Dopamine - Methods and Protocols

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

http://www.springer.com/series/7651
Dopamine
Methods and Protocols
Edited by
Nadine Kabbani
Department of Molecular Neuroscience, Krasnow Institute for Advanced Study,
George Mason University, Fairfax, VA, USA
Editor
Nadine Kabbani
Department of Molecular Neuroscience
Krasnow Institute for Advanced Study
George Mason University
Fairfax, VA, USA
ISSN 1064-3745 ISSN 1940-6029 (electronic)

ISBN 978-1-62703-250-6 ISBN 978-1-62703-251-3 (eBook)
DOI 10.1007/978-1-62703-251-3
Springer New York Heidelberg Dordrecht London
Library of Congress Control Number: 2012950334
© Springer Science+Business Media, LLC 2013

This work is subject to copyright. All rights are reserved by the Publisher, whether the whole or part of the material
is concerned, specifically the rights of translation, reprinting, reuse of illustrations, recitation, broadcasting,
reproduction on microfilms or in any other physical way, and transmission or information storage and retrieval,
electronic adaptation, computer software, or by similar or dissimilar methodology now known or hereafter developed.
Exempted from this legal reservation are brief excerpts in connection with reviews or scholarly analysis or material
supplied specifically for the purpose of being entered and executed on a computer system, for exclusive use by the
purchaser of the work. Duplication of this publication or parts thereof is permitted only under the provisions of the
Copyright Law of the Publisher’s location, in its current version, and permission for use must always be obtained
from Springer. Permissions for use may be obtained through RightsLink at the Copyright Clearance Center. Violations
are liable to prosecution under the respective Copyright Law.
The use of general descriptive names, registered names, trademarks, service marks, etc. in this publication does not
imply, even in the absence of a specific statement, that such names are exempt from the relevant protective laws and
regulations and therefore free for general use.
While the advice and information in this book are believed to be true and accurate at the date of publication, neither
the authors nor the editors nor the publisher can accept any legal responsibility for any errors or omissions that may
be made. The publisher makes no warranty, express or implied, with respect to the material contained herein.
Cover illustration: The image depicts a whole mount adult Drosophila brain triple-labeled with rabbit anti-GFP anti-
body (green), mouse anti-FasII (1D4) antibody (red) and DAPI (blue) which mark the dopaminergic neurons
(revealed by genetically labeling with ple-GAL4,UAS-mCD8::GFP), axon tracts of the mushroom bodies and the
central complex, and all brain cell nuclei, respectively. Note in the merged image the projections of dopamine
neurons to areas of the mushroom body, the Drosophila center for learning and memory, and to the central complex,
which contributes to the regulation of locomotion. (See Chapter 13.)
Printed on acid-free paper
Humana Press is a brand of Springer

Springer is part of Springer Science+Business Media (www.springer.com)
Preface
Dopamine, a catecholamine transmitter, plays a number of important physiological roles in
the brain and body. Clues to dopamine’s role in motivation and learning have come from
over 50 years of studies in laboratory animals, which have included rodents and nonhuman
primates. In more recent years, studies on the role of dopamine in disease have opened new
avenues of research and discovery. Genetic cloning has further enabled studies of dopamine
in other species such as Drosophila melanogaster and Danio rerio. This edition of Methods in
Molecular Biology brings together and provides detailed protocols on leading approaches in
the study of dopamine within biological systems.
In the brain, dopamine functions as a key neurotransmitter in regions such as the cortex
and striatum. Dopamine is also an important modulator of ion balance in the kidney and
adaptation to light in the retina. The many effects of dopamine on physiological systems and
organs are dependent on a class of receptors, which are coupled to heterotrimeric G pro-
teins. In mammals, five dopamine receptors (D1–D5) have been identified. A fundamental
aspect of dopamine function is the localization of these receptors at the membrane, their
interaction with signaling and regulatory molecules, and their ability to assemble into higher-
order receptor oligomers (with dopamine and non-dopamine receptors) within cells.
In many species, dopamine plays a major role in reward-driven learning. Indeed, almost
every type of reward that has been studied increases dopamine transmission in the brain, and
a variety of highly addictive drugs, including stimulants such as cocaine and methamphet-
amine, act directly on the dopamine system. Several prominent diseases of the nervous sys-
tem are associated with dopamine. In particular, alterations in dopamine levels are intimately
linked with the onset and progression of Parkinson’s disease, which results from the death of
dopaminergic neurons within the substantia nigra. Schizophrenia, a disease of multiple genes
and origins, has long been linked to dopamine imbalances within the striatum and cortex
with the majority of classical antipsychotic drugs acting as antagonists at D2 receptors and
many newer generation antipsychotics maintaining an effect on D4 receptors.
This book is of interest to a range of scientists including cellular and molecular biolo-
gists, electrophysiologists, and pharmacologists. The chapters are intended for students and
experts alike and for anyone interested in exploring the vast field of dopamine research. The
book is divided into four parts based on methods: cellular/biochemical, imaging, genetics,
and electrophysiology. Presented are chapters with step-by-step, clear, and precise instruc-
tions for various research procedures. This includes protocols for bioluminescence and
fluorescence imaging, receptor immunoprecipitation and proteomic analysis, creation and
characterization of a mouse model of Parkinson’s disease, real-time measurement of dop-
amine in the brain, and modeling signal transduction in silico.
This volume is the product of contributions from experts and key figures within the
field. I would like to thank the authors for their outstanding work and cooperation during
the preparation of the volume. Specifically, I would like to thank the series editor, Professor
John Walker, for his support during the assembly of this book.
Nadine Kabbani, Ph.D.
v
Contents
Preface . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . v
Contributors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ix
PART I BIOCHEMICAL, PROTEOMIC, AND COMPUTATIONAL TOOLS
1 Detection of Cell Surface Dopamine Receptors . . . . . . . . . . . . . . . . . . . . . . . 3

Jiping Xiao and Clare Bergson
2 Methods for the Study of Dopamine Receptors
Within Lipid Rafts of Kidney Cells . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15
Peiying Yu, Van Anthony Villar, and Pedro A. Jose
3 Methods of Dopamine Research in Retina Cells . . . . . . . . . . . . . . . . . . . . . . . . . 25
Ana Lucia Marques Ventura, Fernando Garcia de Mello,
and Ricardo Augusto de Melo Reis
4 Capture of D2 Dopamine Receptor Signaling Complexes
in Striatal Cells for Mass Spectrometry Proteomic Analysis . . . . . . . . . . . . . . . . . 43
Nadine Kabbani and Jacob C. Nordman
5 Modeling Spatial Aspects of Intracellular Dopamine Signaling . . . . . . . . . . . . 61
Kim T. Blackwell, Lane J. Wallace, BoHung Kim, Rodrigo F. Oliveira,
and Wonryull Koh
PART II CELLULAR IMAGING
6 A Biophysical Approach for the Study of Dopamine

Receptor Oligomerization. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 79
Sylwia Lukasiewicz, Agata Faron-Górecka,
and Marta Dziedzicka-Wasylewska
7 Detection of Receptor Heteromers Involving Dopamine Receptors
by the Sequential BRET-FRET Technology . . . . . . . . . . . . . . . . . . . . . . . . . . . . 95
Gemma Navarro, Peter J. McCormick, Josefa Mallol, Carme Lluís,
Rafael Franco, Antoni Cortés, Vicent Casadó, Enric I. Canela,
and Sergi Ferré
8 BRET Approaches to Characterize Dopamine and TAAR1
Receptor Pharmacology and Signaling . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 107
Stefano Espinoza, Bernard Masri, Ali Salahpour,
and Raul R. Gainetdinov
9 Dopaminergic Regulation of Dendritic Calcium:
Fast Multisite Calcium Imaging . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 123
Wen-Liang Zhou, Katerina D. Oikonomou, Shaina M. Short,
and Srdjan D. Antic
vii
viii Contents
PART III GENETIC MANIPULATION IN CELLS AND ORGANISMS
10 Functional Analysis of Human D1 and D5 Dopaminergic

G Protein-Coupled Receptors: Lessons from Mutagenesis
of a Conserved Serine Residue in the Cytosolic End
of Transmembrane Region 6 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 141
Bianca Plouffe and Mario Tiberi
11 A Molecular Genetic Approach to Uncovering the Differential
Functions of Dopamine D2 Receptor Isoforms . . . . . . . . . . . . . . . . . . . . . . . 181
Yanyan Wang, Toshikuni Sasaoka, and Mai T. Dang
12 Genomic Strategies for the Identification of Dopamine
Receptor Genes in Zebrafish . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 201
Wendy Boehmler, Jessica Petko, Victor A. Canfield,
and Robert Levenson
13 Application of Cell-Specific Isolation to the Study
of Dopamine Signaling in Drosophila . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 215
Eswar Prasad R. Iyer, Srividya Chandramouli Iyer,
and Daniel N. Cox
PART IV ELECTROCHEMICAL, PHYSIOLOGICAL, AND BEHAVIORAL ANALYSIS
14 Regulation of Dopamine Transporter Expression

by Neuronal Activity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 229
Shalini Padmanabhan, Thach Pham, and Balakrishna M. Prasad
15 Monitoring Axonal and Somatodendritic Dopamine Release
Using Fast-Scan Cyclic Voltammetry in Brain Slices . . . . . . . . . . . . . . . . . . . . . . 243
Jyoti C. Patel and Margaret E. Rice
16 Real-Time Chemical Measurements of Dopamine
Release in the Brain . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 275
James G. Roberts, Leyda Z. Lugo-Morales, Philip L. Loziuk,
and Leslie A. Sombers
17 The MPTP/Probenecid Model of Progressive Parkinson’s Disease . . . . . . . . . 295
Anna R. Carta, Ezio Carboni, and Saturnino Spiga
Index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 309
Contributors
SRDJAN D. ANTIC • Department of Neuroscience, University of Connecticut
Health Center, Farmington, CT, USA
RICARDO AUGUSTO DE MELO REIS • Laboratory of Neurochemistry,
Program in Neurobiology IBCCF, UFRJ, Rio de Janeiro, Brazil
KIM T. BLACKWELL • The Krasnow Institute for Advanced Study, George Mason
University, Fairfax, VA, USA
CLARE BERGSON • Department of Pharmacology and Toxicology, Georgia Health
Sciences University, Augusta, GA, USA
WENDY BOEHMLER • Department of Biological Sciences, York College of
Pennsylvania,
York, PA, USA
ENRIC I. CANELA • Department of Biochemistry and Molecular Biology, Faculty of
Biology, University of Barcelona, Barcelona, Spain
VICTOR A. CANFIELD • Department of Pharmacology, Penn State College of
Medicine, Hershey, PA, USA
EZIO CARBONI • Department of Biomedical Sciences, University of Cagliari,
Cagliari, Italy
ANNA R. CARTA • Department of Biomedical Sciences, University of Cagliari,
Cagliari, Italy
VICENT CASADÓ • Department of Biochemistry and Molecular Biology, Faculty of
ANTONI CORTÉS • Department of Biochemistry and Molecular Biology, Faculty of
DANIEL N. COX • School of Systems Biology, Krasnow Institute for Advanced Study,
George Mason University, Fairfax, VA, USA
MAI T. DANG • Department of Neurology, Hospital of University of Pennsylvania,
Philadelphia, PA, USA
MARTA DZIEDZICKA-WASYLEWSKA • Institute of Pharmacology, Polish Academy
of Sciences, Kraków, Poland; Faculty of Biochemistry, Biophysics and Biotechnology,
Jagiellonian University, Krakow, Poland
STEFANO ESPINOZA • Department of Neuroscience and Brain Technologies,
Italian Institute of Technology, Genoa, Italy
AGATA FARON-GÓRECKA • Institute of Pharmacology, Polish Academy of Sciences,
Krakow, Poland
SERGI FERRÉ • Department of Health and Human Services, Intramural Research
Program, National Institute on Drug Abuse, National Institutes of Health,
Baltimore, MD, USA
RAFAEL FRANCO • Department of Biochemistry and Molecular Biology, University
of Barcelona, Barcelona, Spain
ix
x Contributors
RAUL R. GAINETDINOV • Department of Neuroscience and Brain Technologies,

Italian Institute of Technology, Genoa, Italy
FERNANDO GARCIA DE MELLO • Laboratory of Neurochemistry,
Program in Neurobiology IBCCF, UFRJ, Rio de Janeiro, Brazil
ESWAR PRASAD R. IYER • School of Systems Biology, George Mason University,
Manassas, VA, USA
SRIVIDYA CHANDRAMOULI IYER • School of Systems Biology, George Mason University,
Manassas, VA, USA
PEDRO A. JOSE • Department of Pediatrics, Center for Molecular Physiology
Research, Children’s National Medical Center, and School of Medicine, George
Washington University, Washington, DC, USA
NADINE KABBANI • Department of Molecular Neuroscience, Krasnow Institute for
Advanced Study, George Mason University, Fairfax, VA, USA
BOHUNG KIM • The Krasnow Institute for Advanced Study, George Mason
WONRYULL KOH • The Krasnow Institute for Advanced Study, George Mason
ROBERT LEVENSON • Department of Pharmacology, Penn State College of Medicine,
Hershey, PA, USA
CARME LLUÍS • Department of Biochemistry and Molecular Biology,
University of Barcelona, Barcelona, Spain
PHILIP L. LOZIUK • Department of Chemistry, North Carolina State University,
Raleigh, NC, USA
LEYDA Z. LUGO-MORALES • Department of Chemistry, North Carolina State
University, Raleigh, NC, USA
SYLWIA LUKASIEWICZ • Faculty of Biochemistry, Biophysics and Biotechnology,
Jagiellonian University, Krakow, Poland
JOSEFA MALLOL • Department of Biochemistry and Molecular Biology,
BERNARD MASRI • Cancer Research Center of Toulouse, INSERM U1037 -
Université Paul Sabatier Toulouse III, CHU Rangueil, Toulouse, France
PETER J. MCCORMICK • Department of Biochemistry and Molecular Biology,
GEMMA NAVARRO • Department of Biochemistry and Molecular Biology,
JACOB C. NORDMAN • Department of Molecular Neuroscience, Krasnow Institute
for Advanced Study, George Mason University, Fairfax, VA, USA
KATERINA D. OIKONOMOU • Department of Neuroscience, University of
Connecticut Health Center, Farmington, CT, USA
RODRIGO F. OLIVEIRA • The Krasnow Institute for Advanced Study, George Mason
SHALINI PADMANABHAN • Department of Pharmacology, Medical College of
Georgia, Augusta, GA, USA
JYOTI C. PATEL • Departments of Neurosurgery and Physiology & Neuroscience,
New York University School of Medicine, New York, NY, USA
Contributors xi
JESSICA PETKO • Department of Pharmacology, Penn State College of Medicine,

Hershey, PA, USA
THACH PHAM • General Surgery, Dwight D. Eisenhower Army Medical Center,
Fort Gordon, GA, USA
BIANCA PLOUFFE • Departments of Medicine/Cellular and Molecular Medicine/
Psychiatry, Ottawa Hospital Research Institute (Neuroscience Program),
University of Ottawa, Ottawa, ON, Canada
BALAKRISHNA M. PRASAD • Department of Pharmacology, Medical College of
Georgia, Augusta, GA, USA;Clinical Investigation, Dwight D. Eisenhower Army
Medical Center, Fort Gordon, GA, USA
MARGARET E. RICE • Departments of Neurosurgery and Physiology & Neuroscience,
New York University School of Medicine, New York, NY, USA
JAMES G. ROBERTS • Department of Chemistry, North Carolina State University,
Raleigh, NC, USA
ALI SALAHPOUR • Department of Pharmacology and Toxicology, University of
Toronto, Toronto, ON, Canada
TOSHIKUNI SASAOKA • Department of Laboratory Animal Science, Kitasato
University School of Medicine, Kanagawa, Japan
SHAINA M. SHORT • Department of Neuroscience, University of Connecticut Health
Center, Farmington, CT, USA
LESLIE A. SOMBERS • Department of Chemistry, North Carolina State University,
Raleigh, NC, USA
SATURNINO SPIGA • Department of Life and Environmental Sciences,
University of Cagliari, Cagliari, Italy
MARIO TIBERI • Departments of Medicine/Cellular and Molecular Medicine/
Psychiatry, Ottawa Hospital Research Institute (Neuroscience Program),
University of Ottawa, Ottawa, ON, Canada
ANA LUCIA MARQUES VENTURA • Department of Neurobiology, Program in
Neurosciences, Universidade Federal Fluminense, Niterói, Brazil
VAN ANTHONY VILLAR • Department of Pediatrics, Center for Molecular Physiology
Research, Children’s National Medical Center, and School of Medicine, George
Washington University, Washington, DC, USA
LANE J. WALLACE • College of Pharmacy, Ohio State University, Columbus, OH,
USA
YANYAN WANG • Department of Pharmacology, College of Medicine, Beckman
Institute for Advanced Science and Technology, University of Illinois at Urbana-
Champaign, Urbana, IL, USA
JIPING XIAO • Cardiovascular Institute, University of Pennsylvania,
Philadelphia, PA, USA
PEIYING YU • Department of Pediatrics, Center for Molecular Physiology Research,
Children’s National Medical Center, School of Medicine, George Washington
University, Washington, DC, USA
WEN-LIANG ZHOU • Department of Neuroscience, University of Connecticut
Health Center, Farmington, CT, USA
Part I
Biochemical, Proteomic, and Computational Tools

Chapter 1
Detection of Cell Surface Dopamine Receptors

Jiping Xiao and Clare Bergson
Abstract
Dopamine receptors are a class of metabotropic G protein-coupled receptors. Plasma membrane expression
is a key determinant of receptor signaling, and one that is regulated both by extra and intracellular cues.
Abnormal dopamine receptor signaling is implicated in several neuropsychiatric disorders, including
schizophrenia and attention deficit hyperactivity disorder, as well as drug abuse. Here, we describe in detail
the application of two complementary applications of protein biotinylation and enzyme-linked immuno-
absorbent assay (ELISA) for detecting and quantifying levels of dopamine receptors expressed on the cell
surface. In the biotinylation method, cell surface receptors are labeled with Sulfo-NHS-biotin. The charge
on the sulfonyl facilitates water solubility of the reactive biotin compound and prevents its diffusion across
the plasma membrane. In the ELISA method, surface labeling is achieved with antibodies specific to extra-
cellular epitopes on the receptors, and by fixing the cells without detergent such that the plasma membrane
remains intact.
Key words: Schizophrenia, ADHD, DAPI, Biotinylation, ELISA, Plasma membrane
1. Introduction
Dopamine (DA) regulates movement, endocrine function, reward

behavior, and memory processes by stimulating a family of five
subtypes of G protein-coupled receptors (GPCRs) designated the
D1 to D5 receptors (D1R-D5R). Disorders involving DA trans-
mission include Parkinson’s disease (PD), as well as a number of
neuropsychiatric illnesses including attention deficit hyperactivity
disorder (ADHD) and schizophrenia. Several lines of evidence
suggest that plasma membrane levels of D1Rs, in particular, are
critically linked to working memory, an executive function impaired
in schizophrenia (1–5). Since deficits in working memory and
related executive functions are currently treatment-resistant,
reagents which manipulate D1R cell surface expression could rep-
resent an effective therapeutic strategy.
Nadine Kabbani (ed.), Dopamine: Methods and Protocols, Methods in Molecular Biology, vol. 964,
DOI 10.1007/978-1-62703-251-3_1, © Springer Science+Business Media, LLC 2013
3
4 J. Xiao and C. Bergson
A number of factors have been discovered in the past 15 years

or so which can regulate surface levels of D1Rs. For example, both
hyper-and hypo-dopaminergic states produce alterations in surface
D1Rs in vivo (6), and similar effects are observed in cells in culture
with D1R agonists and antagonists (7, 8). Further, activation of
glutamatergic N-methyl-D-aspartic acid (NMDA) receptors in
neurons stimulates accumulation of D1Rs on synaptic membranes
(9). This effect is regulated by physical interaction of NR1 NMDA
receptor subunits with D1Rs (10). In addition, a variety of other
mechanisms regulate D1R surface levels including endocytic recy-
cling (11), receptor phosphorylation (12–14), as well as physical
association with cytoskeletal proteins (15).
Biotinylation and enzyme-linked immunoabsorbent assay
(ELISA) offer a number of advantages for detecting and quantify-
ing cell surface receptors. With either method, it is possible avoid
the use of radioisotopes as is typically required in receptor ligand
binding assays. Both methods are inherently quantitative. While
immunofluorescent detection of DA receptor subtypes is also
straightforward, quantification of surface levels by this method is
not. The isolation of receptors on the cell surface devoid of con-
tamination from other membrane compartments is troublesome
with subcellular fractionation methods involving gradient centrifu-
gation. However, the tools currently available for cell surface
ELISA and biotinylation permit unambiguous assessment of recep-
tors residing specifically on the plasma membrane.
We provide detailed protocols for biotinylation and ELISA
based-methods to quantify the cell surface levels of DA receptors
under basal conditions and agonist stimulation. We use D1Rs to
illustrate application of these approaches. However, these tools can
be easily adapted for other DA receptor subtypes. In the biotinyla-
tion method, cell surface receptors are labeled with non-cleavable
Sulfo-NHS-biotin. At neutral pH, the sulfo-NHS ester reacts rap-
idly with any primary amine-containing protein such that the bio-
tin label is attached via a stable amide bond. As the sulfonyl group
is charged, the compound shows good water-solubility, and poor
ability to cross intact plasma membranes. As a result, labeling is
restricted to the extracellular domains of proteins spanning the
plasma membrane. The sulfo-NHS-biotin compound can also be
used to study endogenous receptors in primary culture or in brain
slices (16, 17). Cleavable biotinylation reagents such as sulfo-NHS-
S-S-biotin include a disulfide group positioned such that biotin
label can be removed by treatment with reducing agents. These
compounds are useful for quantifying agonist-stimulated receptor
internalization as the biotin label can be stripped from receptors
remaining on the cell surface prior to cell lysis (18). In the cells
surface ELISA method, labeling is achieved by fixing the cells with-
out detergent such that the plasma membrane remains intact.
1 Detecting Surface DA Receptors 5
Receptors can be detected with epitope or subtype specific primary

antibodies, followed by enzyme-linked secondary antibodies, and
exposure to chromogenic substrates.
2. Materials
2.1. Biotinylation 1. HEK293 cells.

of Cell Surface DA 2. FLAG-D1R cells: this is an HEK293 cell line which stably
Receptors expresses human D1Rs carrying a FLAG epitope tag inserted
at the N-terminus of the receptor coding sequence.
3. HEK293 culture medium: Dulbecco’s Modified Eagle’s
Medium (DMEM) (Sigma-Aldrich, St. Louis, MO) supple-
mented with 10% fetal bovine serum (FBS) (Sigma-Aldrich), 1%
penicillin-streptomycin (Roche Diagnostics, Indianapolis, IN).
4. FLAG-D1R stable cell line medium: Dulbecco’s Modified
Eagle’s Medium (DMEM) supplemented with 10% fetal bovine
serum (FBS), 1% penicillin-streptomycin, 450 μg/mL G418
(Invitrogen Life Technologies, Grand Island, NY).
5. PBS: 8.5 mM sodium phosphate, 1.5 mM potassium phos-
phate, 137 mM NaCl, pH 7.4.
6. Non-cleavable sulfo-NHS-Biotin (Pierce, Thermo Fisher
Scientific, Rockford, IL).
7. 10 mM glycine in PBS.
8. Lysis buffer: 150 mM NaCl, 20 mM Tris–HCl, pH 7.5, 0.5%
NP-40, 10% glycerol containing protease inhibitor cocktail
(1 tablet/10 mL).
9. Protease inhibitor cocktail (Roche Diagnostics).
10. Sonic dismembrator (Fisher Scientific).
11. Streptavidin slurry (Pierce Biotechnology).
12. 1.5 M guanidine HCl.
13. 1× SDS loading buffer: 63 mM Tris–HCl, 10% glycerol, 2%
SDS, 0.0025% Bromophenol Blue (Sigma), pH 6.8.
14. 7.5% SDS-PAGE gels.
15. Electrophoresis power supply, SDS-PAGE and protein gel
transfer equipment (Bio-Rad, Hercules, CA).
16. 10× transfer buffer: 0.25 M Tris base, 2 M glycine. Dilute with
double distilled H2O and add methanol to 20% for use.
17. PVDF membrane (Protran, GE Healthcare, Waukesha, WI).
18. Whatman 3 M paper.
19. TBS-T buffer: 250 mM Tris–HCl, pH 7.5, 1.5 M NaCl, 1%

(v/v) Tween-20. Dilute from 10× TBS stock with ddH2O and
add Tween-20.
20. Blocking buffer: 5% (w/v) nonfat dry milk in TBS-T.
21. Anti-FLAG M2 monoclonal antibody (mab).
22. Goat anti mouse-HRP antibody (Jackson ImmunoResearch,
West Grove, PA).
23. ECL plus detection kit (Amersham, GE Healthcare, Waukesha,
WI).
24. Kodak X-ray film (Kodak, Rochester, NY).
2.2. Detection of Cell 1. 24-well tissue culture plates.

Surface DA Receptors 2. 1 μg/mL of laminin: (BD bioscience, San Jose, CA).
by ELISA
3. 4% paraformaldehyde solution in PBS.
4. Non-permeabilizing blocking buffer: Tris-buffered saline
(TBS) containing 5% nonfat dry milk and 5% normal goat
serum.
5. Anti-FLAG monoclonal antibody M2 (Sigma-Aldrich).
6. Goat anti mouse antibody conjugated with horseradish peroxi-
dase (HRP) (Jackson ImmunoResearch).
7. Tetramethylbenzidine (TMB) substrate (Pierce).
8. Stop buffer: 2 M H2SO4.
9. 20 nM DAPI solution (Invitrogen).
10. Standard fluorescence and absorbance multi-well plate reader
or spectrophotometer.
3. Methods
3.1. Biotinylation 1. Culture 1 × 106 FLAG-D1R cells and untransfected HEK293

of Cell Surface DA cells in the appropriate medium in 10 cm dishes for 2 days at
Receptors 37°C in a humidified 5% CO2 incubator (see Note 1).
2. Place dishes on ice. All of the following steps are performed on
ice except if noted otherwise.
3. Wash cells three times briefly with 10 mL ice-cold PBS (see
Note 2).
4. Add 5 mL 0.5 mg/mL cell-impermeable, non-cleavable sulfo-
NHS-Biotin in PBS to cells and incubate for 30 min on ice (see
Note 3).
5. Wash cells three times with 10 mL glycine (10 mM) in PBS to
quench the unbound biotin reagent (see Note 4).
6. After the last wash, wash cells once with ice-cold PBS.
7. Harvest cells in 1.5 mL of ice-cold PBS using a cell scraper,
and collect cells by spinning the suspension at 800 × g for 5 min
in a microcentrifuge.
8. Retain the pellet, and resuspend cells in 500 μL of lysis
buffer.
9. Sonicate cells for 10 s on ice (see Note 5).
10. Incubate cells on ice for 30 min.
11. Centrifuge samples at 18,000 × g for 30 min at 4°C.
12. Keep the supernatant.
13. Add the Streptavidin slurry (100 μL) to the supernatant, and
mix by end-over-end rotation for 2 h at 4°C.
14. Pellet the biotinylated protein bound streptavidin resin by cen-
trifugation at 10,500 × g for 2 min (see Note 6).
15. Retain and wash the resin twice with lysis buffer, twice with
1.5 mL guanidine HCl, followed by two additional washes
with lysis buffer (see Note 7).
16. Elute the bound proteins with 50 μL SDS-PAGE loading buf-
fer by boiling the beads for 3–5 min at 100°C (see Note 8).
17. Load 20 μL of the samples and separate on a 7.5% SDS-PAGE
gel by gel electrophoresis (see Note 9).
18. Transfer proteins to PVDF membrane in a gel transfer appara-
tus (see Note 10).
19. After transfer, wash the membrane three times (5 min each)
with TBS-T on a rocking platform.
20. Incubate the membrane with 5% milk in TBS-T for 1–3 h at
RT on rocking platform.
21. Discard block and add anti-FLAG M2 mab diluted 1:2,000 in
blocking solution. Incubate the membrane at 4°C overnight
(or for 2 h at RT) on a nutator (see Note 11).
22. Discard the primary antibody; wash the membrane three times
(10 min each) with TBS-T at RT on rocking platform.
23. Incubate with the secondary antibody, rabbit anti mouse-HRP
antibody (1:10,000) for 1 h at RT on a nutator or rocking
platform.
24. Discard the secondary antibody and wash the membrane three
times (10 min each) with TBS-T.
25. After the final wash, add 1 mL ECL reagent to cover the mem-
brane, incubate for 1 min at RT. Wrap the membrane with saran
wrap sheet, place in a developing cassette, and expose to X-ray
film for a suitable time (typically, from 10 s to several minutes).
Develop film in dark room (see Note 12) (Fig. 1).
Fig. 1. Detection of cell surface D1Rs by biotinylation. 48 h after plating untransfected

HEK293 cells and FLAG-D1R cells expressing human D1Rs, cell surface proteins were
labeled with sulfo-NHS-biotin. Biotinylated proteins were recovered by streptavidin resin.
Cell surface Flag-D1Rs were detected by immunoblotting biotinylated proteins with HRP
conjugated FLAG M2 antibody.
3.2. Detection of Cell 1. Coat a 24-well plate by incubating plate with 1 μg/mL of
Surface DA Receptors laminin at least 2 h (see Note 13).
by ELISA 2. Discard laminin solution and leave the plate in the hood for
30 min to dry.
3. Plate FLAG-D1R and untransfected HEK293 cells at
2 × 104 cells/well and culture cells for 2 days.
4. Wash the cells three times briefly with PBS; and then add 4%
paraformaldehyde and incubate for 20 min at RT to fix cells
(see Note 14).
5. Discard 4% paraformaldehyde and wash cells three times (5 min
each) with PBS.
6. Block cells under non-permeabilizing conditions (PBS con-
taining 5% nonfat dry milk, and 5% normal goat serum) for 1 h
at RT (see Notes 15 and 16).
7. Discard blocking buffer and incubate cells with mouse anti-
FLAG M2 mab (1:250) in blocking buffer (PBS containing 5%
nonfat dry milk, and 5% normal goat serum) for 2 h at RT (see
Note 17).
8. Discard the primary antibody, and wash cells with PBS on a
rocking platform (see Note 18).
9. Incubate cells with the HRP-conjugate secondary antibody
(1:5,000) in blocking buffer for 1 h at RT (see Note 19).
Fig. 2. Determination of cell density using DAPI. HEK293 cells were plated at varying
densities in a 24-well plate. After washing with PBS four times, 100 μL of DAPI (300 nM
in PBS) was added to each well, and the plate was incubated for 5 min at RT. After the
incubation, wells were rinsed several times with PBS. Samples were excited in 358 nm
and the emission at 461 nm was recorded.
10. Discard the secondary antibody, and wash cells four times
(10 min each) with PBS on a rocking platform (see Note 20).
11. Add 500 μL of TMB to each well, and incubate the plate for
15 min at RT (see Note 21).
12. Stop the reaction by adding 50 μL H2SO4 (2 M). The color
will turn from blue into yellow.
13. Transfer 400 μL of the solution and measure the OD at 450 nm
(see Note 22).
14. After HRP detection, add 100 μL DAPI (300 nM) for 5 min.
Measure the DAPI fluorescence by exciting at 350 nm, and
detecting at 470 nm. Cell number can be inferred from a stan-
dard curve of cells plated versus DAPI intensity as shown in
Fig. 2.
4. Notes
1. Cells stably transfected with plasmids containing neomycin

resistant markers such as the FLAG-D1R cells can also be
maintained in HEK293 cell medium containing 250 μg/
mL G418. However, the purity of the cell line is better
maintained with higher concentrations of G418 (e.g.,
450 μg/mL).
2. The cells density should be about 90% confluence. Higher
confluence will decrease the efficiency of cell surface protein
biotinylation.
3. Biotinylation reagents are susceptible to hydrolysis so the

biotin compound should be prepared just prior to use. Optimal
results are obtained when the cell labeling solution is prepared
from newly opened bottles. Alternatively, a stock biotinylation
solution (100–200 mg/mL) could be prepared in DMSO, and
aliquots stored at −20°C until use.
4. This step washes out protein in the culture medium which can
be biotinylated as well as free sulfo-NHS-Biotin. Wash the cells
gently since the plates are not coated with laminin, and
HEK293 cells detach easily.
5. Keep the sonication probe moving slowly in the solution to
avoid local fluxes in temperature, while keeping it submerged
to avoid foaming.
6. Carefully pipette off supernatant. Alternatively, use a spin filter
to retain the streptavidin resin in the upper reservoir.
7. The guanidine HCl wash helps reduce nonspecific binding to
streptavidin. This step does not affect recovery of the avidin–
biotin complexes as the high (10−15 M) affinity of avidin and
biotin renders them fairly insensitive to extremes of pH, deter-
gent, solvents, and temperature.
8. Boiling the samples is necessary to disrupt the non-covalent
association of biotinylated proteins with the streptavidin
beads.
9. Run mini-gels at 100 V for 10 min through stacking portion,
and at 200 V for 40 min through the separating region of the
gels.
10. Transfer can be carried out at 100 V for 1 h, or overnight at
25 V, both at 4°C.
11. The FLAG M2 mab is used to specifically detect D1Rs tagged
with the FLAG epitope among all the biotinylated cell surface
proteins eluted from the streptavidin slurry. If the DA receptor
is not tagged, an alternative approach would be to immuno-
precipitate with polyclonal receptor specific antibodies, and
probe blots of material subsequently eluted from protein A/G
resin with streptavidin-conjugated HRP. Resin washing condi-
tions would need to be adjusted accordingly.
12. Biotinylation efficiency will vary from protein to protein. If
labeling efficiency seems low as gauged from the intensity of
bands in streptavidin recovered lanes versus lysate lanes, con-
sider performing a second round of biotinylation before lysing
the cells. Alternatively, increase the pH of the biotinylation
solution (pH 8–9) to improve labeling by increasing the pro-
portion of lysine ε-amino groups conjugated (19).
13. The laminin coating step helps decrease cell loss as the ELISA
protocol involves several steps with extensive washes.
14. This step can also be carried out following treatment with
agonists such as shown in Fig. 3a. Agonist-induced receptor
internalization can be inferred from the ratio of receptor surface
levels detected before and after agonist treatment (Fig. 3b).
15. This condition assures that anti-FLAG antibodies only bind
the D1R on the cell surface. Nonspecific binding of antibody
to either the plates or the cells will increase background signal.
Blocking buffer composition and volume or blocking time
might need to be adjusted to reduce background noise. We
also suggest plating HEK293 cells which do not express FLAG-
D1Rs. Additionally, include FLAG-D1R negative control wells
where the primary antibody is omitted and only the HRP-
secondary is added; or where FLAG mab is followed by uncon-
jugated secondary antibodies. Negligible TMB signals coming
from such negative control samples are necessary to validate
the results from the experimental samples.
16. An alternative strategy could be the use of D1R subtype selec-
tive antibodies directed at an extracellular epitope.
a
15
Vehicle
(a.u./cell number)
SKF81297
10
HRP
0
HEK293 FLAG-D1R
b
75
Internaliztion Ratio
50
(%)
25
–25
HEK293 FLAG-D1R
Fig. 3. Cells surface D1Rs measured 15 min after addition of D1R agonist SKF81297
(10 nM) or vehicle using the ELISA method. (a) Surface D1Rs by cell surface ELISA assay.
Cells were fixed under non-permeabilizing conditions, and cell surface D1Rs detected
with anti-FLAG and HRP conjugated secondary antibodies, followed by ELISA using the
TMB substrate for HRP. Cell numbers were determined by DAPI staining. (b) The “endocy-
tosis ratio” was determined by the (surface D1Rs in vehicle treated cells- surface D1Rs
treated with SKF for 15 min/surface D1Rs in vehicle treated cells). The bar graphs show
the mean ± SEM of three independent experiments each including four replicates per
group.
17. Be cautious, no detergent!

18. If background signal is high (e.g., non-transfected HEK293
and FLAG-D1R cells give equivalent signals), wash cells six
times, or decrease the incubation time with the anti-FLAG
mab from 2 h to 1 h.
19. Be cautious! No detergent!
20. If background is high, wash cells six times.
21. Blue color should appear after 15 min. TMB is a chromogenic
HRP substrate which absorbs at 450 nm. However, chemilu-
minescent and fluorescent HRP substrates are also available.
22. Remember to keep the plate for cell number counting.
References
1. Vijayraghavan S, Wang M, Birnbaum SG, 8. Brismar H, Asghar M, Carey RM, Greengard P,
Williams GV, Arnsten AFT (2007) Inverted-U Aperia A (1998) Dopamine-induced recruit-
dopamine D1 receptor actions on prefrontal ment of dopamine D1 receptors to the plasma
neurons engaged in working memory. Nat membrane. Proc Natl Acad Sci U S A
Neurosci 10:376–384 95:5573–5578
2. Zahry J, Taylor JR, Mathew RG, Arnsten AF 9. Scott L, Kruse MS, Forssberg H, Brismar H,
(1997) Supranormal stimulation of D1 dop- Greengard P, Aperia A (2002) Selective up-
amine receptors in the rodent prefrontal cortex regulation of dopamine D1 receptors in den-
impairs spatial working memory performance. dritic spines by NMDA receptor activation.
J Neurosci 17:8528–8535 Proc Natl Acad Sci U S A 99:1661–1664
3. McNab F, Varrone A, Farde L, Jucaite A, 10. Pei L, Lee FJS, Moszczynska A, Vukusic B, Liu
Bystritsky P, Forssberg H, Klingberg T (2009) F (2004) Regulation of dopamine D1 receptor
Changes in cortical dopamine D1 receptor function by physical interaction with the
binding associated with cognitive training. NMDA receptors. J Neurosci 24:1149–1158
Science 323:800–802 11. Vargas GA, Von Zastrow M (2004)
4. Abi-Dargham A, Mawlawi O, Lombardo I, Gil Identification of a novel endocytic recycling
R, Martinez D, Huang Y, Hwang DR, Keilp J, signal in the D1 dopamine receptor. J Biol
Kochan L, Van Heertum R, Gorman JM, Chem 279:37461–37469
Laruelle M (2002) Prefrontal dopamine D1 12. Yu P, Asico LD, Luo Y, Andrews P, Eisner GM,
receptors and working memory in schizophre- Hopfer U, Felder RA, Jose PA (2006) D1 dop-
nia. J Neurosci 22:3708–3719 amine receptor hyperphosphorylation in renal
5. Castner SA, Williams GV, Goldman-Rakic PS proximal tubules in hypertension. Kidney Int
(2000) Reversal of anti-psychotic-induced 70:1072–1079
working memory deficits by short-term dop- 13. Kim OJ, Gardner BR, Williams DB, Marinec
amine D1 receptor stimulation. Science PS, Cabrera DM, Peters JD, Mak CC, Kim
287:2020–2022 KM, Sibley DR (2004) The role of phosphory-
6. Dumartin B, Jaber M, Gonon F, Caron MG, lation in D1 dopamine receptor desensitiza-
Giros B, Bloch B (2000) Dopamine tone regu- tion: evidence for a novel mechanism of arrestin
lates D1 receptor trafficking and delivery in association. J Biol Chem 279:7999–8010
striatal neurons in dopamine transporter- 14. Lamey M, Thompson M, Varghese G, Chi H,
deficient mice. Proc Natl Acad Sci U S A Sawzdargo M, George SR, O’Dowd BF (2002)
97:1879–1884 Distinct residues in the carboxyl tail mediate
7. Martin-Negrier M, Charron G, Bloch B (2000) agonist-induced desensitization and internal-
Agonist stimulation provokes dendritic and ization of the human dopamine D1 receptor.
axonal dopamine D(1) receptor redistribution J Biol Chem 277:9415–9421
in primary cultures of striatal neurons. 15. Kim OJ, Ariano MA, Lazzarini RA, Levine MS,
Neuroscience 99:257–266 Sibley DR (2002) Neurofilament-M interacts
with the D1 dopamine receptor to regulate cell receptor erases memory trace. J Neurosci
surface expression and desensitization. 26:8892–8899
J Neurosci 22:5920–5930 18. Ali MK, Bergson C (2003) Elevated intracel-
16. Holman D, Henley JM (2007) A novel method lular calcium triggers recruitment of the recep-
for monitoring the cell surface expression of tor cross-talk accessory protein calcyon to the
heteromeric protein complexes in dispersed plasma membrane. J Biol Chem 278:
neurons and acute hippocampal slices. 51654–51663
J Neurosci Methods 160:302–308 19. Gottardi CJ, Dunbar LA, Caplan MJ (1995)
17. Mao SC, Hsiao YH, Gean PW (2006) Biotinylation and assessment of membrane
Extinction training in conjunction with a par- polarity: caveats and methodological concerns.
tial agonist of the glycine site on the NMDA Am J Physiol 268:F285–F295
Chapter 2
Methods for the Study of Dopamine Receptors

Within Lipid Rafts of Kidney Cells
Peiying Yu, Van Anthony Villar, and Pedro A. Jose
Abstract
There is increasing evidence that G protein-coupled receptor (GPCR) signaling is regulated in lipid raft
microdomains. GPCRs and GPCR-signaling molecules, including G proteins and protein kinases, have
been reported to compartmentalize in these microdomains. Dopamine D1-like receptors (D1R and D5R)
belong to a family of GPCRs that are important in the regulation of renal function. These receptors are
not only localized and regulated in caveolae that contains caveolin-1 but are also distributed in non-
caveolar lipid rafts which do not contain caveolin-1. This chapter describes detergent- and non-detergent-
based methods to obtain lipid raft fractions from renal proximal tubule cells.
Key words: Lipid rafts, Caveolae, Membrane microdomains, Dopamine receptor
1. Introduction
Dopamine receptors belong to the α group of the rhodopsin-like

family of G protein-coupled receptors (GPCRs) and are classified
into two subfamilies depending on their effect on adenylyl cyclase
activity. The D1-like receptors (D1R and D5R) stimulate while the
D2-like receptors (D2R, D3R and D4R) inhibit adenylyl cyclase
activity (1). Dopamine D1-like receptors have been implicated in
the modulation of various neural processes, including learning,
memory, reward, and motor activity (2, 3), and in the regulation
of blood pressure by actions on the adrenergic nervous system,
hormone secretion, and epithelial ion transport (1).
Lipid rafts are membrane microdomains composed of choles-
terol, sphingolipids, glycosylphosphatidylinositol-linked (GPI-
linked) proteins, and other proteins such as caveolin (4–6).
15
16 P. Yu et al.
Caveolae and lipid rafts have been implicated to play a role in

cellular processes like membrane sorting, receptor trafficking, sig-
nal transduction, and cell adhesion. Lipid rafts serve as signaling
platforms for several signaling molecules such as G protein sub-
units, enzymes, and adaptor proteins that play important roles in
signal transduction in a variety of mammalian cells (4–7). Lipid
rafts are characterized by their relative insolubility in nonionic
detergents at 4°C and light buoyant density on sucrose gradient
(4, 7). Among the lipid rafts, caveolae are the best characterized,
are localized on cell surface invaginations, and are formed by
polymerization of caveolin proteins with cholesterol (5–9). Three
caveolin genes encode the caveolin proteins, namely, caveolin-1,
caveolin-2, and caveolin-3. Caveolin-1 has been used as a marker
protein for caveolae (5–9). There are several other markers for lipid
rafts, such as flotillin-1, CD55, and alkaline phosphatase (10, 11).
Flotillin-1 has been used as a lipid raft marker protein in cells that
do not contain caveolae, i.e., blood cells (11), neural cells (12),
and rat kidney tubule cells (13, 14). We have reported that there
are non-caveolar lipid rafts in human embryonic kidney cells since
these cells are devoid of measurable caveolin-1 (13).
There are several ways to prepare lipid rafts using detergent or
detergent-free methods. Detergent-free methods have been devel-
oped to isolate lipid rafts (7–9). Schnitzer et al. described a deter-
gent-free method to isolate lipid rafts from rat lung vasculature by
perfusion with a suspension of cationic colloidal silica particles,
which is a good method for in vivo studies (7). The methods
reported by Smart et al. and Song et al. are also detergent-free
and have been extensively used to isolate lipid rafts membranes
from a variety of cells (8, 9). The method by Smart et al. allows
the isolation of a more purified fraction of lipid rafts because it
uses purified plasma membranes rather than total cell lysates (8),
in contrast to the method by Song et al. which uses total cell
lysates (9). The results obtained using non-detergent extracted
rafts are more reproducible and generate a greater fraction of inner
leaflet-membrane lipids than detergent-extracted rafts (15). The
samples obtained by detergent methods have been termed deter-
gent-resistant membranes (DRMs) or detergent-insoluble fraction
(7, 10, 15). The nonionic detergents, e.g., Triton X-100, are com-
monly used to purify lipid raft fractions (7, 10, 16–18). However,
different detergents may yield different lipid raft components
because different types of raft proteins have varying degrees of
resistance to extraction by specific detergents (10, 15). Differences
between detergent and non-detergent methods for the prepara-
tion of lipid rafts may be responsible for the observed variability in
the lipid composition of the isolated rafts (7, 10, 15). We now
describe non-detergent and detergent methods to isolate lipid raft
membranes.
2 Dopamine Receptors in Lipid Microdomains 17
2. Materials
2.1. Cell Culture 1. HEK-293 cells heterologously expressing human D1 receptor

(HEK-hD1) that have been previously characterized (13).
2. Prepare complete medium for cell culture by adding 5 mL of
Pen/Strep and 50 mL of FBS to 500 mL of DMEM/F12
medium.
3. Grow cells in 150-mm dishes with complete medium in a
humidified incubator in 5% CO2 and 95% air.
2.2. Sucrose Gradient All stock solutions are prepared in distilled water at room tempera-
Centrifugation ture and stored at 4°C.
1. 250 mM 2-N-morpholino ethanesulfonic acid (Mes) stock
solution, pH » 6.7–6.8.
2. 1.5 M sodium chloride (NaCl) stock solution.
3. Mes-buffered saline (MBS) solution: 25 mM Mes and 150 mM
NaCl, pH » 6.7–6.8.
4. 500 mM sodium carbonate, pH 11 (pH need not be
adjusted).
5. 5%, 35%, and 80% sucrose solutions in MBS buffer (see Note 1).
6. Add protease inhibitor cocktail to the sodium carbonate and
sucrose solutions.
7. Protein assay using BCA kit (Pierce Thermo Scientific
(Rockford, IL)).
8. Phosphate-buffered saline (PBS).
9. D1-like receptor agonist fenoldopam (1 mM) (Sigma-Aldrich,
St. Louis, MO) stock solution, aliquoted into small volumes
(50 μL/aliquot), protected from light, and stored at −20°C.
Antioxidants are needed for prolonged incubation of dopamine
and dopamine agonists.
10. Prepare fresh solution of methyl-β-cyclodextrin (βCD) (Sigma)
(2%) in DMEM/F12 serum-free medium (SFM) at room
temperature.
11. Cholesterol-βCD solution (Sigma) for cholesterol repletion
experiment:
(a) Dissolve cholesterol (20 mg/mL) in ethanol by
sonication.
(b) Dissolve βCD (2%) in DMEM/F12 SFM.
(c) Prepare cholesterol-βCD solution by adding 20 μL of
cholesterol solution to 10 mL cyclodextrin solution, mix
by vortexing, and incubating the cholesterol-βCD solu-
tion at 40°C for 30 min (see Note 2).
18 P. Yu et al.
Table 1
Prepare varying concentrations of OptiPrep solutions
Solutions (total 5 mL) 30% 20% 10% 5%

50% OptiPrep (mL) 3.0 2.0 1.0 0.5
MBSTS (mL) 2.0 3.0 4.0 4.5
12. 50% OptiPrep stock solution: 45 mL of 60% OptiPrep mixed

with 9 mL of OptiPrep diluent.
13. MBSTS buffer: MBS with 0.5% (v/v), Triton X-100 in 10%
sucrose, or other nonionic detergents, e.g., β-octyl glucoside,
CHAPS, deoxycholate, Lubrol WX, Lubrol PX, Brij 58, Brij
96, Brij 98 (Sigma), as needed (see Note 3).
14. Prepare 5% and 30% gradient OptiPrep solutions according to
Table 1 using 50% OptiPrep stock solution and MBSTS buffer.
15. 6× sample buffer: 7.5 mL of 0.5 M Tris–HCl, pH 6.8, 1 g of
SDS powder, 3.6 mL of 100% glycerol, 2 mg of bromphenol
blue, 1 g of dithiothreitol in a total 10 mL volume with dis-
tilled water.
2.3. Western Blot 1. Nitrocellulose membranes (0.2 μm pore size) (Invitrogen Life
for Lipid Raft Proteins Technologies, Grand Island, NY).
2. Pre-stained molecular weight markers (Invitrogen Life
Technologies).
3. Vertical midi-format electrophoresis cell, which should include
a buffer tank and lid with power cables.
4. Criterion Precast Gels: 4–20% polyacrylamide gel, 26-well gel
(Bio-Rad, Hercules, CA) or 8–16% polyacrylamide gel, 15-well
gel (Invitrogen).
5. 10× Tris/Glycine/SDS stock buffer, to make 1× running
buffer.
6. 10× Tris/Glycine buffer, to make 1× transfer buffer containing
20% methanol.
7. 10× PBS-tween-20 buffer, to make 1× washing buffer.
8. 0.1% Amido Black, 45% MEOH, 10% acetic acid in distilled
water.
9. 0.1% Ponceau S solution in 5% acetic acid (remove the dye
from the membrane by several washes with distilled water).
10. Primary antibodies and secondary antibodies conjugated to
horseradish peroxidase.
11. Enhanced chemiluminescence (ECL) Western blotting detec-
tion reagents (GE Healthcare, Waukesha, WI).
3. Methods
3.1. Preparation Caveolae and lipid raft proteins are resistant to the solubilizing
of Lipid Raft Fraction actions of detergents and some non-detergent reagents, such as
with Non-detergent sodium carbonate. Therefore, the raft proteins and membranes can
Method be prepared using these detergents or reagents for sucrose gradient
centrifugation. To prepare caveolar-enriched or non-caveolar lipid
rafts, one can use the detergent-free sucrose gradient centrifuga-
tion protocol according to Song et al. (9) with slight modifications
(13). This method can be adopted for all mammalian cells and tis-
sues, including those that do not express caveolin-1 (13, 14), i.e.,
HEK-293 cells. For example, rat renal proximal tubule cells, used
as an example, do not express caveolin-1 and therefore do not have
caveolae (13, 14). We suggest using at least two 150-mm dishes
for a single preparation. All experiments are carried out at 4°C
except for cell culture and cell treatments.
1. Collect cell pellets. Culture cells in 150-mm dishes with
DMEM/F12 complete medium at 37°C until the cells reach
95% confluence. Remove the cell culture medium and wash the
cells twice with PBS. Then, starve the cells in DMEM/F12-
SFM for 1–2 h at 37°C. Treat the cells with vehicle or drugs
(e.g., fenoldopam, 2% βCD, cholesterol–cyclodextrin solution)
at 37°C for 1 h. Wash the cells once with cold PBS or cold
DMEM/F12-SFM. Scrape the cells into a 15 mL tube contain-
ing cold PBS. Pellet the cells by centrifugation at 2,000 × g for
5 min. Discard the supernatant to obtain the cell pellet.
2. Prepare cell homogenates. Add 1.5 mL of 500 mM sodium
carbonate to the cell pellet and mix by vortexing. Place the
15 mL tube containing the cells on ice and homogenize the cell
suspension using a Dounce homogenizer (10 strokes), a Teflon
polytron (three 10-s bursts), and a tip sonicator (three 30-s
bursts). The homogenization steps are carried out on ice (see
Note 4). Add 1.5 mL of 80% sucrose (final volume 3 mL, sucrose
concentration, 40%) and mix the homogenate by vortexing
(three 30 s bursts) and sonicating (three 30 s bursts) on ice.
Determine the protein concentrations by BCA kit (OD 570).
3. Prepare a discontinuous sucrose gradient. Place equal amounts
of cell homogenates (3 mL) into the bottom of each precooled
12 mL ultracentrifuge tubes and overlay 4.5 mL of 35% sucrose
and 4.5 mL of 5% sucrose to each tube. The ultracentrifuge
tubes should be balanced when placed and positioned in SW-41
buckets.
4. Centrifuge the tubes containing the cell homogenates at
180,000 × g (38,000 rpm) for 16 h at 4°C in a Beckman SW41
rotor (see Note 5).
20 P. Yu et al.
5. Remove the tubes from the bucket at the end of the

ultracentrifugation step. A light-scattering band that contains
caveolae-enriched lipid raft membranes is seen at the interface
of the 5–35% sucrose gradient. Carefully collect twelve 1 mL
fractions by pipetting 1 mL starting from the top of the ultra-
centrifuge tube and transfer the fractions into the pre-labeled
1.5 mL microcentrifuge tubes (see Note 6). The light-scatter-
ing band is located at the 3rd to 5th fractions from the top,
with the peak at the 4th fraction.
6. Prepare samples for immunoblotting. Transfer 0.5 mL aliquots
from each fraction into other pre-labeled 1.5 mL microcentri-
fuge tubes. Add 0.1 mL of 6× sample buffer to each sample.
Vortex each tube until dye and samples are mixed well and put
the tubes in boiling water for 5 min. The samples for immuno-
blotting can be saved at −20°C until use. The rest of the
fractionated samples not mixed with the 6× sample buffer are
saved at −80°C (see Note 7).
3.2. Preparation Detergent resistance or detergent insolubility results from the seg-
of Lipid Raft Fraction regation of integral or membrane-associated proteins into
with Detergent Method cholesterol- and glycosphingolipid-enriched membrane microdo-
mains termed lipid rafts. The nonionic detergents such as Triton
X-100, β-octyl glucoside, CHAPS, deoxycholate, Lubrol WX,
Lubrol PX, Brij 58, Brij 96 and Brij 98 have been used to purify
lipid raft fractions (7, 10, 16–18). However, different detergents
may yield different lipid raft components because different types of
raft proteins have varying degrees of resistance to different deter-
gents (10, 15).
1. Collect cell pellets (see Subheading 3.1, step 1) (One 150-mm
dish for one preparation).
2. Prepare cell extract on ice for 30 min in 0.3 mL cold MBSTS
(0.5% Triton X-100 and protease inhibitors) by pushing the
cell suspension through a 25-gauge needle, ten times (cell pel-
let volume is about 0.1 mL/dish and the total cell lysate vol-
ume is about 0.4 mL). Adjust the cell extract (0.4 mL) to 40%
OptiPrep by adding 0.8 mL of cold 60% OptiPrep, mix the cell
extract by vortexing. Determine the protein concentrations
using a BCA kit (OD 570). The total protein amount should
be the same for all centrifuge tubes with the same volume
(1 mL).
3. Prepare a discontinuous OptiPrep gradient. Load 1 mL of the
cell extract into the bottom of precooled 5 mL ultracentrifuge
tubes. Overlay with 1 mL of each 30%, 25%, 20%, and 0%
OptiPrep solutions in MBSTS buffer, as prepared in Table 1
(see Note 8).
4. Ultracentrifuge the OptiPrep gradient solutions at 175,000 × g
(42,000 rpm) at 4°C for 4 h in Beckman SW 50.1 rotor. Other
rotors can be used such as SW 55 (4 h at 170,000 × g), TLS55

rotor (2.5 h at 250,000 × g). However, the equivalent g-force
and centrifugation time should be adjusted according to the
rotor type. Label 1.5 mL microcentrifuge tubes for the next
step.
5. Carefully remove the ultracentrifuge tubes. Collect 0.5 mL
fractions from top to bottom and prepare for immunoblotting,
as in Subheading 3.1.
3.3. Immunoblotting Western blot allows the identification and analysis of the lipid raft
to Analyze Lipid Raft proteins. In general, one should first identify where the peak of
Proteins lipid raft fractions is located using lipid raft marker proteins such as
caveolin-1, caveolin-3, or flotillin-1. To compare the effect of drugs
on lipid raft protein expression, 4–20% Criterion Precast Gradient
Gel with 26 wells per gel is recommended. All the steps are carried
out at room temperature.
1. Run gels. Pre-warm the sucrose gradient samples in a water
bath at 37°C. Mix the samples completely by vortexing (there
should be no precipitate at the bottom of the tubes). Load the
samples and molecular weight marker into a 4–20% Criterion
Precast gradient gel. Run the gel with running buffer at 120 V
for about 2 h. Stop the electrophoresis when the dye migrates
to 0.5–1.0 cm above the bottom edge of the gel.
2. Transfer the proteins from gels onto nitrocellulose membranes.
Prepare the sandwich of gels and nitrocellulose membranes in
transfer buffer and place the sandwich into semidry transfer
equipment and start the transfer at 0.24 mA at constant cur-
rent for 60–90 min.
3. Block the membranes. Rinse the membranes twice with dis-
tilled water after transfer. Verify the protein loading by staining
the membranes with 0.1% Ponceau S solution or 0.1% Amido
Black solution for 10 s and washing the sheets with distilled
water. The stained sheets can be scanned to record the protein
loading information. Block the membranes for 1 h in blocking
buffer (5% nonfat dry milk in PBST washing buffer).
4. Perform the immunoblotting. Incubate the blocked mem-
branes overnight at 4°C with primary antibody diluted in
blocking buffer (see Note 9). Remove the primary antibody
following the overnight incubation and wash the membranes
3× with wash buffer. Incubate the membranes for 1 h with
secondary antibody diluted in blocking buffer.
5. Develop the film. Incubate the membranes for 1 min with ECL
reagent after washing with wash buffer 3×. Visualize the immu-
noreactive bands by autoradiography (see Note 10).
22 P. Yu et al.
4. Notes
1. The sucrose solutions (5%, 35%, and 80%) are prepared in MBS
buffer, pH 6.8 (13) rather then in sodium carbonate solution
(pH 11) (9). This results in a sample pH near 7.0 instead of
pH 11. This may be beneficial to most of the enzyme
proteins.
2. For cholesterol depletion experiment, the cells are incubated
with methyl-β-cyclodextrin (βCD) (2%) for 1 h at 37°C.
However, methyl-α-cyclodextrin has been recommended as a
negative control (19). For cholesterol repletion experiments,
βCD and cholesterol complex is used (Subheading 3.1, step 3).
There is a commercially available cholesterol-cyclodextrin
complex (SIGMA #C4951). However, the complex can also
be prepared as described above (Subheading 2.2, item 11).
An inactive analog of cholesterol (cholestane-3β, 5α, 6β-triol)
has been suggested as a control (20).
3. Extraction using nonionic detergents. In general, Triton X-100
or CHAPS can solubilize the membranes that are extremely
enriched in cholesterol and sphingolipids (15). Different raft
proteins have different sensitivities to the different detergents.
For example, even in the same cell type, different GPI-anchored
proteins which associate with lipid rafts can be distinguished
based on their sensitivity to solubilization in nonionic deter-
gents. A good example of this is the prion protein, a GPI-
linked protein, which was found only in non-raft fractions after
solubilization in 0.5%Brij 96, but was distributed evenly
between the raft and non-raft fractions when 0.5% Triton
X-100 was used (21).
4. To avoid loss of cell samples during homogenization, we use a
tip sonicator (five 20-s bursts, with a 2-min interval after each
burst) instead of using Dounce homogenizer and Teflon poly-
tron. All sonication steps should be performed with the test
tubes on ice. This homogenization procedure can be used for
cells but not for tissues.
5. The SW40 or SW41 rotors can be used for sucrose gradient
centrifugation. The speed of the centrifugation is specific for
each rotor. In general, the rotor speeds are 38,000 rpm
(18,000 × g)/16 h for SW41 rotor and 36,000 rpm
(16,000 × g)/18 h for SW40 rotor.
6. Collect twelve 1 mL fractions starting from the bottom by
inserting a fine plastic tube in to bottom of the centrifuge tube
and withdrawing one 1 mL each time using a 2 mL syringe, or
a peristaltic pump.
7. The sucrose gradient samples with pH 6.8–7.0 can be stored at

−80°C for enzyme assays, e.g., adenylyl cyclase assay. There are
many ways to concentrate the fraction samples such as speed-
vac concentrator or by precipitation with 10% trichloroacetic
acid (TCA). The membranes from lower sucrose gradient frac-
tions can also be concentrated by three-fold dilution of the
samples with MBS and pelleted by centrifugation at 20,000 × g
for 30 min.
8. The OptiPrep discontinuous gradient can be made by overlay-
ing 3 mL of 30% and 0.5 mL of 5% Optiprep solutions (16), or
by overlaying 1 mL of 30%, 1 mL of 25%, 1 mL of 20%, and
1 mL of 0% OptiPrep solutions (17). However, it is best to
prepare an OptiPrep continuous gradient using a machine for
preparing gradients (Bio-Rad) or by overlaying 0.8 mL of each
30%, 25%, 20%, 15%, and 0% OptiPrep solutions and pre-
centrifugation at 175,000 × g (42,000 rpm) at 4°C for 2 h in
Beckman SW 50.1 rotor. Subsequently, load the protein sam-
ples at the bottom of the continuous OptiPrep gradient tube.
9. The primary antibody can be diluted in an antibody diluting
solution (Invitrogen), and the diluted primary antibody can be
collected and saved at −20°C for subsequent usage. The pri-
mary antibody diluted in 5% milk buffer is not recommended
for storage.
10. To visualize the immunoreactive bands, the use of Licor
(Odyssey) is recommended. When using Licor, the membranes
should be blocked using a special blocking solution, such as
casein or BSA blocking buffers (Bio-Rad), and the appropriate
secondary antibody conjugated to IRDye® infrared dyes (in
PBS casein buffer). The immunoreactive bands are visualized
by scanning the membrane using Licor.
Acknowledgments
These studies were supported in part by grants from the National

Institutes of Health (HL023081, HL074940, DK039308,
HL068686, and HL092196).
References
1. Jose PA, Eisner GM, Felder RA (2002) Role of functional specificity of dopamine receptor
dopamine receptors in the kidney in the regula- subtypes. Neuropharmacology 47:
tion of blood pressure. Curr Opin Nephrol 1117–1134
Hypertens 11:87–92 3. Kong MMC, Hasbi A, Mattocks M, Fan T,
2. Holmes A, Lachowicz JE, Sibley DR (2004) O’Dowd BF, George SR (2007) Regulation of
Phenotypic analysis of dopamine receptor D1 dopamine receptor trafficking and signaling
knockout mice; recent insights into the by caveolin-1. Mol Pharmacol 72:1157–1170
24 P. Yu et al.
4. Simons K, Ikonen E (1997) Functional rafts in 13. Yu P, Yang Z, Jones JE, Wang Z, Owens SA,
cell membranes. Nature 387:569–572 Mueller SC, Felder RA, Jose PA (2004) D1
5. Insel PA, Head BP, Ostrom RS, Patel HH, dopamine receptor signaling involves caveo-
Swaney JS, Tang CM, Roth DM (2005) lin-2 in HEK-293 cells. Kidney Int
Caveolae and lipid rafts: G protein-coupled 66:2167–2180
receptor signaling microdomains in cardiac 14. Breton S, Lisanti MP, Tyszkowski R,
myocytes. Ann N Y Acad Sci 1047:166–172 McLaughlin M, Brown D (1998) Basolateral
6. Lingwood D, Simons K (2010) Lipid rafts as a distribution of caveolin-1 in the kidney. Absence
membrane-organizing principle. Science from H+-ATPase-coated endocytic vesicles in
327:46–50 intercalated cells. J Histochem Cytochem
7. Schnitzer JE, McIntosh DP, Dvorak AM, Liu J, 46:205–214
Oh P (1995) Separation of caveolae from asso- 15. Pike LJ (2004) Lipid rafts: heterogeneity on
ciated microdomains of GPI-anchored pro- the high seas. Biochem J 378:281–292
teins. Science 269:1435–1439 16. Waheed AA, Jones TL (2002) Hsp90 interac-
8. Smart EJ, Ying YS, Mineo C, Anderson RG tions and acylation target the G protein Gα12
(1995) A detergent-free method for purifying but not Gα13 to lipid rafts. J Biol Chem
caveolae membrane from tissue culture cells. 277:32409–32412
Proc Natl Acad Sci U S A 92:10104–10108 17. Verkade P, Harder T, Lafont F, Simons K
9. Song KS, Li S, Okamoto T, Quilliam LA, (2000) Induction of caveolae in the apical
Sargiacomo M, Lisanti MP (1996) Co-purification plasma membrane of Madin-Darby canine kid-
and direct interaction of Ras with caveolin, an ney cells. J Cell Biol 148:727–739
integral membrane protein of caveolae microdo- 18. Macdonald JL, Pike LJ (2005) A simplified
mains. Detergent-free purification of caveolae method for the preparation of detergent-free
microdomains. J Biol Chem 271:9690–9697 lipid rafts. J Lipid Res 46:1061–1067
10. Foster LJ, De Hoog CL, Mann M (2003) 19. Vial C, Evans RJ (2005) Disruption of lipid
Unbiased quantitative proteomics of lipid rafts rafts inhibits P2X1 receptor-mediated currents
reveals high specificity for signaling factors. and arterial vasoconstriction. J Biol Chem
Proc Natl Acad Sci U S A 100:5813–5818 280:30705–30711
11. Salzer U, Prohaska R (2001) Stomatin, flotillin-1, 20. Murtazina R, Kovbasnjuk O, Donowitz M, Li
and flotillin-2 are major integral proteins of eryth- X (2006) Na+/H+ exchanger NHE3 activity
rocyte lipid rafts. Blood 97:1141–1143 and trafficking are lipid raft-dependent. J Biol
12. Huang P, Xu W, Yoon SI, Chen C, Chong PL, Chem 281:17845–17855
Liu-Chen LY (2007) Cholesterol reduction by 21. Madre N, Smith KL, Graham CH, Jen A, Brady
methyl-beta-cyclodextrin attenuates the delta K, Hall S, Morris R (1999) Functionally differ-
opioid receptor-mediated signaling in neuronal ent GPI proteins are organized in different
cells but enhances it in non-neuronal cells. domains on the neuronal surface. EMBO J
Biochem Pharmacol 73:534–549 18:6917–6926
Chapter 3
Methods of Dopamine Research in Retina Cells

Ana Lucia Marques Ventura, Fernando Garcia de Mello,
and Ricardo Augusto de Melo Reis
Abstract
Dopamine is the main catecholamine found in the retina of most species, being synthesized from the
L-amino acid tyrosine. Its effects are mediated by G protein coupled receptors subfamilies that are commonly
coupled to adenylyl cyclase in opposite manners. There is evidence that this amine works as a developmen-
tal signal in the embryonic retina and several distinct roles have been attributed to dopamine in the retina
such as proliferation, synaptogenesis, neuroprotection, increased signal transmission in cone, gap junction
modulation, neuronal–pigmented epithelium–glial communication, and neuron–glia interaction. Here we
describe methods that have been used in the study of the dopaminergic function in the retina in the last
40 years. We emphasize the approaches used in the studies on the development of the avian and rodent
retina. The dopaminergic system is one of the first phenotypes to appear in the developing vertebrate
retina.
Key words: Retina, Dopamine, Cyclic AMP, Müller glia, Amacrine, Tyrosine hydroxylase,
Development
1. Introduction
Dopamine is a key catecholamine found in the vertebrate retina,

present mainly in a subtype of amacrine cell where most of the
machinery for synthesis (tyrosine hydroxylase—TH, dopamine
decarboxylase—DDC) and release (vesicular monoamine—VMAT
and membrane—DAT transporters) are found in most vertebrate
species (1). The best way to study the functional response of dop-
amine in retinal cells is to characterize its receptors, since most of
the effects mediated by dopamine are through two basic types of G
protein-coupled receptor, D1 and D2, which stimulate and inhibit,
respectively, the enzyme adenylyl cyclase (2). Dopamine-mediated
cyclic AMP (cAMP) accumulation, via D1-like receptors, is observed
very early during retina ontogeny, before synaptogenesis and, in
25
26 A.L.M. Ventura et al.
some species, before the expression of TH, the enzyme that

characterizes the neuronal dopaminergic phenotype (3, 4). D2-like
receptors appear in the tissue days after D1-like activity is detected
(5). In the embryonic avian retina, before the tissue is capable of
synthesizing its own dopamine via TH, dopamine synthesis is
observed from l-DOPA supplied to the neuroretina from the pig-
mented epithelium which results in dopaminergic communication
in the embryonic tissue before TH expression (6). Recently, Müller
glial cells have also been shown to be able to synthesize and release
dopamine, at least in culture conditions (7, 8). Mixed neuron–glia
cultures obtained from embryonic chick express D1A and D1B recep-
tors mRNA, but not D1D, as detected by RT-PCR (9). Müller glia
cell also expresses the D1 receptor (10). The binding of [3H]-SCH
23390 revealed a significant amount of expressed receptors and
released dopamine was detected in cell extracts of cultured Müller
cells exposed to the DA precursor, l-DOPA (11). Here we describe
practical procedures related to dopamine research in retina cells,
including signaling (cAMP accumulation), immunocytochemistry
(for dopaminergic markers), mRNA quantification and receptor
binding (for D1A, D1B, and D1D receptors, as well as D2) and assays
in neuroprotection, synaptogenesis and dopamine release in early
postnatal retina.
2. Materials
2.1. Cell Culture 1. Dulbecco’s Modified Eagle’s Medium (DMEM), 10% fetal
and Western Blotting calf serum (FCS), and gentamicin (Invitrogen, Life
Technologies, Rockford, IL).
2. Solution of trypsin (Worthington) stored in single use aliquots
(0.1% at 0.5 mL) at −70°C.
3. Epidermal growth factor (1 mg/mL EGF) and B27 supple-
ment (Invitrogen) prepared in single use aliquots of 25 μL
50 μg/mL and 0.5 mL stock, respectively. Both are added to
50 mL DMEM for the preparation of neurosphere retina
culture.
4. CMF (Ca2+ and Mg2+ free solution): 76.55 g/L NaCl, 3.05 g/L
KCl, 1.65 g/L Na2HPO4, 0.610 g/L KH2PO4, 21.95 g/L
glucose, and 7.90 g/L NaHCO3.
5. Plastic dishes: 35 or 60 mm 4-well dishes (Falcon or Nunc.
Int.).
6. Modified Laemmli buffer for cell lysis: 1 mL of 0.5 M Tris–
HCl, pH 6.8 + 1.6 mL of 10% (w/v) sodium dodecyl sulfate
(SDS) + 0.8 mL of glycerol + 0.4 mL of β-mercaptoethanol;
prepare also a 10× solution of bromophenol blue (0.2%).
3 Dopamine in Retina 27
7. Running buffer (1 L): 3 g Trizma base, 14.4 g glycine, 10 mL

10% SDS.
8. Transfer buffer (1 L): 3 g Trizma base, 14.4 g glycine, 1 mL
10% SDS, 100 mL methanol.
9. TBS (1 L): 2.42 g Trizma base, 8 g NaCl; pH 7.6.
10. 9% polyacrylamide mini-gels: mix 2.25 mL of 1.5 M Tris–HCl
buffer, pH 8.8, 2.7 mL of 30% acrylamide/0.8% bis solution,
4.05 mL H2O, 15 μL of TEMED, 15 μL of 10% ammonium
persulfate solution, and 15 μL of 10% SDS. Pour in the gel
apparatus, leaving space for a stacking gel and overlay with a
0.1% SDS solution. Polymerize for 30 min; when the polymer-
ization line appears, pour off SDS solution and make stacking
gel by mixing 625 μL of 0.5 M Tris–HCl, pH 6.8, 375 μL
acrylamide/bis solution, 1.45 mL of H2O, 12.5 μL of TEMED,
25 μL of ammonium persulfate, and 25 μL of 10% SDS. Pour
on the top of the running gel, insert comb, and let polymerize;
remove comb and wash the wells with running buffer.
2.2. mRNA Preparation 1. Trizol and DNAse (Gibco, Life Technologies). Standard pro-
tocols in this section should be done according to kit instruc-
tions or according to ref. 12.
2. First strand synthesis of cDNA is performed using the First–
Strand cDNA Synthesis Kit (Amersham Biosciences, GE
Healthcare, Piscataway, NJ).
3. The following oligonucleotides are specific to amplify retinal
cDNA preparations for avian dopamine D1A and D1B receptors,
as well as for the L27 ribosomal protein (see Fig. 1 and ref. 9.
D1A (GenBank sequence no. L36877):
5¢-CCAAGGGAGCAGAAGCTTTC-3¢ (base position 908)
Fig. 1. PCR products corresponding to D1A (372 bp), D1B (296 bp) and the ribosomal protein
L27 (235 bp, internal control) mRNA separated on a 2% agarose gel and visualized after
staining with 0.5 g/mL ethidium bromide.
5¢-TACCCGACAATGCTGGAGAC-3¢ (base position 1,279)

PCR product = 372 bp
D1B (GenBank sequence no. L36878):
5¢-GAGGACATGAGCACCACATG-3¢ (base position 610)
5¢-GTGTGATGGTGGCAGTCAAC-3¢ (base position 905)
Ribosomal protein L27 as internal control (GeneBank sequence
no. X56852):
5¢-AAGCCGGGGAAGGTGGTG-3¢ (base position 42)
5¢-GGGTGGGCATCAGGTGGT-3¢ (base position 276)
4. 50 μL reaction mixture: 50 pmol of specific oligonucleotides,
200 μM dNTPs, 2.5 U of Taq polymerase, enzyme buffer
(20 mM Tris–HCl, 50 mM KCl, pH 8.4), and 1.5 mM
MgCl2.
2.3. Binding Assays 1. Ice-cold lysis buffer: 5 mM Tris–HCl, 5 mM MgCl2, pH 7.4.

2. Incubation buffer: 50 mM Tris–HCl, pH 7.4, containing
120 mM NaCl, 5 mM KCl, 1.5 mM CaCl2, 4 mM MgCl2, and
1 mM EDTA.
3. [3H]-SCH 23390 (~70.3 Ci/mmol, Perkin-Elmer); 20 μM
(+)-Butaclamol (Sigma).
4. GF/C glass fiber filters immersed in a 0.3% solution of poly-
ethyleneimine (Sigma).
5. 10% Trichloroacetic acid (~50 mL).
2.4. Immuno- 1. Clean spherical microscope coverslips (0.15 mm).

fluorescence 2. Phosphate buffered saline (PBS) in g/L: Mix 0.1 CaCl2
(0.680 mM), 0.2 KCl (2.7 mM), 0.2 KH2PO4 (1.47 mM),
0.12 MgSO4 (0.4896 mM), 8 NaCl (136 mM), and 1.15
Na2HPO4 (8 mM) (adjust to pH 7.4 with HCl if necessary),
and pass through 0.22 μm filters.
3. Paraformaldehyde (Sigma): Prepare a 4% (w/v) solution in
PBS fresh for each experiment. Dissolve in solution in a stirring
hot-plate in a fume hood and then cool to room temperature
for use.
4. Permeabilization solution: 0.25% (v/v) Triton X-100 or tween-
20 in PBS.
5. Antibody dilution buffer: 3% (w/v) BSA in PBS.
6. Secondary antibody: Anti-mouse or anti-rabbit IgG conju-
gated to Cy3 (Jackson Immunoresearch, West Grove, PA).
7. Nuclear stain: 300 nM DAPI (4,6-diamidino-2-phenylindole)

in water.
8. Mounting medium: Antifade (Molecular Probes, Eugene,
OR).
2.5. Dopamine 1. Tris–HCl solution, pH 8.8.

Extraction 2. Dihydroxybenzylamine (DHBA, the internal standard of
extraction).
3. Alumina
4. 100 mM perchloric acid.
5. Reverse phase column: LC-18 column (4.6 mm × 250 mm,
Supelco).
6. Mobile phase: 20 mM sodium dibasic phosphate, 20 mM citric
acid, pH 2.64, containing 10% methanol, 0.12 mM Na2EDTA,
and 566 mg/L heptanesulfonic acid.
3. Methods
3.1. Retinal Cultures 1. Remove the eyes, dissect the retinas free of the pigmented epi-
for Dopamine Assays thelium in Dulbecco’s modified Eagle’s medium (DMEM).
2. Transfer the retinal pieces and wash twice in Ca2+- and Mg2+-
free solution (CMF).
3. Dissociate the tissue using trypsin (Worthington) for 10 min
(37°C).
4. At this point, the experimenter should decide in a number of
choices as listed below.
3.1.1. Mixed Neuronal-Glial Dissociate and plate 1/2 retina per dish (1–1.5 × 107 cells or more).
Cultures (See FIG. 2A) It is not necessary to treat plastic dishes with substrates. Ideal for
functional assays measuring receptor mediated second messenger
shifts, binding or Western blot analysis for specific proteins medi-
ated by dopamine.
3.1.2. Enriched Neuronal Low density neuronal cultures, where approximately 2 × 106 cells
Cultures (See Fig. 2b) or less are seeded onto treated poly L-Lysine (10 μg/mL) plastic
dishes in DMEM medium plus 1% FCS (see Note 1).
3.1.3. Müller Glia Cell 1. Use the amount of 5 × 106 cells over culture dishes in DMEM
Cultures from Embryonic containing 10% FCS.
or Postnatal Retina 2. Medium should be changed every 3 days.
(See Fig. 2c)
3. After approximately 10 days, cell cultures are treated with
4 mM ascorbic acid for 2 h to eliminate neurons (13).
Fig. 2. Dopamine has been investigated as a developmental signal in the retinal tissue or cultures prepared in many
different ways. (a) Mixed neuron–glial cells (prepared in high density, with ~20 × 106 cells) is ideal for functional assays
measuring receptor mediated second messenger shifts, binding or western blot analysis. (b) Enriched neuronal cells pre-
pared in low density, with ~2 × 106 cells or less, seeded on treated poly L-Lysine (10 μg/mL) plastic dishes. (c) Müller glia
culture from embryonic or postnatal retina prepared from progenitors (5 × 106 cells) in DMEM containing 10% FCS and
cultured for 10 days, when neurons are eliminated (13). (d) Neurospheres retinal cultures prepared in the presence of EGF
in an untreated culture dish. On day 5, neurospheres are plated under differentiating conditions that allows the emergence
of all retinal neurons and Müller glia. (e) Photomicrograph (Dr. Marilia Guimarães) of a TH-positive cell in E10C3 chick retina
cell culture (15).
4. Purified glial cultures can be used 3 days later and maintained

for up to 3 weeks. These cultures can be used for different
purposes such as signaling, binding, immunocytochemistry.
3.1.4. Neurospheres Retinal cells are plated in DMEM supplemented with gentamicin,
Retinal Cultures 1% B27 supplement, and 20 ng/mL epidermal growth factor
(See Fig. 2d) (EGF) and then placed in an untreated 35 mm culture dish
(Corning). On day 5, neurospheres are plated under differentiat-
ing conditions onto a poly-D-Lysine matrix (Invitrogen) coverslips
with different substrates such as laminin or fibronectin (Sigma).
3.2. Western Blotting 1. Incubate retinas or retinal cells in culture with drugs of interest
for Detection in DMEM medium
of Dopamine Markers 2. Transfer tissues to ~70 μL of sample buffer without bromophe-
(TH, L-Dopa nol blue. Mix well with vortex.
Decarboxylase, VMAT,
3. For retinal cells in culture, remove medium, wash cells with
DAT, nurr-1) medium without serum, and add ~70 μL of sample buffer.
3.2.1. Preparation Scrape cells with a large bore pipette tip and transfer viscous
of Sample Extracts material to tubes.
4. Boil extracts in boiling water for 10 min and centrifuged at
27,000 × g for 10 min to remove non-soluble material.
5. Estimate protein content in 2 μL samples of extracts by the

Bradford protein assay, using a BSA solution containing 2 μL
of sample buffer as standard; make in duplicate or triplicate.
6. After protein content determination, add 0.1 vol. of a 10× bro-
mophenol blue solution to the remaining volume of samples.
7. Prepare 9% polyacrylamide mini-gels according to
Subheading 2.1, item 10.
8. Mount gels in the running apparatus and add running buffer
in the two chambers.
9. Add retinal extract samples (50 μg/lane) to the lanes of the
SDS polyacrylamide gels and run at ~20 mA for 1–1.5 h, until
dye reaches the end of the gel.
10. Soak PVDF membranes (GE Healthcare) for 10 s in pure
methanol, 5 min in H2O, and 10 min in transfer buffer. Transfer
proteins to PVDF membranes for 1 h, at 100 V.
11. Disassemble the transfer unit and stain membrane with a 0.2%
Ponceau rouge solution in 3% TCA; check the transfer, mark
molecular weight bands; remove Ponceau rouge solution with
Tris-buffered saline (TBS).
3.2.2. Immunodetection 1. Block membranes with 5% nonfat milk in Tris-buffered saline

(pH 7.6) with 0.1% Tween-20 (TBS-T), for 1 h, at room
temperature.
2. Incubate membranes with diluted primary antibodies (1:2,000,
TH; 1:1,500, DDC; 1:1,000, VMAT), overnight, at 4°C.
3. Wash membranes 3× with TBS-T.
4. Incubate with secondary antiserum conjugated to horseradish
peroxidase (Bio-Rad Labs. Inc.) for 1 h at room temperature.
5. Wash membrane 2× with TBS-T for 5 min and 1× with TBS
for 10 min. Results are detailed elsewhere (6, 7, 10, 14).
6. Develop blots using enhanced chemiluminescence (ECL),
according to the manufacturer’s protocol (ECL plus, GE
Healthcare).
7. Strip membranes in glycine 0.2 M, pH 2.2, for 30–40 min, at
room temperature.
8. Re-probe membranes with anti-ERK 2 (Cell Signaling) or
anti-actin (Santa Cruz Biotechnology, Santa Cruz, CA), at
4°C, followed by incubation with the secondary antibody and
detection as described above (7, 10, 14).
3.3. Immuno- Cultures should be prepared in coverslips for the purpose of saving
cytochemistry antibody.
3.3.1. Culture Fixation 1. Briefly rinse retinal cell cultures with PBS and fix in 4% para-
and Staining formaldehyde (PA) in 0.16 M Phosphate buffer (pH 7.2) for
5 min.
2. Rinse cultures extensively with sodium phosphate-buffered

saline (PBS).
3. Rinse cultures with PBS plus Triton X-100 (0.25%).
4. Pre-incubate cultures with 5% bovine serum albumin (BSA) in
PBS plus Triton X-100 (0.25%).
5. Incubate culture coverslips with primary antibody (1:500, TH;
1:700, DDC; 1:300, VMAT; 1:300 Nurr-1).
6. After several rinses in PBS, incubate cultures with appropriate
secondary Alexa fluor 488 conjugated (Molecular Probes) or
598 conjugated (use 1:500 dilution). Alternatively, use
secondary biotinylated antibody (Vector Labs, 1:200) for 2 h,
followed by the avidin–biotin complex (ABC, Vector Labs) for
an additional 2 h (see Subheading 3.3.2).
7. Wash in PBS and counterstain culture cells with DAPI. Mount
slides in sodium n-propyl-gallate 0.2 M (pH 7.2) in glycerol.
8. Control cultures of immunohistochemistry reactions should
be incubated with PBS in the absence of primary antibody.
9. Acquire photomicrographs with standard fluorescence micros-
copy. Example of TH detection in retina cells in culture or
tissue is shown in (15, 16) (Fig. 3).
3.3.2. Retinal Tissue 1. Use alternate radial retina sections for performing immuno-
histochemistry. DDC, TH, VMAT or DAT are commonly
markers used for dopaminergic immunohistochemistry (see
Note 2).
2. Sections are rinsed in PBS and pre-incubated in 5% bovine
serum albumin (BSA) and 3% normal goat serum in PBS, for
1 h. Then, sections are incubated in an antibody against DDC
Fig. 3. Photomicrograph of TH-positive amacrine cell located in the inner nuclear layer
(INL) in a radial section of chick retina. Scale bar: 20 μm; ONL outer nuclear layer, GCL
ganglion cell layer, IPL interplexiform layer. The image is from Dr. Patricia Gardino.
(Chemicon anti-rabbit 1:200 in PBS plus 0.25% Triton X-100,

overnight).
3. Controls are performed by omission of the primary antibody.
Next day, the retinal sections are rinsed in PBS and incubated
in the secondary biotinylated antibody against rabbit IgG
(Vector Labs, 1:200) for 2 h, followed by the avidin–biotin
complex (ABC, Vector Labs) for an additional 2 h. The bind-
ing of the antibody to the sections is revealed by the addition
of a substrate for peroxidase (DAB or SG, Vector Labs
(Southfield, MI)).
4. Sections are coverslipped with 40% glycerol in PBS. Alternate
sections are processed for TH immunohistochemistry however
in this case, there was no need to heat the tissue to improve
antibody penetration. The sections are pre-incubated in 5%
BSA and 3% normal goat serum in PBS for 1 h followed by the
incubation in rabbit antibody against TH (Eugene Tech,
AB1569) at the dilution of 1:3,000 in PBS plus 0.25% Triton
X-100, overnight. As for DDC the same procedure is applied
to visualize the immunohistochemistry reaction with peroxi-
dase. Sections are also coverslipped with 40% glycerol in PBS
(4, 17, 18).
3.4. Detection of 1. Remove eyes, transfer to cold Ca2+, Mg2+ free solution (CMF)
mRNA for Dopamine and dissect retinal tissue free of pigmented epithelium under
D1A and D1B Receptors environmental light.
from Chick Retina by 2. Transfer each retina to a new 1.5 mL tube containing 0.5 mL
RT-PCR (See Fig. 1) Trizol reagent to extract total RNA; mix well in a vortex (see
3.4.1. Extraction
Note 3).
of Retinal RNA 3. Centrifuge for 10 min, at 18,000 × g, at 4°C. Transfer superna-
tant to a new tube and add 0.2 mL pure, high quality chloro-
form. Mix in vortex for 15 s and leave for 2–15 min at room
temperature.
4. Centrifuge for 15 min, at 11,000 × g, at 4°C. Transfer the
aqueous phase to a new tube and add 0.5 mL of high quality
isopropanol; leave for 5–10 min at room temperature.
5. Centrifuge for 10 min, at 15,300 × g, at 4°C. Remove superna-
tant and add carefully 1 mL of high quality 75% ethanol in
DEPC-treated water to wash pellet.
6. Remove supernatant and add 0.5 mL of 100% ethanol.
7. Store at −70°C until use.
3.4.2. DNaseI Treatment 1. Centrifuge total RNA samples for 5 min, at 15,300 × g, at
4°C.
2. Leave the tubes opened on the bench at room temperature
until last traces of fluid have evaporated; add 25 μL of DEPC-
treated water (see Note 4).
3. Incubate for 10 min at 55–60°C; keep at 4°C.

