used for several flushings, but should be made fresh each day.
3.2.2. Collecting Blastocysts

The following steps describe the process of removing blastocysts from the
mother on E 3, PM, so that they can be plated in order to derive TS cells.
1. Set up dissection area with a cloth/diaper for the initial dissection, dissection
tools (scissors, forceps), and spray bottle with 70% ethanol.
2. Place KSOM in syringe in 37°C/5%CO2 incubator.
3. Sacrifice female mouse by cervical dislocation (9).
4. Spray abdomen with ethanol to wet fur.
5. Use scissors to cut through skin and body wall.
6. Remove uterus by cutting below oviduct on both horns and above cervix.
7. Place uterine horns in 10-cm dish and view with dissecting microscope.
8. Insert 1-cc syringe with a 20-gauge needle filled with M2 into oviduct end of the
uterus.
9. Flush with approx 0.3 cc M2, then flush same horn from cervix end (see Note 5).
10. Repeat for second uterine horn.
11. Use mouth pipet with small amount of M2 and a few bubbles to control fluid flow
in order to collect and transfer the blastocysts through four drops of M2 (in a 60-
mm tissue culture dish).
12. Use mouth pipet with small amounts of M2 to transfer blastocysts through four
drops of KSOM heated to 37°C (in 60-mm tissue culture dish).
13. Use mouth pipet to place all embryos in center of an organ culture dish in KSOM.
14. Use mouth pipet to transfer one blastocyst per well onto EMFIs with TS med +
F4H (see Note 6 and Fig. 1).
15. Place in incubator. This is day 0 of the blastocyst outgrowth stage. It is normal
for the blastocyst to remain floating in the medium for awhile (see Note 7).
3.2.3. Blastocyst Outgrowth and First Disaggregation

Within 24–36 h after the initial plating, the blastocyst should attach to the
plate and hatch from the zona pellucida. By the third day, in TS cell conditions,
Fig. 1. Fully expanded blastocyst collected by uterine flushing which can be used to
derive trophoblast stem cells. The polar trophectoderm (PT) overlying the inner cell
mass (ICM) will give rise to most of the trophectoderm derivatives and the mural
trophectoderm (MT) will mediate implantation and give rise to primary giant cells.
Phase contrast, scale bar 50 µm.
a blastocyst outgrowth should form. When this outgrowth is disaggregated, it

will allow the formation of stem cell colonies. If there is no outgrowth, or the
blastocyst has not yet attached, the culture still requires fresh media by the
third day after initial plating (see Note 7).
3.2.3.1. DAY 3 (AFTER INITIAL PLATING) FEEDING
1. Remove media by aspiration.
2. Feed with 500 µL TS med + F4H (1/1000 FGF4 stock, 1/1000 heparin stock).
3. Continue every 48 h until outgrowth has reached an appropriate size for disag-
gregation (Fig. 2).
3.2.3.2. DAYS 4–5 (AFTER INITIAL PLATING) DISAGGREGATE

The blastocyst outgrowth should be formed by this time point (see Note 8).
Disaggregation must occur before the outgrowth becomes too large. Figure 2
outlines an appropriate outgrowth size to harvest TS cells efficiently. Beyond
this point, the TS cells will not be derived as efficiently and an endoderm-like
cell type may form in the cultures (11) (Fig. 3) (see Note 9).
1. Select an appropriate outgrowth for disaggregation (between days 3–8).
2. Aspirate media.
3. Rinse plate with 500 µL PBS.
Fig. 2. Blastocyst outgrowth 4 d after initial plating. This outgrowth is an appropri-

ate size for dissaggregation. Note the field of giant cells (GC) surrounding the out-
growth (trophoblast stem [TS] cells) and the embryonic fibroblasts (EMFIs) covering
the plate. Phase contrast, scale bar 50 µm.
4. Add 100 µL trypsin.

5. Place in the incubator for 5 min at 37°C/5% CO2.
6. Stop this reaction with 400 µL 70% FCM + 1.5X F4H.
7. Pipet the contents of the well to near-single cell suspension.
8. Return to the incubator.
3.2.3.3. DAY 6 (AFTER INITIAL PLATING/48 H AFTER DISAGGREGATION) FEEDING

Between 6 and 10 d after the disaggregation, TS cell colonies begin to form.
TS cell colonies grow as tight, flat epithelial sheets and are present in the cul-
ture along with giant cells, which are differentiated TS cells (Fig. 4). These
cells require fresh 70% FCM + 1.5X F4H every 48 h until the plate becomes
approx 50% confluent.
1. Aspirate the media.
2. Replace with 500 µL of 70% FCM + 1.5X F4H.
Fig. 3. Early trophoblast stem (TS) cell colony surrounded by endoderm-like cells.
These round refractile cells are present in TS cell cultures if the blastocyst outgrowth
is allowed to get too large prior to disaggregation. The endoderm-like cells are diffi-
cult to alleviate since they do not require fibroblast-conditioned medium or fibroblast
growth factor-4 to continue to grow. Phase contrast, scale bar 250 µm.
3. This process needs to be repeated every 48 h until the TS cells reach approxi-
mately 50% confluent and require passaging.
3.2.4. First Passage

The colonies that have arisen from the disaggregated outgrowths require
their first passage when they are approx 50% confluent. This typically occurs
between 15 and 25 d after initial disaggregation. The first passage of TS cells is
the most likely time that these cells can differentiate (see Note 10). For each of
the disaggregated outgrowths that require passaging:
1. Prepare a four-well dish (one well per disaggregated outgrowth) by plating EMFIs
(5 × 104 cells/mL) with fresh TS medium + 1.5X F4H (400 µL).
2. Place new plates in a incubator for at least one h to condition the medium and
limit the amount of TS cell differentiation.
3. Remove media from original well by aspirating.
4. Rinse the well with 500 µL PBS.
5. Aspirate PBS.
Fig. 4. Trophoblast stem (TS) cell colony grown on plastic tissue culture dish in TS
cell conditions. The stem cell colony has tight epithelial borders. Giant cells (GC) can
be seen at the edge of the colony. Phase contrast, scale bar 250 µm.
6. Add 100 µL of trypsin/EDTA.

7. Place in the incubator for 5 min.
8. Add TS medium + 1.5X F4H (400 µL) from new plate to stop trypsinization.
9. Pipet up and down to prepare a near single cell suspension.
10. Transfer all of the cells in suspension to a new well in a four-well plate with
EMFIs.
11. Change the medium approx 8 h after this passage.
12. Continue to feed cells every 2 d with 500 µL TS med + 1.5X F4H.
3.2.5. Early Passages (Passages 2–7)

TS cells are still vulnerable to differentiation throughout the early passages.
Allow the cells to become at least 70% confluent between passages (see Note 10).
1. Repeat steps 1–9 as listed under Subheading 3.2.4.
2. Transfer cell suspension to a 14-mL Falcon tube.
3. Spin at 200g for 3 min.
5. Resuspend in 1 mL 70% FCM + F4H.
6. If the majority of the TS cells have differentiated as a result of passaging or the

culture dish is less than 70% confluent and the cells have stopped expanding,
passage to new well (1:2) (proceed to step 9).
7. If the cells have maintained their stem cell morphology and are growing rapidly,
split into four new wells (1:4).
8. Ensure that each well has a total volume of 500 µL 70% FCM + F4H.
9. Continue to feed cultures every 48 h with fresh medium (see Note 10).
10. Passage as required.
11. By passage 7, it is possible to expand the line to a six-well dish (35 mm), which
can be maintained as described under Subheading 3.3.
3.3. TS Cell Maintenance

Subheadings 3.3.1.–3.3.5. cover protocols required for the maintenance and
storage of a stable TS cell line, including feeding, passaging, plating, and freez-
ing and thawing techniques, which can be used on stable TS cell lines (Fig. 4).
A stable line can be defined as one in which at least 70% of the cells main-
tain typical stem cell morphology. It is normal to have some giant cells in any
TS culture, but if they make up the majority of the population of cells, the
culture is not stable and it may not be possible to recover TS cells through a
freeze/thaw cycle. It is also important to note that, because TS cells and giant
cells are adherent, cells floating in the medium are indicative of dead or dying
cells. The protocols in this section can be applied to cells grown on EMFIs or
to cells grown directly on tissue culture plates. They do not require gelatin as
previously demonstrated (10).
3.3.1. Feeding
In order to maintain TS cells, fresh medium and growth factors are required
every 48 h. TS cells plated on EMFIs tend to recover more rapidly from a thaw
or other detrimental conditions. For TS cell analysis, it is beneficial to have a
pure population without EMFIs. Pure populations of TS cells grown on plastic
require 70% feeder conditioned medium (see Note 11).
3.3.1.1. CULTURING TS CELLS ON EMFIS
EMFIs can effectively condition the TS medium for 10 d; thereafter, it is
best to passage the TS cells onto fresh EMFIs. When these cells are on EMFIs,
they require TS med + F4H. If the cells have not reached 70% confluency on
the original EMFIs beyond 10 d, begin feeding with 70% FCM + F4H as de-
scribed under Subheading 3.3.1.2.
1. Prepare fresh TS medium + F4H.
2. Aspirate old medium.
3. Feed TS cells on EMFIs with new TS med + F4H.
4. Place in a standard tissue culture incubator, 37°C/5% CO2.

5. Replace with fresh TS medium + F4H every 48 h up to 10 d after the initial
plating of EMFIs.
3.3.1.2. CULTURING TS CELLS ON TISSUE CULTURE PLASTIC
TS cells grow well on standard tissue culture dishes if the medium is supple-
mented with 70% FCM + F4H. This is an ideal condition for analysis of DNA
content, RNA isolation, or for visualizing the cells with immunohistochemis-
try techniques, because it provides a pure population of TS cells. These steps must
also be followed in the TS cells grown on EMFIs that are more than 10 d old.
1. Prepare 70% FCM + F4H (e.g.,10 mL of 70% FCM +F4H = 7 mL FCM, 3 mL TS
medium, 10 µL 1000X FGF4 stock, 10 µL 1000X heparin stock).
2. Aspirate old medium.
3. Add fresh medium.
3.3.2. Passaging
When the cells reach approx 80–90% confluency, they must be passaged to
a new plate so they can continue to grow and expand. Stable TS cells can be
passaged at 1:20 every 5–7 d. TS cells that differentiate or are slow growing
can be passaged at 1:5. Each line is unique and requires some level of optimi-
zation. To expand the line to a larger surface area, passage no less than 1:7 of
the total cells. Cells can be passaged directly onto plastic tissue culture dishes
or they may be plated onto EMFIs after they have adhered or co-plated with the
EMFIs at the time of passaging.
Density has an effect on cell growth and differentiation. Cells grown or pas-
saged at too high a density can become overcrowded. This results in cell death
and differentiation (see Note 12). Cells plated at too low a density have high
rates of differentiation even when maintained in stem cell feeding conditions.
1. Aspirate media.
2. Rinse with PBS.
3. Aspirate PBS.
4. Add trypsin (one-half volume of medium) (see Note 11).
5. Incubate at 37°C/5% CO2 for 5 min.
6. Tap plate and cells should lift into suspension. If the cells do not lift, continue to
incubate for another 2–5 min.
7. Stop trypsinization with TS medium (same volume as trypsin).
8. Pipet vigorously to attain near single cell suspension.
9. Lift all media to 14-mL Falcon tube.
12. Resuspend pellet of cells with 1 mL TS medium.
13. Plate into new dishes with 70% FCM at an appropriate volume (see Note 11).
3.3.3. Differential Plating

This technique can be used for two main purposes. The first is to remove
EMFIs from TS cell cultures to provide a pure population of TS cells. Second,
differential plating can enrich either stem cells or giant cells.
3.3.3.1. REMOVING EMFIS FROM TS CELL CULTURES
When switching from EMFI cells to plastic dishes with 70% FCM, it may be
desirable to get rid of the EMFI cells immediately. The different adherence
rates of EMFI cells (fast) and TS cells (slow) can be used to obtain a pure TS
cell population. Because some TS cells do settle along with the EMFIs, the
desired passage density may be increased a bit (e.g., 1:14 instead of 1:16).
1. Passage cells to a new plate following the steps under Subheading 3.3.2.
2. Incubate the culture for 1.5 h at 37°C/5% CO2.
3. Remove the supernatant and plate onto another dish.
3.3.3.2. ENRICHING FOR GIANT CELLS OR TS CELLS

Giant cells can be removed from a culture based on their different adherence
rates. Giant cells adhere more quickly and more strongly than stem cells to the
culture dishes. The supernatant should have a reduced population of giant cells
compared with the initial cell population, whereas the cells remaining on the
plate should have an increased proportion of giant cells.
1. Follow the passaging protocol under Subheading 3.3.2.
2. Incubate for 15–30 min at 37°C/5% CO2.
3. Remove supernatant and passage onto another dish.
3.3.4. Freezing/Thawing
TS cells can be frozen for indefinite periods and then later thawed for use.
This allows a certain level of security, because it is not necessary to derive new
TS cells every time one wants a new plate, and it is possible to expand and
store lines of interest for extended periods of time.
3.3.4.1. FREEZING
1. Obtain cells in suspension (1 mL) using the protocol outlined under Subhead-
ing 3.3.2.
2. Add 1 vol of 2X Freezing medium cooled to 4°C.
3. Place 1 mL in a freezing vial.
4. If freezing several lines, keep those in the 2X freezing medium on ice until all
can go in the freezer.
5. Place tubes in an isopropanol slow-freeze container.
6. Slowly freeze in –70°C Freezer for at least 48 h.
7. Transfer to liquid nitrogen.
3.3.4.2. THAWING
To recover a cell line from a frozen vial, culture and passage freshly thawed
cells at least twice to ensure they have recovered sufficiently before beginning
any experiments. After the initial thaw, it is normal to have a large number of
floating cells, some giant cells and some stem cell colonies (see Note 12). If
the plate appears confluent (it may require aspirating the media and rinsing
with PBS to see through the floaters), the cells require a passage. Otherwise,
continue to feed the cells until they reach 80% confluency then follow the pro-
tocol for passaging. Cells often recover more rapidly when they are thawed
onto EMFIs.
1. Remove the vial of cells from liquid nitrogen or –70°C freezer.
2. Warm in 37°C water bath until just thawed, approx 3 min.
3. Use a 1 mL pipet to transfer contents of vial to a 14-mL Falcon tube containing
1 mL TS medium.
5. Aspirate to remove DMSO contained in the freezing medium.
6. Resuspend into an appropriate medium depending on if they are on EMFIs or on
plastic.
7. Plate all cells from vial onto a surface area, which is smaller than the original
plate.
8. 24 h after initial plating aspirate medium (expect many floaters).
9. Rinse with PBS.
10. Add appropriate fresh medium.
As TS cells differentiate in vitro, markers of later cell types of the trophec-
toderm lineage show an increased expression whereas markers of the blasto-
cyst, extra-embryonic ectoderm, and ectoplacental cone show a decrease (see
Note 1). The changes in gene expression are associated with changes in cell
morphology and DNA content. These changes are indicative of cells that are
changing from tight epithelial TS cell colonies to intermediate cell types and
finally to terminal giant cells with expansive cytoplasm and large polyploid
nuclei. Most TS cell cultures will differentiate to predominantly giant cells by
the sixth day of differentiation although some genotypes require a longer pe-
riod of time (Fig. 5).
The rate of differentiation is influenced by several factors, including geno-
type and cell density. It is important to select a consistent density to work with.
TS cell plated at a very low density differentiate very rapidly to giant cells and
do not show an increased expression of intermediate markers of
spongiotrophoblast and labyrinth cells. Typically, a density that results in the
culture becoming nearly confluent by day 6 of differentiation is used (~2 × 105/
60-mm plate). A protocol for the induction of differentiation is presented.
Fig. 5. Differentiated trophoblast stem cells grown on plastic dishes. These cells
have been without fibroblast-conditioned medium + F4H for 6 d and show a character-
istic giant cell morphology with a large cytoplasm and nucleus. Phase contrast, scale
bar 250 µm.
1. Establish the number of plates required for selected time points.

2. Using the same density, plate all initial dishes with TS cells with 70% FCM +
F4H.
4. After 24 h collect the day 0 culture.
5. Prepare day 0 sample for further analysis.
6. Initiate differentiation of remaining plates by removing 70% FCM + F4H (see
Note 13).
7. Rinse briefly with PBS.
8. Aspirate.
9. Add PBS for 5 min.
10. Aspirate.
11. Replace with TS medium.
12. Incubate.
13. Follow steps 7–12 every 2 d throughout the course of the differentiation experiment.
14. Collect plates at time points throughout differentiation (see Note 14).
3.5. Flow Cytometry Analysis and Sorting

FACS sorting by flow cytometry is a method used to measure ploidy (see
Subheading 3.5.1.) and to sort cells by DNA content or other markers (see
Subheading 3.5.2.). This procedure can be done on living or fixed samples. It
can be used to measure the percentage of diploid, tetraploid, and cells with a
ploidy >8 N to quantify the percentage of giant cells in a population (Fig. 5). It
is also an efficient method to sort for diploid TS cells or green fluorescent
protein (GFP)-positive cells to isolate and expand these populations. The ini-
tial steps of these methods are outlined in this section. These methods have
been adapted from procedures described by Darzynkiewicz and Juan (13).
3.5.1. To Analyze Ploidy
1. Collect cells at selected time points throughout differentiation by trypsinizing
entire plate following protocol to passage. If giant cells are abundant at later time
points in differentiation, use a cell scraper to ensure all giant cells are lifted (see
Note 15).
2. Stop trypsinization with TS medium.
3. Lift all cells to a 14-mL Falcon tube. Ensure that all cells have been collected by
rinsing plate with an additional 2 mL TS medium.
4. Count and record cell number.
5. Spin cells at 200g for 4 min.
6. Remove supernatant.
7. Resuspend cells in 500 µL PBS (see Note 16).
8. Add 6 mL 70% ethanol to fix the cells.
9. Store in –20°C freezer until ready to use.
10. When ready to analyze, make fresh PI/Triton X-100 staining solution with
RNase A.
11. Spin cells at 500g for 5 min.
12. Remove ethanol.
13. Resuspend pellet in 5 mL PBS.
14. Let stand for 1 min.
15. Centrifuge cells for 5 min at approx 200g.
17. Suspend cell pellet, 1 × 106 cells/mL based on cell count in step 4 in PI/Triton
X-100 staining solution with RnaseA.
18. Keep at room temperature for 30 min or in incubator at 37°C for 15 min.
19. Filter cells through a 70-µm filter into 5 mL polypropylene or polystyrene test
tubes.
20. Set up and adjust the flow cytometer for excitation with an argon ion laser (488 nm)
and detection of PI emission using a 675 band pass filter.
21. Measure cell fluorescence using pulse peak–pulse area signal to discriminate
between G2 cells and cell doublets.
22. Analyze the ploidy of cells using DNA content frequency histogram (Fig. 6).
Fig. 6. Fluorescence-activated cell sorting profile of undifferentiated trophoblast

stem (TS) cells (day 0) and differentiated TS (day 12 in TS medium without embry-
onic fibroblasts, fibroblast growth factor-4, and Heparin). These data have been col-
lected from a gated channel to exclude cell doublets. The first peak indicates diploid
cells; the second peak shows tetraploid cells. The peaks included within line I are giant
cells with a ploidy greater than or equal to 8 N. There is an increase in total percentage
of high ploidy cells (>8 N) by day 12 of differentiation indicating an increased per-
centage of giant cells.
3.5.2. To Sort Live Cells

A flow cytometer can sort TS cells for GFP expression and for diploid DNA
content, which is indicative of stem cells.
1. Put cells into single cell suspension following the cell passage protocol.
2. Count cells using hemocytometer.
5. Resuspend cells in TS medium (1 × 106 cells/mL).
6. Add Hoechst 33342 staining solution to cell solution for a final dye concentra-
tion of 2–5 µg/mL.
7. Incubate at 37°C for 20–90 min.
8. Strain cells through 70 µm filter into 5 mL polypropylene or polystyrene tubes.
9. Set up and adjust flow cytometer for ultraviolet excitation at 340–380 nm to
detect Hoechst 33342 and GFP; enhanced green fluorescent protein (EGFP)
can be detected at 488 nm.
10. Measure cell fluorescence using pulse width–pulse area signal to discriminate
between G2 cells and cell doublets.
11. Set a gate to sort out cell doublets.

12. Set a gate to sort out cells with DNA ploidy >8 N (giant cells).
13. Single diploid TS cells will be collected in collection tube.
14. Plate cells on EMFIs + TS med + F4H or on to plastic dishes with 70% FCM + F4H.
15. Incubate and continue to maintain as outlined under Subheading 3.3.
3.6. DNA Transfection

DNA transfection can be used to alter the genome of TS cells by introducing
a specifically designed DNA fragment. The new DNA is only incorporated
into a subpopulation of the cells treated. A positive selectable marker gene is
needed to detect successfully transfected cells. Two methods have been used
to transfect TS cells: lipofectamine (see Subheading 3.6.1.) and electro-
poration (see Subheading 3.6.2. and Note 17).
3.5.1. Transient Transfections With Lipofectamine

1. Obtain a plate of subconfluent cells (1 d after passage) in a six-well dish.
2. Combine circular plasmid carrying gene of interest (1 µg) and plasmid contain-
ing reporter gene (e.g., GFP) (0.2 µg) with RPMI 1640 (200 µL without antibiot-
ics) and PLUS reagent (12 µL) (see Note 18).
3. Incubate for 15 min at room temperature.
4. Add RPMI 1640 (190 µL) and Lipofectamine (10 µL) mixture to the DNA/PLUS.
5. Incubate for 15 min at room temperature.
6. Wash TS cells in PBS.
7. Add 800 µL RPMI 1640.
8. Add DNA/PLUS/Lipofectamine solution.
9. Incubate at 37°C/5% CO2 for 3 h.
10. Add 2 mL FCM with 50 ng/mL FGF4 and 2 µg/mL heparin.
12. Change the medium to 70% FCM with F4H.
13. Incubate 24 h.
14. Assay for reporter gene activity.
In transient transfections the GFP reporter construct has been shown to start
expressing GFP 24 h after the transfection. This expression peaks by 48 h and
the amount of protein shows a dramatic decrease by the fifth day after transfection.
3.5.2. Stable Transfections Using Electroporation

1. Obtain a plate of near confluent cells (5 × 106).
2. Prepare plasmid by isolating DNA using a DNA preparation kit (e.g., Qiagen).
3. Linearize vector DNA with a restriction enzyme digest.
4. Switch on electorporation apparatus and set it to electroporate at 0.25V and a
capacitance of 500 uFD.
5. Follow passaging protocol to pellet cells and remove supernatant.
6. Resuspend in PBS (0.8 mL) (see Note 19).

7. Transfer to GenePulser cuvet (0.4 cm electrode).
8. Add linearized DNA (4–25 µg).
9. Electroporate at 0.25 V and a capacitance of 500 uFD.
10. Incubate cells on ice for 20 min.
11. Plate cells on 100 mm plate with 10 mL 70% FCM +F4H.
13. Start suitable drug selection depending on drug resistance gene on plasmid: G418
(200 µg/mL), puromycin (1 µg/mL), hygromycin (150 µg/mL).
14. Incubate.
15. Feed cells with 70% FCM + F4H + selectable marker every 48 h.
16. Pick colonies 12 d later (see Note 20).
4. Notes
1. TS cells show regulation of different genetic markers throughout differentiation.
Table 1 outlines a few of these genes as well as some genes that can be used to
screen colonies for endoderm-like cells, which might be contaminating a culture.
2. The protocols outlined in this chapter are for deriving TS cells from E 3.5 blasto-
cysts. Embryos flushed at E 2.5 and cultured overnight can be used to plate on
EMFIs for TS cell derivation.
3. TS cells have not been successfully derived from C57BL6 mice. Naturally mated
ICR mice can carry 8–15 embryos.
4. Mouth pipets or finger pipets are required to manipulate blastocysts. Pulling pi-
pets is a delicate process—practice first! To control the blastocysts within the
pipet, ensure that there are at least three air bubbles before attempting to pick up.
In order to keep the mouthpiece clean, place the cap of a 14-mL Falcon tube
(with a hole in it) on the tubing below the mouthpiece. The mouth pipet can be
covered when not in use with a cap of a 14-mL Falcon and the main body of the
Falcon tube. Further directions for mouth pipetting can be found on page 177 in
ref. 9.
5. When flushing blastocysts, the uterine horn will bulge when KSOM is added and
a slightly cloudy liquid will emerge from opposite end. It is important not to
squeeze or puncture uterine horn.
6. EMFIs can “condition” the medium for approx 10 d; after that point 70% FCM is
required.
7. If adding new medium before the blastocysts have fully attached, be careful not
to dislodge or aspirate them.
8. Ensure that each blastocyst and subsequent TS line are kept separate from all
others to avoid contamination.
9. Primitive endoderm-like cells are round and highly refractile. They can be found
in TS cultures if the blastocyst outgrowth becomes too large before the initial
dissociation. These cells grow well in TS medium with or without F4H and are
very difficult to remove.
Table 1
Genes Used to Characterize TS Cell Cultures Throughout Differentiation
TS cell expression
Gene Name Expression profile Reference
Cdx2 Trophectoderm E 3.5 TS cell marker 14,15

Eomesodermin Trophectoderm E 3.5 TS cell marker 10,16,17
Extra-embryonic ectoderm (Exe)
Errβ Exe E 5.5 TS cell marker 10,18
Chorion E 8.5
Esx1 Exe Chorion and 19–21
Chorion E 8.5 labyrinth marker
Labyrinth E 9.5
Wnt7b Exe E 7.5 Chorion marker 22
Chorion E 8.5
4311 Ectoplacental cone E 6.5 Spongiotrophoblast 23,24
Spongiotrophoblast E 8.5–18.5 marker
Nodal Spongiotrophoblast E 10 Spongiotrophoblast 25
marker
Gcm1 Chorion E 7.5–9.5 Chorion and
Labyrinth E 9.5–E 17.5 labyrinth marker 26–29
Placental lactogen 1 Primary giant cells E 5–E 12 Giant cell marker 30–32
Placental lactogen II Secondary giant cells E 12–term Giant cell marker 30–34
α Fetoprotein Visceral and parietal endoderm Endoderm marker 35
Indian hedgehog Visceral endoderm Endoderm marker 36
10. Table 2 provides guidelines for the appropriate dish, amount of medium, and
passage requirements for various stages throughout the derivation of TS cell lines.
11. Table 3 provides the area and requirements of commonly used tissue culture
dishes in the maintenance and differentiation of TS cells.
12. Cultures that have a high rate of floating cells should be rinsed thoroughly.
Aspirate the media, rinse with room temperature PBS, and aspirate. Add PBS
for 5 min, aspirate, and feed cells with fresh medium.
13. Cells cannot be differentiated on EMFIs, because they will condition the media
and will inhibit differentiation.
14. Giant cells are very adherent and are difficult to trypsinize. If after 5 min of
trypsinization the cells remain attached, try to dislodge cells by pipetting up and
down in trypsin only before stopping the reaction with TS medium. If the cells
still remain attached, use a cell scraper to dislodge the rest of the cells.
15. Ensure that all cells are in suspension, especially giant cells, which adhere very
strongly to the plate.
Table 2
A Guideline for Passaging at Different Stages Throughout Derivation
Amount Amount
Stage Size of dish of medium to passage
Blastocyst plating Four-well + feeders 500 µL Not applicable

Disaggregation Same four-well 500 µL All
First passage Four-well + feeders 500 µL
All or 1:2
Passage 2–7 Four-well + feeders 500 µL 1:2–1:10
Passage 8 + to expand Four-well–six-well
Six-well + or – feeders 2 mL 1:4–1:7
Six-well (60-mm) 60-mm 5 mL 1:7
Maintaining on 60-mm 60-mm 5 mL 1:10–1:30
Table 3
Tissue Culture Dishes and Conditions
Well Area cm2 Volume to feed Volume of trypsin
Four-well 0.500 mL 0.100 mL

Six-well (35 mm) 9.62 2 mL 1 mL
60-mm 28.27 5 mL 2.5 mL
100-mm 78.54 10 mL 5 mL
16. Cells used in FACS sorting must be in a single cell suspension or they will be lost
in the filtration step.
17. Transient transfections in TS cell occurs with a success rate of approx 1%.
18. In transient transfections, the ratio of reporter plasmid to gene of interest must be
optimized; often, higher concentrations of DNA are helpful.
19. Approximately 50% cell death is expected with optimal transfection efficiency
when PBS is used.
20. On a 10-cm plate, there are often approx 100 drug-resistant colonies present after
12 d.
References
1. Gardner, R. L. (1982) Investigation of cell lineage and differentiation in the ex-
traembryonic endoderm of the mouse embryo. J. Embryol. Exp. Morphol. 68, 175–
198.
2. Snell, G. D. and Stevens, L. C. (1966) Early embryology, in Biology of the Labo-
ratory Mouse (Green, E. L., ed.). McGraw-Hill, New York: pp. 205–245.
3. Rossant, J. and Cross, J. (2002) Extraembryonic lineages, in Mouse Development;

Patterning, Morphogenesis and Organogenesis (Rossant, J. and Tam, P. P., eds.).
Academic, San Diego: pp. 155–180.
4. Dickson, A. D. (1963) Trophoblastic giant cell transformation of mouse
blastocyts. J. Reprod. Fertil. 169, 465–466.
5. Barlow, P. W. and Sherman, M. I. (1972) The biochemistry of differentiation of
mouse trophoblast: studies on polyploidy. J. Embryol. Exp. Morphol. 27, 447–465.
6. Bradley, A., Evans, M., Kaufman, M. H., and Robertson, E. (1984) Formation of
germ-line chimaeras from embryo-derived teratocarcinoma cell lines. Nature 309,
255–256.
7. Evans, M. J. and Kaufman, M. H. (1981) Establishment in culture of pluripotential
cells from mouse embryos. Nature 292, 154–156.
8. Martin, G. R. (1981) Isolation of a pluripotent cell line from early mouse embryos
cultured in medium conditioned by teratocarcinoma stem cells. Proc. Natl. Acad.
Sci. USA 78, 7634–7638.
9. Nagy, A., Gertsenstein, M., Vintersten, K., and Behringer, R. R. (2003) Manipu-
lating the Mouse Embryo, 3rd Ed. (Inglis, J. and Cuddihy, J., eds.). Cold Spring
Harbor Laboratory, Cold Spring Harbor, NY.
10. Tanaka, S., Kunath, T., Hadjantonakis, A. K., Nagy, A., and Rossant, J. (1998)
Promotion of trophoblast stem cell proliferation by FGF4. Science 282, 2072–
2075.
11. Kunath, T. (2003) PhD thesis in Medical and Molecular Genetics, University of
Toronto, Toronto
12. Uy, G. D., Downs, K. M., and Gardner, R. L. (2002) Inhibition of trophoblast
stem cell potential in chorionic ectoderm coincides with occlusion of the ectopla-
cental cavity in the mouse. Development 129, 3913–3924.
13. Darzynkiewics, Z. and Juan, G. (1997) Nucleic acid analysis, in Current Proto-
cols in Cytometry, John Wiley and Sons, New York: pp. 7.5.1–7.5.23.
14. Beck, F., Erler, T., Russell, A., and James, R. (1995) Expression of Cdx-2 in the
mouse embryo and placenta: possible role in patterning of the extra-embryonic
membranes. Dev. Dyn. 204, 219–227.
15. Chawengsaksophak, K., James, R., Hammond, V. E., Kontgen, F., and Beck, F.
(1997) Homeosis and intestinal tumours in Cdx2 mutant mice. Nature 386, 84–87.
16. Ciruna, B. G. and Rossant, J. (1999) Expression of the T-box gene eomesodermin
during early mouse development. Mech. Dev. 81, 199–203.
17. Russ, A. P., Wattler, S., Colledge, W. H., et al. (2000) Eomesodermin is required
for mouse trophoblast development and mesoderm formation. Nature 404, 95–99.
18. Luo, J., Sladek, R., Bader, J. A., et al. (1997) Placental abnormalities in mouse
embryos lacking the orphan nuclear receptor ERR-beta. Nature 388, 778–782.
19. Li, Y., Lemaire, P., and Behringer, R. R. (1997) Esx1, a novel X chromosome-
linked homeobox gene expressed in mouse extraembryonic tissues and male germ
cells. Dev. Biol. 188, 85–95.
20. Li, Y. and Behringer, R. R. (1998) Esx1 is an X-chromosome-imprinted regulator
of placental development and fetal growth. Nat. Genet. 20, 309–311.
21. Cross, J. C. (2000) Genetic insights into trophoblast differentiation and placental
morphogenesis. Semin. Cell. Dev. Biol. 11, 105–113.
22. Parr, B. A., Cornish, V. A., Cybulsky, M. I., and McMahon, A. P. (2001) Wnt7b
regulates placental development in mice. Dev. Biol. 237, 324–332.
23. Lescisin, K.R., Varmuza, S., and Rossant, J. (1988) Isolation and characterization
of a novel trophoblast-specific cDNA in the mouse. Genes. Dev. 2, 1639–1646.
24. Deussing, J., Kouadio, M., Rehman, S., Werber, I., Schwinde, A., and Peters, C.
(2002) Identification and characterization of a dense cluster of placenta-specific
cysteine peptidase genes and related genes on mouse chromosome 13. Genomics
79, 225–240.
25. Ma, G. T., Soloveva, V., Tzeng, S. J., et al. (2001) Nodal regulates trophoblast
differentiation and placental development. Dev. Biol. 236, 124–135.
26. Basyuk, E., Cross, J. C., Corbin, J., et al. (1999) Murine Gcm1 gene is expressed
in a subset of placental trophoblast cells. Dev. Dyn. 214, 303–311.
27. Anson-Cartwright, L., Dawson, K., Holmyard, D., et al. (2000) The glial cells
missing-1 protein is essential for branching morphogenesis in the chorioallantoic
placenta. Nat. Genet. 25, 311–314.
28. Yu, C., Shen, K., Lin, M., et al. (2002) GCMa regulates the syncytin-mediated
trophoblastic fusion. J. Biol. Chem. 277, 50,062–50,068.
29. Stecca, B., Nait-Oumesmar, B., Kelley, K. A., Voss, A. K., Thomas, T., and
Lazzarini, R. A. (2002) Gcm1 expression defines three stages of chorio-allantoic
interaction during placental development. Mech. Dev. 115, 27–34.
30. Colosi, P., Swiergiel, J. J., Wilder, E. L., Oviedo, A., and Linzer, D. I. (1988)
Characterization of proliferin-related protein. Mol. Endocrinol. 2, 579–586.
31. Faria, T. N., Deb, S., Kwok, S. C., Talamantes, F., and Soares, M. J. (1990) On-
togeny of placental lactogen-I and placental lactogen-II expression in the devel-
oping rat placenta. Dev. Biol. 141, 279–291.
32. Shida, M. M., Jackson-Grusby, L. L., Ross, S. R., and Linzer, D. I. (1992) Placen-
tal-specific expression from the mouse placental lactogen II gene promoter. Proc.
Natl. Acad. Sci. USA 89, 3864–3868.
33. Campbell, W. J., Deb, S., Kwok, S. C., Joslin, J. A., and Soares, M. J. (1989)
Differential expression of placental lactogen-II and prolactin-like protein-A in
the rat chorioallantoic placenta. Endocrinology 125, 1565–1574.
34. Hamlin, G. P., Lu, X. J., Roby, K. F., and Soares, M. J. (1994) Recapitulation of
the pathway for trophoblast giant cell differentiation in vitro: stage-specific ex-
pression of members of the prolactin gene family. Endocrinology 134, 2390–2396.
35. Becker, S., Wang, Z. J., Massey, H., et al. (1997) A role for Indian hedgehog in
extraembryonic endoderm differentiation in F9 cells and the early mouse embryo.
Dev. Biol. 187, 298–310.
36. Dziadek, M. and Adamson, E. (1978) Localization and synthesis of
alphafoetoprotein in post-implantation mouse embryos. J. Embryol. Exp.
Morphol. 43, 289–313.
Connexins and Trophoblast Cell Lineage 149
12
Connexins and Trophoblast Cell Lineage Development
Mark Kibschull and Elke Winterhager
Summary
The mouse is a valuable model for studying basic mechanisms of gene regulation in tropho-
blast cell lineage differentiation. Elements of placental development are conserved across spe-
cies, including trophoblast proliferation, differentiation, migration, and vessel invasion. Among
the regulatory processes, direct intercellular communication between trophoblast cells via gap
junction channels seems to play a crucial role in placental development and physiology. Here
we describe in detail the generation of trophoblast stem (TS) cell lines from connexin-deficient
mice. The design of differentiation and proliferation assays are specified including marker gene
sets which are important for analyzing and comparing the differentiation capacity of the
connexin-deficient TS cell lines. Furthermore, we show that TS cells are capable of forming
tumors after subcutaneous injection into nude mice, providing the opportunity to investigate
trophoblast invasion into host vessels in vivo.
Key Words: Connexins; gap junction; placenta; trophoblast stem cells; trophoblast stem
cell tumors.
1. Introduction
Despite the critical role of the placenta in governing the outcome of preg-
nancy, there is limited information available about the molecules involved in
the differentiation of this organ. Failure of appropriate placental development,
especially in the first trimester, is associated with significant complications in
pregnancy, including miscarriage, preeclampsia, and intrauterine growth
restriction (1). The human placenta is difficult to study for several reasons.
Besides the ethical problems of abortion and the availability of sufficient tis-
sues for research, one major point is that the most important steps of tropho-
blast differentiation occur within the first weeks of gestation. The mouse model
is valuable and helpful for studying basic mechanisms of gene regulation in
trophoblast cell lineage differentiation. Placentation in the mouse and human
involves similar cell biological events, including trophoblast proliferation,
differentiation, migration, and vessel invasion. Among the genes regulating these
149
12_Kibschull_149_158_F 149 8/29/05, 11:19 AM

150 Kibschull and Winterhager
processes, the direct intercellular communication between trophoblast cells via gap
junction channels plays a crucial role in placental development and physiology.
Gap junctions are clusters of intercellular membrane channels that connect
the cytoplasm of two neighboring cells. Each cell contributes one half of the
channel (connexon), which is comprised of six connexin (Cx) subunits. The
hydrophilic central pore allows the transfer of ions and small molecules up to 1 kD
including nucleotides (cAMP, cGMP), inositol 1,4,5-trisphosphate (IP3), Ca2+, and
metabolites. The Cx gene family consists of 19 members in the mouse and 20
members in the human genome (2). Connexins show tissue-specific expression
and are temporally regulated during embryonic and placental development.
Gap junctions play an obligate role in cellular and tissue function that has been
proven by generating knockout mice. Recently, evidence is accumulating for a
dual role of gap junctions in signal transduction mediated not only by their
channel properties but, in addition, by the carboxy-terminus (C-terminus) of
the connexin protein (3). It has been shown that the C-terminus is able to inter-
act with other cellular components and that these protein–protein interactions
mediate intracellular signalling (4). Of interest are connexins that lead to a
placental phenotype if deleted such as Cx26, Cx31, and Cx45, providing a
strong rationale for examining the role of gap junctions in placental develop-
ment (5–7).
Cx26 knockout mice die in utero at day 9.5 post conception (pc) when the
chorioallantoic placenta starts to function. The main reason for this early death
in utero is impaired glucose uptake into the fetal compartment (5). Cx26 chan-
nels connect the two layers of syncytiotrophoblast in the labyrinth layer (5).
No obvious changes in trophoblast differentiation could be detected. Thus, the
Cx26 channel seems to serve as a channel for the diffusion of molecules
across the placental barrier but is not involved in trophoblast cell lineage
development.
Cx31 is expressed in the trophectoderm and, subsequently, in early tropho-
blast derivatives (extra-embryonic ectoderm, ectoplacental cone, chorionic ecto-
derm; see ref. 8). In the mature placenta, Cx31 stays expressed in the
spongiotrophoblast throughout pregnancy. If this channel is deleted, a loss of
more than 60% of the embryos between day 10.5 and 13 pc is observed (6).
The placental phenotype revealed a dramatically reduced size of the placenta
on day 9.5 pc with nearly no labyrinth and spongiotrophoblast but an abun-
dance of trophoblast giant cells. Clearly, the missing channel leads to an im-
balance along the trophoblast cell lineage differentiation in favor of enhanced
differentiation to giant cells. This phenomenon is accompanied by an acceler-
ated decline of proliferating trophoblast stem cells in the placenta. Forty per-
cent of the embryos survive because of a placental rescue starting around day
12 pc (6). Reasons for this partial rescue of Cx31-deficient placentas could be
12_Kibschull_149_158_F 150 8/29/05, 11:19 AM

the induction of the Cx43 channel in the spongiotrophoblast at day 10 pc, which
could serve the function of the Cx31 channel (6). The induction of additional
Cx43 channels accompanies the differentiation of spongiotrophoblast cells to
trophoblast giant cells. The terminally differentiated giant cells only express
Cx43 (6). Thus, the different trophoblast populations of the mouse placenta are
defined by a specific connexin expression pattern with Cx26 in the syncytiotro-
phoblast, Cx31 in the spongiotrophoblast, followed by a coexpression with
Cx43, whereas the terminal differentiated giant cells produce Cx43 exclusively.
Of particular interest is the Cx31 channel, because it is expressed in the early
trophoblast cell lineage and its loss is associated with a failure in trophoblast
differentiation (6).
In the past, it was difficult to study molecular mechanisms in early tropho-
blast development because of a lack of appropriate in vitro systems and the
technical problems of isolating trophoblast tissues without contamination with
maternal and embryonic tissues. Progress in investigating signal cascades re-
sponsible for appropriate placental development has been achieved by generat-
ing trophoblast stem (TS) cells. Rossant and her colleagues have established
permanently growing TS cell lines from the mouse blastocyst or the extra-
embryonic ectoderm in the presence of fibroblast growth factor (FGF)4 (see
ref. 9 and Chapter 11 of this volume). Upon removal of FGF4 or addition of
retinoic acid, TS cells are capable of differentiation (9,10). TS cells also effec-
tively develop into all trophoblast cell lineages in vivo, as shown by aggrega-
tion experiments and by blastocyst injection (9,10). Furthermore, this approach
gives the opportunity to establish TS cells from gene-deficient animals, such
as the Cx mutants, to get more insight in the associated sets of signaling mol-
ecules that are in charge of controlling placental differentiation. In comparison
with placental tissue, TS cells provide an easier tool with which to solve cell
and molecular mechanisms, especially for the application of genomic and
proteomic approaches.
To study the effect of a specific connexin in trophoblast differentiation, we
have established Cx26, Cx31, and Cx43 gene-deficient TS cell lines from blas-
tocysts of the corresponding knockout mice. Using these TS cell lines, the in-
fluence of a specific Cx inactivation on differentiation, proliferation, and
invasion capacity of the TS cell lines was investigated. TS cells provide a tool
with which to understand the null phenotype and a means of distinguishing
specific roles in trophoblast lineage development vs mature trophoblast cell
function. For example, Cx31 is implicated in trophoblast cell lineage develop-
ment, whereas Cx26 regulates transplacental transport (11).
In this chapter, the generation of TS cell lines from Cx knockout mice will be
described in detail, because generation of trophoblast stem cells from knockout
blastocysts seems to be accompanied by more methodological problems, espe-
12_Kibschull_149_158_F 151 8/29/05, 11:19 AM

cially if the genes that are deleted alter the differentiation pathway. Here we
describe marker gene sets, which are important to analyze the differentiation
capacity of Cx-deficient TS cell lines. Furthermore, we show that TS cells are
capable of forming tumors after subcutaneous injection into nude mice. These
tumors, unlike embryonic stem (ES) cell tumors, are only transient because of
the rapid differentiation into the invasive pathway (11). This differentiation
results in the formation of trophoblast giant cells that are not proliferative but
normally invade the decidual compartment and the maternal vessels. The dif-
ferentiation of TS cells into invasive giant cells provides the opportunity to
investigate trophoblast invasion into host vessels using the nude mouse model.
2. Materials
1. TS cell medium: RPMI 1640 (Gibco, Karlsruhe, Germany) supplemented with
20% heat-inactivated fetal bovine serum (Biochrom, Berlin, Germany), 1 mM
sodium pyruvate (Gibco, Karlsruhe, Germany), 100 µM β-mercaptoethanol
(Sigma, Munich, Germany), 2 mM L-glutamine (Gibco, Karlsruhe, Germany),
100 U/mL penicillin, and 100 µg/mL streptomycin (Gibco, Karlsruhe, Germany).
2. Mouse embryonic fibroblast (EMFI)-conditioned TS cell medium (EMFI-CM).
3. C-TS cell medium: 75% EMFI-CM, 25% TS cell medium, 25 ng/mL human
recombinant FGF4 (R&D-systems, Wiesbaden, Germany) and 1 µg/mL heparin
(Sigma, Munich, Germany).
4. TS cell freezing medium: C-TS cell medium containing 10% dimethylsulfoxide
(Merck, Darmstadt, Germany).
5. Cell dissociation medium: 1X trypsin-ethylenediamine tetraacetic acid (EDTA)
solution (Gibco, Karlsruhe, Germany) containing 0.25% trypsin and 1 mM
EDTA.
6. Standard protocol for RNA isolation from cell cultures and Northern blotting
equipment.
7. cDNA probes for Northern blotting. GenBank accession numbers are indicated
in parentheses: Cx26 (BC013634), Cx31 (X63099), Cx31.1 (M91236), Cx43
(NM010288), β-actin (X03672), Mash-2 (NM008553), Pl-1(M35662) and Tpbpa
(NM009411).
8. Male athymic nude mice (Han: NMRI nu/nu), 8–12 wk old (Animal Facility of
the University Hospital Essen, Essen, Germany).
3. Methods
3.1. Preparation of Fibroblast-Conditioned TS Medium
1. For preparation of EMFI-CM, 10.5 mL TS cell medium is incubated on confluent
10-cm plates of mitomycin C arrested EMFI for 72 h.
2. The conditioned medium is centrifuged (4000g, 15 min, room temperature), fil-
tered (0.2 µm) and stored at –20°C (see Note 1).
3. The plates of mitomycin C arrested EMFI can be reused for two more rounds of
EMFI-CM preparation.
12_Kibschull_149_158_F 152 8/29/05, 11:19 AM

3.2. Preparation of FGF4 and Heparin Stock Solutions

Stocks of FGF4 (25 µg/mL) and heparin (1 µg/mL) are prepared in 0.1%
bovine serum albumin (BSA)/phosphate buffered saline (PBS), aliquoted (50 µL/
vial), and stored at –80°C. Unfrozen vials are stored at 4°C.
3.3. Culture of TS Cells

Established TS cells are cultured in the undifferentiated state using C-TS
composed of 75% EMFI-CM, 25% TS cell medium, 25 ng/mL FGF4, and 1 µg/
mL heparin. FGF4 and heparin are added to the culture medium immediately
before use. The cells are routinely grown in six-well plates in a humidified
incubator in a 5% CO2 atmosphere. The medium is changed every second day.
3.4. Passaging TS Cells

TS cells are passaged at 80–90% confluency. The monolayer is washed with
PBS and incubated with a trypsin-EDTA solution for 3 min. To stop the enzy-
matic reaction, TS medium is added to the culture well. After centrifugation the
cells are resuspended in C-TS medium and split at a ratio of 1:10 (see Note 2).
3.5. Generation of TS Cells (As In ref. 9, With Some Modifications)

1. Briefly, a single blastocyst is plated in a four-well plate containing 1 mL C-TS
medium and 3 × 104 preplated mitomycin C treated EMFIs (day 0).
2. The typical trophoblast outgrowth formed on day 3 is released by trypsin by add-
ing 100 µL trypsin-EDTA solution for 15 min and disaggregated using a Pasteur
pipet. Subsequently, 1 mL of C-TS medium is added to the well.
3. The next morning, the medium is changed, and the culture fed every second day.
4. From days 6–10 of culturing onward, a few colonies of TS cells appear, which
are passaged (P1) at 30% confluency to six wells containing 1.6 × 105 mitomycin
C-treated EMFIs.
5. For the next passages, the TS cells are dependent on EMFIs.
6. From P5 to P10, TS cells can be cultured without EMFIs but in the presence of
C-TS medium.
7. Exact genotyping of the generated TS cell clones can be performed when the
EMFIs are removed from the culture.
3.6. TS Cell Cryopreservation

1. For cryopreservation of TS cells, 5 × 105 – 1 × 106 cells/mL are resuspended in
C-TS medium containing 10% dimethylsulfoxide.
2. 1-mL vials are placed in a –80°C freezer for 24 h and then transferred into liquid
nitrogen.
3. Frozen vials are rapidly thawed at 37°C and mixed with TS medium.
4. After centrifugation, the cells are resuspended in C-TS medium and plated into
culture plates.
12_Kibschull_149_158_F 153 8/29/05, 11:19 AM


Differentiation of TS cells can be induced by removal of FGF4, heparin, and
EMFI-CM from the culture.
1. For standard differentiation, 40,000 cells per well in C-TS medium are plated
(see Note 3) in a total of five six-well plates (30 single wells).
2. 24 h after plating, the C-TS medium is removed, the cells are washed twice with
PBS, and TS medium is added to induce differentiation (day 0).
3. The TS medium is changed every second day.
4. Sufficient RNA can be isolated, using standard RNA isolation protocols, at 2-d
(d) intervals from the wells: day 0 (six wells), d1 (four wells), d3 (three wells), d5
(three wells), d7 (three wells), d9 (three wells), d11 (four wells), and d13 (four
wells). The differentiation process can be monitored using marker genes such as
Mash2, Pl-1 and Tpbpα that show a coordinated induction of expression (Fig. 1
and refs. 9,11). The expression of connexins can be quantified using northern
blotting and their cellular source determined using immunocytochemistry (6,11).
3.8. TS Cell Proliferation Assay

Connexins are known to influence cell proliferation (2). The effect of a
connexin knockout on the proliferation capacity of TS cells can be analyzed
using an in vitro proliferation assay.
1. Five thousand TS cells are seeded per well in a 12-well plate either in C-TS
(undifferentiating conditions) or in TS medium (differentiating conditions).
2. Cells are collected on days 1, 3, 5, 7, 9, 11, 13, or 15 by treating with trypsin, and
the total cell number per well is determined.
3. For each day, three wells are plated for independent measurements (see Note 4).
3.9. TS Cell Invasion

The invasive capacity of TS cells can be analyzed in vivo by injection into
athymic nude mice (see Note 5).
1. TS cells are grown in C-TS medium in 75-cm2 flasks.
2. Cells are collected at 90% confluency by trypsination, washed in PBS, and 107
TS cells are resuspended in 250 µL of 100% EMFI-CM containing FGF4 and
heparin.
3. Using a 27-gauge needle, the cells are subcutaneously injected into male nude
mice. The knockout TS cells are injected into one flank of the mouse and the
control cells (wild-type or heterozygous cells) are injected into the other flank of
the same mouse.
4. The largest tumor size is reached 7–10 d after injection, and it is fully resorbed
within the next 2–3 wk. The tumor growth and the dimension of the haemorrhagic
lesions are observed within this time period.
12_Kibschull_149_158_F 154 8/29/05, 11:19 AM

Fig. 1. Expression of connexins and trophoblast marker genes during differentia-

tion of a wild-type trophoblast stem (TS) cell line. (A) Total RNA from a differentiat-
ing TS cell line was analyzed by Northern blot using RNA from liver, heart and skin as
a positive control. For the densitometric analysis of trophoblast marker genes (B) and
connexins (C), the intensity of each signal was normalized to the β-actin signal.
4. Notes
1. We routinely use 75% of EMFI-CM to prepare the C-TS medium for growing
and maintaining TS cells in an undifferentiated state. Others report using 80%
(12), 70% (9,10) or 50% of EMFI-CM (13) to keep TS cells undifferentiated. The
optimal percentage should be empirically determined. In our experience, the FBS
and the FGF4 used are the most critical factors for a successful culturing of TS
cells. Several distributors and much FBS should be tested on established TS cell
lines, because some sera lead to very poor proliferation or enhanced spontaneous
differentiation of the cells. In our hands, FGF4 shows the best results on promot-
ing TS cell proliferation. Established TS cell lines may also be cultured using
FGF1 (14,15) or FGF2 (15), but when problems in TS cell cultures arise, we
recommend using FGF4. We did not observe an influence of the plastic ware
from different companies on cell viability.
12_Kibschull_149_158_F 155 8/29/05, 11:19 AM

Fig. 2. Morphology of a trophoblast stem (TS) cell tumor 7 d after subcutaneous

injection of 107 undifferentiated TS cells into the flank of an athymic nude mouse. (A)
Solid tumor with surrounding hemorrhagic lesion; (B) opened blood vessel in the host
skin caused formation of blood filled lacunas; (C) cytokeratin-8 staining marks tro-
phoblast cells inside the tumor. Arrows, trophoblast cells.
12_Kibschull_149_158_F 156 8/29/05, 11:19 AM

2. Cells are routinely passaged at a ratio of 1:10. If knockout TS cells exhibit an

endogenous tendency to differentiate, then lower splitting ratios should be used.
Short exposure to trypsin-EDTA prevents detachment of giant cells arising from
spontaneous TS cells differentiation, because giant cells are relatively insensitive
to trypsin. Thus, short incubation with trypsin-EDTA can be used to “purify” the
TS cell cultures from differentiated cells when passaging.
3. Differentiation of TS cells is dependent on the density of the culture. A low den-
sity leads to enhanced giant cell formation, whereas high-density cultures show
reduced giant cell formation. Therefore, the same cell number should be seeded
for differentiation experiments and proliferation assays so that the phenotype of
different mutated TS cell lines can be compared. It is also recommended to use
the same batch of EMFI-CM for a series of experiments.
4. For proliferation assays a density of 5000 cells per well in a 12-well plate is
recommended. Lower densities (2500 cells per well) result in poor proliferation
because of increased spontaneous giant cell formation. Higher cell numbers lead
to premature confluency and inhibition of proliferation. Under undifferentiating
conditions, the culture reaches confluency after day 9, whereas under differenti-
ating conditions, cell numbers can be analyzed up to day 13.
5. TS cell differentiation is not restricted to giant cell and spongiotrophoblast path-
ways, as indicated by the marker genes Pl-1 and Tpbpa; it occurs to a lesser
extent in syncytiotrophoblast pathways, as indicated by expression of Cx26.
Therefore TS cells may also be used to study the factors which are necessary for
syncytiotrophoblast differentiation.
6. TS cells form transient hemorrhagic tumors up to 1 cm in size after subcutaneous
injection into nude mice (Fig. 2). It is important to inject at least 107 TS cells,
because lesser numbers of cells do not lead to sufficient tumor formation (11,12).
In addition to the solid tumor, a surrounding hemorrhagic lesion is formed. Both
the tumor and the lesions are formed as a result of opened blood vessels in the
host’s skin (Fig. 2B). Like the giant cells during placental development, TS cells
are invasive in the mouse host tissue. This model can be used to study the influ-
ence of a gene knockout on the invasion properties of TS cells. Male nude mice
are used to avoid the effect of female hormones on the trophoblast cells. To account
for individual differences between the nude mice, the knockout TS cells and the
respective control are each injected into one flank of the same nude mouse and
the tumor size of the knockout tumor is normalized to the control tumor size.
Acknowledgments
The authors would like to thank Dr. J. Rossant for providing the cDNA for
Mash-2, Pl-1, and Tpbpa. We also thank Natalie Knipp, Gabriele Sehn, and
Georgia Rauter for excellent technical assistance in developing these methods.
This work was supported by grants from the National Institutes of Health (NIH)
(1R01 HD42558-01), the Deutsche Forschungsgemeinschaft (DFG Wi 774/
10-3), and the Deutscher Akademischer Austauschdienst (DAAD).
12_Kibschull_149_158_F 157 8/29/05, 11:19 AM

References
1. Kingdom, J., Huppertz, B., Seaward, G. and Kaufmann, P. (2000) Development
of the placental villous tree and its consequences for fetal growth. Eur. J. Obstet.
Gynecol. Reprod. Biol. 92, 35–43.
2. Willecke, K., Eiberger, J., Degen, J., et al. (2002) Structural and functional diver-
sity of connexin genes in the mouse and human genome. Biol. Chem. 383, 725–
737.
3. Moorby, C. and Patel, M. (2001) Dual function for Connexins: Cx43 regulates
growth independently of gap junction formation. Exp. Cell. Res. 271, 238–248.
4. Duffy, H. S., Delmar, M., and Spray, D. C. (2002) Formation of the gap junction
nexus: binding partners for connexins. J. Physiol. Paris 96, 243–249.
5. Gabriel, H. D., Jung, D., Bützler, C., et al. (1998) Transplacental uptake of glu-
cose is decreased in embryonic lethal connexin26 deficient mice. J. Cell Biol.
140, 1453–1461.
6. Plum, A., Winterhager, E., Pesch, J., et al. (2001) Connexin31-null mutation in
mice causes transient placental dysmorphogenesis but does not impair hearing
and skin differentiation. Dev. Biol. 231, 334–347.
7. Kruger, O., Plum, A., Kim, J. S., et al. (2000) Defective vascular development in
connexin 45-deficient mice. Development 127, 4179–4193.
8. Dahl, E., Winterhager, E., Reuss, B., Traub, O., Butterweck, A., and Willecke, K.
(1996) Expression of the gap junction proteins connexin31 and connexin43 corre-
lates with communication compartments in extraembryonic tissues and in the gas-
trulating mouse embryo, respectively. J. Cell Sci. 109, 191–197.
Promotion of tropholast stem cell proliferation by FGF4. Science 282, 2072–2075.
10. Yan, J., Tanaka, S., Oda, M., Makino, T., Ohgane, J. and Shiota, K. (2001)
Retinoic acid promotes differentiation of trophoblast stem cells to a giant cell
fate. Dev. Biol. 235, 422–432.
11. Kibschull, M., Nassiry, M., Dunk, C., et al. (2004) Connexin31-deficient tropho-
blast stem cells: a model to analyse the role of gap junction communication in
mouse placental development. Dev. Biol. 273, 63–75.
12. Erlebacher, A., Lukens, A. K., and Glimcher, L. H. (2002) Intrinsic susceptibility
of mouse trophoblasts to natural killer cell-mediated attack in vivo. Proc. Natl.
Acad. Sci. USA 24, 16,940–16,945.
differentiation and placental development. Dev. Biol. 236, 124–135.
14. Uy, G. D., Downs, K. M., and Gardner, R. L. (2002) Inhibition of trophoblast
stem cell potential in chorionic ectoderm coincides with occlusion of the ectopla-
cental cavity in the mouse. Development 129, 3913–3924.
15. Kunath, T., Strumpf, D., Tanaka, S., and Rossant, J. (2001) Trophoblast stem
cells, in Stem Cell Biology. Cold Spring Harbor Laboratory, Cold Spring Harbor,
NY: pp. 267–287.
12_Kibschull_149_158_F 158 8/29/05, 11:19 AM

Studying Trophoblast Cell Differentiation 159
13
Rcho-1 Trophoblast Stem Cells
A Model System for Studying Trophoblast Cell Differentiation
Namita Sahgal, Lindsey N. Canham, Brent Canham,

and Michael J. Soares
Summary
The biology of trophoblast cell development can be investigated using in vitro model sys-
tems. The Rcho-1 trophoblast stem cell line was derived from a rat choriocarcinoma and is an
effective tool for elucidating regulatory mechanisms controlling trophoblast cell differentia-
tion. In this chapter, we describe methods used in the maintenance and manipulation of the
Rcho-1 trophoblast cell line.
Key Words: Trophoblast differentiation; rat placenta; trophoblast giant cells; Rcho-1 tro-
phoblast stem cells; choriocarcinoma.
1. Introduction
Trophoblast cells possess specialized phenotypes and arise from a common
stem cell population directed along a multi-lineage differentiation pathway (1).
Trophoblast stem cells develop from the blastocyst and are maintained by sig-
nals emanating from the inner cell mass (2,3). In the rat, trophoblast stem cells
can be directed toward at least five recognizable differentiated trophoblast cell
phenotypes: trophoblast giant cells, spongiotrophoblast cells, invasive tropho-
blast cells, glycogen cells, and syncytial trophoblast (Fig. 1) (4,5). Differenti-
ated trophoblast cell populations can be distinguished on the basis of
morphology, location, and patterns of gene expression. These cell types are
arranged into two distinct zones of the chorioallantoic placenta—the junctional
zone and the labyrinth zone—and contribute to a complex uteroplacental struc-
ture prominent during the last week of gestation, the metrial gland (Fig. 1).
Each differentiated cell lineage specializes in activities supportive of preg-
nancy, some of which are well established whereas others are the source of
both speculation and ongoing investigation. Some specific trophoblast func-
159
13_Sahgal_159_178_F 159 8/29/05, 11:19 AM

160
13_Sahgal_159_178_F
160
Fig. 1. A schematic representation of rat trophoblast cell lineages and their location within the mature uteroplacental compart-
8/29/05, 11:19 AM
ment. In the rat, trophoblast stem cells can be directed toward at least five recognizable differentiated trophoblast cell phenotypes:
trophoblast giant cells, spongiotrophoblast cells, invasive trophoblast cells, glycogen cells, and syncytial trophoblast. These cell
types are arranged into two distinct zones of the chorioallantoic placenta, the junctional zone and the labyrinth zone; and contrib-
ute to a complex uteroplacental structure prominent during the last week of gestation: the metrial gland.
Sahgal et al.
tions include remodeling uterine vasculature, hormone/cytokine production,

energy storage, and transcellular transport. The normal growth and differentia-
tion of trophoblast cells is crucial for the establishment and maintenance of
pregnancy.
Insights about placental development have been derived from the genera-
tion of mutant mice by gene targeting (6) and through the use of cell culture
models. The latter efforts have been primarily based on two in vitro systems:
blastocyst-derived trophoblast stem cell lines (2) and trophoblast stem cell lines
derived from a rat choriocarcinoma (7–9). The choriocarcinoma derived cell
lines are remarkable in their ability to differentiate into trophoblast pheno-
types.
More than two decades ago, Dr. Shinichi Teshima and his colleagues at the
National Cancer Institute (Tokyo, Japan) induced a transplantable rat chorio-
carcinoma with extraordinary features (7). Initial observations suggested the
trophoblast tumor contained trophoblast giant cells and produced lactogenic
hormones (7,10,11). Subsequently, trophoblast stem cell lines were established
from the same choriocarcinoma by Dr. Michel Vandeputte’s laboratory at the
University of Leuven (Leuven, Belgium) (8) and by our laboratory (9). The
cell line derived by Dr. Vandeputte and colleagues is termed RCHO, while we
refer to our trophoblast stem cell line as Rcho-1. These trophoblast stem cell
lines are aneuploid, are easy to maintain and expand, and possess the capacity
to differentiate in vitro and in vivo into trophoblast giant cells.
RCHO and Rcho-1 trophoblast stem cell lines have become part of the
experimental arsenal for studying trophoblast cell biology (Table 1). These
trophoblast stem cell lines have been used to investigate the regulation of tro-
phoblast cell cycle (12–15), the regulation of trophoblast cell differentiation
(8,9,16–32), the trophoblast cell phenotype (33–47), trophoblast cell-specific
transcriptional regulation (48–67), trophoblast cell transport processes (68–
72), trophoblast cell DNA methylation (73,74), trophoblast cell invasion
(19,75), and trophoblast tumor development (76,77).
The merit of the RCHO and Rcho-1 trophoblast stem cell models is their
plasticity. These cells can be maintained under conditions that facilitate prolif-
eration, or the culture conditions can be changed to promote robust differentia-
tion. Thus, relatively homogenous populations of proliferating and
differentiating trophoblast cells can be retrieved from the cultures. The most
prominent differentiated phenotype observed in RCHO and Rcho-1 tropho-
blast stem cell cultures is the trophoblast giant cell (7,8). This differentiated
phenotype is easy to track by monitoring cell morphology (large nucleus) or a
variety of functional endpoints. The trophoblast giant cell phenotype is also
the most common direction for in vitro differentiation of blastocyst-derived
trophoblast stem cells (2). Differentiation toward other trophoblast cell pheno-
13_Sahgal_159_178_F 161 8/29/05, 11:19 AM

162 Sahgal et al.
Table 1
Rcho-1 Trophoblast Stem Cell Line Applications for Studying Trophoblast
Cell Biology
Trophoblast cellular process References
Regulation of cell cycle regulation 12–15

Regulation of cell differentiation 8,9,16-32
Characterization of trophoblast cell phenotypes 33–47
Trophoblast cell-specific gene transcription 48–67
Cell transport processes 68–72
DNA methylation 73,74
Cell invasion 19,75
Trophoblast tumor development 76,77
types is possible, but is not optimal using classic monolayer culture practices
(Canham, L. N. and Soares, M. J., unpublished results).
Cancer cells, such as those represented by the RCHO and Rcho-1 tropho-
blast stem cell lines, are caricatures of normal development and represent
potentially important models for dissecting molecular mechanisms control-
ling differentiation (78). The key is in identifying and appreciating which regu-
latory pathways are characteristic of normal development and which are
associated with the transformed phenotype. Thus, it is imperative to perform
complementary experimentation using primary cultures of trophoblast cells and
in vivo models.
In this chapter, we describe methods developed in our laboratory for using
the Rcho-1 trophoblast stem cell model to study various aspects of trophoblast
cell biology.
2. Materials
1. Culture media:
a. Standard Growth Medium: RPMI-1640 culture medium (Mediatech Cellgro,
Herdon, VA) containing 50 µM 2-mercaptoethanol (Bio-Rad Laboratories,
Hercules, CA), 1 mM sodium pyruvate (Sigma Chemical Co., St. Louis,
MO), 100 µg/mL penicillin, and 100 U/mL streptomycin (Mediatech
Cellgro), and 20% heat-inactivated fetal bovine serum (FBS, Altanta
Biologicals, Norcross, GA).
b. Standard Differentiation Medium-Type I: NCTC-135 culture medium
(Sigma) containing 50 µM 2-mercaptoethanol (Bio-Rad), 1 mM sodium pyru-
vate (Sigma), 100 µg/mL penicillin and 100 units/mL streptomycin
(Mediatech Cellgro), and 1–10% heat-inactivated donor horse serum (HS;
Atlanta Biologicals).
13_Sahgal_159_178_F 162 8/29/05, 11:19 AM

c. Standard Differentiation Medium-Type II: RPMI-1640 culture medium

(Mediatech Cellgro) containing 50 µM 2-mercaptoethanol (Bio-Rad), 1 mM
sodium pyruvate (Sigma), 100 µg/mL penicillin and 100 U/mL streptomycin
(Mediatech Cellgro), and 1% heat-inactivated donor HS (Atlanta Biologicals).
2. Hank’s balanced salt solution (HBSS; Sigma).
3. Cell Dissociation Medium: Trypsin-ethylenediamine tetraacetic acid (EDTA)
Solution (0.25% Trypsin/0.1% EDTA in HBSS) (Mediatech Cellgro).
4. Cell Freezing and Storage Medium: Standard Growth Medium containing 10%
dimethylsulfoxide (Sigma) and an additional 25% FBS (Atlanta Biologicals).
5. Cryovials (2-mL, Nalge Company, Rochester, NY).
6. StrataCooler® Cryopreservation Module (Stratagene, La Jolla, CA).
7. Phosphate-buffered saline (PBS).
8. Crystal Violet Solution: 5% formalin, 50% ethanol, 150 mM NaCl, and 0.5%
crystal violet (Sigma).
9. TRIzol reagent (Invitrogen Life Technologies, Carlsbad, CA).
10. 1% Formaldehyde-agarose gels. Formaldehyde (Fisher Scientific, Pittsburgh,
PA); agarose (Sigma).
11. Nylon membranes (Nytran Super Charge, Schleicher & Schuell Biosciences, Inc.,
Keene, NH).
12. Crosslinker (Model XL-1000, Spectronics Corporation, Westbury, NY).
13. [α-P32]dCTP (Perkin Elmer, Boston, MA).
14. cDNAs and polyclonal antibodies for monitoring proliferating and differentiat-
ing trophoblast cells (Tables 2 and 3).
15. Androstenedione and progesterone radioimmunoassay kits (Diagnostic Products
Corporation , Los Angeles, CA).
16. Extracellular matrix-coated BioCoat® Matrigel™ Invasion chambers (BD Bio-
sciences, Bedford, MA).
17. Diff-Quick stain for cells (Allegiance Scientific Products, McGaw Park, IL).
18. Lipofectamine reagent and OPTI-MEM Reduced Serum culture medium
(Invitrogen Life Technologies).
19. Geneticin (Sigma) is prepared as a 40X stock solution (10 mg/mL) in HBSS
(Sigma) and stored at 4°C.
20. Holtzman Sprague-Dawley rats are obtained from Harlan Sprague-Dawley (In-
dianapolis, IN).
3. Methods
3.1. Routine Maintenance and Expansion of Rcho-1 Trophoblast Stem
Cells
1. Rcho-1 trophoblast cells are routinely maintained in 75-cm2 flasks in Growth
Medium, in an atmosphere of 5% CO2/95% air at 37°C in a humidified incuba-
tor. Cells are grown under subconfluent conditions. Initially, cells are plated at
1–2 × 106 cells per flask and fed at two day intervals (see Notes 1–3).
13_Sahgal_159_178_F 163 8/29/05, 11:19 AM

164 Sahgal et al.
Table 2
Genes Expressed in Proliferating Rcho-1 Trophoblast Stem Cells
Gene Functional group GenBank accession no. References
Cdx2 Transcription AJ278466 unpublisheda
Eomes Transcription AY457971 unpublisheda
Id-1 Transcription L23148 17 and unpublisheda,b
Mash2 Transcription X53724 17 and unpublisheda
SOCS 3 Signal transduction AF075383 32 and unpublisheda
Cyclin D3 Cell cycle D16309 14 and unpublishedb
Abbreviations: Eomes, Eomesodermin; Id-1, Inhibitor of DNA binding 1; Mash, mammalian
achaete schute; SOCS3, suppressor of cytokine signaling 3.
aSahgal, N., Canham, L. N., and Soares, M. J., unpublished results.
bCanham, L. N., Sahgal, N., and Soares, M. J., unpublished results.
Table 3
Trophoblast Giant Cell-Associated Genes Expressed in Differentiating Rcho-1
Trophoblast Cellsa
GenBank Antibodies: source
Gene accession no. (cat. no.) References
PRL family
PL-I D21103 Chemicon International, 9,13,26,38,44
Temecula, CA (AB1288)
PL-II M13749 Chemicon (AB1289) 9,13,26,38,44
PLP-A NM_017036 Chemicon (AB1290) 9,13,44
PLP-Fα NM_022530 None currently available 42,44
PLP-M NM_053791 None currently available 44
Steroidogenic regulators
P450scc J05156 Chemicon (AB1244, AB1294) 35,36
3β-HSD L17138 None currently available Unpublishedb
P450c17 NM_012753 See references 37
Others
PSG36 M32474 None currently available Unpublishedb
HAND1 NM_021592 Santa Cruz Biotechnology, 17 and unpublishedc
Santa Cruz, CA (sc-9413)
Abbreviations: PRL, prolactin; PL, placental lactogen, PLP, prolactin-like protein; P450scc, side
chain cleavage; P450c17, 17α hydroxylase; 3βHSD, 3β hydroxysteroid dehydrogenase; PSG, preg-
nancy specific glycoprotein.
aThis list of genes reflects the trophoblast giant cell phenotype of the differentiating Rcho-1 tro-
phoblast stem cells.

bCanham, L. N., Sahgal, N., and Soares, M. J., unpublished results.
cSahgal, N., Canham, L. N., and Soares, M. J., unpublished results.
13_Sahgal_159_178_F 164 8/29/05, 11:19 AM

2. After 48 h of culture, 5 mL of Growth Medium is added to each flask.

3. Following an additional 24 h (72 h from the time of initial plating), the culture
medium is removed, cells are washed with HBSS, and then briefly (1–2 min)
exposed to 3–4 mL of Cell Dissociation Medium, followed by vigorous agitation
of the culture flask.
4. Following dissociation of the cells from the culture flask, an equal volume of
Standard Growth Medium is added to inactivate the trypsin-EDTA.
5. Cells are collected by centrifugation, resuspended in Standard Growth Medium,
and re-plated at a splitting ratio of 1 to 3.
6. Under normal conditions the cells are usually passaged at 3-d intervals.
3.2. Cloning by Limiting Dilution (see Note 4)
Limiting dilution strategies can be used to obtain clones of Rcho-1 tropho-
blast stem cells. Cells are harvested and counted with the aid of a hemacytom-
eter. Cells are distributed into 96-well plates at an estimated concentration of
one-half of a cell per well. The number of cells per well should be verified.
Under standard growth conditions, colonies of cells can be observed within a
week of culture in approx 40–50 wells of the 96-well plate. Colony outgrowths
are then harvested and expanded.
3.3. Freezing, Storage, and Retrieval (see Note 5)
Rcho-1 trophoblast stem cells can be routinely frozen, stored frozen in liq-
uid nitrogen, and retrieved for the establishment of new cultures.
1. Cells are harvested and counted with the aid of a hemacytometer.
2. Cells are equilibrated in Cell Freezing and Storage Medium at a concentration of
1–2 × 106 cells/mL.
3. One milliliter aliquots of the cell suspension are then transferred into 2-mL
cryovials.
4. Cryovials are positioned within a StrataCooler® Cryopreservation Chamber pre-
cooled to 4°C.
5. The Cryopreservation Chamber is transferred to –80°C.
6. After 3 d to 3 wk at –80°C, frozen vials are moved to a liquid nitrogen storage
container, where they can be stored indefinitely.
7. Upon retrieval, frozen aliquots should be rapidly thawed at 37°C, washed once in
Standard Growth Medium, and reseeded into culture plates.
3.4. Method to Monitor Trophoblast Cell Proliferation/Survival (13)

(see Note 6)
2. A total of 500 cells per well are transferred in Standard Growth Medium to a
24-well plate.
3. Following cell attachment overnight, the culture medium is replaced and treat-
ments added. Medium is changed as required over the treatment period. Standard
Growth Medium is used as a positive control for maximal growth.
13_Sahgal_159_178_F 165 8/29/05, 11:19 AM

166 Sahgal et al.
4. After a maximum of seven days, the wells are rinsed with PBS, and stained with
Crystal Violet Solution (300 µL/well) for 10 min with agitation.
5. Cell cultures are then washed repeatedly in tap water, and allowed to dry.
6. Crystal violet dye is then eluted with ethylene glycol.
7. Cell density can be quantified by measuring absorbance of each eluate at 600 nm.
In this assay, cell number is directly correlated with absorbance of the cellular
eluates.
3.5. Induction of Trophoblast Cell Differentiation (see Notes 7 and 8)

Trophoblast giant cell differentiation is induced by growing Rcho-1 tropho-
blast stem cells to confluence in Standard Growth Medium and then replacing
the medium with differentiating conditions. High cell density and the absence
of mitogens (removal of FBS) facilitate trophoblast giant cell differentiation.
2. A total of 1–2 × 106 cells in Standard Growth Medium are plated in a 75 cm2
flask.
3. The cells are fed after 48 h with Standard Growth Medium.
4. After another 24 h, one of two protocols can be used to promote differentiation.
5. Protocol I involves replacing the culture medium with Differentiation Medium
Type I. Cultures are re-fed daily and the appearance of giant cells is evident
within 2–4 d (Fig. 2 ). Differentiation is progressive and differentiated cells main-
tained in culture for up to 3 wk.
6. Protocol II involves replacing the culture medium with Differentiation Medium
Type II. Cultures are re-fed daily for 6 to 8 d and then the cells are returned to
Standard Growth Medium with daily changes for another 6 to 8 d. Trophoblast
giant cells are evident as in Protocol I; however, become more robust in size
during the reintroduction of Standard Growth Medium (Fig. 2).
3.6. Methods to Evaluate Trophoblast Cell Differentiation (see Note 9)

Trophoblast differentiation can be assessed by monitoring changes in cell
morphology/endoreduplication, changes in gene expression, the production of
steroid and polypeptide hormones, and invasiveness.
3.6.2. Morphology/Endoreduplication
Differentiated trophoblast giant cells are easy to recognize and distinguish
from undifferentiated trophoblast stem cells. They are large cells with an en-
larged nucleus and prominent nucleoli. These cells arise by endoreduplication
and their DNA content is polyploid. Nuclear size is proportional to DNA con-
tent. Differentiated trophoblast giant cells can be easily quantified by monitor-
ing nuclear size by image analysis (9) or by monitoring cellular DNA content
by flow cytometry (2).
13_Sahgal_159_178_F 166 8/29/05, 11:19 AM

Fig. 2. Morphology of Rcho-1 trophoblast cells at different stages of differentia-

tion. (A) Proliferative phase, containing primarily trophoblast stem cells; (B) cells
induced to differentiate using Protocol I (withdrawal of the mitogen); (C) cells in-
duced to differentiate using Protocol II (withdrawal of mitogens + reintroduction of
fetal bovine serum [FBS]); (D) development of new trophoblast stem cell colonies
following reintroduction of FBS.
3.6.2. Gene Expression

The differentiation status of the Rcho-1 trophoblast stem cells can be rou-
tinely monitored by Northern blotting.
1. Total RNA is extracted from cells using TRIzol reagent, resolved in 1% formal-
dehyde-agarose gels, transferred to nylon membranes, and crosslinked.
2. Blots are probed with α-P32-labeled cDNAs (Tables 2 and 3).
3. cDNA for a housekeeping gene is used to evaluate the integrity and equal loading
of RNA samples (see Note 10).
13_Sahgal_159_178_F 167 8/29/05, 11:19 AM

168 Sahgal et al.
3.6.3. Hormone Production

Steroid and peptide hormones accumulate in conditioned medium accompa-
nying the differentiation of trophoblast giant cells. Progesterone and andros-
tenedione are the two major steroid products. They can be measured with
commercially available radioimmunoassays (35–37). Production of members
of the prolactin family of polypeptide hormones (placental lactogen-I, placen-
tal lactogen-II, and prolactin-like protein-A) are monitored by Western blot-
ting (34).
3.6.4. Invasion (see Note 11)

The invasive phenotype of trophoblast cells can be assessed by determining
the directional movement of cells through an extracellular matrix (75).
1. Rcho-1 trophoblast stem cells are seeded at 5 × 104 per 3 mL in Standard Growth
Medium on the upper chamber of an extracellular matrix-coated BioCoat Matrigel
Invasion chamber.
2. Cells are incubated at 37°C in a water-jacketed incubator set at 5% CO2.
3. The cultures are continued for various durations.
4. Chambers are then removed and the matrix and cells on the upper surface are
scraped and the membrane fixed and stained with Diff-Quick.
5. Chamber membranes are then excised and placed on slides, overlayed in immer-
sion oil, and cells that invaded and attached to the under surface of the chamber
can be counted using a microscope ocular grid.
3.7. DNA Transfection of Rcho-1 Trophoblast Stem Cells

DNA can be transferred into Rcho-1 trophoblast stem cells using liposome-
mediated procedures. Below is a description of our routine transfection protocol.
1. In a six-well plate, seed 2 × 104 cells per well in 2 mL of Standard Growth Medium.
2. After 2–3 d, the cells are then incubated with a DNA/Lipofectamine mixture
(Lipofectamine reagent 10 µL, DNA construct 2 µg, Opti-MEM culture medium
200 µL) at 37°C for 7 h.
3. Following the incubation the DNA/lipofectamine mixture is removed and the
medium is changed to either Standard Growth Medium or Standard Differentia-
tion Medium.
4. The activity of proteins encoded by the transfected DNA can be monitored 48–60 h
following transfection.
5. Stable DNA transfected Rcho-1 trophoblast stem cell sublines can be generated
through the introduction of DNA plasmids containing cassettes for selectable
genes such as those encoding for neomycin resistance. Effective selection for
neomycin resistance generally requires exposure to geneticin at a concentration
of 250 µg/mL for 2 to 3 wk.
13_Sahgal_159_178_F 168 8/29/05, 11:19 AM

3.8. Transplantation and In Vivo Maintenance of Rcho-1 Trophoblast

Stem Cells (see Note 12)
The kidney capsule serves as an effective growth environment for Rcho-1
trophoblast stem cells.
1. Cells are harvested from cultures and counted with the aid of a hemacytometer.
2. Cells (1–5 × 106) are transferred beneath the kidney capsule of 4-wk-old female
rats (we routinely use Holtzman Sprague-Dawley rats) in a volume of 25–40 µL
using a 27-gauge needle and 1-mL syringe.
3. The cells grow rapidly and must be harvested after 10–12 d.
4. Harvested transplants can also be minced and transferred beneath the kidney cap-
sule of additional recipient animals.
5. Rcho-1 trophoblast stem cells transplanted beneath the kidney capsule have the
potential to exhibit both endocrine and invasive phenotypes.
4. Notes
1. We routinely use RPMI-1640 culture medium as a base growth medium. Rcho-1
trophoblast stem cells grow vigorously in RPMI-1640 culture medium but some-
times at the cost of poor pH regulation. We compensate for the lack of pH control
by changing the culture medium more frequently (daily) and/or by supplement-
ing the cultures with HEPES (10–20 mM). High humidity is essential for optimal
Rcho-1 trophoblast stem cell growth. A serum-free system has not been defined
for propagating the Rcho-1 trophoblast stem cells. At this juncture the inclusion
of FBS is essential. We routinely use high concentrations (20%) of FBS, which
the cells appear to prefer. The high FBS concentration may also minimize some
of the variabilities associated with different lots of serum.
2. Cell density is a key for the appropriate maintenance and expansion of the Rcho-
1 trophoblast stem cell line. The most common problem in working with Rcho-1
trophoblast stem cells is the desire to grow them to confluence. Confluence and
proliferation are not compatible. As the cells become more dense, they begin to
spontaneously differentiate or die. The differentiating cells have a more flattened
appearance and will ultimately develop into trophoblast giant cells, whereas the
dead cells lift from the surface of the culture plate. In order to prevent spontane-
ous cell death or differentiation, the Rcho-1 trophoblast stem cells must be pas-
saged as recommended.
3. Rcho-1 trophoblast stem cell cultures are heterogeneous. Both proliferative and
differentiated cells can be observed in expanding cultures. Manipulating various
aspects of the culture procedure can influence the cellular composition of the cell
line. Cell composition can influence growth rates and features of differentiation.
Maintaining the cells at higher densities or any type of significant stress (humid-
ity, pH, CO2 deprivation, and so on) can lead to differentiation (giant cell forma-
tion) or cell death, both of which result in an irreversible termination of the
culture. Harvesting the Rcho-1 trophoblast cells following brief treatment with
13_Sahgal_159_178_F 169 8/29/05, 11:19 AM

170 Sahgal et al.
trypsin-EDTA results in isolation of a population of cells enriched in stem cells.

This procedure also results in the enrichment of differentiated cells (trophoblast
giant cells) that are more adherent and not removed by brief exposure to the
trypsin-EDTA solution. Harvesting the differentiated cells generally requires
more vigorous dissociation methods such as scraping with a rubber policeman.
Unfortunately, the yield of intact trophoblast giant cells by this technique is not
optimal. Consistency in cell culture practices is extremely important in working
with the Rcho-1 trophoblast stem cell line. Variations in culture densities, pas-
saging methods, and splitting ratios significantly influence the phenotype of the
cell line.
4. Rcho-1 trophoblast stem cells grow well at low density, especially in the pres-
ence of culture medium containing 20% FBS, and clonal lines can be easily
derived. The main concern in isolating clonal lines from Rcho-1 trophoblast
stem cells is obtaining a single cell suspension and preventing cell aggregation
during their dispersal into multi-well plates.
5. Freezing, storage, and retrieval of Rcho-1 trophoblast stem cells require consid-
erable care. In recent years, we have increased the concentration of FBS in the
freezing medium, which seems to improve cell viability at retrieval. We are also
careful to rapidly thaw the cells at 37°C and remove the freezing medium by
centrifugation before culture. If performed well, the cultures are revived within
24 h and ready to passage in another 48 h. Nonetheless, retrieval of cultures from
frozen cell aliquots has been our biggest problem in distributing the Rcho-1 tro-
phoblast stem cells to other laboratories. Because of these problems, we rou-
tinely distribute the cells as live cultures.
6. We have described a simple dye-based colorimetric technique for monitoring
cell proliferation. There are many other strategies that can be used (cell counts,
flow cytometry, and so on). However, it is important to appreciate that a key com-
ponent of differentiation in Rcho-1 trophoblast stem cells is endoreduplication,
e.g., DNA synthesis, without karyokinesis and cytokinesis. Thus, strategies for
monitoring Rcho-1 stem cell proliferation that involve monitoring the incorpora-
tion of a nucleotide or nucleotide analog will not discriminate between DNA
synthesis associated with proliferation and differentiation.
7. One of the experimental advantages of the Rcho-1 trophoblast stem cell line is its
capacity to differentiate. We have developed a couple of protocols for enriching
differentiated trophoblast cells. These involve achieving high cell density and
removal of mitogenic factors. We have the most experience in shifting the cells
to an NCTC 135 basal medium containing HS. Morphological and biochemical
indices of trophoblast giant cell differentiation are evident within a few days.
However, we have noted that the size of the trophoblast giant cells that appear in
these cultures is generally much smaller than those appearing spontaneously in
the expanding cells cultured in FBS. Consequently, we have recently imple-
mented a second protocol for differentiation. The new strategy involves cell
expansion, followed by mitogen withdrawal, and then re-introduction of Stan-
dard Growth Medium. Within a few days large trophoblast giant cells appear
13_Sahgal_159_178_F 170 8/29/05, 11:19 AM

throughout the cultures (Fig. 2). As these cultures are maintained in Standard
Growth Medium, colonies of stem cells will also begin to appear. Cells in these
colonies are tightly packed and rise above the surface of the plate. If needed, the
stem cell colonies can be removed by brief trypsinization without detachment of
the differentiated trophoblast giant cells. In both protocols, mitogen withdrawal
is the key. In the absence of FBS, some cells differentiate, others die, and some
stem cells apparently become dormant. The enhanced trophoblast giant cell for-
mation following re-introduction of Standard Growth Medium suggests that
endoreduplication is stimulated by factors present in FBS.
8. Under our culture conditions, Rcho-1 trophoblast stem cell differentiation is most
prominently directed toward the trophoblast giant cell lineage. Giant cell forma-
tion proceeds over time and may be accelerated by re-introduction of FBS con-
taining medium. Evidence for differentiation along other trophoblast cell lineages
(Fig. 1; spongiotrophoblast cells, glycogen cells, syncytial trophoblast, and the
specialized invasive trophoblast cells of the metrial gland) is apparent but gener-
ally modest to minimal. This restricted differentiation to trophoblast giant cells is
likely, at least in part, a reflection of culture conditions rather than developmen-
tal capabilities of the Rcho-1 trophoblast stem cells. We may be able to learn
from differentiation strategies developed for studying embryonic stem cells (79).
Other cell lineages can be detected by monitoring the expression of genes or gene
products specific for spongiotrophoblast cells, syncytial trophoblast, and the spe-
cialized invasive trophoblast cells of the metrial gland (Table 4). Glycogen cells
are generally identified by their accumulation of glycogen. Exposure of differen-
tiating cells to dimethylsulfoxide can inhibit trophoblast giant cell differentiation
and reactivate part of the trophoblast stem cell phenotype (Sahgal, N., Canham,
L., and Soares, M. J., unpublished results).
9. Balzarini and colleagues use alkaline phosphatase enzyme activity as a measure
of differentiation of RCHO trophoblast stem cells (22,25). The assay is simple
and can readily be adapted to a multi-well format. We have not utilized the assay
mainly because alkaline phosphatase is known to be expressed in many cell types
and thus does not reflect a specific measure of trophoblast cells.
10. We have utilized an assortment of different housekeeping genes to monitor RNA
integrity and loading efficiency. These have included β-actin, glyceraldehyde-3'-
phosphate dehydrogenase (G3PDH), β-tubulin, and 28S ribosomal RNA. Some
of these, including G3PDH and β-tubulin are sometimes problematic in that their
expression is affected by cell differentiation or the treatments employed.
11. Aspects of the invasive phenotype can also be monitored by determining the ex-
pression of gelatinase B and/or α1 integrin and through the analysis of gelatinase
B activity in conditioned medium by substrate gel electrophoresis (zymography;
see ref. 75).
12. Rcho-1 trophoblast stem cells can be maintained in vivo by transplantation into
various host tissues. We have routinely used the kidney capsule but these cells
have also been successfully transplanted to other sites, including the liver, cere-
bral ventricles, lungs, testes, and uteri of rats (7,10,11,85–92). In vivo transplan-
13_Sahgal_159_178_F 171 8/29/05, 11:19 AM

172 Sahgal et al.
Table 4
Other Trophoblast Cell Lineage-Specific Gene Markers
Trophoblast Cell lineage Gene name GenBank accession no. References
Spongiotrophoblast PLP-B M31155 80,81
PLP-Fβ AY741310 Unpublisheda
SSP NM_172073 82
Syncytial trophoblast GCM-1 NM_017186 Unpublishedb
Invasive trophoblast PLP-L NM_138527 5,83
PLP-N NM_153738 84
Abbreviations: PLP, prolactin-like protein; SSP, spongiotrophoblast-specific protein;

GCM-1, Glial cell missing-1.
aHo-Chen, J., Bustamante, J. J., and Soares, M. J., unpublished results.
bSahgal, N., Canham, L. N., and Soares, M. J., unpublished results.
tation of the Rcho-1 trophoblast cells has been effectively used to elevate circu-
lating levels of lactogenic hormones. The predominant lactogen expressed by the
transplants appears to be PL-I. Lactogenic and luteotrophic actions on the mam-
mary glands and ovary, respectively, represent effective indicators of systemic
action of the products of the transplants. Please be aware that Rcho-1 trophoblast
cells are potentially capable of producing other peptide and steroid hormones;
thus the physiological consequences of trophoblast stem cell transplantation may
be complex.
Acknowledgments
We would like to thank past and current members of our laboratory for their
efforts in developing and characterizing the methods described in this chapter.
This work was supported by a National Institutes of Health (NIH) KO8 award
to NS (HD42171) and grants from the NIH (HD20676, HD39878, HD48861)
and the Hall Family Foundation.
References
1. Gardner, R. L. and Beddington, R. S. P. (1988) Multi-lineage ‘stem cells’ in the
mammalian embryo. J. Cell Sci., Suppl. 10, 11–27.
Promotion of trophoblast stem cell proliferation by FGF4. Science 282, 2072–2075.
3. Rossant, J. (2001) Stem cells from the mammalian blastocyst. Stem Cells 19,
477–482.
4. Soares, M. J.,Chapman, B. M., Rasmussen, C. A., Dai, G., Kamei, T., and Orwig,
K. E. (1996). Differentiation of trophoblast endocrine cells. Placenta 17, 277–289.
5. Ain, R., Canham, L. N., and Soares, M. J. (2003) Gestation stage-dependent in-
trauterine trophoblast cell invasion in the rat and mouse: novel endocrine pheno-
type and regulation. Dev. Biol. 260, 176–190.
13_Sahgal_159_178_F 172 8/29/05, 11:19 AM

6. Rossant, J. and Cross, J. C. (2001) Placental development: lessons from mouse

mutants. Nat. Rev. Genet. 2, 538–548.
7. Teshima,S., Shimosato, Y., Koide, T., Kuroki, M., Kikuchi, Y., and Aizawa, M.
(1983) Transplantable choriocarcinoma of rats induced by fetectomy and its bio-
logical activities. Gann. 74, 205–212.
8. Verstuyf, A., Sobis, H., Goebels, J., Fonteyn, E., Cassiman, J. J., and Vandeputte,
M. (1990) Establishment and characterization of a continuous in vitro line from a
rat choriocarcinoma. Int. J. Cancer 45, 752–756.
9. Faria, T. N. and Soares, M. J. (1991) Trophoblast cell differentiation: establish-
ment, characterization, and modulation of a rat trophoblast cell line expressing
members of the placental prolactin family. Endocrinology 129, 2895–2906.
10. Verstuyf, A., Sobis, H., and Vandeputte, M. (1989) Morphological and immuno-
logical characteristics of a rat choriocarcinoma. Int. J. Cancer 44, 879–884.
11. Faria, T.N., Deb, S., Kwok, S. C. M., Vandeputte, M., Talamantes, F., and Soares,
M. J. (1990) Transplantable rat choriocarcinoma cells express placental lactogen:
identification of placental lactogen-I immunoreactive protein and messenger ri-
bonucleic acid. Endocrinology 127, 3131–3137.
12. Verstuyf, A., Goebels, J., Sobis, H., and Vandeputte, M. (1993) Influence of dif-
ferent growth factors on a rat choriocarcinoma cell line. Tumour Biol. 14, 46–54.
13. Hamlin, G. P. and Soares, M. J. (1995) Regulation of DNA synthesis in prolifer-
ating and differentiating trophoblast cells: involvement of transferrin, transform-
ing growth factor-β, and tyrosine kinases. Endocrinology 136, 322–331.
14. MacAuley, A., Cross, J. C., and Werb, Z. (1998) Reprogramming the cell cycle
for endoreduplication in rodent trophoblast cells. Mol. Biol. Cell. 9, 795–807.
15. Hattori, N., Davies, T. C., Anson-Cartwright, L., and Cross, J. C. (2000) Periodic
expression of the cyclin-dependent kinase inhibitor p57(Kip2) in trophoblast giant
cells defines a G2-like gap phase of the endocycle. Mol. Biol. Cell. 11, 1037–1045.
16. Balzarini, J., Verstuyf, A., Hatse, S., et al. (1995) The human immunodeficiency
virus (HIV) inhibitor 9-(2-phosphonylmethoxyethyl)adenine (PMEA) is a strong
inducer of differentiation of several tumor cell lines. Int. J. Cancer 61, 130–137.
17. Cross, J. C., Flannery, M. L., Blanar, M. A., et al. (1995) Hxt encodes a basic
helix-loop-helix transcription factor that regulates trophoblast cell development.
Development 121, 2513–2523.
18. Yamaguchi, M., Kawai, M., Kishi,K., and Miyake, A. (1995) Regulation of rat
placental lactogen (rPL)-II secretion: cAMP inhibits rPL-secretion in vitro. Eur.
J. Endocrinol. 133, 342–346.
19. Grummer, R., Hellmann, P., Traub, O., Soares, M. J., and Winterhager, E. (1996)
Regulation of connexin 31 gene expression upon retinoic acid treatment in rat
choriocarcinoma cells. Exp. Cell Res. 227, 23–32.
20. Kamei, T., Hamlin, G. P., Chapman, B. M., Burkhardt, A. L., Bolen, J. B., and
Soares, M. J. (1997) Signaling pathways controlling trophoblast cell differentia-
tion: Src family protein tyrosine kinases in the rat. Biol. Reprod. 57, 1302–1311.
21. Yamamoto, T., Chapman, B. M., and Soares, M. J. (1997) Protein kinase C de-
pendent and independent mechanisms controlling rat trophoblast cell DNA syn-
thesis and differentiation. J. Reprod. Fertil. 111, 15–20.
13_Sahgal_159_178_F 173 8/29/05, 11:19 AM

