Opioid Receptors Methods and Protocols

Methods in MolecularBiology 1230

SantiM.Spampinato Editor

Opioid

Receptors 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

Edited by

Santi M. Spampinato

University of Bologna, Bologna, Italy

Editor

Santi M. Spampinato
Department of Pharmacy and Biotechnology (FaBiT) University of Bologna
Bologna, Italy

ISSN 1064-3745 ISSN 1940-6029 (electronic) ISBN 978-1-4939-1707-5 ISBN 978-1-4939-1708-2 (eBook) DOI 10.1007/978-1-4939-1708-2
Springer New York Heidelberg Dordrecht London

Library of Congress Control Number: 2014949777

© Springer Science+Business Media New York 2015
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: Typical collective variables used in the study of ligand binding to opioid receptors.

Printed on acid-free paper

Humana Press is a brand of Springer
Springer is part of Springer Science+Business Media (www.springer.com)

Dedication

To my wife, Anna,
for her love, encouragement and patience

Preface

Opioid receptors, beyond their involvement in pain transmission, play a number of relevant physiological roles in the central nervous system and in peripheral organs. Opioid receptors can be considered a crossroads where endogenous opioid peptides and foreign opioids and opiates meet the cell and transmit their messages to another vast array of stimulus–response mechanisms. In recent years, studies on their emerging roles have been favored by numer- ous and fruitful techniques that have opened new avenues of preclinical and clinical research that demands multidisciplinary approaches.

The post-genomic era has opened up novel opportunities for the exploitation of these novel technologies. As an increasing number of investigators seek to harness the fruits of knowledge in these emerging fields, it is essential that well-tested protocols are made avail- able to researchers. With this in mind, it is apposite to provide a collection of protocols to favor innovative studies on opioid receptors written by experts who are routinely employing these techniques in their laboratories. This book presents the protocols in the stepwise “cookbook” style of this well-known book series along with summaries of state-of-the art methods that have been utilized for understanding opioid receptor functionality at a molec- ular, cellular, structural, and organism level. Opioid Receptors: Methods and Protocols is, hence, an invaluable guide for researchers in the fields of neuroscience, biochemistry, phar- macology, and molecular and structural biology.

Part I of this book (Chapters 1–5) focuses on procedures to evaluate genetics and structural biology of opioid receptors as well as their transcriptional and posttranscriptional regulation. An overview of genetic analysis of opioid receptors with the latest sequencing methods is included (Chapter 1). The recent publication of crystal structures of all the three opioid receptors has been instrumental to the development of computational proto- cols, designed to estimate thermodynamic and kinetic parameters describing the receptor binding of small molecule ligands and the formation of supramolecular complexes (Chapter 2). Furthermore, techniques for the epigenetic and posttranscriptional analysis of opioid receptor genes are presented (Chapters 3 and 4). Finally, a protocol is dedicated to the use of DNA microarrays and next-generation sequencing methodologies to obtain a transcrip- tional profile of genes influenced by activation of opioid receptors (Chapter 5).

Part II (Chapters 6–12) illustrates methods for the cellular detection and analysis of opioid receptors. Techniques aimed to monitor the trafficking and interaction of opioid receptors and related signaling molecules are described. Total internal reflection fluores- cence microscopy has been used to investigate, in real-time, surface trafficking events of opioid receptors at the single molecule level (Chapter 6). An innovative protocol aimed at investigating opioid receptor internalization and trafficking events in vivo is reported (Chapter 7). Techniques for monitoring heteromerization between opioid receptors and the interaction of opioid receptors and beta-arrestin in living cells by bioluminescence reso- nance energy transfer are illustrated in Chapters 8 and 9. The study of protein–opioid receptor interactions assists the understanding of biological functions and elucidation of biochemical pathways, and Chapter 10 details procedures to assay the interaction between

vii

viii Preface

protein 14-3-3 zeta and the human kappa opioid receptor by co-immunoprecipitation, pull-down assay, and fluorescence microscopy. Two separate procedures to detect opioid receptors by immunoblot assays in brain areas (Chapter 11) and in peripheral tissues (Chapter 12) are presented.

Part III (Chapters 13–16) covers strategies for the analysis of signaling events modu- lated by opioid receptors activated by agonist ligands. Following opioid receptor activation, GTP will replace GDP on the α-subunit of the G-protein, leading to a dissociation of the βγ-subunit. The [35S]GTPγS autoradiography assay, described in Chapter 13, is useful to monitor opioid receptor activation in discrete brain areas. Chapter 14 describes two real- time fluorescence-based assays of mu-opioid receptor activation by agonists monitoring cell membrane hyperpolarization in AtT-20 cells. These assays may be scaled up for high- throughput screening. The use of imaging assays (Chapter 15) and of the whole cell patch clamp (Chapter 16) to investigate the activation of inwardly rectifying potassium channels and calcium channels by mu- and delta-opioid receptor agonists in cultured mouse dorsal root ganglion neurons are described.

Part IV (Chapters 17–23) covers experimental techniques to investigate opioid receptor- mediated functions at organismal level in a physiological or pathological context. An in vitro skin-saphenous nerve preparation to test the modulatory effects of opioids on the function of cutaneous sensory neurons in experimental models of pain is discussed (Chapter 17), whereas Chapter 18 reports the analysis of cutaneous stimulation-induced sensory input by Von Frey hairs. A protocol to detect drug-stimulated intracellular zinc release in rodent brain slices using time-lapse microscopy and fluorescence imaging is presented in Chapter 19. Chapter 20 provides detailed procedures to measure activation of retinal opioid recep- tors and to assess their roles in retina neuroprotection by electroretinogram. The immuno- suppressive effects mediated by opioids are central to the in vivo activation of opioid receptors, and Chapters 21 and 22 explain strategies towards attaining this objective. Finally, a technique to evaluate the role of opioid receptors in migration and wound recov- ery in cultured human keratinocytes and fibroblasts is presented in Chapter 23.

Part V (Chapters 24–27) showcases methods for the analysis of behavioral effects induced by opioids. Chapter 24 is an overview on a reinstatement animal model that has contributed to disentangling the mechanisms underlying relapse to opioid-seeking in labo- ratory animals. This procedure is useful to investigate the neurobiology of relapse. Analysis of heroin-seeking reinstatement in the rat is useful to study the mechanisms underlying relapse to heroin and vulnerability factors that enhance the resumption of heroin-seeking behavior. This protocol is described in Chapter 25. Chapter 26 presents a procedure to investigate the role of opioid receptors in alcoholism by adopting a model that combines chronic ethanol exposure procedures with voluntary ethanol drinking in rodents. Behavioral tests designed to evaluate in pups the activity and involvement of opioid receptors are described in Chapter 27.

I sincerely hope that these protocols will help both experienced and new entrants in this field to carry out their experiments successfully. Finally, I would like to thank all the authors for their outstanding contributions and the series editor, John M. Walker, for valuable edi- torial help.

Bologna, Italy Santi M. Spampinato

Contents

Dedication. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . v Preface. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . vii Contributors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . xiii

PART I GENETIC, STRUCTURAL BIOLOGY, TRANSCRIPTIONAL
AND POST-TRANSCRIPTIONAL ANALYSIS OF OPIOID RECEPTORS

. 1 Overview of Genetic Analysis of Human Opioid Receptors . . . .
Santi M. Spampinato

. 2 Computational Structural Biology of Opioid Receptors. . . . . . .
Davide Provasi

. 3 Analysis of Epigenetic Mechanisms Regulating Opioid Receptor Gene Transcription . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Cheol Kyu Hwang, Yadav Wagley, Ping-Yee Law, Li-Na Wei, and Horace H. Loh

. 4 Renilla Luciferase Reporter Assay to Study 3′UTR-Driven Posttranscriptional Regulations of OPRM1 . . . . . . . . . . . . … . Gabriele Vincelli and Andrea Bedini

. 5 High-Throughput Gene Expression Profiling
of Opioid-Induced Alterations in Discrete Brain Areas. . . . … . Michal Korostynski, Marcin Piechota, Slawomir Golda,
and Ryszard Przewlocki

PART II CELLULAR DETECTION AND ANALYSIS OF OPIOID RECEPTORS

. 6 Real-Time Imaging of Mu Opioid Receptors
by Total Internal Reflection Fluorescence Microscopy . . . . … . Cristina Roman-Vendrell and Guillermo Ariel Yudowski

. 7 In Vivo Techniques to Investigate the Internalization Profile
of Opioid Receptors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Amynah A. Pradhan, Vivianne L. Tawfik, Alycia F. Tipton, and Grégory Scherrer

. 8 Monitoring Opioid Receptor Dimerization in Living Cells
by Bioluminescence Resonance Energy Transfer (BRET). . . . . . Monica Baiula

. 9 Bioluminescence Resonance Energy Transfer (BRET)
to Detect the Interactions Between Kappa Opioid Receptor
and Non-visual Arrestins . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Andrea Bedini

. ………. 3 . ………. 13

. ………. 39

. ………. 53 . ………. 65

. ………. 79 . ………. 87

. ………. 105

. ………. 115

x Contents

. 10 Identification and Verification of Proteins Interacting
with the Kappa Opioid Receptor (KOPR) . . . . . . . . . . . . . . . . . . . . . . . . . . . . 129 Chongguang Chen, Peng Huang, and Lee-Yuan Liu-Chen

. 11 Detection of Mu Opioid Receptor (MOPR) and Its Glycosylation
in Rat and Mouse Brains by Western Blot with Anti-μC,
an Affinity-Purified Polyclonal Anti-MOPR Antibody . . . . . . . . . . . . . . . . . . . 141 Peng Huang, Chongguang Chen, and Lee-Yuan Liu-Chen

. 12 Immunohistochemical Analysis of Opioid Receptors in Peripheral Tissues . . . . 155
Yvonne Schmidt and Halina Machelska

PART III ANALYSIS OF SIGNALING EVENTS MODULATED BY OPIOID RECEPTORS

. 13 [35S]GTPγS Autoradiography for Studies
of Opioid Receptor Functionality . . . . . . . . . . . . . . . .. . ……………. 169 Alfhild Grönbladh and Mathias Hallberg

. 14 Fluorescence-Based, High-Throughput Assays
for μ-Opioid Receptor Activation Using a Membrane
Potential-Sensitive Dye . . . . . . . . . . . . . . . . . . . . . . . .. . ……………. 177 Alisa Knapman and Mark Connor

. 15 Analysis of Potassium and Calcium Imaging to Assay the Function
of Opioid Receptors . . . . . . . . . . . . . . . . . . . . . . . . . . . . ……………. 187 Viola Spahn, Dinah Nockemann, and Halina Machelska

. 16 Electrophysiological Patch Clamp Assay to Monitor the Action
of Opioid Receptors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 197 Viola Spahn, Dinah Nockemann, and Halina Machelska

PART IV MODEL SYSTEMS TO STUDYING OPIOID RECEPTOR-MEDIATED FUNCTIONS

. 17 Skin–Nerve Preparation to Assay the Function of Opioid Receptors
in Peripheral Endings of Sensory Neurons. . . . . . . . . . . . . . . . . . . . . . . . . . . . 215 Rabih Moshourab, Yvonne Schmidt, and Halina Machelska

. 18 Mechanical Nociception Measurement in Mice and Rats
with Automated Von Frey Equipment. . . . . . . . . . . . . . . . …………… 229 Gabriele Campana and Roberto Rimondini

. 19 Detecting Zinc Release Induced by Mu-Opioid Receptor
Agonists in Brain Slices . . . . . . . . . . . . . . . . . . . . . . . . . . . …………… 233 María Rodríguez-Muñoz, Pilar Sánchez-Blázquez, Concha Bailón,
and Javier Garzón

. 20 Opioid Receptors: Methods for Detection and Their Modes
of Actions in the Eye. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . …………. 243 Shahid Husain

. 21 Evaluation of Murine Macrophage Cytokine Production
After In Vivo Morphine Treatment . . . . . . . . . . . . . . . . . . . . …………. 253 Silvia Franchi, Mara Castelli, Sarah Moretti, Alberto Panerai,
and Paola Sacerdote

. 22 Measurement of Macrophage Toll-Like Receptor 4 Expression
After Morphine Treatment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Mara Castelli, Alberto Panerai, Paola Sacerdote, and Silvia Franchi

. 23 The Role of Opioid Receptors in Migration and Wound Recovery
In Vitro in Cultured Human Keratinocytes and Fibroblasts . . . . . . .. . Mei Bigliardi-Qi and Paul L. Bigliardi

PART V BEHAVIORAL EFFECTS MEDIATED BY OPIOID RECEPTORS

. 24 Role of Opioid Receptors in the Reinstatement
of Opioid-Seeking Behavior: An Overview . . . . . . . . . . . . . . . . . . .. . Liana Fattore, Paola Fadda, Silvia Antinori, and Walter Fratta

. 25 Analysis of Opioid-Seeking Reinstatement in the Rat. . . . . . . . . . . .. .
Liana Fattore, Paola Fadda, Mary Tresa Zanda, and Walter Fratta

. 26 Induction of a High Alcohol Consumption in Rats and Mice:
Role of Opioid Receptors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Roberto Rimondini and Gabriele Campana

. 27 Evaluation of Social and Nonsocial Behaviors Mediated
by Opioids in Mouse Pups . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Francesca R. D’Amato

….. . 263 ….. . 273

….. . 281 ….. . 295

….. . 309

….. . 313

Index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 323

Contents xi

Contributors

SILVIA ANTINORI • Division of Neuroscience and Clinical Pharmacology, Department of Biomedical Sciences, University of Cagliari, Cittadella Universitaria di Monserrato, Monserrato, CA, Italy
CONCHA BAILÓN • Neuropharmacology, Cajal Institute, Consejo Superior de Investigaciones

Científicas (CSIC), Madrid, Spain
MONICA BAIULA • Department of Pharmacy and Biotechnology (FaBiT),

University of Bologna, Bologna, Italy
ANDREA BEDINI • Department of Pharmacy and Biotechnology (FaBiT),

University of Bologna, Bologna, Italy
PAUL L. BIGLIARDI • Agency for Science Technology and Research (A*STAR), Institute of

Medical Biology, Singapore, Singapore; Division of Rheumatology, National University

Hospital University Medicine Cluster, Singapore, Singapore
MEI BIGLIARDI-QI • Agency for Science Technology and Research (A*STAR),

Institute of Medical Biology, Singapore, Singapore; Division of Rheumatology,

National University Hospital, University Medicine Cluster, Singapore, Singapore GABRIELE CAMPANA • Department of Pharmacy and Biotechnology (FaBiT),

University of Bologna, Bologna, Italy
MARA CASTELLI • Department of Pharmacological and Biomolecular Science,

University of Milano, Milan, Italy
CHONGGUANG CHEN • Center for Substance Abuse Research and Department of

Pharmacology, Temple University School of Medicine, Philadelphia, PA, USA
MARK CONNOR • Australian School of Advanced Medicine, Macquarie University, Sydney,

NSW, Australia
FRANCESCA R. D’AMATO • Cell Biology and Neurobiology Institute, CNR/IRCCS Santa

Lucia Foundation, Rome, Italy
PAOLA FADDA • Centre of Excellence Neurobiology of Dependence, Cagliari, Italy;

Division of Neuroscience and Clinical Pharmacology, Department of Biomedical SciencesUniversity of Cagliari, Cittadella Universitaria di Monserrato,
Monserrato, CA, Italy; National Institute of Neuroscience (INN)University of Cagliari, Cagliari, Italy

LIANA FATTORE • CNR National Research Council of Italy, Institute of Neuroscience- Cagliari, Cagliari, Italy; Centre of Excellence Neurobiology of Dependence, Cagliari, Italy

SILVIA FRANCHI • Department of Pharmacological and Biomolecular Science, University of Milano, Milan, Italy

WALTER FRATTA • Centre of Excellence Neurobiology of Dependence, Cagliari, Italy; Division of Neuroscience and Clinical Pharmacology, Department of Biomedical SciencesUniversity of Cagliari, Cittadella Universitaria di Monserrato,
Monserrato, CA, Italy; National Institute of Neuroscience (INN)University of Cagliari, Cagliari, Italy

JAVIER GARZÓN • Neuropharmacology, Cajal Institute, Consejo Superior de Investigaciones Científicas (CSIC), Madrid, Spain

xiii

xiv Contributors

SLAWOMIR GOLDA • Department of Molecular Neuropharmacology, Polish Academy of Sciences, Institute of Pharmacology, Krakow, Poland

ALFHILD GRÖNBLADH • Division of Biological Research on Drug Dependence, Department of Pharmaceutical Biosciences, Beijer Laboratory, Uppsala University, Uppsala, Sweden

MATHIAS HALLBERG • Division of Biological Research on Drug Dependence, Department of Pharmaceutical Biosciences, Beijer Laboratory, Uppsala University, Uppsala, Sweden

PENG HUANG • Center for Substance Abuse Research and Department of Pharmacology, Temple University School of Medicine, Philadelphia, PA, USA

SHAHID HUSAIN • Department of Ophthalmology, Hewitt Laboratory of the Ola B. Williams Glaucoma Center, Storm Eye Institute, Medical University of South Carolina, Charleston, SC, USA

CHEOL KYU HWANG • Department of Pharmacology, University of Minnesota Medical School, Minneapolis, MN, USA

ALISA KNAPMAN • Australian School of Advanced Medicine, Macquarie University, Sydney, NSW, Australia

MICHAL KOROSTYNSKI • Department of Molecular Neuropharmacology, Polish Academy of Sciences, Institute of Pharmacology, Krakow, Poland

PING-YEE LAW • Department of Pharmacology, University of Minnesota Medical School, Minneapolis, MN, USA

LEE-YUAN LIU-CHEN • Center for Substance Abuse Research and Department of Pharmacology, Temple University School of Medicine, Philadelphia, PA, USA

HORACE H. LOH • Department of Pharmacology, University of Minnesota Medical School, Minneapolis, MN, USA

HALINA MACHELSKA • Klinik für Anästhesiologie und Operative Intensivmedizin, Freie Universität Berlin, Charité-Universitätsmedizin Berlin, Berlin, Germany

SARAH MORETTI • Department of Pharmacological and Biomolecular Science, University of Milano, Milan, Italy

RABIH MOSHOURAB • Klinik für Anästhesiologie und Operative Intensivmedizin, Freie Universität Berlin, Charité-Universitätsmedizin Berlin, Berlin, Germany DINAH NOCKEMANN • Klinik für Anästhesiologie und Operative Intensivmedizin, Freie Universität Berlin, Charité-Universitätsmedizin Berlin, Berlin, Germany

ALBERTO PANERAI • Department of Pharmacological and Biomolecular Science, University of Milano, Milan, Italy

MARCIN PIECHOTA • Department of Molecular Neuropharmacology, Polish Academy of Sciences, Institute of Pharmacology, Krakow, Poland

AMYNAH A. PRADHAN • Department of Psychiatry, University of Illinois at Chicago, Chicago, IL, USA

DAVIDE PROVASI • Department of Structural and Chemical Biology, Icahn School of Medicine at Mount Sinai, New York, NY, USA

RYSZARD PRZEWLOCKI • Department of Molecular Neuropharmacology,
Polish Academy of Sciences, Institute of Pharmacology, Krakow, Poland; Department of Neuroscience and Neuropsychology, Institute of Applied PsychologyJagiellonian University, Krakow, Poland

ROBERTO RIMONDINI • Department of Medical and Surgical Sciences, Pharmacology Unit, University of Bologna, Bologna, Italy

MARÍA RODRÍGUEZ-MUÑOZ • Neuropharmacology, Cajal Institute, Consejo Superior de Investigaciones Científicas (CSIC), Madrid, Spain

CRISTINA ROMAN-VENDRELL • Department of Anatomy and Neurobiology, School of Medicine, University of Puerto Rico, San Juan, Puerto Rico; Institute of Neurobiology, University of Puerto Rico, San Juan, Puerto Rico; Department of Physiology, School of Medicine, University of Puerto Rico, San Juan, Puerto Rico

PAOLA SACERDOTE • Department of Pharmacological and Biomolecular Science, University of Milano, Milan, Italy

PILAR SÁNCHEZ-BLÁZQUEZ • Neuropharmacology, Cajal Institute, Consejo Superior de Investigaciones Científicas (CSIC), Madrid, Spain

GRÉGORY SCHERRER • Department of Anesthesiology, Perioperative and Pain Medicine, Department of Molecular and Cellular Physiology, Department of Neurosurgery, Stanford University School of Medicine, Stanford University, Stanford, CA, USA

YVONNE SCHMIDT • Klinik für Anästhesiologie und Operative Intensivmedizin, Freie Universität Berlin, Charité-Universitätsmedizin Berlin, Berlin, Germany

VIOLA SPAHN • Klinik für Anästhesiologie und Operative Intensivmedizin,
Freie Universität Berlin, Charité-Universitätsmedizin Berlin, Berlin, Germany

SANTI M. SPAMPINATO • Department of Pharmacy and Biotechnology (FaBiT), University of Bologna, Bologna, Italy

VIVIANNE L. TAWFIK • Department of Anesthesiology, Perioperative and Pain Medicine, Department of Molecular and Cellular Physiology, Department of Neurosurgery, Stanford University School of Medicine, Stanford University, Stanford, CA, USA

ALYCIA F. TIPTON • Department of Psychiatry, University of Illinois at Chicago, Chicago, IL, USA

GABRIELE VINCELLI • Department of Pharmacy and Biotechnology (FaBiT), University of Bologna, Bologna, Italy

YADAV WAGLEY • Department of Pharmacology, University of Minnesota Medical School, Minneapolis, MN, USA

LI-NA WEI • Department of Pharmacology, University of Minnesota Medical School, Minneapolis, MN, USA

GUILLERMO ARIEL YUDOWSKI • Department of Anatomy and Neurobiology, School of Medicine, University of Puerto Rico, San Juan, Puerto Rico; Institute of Neurobiology, University of Puerto Rico, San Juan, Puerto Rico

MARY TRESA ZANDA • Division of Neuroscience and Clinical Pharmacology, Department of Biomedical Sciences, University of Cagliari, Cittadella Universitaria di Monserrato, Monserrato, CA, Italy

Contributors xv

Part I and Post-transcriptional Analysis of Opioid Receptors

Genetic, Structural Biology, Transcriptional

Chapter 1 Overview of Genetic Analysis of Human Opioid Receptors

Santi M. Spampinato Abstract

The human μ-opioid receptor gene (OPRM1), due to its genetic and structural variation, has been a target of interest in several pharmacogenetic studies. The μ-opioid receptor (MOR), encoded by OPRM1, con- tributes to regulate the analgesic response to pain and also controls the rewarding effects of many drugs of abuse, including opioids, nicotine, and alcohol. Genetic polymorphisms of opioid receptors are candidates for the variability of clinical opioid effects. The non-synonymous polymorphism A118G of the OPRM1 has been repeatedly associated with the efficacy of opioid treatments for pain and various types of depen- dence. Genetic analysis of human opioid receptors has evidenced the presence of numerous polymor- phisms either in exonic or in intronic sequences as well as the presence of synonymous coding variants that may have important effects on transcription, mRNA stability, and splicing, thus affecting gene function despite not directly disrupting any specific residue. Genotyping of opioid receptors is still in its infancy and a relevant progress in this field can be achieved by using advanced gene sequencing techniques described in this review that allow the researchers to obtain vast quantities of data on human genomes and transcrip- tomes in a brief period of time and with affordable costs.

Key words Exon, Gene polymorphism, Intron, Mu-opioid receptor, Mutation, Next generation sequencing, Opioid receptor genes

1 Introduction

Opioid receptors belong to the rhodopsin family of G-protein coupled receptors (GPCRs) and modulate downstream signaling through interactions with heterotrimeric G proteins. They are clas- sified into μ-opioid receptor (MOR), δ-opioid receptor (DOR), and κ-opioid receptor (KOR) and correspond to the OPRM1, OPRD1, and OPRK1 genes, respectively. These receptors have seven transmembrane domains, three intracellular loops, three extracellular loops, an extracellular N-terminus, and an intracellu- lar C-terminus. All three receptors present a high homology within the transmembrane domains, which are arranged in a helical pat- tern, but have less homology in the extracellular regions. There are also many similarities in their binding pockets that, once activated by an agonist, may result in activation of the opioid receptor and

Santi M. Spampinato (ed.), Opioid Receptors: Methods and Protocols, Methods in Molecular Biology, vol. 1230, DOI 10.1007/978-1-4939-1708-2_1, © Springer Science+Business Media New York 2015

Santi M. Spampinato

subsequent downstream signaling. Differences of the extracellular loops contribute to influence ligand–receptor interaction and allow varying degrees of selectivity between different endogenous opioid peptides and opioid receptors. MOR is activated by met-enkephalin, endomorphins, and β-endorphin. Leu-enkephalin and deltorphin have been shown to activate DOR, while dynorphins constitute a family of peptides specific for the KOR. Many of the endogenous opioid peptides show some affinity for more than one opioid recep- tor. Differences among intracellular regions contribute to the spec- ificity of downstream signaling and account for the different pathways activated by the opioid receptors [1–3].

Intracellular receptor domains interact with heterotrimeric Gi/ G0 proteins, which when activated by the binding of an agonist modulate the separation of α and βγ subunits. G protein subunits may alter ion channel activity and decrease cell membrane poten- tial, as well as activate MAPK pathways producing changes in gene expression [4]. Albeit the different opioid receptors display similar signaling pathways, the differences in the intracellular domains of MOR, DOR, and KOR result, when activated by agonists, in the expression of different phenotypes. MOR and DOR activation contributes to analgesia and results in rewarding effects, while KOR activation may cause aversion and dysphoria. Opioid recep- tors may form heterodimers that also occur in vivo and have been shown to regulate unique phenotypes different from those modu- lated by individual receptors, adding further complexity to opioid receptor signaling [5].

Genetic Polymorphisms of Opioid Receptors

Genome sequencing of different ethnic groups has identified 3,324 polymorphisms in the OPRM1 gene, which occupies a region of approximately 200 kb on the long arm of chromosome 6 [6]. The majority of these polymorphisms displays a low frequency and seems to possess a limited relevance at the population level. However, only 1,395 of the known genetic variants have allele fre- quencies greater than 1 % in the considered population. The most common and most studied non-synonymous SNP is rs1799971; this polymorphism is located in exon 1, where a change from ade- nosine (A) to guanosine (G) in nucleotide position 118 (A118G) results in a change in amino acid sequence in which asparagine (Asn) 40 is replaced by aspartic acid (Asp) (designated N40D) and occurs more frequently in non-African populations [6]. Manglik et al. [7] have confirmed that after elimination of this extracellular domain, the basic three-dimensional structure of the receptor is not modified.

Previous studies have reported that this SNP could be associ- ated with addictive behaviors for several drugs, but more extensive

Genetic Analysis of Opioid Receptors 5

studies did not confirm these preliminary observations, and the A118G polymorphism has been reported to either increase or reduce the risk of substance abuse.

Other investigators have showed that pain-related evoked poten- tial responses were lower among individuals with G alleles compared to AA homozygotes [8]. However, other researchers did not confirm any association between OPRM1 SNPs (including A118G) and increased pain sensitivity or chronic widespread pain [9].

As regards ethnic diversity, in comparison to non-Hispanic whites, African Americans report higher levels of pain and disabil- ity associated with several pain conditions [10–15]. In addition, higher experimental pain sensitivity among African Americans has been observed [16, 17].

Though literature is limited regarding Hispanics, higher pain and inadequate analgesia have been reported [18]. Among Asians, carriers of the rare G allele on A118G revealed generally increased clinical pain and analgesic response differences [19], although lim- ited ethnic group comparisons exist.

Previously, Bond et al. [20] have reported that A118G resulted in increased signaling through MOR by the endogenous opioid peptide beta-endorphin; however, recently, other studies did not confirm this hypothesis [21, 22].

Several functional effects have been linked to the A118G poly- morphism. The G allele of A118G creates a novel CpG-methylation site, preventing upregulation of OPRM1 in response to prolonged opioid administration [22]. mRNA with the variant G allele is less abundant in human brain than the A allele and studies on cell lines have ascertained that the A118G variant may reduce expression of MOR at the cell surface [23, 24]. Decreased accumulation of the second-messenger cAMP transfected cells was observed in the presence of morphine, methadone, and DAMGO [24]. This reduced signaling, following DAMGO activation, has also been shown in human postmortem brain tissue [25]. In contrast, other data suggest that β-endorphin has higher binding affinity and increased signaling at the different opioid receptors [20].

In addition to genetic variation, the OPRM1 gene also displays significant structural variations. Alternative splicing of 15 known exons produces at least 23 previously described splice variants; 16 of these variants may be translated into protein products [26]. Despite the large number of total exons, individual splice variants contain only 3–5 exons. The 3′ UTR of OPRM1 is also known to vary in size, with some isoforms in both mice and humans known to have UTRs greater than10 kb in length [27]. Considering the potential role of 3′ UTRs in regulating transcript expression through miRNA binding and other mechanisms, the UTR length in OPRM1 may participate in the regulation of protein levels of the different isoforms [28].

6 Santi M. Spampinato

Fig. 1 Mutations in the OPRM1 gene related to the exonic organization. (a) Locations of polymorphisms in the gene. (b) Exonic organization of mRNA. (c) Sequence of the mu-opioid receptor protein. The position of five mutations that produce an amino acid exchange frequently reported (≥5 %) is indicated in the protein sequence

Other OPRM1 polymorphisms are found at high frequency, and therefore any functional consequence would be relevant for opioid therapy [6]. In the OPRM1 sequence 24 SNPs have been identified that cause an amino acid exchange or were proposed to cause any functional consequence or occurred frequently. However, only five SNPs have a reported frequency of at least 5 %: G-172T, C17T, A118G, IVS2–31G>A, and IVS2–691C>G (Fig. 1).

Lötsch and Geisslinger [29, 30] have showed that the IVS2- 691C>G (44.5 %) and the IVS2-31G>A (8.9 %) SNPs in intron 2 do not affect opioid pharmacodynamics whereas G172T and C17T SNPS have been poorly studied. Therefore, at the moment, the clinical interest remains strictly restricted to the A118G SNP.

The human OPRK1 gene is located on chromosome 8q11.2. It presents at least four major exons and three introns, and the 3′-UTR region of 3,096 nucleotides in length [31]. The G36T SNP (rs1051660) may be associated with drug dependence. Xuei et al. [32] have examined 13 SNPs throughout OPRK1 gene in alcohol-dependent Caucasians and observed a number of the gene variants linked with an increased risk for alcohol dependence. An insertion of 830 bp was found 1,389 bp upstream of the transcrip- tion start site of OPRK1 [33] and it has been proposed that this could be associated with alcohol dependence in Caucasians. OPDR1 gene polymorphisms have been poorly investigated [34].

Genetic Analysis of Opioid Receptors 7

3 Analysis of Gene Variants of Opioid Receptors

Structural variations of opioid receptors comprise different types of genomic variants including deletions, duplications, inversions, transpositions, translocations, and complex rearrangements. Different individuals may differ by thousands of variants. Approaches to identify the genetic basis of rare Mendelian disor- ders have been largely based on well-established techniques such as positional cloning and linkage analysis followed by targeted candidate gene screening.

Recently, investigations of these rare Mendelian disorders have received a great contribution due to the technical advance of a new DNA sequencing technology termed “next generation sequencing,” also known as deep resequencing, or massively parallel sequencing, which is speeding up the investigation of rare disorders [35].

DNA sequencing was originally developed in 1975 by Sanger and Coulson [36] that adopted methods based on polyacrylamide gel electrophoresis, which allows DNA fragments to be distin- guished by their size. This technique is still used widely today. This “Sanger sequencing” or first generation sequencing is founded on the use of oligonucleotide primers on either side of the selected DNA sequence followed by the addition of DNA polymerase and a mixture of nucleotide “building blocks” enabling the generation of multiple copies of the original DNA sample. The use of chain terminating nucleotides in 1977 [37] allowed the generation of a whole array of different copies of the original DNA sequence “chain stopped” at all possible lengths, which are then separated out on gel or capillary system by electrophoresis. Using known specific labeled nucleotides (A, C, T, or G) it is possible to assem- ble the original DNA sequence.

In 1977 Maxam and Gilbert [38] published a sequencing method employing radioactive labeling of double-stranded DNA fragments. The DNA was then cleaved by base-specific chemical reactions and the fragments separated by electrophoresis. In the same year was published another method that was an improvement of Sanger’s method with the dideoxy or chain termination method. Instead of chemical cleavage of the DNA, the process depends on 32P-labeled chain-terminating dideoxy nucleotides, which prevent further extension of the sequence upon incorporation. Each reac- tion generates fragments of increasing size, ending at the base specified by the reaction, i.e., each A, T, C, or G. In 1986, Leroy Hood at Caltech, in collaboration with Applied Biosystems (ABI), published the first report of sequencing data being collected though a computer [39]. This technology, based on Sanger’s dide- oxy method, uses sequencing primers fluorescently end-labeled with four different colors to represent each base. Reactions are then run simultaneously through a polyacrylamide tube gel, with

8 Santi M. Spampinato

the DNA recognized by its fluorescence as it passes a detector. Innovative ABI instruments were released in the following years with dedicated sequencing facilities set up with the eventual aim of sequencing the human genome. The development of the Applied Biosystems capillary sequencer (ABI 3700) in the late 1990s allowed simultaneous sequencing of up to 96 samples through separate capillaries filled with non-cross-linked polymer matrix [40]. This method has been used in several studies focused on human receptor genotyping [41, 42]. A further improvement was the ABI TaqMan SNP genotyping assay (Applied Biosystems) adopted to carry out OPRM1 genotyping by Bortsov et al. [43], Wang et al. [44], and Ashenhurst et al. [34].

Rhodin et al. [45] have employed the Handy Bio-Strand method for the SNP genotyping of OPRM1 A118G (rs1799971). Briefly, the amplified DNA was spotted on a micro-porous nylon thread (Bio-Strand) and hybridized with allele-specific oligonucle- otide competitive hybridization. The Cy5 oligonucleotide Cy5- Tag1 was used as a landmark.

Next generation sequencing may also enable the elucidation of the contribution of rare alleles in common disorders, potentially offering significant breakthroughs in our understanding. Furthermore, based on current thinking, the whole genome does not need to be sequenced to identify polymorphisms restricted to opioid receptors.

Next generation sequencing consists upon multiple, short, overlapping reads of fragments of DNA which can be aligned against a reference genome or assembled “de novo” if no any information of the reference genome is available [46]. It is more faster than the previously available Sanger-sequencing method but, due to the need for overlapping reads to allow fragments to be aligned, the required number of reads per nucleotide position is increased. This means that thousands or millions of pieces of DNA can be sequenced at the same time. This technique is less affordable in genomic regions with extensive nucleotide repeats.

Eighty-five percent of pathogenic mutations causing Mendelian disorders are found within the segments coding for proteins (exons) [47], which collectively are referred to as the “human exome” [48]. Whole “exome” sequencing uses this technique except that complementary strands to known exons (i.e., the protein-coding regions of each gene) are used to extract fragments covering the exonic regions of a gene prior to start sequencing. Initially DNA is fragmented into multiple short segments known as “shotgun library”; thereafter, adaptors are bound to the ends of each fragment. The adaptors consist of short sequences of DNA that have priming sites within them for the subsequent amplifica- tion steps. The segments of DNA (complete with adaptors) are then mixed with probes that correspond to specific regions within the exome. The shotgun library is then “enriched” for the

Genetic Analysis of Opioid Receptors 9

sequences of interest, using beads or a solid plate to allow physical separation of the exome from remaining DNA, and this is washed away. Custom arrays can be designed to enrich for specific groups of genes of interest, the whole exome, and exon-flanking regions.

Several manufacturers, adopting different techniques, sell next generation sequencing platforms [46, 49]. The most common DNA sequencing techniques comprise the Illumina (previously known as Solexa) and the 454 method (also known as Roche FLX).

In the 454 method, DNA is broken into small fragments that are attached to DNA adaptors. Thereafter, small beads containing brief oligonucleotide sequences matching parts of the adaptors are added. Thus, one DNA fragment binds to each bead. DNA stands are amplified on beads and denatured to obtain single-stranded fragments. Single beads are transferred into wells on a plate together with polymerase enzyme beads for sequencing. In this technique, the pyrosequencing method has been adopted that allows shotgun sequencing without cloning any of the DNA. Pyrosequencing involves a DNA synthesis reaction, where each of the four dNTP bases is applied one after another. During the DNA synthesis reaction, nucleotides are added separately (i.e., only A, then T, then C, then G) and a phosphate group is released when a nucleotide is incorporated. Pyrosequencing method mea- sures the amount of phosphate released as each dNTP is added to the reaction and incorporated, allowing determination of the sequence of each fragment [50]. This method can produce one million bases of sequence with 99.5 % accuracy [49, 50].

In the Illumina method, DNA is sheared into short fragments; then adaptors (short DNA sequences) are bound to the DNA frag- ments, and these complexes are put onto a hollow slide with a lawn of primers. DNA fragments bind to a complementary primer on the slide surface and are amplified in clusters before sequencing takes place. Once each segment of DNA is amplified, fluorescent nucleo- tides are added, together with DNA polymerase and sequencing primers. Fluorescently labeled chain-terminating nucleotides are incorporated into the sequence and measured by a detector. However, the incorporation of the chain-terminating nucleotide is reversible, allowing the synthesis to continue until another chain-terminating nucleotide is incorporated, so the bases in each sequence are mea- sured one at a time. The method can produce one billion bases of 30–40 base sequences in a single run [50].

As fluorescent tagged bases are incorporated to each strand on each bead, in real time, laser activation of the fluorescence can be read. Computers monitor each cluster, and can determine the sequence of many clusters at the same time.

Lötsch and Geisslinger [51] have employed a pyrosequencing assay to investigate OPRM1 polymorphisms. Deo and colleagues [52] have genotyped their samples using an beadchip (Illumina, San Diego, California, USA) microarray containing a total of 1,350

Santi M. Spampinato

SNPs within 130 candidate genes implicated in addiction and alcoholism. Finally, Lee et al. [53] have the Illumina GoldenGate platform to investigate OPRM1 polymorphisms.

Opioid Receptor Genotyping and Personalized Medicine

Sequencing technology has advanced massively since its birth in the 1970s. Scientists may use many technologies potentially allow- ing sequencing of whole genomes in a day for less than $1,000. The next step and a current hot topic is to provide further insight into the workings of the body in health and disease by looking at the proteins active in particular cell types. This can be achieved in part by looking at the messenger RNA (transcriptomics) and non- coding RNAs showing how genetics affects the cell system in com- bination with environmental influences. Opioid receptors are linked to different diseases and their pharmacological treatments. Many opioid analgesics are opioid receptor agonists and treatments for addiction include the use of opioid agonists or antagonists.

Pharmacogenetic analysis of OPRM1 polymorphisms could help to guide treatment decisions and patients can receive the ther- apeutic options with the best efficacy and the greatest tolerability. The vast majority of pharmacogenetic studies on OPRM1 have analyzed the effects of A118G; this represents one of the first genetic variants that may be linked with pharmacological outcome. However, intronic and synonymous coding variants in many genes have been shown to have important effects on transcription, mRNA stability, and splicing, thus affecting gene function despite not directly disrupting any specific residue. Opioid receptors have numerous genetic and structural variations, all of which are poten- tial relevant to the field of pharmacogenetics. With the speed at which next generation sequencing technology is becoming increas- ingly common [54], future studies can and must start to focus on all of the genetic variation present in the opioid receptor genes.

References

1. Pasternak GW, Pan YX (2013) Mu opioids and their receptors: evolution of a concept. Pharmacol Rev 65:1257–1317

2. Cox BM (2013) Recent developments in the study of opioid receptors. Mol Pharmacol 83: 723–728

3. Butelman ER, Yuferov V, Kreek MJ (2012) Kappa opioid receptor/dynorphin system: genetic and pharmacotherapeutic implications for addiction. Trends Neurosci 35:587–596

4. Al-Hasani R, Bruchas MR (2011) Molecular mechanisms of opioid receptor-dependent

signaling and behavior. Anesthesiology 115:

1363–1381
5. Stockton SD Jr, Devi LA (2012) Functional

relevance of mu-delta opioid receptor het- eromerization: a role in novel signaling and implications for the treatment of addiction dis- orders: from a symposium on new concepts in mu-opioid pharmacology. Drug Alcohol Depend 121:167–172

6. Crist RC, Berrettini WH (2013) Pharma- cogenetics of OPRM1. Pharmacol Biochem Behav. doi:10.1016/j.pbb.2013.10.018

7. Manglik A, Kruse AC, Kobilka TS et al (2012) Crystal structure of the μ-opioid receptor bound to a morphinan antagonist. Nature 485: 321–326

8. Lötsch J, Stuck B, Hummel T (2006) The human mu-opioid receptor gene polymor- phism 118A>G decreases cortical activation in response to specific nociceptive stimulation. Behav Neurosci 120:1218–1224

9. Holliday KL, Nicholl BI, Macfarlane GJ et al (2009) Do genetic predictors of pain sensitivity associate with persistent widespread pain? Mol Pain 5:56

10. Anderson KO, Green CR, Payne R (2009) Racial and ethnic disparities in pain: causes and consequences of unequal care. J Pain 10: 1187–1204

11. Green CR, Anderson KO, Baker TA et al (2003) The unequal burden of pain: confront- ing racial and ethnic disparities in pain. Pain Med 4:277–294

12. Plesh O, Adams SH, Gansky SA (2011) Racial/ ethnic and gender prevalences in reported common pains in a national sample. J Orofac Pain 25:25–31

13. Green CR, Wheeler JR (2003) Physician vari- ability in the management of acute postoperative and cancer pain: a quantitative analysis of the Michigan experience. Pain Med 4:8–20

14. Green CR, Baker TA, Sato Y et al (2003) Race and chronic pain: a comprehensive study of young black and white Americans presenting for management. J Pain 4:176–183

15. Green CR, Ndao-Brumblay SK, Nagrant AM et al (2004) Race, age, and gender influences among clusters of African American and white patients with chronic pain. J Pain 5:171–182

16. Campbell CM, Edwards RR, Fillingim RB (2005) Ethnic differences in responses to multiple experimental pain stimuli. Pain 113: 20–26

17. Forsythe LP, Thorn B, Day M et al (2011) Race and sex differences in primary appraisals, catastrophizing, and experimental pain out- comes. J Pain 12:563–572

18. Jimenez N, Seidel K, Martin LD et al (2010) Perioperative analgesic treatment in Latino and non-Latino pediatric patients. J Health Care Poor Underserved 21:229–236

19. Fukuda K, Hayashida M, Ikeda K et al (2010) Diversity of opioid requirements for postopera- tive pain control following oral surgery—is it affected by polymorphism of the mu-opioid receptor? Anesth Prog 57:145–149

20. Bond C, LaForge KS, Tian M et al (1998) Single-nucleotide polymorphism in the human mu opioid receptor gene alters beta-endorphin

binding and activity: possible implications for opiate addiction. Proc Natl Acad Sci U S A 95:9608–9613

21. Mague SD, Blendy JA (2010) OPRM1 SNP (A118G): involvement in disease development, treatment response, and animal models. Drug Alcohol Depend 108:172–182

22. Oertel BG, Doehring A, Roskam B et al (2012) Genetic-epigenetic interaction modulates mu- opioid receptor regulation. Hum Mol Genet 21:4751–4760

23. Zhang Y, Wang D, Johnson AD et al (2005) Allelic expression imbalance of human mu opi- oid receptor (OPRM1) caused by variant A118G. J Biol Chem 280:32618–33224

24. Kroslak T, Laforge KS, Gianotti RJ (2007) The single nucleotide polymorphism A118G alters functional properties of the human mu opioid receptor. J Neurochem 103:77–87

25. Oertel BG, Kettner M, Scholich K et al (2009) A common human micro-opioid receptor genetic variant diminishes the receptor signal- ing efficacy in brain regions processing the sensory information of pain. J Biol Chem 284: 6530–6535

26. http://www.ensembl.org. Accessed 26 April 2014

27.Ide S, Han W, Kasai S et al (2005) Characterization of the 3′ untranslated region of the human mu-opioid receptor (MOR-1) mRNA. Gene 364:139–145

28. Wu Q, Law PY, Wei LN et al (2008) Post- transcriptional regulation of mouse mu opioid receptor (MOR1) via its 3′ untranslated region: a role for microRNA23b. FASEB J 22:4085–4095

29. Lötsch J, Geisslinger G (2005) Are mu-opioid receptor polymorphisms important for clinical opioid therapy? Trends Mol Med 11:82–89

30. Lötsch J, Geisslinger G (2010) A critical appraisal of human genotyping for pain ther- apy. Trends Pharmacol Sci 31:312–317

31. Yuferov V, Fussell D, LaForge KS et al (2004) Redefinition of the human kappa opioid receptor gene (OPRK1) structure and association of hap- lotypes with opiate addiction. Pharmacogenetics 14:793–804

32. Xuei X, Dick D, Flury-Wetherill L et al (2006) Association of the kappa-opioid system with alcohol dependence. Mol Psychiatry 11: 1016–1024

33. Edenberg HJ, Wang J, Tian H et al (2008) A regulatory variation in OPRK1, the gene encoding the kappa-opioid receptor, is associ- ated with alcohol dependence. Hum Mol Genet 17:1783–1789

34. Ashenhurst JR, Bujarski S, Ray LA (2012) Delta and kappa opioid receptor polymorphisms

Genetic Analysis of Opioid Receptors 11

12 Santi M. Spampinato

influence the effects of naltrexone on subjective responses to alcohol. Pharmacol Biochem Behav 103:253–259

35. Majewski J, Schwartzentruber J, Lalonde E et al (2011) What can exome sequencing do for you? J Med Genet 48:580–589

36. Sanger F, Coulson AR (1975) A rapid method for determining sequences in DNA by primed synthesis with DNA polymerase. J Mol Biol 94:441–448

37. Sanger F, Nicklen S, Coulson AR (1977) DNA sequencing with chain-terminating inhibitors. Proc Natl Acad Sci U S A 74:5463–5467

38. Maxam AM, Gilbert W (1977) A new method for sequencing DNA. Proc Natl Acad Sci U S A 74:560–564

39. Smith LM, Sanders JZ, Kaiser RJ et al (1986) Fluorescence detection in automated DNA sequence analysis. Nature 321:674–679

40. Hutchinson CA III (2007) DNA sequencing: bench to bedside and beyond. Nucleic Acids Res 35:6227–6237

41. Proudnikov D, Randesi M, Levran O et al (2012) Association of polymorphisms of the mu opioid receptor gene with the severity of HIV infection and response to HIV treatment. J Infect Dis 205:1745–1756

42. Hastie BA, Riley JL 3rd, Kaplan L et al (2012) Ethnicity interacts with the OPRM1 gene in experimental pain sensitivity. Pain 153:1610–1619

43. Bortsov AV, Millikan RC, Belfer I (2012) μ-Opioid receptor gene A118G polymorphism predicts survival in patients with breast cancer. Anesthesiology 116:896–902

44. Wang SC, Tsou HH, Chung RH et al (2014) The association of genetic polymorphisms in the μ-opioid receptor 1 gene with body weight, alcohol use, and withdrawal symptoms in patients with methadone maintenance. J Clin Psychopharmacol 34:205–211

45. Rhodin A, Grönbladh A, Ginya H et al (2013) Combined analysis of circulating β-endorphin

with gene polymorphisms in OPRM1, CACNAD2 and ABCB1 reveals correlation with pain, opioid sensitivity and opioid-related side effects. Mol Brain 6:8

46. Metzker ML (2010) Sequencing technolo- gies—the next generation. Nat Rev Genet 11: 31–46

47. Cooper DN, Krawczak M, Antonorakis SE (1995) The nature and mechanisms of human gene mutation. In: Scriver C, Beaudet AL, Sly WS et al (eds) The metabolic and molecular bases of inherited disease, 7th edn. McGraw- Hill, New York, pp 259–291

48. Stenson PD, Ball EV, Howells K et al (2009) The Human Gene Mutation Database: provid- ing a comprehensive central mutation database for molecular diagnostics and personalized genomics. Hum Genomics 4:69–72

49. Singleton AB (2011) Exome sequencing: a transformative technology. Lancet Neurol 10: 942–946

50. Su Z, Ning B, Fang H et al (2011) Next- generation sequencing and its applications in molecular diagnostics. Expert Rev Mol Diagn 11:333–343

51. Lötsch J, Geisslinger G (2006) Relevance of frequent mu-opioid receptor polymorphisms for opioid activity in healthy volunteers. Pharmacogenomics J 6:200–210

52. Deo AJ, Huang YY, Hodgkinson CA et al (2013) A large-scale candidate gene analysis of mood disorders: evidence of neurotrophic tyrosine kinase receptor and opioid receptor signaling dysfunction. Psychiatr Genet 23: 47–55

53. Lee MR, Gallen CL, Zhang X et al (2011) Functional polymorphism of the mu-opioid receptor gene (OPRM1) influences reinforce- ment learning in humans. PLoS One 6: e24203

54. Foo J-N, Liu J-J, Tan E-K (2012) Whole- genome and whole-exome sequencing in neu- rological diseases. Nat Rev Neurol 8:508–517

Chapter 2 Computational Structural Biology of Opioid Receptors

Davide Provasi Abstract

The publication of high-resolution structures for all of the opioid receptor subfamilies has unveiled exciting opportunities for mechanistic insight into the molecular mechanisms underlying the biology of nocicep- tion, reward, and higher cognitive functions, as well as promises for progress in several clinical areas such as pain management, physiological dependence, addiction, and mood disorders. To turn this promise into novel and improved therapeutic entities, however, this information needs to be supplemented with research strategies that explore the dynamic behavior of the proteins and their interactions with other receptors and ligands in their physiological environment.