4. Dilute 1 μL of RNA preparations in 200 μL to determine the
concentration and purity of RNA in a spectrophotometer by
reading the OD at 260 and 280 nm (see Note 5).
5. Incubate 30 μg of RNA with RNAase free DNase I (0.5 U/μg
RNA) in a 100 μL incubation mixture, containing enzyme
buffer (20 mM Tris–HCl, 50 mM KCl, pH 8.4,) and 2 mM
MgCl2; incubate for 10 min at 37°C. Keep at 4°C.
6. Extract RNA by adding 1 vol. of buffered phenol and mix.
7. Centrifuge at 54,300 × g for 10 min at 4°C. Remove superna-
tant and add 1 vol. of chloroform.
8. Centrifuge at 54,300 × g for 10 min at 4°C. Remove superna-
tant and add 1/10 vol. of 3 M NaOAc.
9. Add 2 vol. of 100% ethanol and incubate for 20 min at
−70°C.
10. Centrifuge at 54,300 × g at 4°C.
11. Wash pellet with 70% ethanol.
12. Discard supernatant and add 20 μL DEPC-treated water.
13. Determine the concentration of RNA in samples using a
spectrophotometer.
3.4.3. First Strand 1. Incubate RNA preparation for 10 min at 65°C to denature
Synthesis of cDNA eventual double strand segments in RNA.
2. Dilute RNA to 1 μg/20 μL with DEPC-treated water.
3. Perform first strand synthesis of cDNA following the proce-
dure described in the First strand cDNA synthesis kit (GE
Healthcare). In brief, incubate 1 μg of RNA in a 33 μL reac-
tion mixture containing Moloney murine leukemia virus
reverse transcriptase, 0.2 μg of random hexamers, and 6 mM
dithiothreitol for 60 min at 37°C.
4. Store at 4–8°C until use.
3.4.4. Amplification 1. Amplify directly 5 μL of cDNA preparation in a 50 μL reaction

of cDNA by PCR mixture prepared according to Subheading 2.2 (see Note 6).
2. Run PCR by initially denaturing samples at 94°C for 4 min
and submitting them to 25–27 cycles of 1 min at 94°C (dena-
turation), 1 min at 60°C (annealing) and 1 min at 72°C (exten-
sion), followed by a final extension of 5 min at 72°C.
3. Prepare a 2% agarose gel without Ethidium Bromide; add 2 μL
of gel loading buffer (0.25% bromophenol blue, 0.25% xylene
cyanol, 30% glycerol) to 10 μL samples.
4. Run samples and ladder in TAE buffer at 75 V for 40 min.
5. Stain gel with Ethidium bromide solution (0.5 μg/mL)
~40 min.
6. Destain in water for 20 min at room temperature; visualize

under U.V. light and photograph with a digital camera using a
yellow filter.
7. To compare the relative amount of the different PCR prod-
ucts, amplify the same amount of cDNA with specific oligo-
nucleotides, photograph gels and determine the density of
each gel band using the 1D Gel Analysis Software (Kodak) or
a similar program. Relate the intensities of the bands corre-
sponding to receptor mRNA with the intensity of the band
corresponding to the internal control L27 ribosomal protein
(see Note 7). Results are detailed (9, 10).
3.5. Dopamine D1 1. Remove eyes, transfer to cold Ca2+, Mg2+ free solution (CMF)
Receptor Binding and dissect retinal tissue free of pigmented epithelium under
Assays in Chick environmental light (see Note 8).
Retinal Membranes 2. Transfer tissues to a Dounce homogenizer (type B) containing
(See Fig. 4) ~3 mL of ice-cold lysis buffer prepared in Subheading 2.3.
3.5.1. Membrane 3. Disrupt cells (using 10–13 stroke movements).
Preparation 4. Transfer material to centrifuge tubes and wash the homoge-
nizer with 1–2 mL of ice-cold incubation buffer (see
Subheading 2.3).
5. Transfer to tubes and centrifuge at 27,000 × g, for 30 min, at
4°C
150
Specific [3H]-SCH23390 binding
(fmol/mg protein)
100 800
Bound/Free
600
400
50 200
0
0 50 100 150
Bound (fmol/mg protein)
0
0.0 0.5 1.0 1.5 2.0
[3H]-SCH 23390 (nM)
Fig. 4. Specific binding of [3H]-SCH 23390 to homogenates of E8C4 (embryonic day 8

retina and 4 days in vitro) mixed neuron–glia cultures. Cell membranes are incubated with
various concentrations of [3H]-SCH 23390 (0.1–2.0 nM) in Tris–HCl buffer, pH 7.4 in a final
volume of 0.2 mL. Data from saturation isotherms are transformed by the method of
Scatchard and submitted to linear regression analysis. [3H]-SCH 23390 bound with high
affinity to homogenates of retinal cells in culture, showing a Kd value of 0.18 nM. The total
number of [3H]-SCH 23390 binding sites revealed by the Scatchard plot (inset) was
129 fmol/mg protein.
6. Discard the supernatant and add ice-cold binding buffer to the

resulting membrane pellet.
7. Transfer material to Dounce homogenizer and homogenize
the clumps until a homogeneous mixture is obtained
8. Store at 4°C (see Note 9).
3.5.2. Incubation 1. Add 50 μL of previously prepared diluted solutions of the

of Membranes with antagonist [3H]-SCH 23390 (~70.3 Ci/mmol, Perkin-Elmer)
[3 H]-SCH 23390 to all incubation tubes (see Note 10).
2. Add 50 μL of buffer to the incubation tubes marked as “Total
binding” and 50 μL of 20 μM (+)-Butaclamol to incubation
tubes named “nonspecific binding” (see Note 11).
3. Begin incubation of 60 min, at room temperature, with the
addition of 100 μL samples of membrane preparation (~0.1 mg
protein).
4. While incubation of membranes is running, immerse GF/C
glass fiber filters in a 0.3% solution of polyethyleneimine; this
procedure decreases nonspecific adsorption of the ligand to the
filter (see Note 12).
5. While incubation is running, precipitate a sample of membrane
preparation with an equal volume of 10% Trichloroacetic acid
to estimate protein content later; store at −20°C until use.
6. Terminate the incubation of tubes in a successive order; dilute
the content of each tube with 3 mL of ice-cold washing buffer
(50 mM Tris–HCl, pH 7.2).
7. Filter content of each tube under negative pressure through
GF/C pretreated filters.
8. Wash tubes three times with 2 mL of washing buffer and
filter.
9. Add 50 μL of [3H]-SCH 23390 diluted solutions directly to
separate dry filters to estimate precisely the concentrations of
ligand that were added to the incubation tubes. Estimate
radioactivity bound to filters by scintillation spectroscopy (see
Note 13).
10. Centrifuge precipitated membrane preparation obtained in
step 5, add 0.1–0.2 N NaOH, and determine protein
content.
11. Calculate the specific binding of the radioactive ligand by sub-
tracting the nonspecific binding from the total binding.
Transform values of radioactivity in cpm to fmol of ligand by
using the specific activity of the radioactive ligand; divide val-
ues by protein content. This procedure will give an amount of
receptors in fmol/mg protein.
12. Estimate radioligand parameters KD and Bmax using Graphpad

Prism software (see Note 14). Representative results are
detailed (11, 19).
3.6. cAMP cAMP is the main signal generated by dopaminergic input in the
Accumulation retina and it can be estimated in this tissue or in retinal cultures
(See Fig. 5) according to a competitive binding assay described previously (3)
(see Note 15).
3.6.1. Stimulation of Cells 1. Retina cells, mixed neuron–glial cells, or confluent Müller glial
cells in culture are pre-incubated for 10 min at 37°C in DMEM
medium buffered with 20 mM HEPES at pH 7.3, containing
0.5 mM isobutylmethylxantine and 100 μM ascorbic acid to
inhibit cAMP-dependent phosphodiesterase and prevent oxi-
dation of dopamine, respectively.
2. Add dopamine, D1-like agonists, or antagonist at the indicated
final concentration and incubate further for 15 min; stop
reaction by adding trichloroacetic acid to a final concentra-
tion of 5%.
3. Scrape cells and transfer all material to tubes; keep frozen until
further use.
3.6.2. Ion Exchange 1. Add trace amounts of [3H]-cAMP (50 nCi in 50 μL) to sample
Chromatography tubes and centrifuge for 30,000 × g for 30 min. Separate
of Samples supernatants and dissolve precipitates with NaOH to measure
protein content later. Add supernatants to ion-exchange resin
columns (Dowex AG50W-X4, 200–400 mesh) previously
equilibrated with 1 N HCl (see Note 16).
Fig. 5. cAMP accumulation in cells of cultures at E8C4 stimulated with dopamine (100 μM)
is completely blocked when pre-incubated with the selective D1-like receptor antagonist,
SCH-23390 (1 μM).
2. Wash columns with 6 mL of water each and discard; elute

columns three times with 3 mL of water and collect. Determine
radioactivity in 150 μL of the fractions by scintillation spec-
troscopy; also determine radioactivity in 50 μL of [3H]-cAMP
tracer solution. Calculate the % recovery of [3H]-cAMP in
tubes; Select the tubes with highest recovery to measure
cAMP.
3.6.3. cAMP Determination 1. Incubate 50 or 100 μL samples with 25 μL solution of PKA /

in Samples BSA and a fixed concentration of [3H]-cAMP (2 pmol/20 μL)
in 50 mM acetate buffer, pH 4.0, at 4°C, for 90 min; add
H2O to complete the volume of the reaction to 200 μL (see
Note 17).
2. At the same time, construct a standard curve with known
concentrations of nonradioactive cAMP (use 0, 1, 3, 5, 10, and
15 pmol). Incubate with PKA and [3H]-cAMP in acetate buffer
as above.
3. Interrupt reaction in samples by adding 2 mL of 200 mM
phosphate buffer, pH 6.0; filter samples through Millipore
acetate filters (0.45 μm), washing tubes 3× with 2 mL phos-
phate buffer; dry filters and quantify the bound radioactivity by
liquid scintillation.
4. Plot data from the standard curve by relating radioactivity
versus concentrations of nonradioactive cAMP. Estimate
nonradioactive cAMP in samples; correct values with the %
recovery from the chromatography columns and divide by
protein content of samples to obtain values in pmol/mg
protein; detailed information are shown in refs.( 3, 10, 11,
14, 20.)
3.7. Dopamine Dopamine in cell extracts can be measured by HPLC analysis

Extraction coupled with electrochemical detection (0.5 V) as described before
and Quantification (6, 7, 21).
1. Cell extracts are added in Tris–HCl solution (pH 8.8) in the
presence of dihydroxybenzylamine (DHBA, the internal stan-
dard of extraction).
2. The amines are precipitated by addition of alumina, then wash
three times with water and eluted with 100 mM perchloric
acid.
3. After centrifugation, filter and inject the supernatant into the
reverse phase column. Fast isocratic separation is obtained
using an LC-18 column eluted with the mobile phase described
in Subheading 2.5.
4. Express the results as nmol/mg cell protein.
In summary, dopamine is a key neurotransmitter in the retina

as its effects guide the development of this tissue and mediates
several distinct roles such as proliferation, signaling, differentiation
and death. Several elements of the dopaminergic system (recep-
tors, enzymes and transporters) emerge in a temporally defined
way in this organ. It is also clear that dopamine regulates retinal
morphogenesis by inhibiting the extension of neurites of retinal
cells (22), reduces retinal apoptosis (23) and restricts retinal cell
divisions (24). Dopamine is also known to inhibit the function of
the N-Methyl D-Aspartate receptor in the avian or rodent retina as
assayed through neurotransmitter release (25) or whole cell currents
(26). This chapter describes key approaches to study dopamine in
the retina, which might be useful to evaluate dopaminergic roles
in the normal retina as well as in deficits such as myopia and
albinism and disorders in the brain such as Parkinson’s disease.
4. Notes
1. It is important to keep serum at a low percentage. At low

density, neurons are the vast majority of cells in the first 72 h.
Glial appearance could be further avoided with cytosine
arabinoside (10 μM) treatment. These cultures are ideal for
neurotoxicity assays as dopamine have been shown as a poten-
tial neuroprotective factor in the retina (23).
2. In order to improve DDC labeling, radial sections are dipped
in 10 mM citrate buffer, pH 6.0, and heated in microwave
oven for 1 min.
3. Use new tubes and tips for RNA extraction and treatment;
solutions always should be made in 0.1% DEPC-treated
water.
4. Working in a laminar flow reduces contamination with RNases
from bacteria and other contaminants.
5. Reading at 260 nm allows calculation of the concentration of
nucleic acid in the sample. An OD of 1 corresponds to approx-
imately 40 μg/mL for single strand DNA or RNA. The ration
between readings at 260 and 280 nm (OD260/OD280) provides
the purity of nucleic acids. Pure preparations of RNA should
have values of 2.0. Contamination with protein or phenol will
give values significantly less than 2.0; If samples are not pure,
extract RNA with buffered phenol as described in step 6 of the
DNase I treatment procedure.
6. For PCR amplification of cDNA, it is useful to prepare first a

master mix containing water, 1× enzyme buffer, dNTPs and
MgCl2.
7. To compare the expression of mRNA for different receptor
subtypes in the tissue it is necessary to work in the linear por-
tion of the PCR amplification curve, where the amount of
amplified PCR products are directly proportional to the initial
amount of cDNA used in the PCR amplification. To determine
this, make a PCR amplification curve for each pair of specific
oligonucleotides by varying the number of PCR amplification
cycles.
8. Membrane preparations of cultured retinal cells can be obtained
by adding lysis buffer to cultures and transferring material to
centrifugation tubes with a pipette.
9. Other homogenizers such as Polytron or Potter can also be
used; prepare incubation tubes while homogenate is under
centrifugation.
10. A stock solution of 400 nM of [3H]-SCH 23390 should be
prepared with binding buffer. For each experiment, successive
dilutions from this stock should be prepared; concentrations of
0.4, 0.8, 1.6, 2.4, and 6.4 nM will result in 0.1, 0.2, 0.4, 0.8,
and 1.6 nM final concentrations. As SCH 23390 is a high
affinity antagonist, final concentrations should be in the range
of 0.1–4 nM.
11. Total binding tubes will contain only membranes and the radio-
active ligand and will provide the specific binding of the ligand
to receptors + nonspecific binding of the ligand. Nonspecific
binding tubes will contain membranes + radioligand + excess of
a nonradioactive ligand to block the specific binding of the
radioactive ligand to receptors, thus providing an estimate of
the nonspecific binding of the ligand. Values obtained in these
tubes should be subtracted from the values obtained in the total
binding tubes in order to estimate the specific binding of the
radioactive ligand to receptors; three total binding + three
nonspecific binding tubes should be made for each ligand
concentration.
12. GF/B glass fiber filters can also be used; however, they have a
higher nonspecific adsorption of the ligand.
13. The reaction mixtures should be filtered under a medium neg-
ative pressure. If the pressure is too high, membranes will pass
through filters; if pressure is too low, the ligand will dissociate
from receptors by dilution. The ideal pressure is the one at
which the buffer flows constantly through filters, thus wash-
ing them.
14. The amount of D1 receptors can also be estimated by performing

displacement experiments; in this case, membranes are incu-
bated with a constant concentration of [3H]-SCH 23390
(around the KD value) and increasing concentrations of nonra-
dioactive compounds.
15. Although cAMP is the major signal stimulated by dopamine,
other second messengers have been reported to be stimulated
by D1 receptors such as inositol triphosphate, in the striatum
(27). Retinal cells should be incubated with 1 μCi myo-[2-3H]
inositol (Perkin-Elmer) for 3 h. Cultures or retina pieces are
treated with selective agonists for 15 min in the presence of
10 mM LiCl (induce accumulation of inositol phosphates).
Stop the reaction by the addition of trichloroacetic acid fol-
lowed by ether extraction; the supernatant containing inositol
polyphosphates are separated by ion exchange chromatogra-
phy (Dowex AG1-X8 resin, formate form; Bio-Rad). Inositol
triphosphate (ip3) is quantified in a liquid scintillation analyzer
(28).
16. Use ~8 cm (height) by 1 cm (diameter) Dowex-AG50W-X4
columns. Equilibrate columns with 10 mL 1 N HCl + 10 mL
H2O immediately before use.
17. A stock solution of the regulatory sub-unit of PKA should be
prepared before cAMP assays. Prepare a 1 mg/mL PKA
solution; keep aliquots frozen until use; prepare a 20 mg/mL
solution of BSA; keep frozen. Dilute the necessary amount of
PKA for the entire assay (five to tenfold) in a working 4 mg/
mL solution of BSA.
Acknowledgements
This work was supported by grants from FAPERJ, CNPq, PROPPi-

UFF and INCT-CNPq (INNT).
References
1. Witkovsky P, Dearry A (1992) Functional roles tyrosine hydroxylase immunoreactive amacrine
of dopamine in the vertebrate retina. Prog cells in the developing chick retina. Brain Res
Retinal Res 11:247–292 Dev Brain Res 72:226–236
2. Reis RAM, Ventura ALV, Kubrusly RC, de 5. Ventura ALM, Klein WL, de Mello FG (1984)
Mello MC, de Mello FG (2007) Dopaminergic Differential ontogenesis of D1 and D2 dop-
signaling in the developing retina. Brain Res aminergic receptors in the chick embryo retina.
Rev 54:181–188 Brain Res 314:217–223
3. de Mello FG (1978) The ontogeny of dop- 6. Kubrusly RCC, Guimarães MPZ, Vieira APB,
amine-dependent increase of adenosine Hokoç JN, Casarini DE, de Mello MC, de
3¢,5¢-cyclic monophosphate in the chick retina. Mello FG (2003) L-DOPA supply to the neuro
J Neurochem 31:1049–1053 retina activates dopaminergic communication
4. Gardino PF, dos Santos RM, Hokoc JN (1993) at the early stages of embryonic development.
Histogenesis and topographical distribution of J Neurochem 86:45–54
7. Kubrusly RC, Panizzutti R, Gardino PF, Stutz 18. Taveira da Silva R, Hokoç JN, de Mello FG
B, Reis RA, Ventura AL, de Mello MC, de et al (2009) Differential immunodetection of
Mello FG (2008) Expression of functional L-DOPA decarboxylase and tyrosine hydroxy-
dopaminergic phenotype in purified cultured lase in the vertebrate retina. Int J Dev Neurosci
Müller cells from vertebrate retina. Neurochem 27:469–476
Int 53:63–70 19. De Mello MC, Pinheiro MC, de Mello FG
8. de Melo Reis RA, Ventura ALV, Schitine CS, (1996) Transient expression of an atypical
de Mello MC, de Mello FG (2008) Müller glia D1-like dopamine receptor system during avian
as an active compartment modulating nervous retina differentiation. Braz J Med Biol Res
activity in the vertebrate retina: neurotransmit- 29:1035–1044
ters and trophic factors. Neurochem Res 20. de Mello MC, Ventura ALM, Paes de Carvalho
33:1466–1474 R, Klein WL, de Mello FG (1982) Regulation
9. Soares HC, Reis RA, De Mello FG, Ventura of dopamine and adenosine-dependent adeny-
AL, Kurtenbach E (2000) Differential expres- late cyclase systems of chick embryo retina cells
sion of D(1A) and D(1B) dopamine receptor in culture. Proc Natl Acad Sci USA
mRNAs in the developing avian retina. 79:5708–5712
J Neurochem 75:1071–1075 21. Arita DY, Di Marco GS, Schor N, Casarini DE
10. Kubrusly RC, da Cunha MC, Reis RA, Soares (2002) Purification and characterization of the
H, Ventura AL, Kurtenbach E, de Mello MC, active form of tyrosine hydroxylase from mesan-
de Mello FG (2005) Expression of functional gial cells in culture. J Cell Biochem 87:58–64
receptors and transmitter enzymes in cultured 22. Lankford KL, De Mello FG, Klein WL (1988)
Müller cells. Brain Res 1038:141–149 D1-type dopamine receptors inhibit growth
11. Ventura ALM, de Mello FG (1990) D1 dop- cone motility in cultured retina neurons: evi-
amine receptors in neurite regions of embry- dence that neurotransmitters act as morpho-
onic and differentiated retina are highly coupled genic growth regulators in the developing
to adenylyl cyclase in the embryonic but not in central nervous system. Proc Natl Acad Sci
the mature tissue. Brain Res 530:301–308 USA 85:2839–2843
12. Sambrook J, Fritsch E, Maniatis T (1989) 23. Varella MH, de Mello FG, Linden R (1999)
Molecular cloning: a laboratory manual. Cold Evidence for an antiapoptotic role of dopamine
Spring Harbor Laboratory, New York in developing retinal tissue. J Neurochem
13. Reis RA, Cabral da Silva MC, Loureiro dos 73:485–492
Santos NE, Bampton E, Taylor JS, de Mello 24. Tibber MS, Whitmore AV, Jeffery G (2006)
FG, Linden R (2002) Sympathetic neuronal Cell division and cleavage orientation in the
survival induced by retinal trophic factors. developing retina are regulated by L-DOPA.
J Neurobiol 50:13–23 J Comp Neurol 496:369–381
14. Kubrusly RC, Ventura AL, Reis RA et al (2007) 25. Do Nascimento JLM, Kubrusly RCC, Reis RAM,
Norepinephrine acts as D1-dopaminergic ago- De Mello MC, De Mello FG (1998) Atypical
nist in the embryonic avian retina: late expres- effect of dopamine in modulating the functional
sion of β1-adrenergic receptor shifts inhibition of NMDA receptors of cultured retina
norepinephrine specificity in the adult tissue. cells. Eur J Pharmacol 343:103–110
Neurochem Int 50:211–218 26. Castro NG, de Mello MC, de Mello FG,
15. Guimarães MZP, Hokoç JN, Duvoisin R et al Aracava Y (1999) Direct inhibition of the
(2001) Dopaminergic retinal cell differentia- N-methyl-D-aspartate receptor channel by dop-
tion in culture: modulation by forskolin and amine and (+)-SKF38393. Br J Pharmacol
dopamine. Eur J Neurosci 13:1931–1937 126:1847–1855
16. Borba JC, Henze IP, Silveira MS et al (2005) 27. Wang HY, Undie AS, Friedman E (1995)
Pituitary adenylate cyclase-activating polypep- Evidence for the coupling of Gq protein to
tide (PACAP) can act as determinant of the D1-like dopamine sites in rat striatum: possible
tyrosine hydroxylase phenotype of dopaminer- role in dopamine-mediated inositol phosphate
gic cells during retina development. Brain Res formation. Mol Pharmacol 48:988–994
Dev Brain Res 156:193–201 28. Reis RA, Kubrusly RC, de Mello MC, de Mello
17. Dos Santos RM, Gardino PF (1998) Differential FG (1995) Transient coupling of NMDA
distribution of a second type of tyrosine receptor with ip3 production in cultured cells
hydroxylase immunoreactive amacrine cell in of the avian retina. Neurochem Int
the chick retina. J Neurocytol 27:33–43 26:375–380
Chapter 4
Capture of D2 Dopamine Receptor Signaling Complexes

in Striatal Cells for Mass Spectrometry Proteomic Analysis
Nadine Kabbani and Jacob C. Nordman
Abstract
In recent years advancements in proteomic techniques have contributed to the understanding of protein
interaction networks (Interactomes) in various cell types. Today, high throughput proteomics promises to
define virtually all of the components of a signaling and a regulatory network within cells for various mol-
ecules including membrane-spanning receptors. The D2 dopamine receptor (D2R) is a primary mediator
of dopamine transmission in the brain. Signaling through D2Rs has been linked to dopamine-mediated
effects on motivation, reward, locomotion and addiction to drugs of abuse. In the striatum, the D2R is a
key mediatory of dopamine transmission. Actions on this receptor are an important pharmacological prop-
erty of various drugs including typical antipsychotics and drugs of abuse. Here we provide an approach for
the identification protein interaction networks of the D2R within striatal cells. We discuss key assays and
techniques, such as cellular membrane protein fractionation, western blot analysis, magnetic bead coim-
munoprecipitation, and liquid chromatography electrospray ionization (LC-ESI) mass spectrometry, that
can be used for the isolation and characterization of D2R protein interaction networks. This approach
presents a reliable method for the identification and characterization of D2R signaling within cells.
Key words: Proteome, Dopamine signaling, Mass spectrometry, Antibody, Membrane spanning
protein, G protein coupled receptors, Interactome
1. Introduction
Dopamine receptors are members of a large family of membrane

spanning proteins that have seven transmembrane domains and
can functionally couple to heterotrimeric GTP binding proteins
(G proteins) (1, 2). The endogenous ligand for these receptors,
dopamine, is one of the main catecholamine transmitters in mam-
mals (2, 3). Dopamine has been found to play a key role in regulat-
ing a range of neural functions including locomotor behavior,
motivational state, and reward processes (4). In addition, it has
been demonstrated that an increase in dopamine transmission
43
44 N. Kabbani and J.C. Nordman
caused by drugs of abuse can lead to addiction in many species (5).

Given their central role in dopamine transmission, the dopamine
receptor (DR) family represents a key class of molecular targets for
pharmacological therapy (6).
Within mammals, DRs are divided into five types (D1-D5).
Based on their pharmacologic profile and genetic sequence similar-
ity, DRs are subdivided into D1-like and D2-like categories (1, 2).
D1-like (D1 and D5) receptors interact with Gs (stimulant) proteins
to increase the production of cellular cAMP by stimulating adenyl
cyclase at the membrane, while D2-like (D2, D3, D4) receptors
interact with Gi (inhibitory) proteins and decrease the production
of cellular cAMP by inhibiting adenyl cyclase activity also at the
membrane (7–9). In addition to their noted effects on cellular
cAMP production, DRs are known to exert effects on cellular
excitability by regulating various ligand and voltage-gated ion
channels, ion transporters, and the sodium-potassium pump (10).
The activation of D2Rs in various cell types has been found to
impact signaling of cellular molecules such as cAMP regulated
Protein Kinase A (PKA), Protein Kinase C (PKC), and calmodulin
regulated protein such as calcineurin (PP2B) and Ca2+/Calmodulin
Dependent Protein Kinase II (CaMKII) (8). The recent discovery
of diverse dopamine receptor interacting proteins (DRIPs) in cells
(9, 11) suggests that the signaling and regulatory properties of
DRs are mediated via their direct protein interactions in cells.
Indeed, certain DRIPs appear to regulate the major signaling and
regulatory features of the D2R including its transport and localiza-
tion within neurons, stability at the plasma membrane, and down-
stream signaling (7, 8).
A growing demand in understanding molecular functions has
spurred efforts in the design of tools for enhanced detection of
intracellular signaling. In addition to advancements in cellular
imaging, which have provided a mean to examine the spatial aspects
of protein expression and trafficking, mass spectrometry techniques
such as matrix assisted laser desorption/ionization-time of flight
(MALDI-TOF) or liquid chromatography electrospray ionization
(LC-ESI) now allow for the identification of signaling and interac-
tion networks within various compartments of living cells (12).
This enables a rapid (high throughput) strategy for the capture of
information on the dynamics and composition of a signaling pro-
tein network within the cell (13, 14). Mass spectrometry analysis
of affinity captured protein complexes has been effectively used by
our group and others to define high and low abundance interac-
tions as well as small and large size protein binding partners of
neurotransmitter receptors in the brain (15–17). The aim of this
chapter is to provide a step-by-step guide for the isolation and
characterization of D2R complexes (D2Rs + DRIPs) from cells.
4 Proteomics of D2 Receptors 45
Since the striatum is an abundant site of D2R expression, we

describe procedures for detecting D2Rs endogenous to striatal
cells. Information gained from understanding receptor–protein
interactions may aid in the treatment of brain disease.
2. Materials
2.1. Protein 1. Cell Culture Medium: Neurobasal (NB) with 2% B-27 supple-
Preparation from ment (Invitrogen, Life Technologies, Grand Island, NY), 5%
Cultured Cells Horse Serum (Gibco, Life Technologies, Grand Island, NY),
1% Pneumococcal Streptomycin (Gibco).
2. Poly-L-Lysine (Invitrogen).
3. Laminin (Invitrogen).
4. Non-denaturing protein extraction solution: 20 mM Tris–HCl,
pH 7.4, 1% Triton X-100, 2 mM EDTA, 137 mM NaCl, 10%
glycerol, and 1× protease inhibitor cocktail (Roche,
Indianapolis, IN).
5. Sterile, optically clear polystyrene petri dishes.
6. Sterile cell scrapers.
2.2. Protein 1. Cold dissection buffer solution: 4 mM Hepes, 1 mM EDTA,

Preparation 0.32 M sucrose, pH 7.4.
from Brain Tissue 2. Cold cutting buffer: 10 mM Tris, 320 mM sucrose, 1 mM
Phenylmethanesulfonylfluoride (PMSF) and 1× protease inhib-
itor cocktail, pH 7.4 (Roche).
3. Bradford dye reagent kit (Bio-Rad, Hercules, CA).
4. Non-denaturing protein extraction solution (20 mM Tris–HCl
(pH 7.4)), 1% Triton X-100.
(VWR), 2 mM EDTA, 137 mM NaCl, 10% glycerol, 1× pro-
tease inhibitor cocktail (Roche).
5. Glass dounce homogenizer.
2.3. Protein 1. Bovine Serum Albumin (BSA).

Quantification 2. Quickstart Bradford Dye Agent (1×) (Bio-Rad).
3. Spectrophotometer.
2.4. Immunoprecipi- 1. Protein G Dynabeads® (Invitrogen).

tation of Dopamine 2. Magnetic Tube Rack (Invitrogen).
Receptors
3. Phosphate Buffer Saline (PBS).
4. Tween-20.
5. Phosphate Buffer Saline with 0.1% Tween 20 (PBST).
6. Monoclonal Anti-D2R (EMD, Millipore, Billerca, MA).

7. NuPAGE LDS Sample Buffer (4×) (Invitrogen).
8. NuPAGE Sample Reducing Agent (10×) (Invitrogen).
9. Mass Spectrometry elution buffer: 49% acetonitrile, 2%
trifluoroacetic acid in deionized water.
2.5. Mass 1. 10 mM DTT (Sigma-Aldrich).

Spectrometry 2. 20 ng Β-casein (Sigma-Aldrich).
3. 8 M Urea (Sigma-Aldrich).
4. Sequencing grade trypsin (Invitrogen).
5. Glacial acetic acid (Sigma-Aldrich).
6. Angiotensin I & II (Sigma-Aldrich).
7. C-18 Zip-Tip Desalting Columns (Millipore, Billerca, MA).
8. Zip-Tip washing buffer: 0.1% trifluoroacetic acid in deionized
water.
9. Zip-Tip elution buffer: 50% acetonitrile, 0.1% trifluoroacetic
acid in deionized water.
10. 0.1% Formic acid (Thermo Fisher Scientific, Rockford, IL).
11. SpeedVac.
2.6. Protein 1. 2 mM DSP (dithiobis[succinimidylpropionate]) (Pierce,

Cross-linking Rockford, IL).
2. 2.5 mM BS3 (Pierce).
3. Glass dounce homogenizer (5–7 mL volume capacity).
4. DMSO (solvent for use with DSP) (Sigma-Aldrich).
5. 20 mM Tris–HCl, pH 7.5.
2.7. Reverse- 1. Coomassie stain: 50% methanol, 10% acetic acid, 0.25%
Cross-linking and Coomassie Blue R-250 (Sigma-Aldrich) in deionized water.
In-Gel Digestion 2. Destaining Solution: 16.5% ethanol, 5% acetic acid in deion-
ized water.
3. 50 mM ammonium bicarbonate (NH4HCO3).
4. In-Gel Reducing Solution: 25 mM DTT, 50 mM NH4HCO3
in deionized water.
5. In-Gel Alkylating Solution: 20 mM iodoacetamide (dissolve in
500 mM NH4HCO3), 50 mM NH4HCO3 in deionized
water.
6. Drying Solution: 80% acetonitrile, 50 mM NH4HCO3 in
deionized water.
7. In-Gel Digestion Solution: 10 ng/μL in ice-cold 50 mM
NH4HCO3.
8. In-Gel Extraction Buffer: 50% acetonitrile, 2% acetic acid in

deionized water.
2.8. Centrifugation 1. Low to medium speed centrifugation was achieved using the
Eppendorf Centrifuge 5810R series (Eppendorf AG,
Hamburg, Germany).
2. High speed ultracentrifugation was achieved using the Sorvall
WX Ultra Series Centrifuge (Thermo Scientific).
3. A variety of rotors can be used for the described centrifugation
procedures. In our experience, both swinging bucket and fixed
angle rotors work well for the isolation of membrane protein
fractions. In general, swinging bucket rotors are considered
better suited for particle separation and advantageous when
working with smaller volume.
3. Methods
The D2R complex is made up of numerous direct and indirect

interacting proteins. Many of these interactions are dynamic, being
sensitive to disruptions in the state of the D2R during various forms
of experimentation. In particular, molecular and biochemical meth-
ods for the dissociation of tissue, lysis of cells, and the efficient solu-
bilization of membrane bound proteins are all known barriers in the
study of receptor–protein interactions. For this reason we have
emphasized experimental conditions that maximize on the preser-
vation of endogenous protein interactions in the proteomic analysis
of D2R protein complexes within striatal cells. This procedure can
be used for D2R analysis from cultured striatal neurons as well as
native striatal tissue. In addition, we present a cross-linking strategy
that can preserve protein–protein interactions in order to define the
D2R complex within cells. The advantages of this procedure include
preservation of the D2R complex in the live cell, a direct visualiza-
tion of several multi-protein complexes on an SDS-PAGE gel using
a western blot or a Coomassie stain, and an identification of the
proteins which make up those complexes using mass spectrometry
peptide detection.
3.1. Protein 1. Primary cultures of striatal neurons can be derived from late
Preparation from embryonic day 18 embryos or, less preferably, neonate day 0 or 1
Primary Striatal rat or mouse. Techniques for tissue dissection, cellular disas-
Neurons sociation and plating of striatal cells have been well described
in the literature.
2. For proteomic analysis, striatal cells are best plated at medium
density concentration of 100 cells/mm2 in NB/B27 medium
(see Note 1). Cells adhere well onto dishes precoated with
poly-L-lysine (12.5 μg/mL) and laminin (5 μg/mL).
3. Perform medium changes every 2–3 days or as needed and
maintain cells in culture for up to 2 weeks for proteomic analy-
sis (see Note 2).
4. For protein identification, gently remove the culturing medium
and replace with 3 mL of room temperature, sterile, PBS. This
volume is per 100 mm size petri dish; however, when using
different size dishes it possible to scale down/up the volume of
PBS according to size.
5. Gently remove cells using a cell scraper (see Note 3). For less
adherent cells, it is possible to remove cells via a gentle
suction.
6. Pool samples into a 15 mL or a 50 mL conical tube as appro-
priate (Falcon).
7. Spin down the cells at 129 × g for 5 min at 4°C.
8. Carefully remove the supernatant and discard.
9. Wash the cells one time using an ice-cold PBS solution. The
volume of PBS is not very critical; however, we suggest a
volume consistent with the pooled final volume in step 6.
10. Gently wash the cells in PBS by inverting the capped tube once
or twice.
11. Spin down the cells at 129 × g for 5 min at 4°C.
12. Carefully remove the supernatant and discard.
13. Resuspend cells into 5 mL of ice-cold PBS and transfer into a
dounce homogenizer.
14. Manually homogenize the cells using gentle, slow, full
strokes.
15. Do not cause bubbles during homogenization.
16. Homogenize until the solution turns consistent and fairly
smooth. On average five to six strokes are sufficient for the
homogenization of cultured cells.
17. Transfer the homogenate into prechilled centrifugation tubes.
18. Spin down the supernatant at 39,000 × g for 1 h for the isola-
tion of membrane proteins.
19. A visible off white colored pellet should appear at the end of
the ultracentrifugation step. This is the membrane fraction.
20. Discard the supernatant by slowly decanting or pipetting. The
pellet should be tightly bound to the centrifugation tube.
21. Resuspend the pellet in 1 mL of the non-denaturing protein
extraction buffer to solubilize membrane proteins (see Note 4).
22. Triturate the pellet gently, several times, by pipetting up and

down in the extraction buffer.
23. Do not try to break up or dissolve the entire pellet in the
buffer.
24. Incubate the pellet in the protein extraction buffer for 1 h (or
overnight) at 4ºC with gentle mixing.
25. Centrifuge at 12,851 × g for 20 min at 4ºC.
26. Remove and keep the supernatant. This is the triton soluble
membrane protein fraction containing D2Rs and their inter-
acting proteins.
27. Optional: retain the pellet and store at −20ºC for up to
3 months. This pellet corresponds to the triton insoluble frac-
tion, which maybe enriched in subcellular domains such as
lipid rafts and postsynaptic densities.
28. Determine the final protein yield and concentration using the
Bradford kit (Bio-Rad) in accordance with the manufacturer’s
protocols and the established method of Bradford (18).
3.2. Protein 1. Fresh striatal tissue can be obtained from either rat or mouse
Preparation from brains of various ages depending on the experimental para-
Striatal Tissue digm. Elaborate protocols on how to dissect the striatum of
rodents can be found elsewhere.
2. For high fidelity protein analysis, rapidly place the striatal tissue
into a freshly made cold dissection buffer solution kept on
ice.
3. We recommend a 1:1 ratio of rat striatum to dissection buffer
(i.e., pool 4 whole striata of adult rats into a 4 mL solution of
cold dissection buffer).
4. Transfer the tissue (in the dissection buffer) into a prechilled
dounce homogenizer.
5. Keep the homogenizer on ice throughout the procedure.
6. Manually homogenize the tissue using gentle, slow, full
strokes.
7. Do not over stroke or cause bubbles during homogenization.
8. Homogenize until the solution turns consistent and most of
the large tissue pieces have broken.
9. On average eight to ten strokes are sufficient for the homoge-
nization of brain tissue.
10. Transfer the homogenate into a prechilled centrifugation
tube.
11. Spin down the homogenate at 700 × g for 10 min at 4°C.
12. Collect the supernatant fraction (S1) and store on ice during
steps 13–18.
13. Resuspend the pellet in the same volume of cold dissection as

before (step 4).
14. Rehomogenize the pellet as before (steps 5–9).
15. Transfer the homogenate into a prechilled centrifugation
tube.
16. Spin down the homogenate at 700 × g for 10 min at 4°C.
17. Collect the supernatant fraction (S2) and keep on ice.
18. Combine the supernatant fractions (S1 + S2) in prechilled
ultracentrifugation tubes. This is the total striatal
homogenate.
19. Ultracentrifuge the S1 + S2 sample at 40,000 × g for 1 h at
4°C.
20. A visible pellet should appear at the end of the ultracentrifuga-
tion step. This is the cellular membrane fraction from striatal
tissue.
21. Discard the supernatant by slowly decanting or pipetting. The
pellet should be tightly bound to the centrifugation tube.
22. Resuspend the pellet in non-denaturing protein extraction
buffer to solubilize membrane proteins. We recommend a vol-
ume of 2 mL extraction buffer for four whole striata of adult
rats for high protein yield content.
23. Triturate the pellet gently, several times, by pipetting up and
down in the extraction buffer and proceed through the solubi-
lization procedure of the membrane fraction as previously indi-
cated in Subheading 3.1.
3.3. Coimmunoprecipi- 1. Place a 1.5 mL microcentrifuge tube on a magnetic rack and

tation of Dopamine add 50 μL of Protein G Dynabeads to the tube.
Receptor Complexes 2. Add 500 μL PBST to the bead matrix and wash the matrix a
total of three times in this volume of PBST. Remove the tube
from the rack during each wash and gently invert the tube sev-
eral times to wash thoroughly.
3. Use the magnetic tube rack to sediment the beads and care-
fully remove the PBST making sure not to remove any of the
bead resin (see Note 5).
4. Add 5 μg of the monoclonal anti-D2R (in 500 μL PBST) onto
the bead matrix.
5. Incubate on a rocking platform for 20 min at room tempera-
ture (see Note 6). This will allow the antibodies to bind to the
beads (Fig. 1a).
6. Place the tube on the magnetic rack for 1 min and carefully
remove the supernatant making sure not to disturb the bead
matrix.
Fig. 1. Major steps in the immunoprecipitation of the D2R complex using the batch method.
(a) Immobilization of the antibody bait on a bead matrix: Prewashed agarose Protein A/G
Dynabeads are incubated with purified monoclonal antibodies. Protein A/G selectively
binds to the IgG region of the antibody. (b) The antibody–bead matrix is incubated with a
solution of solubilized membrane proteins expressing D2R complexes. D2R complexes
(D2R + DRIPs) are then selectively immunoprecipitated from the solution.
7. Wash the beads with 500 μL PBST a total of three times as

described above in steps 2–3.
8. To immunoprecipiate the D2R protein complex, add 100 μg
of solubilized membrane proteins from primary striatal cul-
tures and 50–100 μg of solubilized membrane proteins from
striatal tissue to the bead matrix (proteins obtained from
Subheadings 3.1 and 3.2).
9. For immunoprecipitation controls, we highly recommend per-
forming the same procedures in mice lacking the D2R in order
to gain confidence in the specificity of the anti-D2R antibody
and the results of the immunoprecipitation procedure.
10. Adjust the total volume of each sample tube to 500 μL using
PBST.
11. Place the tube on a rocking platform for 20–30 min at room
temperature to allow the bead complex to bind D2Rs and their
associated proteins (Fig. 1b). Alternatively, it is possible to
incubate the immunoprecipitation overnight at 4°C on the
same rocking platform.
12. Use the magnetic tube rack to carefully remove unbound
supernatant and isolate receptor–antibody–bead complex.
13. Thoroughly wash the bead complex with 500 μL PBST a total
of five times (with 1 min incubation on ice in between washes)
(see Note 7).
14. Make sure to gently invert the tube to assure complete wash-
ing at every step.
15. To elute the protein complex from the Protein G Dynabeads,
add 30 μL of a premixed elution buffer containing 25% LDS,
10% reducing agent (Invitrogen) and 65% room temperature
PBS to the bead matrix.
16. Alternatively if mass spectrometry is necessary use an elution
buffer consisting of 49% acetonitrile + 2% trifluoroacetic acid in
deionized water, and proceed as below.
17. Gently pipette the elution solution with the bead matrix sev-
eral times to make sure they mix well.
18. Incubate the beads in the elution buffer for 10 min at 70ºC.
This should dissociate the antibody–receptor complex from
the Dynabeads.
19. Place the solution back onto the magnetic rack and aim to
recover the entire eluted solution sparing the Dynabeads from
the final eluant.
20. At this point you can analyze your eluant sample using mass
spectrometry or a western blot method (Fig. 2).
3.4. Mass Immunoprecipitated samples must be processed for mass spec-

Spectrometry trometry analysis. Depending on the ion source, different mass
Identification spectrometry methods require the sample to be processed into
of Dopamine Receptor various concentrations, volume, and composition. In addition, for
Interacting Proteins mass spectrometry the protein must be cleaved into peptides before
analysis using either in-gel digestion or proteolysis in solution
procedures. Described here is the method we have used for the
preparation of the immunoprecipitated D2R complex for LC-ESI
mass spectrometry using an LTQ Orbitrap (Thermo Fisher). This
instrument was coupled to a computer that ran the SEQUEST
software (BioWorks software from Thermo Fisher, 3.3.1) in order
to identify the proteins in the NCBI protein database. This proce-
dure is summarized in Fig. 3.
Striatal cell Striatal

culture tissue
Membrane protein Protein Membrane protein

preparation crosslinking preparation
& solubilization in situ & solubilization
D2R Protein Complex

Immunoprecipitation
LC-ESI/
SDS-PAGE MALDI-TOF
electrophoresis Mass spectrometry
Gel
Extraction
Reverse
crosslinking
Western Blot Identification of

Confirmation of Interacting
Interactions Proteins
D2R Protein
Complexes
Fig. 2. A flow chart showing proteomic strategies for the identification of the D2R complexes from cultured cells and
native brain tissue.
1. Reduce the eluted D2R complex by incubating the entire

sample in 10 mM DTT, 20 ng of β-casein, and 8 M urea for
1 h at room temperature.
2. Since protein identification is enhanced by protein abundance,
we strongly recommend using most, if not the entire, immu-
noprecipitated sample from an experiment for mass spectrom-
etry analysis.
Fig. 3. Identification of proteins that coimmunoprecipitate with the D2R using mass spectrometry. (a) Enzymatic digestion
and processing of immunoprecipitated D2R complexes. (b) LC separation of peptide fragments using a C18 chromatogra-
phy column. The sample is then emitted towards the sensor for ESI mass spectrometry. Ions are fragmented by collision
induced dissociation (CID). (c) Tandem mass spectrometry ESI yields a mass to charge (m/z) ion spectra. (d) The fragment
ion spectra are assigned peptide sequences based on database comparison using SEQUEST software. An example of D2R
interacting proteins identified using this approach. (e) Mass spectrometry proteomics can be used to generate data for
determining D2R signaling and interactions within cells.
3. Alklyate the sample with 50 mM iodoacetamide for 20 min in

the dark at room temperature.
4. Enzymatically digest sample with 0.5 μg sequencing grade
trypsin in 50 mM NH4HCO3 (pH 8) at 37°C overnight.
5. Quench the trypsin digestion using 2 μL of 98–100% grade
glacial acetic acid.
6. For detection of phosphorylated proteins, add 500 fmol
Angiotensin II to the sample. If phosphorylated proteins are
not of concern, skip this step.
7. Desalt samples using C-18 elution Zip-Tips (Millipore).
8. Load 20μL of the solution into one Zip-Tip at a time.
9. Activating Zip-Tips with Zip-Tip elution buffer 2× (see
Note 8).
10. Wash Zip-Tip in Zip-Tip washing buffer.
11. Repeat the washing three times.

12. Elute with Zip-Tip elution buffer, leaving a final eluted volume
of 20–30 μL (see Note 9).
13. Dry sample in a SpeedVac concentrator until final volume is
approximately 5 μL (see Note 10).
14. Reconstitute dried samples in 1–5 μL of 0.1% Formic Acid.
15. Add 100 fmol Angiotensin I (see Note 11).
3.5. Cross-linking It is possible to cross-link protein complexes thereby capturing a

of Protein Complexes “snapshot” of the receptor interaction network in space and time
from Cells of Culture during an experiment. A number of cross-linking tools have been
or Tissue Origin developed that complement the range of information that can be
gained by proteomic analysis. Here we describe three main
approaches that can significantly aid in the study of D2R com-
plexes within various cell types.
3.5.1. DSP Cross-linking This method allows for DSP cross-linking of proteins in primary
of D2R Complexes cultures or brain tissue (Fig. 4). Because DSP is membrane-perme-
in Live Cells able, it has been used to cross-link proteins within living cells under
various experimental procedures (19, 20). Below we describe a
method for DSP cross-linking within cultures of primary striatal
neurons. This method can be used for the detection of D2R pro-
tein complexes in living cells under various experimental condi-
tions. Once cross-linked, the proteins can be directly visualized on
an SDS-PAGE gel using a Coomassie stain. In addition, cross-
linked proteins can be identified using a western blot or a mass
spectrometry method.
1. Prepare a 10 mM DSP stock solution by dissolving DSP into
DMSO.
2. Dilute the DSP stock solution into PBS (1:10 dilution) to
create a working solution of 1 mM DSP.
3. Gently remove the culture medium from cultured striatal neu-
rons and wash the cells once with 5 mL of room temperature
PBS.
4. Gently apply 3 mL of the 1 mM DSP solution into each dish
of cells. We recommend using confluent 100 mm petri dishes
for protein analysis since protein yield is rate limiting for accu-
rate detection. This volume of DSP is just enough to cover the
surface area of the dish; however, higher or lower volumes can
be used.
5. Incubate the cells in DSP solution for 2 h at 4°C. We recom-
mend occasional gentle mixing to ensure that the cells are
covered in solution throughout the incubation.
6. A white precipitate will form. This is normal.
Fig. 4. Intracellular protein cross-linking of the D2Rs and their interacting proteins. DSP cross-linking provides a method
that facilitates in the identification of weak and/or transient receptor–protein interactions in cells. (a) The chemical struc-
ture of DSP promotes its association with free amine (NH2) groups on various proteins. The proteins need to be in close
proximity for chemical cross-linking to occur. (b) A model of a cross-linked D2R protein complex at the plasma membrane.
Since DSP is membrane-permeable, the D2R complex can be cross-linked within living cells. The DSP cross-linker can be
reversed with the addition of DTT thereby providing a tool for studies on the dynamics of the D2R complex under various
conditions.
7. End the DSP reaction by diluting the 2× quenching buffer

(50 mM Tris–HCl, pH 7.5) to a final concentration of 25 mM
Tris–HCl, pH 7.5 for 30 min at 4°C. If you used 3 mL DSP
solution per dish, then add the same volume of quenching
buffer (3 mL) to the dish.
8. Remove cells from the dish using a cell scraper.
9. Collect cells in a prechilled 15 mL conical Falcon tube and
keep cells on ice throughout.
10. Spin cells down at 120 × g for 10 min at 4°C.
11. Discard the supernatant and keep the pelleted cells on ice.
12. Cultured cells can now be manually homogenized and
processed for membrane preparation as described in
Subheading 3.1.
13. Cross-linked D2R complexes can then be isolated from mem-
brane preparations using the immunoprecipitation protocol
(see Subheading 3.3), and then analyzed by western blot or

mass spectrometry (see Subheading 3.4).
14. If using gel electrophoresis, it is important to know that the
addition of DTT or β-mercaptoethanol can reverse the DSP
cross link.
3.5.2. BS3 Cross-linking Solubilized membrane proteins derived from either cultured striatal
of Membrane Proteins cells or tissue can be cross-linked in vitro prior to immunoprecipi-
tation with an anti-D2R antibody. In this case, cross-linking of
proteins, prior to immunoprecipitation, provides a way for preserv-
ing weak or transient protein–protein interactions that can be
lost during the immunoprecipitation procedure. This method
may also be useful in eliminating nonspecific interactions during
an experiment.
1. Prepare a 5× BS3 stock solution by dissolving BS3 into ice-cold
deionized water to a concentration of 12.5 mM.
2. Prior to the immunoprecipitation experiment (see
Subheading 3.3), add BS3 to the membrane protein fraction at
a final concentration of 2.5 mM.
3. Incubate the BS3 with the membrane proteins for 2 h at 4°C
with gentle mixing.
4. Quench the BS3 reaction by adding 250 mM Tris–HCl, pH
7.5 to a final concentration of 25 mM Tris–HCl, pH 7.5.
5. Mix the solution for 30 min at 4°C with gentle mixing.
6. Cross-linked samples can now be immunoprecipitated as
described above (see Subheading 3.3) followed by analysis
using western blot or mass spectrometry (see Subheading 3.4).
3.5.3. Reverse- The use of the membrane-permeable, reversible cross-linker DSP

Cross-linking of Protein provides a strong tool in the proteomic analysis of signaling path-
Complexes from ways in cells (19, 20). Protein complexes can be directly visualized
Samples Immobilized on an SDS-PAGE gel using a Coomasie stain. The bands can then
in SDS-PAGE Gel be manually excised and interacting proteins identified using a
reverse cross-linking approach. Here we summarize the steps
involved in reversing the cross-linking of proteins within an SDS-
PAGE gel. This technique has proven useful in detecting specific
changes within a protein–protein interaction network under vari-
ous experimental conditions.
1. A Coomassie-stained SDS-PAGE gel loaded can be processed
for reverse cross-linking using DTT.
2. To do this, the Coomassie-stained gel should be placed in a
plastic container and incubated with a gel in a Destaining
Solution for 1–2 h at room temperature, on an orbital shaker
at a low speed setting.
3. Once protein bands become visible, manually excise the bands

of interest from the Coomassie-stained SDS-PAGE gel. The
bands of interest will be established by the experiment. In our
experiments, we have compared immunoprecipitated D2R
complexes between cross-linked and non-cross-linked cellular
samples. This method has allowed us to identify a series of
higher molecular weight bands, within the cross-linked sample,
that have been identified as direct binding proteins.
4. Cut the desired gel band into 1 mm × 1 mm pieces, placing
individual bands into separate 1.5 mL microcentrifuge tubes.
5. Destain the gel pieces further by adding 500 μL of a Destain
Solution to each tube.
6. Incubate the tubes at 37°C for 15 min with gentle mixing on
a nutator.
7. If after 15 min the gel piece is still blue, dispose of the Destain
solution and replenish with a fresh batch of Destain solution at
the same volume as before.
8. Once the gel piece is destained (a light blue color will remain),
apply 500 μL of the In-Gel Reducing Solution to each tube.
9. Incubate each tube at 37°C for 30 min with gentle mixing.
This is the reverse-cross-linking step.
10. Discard the In-Gel Reducing Solution but slow decanting and
then apply 500 μL of In-Gel Alkylating Solution per tube.
11. Incubate the tubes at room temperature for 20 min in the dark
with gentle mixing.
12. Discard the In-Gel Alkylating Solution by slow decanting and
then add 500 μL NH4HCO3 to each tube.
13. Incubate the tubes at room temperature for 5 min with gentle
mixing.
14. Discard the NH4HCO3 then add 500 μL of the Drying
Solution.
15. Incubate the tube(s) for 15 min at room temperature over
gentle mixing.
16. Remove the Drying Solution by slow decanting and then dry
the gel pieces within the microcentrifuge tubes for 20 min
using a SpeedVac (at low setting).
17. After drying, place the tubes on ice then add 50 μL of In-Gel
Digestion Solution.
18. Incubate the samples for 30 min at 4°C.
19. Transfer the tubes into a 37°C water bath and incubate the
samples overnight. This should free most of the proteins from
the gel matrix.
20. The following day pellet the gel pieces using a quick spin at
room temperature.
21. Transfer the supernatant (S1) to a separate 1.5 mL microcen-
trifuge tube.
22. Add 30 μL of the In-Gel Extraction Buffer to the remaining
gel pieces and incubate for 15 min at room temperature.
23. Pellet the gel pieces using a quick spin.
24. Obtain the supernatant (S2) and combine with S1.
25. SpeedVac the pooled fractions (S1 + S2) at a slow speed to a
volume of 20 μL (see Note 12).
26. Proceed with mass spectrometric analysis as previously described
(see Subheading 3.4) to identify interacting proteins within
the cross-linked complex.
4. Notes
1. Primary striatal cultures are naturally composed of neuronal

and non-neuronal cells. Since D2Rs are expressed in glia, as
well as neurons, we have added 10 μM fluoro-5¢-deoxyuridine
(Sigma) during the first 4 days of culturing in order to mini-
mize glial proliferation.
2. DMEM medium must be room temperature when being intro-
duced to cells. DMEM viability is colorometric: fresh DMEM
is red; overtly basic (high pH) DMEM is purple; overtly acidic
(low pH) DMEM is yellow.
3. Cells should detach from surface with light pressure. Make
sure not to apply too much force as this could damage cell
integrity.
4. For optimal results, ratio of cells to non-denaturing protein
extraction buffer should be 1:2.
5. Taking care to avoid disturbing beads or bead–antibody–
protein complex to ensure that the immunoprecipitated sample
is consistent between different experimental groups.
6. For the immunoprecipitation incubation steps, a nutating
mixer works well.
7. During the last wash and before the elution of protein com-
plexes, a quick centrifugation step is recommended.
8. Avoid creating bubbles when desalting as this compromises
the integrity of the filter.
9. This can be accomplished by simply resuspending the Zip-Tip
sample in the same 20 μL volume of Zip-Tip elution buffer.
10. SpeedVac time using a 20 μL volume is about 3–5 min.

11. Samples can be frozen at this point for up to 1 week before
processing; however, immediate proteomic analysis is most
preferable.
12. SpeedVac time should be about 5 min.
References
1. Missale C, Russel N, Robinson SW, Jaber M, dopamine signaling. Trends Pharmacol Sci
Caron MG (1998) Dopamine receptors: from 24:486–492
structure to function. Physiol Rev 12. Downard KM (2006) Ions of the interactome:
78:189–225 the role of MS in the study of protein interac-
2. Sidhu A, Niznik HB (2000) Coupling of dop- tions in proteomics and structural biology.
amine receptor subtypes to multiple and diverse Proteomics 6:5374–5384
G proteins. Int J Dev Neurosci 18:669–677 13. Baerenfaller K, Grossman J, Grobei MA, Hull
3. Binda AV, Kabbani N, Levenson R (2005) R, Hirsch-Hoffmann M, Yalovsky S,
Regulation of dense core vesicle release from Zimmermann P, Grossninklaus U, Gruissem
PC12 cells by interaction between the D2 dop- W, Baginsky S (2008) Genome-scale proteom-
amine receptor and calcium-dependent activa- ics reveals Arabidopsis thaliana gene models
tor protein for secretion (CAPS). Biochem and proteome dynamics. Science 320:
Pharmacol 69:1451–1461 938–941
4. Goldman-Rakic PS (1998) The cortical 14. Hinkson IV, Joshua EE (2011) The dynamic
dopamine system: role in memory and cogni- state of protein turnover: it’s about time.
tion. Adv Pharmacol 42:707–711 Trends Cell Biol 21(5):293–303
5. Schultz W (2002) Getting formal with 15. Santos SD, Manadas B, Duarte CB, Carvalho
dopamine and reward. Neuron 36:241–263 AL (2010) Proteomic analysis of an interac-
6. Strange PG (2001) Antipsychotic drugs: tome for long-form AMPA receptor subunits.
importance of dopamine receptors for mecha- J Proteome Res 9:1670–1682
nisms of therapeutic actions and side effects. 16. Xiao K, McClatchy DB, Shukla AK, Zhao Y,
Pharmacol Rev 53:119–133 Chen M, Shenoy SK, Yates JR 3rd, Lefkowitz
7. Obidiah J, Avidor-Reiss T, Fishburn CS, RJ (2007) Functional specialization of beta-
Carmon S, Bayewitch M, Vogel Z, Fuchs S, arrestin interactions revealed by proteomic
Levavi-Sivan B (1999) Adenylyl cyclase interac- analysis. Proc Natl Acad Sci U S A 104:
tion with the D2 dopamine receptor family; 12011–12016
differential coupling to Gi, Gz, and Gs. Cell 17. Free RB, Hazelwook LA, Cabrera DM,
Mol Neurobiol 19:653–664 Spalding HN, Namkung Y, Rankin ML, Sibley
8. Neve KA, Seamans JK, Trantham-Davidson H DR (2007) D1 and D2 dopamine receptor
(2004) Dopamine receptor signaling. J Recept expression is regulated by direct interaction
Signal Transduct Res 24:165–205 with the chaperone protein calnexin. J Biol
9. Kabbani N, Levenson R (2007) A proteomic Chem 282:21285–21300
approach to receptor signaling: molecular 18. Bradford MM (1976) Rapid and sensitive
mechanisms and therapeutic implications method for the quantification of microgram
derived from discovery of the dopamine D2 quantities of protein utilizing the principle of
receptor signalplex. Eur J Pharmacol protein dye binding. Anal Biochem
572:83–93 72:248–254
10. Bertorello AM, Hopfield JF, Aperia A, 19. Ollom CM, Denny JB (2008) A crosslinking
Greengard P (1990) Inhibition by dopamine of analysis of GAP-43 interactions with other pro-
(Na+ + K+) ATPase activity in neostriatal neu- teins in differentiated N1E-115 cells. Int J Mol
rons through D1 and D2 dopamine receptor Sci 9:1753–1771
synergism. Nature 347:386–388 20. Thomas DDH, Taft WB, Kaspar KM,
11. Bergson C, Levenson R, Goldman-Rakic PS, Groblewski GE (2001) CRHSP-28 regulates
Lidow MS (2003) Dopamine receptor-inter- Ca2+ stimulated secretion in permeabilized
acting proteins: the Ca(2+) connection in acinar cells. J Biol Chem 276:28866–28872
Chapter 5
Modeling Spatial Aspects of Intracellular

Dopamine Signaling
Kim T. Blackwell, Lane J. Wallace, BoHung Kim, Rodrigo F. Oliveira,
and Wonryull Koh
Abstract
Dopamine binding to various dopamine receptors activates multiple intracellular signaling molecules,
some of which interact with calcium activated signaling pathways. Many experiments measure agonist-
stimulated elevations in signaling molecules using prolonged, diffuse application, whereas the response of
neurons to transient and spatially localized stimuli is more important. Computational modeling is an
approach for investigating the spatial extent, time course, and interaction of postsynaptic signaling mole-
cules activated by dopamine and other transmembrane receptors. NeuroRD is a simulation algorithm
which can simulate large numbers of pathways and molecules in multiple spines attached to a dendrite. We
explain how to gather the information needed to develop computational models, to implement such mod-
els in NeuroRD, to perform simulations, and to analyze the simulated data from these models.
Key words: Computer model, Signaling pathways, Dopamine signaling, Reactions, Diffusion
1. Introduction
Dopamine binding to various dopamine receptors activates

multiple intracellular signaling molecules. The importance of
dopamine activated signaling pathways is evident from
Parkinson’s disease, a degenerative disease in which dopamine
neurons of the substantia nigra pars compacta degenerate. The
lack of dopamine input from the substantia nigra pars compacta
to the basal ganglia produces abnormalities in voluntary motor
activity (1). In vivo recordings have shown that the lack of
dopamine alters the firing dynamics of these neurons and the
striatum as a whole (2). In vitro experiments have shown that
61
62 K.T. Blackwell et al.
dopamine modulates several voltage dependent and synaptic

channels of striatal spiny projection neurons, and that dopamine
activated signaling pathways interact with glutamate and acetyl-
choline activated pathways to produce neuromodulation (3).
Most in vitro experiments demonstrating the effect of dopamine
on signaling pathway molecules or ionic channels use prolonged,
diffuse application of agonists. Because of adaptations such as
receptor desensitization, inferences made from prolonged, dif-
fuse application might not accurately represent the effect of
transient “in vivo like” inputs.
Computational modeling is an approach for investigating the
spatial extent, time course and interaction between dopamine
activated and other signaling pathways (4). Computational model-
ing allows integration of information obtained from biochemical,
pharmacological, and molecular biology experiments with infor-
mation obtained from electrophysiology and calcium imaging
experiments. Simulation experiments using the resulting integra-
tive model provides insight into mechanisms underlying neuro-
modulatory effects in response to realistic synaptic inputs, and can
generate testable hypotheses.
2. Materials
Several software packages are available for modeling signaling

pathways (5–8). They differ in terms of user interface and simula-
tion algorithm. The computational efficiency of the various algo-
rithms depends both on the level of biophysical detail and the
numerical approximations made. The computational efficiency
then constrains the spatial scale and complexity of the signaling
network that can be practically investigated.
The example provided here uses NeuroRD (5), which has a
simple, text-based user interface, and is freely available (http://
krasnow1.gmu.edu/CENlab/software.html). The simulation
algorithm can simulate large numbers of pathways and molecules
in multiple spines attached to a dendrite. The model is defined in a
set of files, each specifying a different attribute of the model: mor-
phology, reaction, initial conditions, stimulation protocol, and
output, and one additional file which specifies the name of the
other files. The following procedure explains how to gather the
information needed to create the model files, and how to run and
analyze the resulting simulation output.
5 Modeling Dopamine Signaling 63
3. Methods
3.1. Signaling The set of reaction pathways is typically determined from multiple
Pathways experiments, some of which demonstrate that blocking specific
molecules prevents a downstream effect, and some of which dem-
3.1.1. Identify Bimolecular
onstrate that two substrate molecules interact to create a product.
and Enzymatic Reactions
Perusal of published literature might be aided by consultation of
that Form the Signaling
online databases.
Pathways of Interest
3.1.2. Find Measures 1. Assays of bimolecular reactions typically provide KD = KB/KF.

of Rate Constants from If separate estimates of KB or KF are not provided, it is neces-
Various Biochemical sary to estimate the rates of forward and reverse reactions.
Experiments 2. Enzyme reactions consist of a pair of reactions in which the
enzyme and substrate form a complex in the first reaction and
free enzyme is regenerated in the second reaction. Under cer-
tain conditions (constant enzyme concentration, irreversible),
which usually apply in enzyme assays, product formation is
given by Eq. 1,
dP K cat ·S
= (1)
dt S + K M
where KM = (KB + Kcat)/KF. Assays typically measure enzyme

activity at various substrate concentrations and provide KM and
Kcat values (9). When a separate estimate of KB is not provided,
it is assumed that KB = 4⋅Kcat (10).
3. One method of estimating rate constants is to create simula-
tions of experimental assays and adjust the rate constants in the
assay simulations until the output is a reasonable match to
experimental assay data. Experimental assays that show dose
response and/or time course relationships are the best. After
choosing the experiment assays to simulate, a separate model is
created to match the conditions of the assay. Enzyme assays are
usually run for many minutes to provide time to accumulate
sufficient substrate to measure. If conditions are such that
product accumulation is a linear function of time, a few sec-
onds of simulation will yield the same enzymatic rate as many
minutes of simulation. In using results from simulations that
run for a few seconds, one must be certain that the system is at
steady state for all species except reaction substrate and prod-
uct. Typically, there are many sets of rate constant values that
will produce a simulation output that matches experimental
data for a single assay. Although the same simulation output
is obtained from these various sets of values, the time for vari-
ous species in the system to reach steady state will be different.
If one can find several different kinds of time course experiments

to simulate, these can be used to constrain the final choice of
rate constants.
4. Constants describing interactions of drugs with various com-
ponents of the system can be estimated from functional data.
When the concentration of agonist in an assay is low, the IC50
concentration for an antagonist is often a reasonable estimate
of the dissociation constant for antagonist binding. If the con-
centration of agonist is higher and the agonist’s dissociation
constant is known, then the Cheng Prusoff equation (11, 12)
is used to calculate the dissociation constant: KI = IC50/
(1 + (substrate/KS)), where KS is the dissociation constant for
the substrate. For agonists, the situation is more complex:
because of signal amplification following interaction with
receptor, EC50 values are often substantially lower than dis-
sociation constants. Thus, EC50 values represent a lower
boundary on the possible range of dissociation constants for an
agonist. Dissociation constants also can be obtained from radi-
oligand binding data. The vast majority of the radioligand
binding studies reported involve equilibrium binding method-
ology. These provide excellent data relative to affinity of inter-
actions but do not provide information on rates of association
and dissociation. Although association and dissociation studies
can be done using radioligand binding methods, they are not
commonly used.
5. Several databases tabulate rate constants and literature sources
of these rate constants (see Note 1). The modeler is always
encouraged to look into the original references and protocols
when trying to understand the source of possible
discrepancies.
3.1.3. Find Measures In some cases the diffusion constant of a molecule in the cytosol is
of Diffusion Constants available. In most cases, it is necessary to estimate diffusion constant
from molecular weight and viscosity of cytosol using Eq. 2 (5)
D = 8.34e −8 ·T (2)
η·M 1/3
where the diffusion coefficient D is in cm2/s, T is temperature in

K, the solution viscosity, η, is in cP (estimated at 1.2–1.4 for
cytosol), and molecular weight M is in g/mol.
3.1.4. Create the The reaction file has two parts. The first part lists all molecular
ReactionScheme File that species. Each molecular species is defined by four attributes: name,
Lists Molecules, Diffusion id, diffusion constant, and units for the diffusion constant. The
Constants, and Reactions second part of the reaction file lists all reactions (Fig. 1). Each
reaction has six attributes: reaction name, id, reactant, product,
Fig. 1. DopamineReac.xml. ReactionScheme file listing molecules, diffusion constants,

and reactions.
forward reaction rate, and reverse reaction rate. Either one or two
reactants and products can be specified. Enzyme reactions are
specified as two bimolecular reactions, with the enzyme regener-
ated in the second step. Additional details are found in the
README file available on the Web site. For example, when the
D1 type of dopamine receptor binds to dopamine, the receptor
becomes an active enzyme that catalyzes the activation of the Gαolf
variety of G protein. The GαolfGTP activated subunit binds to and
activates adenylyl cyclase, which catalyzes the production of cAMP

from ATP. The following reactions describe the first steps of this
pathway:
Da + DIR DaDIR
DaDIR + G G − DaDIR → DaDIR + Ga Olf GTP
Ga Olf GTP → Ga Olf GDP
Reactions are represented by the ReactionScheme file (Fig. 1).

Each enzymatic reaction requires specification of two reactions;
thus forward Rate in the third reaction is turnover or catalytic rate
for the enzymatic reaction.
3.2. Morphology At present, due to computational burden, only a part of the neuron
can be simulated in a reasonable time, e.g., a dendrite with several
3.2.1. Describe the
spines, or several dendrites. The morphology of neurons is avail-
Morphology of the Neuron
able from many publications and various databases. Neuromorpho.
to be Modeled
org provides morphology files of many different neuron types from
many species and brain regions in a standard format in which a
traced neuron is given as a set of points (x, y, z coordinates) with
associated radii. Alternatively, the morphology of whole neurons
or parts of neurons is often available from immunocytochemistry
figures in publications. Lengths and radii of discrete segments can
be extracted from the figures. For simulations of experimental
assays, it is advisable to use a single large compartment.
3.2.2. Create the The NeuroRD morphology file (Fig. 2) specifies the shape of the
Morphology File neuron by specifying the shape of all segments and how they are
connected. Each segment has four attributes: id, region, start and
end. The id attribute is required and must be unique whereas the
optional region attribute does not have to be unique. Regions are
used to group segments with the same initial conditions. The start
and end attributes are specified using x, y, z, (r)adius, and label.
The label is optional and can be used as a site where molecules are
injected into the system. In general, segments are specified with
the starting x, y, z coordinates and radius, and the ending x, y, z
coordinates and radius. To connect the second segment to the first,
it must start from the endpoint of the first compartment using the
attribute start on (Fig. 2) (see Note 2).
3.2.3. Include Spine SpineType specifies a spine prototype and SpineAllocation (Fig. 2)
Specifications in the applies the spine prototype to the surface of a structure. This allows
Morphology File for random placement of spines according to a specified density in
a constrained region or segment of the defined morphology.
Multiple spine prototypes can be defined, e.g., to randomly dis-
tribute long, thin spines among short, stubby spines. SpineType
has an id attribute and is defined using multiple section elements,
Fig. 2. DopamineMorph.xml. Specifications in the morphology file. An example of two dendrites attached to the soma, with
spines on both dendritic branches.
which have four attributes: width, at, regionClass, and label.

The width attribute specifies the radius of that section and the at
attribute indicates the distance from the dendrite at which that
radius begins to apply. The regionClass and label attributes are
optional. SpineAllocation has four attributes: id, spineType, region,
and lengthDensity. The region attribute indicates the region to
which spines of type spineType will be added. The lengthDensity
attribute is the average number of spines per micron. Alternatively,
areaDensity, which is the number per unit area, can be used instead
of lengthDensity (Fig. 2).
3.3. Molecule 1. Find measures or estimates of molecule quantities and, for non-
Quantities and diffusible molecules, their subcellular distribution. For exam-
Initial Conditions ple, in spiny neurons which have glutamatergic synapses on the
spines, metabotropic glutamate receptors are located in the
spine head. Plasma membrane pumps are located in the plasma
membrane and do not diffuse through the cytosol. Once con-
trol conditions are selected, it is possible to determine changes
in molecule quantities that will be used to run simulation
experiments. For example, to simulate pharmacological block
of a particular molecule, its value could be set to zero.
Then, results of the control simulation and the blocked simula-

tion can be compared to evaluate the role of that blocked
molecule.
2. For quantities that are not available, use experiments that
measure time course of a molecule in the signaling pathway as
a constraint. For example, if the amount of substance produced
by a stimulation is known and this substance is produced by an
enzyme that has been purified and tested so that its Kcat is
known but its concentration in the tissue of interest is not
known, one can calculate the minimum concentration of
enzyme that must be present to produce the response. Also,
sufficient substrate must be present to generate the products of
enzymatic reactions. The location of various molecules often is
determined from immunogold labeling (13) or subcellular
fractionation techniques (14).
3. These values are used to create the InitialConditions file, which
specifies the starting concentration or density for each mole-
cule. The file must contain one general ConcentrationSet,
which applies to everything unless overridden. Each
NanoMolarity element names the species and provides a value
for its concentration, entered in nanoMoles per liter. In addi-
tion, non-diffusing molecules can be assigned to a specific
region using additional ConcentrationSets together with a
region attribute. This should correspond to a specified region
from the morphology file and indicates the parts of the structure
to which the initial conditions apply. For membrane localized
molecules, it is possible to specify initial conditions as a density
(picomoles per square meter) using the SurfaceDensitySet
which places these molecules only in the submembrane areas
of the morphology. If the region attribute is specified, then
that submembrane initial condition applies only to that region.
If the region attribute is not specified, then the initial condition
applies to all submembrane areas. In Fig. 3, the dopamine
receptor (D1R) density is zero in general, but non-zero in the
submembrane part of the dendrite region.
3.4. Stimulation 1. Determine the pattern of stimulation for a particular experi-

ment. For example, a brief electrical stimulation leads to release
of dopamine from terminals (15), which then binds to post-
synaptic dopamine receptors. Sometimes, a molecule down-
stream from the experimental stimulation can be used for the
model stimulation, such as with NMDA type of glutamate
receptors. In this case each glutamate release event and bind-
ing to receptors may be approximated as influx of calcium, and
stimulation at high frequency would produce accumulation of
individual calcium transients. Note that neither presynaptic
terminal nor diffusion of dopamine in the extracellular space
can be modeled presently.
Fig. 3. DopamineIC.xml. Receptor quantities and initial conditions.
Fig. 4. DopamineStim.xml. StimulationSet file specifying time and location for injection of
molecules during a simulation.
2. The StimulationSet file specifies the time and location for

(optional) injection of molecules during a simulation (Fig. 4).
Each InjectionStim specifies the molecule injected and its site
of injection. Additional required attributes include onset (in
ms), duration (in ms) and rate (particles/ms). Optional attri-
butes, period and end, can be used to specify a train of input.
Thus, 100 Hz stimulation for 1 s would be specified with a
period of 10 ms, and end equal to onset plus 1,000 ms. Multiple
trains are possible with two more optional attributes: inter-
TrainInterval and numTrains. Variations in stimulation pro-
tocols, e.g., brief, large amplitude (1,000 particles during
1 ms) versus long, low amplitude (10 particles during 100 ms)
dopamine release, are used to run simulation experiments.
Stimulation can be delivered to a single spine or to a range of
spines as explained in more detail at the Web site.
3.5. Output 1. To evaluate the result of the simulation it is necessary to analyze

the concentration or quantity of one or more molecules from
one or more parts of the neuron morphology. For example, if
studying synaptic plasticity, then quantity of activated kinase,
calcium concentration, or phosphorylation state of particular
target molecules may be of interest. To understand mecha-
nisms underlying the simulation results, it may be necessary to
evaluate the time course of several molecular species in the
model. One advantage of simulations is the possibility to exam-
ine every molecular species.
2. The OutputScheme file specifies the file to which output is
written and which molecules from which compartments are
written to the file at which frequency (time interval). Multiple
output files are allowed: for example, information about some
molecules, or information from some regions can be written
more frequently than others. Every OutputSet specifies a single
output file using two required attributes, filename and dt (for
data output interval), and an optional attribute, region. If the
region attribute is not specified, then the concentrations (or
number of particles) for all subvolumes in the system are saved.
Each OutputSpecie in the OutputSet indicates a molecule to be
included in that output file (Fig. 5).
3.6. Additional Details One additional file, called the Model file (Fig. 6), serves to identify
on the Simulation all these other files and provides additional information such as
discretization options, simulation seed(s) and various control
parameters. This model file is input to the software to run the sim-
ulation, as explained in Subheading 3.7.
1. Discretization options
For numerical calculations within the model, each of the mor-
phology segments is subdivided into smaller compartments
called subvolumes. The discretization element indicates the
size of these subvolumes. Smaller sizes produce larger number
of subvolumes which require more calculations and result in a
longer run time. The defaultMaxElementSide specifies the
Fig. 5. DopamineIO.xml.
Fig. 6. DopamineModel.xml.
largest size (in microns) for each side of the subvolume in each
segment. This is the default, and can be overridden using
MaxElementSide with the region attribute: the value supplied
will control the size of subvolumes for that region. Similarly,
spineDeltaX specifies the size of subvolumes in spines. Spines
have a one-dimensional discretization along the spine axis. The
geometry element is used to specify how the morphology is
interpreted. 2D implies that there are multiple voxels in x and
y dimensions, but only a single layer of subvolumes in the
z dimension. Thus, the morphology is three-dimensional, but
diffusion occurs in two dimensions only. For 2D, the user also
specifies the depth of the subvolumes.
2. Additional parameters.
Runtime is used to specify run time in milliseconds. fixedStepDt
specifies the time step, in milliseconds, which must be smaller
than the fastest reactions or diffusion processes. The required
simulationSeed specifies the seed for the random number gen-
erator. In morphologies with spines, a separate seed, spineSeed,
is used to randomly place spines within the specified region.
outputQuantity specifies whether quantity of molecules in the
output is in number of molecules or concentration.
3.7. Run Simulations The key to meaningful experiments is asking good questions that
and Evaluate Output can be answered using data from the simulations. Simulations are
excellent tools for addressing questions such as the effect of
3.7.1. Experimental Design
morphology and spatial constraints, or the conditions which would
be required to produce an experimentally observed response.
Similar to experiments, to address particular effects, it is necessary
to run two or more simulations which differ only in the desired
parameter. For example, the effect of morphology on output may
be evaluated by performing two simulations which differ only in
their morphologies. To evaluate the effect of an antagonist requires
a control simulation (the default simulation with no antagonist)
and a simulation either with antagonist added (simulating
competitive or allosteric binding) or, alternatively, with the antago-
nist target concentration close to zero.
3.7.2. Running Simulations 1. To run a simulation from the command line in Unix, the
following command should be issued: java -jar NeuroRD.jar
DopamineModel.xml Dopamine.out >>Dopamine.log
2. To run a simulation in Windows, type “cmd” in the run
command from the start menu to obtain a command line
window. Then, type the following: java -jar NeuroRD.jar
DopamineModel.xml Dopamine.out
3. The NeuroRD.jar file is the simulation software downloaded
from the Web site (http://krasnow1.gmu.edu/CENlab/
software.html), DopamineModel.xml is the model file (“mas-
ter” file that specifies the other files), and Dopamine.out is the
main output file.
4. The name of additional output files are composed of the
main output file name plus the filename specified in
OutputScheme file.
3.7.3. Evaluating The first set of simulations typically is used to validate the model;
Simulation Output thus simulation output is compared to experimental data. Due to
stochastic variability, multiple simulation runs must be performed to
generate several observations, analogous to experiments which
require multiple trials or samples. Then, conventional statistical tests
are used to compare levels of relevant molecules from simulations to
experimental data. For more quantitative comparison to imaging

data, simulations can include fluorescent indicator molecules and
calculate the change in fluorescence.
Two additional types of simulations are typically performed.
One set of simulations implements the experimental design, e.g.,
evaluating the role of a particular molecule, or the effect of stimu-
lation pattern. Again, multiple simulation runs coupled with statis-
tical analysis are required for hypothesis testing. The other set of
simulations addresses the crucial question of how robust the
conclusions are. It is important to verify that biological conclusions
do not change with small changes in parameter values or different
stimulation protocols. The ideal parameters for robustness simula-
tions are those whose experimental measurements span a wide
range (i.e., molecule quantities and diffusion constants) and those
to which the results are more sensitive. The ability to evaluate
dynamics of many different molecules simultaneously, and to per-
form manipulations that are not possible experimentally such as
changing diffusion coefficients, allows evaluating the biochemical
mechanisms underlying experimental observations.
Evaluating simulation output is challenging due to the amount
of numerical data that is generated. In principle, the quantity of
every molecular species in the system in every subvolume is avail-
able in the numerical experiments. This abundance necessitates
additional processing of the data for efficient presentation of these
vast numerical results generated from computational modeling.
For example, numerical results can be converted to multidimen-
sional visualization (movies) to facilitate understanding and
scientific intuitions, e.g., using VNRD which is available on the
Web site (http://krasnow1.gmu.edu/CENlab/software.html).
Figure 7 shows average concentrations in the dendrite and soma
Fig. 7. Concentration of Dopamine bound D1 receptor and GolfGTP in the dendrite, where it
is produced, and in the soma, to which it diffuses. Fluctuations are due to stochastic
nature of simulation.
for the model presented in Figs. 1–6, and illustrates that the
concentration of GolfGTP (diffusible in the cytosol for the sake of
illustration) increases first in the dendrite (the location of the dop-
amine receptors), and subsequently in the soma. The average con-
centrations were generated by VNRD.
4. Notes
1. These databases can be found at:

http://doqcs.ncbs.res.in
http://www.brenda-enzymes.info
http://www.cyclic-nucleotides.org
2. Subsequent segments, which do not start on an existing
segment, will not be connected.
3. Branching segments are made by creating two segments begin-
ning at the same site but terminating at different points.
4. The two branches need not have the same radius at the point
of connection. In this case, a radius value must be specified.
Acknowledgements
This work was supported by the NIH-NSF CRCNS program on

Collaborative Research in Computational Neuroscience through
NIH grants R01 AA18060 and R01 AA16022.
References
1. Wichmann T, DeLong MR (1998) Models of 5. Oliveira RO, Terrin A, Di Benedetto G,
basal ganglia function and pathophysiology of Cannon RC, Koh W, Zaccolo M, Blackwell KT
movement disorders. Neurosurg Clin N Am (2010) The role of type 4 phosphodiesterases
9:223–236 in generating microdomains of cAMP: large
2. Goldberg JA, Rokni U, Boraud T, Vaadia E, scale stochastic simulations. PLoS One
Bergman H (2004) Spike synchronization in 5:e11725
the cortex/basal-ganglia networks of 6. Andrews SS, Addy NJ, Brent R, Arkin AP
Parkinsonian primates reflects global dynamics (2010) Detailed simulations of cell biology
of the local field potentials. J Neurosci with Smoldyn 2.1. PLoS Comput Biol
24:6003–6010 6:e1000705
3. Kitai ST, Surmeier DJ (1993) Cholinergic and 7. Byrne MJ, Waxham MN, Kubota Y (2010)
dopaminergic modulation of potassium con- Cellular dynamic simulator: an event driven
ductances in neostriatal neurons. Adv Neurol molecular simulation environment for cellular
60(40–52):40–52 physiology. Neuroinformatics 8:63–82
4. Kotaleski JH, Blackwell KT (2010) Modelling 8. Kerr RA, Bartol TM, Kaminsky B, Dittrich M,
the molecular mechanisms of synaptic plasticity Chang JC, Baden SB, Sejnowski TJ, Stiles JR
using systems biology approaches. Nat Rev (2008) Fast Monte Carlo simulation methods
Neurosci 11:239–251 for biological reaction-diffusion systems in
solution and on surfaces. SIAM J Sci Comput enzyme activity and ligand binding. Nucleic
30:3126 Acids Res 37:W441–W445
9. Stenesh J (1993) Core topics in biochemistry. 13. Garzon M, Vaughan RA, Uhl GR, Kuhar MJ,
Cogno Press, Michigan, p 283 Pickel VM (1999) Cholinergic axon terminals in
10. Bower JM, Beeman D (1998) The book of the ventral tegmental area target a subpopulation
genesis: exploring realistic neural models with of neurons expressing low levels of the dopamine
the GEneral NEural SImulation System, 2nd transporter. J Comp Neurol 410:197–210
edn. Springer, New York 14. Xie Z, Adamowicz WO, Eldred WD, Jakowski
11. Cheng Y, Prusoff WH (1973) Relationship AB, Kleiman RJ, Morton DG, Stephenson DT,
between the inhibition constant (K1) and the Strick CA, Williams RD, Menniti FS (2006)
concentration of inhibitor which causes 50 per Cellular and subcellular localization of
cent inhibition (I50) of an enzymatic reaction. PDE10A, a striatum-enriched phosphodi-
Biochem Pharmacol 22:3099–3108 esterase. Neuroscience 139:597–607
12. Cer RZ, Mudunuri U, Stephens R, Lebeda FJ 15. Rice ME, Cragg SJ (2004) Nicotine amplifies
(2009) IC50-to-Ki: a web-based tool for reward-related dopamine signals in striatum.
converting IC50 to Ki values for inhibitors of Nat Neurosci 7:583–584
Part II
Cellular Imaging
Chapter 6
A Biophysical Approach for the Study of Dopamine

Receptor Oligomerization
Sylwia Lukasiewicz, Agata Faron-Górecka,
and Marta Dziedzicka-Wasylewska
Abstract
The ability of certain neurotransmitter receptors to form oligomers provides an additional level of fine-tuning
of intracellular signaling. Among the techniques allowing study of receptor oligomerization as well as
influence of specific ligands on these processes, a biophysical approach with the use of fluorescently tagged
receptors is the most sensitive. Measurement of the fluorescence resonance energy transfer (FRET) phe-
nomenon between two fluorescently tagged receptors is considered a very useful and measurable tool to
study the physical interactions between receptors either in a single cell or in a population of living cells.
Here we describe the use of FRET measurement specifically to monitor protein oligomer formation
between dopamine D1R and D2R, but the same methodology can be used to study other receptor proteins
as well as their mutants.
Key words: FRET, Neurotransmitters, Oligomer formation, D1R, D2R, ECFP, EYFP
1. Introduction
The idea of various neurotransmitter receptors forming dimers

or higher order oligomers has been well documented (1, 2). The
formation of hetero-complexes by certain receptors can take
place only if they are co-expressed in the same cell. It has been
widely accepted that dopamine D1 and D2 receptors are expressed
rather in two separate populations of medium spiny neurons in
the striatum, with about 20% of all striatal neurons co-expressing
the two receptors (3). However, it has also been disputed that
dopamine D1 and D2 receptors are significantly co-localized in
the nucleus acumbens septi (4). Such co-localization provides a
cellular basis for the interaction of these receptors and may be
79
80 S. Lukasiewicz et al.
important for the elucidation of the mechanism of action of

certain drugs (e.g., clozapine).
Among the techniques allowing studies of receptor oligomer-
ization as well as the influence of specific ligands on these pro-
cesses, a biophysical approach with the use of fluorescently tagged
receptors is the most sensitive. Fluorescence resonance energy
transfer (FRET) measurement between two fluorescently tagged
receptors is considered a very useful and measurable tool (5–7).
FRET measurements allow for the study of physical interactions
between receptors either in single cell or in a population of living
cells. The FRET phenomenon is observed between a fluorescent
donor and an acceptor chromophore when they are located within
100 Å of each other and are arranged properly in terms of their
transition dipole moments (8). Here, we describe the use of FRET
to monitor protein oligomer formation between dopamine D1R
and D2R, but the same methodology can be used to study other
receptor proteins as well as their mutants. We labeled dopamine
receptors with enhanced cyan fluorescent protein (ECFP, the
fluorescence donor) and enhanced yellow fluorescent protein
(EYFP, the fluorescence acceptor) and expressed the receptors in
human embryonic kidney (HEK) 293 cells. Generally, receptor oli-
gomerization using the biophysical approach is difficult to analyze
in native cells. The HEK cell line has been widely used in reso-
nance energy transfer studies of membrane receptors because these
cells provide an accepted model in which fluorescently tagged
receptor proteins can be efficiently expressed.
2. Materials
2.1. Construction 1. All molecular biology reagents are obtained from Fermentas
of Fusion Proteins (Vilnius, Lithuania).
and Genetic Variants 2. Oligonucleotides (IBB PAN, Warsaw, Poland).
of the Dopamine
3. pECFP-N1 and pEYFP-N1 vectors (BD Biosciences, Clontech,
Receptors Palo Alto, CA).
4. pcDNA3.1(+) plasmids encoding human dopamine receptor
proteins (UMR cDNA Resource Center, University of
Missouri-Rolla, MO).
5. Escherichia coli DH5α (Dam+) (Novagen, EMD Chemicals,
Merck, Germany).
6. LB: 10 g/L peptone, 10 g/L NaCl, 5 g/L yeast extract, pH
7.0.
2.2. Cell Culture 1. HEK 293 cells (American Type Culture Collection, Manassas,
and Transfection VA).
6 Study of Dopamine Receptor Oligomerization 81
2. HEK 293 cells are grown in Dulbecco’s modified essential

medium (DMEM) supplemented with 1% l-glutamine and 10%
heat-inactivated fetal bovine serum (FBS). The cells were cul-
tured at 37°C in an atmosphere of 5% CO2. All cell culture
materials are purchased from GIBCO (Life Technologies,
Grand Rapids, NY) and Sigma-Aldrich (St. Louis, MO).
3. Solution of trypsin: 137 mM NaCl, 5 mM KCl, 10 mM
NaHCO3, 0.5 mM EDTA, 20 mM Hepes, 5.5 mM glucose,
0.25% trypsin, pH 7.6.
4. Phosphate-buffered saline (PBS): 140 mM NaCl, 10 mM
NaH2PO4, 1.8 KH2PO4, 2.7 mM KCl, pH 7.4.
5. Buffer HBS (transfection): (2×) 280 mM NaCl, 50 mM
HEPES, 1.5 mM Na2HPO4, pH 7.05.
6. Solution of calcium chloride (transfection): 2.5 M CaCl2.
2.3. Radioligand 1. Phosphate-buffered saline (PBS): 140 mM NaCl, 10 mM