174 Sahgal et al.
22. Hatse, S., Naesens,L., De Clercq, E., and Balzarini, J. (1998) Potent differentia-
tion-inducing properties of the antiretroviral agent 9-(2-
phosphonylmethoxyethyl)adenine (PMEA) in the rat choriocarcinoma (RCHO)
tumor cell model. Biochem. Pharmacol. 56, 851–859
23. Nakayama, H., Scott, I. C., and Cross, J. C. (1998) The transition to
endoreduplication in trophoblast giant cells is regulated by the mSNA zinc finger
transcription factor. Dev. Biol. 199, 150–163.
24. Kraut, N., Snider, L., Chen, C. M., Tapscott, S. J., and Groudine, M. (1998) Re-
quirement of the mouse I-mfa gene for placental development and skeletal pat-
terning. EMBO J. 17, 6276–6288.
25. Hatse, S., Naesens, L., De Clercq, E., and Balzarini, J. (1999) N6-cyclopropyl-
PMEDAP: a novel derivative of 9-(2-phenylphonylmethoxyethyl)-2,6-
diaminopurine (PMEDAP) with distinct metabolic, antiproliferative, and
differentiation-inducing properties. Biochem. Pharmacol. 58, 311–323.
26. Peters, T. J., Chapman, B. M., Wolfe, M. W., and Soares, M. J. (2000) Placental
lactogen-I gene activation in differentiating trophoblast cells: extrinsic and intrin-
sic regulation involving mitogen-activated protein kinase signaling pathways. J.
27. Scott, I.C., Anson-Cartwright, L., Riley, P., Reda, D., and Cross, J.C. (2000) The
Hand1 basic helix-loop-helix transcription factor regulates trophpblast giant cell
differentiation via multiple mechanisms. Mol. Cell Biol. 20, 530–541.
differentiation and placental development. Dev Biol. 236, 124–135.
29. Parast, M. M., Aeder, S., and Sutherland, A. E. (2001) Trophoblast giant-cell dif-
ferentiation involves changes in cytoskeleton and cell motility. Dev. Biol. 230,
43–60.
30. Kamei, T., Jones, S. R., Chapman, B. M., McGonigle, K., Dai, G., and Soares, M.
J. (2002) Activation and involvement of the phosphatidylinositol 3-kinase/akt-
signaling pathway in the endocrine differentiation of trophoblast cells. Mol.
Endocrinol. 16, 1469–1481.
31. Hinck, L., Thissen, J. P., Pampfer, S., and De Hertogh, R. (2003) Effect of high
concentrations of glucose on differentiation of rat trophoblast cells in vitro.
Diabetologia 46, 276–283.
32. Takahashi, Y., Carpino, N., Cross, J. C., Torres, M., Parganas, E., and Ihle, J. N.
( 2003) SOCS 3: an essential regulator of LIF receptor signaling in trophoblast
giant cell differentiation. EMBO J. 22, 372–384.
33. Verstuyf, A., Fonteyn, E., Sobis, H., and Vandeputte, M. (1992) A rat cytotropho-
blast antigen defined by a monoclonal antibody. Am. J. Reprod. Immunol. 28, 6–11.
34. Hamlin, G. P., Lu, X.-J., Roby, K. F., and Soares, M. J. (1994) Recapitulation of
35. Yamamoto, T., Roby, K. F., Kwok, S. C. M., and Soares, M. J. (1994) Transcrip-
tional activation of cytochrome P450 side chain cleavage enzyme expression dur-
ing trophoblast cell differentiation. J. Biol. Chem. 269, 6517–6523.
13_Sahgal_159_178_F 174 8/29/05, 11:19 AM

36. Yamamoto, T., Chapman, B. M., Clemens, J. W., Richards, J. S., and Soares, M.
J. (1995) Analysis of cytochrome P450 side-chain cleavage gene promoter activa-
tion during trophoblast cell differentiation. Mol. Cell. Endocrinol. 113, 183–194.
37. Yamamoto, T., Chapman, B. M., Johnson, D. C., Givens, C. R., Mellon, S. H.,
and Soares, M. J. (1996) Cytochrome P450 17α-hydroxylase gene expression in
differentiating rat trophoblast cells. J. Endocrinol. 150, 161–168.
38. Dai, G., Imagawa, W., Liu, B., et al. (1996) Rcho-1 trophoblast cell placental
lactogens: complementary DNAs, heterologous expression, and biological activi-
ties. Endocrinology 137, 5020–5027.
39. Shirai, T., Itonori, S., Soares, M. J., Shiota, K., and Ogawa, T. (1996) Changes in
ganglioside composition of the rat choriocarcinoma cell line, Rcho-1.
Glycoconjugate J. 13, 415–421.
40. Chun, J. Y., Han, Y. J., and Ahn, K. Y. (1999) Psx homeobox gene is X-linked
and specifically expressed in trophoblast cells of mouse placenta. Dev. Dyn. 216,
257–266.
41. Hwang, I. T., Lee, Y. H., Moon, B. C., Ahn, K. Y., Lee, S. W., and Chun, J. Y.
(2000) Identification and characterization of a new member of the placental pro-
lactin-like protein-C (PLP-C) subfamily, PLP-Cβ. Endocrinology 141, 3343–
3352.
42. Sahgal, N., Knipp, G. T., Liu, B., Chapman, B. M., Dai, G., and Soares, M. J.
(2000) Identification of two new nonclassical members of the rat prolactin family.
J. Mol. Endocrinol. 24, 95–108.
43. Oda, M., Sun, W., Hattori, N., Tanaka, S., and Shiota, K. (2001) PAL31 expression
in rat trophoblast giant cells. Biochem. Biophys. Res. Commun. 287, 721–726.
44. Dai, G., Lu, L., Tang, S., Peal, M. J., and Soares, M. J. (2002) The prolactin
family miniarray: a tool for evaluating uteroplacental/trophoblast endocrine cell
phenotypes. Reproduction 124, 755–765.
45. Mehta, D. V., Kim, Y. S., Dixon, D., and Jetten, A. M. (2002) Characterization of
the expression of the retinoid-related, testis-associated receptor (RTR) in tropho-
blasts. Placenta 23, 281–287.
46. Morris Buus, R. and Boockfor, F. R. (2004) Transferrin expression by placental
trophoblastic cells. Placenta 25, 45–52.
47. Novak, D., Quiggle, F., and Matthews, J. (2004) Androgen secretion by Rcho-1 cells
is independent of extracellular glutamate concentration. Placenta 25, 548–552.
48. Shida, M. M., Ng, Y.-K., Soares, M. J., and Linzer, D. I. H. (1993) Trophoblast-
specific transcription from the mouse placental lactogen I gene promoter. Mol
49. Vuille, J. C., Cattini, P. A., Bock, M. E., et al. (1993) Rat prolactin-like protein A
partial gene and promoter structure: promoter activity in placental and pituitary
cells. Mol. Cell. Endocrinol. 96, 91–98.
50. Ng, Y. K., George, K. M., Engel, J. D., and Linzer, D. I. H. (1994) GATA factor
activity is required for the trophoblast-specific transcriptional regulation of the
mouse placental lactogen I gene. Development 120, 3257–3266.
13_Sahgal_159_178_F 175 8/29/05, 11:19 AM

176 Sahgal et al.
51. Dai,G., Liu, B., Szpirer, C., Levan, G., Kwok, S. C. M., and Soares, M. J. (1996)
Prolactin-like protein-C variant: complementary DNA, unique six exon gene struc-
ture, and trophoblast cell-specific expression. Endocrinology 137, 5009–5019.
52. Scatena, C. D. and Adler, S. (1996) Trans-acting factors dictate the species-spe-
cific placental expression of corticotropin-releasing factor genes in choriocarci-
noma cell lines. Endocrinology 137, 3000–3008.
53. Orwig, K. E., Dai,G., Rasmussen,C. A., and Soares, M. J. (1997) Decidual/tro-
phoblast prolactin-related protein: characterization of gene structure and cell-spe-
cific expression. Endocrinology 138, 2491–2500.
54. Lin, J. and Linzer, D. I. H. (1998) Identification of trophoblast-specific regulatory
elements in the mouse placental lactogen II gene. Mol. Endocrinol. 12, 418–427.
55. Matsumoto, K., Yamamoto, T., Kurachi, H., et al. (1998) Human chorionic gona-
dotropin-α gene is transcriptionally activated by epidermal growth factor through
cAMP response element in trophoblast cells. J. Biol. Chem. 273, 7800–7806.
56. Shah, P., Sun, Y., Szpirer, C., and Duckworth, M. L. (1998) Rat placental lacto-
gen II gene: characterization of gene structure and placental expression. Endocri-
nology 139, 967–973.
57. Sun, Y., Robertson, M. C., and Duckworth, M. L. (1998) The effects of epidermal
growth factor/transforming growth factor-α on the regulation of placental lacto-
gen I and II mRNAs in a rat choriocarcinoma cell line. Endocr. J. 45, 297–306.
58. Dai, G., Wolfe, M. W., and Soares, M. J. (1999) Distinct regulatory regions from
the prolactin-like protein-C variant promoter direct trophoblast giant cell versus
spongiotrophoblast cell-specific expression. Endocrinology 140, 4691–4698.
59. Liang, R., Limesand, S. W., and Anthony, R. V. (1999) Structure and transcriptional
regulation of the ovine placental lactogen gene. Eur. J. Biochem. 265, 883–895.
60. Orwig, K. E. and Soares, M. J. (1999) Transcriptional activation of the decidual/
trophoblast prolactin-related protein gene. Endocrinology 140, 4032–4039.
61. Sun, Y. and Duckworth, M. L. (1999) Identification of a placental-specific en-
hancer in the rat placental lactogen II gene that contains binding sites for mem-
bers of the Ets and AP-1 (activator protein 1) families of transcription factors.
Mol. Endocrinol. 13, 385–399.
62. Okamoto, Y., Sakata, M., Yamamoto, T., et al. (2001) Involvement of nuclear tran-
scription factor Sp1 in regulating glucose transporter-1 gene expression during rat
trophoblast differentiation. Biochem. Biophys. Res.Commun. 288, 940–948.
63. Yamamoto, T., Matsumoto, K., Kurachi, H., et al. (2001) Progesterone inhibits tran-
scriptional activation of human chorionic gonadotropin-alpha gene through protein
kinase A pathway in trophoblast cells. Mol. Cell. Endocrinol. 182, 215–224.
64. Szafranska, B., Miura, R., Ghosh, D., et al. (2001) Gene for porcine pregnancy-
associated glycoprotein 2 (poPAG2): its structural organization and analysis of its
promoter. Mol. Reprod. Dev. 60, 137–146.
65. Gabriel, H. D., Strobl, B., Hellmann, P., Buettner, R., and Winterhager, E. (2001)
Organization and regulation of the rat Cx31 gene. Implication for a crucial role of
the intron region. Eur. J. Biochem. 268, 1749–1759.
13_Sahgal_159_178_F 176 8/29/05, 11:19 AM

66. Firulli, B. A., Howard, M. J., McDaid, J. R., et al. (2003) PKA, PKC, and the
protein phosphatase 2A influence HAND factor function: a mechanism for tissue-
specific transcriptional regulation. Mol. Cell 12, 1225–1237.
67. Limesand, S. W., Jeckel, K. M., and Anthony, R. V. (2004) Pur alpha, a single-
stranded deoxyribonucleic acid binding protein, augments placental lactogen gene
transcription. Mol. Endocrinol. 18, 447–457.
68. Hershberger M. E. and Tuan, R. S. (1998) Placental 57-kDa Ca(2+)-binding pro-
tein: regulation of expression and function in trophoblast calcium transport. Dev.
Biol. 199, 80–92.
69. Hershberger, M. E. and Tuan, R. S. (1999) Functional analysis of placental 57-
kDa Ca(2+)-binding protein: overexpression and downregulation in a trophoblas-
tic cell line. Dev. Biol. 215, 107–117.
70. Knipp, G. T., Liu, B., Audus, K. L., Fujii, H., Ono, T., and Soares, M. J. (2000)
Fatty acid transport regulatory proteins in the developing rat placenta and in tro-
phoblast cell culture models. Placenta 21, 367–375.
71. Evans, T. J., James-Kracke, M. R., Kleiboeker, S. B., and Casteel, S. W. (2003)
Lead enters Rcho-1 trophoblastic cells by calcium transport mechanisms and com-
plexes with cytosolic calcium-binding proteins. Toxicol. Appl. Pharmacol. 186,
77–89.
72. Novak, D. and Matthews, J. (2003) Glutamate transport by Rcho-1 cells derived
from rat placenta. Pediatr. Res. 53, 1025–1029.
73. Kimura, H., Takeda, T., Tanaka, S., Ogawa, T., and Shiota, K. (1998) Expression
of rat DNA (cytosine-5) methyltransferase (DNA MTase) in rodent trophoblast
giant cells: molecular cloning and characterization of rat DNA MTase. Biochem.
Biophys. Res. Commun. 253, 495–501.
74. Ohgane, J., Hattori, N., Oda, M., Tanaka, S., and Shiota, K. (2002) Differentia-
tion of trophoblast lineage is associated with DNA methylation and demethylation.
Biochem. Biophys. Res. Commun. 290, 701–706.
75. Peters, T. J., Albieri, A., Bevilaqua, E., et al. (1999) Differentiation-dependant
expression of gelatinase B/ matrix metalloproteinase-9 in trophoblast cells. Cell
Tiss. Res. 295, 287–296.
76. Hatse, S., Naesens, L., Degreve, B., et al. (1998) Potent antitumor activity of the
acyclic nucleoside phosphonate 9-(2-phosphonylmethoxyethyl)adenine in chorio-
carcinoma-bearing rats. Int. J. Cancer 76, 595–600.
77. Naesens, L., Hatse, S., Segers, C., et al. (1999) 9-(2-phosphonylmethoxyethyl)-
N6-cyclopropyl-2,6-diaminopurine: a novel prodrug of 9-(2-
phosphonylmethoxyethyl)guanine with improved antitumor efficacy and
selectivity in choriocarcinoma-bearing rats. Oncol. Res. 11, 195–203.
78. Pierce, G. B. (1985) Carcinoma is to embryology as mutation is to genetics. Am.
Zool. 25, 707–712.
79. Rudnicki, M. A. and McBurney, M. W. (1987) Cell culture methods and induc-
tion of differentiation of embryonal carcinoma cell lines, in Teratocarcinomas
and Embryonic Stem Cells: A Practical Approach, IRL, Oxford, UK: pp. 19–29.
13_Sahgal_159_178_F 177 8/29/05, 11:19 AM

178 Sahgal et al.
80. Duckworth, M. L., Schroedter, I. C., and Friesen, H. G. (1990) Cellular localiza-
tion of rat placental lactogen-II and rat prolactin-like proteins A and B by in situ
hybridization. Placenta 11, 143–155.
81. Cohick, C. B., Xu, L., and Soares, M. J. (1997) Prolactin-like protein-B: heterolo-
gous expression and characterization of placental and decidual species. J.
82. Iwatsuki, K. Shinozaki, M., Sun, W., Yagi, S., Tanaka, S., and Shiota, K. (2000)
A novel secretory protein produced by rat spongiotrophoblast. Biol. Reprod. 62,
1352–1359.
83. Toft, D. J. and Linzer, D. I. (2000) Identification of three prolactin-related hor-
mones as markers of invasive trophoblasts in the rat. Biol. Reprod. 63, 519–525.
84. Wiemers, D. O., Ain, R., Ohboshi, S., and Soares, M. J. (2003) Migratory tropho-
blast cells express a newly identified member of the prolactin gene family. J.
85. Tomogane, H., Mistry, A. M., and Voogt, J. L. (1991) Late pregnancy and rat
choriocarcinoma cells inhibit nocturnal prolactin surges and serotonin-induced
prolactin release. Endocrinology 130, 23–28.
86. Arbogast, L. A., Soares, M. J., Tomogane, H., and Voogt, J. L. (1992) A tropho-
blast-specific factor(s) suppress circulating prolactin levels and increases tyrosine
hydroxylase activity in tuberoinfundibular dopaminergic neurons. Endocrinology
131, 105–113.
87. Mathiasen, J. R., Tomogane, H., and Voogt, J. L. (1992) Serotonin-induced de-
crease in hypothalamic tyrosine hydroxylase activity and corresponding increase
in prolactin release are abolished at midpregnancy and by transplants of rat cho-
riocarcinoma cells. Endocrinology 131, 2527–2532.
88. Arbogast, L. A., Soares, M. J., Robertson, M. C., and Voogt, J. L. (1993) A tro-
phoblast-specific factor increases tyrosine hydroxylase activity in fetal hypotha-
lamic cell cultures. Endocrinology 133, 111–120.
89. Tomogane, H., Arbogast, L. A., Soares, M. J., Robertson, M. C., and Voogt, J. L.
(1993) A factor(s) from a rat trophoblast cell line inhibits prolactin secretion in
vitro and in vivo. Biol. Reprod. 48, 325–332.
90. Sobis, H., Waer, M., and Vandeputte, M. (1996) Normal and malignant tropho-
blasts do not recruit granulated metrial gland cells. Tumour Biol. 17, 13–19.
91. Flietstra, R. J. and Voogt, J. L. (1997) Lactogenic hormones of the placenta and
pituitary inhibit suckling-induced prolactin (PRL) release but not the ante-partum
PRL surge. Proc. Soc. Exp. Biol. Med. 214, 258–264.
92. Lee, Y. and Voogt, J. L. (1999) Feedback effects of placental lactogens on prolac-
tin levels and Fos-related antigen immunoreactivity of tuberoinfundibular dopam-
inergic neurons in the arcuate nucleus during pregnancy in the rat. Endocrinology
140, 2159–2166.
13_Sahgal_159_178_F 178 8/29/05, 11:19 AM

Bovine Trophoblast Cell Culture 179
14
Bovine Trophoblast Cell Culture Systems
A Technique to Culture Bovine Trophoblast Cells
Without Feeder Cells
Kazuyoshi Hashizume, Arata Shimada, Haruo Nakano,

and Toru Takahashi
Summary
Bovine trophoblastic cells are the first cells to differentiate during embryogenesis and play
pivotal role in morphological and physiological development of the placenta. We have devel-
oped culture systems for bovine trophoblast stem cells isolated from in vitro fertilized blasto-
cysts in the absence of feeder cells. These cells continuously proliferate in Dulbecco’s modified
Eagle’s/F12 culture medium supplemented with bovine endometrial fibroblast-conditioned
medium. The cells possess epithelial morphology, express cytokeratin, and form dome-like
structures (vesicles). Methods for the maintenance, subculture, storage, and measurement of
bovine trophoblast stem cell growth are described. The cells exhibit characteristics of bovine
trophoblastic stem cells and possess the ability to differentiate into binucleate cells and express
placental lactogen, prolactin-related protein-1, pregnancy-associated glycoprotein-1, and inter-
feron τ.
Key Words: Trophoblastic cell line; BT-1; binucleate cells; trophoblast differentiation; pla-
cental lactogen; collagen gel; microarray; gene expression; bovine.
1. Introduction
Trophectoderm is the first cell type to differentiate from the embryo at the
blastocyst stage and its cell lineage contributes to placental formation. Factors
controlling early decisions in the development of inner cell mass (ICM) and
trophectoderm cell lineages are not completely understood. Embryonic stem
cells derived the ICM are pluripotent, whereas trophoblast stem cells have a
more restricted developmental capacity (1). Some trophoblast cell lines have
been developed in various species and have been used for cell differentiation
studies (2–4). Fibroblast growth factor (FGF)4 has a critical role in maintain-
ing mouse trophoblast stem cells in an undifferentiated status.
179
14_Hashizume_179_188_F 179 8/29/05, 11:19 AM

180 Hashizume et al.
We have generated bovine trophoblast stem cell populations from in vitro

cultured blastocysts supplemented with fibroblast cell culture conditioned media
(5–8). One of our trophoblast stem cell lines is referred to as bovine tropho-
blast (BT)-1. BT-1 trophoblast cells are not dependent upon the addition of
exogenous FGF4 for maintenance of their stem cell state. These cells have the
ability to differentiate into bovine placental-specific binucleate cells on col-
lagen substrata. The expression of trophoblast cell-specific transcripts, including
placental lactogen (PL), prolactin-related proteins (PRPs), pregnancy-associated
glycoproteins (PAG), and interferon (IFN)-τ , were confirmed by reverse-tran-
scription (RT)-polymerase chain reaction (PCR) and a custom microarray.
In this chapter, we describe: (1) a method for the establishment of a bovine
trophoblast stem population, and (2) a method for the induction of binucleate
cells from the BT-1 cell line.
2. Materials
1. Culture medium: Dulbecco’s modified Eagle’s/F-12 medium (DMEM/F-12;
Sigma, St. Louis, MO, USA) containing 100 international units (IU)/mL penicil-
lin and 100 mg/mL streptomycin (Sigma), and 10% heat-inactivated fetal bovine
serum (FBS; Sigma).
2. Cell substratum: acid-soluble porcine type I collagen solution (3 mg/mL) and gel
reconstitution solution (0.05 N NaOH solution containing 2.2% NaHCO3 and
200 mM HEPES).
3. Tenfold-concentrated physiological salt solution consisting of 1.52 M NaCl, 54 mM
KCl, 10 mM CaCl2, 8 mM MgCl2, 56 mM glucose, and 100 mM HEPES, pH 7.4.
4. Cell extraction medium: 0.1 M crystal violet (Wako Chemical, Osaka, Japan),
0.1% citric acid (Wako chemical), and 0.1% Triton X-100 (Sigma).
5. Collagen solution: acid-extracted collagen, 3 mg/mL (Cell Matrix, Nitta gelatin,
Osaka, Japan).
6. Freezing reagent: CellBanker (Zenyaku kogyo Co, Tokyo, Japan).
7. 5-Bromo 2'-deoxyuridine 5'-triphosphate (BrdU) Labeling and Detection Kit II
(Roche Diagnostics, Mannheim, Germany).
9. Hoechst 33342 (Molecular Probes, OR, USA).
10. Paraformaldehyde (Wako Chemical).
11. Bovine serum albumin (BSA, Sigma).
12. Transfer pipet (Becton Dickinson Labware, Franklin Lake, NJ, cat. no. 357575).
13. Mouse monoclonal and rabbit polyclonal anti-bovine PL (5).
14. Alexa 546-conjugated goat anti-mouse immmunoglobulin (Ig)G antibody (Molecu-
lar Probes).
15. Alexa 488-conjugated goat anti-rabbit IgG antibody (Molecular Probes).
16. 24-well culture plates and 25-cm2 culture flasks (Becton Dickinson).
17. Cell freezing vessel: BICELL (Nihon Freezer Co., Ltd, Tokyo, Japan).
18. Screen Cup with a 80 mesh screen (pore size 180 µm, Sigma).
19. Sterile plastic transfer pipets and 35-mm plastic culture dishes (Becton Dickinson).
14_Hashizume_179_188_F 180 8/29/05, 11:19 AM

3. Methods
3.1. Establishiment of Bovine Trophoblast Stem Cell Cultures
Bovine trophoblast stem cells can be established from in vitro cultured blas-
tocysts.
1. In vitro matured (IVM)/in vitro fertilized (IVF) bovine blastocysts are obtained
as described previously (6,9).
2. Blastocysts are individually plated into 24-well culture dishes that are coated
with acid-extracted collagen and cultured in DMEM/F-12 supplemented with
10% FBS, 50% fibroblast-conditioned medium, and 50 µM beta-mercaptoethanol
(see Note 1).
3. Following attachment cells outgrow from blastocysts in a week. These cells are
collected and maintained as follows under Subheading 3.1.1.
3.1.1. Maintenance of Bovine Trophoblast Stem Cells

1. Bovine trophoblast stem cells are cultured in collagen-coated culture flasks (see
Note 2).
2. The cells are maintained in 25-cm2 flasks until confluent in DMEM/F12 contain-
ing 10% FBS.
3. Media are changed every 2 or 3 d and subcultured every 7 d.
3.1.2. Subculture of BT-1 Cells

1. Pipetting dissociation: when cells have spread out and reached confluence, they
are detached from the culture dish surface by agitation with the transfer pipet.
The cell sheets are dissociated into small cell clumps by further pipetting. The
cell suspension is plated into a new culture flask at 1:2 dilution (see Note 3).
Within 24 h, cellular outgrowths can be observed (Fig. 1).
2. Transfer of vesicles: during continuous culture, some cells form vesicles ranging
in size from 100 µm to more than 1 mm in diameter. When these freely floating
vesicles are transferred into a new flask, they attach and form new outgrowths
(Fig. 1).
3.1.3. Freezing, Storage, and Retrieval of the Bovine Trophoblast Stem

Cells
1. Cells are agitated with pipetting and dissociated into small cell clumps, and then
transferred into sterile conical tubes.
2. After centrifugation at 600g for 1 min, supernatants are removed and cells are
resuspended with CellBanker (cells in 25-cm2 culture flask/1 mL).
3. The cell suspensions are transferred to cryovials and placed into a freezing vessel
(BICELL) and transferred to a –80°C freezer for at least 24 h, then to liquid
nitrogen.
4. Frozen vials are rapidly thawed at 37°C, washed once in standard culture medium
containing FBS, and seeded into culture dishes (see Note 4).
14_Hashizume_179_188_F 181 8/29/05, 11:19 AM

Fig. 1. Bovine trophoblast (BT)-1 cell features. (A) Small cell explants just after
plating in a new culture dish; (B) BT-1 cells 1 d after plating; (C) 2 d after plating; (D) 7
d after plating; (E) freely floating BT-1 vesicles; (F) BT-1 vesicles 1 day after plating.
Note that vesicles attach and exhibit cellular outgrowth. Scale bar = 500 µm
3.1.4. Measurement of Cell Proliferation

1. Nuclear count: cell proliferation is determined by counting nuclei. Cells are dis-
sociated by pipetting and collected by centrifugation at 10,000g for 10 min at
room temperature. Pelleted cells are suspended in nuclei extraction solution. After
extraction, nuclei are counted with a hemocytometer under microscopy.
14_Hashizume_179_188_F 182 8/29/05, 11:19 AM

2. BrdU incorporation assay: cell growth rate is examined by incorporation of BrdU.

Cells are labeled with BrdU for 15 min, and BrdU incorporation is detected with
an antibody to BrdU according to the manufacturer’s instructions. Cells are also
stained with Hoechst 33342 at a concentration of 5 µg/mL to visualize the nuclei.
3.2. Induction of Binucleate Cell Differentiation in BT-1 Cells

Trophoblastic binucleate cells, a bovine trophoblast specific endocrine cell,
possessing two nuclei per cell are induced from bovine trophoblast stem cells
cultured on a collagen gel substratum system.
3.2.1. Collagen Gels
To prepare collagen-gel coated culture dishes, 8 vol of the collagen solution
are gently mixed with 1 vol of 10-fold concentrated physiological salt solution,
and then 1 vol of the gel reconstitution solution with a transfer pipet in a coni-
cal tube on ice. The mixed gel solution is dispensed onto culture dishes (1 mL/
35 mm dish), and incubated at 37°C for 20–30 min for gel reconstitution.
3.2.2. Induction of Binucleate Cell Formation
Trophoblast binucleate cell differentiation is induced by growing BT-1
vesicles in culture medium on collagen gel-coated dishes. BT-1 vesicles (200 µm
to 1 mm in diameter) are isolated from confluent cell cultures by passage through
a Screen Cup (180-µm pore size), resuspended in the fresh medium, and then
plated on collagen gel-coated dishes. About 100 vesicles in 2 mL of the me-
dium are dispensed in each dish. Dishes are gently shaken to disperse vesicles
and are incubated at 37°C in an atmosphere of 5% CO2. The medium is first
changed after 4 d, and then every 2 or 3 d. Vesicles attach to the collagen gel
and cell outgrowth from the vesicles is evident within 2 d. After more than
10 d in culture, clusters of binucleate cells appear, especially at the periph-
eral region of the colony (Fig. 2).
3.2.3. Assessment of Binucleate Cells
Differentiation can be assessed by monitoring endoreduplication and the
expression of PL (Fig. 2).
3.2.3.1. ENDOREDUPLICATION
Endoreduplication can be assessed by monitoring cellular DNA contents.
We measure DNA content in cells with an image analysis system,
AQUACOSMOS (Hamamatsu Photonics, Hamamatsu, Japan) (8).
3.2.3.2. PL IMMUNOCYTOCHEMISTRY
1. For PL immunocytochemistry, cells in dishes are fixed in 4% paraformaldehyde
in 0.1 M phosphate buffer (pH 7.4) at 4°C for 15 min.
14_Hashizume_179_188_F 183 8/29/05, 11:19 AM

Fig. 2. Phase contrast (A), Hoechst 33342 (B), and (C) placental lactogen (PL)
staining (with monoclonal anti-PL antibody) images of bovine trophoblast (BT)-1 cells
cultured on collagen gels for 17 ds. Some binucleate cells are indicated by arrows.
These binucleate cells have an intense Hoechst fluorescence in their nuclei and express
PL. Scale bar =100 µm.
2. After three washes with phosphate-buffered saline (PBS), the cells are blocked
and permeabilized with PBS containing 10% normal goat serum and 0.5% Triton
X-100 for 30 min at room temperature.
3. Incubation with the mouse monoclonal (diluted 1:1000) or the rabbit polyclonal
(1:8,000) anti-PL antibody is carried out in PBS containing 1% BSA, 0.05%
NaN3, and 0.3% Triton X-100 for 2 h at room temperature.
4. After three washes with PBS, Alexa 546-conjugated goat anti-mouse IgG anti-
body (1:400) or Alexa 488-conjugated goat anti-rabbit IgG antibody (1:200) in
PBS containing 1% BSA, 0.05% NaN3, and 0.3% Triton X-100 is applied for 1 h
at room temperature.
5. Hoechst 33342 (5 µg/mL) is added into the secondary antibody solution to stain
nuclei.
6. After three washes with PBS, PL and Hoechst 33342 signals can be viewed using
an inverted epifluorescence microscope with appropriate filters.
3.2.3.3. BOVINE UTEROPLACENTAL CUSTOM DNA MICROARRAY

Bovine trophoblast stem cells can differentiate into binucleate cells and express
an array of transcripts including placental prolactin family proteins, pregnancy-
associated glycoproteins and IFN-τ. Detailed gene profiles can be analyzed using a
custom bovine uteroplacental cDNA microarray (7) (see Note 5).
14_Hashizume_179_188_F 184 8/29/05, 11:19 AM

4. Notes
1. Fibroblast-conditioned medium. We generally use DMEM/F12 containing 10%
FBS culture medium for maintenance of bovine trophoblast stem cells. As previ-
ously described, the BT-1 cells were established from blastocysts using fibro-
blast-conditioned medium (6). However, conditioned medium is not necessary
for the maintenance of cell growth following the establishment of the cell line.
2. Collagen-coated culture dishes. Acid-extracted collagen suspension is diluted
10-fold with distilled water and poured into dishes. After a 60-min incubation
at room temperature, they were rinsed with culture medium and used for culture.
3. Cell concentration. Adequate density of bovine trophoblast stem cells is impor-
tant for sustaining cell growth. When cells are plated at low density, cells tend to
become flatten and stop proliferating. Passaging the cells at a 1:2 ratio is optimal
for subculture.
4. Freezing for storage and thawing for re-culture. After cyropreservation, bovine
trophoblast stem cell viability is low. To successfully establish a culture from a
frozen vial, seed cells from a cryovial into a small culture dish (35 mm or smaller).
Supplementation of the cultures with BT-1 trophoblast stem cell conditioned
medium may facilitate recovery and growth of previously frozen bovine tropho-
blast stem cells.
5. A custom-designed cDNA microarray using uteroplacental cDNAs has been suc-
cessfully utilized for analyzing transcriptome in BT-1 cells as well as in the pla-
centa and uterus (7). Results from DNA microarray studies should be confirmed
by other appropriate procedures (e.g., RT or real-time PCR). Table 1 provides an
overview of the gene expression profile in BT-1 cells. The BT-1 cells express an
array of transcripts including placental prolactin family proteins, pregnancy-
associated glycoproteins and IFN-τ, and other genes involved in steroidogen-
esis and cytokine signaling. In this experiment, BT-1 cells were maintained with
DMEM/F12 supplemented with 20% FBS. The intensity of expression is shown
as a relative value compared to that of the glyseraldehyde-3-phosphate dehydro-
genase (GAPDH) as internal reference. Figure 3 shows the expression profiles
of selected genes under different culture conditions. The cells are routinely grown
in a medium supplemented with 10 % FBS. In this experiment, differences in
gene expression pattern under different culture conditions with 20% FBS, 2%
FBS or 10 % horse serum (HS)-supplemented medium were determined. BT-1
cells express an assortment of trophoblast marker genes (PL, PRP-I, PAG, and
IFN-τ) when maintained in 20% FBS-supplemented medium. However, the cells
cease or decrease expression of these genes when the serum supplementation is
switched to 10% HS. This is in marked contrast to the Rcho-1 rat trophoblast
cell line (11). BT-1 cells represent both proliferative and endocrine phenotypes
depending on the presence of FBS. In contrast, both bone morphogenetic protein
(BMP)4 and Oct 3/4 are stably expressed regardless of sera. Both eomesodermin
and Oct 3/4 are thought to be markers representing the undifferentiating status
in mouse trophoblast stem cell (1). It has been reported that BMP4 triggers
14_Hashizume_179_188_F 185 8/29/05, 11:19 AM

Table 1
Expression Levels of Selected Genes in Bovine Trophoblast (BT)-1 Cells
Accession # Gene name Intensity/GAPDH
M73962 Bovine PAG-1 0.27
L06151 Bovine PAG-2 0.63
AF020506 Bovine PAG-4 0.12
AF020513 Bovine PA G-1l 0.39
J02840 Bovine PL 0.18
J02944 Bovine PRP-1 0.21
M27239 Bovine PRP-2 0.19
X59504 Bovine PRP4 0.16
X15975 Bovine PRP-5 0.15
X53587 Human integrin β 4 0.22
NM000210 Human integrin α 6 0.33
AF074706 Bovine 111-β-HSD type 2 0.00
S54973 Bovine 20 α-HSD 1.02
AF210379 Bovine matrix Gla 0.13
D30750 Bovine Msx-1 0.18
M18344 Bovine Calbindin 0.27
X53553 Bovine IGF-1I 0.19
M76478 Bovine IGFBP-3 0.44
NM001553 Human IGFBP7 0.11
M37210 Bovine IL-1-α 0.93
AF110801 Human IL-18 binding protein c 0.16
AB005148 Bovine IL-1 receptor antagonist 0.18
U16239 Bovine βA inhibin/activin 0.17
L22095 Bovine U-serpin 0.19
S75723 Ovine GRP 0.28
NM003155 Human STC1 0.23
AF196320 Bovine IFN-τ1C 0.22
GAPDH, glyseraldehyde-3-phosphate dehydrogenase; PAG, pregnancy-associated glycopro-

teins; PL, placental lactogen; PRP, prolactin-related protein; HSD, hydroxysteroid dehydroge-
nase; IGF, insulin-like growth factor; IL, interleukin; GRP, glucose-related protein; STC,
Stanniocalcin; IFN, interferon.
14_Hashizume_179_188_F 186 8/29/05, 11:19 AM

Fig. 3. Revers-transcriptase polymerase chain reaction analysis of selected genes

expressed in bovine trophoblast (BT)-1 cells cultured with either 20% (first lane) or
2% (second lane) fetal bovine serum (FBS), or 10% horse serum (HS). PL, placental
lactogen; PRP-I, prolactin-related protein-I; PAG-1, pregnancy-associated glycopro-
tein-1; IFN-τ, interferon τ; BMP4, bone morphogenetic protein 4; GDF9, growth and
differentiation factor 9;. Eomes, eomesodermin.
human trophoblast differentiation (12). There is a mixture of both undifferentiating

and differentiating marker genes in BT-1 cells. Considering the previously described
accounts, it is plausible that BT-1 cells have some characteristics that differ from
those of rat Rcho-1 trophoblast cells and mouse trophoblast stem cells.
Acknowledgments
The authors thank Dr. K. Imai for in vitro fertilization and culture of
bovine embryo. We also thank Drs. H. Ishiwata and K. Kizaki (N.I.A.S.),
and G. Tsujimoto (Kyoto University) for fabricating and analyzing bovine
utero-placental cDNA microarray.
14_Hashizume_179_188_F 187 8/29/05, 11:19 AM

References
2. Flechon. J. E., Laurie, S., and Notarianni, E. (1995) Isolation and characterization
of a feeder-dependent, porcine trophectoderm cell line obtained from a 9-day blas-
tocyst. Placenta 16, 643–658.
3. Talbot, N. C, Caperna, T. J., Edwards, J. L., Garrett, W., Wells, K. D., and Ealy,
A. D. (2000) Bovine blastocyst-derived trophectoderm and endoderm cell cul-
tures: interferon tau and transferrin expression as respective in vitro markers. Biol.
Reprod. 62, 235–247.
4. Miyazaki, H., Imai, M., Hirayama, T., et al. (2002) Establishment of feeder-inde-
pendent cloned caprine trophoblast cell line which expresses placental lactogen
and interferon tau. Placenta 23, 613–630.
5. Nakano, H., Takahashi, T., Imai, K., and Hashizume, K. (2001) Expression of
placental lactogen and cytokeratin in bovine placental binucleate cells in culture.
Cell Tissue Res. 303, 263–270.
6. Shimada, A., Nakano, H., Takahashi, T., Imai, K., and Hashizume, K. (2001)
Isolation and characterization of a bovine blastocyst-derived trophoblastic cell
line, BT-1: development of a culture system in the absence of feeder cell. Pla-
centa 22, 652–662.
7. Ishiwata, H., Katsuma, S., Kizaki, K., et al. (2003) Characterization of gene ex-
pression profiles in early bovine pregnancy using a custom cDNA microarray.
Mol. Reprod. Dev. 65, 9–18.
8. Nakano, H., Shimada, A., Imai, K., Takezawa, T., Takahashi, T., and Hashizume,
K. (2002) Bovine trphoblastic cell differentiation on collagen substrata: formation
of binucleate cells expressing placental lactogen. Cell Tissue Res. 307, 225–235.
9. Konishi, M., Aoyagi, Y., Takedomi, T., Itakura, H., Itoh, T., and Yazawa, S.
(1996) Production and transfer of IVF embryos from individual inhibin-immu-
nized cows by ultrasound-guided transvaginal follicular aspiration. J. Vet. Med.
Sci. 58, 893–896.
10. Quackenbush, J. (2002) Microarray data normalization and transformation. Nat.
Genet. 32(Suppl.), 496–501.
11. Faria, T. N. and Soares, M. J. (1991) Trophoblast cell differentiation: establish-
ment, characterization, and modulation of a rat trophoblast cell line expressing
members of the placental prolactin family. Endocrinology 129, 2895–2906.
12. Xu, R. H., Chen, X., Li, D. S., et al. (2002) BMP4 initiates human embryonic stem
cell differentiation to trophoblast. Nat. Biotechnol. 12, 1261–1264.
14_Hashizume_179_188_F 188 8/29/05, 11:19 AM

Trophoblast From Human ES Cells In Vitro 189
15
In Vitro Induction of Trophoblast from Human Embryonic
Stem Cells
Ren-He Xu
Summary
Human embryonic stem (ES) cells can proliferate without a known limit and can form advanced
derivatives of all three embryonic germ layers. What is less widely appreciated is that human ES
cells can also form the extra-embryonic tissues that differentiate from the embryo before gas-
trulation. The use of human ES cells to derive early human trophoblast is particularly valuable,
because it is difficult to obtain from other sources and is significantly different from mouse
trophoblast. Here we describe a method by using bone morphogenetic protein (BMP)4 , a mem-
ber of the transforming growth factor (TGF)-β superfamily, to induce the differentiation of
human ES cells to trophoblast. Immunoassays (as well as DNA microarray and reverse-tran-
scription polymerase chain reaction analyses—data not shown) demonstrate that the differenti-
ated cells express a range of trophoblast markers and secrete placental hormones. When plated
at low density, the BMP4-treated cells form syncytia that express chorionic gonadotrophin
(CG). This technique underscores fundamental differences between human and mouse ES cells,
which differentiate poorly, if at all, to trophoblast. Human ES cells thus provide a tool for
studying the differentiation and function of early human trophoblast and could provide a new
understanding of some of the earliest differentiation events of human postimplantation devel-
opment.
Key Words: Human embryonic stem cells; trophoblast; bone morphogenetic protein.
1. Introduction
The trophectoderm is the first differentiated cell type in the mammalian
embryo, and it forms the outer epithelium of the blastocyst and later contrib-
utes (as the trophoblast) to the outer layers of the placenta. The trophectoderm
is crucial for implantation and maintenance of pregnancy. When formed into
chimeras with intact preimplantation embryos, mouse embryonic stem (ES)
cells rarely contribute to the trophoblast, and the manipulation of external cul-
ture conditions has, to date, failed to direct mouse ES cells to differentiate to
trophoblast (1). Mixed populations of spontaneously differentiated rhesus mon-
189
15_Xu_189_202_F 189 8/29/05, 11:19 AM

190 Xu
key or human ES cells secrete modest amounts of chorionic gonadotrophin

(CG), which suggests the presence of trophoblast (2,3). The contrast in the
ability to form trophoblast represents a basic difference between mouse and
human ES cells, and may suggest basic differences in the allocation of lineages
in the early embryo. This difference is also suggested by the behavior of embryo-
nal carcinoma cells, the stem cells of teratocarcinomas, which are tumors that
recapitulate events of early embryogenesis. Human teratocarcinomas often
contain a trophoblast component, but mouse teratocarcinomas do not (4). Given
basic differences between mouse placenta and human placenta, and given the
lack of experimental access to early postimplantation human embryos, human
ES cells provide an important new in vitro model to study early trophoblast
differentiation and function, with important implications for infertility, mis-
carriage, and birth defects. Following addition of a transforming growth factor
(TGF)-β superfamily member, bone morphogenetic protein (BMP)4, to cul-
ture conditions that would otherwise support self-renewal of human ES cells,
the cells undergo a highly synchronous differentiation to trophoblast (5). This
features by expression of a range of trophoblast markers, secretion of placental
hormones, and formation of CG-positive syncytia.
2. Materials
1. Human ES cells. Human ES cell lines H1, H7, H9, and H14 were used for the
trophoblast induction by BMPs. Their National Institutes of Health (NIH) regis-
try numbers are WA01, WA07, WA09, and WA14, respectively (http://
stemcells.nih.gov/research/registry/index.asp#warf). These cell lines are avail-
able at the WiCell Research Institute upon application and licensing (http://
www.wicell.org/forresearchers/index.jsp?catid=4).
2. Mouse embryonic fibroblast (MEF) (see below for details).
3. Human uterine fibroblast (HUF) cell line (6).
4. Four-well and six-well culture plates, 50-, 75-, and 90-mm Filter Units (Nalge
Nunc International, Rochester, NY).
5. Six-well Transwell plate (Corning Incorporation, Corning, NY).
6. 0.22-mM Filter Unit (Millipore, Bedford, MA).
7. 40 mM mesh (BD Labware, Bedford, MA).
8. Falcon (35/2054) 5 mL polystyrene round bottom tube (BD Labware).
9. Knockout Dulbecco’s modified Eagle’s/F12 medium (DMEM/F12), knockout
serum replacement (SR), 100X MEM nonessential amino acid solution, L-glutamine,
and Trypsin/ethylenediamine tetraacetic acid (EDTA) solution (Invitrogen,
Carlsbad, CA).
10. Human basic fibroblast growth factor (bFGF) (Invitrogen). Dissolve 10 µg bFGF
in 1 mL of 0.1% bovine serum albumin (BSA) in phosphate-buffered saline (PBS)
(without Ca2+ and Mg2+). Aliquot and store at –80°C.
11. Collagenase type IV (Invitrogen). Dissolve it at 1 mg/mL in DMEM/F12 and
sterilize with a 0.2-µm cellulose acetate filter. Store at 4°C. Use within 2 wk.
15_Xu_189_202_F 190 8/29/05, 11:19 AM

12. Matrigel (BD Labware).

13. BMP2, -4, -7, growth differentiation factor (GDF)5 , noggin, and soluble BMP
receptor IA (R&D Systems, Minneapolis, MN).
14. HUF cell culture medium: RPMI culture medium supplemented with penicillin and
streptomycin, 1 mM sodium pyruvate, and 10% fetal bovine serum (FBS) (6).
15. β-Mercaptoethanol, Brefeldin A, Hoechst 33342 solution, and mouse immuno-
globulin (Ig)G (Sigma, St. Louis, MO).
16. Mouse anti-human CG-antibody (Abcam, Cambridge, UK).
17. Fluorescein-labeled rabbit anti-mouse IgG antibody (Pierce, Rockford, IL).
18. AxSYM Total β-hCG kit (Abbott, Lake Forest, IL).
19. FACSCalibur flow cytometer (Becton Dickson, San Jose, CA).
20. Cellquest acquisition and analysis software (Becton Dickinson).
3. Methods
The methods described below outline (1) cell culture, (2) trophoblast induc-
tion, and (3) biological assays of the induced trophoblast.
3.1. Cell Culture (3,7)

3.1.1. Preparation of Human ES Cell Medium (100 mL)
This medium is also referred to as unconditioned medium in contrast to MEF-
conditioned human ES cell medium. Final Concentration: 80% knockout DMEM/
F12, 20% knockout SR, 1% nonessential amino acid solution, 100X MEM nones-
sential amino acid solution, 1 mM L-glutamine, 0.1 mM β-mercaptoethanol.
1. Add 10 mL PBS (without Ca2+ and Mg2+) to 0.146 g L-glutamine in a 15-mL
tube.
2. Add 7 µL of β-mercaptoethanol to the L-glutamine/PBS, and mix well.
3. Into a 250-mL, 0.2-µm cellulose acetate filtering unit add:
a. 80 mL knockout DMEM/F12.
b. 20 mL knockout SR.
c. 1 mL of the L-glutamine/β-mercaptoethanol solution.
d. 1 mL 100X nonessential amino acid solution
4. Filter to sterilize.
5. Store at 4°C and use within 2 wk.
3.1.2. Preparation of the MEF Medium (100 mL)

1. Into a 250-mL, 0.2-µm cellulose acetate filtering unit add:
a. 90 mL DMEM medium.
b. 10 mL heat-inactivated (30 min at 57°C) FBS.
c. 1 mL 200 mM L-glutamine solution.
d. 1 mL 100X nonessential amino acid solution.
15_Xu_189_202_F 191 8/29/05, 11:19 AM

192 Xu
Fig. 1. Mouse embryonic fibroblast (MEF) cells. MEF cells plated at 2.12 × 105
cells/mL in a T75 flask serve as feeder cells to produce conditioned medium.
2. Filter to sterilize.
3. Store at 4°C and use within 2 wk.
3.1.3. Preparation of MEF Cells and MEF-Conditioned Medium

1. Isolate MEF cells from CF-1 mouse embryos at 13–14 d gestation and store fro-
zen as described (8). Frozen vials of MEF cells are thawed and plated in 20 mL
MEF medium in a T75 flask. Split the cells when confluent.
2. Harvest MEF cells, irradiate at 40 Gy, and seed at 2.12 × 105 cells/mL in MEF
medium (Fig. 1). After at least 4 h, exchange the medium with ES cell medium
(0.5 mL/cm2).
3. Collect the conditioned medium daily and supplement with an additional 4 ng/
mL human bFGF before feeding hES cells.
4. Feed the MEF cells again with ES cell medium daily and use for 7–10 d for the
conditioned medium collection. The conditioned medium can also be frozen for
storage at –20°C for 1 mo and thawed for later use.
3.1.4. Culture of Human ES Cells

1. Coat six-well culture plates by incubating with Matrigel diluted 1:20 in cold knock-
out DMEM at 1 mL per well at 4°C overnight or at room temperature for 1 h.
2. Preparation of 1 mg/mL collagenase solution:
a. Dissolve 10 mg collagenase IV in 10 mL DMEM/F12 medium.
15_Xu_189_202_F 192 8/29/05, 11:19 AM

Fig. 2. Trophoblast induction. Human embryonic stem cell line H1 cells were
treated with (C,D) or without (A,B) 100 ng/mL bone morphogenetic protein-4 for 7 d.
The cells were photographed at 5× (A,C) and 20× (B,D).
b. Sterile filter the solution with a 0.22-mm filter unit.

c. Keep the solution at 4°C.
3. Passage human ES cells as cells become confluent (Fig. 2), i.e., about once a
week, as follows:
a. Add 1 mL collagenase solution to each well of human ES cells cultured in a
six-well plate, and incubate the plate at 37°C for at least 5 min.
b. To confirm colony separation from the plate, look for the edge of the colonies
to peel off under a microscope.
c. Using a glass 5 mL pipet, scrape cells off the surface of the plate.
d. Transfer the cell suspension into a 15 mL tube.
e. Gently pipet cells up and down a few times in the tube to further break-up cell
colonies.
f. Add DMEM/F12 to a final volume of 10 mL and mix gently.
g. Pellet the cells by spinning at 200g for 5 min.
h. Meanwhile, remove unbound Matrigel solution from the coated plate.
i. Resuspend the cell pellet with conditioned medium, plate the cell suspension
into the coated plate at a split ratio of 1:3–1:6 (cells from each original well
are split to three to six new wells).
15_Xu_189_202_F 193 8/29/05, 11:20 AM

194 Xu
j. Return the plate to incubator. Move the plate back-and-forth and side-to-side
several times to evenly disperse the cells across the surface of the wells.
k. Incubate cells overnight without touch to ensure that colonies attach.
l. Add fresh conditioned media daily.
3.2. Trophoblast Induction and Culture (5) (see Note 1)

3.3.1. Trophoblast Induction
1. Split human ES cells into a six-well plate as described previously.
2. When the cells become approx 30% confluent (usually on day 2–3 following
split), treat the cells with or without 100 ng/mL BMP4.
3. Add fresh conditioned media and BMP4 every other day (see Note 2).
4. Observe morphological changes with following features (see ref. 5 [http://ge-
nome-www.stanford.edu/es_cell/supplement.shtml]):
a. At about day 2 (48 h) following the initiation of BMP4 treatment, a synchro-
nous wave of differentiation occurs from the edge of the colonies, character-
ized by flattened and enlarged cell types and reduced proliferation (Fig. 2).
b. Gradually, the differentiation continues inward to the center of the colonies.
c. The morphological changes become obvious by day 2 for BMP4 at 100 ng/
mL, days 3–4 for 10 ng/mL, and days 5–6 for 1 ng/mL.
d. BMP family members such as BMP2 (300 ng/mL), BMP7 (300 ng/mL), and
GDF5 (30 ng/mL) induce similar morphological changes to that induced by
BMP4 (100 ng/mL).
e. Addition of inhibitors of BMP signaling, such as the soluble BMP receptor
IA (500 ng/mL) or the BMP antagonizing protein noggin (500 ng/mL), can
block the morphological changes induced by the BMPs.
3.2.2. Formation of Syncytial Trophoblast

1. Aspirate spent medium from a well of human ES cells cultured in a six-well
plate.
2. Add 1 mL Trypsin/EDTA Solution to the well, and incubate at 37°C for at least
5 min.
3. Scrape the cells from the plate with a glass pipet, and transfer the cells to a 15-mL
tube.
4. Break up the cell colonies by pipetting up and down several times, and add knock-
out DMEM/F12 medium to a final volume of 10 mL.
5. Pellet the cells by spinning at 200g for 5 min.
6. Remove the supernatant and re-suspend the pellet in conditioned medium, which
contains mostly single cells.
7. Plate the cells at 105 cells per well, and incubate at 37°C overnight to allow the
cells to attach.
8. Treat the cells with or without 100 ng/mL BMP4 the next day.
9. Add fresh media and BMP4 every other day.
10. Observe formation of syncytial cells, which usually occurs within 1–2 wk of the
treatment, featuring giant and irregularly shaped cells containing more than two
15_Xu_189_202_F 194 8/29/05, 11:20 AM

Fig. 3. Syncytial cell. The left panel shows a syncytial cell formed after 2 wk of
treatment of individualized embryonic stem cells by bone morphogenetic protein-4.
The right panel shows immunofluorescence for CG-β and Hoechst 33342 fluorescence
for the nuclei in the cell.
nuclei. The observed highest number of nuclei in one single syncytial cell was
100 (Fig. 3). These differentiated phenotypes remain for a long time without
obvious changes (see ref. 5 [http://genome-www.stanford.edu/es_cell/supple
ment.shtml]).
3.3.1. Suspended Culture of Trophoblast Vesicles

1. Induce human ES cells to differentiate into trophoblast with BMP4 in a six-well
plate as described under Subheading 3.2.1.
2. At day 7, aspirate spent media from the plate, and wash the cells with PBS once.
3. Use collagenase solution to detach the cells and prepare cell suspension as described
under Subheading 3.1.4.
4. Pellet the cells, resuspend the pellet in human ES cell medium, and transfer the
cell suspension onto an uncoated six-well plate.
5. Culture the plate on a rotator in incubator.
6. Observe the formation of trophoblast vesicles in the medium (Fig. 4).
7. Refresh 50% of the spent medium twice a week.
8. To passage, break up the vesicles by pipetting them up and down with a glass
pipet, and split the vesicle suspension at 1:2 or 1:3.
3.2.4. Co-Culture of Trophoblast With Human Uterine Fibroblast HUF

Cells
1. Coat the inner chamber of a six-well Transwell plate with Matrigel as described
under Subheading 3.1.4.
15_Xu_189_202_F 195 8/29/05, 11:20 AM

196 Xu
Fig. 4. Trophoblast vesicle. Following bone morphogenetic protein-4 treatment for

7 d, the differentiated human embryonic stem (ES) cells were dislodged from the cul-
ture plate by collagenase treatment. They form vesicles when suspended in human ES
cell medium.
2. Passage human ES cells into the inner chamber, and culture them in a total of 3 mL
conditioned medium added to the inner chamber and the outer chamber of the
Transwell plate (both chambers link through openings on the side wall of the
inner chamber).
3. Two days post passage, add 100 ng/mL BMP4 to the culture for 7 d to induce
trophoblast.
4. HUF cells are maintained and expanded in RPMI culture medium supplemented
with antibiotics, sodium pyruvate, and FBS.
5. Seed 2 × 104/well HUF cells in MEF medium in a new six-well plate, and culture
them for 2 d.
6. Remove the spent media, and wash with PBS once.
7. Aspirate the spent media from the Transwell, transfer each of the inner chambers
that contains BMP4-induced trophoblast to the well that contains HUF cells.
8. Add 3 mL human ES cell medium to the inner and outer chambers.
9. Co-culture the cells for 7 d.
10. Collect spent media daily to test for the production of placental hormones.
15_Xu_189_202_F 196 8/29/05, 11:20 AM

3.4. Characterization of the Induced Trophoblast (5) (see Note 3)

3.4.1. Immunocytochemistry
1. Culture human ES cells in a four-well plate that contains 0.5 mL conditioned
medium.
2. Treat the cells with or without 100 ng/mL BMP4 for 7 d as above.
3. On day 7 of the treatment, remove the spent medium from the treated cells, add
fresh human ES cell medium containing brefeldin A, a Golgi blocker, at 1.25 g/
mL, and incubate for 4 h at 37°C (see Note 4).
4. Remove the medium, wash the cells with PBS once, and fix them with 2%
paraformaldehyde for 10 min at room temperature.
5. Remove the fixative and wash the cells with PBS once.
6. Block and permeabilize the cells by incubating them in PBS containing 0.1%
Triton X-100 (T-PBS) and 5% milk powder at room temperature for 30 min.
7. Remove the blocking solution and wash the cells with T-PBS.
8. Add 0.2 mL T-PBS containing mouse anti-human CG-β antibody at 1:100 dilu-
tion to the test well, and 0.2 mL T-PBS containing equal amount of mouse IgG to
the control well.
9. Incubate the cells at 4°C overnight.
10. Remove the solution, and wash the cells with 0.5 mL T-PBS three times (5 min
each time).
11. Add 0.2 mL T-PBS containing fluorescein-labeled rabbit anti-mouse IgG anti-
body at 1:200 to each well.
12. Incubate the cells at room temperature for 30 min.
13. Remove the solution, and wash the cells with 0.5 mL T-PBS three times (5 min
each time).
14. Incubate the cells with the Hoechst 33342 solution for at least 5 min to stain the
nuclei.
15. Aspirate the solution (to reduce light reflection) and photograph at 20× magnifi-
cation under both phase and epifluorescence microscope (Fig. 3).
3.4.2. Placental Hormone Measurement

1. Collect 2 mL spent media daily from cultures of BMP4-treated ES cells (Fig. 5A),
suspended trophoblast vesicles, or co-culture of trophoblast with human uterine
fibroblast HUF cells (Fig. 5B).
2. Keep the media at –70°C or immediately test for CG-β concentrations using the
AxSYM Total hCG-β kit, and estradiol (9) and progesterone concentrations (10)
by enzyme-linked immunosorbent assay (ELISA) (Fig. 5).
3.4.3. Flow Cytometry Assay of the Induced Trophoblast

1. Culture human ES cells in a six-well plate that contains 2.5 mL conditioned
medium.
2. Treat the cells with or without 100 ng/mL BMP4 for 7 d as described previously.
15_Xu_189_202_F 197 8/29/05, 11:20 AM

198 Xu
Fig. 5 (caption on facing page)
3. Prepare fluorescence-activated cell sorting (FACS) buffer as follows. To cal-

cium- and magnesium-free PBS, add 2% FBS and 0.1% sodium azide.
4. On day 7 of the BMP4 treatment, remove the spent medium from the treated
cells, add fresh human ES cell medium containing brefeldin A at 1.25 g/mL, and
incubate for 4 h at 37°C (see Note 4).
5. Remove the medium, and wash the cells with PBS once.
15_Xu_189_202_F 198 8/29/05, 11:20 AM

Fig. 5. Tests for placental hormone production. chorionic gonadotrophin (CG)-β,

estradiol, and progesterone levels were detected in media collected from the culture of
bone morphogenetic protein (BMP)4-treated human embryonic stem (ES) cells (A) and
from the co-culture of BMP4-induced trophoblast with human uterine fibroblast (HUF)
line cells (B). Mouse embryonic fibroblast (MEF) cells were used to generate the con-
ditioned medium (CM; see Subheading 3.1.3.). BMP4 was added to this medium
(CM + BMP4) to detect hormone production from the BMP4-treated cells. Uncondi-
tioned medium (UM) was used as a control for CM, because ES cells in UM spontane-
ously differentiate and generate low levels of the placental hormones, whereas the
cells remain undifferentiated in CM and do not produce, or produce undetectable lev-
els of, the hormones. Consequently, BMP4 actions are best tested in the CM to exclude
the non-specific differentiation caused by UM.
6. Add 1 mL Trypsin/EDTA solution to the well, and incubate at 37°C for at least
5 min, break up the cell colonies by pipetting up and down several times, and
then add 1 mL ES cell medium to neutralize the Trypsin/EDTA solution.
7. Scrape the cells from the plate with a glass pipet, and transfer the cells to a 15-mL
tube.
8. Break up the cell colonies by pipetting up and down several times, and add human
ES cell medium to a final volume of 10 mL.
9. Pellet the cells by spinning at 200g for 5 min.
10. Remove the supernatant and re-suspend the pellet in FACS buffer and count the
cells.
15_Xu_189_202_F 199 8/29/05, 11:20 AM

200 Xu
11. Pellet the cells by spinning at 200g for 5 min, remove the supernatant (leave
about 0.1 mL supernatant), and briefly mix to re-suspend the cells in the residual
supernatant.
12. Add 1 mL 2% paraformaldehyde to the tube, mix well, and incubate at room
temperature for 10 min.
13. Pellet the cells by spinning at 200g for 5 min, and remove the supernatant.
14. Add FACS buffer plus 0.1% Triton X100 to resuspend and permeabilize the cells,
and also achieve a cell concentration of 5 × 106/mL.
15. Add 100 µL of the cell suspension containing 5 × 105 cells per tube to both a test
tube and a control tube (using a Falcon 5-mL polystyrene round-bottomed tube).
16. Add 1 µL mouse anti-human CG-β antibody (5 mg/mL) to the test tube, and 5 µL
of mouse IgG (1 mg/mL) to the control tube.
17. Briefly vortex the tubes to mix and incubate the tubes at 4°C overnight.
18. Add 1 µL fluorescein-labeled rabbit anti-mouse IgG antibody to both tubes, and
incubate for 30 min on ice.
19. Wash the cells twice with 1-mL FACS buffer plus 0.1% Triton X100 by centrifu-
gation at 200g for 5 min for each wash.
20. Re-suspend the cells in 0.3 mL of FACS buffer.
21. Analyze the samples on a FACSCalibur flow cytometer using the Cellquest
acquisition and analysis software (see Note 5).
4. Notes
1. For synchronous differentiation of human ES cells to trophoblast, ES cells should
be passaged as small colonies (about 200 µm in size), and BMP4 added when the
cells are about 30% confluent. Big colonies often end up with the cells in the
middle of the colonies remaining undifferentiated.
2. We have observed that the potency of BMP4 added to ES cell cultures twice on
alternative days is equivalent to that of BMP4 added daily for 7 d, as evaluated
by morphology and CG secretion.
3. According to microarray and reverse transcription-polymerase chain reaction as-
says (5), the expression levels of ES cell- and trophoblast-related genes change
dynamically in the ES cells during BMP4 treatment. From 3 h through 7 d of
BMP4 treatment, the expression of the following trophoblast-related genes are
elevated: TFAP2A, TFAP2C, MSX2, GATA2, GATA3, SSI3, HEY1, FZD,
PlGF, CGB, CGA, LHB, GCM1, INSL4, PAEP, PAPPE, DEPP, MET, and HLA-
G1 (see ref. 5 [http://genome-www.stanford.edu/es_cell/supplement.shtml]). At
day 7, ES cell marker genes OCT4 and TERT are downregulated.
4. For detection of CG-β expression in trophoblast by immunocytochemistry or flow
cytometry, it is essential to enhance the signal by pretreating the cells with the
Golgi blocker brefeldin A for 4 h, permeabilizing the fixed cells with Triton X-
100, and incubating the cells with anti-CG-β antibody at 4°C overnight.
5. A total of 10,000 events are required. Analysis is restricted to live events based
on light scatter properties. The fluorescein signal is collected through a 530/30
15_Xu_189_202_F 200 8/29/05, 11:20 AM

Fig. 6. Flow cytometry analysis for chorionic gonadotrophin (CG)-β positive cells
in human embryonic stem cells cultured in conditioned medium (CM) with or without
bone morphogenetic protein-4 treatment for 7 d. HCG stands for human CG.
band pass filter, and the mean fluorescence for both the IgG control and the test
samples are determined. All data are normalized via division of the test mean by
the control mean (Fig. 6).
Acknowledgments
I thank Dr. James Thomson, his laboratory and the WiCell Research Institute
for contributions to this work. It was supported by WiCell Research Institute, a
non-profit subsidiary of the Wisconsin Alumni Research Foundation.
References
1. Beddington, R. S. P. and Robertson, E. J. (1989) An assessment of the develop-
mental potential of embryonic stem cells in the midgestation mouse embryo.
2. Thomson, J. A., Kalishman, J., Golos, T. G., et al. (1995) Isolation of a primate
embryonic stem cell line. Proc. Natl. Acad. Sci. USA 92, 7844–7848.
3. Thomson, J. A., Itskovitz-Eldor, J., Shapiro, S. S., et al. (1998) Embryonic stem
cell lines derived from human blastocysts. Science 282, 1145–1147.
4. Andrews, P. W., Oosterhuis, J., and Damjanov, I. (1987) Cell lines from human
germ cell tumors, in Teratocarcinomas and Embryonic Stem Cells: A Practical
Approach (Robertson, E., ed.). IRL, Oxford: pp. 207–246.
5. Xu, R. H., Chen, X., Li, D. S., et al. (2002) BMP4 initiates human embryonic stem
cell differentiation to trophoblast. Nat. Biotechnol. 20, 1261–1264.
15_Xu_189_202_F 201 8/29/05, 11:20 AM

202 Xu
6. Strakova, Z., Srisuparp, S., and Fazleabas, A. T. (2000) Interleukin-1beta induces

the expression of insulin-like growth factor binding protein-1 during
decidualization in the primate. Endocrinology 141, 4664–4706.
7. Xu, C., Inokuma, M. S., Denham, J., et al. (2001) Feeder-free growth of undiffer-
entiated human embryonic stem cells. Nat. Biotechnol. 19, 971–974.
8. Robertson, E. J. (1987) Embryo-derived stem cell lines, in Teratocarcinomas and
Embryonic Stem Cells: A Practical Approach (Robertson, E., ed.). IRL, Oxford:
pp. 71–112.
9. French, J. A., Abbott, D. H., Scheffler, G., Robinson, J. A., and Goy, R. W. (1983)
Cyclic excretion of urinary oestrogens in female tamarins (Saguinus oedipus).
J. Reprod. Fertil. 68, 177–184.
10. Munro, C. and Stabenfeldt, G. (1984) Development of a microtitre plate enzyme
immunoassay for the determination of progesterone. J. Endocrinol. 101, 41–49.
15_Xu_189_202_F 202 8/29/05, 11:20 AM

Isolation and Culture of Human Trophoblast 203
16
Isolation and Culture of Term Human Trophoblast Cells
Margaret G. Petroff, Teresa A. Phillips, Hakhyun Ka, Judith L. Pace,

and Joan S. Hunt
Summary
Experimentation with most human cell types is restricted to the use of cell lines, and this
limits our ability to extrapolate interpretations to the in vivo condition. However, in studying
human trophoblast cells, we have a unique opportunity to obtain large quantities of readily
available human tissue. In this chapter, we outline the methodology for purification of human
trophoblast cells from term placentas. The procedures are based on enzymatic dissociation of
villous placental tissue, followed by gradient centrifugation and immunomagnetic bead purifi-
cation. Purity may be assessed by immunocytochemistry or flow cytometry using a number of
markers to identify both cytotrophoblast cells and cellular contaminants. The resulting cytotro-
phoblast cell populations have excellent viability and purity, and may be subjected to long-
term culture.
Key Words: Human; placenta; cytotrophoblast cell; cell culture.
1. Introduction
The ability to establish primary cell cultures from human organs is a rare
opportunity. Even when it is possible, the investigator must often rely on lim-
ited quantities of tissues obtained from clinical biopsies. Furthermore, it is
implicit that tissues could be diseased, particularly when samples are obtained
from nonelective surgery. In contrast, those who study the trophoblast cell ben-
efit from the large size of a readily available source material: the term placenta.
An individual can easily process up to 50 g of villous placental tissue, harvest-
ing upwards of 250 million cells. A further benefit is that there is little ethical
controversy surrounding the use of human placenta for biomedical research
because it is almost invariably discarded following delivery.
In this chapter, we describe the isolation and purification of villous cy-
totrophoblast cells from the term placenta. In situ, these cells serve as precur-
sor cells for the continually regenerating syncytiotrophoblast, and are located
203
16_Petroff_203_218_F 203 8/29/05, 11:20 AM

204 Petroff et al.
Fig. 1. Histology of a term placental villus. Arrows denote villous cytotrophoblast

cells. sTB, syncytiotrophoblast; IVS, intervillous space.
basally in relation to the syncytiotrophoblast (Fig. 1). Upon histological ex-

amination of the term placenta, the cells are seemingly rare because they are
present in a discontinuous layer beneath the syncytiotrophoblast. However, the
total quantity of cytotrophoblast cells is highest at term in comparison with
earlier stages of pregnancy as a result of their continual proliferation and to the
presence of innumerable villi (1). Thus, the term placenta is a rich source of
trophoblast cells.
Once isolated, term cytotrophoblast cells rapidly lose the capacity to prolif-
erate (2). Cultured cytotrophoblast cells undergo a limited amount of sponta-
neous morphological and endocrinological differentiation, as evidenced by
formation of multinucleated syncytia and a small rise in secretion of hCGβ
(Fig. 2 and ref. 2). In contrast, with the addition of recombinant cytokines that
target epitheloid cells such as epidermal growth factor (EGF), the cells un-
dergo robust differentiation, forming multinucleated syncytiotrophoblast-like
monolayers that secrete high amounts of hCGβ, placental lactogen, and proges-
terone. Cytotrophoblast culture models have proven to be very useful for the
study of endocrinological and immunological functions of the placenta, cy-
totrophoblast differentiation and apoptosis, and mechanisms of infectious and
noninfectious disease (3–9).
16_Petroff_203_218_F 204 8/29/05, 11:20 AM

16_Petroff_203_218_F
205
205
Isolation and Culture of Human Trophoblast
Fig. 2. Immunomagnetically purified from the term placenta. (A) Cells bound to the immunomagnetic column, represent-
ing nontrophoblastic cells. (B) Cells not bound to the column, representing trophoblastic cells in a state of early differentia-
tion; (C) Cells not bound to the column and treated with 10 ng/mL epidermal growth factor (EGF); these cells have undergone
extensive syncytialization. Cells were plated in 60 mm Primaria dishes at a density of approx 5 × 106 per dish (A) or 6 × 106
(B,C) in Iscove’s modified Dulbecco’s medium supplemented with 10% fetal bovine serum. After allowing adherence for 4
h, the nonadherent cells were removed by gentle washing. Thereafter, media were removed and replaced every 48 h 6 d.
8/29/05, 11:20 AM
205
206 Petroff et al.
Table 1
Summary of Immunostaining After Purification
Protocol (Reference) Steps Markers Staining
Kliman et al. (10) Percoll gradient hCG (sTB) 1–5%
hPL (sTB) Absent
SP1 (sTB) Absent
Vimentin (non-TB) Absent
Chymotrypsin (non-TB) Absent
Douglas & King (11) Percoll gradient and
Immunopurification
(Anti-HLA class I/II) Cytokeratin-18 (CTB) 100%
Vimentin (non-TB) Absent
Guilbert et al. (12) Immunopurification PLAP (sTB) 4–46%
(Anti-CD9 and
Anti-MHC class I/II)
Current protocol Percoll gradent and β-hCG (sTB) <1%
immunopurification Cytokeratin-7 (sTB, CTB) ~100%
(Anti-HLA class I) CD14 (Macrophage) <1%
CD9 (non-CTB) <3%
Abbreviations: CTB, cytotrophoblast; hCG, human chorionic gonadotropin; hPL, human placental
lactogen; PI, propidium iodide; PLAP, placental alkaline phosphatase; PS, phosphatidyl serine; SP1,
pregnancy-specific β1-glycoprotein; sTB, syncytiotrophoblast; TB, trophoblast; FACS, fluorescence-
activated cell sorting.
Most published methods for purification of cytotrophoblast cells are based

on methodology originally developed by Kliman et al. (10–13; Table 1). Stan-
dard methods involve digestion of villous tissue in a mixture of trypsin and
DNase I, followed by density gradient centrifugation to remove cellular debris,
red blood cells, and cells of higher density such as leukocytes. Some investiga-
tors utilize the resulting population directly, although purity is relatively low
and cultures will eventually be overrun by proliferating placental fibroblasts
(ref. 14 and unpublished observations). Purity can be improved to nearly 100%
by inclusion of a negative selection procedure to eliminate nontrophoblast cells.
To this end, villous fibroblasts, endothelial cells, and macrophages are bound
and immobilized by antibody-conjugated beads. Antibodies may include anti-
class I and class II human leukocyte antigen (HLA), CD9, or hepatocyte growth
factor activator inhibitor (HAI)-1 (11,13,15,16). A final purification step is
simple adherence to tissue culture vessels, which serves to eliminate contaminat-
ing multinucleated syncytial fragments (12).
16_Petroff_203_218_F 206 8/29/05, 11:20 AM

Fig. 3. Schematic representation of the trophoblast purification method. (1) Villous

tissue is dissected, minced, and dissociated in trypsin/DNase I. (2) The resulting cell
suspension is layered over a Percoll gradient, centrifuged, and thus fractionated on the
basis of size and cellular density. (3) The semi-purified cells are incubated with mouse
anti-human leukocyte antigen (HLA) class I and anti-mouse immunoglobulin (Ig)G-
conjugated magnetic beads, and passed through a magnetic column. Nontrophoblast
cells, which bind the HLA antibody, are retained in the column, whereas trophoblast
cells pass through.
The methods described in this chapter outline the enzymatic dispersion, den-
sity gradient centrifugation, and HLA class I-depletion of trophoblast cells
(Fig. 3). The latter step exploits the fact that virtually every placental cell type
other than the villous cytotrophoblast cell expresses classical HLA class Ia
molecules, so that these contaminating cell types are depleted. We routinely
harvest >97% pure cytotrophoblast cells based on immunocytochemical stain-
ing for cytokeratin-7 and the absence of CD14-positive cells.
16_Petroff_203_218_F 207 8/29/05, 11:20 AM

208 Petroff et al.
2. Materials
2.1. Collection of Semipurified Trophoblast Cells
1. Sterile field sheets (Baxter, Deerfield, IL).
2. Sterile 4 in. × 4 in. gauze pads (Baxter).
3. Two pairs of small straight blade, sharp point scissors (Fine Science Tools, North
Vancouver, BC).
4. Two pairs of fine-toothed forceps (Fine Science Tools).
5. Sterile 250-mL beaker for tissue collection, preweighed.
6. Sterile 1-L beaker for liquid waste.
7. Sterile 500-mL Erlenmeyer flask for tissue dissociation.
8. Sterile 150-mm Petri dish for tissue mincing.
9. Cell dissociation sieve, fitted with 40 mesh screen (Sigma-Aldrich, St. Louis,
MO; cat. no. CD-1).
10. Benchtop water-bath shaker (for example, LabLine model SHKE7000; Barnstead
Intl., Dubuque, IA).
11. Shandon cytospin centrifuge (Shandon, Pittsburgh, PA).
12. 15-mL sterile polypropylene or polystyrene centrifuge tubes.
13. 100-µm Falcon nylon mesh cell strainers (Becton Dickinson, Franklin Lakes, NJ,
cat. no. 2360).
14. Biohazard bags.
15. Bleach.
16. Falcon Opticul 50-mL sterile polypropylene centrifuge tubes (Becton Dickinson).
17. Cryogenic cell storage vials.
18. Cell-freezing chamber.
19. 1 L 0.9% NaCl: dissolve 9 g NaCl in deionized H2O to yield a 1 L solution.
Filter-sterilize (0.2 µm).
20. 10X Hank’s balanced salt solution (10X HBSS), 1 L: 5.36 mM KCl (4 g), 4.4 mM
KH2PO4 (0.6 g), 1.37 M NaCl (80 g), 3.37 mM Na2HPO4 ( 0.4788 g), 55.5 mM
D-glucose (10 g). Dissolve each of the constituents in deionized H2O to yield 1 L
solution. Filter-sterilize (0.2 µm).
21. 1X Ca/Mg-free HBSS (CMF-Hank’s): Combine 100 mL 10X HBSS with 25 mL
1-M HEPES (Sigma, cat. no. H-0887) with deionized H2O to yield 800 mL H2O.
Adjust pH to 7.4 and bring volume to 1 L. Filter sterilize.
22. Enzyme digestion buffer: combine 35 mL 10X HBSS with 1.65 mL 7.5% Na
Bicarbonate (Sigma, cat. no. S-8761), 8.75 mL 1-M HEPES, and 266.1 mL deion-
ized H2O; filter-sterilize and distribute into three bottles containing 133.5 mL,
89 mL, and 66.8 mL. Store at 4°C.
23. 2.5% Trypsin: 10X concentrate (Invitrogen, cat. no. 15090-046). Thaw a 100-mL
bottle of trypsin and distribute 33 mL over three sterile tubes. Store at –20°C, and
thaw just prior to use.
24. DNase I: (Sigma, cat. no. D-5025, 150,000 U, approx 50 mg solid, ~3000 U/mg).
Just before use, add 5 mL sterile 0.9% NaCl to a 150,000-U vial of DNase. Filter-
sterilize and keep on ice until use or store at –20°C for later use.
16_Petroff_203_218_F 208 8/29/05, 11:20 AM

25. Enzyme digestion buffer (working solution):

Batch no. I II III
Digestion buffer 133.5 mL 89 mL 66.8 mL
10X trypsin 15 mL 10 mL 7.5 mL
DNase 1.5 mL 1 mL 0.75 mL
Just prior to each of the three digestion stages, add the appropriate volume of
trypsin and DNase to prewarmed enzyme digestion buffer. When trypsin and
DNase are added, concentrations will be 1X HBSS, 25 mM HEPES, 0.25%
Trypsin, and approx 300 U/mL DNase.
26. Cytotrophoblast cell culture medium: Iscove’s modified Dulbecco’s medium
(IMDM) (Mediatech, Herndon, VA; cat. no. 15-016-CV), containing 10% heat-
inactivated FBS (56°C for 30 min) (Atlanta Biologicals, Norcross, GA; cat. no.
S11150), 2 mM L-glutamine (Mediatech, Cat. No. 25-005-CI), and 1X penicillin/
streptomycin/amphotericin B (100X solution: Mediatech cat. no. 30-004-CI).
27. Percoll gradients (see Note 1).
a. Prior to preparing gradients, be sure to mix Percoll well, as Percoll undergoes
spontaneous formation of gradients. Prepare 90% Percoll stock: 117 mL
Percoll (Sigma, cat. no. P-4937) + 13 mL sterile 10X HBSS.
b. Into sterile 50 mL tubes, prepare 14 dilutions of the Percoll using the 90%
stock solution with sterile CMF-Hank’s, referring to Table 2 for appropriate
volumes.
c. Starting with the 70% solution of Percoll, slowly layer 3 mL of each concen-
tration into the Falcon Opticul 50 mL conical centrifuge tubes (see Note 2).
Expected fluid levels of each concentration are shown in Fig. 4. Store the
gradients at room temperature away from disturbance.
28. Percoll Density Marker Bead Kit (Optional) (Sigma, cat. no. DMB-10). Density
marker beads may be used to verify the accuracy of the Percoll gradient prepara-
tion. To use, resuspend each vial of beads (nine vials total) in water, and mix 40 µL
of each into a centrifuge tube. Slowly pipet the 360-µL mixture onto the surface,
and centrifuge at 1200g for 20 min at room temperature. This can be done along-
side an actual cell preparation. Approximate locations of the density marker beads
after centrifugation are shown in Fig. 4, but may vary according to lot number.
29. Cell freezing medium: 10% dimethylsulfoxide in FBS.
2.2. Immunomagnetic Purification of Cytotrophoblast Cells

1. Human placenta villous mononuclear cells purified by density gradient centrifu-
gation.
2. MACS® separation system (Miltenyi Biotec, Auburn, CA): magnet (separation
unit) (cat. no. 130-042-10); MultiStand (cat no. 130-042-303), MS+ Separation
Columns (cat. no. 130-042-201); preseparation filters (cat. no. 130-041-407).
3. Primary antibody: mouse W6/32 (anti-human HLA-ABC). This antibody recog-
nizes a nonpolymorphic epitope of HLA class I molecules that are associated
16_Petroff_203_218_F 209 8/29/05, 11:20 AM

210 Petroff et al.
Table 2
Dilution Scheme for Preparation of Percoll Gradients
90% Percoll (mL) CMF-Hanks (mL) Final concentration
15.6 4.4 70%
14.4 5.6 65%
13.3 6.7 60%
12.2 7.8 55%
11.1 8.9 50%
10.0 10.0 45%
8.9 11.1 40%
7.8 12.2 35%
6.7 13.3 30%
5.6 14.4 25%
4.4 15.6 20%
3.3 16.7 15%
2.2 17.8 10%
1.1 18.9 5%
with β2-microglobulin (17,18). It can be purchased from several vendors

(eBioscience, San Diego, CA; Novus Biologicals, Littleton, CO); alternatively,
the hybridoma may be obtained from ATCC (Manassas, VA).
4. Secondary antibody: MACS goat anti-mouse IgG microbeads (Miltenyi Biotec,
cat. no. 130-048-401).
5. Cell separation buffer (CSB): Dulbecco’s phosphate-buffered saline (PBS) with-
out Ca2+ or Mg2+ (Sigma, cat. no. D-8537) containing 0.5% bovine serum albu-
min (BSA) (Sigma, cat. no. A-9056) and 2 mM ethylenediamine tetraacetic acid
(EDTA) (Invitrogen, cat. no. 15575-038). Combine solutions and degas under
vacuum.
6. DNase (Sigma, cat. no. D-5025). DNase left over from the villous cytotropho-
blast preparation may be used.
7. Cytotrophoblast culture medium: IMDM supplemented with 10% heat-inacti-
vated FBS (heat-inactivated at 56°C for 30 min; Atlanta Biologicals, cat. no.
S11150), 2 mM L-glutamine (Cellgro, cat. no. 25-005-CI), and 1X penicillin/
streptomycin/amphotericin B (100X solution; Cellgro, cat. no. 0-004-CI).
3. Methods
3.1. Collection of Semipurified Trophoblast Cells (see Note 3)
1. Obtain a human placenta and process as soon as possible after delivery (see Note
4). If variation due to potential effects of labor is a concern, placentas from Cae-
sarian section deliveries may be used. Always use caution when handling human
biological material.
16_Petroff_203_218_F 210 8/29/05, 11:20 AM

Fig. 4. Schematic representation of expected locations of diluted Percoll solutions,

density marker beads, and cell types. PMN, polymorphonuclear cells; RBC, red blood
cells.
2. Place the placenta on a sterile field, with the maternal side (basal plate) facing up.
Prepare histological, RNA or protein samples if desired (see Note 5). Using sharp,
fine-point scissors and blunt forceps, dissect one cotyledon at a time. Remove the
overlying basal plate tissue, about 3 mm from the surface. Avoiding the chori-
onic plate, collect 40–50 g of villous tissue into the preweighed 250-mL beaker.
3. Rinse tissue several times with 0.9% NaCl by swirling with forceps, using the
1-L beaker for liquid waste. Transfer all of the tissue to a 150-mm Petri dish
and mince finely with scissors. Transfer half the tissue to the cell dissociation
sieve and rinse with 0.9% NaCl extensively until the eluate becomes clear. Trans-
fer the minced tissue to a 500-mL sterile Erlenmeyer flask, and repeat with the
second half of the tissue.
16_Petroff_203_218_F 211 8/29/05, 11:20 AM

212 Petroff et al.
4. The dissociation is performed in three stages. During the dissociation of the sec-
ond and third batches, the cell suspensions from the first and second batches are
centrifuged and resuspended in turn. To prewarmed enzyme dilution buffer (labeled
batch I), add DNase and trypsin as indicated above. Add the mixture to the Erlen-
meyer flask containing the tissue, and incubate for 15–20 min at 37°C in a rotat-
ing water-bath shaker (150 rpm). During this incubation, add 1.5 mL FBS to
22 15-mL centrifuge tubes.
5. After the first 15-min dissociation, set the digestion flask at a tilt until tissue
settles. Remove about 13.5 mL of the supernatant, taking care not to collect undis-
sociated tissue. Slowly layer the suspension over the 1.5 mL serum in 15-mL coni-
cal centrifuge tubes. Repeat for seven additional tubes or until most of the
digestion supernatant is transferred. Centrifuge the tubes at 1000g for 15 min at
25°C at room temperature.
6. While batch I is in the centrifuge, add 100 mL warm enzyme digestion solution
containing enzymes (labeled batch II) and repeat digestion. Again collect super-
natant and layer over FBS as for batch I. Repeat digestion a third (final) time.
7. For each batch, following centrifugation, aspirate the supernatant without dis-
turbing the pellet. The trophoblast cells are predominantly in the white portion of
the pellet, overlying the red blood cells.
8. Resuspend the cell pellet in each tube in 1 mL culture medium, and combine the
resuspended pellets for each digestion group. Hold cells at room temperature.
9. After collecting the cells from all three digestion stages, filter the suspension
using a 100-µm nylon cell strainer inserted in the top of a sterile 50-mL conical
centrifuge tube. If the filtration of the cell suspension slows, lift upward on the
filter to draw a vacuum within the tube. Centrifuge at 1000g for 10 min, and
resuspend in 6 mL CMF-Hank’s. The total volume of the resuspended cells
should be approx 8 mL.
10. Carefully layer half of the cells onto each of two preformed Percoll gradients.
Centrifuge the gradients at 1200g for 20 min at room temperature in a swinging
bucket rotor without a brake.
11. Aspirate upper diffuse “band” (Fig. 4; usually down to approx 25–30 mL mark on
tube), and use a Pasteur pipet fitted with a bulb to manually collect the cells that
fractionate near center of tube into each of two sterile 50-mL tubes. The band con-
taining the cytotrophoblast cells is diffuse and usually between the 30-mL and 12-
to 15-mL graduations on tube. If using a separate reference tube containing density
marker beads, the cells will correspond to the location between the 1.048 and 1.060
marker beads. Dilute the cell fractions fourfold with culture medium and centrifuge
at 1000g for 5 min. Resuspend cells in each tube in 10 mL culture medium, com-
bine, and determine yield by trypan blue dye exclusion. Typical yields are between
1.5 and 3 × 108 cells per approx 40 g tissue at greater than 95% viability.
12. To assess cell characteristics by immunocytochemistry, cytocentrifuge slides can
be prepared. For one spot, use 50,000–100,000 cells in a 200-µL vol containing
at least 50% serum. Centrifuge in the Shandon cytospin unit at 700 rpm for 3
min. Remove funnel, turn slide and filter paper end-for-end in holder, replace
funnel and repeat. Place slides on bench to air dry, and store slides at –20°C.
16_Petroff_203_218_F 212 8/29/05, 11:20 AM

13. If not proceeding immediately with immunopurification or cell culture, centri-

fuge remaining cells (approx 250g for 10 min) and resuspend in freezing medium
at 1 × 107 cells/mL. Aliquot 1 mL cells per cryovial, place vials in a room tem-
perature cell freezing unit and store at –80°C overnight. Transfer cryovials to a
liquid nitrogen freezer the next day.
3.2. Immunomagnetic Purification of Cytotrophoblast Cells
1. If starting from a frozen stock of villous mononuclear cells, thaw cells quickly in a
37°C water bath. Transfer the cells into a 50-mL conical centrifuge tube and resus-
pend slowly in about 20 mL cool cytotrophoblast culture medium per 5 × 107 cells.
2. Centrifuge at 200g for 10 min at 4°C and resuspend in 10 mL cold CSB. Keeping
the cells on ice, count the cells and determine viability using trypan blue dye.
3. If cell clumping is evident, treat the cells with DNase as follows: centrifuge at
200g for 10 min at 4°C; resuspend in 20 mL cool (4°C) D-PBS (pH 7.0–7.2)
containing 0.5 mM MgCl2, 1 mM CaCl2, 0.5% BSA, and 100 U/mL DNase. Incu-
bate for 45 min at room temperature (25°C) or 15 min at 37°C in a shaking water
bath. Centrifuge at 200g for 10 min at 4°C. Carefully remove supernatant. Resus-
pend in CSB and check for clumping. Repeat DNase digestion until no DNA
remains (cell clumping). Repeat counting of cells.
4. Transfer cells to sterile 15-mL conical tube(s) (no more than 5 × 107 per tube),
centrifuge again, and resuspend in 500 µL CSB per 5 × 10 7 cells containing
40 µg/mL W6/32. The cell suspension will be very viscous; mix well by flicking
the tube several times and incubate for 10 min in the refrigerator (6–12°C).
5. After the incubation, add 3 mL cold CSB and centrifuge at 300g for 10 min.
Remove and discard the supernatant, add 3 mL cold CSB, and repeat centrifuga-
tion. Resuspend cells in 400 µL (per 5 × 107 cells) CSB.
6. Label the cells with MACS anti-mouse IgG MicroBeads. To the 400 µL cells,
add 100 µL (per 5 × 107 cells) microbeads to the cell suspension and incubate
15 min in the refrigerator.
7. Following the 15 min incubation of secondary antibody with cells, wash the cells
once by adding 3 mL CSB and centrifuging as described above. Resuspend cells
in 0.5 mL CSB per ≤ 5 × 107 cells; keep on ice.
8. Prepare the MACS separation system. Attach the MiniMACS magnet(s) to the
MACS MultiStand and place an MS+ separation column in the MiniMACS mag-
net. To the column, attach a 23–27-gauge needle to act as a flow resistor (we use
25 gauge × five-eighths-inch needles). Prepare the column by applying 500 µL
ice-cold CSB to the top of the column; let the buffer flow through and discard the
effluent. Prewet the preseparation filter by vigorously pipetting 0.5 mL ice-cold
CSB onto the membrane. Place the filter atop the separation column. Place a
clean 15-mL centrifuge tube under the column. This will serve to collect the
purified trophoblast cells.
9. Separate the immunomagnetic bead-labeled cells in the prepared MiniMACS
column. Apply 500 µL cell suspension to the preseparation filter. Allow the cell
suspension to run through the prefilter and column, collecting the effluent in a
sterile tube (see Note 6). A yellow-tipped micropipet can be used to aspirate any
cellular suspension from the underside of the preseparation filter.
16_Petroff_203_218_F 213 8/29/05, 11:20 AM

214 Petroff et al.
10. Rinse the tube that contained the cells with 500 µL ice-cold CSB and apply to
column. Wash the column two to three times with 500 µL ice-cold CSB, each
time collecting in same effluent tube. If using more than one column, combine
effluents from the columns.
11. If desired, recover the bound cells (i.e., noncytotrophoblastic cells that remain
trapped within the column). Remove column from separator by pulling the col-
umn horizontally out of the magnet. Remove flow resistor from bottom of col-
umn, and place column above a fresh collection tube. Add 1 mL ice-cold CSB
onto the column. Using the plunger supplied with the column, flush cells from
column.
12. Centrifuge the cells at 400g and resuspend in cytotrophoblast culture medium.
Prepare a dilution in trypan blue and determine yield and viability. Typical yields
are 60–70% of the starting population. For characterization, prepare cytospin
preparations of each cell type (see Note 7).
13. Plate the purified cells in culture medium at a minimum of 2 × 105 cells/cm2 in
appropriate culture vessels (see Note 8). Allow cells to attach to the surface of
the plate for 4–24 h, after which rinse twice with prewarmed culture medium to
remove the nonattached cells. Refresh medium every 24–48 h.
4. Notes
1. This procedure allows for preparation of six tubes of stepwise 5–70% Percoll
gradients. Each cytotrophoblast preparation from 40 g placental tissues requires
two of these gradients; therefore, six are sufficient for three cytotrophoblast cell
preparations. All six gradients can be prepared at one time, but for best results,
use the gradients within 24 h of preparation. The diluted Percoll solutions may be
stored for later preparation of gradients.
2. Preparation of Percoll gradients is time-consuming. Some laboratories utilize
peristaltic pumps to reduce the workload. We have found that pipetting time can
be greatly reduced by resting the tip of the 5-mL pipet on the side of the tube just
above the liquid level and gently swinging the tip side-to-side against the tube to
induce layering of a broad stream of liquid. The result is that liquid can be pipetted
at a faster rate because the broader stream will reduce the pressure per unit of
surface area of the liquid.
3. We usually reserve 2 or 3 d for isolation of semi-purified cytotrophoblast cells.
The first day is devoted to preparation of reagents and Percoll gradients, and the
second day to the purification procedure itself. At the conclusion of the second
day, one may either cryopreserve the semipurified cytotrophoblast cells, or pro-
ceed directly to the final step of trophoblast isolation, immunomagnetic purifica-
tion (see Subheading 3.2.).
4. In the United States, term placentas are regarded as discarded tissues and can
therefore be obtained without patient consent. Obtaining additional information
about the medical background of the patient will be subject to regulation by the
Health Insurance Portability and Accountability Act (HIPAA), and investigators
must go through the appropriate channels to ensure privacy and consent of the
patient.
16_Petroff_203_218_F 214 8/29/05, 11:20 AM