Here we describe state-of-the-art molecular dynamics computational protocols, based on all-atom and coarse-grained modeling techniques, designed to estimate crucial thermodynamic and kinetic param- eters describing the binding of small-molecule ligands and the formation of supramolecular complexes.

Key words All-atom models, Biased sampling techniques, Coarse-grained models, Dimerization, Ligand binding, Metadynamics, Molecular dynamics, Weighted histogram analysis method, Umbrella sampling

1 Introduction

One century ago M. von Laue won a Nobel Prize for envisioning how x-ray diffraction by crystals could be used to determine their structure to atomic resolution, and for inaugurating a powerful way to look directly at the nest details of matter. Soon, the same techniques were applied to small biological molecules (~1930s) and globular proteins (~1950s), ushering in a new era in the study of life sciences. In the last decade, despite the additional complexi- ties characteristic of membrane protein crystallization, several tech- nological advances have paved the way to the production of high-resolution crystal structures of G-protein coupled receptors (GPCRs) in general, and the opioid receptor (ORs) family in particular, providing priceless insight into the biology of these important proteins.

Santi M. Spampinato (ed.), Opioid Receptors: Methods and Protocols, Methods in Molecular Biology, vol. 1230, DOI 10.1007/978-1-4939-1708-2_2, © Springer Science+Business Media New York 2015

14 Davide Provasi

While it is dif cult to overestimate the importance of crystal- lography data for modern structural biology, high spatial resolution comes at a price.

First, except in very rare cases, biological macromolecules in cells are not in an ordered crystal phase. Thus, while crystal struc- tures are invaluable information, the effects of the crystal packing and of the manipulations needed to produce such highly nonphysi- ological states need to be considered when formulating hypotheses about physiologically relevant cases. For example, the structure may be in uenced by interactions with neighboring proteins across the boundary of the crystal unit cell and by the nonphysiological topology of the hydrophobic phases of lipids and detergents in the cr ystal.

Second, biological processes are intrinsically dynamic. Therefore, the static information obtained from crystallography must necessarily be supplemented by other techniques, including resonance spec- troscopies (such as NMR, EPR, DEER, and ESR), uorescence energy transfer (FRET, BRET), and computation. Despite the pos- sibly lower spatial accuracy, these techniques can extend the explora- tion to the time domain and provide information on the dynamics of the system on different time scales.

The recent resolution of the rst OR crystal structures [1–5] represents a long-awaited breakthrough in the structural biology of opioid signaling that had been based previously on indirect models produced by homology to other GPCRs. These data sup- ply fundamental insight for the validation of existing hypotheses and the formulation of new ones regarding the biology and the pharmacology of pain, addiction, and mental disorders [6].

Of particular relevance to drug design and optimization are the mechanistic details governing ligand recognition and binding, as well as the ow of information from the orthosteric binding site to the intracellular region and the coupling to downstream signal- ing partners (G-protein and arrestins).

Equally important is the clari cation of the role of supramo- lecular assemblies (dimers and oligomers) of ORs, as well as the functional role of both homodimers and heterodimers in opioid biology [7], in particular given the experimental evidence of phar- macologies speci c to these complexes.

Undoubtedly, a complete picture will necessarily bene t from future OR crystals capturing the active forms of the proteins as well as from input from other structural and biophysical experiments. Nonetheless, computational techniques are valuable tools to supply accurate structural information and direct the design of molecules with speci c binding selectivities (including heteromer selective compounds) and kinetics, with selective agonism in speci c signal- ing pathways, or allosterically modulating the action of other drugs and endogenous ligands.

1.1 Opioid Crystal Structures

In the following sections, after a brief review of the OR crystal structures, we will describe how state-of-the-art computational methods can be used to complement crystal structure data to address some of the issues raised above. Speci cally, we will describe molecular dynamics (MD) based computational protocols to study ligand binding to OR and to investigate the stability and kinetics of dimerization interfaces.

Crystal structures of representatives of each of the four opioid subfamilies (the μ-OR [1], δ-OR [4], κ-OR [2], and the noci- ceptin, or ORL-1 receptor [3], respectively MOR, DOR, KOR, and NOR) were simultaneously published in May 2012. These structures were made possible by a skilful combination of several technological breakthroughs, most speci cally the reconstitution of the receptors in lipidic cubic phase [8, 9], and replacement of exible regions of the protein with T4 lysozyme (T4L) [10] or fusion with the thermostabilized apocytochrome b562-RIL (BRIL) [11]. Binding of antagonist ligands was also used to stabilize inac- tive states, thereby reducing the conformational diversity of the proteins. These techniques have allowed an explosive increase in the number of GPCR crystals obtained in the last few years.

The murine (mus musculus) MOR was solved with a morphi- nan antagonist (BF0) covalently linked to K233 (5.39, in the Ballesteros-Weinstein notation, in which the helix number is fol- lowed by a residue index offset so that the most conserved residue is assigned the arbitrary position 50), to a resolution of 2.8 Å (PDB: 4DKL [1]); the human KOR was solved with a highly selec- tive antagonist characterized by a very low dissociation rate (JDTic), to a resolution of 2.9 Å (PDB: 4DJH [2]). In these struc- tures, the long intracellular loop (IC) 3 was substituted with T4L to enhance crystallization. Substitution of the BRIL bundle for the amino-terminal tail of the receptor afforded the crystallization of the human NOR bound to a mimetic of the endogenous peptide antagonist N/OFQ, to a 3.0 Å resolution (PDB: 4EA3 [3]). Finally, a rst structure of the murine DOR was reported to a reso- lution of 3.4 Å (PDB: 4EJ4 [4]); the receptor is in complex with the highly selective morphinan antagonist naltrindole.

In February 2014, the Stevens group published a second, high resolution (1.8 Å, PDB: 4N6H [5]) structure of the human DOR receptor, also in complex with naltrindole, using the same BRIL substitution previously employed for the NOP structure.

Given the high average pairwise sequence identity among the ORs (71 % between the main subfamilies MOR, KOR, and DOR; 61 % between any of those and the NOR receptor), it does not come as a surprise that the transmembrane (TM) bundles exhibit few structural differences. This is also true for some of the less con- served regions; for instance, the long extracellular loop (EC) 2 also displays remarkable structural conservation, adopting a beta-sheet

Computational Biology of Opioid Receptors 15

16 Davide Provasi

conformation in all the crystals. Interestingly, the EC3, although shorter than EC2, could not be solved in the KOP structure and displays high temperature factors indicating high exibility in all the crystals, and one single difference between the murine and the human gene of the DOP receptor in this region (an asparagine N instead of the aspartic acid side chain D290) [5] results in a differ- ent arrangement of the loop and in different chemical properties lining the extracellular rim of the binding pocket.

The importance of the region formed by EC3 and the extracel- lular ends of TM6 and TM7, as a “selectivity lter” for peptide and classic opioid binding, has been reported previously [12], and is now also highlighted by the binding poses of morphinan ligands in the OR crystal structures. In the framework of the address/ message paradigm of opioid receptor pharmacology, classical opi- oids with a morphine-like scaffold bind towards this region of the binding pocket, whereas non-morphinan opioids are selected by addresses on the opposite side of the bundle (TM2/3). This dem- onstrates how an understanding of the exible regions of the bind- ing pocket is necessary for a thorough description of protein–ligand interactions.

Two of the OR crystal structures display interfaces between parallel receptors. It is tempting to infer from such evidence that similar dimeric arrangements would be formed also under physio- logical conditions, and may thus play a role in OR biology. The MOR crystal contains two different interfaces between the pro- teins. The rst is a tightly packed arrangement, involving helices TM5 and 6 from each protomer, similar, albeit not identical, to an interface observed in the CXCR4 crystal (PDB: 3ODU, [13]). We note that, while TM5 has been implicated in dimerization of other GPCRs [14, 15], the conclusion of those experiments also revealed simultaneous involvement of TM4 and IC2 that are not at the interface in the crystal.

The MOR structure also reveals a second, less compact inter- face involving the extracellular halves of helices TM1 and 2 and the amphiphilic helix 8. A similar interface, comprising the same heli- ces but featuring different contacts, appears in the KOR crystal. Notably, the participation of these helices in dimerization has also been reported in the past [15, 16].

Both DOR crystals display antiparallel dimers (i.e., complexes in which the extracellular region of one protomer is in contact with the intracellular region of the other), whose existence and functional relevance under physiological conditions is dif cult to envisage.

Overall, it is clear that the interface data from the crystal structures, while representing an interesting additional piece of information regarding the putative features of the dimers, does not provide a clear answer to questions surrounding the detailed descri- ption of OR quaternary structure. We also note that a complete

2 Materials

characterization of the structural details of the interfaces is hardly enough to reach a full understanding of the biological signi cance of these receptor complexes. If, as data seem to support [7], several dif- ferent interfaces are possible, their relative stability and the kinetic details of their formation will be critical to understanding the func- tional relevance of the dimeric states.

In the next sections we will review some of the computational methodologies that address the stability and kinetics of ligand bind- ing and receptor oligomerization in a fully dynamic framework.

Computer programs implementing state-of-the-art algorithms for molecular dynamics and suitable for the simulations described in the following are widely available. The choice of which package to employ largely depends on the features of the available hardware resources and on personal preference, and will not be discussed here. Among the most popular codes we mention Gromacs (whose core development is currently taking place at the Department of Cellular and Molecular Biology, Uppsala University, and at the Stockholm Bioinformatics Center, Stockholm University, Sweden, http://www.gromacs. org); NAMD (developed by the Theoretical and Computational Biophysics Group at the University of Illinois at Urbana-Champaign, http://www.ks.uiuc.edu/Research/namd/); Amber (developed in an wide joint collaboration between groups at Rutgers University and several other institutions, http://www.ambermd.org); and Desmond (developed by D. E. Shaw Research, freely available for noncommercial use at www. deshawresearch.com and distributed by Schrödinger for commercial purposes). These packages natively implement the algorithms described in the following sections for umbrella sampling; some of them (e.g., NAMD, Desmond) also implement metadynamics. Umbrella sampling and metadynamics, along with a number of addi- tional enhanced sampling strategies, can also be provided through the Plumed plugin code (www.plumed-code.org).

Understanding the molecular mechanisms of ligand binding to ORs is an endeavor of paramount pharmacological relevance. Speci cally, the interactions that drive binding selectivity (i.e., the ability of a molecule to bind to certain OR subtypes and not to oth- ers), ef cacy (i.e., the resulting effect on signaling of the binding event), as well as different agonism pro les (i.e., selective activation of the G-protein vs. the arrestin pathways) are subjects of an intense research effort, with the nal goal of designing molecular entities with optimal clinical ef ciency.

Computational Biology of Opioid Receptors 17

3 Methods

3.1 Reasonable Expectations
for Binding Studies

18 Davide Provasi

3.2 System Preparation

To this end, the insight coming from crystal structures must be extended to the time domain, providing information on the bind- ing pathways, kinetics, and thermodynamics, as well as the modu- lation induced by the ligand on the dynamics and on the conformational free-energy of the receptor.

When designing binding studies, it should be kept in mind that the structural details captured in the crystal structures re ect very speci c conditions determined by the crystallogenesis. As noted above, (1) all the OR structures available so far were solved in com- plex with antagonist ligands; this stabilizes a homogeneous popula- tion of proteins in an inactive conformation, thereby increasing the probability of forming regular crystals. Moreover, (2) the presence of one speci c ligand in uences the binding pocket conformation; nally (3) the exible regions of the protein (e.g., the EC and IC loops) can be in uenced by crystal packing and interactions with fusion partners.

In computational terms, these three orders of problems trans- late into the impracticality of achieving suf cient sampling of the con gurations relevant for the binding process. A second, unre- lated, but critical problem is the accuracy of the model used to describe the molecular interactions between the protein, the ligand, and their environment. Thus, sampling and force- eld accuracy still prevent MD-based methods from being a reliable black-box tool to predict quantitatively accurate absolute binding free-ener- gies and to implement virtual screening techniques.

Nonetheless, if used properly, these techniques can offer valuable insight regarding key details of the binding process. While estimates of the binding thermodynamics can be obtained experi- mentally with relative ease, simulation affords the identi cation of binding pathways, intermediates, and structural details that can help rationalize the stability of the drug–protein complex and the kinetics of its formation, and modulate these crucial pharmacologi- cal parameters by molecular engineering.

Simulation results are obviously only reliable if the chemistry and the physics implemented in the model re ect the interactions rel- evant for the process being studied. A complete description of the interactions between the ligand and the protein requires quantum effects to be included in the simulation model. This, however, adds a layer of complexity to the simulations and drastically increases the computation cost, compromising the sampling of important degrees of freedom such a side chains rearrangements, solvation, and de-solvation of the ligand.

While the development of polarizable force elds could provide an optimal trade-off between model realism and computational cost (e.g., in cases where charge transfer interactions might be important [17]), these strategies are still not ripe for routine use in free-energy prediction studies, and a classical description seems the

3.3 Unbiased Binding Simulations

most reasonable choice for the study of binding to opioid receptors (and to GPCRs in general). Within this approximation, however, extreme care must be taken to model the protein, the ligand, and the membrane lipids with high accuracy.

For small molecules, different parameterization strategies [18–20] have been proposed, requiring different levels of discretion and chemical savvy by the user. Recently, an automated strategy in the framework of the Charmm force eld has been proposed and implemented in the ParamChem web server [20]. The approach automatically identi es the most accurate parameters in a curated set of molecular fragments, and assesses the reliability of the pro- posed parameters through a quantitative penalty score (see Note 1). Finally, protein preparation should follow standard guidelines for high-quality MD simulations (see Note 2).

The recent increase in MD ef ciency on both general and special purpose [21] machines has afforded simulations on time scales (~0.1 ms) that make possible the study of binding of ligands to proteins with unbiased simulations. Such simulations show that molecules placed in the simulation box beyond 20 Å from the pro- tein surface can generally bind to the protein in approximately 1 μs [22–24]. Attempts to obtain numerical estimates of the kinetic and thermodynamic parameters from these simulations [22] have yielded results in qualitative agreement with experiments. The on- rate can be estimated modeling binding as a Poisson process, with probability

Prob(N

n = n) = exp(-koncLt )(koncLt )

Computational Biology of Opioid Receptors 19

where Nt is the number of binding events that occur in time t, and cL is the concentration of ligand molecules in the simulation box. The expected number of binding events in time t is E(Nt) = cL × kon × t, so that an unbiased estimation of the Poisson rate is

k ˆ o n = Nc L t t

that provides a way to estimate the rate based on the number of binding events. To put this value in context, however, we observe that the variance of the same estimator is given by

ˆ
Varkon= 1 =1

ˆ2 ˆN kon cLt ́kon t

This relation reveals why very long simulations need to be per- formed with multiple copies of the ligand in the simulation box to increase the number of observed binding events and reduce the

Davide Provasi

3.4 Simulations

relative variance (see Note 3). Using data from several simulations totaling to more than 120 μs, and observing 12 binding events [22] obtain a kon for alprenolol on the β2-adrenergic receptor in good agreement with measured values. However, we note that such unbiased strategy can hardly be used to estimate ligand dis- sociation rates (see Note 4).

So far, no studies have been published on OR using this approach.

A promising alternative approach is to use unbiased simula- tions to produce data to construct Markov state models, and to effectively extend the description of the system at a coarse-grained level to the time scale of biological processes (see, for instance, reviews [25] and [26]). Within this framework, a large number (on the order of hundreds) of relatively short (on the order of 100 ns) simulations are analyzed to provide a coarse-grained description of the states sampled by the system, and to estimate transition prob- abilities between such states.

Markov state models have been applied with success to a num- ber of binding studies, with varying aggregated simulation lengths (for instance 13 μs [27], 50 μs [28], 148 μs [29]). Since no appli- cation has been reported so far investigating the binding properties of ligands to GPCRs, we will not discuss them further, except for mentioning that they represent interesting and complementary alternatives to biased simulation strategies.

Several biased techniques based on all-atom MD have been pro- posed and applied to study ligand binding and allow the estimation of af nities. A thorough description of computational methods to calculate absolute free-energies of binding is beyond the scope of this work, and we refer the interested reader to one of the many excellent reviews (see for instance [30]). Here, we will limit our- selves to some general considerations, a number of important prin- ciples that should be kept in mind while designing experiments for binding studies at OR, and a few illustrating examples.

Very broadly, practical computational protocols to calculate free-energy differences between two states can be classi ed as either (1) “alchemical” free-energy perturbation strategies, where the end points are connected by states with nonphysical interactions (e.g., states with scaled interactions between the ligand and the protein, as in decoupling methods) or (2) strategies where the ligand is physi- cally separated from the receptor either by nonequilibrium pulling [31, 32] or by a direct calculation of a potential of mean force (PMF) by umbrella sampling [33, 34] or metadynamics [35].

While in their simpler form alchemical protocols are generally simpler to use, methods based on direct calculation of the PMF are better suited to incorporate improved sampling strategies for degrees of freedom such as protein and ligand exibility and solvation [36]. In opioids, the long EL2 can act as a exible lid, modulating the

Biased Binding

Computational Biology of Opioid Receptors 21

accessibility to the orthosteric binding site. Moreover, while several important opioid ligands are rather rigid, ligand exibility must be taken into account for larger small molecules and, especially, for pep- tides. Finally, it is worth noting that alchemical transformations of charged ligands involve additional complications to correctly account for electrostatic self-energy terms [33].

PMF-based methods, moreover, can identify physical interme- diate states, supplying structural information about the binding pathways, which, as we have pointed out earlier, represent a good part of what MD-based computational binding studies are better suited for. As discussed above, binding of diffusible ligands to GPCRs in general and OR in particular most likely involves binding to intermediate entry states in the extracellular vestibule before the ligand can reach the orthosteric binding site in the TM bundle [22].

Among PMF-based strategies, metadynamics techniques allow to adaptively tune the necessary trade-off between the accuracy and scope of the sampling. In the absence of detailed experimental information about the binding process, it is possible to start with a qualitative [37] or low-resolution free-energy estimation to identify relevant states and conformations, followed by a re nement study.

There is no established protocol that has been shown to work better than others for the study of ligand binding to opioids. We will therefore describe the general ideas that can be used to address the technical requirements described above in the context of any PMF-based simulation strategy; the reader should keep in mind that these are only general guidelines.

The primary phase space direction to be sampled in association studies is clearly strongly correlated to the distance between the ligand and the binding site; this distance itself is thus one of the natural starting points for biased simulations.

The inclusion of collective variables that enhance the sampling of loop conformations or ligand internal degrees of freedom must be assessed based on the nature of the ligand and the receptor under study. For the former, a simple choice is to use just the dis- tance between the center of mass of the loop itself and the center of mass of the protein (see Note 5). Once a low resolution descrip- tion of the ligand binding pathway from the initial contact with the protein extracellular region towards the orthosteric binding pocket has been obtained, path variables [38, 39] can be used to obtain more accurate quantitative estimates for the free-energy difference along the binding process.

The techniques described above provide an ef cient computa- tional strategy to sample states with the ligand bound to different regions of the protein. To estimate the absolute binding af nity, the free-energy difference between protein-bound states and unbound states must be calculated, but because of the large volume of phase space corresponding to the latter, extensive sampling is dif cult to achieve. On the other hand, when the ligand is suf ciently

22 Davide Provasi

Fig. 1 Typical collective variables used in the study of ligand binding include (1) the distance between the ligand center of mass and the center of the orthosteric binding pocket (AC), (2) the distance between EL2 center of mass and the ortho- steric binding pocket (BC), (3) the angle de ned by the ligand, the orthosteric binding pocket, and EL2 (ACB). The limits of the conical region sampled to esti- mate the absolute free-energy difference to a reference state in the bulk solvent is also represented

distant from the protein surface, we expect the free-energy to be translationally invariant. We can thus overcome the bulk sampling problem by restricting the ligand to exploring a limited region of space that connects the bound state to an unbound reference. We report here the derivation for the case of a conic restraint used in [39] (Figs. 1 and 2), and refer the reader to the original formula- tion in [33] for the case of a cylindrical constraint. More advanced constraint shapes, such as funnel-shaped controls [40], have also been proposed. We note that similar expressions are also used to calculate the dimerization constant in Subheading 3.8.4.

A biased simulation is set up to obtain the free-energy w(r) as a function of the distance between the ligand and the protein in the presence of additional external potentials constraining the ligand to be within a conical region containing the binding site and extending into the bulk. The binding af nity can be expressed in terms of the full potential of mean force W(x) as

K-1 =òdxe-(W(x)-W(x0))/kBT S

where x0 is a reference position in the bulk, and Σ is the region of the phase space corresponding to the bound ligands. It is impor- tant to observe that the free-energy pro le W(x) is different from

Computational Biology of Opioid Receptors 23

Fig. 2 Reconstruction of the full binding af nity using two separate metadynamics runs for the sampling of protein bound states (1, 2) and protein unbound states (2, 3). The unrestricted free-energy pro le and the PMF reconstructed from the restrained simulation are represented as a dashed and a solid line, respectively

w(r) because of the constraints limiting the explored region; speci cally, switching to polar coordinates, the relation between the two pro les is given by,

e-W(r)/kBT =CòJdwe-W(r,w)/kBT W

where r is the polar distance, ω represents the polar angles, J is the Jacobian determinant, and the integral is calculated on the sampled region Ω. C is a normalization constant that can be xed observing that invariance requires that W does not depend on the angular degrees of freedom ω at r0. Thus

-1 2 W(x0)/kBT C Wr0e

where |Ω| denotes the restraint’s solid angle. Using this expression we can rewrite the relation between the full free-energy W and the sampled pro le as

e-w(r)/k T = 1 òJdwe-éW(r,w)-W(r0,w)ù/kBT Bëû

Wr2 0W

so that the binding af nity can be expressed as

K-1 =Wr02òdre-w(r)/kBT S

24 Davide Provasi

This constant re ects the thermodynamic equilibrium of the bimolecular reaction corresponding to the formation of a protein ligand complex Σ. To obtain a binding constant re ecting the con- centration of ligands in the orthosteric binding pocket, the free- energy difference between the intermediate bound state Σ and the orthosteric site must be calculated. This can be easily accomplished using standard relations (Fig. 2), and expressing the full binding constant K31 as

K-1=1p1=æ1p2öp1=K-1e-DG /kT ç÷ 21B

31 [L ] p [L ] p p 32 3è3ø2

3.5 Reasonable Expectations
for Dimerization Studies

where p1 and p3 are, respectively, the probabilities for the ligand to be in the binding pocket or in the bulk solvent, and p2 is the probability of occupying the intermediate state Σ; [L] is the ligand concentration, K32 is the binding constant to the intermediate state, and ΔG21 is the free-energy difference between the interme- diate and bound state.

It should be kept in mind that any free-energy result is an esti- mate that crucially depends on the sampled molecular probability distributions, and thus that the statistical uncertainty of the nal result should always be identi ed and reported (see Note 6). Finally, we call attention to the fact that the errors calculated only re ect the statistical uncertainty of the estimator used to derive the free-energy: additional contributions to the error due to insuf – cient sampling of relevant degrees of freedom should be assessed separately when possible, and pointed out during the discussion of the results.

Despite large research effort, many of the fundamental questions regarding the physics and the biology of GPCR dimerization are still open. Our understanding of the nature of the physical forces that drive the dimerization remains incomplete, despite its practical importance in designing strategies to disrupt OR dimers [41], and in identifying the structural reasons of homo- and heterodimer selectivity.

Computational methods are promising tools to investigate the structural details of dimerization and oligomerization of mem- brane proteins, although several technical dif culties must be addressed and solved before a complete picture of the dimerization process can be successfully painted.

Solid experimental evidence shows that the characteristics of the lipid environment may play a role in the modulation GPCR function [42] and quaternary structure [43, 44]. First, this modu- lation can be mediated by speci c binding of individual lipids or sterols to crevices on the outer surface of the TM region [45], effectively altering the shape and the physicochemical properties of

3.6 System Setup for Coarse-Grained Dimeric Systems

the protein surface and affecting the dimerization interface details and af nity. Second, membrane–protein interactions can modulate the elastic and rheological properties of the membrane, resulting in association strengths that could depend on membrane thickness [46–48], uidity, and curvature [49]. Heterogeneous lipid com- positions, moreover, introduce a new layer of complexity [50, 51].

Thus, while several implicit effective energy-function models have been proposed, membrane degrees of freedom must be explicitly included in the model to correctly reproduce the proper- ties of the system on scales where the protein–protein distance is of the order of magnitude of a few lipid molecules [52, 53].

Luckily, in the last few years, reliable physical-based coarse- grained (CG) models reproducing the thermodynamics of several lipid phases and the partitioning of amino acids in polar and non- polar phases have been proposed (e.g., the MARTINI force- eld [54]). These models have been used to start investigating, among others, several aspects of membrane protein dimerization and oligomerization. The strategy is based on an approximated 4-1 mapping of heavy atoms to coarse-grained beads, and on a careful parameterization of the nonbonded electrostatic and dispersion terms to reproduce thermodynamics experimental values. Models for lipids, sterols, sugars, and proteins are available [55]. The CG procedure accomplishes a signi cant reduction of the number of particles of the system, thus reducing the number of force compu- tations necessary for the integration of the equations of motion and the computational cost of each of those calculations. Impor- tantly, it also eliminates the fastest degrees of freedom in the system (such as fast bond stretching, small functional groups rotations, etc.) allowing a large increase in the integration time step without appreciable loss of accuracy.

All-atom protein models can be converted to CG representations with scripts (available on the Marrink lab website [56]) implement- ing the MARTINI [54] mapping.

Bonded parameters on the backbone pseudodihedrals are determined by the secondary structure of the input all-atom model and restrict the sampling to such structures. Changes in tertiary structures, albeit possible in principle, likely depend on details not accounted for within the approximations of the CG model, and their accuracy is therefore dif cult to ascertain. For this reason, the desired conformation of the protein is chosen in advance, and the simulation is restricted to sample only small uctuations around this conformation.

A convenient framework to do so is to add a network of elastic constraints to backbone beads pairs within a given distance [57]. The cutoff distance and the strength of the harmonic constraints have to be determined by comparing the RMSD uctuations of the model with those of a corresponding all-atom system (see Note 7).

Computational Biology of Opioid Receptors 25

26 Davide Provasi

3.7 Unbiased Simulations of OR Dimerization

Several methods have been proposed and used to embed CG protein models in CG membranes. While self-assembly of the membrane around the protein [58] is convenient for complex membrane compositions and in cases where the positioning of the protein in the membrane is not known, other strategies are more ef cient in simpler situations. The approach described by [59] involves a series of subsequent compression and equilibration steps of the lipids following an initial expansion of the membrane arti – cially increasing the spacing between the lipid molecules.

Unbiased simulations of GPCR diffusing freely in membrane bilay- ers have been proposed to investigate the dimerization process. Considerations similar to the ones expressed in discussing unbiased ligand binding studies also apply here.

Unbiased simulations can therefore only be effective in a very crowded regime (see Note 8). Furthermore, while dimer dissocia- tion rates at given interfaces are dif cult to obtain experimentally, single-molecule experiments [60] and biased simulations (see below) show that the dissociation is—in some cases—as slow as koff ~ 1.0 s−1, so that we could not expect to see unbinding events in the time scale we can simulate.

Despite these limitations, unbiased strategies represent a very useful tool to generate hypothesis regarding putative dimerization interfaces [43, 61]. Special care should be taken when interpreting the results, making sure that suf cient statistics is accumulated by running replica simulations, and that the in uence of the initial placing of the molecules has been ruled out, or accounted for.

A complete description of the relative orientation of two proteins in a at lipid bilayer requires seven values (Fig. 3): the relative tilt- ing of the two protein axes is described by two polar and azimuthal angles for each protomer (θ1, φ1, θ2, and φ2), while the rotation of each protein around its axis is described by two more angles (α and β). In [61] the values of the six collective variables de ning the relative orientation of the proteins were kept controlled with har- monic potentials [62]. While this assures complete control over the region of the phase space sampled by the simulation, the unbi- asing procedure to extract the nal free-energy needs to account for the restraints imposed using a (six-dimensional) generalization of the common weighted histogram analysis method (WHAM, see below). In at membranes, the axis of GPCR molecules uctuates around the normal of the membrane, so that the average values of θ1, φ1, θ2, and φ2 deviate only slightly from trivial values.

In this situation, the system is described by the distance and two angles, which specify completely the dimeric interface. Given the relative rigidity of the tertiary structure of the TM bundle, a computationally convenient way of specifying the angles α and β is to use the projection on the membrane plane of the angles de ned

3.8 Biased Simulations of OR Dimerization

3.8.1 Collective Variables

Computational Biology of Opioid Receptors 27

3.8.2 Umbrella Sampling Simulations

Fig. 2.3 Collective variables that specify the relative orientation of two protomers in the lipid bilayer. The absolute tilt of the protein principal axis is de ned for each protomer by one polar and one azimuthal angles (θi and φi in dark and light blue, respectively). The description of the rotation of each protein around its principal axis requires two more angles (α and β)

by the center of mass of the two protomers and of one selected helix. Finally, the distance of the centers of mass d completes the de nition of the relative position of the protomers.

The simulation protocol can be made more ef cient by con- straining the absolute position of the pair of proteins in the box. For example, the projection of the center of mass of one protomer on the membrane plane can be xed to the center of the simulation box, and the second one can be allowed to move only along the xy diagonal [63].

Enhanced exploration of the relative distance can be ef ciently achieved using umbrella sampling. Two different choices have been shown to give similar results. In one case (applied with suc- cess to a number of systems including opioids [63–65] and adren- ergic receptors [66]) a relatively large number of simulations with strong harmonic restraints were run (approximately one point every 0.05 nm between 3.00 and 4.90 nm, with a harmonic restraint of 2,400 kcal/mol/nm2). To achieve a proper overlap of the probability distributions, which crucially determines the accu- racy of the resulting free-energy reconstruction, additional points might be inserted where the potential of mean force along the distance is particularly steep.

An alternative strategy was followed in [61]. Windows were spaced of 0.1 nm, and simulated with a much lower harmonic bias (~120 kcal/mol/nm2), with additional windows run with

28 Davide Provasi

3.8.3 Reweighting

240 kcal/mol/nm2 and 1,200 kcal/mol/nm2. In each strategy, the windows were extended to approximately 0.8 μs each.

Different strategies have been experimented for the sampling of the relative orientation of the protomers described by the angles α and β; these can be restrained either with (1) additional har- monic restraints (as described in [61] and [65]), or with (2) square well potentials [63], or (3) their sampling can be enhanced with metadynamics [64, 66].

The rst strategy only allows very limited remodeling of the dimeric interface, and therefore provides a viable method to assess the dimerization free-energy of a speci c interface, de ned by the equilibrium positions α0 and β0 of the harmonic umbrellas (see Note 9). This approach was applied to study the stability of the interfaces inferred in opioid crystal structures in [65], and to quan- tify the stability of putative rhodopsine interfaces [61] identi ed in self-assembly unbiased simulations.

Using weaker harmonic restraints, or a square-well potential, as in [63] in principle allows the system to explore re ned interfaces, reaching a local minimum within the range of orientations allowed by the restraints. It is important to observe that the ef ciency of such sampling is limited by the rotational diffusion constant and the length of the simulations (see Note 10).

To avoid this problem, metadynamics can be applied to the rotational degrees of freedom to accelerate the exploration, providing a converged estimation of the free-energy of different relative orientations of the two protomers. This strategy was applied to adrenergic receptors [64] and to ORs [66].

The biases introduced to enhance the sampling during the simula- tion must be properly accounted for when processing the data and extracting results. In the case of the simulation strategies described above, different approaches must be followed depending on the treatment of the orientation degrees of freedom.

In case (1), the restraints imposed on the system correspond to a three-dimensional umbrella sampling and can therefore be pro- cessed as usual by WHAM [67], Bennett Acceptance Ratio (BAR) [68], or Multistate BAR (MBAR) [69]. Since in this formulation only one umbrella corresponding to the studied interface α and β is applied along each orientation dimension, the unbiasing these in this case trivial, yielding

Pb (r)=Z-1òdadbP(a,b,r)e+U(a-a0)/kBTe+U(b-b0)/kBT i

where U is the usual harmonic umbrella, and Z a normalization constant. The reweighted probability distributions can be used in WHAM codes.

In metadynamics runs, a non-markovian bias is used to enhance the sampling of the relative orientation, so that WHAM or BAR

3.8.4 Binding Af nity

methods cannot be directly applied to remove the bias from the harmonic umbrellas along the distance degree of freedom. Rather, unbiased distance probabilities Pib(r) must be calculated for the distribution of the radial degree of freedom in the presence of the umbrella restraints, but removing the in uence of the angular biases. This can be ef ciently done using reweighting techniques developed to calculate Boltzmann probabilities from metadynam- ics simulations [70]. The usual WHAM or BAR techniques can then be applied to the resulting distributions.

Once a converged reweighted free-energy pro le has been obtained, a formalism paralleling the one introduced for ligand binding can be applied to relate the free-energy as a function of the receptor distance to a binding constant. A standard dimerization free-energy can be calculated only after a reference state (or, equivalently, a scale) has been chosen. Following [71], we use the mole fraction scale, de ning the dimerization constant KX as the ratio of the mole fraction of dimeric and monomeric species in the membrane phase

KX =e-DG°X/RT = nD /nT 2 (nM /nT)

where nD is the number of dimers, nM the number of monomeric proteins, and nT the total number of molecules (lipids and proteins) in the membrane compartment; the expression above also de nes the free-energy ΔG°X on this scale. Assuming diluted samples, the lipid concentration is greater than the protein concentration, so that the number of lipid molecules NL is much bigger than ND + NM and the total number of molecules NT ~ NL. With this assumption, we can express the mole fraction binding af nity KX in terms of the binding af nity KD on the surface concentration scale

K nL ́ nD /A =nL ́K X A(nM/A)2 A D

where A is the surface area of the membrane. As we have seen in the ligand binding case, such binding af nity can be calculated from the calculated PMF, provided that the appropriate correction is included to account for the restraints imposed on the proteins during the simulation:

K = W òdrre-w(r)/kBT D2

Computational Biology of Opioid Receptors 29

(2p) S

where |Ω|=|Δα|×|Δβ| is now the 2D volume in the angle space sampled by the simulation, and the radial integral is calculated over the region of the PMF where the receptors are in contact.

The remarks made above regarding the importance of error

30 Davide Provasi

3.9 Summary and Outlook

estimation are, of course, also valid for dimerization studies. Depending on the method chosen to reconstruct the free-energy w(r), different strategies must be applied to the estimation of its variance (see Note 11) that can then be converted, using general statistical methods, to the error on the binding af nity KD (see Note 12).

Given the recent additions to the crystallographer toolbox, new self-labeling strategies for uorescence spectroscopy, and the ever- increasing computational power available to a growing number of researchers, the outlook for the structural biology of OR looks bright.

For ligand binding, in particular, studies based on unbiased MD techniques, presented here, hold great promise for identifying not only orthosteric binding poses, but, most signi cantly, alterna- tive and dynamically transient binding pockets relevant for alloste- ric modulation of the activation process. Qualitative information about the binding pathways and kinetics can also be obtained with this approach.

Unbinding events, however, cannot be sampled ef ciently in standard MD simulations, and therefore converged quantitative estimates of binding free-energies can be obtained only from either biased simulations or enhanced strategies such as Markov state modeling techniques.

Single-molecule experiments, as well as new crystal structures, will greatly contribute to shedding light on the dimerization process, while the computational techniques described here will hopefully allow bridging the gap between high resolution struc- tural data and single-molecule studies, allowing to compose a com- prehensive picture of what complexes are suf ciently long-lived to contribute to the biology of opioid receptors.

To this end, some technical issues still need to be addressed. In particular, the conformational ensemble of exible loops, which are likely to play a role in some of the proposed interfaces [7], is still inaccurate because of the simple approach used to describe back- bone bonded interactions in the CG strategies presently in use.

1. Interactions at the opioid orthosteric site very often involve charge moieties. Paradigmatically, natural opium alkaloids (e.g., morphine and codeine) contain a weak acidic phenol (pKa2~9.9) and a strongly basic tertiary amine (pKa1~8.0) moiety; these molecules, along with a large class of narcotic analgesics are functionally basic compounds, forming a stable interaction in the binding site with the conserved aspartic acid side chain at position 3.32. For this, special care must be taken

4 Notes

Computational Biology of Opioid Receptors 31

in making sure that the ligand models accurately re ect the electrostatics. Inaccurate partial charges should be identi ed (by a careful analysis of the penalties), and remodeled by following the suggested parameterization strategy or by sup- plementing the fragment subset with additional molecules. Torsion terms with large penalties should also be checked against good ab initio quantum chemistry torsion scans, and modi ed as needed.

2. While crystal structures of constructs mimicking the interac- tions that stabilize the activated state of the receptor (e.g., with agonist ligands, G-protein partners [72] or nanobodies [73]) are well within the reach of available techniques, and will be fundamental to understand the interactions underlying the activation mechanism, there is not yet any direct experimental information about the details of the active states of OR recep- tors. Homology models based on other active GPCRs tem- plates, as well as computational techniques designed to study the activation process using unbiased [74] and biased tech- niques [75–77], can be used to obtain models with binding pockets adapted to agonist binding, but the results must be interpreted with great care.

3. Typical settings [22] comprise up to 10 ligands, corresponding to concentrations as high as cL = 0.05 M. Since for a large frac- tion of the simulation time several ligands are in contact with the membrane, a reduced value of cLt must be used to account for the effective concentration in the solvent phase.

4. Using the same Poisson approximation, the expected number of unbinding events in a simulation of length t is
K(Nt) = koff t
so that typical unbinding rates for opioids—e.g., for MOR [78], koff are in the range ~0.02–2 min-1—put the expected wait timescale for one unbinding event, 1/koff, signi cantly above the simulations length we can afford today even for fast dissociation compounds.

5. This is particularly effective in the case of well-structured loops (such as the EC2 one in opioid receptors), and was applied in the study of the binding of the classical antagonist naloxone in [39] (Fig. 1). More exible loops should be described with collective variable adapted to the natural low-frequency dynam- ics, for example using principal components analysis (PCA) [79] or path-based [80] collective variables.

6. The most appropriate way of doing so depends, of course, on the particular computational method used. A detailed discus- sion of tests to control the convergence of umbrella sampling or metadynamics simulations is beyond the scope of this note,

32 Davide Provasi

and we refer the readers to one of the several review articles describing best practices in the eld (see, for instance [81–83]). Special emphasis must be put in estimating the correlation in the data extracted from the simulations, and accounting for this in calculating the errors. Once the accuracies have been assessed for the different components contributing to the total binding af nity, error propagation can be used to obtain con dence intervals for the nal estimate; in the case of the expression given above involving binding to one intermediate state, we obtain

ædK-1 ö2 ædK-1 ö2 ædDG ö2

ç 31 ÷ =ç 32 ÷ +ç 21 ÷ K -1 K -1 k T
è 31 ø è 32 ø è B ø

When employing complex simulation strategies as the ones described here, in which the binding process is broken up in multiple steps, it is advisable to derive expressions of the nal error in terms of the accuracy of each step, planning the length of each simulation so that the each one contributes similarly to the nal variance, in order to avoid unbalanced scenarios.

7. For the study of GPCRs, where loops are more exible than the transmembrane domain bundle, it is advisable to modulate the strength of the force constant depending on the secondary structure of the receptor residues. Typical values, obtained [63] by comparing a 50 ns long all-atom explicit simulation to several CG runs, are a cutoff dCut=0.9 nm, and elastic con- stants kH = 1,000 kJ/mol/nm2 when both the residues invol- ved are part of helical segments longer than two residues, or kL=250 kJ/mol/nm2 otherwise (i.e., coil, bend, hydrogen bonded turn, or other unde ned structure).

8. We can obtain a very rough idea of the time scales involved by observing that the experimental 2D diffusion constant for ORs in membrane bilayers is of the order of D~0.1 μm2/s [84]. The average square displacement of each protein during the simulation time t is thus expected to be ⟨ds2⟩~Dt, so that to observe dimerization events we need a protein concentration C=NP/A at least of the order of the critical concentration
C~1~1=1

0 2ds 2 2Dt eff 2 ́ 4 ́ D ́ t

where we have accounted for the fact that in CG simulations the effective physical time teff is roughly 4 times the simulation time. For a simulation length of, say, 100 μs, we see that we need a protein concentration larger than C0~1/(10−4 μm2). Assuming simple additive behavior, the area occupied by NP proteins and NL lipids can be easily expressed in terms of the

Computational Biology of Opioid Receptors 33

surface projection of one protein molecule (AP) and the lipid unit area (AL), so that the protein surface concentration in terms of the lipid:protein ratio ρ=nL/nP is

C~1
AP +ALr/2

Approximating AP ~ 7.0 × 10−6 μm2 and AL ~ 1.6 × 10−6 μm2, the constraint that C > C0 entails that ρ < ρ0 = 110.

9. It is useful to estimate the mean torque (i.e., the derivative of the free-energy with respect to the orientation degrees of freedom) exerted on the protomers by the harmonic restraints on α and β, and assess their in uence in maintaining the inter- face. Large torques can signal instabilities of the studied arrangement and provide useful information regarding the closest metastable state. Since the free-energy estimate from each window is given by:

G =-kTlnPb(r,a,b)-w(r,a,b)+F iBi11

where i is the window index, Pib is the biased probability distri- bution sampled by the simulation, wi is the bias, and Fi is a constant resulting from the WHAM.

The mean force on each protomer is obtained by differenti- ating with respect to the corresponding angle. We can roughly estimate the torque in each window at the center of the bias and approximating the probability as a multivariate Gaussian:

Pb = e-(x-m)S-1(x-m)/2 i3

a i ¶a B
h

(2p) detS

where x=(r, α, β), μ is the mean vector, and S the covariance matrix, which can be easily calculated from the collective vari- ables. Thus, the mean torque on the rst protomer is:

t (r)¶Gi =-kTå(x-m)S-1

Errors can be calculated by both block averaging, or by esti- mating errors on both μ and S and propagating.

Nonzero values of the torque re ect the tendency of the system to relax to a different interface, which is contrasted by the harmonic potentials applied to the angles. In other terms, the system could further minimize its free-energy by rearrang- ing the contacts at the dimer interface.

34 Davide Provasi

10. The explored angular range can be estimated assuming a rota- tional diffusion in a at potential, giving the order of magni- tude estimate
da2 ~tDR
For typical values of DR ~ 104 rad2/s [85] and typical simulation length for CG simulations t~1 μs we obtain 〈δα2〉~0.01 rad2, corresponding to a change of ~5° in the orientation. Using wider windows without increasing the simulation length could result in signi cant undersampling of the orientation degrees of freedom.

11. Here we consider the error analyses for the combined use of metadynamics and umbrella sampling, since it represents an unusual simulation strategy. To provide a general framework for error analysis in this scenario, we use the approach described in [86], where the error on the one-dimensional free-energy w(r) is expressed cumulatively in terms of the sample variance of the collective variable average in biased distribution
M Varw(r)=(kDr)2 åVarrˆ

where k is the elastic constant of the umbrella potential, and Δr is the spacing of the umbrella positions. These were assumed to be constant in writing the expression above, but equivalent expressions for the more general case can be obtained trivially. The variance of the average estimator can be expressed in terms of block average [87] when there is no other bias acting on the system. In the case where metadynamics is applied to enhance the angle sampling, block averages of the data does not lead to an unbiased estimator of the average.

An alternative approximated approach is to use instead the following expression

òdrrPi (r,t) òdrP(r,t)

where T is the total length of the simulation, Pi(r, t) is the biased distribution of r, at time t, after the bias on the angles has been removed by reweighting, and the weights gτ are adjusted to account for the convergence of Pi. Empirical analysis of the convergence of the reweighting algorithm [88] shows

T T g t rˆgr= t dt

å i t=0

åò t t=1 t 0

i i=i(r)

Computational Biology of Opioid Receptors 35 that convergence is linear with time, justifying the choice gτ ~ τ.

The variance can now be estimated simply by

T
å g ( rˆ – r ) 2

t i it iT

V a r rˆ = t = 1
åg

12. A useful approximation of the error on the dimerization rate can be obtained by expanding the PMF around the most stable state r0, and approximating it as a harmonic interaction between the two proteins. Indicating with w0 the depth of the PMF and with σr its spread we have that the PMF is w(r)~w0+σr−2 (r−r0)−2 so that the integral can be calculated, and the result- ing dimerization constant becomes:

Ws2
K ~ r e-w0/kBT

D (2p)2
so that the relative error on KD can be expressed in terms of the

uncertainty on the width and the depth of the PMF:

t t =1

ædK ö2 æ2ds ö2 ædw ö2 çD÷=çr÷+ç0÷

KskT èDøèrøèBø

With the protocols described here, the rst term contributes only marginally to the overall error, which is dominated by the relative error on the depth of the PMF.

The author wishes to thank Jennifer Johnston and Sebastian Schneider for several stimulating discussions about GPCR compu- tational biology and for the critical reading of the manuscript.

Acknowledgments

References

1. Manglik A, Kruse AC, Kobilka TS et al (2012) Crystal structure of the μ-opioid receptor bound to a morphinan antagonist. Nature 485:321–326

2. Wu H, Wacker D, Mileni M et al (2012) Structure of the human κ-opioid receptor in complex with JDTic. Nature 485:327–332

3. Thompson AA, Liu W, Chun E et al (2012) Structure of the nociceptin/orphanin FQ

receptor in complex with a peptide mimetic.