Binding Assay NaH2PO4, 1.8 KH2PO4, 2.7 mM KCl, pH 7.4.
2. Binding buffer: 50 mM Tris–HCl, containing 120 mM NaCl,
5 mM KCl, 4 mM MgCl2, pH 7.4.
3. Washing buffer: 50 mM Tris–HCl, pH 7.4.
4. Dopamine D1R specific radioligand: [3H]SCH23390, specific
activity: 86 Ci/mmol (NEN, Boston, MA) (see Note 1).
5. Dopamine D2R specific radioligand: [3H]spiperone, specific
activity: 15.7 Ci/mmol (NEN, Boston, MA) (see Note 1).
6. Specific dopamine receptor ligands cis(Z)-flupentixol
(Lundbeck) and (+) butaclamol (Sigma).
7. Membrane protein was determined by Lowry’s modified
method: Bicinchoninic Acid Protein Assay Kit and Protein
standard, 2 mg BSA/mL (Sigma).
8. Estimation of the radioligand parameters are calculated using
the GraphPad Prism 2.0 curve—fitting program (GraphPad
Software, La Jolla, CA).
9. Glass test tube (15 mL) (see Note 2).
10. Brandel Cell Harvester.
11. Whatman GF/C filter paper.
12. Scintillation fluid (Aqascynt. BioCare, Guangzhou, China).
13. Liquid scintillation counting (Beckman LS 650).
2.4. FRET and Confocal 1. All fluorescence spectra are collected using a spectrofluorimeter
Measurements and 10 mm quartz cuvette (Hellma, Mullheim Germany).
2. Isotonic buffer 1 used as an incubating medium during
measurements: 137.5 mM NaCl, 1.25 mM MgCl2, 1.25 CaCl2,
6 mM KCl, 5.6 mM glucose, 10 mM HEPES, 0.4 mM
NaH2PO4, pH 7.4.
3. Specific dopamine receptor ligands (Hartmann Analytic,

Sigma).
4. TCSPC methodology and lifetime measurements are the most
sensitive for estimation of highly quantitative data of FRET
efficiency (see Note 3).
5. Co-localization analysis is done using Image ProPlus 4.5
software.
3. Methods
3.1. Construction 1. The human dopamine receptor genes cloned into the
of Fusion Proteins pcDNA3.1(+) plasmid are used as the starting point to con-
struct the fusion proteins. Appropriate genes are tagged with
cDNA encoding enhanced cyan or yellow fluorescent proteins
(ECFP or EYFP).
2. The full-length cDNAs encoding the dopamine receptor is
PCR-amplified. The forward primer is universal for pcDNA3.1
(+), and the reverse primers removed the STOP codon and
introduced a unique restriction site, XhoI. The resulting
fragment is inserted, in-frame, into the NheI/XhoI sites of the
pECFP-N1 and pEYFP-N1 vectors. The obtained fusion
proteins constructs are used after expression as the fluorescence
donor (receptor-ECFP) or acceptor (receptor-EYFP) (see
Note 4).
3.2. Construction 1. The genetic variants of the dopamine receptors are generated
of Genetic Variants according to the manufacturer’s protocol Quik-Change II
of the Dopamine Site-Directed Mutagenesis Kit (Stratagene).
Receptors 2. Dopamine receptor genes inserted into pECFP-N1 and
pEYFP-N1 vectors, respectively, are used as the mold for the
PCR-Quik reaction. Incorporating the oligonucleotide prim-
ers, each complementary to the opposite strand of the vector
and containing the desired mutations, generates a mutated
plasmid.
3. The resulting product is treated with endonuclease DpnI,
specific for methylated and hemimethylated DNA, in order to
select synthesized DNA containing the introduced mutations.
E. coli DH5α cells are then transformed with the mutated plas-
mid. All mutated sequences are verified by DNA sequencing.
3.3. Cell Culture 1. HEK 293 cells are cultured at 37°C in an atmosphere of 5%
and Transfection CO2. When the confluence is about 90%, the cells are passaged.
They are washed with phosphate-buffered saline (PBS) and
then treated with trypsin.
2. Transient transfections of HEK 293 cells are performed using

the calcium phosphate precipitation method, as described by
Sambrook et al. (9). HEK 293 cells are transfected with plas-
mids encoding either ECFP-tagged or EYFP-tagged dopamine
receptor proteins separately or co-transfected with different
combinations of both ECFP- and EYFP-tagged plasmids.
3. One day before transfection, the cells are seeded into 100 mm
dishes at a density of 3 × 106 cells/dish for fluorescence spectra
measurements and binding assays. For fluorescence lifetime
measurements and confocal imaging cells are plated on glass
cover slips in 10-mm dishes at a density of 2 × 105 cells/dish.
4. One hour before transfection cells are incubated with the fresh
medium containing 10% FBS in 37°C.
5. The cells are transfected with 15 μg of DNA per 100 mm dish
or 2 μg of DNA per 10-mm dish. The ratio of DNA encoding
the receptor-ECFP fusion protein (fluorescence donor) to
DNA encoding the receptor-EYFP fusion protein (fluorescence
acceptor) is 1:1 (see Note 5).
3.4. Binding Assay Binding of specific radioligands to receptors is an important step

since it allows us to determine whether fluorescently tagged recep-
tors display the same binding parameters (Bmax and Kd) as the native
receptor. To obtain the information, the saturation analysis is
employed. On the other hand, competition analysis is used to
determine the affinity (Ki) of the studied receptors for certain
ligands, e.g., clozapine. For all binding assays (saturation and
competition for D1R and D2R) the steps 1–4 are performed in the
same way.
1. The transfected HEK 293 cells are washed with phosphate-
buffered saline (PBS), scraped from the dish in PBS and centri-
fuged at 160 × g for 5 min. The pellet is then frozen at −30°C
until used.
2. Frozen pellets are resuspended in the binding buffer (50 mM
Tris–HCl, pH 7.4, containing 120 mM NaCl, 5 mM KCl,
4 mM MgCl2, and 1 mM EDTA), using an Ultra Turrax
homogenizer (30 s). The homogenates are centrifuged at
30,000 × g for 10 min. The step should be repeated twice (see
Note 6).
3. Binding studies are performed on a fresh membrane prepara-
tion (final protein concentration usually 20 μg/tube or 40 μg/
tube, for D1 and D2 dopamine receptor, respectively).
Membrane protein is determined by the modified Lowry’s
method.
4. 10 μM cis(Z)-flupentixol, (Lundbeck, Deerfield, IL) and
50 μM (+) butaclamol (Sigma) are added to determine
nonspecific binding for D1R and D2R, respectively.
Fig. 1. Representative saturation data for (a) [3H] SCH23390 binding to dopamine D1R and fusion protein D1R-EYFP; (b) [3H]
spiperone binding to dopamine D2R and fusion protein D2R-ECFP.
3.4.1. Saturation Binding A saturation binding assay for D1R is performed in 200 μL Tris–HCl
Assay buffer (pH 7.4) with 20 μg of membrane homogenate and increas-
ing 12 concentrations (0.06–6 nM) of [3H]SCH23390 (200 μL).
For D2R—200 μL Tris–HCl buffer (pH 7.4) with 40 μg of mem-
brane homogenate and increasing 12 concentrations (0.01–4 nM)
of [3H]spiperone (50 μL) are used (see Note 7) (Fig. 1).
1. Incubation time for D1R is 90 min. at 25°C, and for D2R is
30 min at 37°C.
2. At the end of the incubation, bound ligands are isolated by
rapid filtration through glass fiber filters (GF/C, Whatman).
The filters are washed four times with 5 mL of ice-cold washing
buffer (Tris–HCl, pH 7.4) (see Note 8).
3. Bound radioactivity is determined by liquid scintillation
counting.
4. Estimation of the radioligand binding parameters, Kd (the
equilibrium dissociation constant) and Bmax (maximal binding
capacity) is calculated using the GraphPad Prism version 2.0
(see Note 9).
3.4.2.Competition Binding Competition binding studies are carried out under similar condi-
Assay tions to saturation experiments. Competition analysis can be useful
to see whether the oligomerization of the studied receptors changes

their affinity for the given compound (10). Since the HEK 293 cell
line allows us to express various proteins it is also a useful model to
study the genetic variants of receptors of interest, either to search
for specific regions responsible for receptors oligomerization or to
study the impact of certain polymorphisms linked to the diseases.
The exchange of Ser-311 residue for Cys in the dopamine D2
receptor sequence is often linked to the susceptibility to schizo-
phrenia. Below the affinity for clozapine is presented of dopamine
D1 and D2 receptors as well as of genetic variant, D2Ser311Cys. The
results indicate that the value of Ki depends on whether these
receptors are expressed alone or simultaneously in the same cell.
1. Competition assays should be done on a fresh membrane prep-
aration (final protein concentration was 20 μg/tube or 40 μg/
tube for D1 and D2 dopamine receptor, respectively) using a
fixed concentration of [3H]SCH23390 or [3H]spiperone (see
Note 8).
2. One concentration of radiolabeled ligand is used: [3H]spiper-
one (0.3 nM) (see Fig. 2) and [3H]SCH23390 (1 nM) (see
Fig. 3) (see Note 10). The range of nonradioactive ligand
(clozapine, Sigma) is from 10−3 to 10−12 M.
3. Tubes are incubated for 90 min at room temperature ([3H]
SCH 23390) or for 30 min at 37°C ([3H] spiperone).
Fig. 2. Competition of [3H]SCH23390 with clozapine for binding to membrane preparation

from HEK 293 cells transiently transfected with dopamine receptors—representative
plots. Two-site binding model was fitted to the data. HEK 293 cells transiently transfected
with dopamine D1R (black line ); HEK 293 cells transiently co-transfected with dopamine
D1 and D2 receptors (blue line ); HEK 293 cells transiently co-transfected with dopamine D1
and D2S311C receptors (green line ). The estimated Ki values depended on the presence of
the second dopamine receptor.
Fig. 3. Competition of [3H]spiperone with clozapine for binding to membrane preparation

from HEK 293 cells transiently transfected with dopamine receptors—representative
plots. One- and two-site binding model was fitted to the data. HEK 293 cells transiently
transfected with dopamine D2R (black line); HEK 293 cells transiently co-transfected with
dopamine D1 and D2 receptors (blue line); HEK 293 cells transiently transfected with dop-
amine D2S311C receptors (red line); HEK 293 cells transiently co-transfected with dopamine
D1 and D2S311C receptors (green line).
4. As in the saturation analysis, binding is terminated with rapid

filtration through glass fiber filters (GF/C, Whatman). The
filters are washed four times with 5 mL of ice-cold washing
buffer (50 mM Tris–HCl, pH 7.4) and bound radioactivity is
determined by liquid scintillation counting (Beckman LS 650)
(see Note 7).
5. Estimation of the radioligand binding parameter, Ki, is calcu-
lated using the GraphPad Prism 2.0 curve-fitting program
(GraphPad Software) (see Note 11).
3.5. FRET Qualitative as well as quantitative methods are used to monitor

Measurements whether FRET occurs between the ECFP-tagged protein (donor)
and the EYFP-tagged protein (acceptor of fluorescence).
3.5.1. Fluorescence Although steady state fluorescence spectroscopy measurements in

Spectroscopy cell suspension enable only the qualitative estimation of the FRET
Measurements phenomenon, this approach is very demonstrative and allows for a
quick answer as to whether there is any energy transfer in the exam-
ined sample.
1. Spectrofluorimetric measurements of the cell suspensions are
recorded at 37°C, 48 h after transfection. Cells (expressing
desired combination of receptor fusion proteins) cultured on a
single 100 mm dish are washed and detached from the plate
Fig. 4. Fluorescence emission spectra of HEK 293 cells expressing the ECFP- and EYFP-
tagged proteins coupled to D1R and D2R. FRET control, spectra from a 1:1 mixture of cells
individually expressing the D1-ECFP fusion protein (the cyan line excited at 434 nm) and
the D2-EYFP fusion protein (the yellow line excited at 475 nm). The blue line represents
FRET spectrum of HEK 293 transfected with ECFP-EYFP construct. The black line is the
spectrum of HEK 293 co-transfected with D1-ECFP and D2-EYFP.
using warm (37°C) PBS buffer. Afterwards, the suspensions

are centrifuged at 1,000 rpm for 5 min and resuspended in
1 mL of warm (37°C) isotonic buffer 1.
2. ECFP is excited at 434 nm, and EYFP—at 475 nm. Fluorescence
is detected at 450–550 nm through a double monochromator
(Fig. 4). The spectral contributions arising from light scatter-
ing and nonspecific fluorescence of cells and incubated medium
are eliminated by subtracting the emission spectra of mock-
transfected cells from the fluorescence spectra of cells express-
ing the receptor-ECFP and -EYFP constructs. The analysis is
done according to Stanasilla et al. (11).
3.5.2. FRET Measurement 1. Cells dedicated to TCSPC experiments are grown on cover
by Fluorescence Lifetime slips. The fluorescence decay is measured from single living
Microscopy cells transfected with fusion protein constructs. All measure-
ments are performed at 37°C (see Note 12), 48 h after trans-
fection. Cells are incubated in the same isotonic buffer 1 as the
one used for fluorescence spectra measurements. For each
receptor combination, at least four independent experiments
should be performed and during each experiment, fluorescence
decay from at least 15 cells on the given cover slip should be
measured (Fig. 5).
2. Each fluorescence decay measurement is analyzed with the
multiexponential model, given by the equation:
I (t ) = ∑ i =1 αi e −t / τi
n
Fig. 5. Time-dependent fluorescence intensity decays of ECFP attached to the D1 receptor

with and without EYFP attached to the D2 receptor. The blue dotted curve shows the inten-
sity decay of the donor alone (D), and the green dotted curve shows the intensity decay of
the donor in the presence of acceptor (DA). The green and blue solid lines and weighted
residuals (lower panels) are for the best double exponential fits. The black dotted curve
represents the excitation pulse diode laser profile, set up at 434 nm.
where I(t) stands for fluorescence intensity in time t, αi are

pre-exponential factors representing amplitudes of the compo-
nents at t = 0, τi are the decay times, and n is the number of
decay times. Best fit parameters are obtained by minimizing
the reduced χ 2 value and residual distribution. The average
fluorescence lifetime ‹τ› is calculated from the equation:
τ =
∑ α ·τ
i i
2
i
∑ α ·τ
i i i
The average efficiency of energy transfer ‹E› is calculated from

the average fluorescence lifetime of donor in the absence ‹τD›
or presence ‹τDA› of an acceptor from the equation:
τ DA
E =1−
αD
3. In order to investigate the influence of specific ligands on

oligomerization processes, the cells are incubated in the
presence of agonists and antagonists for 15 min at 37°C before
the measurement (see Note 13).
3.5.3. Control Experiments 1. The often-discussed problem concerning the experiments

performed in heterologous expression systems is the issue con-
cerning “overcrowding” of the protein of interest. Dimerization
might simply be promoted at relatively high expression levels and
may partially be an artifact of over-expression. Moreover FRET
might be observed because of microdomain clustering. Therefore
additional control experiments should be performed in order to
prove that the estimated FRET efficiency reflects specific recep-
tor–receptor interaction and does not result from random molec-
ular interaction within the membrane. A good control is a
measurement of FRET efficiency between membrane-targeted
and noninteracting fusions of EYFP and ECFP, for example
membrane receptor-EYFP and cytosolic protein-ECFP.
2. Measurement of the fluorescence lifetime of ECFP expressed
in the cells alone or together with EYFP—when cells are loaded
with both fluorescent proteins, not linked to any receptors also
should be done as additional control experiment. During the
experiment no FRET should be observed, despite high expres-
sion level of both fluorescent proteins.
3. In order to estimate the highest value of FRET efficiency,
which is possible to be measured in the proposed model, the
fusion construct of two fluorescent proteins used (ECFP_
EYFP) should be prepared and all kinds of the experiments
described above should be done with cells expressing that
construct (Fig. 7).
4. To avoid over-interpretation of FRET efficiency data, the mea-
surements of homo-FRET (occurring between ECFP_ECFP)
should be also conducted.
5. The interpretation of FRET efficiency alterations as a result of
ligands presence should take into account the possible confor-
mational changes upon ligand binding within the receptor
molecule tagged with the fluorescent protein (Fig. 6). In order
to avoid misinterpretation of the data, control experiments
should be conducted. The donor-acceptor distance is calcu-
lated using the following equation:
r = R0 ⎡⎣(E −1 − 1)1/6 ⎤⎦
The anticipated possible alteration in energy transfer induced

by the conformational change is estimated according to the
following equation:
Fig. 6. FRET efficiency measured in HEK 293 cells co-expressing dopamine D1-EYFP and
D2-ECFP receptors in presence of ligand (clozapine) dependent on both clozapine concen-
tration and on the time of ligand presence in the incubation medium.
6r 5
E= r
⎡1 + (r / R0 )6 ⎤ R06
2
⎣ ⎦
6. Moreover, it is very important to show that the used ligands do

not change the fluorescent properties of ECFP labeling the
receptor. Therefore, the lifetime measurements of the recep-
tor-ECFP upon treatment with specific ligands should be done.
In such a case, no changes in the fluorescence lifetime should
be observed (Fig. 7).
3.6. Confocal Confocal microscopy is used to analyze the localization of the

Microscopy fluorescently tagged dopamine receptors in HEK 293 cells. It is
especially important in the studies of mutant protein, often per-
formed in order to identify the role of certain amino acid residues/
regions/domains within the receptor molecule in the oligomeriza-
tion process. Such genetic manipulations within the receptor
sequence might sometime change the cellular localization of the
studied receptor. Cells grown on cover slips are transiently trans-
fected with the cDNA encoding the fluorescently labeled dopamine
receptors.
1. ECFP and EYFP fluorescence is excited by 457 nm and 514 nm
wavelength lights, respectively.
2. For co-localization analysis Image ProPlus 4.5 software is used.
Co-localization describes the existence of two or more
fluorescently labeled molecule types in the same spatial posi-
tions. Pearson’s correlation coefficient is used to measure the
overlap of the pixels and reflects the degree of relationship
Fig. 7. Data obtained in our studies concerning the interactions between the GPCRs (D1R, D2R) and the appropriate alpha
subunit of G protein confirm the specificity of the methodology. Using the same expression system and the same amount
of DNA for transient transfections, FRET did not occur when two noninteracting fusion proteins (bearing ECFP and EYFP,
respectively) were co-expressed in the same cell. Representative fluorescence emission spectra of HEK 293 cells co-
transfected with either D1-EYFP or D2-EYFP and Gα-ECFP or GαI-ECFP fusion proteins. (a) Co-transfection of HEK 293 cells
with D1-EYFP and GαS-ECFP (green line) or D1-EYFP and Gα1-ECFP (blue line); (b) Co-transfection of HEK 293 cells with
D2-EYFP and GαI-ECFP (green line) or D2-EYFP and GαS-ECFP (blue line).
between two variables. It is one of the standard measures in

pattern recognition:
R=
∑ (Ri − Rav)·(Gi − Gav)
i
∑ (Ri − Rav) ·∑ (Gi − Gav)

i
2
i
2
where Ri and Gi are the red and green intensities of voxel I,

respectively, and Rav and Gav the average values of Ri and Gi,
respectively.
It is used for describing the correlation of the intensity
distributions between red and green component of each dual-
channel image. Pearson’s correlation coefficients should be
calculated from randomly selected parts of the image (mem-
brane signal) from individual cells co-transfected with different
construct combinations. The average intensity of the
fluorescence signal is measured for every image in a determined
individual area of interest free of cells, and subtracted as a back-
ground. For analysis these regions are used, of which
fluorescence intensities are correlated. For each combination
of proteins, a minimum of 20 individual regions from differ-
ent, independently transfected cells should be counted.
3. Interpretation of Pearson’s correlation coefficients, especially
relative to each other is difficult, as their relative magnitudes are
not proportional. By that reason coefficients of determination
(which are squared value of correlation coefficients) are

estimated. The resulting coefficient of determination allows
estimating the proportion of overlapping variance between
two sets of pixels, thus making the interpreting correlation
coefficients much easier.
4. Notes
1. Radioligand should be stored at −20°C and diluted in washing

buffer (Tris–HCl, pH 7.4) just before use.
2. The plastic tubes do not promote robust results, probably
owing to adhesion of membrane preparation or drugs to the
test tube wall.
3. The fluorescence lifetime measurements are independent of
any change in fluorophore concentration or excitation inten-
sity; therefore that kind of measurement provides quantitative
information about the interaction between labeled proteins of
interest. In contrast, the steady state fluorescence spectroscopy
measurements in cell suspension enable only the qualitative
estimation of the FRET phenomenon.
4. Construction of fusion protein step-by-step:
(a) PCR—mold pcDNA 3.1 plasmid with desired dopamine
receptor cDNA primers:
Forward—universal for pcDNA 3.1
Reverse—removed the STOP codon and introduced a
unique restriction site
(b) Agarose gel electrophoresis—identification and purification
of PCR product
(c) Enzymatic cleavage of obtained PCR product and plas-
mids pECFP and pEYFP
(d) Agarose gel electrophoresis—identification and purification
of desired DNA
(e) Ligation—introduction of desired DNA fragment encod-
ing appropriate dopamine receptor into vectors encoding
fluorescence proteins
(f) Transformation of E. coli DH5α cells
(g) Bacterial cells cultured, plasmid DNA isolation and
identification of colony containing vector encoding appro-
priate receptor fusion protein.
5. The amount of cDNA used for transfection does not always
correspond with protein expression levels, therefore sometimes
it is necessary to use a different donor/acceptor ratio in order

to obtain the comparable levels of the donor and acceptor
molecules. However, the total amount of cDNA used for trans-
fection cannot exceed 15 μg for 3 × 106 cells.
6. To reduce the degradation of receptors, low temperature (4°C)
must be used throughout these steps. All buffers should be
prepared fresh on the day of analysis.
7. The final volume of tubes was 0.5 mL. Each sample should be
prepared in triplicate. The concentrations of radioligands are
measured using liquid scintillation counting (Beckman LS
650). Calculation of radioligands concentration are analyzed
according to the equation:
DPM
CL =
2, 200 × V P × AS
where: CL [nM]—concentration of radioligand

DPM—disintegrations per minute
Vp [mL]—final volume of tube
As [Ci/mmol]—specific activity of radioligand.
8. Washing buffer should be prepared on the day of analyses and
cooled to 4°C before use. Prior to counting, washed filters
should be incubated with 5 mL of scintillation liquid Aquascynt
(BioCare) for 12 h in room temp.
9. Deduction of the nonspecific disintegrations per minute (dpm)
from the total at each concentration of radioligand gives a
specific binding in dpm. The dpm are then converted to pico-
mole per milligram of protein.
10. The radioligand concentration should be close to the dissocia-
tion constant, Kd, obtained from saturation binding analysis.
11. The dpm remaining for each displacement point can then be
expressed as a percentage of total specific binding.
12. It is very important to control the temperature during the
TCSPC experiments because fluorescence lifetime strongly
depends on temperature.
13. Appropriate stock concentration of ligands required for the
treatment of cells should be made right before the experiment.
Acknowledgments
The authors would like to dedicate this work to the memory of the
late professor Zygmunt Wasylewski, who encouraged us to employ
fluorescence spectroscopy in our studies of dopamine receptors.
References
1. Hansen JL, Sheikh SP (2004) Functional 6. Janetopoulos C, Devreotes P (2002)
consequences of 7TM receptor dimerization. Monitoring receptor-mediated activation of
Eur J Pharm Sci 23:301–317 heterotrimeric G-proteins by fluorescence reso-
2. Prinster SC, Hague C, Hall RA (2005) nance energy transfer. Methods 27:366–373
Heterodimerization of G protein-coupled 7. Elangovan M, Day RN, Periasamy A (2002)
receptors: specificity and functional significance. Nanosecond fluorescence resonance energy
Pharmacol Rev 57:289–298 transfer-fluorescence lifetime imaging micros-
3. Aizman O, Brismar H, Uhlén P, Zettergren E, copy to localize the protein interactions in a
Levey AI, Forssberg H, Greengard P, Aperia A single living cell. J Microsc 205:3–14
(2000) Anatomical and physiological evidence 8. Lakowicz JR (1999) Principles of fluorescence
for D1 and D2 dopamine receptor colocaliza- spectroscopy. Kluwer Academic/Plenum
tion in neostriatal neurons. Nat Neurosci Publishers, New York
3:226–230 9. Sambrook J, Fritsch EF, Maniatis T (1996)
4. Hasbi A, Fan T, Alijaniaram M, Nguyen T, Molecular cloning: a laboratory manual. Cold
Perreault ML, O’Dowd BF, George SR (2009) Spring Harbor Laboratory Press, New York
Calcium signaling cascade links dopamine 10. Faron-Górecka A, Górecki A, Kuśmider M,
D1-D2 receptor heteromer to striatal BDNF Wasylewski Z, Dziedzicka-Wasylewska M (2008)
production and neuronal growth. Proc Natl The role of D1-D2 receptor hetero-dimerization
Acad Sci U S A 106:21377–21382 in the mechanism of action of clozapine. Eur
5. Dziedzicka-Wasylewska M, Faron-Górecka A, Neuropsychopharmacol 18:682–691
Andrecka J, Polit A, Kuśmider M, Wasylewski 11. Stanasila L, Perez JB, Vogel H, Coteechia S
Z (2006) Fluorescence studies reveal heterodi- (2003) Oligomerization of the alpha 1a- and
merization of dopamine D1 and D2 receptors in alpha 1b-adrenergic receptor subtypes.
the plasma membrane. Biochemistry 45: Potential implications in receptor internaliza-
8751–8759 tion. J Biol Chem 278:40239–40251
Chapter 7
Detection of Receptor Heteromers Involving Dopamine

Receptors by the Sequential BRET-FRET Technology
Gemma Navarro, Peter J. McCormick, Josefa Mallol, Carme Lluís, Rafael
Franco, Antoni Cortés, Vicent Casadó, Enric I. Canela, and Sergi Ferré
Abstract
Until very recently, dopamine receptors, like other G-protein-coupled receptors, were believed to function
as individual units on the cell surface. Now it has been described by several groups including ours that
dopamine receptors not only function as homomers but also form heteromers with other receptors at the
membrane level. Bioluminescence and fluorescence resonance energy transfer (BRET and FRET) based
techniques have been very useful to determine the interaction between two receptors, but to demonstrate
the existence of higher-order complexes involving more than two molecules requires more sophisticated
techniques. Combining BRET and FRET in the Sequential BRET-FRET (SRET) technique permits het-
eromers formed by three different proteins to be identified. In SRET experiments, the oxidation of a
Renilla Luciferase substrate triggers acceptor excitation by BRET and subsequent energy transfer to a
FRET acceptor. Using this methodology here we describe the heteromerization between adenosine A2A,
dopamine D2, and cannabinoids CB1 receptors in living cells.
Key words: Dopamine receptors, Dopamine receptors interacting proteins, BRET, FRET, Sequential
resonance energy transfer, GPCR, Receptor oligomerization, Heteromer, Protein–protein
interaction
1. Introduction
Dopamine exerts many of its physiological functions by interacting

with dopamine receptors. Dopamine receptors are classified in D1-
like, with the D1 and D5 receptor subtypes (D1R and D5R), which
usually couple to Gs/olf proteins, and D2-like, with the D2, D3, and
D4 receptor subtypes (D2R, D3R, and D4R), which couple to Gi/o
proteins (1). Like many other G-protein coupled receptors
(GPCRs), dopamine receptors function as oligomers forming
homomers (2–6) and heteromers with other dopamine receptors
(7–10) or other GPCRs (11–13). A receptor heteromer has been
95
96 G. Navarro et al.
recently defined as a macromolecular complex composed of at least

two functional receptor units with biochemical or functional prop-
erties that are demonstrably different from those of its individual
components (14). For this reason, receptor heteromers provide
many implications for pharmacology, since they constitute new tar-
gets for drug development (15–17). The use of biophysical tech-
niques, such as bioluminescence resonance energy transfer (BRET)
and fluorescence resonance energy transfer (FRET) techniques has
been fundamental in taking the issue of GPCR oligomerization to
the front of GPCR research, providing evidence for an increasing
number of receptor heteromers in living cells (18–20). Nevertheless,
to detect higher order receptor oligomers a significant develop-
ment in biophysical energy transfer techniques has been needed.
To this end, two techniques have been developed in our laboratory
to study oligomers formed by three different proteins and applied
to determine heterotrimers involving dopamine receptors. One is
BRET with bimolecular fluorescence complementation (BiFC),
that we used to demonstrate heteromultimerization between ade-
nosine A2A (A2AR), D2R, and cannabinoid CB1 (CB1R) receptors
and between A2AR, D2R, and glutamate mGlu5 receptors (21, 22).
The other technique is Sequential-BRET-FRET (SRET) (23).
In SRET, the oxidation of an RLuc substrate by an RLuc-fusion
protein triggers the excitation of the BRET acceptor (i.e., protein
fused to GFP2) and subsequent energy transfer to the FRET accep-
tor (i.e., protein fused to YFP). SRET will only occur with these
fusion proteins if the two acceptor–donor pairs, Rluc/GFP2 and
GFP2/YFP, are at a distance of less than 10 nm. Here the tech-
nique is described to detect heterotrimers formed by A2AR, D2R,
and CB1R in living cells. In general we conclude that SRET is an
invaluable technique to identify oligomeric complexes of more
than two proteins localized at the plasma membrane, including
more than two GPCRs, which will enable us to better understand
how signals are integrated at the plasma membrane level.
2. Materials
2.1. Fusion Proteins 1. The cDNA for functionally validated fusion proteins in suitable
and Expression mammalian expression vectors are used. The human cDNAs
Vectors for A2AR, D2R, CB1R, and the negative control human
dopamine D4.4 receptor, cloned in pcDNA3.1.
2. pRluc-N1 vector (Rluc expressing vector, PerkinElmer,
Wellesley, MA).
3. pGFP2-N3(h) vector (humanized pGFP2-N3(h) from
PerkinElmer (Waltham, MA)).
7 SRET Technology to Detect Receptor Oligomers 97
4. pEYFP-N1 vector (enhanced yellow variant of GFP; Clontech,

Heidelberg, Germany).
5. DH5α competent bacterial cells (Invitrogen, Carlsbad CA).
6. DH5α growing medium (LB): 10 g/L NaCl, 10 g/L Tryptone,
and 5 g/L Yeast extract in mQ water.
7. Xtra Maxi kit (Nucleobond®, Düren, Germany).
2.2. Cell Culture 1. Human Embrionic Kidney (HEK) 293T cells are grown in
6-well cell culture plates (Techno Plastic Products, Lausanne,
Switzerland).
2. As suitable growth medium, Complete Medium (Dulbecco’s
modified Eagle’s medium (DMEM; Gibco (Carlsbad, CA))
supplemented with 2 mM L-glutamine, 100 U/ml penicillin–
streptomycin, 5% (v/v) heat inactivated Fetal Bovine Serum
(FBS), and 5% (v/v) nonessential amino acids (all supplements
are from Invitrogen, Paisley, Scotland, UK) are used.
3. Protein quantification reagent; Bradford solution (Bio-Rad,
Hercules, CA) diluted 1/5 (v/v) in milliQ (mQ water).
2.3. Transfection 1. Branched PEI (PolyEthylenImine, Sigma, Steinheim,

Germany). Prepare a 40 μM solution in mQ water.
2. NaCl solution: NaCl 150 mM prepared in mQ water.
3. 0.05% trypsin (Gibco).
4. HBSS buffer: 0.185 g of CaCl2·12H2O, 0.370 g of KCl,
0.060 g of KH2PO4, 0.100 g of MgCl2·2H2O, 0.100 g of
MgSO4·7H2O, 8.000 g of NaCl, 0.121 g of Na2HPO4·12H2O,
and 2.385 g of HEPES in 1 L of mQ water. Use 1 M NaOH
to adjust the pH to 7.4.
2.4. SRET 1. Assay buffer: HBSS buffer containing 1 g/L D-glucose. Add
glucose to the buffer 10 min before using.
2. 500 μM DeepBlueC (Perkin Elmer) in anhydrous ethanol as
luciferase substrate stock solution. Store at −20°C protected
from light.
3. 500 μM coelenterazine h (Perkin Elmer) in anhydrous ethanol
as luciferase substrate stock solution (Panreac, Barcelona,
Spain). Store at −20°C protected from light.
2.5. Equipment 1. Multiskan Ascent Photometer (Thermo Labsystems, San Diego

CA).
2. Fluostar Optima Fluorimeter equipped with a high-energy
xenon flash lamp and appropriate filters (excitation filter at
485 nm and 410 nm and emission filter corresponding to
530 nm and 510 nm) (BMG Labtechnologies, Offenburg,
Germany).
3. Mithras LB 940 equipped with detection filters for short-

wavelength (400 nm) and long-wavelength (530 nm) (Berthold
Technologies, DLReady, Germany).
4. 96-well white microplates for BRET (Porvair, Norfolk, UK).
5. 96-well black microplates with transparent bottom for
fluorescence detection (Porvair).
3. Methods
3.1. Preparation Generate fusion constructs in Rluc, GFP2, or YFP expression

of Fusion Proteins vectors consisting of the cDNA for the protein of interest, inserted
in-frame with the cDNA for the bioluminescent or fluorescent
donor or acceptor molecule.
1. Select the donor and acceptor combination to perform SRET
(i.e., A2AR-Rluc, D2R-GFP2, CB1R-YFP, and a negative control
D4.4R-Rluc). Since in SRET experiments, the oxidation of an
Rluc substrate triggers acceptor excitation by BRET and sub-
sequent energy transfer to a FRET acceptor (Fig. 1) it is impor-
tant to select the optimal combination (see Note 4).
Fig. 1. Sequential BRET-FRET (SRET). SRET combines BRET and FRET involving two energy donors and two acceptors. BRET
and FRET techniques are combined to detect heterotrimers at the membrane level. Signal is initiated by oxidation of DeepBlueC
by the Rluc-fused protein (A2AR-Rluc) that generates light emission at the indicated wavelength (blue ). The acceptor in BRET
is a GFP2-fused protein (D2R-GFP2) that, after excitation, results in emission at the indicated wavelength (green ) that excites a
YFP-fused protein (CB1R-YFP) by a FRET process with concomitant light emission peaking at the indicated wavelength
(yellow ). Emission of YFP after addition of the Rluc substrate is only possible if the three fusion proteins are in close proximity
(<10 nm) allowing bioluminescent and fluorescent sequential resonance energy transfer (SRET) to occur. A representation of
excitation (top) and emission (bottom) spectra of fused proteins is shown in the right .
2. The human cDNAs for A2AR, D2R, CB1R, or D4.4R cloned in

pcDNA3.1 are amplified, without their stop codons, using sense
and antisense primers harboring unique EcoRI and BamHI sites
to clone A2AR in the Rluc corresponding vector, EcoRI and KpnI
to clone D2R in the GFP2 corresponding vector and BamHI and
EcoRI to clone CB1R in EYFP corresponding vector or XhoI and
BamHI sites to clone D4.4R in pRluc-N1 vector. To avoid tran-
scription mutations, it is recommendable to use a high-fidelity
DNA polymerase that offers extreme performance for all PCR
applications (i.e., iProof™ from Bio-Rad). It is important to
select the annealing temperature for each primer as indicated by
the supplier and use 15–30 s/kbase for extension times, since
longer times could induce a loss of bases.
3. The amplified fragments are subcloned to be in-frame into restric-
tion sites of a multiple cloning region within pRluc-N1, pGFP2-
N3(h), or pEYFP-N1 vectors respectively yielding the plasmids
corresponding to A2AR-Rluc, D4.4R-Rluc, D2R-GFP2, and CB1R-
YFP. To do this, the amplified fragments (dephosphorylated or
not, depending on the activity of the enzymes) and vectors are
cut with the specific enzymes (see step 3). Then 100 ng of vector
and 60 ng/Kb of insert are mixed with 1 unit of T4 cDNA ligase
(Promega, Fitchburg MA) for 3 h at room temperature. The
ligation product is transformed in DH5α competent bacterial
cells and the subcloned vectors are selected using the specific
selection antibiotics of each vector. The positive colonies are
grown in LB medium in the presence of the selection antibiotic
concentration indicated by the vector supplier.
4. Sequence cDNA to test the correct sequence of the fusion pro-
tein. Obtain enough cDNA to do several transfections using a
Xtra-Maxi kit (usually we obtain 500 μL of cDNA from
1–5 μg/μL concentration).
5. Check that the luminescence or fluorescence is detectable after
fusion protein expression (represent the amount of cDNA
transfected in HEK 293T cells versus bioluminescence or
fluorescence detected). If possible, use confocal microscopy to
visualize (by its own fluorescence or using antibodies) correct
cellular localization of fusion proteins.
6. Validate the fusion proteins of interest, including suitable control
proteins, by comparing fusion and wild-type proteins in func-
tional assays. ERK 1/2 phosphorylation and cAMP production
can be used for this purpose as described previously (13, 24).
7. Generate positive controls for SRET experiments. Rluc-
expressing vector (pRlu-N1) is amplified without its stop
codon using sense and antisense primers harboring unique
HindIII and KpnI sites to clone Rluc in-frame into restrictions
sites of a multiple cloning site of GFP2-YFP vector (pcDNA3.1-
GFP2-YFP; Biosignal Packard) to obtain the fusion protein
Rluc-GFP2-YFP.
8. Reconstitute and store the luciferase substrate stock solution

containing DeepBlueC with anhydrous ethanol. Protect the
solution from light.
3.2. Cell Transient 1. HEK 293 T cells are passaged when approaching confluence
Transfection with trypsin/EDTA to provide new maintenance cultures in
150 cm2 flasks. One 150 cm2 flask is required for transfection
in order to obtain enough transfected cells to perform a SRET
saturation curve (see Subheading 3.4, step 4). Aliquot cells in
6-well cell culture plate in growth medium. They should be
60–80% confluent after 24 h. Maintain at 37°C, 5% CO2 and
90% of humidity.
2. Transfect the expression vectors corresponding to the desired
fusion proteins at the suitable ratios (see legends of Figs. 2 and 3)
using the PolyEthylenImine (PEI) method. Other methods of
transfection may be used as well. To do this, two Falcon tubes
Fig. 2. SRET for A2AR, D2R, and CB1R in living cells. SRET assays are performed 48 h post-
transfection in cells expressing A2AR-Rluc (2 μg of cDNA; approximately 100,000 lumines-
cence units), D2R-GFP2 (3 μg of cDNA; approximately 6,000 fluorescence units), and
CB1R-YFP (9 μg of cDNA; approximately 18,000 fluorescence units) or the equivalent
amounts of the fluorescence or luminescence proteins or transfected with the positive
SRET construct (1 μg of cDNA of Rluc-GFP2-YFP construct). Net SRET was obtained by
monitoring the YFP fluorescence emission after DeepBlueC addition, with subtraction of
the value obtained with cells expressing the same amount of A2AR-Rluc and the corre-
sponding BRET acceptor. Significant net SRET was detected for A2AR-Rluc/D2R-GFP2/
CB1R-YFP coupling or for the positive SRET control, while negligible net SRET was obtained
in cells expressing equivalent amounts of A2AR-Rluc, GFP2, and CB1R-YFP, or A2AR-Rluc,
D2R-GFP2, and YFP. Data are expressed as the mean net SRET ± S.E.M. of four independent
experiments performed in duplicate. One-way ANOVA followed by Newman–Keuls test
showed significant differences with respect to negative controls (***: P < 0.001).
Fig. 3. SRET saturation curve for A2AR-D2R-CB1R heteromers in living cells. SRET satura-
tion curves were obtained using HEK-293T cells transfected with 2 μg of the cDNA for
A2AR-Rluc (approximately 100,000 luminescence units) and 3 μg of the cDNA for D2R-
GFP2 (approximately 6,000 fluorescence units) and increasing amounts of the cDNA for
CB1R-YFP (8,000 to 18,000 fluorescence units). Values, expressed as net SRET, represent
the mean ± S.E.M. of two independent experiments performed in triplicate. Negative con-
trol is constituted by cells expressing the equivalent amounts of D4R-Rluc/A2AR-GFP2/
CB1R-YFP giving linear (nonspecific) SRET with similar amounts of fluorescence and lumi-
nescence as those giving saturable SRET.
are needed. In the Falcon A each μg of cDNA is mixed with

25 μL of NaCl solution. In Falcon B the same amount of NaCl
solution is mixed with 1.25 μL/μg cDNA of 40 μM branched
PEI solution. Both Falcon tubes are vortexed a few seconds.
After this, Falcon B solution is added to Falcon A. The final
solution is strongly vortexed for 10 s. Ten minutes later, 1 mL
of serum starved medium is added. Cells are incubated for 4 h
with the cDNA-PEI solution. Then cells are placed in a fresh
complete culture medium.
3. Around 48 h after transient transfection, detach the cells and
resuspend them in HBSS buffer containing 1 g/L D-glucose.
Wash cells twice with the same buffer for 5 min and resuspend
them in the same buffer.
4. Using 10 μL of cell suspension, quantify the amount of protein
using a Bradford assay and dilute cells to 200 μg/mL. It is
important to maintain the same amount of protein in each
sample. Aliquot cells into 96-well white and black isoplates
(100 μL per black well and 90 μL per white well).
3.3. SRET Detection Using aliquots of transfected cells (20 μg of protein), four different
determinations are performed in parallel:
1. Quantification of protein-YFP expression by determination of
the fluorescence due to protein-YFP. Cells distributed into
96-well microplates (black plates with a transparent bottom),
are read in a Fluostar Optima Fluorimeter using an excitation
filter at 485 nm, and a 10 nm bandwidth emission filter

corresponding to 530 nm (527–536 nm).
2. It is also important to quantify the protein-GFP2 expression to
control for BRET energy transmission and to develop the lin-
ear un-mixing arrangement. Cells distributed into 96-well
microplates (black plates with a transparent bottom), are read
in a Fluostar Optima Fluorimeter using an excitation filter at
410 nm and a 10 nm bandwidth emission filter corresponding
to 510 nm (506–515 nm).
To determine the exact fluorescence amount of protein-
GFP2 and protein-YFP it is necessary to calculate the linear
un-mixing. First, analyze the contribution of GFP2 and YFP
proteins alone using the two detection channels (see Note 6).
Fluorescent determinations are measured in parallel in experi-
ments with cells expressing only one of these proteins and nor-
malized to the sum of the signal obtained in the two detection
channels. Values obtained in 1 and 2 are corrected considering
this linear un-mixing. The sample fluorescence is the emission
at 530 nm corrected as described minus the fluorescence of
cells expressing only protein-Rluc and protein-GFP2. The pro-
tein-GFP2 fluorescence is the emission at 510 nm corrected as
described minus the fluorescence of cells expressing only pro-
tein-Rluc.
3. Quantification of protein-Rluc expression by determination of
the luminescence due to protein-Rluc. Cells are distributed in
96-well microplates (white plates) and luminescence is deter-
mined 10 min after the addition of 5 μM coelenterazine H in
a Mithras LB 940 multimode reader.
4. SRET measurements. Cells are distributed in 96-well
microplates (white plates) and 5 μM DeepBlueC is added.
SRET signal is collected using a Mithras LB 940 reader 30 s
after the substrate addition with detection filters for short-
wavelength (400 nm (370–450 nm)) and long-wavelength
(530 nm (510–590 nm)). Net SRET is defined as [(long-
wavelength emission)/(short-wavelength emission)] − Cf,
where Cf corresponds to [(long-wavelength emission)/
(short-wavelength emission)] for cells expressing protein-
Rluc, protein-GFP2 and the other protein partner not fused
to a fluorescence protein (similar values are obtained mea-
suring Cf in cells expressing protein-Rluc only and protein-
GFP2). Linear un-mixing is done for quantification, taking
into account the spectral signature to separate the two
fluorescence emission spectra (see Note 6). SRET deter-
mined in cells expressing A2AR-Rluc, D2R-GFP2 and CB1R-
YFP or the corresponding positive or negative controls is
shown in Fig. 2.
3.4. SRET Saturation SRET saturation curve is further proof for the specificity of the
Curves interaction observed.
1. Transiently transfect HEK 293T cells with a constant amount of
the constructs corresponding to the protein-Rluc and protein-
GFP2 and with increasing amounts of the construct correspond-
ing to protein-YFP indicated in the legend of Fig. 3.
2. After 48 h of transient transfection determine SRET as indi-
cated (see Subheading 3.3) for each transfection condition.
3. Both fluorescence and luminescence for each sample are mea-
sured to confirm similar donors expression (approximately
100,000 bioluminescence units and 6,000 GFP2 fluorescence
units) while monitoring the increase in acceptor expression
(1,000 to 20,000 fluorescence units) (see Note 7).
4. Represent the net SRET values as a function of the amount of
the acceptor. In each saturation curve, the relative amount of
acceptor is given as the ratio between the fluorescence of the
acceptor (YFP) and the luminescence of the first donor (Rluc).
Curves are fitted to a nonlinear regression equation, assuming
a single phase (i.e., with Graph-Pad Prism software, San Diego,
CA, USA). From these saturation curves, an apparent SRETmax
and an apparent SRET50 can be determined (see Note 8).
A SRET saturation curve obtained by increasing CB1R-YFP
expression while maintaining the same A2AR-Rluc/D2-GFP2
ratio (Fig. 3) allows determination of the following parameters
for the trimer A2AR-Rluc/D2R-GFP2/CB1R-YFP: apparent
SRETmax of 0.18 ± 0.05 and apparent SRET50 of 0.013 ± 0.007.
4. Notes
1. If cells are dying after transfection refer to cytotoxicity of trans-

fection reagent.
2. The amount of fusion proteins transfected must be near the
physiological range.
3. If the donor protein expression is high, then the fluorescence
background will be high and poor SRET signal will be detected.
If there is low donor protein expression, the acceptor protein
will not be excited.
4. The relative orientation of donor and acceptor can be unsuit-
able for SRET.
5. Add DeepBlueC immediately before SRET detection.
6. Given the spectral emission of the GFP2/YFP pair, the contri-
bution of GFP2 or YFP proteins to the two detection channels
is significant. Thus, the spectral signature (25) of these fusion

proteins must be determined and considered for SRET and
YFP expression evaluation.
7. In SRET saturation curves, be sure that the amount of
luminescence due to protein-Rluc and the amount of
fluorescence due to protein-GFP2 are constant while increasing
the fluorescence of the acceptor protein-YFP.
8. It should be noted that the SRETmax and SRET50 will differ
when different ratios of donor–acceptor for BRET are used.
This is reminiscent of what occurs in enzymology, when the
Vmax and KM for an enzyme using two substrates are calculated
by maintaining the concentration of one substrate constant
but varying the concentration of the other. For such enzymes,
the calculated values are known as “apparent Vmax” and
“apparent KM”. Accordingly, we propose the denomination of
“apparent SRETmax” and “apparent SRET50”.
Acknowledgments
Study supported by grants from Spanish Ministerio de Ciencia y

Tecnología (SAF2008-00146, SAF2008-03229-E, and SAF2009-
07276), grant 060110 from Fundació La Marató de TV3 and by
the Intramural Funds of the National Institute on Drug Abuse.
References
1. Missale C, Nash SR, Robinson SW, Jaber M, 6. Han Y, Moreira IS, Urizar E, Weinstein H,
Caron MG (1998) Dopamine receptors: from Javitch JA (2009) Allosteric communication
structure to function. Physiol Rev 78:189–225 between protomers of dopamine class A GPCR
2. Ng GY, Mouillac B, George SR, Caron M, dimers modulates activation. Nat Chem Biol
Dennis M, Bouvier M, O’Dowd BF (1994) 9:688–695
Desensitization, phosphorylation and palmi- 7. Rashid AJ, So CH, Kong MM, Furtak T,
toylation of the human dopamine D1 receptor. El-Ghundi M, Cheng R, O’Dowd BF, George
Eur J Pharmacol 267:7–19 SR (2007) D1-D2 dopamine receptor heteroo-
3. Kong MM, Fan T, Varghese G, O’Dowd BF, ligomers with unique pharmacology are coupled
George SR (2006) Agonist-induced cell sur- to rapid activation of Gq/11 in the striatum.
face trafficking of an intracellularly sequestered Proc Natl Acad Sci U S A 9:654–659
D1 dopamine receptor homo-oligomer. Mol 8. Marcellino D, Ferré S, Casadó V, Cortés A, Le
Pharmacol 70:78–89 Foll B, Mazzola C, Drago F, Saur O, Stark H,
4. George SR, Lee SP, Varghese G, Zeman PR, Soriano A, Barnes C, Goldberg SR, Lluis C,
Seeman P, Ng GY, O’Dowd BF (1998) A Fuxe K, Franco R (2008) Identification of
transmembrane domain-derived peptide inhib- dopamine D1-D3 receptor heteromers.
its D1 dopamine receptor function without Indications for a role of synergistic D1-D3
affecting receptor oligomerization. J Biol receptor interactions in the striatum. J Biol
Chem 273:30244–30248 Chem 283:26016–26025
5. Guo W, Urizar E, Kralikova M, Mobarec JC, 9. Fiorentini C, Busi C, Gorruso E, Gotti C,
Shi L, Filizola M, Javitch JA (2008) Dopamine Spano P, Missale C (2008) Reciprocal regula-
D2 receptors form higher order oligomers at tion of dopamine D1 and D3 receptor func-
physiological expression levels. EMBO J tion and trafficking by heterodimerization.
27:2293–2304 Mol Pharmacol 74:59–69
10. So CH, Verma V, Alijaniaram M, Cheng R, 18. Milligan G (2004) Applications of biolumi-
Rashid AJ, O’Dowd BF, George SR (2009) nescence- and fluorescence resonance energy
Calcium signaling by dopamine D5 receptor transfer to drug discovery at G protein-
and D5-D2 receptor hetero-oligomers occurs coupled receptors. Eur J Pharm Sci 21:
by a mechanism distinct from that for dop- 397–405
amine D1-D2 receptor hetero-oligomers. Mol 19. Pfleger KD, Eidne KA (2005) Monitoring the
Pharmacol 75:843–854 formation of the dynamic G-protein-coupled
11. Ferrada C, Moreno E, Casadó V, Bongers G, receptor-protein complexes in living cells.
Cortés A, Mallol J, Canela EI, Leurs R, Ferré Biochem J 385:625–637
S, Lluís C, Franco R (2009) Marked changes 20. Marullo S, Bouvier M (2007) Resonance
in signal transduction upon heteromerization energy transfer approaches in molecular phar-
of dopamine D1 and histamine H3 receptors. macology and beyond. Trends Pharmacol Sci
Br J Pharmacol 157:64–75 28:362–365
12. Juhasz JR, Hasbi A, Rashid AJ, So CH, George 21. Cabello N, Gandía J, Bertarelli DC, Watanabe
SR, O’Dowd BF (2008) Mu-opioid receptor M, Lluís C, Franco R, Ferré S, Luján R, Ciruela
heterooligomer formation with the dopamine F (2009) Metabotropic glutamate type 5,
D1 receptor as directly visualized in living cells. dopamine D2 and adenosine A2A receptors
Eur J Pharmacol 581:235–243 form higher-order oligomers in living cells.
13. Canals M, Marcellino D, Fanelli F, Ciruela F, de J Neurochem 109:1497–1507
Benedetti P, Goldberg SR, Neve K, Fuxe K, 22. Navarro G, Carriba P, Gandía J, Ciruela F,
Agnati LF, Woods AS, Ferré S, Lluis C, Bouvier Casadó V, Cortés A, Mallol J, Canela EI,
M, Franco R (2003) Adenosine A2A-dopamine Lluis C, Franco R (2008) Detection of heter-
D2 receptor-receptor heteromerization. omers formed by cannabinoid CB1, dop-
Qualitative and quantitative assessment by amine D2, and adenosine A2A
fluorescence and bioluminescence resonance G-protein-coupled receptors by combining
energy transfer. J Biol Chem 278:46741–46749 bimolecular fluorescence complementation
14. Ferré S, Baler R, Bouvier M, Caron MG, Devi and bioluminescence energy transfer.
LA, Durroux T, Fuxe K, George SR, Javitch ScientificWorldJournal 8:1088–1097
JA, Lohse MJ, Mackie K, Milligan G, Pfleger 23. Carriba P, Navarro G, Ciruela F, Ferré S,
KD, Pin JP, Volkow ND, Waldhoer M, Woods Casadó V, Agnati L, Cortés A, Mallol J, Fuxe K,
AS, Franco R (2009) Building a new concep- Canela EI, Lluís C, Franco R (2008) Detection
tual framework for receptor heteromers. Nat of heteromerization of more than two proteins
Chem Biol 5:131–134 by sequential BRET-FRET. Nat Methods 5:
15. Casadó V, Cortés A, Mallol J, Pérez-Capote K, 727–733
Ferré S, Lluis C, Franco R, Canela EI (2009) 24. Carriba P, Ortiz O, Patkar K, Justinova Z,
GPCR homomers and heteromers: a better choice Stroik J, Themann A, Müller C, Woods AS,
as targets for drug development than GPCR Hope BT, Ciruela F, Casadó V, Canela EI,
monomers? Pharmacol Ther 124: 248–257 Lluis C, Goldberg SR, Moratalla R, Franco R,
16. Ferré S, Lluís C, Lanciego JL, Cortés A, Mallol Ferré S (2007) Striatal adenosine A2A and
J, Canela EI, Lluís C, Franco R (2010) G pro- cannabinoid CB1 receptors form functional
tein-coupled receptor heteromers as new tar- heteromeric complexes that mediate the motor
gets for drug development. CNS Neurol effects of cannabinoids. Neuropsychopharma-
Disord Drug Targets 9:596–600 cology 32:2249–2259
17. Ferré S, Navarro G, Casadó V, Cortés A, Mallol 25. Zimmermann T, Rietdorf J, Girod A, Georget
J, Canela EI, Lluís C, Franco R (2010) G V, Pepperkok R (2002) Spectral imaging and
protein-coupled receptor heteromers as new linear un-mixing enables improved FRET
targets for drug development. Prog Mol Biol efficiency with a novel GFP2-YFP FRET pair.
Transl Sci 91:41–52 FEBS Lett 531:245–249
Chapter 8
BRET Approaches to Characterize Dopamine and TAAR1

Receptor Pharmacology and Signaling
Stefano Espinoza, Bernard Masri, Ali Salahpour, and Raul R. Gainetdinov
Abstract
It is evident that G protein-coupled receptors (GPCRs) such as D2 dopamine receptor and functionally
related Trace Amine Associated Receptor 1 (TAAR1) can engage both in G protein-dependent (e.g.,
cAMP-mediated) and -independent β-arrestin-mediated signaling modalities. Both of these signaling
events can be monitored in real-time and in live cells by using new biosensors based on a Bioluminescence
Resonance Energy Transfer (BRET) approach. Here we discuss the practical applications of BRET to ana-
lyze dynamics of cAMP modulation via an EPAC biosensor as well as recruitment of β-arrestin2 to the D2
dopamine receptor. Combination of these approaches allows for a comparison of activity of pharmacologi-
cal compounds on these signaling modalities as demonstrated for various antipsychotics as regard to D2
dopamine receptor. Furthermore, analysis of cAMP concentrations in cells expressing TAAR1 provides a
simple high-throughput screening method to identify new ligands for this receptor. These BRET approaches
could be applied for the characterization of pharmacology and signaling of variety of other GPCRs.
Key words: D2R, TAAR1, cAMP, β-arrestin2, BRET, EPAC, 3-Methoxytyramine,

β-Phenylethylamine
1. Introduction
Dopamine (DA) is a major monoaminergic neurotransmitter in the

mammalian brain and controls many physiological functions, such
as voluntary movement, emotion, reward, cognition and neuroen-
docrine secretion (1). DA is also involved in critical processes in
the periphery, including gastrointestinal activity, regulation of
kidney function and catecholamine secretion. A variety of human
pathologies have been linked to a dysregulation of dopaminergic
neurotransmission ranging from schizophrenia and addiction to
Parkinson’s disease, Tourette’s syndrome, Attention Deficit
Hyperactivity Disorder (ADHD), and hyperprolactinemia (1). DA
exerts its physiological functions via action on five members of a
107
108 S. Espinoza et al.
family of G protein-coupled receptors (GPCR) that are subdivided

into D1-like (D1 and D5) and D2-like (D2, D3, D4) dopamine
receptor subtypes. D1-like dopamine receptors are positively cou-
pled to adenylate cyclase and their stimulation increase cAMP level
inside the cells, while D2-like dopamine receptors negatively regu-
late the production of cAMP in the cell. In addition, recent studies
have demonstrated a novel G protein-independent pathway for D2
dopamine receptors (D2Rs) that involves the multifunctional scaf-
folding protein β-arrestin2 (2). While best known for their role in
GPCR desensitization, β-arrestins are emerging as key signaling
proteins in a variety of physiological processes (3). Regarding D2R,
β-arrestin2-orchestrated formation of a protein complex including
D2R, and other proteins (PP2A and Akt) has been demonstrated
to be important for the full manifestation of certain D2R-related
behaviors (4). Intriguingly, a recent comparison of the ability of
several clinically active antipsychotics to affect G protein-depen-
dent (cAMP accumulation) versus G protein-independent processes
(β-arrestin2 recruitment), has shown that antipsychotics are gener-
ally more effective in preventing β-arrestin2 recruitment to the
D2R than regulating cAMP concentration within the cell (5).
Another GPCR that is emerging as a functionally important
modulator of dopaminergic activity in the brain is Trace Amine
Associated Receptor 1 (TAAR1) (6–8). TAAR1 belongs to the
family of Trace Amine Associated Receptors (TAARs) that is
expressed in several mammalian brain areas including limbic regions
and sites containing monoaminergic cell bodies such as the VTA
(8). It has been shown that TAAR1 can modulate dopaminergic
activity both by altering VTA neurons firing and D2 receptor sig-
naling (9). TAAR1 is a Gs coupled receptor and it is activated by a
class of compounds called trace amines (TAs) that include β-phe-
nylethylamine (β-PEA), tyramine, octopamine, and tryptamine (6,
7, 10). Other interesting molecules that bind TAAR1 are amphet-
amines and the major extracellular dopamine metabolite 3-meth-
oxytyramine (3-MT) (6, 10). This compound has been traditionally
considered as an inactive o-methylated (by COMT) metabolite of
extracellular DA but recently its role in neuromodulation and
DA-independent movement control that is in part mediated by
TAAR1 has been described (11).
Classical assays that evaluate cAMP concentration consist in
detection of the accumulation of the second messenger inside the
cell, or in the evaluation of its signaling as measured by a reporter
gene. The main drawback of these techniques is their low versatil-
ity in terms of temporal monitoring of the sample. In the last few
years, new biosensors based on Fluorescence Resonance Energy
Transfer (FRET) and Bioluminescence Resonance Energy Transfer
(BRET) have been described to measure in real-time physiological
events within living cells (12–14). To monitor cAMP variations, a
genetically encoded BRET biosensor has been developed (10).
The exchange protein activated by cAMP (EPAC) biosensor binds
8 Dopamine/TAAR1 Signaling 109
cAMP in the cell resulting in a conformational change and in an

energy transfer between Rluc and YFP in a concentration depen-
dent manner. With a similar BRET approach, it is also possible to
monitor, in real time, β-arrestin2 recruitment to the D2 receptor
(5). These two systems are increasingly useful in the study of two
important processes in vitro that involve GPCRs signaling: cAMP
levels modulation and β-arrestin recruitment to the receptor. The
major advantages of these approaches are (1) the possibility to
monitor the kinetics of signaling events (second by second) within
live cells, (2) a direct comparison of activity for a given pharmaco-
logical compound on G protein-dependent and independent sig-
naling events within the same cell.
2. Materials
2.1. Cell Culture 1. Dulbecco’s Modified Eagle’s Medium (DMEM) (Invitrogen,

and Compounds Carlsbad, CA) supplemented with 10% (vol/vol) of fetal bovine
serum (FBS, Sigma, St. Louis, MO), 10 mM of HEPES (store
at 4°C) (Invitrogen), and 0.01 mg/mL of gentamicin (store at
room temperature (RT)) (Invitrogen) is used for normal cell
culture purpose and it is stored at 4°C. FBS is aliquoted in
50 mL tube and stored at −20°C.
2. Phenol red free Minimum Essential Medium (MEM) contain-
ing 2% of FBS, 10 mM HEPES, and 2 mM L-glutamine
(Invitrogen) is used for plating the cells before the experiments
and it is stored at 4°C. L-glutamine is a 100× solution stocked
at −20°C in 5 mL aliquots.
3. Poly-D-lysine (Sigma) is a powder dissolved in tissue-culture
water at 0.1 mg/mL and stored at 4°C. Do not discard after
the use since it could be reused several times.
4. Phosphate Buffer Saline (PBS) (Invitrogen) is used to wash the
cells. PBS containing calcium and magnesium (Sigma) and
0.003% (wt/vol) of ascorbic acid (Sigma) is the medium for
the BRET experiments and used to dissolve the compounds
(store at 4°C).
5. Trypsin at 0.25% (wt/vol) with ethylenediamine tetraacetic
acid (EDTA) (1 mM) (Invitrogen) is used to detach the cells
and is kept at 4°C.
6. Burker chamber (Blaubrand, Germany) to count cells and try-
pan blue solution at 0.4% (Sigma) to stain live ones.
7. Human embryonic kidney 293 cells (HEK293T) (Invitrogen).
8. Coelenterazine h (Promega, Madison, WI) is dissolved in anhy-
drous ethanol to a final concentration of 1 mM and it must be
stored at −20°C with some desiccant compound such as silica

gel to keep the humidity at low levels. Coelenterazine h is light
sensitive and the working solutions must be prepared immedi-
ately before the addition to the plate.
9. 3-Isobutyl-1-methylxanthine (IBMX), β-phenylethylamine
(β-PEA), 3-methoxy-tyramine (3-MT), dopamine (DA), for-
skolin, haloperidol, quinpirole, and isoproterenol are from
Sigma. IBMX is prepared at 200 mM in DMSO (1,000×) and
stored at −20°C. Forskolin is dissolved at 10 mM in DMSO as
stock solution and it is stable at RT at least for 6 months. All
the other compounds are made ready at 10 mM in PBS (plus
ascorbic acid—see Subheading 4) and then diluted to the
desired concentration 1:10 each time in PBS except haloperi-
dol that has to be dissolved initially in 10 mM in DMSO.
10. Cells are plated in a culture-treated 96-well plate (Corning,
Corning, NY), white and clear-bottom (catalogue number
3610). Before the experiments a white backing tape (Perkin
Elmer-catalogue number 6005199) has to be placed to cover
the bottom of the plate.
2.2. Plasmids 1. Plasmids containing the cDNA for the human Trace Amine
and Transfection Associated Receptor 1 (hTAAR1) were obtained from the
cDNA Resource Center at the University of Missouri-Rolla
and the American Type Culture Collection (Manassas, VA). A
variant of the wild-type form is used to improve the expression
at the plasma membrane (10).
2. HA-tagged D2 Long dopamine receptor is fused to Renilla
Luciferase at the C-terminus (5).
3. Mouse β-arrestin2 is fused to YFP at the C-terminus (5).
4. The BRET biosensor EPAC is a modification from an existing
FRET biosensor (10).
5. pRluc-N or pRluc-C vectors are from Perkin Elmer (Downers
Grove, IL).
6. pEYFP-N or pEYFP-C vectors are from BD Biosciences (San
Jose, CA).
7. pcDNA 3 is from Invitrogen.
8. Regarding calcium-phosphate transfection, calcium chloride
(Sigma) is dissolved at 2.5 M in Milli-Q water. HEPES buff-
ered saline (HBS) 2× is composed of sodium chloride (Sigma)
(2.5 M), HEPES (0.5 M), and sodium phosphate dibasic
(1 M) (Sigma). The pH of the solution is adjusted to 7.1. Both
solutions should be stored at 4°C and kept sterile by opening
only under the cell culture hood.
2.3. Experimental 1. To read the plate use the Mithras LB940 (Berthold
Machine and Software Technologies, Germany) or the Infinite F500 (Tecan,
Switzerland) with a MicroWin 2000 or an i-control software
respectively.
2. Data are analyzed using GraphPad Prism 5.
3. Methods
EPAC cAMP biosensor, as well as β-arrestin2 recruitment to D2R,

relies on the same principle of function that is BRET. BRET is a
form of non-radiative energy transfer between a donor (Renilla
Luciferase, Rluc) and an acceptor (YFP) (15). Different couples of
donor and acceptor exists, please see ref. (16) for review. Upon
oxidation of its substrate coelenterazine h, Rluc emits light with a
peak at 475 nm. If the donor is in the proximity of the acceptor
(around 0.1 nm), the energy emitted from Rluc can excite the YFP
that in turn emits light with a peak at 530 nm. The change in the
ratio between emissions by Rluc and the YFP reflects a change in
the distance between the two molecules. In the EPAC biosensor,
at basal or low cAMP levels, Rluc and YFP are in close proximity
resulting in a constitutive elevated BRET signal. When cAMP
binds EPAC, a conformational change occurs increasing the dis-
tance between the donor and the acceptor resulting in a decreased
change of BRET ratio. In β-arrestin2-D2R experiments, when DA
or another agonist stimulates the receptor (D2-Rluc), β-arrestin2-
YFP is recruited from the cytosol to the plasma membrane (to bind
D2R) and this translocation results in an increase of BRET ratio.
In this way, it is possible to study in real time these physiological
events by integrating the light emission by the two tags with appro-
priate filters (17). Regarding TAAR1, EPAC biosensor has been
shown to be a useful tool in studying trace amines pharmacology
and in finding new agonists of the receptors (10, 18). It has dem-
onstrated a good sensitivity and very similar results compared to
other techniques (β-PEA and 3-MT potencies to TAAR1).
Similarly, β-arrestin2 recruitment to the D2R as measured by
BRET assay has been validated as an effective method to study this
signaling pathway in antipsychotics pharmacology (5).
3.1. Cells Seeding 1. The wild type HEK-293T cells are cultured in 150-mm dishes
and Transfection with DMEM for the maintenance of a certain number of cells
according to the experiments planned. The medium should be
warmed in a 37°C bath for at least 20 min before using it. Cells
are passaged and seeded in 100-mm dishes the day before the
transfection using trypsin-EDTA. All the media and solutions
must be opened only under the hood (see Note 1).
2. Wash with 10 mL of PBS the 150-mm dishes, add 2 mL of

trypsin-EDTA to each dish and let the trypsin to act for
1–5 min at RT.
3. Add about 5 mL of DMEM in order to inactivate trypsin and
to help cell harvesting. Gently pipette up and down to detach
all the cells from the dish and to break cells lumps.
4. Pour the cells in a 15 mL Falcon tube and count them. Mix
20 μL of cells and 20 μL of trypan blue solution on a piece of
parafilm and then pipette 20 μL in the bunker chamber and
count the cells under the microscope. Make sure to make a
homogeneous cell suspension in the medium with the pipette
before collecting the 20 μL in order to have a representative
concentration of the solution.
5. After counting the number of cells/mL seed about 3 million of
cells for each 100-mm dish in 10 mL of DMEM (see Note 2).
Each dish will be used for a single transfection, including con-
trol mock transfection with an empty vector (see Note 3).
6. The morning after the seeding, cells will be transfected. The
solutions required are the CaCl2 (2.5 M), the HBS 2× and
water. All the solutions must be sterilized by filtration with a
0.22 μm filter and then opened only under the hood. All the
working aliquots of these solutions (10–50 mL) should be kept
in a refrigerator and for more reproducible results equilibrated
to room temperature (RT) for 30 min before using them. The
remaining aliquots of HBS 2× should be stored at −20°C (see
Note 4).
7. Take the DNA tubes out of the freezer (see Note 5) and let
them thaw at RT.
8. Place 15 mL Falcon tubes under the hood in a rack, one for
each transfection.
9. Put the DNA in each tube. It is important to have the same
amount of DNA for each transfection, so equilibrate with
empty vector (pcDNA 3). Usually, the amounts for a 1 mL of
solution, that is enough for a 100-mm dish, are as follows:
EPAC (3 μg), TAAR1 (5 μg), D2R (2 μg), and β-arrestin2
(2 μg). Regarding D1R, they have sufficient endogenous
expression in HEK-293T cells to elicit a response detectable by
BRET upon their stimulation.
10. Add 50 μL of CaCl2 2.5 M and enough water to have a total
volume of 0.5 mL containing the DNA as well.
11. Pipette up and down for few times to mix the solution.
12. Add 0.5 mL of the HBS 2× drop by drop to each tube and mix
by bubbling air inside the solution with the P1000 pipette for
10–15 times.
13. Add the transfection solution drop wise to each dish, gently
and trying to cover the entire plate. Rock the dishes back and
forth for a few times before replacing them in the incubator.
After few hours, a small dark precipitate should be observed in
the cells under the microscope.
3.2. Plating 1. Cells are normally plated in the 96-well plate after 16–24 h,
but for TAAR1 it is better to plate the same day, in the after-
noon, so as to do the experiment 1 day after the transfection
since TAAR1 is partially degraded in vitro (10).
2. After 4–6 h post transfection for TAAR1 and 16–24 h for
dopamine receptors and β-arrestin2 (see Note 6) cells have to
be plated.
3. The assay plate is a culture-treated, white, clear-bottom 96-well
plate. The plate is sterile so it has to be properly handled under
the hood. At this step the phenol red free Minimum Essential
Medium (MEM) containing 2% of FBS, 10 mM HEPES, and
2 mM L-glutamine should be used to avoid color in the medium
that could disturb the light emission during the experiment
(“clear medium”).
4. Place the poly-D-lysine and the PBS under the hood.
5. While pre-warming the medium, pipette approximately 100 μL
of poly-D-lysine in each well of the 96-well plate and incubate
for 10–20 min (see Note 7).
6. After this period re-collect the poly-D-lysine with the pipette
and wash once all the wells with PBS (about 150 μL for each
well). Then remove the remaining PBS with a vacuum
aspirator.
7. Wash twice the transfected cells in the 100-mm dishes with
PBS and detach them by pouring 1 mL of trypsin-EDTA for
each dish.
8. After 5 min of incubation (see Note 8), add 4 mL of clear
medium in each dish. Gently pipette the cells up and down to
detach them. Pour them in a labeled 15-mL Falcon tube, one
for each transfection.
9. Spin the tubes in a centrifuge for 5 min at 200 × g to pellet the
cells.
10. Remove the medium from the tube and resuspend the cells
with 1 mL of clear medium and count the cells for each tube as
described above.
11. Then dilute the cells solutions of each tube with the clear
medium so as to have a concentration of 750,000 cells for each
mL.
12. Put 100 μL of cells suspension for each well in the assay plate
for all the different transfections, according to the experiment
design. This step can be done either using a normal p100

pipette or a repeater pipette (see Note 9).
13. The 96-well plate is stored in the incubator for at least 12 h
before the experiment.
3.3. BRET Experiment 24–48 h post-transfection the EPAC sensor and the other proteins
should be sufficiently expressed. Since Rluc produces a signal inde-
pendently from BRET, an internal control of the level of transfec-
tion is represented by the basal Rluc counts. For Mithras, 100,000
counts are the minimum for a good signal to noise ratio.
1. Pre-warm to RT or to 37°C (depending on the planned exper-
iment) the PBS with calcium and magnesium and add
0.003%(wt/vol) of ascorbic acid. 30–50 mL of PBS solution is
usually enough for one 96-well plate and preparation of the
test compounds.
2. The compounds should be freshly prepared even if some com-
pounds are stable for some time if stored at −20°C in working
aliquots. For example IBMX is dissolved in dimethylsulfoxide
(DMSO) at the concentration of 200 mM (1,000× stock) and
it could be stored at −20°C in small aliquots (e.g., 200 μL).
Prepare all compounds to a concentration of 10 mM in the
same PBS with the ascorbic acid and then make the suitable
dilutions. β-PEA (TAAR1 agonist) is soluble in water and PBS
in the same way as 3-MT (TAAR1 agonist), isoproterenol (β2-
adrenergic receptor agonist), quinpirole hydrochloride (D2R
agonist). Haloperidol (D2R antagonist) is soluble at 10 mM in
DMSO but the subsequent dilutions (1:10, 1:100, 1:1,000,
and so on) should be made in PBS (see Note 10).
3. In case of experiments at RT, all the media and solutions should
be equilibrated at RT. In case of 37°C assay, switch on the plate
reader heating system at least 20 min before the experiments
and keep the PBS warm in a 37°C water bath (see Note 11).
4. Take the plate out of the incubator and check the cells under
the microscope. Before proceeding with BRET, stick the white
backing tape to the bottom of the plate to prevent light
dispersion.
5. Remove the clear medium from all the wells and replace it with
the PBS with calcium, magnesium and ascorbic acid.
6. Dilute the coelenterazine h stock solution 1:20 in PBS (10×
solution) and add to each well to yield a final concentration of
5 μM (see Note 12).
7. For the evaluation of cAMP levels using the EPAC sensor with
TAAR1, D1R or in general with a Gs-coupled receptor, after
adding coelenterazine, if necessary, also add IBMX to a final
concentration of 200 μM (see Note 13).
8. To evaluate agonist activity, such as β-PEA or 3-MT on TAAR1,

add the compounds 10 min after coelenterazine h (see Note
14) and immediately read the plate. If the experiment is con-
ducted at 37°C, place the plate in the incubator during the lag
of time before the agonist addition. β-PEA and 3-MT are
active at low micromolar concentration (see Fig. 1a, b).
9. To evaluate antagonistic activity against a known agonist (e.g.,
putative antagonists of TAAR1), the different compounds are
added 5 min before the agonist.
10. In case of unknown activity of the compounds or of the recep-
tor, a positive control could be made with the endogenously
expressed β2-adrenergic receptor and the agonist isoproterenol
a β-PEA
1.32
β-PEA(10–4M)
1.28 β-PEA(10–5M)
1.24 β-PEA(10–6M)
BRET ratio
β-PEA(10–7M)
1.20
β-PEA(10–8M)
1.16
1.12
0 500 1000 1500 2000

Time (sec.)
b
3-MT 3MT(10–4M)
1.32
3MT(10–5M)
1.28
3MT(10–6M)
1.24
BRET ratio
3MT(10–7M)
1.20 3MT(10–8M)
1.16
1.12
0 500 1000 1500 2000

Time (sec.)
Fig. 1. Evaluation of cAMP levels using cAMP BRET biosensor in HEK-293T cells express-
ing hTAAR1. (a) Time course evaluation of cAMP variations using different concentrations
of β-PEA. BRET ratio is calculated as Rluc/YFP ratio and the reading started after β-PEA
addition. The decrease in BRET ratio indicates an increase in cAMP levels. This experiment
is performed at RT with the addition of 200 μM of IBMX. (b) Similar experiment using
3-MT as TAAR1 agonist.
0.96 Forskolin + Quinpirole (10μM)

Forskolin (25μM)
0.94
Ctrl
0.92
0.90
Rluc/YFP
0.88
0.86
0.84
0.82
0.80
0 400 800 1200
Time (sec.)
Fig. 2. Evaluation of D2R activation using EPAC BRET biosensor at 37°C. Time course of
cAMP variations in HEK-293T cells expressing D2R and EPAC. Forskolin is added at the
beginning of the experiment at the time = 0. Quinpirole at 1 μM is added 5 min before
forskolin and 5 min after coelenterazine h (see Subheading 3). Forskolin, by the activation
of adenylate cyclase, increases cAMP levels and this effect is prevented by D2R activation
with quinpirole.
that is active at nanomolar concentrations. Alternatively, the

activator of adenylate cyclase forskolin may be used
(2–20 μM).
11. For receptors such as D2R, that inhibit adenylate cyclase and
the formation of cAMP, forskolin at 2–20 μM is used to increase
basal cAMP level (see Note 15).
12. Add quinpirole at 1 μM 5 min after Rluc substrate, forskolin
5 min after quinpirole and then read the plate immediately
(Fig. 2) (see Note 16).
13. In case of evaluation of antagonists on quinpirole effect, these
compounds are added simultaneously with coelenterazine h.
For example, for a haloperidol dose–response, add in each well
different concentrations from 10−11 to 10−5 M (Fig. 3a).
14. For the evaluation of β-arrestin2 recruitment to D2R, the
incubations are carried out at RT but the protocol remains the
same, add antagonists 5 min and agonist 10 min after Rluc
substrate and read the plate right after agonist addition.
Haloperidol is active in preventing β-arrestin2 recruitment at
low nanomolar concentrations (Fig. 3b).
15. Usually the plate is read for 20–30 min but, depending on the
receptor, the time course could be longer or shorter (see
Note 17).
3.4. Plate Reader The plate reader used for this assay is the Mithras LB940 that
Settings and Data allows the integration of the luminescent signal detected in the
Analysis 465–505 nm and 515–555 nm windows by using filters with the
appropriate band pass and the MicroWin software. The BRET
a 140
120
100
(% of Quinpirole inhibition)
cAMP accumulation
80
60
40
20
–20
–40
–12 –11 –10 –9 –8 –7 –6 –5 –4
Log [Haloperidol] (M)
b 120
100
β-arrestin 2 recruitment
(% of Quinpirole effect)
80
60
40
20
0
–12 –11 –10 –9 –8 –7 –6 –5 –4
Log [Haloperidol] (M)
Fig. 3. Evaluation of antagonistic activity of haloperidol on Gi activation or on β-arrestin2

recruitment induced by quinpirole. (a) HEK-293 T cells are transfected with EPAC biosen-
sor and D2R and are treated with different concentrations of haloperidol and with 1 μM of
quinpirole in the presence of forskolin. (b) Dose–response of haloperidol to evaluate the
inhibition of β-arrestin2 recruitment induced by 1 μM of quinpirole. Cells are transfected
with D2R-Rluc and β-arrestin2-YFP. Data are plotted as SEM of 3–5 independent
experiments.
signal is determined by calculating the ratio between the light

emitted at 465–505 nm and the light emitted at 515–555 nm. The
data are plotted as BRET ratio over the time in time course experi-
ments. In dose–response graphs, curves are fitted using a non-
linear regression and log vs. response fit using GraphPad Prism 5
(Fig. 4). In the experiment regarding D2R antagonist effect on
140
120
100
(% of maximum response)
cAMP accumulation
80
60
40
20
–20
–40
–12 –11 –10 –9 –8 –7 –6 –5 –4
Log [β-PEA] (M)
Fig. 4. cAMP variations induced by a β-PEA dose–response in cells co-expressing TAAR1

and EPAC. Cells are treated with different concentrations of β-PEA and plotted as a dose–
response experiment. β-PEA is used at concentrations ranging from 10−11 to 10−4 and the
plate is read 15 min after agonist addition. A non-linear regression with one site-specific
binding is used to draw the curve using GraphPad Prism5. Data are plotted as SEM of 3–5
independent experiments.
quinpirole activity, in the y-axis BRET ratio is transformed in the

percent of quinpirole effect (Fig. 3a). In β-arrestin2 recruitment,
the D2R antagonists should be plotted with the percentage of
quinpirole effect over the log of the concentration of the antago-
nist (Fig. 3b) (see Note 18).
4. Notes
1. All the media, solutions and instruments used for cell culturing
should be maintained sterile. It is important to open the medium
only under the hood and clean every object by spraying some
ethanol 70°C before bringing them under the hood. Bacteria
or yeast contaminations of cell dishes are easily detected by
looking at them under the microscope or by looking at the
color and clearness of the medium. The water and the solutions
for the transfection must be sterilized by filtration under the
wood with a 0.22 μm filter. The water used for all procedures is
a Milli-Q grade water, with a resistivity of 18.2 MΩ-cm and
total organic content of less than 5 parts per billion.
2. Depending on the speed of growth of the clone of HEK-293T
cells or the time of the day of seeding it is possible to seed less
or more cells. It is important to have cells at 70–80% confluency

the day of transfection.
3. Since each well (of the 96-well plate) should contain approxi-
mately 75,000 cells, usually one 100-mm plate is enough for
single transfection to yield a full 96-well plate. In case of mul-
tiple transfections and when the experiment design calls for
fewer cells, 35-mm dishes can be used to transfect cells. In this
case the procedures remain the same by just scaling down all
the reagents and solutions used proportionally. In contrast,
when many cells with a single transfection are needed, more
than one 100-mm dish could be used. In order to have a
homogeneous sample, mix all the similar transfections together
before seeding them in the assay plate.
4. The critical parameter for the success of the transfection is the
pH of the HBS 2× solution. It is suggested to prepare a fairly
good amount of solution since it could be tedious to find the
right pH each time the solution has to be prepared. After all
the reagents are dissolved, the pH should be adjusted to
7.1 ± 0.05. Since there are some variables such as incubator or
pH meter dysfunctions or unpredictable ones, it is advisable to
prepare different HBS 2× solutions in a range of pH from 7.0
to 7.2 and test them in different transfections in order to find
the one with the highest transfection efficacy. Then sterilize by
filtration and aliquot into 50 mL Falcon tube and store at
−20°C. Working aliquots can be stored at 4°C.
5. Normally, keep the DNA at −20°C but for long-term storage
put it at −80°C. Be sure that the solution is clear and there are
no precipitate or strange materials that could indicate the pres-
ence of molds. In such a case, it is imperative to re-prepare the
plasmid.
6. For TAAR1 it is better to do the experiments 24 h after trans-
fection. D2R and β-arrestin2 also have a strong protein expres-
sion after 24 h of transfection, so in cases of co-transfecting
TAAR1 and D2R, it is possible to do experiments at 24 h.
7. Sometimes the poly-d-lysine treatment is prolonged for hours,
for low adherent cells, but usually for HEK cells 10–20 min
should be enough. It is possible to reuse the poly-d-lysine up
to three times, so re-collect it and keep it sterile in the refrig-
erator after two uses.
8. Some HEK clones are less or more “attached” to the dish.
In some case, while passaging cells, PBS is enough to detach
the HEK cells; in other cases trypsin is necessary. The latter is
especially the case after a calcium-phosphate transfection.
9. With these dilutions, 75,000 cells/well is seeded and should
be enough to have a good signal to noise ratio. The number
may change from 25,000 to 100,000 per well depending on
the BRET signal, cells availability (e.g., in case of BRET

experiments in primary neurons) and other variables.
10. To minimize the final concentration of DMSO to a maximum
of 0.1–0.5% in order to avoid DMSO toxicity, haloperidol and
other water-insoluble compounds should be diluted as much
as possible in PBS. Appropriate controls should be used to
verify the inactivity of the dissolving medium.
11. For β-arrestin2 recruitment to D2R, RT experiment is more
convenient since the recruitment is slower and more visible
when analyzing the data. cAMP evaluation with EPAC sensor
could be done at 37°C or even at RT. The desensitization of
the cAMP response of the receptor of interest is more readily
observable at 37°C.
12. For BRET experiments, it is convenient, in term of calcula-
tions, to have a final volume of 100 μl. In this case, add 10 μL
of coelenterazine h and prepare all compounds as 10× solu-
tions in Eppendorf tube in order to add 10 μL for each of
them that have to be tested. In this case, make the calculations
of how many microliters of PBS that have to be added at the
beginning of the experiment (e.g., for testing 1 agonist (10 μL)
and 1 antagonist (10 μL), plus coelenterazine h (10 μL) and
IBMX (10 μL), put 60 μL of PBS in each well in order to
attain a final volume of 100 μL, 10 + 10 + 10 + 10 + 60 = 100).
13. To increase cAMP levels, IBMX may be used. IBMX blocks the
degradation of the cAMP by endogenous phosphodiesterases
and may be useful to increase the signal induced by the recep-
tor of interest.
14. The oxidation of coelenterazine h by Rluc takes several min-
utes to equilibrate, so to have a stable and good signal it is
better to wait at least 5 min (ideally 10 min) from the addition
of the substrate. If the transfection is efficient, each well of the
plate consists of one data point and thus it is not necessary to
perform the experiment in duplicates. Duplicate reads however
are often recommended.
15. In vitro, without any stimuli, the inhibition of adenylate cyclase
does not induce a significant decrease of cAMP that is detect-
able by our technique. So, to study Gi-coupled receptors such
as D2R activity, adenylate cyclase has to be activated by
forskolin.
16. Quinpirole has its maximum effect at 1 μM and the EC50 is
5.7 nM.
17. The effect of the agonist on cAMP production is quite fast
even at RT, so it is necessary to work rapidly in order to mea-
sure the beginning of the effect. As an alternative, the injector
in the machine could be used to dispense the agonist to differ-
ent wells, but take into account the number of compounds
that have to be dispensed and the availability as well as the cost

of the substances.
18. Another machine tested for this application is the Infinite F500
(Tecan). It has two appropriate filters to select the light emit-
ted by Rluc and the YFP and as the Mithras two injectors and
the heating system. While the Mithras is fairly more sensitive,
Infinite’s software is more easy-to-use and flexible.
Acknowledgments
Supported in part by research grants from the Michael J. Fox

Foundation for Parkinson’s Research, Fondazione Compagnia di
San Paolo (Torino, Italy) and research grant from F. Hoffmann-La
Roche Ltd. (Basel, Switzerland) to Raul R. Gainetdinov. Bernard
Masri was a recipient of a European Marie-Curie Outgoing
International Fellowship (FP6—2005-Mobility-6). Ali Salahpour
is supported by grants from Canadian Institute of Health Research
(CIHR # 210296) and Natural Sciences and Engineering Council
of Canada (NSERC # 386422).
References
1. Beaulieu JM, Gainetdinov RR (2011) The cholamine neurotransmitters are agonists of a

physiology, signaling and pharmacology of rat trace amine receptor. Mol Pharmacol
dopamine receptors. Pharmacol Rev 63: 60:1181–1188
182–217 7. Borowsky B, Adham N, Jones KA, Raddatz R,
2. Beaulieu JM, Gainetdinov RR, Caron MG Artymyshyn R, Ogozalek KL, Durkin MM,
(2007) The Akt-GSK-3 signaling cascade in Lakhlani PP, Bonini JA, Pathirana S, Boyle N,
the actions of dopamine. Trends Pharmacol Sci Pu X, Kouranova E, Lichtblau H, Ochoa FY,
28:166–172 Branchek TA, Gerald C (2001) Trace amines:
3. Lefkowitz RJ, Shenoy SK (2005) Transduction identification of a family of mammalian G pro-
of receptor signals by beta-arrestins. Science tein-coupled receptors. Proc Natl Acad Sci U S
308:512–517 A 98:8966–8971
4. Beaulieu JM, Gainetdinov RR, Caron MG 8. Lindemann L, Meyer CA, Jeanneau K, Bradaia
(2009) Akt/GSK3 signaling in the action of A, Ozmen L, Bluethmann H, Bettler B,
psychotropic drugs. Annu Rev Pharmacol Wettstein JG, Borroni E, Moreau JL, Hoener
Toxicol 49:327–347 MC (2008) Trace amine-associated receptor 1
5. Masri B, Salahpour A, Didriksen M, Ghisi V, modulates dopaminergic activity. J Pharmacol
Beaulieu JM, Gainetdinov RR, Caron MG Exp Ther 324:948–956
(2008) Antagonism of dopamine D2 receptor/ 9. Bradaia A, Trube G, Stalder H, Norcross RD,
beta-arrestin 2 interaction is a common prop- Ozmen L, Wettstein JG, Pinard A, Buchy D,
erty of clinically effective antipsychotics. Proc Gassmann M, Hoener MC, Bettler B (2009)
Natl Acad Sci U S A 105:13656–13661 The selective antagonist EPPTB reveals TAAR1-
6. Bunzow JR, Sonders MS, Arttamangkul S, mediated regulatory mechanisms in dopaminer-
Harrison LM, Zhang G, Quigley DI, Darland gic neurons of the mesolimbic system. Proc
T, Suchland KL, Pasumamula S, Kennedy JL, Natl Acad Sci U S A 106:20081–20086
Olson SB, Magenis RE, Amara SG, Grandy DK 10. Barak LS, Salahpour A, Zhang X, Masri B,
(2001) Amphetamine, 3,4-methylene- Sotnikova TD, Ramsey AJ, Violin JD,
dioxymethamphetamine, lysergic acid Lefkowitz RJ, Caron MG, Gainetdinov RR
diethylamide, and metabolites of the cate- (2008) Pharmacological characterization of
membrane-expressed human trace amine- 15. Pfleger KD, Seeber RM, Eidne KA (2006)
associated receptor 1 (TAAR1) by a biolumi- Bioluminescence resonance energy transfer
nescence resonance energy transfer cAMP (BRET) for the real-time detection of
biosensor. Mol Pharmacol 74:585–594 protein-protein interactions. Nat Protoc
11. Sotnikova TD, Beaulieu JM, Espinoza S, Masri 1:337–345
B, Zhang X, Salahpour A, Barak LS, Caron 16. Pfleger KD, Eidne KA (2006) Illuminating
MG, Gainetdinov RR (2010) The dopamine insights into protein-protein interactions using
metabolite 3-methoxytyramine is a neuromod- bioluminescence resonance energy transfer
ulator. PLoS One 5:e13452 (BRET). Nat Methods 3:165–174
12. Ayoub MA, Pfleger KD (2010) Recent advances 17. Salahpour A, Espinoza S, Masri B, Lam V,
in bioluminescence resonance energy transfer Barak LS, Gainetdinov RR (2012) BRET bio-
technologies to study GPCR heteromerization. sensors to study GPCR biology, pharmacology,
Curr Opin Pharmacol 10:44–52 and signal transduction. Front Endocrinol
13. Marullo S, Bouvier M (2007) Resonance (Lausanne) 3:105
energy transfer approaches in molecular phar- 18. Espinoza S, Salahpour A, Masri B, Sotnikova
macology and beyond. Trends Pharmacol Sci TD, Messa M, Barak LS, Caron MG,
28:362–365 Gainetdinov RR (2011) Functional interaction
14. Ni Q, Titov DV, Zhang J (2006) Analyzing between Trace Amine Associated Receptor 1
protein kinase dynamics in living cells with (TAAR1) and dopamine D2 receptor. Mol
FRET reporters. Methods 40:279–286 Pharmacol 80:416–425
Chapter 9
Dopaminergic Regulation of Dendritic Calcium:

Fast Multisite Calcium Imaging
Wen-Liang Zhou, Katerina D. Oikonomou, Shaina M. Short,
and Srdjan D. Antic
Abstract
Optimal dopamine tone is required for the normal cortical function; however it is still unclear how cortical-
dopamine-release affects information processing in individual cortical neurons. Thousands of glutamater-
gic inputs impinge onto elaborate dendritic trees of neocortical pyramidal neurons. In the process of
ensuing synaptic integration (information processing), a variety of calcium transients are generated in
remote dendritic compartments. In order to understand the cellular mechanisms of dopaminergic modula-
tion it is important to know whether and how dopaminergic signals affect dendritic calcium transients.
In this chapter, we describe a relatively inexpensive method for monitoring dendritic calcium fluctuations
at multiple loci across the pyramidal dendritic tree, at the same moment of time (simultaneously). The
experiments have been designed to measure the amplitude, time course and spatial extent of action poten-
tial-associated dendritic calcium transients before and after application of dopaminergic drugs. In the
examples provided here the dendritic calcium transients were evoked by triggering the somatic action
potentials (backpropagation-evoked), and puffs of exogenous dopamine were applied locally onto selected
dendritic branches.
Key words: Action potential, Backpropagation, Voltage-gated calcium channels, Dopaminergic

modulation, Dopamine receptors, Dendritic excitability, Phasic dopamine signal
1. Introduction
The level of prefrontal cortex dopamine (DA) is critical for modu-

lating normal cognitive/behavioral processes (1). One important
hypothesis is that deviations from the critical levels can severely dis-
rupt cognition and result in mental disorders (2). At present, it is
not known how such levels of dopamine interact with prefrontal
cortex (PFC) neural circuits. Electron microscope studies in monkey
and human have found that dopamine synaptic contacts onto corti-
cal neurons are located almost exclusively on the distal dendrites
123
124 W.-L. Zhou et al.
and spines of distal dendrites (3). The same distal dendritic seg-
ments of pyramidal neurons primarily receive thousands of excit-
atory glutamatergic inputs (4). The first stage of synaptic integration
takes place at the actual site of glutamatergic inputs, in thin (basal,
oblique and apical tuft) dendrites of pyramidal neurons (5– 7).
Dendritic function is closely linked to local fluctuations in cytosolic
calcium. Calcium ions in the intracellular compartment trigger and
regulate an impressive number of basic cellular functions (8–12).
Based on recent published studies one can identify five main cate-
gories (Fig. 1a–e) of calcium transients that occur in dendrites of
CNS neurons during biological processes:
(A) Calcium sparks—spontaneous miniature releases (13).
(B) Massive synaptically evoked calcium release from internal
stores (14–16).
(C) Action potential-mediated calcium influx (17–20).
(D) Synaptically evoked dendritic calcium influx (21–25).
(E) Local-dendritic-spike-associated calcium influx (26–28).
In addition to dendritic calcium transients (Fig. 1) the modern
experimental techniques have characterized action potential (AP)-
induced calcium transients in axons (29) and more impressively in
axon terminals, both in vitro (30) and in vivo (31) settings. The
experimental tools for monitoring calcium transients in small neu-
ronal compartments have undergone a steady improvement in the
Fig. 1. Five categories of dendritic calcium signaling. Pyramidal neurons have basal, oblique and apical dendrites. The
amplitudes and spatial distributions of neuronal calcium transients are indicated by the intensity and shape of black shad-
ings. (a) Fast but small spontaneous releases of calcium have been detected along the apical dendrites of hippocampal
pyramidal neurons (arrow), as well as in the apical oblique branches and the cell body (not shown but see ref. 13). These
spontaneous events (sparks) represent miniature releases from internal stores (diameter = 2 μm, duration = 100 ms) and
could be important in brain development (11). (b) Synaptically evoked internal release of calcium engulfs the proximal
segment of the apical dendrite and slowly propagates into the cell body (15). (c) Action potential propagates backward
from the axon initial segment into the dendritic tree (backpropagation) causing activation of dendritic voltage-gated
calcium channels (18). AP-associated dendritic calcium signal regulates synaptic plasticity (61). (d) Activation of gluta-
matergic presynaptic axon terminals causes the opening of synaptic receptors and the influx of calcium into dendritic
spines (23). (e) Synchronous activation of spatially segregated glutamatergic presynaptic terminals triggers a local regen-
erative dendritic potential, which is manifested by a spatially restricted peak of calcium influx (27).
9 Dopamine and Dendritic Calcium 125
last 40 years (22, 32–36). An optical measurement of dendritic

calcium fluctuation has become a robust, reliable, and affordable
technique, which is now regularly used in a large number of labora-
tories around the world. The time is right to ask whether and how
stimulation of dendritic dopamine receptors modulate dendritic
calcium signaling as shown in Fig. 1. In the current chapter we
focus on only one of the five aforementioned categories:
AP-associated dendritic calcium transients (Fig. 1c).
The backpropagating action potentials invade distal dendritic
segments, depolarize the dendritic membrane and open voltage-
gated calcium channels (18). The dendrites of PFC neurons express
D1 receptors (37, 38) and N-type calcium channels (18, 39).
A physical link between D1R and N-type calcium channels has
been implicated as a key mechanism in the dopamine-induced
suppression of dendritic calcium influx during action potential
backpropagation (40). In their study the calcium measurements
were performed from only one dendrite class (apical) and at only
one location at the time before and after bath perfusion of dop-
amine (40). Although Kisilevsky et al. (40) provide experimental
evidence for dopamine-induced suppression of dendritic calcium
influx, this issue remains controversial because another group
claimed that in the same brain region, same cell type, same den-
drite (apical), same type of stimulation (dopamine in the bath) has
no effect on AP-mediated calcium influx (41). In this chapter, we
revisit dopaminergic modulation of action potential backpropaga-
tion, and we combine local dopamine applications with multisite
dendritic imaging. We attempt to mimic the phasic DA signal (42)
by locally applying DA from a glass pipette for a brief period of
time (2 s). We address other classes of dendrites (basilar branches
located in the deeper cortical layers where dopaminergic fibers
cluster). We investigate how different regions along the same den-
drite respond to DA receptor stimulation (a spatial aspect of mod-
ulation). Finally, we analyze the voltage waveform of a
backpropagated AP in the same dendritic segments, where phasic
DA stimuli caused suppression of the calcium influx.
2. Materials
2.1. Solutions 1. Extracellular solution. Artificial cerebrospinal fluid (ACSF, in

and Dyes mM): 125 NaCl, 26 NaHCO3, 10 glucose, 2.3 KCl, 1.26
KH2PO4, 2 CaCl2, and 1 MgSO4 (pH 7.4, osmolality
300–310 mosmol/L).
2. Intracellular solution. We use a standard gluconate-based
intracellular solution containing (in mM): 135 potassium glu-
conate, 2 MgCl2, 3Mg-ATP, 10 phosphocreatine, 0.3 GTP,
and 10 Hepes (pH 7.3, adjusted with KOH; osmolality

290–300 mosmol/L).
3. Calcium sensitive dyes: Calcium Green-1 (CG-1), Oregon
Green Bapta-1 (OGB-1), Fluo-5F, and bis-fura-2 were pur-
chased from Invitrogen (Carlsbad, CA) and dissolved directly
into the intracellular solution at 150–250 μM. Voltage-sensitive
dyes JPW1114 and JPW3028 were kindly provided by J.P.
Wuskell and Leslie Loew (University of Connecticut Health
Center, Farmington, CT) and dissolved directly into the intra-
cellular solution at 400 μM.
2.2. Electrophysiology 1. Amplifiers. We use Axon Instruments Multiclamp 700B, a

digitizer, Digidata Series 1322A, and Clampex software for
acquisition, and Clampfit software for the analysis of electrical
recordings (Molecular Devices, Sunnyvale, CA). One personal
computer (PC) is devoted to electrical recordings and for
taking photographs for morphological analysis of fluorescently
labeled neurons.
2. Second analog-to-digital board. The Neuroplex system
(RedShirtImaging LLC, Decatur, GA) is running on a PC
devoted to fast optical imaging only (e.g., calcium imaging).
Neuroplex system is equipped with an A-D board for storing
and synchronizing electrical data with the corresponding opti-
cal data. During a typical experiment the same electrical signal,
which is recorded by Axon Instruments apparata (aforemen-
tioned) on the first PC, is also stored inside the optical file on
a second PC.
3. Micromanipulators. Two motorized micromanipulators
(P-285, Sutter Instrument, Novato, CA) are used for position-
ing the patching pipette onto the cell body, and the dopamine
application pipette near the dendrite.
2.3. Optical 1. Microscope. Standard upright microscope for patching brain

Imaging Station slices (Olympus BX51WI, Olympus Inc., Japan).
2. Lenses. 40× water immersion objective with high N/A. 5×
objective for finding pipettes.
3. Microscope modules. Epi-illumination module, filter cubes,
and two camera ports for fluorescence imaging (Olympus
Inc.).
4. Camera port-1. One camera port is taken by a simple CCD
camera for infrared DIC video microscopy (there are many
models available—we use Dage IR-1000).
5. Camera port-2. The second camera port is taken by a fast low-
noise low-resolution data acquisition CCD camera
(RedShirtImaging LLC, Decatur, GA). We use 80 × 80 pixel
model (NeuroCCD-SMQ), which is suitable for both calcium
and voltage imaging from dendrites of CNS neurons, thanks to

superior sensitivity (low noise) and very fast frame rates (5, 27,
43, 44). The full frame rate is 2,000 frames per second. This
camera can take 5,000 pictures per second after binning or
10,000 pictures per second after a partial reduction of the
frame size (cropping) (45).
6. Software. Data acquisition software Neuroplex
(RedShirtImaging) is endowed with several useful features
including the examination of records immediately following
the acquisition trial, an automatic spatial averaging of individu-
ally selected pixels, and spike-triggered averaging of consecu-
tive sweeps during the actual data acquisition.
7. Light source. For epi-illumination we use 250 or 150 W xenon
arc lamps (Opti-Quip, Highland Mills, NY). These are low-
ripple (low-noise) lamps that have excellent power for whole-
field illumination, and they cover the entire relevant range for
dye excitations; from 340 to 700 nm. This is especially useful
for sequential voltage-calcium imaging using a cocktail of dyes
(27) or for combining AlexaFluor594 with green calcium dyes
(Calcium Green-1, Oregon Green Bapta-1, and Fluo-5F). For
important practical details on the use of Opti-Quip arc lamps
(see Notes 1 and 2).
8. Mechanical shutter. Electro-Programmable shutter system
CS35-Uniblitz was purchased from Vincent Associates
(Rochester, NY). The CS35 shutter is equipped with reflective
blades. This option protects the shutter mechanism from the
light source’s damaging effects by reflecting the energy away
from the blade surface.
9. Optical filters. Optical filters were purchased from Chroma
Technology (Rockingham, VT) and Omega Optical
(Brattleboro, VT). The filter set (cube) for Fluo-5F consists of
an Omega exciter 500AF25 (485–510 nm bandpass), dichroic
525DRLP and emitter 530ALP (530 nm longpass). Filters for
AlexaFluor594 were a Chroma exciter HQ580/20× (570–
590 nm bandpass), dichroic Q595LP and emitter HQ630/60 m
(600–660 nm bandpass). The filter cube for Ca-Green-1 con-
tains an Omega exciter 500AF25, dichroic 525DRLP, and
emitter 530ALP. The same filter set is used for voltage imaging
with JPW1114 or JPW3028 voltage-sensitive dyes. This filter
set consists of a Chroma exciter D510/60 (480–540 nm band-
pass), dichroic 570dcxru and emitter E600lp (600 nm long-
pass). The filter cube for bis-fura-2 contains a Chroma exciter
D380/30× (365–395 nm bandpass), dichroic 400dclp, and
emitter E470lp (470 nm longpass).
10. Vibration isolation table. We use BM-1 platforms made by
Minus K Technology, Inc. (Inglewood, CA).
3. Methods
3.1. Acute Slice Preparation of quality slices is a critical step in the experiment.
Preparation Sprague-Dawley rats (postnatal day 21–35) were deeply anesthe-
tized with isoflurane and decapitated according to an animal
protocol approved by the UConn Health Center Animal Care and
Use Committee. Coronal brain slices (300 μm thick) were har-
vested from the frontal lobe (anterior to genu of corpus callosum)
in gassed (95% O2 and 5% CO2), ice-cold artificial cerebrospinal
fluid (ACSF). Note that we cut slices in the same solution (ACSF)
which is used for experimental recordings (see Note 3). From the
ice-cold cutting chamber the slices were transferred to a warm and
oxygenated holding chamber (35°C) and incubated for 30 min at
35°C. Following a 30 min incubation, the slices were removed
from the warm water-bath and kept at room temperature (1–6 h)
before being transferred to the recording chamber. Prefrontal cor-
tical slices of good quality have an abundance of pyramidal cells in
superficial (Layers 2–3) and deep layers (Layers 5–6). Healthy
pyramidal cells have oval cell bodies, smooth membranes that
appear to shine under infrared differential interference contrast (IR
DIC) video microscopy, and very soft edges. Pyramidal cells that
stand out from the brain slice background, show rugged edges and
strong contrast are often bad and unhealthy cells. One side of the
brain slice is always better than the other. It is important to quickly
examine the slice (using a 40× objective) prior to securing it with a
slice-anchor (see Note 4).
3.2. Patch Electrode Patch pipettes were pulled from borosilicate glass (G150F-3,
Recordings Warner Instruments, Hamden, CT, USA) on a P-97 microelec-
and Dye Injections trode puller (Sutter Instruments, Novato, CA). The ideal pipette
resistance for loading pyramidal neurons with fluorescent dyes is
7 MΩ (see Note 5). All recordings were performed at 33–34°C.
Whole-cell recordings from layer 5 pyramidal neurons were carried
out using a Multiclamp 700B amplifier (Materials) and digitized
with two input boards: (1) Digidata Series 1322A (Subheading 2.2,
item 1) at a 10 kHz sampling rate and (2) Neuroplex
(Subheading 2.2, item 2) at a 4 kHz sampling rate. Only cells hav-
ing a membrane potential more hyperpolarized than −50 mV (not
corrected for liquid junction potential) and AP amplitudes >70 mV
(measured from the base line) were included in the study. Voltage-
sensitive dye (JPW1114 or JPW3028) and calcium-sensitive dyes
(Ca-Green-1, Oregon Green Bapta-1, bis-fura-2, and Fluo-5F), as
well as AlexaFluor594, were dissolved in intracellular solution and
loaded into the patch pipette. Impurities and dust particles in the
intracellular solution represent major obstacles for formation of
the seal and later for dye-injection into the cytosol (see Note 6).
To avoid extracellular deposition of the fluorescent dyes, glass
pipettes were filled from the tip with dye-free solution by applying
negative pressure (front-loading), and were back-filled with dye
solution (back-loading). This procedure is essential for loading
voltage-sensitive dyes into neurons in brain slices (43), and it is
considerably less important for calcium-sensitive dyes; though it
may improve the viability of calcium-loaded neurons in long exper-
iments. Intracellular staining was achieved by free diffusion of the
dye from the pipette into the cell body. Duration of dye loading
depends on the cell type, size and shape of the patch pipette, and
most importantly on the water-solubility of the fluorescent dye.
For example, voltage-sensitive dyes JPW1114 and JPW2030 are
lipophilic, and it takes at least 2 h to fill the apical tuft branches. In
the case of voltage-sensitive dyes the dye loading pipette must not
stay in whole-cell configuration for longer than 60 min; because
the overloading of the cell body compartment causes pharmaco-
logical and photodynamic damage (46). To prevent toxic effects of
voltage-sensitive dyes, after 40–60 min of dye-injection an outside-
out patch was formed and the patch electrode removed (46, 47).
Dye-injected neurons were next incubated for 2–3 h at room tem-
perature and repatched with a dye-free pipette, just prior to the
optical recording session.
Voltage-sensitive dye recordings from dendrites of CNS neu-
rons are beyond the scope of this chapter. For detailed description
of voltage-sensitive dye method see the most recent protocol (48).
3.3. Calcium Imaging Calcium-sensitive dyes are soluble in water and it takes approxi-
mately 30–35 min to properly load the majority of basilar and
oblique dendritic branches in layer 5 pyramidal neurons. For the
most distal apical tuft branches it may take more than 100 min of
dye loading. AlexaFluor594 is co-applied with calcium-sensitive
dyes to allow a quicker and better examination of the dendritic
tree, as well as to aid proper positioning of neurons inside the
visual field of the NeuroCCD-SMQ (Fig. 2), without having to
excite the calcium dye (see Note 7). While the calcium-sensitive
dye is diffusing from the patch pipette into the soma, and from
the soma into the target dendritic branch, the amplitude of
AP-induced calcium transient will change. The unstable ampli-
tude of a calcium transient during control measurements may
compromise the results, therefore, it is important to evaluate the
time-dependence of evoked dendritic calcium transients in the
absence of any conditioning (e.g., before dopaminergic stimula-
tion). We have established that approximately 45 min from the
beginning of dye loading procedure (whole-cell breakthrough)
the baseline measurements in basilar segments of the dendritic
tree become stable and remain at one fixed level for another
30–40 min (Fig. 2).
Fig. 2. Calcium imaging: Establish the baseline prior to testing working hypotheses. A PFC Layer 5 pyramidal neuron was
loaded for 30 min with OGB-1 [200 μM] and Alexa Fluor [60 μM]. The basilar dendritic tree was projected onto the
NeuroCCD using a ×40 objective lens. Action potential was evoked by the somatic current injection and the resulting
dendritic signals were recorded every 2–4 min from the entire visual field. Each panel (a–d) is devoted to one region of
interest marked by white box. The peak amplitude of calcium transient (obtained by averaging outputs of camera pixels
inside the box) is expressed as dF/F (%) and plotted versus time in the graph on the right. Time zero marks the end of the
dye-loading phase and it corresponds to the 30th minute after the whole-cell breakthrough. After 30 min of dye loading
each dendritic segment (a–d) exhibits a relatively stable AP-induced calcium signal for the next 35 min, which is plenty of
time to perform a biological experiment.
3.4. Calcium Imaging In order to mimic phasic dopaminergic signals (42) we loaded
of Dopamine-Induced 5 mM dopamine into a glass micropipette. With the aid of a motor-
Changes ized micromanipulator the DA application micropipette was
positioned in the vicinity of one basal branch. There is 20–30 μm
from the tip of micropipette to the dendritic shaft (Fig. 3a). The
positioning of the glass pipette onto a selected dendritic branch
was done by alternating between IR DIC (not shown) and
fluorescence video microscopy (Fig. 3a). Dopamine was pressure-
ejected for 2 s (computer-driven picospritzer), just prior to a cal-
cium-imaging sweep. Note that [5 mM] refers to the concentration
of dopamine inside the application pipette. The concentration of
dopamine that reaches the dendritic membrane at the end of a 2 s
long puff is likely one or two orders of magnitude lower. The dop-
amine-induced suppression in dendritic calcium transient is only
momentary, as the subsequent sweeps show rapid recovery of the
signal amplitude (Fig. 3b, Wash).
3.5. Spatial Aspect In their current state, the confocal microscopy and two-photon
of Dopamine-Induced imaging methods are not capable of monitoring the spatial distri-
Changes bution of dopamine-induced changes across the dendritic tree at a
200 Hz frequency (5 ms per full frame, Fig. 4). The present experi-
mental setup has been designed to monitor AP-associated tran-
sients from multiple loci (regions of interest, ROIs) across several
dendritic branches (Fig. 4), at the same time (simultaneously).
While the “target” dendrite (Fig. 4, ROI 2) is receiving a phasic
dopaminergic stimulus (DA), the neighboring branches belonging
to the same nerve cell (Fig. 4, ROI 5–7) can be used as an ideal
control (same cell, same instant of time). The spatial resolution of
the system, when used with a 40× objective, is approximately
Fig. 3. Calcium imaging: phasic dopamine stimulation. (a) A PFC Layer 5 pyramidal neuron was loaded for 25 min with
OGB-1 [200 μM] and Alexa Fluor [60 μM]. Time zero marks 25th minute after the whole-cell breakthrough. The basilar
dendritic tree was projected onto the NeuroCCD using a ×40 objective lens and a red fluorescence cube for Alexa Fluor
594 (Materials). Action potential was evoked by a somatic current injection and the resulting dendritic signal was recorded
from 8 camera pixels inside the white box, using the green fluorescence cube (calcium imaging). (b) AP-mediated dendritic
calcium transients were measured in intervals of 1–4 min. Each trace is the product of 8-pixel spatial averaging. Dopamine
was ejected for 2 s (total duration) just prior to the optical recording sweep. Recordings obtained before DA ejection are
considered control recordings (Ctrl.). Recordings obtained after DA ejection are used to estimate the temporal dynamics of
washout.
Fig. 4. Calcium imaging: spatially restricted effect of a phasic dopamine stimulus. A PFC Layer 5 pyramidal neuron was
loaded for 30 min with CG-1 [200 μM] and Alexa Fluor [60 μM]. The basilar dendritic tree was projected onto the NeuroCCD
using a ×40 objective lens. Scale bar, 50 μM. A single action potential was evoked by the somatic current injection and
the resulting dendritic signals were recorded simultaneously from the entire visual field. Only seven regions of interest
(ROIs) are selected for display (1–7). Optical traces were acquired before (Control) and after a local dopamine puff (DA).
Control recordings are marked by a dashed grey line. Dopamine recordings are marked by thick black line. ROI 0 indicates
a somatic whole-cell recording of evoked action potential. Asterisk marks dendritic segment experiencing the most severe
amplitude reduction in response to local dopamine puff (duration, 2 s). Note that ROIs 5, 6 and 7 experience no change at
the same moment of time (Antic lab, unpublished data).
5 μm × 5 μm per pixel. That is to say that each camera pixel covers

an area 5 μm × 5 μm in the object field. With an array composed of
80 × 80 pixels (NeuroCCD-SMQ) we can simultaneously sample
every branch in the visual field of roughly 400 μm × 400 μm. Such
resolution provides ample means to study the spatial extent of the
dopamine effect on dendritic calcium flux. In the example shown

in Fig. 4 the greatest suppression of calcium influx was detected in
dendritic segment “ROI 2,” which is closest to the dopamine
application site (DA). Due to the orientation of the ejection pipette
and the direction of dopamine jet (Fig. 4, arrow), the membranes
in the dopamine path were all affected (ROIs 1, 3, and 4), while
basal dendrites away from the dopamine stream showed no change
in signal amplitude (Fig. 4, ROIs 5–7).
3.6. Voltage–Calcium The amplitude of a backpropagating action potential is not uni-

Imaging of Dopamine- form along a weakly excitable dendrite (49, 50). Depending on a
Induced Changes neuron type (e.g., Mitral cell versus Purkinje neuron), or dendrite
type (e.g., basal dendrite versus apical tuft dendrite) and the physi-
ological state of the dendrite (e.g., recent synaptic history), the
amplitude of backpropagating AP may vary dramatically (51, 52).
Some neurotransmitters/neuromodulators, including dopamine
itself, have been reported to change AP amplitude in dendrites
(53, 54). We have observed a dopamine-induced suppression of
AP-associated dendritic calcium signal (Figs. 3 and 4). The first
logical question that comes to mind is whether this suppression
was caused by dopaminergic modulation of the action potential
waveform? In order to address this issue we performed voltage-
sensitive dye recordings and calcium-sensitive dye recordings from
the same neuron, same dendrite, same dendritic segment (Fig. 5a,
white box) before and after local dopamine application (Fig. 5b).
To achieve this dual-mode optical recording, the neurons were
loaded with a cocktail consisting of one voltage-sensitive dye
Fig. 5. Voltage: Calcium imaging of DA-induced changes. A PFC Layer 5 pyramidal neuron was loaded for 45 min with a
mixture containing bis-fura [200 μM] and JPW3028 [400 μM] and then the loading pipette was removed. Following a
90 min of post-loading incubation the neuron was repatched with solution containing bis-fura but not JPW3028. (a) The
basilar dendritic tree was projected onto the NeuroCCD using a ×40 objective lens. Action potential was evoked by the
somatic current injection and the resulting dendritic signals were recorded from 8 camera pixels inside the white box
(region-of-interest, ROI). Scale bar, 50 μm. (b) The AP-evoked dendritic signal was first recorded using a filter cube for
voltage-sensitive dye JPW3028 (Voltage) before (Control), upon dopamine ejection (DA) and 2 min after the dopamine
ejection (Wash). Five minutes later, the AP-evoked signal was recorded using a filter cube for bis-fura (Calcium) in three
conditions (Control, DA and Wash). The same subset of 8 pixels (inside the ROI) was used to produce a spatial average in
both voltage and calcium modes—traces displayed in (b). Note that during DA stimulus the dendritic transients experience
amplitude reduction in calcium channel, but not so prominent in voltage channel (Antic lab, unpublished data).
(JPW3020) and one calcium-sensitive dye (bis-fura-2). It is

important to note that the excitation spectra of these two dyes do
not overlap. Bis-fura excitation filter is set at 380 nm, while JPW2038
excitation filter is set at 520 nm; more than 100 nm away from each
other. In a typical experiment, AP-associated dendritic transients
were recorded with one filter set (filter cube) and then the measure-
ments were repeated using the same stimulation (single AP) but
different filter cube. In between optical recording sweeps the X-Y
position of the dendritic tree and dendritic focus were kept fixed.
Using the sequential voltage-calcium recordings we found that brief
dopaminergic stimulations strongly affect calcium (Fig. 5b, Calcium)
but not so strongly the voltage signals (Voltage) in the same den-
dritic segment.
3.7. Advantages Recent studies have emphasized the advantages of two-photon

and Disadvantages microscopy for studying calcium dynamics in thin branches of CNS
neurons (23). Here we point out the disadvantages of confocal and
two-photon microscopy approaches, in direct comparison to our
system.
1. The experimental setup described in this chapter is several
times cheaper than a commercially available two-photon sys-
tem. While only a selected few laboratories can afford two-
photon and confocal microscope-based experimental rigs, our
system can be installed on virtually any brain slice electrophysi-
ology setup with a simple addition of one fast CCD camera,
one arc lamp and one mechanical shutter.
2. The maintenance of our system is easier, cheaper and faster.
The same is true for time and effort that must be invested to
train users to perform experiments.
3. High-frequency optical recordings from several dozen loci at
the same moment of time (Fig. 4) are currently not feasible
with two-photon systems, and therefore the complex spatiotem-
poral patterns of synaptic integration (27, 44, 55–59), or neu-
romodulation (60), are difficult to study on these systems.
4. Expensive purchase of laser lines and extensive rebuilding is
necessary to adapt a confocal system to perform sequential
voltage-calcium recordings shown in Fig. 5. In the case of our
system the user only needs to purchase one additional filter
cube.
4. Notes
1. Overheating of the light source. Both 250 and 150 W xenon

arc lamps made by Opti-Quip, Highland Mills, NY, USA have
excellent stability of the photon output but unfortunately they
lack proper cooling systems. Both lamps are prone to frequent

breakdowns, which is a major setback in a research project. To
prevent frequent breakdowns one should install computer fans
on both ends of the lamp house (Model 770U, Opti-Quip).
The bottom fan should direct air upwards into the bottom
vents. The top fan should collect air coming out from top
vents. In this way the lamp house is kept comfortably warm
(no overheating), which dramatically extends the life of the
Xenon bulb and reduces the number of malfunctions. However,
the fans must not be in any physical contact with the lamp
house, because the lamp house is rigidly mounted to the back
of the microscope and provides the direct pathway for mechan-
ical vibrations to get introduced into optical records. To hold
the two fans in position we use metal rods attached to the wall
outside the electrophysiology rig.
2. Mounting of the light source. We do not use optic guides
because regardless of their quality the optic fibers significantly
reduce the number of photons emanating from the light
source. Instead, the arc lamp is mounted directly to the back of
the microscope. Between the lamp and the microscope we
position (secure) a custom-made aluminum box, which pro-
vides a narrow spacing for insertion of the optical shutter. The
optical shutter must not be in any physical contact with the
custom-made aluminum box, because the box is rigidly
mounted to the microscope and provides a direct pathway for
mechanical vibrations to get introduced into optical records.
Instead, the shutter is suspended on a horizontal rod that
branches off a large vertical rod, which is mounted outside the
antivibration table by its bottom end. In this way mechanical
vibrations caused by shutter opening are interrupted on the
way to the microscope by the vibration isolation table
(Subheading 2.3, item 10).
3. Cold ACSF. Note that we cut slices in the same solution
(ACSF) which is used for experimental recordings
(Subheading 2.1, item 1). We obtain good quality slices if we
cool down the ACSF to the point where thin ice flakes are
floating in the beaker. However, this initial cooling is not
sufficient as ACSF rapidly warms during slicing on a tissue
slicer. In order to slow down warming of the ACSF we intro-
duce a cold aluminum block (5 cm × 3 cm × 1 cm) inside the
ASCF filled cutting chamber. The aluminum block and the
cutting chamber (removable parts) are both stored in the
freezer between slicing.
4. Slice flipping. We use a custom-made anchor (nylon grid) to
secure brain slices to the bottom of the recording chamber.
Before final positioning of the anchor, the brain slice needs to
be properly oriented (flipped). One side of the brain slice is
always better than the other. The good side must face the
objective lens. On the good side, the apical dendrites travel
parallel to the surface of the slice, or dive into the slice at a very
shallow angle. If apical dendrites are coming out of the slice
surface, then large portions of the apical dendritic tree have
met the razor blade and were damaged in the brain slice prepa-
ration step. Neurons with severed apical dendrites are not suit-
able for present experiments.
5. Ideal pipette resistance. We have empirically determined that
7 MΩ pipettes are ideal for loading layer 5 pyramidal neurons
with calcium-sensitive and voltage-sensitive dyes (43). Pipette
with smaller tips (pipette resistance > 7 MΩ) produce inade-
quate loading. Pipettes with larger tips (pipette resis-
tance < 7 MΩ) produce too much damage, especially if such
neuron was meant to be repatched (43, 46).
6. Clean pipettes. To prevent pipette clogging Borosilicate
electrode glass (o.d. = 1.5, i.d. = 0.86 mm) was prewashed in
boiling ethanol alcohol, rinsed in acetone and dried. Both dye-
free and dye-rich solutions were filtered through nylon syringe
filters, pore size 0.2 μm (Nalgene 4-mm).
7. Reduce exposure to excitation light. The brightness of calcium
probes (Subheading 2.1, item 3) is poor compared to other
fluorescent markers for intracellular application in modern
neurobiology (e.g., Rhodamine, AlexaFluor). To make things
worse, the excitation-emission spectra of calcium-sensitive dyes
overlaps with the brain slice-autofluorescence; to further
deteriorate image quality. Considerably better images can be
obtained by loading neurons with red dyes Rhodamine or
AlexaFluor594. In our experiments (Figs. 2, 3, 4, and 5)
neurons were loaded with a mixture containing one calcium-
sensitive dye and one red dye (AlexaFluor594 or JPW3028).
During dye loading, positioning and focusing onto the “tar-
get” dendrites we use a filter cube for red dyes. By inserting a
neutral density filter in the epi-illumination light path we
reduced epi-illumination light intensity down to 5–10% of
what is normally used for calcium-imaging sweeps. We used a
manually controlled shutter to keep the illumination episodes
very brief (2–3 s) and very seldom. In summary four steps are
regularly used to minimize the photodynamic damage from
calcium-sensitive dyes prior to the beginning of calcium sensi-
tive dye measurements. Basically, in order to position and focus
fluorescently labeled neurons for calcium imaging:
(a) Use excitation wavelength for AlexaFluor594.
(b) Reduce epi-illumination light intensity.
(c) Reduce shutter open-time to less than 3 s per opening.
(d) Limit the number of positioning and focusing adjustments
to less than 5 per experiment.
Acknowledgments
This work was supported by an R01 grant from National Institutes

of Health (NIH)—grant number MH063503, and the NARSAD
Young Investigator Award to S.D.A.
References
1. Brozoski TJ, Brown RM, Rosvold HE, frequency of spontaneous elementary Ca2+
Goldman PS (1979) Cognitive deficit caused release events in the dendrites of pyramidal
by regional depletion of dopamine in prefrontal neurons. J Neurosci 29:7833–7845
cortex of rhesus monkey. Science 205: 14. Emptage N, Bliss TV, Fine A (1999) Single
929–932 synaptic events evoke NMDA receptor-medi-
2. Carlsson A (1988) The current status of the ated release of calcium from internal stores in
dopamine hypothesis of schizophrenia. hippocampal dendritic spines. Neuron
Neuropsychopharmacology 1:179–186 22:115–124
3. Goldman-Rakic PS, Leranth C, Williams SM, 15. Nakamura T, Barbara JG, Nakamura K, Ross
Mons N, Geffard M (1989) Dopamine synap- WN (1999) Synergistic release of Ca2+ from
tic complex with pyramidal neurons in primate IP3-sensitive stores evoked by synaptic activa-
cerebral cortex. Proc Natl Acad Sci U S A tion of mGluRs paired with backpropagating
86:9015–9019 action potentials. Neuron 24:727–737
4. Elston GN (2003) Cortex, cognition and the 16. Hagenston AM, Fitzpatrick JS, Yeckel MF
cell: new insights into the pyramidal neuron (2007) MGluR-mediated calcium waves that
and prefrontal function. Cereb Cortex invade the soma regulate firing in layer V medial
13:1124–1138 prefrontal cortical pyramidal neurons. Cereb
5. Milojkovic BA, Radojicic MS, Goldman-Rakic Cortex 18(2):407–23
PS, Antic SD (2004) Burst generation in rat 17. Jaffe DB, Johnston D, Lasser-Ross N, Lisman
pyramidal neurones by regenerative potentials JE, Miyakawa H, Ross WN (1992) The spread
elicited in a restricted part of the basilar den- of Na+ spikes determines the pattern of den-
dritic tree. J Physiol 558:193–211 dritic Ca2+ entry into hippocampal neurons.
6. Polsky A, Mel BW, Schiller J (2004) Nature 357:244–246
Computational subunits in thin dendrites of 18. Markram H, Helm PJ, Sakmann B (1995)
pyramidal cells. Nat Neurosci 7:621–627 Dendritic calcium transients evoked by single
7. Larkum ME, Nevian T, Sandler M, Polsky A, back-propagating action potentials in rat neo-
Schiller J (2009) Synaptic integration in tuft cortical pyramidal neurons. J Physiol
dendrites of layer 5 pyramidal neurons: a new 485:1–20
unifying principle. Science 325:756–760 19. Svoboda K, Denk W, Kleinfeld D, Tank DW
8. Bito H, Deisseroth K, Tsien RW (1997) (1997) In vivo dendritic calcium dynamics in
Ca2 + -dependent regulation in neuronal gene neocortical pyramidal neurons. Nature
expression. Curr Opin Neurobiol 7:419–429 385:161–165
9. Lisman J, Malenka RC, Nicoll RA, Malinow R 20. Waters J, Larkum M, Sakmann B, Helmchen F
(1997) Learning mechanisms: the case for (2003) Supralinear Ca2+ influx into dendritic
CaM-KII. Science 276:2001–2002 tufts of layer 2/3 neocortical pyramidal neu-
rons in vitro and in vivo. J Neurosci
10. Zucker RS (1999) Calcium- and activity- 23:8558–8567
dependent synaptic plasticity. Curr Opin
Neurobiol 9:305–313 21. Regehr WG, Tank DW (1990) Postsynaptic
NMDA receptor-mediated calcium accumula-
11. Lohmann C (2009) Calcium signaling and the tion in hippocampal CA1 pyramidal cell den-
development of specific neuronal connections. drites. Nature 345:807–810
Prog Brain Res 175:443–452
22. Miyakawa H, Ross WN, Jaffe D, Callaway JC,
12. Yu LM, Goda Y (2009) Dendritic signaling and Lasser-Ross N, Lisman JE, Johnston D (1992)
homeostatic adaptation. Curr Opin Neurobiol Synaptically activated increases in Ca2+ concen-
19:327–335 tration in hippocampal CA1 pyramidal cells are
13. Manita S, Ross WN (2009) Synaptic activation primarily due to voltage-gated Ca2+ channels.
and membrane potential changes modulate the Neuron 9:1163–1173
23. Denk W, Yuste R, Svoboda K, Tank DW (1996) Wide-field and two-photon imaging of brain
Imaging calcium dynamics in dendritic spines. activity with voltage- and calcium-sensitive
Curr Opin Neurobiol 6:372–378 dyes. Philos Trans R Soc Lond B Biol Sci
24. Mainen ZF, Malinow R, Svoboda K (1999) 364:2453–2467
Synaptic calcium transients in single spines 37. Lidow MS, Goldman-Rakic PS, Gallager DW,
indicate that NMDA receptors are not satu- Rakic P (1991) Distribution of dopaminergic
rated. Nature 399:151–155 receptors in the primate cerebral cortex: quan-
25. Higley MJ, Sabatini BL (2010) Competitive titative autoradiographic analysis using [3H]
regulation of synaptic Ca2+ influx by D2 dop- raclopride, [3H]spiperone and [3H]
amine and A2A adenosine receptors. Nat SCH23390. Neuroscience 40:657–671
Neurosci 13:958–966 38. Goldman-Rakic PS, Muly EC 3rd, Williams
26. Schiller J, Major G, Koester HJ, Schiller Y GV (2000) D(1) receptors in prefrontal cells
(2000) NMDA spikes in basal dendrites of cor- and circuits. Brain Res Brain Res Rev
tical pyramidal neurons. Nature 404:285–289 31:295–301
27. Milojkovic BA, Zhou WL, Antic SD (2007) 39. Westenbroek RE, Hell JW, Warner C, Dubel
Voltage and calcium transients in basal den- SJ, Snutch TP, Catterall WA (1992) Biochemical
drites of the rat prefrontal cortex. J Physiol properties and subcellular distribution of an
585:447–468 N-type calcium channel alpha 1 subunit.
28. Major G, Polsky A, Denk W, Schiller J, Tank Neuron 9:1099–1115
DW (2008) Spatiotemporally graded NMDA 40. Kisilevsky AE, Mulligan SJ, Altier C, Iftinca
spike/plateau potentials in basal dendrites of MC, Varela D, Tai C, Chen L, Hameed S,
neocortical pyramidal neurons. J Neurophysiol Hamid J, Macvicar BA, Zamponi GW (2008)
99:2584–2601 D1 receptors physically interact with N-type
29. Cox CL, Denk W, Tank DW, Svoboda K (2000) calcium channels to regulate channel distribu-
Action potentials reliably invade axonal arbors tion and dendritic calcium entry. Neuron
of rat neocortical neurons. Proc Natl Acad Sci 58:557–570
U S A 97:9724–9728 41. Gulledge AT, Stuart GJ (2003) Action poten-
30. Koester HJ, Sakmann B (2000) Calcium tial initiation and propagation in layer 5 pyra-
dynamics associated with action potentials in midal neurons of the rat prefrontal cortex:
single nerve terminals of pyramidal cells in layer absence of dopamine modulation. J Neurosci
2/3 of the young rat neocortex. J Physiol 23:11363–11372
3:625–646 42. Schultz W (2002) Getting formal with dop-
31. Wachowiak M, Cohen LB (2001) amine and reward. Neuron 36:241–263
Representation of odorants by receptor neuron 43. Antic SD (2003) Action potentials in basal and
input to the mouse olfactory bulb. Neuron oblique dendrites of rat neocortical pyramidal
32:723–735 neurons. J Physiol 550:35–50
32. Brown JE, Cohen LB, De Weer P, Pinto LH, 44. Milojkovic BA, Wuskell JP, Loew LM, Antic
Ross WN, Salzberg BM (1975) Rapid changes SD (2005) Initiation of sodium spikelets in
in intracellular free calcium concentration. basal dendrites of neocortical pyramidal neu-
Detection by metallochromic indicator dyes in rons. J Membr Biol 208:155–169
squid giant axon. Biophys J 15:1155–1160 45. Foust A, Popovic M, Zecevic D, McCormick
33. Ross WN, Arechiga H, Nicholls JG (1987) DA (2010) Action potentials initiate in the
Optical recording of calcium and voltage tran- axon initial segment and propagate through
sients following impulses in cell bodies and axon collaterals reliably in cerebellar Purkinje
processes of identified leech neurons in culture. neurons. J Neurosci 30:6891–6902
J Neurosci 7:3877–3887 46. Antic S, Major G, Zecevic D (1999) Fast opti-
34. Yuste R, Gutnick MJ, Saar D, Delaney KR, cal recordings of membrane potential changes
Tank DW (1994) Ca2+ accumulations in den- from dendrites of pyramidal neurons.
drites of neocortical pyramidal neurons: an api- J Neurophysiol 82:1615–1621
cal band and evidence for two functional 47. Zhou WL, Yan P, Wuskell JP, Loew LM, Antic
compartments. Neuron 13:23–43 SD (2008) Dynamics of action potential back-
35. Yasuda R, Nimchinsky EA, Scheuss V, propagation in basal dendrites of prefrontal
Pologruto TA, Oertner TG, Sabatini BL, cortical pyramidal neurons. Eur J Neurosci
Svoboda K (2004) Imaging calcium concentra- 27(4):923–936
tion dynamics in small neuronal compartments. 48. Canepari M, Popovic M, Vogt K, Holthoff K,
Sci STKE 2004(219):l5 Konnerth A, Salzberg BM, Grinvald A, Antic
36. Homma R, Baker BJ, Jin L, Garaschuk O, SD, Zecevic D (2010) Imaging submillisecond
Konnerth A, Cohen LB, Zecevic D (2009) membrane potential changes from individual
regions of single axons, dendrites and spines. 56. Milojkovic BA, Radojicic MS, Antic SD (2005)
In: Canepari M, Zecevic D (eds) Membrane A strict correlation between dendritic and
potential imaging in the nervous system: meth- somatic plateau depolarizations in the rat pre-
ods and applications. Springer Science+Business frontal cortex pyramidal neurons. J Neurosci
Media, LLC, New York 25:3940–3951
49. Stuart G, Schiller J, Sakmann B (1997) Action 57. Antic SD, Acker CD, Zhou WL, Moore AR,
potential initiation and propagation in rat neocor- Milojkovic BA (2007) The role of dendrites in
tical pyramidal neurons. J Physiol 505:617–632 the maintenance of the UP state. In: Timofeev
50. Spruston N (2008) Pyramidal neurons: den- I (ed) Mechanisms of spontaneous active states
dritic structure and synaptic integration. Nat in the neocortex. Research Signpost, Kerala,
Rev Neurosci 9:206–221 India, pp 45–72
51. Vetter P, Roth A, Hausser M (2001) 58. Acker CD, Antic SD (2009) Quantitative
Propagation of action potentials in dendrites assessment of the distributions of membrane
depends on dendritic morphology. conductances involved in action potential
J Neurophysiol 85:926–937 backpropagation along basal dendrites.
52. Frick A, Magee J, Johnston D (2004) LTP is J Neurophysiol 101:1524–1541
accompanied by an enhanced local excitability 59. Antic SD, Zhou WL, Moore AR, Short SM,
of pyramidal neuron dendrites. Nat Neurosci Ikonomu KD (2010) The decade of the den-
7:126–135 dritic NMDA spike. J Neurosci Res
53. Tsubokawa H, Ross WN (1997) Muscarinic 88:2991–3001
modulation of spike backpropagation in the 60. Antic SD, Acker CD, Zhou WL, Moore AR
apical dendrites of hippocampal CA1 pyramidal (2008) Dopaminergic modulation of dendritic
neurons. J Neurosci 17:5782–5791 excitability in neocortical pyramidal neurons.
54. Hoffman DA, Johnston D (1999) Cell Science Reviews 5:1742–8130
Neuromodulation of dendritic action poten- 61. Nevian T, Sakmann B (2004) Single spine
tials. J Neurophysiol 81:408–411 Ca2+ signals evoked by coincident EPSPs and
55. Djurisic M, Antic S, Chen WR, Zecevic D backpropagating action potentials in spiny
(2004) Voltage imaging from dendrites of stellate cells of layer 4 in the juvenile rat soma-
mitral cells: EPSP attenuation and spike trigger tosensory barrel cortex. J Neurosci
zones. J Neurosci 24:6703–6714 24:1689–1699
Part III
Genetic Manipulation in Cells and Organisms

Chapter 10
Functional Analysis of Human D1 and D5 Dopaminergic

G Protein-Coupled Receptors: Lessons from Mutagenesis
of a Conserved Serine Residue in the Cytosolic End
of Transmembrane Region 6
Bianca Plouffe and Mario Tiberi
Abstract
In mammals, dopamine G protein-coupled receptors (GPCR) are segregated into two categories: D1-like
(D1R and D5R) and D2-like (D2Rshort, D2Rlong, D3R, and D4R) subtypes. D1R and D5R are primarily
coupled to stimulatory heterotrimeric GTP-binding proteins (Gs/olf) leading to activation of adenylyl
cyclase and production of intracellular cAMP. D1R and D5R share high level of amino acid identity in
transmembrane (TM) regions. Yet these two GPCR subtypes display distinct ligand binding and G protein
coupling properties. In fact, our studies suggest that functional properties reported for constitutively active
mutants of GPCRs (e.g., increased basal activity, higher agonist affinity and intrinsic activity) are also
observed in cells expressing wild type D5R when compared with wild type D1R. Herein, we describe an
experimental method based on mutagenesis and transfection of human embryonic kidney 293 (HEK293)
cells to explore the molecular mechanisms regulating ligand affinity, agonist-independent and dependent
activity of D1R and D5R. We will demonstrate how to mutate one conserved residue in the cytosolic end
of TM6 of D1R (Ser263) and D5R (Ser287) by modifying two or three nucleotides in the cDNA of human
D1-like receptors. Genetically modified D1R and D5R cDNAs are prepared using a polymerase chain reac-
tion method, propagated in E. coli, purified and mutations confirmed by DNA sequencing. Receptor
expression constructs are transfected into HEK293 cells cultured in vitro at 37°C in 5% CO2 environment
and used in radioligand binding and whole cAMP assays. In this study, we will test the effect of S263A/
G/D and S287A/G/D mutations on ligand binding and DA-dependent activation of D1R and D5R.
Key words: Dopamine, GPCR, D1-like receptors, Ligand binding, cAMP, Mutagenesis, Third
intracellular loop, TM6, HEK293 cells
1. Introduction
Transmembrane (TM) signaling is fundamental in the homeostatic

regulation of every major physiological function in eukaryotes
ranging from yeast to human (1, 2). A class of integral membrane
141
142 B. Plouffe and M. Tiberi
proteins known as G protein-coupled receptors (GPCRs), which

harbor seven a-helical TM segments, best illustrates this notion (3).
Indeed, over 800 types of GPCRs expressed throughout the body
in humans are stimulated by a variety of ligands such as photons,
protons, ions, odorants, lipids, biogenic amines, peptides and hor-
mones (3). Based on primary sequence and structural similarities,
GPCRs can be grouped into five families: rhodopsin (family A),
secretin (family B), glutamate (family C), adhesion and Frizzled/
Taste2 (3, 4). Classically, heterotrimeric guanine-nucleotide-bind-
ing proteins (G proteins) composed of a, b, and g subunits and
located at the cytoplasmic side of plasma membrane, serve as molec-
ular switches in GPCR signaling pathways by coupling ligand-
induced receptor stimulation to intracellular responses. This is
essentially accomplished by receptor activation of G proteins
through the catalysis of GTP for GDP exchange on Ga promoting
a conformational change in GTP-bound Ga and Gbg subunits,
which culminates in G protein subunit-mediated regulation of
activity of different downstream effector proteins (3). Meanwhile,
studies also show that GPCR signaling can be mediated in a G
protein-independent manner (5, 6). Importantly, at least 50% of
clinical drugs target GPCRs, hence highlighting their human thera-
peutic relevance (7, 8). Given the physiological and clinical impor-
tance of these integral membrane proteins, it is crucial to know how
the conserved seven-a-helix bundle structure imparts subtype-
specific ligand binding and activation properties to the GPCR
members.
Difference in the extent of constitutive activity naturally dis-
played by homologous GPCR subtypes was first highlighted with
D1-like dopaminergic receptors (D1R and D5R), which belong to
the family A GPCRs (9). In contrast to inhibitory G protein (Gi)-
linked D2-like subtypes (D2Rshort, D2Rlong, D3R, D4R), D1R and
D5R couple to stimulatory G proteins (Gs/olf) leading to adenylyl
cyclase (AC) activation and production of intracellular cAMP (10,
11). Interestingly, D5R naturally has a greater constitutive activity
(i.e., increased ability to produce intracellular cAMP in the absence
of agonists) when compared with D1R at similar receptor levels
(9). Furthermore, pharmacological properties of ligands displayed
at D5R are highly reminiscent of those described for constitutively
active mutant (CAM) GPCRs (e.g., higher agonist affinity and
intrinsic activity) (9, 12). The molecular and structural basis under-
lying D1-like subtype-specific functional properties has been
difficult to address experimentally as TM regions of D1R and D5R
are almost indistinguishable (>80% identity) (13). Indeed, the
higher dopamine affinity of D5R relative to D1R cannot be
explained by TM domains interacting with dopamine as critical
amino acids implicated in catecholamine binding (e.g., TM5 serine
residues) are conserved among dopaminergic and adrenergic recep-
tors. Likewise, motifs found within cytosolic surfaces of TM
10 Site-Directed Mutagenesis of D1 and D5 Dopaminergic Receptors 143
domains (e.g., “ionic lock” between TM3 and TM6) regulating

receptor activation are also conserved among family A GPCRs (3).
This view is further underscored by data obtained from recent
crystallographic studies of inactive states of different family A
GPCRs, which display a highly conserved 7TM structural design
(3). Therefore, functional differences between GPCRs displaying
high degree of TM identity likely implicate subtle variations in the
assembly of conserved TM amino acids. These subtle variations are
potentially mediated by GPCR subtype-specific conformation of
remote residues such as those found in receptor intracellular
regions.
While amino acids located in receptor intracellular regions are
unlikely to be directly involved in the binding to extracellular
ligands, mutagenesis studies suggest that the third intracellular
loop (IL3) and cytoplasmic tail (CT) play a major role in mediating
D1-like subtype-specific constitutive activity, agonist affinity and
intrinsic activity (14–18). Notably, substitution of two variant resi-
dues located in the cytosolic end of TM6 of D1R (F264 and R266)
and D5R (I288 and K290) using site-directed mutagenesis show
that F264 and I288 are major determinants in conferring D1-like
subtype-specific agonist binding, constitutive activity and agonist-
dependent G protein-coupling properties (14). Interestingly, F264
(D1R) and I288 (D5R) are located three amino acids downstream
of the glutamate residue, which participates in the formation of an
ionic lock with the highly conserved E/DRY motif of TM3. These
D1R and D5R residues are not found in other catecholaminergic
receptors and hence play potentially an important role in modulat-
ing the arrangement of conserved amino acids forming the ionic
lock. Herein, we describe an experimental approach to study the
role of a conserved serine residue located adjacent to F264 (S263
in D1R) and I288 (S287 in D5R) (Fig. 1), which at this position
does not exist in other catecholamine receptors. We test whether
this conserved serine residue plays a critical role in the modulation
of D1-like receptor binding and activation properties.
2. Materials
2.1. Molecular Biology 1. Custom DNA oligonucleotides (Sigma Genosys, Burlington,

Reagents ON, Canada) as listed in Tables 1 and 2. Lyophilized oligo-
nucleotides are prepared as stock solutions in sterile Milli-Q
water (resistivity of 18.2 MW cm) at 25 pmol/mL for PCR
primers or 2 pmol/mL for DNA sequencing primer.
2. Expand High Fidelity Taq DNA polymerase (3.5 U/mL), PCR
buffer (10×), and MgCl2 (25 mM) (Roche Diagnostics, Laval,
QC, Canada). Store at −20°C.
Fig. 1. Schematic representation of wild type hD1R and hD5R. Putative secondary structure of wild type hD1R and hD5R
are indicated by circles. Distinct residues between hD1R and hD5R are depicted using black circles. Amino acid sequence
of the region encompassing TM5, IL3, and TM6 is also shown. The mutated serine in IL3 of hD1R and hD5R is also indi-
cated. hD1R human D1 receptor, hD5R human D5 receptor.
3. PCR Nucleotides (dATP, dCTP, dGTP, and dTTP) at 100 mM

(Fermentas, Burlington, ON, Canada). Store at −20°C.
4. dNTP Mix: Add 10 mL dATP, 10 mL dCTP, 10 mL dGTP, and
10 mL dTTP to 60 mL sterile Milli-Q water. Store at −20°C.
5. BoxI (PshAI), BsmI (Mva1269I), DraI, Eag I (Eco52I),
EcoRI, HindIII, and XbaI restriction enzymes at 10 U/mL
with 10× Buffers (Y+/TangoTM or yellow; B+ or blue; R+ or red)
from Fermentas. Store at −20°C.
6. Calf Intestinal Alkaline Phosphatase (CIAP) at 1 U/mL and
10× dephosphorylation buffer from Fermentas. Store at
−20°C.
7. T4 DNA ligase (5 U/mL) and T4 DNA ligase buffer (10×)
from Fermentas. Store at −20°C.
8. Tris–acetate–EDTA (TAE) buffer (50×): 2 M Tris–HCl, pH
8.0, 5.71% (v/v) glacial acetic acid, 0.5 M ethylenediamine
tetraacetic acid (EDTA), pH 8.0 in Milli-Q water. Store at
room temperature.
Table 1
Sequences of oligonucleotide primers for the construction of single-point
mutations of hD1R using a PCR-based overlapping approach
Construct Primer sequence 5¢→3¢
hD1R-S263A P1: TTcATcccAgTgcAgcTc

P2: TTcTcTTTTgAATgCcATcTTAAAAgAAcTTTccgg
P3: TTTAAgATgGcATTcAAAAgAgAAAcTAAAgTccTg
P4: TTAggAcAAggcTggTgg
P5: AcTgTgAcTccAgcc
P6: ggccAggAgAggcA
hD1R-S263D P1: TTcATcccAgTgcAgcTc
P2: TTcTcTTTTAAAgTCcATcTTAAAAgAAcTTTccgg
P3: TTTAAgATgGAcTTTAAAAgAgAAAcTAAAgTccTg
P5: AcTgTgAcTccAgcc
P6: ggccAggAgAggcA
hD1R-S263G P1: TTcATcccAgTgcAgcTc
P2: TTcTcTTTTAAAgCCcATcTTAAAAgAAcTTTccgg
P3: TTTAAgATgGGcTTTAAAAgAgAAAcTAAAgTccTg
P5: AcTgTgAcTccAgcc
P6: ggccAggAgAggcA
For each single-point mutant, a set of six primers numbered P1–P6 is used in PCR reactions. Nucleotide
changes leading to S263A, S263D and S263G mutations are indicated in bold. Nucleotide changes intro-
ducing diagnostic restriction sites through silent mutations are underlined or through the S263A mutation
are underlined and bold
9. Agarose gels (1% w/v): Weigh out 0.5 g of agarose (Sigma-

Aldrich, Oakville, ON, Canada) into a plastic 250 mL capped
conical flask and add 50 mL of 1× TAE. Microwave for ~60 s
with loosened cap. Gently swirl and microwave it again for 10 s
and repeat 2–3 times if agarose is not completely dissolved.
Keep watching agarose solution during microwaving as it can
easily boil over. As the solution can become very hot, experi-
menter should wear gloves and hold flask at arms length. Once
agarose is completely dissolved let it cool down briefly (~5 min),
add 2.5 mL of ethidium bromide (10 mg/mL) and gently
swirl. Gloves should be worn when handling ethidium bro-
mide solution as it is mutagenic and to some extent toxic.
Slowly pour the agarose solution into a mini plastic casting
tray, insert sample comb, push away any bubbles to the side
using a pipet tip and allowed gel to solidify for 30–60 min at
room temperature. Rinse out the flask (see Note 1).
10. 50% (v/v) Glycerol. Sterilize through 0.22 mm filter and store
at room temperature.
Table 2
Sequences of oligonucleotide primers for the construction
of single-point mutations of hD5R using a PCR-based
overlapping approach
Construct Primer sequence 5¢→3¢
hD5-S287A P1: TAcggTgggAgg

P2: gATggCAgcgcgcAgAcTggTgTcgggcgcgcAggc
P3: gcgcccgAcAccAgTcTgcgcgcTGccATcAAgAAg
P4: TcATgTggATgTAggcAg
P5: AccTggccAAcTggA
P6: TgTTcAccgTcTccA
hD5-S287D P1: TAcggTgggAgg
P2: gATgTCAgcgcgcAgAcTggTgTcgggcgcgcAggc
P3: gcgcccgAcAccAgTcTgcgcgcTGAcATcAAgAAg
P5: AccTggccAAcTggA
P6: TgTTcAccgTcTccA
hD5-S287G P1: TAcggTgggAgg
P2: gATgCCAgcgcgcAgAcTggTgTcgggcgcgcAggC
P3: gcgcccgAcAccAgTcTgcgcgcTGGcATcAAgAAg
P5: AccTggccAAcTggA
P6: TgTTcAccgTcTccA
For each single-point mutant, a set of six primers numbered P1–P6 is used in
PCR reactions. For each single-point mutant, a set of six primers numbered
P1-P6 is used in PCR reactions. Nucleotide changes leading to S287A, S287D
and S287G mutations are indicated in bold. Nucleotide changes introducing
diagnostic restriction sites through silent mutations are underlined
11. Loading Dye (10×): 0.25% (w/v) bromophenol blue, 0.25%

(w/v) xylene cyanole FF (optional), 50% (v/v) glycerol in
Milli-Q water. Store in aliquots at −20°C.
12. DNA Size Markers: Prepare 1× stock with 250 mL of One
Kilobase Plus DNA Ladder (1 mg/mL; Invitrogen, Burlington,
ON, Canada), 250 mL of 10× loading dye, and 2,000 mL
Milli-Q water. Store in aliquots at −20°C.
13. Sterile Luria Broth (LB) Medium: Weigh out 12.5 g of LB
base (Invitrogen) into glass bottle or Fernbach flask (for DNA
maxipreps) and add 500 mL of Milli-Q water. Sterilize liquid
media using autoclave. Prior to autoclaving loosen cap on glass
bottle and tape aluminum foil on top of flask. Always wear heat
protective gloves when handling bottle or flask at the end of
autoclave cycle. Store at room temperature. If store for a long
time period, check that there is no microorganism contamina-
tion prior to use.
14. Ampicillin (1,000×): 1 g of ampicillin (Sigma-Aldrich) is

dissolved in Milli-Q water at 100 mg/mL and sterilized using
a syringe filter (0.22 mm). Store in 0.5 mL aliquots at −20°C.
15. LB Ampicillin Plates: Weigh out 25 g of LB base (Invitrogen),
15 g of Bacto Agar (Fisher Scientific, Ottawa, ON, Canada)
into glass flask (for DNA maxipreps) and add 1 L of Milli-Q
water. Sterilize using autoclave. Always wear heat protective
gloves when handling flask at the end of autoclave cycle. Let it
cool down for 15 min at room temperature. Thaw two 0.5 mL
aliquots of ampicillin (1,000×), add to molten LB-agar, swirl,
and pour into sterile polystyrene petri dishes (100 × 15 mm,
Fisher Scientific). Quickly pass over the petri dishes a flame
from a Bunsen burner to remove bubbles. Keep moving flame
while removing bubbles to avoid overheating LB-agar and
break up ampicillin. 1 L of LB-agar will make 50 plates. Store
inverted in plastic sleeves at 4°C for no longer than 3 months.
Prior to using plates verify that there is no microorganism
contamination.
16. SOB Medium: 2% (w/v) bacto-tryptone (Fisher Scientific),
0.5% (w/v) bacto-yeast extract (Fisher Scientific), and 10 mM
NaCl. Sterilize by autoclaving. Store at room temperature.
Prior to using SOB medium verify that there is no microorgan-
ism contamination.
17. SOC Medium: SOB medium containing 10 mM MgCl2,
10 mM MgSO4, and 20 mM glucose. Solutions of 1 M MgCl2,
1 M MgSO4, and 2 M glucose are separately made and sterilize
by filtration using 0.22 mm filter. Store at room temperature.
18. XL1-Blue Electroporation-Competent Cells (Agilent
Technologies, Mississauga, ON, Canada). Store at −80°C.
19. Sterile Polypropylene Capped 13 mL Tubes (100 × 16 mm;
Sarstedt, St-Léonard, QC, Canada).
20. Isobutanol (Fisher Scientific).
21. Qiaex II Gel Extraction Kit (Buffer QX1, Buffer PE, Qiaex II
Bead Suspension) (Qiagen, Mississauga, ON, Canada).
22. QIAprep Spin Miniprep and Plasmid Maxiprep Kits from
Qiagen.
2.2. Cell Culture 1. Adenovirus type 5-transformed human embryonic kidney 293
(HEK293) cells (CRL-1573, American Tissue Type Culture
Collection, Manassas, VA).
2. Minimal Essential Medium (MEM) with Earle’s salt
(Invitrogen).
3. Fetal Bovine Serum (FBS) (Invitrogen). Thaw frozen FBS
bottle at 4°C overnight. The next day, warm up bottle in a
37°C water bath, heat-inactivate FBS in a 55°C water bath for
1 h, and sterilize through 0.22 mm filter. Store in 25 and 50 mL

aliquots in sterile capped polypropylene conical tubes at
−20°C.
4. Gentamicin Sulfate (10 mg/mL) (Invitrogen).
5. Trypsin (0.25%) and EDTA (0.05% (w/v)) Buffer Solution
(Invitrogen).
6. Ca2+ and Mg2+-free Phosphate-Buffered Saline (PBS) (Wisent,
St-Bruno, QC, Canada).
7. Tissue Culture Grade Sigma Hybri-Max Dimethylsulfoxide
(DMSO) (Sigma-Aldrich). Store at room temperature.
8. BD Falcon Polystyrene 75 cm2 Flasks with 0.2 mm Vented Blue
Plug Seal Cap (VWR International, Montréal, QC, Canada).
9. Sterile HEPES (4-(2-hydroxyethyl)-1-piperazineethanesulfo-
nic acid) Buffer Solution (1 M, pH 7.4) (Invitrogen).
10. 20 mM HEPES-buffered MEM: In a certified biological safety
cabinet (BSC), add to bottle of MEM (500 mL), 10 mL of
sterile HEPES (pH 7.4) and 0.5 mL of gentamicin (Sigma-
Aldrich). Store at 4°C for up to 2 months.
11. 1 M HEPES (pH 7.0): 37.75 g of HEPES (Fisher Scientific) is
dissolved in 100 mL Milli-Q water, adjust pH to 7.0 and com-
plete to 150 mL with Milli-Q water. Sterilize through 0.22 mm
filter. Store in tissue culture area at room temperature (see
Note 2).
12. 2 M NaCl: Prepare in glass bottle containing 500 mL of
Milli-Q water and sterilize by autoclaving. Store in tissue
culture area at room temperature.
13. 1 M Na2HPO4: Add 35.5 g of Na2HPO4 in a beaker contain-
ing 230 mL of Milli-Q water, stir on a hot plate at low
temperature (setting 2) until completely dissolved, complete
to a final volume of 250 mL, and sterilize through 0.22 mm
filter. Store in tissue culture area at room temperature.
14. 1 M NaH2PO4: Add 34.5 g of NaH2PO4 in a beaker contain-
ing 230 mL of Milli-Q water, stir on a hot plate at low tem-
perature (setting 2) until completely dissolved, complete to a
final volume of 250 mL, and sterilize through 0.22 mm filter.
Store in tissue culture area at room temperature.
15. 1 M Na3PO4: Prepare 25 mL by adding equal volumes of ster-
ile 1 M Na2HPO4 and 1 M NaH2PO4 solutions in a BSC. Store
in tissue culture area at room temperature (see Note 3).
16. Sterile 2× HEPES-buffered saline (0.28 M NaCl, 0.05 M
HEPES, and 1.5 mM Na3PO4, pH 7.1): Add 70 mL of 2 M
NaCl, 25 mL of 1 M HEPES, and 750 mL of 1 M Na3PO4 to
400 mL of Milli-Q water in a glass beaker. Stir and adjust to
pH 7.1 (±0.05). Complete to a final volume of 500 mL and
sterilize through 0.22 mm filter in a BSC. Store in 15 mL ali-

quots at −20°C.
17. Teflon cell lifters (Fisher Scientific).
2.3. Radioligand 1. [3H]-SCH23390 and [3H]-adenine (PerkinElmer NEN,

Binding and Whole Boston, MA). Store at −20°C (see Note 4).
Cell cAMP Assays 2. Ascorbic acid, (+)-SCH23390 hydrochloride, dopamine
hydrochloride, cis-flupenthixol dihydrochloride and (+)-buta-
clamol hydrochloride (Sigma-Aldrich). Store at room temper-
ature (see Note 5).
3. Lysis Buffer: 10 mM Tris–HCl, pH 7.4 and 5 mM EDTA, pH
8.0. Store at 4°C.
4. Resuspension Buffer: 62.5 mM Tris–HCl, pH 7.4 and 1.25 mM
EDTA, pH 8.0. Store at 4°C.
5. Binding Buffer: 62.5 mM Tris–HCl, pH 7.4 and 1.25 mM
EDTA, pH 8.0, 200 mM NaCl, 6.7 mM MgCl2, 2.5 mM
CaCl2 and 8.33 mM KCl. Store at 4°C.
6. Washing Buffer (10×): 500 mM Tris–HCl, pH 7.4 and 1 M
NaCl. Store at 4°C.
7. Whatman GF-C Glassfiber Filter Sheets (Brandel Inc.,
Gaithersburg, MD).
8. Plastic Scintillation Vials (20 mL) (Sarstedt, Newton, NC).
9. Bio-Safe II Biodegradable Scintillation Cocktail (Research
Products International Corp., Mount Prospect, IL).
10. Bio-Rad Protein Assay Dye Concentrate (Bio-Rad Laboratories
Inc., Mississauga, ON, Canada). Store at 4°C.
11. Bovine Serum Albumin (BSA), minimum 96% electrophoresis
(Sigma-Aldrich). BSA is dissolved in sterile Milli-Q water at
1 mg/mL. Store in 1 mL aliquots at −20°C.
12. 3-isobutyl-1-methylxanthine (IBMX) (Sigma-Aldrich) stored
−20°C is dissolved in DMSO at 200 mM. Store stock solution
at 4°C (see Note 6).
13. Neutralizing Solution: 4.2 M KOH. Store in a glass bottle at
room temperature.
14. Cyclic AMP (cAMP) (Sigma-Aldrich). Store desiccated at
−20°C.
15. [14C]-cAMP (Moravek Biochemicals, Brea, CA). Store at
−20°C.
16. cAMP Stop Solution: 2.5% (v/v) perchloric acid, 0.1 mM
cAMP and [14C]-cAMP (~3.3 nCi/mL; ~10,000 dpm) in
1,500 mL Milli-Q water. Store at 4°C (see Note 7).
17. Dowex AG 50 W-4X Resin (hydrogen form, 200–400 dry
mesh, 63–150 mm wet beads) from Bio-Rad Laboratories
(Hercules, CA).
18. Alumina N Super I (MP Biomedicals, Montréal, QC, Canada).

Store at room temperature.
19. Poly-Prep Columns (with 10 mL reservoir and graduated
volume markings) from Bio-Rad Laboratories Inc.
20. Hydrochloric Acid (HCl) and Sodium Hydroxide (NaOH) at
0.1 N.
21. Imidazole (Sigma-Aldrich) is dissolved in Milli-Q water at
2 M, pH 7.5 (see Note 8). Store in plastic bottle at room tem-
perature for up to 4 months.
3. Methods
The following experimental strategy describes how to generate

three single-point mutants of S263 of hD1R (S263A, S263D and
S263G) and S287 of hD5R (S287A, S287D, S287G) using a PCR-
based overlap extension approach, DNA ligation, and DNA auto-
mated sequencing (Fig. 2). The experimental approach entails two
major steps:
1. DNA primers derived from forward and reverse nucleotide
sequences coding for human D1R (hD1R) and D5R (hD5R)
are designed to mutate the candidate serine into alanine, aspar-
tate, and glycine. The hD1R and hD5R DNA templates are
separately mixed with mutagenesis primers and a high proof-
reading DNA polymerase to produce “megaprimers” using
polymerase chain reaction (PCR) and an overlap extension
approach. Pairs of purified overlapping “megaprimers” are
subsequently fused together by PCR and purified by silica-gel
particles to obtain the different mutated receptor DNA cas-
settes. Mutated receptor DNA cassettes and corresponding
wild type receptor DNA in the pCMV5 expression plasmid
(cloning vector) are subjected to restriction digest and ligation
procedures to generate different mammalian mutant receptor
expression constructs.
2. Human embryonic kidney 293 (HEK293) cells are transfected
with wild type and mutant receptor expression constructs using
a DNA-calcium phosphate precipitation procedure, and seeded
in tissue culture dishes and plates for radioligand binding and
whole cell cAMP studies. Ligand binding properties of wild
type and mutant receptors are measured in membrane
preparations of transfected cells with saturation and competi-
tion experiments using the D1-like selective radioligand
[3H]-SCH23390. Agonist-independent and dependent cou-
pling to Gs is assessed in transfected cells metabolically labeled
Fig. 2. Schematic representation of the making of pCMV5 expression constructs for single-point mutant forms of hD1R and
hD5R. The key steps involved in the preparation of hD1R (a) and hD5R (b) single-point mutant constructs in the pCMV5
expression vector are shown (see text for details).
with [3H]-adenine and the amount of intracellular cAMP

production determined from cell lysates with a sequential chro-
matography purification procedure using Dowex and alumina
columns. Radioligand binding and whole cell cAMP data are
analyzed using nonlinear curve fitting program and statistical
tests. Our results suggest that S263 in hD1R and S287 in
hD5R play a differential role in controlling ligand affinity and
DA-induced activation of AC of human D1-like receptors (see
data figures).
3.1. Preparation of 1. Qiagen maxipreps of wild type hD1R and hD5R subcloned in
Single-Point Mutant the expression vector pCMV5 are used as DNA templates
hD1R and hD5R (25 ng/mL) in series of two PCR rounds as depicted in Fig. 3.
Cassettes by Site- For each single-point mutant two PCR products (“megaprim-
Directed Mutagenesis ers”), A and B, are separately amplified in the first round
and PCR using specific set of P1–P2 and P3–P4 primer pairs (see Tables 1
and 2). First-round PCRs are carried out in a final volume of
50 mL containing 50 ng of DNA template (2 mL of stock),
50 pmol of forward primer (2 mL of P1 or P3 stock solution at
25 pmol/mL), 50 pmol of reverse primer (2 mL of P2 or P4
stock solution at 25 pmol/mL), 1.5 mM MgCl2 (3 mL of
25 mM stock solution), 0.2 mM dNTPs (1 mL of 10 mM stock
solution), 3.5 U of Taq DNA polymerase (1 mL of stock
Fig. 3. General scheme for creating single-point mutations using PCR-based overlapping approach. Representative exam-
ple of the experimental strategy used to generate the S263G and S287G mutations in hD1R-pCMV5 (a) and hD5R-pCMV5
(b) constructs is depicted. This strategy remains identical for creating other single point mutations in hD1R (S263A and
S263D) and hD5R (S287A and S287D). The beginning of the polylinker region of pCMV5 (EcoRI site) is arbitrarily set to
position 0. Nucleotide position of 5¢ region of PCR primers (P1-P6) annealing to DNA expression construct template is
indicated. Restriction sites used for generating mutated cassette are shown. The boundaries of mutated cassettes are
illustrated using brackets.
solution), 5 mL of 10× PCR buffer and of 34 mL of sterile

Milli-Q water. DNA is amplified in an Eppendorf Thermal
Mastercycler using the following conditions: 1 cycle (94°C for
3 min, 50°C for 1 min, 72°C for 3 min), 24 cycles (94°C for
45 s, 50°C for 1 min, 72°C for 1 min) completed by an anneal
extension step (50°C for 1 min and 72°C for 8 min).
Importantly, during this first PCR round, an overlapping
region between A and B will be generated, which will allow
amplifying the final PCR product using primers P5–P6 during
the second PCR round (see Subheading 3.1.5). At the end of
the final cycle, add 5.5 mL of 10× loading dye to PCR tubes.
2. Prepare 1× TAE buffer in Milli-Q water and pour a 1% (w/v)
agarose minigel (see Subheading 2.1, item 8). Once agarose
solidifies, put casting tray in electrophoresis apparatus, fill with
1× TAE buffer, gently remove sample comb, and load wells
with PCR samples (55 mL) and one well with 4 mL of DNA
size markers. Separate samples for 1 h at 80 V and visualize
ethidium bromide-stained DNA bands using an UV
Transilluminator equipped with a digital camera.
3. Cut off appropriate sized DNA bands (see Fig. 3) and purify
agarose-embedded DNA with QIAEX beads (Qiagen) accord-
ing to manufacturer’s protocol. Elute purified bands from
QIAEX beads using 20 mL of sterile Milli-Q water. Add 2 mL
of purified bands to 7 mL of sterile Milli-Q water, mix with
1 mL of 10× loading dye and run samples beside a well loaded
with 4 mL of DNA size markers on 1% (w/v) agarose minigel
for 1 h at 80 V.
4. Visualize ethidium bromide-stained DNA bands using an UV
Transilluminator equipped with a digital camera and take pic-
ture to assist in the semi-quantification of purified products.
5. For the second-round PCRs, add equal volume of purified
“megaprimer” A and B (~0.5 mL each) corresponding to the
designated mutant receptor to a mix containing 19.5 mL of
sterile Milli-Q water, 2.5 mL of 10× PCR buffer, 1.5 mM
MgCl2 (1.5 mL of 25 mM stock solution), 0.2 mM dNTPs
(0.5 mL of 10 mM stock solution), and 1.75 U of Taq DNA
polymerase (0.5 mL of stock solution). The overlap PCR is
done in a final volume of 25 mL using 1 cycle at 94°C for
3 min, 50°C for 1 min, 72°C for 10 min, and 20°C for
8 min.
6. During the 8-min period, 25 mL of a mix containing 19.5 mL
of sterile Milli-Q water, 2.5 mL of 10× PCR buffer, 25 pmol of
P5 forward primer (1 mL of 25 pmol/mL stock solution), 25
pmol of P6 reverse primer (1 mL of 25 pmol/mL stock solu-
tion), 1.5 mM MgCl2 (1.5 mL of 25 mM stock solution),
0.2 mM dNTPs (0.5 mL of 10 mM stock solution), and 1.75
U of Taq DNA polymerase (0.5 mL of stock solution) is added

to overlap PCR mixture (final volume of 50 mL). PCR is then
run for 25 cycles (94°C for 45 s, 50°C for 1 min, 72°C for
1 min) and completed by an anneal extension step (50°C for
1 min and 72°C for 8 min). At the end of this step, add 5.5 mL
of 10× loading dye to PCR tubes.
7. Run final PCR products on 1% (w/v) agarose and appropriate
mutant receptor DNA cassettes purified as described (see
Subheadings 3.1.2, 3.1.3, and 3.1.4).
3.2. Preparation of 1. Set up restriction enzyme digestions of the wild type hD1R-
Linearized Wild Type pCMV5 expression construct and purified mutated D1R DNA
Receptor Expression cassettes (849 bp) with HindIII and XbaI (see Fig. 3a), and
Constructs and wild type hD5R-pCMV5 expression construct and purified
Mutated Receptor DNA mutated D5R DNA cassettes (569 bp) with BsmI and EagI
Cassettes by (see Fig. 3b) in a final volume of 30 mL using separate auto-
Digestions with claved 1.5 mL Eppendorf tubes. Prepare reaction tubes for
Restriction Enzymes wild type hD1R and hD5R-pCMV5 expression constructs, in
which restriction enzymes are replaced with an equivalent
amount of sterile Milli-Q water. These tubes are referred to as
uncut DNA (see Note 9).
2. For hD1R DNA digestions, add to tubes 8 mL of wild type
hD1R-pCMV5 (0.125 mg/mL working solution; 1 mg total) or
purified mutant hD1R cassette (see Note 10), 16 mL of sterile
Milli-Q water, 1.5 mL of HindIII (15 U), 1.5 mL of XbaI (15 U),
and 3 mL of 10× Y+/TangoTM buffer (1× final) (see Note 10).
3. For hD5R DNA digestions, add to tubes 8 mL of wild type
hD5R-pCMV5 (0.125 mg/mL working solution; 1 mg total) or
purified mutant hD5R cassette, 13 mL of sterile Milli-Q water,
1.5 mL of BsmI (15 U), 1.5 mL of EagI (15 U), and 6 mL of
10× Y+/TangoTM buffer (2× final) (see Note 11).
4. Gently pipette up and down to mix and float tubes in a 37°C
water bath for 1 h. At the end of incubation, put all tubes on
ice (optional), add 3 mL of 10× loading dye to only the digested
PCR products and mix by gently pipetting up and down.
5. Leave tubes containing digested hD1R-pCMV5 and hD5R-
pCMV5 DNA constructs without loading dye and prepare
linearized wild type hD1R and hD5R-pCMV5 DNA samples
for dephosphorylation (see Note 12).
3.3. Dephosphorylation 1. Carry out dephosphorylation reaction in using a final volume

of Linearized Wild of 50 mL. Add 14 mL of sterile Milli-Q water to tubes contain-
Type hD1R and ing the 30 mL of restriction digestion mix of hD1R-pCMV5
hD5R-pCMV5 DNA (HindIII-XbaI) and hD5R-pCMV5 (BsmI-EagI) expression
Constructs constructs. Then, add 5 mL of 10× dephosphorylation buffer
and 1 mL of CIAP (5 U).

water bath for 30 min.
3. At end of incubation, place dephosphorylation tubes on ice
(optional), put in 5 mL of 10× loading dye and mix by gentle
pipetting up and down.
3.4. Isolation of 1. Digested mutant cassettes and wild type expression constructs
Linearized Wild Type are loaded on individual wells along with a well containing
hD1R and hD5R- 4 mL of DNA size markers of agarose minigels as described
pCMV5 DNA above in Subheading 3.1 (see Subheading 3.1.2). Run hD1R
Constructs and and hD5R samples for 1 h at 80 V on 1% (w/v) and 1.8%
Digested Mutant (w/v) agarose minigels, respectively (see Note 13). Visualize
Receptor DNA ethidium bromide-stained agarose gel using an UV
Cassettes Transilluminator equipped using a digital camera and excise
appropriate bands: (1) HindIII-XbaI linearized hD1R-pCMV5
(~5,400 bp), (2) HindIII-XbaI digested mutant hD1RS263G
cassette (745 bp), (3) HindIII-XbaI digested mutant
hD1RS263A cassette (745 bp), (4) HindIII-XbaI digested
mutant hD1RS263D cassette (745 bp), (5) BsmI-EagI linear-
ized hD5R-pCMV5 (~6,000 bp), (6) BsmI-EagI digested
mutant hD5RS287G cassette (348 bp), (7) BsmI-EagI digested
mutant hD5RS287A cassette (348 bp), and (8) BsmI-EagI
digested mutant hD5RS287D cassette (348 bp).
2. Purify agarose-embedded DNA bands with QIAEX beads
(Qiagen) according to manufacturer’s protocol. Elute purified
bands from QIAEX beads using 50 mL of sterile Milli-Q water
or QIAEX elution buffer. Add 2 mL of purified bands to 7 mL
of sterile Milli-Q water, mix with 1 mL of 10× loading dye and
run samples beside a well loaded with 4 mL of DNA size mark-
ers on 1% (w/v) agarose minigel prepared with thin sample
comb for 1 h at 80 V.
Transilluminator equipped with a digital camera and take pic-
ture to assist in the semi-quantification of purified DNAs to set
up ligations.
3.5. DNA Ligation 1. Thereafter and unless stated otherwise, linearized hD1R-
Reactions pCMV5 (~5,400 bp band) and hD5R-pCMV5 (~6,000 bp
band) will be called “vector” whereas the mutated receptor
cassettes will be referred to as “inserts.” Set up control (vector
alone) and test (vector + insert) ligation reactions in a final vol-
ume of 10 mL in 1.5 mL Eppendorf tubes on ice (optional).
Add to test ligation tubes 0.5 mL vector, 0.5 mL insert (replace
with 0.5 mL sterile Milli-Q water in control ligation tubes),
0.5 mL T4 DNA ligase (2.5 U), 1 mL 10× T4 DNA ligase
buffer, and 7.5 mL sterile Milli-Q water (see Note 14).
2. Float tubes in a 16°C water bath overnight (see Note 15).