5. It is often desirable to obtain histological specimens that include basal plate pla-
centa as well as villous tissue, because the basal plate tissue contains extravillous
trophoblast cells. To do this, score a 1-mm × 3-mm rectangular portion of the
surface of the basal plate with sharp scissors. Make the deeper cuts, blotting away
unclotted blood if necessary, and use blunt, flat-tipped forceps to gently lift the
tissue upwards. Trim the tissue to the desired size and fix or freeze as necessary.
To obtain amniochorion histological samples, lay a section of amniochorion
membrane on a sterile field with the amnion face downward. Trim a 1-mm × 5-mm
slice, and either fix in desired fixative or, for frozen histology, use the scissors
and forceps to roll the membrane. Place the roll on its side for embedding; upon
microscopy, it will be possible to simultaneously observe the amnion, chorion,
and decidua.
For extraction of RNA and/or protein, we usually obtain specimens from the
villous portion of the placenta after removal of the basal plate connective tissue.
This can therefore be done concomitantly with collection of the tissue to be dis-
sociated. We collect and pool these samples from several cotyledons to avoid
potential intercotyledonary differences. To snap-freeze, samples can be dropped
directly into liquid nitrogen in an RNase-free container and subsequently trans-
ferred into sterile, precooled polypropylene tubes for storage at –80°C.
6. If the MACS columns run slowly, there are several steps that may be taken to
improve flow. First, a micropipet fitted with a yellow tip can be used to disrupt
any cells that may have settled on the top of the column. Without introducing
bubbles into the cell suspension over the column, pipet up and down to dislodge
settled cells. If this fails to speed up the column, use the plunger supplied with
the column to introduce light pressure. Place the plunger over the head of the
column and gently press down. Applying too much pressure by forcing the
plunger into the column will shear and detach the nontrophoblast, antibody/bead-
bound cells from the magnet, thus contaminating the eluted cell population. If
application of pressure fails to hasten the elution, we have found that the problem
often resides within the needle, which can get clogged with cells. Thus, the needle
can simply be removed and replaced with a new, sterile needle. If the problem is
indeed due to the needle, the flow will immediately hasten upon removal of the
clogged needle; thus, a new needle must be in place as quickly as the old one is
removed. As a last resort, the column itself can be replaced, and the unfiltered
cells transferred to a new column. However, this will result in the sacrifice of any
cells that may be trapped within the original column.
7. At this point, the cells can be analyzed by flow cytometry (19) or immunocy-
tochemistry on cytospin preparations. By the latter method, we have used anti-
cytokeratin-7 (Dako, Carpinteria, CA; clone OVTL 12/30), anti-CD14 (Zymed,
San Francisco, CA; clone RPA-M1), and anti-hCGβ (Neomarkers, Freemont, CA;
clone CG05) (4,20).
8. Cytotrophoblast cells may be cultured on Permanox Chamber Slides (Nalge Nunc).
Although adherence to this type of plastic is better than glass chamber slides, it is still
relatively poor. Thus, we seed at a very high density to guarantee that sufficient num-
bers of cells attach to the slide (e.g., seed 300,000 cells per well in eight-well slides).
16_Petroff_203_218_F 215 8/29/05, 11:20 AM

216 Petroff et al.
Acknowledgments
The authors thank Mr. Stanton Fernald and Ms. Erin Skorina for their assis-
tance in preparation of graphics. This work was supported by grants R01
HD045611 (M.G.P), P01 HD39878 Project III (J.S.H.), U54 HD33994 (J.S.H)
Project III, and R01 HD24212 (J.S.H) from the National Institutes of Health.
References
1. Benirschke, K. and Kaufmann, P. (2000) Pathology of the Human Placenta.
Springer-Verlag, New York: pp. 55–56.
2. Morrish, D. W., Dakour, J., Li, H., et al. (1997) In vitro cultured human term
cytotrophoblast: a model for normal primary epithelial cells demonstrating a spon-
taneous differentiation program that requires EGF for extensive development of
syncytium. Placenta 18, 577–585.
3. Kamat, A., Alcorn, J. L., Kunczt, C., and Mendelson, C. R. (1998) Characteriza-
tion of the regulatory regions of the human aromatase (P450arom) gene involved
in placenta-specific expression. Mol. Endocrinol. 12, 1764–1777.
4. Petroff, M. G., Chen, L., Phillips, T. A., Azzola, D., Sedlmayr, P., and Hunt, J. S.
(2003) B7 family molecules are favorably positioned at the human maternal-fetal
interface. Biol. Reprod. 68, 1496–1504.
5. Morales, P. J., Pace, J. L., Platt, J. S., et al. (2003) Placental cell expression of
HLA-G2 isoforms is limited to the invasive trophoblast phenotype. J. Immunol.
171, 6215–6224.
6. Huppertz, B., Frank, H. G., Reister, F., Kingdom, J., Korr, H., and Kaufmann, P.
(1999) Apoptosis cascade progresses during turnover of human trophoblast: analy-
sis of villous cytotrophoblast and syncytial fragments in vitro. Lab. Invest. 79,
1687–1702.
7. Ka, H. and Hunt, J. S. (2003) Temporal and spatial patterns of expression of in-
hibitors of apoptosis in human placentas. Am. J. Pathol. 163, 413–422.
8. Abbasi, M., Kowalewska-Grochowska, K., Bahar, M. A., Kilani, R.T., Winkler-
Lowen, B., and Guilbert, L. J. (2003) Infection of placental trophoblast by Toxo-
plasma gondii. J. Infect. Dis. 188, 608–616.
9. Li, H., Dakour J., Kaufman S., Guilbert, L. J., Winkler-Lowen, B., and Morrish,
D.W. (2003) Adrenomedullin is decreased in preeclampsia because of failed
response to epidermal growth factor and impaired syncytialization. Hyperten-
sion 42, 895–900.
10. Kliman, H. J., Nestler, J. E., Sermasi, E., Sanger, J. M., and Strauss, J. F. (1986)
Purification, characterization, and in vitro differentiation of cytotrophoblasts from
human term placentae. Endocrinology 118, 1567–1582.
11. Douglas, G. C., and King, B. F. (1989) Isolation of pure villous cytotrophoblast
from term human placenta using immunomagnetic microspheres. J. Immunol.
Methods 119, 259–268.
12. Guilbert, L. J., Winkler-Lowen, B., Sherburne, R., Rote, N. S., Li, H., and Morrish,
16_Petroff_203_218_F 216 8/29/05, 11:20 AM

D. W. (2002) Preparation and functional characterization of villous cytotrophoblasts

free of syncytial fragments. Placenta 23, 175–183.
13. Potgens, A. J., Kataoka, H., Ferstl, S., Frank, H. G., and Kaufmann, P. (2003) A
positive immunoselection method to isolate villous cytotrophoblast cells from first
trimester and term placenta to high purity. Placenta 24, 412–423.
14. Morrish, D. W., Shaw, A. R., Seehafer. J., Bhardwaj, D., and Paras, M. T. (1991)
Preparation of fibroblast-free cytotrophoblast cultures utilizing differential ex-
pression of the CD9 antigen. In Vitro Cell. Dev. Biol. 27A, 303–306.
15. Yui, J., Garcia-Lloret, M., Brown, A. J., et al. (1994) Functional-long-term cul-
tures of human term trophoblasts purified by column-elimination of CD9 express-
ing cells. Placenta 15, 231–246.
16. Nagamatsu, T., Fuji, T., Ishikawa, T., et al. (2004) A primary cell culture system
for human cytotrophoblasts of proximal cytotrophoblast cell columns enabling in
vitro acquisition of the extra-villous phenotype. Placenta 25, 153–165.
17. Brodsky, F. M., Bodmer, W. F., and Parham, P. (1979) Characterization of a mono-
clonal anti-beta 2-microglobulin antibody and its use in the genetic and biochemical
analysis of major histocompatibility antigens. Eur. J. Immunol. 7, 536–545.
18. Ladasky, J. J., Shum, B. P., Canavez, F., Seuanez, H. N., and Parham, P. (1999)
Residue 3 of beta 2-microglobulin affects binding of class I MHC molecules by
the W6/32 antibody. Immunogenetics 49, 312–320.
19. Potgens, A. J., Gaus, G., Frank, H. G., and Kaufmann, P. (2001) Characterization
of trophoblast cell isolations by a modified flow cytometry assay. Placenta 22,
251–255.
20. Ka, H. and Hunt, J. S. (2003) Temporal and spatial patterns of expression of in-
hibitors of apoptosis in human placentas. Am. J. Pathol. 163, 413–422.
16_Petroff_203_218_F 217 8/29/05, 11:20 AM

16_Petroff_203_218_F 218 8/29/05, 11:20 AM
Human Trophoblast Cell Line Production 219
17
Production of Human Trophoblast Cell Lines
Guy St J. Whitley
Summary
Our understanding of important biological phenomena such as invasion, migration, and
apoptosis has advanced greatly through the prudent use of in vitro models based on isolated
cells in culture. Established cell lines are readily manipulated using simple molecular biologi-
cal techniques and their abundance as homogenous populations allows the rapid accumulation
of statistically significant data. The study of human trophoblast function in vitro has been ham-
pered by the difficulties associated with obtaining and culturing primary human trophoblasts
including the paucity of material and contamination with other cell types. This chapter describes
a cheap and simple method for the production of human trophoblast cell lines using poly-L-orni-
thine. It details the production, isolation and initial characterization of these cells and provides
simple tips on how to store and maintain a bank of cells for future needs.
Key Words: Cytotrophoblasts, transfection, cell line, poly-L-ornithine, SV40.
1. Introduction
The power of interventional over purely observational studies is immense
and our understanding of important biological functions such as cell division,
apoptosis and migration has advanced significantly following studies using
cells grown in culture. The two options for these experiments are either to use
primary cells isolated from fresh tissue or to use established cell lines. It is the
aim of this chapter to describe the methods used to generate cell lines. Cell
lines may be obtained spontaneously as out-growths from malignant tissue or
from normal primary cultures that have been manipulated to express viral
oncogenes which overcome the cells normal growth controls.
The study of human trophoblast function in vitro has been hampered by the
difficulties associated with obtaining and culturing trophoblasts. Although pure
preparations of the different subpopulations of trophoblast can be obtained by
fluorescence-activated cell sorting (1) yields can be low and the procedures
require specialized equipment. Pure extravillous trophoblasts may be obtained
219
17_Whitley_219_228_F 219 8/29/05, 11:20 AM

220 Whitley
following out-growth from chorionic explant cultures (2), although this again
results in relatively low cell yields. Normal human cells cultured in vitro have
a limited proliferative lifespan, dividing a given number of times before they
undergo a permanent growth arrest, known as replicative senescence. This
seems particularly true of trophoblasts cells that grow poorly in culture. Other
problems associated with using primary cells are heterogeneity between prepa-
rations, phenotypic instability of the isolated cells and, in some circumstances,
availability of tissue. Established cell lines would therefore seem to offer a
number of advantages since they are homogenous populations that grow readily
in culture providing a plentiful supply of cells with a stable phenotype over
many passages. Conversely, the methods employed to extend lifespan (trans-
fection or spontaneous immortalization/transformation) may alter regulation
of cell division that could impact on differentiated functions and gene expres-
sion. However, it should be emphasised that no matter which route is taken
(either primary culture or established cell line), no in vitro model is perfect.
Therefore, one of the most important considerations when using in vitro mod-
els is that we recognize their limitations and design the experiments and inter-
pret the results accordingly.
A number of human trophoblast cell lines already exist; some of the more
widely used lines are listed in Table 1. Early trophoblast lines were derived
from choriocarcinomas and these include BeWo, JEG, and JAR (Table 1).
These lines have proved useful in many studies, but may have characteristics
related to their malignant origins. More recently, a number of lines that have
retained phenotypic function have been reported that arose spontaneously from
primary cultures of normal first trimester chorionic villous explants or chori-
onic villous samples; these lines include the HTR-8 (3) and the ED series of
trophoblast cell lines (4,5). Unfortunately, caution should now be taken when us-
ing ED27 as they have at some stage become contaminated with the nontrophoblast
cell line HeLa (6).
Transfection of cells with viral oncogenes from Simian virus-40 (SV40),
adenovirus, or human papilloma virus (HPV) have been used to establish lines.
The most common of these used in the production of trophoblast cell lines is
the early region of SV40 containing the small and large T-antigen. The large
T-antigen binds to and inactivates the protein products of p53 and retinoblas-
toma (RB), two tumor suppressor genes. Expression of the T-antigen increases
the proliferation of the cells in culture and delays the onset of cellular senes-
cence. As increased growth can lead to a reduction in differentiated function,
constructs containing conditional mutants including the temperature sensitive
variants of SV40 have been used to generate these lines. At the permissive
temperature the large T-antigen is expressed and functional and the cells pro-
liferate however at the nonpermissive temperature the stability of the mutant
17_Whitley_219_228_F 220 8/29/05, 11:20 AM

Table 1
Commonly Used Human Trophoblast-Derived Cell Lines
Cell line Source Means of production Reference
BeWo Choriocarcinoma Spontaneous 24
ED27 First trimester Spontaneous 5
ED31 5
ED77 4
HP-A1 Term placenta Infection Adenovirus(ori)SV40tsA209 9
HP-A2 Infection Adenovirus(ori)SV40tsA209
HP-W1 Infection Adenovirus(ori)SV40wt
HT Term placenta Spontaneous 25
HT-116 First trimester Spontaneous 26
HTR-8 First trimester Spontaneous 3
HTR-8/SVneo HTR-8 Electroporation with pSV3neo 3
JAR Choriocarcinoma Spontaneous 27
JEG Choriocarcinoma Spontaneous 28
IST-1 First trimester villous HPV16 E6/E7 11
explant
NHT First trimester Spontaneous 29
NPC First trimester Spontaneous 30
RSVT-2, 2/C HTR-8 Electroporation with pRSVT 30
SGHPL-4 First trimester Poly-L-ornithine with pSV3neo 31
SPA-26 First trimester Infection SV40tsa255 8
TCL-1 Term choriodecidua pZipSV40 32
TL Term Spontaneous 33
gene product is compromised and division of the cells falls. The theory is that
at the nonpermissive temperature, the resulting cell will re-establish lost func-
tions. This type of construct has been used to produce both first and third tri-
mester trophoblast cell lines (7–9).
Although the expression of the T-antigen does extend the lifespan of the
cultures, it seldom produces immortal human cell lines. Eventually, cells
undergo a delayed phase of growth arrest and enter what is termed SV40-
crisis. This is associated with a reduction of telomere length (10). Recent work
has suggested that the expression of human telomerase catalytic component
could overcome this, although further work is required to determine if this has
adverse effects on the phenotype of the resultant cell lines.
Another commonly used construct for the production of cell lines is the
early region of HPV-16. This contains two gene products known as E6 and E7,
expression of which is also known to overcome cellular senescence in human
cells. Although E6 and E7 alone have been used to derive lines in some cell
17_Whitley_219_228_F 221 8/29/05, 11:20 AM

222 Whitley
types, a construct containing both E6 and E7 is considered more efficient and

has been used to produce trophoblast cell lines derived from first trimester
villous explants (11). It is reported that cells expressing HPV-16 E6–E7 are
nontumorigenic in nude mice and retain most of the phenotypic characteristics
of primary cultures.
We have spent a number of years developing a series of cell lines derived
from first-trimester extravillous trophoblasts. The resulting lines express many
markers characteristic of the extravillous trophoblast cells, including HLA-G,
hPL, and CD-9 (12,13). The best-characterized line is called SGHPL-4. This
cell line was produced following transfection with the large T-antigen of SV40.
Unlike some reported trophoblast cell lines, this line is neither immortal (although
rather long-lived) nor transformed, and has been used in a number of studies to
investigate the mechanisms regulating invasion, motility (14–17), and
apoptosis (18,19). In this chapter, we will describe the common ways to pro-
duce cell lines and give specific details of the method we have found most
useful for the production of extravillous trophoblast cell lines. We will discuss
the isolation of clones and the establishment of low-passage frozen stocks. The
characterization of newly produced cell lines will also be discussed.
2. Materials
1. Culture medium:
a. Standard growth medium will depend on the method of primary cell isola-
tion. We use Ham’s-F10 containing 10% (v/v) heat-inactivated fetal calf
serum (FCS), 2 mM glutamine and penicillin (100 U/mL), and streptomycin
(0.1 mg/mL).
b. Transfection medium Dulbecco’s modified Eagle’s medium (DMEM; Sigma,
Dorset UK) containing 2 mM glutamine, 100 U/mL penicillin and 0.1 mg/mL
streptomycin, and 5% (v/v) FCS.
c. Selection medium is Ham’s-F10 plus 0.3 mg/mL geneticin (G418) and 10%
(v/v) FCS.
d. Cell freezing and storage medium: standard growth medium containing 10%
(v/v) dimethylsulfoxide (DMSO; MERCK, Poole, Dorset, UK).
2. Poly-L-ornithine (Sigma Chemicals, cat. no. P4538) is prepared as a 5 mg/mL
stock in water, filter-sterilized through a 0.2-µm syringe and stored as 50-µL
aliquots at –20°C (see Note 1).
3. Chloroquine diphosphate 100 mM stock solutions (60 mg/mL in H2O) filter-ster-
ilized and stored in foil wrapped tubes at –20°C.
4. Geneticin is prepared fresh in medium and filter sterilized through a 0.2-µm sy-
ringe filter.
5. The recombinant plasmid used for transfection, pSV3-neo, contains both the large
and small T-antigens of the early region of SV40 and the bacterial neomycin
17_Whitley_219_228_F 222 8/29/05, 11:20 AM

phosphotransferase gene that confers resistance to the antibiotic geneticin (G418).

The plasmid pSV3-neo was originally described by Southern and Berg (20) and
is available from the American Tissue Culture Collection (Manassas, VA, cat.
no. 37150).
6. Antibodies for the identification of different trophoblast populations are CK7 (Dako
Corporation, Carpinteria, CA, cat. no. M 7018), HLA-G (Serotec, Oxford, UK, cat.
no. MCA2044) and CD9 (Dako, cat. no. M 7055) (see Note 2).
3. Methods
3.1. Transfection
The first and, possibly, most important step in the process of developing a
cell line is to obtain a primary culture rich in the cells of interest. The methods
for isolating and purifying different populations of trophoblast are presented
elsewhere in this volume, and so will not be considered here. Suffice it to say a
pure population of cells is not essential, but the greater the purity the greater
the chance of successfully obtaining useful transfectants. DNA introduced into
mammalian cells can be either retained in the cytoplasm and transiently expressed
or transported to the nucleus and integrated into the genome of the cell. Tran-
sient expression is appropriate for the production of proteins in relatively high
concentrations over short periods of less than 10 d but will not result in the
production of stable lines. Stable integration is a more rare event—perhaps as
few as 0.1% of the transfected cells will result in stable integration. It is there-
fore important to optimize the efficiency of transfection.
1. Seed cells into 9-cm dishes 24 h prior to transfection in order to achieve approx
80% confluence on the day of transfection (see Note 2).
2. For each 9-cm dish, place 3 mL of DMEM into a flat bottomed Bijou con-
tainer and add 10 µg of plasmid DNA and 5 µg/mL of poly- L-ornithine (see
Note 1). Mix gently and allow standing for 30 min at room temperature. Remove
the medium from the cells and add the DNA solution. Return the cells to the
incubator and gently rock the plate every 45 min (see Note 4).
3. After 6 h, aspirate the medium and replace with 2 mL DMEM 5% (v/v) FCS
containing 30% (v/v) DMSO (see Note 5). After 2 min, remove the supernatant
and wash gently with Ham’s-F10 plus 10% (v/v) FCS. Aspirate and replace with
a further 10 mL of Ham’s-10 plus 10% (v/v) FCS and maintain the cells in a
humidified incubator at 37°C in 5 % CO2.
3.2. Selection of Stable Transfectants

Maintain cells in standard growth medium for 48–72 h to allow expression
of the transferred genes to occur. Then aspirate the medium and replace with
standard growth medium containing G418 at a concentration of 0.3 mg/mL.
Replace the medium every 2–3 d for 3–4 wk (see Note 6).
17_Whitley_219_228_F 223 8/29/05, 11:20 AM

224 Whitley
3.3. Isolation and Expansion of Cell Colonies

With time, colonies will form. When they reach a size of approx 2000 cells
or approx 0.5–1 mm in diameter, transfer the colonies into standard medium in
the absence of G418 to a single well of either a 96- or 24-well plate. As the
wells become confluent, transfer the cells into increasingly larger dishes. Keep
a careful record of the number of passages. There are a number of ways for the
initial removal of colonies from the plate. The most common method is to
place a cloning ring around the colony, add trypsin to the inside of the ring, and
gently remove the cells/colony (see Note 7).
3.4. Characterization of Cell Lines
Each of the isolated and expanded clones must be tested and its cellular
origins identified. In many respects, this is an ongoing process. Not only is it
important to establish the characteristics of the clone, but it is also essential to
determine how stable the phenotype is. The latter can only be achieved by
repeated testing after periods in culture. Identification of different subpopula-
tions of cytotrophoblasts has been a topic of great discussion for a number of
years (12,13). As a result, a number of workshops at meetings of the Interna-
tional Federation of Placental Associations (IFPA) have discussed what mark-
ers should be expressed by each subpopulation of trophoblasts. The results of
these discussions can be summarized as follows: the minimum characteristic
requirements for a villous trophoblast cell line would be CK7+, HLA-class I–,
and CD9–, and for an extravillous cell line the requirements would be CK7+,
HLA-G+, and CD9+ (see Note 8).
3.5. Storage
Long-term storage of transfected cells is essential. As with other cells, tro-
phoblasts are readily stored in liquid nitrogen. Cells should be removed from
the plate with trypsin in the normal manner, centrifuged and re-suspended in
standard growth medium containing 10% (v/v) FCS and 10%(v/v) DMSO at a
concentration of 106 cells/mL. Normal freezing protocols should then be employed
(see Note 9).
4. Notes
1. Poly-L-ornithine is available from Sigma in a number of different molecular weights,
a molecular weight of 11,000 is desirable (cat. no. P4538); with a molecular weight
in the range of 5000–15,000 is recommended for this purpose.
2. There has been some controversy over antisera that should be used in the detec-
tion of HLA-G. At the 14th Rochester Trophoblast Conference (12) it was sug-
gested that the two antibodies of choice were 87G (21), and G233 (22). As the
17_Whitley_219_228_F 224 8/29/05, 11:20 AM

commercial availability of these antisera is unclear at the time of writing an alter-

native is MEM-G/9 (23) (Serotec, cat. no. MCA2044).
3. Chloroquine may be used to increase DNA uptake. The concentration of chloro-
quine used and the time it is left with the cells will need to be optimized for each
cell system. Addition of 100 µM 30 min prior to the addition of the DNA would
be an appropriate place to start. This should be left in contact with the cells through-
out the transfection period along with the DNA and for a maximum of 8 h.
4. Three milliliters is just sufficient to cover the dish. The cells must be rocked at
regular intervals to prevent the cells in the center of the dish from drying out.
5. DMSO shock is used to increase the efficiency of transfection and this medium
should not be left in contact with the cells for prolonged periods. Cell viability
during this phase of the transfection will depend on the exact culture conditions.
If there is a significant loss of cell viability using this protocol the concentration
of DMSO and/or the time of exposure can be reduced.
6. Cells will begin to die after 3–4 d of exposure to G418. The regular change of
medium removes cell debris.
7. There are a number of other ways to obtain a clonal population. The easiest way
is by limiting dilution. This method is often used for the cloning of hybridomas
and is therefore not suitable for all cells. After removing cells from the plate they
are re-suspended at a density of 1000 cells/mL, 100 µL of cell suspension is
plated into each well of the first column of a 96-well plate. Then doubling dilu-
tions of the cells is carried out across the plate using a multi-channel pipet. With
this strategy, there will be less than one cell per well in column 8 onward. There
are disadvantages to this approach as cells at low to very low density may not
grow as a result of a lack of growth factors. However, the use of conditioned
trophoblast medium may over come this. An alternative approach that has
proved more appropriate for the production of trophoblast cell lines has been to
removed the cells from the transfection plate the day after the transfection proce-
dure and seed them onto 24 well plates. Once they have recovered (usually fol-
lowing an overnight incubation) the selection medium can be added as above.
This approach relies on the fact that DNA incorporation into the genome is a rare
occurrence and therefore the chance of more than one resistant colony growing
per well is small. Regular inspection of each well throughout the selection proce-
dure will identify those that do have more than one colony. This approach does
have the further advantage that the cells will establish more readily due to the
autocrine/paracrine production of growth factors.
8. Clones should be screened for expression of relevant phenotypic markers. We
have found that SGHPL-4 cells express cytokeratin-7 when grown on collagen,
fibronectin and gelatin. Therefore, for full characterization it is advisable for the
cells to be tested following growth on different extracellular matrices. In our
experience, transfection with the early region of SV40 results in cells with an
extended lifespan; however, prolonged periods in culture will lead eventually to
both phenotypic and proliferative instability (i.e., SV-40 crisis). In the early stages
17_Whitley_219_228_F 225 8/29/05, 11:20 AM

226 Whitley
of characterization, the cellular phenotype should be tested every 5–10 passages.

Changes in phenotype with passage should be documented. It may also be advis-
able to karyotype the cells early and repeat this periodically. Testing the cells
after prolonged periods of storage is also advisable.
9. Cells should be stored from the earliest passage, because this will ensure that
should the clones prove useful there are sufficient early stocks available for many
studies. Each time cells are removed from storage, a proportion of the initial
passages should be frozen down again; by this means, stocks of cells can be rap-
idly established. In this way, if cells are inadvertently contaminated one can trace
back to a time before that happened. A useful cell line will be of interest to other
laboratories. To protect stocks, it is advisable not to give very early stocks but be
willing to supply later passage cells more frequently and never take cells back.
This will maintain cell stocks and reduce the chances of contamination either by
other cell types or micro-organisms.
Acknowledgments
I wish to thank Judith Cartwright and Alan Johnstone for helpful discus-
sions while preparing this chapter.
References
1. King, A., Jokhi, P. P., Smith, S. K., Sharkey, A. M., and Loke, Y. W. (1995)
Screening for cytokine mRNA in human villous and extravillous trophoblasts
using the reverse-transcriptase polymerase chain reaction (RT-PCR). Cytokine 7,
364–371.
2. Aplin, J. D., Haigh, T., Jones, C. J., Church, H. J., and Vicovac, L. (1999) Devel-
opment of cytotrophoblast columns from explanted first-trimester human pla-
cental villi: role of fibronectin and integrin alpha5beta1. Biol. Reprod. 60,
828–838.
3. Graham, C. H., Hawley, T. S., Hawley, R. G., et al. (1993) Establishment and
characterization of first trimester human trophoblast cells with extended lifespan.
Exp. Cell Res. 206, 204–211.
4. Diss, E. M., Gabbe, S. G., Moore, J. W., and Kniss, D. A. (1992) Study of throm-
boxane and prostacyclin metabolism in an in vitro model of first-trimester human
trophoblast. Am. J. Obstet. Gynecol. 167, 1046–1052.
5. Morgan, M., Kniss, D., and McDonnell, S. (1998) Expression of
metalloproteinases and their inhibitors in human trophoblast continuous cell lines.
Exp. Cell Res. 242, 18–26.
6. Kniss, D. A., Xie, Y., Li, Y., et al. (2002) ED(27) trophoblast-like cells isolated
from first-trimester chorionic villi are genetically identical to HeLa cells yet ex-
hibit a distinct phenotype. Placenta 23, 32–43.
7. Chou, J. Y. (1978) Human placental cells transformed by tsA mutants of simian
virus 40: a model system for the study of placental functions. Proc. Natl. Acad.
Sci. USA 75, 1409–1413.
17_Whitley_219_228_F 226 8/29/05, 11:20 AM

8. Chou, J. Y. (1978) Establishment of clonal human placental cells synthesizing

human choriogonadotropin. Proc. Natl. Acad. Sci. USA 75, 1854–1858.
9. Lei, K. J., Gluzman, Y., Pan, C. J., and Chou, J. Y. (1992) Immortalization of
virus-free human placental cells that express tissue-specific functions. Mol.
10. Rhim, J. S. (2000) Development of human cell lines from multiple organs. Ann
NY Acad. Sci. 919, 16–25.
11. Shih, I., Wang, T., Wu, T., Kurman, R. J., and Gearhart, J. D. (1998) Expression
of Mel-CAM in implantation site intermediate trophoblastic cell line, IST-1, lim-
its its migration on uterine smooth muscle cells. J. Cell Sci. 111, 2655–2664.
12. Frank, H. G., Morrish, D. W., Potgens, A., Genbacev, O., Kumpel, B., and
Caniggia, I. (2001) Cell culture models of human trophoblast: primary culture of
trophoblast—a workshop report. Placenta 22(Suppl A), S107–S109.
13. Shiverick, K., King, A., Frank, H., Whitley, G., Cartwright, J. E., and Schneider,
H. (2001) Cell culture models of human trophoblast II: trophoblast cell lines - a
workshop report. Placenta 22(Suppl), s104–s107.
14. Cartwright, J. E., Holden, D. P., and Whitley, G. S. (1999) Hepatocyte growth
factor regulates human trophoblast motility and invasion: a role for nitric oxide.
Br. J. Pharmacol. 128, 181–189.
15. Cartwright, J. E., Kenny, L. C., Dash, P. R., et al. (2002) Trophoblast invasion of
spiral arteries: a novel in vitro model. Placenta 23, 232–235.
16. Cartwright, J. E., Tse, W. K., and Whitley, G. S. (2002) Hepatocyte growth fac-
tor induced human trophoblast motility involves phosphatidylinositol-3-kinase,
mitogen-activated protein kinase, and inducible nitric oxide synthase. Exp. Cell
Res. 279, 219–226.
17. Tse, W. K., Whitley, G. S., and Cartwright, J. E. (2002) Transforming growth
factor-beta1 regulates hepatocyte growth factor-induced trophoblast motility and
invasion. Placenta 23, 699–705.
18. Dash, P. R., Cartwright, J. E., Baker, P. N., Johnstone, A. P., and Whitley, G. S.
(2003) Nitric oxide protects human extravillous trophoblast cells from apoptosis
by a cyclic GMP-dependent mechanism and independently of caspase 3
nitrosylation. Exp. Cell Res. 287, 314–324.
19. Dash, P. R., Cartwright, J. E., and Whitley, G. S. (2003) Nitric oxide inhibits
polyamine-induced apoptosis in the human extravillous trophoblast cell line
SGHPL-4. Hum. Reprod. 18, 959–968.
20. Southern, P. J., and Berg, P. (1982) Transformation of mammalian cells to anti-
biotic resistance with a bacterial gene under control of the SV40 early region
promoter. J Mol Appl Genet. 1, 327–341.
21. Yang, Y., Geraghty, D. E., and Hunt, J. S. (1995) Cytokine regulation of HLA-G
expression in human trophoblast cell lines. J. Reprod. Immunol. 29, 179–195.
22. Loke, Y. W., King, A., Burrows, T., et al. (1997) Evaluation of trophoblast HLA-
G antigen with a specific monoclonal antibody. Tissue Antigens 50, 135-146.
23. Fournel, S., Huc, X., Aguerre-Girr, M., et al. (2000) Comparative reactivity of
different HLA-G monoclonal antibodies to soluble HLA-G molecules. Tissue
Antigens 55, 510–508.
17_Whitley_219_228_F 227 8/29/05, 11:20 AM

228 Whitley
24. Pattillo, R. A. and Gey, G. O. (1968) The establishment of a cell line of human
hormone-synthesizing trophoblastic cells in vitro. Cancer Res. 28, 1231–1236.
25. Ho, C. K., Li, S. Y., Yu, K. J., Wang, C. C., Chiang, H., and Wang, S. Y. (1994)
Characterization of a human tumorigenic, poorly differentiated trophoblast cell
line. In Vitro Cell Dev. Biol. Anim. 30A, 415–417.
26. Zdravkovic, M., Aboagye-Mathiesen, G., Guimond, M. J., Hager, H., Ebbesen,
P., and Lala, P. K. (1999) Susceptibility of MHC class I expressing extravillous
trophoblast cell lines to killing by natural killer cells. Placenta 20, 431–440.
27. Hussa, R. O., Story, M. T., and Pattillo, R. A. (1975) Regulation of human chori-
onic gonadotropin (hCG) secretion by serum and dibutyryl cyclic AMP in malig-
nant trophoblast cells in vitro. J. Clin. Endocrinol. Metab. 40, 401–405.
28. Kohler, P. O. and Bridson, W. E. (1971) Isolation of hormone-producing clonal
lines of human choriocarcinoma. J. Clin. Endocrinol. Metab. 32, 683–687.
29. Rong-Hao, L., Luo, S., and Zhuang, L. Z. (1996) Establishment and characteriza-
tion of a cytotrophoblast cell line from normal placenta of human origin. Hum.
Reprod. 11, 1328–1333.
30. Khoo, N. K., Bechberger, J. F., Shepherd, T., et al. (1998) SV40 Tag transforma-
tion of the normal invasive trophoblast results in a premalignant phenotype. I.
Mechanisms responsible for hyperinvasiveness and resistance to anti-invasive
action of TGFbeta. Int. J. Cancer. 77, 429–439.
31. Choy, M. Y. and Manyonda, I. T. (1998) The phagocytic activity of human first
trimester extravillous trophoblast. Hum. Reprod. 13, 2941–2949.
32. Lewis, M. P., Clements, M., Takeda, S., et al. (1996) Partial characterization of
an immortalized human trophoblast cell-line, TCL-1, which possesses a CSF-1
autocrine loop. Placenta 17, 137–146.
33. Ho, C. K., Chiang, H., Li, S. Y., Yuan, C. C., and Ng, H. T. (1987) Establishment
and characterization of a tumorigenic trophoblast-like cell line from a human
placenta. Cancer Res. 47, 3220–3224.
17_Whitley_219_228_F 228 8/29/05, 11:20 AM

Choriocarcinoma Cell Culture and Transfection 229
18
Culture and Transfection of Human Choriocarcinoma
Cells
Michael W. Wolfe
Summary
In vitro models for human trophoblasts were initially established more than three decades
ago from isolated choriocarcinomas. They have proven to be extremely valuable for the study
of the cellular, molecular, and endocrine aspects of human trophoblasts. This chapter describes
basic methods for culture and maintenance of the Jeg-3, Jar, and BeWo human choriocarci-
noma cell lines as well as an effective paradigm for introducing DNA into the cells.
Key Words: Trophoblast cells; Jeg-3 cell; Jar cell; BeWo cell; medium; transfection;
lipofectamine.
1. Introduction
In vitro models for studying human trophoblast function were initially estab-
lished more than three decades ago. Dr. Roy Hertz (1) isolated choriocarcinomas
from affected women and transferred them to the cheek-pouch of hamsters.
The tumors were maintained over a number of years by serial transfer to addi-
tional animals. Three different strains of tumors were developed. Interestingly,
animals harboring the tumors exhibited endocrine changes consistent with the
presence of human chorionic gonadotropin (hCG), a known endocrine hor-
mone secreted by human trophoblasts and choriocarcinomas. One of these
strains was subsequently used to develop a number of the choriocarcinoma cell
lines that are extant today.
Pattillo and coworkers obtained choriocarcinoma tissue from Dr. Hertz at
serial transfer 304 and placed it in a co-culture with human decidual explants
(2,3). From this co-culture, they established the BeWo choriocarcinoma cell
line. This cell line maintains many of the morphological characteristics of
human trophoblasts including the ability to form syncytia. BeWo cells, like
229
18_Wolfe_229_240_F 229 8/29/05, 11:20 AM

230 Wolfe
human trophoblasts, secrete a number of hormones including hCG, human pla-

cental lactogen, progesterone, estrone and estradiol (3–6).
Serially transferred choriocarcinoma tissue (same strain as used by Pattillo)
was obtained at the 387th transfer by Kohler and coworkers and subsequently
used to establish additional choriocarcinoma lines (7). In this case, explants
were placed in culture and outgrowths were isolated and clonally selected. Six
different clonal cell lines were developed: JEG-1, -2, -3, -4, -7, and -8. The
lines possessed different morphologies, but none were observed to form syn-
cytia (8). Like BeWo cells, the JEG cell lines synthesized and secreted hCG,
human chorionic somatomammotropin (placental lactogen), progesterone, es-
trone, and estradiol (7,8). The lines appear to be distinct in that the rate of
secretion of hCG and progesterone varied across the six clonal lines.
In addition to the choriocarcinoma cell lines originating from the serially
transferred tissue obtained from Dr. Hertz , Pattillo and coworkers established
a second line directly from a trophoblastic tumor of the placenta (9). This line,
Jar choriocarcinoma cells, maintains many of the morphological and endocrine
characteristics found in human trophoblasts (6,9).
Over the last three decades the BeWo, JEG, and Jar choriocarcinoma cell
lines have been adapted to standard cell culture conditions and have proven to
be extremely valuable for studying various aspects of human trophoblast dif-
ferentiation and function. Furthermore, they tend to readily accept DNA when
introduced through many different DNA transfection paradigms. This has
allowed them to serve as the model of choice in studies involving the char-
acterization of promoter function. Instructions for basic culture and mainte-
nance of the JEG-3, Jar, and BeWo human choriocarcinoma cell lines as well
as two paradigms for introducing DNA into the cells follow.
2. Materials
2.1. JEG-3 and Jar Cells
1. The JEG-3 and Jar cell lines are available through American Type Culture Col-
lection (ATCC, Manassas, VA; Cat. Nos. JEG-3 - HTB-36; Jar - HTB-144).
2. Cell culture medium: Dulbecco’s modified Eagle’s medium (DMEM) (with
L -glutamine and 4500 mg glucose/L; Invitrogen, Carlsbad, CA, cat. no.
11965-084), fetal bovine serum (FBS) (Invitrogen, cat. no. 16000-044), and peni-
cillin/streptomycin (10,000 U/mL; Invitrogen, cat. no. 15140-122).
3. Trypsinization solution: Trypsin-ethylenediamine tetraacetic acid (EDTA)
(0.25% trypsin, 1 mM EDTA, 4 Na), 1X liquid (Invitrogen, cat. no. 25200-056).
4. Phosphate-buffered saline (PBS): 137 mM NaCl, 2.7 mM KCl, 10 mM Na2HPO4,
and 2 mM KH2PO4. Adjust pH to 7.4 and filter-sterilize or autoclave.
5. Freezing medium: cell culture medium containing 5% dimethylsulfoxide
(DMSO) (Sigma, St. Louis, MO; cat. no. D-2650), filter-sterilized.
18_Wolfe_229_240_F 230 8/29/05, 11:20 AM

6. Freezing container: Nalgene Cryo 1°C Freezing Container (Nalgene Nunc Inter-
national, Rochester, NY, cat. no. 5100-0001).
7. Tissue culture plates: 100 cm2 (Greiner Bio-One, Longwood, FL).
2.2. BeWo Cells

1. The BeWo cell line is available through ATCC (cat. no. CCL-98; see Note 1).
2. Cell culture medium: DMEM/F12, a 1:1 mixture (with L-glutamine and 15 mM
HEPES; Invitrogen, cat. no. 11330-032), MEM nonessential amino acids solu-
tion (100X, liquid; Invitrogen, cat. no. 11140-050), FBS (Invitrogen, cat. no.
16000-044), and penicillin/streptomycin (10,000 U/mL; Invitrogen, cat. no.
15140-122).
3. Trypsinization solution: Trypsin-EDTA (0.25% trypsin, 1 mM EDTA, 4 Na), 1X
liquid (Invitrogen, cat. no. 25200-056).
4. PBS: 137 mM NaCl, 2.7 mM KCl, 10 mM Na2HPO4, and 2 mM KH2PO4. Adjust
pH to 7.4 and filter-sterilize or autoclave.
5. Freezing medium: cell culture medium containing 5% DMSO (Sigma, cat. no.
D-2650), filter-sterilized.
6. Tissue culture plates: 100 cm2 (Greiner Bio-One).
2.3. Transfections
1. Trypsinization solution: Trypsin-EDTA (0.25% trypsin, 1 mM EDTA, 4 Na), 1X
liquid (Invitrogen, cat. no. 25200-056).
2. PBS: 137 mM NaCl, 2.7 mM KCl, 10 mM Na2HPO4, and 2 mM KH2PO4. Adjust
pH to 7.4 and filter-sterilize or autoclave.
3. Culture medium.
4. Hemacytometer (Fisher, cat. no. 02-671-6).
5. Sterile polypropylene snap-cap culture tubes (Greiner, from ISC Bioexpress, cat.
no. T-2840-5).
6. Opti-MEM I reduced serum media (Invitrogen, cat. no. 31985-070).
7. LipofectAmine transfection reagent (Invitrogen, cat. no. 18324-012).
8. Plus reagent (Invitrogen, cat. no. 11514-015).
9. TK-, CMV-, or SV40-driven Renilla expression plasmid (Promega, Madison, WI).
10. Tissue culture plates: 12-well (Greiner Bio-One).
11. Passive lysis buffer (Promega, cat. no. E1941).
12. Dual-Luciferase reporter assay system (Promega, cat. no. E1960).
2.4. β-Galactosidase Staining

1. 2% paraformaldehyde in PBS.
2. 0.1 M potassium ferricyanide (Sigma, cat. no. P8131).
3. 0.1 M potassium ferrocyanide (Sigma, cat. no. P9387).
4. 1 M MgCl2 (Fisher, cat. no. BP214-500) in PBS.
5. 40 mg/mL of X-Gal (Invitrogen, cat. No. 15520-034) in N,N-diemethylformamide
(DMF; Sigma, cat. no. D4254).
18_Wolfe_229_240_F 231 8/29/05, 11:20 AM

232 Wolfe
3. Methods
3.1. Culture Medium
3.1.1. JEG-3 and Jar Cells
1. In a laminar flow hood or biosafety cabinet (see Note 2), add the following to a
sterile, 500-mL bottle: 445 mL of DMEM, 50 mL of FBS, and 5 mL of penicillin/
streptomycin (see Note 3).
2. The medium can be filtered through a 0.22-µm disposable filter if desired.
3. Store at 4°C in the dark.
3.1.2. BeWo Cells

1. In a laminar flow hood or biosafety cabinet (see Note 2), add the following to a
sterile, 500-mL bottle: 440 mL of DMEM/F12, 50 mL of FBS, 5 mL of MEM
nonessential amino acids, and 5 mL of penicillin/streptomycin (see Note 3).
2. The medium can be filtered through a 0.22-µm disposable filter if desired.
3. Store at 4°C in the dark.
3.2. Cell Removal From Liquid Nitrogen: JEG-3, Jar, and BeWo Cells
1. Warm the cell culture medium to 37°C.
2. Remove a vial of frozen cells from a liquid nitrogen storage tank and thaw the
cells in a 37°C water-bath. Record vial removal in the liquid nitrogen storage log.
3. Immediately after the cells are thawed, transfer the cell suspension to a sterile,
conical 50-mL tube containing 30 mL of culture medium, and mix the cell sus-
pension gently (see Note 4).
4. Place 10 mL of the cell suspension into each of three 100-cm2 tissue culture
plates and label the plates with the cell line name, passage number, and date.
5. Grow cells in a 37°C incubator supplied with 5% CO2 and 95% air.
6. Replace the cell culture medium every 2–3 d.
3.3. Passage of Cells: JEG-3, Jar, and BeWo Cells

1. Warm all solutions (e.g. trypsin/EDTA, culture medium, PBS) to 37°C.
2. When the cells reach 75–90% confluence, the monolayer can be treated with
trypsin to detach the cells (see Note 5).
3. Aspirate media from the plate of cells to be passaged (see Note 6).
4. Add 3 mL trypsin–EDTA solution to each plate (see Note 7). Incubate within the
hood at room temperature for approx 3 min.
5. Aspirate most of the trypsin solution from each plate (leave approx 0.5 mL in the
plate) and place them into a 37°C incubator for an additional 3 min. At this point
the cells should be detaching from the plate. To assist in the detachment, rap the
side of the plate on your hand two to three times.
6. Add approx 5 mL of culture medium to each plate to stop trypsinization. Care-
fully tilt the plate at an angle and detach the remaining adherent cells by flushing
medium across the plate using a pipet. Transfer the cell suspension into a sterile
50-mL conical tube, pooling cells from similar plates.
18_Wolfe_229_240_F 232 8/29/05, 11:20 AM

7. Determine the volume of the cell suspension.

8. These cells can be passaged or split at a ratio of 1:3 to 1:5. Determine the number
of plates needed to passage the cells and the volume of cells per plate (volume of
cell suspension divided by the total number of plates).
9. Aliquot the proper volume of fresh culture medium onto each plate such that the
total volume (cell suspension plus culture medium) is 10 mL.
10. Carefully invert the tube containing the cell suspension to mix and aliquot the
proper volume of the cell suspension onto each plate. Swirl each plate to mix.
11. Label each plate with the cell line name, passage number (one greater than the
previous passage), and date.
12. Grow cells in a 37°C incubator supplied with 5% CO2 and 95% air.
13. Feed the cells the next day and then every 2–3 d subsequently.
14. Visualize cells under a microscope for viability and confluency.
3.4. Liquid Nitrogen Storage of Cells: JEG-3, Jar, and BeWo Cells
1. Harvest the cells from the plates by trypsinization as described in Subheading
3.3., steps 1–6.
2. Centrifuge to pellet the cells at 216g for 10 min at room temperature.
3. Aspirate the supernatant media and gently resuspend the cell pellet in 1 mL of
freezing medium per plate of harvested cells.
4. Label 1.0- or 2.0-mL cryo vials: cell line, passage number (one greater than the
previous passage), date, and your initials. Use vials that are appropriate for liquid
nitrogen storage.
5. Transfer 1.0 mL of cell suspension into each vial.
6. Transfer the vials to the freezing container and place in a –80°C freezer for 6 h to
overnight. Transfer the vials to a liquid nitrogen storage container for long-term
storage (see Note 8). Update the liquid nitrogen storage log with these entries.
3.5. Transient Transfection of Cells

Numerous techniques with which to introduce exogenous DNA into cells
exist: calcium phosphate co-precipitation, electroporation, virus mediated entry,
and so on. Many of these have been successfully used to transiently transfect
DNA into choriocarcinoma cell lines. Our laboratory has utilized a lipid-based
method to introduce DNA into choriocarcinoma cells (10,11). It is a simple,
highly reproducible, and efficient method for transiently transfecting these
cells. Two different approaches are presented below, both of which (with minor
adjustments) are effective. The first approach using LipofectAmine (JEG-3 cells
are used as an example) is simple, cost effective and achieves a moderate level
of transfection efficiency. The second approach using LipofectAmine and Plus
reagent (Jar cells are used as an example) is more costly and labor-intensive,
but achieves a higher level of transfection efficiency (two- to threefold). The
second protocol is the one of choice when trying to overexpress a protein of
interest in a high proportion of the cells in culture.
18_Wolfe_229_240_F 233 8/29/05, 11:20 AM

234 Wolfe
3.5.1. Transfection of JEG-3 Cells Using LipofectAmine

1. Day 1. Harvest the cells from one to three plates by trypsinization as described in
Subheading 3.3., steps 1–6.
2. Bring the total volume of the cell suspension up to 30–40 mL with culture medium.
Determine the density of the cells harvested using a hemacytometer (see Note 9).
3. Seed the appropriate number of 12-well culture plates with cells at a density of
35,000 cells per mL, adding 1 mL per well.
4. Place the plates into a 37°C incubator (supplied with 5% CO2 and 95% relative
humidity) and allow the cells to attach overnight.
5. Day 2. The following morning, warm the Opti-I MEM solution to 37°C.
6. Determine the concentration of all DNAs to be transfected using a spectropho-
tometer (see Note 10).
7. For each DNA to be transfected, add 0.75 mL of Opti-I MEM to separate snap-
cap tubes (see Note 2). This represents the appropriate volume for transfections
performed in triplicate (0.25 mL per well).
8. Add 1.5-µg of DNA to the appropriate tubes (e.g., human α-subunit promoter-
Luciferase construct or Rous sarcoma virus (RSV) β-galactosidase expression
vector) (0.5 µg per well).
9. Add 0.3 µg of a Renilla expression vector (TK-, CMV-, or SV40-driven vector)
to each tube as an internal control (0.1 µg per well).
10. In a sterile conical tube, prepare a solution containing 12 µL of LipofectAmine
and 0.75 mL of Opti-I MEM per DNA to be transfected.
11. Add 0.75 mL of the LipofectAmine/Opti-I solution to each tube prepared in steps
7–9. Swirl to mix and incubate at room temperature for 45 min (4 µL
LipofectAmine and 0.25 mL Opti-I MEM per well).
12. Place in the hood the 12-well plates containing the JEG-3 cells seeded the previous
day. At the end of the 45-min incubation, aspirate the medium from one to two
plates at a time and add 0.5 mL of the DNA/LipofectAmine/Opti-I solution to each
well. Carefully label the plates and record the DNA transfected into each well.
13. Place the plates back into the 37°C incubator.
14. Day 3. In the morning, warm the JEG-3 culture medium to 37°C.
15. Aspirate the DNA/LipofectAmine/Opti-I solution from each well and replace
with 0.5 mL of JEG-3 culture medium (see Note 11). Return the plates to the
37°C incubator.
16. Cultures can be harvested 6–72 h later and assayed for reporter activity (Fig. 1;
time is experiment dependent; method depends on assay used to detect reporter
activity) (see Note 12).
3.5.2. Transfection of Jar Cells Using LipofectAmine and Plus Reagent

1. Harvest the cells from one to three plates by trypsinization as described in Sub-
heading 3.3., steps 1–6.
2. Bring the total volume of the cell suspension up to 30–40 mL with culture medium.
Determine the density of the cells harvested using a hemacytometer (see Note 9).
18_Wolfe_229_240_F 234 8/29/05, 11:20 AM

Fig. 1. JEG-3 and Jar cell lines were transiently transfected with either a human
α-subunit promoter (–845/+48) linked to luciferase or a promoterless control (0.5 µg
per well) along with a Renilla expression plasmid as an internal control (0.1 µg per
well) as described under Subheadings 3.5.1. and 3.5.2. (see Note 12). Approximately
6 h after the transfection medium was removed, cells were washed two times with
phosphate-buffered saline (PBS) and lysed using 100 µL of Passive lysis buffer. Lysates
(30 µL) were assayed for luciferase and renilla activity using the Dual-Luciferase assay
system. Luciferase activity (relative light units [RLU]) of the human α-subunit pro-
moter was adjusted to that of Renilla and presented as the mean + SE (n = 3). Activity
of the promoterless luciferase vector was 293 ± 50 (LipofectAmine only) and 442 ± 58
(LipofectAmine and Plus) RLUs in JEG-3 cells and 1129 ± 175 (LipofectAmine only)
and 610 ± 49 (LipofectAmine and Plus) RLUs in Jar cells.
3. Seed the appropriate number of 12-well culture plates with cells at a density of
30,000 cells per mL, adding 1 mL per well.
4. Place the plates into the 37°C incubator (supplied with 5% CO2 and 95% relative
humidity) and allow the cells to attach overnight.
5. Day 2. The following morning, warm the Opti-I MEM solution to 37°C.
6. Determine the concentration of all DNAs to be transfected using a spectropho-
tometer (see Note 10).
7. For each DNA to be transfected, add 0.75 mL of Opti-I MEM to separate snap-
cap tubes (see Note 2). This represents the appropriate volume for transfections
performed in triplicate (0.25 mL per well).
8. Add 1.5 µg of DNA to the appropriate tubes (e.g., human α-subunit promoter-
luciferase construct or RSV β-galactosidase expression vector) (0.5 µg per well).
9. Add 0.3 µg of a Renilla expression vector (TK-, CMV-, or SV40-driven vector)
to each tube as an internal control (0.1 µg per well).
18_Wolfe_229_240_F 235 8/29/05, 11:20 AM

236 Wolfe
10. Add 9 µL of Plus reagent to each tube and swirl to mix. Incubate at room tem-
perature for 15 min (3 µL per well).
11. In a sterile conical tube, prepare a solution containing 6 µL of LipofectAmine
and 0.75 mL of Opti-I MEM per DNA to be transfected.
12. Add 0.75 mL of the LipofectAmine/Opti-I solution to each tube prepared in steps
7–10. Swirl to mix and incubate at room temperature for 15 min (2 µL
LipofectAmine and 0.25 mL Opti-I MEM per well).
13. Place in the hood the 12-well plates containing the Jar cells seeded the previous
day. At the end of the 15 min incubation, aspirate the medium from one to two
plates at a time and add 0.5 mL of the DNA/Plus/LipofectAmine/Opti-I solution
to each well. Carefully label the plates and record the DNA transfected into each
well.
14. Place the plates back into the 37°C incubator for 5 h.
15. At the end of the 5 h incubation, add 0.5 mL of Jar culture medium (prewarmed)
to each well (see Note 13).
16. Day 3. In the morning, warm the Jar culture medium to 37°C.
17. Aspirate the DNA/Plus/LipofectAmine/Opti-I solution from each well and re-
place with 0.5 mL of Jar culture medium (see Note 11). Return the plates to the
37°C incubator.
18. Cultures can be harvested 6–72 h later and assayed for reporter activity (Fig. 1;
time is experiment dependent; method depends on assay used to detect reporter
activity) (see Note 12).
3.6. β-Galactosidase Staining of Cultures

1. Wash cells three times with cold PBS.
2. Fix cells at 4°C for 5 min with 2% paraformaldehyde.
3. Wash cells three to four times with PBS.
4. Add 400 µL of staining solution (20 mM potassium ferricyanide, 20 mM potas-
sium ferrocyanide, 2 mM MgCl2 in PBS) to each well.
5. Add 10 µL of the X-Gal solution (1 mg/mL final concentration) to each well.
6. Incubate at 37°C overnight or until the appropriate amount of blue staining is
achieved.
4. Notes
1. A clonal variant of the BeWo cell line exists (b30) that has characteristics that
differ slightly from the parental line (forms a tight monolayer) and is cultured
under slightly different conditions (see Volume 2, Chapter 11).
2. All procedures should be performed in a laminar flow hood or biosafety cabinet
to maintain sterile conditions.
3. Alternative growth media suggested by ATCC is as follows: JEG-3 cells: 90%
MEM with 2 mM L-glutamine and Earles’ balanced salt solution to contain 1.5 g/
L sodium bicarbonate, 0.1 mM nonessential amino acids, and 1.0 mM sodium
pyruvate; 10% FBS; Jar cells: 90% RPMI-1640 medium with 2 mM L-glutamine
adjusted to contain 1.5 g/L sodium bicarbonate, 4.5 g/L glucose, 10 mM HEPES,
18_Wolfe_229_240_F 236 8/29/05, 11:20 AM

and 1.0 mM sodium pyruvate; 10% FBS; BeWo cells: 90% - Ham’s F12K medium
with 2 mM L-glutamine adjusted to contain 1.5 g/L sodium bicarbonate; 10% FBS.
4. We typically do not remove the cells from the freezing medium. The 1:30 dilu-
tion is adequate to remove any inhibitory effects that the DMSO may cause on
cell growth and viability. Nonetheless, the cells can be pelleted (216g for 10 min)
at this step to remove the DMSO-containing medium and subsequently resus-
pend the cell pellet in 30 mL of fresh culture medium.
5. The BeWo cells grow much more slowly than do the JEG-3 and Jar cells; there-
fore, it will take longer for plates to reach confluency. In addition, because the
cells are on the plate for a longer period of time, they tend to attach more tightly.
Longer incubations with trypsin/EDTA may be required in order to effectively
detach the cells from the plate.
6. At this point, the cells can be washed two times with 5 mL of PBS to speed up the
subsequent trypsinization.
7. Only trypsinize a maximum of five plates at a time to prevent overtrypsinization.
If there are more than five, tryspinize them in groups with staggered additions.
8. The cells can remain in the –80°C freezer for longer periods of time (a few days),
but extended time at –80°C reduces cell viability when thawed.
9. With the coverslip in place and using a Pasteur pipet, transfer a small amount of
the undiluted cell suspension (invert tube three to five times to ensure that cells
are resuspended) to both chambers of the hemacytometer (allow the chambers to
fill by capillary action; do not overfill). Using an inverted microscope, count all
of the viable cells in four of the 1-mm squares (a 1-mm square is composed of a
4 × 4 grid of smaller squares) in one chamber. Repeat this with the other cham-
ber. Determine the cell density using the following equation: Cells per mL =
(total cell count/eight squares) × dilution (1 in this case) × 104. The total number
of cells harvested would be: Cells per mL × volume of cell suspension.
10. We have determined that more consistent results are obtained when all of the
DNAs to be transfected in a single experiment are quantified at the same time and
near the time (within a few days) of performing the transfection. It is also impor-
tant to use super-coiled DNA that is mycoplasma-free.
11. Alternatively, treatments can be administered at this point.
12. The LipofectAmine protocol is adequate for experiments where a high level of
transfection efficiency is not required. Transfection efficiency in JEG-3 cells can
be boosted two- to threefold (Figs. 1 and 2) by following a modification of the
approach described under Subheading 3.5.2., which uses a combination of
LipofectAmine and Plus reagent (3 µL of LipofectAmine and 3 µL of Plus
reagent per well). Conversely, transfection of the Jar cells can be simplified
(although at the cost of lower transfection efficiency) by modifying the approach
under Subheading 3.5.1. (3 µL of LipofectAmine). Finally, although a specific
protocol for the transfection of BeWo cells is not provided, we have had success
using LipofectAmine in combination with the Plus reagent as described for JEG-
3 cells.
13. Alternatively, add 0.5 mL of serum-free Jar culture medium.
18_Wolfe_229_240_F 237 8/29/05, 11:20 AM

238 Wolfe
Fig. 2 (see companion CD for color version). JEG-3 and Jar cell lines were tran-
siently transfected with a construct containing the Rous sarcoma virus (RSV) pro-
moter linked to a β-galactosidase reporter gene or a promoterless control
β-galactosidase vector (0.5 µg per well) as described under Subheadingss 3.5.1. and
3.5.2. (LipofectAmine alone or LipofectAmine in combination with Plus reagent; see
Note 12). Approximately 6 h after the transfection medium was removed, cells were
stained for β-galactosidase activity (12–14).
Acknowledgments
This work was supported by National Institute of Child Health and Human
Development (NICHD) (HD39695).
References
1. Hertz, R. (1959) Choriocarcinoma of women maintained in serial passage in ham-
ster and rat. Proc. Soc. Exptl. Biol. Med. 102, 77–80.
2. Pattillo, R. A. and Gey, G. O. (1968) The establishment of a cell line of human
hormone-synthesizing trophoblastic cells in vitro. Cancer Res. 28, 1231–1236.
3. Pattillo, R. A., Gey, G. O., Delfs, E., and Mattingly, R. F. (1968) Human hormone
production in vitro. Science 159, 1467–1469.
4. Pattillo, R. A., Hussa, R. O., Delfs, E., et al. (1970) Control mechanisms for gona-
dotropic hormone production in vitro. In Vitro 6, 205–214.
5. Pattillo, R. A., Gey, G. O., Delfs, E., et al. (1971) The hormone-synthesizing tro-
phoblastic cell in vitro: A model for cancer research and placental hormone syn-
thesis. Ann. NY Acad. Sci. 172, 288–298.
18_Wolfe_229_240_F 238 8/29/05, 11:20 AM

6. Pattillo, R. A., Hussa, R. O., Huang, W. Y., Delfs, E., and Mattingly, R. F. (1972)
Estrogen production by trophoblastic tumors in tissue culture. J. Clin. Endocrinol.
Metab. 34, 59–61.
7. Kohler, P. O. and Bridson, W. E. (1971) Isolation of hormone-producing clonal
lines of human choriocarcinoma. J. Clin. Endocrinol. Metab. 32, 683–687.
8. Kohler, P. O., Bridson, W. E., Hammond, J. M., Weintraub, B., Kirschner, M. A.,
and Van Thiel, D. H. (1971) Clonal lines of human choriocarcinoma cells in cul-
ture. Acta Endocrinol. Suppl. 153, 137–153.
9. Pattillo, R. A., Ruckert, A., Hussa, R., Bernstein, R., and Delfs, E. (1971) The Jar
cell line - continuous human multihormone production and controls. In Vitro 6, 398.
10. Thway, T. M. and Wolfe, M. W. (2001) Epidermal growth factor regulation of
equine glycoprotein hormone α subunit expression in trophoblast cells. Biol.
Reprod. 65, 197–203.
11. Thway, T. M. and Wolfe, M. W. (2002) An activator protein-1 complex mediates
epidermal growth factor regulation of equine glycoprotein α subunit expression
in trophoblast cells.Biol Reprod. 67(3), 972–980.
12. Sanes, J. R., Rubenstein, J. L., and Nicolas, J. F. (1986) Use of a recombinant
retrovirus to study post-implantation cell lineage in mouse embryos. EMBO J. 5,
3133–3142.
13. Gray, G. E., Glover, J. C., Majors, J., and Sanes, J. R. (1988) Radical arrange-
ment of clonally related cells in the chicken optic tectum: lineage analysis with
a recombinant retrovirus. Proc. Natl. Acad. Sci. USA 85, 7356–7360.
14. Hamernik, D. L., Keri, R. A., Clay, C. M., et al. (1992) Gonadotrope- and
thyrotrope-specific expression of the human and bovine glycoprotein hormone
α-subunit genes is regulated by distinct cis-acting elements. Mol. Endocrinol.
6, 1745–1755.
18_Wolfe_229_240_F 239 8/29/05, 11:21 AM

18_Wolfe_229_240_F 240 8/29/05, 11:21 AM
In Vitro Methods for Studying Vascularization 241
19
In Vitro Methods for Studying Vascularization
of the Murine Allantois and Allantoic Union
with the Chorion
Karen M. Downs
Summary
Despite the importance of the definitive chorio-allantoic placenta in fetal survival, fetal
development, and long-term health of the adult, little is known about how the placenta’s indi-
vidual components, the allantois and the chorion, proliferate and develop. In this chapter, two
techniques will be described: (1) explanting murine allantoises for culture in isolation, and (2)
grafting murine allantoises into living whole mouse embryos. Together, these will enable study
of differentiation of allantoic mesoderm into the umbilical vasculature, and the mechanism(s)
by which the allantois unites with the chorion to form the chorio-allantoic placenta.
Key Words: Allantois; chorion; chorio-allantoic fusion; explants; placenta; vasculogenesis.
1. Introduction
The definitive chorio-allantoic placenta of eutherian mammals is a compos-
ite of two tissues, the umbilical cord and the chorionic disk. The umbilical
vasculature carries fetal blood to and from the chorionic disk for exchange of
nutrients, wastes, and gases with the mother. Despite their importance in fetal
survival and development, little is known concerning how the two major tis-
sues of the placenta develop, with an especial paucity of information on devel-
opment of the umbilicus.
Although the details of chorio-allantoic placentation vary between species,
humans and rodents share major features (1). First, both possess a hemochorial
placenta, in which maternal blood directly bathes trophoblastic cells of the
chorionic disk. Second, fetal/maternal exchange is carried out within the chori-
onic labyrinth. Third, the umbilicus and future chorionic disk are initially well
separated in the conceptus, uniting to become the chorio-allantoic placenta.
241
19_Downs_241_272_F 241 8/29/05, 11:21 AM

242 Downs
Given the similarities between rodent and human placentae, that genetic muta-
tions are readily available in the mouse, and that microsurgery and other experi-
mental manipulations can be carried out in living mouse embryos during
placentation, the mouse is thus a sound choice for establishing fundamental
paradigms of early placental ontogeny.
The focus of this review will be on the umbilical component of the chorio-
allantoic placenta. In the mouse, the umbilical cord is derived from a wholly
mesodermal organ, the allantois, which grows as an extension of the antero-
posterior axis, the primitive streak (reviewed in ref. 2). Until recently, little
was known about the allantois other than what it looked like and where it came
from in the conceptus (3,4) (Fig. 1).
As a result of studies carried out over the past decade, the timing of forma-
tion and whereabouts of three allantoic cell types have been elucidated.
Mesothelium ensheathes the allantois as soon as the allantoic bud appears
in the exocoelom (early neural plate stage, approx 7.25 d post conception [dpc])
(5). Angioblasts, the precursor cells of endothelium (6), are apparent in the
distal allantoic core slightly later as a small population of Flk-1-containing
cells (late neural plate stage, approx 7.5 dpc) (5,7,8). Chorio-adhesive cells are
a subset of distal mesothelium; they mediate allantoic union with the chorion
(12), and are recognized by their content of vascular cell adhesion molecule
(VCAM)-1 (headfold stage, approx 7.75 dpc), which is required for chorio-
allantoic union (9,10). Recently, thanks in large part to the classical methods of
embryology described in this chapter, a model of differentiation of allantoic
mesoderm into mesothelium, angioblasts/endothelium, and chorio-adhesive
cells has been put forth (5).
Thus, allantoic mesoderm differentiates into the umbilical vasculature,
which is then secured onto the chorionic disk. Detailed methods through which
to study allantoic vascularization and fusion with the chorion are discussed
under Subheadings 3.6. and 3.7. General protocols for addressing these major
biological questions are described under Subheadings 3.1.–3.5., and include
directives for animal husbandry (see Subheading 3.1.), obtaining rat serum
(see Subheading 3.2.), preparation of dissection and culture media (see Sub-
heading 3.3.), sharpening forceps (see Subheading 3.4.), and dissecting and
culturing whole embryos (see Subheading 3.5.).
Please note that, strictly speaking, the correct terminology for the product of
the fertilized egg, which contains both the embryonic and extraembryonic com-
ponents, is “conceptus”; however, because use of the word “embryo” is more
common throughout modern literature, “embryo” will be used in place of “con-
ceptus” in the detailed protocols.
19_Downs_241_272_F 242 8/29/05, 11:21 AM

Fig. 1 (see companion CD for color version). Morphology of the murine allantois.
A–C, Brightfield photomicrographs of pre-fusion allantoic stages in histological sec-
tions (6 µm) stained with hematoxylin and eosin. D, Histological section (6 µm) of the
chorio-allantoic fusion junction immunostained with anti-Flk-1 (white arrowhead,
brown color on the accompanying compact disk) to demonstrate penetration of Flk-1-
positive allantoic blood vessels into chorionic ectoderm. A, Allantoic bud (al), neural
plate stage/early bud stage (7.25 days post conception [dpc]) (20). The mesodermal
allantoic core is covered in a simple squamous epithelium, called mesothelium (5). B,
Enlargement of the allantois in the exocoelomic cavity (x), three-somite pair stage
(8.25 dpc). C, The allantois is just making contact with the chorion (14), in prepara-
tion for enduring fusion (14), five-somite pair stage (8.5 dpc). D, Chorio-allantoic
fusion and penetration of the allantoic vasculature into the chorion, 12-somite pairs
(9.5 dpc). Other abbreviations: ac, amniotic cavity; am, amnion; ps, primitive streak
region; ys, yolk sac. Scale bars: 50 µm (A,D); 75 µm (B,C).
19_Downs_241_272_F 243 8/29/05, 11:21 AM

244 Downs
1.1. Creation of Allantoic Explants

Differentiation of allantoic mesoderm into Flk-1-containing angioblasts
begins in the distal allantoic region, farthest away from the embryo, and
spreads proximally (7). Distal angioblasts form endothelium by flattening and
coalescence, a morphological sequence that is reiterated in the allantoic mid-
region, and finally in the base, with precise temporal regularity. By four- to
six-somite pairs (approx 8.25–8.5 dpc), the nascent endothelial plexus of the
allantois amalgamates with those of the yolk sac and fetus, all of which collec-
tively form a vascular continuum in the conceptus (5,7).
When plated directly onto tissue culture plastic, explanted allantoises behave
similarly to intact ones (8). During the first 24 h in culture, almost all explants
derived from headfold-stage embryos vascularize stereotypically, and main-
tain distal-to-proximal directionality of endothelial tubule formation; allantoic
blood vessels contain proteins typical of angioblasts/endothelial cells (Fig. 2).
With daily feeding, explants proliferate for at least 48 h more, during which the
vasculature undergoes remodeling. When explanted allantoic cells are returned
to intact allantoises, many integrate into appropriate cell lineages, demonstrat-
ing that culture does not adversely affect differentiation. In low serum, allan-
toic angioblasts neither robustly vascularize nor survive beyond 24 h but their
survival can be rescued with exogenous vascular endothelial growth factor
(VEGF165).
Use of allantoic explants has contributed to the discoveries that (1) the
murine allantois vascularizes de novo (7), (2) allantoic vasculogenesis is not
accompanied by erythropoiesis (7), (3) allantoic vascularization depends on
cues internal to the allantois rather than on contact with the chorion (7) or the
underlying primitive streak (5), and (4) allantoic angioblasts contain and respond
to many of the same signaling factors (8) used by the yolk sac and cardiovascular
system (11). Thus, because the murine allantois vascularizes de novo, is dedi-
cated to the establishment of a major vascular system, can be removed from the
conceptus free of contamination, and vascularizes in isolation, there is cur-
rently no whole-organ system as ideally suited to the study of early mamma-
lian blood vessel formation as the murine allantois.
1.2. Chorio-Allantoic Union
When we began study of chorio-allantoic fusion, or “union” (12), the cellu-
lar and genetic mechanisms of early placental ontogeny were unknown. We
hypothesized that chorio-allantoic union might occur by one of three models
(reviewed in ref. 2). In the first, contact between the allantois and the chorion
would stimulate expression of appropriate genes controlling cell adhesion on
both tissues. In the second model, the molecules required would be found con-
stitutively on both the allantois and the chorion prior to fusion. The third model
19_Downs_241_272_F 244 8/29/05, 11:21 AM

Fig. 2 (see companion CD for color version). Vascularization in allantoic explants.

Headfold-stage allantois cultured in the well of a 24-well plate for 24 h. Blood vessels
are darkly stained (arrow; brown color in accompanying compact disk) as a result of
immunostaining with anti-Flk-1; they are sitting on a layer of unstained cells (arrow-
head) derived from the mesothelium (5). Scale bar, 100 µm.
proposed that requisite adhesion molecules might be acquired gradually by

one or both components as they matured.
To distinguish between these possibilities, distal halves of individual labeled
donor allantoises were placed into the exocoelomic cavity of similarly staged
unlabeled hosts whose own allantois had been removed (12) (Fig. 3). Operated
and control conceptuses were then cultured for varying periods and examined.
Use of this grafting method revealed that chorio-allantoic union (1) is specific
to the allantois and the chorion, (2) is mediated by the allantois and chorion’s
mesothelial surfaces, (3) occurs gradually, peaking at six- to eight-somite pairs,
and (4) is dependent upon the developmental maturity of the allantois. Further,
results predicted that the molecules required for chorio-allantoic union would
appear gradually on the allantois, but localize constitutively to the chorion,
19_Downs_241_272_F 245 8/29/05, 11:21 AM

246 Downs
Fig. 3 (see companion CD for color version). Microsurgical technique to intro-

duce labeled donor allantoises (al) or donor allantoic subregions into the exocoelom
of unlabeled host embryos. A, The allantois (al) of a labeled donor embryo is aspirated
from the exocoelomic cavity (x) by anterior yolk sac puncture (long arrow) and placed
into the exocoelomic cavity (B) of an unlabeled host via the anterior yolk sac puncture
(long arrow) through which its own allantois has been removed. The small arrows
within the host’s exocoelom indicate the possible sites of donor fusion within the
exocoelom, which include the chorion (ch) and the host’s allantoic regenerating allan-
tois. C, The operated conceptus is then cultured for up to 24 h.
according to the third model. Serendipitous knockouts of the receptor/counter-

receptor pair, VCAM1 and α4-integrin (9,10,13), revealed a major role for each
gene’s protein product in chorio-allantoic union. Moreover, VCAM-1 was
localized to the allantois, whereas α4-integrin was confined to the mesoder-
mal component of the chorion. Later studies spatio-temporally fine-mapped
VCAM-1/α4-integrin to the allantois and chorion, and confirmed gradual acqui-
sition of VCAM-1 by allantoic mesothelium, whereas chorionic mesoderm con-
tained α4-integrin throughout its development (14). Together, these data validated
the study of chorio-allantoic union by classical methods of embryology.
In later experiments, grafting whole allantoises rather than distal halves
resulted in the creation of chimeric allantoises with the host allantois (Fig. 4).
Staining chimeric allantoises and allantoic explants with benzidine, a marker of
hemoglobinized red blood cells, supported the conclusion that allantoic
vasculogenesis is not accompanied by erythropoiesis (7), as it is in the yolk sac
(15,16).
2. Materials
2.1. Mice
1. F2 generation of embryos from intercrosses of the F1 inbred hybrid strain (C57Bl/
6 × CBA) (B6CBAF1/J; Jackson Laboratories, Bar Harbor, Maine) (see Note 1).
19_Downs_241_272_F 246 8/29/05, 11:21 AM

Fig. 4 (see compaion CD for color version). Chimeric allantois. Donor allantois
(dal) is darkly stained (blue on the accompanying compact disk) and has fused with
both the host’s regenerating allantois (hal-r) and chorion (ch). Other abbreviations:
am, amnion; b, embryonic brain; ys, yolk sac. Scale bar, 100 µm.
2. Embryos hemizygous for the Rosa26 lacZ transgene (17), which are produced by
mating an F1 hybrid female (see Subheading 2.1., item 1) with a male of similar
(C57Bl/6 × CBA) genetic background but which bears two copies of the Rosa26
lacZ transgene (18) (see Note 2).
2.2. Rat Bleeding

1. Male or nonpregnant female Sprague-Dawley rats ⱖ 300 g (Harlan Teklad, Madi-
son, WI), are temporarily housed in an AALAC-approved animal quarters until
exsanguination.
2. 70–100% ethanol in a squeeze bottle.
3. Bunsen burner.
4. Carcass bag.
5. Two clinical tabletop centrifuges, swinging-bucket style.
6. Container for used syringes and test tubes.
7. Cryogenic vials: Nalgene, 5-mL capacity, sterile (Fisher Scientific, Pittsburgh,
PA, cat. no. 03-337-7H).
8. Long, 6-in., straight (“dressing”) forceps: for squeezing serum clots (Fisher, Cat.
No. 13-812-40).
9. Two pairs short (4-inch) straight, fine-pointed forceps , (Fisher, cat. no. 08-880)
for grabbing the rat’s skin and removing fat from descending dorsal aorta.
19_Downs_241_272_F 247 8/29/05, 11:21 AM

248 Downs
10. Disposable gloves.

11. Five-inch, straight hemostat (“Halstead Mosquito Forceps”) (Fine Science
Tools, Foster City, CA, cat. no. 13008-12), for safe removal and disposal of
syringe needles.
12. Isofluorane (MidWest Veterinary Distribution, Winnipeg, Manitoba, Canada, cat.
no. 558.09058.3).
13. Isofluorane anesthetizing apparatus (Surgivet/Anesco, Waukesha, WI).
14. Kimwipes.
15. Lab coat.
16. Lamp (table-poise) or cool lighting.
17. Two tanks pure oxygen.
18. Nine-inch, sterile Pasteur pipets.
19. Two scavenging filter canisters (Surgivet).
20. Six-inch, straight scissors (Fisher, cat. no. 08-951-20) for making incisions into
the rat’s abdomen.
21. Sharps container for needle disposal.
22. 21-gauge × 1.5- in. syringe needles.
23. 10- and 20-mL sterile syringes.
24. 15-mL polypropylene test tubes (Fisher, cat. no. 05-538-53D).
25. Timer (multichannel).
2.3. Dissection and Culture Media
1. Double-distilled (tissue culture-grade) sterile water (Sigma Chemical Co., St. Louis,
MO, cat. no. W-3500).
2. Color pHast indicator strips, 5.0–-10 (VWR International, West Chester, PA, cat.
no. EM-9588-3); 6.5–10 (VWR, cat. no. EM-9583-3).
3. Two sterile, 1000-mL Erlenmeyer flasks.
4. 500-mL–0.22-µm cellulose acetate filter units (Fisher, cat. no. 09 740 28C).
5. Freezer (–86°C).
6. Pure carbon dioxide.
7. Gas regulator for CO2 tank.
8. 1000-mL, tissue culture-dedicated, sterile graduated cylinder.
9. 500-mL, tissue culture-dedicated, sterile graduated cylinder.
10. Nine-inch, sterile, nonabsorbent, cotton-plugged Pasteur pipets for bubbling CO2
gas to medium.
11. Volumetric, disposable, sterile pipets.
12. Tissue culture-dedicated automatic pipettor.
13. Chemical and biological reagents (see Table 1).
14. Tissue culture-dedicated stir flea.
15. 15-mL, sterile, polypropylene conical test tubes (Fisher, cat. no. 05 538 53D).
16. 50-mL, sterile, polypropylene conical test tubes (Fisher, cat. no. 14 959 49A).
17. Laminar-flow tissue culture hood.
18. Clear tubing for delivering CO2 gas to medium.
19. Vacuum apparatus.
20. Weigh boats and glass paper.
19_Downs_241_272_F 248 8/29/05, 11:21 AM

Table 1
Reagents and Suppliers for Dissection and Culture Media for Mouse Embryos
Between 7.25 and 8.5 Days Post Coitum
Dissection medium Culture medium
(DMEM + H + S) (DMEM–S)
Media component (per liter) (per liter)a
DMEM (Invitrogen-Gibco, Carlsbad, CA, 13.5 g 13.5 g
12800-017) with 4500 g/L glucose,
L-glutamine (584 mg/L) and NaPyruvate
(110 mg/L); w/o NaHCO3
HEPES, free acid (Sigma H-3375) 2.385 g –
NaHCO3 (Sigma S-8875) – 3.7 g
NaCl 0.8 g –
Amino acids (Sigma)b:
cysteine-HCl (C-1276) 4.8 mg 4.8 mg
L-alanine (A-7469) 3.6 mg 3.6 mg
L-asparagine (A-0884) 6.0 mg 6.0 mg
L-aspartic acid (A-8949) 5.3 mg 5.3 mg
L-proline (P-5607) 4.6 mg 4.6 mg
L-glutamic acid (G-1626) 5.9 mg 5.9 mg
Penicillin-streptomycinc (Invitrogen-Gibco 15070-063) 2 mL 2 mL
(100 U/mL pen, 100 mg/mL strep)
Heat-inactivated Fetal Calf Serum (FCS)d 7.5% –
(Invitrogen-Gibco 16000-044)
Heat-inactivated rat serum (prepared in house) – 50.0%e
DMEM, Dulbecco’s modified Eagle’s medium.
aBefore rat serum is added, all components are combined and the pH of the culture medium is adjusted
to 7.3 by bubbling in CO2. The solution is then filter-sterilized, aliquoted, and stored indefinitely at –86°C.
bThe amino acids are made as a 100X stock, i.e., 100 times the amounts shown above are combined
in approx 980 mL sterile double-distilled water, the pH is adjusted to 9.0 with 5 N NaOH. Bring the
volume to 1000 mL with double-distilled sterile water. Dispense 10 mL aliquots to 15 mL sterile
polypropylene tubes and store, unfiltered, indefinitely at –86°C.
cPenicillin-streptomycin stock: 5000 U/mL penicillin G sodium and 5000 µg/mL streptomycin sul-
fate in 0.85% (physiological) saline. Upon delivery from the manufacturer, thaw solution in warm water
and dispense in 6 mL aliquots to 15 mL sterile polypropylene tubes via a sterile volumetric disposable
pipette. Store the aliquots indefinitely at –86°C. When making dissection or culture media, completely
thaw a 6-mL aliquot and remove 2 mL; refreeze the remaining 4 mL.
dAs soon as the fetal calf serum (FCS) is delivered to the lab, thaw it in warm water; monitor the
bottle for breakage during the thaw. If the FCS has not been heat-inactivated from the supplier, dispense
37.5 mL aliquots of FCS to 50 mL sterile polypropylene conical tubes and heat-inactivate it at 56°C for
1 h. Store the FCS indefinitely at –86°C.
eHeat-inactivated rat serum is prepared by bleeding isofluorane-anesthetised rats from the descend-
ing dorsal aorta and immediately centrifuging the blood (see Subheading 3.2.2.). The serum is stored at
–86°C. Just before use, the serum is heat-inactivated at 56°C for 30 or 60 min, spun at 1625g for 5 min,
and diluted 1:1 with culture medium. It is then distributed to embryo culture tubes or 24-well plates,
which are gas- and temperature-equilibrated for at least 1 h before embryos or allantoises are added.
19_Downs_241_272_F 249 8/29/05, 11:21 AM

250 Downs
2.4. Sharpening Forceps

1. Arkansas stone (Fine Science Tools, cat. no. 29008-01).
2. Dissection microscope, total magnification 7.5–35.0×, equipped with trans- and
epi-illumination as well as an eyepiece reticule for measuring embryonic fea-
tures of interest before and after culture.
3. Fine, straight, #5 Dumoxel forceps (Fine Science Tools, cat. no. 11252-30).
4. Robust, straight, Inox forceps (Fine Science Tools, cat. no. 11251-20).
5. Very fine sandpaper (any local hardware store).
6. Squeeze bottle with distilled water.
2.5. Dissecting and Culturing Embryos

1. Clinical table-top centrifuge.
2. Culture medium (see Subheadings 2.3., 3.3.2., and 3.3.3.; reagents in Table 1).
3. Dissection medium (see Subheadings 2.3. and 3.3.1.; reagents in Table 1).
4. One pair “Jeweller’s” straight forceps (Fisher, cat. no. 09 953E).
5. Two pairs fine, straight, #5 Dumoxel forceps (Fine Science Tools, cat. no.
11252-30).
6. Two pairs robust, straight, Inox forceps (Fine Science Tools, cat. no. 11251-20).
7. Water-jacketed incubator set to 37°C and 6.2% CO2
8. Nine-inch, sterile Pasteur pipets.
9. 35 mm, sterile Petri dishes.
10. 60-mm, sterile Petri dishes.
11. Phosphate-buffered saline (PBS; Sigma, cat. no. P-4417), dissolved in double-
distilled water for tissue culture (see Subheading 2.3., item 1), and filtered
through 0.22-µm filter (see Subheading 2.3., item 4); stored at 4°C.
12. 5-mL disposable, sterile pipets.
13. Tissue culture-dedicated automatic pipettor (see Subheading 2.3., item 12).
14. Roller apparatus (Fisher, cat. nos. 14-277-4, 14-277-3) inserted into the incuba-
tor, set at 8° off the horizontal, and rotating continuously at 0.5 rpm.
15. One pair small (4.5-in), straight scissors (Fisher, cat. no. 08-940).
16. 12- × 75-mm sterile, individually-wrapped test tubes (Falcon 2003, Fisher, cat.
no. 14 959A) for culture of whole embryos or suspended allantoises.
17. Disposable glass test tubes (Fisher, cat. no. 14 961 32) inserted into the rollers as
“adaptors” for embryo culture tubes.
18. Laminar-flow tissue culture hood (see Subheading 2.3., item 17).
19. Water bath (56˚C).
20. Two 28-gauge hypodermic needles for trimming the ectoplacental cone.
2.6. Isolation of Allantoises

1. Mouth aspirator (requires assembly).
a. “Tube assemblies” (Sigma, cat. no. A-5177-5EA).
b. A flat mouthpiece (HPI Hospital Products, Altamonte Springs, FL).
c. A 0.45-µm filter (Fisher, cat. no. 09-754-21).
19_Downs_241_272_F 250 8/29/05, 11:21 AM

d. Latex tubing: OD, three-sixteenths inch; ID, one-eighth inch; W, one-thirty-

second inch (Sigma, cat. no. Z25, 577-7).
2. Microflame (these are not commercially available, but made in the lab)
a. Five-inch, straight hemostat (“Halstead Mosquito Forceps”; see Subheading
2.2., item 11) for breaking off beveled tip of hypodermic needle.
b. Clamp hosecock (Fisher, cat. no. 05 847) for regulating gas flow.
c. Ring stand.
d. Clamp to secure needle onto ring stand.
e. Flint striker.
f. Five-thirtyseconds by three-sixty-fourths inch rubber tubing (12-ft roll;
Fisher, Cat. No. 14178A).
g. One-fourth by one-sixteenth inch rubber tubing (12-ft roll; Fisher, Cat. No.
14178C).
h. One 19-gauge needle (1.5 in. long).
3. Electrode puller (Model P-87, Sutter Instruments, Novato, CA) with a trough
filament (0.45-mm; Sutter Instruments, cat. no. FT345B).
4. Dissection medium (see Subheadings 2.3. and 3.1.) is used for embryo and/or
allantois manipulations prior to culture.
5. Dissection microscope (see Subheading 2.4., item 2).
6. Diamond-tip glass scorer (Fisher, cat. no. 11-315).
7. Thick-walled glass tubing (outer diameter [O.D.], 1.00 mm, inner diameter [ID],
0.75 mm; 150 mm long; World Precision Instruments, Sarasota, FL, cat. no.
TW100-6,) (see Note 3).
8. Thin-walled glass tubing (O.D. 1.00 mm, ID 0.85 mm; 150 mm long; Atlantic
International Technologies, Rockaway, NJ) (see Note 4).
9. Leica instrument tube plus rubber insert to hold a glass scalpel.
10. Sterile lids of 60-mm Petri dishes. These are used for trimming or subdividing
allantoises, as they provide ample dissection space.
11. Storage boxes for glass instruments (Fisher, cat. no. 15 350 55), containing an
insert of flexible foam scored with a razor blade.
2.7. Culturing Isolated Allantoises

1. 24-well tissue culture plate (Fisher, cat. no. 08-772-1).
2. Complete culture medium (see Subheading 3.3.3.); 0.5 mL per well.
3. Dissection microscope equipped with trans- and epi-illumination as well as an
eyepiece reticule (see Subheading 2.4., item 2).
4. 12- × 75-mm test tubes (see Subheading 2.5., item 16) for culture of unoperated
control embryos; the latter are also used for culture of suspended allantoises.
5. Water-jacketed incubator set to 37°C and 6.2% CO2 (see Subheading 2.5., item 7).
6. Inverted compound microscope or tissue culture microscope with phase-contrast
optics.
7. 5-mL, disposable, sterile pipets.
8. 12-mm, round glass cover slips (Fisher, cat. no. 12-545-80)
19_Downs_241_272_F 251 8/29/05, 11:21 AM

252 Downs
9. Poly-D-lysine (Sigma, cat. no. P1024).

10. Water (see Subheading 2.3, item 1).
11. Vacuum apparatus.
2.8. Grafting Labeled Donor Allantoises into Host Exocoelomic Cavities

1. Mouth aspirator (see Subheading 3.6.2.).
2. Dissection medium (see Subheading 3.3.1.).
3. Complete culture medium (see Subheading 3.3.3.).
4. 12- × 75-mm test tubes (see Subheading 2.5., item 16) for embryo culture.
5. Other equipment as described under Subheading 2.7., with the exception of item 6.
2.9. X-Gal-Staining for β-Galactosidase (lacZ) Activity

1. 4% paraformaldehyde (see Note 5).
2. 150- to 200-mL clean glass bottles.
3. Whatman No. 1 filter paper.
4. Freezer (–86°C).
5. Clean glass funnel.
6. Hot plate (68°C).
7. Incubator (37°C).
8. Paraformaldehyde (Sigma, cat. no. P-6148).
9. Rack for holding shell vials (Fisher, cat. no. 03 389A).
10. Stock solutions (see Table 2).
11. Disposable shell vials (Fisher, cat. no. 03 339 15A).
12. Working solution of X-gal (Table 3).
13. X-gal (Tables 2 and 3).
3. Methods
3.1. Maintenance of Mouse Strains
The F1 inbred hybrid strain (B6CBA/J) provides not only standard embry-
onic material for most allantoic explant experiments, but host recipient embryos
for allantoic grafting experiments, as well.
1. F1 hybrids are purchased from the Jackson Laboratories when they are 6–8 wk of
(breeding) age, and are maintained on a 12-h light/dark cycle.
2. The lights are turned off at 1300 h so that most experiments can be carried out in
the late morning/early afternoon, rather than late into the night.
3. Estrous females are selected (19) just before the lights are turned off; single
estrous females are then paired with a single stud male.
4. Copulation plugs are checked during the dark cycle at 1630 h on the day of pair-
ing with the aid of a red light.
5. If a plug is detected at 1630 h, pregnant females are dissected 8 d later, at 1100 h,
when the majority of conceptuses are at the headfold stage (20). If no copulation
plug is detected at 1630 h, females are checked again at 0830 h the next morning,
after which embryos are dissected between 1300–1600 h seven days later.
19_Downs_241_272_F 252 8/29/05, 11:21 AM

Table 2
Stock Solutions for β-Galactosidase Activity*
Number Chemical FW Concentration Preparation
I K3Fe(CN)6 329.2 200 mM 65.85 mg/mL PBS
II K4Fe(CN)6.3H2O 422.4 200 mM 84.48 mg/mL PBS
III MgCl2 203.3 1M 10.16 g/50 mL H2O
IV X-gal 408.6 1M 40 mg/mL DMSO
(5-bromo-4-chloro-3-indolyl (dimethyl sulfoxide)
β-D-galactopyranoside)
*All reagents are purchased from Sigma. Nonsterile distilled water is used to make solutions I–III,
which are stored indefinitely at 4°C. Aliquot 250 µL of solution IV into sterile Eppendorf tubes, cover
each one with foil, and store indefinitely at –86°C. X-gal can be thawed and re-frozen five times.
Table 3
Working X-Gal Solution*
Amount of working solution
Stock solution 50 mL 10 mL Final concentration
PBS 46.15 mL 9.23 mL

I 1.25 mL 250 µL 5 mM
II 1.25 mL 250 µL 5 mM
III 0.1 mL 20 µL 2 mM
X-gal 1.25 mL 250 µL 1 mg/mL
*Make fresh just before use. Add each component to a disposable test tube in the order of presenta-
tion in this Table. The phosphate-buffered saline (PBS) is made in nonsterile distilled water; use it at
room temperature to ensure that all components go into solution.
Although most of our experiments are initiated at the headfold stage (approx
7.75–8.0 dpc), a variety of other prefusion stage embryos (neural plate/early bud,
EB, through five to sixsomite pairs (20); approx 7.25–8.5 dpc) can be obtained
by following these guidelines of animal husbandry.
To distinguish allantoic cells from those of the host in allantoic grafting
experiments, use lacZ-labeled allantoises as donor material. Adult Rosa26 lacZ/
lacZ homozygotes are maintained on the same lighting regime as the F1 hybrid
animals, described in the previous section. To obtain hemizygous lacZ/+ embryos,
F1 hybrid females, described above, are selected for estrus (described above), and
mated with individual stud Rosa26 lacZ/lacZ males, at the times described
above. Plugs are checked as described above.
19_Downs_241_272_F 253 8/29/05, 11:21 AM

254 Downs
3.2. Rat Bleeding

Rat serum is required for culture of both whole embryos and allantoic
explants. Five percent fetal calf serum (FCS), often used in cell culture and
considerably easier to obtain than rat serum, does not support optimal allantoic
growth and maintenance of the vasculature (8).
The serum for culturing whole mouse embryos and allantoic explants comes
from rats rather than mice because of higher yield: 3–5 mL of serum can be
obtained from one medium-sized rat (approx 300–350 g), enough to culture 24
embryos (approx three mouse litters’ worth). With experience, 20 rats can be
exsanguinated in a single morning. Preparing rat serum yourself, although it is
time consuming, is superior to obtaining it commercially, because commercial
rat serum is typically hemolyzed, which leads to embryo growth retardation
and severe developmental anomalies. A description of rat bleeding has previ-
ously been published (21). Unfortunately, this excellent publication is now out
of print, and difficult to obtain.
Until recently, ether was the most commonly employed anesthetic, because
it readily evaporates from the serum. However, ether is now strongly discour-
aged by most academic institutions because of its explosive properties and,
similar to halothane, it is easy to overanesthetize the animal. Thus, we have
turned to isofluorane, which is relatively safe and rarely leads to anesthesia
overdose, because it is exhaled rather than stored in the liver. We have not
noticed any aberrant affects on embryonic development as a result of
isofluorane. An anesthesia machine is used to deliver isofluorane to rats.
Exsanguination is carried out in the hood, and isofluorane is stored in a
locked metal chemicals box, its expiration date is noted, and the oxygen cylin-
ders are secured and stored in a portable dolly.
3.2.1. Set Up the Work Area

1. Illuminate, clean, and apply ethanol to the entire work area.
2. Cover the work surface with clean paper, e.g., plastic-backed absorbent paper
such as that used for radiation safety.
3. In the center of the work area, tape plastic-backed absorbent paper, plastic side
up, to a re-usable platform. Our platform is an old cardboard box, about 1 in.
deep and 8 in. square; at each use, it is covered with the fresh absorbent paper.
4. Nearby, place a box of Kimwipes, disposable gloves, the sharps container,
syringes, needles, 15-mL test tubes, a timer, and the dissection instruments.
5. Loosen the caps on as many 15-mL test tubes as there are rats. Place the centri-
fuges nearby.
6. On one side of the platform, set up the anesthesia machine.
7. The cage of live rats should be placed out of sight of the work surface.
19_Downs_241_272_F 254 8/29/05, 11:21 AM

3.2.2. Exsanguinate the Anesthetized Rat

1. Once the anesthesia machine is set up according to the manufacturer’s instructions,
glove up, and place one rat into the anesthesia chamber for approx 1–2 min, moni-
toring it continuously until its breathing is slow and steady, and its whiskers are
no longer twitching.
2. During this time, prepare the syringe by twisting a 21-gauge needle onto it, check-
ing the plunger for free movement, and expelling all the air from it.
3. Carefully re-sheathe the needle without dulling the needle’s tip, and place the
syringe to the left or right of the platform—e.g., if one is left-handed, place the
syringe to the right side of the animal, then grasp the syringe with the right hand,
and operate the plunger with the left, because this requires steadiness and some
strength.
4. While standing, remove the rat from the anesthesia chamber and close the inflow
and outflow manifold levers to the chamber.
5. Open the nosecone levers, set the vaporizer to 2, and quickly place the nosecone
over the rat’s nose, positioning the rat on its back on the platform, its head away
from you.
6. Examine the rat’s footpad reflex, applying pressure especially to the tissue between
the toes, and spray the abdomen with ethanol to matten down the fur.
7. While still standing, use one of the small forceps to grab the rat’s skin in its
abdominal midline at the level of the top of the hind legs, and elevate it.
8. With the scissors, make a horizontal incision just beneath the forceps, penetrat-
ing first the skin, then the peritoneum. From there, cut through the skin and peri-
toneum anteriorly to the level of the rib cage, and posteriorly, to just above the
animal’s rectum.
9. Returning to the site of your initial midline incision, cut through the skin and
peritoneum at the level of the top of the hind legs, and toward each one, thereby
exposing the viscera.
10. Now comfortably seated, push the viscera to the animal’s left side to reveal the
location of the descending dorsal aorta, which is covered by retroperitoneal fat.
Sometimes a hemostat is useful for retracting the colon when the latter is full.
The dorsal aorta is silvery, and pulsating, whereas the vein is dark purple.
11. Using both pairs of small forceps, carefully dissect the peritoneum and expose
about one inch of the aorta. If the aorta is not adequately exposed, its fatty mesen-
tery may clog your needle.
12. When you are ready, hold the dorsal aorta taut by grabbing tissue lateral to the
vessel with one of the small forceps and gently pulling it horizontally. Then, with
the opposite hand, grab your syringe, and let the sheath fall off.
13. At this point, hold the syringe barrel squarely from above, and orient the needle
bevel-side-down. Then, directing the needle as parallel to the supine animal as
possible, insert the tip into the lumen of the dorsal aorta. The tip of the needle
should be visible through the artery wall; blood will immediately enter the neck
of the syringe.
19_Downs_241_272_F 255 8/29/05, 11:21 AM