Nature 485:395–399

4. Granier S, Manglik A, Kruse AC et al (2012)
Structure of the δ-opioid receptor bound to
naltrindole. Nature 485:400–404

5. Fenalti G, Giguere PM, Katritch V et al (2014)
Molecular control of δ-opioid receptor signal- ling. Nature 506:191–196

36 Davide Provasi

6. Filizola M, Devi LA (2012) Structural biology: how opioid drugs bind to receptors. Nature 485:314–317

7. Ferré S, Casadó V, Devi LA et al (2014) G protein-coupled receptor oligomerization revisited: functional and pharmacological per- spectives. Pharmacol Rev 66:413–434

8. Landau EM, Rosenbusch JP (1996) Lipidic cubic phases: a novel concept for the crystalli- zation of membrane proteins. Proc Natl Acad Sci U S A 93:14532–14535

9. Cherezov V (2011) Lipidic cubic phase tech- nologies for membrane protein structural stud- ies. Curr Opin Struct Biol 21:559–566

10. Zou Y, Weis WI, Kobilka BK (2012) N-terminal T4 lysozyme fusion facilitates crystallization of a G protein coupled receptor. PLoS One 7: e46039

11. Chun E, Thompson AA, Liu W et al (2012) Fusion partner toolchest for the stabilization and crystallization of G protein-coupled recep- tors. Structure 20:967–976

12. Bonner G, Meng F, Akil H (2000) Selectivity of mu-opioid receptor determined by interfa- cial residues near third extracellular loop. Eur J Pharmacol 403:37–44

13. Wu B, Chien EY, Mol CD et al (2010) Structures of the CXCR4 chemokine GPCR with small-molecule and cyclic peptide antago- nists. Science 330:1066–10671

14. Hu J, Thor D, Zhou Y et al (2012) Structural aspects of M3 muscarinic acetylcholine receptor dimer formation and activation. FASEB J 26: 604–616

15. Huang J, Chen S, Zhang JJ et al (2013) Crystal structure of oligomeric beta1-adrenergic G protein-coupled receptors in ligand-free basal state. Nat Struct Mol Biol 20:419–425

16. Hu J, Hu K, Liu T et al (2013) Novel structural and functional insights into M3 muscarinic receptor dimer/oligomer formation. J Biol Chem 288:34777–34790

17.Dastmalchi S, Hamzeh-Mivehroud M, Ghafourian T et al (2008) Molecular modeling of histamine H3 receptor and QSAR studies on arylbenzofuran derived H3 antagonists. J Mol Graph Model 26:834–844

18. Jorgensen WL, Maxwell DS, Tirado-Rives J (1996) Development and testing of the OPLS all-atom force eld on conformational energet- ics and properties of organic liquids. J Am Chem Soc 118:11225–11236

19. Wang J, Wolf RM, Caldwell JW et al (2004) Development and testing of a general amber force eld. J Comput Chem 25:1157–1174

20. Vanommeslaeghe K, Hatcher E, Acharya C et al (2010) CHARMM general force eld:

a force eld for drug-like molecules compatible with the CHARMM all-atom additive biologi- cal force elds. J Comput Chem 31:671–690

21. Shaw DE, Dror RO, Salmon JK, et al (2009) Millisecond-scale molecular dynamics simulation on Anton. In: Proceedings of the conference on high performance computing, networking, stor- age, and analysis. IEEE, Piscataway, pp. 1–11

22. Dror RO, Pan AC, Arlow DH et al (2011) Pathway and mechanism of drug binding to G-protein-coupled receptors. Proc Natl Acad Sci U S A 108:13118–13123

23. Shan Y, Kim ET, Eastwood MP et al (2011) How does a drug molecule nd its target bind- ing site? J Am Chem Soc 133:9181–9183

24. Dror RO, Green HF, Valant C et al (2013) Structural basis for modulation of a G-protein- coupled receptor by allosteric drugs. Nature 503:295–299

25. Noe F, Fischer S (2008) Transition networks for modeling the kinetics of conformational change in macromolecules. Curr Opin Struct Biol 18:154–162

26. Pande VS, Beauchamp K, Bowman GR (2010) Everything you wanted to know about Markov State Models but were afraid to ask. Methods 52:99–105

27. Silva D-A, Bowman GR, Sosa-Peinado A et al (2011) A role for both conformational selec- tion and induced t in ligand binding by the LAO protein. PLoS Comput Biol 7:e1002054

28. Buch I, Giorgino T, De Fabritiis G (2011) Complete reconstruction of an enzyme-inhibi- tor binding process by molecular dynamics simulations. Proc Natl Acad Sci U S A 108: 10184–10189

29. Bisignano P, Doerr S, Harvey MJ et al (2014) Kinetic characterization of fragment binding in AmpC β-lactamase by high-throughput molecular simulations. J Chem Inf Model 54: 362–366

30. Shirts MR (2012) Best practices in free energy calculations for drug design. Methods Mol Biol 819:425–467

31. Ytreberg F (2009) Absolute FKBP binding af nities obtained via nonequilibrium unbind- ing simulations. J Chem Phys 130:164906

32. Patel JS, Branduardi D, Masetti M et al (2011) Insights into ligand–protein binding from local mechanical response. J Chem Theory Comput 7:3368–3378

33. Woo H-J, Roux B (2005) Calculation of abso- lute protein-ligand binding free energy from computer simulation. Proc Natl Acad Sci U S A 102:6825–6830

34. Lee MS, Olson MA (2006) Calculation of absolute protein-ligand binding af nity using

path and endpoint approaches. Biophys J 90:

864–877

35. Gervasio FL, Laio A, Parrinello M (2005)
Flexible docking in solution using metadynam-
ics. J Am Chem Soc 127:2600–2607

36. Limongelli V, Marinelli L, Cosconati S et al (2012) Sampling protein motion and solvent effect during ligand binding. Proc Natl Acad
Sci U S A 109:1467–1472

37. Soederhjelm P, Tribello GA, Parrinello M
(2012) Locating binding poses in protein-ligand systems using reconnaissance metadynamics. Proc Natl Acad Sci U S A 109:5170–5175

38. Saladino G, Gauthier L, Bianciotto M et al (2012) Assessing the performance of metady- namics and path variables in predicting the binding free energies of p38 inhibitors. J Chem Theory Comput 8:1165–1170

39. Provasi D, Bortolato A, Filizola M (2009) Exploring molecular mechanisms of ligand rec- ognition by opioid receptors with metadynam- ics. Biochemistry 48:10020–10029

40. Limongelli V, Bonomi M, Parrinello M (2013) Funnel metadynamics as accurate binding free- energy method. Proc Natl Acad Sci U S A 110: 6358–6363

41. O’Dowd BF, Ji X, O’Dowd PB (2012) Disruption of the mu-delta opioid receptor heteromer. Biochem Biophys Res Commun 422:556–560

42. Escribá PV, Wedegaertner PB, Goñi FM et al (2007) Lipid-protein interactions in GPCR- associated signaling. Biochim Biophys Acta 1768:836–852

43. Periole X, Huber T, Marrink SJ et al (2007) G protein-coupled receptors self-assemble in dynamics simulations of model bilayers. J Am Chem Soc 129:10126–10132

44. Goddard AD, Dijkman PM, Adamson RJ et al (2013) Lipid-dependent GPCR dimerization. Methods Cell Biol 117:341–357

45. Zheng H, Pearsall EA, Hurst DP et al (2012) Palmitoylation and membrane cholesterol sta- bilize μ-opioid receptor homodimerization and G protein coupling. BMC Cell Biol 13:6

46. Kusumi A, Hyde JS (1982) Spin-label satura- tion-transfer electron spin resonance detection of transient association of rhodopsin in reconsti- tuted membranes. Biochemistry 21:5978–5983

47. Ryba NJP, Marsh D (1992) Protein rotational diffusion and lipid/protein interactions in recombinants of bovine rhodopsin with satu- rated diacylphosphatidylcholines of different chain lengths studied by conventional and sat- uration-transfer electron spin resonance. Bio- chemistry 31:7511–7518

48. Pearson LT, Chan SI, Lewis BA et al (1983) Pair distribution functions of bacteriorhodop- sin and rhodopsin in model bilayers. Biophys J 43:167–174

49. Botelho AV, Huber T, Sakmar TP (2006) Curvature and hydrophobic forces drive oligo- merization and modulate activity of rhodopsin in membranes. Biophys J 91:4464–4477

50. Barnett-Norris J, Lynch D, Reggio PH (2005) Lipids, lipid rafts and caveolae: their impor- tance for GPCR signaling and their centrality to the endocannabinoid system. Life Sci 77: 1625–1639

51. Fallahi-Sichani M, Linderman JJ (2009) Lipid raft-mediated regulation of G-protein coupled receptor signaling by ligands which in uence receptor dimerization: a computational study. PLoS One 4:e6604

52. Mondal S, Johnston JM, Wang H et al (2013) Membrane driven spatial organization of GPCRs. Sci Rep 3:2909

53. Mondal S, Khelashvili G, Johner N et al (2014) How the dynamic properties and functional mechanisms of GPCRs are modulated by their coupling to the membrane environment. Adv Exp Med Biol 796:55–74

54. Marrink SJ, Tieleman DP (2013) Perspective on the Martini model. Chem Soc Rev 42: 6801–6822

55. de Jong DH, Singh G, Bennett WFD et al (2013) Improved parameters for the martini coarse-grained protein force eld. J Chem Theory Comp 9:687–697

56. Coarse grain force eld for biomolecular simula- tions. Tools: proteins and bilayers. 18 Oct. 2013. http://md.chem.rug.nl/cgmartini/index. php/tools2/proteins-and-bilayers. Accessed 20 Mar 2014

57. Periole X, Cavalli M, Marrink SJ et al (2009) Combining an elastic network with a coarse- grained molecular force eld: Structure, dynam- ics and intermolecular recognition. J Chem Theory Comput 5:2531–2543

58. Sansom MS, Scott KA, Bond PJ (2008) Coarse- grained simulation: a high-throughput compu- tational approach to membrane proteins. Biochem Soc Trans 36:27–32

59. Kandt C, Ash WL, Tieleman DP (2007) Setting up and running molecular dynamics simulations of membrane proteins. Methods 41:475–488

60. Hern JA, Baig AH, Mashanov GI et al (2010) Formation and dissociation of M1 muscarinic receptor dimers seen by total internal re ection uorescence imaging of single molecules. Proc Natl Acad Sci U S A 107:2693–2698

Computational Biology of Opioid Receptors 37

38 Davide Provasi

61. Periole X, Knepp AM, Sakmar TP et al (2012) Structural determinants of the supramolecular organization of G protein-coupled receptors in bilayers. J Am Chem Soc 134:10959–10965

62. Boresch S, Tettinger F, Leitgeb M et al (2003) Absolute binding free energies: a quantitative approach for their calculation. J Phys Chem B 107:9535

63. Provasi D, Johnston JM, Filizola M (2010) Lessons from free energy simulations of delta- opioid receptor homodimers involving the fourth transmembrane helix. Biochemistry 49: 6771–6776

64. Johnston JM, Aburi M, Provasi D et al (2011) Making structural sense of dimerization inter- faces of delta opioid receptor homodimers. Biochemistry 50:1682–1690

65. Johnston JM, Filizola M (2014) Differential stability of the crystallographic interfaces of mu- and kappa-opioid receptors. PLoS One 9:e90694

66. Johnston JM, Wang H, Provasi D et al (2012) Assessing the relative stability of dimer inter- faces in G protein-coupled receptors. PLoS Comput Biol 8:e1002649

67. Kumar S, Bouzida D, Swendsen RH et al (1992) The weighted histogram analysis method for free-energy calculations on biomolecules. I The method. J Comput Chem 13:1011–1021

68. Bennett CH (1976) Ef cient estimation of free energy differences from Monte Carlo data. J Comput Phys 22:245–268

69. Shirts MR, Chodera JD (2008) Statistically optimal analysis of samples from multiple equi- librium states. J Chem Phys 129:129105

70. Bonomi M, Barducci A, Parrinello M (2009) Reconstructing the equilibrium Boltzmann distribution from well-tempered metadynam- ics. J Comput Chem 30:1615–1621

71. Fleming KG (2002) Standardizing the free energy change of transmembrane helix-helix interactions. J Mol Biol 323:563–571

72. Rasmussen SG, DeVree BT, Zou Y et al (2011) Crystal structure of the β2 adrenergic receptor- Gs protein complex. Nature 477:549–555

73. Rasmussen SG, Choi HJ, Fung JJ et al (2011) Structure of a nanobody-stabilized active state of the β2 adrenoceptor. Nature 469:175–180

74. Nygaard R, Zou Y, Dror RO et al (2013) The dynamic process of β2-adrenergic receptor acti- vation. Cell 152:532–542

75. Provasi D, Artacho MC, Negri A et al (2011) Ligand-induced modulation of the free-energy landscape of G protein-coupled receptors explored by adaptive biasing techniques. PLoS Comput Biol 7:e1002193

76. Miao Y, Nichols SE, Gasper PM et al (2013) Activation and dynamic network of the M2 muscarinic receptor. Proc Natl Acad Sci U S A 110:10982–10987

77. Tikhonova IG, Selvam B, Ivetac A et al (2013) Simulations of biased agonists in the β2 adren- ergic receptor with accelerated molecular dynamics. Biochemistry 52:5593–5603

78. Cassel JA, Daubert JD, DeHaven RN et al (2005) [3H]Alvimopan binding to the μ opi- oid receptor: comparative binding kinetics of opioid antagonists. Eur J Pharmacol 520: 29–36

79. Sutto L, D’Abramo M, Gervasio FL (2009) Comparing the ef ciency of biased and unbi- ased molecular dynamics in reconstructing the free energy landscape of met-enkephalin. J Chem Theory Comp 6:3640–3646

80. Branduardi D, Gervasio FL, Parrinello M (2007) From A to B in free energy space. J Chem Phys 126:054103

81. Shirts MR, Mobley DL (2013) An introduc- tion to best practices in free energy calculations. In: Monticelli L, Salonen E (eds) Biomolecular simulations: methods and protocols. Humana Press, New York, pp 271–311

82. Raiteri P, Laio A, Gervasio FL et al (2006) Ef cient reconstruction of complex free energy landscapes by multiple walkers metadynamics. J Phys Chem B 110:3533–3539

83. Crespo Y, Marinelli F, Pietrucci F et al (2010) Metadynamics convergence law in a multidi- mensional system. Phys Rev E Stat Nonlin Soft Matter Phys 81:055701

84. Saulière-Nzeh NA, Millot C, Corbani M et al (2010) Agonist-selective dynamic compart- mentalization of human Mu opioid receptor as revealed by resolutive FRAP analysis. J Biol Chem 285:14514–14520

85. Peters R, Cherry RJ (1982) Lateral and rota- tional diffusion of bacteriorhodopsin in lipid bilayers: experimental test of the Saffman- Delbrück equations. Proc Natl Acad Sci U S A 79:4317–4321

86. Zhu F, Hummer G (2012) Convergence and error estimation in free energy calculations using the weighted histogram analysis method. J Comput Chem 33:453–465

87. Flyvbjerg H, Petersen HG (1989) Error esti- mates on averages of correlated data. J Chem Phys 91:461–466

88. Berteotti A, Barducci A, Parrinello M (2011) Effect of urea on the β-hairpin conformational ensemble and protein denaturation mecha- nism. J Am Chem Soc 133:17200–17206

Chapter 3

Analysis of Epigenetic Mechanisms Regulating Opioid Receptor Gene Transcription

Cheol Kyu Hwang, Yadav Wagley, Ping-Yee Law, Li-Na Wei, and Horace H. Loh

Abstract

Opioid drugs are generally used for moderate and severe pain reductions which act through opioid receptors. Studies on transcriptional regulation of opioid receptors are still invaluable because not only transcription is the first step to produce protein products in cells, but the receptor transcription levels also affect the pain reduction by opioids, as observed in studies of heterozygous opioid receptor knockout mice.

There are growing evidences that epigenetic regulation has played significant roles in transcriptional regulation of genes, including opioid receptors. In general, epigenetic mechanisms include three main regulatory factors: DNA methylation, chromatin modification, and noncoding RNAs (such as microRNA). From previous studies of ours and others on opioid receptors, those epigenetic factors were clearly involved in regulating opioid receptor expression in vivo and in vitro. In this chapter, among those three techniques we describe more details of DNA methylation methods because of emerging concepts of DNA methyla- tion with the recent discovery of 5-hydroxymethylcytosine converting enzyme, TET1. Another analytical method of the epigenetic factors, chromatin modification, will be described briefly and information of analyzing noncoding RNAs is briefly mentioned in Subheading 1.

Key words ChIP, Chromatin, CpG, DNA methylation, Epigenetic regulation, Histone, Opioid receptor, Promoter, TET1, Transcription, 5-Hydroxymethylcytosine

1 Introduction

DNA methylation is an epigenetic event that has been frequently associated with gene silencing, and refers to the covalent addition of a methyl group, catalyzed by DNA methyltransferase (DNMT), to the 5-carbon of cytosine (5-methylcytosine, 5-mC) in a CpG dinucleotide which is preferentially located at the promoter region of about 60 % of human genes. The methyl CpG recruits methyl-CpG-binding proteins [e.g., MeCP2, methyl CpG-binding domain (MBD) 1, MBD2, and MBD3], and leads to the binding of transcriptional repressor-forming complex which results in tar- get gene silencing [1, 2]. DNA methylation analysis can be either

Santi M. Spampinato (ed.), Opioid Receptors: Methods and Protocols, Methods in Molecular Biology, vol. 1230, DOI 10.1007/978-1-4939-1708-2_3, © Springer Science+Business Media New York 2015

40 Cheol Kyu Hwang et al.

gene-specific or global genomic analyses. Global methylation anal- ysis measures the overall level of methyl cytosines in genome, while analysis of gene-specific methylation measures the level of methyl cytosines in an individual gene. Since this chapter limits epigen- etic studies of opioid receptors, we provide here the methods of gene-specific methylation analysis as obtained from our epigenetic analyses of mu opioid receptor (MOR), which primarily mediates pharmacological effects of the opioid drugs commonly used in the clinics, such as morphine, hydrocodone, and fentanyl. Most of the early studies used the methods which include methylation- sensitive restriction enzymes (e.g., DpnII, HpaII, and MspI) for digesting DNA differently, depending on methylation status of the DNAs, followed by PCR amplification. Typical examples include bisulfite conversion reaction based methods [methylation-specific PCR (MSP)] and bisulfite genomic sequencing PCR. Additionally, in order to identify a new paradigm of DNA methylation mecha- nism based on recent finding of the enzyme ten-eleven transloca- tion 1 (TET1; one of the three enzymes of the TET family) [3] that oxidizes 5-mC to 5-hydroxymethylcytosine (5-hmC), global analytic methods for 5-hmC detection using 5-hmC-specific anti- body, such as ELISA and dot blot, as well as gene-specific analy- sis using hydroxymethylation-sensitive restriction enzymes, will be described. There are several commercial kits available for these DNA methylation/hydroxymethylation studies. It is the research- ers’ choices which products they would like to use but we have tried some kits from Zymo Research, such as DNA methylation, 5-hmC DNA enrichment, and 5-hmC detection kits, and found that they performed well in our laboratory.

DNA methylation is tightly correlated with chromatin modi- fication, especially in promoter regions of genes. When cytosine residues of the CpG in promoter regions are hypermethylated, repressive transcription factors are recruited that causes the pro- moters to be inaccessible to the transcription factors by compact- ing chromatin structure, leading to gene silencing. On the other hand, when the CpGs are hypomethylated and the regions become accessible to active transcription factors with the respec- tive chromatin modification, it leads to active gene expression. The chromatin modification consists of dynamic and reversible posttranslational modifications of the residues at N-terminal tails of histones (mostly on histones 3 and 4) that are mediated by sets of enzymatic complexes that site-specifically attach or remove the corresponding chemical groups [2, 4, 5]. The his- tone modifications include acetylation, methylation, phosphory- lation, ubiquitination, SUMOylation, and ADP-ribosylation, and play a key role in important cellular processes such as DNA repair, DNA replication, alternative splicing, and chromosome conden- sation. Numerous antibodies are commercially available which either detect the specific modifications on histones or detect the

2 Materials

2.1 ELISA Components

histone modifying enzymes. Among several methods for the analysis of chromatin, chromatin-immunoprecipitation (ChIP) assay is an easy and convenient method to determine changes of chromatin structure despite its intrinsic nature of undesired high background that researcher must find ways to avoid [see Notes for ChIP assay]. There are well-known histone antibodies frequently being used in analysis of chromatin structure, for example, H3K4me3 (histone H3 trimethyl lys4) serves as a hallmark of gene activation, and H3K9me3 (histone H3 trimethyl lys9) and H3K27me3 (histone H3 trimethyl lys27) serve as hallmarks of gene repression. For genome-wide studies, ChIP on Chip assays (microarray) and ChIP-Seq (next generation sequencing) methods can be used. Furthermore, there are additional methods for studying chromatin structure, such as positioning of nucleosomes on DNA by nucleo- some foot-printing using CpG methyltransferases and DNase I [6, 7], detecting open regions of the genome or promoter region by their sensitivity to nucleases, such as DNAse I or micrococcal nuclease [8], and high resolution mapping of protein factors bind- ing by combining ChIP with exonuclease I (ChIP-exo) [9]. Since the ChIP assay is an initial step of analyzing chromatin modifica- tion on most studies, it will be described in this chapter.

For the third epigenetic factor (noncoding RNAs), most eukaryotes contain small noncoding RNAs, especially microRNAs (miRNA). miRNAs are typically 21–24 nucleotides long and regu- late gene expression posttranscriptionally and with sequence spe- cifically. miRNAs affect both the stability and translation of mRNAs. miRNAs are predicted to target over 30 % of protein- coding genes [10], and are involved in almost all cellular functions and regulate gene expression mainly by binding to the 3′-UTRs of targeted mRNAs. Studies on miRNAs’ relationship with opioid receptors function are increasing [11–14], and it has been reported that several miRNAs, including miR-23b, let-7, miR-134, and miR-339-3p, regulate MOR level [12–15]. From our studies of two miRNAs, miR-23b and miR-339-3p, we have established some methods, although they are somewhat not unique compared to conventional known methods. Those methods are construction of miRNA expression plasmid, analysis of miRNA expression by real-time qRT-PCR, immunocytochemistry and image data analy- sis, and flow cytometry (refer to refs. 11, 13).

1. TMB (3,3′,5,5-tetramethylbenzidine, Sigma T5525) sub- strate solution (0.1 mg/ml): dissolve 1 mg of TMB in 1 ml of DMSO completely and add 9 ml of 0.05 M phosphate-citrate buffer (pH 5.0). Add 2 μl of 30 % hydrogen peroxide before use (see Note 1).

Epigenetic Methods for Opioid Receptor Regulation 41

42 Cheol Kyu Hwang et al.

2.2 ChIP Components

2. 0.05 M phosphate-citrate buffer (pH 5.0): add 25.7 ml of 0.2 M dibasic sodium phosphate [dissolve 1.42 g (FW 141.96) in 50 ml of distilled water] and 24.3 ml of 0.1 M citric acid [dissolve 1.47 g (FW 294.1) in 50 ml of distilled water], and adjust pH to 5.0.

3. Protamine sulfate stock solution (100×, 0.3 %): Weigh 30 mg of protamine sulfate (salt form from salmon, Sigma P4020) and dissolve in 10 ml of distilled water by stirring for 1 h. Working concentration is 0.003 %.

1. ChIP cell suspension buffer: 25 mM HEPES [pH 7.8], 1.5 mM MgCl2, 10 mM KCl, 0.1 % NP-40, 1 mM DTT, 0.5 mM PMSF, 1× protease inhibitor cocktail (Roche).

2. ChIP lysis buffer: 50 mM Hepes [pH 7.9], 140 mM NaCl, 1 mM EDTA, 1 % Triton X-100, 0.1 % sodium deoxycholate, 0.1 % SDS, 0.5 mM PMSF, 1× protease inhibitor cocktail.

3. Wash buffer A: ChIP lysis buffer containing 500 mM NaCl.

4. Wash buffer B: 20 mM Tris–HCl [pH 8.0], 1 mM EDTA, 250 mM LiCl, 0.5 % NP-40, 0.5 % sodium deoxycholate, 0.5 mM PMSF, 1× protease inhibitor cocktail. TE buffer: 10 mM Tris–HCl [pH 8.0], 1 mM EDTA.

5. Elution buffer: 50 mM Tris–HCl [pH 8.0], 1 mM EDTA, 1 % SDS, 50 mM NaHCO3.

1. For coating microwell plates with protamine sulfate (see Note 2), add 50 μl per well of 0.003 % protamine sulfate solution to 96-well plates (flat-bottom, MaxiSorp treated clear polystyrene from Nunc 442404).

2. Incubate the plates at 37 °C overnight (see Note 3). This will coat plates with protamine sulfate after complete drying.

Wash the plates three times with 150 μl of distilled water per well. These plates can be stored for several months pro- tected from light.

3. For coating DNA samples (see Note 4) to the microwell plates precoated with the protamine sulfate, prepare sample DNA solutions in PBS at the concentration of 0.2 μg per ml (10 ng per well).

4. Denature DNA by heating DNA solutions in a hot plate at 100 °C for 5 min and chill rapidly in an ice bath for 5 min.

5. Add 50 μl of each denatured DNA solution per well to protamine sulfate precoated 96-well plates (use at least two wells for each sample) and dry completely overnight in an incubator at 37 °C.

3 Methods

3.1 ELISA Immunoassay
for 5-mC and 5-hmC Detections

Epigenetic Methods for Opioid Receptor Regulation 43

6. Wash the DNA-coated plates three times with 150 μl of TTBS
(0.05 % Tween-20 in TBS) per well (see Note 5).

7. To block nonspecific binding, add 100 μl of 3 % bovine serum albumin (BSA) in PBS to each well and incubate for 30 min at 37 °C.

8. Wash the plates two times with 100 μl of TTBS per well.

9. Add 100 μl of first antibody diluted with 1 % BSA in PBS to each well and incubate for 30 min at 37 °C. We used 1:5,000 dilutions for 5-hmC antibody (Zymo Research, A4001) (see Note 6).

10. Wash the plates three times with 150 μl of TTBS per well.

11. Add 100 μl of 1:5,000 second antibody-HRP (horseradish peroxidase) diluted with 1 % BSA in PBS to each well and incu- bate for 30 min at 37 °C (goat anti-rabbit-HRP, 1:5,000 dilu- tion, Santa Cruz sc-2054).

12. Wash the plates three times with 150 μl of TTBS per well.

13. For colorimetric development, add 100 μl of freshly prepared TMB substrate solution to each well. Develop at room tem- perature for 3 min (see Note 7).

14. Stop the reaction by 25 μl of 2 M sulfuric acid (H2SO4) and read absorbance at 450 nm in an ELISA plate reader. See Fig. 1 for the result of ELISA immunoassay.

0.4 0.3 0.2 0.1 0.0

ELISA assay

5-hmC level

5-mC level

100 95

6 4 2 0

Fig. 1 ELISA immunoassay with 5-mC and 5-hmC antibodies. See the informa- tion of PCR control DNAs containing the normal C, 5-mC DNA, and 5-hmC DNA in Note 4. STM indicates genomic DNA from striatum (STM) of mouse brain. Either rabbit 5-hmC (A4001, Zymo Research) or mouse monoclonal 5-mC antibodies (A3001, Zymo Research) were used to detect 5-hmC or 5-mC levels, respec- tively, with both 1:5,000 dilutions. The percentage of each modified DNA level is relative to each respective control DNA

% 5-hmC amount

% 5-mC amount

STM normal DNA

5-mC DNA 5-hmC DNA

STM normal DNA

5-mC DNA 5-hmC DNA

44 Cheol Kyu Hwang et al.

3.2 Dot Blot Method for 5-mC and 5-hmC Detections

In general, a dot blot is a simple and quick assay to determine if antibodies and detection system are effective. To quantify global levels of 5-mC and 5-hmC, we employed the dot blot assay in addition to the ELISA assay described above.

1. Prepare control DNAs (see Note 4) and experimental DNA samples in 5 μl of 0.1 M NaOH.

2. Boil for 5 min to denature DNAs. Cool down on ice for 2 min and neutralize with 0.5 μl of 6 M ammonium acetate (pH 7.0).

3. Spot on membrane (Hybond-N+, Amersham) and let the membrane dry at room temperature for 10 min.

4. Cross-link DNA by UV (1× auto cross-link, UV Stratalinker 1800, Stratagene).

5. Block nonspecific sites by soaking in blocking solution [10 % dry milk, 1 % BSA in PBST (0.1 % Tween-20 in PBS)] for 1 h at room temperature. Pour off the blocking buffer, but keep membrane wet at all times for the remainder of the procedure.

6. Incubate with first antibody (1:1,000 dilution in blocking solution, see Fig. 2) for 1 h in at room temperature.

7. Wash three times with PBST (5 min each wash).

8. Incubate with second antibody conjugated with HRP (for optimum dilution, follow the manufacturer’s recommenda- tion) for 30–60 min maximum at room temperature.

9. Wash three times with PBST (5 min each wash), then once with PBS (5 min).

10. Develop with TMB substrate solution for 2–3 min and stop the reaction in distilled water. Scan the membrane using a PhosphorImager (Storm 860; Molecular Dynamics). See our result in Fig. 2.
IB: 5-hmc Ab IB: 5-mc Ab
40 ng 250 ng

Fig. 2 Dot blot analysis of DNA samples with 5-mC and 5-hmC antibodies. Control DNAs (normal, 5-mC, and 5-hmC DNAs) were described in Note 4. The membrane was incubated with either rabbit 5-hmC or mouse monoclonal 5-mC antibodies (Zymo Research) with both 1:1,000 dilutions. P19 and AP4d indicate genomic DNAs from embryonal P19 and neuronal differentiated P19 cells (AP4d), respectively [2]. STM indicates genomic DNA from striatum of mouse brain

normal DNA 5-hmC DNA

5-mC DNA P19

AP4d STM

3.3 Methods
of 5-hmC Detection Using Hydroxymethylation- Sensitive Restriction Enzymes

Gene-specific analysis of 5-hmC can be done by using hydroxymethylation-sensitive restriction enzymes, such as PvuRts1I. Since the PvuRts1I restriction enzyme is specific to 5-hmC DNA (recognizes the sequence hmCN11-12/N9-10G), and will not digest 5-mC residues or unmethylated DNA, this enzyme is very useful to detect 5-hmC levels in the promoter region of a specific gene.

MspJI restriction enzyme recognizes 5-hmC residues as well as 5-mC residues. Using the digested DNA, quantitative PCR-based methods can be used to determine relative abundance of C, 5-mC, and 5-hmC residues at a promoter sequence. Results from these two restriction enzymatic analyses would provide comparative lev- els of both modified cytosine residues in a gene promoter. In case low levels of 5-hmC residues on a gene promoter exist, an affinity- based enrichment method can be employed. β-Glucosyltransferase glycosylates 5-hmC to β-glucosyl-5-hmC (5-ghmC) which can be selectively precipitated by J-binding protein (JBP) coupled to mag- netic beads. This J-binding protein pull down assay can enrich 5-hmC DNA from genomic DNA and can be followed by down- stream analyses, such as qPCR, microarray, or sequencing. The PvuRts1I restriction enzyme also digests 5-ghmC DNA so that PvuRts1I can be used with the above methods. In this section, a 5-hmC detection method using hydroxymethylation-sensitive restriction enzyme (PvuRts1I) is described (Fig. 3).

Epigenetic Methods for Opioid Receptor Regulation 45

Analysis of 5-hmC level by PvuRts1 I restriction enzyme

PvuRts1 I : hmCN11-12/N9-10G, recognizes most CpG sites

5-hmC

5-mC

PvuRts1 I

PCR reaction

up or down

genomic DNA

PCR primers

agarose DNA gel

mRNA

Fig. 3 Experimental scheme of 5-hmC detection in a gene promoter region using 5-hmC-sensitive-restriction enzyme PvuRts1I

Cheol Kyu Hwang et al.

3.4
Detection Using Methylation-Sensitive Restriction Enzymes

1. In a total reaction volume of 30 μl, add 100 ng of isolated genomic DNA or enriched 5-ghmC DNA, 1× PvuRts1I reac- tion buffer, 2 U of PvuRts1I, 3 μl of 10 mM DTT.

2. Incubate at 22 °C for 1 h and stop the reaction at 65 °C for 20 min.

3. Purify the digested DNA by using a DNA clean-up kit (DNA Clean & Concentrator from Zymo Research). Elute DNA from a column in 20 μl of elution buffer.

4. Use 2 μl of eluted DNA for PCR reaction.

To detect 5-mC levels in a genome-wide or gene-specific way, both

requires bisulfite reaction which converts cytosine residues to ura- cil, but leaves 5-methylcytosine residues unaffected. In DNA sequencing or methylation-specific PCR, the uracil can be replaced by thymine which reflects a demethylated form of cytosine. The bisulfite reaction-based methods such as methylation-specific PCR and bisulfite genomic sequencing PCR have become very popular for DNA methylation analysis [2, 16]. The bisulfite conversion- based PCR method is described below briefly.

1. Use 1 μg of purified genomic DNA to linearize with the restric- tion enzyme EcoRV. Bisulfite treatment of DNA is performed using an EZ DNA Methylation-Gold KitTM (Zymo Research) according to the manufacturer’s recommendations.

2. Amplify the resulting bisulfite-modified DNA by PCR using methylation-specific primers.

3. Clone the PCR products into a pCRII-TOPO vector (Invitrogen) and isolate clones containing a correctly sized insert.

4. Analyze the insert by DNA sequencing.

5. The same PCR products can be used for methylation-sensitive restriction enzyme analyses.

6. Find appropriate methylation-sensitive restriction enzymes based on DNA sequence information for the gene of interest (e.g., BstBI, ClaI, Hpy188I, and HpyCH4IV for the MOR gene).

7. Incubate the modified DNA samples for 1 h at the recom- mended temperature. (Input of each PCR product should be adjusted for equivalent amounts in the restriction enzyme reactions.)

8. Analyze the resulting DNA in a 2 % agarose gel and quantify by molecular imaging software.

Chromatin immunoprecipitation (ChIP) is one of most valuable methods for detecting the presence of proteins, including chromatin modifications, recruited to a specific genomic locus. ChIP can be used to examine transcription factor occupancy, co-factor recruitment,

Methods of 5-mC

3.5 Chromatin Immunoprecipitation

Epigenetic Methods for Opioid Receptor Regulation 47

and histone modifications status in vivo with posttranslational modifications at specific genomic locations. After using formalde- hyde to cross-link chromatin and associated proteins, the cross- linked chromatin is either sonicated or cleaved with restriction enzymes to generate smaller DNA fragments, followed by immu- noprecipitation with the desired antibodies. The resulting “pull- down” is then analyzed by PCR [or by real-time quantitative PCR (qPCR)] or array analysis to assess the relative amount of DNA corresponding to specific genomic loci. This ChIP technique has been frequently used to study the transcriptional regulation of opi- oid receptor genes [2, 17, 18]. Although the following protocol is based on the use of mouse cell lines (i.e., P19 cells), the method can be applied to mammalian tissues, such as mouse brain regions. The procedure is designed for cells in 100-mm dishes, so the vol- umes listed below should be modified as necessary.

1. Remove cell culture medium and replace with 10 ml of fresh culture medium. Incubate at 37 °C for 10 min.

2. Cross-link DNA–protein complexes with 270 μl of 37 % form- aldehyde to the culture medium (1 % final conc.). Incubate at room temperature for 15 min (see Note 8).

3. To stop the cross-linking, add 1.47 ml of 1 M glycine per 10 ml of culture medium.

4. Transfer the cell culture dish on ice. Wash twice with 10 ml of ice-cold PBS.

5. Add 1 ml of ChIP cell suspension buffer (see Subheading 2) to the cells. Harvest using a cell scraper.

6. After incubating on ice for 10 min, homogenize the cells using 15 strokes in a Dounce homogenizer and harvest the pellets by centrifuging at 2,000 rpm for 5 min.

7. Resuspend the pelleted nuclei in 5 ml of ChIP lysis buffer (see Subheading 2).

8. Sonicate 15 times using a Heat Systems sonicator at a setting of 40 %, for 20 s each burst. Keep the sample on ice and allow cooling for 1 min between each sonication. The fragment size produced should be 200–500 nucleotides.

9. Centrifuge at 14,000 rpm for 15 min and keep the supernatant only. Measure the protein concentration of the supernatant (using Bio-Rad BCA protein assay kit). Use 1–5 mg protein per immunoprecipitation.

10. Preclear the lysate with 40 μl of Dynabeads Protein G (Invitrogen) per ml of lysate by incubating in cold room for 1 h with constant rotation.

11. Place the tube on a magnet where the beads attach to the side of the tube facing the magnet. Transfer the supernatant to a new tube (see Note 9).

48 Cheol Kyu Hwang et al.

12. Reserve a 50 μl aliquot (5 % of the amount used per immunoprecipitation) at −20 °C for preparation of input DNA.

13. Prepare the sample in 1 ml of ChIP lysis buffer in Eppendorf tubes for immunoprecipitation. Add 2–5 μg antibody. Rotate for 2 h in the cold room (see Note 10).

14. Add 50 μl of Dynabeads Protein G (see Note 11) per immuno- precipitation. Incubate overnight by constant rotation in the cold room.

15. Place the tube on a magnet as above step 11. Remove the supernatant by careful aspiration without disturbing the attached magnet beads.

16. Wash twice with 1 ml of ChIP lysis buffer. All washing steps from here need 10 min of constant rotation in the cold room.

17. Wash twice with 1 ml of wash buffer A (see Subheading 2). Wash twice with 1 ml of wash buffer B (see Subheading 2). Wash twice with 1 ml of TE buffer (see Subheading 2).

18. Add 200 μl of elution buffer (see Subheading 2) to the beads. Incubate for 10 min at 65 °C.

19. Place the tube on a magnet as above step 11. Transfer the supernatant to a new tube. Elute the beads again as above. Combine the eluates to a final volume of 400 μl (adjust with elution buffer, if necessary).

20. Thaw the 50 μl input sample and supplement with 350 μl of elution buffer. Add 21 μl of 4 M NaCl to both the input and immunoprecipitated samples.

21. Incubate the samples for at least 5 h at 65 °C.

22. Add 1 μl of RNase A (from a 10 mg/ml, DNase-free stock). Incubate for 1 h at 37 °C.

23. Add 4 μl of 0.5 M EDTA and 2 μl of 10 mg/ml proteinase K. Incubate for 2 h at 42 °C.

24. Recover the DNA by phenol/chloroform extraction.

25. Add 1 μl of 20 mg/ml glycogen, 40 μl of 3 M sodium acetate, and 1 ml of 100 % ethanol. Mix gently and allow precipitating overnight at −20 °C.

26. Centrifuge at 14,000 rpm for 30 min. Wash once with 70 % ethanol.

27. Resuspend the immunoprecipitate and input samples in 100 μl of 10 mM Tris–HCl (pH 7.5).

28. Analyze the immunoprecipitated DNA by PCR and real-time qPCR. Details of the PCR conditions and analyses of results can be found in our previous studies of opioid receptors [2, 8, 16, 17].

4 Notes

Epigenetic Methods for Opioid Receptor Regulation 49

1. Always use freshly prepared TMB since it gradually deterio- rates. The solution should be stored in dark before use.

2. We have tried other materials for coating the plates, such as poly-L-lysine. Protamine sulfate provided better sensitivity with DNAs on the coated plates.

3. Do not use a cell culture incubator which is moisturized. Use an incubator for bacterial cultures without moisture control for better drying of protamine sulfate solution on plates.

4. Genomic DNA was isolated using a genomic DNA purification kit (Promega, A1125). DNA concentrations were calculated from the absorbance at 260 nm. Be sure to include positive and negative controls as well as DNA samples for a standard curve. Control DNA samples were generated by PCR using different dNTP mix containing normal cytosine, 5-methylcytosine, or 5-hydroxymethylcytosine.

5. During each washing step, avoid drying of the wells. Wash one 96-well plate at a time.

6. For optimal dilution of each antibody, follow manufacturers’ recommendations or optimize empirically.

7. There are more substrates available for the ELISA detections: ABTS (2,2′-azinobis [3-ethylbenzothiazoline-6-sulfonic acid]- diammonium salt, a HRP substrate, less sensitive than OPD and TMB but with background signal, and detectable at 405– 410 nm); OPD (o-phenylenediamine, a HRP substrate, and detectable at 492 nm); PNPP (p-nitrophenyl phosphate, a widely used alkaline phosphatase substrate, and detectable at 405–410 nm). However, TMB is the most popular chromo- genic substrate for HRP detection in ELISA.

8. Cross-linking time is important for consistence and efficiency of ChIP experiments. 10 min incubation is suitable for ChIP analyses of histone modifications, but longer cross-linking times (up to 30 min) are required for ChIP studies on tran- scription factors. In most of our ChIPs, we used 15 min of incubation time in our laboratory.

9. At this step, the samples can be stored at −80 °C for later use. However, the use of fresh chromatin increases the ChIP effi- ciency. Storage for longer than 1–2 months is not recom- mended. Chromatin prepared in ChIP lysis buffer containing 1 % SDS can be stored at 4 °C for 1–2 days until use.

10. Theoptimalratiofortranscriptionalfactorsis~2–4μgofantibody per 1–2×106 cells, although a different ratio was observed for

50 Cheol Kyu Hwang et al.

anti-histone antibodies. Generally, the concentration of the antibody should be determined empirically.

11. For many years, we had used an agarose-based Protein G (Protein G agarose/salmon sperm DNA) immunoprecipita- tion for ChIP assays. Since we switched it to a magnet-based Protein G (Dynabeads Protein G), background signals were greatly reduced probably because the magnet-based Protein G precipitation does not leave much residual amount of solution in each wash step which is the case of the agarose-based method. Magnet brings most of aggregates to the side of a tube and then it is easier to remove almost all solution at each washing step by aspiration.

This work was supported by NIH Grants DA000564, DA001583, DA011806, DA011190, DA013926, and by the A&F Stark Fund of the Minnesota Medical Foundation. The authors declare no conflicts of interest.

Acknowledgments

References

1. Klose RJ, Bird AP (2006) Genomic DNA methylation: the mark and its mediators. Trends Biochem Sci 31:89–97

2. Hwang CK, Song KY, Kim CS et al (2007) Evidence of endogenous mu opioid receptor regulation by epigenetic control of the promot- ers. Mol Cell Biol 27:4720–4736

3. Tahiliani M, Koh KP, Shen Y et al (2009) Conversion of 5-methylcytosine to 5-hydroxymethylcytosine in mammalian DNA by MLL partner TET1. Science 324:930–935

4. Jenuwein T, Allis CD (2001) Translating the histone code. Science 293:1074–1080

5. Wang G, Liu T, Wei LN et al (2005) DNA methylation-related chromatin modification in the regulation of mouse delta-opioid receptor gene. Mol Pharmacol 67:2032–2039

6. Fatemi M, Pao MM, Jeong S et al (2005) Footprinting of mammalian promoters: use of a CpG DNA methyltransferase revealing nucleo- some positions at a single molecule level. Nucleic Acids Res 33:e176

7. Brown PM, Fox KR (1997) Footprinting stud- ies with nucleosome-bound DNA. Methods Mol Biol 90:81–93

8. Hwang CK, Kim CS, Kim do K et al (2010) Up-regulation of the mu-opioid receptor gene is mediated through chromatin remodeling and transcriptional factors in differentiated neuronal cells. Mol Pharmacol 78:58–68

9. Rhee HS, Pugh BF (2011) Comprehensive genome-wide protein-DNA interactions detected at single-nucleotide resolution. Cell 147:1408–1419

10. van Rooij E (2011) The art of microRNA research. Circ Res 108:219–234

11. Hwang CK, Wagley Y, Law PY et al (2012) MicroRNAs in opioid pharmacology. J Neuro- immune Pharmacol 7:808–819

12. He Y, Yang C, Kirkmire CM et al (2010) Regulation of opioid tolerance by let-7 family microRNA targeting the mu opioid receptor. J Neurosci 30:10251–10258

13. Wu Q, Hwang CK, Zheng H et al (2013) MicroRNA 339 down-regulates mu-opioid receptor at the post-transcriptional level in response to opioid treatment. FASEB J 27: 522–535

14. Wu Q, Law PY, Wei LN et al (2008) Post- transcriptional regulation of mouse mu opioid

receptor (MOR1) via its 3′ untranslated region: a role for microRNA23b. FASEB J 22: 4085–4095

15. Ni J, Gao Y, Gong S et al (2013) Regulation of mu-opioid type 1 receptors by microRNA134 in dorsal root ganglion neurons following periph- eral inflammation. Eur J Pain 17:313–323

16. Hwang CK, Song KY, Kim CS et al (2009) Epigenetic programming of mu opioid recep- tor gene in mouse brain is regulated by MeCP2

and Brg1 chromatin remodeling factor. J Cell

Mol Med 13:3591–3615

17. Park SW, Huq MD, Loh HH et al (2005)
Retinoic acid-induced chromatin remodeling of mouse kappa opioid receptor gene. J Neurosci 25:3350–3357

18. Wei LN, Loh HH (2011) Transcriptional and epigenetic regulation of opioid receptor genes: present and future. Annu Rev Pharmacol Toxicol 51:75–97

Epigenetic Methods for Opioid Receptor Regulation 51

Chapter 4

Renilla Luciferase Reporter Assay to Study 3′UTR-Driven Posttranscriptional Regulations of OPRM1

Gabriele Vincelli and Andrea Bedini Abstract

The regulation of MOR expression at the level of mRNA is relevant for its role in pain transmission and in other functions involving opioid receptors. Recently, the role of the 3′UTR in the posttranscriptional regu- lation of MOR expression has been highlighted. Here we describe a Renilla luciferase reporter assay for the study of the effect of any selective treatment on the 3′UTR-dependent regulation of OPRM1 in a model of glial cells.

Key words Mu opioid receptor, 3′UTR, Posttranscriptional regulation, Gene reporter assay, Renilla luciferase

1 Introduction

The regulation of OPRM1 is crucial since the expression of this receptor modulates the activity of endogenous and exogenous ligands. MOR, the primary product of OPRM1, is expressed not only in neurons but also in immune cells [1] and in glial cells [2]. The importance of MOR regulation in these cell types emerges from the influence that they can exert on opioid analgesia: for example glia is a key modulator of neuropathic pain, a chronic status where morphine becomes not only inactive, but sometimes even noxious [3].

MOR expression is regulated at nearly all levels: epigenetic [4], transcriptional [5], posttranscriptional [5], translational [6], and posttranslational [7]. The regulation at mRNA level plays a relevant role in modulating MOR expression. Several studies have been focused on the OPRM1 promoter to elucidate the transcriptional regulation of MOR [8]; however, interest for the 3′UTR of mRNA has recently arisen. This region is about 13,000 nucleotides long and thus represents a potential site for an exten- sive posttranscriptional regulation [9]. Till now, several miRNAs

Santi M. Spampinato (ed.), Opioid Receptors: Methods and Protocols, Methods in Molecular Biology, vol. 1230, DOI 10.1007/978-1-4939-1708-2_4, © Springer Science+Business Media New York 2015

54 Gabriele Vincelli and Andrea Bedini

Fig. 1 Schematic representation of the Gene Reporter Assay described in this chapter. The transfected plas- mids will produce mRNA molecules, in a quantity depending on the promoter and on the effect of the treat- ments. The mRNA will be translated into protein that will give a luminescent readout. The observed difference between the signals emerging from cells transfected with the regulated or unregulated plasmids is due to the 3′UTR, as this is the only sequence that varies between the two plasmids

have been shown to regulate MOR expression through the binding to the 3′UTR [10–12].

A gene reporter assay is a technique that uses a reporter gene, which encodes for an easily detectable protein that is not expressed in humans and does not need posttranslational modifications, fused to the regulatory element of interest. The activity of the reporter protein (for instance the emitted light) is used as a readout of gene expression, so that the effect of the regulatory element can be iso- lated from the native context and studied individually (Fig. 1).

The same reporter gene without any regulatory element inserted is used as control. So far, several reporter genes have been constructed, such as β-galactosidase, firefly luciferase, Renilla lucif- erase, and Cypridina luciferase. The substrate of Renilla luciferase (coelenterazine h) is distinct from that used by firefly luciferase (D-luciferin), making it possible to use the two proteins in the same assay: in dual-luciferase assays one of them is used as the reporter and the other as an internal reference expressed from a constitutive promoter [13]. The presence of an internal reference is useful to normalize sample variations, especially those due to transfection efficiencies. However, attention has to be paid on this issue, as a perfectly constitutive and universally non-inducible promoter has not yet been described. The most commonly used ones can some- times lead to misinterpretation of the results [13–15]. Moreover, the modern advancement of the available commercial transfection reagents has allowed the improvement of transfection efficiency so that experiments carried out with one luciferase often present a lower coefficient of variation when compared to normalized ones [16]. Therefore, in many cases it is possible to design the

2 Materials

2.1 Cell Plating

experiment without an internal reference (see Note 1), which also displays the advantage of reducing the number of plasmids that are transfected in the cells to carry out the experiments. As for the dual-luciferase assays, an independent assay using the unregulated luciferase control (i.e., the same vector lacking the 3′UTR regula- tory element) is still needed to allow comparison across the treat- ments (Fig. 1).

Here we describe a gene reporter assay-based method to study the 3′UTR regulation of MOR expression after a treatment of U87-MG cells, a widely used cell model of human glia. This method employs an optimized Renilla luciferase developed by Switchgear genomics, called RenSP, that presents an enhanced overall enzymatic activity; moreover, this luciferase is fused with a PEST domain that increases the protein turnover, reducing therefore its accumulation that could otherwise interfere with the measurement of expression changes [17]. Switchgear has a library of plasmids expressing RenSP under the regulation of different 3′UTRs, including the one of OPRM1 (see Note 2). As an unreg- ulated control we used the empty vector, which expresses RenSP from the same promoter, but has a minimal 3′UTR sequence (see Note 3). This method is also suitable for high-throughput applications.

1. 70–80 % confluent U87-MG cells (see Note 6).

2. Neubauer cell counting chamber (see Note 4).

3. Trypan blue solution (0.4 %) (Sigma).

4. Sterile 96-well polystyrene black plate with clear bottom, tissue culture treated (Costar).

5. 100× Antibiotic-Antimycotic (Life Technologies).

6. U87-MG medium: Eagle’s minimal essential medium (EMEM, Lonza), 10 % FBS (Lonza), 2 mML-glutamine (Lonza). Add to 445 ml of EMEM, 10 ml of filtered FBS, and 5 ml of L-glutamine 200 mM. Store at 4 °C. Before usage heat the medium at 37 °C and add 1× Antibiotic-Antimycotic.

7. 1× Trypsin/EDTA solution (Lonza). 8. Sterile PBS (Lonza).

1. Opti-MEM® (Life Technologies).

2. Lipofectamine® 2000 transfection reagent (Life Technologies).

3. pLightSwitch_3UTR vector DNA (SwitchGear genomics): one control vector and one vector regulated by the OPRM1 3′UTR (see Notes 2 and 3).