Store ligation tubes at −20°C until use for transformation of
XL-1 Blue electroporation-competent cells.
3.6. Desalting of DNA 1. Add 40 mL of sterile Milli-Q water to fresh or thawed ligation
Ligation Samples tubes (10 mL) at room temperature and gently tap tubes to
mix.
2. In a fume hood, add 500 mL of isobutanol to 50 mL ligation
samples.
3. Mix by gently inverting tubes several times until isobutanol is
fully miscible with aqueous ligation samples (no detection of
isobutanol phase remnant or bubbles).
4. Spin tubes in a microfuge at 16,000 × g for 10 min at room
temperature.
5. In a fume hood, decant supernatant in waste glass bottle and
spin again tubes at 16,000 × g for 30 s.
6. In a fume hood, carefully remove supernatant using a P200
pipette and discard supernatant in waste glass bottle.
7. Let air dry the small DNA pellet in fume hood for 5–10 min,
add 10 mL of sterile Milli-Q water to tubes and carefully resus-
pend DNA pellet by washing sides of tubes.
3.7. Transformation 1. Take 5 mL of desalted DNA samples from control and test
of XL1-Blue ligation reactions and separately mix with 40 mL of XL1-Blue
Electroporation- electroporation-competent cells on ice by gently pipetting up
Competent Cells with and down in 1.5 mL Eppendorf tubes.
Ligated DNA Samples 2. Transfer 45 mL of DNA-bacteria mixtures into ice-cold elec-
troporation cuvettes. Carefully wipe side of electroporation
cuvettes to remove any condensation prior to inserting into
electroporator.
3. Shock cells at 1,800 V for 5 ms (see Note 16).
4. Add 1 mL of freshly made SOC in each cuvette, gently pipette
up and down and transfer 1 mL to sterile polypropylene capped
13 mL tubes (100 × 16 mm).
5. Incubate with loosened cap in a 37°C shaking incubator at a
velocity of 300 rpm for 1 h.
6. Pour bacterial cultures into 1.5 mL Eppendorf tubes and spin
at 6,000 × g for 30 s at room temperature. Discard ~900 mL
supernatant and gently resuspend bacterial pellet with leftover
supernatant (~100 mL) by pipetting up and down.
7. Spread ~100 mL of bacterial cultures on pre-warmed (37°C)
LB-ampicillin plates and grow bacteria in a 37°C incubator
overnight (see Note 17).
3.8. Preparation 1. Prepare sterile polypropylene capped 13 mL tubes containing

of Plasmid DNA 5 mL of LB with 1× ampicilin (100 mg/mL).
Minipreps and 2. Pick two isolated colonies from each test ligation plates using
Identification of small sterile pipette tips and inoculate a LB-ampicillin tube
Single-Point Mutants with a single colony by ejecting tip in it. Incubate tubes with
by Restriction loosened caps in a 37°C shaking incubator at a velocity of
Digestions 300 rpm overnight.
3. Prepare backup bacterial glycerol stocks with overnight
cultures. Add 0.7 mL of bacterial cultures to 0.3 mL of sterile
50% (v/v) glycerol (final concentration: 15% (v/v)) in 1.5 mL
Eppendorf tubes and gently mix by inverting tubes several
times. Snap-freeze in liquid nitrogen. Store at −80°C (see
Note 18).
4. Make miniprep DNA with the rest of bacterial cultures
(~4.3 mL) using QIAprep Spin Columns according to manu-
facturer’s protocol (see Note 19).
5. Measure DNA concentration and purity of plasmid miniprep
(total volume of 50 mL). Add 5 mL of plasmid miniprep DNAs
in a final volume of 1 mL of sterile Milli-Q water in quartz UV
cuvettes. Read optical density (OD) against a blank solution
(1 mL of sterile Milli-Q water) at wavelengths of 260 nm and
280 nm using spectrometer (see Note 20).
6. Set up diagnostic digestion reactions in a final volume of 15 mL
with and without appropriate restriction enzymes as follows
(see Table 3).
7. For the screening of positive hD1R-S263A plasmid DNAs,
carry out restriction digestions in 1.5 mL Eppendorf tubes
containing 2 mL of plasmid DNA (0.5 mg; stock solution of
0.25 mg/mL), 1.5 mL of 10× red buffer, 0.5 mL of BsmI (5U),
Table 3
Expected band size pattern of wild type and single-point mutants of hD1R
and hD5R following digestion with restriction enzymes
Constructs Restriction enzymes Band sizes (base pairs)
hD1R-S263A BsmI + EcoRI Wild Type hD1R: 6025

Mutant hD1R: 804, 5221
hD1R-S263G DraI Wild Type hD1R: 19, 692, 1666, 3648
hD1R-S263D Mutant hD1R: 19, 692, 1112, 1666, 2536
hD5R-S287A BoxI Wild Type hD5R: 6322
hD5R-S287G Mutant hD5R: 612, 5710
hD5R-S287D
For each single-point mutant, the diagnostic band sizes of digested positive plasmid DNAs are bold and
underlined
0.5 mL of EcoRI (5U), and 10.5 mL of sterile Milli-Q water

(11.5 mL for uncut condition).
8. For the screening of positive hD1R-S263G and hD1R-S263D
plasmid DNAs, carry out restriction digestions in 1.5 mL
Eppendorf tubes containing 2 mL of plasmid DNA (0.5 mg;
stock solution of 0.25 mg/mL), 1.5 mL of 10× blue buffer,
0.5 mL of DraI (5U), and 11 mL of sterile Milli-Q water
(11.5 mL for uncut condition).
9. For the screening of positive hD5R-S287A, hD5R-S287G and
hD5R-S287D plasmid DNAs, carry out restriction digestions
in 1.5 mL Eppendorf tubes containing 2 mL of plasmid DNA
(0.5 mg; stock solution of 0.25 mg/mL), 1.5 mL of 10× Y+/
TangoTM buffer, 0.5 mL of BoxI (5U), and 11 mL of sterile
Milli-Q water (11.5 mL for uncut condition).
water bath for 1 h. At the end of incubation, add 1.5 mL of 10×
loading dye, mix by gently pipetting up and down and run
samples beside a well loaded with 4 mL of DNA size markers
on 1% (w/v) agarose minigel for 1 h at 80 V.
Transilluminator equipped with a digital camera, take picture
and identify positive mutant plasmid DNAs according to
expected band sizes of digested wild type and mutated DNA
(see Table 3).
3.9. Preparation 1. Prepare samples for automated DNA sequencing as follows

of Samples for (see Note 21).
Automated DNA 2. Make working solution of plasmid DNA minipreps at a final
Sequencing and concentration of 12.5 ng/mL with sterile Milli-Q water in
Plasmid DNA autoclaved 1.5 mL Eppendorf tubes.
Maxipreps 3. For each single-point mutants of hD1R, set up autoclaved
1.5 mL Eppendorf tubes containing 10 mL (125 ng) of plas-
mid DNA and 5 mL (10 pmol) of hD1R-P1 forward primer or
hD1R-P4 reverse primer (see Table 1) (see Note 22).
4. For each single-point mutants of hD5R, set up autoclaved
1.5 mL Eppendorf tubes containing 10 mL (125 ng) of plasmid
DNA and 5 mL (10 pmol) of hD5R-P6 reverse primer (see
Table 2).
5. Align sequenced DNAs with predicted nucleotide sequences
and confirm that (1) serine is mutated, (2) mutated receptor
DNA cassette is ligated in-frame with the expression vector
pcCMV5 containing wild type hD1R or hD5R DNA sequences,
and (3) integrity of restriction sites used for subcloning the
mutated hD1R (HindIII and XbaI) and hD5R (BsmI and
EagI) cassettes.
6. Once positive clone identify and DNA integrity confirmed,

make large-scale plasmid DNA preparations as follows.
7. Stab glycerol stocks of positive bacterial clones with a sterile
straight wire and streak pre-warmed (37°C) LB-ampicillin
plates. Grow bacteria in a 37°C incubator overnight.
8. The next day, prepare sterile polypropylene capped 13 mL
tubes containing 1 mL of LB with 1× ampicilin (100 mg/mL)
and inoculate each tube with a single colony using sterile small
pipette tips as described above (see Subheading 3.8, step 2).
Incubate tubes with loosened caps in a 37°C shaking incubator
at a velocity of 300 rpm for 6 h.
9. Subsequently, add 10 mL of the small bacterial cultures to 1 mL
of sterile LB in 1.5 mL Eppendorf tubes (see Note 23). Open
aluminum foil on top of already made sterile Fernbach flasks
(see Subheading 2.1, item 13) containing 500 mL of LB and
1× ampicillin with 1 mL of the diluted small bacterial cultures.
Incubate flasks in a 37°C shaking incubator at a velocity of
300 rpm overnight.
10. Prepare large-scale plasmid DNAs with Maxiprep Column Kit
according to Qiagen’s protocol. Measure plasmid DNA
concentration and purity (see Subheading 3.8, step 5).
11. Verify integrity of plasmid receptor DNAs using digestions
with restriction enzymes (see Subheading 3.8, steps 6–11).
3.10. Preparation 1. Make frozen stocks of HEK293 cells in Nalgene cryovials at a

and Transfection density of 5 × 106 cells per mL of sterile freezing medium (10%
of HEK293 Cells tissue culture grade DMSO, 20% FBS, 70% MEM with Earle’s
salts and 40 mg/mL gentamicin) (see Note 24).
2. Grow working stocks of HEK293 cells in polystyrene 75 cm2
flasks containing 20 mL of complete MEM (10% (v/v) FBS
and 40 mg/mL gentamicin) at 37°C in a humidified 5% CO2
incubator and maintain stocks as described previously (19).
3. Prepare HEK293 cells for transfection as follows.
4. Aspirate medium from 75 cm2 flasks, add 5 mL of room
temperature PBS and wash cells by gently rocking flasks (see
Note 25).
5. Aspirate PBS, add 1 mL trypsin to flasks, briefly incubate cells
at room temperature (<1 min) (see Note 26).
6. Stop trypsinization by adding 20 mL of complete MEM per
flask (see Note 27).
7. Mix cells by gentle trituration using 10 mL pipette. Pipette up
and down 10–15 times to detach cells and avoid clump forma-
tion. Count cells using hemacytometer.
8. For the transfection procedure, seed polystyrene tissue culture

dishes (100 × 20 mm) with 2 × 106 cells in final volume of
10 mL of complete MEM and grow cells until the next day
(24–30 h) at 37°C in a humidified 5% CO2 incubator (see
Note 28).
9. Prepare transfection solutions in sterile 13 mL polypropylene
tubes (100 × 16 mm) in a BSC as follows.
10. Add a total quantity of 10 mg of different plasmid DNAs to be
transfected in a final volume of 50 mL or less of Milli-Q water
to 13 mL tubes (see Note 29).
11. Add sterile Milli-Q water to each tube containing plasmid
DNA to final volume of 900 mL.
12. Add 100 mL of sterile 2.5 M CaCl2 to each tube (final volume
1 mL) (see Note 30).
13. Add 1 mL of sterile 2× HEPES-buffered saline solution to
DNA-calcium phosphate mixture in a dropwise manner using
a P1000 pipette. Mix by gentle flicking of tubes (final volume
of 2 mL) (see Note 31).
14. Transfect two 100 × 20 mm dishes of HEK293 using 1 mL per
dish as follows.
15. Open the lid of a dish, add dropwise 1 mL of transfection
solution to the whole medium surface, close lid, and incubate
HEK293 cells overnight at 37°C in humidified 5% CO2
incubator (see Note 32).
16. The next day, split cells as follows (see Note 33).
3.11. Seeding of 1. For each transfection condition, aspirate culture medium from
Transfected HEK293 the four dishes in a BSC, add 5 mL of room temperature PBS
Cells for Radioligand per dish to wash cells (see previous Note 25), aspirate PBS, add
Binding and Whole 0.5 mL of 1× trypsin per dish, incubate briefly, add 10 mL of
Cell cAMP Assays complete MEM and triturate cells using gentle pipetting up
and down (see Subheading 3.10, step 7).
2. For radioligand binding studies, pool cells from four
100 × 20 mm dishes into a 150 × 25 mm dish (final volume
~40 mL) and grow transfected HEK293 cells for ~48 h at
37°C in humidified 5% CO2 incubator (see Note 34).
3. For dose–response curves using whole cell cAMP assays, pool
cells from four 100 × 20 mm dishes (final volume ~40 mL) and
seed two 12-well plates with 1 mL of cells per well (total vol-
ume required is 24 mL) for each experimental condition. Seed
also one 100 × 20 mm dish with 10–15 mL of cells per dish for
determination of receptor levels in membrane preparations
from cells used for dose–response curves (see Note 35).
3.12. Preparation 1. On the day of the experiment, put 150 × 25 mm dishes on ice,
of Crude Membranes aspirate culture medium, add 10 mL of cold PBS to side of
from Transfected dishes and wash cells by gentle rocking of dishes.
HEK293 Cells for 2. Aspirate PBS, add 15 mL of ice-cold lysis buffer, detach cells
Saturation Studies with a cell lifter by scraping off the dish surface and transfer
lysates to 50 mL polycarbonate centrifuge tubes (29 × 104 mm,
Beckman Coulter). Wash dishes again with 15 mL of ice-cold
lysis buffer, harvest 5 mL wash and put in centrifuge tubes.
3. Centrifuge samples at 40,000 g for 20 min at 4°C and put
tubes on ice.
4. Discard supernatant, add 3 mL of ice-cold lysis buffer to
centrifuge tubes, pipette up and down to detach pellets and
homogenize pellets in centrifuge tubes with a Brinkmann
Polytron at a velocity of 17,000 rpm for 15 s. Adjust final
volume in tubes to 30 mL and centrifuge at 40,000 g for
20 min at 4°C.
5. Discard supernatant, add 3 mL of ice-cold lysis buffer to cen-
trifuge tubes, pipette up and down to detach pellets and
homogenize pellets in centrifuge tubes with a Brinkmann
Polytron at a velocity of 17,000 rpm for 15 s.
6. Add 0.6 mL of membrane preparations to tubes containing
3 mL of cold resuspension buffer (dilution factor of 1:6) and
leave tubes on ice until used for saturation studies. Add the
remnant of membrane preparations in lysis buffer in two
1.5 mL Eppendorf tubes (~1.2 mL per tube), snap-freeze in
liquid nitrogen and store at −80 C until used for competition
studies.
3.13. Saturation 1. Set up four three-tier polypropylene racks (6 × 12 holes per

Studies row, Fisher Scientific) containing 48 polystyrene test tubes
(12 × 75 mm) to carry binding reactions in a final volume of
500 mL (see Note 36).
2. Prepare six 10× concentrations of [3H]-SCH23390 (84 Ci/
mmol) in Milli-Q water ranging from ~0.1 to 100 nM (final in
assays: ~0.01–10 nM) using a dilution factor of 1:3.
3. Prepare 10× cis-flupenthixol (100 mM) in Milli-Q water using
frozen stock (final in assays: 10 mM) (see Note 5).
4. Add 300 mL of binding buffer to all tubes (final in assays:
50 mM Tris–HCl, pH 7.4, 120 mM NaCl, 4 mM MgCl2,
1.5 mM CaCl2, 5 mM KCl, and 1 mM EDTA, pH 8.0).
5. Add 50 mL of Milli-Q water to total binding tubes or 50 mL of
10× cis-flupenthixol (100 mM) to nonspecific binding tubes.
6. Add 50 mL of different 10× [3H]-SCH23390 concentrations
to a set of four tubes from the lowest to highest concentration
(see Note 36).
7. Add 100 mL of membrane preparations to tubes (24 tubes for

each receptor tested), shake racks to mix and incubate at room
temperature (~20°C) for 90–120 min.
8. Measure protein concentration in leftover membrane prepara-
tions using Bio-Rad assay kit with BSA as standard according
to manufacturer’s protocol (see Note 37).
9. Terminate binding reactions by rapid filtration through precut
Whatmann GF/C glassfiber filter sheets (11.4 cm × 31.1 cm)
using Brandel semi-automated harvesting system (equipped
with 48 harvesting probes) and wash membranes bound to
filters three times with 5 mL of cold washing buffer (tubes are
subjected to three fill up and aspiration cycles).
10. Put filter circles in plastic scintillation vials, add 5 mL of scintilla-
tion cocktail and determine tritium-bound radioactivity in beta
liquid scintillation counter with 30–40% counting efficiency.
11. Analyze binding isotherms with a nonlinear curve-fitting soft-
ware from GraphPad Prism (GraphPad Software, San Diego,
CA). Determine equilibrium dissociation constant (Kd, nM)
and maximal binding capacity (Bmax, pmol/mg membrane pro-
teins) of [3H]-SCH23390 with saturation curves (see Note
38). A representative example of saturation curves obtained in
membrane preparations from transfected HEK293 cells with
wild type (WT) and single-point mutant forms of hD1R and
hD5R is shown in Fig. 4. Averaged Kd and Bmax values of
[3H]-SCH23390 are reported in Table 4. Data are also
expressed relative to hD1R-WT and hD5R-WT (Fig. 5). The
S287A, S287G and S287D mutations have no significant effect
on Kd of [3H]-SCH23390 while Bmax values are differentially
modulated. Meanwhile, the S263G mutation in hD1R leads to
a small but significant reduction of Kd of [3H]-SCH23390.
This increased affinity of hD1R-S263G for [3H]-SCH23390 is
not observed in cells transfected with hD1R-S263A and hD1R-
S263D. Moreover, while Bmax of hD1R-S263D is not changed
relative to hD1R-WT, hD1R-S263G and hD1R-S263A dis-
play a reduction and augmentation of Bmax, respectively.
3.14. Competition 1. Set up 16 three-tier polypropylene racks (6 × 12 holes per row,

Studies Fisher Scientific) containing 48 polystyrene test tubes
(12 × 75 mm) to carry binding reactions in a final volume of
500 mL (see Note 39).
2. Thaw frozen tubes of membranes on ice and mix with 11 mL
of resuspension buffer using Brinkmann Polytron at velocity of
17,000 rpm for 15 s. Leave membrane resuspension on ice
until use (see Note 40).
3. Prepare a 10× concentration of [3H]-SCH23390 (84 Ci/
mmol) in Milli-Q water ranging from ~7 to 13 nM (final in
assays: ~0.7–1.3 nM) (see Note 41).
Fig. 4. Saturation curves for HEK293 cells transfected with wild type and single-point mutant forms of hD1R and hD5R.
Representative examples for saturation curves generating data for Table 4 are shown. Saturation curves of [3H]-SCH23390
on crude membrane preparations were determined in the presence of 10 mM cis-flupenthixol. TOTAL total binding, NS
nonspecific binding, hD1R human D1 receptor, hD5R human D5 receptor, WT wild type.
164
Table 4
Ligand binding properties of wild type and single-point mutant forms of dopamine hD1R and hD5R
Saturation studies Competition studies

B. Plouffe and M. Tiberi
[3H]-SCH23390 Ki (nM)
Receptor Kd (nM) Bmax (pmol/mg prot.) SCH23390 Dopamine cis-Flupenthixol (+)-Butaclamol
hD1R-WT 0.82 (0.64–1.05) 13.5 (9.2–17.8) 0.80 (0.56–1.13) 7,420 (4,019–13,699) 9.42 (5.89–15.1) 5.15 (3.35–7.93)
hD1R-S263A 0.64 (0.43–0.96) 18.7 (12.5–24.9) 0.66 (0.57–0.78) 11,085 (9,273–13,252) 8.37 (6.04–11.6) 4.98 (3.51–7.05)
hD1R-S263G 0.59 (0.40–0.86) 7.8 (5.1–10.6) 0.57 (0.42–0.77) 2,815 (2,341–3,385) 7.25 (5.49–9.58) 4.39 (3.46–5.55)
hD1R-S263D 0.69 (0.55–0.86) 13.4 (9.6–17.2) 0.70 (0.60–0.82) 6,734 (5,369–8,446) 8.50 (6.11–11.8) 4.73 (3.70–6.03)
hD5R-WT 1.22 (0.90–1.65) 13.9 (10.1–17.7) 0.98 (0.84–1.15) 735 (542–997) 12.1 (8.69–16.9) 29.3 (18.2–47.3)
hD5R-S287A 1.03 (0.87–1.22) 17.6 (15.6–19.7) 0.91 (0.75–1.10) 916 (605–1,388) 11.8 (9.73–14.3) 24.7 (15.4–39.4)
hD5R-S287G 0.95 (0.69–1.31) 12.1 (10.3–14.0) 0.89 (0.68–1.15) 423 (314–572) 11.0 (9.27–13.0) 23.6 (14.0–39.8)
hD5R-S287D 1.32 (1.11–1.56) 17.6 (14.3–20.8) 1.08 (0.95–1.23) 841 (636–1,113) 15.2 (13.1–17.7) 30.0 (16.3–55.2)
Data are expressed as geometric (Kd, Ki) and arithmetic (Bmax) means with 95% lower and upper confidence intervals from 5 to 6 experiments done in dupli-
cate determinations. Best-fitted parameters were obtained from binding isotherms analyzed using nonlinear curve regression programs from GraphPad Prism.
hD1R-WT, wild type human D1R; hD5R-WT, wild type human D5R; Kd, equilibrium dissociation constant; Bmax, maximal binding capacity; Ki, equilibrium
dissociation constant of unlabeled drug
Fig. 5. Relative equilibrium dissociation constant (Kd) and maximal binding capacity (Bmax ) values of hD1R and hD5R single-point
mutants. Arithmetic means ± S.E. of Kd (a) and Bmax (b) values of hD1R and hD5R single-point mutants were calculated rela-
tive to their respective wild type receptor counterpart. *P < 0.05 when compared with a value of 1 (wild type receptor) using
one-sample t test. hD1R-WT wild type human D1 receptor, hD5R-WT wild type human D5 receptor.
4. Prepare 11 increasing 10× concentrations of SCH23390

(~3 × 10−11 to 3 × 10−6 M; ~3 × 10−12 to 3 × 10−7 M final in assays),
cis-flupenthixol (~4 × 10−10 to 5 × 10−5 M; ~4 × 10−11 to
5 × 10−6 M final in assays) and (+)-butaclamol (~3 × 10−10 to
3 × 10−5 M; ~3 × 10−11 to 3 × 10−6 M final in assays) in Milli-Q
water using frozen stock (see Note 5).
5. Prepare a fresh 10× ascorbic acid solution (1 mM) in Milli-Q
water.
6. Prepare 11 increasing 10× concentrations of dopamine
(~3 × 10−8 to 1 × 10−2 M; final in assays) in 10× ascorbic acid
solution (see Note 42).
7. Add 300 mL of binding buffer (final in assays: 50 mM Tris–
HCl, pH 7.4, 120 mM NaCl, 4 mM MgCl2, 1.5 mM CaCl2,
5 mM KCl, and 1 mM EDTA, pH 8.0).
8. Add 50 mL of Milli-Q water or 10× ascorbic acid to total binding
tubes.
9. Add 50 mL of different 10× concentrations of cold drugs from

the lowest to highest concentration.
10. Add 50 mL of the fixed 10× concentration of
[3H]-SCH23390.
11. Add 100 mL of membrane preparations to tubes (96 tubes for
each receptor tested), shake racks to mix and incubate at room
temperature (~20°C) for 90–120 min.
12. Terminate binding reactions by rapid filtration through precut
Whatmann GF/C glassfiber filter sheets (11.4 cm × 31.1 cm)
using Brandel semi-automated harvesting system and wash
membranes bound to filters three times with ~5 mL of cold
washing buffer (tubes are subjected to three fill up and aspira-
tion cycles).
13. Put filter circles in plastic scintillation vials, add 5 mL of scintil-
lation cocktail and determine tritium-bound radioactivity in beta
liquid scintillation counter with 30–40% counting efficiency.
14. Analyze binding isotherms using GraphPad Prism software.
Determine equilibrium dissociation constant of cold (unla-
beled) drugs (Ki, nM). A representative example of competi-
tion curves obtained in membrane preparations from transfected
HEK293 cells with WT and single-point mutant forms of
hD1R and hD5R is depicted in Fig. 6. Averaged Ki values of
unlabeled drugs are reported in Table 4. Data are also expressed
relative to hD1R-WT and hD5R-WT (Fig. 7). None of the
mutations has a major effect on the affinity of cis-flupenthixol
and (+)-butaclamol, two inverse agonists. Interestingly, in both
hD1R and hD5R, the S-to-G mutation leads to a significant
increase in dopamine affinity (lower Ki value) of receptors
while S-to-D mutation had no effect (Table 4). Additionally,
we observe that the S-to-A mutation promotes a decrease in
dopamine affinity of hD1R and hD5R, a trend that is not how-
ever statistically significant. Importantly, Ki values measured
with unlabelled SCH23390 are in line with Kd values of
[3H]-SCH23390 (Table 4). It is also worth mentioning that
the increased affinity of hD1R-S263G for [3H]-SCH23390
was also observed with unlabelled SCH23390 using competi-
tion studies (Table 4).
3.15. Establishment 1. Prepare MEM containing 5% (v/v) FBS and 40 mg/mL gen-
of Dopamine Dose– tamicin in a BSC and add sterile [3H]-adenine (stock at 1 mCi/
Response Curves mL) using a dilution factor 1:1,000 to obtain a final activity of
using Whole Cell 1 mCi/mL.
cAMP Assays 2. Aspirate medium from 12-well plates, add 1 mL of prewarmed
(37°C) labeling media per well and incubate HEK293 cells
with [3H]-adenine overnight at 37°C in humidified 5% CO2
incubator (see Note 43).
Fig. 6. Competition curves for HEK293 cells transfected with wild type and single-point mutant forms of hD1R and hD5R.
Representative examples for competition curves generating data for Table 4 are shown. Competition curves on crude
membrane preparations from wild type and single-point mutant forms of hD1R (left panels) and hD5R (right panels) were
performed with 0.5–1.2 nM of [3H]-SCH23390 in the absence and presence of increasing concentrations of unlabeled
drugs. Concentration of unlabeled drugs (M) is given as log values. hD1R-WT wild type human D1 receptor, hD5R-WT wild
type human D5 receptor.
Fig. 7. Relative equilibrium dissociation constant of unlabeled drug (Ki) values of hD1R and hD5R single-point mutants.
Arithmetic means ± S.E. of Ki values of hD1R and hD5R single-point mutants were calculated relative to their respective
wild type receptor counterpart. *P < 0.05 when compared with a value of 1 (wild type receptor) using one-sample t test.
hD1R-WT wild type human D1 receptor, hD5R-WT wild type human D5 receptor.
3. Prepare two scintillation vials containing 50 mL of labeling

media and 5 mL of scintillation cocktail. Count in a beta liquid
scintillation counter (see Note 44).
4. After an overnight labeling with [3H]-adenine, perform whole
cell cAMP assays as follows.
5. Thaw 200 mM IBMX stock and prepare 20 mM HEPES-
buffered MEM containing 1 mM IBMX. Keep in a 37°C water
bath until use.
6. Prepare racks of polystyrene test tubes (12 × 75 mm) containing
100 mL of neutralizing solution.
7. Prepare a fresh 100× ascorbic acid solution (10 mM) in Milli-Q
water.
8. Prepare seven increasing 100× concentrations of dopamine
(10−9 to 10−3 M; 10−11 to 10−5 M final in assays) in 100× ascorbic
acid solution using 1:10 dilution factor (see Note 42).
9. Aspirate labeling medium and add 1 mL of 20 mM HEPES-
buffered MEM containing 1 mM IBMX per well.
10. Add 10 mL of 100× ascorbic acid to three wells (triplicate
determinations).
11. Add 10 mL of increasing 100× concentrations of dopamine in
triplicate to remaining wells of the two 12-well plates.
12. Incubate 12-well plates at 37°C for 30 min (see Note 45).
13. At the end of treatment period, put 12-well plates on ice tray,
aspirate medium, add 1 mL of cAMP stop solution and incubate
cells at 4°C for 30 min.
14. Transfer lysates from wells to tubes containing neutralizing
solution, mix using vortex and put a parafilm sheet to seal top
of tubes in racks. Store at 4°C until use (see Note 46).
15. Prepare membranes from the binding dish as follows.
16. Put dishes on ice tray, aspirate culture medium and add 5 mL
of cold PBS to side of dishes and wash by gentle rocking of
dishes.
17. Aspirate PBS, add 5 mL lysis buffer, detach cells with a cell
lifter by scraping off the dish surface and transfer lysates to a
15 mL polycarbonate centrifuge tubes (18 × 100 mm, Beckman
Coulter). Wash dishes again with 5 mL of lysis buffer and trans-
fer whole volume to centrifuge tubes (final volume of 10 mL).
Prepare membranes essentially as described in Subheading 3.12
(steps 3–6). Resuspend final pellets in 0.6–1 mL of resuspen-
sion buffer and perform binding reactions using 100 mL of
membranes and a saturating concentration of [3H]-SCH23390
in the absence and presence of cold 10 mM cis-flupenthixol as
described above (see Subheading 3.13). Measure protein
concentrations with the Bio-Rad assay kit with BSA as standard
to determine Bmax in pmol/mg membrane proteins.
3.16. Purification 1. Mount Bio-Rad Poly-Prep columns on custom-made Plexiglas

of [3H]-cAMP with racks (100 per rack), fill separate set of columns with Dowex
Sequential and alumina and establish elution profile of columns using
Chromatography 0.1 N NaOH/ 0.1 N HCl/distilled water (Dowex) and 0.1 M
on Dowex and imidazole (alumina) as previously described (19).
Alumina Columns 2. Prepare KOH-neutralized cAMP samples for sequential chro-
and Quantification matography as follows.
of Intracellular 3. Prepare also triplicate samples of 1 mL of cAMP stop solution
[3H]-cAMP Mediated in tubes containing neutralizing solution, mix using vortex and
by Dopamine store with [3H]-cAMP lysate samples at 4°C (see Note 47).
4. Centrifuge test tubes containing KOH-neutralized cAMP
samples at ~500 g (1,500 rpm) in Beckman Coulter AllegraTM
6R Centrifuge for 10 min (see Note 48).
5. Load 0.85 mL of samples on separate Dowex columns and let
drain the flow through fraction (see Note 49).
6. Add 1 mL of distilled water to each column, let drain the water
wash fraction from Dowex and elute samples from Dowex on
alumina columns using 4 mL of distilled water.
7. Let drain water wash fraction from alumina, wash with 1 mL of
0.1 M imidazole and let drain the imidazole wash fraction.
8. Elute samples from alumina columns into plastic scintillation
vials using 3 mL of 0.1 M imidazole. Add 18 mL of scintilla-
tion cocktail. Shake vials vigorously to mix aqueous samples
(~3 mL) with scintillation cocktail and count vials in beta liq-
uid scintillation counter using dual 3H and 14C channels (see
Note 50).
9. Take a 50 mL aliquot from leftover volume (~200 mL) of each
KOH-neutralized cAMP lysate samples, add to plastic scintilla-
tion vials, fill vials with 5 mL of scintillation cocktail and count
in beta liquid scintillation counter using dual 3H and 14C channels
(see Note 51).
10. Enter 3H and 14C counts in Excel spreadsheet and compute
intracellular [3H]-cAMP levels in the absence and presence of
increasing concentrations of dopamine using formulas described
in Plouffe et al. (19).
11. Compute intracellular cAMP levels expressed in arbitrary units
(CA/TU × 1,000) and analyze averaged dose–response curves
with a four-logistic parameter equation using a nonlinear curve-
fitting program from GraphPad Prism to determine effective
concentration of dopamine that produces 50% of maximal stim-
ulation (EC50, an index of potency) and dopamine-mediated
maximal stimulation of AC (Emax). The best-fitted dose–response
curves to dopamine obtained with cells expressing similar recep-
tor levels are reported in Fig. 8. Our data show that S-to-A,
S-to-G, and S-to-D mutations modulate EC50 and Emax.
Fig. 8. DA-induced formation of intracellular cAMP by wild type and single-point mutant forms of hD1R and hD5R in intact
HEK293 cells. Cells were grown in medium containing 1 mCi/mL of [3H]-adenine overnight. Intracellular cAMP levels were
determined in 12-well dishes in the absence and presence of increasing concentration of DA (10−11 to 10−5 M). Each point
represents the arithmetic mean ± S.E. of three to four experiments done in triplicate determinations. (a) Dose–response
curves to DA were simultaneously analyzed by a four-parameter logistic equation using GraphPad Prism version 5.03 using
constrained or unconstrained parameters. Best-fitted values for effective concentration that elicits 50% of maximal stimu-
lation (EC50, nM) with 95% lower and upper confidence intervals are as follows: 14.7 (6.87–31.4) for hD1R, 43.8 (18.5–104)
for hD1R-S263A, 9.90 (5.20–18.9) for hD1R-S263G, 30.1 (13.2–68.4) for hD1R-S263D, 1.50 (0.64–3.49) for hD5R, 3.15
(1.11–8.94) for hD5R-S287A, 0.72 (0.28–1.85) for hD5R-S287D, and 1.96 (0.68–4.98) for hD5R-S287G. (b) Best-fitted
values for Emax ± S.E. obtained from dose–response curves to dopamine in HEK293 cells expressing wild type and single-
point mutant forms of hD1R and hD5R are shown. (c) EC50 shift (expressed as arithmetic mean ± S.E.) of single-point
mutants were calculated relative to wild type receptor using EC50 values derived from individual fitted dose–response
curves. *P < 0.05 when compared with wild type receptor. The Bmax values in pmol/mg/membrane proteins (expressed as
arithmetic mean ± S.E.) are as follows: 1.04 ± 0.18 (hD1R-WT), 1.34 ± 0.32 (hD1R-S263A), 1.19 ± 0.35 (hD1R-S263G),
1.15 ± 0.20 (hD1R-S263D), 1.12 ± 0.34 (hD5R), 1.31 ± 0.28 (hD5R-S287A), 0.83 ± 0.04 (hD5R-S287G), and 1.04 ± 0.26
(hD5R-S287D). hD1R-WT wild type human D1 receptor, hD5R-WT wild type human D5 receptor.
Interestingly, the trend in mutation-specific effects is generally

recapitulated in cells expressing hD1R and hD5R mutants
(Fig. 8). Data obtained with the S-to-G mutation are in agree-
ment with the increased dopamine affinity for hD1R-S263G
and hD5R-S287G (Table 4 and Fig. 7). The S-to-G mutation
may thus promote the release of molecular constraints within
the cytosolic end of TM6, leading to a more “relaxed” confor-
mation for agonist binding and agonist-dependent G protein
coupling. This idea is to some extent supported by our binding
data showing increased affinity of hD1R-S263G for SCH23390
(Figs. 6 and 7). Indeed, SCH23390 is a classical D1-like antag-
onist that behaves pharmacologically as a partial agonist in
HEK293 cells (9, 20). In addition, the decreased potency (EC50
rightward shift) observed with the S-to-A mutation (Fig. 8)
suggests a potential role of the hydroxyl group on the side chain
of serine in the formation of hydrogen bonds, which may be
only involved in agonist-dependent G protein coupling. Indeed,
in contrast to S-to-G mutation, the S-to-A mutation has no
effect on agonist binding (Table 4 and Fig. 7). Therefore,
S-to-A mutation may impart subtle structural changes to the
cytosolic end of TM6 that specifically impact the agonist-
dependent G protein-coupling conformation. Meanwhile, the
S-to-D mutation in hD1R (and to a lesser extent in hD5R)
paradoxically increases EC50 (lower potency, i.e., reduced G
protein coupling) and Emax (greater dopamine efficacy) (Fig. 8).
These results may imply that the phosphomimetic S-to-D
mutation promotes the formation of an ionic bond between
the negatively charged side chain in aspartic acid and a posi-
tively charged amino acid (arginine, histidine or lysine).
Ultimately, this may lead to the formation of a new GPCR
conformation with distinct G protein coupling properties when
compared with wild type receptors. However, it remains to be
established whether results obtained with the phosphomimetic
S-to-D mutation can also be recapitulated with the phosphory-
lation of S263 and S287 in hD1R and hD5R, respectively. In
closing, it is also worth mentioning that the role of S263
(hD1R) and S287 (hD5R) in regulating the constitutive activity
will require additional studies. In fact, dose–response curves in
12-well dishes do not provide the best experimental frame-
work to address this important question (see Note 52).
4. Notes
1. Ethidium bromide wastes (tip, gel) should be disposed as per

the experimenter’s research institution rules.
2. The 1 M HEPES solution is utilized to prepare 2× HEPES-
buffered saline solution, which is used in the transfection
procedure. Note that the pH of this HEPES solution is different

from that of the sterile HEPES buffer solution (pH 7.4)
employed in the preparation of 20 mM HEPES-buffered MEM
(whole cAMP assays).
3. Na3PO4 precipitates can form over time. Precipitates can be
easily dissolved by swirling solution in a 37°C water bath prior
to preparing stock of 2× HEPES-buffered saline solution.
4. Handling and disposal of radioactive material should be done
as per radiation safety office rules of the experimenter’s research
institution.
5. Experimenter can make stock solution of cold drugs in Milli-Q
water (SCH23390, cis-flupenthixol) or ethanol ((+)-butacla-
mol) at a final concentration of 10 mM. Store stock solution of
drugs at −20°C. However, ascorbic acid and dopamine solu-
tions are always made fresh.
6. The IBMX-DMSO stock solution will freeze at 4°C. The
200 mM IBMX is made in a 50 mL polypropylene conical
tube, which facilitates rapid thawing in 37°C water bath. We
recommend preparing small volume of solution, typically using
5 mL of DMSO, to avoid multiple “freeze–thaw” cycles. The
IBMX stock solution should be discarded if experimenters note
that it is unfrozen when stored at 4°C.
7. Typically, stop solution is prepared in large polypropylene
graduated cylinder (2 L) and carefully poured in two dark glass
bottles (~750 mL) fitted with a bottletop dispenser. The stop
solution can be kept for 4 months at 4°C. Unlabelled cAMP is
used to saturate phosphodiesterases and hence to prevent deg-
radation of [3H]-cAMP formed by receptor stimulation.
[14C]-cAMP is used as a tracer to measure [3H]-cAMP recov-
ery during chromatography column procedure.
8. The imidazole working solution is diluted in Milli-Q water at
0.1 M. 0.1 M imidazole is stored at room temperature in a
glass bottle equipped with bottletop dispenser and used during
alumina column washing and elution procedure.
9. The electrophoresis of uncut DNA is recommended and useful
in assessing the efficiency of digestion with selected restriction
enzymes and isolation of positive clones for single-point
mutants.
10. The volume of purified mutant receptor cassette can be
increased to 16 mL if the recovery yield of mutant receptor cas-
sette is low after purification using QIAEX beads.
11. Double digestions using restriction enzymes from Fermentas
are typically performed in either 2× or 1× buffer Y+/TangoTM
(yellow) according to optimal enzyme activity. For a double
digestion using BsmI and EagI, 2× buffer Y+/TangoTM gives
the highest enzyme activities.
12. Do not add loading dye following digestion of hD1R-pCMV5

and hD5R-pCMV5 DNA constructs because it will compromise
CIAP dephosphorylation. This step is critical for the ligation
reaction. It allows dephosphorylation of 5¢-ends of digested
constructs DNA that have not be fully cut with both enzymes
and hence reducing the self-ligation of single enzyme-digested
DNA molecules during ligation reactions. Ultimately, this
will decrease the number of false positive on your test ligation
plates.
13. A fraction of the hD5R DNAs is partially digested with BsmI
and EagI. It is thus recommended to run side by side uncut
and cut samples on 1.8% (w/v) agarose gel to visualize and
isolate DNA bands fully digested by BsmI and EagI.
14. We usually prepare ligation reactions using a vector and insert
ratio of 1:3. The volume of vector and insert used in our liga-
tion reactions is based on relative size and intensity between
vector and insert bands. If yield of purified vector is far in excess
of your purified insert, we recommend adding 0.5 mL of a 1:10
dilution of the purified vector in sterile Milli-Q water.
15. Typically, a water bath is placed in cold room or chromatogra-
phy refrigerator (4°C) and temperature of water bath is set to
reach 16°C.
16. Optimal desalting is critical for electroporation of bacteria. If
high salt concentrations remain in your ligation reactions, this
will create an electrical discharge or arcing during electropora-
tion. Arcing will significantly decrease the viability of bacteria
and ultimately lead to a poor efficiency of transformations.
17. We generally obtain less than 5 colonies following transforma-
tion with control ligations while test ligation yield between 25
and 50 colonies. The rate of positive colonies ranged between
80 and 90%.
18. The preparation of bacterial glycerol stocks of selected colonies
prior to making plasmid DNA minipreps for analysis using
restriction digestions and automated DNA sequencing will
avoid retransforming bacteria with the positive plasmid DNA
following the confirmation of the positive mutation. Overall,
having readily available bacterial glycerol stocks will save the
experimenter from redoing steps described in Subheading 3.7
and thus work more efficiently.
19. Alternatively, boiling methods can be used to prepare plasmid
DNA minipreps. The boiling methods are perhaps more cost-
efficient in comparison to using QIAprep Spin columns and as
suitable for restriction digestions. However, in our experience,
boiling methods do not give high enough DNA quality for
automated DNA sequencing and thus the experimenter will
have to rely on other methods to make higher quality plasmid
DNA preps. In our opinion, the cost does not justify using
boiling methods.
20. Concentrations of plasmid DNA minipreps range from 0.25 to
0.5 mg/mL with a purity (OD260/OD280 ratio) of ~1.8–1.9.
21. Automated DNA sequencing protocol may vary according to
company or core facility’s protocol. Make sure that method
described herein is suitable with core facility or company per-
forming automated DNA sequencing.
22. If the same oligonucleotides is used as PCR and sequencing
primers, experimenter will make sure that primers are made at
appropriate concentrations for PCR and sequencing reactions.
23. This step does not require the addition of ampicillin as diluted
bacterial cultures are immediately inoculated into LB contain-
ing 1× ampicillin.
24. ATTC suggest that HEK293 cells be grown in 15% (v/v) horse
serum. However, we have replace horse serum with heat-inac-
tivated FBS without aberrant effect on cell growth.
25. Do not add PBS directly on cells as they may detach from flasks
before adding trypsin.
26. You can gently rock flasks to accelerate trypsinization of cells.
27. Prior to begin cell culture work, put trypsin and complete
MEM 37°C water bath to warm up solutions. When ready to
start cell culture, trypsin and MEM bottles are wiped with 70%
(v/v) ethanol and put in level 2-certified biological safety cabi-
net (BSC). Trypsin and MEM can be left at room temperature
in BSC during cell culture procedures.
28. Dishes can also be seeded at a cell density up to 2.5 × 106 cells/
dish. If you observed that cells seeded at 2.5 × 106 cells/dish
grow too fast at high number of cell passages (>P48), we rec-
ommend that dishes be seeded at a lower cell density (2 × 106
cells). In our experience, cells transfected at a density ranging
from 2 to 2.5 × 106 cells does not significantly impact the
D1-like receptor expression. However, our unpublished data
suggest that when cells are seeded at higher cell density than
2.5 × 106 cells/dish (>3 × 106 cells/dish), there is a greater vari-
ability in receptor expression. Experimenter should bear in
mind that the aforementioned guidelines are those that have
been optimal for our laboratory.
29. For radioligand binding studies, we typically transfect each
dish of cells with 5 mg of plasmid DNA. In our hands, this
amount of DNA yields in transfected HEK293 cells the
maximal achievable receptor expression as measured with
[3H]-SCH23390 (~15–20 pmol/mg membrane proteins).
Indeed, the use of higher amounts of DNA (10–20 mg/dish)

does not lead to greater expression levels of wild type (WT) or
low-expressing mutant forms of hD1R or hD5R. With respect
to performing dose–response curves in intact cells, we titrate
the amount of receptor expression construct DNA to be used in
transfection to achieve lower levels of receptor (1–3 pmol/mg
membrane proteins). For instance, the amounts of receptor
expression construct DNA used herein for dose–response
curves are as follows: hD1R-WT (0.04–0.07 mg/dish), hD1R-
S263A (0.04 mg/dish), hD1R-S263G (0.08–12 mg/dish),
hD1R-S263D (0.06–0.08 mg/dish), hD5R-WT (0.04–
0.08 mg/dish), hD5R-S287A (0.06 mg/dish), hD5R-S287G
(0.06–0.08 mg/dish), and hD5R-S287D (0.08 mg/dish).
Importantly, when using lower amount of 5 mg/dish of receptor
expression construct DNAs, the total amount of plasmid DNA
must be normalized at a constant amount per transfected dish
(5 mg) using empty plasmid (e.g., pCMV5). This will mitigate
variations in transfection efficiency between different condi-
tions. The efficiency in HEK293 cells obtained using our trans-
fection method will not be covered here as this issue has been
previously discussed elsewhere (17, 19).
30. Remove drips left on side of tubes by gentle tapping on a hard
surface inside the BSC.
31. The pH of 2× HEPES-buffered saline solution (pH to 7.1 ±
0.05) is critical for optimal transfection of plasmid DNA. Higher
and lower pH will significantly impact the formation DNA-
calcium phosphate precipitates. For instance, we observed a
drastic reduction in receptor expression with a solution at pH
of 7.3.
32. We observe that the proliferation rate of HEK293 cells slowly
get higher up to 52 passages under our cell culture conditions.
After 52 passages, HEK293 cells grow more rapidly and some-
times as foci. In addition, cells adhere less on dishes at older
passages. In our laboratory, we generally utilize HEK293 cells
between 40 and 52 passages.
33. Typically, we use four transfection dishes per experimental
condition in radioligand binding and whole cell cAMP assays.
Meanwhile, the number of transfected dishes can be scaled up
if mutant receptors display low expression levels in transfected
HEK293 cells (<1 pmol/mg membrane proteins) or if larger
amounts of membrane preparations are needed for radioligand
binding assays.
34. For radioligand binding studies, we normally let transfected
HEK293 cells grow for 48 h following seeding in 150 × 25 mm
dishes. For instance, if the transfection is performed on Tuesday,
cells are split and seeded in new dishes on Wednesday and then
used to make membrane preparations on Friday. Alternatively,

the experimenter can use cells on Thursday (cells grown for
24 h) as we observe that expression is high enough to detect
receptor binding.
35. Cells in 100 × 20 mm dishes are used to prepare crude mem-
branes to determine receptor levels for each transfection condi-
tion using radioligand binding.
36. In general, 4 rows of 12 holes are used to set up two saturation
curves with six concentrations of [3H]-SCH23390 using dupli-
cate determinations for total and nonspecific binding (24 tubes
for each receptor tested) for a total of 48 tubes per rack. This
set up is optimal for terminating reactions using our Brandel
harvesting system. For the purpose of our study, four racks
were used to perform saturation studies with wild type and
single-point mutants of hD1R and hD5R. A set of four tubes
is used for each concentration of [3H]-SCH23390 (two tubes
for total binding followed by two tubes for nonspecific
binding).
37. We usually carry out assays with 25–100 mL of membrane
resuspension in a final volume of 800 mL of Milli-Q water in
glass test tubes (10 × 75 mm) and 200 mL of protein dye. Mix
well by vortexing several times.
38. Bmax value is used as an index of total receptor levels.
39. Typically, one rack (4 rows of 12 holes) allows performing two
competition curves. For the purpose of the present study, each
receptor construct to be tested requires two racks of 48 tubes
to test four dopaminergic compounds (SCH23390, dopamine,
cis-flupenthixol, (+)-butaclamol). Each competition curve is
done in the absence (total binding) and presence of 11 increas-
ing concentrations of cold drugs using duplicate determina-
tions (24 tubes for each drug tested).
40. Volume of resuspension can be reduced to increase the specific
binding signal if receptor constructs are expressed at a Bmax
lower than 1 pmol/mg of membrane proteins. If so, the num-
ber of drugs to be tested will be reduced accordingly.
Alternatively, to test four drugs in the same experiment, one
can prepare a larger number of frozen stocks of membrane
preparations of the given receptor expressed at low Bmax.
41. Competition curves are performed using a fixed concentration
of [3H]-SCH23390 similar to Kd values measured at wild type
and mutant receptors.
42. Competition studies and dose–response curves with dopamine
are done in the presence of ascorbic acid at a final concentration
of 0.1 mM in the assays to prevent dopamine oxidation.
43. Values obtained from dose–response curves remain essentially

unchanged whether HEK923 cells are metabolically labeled
with 1 or 2 mCi/mL of [3H]-adenine. Therefore, the use of
1 mCi/mL of [3H]-adenine will considerably decrease the
amount of handled radioactivity and cost of experiments. Make
also a note that if cells are not labeled on the following day of
seeding, cells can be labeled the same day of experiment using
an amount of ³2 mCi/mL of [3H]-adenine to increase signal
detection. Under these circumstances, cells are metabolically
labeled for a minimum of 4 h prior to doing whole cell cAMP
assays.
44. These samples are used to assess the amount of [3H]-adenine
in labeling media. Vials can be put aside until counting the
other vials with radiolabeled eluates obtained from double
column chromatography.
45. Cells can also be incubated with drugs for a shorter time
period.
46. Samples can be stored at 4°C up to 5 days. In our hands, no
differences have been measured between samples processed
immediately for double chromatography columns and samples
processed after being left at 4°C for 5 days. Samples should be
frozen at −20°C if longer time of storage is required. However,
experimenter should first verify that the used test tubes do not
crack −20°C.
47. The average [14C]-cAMP counts from triplicate samples (column
input) will be computed to determine the column recovery
and establish a correction factor (CF) for each set of paired
Dowex and alumina columns. The column input (C14total) is
calculated from counting the radioactivity in vials containing
0.85 mL of KOH-neutralized cAMP stop solution mixed with
2 mL of 0.1 M imidazole (to make a 3 mL aqueous sample)
and 18 mL of scintillation cocktail.
48. Samples are clarified using a low-speed centrifugation to pellet
down potassium perchlorate salts prior to applying samples on
Dowex columns. Loading salts on Dowex columns will inter-
fere with sample elution and lead potentially to spurious radio-
active counts.
49. We choose to apply a volume of 0.85 mL on Dowex columns
to prevent touching and pipetting salt precipitates from the
bottom of test tubes.
50. While the volumes for washes and elution of samples on Dowex
and alumina columns reported herein are those routinely
obtained in our laboratory, it is strongly recommended that the
column profiles be thoroughly reassessed in the experimenter’s
laboratory as described (19).
51. [3H] counts measured in 50 mL aliquots will be used to

determine the total amount of [3H]-adenine uptake (TU) in
each well (an index of cell number).
52. The assessment of constitutive activity of wild type and mutant
forms of hD1R and hD5R is beyond the scope of this study.
The determination of constitutive activity of wild type and
mutant forms of hD1R and hD5R is typically performed with
cells seeded in 6-well plates and incubated in labeling medium
containing 2 mCi/mL of [3H]-adenine overnight at 37°C in
humidified 5% CO2 incubator. Generally, cells used for these
assays express receptors at higher Bmax values. Methods pertain-
ing to the study of constitutive activity of D1-like receptors
and their mutants have been discussed elsewhere (19).
Acknowledgments
We would to thank Dr. Kursad Turksen for his advice and Andrew
Charrette for reading the manuscript. Bianca Plouffe was a recipi-
ent of a graduate scholarship from Fonds de la recherche en santé
du Québec. This work was supported by an operating grant from
Canadian Institutes of Health Research (MOP-81341).
References
1. Arinaminpathy Y, Khurana E, Engelman DM, 7. Davey J (2004) G-protein-coupled receptors:
Gerstein MB (2009) Computational analysis new approaches to maximise the impact of
of membrane proteins: the largest class of drug GPCRS in drug discovery. Expert Opin Ther
targets. Drug Discov Today 14:1130–1135 Targets 8:165–170
2. Hubert P, Sawma P, Duneau JP, Khao J, Henin 8. Yildirim MA, Goh KI, Cusick ME, Barabasi
J, Bagnard D, Sturgis J (2010) Single-spanning AL, Vidal M (2007) Drug-target network. Nat
transmembrane domains in cell growth and Biotechnol 25:1119–1126
cell-cell interactions: more than meets the eye? 9. Tiberi M, Caron MG (1994) High agonist-
Cell Adh Migr 4:313–324 independent activity is a distinguishing feature
3. Rosenbaum DM, Rasmussen SG, Kobilka BK of the dopamine D1B receptor subtype. J Biol
(2009) The structure and function of G-protein- Chem 269:27925–27931
coupled receptors. Nature 459:356–363 10. Missale C, Nash SR, Robinson SW, Jaber M,
4. Fredriksson R, Lagerstrom MC, Lundin LG, Caron MG (1998) Dopamine receptors: from
Schioth HB (2003) The G-protein-coupled structure to function. Physiol Rev 78:189–225
receptors in the human genome form five main 11. Beaulieu JM, Gainetdinov RR (2011) The
families. Phylogenetic analysis, paralogon physiology, signaling, and pharmacology of
groups, and fingerprints. Mol Pharmacol dopamine receptors. Pharmacol Rev 63:
63:1256–1272 182–217
5. Pineyro G (2009) Membrane signalling com- 12. Cotecchia S (2007) Constitutive activity and
plexes: implications for development of func- inverse agonism at the alpha(1)adrenoceptors.
tionally selective ligands modulating heptahelical Biochem Pharmacol 73:1076–1083
receptor signalling. Cell Signal 21:179–185 13. Jarvie KR, Caron MG (1993) Heterogeneity of
6. Ritter SL, Hall RA (2009) Fine-tuning of GPCR dopamine receptors. Adv Neurol 60:325–333
activity by receptor-interacting proteins. Nat 14. Charpentier S, Jarvie KR, Severynse DM,
Rev Mol Cell Biol 10:819–830 Caron MG, Tiberi M (1996) Silencing of the
constitutive activity of the dopamine D1B molecular interplay between the third extracel-
receptor. Reciprocal mutations between D1 lular loop and the cytoplasmic tail. J Biol Chem
receptor subtypes delineate residues underlying 278:8146–8153
activation properties. J Biol Chem 18. Tumova K, Zhang D, Tiberi M (2004) Role of
271:28071–28076 the fourth intracellular loop of D1-like dopamin-
15. Jackson A, Iwasiow RM, Tiberi M (2000) ergic receptors in conferring subtype-specific sig-
Distinct function of the cytoplasmic tail in naling properties. FEBS Lett 576:461–467
human D1-like receptor ligand binding and 19. Plouffe B, D’Aoust JP, Laquerre V, Liang B,
coupling. FEBS Lett 470:183–188 Tiberi M (2010) Probing the constitutive
16. Demchyshyn LL, McConkey F, Niznik HB activity among dopamine D1 and D5 recep-
(2000) Dopamine D5 receptor agonist high tors and their mutants. Methods Enzymol
affinity and constitutive activity profile con- 484:295–328
ferred by carboxyl-terminal tail sequence. J Biol 20. D’Aoust JP, Tiberi M (2010) Role of the
Chem 275:23446–23455 extracellular amino terminus and first mem-
17. Tumova K, Iwasiow RM, Tiberi M (2003) brane-spanning helix of dopamine D1 and D5
Insight into the mechanism of dopamine receptors in shaping ligand selectivity and
D1-like receptor activation. Evidence for a efficacy. Cell Signal 22:106–116
Chapter 11
A Molecular Genetic Approach to Uncovering

the Differential Functions of Dopamine D2 Receptor
Isoforms
Yanyan Wang, Toshikuni Sasaoka, and Mai T. Dang
Abstract
Alterations in the activity of the dopamine D2 receptor (D2R) have been implicated in several neurological
and psychiatric disorders, including schizophrenia, Parkinson’s disease, Huntington’s disease, Tourette
syndrome, attention-deficit hyperactivity disorder (ADHD), and drug addiction. Two isoforms of D2R,
long form (D2LR) and short form (D2SR), have been identified. The specific function of each D2R iso-
form is poorly understood, primarily because isoform-selective pharmacological agents are not available.
Using homologous recombination, we have generated D2LR knockout (KO) mice. D2LR KO mice are
completely deficient in D2LR, but still express functional D2SR at a level similar to the total D2R level in
wild-type (WT) mice. D2LR is generally the predominant isoform expressed in WT mice. We showed that
D2LR KO mice displayed a number of robust behavioral phenotypes distinct from WT mice, indicating
that D2LR and D2SR have differential functions. In this chapter we describe the generation and charac-
terization of the D2LR KO mouse. This genetic approach provides a valuable research tool to investigate
the functional role of individual D2R isoforms in the mammalian central nervous system (CNS).
Key words: Dopamine D2 receptor, Knockout mouse, Homologous recombination, Transfection of

embryonic stem cell, Genotyping
1. Introduction
Dopamine (DA) receptors are classified into two categories: D1-like

receptors (D1 and D5 subtypes) and D2-like receptors (D2, D3,
and D4 subtypes). Additionally, two isoforms of dopamine D2
receptor (D2R) have been identified by molecular cloning in the
brain of rats, mice, and humans and are termed as the long recep-
tor form (D2LR) and short receptor form (D2SR) (1–4). D2LR
and D2SR are generated by alternative splicing of the same gene
181
182 Y. Wang et al.
Fig.1. Schematic illustration of the gene and mRNA structures of D2LR and D2SR. Open
and filled boxes on gene structure represent exons, and lines between boxes represent
introns. The Exon 6 (filled box) is 87 bp that encodes the D2LR insert region of 29 amino
acids in the third intracellular loop.
(Fig.1). No selective pharmacological ligands are available for these

two types of D2R, which has hampered the progress towards
uncovering their specific function in the CNS. To overcome this
challenge, we decided to generate mutant mice expressing only a
single D2R isoform using gene-targeting technology. We made a
line of mice that lack D2LR (D2LR KO mice) by selectively delet-
ing exon 6 in the D2 gene (5). We previously showed that D2LR
KO mice still express a functional D2SR isoform on the cell surface
at a level similar to the native level of the total D2R in wild-type
(WT) mice(5). The mice that express only D2SR allow us to exam-
ine the biological responses mediated by D2SR. Because D2LR is
the predominant isoform expressed in the brain of WT mice (5, 6),
the comparative studies between D2LR KO and WT mice should
help elucidate the functional role of D2LR (5–8).
Making a knockout mouse involves five major steps: (1) making a
gene-targeting vector, (2) transfecting the targeting vector into
embryonic stem (ES) cells and screening targeted ES clones by homol-
ogous recombination, (3) injecting the targeted ES cells into blasto-
cysts and returning the blastocysts to the foster mother to generate
chimeric mice, (4) breeding chimeric mice with WT mice to generate
heterozygous mice, and (5) genotyping to identify germline transmit-
ted mice and intercrossing heterozygotes to yield homozygous mice.
2. Materials
2.1. Making the 1. Mouse genomic DNA library can be obtained from Agilent
Targeting Vector Technologies, Inc. (Santa Clara, CA).
2. PGK-neo cassette can be obtained from: Gene Bridges GmbH,
Addgene (Cambridge, MA).
3. Cloning materials: Escherichia coli, pBluescript, Quick Ligation
Kit (New England Biolabs, Ipswich, MA), an assortment of
11 Functions of D2 Receptor Isoforms 183
restriction enzymes (New England Biolabs or Invitrogen Life

Science, Grand Island, NY).
4. Mini-plasmid preparation material: mini-prep solution, agarose
electrophoresis gel.
5. LB solution containing antibiotic.
6. Alkaline lysis solution I: 50 mM glucose, 25 mM Tris–HCl
(pH 8.0), 10 mM EDTA.
7. Alkaline lysis solution II: 0.2 M NaOH, 0.1% SDS, pre-
pared fresh.
8. Alkaline lysis solution III: 3 M potassium, 5 M acetate.
9. Isopropanol.
10. Isobutanol.
11. Cesium chloride.
12. Ethidium bromide.
13. 1 M Tris–HCl (pH 8.0).
14. 1 M Tris–HCl (pH 8.5).
15. 0.5 M EDTA (pH 8.0).
16. 20× SCC.
17. 50× TAE (Tris–acetate–EDTA).
18. TE (pH 8.0) (10 mM Tris–HCl—1 mM EDTA).
19. 10% SDS.
20. Ethidium bromide (10 mg/mL).
21. 100× Denhardt’s solution.
22. 5 M NaCl.
23. 1 M CaCl2.
24. 1 M potassium acetate.
25. 3 M sodium acetate.
26. Tris–EDTA buffer.
27. LB solution.
See “Molecular Cloning” (9) regarding how to make various
buffers (see Note 1). Pre-mixed buffers are also commercially
available.
2.2. Transfection of 1. Embryonic stem (ES) cells: J1 ES cell established from

Embryonic Stem Cells 129 S4/SvJae provided by Dr. En Li (Novartis Institute for
Biomedical Research).
2. Embryonic fibroblast (EF) cells, used to provide nourishment
to ES cells (American Type Culture Collection, Manassas,
VA).
3. ES medium (500 mL): 75 mL fetal calf serum (FCS), 2.5 mL
100× penicillin–streptomycin (Gibco, Invitrogen), 5 mL 100×
184 Y. Wang et al.
glutamine (Gibco, Invitrogen), 5 mL 100× nonessential amino

acids (Gibco, Invitrogen), 5 mL 100× nucleosides, 5 mL 100×
2-mercaptoethanol, 400 mL DMEM; filter with 0.2 μm filter
to sterilize and store at 4°.
4. 100× Nucleosides: 80 mg adenosine, 85 mg guanosine, 73 mg
cytidine, 73 mg uridine, 24 mg thymidine, 100 mL Milli-Q
H2O (mQH2O); dissolve for 15 min at 65°, filter (0.2 μm),
aliquot and store at −20°.
5. 100× 2-mercaptoethanol (Sigma-Aldrich, St. Louis, MO):
70 μL of 2-mercaptoethanol in 100 mL PBS, filter (0.2 μm).
This solution needs to be made every 2 weeks.
6. EF medium (500 mL): 50 mL FCS, 5 mL 100× penicillin–
streptomycin, 5 mL 100× glutamine, 450 mL DMEM with
HEPES, filter (0.2 μm) and store at 4°.
7. Plates and flasks for EF cells; treat the container with 0.1%
gelatin solution for 20 min at room temperature and then aspi-
rate the gelatin.
8. Trypsin (Gibco, Invitrogen), Geneticin (G418) (Gibco,
Invitrogen), Leukemia inhibitory factor (LIF, BRL; ESGRO
(mLIF), Invitrogen).
9. 2× Freezing medium: 80% FCS with 20% DMSO, filter to
sterilize (0.2 μm) and store at 4°.
10. Electroporation apparatus (Bio-Rad Gene Pulser, Bio-Rad,
Hercules, CA).
2.3. Injection of ES 1. Female C57BL/6 donor mice (2- to 3-month-old) (The

Cells into Blastocysts Jackson Laboratory, Bar Harbor, ME).
and Transfer 2. C57BL/6 male breeder mice (3- to 6-month-old, should be
of Blastocysts less than 1 year old) (Jackson).
to the Uterus
3. Female foster mice (2- to 3-month-old, B6CBAF1 or ICR).
2.3.1. Breeders 4. Vasectomized male mice (3- to 12-month-old, B6CBAF1
for Harvesting Blastocysts or ICR).
and Foster Mother Mice
5. ES medium.
6. Hand-pulled pipette and bulb or tubing with mouth piece.
7. Sterile surgical tools: scissors and forceps.
8. Pregnant mare’s serum gonadotropin (PMSG) (Sigma-
Aldrich), Human chorionic gonadotropin (HCG) (Sigma-
Aldrich), Phosphate-buffered saline (PBS).
2.3.2. Blastocysts Injection 1. ES medium.

2. Small culture plates.
3. Injection and holding pipettes.
4. Microinjector apparatus (Leica Microsystems, Wetzlar, Germany).
2.3.3. Surgery to Return 1. Sterile surgical tools: scissors (1 serrated for cutting through
Injected Blastocysts pelt and 1 fine for body wall), forceps, 2 Serafin clips, surgical
to Foster Mother wound clips. Hand-pulled pipette for transferring blastocyst.
2. Pentobarbital sodium (Somnopentyl injection, Schering-
Plough Animal Health Corp., Union, NJ).
3. Needle and syringe for injection. 70% ethanol.
4. Heating pad.
2.4. Genotyping 1. Lysis buffer (500 mL): 50 mL of 1 M Tris–HCl (pH 8.5),

the Mice 5 mL of 0.5 M EDTA (pH 8.0), 10 mL of 10% SDS, 20 mL of
5 M NaCl, 0.5 mL of 1 M CaCl2, and bring to a final volume
of 500 mL with mQ H2O or equivalent.
2. Proteinase K (20 mg/mL) (New England Biolabs,
Invitrogen).
3. dNTP (QIAGEN, Valencia, CA).
4. Mineral oil (Sigma).
5. AmpliTaq DNA Polymerase Stoffel Fragment (Applied
Biosystems Life Technologies).
6. PCR Thermo Cycler (MJ Research, Inc./ Bio-Rad). PCR
tubes (USA Scientific, Ocala, FL).
7. Agarose gel electrophoresis apparatus (Owl Scientific, Thermo
Fisher, Rockford, IL).
8. Agarose (Invitrogen)
9. 1× TAE buffer.
10. Ethidium bromide (Sigma).
11. Ultraviolet transilluminator and camera (or computer) for
taking the gel picture.
3. Methods
3.1. Making the Gene-targeting relies on homologous recombination, which is

Targeting Vector done by a cellular endogenous recombinase that recognizes the
for D2 Long Receptor similarity between the DNA sequence fragments of the targeting
Knockout vector (or construct) and that of the native gene. Recombination
occurs at the sites of similarity, and the altered genetic material
within those flanking sites is incorporated into the native gene.
The construct design determines the type of knockout mouse.
The simplest construct for producing a knockout is to replace a cod-
ing region of the gene of interest with an antibiotic resistant gene.
The removal of a coding region will interfere with transcription and
protein expression, disrupting the production of the protein or result-
ing in the translation of an unstable protein. Recombination efficiency
186 Y. Wang et al.
is sensitive to the homologous sequence. Generally, the size of the

deleted segment of the gene should be close to the size of the
insertion between the left homologous arm and the right homologous
arm of the targeting vector, although there are successful exceptions
(10). The antibiotic resistant gene most commonly used is the
neomycin (neo) resistant gene driven by a ubiquitous 3-phospho-
glycerate kinase (PGK) promoter, together referred to as a PGK-neo
cassette (11). The PGK-neo cassette is used as a positive selection
marker for homologous recombination in ES cells. Here we provide
mainly general guidelines in designing and making a gene-targeting
vector because this process is unique to the gene of interest.
Because D2LR and D2SR are generated from the same gene
by alternatively splicing, it is important to not disrupt transcrip-
tion of the D2 gene; otherwise you would produce a complete
D2R KO mouse. D2LR has a 29 amino acid insertion located
within the third intracellular loop, which is encoded by exon 6;
this insertion is absent in D2SR (Fig. 1). To generate the mice
deficient in D2LR, we selectively replaced exon 6 with the PGK-
neo cassette. In the D2LR KO mouse, D2SR is over-expressed to
a level similar to that of total D2R in WT mice. This is because
our manipulation of the D2 gene eliminated the possibility of
transcription of D2LR, and therefore the D2 gene was exclusively
available for making D2SR.
1. Find the genomic sequence from: www.ncbi.nlm.nih.gov/
genome/guide/mouse/
2. The mouse D2 gene containing exons 3–8 was cloned from a
mouse 129/SV genomic DNA library (see Note 2). Subclone
the gene into pBluescript in E. coli.
3. To verify the genomic clones, map some of the major restric-
tion enzyme sites on the exons and introns according to the
restriction map and exon/intron structure of the D2 gene
obtained from the database. This mapping will also identify
the orientation of the fragment. Sequence some important
gene regions (see Note 3). For example, we sequenced exon 6
and the exon/intron junctions between exon 6/intron 5 and
between exon 6/intron 6 on the D2 gene.
4. Replace exon 6 of the D2 gene fragment with a PGK-neo
cassette (Fig. 2a). The left homology arm of this construct is
7 kb and the right homology arm is 8.5 kb, which flank the
1.8 kb PGK-neo cassette (see Note 4).
5. While making the targeting vector, you should also design and
make probes for screening homologous recombination in ES
cells via Southern blot analysis. Usually the sequence of the probe
for Southern blot analysis should be outside of the targeting
vector (Fig. 2a). The probe should be able to distinguish homol-
ogous recombination from non-homologous recombination.
It is better to prepare two probes: one is external to the 5¢ end of

the left homology arm and the other is external to the 3¢ end
of the right homology arm.
6. Verify the accuracy of the targeting vector against the design
with restriction mapping, nucleotide sequencing, and Southern
blot analysis.
7. Produce a large-scale plasmid preparation of the finished tar-
geting vector with cesium chloride. Aliquot the targeting vec-
tor DNA into multiple tubes and store them at −20°.
8. For standard molecular biology protocols such as cloning,
plasmid DNA preparation, and Southern blot analysis, see
ref. 9, 12. Miniprep Kit and Plasmid Maxi Kit for purifying
DNA are also commercially available (QIAGEN).
3.2. Transfection 1. Set up timed mating between a male mouse carrying a neomy-
of Embryonic Stem cin-resistant gene and several females.
Cells 2. Remove the uterus with the fetuses and put in a petri dish (this
3.2.1. Isolation can be done on your working bench). Aim to isolate about
of Embryonic Fibroblasts 5–10 of 13- to 16-day-old fetuses (see Note 5).
from Mice 3. Dissect out the fetuses in the clean hood and wash 5–7 times
with 50 mL PBS.
4. Cut off the heads. Remove and discard the inside organs from
thorax and abdomen.
5. Wash the tissue 5 times with PBS.
6. Cut up as finely as possible with sharp scissors.
7. Add 5 mL of trypsin solution (0.25% trypsin, 1 mM EDTA,
Gibco), incubate for 10 min at 37° with gentle shaking and
then pipet vigorously.
8. Add 30 mL of EF medium, allow the debris to settle, and
transfer the supernatant to a new tube.
9. Re-trypsinize the debris, collect the supernatant, and combine
the collected supernatants.
10. Spin down the cells for 5 min at 239 × g. Resuspend the pellet
in 100 mL of EF medium.
11. Plate onto three T175 flasks. This plating is considered to be
passage zero (P = 0).
12. Grow to confluent culture (1–4 days) and split into ten T175
flasks (P = 1). Add G418 (150 μg/mL) to select for neomycin
resistant clones and grow to confluent culture (2–3 days).
13. Split into 30 T175 flasks (P = 2) and continue the selection
process with G418. Continue growing to confluent culture.
14. Trypsinize the EF cells and freeze cells from one T175 flask per
tube (about 3 × 107 cells).
188 Y. Wang et al.
Fig.2. Targeted disruption of the D2L gene. (a) Targeting vector. For gene targeting of the
D2L locus, exon 6 (filled box, enlarged for visualization) was replaced by a PGK-neo cas-
sette, which was flanked by loxP (arrowhead). H, HindIII; K, KpnI; Sc, ScaI. (b) Southern
blot analysis of genomic DNA from mouse tails (−/−: homozygote; +/−: heterozygote;
+/+: wild-type). Tail DNA was either digested with KpnI and hybridized with the external
probe A or digested with ScaI and hybridized with the external probe B. (c) Detection of
two forms of D2 mRNA by RT-PCR. PCR products, differing by 87 bp and corresponding to
the alternatively spliced isoforms of D2 mRNAs, were shown in lanes 2–4 (+/+ mice:
235 bp—D2S and 322 bp—D2L) and in lanes 5–7 (−/− mice). The size marker (M) was
a 123 bp DNA ladder (lane 1). (d) Northern blot analysis of D2 RNA levels in striatum of
+/+ (lane 1) and −/− (lane 2) mice. The probe used was a DNA fragment complementary
to exon 7 of the D2 gene and thus hybridized to RNA from both WT and D2L−/− mice. Total
striatal RNA was obtained from 3 to 4 mice for each genotype. The data were quantified
with densitometric analysis using a scanner (UMAX Astra 2400S) and IQ Mae program.
Note that D2 mRNA from mutant mice was present at the predicted smaller size (similar
to D2S mRNA size). The size marker indicated by the arrows was Millennium RNA size
marker (Ambion). β-Actin RNA levels were simultaneously monitored as an internal con-
trol. (This figure is reproduced from ref. 5 with permission).
15. Grow a small amount of the cells to high confluent culture in

antibiotic-free medium and check for mycoplasma contamina-
tion using the mycoplasma detection kit from Gen-Probe, Inc.
Fresh medium can be used as a negative control.
3.2.2. Expanding 1. Day 0: thaw 1 tube of 3 × 107 EF cells (or feeders) at P = 2 and
Embryonic Fibroblasts plate onto three T175 flasks (use about 30 mL of EF medium
per flask).
2. Day 4: split the cells and plate onto ten T175 flasks. To split
the cells, wash the cells with PBS, add 2 mL of trypsin, 5 min
37°, pipet up and down, and resuspend in EF medium.
3. Day 7: inactivate the cells by mitomycin C (mit C) treatment
(see Note 6):
(a) Dissolve 1 vial of mit C (2 mg/vial, Sigma) in 10 mL of
PBS (20× solutions).
(b) Feed with EF medium containing 1× mit C for 2 h at
37°.
(c) Aspirate the medium and wash 3 times using PBS.
(d) Trypsinize the cells and freeze at 3 × 107 cells/mL
(10 tubes of 1 mL).
3.2.3. Plating of Embryonic 1. Treat the plates with a 0.1% gelatin solution for at least 20 min
Fibroblasts at room temperature, and aspirate the gelatin solution.
2. Thaw a frozen stock of mit C-treated EF cells at 37°, add
10 mL EF medium, spin, and resuspend. One vial of mit
C-treated EF (3 × 107 cells) should be resuspended in 30 mL of
EF medium.
3. Plate EF as follows: 500 μL per 1 well of 24-well plate, 200 μL
per 1 well of 96-well, 5 mL for a T25 flask, 30 mL for a T175
flask, and 9 mL for a 10 cm petri dish.
4. Wait at least 4 h to plate ES cells; do not wait longer than
3 days.
3.2.4. Plating of Embryonic 1. Prepare a flask with mit C-treated EF feeders the day before
Stem Cells (13) plating the ES cells. Feed the cells with ES medium + LIF just
before plating.
2. Thaw ES cells at 37°, add 10 mL of ES medium, spin, and
resuspend in ES medium.
About 3 × 106 cells should be plated onto a T25 flask. The
cells should reach 50% confluent after 2–3 days (see Note 7).
3. ES cells are split about 1:7 (e.g., a T25 flask can be split into a
T175 flask).
190 Y. Wang et al.
3.2.5. Electroporation 1. The day before electroporation, prepare ten 10 cm petri dishes
Conditions with EF.
2. Linearize approximately 50 μg of the targeting vector, ethanol
precipitated, wash 2 times with 70% ethanol, dry in the hood,
and further dry for 30 min at 65° in a heating block with the
tube cap closed.
3. Resuspend the DNA in 600 μL warm PBS.
4. Trypsinize about 5 × 107 ES cells (about 1 × 50% confluent
T175), resuspend in DNA solution to a total volume of
800 μL.
5. Add the mixture to a cuvette (0.4 cm gap, #165-2088, Bio-
Red Gene Pulser).
6. Electroporate at 3.0 μF, 800 V (Bio-Rad). These cells are con-
sidered to be at day 0.
7. Resuspend the electroporated mixture in 100 mL ES medium +
LIF and plate onto ten 10 cm plates with about 5 × 106 EF previ-
ously plated. Do not add G418 to the medium on day 0.
8. On the next day, begin the selection with the antibiotic. Add
150 μg G418/mL to five plates and 125 μg G418/mL to the
other five plates (see Note 8).
9. Keep changing the medium with ES medium + G418 every day
(see Note 9).
10. Start picking antibiotic-resistant colonies around day 7 or 8.
3.2.6. Picking 1. The day before picking the colonies, prepare several 96-well
of Colonies (14) flat-bottom plates with EF. Change the medium with ES
medium + LIF before picking.
2. Pick colonies under the microscope with an Eppendorf pipet-
man (keep the volume smaller than 5 μL) (see Note 10).
3. Transfer the clones to 96-well U-bottom plates. Add 15 μL
trypsin and incubate for 5 min at 37°.
4. Add 35 μL ES medium (use medium from the 96-well plates
with EF), pipet 10 times up and down and transfer to the
96-well flat-bottom plate with EF.
5. Change the medium every day.
6. Two to three days after picking, the ES cells should be ready to
split into two series of 24-well plates; one set of plates will be
used for freezing, the other will be used for DNA preparation
for Southern blot analysis. The plates to hold ES cells for freez-
ing should contain EF feeders. The plates for DNA prepara-
tion do not need feeders. All plates should be gelatinized.
7. Passage method: (a) Aspirate the medium, add 50 μL trypsin,
and incubate for 5 min at 37°; (b) Add 150 μL of ES medium,
pipet 10 times up and down; (c) Transfer 100 μL to the 24-well
plate without feeders and 100 μL to the 24-well plate with

feeders; (d) Change the medium every day. The cells for DNA
preparation can be fed with ES medium without LIF.
3.2.7. Freezing of Clones 1. Aspirate the medium from the 24-well plates.
(Method 1) (See Note 11) 2. Add 1 mL of ice-cold 1× freezing medium (90% FCS + 10%
DMSO, filtered (0.2 μm) and stored at 4°).
3. Seal the plate with Parafilm and put in a styrofoam box to store
at −70° (see Note 12).
3.2.8. Screening of ES Cell 1. For preparation of genomic DNA, aspirate the medium. Add
Clones with Homologous 500 μL of lysis buffer (100 mM Tris–HCl pH 8.5, 5 mM
Recombination EDTA, 0.2% SDS, 200 mM NaCl, 100 μg/mL Proteinase K);
(See Note 13) Shake overnight at 56°.
2. Put the plate on a swirling table for 15 min, add 1 volume of
isopropanol, and keep swirling until the DNA has aggregated.
This usually takes about 15 min.
3. Recover the DNA by lifting the aggregated precipitate from
the solution using an Eppendorf tip, remove excess liquid and
transfer to a pre-labeled microcentrifuge tube.
4. Spin briefly to collect the pellet at the bottom of the tube.
5. Dry for 5 min in the Speed Vac evaporator or for 1 h in a 37°
warm room.
6. Add 50 μL of TE-buffer or sterile H2O and dissolve overnight
at 56°.
7. Store at 4° or −20°.
8. To digest the DNA (see Note 14), prepare 10 μL DNA, 2.5 μL
10× restriction buffer, 2.5 μL 0.1 M 2-mercaptoethanol,
2.5 μL restriction enzyme, 7.5 μL mQH2O (total volume:
25 μL).
9. Mix well and digest overnight at 37°.
3.2.9. Southern Blotting 1. Take a picture of the stained gel. Cut out the region of the gel
where the bands are expected.
2. Shake the gel in 0.5 M NaOH + 1.5 M NaCl for 30 min.
3. Pre-wet the nylon filter (Zeta-Probe GT, Bio-Red) in water for
5 min.
4. Transfer the gel in 0.5 M NaOH + 1.5 M NaCl overnight.
5. Rinse the filter in 2× SSC.
6. Air dry the filter, put the gel between Whatmann 3MM paper
and bake at 80° under vacuum for 30 min to 1 h.
3.2.10. Hybridization 1. Hybridization mix: 1% SDS, 6× SSC, 10% dextran sulfate, and
at least 100 μg salmon sperm DNA/mL. Alternatively, you can
192 Y. Wang et al.
use commercially available hybridization solution (Clonetch,

Stratagene/Agilent Technologies).
2. Pre-hybridize at least 4 h at 65°, add radioactive probe and
hybridize overnight.
3. Wash the filter: 1 time for 15 min with 2× SSC + 1% SDS at
room temperature, and then 4 times for 15 min with 0.1×
SSC + 0.1% SDS at 65°.
4. Remove the liquid, wrap in plastic wrap, and expose to X-ray
film with DNA side up.
3.3. Injection of ES 1. On day 1 at 4:00 pm, intraperitoneally (IP) inject 4–10 female
Cells into Blastocysts donor mice with PMSG (5 IU/mouse).
for Production 2. On day 3 at 1:00 pm, inject (IP) the female donor mice with
of Chimeric Mice HCG (5 IU/mouse), and immediately mate the donor mice
3.3.1. Breeding of Donor
with intact male breeders (one to one).
Mice and Harvesting 3. On the morning of day 4, check females for semen plug, sepa-
of Blastocysts rate plugged females into a new cage from non-plugged females
and hold them until day 7.
4. On day 7, collect blastocysts from the female donor mice.
Remove the uterus, cut the distal ends, and flush out blastocysts
by expelling ES medium from a pipette into a dish. Shake the
dish gently on a shaker for a few minutes. Under a microscope,
collect the blastocysts and transfer to a new petri dish contain-
ing 50 μL of ES medium underneath the light mineral oil.
3.3.2. Breeding of Foster 1. In the evening of day 4, breed 6 foster mothers (see Note 15)
Mothers with vasectomized male mice.
2. On the morning of day 5, check females for semen plug, sepa-
rate plugged females into a new cage from non-plugged
females, and hold them until day 7.
3.3.3. Thawing of Targeted 1. Plate EF feeders in gelatinized 24-well plates.