256 Downs
14. At this point, release the forceps and gently grab the end of the plunger with the
same hand; this allows the other hand to switch grip on the syringe barrel and
cradle it from underneath.
15. From this point on, slowly pull the plunger toward you, while maintaining an eye
at all times on the tip of the needle. By this time, about 1–2 mm of the needle tip
will have penetrated the aorta. The blood will look cherry red; it is highly oxy-
genated. However, you might mistakenly enter the vein, in which case the venous
blood is quite dark purple. Serum obtained from venous blood seems not to have
an adverse affect on embryo culture (K. Downs, unpublished). If entry into the
dorsal aorta is too superficial, blood will not enter your syringe. In that case, you
can safely remove the needle, and try again.
16. During exsanguination, and as you gently pull the plunger toward you, the needle
may clog. Stop pulling the plunger, and gently rotate the needle 90° within the
lumen of the aorta, then resume pulling on the plunger. As the rat becomes exsan-
guinated, it will eventually stop breathing. From this point on, only approx
1–2 mL more of blood might be collected, bringing the total to 10–15 mL.
The procedure, from the time of removing the rat from the anesthesia chamber to
complete exsanguination, takes approx 3–4 min.
17. As you withdraw the needle from the aorta, place a Kimwipe over the entry site
to absorb any residual blood that might spurt out.
18. Using the hemostat, disconnect the needle from the syringe, dispose of it in a
sharps container, and gently expel the fresh blood down the side of a sterile
15-mL tube.
19. Immediately euthanize the rat by cutting open its rib cage and piercing its heart.
Immediately thereafter, spin the collected blood in a tabletop centrifuge at 1625g
for at least 10 min. Place the rat in a nearby carcass bag. If only one centrifuge is
available, delay procedure on the next rat. If two centrifuges are available, you
can immediately prepare the next rat. Do not worry if the tubes spin for longer
than 10 min.
3.2.3. Serum Collection

1. Continue bleeding each rat as described above. Gloves are typically changed for
every rat, so as not to cause distress when removing each one in turn from the
communal cage.
2. At the end of bleeding, weigh the filter unit; if its weight exceeds 316 g, throw it
away and replace it with a fresh one.
3. After centrifugation, the blood will have separated into a bottom layer of red
blood cells, and an upper layer of plasma containing a fibrin clot. Do not worry if
the separation isn’t perfectly “clean” and you find red blood cells in the upper
layer.
4. Maintain the tubes of separated blood and plasma components at room tempera-
ture until the last rat has been bled.
5. Alcohol-flame the clean 6-in. dressing forceps, cool, and straddle the clot with the
tines of the forceps, squeezing the clot from the top, and working down toward the
19_Downs_241_272_F 256 8/29/05, 11:21 AM

interface. This procedure releases the serum and results in collapse of the clot
toward the interface. Take care to minimize contact with red blood cells. Some-
times a clot is not visible; nonetheless, “squeeze” the plasma as if it contained a
clot.
6. Spin the tubes containing the squeezed clots for 10 min at 1625g, and decant the
serum to fresh 15-mL tubes via a sterile 9-in. Pasteur pipet. If the clot reappears,
squeeze it again, and spin it down.
7. It is desirable to mix the serum from rat to rat, so as to partially normalize it.
Thus, serum from several rats can be collected in one tube.
8. To remove residual red blood cells from the serum, spin the tubes containing
pooled serum for 10 min at 1625g.
9. Avoiding the typically tiny pellet of red blood cells at the bottom of the conical
tube, use a sterile 5-mL pipette to aliquot serum to 6-mL labeled cryotubes in
1-, 2-, and 3-mL volumes.
10. These are then frozen at –20°˚C overnight, after which they are placed indefi-
nitely at –86˚°C (see Note 6).
3.3. Preparation of Dissection and Culture Media

Table 1 outlines the reagents for Dissection and Culture Media. These reci-
pes were previously communicated by Dr. Kirstie A. Lawson (4,22,23). They
have been used exactly as described here in all papers from my laboratory
between 1994 and the present time. For a thorough description of the various
media components and their use in whole embryo culture, refer to (24).
3.3.1 Dissection Medium (Dulbecco’s modified Eagle’s medium +
HEPES [H] + FCS [S]) (see Note 7)
1. Add approx 600 mL of sterile double-distilled water to a 1000-mL sterile gradu-
ated cylinder. Drop in a sterile stir flea, and place the cylinder on a magnetic
stirrer.
2. Turn on the magnetic stirrer, and add the following components, making sure
that each one is in solution before adding the next one:
• 13.5 g Dulbecco’s modified Eagle’s medium (DMEM) (color of solution will
be light amber).
• 2.385 g HEPES.
• 0.8 g NaCl.
• 10 mL 100X amino acids, pH 9.0.
• 2 mL pen-strep solution.
3. Bring volume to approx 910 mL with sterile double-distilled water. Adjust the
pH to 7.4 with approx 25 drops of 5 N NaOH delivered via a 9-in. Pasteur pipet.
Use color pHast indicator strips. Solution will turn red.
4. Bring volume to 925 mL with sterile double-distilled water. In the tissue culture
hood, divide the solution into two 462.5 mL aliquots and filter-sterilize each
through a separate 500-mL filter unit (0.22 µm cellulose acetate filter) under
19_Downs_241_272_F 257 8/29/05, 11:21 AM

258 Downs
vacuum. Add 37.5 mL of heat-inactivated FCS to each aliquot, bringing the total
volume of each half to 500 mL. Dispense approx 40 mL of the dissection medium
into each of about 25 sterile 50-mL polypropylene conical tubes labeled “DMEM
+ H + S” and the date. Store indefinitely at –86°C.
5. After thawing, store dissection medium at 4°C for no longer than 1 mo.
3.3.2. Culture Medium (DMEM–Rat Serum [S]) (see Note 7)

1. Measure out approx 800 mL of sterile double-distilled water into a sterile
1000 mL graduated cylinder reserved for tissue culture and containing a tis-
sue culture dedicated sterile stir flea.
2. Turn the magnetic stirrer on and add the following components:
• 13.5 g DMEM (color of solution will be light amber).
• 3.7 g NaHCO3
• 10 mL 100X amino acids, pH 9.0.
• 2 mL pen-strep solution.
3. Bring volume to 1000 mL with sterile double-distilled water and mix thoroughly.
4. In the tissue culture hood, distribute 500 mL of the solution to each of two sterile
1000 mL Erlenmeyer flasks. Bubble CO2 into each one until the pH is 7.3 (it can
go as low as 6.5). The color of the solution will turn to yellow-orange.
5. Filter the medium through two 500-mL filter units (0.22 µm cellulose acetate)
under vacuum and aliquot approx 40 mL into each of about 25 sterile 50-mL
polypropylene tubes. Freeze the medium indefinitely at –86°C.
6. When frozen, the color of the medium will be yellow; when thawed, the color
will turn red and many of the heavier components will sink to the bottom of the
tube - mix well by gentle inversion. DMEM–S, once thawed, is ready to be mixed
with rat serum (see Subheading 3.3.3.) and used immediately for embryo or allan-
toic explant culture. Once thawed, DMEM–S can be stored without serum at 4°C
for 1 mo only.
3.3.3. Preparation of Complete Culture Medium (see Note 8)

Just before each experiment, “complete” culture medium is made by mixing
DMEM–S with heat-inactivated and spun rat serum in a ratio of 1:1.
1. Turn on the 56°C water-bath. If water is at room temperature, then about 2 h are
required to reach temperature.
2. Estimate the number of embryos that you will culture. The (C57Bl/6 × CBA)
F1 hybrid yields, on average, nine implantation sites (K. Downs, unpublished).
Embryos are typically cultured singly or in pairs in 1 mL of complete culture
medium.
3. Thaw the rat serum in a clean beaker of warm water (this will take about 10 min),
then heat-inactivate the serum by placing it at 56°C for 30 min. Be sure that the
water bath entirely covers the serum. Occasionally, after thawing, some particu-
late will be found in the serum, which can adhere to embryos and prevent them
from developing properly. Thus, we routinely centrifuge heat-inactivated serum
19_Downs_241_272_F 258 8/29/05, 11:21 AM

for 5 min at 1625g at room temperature. With a sterile 9-in. Pasteur pipet, trans-
fer all but the bottom-most serum into a sterile 50-mL polypropylene tube. It is
important that one add the serum to the test tube before adding the culture medium
(DMEM–S) in case some is spilled. Then, add an equal volume of culture medium
(DMEM–S) via a sterile 5-mL disposable volumetric pipet attached to a dedi-
cated automated pipettor. Gently mix.
4. Label embryo culture tubes and distribute 1 mL of complete culture medium to
each one. With caps in the loose position, balance the culture tubes against each
other in the incubator’s roller apparatus, placed at 8° off the horizontal and roll-
ing at 0.5 rpm (25). We never turn off the roller apparatus. Gas- and temperature-
equilibrate the complete culture medium for at least 1 h before placing embryos
into the tubes. When equilibrated, the color of the medium will be fleshy-pink.
Any unused medium can be stored for up to 24 h at 4°C and used the next day; if
you do this, mix it with fresh medium, rather than using it separately, because in this
way, all embryos in that experiment will be exposed to similar culture conditions.
3.4. Protocol for Sharpening Forceps (Modified From ref. 21)
Assess the state of your forceps. Forceps should have the following charac-
teristics:
• Tines are of similar thickness.
• Tips of tines meet in a point when examined in profile. There is no gap between
them.
• When examined one above the other, tines do not substantially overlap.
• When examined from the inside, the tips of the tines are pointed.
With the aid of the dissection microscope, carry out the following steps, but
note that fine forceps will require a much more gentle touch during the sharp-
ening procedure than the robust forceps:
1. Place the Arkansas stone on a horizontal surface and squirt it with distilled water.
Hold the forceps vertically over the wet part of the stone with the tips down and
lightly squeezed together. Move the forceps across the stone, applying minimum
pressure. After grinding, dry the tines with a Kimwipe and check them in a dis-
section microscope with ×25 magnification. Continue grinding until the tips are
the same length.
2. Rotate the forceps 90°, and determine whether and where the tines overlap. With
the tips of the tines apart, position the forceps horizontally so that the edges of
both tines are in contact with the stone along their length, and move the forceps
from side to side, applying more pressure to one or the other prong if it overhangs
when the tips are brought together. The tines may also be slightly angled toward
their tip during this procedure to shave off more metal from the tip.
3. Rotate the forceps 180°, examine them for overlap, and repeat step 2 with the
other side of the tines.
4. Rotate the forceps 90°, and examine the thickness of the tines. A gradual tapering
toward the tip, and tines of similar thickness along this length, are desired. For
19_Downs_241_272_F 259 8/29/05, 11:21 AM

260 Downs
this, grind the outer surface of each tine from side to side on the stone until the
thickness is reduced and tapered toward the tip.
5. Turn the forceps 180° and repeat step 4 on the other tine.
6. Check that the tips are still of the same length when just touched together, repeat-
ing step 1 if necessary.
7. With very light strokes, round off the inner edges and tip so that, when apposed,
the tips together form a tapered probe. Very fine sandpaper may be used for this
step and for future periodic touch-ups.
8. Check the inner edges of the tines to make sure that they make a point. If not,
gently stroke each tine near the tip along the stone until a point is made; if the tips
no longer meet, repeat step 6 and/or 7.
Before each dissection, make sure that the forceps are in good condition.
Once sharpened as desired, forceps should need only minor touch-ups, which
can be carried out with very fine sandpaper. When not in use, protect the for-
ceps by sheathing the tips with a cut-off yellow pipet tip. After use, and also
before the next dissection, wash the forceps in soapy water, rinse in tap water
and then distilled water, apply a final rinse in absolute alcohol, and wipe dry
with a Kimwipe. Never flame dissection forceps.
3.5. Dissection of Embryos (7.25–8.5 dpc)

3.5.1. Dissect Embryos (see Note 9)
Prior to dissecting embryos, prepare an appropriate amount of complete cul-
ture medium (see Subheading 3.3.3.), and gas- and heat-equilibrate it for 1–3 h
prior to introducing dissected whole embryos, allantoic explants, and/or oper-
ated whole embryos.
Pregnant females are sacrificed by cervical dislocation (26), although spe-
cial justification may be necessary in order to do this without anesthesia. A
partial schematic guide to embryo dissections, with modifications that might
be preferred, can be found in (21,26).
3.5.1.1.REMOVE IMPLANTATION SITES FROM THE PREGNANT DAM
1. Using the Jeweller’s forceps and the small scissors, locate the ovaries on one side
of the female, and snip the tissue between the ovary and the uterine horn.
2. Grasping the free uterine horn by an edge, sever the mesometrial connection down
one horn, snip the cervix, and cut through the mesometrium on the other horn,
releasing the entire uterus by cutting it away from the other ovary.
3. Place the uterus into the lid of a sterile 60-mm Petri dish containing sterile PBS.
4. Trim away the fat with clean small scissors, and transfer the uterus to the deep
portion of the 60-mm Petri dish, which should contain enough sterile PBS to
cover the tissue.
19_Downs_241_272_F 260 8/29/05, 11:21 AM

3.5.1.2. REMOVE UTERINE MUSCLE AND LIBERATE DECIDUAL SWELLINGS

1. Bring the dish to the dissection microscope.
2. Turn on both sources of illumination and, at the lowest magnification possible
(typically ×7.5), locate the puffiest side of the first implantation site, closest to
where the uterus was snipped from one of the ovaries.
3. With both forceps and arms resting on the glove boxes or other similar supports,
tear through the two uterine muscle layers on one face and pick away at these
until the decidual half is exposed.
4. With one forceps, grip the uterus between the exposed implantation site and the
next unexposed one.
5. With the other forceps, squeeze the uterine tissue near the same site but closest to
the exposed deciduum, tines together, and shell out the exposed deciduum.
6. Repeat this with each implantation site.
7. Count the implantation sites.
3.5.1.3. REMOVE EMBRYOS FROM DECIDUAL SWELLINGS

1. With forceps, transfer all decidua to the deep part of a 35-mm dish containing
dissection medium.
2. Count them to make sure that all have been transferred.
3. Increase the magnification slightly and gently stabilize the wider part of each
deciduum between the tines of one forceps.
4. With the other forceps, tines together, impale the decidual swelling by piercing
the decidual dimple at a 60° angle to the bottom of the dish.
5. Slightly separate the decidual halves with the same forceps.
6. With the other forceps, clip the wider part of the deciduum that does not contain
the embryo.
7. Rotate the deciduum 90°, so that the wider part is at the top.
8. Firmly but gently grip each separated decidual portion close to the base of the split.
9. Applying even pressure to each decidual wall, split the deciduum into two halves
with a vertical downward motion. The embryo with its associated trophoblast
will stay with one of the halves.
10. Now, with the tines of one forceps separated, impale the exposed decidual half
and its embryo to the bottom of the dish.
11. With the tines of the other forceps together, scrape the embryo out.
12. When all of the embryos have been scraped out, transfer them via a 9-in. sterile
Pasteur pipet into the lid of the 35-mm dish containing dissection medium and
make sure that the number transferred is the same as the number of decidua dis-
sected.
3.5.1.4. REFLECT REICHERT’S MEMBRANE

1. Turn off the epi-illumination. This will allow Reichert’s membrane to be more
readily visualized by trans-illumination only.
19_Downs_241_272_F 261 8/29/05, 11:21 AM

262 Downs
2. Using both pairs of fine forceps, pinch the trophoblast, associated parietal endo-
derm, and its Reichert’s membrane (all three tissues are collectively, though col-
loquially, called “Reichert’s membrane) at the embryonic/extraembryonic
junction.
3. With one of the forceps, reflect Reichert’s membrane toward the embryo’s distal
end, rounding it and allowing Reichert’s membrane to retract toward the extraem-
bryonic region.
4. Trim away enough of this reflected membrane so that the embryos can be
staged (20).
3.5.1.5. STAGE EMBRYOS

1. Place the reflected embryos into a 35-mm dish containing fresh dissection
medium—they will be stable in this dish at room temperature (18–20°C) for
about 90 min.
2. Stage the embryos (4,20,22,23,27) and form pairs in microdrops of dissection
medium in 35- or 60-mm dishes.
3. If it is necessary to distinguish one member of the pair from the other, trim its
ectoplacental cone by scissor action with a pair of 28-gauge insulin hypodermic
needles.
3.5.2. Culturing Dissected Embryos

1. Remove each culture tube from the incubator, one at a time, pipet a single embryo
or pair of embryos into the tube with the aid of a sterile 9-inch Pasteur pipet, and
return the tube to the roller apparatus in the incubator as quickly as possible.
2. Do the same with the next tube, and so on.
3. Record the time at which each tube of embryos went into culture.
4. Once all of the embryos are in the culture tubes, it is important that the incubator
not be opened at any time during the remainder of the culture period, because
temperature and gas fluctuations compromise embryo growth and development.
3.5.3. Scoring Cultured Embryos (see Note 10)

It is important that cultured embryos not bear any visible morphological
defects as a result of culture, particularly those embryos that have been oper-
ated on. Hence, we employ a classic morphological staging system (28) to
ensure that operated cultured embryos grow and develop to the same extent
as their unoperated cultured counterparts in the same experiment.
3.6. Isolation and Culture of Allantoises From Dissected Mouse Embryos

3.6.1. Outline of Method for Explanting Allantoises
An outline for creating allantoic explants is presented below. Full details for
explanting allantoises follow under Subheadings 3.6.2.–3.6.4. Although all
prefusion stage allantoises (7.25–8.5 dpc) may be explanted, most of our stud-
19_Downs_241_272_F 262 8/29/05, 11:21 AM

ies have focused on headfold-stage allantoises (7.75–8.0 dpc), because these

do not contain morphologically obvious blood vessels, and hence, are a rela-
tively clean slate for the study of vasculogenesis. Controls for explant cultures
should include cultured whole embryos (see Subheading 3.5.), scored for
morphology and chorio-allantoic union (27) to verify the quality of the culture
medium.
1. Prepare complete culture medium and distribute 0.5 mL to each well of a 24-well
tissue culture dish; gas- and temperature-equilibrate the medium for at least 1 h.
2. Dissect embryos from a pregnant dam and reflect each one’s Reichert’s mem-
brane—this entire procedure should take no longer than 30 min. Stage each
embryo (20).
3. With a microcapillary secured onto a mouth aspirator, aspirate out the allantois.
4. Gently release one allantois per well of the 24-well plate.
5. Culture the allantoic explants.
6. After culture, vascularization of the allantois can be viewed in phase-contrast
optics, photographed, fixed and stained (8).
3.6.2. Assembling Equipment and Glass Instruments for Isolating

Allantoises
3.6.2.1. ASSEMBLE THE MOUTH ASPIRATOR
1. Measure out approx 2 ft of tubing.
2. To one end, attach a microcapillary holder, and to the other, a 0.45-µm filter.
3. Insert the flat mouthpiece into the filter.
4. It may be necessary to make a mouthpiece adapter by wrapping its stem in
parafilm. Round mouthpieces are generally sold with the Sigma aspirator assem-
blies but, in terms of suction control, they are inferior to the flat ones.
3.6.2.2. PREPARE THE MICROFLAME

1. Cut off the beveled end of a 19-gauge hypodermic needle with a hemostat.
2. Insert the base of the hypodermic needle into an adequate length of tubing.
Because the diameter of this tubing may be too narrow for the gas outlet, you
may insert an adapter made of a cut-off Pasteur pipet onto which a length of
wider tubing is fixed that will snugly fit over the gas outlet.
3. The needle is vertically supported in a clamp secured onto a ring stand.
4. Place the hosecock clamp near the middle of the tubing, and tighten slightly.
3.6.2.3. PULL MICROCAPILLARIES TO ISOLATE ALLANTOISES

1. Make a microflame, adjusting the hosecock clamp until the flame is about 1 cm high.
2. Place the thin-walled glass tubing into the flame, hold each end between your
index finger and thumb, and rotate the glass tubing through the tip of the flame.
3. When the glass is red-hot, remove it, cool for about 1–2 s, and then pull horizon-
tally so that the molten portion stretches and decreases slightly in diameter.
19_Downs_241_272_F 263 8/29/05, 11:21 AM

264 Downs
4. If the stretched portion of the tubing is less than 120 µm thick, it should snap
smartly, and break in half. If not, throw the tubing away and try another piece.
5. If the glass breaks, the broken ends should be flush, rather than beveled. Discard
any halves that have beveled ends, because these will leave behind bits of the
base of the allantois during aspiration.
6. Score the base (nonpulled end) of the glass microcapillary with the diamond-tip
glass scorer and break cleanly so that the shaft of the final product is approx 1–2
inches long. This length will allow comfortable manipulation of the allantois in
the dissection microscope.
7. Measure the inner diameter of the microcapillary in an eyepiece reticule. It should
be between 60 and 120 µm.
8. Store the microcapillaries in boxes into which has been inserted a strip of thick
foam scored at regular intervals with a razor blade. Use one box each for 60-µm,
90-µm, and 120-µm microcapillaries and label appropriately.
9. Repeat the above procedure on the shorter length of glass, thereby conserving
this expensive commodity.
3.6.2.4. FORGE GLASS SCALPELS

Glass scalpels are solid dissection needles produced by fusing the central
portion of thick-walled glass capillary tubing on a microflame, and then pro-
ducing a tapered cutting edge on an electrode puller (29). Glass scalpels are
used to trim away visceral endoderm from the allantois after aspiration. They
can also be used to subdivide allantoises into distinct regions.
1. Hold the thick-walled glass tubing horizontally within the flame, and rotate the
central portion until molten.
2. When the tubing is red-hot, remove it from the flame, wait a few seconds, then
gently pull the tubing straight across just a few centimeters, limiting reduction in
its diameter.
3. Insert the microcapillary into the electrode puller, so that the fused mid-portion
lies over the trough filament. On our electrode puller, the settings for long,
tapered glass scalpels are: Heat, 500 mA; Pull, 0 mA; Velocity, 30 mV; Time,
0 ms; Air Flow, 470.
4. Remove tapered microcapillaries from the electrode puller, and bend them in the
microflame so that they assume a sigmoidal profile (26,29).
5. For trimming allantoises, one of these scalpels is inserted into the Leica instru-
ment tube.
3.6.3. Protocol for Isolating Allantoises

1. Pipet a prefusion stage embryo into the lid of a sterile 60-mm tissue culture dish
that contains enough dissection medium to cover the bottom. Estimate the diam-
eter of the base of the allantois using an eyepiece reticule.
2. Load the mouth aspirator with a microcapillary of diameter similar to that of the
base of the allantois. Place the tip of the microcapillary into the medium and
19_Downs_241_272_F 264 8/29/05, 11:21 AM

allow it to equilibrate, so that no more medium enters the capillary. Aspirate

more medium into the microcapillary to guard against blowing air bubbles into
the exocoelomic cavity in step 3.
3. Using one pair of robust forceps to buttress the embryo’s posterior side, push the
tip of the microcapillary against the anterior yolk sac, above the headfolds, while
at the same time, gently blowing medium at the yolk sac. This will result in a
small anterior yolk sac puncture. The exocoelomic cavity will inflate when the
yolk sac has been pierced. If culture of the embryo after allantois removal is not
desired, an alternative method for gaining access to the allantois is to remove the
ectoplacental cone and chorion. For this, drag the conceptus to the meniscus with
opposing forceps; at the meniscus, the ectoplacental cone/chorion should cleanly
break off, leaving an opening to the exocoelomic cavity for easy aspiration of the
allantois.
4. Immediately after inflating the exocoelom, which allows better visualization of
the allantois through the yolk sac, aim the tip of the microcapillary toward the
distal tip of the allantois and, while gently mouth aspirating, “sheathe” the entire
length of the allantois with the tip of the microcapillary.
5. While maintaining minimal aspiration, lift the allantois, still in the exocoelom,
toward the meniscus. At the meniscus, gently suction the allantois into the
microcapillary, thus leaving behind the embryo, which will drop to the bottom of
the dish. Examine the embryo for any remaining bit of allantois. Any incomplete
allantoic isolations should be noted. Also, the allantois may contain some vis-
ceral endoderm. This can be removed by trimming with a glass scalpel. The
allantois can now be manipulated. Note that whole allantoises (7,8,18) as well
as allantoic subregions (5) can be cultured. To produce allantoic subregions, cut
whole allantoises with a glass scalpel held in a Leica instrument tube.
6. Collect all similar-staged allantoises or allantoic subregions in small drops of
dissection medium. They will remain healthy for at least 30 min at room tem-
perature (18–20°C).
3.6.4. Protocol for Culturing Isolated Allantoises

3.6.4.1. STANDARD PROCEDURE
1. Prepare complete culture medium as described for whole embryos (see Subhead-
ing 3.3.3.), with the modification that rat serum is heat-inactivated for 1 h instead
of 30 min.
2. Using a sterile 5-mL disposable pipet, distribute 0.5 mL culture medium into
each well of the 24-well tissue culture plate. Gas- and temperature-equilibrate
the complete culture medium for 1–3 h before culturing isolated allantoises.
3. After all allantoises and/or subregions have been collected, remove the 24-well
plate from the incubator, and place it near the dissection microscope. Using the
mouth aspirator and appropriately sized microcapillary, aspirate one to several
allantoises into the microcapillary, taking care that the allantoises remain near
the tip. Otherwise, they will get hopelessly stuck in the capillary shaft.
19_Downs_241_272_F 265 8/29/05, 11:21 AM

266 Downs
4. Replace the dish containing the explanted allantoises with the 24-well plate; in
the dissection microscope, release a single allantois into a single well of the plate.
5. Repeat steps 3 and 4.
6. Culture the allantoic explants for up to 24 h. If you wish to culture the explants
for longer periods, change the complete culture medium every 24 h with fresh
gas- and temperature-equilibrated fresh complete culture medium. Do not rinse
allantoises between changes of medium.
7. Examine the explants in the inverted compound or tissue culture microscope.
3.6.4.2. CULTURING ALLANTOISES ON GLASS COVER SLIPS
You may want to mount immunostained allantoic explants or subregions on
a glass microscope slide for closer analysis. Although chamber slides are avail-
able, we have found that these do not support allantoic growth and develop-
ment as well as culture in 24-well dishes (K. Downs, unpublished data).
1. Thus, at least 1 d before allantoic explantation, insert autoclaved glass cover slips
into each well of a sterile 24-well plate, and coat each cover slip with poly-D-
lysine, as described later. Vascularization on poly-D-lysine-coated glass cover-
slips appears to take place as well as on tissue culture plastic, although with a
delay of 4–5 h (8).
2. Prepare poly-D-lysine by dissolving 50 mg poly-D-lysine in 50 mL sterile double-
distilled water for a final solution of 1 mg/mL. Filter poly-D-lysine solution
through a 0.22-µm filter and store at 4°C.
3. Place autoclaved cover slips into individual wells of a 24-well plate using heat-
flamed nonembryo dissection forceps.
4. Pipet 200 µL poly-D-lysine onto each cover slip.
5. Incubate cover slips for 30 min at room temperature.
6. Aspirate poly-D-lysine via a vacuum apparatus, and rinse wells three times each
with at least 2 mL of sterile double-distilled water per well each time. Do not flip
over the cover slips during this procedure.
7. Cover the 24-well plate, and air dry the poly-D-lysine coated glass inserts over-
night up to 2 wk.
8. Proceed as described above for culturing allantoic explants, and carry out stain-
ing as normal. At the last step of the staining procedure, remove the cover slip
with a forceps, place it explant side up on a glass slide, and overlay the explant
with water-based mounting medium and a standard glass cover slip. View the
explant in a compound microscope.
3.6.4.3. CULTURING ALLANTOISES IN SUSPENSION
One may wish to culture whole allantoises or allantoic subregions in sus-
pension, because cell relations after double-immunostaining are more readily
apparent in sectioned specimens than in plated allantoic explants.
1. Isolated allantoises can be cultured in suspension in embryo culture tubes (see
Subheading 2.5., item 16) containing 0.5 mL medium and inserted into the roller
apparatus.
19_Downs_241_272_F 266 8/29/05, 11:21 AM

2. Roll the tubes at approx 1 rpm, instead of 0.5 rpm, to prevent the explants from
sticking to the walls of the test tube.
3. Suspended allantoises can be whole mount stained, squashed beneath a glass
cover slip on a glass microscope slide, or especially, prepared for immunohis-
tochemistry in histological sections (7,18).
3.7. Grafting Labeled Donor Allantoises into the Exocoelomic Cavity

of Host Embryos
3.7.1. Outline of Method for Grafting Labeled Donor Allantoises into
the Exocoelom of Host Embryos
1. Dissect donor and host embryos. Stage-match a donor with a host embryo.
2. Remove host allantois, and replace it with a stage-matched donor whole allantois
or allantoic subregion.
3. Culture operated embryos and controls for up to 24 h.
4. Score embryos, fix them, and X-gal stain.
3.7.2. Protocol for Introducing Donor Allantoises into Host

Exocoelomic Cavities (see Note 11)
Unoperated cultured embryos are used as controls to ensure that culture con-
ditions support chorio-allantoic union.
1. Make complete culture medium (see Subheading 3.3.3.), pipet 1 mL into each
embryo culture tube, place the tubes into the roller apparatus, and gas- and tem-
perature-equilibrate the complete culture medium for 1–3 h.
2. Dissect donor embryos (see Subheading 3.5.1.) during prechorionic fusion stages
(neural plate–six-somite pairs, 7.25–8.5 dpc); reflect Reichert’s membrane (see
Subheading 3.5.1.4.), and hold at room temperature (18–20°C) until host em-
bryos have been dissected (see step 3).
3. Dissect host embryos (see Subheading 3.5.1.) during prechorionic fusion stages
(neural plate–six-somite pairs, 7.25–8.5 dpc); reflect Reichert’s membrane (see
Subheading 3.5.1.4.).
4. Stage donor embryos (see Subheading 3.5.1.5.). Pool similar stages of donor
embryos into small microdrops. Label the dish appropriately.
5. Stage host embryos (see Subheading 3.5.1.5.). Pool similar stages of host em-
bryos into small microdrops and label the dish appropriately.
6. Stage-match pairs of donor and host embryos while maintaining them in separate
labeled dishes to distinguish them from each other. Set aside an appropriate num-
ber of donor and host embryos to serve as unoperated controls. Fill the lid of a
60-mm dish with dissection medium and transfer a single host embryo into it.
Remove the host’s allantois via anterior yolk sac puncture (see Subheading
3.6.3., steps 3 and 4), and discard, culture, or use it for genotyping. Transfer a
stage-matched donor embryo into the lid, remove its allantois by one of the two
methods described in Subheading 3.6.3., step 3. This allantois can be transferred
to the host exocoelom in its entirety, or divided into subregions.
19_Downs_241_272_F 267 8/29/05, 11:21 AM

268 Downs
7. Aspirate the donor material into the tip of the microcapillary. Directing the
microcapillary into the host’s anterior yolk sac puncture, gently release the donor
material into the host’s exocoelom.
8. Place the operated host embryo and the operated donor embryo from which
you’ve removed the donor allantois (by yolk sac puncture) into an embryo cul-
ture tube containing gas- and temperature-equilibrated culture medium, and note
the time that culture was begun. The advantages of this culture arrangement are
that first, at the end of culture, the numbers of somite pairs are compared—they
should be similar, thus confirming synchronicity of the donor allantois and host
embryo; and second, the donor embryo serves as a positive control for X-gal
staining.
9. Continue in this manner until all desired donor and host embryos have been operated.
10. Place unoperated donor and host embryos into separate culture tubes; these serve
as controls for chorio-allantoic union. Embryos of each pair can be distinguished
by trimming the ectoplacental cone of one member of the pair (see Subheading
3.5.1.5., step 3).
11. Culture embryos for up to 24 h.
12. At the end of the culture period, score the operated embryos for the location of
the donor allantois (sometimes this is not possible until after applying appropri-
ate methods of visualization; see Subheading 3.7.2.), and score both operated
and control embryos by morphological features (27) (see Subheading 3.5.3.).
13. Proceed to Subheading 3.7.3.
3.7.3. Protocol for Staining Embryos for β-Galactosidase (lacZ) Activity

1. Fix embryos in 4% paraformaldehyde for 2 h at 4°C in shell vials.
2. Rinse embryos three times for 20 min each in PBS at room temperature. Embryos
may be stored at 4°C overnight.
3. Prepare working solution of X-gal (Table 3).
4. X-gal stain the embryos at 37°C. Before immunostaining X-gal-treated material,
incubate it in the X-gal solution for no longer than 2–6 h (5). Otherwise, the
specimens can be X-gal-stained for up to 18 h (7,14,18).
5. Rinse the X-gal-stained embryos three times successively in PBS at room tem-
perature. The specimens can be held in PBS at 4°C for up to 3 d at this point
before dehydrating, clearing, and embedding. Before dehydration is begun,the
yolk sac of each embryo MUST be punctured; if not, the embryos will collapse in
the clearing agent.
4. Notes
1. The F2 inbred hybrid strain (B6CBA/J) was used to compile an extensive histo-
logical atlas (30) as well as fate maps of the mouse gastrula (4). Thus, an exten-
sive foundational literature exists for this strain.
2. All cells of hemizygous Rosa26 embryos stain positively for β-galactosidase
activity between 7.5–9.5 dpc, turning blue, with the exception that cells of the
ectoplacental cone are not as intensely blue as the other cells in the embryo, even
19_Downs_241_272_F 268 8/29/05, 11:21 AM

after 18 h of X-gal staining. Moreover, hemizygous Rosa26 lacZ/+ embryos grow

and develop with similar kinetics (18) as the nontransgenic F2 embryos, described
previously, thereby ensuring adequate synchrony between donor and host tissue.
3. Hand-pulled glass tubing is further pulled on an electrode puller to achieve long ta-
pered ends for cutting away residual visceral endoderm from the base of the allantois
or for dividing the allantois into subregions (see Subheading 3.6.3., step 5).
4. Glass tubing will be hand-pulled to achieve an ID of 60–120 µm (see Subhead-
ings 2.6. and 3.6.2.3.). Hand-pulled glass capillaries are used to aspirate
allantoises from exocoelomic cavities, and introduce them into wells for explant
culture (see Subheading 3.6.4.) or into host exocoeloms (see Subheading 3.7.2.).
5. Transfer a small portion of stock paraformaldehyde powder to a small bottle, and
store both the stock and working powders at 4°C. When needed, bring the work-
ing powder to room temperature, and weigh out 4 g. Warm 100 mL of PBS for
about 30 min on a hot plate set to 68°C. Add 4 g of paraformaldehyde, cover, and
stir at 68°C for 15–20 min. Cool and then filter through No. 1 Whatman paper
into a clean glass bottle and store for up to 2 wk at 4°C.
6. Note that some authors store serum at –20°C or –86°C for only 4–6 mo (28),
some evaporate the anesthesia at the time of heat-inactivation (21,26), some add
antibiotics to it (21), and some filter the culture medium after adding rat serum to
it (21,26). We do none of these things.
7. To adjust the pH of the dissection and culture media, we use 5 N NaOH and CO2.
For a 50 mL stock solution of 5 N NaOH, mix 10.0 g NaOH pellets and double-
distilled sterile water and bring to volume in a 50-mL sterile polypropylene coni-
cal tube. The tube will be very hot, but will cool in about 30 min. Store 5 N
NaOH stock indefinitely at 4°C. A CO2 tank is set up with a regulator, and clear
plastic tubing that is cleaned with absolute alcohol just prior to use. For bubbling,
attach a sterile Pasteur pipet with a non-absorbent cotton plug to the tubing end.
Verify that the stream of CO2 is not too violent before applying it to the medium.
8. For reproducibility within and between experiments, it is important that com-
plete culture medium be prepared in a consistent manner, e.g., heat-inactivation
of rat serum for exactly 30 min (whole embryo culture) or exactly 1 h (allantoic
explants), with subsequent centrifugation for exactly 5 min. Each tube or well of
a 24-well plate should contain equal amounts of culture medium. Gas- and heat-
equilibration thereafter should take place over similar well-defined time periods,
for 1–3 h.
9. From the time of removal of the uterine horns to reflection of Reichert’s mem-
brane, dissections must take no longer than 20–30 min, or subsequent growth and
development of the embryos will be compromised. The following points are key
to successful dissections and embryo culture:
• Invest in the time it requires to do perfect and timely dissections. Generally,
the novice should dissect two litters per day on at least three consecutive days
for 1 mo. Start time, the number of implantation sites, finish time, and num-
ber of perfectly dissected embryos are noted. Ultimately, all embryos should
be in perfect condition, with no nicks or deflated cavities.
19_Downs_241_272_F 269 8/29/05, 11:21 AM

270 Downs
• Work at the lowest magnification possible at each step of the technique.

• Always rest arms on glove boxes or other appropriate arm rests during the
dissections.
• Touch up forceps and clean them the day before the dissection.
• Robust forceps are used for the removal of the uterine muscle layers, shelling
out and splitting the decidua in half, and scraping out embryos.
• Fine forceps are used for reflection of Reichert’s membrane; however, if these
are too pointy, they will pierce the yolk sac and deflate it.
• Enjoy a milk product before dissecting—this has a calming effect. Avoid caf-
feine or other stimulants until proficiency is achieved.
• Although sterility will not be achieved, contamination can be minimized by
using sterile PBS, re-washing forceps before dissecting, and ensuring that
pipet tips do not contact bench surfaces.
10. On the basis of morphological scoring and comparison of freshly dissected with
cultured embryos at equivalent stages, we know that these culture conditions sup-
port (a) appropriate timing of appearance of the allantoic bud (4), (b) vascular-
ization of the allantois at the right time, and with appropriate distal-to-proximal
directionality (7), (c) entry of primitive erythroid cells into the allantois at the
right time and in numbers similar to ex vivo counterparts (7), and (d) contact and
fusion with the chorion with kinetics similar to ex vivo allantoises (12,14). Only
a very small number, approx 0.65% (18) to 1.7% (8), of embryos exhibit mor-
phological defects, typically allantoises that have failed to fuse with the chorion
at the appropriate stage. These are always discarded.
11. Some comments are presented on grafting distal allantoic tips into host
exocoelomic cavities. Depending on the biological question being addressed,
heterosynchronous grafting in which the donor allantois is either younger or older
than the host (13) might be desirable. Heterosynchronous grafts allowed us to
discover that chorio-allantoic union is dependent upon the developmental age of
the allantois; the chorion was always receptive to the allantois provided the latter
was appropriately mature. Maturity is dependent, in part, on expression of
VCAM-1 (2,9,10,14).
Acknowledgments
The author is deeply grateful to the following scientists who contributed to
the foundational success of the protocols described in this document: Professor
Sir Richard Gardner, Dr. Kirstie Lawson, the late Dr. Rosa Beddington, and
Dr. David Cockroft.
References
developmental anatomy of the murine human definitive placentae. Placenta 23,
3–19.
2. Downs, K. M. (1998) The murine allantois. Curr. Top. Dev. Biol. 39, 1–33.
19_Downs_241_272_F 270 8/29/05, 11:21 AM

3. Gardner, R. L., Lyon, M. F., Evans, E. P., and Burtenshaw, M. D. (1985) Clonal
analysis of X-chromosome inactivation and the origin of the germ line in the
mouse embryo. J. Embryol. Exp. Morph. 52, 141–152.
4. Lawson, K. A., Meneses, J., and Pedersen, R. A. (1991) Clonal analysis of epi-
blast fate during germ layer formation in the mouse embryo. Development 113,
891–911.
5. Downs, K. M., Hellman, E. R., McHugh, J., Barrickman, K., and Inman, K. (2004)
Investigation into a role for the primitive streak in development of the murine
allantois. Development 131, 37–55.
6. Sabin, F. R. (1920) Studies on the origin of blood-vessels and of red blood-cor-
puscles as seen in the living blastoderm of chicks during the second day of incu-
bation. Contr. Embryol. 9, 215–262.
7. Downs, K. M., Gifford, S., Blahnik, M., and Gardner, R. L. (1998) The murine
allantois undergoes vasculogenesis that is not accompanied by erythropoiesis. De-
velopment 125, 4507–4521.
8. Downs, K.M., Temkin, R., Gifford, S., and McHugh, J. (2001) Study of the
murine allantois by allantoic explants. Dev. Biol. 233, 347–364.
9. Gurtner, G. C., Davis, V., Li, H., McCoy, M. J., Sharpe, A., and Cybulsky, M. I.
(1995) Targeted disruption of the murine VCAM1 gene: essential role of VCAM-
1 in chorioallantoic fusion and placentation. Genes Dev. 9, 1–14.
10. Kwee, L., Baldwin, H. S., Shen, H. M., et al. (1995) Defective development of the
embryonic and extraembryonic circulatory systems in vascular cell adhesion mol-
ecule (VCAM-1) deficient mice. Development 121, 489–503.
11. Carmeliet, P., Ferreira, V., Breier, G., et al. (1996) Abnormal blood vessel develop-
ment and lethality in embryos lacking a single VEGF allele. Nature 380, 435–439.
12. Downs, K. M. and Gardner, R. L. (1995) An investigation into early placental
ontogeny: allantoic attachment to the chorion is selective and developmentally
regulated. Development 121, 407–416.
13. Yang, J. T., Rayburn, H., and Hynes, R. O. (1995) Cell adhesion events mediated
by α4 integrins are essential in placental and cardiac development. Development
121, 549–560.
14. Downs, K. M. (2002) Early placentation in the mouse. Placenta 23, 116–131.
15. Belaoussoff, M., Farrington, S. M., and Baron, M.H. (1998) Hematopoietic in-
duction and respecification of A-P identity by visceral endoderm signaling in the
mouse embryo. Development 125, 5009–5018.
16. Wilt, F. H. (1965) Erythropoiesis in the chick embryo: the role of endoderm. Sci-
ence 147, 1588–1590.
17. Friedrich, G. and Soriano, P. (1991) Promoter traps in embryonic stem cells: a
genetic screen to identify and mutate developmental genes in mice. Genes Dev. 5,
1513–1523.
18. Downs, K. M. and Harmann, C. (1997) Developmental potency of the murine
allantois. Development 124, 2769–2780.
19. Champlin, A. K., Dorr, D. L., and Gates, A. H. (1973) Determining the stage of
the estrous cycle in the mouse by the appearance of the vagina. Biol. Reprod. 8,
491–494.
19_Downs_241_272_F 271 8/29/05, 11:21 AM

272 Downs
20. Downs, K. M. and Davies, T. (1993) Staging of gastrulation in mouse embryos by

morphological landmarks in the dissection microscope. Development 118, 1255–
1266.
21. Cockroft, D. L. (1990) Dissection and culture of post-implantation mouse em-
bryos, in Postimplantation Mammalian Embryos: A Practical Approach (Copp,
A. J. and Cockroft, D. L., eds.). IRL, Oxford, UK: pp. 15–40.
22. Lawson, K. A., Meneses, J. J., and Pedersen, R. A. (1986) Cell fate and cell lin-
eage in the endoderm of the presomite mouse embryo, studied with an intracellu-
lar tracer. Dev. Biol. 115, 325–339.
23. Lawson, K. A. and Pedersen, R. A. (1987) Cell fate, morphogenetic movement
and population kinetics of embryonic endoderm at the time of germ layer forma-
tion in the mouse. Development 101, 627–652.
24. Gardner, D. K. and Lane, M. (2000) Embryo culture systems, in Handbook of In
Vitro Fertilization (Trounson, A. O. and Gardner, D. K., eds.). CRC, Boca Raton,
FL: pp. 205–263.
25. Beddington, R. S. P. and Lawson, K. A. (1990) Clonal analysis of cell lineages, in
Postimplantation Mouse Embryos: A Practical Approach (Copp, A. J. and
Cockroft, D. L., eds.). IRL, Oxford, UK: pp. 267–292.
26. Nagy, A., Gertsenstein, M., Vintersten, K., and Behringer, R. R. (2003) Manipu-
lating the Mouse Embryo: A Laboratory Manual, Third Edition Edition. Cold
Spring Harbor Laboratory, Cold Spring Harbor, NY.
27. Brown, N. A. (1990) Routine assessment of morphology and growth: scoring sys-
tems and measurements of size, in Postimplantation Mammalian Embryos: A
Practical Approach (Copp, A. J. and Cockroft, D. L., eds.). IRL Press, Oxford,
UK: pp. 93–108.
28. Brown, N. A. and Fabro, S. (1981) Quantitation of rat embryonic development in
vitro: a morphological scoring system. Teratology 24, 65–78.
29. Beddington, R. S. P. (1987) Isolation, culture and manipulation of post-implanta-
tion mouse embryos, in Mammalian Development: A Practical Approach (Monk,
M., ed.). IRL, Oxford, UK: pp. 43–70.
30. Kaufman, M. H. (1992) The Atlas of Mouse Development. Academic, London, UK.
19_Downs_241_272_F 272 8/29/05, 11:21 AM

Phenotypic Analysis of Mouse Placenta 273
IV
PHENOTYPIC ANALYSIS OF THE PLACENTA
20_Natale_273_294_F 273 8/29/05, 11:21 AM

274 Natale, Starovic, and Cross
20_Natale_273_294_F 274 8/29/05, 11:21 AM

20
Phenotypic Analysis of the Mouse Placenta
David R. C. Natale, Maja Starovic, and James C. Cross
Summary
Placental development is a dynamic and complex process and much of our current under-
standing of the underlying molecular processes comes from analysis of targeted gene mutations
in mice. There are more than 50 strains of mutant mice that have placental defects, and it has
become widely appreciated that placental defects should be suspected in cases where embry-
onic lethality is observed. The degree to which these phenotypes are investigated is highly
variable, owing to a general lack of expertise in the field. However, there has been considerable
progress in developing techniques and reagents for analyzing placental phenotypes that are
relatively simple to apply and that should be accessible to all investigators. This chapter pro-
vides a basic outline of the strategies for the general identification and then the subsequent
detailed investigation of placental phenotypes.
Key Words: Transgenic mice; knockout mice; histology; gene.
1. Introduction
The mouse has become a dominant model system for studying the function
of genes in mammals because of the ability to alter gene functions through
transgenic and gene knockout approaches. The vast majority of embryonic
lethal phenotypes that are due to loss-of-function mutations are associated
with placental defects and, as such, we now know the molecular basis of many
aspects of placental development and function (1–5). In most cases in which
investigators have made knockout mice, a placental phenotype was completely
unsuspected. Indeed, in some cases, phenotypes were initially attributed to
other defects but have been more recently re-interpreted as being due to the
placenta. These problems have highlighted the fact that the placenta, until
recently, has been underappreciated in the broad biomedical research field.
Considerable progress has been made in the last decade, however, in our under-
standing of how the mouse placenta develops and in development of approaches
and methods for studying it in detail. Given this progress, it is now possible to
275
20_Natale_273_294_F 275 8/29/05, 11:21 AM

expect that detailed analysis of placental phenotypes in mutant mice should

be routine in most laboratories and, therefore, that the standards for pheno-
type analysis should rise. This chapter outlines the general approaches that we
use in our lab for determining whether a placental phenotype exists in a mutant
mouse and what specific cell type or structure is affected.
The mouse placenta develops through four distinct milestone stages that rep-
resent the major morphological phases of placental development and also the
four major classes of mouse placental phenotypes (for more extensive reviews,
see refs. 1 and 3–5):
1. Post-implantation period (embryonic days [E] 4.5 to 8). Immediately after implan-
tation, the mural trophectoderm transforms into primary trophoblast giant cells that
invade into the uterus (1). The polar trophectoderm proliferates in response to
mitogenic signals from the epiblast to form the extraembryonic ectoderm (proxi-
mal) and ectoplacental cone (distal). The trophoblast stem cell population resides
in the extraembryonic ectoderm compartment (6), whereas cells of the ectopla-
cental cone have a more limited proliferation potential and also differentiate into
secondary trophoblast giant cells (1). After gastrulation begins (E 6.5), the ex-
traembryonic ectoderm becomes separated from the epiblast as a result of migra-
tion of mesoderm and forms a distinct flat plate of cells called the chorion (Fig. 1).
The chorionic trophoblast cells continue to express genes typical of extraembry-
onic ectoderm at earlier stages (e.g., Cdx2, Eomes) and so are also presumed to
be trophoblast stem cells.
2. Chorioallantoic attachment at E 8.5. Mesodermal cells in the allantois grow out
from the primitive streak and make contact with the basal surface of the chorion
at E 8.5 (7). The initial contact surface is actually a thin mesothelial layer that
underlies the chorionic trophoblast cells (Fig. 1).
3. Labyrinth morphogenesis from E 9 to 17. After chorioallantoic attachment, the
initially flat chorion begins to fold into primary villi that are filled with blood
vessels and stromal cells that differentiate from the allantoic mesoderm (Fig. 1)
(3,5,7). Through extensive branching morphogenesis, the villi elongate and
develop a complex array of secondary branches. There are multiple distinct
trophoblast cell subtypes within the labyrinth and though it is common for people
to refer to “labyrinthine trophoblast,” there is no such single cell type. Chorionic
trophoblast cells differentiate into two layers of multinucleated syncytiotropho-
blast that form the major surface barrier for nutrient exchange between the mater-
nal blood sinuses in the labyrinth and the fetal capillaries. Mononuclear trophoblast
cells of unknown function sit within the lumen of the maternal blood sinuses.
Finally, small islands of densely packed, cuboidal trophoblast cells persist within
the labyrinth layer well into late placental development. The morphology of these
cells is typical of extraembryonic ectoderm/chorionic trophoblast and they also
express markers typical of these cells (8).
4. Endovascular and interstitial trophoblast invasion. There are two types of tro-
phoblast invasion in mice that are characterized by different patterns and timing
(9). A subtype of trophoblast giant cell invades into the maternal spiral arteries
20_Natale_273_294_F 276 8/29/05, 11:21 AM

Fig. 1. Low and high magnification micrographs of sagittal histological sections of

the placenta at embryonic day (E) 7.5, E 8.0 and E 9.5. The sections were stained with
hemotoxylin and eosin and show the basic structures of the developing placenta. At
E 8.0, the allantois is shown in approximation to the mesothelial layer that underlies
the chorionic trophoblast cells. By E 9.5, chorioallantoic attachment has occurred and
the chorion layer is starting to fold (arrowheads) to form the primary villi of the laby-
rinth layer. The location of the trophoblast giant cell layer is shown by the dotted line.
The boxes in low magnification micrographs indicate the location of the correspond-
ing high magnification micrograph. Al, allantois; Ch, chorion; EPC, ectoplacental
cone; FC, fetal capillary; Me, mesothelium; TGC, trophoblast giant cell.
20_Natale_273_294_F 277 8/29/05, 11:21 AM

that bring maternal blood to the implantation site, and these cells are detectable
soon after implantation. A subtype of spongiotrophoblast cell, called glycogen tro-
phoblast cell, invades interstitially into the uterus but only starting after E 12.5.
In analyzing placental development in mouse mutants, we take a systematic
approach that first involves assessing the stage(s) that may be affected. In gen-
eral, this is done by defining if there is a specific period during gestation when
placental function is abnormal and by noting whether the major developmental
milestones have been achieved. This is usually best accomplished by doing
gross dissection of tissues. This is then followed by histological and marker
analysis. It is important to take this two-stepped approach, and a common mis-
take is to immediately dive into histological analysis, which is very time-con-
suming if done properly. In this chapter we outline the general approaches and
refer to the specific techniques that we use for detailed analysis. Many of the
detailed protocols have been published elsewhere and as such will not be
repeated here.
2. Materials
1. Dissecting scissors.
2. Dissecting stereomicroscope.
3. Watchmaker’s forceps: Dumont #55 (Fine Science Tools Inc., Foster City, CA).
4. 60-mm and 100-mm Petri dishes for dissections.
5. Ice-cold phosphate-buffered saline (PBS).
6. Ice-cold 4% paraformaldehyde (PFA) in PBS, prepared fresh. Made from a 20%
stock solution that can be stored at –20°C.
7. Proteinase K lysis buffer: 50 mM KCl; 10 mM Tris-HCl, pH 8.3; 2.0 mM MgCl2;
0.1 mg/mL gelatin; 0.45% Tween-20; 0.45% Nonidet P-40; 0.1 mg/mL proteinase K.
8. NT solution: 0.15 M NaCl; 0.1 M Tris, pH 7.5.
9. NTMT solution: 0.1 M NaCl; 0.1 M Tris, pH 9.5; 0.05 M MgCl2; 0.1% Tween-
20. Prepare fresh daily.
10. Alkaline phosphatase substrate: 5-bromo,4-chloro,3-indolylphosphate (BCIP)/
nitroblue tetrazolium (NBT) substrate kit (Vector Laboratories, Burlington, ON,
Canada, Cat. No. SK-5400).
11. Nuclear Fast Red (Vector Laboratories, Cat. No. H-3404).
12. Peroxidase-conjugated Isolectin BS1 (Sigma-Aldrich, St Louis, MO, cat no.
L-5391).
13. Rabbit polyclonal anti-laminin antibody (Sigma-Aldrich, cat. no. L-9393).
3. Methods
3.2. Gross Dissections to Survey Stages of Placental Development
and Function
The most critical first step in determining whether you have a placental phe-
notype is to look for signs of placental dysfunction. Mild placental dysfunction
20_Natale_273_294_F 278 8/29/05, 11:21 AM

or mal-development might manifest as fetal growth restriction resulting in

lower birthweight, whereas severe dysfunction would result in embryonic
lethality corresponding to a deficiency of pups at birth. For any new mutant
or transgenic line, we always take the approach of letting litters go to term,
quietly observing the litters soon after birth to look for uniformity of birth
weight and to assess litter size, and then letting the pups reach weaning age
(3 wk postnatally) before the pups are individually identified (e.g., by ear
notches, tag, or tattoo) and genotyped. We will do this until at least 40 pups
from four different dams have been observed before doing anything more inva-
sive. With this number of progeny, you should see the emergence of Mendelian
ratios of genotypes.
If there is evidence of a significant proportion of pups born with low
birthweight, but that by 3 wk there is no such variation, then pups can be tat-
tooed at birth and followed until weaning (see Note 1). If the litter size at birth
and at weaning is the same, but there are no animals with a mutant genotype
represented, then it is safe to assume embryonic lethality (see Note 2).
If there is evidence of embryonic lethality, we begin our survey by observ-
ing conceptuses in utero and collecting DNA for genotyping at E 10.5 (see
Note 3). Once at least 40 conceptuses from four different dams have been ob-
served on gestation day 10.5, and correlations made between genotypes and gross
morphology, then the decision simply becomes whether to look earlier or later
than E 10.5. If no resorptions are observed and yet the mutant genotype class is
not represented, then it is likely that mutants failed to even implant.
The analysis of the pregnant uterus begins by observation in situ to look for
the number of implantation sites, and signs of hemorrhage or overt resorption.
The uterus is then removed by using scissors to cut through the cervix and then
using forceps to sharply lift up the cervix in order to tear the membranous
attachment of the uterus to the upper abdomen (the mesometrium), leaving the
uterus attached to the carcass only at points where it connects with the oviducts
and ovaries. The mesometrial attachment site is a major landmark as this repre-
sents the side of the uterus into which the uterine artery flows and, as such, the
fetal placenta is oriented towards that side. Trim away as much of the fatty
membrane as possible as it interferes with the subsequent dissections. Then cut
through the oviducts and transfer the intact uterus to a Petri dish containing
PBS. Use scissors to cut between each implantation site (Fig. 2). The uterus is
removed from the around the decidual tissue by using fine forceps (Fig. 2).
One forcep is used to grasp the cut edge and the other is used to tear along the
antimesometrial surface. With experience, it is possible to leave the uterus
intact and to tear along the entire length by alternating the holding and tear-
ing forceps and working back and forth down the length. The uterus shells
away from the decidua easily on the antimesometrial side and a gentle squeez-
20_Natale_273_294_F 279 8/29/05, 11:21 AM

Fig. 2. Anatomy and dissection of the pregnant mouse uterus to remove the decidual
swellings. Briefly, individual implantation sites are separated. The uterine muscle and
membranes are removed by using forceps to carefully tear them away along the
antimesometrial side of the implantation site. The implantation site can then be
removed from the uterus on the mesometrial side by using one set of forceps to
hold the uterus while using a second set to gently shell the implantation site away.
ing motion with the forceps can lift the decidua away from the mesometrial
side. This same general approach is used to remove implantation sites at all
stages of gestation.
3.2. Analysis of the Placenta Between E 5.5 and 8.5
3.2.1. Examination of the Placental Structures
During the first few days after implantation, the conceptus is surrounded by
a relatively thick layer of decidual tissue. The conceptus is positioned within
the tissue in a stereotypical pattern. The conceptus is located centrally in the
radial axis. The ectoplacental cone sits at roughly the mid-point along the
mesometrial to antimesometrial axis and one can usually see an accumula-
tion of maternal blood in a “red band” at this position (Fig. 3). As the concep-
tus grows, the decidual tissue thins around the radial diameter and therefore the
outermost layer of the conceptus (trophoblast giant cells of the parietal yolk
sac), are located closer to the outside. Using these general rules and landmarks
we use one of two approaches to remove the conceptus from the decidua
between E 5.5 and 8.5. In order to visualize the conceptus intact we split the
deciduas along the mesometrial to antimesometrial axis (Fig. 3, longitudinal
bisection). For conceptuses at E 8.5, we position forceps at about the 30:70
position and slowly squeeze the points together in a scissor-like cut. This will
20_Natale_273_294_F 280 8/29/05, 11:21 AM

20_Natale_273_294_F
281
281
Phenotypic Analysis of Mouse Placenta
Fig. 3. Alternative approaches for opening the decidual swellings. Implantation sites can be bisected either transversely
or longitudinally with respect to the mesometrial–antimesometrial axis to facilitate the separation of the embryo and pla-
centa into separate compartments of the implantation site or to allow observation of the intact embryo and placenta respec-
tively. To bisect the implantation site, place it gently between fine forceps and carefully close them along the axis you wish
cut. Using a scalpel blade, or one side of a second set of forceps as a blade, make a cut along the line of the first set of forceps
8/29/05, 11:21 AM
to bisect the implantation site as shown.
281
Table 1
Molecular Markers of Placental Development From Early Postimplantation to
Embryonic Day (E) 8.5
Gene Site
of expression Technique Comments Reference
Pem ExE ISH, IHC E 5.5 in extraembryonic tissues; 19
later in EPC, Ch, SGC
Eomes ExE ISH E 5.5 in extraembryonic tissues; 20
later in EPC, Ch and embryo;
required for trophoblast devel-
opment following blastocyst
formation
Estrrb ExE ISH E 5.5—in extraembryonic 21
ectoderm; E 7.5—specific to
chorion; diminishes following
chorioallantoic attachment
Limk TGC ISH Expressed exclusively in giant 22
cells from E 4.5 onwards
Mash2 EPC, Ch ISH 23
Pl1 TGC ISH Detectable in primary GCs 24
lining the embryonic cavity
Plf TGC ISH Detectable in primary and 25
secondary GCs
ISH, in situ hybridization; IHC, immunohistochemistry; ExE, extraembryonic ectoderm; EPC,

ectoplacental cone; TGC, trophoblast giant cell; SGC, secondary giant cell; Ch, chorion.
make a glancing cut through the parietal yolk sac but leave the ectoplacental
cone and embryo intact. The ectoplacental cone is recognizable because it is
red as a result of the presence of maternal blood. The smaller piece of decidua
will contain part of the parietal yolk sac and can be used to examine tropho-
blast giant cells. For conceptuses younger than E 8.5, the forceps should be
positioned closer to the midline since the conceptuses are much smaller.
The ectoplacental cone and extraembryonic ectoderm/chorion are best stud-
ied in detail using histological and marker analysis (Fig. 1; Table 1). Most
current markers that are available are mRNAs and therefore must be detected using
RNA in situ hybridization. Given that both ectoplacental cone and extraembry-
onic ectoderm/chorion cells proliferate, if the relative size of these tissues is
reduced, it is useful to assess the cell proliferation index by using standard
approaches such as BrdU incorporation, or Ki67 or proliferating cell nuclear
antigen (PCNA) immunostaining which all identify cells undergoing or that
20_Natale_273_294_F 282 8/29/05, 11:21 AM

have recently undergone DNA replication. It is important to supplement these

methods with an assessment of mitosis (e.g., phospho-histone immunostaining),
however, in order to distinguish proliferating cells from those undergoing
endoreduplication. Trophoblast giant cells get their name from the fact that
they are large polyploid cells that arise from repeated rounds of DNA replica-
tion without intervening mitoses (endoreduplication) (10).
To study trophoblast giant cells, assess their numbers, size, and gene
expression, it is convenient to stain them using the split pieces of decidual
tissue as described above. During the early postimplantation period, the Pl1
gene is expressed in a trophoblast giant cell-specific manner and in situ hybrid-
ization gives very robust signals. In whole mount in situ hybridization prepara-
tions, the mRNA is localized in a perinuclear pattern and therefore the signal
gives a clear outline of the nucleus (11). The major features to note are the
overall expression level, number of trophoblast giant cells per implantation
site, and cell/nuclear size. A variety of other markers can be used for follow-up
studies (Table 1). Another very useful approach for studying trophoblast giant
cell differentiation is to perform cultures with either dispersed ectoplacental
cone cells (digested in 0.25% trypsin for 5 min) or intact blastocysts in
Dulbecco’s modified Eagle’s medium (DMEM) containing 5% fetal bovine serum
(12). These cells are very hearty and will readily attach to regular tissue culture
dishes. Fixed cells can be stained with 4',6-diamidino-2-phenylindole (DAPI)
in order to visualize the nucleus of the cells and fluorescence intensity can be
measured as an estimate of DNA content (13,14) (see Note 4).
3.2.2. Preparation of DNA From Yolk Sac or Small Numbers of Cells

Genotyping embryos at early postimplantation stages is challenging because
of their small size and the limited amount of tissue that is available for sam-
pling. For embryos at E 8.5 and older, we routinely collect the visceral yolk
sac. For smaller embryos, DNA can be isolated from fixed tissue and therefore
we usually complete our analysis (including in situ hybridization) and then
digest the entire specimen. For material that is sectioned, we scrape cells off of
one of the sections using a 30-gage needle and transfer to lysis buffer.
DNA for polymerase chain reaction (PCR) analysis is prepared using a
simple cell lysis protocol.
1. The tissue or cells are placed into 50 to 100 µL of proteinase K lysis buffer and
incubated for at least four h (or overnight) at 55°C.
2. The samples are then incubated for 10 to 15 min at 95°C to inactivate the protein-
ase K, after which 1 to 2 µL can then be used directly in a PCR reaction.
3. For very small numbers of cells, the volume of lysis buffer can be reduced by
half.
20_Natale_273_294_F 283 8/29/05, 11:21 AM

Table 2
Molecular Markers of Placental Development at the Time of Chorioallantoic
Attachment
Gene Site of expression Technique Comments Reference
α4 integrin Basal surface IHC Required for chorioallantoic 15

of chorion attachment
Vcam1 Distal two-thirds
of allantois IHC Required for chorioallantoic 16,26
attachment
IHC, immunohistochemistry.
3.3. Analysis of the Placenta at the Time of Chorioallantoic Attachment

Whenever examining conceptuses at E 8.5 or later, one should always deter-
mine whether or not chorioallantoic attachment has occurred correctly. By dis-
secting the decidua/conceptus as described above, it will provide the
investigator with a clear view of the chorioallantoic interface (Fig. 1). Impor-
tantly, the allantois and later the umbilical cord are delicate and so it is impor-
tant to be careful when doing these dissections as they can easily tear. If the
allantois fails to attach to the chorion, it will usually retract and appear as small
ball of tissue attached to the abdominal wall of the embryo. The cell adhesion
molecules α4 integrin (15) and vascular cell adhesion molecule (VCAM)1 (16)
are essential for chorioallantoic attachment and therefore their expression
should be assessed if attachment fails to occur, and this is best accomplished
using immunostaining of sectioned material (Table 2).
3.4. Examination of the “Mature” Placenta: Spongiotrophoblast and
Labyrinth
3.4.1. General Dissection and Routine Histological Analysis
After chorioallantoic attachment at E 8.5, there is extensive vascular devel-
opment within the allantoic mesoderm and one should normally see a promi-
nent umbilical artery and vein within the umbilical cord, as well as prominent
placental vessels fanning out across the bottom surface of the placenta. Other
than these superficial vessels, however, it is difficult to assess if the labyrinth
is developing correctly upon gross dissection and any further study requires
histological analysis. There are two general dissection approaches that can be
taken for these older conceptuses. By E 9.5 and later, the decidual layer is
extremely thin and, therefore, the technique should be modified. Using the for-
ceps like scissors, a one-quarter thickness cut should be placed just below and
parallel to the base of the placenta. This will cut through both parietal and
20_Natale_273_294_F 284 8/29/05, 11:21 AM

visceral yolk sac creating a window into the extraembryonic cavity. The cut
can then be extended around the circumference thus removing the cup-shaped
decidua/yolk sac tissue, leaving the amnion-covered fetus attached to the pla-
centa by the umbilical cord. A second and even simpler technique can be used
to separate the fetus from placenta, once you are assured that chorioallantoic
attachment is normal based on dissection of other litters—simply place forceps
across the full thickness of the decidua, just below and parallel to the base of
the placenta, slowly squeeze shut and make a cut along the edge of the forceps.
This separates the placenta from the fetus/yolk sac/decidual tissue in one rapid
step (Fig. 3, transverse bisection).
For routine purposes, we fix tissues in 4% PFA, embed in paraffin, and then
make serial sagittal sections through the entire block, mounting two sections
per slide. We then stain every 10th slide with hematoxylin/eosin (H&E). Slides
are examined in order to find the midpoint of the placenta (site of umbilical
attachment), which is used as the major reference point for comparisons
between mutants and wild-type littermates. The major structures to note
are the number and size of trophoblast giant cells, and the thickness of the
spongiotrophoblast and labyrinth layers. These layers have distinct morpho-
logical features that are clear even from H&E-stained materials, though a num-
ber of molecular markers are helpful in defining the layers at a gross level (see
Table 3, Fig. 4, and Note 5). A systematic approach is required in order to
evaluate the development of both the trophoblast and vascular compartments
of the placenta, and a common mistake is to only describe the presence or
absence of fetal capillaries (see Note 6).
3.4.2. Alkaline Phosphatase Histochemical Staining for Trophoblast
Cells Lining the Maternal Blood Sinuses
Trophoblast cells that line the maternal blood spaces express endogenous
alkaline phosphatase activity that can be detected by histochemical staining of
histological sections (Fig. 4).
1. Briefly, following de-waxing and rehydration of tissue sections, slides are washed
in NT solution for 20 min at room temperature followed by washing in NTMT
solution for 10 min at room temperature.
2. To visualize enzymatic activity, the color is developed using a standard alkaline
phosphatase substrate (e.g., BCIP/NBT).
3. Sections are then stained with Nuclear Fast Red and mounted following dehydra-
tion and clearing through a graded series of ethanol and xylene washes.
3.4.3. Lectin Histochemistry

As an alternative, trophoblast cells lining maternal blood spaces of the laby-
rinth layer can also be identified by the presence of a cell-surface carbohydrate
structure that is recognized by Isolectin BS1. Staining is done by incubating
20_Natale_273_294_F 285 8/29/05, 11:21 AM

286
20_Natale_273_294_F
Table 3
Molecular Markers of Placental Development Associated With Placental Labyrinth Development
Gene Site of expression Technique Comments Reference
Gcm1 Ch, Lab ISH— E 8.5—expressed in subsets of trophoblast 18,27
286
wholemount on chorionic plate; E 9.0 demarcates
chorioallantoic branching and then
expressed in syncytiotroph. in labyrinth
layer; required for chorioallantoic
branching and labyrinth formation
Eomes Lab ISH Marker of trophoblast stem cells 20
Tpbpα EPC, Sp ISH 28
286
Mash2 EPC, Sp ISH Mutant results in loss of spongiotrophoblast
layer and compact labyrinth 23
Hand1 EPC, TGC ISH Expression overlaps with Mash2 in EPC;
required for TGC differentiation 11,29,30
Pl1 TGC ISH, IHC PL1 protein detectable E 9–E 10, not after E 10 24
Pl2 TGC ISH, IHC PL2 protein detectable beginning E 10 31
Plf TGC ISH Expressed strongly E 8–E 10 25
ISH, in situ hybridization; IHC, immunohistochemistry; EPC, ectoplacental cone; TGC, trophoblast giant cell; Ch, chorion; Lab,
8/29/05, 11:21 AM
labyrinth layer; Sp, spongiotrophoblast
Natale, Starovic, and Cross
Fig. 4. Low and high magnification micrographs of sagittal sections from a pla-
centa at embryonic day (E) 11.5 stained for various markers. Alkaline phosphatase
activity and laminin immunoreactivity are used to identify maternal and fetal-derived
blood spaces in the labyrinth layer, respectively. Tpbpα and proliferin (Plf) mRNAs
are used to identify the spongiotrophoblast and trophoblast giant cell layers respec-
tively. F, fetal capillary; M, maternal blood space; Sp, spongiotrophoblast; TGC, tro-
phoblast giant cell.
20_Natale_273_294_F 287 8/29/05, 11:21 AM

Isolectin BS1 conjugated to peroxidase using a protocol similar to a standard

immunohistochemistry protocol.
1. Briefly, following a blocking step, sections are incubated in isolectin BS1 diluted
to 25 µg/mL in PBS for 60 min at room temperature.
2. Sections are washed in PBS + 1.0 % bovine serum albumin (BSA).
3. Detection of lectin binding is by a standard peroxidase substrate reaction.
3.4.4. Detection of Capillary Endothelium

Laminin staining of basement membrane in the labyrinth can be used for
identification of fetal capillaries (Fig. 4). Rabbit polyclonal anti-laminin anti-
body is used in a standard immunohistochemistry protocol diluted to a concen-
tration of 1:500 from stock concentration. Importantly, the endothelial cells in
the labyrinth do not stain with anti-Factor VIII antibody, though they do react
with anti-platelet/endothelial cell adhesion molecule (PECAM)1/CD31.
3.4.5. Electron Microscopy
In addition to the marker analysis described above, transmission electron
microscopy is a very powerful but under-utilized technique for studying the
labyrinth. Its value is in the ability to observe the array of cell types within the
labyrinth that separate the maternal blood sinuses from the fetal capillaries
which are so thin that they cannot adequately be resolved with light micros-
copy (3). Also, it allows the investigator to closely examine the ultrastructure
of intercellular junctions and basement membranes.
3.4.6. Distinguishing Trophoblast vs Vascular Phenotypes in the Labyrinth
Because the labyrinth is composed of multiple cell types that are derived
from two distinct cell lineages (trophoblast and mesoderm) and their develop-
ment is interdependent, it can be difficult at times to sort out which specific
cell type(s) are responsible for a morphological phenotype (see Note 6). How-
ever, chimeric embryo and conditional knockout approaches are two very pow-
erful means of at least sorting out which cell lineage is responsible. Briefly,
tetraploid chimeras can be used to provide wild-type cells to trophoblast (and
extraembryonic endoderm), whereas embryonic stem (ES) cells contribute to
fetus as well as extraembryonic mesoderm (including allantois) but not tropho-
blast (17). Mox2-Cre transgenic mice are useful because Cre recombinase is
expressed in the fetus as well as extraembryonic mesoderm but not trophoblast
(8). To date, a pan-trophoblast specific Cre mouse has not been described.
3.5. Trophoblast Invasion
Mouse mutants in which trophoblast invasion is affected have not yet been
described. However, this may be a simple oversight, because it has been only
20_Natale_273_294_F 288 8/29/05, 11:21 AM

Fig. 5. Anatomical organization of the mature placenta and the maternal spiral
arteries. The arrows highlight the orientation of tissue sectioning for sagittal and
transverse sections. Sagittal tissue sections are used for the routine analysis of mor-
phology and marker gene expression. Serial transverse sections are used for detailed
assessment of endovascular trophoblast invasion.
recently that the normal patterns of invasion have been described and methods
developed to identify the two distinct invading cell populations (9). Sagittal
sections can be used to generally survey the extent of invasion, but serial cross
sections (Fig. 5) are needed in order to study the course of invasion of the
endovascular trophoblast giant cells because they traffic along spiral shaped
arteries that come in and out of the planes of sagittal sections. The endovascular
giant cells are apparent from the early postimplantation period to term and are
characterized by positive staining for Plf mRNA, but negative for Pl1 and
Tpbpα mRNA and periodic acid-Schiff (PAS) staining. The interstitial invad-
ing glycogen trophoblast cells are negative for Plf and Pl1 mRNA, but positive
for Tpbpα mRNA and PAS staining (9). Because of the close relationship be-
tween endovascular trophoblast giant cells and maternal spiral artery develop-
ment, the information in a survey of these trophoblast cell subtypes in serial
sections is greatly enhanced by doing a comparative analysis with markers for
endothelium (e.g., Factor VIII or CD31/PECAM1), smooth muscle actin and
NK cells (which are implicated in spiral artery dilation) (9).
20_Natale_273_294_F 289 8/29/05, 11:21 AM

4. Notes
1. Identification of pups. Tatooing pups should not routinely be done at birth because
many females are stressed by the intrusion and will reject or cannibalize their
young.
2. Verification of an embryonic lethal phenotype. An embryonic-lethal phenotype
can be difficult to prove as pups that are born stillborn, or that die within hours of
birth, are usually cannibalized by the female. For this reason, it is a good idea to
check the cage in the morning after a delivery to look for signs of dead pups,
before putting a lot of effort into assessing intrauterine development. This is
important because observing intrauterine development involves having to sac-
rifice the pregnant female, and such females are usually in short supply when a
phenotyping project first begins.
3. Initial intrauterine investigation of an embryonic lethal phenotype. We begin our
survey of embryonic lethal phenotypes by examining conceptuses at E 10.5. This
time is chosen for three reasons. First, it is half way through gestation and there-
fore it is convenient to address if lethality occurs in the first or second half of
gestation. Second, the vast majority of placental mutants that have been described
to date occur around mid-gestation. Third, if mutant conceptuses were able to
implant (E 4.5) and therefore initiate a decidual response, there should still be
evidence of it within the uterus even if development of the conceptuses stalled
immediately after implantation. By gestation day 10.5, such implantation sites
would show signs of advanced resorption (hemorrhage, necrosis) and usually no
useful tissue for genotyping can be recovered. However, their presence should
be noted and the objective is to see if the frequency of resorptions accounts for
the mutant class. It is important to remember that there is a normal background
rate of resorptions even in wild-type mice, but in outbred strains of mice the
normal resorption rate is usually less than 5%.
4. Fluorescence-activated cell sorting (FACS) and trophoblast giant cells. In our
experience, trophoblast giant cells cannot reliably be analyzed using FACS. Their
large size results in cell fragmentation during sorting and they also tend to clump.
5. Complexity of the labyrinth layer. The labyrinth layer of the placenta is very com-
plex in that, as described in the Introduction, it contains several trophoblast sub-
types as well as stromal and vascular cell types. H&E staining does not provide
sufficient detail to accurately describe the relative numbers of these cell types.
Therefore, detailed analysis is critical and there are now several robust markers
available (Table 3). Morphometric techniques should also be used to quantitate the
relative volumes of each layer, and relative densities of various cell types and struc-
tures particularly within the labyrinth (e.g., maternal blood sinus, syncytiotropho-
blast layers, mononuclear trophoblast, feto-placental capillaries).
6. Phenotypes affecting the labyrinth layer. Of all the parts in the placenta, the laby-
rinth layer is the one that is most commonly misunderstood and poorly described
in descriptions of mutant phenotypes. A very common mistake is that investiga-
tors simply look at whether the labyrinth is vascularized. Because fetal red blood
cells are nucleated, they are certainly easy to spot. However, it is important to
20_Natale_273_294_F 290 8/29/05, 11:21 AM

remember that the fetal vascular network cannot develop unless chorioallantoic
morphogenesis and trophoblast differentiation has occurred to create the villi into
which the vessels develop. Therefore, a simple reduction in the thickness of the
labyrinth layer due to reduced formation of villi would have secondary effect that
would superficially look like “reduced vascularization” of the labyrinth. In Gcm1
mutants, for example, the labyrinth layer completely fails to form as a result of a
block in chorioallantoic morphogenesis (18). In many other cases, the labyrinth
defect is less dramatic and morphometric analysis can be very helpful in identify-
ing the true cause of the phenotype. A case in point is the phenotype of Rb
mutants in which trophoblast differentiation is impaired because of a reduction
in the ability of trophoblast stem cells to exit the cell cycle (8). Using morpho-
metric analysis to assess the volume densities of differentiated trophoblast (villi)
and fetal capillaries, significant reductions are observed in both. However, the
reduction in capillary density within the overall labyrinth is less than the decrease
in villous density. Indeed, in areas of the labyrinth in which villi do form, the rela-
tive capillary density is higher (8).
References
1. Cross, J. C., Werb, Z., and Fisher, S. J. (1994) Implantation and the placenta: key
pieces of the development puzzle. Science 266, 1508–1518.
2. Copp, A. J. (1995) Death before birth: clues from gene knockouts and mutations.
Trends Genet. 11, 87–93.
3. Cross, J. C. (2000) Genetic insights into trophoblast differentiation and placental
morphogenesis. Semin. Cell Dev. Biol. 11, 105–113.
4. Hemberger, M. and Cross, J. C. (2001) Genes governing placental development.
Trends Endocrinol. Metab. 12, 162–168.
5. Rossant, J. and Cross, J. C. (2001) Placental development: lessons from mouse
mutants. Nat. Rev. Genet. 2, 538–548.
7. Downs, K. M. (2002) Early placental ontogeny in the mouse. Placenta 23, 116–131.
8. Wu, L., de Bruin, A., Saavedra, H. I., et al. (2003) Extra-embryonic function of
Rb is essential for embryonic development and viability. Nature 421, 942–947.
9. Adamson, S. L., Lu, Y., Whiteley, K. J., et al. (2002) Interactions between tropho-
blast cells and the maternal and fetal circulation in the mouse placenta. Dev. Biol.
250, 358–373.
10. Cross, J. C., Baczyk, D., Dobric, N., et al. (2003) Genes, development and evolu-
tion of the placenta. Placenta 24, 123–130.
11. Scott, I. C., Anson-Cartwright, L., Riley, P., Reda, D., and Cross, J. C. (2000) The
HAND1 basic helix-loop-helix transcription factor regulates trophoblast differen-
tiation via multiple mechanisms. Mol. Cell. Biol. 20, 530–541.
12. Wang, J., Paria, B. C., Dey, S. K., and Armant, D. R. (1999) Stage-specific exci-
tation of cannabinoid receptor exhibits differential effects on mouse embryonic
development. Biol. Reprod. 60, 839–844.
20_Natale_273_294_F 291 8/29/05, 11:21 AM

13. Nakayama, H., Scott, I. C., and Cross, J. C. (1998) The transition to
endoreduplication in trophoblast giant cells is regulated by the mSNA zinc finger
transcription factor. Dev. Biol. 199, 150–163.
14. MacAuley, A., Cross, J. C., and Werb, Z. (1998) Reprogramming the cell cycle
for endoreduplication in rodent trophoblast cells. Mol. Biol. Cell 9, 795–807.
15. Yang, J. T., Rayburn, H., and Hynes, R. O. (1995) Cell adhesion events mediated
by alpha 4 integrins are essential in placental and cardiac development. Develop-
ment 121, 549–560.
16. Gurtner, G. C., Davis, V., Li, H., McCoy, M. J., Sharpe, A., and Cybulsky, M. I.
(1995) Targeted disruption of the murine VCAM1 gene: essential role of VCAM-
1 in chorioallantoic fusion and placentation. Genes Dev. 9, 1–14.
17. Rossant, J. (2001) Stem cells from the Mammalian blastocyst. Stem Cells 19,
477–482.
18. Anson-Cartwright, L., Dawson, K., Holmyard, D., Fisher, S. J., Lazzarini, R. A.,
and Cross, J. C. (2000) The glial cells missing-1 protein is essential for branching
morphogenesis in the chorioallantoic placenta. Nat. Genet. 25, 311–314.
19. Lin, T. P., Labosky, P. A., Grabel, L. B., et al. (1994) The Pem homeobox gene is
X-linked and exclusively expressed in extraembryonic tissues during early mu-
rine development. Dev. Biol. 166, 170–179.
20. Russ, A. P., Wattler, S., Colledge, W. H., et al. (2000) Eomesodermin is required
for mouse trophoblast development and mesoderm formation. Nature 404, 95–99.
21. Luo, J., Sladek, R., Bader, J. A., Matthyssen, A., Rossant, J., and Giguere, V.
(1997) Placental abnormalities in mouse embryos lacking the orphan nuclear re-
ceptor ERR-beta. Nature 388, 778–782.
22. Cheng, A. K. and Robertson, E. J. (1995) The murine LIM-kinase gene (limk)
encodes a novel serine threonine kinase expressed predominantly in trophoblast
giant cells and the developing nervous system. Mech. Dev. 52, 187–197.
23. Guillemot, F., Nagy, A., Auerbach, A., Rossant, J., and Joyner, A. L. (1994) Es-
sential role of Mash-2 in extraembryonic development. Nature 371, 333–336.
24. Faria, T. N., Ogren, L., Talamantes, F., Linzer, D. I., and Soares, M. J. (1991)
Localization of placental lactogen-I in trophoblast giant cells of the mouse pla-
centa. Biol. Reprod. 44, 327–331.
25. Linzer, D. I., Lee, S. J., Ogren, L., Talamantes, F., and Nathans, D. (1985) Identi-
fication of proliferin mRNA and protein in mouse placenta. Proc. Natl. Acad. Sci.
USA 82, 4356–4359.
26. Kwee, L., Baldwin, H. S., Shen, H. M., et al. (1995) Defective development of the
embryonic and extraembryonic circulatory systems in vascular cell adhesion mol-
ecule (VCAM-1) deficient mice. Development 121, 489–503.
27. Stecca, B., Nait-Oumesmar, B., Kelley, K. A., Voss, A. K., Thomas, T., and
Lazzarini, R. A. (2002) Gcm1 expression defines three stages of chorio-allantoic
interaction during placental development. Mech. Dev. 115, 27–34.
28. Lescisin, K. R., Varmuza, S., and Rossant, J. (1988) Isolation and characterization
of a novel trophoblast-specific cDNA in the mouse. Genes Dev. 2, 1639–1646.
20_Natale_273_294_F 292 8/29/05, 11:21 AM

29. Cross, J. C., Flannery, M. L., Blanar, M. A., et al. (1995) Hxt encodes a basic
helix-loop-helix transcription factor that regulates trophoblast cell development.
30. Riley, P., Anson-Cartwright, L., and Cross, J. C. (1998) The Hand1 bHLH tran-
scription factor is essential for placentation and cardiac morphogenesis. Nat.
Genet. 18, 271–275.
31. Jackson, L. L., Colosi, P., Talamantes, F., and Linzer, D. I. (1986) Molecular clon-
ing of mouse placental lactogen cDNA. Proc. Natl. Acad. Sci. USA 83, 8496–8500
20_Natale_273_294_F 293 8/29/05, 11:21 AM

20_Natale_273_294_F 294 8/29/05, 11:21 AM
Phenotypic Analysis of Rat Placenta 295
21
Phenotypic Analysis of the Rat Placenta
Rupasri Ain, Toshihiro Konno, Lindsey N. Canham,

and Michael J. Soares
Summary
The rat is an important model for studying the biology of trophoblast-uterine development.
This chapter describes methods that are useful for the characterization of the rat uteroplacental
compartment.
Key Words: Rat; chorioallantoic placenta; choriovitelline placenta; junctional zone; laby-
rinth zone; metrial gland; spongiotrophoblast; trophoblast giant cells; trophoblast cell inva-
sion; prolactin gene family.
1. Introduction
In this chapter, we discuss the rat as an experimental model for investiga-
tions directed toward pregnancy and the uteroplacental interface. We present
an overview on the organization of the rat maternal–fetal interface and present
methods for studying the rat placenta and trophoblast cells.
1.1. Merits of the Rat as an Animal Model for Placental Research
Historically, the rat has been a valuable model for studying most aspects of
reproduction and, in many areas, still remains the preferred model system. The
rat is the dominant preclinical model system used by the pharmaceutical and
agro-chemical industries. Genetic approaches for analysis of the rat are rapidly
advancing. The first draft of the rat genome is completed (1) and significant
progress is being made in developing strategies for the genetic manipulation of
the rat, including transgenic (2–7), in vivo mutagenesis (8), and nuclear trans-
fer (9,10). Finally, panels of consomic rat strains have been established, which
will permit a systematic and physiologically relevant dissection of the rat
genome (11).
295
21_Ain_295_314_F 295 8/29/05, 11:22 AM

296 Ain et al.
The use of animal model systems provides an essential tool for dissecting
molecular mechanisms controlling cellular development. The maternal–fetal
interface is no exception. The premise of employing any animal model system
is that if the process being studied is fundamental it will likely demonstrate
conservation across species. Although there are some differences in the orga-
nization of the rodent vs the primate maternal–fetal interface, there are over-
riding similarities in the functions and lineages of cells comprising the
maternal–fetal interface. Among these are events transpiring during the last
week of gestation in the rat. During this latter phase of pregnancy in the rat,
trophoblast cells exit the chorioallantoic placenta and invade into the uterine
endometrium, where they establish intimate relationships with the uterine vas-
culature (12). These invasive events in the rat are remarkably similar to inva-
sive events occurring during the latter stages of the first trimester and
throughout the second trimester of pregnancy in the human (13).
If we can understand and appreciate biological processes at the maternal–
fetal interface in species that can be experimentally manipulated (e.g., rat,
mouse), then we can more intelligently study the development of the human
maternal–fetal interface and identify pivotal junctures of cellular control, fa-
cilitating diagnosis and therapeutic intervention.
In some instances, cross-species similarities may prevail, whereas in other
cases, the differences may be most compelling. Nonetheless, our appreciation
for the biology of pregnancy increases. Animal models provide us with a means
of studying biological phenomena at levels that are not ethically permissible
with humans. Viviparity is vital to the success of our species. Mechanisms that
underlie viviparity will exhibit some level of conservation.
1.2. Overview of the Organization of the Uteroplacental Compartment
of the Rat
The uteroplacental compartment of the rat is similar to the mouse and pos-
sesses similarities and differences to the organization of the uteroplacental
compartment of other species with hemochorial placentation (14–17). This has
led to the utilization of an assortment of terms to describe components of the
uteroplacental compartment. Schematic representations of the rat uteroplacental
compartment are presented in Figs. 1 and 2.
The site where blood enters the uterus determines the orientation of the
uteroplacental compartment. This region is referred to as the mesometrial com-
partment, and the opposite end is termed the antimesometrial compartment.
The uterine mesometrial compartment is prominently comprised of stromal
cells, blood vessels (endothelial cells, smooth muscle cells), immune/inflam-
matory cells (natural killer cells, macrophages), smooth muscle cells of the
myometrium, and trophoblast cells. Cellular composition of this compartment
21_Ain_295_314_F 296 8/29/05, 11:22 AM

Fig. 1 (see companion CD for color version). Hematoxylin and eosin-stained tissue
section of the midgestation rat uteroplacental compartment (left panel, day 11 of ges-
tation) and a corresponding schematic diagram (right panel).
is dynamic and has gestation-stage dependent and species-specific characteris-

tics. Following implantation, natural killer cells expand in number, and infil-
trate the mesometrial decidua, located adjacent to the developing
chorioallantoic placenta. Decidual cells are derived from uterine stromal cells
and exhibit functional differences depending upon their location (18,19). Me-
sometrial decidua is the site of extensive vascular remodeling, whereas
antimesometrial decidual cells are conspicuous in their elaboration of
cytokines, including members of the prolactin (PRL) family (20). A triangle-
shaped area rich in blood vessels is situated between the mesometrial decidua
and the surface of the uterus (21,22). This region has been referred to by the
terms “mesometrial triangle,” “metrial gland,” and several others (23). As ges-
tation advances, extraembryonic and embryonic structures expand in size and
the decidua thins. Accompanying these events, natural killer cells vacate the
mesometrial decidua and infiltrate the mesometrial triangle where they associ-
ate with the resident vasculature. Subsequently, the antimesometrial deciduum
and mesometrial-associated natural killer cells degenerate. As natural killer
cells depart, a specialized population of trophoblast cells exits the chorioallan-
toic placenta and invades into the mesometrial decidua (12). In the mouse,
trophoblast invasion is limited to the mesometrial decidua, whereas in the rat,
trophoblast cells penetrate through the mesometrial decidua and infiltrate the
mesometrial triangle.
21_Ain_295_314_F 297 8/29/05, 11:22 AM

298
21_Ain_295_314_F
298
298
8/29/05, 11:22 AM
Fig. 2 (see companion CD for color version). Hematoxylin and eosin-stained tissue section of the late gestation
rat uteroplacental compartment (left panel, day 18 of gestation) and a corresponding schematic diagram (right
Ain et al.
panel) with highlighted expanded views of the labyrinth and junctional zones (lower panels).
The chorioallantoic placenta is situated at the base of the mesometrial com-

partment. It develops from trophoblast stem cells present in the ectoplacental
cone and generates two recognizable structures: (a) labyrinth zone and (b) junc-
tional zone. The labyrinth zone arises from the interaction of allantoic meso-
derm with the trophoblast stem cell population (24), yielding trophoblast cell
syncytialization and establishment of the barrier for maternal–fetal exchange
(17). Once the barrier is established, endocrinologically active trophoblast
giant cells appear within the labyrinth zone. Four trophoblast cell lineages
differentiate from trophoblast stem cells within the junctional zone: (a) tro-
phoblast giant cells, (b) spongiotrophoblast cells, (c) glycogen cells, and (d)
invasive trophoblast cells. Trophoblast giant cells are the first trophoblast cell
lineage to develop and, until the last week of gestation, are the most distally
located trophoblast cell types within the uterus. Spongiotrophoblast cells are
the main constituents of the junctional zone. Trophoblast giant cells,
spongiotrophoblast cells, and invasive trophoblast cells are the major endo-
crine cells of the rat and mouse placenta. Glycogen cells appear during the last
week of pregnancy, notably accumulate glycogen, and disappear before the
end of pregnancy. The invasive trophoblast cell population first appears at mid-
gestation. It consists of trophoblast cells that penetrate and surround the uter-
ine vasculature present within the developing chorioallantoic placenta. As
gestation advances, invasive trophoblast cells exit the junctional zone and
enter the mesometrial compartment. Three terms have been used to describe
the invasive trophoblast cells (25). Endovascular trophoblast cells replace the
endothelium, intramural trophoblast cells are embedded within the vascular
wall, and interstitial trophoblast cells are situated between the vasculature.
1.3. Investigation of the Rat Placenta
In this chapter, we describe methods for (1) mating and gestational staging;
(2) detection of pregnancy during early postimplantation stages; (3) dissection
of the midgestation uteroplacental compartment; (4) dissection of the chorioal-
lantoic placenta; (5) establishment of spongiotrophoblast cell primary cultures;
(6) isolation of the metrial gland; (7) chemical induction of fetal death; and (8)
monitoring the phenotype of cells within the uteroplacental compartment. The
outlined protocols are based on our experience in working with the rat placenta
over the past two decades.
2. Materials
2.1. Mating and Gestational Staging
1. Holtzman rats are obtained from Harlan Sprague-Dawley (Indianapolis, IN).
2. Saline solution (0.9% NaCl).
3. Glass slide with wells.
4. Microscope (×40–100 magnification).
21_Ain_295_314_F 299 8/29/05, 11:22 AM

300 Ain et al.
2.2. Detection of Pregnancy During Early Post Implantation Stages

Chicago Blue B (1% solution; Matheson Coleman & Bell Manufacturing
Chemists, Norwood, OH, cat. no. CX685 B364).
2.3. Mid-Gestation Placental Dissection
1. Dissecting microscope (×10–20 magnification).
2. Hank’s balanced salt solution (HBSS; Sigma Chemical Company, St. Louis, MO,
cat. no. H-387).
3. Fine forceps and microdissecting spring scissors (Roboz Surgical Instrument Co.,
Gaithersburg, MD, cat. nos. RS-5155 and RS-5602, respectively).
2.4. Chorioallantoic Placental Dissection

1. Dissecting microscope (×10–20 magnification).
2. HBSS (Sigma).
3. Fine forceps and microdissecting spring scissors (Roboz).
4. 23-gauge needles (BD Biosciences, Franklin Lakes, NJ, cat. no. 305134).
2.5. Primary Culture of Spongiotrophoblast Cells

2. Dispase II (Roche Diagnostic Corporation Indianapolis, IN, cat. no. 295825).
3. DNase I (Sigma, cat. no. D4263).
4. HBSS (Sigma).
5. Dulbecco’s modified Eagle’s medium (DMEM) culture medium (Mediatech
Cellgro, Herdon, VA, cat. no. 10-017-CV) supplemented with penicillin and
streptomycin (Mediatech Cellgro) and 10% fetal bovine serum (FBS; Altanta
Biologicals, Norcross, GA , cat. no. S11150).
6. Nylon mesh cell strainers (70 µm; BD Biosciences, cat. no. 352350).
7. Percoll (Amersham Biosciences, Uppsala, Sweden, cat. no. 17-089-02).
2.6. Metrial Gland Isolation

2. HBSS (Sigma).
2.7. Chemical Induction of Fetal Death

Digoxin (Elkin-Sinn, Cherry Hill, NJ).
2.8. Monitoring the Phenotype of Cells Within the Uteroplacental
Compartment
1. TRIzol reagent (Invitrogen Life Technologies, Carlsbad, CA, cat. no. 15596-018).
2. 1% Formaldehyde-agarose gels. Formaldehyde (Fisher Scientific, Pittsburgh, PA,
cat. no. F79-4); agarose (Sigma, cat. no. A-9539).
3. Nylon membranes (Nytran Super Charge, Schleicher & Schuell Biosciences, Inc.,
Keene, NH, cat. no. 10416296).
21_Ain_295_314_F 300 8/29/05, 11:22 AM