2.2 Cell Transfection

3 ́UTR-Mediated Regulation of OPRM1 55

56 Gabriele Vincelli and Andrea Bedini

2.3 Cell Treatments 1. U87-MG medium (see above).

and Readings

3 Methods

2. The compound, whose effect has to be evaluated, is dissolved in a sterile solution suitable for cell culture.

3. LightSwitch Assay Reagent (SwitchGear genomics). 4. A plate luminometer.

The following protocol is set to evaluate the effect of treatments at 24 h performing the assay in triplicate. The design of the experi- ment is shown in Fig. 2. Twelve wells are needed in total: three that will contain untransfected and vehicle-treated (see Note 5) cells for background measurement (white wells); six wells with cells that will be transfected with the unregulated control plasmid (pat- terned wells); and six wells with cells that will be transfected with the 3′UTR regulated plasmid (grey wells). For each transfected plasmid, three wells will be treated with the vehicle alone (see Note 5), while three wells will be treated with the compound of interest (black box). To add different time points or different treatments, the number of wells has to be scaled-up adding, for each one, three

Fig. 2 Schematic representation of the experimental design proposed in this chap- ter. White wells: cells not transfected and treated with only the vehicle for back- ground measurements. Patterned wells: cells to be transfected with the unregulated control plasmid. Grey wells: cells to be transfected with the 3′UTR-regulated plasmid. Boxed wells: cells that have to be treated with the compound of inter- est. All the other cells have to be treated only with the vehicle (see Note 5)

3.1 Cell Plating

wells for the unregulated plasmid and three for the regulated one. However, no further background wells have to be added.

Always mix thoroughly all the solutions (unless indicated oth- erwise) and prepare a master mix when it’s possible to increase uniformity in the assay conditions. Perform all the passages under a sterile tissue culture hood until the passage #5 of Subheading 3.3.

1. Replace the medium of a 70–80 % confluent U87-MG flask (see Note 6) with 10 ml of trypsin and incubate for 5 min at 37 °C (see Note 7).

2. Transfer the content of the flask in a 15 ml tube.

3. Spin the cells and resuspend the pellet in 1 ml of PBS.

4. Counts the number of cells using a Neubauer chamber. Use Trypan blue staining to count viable cells (see ref. 18 and Note 4).

5. Per each well that has to be seeded add in a tube 15,000 cells suspended in U87-MG medium in a final volume of 100 μl per well. Prepare the mix for one extra well. In the reported example, 240,000 cells in 1.6 ml are needed for 16 wells (see Note 8).

6. Mix thoroughly and seed the cells by pipetting 100 μl of the cell suspension in each well (see Note 9).

7. Incubate the plate for 24 h at 37 °C in 95 % air and 5 % CO2 atmosphere.

The transfection step is carried out using Lipofectamine® 2000. This reagent is guaranteed by the manufacturer to give a very high efficiency of transfection, which is a prerequisite for this experi- ment (see Note 10).

The following protocol (steps 1–6) is used for the transfection of one of the two plasmids (e.g., the control one) in six wells (Fig. 2). The same procedure has to be used for the other plasmid as well. Prepare all the mixes for one extra well: in this example for seven wells.

1. Check on an inverted microscope the condition of the cells in each well. Check that they are all healthy, attached, with a nor- mal morphology and at similar confluence.

2. Prepare in a tube 500 ng of DNA per each well that has to be transfected diluted in Opti-MEM® to a final volume of 65 μl. In the example 3.5 μg of DNA is diluted in 65 μl of Opti- MEM® (see Notes 11 and 12).

3. Prepare in a second tube 1.5 μl of Lipofectamine® 2000 reagent per each well that has to be transfected diluted in Opti-MEM® to a final volume of 65 μl. For instance, 10.5 μl of Lipofectamine® 2000 is diluted in 65 μl of Opti-MEM® (see Note 11).

3.2 Cell Transfection

3 ́UTR-Mediated Regulation of OPRM1 57

58 Gabriele Vincelli and Andrea Bedini

3.3 Treatments and Readings

4. Add the diluted DNA to the diluted transfection reagent, vortex for 20 s and incubate for 20 min at room temperature.

5. Add Opti-MEM® to reach a final volume of 100 μl per each well that has to be transfected. In the example the final volume is 700 μl.

6. Remove the medium from the wells that have to be transfected (see Fig. 2) and add to each one 100 μl of the transfection mix, paying attention to not peel off the cells (see Note 13).

7. Replace the medium of the untransfected background wells with 100 μl of Opti-MEM® each (Fig. 2, white wells).

8. Incubate the plate for 24 h at 37 °C in 95 % air and 5 % CO2 atmosphere.

The day before performing the first Gene Reporter Experiment reconstitute the 100× LightSwitch Assay Reagent. Add 100 μl of the provided “Substrate Solvent” to the tube of lyophilized “Assay Substrate.” Dissolve and aliquot in tubes protected from light, so that each tube has the substrate needed for one experiment: con- sider that, being a 100× substrate, 1 μl for each well is needed. Store each aliquot at −80 °C and thaw aliquots few minutes before use. Avoid additional freeze-thaw cycles.

1. Check on an inverted microscope the condition of the cells in each well. Check that they are all healthy, attached, and with a normal morphology.

2. Prepare two tubes, one containing fresh U87-MG medium (basal medium, see Note 5) and the other containing U87-MG medium adding the compound to be tested at the desired con- centration (treatment medium). Consider at least 100 μl per well. Prepare in excess.

3. Replace the medium in all the untransfected wells and in those that are not going to be treated with the compound of interest (Fig. 2, unboxed wells, vehicle treated) with 100 μl of basal medium (see Note 5).

4. Replace the medium in the wells that have to be treated with the compound of interest (Fig. 2, boxed wells) with 100 μl of treatment medium.

5. Incubate the plate at 37 °C in 95 % air and 5 % CO2 atmo- sphere for the desired amount of time (24 h in this example. See also Note 14).

6. Thaw the “Assay Buffer” provided with the LightSwitch substrate.

7. Few minutes before the treatment time is over dilute 1:100 the provided 100× substrate in “Assay Buffer” for a final volume of 100 μl per well (see Note 15).

3.4 Data Analysis

8. When the treatment time has ended add 100 μl of 1× substrate to each well (see Notes 16–18).

9. Incubate for 30 min at room temperature protected from light (see Note 19).

10. Read in a plate luminometer. For each well read the light emis- sion for 10 s and report the value of counts/second as the total counts divided by 10 (see Note 20).

1. Report the values of counts/second of each well in an excel worksheet (or equivalent).

2. Calculate the average of the signal from background wells.

3. Subtract the background average signal from the signals of all the other wells.

4. Calculate the average of the background-corrected signal for the three wells whose cells were transfected with the control plasmid and treated with only the vehicle; afterwards, express the corrected signal of every well in proportion.

5. Calculate the mean and standard deviation of the signals for each experimental group.

6. For each plasmid employed in the experiment (i.e., the unreg- ulated control plasmid and the regulated one) calculate the ratio of the means of treatment/vehicle. A difference between these values reflects a specific effect of the treatment on the 3′UTR-mediated modulation of gene expression.

7. Perform statistical analysis (e.g., a t-test) to evaluate the signifi- cance of the difference between these two values observed. As an example, Fig. 3 reports the results of an experiment in which the treatment stabilized the OPRM1 mRNA, through the 3′UTR.

1. Check the Coefficient of Variation (CV) as the ratio between the standard deviation and the mean within each experimental group: this should be around 10 %. If the CV observed is sig- nificantly greater than 10 %, it is advisable to use an internal reference control [16]. Many of them are available. Switchgear genomics offers a dual luciferase system (LightSwitch Dual Assay System) that uses a secreted Cypridina luciferase expressed from a constitutive TK promoter. This method has the advantage that the reference luciferase can be measured from a small aliquot taken from each well, while the rest of the assay remains unchanged. Another easy method is to use a dual-luciferase approach, as for example the Dual-Glo® system

3 ́UTR-Mediated Regulation of OPRM1 59

4 Notes

60 Gabriele Vincelli and Andrea Bedini

Fig. 3 Results of a pilot experiment. A stabilizing effect of the treatment on the mRNA level is observed. The graph is reported using mean ± standard deviation of each calculated treatment/vehicle ratio (n = 3)

(Promega). This consists of a homogeneous reagent system that allows the subsequent detection of two different lucifer- ases (firefly and Renilla) within the same well. In this case a control vector expressing an unregulated and constitutively expressed firefly luciferase is needed (as for example the vector pGL4.13 by Promega). Finally, a vector expressing an unregu- lated and constitutively expressed β-galactosidase can be used (as for example pSV-β-galactosidase Control Vector by Promega, using the Beta-Glo® Assay System to detect the sig- nal). The last option has the disadvantage that the two reporter signals have to be detected in two different plates.

In case an endogenous control is included, perform an independent experiment transfecting only the plasmid encoding for this protein to evaluate the effect of the treat- ment on its expression. Check that no significant difference is observed in endogenous control expression between the cells treated with the compound of interest and those treated with vehicle only.

2. The OPRM1 3′UTR is more than 13,000 nucleotides (nt) long [9]. Studying the whole regulatory element at once can give several problems (e.g., the resulting plasmid would be about 16 kbp long and it would be difficult to manipulate, transfect, etc.). Moreover, the plasmid available from Switchgear genomics contains only the first 500 nt of the

3 ́UTR-Mediated Regulation of OPRM1 61

OPRM1 3′UTR. It is advisable to perform experiments using only portions of the 3′UTR, maybe trying to narrow the portion of interest. If interested in studying different portions of the OPRM1 3′UTR, it may be necessary to clone it in the multiple cloning site of the empty pLightSwitch_3UTR vector.

3. Several control plasmids are available from SwitchGear genomics: some contain 3′UTRs of housekeeping genes with different length, while others contain non-conserved, noncod- ing, and non-repetitive human genomic fragments. The easiest control consists instead of the empty vector, containing only a minimal 3′UTR before the poly-A signal. The use of this vec- tor would avoid any unwanted and unexpected unspecific effect that a longer 3′UTR can provide.

4. Other counting methods are suitable. It is advisable to always include a viability control to exclude possible dead cells from the count.

5. In this example the vehicle is U87-MG medium alone. If the compound of interest is dissolved in a particular solvent (e.g., ethanol or DMSO) an equal amount of that solvent has to be added to the vehicle. If possible always prefer to dissolve the compound in cell culture water.

6. For U87-MG the use of cells over the 17th passage should be avoided as a change in some signaling pathways can be obser ved.

7. The use of trypsin to detach cells from the flask will also help to avoid cell clamps that would make the counting step diffi- cult. Always check the flask content at the microscope before spinning the cells down, to control if a longer exposure to trypsin is needed.

8. It is advisable to plate three to six extra wells for backup pur- pose. During all the operations, uniformity among wells is cru- cial. It can sometimes happen that some wells present not well-attached cells or clamps of cells.

9. Plating the cells in the wells of the external perimeter of the plate should be avoided to prevent evaporation of samples that could produce artifacts. It is also preferable to fill these wells with medium.

10. Other reagents yielding high transfection efficiency could be used according to their manufacturer’s recommendations.

11. These steps are designed to be carried out in a small volume to facilitate the DNA-liposome formation. If a bigger amount of wells has to be transfected, the 65 μl volume can be increased, keeping the 1:1 volume proportion of DNA:lipofectamine.

62 Gabriele Vincelli and Andrea Bedini

Try to keep the volume of this mix to about 10–20 % of the final transfection master mix volume.

12. If an internal reference is necessary, the DNA of the plasmid expressing this luciferase has to be considered in the 500 ng of DNA. A usually good ratio between the reporter and reference plasmids is 20:1 in mass. If the total 500 ng of DNA is not changed the quantity of needed Lipofectamine® 2000 remains the same.

13. When removing the medium pay attention not to touch the bottom of the wells with the tip to avoid the detachment of cells. To facilitate the medium removal process it is possible to lean the plate forward. When adding solutions, instead, pipette on the wall of the well to avoid the detachment of cells.

14. If different timings are included in the experiment, replace the medium of each well with basal medium after 24 h from the transfection to block this process. Afterwards, treat the cells so that the treatment times will expire all at the same moment, to increase uniformity in the plate readings.
If treatments longer than 24 h are performed plate 10,000 cells instead of 15,000 on day 1. If a combination of long and short times are performed include consistent vehicle-treated controls.

15. Prepare in small excess. Mix thoroughly, but avoid vortexing to reduce the formation of bubbles.

16. If possible use a multichannel pipette. If it is not possible try to be as precise as possible in pipetting. Rapidity is also important as otherwise the incubation time can vary from the first and last well.

17. The substrate also contains a lysis agent. Therefore, it is suffi- cient to add it in a 1:1 ratio in each well (i.e., 100 μl in each well). If the luciferase signal is low, it is possible to freeze and thaw the plate to help the lysis of the cells and the release of the luciferase protein.

18. In case an internal reference has been added and a dual- luciferase approach has not been chosen, it is necessary to split the samples (see Note 1). In case the Lightswitch dual assay system has been chosen, it is sufficient to take a 20 μl sample for the detection of the Cypridina luciferase signal to be tested in a new plate. In case the galactosidase approach has been chosen, instead, each sample has to be divided in two halves: freeze and thaw the plate and use the pipette tip as a scraper to detach the cells from the well. Collect the 100 μl samples in clean tubes and use 50 μl for the luciferase assay in a new plate, mixing the sample with 50 μl of LightSwitch Assay Reagent. Use the other 50 μl of sample for the assessment of the other reporter activity with the Beta-Glo® Assay System.
In all cases refer to the protocols provided by the manufacturer.

References

1. Börner C, Kraus J, Bedini A et al (2008) T-cell receptor/CD28-mediated activation of human T lymphocytes induces expression of func- tional μ-opioid receptors. Mol Pharmacol 74:496–504

2. Watkins LR, Hutchinson MR, Johnston IN et al (2005) Glia: novel counter-regulators of opioid analgesia. Trends Neurosci 28: 661–669

3. Watkins LR, Hutchinson MR, Milligan ED et al (2007) “Listening” and “talking” to neu- rons: implications of immune activation for pain control and increasing the efficacy of opi- oids. Brain Res Rev 56:148–169

4. Hwang CK, Song KY, Kim CS et al (2007) Evidence of endogenous mu opioid receptor regulation by epigenetic control of the pro- moters. Mol Cell Biol 27:4720–5736

5. Wei LN, Loh HH (2002) Regulation of opioid receptor expression. Curr Opin Pharmacol 2: 69–75

6. Song KY, Hwang CK, Kim CS et al (2007) Translational repression of mouse mu opioid receptor expression via leaky scanning. Nucleic Acids Res 35:1501–1513

7. Williams JT, Ingram SL, Henderson G et al (2013) Regulation of μ-opioid receptors: desensitization, phosphorylation, internaliza- tion, and tolerance. Pharmacol Rev 65: 223–254

8. Bedini A, Baiula M, Spampinato S (2008) Transcriptional activation of human mu-opi- oid receptor gene by insulin-like growth fac- tor-I in neuronal cells is modulated by the transcription factor REST. J Neurochem 105:2166–2178

9. Ide S, Han W, Kasai S et al (2005) Characterization of the 3′ untranslated region of the human mu-opioid receptor (MOR-1) mRNA. Gene 364:139–145

10. Wu Q, Law PY, Wei LN et al (2008) Post- transcriptional regulation of mouse mu opioid

receptor (MOR1) via its 3′ untranslated region: a role for microRNA23b. FASEB J 22:4085–4095

11. Ni J, Gao Y, Gong S et al (2013) Regulation of μ-opioid type 1 receptors by microRNA134 in dorsal root ganglion neurons following periph- eral inflammation. Eur J Pain 17:313–323

12. Lu Z, Xu J, Xu M et al (2014) Morphine regu- lates expression of μ-opioid receptor MOR-1A, an intron-retention carboxyl terminal splice variant of the μ-opioid receptor (OPRM1) gene via miR-103/miR-107. Mol Pharmacol 85:368–380

13. Shifera AS, Hardin JA (2010) Factors modu- lating expression of Renilla luciferase from control plasmids used in luciferase reporter gene assays. Anal Biochem 396:167–172

14. Matuszyk J, Ziolo E, Cebrat M et al (2002) Nurr1 affects pRL-TK but not phRG-B inter- nal control plasmid in genetic reporter system. Biochem Biophys Res Commun 28(1036): 1039

15. Ho CK, Strauss JF III (2004) Activation of the control reporter plasmids pRL-TK and pRL- SV40 by multiple GATA transcription factors can lead to aberrant normalization of transfec- tion efficiency. BMC Biotechnol 30:4–10

16. Switchgear technical note. Co-transfection controls with reporter assays. http://switch- geargenomics.com/sites/default/files/pdf/ LightSwitch_CoTfx.pdf. Accessed 23 April 2014

17. Switchgear technical note. LightSwitch System Over view http://switchgeargenomics.com/ sites/default/files/pdf/LightSwitch_Manual. pdf. Accessed 23 April 2014

18. Bastidas O. Cell Counting with Neubauer Chamber. Basic Hemocytometer Usage. Celeromics technical note. http://www.celeromics. com/en/resources/docs/Articles/Cell- counting-Neubauer-chamber.pdf. Accessed 23 April 2014

3 ́UTR-Mediated Regulation of OPRM1 63

19. If more than one plate is used, add the substrate at different times, so that every plate will be read at the luminometer after exactly 30 min.

20. If there is a high number of wells to be read it can be possible to reduce the reading time of each well to reduce the gap between the readings of the first and last well.

Chapter 5 of Opioid-Induced Alterations in Discrete Brain Areas

Michal Korostynski, Marcin Piechota, Slawomir Golda, and Ryszard Przewlocki

Abstract

Whole-genome screening methods are unique approach to search for novel genes and molecular pathways involved in drug action. High-throughput profiling allows the gene expression levels of tens of thousands of transcripts to be measured simultaneously. Here, we describe transcriptional profiling in a specific area of the brain using DNA microarrays and next-generation sequencing.

Key words DNA microarrays, Genomic signature, Mouse model, Next-generation sequencing, Transcriptome resequencing

1 Introduction

Whole-genome sequencing projects of human, mouse, and other model organisms have been recently completed [1]. These sequences encode the genetic instructions for development, struc- ture, and function of the organism. The next challenge is to determine how and when cells use their genetic information to control gene expression and the synthesis of new proteins. The temporal profile of gene transcription alterations provides markers of biological processes taking place in a particular cell, tissue, or system. Whole transcriptome profiling in functionally distinct regions of the brain allows to search for candidate genes and molecular factors involved in mechanism of psychotropic drug action. Regulation of gene expression is complex and a dynamic process [2]. Moreover, half-life of mRNAs is relatively short; there- fore, experimental design and selection of optimal time points is particularly important in this type of studies [3]. The results pro- vide a researcher with a global picture of cellular functions during precisely selected conditions and within a specific time window.

Santi M. Spampinato (ed.), Opioid Receptors: Methods and Protocols, Methods in Molecular Biology, vol. 1230, DOI 10.1007/978-1-4939-1708-2_5, © Springer Science+Business Media New York 2015

High-Throughput Gene Expression Profiling

66 Michal Korostynski et al.

High-throughput methods are unique tools that allow evaluat- ing patterns of co-expressed transcripts and investigating networks of functionally related genes, including regulatory factors and their effectors [4]. The introduction of DNA microarrays has opened a new era of whole-genome transcriptional analysis in molecular biology [5]. The core principle behind microarrays is hybridization between two DNA strands. The technology has subsequently evolved from self-spotted cDNA arrays to silicon chips printed with short oligonucleotide probes. Probes are selected on the basis of their sequence specificities and synthesized in situ on a solid surface (from Affymetrix) or attached to the beads (from Illumina). The main advantages of the gene expression profiling using micro- arrays are standardization and whole transcriptome coverage. However, this method allows detecting and analyzing only previ- ously annotated transcripts. Technological advances have resulted in the development of transcriptome resequencing methods based on next-generation sequencing technologies [6]. This technique provides an opportunity to gather a full profile of expressed genes that includes quantitative determination of transcripts which use alternative start, termination sites, and splicing variants. The main advantage of this technology is the fact that the mRNA sequences sought need not be known a priori; thus, unknown genes or expression of novel gene variants can be discovered.

Here, we demonstrate that opioid-induced changes in brain gene expression can be effectively measured using whole-genome microarrays [7–9]. We were able to dissect (Fig. 1) a major

Fig. 1 Gene expression profiling of drug-induced alterations in the brain using microarrays. (a) Hierarchical clus- tering of morphine-dependent transcriptional alterations in mouse striatum. Microarray results shown as a heat map with colored rectangles represent transcript abundance. (b) Morphine-induced gene expression networks in the striatum. Gene networks α, and β identified by hierarchical clustering of top drug-regulated transcripts. Statistical analysis, clustering, and network analysis were performed using genes2mind online resource

2 Materials

2.1 Animals

opioid-regulated gene networks implicated in the control of neuronal signaling, brain metabolism, and organization of cell projections in mouse striatum, define time-course of the altera- tions, and compare to effects of several psychotropic drugs. Moreover, integration of bioinformatic tools with expression profiling provides opportunity to identify functional enrich- ments and putative regulatory factors for the gene expression patterns. The research was further completed by implementation of RNA sequencing methodology. Novel drug-regulated tran- scriptional forms were identified and cataloged.

Despite the limitations of an animal model, the comparison of drug-induced dynamic alterations in the rodent (mouse or rat) striatal gene expression profile provides insights into molecular mechanisms of action of psychotropic drugs.

1. Adult male (8–10 weeks old) C57BL/6J inbred mice (Jackson Laboratory, Bar Harbor, ME, USA) were housed 6–10 per cage under a 12-h dark/light cycle with free access to food and water.

2. Animals weighing 20–30 g were used throughout the experiments.

The effective dose of morphine was based on the literature, par- ticular attention being paid to the pharmacokinetics and pharma- cological effects in C57BL/6J mice.

1. Mice were intraperitoneally injected (vol. 10 ml/kg) with 20 mg/kg morphine dissolved in saline.

2. Then the animals were sacrificed by decapitation 1, 2, 4, or 8 h after a single injection along with the appropriate vehicle and naïve control groups (six animals per every drug-treated and control group) (see Note 1).

The striatum is a brain region responsible for control of motiva- tion, reward-based learning, and decision-making. The striatum as an evolutionarily ancient brain region reveals comparable functions and gene expression profiles between rodents and humans.

1. After decapitation, brains were removed from the skulls and dissected rapidly (see Note 2).

2. Samples containing the rostral part of the caudate putamen and the nucleus accumbens, referred as the striatum, were col- lected [10].

3. Tissue samples were placed in RNAlater reagent (Qiagen Inc., Valencia, CA, USA) and preserved at −70 °C.

2.2 Drugs

2.3 Tissue Collection

Opioid-Induced Gene Expression Profile 67

68 Michal Korostynski et al.

3 Methods

3.1 RNA Isolation and Quality Check

Sufficient quality and quantity of total RNA are essential for gene expression analysis using genome-wide technologies (see Note 3).

1. Samples were homogenized in 1 ml Trizol reagent (Invitrogen, Carlsbad, CA, USA). RNA was isolated following the manu- facturer’s protocol and further purified using the RNeasy Mini Kit (Qiagen Inc.).

2. The total RNA concentration was measured using an ND-1000 Spectrometer (NanoDrop Technologies Inc., Montchanin, DE, USA).

3. RNA quality was determined using a microfluidics-based plat- form—Agilent Bioanalyzer 2100 (Agilent, Palo Alto, CA, USA) (Fig. 2).

After extraction from the tissue mRNA is converted to cDNA (complementary DNA) using reverse transcriptase. The cDNA works as a complementary copy of the expressed mRNA in a DNA form. The oligonucleotides covering the microarray slide are hybridized to fluorescently labeled cRNA (RNA complementary to cDNA)

3.2 Microarray Hybridization

Fig. 2 High quality of 12 total RNA samples extracted from mouse brain, as measured by the appearance of bands on electropherograms of RNA samples after capillary electrophoresis

3.3 Microarray Data Analysis

samples and the fluorophore emission from the hybridized samples is registered by the microarray scanner. The signal intensity is con- sidered to be a relative measurement of mRNA abundance.

1. A starting amount of 200 ng high-quality total RNA (pooled 1:1 from two animals) was used to generate cDNA and cRNA with the Illumina TotalPrep RNA Amplification Kit (Illumina Inc., San Diego, CA, USA) (see Note 4).

2. The cDNA served as a template for in vitro transcription with T7 RNA polymerase and biotinylated UTP to generate multi- ple copies of biotinylated cRNA. Such cRNA is conjugated with the streptavidin-bound fluorophore indirectly through the biotin–streptavidin interactions.

3. Every cRNA sample (1.5 μg) was hybridized overnight to the MouseWG-6 BeadChip array (Illumina). Subsequently, chips were washed, dried, and scanned with the BeadArray Reader (Illumina).

4. Three independent biological replicates were used per time- point. To provide an overall appropriately balanced dataset, samples from treatment group were evenly distributed between arrays and hybridization batches (see Note 5).

In contrast to many of the techniques used in molecular biology, a relatively short time is required to obtain raw microarray data com- paring the time spent on analyses. To obtain comprehensive results and biologically relevant answers from such data, the proper use of statistical methods is obligatory [11].

1. Microarray data preprocessing and quality control were per- formed using BeadArray R package v1.10.0 [12]. After back- ground subtraction, data was normalized using quantile normalization and then log2-transformed (see Note 6).

2. Statistical analyses were performed in R software [13]. Two- way ANOVA with fixed effects for drug factor, time factor, and interaction was followed by appropriate correction for multiple testing (using the Bonferroni and Benjamini-Hochberg proce- dures) (see Note 7) (Fig. 1a).

3. Identification of co-expressed gene transcription modules and further bioinformatic analyses were performed using available online resources. Clusters of co-expressed genes were iden- tified using the single-linkage clustering method using our interactive database genes2mind (www.genes2mind.org) (Fig.1b) [7]. The functional annotation and analysis tool DAVID was used to identify overrepresented ontologic groups among the gene expression patterns (david.abcc.ncifcrf.gov) [14]. The analysis of common motifs in promoter regions of co-expressed genes was performed with seqinspector (seqinspector. cremag.org) (Table 1).

Opioid-Induced Gene Expression Profile 69

70 Michal Korostynski et al.

Table 1
Putative molecular regulators of morphine-induced gene expression alterations in the striatum were identified using seqinspector online tool

Transcriptional regulator (track name)

Query

Background

Fold difference

P value

Bonferroni corrected P value

GR_04

1.7

0.56

3.0

1.3E−5

0.003

TBP_01

17.0

8.0

2.1

1.5E−4

0.034

TBP_02

30.0

14.0

2.1

6.5E−4

0.14

GR_03

1.9

0.62

3.1

8.6E−4

0.19

STAT5A_01

0.85

0.47

1.8

0.0065

1.0

PPARG_02

1.5

1.1

1.3

0.0097

1.0

NPAS4_01

0.074

0.054

1.4

0.01

1.0

CEBPB_06

1.5

0.7

2.2

0.011

1.0

E2F1_01

2.0

0.41

4.9

0.013

1.0

GATA1_01

6.2

4.0

1.5

0.014

1.0

3.4 Transcriptome Resequencing

3.4.1 Library Preparation

3.4.2 Template Preparation

The development of NGS technology leads to benchtop sequencing systems that are capable of human-scale exome or transcriptome resequencing in a few hours [15]. We present simplified protocol for total RNA expression profiling using Ion Proton instrument.

The library preparation starts from total RNA. After digestion with DNAse III and rRNA depletion digest the RNA with RNAse III (see Note 8), purify on magnetic beads, hybridize and ligate to the Ion Adapter Mix, and then set the reaction for reverse transcrip- tion (Applied Biosystems). Purify the obtained cDNA, amplify and assess the yield and size distribution by chip-based capillary elec- trophoresis using Agilent High Sensitivity DNA Kit and Agilent Bioanalyzer 2100 (Agilent) (Fig. 3a).

For the template preparation we perform emulsion PCR using the Ion OneTouch 2 instrument (Applied Biosystems) (see Note 9). Briefly, dilute the amplified cDNA library to obtain 11pM final concentration, load the Ion PI Plus Reaction Filter Assembly with the diluted library, Ion Sphere Particles (ISPs), and all the neces- sary reagents delivered by the manufacturer, and run the templa- tion reaction. The reaction takes about 6.5 h. Then purify and enrich the positively templated Ion Sphere Particles on the Ion OneTouch ES System with biotinylated magnetic beads (Applied Biosystems). Such ISPs are ready for sequencing within 3 days.

Fig. 3 Whole transcriptome resequencing of mouse striatum using NGS. (a) Critical quality check points for NGS sample preparations: total RNA, RNA after rRNA depletion, RNA after RNAase III digestion, and an ampli- fied DNA library. (b) Ion Proton chip quality check after the sequencing

3.4.3 Next-Generation Sequencing (NGS)

3.4.4 NGS Data Analysis

For the sequencing of template positive ISPs we use Ion Proton Next Generation Semiconductor sequencer (Applied Biosystems) (see Note 10). After cleaning and initialization the machine is ready for the sequencing run. Load the positively templated ISPs on the Ion PI Chip according to the manufacturer’s protocol, introduce the chip into machine, and start new run. After approximately 3 h preliminary report will be ready. In order to obtain complete NGS raw data, additional several hours is required (Fig. 3b).

Analysis of RNA-seq data is still a challenging task. Bioinformaticians can choose now from a variety of tools. For read mapping the most popular tool is Tophat (if genome is well annotated like human or mouse). For differential expression testing cuffdiff and deseq are the most frequently used tools.

1. Quality control
Run a quality control check on your data using the FASTQC tool. The covered quality measures include per base sequence quality, per sequence quality scores, per base sequence con- tent, per base GC content, per sequence GC content, per base N content, sequence length distribution, sequence duplica- tion levels, overrepresented sequences, kmer and content (see Note 11). Other quality measures including percent of intronic reads, percent of intergenic reads, and percent of unaligned reads can be calculated after alignment using bed- tools and samtools packages.

2. Alignment of reads
The next step is mapping the reads to the reference genome (Fig. 4). The major challenge is that the reads span over the splice junction boundaries. Thus, they cannot be directly aligned to reference genome in which splice junctions are not present.

Opioid-Induced Gene Expression Profile 71

72 Michal Korostynski et al.

Fig. 4 Example NGS reads after aligned to mouse genome at Tsc22d3 gene position

4 Notes

Hence typical short read mappers such as Bowtie and BWA are not suitable for this task. Instead, it is better to use splice junction mappers such as Tophat or STAR. Use the Tophat tool to map RNA-seq reads to recent genome assembly (mm10 for mouse) (see Note 12) [16]. Use gene annotations to inform Tophat where the splice junctions are. These annotations can be down- loaded from UCSC Table Browser in “gtf” format. The results of this step are aligned reads in a bam format.

3. Transcript quantification and differential expression testing Use cufflinks to estimate transcripts abundance from bam files with previously downloaded annotations as a guide [17]. Use fpkm values from “isoforms.fpkm_tracking” file for differential expression testing. FPKM values can be used for parametric sta- tistic after log 2(1+n) transformation. Alternatively use cuffdiff as a differential expression testing tool (see Note 13) (Fig. 5).

1. To minimize influence of additional factors on the gene expres- sion profile all the animals in the experiment should be treated in accordance to the same procedure. The design of experiment

Opioid-Induced Gene Expression Profile 73

Fig. 5 Expression of short drug-regulated transcriptional variants of Sgk1 identi- fied using NGS (regulated form was indicated by red color)

should take into consideration important biological factors such as circadian rhythm and the effects of vehicle or different genetic backgrounds. They can significantly influence gene expression profiles. Planning the experiments with detailed timetable is critical for the drug administration and tissue collection.

2. Precautions for RNA handling are critical for successful gene expression analysis. Tissue dissection should be performed as quickly as possible and at low temperature (on ice). Moreover, RNA should be stabilized during storage and transportation.

3. The quality of the extracted total RNA must always be esti- mated without bias. The best method to assess RNA quality is based on a chip-based electrophoresis on the Agilent Bioanalyzer platform (Fig. 2). This method is suitable for qualitative and quantitative analysis of RNA and allows evaluating the integrity of the RNA scored as an RNA Integrity Number (RIN). Typically, total RNA samples with a RIN between 8 and 10 are considered to be of a good quality. All samples in experiment should be of a similar quality. The use of RNA stabilization reagent such as RNAlater (Qiagen Inc.) is recommended to maximize quality during tissue collection and extraction.

4. For oligonucleotide microarrays, at least 100 ng of high-quality total RNA is needed, and about 1 μg is optimal. When the starting material is highly limited, a linear amplification of the RNAs using a two-cycle protocol is possible. However, to ensure that the initial differences between samples are preserved, it is highly recommended that RNA quality and quantity should be carefully checked prior to the experiment.

74 Michal Korostynski et al.

5. Three independent biological replicates are considered to be the minimal number of samples per experiments. Six to nine replicates would be close to the optimal and most cost-effective range (behavioral and pharmacological experiments usually consists of six to nine replicates in parallel). The number of replicates and distribution of samples across hybridization batches require careful consideration. To reduce variation or to obtain a sufficient amount of starting material, samples from two or three individuals might be pooled and hybridized on a single microarray.

6. Variations in experimental and hybridization conditions require data normalization to reduce systematic variation in gene expres- sion measurements. It is also possible to reduce the batch to batch hybridization variability using z-score transformation.

7. Usually, it is better to use a combination of statistical tests as a primary means of filtering rather than simple fold changes (ratio of mRNA abundance to mean levels in control and experimen- tal groups). Second, correction for multiple comparisons is nec- essary. Because it is not always easy to define the balance between the levels of false positive and false negative results, the optimal method of correction of the data should be ascertained. For instance, use of the Bonferroni correction (λ=λ/n, when an experimenter is testing n hypotheses) might be too stringent for experiments with a low number of replicates.

8. For the NGS library generation extract total RNA from tissue with Trizol Reagent. Always remember to use columns for digestion with DNAse I. Traces of DNA may interfere with the following steps in protocol. As a starting amount use 4 μg of total RNA. After the rRNA depletion (RNAse III treat- ment for 3 min) 70–80 ng of RNA is obtained. Digestion rRNA depleted RNA with RNAse III for 3 min. Do not size select the RNase III treated samples in agarose gels. Size selec- tion is not recommended because of high probability of losing shorter transcripts. Purify and concentrate samples with Speedvac to ~3 μl. While amplifying the library set the PCR to 18 cycles (not 16 as stated in automatic protocol).

9. For Ion PI template preparation always use the newest chemis- try. Older batches of regents may give low key signal and high number of low-quality reads in sequencing. Proper calculation of template dilution factor (TDF) is essential to keep balance between mono- and polyclonal Ion Sphere Particles (ISPs). Library concentration of 11 pM usually gives 15–18 % tem- plated ISPs.

10. While sequencing only use the newest chemistry for longer reads (200 bp or longer). Follow the protocol except for: use 20 μl of Annealing Buffer (Applied Biosystems) instead of

Opioid-Induced Gene Expression Profile 75

15 μl while annealing sequencing primer. The final volume should be 50 μl not 45 μl. Beware of Ion PI Chip loading issues. Load the chip in air conditioned area and do not shorten the centrifugation time below 10 min.

11. After quality check it is recommended to trim reads to remove base positions that have a low median score (softclipping). Trimmed reads are more likely to be aligned to the genome.

12. Tophat splice junction mapper behaves better if guided by genome annotations about positions of splice junctions. If genome positions are poor or unavailable use STAR map- per instead.

13. The bioinformatics field still does not have a good handle for differential expression at isoform-level resolution. At gene level you can use a tool like HTSeq to count the number of reads overlapping a gene, then use a count-based differential-expression approach such as edgeR or DESeq to assess differential expression.

This work was supported by NCN 2011/03/D/NZ3/01686 SONATA, POIG De-Me-Ter 3.1, and 2013/08/A/NZ3/00848 Maestro.

Acknowledgments

References

1. Butler D (2010) Human genome at ten: science after the sequence. Nature 465:1000–1001

2. Lamb J, Crawford ED, Peck D et al (2006) The Connectivity Map: using gene-expression signatures to connect small molecules, genes, and disease. Science 313:1929–1935

3. Yu H, Luscombe NM, Qian J et al (2003) Genomic analysis of gene expression relation- ships in transcriptional regulatory networks. Trends Genet 19:422–427

4. Liang M, Cowley AW, Greene AS (2004) High throughput gene expression profiling: a molecular approach to integrative physiology. J Physiol 554(Pt 1):22–30

5. Draghici S, Khatri P, Eklund AC et al (2006) Reliability and reproducibility issues in DNA microarray measurements. Trends Genet 22:101–109

6. Wang Z, Gerstein M, Snyder M (2009) RNA- Seq: a revolutionary tool for transcriptomics. Nat Rev Genet 10:57–63

7. Korostynski M, Piechota M, Dzbek J et al (2013) Novel drug-regulated transcriptional networks in brain reveal pharmacological

properties of psychotropic drugs. BMC

Genomics 14:606
8. Korostynski M, Piechota M, Kaminska D et al

(2007) Morphine effects on striatal transcrip-

tome in mice. Genome Biol 8:R128
9. Piechota M, Korostynski M, Solecki W et al (2010) The dissection of transcriptional mod- ules regulated by various drugs of abuse in the

mouse striatum. Genome Biol 11:R48
10. Ziolkowska B, Gieryk A, Bilecki W et al (2005) Regulation of alpha-synuclein expression in limbic and motor brain regions of morphine-

treated mice. J Neurosci 25:4996–5003
11. Slonim DK (2002) From patterns to pathways: gene expression data analysis comes of age.

Nat Genet 32(Suppl):502–508
12. Dunning MJ, Smith ML, Ritchie ME et al

(2007) R classes and methods for Illumina bead-based data. Bioinformatics 23: 2183–2184

13. Gentleman RC, Carey VJ, Bates DM et al (2004) Bioconductor: open software develop- ment for computational biology and bioinfor- matics. Genome Biol 5:R80

Michal Korostynski et al.

14.

15.

da Huang W, Sherman BT, Lempicki RA (2009) Systematic and integrative analysis of large gene lists using DAVID bioinformatics resources. Nat Protoc 4:44–57

Rothberg JM, Hinz W, Rearick TM et al (2011) An integrated semiconductor device enabling non-optical genome sequencing. Nature 475:348–352

16. Trapnell C, Pachter L, Salzberg SL (2009) TopHat: discovering splice junctions with RNA-Seq. Bioinformatics 25:1105–1111

17. Trapnell C, Williams BA, Pertea G et al (2010) Transcript assembly and quantification by RNA-Seq reveals unannotated transcripts and isoform switching during cell differentiation. Nat Biotechnol 28:511–515

Part II Cellular Detection and Analysis of Opioid Receptors

Chapter 6

Real-Time Imaging of Mu Opioid Receptors by Total Internal Reflection Fluorescence Microscopy

Cristina Roman-Vendrell and Guillermo Ariel Yudowski Abstract

Receptor trafficking and signaling are intimately linked, especially in the Mu opioid receptor (MOR) where ligand-dependent endocytosis and recycling have been associated with opioid tolerance and depen- dence. Ligands of MOR can induce receptor endocytosis and recycling within minutes of exposure in heterologous systems and cultured neurons. Endocytosis removes desensitized receptors after their activa- tion from the plasma membrane, while recycling promotes resensitization by delivering functional recep- tors to the cell surface. These rapid mechanisms can escape traditional analytical methods where only snapshots are obtained from highly dynamic events.

Total internal reflection fluorescence (TIRF) microscopy is a powerful tool that can be used to inves- tigate, in real time, surface trafficking events at the single molecule level. The restricted excitation of fluo- rophores located at or near the plasma membrane in combination with high sensitivity quantitative cameras makes it possible to record and analyze individual endocytic and recycling event in real time. In this chapter, we describe a TIRF microscopy protocol to investigate in real time, the ligand-dependent MOR trafficking in Human Embryonic Kidney 293 cells and dissociated striatal neuronal cultures. This approach can pro- vide unique spatio-temporal resolution to understand the fundamental events controlling MOR trafficking at the plasma membrane.

Key words G protein-coupled receptor, MOR, TIRF, Live cell imaging, Endocytosis, Recycling, Resensitization

1 Introduction

Ligand-induced receptor endocytosis and recycling have been proposed to have important roles on physiological events such as the development of tolerance and resensitization [1, 2]. Studies carried out in different neuronal and non-neuronal cell models have demonstrated that endocytosis of the Mu opioid receptor (MOR) is cell type and ligand specific [3–6]. For instance, mor- phine fails to rapidly internalize receptors in most cells, whereas [D-Ala2, N-MePhe4, Gly-ol]-enkephalin (DAMGO) has been shown to induce receptor internalization and recycling [2].

Santi M. Spampinato (ed.), Opioid Receptors: Methods and Protocols, Methods in Molecular Biology, vol. 1230, DOI 10.1007/978-1-4939-1708-2_6, © Springer Science+Business Media New York 2015

80 Cristina Roman-Vendrell and Guillermo Ariel Yudowski

2 Materials

2.1 Cells
and Culture Media

Moreover, several studies correlate the low endocytic efficacy of morphine with the development of tolerance in vivo [7, 8]. These trafficking mechanisms have been mostly studied by biochemical and time-lapse imaging approaches. However, endocytosis and recycling occur within minutes of receptor activation and single and rapid events can escape traditional techniques. Furthermore, we are only beginning to grasp the complexity of the mechanisms and kinetics of individual trafficking events.

New technologies like quantitative live cell imaging can pro- vide new insight into these complex mechanisms during receptor trafficking with high spatio-temporal resolution [9–11]. In our laboratory, we use total internal reflection fluorescence (TIRF) microscopy to investigate MOR trafficking at the single molecule resolution [12]. Relying on the selective illumination of a thin layer of the sample, TIRF allows visualization of the events occurring at plasma membrane with high signal-to-noise ratio. We and others have previously utilized TIRF to investigate the ligand- dependent trafficking of the MOR in heterologous systems and neuronal cultures [12–15].

In these studies, we use Human Embryonic Kidney 293 cells (HEK293) and striatal neurons expressing a pH-sensitive GFP variant called superecliptic phluorin (SEP) fused to the amino ter- minal of MOR [16]. SEP-MOR fluorescence is visible at the cell surface, where the extracellular pH is neutral. The fluorescence is reversibly quenched when the receptor enters the mildly acidic endocytic and recycling vesicles. This property of SEP-MOR facili- tates the detection of discrete events mediating the removal and delivery of receptors at the cell surface [16, 17]. Here, we describe the use of TIRF microscopy to investigate the kinetics and molecular mechanisms during MOR trafficking at the plasma membrane. This protocol describes the necessary steps, and important controls, to explore the ligand-dependent endocytosis and recycling of MOR at the single vesicle level.

1. Human embryonic kidney cells (HEK293; ATCC CRL-1573) (see Note 1).

2. Striatal primary cultures obtained from embryonic day 17–18 Sprague-Dawley rat embryos. Alternatively, brain tissue can be purchased from BrainBits LLC (Springfield, IL).

3. HEK293 culture media: Dulbecco’s Modified Eagle’s medium supplemented with 10 % fetal bovine serum and 1 % penicillin/ streptomycin (Life Technologies).

4. Neuron culture media: Neurobasal medium supplemented with B27 (according to manufacturer’s protocol) and 0.5 mM glutaMAXTM (Life Technologies).

2.2 TIRF Microscopy Equipment
and Settings

5. Imaging media: Opti-MEM® reduced serum medium supple- mented with 20 mM HEPES (Life Technologies) (see Note 2).

6. 30 mm coverslips #1.5 thickness (Bioptechs) acid wash treated and coated with poly D-Lysine (Sigma).

7. Transfection reagents: Lipofectamine 2000 (Life Technologies) or Effectene (Qiagen).

8. MOR-SEP cDNA [13] construct (see Note 3). Multiple fluo- rescently tagged proteins can be imaged by co-transfection.

9. DAMGO, [D-Ala2, N-MePhe4, Gly-ol]-enkephalin. 10. Morphine sulfate salt.

1. Motorized Nikon (Melville, NY) Ti-E inverted microscope with a CFI-Apo × 100 1.49 oil TIRF objective lens with color correc- tion and a motorized stage with perfect focus (see Note 4).

2. Lightsource:488and561nmCoherentsapphirelasers(Coherent Inc. Santa Clara, CA) 50 and 100 mW lasers, respectively.

3. Temperature control is utilized to keep cells at 37 °C with a Stable Z stage and objective warmer (Bioptechs, Butler, PA).

4. Interchangeable Coverslip Dish (Bioptechs) (http://www. bioptechs.com/Products/ICD/coverslipdish.html).

5. Camera: iXonEM, DU897 back illuminated EMCCD camera (Andor, Belfast, UK).

6. Readout speed: 10 Hz, exposure time: continuous 100 ms exposure for receptor recycling, EM gain 300, binning: 1×1, image: 512 × 512, BitDepth = 14 bits, camera temperature: −75 °C, and laser power: 10 % for 488 nm.

7. Software: NIS-Elements.

1. Acid clean coverslips by incubation in 1 M HCl shaking over- night. Rinse the coverslips two times with abundant ddH2O and once with 70 % ethanol. Leave in 95 % ethanol until ready to use. Before coating, UV and dry in the culture hood.

2. Coat acid clean coverslips with 100 μg/mL poly-d-lysine for 2–4 h at 37 °C. Wash 3 times with sterile water and air-dry in sterile environment.

3. Passage HEK293 cells onto the prepared 35 mm plates so that they reach 50–60 % confluence on the day of transfection (see Note 5).

4. Plate 300,000 neurons per 35 mm dish. Neurons are trans- fected at 4–5 DIV and imaged >15 DIV.

Imaging Individual Trafficking Events of the Mu Opioid Receptor 81

3 Methods

3.1 Cell Culture

Cristina Roman-Vendrell and Guillermo Ariel Yudowski

3.2

Live Cell Imaging

5. Transfect cells with DNA constructs using Lipofectamine 2000 or Effectene according to the manufacturer’s instructions. We perform experiments 48–72 h after transfection to allow cell recovery from the transfection and achieve optimal expres- sion levels. (High expression levels will impair observation of individual events.)

6. On the day of imaging, carefully transfer the coverslip to the interchangeable coverslip dish and add 2 mL of freshly pre- pared Opti-MEM with HEPES replacing the incubation media 15–30 min before imaging sessions (see Note 6).

7. Incubate the cells at 37 °C for 10 min to allow acclimatization.

1. At least 30 min before any acquisition, turn on the microscope and the temperature controllers. Turn on the laser key and let the laser warm up.

2. Select TIRF objective, add a drop of immersion oil (Type LDF, RI: ~1.515) and carefully place the interchangeable coverslip dish on the stage (see Note 7).

3. Temperature of the imaging media must be controlled regu- larly and kept constant (37 °C).

4. To reduce the effects of photobleaching, it is important to find and focus the cells using transmission light first. Then, find cells expressing tagged receptors using epifluorescence and then switch to TIRF illumination (see Note 8) (Fig. 1).

5. Add agonist (DAMGO 10 μM) diluted in warm imaging media by automated perfusion system or manually outside the imag- ing area to minimize artifacts from media changes (see Note 9).

6. Acquisition settings for endocytosis: 100–300 ms exposures every 2–3 s. Total time: 10–30 min.

7. Acquisition settings for recycling: Continuous illumination and acquisition at 100 ms exposures for 1–2 min (see Note 10).

8. Imaging sessions will generate large amounts of data. Careful data management must be implemented in advance. Standardized electronic notebooks or spreadsheets are recommended.

1. To obtain single molecule information, single EGFP analysis (see Note 11) is performed regularly to help compensate for day-to-day variability. Mean fluorescent intensity of single EGFPs is obtained by combining all single measurements from multiple experiments (Fig. 2). This information is used to cor- relate fluorescence intensity with the number of SEP-MOR receptors per recycling vesicle.

2. Endocytic events are analyzed by double blind analysis, multiple times manually and using the particle tracking algorithm 2D spot tracker [18]. Individual event location, time, and fluorescence

3.3 Analysis

Imaging Individual Trafficking Events of the Mu Opioid Receptor 83

Fig. 1 Example of a neuron imaged using epifluorescence illumination (left panel) and TIFR microscopy (right panel)

Fig. 2 Single molecule quantification of EGFPs. Left panel, representative fluorescence intensity measurement of a single EGFP. Right panel, histogram depicting single GFP intensity distribution

profile are logged and recorded electronically. Endocytic events are identified and scored following previously described behav- iors, briefly: (a) events must appear and disappear within the time series; (b) events must display limited movement (no more than 4 by 4 pixels through their lifetime), and (c) events must not fuse or collide with each other [19, 20].

3. Recycling events are analyzed using the open source program, ImageJ and FIJI (NIH). Recycling receptors are observed as abrupt increases of surface fluorescence in diffraction-limited spots. Maximum intensity projections of each treatment can be compared for changes in recycling frequency (Fig. 3).