ES Clone (See Note 16) 2. For Method 1 of freezing cells, thaw the targeted ES cells in
the following way:
(a) Take the plate out of the −70° freezer and put in the hood.
(b) Add pre-warmed medium to the well and pipet up and
down with a Pasteur pipette until everything is thawed.
Make sure all the cells are detached from the bottom of
the well.
(c) Suspend cells in 10 mL of medium, spin, resuspend in ES
medium, and transfer to two 24-wells.
3. Change the medium every day and grow to 50% confluent
culture. The time to confluence usually takes 2–5 days, depend-
ing on the clone.
4. Freeze the cells from 1 well. Split the cells from the other well
to 2 wells on a 12-well plate and grow to 50% confluent.
5. Freeze 1 well, split the other to 2 wells on a 6-well plate. Use
also some of the cells to startup 1 or 2 wells on a 24-well plate;
these cells will be used to prepare DNA to verify the clone.
6. Freeze 1 well, and split the other well into two T25 flasks.
7. Freeze cells from the two flasks in ten tubes.
8. Prepare DNA from the clone and verify by Southern blotting.
9. Select correctly targeted ES clones for injection that (a) have a
close to 1:1 ratio of wild-type: knockout band on the Southern
blot, (b) look healthy and undifferentiated, and (c) grow fast.
3.3.4. Blastocysts Injection 1. Thaw one tube of frozen ES cells 2–3 days before injection and
(13, 15) plate onto 3 wells of a 12-well plate containing ES medium.
The plate should be gelatinized but without EF.
2. In the afternoon of the injection after blastocysts have been
collected, trypsinize ES cells, gently manually dissociate, spin
down, resuspend in fresh ES medium, and maintain them as a
single cell suspension on ice.
3. Pre-cool the injection chamber on the microscope stage.
4. Use a transfer pipette to transfer the expanded blastocysts in
groups of about ten into the pre-cooled injection chamber.
Then, introduce a few hundred ES cells (a “cloud” of cells,
filling, at most, half the drop) into the injection chamber and
allow them to settle on the bottom.
5. Using high-power magnification, select individual ES cells
carefully on the basis of size (small, compared to the feeder
cells) and shape (uniformly round, compared with more ragged
or “rough” feeder cells). Draw about a hundred ES cells into
the injection pipette and position them near the tip in a mini-
mal amount of medium.
6. Hold a single blastocyst by applying suction to the holding
pipette and move it toward the center of the microscope field.
Position the embryo with the inner cell mass (ICM) at either
the 6 or 12 o’clock position.
7. Adjust the focus precisely at the equator of the blastocyst. Align
the tip of the injection pipette in the same focal plane as the
equator of the blastocyst.
8. With a continuous movement, introduce the loaded injection
pipette into the blastocyst cavity. Aim to insert the injection
needle at a junction between two trophoblast cells. When a
half of the needle of the injection pipette enters the blastocyst
cavity, slowly expel medium into the cavity so as to expand the
blastocyst and insert the pipette further into the cavity.
194 Y. Wang et al.
Take care not to touch the ICM with the injection needle.
If the first attempt to penetrate the trophoblast layer is unsuc-
cessful, and the blastocyst is not collapsed, try to insert the
needle exactly at the same position again. When the blastocyst
starts collapsing, expel a small amount of medium slowly to
keep the blastocyst cavity expanded.
9. Slowly expel 8–15 cells into the blastocyst cavity. Take care not
to insert any oil bubbles or lysed cells into the blastocyst.
10. Withdraw the injection pipette slowly. If the pressure is high
inside the blastocyst cavity, keep a half of the tapered needle of
the injection pipette inside the blastocyst to reduce the pres-
sure. This is to prevent injected cells from being pushed out
while the needle is being withdrawn. The blastocyst will col-
lapse once the pipette is removed, resulting in the cells coming
into close contact with the surface of the ICM.
11. Remove the injected blastocysts periodically and incubate them
in a humidified incubator with 5% CO2 at 37° in microdrops of
M16 or KSOM-AA medium (or ES medium). The blastocysts
will re-expand after 1–3 h in culture (see Note 17).
3.3.5. Surgery to Return 1. Anesthetize the foster mother mouse with pentobarbital
Injected Blastocysts sodium injected (IP) at a dose of 50–75 mg/kg.
to Foster Mother (13) 2. When the foster mouse is fully anesthetized with no reflex
response to pinching of paws, spray the lower back with 70%
ethanol.
3. Make a horizontal incision in the flank approximately 0.5 cm
away from the midline and between the natural hump of the
back and the point where the rear leg joins the abdomen.
4. Remove hair by wiping the area again with ethanol.
5. Locate under the skin, the ovarian fat pad and gently nick the
body wall (the peritoneum) overlying the pad.
6. Pull out the fat pad with the forceps and the uterus attached to
it. Attach the serafin clips to the fat pad and lay the clips down
across the animal’s body to stabilize the tissue mass.
7. Locate the uterine horns.
8. Take up 8–10 blastocysts and small air-bubbles as a marker of
where the difficult-to-see blastocysts are in the pipette.
9. Insert a 26 G needle through the muscle layers of the uterus
2–3 mm distant from the utero-tubal junction. Remove the
needle and keep your eye on the resulting hole in the uterine
wall.
10. Insert the tip of the pipette into the lumen of the uterus
through the hole. Transfer blastocysts and air-bubbles into the
uterine.
11. After removing the pipette, expel residual solution from the
pipette and check under the microscope to ensure no blasto-
cysts remained in the pipette.
12. Return the uterus and fat pad to the body cavity.
13. Repeat step 3–12 on the other side.
14. Close the wound with surgical clips.
15. Put the mouse on heating pad until it wakes up, and then
return it to its home cage.
3.4. Breeding 1. Pups developed from chimeric fertilized eggs will have a mosaic
of Chimeric Mice coat color pattern. Mice containing a greater percentage of the
for Determining ES cell strain color (i.e., agouti in the case of J1 ES cell A/A
Germline Contribution C/C) are the more useful progenies, since in these mice there
is a greater likelihood that their germ cells differentiated from
the targeted ES cells. In addition, since the majority of the
established ES cell lines are of the XY karyotype, the gender
ratio of the chimeras is usually biased towards males.
2. Mate male chimeras with mice of the strain identical to that of
the host blastocysts (i.e., C57BL/6 females). Offspring of this
breeding that have the color of the blastocyst host strain (i.e.,
non-agouti black color a/a C/C) do not carry the transgene.
Pups with a homogenous hybrid color, such as agouti (A/a
C/C in the case of J1 ES cell as a donor ES cell and C57BL/6
as a female strain) represent germline contribution of ES cells.
Since one of two allele of the relevant gene is targeted in the
ES cell, a half of ES cell-derived progeny has potential ger-
mline transmission.
3. Identify mice with germline transmission (i.e., heterozygous
mice) by Southern hybridization (or blot) analysis as described
in the next section.
4. Mate heterozygous males and females to yield homozygous
offspring, and verify the pups genotype by Southern blot
analysis (Fig. 2b.).
3.5. Genotyping to 1. Cut 3–4 mm of tail and drop it into a pre-labeled 1.5 mL
Identify Mice with microcentrifuge tube.
Germline Transmission 2. Add 500 μL of lysis buffer with 2.5 μL of proteinase K into
of the Targeted Allele each tube containing the tail.
3.5.1. Extract Genomic 3. Incubate the tubes in the oven (54–56°) overnight. If possible,
DNA from a Tail Tip put tubes in a rotation rack (see Note 18) or in a foam box on
a shaker to facilitate the lysis process (see Note 19).
4. Put the tubes in a microcentrifuge and spin at 16,500 × g for
10 min. Upon completion, a pellet should be firmly aggre-
gated at the bottom of the tubes. The DNA is in the
supernatant.
196 Y. Wang et al.
5. Transfer by pouring the supernatant (about 450 μL) to a new

1.5 mL tube without disturbing the pellet.
6. Add equal volume of isopropanol (about 450 μL) to each tube
containing the supernatant. Invert the tubes 10 or more times
until the DNA aggregates.
7. Spin the tubes in a microcentrifuge for 5 min to collect the
pellet (DNA) at the bottom of the tubes. Discard the superna-
tant by pouring without disturbing the pellet.
8. Add 1 mL of 70% ethanol to each tube containing the DNA
and rinse briefly.
9. Spin the tubes in a microcentrifuge for 3–5 min to collect the
pellet at the bottom of the tubes, and discard the ethanol by
pouring without disturbing the pellet. Spin the tubes briefly
again, and put the tubes with the lid open on Kim-Wipes
(Kimberly-Clark Co.) in an inverted position to drain the
residual ethanol.
10. Air dry the DNA for 5–10 min at room temperature.
11. Add 50 μL of TE buffer (pH 8.0), and put the tubes on a
rotating rack or shaker in the oven (54–56°) overnight to dis-
solve the DNA (see Note 20).
12. After the DNA is dissolved, store it at 4°.
3.5.2. Southern Blot 1. Heterozygous (+/−), homozygous (−/−), and wild-type (+/+)
Analysis to Identify Pups mice should be determined by Southern hybridization analysis
with the Targeted Allele using appropriate DNA probes made during the construction
of the targeting vector (Fig. 2b).
2. For Southern Hybridization method, see previous descriptions
(Subheading 3.2).
3.5.3. RT-PCR and 1. To verify the D2L mRNA is deficient and determine the expres-
Northern Blot Analysis sion level of D2S mRNA, perform reverse transcriptase-PCR
(RT-PCR) (Fig. 2c). Isolate total RNA from the striatum of
D2LR KO or WT littermates (TRIzol Reagent, Molecular Res
Center/ Invitrogen). Reverse-transcribed it into cDNA using
oligonucleotide primers complementary to exon 8 of the D2
gene (Superscript III, Invitrogen). Use the resulting cDNA in
a PCR reaction with oligonucleotide primers complementary
to exon 5 (5¢-(d GTGTATCATTGCCAACCCTGCC)) and
exon 7 (5¢-(d TGGTGCTTGACAGCATCTCC)) on the D2
gene. Conditions for the PCR should be 94° for 30 s, 60° for
1 min, and 72° for 3.5 min for 35 cycles.
2. Perform Northern blot analysis to quantify the expression level
of D2S mRNA (Fig. 2d).
3. If you obtain the transgenic mice made by others, no Southern
analysis, RT-PCR and Northern blot analyses are necessary.
4. Receptor binding assays, receptor autoradiography, and

Western blot analysis (if the specific antibody is available) can
determine the expression level of receptor protein (5).
3.5.4. PCR Analysis to 1. Make a master mix of the following composition on ice:
Identify Homozygous, 28.75 μL mQ H2O, 5 μL 10× polymerase buffer, 4 μL dNTP
Heterozygous, and (2.5 mM), 2 μL Primer mix (50 μM), 8 μL MgCl2 (25 mM),
Wild-Type Mice 0.25 μL DNA polymerase Stoffel (see Notes 21 and 22). Add
48 μL of the master mix to each PCR tube. Add 2 μL DNA
(tail) to each tube and pipette up and down. Add one drop of
mineral oil to each tube (optional).
2. PCR setting (see Note 23): step 1: 96° for 20 s, step 2: 60° for
1 min, step 3: 72° for 2 min. Repeat above cycle 35 times.
After finishing, keep at 4° (see Note 24).
3. Identify the PCR products by running it on a 1.5% agarose gel.
4. Notes
1. Always use Milli-Q H2O (mQ H2O) or water equivalent in

purity. Most buffers need to be autoclaved. For buffers that
cannot be autoclaved (e.g., 10% SDS, ethidium bromide,
Denhardts solution), use autoclaved mQ H2O or equivalent.
Sometimes it is more appropriate to sterilize the solution with
a filter (0.2 μm).
2. Genomic DNA fragments are also available from various com-
panies (e.g., BACPAC Resource Center (BPRC), the Children’s
Hospital Oakland Research Institute in Oakland, California,
USA).
3. If Ensembl is used to predict the restriction enzyme sites,
always confirm with your own mapping of the restriction sites.
This is especially important if your gene sequence was obtained
from a strain different from C57BL/6J, which is the genome
represented on Ensembl (16).
4. The efficiency of homologous recombination depends on sev-
eral factors. The first is similarity between the construct DNA
and its corresponding endogenous counter (17) that can be
best achieved using DNA for the construct that comes from
the same mouse strain as the ES cells. The second is the length
of the homologous arms (18) which generally should be sev-
eral kilobases long on each end. Longer homologous
sequences will result in recombination at high frequencies.
The third is the sequence itself. Hotspots, defined as specific
sequences that promote recombinase interaction with the
nucleotide to induce homologous recombination, have also
198 Y. Wang et al.
been identified in some genes (19). These hotspots are

difficult to predict and so as a result difficult to avoid. Keep in
mind that these exist if you find that the antibiotic gene is
present, but segments of your gene after homologous recom-
bination is unexpectedly missing.
5. Fourteen-day-old fetuses are the best.
6. The EF cells can also be inactivated by irradiation with the
following process: (a) Trypsinize the cell, wash, and resuspend
in medium; (b) Irradiate at 6,000 rads for 1 h; (c) Spin down
and freeze at 3 × 107 cells/mL (10 tubes of 1 mL).
7. Technique must be absolutely meticulous while working with
ES cells to prevent contamination with any substances that
could trigger the differentiation of the pluripotent cells in cul-
ture. Some investigators try to use the exact same lot of fetal
calf serum that were previously used with success to generate a
gene-targeted mouse line.
8. The optimal amount of G418 might vary for each cell line.
9. On day 2, small numbers of cell death can be seen. On day 3
many cells should start dying. By day 4 there should be a lot of
cell death.
10. When picking antibiotic-resistant colonies, be conscientious to
pick only well-shaped and undifferentiated colonies, and avoid
the flat, differentiated colonies. Healthy ES colonies that have
not differentiated have relatively round and well-defined edges
and grow in three dimensions. Differentiated colonies usually
have an irregular shape and look flat, often differentiated cells
(muscle or fibroblast-like cells) can be seen. Even in a normal
culture some colonies are differentiated.
11. Method 2 for freezing ES clones: (a) Aspirate the medium
from the 24-well plates; (b) Add 150 μL of trypsin and
incubate for 5 min at 37°; (c) Add 100 μL of ES medium,
pipet up and down (10×) and transfer to chilled pre-labeled
freezing tubes; (d) Add 200 μL of ice-cold 2× freezing
medium (80% FCS + 20% DMSO, filter sterilized (0.2 μm)
and stored at 4°; (e) Close the tubes, shake to mix, and
freeze slowly at −70°; (f) Transfer tubes to liquid nitrogen
the next day.
12. The plates can be kept at −70° for several months.
13. Aim to have at least 5 or 6 colonies of ES cells that have the
gene targeted as expected.
14. For most enzymes, a universal buffer can be used. The compo-
sition of 10× concentrated universal buffer is as follows:
200 mM Tris–HCl (pH 7.5), 70 mM MgCl2, 500 mM KCl,
10 mM 2-me (0.875 μL 2-mercaptoethanol/mL), 0.5 mM
bovine serum albumin/mL. For some enzyme (EcoRI),

additional salt (1 M NaCl for 10× solution) can be added.
15. Use foster mothers that are experienced in supporting their
litter in utero and ex utero. Overweight mice should be
avoided because they tend to have large fat pads that contain
blood vessels that may be easily nicked, thus making the sur-
gery site bloody, which can block your view during the trans-
fer of the blastocysts.
16. For Method 2 of freezing (see Note 11): thaw the tube at
37°. Suspend cells in 10 mL of medium, spin down to collect
the cells, and resuspend in ES medium. Transfer the cells to
two 24 wells.
17. To maximize your chances to obtain a germline transmitted
pup, aim to devote at least two separate cycles of injection and
embryo transfer for each colony of gene-targeted ES cells.
18. If using a rotating rack, wrap the tube with Parafilm to prevent
the solution leak out.
19. If there is no rotating rack or shaker available, you can put
tubes in a foam box in oven overnight. Vortex the tubes at least
twice during the incubation period to help break down the tis-
sue to release the DNA. The tail tissue should be completely
dissolved with only the bones left.
20. Alternatively, you can put tubes in a foam box in the oven to
warm up for at least 30 min, and then pipet up and down the
DNA using a yellow pipette tip to disperse the pellet. Then,
put tubes back in the oven for overnight. If necessary, pipet
up and down to disperse the DNA again at a later time.
21. When preparing the master mix, it is better to prepare one
extra sample for every 10–20 samples because pipetting can
have error.
22. Primer mix: 4 oligonucleotide primers are used—two with
sequence complementary to the D2 gene (around exon 6 regions)
and two with sequence complementary to the neo gene.
23. PCR conditions may vary, depending on the gene. For exam-
ple, Dr Sasaoka uses the following PCR setting: step 1: 94°
1 min, step 2: 94° 1 min, step 3: 60° 2 min, step 4: 72°
2 min, repeat steps 2–4 30 times, step 5: 72° 5 min, step 6:
hold at 4° (20).
24. The DNA in PCR tubes should be denatured as soon as pos-
sible; this can be achieved by starting the program, letting
the temperature rise to 96°, pausing the program, loading
the PCR samples, and then resuming the program. For some
genes, an extra period (20 s or longer) of denaturation at the
beginning is necessary.
200 Y. Wang et al.
Acknowledgments
We thank Dr. Luc Van Kaer (Vanderbilt University) for helping

establish the ES cell transfection protocol.
References
1. Dal Toso R, Sommer B, Ewert M, Herb A, 11. Adra CN, Boer PH, McBurney MW (1987)
Pritchett DB, Bach A, Shivers BD, Seeburg Cloning and expression of the mouse pgk-1
PH (1989) The dopamine D2 receptor: two gene and the nucleotide sequence of its
molecular forms generated by alternative splic- promoter. Gene 60:65–74
ing. EMBO J 8:4025–4034 12. Ausubel F, Brent R, Kingston RE, Moore
2. Giros B, Sokoloff P, Martres MP, Riou JF, DD, Seidman JG, Smith JA, Struhl K (1995)
Emorine LJ, Schwartz JC (1989) Alternative Short protocols in molecular biology, 3rd
splicing directs the expression of two D2 dop- edn. Wiley, New York
amine receptor isoforms. Nature 342:923–926 13. Nagy A, Gertsenstein M, Vintersten K,
3. Monsma FJ Jr, McVittie LD, Gerfen CR, Behringer R (2003) Manipulating the mouse
Mahan LC, Sibley DR (1989) Multiple D2 embryo: a laboratory manual, 3rd edn. Cold
dopamine receptors produced by alternative Spring Harbor Laboratory Press, Cold
RNA splicing. Nature 342:926–929 Spring Harbor
4. Chio CL, Hess GF, Graham RS, Huff RM 14. Matise M, Auerbach W, Joyner AL (2000)
(1990) A second molecular form of D2 dop- Production of targeted embryonic stem cell
amine receptor in rat and bovine caudate clones. In: Joyner A (ed) Gene targeting: a
nucleus. Nature 343:266–269 practical approach, 2nd edn. Oxford University
5. Wang Y, Xu R, Sasaoka T, Tonegawa S, Kung Press, Oxford
M-P, Sankoorikal E-B (2000) Dopamine D2 15. Papaioannou V, Johnson R (2000) Production
long receptor-deficient mice display alterations of chimeras by blastocyst and morula injection
in striatum-dependent functions. J Neurosci of targeted ES cells. In: Joyner A (ed) Gene
20:8305–8314 targeting: a practice approach, 2nd edn.
6. Usiello A, Baik J-H, Rouge-Pont F, Picetti R, Oxford University Press, Oxford
Dierich A, LeMeur M, Piazza PV, Borrelli E 16. Dang MT, Yokoi F, Yin HH, Lovinger DM,
(2000) Distinct functions of the two isoforms Wang Y, Li Y (2006) Disrupted motor learn-
of dopamine D2 receptors. Nature 408: ing and long-term synaptic plasticity in mice
199–203 lacking NMDAR1 in the striatum. Proc Natl
7. Smith JW, Fetsko LA, Xu R, Wang Y (2002) Acad Sci USA 103:15254–15259
Dopamine D2L receptor knockout mice dis- 17. te Riele H, Maandag ER, Berns A (1992)
play deficits in positive and negative reinforc- Highly efficient gene targeting in embryonic
ing properties of morphine and in avoidance stem cells through homologous recombina-
learning. Neuroscience 113:755–765 tion with isogenic DNA constructs. Proc Natl
8. Xu R, Hranilovic D, Fetsko LA, Bucan M, Acad Sci USA 89:5128–5132
Wang Y (2002) Dopamine D2S and D2L 18. Thomas KR, Deng C, Capecchi MR (1992)
receptors may differentially contribute to the High-fidelity gene targeting in embryonic
actions of antipsychotic and psychotic agents stem cells by using sequence replacement vec-
in mice. Mol Psychiatry 7:1075–1082 tors. Mol Cell Biol 12:2919–2923
9. Sambrook J, Fritsch EF, Maniatis T (1989) 19. Smith GR (1994) Hotspots of homologous
Molecular cloning: a laboratory manual, 2nd recombination. Experientia 50:234–241
edn. Cold Spring Harbor Laboratory Press, 20. Sasaoka T, Imamura M, Araishi K, Noguchi S,
Cold Spring Harbor, NY Mizuno Y, Takagoshi N, Hama H,
10. Mombaerts P, Clarke AR, Hooper ML, Wakabayashi-Takai E, Yoshimoto-Matsuda Y,
Tonegawa S (1991) Creation of a large Nonaka I, Kaneko K, Yoshida M, Ozawa E
genomic deletion at the T-cell antigen receptor (2003) Pathological analysis of muscle hyper-
beta-subunit locus in mouse embryonic stem trophy and degeneration in muscular dystro-
cells by gene targeting. Proc Natl Acad Sci phy in gamma-sarcoglycan-deficient mice.
USA 88:3084–3087 Neuromuscul Disord 13:193–206
Chapter 12
Genomic Strategies for the Identification of Dopamine

Receptor Genes in Zebrafish
Wendy Boehmler, Jessica Petko, Victor A. Canfield, and Robert Levenson
Abstract
In this chapter, we describe the identification and cloning of D2-like dopamine receptor (DR) genes in
zebrafish, a vertebrate model genetic organism. To identify DR genes, we performed searches of the
zebrafish genomic sequence database that yielded contig segments of several D2-like DR genes. From
these sequences, we amplified full-length cDNAs encoding three D2, one D3, and three D4 DR receptor
subtypes via RT-PCR. The predicted proteins displayed 57–72% amino acid identity when compared to
their human DR counterparts. To validate the identity of zebrafish DR genes, each of the genes was
mapped by using the T51 radiation hybrid panel. With the exception of drd2b and drd4b, each of the
zebrafish DR genes mapped to chromosomal positions that were syntenic with regions of human chromo-
somes containing orthologs of the zebrafish DR genes. To further validate the identity of the D2-like DR
genes in zebrafish, we conducted phylogenetic analysis which supported the predicted identities of the
cloned DR receptor cDNAs.
Key words: Cloning, D2-like DR, RT-PCR, Phylogenetic analysis, Contig, Dopamine
1. Introduction
Dopamine is a catecholamine neurotransmitter that in mammals

controls a wide variety of functions including locomotion and
learned behavior. Dysfunction in dopaminergic signaling is believed
to underlie several neuropathologies including schizophrenia (SZ),
Tourette’s syndrome, and Parkinson’s disease (PD). It has been
postulated that manifestation of SZ involves dopamine hyperactiv-
ity in the brain (1) and this hyperactivity may be due to the dys-
regulation of dopamine receptor signaling. Much of what is known
about SZ has been driven by the development of antipyschotic
drugs that antagonize D2-like DRs. Even though such antipsy-
chotic drugs effectively treat the positive symptoms of SZ, several
201
202 W. Boehmler et al.
classes can cause discomforting side effects such as unwanted

sedation, movement disorders including tardive dyskinesia, and
pseudopregnancy. In contrast to SZ, PD is characterized by a
decrease in dopaminergic signaling due to the degeneration of
dopaminergic neurons in the substantia nigra (2). The lack of dop-
amine signaling in PD patients results in symptoms that include
tremors, rigidity, and bradykinesia.
To better understand DR function, we initiated studies
designed to characterize D2-like DR genes in zebrafish. Zebrafish
is an attractive model genetic system for the study of many biologi-
cal processes. The zebrafish model system provides several advan-
tages for identifying and characterizing genes and their function.
The zebrafish embryo develops ex utero and is transparent, which
allows for gene expression analysis using whole-mount in situ
hybridization. Zebrafish are amenable to forward and reverse
genetic techniques. Mutant zebrafish can be screened morphologi-
cally and behaviorally to identify new genes and their function.
Antisense morpholino oligonucleotides make it possible to knock
down expression of a specified gene during the first several days of
development in order to determine its function. Zebrafish display
a cohort of behaviors related to DA signaling including locomo-
tion and conditioned place preference. Additionally, chemicals and
drugs can be added directly to tank water, thus making zebrafish
an attractive model system for toxicology and drug screening.
We have identified and cloned the complete cohort of D2-like
DR genes in zebrafish. Mining the zebrafish genomic sequence
database allowed us to identify contigs containing segments of sev-
eral D2-like DR genes. From these sequences, we were able to
amplify full-length cDNAs encoding three D2, one D3, and three
D4 DR subtypes. When these studies were originally performed,
extensive searches of the expressed sequence tags (EST) database
retrieved no results for zebrafish DRs. In retrospect, this probably
reflected low expression of the receptors in zebrafish. In addition,
the first zebrafish genome assemblies had extensive gaps that made
it difficult to identify complete exons of the DR genes. Subsequent
releases of preliminary genomic assemblies and tools for performing
BLAST searches of the zebrafish genome (http://www.ensembl.
org/Danio_rerio/blastview) have greatly facilitated identification
of additional exons and assignment of exons to genes.
2. Materials
1. TRIzol® Reagent (Invitrogen, Life Technologies, Grand Island,

NY).
2. SuperScript® II Reverse Transcriptase (Invitrogen).
12 Dopamine Receptors in Zebrafish 203
3. REDTaq® DNA Polymerase (Sigma-Aldrich, St. Louis, MO).

4. TOPO® TA Cloning® Kit (Invitrogen).
5. Zebrafish are sold by various vendors (e.g., SoBran, Inc.,
Fairfax, VA and Charles River, Wilmington, MA).
3. Methods
3.1. Cloning The general strategy we used to identify, clone, and validate
and Characterization zebrafish D2-like DR genes is outlined in Fig. 1. To identify
of Zebrafish DR Genes zebrafish DR genes, we first searched the available EST database.
Initial BLAST (Basic Local Alignment Search Tool) queries of the
EST database using human DR nucleotide sequences failed to
identify any zebrafish cDNAs with sequence homology to human
Fig. 1. Schematic representation of the strategy used to identify, clone, and validate
zebrafish D2-like DR genes.
DR genes. The Wellcome Trust Sanger Institute began sequencing

the zebrafish genome in early 2001. Segments of zebrafish dopamine
receptor genes were identified by using mammalian and fish poly-
peptide sequences as probes in low stringency SSAHA (3) and
BLAST searches of the zebrafish genomic reads generated by the
Sanger Center (http://www.sanger.ac.uk/Software/analysis/
SSAHA/). One of the polypeptide sequences used was the human
D2 DR (DRD2; NM_000795). Specifically, this sequence was
used in a TBLASTN search, which will search a translated nucle-
otide database using an amino acid sequence. The results of these
searches allowed for the identification of portions of exons repre-
senting a minimum of three zebrafish D2-like, one D3-like, and
three D4-like DR genes. All genomic sequence reads encompassed
entire zebrafish DR genes except for D2b and D2c, which were
missing their 5¢ and 3¢ ends, respectively. Subsequent releases of
preliminary genomic assemblies and enhanced tools for perform-
ing BLAST searches (http://www.ensembl.org/Danio_rerio/
blastview) allowed for the identification of those additional exons.
Polymerase chain reaction (PCR) primers that overlapped
conserved initiation and termination codons were designed based
on our genomic sequence reads (see Note 1). The corresponding
full-length cDNAs were generated by RT (reverse transcriptase)-
PCR. First-strand cDNA synthesis was performed using
SuperScript RT (Life Technologies) according to the manufac-
turer’s protocols (see Note 2). Random hexamers were used to
prime cDNA synthesis and total RNA from adult zebrafish was
used as template. REDTaq DNA Polymerase (Sigma) was used
for subsequent amplification. For each cDNA, it was necessary to
use nested primers for PCR because the first round of PCR invari-
ably failed to amplify the transcripts. We believe this may be due
to low expression of DR transcripts in whole adult zebrafish.
Nested PCR is a technique that employs two primer pairs and
two sequential PCR runs to amplify a single target sequence.
Primers and annealing temperatures for the D2-like DR are listed
in Table 1. PCR was carried out using a TD-7500 Thermal Cycler
(Hybaid). An initial 4 min denaturation step at 94°C was fol-
lowed by 34 cycles at 94°C for 30 s, 45–52°C for 30 s, and 72°C
for 90 s. A final elongation step was carried out at 72°C for
10 min. All cDNAs were TOPO-cloned and then sequenced
using an ABI 377-automated DNA sequencer. Sequences were
verified by sequencing at least three independent clones and com-
parison to genomic sequences.
3.2. Validation We utilized multiple parameters to confirm the identities of the

of DR Gene Identity predicted DR polypeptides. These included BLAST analysis,
comparison of gene structure (i.e., intron/exon organization),
phylogeny, and conserved synteny.
Table 1
PCR primers for the amplification of Zebrafish D2-like dopamine receptors
Clone Forward primer Reverse primer Ta (°C)

drd2a
−34
PCR 1 ggttctagggtctcagcttg−15 1772
caccccatacacagtaatgttatg1749 49
−3
PCR 2 ctgatggaagtcttcacagcg18 1772
caccccatacacagtaatgttatg1749 50
drd2b
−15
PCR 1 cagaggatcttcatcatgcct6 1320
gtgtgttcaacagtgcaggatcttgat1294 50
−7
PCR 2 cttcatcatgcctgtcctgaac15 1320
gtgtgttcaacagtgcaggatcttgat1294 52
drd2c
−44
PCR 1 gaattatgtctctcagtttcaggcttc−18 1363
atcctcagcagtgcaatatc1344 50
−11
PCR 2 ggccacagctcatggatttc9 1363
atcctcagcagtgcaatatc1344 52
drd3
−75
PCR 1 gttacactgcatgttgtcaag−55 1379
atatagcccatggtttagca1360 45
−35
PCR 2 aaaatctgtccaccctctcc−16 1362
gcagctcaggattttaatga1343 47
drd4a
1
PCR 1 atggtagaggcagacatgcca21 1157
ttagcatgctcaggctagcag1137 54
1
PCR 2 atggtagaggcagacatgcca21 1143
ctagcagcagccaggcagcgt1123 52
drd4b
−5
PCR 1 gcatcatggtcaatgtgacgcc17 1262
caacctcaggaacgacagcagag1240 54
1
PCR 2 atggtcaatgtgacgcccagt21 1257
tcaggaacgacagcagagaag1237 52
drd4c
−90
PCR 1 gcagagatggaccacagtggaca−68 1184
aatcagtacagtcttcagcatc1163 52
−15
PCR 2 ggatcaagaaggacaatgtctgc8 1160
tatggtggatgtcagcagcagc1139 52
Nucleotide +1 si the A of the ATG codon for the initiating methionine
3.2.1. BLAST Analysis BLAST analysis indicates that each of the cDNAs we cloned
encodes a polypeptide with a high degree of sequence similarity to
mammalian dopamine receptors. Sequence comparisons indicate
that the zebrafish polypeptides show 57–71% amino acid sequence
identity to human D2, D3, and D4 receptors (Table 2). One
zebrafish clone (drd3) shows highest identity (67%) to the human
D3 DR, while the D2 zebrafish clones (drd2a, drd2b, and drd2c)
share highest identity (66–71%) with the human D2 DR (Table 2).
In contrast, drd3 exhibits 52% identity to zebrafish D2 DRs, simi-
lar to the 54–62% identity between mammalian D2 and D3 DRs.
Sequence comparisons indicate that the zebrafish D4 DR polypep-
tides show 57–59% amino acid identity to mammalian D4 DRs.
In contrast, the zebrafish D4 DRs share very low sequence identity
with human D2 and D3 DRs (35–40%) (Table 2). Over the years,
continued mining of the zebrafish genomic and EST databases has
Table 2
Pairwise comparisons between Zebrafish and human D2-like
receptors
Zebrafish D2a D2b D2c D3 D4a D4b D4c

Human D2 71 66 71 52 38 38 37
Human D3 58 58 59 67 37 39 39
Human D4 37 40 35 39 59 57 58
Numbers represent percent amino acid identity
failed to uncover additional D2, D3, or D4 dopamine receptor

genes suggesting that zebrafish likely possess three D2 DR genes,
a single D3 DR gene, and three D4 DR genes (see Note 3).
3.2.2. Structural Sequence comparisons between the human, rat, and zebrafish D2, D3,
Comparisons Between and D4 DRs have been previously described (4, 5). By aligning the
Zebrafish and Human DRs predicted zebrafish and human DR polypeptides, we identified
seven putative transmembrane (TM) domains that are highly con-
served with the TM segments of the human D2-like DRs. In addi-
tion to the TM segments, several other regions are highly conserved
between mammalian and zebrafish DRs. For D2 DRs, these regions
include the first and second intracellular loops, three short seg-
ments within the third intracellular loop, and the C-terminal tail.
D3 receptors exhibit fewer highly conserved regions which include
the first (but not the second) intracellular loop and the C-terminus.
For D4 receptors, the majority of the conserved regions are the
TM segments. In addition, the intron/exon organization of the
zebrafish D2, D3, and D4 DR genes is virtually identical with
respect to their mammalian counterparts, suggesting that the
zebrafish and mammalian genes arose from a common ancestral
gene. It is interesting to note that many GPCRs lack introns, how-
ever the human and zebrafish D2 and D3 DR genes each contain
seven exons, while the human and zebrafish D4 DR genes each
contain 4 exons. Overall, the third intracellular loop of the D2-like
DRs is highly divergent between zebrafish and mammals, as well as
between each of the zebrafish DR subtypes. It is possible that this
sequence divergence may reflect functional differences between
the various zebrafish DRs.
3.2.3. Phylogenetic The procedures originally described for phylogenetic analysis (4, 5)
Analysis were repeated, combining D4 with D2/D3 sequences, and using
additional sequences not available at the time of the initial
analysis.
1. Using the human D2, D3, and D4 protein sequences as probes,

we performed BLAST searches (6) to identify dopamine receptor
sequences from additional species. These searches employed
the blastp program to search nonredundant peptide databases
in ncbi/GenBank (http://blast.ncbi.nlm.nih.gov/Blast.cgi).
Including the seven zebrafish clones and several additional
sequences previously collected, this procedure yielded 31
sequences from nonmammalian vertebrates. All sequences that
were not obviously incomplete were collected at this stage.
Using the same procedure, we also collected eight outgroup
sequences, comprising three invertebrate sequences and five
human sequences representing GPCRs highly similar to D2,
D3, and D4.
2. These 42 sequences were aligned using the GCG program
“pileup” (7). Alternatively, the freely available ClustalW pro-
gram (8) (http://www.clustal.org/) could have been used for
multiple sequence alignment. Nine sequences that were trun-
cated or contained large insertions or deletions in the align-
ment were discarded, and the alignment process was repeated.
3. The GCG program “pretty” was used to generate a consensus
sequence for the revised alignment. Aligned sequences were
inspected to locate regions in which alignments were ambigu-
ous. These regions were excluded from phylogenetic analysis
either by removing individual sequences that could not be
aligned (three cases) or by using only unambiguously aligned
regions. The final analysis contained 211 aligned amino acids
from 30 sequences (23 vertebrate D2, D3, or D4 sequences
plus 7 outgroup sequences).
4. The consensus sequence was edited to distinguish retained
from excluded amino acid positions. A simple perl filter was
then used to replace excluded amino acids with blanks and to
reformat the file for use by the Phylip (3.68) suite of programs
(9) (http://www.phylip.com). The equivalent changes could
have been performed by hand using a word processing pro-
gram. Note that ClustalW provides the option of writing
aligned sequences in a Phylip-compatible format.
5. Aligned sequences were analyzed using Maximum Parsimony
and Distance Matrix methods, using half-jacknife replications
to estimate reliability of the deduced phylogenies. The sequence
of programs used for the former was seqboot, protpars, then
consense; while for the latter, we used seqboot, protdist, fitch,
and consense.
6. Results of the phylogenetic analysis are shown in Fig. 2. D2,
D3, and D4 clusters were obtained in 81, 76, and 100% of
replicates, respectively (maximum parsimony) or in 94, 87, and
100%, respectively (using the distance method). Groupings
Fig. 2. Phylogenetic analysis of zebrafish D2, D3, and D4 DRs. Consensus trees were generated using maximum parsimony
(left) and distance methods (right ). Numbers represent percent of trees in which the clustering depicted was observed.
Included sequences, in addition to zebrafish: hud2, human dopamine D2, NP_000786; hud3, human D3, AAB08750; hud4,
human D4, AAB59386; parus, great tit (Parus major), AAY56686; chickd3 (Gallus gallus), ACR48171; turkeyd2 (Meleagris
gallopavo), AAD03818; xend21 (Xenopus laevis), NP_001095212; carpd4b (Cyprinus carpio), CAA74977; troutd2, rainbow
trout (Oncorhynchus mykiss), NP_001117843; tilpiad2, (Oreochromis niloticus), AAU87971; tetra_e, (Tetraodon nigro-
viridis), CAF97490; mulletd2 (Mugil cephalus), AAU87970; eeld2a (Anguilla anguilla), ABH06893; eeld2b (Anguilla anguilla),
ABH06894; d215 (Takifugu rubripes), CAA56456; d222 (Takifugu rubripes; (17)). Outgroup sequences (only first is shown
on tree): drod2 (Drosophila melanogaster), NP_001014760; Drosophila simulans, XP_002103025; human adrenergic
alpha2A, AAF91441; human serotonin 5HT4, NP_001035261; human dopamine D5, NP_000789; human adrenergic
alpha1D, NP_000669; human histamine H2, NP_071640. Excluded sequences: chicken (Gallus gallus), NP_001136321;
Taeniopygia guttata (zebra finch), XP_002196676; Xenopus laevis, P34973; Tetraodon nigroviridis, CAF98376, CAF95731,
CAG04235, CAF92229; minnow (Pimephales promelas), ABS30830; carp (Cyprinus carpio), CAA74976, CAB40081,
CAA74974; Branchiostoma floridae, XP_002211994.
within the D2 cluster were supported less strongly, with one

exception, and differed somewhat between the two methods.
Both methods identified zebrafish d2b as the most distant
member of the D2 cluster, while the remaining D2 sequences
clustered in 99% of trees with the distance method (80% with
maximum parisomony). Within the D3 group, the fish and
mammal/bird clusters were strongly supported. In the D4
cluster, zebrafish d4b was an outlier, although separation of
this sequence from the remaining sequences was supported by
only 57% (maximum parsimony) or 65% (distance method) of
trees. Outgroup sequences located the root of the tree between
the D4 and D2/D3 clusters, consistent with the differences in
intron/exon organization, while D2 and D3 clustered together
(100% of trees, both methods). The only exception among the
outgroup sequences was the Drosophila D2-like sequence,
which clustered with D4 sequences using the distance method
(71% support). The Drosophila sequence clustered with other
outgroup sequences using maximum parisomony (62% of
trees), while clustering with D4 sequences occurred in only
29% of trees obtained using this method.
3.2.4. Chromosomal Each zebrafish D2-like DR gene was localized to a zebrafish chro-
Mapping of Zebrafish mosome via radiation hybrid mapping. This type of comparative
Dopamine Receptor Genes mapping study allowed us to identify conserved synteny between
zebrafish and human DR genes. Zebrafish DR genes were mapped
using the Goodfellow T51 radiation hybrid (RH) panel (10). PCR
products specific for each zebrafish DR gene were amplified using
primers corresponding to unique sequences within each gene. PCR
reactions were performed in duplicate on the RH panel using con-
ditions optimized for each primer pair. PCR reaction products
were fractionated on 2% agarose gels, and each sample scored for
presence or absence of the zebrafish-specific amplicon. Linkage
assignments were computed using the Zon RH mapper resource
(http://zfrhmaps.tch.harvard.edu/ZonRHmapper/).
Our mapping data shows that the zebrafish drd3 gene is located
on chromosome 24, while the human D3 dopamine receptor
DRD3 gene maps to chromosome 3 (11). Zebrafish chromosome
24 contains orthologs of additional human genes located on human
chromosome 3 (Table 3). Comparative gene mapping thus pro-
vides strong additional evidence that the zebrafish drd3 DR gene
and the mammalian D3 DR gene are in fact orthologous. The
zebrafish drd2a and drd2c DR genes map to chromosomes 15 and
5, respectively, while the human D2 DR gene (DRD2) has been
localized to chromosome 11q23 (12). Both zebrafish chromosome
15 and chromosome 5 contain multiple orthologs of human genes
located on chromosome 11 (Table 3). The presence of lim1 and
lim6 (zebrafish duplicates of the single human LHX1 gene) on
zebrafish chromosome 15 and chromosome 5 (13) is consistent
Table 3
Markers Syntenic with dopamine receptor genes in Zebrafish
and human
Zebrafish gene Human ortholog
Name Accession no Map position Name Map position

drd2a AY183456 15 DRD2 11q23
acad8 BC044203 15 ACAD8 11
arcn1l BC050499 15 ARCN1 11
cryab AF159089 15 CRYAB 11q
ctsc AI436938 15 CTSC 11q14.1-q14.3
hsp47 U31079 15 CBP2 11q13.5
hsp70 AF006006 15 HSPA10 11
kiaa0102 AA494919 15 KIAA0102 11q13.3
mgc10485 BC046056 15 MGC10485 11q25
mpzl3 BC067698 15 MPZL3 11q23.3
mre11a BC053202 15 MRE11a 11q21
or13.1 AF012746 15 OR2AT4 11q13.4
sesn3 BC045518 15 SESN3 11q21
tyr NM_131013 15 TYR 11q14-q21
drd2b AY333791 16 DRD2 11q23
fli1b BC055627 16 FLI1 11q24.1-q24.3
drd2c AY333792 5 DRD2 11q23
apoa Y13653 5 APOA1 11q23-q24
atdc AI721600 5 ATDC 11q22-q23
htatip AI477057 5 HTATIP 11q12.1
ins AF036326 5 INS 11p15.5
mizf BC057479 5 MIZF 11q23.3
slc37a2 BC055147 5 SLC37A2 11q24.2
spon1a AB006086 5 SPON1 11p15.2
stip1 BC085642 5 STIP1 11q13
tmprss4 BC048135 5 TMPRSS4 11q23.3
wnt11 AF067429 5 WNT11 11q13.5
drd3 AY183455 24 DRD3 3q13.3
boc AF461120 24 BOC 3q13.2
c3orf3b BI887394 24 C3orf3 3p22.1
cldn11 AF359429 24 CLDN11 3q26.2-q26.3
cpb1 BC067637 24 CPB1 3q24
ek1 U89295 24 EPHB3 3q21-qter
nktr AI584240 24 NKTR 3p23-p21
ostalpha BC081597 24 OSTalpha 3q29
slc22a13 AW019519 24 SLC22A13 3p21.3
zic1 NM_130933 24 ZIC1 3q24
(continued)
Table 3
(continued)

drd4a AY750152 25 DRD4 11p15.5
c11orf24 AW154701 25 C11orf24 11q13
cat AJ007505 25 CAT 11p13
ckap5 NM_001037667 25 CKAP5 11p11.2
fancf CN014287 25 FANCF 11p15
irf7 NM_200677 25 IRF7 11p15.5
ldha AF067201 25 LDHA 11p15.4
mdkb BC059575 25 MDK 11p11.2
myod Z36945 25 MYOD1 11p15.4
nap1l4 NM_001089347 25 NAP1L4 11p15.5
pax6a AJ507427 25 PAX6 11p13
rag1 U71093 25 RAG1 11p13
rag2 U71094 25 RAG2 11p13
rnh CF997807 25 RNH 11p15.5
tnni2 BC071462 25 TNNI2 11p15.5
wt1 NM_131046 25 WT1 11p13
zdhhc13 BC086723 25 ZDHHC13 11p15.1
drd4b AY750152 4 DRD4 11p15.5
ms4a4a AW078034 4 MS4A4A 11q12
drd4c AY750154 7 DRD4 11p15.5
c11orf15 BC050159 7 C11orf15 11p15.3
ccnd1 NM_131025 7 CCND1 11q13
cd81 BC057419 7 CD81 11p15.5
copb1 AY294010 7 COPB 11p15.2
cugbp1 AB032726 7 CUGBP1 11p11
dbx1a AF030284 7 DBX1 11p15.1
dhcr7 BC055631 7 DHCR7 11q13.2-q13.5
eif4g2a BC059195 7 EIF4G2 11p15
ext2 AY786508 7 EXT2 11p12-p11
f2 AW115527 7 F2 11p11
fgf4 AF283555 7 FGF4 11q13.3
hsd17b12b BC059617 7 HSD17B12 11p11.2
htatip2 NM_131681 7 HTATIP2 11p15.1
igf2 AF194333 7 IGF2 11p15.5
lmo1 AF398514 7 LMO1 11p15
men1 AF212919 7 MEN1 11q13
mtch2 AF176010 7 MTCH2 11p11.2
mus81 BC055679 7 MUS81 11q13
nucb2a NM_201493 7 NUCB2 11p15.1-p14
pax6b AF061252 7 PAX6 11p13
pc AF295372 7 PC 11q13.4–13.5
ppp1r14b BC076513 7 PPP14B1B 11q13
psmc3 BC071390 7 PSMC3 11p12–p13
rbm4l BC067187 7 RBM4L 11q13
(continued)
Table 3
(continued)

rras2 AI882911 7 RRAS2 11p15.2
scube2 NM_001014813 7 SCUBE2 11p15.3
sf1 BC056288 7 SF1 11q13
slc17a6 AL627163 7 SLC17A6 11p14.3
st5 BC076471 7 ST5 11p15
stk33 BQ617398 7 STK33 11p15.3
with the view that these are duplicate chromosomal segments.

Taken together, these data support the conclusion that drd2a and
drd2c are duplicate DR genes orthologous to the single mamma-
lian D2 DR gene. Our mapping experiments placed drd2b, the
third zebrafish D2 DR gene, on chromosome 16. Although we
have only been able to identify one additional gene that is syntenic
between zebrafish chromosome 16 and human chromosome 11
(Table 3), improvements in the density of the zebrafish chromo-
some 16 map may eventually reveal whether or not these are related
chromosomal segments. Our gene mapping data indicate that the
zebrafish drd4a and drd4c DR genes map to chromosomes 25 and
7, respectively, whereas the human D4R DR gene (DRD4) has
been localized to chromosome 11p15.5 (14). Chromosomes 25
and 7 contain multiple orthologs of human genes located on chro-
mosome 11 (15). The fact that these zebrafish chromosomes share
significant synteny with human chromosome 11 is consistent with
the view that these are duplicate chromosome segments and that
drd4a and drd4c are true orthologs of the single mammalian D4R
DR gene. Our mapping assignments placed drd4b, the third
zebrafish D4R DR gene, on chromosome 4. We were able to iden-
tify only one additional gene that is syntenic between zebrafish
chromosome 4 and human chromosome 11. Again, improvements
in the density of the zebrafish map may eventually reveal whether
or not these are related chromosomal segments. These results are
consistent with the view that zebrafish posses three paralogous
D4R DR genes.
4. Notes
1. Primer Design—It is important to consider primer

length(18–22 bp), GC content (<60%), and annealing temperatures
(>50°C). It is also important to avoid primer secondary structure
by avoiding repeats and runs in a sequence. There are a variety of

primer design software tools available. One program that is avail-
able is OligoAnalyzer by Integrated DNA Technologies (http://
www.idtdna.com/analyzer/Applicati ons/OligoAnalyzer/
Default.aspx).
2. PCR conditions—There are several ways to optimize PCR
conditions. Adding enhancing agents such as dimethyl sulfox-
ide (DMSO) can prevent secondary structure in difficult tar-
gets for amplification. Adjusting the amount of MgCl2 can
influence the specificity of the target DNA sequence and primer
interaction. The typical final concentration of MgCl2 is 1.5 mM,
but ranges above and below that can work as well.
3. Functional Analysis—The homology, phylogeny, and synteny
between the mammalian and zebrafish D2-like DR genes is
highly supported, however this is not always the case. For
example, several D2-like DR genes were identified in Drosophila
with only ~33% amino acid identities shared with their human
counterparts. In these particular cases, it is important to
uncover the pharmacological profiles of the cloned receptors in
order to verify their identities (16).
References
1. Matthysse S (1974) Dopamine and the phar- Clustal X version 2.0. Bioinformatics 23:
macology of schizophrenia: the state of the 2947–2948
evidence. J Psychiatr Res 11:107–113 9. Felsenstein J (1981) Evolutionary trees from
2. Fahn S (2003) Description of Parkinson’s DNA sequences: a maximum likelihood
disease as a clinical syndrome. Ann N Y Acad approach. J Mol Evol 17:368–376
Sci 991:1–14 10. Kwok C, Korn RM, Davis ME, Burt DW,
3. Ning ZCA, Mullikin JC (2001) SSAHA: a fast Critcher R, McCarthy L, Paw BH, Zon LI,
search method for large DNA databases. Goodfellow PN, Schmitt K (1998)
Genome Res 11:1725–1729 Characterization of whole genome radiation
4. Boehmler W, Obrecht-Pflumio S, Canfield V, hybrid mapping resources for non-mammalian
Thisse C, Thisse B, Levenson R (2004) vertebrates. Nucleic Acids Res 26:3562–3566
Evolution and expression of D2 and D3 dop- 11. Le Coniat M, Sokoloff P, Hillion J, Martres
amine receptor genes in zebrafish. Dev Dyn MP, Giros B, Pilon C, Schwartz JC, Berger R
230:481–493 (1991) Chromosomal localization of the
5. Boehmler W, Carr T, Canfield V, Thisse C, human D3 dopamine receptor gene. Hum
Thisse B, Levenson R (2007) D4 dopamine Genet 87:618–620
receptor genes of zebrafish and effects of the 12. Eubanks JH, Djabali M, Selleri L, Grandy DK,
antipsychotic clozapine on larval swimming Civelli O, McElligott DL, Evans GA (1992)
behavior. Genes Brain Behav 6:155–166 Structure and linkage of the D2 dopamine
6. Altschul SF, Gish W, Miller W, Myers EW, receptor and neural cell adhesion molecule
Lipman DJ (1990) Basic local alignment search genes on human chromosome 11q23.
tool. J Mol Biol 215:403–410 Genomics 14:1010–1018
7. Devreux JHP, Smithies O (1984) A compre- 13. Postlethwait JH, Woods IG, Ngo-Hazelett P,
hensive set of sequence analysis programs for Yan YL, Kelly PD, Chu F, Huang H, Hill-
the VAX. Nucleic Acids Res 12:387–395 Force A, Talbot WS (2000) Zebrafish com-
8. Larkin MA, Blackshields G, Brown NP, Chenna parative genomics and the origins of vertebrate
R, McGettigan PA, McWilliam H, Valentin F, chromosomes. Genome Res 10:1890–1902
Wallace IM, Wilm A, Lopez R, Thompson JD, 14. Gelernter J, Kennedy JL, van Tol HH, Civelli
Gibson TJ, Higgins DG (2007) Clustal W and O, Kidd KK (1992) The D4 dopamine receptor
(DRD4) maps to distal 11p close to HRAS. dopamine 2-like receptor: molecular charac-
Genomics 13:208–210 terization and identification of multiple alter-
15. Woods IG, Wilson C, Friedlander B, Chang P, natively spliced variants. Proc Natl Acad Sci
Reyes DK, Nix R, Kelly PD, Chu F, Postlethwait USA 99:14554–14559
JH, Talbot WS (2005) The zebrafish gene map 17. Macrae AD, Brenner S (1995) Analysis of the
defines ancestral vertebrate chromosomes. dopamine receptor family in the compact
Genome Res 15:1307–1314 genome of the puffer fish Fugu rubripes.
16. Hearn MG, Ren Y, McBride EW, Reveillaud I, Genomics 25:436–446
Beinborn M, Kopin AS (2002) A Drosophila
Chapter 13
Application of Cell-Specific Isolation to the Study

of Dopamine Signaling in Drosophila
Eswar Prasad R. Iyer, Srividya Chandramouli Iyer, and Daniel N. Cox
Abstract
Dopamine neurotransmission accounts for a number of important brain functions across species including
memory formation, the anticipation of reward, cognitive facilities, and drug addiction. Despite this func-
tional significance, relatively little is known of the cellular pathways associated with drug-induced molecular
adaptations within individual neurons. Due to its genetic tractability, simplicity, and economy of scale,
Drosophila melanogaster has become an important tool in the study of neurological disease states, includ-
ing drug addiction. To facilitate high-resolution functional analyses of dopamine signaling, it is highly
advantageous to obtain genetic material, such as RNA or protein, from a homogeneous cell source. This
process can be particularly challenging in most organisms including small model system organisms such as
Drosophila melanogaster. Magnetic bead-based cell sorting has emerged as a powerful tool that can be used
to isolate select populations of cells, from a whole organism or tissue such as the brain, for genomic as well
as proteomic expression profiling. Coupled with the temporal and spatial specificity of the GAL4/UAS
system, we demonstrate the application of magnetic bead-based cell sorting towards the isolation of dop-
aminergic neurons from the Drosophila adult nervous system. RNA derived from these neurons is of high
quality and suitable for downstream applications such as microarray expression profiling or quantitative
rtPCR. The versatility of this methodology stems from the fact that the cell-specific isolation method
employed can be used under a variety of experimental conditions designed to survey molecular adaptations
in dopamine signaling neurons including in response to drugs of abuse.
Key words: Drosophila, Dopamine signaling, GAL4/UAS, Magnetic bead cell sorting, Brain, RNA
isolation
1. Introduction
Dopamine neurotransmission systems are found in various species

where they signal the reinforcing properties of natural rewards
such as food, sex, and social interaction (1–4). Genes involved
directly or indirectly in the actions of drugs of abuse appear to also
be conserved across evolution (5, 6), suggesting a preservation of
215
216 E.P.R. Iyer et al.
dopamine mechanisms on addictive behaviors in various species.

Due to its genetic tractability, simplicity, and economy of scale,
Drosophila melanogaster has become an important tool in the study
of neurological disease states, including drug addiction (7).
Previous findings indicate that dopamine plays a key role in
responses of Drosophila to addictive drugs such as ethanol, nico-
tine, and cocaine (5, 8–11).
Magnetic bead-based cell sorting has emerged as a powerful
tool that can be used to isolate select populations of cells, from a
whole organism or specific tissue type such as brain, for genomic,
as well as proteomic expression profiling (12, 13). Coupled with
the temporal and spatial specificity of the GAL4/UAS system, we
demonstrate the application of magnetic bead-based cell sorting to
the specific isolation of genetically labeled, wild-type populations
of dopaminergic neurons from the Drosophila adult brain. This
technique provides a significant advantage over existing method-
ologies wherein samples derived from the brain may be of hetero-
geneous cell populations thereby confounding molecular analyses
of dopamine signaling in either wild-type or experimentally chal-
lenged dopaminergic neurons. RNA derived from these neurons is
suitable for downstream applications such as microarray expression
profiling or quantitative rtPCR. The versatility of this methodol-
ogy stems from the fact that cell-specific isolation can be used
under a wide variety of experimental conditions designed to survey
molecular signaling adaptations in dopaminergic neurons such as
in response to drugs of abuse. Moreover, the availability of
Drosophila strains and commercially available reagents (including
antibodies, magnetic beads, and RNA purification kits optimized
for recovery from relatively small, pure populations of cells at low
elution volumes) provides a unique tool for investigating dopamine
signaling at the level of cellular resolution.
2. Materials
2.1. Magnetic Bead 1. 10× Phosphate-buffered saline (PBS) (MP Bioproducts, Solon,
Preparation OH) diluted to 1× working solution with RNase-free double-
distilled water (ddH2O) and stored at room temperature (see
Note 1).
2. RNase-AWAY (Sigma-Aldrich, St. Louis, MO) stored in a
spray-bottle at room temperature for rendering the working
surfaces RNAse free.
3. Biotinylated Rat anti-Mouse-CD8a antibody (eBioscience, San
Diego, CA) 500 μg/mL stored at 4°C.
4. Dynabeads MyOne Streptavidin T1 (Invitrogen, Life
Technologies, Grand Island, NY), stored at 4°C. Once coupled
13 Cell-Specific Isolation to Study Dopamine Signaling in Drosophila 217
to antibody, the magnetic beads are stored at 4°C until needed

and maximally up to 1 month. 1 mg of beads can bind up to
20 μg of biotinylated antibody.
5. (8) Rare Earth Magnet Blocks: 2 in. (length) × 0.5 in.
(width) × 0.125 in. (thickness) (Magcraft, Vienna, VA).
Alternatively DynaMag™-2 (Invitrogen) may be used.
6. Tabletop Vortex.
7. Crushed ice for incubating the antibody bead mixture.
2.2. Dissecting 1. 10× Phosphate-buffered saline (PBS) diluted to 1× working

and Washing solution in RNase-free ddH2O and chilled on ice (see Note 2).
Adult Fly Heads 2. Transgenic flies expressing UAS-mCD8::GFP (Bloomington
Stock Center, stock# 30125) under the control of ple-GAL4
(Bloomington Stock Center, stock# 8848). ple-GAL4 (also
known as TH-GAL4) drives the expression of GAL4 under the
control of the pale (ple) promoter specifically within dopaminer-
gic neuron clusters in the adult Drosophila brain which project
to specific regions within the mushroom bodies and to the
central complex (14) (Fig. 1). The Drosophila ple gene encodes
the enzyme tyrosine hydroxylase, which is required for dop-
amine biosynthesis.
3. Glass Pasteur pipettes, fire-polished by quickly rotating the
pipette tip over a hot Bunsen burner flame until the pipette
edges become smooth.
4. Dissecting tools: two pairs of sharp Dumont No. 5 forceps.
2.3. Dissociating the 1. 10× Phosphate-buffered saline (PBS) diluted to 1× working

Tissue into a Single solution in RNase-free ddH2O and chilled on ice.
Cell Suspension 2. Kontes Glass Tissue Grinder, 7 mL working capacity, with large
clearance pestle (Kimble Chase, Vineland, NJ) (see Note 3).
Fig. 1. Dopaminergic neuron clusters in adult Drosophila brain. Confocal image of whole
mount adult Drosophila brain from a strain bearing the ple-GAL4 and UAS-mCD8::GFP
transgenes. ple-GAL4 drives expression of UAS-mCD8::GFP specifically in dopaminergic
neurons of the adult brain.
3. BSA (Bovine Serum Albumin), Fraction V. Freshly prepare

10–15 mL of 1% BSA solution in 1× PBS and store on ice (see
Note 1). This is used for coating the tubes and pipettes to
reduce cell loss due to adhesion to glass surfaces.
4. 30 μm cell filters, e.g., MACS Pre-Separation Filter (Miltenyi
Biotec, Cambridge, MA).
5. Pipettes (P-1000, P-100, P-10) with disposable, low-retention
plastic tips.
6. Pasteur pipettes with fire polished tips of standard diameter,
narrowed to ~50% of standard diameter for trituration. Fire
polishing and narrowing of Pasteur pipettes is achieved over a
Bunsen burner flame by rotating the tip in the flame to polish
edges and narrow tip diameter.
7. Fluorescent stereomicroscope equipped with GFP filter-set
(a Leica MZ16FA was used in this protocol).
8. Table top micro-centrifuge (1–16,000 × g).
9. Tabletop Vortex.
10. Crushed ice.
2.4. Magnetic Bead 1. 10× Phosphate-buffered saline (PBS) diluted to 1× working

Cell Sorting solution in RNase-free ddH2O and chilled on ice.
2. Anti-CD8a antibody-coated magnetic beads from
Subheading 3.1, step 4.
3. BSA (Bovine Serum Albumin), Fraction V. Freshly prepare
10–15 mL of 1% BSA solution in 1× PBS and store on ice. This
is used for coating the tubes and pipettes to reduce cell loss
due to adhesion to glass surfaces.
(width) × 0.125 in. (thickness) (Magcraft). Alternatively
DynaMag™-2 (Invitrogen) may be used.
5. Pipettes (P-1000, P-100, P-10) with disposable, low retention
plastic tips.
6. Standard hemocytometer to count the cells and estimate purity.
7. Pasteur pipettes with fire polished tips of standard diameter.
8. Fluorescent stereomicroscope equipped with GFP filter set
(a Leica MZ16FA was used in this protocol).
10. Tabletop Vortex.
11. Crushed ice.
2.5. RNA Isolation 1. PicoPure® RNA Isolation Kit (Applied Biosystems, Life
from Magnetic Bead Technologies).
Sorted Cells 2. Pipettes (P-1000, P-100, P-10) with disposable, low retention
plastic tips.

(width) × 0.125 in. (thickness) (Magcraft). Alternatively
DynaMag™-2 (Invitrogen) may be used.
4. Heating block pre-heated to 42°C stored inside an incubator,
pre-set to 42°C.
6. Tabletop Vortex.
7. RNAse-Free DNAse set (Qiagen, Valencia, CA) (Optional
step. Needed if performing DNAse treatment prior to RNA
column purification).
3. Methods
To ensure high-quality RNA isolation from dopaminergic neurons,

it is imperative that standard lab procedures for maintaining a
clean, RNase-free environment be observed throughout this pro-
tocol in order to prevent RNA degradation (see Note 1). Preparation
for this method requires isolation of age-matched Drosophila adults
bearing a ple-GAL4 transgene and a UAS-mCD8::GFP transgene
in order to genetically label dopaminergic neurons for visualization
and CD8-mediated cell isolation via magnetic bead cell sorting. Prior
to initiating dissection of adult Drosophila heads, it is important
to prepare magnetic beads in advance for use in cell binding. The
antibody-bound magnetic beads are then stored at 4°C until ready
for cell isolation. Following dissection of heads from ple-
GAL4,UASmCD8::GFP adults, the tissue is dissociated and sub-
jected to filtration into single cell suspensions. These single cell
suspensions are verified by epi-fluorescent microscopy for the pres-
ence of the genetically labeled dopaminergic neurons, which
uniquely express green fluorescent protein (GFP). Once verified,
antibody-coated magnetic beads are added to the single cell sus-
pensions and allowed to incubate for a limited time to promote
high affinity and specificity interactions between the antibody-
coated beads and the genetically labeled dopaminergic neurons.
Homogenous, cell-specific isolation is achieved by placing the cell
suspension in a strong magnetic field to pellet the target cells, fol-
lowed by rigorous washing to remove any non-specific contami-
nating cells. The isolated, purified population of cells can be visually
verified for specificity by epi-fluorescent assessment of GFP expres-
sion in all cells. Ultimately, the purified population of dopaminer-
gic neurons is subjected to RNA isolation for downstream
applications. Figure 2 schematically summarizes the procedures
involved in this protocol and Table 1 summarizes the timing
required for each step of the protocol.
Fig. 2. Schematic of procedure for magnetic bead cell sorting of Drosophila dopaminergic neurons. (a) Age-matched, intact
adult Drosophila heads carrying the ple-GAL4 and UAS-mCD8::GFP transgenes are dissected. (b) The heads are transferred
to a microfuge tube and washed/vortexed to remove any debris. (c) The heads are then transferred into a 7 mL dounce to
homogenize the brain tissue followed by trituration to further dissociate the tissue into a cell suspension. (d, e) The homo-
genate is then filtered using a 30 μm cell filter to remove any large cellular debris, including the cuticle from the head and
filter out the single cell suspension. The filtrate solution contains a single cell suspension of different cell types including
neurons and glia. (f) Biotinylated anti-mouse-CD8-antibody-coated magnetic Streptavidin T1 Dynabead are added to the
cell suspension, and incubated on ice for 30–60 min. (g) The magnetic beads specifically bind to the dopaminergic neurons
that express a mouse CD8-tagged GFP fusion protein under the control of the ple-GAL4 driver. (h, i) The magnetic bead-
coated cells are separated by placing the cell solution in a strong magnetic field. The supernatant is discarded, and the
cells are washed three times to remove any nonspecific cells, resulting in (j) highly enrichment of specific dopaminergic
neurons. Panels (b–j) are adapted in part (14).
Table 1
Summary of experimental work flow and timings
Procedure Timing
Preparing magnetic beads for binding (prepared prior to cell isolation experiment) 75–90 min
Dissecting and washing adult fly heads 10–15 min
Dissociating the tissue into a single cell suspension 18–20 min
Magnetic bead cell sorting 45–75 min
RNA isolation from magnetic bead sorted cells 60–75 min
3.1. Preparing This step must be completed prior to the start of the experiment.
Magnetic Beads The prepared beads can be stored at 4°C until needed (see Note 4).
for Binding Cells:
1. Wash 100 μL of Dynabeads StreptavidinT1-coated beads three
(75–90 min)
times in PBS by resuspending the magnetic beads in 1 mL of
fresh 1× PBS and pelleting the beads in a strong magnetic field
after each wash. Discard the supernatant after each wash step.
2. Resuspend the beads directly in 100 μL of undiluted biotinylated

rat anti-mouse-CD8a antibody (antibody concentration is
500 μg/mL).
3. Incubate the mixture for 1 h on ice with occasional mild vortexing
to prevent sedimentation [Dynabeads T1 can bind 20 μg/mg
of biotinylated Ig].
4. Wash the bead–antibody mixture three times as described in
step 1 above to remove any unbound excess antibody. The
magnetic beads are now coated with antibody and ready to
be used. Store the bead-antibody mixture in 100 μL of 1× PBS
at 4°C until use.
3.2. Dissecting 1. Dissect 30–50 age-matched adult Drosophila heads from the
and Washing Adult Fly ple-GAL4,UAS-mCD8::GFP strain. Dissection of adult heads
Heads: (10–15 min) is achieved by using two pairs of sharp Dumont No. 5 forceps
in cold 1× PBS (see Note 2). One pair of forceps is used to
hold the fly in place at the thorax, while the other pair of for-
ceps is placed around the rear side of the fly head where it
meets the thorax. Using the pair of forceps placed behind the
fly head, gently separate the head from the body of the fly and
transfer the intact head to a 1.6 mL microfuge tube filled with
1–1.2 mL of 1× PBS on ice (see Note 5).
2. Once all fly heads have been dissected and transferred to the
microfuge tube, close the tube and vortex it at the maximum
setting three times for 1 s each.
3. Using a fire-polished Pasteur pipette discard the supernatant
completely. Repeat the wash (Subheading 3.2, step 1) and vortex
(Subheading 3.2, step 2) steps 3–4 times until the supernatant
is visibly clear of any and debris.
3.3. Dissociating the 1. To avoid the cells from sticking to the glass surface of the pre-
Tissue Into a Single chilled 7 mL Kontes tissue grinder and large clearance pestle
Cell Suspension: (see Note 3), pre-coat the tissue grinder and pestle with a 1%
(18–20 min) BSA in 1× PBS solution and after a brief rinse discard the BSA
solution. Subsequently, using a P-1000 pipette with a cut low
retention filter plastic filter tip, transfer the heads in 1 mL 1×
PBS from step Subheading 3.2, step 3, and add an additional
3–4 mL of 1× PBS to the BSA-coated tissue grinder.
2. Homogenize the tissue with slow and steady strokes, avoiding
frothing (approximately 20–30 strokes) (see Note 6).
3. To assess the cell dissociation levels, pipette out a small sample
of the solution (up to 50 μL), and observe it under a fluorescent
stereomicroscope using the GFP filter set. If one still observes
intact or incompletely dissociated tissue, homogenize further.
4. Triturate the solution five times with a fire-polished Pasteur
pipette narrowed to approximately 50% of the standard tip
diameter (see Note 7).
5. Filter the solution through a 30-μm cell filter and collect the
cell filtrate in a 1.6 mL microfuge tube. The resulting solution
should consist of a single cell suspension and is now ready for
magnetic cell sorting. Take a small sample of the dissociated
cells and observe it under a fluorescent stereomicroscope.
If cell-clumps are observed, pass the suspension again slowly
through a new 30 μm filter.
3.4. Magnetic Bead Cell 1. Add 15 μL of antibody-coated magnetic beads per 1 mL of cell
Sorting: (45–75 min, suspension (Subheading 3.3, step 5). The remaining antibody-
Depending on conjugated magnetic beads can be stored at 4°C until needed
Antibody Incubation for subsequent cell isolations.
Time) 2. Incubate the cells with antibody-coated magnetic beads for
30–60 min on ice with occasional hand-mixing via inversion of
the microfuge tube (see Note 8).
3. Place the microfuge tube in a strong magnetic field for 2 min.
All positively selected cells along with unbound antibody-
coated magnetic beads will separate to the side of the tube.
4. Slowly pipette the supernatant, making sure not to disturb the
cell pellet and discard.
5. Wash the cells 3–4 times in fresh, ice-cold 1× PBS followed by
magnet-induced pelleting to remove any remaining nonspecific
cells.
6. Resuspend the specifically bound cells coupled to the magnetic
beads in 30 μL of fresh, ice-cold 1× PBS.
7. To approximate the purity and yield of cells, pipette 5 μL of
the cell suspension on the polished surface of a hemocytometer
and count all the visible fluorescent cells under a fluorescent
stereomicroscope equipped with a GFP filter set. Also assess
the relative amount of nonfluorescent cells or any other signs
of impurities. Typically the sample will be highly enriched for
fluorescently labeled, GFP-positive cells (Fig. 3).
3.5. RNA Isolation Although this protocol is focused on RNA isolation as described
from Magnetic Bead here, this step of the overall dopaminergic cell isolation can be
Sorted Cells: replaced by other protein/DNA/chromatin isolation methods
(60–75 min) depending on nature of the study.
1. After counting, pellet the cells in a magnetic field, discard the
supernatant and add 20 μL of extraction buffer from the
PicoPure™ RNA Isolation Kit (Molecular Devices). Depending
upon cell number one may need to add a higher volume of
extraction buffer.
2. Vortex the tube at maximum speed to enable the mixing of cell
pellet with extraction buffer.
3. Incubate the microfuge tube at 42°C for 30 min.
Fig. 3. Magnetic bead isolated dopaminergic neurons. Representative epi-fluorescent

image of a field of positively selected, GFP fluorescent dopaminergic neurons isolated by
magnetic bead cell sorting from the adult Drosophila brain. The purified cell population of
neurons was determined to have little or no contaminating cell impurities. The variation in
cell body size reflects natural variation seen in the adult brain among distinct dopaminer-
gic neuron clusters.
4. To ensure removal of the magnetic beads prior to column

purification of the RNA, the tube is briefly centrifuged at
2,000 × g for 2 min to pellet the magnetic beads. The tube is
then placed in a strong magnetic field to retain the pellet, and
the supernatant is transferred to a new microfuge tube.
5. Extract and column purify the RNA according to the PicoPure™
RNA extraction kit manufacturer’s instructions. DNAse treatment
is optional and can be performed on column during the RNA
purification according to the analysis requirement. Finally,
elute the bound total RNA in a small volume (11–30 μL) of
elution buffer and store at −80°C until ready for use. If desired
a 1 μL aliquot may be used to assess total RNA quality on a
Bioanalyzer 2100 (Agilent Technologies, Inc.) (Fig. 4).
4. Notes
1. All solutions must be prepared in RNase-free double-distilled

water (not DEPC treated). BSA can be prepared prior to each
experiment by dissolving the appropriate, pre-weighed quantity
of BSA in PBS and vortexing at maximum speed.
Fig. 4. Bioanalyzer analysis reveals high-quality RNA purification from isolated neurons.
Representative Agilent 2100 Bioanalyzer (Agilent Technologies, Inc.) electropherogram of
total RNA isolated from magnetic bead sorted dopaminergic neurons. The electrophero-
gram reveals excellent RNA quality and integrity suitable for a wide range of downstream
applications including microarray analyses and qPCR as indicated by the presence of the
5.8S, 18S, and 28S rRNAs.
2. It is important to pre-cool the PBS used for dissection of fly

heads. Placing the bottle of PBS covered in crushed ice for an
hour before the experiment starts can do this. Alternatively,
PBS can be stored at 4°C overnight.
3. Pre-cool the tissue grinder/pestle by placing it on ice for a few
minutes to prevent cell damage/lysis during tissue
homogenization.
4. The volume of antibody–bead mixture prepared depends on
the number of CD8a-positive cells to be isolated given that
1 mg of beads can bind up to 20 mg of biotinylated antibody.
Once prepared, it is recommended that the antibody–bead
mixture be used within 4 weeks.
5. This protocol describes a high-throughput method of isolating
specific neuronal cells from the Drosophila brain beginning
with intact heads. We find that this method is highly efficient
and isolation of single cell suspensions is not impeded by the
head cuticle, which is filtered out when the sample is run
through the 30 mm filter. Alternatively, if desired, the fly brains
can be dissected according to the method described (15) prior
to dissociation, however this will substantially increase the time
required during the dissection stage of the protocol.
6. It is important to homogenize the tissue slowly and steadily
without any sudden movements, to avoid unwanted stress and
damage to the brain cells.
7. Trituration must be performed slowly, as forceful trituration

may damage cells.
8. It is important to note that during the incubation of the anti-
body step, high temperatures or long incubation periods can
result in a loss of antibody specificity thereby compromising
purity in the isolation procedure.
Acknowledgments
The authors thank the Bloomington Stock Center for providing

Drosophila transgenic strains and Dr. Nadine Kabbani for contribu-
tions to this protocol. This work was supported by the Thomas F.
and Kate Miller Jeffress Memorial Trust and NIH MH086928
(DNC).
References
1. Kim YC, Lee HG, Han KA (2007) D1 dop- development of behavioral sensitization in
amine receptor dDA1 is required in the mush- Drosophila. Curr Biol 8:109–112
room body neurons for aversive and appetitive 9. Rothenfluh A, Heberlein U (2002) Drugs,
learning in Drosophila. J Neurosci flies, and videotape: the effects of ethanol and
27:7640–7647 cocaine on Drosophila locomotion. Curr Opin
2. Selcho M, Pauls D, Han K-A, Stocker RF, Neurobiol 12:639–645
Thum AS (2009) The role of dopamine in 10. Heberlein U, Wolf FW, Rothenfluh A,
Drosophila larval classical olfactory condition- Guarnieri DJ (2004) Molecular genetic analy-
ing. PLoS One 4:e5897. doi:10.1371/jour- sis of ethanol intoxication in Drosophila mela-
nal.pone.0005897 nogaster. Integr Comp Biol 44:269–274
3. Ward A, Walker VJ, Feng Z, Xu XZS (2009) 11. Heberlein U, Tsai LT, Kapfhamer D, Lasek
Cocaine modulates locomotion behavior in C. AW (2009) Drosophila, a genetic model system
elegans. PLoS One 4:e5946. doi:10.1371/ to study cocaine-related behaviors: a review
journal.pone.0005946 with focus on LIM-only proteins.
4. Lebestky T, Chang JS, Dankert H, Zelnik L, Neuropharmacology 56:97–106
Kim YC, Han KA, Wolf FW, Perona P, Anderson 12. Iyer EPR, Iyer SC, Sulkowski MJ, Cox DN
DJ (2009) Two different forms of arousal in (2009) Isolation and purification of Drosophila
Drosophila are oppositely regulated by the dop- peripheral neurons by magnetic bead sorting. J
amine D1 receptor ortholog DopR via distinct Vis Exp 34:e1599. doi: 10.3791/1599
neural circuits. Neuron 64:522–536 (2009).
5. Bainton RJ, Tsai LT-Y, Singh CM, Moore MS, 13. Wang X, Bo J, Bridges T, Dugan KD, Pan T-C,
Neckameyer WS, Heberlein U (2000) Chodosh L, Montell DJ (2006) Analysis of cell
Dopamine modulates acute responses to migration using whole genome expression
cocaine, nicotine and ethanol in Drosophila. profiling of migratory cells in the Drosophila
Curr Biol 10:187–194 ovary. Dev Cell 10:483–495
6. Blenau W, Baumann A (2001) Molecular and 14. Friggi-Grelin F, Coulom H, Meller M, Gomez
pharmacological properties of insect biogenic D, Hirsh J, Birman S (2003) Targeted gene
amine receptors: lessons from Drosophila mela- expression in Drosophila dopaminergic cells
nogaster and Apis mellifera. Arch Insect using regulatory sequences from tyrosine
Biochem Physiol 48:13–38 hydroxylase. J Neurobiol 54:618–627
7. Wolf FW, Heberlein U (2003) Invertebrate 15. Wu JS, Luo L (2006) A protocol for dissect-
models of drug abuse. J Neurobiol 54:161–178 ing Drosophila melanogaster brains for live
8. McClung C, Hirsh J (1998) Stereotypic behav- imaging or immunostaining. Nat Protoc
ioral responses to free-base cocaine and the 1:2110–2115
Part IV
Electrochemical, Physiological, and Behavioral Analysis

Chapter 14
Regulation of Dopamine Transporter Expression

by Neuronal Activity
Shalini Padmanabhan, Thach Pham, and Balakrishna M. Prasad
Abstract
Actions of extracellular dopamine released in the central nervous system are primarily terminated by the
dopamine transporter. This protein is also a target for therapeutic and abused psychostimulant drugs.
Different methods used to study dopamine transporter function, its expression, and intracellular signaling
in neurons are described in this chapter. Function of the dopamine transporter in mesencephalic primary
cultures can be measured by dopamine uptake assay. Expression of the transporter protein and mRNA are
analyzed by western blots and quantitative RT-PCR, respectively. Finally, methods to study neuronal activ-
ity-dependent changes in Ca2+⁄calmodulin-dependent protein (CaM) kinase activity in dopamine neurons
are described.
Key words: Mesencephalic, Dopamine, Uptake, Transporter, Western blot, Quantitative RT-PCR,
Action potential, Tyrosine hydroxylase, CaM kinase, MAP kinase
1. Introduction
The primary physiological function of the dopamine transporter is

removal of dopamine from the extracellular space. Dopamine
transporter activity can shape the anatomical distribution and dura-
tion of action of dopamine released from nerve terminals (1, 2).
Blockade of dopamine transporter by psychostimulant drugs can
lead to an enhanced activation of dopamine receptors and result in
overt behavioral changes (3, 4). There is significant interindividual
variability in dopamine transporter expression and this is associated
with measures of attention in humans (5, 6). Furthermore, phar-
macological and environmental factors can reversibly alter the
abundance of dopamine transporter (7–9). However, mechanisms
229
230 S. Padmanabhan et al.
regulating the expression of dopamine transporter are poorly

understood.
We discovered that neuronal activity is one of the factors that
can regulate dopamine transporter expression (10, 11). Calcium
and CaM kinase II signaling mediate neuronal activity-dependent
changes of dopamine transporter expression (10). Primary cultures
of dopamine neurons are the best experimental system to study
neuronal activity-dependent regulatory mechanisms. Heterologous
expression of cDNA constructs is typically used to study uptake
activity of dopamine and other monoamine transporters. These
expression systems offer the advantages of easy data interpretation,
availability of molecular tools, and feasibility of structure–function
analysis. However, the intracellular environment, electrical activity,
and signaling machinery in heterologous expression systems such
as HEK 293, MDCK, or HeLa cells are significantly different from
that of dopamine neurons. Ex vivo preparations such as synapto-
somes provide an alternative to overcome these limitations.
However, ex vivo preparations are not suitable to perform long-
term experiments, especially those involving gene expression
changes. Thus, primary cultures are the best available option at
present to reliably study changes in dopamine transporter
expression.
2. Materials
2.1. Mesencephalic 1. Time-mated female Sprague Dawley rats (Harlan, Indianapolis,

Primary Cultures IN) are typically obtained at 14–16 day gestation.
2. Ketamine HCl/Ketaset (10 mg/mL, Fort Dodge Laboratories,
Fort Dodge, IA).
3. Dissociation medium salt concentrate (5×): 2,900 mM NaCl,
135 mM KCl, 650 mM NaHCO3, 100 mM NaH2PO4 H2O,
20 mM MgSO4, 25 mM EDTA, and 1200 mM d-glucose
made in water and stored at 4°C (see Note 1).
4. Cysteine water: 1.25 mM cysteine in water supplemented with
1.9 mM CaCl2.
5. Papain (Worthington Biochemicals, Lakewood, NJ) is activated
by adding it to cysteine water (20 units/mL) (see Note 2).
6. Dissociation medium is prepared by adding dissociation
medium salt concentrate (1×) and kynurenate (0.5 mM) to
activated papain solution. pH is adjusted to 7.2–7.4.
7. Carbogen cylinder (95% oxygen and 5% carbon dioxide).
8. Poly-D-lysine (Sigma-Aldrich, St. Louis, MO): 10 mg/mL
(100×) single-use aliquots are made up in Hank’s balanced salt
solution (Invitrogen, Carlsbad, CA) and stored at −20°C. 1×
14 Dopamine Transporter Expression 231
(100 μg/mL) Poly-D-lysine is made by diluting 100× stocks in

tissue-culture grade water, filter-sterilized and used fresh.
9. Laminin (Sigma-Aldrich): 1 mg/mL single-use aliquots are
stored at −80°C. 5 μg/mL laminin is made up in minimum
essential medium (Invitrogen, Carlsbad, CA) and used fresh.
10. 12-well, 24-well, and 48-well tissue culture plates (Costar).
These plates are coated with poly-d-lysine by incubating in 1×
solution for 1 h at room temperature. The wells are washed
twice with sterile water and laminin solution is added to the
wells for 1 h. Dissociated cells are plated immediately after
aspirating the laminin solution out of the plates. 48-Well plates
are typically used for uptake assays, while 24-well and 12-well
plates are used for western and RT-PCR assays, respectively.
11. Glial medium: Minimum essential medium with 10% new-born
calf serum (Invitrogen), 0.45% D-glucose, 5 pg/mL insulin,
0.5 mM glutamine, 100 U/mL penicillin, and 100 μg/mL
streptomycin. All media components are added to minimum
essential medium and filter sterilized. This medium is used to
maintain cultures of cortical glia and for initial plating of mes-
encephalic cells.
12. Neuronal medium: 50% minimum essential medium, 39%
Ham’s-F12 medium, 10% heat-inactivated horse serum
(Invitrogen), 1% heat-inactivated fetal bovine serum
(Invitrogen), 0.45% d-glucose, 5 pg/mL insulin, and 0.1 mg/
mL apotransferrin (ICN Biomedicals, Irvine, CA).
13. Conditioned neuronal medium: 25 mL of neuronal medium is
added to cortical glia grown in T75 tissue culture flasks. After
overnight conditioning, the media is collected and supple-
mented with 1 ng/mL glial cell line-derived neurotrophic fac-
tor (1 ng/μL stock solution is made in 5% BSA water and stored
as single-use aliquots at −80°C) and 500 μM kynurenate (0.5 M
stock solution is made in 1 N NaOH and stored at −80°C as
single-use aliquots). The medium is filter-sterilized and can be
stored at −80°C, if not used immediately (see Note 3).
14. Flea micro stir bar (Fisher).
15. Sterile petri dishes (Fisher).
16. Tetrodotoxin (1 mM, PH 4–5) and 4-aminopyridine (100 mM)
(Sigma-Aldrich) are made up in PBS, filter-sterilized and stored
as single-use aliquots at −80°C.
2.2. Dopamine Uptake 1. Ringer’s solution: 130 mM NaCl, 3 mM KCl, 1.2 mM MgSO4,
Assay 1 mM CaCl2, 10 mM d-glucose, and 1 mM ascorbic acid, pH
7.4 (one 50 mL tube warmed to 37°C and another tube chilled
over ice).
2. (2, 5, 6 3H)-dopamine (GE Healthcare, Piscataway, NJ) is

diluted to appropriate concentration in warm Ringer’s solu-
tion before use.
3. Unlabeled dopamine stocks (10 mM) are made up in Ringer’s
solution and stored at −80°C. For determining uptake kinetics,
4.05 μM dopamine is made with 20% radiolabeled and 80%
unlabeled dopamine. Other dopamine concentrations are
obtained by serial dilution of the 4.05 μM dopamine stock.
4. 3% trichloroacetic acid in water.
5. GBR12909 (Tocris Cookson, Ellisville, MO) is dissolved to
1 mM in water and stored at −80°C.
6. ScintiVerse scintillation liquid (Fisher Scientific, Pittsburgh, PA).
2.3. Western Blots 1. 12% Polyacrylamide gels (Bio-Rad, Hercules, CA).

2. Electrophoresis buffer: 25 mM Tris base, 192 mM glycine, and
0.5% SDS.
3. Electroblotting buffer: 20 mM Tris, 150 mM glycine, and 20%
methanol. Adjust pH to 8.0 before adding methanol.
4. Sample buffer 6×: 60 mM TRIS, 1 mM EDTA, 2% SDS, 10%
glycerol, 100 mM DTT, and 0.004% bromophenol blue.
5. Nitrocellulose membrane (0.45 μm NitroPure; Fisher Scientific,
Pittsburgh, PA).
6. Ponceau S solution (0.1% in 5% acetic acid).
7. Prestained molecular Marker (Bio-Rad).
8. Blocking buffer: 5% nonfat dry milk in PBST.
9. Wash buffer (PBST): 1× PBS with 0.1% Tween-20.
10. SuperSignal West Pico Chemiluminescent Substrate (Pierce,
Rockford, IL).
11. RIPA buffer: 50 mM Tris (pH 6.8), 150 mM NaCl, 4 mM
EDTA, 100 mM sodium fluoride, 1 mM sodium orthovana-
date, 1% sodium deoxycholate, 0.1% SDS, 1% Triton-100×.
A Complete mini protease Inhibitor tablet (Roche Diagnostics,
Indianapolis, IN) is added to 10 mL buffer, right before use.
12. Stripping buffer (Pierce).
13. Primary antibodies:
(a) Rabbit polyclonal dopamine transporter antibody (10).
(b) Monoclonal TH antibody (clone TH16, Sigma-Aldrich).
(c) β-Actin antibody (goat polyclonal, Santa Cruz
Biotechnology, Santa Cruz, CA).
(d) Phospho-CaM kinase II (Thr286/7) and CaM kinase II anti-
bodies (Cell Signaling, Danvers, MA).
(e) Phospho-TH (Ser19) antibody (Invitrogen).
14. HRP-conjugated secondary antibodies:

(a) Goat Anti-Mouse (Jackson ImmunoResearch Laboratories,
West Grove, PA)
(b) Donkey Anti-Rabbit (Jackson ImmunoResearch)
2.4. Quantitative 1. Trizol reagent (Invitrogen).

RT-PCR 2. Chloroform.
3. Isopropanol.
4. 75% Ethanol.
5. RNAse-free water (Invitrogen).
6. Superscript III reverse transcriptase kit (Invitrogen).
7. SYBR® Green PCR supermix (Bio-Rad).
8. Bio-Rad iQ iCycler.
9. iQ 96-Well PCR Plates (Bio-Rad).
10. Optical Sealing Tape (Bio-Rad).
11. Primers (Integrated DNA Technologies, Coralville, IA):
DAT (F)—5¢-TGGGCCTCAATGACACCTTT-3¢
DAT (R)—5¢-AGCAGAACAATGACCAGCACCA-3¢
Actin (F)—5¢-TTGCTGACAGGATGCAGAAGGAGA-3¢
Actin (R)—5¢-TAGAAGCATTTGCGGTGCACGATG-3¢
3. Methods
3.1. Mesencephalic As described above, mesencephalic primary cultures are the best
Primary Cultures available experimental system to study the regulation of dopamine
transporter expression. Currently available cell-lines of dopamine
neurons do not express appreciable amount of dopamine trans-
porter. We adapted the dopamine neuronal culture methods origi-
nally described by Rayport et al. (12), for measuring dopamine
transporter function and expression.
1. 2–4-day-old Sprague-Dawley rat pups are anesthetized with
intraperitoneal injection of ketamine (3 mg per pup).
2. Brains are aseptically dissected out and placed in a sterile petri
dish with Ringer’s solution. Approximately 3 mm thick coronal
sections of mid-brain just caudal to hypothalamus are dissected.
Ventral one-third of the coronal sections are cut into two pieces
and placed in the dissociation medium (see Note 4).
3. Ventral mid-brain sections in dissociation medium are gently
stirred using a micro-stir bar and stir plate. The vial is placed in
Carbogen Vent
18g Needle
Thermometer
0.2 µm Filter
Water
Hot Plate
Water
Micro Stir Bar
Fig. 1. Tissue dissociation setup. Carbogen is humidified via sterile water in a conical flask and passed through a 0.2 μm
filter into dissociation medium. Dissociation vial with mesencephalic tissue is placed in a water bath maintained at 35–37°C.
Tissue in dissociation vial is continuously bubbled with humidified carbogen and slowly stirred using fleabar for 2 h.
a 37°C water bath and tissue is constantly bubbled with

carbogen (Fig. 1).
4. After 2 h of incubation, the tissue sections are transferred into
a sterile 15 mL conical tube. Sections are dissociated using
10 mL plastic serological pipette, then by fire polished glass
Pasteur pipette. Complete dissociation of tissue is ensured by
the absence of visible tissue mass.
5. Dissociated cells are pelleted by centrifuging at 1,000 × g for
10 min at 4°C.
6. Dissociation medium is aspirated out and cells are resuspended
in glial medium. Cell number is calculated using hemocytom-
eter and the medium is diluted to obtain 300,000 cells per
mL.
7. Cells are plated in 48-well plates previously coated with poly-
d-lysine and laminin. 150,000 cells are plated per well using
repeat pipetter to minimize inter-well variability in cell density.
For western blots 300,000 cells are plated per well in 24-well
plates and for quantitative RT-PCR assays 600,000 cells are
plated in 12-well plates.
8. An hour after plating, glial medium is aspirated and replaced
with conditioned neuronal medium.
9. After 7 days in vitro, medium is changed to appropriate treat-
ment medium (with vehicle or drugs).
10. Different assays for dopamine transporter function and abun-
dance are performed as described below.
3.2. Dopamine Uptake The primary function of dopamine transporter is the reuptake of
Assay extracellular dopamine. The simplest and most commonly used
assay to measure dopamine transporter function is to assess the
intracellular accumulation of radiolabeled dopamine. Translocation
of substrates by dopamine transporter is analogous to enzymatic
action of converting substrates into products. Thus, kinetics of sub-
strate translocation are typically analyzed using Michaelis-Menten
equation. Dopamine accumulation is typically measured at multiple
concentrations (with the highest concentration being about tenfold
greater than expected apparent affinity of dopamine, Km).
1. Cells are washed once in Ringer’s solution.
2. Radiolabeled dopamine in a total volume of 300 μL is added
to wells. Five different concentrations of tritiated dopamine
(0.05, 0.15, 0.45, 1.35, and 4.05) are used in duplicate wells
(see Note 5).
3. After 2 min of incubation at room temperature, uptake is ter-
minated by washing twice with 500 μL ice-cold Ringer’s solu-
tion (see Note 6).
4. Tritiated dopamine is extracted into 300 μL of 3% trichloroa-
cetic acid by incubating 30 min at room temperature.
5. Trichloroacetic acid samples are transferred to scintillation vials
and 5.0 mL of scintiverse fluid is added.
6. Radioactivity in each tube is measured by a liquid scintillation
counter.
7. Nonspecific dopamine accumulation at each concentration
(0.05–4.05 μM) is determined by incubating in the presence
10 μM GBR12909. Specific uptake is calculated as the differ-
ence between total (in absence of GBR12909) and nonspecific
uptake.
8. Data are analyzed by the equation V = Vmax· (DA)/(Km + (DA))
using PRISM (GraphPad, La Jolla, CA). V is the velocity of
uptake at a given concentration of dopamine (DA). Vmax repre-
sents maximal velocity, while Km is the apparent affinity of the
substrate to transporter (see Fig. 2).
3.3. Measuring Dopamine neurons in culture conditions described above are toni-
Dopamine Transporter cally active and fire action potentials at an average rate of 2.3 Hz
Abundance (13). Their firing can be abolished in the presence of 1 μM tetro-
dotoxin (TTX) (a sodium channel blocker) or increased by 1 mM
4-aminopyridine (a potassium channel blocker). In order to test
the effect of neuronal activity on dopamine transporter function
and abundance, we maintained mesencephalic cultures in the pres-
ence of control, TTX or 4-aminopyridine media for 5 days.
Dopamine uptake data shown in Fig. 3 indicate that TTX decreases
dopamine uptake while, 4-aminopyridine increases uptake.
12
10
pmoles / well
8
0
0.01 0.1 1 10
Dopamine [µM]
Fig. 2. Dopamine uptake kinetics in mesencephalic cultures. Picomoles of dopamine

accumulated per well in 2 min at different substrate concentrations are shown. Vmax of
dopamine uptake is11.6 ± 0.6 pmol/well and Km is 0.42 ± 0.08 μM. Average data from
four independent experiments are shown.
4-AP
140 Control
*
TTX
Dopamine Uptake (% control)
120
100
80 *
60
40
20
4 3 6 6
0
30minutes 5 days
Fig. 3. The effect of neuronal activity on dopamine uptake in mesencephalic cultures.