4. Crosslinker (Model XL-1000, Spectronics Corporation, Westbury, NY).

5. [α-P32]dCTP (Perkin Elmer, Boston, MA, cat. no. Blu/NEG/0134).
6. Plasmids containing cDNAs for the transcripts of interest (Table 1).
2.8.1. PRL Family DNA Miniarray (32) (see Note 12)

1. Polymerase chain reaction (PCR)-amplified cDNA for each of the members of
PRL family is spotted, in duplicate, onto nylon membranes (Schleicher & Schuell
Biosciences, Inc.), crosslinked, and stored at 4°C until used.
2. TRIzol reagent (Invitrogen).
3. [αP32]dCTP (Perkin Elmer, Boston, MA).
4. Micro bio-spin columns (Bio-Rad Laboratories, Hercules, CA, cat. no. 732-
6223).
5. Denhardt’s reagent (50X Denhard’s reagent): 1% ficoll (Sigma, cat. no. F-4375),
1% polyvinylpyrrolidone (Sigma, cat. no. P-5288), 1% bovine serum albumin
(BSA; Fraction V, Fisher, cat. no. BP1600-100) diluted in distilled water.
6. Deionized formamide (Sigma, cat. no. F9037).
7. Salmon sperm DNA (Invitrogen, cat. no. 15632-011).
8. SSPE buffer (20X SSPE): add 175 g sodium chloride, 27.6 g sodium phosphate
monobasic, 7.4 g ethylenediamine tetraacetic acid (EDTA) to 1 L of distilled
water.
9. 0.1% sodium dodecyl sulfate (SDS; Fisher, cat. no. BP 166-500).
10. Kodak Bio-Max film (Kodak, Rochester, NY, cat. no. 829-4985).
2.8.2. In Situ Hybridization (33,34) (see Note 13)

1. Plasmids containing cDNAs for the transcript of interest are used as templates to
synthesize sense and anti-sense digoxigenin-labeled riboprobes (Table 1).
2. Dry-ice-cooled heptane (Fisher, cat. no. 03008-1).
3. Cryostat (Model No. CM 1850-3-1, Lieca Microsystems, Germany).
4. 4% paraformaldehyde (P-6148, Sigma) in phosphate-buffered saline (PBS) at 4°C.
5. Deionized formamide (Sigma).
6. 1X Denhardt’s reagent.
7. 10% Dextran sulfate (Fisher, cat. no. BP 1585-100).
8. Salmon sperm DNA (Invitrogen).
9. Standard saline citrate (SSC; 20X SSC): 175.3 g sodium chloride, 88.2 g sodium
citrate in 1 L of distilled water.
10. RNase-A (Sigma, cat. no. R-6513).
11. Dig-High Prime DNA Labeling and Detection Starter Kit II (Roche Diagnostic
Corporation, cat. no. 1585614).
2.9. Monitoring Protein Expression

2.9.1. Western Blotting
1. Reagents for polyacrylamide gel electrophoresis and electrophoretic transfer
(BioRad Laboratories).
21_Ain_295_314_F 301 8/29/05, 11:22 AM

302 Ain et al.
Table 1
cDNAs Used in the Phenotypic Analysis of Cells Within the Uteroplacental Com-
partment
GenBank
Cell type Gene Gestation stage accession no. Reference
Trophoblast
Ectoplacental cone PLF-RP Early NM_053364 37
Trophoblast giant cell PL-I Early to mid D21103 31
PL-II Mid to late M13749 31,38
PLP-Fα Early to mid NM_022530 37,
unpublisheda
P450scc Early to late J05156 39,40
P450c17 Mid to late NM_012753 40,41
3βHSD Early to late L17138 Unpublishedb
Spongiotrophoblast PLP-B Mid to late M31155 42,43
PLP-Fβ Mid to late AY741310 Unpublisheda
SSP Mid to late NM_172073 44
Labyrinthine TGC PL-II Mid to late M13749 42,45
PLP-K Mid to late NM_138861 32
PLF-RP Mid to late NM_053364 37
Syncytial trophoblast FABP3 Mid to late NM_024162 46
Alk Phos Mid to late NM_013059 47
Invasive trophoblast PLP-L Mid to late NM_138527 12
PLP-M Mid to late NM_053791 12
PLP-N Mid to late NM_153738 48
IGF-II Mid to late X17012 49,
unpublishedc
Decidual cells
Mesometrial α2-MG Early to mid NM_012488 50
Antimesometrial dPRP Early to mid NM_022846 51,52
PLP-J Early to mid NM_031316 32
Natural killer cells Osteopontin Mid NM_012881 Unpublishedd
Abbreviations: TGC, Trophoblast giant cell; PLF-RP, proliferin-related protein; PL, placental lacto-
gen, PLP, prolactin-like protein; P450scc, side chain cleavage; P450c17, 17α hydroxylase; 3βHSD, 3β
hydroxysteroid dehydrogenase; SSP, spongiotrophoblast-specific protein; FABP3, fatty acid binding
protein-3; IGF-II, insulin-like growth factor-II; α2-MG, α2-macroglobulin; dPRP, decidual prolactin-
related protein.
aHo-Chen, J., Bustamante, J. J., and Soares, M. J., unpublished results.
bCanham, L. N. and Soares, M. J., unpublished results.
cAin, R. and Soares, M.J., unpublished results.
dLiu, B. and Soares, M.J., unpublished results.
21_Ain_295_314_F 302 8/29/05, 11:22 AM

Table 2
Antibodies Used in the Phenotypic Analysis of Cells Within the Uteroplacental
Compartment
Cell type Antigen Gestation stage Source (Cat. No.) Reference
Trophoblast Cytokeratin All stages Sigma Chemical Co., 12

St. Louis, MO (C2931)
Trophoblast giant cell PL-I Early to mid Chemicon International, 31,53
Temecula, CA (AB1288)
PL-II Mid to late Chemicon (AB1289) 31,38,54
PLP-A Mid to late Chemicon (AB1290) 55,56
P450scc Early to late Chemicon (AB1244, 39,57
AB1294)
P450c17 Mid to late see refs. 40,58
Spongiotrophoblast PLP-B Mid to late Chemicon (AB1291) 43
SSP Mid to late see ref. 44
Labyrinthine TGC PL-II Mid to late Chemicon (AB1289) 38
Syncytial trophoblast FABP3 Mid to late see ref. 46
Alk Phos Mid to late see ref. 47
Decidual cells
Mesometrial α2-MG Early to mid see ref. 50
Antimesometrial dPRP Early to mid Chemicon (AB1293) 52,59
PLP-J Early to mid see ref. Unpublisheda
Natural killer cells Perforin Mid Torrey Pines Biolabs, 12
Houston, TX (TP251)
Macrophages ED-1/ED-2 All stages Serotec Inc., Raleigh, NC Unpublishedb
(MCA341R/ MCA342R)
Smooth muscle cells Sm actin All stages Sigma (A2547) Unpublishedb
Abbreviations: TGC, trophoblast giant cells; PL, placental lactogen, PLP, prolactin-like protein; P450scc,
side chain cleavage; P450c17, 17α hydroxylase; SSP, spongiotrophoblast-specific protein; FABP3, fatty acid
binding protein-3; β2-MG, α2-macroglobulin; dPRP, decidual prolactin-related protein; Sm, smooth muscle.
aAlam, S. M. K., Konno, T., and Soares, M. J., unpublished results.
bAin, R. and Soares, M. J., unpublished results.
2. Nitrocellulose (Optitran, Schleicher & Schuell Biosciences, Inc., Cat. No. BA-S 85).
3. Antibodies to the protein of interest (Table 2).
2.9.2. Immunocytochemistry
1. Dry-ice-cooled heptane (Fisher).
2. Cryostat (Leica).
3. Antibodies to the antigen(s) of interest (Table 2).
21_Ain_295_314_F 303 8/29/05, 11:22 AM

304 Ain et al.
3. Methods
3.1. Mating and Gestational Staging
1. The rats are maintained on a 14 h light:10 h dark lighting schedule with lights on
at 0600 h (see Note 1).
2. Adult males, preferably older than 10 wk of age, are placed one per cage.
3. Adult females, generally 7–10 wk of age, are transferred to a cage with a male
(no more than two females per male). The fur on one of the females is generally
marked with a dye to distinguish it from the other female.
4. Every morning between 0800 and 0900 h, each female cohabiting a cage with a
male, is removed from the cage for the purpose of obtaining a vaginal lavage.
5. A few hundred microliters of saline are delivered with a pipette into the vagina of
the female and recovered with the same pipet.
6. The contents of each saline lavage are transferred to a well within a multi-well
glass plate.
7. After all of the vaginal lavages are collected, they are examined with the aid of a
microscope (×40–100 magnification).
8. The presence of sperm in the lavage is recorded, as is the cellular content of the
lavage (see Notes 2 and 3).
9. The sperm positive females are transferred to separate cages. The presence of
sperm in the vaginal lavage is considered day 0 of pregnancy (see Note 4).
3.2. Detection of Pregnancy During Early Post Implantation Stages (26)
(see Note 5)
1. Pregnancy detection within the first 48 h post implantation requires intravenous
injection of a vital blue dye, such as Chicago Blue B.
2. A volume of 0.25–0.5 mL/100 g body weight of a 1.0% solution of Chicago Blue
B can be injected into the tail vein of the rat.
3. Implantation reactions are identified by the accumulation of blue bands within
the uterus after 15 min.
3.3. Mid-Gestation Placental Dissection (27) (see Note 6)
1. Embryos with their encapsulating decidual tissues (conceptuses) are retrieved
from the uterus from days 10–13 of gestation.
2. Conceptuses are dissected with the aid of a dissecting microscope (×10–20 mag-
nification).
3. Tissues are collected and washed with HBSS.
4. The overlying decidua basalis and decidua capsularis are removed with fine for-
ceps and gentle dissection.
5. A cut through the mural pole of the trophoblast layer is made and the trophoblast
retracted. Be careful not to cut through the yolk sac/amnion.
6. The visceral yolk, amnion, and embryo are separated from the developing chorio-
allantoic placenta by cutting at the insertion site of the allantois with microdis-
section spring scissors.
7. The entire trophoblast component is flattened (allantoic insertion site is up) and
can be further separated into chorioallantoic and choriovitelline layers with the
21_Ain_295_314_F 304 8/29/05, 11:22 AM

aid of microdissection spring scissors. The inner dark circle of tissue (more vas-
cular) comprising the chorioallantoic tissue is cut away from the lighter surround-
ing tissue (less vascular) consisting of the choriovitelline tissue.
8. Dissected decidua basalis, decidua capsularis, and trophoblast components can
each be processed as required, and/or stored at –80°C, until further use.
3.4. Chorioallantoic Placental Dissection (28) (see Note 7)

1. Embryos with their encapsulating decidual tissues (conceptuses) can be retrieved
from the uterus on days 13 to 21 of gestation.
2. Conceptuses are dissected with the aid of a dissecting microscope (×10–20 mag-
nification).
3. The tissues are collected into and washed with HBSS.
4. The overlying decidual basalis tissue and underlying yolk sac/umbilical insertion
are removed with fine forceps and microdissection spring scissors.
5. The junctional zone is identified by its pale appearance, due to the absence of
fetal blood, and separated from the labyrinth zone, a richly vascularized tissue,
with fine forceps and 23-gauge needles.
6. Recovered tissues are rinsed in HBSS, processed as required, and/or stored at
–80°C, until further use.
3.5. Primary Culture of Spongiotrophoblast Cells (29) (see Note 8)

1. Junctional zones from day-13 rat chorioallantoic placentas are dissected under
sterile conditions (see Subheading 3.4.).
2. Tissues are cut into small pieces with microdissection spring scissors and disso-
ciated with Dispase II (4.8 mg/mL) and DNase I for 1 h at 37°C with continuous
shaking.
3. At the end of the digestion, the suspension of cells and tissue fragments are mixed
several times with the aid of a Pasteur pipet, and centrifuged.
4. The harvested cells are then resuspended in DMEM culture medium supple-
mented with 10% FBS and filtered through a nylon mesh (70 µm).
5. The cell suspension is then centrifuged through a 60% cushion of Percoll for
15 min at room temperature.
6. Cells at the interface are collected, washed with DMEM supplemented with 10%
FBS, and cultured in the same medium.
7. The cells can be maintained in DMEM medium containing FBS (1–10%) for
7–10 d.
3.6. Metrial Gland Isolation (see Note 9)

The metrial gland can be isolated from midgestation onward (Fig. 1).
1. The uterus is removed from the female rat.
2. Fat and mesenteries are removed from the uterus.
3. Microdissection spring scissors are used to cut each uterine horn along its
antimesometrial surface.
21_Ain_295_314_F 305 8/29/05, 11:22 AM

306 Ain et al.
4. The decidual–placental–embryo units are removed by gentle dissection from the

opened uterus with fine curved forceps.
5. The attachment site of the mesometrial decidua with the uterine wall can be readily
identified by its pale appearance and defines the limits of the metrial gland.
6. The wall of the uterus associated with the mesometrial attachment site is removed
with microdissection spring scissors and processed as required.
3.7. Chemical Induction of Fetal Death (30) (see Note 10)

1. Pregnant female rats (days 10–13 of gestation) are anesthetized.
2. A midline abdominal incision is made to expose the uterine horns.
3. Digoxin, a cardiac glycoside, is injected into each amniotic cavity (12.5 µg/
50 µL in a saline vehicle) through the antimesometrial aspect of the uterus.
4. The abdominal muscle is sutured and the skin is secured with wound clips.
3.8. Monitoring the Phenotype of Cells Within the Uteroplacental

Compartment
Several methods can be used to assess the phenotype of the rat uteroplacental
compartment, including assessment of mRNA and protein expression.
3.8.1. Monitoring mRNA Expression
We routinely use northern blotting, the PRL family DNA miniarray, and in
situ hybridization for monitoring the phenotypes of cells within the rat
uteroplacental compartment.
3.8.1.1. NORTHERN BLOTTING (31)
1. Total RNA is extracted from tissues using TRIzol reagent, resolved in 1% form-
aldehyde-agarose gels, transferred to nylon membranes, and crosslinked.
2. Blots are probed with α-32P-labeled cDNAs.
3. Glyceraldehyde-3'-phosphate dehydrogenase (G3PDH) cDNA is used to evalu-
ate the integrity and equal loading of RNA samples (see Note 11).
3.8.1.2. PRL FAMILY DNA MINIARRAY (32) (SEE NOTE 12)

1. Twenty ng of PCR-amplified cDNA for each of the members of PRL family is
spotted, in duplicate, onto nylon membranes, crosslinked, and stored at 4°C until
used.
2. Total RNA is extracted from tissues using TRIzol reagent.
3. [α-32P]dCTP labeled cDNA probes are generated by reverse transcription using
5 µg of total RNA.
4. Probes are purified using micro bio-spin columns.
5. Membrane filters are briefly rinsed with water and prehybridized for 2 h at 42°C
with 5X SSPE containing 5X Denhardt’s reagent, 50% deionized formamide, 1%
SDS, and salmon sperm DNA (100 µg/mL).
6. Hybridizations are performed overnight with the labeled probes at 42°C.
21_Ain_295_314_F 306 8/29/05, 11:22 AM

7. Membranes are washed once with 2X SSPE and 0.1% SDS for 30 min at 42°C
and twice with 0.1X SSPE and 0.5% SDS at 60°C for 30 min each.
8. Membranes are then wrapped with plastic wrap and exposed to Kodak Bio-Max
film for 1–4 h and developed.
3.8.1.3. IN SITU HYBRIDIZATION (33,34) (SEE NOTE 13)

1. Ten-micron cryosections of tissues are prepared and stored at –80°C until used.
2. Plasmids containing cDNAs for the transcript of interest are used as templates to
synthesize sense and anti-sense digoxigenin-labeled riboprobes.
3. Tissue sections are air dried and fixed in ice cold 4% paraformaldehyde in PBS
for 15 min.
4. Prehybridization is carried out in a humidified chamber at 50°C in 5X SSC, 50%
deionized formamide, 1X Denhardt’s reagent, 10% dextran sulfate and salmon
sperm DNA (100 µg/mL).
5. Hybridizations are performed in the same incubation conditions overnight.
6. Slides are washed in 2X SSC at room temperature for 30 min followed by treat-
ment with RNase-A (100 ng/mL) and additional washes with 2X SSC for 30
min at room temperature, with 2X SSC for 1 h at 65°C, with 0.1X SSC for 1 h
at 65°C.
7. Tissue samples are then blocked for 30 min and incubated with alkaline phos-
phatase-conjugated anti-digoxigenin antibody (1:500) in blocking buffer for 2 h
at room temperature.
8. Slides are then washed and detection is performed using nitro blue tetrazolium
(250 µg/mL) and 5-bromo-4-chloro-3-indolyl-phosphate (225 µg/mL).
3.8.2. Monitoring Protein Expression

Antibodies can be used to monitor specific protein expression by Western
blotting or used to localize the expression of a specific protein/antigen to a cell
type within the uteroplacental compartment. Table 2 contains a list of antibod-
ies useful in the analysis of the rat uteroplacental compartment.
1. Western blotting. Uteroplacental tissues are isolated as described above and ex-
tracted consistent with the protein being investigated. Extracts are separated by
polyacrylamide gel electrophoresis, transferred to nitrocellulose, and probed with
antibodies specific to the protein of interest.
2. Immunocytochemistry. The intact uteroplacental unit is isolated, frozen in dry-
ice-cooled heptane, and stored at –80°C. Cryosections are prepared at 8–10 µm,
adhered to glass slides, and probed with antibodies specific to the antigen of
interest (see Note 14).
4. Notes
1. The Holtzman rat is an outbred strain closely related to the Sprague-Dawley rat.
The rats are easy to handle. Under the 14 h light:10 h dark lighting schedule the
female rats tend to have a 5-d estrous cycle. Our approach for mating and the
21_Ain_295_314_F 307 8/29/05, 11:22 AM

308 Ain et al.
pregnancy dating system also applies to other strains. Males used for breeding
are obtained at 10 wk of age and usually continue to be effective breeders until
approx 1 yr of age. Although, most vendors provide timed-pregnant female rats,
we have not found their dating of pregnancies to be reliable and prefer to gener-
ate our own timed pregnancies.
2. The process is repeated daily until mating is confirmed by the presence of sperm.
Occasionally, seminal plugs are present in the vagina. Unlike the mouse, seminal
plugs tend to fall out of the rat vagina. Special suspended caging, if permitted,
can be used and dark paper placed in the trays beneath the cages each evening
and checked the following morning for the presence of the whitish-yellow,
opaque seminal plugs. This is generally the procedure used for generating
pseudopregnant rats with vasectomized males.
3. Inspection of the cellular content is helpful. Based on the cellular composition of
the vaginal lavage it is possible to determine whether the animal is cycling. Cells
present in the vaginal lavage are impacted by circulating concentrations of the
ovarian steroid hormones, estrogen and progesterone. Vaginal lavage’s contain-
ing nucleated and/or nucleated with cornified cells characterize the estrous cycle
stage coinciding with behavioral estrus and receptivity. Cornified cells are char-
acteristic at the time of ovulation and lavages containing leukocytes are domi-
nant post-ovulation during the luteal phase. A cycling female cohabiting a cage
with a male would raise the question of the male’s fertility.
4. It is always important to determine the pregnancy dating system used when re-
viewing any experimentation with pregnant rats. For us, classifying the presence
of vaginal sperm as day 0 of pregnancy is historical. Others consider the presence
of sperm in the vagina as day 1 of pregnancy. It would likely be most accurate to
consider 1200 h on the day sperm is found to be day 0.5 of pregnancy. It is criti-
cal to appreciate that trophoblast/placental development can differ markedly
within a day. Thus, regardless, of the pregnancy dating system used, it is impera-
tive to provide the information in any scientific report forthcoming from the re-
search. De Rijk and colleagues have published a useful guide for assessing placental
morphology and pregnancy-dependent maternal hematological indices (35).
5. The detection of early postimplantation pregnancy by intravenous injection of a
vital dye is based on capillary permeability changes. The vital dyes bind to circu-
lating proteins such as albumin. As the capillary permeability increases after im-
plantation, the circulating dye-bound albumin exits the capillaries and
concentrates in the surrounding tissue.
6. The dissection of the midgestation placenta requires practice. In order to main-
tain the appropriate orientation of maternal, extraembryonic, and embryonic tis-
sues, it is best not to disrupt the amniotic contents. It is also essential to identify
mesometrial vs antimesometrial poles. The mesometrial decidua is thicker than
the antimesometrial decidua. The shape of the mid-gestation conceptus can be
compared to an ice-cream cone (more evident on days 10–11 of gestation). Using
such an analogy, the mesometrial pole would be the location of the ice cream. We
21_Ain_295_314_F 308 8/29/05, 11:22 AM

have used chorioallantoic and choriovitelline to discriminate between the polar

and mural trophoblast components of the midgestation conceptus. Some have
argued that the relationship between the mural trophoblast and the yolk sac does
not constitute a true choriovitelline or yolk sac placenta (36).
7. Prior to day 13 of gestation, it is difficult to separate the junctional and labyrinth
zones. After day 15, it is difficult to completely remove the thinning decidua
basalis, which becomes firmly adherent to the junctional zone. Beginning on day
15 of gestation, it also becomes possible to dissect the chorioallantoic zones with
fine forceps; needles are not required. Separation of junctional and labyrinth
zones is not readily achieved in the mouse and hamster as a result of extensive
interdigitation of the zones. It is important to appreciate that even in the rat, the
dissection of the junctional and labyrinth zones is not perfect and usually, a small
amount of residual contamination is evident.
8. Dissection of the junctional zone from tissue obtained earlier than day 13 of ges-
tation is more time-consuming and yields a nominal amount of starting tissue for
the enzymatic dissociation. Junctional zones from later in gestation are easy to
dissect but do not yield a good starting cell population. In a limited series of
experiments, we have found that the establishment of junctional zone cultures
from placentas obtained later during gestation is difficult and the cells show poor
viability. The rat day-13 junctional zone cell cultures show no evidence of prolif-
eration of any cell type. These cultures also show modest contamination of
vimentin-positive cells (2–3%). The rat day-13 junctional zone cells in primary
culture spontaneously differentiate as demonstrated by their expression of members
of the placental PRL gene family (29). Most of the cells exhibit a spongiotrophoblast
cell phenotype; however, some trophoblast giant cells and syncytial trophoblast cells
are also detected.
9. The metrial gland is a heterogeneous tissue. Its landmarks are not well defined.
Thus consistency in dissection is a necessity.
10. Digoxin is an effective tool for inducing fetal death at midgestation but does not
work well as gestation advances. We have had some success using intra-amniotic
injections of potassium chloride to kill fetuses from later in gestation.
11. Interpretation of RNA measurements in the uterus or placenta is directly depen-
dent upon the quality of tissue dissection. Simply labeling a sample containing
some placental tissue as placenta does not make it placenta. Most placental
samples contain varying amounts of decidual tissue and yolk sac. We have uti-
lized an assortment of different housekeeping genes to monitor RNA integrity
and loading efficiency. These have included β-actin, G3PDH, β-tubulin, and 28S
ribosomal RNA.
12. The PRL family miniarray assay represents an effective screening tool for
monitoring trophoblast cell development. Information can be retrieved regard-
ing decidual and trophoblast cell lineages and temporal aspects of differentia-
tion. The assay is most effective when complementary procedures are employed
for monitoring changes in gene expression.
21_Ain_295_314_F 309 8/29/05, 11:22 AM

310 Ain et al.
13. In our hands, in situ hybridization is a reliable method for identifying cell types
contributing to the expression of a specific gene within the uteroplacental com-
partment. An appreciation of the dynamic changes in uteroplacental morphology
is essential to maximize the benefit of the approach.
14. Antibodies are effective tools for identifying and localizing proteins. Each antibody–
antigen interaction needs to be optimized for the specific technique employed.
Acknowledgments
We would like to thank past and current members of our laboratory for their
efforts in developing and characterizing the methods described in this chapter.
This work was supported by grants from the National Institutes of Health (NIH)
(HD20676, HD39878, HD48861) and the Hall Family Foundation.
References
1. Rat Genome Sequencing Project Consortium. (2004) Genome sequence of the
Brown Norway rat yields insights into mammalian evolution. Nature 428, 493–521.
2. Hammer, R. E., Maika, S. D., Richardson, J. A., Tang, J. P., and Taurog, J. D.
(1990) Spontaneous inflammatory disease in transgenic rats expressing HLA-B27
and human beta 2m: an animal model of HLA-B27-associated human disorders.
Cell 63, 1099–1112.
3. Mullins, J. J., Peters, J., and Ganten, D. (1990) Fulminant hypertension in
transgenic rats harbouring the mouse Ren-2 gene. Nature 344, 541–544.
4. Hamra, F. K., Gatlin, J., Chapman, K. M., et al. (2002) Production of transgenic
rats by lentiviral transduction of male germ-line stem cells. Proc. Natl. Acad. Sci.
USA 99, 14,931–14,936.
5. Hasuwa, H., Kaseda, K., Einarsdottir, T., and Okabe, M. (2002) Small interfering
RNA and gene silencing in transgenic mice and rats. FEBS Lett. 532, 227–230.
6. Lois, C., Hong, E. J., Pease, S., Brown, E. J., and Baltimore, D. (2002) Germline
transmission and tissue-specific expression of transgenes delivered by lentiviral
vectors. Science 295, 868–872.
7. Orwig, K. E., Avarbock, M. R., and Brinster, R. L. (2002) Retrovirus-mediated
modification of male germline stem cells in rats. Biol. Reprod. 67, 874–879.
8. Zan, Y., Haag, J. D., Chen, K.-S., et al. (2003) Production of knockout rats using
ENU mutagenesis and a yeast-based screening assay. Nat. Biotechnol. 21, 645–651.
9. Mullins, L. J., Wilmut, I., and Mullins, J. J. (2004) Nuclear transfer in rodents. J.
Physiol. 554, 4–12.
10. Zhou, Q., Renard, J.-P., Le Friec, G., et al. (2003) Generation of fertile cloned rats
by regulating oocyte activation. Science 302, 1179.
11. Cowley, A. W., Roman, R. J., and Jacob, H. J. (2004) Application of chromo-
somal substitution techniques in gene-function discovery. J. Physiol. 554, 46–55.
12. Ain, R., Canham, L. N., and Soares, M. J. (2003) Gestational stage-dependent
intrauterine trophoblast cell invasion in the rat and mouse: novel endocrine phe-
notype and regulation. Dev. Biol. 260, 176–190.
21_Ain_295_314_F 310 8/29/05, 11:22 AM

13. Pijnenborg, R., Robertson, W. B., Brosens, I., and Dixon, G. (1981) Trophoblast
invasion and the establishment of haemochorial placentation in man and labora-
tory animals. Placenta 2, 71–92.
14. Bridgman, J. (1949) A morphological study of the development of the placenta of
the rat. II. An histological and cytological study of the development of the chorio-
allantoic placenta of the white rat. J. Morphol. 83, 195–224.
15. Davies, J. and Glasser, S. R. (1968) Histological and fine structural observations
on the placenta of the rat. Acta Anat. 69, 542–608.
developmental anatomy of the murine and human definitive placentae. Placenta
23, 3–19.
17. Rossant, J. and Cross, J. C. (2002) Extraembryonic lineages, in Mouse Develop-
ment (Rossant, J. and Tam, P. P. L., eds.). Academic, New York: pp. 155–180.
18. De Feo, V. J. (1967) Decidualization, in Biology of the Uterus (Wynn, R. M., ed.).
Appleton-Century-Crofts, New York: pp. 191–290.
19. Bell, S. C. (1983). Decidualization: regional differentiation and associated func-
tion. Oxford Rev. Reprod. Biol. 5, 220–271.
20. Orwig, K. E., Rasmussen, C. A., and Soares, M. J. (1997) Decidual signals in the
establishment of pregnancy: the prolactin family. Trophoblast Res. 10, 329–343.
21. Selye, H. and McKeown, T. (1935) Studies on the physiology of the maternal
placenta in the rat. Proc. R. Soc. Lond. Biol. 119, 1–31.
22. Peel, S. (1989) Granulated metrial gland cells. Adv. Anat. Embryol. Cell Biol.
115, 1–112.
23. Ain, R. and Soares, M. J. (2004) Is the metrial gland really a gland? J. Reprod.
Immunol. 61, 129–131.
24. Downs, K. M. (2002) Early placental ontogeny in the mouse. Placenta 23, 116–131.
25. Kaufmann, P., Black, S., and Huppertz, B. (2003) Endovascular trophoblast inva-
sion: implications for the pathogenesis of intrauterine growth retardation and
preeclampsia. Biol. Reprod. 69, 1–7.
26. Psychoyos, A. (1971) Methods for studying changes in capillary permeability of
the rat endometrium, in Methods in Mammalian Embryology (Daniel, J. C., ed.).
W.H. Freeman, San Francisco: pp. 334–338.
27. Soares, M. J., Julian, J. A. and Glasser, S. R. (1985). Trophoblast giant cell re-
lease of placental lactogens: temporal and regional characteristics. Dev. Biol. 107,
520–526.
28. Soares, M. J. (1987). Developmental changes in the intraplacental distribution of
placental lactogen and alkaline phosphatase in the rat. J. Reprod. Fertil. 79, 93–98.
29. Lu, X.-J., Deb, S., and Soares, M. J. (1994) Spontaneous differentiation of tro-
phoblast cells along the spongiotrophoblast pathway: expression of members of
the placental prolactin gene family and modulation by retinoic acid. Dev. Biol.
163, 86–97.
30. Roby, K. F. and Soares, M. J. (1993) Trophoblast cell differentiation and organi-
zation: role of fetal and ovarian signals. Placenta 14, 529–545.
21_Ain_295_314_F 311 8/29/05, 11:22 AM

312 Ain et al.
31. Faria, T. N., Deb, S., Kwok, S. C. M., Talamantes, F., and Soares, M. J. (1990).
Ontogeny of placental lactogen-I and placental lactogen-II expression in the de-
veloping rat placenta. Dev. Biol. 141, 279–291.
32. Dai, G., Lu, L., Tang, S., Peal, M. J., and Soares, M. J. (2002) The prolactin
family miniarray: a tool for evaluating uteroplacental/trophoblast endocrine cell
phenotypes. Reproduction 124, 755–765.
33. Braissant, O. and Wahli, W. (1998) A simplified in situ hybridization protocol
using non-radioactively labeled probes to detect abundant and rare mRNAs on
tissue sections. Biochemica 1, 10–16.
34. Wiemers, D.O., Shao, L.-J., Ain, R., Dai, G., and Soares, M. J. (2003) The mouse
prolactin gen e family locus. Endocrinology 144, 313–325.
35. De Rijk, E. P. C. T., van Esch, E., and Flik, G. (2002) Pregnancy dating in the rat:
placental morphology and maternal blood parameters. Toxicol. Pathol. 30, 271–282.
36. Wooding, F. B. P. and Flint, A. P. F. (1994) Placentation, in Marshall’s Physiol-
ogy of Reproduction, Fourth Edition, Volume 3 Pregnancy and Lactation (Lam-
ming, G. E., ed.). Chapman & Hall, London: pp. 235–460.
37. Sahgal, N., Knipp, G. T., Liu, B., Chapman, B. M., Dai, G., and Soares, M. J.
(2000) Identification of two new nonclassical members of the rat prolactin family.
J. Mol. Endocrinol. 24, 95–108.
38. Campbell, W.J., Deb, S., Kwok, S. C. M., Joslin, J., and Soares, M. J. (1989).
Differential ex pression of placental lactogen-II and prolactin-like protein-A in
the rat chorioallantoic placenta. Endocrinology 125, 1565–1574.
39. Yamamoto, T., Roby, K. F., Kwok, S. C. M., and Soares, M. J. (1994) Transcrip-
tional activation of cytochrome P450 side chain cleavage enzyme expression dur-
ing trophoblast cell differentiation. J. Biol. Chem. 269, 6517–6523.
40. Durkee, T. J., McLean, M. P., Hales, D. B., et al. (1992) P450 (17α) and P450scc
gene expression and regulation in the rat placenta. Endocrinology 130, 1309–1317.
41. Yamamoto, T., Chapman, B.M., Johnson, D.C., Givens, C.R., Mellon, S.H., and
Soares, M.J. (1996) Cytochrome P450 17α-hydroxylase gene expression in dif-
ferentiating rat trophoblast cells. J. Endocrinol. 150, 161–168.
42. Duckworth, M. L., Schroedter, I. C., and Friesen, H. G. (1990) Cellular localiza-
tion of rat placental lactogen-II and rat prolactin-like proteins A and B by in situ
hybridization. Placenta 11, 143–155.
43. Cohick, C. B., Xu, L., and Soares, M. J. (1997) Prolactin-like protein-B: heterolo-
gous expression and characterization of placental and decidual species. J.
44. Iwatsuki, K. Shinozaki, M., Sun, W., Yagi, S., Tanaka, S., and Shiota, K. (2000)
A novel secretory protein produced by rat spongiotrophoblast. Biol. Reprod. 62,
1352–1359.
45. Deb, S., Faria, T. N., Roby, K. F., et al. (1991) Identification and characterization
of a new member of the placental prolactin family: placental lactogen-I variant. J.
Biol. Chem. 266, 1605–1610.
21_Ain_295_314_F 312 8/29/05, 11:22 AM

46. Knipp, G. T., Liu, B., Audus, K. L., Fujii, H., Ono, T., and Soares, M. J. (2000)
Fatty acid transport regulatory proteins in the developing rat placenta and in tro-
phoblast cell culture models. Placenta 21, 367–375.
47. Campbell, W. J., Larsen, D., Deb., S., Kwok, S. C. M., and Soares, M. J. (1991)
Expression of alkaline phosphatase in differentiated rat labyrinthine trophoblast
tissue. Placenta 12, 227–237.
48. Wiemers, D. O., Ain R, Ohboshi, S., and Soares, M. J. (2003) Migratory tropho-
blast cells express a newly identified member of the prolactin gene family. J.
49. Correia-da-Silva, G., Bell, S. C., Pringle, J. H., and Teixeira, N. (1999) Expres-
sion of mRNA encoding insulin-like growth factors I and II by uterine tissues and
placenta during pregnancy in the rat. Mol. Reprod. Dev. 53, 294–305.
50. Gu, Y., Jayatilak, P. G., Parmer, T. G., Gauldie, J., Fey, G. H., and Gibori, G.
(1992) Alpha 2-macroglobulin expression in the mesometrial decidua and its regu-
lation by decidual luteotropin and prolactin. Endocrinology 131, 1321–1328.
51. Roby, K. F., Deb, S., Gibori, G., et al (1993) Decidual prolactin related protein:
identification, molecular cloning and characterization. J. Biol. Chem. 268, 3136–
3142.
52. Rasmussen, C. A., Orwig, K. E., Vellucci, S., and Soares, M. J. (1997) Dual ex-
pression of prolactin-related protein in decidua and trophoblast tissues during
pregnancy. Biol. Reprod. 55, 647–654.
53. Hamlin, G. P., Lu, X.-J., Roby, K. F., and Soares, M. J. (1994) Recapitulation of
54. Deb, S., Hashizume, K., Boone, K., et al. (1989) Antipeptide antibodies reveal
structural and functional characteristics of rat placental lactogen-II. Mol. Cell.
55. Deb, S., Youngblood, T., Rawitch, A., and Soares, M. J. (1989) Placental prolac-
tin-like protein-A: identification and characterization of two major glycoprotein
species with antipeptide antibodies. J. Biol. Chem. 264, 14,348–14,353.
56. Deb, S., Hamlin, G. P., Kwok, S. C. M., and Soares, M. J. (1993) Heterologous
expression and characterization of prolactin-like protein-A: identification of se-
rum binding proteins. J. Biol. Chem. 268, 3298–3305.
57. Roby, K. F., Larsen, D., Deb, S., and Soares, M. J. (1991) Generation and charac-
terization of antipeptide antibodies to rat cytochrome P-450 side chain cleavage
enzyme. Mol. Cell. Endocrinol. 79, 13–20.
58. Johnson, D. C. (1992) Cellular localization and factors controlling rat placental
cytochrome P45017 alpha (CYP17): 17 alpha-hydroxylase/C17,20-lyase activity.
Biol. Reprod. 46, 30–38.
59. Rasmussen, C. A., Hashizume, K., Orwig, K. E., Xu, L., and Soares, M. J. (1996)
Decidual prolactin-related protein: heterologous expression and characterization.
Endocrinology 137, 5558–5566.
21_Ain_295_314_F 313 8/29/05, 11:22 AM

21_Ain_295_314_F 314 8/29/05, 11:22 AM
Ruminant Placenta Structure Analysis 315
22
Analysis of the Structure of the Ruminant Placenta
Methods of Fixation, Embedding, and Antibody Localization
at Light and Electron Microscope Levels
F. B. P. Wooding
Key Words: Ruminant; placental structure; embedding; immunocytochemistry.
1. Introduction
The basic structure of all ruminant placentas described so far is cotyledon-
ary (or placentomal), with local proliferations of apposed fetomaternal mem-
branes forming placentomes that are linked by flat interplacentomal areas. In
vivo, the placenta consists of an intimate apposition of trophoblast to uterine
epithelium or derivative, with interdigitation of microvilli on both sides. Any
separation of these two surfaces is an artifact of preparation except in hemopha-
gous zones, where maternal blood is released between the surfaces and phago-
cytosed by the trophoblast.
Placentomes develop from a flat membrane apposition at implantation (20 d
post coitum [dpc] in cow and 16 dpc in sheep) by a mutual growth of
fetomaternal surfaces to form villi, remodeling the endometrium rather than
invading it. Claims of maternal crypts forming into which the trophoblast villi
grow are based on poor fixation and processing, the two surfaces separate very
quickly after death.
The trophoblast epithelium uninucleate cell (UNC) gives rise to binucleate
cells (BNC) with characteristic cytoplasmic granules at the earliest stage of
implantation. The BNC develop in the trophoblast epithelium out of contact
with the basement membrane or the tight junction. When mature (fully granu-
lated), they migrate up to and through the tight junction while maintaining its
barrier function. The BNC apex now fuses with the uterine cell or derivative to
which it is apposed and its cytoplasmic contents are injected into the uterine
315
22_Wooding_315_322_F 315 8/29/05, 11:22 AM

316 Wooding
cell (1). The BNC plasma membrane on the fetal side of the tight junction is
resorbed by the trophoblast UNC. The microvillar junction (MVJ) reforms and
in the cow/deer placental types, the trinucleate cell formed by the injection,
releases the BNC granules to the maternal side by exocytosis, then dies and is
resorbed by the trophoblast. In the sheep/goat placental types, the continuous
injection of BNC content forms a persistent fetomaternal syncytium (“uterine
derivative”) replacing the original uterine epithelial cells (2).
Throughout pregnancy in all ruminants, approx 20% of the trophoblast are
BNC, and approx 20% of these BNC are in the process of migration and injec-
tion. The number of BNC falls in the days before parturition, and this fall can
be induced prematurely by injection of cortisol into the fetal circulation in
vivo (3).
The BNC migration and injection process allows delivery throughout preg-
nancy of the fetal molecules in the granules (which contain placental lactogen
hormone and many different pregnancy-associated glycoproteins [PAGs]) to
the maternal circulation while maintaining the placental barrier between the
two circulations. The width of this barrier is reduced progressively by indenta-
tion of maternal and fetal capillaries into the respective trophoblast and uterine
epithelia, down to between 1 and 2 µm in many places, but the number of
membrane layers in the barrier stays the same throughout pregnancy.
It should be emphasized that the above analysis is the view of the author,
based on a relatively small sample of the enormous variety of ruminant spe-
cies. The author looks forward to more studies with a greater variety of modern
techniques to fully test the interpretation presented previously.
The two main problems to be addressed in investigating ruminant placental
structure and function are the preservation of the fetomaternal interface as it is
in vivo, and the maintenance of the antigenicity of the material during fixation
and embedding. The best (and probably the only) method of achieving the first
objective is to doubly perfuse the placenta via both fetal and maternal circula-
tions as quickly as possible after death of the animal. The uterus should be
removed and perfused via the uterine arteries, first tying or clamping off part
or the whole of one uterine horn if frozen and/or unfixed samples are required.
After 5 min or so of perfusion the uterus is opened and the placenta is perfused
via the umbilical arteries. Perfusion of different food colorings finally allows
identification of the best-fixed areas when the placenta is cut into smaller
samples for processing. Successful perfusion produces a tissue that is firm
enough to maintain its structure when cut into pieces small enough to process
for light microscopy (LM) and electron microscopy (EM).
One can start immersion fixation more rapidly, but cannot avoid crush-
ing and separating the initially soft tissues, and penetration of the fixative
is unavoidably slower and much more uneven. The process can produce useful
22_Wooding_315_322_F 316 8/29/05, 11:22 AM

results but the drawbacks of the range of fixation quality and the lack of pres-
ervation of the volume relationship between blood and tissue must always be
borne in mind when assessing the results.
For optimal ultrastructure and membrane preservation, a fixative with more
than 2% glutaraldehyde is advisable, whereas for immunocytochemistry
(IMCYT), 4% formaldehyde is optimal. Because the cost of animals is high,
perfusing with 4% formaldehyde and, subsequently, cutting the placental
sample up into several different fixatives can help maximize limited animal
numbers. Fortunately, with most ruminants, the amount of tissue available from
a single placenta is considerable. The use of several types of embedding media
is also recommended: wax, acrylic, and epoxy nonosmicated blocks form a
useful basic resource for subsequent work.
To simplify the transition from LM to EM, use a pieces-of-coverslip (2 to
6 mm2) carrier system for LM IMCYT. It reduces antibody-volume require-
ments and storage problems. Then, one can apply in Petri dishes the same pro-
cedures as those used for EM IMCYT with grids. However, this is the
preference of the author, and produces no better results than does conventional
slide processing. To further simplify the number of procedures, the author uses
the immunogold–silver enhancement system for all IMCYT, but this is not
necessarily any more sensitive than more conventional ABC or peroxidase LM
IMCYT methods.
2. Materials
2.1. Fixatives
1. 4% Paraformaldehyde in 0.1 M phosphate-buffered saline (PBS), pH 7.2. To
depolymerize the paraformaldehyde:
a. Add 2 mL NaOH to 50 mL water in a 250-mL conical flask.
b. Add magnetic stirrer bar and stir on magnetic stirrer hotplate.
c. Add 20 g of solid paraformaldehyde (Agar Scientific, Stansted, UK).
d. Heat water to 60°C until solution is clear.
e. Cool and make up to 500 mL with PBS.
2. 1% Glutaraldehyde–4% paraformaldehyde in PBS (in step 1 above, with 21 mL
of 25% aqueous glutaraldehyde [Agar Scientific] added to 479 mL formaldehyde
solution).
3. 4% glutaraldehyde in PBS plus 5% sucrose.
2.2. Embedding Media

1. Epoxy resin: Araldite CY212 kit (Agar Scientific).
2. Acrylic resin: Lowicryl K4M kit (Agar Scientific).
3. Paraffin wax (Agar Scientific).
22_Wooding_315_322_F 317 8/29/05, 11:22 AM

318 Wooding
2.3. Antibodies and IMCYT Reagents

1. Primary antibodies, polyclonal and monoclonal (numerous suppliers: e.g., Dako
Cytomation, Santa Cruz Biotechnology, Chemicon International, Abcam and
so on).
2. Secondary antibodies, labeled with gold colloid. Minimum requirement: goat
anti-rabbit 5 nm gold (detects rabbit polyclonal primary antibodies); goat anti-
mouse 5 nm gold (detects mouse monoclonals). These two are used for LM and
EM localization. Ten- and 15-nm gold can also be used for EM localization. All
of these are from Agar Scientific.
3. Silver enhancement reagents, which make the gold label larger and thus more
visible (Intense M, Amersham International).
4. Petri dishes (9-cm glass).
5. Parafilm roll.
6. Fine curved forceps (no. 7, nonmagnetic).
3. Methods
3.1. Perfusion Fixation
1. Kill animal by shooting (e.g., at slaughterhouse) or barbiturate overdose via
injection.
2. Remove uterine tract as quickly as possible, leaving uterine arteries as long as
possible—trace them as far back to the dorsal aorta as you can before cutting.
3. Wear disposable plastic gloves for all manipulations with fixative. Tie a blunted
18-gauge (or as large as practicable) disposable needle into each uterine artery.
4. Inject fixative at 18 to 20°C through the needles with a 50-mL disposable
syringe. Do not flush the vasculature first with buffer or saline. Manually
apply as much pressure as is necessary to force blood out of the uterine vein until
it runs clear; do not worry about total pressure exerted. With sheep, this will take
approx 150 to 200 mL fixative for each side.
5. Clamp uterine vein outflow and inject a further 25 mL of fixative. This ensures
that as many arteries and veins are perfused as possible.
6. Cut uterus open just above cervix, avoiding obvious blood vessels, and remove
the fetus. Cannulate both umbilical arteries in the placental direction with blunted
needles. The arteries are thicker and more solid than the veins. Inject 100–200 mL
fixative down each, cutting the umbilical veins to allow outflow if necessary. Con-
tinue to trickle fixative through all cannulae for 10 min.
7. Inject 2 mL of aqueous 1% malachite green or green food dye down each uterine
artery and 2 mL of a different color (e.g., 0.1% nigrosin or blue food dye) down
the umbilical arteries. Push the dyes through with 25 mL of fixative (see Note 1).
8. Open uterus out flat, slice across the placentomes centrally to find the best double-
colored ones, and take a 2- to 3-mm thick whole slice across the center of the
placentome. Cut into 2- to 3-mm cross section “matchsticks” running the full
depth of the placentome. This should be accomplished in a pool of fixative on a
wax or Sylgard elastomer resilient layer in a Petri dish. The samples are suitable
22_Wooding_315_322_F 318 8/29/05, 11:22 AM

in size for resin processing; wax processing can accommodate much larger pieces.
For interplacentomal samples, cut 1- × 1-cm squares through the full depth of the
uterine wall, but always cut from the fetal side, again selecting the best-dyed areas.
9. Divide the samples into a chosen range of fixatives and fix for a further 2 h or up
to overnight if it is more convenient.
10. Rinse all samples in buffer (PBS) and store at 4°C until processed. The samples
can be stored for several years at this stage, as long as a bacteriostatic/antifungal
(e.g. 0.05% Thimerosal) is incorporated.
3.2. Immersion Fixation

1. Kill animal, remove uterus.
2. Open uterus and remove fetus.
3. Cut free a whole placentome; grip one side firmly with rat tooth forceps or equiva-
lent and position on a half-filled wax or Sylgard Petri dish with the fetal side
uppermost. Use a fresh, single-edged razor blade to slice (try not to push down-
ward) across the center of the placentome, then repeat the slicing to produce a 2-
to 3-mm thick slice that is the full depth of the placentome. Repeat this with
several placentomes, immersing the slices in fixative as soon as they are cut.
4. For interplacentomal samples, cut 1-cm squares that are the full depth of the
uterine wall and leave these in fixative. After about 5 min fixation, place each
placentomal slice flat on the wax or Sylgard Petri dish in a pool of fixative and
cut “matchsticks” 2 × 2 mm in cross-section across the full depth of the
placentome from the fetal to the maternal surfaces. After 2 h to overnight fixa-
tion, the interplacentomal samples can be dissected free of the uterine wall and
some of the myometrium and transferred to buffer ready for embedding.
3.3. Embedding
1. 2-mm cubes are best, but up to 4-mm cubes will embed satisfactorily. If you use
larger areas, keep one dimension down to 2 mm to allow sufficient liquid resin
permeation prior to curing.
2. For conventional ultrastructural studies only, with no IMCYT, use the glutaral-
dehyde-fixed material and, in a fume cupboard, transfer to freshly made 1% osmic
acid plus 1.5% potassium ferrocyanide in PBS for 1 h, rinse with distilled water,
and then incubate in 2% uranyl acetate in distilled water for 15 min, finally rins-
ing with distilled water.
3. Dehydrate in 50% alcohol, 75% alcohol, and 2X 100% alcohol for 15 min each.
The author uses disposable 14- × 50-mm glass tubes with plastic stoppers for all
embedding procedures.
4. For epoxy (Araldite): 100% Alco to propylene oxide (15 min); overnight in 66%
araldite/33% propylene oxide; 4 h in 100% araldite. Place in plastic (Beem) cap-
sules or trays (Agar Scientific) in fresh araldite, orienting as required. Cure at
60°C for 8 h or until the surface, when cold, is not sticky to touch.
5. For acrylic (K4M): 100% Alco to 33% K4M/66% Alco (40 min); 66% K4M/
33% Alco (40 min); 100% K4M overnight. Place in fresh K4M in plastic (Beem)
22_Wooding_315_322_F 319 8/29/05, 11:22 AM

320 Wooding
capsules with minimum air space, close tops, and cure at a distance of 30 cm
from an ultraviolet (UV) light (wavelength 375 nm) for 2 to 3 h at room tempera-
ture, or as long as it takes—check periodically with a mounted needle (see Note 2).
6. For wax: 100% Alco to 100% Xylene or equivalent (overnight), 100% wax at 55
–60°C (3X 60 min). Place into moulds at 60°C and, finally, cool rapidly to pro-
duce best cutting texture.
7. All of the above processing is performed at room temperature except where noted.
The K4M can also be processed at –20°C to preserve antigenicity (use a –20°C
freezer and dry ice in a polystyrene box for solution changes). The curing takes
much longer at –20°C.
3.4. Sectioning
1. Fill a rack with 22- × 22-mm coverslips or slides and put into 2% 3-Aminopropyl
trimethoxy silane (APTES; Sigma) in 100% alcohol for 10 min, rinse in 100%
alcohol, leave in distilled water for 1 min, blow dry or blot edges thoroughly, and
dry in oven at 60°C for 60 min. APTES ensures that the sections stay firmly stuck
to the glass during processing. Place a coverslip on a hard, flat surface, and
score in two directions at right angles with a diamond pencil to produce pieces
2–9 mm x 2–9 mm (in various sizes to accommodate a variety of section sizes).
Break into the individually scored pieces using gentle pressure with the “wrong”
end of a pair of forceps on filter paper.
2. Cut 0.5- to 1.0-µm thick resin sections or 4- to 8-µm thick wax sections for LM,
or 80- to 110-nm thick sections for EM, onto a water surface using an ultramicro-
tome. For LM, pick up at least two sections per cover slip piece or slide-transfer
each of the sections with a metal loop to a drop of water on a slide. LM IMCYT
is inherently variable—do not be satisfied with what is localized on only one of a
pair of sections without checking further. For EM, pick up on 300 mesh naked
or formvar film-coated nickel (not copper) grids or slotted supports. Dry all at
60°C for 30 min before further processing.
3.5. Immunocytochemistry
1. Initially, removal of wax or araldite from the sections is necessary. K4M is used
without any pretreatment.
a. Dewaxing. Immerse coverslip pieces (use one eppendorf tube for each piece)
or slides in xylene or equivalent for 10 min, transfer to 100% alcohol for
10 min, wash in running tap water for 15 min. Dry around the sections on
slides and encircle them with a wax pencil or PAP pen to minimize the vol-
umes of reagents required subsequently to cover the sections.
b. Resin removal. Float coverslip pieces section side down on a drop of sodium
ethoxide (dissolve 15 g of NaOH pellets in 15 mL of 100% alcohol using a
magnetic stirrer bar for 4 h; keep in the dark). Handle this carefully; it is a
very corrosive solution. On slides, use a drop on top of the sections. All float-
ing on drops is performed on parafilm sheets secured in Petri dishes by press-
ing along the film edges with the “wrong” end of a pair of forceps. Treat with
22_Wooding_315_322_F 320 8/29/05, 11:22 AM

sodium ethoxide for 15 min, jet-wash (from a washbottle with a narrow jet)
the coverslip pieces or slides with 100% alcohol, float on (or drop on) 50%
alcohol, leave for 5 min, then wash with water for 5 min.
2. K4M sections start here with the dewaxed or deresinated sections. Float sections
on coverslip piece or grid (or place a drop on a slide) for 10 min on a drop of PBS
containing 1% bovine serum albumin (Sigma) and 0.05% Thimerosal (Sigma).
This solution is used for all antibody dilutions.
3. Sections on grids and coverslip pieces (subsequently referred to as sections) are
touched at the side to filter paper to draw off excess fluid (do not wash), floated
on the primary antibody (e.g., rabbit or mouse) (see Note 3) overnight at 4°C, jet-
washed with PBS, and incubated for 40 min on the relevant secondary gold col-
loid-labeled antibody (e.g., goat anti-rabbit or goat anti-mouse) using 4-nm gold
for the LM and 4-, 10-, or 15-nm gold for EM.
4. After the immunoreaction, the sections are jet-washed with PBS and distilled
water and left in distilled water for 10 min. The 4 nm gold is now intensified by
floating the sections on the intensification reagent for 7 min for grids (EM) and
15–20 min for LM (slides or coverslip pieces). The level of LM label can be
monitored continuously by eye or on a microscope. The reaction is stopped by
jet-washing with distilled water and the LM sections dried. For counterstaining the
EM grids can be transferred successively to uranyl acetate (5% aqueous, 15 min),
jet-washed with distilled water, then with lead citrate (0.1% lead citrate in 0.05%
NaOH), jet-washed, and finally dried. Reduce the staining times if the electron
density obscures the label. LM counterstains can be applied once the
immunoreaction has been assessed. Try 1% toluidine blue in 1% sodium borate,
and reduce the stain level (differentiate) with acid alcohol (1 mL 1N HCl per 15
mL 100% alcohol). Alternatively, try 0.2% Fast green in 0.03 N HCl and reduce
with warm water.
5. Permanently mount the LM sections using Biomount (BB-International, Cardiff,
UK). In conventional mountants (e.g., Depex), the metallic black silver depos-
ited in the intensification reaction oxidizes quite rapidly to an invisible silver
carbonate.
4. Notes
1. Perfusion fixation is rarely 100%; the use of dyes is essential to monitor this and
select the best areas of fixation.
2. The time required for curing resins can be unpredictable. With K4M, always cure
initially from underneath the Beem capsules, targeting the end with the speci-
men. Suspend the capsules in a perforated sheet of clear plastic above the UV
light source. If the top half/portion is slow to cure (check with a mounted needle),
suspend the UV light above the capsules to finish. When using Araldite, cure at
60°C until it resists a mounted needle and is not sticky on the surface after cool-
ing to room temperature.
3. Immunocytochemistry: start with one-tenth and one-hundredth dilutions of any
new antibody, then dilute according to results. Never trust localizations seen on
22_Wooding_315_322_F 321 8/29/05, 11:22 AM

322 Wooding
only one of a pair of sections—IMCYT results are inherently variable. The au-
thor has not found antigen retrieval (microwave, pressure cooking, heat treat-
ment) prior to IMCYT of the sections useful, but many people do—consult
Chaiwun et al. (4) and Groos et al. (5) for details.
References
1. Wooding, F. B. P. and Flint, A. P. F. (1994) Placentation, in Marshall’s Physiol-
ogy of Reproduction, 4th edition, Vol. 3, Pregnancy and Lactation (Lamming, G.
E., ed.). Chapman and Hall, London: pp. 242–460.
2. Wooding, F .B .P., Morgan, G., Brandon, M. R., and Camous, S. (1994) Mem-
brane dynamics during migration of placental cells through trophectodermal tight
junctions in sheep and goats. Cell Tissue Res. 276, 387–397.
3. Ward, J. W., Wooding, F. B. P. and Fowden, A. L. (2002) Effects of cortisol on
the binucleate cell population in the ovine placenta during late gestation. Placenta
23, 451–458.
4. Chaiwun, B., Shi, S. R., Cote, R. J., and Taylor, C. R. (2002) Major factors influ-
encing the effectiveness of Antigen Retrieval Immunohistochemistry, in Antigen
Retrieval Techniques (Shi, S. R., Gu, J., and Taylor, C. R., eds.). Eaton Publish-
ing, Natick, MA: pp 41–53.
5. Groos, S., Reale, E., and Luciano, L. (2001) Reevaluation of epoxy resin sections
for LM and EM immunostaining. J. Histochem. Cytochem. 49, 397–406.
22_Wooding_315_322_F 322 8/29/05, 11:22 AM

Bovine Placenta Characterization 323
23
Characterization of the Bovine Placenta
by Cytoskeleton, Integrin Receptors, and Extracellular
Matrix
Christiane D. Pfarrer
Summary
The cytoskeleton together with integrin receptors and proteins of the extracellular matrix
provide sensitive indices of the development and organization of the bovine placenta. The bo-
vine placenta is classified as synepitheliochorial because migrating trophoblast giant cells fuse
with single uterine epithelial cells. This phenomenon may be interpreted as a restricted tropho-
blast invasion. Bovine placentomes from early placentation until term can be characterized by
indirect immunohistochemical methods. In order to do so, placental tissues are snap-frozen in
liquid nitrogen or perfusion-fixed in formalin and embedded in paraffin. Depending on the
antibodies used, the different cell types within the cow placenta are identified either on frozen
sections or on paraffin sections according to the expression of different cytoskeletal filaments
(α smooth muscle actin, different cytokeratins, desmin and vimentin). The specific expression
of integrin receptors (subunits α1, α2, α3, α4, α5, α6, αv, β1, β3, and α4) as well as proteins of the
extracellular matrix (collagens type I and IV, fibronectin, and laminin) in the different cell
populations is also examined.
Key Words: Placenta; bovine; trophoblast giant cell; migration; cytoskeleton; integrin;
extracellular matrix.
1. Introduction
The cow placenta is cotyledonary and synepitheliochorial. That means fetal
cotyledons interdigitate with maternal caruncles, thus forming so-called
placentomes, where the chorionic epithelium (or trophoblast) is in intimate
contact with the uterine or caruncular epithelium. The bovine chorionic epithe-
lium consists of “normal” polarized cytotrophoblast cells and a second popula-
tion of trophoblast cells, namely nonpolarized, mostly binucleated trophoblast
giant cells (TGCs) (1). These TGC are generated by unidentified stem cells
323
23_Pfarrer_323_336_F 323 8/29/05, 11:22 AM

324 Pfarrer
through acytokinetic mitosis (2), and have migratory properties. TGC migra-
tion is characterized by a sequence of events: first, TGCs “squeeze through”
chorionic tight junctions (3) and, in a next step, fuse with single uterine epithe-
lial cells via specifically formed pseudopodia (1). Finally, the resulting
trinucleated feto–maternal hybrid cells degenerate (3,4), maybe partly by
apoptosis and/or through exfoliation toward the feto-maternal contact inter-
face, where they may be phagocytosed by mononuclear trophoblast cells. Be-
cause this TGC migration does not continue beyond the maternal basement
membrane (BM), it is regarded as a restricted type of trophoblast invasion (5).
Highly invasive trophoblast during hemochorial placentation of humans and
rodents is characterized by specific binding patterns of integrin receptors to
proteins of the extracellular matrix (ECM) (6), and other invasive processes
like tumor growth are associated with similar/comparable expression patterns
(7). Integrins are heterodimeric transmembrane glycoproteins serving as adhe-
sion receptors at the cell surface (8). Twenty-four different heterodimers that
are formed by the dimerization of various α- and β-chains have been identified
to date (9). Integrins interact with a variety of ligands, including ECM glyco-
proteins and molecules on the surface of neighboring cells (10,11). The β sub-
unit is linked to components of the cytoskeleton, and besides promoting cellular
conformational changes, integrin activation may also induce phosphorylation
cascades of signalling proteins (12). This results in bidirectional signal trans-
duction (“outside-in” and “inside-out” signaling) (8).
Prior to implantation, the integrin β1 subunit occurs basolaterally on fetal
mononuclear trophoblast cells and trophoblast giant cells of day 24 post insemi-
nation, which implicates β1 integrin in the process of trophoblast migration and
cell development (13). As attachment proceeds, implantation-associated changes
in the expression patterns of integrin subunits α1, α3, and α6, as well as extra-
cellular matrix proteins collagen IV and laminin, have been observed in the
bovine endometrium and in isolated binucleate trophoblast cells (14). The alter-
ation of integrin and ECM expression patterns may be induced by the fusion of
TGC with uterine epithelial cells (14). Implantation in the closely related goat
differs, because collagen type IV decreases in maternal BM and is conse-
quently interpreted as a modification of the BM composition rather than its
destruction (15).
In the definitive bovine placenta, around day 80 until near term (around day
270), the expression of the respective integrin subunits suggests the presence
of collagen receptor α2β1 and laminin receptor α6β1, which may be responsible
for the adhesion of fetal and maternal epithelial cells to BM (5) (Fig 2).
Coexpression of laminin, α2, α6, and β1 in nonpolarized TGC supports the
concept that TGC migrate along their own laminin matrix utilizing integrin
23_Pfarrer_323_336_F 324 8/29/05, 11:22 AM

α6β1, and maintain cell–cell contacts with neighboring cells via α2β1 integrin
(5). In maternal stroma and fetal mesenchyme, a pool of integrins is present
which may be involved in the regulation of proliferation and differentiation of
maternal septa and fetal villi (5) (Fig 3).
During the estrous cycle of cows, changes in the localization of collagens I,
III, IV, and VI in the uterine wall are also observed (16). In contrast, the por-
cine uterine epithelium, lacking TGC, shows no detectable structural changes
in the basement membranes during epitheliochorial implantation (17,18).
Cell migration is dependent on the mechanical properties of the actin cytosk-
eleton, and changes in cell shape, anchorage, and motility are associated with the
dynamic reorganization of the actin-cytoskeleton (19,20). Migratory TGC pos-
sess specifically arranged actin filaments (21).
Integrin binding leads to cytoskeletal responses including induction of focal
accumulations of a variety of cytoskeleton-associated molecules, vinculin,
talin, α-actinin, and F-actin, which may lead to conformational changes of the
actin cytoskeleton and the transduction of signals to the nucleus and altered
gene transcription (22). In vitro, beads coated with the ECM glycoprotein
osteopontin induced a transmembrane accumulation of focal adhesion com-
plexes (talin and α-actinin) at the apical surface of ovine luminal epithelial and
conceptus trophectoderm cells, revealing functional integrin activation and
cytoskeletal reorganization and potentially simulating events occurring during
embryo implantation in sheep (23).
Identification of the tissue components within the bovine placentome (fetal
cotyledon and maternal caruncle—both consist of connective tissue and epi-
thelium) may be facilitated by the analysis of the expression of different
cytoskeletal filaments (Fig. 1). The method is well established, and has been
used for the characterization of other placental types, such as the endothelio-
chorial mink and hemochorial guinea pig, macaque, and human (24–29). In the
bovine placenta, cytokeratin is used for the identification of fetal and/or mater-
nal epithelial cells, whereas tissue of mesenchymal origin can be detected by
vimentin (connective tissue, including endothelial cells), α-smooth muscle
actin and desmin (smooth muscle cells of blood vessel walls and pericytes).
2. Materials
2.1. Tissue Collection
1. Liquid nitrogen (N2).
2. Tissue-Tek® O.C.T.™ Compound (Sakura, Torrance, CA).
3. Fixative: 4 % buffered formaldehyde (pH 7.4).
4. Washing step: phosphate-buffered saline for immunohistochemistry (PBS-IHC)
(see Subheading 2.2., step 1).
23_Pfarrer_323_336_F 325 8/29/05, 11:22 AM

326 Pfarrer
Fig.1. Immunohistochemistry for cytoskeletal filaments in bovine placentomes of

late gestation (A,C,D: without counterstaining; B: weak counterstaining with hema-
toxylin). (A) α-sm actin is observed in smooth muscle cells of large blood vessel walls
and pericytes of microvasculature (arrows). (B) Cytokeratin immunostains tropho-
blast cells (T) in a differentiated pattern, mononuclear trophoblast cells display cyto-
plasmic staining (arrowheads), while trophoblast giant cells display membrane staining
(arrow). (C) Vimentin is localized in cells of the maternal stroma (MS) and fetal mes-
enchyme (FM), including endothelial cells. (C) Desmin is found exclusively in smooth
muscle cells of small arterioles in the tips of the maternal septa (arrows), and in the
walls of large allantochorionic blood vessels (not shown). FM, fetal mesenchyme;
MS, maternal stroma; T, trophoblast; UE, uterine (maternal) epithelium.
2.2. Immunohistochemistry
1. PBS-IHC stock solution: 41 g NaCl, 11 g Na2HPO4 · 2H2O, 2.75 g KH2PO4. Add
up to 1000 mL with distilled water, adjust to pH 7.2. Working solution: one part
stock solution and four parts distilled water give 0.1 M PBS-IHC.
2. Primary antibodies and their sources are listed in Table 1.
3. Secondary antibodies generally detect the species the primary antibody is raised
in. In this case: biotinylated horse anti-mouse/anti-rabbit secondary antibody
(Vector Laboratories, Burlingame, CA, USA).
23_Pfarrer_323_336_F 326 8/29/05, 11:22 AM

Table 1
Antibodies
Antibodies Clone Host Dilution Supplier (Cat. No.)
α-sm Actin 1A4 mouse 1:100 Dako (M0851)
Collagen IV CIV 22 mouse 1:100 Dako (M785)
Cytokeratin AE1, AE3, Ks13.1 mouse 1:20a Linaris (E020)
Desmin DE-R-11 mouse 1:50a Dako (M0724)
Fibronectin IST-3 mouse 1:250b Sigma (F0791)
Integrin α1 polyclonal rabbit 1:800 b Chemicon (AB1934)
Integrin α2 polyclonal rabbit 1:400b Chemicon (AB1936)
Integrin α6 NKl-GoH3 rat 1:25b Chemicon (MAB1378)
Integrin αv polyclonal rabbit 1:100b Chemicon (AB1930)
Integrin β1 polyclonal rabbit 1:100b Chemicon (AB1952)
Integrin β3 PM 6/13 mouse 1:80b Chemicon (MAB1381)
Integrin β4 polyclonal rabbit 1:1600b Chemicon (AB1922)
Laminin polyclonal rabbit 1:20b BioGenex (AR078)
Vimentin Vim 3B4 mouse 1:100 Dako (M7020)
Secondary Antibodies:
Anti mouse/ IgG (H+L) horse 10–20 µL/mL Vector (BA-1400)
anti rabbit
(biotinylated)
Anti-mouse FITC IgG donkey 1:200 Chemicon (AP192F)
Anti-rabbit FITC IgG donkey 1:100 Chemicon (AP182F)
Anti-rat CY3 IgG donkey 1:100 Chemicon (AP189C)
Anti-rabbit CY3 IgG donkey 1:300 Chemicon (AP182C)
FITC, fluorescein isothiocyanate; IgG, immunoglobulin G.

aParaffin-embedded sections are pretreated with citrate buffer, three times for 5 min, in microwave
oven at 700 W to demask antigens.

bWorks only with frozen sections.
4. Signal amplification is accomplished by incubation with avidin/biotinylated

horseradish peroxidase complex (ABC-Method; Vectastain-Universal-ELITE-
ABC-Kit®, Vector Laboratories).
5. A chromogen is used to visualize the immunoreaction: 3-amino-9-ethylcarbazole
(AEC; Vector Laboratories).
6. Mounting and sealing is performed with Kaiser’s glycerol gelatine (Merck,
Darmstadt, Germany).
7. Proteinase K (Roth, Karlsruhe, Germany).
23_Pfarrer_323_336_F 327 8/29/05, 11:22 AM

328 Pfarrer
2.3. Immunofluorescence
1. PBS for immunofluorescence (PBS-IF) stock solution: 16 g NaCl, 0.4 g KH2PO4,
2.28 g Na2HPO4. Add up to 1000 mL with distilled water, adjust pH to 7.3. Work-
ing solution: one part stock solution and one part distilled water give 0.02 M
PBS-IF.
2. PBS-IF/Tween: 0.3 % Tween (Polyoxyethylene-Sorbitan Monolaurate) in PBS.
3. Antibody dilution buffer (100 mL): 0.3 % Tween in PBS-IF, 1.0 g bovine serum
albumin (BSA), 45 mL glycerol (pH 8.0). Economize production of antibody
dilution buffer by dividing into aliqots: produce aliquots with 0.5 M Na2CO3 (pH
9.5) and freeze in portions of 5 mL.
4. Mounting media: e.g., Vectashield (H-1200, Vector Laboratories) or ProLong®
Antifade Kit (Molecular Probes Inc., Eugene, OR, USA) or Mowiol (Sigma)-
propyl gallate (Sigma) mounting medium.
3. Methods
3.1. Tissue Collection
1. After removal from the cow (see Note 1), the uterus is opened along the large
curvature with a pair of scissors. The fetus is removed and crown–rump length is
recorded (see Note 2).
2. Snap-freezing: select placentomes, gently excise whole placentomes (see Notes
3 and 4), cut into smaller pieces (around 1 cm3), wrap in aluminum foil or use
Tissue-Tek for embedding, snap-freeze in liquid nitrogen, and store at –80°C
until use.
3. Perfusion fixation: select placentomes (see Note 5), separate the artery, make a
small incision into wall of artery, enter a blunt cannula (see Note 6), secure with
clamp, and inject approx 80 mL of fixative with gentle and steady manual pres-
sure (see Note 7).
4. Excise placentomes, cut into slices of approx 0.5 cm, and postfix in the same fixa-
tive as mentioned above for 24 h and wash in PBS-IHC (three times for 10 min).
5. Cut into smaller pieces of around 1 cm2. Thickness of slices remains 0.5 cm.
Make sure that one tissue block includes the total height of the placentome from
allantochorion to caruncular stalk.
6. Paraffin-embed the tissue.
7. Depending on the properties of the antibodies used, either frozen or paraffin sec-
tions are used for the immunohistochemical detection of selected antigens.
8. Frozen sections: mount tissue blocks of approx 1 cm3 on specimen holders with
Tissue-Tek, produce cryostat sections of 10–12 µm at approx –22°C, and mount
sections on Superfrost Plus glass slides.
9. Paraffin sections: prepare sections of approx 3 µm, mount these on Superfrost
Plus glass slides, deparaffinize in xylene (once for 5 min at 60°C, twice for 5 min
at room temperature), and rehydrate in a series of graded alcohol (5 min in each
of 100 %, 100 %, 96 %, and 70 %).
23_Pfarrer_323_336_F 328 8/29/05, 11:22 AM

3.2. Immunohistochemistry (Figs. 2 and 3)

3.2.1. Cryostat Sections
1. Air-dry cryostat sections for 1 h at room temperature.
2. Fix with acetone/methanol (70:30) for 10 s at room temperature.
3. Perform all following incubation steps in moist chamber at room temperature
(see Note 8).
4. Quench endogenous peroxidase with 1% H2O2 in PBS-IHC for 10 min.
5. Rinse sections in PBS-IHC (three times for 5 min).
6. Incubate in 0.02–5% BSA in PBS-IHC for 20 min to reduce nonspecific binding
(see Note 9).
7. Incubate with primary antibodies for 1 h at room temperature (see Note 10). The
details of the antibodies used are shown in Table 1.
8. Rinse in PBS-IHC (three times for 5 min).
9. Incubate with secondary antibody (Table 1) for 20 min.
10. Rinse in PBS-IHC (three times for 5min).
11. Incubate with ABC complex for signal amplification (45 min).
12. Rinse in PBS-IHC (three times for 5 min),
13. Visualize with AEC for 10 min.
14. Rinse in distilled water (three times for 5 min).
15. Optional: counterstain with hematoxylin (30 s) (see Note 11) .
16. Rinse with tap water (10 min).
17. Mount coverslips with Kaiser’s Gelatine.
18. Carry out control reactions in parallel (see Note 12).
19. View slides and store dry, dark, and at moderate temperatures; the chromogen
will not fade out.
3.2.2. Paraffin-Embedded Sections

Paraffin-embedded sections may also be used for immunohistochemistry.
However, steps 1 and 2 must be omitted. Additionally, it may be necessary to
pretreat the sections with either proteinases or citrate buffer and heat to unmask
antigens (see Note 13).
3.3. Immunofluorescence (see Note 14) (Figs. 2 and 3)
3.3.1. Cryostat Sections
1. Fix tissue sections at –20°C in 100 % methanol for 10 min.
2. Air-dry methanol fixed sections.
3. Wash quickly in PBS-IF/Tween.
4. Dilute protein blocking solution 1:20 in antibody dilution buffer and add to sec-
tions (see Note 15) and incubate for 1 h at room temperature in a humidified
chamber.
5. Wash in PBS-IF/Tween (short).
23_Pfarrer_323_336_F 329 8/29/05, 11:22 AM

330 Pfarrer
Fig. 2. Immunohistochemistry and immunofluorescence for laminin, collagen type

IV and integrin subunits α6 and β1 in bovine placentomes of late gestation (A,C,E,G:
immunohistochemistry with counterstained nuclei, which are not positive; B,D,F,H:
immunofluorescence). (A,B) Laminin is detected in basement membranes of fetal tro-
phoblast and maternal epithelium as well as in TGC. (C,D) α6 integrin subunit. (E,F)
β1 integrin subunit. (C–F) α6 and β1 integrin are colocalized with laminin at the
basal aspects of fetal and maternal epithelium and in TGC. (G,H) Collagen type IV
23_Pfarrer_323_336_F 330 8/29/05, 11:22 AM

Fig. 3. Immunohistochemistry and immunofluorescence for fibronectin and the

integrin subunits αv and β3 in bovine placentomes of late gestation (A,C,E: Immuno-
histochemistry with counterstained nuclei, which are not positive; B,D,F: Immunof-
luorescence). (A,B) Fibronectin. (C,D) αv integrin subunit. (E,F) β3 integrin subunit.
Fibronectin and the integrin subunits αv and β3 are colocalized in maternal stroma,
whereas the fetal mesenchyme expresses fibronectin but not the integrin subunits αv
and β3. FM, fetal mesenchyme; MS, maternal stroma; T, trophoblast; UE, uterine (ma-
ternal) epithelium. Fluorescence pictures are provided through the courtesy of Martina
Zeiler, Justus-Liebig-University Giessen, Germany.
(Fig 2. continued) is observed in fetal mesenchyme and to lesser extent in maternal

stroma. FM, fetal mesenchyme; MS, maternal stroma; T, trophoblast; UE, uterine (mater-
nal) epithelium. Fluorescence pictures are provided through the courtesy of Martina Zeiler,
Justus-Liebig-University Giessen, Germany.
23_Pfarrer_323_336_F 331 8/29/05, 11:22 AM

332 Pfarrer
6. Dilute primary antibodies with antibody dilution buffer (e.g., 1:200) and add to
sections and incubate at 4°C in a humidified chamber overnight.
7. Rinse in PBS-IF/Tween, three times for 10 min.
8. Dilute secondary antibodies (fluorescein isothiocyanate [FITC]- or CY3-conju-
gated) 1:200 in antibody dilution buffer and apply to sections (from now on, all
steps must be conducted in darkness) and incubate at room temperature in a humidi-
fied chamber for 1 h.
9. Rinse in PBS-IF/Tween, three times for 10 min.
10. Wash in distilled water (short).
11. Apply mounting medium (see Note 16) and seal with nail polish.
12. Examine within a week with fluorescence microscope (store sections at 4°C).
3.3.2. Paraffin-Embedded Sections

See Subheading 3.2.2., immunohistochemistry with paraffin.
4. Notes
1. If the material is taken during routine slaughtering, it is very important to remove
the uterus from the cow as fast as possible to avoid separation of fetal and mater-
nal tissues. Excision and/or fixation should be done immediately.
2. If cows of defined gestational ages are not available, the measurement of the fetal
crown-rump length can be done. This will provide an approximation of the stage
of gestation.
3. Gentle manipulation of placentomes is essential to avoid separation of fetal and
maternal components of the placentome. Generally, it is very hard to prevent this
separation, especially when cutting the placentomes into smaller pieces. It can be
helpful to freeze complete placentomes and to split them with a knife and a ham-
mer when frozen.
4. Select and excise material for snap-freezing first to avoid contamination with
fixative.
5. If you wish to take material for examination with molecular biological methods
also, use gloves during the preparation and handling of the equipment and tissue
to decrease RNA degradation.
6. Try to select placentomes with only one supplying allantochorionic artery and
vein to avoid incomplete fixation and stay away from areas where placentomes
were already excised, because fixative will leak from cut blood vessels.
7. The use of butterfly cannulae is most convenient, because manipulation can be
done from a distance.
8. Successful fixation will be obvious, because placentomes will turn pale and hard,
and venous efflux will consist of pure fixative in the end. If this is not the case,
check for serial connection to other placentomes and clamp these. Then repeat
fixation.
9. Incubation with primary antibodies may be done either at room temperature (or
37°C) for 1 h or at 4°C overnight. Please note that the specificity of antibody
binding may be decreased at room temperature and thus can be associated with
23_Pfarrer_323_336_F 332 8/29/05, 11:22 AM

increased background or unspecific staining. Each antibody has to be evaluated

separately.
10. The BSA concentration must be empirically determined for each antibody. The
concentration may vary from 0.02 to 5 % in PBS-IHC.
11. Unmasking of antigens can be achieved by boiling in citrate buffer or incubation
with Proteinase K or trypsin; see data sheet of antibodies.
12. Counterstaining facilitates orientation within the tissue. However, immunohisto-
logical staining without counterstaining may be preferred for the preparation of
images in black and white.
13. Appropriate controls must be utilized with normal mouse immunoglobulins (for
monoclonal antibodies raised in the mouse) or normal rabbit sera (for polyclonal
antibodies raised in the rabbit). Further control sections may be incubated with
PBS instead of primary antibodies.
14. Immunohistochemistry vs immunofluorescence—each method has its advantages
and disadvantages, which have to be taken into account:
• Immunohistochemistry may be viewed at daylight, does not fade out, and,
when counterstaining is applied, orientation is very easy; however, the reac-
tion product may diffuse away from its original location and small amounts of
antigens may not be visualized.
• Immunofluorescence must be handled in darkness; rapid documentation is
necessary, because fluorescence fades out, but very small amounts of antigen
can be detected; and double staining is easy to apply, whenever antibodies
generated in different species are available.
15. The blocking solution (serum or immunoglobulin G) should be taken from the
species the secondary antibody was raised (e.g., goat).
16. Different commercial mounting media are available, e.g., Vectashield (contains
4',6-diamidino-2-phenylindole [DAPI] as counterstain for nuclei), Prolong
Antifade Kit (prevents fading very effectively), and Mowiol (inexpensive, but
fades out within few days).
Acknowledgments
The author dedicates this book chapter to her scientific mentor Professor Dr.
Rudolf Leiser, and gratefully acknowledges the generous donation of fluores-
cence pictures by Martina Zeiler (both Justus-Liebig-University Giessen, Ger-
many). The author further acknowledges the fruitful collaboration with Drs.
M. Guillomot (INRA, Jouy-en-Josas, France), G. Johnson (Texas A&M Uni-
versity, College Station, TX, USA), P. Hirsch, and C. Y. Lang (both Justus-
Liebig-University Giessen, Germany).
References
1. Wooding, F. B. and Wathes, D. C. (1980) Binucleate cell migration in the bovine
placentome. J. Reprod. Fertil. 59, 425–430.
23_Pfarrer_323_336_F 333 8/29/05, 11:22 AM

334 Pfarrer
2. Klisch, K., Hecht, W., Pfarrer, C., Schuler, G., Hoffmann, B., and Leiser, R.
(1999) DNA content and ploidy level of bovine placentomal trophoblast giant
cells. Placenta 20, 451–458.
3. Wooding, F. B. (1992) Current topic: the synepitheliochorial placenta of rumi-
nants: binucleate cell fusions and hormone production. Placenta 13, 101–113.
4. Hoffman, L. H. and Wooding, F. B. (1993) Giant and binucleate trophoblast cells
of mammals. J. Exp. Zool. 266, 559–577.
5. Pfarrer, C., Hirsch, P., Guillomot, M., and Leiser, R. (2003) Interaction of integrin
receptors with extracellular matrix is involved in trophoblast giant cell migration
in bovine placentomes. Placenta 24, 588–597.
6. Damsky, C. H., Librach, C., Lim, K. H., et al. (1994) Integrin switching regulates
normal trophoblast invasion. Development 120, 3657–3666.
7. Lohi, J., Oivula, J., Kivilaakso, E., et al. (2000) Basement membrane laminin-5 is
deposited in colorectal adenomas and carcinomas and serves as a ligand for
alpha3beta1 integrin. APMIS 108, 161–172.
8. Hynes, R. O. (1987) Integrins: a family of cell surface receptors. Cell 48, 549–554.
9. Matlin, K. S., Haus, B., and Zuk, A. (2003) Integrins in epithelial cell polarity:
using antibodies to analyze adhesive function and morphogenesis. Methods 30,
235–246.
10. Ruoslahti, E. (1991) Integrins. J. Clin. Invest. 87, 1–5.
11. Bosman, F. T. (1993) Integrins: cell adhesives and modulators of cell function.
Histochem. J. 25, 469–477.
12. Giancotti, F. G. and Ruoslahti, E. (1999) Integrin signaling. Science 285, 1028–1032.
13. MacLaren, L. A. and Wildeman, A. G. (1995) Fibronectin receptors in preimplan-
tation development: cloning, expression, and localization of the alpha 5 and beta
1 integrin subunits in bovine trophoblast. Biol. Reprod. 53, 153–165.
14. MacIntyre, D. M., Lim, H. C., Ryan, K., Kimmins, S., Small, J. A., and MacLaren,
L. A. (2002) Implantation-associated changes in bovine uterine expression of
integrins and extracellular matrix. Biol. Reprod. 66, 1430–1436.
15. Guillomot, M. (1999) Changes in extracellular matrix components and
cytokeratins in the endometrium during goat implantation. Placenta 20, 339–345.
16. Boos, A. (2000) Immunohistochemical assessment of collagen types I, III, IV and
VI in biopsy samples of the bovine uterine wall collected during the oestrous
cycle. Cells Tissues Organs 167, 225–238.
17. Bowen, J. A. and Hunt, J. S. (1999) Expression of cell adhesion molecules in
murine placentas and a placental cell line. Biol. Reprod. 60, 428–434.
18. Burghardt, R. C., Bowen, J. A., Newton, G. R., and Bazer, F. W. (1997) Extracel-
lular matrix and the implantation cascade in pigs. J. Reprod. Fertil. Suppl. 52,
151–164.
19. Pavalko, F. M. and Otey, C. A. (1994) Role of adhesion molecule cytoplasmic
domains in mediating interactions with the cytoskeleton. Proc. Soc. Exp. Biol.
Med. 205, 282–293.
20. Small, J. V., Rottner, K., and Kaverina, I. (1999) Functional design in the actin
cytoskeleton. Curr. Opin. Cell Biol. 11, 54–60.
23_Pfarrer_323_336_F 334 8/29/05, 11:22 AM

21. Lang, C. Y., Hallack, S., Leiser, R., and Pfarrer, C. (2004) Cytoskeletal filaments
and associated proteins during restricted trophoblast invasion in bovine
placentomes: light and transmission electron microscopy and RT-PCR. Cell Tis-
sue Res. 315, 339–348.
22. Miyamoto, S., Teramoto, H., Coso, O. A., et al. (1995) Integrin function: molecu-
lar hierarchies of cytoskeletal and signaling molecules. J Cell Biol. 131, 791–805.
23. Johnson, G. A., Bazer, F. W., Jaeger, L. A., et al. (2001) Muc-1, integrin, and
osteopontin expression during the implantation cascade in sheep. Biol. Reprod.
65, 820–828.
24. Khong, T. Y., Lane, E. B., and Robertson, W. B. (1986) An immunocytochemical
study of fetal cells at the maternal-placental interface using monoclonal antibod-
ies to keratins, vimentin and desmin. Cell Tissue Res. 246, 189–195.
25. Beham, A., Denk, H., and Desoye, G. (1988) The distribution of intermediate
filament proteins, actin and desmoplakins in human placental tissue as revealed
by polyclonal and monoclonal antibodies. Placenta 9, 479–492.
26. Carter, A. M., Tanswell, B., Thompson, K., and Han, V. K. (1998) Immunohis-
tochemical identification of epithelial and mesenchymal cell types in the chorio-
allantoic and yolk sac placentae of the guinea-pig. Placenta 19, 489–500.
27. Winther, H., Leiser, R., Pfarrer, C., and Dantzer, V. (1999) Localization of micro-
and intermediate filaments in non-pregnant uterus and placenta of the mink sug-
gests involvement of maternal endothelial cells and periendothelial cells in blood
flow regulation. Anat. Embryol. 200, 253–263.
28. Blankenship, T. N. and King, B. F. (1993) Developmental changes in the cell
columns and trophoblastic shell of the macaque placenta: an immunohistochemi-
cal study localizing type IV collagen, laminin, fibronectin and cytokeratins. Cell
Tissue Res. 274, 457–466.
29. Blankenship, T. N., Enders, A. C., and King, B. F. (1993) Trophoblastic invasion
and the development of uteroplacental arteries in the macaque: immunohis-
tochemical localization of cytokeratins, desmin, type IV collagen, laminin, and
fibronectin. Cell Tissue Res. 272, 227–236.
23_Pfarrer_323_336_F 335 8/29/05, 11:22 AM

23_Pfarrer_323_336_F 336 8/29/05, 11:22 AM
Human Placental Investigation 337
24
Molecular Markers for Human Placental Investigation
Berthold Huppertz
Summary
The human placenta is a source for a variety of growth factors, hormones, and other pro-
teins. The cellular source of the proteins can be best determined by immunohistochemistry.
Furthermore, immunohistochemistry can also be used to identify a specific cell type and to
differentiate it from other types of cells. Thus, there is the need for specific markers of those
cell types that are present in the placenta. In this chapter, the basic protocols for the identifica-
tion of proteins in a tissue section are described. This chapter focuses on methods that are
available in the majority of laboratories, and therefore concentrates on methods that are used
together with light microscopy.
Key Words: Immunohistochemistry; morphology; antibody; M30; TUNEL; ssDNA;
marker; syncytiotrophoblast; cytotrophoblast; Hofbauer cell; macrophage; fibroblast;
myofibroblast; endothelial cell.
1. Introduction
With its specific location between mother and fetus, the placenta comes in
direct contact with maternal blood and endometrial tissues. Although defined
as an allograft that should be recognized as nonself by the mother, the placenta
is normally not rejected but remains within the uterus for 40 wk. The placenta
is composed of different tissues, comprising (1) the villous trophoblast as the
epithelial cover of the villous tree, (2) the villous stroma with mesenchymal
cells, fetal vessels, and free connective tissue cells such as macrophages
(Hofbauer cells), mast cells and plasma cells, (3) fetal blood that enters the
placenta via the two umbilical arteries and leaves the placenta via the umbilical
vein. Another tissue derived from trophoblast is the extravillous trophoblast,
which invades maternal tissues, finally reaching the walls of spiral arteries as
deep as the inner third of the myometrium.
Both trophoblast populations, villous and extravillous trophoblast, are de-
rived from the trophoblast layer of the blastocyst and maintain all characteris-
337
24_Huppertz_337_350_F 337 8/29/05, 11:22 AM

338 Huppertz
tics of epithelial cells. The villous stroma develops from the extraembryonic
mesenchyme, which is derived from the two-layered embryonic disk and fur-
ther develops a mesodermal phenotype. The first free connective cells, the
Hofbauer cells, directly derive from the placental mesenchymal cells within the
villous stroma, which is also true for the endothelial and first blood cells (1).
Basic protocols for the identification of proteins in a tissue section are
described in this chapter. In order to focus on methods that are available in
the majority of laboratories, this chapter will concentrate on methods that are
used together with light microscopy. In some cases, the use of electron micros-
copy is the better choice, e.g., if the localization of a membrane protein in the
plasma membranes of two adjacent cells must be defined.
Immunohistochemistry of placental tissues is often used to differentiate dif-
ferent cell types from each other. Thus there is a need for specific markers of
the cell types present in the placenta. In Table 1, a list of potential markers for
various cell types within the human placenta can be found. Protocols frequently
used to identify apoptotic cells within the human placenta are also provided
(2,3).
2. Materials
2.1. General Histology
2.1.1. Tissue Preparation
1. Transfer container for the transport of the placenta (isolated to keep the placenta
cool).
2. Plastic bags to keep a term placenta during transport.
3. Small plastic vial to keep a first-trimester placenta during transport.
4. Sterile scalpels, scissors and foreceps.
5. 250-mL bottles.
2.1.2. Fixation
1. 4% neutrally-buffered formalin solution. For 1 L: solve 4 g NaH2PO4 and 6.5 g
Na2HPO4 in 900 mL double distilled water, add 100 mL 37% formaldehyde solu-
tion (Merck, Darmstadt Germany; for analysis, stabilized with about 10% metha-
nol) and adjust pH to 7.0 with HCl or NaOH. Filter fixative and store at 4°C.
2. 250-mL bottles.
3. Embedding cassettes.
2.1.3. Embedding and Sectioning

1. Alcohol series: acetone, 100% ethanol, 96% ethanol, 90% ethanol, 80% ethanol,
70% ethanol.
2. Paraffin with a melting temperature between 52°C and 54°C (e.g., Paraplast x-tra;
Fluka, Seelze, Germany).
24_Huppertz_337_350_F 338 8/29/05, 11:22 AM

Table 1
Molecular Markers for Human Placental Investigation
Tissue/cell type Marker Reference
1. villous trophoblast cytokeratin 4 (review)
cytokeratin 18 neoepitope (M30) 2
syncytin 5
CD10 6
AFP (first trimester) 7
GB25 8
PLAP (third trimester) 9
CD133 10
A. syncytiotrophoblast hCG 11
cadherin-11 12
endoglin/CD105 13,14
β-microglobulin, HFE (third trimester) 15
Thomsen-Friedenreich antigen 16
mucin 1 16
PPAR-γ 17
hPL 4 (review)
B. cytotrophoblast HAI-1 14,18
E-cadherin 12
2. villous stromal cells CD9, CD45, vimentin 4 (review)
CTLA-4 19
A. mesenchymal cells Vimentin 20,21
B. fibroblasts vimentin, desmin, α-sm-actin 20,22
cytokeratin 8/18 (few cells) 23
C. myofibroblasts/ vimentin, desmin, α-sm-actin, α-sm-actin 20–22
smooth muscle cells sm-myosin 24
caveolin-1, -2 25
D. endothelial cells CD34 26,27
caveolin-1, -2 25
von Willebrand factor, Ulex europaeus lectin 28
1F10, PAL-E 29,30,31
E. macrophages CD68 32
CD14 (anti-leu-M3) 33,34
Occurrence of markers may be of importance for those who want to isolate and culture the respective cells or
simply need a marker to identify cells. Abbreviations and CD numbers: 1F10, monoclonal antibody recognizing an
unknown endothelial protein; AFP, alpha-fetoprotein; CD9, p24/motility-related protein-1 (MRP-1), DRAP-27, type
III membrane protein, 228 amino acids; CD10, common acute lymphoblastic leukemia antigen (CALLA), EC
3.4.24.11, neprilysin, enkephalinase, gp100, neutral endopeptidase metalloendopeptidase (NEP); CD14, lipopolysac-
charide receptor (LPS-R), anchored by glycosylphosphatidylinositol (GPI), 356 amino acids; CD34, gp105-120,
heavily glycosylated type I transmembrane protein, 385 amino acids; CD45, leukocyte common antigen (LCA),
tyrosine phosphatase (EC 3.1.34), long single chain type I transmembrane protein, 1120-1281 amino acids; CD68,
macrosialin, gp110, type I transmembrane glycoprotein, 354 amino acids; CD105, endoglin, type I integral mem-
brane protein, 633 amino acids; CD133, AC133, PROML1, hematopoietic stem cell antigen, pentaspan transmem-
brane glycoprotein, 865 amino acids; CTLA-4, cytotoxic T lymphocyte-associated protein-4; GB25, monoclonal
antibody recognizing an unknown trophoblast protein; HAI-1, hepatocyte growth factor activator inhibitor type 1;
hCG, human choriogonadotropin; HFE, hemochromatosis protein; hPL, human placental lactogen; PAL-E, mono-
clonal antibody recognizing an unknown endothelial protein; PLAP, placental alkaline phosphatase; PPAR-γ, per-
oxisome proliferator-activated receptor γ.
24_Huppertz_337_350_F 339 8/29/05, 11:23 AM

340 Huppertz
3. 250-mL bottles.
4. Incubator adjusted to 54°C.
5. Heating plate adjusted to 54°C.
6. Embedding molds.
7. Forceps.
8. Microtome.
9. Glass slides.
2.2. Standard Immunohistochemistry

1. Humidified chamber.
2. Staining dishes.
3. Xylene.
4. Ethanol.
5. 1.25 M Tris stock solution. For 100 mL: dissolve 15.1 g Tris base in 50 mL
double distilled water, adjust pH to 7.6 with HCl and add distilled water up to
100 mL. Store at 4°C.
6. Tris buffer. The ready-to-use buffer is achieved by a one-twenty-fifth dilution of
the stock solution with distilled water leading to a final concentration of 0.05 M
Tris.
7. 0.1 M Acetate buffer. For 100 mL: dissolve 1.36 g Na-acetate in 80 mL double
distilled water, adjust pH to 5.2 with 1 M, acetic acid and add distilled water up to
100 mL. Store at 4°C.
8. 3% H2O2 in methanol. Dilute the H2O2 solution (30%; Merck) one-tenth with
methanol just prior before use.
9. Block solution for unspecific binding sites. Take a serum from the species in
which the secondary antibody has been raised. Use this serum diluted one-twen-
tieth in Tris buffer and add 60 mg/mL bovine serum albumin (BSA).
10. Primary antibody solution. The primary antibody is diluted in Tris buffer con-
taining 15 mg/mL BSA.
11. Secondary antibody solution. The secondary antibody is diluted in Tris buffer
containing 125 mg/mL BSA.
12. Strepavidin conjugate. The streptavidin conjugate is diluted one-four hundredth:
add 10 µL streptavidin (Dako, Hamburg, Germany) to 4 mL Tris buffer.
13. 3-amino-9-ethylcarbazole (AEC) chromogen. Using the Zymed kit (Zymed,
Berlin, Germany) the following protocol has to be performed. Take one drop of
reagent A (substrate buffer 20X), one drop of reagent B (AEC chromogen 20X)
and one drop of reagent C (12% hydrogen peroxide; 0.6% 20X) and add all three
drops to 1 mL distilled water.
14. Glycerin gelatin.
15. Coverslips.
2.3. M30 Immunohistochemistry