84 Cristina Roman-Vendrell and Guillermo Ariel Yudowski

Fig. 3 Example of SEP-MOR expressing HEK293 cell. Panel a, vesicles at the cell surface and recycling events can be visualized by maximum intensity projection for a HEK293 cell. The image represents 60 s acquired at 10 Hz. Each fluorescence spot surrounded by the circle represents a recycling event. Panel b, kymograph of the representative cell, with increasing time from left to right. An example of recycling is indicated by the arrow

4 Notes

1. We do not recommend HEK293T; these cells achieve high expression levels preventing imaging of vesicular events.

2. HEPES is used to maintain the pH constant for up to 45–60 min outside a CO2 incubator.

3. High-quality cDNA is required especially for neuronal trans- fection. To investigate mechanism involved in MOR traffick- ing, fluorescently tagged dominant negative versions of known recycling players can be co-transfected with SEP-MOR. These mutations produce an altered gene product that acts antagonis- tically to the wild-type allele [21]. Fluorescently labeled siRNA can be used to selectively knock-down target proteins from individual cells while investigating SEP-MOR trafficking.

4. Focal plane must be kept constant during imaging sessions.

5. It is very important that the cells grow in monolayer and are not more than 80–90 % confluent on the day of imaging.

6. With forceps, take the coverslip from the 35 mm dish and place on the dish base, tighten the threaded insert with O-ring, and make sure that the media does not leak.

7. It is very important that the bottom of the imaging dish is completely dry and clean. Any liquid or dirt will interfere during TIRF imaging.

8. The most critical step is to find the exact angle for TIRF. To align the laser properly, focus on the plasma membrane. You can find the cell sharp edges and use them as reference.

Imaging Individual Trafficking Events of the Mu Opioid Receptor 85

9. DAMGO is dissolved in DMSO, a highly viscous solvent, and must be mixed well with imaging media before adding to the cells. If adding manually, be very careful not to disturb the cells within the imaging area. Controls should be performed to test the effects of DMSO on surface fluorescence and basal cell activity.

10. Agonist-induced MOR recycling can be observed 2–3 min after agonist exposure. A constant rate of vesicular fusion is generally observed at ~10 min.

11. Utilizing the linear range of our EMCCD camera, we corre- lated the number of single EGFPs to the number of SEP- MORs observed during our imaging.

This work was supported by research grants from NIH DA023444, R01DA037924, Puerto Rico Science Trust, and NIMHD 8G12-MD007600 (RCMI). We would also like to thank Stephanie Palacio for providing control epifluorescence versus TIRF images.

Acknowledgments

References

1. Koch T, Widera A, Bartzsch K et al (2005) Receptor endocytosis counteracts the develop- ment of opioid tolerance. Mol Pharmacol 67: 280–287

2. Koch T, Höllt V (2008) Role of receptor inter- nalization in opioid tolerance and dependence. Pharmacol Ther 117:199–206

3. Whistler JL, Chuang HH, Chu P et al (1999) Functional dissociation of mu opioid receptor signaling and endocytosis: implications for the biology of opiate tolerance and addiction. Neuron 23:737–746

4. Bushell T, Endoh T, Simen AA et al (2002) Molecular components of tolerance to opiates in single hippocampal neurons. Mol Pharmacol 61:55–64

5. Bailey CP, Couch D, Johnson E et al (2003) Mu-opioid receptor desensitization in mature rat neurons: lack of interaction between DAMGO and morphine. J Neurosci 23:10515–10520

6. Haberstock-Debic H, Kim K-A, Yu YJ et al (2005) Morphine promotes rapid, arrestin- dependent endocytosis of mu-opioid receptors in striatal neurons. J Neurosci 25:7847–7857

7. Grecksch G, Bartzsch K, Widera A et al (2006) Development of tolerance and sensitization to different opioid agonists in rats. Psychophar- macology 186:177–184

8. Enquist J, Kim J, Bartlett S (2011) A novel knock-in mouse reveals mechanistically distinct

forms of morphine tolerance. J Pharmacol 338:

633–640
9. Schmoranzer J, Goulian M, Axelrod D et al

(2000) Imaging constitutive exocytosis with total internal reflection fluorescence micros- copy. J Cell Biol 149:23–32

10. Steyer JA, Almers W (2001) A real-time view of life within 100 nm of the plasma membrane. Nat Rev Mol Cell Biol 2:268–275

11. Wennmalm S, Simon SM (2007) Studying individual events in biology. Annu Rev Biochem 76:419–446

12. Roman-Vendrell C, Yu YJ, Yudowski GA (2012) Fast modulation of μ-opioid receptor (MOR) recycling is mediated by receptor ago- nists. J Biol Chem 287:14782–14791

13. Yu YJ, Dhavan R, Chevalier MW et al (2010) Rapid delivery of internalized signaling recep- tors to the somatodendritic surface by sequence-specific local insertion. J Neurosci 30:11703–11714

14. Soohoo AL, Puthenveedu MA (2013) Divergent modes for cargo-mediated control of clathrin-coated pit dynamics. Mol Biol Cell 24:1725–1734

15. Henry AG, Hislop JN, Grove J et al (2012) Regulation of endocytic clathrin dynamics by cargo ubiquitination. Dev Cell 23:519–532

16. Miesenbock G, De Angelis DA, Rothman JE (1998) Visualizing secretion and synaptic

86 Cristina Roman-Vendrell and Guillermo Ariel Yudowski

transmission with pH-sensitive green fluorescent

proteins. Nature 394:192–195
17. Sankaranarayanan S, De Angelis D, Rothman

JE et al (2000) The use of pHluorins for opti- cal measurements of presynaptic activity. Biophys J 79:2199–2208

18. Sage D, Neumann FR, Hediger F et al (2005) Automatic tracking of individual fluorescence particles: application to the study of chromosome dynamics. IEEE Trans Image Process 14: 1372–1383

19. Saffarian S, Cocucci E, Kirchhausen T (2009) Distinct dynamics of endocytic clathrin- coated pits and coated plaques. PLoS Biol 7: e1000191

20. Flores-Otero J et al (2014) Ligand-specific endocytic dwell times control functional selectivity of the cannabinoid receptor 1. Nat Commun 5:4589 doi:10.1038/ncomms5589

21. Herskowitz I (1987) Functional inactivation of genes by dominant negative mutations. Nature 329:219–222

Chapter 7

In Vivo Techniques to Investigate the Internalization Profile of Opioid Receptors

Amynah A. Pradhan, Vivianne L. Tawfik, Alycia F. Tipton, and Grégory Scherrer

Abstract

G-protein-coupled receptors (GPCRs) regulate a remarkable diversity of biological functions, and are thus often targeted for drug therapies. Receptor internalization is commonly observed following agonist bind- ing and activation. Receptor trafficking events have been well characterized in cell systems, but the in vivo significance of GPCR internalization is still poorly understood. To address this issue, we have developed an innovative knock-in mouse model, where an opioid receptor is directly visible in vivo. These knockin mice express functional fluorescent delta opioid receptors (DOR-eGFP) in place of the endogenous recep- tor, and these receptors are expressed at physiological levels within their native environment. DOR-eGFP mice have proven to be an extraordinary tool in studying receptor neuroanatomy, real-time receptor traf- ficking in live neurons, and in vivo receptor internalization. We have used this animal model to determine the relationship between receptor trafficking in neurons and receptor function at a behavioral level. Here, we describe in detail the construction and characterization of this knockin mouse. We also outline how to use these mice to examine the behavioral consequences of agonist-specific trafficking at the delta opioid receptor. These techniques are potentially applicable to any GPCR, and highlight the powerful nature of this imaging tool.

Key words Behavior, Delta opioid receptor, G-protein-coupled receptor, Immunohistochemistry, Ligand-directed signaling, Mouse, Pain, Receptor trafficking

1 Introduction

G-protein-coupled receptors (GPCRs) form the most abundant receptor class in the human genome [1]. A variety of biological functions are regulated by GPCRs, and this receptor class is most commonly targeted for pharmacological therapies. Activation of a GPCR by endogenous or synthetic agonists elicits receptor signal- ing via heterotrimeric G proteins. This process is highly regulated and receptor activation is often accompanied by receptor inter- nalization, a key process in the modulation of receptor signaling. Internalization is a complex regulatory process which terminates

Santi M. Spampinato (ed.), Opioid Receptors: Methods and Protocols, Methods in Molecular Biology, vol. 1230, DOI 10.1007/978-1-4939-1708-2_7, © Springer Science+Business Media New York 2015

Amynah A. Pradhan et al.

1.1
of Opioid Receptors

receptor cell surface signaling, but can also initiate different signaling cascades intracellularly (for review see [2]). In addition, receptor internalization and subsequent receptor trafficking also serves to recycle receptors back to the cell surface, or designate activated receptors for degradation [3]. The specific trafficking events that occur following receptor internalization are dependent on a number of factors, including the specific receptor, the ago- nist, duration of exposure, and cell type. Ligand-induced recep- tor internalization serves as an important modulator of receptor signaling.

Importantly, not all agonists elicit the same signaling and traf- ficking events. GPCRs can exist in multiple conformations, and different agonists can stabilize different activation states [4]. Each receptor conformation in turn produces distinct receptor–effector complexes, which can initiate differing signaling and receptor traf- ficking events [5, 6]. Therefore, not all agonists produce receptor internalization, and ligand-specific differences in receptor traffick- ing have acute and long-term consequences. This concept, referred to as ligand-directed signaling/trafficking, functional selectivity, or biased agonism [7] has important biological and pharmacological implications.

The opioid receptor family regulates a number of important physi- ological processes, including pain, reward, mood, and stress [8, 9]. The most clinically effective analgesics are mu opioid receptor (MOR) agonists (e.g. morphine and fentanyl). These are limited in their clinical applications as they can produce a number of adverse effects, including addiction. Delta opioid receptors (DOR) offer a promising alternative to MOR ligands. Stimulation of delta opioid receptors does not result in any of the severe adverse effects associ- ated with MOR agonists, including euphoria (for review see [10]). DOR agonists are also effective in chronic inflammatory and neu- ropathic pain models [1–16]. In addition, DORs also modulate emotional state. Genetic deletion of DOR or its endogenous ligand, enkephalin, results in anxiogenic and depressive-like behav- iors [17, 18]. Further, DOR agonists also produce anxiolytic and antidepressant effects [19–21]. Considering these preclinical find- ings, DOR agonists are currently being developed for the treat- ments of pain, and emotional disorders; and a greater understanding of the consequences of DOR activation and internalization is imperative.

The majority of studies characterizing receptor internalization have been performed in transfected cellular systems. These in vitro systems are highly useful in identifying specific receptor trafficking and signaling events that occur within the cell. However, these in vitro models do not always reflect in vivo systems in terms of receptor density, protein content of receptor-expressing cells, or even receptor localization within subcellular compartments, as

Cell Trafficking

1.2 Future Applications of In Vivo Analysis of Opioid Receptor Trafficking

must be considered for neurons [22]. In addition, cellular models cannot provide insight into how receptor trafficking influences integrated responses in the living organism. Compared to our cel- lular understanding of receptor trafficking, the in vivo significance of this highly regulated process is still in its infancy.

In vivo trafficking of mu opioid receptors has been previously investigated using immunohistochemical techniques [23–25]. However, a lack of specific antibodies has limited this strategy for characterization of the delta opioid receptor [26]. To address this issue, we have created knockin mice expressing a fully functional fluorescent delta opioid receptor (DOR-eGFP) in place of the endogenous DOR [27]. In these animals, DOR-eGFP receptors are expressed at physiological levels within their native environ- ment. Furthermore, these receptors are directly visible in vivo. These mutant mice have proven to be an exceptional tool for studying receptor neuroanatomy, real-time receptor trafficking in live neurons, and receptor movements in vivo [26–31].

The DOR-eGFP mice have proven to be a unique tool to directly monitor in vivo delta opioid receptor trafficking events. These mice are particularly valuable considering the paucity of specific antibodies for immunohistochemical detection of DOR, and we have shown that they provide a dynamic, noninvasive way to moni- tor receptor internalization. There are a number of exciting uses for these knockin mice that are only just being explored. In this chapter, we focus on receptor internalization following exogenous agonist administration; however, these mice can also be used to track release of endogenous opioids. Faget et al. [32] recently found that in a behavioral model of context-dependent withdrawal, DOR-eGFP internalization was transiently detected in a subset of hippocampal neurons. This receptor trafficking revealed regionally restricted endogenous opioid peptide release. Importantly, inter- nalization to endogenous opioid release consistently showed a pool of receptors remaining on the cell surface, which was in marked contrast to internalization following stimulation with the small molecule DOR agonist, SNC80 [32]. These studies show that, in vivo, physiological and pharmacological stimulation pro- duces distinct delta receptor regulation.

DOR-eGFP mice can also be used to examine receptor trans- location to the cell surface. Several lines of evidence indicate that the cell surface population of DORs may be pliable within specific cell types, and is increased following certain stimuli. Peripheral injury [11, 33, 34], morphine treatment [35–37], ablation of MOR [38], and chronic ethanol exposure [39] are a few of the events that have been proposed to increase functional DORs on the cell surface. Recently, DOR-eGFP mice were used to show learning-related plasticity of delta opioid receptors. Increased pre- dictive learning was directly associated with increased translocation

Examining Receptor Internalization In Vivo 89

90 Amynah A. Pradhan et al.

2 Materials

2.1 In Vivo Trafficking of DOR- eGFP (Pradhan Lab)

of DOR-eGFP to the cell surface in cholinergic interneurons of the nucleus accumbens shell [28]. Receptor trafficking is an active process of translocation of receptor to the cell membrane, and sub- sequent activation and internalization of the receptor away from the surface; and future studies will focus on clarifying this in vivo dynamic.

G-protein-coupled receptor trafficking is regulated by a num- ber of modulatory proteins, and interactions of this nature have primarily been studied in vitro. Along with several other modula- tory proteins, DORs have been shown to interact with β-arrestins [40–42], regulator of G protein signaling 4 [43–45], and G-protein- coupled receptor-associated sorting proteins [46, 47]. Mutant mouse lines exist for these proteins, and there is the opportunity to cross DOR-eGFP knockin mice with these lines. Such crosses would provide insight on the role of these regulatory proteins on in vivo DOR trafficking.

More broadly, the DOR-eGFP knockin strategy highlights the potential of labeled receptors in vivo. Brain receptor imaging is a major focus of research in neuroscience. Standard imaging tech- niques such as positron emission tomography and magnetic reso- nance imaging provide a wealth of information on anatomy and receptor occupancy, but they have limited resolution in the small animal. At the other end of the spectrum, receptor internalization and trafficking have important functional implications, but work in this field has primarily been restricted to cell systems. Mouse models where proteins of interest are replaced by labeled versions provide a powerful tool to observe specific molecular events within the complex circuitry of in vivo physiological processes.

In this chapter we outline how to construct and characterize a DOR-eGFP knockin mouse, and how to use this knockin mouse model to specifically examine in vivo receptor internalization in a pain-based behavioral assay.

1. Manual von Frey hairs (weight in grams: 0.008, 0.04, 0.07, 0.16, 0.4, 0.6, 1, 2).

2. Wire testing rack (IITC or made in house).
3. Complete Freund’s Adjuvant (CFA, Sigma).
4. 25 μl Hamilton syringe with 27 gauge beveled needle. 5. Rodent anesthesia machine with isoflurane and O2.
6. SNC80, ARM390 (Tocris), distilled H2O, saline.

7. Syringes (1 cc), needles (27 gauge), and feeding needles for oral gavage.

2.2 Transcardial Perfusion and Tissue Preparation (Scherrer Lab)

2.3 Immuno- histochemistry

2.3.1 Antibodies

3 Methods

3.1 Construction and Characterization of the DOR-eGFP Mouse

3.1.1 DOR-eGFP Receptor

0.2 M phosphate buffer stock: Na2PO4 122 g, NaH2PO4 21 g, ddH2O to 4 l.

0.1 M phosphate buffered saline: 0.2 M PB 500 ml, ddH2O 500 ml, NaCl 9 g.

30 % (w/v) sucrose: Sucrose 150 g, 0.1 M PB to 500 ml.

Glycerol cryoprotectant: Ethylene glycol 150 ml, glycerol 150 ml, 0.2 M PB 50 ml, ddH2O 150 ml.

4 % Formaldehyde: 0.1 M PBS 27 ml, formaldehyde 3 ml.

1. 0.1 M PBS-0.3 % Triton X-100 (PBST): 0.1 M PBS 500 ml, Triton X-100 1.5 ml.

2. 5 % NDST block: PBST 19 ml, normal donkey (or goat) serum 1 ml.

3. 1 % NDST block: PBST 99 ml, normal donkey (or goat) serum 1 ml.

1. Rabbit anti-green fluorescent protein, Life technologies, 1:1,000.

2. Donkey anti-rabbit Alexa Fluor 488-conjugated, Life technol- ogies, 1:1,000.

The key to generating a knockin mouse expressing a fluorescently labeled GPCR is to ensure that the fusion protein maintains its endogenous properties and function. When designing the knockin strategy, there are three main parameters that can be modified: (1) the nature of the tag (fluorescent protein, FLAG-tag, HA-tag, etc.); (2) the position of the tag in the native protein (C-terminus, loop, N-terminus); and (3) the composition of the linker, amino acid identity and length, between the GPCR and the tag. We elected to use the enhanced green fluorescent protein (eGFP) as a tag because of its brightness, photostability and functionality as a monomer, compared to other fluorescent proteins available at the time (CFP, YFP, RFP). Numerous other fluorescent proteins are now available and it is essential to select the fluorescent tag based on future applications [48]. Previous work indicated that C-terminal fusions allow correct receptor folding, insertion, and export to the cell surface with minimal alterations to ligand bind- ing or receptor signaling [49]. We therefore tested multiple con- structs in vitro and determined that linking the eGFP protein to the DOR C-terminus via a five amino-acid linker (GSIAT) resulted in retained receptor function. This was confirmed in radioligand binding experiments on HEK293 cell lines that stably expressed either the native DOR or the DOR-eGFP fusion. Our results dem- onstrated that a variety of opioid ligands (endogenous opioid

Examining Receptor Internalization In Vivo 91

Amynah A. Pradhan et al.

3.1.2

Targeting Strategy

peptides, agonists, antagonists) all displayed similar affinities for the two receptors (native and fluorescent) [27], suggesting that the addition of the eGFP tag did not alter receptor conformation.

Next, we determined whether DOR signaling through Gi/o protein coupling was intact, given the possibility of steric hindrance from the fluorescent tag. We treated membrane preparations expressing either native or DOR-eGFP receptors with the agonist deltorphin II and found that the potency and maximal effect of this agonist was identical for the two receptors [50]. Following receptor activation, binding of multiple molecules to the DOR intracellular domains has been shown to lead to internalization and subsequent lysosomal or proteasomal degradation [41, 51–54]. To confirm that the DOR-eGFP fusion protein maintained native internalization and desensitization we used the stable DOR-eGFP HEK293 cell line and found that under basal conditions the fluo- rescent signal was primarily at the cell surface, however, following the application of DOR agonists (met-enkephalin, deltorphin II or SNC80), internalization of the fusion protein occurred as expected. In agreement with previous reports [55, 56], and with native receptor action, DOR-eGFP was downregulated following pro- longed exposure to deltorphin II. In summary, we demonstrated that addition of the eGFP tag did not alter DOR conformation, G protein binding, interaction with adaptor proteins or internal- ization, and we therefore selected this construct to create the knockin mouse line.

Our goal when designing the knockin targeting strategy was to minimize modifications to the genome that might disrupt gene transcription or RNA stability. The targeting vector was con- structed by inserting the sequence encoding the GSIAT-eGFP tag to replace the STOP codon in exon 3 of the DOR gene (Oprd1; Fig. 1). In this construct, all the Oprd1 gene sequences regulating gene transcription, mRNA stability and protein translation were unchanged and therefore intact in the final recombined allele. To select for recombined alleles in ES cells, we inserted a floxed hygromycin resistance gene as close as possible (less than 300 base pairs) to the GSIAT-eGFP sequence to ensure selection of clones that had integrated both the resistance gene and the tagged sequence. Once the clones which had integrated the targeting con- struct by homologous recombination were identified, we removed the hygromycin cassette by transfecting ES cells with a Cre recom- binase-expressing vector. The final mutant DOR-eGFP allele con- tained the desired GSIAT-eGFP vector and one remaining loxP sequence of 34 base pairs.

We confirmed the expression of the Oprd1 gene in the DOR- eGFP mice and found a slight alteration in processing that led to an approximate twofold increase in DOR-eGFP mRNA levels in the knockin mouse compared to wild-type controls. It is possible

Examining Receptor Internalization In Vivo 93

Fig. 1 Tagging delta opioid receptors in vivo with eGFP by homologous recombination in mice. (a) Schematic structure of the DOR-eGFP receptor. The eGFP tag is fused to the C-terminus of the receptor by means of a five amino acid linker (GSIAT). (b) Strategy used to insert the eGFP cDNA in place of the stop codon within the Oprd1 gene. Oprd1 exons, eGFP cDNA, and the floxed hygromycin cassette are displayed as empty box, gray box and black triangles, respectively. Homologous recombination (HR) was followed by Cre recombinase treat- ment (Cre) in ES cells

that the exogenous GSIAT-eGFP sequence increases mRNA sta- bility as we also found a parallel increase in receptor density (320 vs. 160 fmol/mg brain membrane protein), and maximal level of G protein activation (Emax) to deltorphin II and SNC80, but not met-enkephalin. In transfected systems, overexpression levels are in the picomolar range suggesting that our observed increase in expression level is comparatively subtle. In addition, this increase in receptor number did not impact drug responsiveness, suggest- ing the existence of a receptor reserve in wild-type animals.

94 Amynah A. Pradhan et al.

3.1.3 Receptor Expression and Function in DOR-eGFP Knockin Mouse Line

Overall, our data suggest that the DOR-eGFP mice recapitulate the DOR expression found in physiological conditions and there- fore can serve as a tool for probing the distribution of the DOR in the central nervous system (CNS).

Once the DOR-eGFP knockin mouse line was established, we first tested whether the DOR-eGFP fusion protein was functional in mutant animals. We utilized membrane protein preparations from DOR-eGFP mice and control littermates to perform a series of biochemical and pharmacological assays. Western blot analysis demonstrated an 80 kDa eGFP immunoreactive band, correspond- ing to the predicted size of the fusion protein, and found only in homozygous and heterozygous mutant mice, but not in wild-type controls. Furthermore, ligand-binding experiments confirmed that ligand affinity for the DOR-eGFP receptor was identical to that of the wild-type receptor, regardless of the nature (partial or full ago- nists, antagonists) or structure (peptides or alkaloids) of the ligand tested. Finally, using a G protein activation assay, we found that potencies for prototypical DOR agonists, SNC80, deltorphin II and met-enkephalin were similar in knockin and wild-type mice. Together, these results indicate that DOR-eGFP is expressed and functional in tissues from DOR-eGFP knockin mice.

We next needed to confirm that the fusion protein was expressed in the correct anatomic distribution and remained func- tional. To do this we used fluorescence microscopy on CNS tissue and directly compared the pattern of eGFP fluorescence to the known distribution of wild-type DOR as determined by autoradio- graphic binding [57–59]. Fluorescence microscopy demonstrated that the DOR-eGFP was expressed at particularly high levels in the striatum, cortex, basolateral nucleus of the amygdala, similar to previously published results [57]. The agreement in expression pattern between the eGFP-labeled and native receptor indicates that the eGFP fusion protein is found in the same neuronal popu- lation as the native receptor and that it is correctly transported to cellular compartments such as dendrites and axon terminals. Finally, in agreement with previous data using transfected cells [41, 52, 54], DOR-eGFP was internalized upon exposure to an agonist either in striatal or hippocampal primary cultures or in vivo (see below, and [27, 30]). Taken together, these data strongly sup- port the notion that DOR-eGFP is expressed and traffics as the native DOR does in wild-type mice.

Next, we tested whether DOR-eGFP mice exhibited the known behavioral responses to DOR agonists using the com- pound SNC80 for its well-documented stimulant effect. We found a dose-dependent increase in locomotor activity that was comparable in DOR-eGFP mice and their wild-type littermates, but absent in DOR knockout mice [27]. Furthermore, in a

3.1.4 Anatomical Characterization of DOR Receptor Distribution

model of inflammatory pain, we determined that the analgesic effects of SNC80 (the prototypical nonpeptide DOR agonist) and AR-M100390 (ARM390) were identical in DOR-eGFP mutant mice and wild-type controls (see below and [60]).

Taken together, our biochemical, pharmacological, histologi- cal, and behavioral studies have demonstrated that the DOR-eGFP fusion protein is a fully functional receptor with unchanged distri- bution and biology. Importantly, in spite of a slightly higher recep- tor number in DOR-eGFP mice, DOR-mediated responses in vivo are unchanged in the knockin mouse line. These data support the use of the DOR-eGFP mouse as a unique tool to gain insights into several aspects of DOR function at a resolution that was previously not possible.

As described above, the DOR-eGFP mouse allows direct visualiza- tion of receptor distribution in all CNS tissue and allows for immu- nohistochemical colocalization experiments that can discern subpopulations of neurons expressing DOR (Fig. 2). For example, the distribution of Oprd1 (DOR gene) and choline acetyl- transferase mRNAs was investigated in adjacent sections by in situ hybridization and a large overlap was noted between the two pop- ulations [50, 51]. In agreement with this work, immunostaining of striatal sections from DOR-eGFP knockin mice showed that 76 % of cholinergic neurons expressed the DOR [27]. Recently, we used DOR-eGFP mice to resolve the molecular identity of DOR- expressing hippocampal neurons. These studies indicated that DOR is predominantly expressed in GABAergic inhibitory interneurons in the oriens and principal layers [52]; further, colocalization experiments in many brain areas are currently underway to provide a more complete description of DOR anatomy within neural circuits.

The DOR-eGFP mouse was also used to study receptor distri- bution in the spinal cord and peripheral nervous system. Molecularly diverse primary afferent neurons of the dorsal root ganglia (DRG) detect a wide range of stimulus types and intensities to encode a variety of somatosensory modalities such as heat pain or light touch [61]. In situ hybridization studies have demonstrated DOR mRNA in large diameter DRG neurons with myelinated axons, and in a restricted population of small-diameter DRG neurons [59, 62]. The molecular identity of these DOR-expressing cells is key to understanding the function of the receptor and we took advantage of the DOR-eGFP mice to perform colocalization studies with markers for the various neuronal populations. Consistent with pre- vious in situ hybridization studies, we found that about 60 % of DOR-eGFP-expressing neurons were large-diameter DRG neurons that immunostain for neurofilament 200, a marker of myelinated neurons (A fibers) [26, 63]. Some of these cells transmit mechanical

Examining Receptor Internalization In Vivo 95

96 Amynah A. Pradhan et al.

Fig. 2 Identification of neurons expressing the delta opioid receptor using DOR-eGFP mice. (a) Cholinergic (ChAT) neurons (NeuN) in the striatum express the delta opioid receptor (DOR-eGFP). DOR-eGFP mice were treated with SNC80 to cause receptor internalization. (b) In the hippocampus of SNC80-treated mice DOReGFP is concentrated in inhibitory interneurons, marked with an antibody against glutamate acid decarboxylase (GAD).

3.2 In Vivo Trafficking of Delta Opioid Receptors

pain and touch information, and play a role in the development of chronic mechanical hypersensitivity after tissue or nerve injury (inflammatory or neuropathic pain, respectively). In addition, DOR-expressing small-diameter DRG neurons were identified as the nonpeptidergic population that detects noxious stimuli as indicated by colocalization with the lectin IB4. In contrast, DOR- eGFP was found to only rarely colocalize with markers of the pep- tidergic small-diameter population of neurons: only about 2 % of DOR-eGFP DRG neurons expressed substance P and only 5 % of DOR-eGFP DRG neurons expressed the heat sensor transient receptor potential vanilloid 1 (TRPV1) [26, 63].

In the spinal cord, in situ hybridization and autoradiographic studies have demonstrated Oprd1 mRNA and DOR binding in neurons throughout the gray matter (laminae I-X) of the ventral and dorsal horns [59, 64]. Our studies with the DOR-eGFP mice confirm this expression pattern with preliminary work suggesting co-labeling with PKCγ, a marker of interneurons critical for neuro- pathic pain [65] in the ventral part of lamina II. Identification of the DOR-eGFP expressing neurons throughout the rest of the spinal cord is currently underway.

Taken together, these data show that the knockin of DOR- eGFP does not adversely affect receptor binding or expression. In addition, these mice are viable and behave similar to wild-type controls. We next used these mice to correlate receptor trafficking with in vivo behavioral effects.

As described above, the DOR-eGFP mouse allows direct visualiza- tion of the receptor, which we then used to characterize in vivo receptor trafficking events. Evidence from a number of in vitro cel- lular studies has revealed functional selectivity at the delta opioid receptor (for review see [66]). We used DOR-eGFP knockin mice to explore the relationship between agonist-induced receptor internalization and receptor function further. We compared two structurally related DOR agonists, SNC80 [67], and ARM390 [68] in several in vivo and ex vivo assays. These two ligands show similar binding, and G protein coupling [26], and like other sys- temically administered DOR agonists [69, 70], both SNC80 and ARM390 produce analgesia in a model of inflammatory pain, and

Examining Receptor Internalization In Vivo 97

Fig.2 (continued)(c)Inthespinalcord,DOR-eGFPfluorescenceispresentthroughoutthegreymatter(GM)and enriched in lamina I and in the ventral portion of inner lamina II (IIi) of the dorsal horn (DH). WM white matter. VH ventral horn. (d) In the dorsal root ganglia, DOR-eGFP is expressed in subsets of small (arrowhead) and large (arrow) diameter neurons. (e) Labeling of dorsal root ganglion (DRG) neurons with the marker isolectin B4 (IB4) reveals that small-diameter neurons expressing DOReGFP belong mainly to the nonpeptidergic subset of DRG neurons (IB4 positive). Arrow, large-diameter DRG neuron expressing DOR-eGFP

98 Amynah A. Pradhan et al.

3.2.1 In Vivo Observation of DOR-eGFP Trafficking

have anxiolytic effects [30, 31]. However, in primary striatal and hippocampal cultures made from DOR-eGFP mice, SNC80 and ARM390 had profoundly different internalization profiles [30]. SNC80 produced internalization at concentrations as low as 10 nM [27, 30]. In contrast, internalization with ARM390 was only observed in cultures at a 100× higher concentration (1,000 nM). These results show biased agonism at the delta opioid receptor, where the binding of two different agonists induces highly diver- gent receptor trafficking events.

We next examined the effect of ligand-directed trafficking in vivo, and we compared these two DOR agonists in the Complete Freund’s Adjuvant (CFA) model of inflammatory pain. Following repeat injection, SNC80, but not ARM390, produced acute behav- ioral desensitization. Upon examination of receptor trafficking, we showed that inhibition of agonist-induced behavior and decreased DOR-G protein coupling were directly correlated with receptor internalization [30]. In addition, we also examined the long-term effects of treatment with these two different DOR agonists. We found that distinct forms of tolerance were observed depending on the internalization property of the drug [31]. Chronic treatment with the high-internalizing agonist SNC80 produced widespread receptor downregulation and a generalized behavioral tolerance. In contrast, the low-internalizing agonist ARM390 produced no detectable change in receptor localization, number, or binding. Only an analgesic tolerance was observed following chronic ARM390, which corresponded to an uncoupling of DOR from voltage- dependent Ca2+ channels in DRG neurons (Fig. 3). Taken together, these results show that ligand-specific internalization of DORs in vivo needs to be considered during the development of these agonists as novel drug therapies.

1. Habituate DOR-eGFP mice to testing rack for 20 min for 2 days (see Note 1).

2. Determine baseline mechanical responses in the plantar surface of either the left or right hindpaw. Mechanical responses are determined by using von Frey hairs. The 50 % threshold for response to these punctate mechanical stimuli is determined according to the up-and-down method [71]. In this case, the plantar surface of the hindpaw is stimulated with a series of eight von Frey filaments. The first filament tested is 0.4 g. In the absence of a response a heavier filament (up) is tried, and in the presence of a response a lighter filament (down) is tested. This pattern is followed for a maximum of four filaments following the first response. A response is defined as a lifting, shaking, or licking of the paw upon stimulation.

3. Inflammatory pain is induced by injecting 13 μl of CFA into the plantar surface of the hindpaw. Inject the same hindpaw in which

Examining Receptor Internalization In Vivo 99

Fig. 3 Chronic ligand-directed trafficking at the delta opioid receptor results in two distinct forms of tolerance. SNC80 and ARM390 (ARM) have comparable selectivity and potencies for the delta opioid receptor, but result in highly distinct receptor trafficking events. Systemic SNC80, but not ARM390, produces clear receptor inter- nalization in vivo as shown in slices from DOR-eGFP knockin mice (representative images from the hippocam- pus and dorsal root ganglia) [30]. Chronic administration of either agonist produces two distinct forms of tolerance. Repeated administration of SNC80 produces widespread receptor downregulation, thus resulting in a generalized tolerance where all DOR agonist-induced behaviors are inhibited. In contrast, chronic adminis- tration of the low-internalizing agonist, ARM390, affects DOR coupling to Ca2+ channels in dorsal root ganglia, thus producing tolerance only at the level of pain processing [31]

baseline mechanical responses were determined. Lightly anes- thetize the mice with isoflurane (5 % to induce anesthesia, and 0.8–1.5 % for maintenance). Gently insert the needle (bevel side up) of the filled Hamilton syringe subcutaneously into the center of the plantar surface of the paw. Pull the plunger up 1–2 μl to ensure that the needle is not in an artery. Slowly inject CFA, remove needle, and apply mild pressure to the wound opening for 3–5 s (see Note 2).

4. On the test day, inject mice with SNC80 (10 mg/kg IP), ARM390 (10 mg/kg PO), or vehicle (see Note 3). To ensure that all animals are treated equally, all mice receive an intra- peritoneal (SNC80 or saline) and a per os (ARM390 or distilled H2O) injection. Immediately place them on the testing rack.

5. Mechanical responses for each mouse are determined 45 min later. As above, use mechanical von Frey filaments on the ipsi- lateral paw.

6. Mice may be euthanized immediately after testing or up to 24 h later. Animals may also be tested further, see below.

100 Amynah A. Pradhan et al.

3.3 Acute Behavioral Desensitization
and Chronic Tolerance

3.3.1 Analysis of Receptor Internalization State at
the Time of Behavioral Desensitization

3.3.2 Analysis
of Receptor Internalization Following Chronic Drug Treatment

3.4 Transcardial Perfusion and Tissue Preparation

1. 4–12 h after the first injection of vehicle, SNC80 or ARM390, mice are re-injected with the same drug, placed on the testing rack; and tested 45 min later.

2. Another cohort of animals are euthanized/perfused at the time that the tested group receives the second injection.

1. Mice are treated as on the test day outlined above (numbers 4 and 5), every day for 5 days.

2. Mice are euthanized/perfused 24 h after the final injection (day 6) (see Note 4).

1. Using standard protocol for transcardial perfusion, infuse: (a) 10 ml 0.1 M PBS.
(b) 30 ml 4 % formaldehyde.

2. Remove tissue of interest by dissection.

3. Postfix tissue in 4 % formaldehyde for 4 h at 4 °C.

4. Transfer to 30 % sucrose and store at 4 °C overnight.

5. Freeze tissue in OCT.

6. Cut sections on cryostat (40 μm thickness for standard immu- nohistochemistry) into PBS.

7. Transfer free floating sections into glycerol cryoprotectant and store at −20 °C until use.

1. Block tissue with 5 % NDST for 1 h at room temperature.

2. Prepare primary antibody solution by diluting in 1 % NDST buffer.

3. Remove blocking solution and apply primary antibody solu- tion (no washes). Incubate overnight at room temperature.

4. Wash 3×10 min with 1 % NDST buffer.

5. Meanwhile, prepare secondary antibody solution by diluting in 1 % NDST buffer.

6. Apply secondary antibody solution and incubate for 2 h at room temperature (see Note 5).

7. Wash 3×10 min with 0.1 M PB.

8. Mount sections on slides using a paintbrush.

9. Coverslip with Fluoromount G and allow slides to dry prior to imaging.

3.5 Immuno- histochemistry

Examining Receptor Internalization In Vivo 101

4 Notes

Acknowledgments

1. For pain testing, we have observed less variability when we test under low light conditions, preferably at the beginning of the light cycle.

2. Animals are subsequently tested 72 h post-CFA administra- tion, to ensure increased functionality of delta opioid receptors in vivo [11, 72].

3. Inrodents,theoriginalmanufacturerofARM390(AstraZeneca) found that the bioavailability of this compound was greatest when given per os. Intraperitoneal administration of ARM390 does not appear to have pain-relieving effects when tested by us or by others [73].

4. To ensure that all of the behavioral results are generalizable to wild-type mice, all behavioral experiments should also be per- formed in commercially available C57Bl/6 mice, which we have shown to have a similar dose response to SNC80 and ARM390 [30].

5. To avoid photobleaching of the fluorophore, it is best to per- form secondary antibody incubation and subsequent steps with as little exposure to light as possible.

AAP is supported by NIH grant DA031243, University of Illinois at Chicago Department of Psychiatry start-up funds. VLT is sup- ported by a FAER Research Fellowship Grant. GS is supported by NIH grant DA031777, Stanford University Department of Anesthesiology, Perioperative and Pain Medicine and Stanford Institute for Neuro-Innovation and Translational Neurosciences start-up funds.

References

1. Lagerstrom MC, Schioth HB (2008) Structural diversity of G protein-coupled receptors and significance for drug discovery. Nat Rev Drug Discov 7:339–357

2. von Zastrow M, Williams JT (2012) Modulating neuromodulation by receptor membrane traffic in the endocytic pathway. Neuron 76:22–32

3. Hanyaloglu AC, von Zastrow M (2008) Regulation of GPCRs by endocytic membrane trafficking and its potential implications. Annu Rev Pharmacol Toxicol 48:537–568

4. Kenakin T (2003) Ligand-selective receptor conformations revisited: the promise and the problem. Trends Pharmacol Sci 24:346–354

5. Kenakin TP (2012) Biased signalling and allosteric machines: new vistas and challenges for drug discovery. Br J Pharmacol 165: 1659–1669

6. Reiter E, Ahn S, Shukla AK et al (2012) Molecular mechanism of beta-arrestin-biased agonism at seven-transmembrane receptors. Annu Rev Pharmacol Toxicol 52:179–197

102 Amynah A. Pradhan et al.

7. Galandrin S, Oligny-Longpre G, Bouvier M (2007) The evasive nature of drug efficacy: implications for drug discovery. Trends Phar- macol Sci 28:423–430

8. Al HR, Bruchas MR (2011) Molecular mecha- nisms of opioid receptor-dependent signaling and behavior. Anesthesiology 115:1363–1381

9. Sauriyal DS, Jaggi AS, Singh N (2011) Exten- ding pharmacological spectrum of opioids beyond analgesia: multifunctional aspects in different pathophysiological states. Neuropep- tides 45:175–188

10. Pradhan AA, Befort K, Nozaki C et al (2011) The delta opioid receptor: an evolving target for the treatment of brain disorders. Trends Pharmacol Sci 32:581–590

11. Cahill CM, Morinville A, Hoffert C et al (2003) Up-regulation and trafficking of delta opioid receptor in a model of chronic inflam- mation: implications for pain control. Pain 101:199–208

12. Fraser GL, Gaudreau GA, Clarke PB et al (2000) Antihyperalgesic effects of delta opioid agonists in a rat model of chronic inflamma- tion. Br J Pharmacol 129:1668–1672

13. Gaveriaux-Ruff C, Karchewski LA, Hever X et al (2008) Inflammatory pain is enhanced in delta opioid receptor-knockout mice. Eur J Neurosci 27:2558–2567

14. Hurley RW, Hammond DL (2000) The anal- gesic effects of supraspinal mu and delta opioid receptor agonists are potentiated during persis- tent inflammation. J Neurosci 20:1249–1259

15. Nadal X, Banos JE, Kieffer BL et al (2006) Neuropathic pain is enhanced in delta-opioid receptor knockout mice. Eur J Neurosci 23: 830–834

16. Pradhan A, Smith M, McGuire B et al (2013) Chronic inflammatory injury results in increa- sed coupling of delta opioid receptors to voltage- gated Ca2+ channels. Mol Pain 9:8

17. Filliol D, Ghozland S, Chluba J et al (2000) Mice deficient for delta- and mu-opioid recep- tors exhibit opposing alterations of emotional responses. Nat Genet 25:195–200

18. Konig M, Zimmer AM, Steiner H et al (1996) Pain responses, anxiety and aggression in mice deficient in pre-proenkephalin. Nature 383: 535–538

19. Broom DC, Jutkiewicz EM, Folk JE et al (2002) Convulsant activity of a non-peptidic delta-opioid receptor agonist is not required for its antidepressant-like effects in Sprague- Dawley rats. Psychopharmacology 164:42–48

20. Perrine SA, Hoshaw BA, Unterwald EM (2006) Delta opioid receptor ligands modulate

anxiety-like behaviors in the rat. Br J Pharmacol

147:864–887
21. Saitoh A, Kimura Y, Suzuki T et al (2004)

Potential anxiolytic and antidepressant-like activities of SNC80, a selective delta-opioid agonist, in behavioral models in rodents. J Pharmacol Sci 95:374–380

22. Bernard V, Decossas M, Liste I (2006) Intraneuronal trafficking of G-protein-coupled receptors in vivo. Trends Neurosci 29:140–147

23. Keith DE, Anton B, Murray SR et al (1998) mu-Opioid receptor internalization: opiate drugs have differential effects on a conserved endocytic mechanism in vitro and in the mam- malian brain. Mol Pharmacol 53:377–384

24. Sternini C, Spann M, Anton B et al (1996) Agonist-selective endocytosis of mu opioid receptor by neurons in vivo. Proc Natl Acad Sci U S A 93:9241–9246

25. Van Bockstaele EJ, Commons KG (2001) Internalization of mu-opioid receptors pro- duced by etorphine in the rat locus coeruleus. Neuroscience 108(3):467–477

26. Scherrer G, Imamachi N, Cao YQ et al (2009) Dissociation of the opioid receptor mecha- nisms that control mechanical and heat pain. Cell 137:1148–1159

27. Scherrer G, Tryoen-Toth P, Filliol D et al (2006) Knockin mice expressing fluorescent delta-opioid receptors uncover G protein- coupled receptor dynamics in vivo. Proc Natl Acad Sci U S A 103:9691–9696

28. Bertran-Gonzalez J, Laurent V, Chieng BC et al (2013) Learning-related translocation of delta-opioid receptors on ventral striatal cholinergic interneurons mediates choice between goal-directed actions. J Neurosci 33: 16060–16071

29. Poole DP, Pelayo JC, Scherrer G et al (2011) Localization and regulation of fluorescently labeled delta opioid receptor, expressed in enteric neurons of mice. Gastroenterology 141:982–991

30. Pradhan AA, Becker JA, Scherrer G et al (2009) In vivo delta opioid receptor internalization controls behavioral effects of agonists. PLoS One 4(5):e5425

31. Pradhan AA, Walwyn W, Nozaki C et al (2010) Ligand-directed trafficking of the delta-opioid receptor in vivo: two paths toward analgesic tolerance. J Neurosci 30:16459–16468

32. Faget L, Erbs E, Le Merrer J et al (2012) In vivo visualization of delta opioid receptors upon physiological activation uncovers a dis- tinct internalization profile. J Neurosci 32: 7301–7310

33. Beaudry H, Proteau-Gagne A, Li S et al (2009) Differential noxious and motor tolerance of chronic delta opioid receptor agonists in rodents. Neuroscience 161:381–391

34. Kabli N, Cahill CM (2007) Anti-allodynic effects of peripheral delta opioid receptors in neuropathic pain. Pain 127:84–93

35. Cahill CM, Morinville A, Lee MC et al (2001) Prolonged morphine treatment targets delta opioid receptors to neuronal plasma mem- branes and enhances delta-mediated antinoci- ception. J Neurosci 21:7598–7607

36. Gendron L, Lucido AL, Mennicken F et al (2006) Morphine and pain-related stimuli enhance cell surface availability of somatic delta-opioid receptors in rat dorsal root gan- glia. J Neurosci 26:953–962

37. Morinville A, Cahill CM, Aibak H et al (2004) Morphine-induced changes in delta opioid receptor trafficking are linked to somatosen- sory processing in the rat spinal cord. J Neurosci 24:5549–5559

38. Morinville A, Cahill CM, Esdaile MJ et al (2003) Regulation of delta-opioid receptor trafficking via mu-opioid receptor stimulation: evidence from mu-opioid receptor knock-out mice. J Neurosci 23:4888–4898

39. van Rijn RM, Brissett DI, Whistler JL (2012) Emergence of functional spinal delta opioid receptors after chronic ethanol exposure. Biol Psychiatry 71:232–238

40. Qiu Y, Loh HH, Law PY (2007) Phosphor- ylation of the delta-opioid receptor regulates its beta-arrestins selectivity and subsequent recep- tor internalization and adenylyl cyclase desensi- tization. J Biol Chem 282:22315–22323

41. Whistler JL, Tsao P, von Zastrow M (2001) A phosphorylation-regulated brake mechanism controls the initial endocytosis of opioid recep- tors but is not required for post-endocytic sorting to lysosomes. J Biol Chem 276: 34331–34338

42. Zhang X, Wang F, Chen X et al (2008) Post- endocytic fates of delta-opioid receptor are regulated by GRK2-mediated receptor phos- phorylation and distinct beta-arrestin isoforms. J Neurochem 106:781–792

43. Leontiadis LJ, Papakonstantinou MP, Georgoussi Z (2009) Regulator of G protein signaling 4 confers selectivity to specific G proteins to mod- ulate mu- and delta-opioid receptor signaling. Cell signal 21:1218–1228

44. Stratinaki M, Varidaki A, Mitsi V et al (2013) Regulator of G protein signaling 4 [corrected] is a crucial modulator of antidepressant drug action in depression and neuropathic

pain models. Proc Natl Acad Sci U S A 110:

8254–8259
45. Wang Q, Liu-Chen LY, Traynor JR (2009)

Differential modulation of mu- and delta- opioid receptor agonists by endogenous RGS4 protein in SH-SY5Y cells. J Biol Chem 284: 18357–18367

46. Simonin F, Karcher P, Boeuf JJ (2004) Identification of a novel family of G protein- coupled receptor associated sorting proteins. J Neurochem 89:766–775

47. Whistler JL, Enquist J, Marley A et al (2002) Modulation of postendocytic sorting of G pro- tein-coupled receptors. Science 297: 615–620

48. Shaner NC, Steinbach PA, Tsien RY (2005) A guide to choosing fluorescent proteins. Nat Methods 2:905–909

49. Kallal L, Benovic JL (2002) Fluorescence micros- copy techniques for the study of G-protein- coupled receptor trafficking. Methods Enzymol 343:492–506

50. Scherrer G, Befort K, Contet C (2004) The delta agonists DPDPE and deltorphin II recruit predominantly mu receptors to produce ther- mal analgesia: a parallel study of mu, delta and combinatorial opioid receptor knockout mice. Eur J Neurosci 19:2239–2248

51. Eisinger DA, Schulz R (2005) Mechanism and consequences of delta-opioid receptor internal- ization. Cri Rev Neurobiol 17:1–26

52. Ko JL, Arvidsson U, Williams FG et al (1999) Visualization of time-dependent redistribution of delta-opioid receptors in neuronal cells dur- ing prolonged agonist exposure. Brain Res Mol Brain Res 69:171–185

53. Law PY, Erickson LJ, El Kouhen R (2000) Receptor density and recycling affect the rate of agonist-induced desensitization of mu-opioid receptor. Mol Pharmacol 58: 388–398

54. Lecoq I, Marie N, Jauzac P et al (2004) Different regulation of human delta-opioid receptors by SNC-80 [(+)-4-[(alphaR)-alpha- ((2S,5R)-4-allyl-2,5-dimethyl-1-piperazinyl)- 3-methoxybenz yl]-N, N-diethylbenzamide] and endogenous enkephalins. J Pharmacol Exp Ther 310:666–677

55. Afify EA, Law PY, Riedl M et al (1998) Role of carboxyl terminus of mu-and delta-opioid receptor in agonist-induced down-regulation. Brain Res Mol Brain Res 54:24–34

56. Trapaidze N, Keith DE, Cvejic S et al (1996) Sequestration of the delta opioid receptor. Role of the C terminus in agonist-mediated internal- ization. J Biol Chem 271:29279–29285

57. Goody RJ, Oakley SM, Filliol D et al (2002) Quantitative autoradiographic mapping of

Examining Receptor Internalization In Vivo 103

104 Amynah A. Pradhan et al.

opioid receptors in the brain of delta-opioid receptor gene knockout mice. Brain Res 945: 9–19

58. Mansour A, Khachaturian H, Lewis ME et al (1987) Autoradiographic differentiation of mu, delta, and kappa opioid receptors in the rat fore- brain and midbrain. J Neurosci 7:2445–2464

59. Mennicken F, Zhang J, Hoffert C et al (2003) Phylogenetic changes in the expression of delta opioid receptors in spinal cord and dorsal root ganglia. J Comp Neurol 465:349–360

60. Pradhan AA, Yu XH, Laird JM (2010) Modality of hyperalgesia tested, not type of nerve dam- age, predicts pharmacological sensitivity in rat models of neuropathic pain. Eur J Pain 14: 503–509

61. Basbaum AI, Bautista DM, Scherrer G et al (2009) Cellular and molecular mechanisms of pain. Cell 139:267–284

62. Minami M, Maekawa K, Yabuuchi K et al (1995) Double in situ hybridization study on coexis- tence of mu-, delta- and kappa-opioid receptor mRNAs with preprotachykinin A mRNA in the rat dorsal root ganglia. Brain Res Mol Brain Res 30:203–210

63. Bardoni R, Tawfik VL, Wang D et al (2014) Delta opioid receptors presynaptically regulate cutaneous mechanosensory neuron input to the spinal cord dorsal horn. Neuron 81(6):1312–1327, PMID: 24583022

64. Cahill CM, McClellan KA, Morinville A et al (2001) Immunohistochemical distribution of delta opioid receptors in the rat central nervous system: evidence for somatodendritic labeling and antigen-specific cellular compartmentaliza- tion. J Comp Neurol 440:65–84

65. Malmberg AB, Chen C,Tonegawa S et al (1997) Preserved acute pain and reduced neu- ropathic pain in mice lacking PKCgamma. Science 278:279–283

66. Pradhan AA, Smith ML, Kieffer BL et al (2012) Ligand-directed signalling within the

opioid receptor family. Br J Pharmacol 167:

960–969
67. Calderon SN, Rothman RB, Porreca F et al

(1994) Probes for narcotic receptor mediated phenomena. 19. Synthesis of (+)-4-[(alpha R)-alpha-((2S,5R)-4-allyl-2,5-dimethyl-1- piperazinyl)-3- methoxybenzyl]-N, N-diethyl- benzamide (SNC 80): a highly selective, nonpeptide delta opioid receptor agonist. J Med Chem 37:2125–2128

68. Wei ZY, Brown W, Takasaki B et al (2000) N, N-Diethyl-4-(phenylpiperidin-4-ylidene- methyl)benzamide: a novel, exceptionally selective, potent delta opioid receptor agonist with oral bioavailability and its analogues. J Med Chem 43:3895–3905

69. Aguila B, Coulbault L, Boulouard M et al (2007) In vitro and in vivo pharmacological profile of UFP-512, a novel selective delta- opioid receptor agonist; correlations between desensitization and tolerance. Br J Pharmacol 152:1312–1324

70. Codd EE, Carson JR, Colburn RW et al (2009) JNJ-20788560 [9-(8-azabicyclo[3.2.1]oct-3- ylidene)-9H-xanthene-3-carboxylic acid diethyl- amide], a selective delta opioid receptor agonist, is a potent and efficacious antihyperalgesic agent that does not produce respiratory depression, pharmacologic tolerance, or physical depen- dence. J Pharmacol Exp Ther 329: 241–251

71. Chaplan SR, Bach FW, Pogrel JW et al (1994) Quantitative assessment of tactile allodynia in the rat paw. J Neurosci Methods 53:55–63

72. Erbs E, Faget L, Scherrer G (2012) Distribution of delta opioid receptor-expressing neurons in the mouse hippocampus. Neuroscience 221:203–213

73. Negus SS, Rosenberg MB, Altarifi AA et al (2012) Effects of the delta opioid receptor agonist SNC80 on pain-related depression of intracranial self-stimulation (ICSS) in rats. J Pain 13:317–327

Chapter 8

Monitoring Opioid Receptor Dimerization in Living Cells by Bioluminescence Resonance Energy Transfer (BRET)

Monica Baiula Abstract

Bioluminescence resonance energy transfer (BRET) is a natural phenomenon that has been successfully applied for the study of protein–protein interactions, including opioid receptor oligomers. The discovery of opioid receptor homomers and heteromers has brought to the finding of new functions and new way of signaling and trafficking; therefore, opioid receptor oligomers may be considered as novel drug targets. Fusing receptors of interest with Renilla luciferase and with a fluorescent protein (such as EYFP), it is possible to study opioid receptor dimerization using BRET.