Uptake of 100 nM 3 H-dopamine in cultures treated with 1 μM TTX or 1 mM 4-aminopyridine
(4-AP) for 30 min or 5 days is shown as percentage of control. n = 3–6 as indicated.
*P < 0.05, compared with control. Adapted from (10) with permission.
In order to determine if these changes in dopamine uptake are

accompanied by changes in dopamine transporter abundance, we
analyzed cell lysates using western blots. A typical western blot
assay is described below.
1. After appropriate treatments, cultures are washed once in
Ringer’s solution.
2. Cells are extracted into 200 μL RIPA buffer by gentle agitation
of culture plate on a shaker.
3. Cell lysates are mixed with 6× sample buffer and run on 10%
polyacrylamide gels. A molecular weight marker is run in the
first lane of the gel. This will be used to ensure appropriate
molecular mass of the protein of interest and also to orient the
samples when they are transferred to the membrane.
4. Protein samples resolved on the gels are transferred to nitrocel-
lulose membrane using 600 mA-h of power (15 h at 40 mA or
3 h at 200 mA).
5. Membranes with protein samples are stained with Ponceau-S
by incubating in the staining solution for 5 min and then wash-
ing in water.
6. Sample lanes are labeled and the molecular weight bands are
marked.
7. Membrane is incubated in blocking buffer for 30 min at room
temperature.
8. Dopamine transporter antibody is diluted in the blocking buf-
fer (1:1,000) and the membrane blot is incubated in the anti-
body solution for 1 h at room temperature.
9. Blot is washed four times in PBST with 5 min per wash.
10. Blot is incubated in an HRP-conjugated anti-rabbit secondary
antibody (1:10,000 dilution) for 1 h at room temperature.
11. After five washes in PBST, a chemiluminescent substrate solu-
tion is added to the blot.
12. After 1 min of incubation, the chemiluminescence from the
blot is captured by using an X-ray film or an imaging system
(Kodak in vivo Fx Pro) (see Note 7).
13. Intensities of the bands corresponding to the dopamine trans-
porter are analyzed by NIH image.
14. Data can be expressed as a percentage of control within each
experiment. β-Actin is used as an internal control to normalize
for inter-sample variability in protein concentration.
3.4. Measuring Data presented in Fig. 4 show that dopamine transporter abun-
Dopamine Transporter dance is altered by neuronal activity. In order to determine if these
mRNA Abundance changes are accompanied by changes in dopamine transporter
mRNA abundance, we used quantitative RT-PCR assay.
1. After appropriate treatments, cultures are washed once in
Ringer’s solution.
2. Cells are extracted into 0.5 mL Trizol reagent by passing sev-
eral times through a pipette. Samples are incubated for 5 min
at room temperature (see Note 8).
3. 100 μL of chloroform is added per sample, shaken vigorously,
and incubated at room temperature for 3 min.
4-AP
b
150 Control
Densitometric Units (% control)

*
TTX
125
a
Control TTX Control 4-AP
100
DAT *
75
TH 50
25
b-Actin
0
DAT TH Actin
Fig. 4. The effect of chronic changes in neuronal activity on dopamine transporter abundance. (a) Representative images
of western blots from control, TTX- or 4-AP-treated cultures that were probed with dopamine transporter (DAT), tyrosine
hydroxylase (TH), and β-actin antibodies. (b) Densitometric analysis of bands corresponding to DAT, TH, and β-actin. Data
from NIH image analysis were normalized to respective control values (n = 5). *P < 0.05, compared with control cultures.
Adapted from ref. (10) with permission.
4. Samples are centrifuged at 12,000 × g for 15 min at 4°C. Upper

aqueous phase is collected into a new microfuge tube and
250 μL isopropyl alcohol is added. Samples are incubated at
room temperature for 10 min.
5. Tubes are centrifuged at 12,000 × g for 10 min at 4°C.
Supernatant is removed and RNA pellet is gently washed by
adding 0.5 mL 75% ethanol.
6. The pellet is dissolved in RNAse-free water and RNA concen-
tration of each sample is measured by a spectrophotometer at
260 nm wavelength.
7. 2 μg of total RNA is reverse transcribed by Superscript III assay
kit, using oligo-dT primers.
8. 20 μL of RT samples are diluted to 100 μL using RNAse-free
water.
9. Diluted RT samples (10 μL), SYBR GreenII mastermix
(12.5 μL), and gene-specific primers (2.5 μL) are added to the
wells in PCR plate.
10. PCR reactions are carried out in Bio-Rad iQ iCycler using the
following parameters; Initial denaturation for 4 min at 95°C.
40 PCR cycles are performed with 95°C melting, 57°C anneal-
ing, and 72°C extension, with 30 s for each step (see Note 9).
11. Threshold cycles (CT) are calculated using baseline subtracted
curves.
12. Fold change in mRNA abundance is calculated by comparative
threshold cycle method (Scmittgen and Livak, 2008). Briefly,
a b 3
DAT mRNA fold change

*
2
1
*
4 6
0
ol
X
C AP
TT
tr
4-
on
Fig. 5. Effects of neuronal activity on dopamine transporter mRNA abundance. (a) RT-PCR assay validation: Different
volume of pooled cDNA samples that correspond to 0.05, 0.1, 0.2, 0.4, and 0.8 μg of total RNA were used to generate DAT
and actin threshold cycles (CT). Standard curve graphs were plotted with RNA (μg) concentration and threshold cycles.
Insets in these graphs are curves of relative fluorescence units (of cDNA samples corresponding to 0.05, 0.1, 0.2, 0.4, and
0.8 μg of RNA) plotted against PCR cycle numbers. The correlation coefficients for DAT and actin standard curves are
greater than 0.99. The data are best fit to the following equations: DAT: CT = −3.189 log (μg RNA) + 20.98 and Actin:
CT = −3.172 log (μg RNA) + 10.56. Amplification efficiencies defined as product increase per cycle were calculated by
10(−1⁄slope). PCR efficiency for DAT and actin were 2.058 and 2.066, respectively. These data demonstrate that amplification
efficiency for DAT and actin are similar and are close to 100% (102.9 and 103.3%, respectively). (b) Changes in DAT mRNA
abundance normalized to actin mRNA are expressed as fold change compared to control cultures. N values of RT-PCR
assay data are shown in the graph. *P < 0.05, compared with control cultures. Adapted from ref. (10) with permission.
dopamine transporter CT values of each sample are subtracted

from actin CT values to get delta CT values. The difference
(delta-delta CT value) between treatment delta CT values (TTX
or 4-AP) and control delta CT values is used to calculate the fold
change. Fold change = 2(delta-delta CT value) (see Fig. 5, Note 10).
3.5. Measuring CaM CaM kinase II and mitogen-activated protein (MAP) kinase are
Kinase II Activity in two major signaling molecules that mediate activity dependent
Dopamine Neurons changes in neurons (14, 15). Using pharmacological inhibitors we
identified that CaM kinase II, but not MAP kinase is involved in
activity-dependent changes in dopamine transporter expression (10).
We developed assays for direct measurement of the activity of these
two kinases in mesencephalic cultures (10, 11). Dopamine neurons
are the only cells that express tyrosine hydroxylase in our culture
conditions. Tyrosine hydroxylase can be phosphorylated at Ser19 by
CaM kinase II and by MAP kinase at Ser31 (16). We used western
blot assays to measure the phosphorylation state of TH at these
two sites as a read out of CaM kinase and MAP kinase. Data pre-
sented in Fig. 6 show that phosphorylation of tyrosine hydroxylase
(Ser19) can be used to monitor neuronal activity-dependent changes
in CaM kinase II activity in dopamine neurons.
a b
CaM KII Activation
CaM K II Phospho-CaMK II 250 4-AP
Densitometric Units (% control)

* Control
200 TTX
TH Phospho-TH *
CaM KII Activity 150
100
*
4-AP Control TTX
50
p-CaMKII
*
CaMKII 0
p-TH
H
II
K
/T
aM
TH
TH
p-
II/
K
aM
C
p-
Fig. 6. Effect of neuronal activity on CaM kinase II activity in dopamine neurons. (a) Representative immunoblots showing
the effects of a 30 min treatment with 1 μM TTX and 1 mM 4-AP on phospho-CaM kinase II, CaM kinase II, phospho-TH,
and TH. (b) Densitometric analysis shows that ratios of phospho-CaM kinase II:CaM kinase II (a measure of CaM kinase II
activation) and phospho-TH:TH (a measure of CaM kinase II activity in dopamine neurons) are altered by TTX and 4-AP
(n = 5). *P < 0.05, compared with control cultures. Modified from ref. (10).
4. Notes
1. Unless otherwise noted, all reagents are obtained from Sigma-

Aldrich.
2. Papain activity expressed in units per milligram protein varies
between lots. The protein content per unit volume also varies
between lots. Both these factors need to be used in determin-
ing the volume of enzyme stock to be used.
3. Cortical glia is cultured from postnatal rat pups using the
same protocol described for mesencephalic cultures. Neurons
in these cultures are eliminated by feeding glial medium
containing 100 μM l-glutamate for 24 h on the day of
plating.
4. Correctly dissected mid-brain sections have a visible groove
(mid-brain flexure) in the middle of the ventral surface. Proper
dissection ensures the presence of sufficient dopamine neurons
without any norepinephrine neurons. Presence of norepineph-
rine neurons can potentially be a major problem as these neu-
rons can accumulate dopamine and contain tyrosine
hydroxylase. An easy way to ensure lack of norepinephrine
neurons in culture is to test the susceptibility of dopamine
uptake to 100 nM GBR12909 (dopamine transporter inhibitor)
and 100 nM desipramine (norepinephrine transporter

inhibitor) (13).
5. A commonly used method to obtain different substrate
concentrations for kinetic analysis is to use a low (100 nM)
concentration of radiolabeled dopamine and dilute it with
unlabeled dopamine to obtain desired concentration. Quantity
of radiolabeled dopamine accumulated is multiplied by the
dilution factor to obtain an estimate of total dopamine
accumulated at each concentration. An alternative method is to
use radiolabeled dopamine as a fixed proportion of the total
dopamine. In the uptake assay described above 20% of total
dopamine is radiolabeled. Although the latter method requires
the use of more radiolabeled substrate, the data obtained are
likely to be more accurate. This is because the data at the high
end of concentration range are not multiplied by large dilution
factors, as in the commonly used method.
6. To obtain accurate transport kinetic parameters, it is important
to ensure that uptake of highest substrate concentration used
is linear within the duration of uptake assay. Uptake of 4.05 μM
dopamine is linear with time for at least 4 min. When single
low concentration of tritiated dopamine (100 nM) is used
uptake can be carried out for 4–6 min.
7. When optimizing the imaging conditions, western blots of
serially diluted samples with different exposure times can be
collected. Appropriate exposure timing is important to ensure
that the bands corresponding to different samples are in the
dynamic range of image analyzing program used. X-ray film
images can be digitized by commercially available scanners.
8. RNAse-free pipette tips and nuclease-free molecular grade
reagents are critical for this assay. Samples and reagent vials are
to be handled using gloves.
9. Agarose gel analysis of PCR products can be used to confirm
that only one product of predicted size is amplified.
10. The underlying assumption of this method is that different
PCR products are amplified with equal efficiency. Amplification
efficiency for each PCR product can be measured by using
serial dilution of pooled cDNA samples from different cultures
(Fig. 5).
References
1. Gainetdinov RR, Jones SR, Fumagalli F, mice lacking the dopamine transporter: functional
Wightman RM, Caron MG (1998) consequences. Eur J Neurosci 12:2985–2992
Re-evaluation of the role of the dopamine 3. Di Chiara G, Imperato A (1988) Drugs abused
transporter in dopamine system homeostasis. by humans preferentially increase synaptic dop-
Brain Res Rev 26:148–153 amine concentrations in the mesolimbic system
2. Benoit-Marand M, Jaber M, Gonon F (2000) of freely moving rats. Proc Natl Acad Sci USA
Release and elimination of dopamine in vivo in 85:5274–5278
4. Volkow ND, Wang G, Fowler JS, Logan J, involvement of dopamine transporter and
Gerasimov M, Maynard L, Ding Y, Gatley SJ, trophic factors. J Neurosci 23:10999–11007
Gifford A, Franceschi D (2001) Therapeutic 10. Padmanabhan S, Lambert NA, Prasad BM
doses of oral methylphenidate significantly (2008) Activity-dependent regulation of the
increase extracellular dopamine in the human dopamine transporter is mediated by CaM kinase
brain. J Neurosci 21(2):RC121, 1–5 signaling. Eur J Neurosci 28:2017–2027
5. Newberg A, Amsterdam J, Shults J (2007) 11. Padmanabhan S, Prasad BM (2009) Sustained
Dopamine transporter density may be associ- depolarization decreases calcium/calmodulin-
ated with the depressed affect in healthy sub- dependent protein kinase II activity and gene
jects. Nucl Med Commun 28:3–6 expression in dopamine neurons. Neuroscience
6. Volkow ND, Wang GJ, Newcorn J, Fowler JS, 163:277–285
Telang F, Solanto MV, Logan J, Wong C, Ma 12. Rayport S, Sulzer D, Shi WX, Sawasdikosol S,
Y, Swanson JM, Schulz K, Pradhan K (2007) Monaco J, Batson D, Rajendran G (1992)
Brain dopamine transporter levels in treatment Identified postnatal mesolimbic dopamine neu-
and drug naive adults with ADHD. Neuroimage rons in culture: morphology and electrophysi-
34:1182–1190 ology. J Neurosci 12:4264–4280
7. Dresel S, Krause J, Krause KH, LaFougere C, 13. Prasad BM, Amara SG (2001) The dopamine
Brinkbaumer K, Kung HF, Hahn K, Tatsch K transporter in mesencephalic cultures is refrac-
(2000) Attention deficit hyperactivity disorder: tory to physiological changes in membrane
binding of (99mTc)TRODAT-1 to the dop- voltage. J Neurosci 21:7561–7567
amine transporter before and after meth- 14. Vaillant AR, Zanassi P, Walsh GS, Aumont A,
ylphenidate treatment. Eur J Nucl Med Alonso A, Miller FD (2002) Signaling mecha-
27:1518–1524 nisms underlying reversible, activity-dependent
8. Feron FJ, Hendriksen JG, van Kroonenburgh dendrite formation. Neuron 34:985–998
MJ, Blom-Coenjaerts C, Kessels AG, Jolles J, 15. Kelleher RJ III, Govindarajan A, Jung HY, Kang
Weber WE, Vles JS (2005) Dopamine trans- H, Tonegawa S (2004) Translational control by
porter in attention-deficit hyperactivity disor- MAPK signaling in long-term synaptic plasticity
der normalizes after cessation of and memory. Cell 116:467–479
methylphenidate. Pediatr Neurol 33:179–183
16. Haycock JW, Haycock DA (1991) Tyrosine
9. Bezard E, Dovero S, Belin D, Duconger S, hydroxylase in rat brain dopaminergic nerve
Jackson-Lewis V, Przedborski S, Piazza PV, terminals. Multiple-site phosphorylation in vivo
Gross CE, Jaber M (2003) Enriched environ- and in synaptosomes. J Biol Chem
ment confers resistance to 1-methyl-4-phenyl- 266:5650–5657
1,2,3,6-tetrahydropyridine and cocaine:
Chapter 15
Monitoring Axonal and Somatodendritic Dopamine Release

Using Fast-Scan Cyclic Voltammetry in Brain Slices
Jyoti C. Patel and Margaret E. Rice
Abstract
Brain dopamine pathways serve wide-ranging functions including the control of movement, reward, cog-
nition, learning, and mood. Consequently, dysfunction of dopamine transmission has been implicated in
clinical conditions such as Parkinson’s disease, schizophrenia, addiction, and depression. Establishing fac-
tors that regulate dopamine release can provide novel insights into dopaminergic communication under
normal conditions, as well as in animal models of disease in the brain. Here we describe methods for the
study of somatodendritic and axonal dopamine release in brain slice preparations. Topics covered include
preparation and calibration of carbon-fiber microelectrodes for use with fast-scan cyclic voltammetry, prep-
aration of midbrain and forebrain slices, and procedures of eliciting and recording electrically evoked
dopamine release from in vitro brain slices.
Key words: Dopamine, Brain slices, Voltammetry, Carbon-fiber microelectrodes, Striatum, Substantia
nigra pars compacta, Ventral tegmental area
1. Introduction
The principal dopamine (DA) pathways in the brain arise from

midbrain nuclei designated as the A9 and A10 cell groups, which
are the substantia nigra pars compacta (SNc) and ventral tegmental
area (VTA), respectively (1). Dopamine neurons in these regions
project ipsilaterally via the medial forebrain bundle (MFB) to vari-
ous forebrain structures (Fig. 1a). Neurons of the SNc send axon
projections via the nigrostriatal pathway to the dorsolateral stria-
tum (caudate putamen, CPu), which is critically involved in the
control of movement (2, 3). Neurons of the VTA project via the
mesolimbic pathway to the ventromedial striatum (nucleus accum-
bens, NAc), which is involved in motivation and reward (4) and
243
244 J.C. Patel and M.E. Rice
a Forebrain slice Midbrain slice
2 mm
cc
CPu
NAc ac
SN
mfb
OT u
lateral 2.4 mm
b c
2 mm cc
cc
CPu
NAc (core) SNr

SNc
NAc (shell) VTA
OTu
bregma +1.0 mm bregma –4.8 mm
Fig. 1. Rat brain sections showing major forebrain and midbrain dopamine (DA) regions.
(a) Sagittal section of rat brain, stained with cresyl violet, showing main forebrain (axon
terminal) and midbrain (somatodendritic) regions from which DA release can be recorded.
Lateral coordinate indicates distance from rat central suture, a landmark on the skull.
Dashed lines show the relative positions at which coronal forebrain or midbrain slices are
typically made. (b) Typical level of a coronal forebrain slice used to study axonal DA
release. At this level, it is possible to record DA release from the dorsal striatum (CPu) as
well as the ventral striatum (NAc, core and shell or OTu) in the same forebrain slice. (c)
Typical level of a midbrain slice to study somatodendritic DA release. At this level, it is
possible to record DA release in the SNc (A9 region) as well as the VTA (A10 region).
Bregma coordinates indicate distance anterior (+) and posterior (−) to this landmark on
the skull. Abbreviations: ac anterior commissure, Cer cerebellum, cc corpus callosum,
CPu caudate putamen, mfb medial forebrain bundle, NAc nucleus accumbens, OTu olfac-
tory tubercle, SN substantia nigra, SNc SN pars compacta, SNr SN pars reticulata, VTA
ventral tegmental area (from ref. (60) with permission from Elsevier).
to the olfactory tubercle (OTu), which is involved in olfactory

processing. Additionally, VTA DA neurons project via the meso-
cortical pathway to the amygdala, which is involved in the control
of emotion, and to prefrontal cortex, which is involved in cogni-
tion, including working memory. Consequently, dysfunction of
15 Monitoring Dopamine Release 245
DA transmission has been implicated in several significant brain

disorders, including the movement deficits of Parkinson’s disease,
addiction to substances like alcohol and cocaine as well as natural
rewards including food and sex, and the pathophysiology of schizo-
phrenia and depression.
Studies of DA release regulation using voltammetric methods
with carbon-fiber microelectrodes have made significant contribu-
tions to the present understanding of DA regulation and role of
DA in normal and pathological conditions (5). A number of key
mechanistic insights have been obtained using these methods in
brain slices in vitro, which readily permit voltammetric detection of
DA release in discrete brain regions, without complicating factors
inherent to in vivo studies, including animal behavior, anesthesia,
and indirect effects from distant structures via long pathways (5, 6).
The use of midbrain slices permits examination of somatodendritic
DA release in the SNc and VTA, whereas the use of forebrain slices
permits monitoring of axonal DA release in the dorsal striatum and
nucleus accumbens shell and core. These techniques are used pri-
marily to study spontaneous or stimulated DA release, rather than
basal [DA]o for two main reasons. First, the rapid sampling time
and small probe size are ideally suited for very localized monitor-
ing of transient changes in extracellular DA concentration ([DA]o).
Second, basal [DA]o is in the low nM range, which is below the
detection limits for most carbon-fiber electrodes. This contrasts
with another popular method for monitoring [DA]o, which is
microdialysis. That method involves extraction of substances, like
DA, from brain extracellular fluid that are then analyzed off-line
using HPLC or other analytical methods. This extraction takes
time, so that microdialysis is not suited for monitoring dynamic
changes in [DA]o, but can be adapted to determine basal [DA]o.
Therefore these methods provide complementary information
about DA transmission. Moreover, other electroactive substances
found in vivo, especially DOPAC, a DA metabolite, and ascorbate,
an abundant extracellular antioxidant, are readily lost from the
extracellular compartment of brain slices, so do not interfere with
evoked DA detection in vitro.
A variety of voltammetric methods have been developed; the
three most commonly used are amperometry, chronoamperome-
try, and fast-scan cyclic voltammetry (FCV) (5). Each involves the
measurement of current generated by the electrochemical oxida-
tion of DA at a suitable recording electrode. The primary focus of
this chapter is to describe methods and methodological issues
required for monitoring DA release in brain slices using FCV and
carbon-fiber microelectrodes. However, the basic methods can be
used to monitor the other primary biogenic amines, serotonin
(5-HT), and norepinephrine (NE).
2. Materials
2.1. Instrumentation 1. Millar Voltammeter (obtained by special request from Dr Julian

for FCV Millar at St. Bartholomew’s and The Royal London School of
Medicine and Dentistry, University of London, UK).
2. Digital storage oscilloscope.
3. Analogue to Digital (A-to-D) converter (e.g., Digidata A-to-D
board, Molecular Devices, Foster City, CA, USA) with Clampex
Software (Molecular Devices) for data acquisition.
4. External timing unit (e.g., Master-8, A.M.P.I., Jerusalem, Israel).
5. Personal computer.
2.2. Manufacture 1. Bundle of unsized 7–8 mm diameter carbon fibers (Goodfellow

of Carbon-Fiber Corporation, Berwyn, PA, USA). The fibers should be stored
Microelectrodes so that they are protected from dirt and dust in the
atmosphere.
2. Borosilicate glass capillaries, 10 cm length, 2.0 mm outer
diameter, 1.16 mm inner diameter (Part No GC200-10;
Harvard Apparatus Ltd. or Part No B200-116-10; Sutter
Instrument Company).
3. Tools for electrode fabrication, including fine forceps modified
by covering each tip with a plastic insulation sleeve to permit
handling of carbon fibers without crushing and scissors for
trimming the carbon fibers to a length of about 15 cm.
4. Clean 13 × 80 mm glass test-tube angled at 30° and held with
modeling clay on a glass slide or other solid surface.
5. Acetone (General grade, Fischer Scientific, Rockford, IL).
6. Vertical microelectrode puller (e.g., Model PE-2, Narishige
Scientific Instruments, Japan).
7. Microscope with 20× eyepieces and 10× objective to view and
cut the carbon fiber extending from the glass pipette after
pulling.
8. Apparatus for cutting carbon-fiber tips, which includes two
micromanipulators set at right angles to one another; one for
holding the carbon-fiber electrode and the other to hold a
mounted scalpel blade.
9. Woods metal (Goodfellow, Oakdale, PA), electrically conduc-
tive silver paint (e.g., Silver Print II, GC Electronics, Rockford,
IL), silver epoxy (MG Chemicals, Surrey, BC, Canada), or 4 M
K-acetate/0.15 M KCl solution for electrical contact between
the carbon fiber and a connection wire.
10. Insulated electrical stranded copper wire (1 mm diameter)
(e.g., Extraflexible Miniature Vinyl Insulated Lead Wire, 26–28
gauge, Cooner Wire, Chatsworth, CA).
11. Gold-plated connector pin (1 mm) (e.g., model WC1-10,

Harvard Apparatus, Holliston, MA) or small alligator clip to
connect the electrode wire to the voltammeter headstage.
2.3. Testing and 1. Dopamine (Sigma-Aldrich, St. Louis, MO) stock solution
Calibrating Carbon- (2 mM in 0.1 M perchloric acid, Fisher).
Fiber Microelectrodes 2. 0.1 M nitric acid solution for weekly washing of glassware; this
minimizes DA oxidation in solution. Be sure to wash out the
acid afterwards with deionized water.
3. Bicarbonate-buffered artificial cerebrospinal fluid (aCSF) con-
taining 124 mM NaCl, 3.7 mM KCl, 26 mM NaHCO3,
1.3 mM MgSO4, 1.3 mM KH2PO4, 10 mM glucose, 2.4 mM
CaCl2 and saturated with 95% O2/5% CO2 (Sigma-Aldrich).
4. Uncoated silver wire to make Ag/AgCl reference electrodes
(0.015 in. bare Ag wire; A–M Systems, Sequm, WA); 9 V bat-
tery and 0.1 M KCl solution for chloride plating.
2.4. Preparation 1. Vibratome (e.g., Ted Pellar, Inc., Redding, CA) or a vibrating
of Brain Slices blade microtome (e.g., VT1200S, Leica Microsystems, Buffalo
Grove, IL).
2. Platinum-edge razor blades (Gillette, Boston, MA).
3. Appropriate stereotaxic brain atlas for rodent species examined.
4. Dissection tools including scalpel blade, rongeurs, sharp scissors,
fine iris scissors, fine forceps (e.g., Fine Science Tools, Foster
City, CA), small spatula, glass Petri dish, and filter paper.
5. Ice-cold HEPES-bicarbonate-buffered artificial cerebrospinal
fluid (aCSF) containing in mM NaCl (120), KCl (5), NaHCO3
(20), HEPES acid (6.7), HEPES salt (3.3), MgSO4 (2), glu-
cose (10), CaCl2 (2) and saturated with 95% O2/5% CO2.
6. Brain slice holding chamber consisting of a 15 cm disposable
Petri dish with lid (Fisher). The chamber is filled with HEPES-
buffered aCSF at room temperature and receives a constant
supply of 95% O2/5% CO2, which gently bubbles the aCSF via
a 1 mm plastic tube inserted through a small hole made in the
cover or lid.
2.5. Stimulating 1. Anti-vibration table (e.g., TMC, Peabody, MA) or a solid surface.
and Recording DA 2. Dissection microscope (e.g., Olympus, Center Valley, PA).
Release
3. Brain slice submersion recording chamber (model RC-26G
with series 20 P-1 chamber platform, Warner Instruments from
Harvard Apparatus, Holliston, MA).
4. Peristaltic pump (Minipuls 3, Gilson, Middleton, WI) for
controlled perfusion of the brain slice chamber with aCSF.
5. Inline heater (SH-27B, Warner Instruments) with an auto-
matic temperature controller (TC-324B, Warner Instruments,
Hamden, CT).
6. Fiber optic cold-light source for slice illumination (e.g., Cuda

Products Co., Jacksonville, FL).
7. Two micromanipulators for placement of the carbon fiber
recording electrode and stimulating electrode (e.g., Narishige
from Tritech Research, Los Angeles, CA).
8. Bipolar platinum or platinum/iridium stimulating electrode
(twisted-wire or concentric) (e.g., Fredrick Haer & Co. (FHC),
Bowdoinham, ME).
9. Stimulus isolator (Isoflex, A.M.P.I) with stimulus pulse dura-
tion and pulse-train frequency controlled by the Master-8 pulse
generator (A.M.P.I).
3. Methods
3.1. Basic Principles “Fast-scan cyclic voltammetry” is a very descriptive name for this
and Instrumentation method to monitor the release of biogenic amines. The term “fast”
for FCV refers to the scan rate, typically 300–900 V/s. The standard volt-
age waveform is triangular (Fig. 2a), beginning with a positive
ramp from a negative starting potential (we use −0.7 V vs. Ag/
Ag/Cl) to a potential that exceeds that required to oxidize DA (we
scan to +1.3 V vs. Ag/AgCl), followed by a negative going ramp
that returns to the starting potential, passing the potential required to
reduce just-oxidized DA en route. At a scan rate of 800 V/s (7, 8),
each voltage cycle, or “scan” takes ~5 ms (Fig. 2). Scans are
repeated continually throughout an experiment, typically at 100 ms
intervals (or 10 Hz), which maintains a stable “background” current,
discussed further below. The Millar Voltammeter takes the recording
electrode out of circuit between scans, whereas other voltamme-
ters, like the Ensman EI400, keep the electrode at a negative
“holding” potential. Taking the electrode out of circuit helps
maintain electrode sensitivity during prolonged recording, whereas
a negative holding potential may enhance sensitivity by promoting
biogenic amine adsorption, however, this may also slow the clearance
phase release records.
The last term, “voltammetry” refers to any method in which a
voltage is applied and current is measured. In the case of DA, each
molecule produces two electrons during oxidation, resulting in
current flow through the conductive carbon-fiber microelectrode
that is proportional to the concentration of DA in a calibration
solution or released into the extracellular space of brain tissue.
It should be noted that other key transmitters, glutamate, GABA,
and acetylcholine (ACh) are not electroactive. During the positive
voltage scan, current begins to flow as the DA oxidation potential
is approached (Fig. 2b). Once a potential has been reached that is
sufficient to oxidize all DA molecules at the electrode surface, the
Oxidation Reduction
a +1.5
(V vs. Ag/AgCl)
Eapp
/s
0V
0
80
–1.0
b
25 nA
c
500 nA
Eapp
(V vs. Ag/AgCl) –0.7 1.3 –0.7
Time (ms) 0 2.5 5 7.5 10
Fig. 2. Fast-scan cyclic voltammetry (FCV) voltage waveform and dopamine (DA) voltam-
mograms. (a) Typical triangular voltage waveform applied to a carbon-fiber microelec-
trode with FCV. (b) Voltammogram for 5 mM DA obtained by subtracting the background
current in aCSF alone from that obtained in the presence of DA in panel C. Electrode
sensitivity to DA in this example was 7 nA/mM; oxidation and reduction peak potentials for
DA were +580 mV and −190 mV vs. Ag/AgCl, respectively. (c) Background current gener-
ated at a carbon-fiber microelectrode with FCV in the absence (solid-line) and presence
(broken line) of 5 mM DA.
local concentration of unoxidized DA is rapidly depleted, resulting

in a current peak that is seen at the DA oxidation potential (typi-
cally ~0.6 V vs. Ag/AgCl) (Fig. 2b). A peak is seen with FCV but
not with slow voltammetric scans because the scan rates used are
faster than the rate of replenishment of DA from spherical diffu-
sion to the electrode surface is depleted. When the voltage reverses,
oxidized DA is then reduced, resulting in a reduction current peak
(typically ~0.2 V vs. Ag/AgCl) (Fig 2b). The oxidation and reduc-
tion peak potentials are characteristic for DA (and other cate-
cholamines) and can therefore be used to identify it as the monitored
substance, at least in a DA-rich region like dorsal striatum (6, 9, 10).
The characteristics of rapid monitoring and signal identification

have made FCV the method of choice for many in vitro as well as
in vivo studies of endogenous DA release. There are some caveats
to the use of this method, as well, however.
1. There is a high non-Faradaic background current in all FCV
scans (11, 12), which is directly proportional to scan rate. The
Faradaic (DA) oxidation current, on the other hand, is propor-
tional to the square root of scan rate (13).
2. The background current is usually ~1 mA (Fig. 2c), whereas a
typically 1 mM increase in extracellular DA concentration
([DA]o) generates oxidation currents that are in the low nA range
at most carbon-fiber microelectrodes. Therefore this large back-
ground must be subtracted (Fig. 2c). This introduces a challenge
in the analysis of FCV data, which can be handled either by the
voltammetric instrument used (e.g., the Millar voltammeter) or
by post hoc analysis, as discussed further below.
Voltammetric instrumentation required for FCV can be pur-
chased made in-house by a good electronics workshop using pub-
lished circuit diagrams (14) or generated digitally. There are a few
commercially available instruments that are specifically designed
for FCV, including the Millar Voltammeter (e.g., (6–8)) and the
Ensman EI 400 (most recently available through ESA, now part of
Dionex, but discontinued). Alternatively, voltammetric waveform
generation as well as data acquisition and analysis can be carried
out on a PC-based system using two multifunction data acquisition
cards and software written in LabView (National Instruments,
Austin, TX, USA) (15). The Millar voltammeter has a convenient
feature (a “refresh” button) that stores a background voltammo-
gram immediately before initiating a calibration or stimulation,
which is then subtracted from subsequent scans, and allows direct
monitoring of DA voltammograms.
A schematic diagram showing the basic instrumentation used in
our laboratory for determination of electrically evoked DA release
using FCV with a Millar voltammeter is shown in Fig. 3. The Millar
voltammeter consists of a function generator (which produces a con-
tinuous triangular voltage waveform), a waveform gate (which dis-
sects the voltage waveform into discrete segments), a potentiostat
(which applies the voltage waveform to the recording electrode and
measures the current generated), and amplifiers (to amplify the cur-
rent output from the recording electrode). The voltammeter can be
used in three-electrode mode with a platinum wire auxiliary elec-
trode (6, 10), but is commonly used in two-electrode mode, with
only a carbon-fiber microelectrode and Ag/AgCl reference. The
third auxiliary electrode provides feedback about current flow dur-
ing an electrochemical measurement; this is used to maintain the
applied potential when large current flow might otherwise lead to a
voltage drop at the time of an oxidation peak. However, this is not a
Oscilloscope
Millar Voltammeter Solution

Microscope Pump
Master-8
Pulse Generator Heater
Stimulus
Isolator
aCSF
A-to-D Interface
Computer
Cold-Light
Source
Fig. 3. Schematic diagram showing the basic instrumentation for detection of dopamine release using FCV in a brain slice.
A brain slice is continuously superfused with oxygenated aCSF at 32°C (arrows) via a peristaltic solution pump and in-line
heater. The slice is illuminated by a cold light-source and the carbon-fiber recording electrode and bipolar stimulating
electrodes positioned in a DA-rich region of the slice using two micromanipulators under microscopic visualization. A Millar
voltammeter is triggered continually by a Master-8 pulse generator to apply a triangular voltage waveform every 100 ms
via a headstage to the carbon-fiber microelectrode. The background non-Faradaic current detected at the surface of
the electrode (versus a Ag/AgCl electrode located in aCSF near the slice) is amplified by the headstage and fed back to the
voltammeter for background subtraction. Electrical stimulation is applied via a stimulus isolator controlled by the Master-8.
Release of DA can also be evoked from DA somata, dendrites, and/or axons that express channelrhodopsin (e.g., 41). In
that case, the Master-8 would be used to control the duration and frequency of light pulses from an LED light source or a
laser and fiber optic cable. When the slice is stimulated to release DA, the Faradaic current increases at the characteristic
DA oxidation and reduction peak potentials. The triangular voltage waveform and background subtracted voltammogram are
both monitored on a digital storage oscilloscope and also sent to an analogue to digital (A-to-D) interface connected to a
computer for data capture and off-line analysis.
concern for the usual nA–mA currents generated at carbon-fiber

microelectrodes, eliminating the need for an auxiliary electrode.
To monitor the time course of a DA release response, current
detected at the DA oxidation peak potential can be followed using
a sample-and-hold circuit (as a difference signal, this can be
amplified as needed. Signal identification, however, requires evalu-
ation of background-subtracted voltammograms of the released
substance (Fig. 2c). This can be done off-line and is accomplished
using a digital storage oscilloscope or data capture software with
digital subtraction capabilities that permits identification of the DA
signal (9, 10, 12). Labview-driven software also permits off-line sub-
traction of all scans. As already noted, however, the Millar
Voltammeter contains circuitry that stores a background scan which
is then subtracted from subsequent scans (i.e., analogue subtrac-
tion), so the Faradaic voltammogram corresponding to an increase
in [DA]o can be monitored in real time.
3.2. Fabrication The earliest electrodes developed for the detection of DA release in
of Carbon-Fiber brain tissue were made using carbon paste. These were superseded
Microelectrodes by carbon-fiber microelectrodes (16), which are now the electrode
of choice in almost all in vitro or in vivo voltammetric DA release
studies. A number of methods for the fabrication of carbon-fiber
microelectrodes have been described (12, 16–22). In all of these,
the fundamental design consideration is to get the highest signal-
to-noise ratio and lowest detection limit possible. The basic proce-
dure involves inserting a carbon fiber, ~7 mm in diameter, in a glass
capillary tube, then pulling the filled capillary on a conventional
microelectrode puller to provide a seal and insulate the fiber with
glass (Fig. 4). Every attempt is made to ensure that the seal is tight,
Fig. 4. Photographic summary of steps to make a cylindrical carbon-fiber microelectrode. (a) Loading: a glass capillary is
placed in an angled test tube containing a few mL of acetone. A 7 mm carbon fiber is selected, cut, then loaded into the
acetone-filled capillary. (b) Pulling: after the acetone has been removed and the loaded capillary allowed to air dry for
several hours, the capillary is then placed in an electrode puller, with heat and magnet settings optimized to form a good
seal between the glass and the carbon fiber, without burning the fiber. (c) Cutting: the pulled electrode is placed in an
electrode holder in a micromanipulator oriented perpendicular to another manipulator with an electrode holder equipped
with a scalpel blade. The optimal length of fiber extending beyond the glass seal is 30–70 mm. (d) Tip of a cylindrical
carbon-fiber microelectrode after cutting; the junction at which the glass insulation ends us indicated by a white arrow ;
the fiber extends ~40 mm beyond this junction (each small division is 10 mm). (e) Connecting: in this completed electrode,
Woods metal was used to make electrical contact between the carbon fiber and the insulated lead wire that will be used
to link the electrode to the headstage of a voltammeter. This figure is in color in the on-line version only.
as small gaps or cracks can lead to background noise as solution is

drawn into them by capillary action.
The design of carbon-fiber microelectrodes used for brain slice
recording must include an active surface that is completely con-
tained within a £400 mm thick slice. Such surfaces can be prepared
by beveling or cutting or by chemical or electrical etching of the
carbon-fiber tip. The resulting working surface can either be a disk
that is flush with surrounding insulation, a slightly extended fiber
with a conical shape, or a cylinder that extends <70 mm beyond the
insulation (5). Longer electrodes work well in vivo, but are not
practical for use in slices. Methods for making the short-cylinder
electrodes we use are given below.
One final issue in making carbon-fiber microelectrodes is how
to make electrical contact between a typically 5–10 mm carbon fiber
and the electrode lead that will make contact with the instrumenta-
tion required for voltammetric recording. Poor contact with the
fiber can be a significant source of noise. Materials used include elec-
trically conducting silver paint or epoxy, Woods metal or conductive
solutions like potassium acetate mixed with KCl (12, 16, 20, 22),
each of which can link the fiber to a larger diameter wire that is then
used for connection to the voltammeter headstage. The manufac-
ture of carbon-fiber microelectrodes in the authors’ laboratory are
based on the procedures described in refs. (16, 20). We make several
electrodes in a given session although not all of the electrodes made
will end up being suitable for experimental use. Electrodes are care-
fully stored in a box to protect them from mechanical or atmo-
spheric damage at all stages after construction.
3.2.1. Loading 1. Loading a capillary tube with a carbon fiber is most easily
accomplished when the tube is filled with acetone. The acetone
serves two purposes; first it facilitates insertion of the carbon
fiber into the capillary and second it removes any coatings or
residue that may have formed on the carbon fiber surface.
Using a Pasteur pipette, add a few mLs of acetone to a glass
test tube £8 cm long. This should be held at a ~30° angle from
horizontal, e.g., with modeling clay (Fig. 4a).
2. Place a single capillary in the test tube; it should extend ~2 cm
beyond the end of the test tube (Fig. 4a). The acetone should
immediately fill the glass barrel by capillary action.
3. With the capillary still in the test tube, use fine forceps with
insulated tips to extract a single carbon fiber from the fiber
bundle and gently insert it into the acetone-filled capillary
about 1 cm at a time. A low power microscope or magnifying
glass may be useful for this stage.
4. Remove the loaded capillary from the acetone, then touch one
end to a tissue to remove the acetone from the capillary (with-
out removing the fiber). As soon as the acetone is removed,
the carbon fiber will adhere to the inside wall of the capillary
which will prevent the fiber from falling out. The protruding
ends of the carbon fiber are then cut to about 1 cm beyond the
ends of the glass capillary with fine scissors. Any remaining
acetone should be allowed to evaporate for at least a few hours,
but usually overnight, before pulling.
3.2.2. Pulling 1. Place the carbon-fiber filled capillary into an electrode puller
(Fig. 4b). The combination of heat and magnet settings must
be determined empirically for each puller and adjusted as the
heating coil ages or when it is replaced. The ideal settings make
a taper of 1–2 cm that seals the carbon fiber tightly. With a
good seal, it is difficult to see where the glass ends and bare
fiber begins. If the heat is too high, this can burn the fiber,
which becomes wavy. If the heat is too low, the seal may not be
good. Too little pull from the magnet may make the taper too
short, whereas too strong a pull can result in a flare at the end
of the glass where the seal should be.
2. When puller settings have been optimized, the capillary will be
pulled into two halves with each tapered around the carbon
fiber. The high tensile strength of the carbon fiber usually pre-
vents it from breaking when the halves separate. Cut the fiber
between the glass tapers to form two separate electrodes.
Depending on the puller, the fiber may seal in only one half.
3.2.3. Cutting 1. The optimal length for slice recording is ~50 mm. Cutting the
fiber is a precise and delicate process that should be done using
a high-powered binocular microscope (100× to 200× total
magnification).
2. Position two micromanipulators at right angles to one another
to enable fine positioning of the carbon-fiber microelectrode
and the cutting device (e.g., a mounted scalpel blade) (Fig. 4c).
Both should be adjusted to advance a microelectrode holder
with the electrode or cutting blade in a roughly horizontal
plane to reach the visual field of the microscope. Attach the
blade to one holder and the electrode to the other.
3. Visually inspect the glass seal, which should be tight and almost
invisible (Fig. 4d); if not the electrode will likely be noisy. Then
check that a sufficient length of the fiber protrudes back into the
shaft of the capillary to enable electrical contact to be made.
4. Lower the electrode onto a glass slide so that the tip lightly
rests on the surface of the slide. Using the edge of the mounted
scalpel blade cut the carbon fiber to a length of 30–70 mm
from the glass seal (Fig. 4d). In theory the second half of the
pulled capillary can also be used but in practice this is rare and
is usually because of a poor seal or that the fiber preferentially
ends up in one half.
3.2.4. Connecting Electrical contact between the carbon fiber and the headstage of
the voltammeter can be made using insulated, multi-stranded copper,
or other flexible wire with ~1 mm total diameter that is secured to
the fiber with a conductive sealant (Fig. 4e).
1. Cut the wire to a length of ~15 cm; this should be long enough
to connect to the headstage, but short enough to minimize
picking up electrical noise in the system which would be
amplified by the headstage. Strip ~1 cm of insulation from one
end of the wire and <0.5 cm from the other.
2. Solder a 1 mm gold pin to the shorter end for connection to
the voltammeter headstage. This connection can be insulated
with heat-shrink tubing if desired (Fig. 4e).
3. Dip the longer stripped end of the wire into electrically con-
ductive silver paint or silver epoxy and push in as far as possible
into the shank of the electrode to make contact with the
carbon fiber.
4. Alternatively, low-melting point Woods metal can be used to
make electrical contact. For this method, melting the metal in
a beaker on a hot plate and drawing it into a plastic tube can
make a Woods metal plug, 1 mm in diameter. Drop a plug that
is a few millimeters in length into the open end of the electrode
and then push in the stripped wire as in step 2. The metal plug
is melted using a heat gun or hair dryer to connect the wire
with the fiber (see Note 1).
3.3. Testing and Each carbon-fiber microelectrode must be screened for quality.
Calibrating Carbon- This involves ensuring that the electrode has a suitable and stable
Fiber Microelectrodes background current profile, low electrical noise, and adequate sen-
sitivity to detect low mM DA concentrations during calibration. We
typically test and calibrate our electrodes in aCSF in the brain slice-
recording chamber at the same temperature (~32°C) used during
our experiment. With experience, one can determine very quickly
whether an electrode will be viable for reliable recordings in brain
tissue. It is best to test and calibrate electrodes in the aCSF solu-
tion used for brain slice recordings rather than in standard phos-
phate-buffered saline (PBS) because the sensitivity to DA can differ
considerably between media. For example the absence of divalent
Ca2+ and Mg2+ ions in PBS leads to a twofold higher peak DA oxi-
dation current than that seen in aCSF for a given DA concentra-
tion (23). This presumably is because divalent ions associate with
anionic functionalities (e.g., carboxylic acid) on the carbon surface
thereby decreasing DA adsorption and thus oxidation current. As
a consequence, calibration in PBS or other standard buffers that
lack these divalent ions will lead to an underestimate of [DA]o in
the brain microenvironment, which includes Ca2+ and Mg2+.
3.3.1. Setting Up 1. Prepare standard aCSF, continually bubbled with 95% O2/5%
Calibration in the Brain CO2. Use a peristaltic pump to flow the aCSF through a tem-
Slice Chamber perature controller into the brain slice recording chamber;
chamber temperature should be set to ~32°C.
2. Anodize half of a 4–5 cm length of silver wire in 0.1 M KCl,
using a 9 V battery to make the Ag/AgCl reference electrode.
Attach the wire to be anodized to the (+) pole of the battery
and another Ag (or Pt) wire to the (−) pole. A smooth, light
gray AgCl coating is optimal. When this becomes black (Ag2O)
and/or begins to flake off, it is time to remake.
3. For two-electrode recording with a Millar Voltammeter, make a
short connection between the reference and auxiliary electrode
pins on the headstage before connecting a reference or carbon
fiber microelectrode to the headstage.
4. Secure the Ag/AgCl reference electrode in the corner of the
chamber and connect it (using a small alligator clip attached to
a flexible, insulated wire) to the appropriate input socket of the
voltammeter headstage. Make sure that the electrode and clip
do not touch any other metal surface.
5. Using a micromanipulator equipped with a microelectrode
holder, lower a carbon-fiber electrode into the superfusing
aCSF. Turn on the voltammeter and then connect the electrode
to the working electrode socket of the voltammeter headstage
to avoid an electrical surge that could damage the headstage or
the electrode.
6. Switch on the scan trigger (or initiate control of scans from a
digital timing circuit, e.g., Master-8) and monitor the voltage
and current waveforms on a digital storage oscilloscope. Ensure
that the starting potential and the reversal potential of the tri-
angle voltage waveform are appropriate. Adjust the current
gain so that the background current signal monitored in “Full”
mode is just below the point at which the amplifiers become
saturated, which can be seen when the current waveform is
flattened at the top or bottom, even though the oscilloscope
scale is appropriate.
7. Assess the shape of the background current. A carbon fiber in
contact with an electrolyte behaves electrically as a resistor–
capacitor (RC) network (24) and is further influenced by the
electrolytic composition of the test solution because of surface
functionalities already discussed. If the shape of the back-
ground current is rounded, much as one would expect for
charging a simple RC circuit, the electrode will likely have little
sensitivity for DA and should be discarded. The shape of the
background current for a “good” electrode will have well-
defined additional features, including a broad peak at around
0.3 V vs. Ag/Ag/Cl and some evidence of water oxidation
near the switching potential of the input waveform (Fig. 2c).

These features usually indicate a good surface for DA oxida-
tion and predict good electrode sensitivity.
8. Assess the stability of the background current. A slow drift in
the shape of the background current over a period of a few
hours has a negligible effect, whereas large changes occurring
over a few minutes usually indicates that the electrode is unreli-
able. Large changes in background current are often caused by
a poor glass-carbon seal which allows electrolytes to creep up
into the electrode shaft.
9. Assess electrical noise. Sufficient grounding should minimize
the inherent noise of the apparatus and environment when
the rig is set up and throughout an experiment. Electrical noise
associated with an electrode should be minimal, as well. This
can be assessed by switching to “Faradaic” mode, in which dif-
ferences from a background scan can be readily seen and
quantified in comparison to a built in 10 nA calibration pulse.
If a calibration pulse is not available with the instrumentation
used, this can be assessed from the current axis of voltammo-
grams monitored on a PC with appropriate data acquisition
software. Noisy electrodes are usually the result of poor electri-
cal contact between the connecting wire and the carbon fiber
or a poor glass-carbon seal. If the level of electrical noise is
unacceptable (typically >1–2 nA) the electrode should be
discarded.
3.3.2. Calibrating 1. Position the electrode at the superfusing inlet of the chamber
so that the carbon-fiber tip is directly in the incoming flow-
stream. When the background-subtracted current is stable,
switch the flow to aCSF with a known concentration of DA
(e.g., 1 mM) in aCSF, remembering to refresh the background
current immediately before the DA solution comes in contact
with the electrode surface. It is important to match the rate of
bubbling of the calibration solution with that of the back-
ground aCSF or background shifts from differences in solution
pH (e.g., (12)) can distort the DA voltammogram. The appear-
ance of a single oxidation peak (at ~+600 mV) and a single
reduction peak (at ~−200 mV) can be monitored on the oscil-
loscope. The time that the DA is in contact with the electrode
should be recorded using a stop-watch and limited to a set
time between 30 and 60 s, after which the Faradaic signal
should be electronically recorded for analysis and the solution
changed back to control aCSF.
2. Several DA concentrations can be used to confirm the linearity
of the current response over the range of [DA]o expected in a
brain slice.
3. The detection limit can be defined as the concentration at

which the Faradaic current is twice that of the background
noise. Electrodes made according to these protocols have a
detection limit of ~30–50 nM.
4. Although preexperiment calibration is necessary to assess elec-
trode quality and sensitivity, calibration factors determined
after the electrode has been in brain tissue are the most reliable
for quantification of evoked [DA]o during an experiment.
Typical postexperiment calibration factors are 5–20 nA/mM.
5. If the composition of aCSF will be changed in an experiment
or if a new drug is to be used, the effect of these changes on
electrode sensitivity to DA must be assessed. Whether a new
drug is electroactive must also be determined. This is a three-
step process. First, determine the calibration factor for 1 mM
DA in aCSF alone. Second, switch the calibration solution
back to aCSF alone until the background current is again sta-
ble, then switch to aCSF plus drug at the concentration to be
used experimentally; be sure to “refresh” the scan immediately
before drug-aCSF enters the recording chamber. Oxidation
peaks if the drug is electroactive or non-Faradaic current
changes (e.g., from electrode filming) can interfere with DA
detection and may preclude the use of the drug. Third, to
assess whether the presence of a drug alters electrode sensitiv-
ity to DA, refresh the background again in the continued pres-
ence of drug-aCSF and switch to drug-aCSF plus 1 mM DA. If
the sensitivity of the electrode to DA is lost in the presence of
the drug, this again precludes the use of the drug for FCV
experiments. If the drug causes a consistent increase or decrease
in the calibration factor from that in aCSF alone, then separate
calibration factors to quantify evoked release in aCSF alone
versus in drug can be used, although quantitation may be com-
promised. This should only be done if a similar drug that does
not interfere with electrode sensitivity is not available.
3.4. Preparation Optimal conditions for preparing and maintaining viable brain
of Brain Slices slices have been developed over many years and are discussed in a
comprehensive book edited by Dingledine (25). A few basic points
relevant for midbrain and forebrain slices are summarized here.
1. Slices are made from deeply anesthetized animals after decapi-
tation and cut while submerged in ice-cold aCSF or modified
aCSF. For example, using an oxygenated HEPES-bicarbonate-
buffered aCSF for slice preparation (26) and during the hold-
ing period (7, 8) improves slice quality by minimizing water
gain (edema) (27).
2. Slice quality is best when slices are cut using a classic Vibratome
or a Leica vibrating microtome (Leica Microsystems, Buffalo
Grove, IL). Both instruments separate tissue into slices using a
plane of turbulent solution that precedes a vibrating blade,

thereby minimizing compression injury to the tissue.
3. Slice thickness should not exceed 400 mm, which is the maximum
at which the core of the slice can still receive adequate oxygen
and glucose. On the other hand, the minimal thickness of
acute slices is typically considered to be 200 mm, which pro-
vides sufficient viable tissue between the cut surfaces, where
there are damaged cells and severed processes.
4. Forebrain and midbrain slices are most commonly cut in the
coronal plane (Fig. 1). However, horizontal, sagittal, or paras-
agittal slices can be cut in order to preserve a few millimeters of
the DA pathway itself (28) or a non-DA input for stimulation
purposes. For example horizontal slices from the small mouse
brain can preserve DA connections between the midbrain and
ventral striatum such that spontaneous [DA]o transients can be
detected (29).
5. Once prepared, viable slices can be maintained in a holding
chamber for 8–10 h at room temperature, although slice via-
bility may not be as robust with older animals and/or in some
experimental models (e.g., after DA system lesion or genetic
manipulation)
3.4.1. Before Anesthetizing 1. Make sure all surgical equipment required (scalpel and razor
an Animal blades, small scissors, iris scissors, small rongeurs, etc.) is available
and clean.
2. aCSF (e.g., HEPES-bicarbonate buffered aCSF) is prepared
and chilled on ice to ~2°C (allow 30 min for 200 mL in a
beaker).
3. Clean a razor blade and the blade for the vibrating microtome
with acetone, then ethanol to remove greasy lubricant manu-
facturers usually use to prevent rusting. This can be done by
sequential immersion in each solvent. The solvents can be
reused if capped to prevent evaporation, but should be changed
at least bi-weekly.
4. Optimize the settings on the vibrating microtome for cutting
the brain region of interest. These settings can vary among spe-
cies. Parameters include the angle of the blade, the speed at
which the blade advances, and amplitude at which the blade
vibrates. A general rule is that slower advancement and higher
vibration amplitude minimizes tissue damage, with the caveat
that taking too long to slice can compromise slice viability.
5. Prepare a tissue holding chamber (e.g., a large plastic Petri
dish) containing the desired aCSF at room temperature, with
gentle bubbling of 95% O2/5% CO2 that is sufficient to main-
tain oxygenation and proper pH (7.4–7.6), but not so vigor-
ous that the tissue moves around the container.
3.4.2. Dissecting 1. Anesthetize an animal with an appropriate anesthetic until

pedal reflexes are absent (e.g., sodium pentobarbital, 50 mg/
Kg/i.p).
2. Using a scalpel, make an incision along the midline of the scalp
from neck to snout of the anesthetized animal. Pull back the
skin flaps and remove underlying connective tissue to expose
the skull surface. As soon as the skull is exposed, use a Pasteur
pipette to bathe it with a few milliliters of ice-cold oxygenated
HEPES-buffered aCSF.
3. Insert the lower blade of sharp, pointed scissors beneath the bone
at the back of the skull and make a short cut along the midline.
Bathe exposed tissue with cold aCSF. Remove the bone on
each side of the cut using fine rongeurs and repeat the proce-
dure until all the bone posterior to Bregma is removed. Tips of
scissors and rongeurs should be pointed upwards to prevent
tissue damage. Indeed, care is more important than speed for
this and all other steps, although efficiency will come with
experience.
4. Cut and remove the dural membrane using fine iris scissors and
fine forceps and continue removing bone until olfactory bulbs
are revealed.
5. Gently invert the skull and then release the brain into a beaker
of ice-cold oxygenated aCSF by carefully inserting a small spat-
ula between the brain and the base of the skull to sever the
cranial and optic nerves.
6. Allow the brain to chill 1–2 min to minimize on-going metab-
olism and the consequences of anoxia. Chilling procedure also
makes the brain firmer for blocking and cutting.
3.4.3. Blocking and Slicing 1. To cut coronal slices, remove the chilled brain from the beaker
of aCSF and place it ventral side-up on an inverted glass Petri
dish (or other surface) covered with filter paper soaked with
ice-cold aCSF.
2. Remove any remaining membranes that could snag the blade
during slicing. Using a cleaned razor blade cut a block of tissue
that is 3–7 mm in anterior-to-posterior length that incorpo-
rates the brain region of interest (e.g., striatum in the forebrain
or substantia nigra in the midbrain).
3. Affix one cut surface of the tissue block to the specimen plate
of the tissue slicer using cyanoacrylate glue (Krazy glue). For
forebrain slices, glue the posterior surface so that slices are cut
from anterior-to-posterior locations; orient the block so that
the dorsal surface faces the blade for dorsal-to-ventral slicing.
For midbrain slices, glue the cut anterior surface so that slices
are cut from posterior-to-anterior locations; orient the mid-
brain block so that the ventral surface faces the blade for ven-
tral-to-dorsal slicing. These procedures have been found
empirically to produce the best slices for voltammetric

recording.
4. Fit the specimen plate into the buffer well of the tissue slicer as
quickly as possible after gluing, and fill the buffer well with
oxygenated ice-cold aCSF.
5. Attach a clean slicing blade to the vibrating microtome.
6. Position the starting point of the blade to allow ~500 mm of
tissue to be cut from the block to leave a flat, horizontal plane
for subsequent slicing; discard this section.
7. Begin slicing of tissue sections of the desired thickness (200–
400 mm). Using a small spatula or a glass pipette, carefully
transfer each slice as it is made to the tissue holding chamber.
The number of slices prepared also depends on the region of
interest and the species examined. For example, more slices can
be made from striatum than from the smaller midbrain DA cell
body regions, and more from either region using a guinea pig
than the smaller mouse brain.
8. Allow at least 1 h for recovery in the holding chamber at room
temperature before experimentation.
3.5. Stimulating and In the striatum, in vivo, rapid, spontaneous [DA]o transients can be
Recording DA Release detected using FCV, which reflect axonal DA release in response to
in Brain Slices presumed bursting of DA neurons in the midbrain (30). Most
in vitro brain slices preparations, however, contain either DA axons
(forebrain slices) or DA neurons (midbrain slices), so such sponta-
neous transients are not seen in striatal slices. Moreover, DA neu-
rons in vitro display steady pacemaker activity rather than bursting
behavior (31), so that somatodendritic [DA]o transients are also
not seen in slices. However, axonal and somatodendritic DA release
can be readily evoked using local electrical stimulation (10, 32–36)
both of which are sensitive to Na+ channel blockade, indicating
action-potential dependence, and are Ca2+ dependent. An insulated
bipolar stimulating electrode is used to evoke release, which can be
either concentric or parallel in style. Stimulating electrodes are
typically made from tungsten, platinum, or platinum–iridium and
can be constructed locally or purchased from a commercial supplier.
For example, an effective twisted wire electrode can be constructed
from insulated narrow gauge wire (e.g., 75 mm diameter wire only),
with tip separation of ~100 mm.
Methods in this section will focus on local electrical stimula-
tion, but the basic procedures for voltammetric recording in slices
can be used with other stimulation methods, as well. For example,
superfusion or local pressure ejection of depolarizing agents, like
high K+ or veratridine, can also be use to elicit DA release (21, 26,
37), with the caveat that these can cause large background shifts
from changes in extracellular Ca2+ and pH. The effect of other
releasing agents like amphetamine can also be examined using FCV
(38–40). Most recently, optogenetic techniques that permit selective
activation of neuronal populations have been introduced and

applied to study DA transmission. This methodology involves
transfecting DA neurons with photo-sensitive channelrhodopsin
using a viral vector injected into the SNc or VTA of DAT-Cre or
TH-Cre transgenic mice. Once the virus has fully penetrated and
has been transported throughout the DA neuron, including axon
terminals (usually after a few weeks), slices are prepared as usual
and a blue laser or LED is used to elicit DA release selectively (41).
3.5.1. Brain Slice Recording For voltammetric recording of DA release using carbon-fiber
microelectrodes in brain slices, it is preferable to use a submersion
chamber in which slices (and electrode tips) can be submerged in
aCSF at all times, rather than an interface chamber in which the
slice surface is exposed to a humidified atmosphere of 95% O2/5%
CO2 into which the electrode must be drawn after recording.
Keeping the carbon-fiber microelectrode in solution at all times
when not in tissue helps maintain a constant electrode surface state
and thereby optimal reliability of electrochemical measurements.
Background currents are also kept as constant as possible by con-
tinuous triggering of the FCV voltage waveform from the time the
electrode is first positioned in the chamber for preexperiment cali-
bration until the end of the experimental day. A few basic points
about brain slice recording in a submersion chamber are summa-
rized here.
1. Slices are superfused at a rapid but constant rate (typically
>1–2 mL/min) with freshly prepared oxygenated bicarbonate-
buffered aCSF.
2. To maintain slice viability, aCSF used for in vitro brain slice
recording differs from CSF in vivo in that it is hyperoxic, when
equilibrated with 95% O2/5% CO2, and hyperglycemic, with
10 mM glucose. Both conditions are necessary to provide
sufficient O2 and glucose to maintain the viability of cells below
the slice surface, as superficial cells consume these metabolic
substrates. To preserve O2 and CO2 saturation of aCSF en route
to a slice, the inlet tubing should be made of gas-impermeable
material, although this is not always done.
3. Tubing should also be narrow and kept as short as practical to
ensure rapid delivery of oxygenated aCSF. This also facilitates
efficient solution changes for drug studies as well as for elec-
trode calibration.
4. The slice recording environment is slightly hypothermic, typi-
cally ~32°C, which helps preserve viability for several hours of
recording, and also decreases temperature-dependent DA
uptake via the DA transporter (DAT), enhancing [DA]o dur-
ing stimulation (6).
5. The aCSF used for recording usually has slightly elevated con-
centrations of K+ and/or Ca2+ compared to those in CSF in vivo
to enhance slice excitability and Ca2+-dependent DA release.
3.5.2. Brain Slice 1. Once the brain slice recording chamber is set up with aCSF
Orientation and Electrode (at ~32°C) flowing through at a rate of at least 1.2 mL/min
Placement and the Ag/AgCl reference electrode secured in place, as
described in Subheading 3.3, carefully transfer a suitable slice
from the holding chamber to the recording chamber using a
spoon-like spatula or a glass pipette with an open tip greater than
the dimensions of the slice. Keep the slice immersed in aCSF and
flat at all times during transfer to prevent mechanical trauma.
2. Orient the slice in the recording chamber to allow optimal
placement of the stimulating and recording electrodes (more
below). Secure the slice by placing 3 or 4 small lengths
(~2–4 mm) of bare silver wire around the edges of the slice to
hold it down without blocking the flow of aCSF over the slice
surface or interfering with electrode placement.
3. Position a fiber optic (cold) light source below the slice to
allow transillumination of the tissue for viewing with a dissection
microscope. Under these illumination conditions gray matter
appears translucent, while the white matter containing myeli-
nated fiber bundles appears dark.
4. Connect the stimulating electrode to a stimulus isolator con-
trolled by a Master-8 or other pulse-timing device. The use of
a stimulus isolator avoids interference with the ground of the
voltammeter.
5. Allow 20–30 min for equilibration with the recording medium
before stimulating the tissue. The voltammetric electrode can
be positioned in the tissue at any time during this equilibration
period to facilitate background current stabilization; scan trig-
ger is on. Insert the electrode tip 50–100 mm below the slice
surface under the control of a micromanipulator. The stimulat-
ing electrode can be positioned using a micromanipulator at
this time as well, with the electrode tip(s) just resting on the
surface of the slice. An appropriate rodent stereotaxic atlas can
be used for guidance in optimal electrode orientation, although
this must be confirmed empirically.
6. To monitor evoked axonal DA release in the striatal complex
(Fig. 1b), the carbon-fiber microelectrode is typically posi-
tioned ~100 mm dorsolateral to the stimulating electrode. Both
recording and stimulating electrodes should be placed so as to
avoid the non-dopaminergic myelinated fiber bundles that run
through the striatum. In the dorsolateral (motor) striatum and
nucleus accumbens shell and core, which can all be contained
in a coronal plane of section (see Fig. 1b), reproducible DA
release can be evoked in the same site for several hours at
2–5 min intervals for single-pulse stimulation or 5–10 min for
pulse-train stimulation. Single-pulse stimulation is ~1 mM in
the guinea-pig dorsal striatum but is lower in the NAc core and
even lower in the NAc shell.
7. To monitor somatodendritic DA release in the SNc and VTA,

optimal release is usually seen when the recording electrode is
positioned between the poles of a non-concentric bipolar stim-
ulated electrode. The location of these cell body regions at
similar midbrain levels make it possible to record from both
regions in a single coronal slice (see Fig. 1b). Moreover, the
lateral projecting dendrites of SNc DA neurons lie in this plane.
In the SNc, both poles of the stimulating electrode should be
placed within this band of DA somata, with the carbon-fiber
tip equidistant between them. Note that this band is not pig-
mented in rodents (as it is in humans), so that a brain atlas can
be very useful for proper electrode placement. Detection of
DA release provides confirmation of location. In both SNc and
VTA multiple-pulse stimulation is usually required to detect
DA release, which is typically much smaller (<0.5 mM) than
that evoked in dorsal striatum.
3.5.3. Stimulating 1. After equilibration in the recording chamber, check the shape
and Recording Axonal DA and stability of the background current. The shape should be
Release (Forebrain Slice) similar to that seen in calibration. Stability is indicated by little
drift in background-subtracted scans monitored on a digital
oscilloscope or data acquisition program over a few seconds,
i.e., the usual recording time for DA release measurements.
2. Set stimulus pulse duration on the timing system (e.g.,
Master-8); this is often 0.1 ms, but should ideally be £1 ms to
remain shorter than the duration of action potentials initiated
by each pulse. The same pulse duration should be used
throughout an experimental series, and ideally used for both
single-pulse and pulse-train stimulation in a given experiment.
3. The timing of stimulus pulse should be synchronized with the
on-going FCV scan trigger, but adjusted to avoid “collision”
of a stimulus pulse with an FCV scan. This can be achieved by
delaying the stimulus trigger by 7–8 ms after initiation of a
scan. This brief delay also can allow a stimulus artifact to be
captured with each voltammogram to ensure that the stimulus
was turned on, etc. This works perfectly for single-pulse or
10-Hz stimulation when FCV sampling interval is 100 ms, but
may need adjustment for other pulse-train frequencies.
4. Set the stimulus isolator to apply current pulses. Although
either current or voltage can be used, current pulses provide
more consistent stimulation than voltage pulses because of
possible voltage drop at the electrode surface from tissue build-
up after exposure to slices over many days of experimentation.
5. Using the data acquisition software described (see Subheading
3.3), set parameters to capture the input voltage waveform and
output current for each voltammetric scan. For single-pulse
stimulation or pulse trains of up to 30 pulses, set the timing
circuit to trigger data acquisition to provide ~2 s of baseline

records before the stimulus pulse or train is triggered. Allow a
~9 s time window for data acquisition; with a scan trigger rate
of 10 scans/s, this records 90 voltammograms. Excess baseline
or recovery data can be omitted when making figures from
release records, as appropriate.
6. Using single-pulse stimulation, determine the lowest stimulus
intensity (amplitude) that gives the highest peak [DA]o.
Determination of this “perimaximal” stimulus intensity is best
done in dorsolateral striatum, which has the largest single-pulse
evoked peak [DA]o of any dopaminergic region. Be sure to
“refresh” the background immediately before stimulation and
turn on the 10 nA calibration pulse. Start with a low current
amplitude, e.g., 0.3 mA. With most stimulating electrodes,
0.3 mA will be subthreshold for initiating action potential-
dependent DA release. Increment this sequentially using 0.05–
0.1 mA steps. As long as the intensity is subthreshold (i.e., no
DA release is evoked), another measurement can be taken
immediately. Once DA release is seen, however, wait 5 min
before examining the next current increment. Usually after
2–5 more increments, no further increase in peak [DA]o is
seen. Lower the current setting to the first setting that gave
this maximal [DA]o. The rationale is that supramaximal stimu-
lation can mask regulatory processes that might decrease DA
release, e.g., from DA D2 autoreceptors that activate a hyper-
polarizing K+ channel. Moreover, larger currents can cause
electrolytic damage to the tissue (6) (see Note 2).
7. Single-pulse stimulation can be sampled at different sites of the
slice or can be sampled at regular intervals at the same site.
Peak [DA]o evoked by the first stimulus pulse in a slice is usu-
ally higher than subsequent pulses. However, this is usually
constant (±10%) after 2–3 stimulations repeated at a constant
interval of 4 or 5 min. Shorter interstimulus intervals may take
slightly longer to become constant. However, this characteris-
tic can be used deliberately to compare DA release “sustain-
ability” among groups (42). Regardless of the stimulus-interval,
release in some slices or recording sites can result in a progres-
sive decrement in peak [DA]o that never becomes constant.
Conducting experiments in these sites should be avoided.
Single-pulse stimulation evokes DA release that is independent
of regulation by transmitters released by the pulse, including
glutamate and GABA, as well as DA itself (10, 32, 33, 43, 44).
This is because the DA release event has already occurred by
the time receptor-dependent regulation might have influenced
it. One caveat is that in the striatal complex single-pulse DA
release is influenced by ACh release from cholinergic interneu-
rons (29, 45). Single-pulse stimulation is well-suited for
comparison of absolute DA release levels and/or DA uptake

between groups of transgenic mice (e.g., (40, 42, 46)), deter-
mination of whether a drug has a direct effect on axonal DA
release (e.g., nicotine, (29, 45), or to compare D2 autorecep-
tor sensitivity among groups (40, 46). For example, to assess
D2 autoreceptor sensitivity, single-pulse stimulation is repeated
until 5 records with peak evoked [DA]o that differ by £10%
have been obtained, then the superfusing aCSF switched to
aCSF plus a low concentration of a selective D2 agonist, e.g.,
quinpirole. Once a maximal effect on peak [DA]o is seen (~30–
40 min), a slightly higher concentration of the drug is applied
and so on until responses in several concentrations are obtained.
The efficacy of the drug (i.e., concentration of quinpirole at
which a half maximal effect is seen can then be extracted from
the sigmoidal concentration–response curve (40, 46).
8. Pulse-train stimulation. The number of pulses in the train and
the frequency depend on the questions asked. With shorter
pulse trains (4–5 pulses), a 5 min interval between stimulation
is usually sufficient to maintain stable release after the first 2–3
stimulations. With longer trains (10–50 pulses), a 10 min
interval is required. The simplest use of pulse-train stimulation
is to assess the frequency dependence of DA release using con-
stant pulse number and varying frequency in a range from 5 to
100 Hz (32, 33, 45, 46). Pulse-train stimulation can be used
to assess how endogenous transmitters, including DA, gluta-
mate, and GABA, influence DA released by subsequent pulses
in a train; this contrasts with single-pulse evoked [DA]o which
is independent of regulation by concurrently released transmit-
ters. For this analysis, train stimulation with constant pulse
number and frequency (e.g., 30 pulses at 10 Hz) is repeated
until three records with peak evoked [DA]o that differ by £10%
have been obtained, then the superfusing aCSF switched to
aCSF plus a selective antagonist, e.g., for glutamatergic AMPA
receptors (28, 47), and the effect on peak [DA]o assessed after
a maximal effect is seen (30–60 min) (see Note 3).
9. Regional dependence of single-pulse and pulse-train DA release
characteristics. The amplitude of single-pulse evoked [DA]o
varies among striatal subregions with highest peak single-pulse
[DA]o in dorsal striatum, intermediate in NAc core, and lowest
in NAc shell (Fig. 5a–c). The shape of the release response also
differs among these regions. In dorsal striatum, peak [DA]o is
seen within 2–3 stimulus pulses, then declines during contin-
ued stimulation because of D2 DA autoreceptor activation and
DAT-dependent uptake (Fig. 5a). By contrast, in NAc shell, a
progressive increase in [DA]o is usually seen throughout a pulse
train (Fig. 5c), so that although peak single-pulse [DA]o is much
lower in NAc shell than in dorsal striatum, peak pulse-train
a Dorsal striatum
b NAc core
c NAc shell
2.4
1.4 0.6
1.2 2.0
0.5
[DA]o (µM)
[DA]o (µM)
[DA]o (µM)
1.0 1.6
0.4
0.8 1.2
0.3
0.6
0.2 0.8
0.4
0.2 0.1 0.4
0 0 0
10 Hz, 3 s 10 Hz, 3 s 10 Hz, 3 s
1P 1P 1P
d 0.6 SNc
e 0.6 VTA
0.5 0.5
[DA]o (µM)
[DA]o (µM)
0.4 0.4
0.3 0.3
0.2 0.2
0.1 0.1
0 0
10 Hz, 3 s 10 Hz, 3 s
Fig. 5. Comparison of DA release records obtained using FCV in different regions within forebrain and midbrain slices. (a–c)
Representative [DA]o versus time records showing axonal DA release evoked by single-pulse stimulation (1P; gray traces)
or pulse-train stimulation (30 pulses at 10 Hz; black traces) in the mouse dorsal striatum (CPu) (a), NAc core (b), and NAc
shell (c) in a forebrain slice. (d–e) Representative [DA]o versus time records showing somatodendritic DA release evoked
by pulse-train stimulation (30 pulses at 10 Hz) in the SNc and VTA in a guinea-pig midbrain slice.
[DA]o can be similar in these regions. The response in NAc

core is intermediate between dorsal striatum and NAc shell
(Fig. 5b), reflecting mixed DA input from SNc and VTA (48).
10. FCV can also be used to examine DA release in other axon-
terminal regions, like the olfactory tubercle (33), amygdala
(49, 50), prefrontal cortex (51), and subthalamic nucleus (52).
The small size of some of these structures as well as sparse DA
innervation and significant NE input can present challenges,
however. Confirmation that the monitored substance is DA
rather than NE contributions must be made using pharmaco-
logical and anatomical methods to complement voltammetric
data (49).
3.5.4. Stimulating and Initial steps for stimulating and recording somatodendritic DA
Recording Somatodendritic release in the SNc and VTA are identical to steps 1–5 for assessing
DA Release (Midbrain axonal DA release in the previous section. Step 6, determining
Slice) perimaximal stimulation intensity, is also important for assessing
DA release from somata and dendrites. However, the typically
small amplitude of single-pulse evoked [DA]o in these regions
(~100 nM) make this tricky to assess. Consequently, stimulus
parameters determined in dorsal striatum with the electrode
orientation for midbrain recording can be used as a starting point.
It should be noted that guinea pigs are the species of choice for
studies of electrically evoked somatodendritic DA release in the

SNc because FCV measurements detect only DA, whereas stimula-
tion of the SNc in rats and mice evokes concurrent release of 5-HT
(26, 53–55). In VTA, however, DA is the only electroactive sub-
stance detected in the VTA of rats and mice (34, 55) as well as
guinea pigs (35, 54). On the other hand, DA release in SNc is
entirely somatodendritic, based on anatomical studies showing no
synaptic DA input, whereas somatodendritic DA release in the
VTA is mixed with synaptic DA release from its own axon collater-
als as well as synaptic input from SNc (36).
1. Single-pulse stimulation. Like axonal DA release in the striatal
complex, single-pulse evoked somatodendritic DA release in
the midbrain is free from regulation by concurrently released
transmitters (44). However, as already noted, midbrain DA
release levels are often much lower than in striatum. To improve
detection of single-pulse-evoked DA release in midbrain, [DA]o
can be amplified by using a stimulus pulse duration of 1 ms,
blocking D2 DA autoreceptors and inhibiting DA uptake
(36, 44), with the caveat that this may also alter DA neuron
excitability.
2. Pulse-train stimulation. As a consequence of the low levels of
evoked [DA]o in SNc and VTA without pharmacological
manipulation, pulse-train stimulation is most commonly used
for studies of somatodendritic DA release (34–36, 54–57).
Higher peak [DA]o is seen with longer trains (e.g., 30 pulses),
so these are often used. The concentration–time profiles of
pulse-train evoked [DA]o differs again from axonal release in
forebrain structures (Fig. 5d, e). Indeed, [DA]o typically con-
tinues to rise after the stimulation train ends, consistent with
limited DA uptake in these regions compared to striatum (57).
Perhaps as a consequence of lower reuptake, somatodendritic
DA release is not as robust as axonal release in striatum, with a
progressive decrease in evoked [DA]o seen with repetitive stim-
ulation. However, DA release regulation in the SNc can still be
assessed by recording evoked [DA]o in 2–3 sites, then compar-
ing these with a drug response in paired sites in the opposite
hemisphere. Although this paradigm lacks the advantages of
same-site controls, including requiring larger sample sizes
because of greater variability, valuable information can still be
obtained (8, 44, 56). Release in the VTA can be elicited repeti-
tively in the same site (34), possibly reflecting the mix of axonal
and somatodendritic release in this region, although paired
measurements can also be used effectively in VTA. As in striatal
regions, the use of selective receptor antagonists can be used to
indicate the influence of concurrently release glutamate and
GABA on somatodendritic DA release in both SNc and VTA
(8, 56) (see Note 4).
3.5.5. Data Analysis 1. The simplest form to data analysis is to express data as absolute
levels of peak [DA]o obtained by converting the current detected
into a DA concentration using the post-experimental calibration
factor obtained for a given electrode. This is particularly useful for
comparing release levels between brain regions in the same spe-
cies, release in the same brain region between species, or release in
the same brain region between a transgenic and non-transgenic
mouse.
2. To examine the effect of a drug on evoked DA release, data are
often expressed as a percentage of the control response, where
control responses are set as 100%. This analysis is also useful to
compare if a drug alters the time course of the response or if
the profile differs between brain regions.
3. Data can also be expressed as a ratio of peak [DA]o evoked by
a pulse-train stimulation relative to that evoked by a single-
pulse stimulation when they are evoked in the same recording
site. This analysis can provide an indication of the sensitivity of
release to tonic versus phasic stimulation and can vary through-
out the striatal complex (e.g., (10, 33)) or in the presence of
drugs including D2 autoreceptor antagonists (e.g., (10, 33))
or drugs that act at striatal nicotinic receptors (45, 46).
4. Evoked [DA]o data can be used to evaluate the kinetics of DA
release and uptake in axon terminal and cell body regions.
A number of studies determine Michaelis-Menten uptake
parameters using appropriate curve-fitting procedures (42, 59).
In this analysis, evoked DA release (using electrical stimula-
tion) is assumed to provide a homogeneous increase in extra-
cellular concentration, such that diffusion plays no role in DA
clearance. Typically, the Michaelis-Menten constant, Km, which
indicates the [DA]o at which uptake is half-maximal, is set from
literature values with experimental assessment of the maximal
rate, Vmax, and the increase in [DA]o per stimulus pulse. This
method can be used to examine regional or species variation in
DA dynamics, to examine differences in transgenic versus non-
transgenic mice or to characterize the effect of drugs acting at
the DAT, like cocaine or amphetamine.
4. Notes
1. Individual users may chose to use a fresh electrode for each

experimental day because of loss of sensitivity and possible
increase in noise. However, we tend to re-use ours if they are
still in good condition and the response kinetics of the surface
has not deteriorated. Given the initial loss of sensitivity that
can occur upon initial electrode exposure to brain tissue, the
calibration factor of re-used electrodes can be relatively stable.
After use in tissue, the surface can be washed; further washing

can take place when water is run through the chamber to
remove aCSF after an experiment. Optimal methods are user
dependent and should be determined empirically.
2. Perimaximal stimulation intensity should be assessed initially
for every slice tested, because this can vary with the distance
between stimulating and working electrode, as well as record-
ing electrode depth, etc. Once an experimenter is proficient
and consistent in electrode placement, perimaximal intensity
will be consistent, as well. Nevertheless, this should still be
assessed regularly, because the stimulus electrode surface and
pole spacing can change over time, influencing current density
in the surrounding tissue. Perimaximal stimulus intensity
should also be assessed whenever a new stimulating electrode
is used, a new slice thickness is used or when a slice from a dif-
ferent species is used. Another important test when setting up
a new voltammetry system is to confirm that the stimulus used
produces action-potential dependent DA release using
0.1–1 mM tetrodotoxin (TTX) (6, 36). Large gauge stimulat-
ing electrodes (³150 mm diameter) can produce TTX-insensitive
release (35), which is not ideal. If the distance between the
stimulating and recording electrode is too large, current levels
³1 mA may be required to elicit DA release, which as men-
tioned above can lesion to the tissue and result in a progressive
decrease in peak evoked [DA]o (35). If the distance is too
short, however, the stimulation voltammogram may be domi-
nated by stimulation artifact (i.e., large background current
shifts that do not correspond to DA oxidation or reduction
peaks). Background-dominated shifts can also be seen when
the carbon-fiber microelectrode tip is too close to one pole of
the stimulating electrode.
3. The basic procedure for assessing the effects of receptor ago-
nists or antagonists can be used with either single- or multiple-
pulse stimulation to address the effect of any drug or other
agent on DA release. Ideally, the effects observed should
reverse upon drug washout for ~30 min, unless a drug is hydro-
phobic or is known to have an irreversible action.
4. One final point about voltammetric recordings is that back-
ground shifts during stimulation can also arise from non-
electroactive interferents, especially Ca2+ and pH (H+), which
can change dynamically with depolarizing stimuli (see ref. (5),
for review). The potentials of these background shifts can read-
ily be distinguished from the DA oxidation peak at ~+0.6 V vs.
Ag/AgCl) (12, 58); however, this can interfere with quantita-
tion in regions with low evoked [DA]o like the amygdala (58)
as well as SNc and VTA. If necessary a method that uses princi-
pal component regression coupled with appropriate calibration
data can be used to resolve substances that give overlapping

voltammetric waveforms, including DA and pH (15).
Acknowledgments
We are grateful for support from NIH/NINDS grant NS036362

and the Attilio and Olympia Ricciardi Research Fund (M.E.R). We
also appreciate critical reading of the manuscript by Melissa
A. Stouffer and Harry Xenias.
References
1. Dahlström A, Fuxe K (1964) Evidence of the 10. Patel J, Trout SJ, Kruk ZL (1992) Regional
existence of monoamine-containing neurons in differences in evoked dopamine efflux in brain
the central nervous system. I: demonstration of slices of rat anterior and posterior caudate puta-
monoamines in the cell bodies of brain stem men. Naunyn Schmiedebergs Arch Pharmacol
neurons. Acta Physiol Scand 62:1–55 346:267–276
2. Carlsson A (2002) Treatment of Parkinson’s 11. Armstrong-James M, Millar J, Kruk ZL (1980)
with L-DOPA. The early discovery phase, and a Quantification of noradrenaline iontophoresis.
comment on current problems. J Neural Nature 288:181–183
Transm 109:777–787 12. Rice ME, Nicholson C (1989) Measurement of
3. Evans AH, Lees AJ (2004) Dopamine dysregu- nanomolar dopamine diffusion using low-noise
lation syndrome in Parkinson’s disease. Curr perfluorinated ionomer coated carbon fiber
Opin Neurol 17:393–398 microelectrodes and high-speed cyclic voltam-
4. Cannon CM, Bseikri MR (2004) Is dopamine metry. Anal Chem 61:1805–1810
required for natural reward? Physiol Behav 13. Adams RN (1969) Electrochemistry at solid
81:741–748 electrodes. Marcel Dekker, Inc., New York,
5. Patel JC, Rice ME (2006) Dopamine in brain p 124
slices. In: Grimes CA, Dickey EC, Pishko MV 14. Millar J, Barnett T (1988) Basic instrumenta-
(eds) Encyclopedia of sensors, vol 6. American tion for fast cyclic voltammetry. J Neurosci
Scientific Publishers, Stevenson Ranch, Methods 25:91–95
California, USA, pp 313–334 15. Heien ML, Johnson MA, Wightman RM
6. Bull DR, Palij P, Sheehan MJ, Millar J, Stamford (2004) Resolving neurotransmitters detected
JA, Kruk ZL, Humphrey PP (1990) Application by fast-scan cyclic voltammetry. Anal Chem
of fast cyclic voltammetry to measurement of 76:5697–5704
electrically evoked dopamine overflow from 16. Armstrong-James M, Millar J (1979) Carbon
brain slices in vitro. J Neurosci Methods fibre microelectrodes. J Neurosci Methods
32:37–44 1:279–287
7. Chen BT, Avshalumov MV, Rice ME (2001) 17. Kawagoe KT, Zimmerman JB, Wightman RM
H2O2 is a novel, endogenous modulator of syn- (1993) Principles of voltammetry and micro-
aptic dopamine release. J Neurophysiol electrode surface states. J Neurosci Methods
85:2468–2476 49:225–240
8. Patel JC, Witkovsky P, Avshalumov MV, Rice 18. Cahill PS, Walker QD, Finnegan JM, Mickelson
ME (2009) Mobilization of calcium from intra- GE, Travis ER, Wightman RM (1996)
cellular stores facilitates somatodendritic dop- Microelectrodes for the measurement of cate-
amine release. J Neurosci 29:6568–6579 cholamines in biological systems. Anal Chem
9. Millar J, Stamford JA, Kruk ZL, Wightman 68:3180–3186
RM (1985) Electrochemical, pharmacological 19. Koh DS, Hille B (1999) Rapid fabrication of
and electrophysiological evidence of rapid dop- plastic-insulated carbon-fiber electrodes for
amine release and removal in the rat caudate micro-amperometry. J Neurosci Methods
nucleus following electrical stimulation of the 88:83–91
median forebrain bundle. Eur J Pharmacol 20. Millar J, Pelling CW (2001) Improved meth-
109:341–348 ods for construction of carbon fibre electrodes
for extracellular spike recording. J Neurosci 33. Trout SJ, Kruk ZL (1992) Differences in
Methods 110:1–8 evoked dopamine efflux in rat caudate puta-
21. Gerhardt GA, Hoffman AF (2001) Effects of men, nucleus accumbens and tuberculum olfac-
recording media composition on the responses torium in the absence of uptake inhibition:
of Nafion-coated carbon fiber microelectrodes influence of autoreceptors. Br J Pharmacol
measured using high-speed chronoamperome- 106:452–458
try. J Neurosci Methods 109:13–21 34. Iravani MM, Muscat R, Kruk ZL (1996)
22. Clark JJ, Sandberg SG, Wanat MJ, Gan JO, Comparison of somatodendritic and axon
Horne EA, Hart AS, Akers CA, Parker JG, terminal dopamine release in the ventral teg-
Willuhn I, Martinez V, Evans SB, Stella N, mental area and the nucleus accumbens.
Phillips PE (2010) Chronic microsensors for Neuroscience 70:1025–1037
longitudinal, subsecond dopamine detection in 35. Rice ME, Cragg SJ, Greenfield SA (1997)
behaving animals. Nat Methods 7:126–129 Characteristics of electrically evoked somatoden-
23. Kume-Kick J, Rice ME (1998) Dependence of dritic dopamine release in substantia nigra and
dopamine calibration factors on media Ca2+ and ventral tegmental area in vitro. J Neurophysiol
Mg2+ at carbon-fiber microelectrodes used with 77:853–862
fast-scan cyclic voltammetry. J Neurosci 36. Chen BT, Rice ME (2001) Novel Ca2+ depen-
Methods 84:55–62 dence and time course of somatodendritic dop-
24. Fox K, Armstrong-James M, Millar J (1980) amine release: substantia nigra vs. striatum. J
The electrical characteristics of carbon fibre Neurosci 21:7841–7847
microelectrodes. J Neurosci Methods 3:37–48 37. Elverfors A, Jonason J, Jonason G, Nissbrandt
25. Dingledine R (1984) Brain slices. Plenum H (1997) Effects of drugs interfering with
Publishing Corporation, New York sodium channels and calcium channels on the
26. Rice ME, Richards CD, Nedergaard S, release of endogenous dopamine from super-
Hounsgaard J, Nicholson C, Greenfield SA fused substantia nigra slices. Synapse
(1994) Direct monitoring of dopamine and 26:359–369
5-HT release from substantia nigra and ventral 38. Iravani MM, Kruk ZL (1995) Effects of
tegmental area in vitro. Exp Brain Res 100: amphetamine on carrier-mediated and electri-
395–406 cally stimulated dopamine release in slices of rat
27. MacGregor DG, Chesler M, Rice ME (2001) caudate putamen and nucleus accumbens. J
HEPES prevents edema in rat brain slices. Neurochem 64:1161–1168
Neurosci Lett 303:141–144 39. Schmitz Y, Lee CJ, Schmauss C, Gonon F,
28. Avshalumov MV, Patel JC, Rice ME (2008) Sulzer D (2001) Amphetamine distorts stim-
AMPA receptor-dependent H2O2 generation in ulation-dependent dopamine overflow: effects
striatal medium spiny neurons, but not dop- on D2 autoreceptors, transporters, and syn-
amine axons: one source of a retrograde signal aptic vesicle stores. J Neurosci 21:
that can inhibit dopamine release. J 5916–5924
Neurophysiol 100:1590–1601 40. Patel J, Mooslehner KA, Chan PM, Emson
29. Zhou FM, Liang Y, Dani JA (2001) Endogenous PM, Stamford JA (2003) Presynaptic control
nicotinic cholinergic activity regulates dop- of striatal dopamine neurotransmission in
amine release in the striatum. Nat Neurosci adult vesicular monoamine transporter 2
4:1224–1229 (VMAT2) mutant mice. J Neurochem 85:
898–910
30. Sombers LA, Beyene M, Carelli RM, Wightman
RM (2009) Synaptic overflow of dopamine in 41. Tecuapetla F, Patel JC, Xenias H, English D,
the nucleus accumbens arises from neuronal Tadrosi I, Shah F, Berlin J, Deisseroth K, Rice
activity in the ventral tegmental area. J Neurosci ME, Tepper JM, Koos T (2010) Glutamatergic
29:1735–1742 signaling by mesolimbic dopamine neurons
in the nucleus accumbens. J Neurosci 30:
31. Grace AA, Onn S-P (1989) Morphology and 7105–7110
electrophysiological properties of immunocy-
tochemically identified rat dopamine neurons 42. Li X, Patel JC, Wang J, Avshalumov MV,
recorded in vitro. J Neurosci 9:3463–3481 Nicholson C, Buxbaum JD, Elder GA, Rice
ME, Yue Z (2010) Enhanced striatal dopamine
32. Limberger N, Trout SJ, Kruk ZL, Starke K transmission and motor performance with
(1991) “Real time” measurement of endoge- LRRK2 overexpression in mice is eliminated by
nous dopamine release during short trains of familial Parkinson’s disease mutation G2019S.
pulses in slices of rat neostriatum and nucleus J Neurosci 30:1788–1797
accumbens: role of autoinhibition. Naunyn
Schmiedebergs Arch Pharmacol 344:623–629 43. Phillips PEM, Hancock PJ, Stamford JA
(2002) Time window of autoreceptor-mediated
inhibition of limbic and striatal dopamine the subthalamic nucleus. Eur J Neurosci
release. Synapse 44:15–22 20:1788–1802
44. Chen BT, Moran KA, Avshalumov MV, Rice 53. Iravani MM, Kruk ZL (1997) Real-time mea-
ME (2006) Limited regulation of somatoden- surement of stimulated 5-hydroxytryptamine
dritic dopamine release by voltage-sensitive Ca release in rat substantia nigra pars reticulata
channels contrasted with strong regulation of brain slices. Synapse 25:93–102
axonal dopamine release. J Neurochem 96: 54. Cragg SJ, Hawkey CR, Greenfield SA (1997)
645–655 Comparison of serotonin and dopamine release
45. Rice ME, Cragg SJ (2004) Nicotine amplifies in substantia nigra and ventral tegmental area:
reward-related dopamine signals in striatum. region and species differences. J Neurochem
Nat Neurosci 7:583–584 69:2378–2386
46. Bao L, Patel JC, Walker RH, Shashidharan P, 55. John CE, Budygin EA, Mateo Y, Jones SR
Rice ME (2010) Dysregulation of striatal dop- (2006) Neurochemical characterization of the
amine release in a mouse model of dystonia. J release and uptake of dopamine in ventral teg-
Neurochem 114:1781–1791 mental area and serotonin in substantia nigra of
47. Avshalumov MV, Chen BT, Marshall SP, Peña the mouse. J Neurochem 96:267–282
DM, Rice ME (2003) Glutamate-dependent 56. Chen BT, Rice ME (2002) Synaptic regulation
inhibition of dopamine release in striatum is of somatodendritic dopamine release by gluta-
mediated by a new diffusible messenger, H2O2. mate and GABA differs between substantia
J Neurosci 23:2744–2750 nigra and ventral tegmental area. J Neurochem
48. Voorn P, Vanderschuren LJMJ, Groenewegen 8:158–169
HJ, Robbins TR, Pannartz CMA (2004) 57. Cragg SJ, Rice ME, Greenfield SA (1997)
Putting the spin on the dorsal-ventral divide Heterogeneity of electrically-evoked dopamine
of the striatum. Trends Neurosci 27: release and uptake in substantia nigra, ventral
468–474 tegmental area, and striatum. J Neurophysiol
49. Bull DR, Bakhtiar R, Sheehan MJ (1991) 77:863–873
Characterization of dopamine autoreceptors in 58. Jones SR, Mickelson GE, Collins LB, Kawagoe
the amygdala: a fast cyclic voltammetric study KT, Wightman RM (1994) Interference by pH
in vitro. Neurosci Lett 134:41–44 and Ca2+ ions during measurements of cate-
50. Jones SR, Garris PA, Kilts CD, Wightman RM cholamine release in slices of rat amygdala with
(1995) Comparison of dopamine uptake in the fast-scan cyclic voltammetry. J Neurosci
basolateral amydaloid nucleus, caudate-putamen, Methods 52:1–10
and nucleus accumbens of the rat. J Neurochem 59. Wu Q, Reith ME, Wightman RM, Kawagoe KT,
64:2581–2589 Garris PA (2001) Determination of release and
51. Mundorf ML, Joseph JD, Austin CM, Caron uptake parameters from electrically evoked dop-
MG, Wightman RM (2001) Catecholamine amine dynamics measured by real-time voltam-
release and uptake in the mouse prefrontal cor- metry. J Neurosci Methods 112:119–133
tex. J Neurochem 79:130–142 60. Paxinos G, Watson C (1998) The Rat
52. Cragg SJ, Baufreton J, Xue Y, Bolam JP, Bevan Brain in Stereotaxic Coordinates, 4th edn.
MD (2004) Synaptic release of dopamine in Academic, Inc
Chapter 16
Real-Time Chemical Measurements of Dopamine

Release in the Brain
James G. Roberts*, Leyda Z. Lugo-Morales*, Philip L. Loziuk,
and Leslie A. Sombers
Abstract
Rapid changes in extracellular dopamine concentrations in freely moving or anesthetized rats can be
detected using fast-scan cyclic voltammetry (FSCV). Background-subtracted FSCV is a real-time electro-
chemical technique that can monitor neurochemical transmission in the brain on a subsecond timescale,
while providing chemical information on the analyte. Also, this voltammetric approach allows for the
investigation of the kinetics of release and uptake of molecules in the brain. This chapter describes, com-
pletely, how to make these measurements and the properties of FSCV that make it uniquely suitable for
performing chemical measurements of dopaminergic neurotransmission in vivo.
Key words: Fast scan cyclic voltammetry, In vivo, Electrochemistry, Carbon fiber microelectrode
1. Introduction
Dopamine is a neurotransmitter of particular interest due to its

involvement in motivated behavior and reward-driven learning
(1, 2), as well as various neurological disorders such as Parkinson’s
disease (3), schizophrenia (4), and drug addiction (5, 6).
Extracellular dopamine concentrations in the brain vary on a sub-
second timescale due to phasic firing of dopamine neurons (7).
These naturally occurring dopamine transients are enhanced upon
administration of drugs of abuse (8, 9), and become time-locked
to cues that predict reward availability (1, 9–12). The ability to
characterize and provide real-time measurements of rapidly
* These two authors contributed equally to this work.
275
276 J.G. Roberts et al.
fluctuating dopamine in vivo requires an analytical method with

rapid temporal resolution. To date, the most widely used technique
to monitor in vivo neurotransmitter release is microdialysis (13).
This method provides excellent chemical selectivity and is well
suited for measuring dopamine levels averaged over the course of
minutes to hours. However, microdialysis lacks the temporal reso-
lution to detect phasic dopamine fluctuations that occur on the
subsecond to second timescale (14). In contrast, electrochemical
techniques are especially useful for monitoring rapid chemical
changes resulting from discrete neurochemical events due to rapid
sampling rates (micro-to-millisecond timescale). Broadly speaking,
the techniques that involve current flow at an electrode under
potential control can be divided into two groups: voltammetric
and amperometric methods (15). Of these techniques, fast-scan
cyclic voltammetry (FSCV) provides the best combination of tem-
poral resolution, sensitivity, and chemical selectivity—features that
are essential for the detection of rapid neurotransmitter fluctuations
in vivo (16). Additionally, FSCV has been combined with other
techniques including microinjection (7), iontophoresis (12, 17),
and electrophysiology (12, 18, 19) to provide a great deal of new
information regarding dopamine and its role in the brain.
1.1. Use of FSCV in In FSCV, a dynamic potential is applied to a carbon fiber micro-
Dopamine Detection electrode (20). As the voltage is cycled through a triangular poten-
tial pattern, current is generated and recorded as a function of
potential. The current generated over a single scan is plotted versus
the applied potential. This resulting voltammogram is characteris-
tic of the analyte, allowing it to be distinguished from many other
electroactive species. This principle by which FSCV operates
allows the identification of dopamine by the location of the
potentials at which it oxidizes/reduces and the characteristic peak
shape (Fig. 1). This also allows dopamine to be distinguished
from many other interferents in the brain, such as ascorbic acid
and changes in pH (16).
Although FSCV generates characteristic voltammograms that
can serve as qualitative identifiers for a molecule of interest, selectiv-
ity is always a concern. Several criteria need to be followed in order
to positively confirm the identity of a voltammetric signal (21).
These steps include electrochemical, anatomical, physiological, and
pharmacological verification.
1. The in vivo signal should match the standard voltammogram
collected in vitro.
2. The signal must be collected in an anatomical region known to
be analyte rich.
3. Stimulation of cell bodies should illicit analyte release.
4. Pharmacological manipulation should be used to reduce or/and
increase analyte release.
16 Measurements of Dopamine Release in the Brain 277
Fig. 1. Principles of background subtracted FSCV. (a) First, a potential waveform is applied to a working electrode. (b) This
generates a stable non-faradaic background current. (c) The redox reaction will exchange electrons at the electrode sur-
face, generating faradaic current. (d) At low analyte concentrations, the faradaic response (dashed red line) is small com-
pared to the background current. (e) The stable non-faradaic background current can be subtracted from the faradaic
current arising from the redox reaction. The resulting cyclic voltammogram is an electrochemical fingerprint of the
analyte.
These steps must be taken in order to perform reliable electrochemical

measurements and authenticate the identity of an analyte.
The microelectrodes typically used with FSCV are cylindrically-
shaped carbon-fiber microelectrodes. With this approach, a
5–7 μm diameter carbon fiber electrode is sealed in a glass capil-
lary with a portion of the fiber (75–125 μm) extending from the
tip. This carbon fiber electrode is approximately 40-fold shorter
and 50-fold smaller in diameter than a typical microdialysis
probe, and thus it is particularly well suited to probe brain
regions that have gradations in the density of neuronal terminals
over these dimensions (22). This small size results in minimal
tissue damage during in vivo experiments, allows for character-
ization of specific brain regions, and its cylindrical shape enables
detection from all sides of the electrode by way of hemispherical
diffusion to the recording surface, which enhances sensitivity.
2. Materials
2.1. Electrode 1. Carbon fiber: T-650 (GoodFellow, Huntingdon, England).