1. Microwave oven.
2. Staining dishes.
24_Huppertz_337_350_F 340 8/29/05, 11:23 AM

4. 3% H2O2 (see Section 2.2.8.).
5. Acetic acid buffer. For 500 mL: solve 1 g acetic acid in 400 mL distilled water,
adjust pH to 6.0 with NaOH and add distilled water up to 500 mL.
6. Blocking solution: phosphate buffered saline (PBS) with 1% BSA and 0.1%
Tween 20.
7. PBS with 0.1% Tween 20.
8. Starting with the secondary antibody solution the materials are identical with the
materials for the standard immunohistochemistry (see Subheadings 2.2.11.–
2.2.15.).
2.4. Single-Stranded DNA Immunohistochemistry

1. Staining dishes.
3. Heating block (99°C).
4. 50 mL tubes.
5. Ice-cold 0.1 M HCl.
6. PBS with 0.2% Triton X-100 and 5 mM MgCl2.
7. PBS with 5 mM MgCl2.
8. PBS with 0.1% BSA.
9. PBS with 4.5% BSA.
10. 3% H2O2 (see Subheading 2.2.8.).
11. Starting with the secondary antibody solution the materials are identical with the
materials for the standard immunohistochemistry (see Subheading 2.2.11.–
2.2.13.).
2.5. Terminal Deoxynucleotidyl Transferase-Mediated dUTP Nick-End

Labeling Test (TUNEL Test)
1. Staining dishes.
3. Parafilm.
4. Terminal deoxynucleotidyl transferase (TdT)-FragEl kit from Calbiochem
(Darmstadt, Germany, Cat. No. QIA33) with some slight adaptations.
5. Alcohol series: Xylene, 100% ethanol, 90% ethanol, 80% ethanol, 70% ethanol.
6. Tris-buffered saline (TBS) : 20 mM Tris, pH 7.6, 140 mM NaCl.
7. Tris buffer : 10 mM Tris, pH 8.0.
8. Proteinase K solution: add 20 µg/mL proteinase K to Tris buffer (see Subhead-
ing 2.5.7.).
9. 3% H2O2 in methanol (see Subheading 2.2.8.).
10. Equilibration buffer. 200 mM Na-cacodylate, 30 mM Tris, 0.3 mg/mL BSA,
0.75 mM CoCl2, pH 6.6.
11. TdT labeling solution. Ready-to-use solution provided with the kit.
12. TdT enzyme. Ready-to-use solution provided with the kit.
13. Stop solution. 0.5 M ethylenediamine tetraacetic acid (EDTA), pH 8.0.
24_Huppertz_337_350_F 341 8/29/05, 11:23 AM

342 Huppertz
14. Blocking buffer. PBS with 4% BSA.

15. Peroxidase-streptavidin conjugate solution. Dilute the stock solution (50X) one-
fiftieth with blocking buffer.
16. 3,3'-Diaminobenzidine tetrahydrochloride (DAB) solution. Dissolve 1 tablet of
DAB and I tablet of H2O2/urea in 1 mL of tab water.
17. Counterstain solution methyl green.
3. Methods
3.1. General Methods
3.1.1. Tissue Preparation
1. Directly after delivery or termination of pregnancy, take the placental material
and put it into a plastic bag (third-trimester material) or into a small plastic vial
(first-trimester material). Place the bag/vial on ice in an isolated container and
transfer the placenta to the laboratory (see Note 1).
2. Place the placenta on a tray and cut out the tissues used for your experiments (see
Note 2). Take a scalpel or scissors and cut pieces with a maximal width of 5 mm
from the placenta. The other dimensions should be in the range of 1–3 cm (see
Note 3).
3.1.2. Fixation (see Note 4)

Pour about 200 mL of fixative into a 250-mL bottle, place the placental
specimens in prelabeled embedding cassettes, and transfer the cassettes into
the 4% formalin solution. Close the bottle with a lid or with parafilm and fix
the specimens overnight at room temperature (see Note 5).
3.1.3. Embedding
1. Dehydrate the samples before embedding into paraffin. Pour the alcohol series in
250-mL bottles and place the embedding cassettes containing the specimens in
the 70% ethanol bottle for 24 h, in the 80% ethanol bottle for 24 h, in the 96%
ethanol bottle for 12 h, in the 100% ethanol bottle for 12 h (twice), and in the
acetone bottle for 2 h (twice) (see Note 6).
2. Then place the specimens into prewarmed paraffin and leave it at 54°C overnight
(see Note 7).
3. The next day transfer the specimens into fresh prewarmed paraffin for 12 h at
54°C and replace the paraffin for an overnight incubation at 54°C.
4. Finally, pour prewarmed paraffin into the embedding molds (on a heating plate;
54°C), remove the tissues from the cassettes and place the specimens inside the
paraffin in the molds.
5. Place the molds at room temperature until the paraffin is hard and remove the
paraffin blocks from the molds.
6. Cut sections from the paraffin block (3–6 µm thick) and place them on glass
slides.
24_Huppertz_337_350_F 342 8/29/05, 11:23 AM

3.2. Standard Protocol for Immunohistochemistry (see Note 8)

1. To rehydrate and deparaffinize the sections (on glass slides), use an alcohol series
with xylene two times (10 min each) and 100% ethanol two times (5 min each).
2. Transfer the specimens into the 3% H2O2 solution for 10 min to block endo-
genous peroxidase activity (see Note 9).
All of the following incubation steps are performed at room temperature in a
humidified chamber.
3. Wash the slides two times with Tris buffer (5 min each) and incubate them for
10 min with blocking solution (see Note 10).
4. Without additional washing, apply the primary antibody solution and incubate
the slides for 60 min (see Note 11).
5. Wash the slides three times with Tris buffer (5 min each) and incubate them for
30 min with the secondary antibody solution.
6. Wash the slides three times with Tris buffer (5 min each) and incubate them for
10 min with the streptavidin conjugate.
7. Wash the slides briefly with Tris buffer and incubate the slides for 10 min (see
Note 12) with the chromogenic substrate AEC (see Note 13).
8. Wash the slides briefly with Tris buffer and cover the section with glycerin gela-
tin using a coverslip.
3.3. Specific Protocol for M30 (see Note 14)

1. To rehydrate and deparaffinize the sections (on glass slides), use an alcohol se-
ries with xylene two times (10 min each) and 100% ethanol two times (5 min
each).
2. Transfer the specimens into the 3% H2O2 solution for 10 min to block endo-
genous peroxidase activity (see Note 9).
3. Prewarm acetic acid buffer in a microwave oven until cooking. Put the sections
into the hot buffer and place them in the microwave oven. Warm at 750 W until
the buffer is cooking again, then turn the microwave oven to “warming” (about
100 W) for another 15 min. Remove the slides from the microwave oven and let
them cool down to room temperature on the bench.
All of the following incubation steps are performed at room temperature in a
humidified chamber.
4. Wash the sections three times with PBS (5 min each) and incubate them in block-
ing solution for 10 min.
5. Dilute the primary M30 antibody 1:100 in blocking solution and incubate the
sections for 1 h.
6. Wash the slides two times with PBS with 0.1% Tween 20 (2 min each).
7. Beginning with the secondary antibody the protocol is identical with the general
immunohistochemistry protocol (see Subheadings 3.2.5.–3.2.8.).
24_Huppertz_337_350_F 343 8/29/05, 11:23 AM

344 Huppertz
3.4. Specific Protocol for Single-Stranded DNA (see Note 15)

1. To rehydrate and deparaffinize the sections (on glass slides), use an alcohol series
with xylene two times (10 min each) and 100% ethanol two times (5 min each).
2. Cover the sections with PBS for 5 min and wash them two times with double
distilled water (2 min each).
3. Incubate the sections in ice-cold 0.1 M HCl for 15 min and wash two times with
water and one time with PBS (2 min each).
4. Incubate the sections in PBS containing 0.2% Triton X-100 and 5 mM MgCl2 for
5 min.
5. Transfer the slides into 50-mL tubes filled with 30 mL PBS containing 5 mM
MgCl2. place the tubes into a prewarmed water bath and heat them for 5 min at
99°C. Take the slides out of the tubes directly and put them into a new tube filled
with the same buffer prechilled to 4°C and incubate for 10 min on ice.
All following incubation steps are performed at room temperature in a humidi-
fied chamber.
6. Incubate the sections in 3% H2O2 for 5 min (see Note 9), wash two times with
PBS, and incubate in PBS with 0.1% BSA for 30 min.
7. Wash the slides with PBS and incubate the sections with the primary F7-26 anti-
body (10 µg/mL in PBS with 4.5% BSA; 100 µL per section) for 15 min.
8. Wash the slides two times with PBS and use a secondary antibody that recog-
nizes immunoglobulin (Ig)M (the primary antibody). Beginning with the second-
ary antibody, the protocol is identical with the general immunohistochemistry
protocol (see Subheadings 3.2.5.–3.2.8.).
3.5. Specific Protocol for the TUNEL Test (see Note 16)
The use of a commercial kit with some minor adaptations is described (see
Note 17).
1. Rehydrate and deparaffinize the sections (on glass slides) in an alcohol series
with two times xylene (5 min each), two times 100% ethanol (5 min each), one
time 90% ethanol (3 min), one time 80% ethanol (3 min), one time 70% ethanol
(3 min), and a short washing step in Tris buffer.
For all of the following steps of the protocol, it is important not to over
incubate the specimens, otherwise false positives cannot be excluded (see Note
18). It is also very important not to let the specimens dry out during or between
any steps.
2. Permeabilize the specimens with proteinase K solution for 20 min at room tem-
perature and rinse specimens in TBS (see Note 18).
3. Inactivate endogenous peroxidases by incubating the specimens in 3% H2O2 for
5 min at room temperature and rinse them with TBS (see Note 19).
4. Cover the specimens with equilibration buffer for 20 min.
24_Huppertz_337_350_F 344 8/29/05, 11:23 AM

5. During this time prepare the TdT reaction mix on ice. Mix 57 µL TdT labeling
reaction mix with 3 µL TdT enzyme for each specimen and apply the solution
onto the specimens (see Note 20). The specimens are incubated in a humidified
chamber for 90 min at 37°C.
6. Rinse the specimens with TBS buffer and apply the stop solution for 5 min at
room temperature.
7. Rinse the specimens with TBS buffer again and cover them with blocking buffer
for 10 min at room temperature.
8. Apply the peroxidase-streptavidin conjugate onto the specimens for 30 min at
room temperature in a humidified chamber.
9. Wash the specimens in TBS buffer and apply 100 µL DAB solution onto the
specimens. A brown precipitate in apoptotic nuclei will appear in 10–15 min (see
Notes 21 and 22).
10. Rinse slides with distilled water and cover the specimens with 100 µL methyl
green for 3 min to counter stain the nuclei.
11. Wash the specimens with distilled water and cover the section with glycerin gela-
tin using a coverslip.
4. Notes
1. Normally, there is a delay between delivery of the placenta and availability and
usage of this material in the laboratory. There are a few scientists who are able to
obtain this material fresh from delivery and fix or freeze it within a few minutes.
But mostly, there is at least a 10- to 30-min gap between delivery room and labo-
ratory as a result of the distance between both rooms. A term placenta can be kept
on ice (without direct contact) for this time without additional solutions. A first-
trimester placenta is kept in a small vial and normally also does not need addi-
tional solutions for this duration of storage.
2. When cutting pieces of the placenta, take care not to destroy the fragile villi
inside the placenta. If you use foreceps to hold the tissue during cutting, take the
side that will not be used for fixation. When taking the piece of tissue of interest,
hold it at the edge without compressing the villous tissue.
3. Fixation in formalin solutions requires a minimal diffusion distance of the fixa-
tive. The same is valid if you freeze the samples in liquid nitrogen; the thicker the
sample, the longer it takes to freeze it; thus, with thicker samples, generation of
ice crystals will destroy the tissue. Therefore, samples obtained from the pla-
centa should have a maximal width of 5 mm. Other dimensions may be chosen
“without limitations,” e.g., if samples from a term placenta are required, then a
sample covering the whole thickness of the placenta can be obtained, but one
must keep in mind that the width of one side is restricted to 5 mm. One must keep
in mind that fixation of the samples is performed with embedding cassettes. This
will restrict the size of the samples to about 3 × 1–2 × 0.5 mm.
4. Tissues can be shock-frozen in liquid nitrogen or fixed in formalin for paraffin
embedding. Because most of the sections that are available in pathology depart-
ments and other archives are paraffin sections, this chapter concentrates on the
fixation and handling of those materials.
24_Huppertz_337_350_F 345 8/29/05, 11:23 AM

346 Huppertz
5. Fixation in formalin should not exceed 24 h. Longer fixation times lead to loss of
immunoreactivity; also, other fixatives may reduce the binding capacity of anti-
bodies used in immunohistochemistry. Smaller samples from a first-trimester
placenta or from villous explant cultures need much shorter fixation times. Very
small samples (up to 3 mm in diameter) only require 1 to 2 h of fixation.
6. After fixation, the samples must be dehydrated before embedding into paraffin.
The times for the alcohol series have to be adapted depending on the volume of
the samples. For small sample sizes, the following times are sufficient: 45 min in
70% ethanol (or storage for longer times), 45 min in 95% ethanol, three times
45 min in 100% ethanol, three times 45 min in acetone, 60 min in paraffin (or
overnight).
7. The higher the temperature during embedding into paraffin, the worse is the anti-
genic stability. Thus, antigenicity decreases, and some antibodies may not be
able to bind to the altered antigens. Therefore, the use of low-melting paraffin is
recommended. Paraffin with melting temperatures between 52°C and 54°C (e.g.,
Paraplast x-tra; Fluka, 76259) are recommended; however, paraffin with melting
temperatures up to 58°C may be used.
8. Any detection protocol or kit may be used here. A variety of protocols are used
for immunohistochemistry and it is always necessary to adapt a protocol to the
needs of the respective laboratory. A representative protocol used in our labora-
tory is presented.
9. Incubation times of the H2O2 solution to block endogenous peroxidase activity
have to be adapted according to the concentration of H2O2. Incubation with a 3%
H2O2 solution requires only 5–10 min incubation whereas a 0.3% H2O2 solution
requires an incubation time of 30 min. The H2O2 solution should always be pre-
pared and used fresh.
10. Normally, Tris buffer is used in all steps, but other buffers such as PBS may be
used. For some specific antibodies and staining procedures, alternative protocols
have to be used (see Subheadings 3.3. and 3.4.).
11. For most primary antibodies an incubation time of 60 min at room temperature is
sufficient to result in a clear staining with low background. But depending on the
antibody, changes of the times and temperature may be necessary. Another
often-used protocol combines incubation overnight at a temperature of 4°C.
Please note that conditions need to be optimized for every single antibody.
12. Incubation times for AEC may vary depending on the antibodies used. To stan-
dardize the protocol, a fixed incubation time (e.g., 10 min) should be used. But
sometimes it may be necessary to wait longer (up to 30 min) or to stop the incu-
bation already after a few minutes, depending on when the color development is
complete.
13. Two chromogens are classically used in immunohistochemistry (AEC and DAB)
although other substrates or fluorochromes can be used. In paraffin sections, espe-
cially from archives, the use of the classical chromogens is superior to the use of
other substrates. Both AEC and DAB are chromogens used for staining peroxi-
dase labeled compounds in immunohistochemistry. AEC produces an insoluble
24_Huppertz_337_350_F 346 8/29/05, 11:23 AM

end product that is red in color while DAB produces a brown water insoluble end
product.
14. The monoclonal antibody with the clone number M30 specifically recognizes an
epitope of the cytokeratin 18 protein generated during cleavage by caspases.
Thus, this antibody should recognize only cytokeratin 18 cleavage products.
However, under denaturing conditions (Western blots), it is apparent the M30
antibody also recognizes intact cytokeratin 18. Denaturation of native cytokeratin
18 protein may occur during from the increased temperatures used during embed-
ding into paraffin. Therefore, low temperatures are essential when using the M30
antibody. It must be kept in mind that cytokeratin 18 is only found in some epi-
thelia. Thus, this antibody cannot be used to detect apoptosis in cells of mesoder-
mal origin such as villous stromal cells.
15. Similarly to the M30 antibody that recognizes a caspase-generated cleavage prod-
uct of cytokeratin 18, the F7-26 antibody (Alexis Corporation, Lausen, Switzer-
land) specifically binds to apoptotic single-stranded DNA (ssDNA). The stability
of apoptotic DNA is decreased during thermal denaturation due to proteolysis of
DNA-bound proteins by effector caspases. The antibody is specific for ssDNA
and does not bind to double-stranded DNA. The advantage of this antibody over
the widely used TUNEL test (discussed later) is its high sensitivity to apoptosis
with nearly no staining of necrotic nuclei.
16. Similar to the antibodies M30 and F7-26, the TUNEL test is used to identify late
apoptotic cells. The TUNEL test is not based on an antibody recognizing its spe-
cific antigen. This time, an enzyme (TdT) recognizes nicks inside the DNA
strands and links nucleotides to the ends. Using an excess of labeled nucleotides,
the nicks can be visualized.
17. A representative protocol is provided that generally gives reproducible and con-
vincing results. The protocol uses the TdT-FragEl kit from Calbiochem (cat. no.
QIA33) with some slight adaptations. We have tested a variety of different kits
and have tried to create our own protocol. We have found that the protocol pro-
vided with the above mentioned kit is satisfactory, with some minor modifica-
tions.
18. Care should be taken using this assay since it is very easy to produce false posi-
tive results. Not only does the TUNEL test also stain necrotic DNA, but, depend-
ing on the protocol used, even mitotic cells or cells in interphase may be labeled.
Thus, close inspection of the morphology of the positive nuclei is always recom-
mended. A TUNEL positive nucleus should always display a morphology that is
condensed (higher density of chromatin), irregular in shape and smaller in size.
Large ovoid nuclei that show little densely packed chromatin but that show
TUNEL positivity are most likely a false positive. And even within one section
and one incubation, there may be areas where the staining is reliable, whereas the
area next to it shows 100% positive nuclei (which is clearly false positive).
19. The digestion with proteinase K is one of the crucial steps of this protocol.
Overincubation for only a few min results in a dramatic increase in false-posi-
tive nuclei. Thus, one must check first whether the time given in this protocol is
24_Huppertz_337_350_F 347 8/29/05, 11:23 AM

348 Huppertz
appropriate or whether the incubation time with proteinase K has to be modified.

20. If the specimen is too large to be covered by 60 µL, a piece of parafilm can be
used to cover the specimen. A further advantage of this method is that the section
will not run dry during incubation.
21. A measure of the reliability of the TUNEL stain may be the quantification of the
percentage of positive nuclei. One must keep in mind that a late apoptotic cell
that already displays DNA strand breaks will be present within a tissue for a few
min or up to a few h. Then it will be phagocytozed by macrophages or neighbor-
ing cells. Thus, the percentage of positive nuclei should be in the range of a few
percent (for high rates of apoptosis) down to 0.1% and less (for normal to low
rates of apoptosis). If a cell culture is analyzed, apoptotic values may be higher
because here are no other cells to remove the dying cells from the culture.
22. Even if the reaction has resulted in a seemingly optimal result, one must remem-
ber that the TUNEL-positive nuclei are really smaller and more condensed than
the unstained nuclei. Thus, morphology is a direct control of the TUNEL staining.
Acknowledgments
The author thanks Dr. Maria Kokozidou for critically reading the manu-
script and Uta Zahn for help with the exact reading of the protocols.
References
1. Demir, R., Kayisli, U. A., Seval, Y., et al. (2004) Sequential expression of VEGF
and its receptors in human placental villi during very early pregnancy: differences
between placental vasculogenesis and angiogenesis. Placenta 25, 560–572.
2. Kadyrov, M., Kaufmann, P., and Huppertz, B. (2001) Expression of a cytokeratin
18 neo-epitope is a specific marker for trophoblast apoptosis in human placenta.
Placenta 22, 44–48.
3. Huppertz, B., Kingdom, J., Caniggia, I., et al. (2003) Hypoxia favours necrotic
versus apoptotic shedding of placental syncytiotrophoblast into the maternal cir-
culation. Placenta 24, 181–190.
4. Frank, H. G., Morrish, D. W., Potgens, A., Genbacev, O., Kumpel, B., and
Caniggia, I. (2001) Cell culture models of human trophoblast: primary culture of
trophoblast—a workshop report. Placenta 22(Suppl A), S107–S109.
5. Mi, S., Lee, X., Li, X., et al. (2000) Syncytin is a captive retroviral envelope
protein involved in human placental morphogenesis. Nature 403, 785–789.
6. Ordi, J., Romagosa, C., Tavassoli, F. A., et al. (2003) CD10 expression in epi-
thelial tissues and tumors of the gynecologic tract: a useful marker in the diag-
nosis of mesonephric, trophoblastic, and clear cell tumors. Am. J. Surg. Pathol.
27, 178–186.
7. Lafuste, P., Robert, B., Mondon, F., et al. (2002) Alpha-fetoprotein gene expres-
sion in early and full-term human trophoblast. Placenta 23, 600–612.
8. Hsi, B. L. and Yeh, C. J. (1986) Monoclonal antibody GB25 recognizes human
villous trophoblasts. Am. J. Reprod. Immunol. Microbiol. 12, 1–3.
24_Huppertz_337_350_F 348 8/29/05, 11:23 AM

9. Leitner, K., Szlauer, R., Ellinger, I., Ellinger, A., Zimmer, K.P., and Fuchs, R.
(2001) Placental alkaline phosphatase expression at the apical and basal plasma
membrane in term villous trophoblasts. J. Histochem. Cytochem. 49, 1155–1164.
10. Pötgens, A. J., Bolte, M., Huppertz, B., Kaufmann, P., and Frank, H. G. (2001)
Human trophoblast contains an intracellular protein reactive with an antibody
against CD133—a novel marker for trophoblast. Placenta 22, 639–645.
11. Peleg, D., Peleg, A., and Shalev, E. (2000) Immunodetection of living tropho-
blast. Isr. Med. Assoc. J. 2, 821–822.
12. MacCalman, C. D., Furth, E. E., Omigbodun, A., Bronner, M., Coutifaris, C., and
Strauss, J. F. 3rd. (1996) Regulated expression of cadherin-11 in human epithelial
cells: a role for cadherin-11 in trophoblast-endometrium interactions? Dev. Dyn.
206, 201–211.
13. Dagdeviren, A., Muftuoglu, S. F., Cakar, A. N., and Ors, U. (1998) Endoglin (CD
105) expression in human lymphoid organs and placenta. Anat. Anz. 180, 461–469.
14. Pötgens, A. J., Kataoka, H., Ferstl, S., Frank, H. G., and Kaufmann, P. (2003) A
positive immunoselection method to isolate villous cytotrophoblast cells from first
trimester and term placenta to high purity. Placenta 24, 412–423.
15. Leitner, K., Ellinger, A., Zimmer, K. P., Ellinger, I., and Fuchs, R. (2002) Local-
ization of beta 2-microglobulin in the term villous syncytiotrophoblast.
Histochem. Cell Biol. 117, 187–193.
16. Jeschke, U., Richter, D. U., Hammer, A., Briese, V., Friese, K., and Karsten, U.
(2002) Expression of the Thomsen-Friedenreich antigen and of its putative carrier
protein mucin 1 in the human placenta and in trophoblast cells in vitro. Histochem.
Cell Biol. 117, 219–226.
17. Tarrade, A., Schoonjans, K., Guibourdenche, J., et al. (2001) PPAR gamma/RXR
alpha heterodimers are involved in human CG beta synthesis and human tropho-
blast differentiation. Endocrinology 142, 4504–4514.
18. Kataoka, H., Meng, J. Y., Itoh, H., et al. (2000) Localization of hepatocyte growth
factor activator inhibitor type 1 in Langhans’ cells of human placenta. Histochem.
Cell Biol. 114, 469–475.
19. Kaufman, K. A., Bowen, J. A., Tsai, A. F., Bluestone, J. A., Hunt, J. S., and Ober,
C. (1999) The CTLA-4 gene is expressed in placental fibroblasts. Mol. Hum.
Reprod. 5, 84–87.
20. Kohnen, G., Kertschanska, S., Demir, R., and Kaufmann, P. (1996) Placental vil-
lous stroma as a model system for myofibroblast differentiation. Histochem. Cell
Biol. 105, 415–429.
21. Demir, R., Kosanke, G., Kohnen, G., Kertschanska, S., and Kaufmann, P. (1997)
Classification of human placental stem villi: review of structural and functional
aspects. Microsc. Res. Tech. 38, 29–41.
22. Kohnen, G., Castellucci, M., His, B. L., Yeh, C. J., and Kaufmann, P. (1995) The
monoclonal antibody GB 42—a useful marker for the differentiation of
myofibroblasts. Cell Tissue Res. 281, 231–242.
23. Haigh, T., Chen, C., Jones, C. J., and Aplin, J. D. (1999) Studies of mesenchymal
cells from 1st trimester human placenta: expression of cytokeratin outside the
trophoblast lineage. Placenta 20, 615–625.
24_Huppertz_337_350_F 349 8/29/05, 11:23 AM

350 Huppertz
24. Graf, R., Matejevic, D., Schuppan, D., Neudeck, H., Shakibaei, M., and Vetter, K.
(1997) Molecular anatomy of the perivascular sheath in human placental stem
villi: the contractile apparatus and its association to the extracellular matrix. Cell
Tissue Res. 290, 601–607.
25. Lyden, T. W., Anderson, C. L., and Robinson, J. M. (2002) The endothelium but
not the syncytiotrophoblast of human placenta expresses caveolae. Placenta 23,
640–652.
26. Qiao, S., Nagasaka, T., and Nakashima, N. (1997) Numerous vessels detected by
CD34 in the villous stroma of complete hydatidiform moles. Int. J. Gynecol.
Pathol. 16, 233–238.
27. Fina, L., Molgaard, H. V., Robertson, D., et al (1990) Expression of the CD34
gene in vascular endothelial cells. Blood 75, 2417–2426.
28. Lang, I., Pabst, M. A., Hiden, U., et al. (2003) Heterogeneity of microvascular
endothelial cells isolated from human term placenta and macrovascular umbilical
vein endothelial cells. Eur. J. Cell Biol. 82, 163–173.
29. Goerdt, S., Steckel, F., Schulze-Osthoff, K., Hagemeier, H. H., Macher, E., and
Sorg, C. (1989) Characterization and differential expression of an endothelial cell-
specific surface antigen in continuous and sinusoidal endothelial, in skin vascular
lesions and in vitro. Exp. Cell Biol. 57, 185–192.
30. Schlingemann, R. O., Dingjan, G. M., Emeis, J. J., Blok, J., Warnaar, S. O., and
Ruiter, D. J. (1985) Monoclonal antibody PAL-E specific for endothelium. Lab.
Invest. 52, 71–76.
31. Lang, I., Hartmann, M., Blaschitz, A., Dohr, G., Skofitsch, G., and Desoye, G.
(1993) Immunohistochemical evidence for the heterogeneity of maternal and fe-
tal vascular endothelial cells in human full-term placenta. Cell Tissue Res. 274,
211–218.
32. Wetzka, B., Clark, D. E., Charnock-Jones, D. S., Zahradnik, H. P., and Smith, S. K.
(1997) Isolation of macrophages (Hofbauer cells) from human term placenta and
their prostaglandin E2 and thromboxane production. Hum. Reprod. 12, 847–852.
33. Bulmer, J. N., and Johnson, P. M. (1984) Macrophage populations in the human
placenta and amniochorion. Clin. Exp. Immunol. 57, 393–403.
34. Zaccheo, D., Pistoia, V., Castellucci, M., and Martinoli, C. (1989) Isolation and
characterization of Hofbauer cells from human placental villi. Arch. Gynecol.
Obstet. 246, 189–200.
24_Huppertz_337_350_F 350 8/29/05, 11:23 AM

Correlative Microscopy of Cryosections 351
25
Correlative Microscopy of Ultrathin Cryosections
in Placental Research
Toshihiro Takizawa and John M. Robinson
Summary
In this chapter, we describe procedures for correlative microscopy in immunocytochemical
studies on the human placenta. We have adapted ultrathin cryosections for use in high-resolu-
tion immunofluorescence microscopy (IFM) and for correlative immunocytochemical local-
ization using fluorescence and electron microscopy. High-resolution IFM of ultrathin
cryosections (50–100 nm in thickness) can be important because these physical sections mini-
mize the potential for false co-localization in the z-dimension. In addition, IFM of these sec-
tions affords greater sampling efficiency than does immunoelectron microscopy (IEM). These
ultrathin cryosections are compatible with conventional electron microscopy because a rela-
tively low-voltage electron beam can penetrate them. Thus, the same ultrathin cryosections of
placenta can be viewed in both fluorescence and electron microscopes. This latter point can be
of importance because it may be necessary to know the true size and shape of objects observed
by IFM; this can be determined best by IEM. Additionally, IEM can provide the “reference
space” lacking in IFM. The use of ultrathin cryosections is a powerful approach for placental
research, especially for the investigation of the in situ localization of antigens in the complex
structure of the human placenta.
Key Words: Placenta; villi; immunocytochemistry; correlative microscopy; immunofluo-
rescence; immunoelectron microscopy; ultrathin cryosections; Alexa; FluoroNanogold; silver
enhancement; caveolin-1α; early endosome antigen 1 (EEA1).
1. Introduction
Immunocytochemistry is a powerful and diverse set of methods directed
toward obtaining spatial and temporal information concerning the expression
and distribution of antigens in situ. Immunocytochemistry can provide unique
information that cannot be gained readily with biochemical, immunochemical,
or morphological methods alone. Ideally, the localization of antigens is highly
specific, with minimal background signal; this relies upon the inherent speci-
ficity of the antigen–antibody reaction. However, specificity can vary among
351
25_Takizawa_351_370_F 351 8/29/05, 11:23 AM

352 Takizawa and Robinson
antibodies, ranging from those with exquisite specificity to those that cross-
react with molecules other than the one to which the antibody was generated. It
is therefore prudent to test the specificity of antibodies in as many ways as
possible to determine if a particular antibody will be useful for immunocy-
tochemistry. Another important consideration in all immunocytochemical
experiments is the use of proper control reactions. This is crucial in order
to prevent spurious immunolocalization so that the genuine antigen distribu-
tion is ascertained. A delineation of the range of controls for immunocy-
tochemical preparations will not be given due to space limitations. The reader
is referred to the work of other authors who have dealt with this issue directly
(1,2). In addition to antibody specificity and the use of proper controls, there
are conditions that should be met before completely successful immunocy-
tochemistry can be achieved. These include: (a) preservation of immunoreac-
tivity; (b) retention of antigens in the proper location; (c) retention of morphological
detail (this is especially the case for electron microscopy); (d) maintenance of
equal accessibility of antibodies to antigen molecules at different locations
within the specimen; and (e) ability to label multiple antigens in the same
sample (3).
The nature of the biological sample to be examined by immunocytochemis-
try can set constraints on the methods that must be employed. Single cells,
such as those grown in tissue culture, require minimal sample preparation for
light microscope level immunocytochemistry since they are relatively thin and
are directly amenable to certain kinds of optical microscopy. However, these
same cells require special preparative methods for immunoelectron micros-
copy (IEM). Typically, single cells must be embedded and subsequently cut
into extremely thin sections (50–100 nm in thickness) before examination in
an electron microscope. Various plastic resins have been used to embed cells
for the preparation of conventional thin sections. Alternatively, these cells can
be “embedded” in concentrated sucrose solutions and subsequently frozen for
the preparation of ultrathin cryosections (see Subheading 3.1.) (4).
Intact tissue such as human placenta, on the other hand, presents additional
problems for immunocytochemical analysis. Diffusion of antibodies into intact
tissue as well as subsequent visualization is problematic. These difficulties can
be overcome by sectioning the tissue into thin slices. Useful sections range
from 5–20 µm in thickness for light microscopy to 50–100 nm thickness for
electron microscopy (EM). In order to cut usable sections, the tissue is gener-
ally embedded in a matrix of some sort prior to sectioning; paraffin or plastic
resins are commonly used to embed tissues. For retention of antigenicity within
tissues, preparation of frozen sections is generally less intrusive than other
preparative methods. Tissue to be frozen is routinely fixed and subsequently
infused with sucrose solutions prior to freezing. The use of sucrose has two pur-
25_Takizawa_351_370_F 352 8/29/05, 11:23 AM

Fig. 1. Flowchart of immunocytochemistry using cryosections. Use of cryostat sec-

tions (5–20 µm in thickness) is for conventional immunofluorescence microscopy
(IFM) (left). Employment of ultrathin cryosections (50–100 nm in thickness) is for use
in high-resolution IFM (middle) and for correlative immunocytochemical localization
using fluorescence and electron microscopy (EM) (right).
poses: it serves as a cryoprotectant to minimize ice crystal damage that may

occur during freezing and once frozen provides a uniform consistency within
the tissue that is desirable for obtaining good sections. Frozen sections (5–20 µm)
can be can be cut with the aid of a conventional cryostat while the much thinner
ultrathin cryosection (50–100 nm) can be cut with a cryoultramicrotome that is
cooled with liquid nitrogen (Fig. 1).
25_Takizawa_351_370_F 353 8/29/05, 11:23 AM

Ultrathin cryosections of cells and tissues have been used for IEM using
colloidal gold particles as the detection system for more than 20 yr. This approach,
sometimes called the Tokuyasu method, has been extremely valuable for detect-
ing the distribution of a number of antigens at the ultrastructural level (4).
Indeed, our understanding of some fundamental aspects of cell biology is
largely dependent on this methodology. However, this technique, like all tech-
niques, is subject to limitations. The exclusive use of ultrathin cryosections for
EM limits their utility as a result of the sampling limitations imposed by the
small amount of material that can be examined in the electron microscope at
any one time. The use of colloidal gold as the reporter system also imposes
limitations. Colloidal gold particles are discrete structures that can be counted
for quantitative analysis. However, various investigators have addressed the
question of whether the density of colloidal gold immunolabeling correlates
with antigen concentration. In such studies, the determination was that a one-
to-one relationship between colloidal gold particles and antigens did not occur
(5,6). It has been proposed that the greatest labeling efficiency achieved with
colloidal gold probes is 20% or less (7). The major reasons proposed for the
relative inefficiency of labeling with colloidal gold immunoprobes are: (a) poor
penetration of the colloidal gold into the section thus only the most superficial
antigens are detected and (b) steric hindrance effects (8). Another important
consideration in using colloidal gold is the recognition that labeling efficiency
varies with the size of the gold particles. Smaller gold particles lead to greater
labeling efficiency than do larger ones (i.e., 5-nm gold particles label more
efficiently than do 10-nm particles, and so forth) (9–11).
These limitations aside, ultrathin cryosections offer an excellent substrate
for immunocytochemical localization of antigens. One important consideration
is that they are thin enough for antibodies to readily penetrate throughout their
volume (in the absence of colloidal gold particles). We have used ultrathin
cryosections for high-resolution immunofluorescence microscopy (IFM) (Fig. 1
and Subheading 3.2.). These sections have a real advantage in eliminating out-
of-focus fluorescence since all of the fluorescence must come from within the
section. This is in contrast to using conventional sections of 5 µm in thickness
and then imaging them with a confocal microscope. In the confocal micro-
scope, the out-of-focus fluorescence signal is minimized by optical sectioning.
The resolution in the z-dimension usually achieved with confocal microscopy
of biological samples is about 500 nm (12). The ultrathin cryosections we
employ are about 100 nm in thickness (Figs. 2,3). Thus, the use of ultrathin
cryosections can minimize the possibility of false co-localization in two-color
IFM (Fig. 4) (13). In addition, in IFM the sampling efficiency is increased
when compared with IEM. These are real advantages that we have utilized in
our study of the distribution of a number of antigens in the human placenta.
25_Takizawa_351_370_F 354 8/29/05, 11:23 AM

Fig. 2 (see companion CD for color version). Diagram summarizing a model illus-
trating the advantages of using ultrathin cryosections for high-resolution immunofluo-
rescence microscopy (IFM). The left side shows a diagram of a terminal villus cut in
cross section. Three cell types are shown: endothelium (#1), pericyte (#2), and syncy-
tiotrophoblast (STB) (#3). The black and gray dots indicate two different structures
labeled with two different fluorochromes in IFM. The gray bar indicates the volume of
an ultrathin cryosection (about 100 nm). The right side shows the fluorescence signals
in the z-dimension (side view) and the x- and y-dimensions (top view). In 5 µm sec-
tions, the fluorescence signal from the individual structure may be stacked in the vol-
ume of the section; this may lead to false co-localization as indicated by the white
color. In confocal optical sections (approx 500 nm), the potential for false co-localiza-
tion is reduced when compared to thicker conventional cryostat sections. However,
separate small organelles (50–200 nm size range) that are labeled with two different
fluorochromes and that are stacked within the section may appear to be co-localized in
the same structure as indicated by the white color (arrowhead). In ultrathin
cryosections (100 nm or less in thickness), the possibility for false co-localization is
minimized further since small structures (50–200 nm in size) could occupy the entire
volume of the section as indicated by the gray color (arrowhead).
25_Takizawa_351_370_F 355 8/29/05, 11:23 AM

Fig. 3. Comparison between the immunofluorescence localization of early endosome

antigen (EEA)1 in 5-µm and 100-nm thick sections. (A) Immunofluorescence localization
of EEA1 in a 5-µm thick cryostat section of a terminal villus. The fluorescence signal is
primarily located in the syncytiotrophoblast (STB) layer (arrowheads). It is difficult to
identify individual EEA1-positive structures. Note that endothelial cells have little label-
ing for EEA1 (arrow). Capillary lumens are evident (*). (B) The differential interference
contrast (DIC) image of the same section shown in A is presented so that the morphology
of the tissue can be evaluated. The same labels (arrowheads, arrow, and asterisks) used in
A are presented here to provide reference points. (C) Immunofluorescence localization of
EEA1 in a 100-nm thick ultrathin cryosection of terminal villi. The fluorescence signal
indicating EEA1 is primarily in the apical portion of the STB in vesicle-like structures
(arrowheads) (13). EEA1-positive structures are also present in the endothelial cells (ar-
rows) and in stroma cells (stars) but are less abundant than in the STB. A capillary lumen
is indicated (*). (D) The DIC image of the same section shown in C is presented so that the
morphology of the section can be evaluated. The same labels (arrowheads, arrows, and
stars) used in C are presented to provide reference points. Further reference points are
provided by the fluorescence image of the 4',6-diamidino-2-phenylindole (DAPI)-stained
nuclei (n) that has been merged with the DIC image. Bar = 10 µm. All panels are at the
same magnification. (Figs. 3C,D reproduced with permission from Takizawa, T., Ander-
son, C. L., and Robinson, J. M. (2005) J. Immunol. 175, 2331–2339. Copyright 2005. The
American Association of Immunologists, Inc.)
25_Takizawa_351_370_F 356 8/29/05, 11:23 AM

25_Takizawa_351_370_F
357
Correlative Microscopy of Cryosections
Fig. 4 (see companion CD for color version). Double-label immunofluorescence microscopy (IFM) detection
of caveolin (CAV)-1α and early endosome antigen (EEA)1 on an ultrathin cryosection. (A) In a terminal villus
of the placenta, CAV-1α is localized to capillary endothelium (a star indicates the capillary lumen) and adjacent
pericytes (P) and is seen as small punctate structures (13,16,21). In the syncytiotrophoblast (STB) (arrowheads),
CAV-1α is not detected. (B) EEA1 is primarily in the apical portion of the STB in vesicle-like structures (arrow-
heads) (13). EEA1-positive structures are also present in the endothelium and pericytes (arrows) but are less
8/29/05, 11:23 AM
abundant than in the STB (arrowheads). (C) The differential interference contrast (DIC) image of the same
section shown in A and B illustrates the morphology of the section. The fluorescence image of the 4',6-diamidino-
2-phenylindole (DAPI)-labeled nuclei has been merged with the DIC image to facilitate orientation. The lumen
of the capillary (star) and the STB (arrowheads) are indicated. (D) The merged image shows the distribution of
357
CAV-1α and EEA1 and illustrates the relationship between the two IFM signals. The same labels (arrowheads,
arrows, P, and star) used in each panel are presented to provide reference points. Bar = 10 µm.
Ultrathin cryosections are also very useful for correlative fluorescence and
electron microscopy (Fig. 1 and Subheading 3.3.); that is imaging the same
exact structures (in the same ultrathin cryosection) by both fluorescence and
electron microscopy. This methodology is important becauses it can bridge the
resolution gap between fluorescence and electron microscopy. In our studies,
we have used a single reporter system that contains both a fluorochrome and a
gold-cluster compound. This probe is known as FluoroNanogold. In addition
to containing a fluorochrome and a gold probe in the same reagent, it has the
further advantage of being extremely small (approx 1.4 nm for the gold probe).
This probe appears to behave more like a fluorochrome-conjugated secondary
antibody than a colloidal gold-coupled immunoprobe. It appears to penetrate
fully into ultrathin cryosections (11). We have used this reagent in correlative
microscopy in which we first collect a fluorescence image and then following
a silver-enhancement reaction to enlarge the gold cluster compound, image the
same structures in an electron microscope (Fig. 5). This approach is valuable
when it is important to know the true size and shape of a structure and when it
is important to see the “reference space.” In IFM, the only structures evident
are those tagged with the fluorochrome; all other parts of the cell or tissue are
invisible under these conditions. In EM, on the other hand, all structures are seen
not just those positively labeled. Imaging this reference space may be vital for
understanding the positive immunolabeling pattern (13). The use of ultrathin
cryosections is essential in these types of experiments.
2. Materials
1. Carbon steel blades (Feather Safety Razor, Osaka, Japan).
2. Pink base plate wax (Coltene/Whaledent Inc., Cuyahoga Falls, OH).
3. 60 × 15 mm and 35 × 10 mm cell culture dishs (BD Falcon, Franklin Lakes, NJ).
4. Hardened filter paper (Grade 50, Whatman, Clifton, NJ).
5. SS-style tweezers (Ted Pella, Redding, CA).
6. Groove type specimen carrier pins (Mager Scientific, Dexter, MI).
7. Cryo P diamond knife (Diatome-US, Fort Washington, PA).
8. Reichert Ultracut E equipped with an FC 4D cryounit (Leica, Vienna, Austria).
9. Mouse anti-early endosome antigen (EEA)1 monoclonal antibody is available
from BD Transduction Laboratories (San Diego, CA). EEA1 is a 180-kDa coiled-
coil dimer that is crucial for endosome fusion (14).
10. Anti-peptide antibody specific for caveolin (CAV)-1α was generated in chickens.
Immunocytochemical characterization of this antibody has been reported (15).
11. Alexa Fluor 488 and 594 goat anti-chicken and goat anti-mouse immunoglobulin
(Ig)G as well as the ProLong antifade kit can be obtained from Molecular Probes
(Eugene, OR).
12. Biotin-labeled goat anti-mouse F(ab)' 2 antibody is available from Jackson
ImmunoResearch (West Grove, PA).
25_Takizawa_351_370_F 358 8/29/05, 11:23 AM

Fig. 5. The immunocytochemical localization of early endosome antigen (EEA)1

by correlative microscopy on the same ultrathin cryosection. (A) The immunofluores-
cence microscopy (IFM) localization of EEA1 in a terminal villus of the placenta using
FluoroNanogold (FNG) as the detection system. The section was collected on an elec-
tron microscopy (EM) grid; the grid bars (*) are not particularly evident in this
epifluorescence image. EEA1 is primarily in the apical portion of the syncytiotropho-
blast (STB) in vesicle-like structures (arrowheads). (Inset) Higher-magnification view
of the EEA1 indicated with the box in A. Individual fluorescence signals in the STB (# 1,
2, and 3) are indicated. Bar = 1 µm. (B) The differential interference contrast (DIC)
image of the same section shown in A illustrates the morphology of the ultrathin
cryosection. The opaque grid bars are evident (*). Note that diffraction around the grid
bars degrades a portion of the DIC image. A and B are at the same magnification. Bar
= 10 µm. (C) Higher-magnification fluorescence image of the inset of A. Individual
fluorescence signals in the STB are labeled #1–3. (D) An electron micrograph of the
same region shown in C illustrates the distribution of EEA1 detected with silver-en-
hanced FNG. Early endosome clusters tagged with FNGs are labeled #1–3. Note the
precise correspondence between the fluorescence spots and the FNG-labeled early
endosome clusters. Nucleus of the STB is evident (n). C and D are at the same magni-
fication. Bar = 500 nm. (Inset) Higher-magnification view of the FNG-labeled early
endosome cluster indicated with an arrow in D. Bar = 100 nm.
25_Takizawa_351_370_F 359 8/29/05, 11:23 AM

13. FNG: Alexa Fluor 594 FluoroNanogold (FNG)-streptavidin is available from

Nanoprobes (Yaphank, NY).
14. Paraformaldehyde fixative: 4% paraformaldehyde (Polysciences, Warrington,
PA) in 0.1 M sodium cacodylate buffer, pH 7.4, containing 5% sucrose. Fixative
is freshly prepared on the day of use.
15. Cacodylate buffer: 0.1 M sodium cacodylate buffer, pH 7.4, containing 5% sucrose.
16. 20% gelatin solution: 3 mL of 0.1 M sodium cacodylate buffer, pH 7.4, contain-
ing 5% sucrose is prewarmed at room temperature (22°C) and then added in a
glass test tube. 600 mg of gelatin (300 bloom, product No. G2500, Sigma-Aldrich,
St. Louis, MO) is added. The test tube is gently mixed with a vortex, incubated in
80°C hot water until gelatin is dissolved completely, and subsequently kept in a
water bath at 37°C for more than 5 min prior to use.
17. 2.3 M sucrose solution: 393.6 g of sucrose (Sigma-Aldrich) is added to a 500 mL
volumetric flask and then 0.1 M sodium cacodylate buffer, pH 7.4 is added to
make the volume 500 mL. After dissolving sucrose, the solution is aliquoted to
50 mL conical centrifuge tubes (BD Falcon, Bedford, MA) and stored at –20°C.
18. 0.75% gelatin–2.0 M sucrose pick-up solution: 377 mg of gelatin (300 bloom) is
added to 6.5 mL of 0.1 M sodium cacodylate buffer, pH 7.4 in a 100 mL beaker
and then dissolved on a hot plate at 60°C. 43.5 mL of 2.3 M sucrose solution
containing 0.05 % sodium azide (Sigma-Aldrich) is gradually added to the bea-
ker with continuous stirring in a water bath of a large Petri dish on the hot plate at
60°C and the gelatin–sucrose solution is kept at 60°C.
19. Phosphate-buffered saline (PBS): 137 mM NaCl, 2.7 mM KCl, 6.45 mM
Na2HPO4, 1.47 mM KH2PO4, 0.05% NaN3, pH 7.35
20. MFBS-PBS: PBS containing 1% nonfat dry milk (Carnation, Los Angeles, CA)
and 5% fetal bovine serum (Invitrogen, Carlsbad, CA).
21. Antifading medium: 1% n-propyl gallate (NPG, Kodak, Rochester, NY) and 45%
glycerol (Sigma-Aldrich) in PBS, pH 8.0. 10 mg of NPG is dissolved with 50 µL
of ethanol in a 1.5-mL microcentrifuge tube. 450 µL of glycerol and 500 µL of
PBS (NaN3 (-)) is added and mixed with a vortex. The pH is adjusted to 8.0–8.2
with 1 N NaOH and checked using pH indicator strips. PBS(NaN3(-)), PBS with-
out NaN3.
22. 2% glutaraldehyde in PBS (NaN3 (-)): Fixative is freshly prepared on the day of use.
23. 0.5 M MES (Sigma-Aldrich), pH 6.15 is prepared as a stock solution and stored at
4°C. The pH is adjusted without the use of HCl because chlorine should be removed.
24. 50 mM MES buffer: 0.5 M MES stock solution is diluted 1:10 with distilled water.
25. Gum arabic stock solution: 50 g of gum arabic powder (Sigma-Aldrich) is dis-
solved in 100 mL of distilled water in a 300 mL beaker; it takes 2 or 3 d to
dissolve it. When dissolved, the gum arabic solution is degassed with a vacuum
pump, aliquoted ca.11 ml to 15 mL conical centrifuge tubes (BD Falcon), and
then stored at –20°C.
26. NPG stock solution (10 mg/5 mL): 10 mg of NPG is dissolved with 250 µL of
ethanol in a 15 mL conical centrifuge tube with a vortex. 4.75 mL of distilled
water is added and mixed. This is freshly prepared on the day of use.
27. Silver lactate stock solution (10.95 mg/1.5 mL distilled water): 10.95 mg of sil-
ver lactate (Fluka, Ronkonkoma, NY) is added to 1.5 mL microcentrifuge tubes,
25_Takizawa_351_370_F 360 8/29/05, 11:23 AM

wrapped in aluminum foil and then stored in light tight box. When the silver
enhancement is ready, 1.5 mL of distilled water is added to a stock tube and then
mixed well in a darkroom with a sodium vapor safelight. This is freshly prepared
on the day of use.
28. NPG silver enhancement stock solution: 5 mL of Gum arabic stock, 2 mL of 0.5 M
MES stock, and 1.5 mL of NPG stock are combined and mixed well for 5 min.
29. Working NPG silver enhancement solution: 8.5 mL of NPG silver enhancement
stock solution is well mixed for 5 min under a room light and then 1.5 mL of the
silver lactate stock solution is mixed with the NPG silver enhancement stock
solution in a darkroom with a sodium vapor safelight. Immediately after mixing
for 1 min, the working NPG solution is applied to sections labeled with FNG.
30. Neutral fixer solution: 250 mM sodium thiosulfate (Sigma-Aldrich) and 20 mM
HEPES (Sigma-Aldrich), pH 7.4. The pH is adjusted with 1 N NaOH; the solu-
tion is stored at 4°C.
31. Reduced osmium fixative: 2% osmium tetroxide–1.6% potassium ferrocyanide
in 0.1 M cacodylate buffer, pH 7.4. This is freshly prepared on the day of use and
handled in a chemical fume hood.
32. Uranyl acetate (UA) solution: 4% uranyl acetate (Mallinckrodt, Paris, KY) in
distilled water. UA is wrapped in aluminum foil and then stored at 4°C. UA is
passed through a syringe filter with 0.2-µm pore size (Acrodisc, Pall Corp., Ann
Arbor, MI) before use.
33. 2% polyvinyl alcohol (PVA), and 0.0015% lead citrate–2% PVA (LC-PVA) so-
lutions: 120 mL of distilled water is boiled in a beaker and cooled at 4°C. 2.4 g of
PVA (MW 30-70k, Sigma-Aldrich, product no. P8136) is added to the beaker
and mixed well at 4°C. 20 mL of 2% PVA is taken to a 60 mL disposable syringe
and can stored at 22°C for at least 2 wk. 1.5 mg of lead citrate (trihydrate, carbon-
ate-free, Polysciences) is added to the remaining PVA, ultra-sonicated for 15
min, mixed with stirring for 5 min and then left unstirred for 15 min at 22°C.
Small amount of precipitates may be present in the bottom of the beaker. Super-
natant of the LC-PVA is taken into a 60-mL disposable syringe and can be stored
at 22°C for at least 2 wk. PVA and LC-PVA stock solutions as well as UA are
filtered before use.
34. 0.8% uranyl acetate-1.6% polyvinyl alcohol (UA-PVA) solution: 200 µL of UA
and 800 µL of PVA are mixed in a 1.5-mL microcentrifuge tube with a vortex.
35. Nikon Optiphot microscope equipped with a Photometrics Cool Snap fx charge-
coupled device (CCD) camera (Roper Scientific, Tucson, AZ).
36. MetaMorph image analysis software system (Universal Imaging Corp.,
Downingtown, PA).
37. Photoshop software (Adobe, Mountain View, CA).
3. Methods
The methods described below outline (1) the preparation of ultrathin
cryosections of human placenta, (2) the technique of high-resolution immuno-
fluorescence microscopy using ultrathin cryosections, and (3) the procedure of
correlative microscopy using FNG.
25_Takizawa_351_370_F 361 8/29/05, 11:23 AM

3.1. Preparation of Ultrathin Cryosections for Correlative Microscopy

1. Samples. Human placentas from uncomplicated Cesarean deliveries are obtained
for fixation as rapidly as possible (no more than 20 min after delivery).
2. Fixation. A tissue block (1 cm3) is cut from placenta with a pair of fine surgical
scissors, washed in 50 mL of 4% paraformaldehyde in a disposable plastic cup
within 5 s at 22°C in order to remove excess maternal blood as much as possible,
and then cut into thin strips like match-sticks as fast as possible with double
edge, carbon steel blades on a pink base plate wax for no longer than 5 min (see
Note 1). Samples are then transferred into a new disposable cup containing 50 mL
of the same fixative for 2 h at 22°C (16).
3. Collection of villi. After fixation, the samples are washed in 50 mL of cacodylate
buffer at 4°C for 10 min, transferred to a 60 × 15 mm cell culture dish containing
the same buffer, and then further dissected in order to collect segments of villi
with a No.10 blade and a pair of SS-style tweezers under a stereo microscope
within 1 h.
4. Gelatin solidification. Collected villi are transferred into a 1.5-mL microcentri-
fuge tube containing the same buffer and then pelleted for 3 s at 8000g using a
microcentrifuge. Supernatant is carefully discarded using a 200-µL round gel
loading tip attached to a suction tube. Two hundred microliters of the same buffer
is added into the tube, mixed gently, and then placed on ice until 20% gelatin
solution is prepared (see Note 2). After the microcentrifuge tube is placed in a
water bath for 1 min at 37°C, 200 µL of preheated (37°C) 20% gelatin solution is
added and mixed by gently pipeting for no longer than 10 s (see Note 2). Imme-
diately after pipeting, the tube is centrifuged for 6 s at 8000g (22°C) and then
transferred to ice for at least 15 min.
5. Sucrose infiltration. The microcentrifuge tube is cut with a single-edged blade
and then the solidified gelatin containing villus pellet is carefully taken from the
tube into a droplet of the cacodylate buffer on a dental wax plate with SS-style
tweezers. The solidified gelatin is cut into small rectangular parallelepipeds (1 x
1 × 3–4 mm) or prisms with double edge blades under a stereo microscope in
order to fit into groove type specimen carrier pins for a Reichert cryoultramicrotome
(see Note 3). The specimen is transferred to 1 mL of the cacodylate buffer in a
small snap-cap vial (BD Falcon polystyrene round bottom test tubes, 12 × 75 mm,
5 mL) and then infiltrated with 2.3 M sucrose solution (while being intermittently
stirred) with the following mixtures:
5% sucrose plus 2.3 M sucrose (2:1) 10 min at 22°C.
5% sucrose plus 2.3 M sucrose (1: 3) 10 min at 22°C.
2.3. M sucrose overnight at 4°C.
6. Sample freezing. After sucrose infiltration, the samples are mounted on the speci-
men pins, frozen in liquid nitrogen and subsequently stored in a liquid nitrogen
tank until they are sectioned (see Note 3).
25_Takizawa_351_370_F 362 8/29/05, 11:23 AM

7. Sectioning. Ultrathin cryosections are cut on a Cryo P diamond knife with a

Reichert Ultracut E equipped with an FC 4D cryounit. The temperature of the
sample, the knife, and the chamber are set at –130°C, –120°C, and –135°C
respectively, the cutting speed at 0.8–10 mm/s, and the section thickness at
50–100 nm.
8. Recovery of sections. The ultrathin section is collected on a droplet of 0.75%
gelatin–2.0 M sucrose (4,17) and then transferred to a 2.0% 3-aminopropyltriethoxy-
silane-coated round glass coverslip (12-mm diameter, No. 1 thickness) or to a
formvar film-coated, carbon stabilized, glow-discharged nickel grid (see Note 4).
The pick-up solutions contain 0.05% sodium azide, so that the cryosections on
coverslips or grids can be stored at 4°C until they are immunolabeled (18) (see
Note 5).
3.2. High-Resolution Immunofluorescence Microscopy Using Ultrathin

Cryosections
1. Removal of pick-up solution. Coverslips containing sections are transferred to
1 mL MFBS-PBS in a 24-well cell culture plate in a 4°C cold-room and immersed
for 15 min at 37°C (see Note 6).
2. Immunostaining. The coverslips are washed in PBS in the 24-well cell culture
plate three times, and then incubated in MFBS-PBS to block nonspecific protein
binding sites for 60 min at 22°C. The cryosections are subsequently incubated
with primary antibodies (e.g., chicken anti-CAV-1α, diluted 1:500 in MFBS-
PBS; and mouse anti-EEA1, diluted to 10 µg/mL in MFBS-PBS) on Parafilm in
a 100- × 20-mm cell culture dish (BD Falcon) for 30 min at 37°C (see Note 7).
The coverslips are transferred to PBS in the 24-well plate, rinsed in PBS four
times over 15 min, immersed in MFBS-PBS, and then incubated with secondary
antibodies (e.g., either Alexa Fluor 488-labeled goat anti-chicken IgY [diluted
1:200] and Alexa Fluor 594-labeled goat anti-mouse IgG [diluted 1:200] in
MFBS-PBS for 30 min at 37°C for detection of anti-CAV-1α and anti-EEA1
binding, respectively) as done with primary antibodies. After immunolabeling,
the cryosections are washed in PBS three times in the 24-well, stained with 4',6-
diamidino-2-phenylindole (DAPI) (1 µg/mL)-PBS for 10 min at 22°C, washed in
PBS five times over 10 min, and then mounted on glass microscope slides in
ProLong antifade medium (7 µL per coverslip). The mounting medium is dried
on a flat surface in the dark overnight at 22°C. Control sections receive the same
treatment except for omission of the primary antibody or substitution of pre-
immune antibody for the primary antibody.
3. Capture of images. Fluorescence and differential interference contrast (DIC)
images are collected with a Nikon Optiphot microscope equipped with a
Photometrics Cool Snap fx CCD camera (see Note 8 and Figs. 3,4). Images are
captured with a MetaMorph image analysis software system and then analyzed.
Captured images may be compiled with Photoshop software, as well as in
MetaMorph.
25_Takizawa_351_370_F 363 8/29/05, 11:23 AM

3.3. Correlative Microscopy Using Ultrathin Cryosections

1. Removal of pick-up solution. Grids in the gelatin–sucrose solutions are carefully
picked up from a microscope slide with a pair of No. 5 style tweezers in a 4°C
cold-room, floated on MFBS-PBS in a 35- × 10-mm cell culture dish, and then
incubated for 15 min at 37°C.
2. Immunostaining. The girds are transferred with a pair of No. 5 style nonmagnetic
tweezers (see Note 9) to droplets of PBS, and given two more PBS washes on
Parafilm. They are then floated on MFBS-PBS droplets to block nonspecific pro-
tein binding sites for 60 min at 22°C. The grids are subsequently incubated with
primary antibody (e.g., mouse anti-EEA1, diluted to 10 µg/mL in MFBS-PBS, in
a 100- × 20-mm cell culture dish in the same manner as are the 12-mm coverslips
containing ultrathin cryosections). Following washing by floating each grid suc-
cessively on four droplets of PBS and a droplet of MFBS-PBS for at least 3 min
in each droplet, the grids are incubated with biotin-labeled secondary antibody
(e.g., biotin-labeled goat anti-mouse, diluted to 25 µg/mL in MFBS-PBS, in the
cell culture dish for 30 min at 37°C. Washing with PBS and MFBS-PBS are
performed in the same manner. The grids are subsequently incubated with
streptavidin-labeled FluoroNanogold conjugated with Alexa Fluor 594 (diluted
1:50 in MFBS-PBS) for 30 min at 22°C. The grids are then washed by floating
them successively on five droplets of PBS for 3 min in each droplet.
3. Temporary Mounting. Following the immunolabeling procedure, the grid is not
floated but immersed in a droplet of an antifading medium, transferred onto a
glass microscope slide, and then temporarily mounted in a very thin layer of the
antifading medium (6 µL/grid) between a 18-mm round glass coverslip (#1
thickness) and the glass microscope slide without any special spacers (19) (see
Note 10).
4. Immunofluorescence microscopic observation. Once mounted, the grid is exam-
ined immediately by optical microscopy (Fig. 5). Images are collected in the
same manner as those from 12-mm coverslips containing ultrathin cryosections;
the locations of regions of interest on the “finder” grid are noted for relocation in
an electron microscope.
5. Disassembly of temporary slide. The temporary slide preparation is subsequently
disassembled (see Note 11) and the grids are immersed in droplets of PBS until
light microscopic examination of every grid is ended. The grids are washed not
on but in five droplets of PBS for 3 min in each droplet. The ultrathin cryosections
on the grids are then fixed in droplets of 2% glutaraldehyde in PBS for 30 min to
further stabilize the sections. The grids are then washed in five droplets of dis-
tilled water for 1 min in each droplet. The grids are carefully picked-up with a
pair of SS-style tweezers and the side of the grids opposite the sections is barely
touched with the tip of a piece of hardened filter paper (Grade 50) to remove
excess distilled water on the backside of the grids. Immediately after the treat-
ment, the grids are floated on a droplet of distilled water not to dry the sections on
the grids.
25_Takizawa_351_370_F 364 8/29/05, 11:23 AM

6. Silver enhancement. 1.4-nm FNGs bound to the cryosections are then subjected
to a silver enhancement procedure in order to render them visible in the sections
under an electron microscope (20). The grids are then floated on three droplets of
50 mM MES buffer for 1 min in each droplet on Parafilm. Immediately after
making working NPG silver enhancement solution, the grids are washed on a
droplet of the working NPG solution within a few seconds to minimize the dilu-
tion of the silver enhancement solution with MES buffer and then floated on
another droplet of the working NPG solution for 3–3.5 min at 22°C under a sodium
vapor safelight (see Note 12). The grids are immediately washed by floating them
successively on droplets of neutral fixer solution for 5 min at 22°C under a
sodium vapor safelight and three droplets of PBS for 6 min.
7. Positive contrast enhancement and PVA embedding. Visualization and preserva-
tion of ultrastructure of cryosections is achieved by the positive contrast enhance-
ment method (16). After silver enhancement, the ultrathin cryosections on grids
are postfixed on droplets of reduced osmium fixative on Parafilm for 15 min at
22°C in a chemical fume hood. The grids are washed on three droplets of distilled
water for 1 min each droplet and then floated on droplets of UA-PVA for 15 min
at 22°C. The grids are then washed on droplets of PVA for 10 s and droplets of
LC-PVA for 10 s and then floated on droplets of LC-PVA for 15 min at 22°C.
The EM grids on droplets of LC-PVA are collected with a wire loop (3–3.5 mm
in inner diameter) and the excess LC-PVA fluid is removed with a small piece of
hardened filter paper (see Note 13). The grids are then dried in air.
8. Electron Microscopic Observation. Grids are examined with a Philips CM-12
transmission electron microscope operated at 60 kV. The same regions examined
by fluorescence microscopy are relocated and then electron micrographs are col-
lected (Fig. 5).
4. Notes
1. We routinely cut a tissue block from tissue in the half or one third depths from
the maternal surface of the central region of placenta because it is rich in terminal
villi. Initial fixation during mincing on dental wax, as well as additional 2 h fixa-
tion, is important to meet the conditions for achievement of successful immunocy-
tochemistry as described in the Introduction. Fixative is freshly prepared on the
day of use. It should be noted that 2 h fixation in 4% paraformaldehyde at room
temperature is minimally essential for preservation of placenta ultrastructure for
immunoelectron microscopy; in other words, this fixation is needed for correla-
tive microscopy (16,21).
2. Before adding the 20% gelatin solution, a pellet of villi is resuspended with the
cacodylate buffer since the gelatin solution is viscous. In addition, for pipeting
of the 20% gelatin solution, 3–4 mm is cut from a 1000 µL tip to permit easy
mixing.
3. It is better to finish the necessary trimming of samples prior to sucrose infiltra-
tion since sucrose-infiltrated samples are very sticky. Furthermore, postfreezing
25_Takizawa_351_370_F 365 8/29/05, 11:23 AM

trimming may crack the frozen samples. For the prevention of sample dissocia-
tion from the pins during storage in liquid nitrogen, the mounting surface of the
pins is well scratched with a single edge blade and then ultra-sonicated in acetone
in a beaker for 15 min prior to use.
4. The cutting of ultrathin cryosections and transfer from a knife to either coverslips
or EM-grids is one of the most crucial steps in this technique. It is key to make a
flat (i.e., uncompressed and unwrinkled) thin section with a diamond knife and
then collect it on a droplet of the gelatin–sucrose pick-up solution. Many investi-
gators think it necessary to make sections stretch using the surface tension of a
2.3 M sucrose pick-up solution developed by Tokuyasu (22), because our samples
are fixed only with 4% paraformaldehyde, the higher surface tension of the 2.3 M
sucrose pick-up solution results in alterations in the ultrastructural integrity of
ultrathin cryosections (4,22,23). This leads to the failure of immunocytochemis-
try (i.e., poor labeling, increase of background, dislocation of antigen sites) as
well as poor visualization of cell ultrastructure (17). Another alternative pick-up
solution is 1% methylcellulose–1.15 M sucrose in order to reduce the surface
tension (24). For correlative microscopy, ultrathin cryosections are collected on
Maxtaform “finder” grids (200 mesh, nickel; Graticules, Tonbridge, Kent, UK)
in order to facilitate location of specific structures when going from the optical to
the electron microscope. Nickel grids are better than copper ones since nickel
grids are stronger.
5. The ability to store cryosections in pick-up solutions contain 0.05% sodium azide
at 4ºC greatly increases efficiency. When we examined the in situ distribution of
CAV-1α in human placenta, there was no significant difference in its immunore-
activity and the ultrastructure of placenta between immediately processed sec-
tions and ones stored for 1 yr (unpublished data).
6. This step is designed to gently dissolve the gelatin–sucrose solutions that cov-
ered sections on coverslips. It is convenient to use a 24-well plate for the treat-
ment (i.e., washing and blocking) of coverslips during immunocytochemistry.
You can discard used PBS from a 24-well cell culture plate using a 200-µL tip
attached to a suction tube and add new PBS with a disposal transfer pipet.
7. Each coverslip is carefully picked up from a well with a pair of No. 5 tweezers,
excess MFBS-PBS on the side of the coverslip opposite the section is wiped with
a piece of Kimwipes XL Delicate Task Wiper (Kimberly-Clark, Neenah, WI),
and then the coverslip is floated on a droplet of primarily antibody (approx 25–
50 µL/coverslip) on Parafilm in the cell culture dish. Beginners may need to
practice handling coverslips with tweezers. A piece of wet filter paper is attached
on the inside surface of the lid of the dish in order to prevent evaporation of
primary antibody solution. The use of these small coverslips for immunocy-
tochemistry minimizes the amount of primary antibody solution required.
8. A research-grade fluorescence microscope equipped with the proper fluorescence
filter sets, high magnification objective lens, and differential interference con-
trast optics is required for the imaging shown in this chapter. In addition, the
microscope should have a high quality electronic camera and an image analysis
25_Takizawa_351_370_F 366 8/29/05, 11:23 AM

system. It should take CCD camera features into consideration, including: quan-
tum efficiency, scan mode (bit and MHz), noise, sensitivity, resolution, and price.
9. For washing and blocking of the grids, the use of a 0.5-mL microcentrifuge tube
rack (50-place capacity; Ieda Trading Corp., Tokyo, Japan; cat. no. 9901) would
be helpful. The rack is covered with a Parafilm sheet and then dents are made on
the Parafilm at the openings with the bulb of a disposal transfer pipet.
10. Three microliters of the antifading medium is applied to the center of both a
coverslip and a microscope slide. The grid is carefully placed face-up on the
droplet of the antifading medium on the microscope slide not to leave air bubbles
in the medium. The coverslip then overlies the grid to keep the grid in the center
of the coverslip. The use of an 18-mm coverslip is large enough to minimize
contamination with immersion oil under optical microscopic observation using a
×100 objective lens. The choice of antifade reagents is important for correlative
microscopy with FNG. When 1,4diazabicyclo[2.2.2]octane (DABCO) is employed
instead of NPG, the gold signal from FNG is dramatically diminished but the
fluorescence signal is unaffected (19). The gold signal of DABCO-treated
samples decreases to approx 30% of that of the sample that is treated with NPG
(19). We recommend that NPG be used and that DABCO be avoided as an anti-
photobleaching reagent for this technique.
11. When the coverslip is removed with a pair of SS-style tweezers, a few droplets of
PBS are carefully added to the temporary slide to float the coverslip. It is neces-
sary to keep damage of ultrathin cryosections at a minimum.
12. Each batch of gum arabic stock solution, NPG, and silver lactate should be pre-
liminarily tested to determine the best enhancement incubation time. FNG are
adhered to 0.25% poly-L-lysine (Sigma-Aldrich)-coated formvar grids. These
grids are exposed to the silver enhancement procedure for various time (e.g.,
1–5 min) and then examined by electron microscopy (11).
13. Removal of excess of LC-PVA is important to attain adequate morphological
detail in ultrathin cryosections. As Tokuyasu reported (25), the removal of the
excess is terminated immediately after the remaining amount of LC-PVA forms a
concave meniscus with its center barely touching the center of the grids. The
transfer loop is then left in air. The embedding medium begins to dry from the
center, becomes a thin film in the loop and then shows a gold-blue color in
reflected light. The dried film in the loop is carefully cut along the rim of the
grids with a pair of SS style tweezers prior to picking up the grids.
Acknowledgments
This work was supported in part by grants from the National Institutes of
Health (HD38764 [JMR]). We are indebted to Ms. Heather Richard and the
Campus Microscopy and Imaging Facility at Ohio State University for assis-
tance. We thank Drs. Shigeki Matsubara and Takeshi Takayama of Jichi Medi-
cal School for their technical support. We also thank Dr. Fumimaro Takaku,
President of Jichi Medical School for his encouragement.
25_Takizawa_351_370_F 367 8/29/05, 11:23 AM

References
1. Stirling, J. W. (1993) Controls for immunogold labeling. J. Histochem. Cytochem.
41, 1869–1870.
2. Burry, R. W. (2000) Specificity controls for immunocytochemical methods. J.
Histochem. Cytochem. 48, 163–165.
3. Robinson, J. M., Takizawa, T., Vandre, D. D., and Burry, R. W. (1998) Ultrasmall
immunogold particles: Important probes for immunocytochemistry. Microsc. Res.
Tech. 42, 1–11.
4. Tokuyasu, K. T. (1980) Immunochemistry on ultrathin frozen sections.
Histochem. J. 12, 381–403.
5. Howell, K. E., Reuter-Carlson, U., Devaney, E., Luzio, J. P., and Fuller, S. D.
(1987) One antigen, one gold? A quantitative analysis of immunogold labeling of
plasma membrane 5'-nucleotidase in frozen thin sections. Eur. J. Cell Biol. 44,
318–327.
6. Dulhunty, A. F., Junankar, P. A., and Stanhope, C. (1993) Immunogold labeling
of calcium ATPase in sarcoplasmic reticulum of skeletal muscle: use of 1-nm, 5-nm,
and 10-nm gold. J. Histochem. Cytochem. 41, 1459–1466.
7. Griffiths, G. and Hoppeler, H. (1986) Quantitation in immunocytochemistry:
correlation of immunogold labeling to absolute number of membrane antigens.
J. Histochem. Cytochem. 34, 1389–1398.
8. Robinson, J. M., Takizawa, T., and Vandre, D. D. (2000) Applications of gold
cluster compounds in immunocytochemistry and correlative microscopy: com-
parison with colloidal gold. J. Microsc. 199, 163–179.
9. Yokota, S. (1988) Effect of particle size on labeling density for catalase in protein
A-gold immunocytochemistry. J. Histochem. Cytochem. 36, 107–109.
10. Ghitescu, L. and Bendayan, M. (1990) Immunolabeling efficiency of protein A-gold
complexes. J. Histochem. Cytochem. 38, 1523–1530.
11. Takizawa, T. and Robinson, J. M. (1994) Use of 1.4-nm immunogold particles for
immunocytochemistry on ultrathin cryosections. J. Histochem. Cytochem. 47,
569–573.
12. Majlof, L. and Forsgren, P. O. (1993) Confocal microscopy: important consider-
ations for accurate imaging. Meth. Cell Biol. 38, 79–95.
13. Takizawa, T. and Robinson, J. M. (2003) Ultrathin cryosections: An important
tool for immunofluorescence and correlative microscopy. J. Histochem.
Cytochem. 51, 707–714.
14. Mills, I. G., Jones, A. T., and Clague, M. J. (1998) Involvement of the endosomal
autoantigen EEA1 in homotypic fusion of early endosomes. Curr. Biol. 8, 881–884.
15. Lyden, T. W., Anderson, C. L., and Robinson, J. M. (2002) The endothelium but
not the syncytiotrophoblast of human placenta expresses caveolae. Placenta 23,
640–652.
16. Takizawa, T., Anderson, C. L., and Robinson, J. M. (2003) A new method to enhance
contrast of ultrathin cryosections for immunoelectron microscopy. J. Histochem.
Cytochem. 51, 31–39.
25_Takizawa_351_370_F 368 8/29/05, 11:23 AM

17. Takizawa, T. and Robinson, J. M. (1994) Composition of the transfer medium is

crucial for high-resolution immunocytochemistry of cryosectioned human neu-
trophils. J. Histochem. Cytochem. 42, 1157–1159.
18. Griffith, J. M. and Posthuma, G. (2002) A reliable and convenient method to store
ultrathin thawed cryosections prior to immunolabeling. J. Histochem. Cytochem.
50, 57–62.
19. Takizawa, T. and Robinson, J. M. (2000) Analysis of antiphotobleaching reagents
for use with FluoroNanogold in correlative microscopy. J. Histochem. Cytochem.
48, 433–436.
20. Burry, R. W. (1995) Pre-embedding immunocytochemistry with silver enhanced
small gold particles, in Immunogold-Silver Staining. Principles, Methods, and
Applications (Hayat, M. A., ed.). CRC, Boca Raton, FL: pp. 217–230.
21. Takizawa, T. and Robinson, J. M. (2003) Correlative microscopy of ultrathin
cryosections is a powerful tool for placental research. Placenta 24, 557–565.
22. Tokuyasu. K. T. (1973) A technique for ultracryotomy of cell suspensions and
tissues. J. Cell Biol. 57, 551–565.
23. Tokuyasu, K. T. (1986) Application of cryoultramicrotomy to immunocytochem-
istry. J. Microsc. 43 (Pt 2), 139–149.
24. Liou, W., Geuze, H. J., and Slot, J. W. (1996) Improving structural integrity of
cryosections for immunogold labeling. Histochem. Cell Biol. 106, 41–58.
25. Tokuyasu, K. T. (1989) Use of poly(vinylpyrrolidone) and poly(vinyl alcohol) for
cryoultramicrotomy. Histochem. J. 21, 163–171.
25_Takizawa_351_370_F 369 8/29/05, 11:23 AM

25_Takizawa_351_370_F 370 8/29/05, 11:23 AM
Vascular Corrosion Casting in Mice 371
26
Vascular Corrosion Casting of the Uteroplacental
and Fetoplacental Vasculature in Mice
Kathie J. Whiteley, Christiane D. Pfarrer, and S. Lee Adamson
Summary
This chapter describes methods for making vascular corrosion casts of the uteroplacental
and fetoplacental vasculature of the mouse placenta. A catheter placed in the ascending tho-
racic aorta of a pregnant mouse permits the introduction of a methyl methacrylate casting com-
pound into the lower body vasculature, including the uterus and placenta. A fine-tipped glass
cannula attached to a double-lumen catheter is used to instill the same casting compound in the
fetoplacental vessels of mouse placentas. Following polymerization of the casting compound,
tissue is digested off of the placental casts using 20% KOH. The washed and dried casts are
then available for light or scanning electron microscopy. The methods described have been
used to cast the mouse uteroplacental vasculature from 5.5–18.5 d gestation and the
fetoplacental vasculature from 12.5 d gestation to term.
Key Words: Mouse; placenta; umbilical vessels; uterus; pregnancy; embryo; decidua; cir-
culation; microvasculature.
1. Introduction
The availability of mutant strains of mice with placental insufficiencies has
emphasized the need for methods to investigate the normal and abnormal vascu-
lar anatomy of the mouse placental circulation. One method we found invaluable
was vascular corrosion casting of the uteroplacental and fetoplacental vasculature.
We used this method to help establish the normal maternal and fetal circula-
tions in the mouse placenta in the last half of pregnancy (1). Vascular casts
permitted three-dimensional visualization of the structure of the blood spaces
in the placenta, and by partially filling the circulations from the arterial or
venous sides, we were able to conclusively identify and separately describe the
arterial and venous supply. We found that blood spaces so obvious in vascular
casts often collapsed during tissue dissection and processing for histology and
were barely detectable on histological sections. Qualitative and quantitative
information on vessel number and diameter can also be obtained from the casts.
371
26_Whiteley_371_392_F 371 8/29/05, 11:23 AM

372 Whiteley, Pfarrer, and Adamson
372
26_Whiteley_371_392_F 372 8/29/05, 11:23 AM

In addition, casts provide information on the surface morphology of the cells

lining the blood-perfused spaces. The endothelial cell imprints of arteries dif-
fer from those of veins, and the imprints of trophoblast cells are also dissimilar
(Figs. 1C and 1D).
This chapter details our protocols for the casting of the uteroplacental and
fetoplacental vasculature of the mouse placenta. Examples of casts are shown
in two of the figures. Within each protocol, methods are described for the sur-
gical preparation of the samples, the perfusion and casting of the placental
vessels, and the processing of the vascular casts. These methods were devel-
oped through modifications of previous work (2,3,4).
This chapter describes methods for filling the placental circulation with
methyl methacrylate for vascular corrosion casts. However, these methods
could be easily adapted to fill the circulation with other substances such as
contrast agents for magnetic resonance imaging or for micro-computed tomog-
raphy, or for perfusion fixation of the placenta.
2. Materials
2.1. Vascular Corrosion Casting of the Uteroplacental Vasculature
in Mice
1. Pregnant mouse.
2. Warming mattress and/or warming light (VWR International, Inc., West
Chester, PA).
Fig. 1. (opposite page) Vascular corrosion casts of the uteroplacental vasculature in

mice. (A) A light microscope photograph of a complete, undissected cast of the mater-
nal placental circulation. (B) A scanning electron microscopy (EM) photo of a cast
which has been dissected to show the central canals traversing the labyrinth and
branching near the embryonic surface of the placenta. (C) The region indicated by the
upper box in B is enlarged to show the indentations in the cast surface caused by
endothelial cells lining a uterine artery (vessel on lower left) and vein (vessel on upper
right). (D) The lower box in B is enlarged to show the punctate nuclear impressions
left by a spiral artery (upper vessel) and the relatively smooth surface of the tropho-
blast-lined central canal (lower vessel). (E) A scanning EM photograph of a complete,
undissected cast of the maternal placental circulation viewed from the embryonic side.
The arrow points to the labyrinth, the site of exchange between maternal and embry-
onic blood. The trophoblastic tubules of the labyrinth form a ring around a circular
central region, which is bordered by the site of attachment of the yolk sac membrane.
(F) A scanning EM photograph showing the radial arteries of the maternal arterial
supply to the uterus. (G) The region indicated by the box in F is enlarged to show the
circumferential bands of vasculature and localized site of constriction (arrows) that
suggest the presence of sphincters around the radial arteries.
26_Whiteley_371_392_F 373 8/29/05, 11:23 AM

3. Small metal tray (approx 16 cm × 25 cm).

4. Surgical microscope (at 5–8× magnification).
5. Warming oven.
6. Anesthetic.
7. Masking tape.
8. Gauze.
9. Cotton swabs.
10. Two fine forceps (e.g. No. 5 Dumont forceps).
11. Four Hartman hemostatic forceps (curved or straight).
12. Fine iris scissors.
13. ToughCut Metzenbaum scissors.
14. Extra fine or delicate scissors (e.g., mini-Vannas or spring scissors).
15. Small needle holders (e.g., Castroviejo).
16. Aortic catheter: 10 cm of P.E. 50 tubing (Intramedic®, Becton Dickinson and Com-
pany, Parsippany, NJ) with a hand-stretched tapered tip cut blunt, on a 23-gage
blunted needle remove.
17. 5'0' silk suture.
18. 6'0' silk suture with a three-eighths-inch tapered, curved needle.
19. Plastic syringes: three 10-cc syringes, two 5-cc syringes, two 3-cc syringes, one
1-cc syringe.
20. 1-cc plastic syringe with 27-gauge needle.
21. 0.2-µm syringe filter (Arrow International, Reading, PA).
22. Intracardiac heparin stock solution (100 international units [IU]/mL): 0.1 mL of
10,000 IU/mL heparin and 10 mL of 0.9% NaCl. If prepared and stored using
sterile techniques, this solution can be stored at 4°C for up to 1 mo.
23. 2% xylocaine (20 mg/mL; AstraZeneca Canada, Inc., Missisauaga, ON).
24. Perfusate: 15 mL of 2% xylocaine (20 mg/mL), 15 mL of 0.9% NaCl, and 0.3 mL
of intracardiac heparin (100 IU/mL). Mix in a small beaker. Draw up 10 mL
through a 0.2-µm syringe filter into each of three 10-cc syringes. Label two sy-
ringes as “warm perfusate” and place in a warming oven at 45°C. Label the third
syringe as “cold perfusate” and place on crushed ice (chill to 4°C).
25. Methyl methacrylate casting compound (Polysciences, Inc., Warrington, PA): 5 mL
of Batson’s No. 17 Monomer base solution at 4°C, 1.5 mL Catalyst at 20°C, 0.1 mL
Promoter at 4°C.
26. Jet Acrylic Liquid (Lang Dental, Wheeling, IL): 2.4 mL at 4°C.
27. Wooden applicator stick.
28. 20-mL glass vial.
29. Infusion pump.
30. Crushed ice in a small styrofoam box (for cooling casting compound supplies
to 4°C).
31. Pressure-head system (to maintain 20 mmHg pressure during curing of casting
compound).
32. Stopwatch or watch with a second hand.
33. Glass test tubes or vials (7 to 20 mL, depending on size of samples being col-
lected).
26_Whiteley_371_392_F 374 8/29/05, 11:23 AM

34. 20% KOH (20% KOH is corrosive. Wear protective gloves and eyewear and
avoid contact with skin and clothing).
35. Distilled water.
36. Pasteur pipets with bulb.
2.2. Vascular Corrosion Casting of the Fetoplacental Vasculature

in Mice
1. Pregnant mouse.
2. Warming mattress and/or warming light (VWR International, Inc.).
3. Small metal tray (approx 16 cm × 25 cm).
4. Surgical microscope (at 8× magnification) with an observer port.
5. Warming oven.
6. Small, wide-mouthed container, approx 500 mL vol (e.g., plastic margarine dish).
7. Styrofoam box with lid, filled with crushed ice.
8. Ice-cold phosphate-buffered saline (PBS): 0.14 M NaCl, 2.7 mM KCl, 8.1 mM
Na2HPO4, 1.5 mM KH2PO4, pH 7.2, at 4°C.
9. 100-mm × 15-mm plastic Petri dish.
10. 2-inch × 2-inch gauze squares.
11. Warming plate with 250- to 500-mL beaker.
12. Small thermometer (must include 45°C and surrounding temperatures).
13. Ice pack or resealable plastic bag filled with crushed ice.
14. 7'0' silk suture.
15. Conchatherm humidifier heater (Respiratory Care Inc., Arlington Heights, IL).
16. Glass cannula attached to a double-lumen catheter (one per placenta to be cast)
17. Glass test tubes or vials (7–20 mL, depending on size of samples being collected).
18. Toughcut Metzenbaum scissors.
19. Fine iris scissors.
20. Extra-fine or delicate scissors (e.g., mini-Vannas or spring scissors).
21. Two very fine forceps (e.g., No. 55 Dumonts).
22. Hayman-style microspatula.
23. Crushed ice in a small styrofoam box (for casting compound supplies).
24. Methyl methacrylate casting compound (Polysciences Inc.): 0.5 mL of Batson’s
No. 17 Monomer base solution at 4°C, 0.15 mL of Catalyst at 20°C, 10 µL Pro-
moter at 4°C.
25. Jet Acrylic Liquid (Lang Dental): 0.24 mL at 4°C.
26. Red pigment (Polysciences, Inc.).
27. Wooden applicator sticks (one per placenta to be cast).
28. 12- × 75-mm glass test tubes (one per placenta to be cast) in a test tube rack.
29. 1- to 10-µL pipet with tips.
30. 0.3-cc insulin syringes (two per placenta to be cast).
31. 1-cc syringes (one per placenta to be cast).
32. 3-cc syringes (one per placenta to be cast).
33. Stopwatch or watch with second hand.
34. 0.2-µm syringe filter (Arrow International).
26_Whiteley_371_392_F 375 8/29/05, 11:23 AM

35. Intracardiac heparin stock solution (100 IU/mL): 0.1 mL of 10,000 IU/mL hep-
arin and 10 mL of 0.9% NaCl. If prepared and stored using sterile techniques,
this solution can be stored at 4°C for up to 1 mo.
36. Perfusate: 10 mL of 2% xylocaine (20 mg/mL), 10 mL of 0.9% NaCl, and 0.2 mL
of intracardiac heparin (100 IU/mL). Mix in a small beaker. Draw up 10 mL
through a 0.2-µm syringe filter into each of two 10-cc syringes. Cap the syringes
and heat to 40–45°C.
37. Warm PBS: 0.14 M NaCl, 2.7 mM KCl, 8.1 mM Na2HPO4, 1.5 mM KH2PO4,
pH 7.2, at 40–45°C.
38. 10-cc syringes (eight syringes are required).
39. Warm 3% paraformaldehyde in a labeled 3-cc syringe with a 23-gage needle.
40. 20% KOH (20% KOH is corrosive. Wear protective gloves and eyewear and
avoid contact with skin and clothing).
41. Distilled water.
42. Pasteur pipets with bulb.
3. Methods
3.1. Vascular Corrosion Casting of the Uteroplacental Vasculature
in Mice
The methods described below outline (1) the surgical procedure, (2) the per-
fusion and casting of the lower body vasculature (including the uterus and pla-
centa), and (3) the processing of the vascular casts. The casting methods
detailed here are used to cast the entire uteroplacental vasculature (arteries,
microcirculation, and veins). However, we have used modifications of these
methods to cast the veins in blue and arteries in red or to prepare partial casts of
the venous or arterial circulations by infusing via the vena cava or aorta, respec-
tively (see Note 1).
3.1.1. Surgical Procedure
1. Anesthetize the mouse (see Note 2).
2. Lay the mouse on its back on a warmed tray to maintain body heat during the
surgery (see Note 3). Tape the mouse’s front paws to the tray to secure the animal.
3. Inject 0.05 mL of intracardiac heparin (see Note 4) combined with 0.1 mL of
2% xylocaine (see Note 5) into the mouse’s heart using a 1-cc syringe and a
27-gauge needle (see Note 6).
4. Open the mouse’s chest using Toughcut Metzenbaum scissors, taking care to leave
the diaphragm intact (i.e., do not enter the abdominal cavity) (see Note 7). Cut
through the sternum near the fifth rib. Cut along the rib towards the back of the
mouse on each side. Cut across the ribs as far dorsally as possible in a rostral direc-
tion. Once cut, the rib cage can be deflected rostrally to expose the heart and lungs
(Fig. 2). Using gauze and cotton swabs, clear any blood from the chest cavity.
5. The heart should still be beating (allowing the heparin to circulate). Xylocaine
will stop the breathing so the mouse will die quickly. When the heart stops, the
anesthesia is discontinued.
26_Whiteley_371_392_F 376 8/29/05, 11:23 AM

Fig. 2. Placement of the catheter in the aorta of the pregnant mouse which allows
perfusion and casting of the lower body vasculature (including the uterus and pla-
centa). (A) Retracted heart. (B) Lung. (C) Inferior vena cava with loose tie. (D) Intact
diaphragm. (E) Esophagus. (F) Ribs. (G) Azygos vein. (H) Aorta. (I) Catheter tied in
place in the aorta.
The following steps are performed under a surgical microscope:

6. Locate the inferior vena cava and position a loose 5'0' tie around the vessel (Fig. 2).
Secure the loose tie to the mouse’s skin with a hemostatic forcep to prevent the tie
from accidentally tightening around the vessel. This tie will be tightened after the
casting compound has been infused and is curing.
7. Expose the descending thoracic aorta and clear enough of its length to allow two
8-cm lengths of 5'0' ties to be placed approx 5 mm apart along the length of the
aorta. The top tie is positioned as close as possible to the Azygos vein. Tighten
this tie around the aorta. Attach a pair of hemostatic forceps to the ends of the tie.
This tie is used to gently retract the vessel to provide tension on the vessel during
catheterization. The downstream tie is left loose until the catheter is in place in
the aorta. Attach hemostatic forceps to the ends of this tie to keep them together.
To improve access to the aorta, it is helpful to retract the heart using a hemostatic
forcep on the tip of the ventricle (Fig. 2).
26_Whiteley_371_392_F 377 8/29/05, 11:23 AM

8. Attach a 10-cc syringe filled with warmed (45°C), filtered perfusate to the cath-
eter (see Note 8). Fill the catheter with perfusate and expel a small volume of the
perfusate into the chest cavity. Lay the catheter alongside the aorta with the tip of
the catheter in the expelled perfusate to prevent air from travelling up the catheter.
9. Using the two ties around the aorta, extend the aorta using a small amount of
tension. Cut a small hole in the aorta, between the two ties, using extra fine or
delicate scissors. Take care not to cut through the vessel.
10. Using forceps, insert the catheter in an anterograde direction through the hole
into the aorta. Advance the catheter until its tip is lying just inside the abdominal
cavity. Secure the catheter in place with the downstream tie.
11. Ensure that the catheter is in the aorta and has not perforated the vessel by gently
pulling back on the plunger of the 10-cc syringe (see Note 9). Some blood should
come back in the catheter. Flush this blood back into the mouse.
12. When satisfied that the catheter is positioned correctly, secure it in place using
the upstream tie. Wrap the tie around the catheter and tighten it without occlud-
ing the catheter (Fig. 2).
13. Tether the catheter to the ribcage using 6'0' suture with a three-eighths-inch
tapered, curved needle.
14. Trim the aortic catheter ties.
15. Locate the right atrium and, using iris scissors, make a cut through the wall of the
chamber to act as a vent.
16. Gently infuse approx 1 mL of warmed perfusate through the catheter and ensure
that the blood being pushed by this influx of perfusate is exiting through the vent
in the right atrium. Ensure that the tie around the inferior vena cava is still loose
and not occluding the vessel.
17. Transfer the mouse preparation on its tray to a fumehoood where an infusion
pump, the casting compound supplies, and a pressure-head system (for maintain-
ing a constant 20 mmHg pressure during curing of the casting compound) have
been set up.
3.1.2. Perfusion and Casting of the Lower-Body Vasculature, Including
the Uterus and Placenta
The following steps are carried out in a fume hood.
It is critical that air is not allowed to enter the vascular system during perfu-
sion and casting. Air bubbles will decrease the quality of the vascular casts.
Preliminary experiments are important to measure femoral arterial blood pres-
sure during infusion of the perfusate and casting compound. Ensure that pres-
sures do not exceed normal systolic arterial blood pressure. Rates and times
given below are suitable for our pregnant ICR (CD-1) mice (1). It may be nec-
essary to adjust infusion times and/or rates depending on the size and/or strain
of mouse.
1. Infuse approx 9–10 mL of warmed (45°C), filtered perfusate using an infusion
pump (see Note 8). Begin at a rate of 0.5 mL/min increasing gradually to a final
rate of 4 mL/min. It should take approx 2–2.5 min to infuse 9–10 mL of perfusate.
26_Whiteley_371_392_F 378 8/29/05, 11:23 AM

2. Repeat step 1 using a second syringe containing 9–10 mL of warmed (45°C),

filtered perfusate. The infusion of the warmed perfusate should reopen the vascu-
lature and clear it of blood.
3. Repeat step 1 with 10 mL of cold (4°C), filtered perfusate. The cold perfusate
will prolong the setting time of the casting compound as well as harden the ves-
sels so they are less likely to rupture.
4. Prepare the casting compound while the cold perfusate is being infused. The cast-
ing compound is a mixture of Batson’s No. 17 monomer base solution (5 mL at
4°C), catalyst (1.5 mL at 20°C) and promoter (0.1 mL at 4°C) with the addition
of jet liquid (2.4 mL at 4°C) (see Note 10). All of the components of the casting
compound (except the catalyst), the syringes used to measure the components
and the mixing vial are kept at 4°C on crushed ice. The first three components
can be premeasured and stored in their respective syringes. The jet liquid must be
measured immediately prior to use. If left in a plastic syringe, the Jet Liquid will
digest the syringe. Mix the components on ice in a glass vial (20 mL) in the order
given above. Mix gently with a wooden applicator stick. Avoid introducing
bubbles or ice chips into the mixture, as they will decrease the quality of the
casts. Let the casting compound stand on ice for 30 s once completely mixed.
Draw up the casting compound into a chilled 5-cc syringe. Expel any air and
attach the syringe to the aortic catheter and to the infusion pump.
5. Infuse the chilled casting compound beginning at a rate of 0.14 mL/min, infuse
for 2 min at 0.4 mL/min and increase to a final rate of 0.7 mL/min. It should take
approx 9–10 min to infuse 5 mL of the casting compound.
6. Turn the pump off when done then immediately tighten the tie around the inferior
vena cava.
7. Apply a pressure of approx 20 mmHg to the infusion syringe (which should still
contain approx 0.5 mL of casting compound) to sustain vessel inflation while the
casting compound polymerizes (cures). We use pressure generated by a column
of water for this purpose. It will take approx 60–90 min for polymerization. One
can determine when the compound has cured by observing the hardness of the
excess compound remaining in the mixing vial. When the casting compound has
cured, remove the mouse from the infusion system for dissection.
8. Cut through the skin to expose the abdominal contents. Organs should be well
cleared of blood. Detach the uterus by severing the uterine and ovarian vessels
and the vagina. Gently remove the uterus, taking care not to grip it or bend it to
avoid breaking the cast. One can choose to either digest the tissue off of the cast
in one piece or to cut the uterus and its contents into individual implantation sites
and digest each site individually.
3.1.3. Processing the Casts