Key words BRET, Enhanced yellow fluorescent protein, Receptor dimerization, Renilla luciferase, Opioid receptor

1 Introduction

Biological processes proceed through a sequence of specific protein– protein interactions along intracellular signaling cascades. Charac- terization of these interactions is essential to the understanding of cellular mechanisms.

Bioluminescence resonance energy transfer (BRET) is a natural phenomenon that has been successfully applied to the identifica- tion and the characterization of protein–protein interactions in liv- ing cells and displays an important role in the discovery of novel drugs [1, 2]. Further development of BRET application has greatly improved knowledge on G-protein-coupled receptors (GPCRs) and especially on opioid receptor dimerization [3].

BRET was first described in marine organisms such as Renilla reniformis and Aequorea Victoria, in which an enzyme (Renilla lucif- erase and aequorin, respectively) catalyzes the oxidation of the endogenous substrate coelenterazine to coelenteramide resulting in bioluminescence [4, 5]; if endogenous green fluorescent protein (GFP) is in close proximity then part of energy deriving from

Santi M. Spampinato (ed.), Opioid Receptors: Methods and Protocols, Methods in Molecular Biology, vol. 1230, DOI 10.1007/978-1-4939-1708-2_8, © Springer Science+Business Media New York 2015

105

106 Monica Baiula

Fig. 1 BRET assay to monitor opioid receptor homo/heterodimerization. To study any possible interaction the first receptor is fused to Rluc, as energy donor, and the second to EYFP, as energy acceptor. (a) If the receptors do not interact (distance > 100 Å) there is no energy transfer from donor to acceptor and the only signal detect- able is from coelenterazine h oxidation. (b) After adding luciferase substrate, coelenterazine h, if the receptors are in close proximity (<100 Å) the energy deriving from substrate oxidation (480 nm) can be transferred to acceptor (EYFP) which re-emits light at a longer wavelength (535 nm)

bioluminescent reaction is transferred to GFP which emits light at a characteristic wavelength.

BRET is a nonradiative dipole–dipole energy transfer that occurs between a luminescent donor enzyme and an acceptor fluo- rescent protein when they are in close proximity, within 100 Å (Fig. 1).

The efficiency of energy transfer depends on several factors such as the distance between donor and acceptor molecule, relative dipole orientation and the need of overlap between emission spec- trum of the donor and the absorption spectrum of the acceptor.

To study receptor interactions using BRET, the proteins of interest are fused either to the donor Rluc or to an acceptor fluorescent protein, usually a variant of GFP. Depending on Rluc variant, substrate and acceptor fluorescent protein, several types

2 Materials

2.1 Generation of Fusion Proteins

of BRET have been developed: the most common techniques are BRET1 and BRET2 [6]. BRET1 uses Rluc as donor and GFP or enhanced yellow fluorescent protein (EYFP) as acceptor, while in BRET2 the acceptor is GFP2 and the substrate is DeepBlueC (coel- enterazine 400a), resulting in a better separation of donor and acceptor emission spectra, although BRET2 signal is weaker.

As coelenterazine is extremely instable in aqueous environ- ment, the use of a protected form of coelenterazine h, named EnduRen, provides stability and ensures real-time monitoring of BRET signal for several hours: “extended BRET” (eBRET) uses EnduRen substrate that penetrates and is cleaved to coelenterazine h by esterases only in living cells [7, 8].

BRET has been extensively used to demonstrate homo- and heterodimerization of GPCR [9] and in particular of opioid receptors [9, 10]. Over the last two decades, it has become clear that GPCRs do not act solely as monomer but can interact with them- selves or with other receptor types to form homomers and het- eromers. Therefore, it is possible that in a context-dependent way different effects mediated by a receptor originate from homomer/ heteromer-specific signaling cascade, as they display different signal- ing and novel pharmacological properties. Opioid homomers and heteromers have been among the first studied [11, 12]. All three opioid receptors can form homomers while heteromers have been observed between δ opioid receptor (DOR) and κ opioid receptor (KOR) or μ opioid receptor (MOR) [13]. Moreover, opioid recep- tor can form heteromers with other types of receptors: DOR can heterodimerize with nociceptin receptor [14] and can form oligo- mers with β2-adrenergic receptors [15]; in addition, MOR, DOR and KOR oligomerize with cannabinoid receptor CB1 [16]. Opioid receptor oligomers can be considered as novel drug targets: in fact, ligands targeting selectively oligomers could be therapeutically valu- able for the development of improved analgesic drugs with reduced tolerance and dependence.

1. Plasmid containing the coding sequence of Renilla reniformis luciferase (Rluc).

2. Plasmid containing the coding sequence of enhanced yellow fluorescent protein (EYFP).

3. cDNA of KOR.

4. cDNA of DOR.

5. Materials and reagents necessary for cloning and validation of created fusion constructs.

BRET Assay to Study Opioid Receptors Dimerization 107

108 Monica Baiula

2.2 Cell Culture and Transfection

1. HEK293 cell line.

2. Cell culture medium: Eagle’s minimum essential medium (EMEM) supplemented with 2 mM l-glutamine, 1 % nonessen- tial amino acids (NEAA), 10 % fetal bovine serum and 1 % antibiotic-antimycotic solution (containing 100 U/mL penicil- lin, 100 μg/mL streptomycin and 0.25 μg/mL amphotericin B).

3. Cell culture medium (composed as previously described) with- out phenol red.

4. 6-Well tissue culture plates. 5. 96-Well white plate.
6. Transfection reagent.

1. Trypsin/EDTA solution or PBS+5 mM EDTA.

2. Phosphate-buffered saline (PBS) pH 7.4, without calcium and magnesium.

3. Coelenterazine h: stock solution 500 μM in methanol, store at −80 °C; dilute to 5 μM in PBS immediately before the experiment. Coelenterazine h should be protected from light (see Note 1).

4. 96-Well white plate.

5. Plate reader able to measure luminescence (440–500) and fluo- rescence (excitation 400–480 nm and emission 520–540 nm).

1. As a first step in the set up of a BRET assay, it is necessary to subclone proteins of interest together with donor (Rluc) or acceptor (EYFP) proteins. Through DNA recombinant tech- niques, KOR and DOR are genetically fused with Rluc or EYFP in frame in a way that the stop codon between the two coding sequence should be removed, generating pRluc-KOR and pEYFP-DOR constructs (see Note 2).

It is also important to include negative and positive con- trols for BRET assay. For negative control plasmid containing Rluc alone can be used; to demonstrate that donor and accep- tor molecules do not interact if they are not fused with pro- teins of interest it could be useful to co-express EYFP not fused to another protein with donor fusion construct. Otherwise, a fusion construct could be generated using a similar protein known not to interact with protein of interest (such as for example muscarinic receptor M2, that does not interact with opioid receptor) [10].

2.3 Evaluation of Fusion Proteins Expression
and Detection
of BRET Signal

3 Methods

3.1 Generation of Fusion Proteins for Monitoring KOR and DOR Heterodimerization

3.2 Cell Culture and Transfection

2. For positive control it is possible to use donor–acceptor fusion construct composed by Rluc fused in frame with EYFP (pRluc- EYFP); this construct yields a high and robust BRET signal easily distinguishable from background.

3. The sequences of all constructs should be confirmed by DNA sequencing.

4. After generation of fusion constructs it is necessary to validate them through functionality assay: for opioid receptor it is pos- sible to perform binding assay or signaling assay to compare the behavior of fused receptor to that of its native form. Moreover, correct localization and trafficking should also be assessed.

1. Plate HEK293 cells in a 6-well tissue culture plate in cell cul- ture medium: to have cells ready for transfection after 24 h seed about 300,000 HEK293 cells/well.

2. Incubate 24 h in a humidified incubator at 37 °C, 5 % CO2.

3. Transfect transiently HEK293 cells with fusion proteins (pRluc-KOR and pEYFP-DOR) using an appropriate transfec- tion reagent according to the manufacturer’s instructions (see Note 3).

4. Transfect transiently HEK293 cells with positive (pRluc-EYFP) and negative (Rluc alone or pM2-EYFP) controls.

5. After 48 h it is possible to evaluate the relative expression of fusion constructs measuring separately luminescence and fluorescence.

6. To measure BRET on adherent cells, they can be detached after 24 h from transfection (with trypsin/EDTA solution or PBS + 5 mM EDTA) and re-plated in a 96-well white plate cell culture treated at a concentration of 50,000–100,000 cells/ well. Allow attachment of cells for additional 24 h in a humidi- fied incubator at 37 °C, 5 % CO2.

1. To validate luciferase functionality, just before the experiment dilute coelenterazine h in PBS without calcium and magne- sium (see Note 4) to a final concentration of 5 μM; protect the substrate from light (see Note 1).

2. Remove cell culture medium and add gently luciferase sub- strate avoiding cell detachment (see Note 5).

3. Read luminescence at 485 nm.

4. Measure fluorescence exciting at 485 nm and registering EYFP emission at 535 nm.

3.3 Expression of Fusion Proteins

BRET Assay to Study Opioid Receptors Dimerization 109

110 Monica Baiula

3.4 BRET Signal 1. Prepare fresh coelenterazine h dilution (5 μM) (see Note 1).

Measurement

2. Add luciferase substrate freshly diluted to the cells after remov- ing cell culture medium (be careful not to detach cells).

3. Read immediately BRET signal measuring light from each well for 1 s using filters at 485 nm (Rluc signal) and 535 nm (BRET signal) (see Note 6).

4. Calculate BRET ratio: first calculate the ratio between fluores- cence of EYFP at 535 nm and luminescence of Rluc at 485 nm for cells transfected with Rluc alone (background signal: BRETbackground); calculate the ratio between fluorescence of EYFP at 535 nm and luminescence of Rluc at 485 nm for cells transfected with both fusion constructs (Rluc-KOR and EYFP- DOR; BRETRluc-KOR- +EYFP-DOR). Then, subtract the ratios as follows:

BRET ratio=[(fluorescence of EYFP at 535 nm)/(lumines- cence of Rluc at 485 nm)] Rluc-KOR- +EYFP-DOR−[(fluorescence of EYFP at 535 nm)/(luminescence of Rluc at 485 nm)]background. In Fig. 2, representative results for BRET assay evaluating het- erodimerization of KOR and DOR opioid receptors are shown.

Fig. 2 Detection of KOR-DOR heterodimerization in HEK293 cells using BRET assay. Transfected cells with pRluc-KOR alone or with pRluc-KOR + pEYFP-M2 represent negative controls, while pRluc-EYFP is a positive control. The BRET ratio has been calculated subtracting the ratio obtained in cells transfected with pRluc-KOR alone from the ratio generated in cells co-expressing both fusion constructs (pRluc-KOR + pEYFP-DOR)

4 Notes

5. Other than the evaluation of constitutive protein–protein interaction, it is possible to analyze ligand-induced opioid receptor dimerization using BRET assay. BRET signal is mea- sured before and after ligand treatment [17, 18].

6. To calculate BRET ratio of ligand-induced BRET signal: BRET ratio ligand-induced=BRET ratio (ligand-treated cells)– BRET ratio (vehicle-treated cells).

1. In alternative to coelenterazine h, other substrates have been developed for BRET measurements in live cells [7, 8]. ViviRenTM and EnduRenTM (both from Promega) are protected forms of coelenterazine h with increased half-life in live cells and with low autoluminescence. If it is necessary to get a long- lasting signal EnduRenTM generates a stable signal for up to 24 h; alternatively, ViviRenTM ensures a very bright lumines- cent signal but that is short-lived. ViviRenTM: stock solution 60 mM in DMSO (stored at −80 °C for up to 1 year); dilute to a final concentration of 60 μM in cell culture medium without phenol red. EnduRenTM: stock solution 60 mM in DMSO, vortexing and if it is necessary heating to complete dissolution; dilute just prior to use to final concentration of 60 μM in cell culture medium without phenol red.

2. Determining the optimal orientation (N- or C-terminus) and nature (Rluc or EYFP) of fused recombinant proteins is neces- sary to generate several plasmids for each protein of interest. In fact, the activity and/or the expression of recombinant protein could be influenced by both orientation and composition (Table 1).
It should be also considered that when studying interac- tions between receptors or with intracellular proteins donor or acceptor need to be fused at C-terminus, which is cytosolic. This strategy increases also the possibility of correct receptor folding, membrane localization, and ligand binding. Addi- tionally, a linker (usually 5–10 amino acids) can be inserted between fused proteins in order to optimize relative orienta- tion that can influence BRET signal.

3. It is necessary to define empirically the optimal ratio of fusion proteins to transfect: usually to get most favorable energy transfer it could be useful to transfect a higher amount of acceptor than donor. To avoid nonspecific BRET signal, it is suggested not to overexpress fusion constructs. Nevertheless, it is recommended to perform a matrix transfection using dif- ferent amounts of both plasmids to determinate the optimal protein expression levels.

BRET Assay to Study Opioid Receptors Dimerization 111

112 Monica Baiula Table 1

Combinations of fusion construct pairs to optimize BRET assay

N-terminal Rluc fused protein

C-terminal Rluc fused protein

N-terminal EYFP fused protein

C-terminal EYFP fused protein

Protein A

Protein B

Protein A

Protein B

Protein A

Protein B

Protein A

Protein B

Protein A

Protein B

Protein A

Protein B

Protein A

Protein B

Protein A

References

4. It has been shown that divalent cations such as calcium and magnesium induce a decrease in BRET signal [19]. In addi- tion, a shift in BRET signal is expected also using medium containing phenol red probably due to changes in propagation of light in the medium [20]. For this reason, it is recommended to use as BRET assay buffer PBS without calcium and magne- sium or cell culture medium phenol red free.

5. Measurements of luminescence, fluorescence, and BRET signal can be done also on detached cells. Using trypsin/EDTA solu- tion or PBS+5 mM EDTA detach the cells from the plate and resuspend in PBS containing 0.1 % glucose, to increase detached cells viability. Then to perform measurements put 50,000–100,000 cells/well in a 96-well white plate and con- tinue as described for adherent cells.

6. If EnduRenTM is employed as luciferase substrate BRET mea- surement is done at least 90 min after substrate addition, while for coelenterazine h it is necessary to read immediately.
targets for drug development. Prog Mol Biol

Transl Sci 91:41–52
Hart RC, Stempel KE, Boyer PD et al (1978) Mechanism of the enzyme-catalyzed biolumi- nescent oxidation of coelenteratetype luciferin. Biochem Biophys Res Commun 81:980–986 Wilson T, Hastings JW (1998) Bioluminescence. Annu Rev Cell Dev Biol 14:197–230

Ayoub MA, Pfleger KD (2010) Recent advances in bioluminescence resonance energy transfer

1. Bacart J, Corbel C, Jockers R et al (2008) The BRET technology and its application to screen- ing assays. Biotechnol J 3:311–324

2. Lohse MJ, Nuber S, Hoffmann C (2012) Fluorescence/bioluminescence resonance energy transfer techniques to study G-protein-coupled receptor activation and signaling. Pharmacol Rev
64:299–336

3. Ferré S, Navarro G, Casadó V et al (2010) G protein-coupled receptor heteromers as new

5. 6.

technologies to study GPCR heteromerization.

Curr Opin Pharmacol 10:44–52

7. Pfleger KD, Dromey JR, Dalrymple MB et al
(2006) Extended bioluminescence resonance energy transfer (eBRET) for monitoring pro- longed protein–protein interactions in live cells. Cell Signal 18:1664–1670

8. Pfleger KD (2009) Analysis of protein–protein interactions using bioluminescence resonance energy transfer. Methods Mol Biol 574: 173–183

9. Pfleger KD, Eidne KA (2003) New technolo- gies: bioluminescence resonance energy trans- fer (BRET) for the detection of real time interactions involving G-protein coupled recep- tors. Pituitary 6:141–151

10. Wang D, Sun X, Bohn LM et al (2005) Opioid receptor homo- and heterodimerization in living cells by quantitative bioluminescence resonance energy transfer. Mol Pharmacol 67: 2173–2184

11. Gomes I, Filipovska J, Jordan BA et al (2002) Oligomerization of opioid receptors. Methods 27:358–365

12. George SR, Fan T, Xie Z et al (2000) Oligomerization of mu- and delta-opioid recep- tors. Generation of novel functional properties. J Biol Chem 275:26128–26135

13. van Rijn RM, Whistler JL, Waldhoer M (2010) Opioid-receptor-heteromer-specific trafficking and pharmacology. Curr Opin Pharmacol 10: 73–79

14. Pan YX, Bolan E, Pasternak GW (2002) Dimerization of morphine and orphanin FQ/ nociceptin receptors: generation of a novel

opioid receptor subtype. Biochem Biophys Res Commun 297:659–663

15. McVey M, Ramsay D, Kellett E et al (2001) Monitoring receptor oligomerization using time resolved fluorescence resonance energy transfer and bioluminescence resonance energy transfer. The human delta opioid receptor displays constitutive oligomerization at the cell surface, which is not regulated by receptor occupancy. J Biol Chem 276:14092–14099

16. Rios C, Gomes I, Devi LA (2006) mu opioid and CB1 cannabinoid receptor interactions: reciprocal inhibition of receptor signaling and neuritogenesis. Br J Pharmacol 148:387–395

17. Kocan M, See HB, Seeber RM et al (2008) Demonstration of improvements to the biolu- minescence resonance energy transfer (BRET) technology for the monitoring of G protein- coupled receptors in live cells. J Biomol Screen 13:888–898

18. Li Y, Chen J, Bai B et al (2012) Hetero- dimerization of human apelin and kappa opioid receptors: roles in signal transduction. Cell Signal 24:991–1001

19. Li F, Yu J, Zhang Z et al (2012) Buffer enhanced bioluminescence resonance energy transfer sensor based on Gaussia luciferase for in vitro detection of protease. Anal Chim Acta 724:104–110

20. Couturier C, Deprez B (2012) Setting up a bioluminescence resonance energy transfer high throughput screening assay to search for protein/protein interaction inhibitors in mammalian cells. Front Endocrinol (Lausanne) 3:1–13

BRET Assay to Study Opioid Receptors Dimerization 113

Chapter 9

Bioluminescence Resonance Energy Transfer (BRET) to Detect the Interactions Between Kappa Opioid Receptor and Non visual Arrestins

Andrea Bedini Abstract

Bioluminescence resonance energy transfer (BRET) is a very sensitive technique employed to study protein–protein interactions, including G-protein-coupled receptors (GPCRs) hetero- and homo-dimerization. Recently, BRET has also been used to investigate the interaction between GPCRs (e.g., β2 adrenergic receptor, muscarinic M2 receptor, dopaminergic D2 receptor) and non-visual arrestins. Here a BRET protocol is described to investigate interactions between the kappa opioid receptor (KOR) and non visual arrestins (arrestin-2 and arrestin-3) in HEK-293 cells, both under basal conditions and after expo- sure to KOR ligands.

Key words BRET, Arrestins, Kappa opioid receptor, Ligand directed signaling, Renilla Luciferase

1 Introduction

Bioluminescence resonance energy transfer (BRET) is a natural phenomenon known in several marine organisms that relies on the non-radiative transfer of energy from an appropriate energy donor to an energy acceptor, provided that both are located at a distance less than 10 nm [1]: by hydrolyzing a specific substrate the donor protein produces a luminescent signal that may excite the acceptor protein (if the two proteins are close enough) and the acceptor pro- tein, in turn, emits a fluorescent signal at a specific wavelength.

BRET was first applied for the detection of protein–protein interactions (PPIs) in 1999 [2]. When one of the two proteins whose interaction is under investigation is fused to a certain donor and the other protein is fused to the corresponding acceptor, a specific BRET signal will be detectable if there is an interaction between these two. The fundamental step in order to successfully employ BRET analysis to investigate PPIs is to generate recombinant fusion

Santi M. Spampinato (ed.), Opioid Receptors: Methods and Protocols, Methods in Molecular Biology, vol. 1230, DOI 10.1007/978-1-4939-1708-2_9, © Springer Science+Business Media New York 2015

115

116 Andrea Bedini

constructs between the proteins of interest and either a donor or an acceptor; the choice of the proper donor/acceptor couple is another crucial aspect of BRET employment in the study of PPIs.

Since the first use of BRET to study PPIs, this technology has been constantly evolved by combining new donor and acceptor couples and developing various BRET assay formats [3].

Recently, many BRET assays have been developed to study the oligomerization of seven-transmembrane-spanning G-protein- coupled receptors (GPCRs), as the BRET technology provides an attractive way to study this phenomenon in intact cells without the need to solubilize receptors from their natural membrane environ- ment [4].

Moving from these considerations, BRET has been also been adapted to investigate the interactions between GPCRs (e.g., β2 adrenergic receptor, muscarinic M2 receptor, dopaminergic D2 receptor) and non visual arrestins [5] and to characterize how such interactions are affected by selective receptor ligands (e.g., isopro- terenol for β2 adrenergic receptor).

Investigating the recruitment of non visual arrestins (arrestin-2 and arrestin-3) at a GPCR provides unique insight into GPCR pharmacology: arrestins, in fact, albeit initially found to mainly control receptor desensitization, internalization and recycling, are now well known as multifunctional scaffolds able to interact with kinases and signal transducers, thereby triggering a plethora of spe- cific signaling events.

GPCRs, in fact, are dynamic entities that exist in multiple con- formations: diverse ligands may stabilize different receptor active states, thus resulting in different receptor–effector complexes and diverse levels of activation or inhibition of signaling cascades [6].

Some ligands may therefore activate all the pathways down- stream of a GPCR, whereas others may elicit distinct conformational changes within the receptor, promoting the interaction with a specific set of signal transducers and resulting in a selective modula- tion of intracellular signaling [7]: therefore, a GPCR activated by a ligand may result in the activation of both G-protein and arrestin- mediated signaling cascades, leading to a full response, whereas some ligands may preferentially activate a biased G-protein-mediated signal transduction cascade and not an arrestin-dependent one.

This phenomenon is defined as “ligand-directed signaling” and represents an intriguing opportunity in pharmacological research, as specific transduction pathways may produce a thera- peutic response, whereas others may controbute to the adverse effects of a drug.

Ligand-directed signaling has been clearly demonstrated at kappa opioid receptor (KOR): KOR, in fact, may couple to many different signaling cascades, being tuned toward specific signaling states by different ligands [8]. KOR classic agonists elicit analgesic responses, but their arrestin-dependent activation of p38 MAPK

BRET Analysis of KOR/Non Visual Arrestins Interactions 117

has been shown to contribute to adverse effects such as dysphoria, therefore limiting the utility of KOR classic agonists as analgesics; for these reasons, KOR partial agonists devoid of arrestin recruitment and arrestin-dependent activation of p38 MAPK could represent useful and innovative drugs as they might retain sufficient analgesic activity for the treatment of pain-related disorders without causing arrestin-dependent side effects such as dysphoria [8]. This is just one example, among many possible ones, of the relevant impact that ligand-directed signaling at KOR may have on opioid pharmacology and highlights the need for reli- able methods to investigate ligand-specific effects on KOR/arres- tin interactions.

In this chapter a sensitive and reproducible BRET assay aimed to investigate the interaction between non visual arrestins (arrestin-2 and arrestin-3) and KOR in HEK-293, under both basal condi- tions and upon KOR ligands administration, is described. This experimental approach could be used to investigate any effect of well-known KOR ligands on receptor/arrestin interactions as well as to screen libraries of novel compounds in a high-throughput format for their ability to modulate arrestin recruitment at KOR.

The BRET assay described in this chapter is adapted from the studies performed by Gurevich and co-workers on β2 adrenergic receptor/arrestin interactions and employs Renilla Luciferase (RLuc) as donor and Venus as acceptor; in this assay, RLuc is fused with the C-terminus of KOR and Venus is fused with the N-terminus of either arrestin-2 or arrestin-3 (see Note 1). Both the fusion proteins are expressed in a selected cell model (HEK-293 cells; see Note 2). Upon the administration of its specific substrate (colenterazine h), the RLuc fused to KOR triggers substrate hydro- lysis, thus generating a luminsecent signal; if the Venus/arrestin fusion protein is within 10 nm of the KOR/RLuc (i.e., if there is an interaction between the two), the light produced by coelentera- zine h hydrolysis will excite the Venus tag fused to the arrestin, thereby producing a second emission of light at a higher wave- length (Fig. 1). By detecting the emissions at the wavelengths typi- cal of coelenterazine h hydrolysis and Venus emission, it is possible to evaluate whether interactions occurred between donor (KOR/ RLuc) and acceptor (Venus/arrestin). Performing the above- described procedure in the presence or absence of KOR ligands makes it possible to evaluate any ligand-specific effect on KOR/ arrestin interactions.

A general outline of the whole methodology described in this chapter is reported below: HEK-293 cells are transfected with a fixed amount of donor plasmid (the one encoding for the KOR/ RLuc chimera) and increasing amounts of acceptor plasmid (the one encoding for the Venus/arrestin-2 or Venus/arrestin-3 fusion protein); this allows for a concentration–response curve of the BRET signal (pcDNA3 empty vector or similar plasmids will be

118 Andrea Bedini

Fig. 1 Representative cartoon of the BRET format presented in this chapter. (a) RLuc fused to KOR hydrolyses coelenterazine h to celenteramide thus producing a luminous signal at a specific wavelength; if there is no inter- action between KOR/RLuc and Venus/arrestin, the Venus tag fused to the arrestin cannot be excited by the light emitted upon colenterazine h hydrolysis. (b) RLuc fused to KOR hydrolyses coelenterazine h to celenteramide thus producing a luminous signal at a specific wavelength; if KOR/RLuc and Venus/arrestin are closer than 10 nm (i.e., there is interaction between the two), the light produced upon colenterazine h hydrolysis can be transferred to the Venus tag fused to the arrestin, which in turn is excited and emits at its specific wavelength

2 Materials

2.1 Cell Transfection (Day 1)

used as “junk” DNA to compensate for the different amount of DNA to be transfected). Twenty-four hours after transfection, HEK-293 cells are plated into 96-well plates with white walls (to measure BRET signal and basal luminescence) or black walls (to detect basal fluorescence); after further 24 h, transfected cells are incubated in warm DPBS (containing 1 g/l glucose and 25 mM HEPES), in the presence or absence of the KOR ligand of interest, and the Rluc-specific substrate (colenterazine h) is added. BRET signal, basal luminescence, and basal fluorescence are then measured at 37 °C in a luminometer/fluorimeter properly set up.

1. Donor plasmid: kappa opioid receptor in pcDNA3 fused at C-terminus with RLuc (see Note 1).

2. Acceptor plasmid: arrestin-2 fused at N-terminus with Venus or arrestin-3 fused at N-terminus with Venus (see Note 1).

3. “Junk” DNA: pcDNA3.1 empty vector.
4. OPTI-MEM® (Life Technologies).
5. Lipofectamine® 2000 (Life Technologies) (see Note 3).

6. HEK-293 cells (see Note 2) at 60–80 % confluence (one 6-well plate per each combination of donor and acceptor plasmid to be transfected).

7. HEK-293 complete growth medium: DMEM/F12 1:1, 10 % FBS, 1 % L-Glutamine, 1 % Pen/Strep.

8. Six 1.5 ml tubes per each combination of donor and acceptor plasmid to be transfected.

9. One 15 ml tube.

2.2 Plating Transfected Cells into 96-Well Plates (Day 2)

1. Two 96-well plates with white walls (Nunc cat. #: 136101) per each combination of donor and acceptor plasmid transfected on day 1 (Fig. 2a, b).

2. One 96-well plate with black walls (Nunc cat. #: 137101) per each combination of donor and acceptor plasmid transfected on day 1 (Fig. 2c).

3. HEK-293 complete growth medium: DMEM/F12 1:1, 10 % FBS, 1 % L-Glutamine, 1 % Pen/Strep.

BRET Analysis of KOR/Non Visual Arrestins Interactions 119

Fig. 2 Plating scheme for the BRET format presented in this chapter. (a) Plating scheme for the 96-well plate with white walls destined to measure the BRET signal. Per each of the transfection groups prepared as indicated in day 1 (numbered from #1 to #6) three columns of four wells each are plated as represented in the plate scheme; raws A, B, E, and F are treated with the KOR ligand of interest, whereas raws C, D, G, and H are treated with vehicle. (b) Plating scheme for the 96-well plate with white walls destined to measure the basal lumines- cence. Per each of the transfection groups prepared as indicated in day 1 (numbered from #1 to #6) one column of four wells is plated as represented in the plate scheme. (c) Plating scheme for the 96-well plate with black walls destined to measure the basal fluorescence. Per each of the transfection groups prepared as indicated in day 1 (numbered from #1 to #6), one column of four wells is plated as represented in the plate scheme

120 Andrea Bedini

2.3 CellStimulation with Opioid Ligands or Vehicle and BRET Signal Measurement (Day 3)

3 Methods

3.1 Cell Transfection (Day 1)

1. DPBS+Glucose+HEPES: 1× DPBS, 1 g/l glucose, 0.1 g/l magnesium chloride, 0.1 g/l calcium chloride, 25 mM HEPES; pH 7.3 (see Note 4).

2. Stock solutions of KOR ligands of interest (e.g., U50,488 or norBNI).

3. Coelenterazine h (DiscoverX Cat. #: 90-0084) (see Note 5).

4. Plate reader with luminometer reading protocols with filters at 460 and 542 nm and fluorimeter reading protocols with exci- tation filter at 485 nm and emission filter at 535 nm.

Work under a sterile hood for cell culture and prepare a set of six 1.5 ml tubes per each combination of BRET plasmids (donor plas- mid and acceptor plasmid) to be transfected; employ one six-well plate containing HEK-293 cells at 60–80 % of confluence per each combination of BRET plasmids to be transfected. Be careful in preparing the various dilutions of plasmids and in very accurately pipetting the right volumes.

1. Label the six 1.5 ml tubes from #1 to #6.

2. Calculate the volumes corresponding to each of the following plasmid amounts:

– 200 ng of donor plasmid (see Note 6).

– 24 μg of acceptor plasmid (see Note 7).

– 0, 6, 9, 10.5, 11.25, and 12 μg of “junk” DNA (see Note 8).

3. Add 600 μl of OPTIMEM to each 1.5 ml tube.

4. Add to tube #6 a volume of acceptor plasmid corresponding to
24 μg (see Note 7).

5. Add to tube #6 a volume of OPTIMEM so that the final volume
is 1,200 μl.

6. Mix gently tube #6 by inverting and tapping it with finger,
then transfer 600 μl to tube #5.

7. Mix gently tube #5 by inverting and tapping it with finger,
then transfer 600 μl to tube #4.

8. Mix gently tube #4 by inverting and tapping it with finger,
then transfer 600 μl to tube #3.

9. Mix gently tube #3 by inverting and tapping it with finger,
then transfer 600 μl to tube #2.

10. Mix gently tube #2 by inverting and tapping it with finger, then aspire and discard 600 μl.

3.2 Plating Transfected Cells into 96-Well Plates (Day 2)

11. Add to each 1.5 ml tube a volume corresponding to the right amount of “junk” DNA, to compensate for different amounts of acceptor plasmid, as follows (see Note 8):

– 0 μg to tube #6.

– 6 μg to tube #5.

– 9 μg to tube #4.

– 10.5 μg to tube #3.

– 11.25 μg to tube #2.

– 12 μg to tube #1.

12. Add to each 1.5 ml tube a volume corresponding to the right amount of donor plasmid (200 ng) and mix the tubes gently (see Note 6).

13. In a 15 ml tube dilute Lipofectamine® 2000 in OPTIMEM as follows: 23 μl of Lipofectamine® 2000 in 3.6 ml of OPTIMEM (see Note 9).

14. Incubate the Lipofectamine® 2000/OPTIMEM mixture for 5 min under the hood.

15. Add 600 μl of Lipofectamine® 2000/OPTIMEM mixture to each of the six 1.5 ml tubes previously prepared with the dif- ferent plasmids, mix gently but thoroughly and incubate under the hood for 20 min.

16. Suck off the cell culture medium from the 6-well plate and pour or gently pipet the contents of every tube in each of the six wells of the plate containing the cells to be transfected, being very careful not to detach the cells to be transfected.

17. Place the plate in a humidified incubator for 5 h (37 °C; 5 % of CO2).

18. At the end of the incubation, suck off the transfection mixture from each well and replace it with complete growth medium (2 ml per well).

Work under a sterile hood for cell culture; a general plating scheme is reported in Fig. 2. Detach the cells (by pipetting them off in their medium, see Note 10) and centrifuge them at 300×g for 4 min; discard the supernatant and resuspend each cell pellet in 2.7 ml of complete growth medium. For each transfection condi- tion transfer 100 μl of each cell suspension into one well of a 96-well plates, as below indicated (Fig. 2).

1. Into the first 96-well plate with white walls: plate three col- umns of four wells each per every transfection group prepared on day 1 (Fig. 2a). This plate will be used to measure the BRET signal of each sample.

BRET Analysis of KOR/Non Visual Arrestins Interactions 121

122

Andrea Bedini

3.3
with Opioid Ligands or Vehicle and BRET Signal Measurement (Day 3)

2. Into the second 96-well plate with white walls: plate one column of four wells per each transfection group prepared on day 1 (Fig. 2b). This plate will be used to measure the basal donor signal (RLuc luminescence) of each sample.

3. Into the 96-well plate with black walls: plate one column of four wells per each of the transfection groups prepared on day 1 (Fig. 2c). This plate will be used to measure the basal accep- tor signal (Venus fluorescence) of each sample.

4. Incubate the 96-well plates in a humidified incubator (37 °C; 5 % of CO2) overnight.

1. Pre-warm DPBS + Glucose + HEPES (see Note 4) at 37 °C (30 ml of DPBS should be sufficient for each six-well plate transfected on day 1).

2. Dilute the KOR ligand of interest (e.g., U50.488) in warm DPBS+Glucose+HEPES so that the ligand concentration is fourfold higher than the tested concetration (e.g., to test a 10 μM ligand does make the stock in DPBS at 40 μM).

3. Take out of the incubator the 96-well plate with white walls destined to BRET measurement (Fig. 2a) and discard the growth medium.

4. Add 100 μl of warm DPBS+Glucose+HEPES solution to each well (see Note 11).

5. Add 50 μl of warm DPBS+Glucose+HEPES to each wells of raws C, D, G, H (these will be the vehicle-treated wells; Fig. 2a) (see Note 11).

6. Turn the plate reader on and set it at 37 °C.

7.Add 50 μl of the ligand of interest diluted in DPBS+Glucose+HEPES (as prepared in step 2) to each well of raws A, B, E, F (these will be the ligand-treated wells; Fig. 2a) (see Note 11) and note down the time (see Note 12).

8. Dilute 27.3 μl of 2.45 mM coelenterazine h in 4 ml of warm DPBS + Glucose + HEPES (see Notes 12 and 13) (this way a working solution of 16.7 μM coelenterazine h is obtained).

9. Add 50 μl of 16.7 μM coelenterazine h to each well (see Note 12).

10. Read the plate at 37 °C at 5 min after agonist administration by employing a luminometer counting protocol with filters at 460 and 542 nm (1 s per well) (see Note 12).

11. Repeat the reading at 15, 25, 35, and 45 min after agonist administration.

12. Take out of the incubator the 96-well plate with white walls destined to the measurement of basal luminescence (Fig. 2b),

Cell Stimulation

3.4 Data Analysis

remove the growth medium and add 150 μl of warm DPBS + Glucose + HEPES to each well.

13. Prepare 1.3 ml of 16.7 μM colenterazine h diluted in DPBS + Glucose + HEPES (similarly to step 8) and add 50 μl of it at each well (see Note 13).

14. Read the plate at 37 °C employing a luminometer counting protocol with filter at 460 nm (1 s per well).

15. Take out of the incubator the 96-well plate with black walls destined to the measurement of basal fluorescence (Fig. 2c), remove the growth medium and add 200 μl of warm DPBS + Glucose + HEPES to each well.

16. Read the plate at 37 °C employing a fluorimeter counting pro- tocol with excitation filter set at 485 nm and emission filter set at 535 nm (1 s per well).

Perform data analysis as follows (representative data obtained with the BRET assay described in this chapter is reported in Fig. 3):

1. Calculate the ratio between the basal fluorescence and the basal luminescence values obtained per each well read in steps 14 and 16, respectively: this way four F/L values will be obtained per each of the six transfection groups prepared on day 1.

2. Calculate the F/L mean value for each of the six transfection groups.

3. Per each of the wells read as indicated in step 10, calculate the ratio between the higher wavelength signal (542 nm) and the

BRET Analysis of KOR/Non Visual Arrestins Interactions 123

Fig. 3 Interaction between KOR and arrestin (pilot experiment carried out according to the protocol presented in this chapter). (a) BRET ratio graph of both vehicle-treated and ligand-treated HEK-293 cells, built as indicated in Subheading 3.4, step 4. (b) Net BRET graph showing the effect of the KOR ligand of interest on the KOR/arrestin interaction, built as indicated in Subheading 3.4, step 7

124 Andrea Bedini

lower one (460 nm); this way, BRET ratio values will be calculated for each well. Then, within each of the six transfection groups, determine the mean of the six BRET ratio values of ligand- treated wells and the mean of the six BRET ratio values of vehicle-treated wells. These calculations are to be performed per each of the time points of interest (see step 11 of Subheading 3.3).

4. Build an XY graph with on the X axis (use a log10 scale) the F/L values calculated in step 2 of Subheading 3.4, and on the Y axes the corresponding BRET ratio values (as calculated in step 3 of Subheading 3.4) (Fig. 3a): this way two BRET ratio curves will be obtained, one for the vehicle-treated wells and one for the ligand-treated wells (Fig. 3a). These curves will allow the experiment to determine if opioid receptor and arres- tin interacted in the absence of the ligand of interest and if the administered ligand influenced this interaction or lack there of (see Note 14).

5. Perform the statistical analysis of the obtained BRET ratio curves by interpolating them with a sigmoidal dose–response curve model; the obtained R2 values will be informative of the goodness of this interpolation (>0.9).

6. Subtract the mean BRET ratio value of vehicle-treated samples to their corresponding ligand-treated mean BRET ratio value and propagate the error accordingly (see Note 15): this way Net BRET values will be obtained, representing the net effect of the ligand administered on the interaction between KOR and arrestin.

7. Plot the Net BRET values obtained as in step 6 of Subheading 3.4 in an XY graph having on the X axes (this time use a linear scale) the F/L values (as calculated in step 2 of this section), and on the Y axes the Net BRET values deter- mined in step 6 of this section (Fig. 3b); this way a Net BRET curve will be obtained. This curve will allow to quantify the effect of the ligand of interest on the opioid receptor/arrestin interaction (see Note 16).

8. Perform the statistical analysis of the obtained Net BRET curve by interpolating it with a one site binding (hyperbola) curve model; the R2 value will be informative of the goodness of this interpolation (>0.9).

9. Determine the Net BRET Max value from the curve obtained as indicated in step 7 of this section (this value corresponds to the BMax in the one site binding curve model); the Net BRET Max value is a quantitative evaluation of the ligand of interest’s ability to influence the interaction between KOR and arrestin, allowing comparison between different experimental replicates of the same treatment or between different treatments.

4 Notes

BRET Analysis of KOR/Non Visual Arrestins Interactions 125

1. A relevant issue in generating chimeric constructs for BRET is whether to fuse the donor/acceptor with the N-terminus or C-terminus of the protein of interest: one of the preliminary procedures in setting up a BRET assay, in fact, is to find out the best configuration for the chimeric protein to be employed. GPCRs are usually fused to RLuc at their C-terminus, whereas arrestin-2 and arrestin-3 with Venus fused at their N-terminus have been successfully employed in BRET assays [9]. This kind of configuration for KOR/RLuc and Venus/arrestin chimeras should be effective in most of BRET assays.

2. HEK-293 cells have been selected for the BRET assay presented in this chapter as they are a model of human cells that can be easily cultured and transfected with the various plasmids required for the described technique; furthermore, they express neither endogenous opioid receptors nor other related GPCRs that can interfere with the BRET assay. Other cell models could be used instead of HEK-293 (e.g., COS-7); however, it is important that alternative cell models are easily transfected and are similar to HEK-293 cells in terms of GPCR expression. If using cells different than HEK-293 it could be necessary to tailor both plasmid amounts and transfection conditions.

3. The transfection step is carried out employing Lipofectamine® 2000. This reagent is guaranteed by the manufacturer to yield a very high transfection efficiency, that is a prerequisite for the BRET assay described in this chapter. Other reagents yielding high transfection efficiency could be used according to their manufacturer’s recommendations.

4. Calcium and magnesium divalent cations are crucial for cell adhesion, therefore it is important that the DPBS solution employed in this BRET assay contains the correct amount of both. Different DPBS formulations are available, both with and without magnesium chloride and calcium chloride: check the formulation of the selected DPBS and adjust for calcium chloride and magnesium chloride concentration if necessary.

5. The BRET protocol described in this chapter has been set up using the coelenterazine h substrate provided by DiscoverX; other providers or types of substrate suitable for RLuc could be used according to their manufacturer’s recommendations.

6. Transfecting 200 ng/well of KOR/RLuc (donor) plasmid should yield expression levels of the corresponding protein sufficient to successfully perform the BRET assay described in this chapter (the optimal amount of donor plasmid in the proposed BRET format should range from 50 to 250 ng); in

126 Andrea Bedini

case of weak luminescence signal it could be useful to adjust the amount of donor plasmid to be transfected. If this is the case, prepare a six-well plate of HEK-293 cells and transfect each well with various amount of donor plasmid (include also an untransfected well); 24 h after transfection plate the cells into a 96-well plate with white walls (as indicated in day 2) and go on with the detection of basal luminescence after fur- ther 24 h (as indicated in day 3). This will allow the experi- menter to determine the optimal concentration of donor plasmid to be transfected to obtain KOR/RLuc expression levels that are significantly higher than those detected in untransfected cells.

7. The transfection of 24 μg of Venus/arrestin-3 encoding plas- mid may sometimes yield to expression levels of the corre- sponding chimeric protein lower than those obtained by transfecting the same amount of Venus/arrestin-2 encoding plasmid. To overcome this issue, it could be useful to increase the amount of Venus/arrestin-3 encoding plasmid to be trans- fected (e.g., use 48 μg instead of 24 μg).

8. In case of transfecting increased amounts of Venus/arrestin-3, remember to adjust the quantities of “junk” DNA accordingly (i.e., they should also be doubled).

9. In case of transfecting increased amounts of Venus/arrestin-3, remember to adjust the quantity of Lipofectamine® 2000 accordingly (i.e., it should also be doubled). In case of cell toxicity due to the increased amount of the transfection reagent employed, reduce the amount of Lipofectamine® 2000 reagent to 35–40 μl.

10. HEK-293 cells grow weakly adherent to the bottom of tissue culture plates, therefore they can be easily detached by pipet- ting up and down the medium contained in the plate; alterna- tively, a cell scraper can be used to gently detach the cells. If other cell types are used in the BRET format presented in this chapter, they have to be detached according to their standard culture procedures.

11. To minimize well-to-well variability as well as pipetting errors, it is recommended to use a well-calibrated multichannel pipette. Add the solutions by carefully pipetting them on the walls of the wells in order not to detach the transfected cells.

12. In the BRET format described in this chapter, the timing is crucial to precisely evaluate any effect elicited by the KOR ligand of interest on KOR/arrestin interactions; therefore it is important to estimate in advance the amount of time required for substrate dilution and administration, so the plate readings occur at the desired time points after ligand administration.

Acknowledgements

13. Coelenterazine h is light-sensitive: dilute the stock solution immediately before using it and keep it protected from direct light sources; diluted coelenterazine h solution may be added to the wells by using a multichannel pipette if direct illumina- tion of the tank containing the solution is prevented.

14. In case of interaction between KOR and arrestin, the data will result in sigmoidal curves on the BRET ratio graph built as indicated in Subheading 3.4, step 4 (Fig. 3a); if there is no interaction between KOR and arrestin, the data will result in a flat line parallel to X axes should appear. If the KOR ligand of interest favored the interaction between KOR and arrestin, its sigmoidal curve should be shifted leftward (Fig. 3a).

15. Alternatively to the indications reported in Subheading 3.4, step 6, it is possible to calculate the Net BRET value for each of the wells read as indicated in Subheading 3.3, step 9 as follows: calculate the ratio between the higher wavelength signal (542 nm) and the lower one (460 nm); this way, BRET ratio values will be calculated for each well. After that, subtract the BRET ratio values of vehicle-treated wells to those of the ligand- treated wells: this way six Net BRET values will be obtained per each of the transfection groups prepared on day 1. Then, calcu- late the Net BRET mean values for every transfection group and proceed with Subheading 3.4, step 7 and subsequent ones.

16. If the KOR ligand of interest promoted the interaction between KOR and arrestin the data will result in a hyperbolic saturation curve for the Net BRET graph built as indicated in Subheading 3.4, step 7 (Fig. 3b), otherwise either no curve or a flat line parallel to the X axis should be obtained.

The author would like to acknowledge Prof. V. Gurevich for pro- viding the plasmids backbones and for the precious advice in set- ting up the protocol presented in this chapter; authors would like also to acknowledge Prof. C. Chavkin, Dr. Selena S. Shattauer and Mrs. Jamie R. Kuhar for their practical support with KOR sub- cloning into the donor plasmid and for their critical review of the experimental procedures presented in this chapter.