Fabrication 2. Borosilicate capillary glass: 0.6 mm O.D., 0.4 mm I.D. for
freely-moving experiments and 1.0 mm O.D., 0.5 mm I.D. for
anesthetized experiments (A-M Systems, Sequim, WA).
3. Vertical electrode puller: PE-21 (Narishige, Tokyo, Japan).
4. Vacuum pump.
5. Optical microscope.
6. Surgical scalpel to cut the carbon fiber.
7. Micromanipulator: for freely-moving experiments (custom
made, UNC-CH Chemistry, Machine Shop).
8. Silver paint: Silver Print II (GC Electronics, Rockford, IL).
9. Heat shrink tubing: EPS-200-1/8″ and FP-301-3/32″ (3 M
Electronics, Austin, TX).
10. NORIT A® activated carbon is used for isopropyl alcohol
purification (MP Biomedicals, LLC, Solon, OH).
11. Silver wire: 0.5 mm diameter for Ag/AgCl reference electrode
(Sigma-Aldrich, St. Louis, MO).
12. Gold connector: PCB socket (Newark Electronics, Chicago,
IL).
13. Insulated leads: 30 gauge (Squires Electronics, Inc., Cornelius,
OR).
2.2. Surgery 1. Anesthetic: xylazine and ketamine for freely-moving experi-

ments, urethane for anesthetized experiments, and 0.25%
bupivacaine is used as a local anesthetic.
2. Heating pads.
3. Stereotaxic frame: such as, Model 900 Small Animal Stereotaxic
(David Kopf Instruments, Tujunga, CA).
4. Guide cannula (Bioanalytical Systems, Inc., West Lafayette, IN).
5. Anchor screws (Gexpro, Indianapolis, IN).
6. Cranioplastic cement: Grip Cement (Dentsply International,
Inc., Milford, DE).
7. Stimulating electrode: 20 mm long bipolar stainless steel
(Plastics One, Roanoke, VA).
2.3. Electrochemistry 1. Multifunction input/output cards: PCI-6251 and PCI-6711

(16 bit, 333 kHz) (National Instruments, Austin, TX).
2. Software for data collection and analysis: TH-1 (ESA,
Chelmsford, MA), or custom written in house using LabVIEW
(National Instruments, Austin, TX).
Fig. 2. Headstage (UNC-CH Electronics Design Facility). A miniaturized current-to-voltage

convertor that consists of (a) operational amplifier, (b) threaded connection to stimulating
electrode, (c) lead for reference electrode, (d) lead for working electrode, (e) DB-25
connector.
3. Potentiostat. One of the following is appropriate: EI-400 bio-

potentiostat (Cypress Systems, Lawrence, KS), Universal
Electrochemistry Instrument (UEI, UNC-Chapel Hill,
Electronics Design Facility), or Universal Headstage Controller
(United World Domination, Mebane, NC).
4. Headstage: miniaturized current-to-voltage converter
(UNC-CH Electronics Design Facility; or United World
Domination, Mebane, NC) (Fig. 2). A larger version can be
used for anesthetized experiments and postexperiment
calibration.
5. Commutator: 25 channel (Crist Instruments, Hagerstown,
MD).
6. Screened behavioral chamber: custom made for in vivo experi-
ments (Med Associates Inc., St. Albans, VT).
7. Optional equipment: TV, DVD-R, and video character gen-
erator (for monitoring, recording, and time-stamping animal
behavior).
2.4. Stimulation 1. Multifunction input/output card: PCI-6711 (National

Instruments, Austin, TX).
2. Bi-phasic stimulus isolator: DS4 (Digitimer, Ltd, Hertfordshire,
England).
2.5. Electrode 1. Dopamine HCl: 1 mM dopamine in 0.1 N HClO4 for stock

Postcalibration solutions, and dilutions are made in buffer (Sigma-Aldrich, St.
Louis, MO).
2. Flow injection apparatus: six-port, two-position high-performance

liquid chromatography (HPLC) valve, with air actuator, and
digital valve interface (VICI, Houston, TX).
3. Grounded Faraday cage: custom built in house.
4. Tris buffer: 3.25 mM KCl, 1.2 mM CaCl2, 1.2 mM MgCl2,
2.0 mM Na2SO4, 1.25 mM NaH2PO4, 145 mM NaCl, and
15 mM Trizma ® HCl at pH 7.4 (Sigma-Aldrich, St. Louis,
MO).
3. Methods
3.1. Electrochemistry, Dopamine release in brain tissue can be monitored in real-time

Instrumentation, with high spatial and temporal resolution when micron-scale elec-
and Software trodes and low-noise instrumentation are implemented. Dopamine
is electrochemically detected at carbon-fiber microelectrodes by
applying a potential sufficient to liberate two electrons from dop-
amine to form dopamine ortho-quinone. This provides a current
that can be converted to a voltage and measured using a current
transducer. Instrumentation includes the Universal Electrochemical
Instrument, the Universal Headstage Controller, or a EI-400 bio-
potentiostat. These instruments are generally used with computer-
controlled interface boards and locally written software (LabVIEW,
National Instruments, Austin, TX). Software is commercially avail-
able from ESA. The instrument provides all inputs and supplies
power to the headstage, and usually consists of two main compo-
nents: a low-pass filter and a headamp module. The computer-
generated waveform contains digitization noise that must be
smoothed by a low-pass filter before the signal reaches the working
electrode. The output voltage from the current transducer is
further amplified and conditioned by the headstage amplifier. The
interface boards are responsible for the digital-to-analog and
analog-to-digital conversions that are transmitted to and from the
headstage, respectively.
The use of FSCV for the electrochemical detection of dop-
amine at carbon-fiber microelectrodes requires a waveform that
optimizes peak currents, response time, and chemical selectivity.
The most commonly used waveform holds the working electrode
at −0.4 V vs. Ag/AgCl with periodic ramping to +1.3 V and back
at a rate of 400 V/s and a frequency of 10 Hz (Fig. 1). The time
between scans when the working electrode is held at a negative
potential allows positively-charged dopamine to preconcentrate
at the electrode surface (23). Due to the fast scan rate, scanning
generates a large capacitive charging current at the electrode sur-
face (24), which is significantly larger than faradaic currents
resulting from redox processes at the microelectrode surface.
These background currents are stable over tens of seconds.

This allows for subtraction, revealing the interesting faradaic
responses. The resulting background-subtracted cyclic voltam-
mograms provide information on the analyte identity, redox
potentials, reversibility, and electron transfer kinetics. The shape
of the peaks allows for the discrimination of multiple species (how-
ever all catecholamines produce similar voltammograms) and can
be used to assess the role of mass transfer. The amplitude of the
peaks can be correlated to the concentration of the analyte at
the electrode surface. Under the conditions described, the cyclic
voltammograms for dopamine should have a peak for the oxida-
tive current at around +0.6 V.
3.2. Bipolar Electrical Electrical stimulation of dopaminergic cell bodies evokes dopamine
Stimulation release from the terminals in a time-locked manner, enabling the
experimenter to monitor the kinetics of dopamine release and
uptake with FSCV (25). The computer controlled stimulation is
delivered with a biphasic stimulus isolator to the stimulating elec-
trode. The device must be calibrated before use to ensure proper
function. The waveform applied to the stimulating electrode is a
biphasic square wave that is applied with a frequency, amplitude,
pulse width, and number of pulses consistent with the experimen-
tal goals. Typical stimulation parameters for dopamine neuron cell
bodies are 125 biphasic pulses, 60 Hz, ±125–150 μA, and 2 ms/
phase. This stimulation must be applied between the ramps of the
electrochemical waveform, such that the electrochemical data is
not disturbed by the current stimulation (Fig. 3).
3.3. Electrode 1. A single carbon fiber is placed on a flat and clean surface that is
Fabrication well illuminated. The fiber is then aspirated into a borosilicate
glass capillary, so that it extends from both ends.
3.3.1. Carbon Fiber
Microelectrode 2. The filled capillary is tapered in an electrode puller. This forms
two electrodes from a single filled capillary. Each is inspected
Fig. 3. Electrical stimulation. The bipolar electrical stimulation (gray), must not overlap
with the applied electrochemical waveform (black ).
Fig. 4. Micromanipulator (UNC-CH Machine shop). An illustration of a loaded micromanipulator, ready for an experiment.
under the microscope to ensure a tight glass seal around the

carbon fiber.
3. The exposed carbon fiber is then cut to length (~100 mm) with
a sharp scalpel under a microscope using a magnification of
10×. The electrode should also be inspected under the micro-
scope using a higher magnification of at least 40× for visible
cracks or abnormalities in the fiber or glass seal, and discarded
if any are present (see Note 1).
4. In freely-moving experiments:
(a) An inspected 100 mm carbon fiber microelectrode is loaded
into a custom micromanipulator and secured with heat
shrink tubing (Fig. 4).
(b) A small diameter insulated wire is painted with silver paint,
and fed into the back of the capillary to make an electrical
connection with the carbon fiber. A slight rotation of the
wire ensures connectivity with the carbon fiber. The wire
is secured to the micromanipulator with additional heat
shrink tubing.
(c) All loaded manipulators are stored with the exposed carbon
in purified and filtered isopropyl alcohol.
5. In anesthetized experiments:
(a) Larger diameter glass capillaries can be used.
(b) The carbon fiber microelectrode is backfilled using a sat-
urated solution of 150 mM potassium chloride and 4 M
potassium acetate. A small diameter insulated wire is fed
into the back of the capillary to make electrical
connection.
3.3.2. Ag/AgCl Reference 1. A piece of silver wire is cut to approximately 10 mm, inserted
Electrode into the socket of a gold connector, and soldered in place.
2. The solder is then covered with quick dry epoxy to avoid the
contact of the soldering material with tissue.
3. On the day of surgery, the reference is chlorinated by connect-
ing the positive terminal of a 2.5 V power supply to the gold
pin on the silver wire and the negative terminal to a wire, with
both leads immersed in 0.1 M hydrochloric acid. Chlorination
is performed for about 1 min until the surface of the silver wire
turns slightly white.
3.4. Surgery 1. The rat is anesthetized with urethane (3 g/kg i.p.), the top of the
head is shaved, and the animal is placed in a stereotaxic frame.
3.4.1. Anesthetized
Preparation 2. The scalp is locally anesthetized with a subcutaneous injection
of 0.25% bupivacaine. An incision is made in the scalp, and the
skin retracted to expose a 15–20 mm longitudinal and
10–15 mm lateral area of cranium.
3. Holes are drilled through the skull for stereotaxic placement of
electrodes (stimulating, working, reference) (Fig. 5). The
stimulating electrode can be positioned either in regions
Fig. 5. A top view illustration of a rat skull, highlighting the general placement of holes
(dotted lines) for electrode and surgical screw placement.
containing dopaminergic cell bodies (substantia nigra/ventral

tegmental area) or at the ascending fibers of the medial fore-
brain bundle. The hole for the working electrode is drilled
above the target terminal region (e.g., 1.3 mm lateral and
1.3 mm rostral from bregma for the caudate-putamen and the
core of the nucleus accumbens, and +1.7 mm anterior and
+0.8 mm lateral for the nucleus accumbens shell). The hole for
the reference electrode is drilled in the contralateral hemi-
sphere, opposite the working electrode.
4. Electrodes are lowered and secured in select areas of the brain
using micromanipulators mounted on the stereotaxic frame.
The reference electrode is secured with cranioplastic cement.
Mild electrical stimulations are applied to evoke neurotrans-
mitter release that is monitored at the microelectrode using
FSCV. The rat is maintained on a heated pad for the duration
of the experiment.
3.4.2. Freely-Moving 1. The rat is anesthetized with intramuscular or intraperitoneal

Preparation ketamine (100 mg/kg) and intramuscular xylazine (10 mg/kg),
the top of the head is shaved, and the animal is placed in a
stereotaxic frame.
2. The scalp is locally anesthetized with a subcutaneous injection
of 0.25% bupivacaine. An incision is made in the scalp, and the
skin retracted to expose a 15–20 mm longitudinal and
10–15 mm lateral area of cranium.
3. Holes are drilled for the working electrode guide cannula,
stimulating and reference electrodes (Fig. 5). In addition,
four holes are drilled at a 45° angle into which anchor screws
are secured.
4. Reference electrode and guide cannula are lowered using
micromanipulators mounted on the stereotaxic frame.
5. Once the components are in place, they are secured with cran-
ioplastic cement, leaving the stimulating electrode hole
exposed.
6. The stimulating electrode is modified in order to provide
adequate space between the plastic hub of the stimulating elec-
trode and the guide cannula (for the working electrode) on the
animal’s head cap (Fig. 6). The stimulating electrode wires are
bent at a 90° angle from the plastic hub and then bent back
down at another 90° angle, to give a horizontal distance of
~5 mm between the hub and the main axis of the wires. Next,
the tips are separated by 0.8–1.0 mm and carefully cut to a
uniform length without disturbing the insulation. Dura mater
is thoroughly cleared and the electrode is stereotaxically low-
ered into the tissue, oriented so that the tips of the electrode
are splayed on the coronal plane. The electrode is lowered to
1 mm above the target brain region.
Fig. 6. Stimulating electrode.
7. The stimulating electrode is connected to the stimulator and a

mild electrical stimulation is applied through the stimulating
electrode. The animal’s tail should respond to this stimulation
by rapidly rising and then slowly falling back to the resting
position. The stimulating electrode is lowered in 0.2 mm incre-
ments until this response is diminished. It is then lowered
further in 0.1 mm increments until this tail response is almost
non-detectable.
8. Finally, cranioplastic cement is applied to the exposed cranium,
carefully covering the stimulating electrode and lower half of
its plastic hub.
9. Immediately following surgery, the animal is placed on a warm
heating pad until fully recovered. Once fully awake, soft food
and fresh water are offered with a fruit-flavored analgesic, such
as acetaminophen (0.1–0.3 g/kg) that the rat will readily lick.
10. The animal is monitored daily and gently handled to facilitate
experimental procedures. While handling, the stylet should be
gently removed from the guide cannula, cleaned with an alco-
hol wipe, and reinserted. Experiments can be conducted within
2–5 days of surgery.
3.5. Freely-Moving Rat Two to five days after surgery, depending on the rat’s postsurgical
Experiment recovery, the animal is prepared for the experiment. The animal is
placed in the behavioral chamber, tethered using the stimulator
3.5.1. Making
cable on which the headstage is mounted (Fig. 2) and allowed to
the Connections
acclimate for about 10 min. Before the loaded micromanipulator is
placed into the guide cannula, the electrode is inspected once again
under a microscope to double check the condition of the seal. The
electrode is retracted inside the micromanipulator as the tip of the
electrode is monitored. Once the electrode tip disappears, each
turn is counted until the electrode is fully retracted. This protects
the electrode integrity as it is loaded into the cannula and allows
the experimenter to index the tip location inside the manipulator.
All connections are cleaned, and the guide cannula stylet is removed
and replaced with the micromanipulator containing the retracted
microelectrode. The manipulator is locked in place and the work-
ing and reference electrodes are connected to the headstage.
3.5.2. Lowering the Carbon The electrode is slowly lowered into tissue as its output is moni-
Fiber Microelectrode tored on an oscilloscope. To do this, the waveform is applied.
As soon as the carbon fiber electrode comes in contact with tissue,
the non-faradaic background current appears, and is monitored
for stability as the electrode is lowered through the tissue (Fig. 7).
Fig. 7. Oscilloscope output. (a) Diagram of applied waveform. (b) Electrode response when
circuit is completed in tissue or buffer.
A break in the electrode is evident by a sudden change in the shape

of the background current to a more resistive profile (approximat-
ing a triangular wave), and it should be removed and replaced with
a fresh carbon fiber electrode. Once in place at the target region,
the electrode is conditioned for about 20 min to stabilize the signal.
Electrochemical conditioning consists of applying the triangular
waveform mentioned above for at least 10 min at a frequency of
60 Hz and then changing it to 10 Hz for 10 additional minutes of
potential cycling.
A mild electrical stimulation is applied to the stimulating bipo-
lar electrode while the current output is monitored at the carbon
fiber microelectrode. If a dopamine signal is not obtained, the
working electrode is lowered in small increments and stimulation
repeated until electrically evoked dopamine release is observed.
The electrode is then secured in position by a locking device on the
micromanipulator and the experiment is initiated (see Note 2).
3.6. Anesthetized Rat 1. Immediately following surgery, the stereotaxic frame is placed
Experiment into the grounded Faraday cage.
2. The electrodes (carbon fiber, stimulating, and Ag/AgCl refer-
ence) are lowered into the appropriate holes using the stereo-
taxic frame. There is no need to use screws and cranioplastic
cement to secure the electrodes in an anesthetized experiment;
however, the reference can be secured in place for stability.
3. The stimulating electrode is connected to the biphasic stimulus
isolator and the working and reference electrodes are con-
nected to the headstage.
4. As described above, the microelectrode is lowered in small
increments (0.1 mm) into a brain region rich in dopamine
terminals.
5. Dopamine neurons are electrically stimulated to illicit dop-
amine release at the terminals in a time-locked fashion (see
Note 3).
3.7. After the Upon completion of the experiment(s), there are two options
Experiment depending on the objective of the experiment and the investiga-
tor’s primary interest. These two options are described below.
3.7.1. Verification of The electrode tip is too small to leave a visible mark in tissue, thus
Electrode Placement an electrical lesion is made at the carbon fiber tip by applying a
high current to the microelectrode. This unequivocally shows the
location of the electrode in the tissue; however, this renders the
electrode useless and it cannot be calibrated. The rat is transcardi-
ally perfused with 0.9% saline and 10% formalin solution to fix
brain tissue. Finally, the animal is decapitated and the brain is
removed from the skull and stored in formalin solution at 4°C,
until it is sliced for histology.
Fig. 8. Flow-injection analysis system. A syringe pump supplies a constant buffer flow
across the working and reference electrodes. An HPLC valve controls the introduction of
an analyte to the working electrode surface.
3.7.2. Electrode Alternatively, the microelectrode is carefully removed from the

Postcalibration brain, replaced with a sacrificial carbon fiber microelectrode, and
an electrical lesion is made as described above. The electrode for
calibration is rinsed in water and calibrated in vitro on a flow-
injection apparatus using known physiological concentrations of
dopamine (usually between 200 and 1,000 nM). This system con-
sists of a custom-made electrochemical cell and a sample loop by
which small volumes of analyte are rapidly injected into the cell
using a six-port HPLC valve and a computer-controlled pneumatic
actuator. A syringe pump is used to continuously supply physiolog-
ical buffer at a constant flow rate through the electrochemical cell
(Fig. 8.). The working electrode is lowered with a micromanipula-
tor into the stream of buffer flowing at 1–3 mL/min. The Ag/
AgCl reference electrode is submerged in the buffer as well and
both are connected to the headstage. The same waveform used for
the in vivo experiment is applied for the calibration of the elec-
trode. Concentrations of the analyte of interest are loaded into the
sample loop and introduced into the electrochemical cell with
the digitally-controlled pneumatic actuator. The injection is soft-
ware controlled. Each concentration of dopamine is sampled at
least in triplicate and the averaged peak oxidative current is plotted
against concentration. The resulting calibration plot is used to
relate the current collected in vivo to corresponding dopamine
concentrations.
3.8. Data Analysis TH-1 (ESA, Chelmsford, MA) software is commercially available
and can be used for data analysis. Additionally, custom software
written with Matlab (MathWorks, Inc., Natick, MA) can mathematically

extract information from chemical data for quantitative analysis. The
current method of multivariate statistical analysis involves the use of
principle component regression (PCR) (16, 28). PCR has the ability
to separate intensity based data into relevant components and noise,
so that noise can be discarded. A training (calibration) set is used
that includes individual cyclic voltammograms for the major spe-
cies (typically dopamine and pH shifts, depending on the local
microcircuitry) at various concentrations. Principle components
that best describe the data are chosen. A principle component can
be described as a vector that passes through the data that includes
the most information. These principle components are then used
to predict unknown concentrations from individual cyclic voltam-
mograms collected in vivo, as long as the unknown concentrations
fit within the training set.
3.9. FSCV Combined FSCV can be combined with more traditional neuroscience tools
with Electrophysiology such as electrophysiology, a technique that uses an electrode to
measure action potentials (12, 18, 29). With this combined
approach, the microelectrode employed for electrochemical detec-
tion is also used to monitor local synaptic activity. Between scans
the holding potential is abbreviated and the electrode is allowed to
float, thereby adopting the potential of its local environment,
which is digitally recorded. The use of this method has allowed
dopamine release to be correlated with changes in the firing of
specific neurons in the vicinity of the electrode, shedding light on
dopamine function in discrete brain microcircuits.
3.10. FSCV Combined The behavioral paradigm of intracranial self-stimulation (ICSS) is

with Intracranial an intensely rewarding experimental model (30). It has also been
Self-Stimulation combined with FSCV (10, 12, 31). In ICSS a stimulating elec-
trode is implanted into a specific brain nucleus and the animal is
taught to use a lever to deliver a mild electrical stimulation to the
chosen region. This serves as a powerful operant reinforcer and is
often used in studies of motivated behavior. The use of ICSS in
combination with FSCV has led to the association of rapid dop-
amine signaling with learned cues (such as audio or visual cues)
that precede an electrical stimulation or reward availability (10, 12).
This technique reveals information on the chemical mechanisms
underlying reward based learning.
3.11. FSCV Microinjection and iontophoresis are two methods that have
and Methods been implemented for administering small quantities of a com-
of Localized pound into a specific region of the brain. While systemic applica-
Pharmacological tion of drugs affects global brain circuitry, localized drug delivery
Manipulation techniques allow the experimenter to pharmacologically manipu-
late one discrete brain region. Microinjection involves the place-
ment of a small needle into the desired brain location, and the
Fig. 9. Iontophoresis probe. Scanning electron micrograph of a five-barrel probe coupling

FSCV with iontophoresis, using a carbon fiber microelectrode. Reprinted with permission
from ref. (36). Copyright 2008 American Chemical Society.
subsequent pressure-driven infusion of a compound. Intracranial

self-administration using microinjection has been used to elucidate
the reinforcing action of specific agents in precise brain nuclei
(32–34). Additionally, microinjection combined with FSCV has
established that dopamine transients recorded in the nucleus
accumbens shell require phasic neuronal activity in the ventral teg-
mental area (7), linking the activity of these two regions.
Iontophoresis can be used to locally apply compounds in an
ionic solution using an applied current (35). When combined with
FSCV (17, 36), capillary barrels are attached to a working elec-
trode to deliver small quantities of a compound into tissue (Fig. 9).
This approach enables drug administration to the same site as the
working electrode. While it has its advantages, iontophoresis is a
largely nonquantitative technique. However, recent advances have
allowed researchers to accurately quantify the amount of drug
delivered during an iontophoretic ejection by the use of an elec-
troosmotic flow marker (17, 36). This marker allows quantitative
analysis by taking into account the variability due to inconsistent
barrel dimensions that affect electroosmosis, which in turn affects
the observed iontophoretic ejection.
3.12. Recent Advances One drawback to the use of FSCV in freely-moving experiments has
been the cable which tethers the animal. The introduction of wire-
less integrated circuits has created new opportunities for studying
dopamine function in freely-moving animals (37). Advantages of
this technology include the ability to perform measurements during
multiple animal social interactions, investigation of more natural
behaviors and more complex environments, and fewer artifacts
introduced during movement of electrical connections. Another
Fig. 10. Microelectrode array. An array of four carbon fiber microelectrodes, with fused
silica insulation, secured with a fixed separation of 250 μm. Reprinted with permission
from ref. 41 Copyright 2010 Elsevier.
recent development has been the incorporation of analog background

subtraction to enable recordings over 30 min time intervals before
distortion of dopamine signals occurs due to background drift (38).
Other developments, such as microelectrode arrays (Fig. 10), allow
multiple electrodes to be used in a single experiment (39–41). This
has opened up the opportunity for researchers to simultaneously
measure dopamine release at spatially discrete brain locations (39,
41). This approach also allows for the simultaneous detection of
multiple signaling agents at various locations (40) and allows for
more representative data to be obtained because of the increased
number of recordings that can be acquired in a given experiment.
Another recent advance has enabled chronic implantation of micro-
electrodes, enabling recording at the same electrode over months,
rather than hours (42, 43). Finally, as an alternative to electrical
stimulation, optogenetics can be used to stimulate specific neuronal
populations using light-activated ion channels (44).
4. Notes
1. Optimal electrode length is determined by instrument limita-

tions and experimental goals.
2. Many experiments monitor naturally-occurring dopamine
fluctuations or transients. These dopamine release events are
evident at some but not all sites that support electrically-evoked

dopamine release (26). Also, after several days of implantation,
the reference electrode may drift by about 0.2 V, requiring the
applied potential to be offset by 0.2 V.
3. Naturally-occurring transient dopamine release events are not
generally detected in anesthetized animals, unless pharmaco-
logically evoked (27).
Acknowledgments
This work was funded in part by grants from the National Institutes
of Health, the National Science Foundation, and NCSU
Department of Chemistry. In addition, we gratefully acknowledge
our coworkers, past and present, for the studies cited in this
review.
References
1. Day JJ, Roitman MF, Wightman RM, Carelli cannabinoid receptor activation. J Neurosci
RM (2007) Associative learning mediates 27:791–795
dynamic shifts in dopamine signaling in the 9. Phillips PE, Stuber GD, Heien ML, Wightman
nucleus accumbens. Nat Neurosci RM, Carelli RM (2003) Subsecond dopamine
10:1020–1028 release promotes cocaine seeking. Nature
2. Schultz W (2007) Behavioral dopamine sig- 422:614–618
nals. Trends Neurosci 30:203–210 10. Owesson-White CA, Cheer JF, Beyene M,
3. Obeso JA, Rodriguez-Oroz MC, Goetz CG, Carelli RM, Wightman RM (2008) Dynamic
Marin C, Kordower JH, Rodriguez M, Hirsch changes in accumbens dopamine correlate with
EC, Farrer M, Schapira AH, Halliday G (2010) learning during intracranial self-stimulation.
Missing pieces in the Parkinson’s disease puz- Proc Natl Acad Sci U S A 105:11957–11962
zle. Nat Med 16:653–661 11. Roitman MF, Stuber GD, Phillips PE,
4. Grace AA (1991) Phasic versus tonic dopamine Wightman RM, Carelli RM (2004) Dopamine
release and the modulation of dopamine sys- operates as a subsecond modulator of food
tem responsivity: a hypothesis for the etiology seeking. J Neurosci 24:1265–1271
of schizophrenia. Neuroscience 41:1–24 12. Cheer JF, Aragona BJ, Heien ML, Seipel AT,
5. Carelli RM, Wightman RM (2004) Functional Carelli RM, Wightman RM (2007)
microcircuitry in the accumbens underlying Coordinated accumbal dopamine release and
drug addiction: insights from real-time signal- neural activity drive goal-directed behavior.
ing during behavior. Curr Opin Neurobiol Neuron 54:237–244
14:763–768 13. Justice JB Jr (1993) Quantitative microdialysis
6. Di Chiara G, Bassareo V (2007) Reward sys- of neurotransmitters. J Neurosci Methods
tem and addiction: what dopamine does and 48:263–276
doesn’t do. Curr Opin Pharmacol 7:69–76 14. Lu Y, Peters JL, Michael AC (1998) Direct
7. Sombers LA, Beyene M, Carelli RM, Wightman comparison of the response of voltammetry
RM (2009) Synaptic overflow of dopamine in and microdialysis to electrically evoked release
the nucleus accumbens arises from neuronal of striatal dopamine. J Neurochem
activity in the ventral tegmental area. J Neurosci 70:584–593
29:1735–1742 15. Michael AC, Borland LM (eds) (2007)
8. Cheer JF, Wassum KM, Sombers LA, Heien Electrochemical methods for neuroscience, vol
ML, Ariansen JL, Aragona BJ, Phillips PE, 1, 1st edn. CRC Press, Boca Raton
Wightman RM (2007) Phasic dopamine 16. Heien ML, Johnson MA, Wightman RM
release evoked by abused substances requires (2004) Resolving neurotransmitters detected
by fast-scan cyclic voltammetry. Anal Chem 29. Williams GV, Millar J (1990) Concentration-
76:5697–5704 dependent actions of stimulated dopamine
17. Herr NR, Belle AM, Daniel KB, Carelli RM, release on neuronal activity in rat striatum.
Wightman RM (2010) Probing presynaptic Neuroscience 39:1–16
regulation of extracellular dopamine with 30. Olds J, Milner P (1954) Positive reinforce-
iontophoresis. ACS Chem Neurosci ment produced by electrical stimulation of sep-
1:627–638 tal area and other regions of rat brain. J Comp
18. Kuhr WG, Wightman RM, Rebec GV (1987) Physiol Psychol 47:419–427
Dopaminergic neurons: simultaneous mea- 31. Garris PA, Kilpatrick M, Bunin MA, Michael
surements of dopamine release and single-unit D, Walker QD, Wightman RM (1999)
activity during stimulation of the medial fore- Dissociation of dopamine release in the nucleus
brain bundle. Brain Res 418:122–128 accumbens from intracranial self-stimulation.
19. Millar J, Stamford JA, Kruk ZL, Wightman Nature 398:67–69
RM (1985) Electrochemical, pharmacological 32. Ikemoto S, Qin M, Liu ZH (2005) The func-
and electrophysiological evidence of rapid tional divide for primary reinforcement of
dopamine release and removal in the rat cau- D-amphetamine lies between the medial and
date nucleus following electrical stimulation of lateral ventral striatum: is the division of the
the median forebrain bundle. Eur J Pharmacol accumbens core, shell, and olfactory tubercle
109:341–348 valid? J Neurosci 25:5061–5065
20. Robinson DL, Hermans A, Seipel AT, 33. Ikemoto S, Qin M, Liu ZH (2006) Primary
Wightman RM (2008) Monitoring rapid reinforcing effects of nicotine are triggered from
chemical communication in the brain. Chem multiple regions both inside and outside the
Rev 108:2554–2584 ventral tegmental area. J Neurosci 26:723–730
21. Phillips PEM, Wightman RM (2003) Critical 34. Ikemoto S, Sharpe LG (2001) A head-attach-
guidelines for validation of the selectivity of in- able device for injecting nanoliter volumes of
vivo chemical microsensors. Trac-Trend Anal drug solutions into brain sites of freely moving
Chem 22:509–514 rats. J Neurosci Methods 110:135–140
22. Arbuthnott GW, Wickens J (2007) Space, time 35. Rebec GV, Bashore TR (1984) Critical issues
and dopamine. Trends Neurosci 30:62–69 in assessing the behavioral effects of amphet-
23. Bath BD, Michael DJ, Trafton BJ, Joseph JD, amine. Neurosci Biobehav Rev 8:153–159
Runnels PL, Wightman RM (2000) Subsecond 36. Herr NR, Kile BM, Carelli RM, Wightman
adsorption and desorption of dopamine at RM (2008) Electroosmotic flow and its contri-
carbon-fiber microelectrodes. Anal Chem bution to iontophoretic delivery. Anal Chem
72:5994–6002 80:8635–8641
24. Bard AJ, Faulkner LR (2001) Electrochemical 37. Garris PA, Ensman R, Poehlman J, Alexander
methods: fundamentals and applications, 2nd A, Langley PE, Sandberg SG, Greco PG,
edn. John Wiley, New York Wightman RM, Rebec GV (2004) Wireless
25. Wightman RM, Amatore C, Engstrom RC, transmission of fast-scan cyclic voltammetry at
Hale PD, Kristensen EW, Kuhr WG, May LJ a carbon-fiber microelectrode: proof of princi-
(1988) Real-time characterization of dopamine ple. J Neurosci Methods 140:103–115
overflow and uptake in the rat striatum. 38. Hermans A, Keithley RB, Kita JM, Sombers
Neuroscience 25:513–523 LA, Wightman RM (2008) Dopamine detec-
26. Wightman RM, Heien ML, Wassum KM, tion with fast-scan cyclic voltammetry used
Sombers LA, Aragona BJ, Khan AS, Ariansen with analog background subtraction. Anal
JL, Cheer JF, Phillips PE, Carelli RM (2007) Chem 80:4040–4048
Dopamine release is heterogeneous within 39. Zachek MK, Park J, Takmakov P, Wightman
microenvironments of the rat nucleus accum- RM, McCarty GS (2010) Microfabricated
bens. Eur J Neurosci 26:2046–2054 FSCV-compatible microelectrode array for
27. Venton BJ, Wightman RM (2007) real-time monitoring of heterogeneous dop-
Pharmacologically induced, subsecond dopamine release. Analyst 135:1556–1563
amine transients in the caudate-putamen of the 40. Zachek MK, Takmakov P, Moody B, Wightman
anesthetized rat. Synapse 61:37–39 RM, McCarty GS (2009) Simultaneous decou-
28. Keithley RB, Heien ML, Wightman RM pled detection of dopamine and oxygen using
(2009) Multivariate concentration determina- pyrolyzed carbon microarrays and fast-scan
tion using principal component regression cyclic voltammetry. Anal Chem 81:6258–6265
with residual analysis. Trends Anal Chem 41. Zachek MK, Takmakov P, Park J, Wightman
28:1127–1136 RM, McCarty GS (2010) Simultaneous moni-
toring of dopamine concentration at spatially in behaving animals. Nat Methods

different brain locations in vivo. Biosens 7:126–129
Bioelectron 25:1179–1185 43. Garris PA (2010) Advancing neurochemical
42. Clark JJ, Sandberg SG, Wanat MJ, Gan JO, monitoring. Nat Methods 7:106–108
Horne EA, Hart AS, Akers CA, Parker JG, 44. Stuber GD, Hnasko TS, Britt JP, Edwards RH,
Willuhn I, Martinez V, Evans SB, Stella N, Bonci A (2010) Dopaminergic terminals in the
Phillips PE (2010) Chronic microsensors for nucleus accumbens but not the dorsal striatum
longitudinal, subsecond dopamine detection corelease glutamate. J Neurosci 30:8229–8233
Chapter 17
The MPTP/Probenecid Model of Progressive

Parkinson’s Disease
Anna R. Carta, Ezio Carboni, and Saturnino Spiga
Abstract
Parkinson’s disease (PD) is characterized by a progressive degeneration of dopamine (DA) neurons and a
chronic loss of motor functions. The investigation of progressive degenerative mechanisms and possible
neuroprotective approaches for PD depends upon the development of an experimental animal model that
reproduces the neuropathology observed in humans. This chapter describes the generation of the 1-methyl-
4-phenyl-1,2,3,6-tetrahydropyridine/probenecid (MPTPp) chronic mouse model of PD. This model dis-
plays key features of PD, including impairment of motor and olfactory functions associated with partial loss
of tyrosine hydroxylase-positive neurons and DA levels in the brain. The MPTPp mouse model provides
an important tool for the study of mechanisms contributing to the pathological dysfunction of PD at the
cellular and whole animal level.
Key words: Neuropathology, Parkinson’s disease, Neuroprotection, Animal model, Preclinical

studies
1. Introduction
Parkinson’s disease (PD) is a neurodegenerative disorder

characterized by a progressive death of nigrostriatal dopamine
(DA) neurons and loss of striatal DA content. Symptoms of the
disease also have a gradual onset and progression (1). Specific non-
motor deficits, such as olfactory dysfunction and gastrointestinal
constipation, are generally considered early symptoms that often
precede motor deficits. Appearance of motor symptoms such as
bradykinesia, muscle rigidity and tremors, reflect the progressive
nigrostriatal degeneration, generally becoming manifest when
striatal DA drops to 20–30% of physiological levels, and nigral neu-
rons are largely lost (2). The time-lapse between initial neuronal
degeneration and occurrence of motor symptoms is likely attributable
295
296 A.R. Carta et al.
to the development of compensatory mechanism at the striatal

level (3). Moreover, motor deficits are associated with postsynaptic
functional changes in striatal medium-sized neurons (4, 5).
Classically, a reduced expression of mRNA for dynorphin peptide
in striatonigral neurons and an overexpression of mRNA for
enkephalin in striatopallidal neurons are thought to represent mea-
surable markers of these changes. On a neuropathological basis,
diseased dopaminergic neurons display an unbalanced neuronal
network, due to a complex scenario of malfunctioning cellular
components, including oxidative stress, impaired protein disposal
systems, and chronic neuroinflammation (6–10). Moreover, intra-
cellular formations named Lewy bodies and dystrophic neuritis are
found in nigral neurons as well as in other brain areas such as the
olfactory tubercle and locus ceruleous, and represent a cellular
hallmark of PD (11).
Given its progressive nature, any neuroprotective intervention
aimed at modifying the prognosis of the disease should optimally
be commenced at the early stages of PD. For this reason, preclini-
cal investigation of PD neuropathology and testing of neuropro-
tective strategies should rely on experimental models that reproduce
the progressive nature of PD pathology. More generally, animal
models of PD should possess the highest number of features of
human PD (face validity), the underlying neuropathology should
evolve as much as possible as PD and should respond to treatments
in a manner comparable to human PD (predictive validity). Lastly,
they should also reproduce the complex scenario of multiple inter-
actions between neuronal elements and surrounding cells (con-
struct validity) (12, 13).
A number of neurotoxin-based or genetic models that imply
the degeneration of dopaminergic nigrostriatal neurons have been
developed, although most of them present limitations (14–18).
One important limit is the rapid and transient neurodegeneration,
which occlude the development of chronic pathogenic mechanisms
and/or motor disabilities. Models based on the delivery of neuro-
toxin 1-methyl-4-phenyl-1,2,3,6-tetrahydropyridine (MPTP), that
offer the advantage of selectively targeting dopaminergic neurons,
have been largely characterized in the past decade (19–23). MPTP
treatment in mice, being easy to handle, affordable and highly
reproducible, remains to date the most widely used in vivo model
of PD in preclinical studies.
MPTP enters the brain after systemic administration. Here,
MPTP undergoes a two-step biotransformation process to form
the toxic metabolite MPP + (24). First, MPTP is converted to
MPDP + by MAO-B in astrocytes. In a second step MPDP + spon-
taneously oxidizes to MPP+. Released MPP + is specifically taken
up by dopaminegic cells via the dopamine transporter (DAT).
Importantly, both dopaminergic terminals in the striatum and
17 MPTPp Model of PD 297
dopaminergic dendrites in the SN express DAT, therefore

MPP + toxicity is possibly directed at both sites. Inside the cell,
MPP + can be sequestered into synaptosomal vesicles via the vesicular
monoamine transporters (VMATs), or it can accumulate into mito-
chondria, which are the subcellular target of toxicity. Hence MPP+,
by binding to complex I of the respiratory chain, blocks electron
transport leading to ATP depletion, increased oxidative stress, and
finally disruption of cellular energy metabolism and cell death (24).
Rodents are less sensitive to MPTP toxicity when compared to
primates. However, the C57black/6 strain is sensitive to appropri-
ate MPTP protocols of administration, also offering the advantage
of being selective in terms of targeting the nigrostriatal neurons
(25, 26). MPTP toxicity in mouse is strictly dependent on the dose
and paradigm of administration. Thus, the acute and subacute
administration models often fail to induce motor disabilities, loss
of TH-immunopositive cells in the SNc or persistency of the neu-
rodegeneration after the neurotoxin treatment has been completed
(24, 27). In the last decade, several research groups have devoted
efforts to the characterization of a chronic protocol of MPTP
administration in mice, delivered in association with the clearance
inhibitor probenecid (MPTPp). As enumerated below, this proto-
col overcomes most of the limitations presented by shorter MPTP
administration models, leading to validation of a preclinical model
of progressive PD (23, 28–30).
1. Administration of MPTPp, in a 5 week chronic protocol,
induces the gradual appearance of non-motor and motor symp-
toms, underlined by a progressive depletion of nigral dopamin-
ergic cells, as shown by TH-immunoreactivity (23, 28). The
olfactory dysfunction is detectable as early as after 1 MPTPp
administration, whereas motor impairment can be assessed
after the third administration, becoming overtly evident only
after 7 MPTPp administrations (i.e., in the fourth week of the
treatment) (23) (Fig. 1).
2. Dopaminergic cell loss in the SNc is observed after 3 MPTPp
administrations and gradually increases at the end of treatment
(23) (Fig. 2). Unlike shorter MPTP protocols, the MPTPp
chronic administration induced an enduring depletion of
dopaminergic neurons, as assessed up to 6 months after the
end of treatment (23, 28).
3. After chronic MPTPp typical PD hallmarks such as α-synuclein-
immunopositive deposits are present in the SNc (29), (Fig. 2),
whereas progressively increasing CD11b-immunoreactivity
reflecting microgliosis has been described, starting before
neurodegeneration can be measured (23). In addition signs of
neuronal apoptosis are detected in the SNc at the end of treat-
ment (30).
100 vehicle
Olfactory test * MPTPp *
pellet retrieval time (sec)

*
75
*
50
25
0
1 3 7 10 60
administrations days after treatment
3
5 vehicle
Beam traversal test *
number of errors per step MPTPp
4 3
3
* 3
3 *
0
1 3 7 10 60
Pole test 3
100 vehicle 3 3
*
MPTPp * *
time to descend (sec)
50
0
1 3 7 10 60
Fig. 1. Progressive impairment of olfactory and motor functions induced by chronic MPTPp administration. Retrieval of a
buried smelly pellet is used as olfactory test; the beam traversal test and the pole test are used to detect motor impair-
ments. *p < 0.05 as compared with respective vehicle-treated mice, 3p < 0.05 versus 1 and 3 MPTPp administrations; by
Newman-Keuls post hoc test.
4. In order to estimate cell loss in the entire SNc volume,

stereological counting is usually required. Alternatively, den-
sity of dopaminergic neurons in the SNc can be evaluated by
confocal laser scanning microscopy (CLSM) analysis of this
area, which provides 3-dimensional measures with accuracy
Fig. 2. Upper panels: confocal 3D reconstraction (40×) of dopaminergic neurons in the

substantia nigra compacta, double stained for TH (green) and α-synuclein (orange). Lower
panels: confocal 3D reconstraction (100×) of TH-positive fibers (green) in the striatum,
colocalized with synapsin-I (yellow ). Images are from the substantia nigra and striatum of
mice administered with vehicle or with chronic MPTPp.
higher than conventional light microscopy, allowing the assess-

ment of density as well as morphological changes of cells (31,
32).
5. In the striatum, DA and DA metabolites content are dramatically
reduced since the first MPTPp administration, as assessed by
HPLC measurement (23) (Fig. 3). In addition, at the postsyn-
aptic level, gradual changes in neuropeptides mRNA expres-
sion in medium-sized striatal neurons occur, indicating an
abnormal postsynaptic neuronal activity (23). Interestingly,
these changes show a delayed development if compared to the
drop of presynaptic DA content, possibly reflecting the tempo-
rarily development of compensatory mechanisms (23).
Based on these observations, the chronic MPTPp protocol of
administration represents a valid model of progressive PD to inves-
tigate the chronic mechanisms and to test neuroprotective strate-
gies for the disease. Behavioral tests, including non-motor and
motor tests, are advised as crucial tools to confer face validity to the
preclinical model of a motor disorder.
DOPAC/Dopamine
4
6000
3
striatal DA content
1
(pg/mg tissue)
4000
0
2000
* *
*
*
0
0 1 3 7 10
MPTP administrations
3200
striatal DOPAC content
2400
*
(pg/mg tissue)
* *
1600
^
*
800
0
0 1 3 7 10
MPTP administrations
Fig. 3. Loss of striatal dopamine and DOPAC content from mice chronically treated with
MPTPp. *p < 0.05 versus vehicle-treated mice.
2. Materials
2.1. Animals and Drug 1. Male C57Bl/6 J mice, 3 months old (25–30 g).
Treatment 2. MPTP-HCl (25 mg/kg i.p.) (Sigma-Aldrich, St. Louis, MO)
dissolved in water.
3. Probenecid (100 mg/kg i.p.) dissolved in 5% NaHCO3.
2.2. Olfactory Test 1. Clean plastic cage (24 width, 42 length, 15 height in
centimeters).
2. Cheese-smelly pellet.
2.3. Beam 1. Plexiglas Beam: the beam consists of four sections (25 cm each,
Traversal Test 1 m total length) of different width, starting at a width of
3.5 cm and gradually narrowing to 1 cm.
2. Mesh grid (1 cm square) of corresponding width to be placed
over the beam, leaving a 1 cm space between the grid and the
beam surface.
3. Video recording device.
2.4. Pole Test Vertical pole (diameter 8 mm, height 55).
2.5. Fluorescence 1. Surgical tools for the transcardial perfusion.

Immunohisto- 2. Brain removing and slices processing (Fine Science Tools):
chemistry and 22-gauge needles, scissors, scalpel.
Confocal Microscopy
3. Vibratome or Cryostat.
4. Multiwell plates and pipettes.
5. Microscope slides (superfrost plus for cryostat) and coverslips
(VWR, Radnor, PA).
6. 2× Phosphate-buffer saline (pH 7.4) (PBS).
7. 4% Paraformaldehyde (PFA) in 1× PBS. Heat H2O to 50°C
and add 8% PFA under a hood. Filter and store at 4°C. Add 8%
PFA to 2× PBS 1:1 ratio.
8. 0.1% Sodium azide in 1× PBS (Sigma-Aldrich).
9. Antifreezing solution: 30% glycerol, 30% ethylene glycol in 1×
PBS, 30% sucrose (for cryostat sections).
2.6. High Pressure The process of measuring neurotransmitters and metabolites from
Liquid Chromatography biological samples can be divided into two steps, extraction and
(HPLC) assessment.
2.6.1. Extraction 1. Ultrasonic homogenizer.

2. Centrifuge: to separate the analytes from the cell material a
centrifuge is required. The samples can be kept in ice at dark
and analyzed immediately after centrifugation. Alternatively, brain
tissue can be stored in −80°C in a deep freezer until use.
2.6.2. Assessment 1. HPLC pump (e.g., Jasco PU 1580, Great Dunmow, Essex UK).
2. Chromatographic column (e.g., LC −18 DB, 15 cm, 5 μm
particle size, Supelco, Milano, Italy, or Simmetry C-8 Waters,
Milford, MA).
3. Injector (e.g., Valco Instruments Co., Inc., Huston, TX, USA)

or an autosampler (e.g., ANTEC AS 110 autosampler,
Leyden NL).
4. Electrochemical detector (e.g., ESA Coluchem II Chelmsford
MA) with its electrode (e.g., ESA 50514 B).
5. Integrator (e.g., Agilent 3396, Santa Clara CA) or a detection
software (e.g., Kromatek Dunmow, Essex UK).
3. Methods
3.1. Animals and Drug MPTP and probenecid are administered twice a week for 5 weeks,
Treatment for a total of 10 administrations. Probenecid is given 30 min before
MPTP.
3.2. Behavioral Tests Olfactory function and motor performance are evaluated after 1, 3,
7, and 10 MPTPp injections, and 2 months after treatment discon-
tinuation. Behavioral tests should not detect acute pharmacologi-
cal actions of MPTP/MPP+, unrelated to neuronal damage, while
should reflect the neurodegenerative effect. For this reason, in our
studies we performed all behavioral tests on the third day after each
time point (23).
3.2.1. Olfactory Test Mice are food-deprived for 20 h before test. The test is conducted
in a clean plastic cage (24 w, 42 l, 15 h cm). A cheese-smelly pellet
is buried 1 cm under the bedding in a cage corner, and the mouse
is positioned in the center of the cage. Time spent to retrieve the
pellet and bite it is measured (23) (Fig. 1).
3.2.2. Beam Traversal Test The beam traversal test used to measure motor performance has
been adapted from the traditional beam-walking tests (33–35)
(Fig. 1). The beam is made of Plexiglas and consists of four sec-
tions (25 cm each, 1 m total length) of different width, starting at
a width of 3.5 cm and gradually narrowing to 1 cm. Mice are
trained for 2 consecutive days, and for five trials each day, to tra-
verse the beam from the widest to the narrowest side. On the test
day, a mesh grid (1 cm square) of corresponding width is placed
over the beam, leaving a 1 cm space between the grid and the beam
surface. Mice are videotaped while traversing the grid-surfaced
beam for a total of five trials. Videotapes are viewed and rated in
slow motion to detect errors. An error consists of a limb (forelimb
or hindlimb) slipping through the grid during a forward move-
ment that is visible between the grid and the beam surface. The
severity of the error is measured by scoring each limb slip individu-
ally (23, 35).
3.2.3. Pole Test The pole test is used to assess basal ganglia related movement
disorders in mice (36–39) (Fig. 1). Mouse is placed head-up on
top of a vertical pole (diameter 8 mm, height 55) placed in the
cage. When placed on the pole, mouse orients himself downward
and descends the length of the pole back into the cage. This test
requires motor coordination to turn downward and to climb to the
ground. Mice are trained for 2 consecutive days, for five trials each
day. On the test day, animals receive five trials, and total time to
orient downward and descend the pole is measured with a maxi-
mum duration of 120 s.
3.3. Fluorescence The procedure can be subdivided into three phases: Phase 1:
Immunohisto- Fluorescence immunocytochemistry, 1–3 days; Phase 2: CLSM,
chemistry and time required variable; Phase 3: Computer analysis, time required
Confocal Microscopy variable.
3.3.1. Immunofluorescence Mice are anesthetized at selected time points with chloral hydrate
(400 mg/kg i.p.) and transcardially perfused with ice-cold 4%
paraformaldehyde in 1× PBS. Brains, carefully removed from the
skull, are postfixed in 4% PFA/PBS for 2 h, and stored in SA at
4°C until cut. For cryostat sectioning, brains are PBS rinsed
(5 × 2 h), and cryoprotected in 30% sucrose for 2–3 days. Sections
(50 μm thick) are vibratome or cryostat-cut, and collected in mul-
tiwell plates (in PBS) for free-floating procedure. For the SNc,
consecutive sections are collected starting at −2.85 mm anterior
from bregma to the posterior ending of the area (18 total sections)
(40). Every third other section is processed and analyzed for TH
immunohistochemistry. Adjacent SNc sections are Nissl-stained, in
order to evaluate real cell loss, or processed to visualize different
proteins (see below).
1. Rinse in PBS (3 × 15 min).
2. Incubate for 30–60 min in a solution containing 5% normal
goat serum (NGS), 0.5% Triton X-100, and 5% Bovine serum
albumin in PBS.
3. Incubate with primary antibodies, overnight at 4°C. Process
sections for double labeling in various combinations (TH and
PSD-95/synapsin I/alpha-synuclein).
4. Incubate in a cocktail-containing rabbit anti-TH antibody
(1:400, Sigma-Aldrich Mo, USA) and mouse anti-PSD-95
(1:400, Santa Cruz Biotec. CA, USA), or rabbit anti-synapsin
I (1:400, Santa Cruz Biotec. CA, USA), or mouse anti-alpha-
synuclein (1:400), BD Transduction Laboratories, USA) in
PBS (pH 7.4).
6. Incubate with secondary antibodies.
7. To enhance the signal (for small and scattered structures like

PSD-95 antigen), incubate in biotinylated anti-mouse (rabbit)
IgG (1:400) for 2 h at RT. PBS rinse (3 × 15 min) and incubate
in a cocktail containing streptavidin fluorescein conjugate
(1:400, Vector, CA, USA) and goat anti-rabbit (mouse) IgG
Alexa Fluor 594 (1:400, Invitrogen, CA, USA) for 4 h at RT.
8. For normal signal, incubate in a cocktail containing anti-mouse
IgG TRITC conjugate (1:400 Sigma, Mo USA) and anti-rabbit
IgG FITC conjugate (1:400 Vector , CA, USA) for 4 h at RT.
10. Glass mount and coverslip with vectashield (Vector, CA, USA)
mounting medium.
11. For control procedure follow step-by-step the previous procedure
without incubation with primary antibodies (step 3, replace
the primary antibody with PBS).
3.3.2. Microscopy Analysis A main problem with conventional microscopy is the out-of-focus
light that leads to reduction in image contrast and a decrease in
resolution. The resulting blurring effect degrades the image by
shadowing important structures of interest, particularly in thick
specimens. In CLSM, out-of-focus structures do not contribute to
image shaping. By visualizing fluorescence associated with multiple
markers, CLSM is extremely valuable for co-localization studies.
CLSM scans can be collected in different datasets for computer
combination, volume and surface rendering (Fig. 3).
3.3.3. Computer Analysis The rendering technique is a computer algorithm used to transform
Via Rendering serially acquired images into 2D images containing 3D information.
Maximum Intensity Projection (MIP), Extended (depth-of) Focus
(EdF), Simulated Fluorescence Process (SFP), Surface rendering
and Colocalization algorithms are frequently used to display, ana-
lyze, and counting.
In particular, the MIP technique considers the brightest point
(the pixel with maximum intensity value) along the viewing direction.
This rendering process is fast and provides an estimation of the
amount of fluorescence without 3-D information. On the contrary,
the EdF synthesizes the three-dimensional information of the spec-
imen in a projection. The SFP renders a “realistic” view by surface
shading, transparency, and various lighting effects. This algorithm
simulates how the material would appear as if it was excited and
how the emitted light would travel. Surface rendering algorithm,
based on triangulated surfaces, is best used for volumetric counts,
required in small structures as the substantia nigra, to create shaded
solid 3-D objects. This method utilizes information from sequenced
image slices, in a topologically consistent way. The rendered sur-
faces are interactively displayed and analyzed for global structure
properties. Accordingly, automatic measurements of geometrical

characteristics, as area, volume, length, etc., can be directly obtained
by dedicated softwares. Appropriate spatial resolution (X, Y, and Z)
and high image quality are required for consistency and accuracy of
the results. In addition, high quality contrast images enable seg-
mentation simply by thresholding, while low contrast images, due
to uneven staining or heterogeneous acquisition, require more
advanced techniques of image analysis and processing.
3.4. High Pressure The possibility of detecting biogenic amines such as catecholamines
Liquid (dopamine (DA) and noradrenaline (NA)) or indolamines (sero-
Chromatography tonin (5-HT)) and relative metabolites with HPLC coupled with
(HPLC) electrochemical detection (ED) has provided a very powerful tool
for investigating peripheral and central nervous system (40–42).
The essential feature that allows the detection of these molecules is
their readiness to be oxidized, in fact oxidation of DA, NE, and
5-HT occurs spontaneously in solution. Nevertheless, if spontane-
ous oxidation of DA, NA, 5-HT, and metabolites is prevented the
quantitative detection of these molecules in a biologic fluid or in a
biological extract could be successfully managed by oxidizing
them after separation on a chromatographic column. Briefly, the
application of an electrical oxidation potential to a carbon-based
electrode on which the substance to be oxidized flows, will pro-
duce a chemical derivative with the giving up of electrons that will
alter the basal current detected, generating a signal. The electronic
elaboration of this signal through an electrochemical detector will
allow the quantitative detection of oxidable substances because the
current alteration is proportional to the concentration of the sub-
stance in the sample.
1. Mice are sacrificed by CO2 inhalation one at a time.
2. The brain is rapidly removed and the striatum is dissected on
an iced surface prepared by freezing saline in an expanded
polystyrene container.
3. The striatum is quickly introduced in microcentrifuge tube
previously weighted. Tubes are stored in dry ice until sup-
plementary storage at −80°C or immediate processing (see
Note 1).
4. In the latter case the number of sample to be processed depends
on the capability of the assaying procedure through the HPLC
(see Note 2).
5. Keeping all the test tubes in ice, 250 μl of previously cooled
0.2 M perchloric acid is added in each test tube.
6. Striatal tissue is sonicated (see Note 3) and then centrifuged at
9,000 × g for 15 min at 4°C.
7. Supernatant is filtered (0.6 μm) and diluted 1: 200 (see Note 4).
8. Twenty microliters are injected into an HPLC apparatus,

equipped with a reverse-phase column (LC −18 DB, 15 cm,
5 μm particle size Supelco, Milano, Italy) and coulometric
detector (Esa Coulochem II and 5014B electrode) to quantitate
DA and DOPAC. Electrodes are set at +150 mV (oxidation)
and −200 mV (reduction).
9. The mobile phase (in mM: CH3COONa, 0.23 M; Citric acid,
0.15 M; Na2EDTA, 100 mg/mL; pH 5.5) is pumped by a
Jasco PU 1580 pump at 1 mL/min flow rate. The assay sensi-
tivity for DA and DOPAC is 5 and 10 fmol/sample, respec-
tively (see Note 4).
4. Notes
1. Extreme care must be taken to prevent neurotransmitter

degradation by heat or light. Degradation of neurotransmit-
ters and metabolites during extraction are the most likely cause
for variability in substance values (i.e., pg/mg of fresh tissue)
among samples or found by different laboratories in the same
animal species. During the entire procedure test tubes must be
kept on ice and in dark conditions.
2. A typical HPLC run lasts from 10 to 12 min or 15–20 min.
Therefore, the number of samples to be processed must be
carefully scheduled, considering that up to 3–4 samples can be
processed per hour.
3. The selection of sonication parameters is crucial. We usually
use a probe with 2 mm diameter, sonicating each sample 4 × 5 s,
with 3 s intervals, at 30% power (130 W).
4. The sample dilution depends on the HPLC system sensibility
for DA and metabolites and may vary from 1:50 to 1:200.
Considering that an HPLC apparatus can detect as little as
1 pg of DA in a 20 μL sample and the average content of
striatal tissue is about 5,000 pg/mg of fresh tissue, it is pos-
sible to assess DA and metabolites in each striatum in order
to have an internal control for the extraction and detection
procedure.
5. The absolute amount of neurotransmitter and metabolite
detected can be obtained by using a standard curve reference.
However, standards may undergo partial degradation, result-
ing in the detection of higher amounts of neurotransmitters
and metabolites. Keeping them on ice and in the dark will min-
imize oxidative degradation. Additionally sodium metabisulfite
can be added (41). Each standard must be injected in
duplicate.
References
1. Hawkes CH, Del Tredici K, Braak H (2007) models for Parkinson’s disease. Neurotox Res
Parkinson’s disease: a dual-hit hypothesis. 11:219–240
Neuropathol Appl Neurobiol 33:599–614 15. Blandini F, Armentero MT, Martignoni E
2. Hornykiewicz O, Kish SJ (1987) Biochemical (2008) The 6-hydroxydopamine model: news
pathophysiology of Parkinson’s disease. Adv from the past. Parkinsonism Relat Disord
Neurol 45:19–34 14(Suppl 2):S124–S129
3. Hirsch EC (2000) Nigrostriatal system plastic- 16. Chesselet MF (2008) In vivo alpha-synuclein
ity in Parkinson’s disease: effect of dopaminer- overexpression in rodents: a useful model of
gic denervation and treatment. Ann Neurol Parkinson’s disease? Exp Neurol 209:22–27
47:S115–S120 17. Jenner P (2008) Functional models of
4. Nisbet AP, Foster OJ, Kingsbury A, Eve DJ, Parkinson’s disease: a valuable tool in the
Daniel SE, Marsden CD, Lees AJ (1995) development of novel therapies. Ann Neurol
Preproenkephalin and preprotachykinin mes- 64(Suppl 2):S16–S29
senger RNA expression in normal human basal 18. Schneider B, Zufferey R, Aebischer P (2008)
ganglia and in Parkinson’s disease. Neuroscience Viral vectors, animal models and new therapies
66:361–376 for Parkinson’s disease. Parkinsonism Relat
5. Goto S, Hirano A, Matsumoto S (1990) Met- Disord 14(Suppl 2):S169–S171
enkephalin immunoreactivity in the basal gan- 19. Bezard E, Dovero S, Bioulac B, Gross C
glia in Parkinson’s disease and striatonigral (1997) Effects of different schedules of MPTP
degeneration. Neurology 40:1051–1056 administration on dopaminergic neurodegen-
6. Barcia C, Fernández Barreiro A, Poza M, eration in mice. Exp Neurol 148:288–292
Herrero MT (2003) Parkinson’s disease and 20. Fornai F, Schlüter OM, Lenzi P, Gesi M, Ruffoli
inflammatory changes. Neurotox Res R, Ferrucci M, Lazzeri G, Busceti CL, Pontarelli
5:411–418 F, Battaglia G, Pellegrini A, Nicoletti F, Ruggieri
7. Hirsch EC, Hunot S (2009) Neuroinflammation S, Paparelli A, Südhof TC (2005) Parkinson-like
in Parkinson’s disease: a target for neuropro- syndrome induced by continuous MPTP infusion:
tection? Lancet Neurol 8:382–397 convergent roles of the ubiquitin-proteasome
8. Imamura K, Hishikawa N, Sawada M, Nagatsu system and alpha-synuclein. Proc Natl Acad Sci
T, Yoshida M, Hashizume Y (2003) U S A 102:3413–3418
Distribution of major histocompatibility com- 21. Smeyne RJ, Jackson-Lewis V (2005) The
plex class II-positive microglia and cytokine MPTP model of Parkinson’s disease. Brain Res
profile of Parkinson’s disease brains. Acta Mol Brain Res 134:57–66
Neuropathol 106:518–526 22. Yazdani U, German DC, Liang CL, Manzino
9. Olanow CW (2007) The pathogenesis of cell L, Sonsalla PK, Zeevalk GD (2006) Rat model
death in Parkinson’s disease (2007). Mov of Parkinson’s disease: chronic central delivery
Disord 22(17):S335–S342 of 1-methyl-4-phenylpyridinium (MPP+). Exp
10. Zhou C, Huang Y, Przedborski S (2008) Neurol 200:172–183
Oxidative stress in Parkinson’s disease: a mech- 23. Schintu N, Frau L, Ibba M, Garau A, Carboni
anism of pathogenic and therapeutic E, Carta AR (2009) Progressive dopaminergic
significance. Ann N Y Acad Sci 1147:93–104 degeneration in the chronic MPTPp mouse
11. Wakabayashi K, Tanji K, Mori F, Takahashi H model of Parkinson’s disease. Neurotox Res
(2007) The Lewy body in Parkinson’s disease: 16:127–139
molecules implicated in the formation and 24. Schmidt N, Ferger B (2001) Neurochemical
degradation of alpha-synuclein aggregates. findings in the MPTP model of Parkinson’s
Neuropathology 27:494–506 disease. J Neural Transm 108:1263–1282
12. Meredith GE, Sonsalla PK, Chesselet MF 25. Heikkila RE, Hess A, Duvoisin RC (1984)
(2008) Animal models of Parkinson’s disease Dopaminergic neurotoxicity of 1-methyl-4-
progression. Acta Neuropathol 115: phenyl-1,2,5,6-tetrahydropyridine in mice.
385–398 Science 224(4656):1451–1453
13. Olanow CW, Kieburtz K, Schapira AH (2008) 26. Zuddas A, Fascetti F, Corsini GU, Piccardi MP
Why have we failed to achieve neuroprotection (1994) In brown Norway rats, MPP + is accu-
in Parkinson’s disease? Ann Neurol 64(Suppl mulated in the nigrostriatal dopaminergic ter-
2):S101–S110 minals but it is not neurotoxic: a model of
14. Manning-Bog AB, Langston JW (2007) Model natural resistance to MPTP toxicity. Exp
fusion, the next phase in developing animal Neurol 127:54–61
27. Ricaurte GA, Langston JW, Delanney LE, deficits in mice transgenic for the human
Irwin I, Peroutka SJ, Forno LS (1986) Fate of Huntington’s disease mutation. J Neurosci
nigrostriatal neurons in young mature mice 19:3248–3257
given 1-methyl-4-phenyl-1,2,3,6-tetrahydro- 35. Fleming SM, Salcedo J, Fernagut PO,
pyridine: a neurochemical and morphological Rockenstein E, Masliah E, Levine MS,
reassessment. Brain Res 376:117–124 Chesselet MF (2004) Early and progressive
28. Petroske E, Meredith GE, Callen S, Totterdell sensorimotor anomalies in mice overexpressing
S, Lau YS (2001) Mouse model of Parkinsonism: wild-type human alpha-synuclein. J Neurosci
a comparison between subacute MPTP and 24:9434–9440
chronic MPTP/probenecid treatment. 36. Ogawa N, Hirose Y, Ohara S, Ono T, Watanabe
Neuroscience 106:589–601 Y (1985) A simple quantitative bradykinesia
29. Meredith GE, Totterdell S, Petroske E, Santa test in MPTP-treated mice. Res Commun
Cruz K, Callison RC Jr, Lau YS (2002) Chem Pathol Pharmacol 50:435–441
Lysosomal malfunction accompanies alpha- 37. Matsuura K, Kabuto H, Makino H, Ogawa N
synuclein aggregation in a progressive mouse (1997) Pole test is a useful method for evaluat-
model of Parkinson’s disease. Brain Res ing the mouse movement disorder caused by
956:156–165 striatal dopamine depletion. J Neurosci
30. Novikova L, Garris BL, Garris DR, Lau YS Methods 73:45–48
(2006) Early signs of neuronal apoptosis in the 38. Sedelis M, Hofele K, Auburger GW, Morgan
substantia nigra pars compacta of the progres- S, Huston JP, Schwarting RK (2000) MPTP
sive neurodegenerative mouse 1-methyl-4- susceptibility in the mouse: behavioral, neuro-
phenyl-1,2,3,6-tetrahydropyridine/ chemical, and histological analysis of gender
probenecid model of Parkinson’s disease. and strain differences. Behav Genet
Neuroscience 140:67–76 30:171–182
31. Spiga S, Serra GP, Puddu MC, Foddai M, 39. Fernagut PO, Chalon S, Diguet E, Guilloteau
Diana M (2006) Morphine withdrawal- D, Tison F, Jaber M (2003) Motor behaviour
induced abnormalities in the VTA: confocal deficits and their histopathological and func-
laser scanning microscopy. Eur J Neurosci tional correlates in the nigrostriatal system of
17:605–612 dopamine transporter knockout mice.
32. Spiga S, Lintas A, Migliore M, Diana M (2010) Neuroscience 116:1123–1130
Altered architecture and functional conse- 40. Paxinos G, Franklin KBJ (2001) The Mouse
quences of the mesolimbic dopamine system in Brain in Stereotaxic Coordinates, 2nd edn.
cannabis dependence. Addict Biol Academic, San Diego
15:266–276 41. Marsden CA, Joseph MH (1986) Biogenic
33. Drucker-Colín R, García-Hernández F (1991) amines. In: Kim CK (ed) HPLC of small mol-
A new motor test sensitive to aging and dop- ecules, a practical approach. IRL Press,
aminergic function. J Neurosci Methods Oxford
39:153–161 42. Carboni E (2003) Microdialysis coupled with
34. Carter RJ, Lione LA, Humby T, Mangiarini L, electrochemical detection: a way to investigate
Mahal A, Bates GP, Dunnett SB, Morton AJ brain monoamine role in freely moving ani-
(1999) Characterization of progressive motor mals. Methods Mol Med 79:415–432
INDEX
A C
Action potential ............................... 124, 125, 130–132, 235, Calcium imaging
261, 264, 265, 270, 289 advantages and disadvantages .................................... 133
Acute slice preparation .................................................... 128 dendritic calcium ............................................... 123–135
Adenosine A2A ................................................................... 96 voltage calcium imaging ............................ 127, 132–133
ADHD. See Attention deficit hyperactivity disorder Calmodulin kinase ............................................................. 44
(ADHD) CaMKII activity .......................................................... 44
Animal model .................................................................. 296 Cannabinoids CB1 ............................................................ 96
Parkinson’s disease (PD) ............................................ 296 Carbon-fiber microelectrode .... 245–258, 262, 263, 270, 280
β-Arrestin2 ...............................................108–113, 116–120 Caveolae ................................................................ 16, 19, 20
Attention deficit hyperactivity Cell culture ................................ 8, 17, 19, 26–27, 29–32, 45,
disorder (ADHD)............................................ 3, 107 80–84, 99, 100, 109–110, 147–148, 175, 176
Axonal dopamine release ......................................... 243–271 Chimeric mice
breeding ............................................................. 182, 195
B production ......................................................... 192–195
Backpropagation ...................................................... 124, 125 Chromosomal mapping ........................................... 209–212
Beam traversal test ............................................298, 301, 302 zebrafish dopamine receptor genes .................... 209–212
Binding assays ................................................... 4, 28, 35–37 Cloning....................... 99, 150, 158, 181–183, 187, 203–204
Bioluminescence resonance energy dopamine receptor genes ........................................... 182
transfer (BRET) .................................... 95–104, 108 signaling pathways ....................................................... 62
Biotinylation Confocal microscopy .....................................90–92, 99, 130,
biotinylation of cell surface 133, 301, 303
receptors ..................................................................4 Contigs ............................................................................ 202
sulfo-NHS-biotin ............................................ 4–6, 8, 10 Crosslinking .................................................... 46–47, 55–59
Bipolar electrical stimulation, 281 cell permeable crosslinking .................................... 55–57
Blastocysts protein crosslinking ............................................... 46, 56
harvesting .......................................................... 184, 192 reverse crosslinking .................................... 46–47, 57–59
injection ..............................................184–185, 192–195 Cyclic AMP (cAMP) .......................... 25, 26, 37–38, 41, 44,
Brain ...........................................4, 39, 44, 45, 49, 53, 55, 66, 66, 99, 108, 109, 111, 114–118, 120, 142, 149–151,
107, 108, 124–126, 128, 129, 133–136, 181, 182, 160, 166, 169–173, 176, 178
201, 216, 217, 220, 223, 224, 233, 240, 243–271, whole-cell cAMP assay..............................149–150, 160,
275–292 166, 169, 173, 176, 178
protein preparation ...................................................... 45
D
Brain slice
forebrain slices ....................................244, 245, 258–261 DAPI ..............................................................6, 9, 11, 29, 32
midbrain slices .................................................. 244, 245, DAT. See Dopamine transporter (DAT)
259, 261 Dendritic
Breeding ...........................................................182, 192, 195 excitability.................................................................. 268
BRET. See Bioluminescence resonance energy spike........................................................................... 124
transfer (BRET) Diffusion constant ................................................. 64–66, 73
BS3..................................................................................... 57 DNA ligation .................................................. 150, 155–156
DOI 10.1007/978-1-62703-251-3, © Springer Science+Business Media, LLC 2013
309
DOPAMINE
310 Index
Dopamine (DA) Fluorescence resonance energy transfer

binding ....................................................................... 61 (FRET)...................80–82, 86–92, 95–104, 108, 110
binding assays ........................................................35–37 Freely moving rat .....................................................286–287
dose-response curve ...................................166–171, 177 FRET. See Fluorescence resonance energy transfer (FRET)
extraction .........................................................29, 38–39 FSCV. See Fast-scan cyclic voltammetry (FSCV)
[3H]-SCH 23390 labeling .............................. 81, 84, 85, Fusion protein ...............................80, 82–84, 86, 87, 91, 92,
164, 167 96–100, 103, 104, 220
oxidation peak....................................................251, 270
quantification .........................................................38–39 G
release .............................26, 69, 229, 243–271, 275–292 GAL4/UAS ...................................................... 216, 217, 220
signaling ................................ 61–74, 202, 215–225, 289 Genotyping ............................................. 182, 185, 195–197
uptake assay ....................................... 231–232, 235, 241 Green fluorescent protein (GFP)............... 97, 217, 219–223
Dopamine (DA) receptor dopamine cell labeling ...............................................219
D1-like ..............................15, 25, 95, 108, 142, 151, 181
D2-like .......................................... 15, 95, 108, 181, 205 H
D2long .........................................................................110
High pressure liquid chromatography (HPLC) ................. 7
D2Rshort ......................................................................142
dopamine measurements .....................................38, 299
interacting protein, DRIP................................44, 52–55
dopamine metabolite measurement ...................305, 306
oligomerization ......................................................79–93
Homologous recombination ........................... 182, 185, 186,
Dopamine release
191, 197, 198
axonal.................................................................243–271
somatodendritic .................................................243–271 I
Dopamine transporter (DAT) .............. 25, 30–32, 229–241,
262, 269, 296, 297 ICSS. See Intracranial self-stimulation (ICSS)
Drosophila ....................................... 208, 209, 213, 215–225 Immunocytochemistry..................... 26, 30–32, 66, 303–305
Drosophila melanogaster ......................................208, 216 immunofluorescence ..........................................303–304
Dye injection ...........................................................128–129 Immunoprecipitation ........................... 45–46, 51, 56, 57, 59
Dynabeads ......................................45, 50–52, 216, 220, 221 co-immunoprecipitation ........................................50–52
In gel digestion ................................................ 46–47, 52, 58
E In situ hybridization.........................................................202
Intracranial self-stimulation (ICSS) ................................289
Electrochemistry .....................................................278–281
Ion exchange chromatography ...............................37–38, 41
ELISA. See Enzyme-linked immunoabsorbant
assay (ELISA) K
Embryonic stem (ES) cells
cloning .......................................................182, 190–191 Kidney cells .................................................................15–23
electroporation ...........................................................190 Knockout (KO) mouse ............................................182, 185
isolation ...............................................................87–188 D2LR ................................................................182, 186
plating........................................................................189
screening ............................................................182, 191
L
transfection .......................................................183–184, Ligand binding assay
187–192 competition assay ...................................................84–86
Enriched neuronal cultures ............................................... 29 saturation assay ........................................................... 84
Enzyme-linked immunoabsorbant assay Lipid raft
(ELISA) ....................................................4, 6, 8–11 immunoblotting to analyze ......................................... 21
Exchange protein activated by cAMP (EPAC) preparation with detergent method .......................20–21
sensor ...........................................................114, 120 preparation with non-detergent method................19–20
Expressed sequence tag (EST) ........................ 202, 203, 205
EST database............................................. 202, 203, 205 M
F Macromolecular complex.................................................. 96
Magnetic bead
Fast-scan cyclic voltammetry (FSCV)....................243–271, cell sorting ................................. 216, 218–220, 222, 223
276–278, 280, 281, 284, 289–290 MAP kinase.....................................................................239
DOPAMINE
Index
311
Mass spectrometry Phylogenetic analysis ...............................................206–209
liquid chromatography electrospray Plate reader .................................................. 6, 114, 116–118
ionization (LC-ESI) ........................................44, 52 BRET settings ...................................................116–118
matrix assisted laser desorption ionization.................. 44 Pole test ........................................................... 298, 301, 303
maximum intensity projection ...................................304 Polymerase chain reaction (PCR) ................... 27, 28, 34–35,
tandem mass spectrometry.......................................... 54 40, 82, 92, 99, 143–146, 150, 152–154, 175, 185,
Maximum parsimony...............................................207–209 188, 196–197, 204, 205, 209, 213, 231, 233, 234,
MEM. See Minimal essential media (MEM) 237–239, 241, 289
Membrane preparation Preclinical studies ............................................................296
crude membrane from HEK293 cells ................162, 166 Probenecid ...............................................................295–306
preparation from cultured striatal cells........................ 47 Proteomics
Mesencephalic primary cultures .............. 230–231, 233–234 protein-protein interaction ......................................... 57
3-Methoxytyramine (3-MT) ..................................108, 110, sample preparations .................................................... 45
111, 114, 115 trypsin digestion ......................................................... 54
Methyl-β-cyclodextrin (βCD) ..............................17, 19, 22
1-Methyl-4-phenyl-1,2,3,6-tetrahydropyridine R
(MPTP).......................................................295–306 Radioligand binding
Microdomain ....................................... 15, 17, 19–21, 23, 89 Bmax ............................................................................. 84
Minimal essential media (MEM) ................... 109, 113, 147, saturation studies .......................................................160
148, 159, 160, 166, 169, 173, 175 Rate constant .......................................................63–64, 262
MPTP. See 1-Methyl-4-phenyl-1,2,3,6-tetrahydropyridine Receptor
(MPTP) dopamine receptor oligomerization .......................79–93
3-MT. See 3-Methoxytyramine (3-MT) heteromer ............................................................95–104
Müller glia ....................................................... 26, 29–30, 37 Renilla Luciferase (Rluc) .................................................111
Mutagenesis.......................................................82, 141–179 Restriction enzyme ..........................154, 157, 159, 173, 183,
mutant receptor .................................................153–155 186, 191, 196, 1447
Mutant cassettes ..............................................................155 Retina
retinal cultures .................................................29–30, 37
N
RNA extraction .......................................................... 39
Neuronal-glial cultures ..................................................... 29 Reverse transcriptase-PCR (RT-PCR)
Neuroprotection ............................................................... 26 D1A and D1B receptors .............................................33–35
Neurospheres ...............................................................26, 30 dopamine receptor .................................................33–35
retinal culture .............................................................. 30 dopamine transporter (DAT) ....................................237
Northern blot ..................................................188, 196–197 quantitative RT-PCR ................................ 233, 234, 237
RNA isolation ......................................... 218–220, 222–223
O
S
Olfactory test ................................................... 298, 301, 302
Oligonucleotides ......................... 27, 28, 35, 40, 80, 82, 143, Schizophrenia (SZ) ................3, 85, 107, 201, 202, 245, 275
145, 146, 175, 196, 199, 202 Scintillation ........................... 36, 38, 41, 81, 84, 86, 93, 149,
Optical imaging .......................................................126–127 162, 166, 169, 170, 178, 232, 235
Ortholog Sequential BRET-FRET (SRET)
dopamine receptor genes in zebrafish .......... 82, 201, 209 saturation curves ........................................ 101, 103, 104
SRET detection .................................................101–103
P Sequential chromatography .............................151, 170–172
Parkinson’s disease (PD) ..........................3, 39, 61, 107, 201, Site-directed mutagenesis ......................... 82, 143, 145, 147,
202, 245, 275, 295–306 149, 151–155, 157, 159, 161, 163, 165, 167, 169,
Patch electrode recording ........................................128–129 171, 173, 175, 177, 179
PCR. See Polymerase chain reaction (PCR) Striatum..................... 41, 45, 49, 61, 79, 188, 196, 243–245,
Phasic dopamine 249, 259–261, 263–268, 296, 299, 305, 306
firing ..........................................................................275 Substantia nigra .......... 61, 202, 243, 244, 260, 284, 299, 304
signal .........................................................................130 substantia nigra pars compacta .................... 61, 243, 244
β-Phenylethylamine (β-PEA) ........................ 108, 110, 111, Sucrose gradient ..........................................................16–23
114, 115, 118 sucrose gradient centrifugation ........................17–19, 22
DOPAMINE
312 Index
T V
TAAR1....................................................................107–121 Ventral tegmental area (VTA) ........................ 108, 243–245,
Targeting vector...................................... 182–183, 185–188, 262, 264, 267, 268, 270, 284, 290
190, 196 Voltage calcium imaging..................................127, 132–133
Third intracellular loop ........................... 143, 182, 186, 206 Voltammetry ....................................................243–271, 276
Toxicity ....................................................................120, 297 VTA. See Ventral tegmental area (VTA)
MPTP toxicity ..........................................................297
Trace amines .................................................... 108, 110, 111
Transfection W
embryonic stem cell ..........................................183–184, Western blot
187–192 dopamine markers .................................................30–31
transient transfection ..................... 83, 91, 100–101, 103 dopamine transporter ................................................238
Transformation .......................................... 92, 156, 174, 296 lipid raft proteins ........................................................ 18
Transgene ................................................ 195, 217, 219, 220
Transmembrane (TM) domain
transmembrane region ...............................................142 Z
Transporter ........................................... 25, 39, 44, 229–241,
Zebrafish
262, 296, 297
chromosomal mapping ......................................209–212
Tyrosine hydroxylase (TH) ............................. 25, 26, 30–33,
phylogenetic analysis .................................................208
217, 232, 238–240, 262, 278, 288, 297, 299, 303

Dopamine - Methods and Protocols

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Dopamine - Methods and Protocols

Uploaded by

Copyright:

Available Formats

METHODS IN MOLECULAR BIOLOGY™

For further volumes:

Methods and Protocols

ISSN 1064-3745 ISSN 1940-6029 (electronic)

Library of Congress Control Number: 2012950334

© Springer Science+Business Media, LLC 2013

Printed on acid-free paper

Humana Press is a brand of Springer

Nadine Kabbani, Ph.D.

PART I BIOCHEMICAL, PROTEOMIC, AND COMPUTATIONAL TOOLS

1 Detection of Cell Surface Dopamine Receptors . . . . . . . . . . . . . . . . . . . . . . . 3

PART II CELLULAR IMAGING

6 A Biophysical Approach for the Study of Dopamine

PART III GENETIC MANIPULATION IN CELLS AND ORGANISMS

10 Functional Analysis of Human D1 and D5 Dopaminergic

PART IV ELECTROCHEMICAL, PHYSIOLOGICAL, AND BEHAVIORAL ANALYSIS

14 Regulation of Dopamine Transporter Expression

RAUL R. GAINETDINOV • Department of Neuroscience and Brain Technologies,

JESSICA PETKO • Department of Pharmacology, Penn State College of Medicine,

Biochemical, Proteomic, and Computational Tools

Detection of Cell Surface Dopamine Receptors

Key words: Schizophrenia, ADHD, DAPI, Biotinylation, ELISA, Plasma membrane

Dopamine (DA) regulates movement, endocrine function, reward

A number of factors have been discovered in the past 15 years

Receptors can be detected with epitope or subtype specific primary

2.1. Biotinylation 1. HEK293 cells.

19. TBS-T buffer: 250 mM Tris–HCl, pH 7.5, 1.5 M NaCl, 1%

2.2. Detection of Cell 1. 24-well tissue culture plates.

3.1. Biotinylation 1. Culture 1 × 106 FLAG-D1R cells and untransfected HEK293

Fig. 1. Detection of cell surface D1Rs by biotinylation. 48 h after plating untransfected

1. Cells stably transfected with plasmids containing neomycin

3. Biotinylation reagents are susceptible to hydrolysis so the

17. Be cautious, no detergent!

Methods for the Study of Dopamine Receptors

Key words: Lipid rafts, Caveolae, Membrane microdomains, Dopamine receptor

Dopamine receptors belong to the α group of the rhodopsin-like

Caveolae and lipid rafts have been implicated to play a role in

2.1. Cell Culture 1. HEK-293 cells heterologously expressing human D1 receptor

Solutions (total 5 mL) 30% 20% 10% 5%

12. 50% OptiPrep stock solution: 45 mL of 60% OptiPrep mixed

5. Remove the tubes from the bucket at the end of the

rotors can be used such as SW 55 (4 h at 170,000 × g), TLS55

7. The sucrose gradient samples with pH 6.8–7.0 can be stored at

These studies were supported in part by grants from the National

Methods of Dopamine Research in Retina Cells

Dopamine is a key catecholamine found in the vertebrate retina,

some species, before the expression of TH, the enzyme that

7. Running buffer (1 L): 3 g Trizma base, 14.4 g glycine, 10 mL

5¢-TACCCGACAATGCTGGAGAC-3¢ (base position 1,279)

2.3. Binding Assays 1. Ice-cold lysis buffer: 5 mM Tris–HCl, 5 mM MgCl2, pH 7.4.

2.4. Immuno- 1. Clean spherical microscope coverslips (0.15 mm).

7. Nuclear stain: 300 nM DAPI (4,6-diamidino-2-phenylindole)

2.5. Dopamine 1. Tris–HCl solution, pH 8.8.

4. Purified glial cultures can be used 3 days later and maintained

5. Estimate protein content in 2 μL samples of extracts by the

3.2.2. Immunodetection 1. Block membranes with 5% nonfat milk in Tris-buffered saline

2. Rinse cultures extensively with sodium phosphate-buffered

(Chemicon anti-rabbit 1:200 in PBS plus 0.25% Triton X-100,

3. Incubate for 10 min at 55–60°C; keep at 4°C.

3.4.4. Amplification 1. Amplify directly 5 μL of cDNA preparation in a 50 μL reaction

6. Destain in water for 20 min at room temperature; visualize

[3H]-SCH 23390 (nM)

Fig. 4. Specific binding of [3H]-SCH 23390 to homogenates of E8C4 (embryonic day 8

6. Discard the supernatant and add ice-cold binding buffer to the