The following steps are carried out in a fume hood.
1. Immerse the dissected tissue containing the cast in 20% KOH to digest the tissue
off of the cast. Digestion can be done in small test tubes or larger glass vials
depending on the size of the samples. The tissue should be completely digested
26_Whiteley_371_392_F 379 8/29/05, 11:23 AM

within 24–48 h depending on the size of the samples. Digestion can be aided by
removing the old KOH (with a Pasteur pipet) and adding fresh KOH if digestion
is not completed in the first 24 h.
2. Following complete tissue digestion, wash the cast with distilled water. It is
imperative that this washing is done very carefully to avoid damaging the deli-
cate vascular cast. With a Pasteur pipet, add and remove at least three cycles of
distilled water from the vial containing the cast. Ensure that all tissue is washed
off of the cast.
3. The cleaned cast can be stored in distilled water (see Note 11). For viewing with
light or scanning electron microscopy (SEM) (see Note 12), casts must be air-
dried. It may be necessary to dissect away portions of the cast to expose the
vessels of interest (Fig. 1A). Dissection can be done under a light microscope
using Iris scissors. Dried casts should be stored covered to protect them from
damage and dust.
3.2. Vascular Corrosion Casting of the Fetoplacental Vasculature

in Mice
Before beginning the surgical portion of this protocol, ensure that all equip-
ment and supplies are set up and ready to use. This protocol requires two
people. Fill a styrofoam box with a lid two-thirds full with crushed ice. Set a
500-mL container full of ice-cold PBS into the crushed ice. The small metal
tray that we work on is placed on a warming mattress to provide a warm work
surface. In addition, our work surface is heated from above with a warming
light. Place the bottom half of a 100-mm × 15-mm Petri dish on the warm
metal tray. Set a 2-inch × 2-inch gauze square in the middle of the Petri dish.
We use a Conchatherm humidifier heater to continuously pass humidified
warm air over the petri dish containing our sample. The warm humidified air
passing over the embryo revives the embryo (see Note 13). The perfusate used
to flush blood from the vessels and the paraformaldehyde used to fix the holes
cut in the umbilical vessels must also be warm. We warm these solutions in a
warming oven and keep them warm at the work surface using a mug on a small
cup warming plate. Any regulated warming system can be used. The mug is
filled with warm PBS to keep a supply of warm PBS at the work site. In the
mug, we keep a labeled 10-cc syringe for drawing up the PBS. Also in the mug
are a labelled 3-cc syringe with a capped 23-gauge needle containing 3%
paraformaldehyde, and two capped and labeled 10-cc syringes containing per-
fusate. Additional syringes of perfusate and a supply of PBS are kept in a warm-
ing oven until they are needed. Monitor the temperature of the PBS in the mug
and maintain it between 40–45°C. On a bench away from the heated work area,
set up all of the supplies required for preparing the casting compound. The
components of the casting compound (with the exception of the catalyst and
the pigment) are kept on ice ready for use. The casting compound is mixed
26_Whiteley_371_392_F 380 8/29/05, 11:23 AM

fresh for each placenta being cast. The casting compound is prepared once an
embryonic heartbeat has been observed and while the umbilical vessels are
being prepared for infusion. The surgeon prepares the sample for casting while
the assistant prepares the casting compound. The surgeon inserts the cannula
into the umbilical vessel and holds it in place while the assistant acts as an
infusion pump and pushes the casting compound into the vessel. With the
exception of the removal of the pregnant uterus and the preparation of the
casting compound, all work is carried out using a surgical microscope with an
observer port allowing both the surgeon and the assistant to view the sample.
The methods described as follows outline the surgical procedure, perfusion
and casting of the placenta, and processing of the casts.
3.2.1. Surgical Procedure

1. Kill the pregnant mouse by cervical dislocation and record the time of death.
Steps 2 to 4 should be performed as quickly as possible. The goal is to get
the pregnant uterus into ice-cold PBS as quickly as possible without damaging
its contents.
2. Using Toughcut Metzenbaum scissors, open the mouse’s abdomen to expose the
intact pregnant uterus.
3. Carefully but quickly remove the entire uterus with its contents intact from the
mouse by severing the uterine and ovarian vessels and the vagina.
4. Immediately drop the pregnant uterus into the dish filled with ice-cold PBS. This
dish should be sitting in ice in a styrofoam box with a lid. The pregnant uterus
remains in the chilled PBS throughout the course of the experiment with indi-
vidual implantation sites being cut off as required for casting. We have success-
fully cast placentas up to 3 h after the death of the mother. Discard the mouse
carcass.
5. Select one implantation site to work on first. It is best to begin selecting from one
ovarian end in order to keep all of the remaining sites intact in the uterus. Cut the
section of uterus containing the selected implantation site off of the uterus and
set the selected segment on the gauze in the Petri dish. The remainder of the
pregnant uterus is returned to the dish of ice-cold PBS and kept on ice in the
styrofoam box until another site is required.
Work will be performed on individual, selected implantation sites. The fol-
lowing steps are carried out in the heated environment with the sample posi-
tioned on the gauze in the petri dish. These steps are repeated for each site as it
is selected for casting.
6. Identify the area of the uterus that appears least vascularized (the antimesometrial
side) and cut through the uterine muscle using iris scissors to expose the embryo
and placenta still within the membranes.
26_Whiteley_371_392_F 381 8/29/05, 11:23 AM

7. Using fine forceps and iris scissors, cut through the yolk sac and amniotic mem-
branes along the placental margin. Cut one-third to one-half of the way around
the placenta. Choose the side with the least yolk sac vasculature for the incision.
8. Put down the sharp instruments to avoid damaging the embryo and placenta. We
use a blunt Hayman-style microspatula to gently manipulate the embryo and pla-
centa. Gently lift the embryo out from within the membranes. Position the embryo
so it lies next to but faces away from the placenta (Fig. 3). Use the body of the
embryo to support the umbilical vessels at one end, and the position of the pla-
centa to put tension on the umbilical vessels. The vessels should be extended but
not stretched.
9. Revive the selected embryo by gradually warming it. Position the stream of warm,
humidified air over the embryo. It is important to keep the preparation warm
from this point on. Prior to being warmed, the umbilical vessels will be empty of
blood. As the embryo is warmed, its heart will begin to beat and the umbilical
vessels will fill with blood. When blood first appears in the vessels, the umbilical
vein will be a bright red color. Very quickly, the color of the blood in both ves-
sels will appear identical. A pulse will be visible in the umbilical artery. The
branches arising from the umbilical vein usually overlay those from the umbili-
cal artery on the placental surface (Fig. 4A) (see Note 14).
10. As soon as there is a visible heartbeat, drop 1–2 drops of warm 3% paraformalde-
hyde onto the umbilical vessels (using a 3-cc syringe and 23-gauge needle). The
paraformaldehyde helps to prevent vasospasm of the umbilical vessels during
handling and helps maintain the hole that will be cut in the vessel for cannula
insertion.
11. As soon as a heartbeat is detected, the assistant should begin mixing the casting
compound. The casting compound is a mixture of Batson’s No. 17 monomer
base solution (0.5 mL at 4°C), catalyst (0.15 mL at 20°C), and promoter (10 µL at
4°C) with the addition of jet liquid (0.24 mL at 4°C). All of the components of
the casting compound (except the catalyst), the syringes to measure the compo-
nents and the test tubes used to mix the casting compound in are kept at 4°C on
crushed ice. The first two components can be premeasured and stored in their
respective syringes. The promoter is measured as required. The jet liquid must be
measured immediately prior to use. If left in a plastic syringe, the jet liquid will
digest the syringe. If desired, pigment can be added to the casting compound (see
Note 15). The amount of pigment added to the casting compound depends on the
desired darknesss of the color in the cast. Start with the least amount of pigment
paste possible and adjust accordingly (see Note 16). Mix the components on ice
in a 12-mm × 75-mm glass test tube in the order described previously. Mix the
components thoroughly but gently with a wooden applicator stick. Avoid intro-
ducing air bubbles or ice chips into the mixture, because they will decrease the
quality of the casts. Let the casting compound stand on ice for 30 s once it is
completely mixed. Draw up the casting compound into a chilled 3-cc syringe.
Expel any air and attach the syringe to the catheter (see Note 17). Return to the
mouse preparation.
26_Whiteley_371_392_F 382 8/29/05, 11:23 AM

Fig. 3. Cannulation of the umbilical vessel for perfusion and casting of the
fetoplacental vasculature. The body of the embryo supports the umbilical vessels while
the position of the placenta places tension on the vessels. Perfusate and casting com-
pound are infused into one umbilical vessel with the second umbilical vessel acting as
a vent.
12. In order to slow down the setting of the casting compound, rest the catheter sys-
tem on an ice pack or a bag of crushed ice. Keep the catheter system as cool as
possible during infusion of the casting compound.
13. While the assistant is preparing the casting compound, the surgeon must prepare
the sample for infusion of the casting compound. Ensure that the umbilical ves-
sels are stable and under slight tension. Position the sample by rotating the Petri
dish so that the work can be done comfortably. The casting compound will be
infused into the placenta through a hole cut in one of the umbilical vessels. A
hole cut in the other umbilical vessel will act as a vent. Cut holes in both the
umbilical artery and the umbilical vein approximately halfway between the em-
bryo and the placenta but a distance apart so that leakage from one hole does not
obscure the other hole (see Note 18). Do not sever either vessel completely. Blood
will flow out of the vessels as holes are cut. It is crucial that the surgeon keeps
track of the location of the holes in the vessels. It is often difficult to see the holes
once the blood has drained out of the vessels. After cutting both holes, drop 1–2
drops of warm 3% paraformaldehyde onto the umbilical vessels. This helps to fix
the holes in an open position.
3.2.2. Perfusion and Casting of the Placenta

The following steps must be performed by two people. After priming the
infusion cannula, the surgeon inserts the glass cannula into the umbilical ves-
sel and holds the cannula in place. The assistant will infuse the perfusate and
26_Whiteley_371_392_F 383 8/29/05, 11:23 AM

Fig. 4. Vascular corrosion casts of the fetoplacental vasculature in mice. The cast of
the embryonic circulation of the placenta shown in A and B was prepared by infusing
plastic into one umbilical vessel until it drained out the other. The casts in C and D
were prepared by partially filling the circulation via the umbilical artery (C) or umbili-
cal vein (D). (A) A scanning electron microscopy (EM) photograph of a complete,
undissected cast. The cast shows a central circular region and a peripheral labyrinthine
ring when viewed from the embryonic side. The branches arising from the umbilical
vein (v) usually overlay those from the umbilical artery (a) on the placental surface.
(B) The “tuft-like” capillaries of the labyrinth drain into veins at the edge of the cen-
tral circular region. (C) The umbilical artery branches into many widely distributed
arteries that branch a few times while traversing the labyrinth. At the mesometrial side
of the placenta, these arteries branch abruptly to form the capillaries of the labyrinth.
(D) The capillaries drain into evenly distributed, superficial veins located in the cen-
tral circular region.
26_Whiteley_371_392_F 384 8/29/05, 11:23 AM

casting compound into the preparation through the cannula via an attached
double lumen catheter (Fig. 5). The perfusate will flush the blood out of the
placental vessels prior to casting.
1. The glass cannula has a tapered tip. The surgeon should hold the cannula with its
tip next to the umbilical vessel that the cannula will be inserted into in order to
compare their diameters. The glass tip of the cannula should then be broken at the
appropriate spot along its length leaving a tip with a diameter that can be inserted
into the hole made in the umbilical vessel. Ideally, the cannula is easily inserted
into the vessel and the taper along the length of the cannula will seal the cannula
within the vessel. To break the tip of the cannula, hold the glass tip in the end of
a pair of Dumont forceps and close down around the tip. This will break off the
tip where it is held. It may be necessary to tap the tip of the cannula against the
edge of the forceps to remove jagged edges in the glass.
2. The assistant should then attempt to flush first perfusate and then casting com-
pound through the cannula to clear all air and to ensure that the tip of the cannula
is large enough to allow their passage. The casting compound is the more viscous
of the two solutions and will ultimately determine how large the tip of the can-
nula must be. The tip must be large enough to allow passage of the casting com-
pound but small enough that it can still be inserted into the hole in the umbilical
vessel. If the tip is not large enough to allow passage of the solutions, repeat
Step 1. Avoid contaminating the sample with casting compound during this
process.
3. After the cannula tip has been broken to an appropriate diameter, flush the cast-
ing compound out of the cannula using the perfusate.
4. The surgeon inserts the tapered tip of the cannula into the hole cut in one of the
umbilical vessels and advances the cannula in the direction of the placenta until
the vessel fits snugly around the cannula (Fig. 3) (see Note 19). The assistant
may exert a very slight pressure on the perfusate syringe to expel a minimal
amount of perfusate from the tip of the cannula during its insertion. This will help
to open up the hole in the vessel and aid in the insertion of the cannula. Often
there is already pressure built up within the catheter system from the flushing
procedure and it is unnecessary for the assistant to exert any additional pressure
during cannula insertion. It is critical that all pressures exerted with this system
are very slight to prevent damage to the delicate placental vessels. Pressure can
be added as required but it is impossible to reverse damage done by initially
using too much pressure.
5. When the cannula is in place, the assistant very gently pushes perfusate through
the catheter. The perfusate is expelled through the tip of the cannula and into the
umbilical vessel and the placenta. The perfusate will flush the blood out of the
placental vasculature via the vent in the other umbilical vessel. Carefully monitor
the size of the placenta during the infusion of the perfusate. If the placenta
appears to swell, immediately reduce the infusion pressure. When the fluid
flowing out of the vent is colorless, one can assume that the blood has been
26_Whiteley_371_392_F 385 8/29/05, 11:23 AM

386
26_Whiteley_371_392_F
386
Fig. 5. The double lumen catheter system used for casting the fetoplacental vasculature. (A) The glass cannula and the outer
tubing through which the casting compound passes. Note the tapered tip on the glass cannula and the distance between the end of
the inner tubing and the tip of the cannula in an assembled catheter, as shown in the detail. (B) The inner tubing through which the
perfusate passes. (C) The assembled catheter system.
8/29/05, 11:23 AM
Whiteley, Pfarrer, and Adamson
flushed out of the placenta. The vessels on the surface of the placenta should now
be cleared of blood.
6. Remove the syringe containing the perfusate from the catheter system. This pro-
vides a pressure vent within the system during infusion of the casting compound.
7. Start infusing the casting compound very slowly. A very slight pressure is all that
is required to begin moving the casting compound through the catheter. It should
take 10–20 s for the casting compound to begin to enter the placenta. Do not
become impatient waiting for the casting compound to begin moving down
through the cannula. When the casting compound begins to enter the umbilical
vessel, the assistant should stop pushing on the casting compound syringe momen-
tarily to ensure that excessive pressure is not being exerted on the placental vascu-
lature. Gradually begin to increase the infusion pressure to control the rate of
entry of the casting compound. Specks of color in the casting compound are moni-
tored to ensure the infusion pressure is high enough to sustain flow into the pla-
centa. Carefully monitor the size of the placenta during infusion of the casting
compound. If the placenta appears to swell, immediately reduce the infusion pres-
sure. The appearance of leaked plastic under the surface of the placenta is indica-
tive of a vessel rupture caused by the use of excessive pressure.
8. Continue infusing the casting compound until either the casting compound runs
out, the casting compound becomes too viscous to infuse, casting compound is
seen exiting through the vent, or until a rupture is observed. Ideally, casting com-
pound should be seen exiting through the vent before one is finished infusing the
casting compound and before it becomes too viscous to infuse. If a rupture is
observed, stop infusing the casting compound to avoid excess artefact caused by
plastic curing outside of the placental vessels. To obtain casts of the venous or
arterial sides of the circulation, infuse via the umbilical vein or artery respec-
tively and stop infusing the casting compound shortly after it enters the placental
microcirculation (Fig. 4C,D).
9. After infusing the casting compound, carefully remove the cannula from the um-
bilical vessel. Tighten a 7'0' silk tie around the umbilical vessels if there is con-
cern that the casting compound is still fluid enough to flow out of the placenta. In
order to reduce the amount of tissue to be digested, the embryo may be removed
at this time. Transfer the placenta to an appropriately sized vial or test tube and
allow a minimum curing time of 2 h.
3.2.3. Processing the Casts

1. After a minimum curing time of 2 h, immerse the placenta in 20% KOH to digest
the tissue off of the cast. Digest in small test tubes or larger glass vials depending
on the size of the samples. The tissue should be completely digested within 24–
48 h, depending on the size of the samples. If digestion is not completed in the
first 24 h, it can be aided by removing the old KOH (with a Pasteur pipet) and
adding fresh KOH.
2. Following complete tissue digestion, wash the cast with distilled water. The vas-
cular cast is delicate, so it is imperative that it be washed very carefully to avoid
26_Whiteley_371_392_F 387 8/29/05, 11:23 AM

damaging it. With a Pasteur pipet, add and remove at least three cycles of dis-
tilled water from the vial containing the cast. Ensure that all tissue is washed off
of the cast.
3. The cleaned cast can be stored in distilled water (see Note 20). For viewing with
light or SEM (see Note 21), casts must be air-dried. Dried casts should be stored
covered to protect them from damage and dust.
4. Notes
1. The casting methods detailed here are used to cast the entire uteroplacental vas-
culature (arteries, microcirculation, and veins). However, we have used modifi-
cations of these methods to cast the veins in blue and arteries in red. We
simultaneously infused red casting compound into the maternal aorta and blue
casting compound retrogradely into the maternal intra-abdominal inferior vena
cava. The uterus was exposed in a bath of warm PBS so that the clearing and
filling of the uterine vasculature could be visually monitored. A dual infusion
pump was used to simultaneously infuse the two colors of casting compound at
the same rate. The infusion was stopped once both colors were observed entering
the intrauterine vasculature. We also prepared partial casts of the venous or arte-
rial circulations by infusing via the vena cava or aorta respectively and stopping
before the casting compound exited from the microcirculation.
2. We use isoflurane inhalation anesthetic (5% for induction and 1.5% for mainte-
nance), but injectable anesthetics could be used. We use a Vaporstick Small Ani-
mal Anesthesia Machine with a Flush Valve and a Modified Jackson Rees
breathing circuit (Anesco, Georgetown, KY). Excess anesthetic is scavenged.
We have made a mouse facemask using the end connector from an endotracheal
tube. The mouse’s face fits snugly inside the larger end of the connector.
3. We lay our mouse on a small metal tray resting on a warming mattress. Our work
surface is heated from above with a warming light. The intent is to maintain the
mouse’s body temperature at 37–38°C and to maintain a normal blood pressure
and heart rate. This will help to ensure that the intracardiac heparin is circulated
throughout the mouse’s body before death.
4. Heparin is injected into the mouse to prevent clotting of the blood. It is critical
for the production of good quality casts to flush the blood from the vessels to be
cast. Injecting the heparin directly into the heart allows for quick circulation of
the heparin throughout the mouse’s body prior to death.
5. Xylocaine is also injected into the heart for immediate circulation. We have found
that xylocaine causes the cessation of respiration but the heart continues to beat,
circulating the heparin and xylocaine. The inclusion of xylocaine in the perfusate
enhanced the clearance of blood from the uterine vessels and improved vascular
filling possibly by “anesthetizing” the vascular musculature. Nevertheless, incom-
plete vascular filling of the uteroplacental circulation occurred in at least some
placentas of each mouse. Incomplete filling may be due to constriction of the
26_Whiteley_371_392_F 388 8/29/05, 11:23 AM

sphincter-like structures visible around the extrauterine arteries; they are some-
times partially constricted even in well filled casts (Fig. 1G).
6. Intracardiac injection can be performed using two methods. Both are performed
with the mouse lying on its back with its front paws secured to the metal tray.
Both methods are done before the chest cavity is opened. The intracardiac hep-
arin and xylocaine are combined and injected using a 1-cc plastic syringe and a
three-quarter-inch, 27-gauge needle. The needle is inserted through the mouse’s
skin either through the diaphragm lateral to the xiphoid cartilage and directed
forward and medially toward the heart or between the fifth and sixth ribs on the
left side and directed forward toward the heart. When the needle has been inserted,
aspirate a small volume of blood from the heart to ensure that the needle is in the
heart before injecting the heparin and xylocaine. Breathing will stop almost
immediately if this has been successful.
7. By not entering the abdominal cavity, one can check for punctures in the aorta
because the appearance of casting compound in the abdominal cavity would then
be the result of a leak through a vessel puncture or tear.
8. The perfusate is filtered through a 0.2-µm syringe filter to remove any particulate
matter that may be in the perfusate. Particles introduced into the blood vessels
during perfusion may interfere with the passage of the casting compound through
the catheter. Particles can also interfere with the polymerization of the casting
compound and can cause weak points or breakages in the casts.
9. The catheter tip must be cut blunt. If the tip of the catheter has sharp edges or if it
is cut to a point, it is likely to perforate the aorta as it is passes through the vessel.
10. The Jet Acrylic Liquid was added to the casting compound to decrease the overall
viscosity of the casting compound. Decreasing the viscosity improved the flow
of the casting compound and the filling of the placental microvasculature.
11. If a dried cast floats when returned to distilled water for storage, apply one to two
drops of 70% ethanol to the cast to reduce surface tension. We use a plastic
syringe and a 23-gauge needle to drop the alcohol onto the cast. The cast will
then sink in water. Use a minimal amount of alcohol to avoid having the alcohol
soften the cast.
12. To view the casts using SEM, we mount the casts on SEM stubs using 5-min
epoxy glue. The glue must dry overnight before the casts are sputtercoated with
gold. We have found that the casts require extensive coating with gold to elimi-
nate problems of charging when viewing the casts with the scanning electron
microscope. In order to adequately coat the casts with gold, we initially
sputtercoat the casts in an upright position. We then tip the stubs on their sides
and rotate the stubs in one-third turns through three cycles of sputtercoating.
13. We have also used a continuous drip of warm PBS to warm the embryo and
restore a heartbeat. We prefer the humidified warm air, because it does not cause
a build-up of liquid at the surgical site. Also, the force of the drops of PBS falling
onto the embryo can damage younger embryos. We do keep a supply of warm
26_Whiteley_371_392_F 389 8/29/05, 11:23 AM

PBS at the work site and have the option of using it along with the humidified
warm air to revive the embryo.
14. It is helpful to have the assistant record the orientation of the umbilical vessels
(e.g., which vessel is rostral) because it can be difficult to distinguish them once
the vessels are cut and the blood drains out.
15. We have found it useful to add a very slight hint of pigment (red) to our casting
compound. The color aids in the visualization of the casting compound as it passes
through the cannula and the placenta. The amount of pressure used to infuse the
casting compound is critical. Infusion pressures that are too high can rupture the
delicate vessels within the placenta, causing a leakage of the casting compound
throughout. With pressure that is too low, the casting compound sets before the
vasculature has filled. Being able to track the casting compound as it moves
through the cannula and the placenta allows for greater control over the rate and
pressure at which the casting compound is infused. Pigment is available in red,
white, yellow, blue, and green (Polysciences, Inc., Warrington, PA).
16. An excessive amount of pigment in the casting compound can interfere with the
curing of the casts. Add the least amount of pigment possible to achieve the desired
effect. Preliminary trials mixing small amounts of the casting compound with
varying amounts of pigment are useful to determine the amount of pigment
needed and the effect of varying amounts of pigment on the curing of the casting
compound.
17. Much of the refinement of this technique involved perfecting the double-lumen
catheter system used to infuse the perfusate and casting compound in sequence
(Fig. 5). The double lumen eliminated the need to remove and reinsert cannulae;
we found this extremely difficult because of the viscous, sticky properties of the
casting compound. The system is comprised of a double-lumen catheter and a
glass cannula with a tapered-tip.
a. To create the cannula, a borosilicate glass capillary tube (0.75 mm inner
diameter [I.D.], 1.0 mm outer diameter [O.D.]; F.H.C., Inc., Bowdoinham,
ME) is drawn to a fine point with a pipet puller (Model P-97 Brown-Flaming
Micropipette Puller, Sutter Instrument Company, Novato, CA). The large end
of the glass cannula is attached to 28 cm of Tygon microbore tubing (0.04 in.
I.D., 0.07 in. O.D., Cole Parmer Instrument Company, Vernon Hills, IL), and
glued in place with 5-min epoxy glue. This tubing is referred to as the outer
tubing and is the tubing through which the casting compound flows. An 18-gauge
needle is inserted into the other end of the outer tubing and is glued in place with
5-min epoxy glue. A 2-cm length of Silastic medical grade tubing (0.062 in.
I.D., 0.125 in. O.D., Fisher Scientific Ltd., Ottawa, Ont.) covers the junction
between the glass cannula and the outer tubing and is glued in place with 5-min
epoxy glue. The Silastic tubing helps to strengthen the junction between the
glass cannula and the microbore tubing.
b. The inner tubing of the catheter system is comprised of 40.5 cm of PE 10
tubing (Intramedic®, Becton Dickinson and Company, Parsippany, NJ). This
is the tubing through which the perfusate flows. The inner tubing is threaded
26_Whiteley_371_392_F 390 8/29/05, 11:23 AM

through a double male leur lock adapter (Cobe Canada, Ltd., Scarborough,
ON, Canada) with approximately 3 mm of the PE tubing left sticking out one
end of the adapter. The tubing is glued in place by filling the lumen of the
adapter with Silastic Medical Adhesive-Silicone Type A (Dow Corning Cor-
poration, Midland, MI). The Silastic adhesive must cure overnight. The PE 10
tubing is then threaded through a clear, four-way stopcock with male leur lock
(Cobe Canada, Ltd., Scarborough, ON, Canada) with ports that are all open. The
double male leur lock adapter is attached to the female port on the stopcock.
c. To assemble the catheter, the long end of the PE 10 tubing is threaded through
the 18-gauge needle on the outer tubing and passed down through the lumen
of the outer tubing. It may be necessary to trim the length of the PE 10 tubing
so that the tip of the PE 10 tubing sits approx 1–1.5 cm from the tip of the
glass cannula. The hub of the 18-gauge needle is then attached to the male
port on the stopcock. The 10-cc perfusate syringe is attached to the double
male leur lock adapter using a female-to-female leur adapter (Cole Parmer
Instrument Company, Vernon Hills, IL). The 3-cc casting compound syringe
is attached to the free port on the single stopcock.
18. We have used two methods to cut the holes in the umbilical vessels. The first is to
tear holes in the vessels using a pair of No. 55 Dumont forceps. Alternatively,
holes can be cut in the vessels using a pair of very delicate Vannas or spring
scissors.
19. We have produced excellent placental casts by infusing the casting compound
either into the umbilical vein or the artery. However, the umbilical vein seems to
provide more reliable filling.
20. If a dried cast floats when returned to distilled water for storage, apply one to two
drops of 70% ethanol to the cast to reduce surface tension. We use a plastic sy-
ringe and a 23-gauge needle to drop the alcohol onto the cast. The cast will then
sink in water. Use a minimal amount of alcohol to avoid having the alcohol soften
the cast.
21. To view the casts using SEM, we mount the casts on SEM stubs using 5-min
epoxy glue. The glue must dry overnight before the casts are sputtercoated with
gold. We have found that the casts require extensive coating with gold to elimi-
nate problems of charging when viewing the casts with the scanning electron
microscope. In order to adequately coat the casts with gold, we initially
sputtercoat the casts in an upright position. We then tip the stubs on their sides
and rotate the stubs in one-third turns through three cycles of sputtercoating.
Acknowledgments
We thank Dr. Bradley Smith for his valuable advice on fetal perfusion tech-
niques, Dora Chan for assisting in the establishment of embryonic casting
methods in our laboratory, Yong Lu for his work dissecting casts, and Doug
Holmyard for his SEM expertise. This work was supported by a grant from the
Canadian Institutes of Health Research.
26_Whiteley_371_392_F 391 8/29/05, 11:23 AM

References
1. Adamson, S. L., Lu, Y., Whiteley, K. J., et al. (2002) Interactions between tropho-
blast cells and the maternal and fetal circulation in the mouse placenta. Dev. Biol.
250, 358–373.
2. Smith, B. R. (2000) Magnetic resonance imaging analysis of embryos, in Devel-
opmental Biology Protocols, Volume 1 (Tuan, R. S. and Lo, C. W., eds.). Humana,
Totowa, NJ: pp. 211–216.
3. Pfarrer, C., Winther, H., Leiser, R., and Dantzer, V. (1999) The development of
the endotheliochorial mink placenta: light microscopy and scanning electron mi-
croscopical morphometry of maternal vascular casts. Anat. Embryol. (Berl.) 199,
63–74.
4. Gannon, B. J. (1978) Vascular casting, in Principles and Techniques of Scanning
Electron Microscopy, Volume 6 (Hayat, M. A., ed.). Van Nostrand Reinold, New
York, NY: pp. 170–193.
26_Whiteley_371_392_F 392 8/29/05, 11:23 AM

Fetal and Maternal Microvasculature 393
27
Analysis of Fetal and Maternal Microvasculature
in Ruminant Placentomes by Corrosion Casting
Rudolf Leiser and Christiane D. Pfarrer
Summary
Vascular corrosion casting is a useful tool for studying the vascular architecture of complex
organs. The synepitheliochorial placenta of ruminants is composed of two closed blood cir-
cuits, a fetal and a maternal one. The microvasculature of each circuit has the shape of the
corresponding cotyledon (villous trees) and caruncle (crypts). These two compartments inter-
digitate with each other in a complementary fashion. Understanding three-dimensional vascu-
lar arrangements is facilitated by scanning electron microscopy of vascular corrosion casts.
Methods to be used in the generation of vascular casts from fetal and maternal placentomal
blood vessels are described, with special emphasis on casting resins and corrosion using potas-
sium hydroxide. The procedure of splitting larger casts following gelatin embedding and freez-
ing is also presented.
Key Words: Placenta; ruminants; corrosion cast; microvasculature.
1. Introduction
The application of scanning electron microscopic analysis to microvascular
corrosion casting (1) has enormously widened the scope of morphological
blood vessel research. Novell and coworkers first applied this technique to the
avian lung (2). In the placenta, this method was introduced by MacDonald in
the epitheliochorial pig 1976 (3), then, it was modified by Leiser and cowork-
ers for many other species with different placental types in comparative pla-
cental research (4–6).
The placenta is a unique organ with maternal and fetal blood vessels devel-
oping in close proximity. The coordinated establishment of these vascular sys-
tems is essential for materno–fetal exchange (7). Three-dimensional
materno–fetal capillary interrelationships vary among species (4,6). Vascular
organization at the materno–fetal interface influences diffusion efficiency and
thus nutrient delivery to the developing fetus (8–10).
393
27_Leiser_393_408_F 393 8/29/05, 11:24 AM

394 Leiser and Pfarrer
Fig. 1. Cracked materno–fetally combined vessel cast with Batson-filling of goat

placenta; day 120 of pregnancy. Irregularly shaped placentomes are supplied by ma-
ternal vessels on their convex side; whereas on their concave side, there are fetal
allantochorionic vessels. The arrows point to the main direction of maternal (left) and
fetal (right) intraplacentomal stem vessels. From ref. 4, with permission.
The three-dimensional organization of placental vasculature is extensive,

extending throughout the placenta (see ref. 4 and Fig. 1). Furthermore, the
interrelationships of the maternal and fetal vasculature provide insights into
mechanisms underlying the anchoring of the conceptus within the uterus (6).
Placental attachment is simple in epitheliochorial placental types, but becomes
increasingly more complex in endotheliochorial and hemochorial types of pla-
centation (6,10,11).
In this chapter, we provide methodology for examining the three-dimen-
sional organization of the placentomal vasculature of synepitheliochorial pla-
centae from three ruminants (cattle, goat, and sheep) using vascular corrosion
casting. Methodologically, this is not easy, because fetal and maternal com-
27_Leiser_393_408_F 394 8/29/05, 11:24 AM

partments interdigitate with each other in a villous-crypt pattern, which is mor-

phologically the most complicated placental type within the epitheliochorial
category (6). The villous structure of the placenta in both ruminants and humans
is determined by the architecture of the underlying blood vessels (12). The gen-
eration of accurate and scientifically useful vascular casts requires experimen-
tal precision.
2. Materials
2.1. Anesthetics for Small Ruminants—Goat and Sheep
1. During the initial phases of the experimentation the animals are administered a
tranquilizer, Rompun® (2 mL, intramuscularly, Bayer, Leverkusen, Germany)
and a barbiturate (Narcoren®, Merial, Hallbergmoos, Germany, intravenously).
2. Barbiturates are used for prolonged anesthesia (20–30 mg/kg body mass) and
eventual euthanasia (50–60 mg/kg body mass) during perfusion of the placenta
(intravenously).
2.2. Solutions for Perfusion of Vascular Systems

Phosphate buffer, 0.1 M, pH 7.3, containing 1 mL/L heparin as an antico-
agulant (Liquemin®, 10,000 IU, Roche, Basel, Switzerland), and 0.5% procaine
(M Curasan, Aventis Pharma, Bad Soden, Germany) for vasodilation. The
buffer is used at 38°C and at 4°C.
2.3. Liquid Plastic Instillation Media for Casting (Toxic—Store and Use
in a Fume Hood)
2.3.1. Batson® No. 17 Compound (Polysciences, Warrington, PA)
With Three Components (1–3) and Sevriton® (From Any Dental Supplier)
1. Monomer Base solution, 25 mg, 4°C.
2. Catalyst, 7.5 mL, room temperature.
3. Promoter, 0.5 mL, 4°C.
4. Sevriton® (available from any dental supplier), 12 mL, 4°C.
5. Preparation of the casting material. Mix the Monomer Base solution, the Cata-
lyst, and Sevriton as a minimal batch (45 mL) with the Promoter added immedi-
ately before use.
2.3.2. Mercox CL-2R® (Vilene, Tokyo, Japan)

1. Catalyst, 0.5 g, room temperature.
2. Base resin, 20 mL, room temperature.
3. Red and blue dyes.
4. Preparation of the Mercox CL-2R®. The Catalyst and Base resin are mixed im-
mediately before use. The dyes provided with the kit are not necessary for scan-
ning electron microscopy. When stirring each of the mixtures take care not to
generate air bubbles (see Note 1).
27_Leiser_393_408_F 395 8/29/05, 11:24 AM

2.4. Solutions/Liquids for Corrosion Procedure and Cutting/Cracking

of Casts
1. Tap water for water baths, 20°C and 60°C.
2. 40% KOH (Sigma, Taufkirchen, Germany) at 60°C.
3. Distilled water, room temperature.
4. N2 (in liquid form), –196°C.
5. 20% Gelatin, dissolved in water (Sigma, Taufkirchen, Germany) at 50°C.
6. 10% Extran®-detergent (Merck, Darmstadt, Germany) at room temperature.
2.5. Scanning Electron Microscopy of Vascular Casts

1. Stereo-microscope.
2. Specimen holders (stubs) for scanning electron microscopy (Plano GmbH,
Wetzlar, Germany).
3. Conductive adhesive carbon cement (Goecke, Neubauer Chemikalien, D-4000
Münster, Germany).
4. Sputtercoating device to cover the vascular cast with gold or platinum (in argon-
vacuum chamber; Balzer, FL-Lichtenstein).
5. Scanning electron microscope, including equipment and supplies for photog-
raphy.
3. Methods
3.1. Preparation of the Animals
1. For each pregnant animal it is essential to have an accurate history, including
age, breed, feeding records, health status, as well as the exact gestational age,
which should be recorded as day and h of insemination and/or, later in the study,
by measuring the crown–rump length (CRL) of the fetus (13).
2. The pre-experimental treatment of animals is different for cows vs goats or sheep.
In cows, which are generally slaughtered for commercial purposes, the pregnant
uterus must be excised as soon as possible after killing of the animal. The uterus
is opened and depending on the stage of gestation the fetus may be delivered
alive or is euthanized. Goats and sheep are usually sacrificed exclusively for
experimental purposes. Initially, the animal is tranquilized and anesthetized
(see Note 2) and the uterus is left inside the body of the animal at the beginning
of the experiment. As buffer perfusion (discussed later) is initiated, the anesthe-
sia is increased, and finally the animal is euthanized by an intravenous overdose
with the same anesthetic, and the uterus (including the placenta) excised (dis-
cussed later).
3.2. Vessel Casting

Vessel casting in ruminants is restricted to a small region of the placenta
(see Notes 3 and 4). For routine research, the most developed placentomes are
preferred, which are located nearby the umbilical cord (Fig. 1). The size of the
27_Leiser_393_408_F 396 8/29/05, 11:24 AM

area should not be too large, because injected fluids (buffers and liquid plastic)
may be lost in too big a volume of blood vessels. On the other hand, too small
of an area may favor extravasation of these solutions, caused by rupture of
vessel walls, due to high and/or uncontrolled pressure (see below) in the selected
area. The vascular casting of ruminant placentomes may be applied to maternal
blood vessels, fetal vasculature, or both maternal and fetal systems.
1. The method is initiated by perfusion with phosphate buffer. To do so, well visible
arteries (which are preferred to veins) with a caliber not less than 0.3 mm are
cannulated. The maternal arteries at the external side of uterus must be carefully
prepared.
2. The fetal vasculature is casted, after opening the uterus, through the umbilical
arteries or their distinct branches.
3. For combined materno–fetal casts overlapping areas must be perfused in parallel.
4. The vessels should be cannulated as close as possible to the selected area. The
selected areas are perfused with phosphate buffer (38°C) by syringe (20 mL) and
cannula using manual pressure (about 10 mL/min) for a short time (about 1 min
until “blanching”; discussed later), then, the buffer is changed to ice-cold (4°C)
phosphate buffer to retard postmortem degenerative changes of the vessel wall
and too slow polymerization of the instilled liquid plastic resin. Putting the whole
uterus on ice or into an ice-cold water bath can also enhance the cooling effect.
5. Successful perfusion is visible as “blanching” of the placentome, because the
buffer replaces the blood in placental areas, which are tributaries of the cannu-
lated vessel(s). Buffer perfusion of the maternal system may not be visible,
because of the highly anastomosed vasculature; therefore the selected area must
be delimited by clamps.
3.3. Instillation of Liquid Plastic
The instillation of liquid plastic into the buffer-perfused vascular areas of
ruminant placentomes may be conducted with two different resins, either
Batson/Sevriton (5) or Mercox. Both resins work well in vascular casting ex-
periments but require different protocols (14,15).
3.3.1. Batson/Sevriton Protocol
1. Batson No. 17 compound/Sevriton has a hydrophilic quality resulting in very
detailed casts, mirroring exactly the luminal side of the vessel wall by scanning
electron microscopy (including endothelial cell impressions, discussed later). Its
viscosity, however, is relatively high; therefore, it cannot be used with cannulas
below 0.7 mm diameter and an instillation rate of more than 5 mL/min by manual
pressure (see Note 5).
2. The different components of the resin are mixed immediately before use (see
Section 2).
3. The instillation of the liquid plastic takes up to several min for filling the buffer-
perfused placental area through the same arteries (veins) as used for the buffer
27_Leiser_393_408_F 397 8/29/05, 11:24 AM

perfusion until the venous outflow consists of pure plastic. Instillation of this
plastic resin is limited to 3 to 5 min before polymerization of the plastic starts.
4. Because this period of time may be too short for a complete filling of the per-
fused vascular system, the plastic and all other materials (buffer), tools (syringes,
cannulas, scissors, clamps, knives) and the tissue itself (discussed earlier) must
be cooled down (4°C) before use. This procedure extends the time for instillation
up to approx 8 min.
3.3.2. Mercox Protocol

1. Mercox CL-2R has a hydrophobic, rather oily quality and is much less viscous
than Batson resin. Therefore, it may be applied with cannulas of smaller diameter
(>0.3 mm). Like the Batson resin, each batch must be freshly prepared immedi-
ately before injection. Mercox can be handled at room temperature, because it
polymerizes slower, leaving up to 10 min for instillation. However, with Mercox
the finest details of the interior of the vascular walls can only be shown by stable
instillation pressure inside the vascular system (see Note 5). That means that, at
the end of vascular instillation, all vessel outlets of the system must be carefully
clamped to prevent efflux of the resin.
2. In case of combined materno–fetal casting, the instillation of liquid plastic must
be done simultaneously with about the same pressure in both systems in order to
prevent unequal filling (Fig. 2).
3. A successful instillation of plastic in the selected area is soon visible, because the
color changes to yellowish and can also be detected by palpation, because the
rigidity of the tissue increases significantly.
3.4. Corrosion Procedure
1. For the corrosion procedure, the plastic instilled placental areas must be excised
from the uterus.
2. The plastic is allowed to polymerize or harden completely in a 20°C water bath
for 30 min, which is followed by warming in a 60°C water bath for 4 h.
3. Subsequently, the preparation is immersed in the corrosion solution (40% KOH
at 60°C).
4. The resin-instilled tissues are then alternately incubated with distilled water at
60°C or the corrosion solution (at least twice a day) until the yellow-whitish
“crude” casts become visible.
5. Careful rinses with running tap water accelerate the cleaning process, but are
only recommended with compact casts.
3.5. Preparation of the Casts for Scanning Electron Microscopy
1. Prior to analysis by scanning electron microscopy, the casted areas—whole
placentomes (Fig. 1), caruncles (Fig. 3), and cotyledons (Fig. 4)—may require
cutting into smaller pieces.
2. To avoid uncontrolled breaking of the casts, they are first embedded in 20% gela-
tin solution (50°C), which hardens when cooled to about –5°C in a freezer. In this
27_Leiser_393_408_F 398 8/29/05, 11:24 AM

Fig. 2. Goat placental cast. Magnification of Fig. 1. (A) Part of a perpendicular

section across a placentome with fetal (FS) and maternal side (MS). (B) Detail from
(A) showing: fetal arterioles (FAl), fetal capillary coilings (F) and maternal capillary
network (M). (C) Cross-section of a placentome corresponding to (B). Maternal capil-
lary network forming basket-like crypts (M). Within two of the crypts a fetal arteriole
(FAl) and capillary coilings (F) can be observed at the cracked surface of the cast.
From ref. 5, with permission.
state, all the little plastic branches are fixed and are then cut with a knife.
3. To achieve the smoothest fracture lines, the gelatin-embedded casts can be fur-
ther hardened in liquid nitrogen (N2) and cracked with hammer and knife. How-
ever, the disadvantage of this method is that it is not as easy to control (see Note 6).
4. After thawing, the gelatin is removed by a second corrosion procedure, which
follows the same regime like the first (see Subeading 3.4.).
5. Although the gelatin treatment facilitates cleaning, the casts, again, need a very
thorough and repeated washing in distilled water at room temperature, followed
by washes in 5% Extran solution, and distilled water again in order to have all
remnants of tissue removed from very dense vascular casts.
6. The cut or cracked casts are air-dried and suitable specimens are selected by
means of stereomicroscopy and mounted onto stubs using fast-drying conductive
adhesive carbon cement.
7. The specimens are dried again and stored in a dust-free dessicator until used.
8. Finally, the casts are sputter-coated with gold or platinum (30 nm) and examined
with a scanning electron microscope (see Note 7).
27_Leiser_393_408_F 399 8/29/05, 11:24 AM

Fig. 3. Perpendicularly cut maternal vessel casts of bovine placentomal caruncles

with Mercox-filling. (A) Overview of the intercaruncular (arrows) and the caruncular
vasculature (top right) which extends from the stalk (below right) to the base of the
caruncle (CB); fetus of 8-cm crown-rump length (CRL). Note spiral arteries running
from the caruncle stalk to the CB. Blood vessels of main septa (S) and primary crypts
(asterisks) are indicated. (B) Detail of a caruncle stalk; fetus of 23-cm CRL. The coil-
ing of the spiral artery (A) contrasts with the veins (V) and venules (VI), which are
straighter and have a relatively larger diameter. Impressions of endothelial perikarya
can be observed (arrowheads). A and B from ref. 15, with permission.
4. Notes
1. Air bubbles must be avoided. Air bubbles can enter the vascular system during
the buffer and resin perfusion, especially when cannulas are incorrectly inserted
into the vessels or when the resin is mixed or stirred too vigorously. Once air is
trapped inside the vascular system there is no way to remove the air. Air bubbles
cause incomplete formation of vascular casts. Vascular casts with air bubbles do
not represent a replica of the vessel and may be more susceptible to fracture or
damage. Air bubbles tend to be a greater problem when using Mercox.
2. The slaughter process and anesthesia require careful attention. Anesthesia during
in situ perfusion of the placenta in goats and sheep should be maintained as long
as possible, since the solutions are driven through the body with the help of the
heart. Animals should be euthanized at the time of infusion of the toxic com-
pounds (liquid plastic resin and fixatives; see Note 3).
27_Leiser_393_408_F 400 8/29/05, 11:24 AM

Fig. 4. Appearance of casts with (A) and without (B) extravasations. (A) Corrosion
cast of bovine fetal cotyledonary villous tree (day 220 of pregnancy) where only the
villous tip shows the vasculature (lower left). All other parts have a sculpture-like
appearance resulting from extravasation of Batson resin. The complex ramification of
intermediate villi into terminal villi is clearly visible (arrows). (B) Ovine fetal cast
with Mercox-filling; day 140 (near term) of pregnancy. The stem vessels, fetal stem
artery (FSA), and vein (FSV), are usually found within the center of a vascular tree.
Few arterioles branch off the stem arteries (arrows) to enter the capillary system, where
it is easy to follow single vessels until they enter numerous venules (arrowheads),
which run towards the stem of the vascular tree in a parallel manner. The images in A
are from ref. 16, with permission; the images in B are from ref. 18, with permission.
3. Perfusion-fixation with formalin (2%) or with other fixatives may minimize post-
mortem degeneration of placental vessel walls and enhance their rigidity, which
allows casting of early fetal placental stages. Extravasation during casting with
Mercox may be decreased following fixation. Fixation prior to resin instillation
has its limiations. It takes time and increases handling stress of the delicate tis-
sue. Consequently, the most successful approach is often a short buffer perfu-
sion, quickly followed by the instillation of plastic.
4. In order to achieve good casting results, the size of the selected area should be
restricted to less than 10 × 10 cm (Figs. 4 and 5). The vasculature at the periphery
of the selected area must be adequately clamped. It is important that care is taken
27_Leiser_393_408_F 401 8/29/05, 11:24 AM

Fig. 5. Bovine fetal BatsonR cast showing serially linked capillary loops of termi-
nal villi day 270 (near term) of pregnancy. Highly anastomosing capillaries in the
neck region (top area) of three terminal villi are serially bridged (dashed lines). A
fourth terminal villus is obviously still growing (arrow). Note the increase of the cap-
illary diameter towards the terminal loops, which form sinusoidal dilations (stars) at
the tips of the loops. The image is from ref. 12, with permission.
Fig. 6. (opposite page) Schematic drawings of bovine placentomal microvascula-

ture in early (A) and late (B) gestation. (A) The maternal, caruncular part, at around
day 100 of pregnancy, already shows all typical components. (B) Fetal, cotyledonary
part, near term. Villous trees in full-grown (b1) and budding (b2) stages illustrating a
cotyledonary artery (1) and vein (2) of the chorionic plate, stem arteries (3) and veins
(4) of a stem villus; branching arteries/arterioles (5), arterioles of intermediate villi
(6), and capillary convolutions of terminal villi consisting of arterial capillary limbs
(7), capillary loops (8) with dilations (8a) and anastomoses (8b), and venous capillary
limbs (9), capillaries or venules serially connecting capillary convolutions (9a),
venules of intermediate villi (10), and branching venules/veins (11). The drawings are
from ref. 15 and 17, with permission.
27_Leiser_393_408_F 402 8/29/05, 11:24 AM

403
27_Leiser_393_408_F 403 8/29/05, 11:24 AM

in the perfusion of buffer and instillation of resin. This can be accomplished by

using manual pressure. High pressure will result in extravasation of plastic out-
side the vasculature, whereas low pressure results in partial filling and the forma-
tion of incomplete casts that readily crumble. The size of the batch of resin must
be accurately matched with the size of the perfused area. It is not possible to mix
and inject another batch of resin into the same area at a later time. These instruc-
tions are particularly important for the method of materno–fetally combined cast-
ing. Fetal vascular casting of early placentae is problematic because the vessels
being cannulated are small and fragile.
5. Extravasation of plastic is a general problem in casting blood vessels. Effective-
ness of the instillation of the resin is impacted by the developmental stage of the
uteroplacental tissue and its water content. Early in gestation, the blood vessels
of fetal cotyledons have a soft consistency with high water content and are more
susceptible to extravasation when high perfusion and instillation pressures are
used. Extravasations of resins outside the vessels may give a false impression of
vessel architecture (Fig. 4A). The use of a pump for controlling instillation pres-
sure is not recommended. The resins have a high viscosity and there is significant
variability among animals. The best results in combined materno–fetal vessel
casting have been achieved by using manually controlled instillation pressures.
6. For a complete understanding of the three-dimensional placental vascular archi-
tecture, splitting the vascular casts along or across the materno–fetal plane may
be necessary (Figs. 1 and 2). Such preparations can be best generated via embed-
ding the vascular casts in gelatin. The frozen (–5°C) gelatin-embedded vascular
casts are transparent. The gelatin blocks are scored along the plane of the desired
cuts. The gelatin blocks are then placed in liquid N2 and then the casts are split
along the chosen plane. After thawing, gelatin is removed from the casts by a
second corrosion process. The second corrosion treatment greatly facilitates the
cleaning of any remaining tissue from the casts. The more precise the orientation
of the gelatin block fractures, the more helpful they will be for analysis.
7. Examples of combined materno–fetal placental vascular corrosion casts in rumi-
nants are presented in Figs. 1 and 2 and further clarified in the schematic drawing
of Fig. 6. The characteristics of either fetal or maternal blood vessels are shown
in Figs. 4 and 5 and Table 1 (4,5). The interrelationship of materno–fetal tissues
is evident in Figs. 2C and 6A,B (6). Materno–fetal blood flow is a mixture of
crosscurrent and countercurrent. In comparison with that other mammals, rumi-
nant placentation provides a relatively efficient transfer process between mother
and fetus (10,16). The materno–fetal contact surface for transplacental exchange
increases with the progression of pregnancy (6,9). Additionally, the materno–
fetal interhemal distance or placental barrier can be measured directly on
materno–fetally combined casts (Fig. 2C). The functional significance of this
parameter has been discussed by many authors (17) and is associated with an
increasing substance transfer efficiency from epitheliochorial to endotheliocho-
rial to hemochorial types of placentation (10,11).
27_Leiser_393_408_F 404 8/29/05, 11:24 AM

27_Leiser_393_408_F
Table 1
Vascular Characteristics by Corrosion Casting of Ruminant Placenta
405
Superficial
Network/coiling Size/form capillary Materno–fetal
Classification of capillary of capillary structure interhemal
of placental type Architecture (vascular) system cross-section (roughness) distance
macroscopic microscopic
Fetal and Maternal Microvasculature
Multiple placentomes Maternal Caruncles Branched crypts Irregular-coarse- Irregular Smooth Highly variable
(Fig. 1) (Figs. 3A, 6A) (Fig. 3A) network (Fig. 2B,C) (Figs. 2B, 5) (Fig. 2C)
(Fig. 2B,C)
Epitheliochorial Fetal Cotyledons Branched Terminal Roundish Rough
(10,16) (Fig. 6B) Villi (Fig. 4) coilings approx 8 µm (Fig. 2B,C)
(Figs. 2B, (Figs. 2B
4B, 5) 4B, 5)
8/29/05, 11:24 AM
405
Studying the angio-architecture of blood vessels by corrosion casting requires

clear and exact methodological standards and extensive scanning electron micro-
scopic examination. Such an approach permits the re-creation of the three-dimen-
sional vascular organization of the ruminant placenta, as shown in Fig. 6A,B
(12,15,16,18).
References
1. Aharinejad, S. H. and Lametschwandtner, A. (eds.) (1992) Microvascular Corro-
sion Casting in Scanning Electron Microscopy. Techniques and Applications.
Springer-Verlag, Wien.
2. Nowell, J., Pangborn, J., and Tyler, W. S. (1970) Scanning electron microscopy
of the avian lung. Scan. Electron Microsc. 1, 249–256.
3. MacDonald, A. A. (1976) Uterine vasculature of the pregnant pig: a scanning
electron microscope study. Anat. Rec. 184, 689–697.
4. Leiser, R. and Koob, B. (1992) Structural and functional aspects of placenta mi-
crovasculature studied from corrosion casts, in Scanning Electron Microscopy of
Vascular Casts: Methods and Applications (Motta, P. M., Murakami, T., and
Fujita, H., eds.). Kluwer, Boston: 261–277.
5. Leiser, R., Dantzer, V., and Kaufmann, P. (1989) Combined microcorrosion casts
of maternal and fetal vasculature. A new method of characterizing different pla-
cental types, in Developments in Ultrastructure of Reproduction (Motta, P. M.,
ed.). Alan R. Liss, New York: 421–433.
6. Leiser R., Pfarrer C., Abd-Elnaeim M., and Dantzer V. (1998) Feto-maternal an-
chorage in epitheliochorial and endotheliochorial placental types studied by his-
tology and microvascular corrosion casts. Troph. Res. 12, 21–39.
7. Moll, W. (1981) Theorie des plazentaren Transfers durch Diffusion, in Die
Plazenta des Menschen (Becker, V., Schiebler, T. H., Kubli, F., eds.). Thieme,
New York: 140–152.
8. Carter, A. M. (1975) Placental circulation, in Comparative Placentation (Steven,
D. H., ed.). Academic, New York: 108–160.
9. Faber, J. and Thornburg, K. (1983) Placental Physiology. Structure and Function
of Feto-Maternal Exchange. Raven, New York.
10. Leiser, R. and Kaufmann, P. (1994) Placental structure: in a comparative aspect.
Exp. Clin. Endocrinol. 102, 122–134.
11. Mossmann, H. W. (1987) Vertebrate Fetal Membranes. Macmillan, Basingstoke, UK.
12. Leiser, R., Krebs, C., Ebert, B., and Dantzer, V. (1997) Placental vascular corro-
sion cast studies: a comparison between ruminants and humans. Microsc. Res.
Tech. 38, 76–87.
13. Richter, J. and Goetze, R. (eds.) (1978) Tiergeburtshilfe. Parey, Berlin.
14. Abd-Elnaeim, M. M., Miglino, M. A., Pfarrer, C., and Leiser, R. (2003) Microvas-
cular architecture of the fetal cotyledons in water buffaloes (Bubalus bubalis) dur-
ing different stages of pregnancy. Ann. Anat. 185, 325–334.
27_Leiser_393_408_F 406 8/29/05, 11:24 AM

15. Pfarrer, C., Ebert, B., Miglino, M. A., Klisch, K., and Leiser, R. (2001) The three-
dimensional feto-maternal vascular interrelationship during early bovine placen-
tal development: a scanning electron microscopical study. J. Anat. 198, 591–602.
16. Leiser, R., Krebs, C., Klisch, K., et al. (1997) Fetal villosity and microvasculature
of the bovine placentome in the second half of gestation. J. Anat. 191, 517–527.
17. Grosser, O. (1927) Frühentwicklung, Eihautbildung und Placentation des
Menschen und der Säugetiere, in Deutsche Frauenheilkunde, Geburtshilfe,
Gynäkologie und Nachbargebiete in Einzeldarstellungen (Jaschke, R. T., ed.).
Bergmann, Munich.
18. Krebs, C., Longo, L. D., and Leiser, R. (1997) Term ovine placental vasculature:
comparison of sea level and high altitude conditions by corrosion cast and
histomorphometry. Placenta 18, 43–51.
27_Leiser_393_408_F 407 8/29/05, 11:24 AM

27_Leiser_393_408_F 408 8/29/05, 11:24 AM
Trophoblast Microarrays 409
V
MOLECULAR ANALYSIS AND GENE TRANSFER
TECHNIQUES
28_Budhraja_409_424_F 409 8/29/05, 11:24 AM

410 Budhraja and Sadovsky
28_Budhraja_409_424_F 410 8/29/05, 11:24 AM

28
Microarray Analysis of Trophoblast Cells
Vikram Budhraja and Yoel Sadovsky
Summary
A complex repertoire of trophoblast gene products governs the multifaceted functions per-
formed by the placenta during the relatively short period of pregnancy. Cloning and sequencing
the human as well as other mammalian genomes allow investigators to gain better insight into
the function of trophoblast genes. Our ability to identify transcripts by their nucleotide sequences
and determine their expression patterns enables us to glean information on gene function.
Although the molecular principles underlying microarray are not new to biology, the high
throughput, low reaction volumes, fluorescent labeling, accurate detection, and robust analysis
software makes this approach most appealing to today’s researchers, when compared with stan-
dard filter blotting techniques. This chapter focuses on DNA microarray of the human placental
transcriptome as a means to identify alterations in gene expression in different physiological or
pathological conditions.
Key Words: Placenta; trophoblasts; oligonucleotide microarray; cDNA microarray;
transcriptome; RNA; variability; precision; normalization; replicates.
1. Introduction
Several techniques have been available for analysis of differential gene
expression. These include Northern analysis, S1 nuclease protection, differ-
ential display, subtraction hybridization, representational difference analysis
(RDA), library screens, and serial analysis of gene expression (SAGE). Over
the recent years, additional technologies, including microarray, have been
added to the arsenal of the investigator, allowing research into the presence of
a large number of transcripts in a tissue or a dynamic probe into quantitative
changes in expression between different tissues, normal or diseased tissue, or
among cells exposed to single or several conditions (1). Array technology has
recently improved to enable a greater number of elements per chip surface
area. Currently, the entire human transcriptome is represented on the surface
area of modern chips (2–6 cm2). Arrays are available for more than 100 organ-
411
28_Budhraja_409_424_F 411 8/29/05, 11:24 AM

isms, and arrays from one organism are currently used to examine expression
in phylogenetically related species.
DNA microarray is based on the fact that matrix-immobilized DNA sequences
bind their complementary transcripts in a highly sensitive and specific manner
(recently reviewed in refs. 2–5). Manufacturing approaches to DNA arrays
include either “on slide synthesis,” in which DNA sequences are assembled
directly on the array substrate, or delivery approach, in which pre-assembled
oligonucleotides or cDNAs are attached to the microarray. Most noncommer-
cial array producers utilize the second approach, as long as a contaminant-free
chamber with regulated temperature and humidity levels can be assured. Hybrid-
ization takes place between the glass-bound arrayed polynucleotide sequences
and fluorescently labeled cellular mRNA, or after RNA conversion into cDNA
or cRNA using in vitro transcription (IVT). After extensive washings, the rela-
tive abundance of every transcript from each of the two populations is imaged
with a digitally operated confocal-scanning microscope, and analyzed using
image analysis algorithms.
Expression profiling is based on one-dye or two-dye approaches (Fig. 1). In
the one-dye approach, each RNA sample is labeled with a single label (e.g.,
phycoerythrin), and signals from the two experiments are compared for deter-
mination of gene activation or repression. In the two-dye approach, each of the
two RNA populations is labeled with a different fluorescent dye, such as cya-
nine-3 and cyanine-5 (Cy3 and Cy5), applied to a single chip. Cy3 and Cy5
dyes are stable and easy to incorporate into the probe during reverse transcrip-
tion. In addition, these dyes emit easily separable spectra. A fluorescent two-
channel dye detector, coupled to photomultiplier tubes (PMT) or a charged-coupled
device (CCD), scans each array for emitted signals. The expression level of a
gene is determined by the ratio of Cy3/Cy5 signals. Detector elements are
designed to yield a clear signal from robotic imaging of the arrayed probes.
Sampling of signal-to-noise ratio defines the threshold above which a signal is
determined positive. All experiments are typically pseudo first-order reactions,
in which there is a large excess of the immobilized probe relative to the labeled
target, and thereby eliminating the problem of probe competition. Overall, dif-
ferences in signal intensity represent the amount of applied target mRNA, tar-
get labeling, hybridization efficiency, efficiency of fluorescent excitation and
emission, and detection efficiency. Signals are converted to an expression level,
which is saved in a tab-delimited format and can be imported into spreadsheets
for further analysis. Results are reported as relative changes in a set of ex-
pressed genes, defining one transcriptome as control.
Commonly used microarray technologies include oligonucleotide arrays and
cDNA arrays (1,6). In oligonucleotide array technology, short (15–25) or long
(50–120) nucleotides are attached to the glass matrix. Sequences can be designed
28_Budhraja_409_424_F 412 8/29/05, 11:24 AM

Fig. 1 (see companion CD for color version). Differential gene expression can be
determined using a one-dye or two-dye approach. Computerized analysis of signal
intensity between two arrays (one-dye) or of combined signals in one array (two-dye)
indicates differences in gene expression between two experimental paradigms.
to identify mRNA expression or DNA polymorphism. Selection of probes for

the array is based on existing databases, such as Genbank, dbEST, and
UniGene. Although oligonucleotide arrays are easier to optimize for binding
specificity and melting temperature, they are associated with a greater risk of
nonspecific hybridization, particularly by abundant transcripts. This risk is
reduced by development of a series of oligonucleotides for each transcript
such that hybridization is determined to a series of specific oligonucleotides
(“match”) as well as to oligonucleotides that harbor a mutation within the cen-
ter region of the oligonucleotide (“mismatch”), designed to bind equally well
to nonspecific transcripts. Therefore, analysis of these arrays includes signal
comparison between transcripts bound to the series of “match” oligonucleotides
and those bound to the series of “mismatch” oligonucleotides. The presence of
multiple oligonucleotides (usually 16–20) representing each transcript also
reduces the risk of altered signal intensity due to base composition. In cDNA
array technology, whole or partial cDNA fragments (0.6–2.4 kb each, dissolved
in nL volume) derived from the 3'-end of the transcripts are robotically placed
28_Budhraja_409_424_F 413 8/29/05, 11:24 AM

Fig. 2. An overview of design principles guiding microarray experiments.
on a glass slide. The technology is commonly performed using two-color fluo-

rescence and mixed samples (at a ratio of 1:1), which diminishes intra-experi-
mental variations. Whereas cDNA technology tends to produce strong specific
signals, it requires full knowledge of transcript sequences. Additionally, cDNA
arrays tend to form double strands, which may reduce binding capacity. This
problem may be resolved with the use of heat or alkali.
Although microarray is a powerful high-throughput technology designed to
identify changes in gene expression, success of microarray screens depends on
carefully crafted experimental design and realistic expectations (discussed
later). The hypothesis as well as choice of array should be influenced by exist-
ing transcriptome databases that focus on similar physiological questions and
tissues. Data derived from the expression screen should be analyzed based on
updated data warehouses (Fig. 2). Importantly, results regarding the rela-
tive expression of mRNA should be confirmed by additional, down-stream
approaches. Northern blot, although commonly regarded as a gold standard, is
relatively time-consuming and cumbersome, and therefore less efficient. Real-
time quantitative polymerase chain reaction (PCR) is based on simultaneous
performance of rapid amplification and quantitative detection. It is fast and
convenient, and allows quantification of mRNA changes in a large number of
paradigms.
Whereas performance of microarray experiments is within reach for many
laboratories, analysis of microarray data has been one of the most challenging
aspects of this new technology. Although precise automation has significantly
reduced experimental variability, microarray data is still widely regarded as a
“screen,” requiring validation by established technologies. Rapid progress in
technology and data analysis is likely to challenge that view. To understand the
true limits of gene expression experiments, one must consider contributing
sources of error. The two main sources of variability in microarray experi-
28_Budhraja_409_424_F 414 8/29/05, 11:24 AM

ments can be classified as technological and biological. Technological vari-

ability refers to all experimental errors resulting from the combined processes
of the experiment: array quality, RNA quality and extraction, labeling, hybrid-
ization, scanning, processing, and normalization. Because some of these
sources of error are platform-specific, they are generally out of the hands of the
typical researcher. However, more precise protocols and even automation of
common laboratory techniques can decrease the error during the prehybridization
steps. The expected value for precision in signal intensity is estimated at ⱕ10%
within chip, and 10–30% between chips (7). Biological variability, which
refers to variability inherent to the experimental design, exceeds technologi-
cal variability and requires careful consideration of the design goals of the
particular experiment. Biological variability is associated with factors such as
species, genetic differences, individual environmental effects, and temporal
changes. Variability associated with a cell line is usually different from vari-
ability associated with primary tumors taken from different patients with a simi-
lar diagnosis. These issues are dealt by primarily via replicates. Replicates
should be designed a priori, with the number of replicates dictated by the type
of biological samples (more replicates in more complex or heterogeneous
samples), anticipating differences in gene expression and precision. When rep-
licates are considered, researchers typically attempt to balance the need for
reliable data against the time and cost of the experiment. Consideration should
include the type of desired replicates, and their ability to address the sources of
error (8). For example, technological variability can be assessed by dividing a
sample at any phase prior to hybridization, thereby assessing variability intro-
duced at all stages after the samples were divided. Generation of replicates at
different experimental steps can therefore isolate the contribution of an indi-
vidual step to total variability. Unlike technological replicates, biological rep-
licates are more subjective, and require careful evaluation of different
contributions. It is important to note that experimental precision can be esti-
mated only once both types of variability are evaluated.
Normalization refers to the process of removing systematic errors in array
processing or data acquisition (9). Because many factors that contribute to sys-
tematic error are platform specific, the process of normalization must rely on
internal controls (10). When normalization is globally applied to an array, it is
commonly referred to as scaling. Comparison of expression level for each gene
between samples and control can only be done after a uniform scale is ensured,
implying that the expression level measurement in the samples and control is
comparable. Scaling can be performed by multiplying expression values of all
probe sets by a central estimate of expression level. Such an estimate can be
derived from the expression level of housekeeping genes within the array, under
the assumption that the expression of these genes is unchanged across the para-
28_Budhraja_409_424_F 415 8/29/05, 11:24 AM

digms. The validity of this assumption, however, has been questioned (11). An
alternative approach is adjustment to total mRNA expression, under the assump-
tion that the total mRNA in the cell is constant. However, divergent expression
levels among transcripts may increase the chance that global normalization
will result in false estimates.
Although the procedure described in this chapter is based on a one-dye sys-
tem, the use of a two-dye system introduces an additional source of bias. It has
been observed that different dyes can display differential efficiencies in label-
ing and detection, and are influenced by signal intensity (12–14). Such differ-
ences can be normalized based on the assumption that overall gene expression
is the same between experimental and control samples. A preferred method of
normalization utilizes locally weighted regression (LOESS), where local esti-
mates of mean ratios are found over small intervals of total signal intensity.
This process is repeated along a sliding scale of intensities (15,16).
Researchers seek to answer diverse questions with microarray experiments.
These questions commonly center on finding genes whose expression changes
from control to experimental paradigms, or exhibit an expression profile that
fits a certain pattern (e.g., time course or disease progression). A common
approach to identification of expression changes between two experiments
depends on an arbitrary fold change as a threshold for expression difference.
Although simple, this approach has no associated level of statistical confidence,
and depends on the assumption of a constant coefficient of variance (17). When
this assumption is violated (as typically happens in microarray data), the fre-
quency of false positive results increases dramatically. This is particularly true
for transcripts that exhibit low expression intensities, where fold differences
can be amplified by a denominator’s low expression intensity. Moreover, these
signals tend to be near noise level, which may mask expression differences.
Newer approaches to decrease such bias adjust fold-change threshold to signal
intensity (17–21).
The use of replicate paradigms allows investigators to estimate experimen-
tal variance (22). Methods that are based on a t-test or analysis of variance
(ANOVA) can assign a confidence level to each change in gene expression
(23,24). These effective approaches have been further developed to adjust for
intensity-specific variance (17–21), and even for variance that is specific for
each probe-set (21). With the use of a large set of replicates, one can estimate
the expected variances from these factors, and use them as reference values
(“correction factors”) in future experiments, when samples are compared with-
out replicates (21).
Identification of expression profile patterns can be achieved using cluster-
ing programs, which are based on the premise that similar expression changes
may imply similar functions. For example, when compared with all other genes
28_Budhraja_409_424_F 416 8/29/05, 11:24 AM

in the experiment, all genes that exhibit particular expression values are clus-
tered. Algorithms such as k-means and self-organizing maps cluster transcripts
by expression values based on a pre-defined number of expression patterns.
Hierarchical clustering defines “gene-tree” based on relative expression values
such that genes with the most similar expression values are clustered further
down on the tree (16,25). Techniques are available with which to utilize infor-
mation from array analysis in characterization of a new sample or paradigm.
Diverse methods can define a set of predictor genes, selected based on their
defined and discriminatory behavior in a known type of sample. This set of
predictor (signature) genes can be used to define an unknown sample or bio-
logical response.
Procedures presented in this chapter are based on prefabricated high-density
oligonucleotide arrays (Affymetrix) using a single dye (info can be found at
www.affymetrix.com). We therefore focus on sample preparation for array
hybridization and provide tips for data analysis. The experimental principles
provided here are applicable to other array platforms. It should be noted that
rapid progress in microarray technology is likely to impact many of the proce-
dural and technical notes presented in this chapter. The reader is therefore
advised to obtain platform-specific updates prior to performance of the experi-
ments. Additional research based on placental microarray has been published
(17,21,26,27).
2. Materials
1. Cell culture. Standard growth medium: Earl’s Medium 199 (M199), Ham’s/
Waymouth (H/W; composed of equal volumes of Ham’s F12 medium and
Waymouth medium), or Dulbecco’s modified Eagle’s medium containing 10%
fetal bovine serum (Hyclone, Logan, UT), 20 mM HEPES (Sigma, St. Louis,
MO), pH 7.4, 0.5 mM L-glutamine, penicillin (10 U/mL), streptomycin (10 µg/
mL), and Fungizone® (0.25 µg/mL).
2. RNA preparation and cleaning. Tri Reagent® (Molecular Research Center, Inc.,
Cincinnati, OH), chloroform (without isoamyl-alcohol or other additives), iso-
propanol, diethylpyrocarbonate (DEPC)-treated dH2O. RNA is cleaned using
QIAgen’s RNeasy Mini Kit (Qiagen, Inc., Valencia, CA). It is noted that
β-mercaptoethanol must be added to Buffer RLT before use. β-Mercaptoethanol
is toxic, and should be dispensed in a fume hood with protective clothing. In
addition, Buffer RPE should be diluted in 4 vol of ethanol (96–100%).
3. cDNA synthesis and cleaning. SuperScript™ Choice System for cDNA Synthe-
sis from Invitrogen Life Technologies (Carlsbad, CA) and a T7T21 oligonucle-
otide primer (GenSet, La Jolla, CA). For clean up: Phase Lock Gel® Light
(1.5 mL Eppendorf/Brinkmann, Westbury, NY), Phenol:chloroform:isoamyl
alcohol (25:24:1) saturated with 10 mM Tris-HCL pH 8.0/1 mM ethylenediamine
tetraacetic acid (EDTA), NH4Ac, absolute ethanol and 80% ethanol.
28_Budhraja_409_424_F 417 8/29/05, 11:24 AM

4. Synthesis of biotin-labeled cRNA by IVT. We use Enzo BioArray™ HighYield™

RNA Transcript Labeling Kit 3 (Enzo Life Sciences, Farmingdale, NY).
5. cRNA fragmentation. Fragmentation buffer is available in the GeneChip® Sample
Cleanup Module from Affymetrix (Affymetrix, Santa Clara, CA) (see Note 1).
3. Methods
3.1. Cell Culture or Tissues
Microarray analysis can be performed on transcriptomes derived from any
tissue or cell type. Analysis of placental tissues can be typically performed
using whole placental tissue samples (biopsies), placental explants, plated pri-
mary placental trophoblasts or trophoblast cell lines (see Note 2).
1. To prepare primary human cytotrophoblasts for our experiments, we typically
digest normal term human placentas using the trypsin-DNAse-Dispase/Percoll
method as described by Kliman (28), with modifications (29,30).
2. Cultures are routinely plated at a density of 350,000 cells/cm2 and maintained in
Earl’s M199 with additives as noted under Subheading 2.
3. To modulate trophoblast differentiation, we also culture cells in H/W, supple-
mented as described previously for M199 (31,32).
4. All cultures are maintained at 37°C, in a 5% CO2 atmosphere. Medium is changed
every 24 h.
3.2. RNA Preparation and Cleaning (see Note 3)

1. For tissue, we place 50–100 mg of samples frozen in –80°C in 1 mL of Tri Reagent.
The sample volume should not exceed 10% of the Tri Reagent’s volume. It is
essential to place snap frozen tissue directly into Tri Reagent and homogenize
immediately, to minimize the risk of RNA degradation (see Note 4).
2. Plated cells are washed once in warm PBS, followed by direct lysis using 1 mL of
Tri Reagent per 10 cm2 of culture plate.
3. After 5 min in Tri Reagent, we scrape off the cells using the back side of blue
pipet tip, and homogenize the sample by repeat up-and-down pipetting.
4. Homogenates can be processed directly, or stored in –80°C.
5. After 5 min in room temperature (RT), 0.2 mL chloroform is added per 1 mL Tri
Reagent.
6. Samples are vortexed vigorously for 15 s, and incubated at RT for 10 min.
Samples are then centrifuged at 12,000g for 15 min at 4°C, and the aqueous (RNA
containing) phase transferred to a new tube.
7. The RNA is precipitated using isopropanol (0.5 mL isopropanol per 1 mL Tri
Reagent used), mixed by inverting the tubes several times and incubated at RT
for 10 min.
8. Samples are centrifuged at 12,000g for 10 min. The RNA forms a white precipi-
tate on the side and bottom of the tube.
9. The RNA is washed by removal of the supernatant, followed by addition of 1 mL
of 75% ethanol, vortexed, and subsequent centrifugation at 12,000g for 5 min.
28_Budhraja_409_424_F 418 8/29/05, 11:24 AM

10. The ethanol is removed, the RNA pellet air-dried, and then dissolved in 50–100 µL
of RNase-free (DEPC-treated) dH2O.
11. Finally, the RNA yield should be quantified. RNA yield and purity is determined
by spectrophotometry at 260 and 280 nm. A260/A280 ratios between 1.9 and 2.2
are acceptable. If lower, repeat cleanup may be needed (see Note 5). We also
recommend checking RNA quality by running 2 µL of each sample on 1% agar-
ose gel. There should be two distinct bands (28S and 18S), without smearing (see
Note 4).
12. It is essential to clean the RNA samples before cDNA synthesis (see Note 5). For
RNA cleanup, we use QIAgen’s RNeasy Mini Kit, following the manufacturer’s
protocol. Because the binding capacity of each RNeasy mini column is 100 µg of
RNA, we usually use 50–100 µg of RNA per column. At the end of the proce-
dure, we elute RNA with 30 µg RNase-free dH2O (see Note 6).
3.3. Double-Stranded cDNA Synthesis and Cleaning

1. We use SuperScript™ Choice system (Invitrogen Life Technologies) and a T7T21
oligonucleotide primer (GenSet). We follow the manufacturer’s and Affymetrix’
instructions for first strand and second strand cDNA synthesis. At the end of
this section, we proceed to cleanup procedure for cDNA or store at –20°C for
later use.
2. We clean up the double-stranded DNA using phenol/chloroform extraction and
ethanol precipitation (see Note 1).
a. We first pellet the Phase Lock Gel (1.5 mL microcentrifuge tube with PLG
light) at ⱖ12,000g for 30 s.
b. In a separate tube, we mix the DNA with an equal volume of buffer-saturated
phenol:chloroform:isoamyl-alcohol, vortex, and add to the Phase Lock Gel
tube.
c. The tube is centrifuged at ⱖ12,000g for 2 min, and the aqueous, upper phase
transferred to a new 1.5-mL tube.
d. To precipitate the DNA, we add 0.5 vol of 7.5 NH4Ac and 2.5 vol of absolute
cold ethanol to the sample and vortex, followed immediately by centrifuga-
tion at >12,000g at RT for 20 min.
e. The supernatant is removed, the pellet washed twice with 0.5 mL of cold 80%
ethanol, and the pellet air-dried and resuspended in 12 µL of RNase-free
dH2O.
3.4. Synthesis of Biotin-Labeled cRNA by In Vitro Transcription

1. We follow the instructions from the Enzo Bioarray RNA labeling kit and
Affymetrix for synthesis of biotin-labeled cRNA, designed to bind streptavidin-
phycoerythrin during array staining.
2. At the end of the procedure, we either store the labeled cRNA at –20°C (or –70°C)
or proceed to cRNA cleanup.
3. We clean the cRNA using the reagents and protocol as described under Sub-
heading 3.2. To increase RNA yield, we pass the sample twice over the column
28_Budhraja_409_424_F 419 8/29/05, 11:24 AM

Fig. 3. An agarose gel analysis of purified in vitro transcription (IVT) product and
fragmented product. Fragment size (kb) is indicated in the margin.
before the wash and elution steps, and wait 5 min after adding water to the col-
umn for RNA elution, prior to centrifugation.
4. Note that the protocols from Affymetrix provide a formula for calculating the
relative amount of labeled cRNA yield, needed prior to array hybridization. The
difference between unpurified and purified RNA can be assessed by electrophore-
sis using a 1% agarose gel.
3.5. RNA Fragmentation

1. To optimize hybridization signal it is highly recommended by Affymetrix to frag-
ment cRNA targets before hybridization onto the array. Fragmentation buffer
and protocol are supplied by Affymetrix.
2. After fragmentation it is recommended that ⱖ 1 µg of fragmented cRNA is ana-
lyzed using electrophoresis with a 1% agarose gel, and visualized with ethidium
bromide.
3. At this point, we routinely compare cleaned IVT product with fragmented prod-
uct (Fig. 3), to ensure adequate fragmentation.
4. Samples are now ready for gene-chip hybridization, or can be stored in –20°C.
4. Notes
1. A GeneChip Sample Cleanup Module is available from Affymetrix, and may be
suitable for sample cleaning as an alternative to the described procedure.
2. Because biological variability is directly related to sample homogeneity, results
from homogenous samples exhibit lower variability, whereas results from tissue
28_Budhraja_409_424_F 420 8/29/05, 11:24 AM

biopsies (e.g., whole placental samples) exhibit greater variability, and therefore
require more replicates.
3. Although mRNA can be extracted and used for subsequent processing and analy-
sis, this is not necessary. Furthermore, it is more difficult to isolate mRNA, as all
precipitates are practically invisible. If needed, mRNA can be isolated using kits
such as Qiagen’s Oligotex mRNA Midi Kit (Cat. No. 70042). If mRNA is used,
amounts should be adjusted according to protocols from Affymetrix or other chip
manufacturers.
4. The placenta is a rich source of RNAse, necessitating careful isolation and puri-
fication procedures in order to avoid RNA degradation. RNA degradation can be
determined using 28S/18S rRNA signal ratio. If needed, capillary electrophore-
sis allows quantification of additional degradation species of RNA (33). Mini-
mizing differences in RNA degradation among the samples is essential for
comparison of transcript expression. This issue is particularly relevant for stud-
ies of apoptosis, where 28S rRNA is cleaved more rapidly than 18S (34). Note
that alternative RNA preparation protocols can also be used for extraction of
placental RNA.
5. The amount of RNA is critical for adequate microarray experiments. Low
amounts of RNA my result in signals at the lower limit of fluorescence detection,
with a low signal to noise ratio. At least 10 µg of RNA are needed for fluorescent
signal detection without amplification. Typically, the initial amount of RNA
needed for an experiment is 50 µg of total RNA or 2 µg of poly(A) mRNA.
Therefore, in studies using the human placenta, this is unlikely to be a limiting
factor. We usually extract 30–50 µg of RNA from 20 × 106 human term tropho-
blast cells or from 50 mg human placental tissue. When the amount of RNA is
insufficient, amplification using T7 polymerase, which exhibits linear amplifica-
tion (thereby sustaining relative RNA amounts), can be used. RNA purity is also
critical for signal reproducibility. Contaminants such as lipids, proteins, or sug-
ars can affect target hybridization to the slide surface. Microarray experiments
are based on the assumption that the level of RNA in the sample closely repre-
sents its relative amount in the tissue or cells.
6. To obtain a higher RNA concentration, a second elution step using the first eluate
can be performed. Although only a small amount of RNA will be used, it is diffi-
cult to obtain a high enough concentration of RNA if initial RNA quantity is less
then 50 µg. If the final concentration of RNA is too low, RNA can be precipitated
and resuspended in a smaller volume, but up to 50% of the product may be lost
during the process. The remaining RNA can be stored and used in other essays.
Acknowledgments
This research was supported by National Institutes of Health (NIH) R01
ES11597-01 and the Siteman Cancer Center GeneChip Core Facility, Wash-
ington University School of Medicine, St. Louis, MO, USA. We thank Elena
Sadovsky and Lori Rideout for technical assistance.
28_Budhraja_409_424_F 421 8/29/05, 11:24 AM

References
1. Hegde, P., Qi, R., Abernathy, K., et al. (2000) A concise guide to cDNA
microarray analysis. Biotechniques 29, 548–562.
2. Moreau, Y., Aerts, S., De Moor, B., De Strooper, B., and Dabrowski, M. (2003)
Comparison and meta-analysis of microarray data: from the bench to the com-
puter desk. Trends Genet. 19, 570–577.
3. Cui, X. and Churchill, G. A. (2003) Statistical tests for differential expression in
cDNA microarray experiments. Genome Biol. 4, 210.
4. Leung, Y. F. and Cavalieri, D. (2003) Fundamentals of cDNA microarray data
analysis. Trends Genet. 19, 649–659.
5. Saeed, A. I., Sharov, V., White, J., et al. (2003) TM4: a free, open-source system
for microarray data management and analysis. Biotechniques 34, 374–378.
6. Heller, M. J. (2002) DNA microarray technology: devices, systems, and applica-
tions. Annu. Rev. Biomed. Eng. 4, 129–153.
7. Yue, H., Eastman, P. S., Wang, B. B., et al. (2001) An evaluation of the perfor-
mance of cDNA microarrays for detecting changes in global mRNA expression.
Nucleic Acids Res. 29, E41.
8. Kerr, M. K. (2003) Design considerations for efficient and effective microarray
studies. Biometrics 59, 822–828.
9. Bolstad, B. M., Irizarry, R. A., Astrand, M., and Speed, T. P. (2003) A compari-
son of normalization methods for high density oligonucleotide array data based
on variance and bias. Bioinformatics 19, 185–193.
10. Quackenbush, J. (2002) Microarray data normalization and transformation. Nat.
Genet. 32(Suppl), 496–501.
11. Lee, P. D., Sladek, R., Greenwood, C. M., and Hudson, T. J. (2002) Control genes
and variability: absence of ubiquitous reference transcripts in diverse mammalian
expression studies. Genome Res. 12, 292–297.
12. Yang, Y. H. and Speed, T. (2002) Design issues for cDNA microarray experi-
ments. Nat. Rev. Genet. 3, 579–588.
13. Chen, Y., Dougherty, E. R., and Bittner, M. L. (1997) Ratio-based decisions
and the quantitative analysis of cDNAmicroarray images. J. Biomed. Optics 2,
364–374.
14. Dombkowski, A. A., Thibodeau, B. J., Starcevic, S. L., and Novak, R. F.
(2004) Gene-specific dye bias in microarray reference designs. FEBS Lett.
560, 120–124.
15. Cleveland, W. and Devlin, S. (1988) Locally-weighted regression: an approach to
regression analysis by local fitting. J. Am. Statistical Assoc. 83, 596–610.
16. Tamayo, P., Slonim, D., Mesirov, J., et al. (1999) Interpreting patterns of gene
expression with self-organizing maps: methods and application to hematopoietic
differentiation. Proc. Natl. Acad. Sci. USA 96, 2907–2912.
17. Mariani, T. J., Budhraja, V., Mecham, B. H., Gu, C. C., Watson, M. A., and
Sadovsky, Y. (2002) A variable fold change threshold determines significance for
expression microarrays. FASEB J. Express 17, e10.1096. Summary in FASEB J.
2003;1321–1323.
28_Budhraja_409_424_F 422 8/29/05, 11:24 AM

18. Tsien, C. L., Libermann, T. A., Gu, X., and Kohane, I. S. (2001) On reporting fold
differences. Pac. Symp. Biocomput. 496–507.
19. Baldi, P. and Long, A. D. (2001) A Bayesian framework for the analysis of
microarray expression data: regularized t-test and statistical inferences of gene
changes. Bioinformatics 17, 509–519.
20. Tusher, V. G., Tibshirani, R., and Chu, G. (2001) Significance analysis of
microarrays applied to the ionizing radiation response. Proc. Natl. Acad. Sci. USA
98, 5116–5121.
21. Budhraja, V., Spitznagel, E., Schaiff, W. T., and Sadovsky, Y. (2003) Incorpora-
tion of gene-specific variability improves expression analysis using high-density
DNA microarrays. BMC Biol. 1, 1.
22. Durbin, B. P., Hardin, J. S., Hawkins, D. M., and Rocke, D. M. (2002) A variance-
stabilizing transformation for gene-expression microarray data. Bioinformatics
18(Suppl 1), S105–S110.
23. Kerr, M. K., Martin, M., and Churchill, G. A. (2000) Analysis of variance for
gene expression microarray data. J. Comput. Biol. 7, 819–837.
24. Reiner, A., Yekutieli, D., and Benjamini, Y. (2003) Identifying differentially ex-
pressed genes using false discovery rate controlling procedures. Bioinformatics
19, 368–375.
25. Sherlock, G. (2000) Analysis of large-scale gene expression data. Curr. Opin.
Immunol. 12, 201–205.
26. Aronow, B. J., Richardson, B. D., and Handwerger, S. (2001) Microarray analysis
of trophoblast differentiation: gene expression reprogramming in key gene func-
tion categories. Physiol. Genomics 6, 105–116.
27. Kato, H. D., Terao, Y., Ogawa, M., et al. (2002) Growth-associated gene expres-
sion profiles by microarray analysis of trophoblast of molar pregnancies and nor-
mal villi. Int. J. Gynecol. Pathol. 21, 255–260.
28. Kliman, H. J., Nestler, J. E., Sermasi, E., Sanger, J. M., and Strauss, J. M. (1986)
Purification, characterization and in vitro differentiation of cytotrophoblasts from
human term placentae. Endocrinology 118, 1567–1582.
29. Nelson, D. M., Johnson, R. D., Smith, S. D., Anteby, E. Y., and Sadovsky, Y.
(1999) Hypoxia limits differentiation and up-regulates expression and activity of
prostaglandin H synthase 2 in cultured trophoblast from term human placenta.
Am. J. Obstet. Gynecol. 180, 896–902.
30. Schaiff, W. T., Carlson, M. G., Smith, S. D., Levy, R., Nelson, D. M., and
Sadovsky, Y. (2000) Peroxisome proliferator-activated receptor-γ modulates dif-
ferentiation of human trophoblast in a ligand-specific manner. J. Clin. Endocrinol.
Metab. 85, 3874–3881.
31. Douglas, G. C. and King, B. F. (1989) Isolation of pure villous cytotrophoblast
from term human placenta using immunomagnetic microspheres. J. Immunol.
Methods 119, 259–268.
32. Douglas, G. C. and King, B. F. (1990) Differentiation of human trophoblast cells
in vitro as revealed by immunocytochemical staining of desmoplakin and nuclei.
J. Cell Sci. 96, 131–141.
28_Budhraja_409_424_F 423 8/29/05, 11:24 AM

33. Auer, H., Lyianarachchi, S., Newsom, D., et al. (2003) Chipping away at the chip
bias: RNA degradation in microarray analysis. Nat. Genet. 35, 292–293.
34. Nadano, D. and Sato, T. A. (2000) Caspase-3-dependent and -independent degra-
dation of 28 S ribosomal RNA may be involved in the inhibition of protein syn-
thesis during apoptosis initiated by death receptor engagement. J. Biol. Chem.
275, 13,967–13,973.
28_Budhraja_409_424_F 424 8/29/05, 11:24 AM

Gene Expression Microarray Data Analysis 425
29
Gene Expression Microarray Data Analysis of Decidual
and Placental Cell Differentiation
Sue Kong, Bruce J. Aronow, and Stuart Handwerger
Summary
Gene expression analysis using DNA microarray approaches have provided new insights
into the physiology and pathophysiology of many biological processes. These include identifi-
cation of genetic programs and pathways that underlie cell and tissue differentiation and gene
expression programs responsive to genetic perturbations, drugs, toxins, and infectious agents.
In this chapter, we present methods for the analysis of microarray data using earlier investiga-
tions from our laboratory as examples of how gene expression patterns for cellular differentia-
tion may be detected and analyzed for biological significance and how regulated genes may be
classified into functional categories and pathways.
Key Words: Gene expression; clustering; microarray; decidualization; trophoblast; pla-
centa; differentiation.
1. Introduction
DNA microarray technology permits qualitative analysis of mRNA expres-
sion of multiple genes in a single specimen. Because large numbers of genes
can be assessed, microarray studies have provided considerable insights into
physiological processes such as cell proliferation, differentiation, apoptosis,
and malignant transformation. In addition, microarrays have provided insight
into the cellular responses to drug treatment, environmental toxins, and infec-
tious agents.
There are two main types of microarray technologies: the single-channel
array and the two-channel array. Two-channel arrays employ a reference RNA
in which the relative signal of a given gene is detected as the ratio of the signal
for the reference vs that of the sample. In contrast, single-channel arrays mea-
sure relative intensities of each gene per array and per RNA sample. Thus,
single-channel arrays require that each RNA (e.g., experimental or control
425
29_Kong_425_438_F 425 8/29/05, 11:24 AM

426 Kong, Aronow and Handwerger
sample) be measured on their own separate microarray (for a detailed discus-

sion, see ref. 1). Codelink (Amersham) microarrays and Affymetrix GeneChips
are examples of single-channel technologies. Affymetrix GeneChips uses mul-
tiple spots that contain independent oligonucleotide probes for each gene, divided
into perfect matches and mismatches. Each perfect match is a short oligonucle-
otide probe corresponding to a specific gene transcript, and the mismatches
contain a single point mutation at the midpoint of each sequence. Codelink
mechanically spots a single 30-mer oligonucleotide using highly reproducible
methodologies. NimbleGen, another novel single channel technology, uses
either 24-mer or 60-mer oligonucleotides that are synthesized photolitho-
graphically onto a 390,000-feature array. Most other microarrays, such as
Stanford-type spotted cDNA arrays (2), use two-channel labeling approaches
in which cRNA hybridized with Cy3 dye is used for one channel and cRNA
hybridized with Cy5 is used for the other channel.
During the past 3 yr, our laboratory has used gene expression microarrays to
identify genes that are regulated during decidualization and trophoblast differ-
entiation of cells purified from human decidua and placenta (3–6). In studies of
decidualization, human decidual fibroblast cells were differentiated in vitro by
treatment with progesterone, estradiol, and dibutyryl cAMP. Gene expression
microarray analyses of decidualization were performed from RNAs purified
after 0, 2, 4, 6, 9, and 12 d of differentiation. Trophoblast differentiation, which
occurs somewhat more rapidly in vitro, was analyzed using RNAs isolated
from an enriched fraction of human cytotrophoblast cells undergoing sponta-
neous differentiation to syncytiotrophoblast cells over 6 d of cell culture. The
expression patterns of the induced and repressed decidual and placental genes
were determined, and the biological functions of the regulated genes were cat-
egorized into major groups using Gene Ontology and Medline data sources.
In this chapter, we will focus on the procedure of DNA microarray data
analysis, emphasizing methods for the identification of differentially expressed
genes, clustering, and functional classification.
2. Materials
1. Microarray Suite/GeneChip Operating System (GCOS) software (Affymetrix,
Inc., Santa Clara, CA).
2. RMAExpress (written by Ben Bolstad, University of California, Berkeley) http:/
/stat-www.berkeley.edu/~bolstad/RMAExpress/RMAExpress.html
3. GeneSpring software (Silicon Genetics, Redwood City, CA) (see Note 1).
3. Methods
The general scheme for DNA microarray analysis is illustrated in Fig. 1.
Each of the individual steps of the analysis is described in greater detail below.
29_Kong_425_438_F 426 8/29/05, 11:24 AM

Fig. 1. General scheme for microarray data analysis.
3.1. Probe Level Data Summarization (One-Channel Arrays)

Because Affymetrix relies on the hybridization signal strength from a series
of oligonucleotides, an estimate of the relative signal strength for each gene
requires a summarization of the probe-level intensities across the oligonucle-
otide set. Affymetrix Microarray Suite software version 5.0 (MAS5), now
GCOS is used to scan and quantify GeneChips to produce “.cel” files which
contain the individual probe level signals (see Note 2). Relative gene expres-
29_Kong_425_438_F 427 8/29/05, 11:24 AM

sion is then determined from the Affymetrix .cel files by either the Affymetrix
algorithm contained in MAS5 (which, among other things, subtracts the mis-
match signals) or the Robust Multi-array Analysis protocol (7) using the
RMAExpress program with default settings. The RMA program uses the per-
fect match data and not the mismatch data, but provides background adjust-
ment, quantile normalization, and then a summarization of gene-specific
relative signal strength. The probe set signals are then saved into a text file that
can be imported into GeneSpring. Using a series of dilution and calibration
experiments (8), we earlier observed that RMA reduces spurious signals and
eliminates false positive estimations of induced and repressed genes as com-
pared to signal strength estimated through MAS5.0.
3.2. Loading Data into GeneSpring
1. Importing data. GeneSpring recognizes the format of data files obtained from
most expression analysis programs. Although RMAExpress summary data is not
directly recognized by GeneSpring, a column editor permits a custom format to
specify gene identifier and signal columns in a tab-delimited text file.
2. Create experiment. Upon importing the data files, a new data set for each
microarray (referred to as new sample in GeneSpring) is created and saved. If
necessary, the new data set can be easily combined with another newly created
data set or an older data set using the GeneSpring sample manager that is part of
the software package.
3.3. Data Preprocessing