References

1. Jockers R (2014) Comment on the use of BRET to study receptor-protein interactions. Front Endocrinol. doi:10.3389/fendo.2014.00003

2. Xu Y, Piston DW, Johnson CH (1999) A biolu- minescence resonance energy transfer (BRET) system: application to interacting circadian clock proteins. Proc Natl Acad Sci U S A 96:151–156

3. De A, Jasani A, Arora R et al (2013) Evolution of BRET biosensors from live cell to tissue-scale in vivo imaging. Front Endocrinol. doi:10.3389/ fendo.2013.00131

4. Drinovec L, Kubale V, Nøhr LJ et al (2012) Mathematical models for quantitative assess- ment of bioluminescence resonance energy

BRET Analysis of KOR/Non Visual Arrestins Interactions 127

128 Andrea Bedini

transfer: application to seven transmembrane receptors oligomerization. Front Endocrinol. doi:10.3389/fendo.2012.00104

5. Gimenez LE, Kook S, Vishnivetskiy SA et al (2012) Role of receptor-attached phosphates in binding of visual and non-visual arrestins to G protein-coupled receptors. J Biol Chem 287: 9028–9040

6. Pradhan AA, Smith ML, Kieffer BL et al (2012) Ligand-directed signalling within the opioid receptor family. Br J Pharmacol 167:960–969

7. Kenakin T, Christopoulos A (2013) Signalling bias in new drug discovery: detection, quantifi- cation and therapeutic impact. Nat Rev Drug Discov 12:205–221

8. Bruchas MR, Chavkin C (2010) Kinase cascades and ligand-directed signaling at the kappa opioid receptor. Psychopharmacology 210:137–147

9. Gimenez LE, Vishnivetskiy SA, Gurevich VV (2014) Targeting individual GPCRs with rede- signed nonvisual arrestins. Handb Exp Pharmacol 219:153–170

Chapter 10

Identification and Verification of Proteins Interacting with the Kappa Opioid Receptor (KOPR)

Chongguang Chen, Peng Huang, and Lee-Yuan Liu-Chen Abstract

Proteins that interact with the human kappa opioid receptor (hKOPR) may contribute to regulation and signaling of the receptor. In this paper, we focus on the protein 14-3-3zeta that regulates anterograde trans- port of the hKOPR from the endoplasmic reticulum (ER) to the Golgi apparatus. 14-3-3zeta interacts with the C-terminal domain of the receptor and promotes cell surface expression of the hKOPR by inhibiting coatomer protein I (COPI) and RVR motif-mediated ER retension of the hKOPR. Here we describe three experimental procedures we used to evaluate the interaction between hKOPR and 14-3-3zeta: co-immuno- precipitation, pull-down assay and immunofluorescence microscopy.

Key words Co-immunoprecipitation, Endoplasmic reticulum, Fluorescence microscopy, Immuno- blots, Kappa opioid receptor, Pull-down assay

1 Introduction

Here, we describe the strategy and protocols developed or adopted in our laboratory for identification and verification of proteins that interact with the human kappa opioid receptor (hKOPR). We then focus on proteins that play important roles in the anterograde transport of the hKOPR from the ER to Golgi apparatus [1–5]. One of the proteins, 14-3-3zeta, is presented as an example in the description of experimental procedures and in the figures. 14-3- 3zeta, a 28 kDa protein, is one of the seven 14-3-3 proteins and belongs to a family of conserved regulatory molecules in eukary- otic cells. We have demonstrated that siRNA knockdown of 14-3-3 zeta reduces cell surface expression of the hKOPR, which is reversed by overexpression of 14-3-3zeta. R354A/S358A substi- tutions in the putative 14-3-3 interaction motif R354QSTS358 in the hKOPR C-tail reduces interaction of the hKOPR with 14-3-3zeta and abolishes the effect of 14-3-3zeta knockdown on hKOPR

Santi M. Spampinato (ed.), Opioid Receptors: Methods and Protocols, Methods in Molecular Biology, vol. 1230, DOI 10.1007/978-1-4939-1708-2_10, © Springer Science+Business Media New York 2015

129

130 Chongguang Chen et al.

expression. Mutation of the ER retention motif R359VR361 adjacent to the 14-3-3 interaction motif in the hKOPR C-tail decreases interaction of coatomer protein I (COPI) with the hKOPR and abolishes 14-3-3zeta-mediated regulation of hKOPR expression. 14-3-3zeta knockdown increases association of COPI with the hKOPR, leading to retension of the hKOPR in the ER. Thus, 14-3-3zeta interaction with R354QSTS358 in the hKOPR C-tail enhances cell surface expression of the hKOPR by suppressing COPI interaction with the ER retension motif R359VR361 [1].

Three methods are described: co-immunoprecipitation used for both discovery proteomics and single protein verification, pull- down assay for determination of direct interaction, and fluores- cence microscopy for examination of co-localization. Functional studies are not included because they are beyond the scope of this chapter.

Co-immunoprecipitation or co-immunopurification technique is based on the assumption that protein–protein interaction in live cell compartments can be preserved after cells are lysed in a solu- bilization buffer. Since proteins remain associated with one another in the lysis solution, an antibody against the protein is used as “bait” to purify the target protein-interacting protein complexes. This is accomplished by using the antibody immobi- lized to agarose beads or using the antibody followed by Pansorbin for co-immunoprecipitation or using antibody-agarose packed in a column for co-immunopurification. The proteins co-eluted with the target protein are subsequently resolved by SDS-PAGE and identified by mass spectrometry and/or immunoblotting. Although it is a widely used technique for proteomics studies and for pro- tein–protein interactions, there are limitations and pitfalls inherent to co-immunoprecipitation. After solubilization, proteins origi- nally localized in different cellular compartments are brought to the same pool and thus give rise to physiologically irrelevant pro- tein complexes. In addition, weak or dynamic protein–protein interactions may be lost or reduced because of low affinity or dis- ruption of cytosolic environments. Moreover, strong binding to insoluble cellular matrix, e.g., microtubules, renders immunopre- cipitated proteins difficult to detect. It is crucial to carry out addi- tional experiments for verification of the initial results, such as repeating the co-immunoprecipitation in small scales for the pro- teins with rigorous controls, performing pull-down assay to test for direct interaction and immunofluorescence to examine co-localiza- tion in cell organelles.

Analysis of Proteins Interacting with the Human KOPR 131

2 Materials

2.1 Co-immuno- precipitation

2.1.1 Cells Stably Expressing FLAG-hKOPR and Its Mutants (N2A-FLAG-hKOPR)

2.1.2 Buffers
and Materials
for Immunoprecipitation

2.1.3 Materials
and Buffers for SDS-PAGE and Immunoblotting

All cells are cultured in an incubator at 37 °C and 5 % CO2 in humidified atmosphere in the indicated medium. All the buffers and solutions used are made in Milli-Q water.

Mouse Neuro2A (N2A) neuroblastoma cells are transfected with hKORP tagged at the N-terminus with FLAG (FLAG-hKOPR) in pcDNA3 plasmid (Invitrogen, neomycin-resistant) using Lipofectamine 2000 (Invitrogen) and selected in 0.5 mg G418/ ml in MEM (for N2A cell) media supplemented with 10 % FBS. Clonal cells stably expressing FLAG-hKOPR at a level of ~2 pmol/mg protein were used for experiments.

1. Cell Lysis buffer: 50 mM Tris adjusted with HCl to pH 7.4, 150 mM NaCl, 2 % Triton X-100 (Sigma-Aldrich), 5 mM EDTA and protease inhibitor cocktail tablets (Roche, 1 tablet/ 10 ml) (TTSEC).

2. Wash buffer: 25 mM Tris–HCl, 150 mM NaCl, pH 7.4, 1 % Triton X-100.

3. Elution buffer: 100 μM FLAG peptide (Sigma-Aldrich) in 0.1 % Triton X-100, 200 mM NaCl, 20 mM Tris–HCl, pH 7.4.

4. Anti-FLAG M2 affinity gel (Sigma-Aldrich).

1. Materials: Anti-FLAG (rabbit polyclonal antibody, F7425, Sigma-Aldrich); Anti-14-3-3zeta and anti-HA (rabbit poly- clonal antibody, Santa Cruz, CA); Anti-HA (mouse monoclonal antibody, Covance, Denver, PA); HRP-conjugated goat anti- rabbit IgG or anti-mouse IgG (pre-absorbed, Jackson ImmunoTechnology); Immobilon PVDF transfer membrane (Millipore); SuperSignal West Pico chemiluminescent reagent (ECL) (Pierce); GelCode Blue protein staining reagent (Pierce); ProteomIQ Blue protein staining reagent (Proteom System).

2. SDS-PAGE and immunoblotting solutions: Tricine-SDS- PAGE was used throughout the study [6]. See Huang et al. (in this book) for compositions of the cathode, anode and gel buffers and preparation of stacking and separation gels.

3. 2× Sample loading buffer: 4 % SDS, 100 mM Tris–HCl, pH 7.0, 0.03 % bromophenol blue, 20 % glycerol and 100 mM DTT (add fresh).

4. 5× Sample loading buffer: 10 % SDS, 250 mM Tris–HCl, pH 7.0, 0.07 % bromophenol blue, 50 % glycerol and 250 mM DTT (add fresh).

5. Transfer buffer: 25 mM Tris base, 200 mM Glycine and 20 % methanol.

132

Chongguang Chen et al.

2.2

Pull-Down Assay

6. Wash buffer (TBS-T): 20 mM Tris adjusted with HCl to pH 7.4, 150 mM NaCl and 0.1 % Tween 20 (Sigma-Aldrich).

7. Blocking buffer: 5 % non-fat milk powder or 5 % BSA in wash buffer.

1. Materials to generate Glutathione S-Transferase (GST)-KCL and 14-3-3 zeta: From GE Healthcare, pGEX-4 T-1, Glutathione Sepharose 4B, IPTG, E. coli BL21-CodonPlus (DE3)-RP. From Novagen, Thrombin Cleavage/Capture Kit, BugBuster Master Mix.

2. Binding buffer: PBS containing 1 mg/ml BSA, 0.1 % Triton X100, 1 mM DTT (added immediately before use).

3. Ponseau S membrane staining solution: 0.1 % Ponseau S in 5 % acetic acid.

N2A cells stably expressing 3XHA-hKOP was established as described earlier.

1. Coverslips (12CIR-1D, Fisher Sci.) were pre-coated with poly- D-lysine (Sigma) by soaking in sterile 0.01 % poly-D-lysine in water following by drying in a Laminar flow hood.

2. Blocking buffer, 1× PBS containing 2 % normal goat serum and 0.1 % Triton X100, pH 7.4.

3. Primary antibodies, mouse anti-HA (monoclonal, Covance); rabbit anti-14-3-3zeta (polyclonal, Santa Cruz).

4. Secondary antibodies (Molecular Probes, Invitrogen), Alexa 488-goat anti-mouse IgG; Texas Red-goat anti-rabbit IgG.

End-to-end mixer, Vari Mix (Thermolyne); Beckman Ultra- centrifuge; Beckman High Speed centrifuge; Bench top micro centrifuge (Eppendorf); Refrigerated bench top microcentrifuge (SFR13K, Savant); Electrophoresis apparatus (Hoffer Scientific Inc.); Electro-transfer apparatus (Bio-Rad); Cell culture incubator, Isotemp (Fisher Sci.); FUJIFILM LAS 1000 Imaging system; Microscope, Nikon TE300 fluorescence microscope with a 40×, 60× objective lens and a Magnifier digital camera; NIH Image and Adobe Photoshop.

The experimental procedures have been reproduced from Chen et al. [2–5] and Li et al. [1] with modifications (with permission from The American Society for Biochemistry and Molecular Biology).

2.3 Immuno- fluorescence

2.4 Instruments and Software

3 Methods

3.1 Co-immuno- precipitation
for Identification
of Interacting Proteins by Proteomics

All procedures are carried out at 4 °C unless otherwise indicated. Centrifuge the gel beads at no greater than 500 × g in all steps.

1. Subculture N2A-FLAG-hKOPR and null N2A cells at 80 % confluence at 1:3 ratio into ten 100-mm plates each. After 48 h, when cells grew to ~80 % density, remove medium, wash twice with PBS buffer, add 1 ml TTSEC/plate directly to cells and lyse for 5 min. Swirl plates to facilitate solubilization of cells (see Note 1).

2. Lift the plates one end higher for 1 min to let crude lysate flow to the lower end completely, collect the lysate using a 1 ml pipetter into a 15 ml screw cap tube, put the tubes on an end- to-end mixer and mix for 30 min to complete solubilization.

3. Transfer the lysate to centrifuge tubes (12 ml) for Beckman SW41 rotor, centrifuge at 100,000×g for 30 min. Pass the supernatants through a 0.22-μm syringe filter and into a fresh 15-ml tube (see Note 2).

4. Washing of M2-gel beads: Pipet 25 μl anti-FLAG M2 gel beads (50 μl suspensions) into a 1.5-ml Eppendorf tube. Wash three times with TTSEC by centrifugation and re-suspension. Use care to avoid damaging the gel beads (see Note 3).

5. Suspend washed gel beads with 1 ml supernatant, transfer the mixture to the bulk supernatant in 15 ml tube. Place the tube on an end-to-end mixer set at low speed, incubate for 4 h to overnight to immunoprecipitate FLAG-hKOPR (see Note 4).

6. Spin down the gel beads for 2 min, decant the supernatant, transfer the beads with 1 ml wash buffer to a fresh 1.5-ml Eppendorf tube. Wash the gel beads by placing the tubes on the end-to-end mixer and mixing for 5 min. Spin and aspirate the wash buffer with a capillary tip. Remove as much wash buffer as possible. Repeat the washing for three more times (see Note 5).

7. Suspend with 0.5 ml wash buffer and transfer the gel slurry to a 0.22 μm, 0.5 ml centrifugal filter (Ultrafree-MC, Millipore), centrifuge for 30 s to remove the buffer. Pipet 25 μl elution buffer (containing 100 μM FLAG peptide) to the gel pellet, tap the filter tube to help mix thoroughly. Place the filter in a 37 °C oven and incubate for 15 min. Tap the filter tube every few minutes. Centrifuge for 30 s to collect the elution solu- tion. Repeat the elution one more time and combine the eluate which should be measured ~50 μl. Mix 12.5 μl 5× SDS-PAGE sample loading buffer with the eluate so that the ratio of SDS:Triton X100 should be no less than 20:1 (w:w). Let the mixed samples sit at room temperature for 1 h (see Note 6).

8. Centrifuge the samples at maximum speed in a bench top cen- trifuge (~12.000×g) and load the supernatants onto 10 %

Analysis of Proteins Interacting with the Human KOPR 133

134 Chongguang Chen et al.

Fig. 1 (a) SDS-PAGE separation of FLAG-hKOPR-associated proteins. N2A-FLAG-hKOPR cells and untrans- fected N2A cells (control) were solubilized and FLAG-hKOPR was immunoprecipitated with anti-FLAG antibod- ies as described in Subheading 3. The immunoprecipitated complex was subjected to 10 % SDS-PAGE separation. The gel was stained with ProteomIQ Blue. Twenty-nine gel slices were collected for mass spec- trometry analysis. Five 14-3-3 isoforms were identified in bands 8 and 9. (b) LC-MS identification of 14-3-3 zeta peptides generated from tryptic digestion. The five 14-3-3 isomers were scored from 56 to 189 (>30), of which 14-3-3 zeta scored highest and had the highest degree of confidence in identification. The m/z ratios observed were 454.5611 (M + 2H)2+, 427.0818 (M + 3H)3+ and 774.9987 (M + 2H)2+. The underlined numbers indicate the queries matched with 14-3-3 zeta sequences. (Reproduced from Li et al. [1] with modifications, with permission from The American Society for Biochemistry and Molecular Biology)

3.2 Verification

SDS-PAGE gel. Set a 5 min pre-run at 40 V, then increase to 100 V and run for ~3 h or until the frontline of bromophenol blue reach the lower edge of the gel slab.

9. Stain the gel with GelCode blue reagent. Document the gel image and excise the gel bands for mass spectrometry analysis (Fig. 1).

All procedures are carried out at 4 °C unless otherwise indicated. Centrifuge the gel beads at no greater than 500 × g in all steps.

3.2.1 Small-scale Co-immunoprecipitation

Once a protein is chosen from the list of identified interacting protein candidates, immunoblot can be applied so that co- immunoprecipitation can be scaled down to save time and costs. Small formats also allow performing an array of tests which are either impractical or impossible in the discovery stage. If antibodies are not available for the chosen protein, the protein can be tagged with popular epitopes such as FLAG or HA and expressed by trans- fection into cultured cells.

1. Subculture N2A-FLAG-hKOPR or -hKOPR-AQSTA mutant (interacting site) and null N2A cells (control) into one 100- mm plates each. After 48 h, wash cells twice with PBS and dissolve in TTSEC at 1 ml/plate.

2. If epitope tagged protein is to be used (e.g., HA-14-3-3zeta, served as an additional control), transfection should be performed 24 h after subculture when cells grow to ~70 % confluence. Replace the complete medium with 5 ml Opti-MEM medium and let the cell conditioned for 1 h in the incubator. For a 100-mm plate, we typically use a ratio of DNA:Lipofect amine = 10 μg:30 μl. Dilute DNA and Lipofectamine sepa- rately into 1 ml Opti-MEM medium each, mix and incubate at room temperature for 20 min, add the mixture dropwise to the plate. Swirl the plate gently every 1 h or so to disperse the DNA/Lipofectamine complex. Replace the Opti-MEM with 10 ml complete medium after 4 h or overnight. Cells will be ready 24 h after transfection. Treat the cells as in step 1 (see Note 7).

3. Transfer the cell lysates to 1.5 ml Eppendorf tubes, centrifuge at the maximum speed on a bench top microcentrifuge (13,000–20,000 × g) at 4 °C for 10 min. Filter supernatants with 0.22 μm centrifugal filter. Mix the cleared supernatants with 10 μl pre-washed M2-gel in 1.5 ml tubes. Incubate on an end-to-end mixer for 2 h or longer.

4. Wash the gel beads with wash buffer four times by centrifuga- tion and re-suspension.

5. Pipet 10–20 μl 2× sample loading buffer to the gel beads, tap the tube to mix, incubate for 10 min at room temperature. Centrifuge for 1 min (see Note 8).

6. Immediately load the sample onto 10 % SDS-PAGE and elec- trophoresed at settings described earlier.

7. Electro-transfer separated protein bands to PVDF membrane at100 V for 1 h. Wash the membrane with blocking buffer for 10 min. Make blotting solution with rabbit anti-14-3-3zeta and/or rabbit anti-HA in blocking buffer at ratio of 1:1,000– 1:5,000. Incubate the membrane in the blotting solution on shaker overnight (see Note 9).

Analysis of Proteins Interacting with the Human KOPR 135

136

Chongguang Chen et al.

3.3

Pull-Down Assay

8. Wash the membrane with wash buffer (TBS-T) three times, 10 min each time. Blot the membrane with secondary anti- body HRP-goat anti-rabbit IgG at 1:5,000 in wash buffer for 1 h at room temperature.

9. Wash three times with wash buffer, 5 min each time.

10. Incubate in ECL for 5 min. Place the membrane in-between two pieces of transparency sheets, lay on paper towel and press to remove air bubbles and extra reagents.

11. Capture the gel image with an Imaging system or expose to X-ray film.

12. Strip the membrane with stripping buffer and re-blot with rab- bit anti-FLAG and/or anti-GAPDH.

13. Repeat steps 8–11 (see Fig. 2).

Two GST fusion proteins were generated for the studies: GST-14- 3-3zeta and GST-KCL. 14-3-3zeta and C-terminal domain of the hKOPR-(334–380) were cloned into the pGEX-4T-1 bacterial expression system, which contains the GST sequence 5′ to the inserted sequence, and transformed into E. coli BL21-CodonPlus (DE3)-RP to express. The E. coli cells overexpressing GST fusion proteins or GST were solubilized in BugBuster protein extraction reagent at room temperature for 30 min according to the Novagen manual, and then adsorbed onto glutathione-Sepharose 4B beads (15 min at room temperature) and washed three times with phosphate-buffered saline. The beads loaded with GST and GST- KCL were ready for pull-down analysis, and the amounts of loaded proteins (~1 μg/μl) were semiquantified by use of Coomassie Blue staining (SDS-gels) and Ponceau S staining (PVDF membranes), respectively. 14-3-3zeta was further cleaved off from GST at the thrombin site in 1× thrombin cleavage buffer (20 mM Tris–HCl pH 8.4, 150 mM NaCl, 2.5 mM CaCl2) by biotinylated thrombin, which was removed with Streptavidin Agarose later on, according to the manufacture’s protocol.

To test if there is a direct interaction between two proteins, pure proteins will be mixed one to one. We use recombinant proteins 14-3-3zeta and GST-KCL. GST protein is used as control.

1. Pipet 10 μl GST-KCL or GST beads to 300 μl binding buffer containing 3 μg 14-3-3zeta protein in 1.5 ml tube.

2. Incubate on an end-to-end mixer overnight.

3. Wash with binding buffer four times, 5 min each time.

4. Dissociate the bound proteins by adding 20 μl of 2× sample load- ing buffer and incubating for 10 min at room temperature.

5. Perform the SDS-PAGE and immunoblotting for 14-3-3zeta as in steps 6–11 of Subheading 3.2.1.

3.3.1
and Purification
of Recombinant Proteins

cDNA Construction

3.3.2 GST Pull-Down

Analysis of Proteins Interacting with the Human KOPR 137

Fig. 2 14-3-3 zeta co-immunoprecipitated with the hKOPR. (a) Endogenous 14-3-3 zeta: N2A-FLAG-hKOPR or blank N2A cells were solubilized and immunoprecipitated with agarose beads conjugated with M2 anti-FLAG antibody. (b) Transfected 14-3-3 zeta: N2A-FLAG-hKOPR cells in a 100-mm plate were transiently transfected with 10 μg HA-14-3-3 zeta cDNA or the vector (control) with Lipofectamine 2000. Approximately 2 days later, cells were solubilized and immunoprecipitated with M2 anti-FLAG antibody agarose beads. (a, b). Immunoprecipitated materials were then separated with SDS-PAGE, transferred and immunoblotted with antibodies as indicated. Each immunoblot represents one of three independent experiments performed with similar results. (Reproduced from Li et al. [1] with modifications, with permission from The American Society for Biochemistry and Molecular Biology)

3.4 Immuno- fluorescence

6. Wash the membrane with water for 5 min. Stain the membrane with Ponseau S for GST and GST-KCL.

7. Capture an image of the membrane with an imaging system (see Fig. 3).

Finally, whether the pair of interacting proteins co-localizes in live cell compartment needs to be determined.

1. Subculture N2A-3xHA-hKOPR cells on coverslips placed in a 12-well plate (2×105 cells/well). Incubate for 24 h (see Note 10).

138 Chongguang Chen et al.

Fig. 3 Direct interaction between the hKOR C-tail and 14-3-3 zeta. Purified 14-3-3 zeta was incubated with glutathione-Sepharose 4B beads pre-loaded with GST and GST-hKOPR C-tail overnight at 4 °C. The beads were washed extensively, and the bound proteins were eluted from the beads, resolved by 8 % SDS-PAGE, and transferred onto ImmobilonTM-P PVDF membranes. A 1/250 supernatant was also loaded as an input control. Upper panel, 14-3-3 was detected by a rabbit anti-14-3-3 zeta antibody. Lower panel, the same mem- brane was stained with Ponceau S, showing the relative sizes and the amounts of the GST and GST-hKOPR C-tail loaded. The figure represents one of the three experiments with similar results. (Reproduced from Li et al. [1] with modifica- tions, with permission from The American Society for Biochemistry and Molecular Biology)

2. Place a fresh 12-well plate filled with methanol (2 ml/well) in −20 °C freezer for 30 min or longer. Keep the plate in freezer (see Note 11).

3. Wash the cells clean of medium with PBS. Use a sharp tweezers to pick up the coverslips, dip in the cold methanol. Keep the plate in freezer for 10 min.

4. Take the plate to bench, remove methanol by aspiration, immediately wash once with ice-cold PBS, add blocking buffer containing mouse anti-HA at 1:2,000 and rabbit anti-14-3- 3zeta at 1:62.5 (0.3 ml/well). Incubate overnight at 4 °C.

5. Wash three times with PBS, 5 min each time. Add blocking buffer containing Alexa Fluor 488-goat anti-mouse IgG (green) and Texas Red-goat anti-rabbit IgG (red) (both 1:1,000, 0.3 ml/well). Shield from ambient light from this point forward. Incubate on an end-to-end mixer for 30 min at room temperature.

6. Wash three times with PBS, 5 min each time. Mount the coverslips to glass slides with mounting solution.

7. Examine the cells under a fluorescence microscope and capture the images (see Fig. 4).

Analysis of Proteins Interacting with the Human KOPR 139

Fig. 4 Co-localization of hKOPR and 14-3-3 zeta in Neuro2A cells. N2A-3HA-hKOPR cells cultured on cover slips were fixed with methanol and immunofluorescence was performed as described in the text. 3HA-hKOPR was detected with mouse anti-HA antibody (1:2,000) followed by Alexa Fluor 488-goat anti-mouse IgG (green), whereas 14-3-3 zeta was revealed by rabbit anti-14-3-3 zeta antibody (1:62.5) followed by Texas Red-goat anti-rabbit IgG (red). Images were taken at 40× objectives using Nikon Eclipse TE300 fluorescence microscope mounted with Optronics MagnaFire digital camera. The experiment was performed three times with similar results. Scale bar=20 μm. (Reproduced from Li et al. [1] with modifications, with permission from The American Society for Biochemistry and Molecular Biology)

4 Notes

1. If the lysates appear too viscous, increase the volume of lysis buffer by 1/10 at a time. Do not sonicate the cell lysate because it breaks down nucleus and cell matrix and complicate co-IP outcomes.

2. Light and fluffy particulates are often noticeable on top of the supernatant which should be cleared by filtration.

3. Make sure to suspend the gel beads thoroughly. Use a 200 μl tip cut at end with a sharp razor so the tip has a ~1-mm pore when pipette the gels.

4. If precipitates are observed after overnight incubation, use shorter time. It is critical to have a crystal clear supernatant in this step.

5. Avoid losing gel beads in this step while removing wash buffer completely. Use a capillary tip and squeeze the tip with a flat tweezers.

6. SDS to Triton X100 (or any detergent) ratio is important for efficient SDS binding to protein which, in turn, is vital for good separation on SDS-PAGE. This is especially an issue for partially purified membrane proteins such as hKOPR since detergents have to be included in all the steps until the samples are ready for SDS-PAGE, and detergent exchange is often not an option given the final sample size (~50 μl). So detergent “depletion” is our choice. In our experience, using fourfold or greater micellar numbers of SDS over Triton X100 is sufficient. Their CMC values should be used as reference to calculate the SDS concentration.

140 Chongguang Chen et al.

Acknowledgments

References

In this case, SDS/Triton X100(%/%)=4×CMCSDS/CMCTriton X100=4×0.075/0.014=~20. Since 0.1 % Triton X100 is used, so 2 % or greater of SDS is required to deplete or replace the Triton X100 efficiently in the sample.

7. DNA–Lipofectamine complex precipitates on cells surface, so transfection can take place. Swirling the plates periodically help brings up those complexes that fall to empty space and thus increase the chances of making a contact with cells. Doing so increases the transfection efficiency.

8. Prolonged incubation in DTT-containing sample buffer releases a significant amount of antibody fragments and increases the background of immunoblot. Boiling the sample- bound gel beads will make the issue worse.

9. Note that since the anti-FLAG M2 gel is a mouse monoclonal antibody and the antibody leaks from the gel beads are inevi- table, so the primary and secondary antibodies pair for immu- noblot has to be rabbit/anti-rabbit.

10. To prevent the coverslips from floating upon addition of medium, use a small drop of nail polish at a corner to glue the coverslip to the plate. Let the nail polish dry completely by leaving the plates open in the laminar flow hood.

11. The fixation condition is determined by the nature of primary antibodies. In this case, the anti-14-3-3zeta works only on methanol fixed cells while other antibodies may work better under other fixation conditions.

This work was supported by the National Institutes of Health (grant numbers R01 DA017302, R03 DA036802 and P30 DA013429).

1. Li JG, Chen C, Huang P et al (2012) 14-3- 3zeta Protein regulates anterograde transport of the human kappa-opioid receptor (hKOPR). J Biol Chem 287:37778–37792

2. Chen C, Wang Y, Huang P et al (2011) Effects of C-terminal modifications of GEC1 protein and gamma-aminobutyric acid type A (GABA(A)) receptor-associated protein (GABARAP), two microtubule-associated pro- teins, on kappa opioid receptor expression. J Biol Chem 286:15106–15115

3. Chen Y, Chen C, Kotsikorou E et al (2009) GEC1-kappa opioid receptor binding involves

5. 6.

hydrophobic interactions: GEC1 has chaperone- like effect. J Biol Chem 284:1673–1685
Chen C, Li JG, Chen Y et al (2006) GEC1 inter- acts with the kappa opioid receptor and enhances expression of the receptor. J Biol Chem 281:7983–7993 WangY,DunSL,HuangPetal(2006)Distribution and ultrastructural localization of GEC1 in the rat CNS. Neuroscience 140:1265–1276

Chen C, Xue JC, Zhu J et al (1995) Characterization of irreversible binding of beta- funaltrexamine to the cloned rat mu opioid receptor. J Biol Chem 270:17866–17870

Chapter 11

Detection of Mu Opioid Receptor (MOPR)
and Its Glycosylation in Rat and Mouse Brains by Western Blot with Anti-μC, an Affinity-Purified Polyclonal Anti-MOPR Antibody

Peng Huang, Chongguang Chen, and Lee-Yuan Liu-Chen Abstract

Our experience demonstrates that it is difficult to identify MOPR in rat and mouse brains by western blot, in part due to low abundance of the receptor and a wide relative molecular mass (Mr) range of the receptor associated with its heterogeneous glycosylation states. Here, we describe generation and purification of anti-μC (a rabbit polyclonal anti-MOPR antibody), characterization of its specificity in immunoblotting of HA-tagged MOPR expressed in a cell line, and ultimately, unequivocal detection of the MOPR in brain tissues by western blot with multiple rigorous controls. In particular, using brain tissues from MOPR knockout (K/O) mice as the negative controls allowed unambiguous identification of the MOPR band, since the anti-MOPR antibody, even after affinity purification, recognizes nonspecific protein bands. The MOPR was resolved as a faint, broad, and diffuse band with a wide Mr range of 58–84 kDa depending on brain regions and species. Upon deglycosylation to remove N-linked glycans by PNGase F (but not Endo H), the MOPR became a dense and sharp band with Mr of ~43 kDa, close to the theoretical Mr of its deduced amino acid sequences. Thus, MOPRs in rodent brains are differentially glycosylated by complex type of N-linked glycans in brain region- and species-specific manners. Furthermore, we characterized the MOPR in an A112G/N38D-MOPR knockin mouse model that possesses the equivalent substitution of the A118G/N40D SNP in the human MOPR gene. The substitution removes one of the four and five N-linked consensus glycosylation sites of the mouse and human MOPR, respectively. We demonstrated that the Mr of the MOPR in A112G mouse brains was lower than that in wild-type mouse brains, and that the difference was due to lower degrees of N-linked glycosylation.

Key words A118G, Brain, Receptor glycosylation, Immunoblotting, Mu opioid receptor

1 Introduction

Despite the assertions that many companies made, most of commercially available antibodies against seven-transmembrane receptors/G protein-coupled receptors (7TMRs/GPCRs) are not suitable for western blot. Lack of specificity is a common problem for 7TMR antibodies [1–7]. It has been suggested by a number of

Santi M. Spampinato (ed.), Opioid Receptors: Methods and Protocols, Methods in Molecular Biology, vol. 1230, DOI 10.1007/978-1-4939-1708-2_11, © Springer Science+Business Media New York 2015

141

142 Peng Huang et al.

2 Materials

2.1 Reagents

reports from different groups, including ours (see [6, 8] for reviews), that even with good choices of peptide epitopes as antigens, it is often serendipitous to obtain good antibodies for western blot of 7TMRs/GPCRs, including the MOPR.

By detecting MOPR with multiple approaches following SDS- PAGE, several groups, including ours, demonstrated the MOPR to be a broad and diffuse band, with a Mr range between 58 and 97 kDa, depending on cell lines, brain regions and species. The approaches include specific covalent labeling with [3H]beta- funaltrexamine ([3H]beta-FNA)] [9, 10], agonist-induced MOPR phosphorylation [11–13], and western blot with anti-MOPR anti- bodies [14–18]. The broad and diffuse nature of the band was shown to be due to heterogeneity in N-linked glycosylation (see [8] for review). Most important for the immunoblotting approach is the use of cells not transfected with the MOPR and brain tissues of MOPR-K/O mouse as the in vitro and in vivo negative con- trols. The convergence of pharmacological and biochemical data renders the unequivocal identification of MOPR (see [8] for review). To the best of our knowledge, most of the commercially available antibodies were characterized by preincubation with the peptide or protein antigen, which by itself does not constitute proof of speci- ficity and may be misleading (see below). In addition, none has been characterized against MOPR-K/O mouse brain tissues. Therefore, results obtained with such antibodies should be inter- preted with caution.

We generated and purified an anti-MOPR antibody. The immuno- blotting specificity of the antibody was examined rigorously with mul- tiple controls, such as the human influenza hemagglutinin (HA)-tagged MOPR expressed in CHO cells (the positive control) versus CHO cells transfected with the κ or δ opioid receptor (negative controls) [15] and brain tissues from wild-type versus MOPR-K/O mice (neg- ative controls) [15–17]. Thus, MOPRs in rat and mouse brain tissues were identified unambiguously by western blot. The materials and methods are detailed as it follows.

1. The μC peptide (TNHQLENLEAETAPLP), corresponding to MOPR 383–398 in the C-terminal domain, with an added cysteine at the N-terminus was used as the antigen. The pep- tide was custom-synthesized by EZBiolab (Carmel, IN).

2. The following reagents were purchased from indicated sup- pliers: AffiGel-15 (Bio-Rad); HA.11 antibody (Covance); Immobilon-P PVDF transfer membrane (Millipore); Mini CompleteTM protease inhibitor cocktail (Roche); PANSORBIN (Calbiochem); PNGase F (peptide N-glycosidase F), Endo H and buffers (New England Biolabs); SeeBlue Prestained protein

. 2.2 Cell Lines

. 2.3 Rat Brain

Collection

2.4 Solutions

2.5 SDS-PAGE Buffers

standards (Invitrogen); m-maleimidobenzoyl-N-hydroxysuc- cinimide ester, sulfo-NHS-LC-Biotin and SuperSignal West Pico Chemiluminescent Substrate solutions (Pierce); Lectin from Triticum vulgaris (wheat germ agglutinin/WGA)-agarose (Sigma-Aldrich); WGA-Sepharose 6 MB (Amersham Pharmacia); Reagent-grade chemicals were purchased from Sigma-Aldrich or Fisher Scientific.

Clonal Chinese hamster ovary (CHO) cell line stably expressing HA-rMOPR (CHO-HA-rMOR) was established and cultured as described previously [19] (Bmax value of [3H]diprenorphine bind- ing=1.8 pmol/mg membrane protein).

Frozen meninges-stripped brains of Sprague–Dawley rats (mixed gender) were purchased from Pel-Freeze Biologicals (Rogers, AR). Brains were also collected from male Sprague–Dawley rats (Charles River), littermates of male and female wild-type and MOPR knock- out mice originally generated by Dr. John Pintar’s group [20] and bred in the Central Animal Facility in Temple University, and male and female wild-type and A112G-MOPR mice produced by Dr. Julie Blendy’s group [21].

All the solutions were prepared with deionized water (18.2 MΩ cm at 25 °C). Buffers were stored at room temperature unless indi- cated otherwise.

1. Cathode buffer (0.1 M Tris, 0.1 M Tricine, 0.1 % SDS, pH 8.25): to prepare 2 L of 5× stock solution, dissolve the follow- ing chemicals in 1.5 L water: 121 g Tris, 180 g Tricine and 10 g SDS and bring up to 2 L. There is no need to adjust pH.

2. Anode buffer (0.2 M Tris–HCl, pH 8.9): to prepare 2 L of 10× stock solution, dissolve 484 g Tris in 1 L water, adjust pH with ~70 mL concentrated HCl (37 %), bring up to 2 L with water.

3. Gel buffer (1.0 M Tris–HCl, 0.1 % SDS, pH 8.45): to prepare 1 L of 3× stock solution, dissolve 363 g Tris and 3 g SDS in 500 mL water, adjust pH with ~90 mL concentrated HCl (37 %), bring up to 1 L with water.

4. Gel stock (49.5 % T, 3 % C): dissolve 48 g acrylamide and 1.5 g biascrylamide, bring up to 100 mL with water.

5. AP stock: 10 % ammonium persulfate in water (store at 4 °C for up to a month and at −20 °C for years).

6. 8 % separation gel preparation: 2.5 mL gel stock+5 mL gel buffer + 7.5 mL water + 90 μL AP stock + 12 μL TEMED.

7. 4 % Stacking gel preparation: 0.5 mL gel stock+1.5 mL gel buffer + 4.2 mL water + 120 μL AP stock + 12 μL TEMED.

Mu-Opioid Receptor Detection by Western Blot 143

144 Peng Huang et al.

2.6 Western Blot Buffers

8. 2× Laemmli sample buffer (125 mM Tris–HCl/pH 6.8, 4 % SDS, 40 % glycerol, 0.02 % bromphenol blue): to prepare 200 mL, add 20 mL 1 M Tris–HCl/pH 6.8, 40 mL 20 % SDS, 80 mL glycerol, 0.04 g bromphenol blue, 60 mL water. Before preparing the protein loading samples, add 0.1 M DTT (dithio- threitol) freshly to the 2× Laemmli sample buffer (154 mg DTT/10 mL).

1. Transfer buffer (25 mM Tris, 0.2 M glycine, pH 8.5): to pre- pare 1 L of 10× stock solution, dissolve 30.2 g Tris and 150 g glycine, make up to 1 L with water, no need to adjust pH.

2. TBS buffer (20 mM Tris–HCl, 0.9 % NaCl, pH 7.4): to pre- pare 2 L of 10× stock solution, dissolve 48.4 g Tris and 180 g NaCl, adjust pH with ~31 mL concentrated HCl (37 %), make up to 2 L with water.

3. TBS-T buffer: 1× TBS plus 0.05 % tween-20.

4. Blocking buffer: TBS-T buffer containing 5 % Nestle Carnation instant nonfat dry milk (store at 4 °C for up to a week).

5. Ponceu S (0.5 %) for blot staining: 0.5 g ponceu S in 100 mL water.

1. Breaking buffer (25 mM Tris–HCl buffer/pH 7.4, 1 mM EDTA, 0.1 mM PMSF freshly added).

2. Wash buffer 1 (25 mM Tris–HCl buffer/pH 7.4).

3. Suspension buffer (50 mM Tris–HCl buffer/pH 7.0, 0.32 M sucrose).

1. TTSEC buffer (50 mM Tris–HCl/pH 7.4, 2 % Triton X-100, 150 mM NaCl, 5 mM EDTA, Mini protease inhibitors (1 tablet/ 10 mL) (store at −20 °C)). Thaw and add 1 mM PMSF freshly to TTSEC buffer before use.

2. Wash buffer 2 (TTSEC buffer containing 0.2 % Triton X-100) (store at 4 °C).

3. Elution buffer A (5 % SDS, 0.4 M DTT) (store at −20 °C)

4. Elution buffer B (Wash buffer 2 plus 0.25 M N-acetyl- d-glucosamine).

1. For SDS-PAGE: SE660 Tall Standard Dual Cooled Vertical Unit (Hoefer), glass plates (18×8 cm), 0.75-mm spacers and 10- or 15-well combs.

2. For gel transfer: CriterionTM Blotter (Bio-Rad).

3. For chemiluminescence imaging: FUJIFILM LAS-1000 imag- ing system.

2.7 Membrane Preparation Buffers (Store at 4 °C)

2.8 Solubilization Buffers

2.9 Apparatus

3 Methods

3.1 Generation and Purification
of a Polyclonal Anti-MOPR Antibody Anti-μC

3.1.1 Peptide Synthesis and Conjugation to KLH or BSA

All the methods have been reported in our previous publications, which are reproduced with some modifications.

This method has been reproduced from the methods of Chen et al. [22] with modifications (with permission of The American Society for Biochemistry and Molecular Biology).

TNHQLENLEAETAPLP (μC peptide), which corresponds to the amino acids 383–398 of the C-terminal domain of the cloned rat MOPR-1 (Gen-Bank: NM_013071), was chosen to be the antigen because of its high antigenicity and unique sequence. A Swiss Prot search indicated that no other proteins have this sequence. The sequence is identical among the rat, mouse, and human MOPRs. The peptide with an added cysteine residue at the N-terminus (CTNHQLENLEAETAPLP) was custom-synthesized by EZBiolab (Carmel, IN). The addition of a cysteine facilitated conjugation to keyhole limpet hemocyanin (KLH) or bovine serum albumin (BSA) [23]. KLH or BSA was activated with m-maleimidobenzoyl- N-hydroxysuccinimide ester (MBS) according to the manufactur- er’s instructions (Pierce) and passed through a G-25 Sephadex column to remove free MBS. Activated KLH or BSA was collected and incubated with μC peptide. Reaction mixtures were then used as the antigen.

1. Antiserum production was carried out by Covance, Inc. (Denver, PA). Two female New Zealand white rabbits (3–3.5 kg) were immunized with the μC peptide-KLH or -BSA according to standard protocols [23]. The peptide-KLH conjugate was used in the primary injection and the peptide-BSA conjugate was used in booster injections. Antiserum was collected 10–14 days after each booster injection. Titer of immunoreactivity of antiserum was determined by ELISA assay.

2. Partial purification of Antiserum. The μC peptide was conju- gated to AffiGel-15 for generation of μC-AffiGel-15 affinity gel according to the manufacture’s instructions (Bio-Rad). The antiserum generated against the μC peptide was diluted with 1 volume of 10 mM Tris–HCl buffered saline (pH 7.5) and then passed through a 1-mL μC-AffiGel-15 column. After extensive washing, adsorbed antibodies were eluted with 0.1 M glycine- HCl, 10 % ethylene glycol, pH 2.5 and immediately neutralized with 1 M Tris. Eluted antibodies were precipitated with ammo- nium sulfate, desalted over a Sephadex G-25 column equilibrated

3.1.2 Generation of Antiserum Against the μC Peptide

Mu-Opioid Receptor Detection by Western Blot 145

146 Peng Huang et al.

3.1.3 Preparation
of Biotinylated Anti-μC (Anti-μC-Biotin)

3.2 Brain Membrane Preparation

in 0.1 M MOPS buffer (pH 7.5) and concentrated. Recovery of immunoreactivity was monitored by immunoprecipitation of HA-tagged MOPR before and after the purification process followed by immunoblotting with anti-HA mouse antibody. The partially purified antibodies are termed anti-μC.

Biotinylation of anti-μC by sulfo-NHS-LC-Biotin was carried out according to the manufacturer’s instructions (Pierce).

This method has been reproduced from the methods of Huang et al. [17] with some modifications, (with permission of Biochemical Society).

Mouse or rat brains were collected. The striatum or thalamus tissues were dissected and homogenized in 8–10× volumes (v/w) of 25 mM Tris–HCl buffer (pH 7.4) containing 1 mM EDTA and 0.1 mM PMSF on ice and then centrifuged at 30,000 rev./min (100,000 × g) for 30 min in a Beckman 50.4 Ti rotor. Pellets were rinsed twice with 25 mM Tris–HCl buffer (pH 7.4) and resus- pended in 0.32 M sucrose buffered with 50 mM Tris–HCl (pH 7.0). Suspended membranes were passed through a 26.5-gauge needle five times and then frozen at −80 °C until use.

This method has been reproduced from the methods of [17] (with modifications, with permission of Biochemical Society).

Brain membrane proteins (2–3 mg) were solubilized in 0.8 mL of TTSEC buffer [50 mM Tris–HCl (pH 7.4), 2 % Triton X-100, 150 mM NaCl, 5 mM EDTA and Mini CompleteTM protease inhibitors (Roche; 1 tablet/10 mL)] with 1 mM PMSF at 4 °C for 3 h. Supernatants were collected after centrifugation at 30,000 rev./ min (150,000×g) for 30 min in a Beckman 50.4 Ti rotor and mixed with 50 μL of WGA-Sepharose 6 MB or WGA-agarose at 4 °C for 1 h. The beads were washed three times with ice-cold TTSEC buffer containing 0.2 % Triton X-100. The WGA-bead- associated proteins were dissociated/eluted in Elution Buffer A (5 % SDS and 0.4 M dithiothreitol) or Elution buffer B (wash buf- fer 2 containing 0.25 M N-acetyl-d-glucosamine), and subjected to deglycosylation as follows. Some solubilized supernatants were also immunoprecipitated with anti-μC followed by PANSORBIN (Calbiochem). After washing, the PANSORBIN-bound proteins were dissociated in Elution Buffer A and left untreated or treated with PNGase F as follows.

Treatments of WGA-affinity purified eluates or anti-μC- PANSORBIN immunoprecipitated materials with PNGase F or Endo H has been reproduced from the methods of Huang et al. [17] with modifications (with permission of Biochemical Society).

3.3 Solubilization and Enrichment
of MOPRs by WGA- Affinity Purification or Immunoprecipitation

3.4 Deglycosylation of MOPRs

3.5 SDS-PAGE and Western Blot

Treatment of MOPRs with PNGase F (or Endo H) was performed according to the manufacturer’s protocols (New England Biolabs). To 3 μL of the dissociated proteins described above, 3 μL of 10× G7 (or G5) reaction buffer [0.5 M sodium phosphate (pH 7.5)] or [0.5 M sodium citrate (pH 5.5)], and 3 (or 0) μL of 10 % Nonidet P40 was added, followed by 21 (24) μL of water. The 30 μL reaction mixture was incubated at 37 °C over- night in the absence or presence of 1 μL of PNGase F (or Endo H) (500 Units). An equal volume (30 μL) of 2× Laemmli sample buffer was added to the reaction mixture. The 60 μL sample was incubated at 37 °C for 10 min and loaded on to a SDS-PAGE gel for separation.

1. SDS-PAGE was carried out as we previously described [9], which was modified from the method of Schagger and von Jagow [24]. It is noteworthy that, in this system, (1) tricine replaced glycine in the Laemmli SDS-PAGE system, (2) the cathode and anode buffers had different chemical components and pH, and (3) the same gel buffer was used for preparing both stacking gel and separating gel, as shown in the Materials. Electrophoresis was carried out at 40 V for 30 min and then at 90 V for ~3 h using 8 cm gels (Hoefer protein electrophoresis vertical gel unit SE660, 0.75 mm spacers and 10- or 15-well combs). Cells or brain membranes were solubilized in 2× Laemmli sample buffer with brief sonication, incubated at RT or 37 °C for 10–15 min (see Note 1) before loading onto the gel with a protein amount of 15–30 μg or storing at −80 °C. Protein separation was performed with a 8 % poly- acrylamide separation gel with a 4 % stacking gel. For protein molecular mass standards, prestained markers were used.

2. Western blot was performed as previously described (e.g. [25]. After SDS-PAGE, protein bands on the gel were transferred to Immobilon-P PVDF membranes. Membranes were incubated with blocking buffer for 30 min at room temperature on an orbital shaker to block nonspecific binding and then incubated with one of the following primary antibodies (1:5,000) in the blocking buffer at 4 °C overnight on an orbital shaker: partially purified rabbit anti-μC polyclonal antibody, anti-μC-biotin or mouse anti-HA monoclonal antibody (HA.11). After three washes with TBS-T for 10 min each time, blots were incubated with the following secondary antibodies (1:5,000) for 1 h at room temperature: goat anti-rabbit IgG conjugated with HRP, anti-biotin-HRP conjugate or goat anti-mouse IgG conju- gated with HRP. Membranes were washed three times with TBS-T and then reacted with SuperSignal West Pico Chemi- luminescence Substrate Solution. Images were captured with a FUJIFILM LAS-1000 imaging system.