1. Define parameters: The first step in the preprocessing of a data set (sample) is to
define the experimental parameters or variables that describe the data. A param-
eter value is then assigned for each defined parameter. For example, “Cell Type”
can be a parameter with “trophoblast” and “decidual” as two parameter values. A
“Time” parameter could contain “2,” “4,” “6,” “9,” “12,” and “15” as parameter
values for different days.
2. Normalization: Once the parameters have been defined and assigned values, the
data is normalized in order to remove unwanted technical variation introduced in
the measurement process. Normalizing also scales data so that relative gene
expression levels can be obtained from different chips. The particular nor-
malization and baseline reference methods depend on the design and goals of
each experiment (9).
a. One-color experiment: Many normalization options are available in
GeneSpring. The first consideration is the relative signal strength overall of
each chip within the experimental series using a so-called “per-chip normal-
ization.” Such variations in overall labeling intensity from chip to chip may
be due to inconsistent washing, inconsistent sample preparation, or microarray
production or microfluidics imperfections. Because the quantile normaliza-
tion step in the RMA summarizing algorithm addresses the issue of chip-wide
variation (7), no additional per-chip normalization is necessary. After retrieval
29_Kong_425_438_F 428 8/29/05, 11:24 AM

into GeneSpring, RMA signal strength measures are first transformed from
log base 2 to linear values and then normalized to the median or mean of all
measurements for each gene across all samples or sometimes just to the con-
trol samples as a function of the experiment design.
b. Two-color experiment: In experiments that use two-channel microarray data,
a control is performed to account for dye-based gene labeling differences.
This is done by reversing the signal channel and control channel measure-
ments for selected samples and calculating a correction factor that shifts the
relative gene expression ratio as a function of which dye was used for which
sample. Then a combined per-Spot and per-Chip normalization is performed
with intensity-dependent (locally weighted regression [LOESS]) normaliza-
tion. To counter intensity-dependent labeling and hybridization differences, a
LOESS fit at each point is calculated and a LOESS curve is fit to the log-
intensity vs log-ratio plot to adjust the control value for each measurement.
3. Set up an experimental interpretation: Further data analysis is dependent on the
specific questions being addressed. For example, the numerical display mode
permits data within an experiment to be represented as ratio, log of ratio, or fold-
change. In log of ratio mode, normalized intensity values are plotted against a log
scale so that underexpressed and overexpressed genes are considered equally sig-
nificant. The display setting may also be customized by defining a parameter to
be continuous, noncontinuous, or not displayed. When a parameter is not dis-
played, samples with the same parameter values will be grouped together, which
can be important in the following statistical analysis techniques. Another option
in the interpretation setup is whether to turn on the cross-gene error model (see
Note 3).
3.4. Identification of Differentially Expressed Genes

The goal of gene filtering for differential expression is to identify those genes
that are differentially expressed between two conditions. This can be done
using a variety of approaches that can provide additional confidence that there
is differential gene expression. Although it is generally agreed that microarray
data should be independently corroborated by the use of polymerase chain
reaction (PCR) analysis of selected mRNAs used in the microarray experi-
ments, we have also seen cases where multiple microarray platforms consis-
tently call gene differences among genes that fail to be measured accurately in
reverse-transcription (RT)-PCR analyses. Thus, there is no such thing as a guar-
anteed measurement method.
1. Filter on fold change. This method is used to identify genes with different expres-
sion profiles based on a comparison of two samples or conditions. Two conditions
(groups of samples) under an experiment interpretation are selected, and mean
expression levels are used in comparing the two groups. The program will then
identify the genes whose normalized intensity in the first condition is greater
than or less than that in the second condition by a specified fold factor. Because
29_Kong_425_438_F 429 8/29/05, 11:25 AM

the RMA algorithm compresses the expression range, a fold factor of 1.2, in our
experience, can indicate a significant change in expression level.
2. Statistical analysis. Statistical analysis (analysis of variation [ANOVA]) is a fil-
tering tool that can be applied to a gene list obtained from other filters such as
fold change. This filter compares mean expression levels between two or more
groups of samples (conditions) to detect subsets of genes that show statistically
significant differences in mean normalized expression levels. The interpretation
is important as it defines the data mode and the grouping of the samples. Both
one-way ANOVA and two-way ANOVA are permitted in GeneSpring. Compari-
sons can be performed with parametric or non-parametric methods at a specified
p-value cutoff, with or without multiple testing corrections.
Student’s t-test/ANOVA assumes variances to be equal while Welch t-test/
ANOVA assumes variances not equal across groups. The specific test to choose
depends on the variance across the data and practically the number of replicates
in the experiment. In our experience, Student’s t-test gives better results when
there are only a few replicates with the multiple testing correction option to con-
trol the false positive rate.
When testing the statistical significance of group comparisons for many genes,
a certain number of genes will pass the filter by chance alone and be considered
statistically significant. Multiple testing corrections can adjust the individual
p-value to account for this effect. The Benjamini and Hochberg test controls
the false discovery rate, defined as the proportion of genes expected to be identi-
fied by chance relative to the total number of genes called significant. However,
with too few replicates, the test itself may not have enough power to differentiate
false-positives or -negatives. That is, by applying multiple testing correction,
some potentially interesting genes could be incorrectly labeled false-positives
and removed because of a lack of statistical power. Figure 2A shows the differ-
ence in gene lists identified with different comparison options. Some clusters
identified statistically significant by Student’s t-test did not pass the Welch t-test,
although different expresison profiles are detected vissually. Figure 2B shows
the genes that passed the Student’s t-test but were not identified when using mul-
tiple testing correction. No obvious difference can be seen between the decidua
and trophoblast groups for these genes, many of which are false positives that
made the list by chance alone.
3. Combining filtering and statistical analysis. In earlier versions of GeneSpring, a
sequential maneuver was used to identify significance and magnitude of change
in expression of a set of genes between two conditions. With Bioscripts imple-
mented in the current release of GeneSpring V6.2, filtering and statistical analy-
sis can be combined and genes identified with one script by generating a volcano
plot (see Note 4). A volcano plot displays the negative log of p-values from a t-test
on one axis and the log2 of fold change between two conditions on the other axis
on the Scatter Plot view. Figure 3 shows a volcano plot that presents upregulated
(yellow) and downregulated (blue) genes that are also statistically significant in
one plot
29_Kong_425_438_F 430 8/29/05, 11:25 AM

Figure 2
Figure 3
29_Kong_425_438_F 431 8/29/05, 11:25 AM

Figure 4
4. Other analyses. GeneSpring is a rich source of analysis tools to assist in the iden-
tification of biologically meaningful expression data (see Note 5). One other op-
tion is to find genes with similar expression patterns to a selected gene generated
from the average of a group of genes. If a gene or group of genes is selected based
on expression level filtering and/or statistical tests, other genes whose expression
profiles are similar but did not pass the filters may be identified. Figure 4 shows
an example of genes selected because of similarities in expression pattern to a
specific gene.
3.5. Clustering
GeneSpring’s clustering algorithms are designed to form groups of genes or
conditions with similar expression patterns. GeneSpring supports a variety of
clustering methods—K-means, Gene Tree, Condition Tree, Self-Organizing
Map, and QT Clustering. Each method uses a series of different distance
metrics to define relative similarity, such as Pearson Correlation, Standard
Correlation, Distance, and others. These are useful tools to identify genes that
are potentially co-regulated as well as to reveal coordinated responses shared by
sets of samples to various experimental treatments. Figure 5A is an example of
29_Kong_425_438_F 432 8/29/05, 11:25 AM

Figure 5
gene tree clustering, where the closeness of the branches in the trees is a mea-
sure of the correlation of the genes’ expression. Clusters of genes (Fig. 5B–E)
that are co-regulated can be identified from the tree and further analyzed.
K-means clustering divides genes into groups with a high degree of similar-
ity based on their expression levels. In the time series experiment of
decidualization shown in Fig. 6, K-means clustering was used to identify 9
unique classes of genes that are upregulated or downregulated in a time-depen-
dent manner. Using a lower number of clusters resulted in groups with less
consistent patterns of expression, whereas using a higher number of clusters
resulted in groups that appeared to overlap with patterns of expression ob-
served in other groups.
3.6. Functional Classification
Genes can be categorized using shared attributes in the description of their
function or structure. This allows for genes to be grouped within common cat-
egories that can then be combined or contrasted to other categories. These cat-
egories include biological process, cellular component, and molecular function.
A useful approach to gene categorization is provided by the Simplified Gene
Ontology tool in GeneSpring using information stored in the annotations fields
of the genome features file. Combined with annotations retrieved by the
GeneSpider tool within GeneSpring, which updates gene annotations from
Unigene, LocusLink and Genbank based on Genbank accession numbers, we
29_Kong_425_438_F 433 8/29/05, 11:25 AM

Figure 6
have added annotations from Affymetrix annotation releases and Stanford pub-
lic database SOURCE (http://genome-www5.stanford.edu/) (see Note 6). The
Build Simplified Ontology tool groups genes hierarchically into biological cat-
egories (gene lists) based on the Gene Ontology Consortium Classifications
(http://geneontology.org/). GeneSpring’s ontology tool parses all of the anno-
tations in the genome and then assigns each gene to one or more ontology
groups. Additional gene categories can be constructed using selected annota-
tions in the program, such as chromosome location and pathways (see Sub-
heading 3.7.). The scripting environment of GeneSpring allows for automation
of the process of comparing a list of regulated genes in an experiment to each
29_Kong_425_438_F 434 8/29/05, 11:25 AM

of the gene categories. Many scripts are included in the BioScripts library
(BioScript Library 2.0\Biological Queries\Gene Ontology (GO) analysis)
released by SiliconGenetics.
3.7. Pathway Analysis
A pathway is a graphical representation of the interaction between gene
products in a biological system. User-drawn pathways as well as publicly avail-
able pathways such as Kyoto Encyclopedia of Genes and Genomes (KEGG,
http://www.genome.ad.jp/kegg/) can be imported to GeneSpring (see Note 7).
The expression change of the genes participating in a pathway can be viewed
on the graphical representation. This analysis can be very useful if you are
trying to identify a class of genes that are associated with a particular step or
regulatory element within a pathway. Androgen and estrogen metabolism path-
way, which is important in decidualization, is illustrated in Fig. 7.
3.8. Publication of Microarray Data-Based Experiments
A consistent procedure for the description and publication of microarray
data-based experiments is critical. It is important that a sufficiently detailed
description of the experiment and its analysis accompanies a microarray-based
publication so as to allow corroboration and re-analysis. It is also extremely
useful that there is the web-accessible release of primary microarray data. The
advantage of microarray data release is that it can permit others to corroborate
the authors’ interpretations as well as to permit additional questions to be posed
of the data set by different investigators. In order to accomplish this, the
Microarray Gene Expression Data Society (MGED) has produced a general
guideline document to aid authors in the presentation of relevant details that
can allow another investigator to understand the experiment and how it was set
up and analyzed using microarray technology. The guideline is called the Mini-
mal Information About Microarray Experiment (MIAME) checklist, and it is
currently available at the MGED website at http://www.mged.org/ as the
MIAME 2.0 document.
4. Notes
1. Other software such as Spotfire (Spotfire, Inc., Somerville, MA) and Genetraffic
(Iobion Informatics LLC, La Jolla, CA) are alternatives for analyzing microarray
data.
2. Quality control may be applied before analyzing the microarray data by checking
significant parameters in the report file generated from Affymetrix chip analysis.
Within GeneSpring software, the “All Samples” interpretation and clustering by
condition tree can also be used to check the quality of the data obtained from
microarray experiment.
29_Kong_425_438_F 435 8/29/05, 11:25 AM

Figure 7
436
29_Kong_425_438_F 436 8/29/05, 11:25 AM

3. The GeneSpring error model can be used to estimate either measurement varia-
tion or sample-to-sample variation. The estimates of these two components of
variation are used to estimate standard errors and compare mean expression lev-
els between experimental conditions. In case there are no replicates for a condi-
tion, statistical analysis can still be performed with the GeneSpring error model
turned on. However, using sufficient biological replicates is recommended in
microarray studies to obtain the most statistical power.
4. If multiple testing correction is not applied, the order of filtering and statistical
analysis is not important. However, using previously filtered gene lists in t-test/
ANOVA with multiple testing correction can result in a larger gene list due to
less false positives and smaller variance.
5. We only chose to cover the most basic and important tools that are frequently
used in our laboratory in this chapter. Many other tools implemented in
GeneSpring can be very useful in the process of identifying of significant genes.
6. The annotations for the ontology tool are regularly updated by SiliconGenetics.
Combining annotations from different public data sources provides the most com-
plete ontology analysis.
7. Stand-alone applications for pathway analysis with more functionality than
GeneSpring are available. We have used Ingenuity Pathways Analysis (Ingenu-
ity, Mountain View, CA).
Acknowledgments
We thank Cherie Kessler, Anoop Brar, and You-Hong Cheng for their con-
tributions to the DNA microarray studies cited in this chapter. Supported by
National Institutes of Health (NIH) grant HD-15201.
References
1. Yang, Y. H. and Speed, T. (2002) Design issues for cDNA microarray experi-
ments. Nat. Rev. Genet. 3, 579–588.
2. Brown, P. O. and Botstein, D. (1999) Exploring the new world of the genome
with DNA microarrays. Nat. Genet. 21, 33–37.
3. Cheng, Y. H., Aronow, B. J., Hossain, S., Trapnell, B., Kong, S., and
Handwerger, S. (2004) Critical role for transcription factor AP-2 in human tro-
phoblast differentiation. Physiol. Genomics 18, 99–107.
4. Brar, A. K., Handwerger, S., Kessler, C. A., and Aronow, B. J. (2001) Gene
induction and categorical reprogramming during in vitro human endometrial fi-
broblast decidualization. Physiol. Genomics 7, 135–148.
5. Aronow, B. J., Richardson, B. D., and Handwerger, S. (2001) Microarray analy-
sis of trophoblast differentiation: gene expression reprogramming in key gene
function categories. Physiol. Genomics 6, 105–116.
6. Handwerger, S. and Aronow, B. (2003) Dynamic changes in gene expression
during human trophoblast differentiation. Recent Prog. Horm. Res. 58, 263–281.
29_Kong_425_438_F 437 8/29/05, 11:25 AM

7. Irizarry, R. A., Bolstad, B. M., Collin, F., Cope, L. M., Hobbs, B., and Speed, T.
P. (2003) Summaries of Affymetrix GeneChip probe level data. Nucleic Acids
Res. 31, e15.
8. Freudenberg, J., Kong, S., Jegga, A., et al. Experimental design, data analysis,
and quality evaluation approaches to maximize cross-platform and cross-proto-
col inter-comparability of gene expression microarray data, Manuscript in
preparation.
9. Sartor, M. A., Medvedovic, M., and Aronow, B. J. (2003) in A Beginner’s Guide
to Microarrays (Blalock, E. M., ed), Technical Books, San Diego: pp. 151–178.
29_Kong_425_438_F 438 8/29/05, 11:25 AM

Gene Expression Assays 439
30
Assays to Determine Allelic Usage of Gene Expression
in the Placenta
Paul B. Vrana
Summary
Mammalian placentas express a large number of so-called imprinted genes. Imprinting refers
to mono-allelic or biased expression based on which parent contributed the allele. Many of these
imprinted loci encode factors involved in growth and cell-cycle regulation, as well as maternal
behavior. In general, paternally expressed genes tend to enhance growth, whereas maternally
expressed genes inhibit growth. Methods are described for developing assays to test the allelic
usage of a gene. The approaches described are best utilized within a system where multiple
strains are available, and it is possible to perform reciprocal crosses. Only polymerase chain
reaction-based methods are examined in any detail.
Key Words: Imprinting; mono-allelic gene expression; Peromyscus; placenta; polymerase
chain reaction.
1. Introduction
The last 20 yr have revealed that a number of autosomal mammalian genes
are primarily expressed from one allele. Typically, this mono-allelic expres-
sion is dependent on which parent contributed the allele. This phenomenon is
termed genomic imprinting (1). The mammalian placenta and extraembryonic
tissues are particularly rich in the expression of these imprinted genes, many of
which regulate growth (2,3). Also, much of the X chromosome is expressed
from the maternal allele exclusively in extraembryonic tissues in various mam-
malian species including rodents and cows. Whether or not this placental X
imprinting occurs in any human trophoblast cells is controversial, but it is clear
that the entire human placenta is not subject to this phenomenon (4). However,
skewing of X-inactivation, such that one allele is preferentially expressed occurs
regularly. Such skewing is often associated with spontaneous abortions (5).
439
30_Vrana_439_450_F 439 8/29/05, 11:25 AM

440 Vrana
Similar to the X chromosome, autosomal imprinted loci are also found in

clusters. The number of clusters found has been growing such that greater than
70 imprinted transcripts have been identified located in at least 15 domains. The
latter number is debatable because there are species differences and the bound-
aries of these domains are often unclear.
The importance of the imprinting process(es) in the placenta is underscored
by the dramatic effects that perturbations have on its growth and morphology.
For example, the first imprinted gene discovered, insulin-like growth factor 2
(Igf2) is expressed at high levels both in embryonic and extraembryonic tis-
sues. However, an Igf2 targeted mutation in mice (“knockout”) that was restricted
to the placenta (i.e., only placental Igf2 was lacking) had a growth reduction
nearly equal to that of complete zygotic lack of the gene product (6).
Another example is that of hydatidiform moles (HMs), which are over-
growths of trophoblast-like tissue in humans. HMs may be caused by dispermy
(an overabundance of paternally expressed genes) and/or lack of maternal
expressed genes (7). While this evidence indirectly linked HM trophoblast
overgrowth to imprinting, it has recently been shown that disruptions of the
imprinting process lead to very similar phenotypes (8).
Finally, I work on a rodent system of the genus Peromyscus (native Ameri-
can deer mice) in which reciprocal hybrids of two closely related species yield
opposite phenotypes on growth. That is, female A × male B yields overgrowth
while female B × male A results in undergrowth. The placenta appears to be
the primary target of these phenotypes: the undergrown hybrids appear to be
lacking most of the spongiotrophoblast layer, while the overgrown hybrids
show an expansion of this layer, but also general placental disorganization
(9,10).
The overgrown hybrids frequently lack an embryo proper, consisting entirely
of extra-embryonic tissue (as in HMs). The Peromyscus placental phenotypes
are also associated with perturbations of imprinting (11). Two loci involved in
the placental phenotypes have been mapped: one to an imprinted domain which
also harbors a susceptibility locus for HM (12), and another which maps to the
X chromosome (13). A very interesting locus, Esx1, is located on this portion of
the X. Esx1 encodes a homeodomain protein involved in placental patterning
and growth, yet is rapidly evolving (14,15).
The most compelling explanation to date for the existence of imprinting is
the Parental Competition hypothesis (2,16), which proposes that in any
nonmonogamous population, parental interests differ: males “want” their off-
spring to extract as many maternal resources as possible during gestation and
postnatal care, particularly if other offspring from this female are not likely to
be their own. This antagonism is likely to be realized in behavior and growth.
30_Vrana_439_450_F 440 8/29/05, 11:25 AM

Enhanced prenatal/neonatal growth (and hence survival) at the cost of others

offspring is desirable.
Conversely, females should value fitness of all their offspring equally, regard-
less of paternity. Consequently, maternal resources should usually be allocated
equally among all offspring. According to this theory, males use imprinting to
repress alleles of growth-inhibiting genes. In response, females repress their
own alleles of growth promoting genes. This conflict is predicted to have the
placenta as a primary battleground due to its function as both a major endo-
crine organ and the source of maternal–fetal nutrient transfer.
Here I will outline what I feel is the easiest method for developing an allele
specific gene expression assay. Whether this approach proves the easiest for
the reader will depend on a number of variables. Probably the most important
of these variables is the system the investigator works on. The optimal system
here would be one in which: (1) one has multiple relatively homogenous strains
of the organism in question; (2) polymorphisms between strains are abundant;
(3) it is feasible to breed these strains, and to do both reciprocal crosses.
2. Materials
1. Acrylamide and agarose electrophoresis apparatuses and reagents.
2. Dissection equipment and tissue homogenizer.
3. DNA sequencing reagents and apparatus or access to sequencing service.
4. Oligonucleotide primer synthesis and design programs.
5. Polymerase chain reaction (PCR) thermocycler.
6. Gel documentation system with quantification software and/or phosphorimager.
7. PCR reagents.
8. Reagents for RNA isolation, e.g., lithium chloride/urea or kit.
9. Reverse-transcription (RT) and PCR and associated reagents and/or kit.
10. Restriction enzymes and associated reagents.
11. Phosphorimager and/or gel documentation system with quantification software.
3. Methods
The methods described outline various strategies and considerations in devel-
oping allelic expression assays using placental tissues. Standard molecular biol-
ogy procedures, such as RNA purification and cDNA synthesis, are not covered
in any detail.
3.1. General Considerations
Because imprinting may be tissue-specific, if the gene effect is thought to be
associated with the placenta, one must examine placental RNA for allelic expres-
sion status (see Note 1). A parental polymorphism is generally necessary to
determine parent-of-origin allelic expression, although fluorescence in situ
hybridization (FISH) techniques may also be used to determine mono-alleic
30_Vrana_439_450_F 441 8/29/05, 11:25 AM

442 Vrana
expression. However, FISH assays are generally prohibitively labor-intensive

and impractical when dealing with many samples. Moreover, these assays will
generally not reveal the parent-of-origin of the expressed allele. To determine
which parent contributed the active allele, a polymorphism is required (see
Note 2). Sequencing the individual to ascertain heterozygosity, or sequencing
the potential animals to be bred is the method of choice for finding such a
polymorphism. With the costs of DNA sequencing having dropped dramati-
cally, and commercial services readily available, this should not be a hardship.
3.2. Developing a PCR-Based Assay
Perhaps the most straightforward method of developing an allele-specific
assay is to find a polymorphism between two relatively inbred strains of ani-
mal, then develop an RT- PCR assay which exploits that polymorphism (see
Note 3). Indeed, this will be the only method described in detail in this chapter.
Length differences in the amplicon provide the easiest assays to develop (in
that one must simply subject the RT-PCR products to electrophoresis to ascer-
tain allelic usage) but may be difficult to find. The 3' untranslated region is
usually fertile ground for such polymorphisms.
Restriction fragment length polymorphisms (RFLPs) are generally the next
most desirable method of choice (Fig. 1). Optimally, the restriction enzyme in
question will cut both alleles (albeit differentially), such that a control for the
digestion is inherent in every experiment. Otherwise, one should always
include a pure sample of the allele that cuts as a control for the restriction
enzyme digest.
Sequencing the individual alleles (when possible) is the most sure-fire way
to assess potential RFLP assays. However, if the RT-PCR product is larger
than several hundred base pairs, and there are a number of restriction enzymes
on hand, one may elect just to cut the potential alleles with selected enzymes.
In general, endonucleases with either four base-pair recognition sequences or
ambiguities in the recognition sequence are more likely to uncover allelic varia-
tion. We have found the New England Biolabs (Beverly, MA) catalog an in-
valuable guide to restriction digests.
Restriction digests of these alleles are then subjected to gel electrophoresis
under conditions appropriate to the size of the resulting fragments. Generally,
we find that small (7.5 cm × 7.5 cm) approx 7.5% acrylamide gels yield better
resolution and more sensitivity than even high-resolution agarose. We have
routinely developed assays in which fragments differed by 20 bp or less.
Developing an assay that works identically on both genomic and cDNA has
both positive and negative aspects. The positive aspects being that one can
work out the kinks on the genomic DNA without wasting potentially hard-to-
obtain cDNA. Typically, the assays behave identically on cDNA (i.e., assum-
ing no introns are spanned). In testing for allelic bias, showing the same assay
30_Vrana_439_450_F 442 8/29/05, 11:25 AM

Fig. 1. Illustration of reverse-transcription polymerase chain reaction-based allelic

usage assay that exploits an restriction fragment length polymorphism between two
strains of rodent. A paternally expressed gene (Igf2) is shown. Strains A and B have
been reciprocally crossed. Female is shown first in each cross. Bars overlay lanes with
samples from one cross direction (i.e., A × B or B × A). Arrows indicate the allele-
specific bands after restriction digestion. A+B indicates lane where pure A and B RNA
samples have been mixed to show lack of amplification bias.
performed on genomic side-by-side with the cDNA assay makes an excellent

visual.
The negative aspect of developing identical genomic/cDNA assays is the
ever-present danger of genomic DNA contaminating the RNA/cDNA samples.
Typically, one must treat the RNA with DNAse prior to the reverse transcrip-
tion step. This also necessitates an additional purification step to remove the
DNAse enzyme, through precipitation (and subsequent centrifugation) or
through column purification. These additional steps can result in sample loss
as well as additional chances for human error.
Perhaps the optimal situation is when the primers span a small intron such
that genomic and cDNA may be easily distinguished (and the cDNA preferen-
tially amplified due to the smaller amplicon). In this situation, one can use the
same primer pair/polymorphism for genotyping and allelic expression assays.
There are many computer software programs available for picking PCR
primers. One should consider designing the primers so that the RFLP will be
easy to distinguish, and will only amplify the gene of interest. The latter can
usually be ascertained by sequence comparisons using the Basic Local Align-
ment Search Tool (BLAST) at the National Center for Biotechnology Informa-
30_Vrana_439_450_F 443 8/29/05, 11:25 AM

444 Vrana
tion (NCBI, Bethesda, MD) database (http://www.ncbi.nlm.nih.gov/BLAST/)

(a sort of “electronic PCR”).
3.3. RNA Preparation and Reverse Transcription
There are several RNA preparation protocols that are satisfactory for RT-PCR.
Our method of choice for samples of reasonable mass (1 mg or greater) is the
“old-school” lithium chloride/urea prep (17). Many people prefer the “Trizol”
reagent (Invitrogen, Carlsbad, CA) because its use typically requires fewer
steps. However, we have found that it is typically much dirtier RNA, and is
more likely to be refractory to the RT step, and to have much more DNA con-
tamination.
For smaller samples and for clean-up after DNAsing RNA samples, we uti-
lize the QIAgen “RNAeasy” kit (Valencia, CA). A number of companies now
make similar columns that selectively bind RNA. There are also many kits
available to selectively isolate polyA RNA, but this should be unnecessary
unless the mRNA to be assayed is exceedingly rare.
A number of RT enzymes and kits are also available on the market. Although
we routinely use a kit from Invitrogen, the specific enzyme is one parameter we
will change in the case of messy PCRs or low-copy messages. Yet another
variable is the primer used to make the cDNA. Our first course is generally to
use the poly-T primer, which of course binds to the poly-A tract found at the
end of mature messenger RNA. In other cases, it may be advisable to prime
with random sequence oligonucleotides. Random priming is specifically rec-
ommended in cases in which the region of the mRNA to be amplified is far
from the 3' end, or the RNA is somewhat degraded.
3.4. Assay Controls
Standard negative controls when performing RT-PCR include RNA
samples, which have had no RT enzyme added (these can be dropped once it
has been established that the amplicon spans a large intron), as well as controls
where template is not added to the PCR reactions. I recommend keeping a set
of pipets exclusively for RNA/PCR work, which are never brought near DNA
clones, or handled without gloves. A dedicated area of the lab is also preferable,
but not essential. Equipment may be decontaminated by treatment (10–20 min)
in a standard ultraviolet (UV) crosslinker, which damages contaminating DNA
sufficiently to generally prevent its amplification.
The bête noir of these assays is inherent bias in which an allele is amplified.
To test for this bias, careful mixing experiments must be performed. That is,
one should mix equal amounts of strain RNAs (i.e., the two alleles one hopes
to distinguish between), and subject them to the RT-PCR/RFLP assay. Both
alleles should show up roughly equally. One difficulty here can be if there is
significant genomic DNA in either of the samples, which may render getting
equivalent amounts of RNA difficult. In this case, one can mix cDNAs made
30_Vrana_439_450_F 444 8/29/05, 11:25 AM

from the two samples, using amounts that have yielded roughly equivalent PCR
products.
The assumption one makes when performing these allelic usage assays is
that the ratio of one allele to the other on the gel will reflect the actual usage on
the two parental chromosomes. Another danger is that as the PCR goes beyond
the linear amplification range, the lower expressed allele may “catch-up” with
the dominant allele. This again appears more likely with more rare transcripts.
First, the assay should be repeated several times, and preferably in both direc-
tions—that is, if one has alleles A and B, once in which A is the maternal
allele, and once in which B is the maternal allele. (This is a desirable experi-
ment regardless; there may be leakier imprinting in one direction.)
One way to prevent the “catching-up” phenomenon is to lower the number of
PCR cycles to approx 18–20, so that one is well within the linear range. To
visualize the PCR products after such low cycle numbers, one can add P32-
labeled CTP to the PCR cocktail. One can then use a phosphorimager to quan-
tify the ratio of one allele to the other. I should note here that one could use this
trick for quantification if a gel documentation system with software for ethidium
bromide-stained gels is not available. Often, imprinting will not be an all-or-
none situation, but rather a pronounced bias. It is advisable to quantify the ratio
between alleles on several different experiments to ascertain consistency.
3.5. RFLP Alternatives

Unfortunately, finding an RFLP between two alleles is not always possible.
If there are multiple (non-RFLP) polymorphisms, one alternative is to design a
pair of allele-specific primers. These primers are designed so that each primer
ends on a polymorphism: thus, two pairs of primers are designed, one for allele
A, and one for allele B. These assays may take some tweaking; primer concen-
tration is especially important for these assays. Because most oligonucleotide
preparations contain some non-full-length oligonucleotides, both alleles may
amplify if too much is used. An important consideration is the nature of the
(presumably) single nucleotide polymorphism (SNP). Transversions are
heavily favored over transitions for the final primer base, as a result of the fact
that G:T and A:C base-pairs can pair to some extent. Switches such as
purine:purine are optimal, because of the bulge they produce (18).
Readers are urged to consult the books PCR Protocols (18) and PCR Applica-
tions (19) for more in-depth comments and considerations on this sort of assay.
Perhaps the quickest (although not the least expensive) method of develop-
ing an allelic usage assay is simply to have the RT-PCR products sequenced.
Given that the purified PCR product must be directly sequenced (unless one is
willing to sequence many clones of each sample to ensure adequate representa-
tion), care should be taken to make sure that the primers are at least 20–30
30_Vrana_439_450_F 445 8/29/05, 11:25 AM

446 Vrana
bases from the polymorphism. This distance is to ensure that the polymorphic
base is read by the sequencing polymerase. Occasionally, PCR primers need to
be modified for sequencing purposes, particularly if one is not using a PCR-
based sequencing reaction. The mixing control in this case may be the genomic
DNA of a heterozygote, or an artificial mix. For certain model organisms such
as humans and mice, for which there are SNP databases, microarray based
allelic usage assays will likely emerge in the next decade.
Another method often used by human geneticists is the single-stranded con-
formation polymorphism (SSCP) assay. This method utilizes the fact that
single-stranded DNA of the same length, but different base composition, will
migrate differently in certain matrices. Even SNPs can result in mobility dif-
ferences between alleles. After the RT-PCR product is denatured, it is run on
long gels to allow ample room for separation. Typically these gels are either
acyrlamide or a matrix designed for this procedure termed mutation detection
enhancement gel solution. Protocols are readily available on the Internet (e.g.
http://www.cambrex.com/Content/Documents/Bioscience/MDE(18153).pdf).
The SSCP assay is often done “hot,” which adds the cost of the P32 nucleotide
and radioactive waste disposal.
Another option is the single nucleotide primer extension (SNuPE) assay
(20). In this method, an oligonucleotide primer is designed to bind immedi-
ately adjacent to a known SNP. This primer is then extended by a polymerase
in the presence of a complementary radio-labeled nucleotide to one of the two
bases present at the polymorphic site. The amount of each of the two DNTPs
incorporated from the extension of the RT-PCR of the message is then quanti-
fied after gel electrophoresis. Alternatively, fluorescently labeled DNTPs may
be combined with capillary electrophoresis.
The future of PCR-based allele-specific assays will likely see more allele-
specific real-time reverse-transcribed (or AS-RT2, as we refer to it) PCR. This
technique is a variation of the common “TaqMan” strategy. Normally, a probe
is hybridized to nascent RT-PCR products. This probe has both a reporter fluo-
rescent dye molecule and a quencher. The quencher acts only when it is in
close proximity to the dye. As the product is used as a template for the next
round of amplification, the quencher is detached, and the reporter fluoresces.
The amount of the fluorescence is then quantified. Applied Bio Systems (ABI
PRISM) machines (Foster City, CA) are among the most commonly used for
such detection. An excellent diagram of the process is viewable at http://
www.med.unc.edu/anclinic/Tm.htm.
In this allele-specific TaqMan reaction, oligonucletide probes (with differ-
ent fluorescent labels attached) must be designed so that they bind to the
amplicons in an allele specific fashion. AS- RT2 would seem optimal in that
the levels of message coming from each allele can be quantified. This tech-
30_Vrana_439_450_F 446 8/29/05, 11:25 AM

nique has only recently been implemented (21), but its use will likely grow
rapidly.
3.6. Non-PCR-Based Methods of Detecting Differences in Allelic Usage
There are at least two other methods that may be used for allelic usage. The
first allele-specific FISH is certainly the most visually compelling method in
that one can see the expression of the actual chromosome. This technique
requires a polymorphism sufficient to prevent the probe (which, in the case
of FISH, tends to be large) from hybridizing to both alleles. Again, this is tech-
nically quite demanding.
Another method that can show allele usage and is quantitative is the RNase
protection assay (RPA). The RPA method uses a labeled probe, which binds to
a region of the message of interest, which then “protects” it from subsequently
added RNases. The mix is then run out on an acyrlamide gel, dried, and the
intensity of the band can be used to estimate the amount of message originally
present. The difficulty with this assay comes with developing two probes, each
of which are allele-specific, and this explains why, with certain exceptions
(e.g., there is an excellent H19 allele-specific assay in house mice [22]), it is
not used often in imprinting studies. General RPA protocols can be found on
the web (http://micro.nwfsc.noaa.gov/protocols/) and kits are available from
such companies as Ambion Inc. (Austin, TX).
3.7. Assessing Allelic Usage in Cases of Gene “Knockouts”
and Uncharacterized Genes
Two cases where typical allelic usage assays may not be possible are when
(1) the gene itself has been localized, but not identified, and (2) the gene in
question has been deleted, either by targeted mutation or via other means.
Imprinting of genes in these cases may be inferred depending on the inherit-
ance patterns. For example, the imprinting of Igf2 was discovered when it was
the subject of one of the first mouse targeted mutations. While the resulting
growth retardation phenotype was expected with reduction in Igf2 expression,
the genetics of the phenotype were not. When the mutation was passed mater-
nally, there was no effect on growth, while paternal inheritance gave a dra-
matic growth reduction (23). The homozygous null animals correspondingly
showed a phenotype equivalent to those offspring who had only received the
null mutation paternally.
Passing a mutant or novel allele both paternally and maternally is certainly
prudent in cases where the gene may not be directly assayed. The results of a
genetic test may not be as clear-cut as was the Igf2 situation. That is, although
one parental inheritance will likely yield more severe phenotypes than the
other, neither will result in wild-type offspring. Such a result is likely due to
30_Vrana_439_450_F 447 8/29/05, 11:25 AM

448 Vrana
the fact that many imprinted genes do not exhibit complete (all or none)
monoallelic expression. Such genetic tests are sometimes termed “functional
imprinting assays” in that they demonstrate the effects and/or consequences of
an imprinted gene. Assaying allelic usage of the endogenous gene, however, is
still necessary to confirm the effect.
3.8. Summary Flow Chart of Developing a PCR-Based Assay
to Determine Allelic Usage
Typically, this process starts with a notion that a gene might be imprinted
because of it is located near an imprinted domain, it is a candidate for involve-
ment in a parent-of-origin effect, or it is expressed in a tissue where imprinted
genes are known to be very abundant (e.g., the placenta).
1. Work out a protocol for amplifying the gene of interest.
2. Amplify as large a piece of the mRNA as possible from individuals likely to have
genetic differences (e.g., different strains). If a length difference is apparent, you
are set!
3. Digest these RT-PCR products with various restriction enzymes, particularly
those with nonspecific or four-base recognition sequences.
4. If no RFLP is apparent in several tries, ascertain an expressed polymorphism in
the gene of interest through (re-)sequencing multiple individuals.
5. Analyze the sequences for polymorphisms, and potential RFLPs.
6. If potential RFLPs are present, test them with the appropriate enzymes.
7. If there are no potential RFLPs, but multiple SNPs, one may try pairs of allele-
specific primers.
8. Alternatively, if there is only one SNP, or good primer designs are not possible,
sequence the RT-PCR products.
9. If there appears to be mono-allelic (or biased) expression, carefully test for inher-
ent amplification bias by performing a “mixing experiment.”
10. Quantify the ratio of one allele to the other.
4. Notes
1. One potential problem with these assays that is unique to the placenta is the poten-
tial contamination with maternal tissues. In the case of paternally expressed genes,
this is of course not an issue. One easy way to rule out maternal contamination is
to ascertain whether the gene of interest is expressed in (pregnant) uterine tissue.
If the gene is not expressed here, it makes a contamination artifact unlikely
(unless the gene is only expressed in uterine tissue in close contact with fetal
tissue, or is expressed in blood). Showing that the gene is imprinted in other
tissues (e.g., yolk sac or embryonic tissue) also strengthens a placental imprint-
ing argument. Finally, if the gene is only expressed/imprinted in the placenta,
one must be particularly careful about dissection. Use as late-stage a placenta as
possible, and utilize tissue farthest from the maternal surface.
30_Vrana_439_450_F 448 8/29/05, 11:25 AM

2. Most of the potential pitfalls in the process have been previously described. How-
ever, perhaps the most common and (most difficult) problem in developing these
assays is finding a polymorphism. If the gene of interest has many small exons, it
is advisable to sequence across as many as possible by sequencing RT-PCR products
rather than utilizing genomic DNA. Again, focusing on the untranslated regions is
advisable. If no polymorphisms are present, one should examine alternative
strains or individuals. Indeed, if one has this luxury (i.e., multiple strains or very
polymorphic population), start the process by examining 3–4 strains/individuals.
3. Another problem that may occur is co-amplification of a closely related gene.
One must then redesign primers such that only the gene of interest is amplified,
or choose a time point when the contaminating family member is not expressed.
This situation can be vexing, as we realized while characterizing an imprinted
placental lactogen (PL), and then testing other PLs for imprinting (10).
References
1. Tilghman, S. (1999) The sins of the fathers and mothers: genomic imprinting in
mammalian development. Cell 96, 185–193.
2. Haig, D. (1996) Placental hormones, genomic imprinting and maternal-fetal
comunication. J. Evol. Biol. 9, 357–380.
3. Moore, T. and Reik, W. (1996) Genetic conflict in early development: parental
imprinting in normal and abnormal growth. Rev. Reprod. 1, 73–77.
4. Zeng, S. M. and Yankowitz, J. (2003) X-inactivation patterns in human embry-
onic and extra-embryonic tissues. Placenta 24, 270–275.
5. Sangha, K. K., Stephenson M. D., Brown, C. J., and Robinson, W. P. (1999) Ex-
tremely skewed X-chromosome inactivation is increased in women with recurrent
spontaneous abortion. Am. J. Human Genet. 65, 913–917.
6. Constancia, M., Hemberger, M., Hughes, J., et al. (2002) Placental-specific IGF-
II is a major modulator of placental and fetal growth. Nature 417, 945–948.
7. Wake, N., Takagi, N., and Sasaki, M. (1978) Androgenesis as a cause of hydatidi-
form mole. J. Natl. Cancer Inst. 60, 51–57.
8. Judson, H., Hayward, B. E., Sheridan, E., and Bonthron, D. T. (2002) A global
disorder of imprinting in the human female germ line. Nature 416, 539–542.
9. Rogers, J. F. and Dawson, W. D. (1970) Foetal and placental size in a Peromyscus
species cross. J. Reprod. Fertil. 21, 255–262.
10. Vrana, P. B., Matteson, P. G., Schmidt, J. V., et al. (2001) Genomic imprinting of
a placental lactogen gene in Peromyscus. Dev. Genes Evol. 211, 523–532.
11. Vrana, P., Guan, X.-J., Ingram, R. S., and Tilghman, S. M. (1998) Genomic imprint-
ing is disrupted in interspecific Peromyscus hybrids. Nature Genet. 20, 362–365.
12. Moglabey, Y. B., Kircheisen, R., Seoud, M., El Mogharbel, N., Van den Veyver,
I., and Slim, R. (1999) Genetic mapping of a maternal locus responsible for famil-
ial hydatidiform moles. Human Mol. Genet. 8, 667–671.
13. Vrana, P., Fossella, J. A., Matteson, P., del Rio, T., O’Neill, M. J., and Tilghman,
S. M. (2000) Genetic and epigenetic incompatibilities underlie hybrid dysgenesis
in Peromyscus. Nat. Genet. 25, 120–124.
30_Vrana_439_450_F 449 8/29/05, 11:25 AM

450 Vrana
14. Li, Y. and Behringer, R. R. (1998) Esx1 is an X-chromosome-imprinted regulator

of placental development and fetal growth. Nat. Genet. 20, 309–311.
15. Fohn, L. E. and Behringer, R. R. (2001) ESX1L, a novel X chromosome-linked
human homeobox gene expressed in the placenta and testis. Genomics 74, 105–108.
16. Moore, T. and Haig, D. (1991) Genomic imprinting in mammalian development:
a parental tug-of-war. Trends Genet. 7, 45–49.
17. Auffray, C. and Rougeon, F. (1980) Purification of mouse immunoglobulin heavy-
chain messenger RNAs from total myeloma tumor RNA. Eur. J. Biochem. 107,
303–314.
18. Innis, M. A., Gelfand, D. H., Sninsky, J. J., and White, T. J. (eds.) (1990) PCR
Protocols: A Guide to Methods and Applications. Academic, San Diego, CA.
19. Innis, M. A., Gelfand, D. H., and Sninsky, J. J. (eds.) (1999) PCR Applications:
Protocols for Functional Genomics. Academic, San Diego, CA.
20. Singer-Sam, J., LeBon, J. M., Dai, A., and Riggs, A. D. (1992) A sensitive, quan-
titative assay for measurement of allele-specific transcripts differing by a single
nucleotide. PCR Methods Appl. 1, 160–163.
21. Weber, M., Hagege, H., Lutfalla, G., et al. (2003) A real-time polymerase chain
reaction assay for quantification of allele ratios and correction of amplification
bias. Anal. Biochem. 320, 252–258.
22. Bartolomei, M. S., Zemel, S., and Tilghman, S. M. (1991) Parental imprinting of
the mouse H19 gene. Nature 351, 153–155.
23. DeChiara, T. M., Robertson, E. J., and Efstratiadis, A. (1991) Parental imprinting
of the mouse insulin-like growth factor II gene. Cell 64, 849–859.
30_Vrana_439_450_F 450 8/29/05, 11:25 AM

Gene Delivery to Trophoblast 451
31
Adenoviral-Mediated Gene Delivery to Trophoblast
Cells
Bing Jiang and Carole R. Mendelson
Summary
This chapter focuses on technology for construction of recombinant adenoviruses contain-
ing reporter genes under the control of putative regulatory regions of the human (h)CYP19
(aromatase) gene, as well as expression vectors. These recombinant adenoviruses have been
used in primary cultures of human placental cells to characterize regulatory regions of the
hCYP19 gene and to analyze the function of transcription factors on hCYP19 expression and on
trophoblast differentiation.
Key Words: Trophoblast; recombinant adenoviruses; CYP19 gene; placenta; aromatase.
1. Introduction
Cytotrophoblast proliferation and differentiation to syncytiotrophoblast is
key to implantation and human placental development. As cytotrophoblasts
mature, they stop dividing and spontaneously fuse to form the terminally dif-
ferentiated syncytiotrophoblast layer that functions in gas and nutrient exchange
and in biosynthesis of steroid and polypeptide hormones (1). In previous studies
using trophoblast cells from human mid-gestation placenta in primary culture,
we observed that differentiation of cytotrophoblasts to syncytiotrophoblast was
associated with a marked induction of aromatase activity and of CYP19
(aromatase P450) gene expression (2). In humans, aromatase P450—the key
regulatory enzyme in estrogen biosynthesis—is expressed in a number of tis-
sues, including ovary and testis, brain, adipose stromal cells, and the syncy-
tiotrophoblast cells of the placenta (3). CYP19 gene expression in these tissues
is driven by tissue-specific promoters upstream of tissue-specific alternative
first exons, which encode the 5'-untranslated regions of CYP19 mRNA tran-
scripts. These alternative first exons, which are located from approx 110 to
451
31_Jiang_451_462_F 451 8/29/05, 11:25 AM

452 Jiang and Mendelson
approx 100,000 bp upstream of the CYP19 translation initiation site in exon II,
are alternatively spliced onto a common site just upstream of the translation
start site in exon II so that the protein encoded in each of these tissues is identi-
cal. In placenta, the majority of the CYP19 mRNA transcripts contain sequences
encoded by exon I.1, which lies approx 100,000 bp upstream of the start site of
translation in exon II (4). To analyze the genomic regions and response ele-
ments that mediate syncytiotrophoblast-specific hCYP19 gene expression, we
have transfected human trophoblast cells in primary culture with fusion genes
containing various amounts of DNA upstream of placenta-specific exon 1.1
(with or without mutations of putative response elements), fused to the human
growth hormone (hGH) structural gene, as a reporter. In some cases, expres-
sion vectors containing transcription factors that play a potential regulatory
role in hCYP19 gene expression and trophoblast differentiation are co-trans-
fected. Primary cultures of human trophoblast cells are highly resistant to stan-
dard gene transfection methods, such as DEAE-dextran, calcium phosphate,
lipofection, and electroporation. To circumvent this barrier, we have incorpo-
rated the fusion genes and expression vectors of interest into the genome of a
replication-defective human adenovirus and introduced these DNA constructs
into the human placental cells by infection (2,5). In our earlier studies, the
recombinant adenoviral particles were produced by in vivo recombination in
293 cells (2,5). In vivo recombination in mammalian cells is relatively ineffi-
cient and time-consuming. Therefore, more recently, we have utilized a highly
efficacious method developed by Vogelstein and colleagues (6) in which the in
vivo recombination step is carried out in bacteria rather than in mammalian
cells (7,8). The recombinant adenoviral plasmids are then transfected into 293
cells for production of recombinant adenoviral particles, which are titered to
ascertain the concentration of infectious viral particles (multiplicity of infec-
tion [MOI]) and used to transfer gene constructs of interest into the placental
cells in primary monolayer culture by infection. In experiments in which an
expression vector containing a transcription factor or other putative regulatory
factor is introduced into the placental cells, a control adenovirus containing the
gene for bacterial β-galactosidase under control of the human cytomegalovirus
(hCMV) promoter is used to infect a parallel set of dishes at the same MOI to
control for nonspecific effects of the adenoviral infection. In this chapter, we
describe our methods for isolation and primary culture of human trophoblast
cells, preparation of recombinant adenoviral particles, and their use for intro-
duction of gene constructs into the primary cultures of trophoblast cells by
infection.
31_Jiang_451_462_F 452 8/29/05, 11:25 AM

2. Materials
2.1. Generation of Recombinant Adenoviruses
1. Shuttle vectors: pShuttle, pShuttle-CMV, pAdTrack, and pAdTrack-CMV
(kindly provided by Dr. Bert Vogelstein, Johns Hopkins Oncology Center, Balti-
more, MD; also available from American Type Culture Collection [ATCC,
Manassas, VA] and from Stratagene [La Jolla, CA]).
2. Adenoviral plasmids: pAdEasy-1 and pAdEasy-2 (kindly provided by Dr. Vogelstein;
also available from ATCC and from Stratagene).
3. Escherichia coli BJ5183 (kindly provided by Dr. Vogelstein), E. coli DH5α
(Invitrogen, Carlsbad, CA; also available from ATCC and from Stratagene).
4. Restriction enzymes Pme I and Pac I (New England Biolabs Inc., Beverly, MA).
5. Ampicillin and kanamycin (Sigma, St. Louis, MO).
6. Luria-Bertani (LB) medium (BD Biosciences, Franklin Lakes, NJ).
7. 10% glycerol (Fisher Scientific, Fair Lawn, NJ).
8. Bio-Rad Gene Pulser electroporator (Bio-Rad Laboratories, Hercules, CA).
9. LB/kanamycin plates.
10. Promega Miniprep kits (Promega Corporation, Madison, WI).
11. SuperFect transfection reagent (Qiagen Inc., Valencia, CA) or other transfection
reagent.
12. Phosphate-buffered saline (PBS).
13. Hank’s balanced salt solution (HBSS), pH 7.4 (GIBCO, Grand Island, NY).
14. 293 cells (ATCC, Manassas, VA) are cultured in Dulbecco’s modified Eagle’s
nedium (DMEM; Mediatech, Inc. , Herndon, VA).
15. Overlay Agarose (BioWhittaker Molecular Applications, Rockland, MD), a ster-
ile solution of melted 1% agarose in 25 mM HEPES, pH 7.4 (50°C).
16. 2X DMEM (GIBCO, Grand Island, NY) is used for titration of recombinant
adenoviruses.
17. Neutral red (Sigma, St. Louis, MO).
2.2. Isolation and Primary Culture of Cytotrophoblast Cells

and Infection of Trophoblast Cells With Recombinant Adenoviruses
1. Mid-trimester human placenta (20- to 24-wk gestation; Advanced Bioscience
Resources, Inc., Alameda, CA).
2. Trypsin (Invitrogen, Carlsbad, CA).
3. Fetal bovine serum (FBS; Gemini Bio-Products, Woodland, CA).
4. Percoll (Amersham Pharmacia Biotech AB, Uppsala, Sweden).
5. HBSS, pH 7.4 (GIBCO, Grand Island, NY).
6. DMEM is the medium used for primary culture of human trophoblasts.
7. DMEM supplemented with 10% FBS and 1.2% antibiotic/antimycotic solution
(GIBCO, Grand Island, NY).
31_Jiang_451_462_F 453 8/29/05, 11:25 AM

3. Methods
The methods described below outline: (1) modified adenoviral generation
method diagrammed schematically in Fig. 1; and (2) isolation and infection of
trophoblast cells.
3.1. Generation of Recombinant Adenoviruses
3.1.1. Preparation of Electrocompetent Bacterial Cells
1. Inoculate a fresh colony of BJ5183 into 10 mL LB medium. Shake the cells over-
night at 37°C.
2. Add 1 mL of cells into 1000 mL of LB medium in a 5-L flask. Grow for 4 to 5 h
at 37°C, until A550 is approx 0.8.
3. Collect cells in two 500-mL conical centrifuge bottles and incubate on ice for 10 min
to 1 h (the longer the cells are incubated the higher the competency).
4. Centrifuge at 2600g at 4°C for 10 min to pellet cells.
5. Resuspend the cell pellet in 1000 mL of sterilized, ice-cold 10% glycerol.
6. Centrifuge and pellet the cell suspension at 2500g for 30 min.
7. Repeat steps 5 and 6.
8. Pour off most of the supernatant; gently pipet off the remaining supernatant, leav-
ing about 20 mL. Resuspend the cells in the remaining supernatant and transfer
the cell suspension to a 50-mL tube. Spin at 2600g for 10 min, and pipet off all
but 5 mL of the supernatant.
9. Resuspend the cell pellet in the remaining supernatant. Aliquot in 50-µL aliquots
and store at –80°C.
3.1.2. Subcloning the Gene of Interest into the Shuttle Vector

The multiple cloning sites of pShuttle, pShuttleCMV, pAdTrack, and
pAdTrack-CMV vectors are shown Fig. 2. To investigate the regulatory regions
Fig. 1. (opposite page) Schematic outline of the AdEasy system. The gene of inter-
est is first cloned into a shuttle vector, e.g., pAdTrack-CMV. The resultant recombi-
nant plasmid is linearized by digesting with restriction endonuclease Pme I, and
subsequently cotransformed into Escherichia Coli BJ5183 cells with an adenoviral
backbone plasmid, e.g., pAdEasy-1. Recombinants are selected for kanamycin resis-
tance, and recombination is confirmed by restriction endonuclease analyses. Finally,
the linearized recombinant plasmid is transfected into adenovirus packaging cell lines,
e.g., 911 or 293 cells. Recombinant adenoviruses typically are generated within 7–10 d.
The ‘’left arm’’ and ‘’right arm’’ represent the regions mediating homologous recombi-
nation between the shuttle vector and the adenoviral backbone vector. A,
polyadenylation site; Bm, BamHI; RI, EcoRI; LITR, left-hand ITR and packaging
signal; RITR, right-hand ITR; Sp, Spe I. (Reprinted from ref. 6, Copyright 1998
National Academy of Sciences, USA).
31_Jiang_451_462_F 454 8/29/05, 11:25 AM

31_Jiang_451_462_F 455 8/29/05, 11:25 AM

Fig. 2. Shuttle vectors and adenoviral plasmids. (Reprinted from ref. 6, Copyright
1998 National Academy of Sciences, USA).
31_Jiang_451_462_F 456 8/29/05, 11:25 AM

of a gene of interest involved in tissue-specific expression, a fusion gene com-

prised of various amounts of DNA containing the putative regulatory region
fused to a reporter gene (e.g., hGH, β-galactosidase, luciferase) is subcloned
into the pShuttle or pAdTrack vector. To investigate the effects of overexpression
of a putative regulatory factor, the cDNA encoding this factor is subcloned into
pShuttle-CMV or pAdTrack-CMV vector. Restriction digestion and/or DNA
sequence analysis is carried out to ensure correct orientation of reporter gene
and expression vectors. The 5' CMV primer can be used to sequence the insert
within the pShuttleCMV vector, but internal primers must be used to sequence
the recombinant pAdTrack-CMV vector, which has a second CMV promoter
that drives green fluorescent protein (GFP) expression.
3.1.3. Cutting and Linearization of the Recombinant Shuttle Plasmid

With Pme I
Usually, one-fifth of a miniprep (typically 500 ng) is sufficient for this step.
After digestion, DNA is extracted with phenol-chloroform, precipitated with
ethanol, and resuspended in 7.0 µL of double-distilled (dd) H2O. One microli-
ter of the digested DNA and of undigested DNA is then loaded on an agarose
gel to determine if digestion is complete.
3.1.4. Co-Transformation of the Shuttle Plasmid and Adenoviral

Genomic Plasmid into BJ5183 Cells to Generate a Recombinant
Adenoviral Plasmid
Pme I-digested shuttle plasmid is co-transformed with 1 µL of adenoviral
backbone vector (100 ng/µL) pAdEasy-1 or pAdEasy-2 (Fig. 2). Twenty to
fifty microliters of electrocompetent E. coli BJ5183 cells are added and
electroporation is performed in 2.0-mm cuvets at 2500 V, 200 Ohms, and 25 µFD
in a Bio-Rad Gene Pulser electroporator. The transformation mix is resuspended
in 500 µL of LB medium (incubation at 37°C for 10–20 min is optional), plated
onto three to five LB/kanamycin plates, and grown at 37°C overnight (see
Notes 1 and 2).
3.1.5. Screening of Recombinant Adenoviral Plasmids

1. Ten to twenty of the smallest colonies (the smallest colonies usually contain the
recombinant adenoviral plasmids) are picked and grown overnight in 3 mL LB
containing 25 µg/mL kanamycin. Minipreps are performed using the conven-
tional alkaline lysis method (e.g., Promega Miniprep kit). One-fifth of the
miniprep DNA is analyzed by restriction enzyme digestion using Pac I, followed
by electrophoresis on a 0.8% agarose gel. Candidate clones usually yield a large
fragment (~30 kb), plus a smaller fragment of 3.0 or 4.5 kb (see Note 3).
31_Jiang_451_462_F 457 8/29/05, 11:25 AM

2. The insert is sequenced using appropriate primers. Two microliters of the recom-
binant miniprep DNA is re-transformed into DH5α (or a comparable plasmid
propagation strain). Plasmids are then purified for adenoviral production in 293
cells.
3.1.6. Adenoviral Production by Transfection of Recombinant

Adenoviral Plasmids into 293 Cells
1. 293 cells (E1-transformed human embryonic kidney cells) are cultured in one
60-mm dish at 20–30 × 104 cells per flask the day before transfection. Confluency
should be approx 50% to 80% at the time of transfection.
2. Before transfection, recombinant adenoviral plasmids are digested with Pac I
(usually 4 µg DNA are required to transfect one 60-mm dish). The plasmid DNA
is ethanol precipitated and resuspended in 20 µL of sterile H2O. Complete diges-
tion of the DNA is ascertained on an agarose gel.
3. The Pac I-digested recombinant adenoviral plasmids are transfected into the 293
cells using SuperFect or a comparable transfection reagent according to
manufacturer’s instructions. Briefly, 4 µg of Pac I-digested plasmid and 30 µL of
SuperFect are combined for each 60-mm dish and incubated at room temperature
for 5–10 min before adding to the cultured 293 cells.
4. While the digested plasmid DNA and SuperFect are incubating, culture medium is
removed from the 293 cells and the cells are washed once with 4 mL sterile PBS.
5. One milliliter of DMEM containing serum and antibiotics is added to the reaction
tube containing digested adenoviral DNA and SuperFect reagent. This is then
mixed well and added to the washed cells, which are then incubated in a 37°C
CO2 incubator for 2–3 h.
6. Remove medium and wash the cells once with 4 mL of PBS. Add 4 mL of fresh
medium (containing serum and antibiotics).
7. Transfection efficiency and adenoviral production can be monitored by GFP
expression if pAdTrack-based vectors are used. Plaques are readily observed
under fluorescence using the GFP marker.
8. Seven to ten days post-transfection, the cells are scraped from the flasks and
transferred to 10-mL conical tubes. The cells are sedimented in a benchtop cen-
trifuge and the pellet resuspended in 2.0-mL sterile HBSS or PBS. The resus-
pended cells are frozen in a dry ice/ethanol bath, and then thawed in a 37°C water
bath and vortexed vigorously. The freeze/thaw/vortexing is repeated for three
more cycles (four cycles total). The supernatants should not be allowed to warm
up. The cell debris is then sedimented by centrifugation at 1000g and the aden-
ovirus containing supernatants stored at –80°C.
9. To propagate the recombinant adenoviruses, two 50% to 70% confluent 60-mm
dishes of 293 cells are infected using 30–50% of the collected adenoviral super-
natants for each flask. Cell lysis should be evident within 2 to 3 d post infection.
Productive infections are easily observed with the AdTrack vectors.
10. When one-third to one-half of the cells are detached (usually 3 to 5 d post infec-
tion), the virus-containing medium is collected and analyzed for the presence of
31_Jiang_451_462_F 458 8/29/05, 11:25 AM

recombinant adenoviruses by Northern blot/Western blot and/or polymerase

chain reaction (PCR). For PCR, take 5 µL virus supernatant plus 10 µL PCR-
grade Proteinase K at 55°C for 1 h, then boil samples for 5 min. Spin briefly and
use 1 to 2 µL for PCR using primers corresponding to the insert.
11. If virus production is evident, the cells are scraped from the dishes and viral
supernatants are prepared as described above. There should be at least 107 infec-
tious viral particles/mL at this stage, and ideally much more. Each round of ampli-
fication should yield at least 10-fold more virus than present in the previous round.
12. To amplify further, infection of the 293 cells is repeated using 30–50% of the
viral supernatant from step 11, using 100-mm dishes instead of 60-mm dishes.
Viral titers can be measured at any time, and this is particularly easy with AD-
Track vectors. The 293 cells are simply infected with various dilutions of viral
supernatant and the number of plaques that fluoresce green within 18 h is ascer-
tained. If AdTrack is not used, viruses can be titered using standard methods,
described as follows.
3.1.7. Titration of Recombinant Adenoviruses
1. Various dilutions of the recombinant adenoviruses are prepared. One millilter is added
to each 60-mm dish containing 80–90% confluent 293 cells and incubated for 1 h.
2. To prepare the overlay agarose, a sterile solution of melted 1% agarose in 25 mM
HEPES, pH 7.4 (50°C) is combined with an equal volume of 2X DMEM medium
in a 37°C waterbath.
3. One hour after virus infection, the medium is removed and the cells are overlaid
with 4 mL of 0.5% agarose in 1X DMEM.
4. The cells are incubated at 37°C for 7–10 d. Neutral red (100 µL) is then added to
dishes containing 4 mL medium to achieve a final concentration of 0.025% be-
fore counting plaques.
3.2. Isolation and Primary Culture of Cytotrophoblast Cells
and Infection of Trophoblast Cells With Recombinant Adenoviruses
3.2.1. Primary Culture of Human Trophoblast Cells
Mid-trimester human placental tissues are obtained in accordance with the
Donors Anatomical Gift Act of the State of Texas after obtaining consent in
writing. In all cases, consent forms and protocols are approved by the Institu-
tional Review Board of the University of Texas Southwestern Medical Center
at Dallas. A placental primary culture system has been modified for isolation
and culture of cytotrophoblasts from mid-gestation human placenta (9).
1. Briefly, the placental tissues are washed with HBSS, pH 7.4, then finely minced
and digested with 0.125% trypsin in HBSS at 37°C for 20 min. This procedure is
repeated three times.
2. At the end of each digestion step, the supernatant is collected, layered over 10 mL
FBS, and then briefly centrifuged at 1200g.
3. The pellet is suspended in DMEM, filtered, and layered over a Percoll gradient
31_Jiang_451_462_F 459 8/29/05, 11:25 AM

(70%–5%).
4. The gradients are centrifuged at 1200g for 20 min at room temperature, and cells
in the middle layer (density 1.045–1.062) are collected, washed, and counted.
5. The cells are then resuspended in DMEM supplemented with 10% FBS and 1.2%
antibiotic/antimycotic solution and plated at a density of 2 × 106 cells per dish in
35-mm culture dishes or 15 × 106 cells per dish in 100-mm dishes.
3.2.2. Infection of Human Trophoblast Cells With Recombinant

Adenoviruses
1. To characterize the regulatory regions of a gene of interest involved in directing
placenta-specific expression, cytotrophoblasts are plated at a density of 2 × 106
cells per dish in 35-mm culture dishes and infected with recombinant adenoviruses
(MOI = 0.5) containing fusion genes comprised of various amounts of potential
regulatory sequence fused to a reporter gene, such as luciferase, β-galactosidase,
or hGH.
2. After overnight incubation, the medium is changed to DMEM containing 2%
FBS. At this point, hormones or factors can be added and the incubation contin-
ued.
3. Cells are harvested after various periods of incubation and assayed for β-galac-
tosidase or luciferase activity. If hGH is used as the reporter, media are collected
at various intervals and analyzed for secreted hGH by radioimmunoassay (see
Note 4).
4. To investigate the roles of potential regulatory factors on human trophoblast dif-
ferentiation and expression of trophoblast-specific genes, cytotrophoblasts are
plated at a density of 15 × 106 cells per dish in 100-mm culture dishes and infected
within 1 h of plating with recombinant adenoviruses containing expression vectors
for these putative regulatory factors at an MOI between 0.5 and 10.0 (see Note 5)
5. After various periods of incubation, the cells are harvested for analysis of mor-
phological changes as well as alterations in expression of specific genes using a
variety of techniques.
4. Notes
1. In the generation of recombinant adenoviruses, Pme I and Pac I are used to lin-
earize the final constructs for transformation and transfection. Therefore, use of
inserts containing these sites will be problematical.
2. It is critical that recombinant adenoviral plasmids be generated in bacteria using
electroporation rather than other methods of transformation.
3. All the constructs (including recombinant adenoviral plasmids) are resistant to
kanamycin (NOT ampicilin) except pAdEasy-1 and pAdEasy-2.
4. It is essential that the cytotrophoblasts be exposed to recombinant adenovirus
within 1 h of plating, because the cells become resistant to adenoviral infection
upon differentiation to syncytiotrophoblast (10).
5. It is always important to infect parallel cultures with a “control” recombinant
31_Jiang_451_462_F 460 8/29/05, 11:25 AM

adenovirus at similar MOI to evaluate nonspecific effects of adenoviral infec-

tion. We have used a recombinant adenovirus expressing β-galactosidase under
control of the CMV promoter as a control.
Acknowledgments
We thank Vickey Chau and Jo Smith for isolation of cytotrophoblast cells.
Research using these techniques was supported by National Institutes of Health
(NIH) R01 DK-31206.
References
1. Ringler, G. E. and Strauss, J. F., III (1990) In vitro systems for the study of human
placental endocrine function. Endocr. Rev. 11, 105–123.
2. Kamat, A., Alcorn, J. L., Kunczt, C., and Mendelson, C. R. (1998) Characteriza-
tion of the regulatory regions of the human aromatase (P450arom) gene involved
in placenta-specific expression. Mol. Endocrinol. 12, 1764–1777.
3. Kamat, A., Hinshelwood, M. M., Murry, B. A., and Mendelson, C. R. (2002)
Mechanisms in tissue-specific regulation of estrogen biosynthesis in humans.
Trends Endocrinol. Metab. 13, 122–128.
4. Means, G. D., Mahendroo, M. S., Corbin, C. J., et al. (1989) Structural analysis of
the gene encoding human aromatase cytochrome P-450, the enzyme responsible
for estrogen biosynthesis. J. Biol. Chem. 264, 19385–19391.
5. Alcorn, J. L., Gao, E., Chen, Q., Smith, M. E., Gerard, R. D., and Mendelson, C.
R. (1993) Genomic elements involved in transcriptional regulation of the rabbit
surfactant protein-A gene. Mol. Endocrinol. 7, 1072–1085.
6. He, T. C., Zhou, S., da Costa, L. T., Yu, J., Kinzler, K. W., and Vogelstein, B.
(1998) A simplified system for generating recombinant adenoviruses. Proc. Natl.
Acad. Sci. USA 95, 2509–2514.
7. Jiang, B., Kamat, A., and Mendelson, C. R. (2000) Hypoxia prevents induction of
aromatase expression in human trophoblast cells in culture: potential inhibitory
role of the hypoxia-inducible transcription factor Mash-2 (mammalian achaete-
scute homologous protein-2). Mol. Endocrinol. 14, 1661–1673.
8. Jiang, B. and Mendelson, C. R. (2003) USF1 and USF2 mediate inhibition of
human trophoblast differentiation and CYP19 gene expression by Mash-2 and
hypoxia. Mol. Cell. Biol. 23, 6117–6128.
9. Kliman, H. J., Nestler, J. E., Sermasi, E., Sanger, J. M., and Strauss, J. F., III
(1986) Purification, characterization, and in vitro differentiation of
cytotrophoblasts from human term placentae. Endocrinology 118, 1567–1582.
10. MacCalman, C. D., Furth, E. E., Omigbodun, A., Kozarsky, K. F., Coutifaris, C.,
and Strauss, J. F., III (1996) Transduction of human trophoblast cells by recombi-
nant adenoviruses is differentiation dependent. Biol. Reprod. 54, 682–691.
31_Jiang_451_462_F 461 8/29/05, 11:25 AM

31_Jiang_451_462_F 462 8/29/05, 11:25 AM
Index 463
Index
A blastocyst, 4, 16, 26–29, 35–53, 130–136

adenogenesis, 86 adhesion, 40–42
adenoviral-mediated gene delivery, 6, attachment, 38–40, 45
451–461 collection of, 16, 19–22, 43, 44, 130,
Alexa, 358, 363, 364 131
alkaline phosphatase, 171, 206, 285, culture of, 35–53, 131–136, 181
287, 302, 303, 339 hatching, 37, 38
histochemistry, 285 structure of, 36, 132
alpha-fetoprotein (AFP), 145, 339 uterine transfer of, 16, 26–29
alpha-2 macroglobulin (α2-MG), 302, bone morphogenetic proteins (BMP),
303 187, 194
allantochorionic vessels, 394 bovine, 179–187, 323–333
allantoic explant cultures, 5, 241–270 placental characterization, 323–333
Alzet minipump, 102-104, 106, 108 trophoblast cell culture, 179–187
androstenedione, 168 breeding, 11, 13, 17, 18, 43, 252, 253,
anesthesia, 14–16, 30, 395, 396 298, 304
mouse, 14–16, 30, 388 mouse, 11, 13, 17, 18, 43, 252, 253
rat, 255 rat, 298, 304
ruminant, 395, 396 BT-1 cells, 179–187
AP2, activator protein-2, 200
apoptosis, detection techniques, C
338–348 cadherin-11, 339
aromatase (CYP19), 113, 114, 116, Callithrix jacchus, 111–121
118–120, 451–461 caruncle, ruminant uterus, 86, 315–317,
323, 400, 405
B caveolin, 339, 357, 363
baboon, 95–99, 101–109 cdx2, 145, 164
early pregnancy model, 101–109 CD9, 206, 222, 223, 224, 339
endometriosis model, 95–99 CD10, 339
BeWo, choriocarcinoma cells, 220, 221, CD14, lipopolysaccharide receptor,
229–238 206, 339
binucleate cells, ruminant trophoblast, CD31, PeCAM1, 288, 289
183, 184, 315–322, 323–333 CD34, 339
biotin-labeling, RNA, 419, 420 CD45, leukocyte common antigen, 339
blastocoel, 36, 37 CD68, 339
463
464 Index
CD105, endoglin, 339 choriocarcinoma cells, human,

CD133, 339 232, 233
cDNA synthesis, 417, 419 endometrial stromal cells, human, 82
cell cycle, 58 Rcho-1 trophoblast stem cells, rat,
Chicago blue, intravital dye, 14, 18, 22, 165
23, 300, 304 trophoblast-derived cell lines, human,
chorioallantoic placenta, 159, 160, 224
241–270, 275–285, 289, trophoblast stem cells
296–299, 300, 304, 305, bovine, 181
315–317, 355–359 mouse, 138, 153
dissection uterine stromal cells, rat, 59, 61, 62
mouse, 278–285 cryosections, ultrathin, 362-363
rat, 300, 304, 305 CTLA-4 (cytotoxic T lymphocyte-
fusion, 241–270, 276, 284 associated protein-4), 339
structure of cyclin D3, 164
human, 204 CYP19, aromatase, 113, 114, 116,
mouse, 243, 275–278, 289 118–120, 451–461
rat, 159, 160, 296–299 cytochrome P450 17α-hydroxylase
ruminant, 315–317 (P450c17), 164, 302, 303
villus structure, human, 355-359 cytochrome P450 side chain cleavage
choriocarcinoma, 5, 159–172, 221, (P450scc), 164, 302, 303
229–238 cytokeratin, 114, 120, 206, 215, 223,
cell lines, human, 221, 229–238 224, 303, 325–327, 339
Rcho-1 trophoblast stem cell line, cytoskeleton, 325-327
159–172 cytotrophoblast, 203–216, 220–226,
chorionic ectoderm, 243 337–339, 451, 452
chorionic gonadotropin, 101–108, 190,
195, 197–201, 204, 206, 215, D
230, 339 decidua, 13, 57–65, 69–77, 74–76,
choriovitelline placenta, 297 297, 302
chymotrypsin, 206 antimesometrial, 58, 69–70, 74–76,
cloning, cells, 165 297, 302
cluster analysis, DNA microarray, culture of, rat, 69–77
432–433 DNA microarray analysis of,
collagen, 46, 324, 325, 327, 330 425–431
colloidal gold, 354, mesometrial, 58, 69–70, 74–76,
connexin, 149–157 297, 302
copulatory plug, 11, 13, 18, 43, 252 primary decidual zone, mouse, 13
correlative microscopy, 351-367 secondary decidual zone, mouse 13
cow, 179–187, 323–333 decidual prolactin-related protein, 302
placental characterization, 323–333 decidualization, 12, 15, 16, 24–26, 32,
trophoblast cell culture, 179–187 69–77
cryopreservation, 59, 61, 62, 82, 138, induction of
153, 165, 181, 224, 232, 233 mouse, 12, 15, 16, 24–26, 32
Index 465
rat, 71, 75, 76 endothelium, capillary, 288

delayed implantation, 14, 15, 22-24 entactin, 46
desmin, 325-327, 339 eosmesodermin, 145, 164, 187, 282, 286
DNA methylation, 161 epidermal growth factor (EGF), 204
DNA microarray analysis, 6, 184–186, epitheliochorial, placentation,
411–421, 425–437 4, 393, 405
clustering, 432, 433 ERRβ, 145, 282
decidual and placental cells, human, estrogen, 12, 15, 25–27, 57–60, 95,
425–437 102, 113, 116, 118–120, 197
trophoblast cells endometriosis, baboon, 95
bovine, 184–186 hormonal priming, mouse, 12, 15,
human, 411–421 25–27
measurement, 102, 197
E receptor, 112, 113, 116, 118–120
E-cadherin, 339 treatment of uterine stromal cells,
ectoplacental cone, 126, 139, 280 57–60
ED-1/ED-2, 303 estrous cycle, 10, 13, 17, 90, 308
endoglin (CD105), 339 mouse, 10, 13, 17
electron microscopy, 288, 317–322, rat, 308
351–367, 373, 380, 384, 388, sheep, 90
389, 396, 398–402 Esx1, 145
placenta euthanasia, 16, 31, 32
human, 351–367 mouse, 16, 31, 32, 381
mouse, 288 rat, 256
ruminant, 317–322 ruminant, 318, 396
scanning, 373, 380, 384, 388, 389, extracellular matrix (ECM), 38–42,
396, 398–402 323–333
elutriation, cell separation, 72-73, 76 extraembryonic ectoderm, 126, 282
embryo, 16, 26–29, 260–262 extravillous trophoblast, 221, 222, 337
dissection, 262–262
transfer, 16, 26–29 F
embryonic stem cells, 126, 189–201 1F10, 339
human, 189–201 factor VIII, 289
mouse, 126 fatty acid binding protein-3 (FABP3),
embryonic fibroblast feeder layers, 126, 302, 303
127, 129, 130, 152, 191, 192 fetal death, experimental, 300, 306
endometriectomy, 104 fetoplacental vasculature, 375, 376,
endometriomas, 98 380–388, 397–406
endometriosis, 95-99 mouse, 375, 376, 380-388
endoreduplication, 126, 166, 183 ruminant, 397-406
endosome antigen-1 (EEA1), 356–358, α-fetoprotein (AFP), 145, 339
363 fibroblast growth factor-4 (FGF4),
endothelial cells, 339 126–128, 151, 155, 179
endotheliochorial, placentation, 394 fibroblasts, 339
466 Index
fibronectin, 41, 42, 46-50, 53, 324, 327 derivation of trophoblast cell lines,
binding assay, 48, 53 220–226
FISH, fluorescence in situ hybridization, embryonic stem cells, 5, 189–201
447 endometrial cells, 79–83
fixation, tissue, 318, 319, 328, 338, 342 primary trophoblast cell culture,
Flk-1, 243 203–216
flow cytometry, 128, 141–143, 197–200 3β hydroxysteroid dehydrogenase
FluroNanogold, 353, 359–361, 363, 364 (3βHSD), 164, 302
17β hydroxysteroid dehydrogenase
G (17βHSD), 113, 114, 116–120
β-galactosidase staining, 231, 236–238,
252, 253, 268 I
gap junction, 149–152 Id-1, 164
GATA, 200 image analysis, 47–51, 53, 361, 363
GB25, 339 immortalization, cell, 79–83
Gcm1, 145, 172, 200, 286 immunocytochemistry, 5, 115–117,
gene delivery, 231, 233–236, 451–461 303, 307, 320, 321, 325–333,
GeneSpring, 428, 433 338–348
glycogen cells, 159, 160, 298 protocol
glycosylated cell adhesion molecule-1 bovine placentome, 325–333
(GlyCAM-1), 89 human
goat, vascular corrosion casting, embryonic stem cells, 197
393–406 placenta, 337–348
marmoset uterus, 115–117
H ruminant placenta, 320, 322
HAI-1 (hepatocyte growth factor immunoelectron microscopy,
activator inhibitor type 1), 351–367
206, 339 immunofluorescence, 328, 329, 363
HAND1, 164, 286 immunomagnetic cell isolation, 205,
hemochorial, placentation, 4 207–210, 213–215
heparan sulfate, 49 implantation, 4, 9–34, 35–53, 101–109,
heparin, 126–128, 153 315–322, 323–333
Hertz, Roy, 229 analysis, baboon, 101–109
HFE (hemochromatosis protein), 339 induction of delayed implantation,
histotroph, 86, 89 14, 15, 22–24
HLA, 206, 207 in vitro, 35–53
HLA-G, 200, 223, 224 procedures, mouse, 9–34
Hoechst 33342, 128, 184 morphological analysis
Hofbauer cell, 338 bovine, 323–333
human, 5, 79–83, 189–201, 203–216, ruminant, 315–322
220–226, 229–238, 451–461 imprinting, genomic, 6, 439–448
adenoviral-mediated gene delivery, Indian hedgehog, 145
451–461 infection, adenoviral gene delivery,
choriocarcinoma cell culture, 229–238 451–461
Index 467
injections, 14, 15, 18, 25–27, 30, 31, Matrigel, 46

300, 304 matrix metalloproteinase, 118, 171
hormone, mouse, 15, 25–27 medroxyprogesterone acetate, 90
intravital dye, early pregnancy menstrual cycle, 99
detection, 14, 18, 22, 23, menstrual fluid, 96
300, 304 mesenchymal cells, 339
procedure, 16, 30, 31 metrial gland, 159, 160, 297, 298, 300,
inner cell mass, 36, 37, 126, 179 305, 306
in situ hybridization, 5, 301, 307 dissection of, 300, 305, 306
insulin-like growth factor-II (IGF-II, microcirculation, uteroplacental, 371–391,
Igf2), 302, 440 393–406
integrin, 41, 42, 89, 171, 246, 284, 324, microdrop cultures, 46, 52
325, 327, 330, 331 β-microglobulin, 339
interferon-tau (IFNτ), 88, 184, 187 microscopy, 351–367
invasive trophoblast, 159, 160, 172, correlative, 364–367
276–278, 288, 289, 298 fluorescence, 355
in vitro fertilization, 181 microsphere cultures, 48, 49, 52
microvasculature, placental, 371–391,
J 393–406
JAR, choriocarcinoma cell line, 221, mitomycin, 127
229–238 monoallelic gene expression, 439–448
JEG, choriocarcinoma cell line, 221, mouse, 9–34, 35–53, 125–146, 149–157,
229–238 241–270, 275-291, 371–391
junctional zone, chorioallantoic allantoic explant cultures, 241–270
placenta, 159, 160, 298, 299 blastocyst culture, 35–53
chorioallantoic fusion, 241–270
K implantation, 9–34
K4M, acrylic, 319 placental phenotypic analysis, 275–291
trophoblast stem cell culture, 125–146,
L 149–157
labyrinth zone, chorioallantoic vascular corrosion casting, 371–391
placenta, 126, 139, 159, 160, MTT assay, 60, 63
276, 284–289, 298, 372, 373 mucin glycoprotein-1 (Muc-1), 88, 339
laminin, 46, 287, 324, 325, 327, 330 myofibroblasts, 339
lectin histochemistry, 285–288
Lim kinase (Limk), 282 N
luteotrophic hormone, 101 natural killer cells, 297, 302, 303
nodal, 145
M northern blotting, 306
M30 immunohistochemistry,
340, 341, 343 O
macrophage, 206, 303, 339 Oct4, 187, 200
marmoset, 111–121 oligonucleotide microarray, 412–414
Mash2, 154, 164, 282, 286 osteopontin (OPN), 89, 302
468 Index
ovariectomy, 14–15, 22, 60 hormonal priming, mouse, 12, 15,

procedure mouse, 14-15, 22 25–27
rat, 60 induction of uterine gland knockout,
oxytocin, 88 86, 88–91
measurement, 168, 197
P receptor, 112, 113, 116, 118–120
P450c17, 164, 302, 303 treatment of uterine stromal cells,
P450scc, 164, 302, 303 57–60
PAL-E, 339 prolactin family miniarray, 301, 306, 307
Papio anubis, 95-99, 101-109 prolactin receptor, 91
early pregnancy model, 101–109 prolactin-like protein-A (PLP-A), 164,
endometriosis model, 95–99 303
Pem, 282 prolactin-like protein-B (PLP-B), 172,
Percoll density gradients, 209-214 302, 303
perforin, 303 prolactin-like protein-Fα (PLP-Fα),
Peromyscus (deer mice), 440 164, 302
placental circulation, 288, 299, 371–391 prolactin-like protein-Fβ (PLP-Fβ),
placental growth factor (PlGF) 200 172, 302
placental lactogen (PL), 180, 184, 187, prolactin-like protein-J (PLP-J), 302
204, 206, 222, 316, 339, 448 prolactin-like protein-K (PLP-K), 302
bovine, 180, 184, 187 prolactin-like protein-L (PLP-L), 172, 302
human (hPL), 206, 222, 339 prolactin-like protein-M (PLP-M), 164, 302
Peromyscus, 448 prolactin-like protein-N (PLP-N), 172, 302
placental lactogen-I (PL-I), 145, 154, prolactin-related protein-1 (PRP-1),
164, 282, 286, 289, 302, 303 180, 184, 187
placental lactogen-II (PL-II), 145, 164, proliferation, cell assays, 63, 154, 165,
286, 302, 303 166, 182, 183
placentomal microcirculation, ruminant, proliferin (PLF), 282, 286, 287, 289
402–403 proliferin-related protein (PLF-RP), 302
placentome, 315, 323–325, 393–395 prostaglandin F2α, 88, 90, 111, 112
platelet/endothelial cell adhesion pseudopregnancy, 15, 25, 71, 75
molecule-1 (PeCAM1), 288, 289 induction of
poly-L-ornithine, 222 mouse, 15, 25
polymerase chain reaction (PCR) based rat, 71, 75
assays, 442, 443
polyploidy, 126, 166 R
PPAR-γ (peroxisome proliferator- rat, 57–65, 69–77, 159–172, 247, 248,
activated receptor-γ), 339 254–257, 295–310
pregnancy-associated glycoprotein, blood collection, 247, 248, 254–257
180, 184, 187, 316 decidual cell culture, 69–77
pregnancy-specific glycoprotein, 164 placental phenotypic analysis, 295–310
progesterone, 12, 15, 25–27, 57–60, 86, Rcho-1 trophoblast stem cells, 159–172
88–91, 112, 113, 116, 118–120, uterine stromal cell culture, 57–65
168, 197, 204 Rcho-1 trophoblast stem cells, 159–172
Index 469
recombinant adenoviruses, 453–459 single-stranded DNA (ssDNA)

Reichert’s membrane, 261, 262 immunohistochemistry,
relaxin, 112, 114, 117 341, 344
retinoic acid, 151 smooth muscle
reverse transcription polymerase chain actin, 303, 326, 327, 339
reaction (RT-PCR), 112, 113, muscle myosin, 339
115, 444 SOCS 3, 164
aromatase, 113 SP1, 206
estrogen receptor, 113 spiral arteries, uterine, 289
imprinting analysis, 444 spongiotrophoblast, 126, 139, 151, 157,
progesterone receptor, 113 159, 160, 172, 284-289, 298,
protocol uterus, marmoset, 115 300, 305
17β hydroxysteroid dehydrogenase, culture of, 300, 305
113 spongiotrophoblast-specific protein,
RFLP (restriction fragment polymorphism), 172, 302
analysis, 445–446 stromal cells, uterine, 57–65, 79–83,
RGD peptides, 46 195, 196
RNA, 417–421, 444 culture of
biotin labeling of, 419, 420 human, 79–83, 195, 196
fragmentation of, 420 rat, 57–65
preparation superovulation, mouse, 43
DNA microarray, 417–419 syncytial trophoblast, 157, 159, 172, 195
imprinting analysis, 444 syncytin, 339
RNase protection assay (RPA), 447 syncytiotrophoblast, 204, 337–339, 451,
Rosa26 lacZ transgene, 247 452
ruminant placentation, 315–322 synepitheliochorial, placentation, 4, 86,
315–317, 323, 394
S
Sampson’s theory, 95, 96 T
scanning electron microscopy of telomerase, 79–83, 200
vascular casts, 373, 380, 384, terminal deoxynucleotidyl transferase-
388, 389, 396, 398–402 mediated dUTP nick-end
SGHPL-4, trophoblast-derived cell line, labeling (TUNEL), 341, 342,
220–226 344–348
sheep, 85–91, 393–406 Tpbpα (4311), 145, 154, 286, 287, 289
uterine glands, 85-91 Thomas–Friedenreich antigen, 339
vascular corrosion casting, 393–406 transcriptome, 411, 418
silver enhancement, 359-361,364, 365 transfection, 81–82, 128, 143, 144, 168,
Simian virus-40 (SV40), 220, 222 223, 224, 231–238
single nucleotide polymorphism (SNP), choriocarcinoma cells, human, 231,
445, 446, 448 233–238
single nucleotide primer extension, 446 endometrial stromal cells, 81–82
single-stranded conformation trophoblast stem cells, 128, 143, 144
polymorphism (SSCP), 446 Rcho-1 trophoblast stem cells, 168
470 Index
trophoblast-derived cell lines, uninucleate cells, ruminant trophoblast,

223, 224 315–316, 323–325
transforming growth factor-β, 190 uterine gland, 4, 85–91
trinucleate cells, ruminant trophoblast, sheep knockout model, 85–91
316 uterine mesometrial compartment, 297
trophectoderm, 36, 37, 125, 126, 179 uterine milk protein (UTMP), 89
mural, 36, 37 uterine stromal cell culture, 57–65,
polar, 36, 37 79–83, 195, 196
trophoblast, 5, 40, 42, 45–48, 125–146, uteroplacental vasculature, 288, 289,
149–157, 159–172, 166–168, 372–380, 393–406
179–187, 194–200, 203–216, mouse, 288, 289, 372–380
219–226, 280, 289, 298, 315– ruminant, 393–406
317, 322–324, 451–461
adenoviral-mediated gene delivery, V
451–461 vaginal plug, mouse, 11, 13, 18, 43, 252
derived cell lines, 220-226 vaginal lavage, mouse, 17
differentiation, 128, 139–143, 154, vasectomy, 15, 24–25, 71
159–172, 166–168, 183–185, procedure, mouse, 15, 24-25
194–200 rat, 71
establishment of cell lines, 219–226 vasopressin, 104
extravillous, 219 vascular corrosion casting, 371–391,
giant cells, 5, 125, 126, 138–142, 393–406
151, 157, 159–161, 280, mouse, 371-391
289, 298 ruminant, 393-406
invasion of, 40–42, 154, 168, 288, 289 vascularization, placental, 245, 288–291,
migration of, 315–317, 322–324 371–391, 393–406
outgrowth of, 45-48 VCAM-1, 246, 270, 284
primary culture of, human, 203–216, villous stromal cells, 203–207, 339
453, 459–461 villous trophoblast, 203–207, 339
stem cells, 5, 125–146, 149–157, villus, placenta, 355-359
159–172, 179–187 vimentin, 206, 325–327, 339
bovine, 179-187 vitronectin, 46
mouse, 125-146, 149-157 viviparity, 3
rat, 159-172 von Willebrand factor, 339
stem cell in vivo transplantation,
156, 161, 169 W
vesicles, 182, 195, 196 Western blotting, 301, 303, 307
Trypan blue, 59, 214 Whitten’s culture medium, 14
TUNEL, 341, 342, 344–348 Wnt7b, 145
U X, Y, Z
Ulex europaeus lectin, 339 xylocaine, 388
umbilical vessels, 380–387, 396–397 yolk sac, 283, 382
Unimar Pipelle, 96 zona pellucida, 37
M E T H O D S I N M O L E C U L A R M E D I C I N E TM
Series Editor: John M. Walker
Placenta and Trophoblast

Volume I
Edited by
Michael J. Soares
Institute of Maternal–Fetal Biology, Division of Cancer and Developmental Biology,
Department of Pathology and Laboratory Medicine, University of Kansas Medical Center, Kansas City, KS
Joan S. Hunt
University Distinguished Professor, Vice Chancellor for Research, Department of Anatomy and Cell Biology,
Placenta research has progressed rapidly over the past several decades by taking advantage of
technical advances, such as microarray analysis, reverse transcriptase polymerase chain reaction,
protein analysis, and in situ hybridization. In Placenta and Trophoblast: Methods and Protocols,
Volumes 1 & 2, internationally recognized investigators describe cutting-edge laboratory techniques
for the study of the trophoblast and placental biology. The techniques presented range from
experimental animal models, to animal and human placental organ and cell culture systems, to
morphological, biochemical, and molecular strategies for assessing trophoblast/placental growth,
differentiation, and function. Volume 1 provides readily reproducible protocols for studying embryo–
uterine implantation, trophoblast cell development, and the organization and molecular character-
ization of the placenta. Highlights include strategies for the isolation and culture of trophoblast cells
from primates, ruminants, and rodents, and precise guidance to the molecular and cellular analysis
of the placental phenotype. A companion second volume concentrates on methods for investigating
placental function.
Comprehensive and state-of-the-art, Placenta and Trophoblast: Methods and Protocols, Volumes
1 & 2 provide researchers a firm foundation for successful cellular and molecular analysis of the
placenta and the establishment of pregnancy.
Features
• Readily reproducible methods for • New ides for investigating gene imprint-
studying the trophoblast and placental ing and gene transfer via viral vectors
biology • Step-by-step instructions to ensure
• Detailed techniques for studying successful results
embryo implantation • Tricks of the trade and notes on trouble-
• Strategies for the isolation and culture shooting and avoiding known pitfalls
of trophoblast cells
• Expert guidance on the molecular and
cellular analysis of the placental
phenotype
Methods in Molecular Medicine™ • 121

ISSN 1543–1894
Placenta and Trophoblast: Methods and Protocols, Volume I
ISBN: 1-58829-404-8 E-ISBN: 1-59259-983-4
humanapress.com

Placenta and Trophoblast - Methods and Protocols (Vol 1) - M. Soares, J. Hunt (Humana, 2006) WW PDF

Uploaded by

Document Information

Original Title

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

Placenta and Trophoblast - Methods and Protocols (Vol 1) - M. Soares, J. Hunt (Humana, 2006) WW PDF

Uploaded by

Copyright:

Available Formats

M E T H O D S I N M O L E C U L A R M E D I C I N E TM

John M. Walker, SERIES EDITOR

This publication is printed on acid-free paper. ∞

Cover design by Patricia F. Cleary.

PART II. METHODS FOR STUDYING EMBRYO IMPLANTATION

PART III. IN VITRO TROPHOBLAST AND PLACENTAL MODEL SYSTEMS

PART IV. PHENOTYPIC ANALYSIS OF THE PLACENTA

23 Characterization of the Bovine Placenta by Cytoskeleton,

PART V. MOLECULAR ANALYSIS AND GENE TRANSFER TECHNIQUES

HARUO NAKANO • Reproductive Biology and Technology Laboratory,

MAJA STAROVIC • Genes and Development Research Group, Department of

PART II. ANALYSIS OF TROPHOBLAST INVASION

PART III. ANALYSIS OF UTEROPLACENTAL IMMUNE

8 In Vitro Models for Studying Human Uterine

PART IV. ANALYSIS OF PLACENTA FUNCTION:

20 Alkaline Phosphatase Fusion Proteins as Tags for Identifying

PART V. ANALYSIS OF PLACENTA ADAPTATION TO DISEASE

Michael J. Soares and Joan S. Hunt

01_Soares_1_6_F 3 8/29/05, 11:14 AM

2. Methods for Studying Embryo Implantation and Uterine Biology

01_Soares_1_6_F 4 8/29/05, 11:14 AM

concludes with presentations of nonhuman primate models for investigating

3. In Vitro Trophoblast and Placental Model Systems

4. Phenotypic Analysis of the Placenta

01_Soares_1_6_F 5 8/29/05, 11:14 AM

development. Uteroplacental vascular beds dictate nutrient delivery and are

5. Molecular Analysis and Gene Transfer Techniques

01_Soares_1_6_F 6 8/29/05, 11:14 AM

02_Deb_7_34_F 7 8/29/05, 11:15 AM

02_Deb_7_34_F 8 8/29/05, 11:15 AM

Kaushik Deb, Jeff Reese, and Bibhash C. Paria

02_Deb_7_34_F 9 8/29/05, 11:15 AM

The following description and methodologies are intended for investigators

02_Deb_7_34_F 10 8/29/05, 11:15 AM

1.2. Breeding of Mice

02_Deb_7_34_F 11 8/29/05, 11:15 AM

02_Deb_7_34_F 12 8/29/05, 11:15 AM

zone (PDZ). This is an avascular zone of the endometrium. The secondary

2.1.2. Identification of the Stage of Estrous Cycle

2.2. Breeding and Plug Checking

2.2.2. Checking Copulatory Plug

2.3. Early Pregnancy Determination

02_Deb_7_34_F 13 8/29/05, 11:15 AM

2.3.2. Invasive Method

2.3.2.2. DETECTION OF EARLY IMPLANTATION SITES BY INTRAVENOUS DYE INJECTION

2.4. Experimental Delay in Implantation

2.4.2. Experimentally Delayed Implantation

02_Deb_7_34_F 14 8/29/05, 11:15 AM

3. BD PrecisionGlide 1-mL syringe with 27-gauge, one-half-inch needle (Becton

2.5. Artificial Decidualization

2.5.1.2. INDUCTION OF PSEUDOPREGNANCY

2.5.1.3. HORMONAL PRIMING OF OVARIECTOMIZED MICE

2.5.1.4. INDUCTION OF DECIDUALIZATION BY ARTIFICIAL MEANS

02_Deb_7_34_F 15 8/29/05, 11:15 AM

8. BD PrecisionGlide 1-mL syringe with 27-gauge, one-half-inch needle (Becton

2.6. Intrauterine Blastocyst Transfer

2.7. Commonly Used Anesthetics

2.8. Common Injection Techniques

02_Deb_7_34_F 16 8/29/05, 11:15 AM

3.2. Breeding and Plug Checking

02_Deb_7_34_F 17 8/29/05, 11:15 AM

3.2.2. Checking Copulatory Plugs

3.3. Detection of Early Pregnancy