Mu-Opioid Receptor Detection by Western Blot 147

148 Peng Huang et al.

3.6 Determination of the Specificity
of Anti-μC to Detect MOPRs by Western Blot with Various Controls

1. Opioid receptors expressed in cultured cells [15]. HA-rMOPR expressed in CHO cells were used as the positive control, and FLAG-hKOPR and FLAG-mDOPR expressed in CHO cells were used as the negative controls. The three cell lines express rMOPR, hKOPR and mDOPR at similar levels, with Bmax val- ues of [3H]diprenorphine binding at 1–2 pmol/mg membrane protein. For CHO-HA-rMOPR cells, anti-μC-labeled rat MOPR proteins migrated as a major broad and diffused band with a median Mr of 78 kDa and a minor lower band of Mr 52 kDa (Fig. 1a, left panel), which are similar to the bands detected by HA.11 (Fig. 1a, right panel). Both antibodies detected no spe- cific bands in either CHO-FLAG-hKOPR or CHO-FLAG- mDOPR cells (Fig. 1a). (This procedure has been reproduced from Huang et al. [15]; with modifications, with permission from Elsevier.)

2. Brain tissues from wild-type and MOPR knockout mice. This procedure has been reproduced from Huang et al. [15] with modifications (with permission from Elsevier).
To detect endogenous MOPR, caudate putamen (CPu), thala- mus, and cerebellum were dissected from brains of wild-type and MOPR-K/O mouse littermates, and brain membranes were prepared. Western blotting revealed that in wild-type mice, anti-μC labeled several bands (see Note 2) in the CPu, one of which was absent in the MOPR-K/O mice (Fig. 1b, upper panel, lanes 1 and 2), indicating that this pro- tein band, with an Mr of 60–84 kDa (median, 74 kDa), repre- sents the MOPR. Similarly, in the thalamus, one of the bands labeled by anti-μC in the wild type was not present in the MOPR-K/O mice (Fig. 1b, upper panel, lanes 3 and 4). However, surprisingly, the MOPR band in the thalamus was narrower and had lower Mr [58–68 kDa (median, 63 kDa)] (Fig. 1b, upper panel, lane 3). The labeling of both bands was completely blocked by preadsorption of anti-μC with the μC peptide (Fig. 1b, middle panel) (see Note 3). The cerebellum does not express the MOPR and there was no difference in labeling in cerebella of wild-type and MOPR-K/O mice (Fig. 1b, upper panel, lanes 5 and 6). The amount of protein loaded (30 μg per lane) for each sample was similar as demon- strated by Ponceu S staining of the membranes (Fig. 1b, lower panel).

Immunoblotting with anti-μC was also performed on brain membranes of the rat CPu and thalamus. MOPR in the rat CPu migrated as a broad and diffuse band with a Mr range of 61–84 kDa (median, 75 kDa), while MOPR in the rat thala- mus was resolved as a narrow and diffuse band (60–72 kDa) with a smaller median Mr of 66 kDa (Fig. 1c, left panel). Both bands were completely blocked by preadsorption with the μC

Mu-Opioid Receptor Detection by Western Blot 149

Fig. 1 Immunoblotting of the HA-rMOPR stably expressed in CHO cells (a) and the endogenous MOPR in CPu and thalamus of the mouse (b) or rat (c) brain. (a) HA-rMOPR, FLAG-hKOPR, and FLAG-mDOPR stably expressed in CHO cells were blotted by anti-mC, a polyclonal anti-MOPR antibody (1.3 mg/mL) (1:5,000) (a, left panel). The same membrane was stripped and blotted with an anti-HA monoclonal antibody (HA.11) (1:5,000) as described in Subheading 3 (a, right panel). (b) Membranes prepared from CPu, thalamus, and cerebellum of wild-type and MOPR-K/O mice were blotted with anti-mC (1:5,000) preincubated without (b, upper panel) or with mC peptide (0.6 mg/mL) (b, middle panel) as described in Subheading 3. The same blot used in the upper panel of b was stained by Ponceu S to show protein loading amounts (b, lower panel). (c) CPu and thalamus were dissected from frozen rat brains and membranes were prepared. Western blot was performed with anti- mC (1:5,000) (c, left panel) or anti-mC (1:5,000) preincubated with the mC peptide (0.6 mg/mL) as described in Subheading 3 (c, right panel). Each of these figures represents one of the three experiments performed with separate batches of tissues. (Reprinted from Fig. 1 of Huang et al. ref. [15], with permission of Elsevier)

peptide (Fig. 1c, right panel). Thus, a similar discrepancy in Mr of the MOPR between the CPu and thalamus was observed in the rat.

3. ExperimentstorevealcomplexN-linkedglycosylationofMOPRs. De-glycosylated MOPRs (Figs. 2 and 3) and brain tissues from A112G-MOPR knockin mice (Fig. 3).

In Fig. 2, western blotting of WGA affinity-purified mate- rials (see Note 4) with anti-μC showed that the MOPR in the

150 Peng Huang et al.

Fig. 2 Deglycosylation of MOPR from rat CPu and thalamus. Membranes of the rat CPu or thalamus were solubilized with 2 % Triton X-100. (a) The solubilized preparations were applied to WGA-agarose beads. After washing, the bound glycoproteins were eluted with Elution Buffer B (wash buffer 2 with 0.25 M N-acetyl-d- glucosamine). The eluate was left untreated or treated with PNGase F, resolved with 8 % SDS-PAGE and immu- noblotted with anti-μC (1:5,000). (b) The solubilized preparations were immunoprecipitated with anti-mC followed by PANSORBIN (Calbiochem), dissolved in Elution Buffer A (5 % SDS, 0.4 M DTT) and left untreated or treated with PNGase F or Endo H. Samples were analyzed by 8 % SDS-PAGE and immunoblotting was performed with anti-μC-biotin (1:5,000) and anti-biotin-HRP-conjugate (1:5,000). Each figure shown represents one of the two independent experiments (Reprinted from Fig. 2 of Huang et al. ref. [15], with permission of Elsevier)

Fig. 3 Deglycosylation of thalamic MOPR from AA and GG mice. Thalamic mem- branes of the AA or GG mice (50 % of each sex, see Note 5) were pooled and solubilized with 2 % Triton X-100. The solubilized preparations were applied to WGA-Sepharose 6 MB beads (Amersham Pharmacia) and the bound glycopro- teins were eluted with Elution Buffer A (5 % SDS and 0.4 M dithiothreitol). The eluate was left untreated or treated with PNGase F, resolved with 8 % SDS-PAGE, and immunoblotted with anti-μC (1:5,000). Each figure shown represents one of two independent experiments. (Reprinted from Fig. 3 of Huang et al. ref. [17], with permission of Biochemical Society)

Mu-Opioid Receptor Detection by Western Blot 151

rat CPu migrated as a broad and diffuse band with a median M of 75 kDa (Fig. 2a, lane 1). Blotting of immunoprecipitated complex with biotinylated anti-μC yielded similar results (Fig. 2b, lane 3). In contrast, the MOPR in the rat thalamus migrated as a narrow and diffuse band with a median Mr of 66 kDa (Fig. 2a, lane 2 and Fig. 2b, lane 4). PNGase F treat- ment of the WGA affinity-purified materials, which removes all N-linked glycans, resulted in an increase in the mobility of MOPR in the rat CPu or thalamus on SDS-PAGE (Fig. 2a, lanes 3 and 4), compared with the untreated controls (Fig. 2a, lanes 1 and 2). Importantly, the diffuse bands with different widths and median Mr’s (Fig. 2a, lanes 1 and 2) in the two brain regions became sharp bands with identical Mr’s (43 kDa) (Fig. 2a, lanes 3 and 4).

Anti-μC-immunoprecipitated MOPRs (see Note 4) of the CPu or thalamus was also treated with PNGase F, yielding a similar observation (Fig. 2b, lanes 1–4). Thus, the difference in Mr of the MOPR in the thalamus and CPu is due to different degrees of N-linked glycosylation. In addition, Endo H treat- ment, which cleaves N-linked glycans of high-mannose (ER forms) and some hybrid types, caused no mobility changes of the MOPR of CPu and thalamus (Fig. 2b, lanes 5 and 6). These results indicate that the MOPR in the CPu and thalamus contains different complex type N-linked glycans and is likely to be located in trans-Golgi and/or plasma membranes (see Note 4) (Reproduced from Huang et al. [15] with modifications, with permission from Elsevier).

OPRM1 A118G is a common SNP in the coding region of the human MOPR gene OPRM1 [26, 27]. This SNP is associ- ated with higher morphine doses required for postoperative analgesia and better treatment outcome for alcohol addiction (reviewed in [21]). A mouse model possessing the equivalent substitution (A112G) in the oprm1 gene was generated by Dr. Julie Blendy’s group of University of Pennsylvania [16]. Mice homozygous for the G112 allele (G/G) displayed lower anti- nociception to morphine compared with those homozygous for A112 allele (A/A), similar to humans, suggesting that the mice are a good model to further characterization of this SNP [16]. This SNP results in N40D substitution and thus elimi- nates one of the consensus N-linked glycosylation sites. We investigated the N-linked glycosylation status of the MOPR in G/G mice (mice homozygous for the 112G allele of MOPR), compared with A/A mice (wild-type mice homozygous for the 112A allele of MOPR). Western blotting of WGA-affinity- purified materials (see Note 4) with the anti-μC antibody showed that the thalamic MOPRs in A/A and G/G mice migrated as a single diffuse band with a median Mr of 62 and 55 kDa (Fig. 3, lanes 1 and 2 from left), respectively. Treatment of the WGA-affinity-purified materials with PNGase F resulted

152 Peng Huang et al.

4 Notes

in an increase in the mobility of thalamic MOPRs from both A/A and G/G mice on SDS/PAGE gels (Fig. 3, lanes 3 and 4 from left), compared with the untreated controls (Fig. 3, lanes 1 and 2 from left). More importantly, the diffuse MOPR bands with different median Mr’s in the two mouse lines (Fig. 3, lanes 1 and 2 from left) became sharp bands with a lower and identi- cal molecular mass (41 kDa) (Fig. 3, lanes 3 and 4 from left). Thus the difference in Mr’s of the MOPRs in A/A and G/G mice is due to differential N-linked glycosylation (i.e., a lower level of N-glycosylation of the MOPR in G/G mice) (Reproduced from Huang et al. [17] with modifications, with permission from Biochemical Society).

1. Do not boil the protein loading samples. Otherwise, MOPRs (or KOPRs, DOPRs) would aggregate and become too big to get into the separation gel.

2. Due to the low abundance of MOPRs in brain tissues, without any enrichment of MOPRs anti-μC always recognizes multiple nonspecific bands. See Note 4 for different enrichment strate- gies we have adopted.

3. Preincubation of anti-μC antibody with the μC peptides also partially blocked some of the nonspecific bands. Thus, using preincubation of antibody with the antigen peptide alone as the negative control is misleading.

4. Enrichment of brain MOPRs by WGA-affinity purification or immunoprecipitation resulted in much fewer nonspecific bands and made more prominent the single specific band which was broad and diffuse with high Mr’s. In addition, partial purifica- tion of MOPRs by lipid rafts preparation had similar effects as shown in our publications [14, 15]. The MOPR in this single band contains complex type N-linked glycans and is likely to be fully glycosylated and likely located in trans-Golgi and/or plasma membranes. In contrast, the glycosylated intermediates are located in the ER and cis-Golgi.

5. No sex differences were found regarding the band pattern and glycosylation status of the MOPR recognized by anti-μC in CPu or thalamus of mice with C57/BL6 background [17].

This work was supported by the National Institutes of Health (grant numbers R01 DA017302 and P30 DA013429).

Acknowledgement

References

1. Bodei S, Arrighi N, Spano P et al (2009) Should we be cautious on the use of commer- cially available antibodies to dopamine recep- tors? Naunyn Schmiedebergs Arch Pharmacol 379:413–441

2. Hamdani N, van der Velden J (2009) Lack of specificity of antibodies directed against human beta-adrenergic receptors. Naunyn Schmie- debergs Arch Pharmacol 379:403–407

3. Jensen BC, Swigart PM, Simpson PC (2009) Ten commercial antibodies for alpha-1-adre- nergic receptor subtypes are nonspecific. Naunyn Schmiedebergs Arch Pharmacol 379: 409–412

4. Jositsch G, Papadakis T, Haberberger RV et al (2009) Suitability of muscarinic acetylcholine receptor antibodies for immunohistochemistry evaluated on tissue sections of receptor gene- deficient mice. Naunyn Schmiedebergs Arch Pharmacol 379:389–395

5. Lu X, Bartfai T (2009) Analyzing the validity of GalR1 and GalR2 antibodies using knock- out mice. Naunyn Schmiedebergs Arch Pharmacol 379:417–420

6. Michel MC, Wieland T, Tsujimoto G (2009) How reliable are G-protein-coupled receptor antibodies? Naunyn Schmiedebergs Arch Pharmacol 379:385–388

7. Pradidarcheep W, Stallen J, Labruyere WT et al (2009) Lack of specificity of commercially available antisera against muscarinergic and adrenergic receptors. Naunyn Schmiedebergs Arch Pharmacol 379:397–402

8. Huang P, Liu-Chen LY (2009) Detection of the endogenous mu opioid receptor (mopr) in brain. Front Biosci 1:220–227

9. Liu-Chen L-Y, Chen C, Phillips CA (1993) Beta-[3H]funaltrexamine-labeled mu-opioid receptors: species variations in molecular mass and glycosylation by complex-type, N-linked oligosaccharides. Mol Pharmacol 44:749–756

10. Chen C, Xue JC, Zhu J et al (1995) Charac- terization of irreversible binding of beta-funal- trexamine to the cloned rat mu opioid receptor. J Biol Chem 270:17866–17870

11. Deng HB, Yu Y, Wang H et al (2001) Agonist- induced mu opioid receptor phosphorylation and functional desensitization in rat thalamus. Brain Res 898:204–214

12. Yu YK, Zhang L, Yin XX et al (1997) Mu-opioid receptor phosphorylation, desensitization, and ligand efficacy. J Biol Chem 272: 28869–28874

13. Carman CV, Barak LS, Chen C et al (2000) Mutational analysis of Gbeta gamma and phospholipid interaction with G Protein-

coupled receptor kinase 2. J Biol Chem 275: 10443–10452

14. Huang P, Xu W, Yoon SI et al (2007) Agonist treatment did not affect association of mu opi- oid receptors with lipid rafts and cholesterol reduction had opposite effects on the receptor- mediated signaling in rat brain and CHO cells. Brain Res 1184:46–56

15. Huang P, Chen C, Xu W et al (2008) Brain region-specific N-glycosylation and lipid rafts association of the rat mu opioid receptor. Biochem Biophys Res Commun 365:82–88

16. Mague SD, Isiegas C, Huang P et al (2009) Mouse model of OPRM1 (A118G) polymor- phism has sex-specific effects on drug-mediated behavior. Proc Natl Acad Sci U S A 106: 10847–10852

17. Huang P, Chen C, Mague SD et al (2012) A common single nucleotide polymorphism A118G of the mu opioid receptor alters its N-glycosylation and protein stability. Biochem J 441:379–386

18. Lupp A, Richter N, Dol C et al (2011) UMB- 3, a novel rabbit monoclonal antibody, for assessing mu-opioid receptor expression in mouse, rat and human formalin-fixed and paraffin-embedded tissues. Regul Pept 167: 9–13

19. Huang P, Li J, Chen C et al (2001) Functional role of a conserved motif in TM6 of the rat mu opioid receptor: constitutively active and inac- tive receptors result from substitutions of Thr6.34(279) with Lys and Asp. Biochem 40: 13501–13509

20. Schuller AG, King MA, Zhang J et al (1999) Retention of heroin and morphine-6 beta- glucuronide analgesia in a new line of mice lacking exon 1 of MOR-1. Nat Neurosci 2: 151–156

21. Mague SD, Blendy JA (2010) OPRM1 SNP (A118G): involvement in disease development, treatment response, and animal models. Drug Alcohol Depend 108:172–182

22. Chen C, Yin J, Riel JK et al (1996) Determination of the amino acid residue involved in [3H]beta-funaltrexamine covalent binding in the cloned rat mu-opioid receptor. J Biol Chem 271:21422–21429

23. Harlow E, Lane D (1988) Antibodies: a labo- ratory manual. Cold Spring Harbor Laboratory, Cold Spring Harbor, NY

24. Schagger H, von Jagow G (1987) Tricine- sodium dodecyl sulfate-polyacrylamide gel electrophoresis for the separation of proteins in

Mu-Opioid Receptor Detection by Western Blot 153

154 Peng Huang et al.

the range from 1 to 100 kDa. Anal Biochem

166:368–379
25. Xu W, Yoon SI, Huang P et al (2006)

Localization of the kappa opioid receptor in lipid rafts. J Pharmacol Exp Ther 317: 1295–1306

26. Bergen AW, Kokoszka J, Peterson R et al (1997) Mu opioid receptor gene variants: lack

of association with alcohol dependence. Mol Psychiatry 2:490–494

27. Bond C, LaForge KS, Tian M et al (1998) Single-nucleotide polymorphism in the human mu opioid receptor gene alters beta-endorphin binding and activity: possible implications for opiate addiction. Proc Natl Acad Sci U S A 95: 9608–9613

Immunohistochemical Analysis of Opioid Receptors in Peripheral Tissues

Yvonne Schmidt and Halina Machelska Abstract

Immunohistochemical staining is widely used to identify opioid receptors in specific cell types or anatomical structures throughout the nervous system. Opioid receptors are not restricted to the central nervous system, but are also present in peripheral sensory neurons, where their activation exerts analgesic effects without inducing centrally mediated side effects. Here, we describe immunohistochemical analysis of opioid receptors in the peripheral sensory neuron cell bodies, along the axons and their peripheral endings in the hind paw skin, as well as in the spinal cord, under naïve and sciatic nerve damage conditions in mice. Moreover, we consider the current debate on the specificity of antibodies.

Key words Antibodies, Dorsal root ganglia, Immunohistochemistry, Immunofluorescence, Opioid receptors, Peripheral neurons, Specificity controls

1 Introduction

Apart from the central nervous system (CNS), all three opioid receptors (μ, δ, and κ) are also localized in peripheral sensory neurons and in neuroendocrine (pituitary, adrenals), immune, and ectodermal cells [1]. In the peripheral nervous system, opioid receptors are mainly expressed in small- and medium-size dorsal root ganglion (DRG) neurons and their axons, which co-express prototypical sensory neuropeptides, such as substance P and calci- tonin gene-related peptide (CGRP) [2–5]. From the production site, the DRG cell bodies, opioid receptors are transported to the central and peripheral neuronal terminals [6, 7]. Similar to the CNS, activation of peripheral opioid receptors reduces neurotrans- mitter release, for example, substance P and CGRP, from central and peripheral terminals of DRG neurons [1, 8]. Peripheral opioid receptors are also coupled to Gi/o proteins that inhibit adenylyl cyclase activity and modulate ion channels [9, 10]. However, in contrast to the CNS, activation of peripheral opioid receptors can efficiently alleviate pain without CNS side effects (e.g., nausea,

Santi M. Spampinato (ed.), Opioid Receptors: Methods and Protocols, Methods in Molecular Biology, vol. 1230, DOI 10.1007/978-1-4939-1708-2_12, © Springer Science+Business Media New York 2015

Chapter 12

155

156 Yvonne Schmidt and Halina Machelska

2 Materials

2.1 Buffers and Solutions

dependence, and addiction) [10]. Indeed, many studies have provided strong evidence for peripheral analgesic effects of opioids in injured tissues, including neuropathic conditions. Injection of μ-, δ-, or κ-receptor agonists into the paw (intraplantarly; i.pl.) innervated by an injured nerve attenuated mechanical or heat hypersensitivity in various traumatic neuropathy animal models, including a chronic constriction injury (CCI) of the sciatic nerve, its resembling Mosconi and Kruger model, a partial sciatic nerve ligation, and the spinal nerve ligation [10–14]. Interestingly, our recent study suggests that opioid receptors directly at the site of neuronal injury might be a preferred target since μ-, δ-, or κ-receptor agonists produced stronger analgesia following applica- tion at the CCI site than after i.pl. injection [15]. Additionally, peripheral analgesia can be produced by endogenous opioid peptides derived from immune cells accumulating at the site of nerve damage [16].

The use of immunohistochemistry has become ubiquitous in neuroscience. In contrast to, for example, Western blot, immuno- histochemistry offers the advantage of identifying the anatomical structure, the cell type, and the subcellular localization of a given protein. Although the principles of immunohistochemical reac- tions are relatively simple, many studies may have led to flawed conclusions, mostly because of often overlooked nonspecific stain- ing by antibodies, including those to opioid receptors [17, 18]. Moreover, the staining specificity might be even tissue-dependent [19, 20], suggesting the need for the verification of antibody spec- ificity in each tissue of interest. In this chapter, we describe immu- nohistochemical analysis of opioid receptors along the peripheral neuronal pathways, the DRG, the nerve trunk, and its endings in the hind paw skin and in the spinal cord, under naïve and sciatic nerve damage conditions. We also put a strong emphasis on the antibodies’ specificity in control experiments.

1. Wash buffer (1× phosphate buffered saline [PBS]): For 1 l use 8 g of NaCl (final concentration 137 mM), 0.2 g of KCl (final concentration 2.7 mM), 1.44 g of Na2HPO4 (final concentration 10 mM), and 0.24 g of KH2PO4 (final concentration 2 mM), add up to a volume of approximately 800 ml with distilled water, adjust the pH to 7.4, and add the remaining distilled water to obtain a total volume of 1 l.

2. Dilution buffer (PBS+): To 1× PBS add 0.3 % Triton X-100 and 1 % bovine serum albumin (BSA); stir constantly until the solution is cleared.

2.2 Antibodies and Antigens

3. Paraformaldehyde (PFA) fixative (4 %): For 1 l of final solution, add 800 ml of 1× PBS to a glass beaker on a stir plate in a ventilated hood. Dissolve 40 g of PFA powder by heating the solution to approximately 60 °C while constantly stirring. It may be necessary to add a few drops of 1 M NaOH to the solution to make it clear. Afterwards, cool it down to room temperature, filter through a standard filter paper, and adjust the pH to 7.4.

4. Zamboni’s fixative: Mix 75 ml of saturated aqueous picric acid and 75 ml of distilled water, and filter it. Add 18 g of PFA and heat the solution to 60 °C while stirring constantly. Dropwise add 10–12 ml of 2.5 % NaOH until the solution clears. Filter the solution again and cool it to room temperature. Fill up to 1 l with PO4 buffer (3 g of NaH2PO4 and 33.77 g of Na2HPO4 dissolved in 1 l of distilled water), and adjust the pH to 7.4 (see Note 1).

5. Sucrose solution: Dissolve 10–30 % sucrose in 1× PBS, and adjust to pH 7.4.

1. Primary antibodies to opioid receptors: In our experimental conditions, rabbit polyclonal anti-μ-receptor (Ab10275; Abcam) specifically stained μ-receptors in the spinal cord, sciatic nerve, and skin dermis, but not in the DRG, as judged by the use of tissue from mice lacking all three opioid receptors (μ-, δ-, and κ-receptor knockout mice) [20] (see Note 2). Other opioid receptor antibodies we used still await specificity validation in knockout mouse tissue. However, it is advisable to choose affinity purified antibodies that were validated in the desired staining method by the company.

2. Antibodies to neuronal markers: For example, polyclonal guinea pig anti-α-CGRP (Bachem; see Note 3) to stain pepti- dergic sensory neurons, and chicken anti-neurofilament 200 (NF200; Millipore) to stain myelinated sensory neurons.

3. Immunizing peptides: Use specific immunizing opioid recep- tor peptides (ideally from the same company from which you acquired the antibodies) for a preabsorption control staining (see Subheading 3.6).

4. Fluorescent secondary antibodies: We used TexasRed alone or in combination with fluorescein isothiocyanate (FITC) or Alexa Flour dyes (e.g., Alexa 488 and Alexa 568). The Alexa Flour dyes are recommended when greater photostability or higher fluorescence intensity are needed. Isolectin B4 (IB4) coupled to FITC (e.g., Sigma-Aldrich) can be used as a marker of nonpeptidergic sensory neurons.

Immunohistochemistry of Opioid Receptors 157

158

Yvonne Schmidt and Halina Machelska

2.3

Other Materials

5. Biotinylated secondary antibodies: We use a Vectastain Elite ABC kit (Vector Laboratories) for the respective species IgG, which contain a secondary antibody coupled to biotin and an avidin–peroxidase complex.

6. Peroxidase substrate: We use a 3,3′-diaminobenzidine (DAB) peroxidase substrate kit (Vector Laboratories), according to the manufacturer’s instructions.

1. Surgical dissection tools (e.g., Fine Science Tools): Surgical scissors and forceps (e.g., student standard pattern forceps, student surgical scissors, student Vannas spring scissors, fine Vannas spring scissors, Dumont #5 forceps, fine forceps with curved tips).

2. Dissection microscope (e.g., Zeiss).
3. Cryostat (e.g., Microm).
4. Polysine slides (e.g., Thermo Scientific, Menzel Gläser).

5. PAP pen: It can be used to create a water-repellent barrier that keeps reagents localized on tissue specimens (prevents wasting reagents by keeping liquid pooled in a small droplet).

6. Mounting medium for fluorescent staining: We routinely use Mowiol (e.g., Sigma-Aldrich), prepared according to the man- ufacturer’s instructions.

7. Mounting medium for DAB staining (e.g., Entellan; Millipore). 8. Fluorescence microscope (e.g., Zeiss).

1. To perform the CCI and sham surgery, anesthetize the mouse by placing it in a glass chamber on a ceramic perforated plate located above paper tissues soaked with approximately 15 ml of isoflurane, until the animal loses consciousness. Subsequently, cover the animal’s nose using a tube attached to an anesthesia machine delivering a gaseous mixture of isoflurane (3–4 %) and oxygen.

2. Open the skin at the level of the right mid-thigh by making an incision (approximately 1 cm long), and cut the underlying muscle.

3. Expose the sciatic nerve using fine forceps with curved, smooth tips. For the CCI, place three loose silk sutures (4/0) around the nerve with about 1 mm spacing (see Note 4) and tie them carefully, until they elicit a brief twitch in the respective hind limb. For sham surgery leave the nerve intact.

4. Close the wound with two or three silk sutures.

3 Methods

3.1 Surgeries

3.2 Animal Perfusion

1. It is recommended to perform the perfusion in a chemical fume hood.

2. Deeply anesthetize the animal (by placing it in a glass chamber, as described in Subheading 3.1), and make a midline skin incision from the thoracic inlet to the pelvis. Hold the tip of the sternum with forceps and make an incision on the left and right side of the thoracic cavity to expose the heart.

3. Gently grasp the heart and identify the left ventricle. Place the needle (21-gauge) into the left ventricle toward the aorta and clamp or hold it in place. Start to perfuse the animal with about 40–50 ml of 1× PBS at room temperature (using a pump or a 50 ml syringe). Immediately afterward incise the right atrium to allow the perfusate to exit the circulation.

4. When the fluid exiting the mouse is clear of blood (see Note 5), stop the PBS perfusion and change to cold 4 % PFA (keep the bottle on ice). Slowly perfuse the animal with approximately 40–50 ml of PFA. This should result in visible extension/ stretching of the limbs. After perfusion is complete, immedi- ately start to remove the tissue (see Subheading 3.3).

1. Start with dissecting the paw tissue; use a sharp razor blade. Take off the skin and subcutaneous tissue from the plantar side of the hind paw by cutting just below the bone.

2. To isolate the sciatic nerve, remove the fur at the level of the mid-thigh. Proceed with step 2 described in Subheading 3.1 to isolate a piece of the sciatic nerve containing the CCI liga- tures, the corresponding part of sham- and/or nonoperated nerves. The part of injured nerves should include the ligation site and sites proximal and distal to it.

3. To isolate the spinal cord, remove the fur and skin on the back of the mouse. Cut the spine (with surrounding muscle) as far posterior as possible (until the level at which the femur head joints the pelvis). Then, cut along the spine on both sides and remove the spine at the level of the ribs (including at least one pair of ribs).

4. Pin the spine on a rubber dish (bearing in mind the cranial– caudal orientation) under a dissection microscope and open the lamina of the vertebral arch starting at the cranial end by cutting it alternately on the right and left sides.

5. Take out the spinal cord by gently holding the cranial part of the spine with forceps, and carefully cut off the spinal nerves.

6. Identify the lumbar enlargement of the spinal cord and isolate it.

7. To isolate the DRG, identify the most cranial vertebra that lacks an articulation with a rib and mark it as the first lumbar (L1) vertebra [21].

3.3 Tissue Isolation

Immunohistochemistry of Opioid Receptors 159

160 Yvonne Schmidt and Halina Machelska

3.4 Single
and Double Immunofluorescence

8. Isolate the DRG that supply the sciatic nerve (in mice mainly L3 and L4; a minor supply comes from L5 DRG) [21] (see Note 6).

9. Postfix tissue for 2 h in 4 % PFA in 1× PBS (DRG, sciatic nerve, and spinal cord) or in Zamboni’s fixative (paw skin) for 5–8 h at 4 °C.

10. Exchange the fixative solution with 10 % sucrose solution (in 1× PBS) at 4 °C overnight; optionally, change to a 30 % sucrose solution after 1 h.

11. On the following day, freeze the tissue in a water-soluble frozen section medium and cut them on a cryostat, or store at −80 °C until further processing.

12. Prepare approximately 10 μm-thick sections of the DRG, sciatic nerves, and spinal cord, and approximately 12 μm-thick sections of the paw tissue. Mount sections on polysine-coated slides.

13. Let the slides dry for at least half an hour at room temperature.

1. Place the slides horizontally in a plastic slide box (covered with wet paper to create a “moist chamber”; alternatively, use commercially available slide moist chamber).

2. Wash the slides twice for 5 min with 1× PBS.

3. Expose the slides to the dilution buffer (including 5 % normal serum from the host species of the secondary antibody) for 1 h.

4. To examine single staining or the co-expression of opioid receptors with neuronal markers, incubate the sections over- night with rabbit polyclonal antibodies to the respective opioid receptor (we used the μ-receptor Ab10275 from Abcam (see Subheading 2.2) at a concentration of 1:800) [20] alone or in combination with polyclonal guinea pig anti-α-CGRP (1:800) or chicken anti-NF200 (1:500) appropriately diluted in the dilution buffer.

5. On the following day, wash the slides (3–4 times for 10 min, preferably under mild agitation, e.g., on a shaker) with 1× PBS; the most convenient is to use a glass box for microscopic slides filled with 1× PBS.

6. Incubate the slides for 1 h with the secondary antibodies, e.g., goat anti-rabbit conjugated to TexasRed alone or combined with goat anti-guinea pig conjugated to FITC, or goat anti- chicken conjugated to FITC (both at a dilution of 1:200). Secondary antibodies coupled to Alexa Flour dyes can be used at a concentration of 1:1,000.

7. To identify opioid receptors in nonpeptidergic C fibers, apply IB4 conjugated to FITC (1:150), according to the procedure described for the secondary antibodies (see step 6).

3.5 DAB Immuno- histochemistry

8. Thereafter, wash the sections with PBS (3–4 times for 10 min), carefully remove as much remaining PBS as possible, and mount the slides in Mowiol (see Note 7).

1. Follow step 1 in Subheading 3.4.

2. Incubate the slides for 45 min in PBS+with 0.3–0.6 % H2O2 and 40–50 % methanol to block endogenous peroxidase; alter- natively use 0.3 % H2O2 and 0.1 % sodium azide in PBS, because some antigens can be damaged by the use of methanol.

3. Wash the slides (3–4 times for 10 min, preferably under mild agitation) with 1× PBS.

4. Expose the slides to the dilution buffer (including 5 % normal serum from the host species of the secondary antibody) for 1 h.

5. To examine the expression of opioid receptors, incubate the sections overnight with rabbit polyclonal antibodies to the respective opioid receptor (we used μ-receptor Ab10275 from Abcam (see Subheading 2.2) at a concentration of 1:1,500) [20], diluted in PBS+.

6. On the following day, wash the slides (3–4 times for 10 min, preferably under mild agitation) with 1× PBS.

7. Expose the slides to the secondary antibody solution (follow the instructions on the ABC Elite kit) for approximately 1 h.

8. Wash the slides (3–4 times for 10 min, preferably under mild agitation) with 1× PBS (no PBS+at this step).

9. Incubate the slides in the biotin-peroxidase solution (follow the instructions on the kit; instead of PBS+use 1× PBS as a buffer) for 30–60 min.

10. Wash the slides with 1× PBS (3–4 times for 10 min).

11. Prepare DAB (follow the instructions on the kit), and stain the slides for 30 s up to 2 min (the time should be determined on a “positive” slide and should be similar for all slides; see Note 8).

12. Rinse the slides twice with tap water after DAB staining.

13. Dehydrate the slides in alcohol of increasing concentrations (70, 80, and 100 %), and clear them in xylene solutions of increasing concentrations (70, 80, and 100 %).

14. Mount the slides in Entellan and air dry them under the hood.

1. Include slides without staining with the primary antibody in each staining procedure to verify the staining specificity of the secondary antibody.

2. Preabsorption test: To exclude antigen-independent nonspe- cific interactions of the primary antibody, incubate the primary

3.6 Antibody Specificity Controls

Immunohistochemistry of Opioid Receptors 161

162 Yvonne Schmidt and Halina Machelska

4 Notes

antibody with the respective immunizing peptide: add the peptide to the primary antibody solution in a five- to tenfold excess (or according to the instructions on the data sheet) and leave it on a shaker at room temperature for at least 3 h before applying the solution to a slide (see Note 9). The lack of the staining in this experiment will confirm that the antibody selectively binds to its commercial immunizing peptide. However, this does not guarantee specific staining of the native protein, and it is now clear that the preabsorption test must be supported by additional control experiments (see Note 10).

3. Cell lines with and without a protein of interest can be employed; we have used human embryonic kidney (HEK) 293 cells [20]. Transiently transfect HEK 293 cells with plasmids containing the full-length cDNA (approximately 2 μg) of the respective mouse opioid receptor. Ideally, transfect HEK 293 cells with μ, δ or κ-opioid receptor cDNA, respectively. Use a transfection agent (e.g., X-tremeGENE HP DNA transfection reagent; Roche) following the protocol of the manufacturer. Test the opioid receptor antibody in question on all transfected and on untransfected HEK 293 cells. Wash cells in 1× PBS (in a cell plate dish), fix them in 4 % PFA and 4 % sucrose in PBS for 15 min at room temperature, wash again, and permeabilize in 0.25 % TritonX-100 in PBS for 5 min. Wash again and block cells with 10 % BSA in PBS for 30 min at 37 °C, and incubate with the primary antibody in 3 % BSA/PBS for 2 h at 37 °C. After washing, incubate the sections with a fluorophore- conjugated secondary antibody in 3 % BSA/PBS for 45 min at 37 °C. Wash again and mount in Mowiol. However, the data obtained from experiments using cell lines might also not be predictive for post-in vivo antibody staining (see Note 11).

4. Ideally, use tissue from animals genetically lacking the opioid receptor in question (see Note 12). We used tissue from μ-, δ-, and κ-receptor knockout mice (see Note 2). To check for possible cross-reaction of a given opioid receptor antibody with the other opioid receptors, it is recommended to use tissue from mice lacking only one opioid receptor type of interest.

1. We prepared the phosphate buffer and Zamboni’s fixative according to the instructions kindly provided by Dr. Stanley J. Watson (Department of Pharmacology, University of Michigan Medical School, Ann Arbor, MI, USA).

2. Regardless of the antibody specificity, there is also no guaran- tee that each batch of the same antibody type will produce

Immunohistochemistry of Opioid Receptors 163

satisfactory staining quality. We as well as other researches had batches producing good intensity staining and batches producing no staining at all [20, 22]. The μ, δ, and κ-opioid receptor knockout mice were provided by Drs. Brigitte L. Kieffer and Claire Gavériaux-Ruff (Institut de Genetique et de Biologie Moleculaire et Cellulaire, CNRS/INSERM/ULP, Strasbourg, France) [23].

3. This antibody is also commonly used by other researchers, however, we had no opportunity to test it in tissue from mice lacking CGRP.

4. Close the fine forceps and place them underneath the nerve, and lift the nerve up; while doing this, slowly open the forceps (ideally the forceps are wrapped with a rubber band that limits the opening according to the desired width of the ligated nerve part), thereby freeing the nerve part of surrounding tis- sue. We usually place the two outer ligatures first, which results in a twitch of the corresponding hind limb. Then we place the middle ligature; this is often not accompanied by a visible limb twitch.

5. The color change of the liver from red to skin color is nor- mally easy visible and a good sign that the perfusion is working properly.

6. The DRG can be identified within the intervertebral foramen. If necessary, carefully cut the remaining vertebral arch behind the DRG on both sides and remove the arch to have better access to the DRG. Use fine scissors and forceps to free the DRG of interest from the spinal nerves, and place it in the fixa- tive solution.

7. Use between 40 and 50 μl of Mowiol and apply it in a T-shape onto the slide (the shorter line of the T in direction to the left end of the slide). Then, carefully put a cover slip onto the slide and lower it from left to right trying to avoid air bubbles under the slip.

8. The DAB kits we used suggested incubation times between 2 and 10 min, however, we experienced that a shorter time of about 30–60 s is sufficient to develop a clear signal. Longer incubation times may lead to increases in background staining.

9. Per concentration of the antibody, tissue type, and treatment, use at least one slide with the respective opioid receptor anti- body alone and one slide with the opioid receptor antibody preincubated with the corresponding immunizing peptide mixture.

10. The current debate about the lack of specificity of opioid receptor antibodies and G-protein-coupled receptor antibod- ies in general, indicates however, that the disappearance of staining after preabsorption with immunizing peptides is an

164 Yvonne Schmidt and Halina Machelska

insufficient indicator for specific labeling in immunohisto- chemistry [19, 24]. We have made similar observations. Thus, although the preabsorption of the anti-μ-receptor (Ab10275; Abcam) with the μ-receptor peptide (Abcam) resulted in the absence of the staining, the antibody similarly stained DRG from wild-type and μ-, δ-, and κ-receptor knockout mice [20].

11. Although in our experiments the anti-μ-receptor positively stained HEK 293 cells transfected with the mouse μ-receptor and did not stain untransfected or δ-receptor transfected HEK 293 cells, it was still not specific to μ receptors in mouse DRG [20] (see Note 10).

12. Since the antibody specificity might depend on the tissue type [19, 20], we strongly recommend to test all tissues of interest both in wild type and knockout animals. Per concentration of the antibody and per tissue type, use at least one slide with opioid receptor knockout tissue and one slide with opioid receptor wild-type tissue.

This work was supported by the Deutsche Forschungsgemeinschaft grant (MA 2437/1-4; H.M).

Acknowledgements

References

1. Zollner C, Stein C (2007) Opioids. Handb Exp Pharmacol 177:31–63

2. Zhang X, Bao L, Arvidsson U et al (1998) Localization and regulation of the delta-opioid receptor in dorsal root ganglia and spinal cord of the rat and monkey: evidence for association with the membrane of large dense-core vesi- cles. Neuroscience 82:1225–1242

3. Minami M, Maekawa K, Yabuuchi K et al (1995) Double in situ hybridization study on coexistence of mu-, delta- and kappa-opioid receptor mRNAs with preprotachykinin A mRNA in the rat dorsal root ganglia. Brain Res Mol Brain Res 30:203–210

4. Li JL, Ding YQ, Li YQ et al (1998) Immunocytochemical localization of mu-opioid receptor in primary afferent neurons containing substance P or calcitonin gene-related peptide. A light and electron microscope study in the rat. Brain Res 794:347–352

5. Mousa SA, Cheppudira BP, Shaqura M et al (2007) Nerve growth factor governs the enhanced ability of opioids to suppress inflam- matory pain. Brain 130:502–513

6. Hassan AH, Ableitner A, Stein C et al (1993) Inflammation of the rat paw enhances axonal transport of opioid receptors in the sciatic nerve and increases their density in the inflamed tissue. Neuroscience 55:185–195

7. Li JL, Kaneko T, Mizuno N (1996) Effects of peripheral nerve ligation on expression of mu- opioid receptor in sensory ganglion neurons: an immunohistochemical study in dorsal root and nodose ganglion neurons of the rat. Neurosci Lett 214:91–94

8. Kondo I, Marvizon JC, Song B et al (2005) Inhibition by spinal mu- and delta-opioid ago- nists of afferent-evoked substance P release. J Neurosci 25:3651–3660

9. Zollner C, Shaqura MA, Bopaiah CP et al (2003) Painful inflammation-induced increase in mu-opioid receptor binding and G-protein coupling in primary afferent neurons. Mol Pharmacol 64:202–210

10. Stein C, Machelska H (2011) Modulation of peripheral sensory neurons by the immune sys- tem: implications for pain therapy. Pharmacol Rev 63:860–881

11. Guan Y, Johanek LM, Hartke TV et al (2008) Peripherally acting mu-opioid receptor agonist attenuates neuropathic pain in rats after L5 spinal ner ve injur y. Pain 138: 318–329

12. Kabli N, Cahill CM (2007) Anti-allodynic effects of peripheral delta opioid receptors in neuropathic pain. Pain 127:84–93

13. Obara I, Przewlocki R, Przewlocka B (2004) Local peripheral effects of mu-opioid receptor agonists in neuropathic pain in rats. Neurosci Lett 360:85–89

14. Obara I, Parkitna JR, Korostynski M et al (2009) Local peripheral opioid effects and expression of opioid genes in the spinal cord and dorsal root ganglia in neuropathic and inflammatory pain. Pain 141:283–291

15. Labuz D, Machelska H (2013) Stronger anti- nociceptive efficacy of opioids at the injured nerve trunk than at its peripheral terminals in neuropathic pain. J Pharmacol Exp Ther 346:535–544

16. Labuz D, Schmidt Y, Schreiter A et al (2009) Immune cell-derived opioids protect against neuropathic pain in mice. J Clin Invest 119:278–286

17. Michel MC, Wieland T, Tsujimoto G (2009) How reliable are G-protein-coupled receptor antibodies? Naunyn Schmiedebergs Arch Pharmacol 379:385–388

18. Scherrer G, Imamachi N, Cao YQ et al (2009) Dissociation of the opioid receptor mecha- nisms that control mechanical and heat pain. Cell 137:1148–1159

19. Jositsch G, Papadakis T, Haberberger RV et al (2009) Suitability of muscarinic acetylcholine receptor antibodies for immunohistochemistry evaluated on tissue sections of receptor gene- deficient mice. Naunyn Schmiedebergs Arch Pharmacol 379:389–395

20. Schmidt Y, Gavériaux-Ruff C, Machelska M (2013) μ-Opioid receptor antibody reveals tissue-dependent specific staining and increased neuronal μ-receptor immunoreactivity at the injured nerve trunk in mice. PLoS One 8:e79099

21. Rigaud M, Gemes G, Barabas ME et al (2008) Species and strain differences in rodent sciatic nerve anatomy: implications for studies of neu- ropathic pain. Pain 136:188–201

22. Couchman JR (2009) Commercial antibodies: the good, bad, and really ugly. J Histochem Cytochem 57:7–8

23. Karaji AG, Reiss D, Matifas A et al (2011) Influence of endogenous opioid systems on T lymphocytes as assessed by the knockout of mu, delta and kappa opioid receptors. J Neuroimmune Pharmacol 6:608–616

24. Saper CB (2005) An open letter to our readers on the use of antibodies. J Comp Neurol 493:477–478

Immunohistochemistry of Opioid Receptors 165

Part III Analysis of Signaling Events Modulated by Opioid Receptors

Chapter 13

[35S]GTPγS Autoradiography for Studies of Opioid Receptor Functionality

Alfhild Grönbladh and Mathias Hallberg Abstract

The opioid receptors have been an interesting target for the drug industry for decades. These receptors were pharmacologically characterized in the 1970s and several drugs and peptides have emerged over the years. In 2012, the crystal structures were also demonstrated, with new data on the receptor sites, and thus new possibilities will appear. The role of opioids in the brain has attracted considerable interest in several diseases, especially pain and drug dependence. The opioid receptors are G-protein-coupled receptors (GPCR) that are Gi-coupled which make them suitable for studying the receptor functionality. The [35S] GTPγS autoradiography assay is a good option that has the benefit of generating both anatomical and functional data in the area of interest. It is based on the first step of the signaling mechanism of GPCRs. When a ligand binds to the receptor GTP will replace GDP on the α-subunit of the G protein, leading to a dissociation of the βγ-subunit. These subunits will start a cascade of second messengers and subsequently a physiological response.

Key words Brain, Functional autoradiography, G proteins, GTPgammaS, Opioid receptors

1 Introduction

The marketed drugs in the world today focus on over 100 different targets. Most of these are enzymes and receptors, whereas the G-protein-coupled receptors (GPCR) seem to be the largest and predominating target family. Approximately 800 human GPCRs have been verified and reported [1]. Five main families of human GPCRs have been reported. These are termed Adhesion, Frizzled/ Taste2, Glutamate, Rhodopsin, and Secretin [2]. A majority of these are not fully characterized. Interestingly, today many of the newly discovered GPCRs lack natural ligands [3]. In the last years GPCRs have been particularly highlighted through the Nobel prize in Chemistry in 2012 since Robert J. Lefkowitz and Brian K. Kobilka shared the prize for their work on GPCRs. Furthermore, with regards to opioids, in the same year the first crystal structures of the mu, delta and kappa receptors binding to various ligands

Santi M. Spampinato (ed.), Opioid Receptors: Methods and Protocols, Methods in Molecular Biology, vol. 1230, DOI 10.1007/978-1-4939-1708-2_13, © Springer Science+Business Media New York 2015

169

170 Alfhild Grönbladh and Mathias Hallberg

were disclosed [4–6]. For the mu opioid receptor, the binding sites were reported to be wider and larger than previously estimated. Thus, these data bring more understanding to how alkaloids such as morphine and large peptides such as β-endorphin can bind and activate the same receptor.

The opioid receptors that were cloned during the 1990s [7–10] are GPCRs that originally were classified based on their pharmaco- logical profile as well as tissue distribution. The opioid receptors are divided into three different subclasses: the mu, delta, and kappa opioid receptors. These receptor subclasses all contain seven trans- membrane helixes and one of the intracellular loops interacts with G proteins. There are four different subclasses of G proteins: Gs, Go, Gq, and Gi. Overall, different receptors bind to specific G proteins. The focus of this protocol is however on the opioid recep- tors which are all Gi-coupled.

Opioid peptides as well as opioid receptors are involved in the processing of pain and also in several behavioral processes such as dependence, reward, stress, and sedation. To understand the roles of the opioid system it is important to study new ligands, the func- tionality as well as the distribution of the receptors. Nevertheless, when developing new ligands or studying a certain receptor system it is important to have a reliable functional assay. In the case of GPCR signaling, the first step is mediated through G proteins. These are well characterized and can be studied through assays. The GTPγS binding assay is a suitable technique for determining whether in particular Gi-coupled receptors are activated by differ- ent potential agonists. The [35S]GTPγS binding assay was first described in the 1980s for β-adrenergic and muscarinic receptors [11, 12]. The technique was later also shown to be useful in order to map the anatomical distributions of receptors, including the mu opioid receptor [13, 14]. Comprehensive reviews of the develop- ment of the [35S]GTPγS binding assay as well as applications have been published over the years [15–18].

The GTPγS assay measures the increase in the guanine nucleo- tide exchange in the heterotrimeric G protein. Briefly, in the absence of ligand the receptor will be inactive and the α-units of the G proteins will be bound to guanosine diphosphate (GDP). When the receptor is ligand-activated, guanosine triphosphate (GTP) will replace the GDP, which will cause the GTP-bound α-subunit to dissociate from the βγ-subunit (see Fig. 1). These sub- units will later trigger new targets and thus activate the second messenger system. By replacing the GTP with the nonhydrolyzable [35S]GTPγS and by administering high concentrations of GDP in the assay, a ligand activated response can be measured.

This chapter will describe the GTPγS binding assay. We aim to highlight crucial factors with regard to the GTPγS binding assay and will have a focus on the brain of rodents (see Note 1).

[35S]GTPγS Autoradiography for Studies of Opioid Receptor Functionality 171

Fig. 1 Binding of the agonist to the active receptor site generates an exchange of GDP for GTP and subse- quently an activation of the second messenger system. (1) An agonist binding to the receptor; (2) Guanosine triphosphate (GTP) will replace the GDP; (3) The GTP bound α-subunit dissociates from the βγ-subunit and activates a target protein. These subunits will later trigger new targets and thus activate the second messen- ger system. (4) The GDP/α-subunit will “return” to a receptor. The GTPγS assay utilize nonhydrolyzable [35S] GTPγS instead of GTP to measure the activity

2 Materials

2.1 Coating of Slides

2.1.1 Preparation
of a Gelatin Buffer Containing 0.5 % Gelatin, 0.05 % KCr(SO4)2

2.1.2 Cryostat/ Sectioning

We retain that the GTPγS binding assay or GTPγS autoradiography is a well functioning and reliable assay. The technique is straight- forward and suitable for the Gi-coupled opioid receptors.

1. Add the gelatin to q.s. 100 % water and heat to approximately 60 °C. Dissolve the 0.05 % KCr(SO4)2 in a separate beaker.

2. Mix the two solutions when the gelatin solution is ready and has started to cool down.

3. Dip the slides when the solution has reached room tempera- ture and dry under a fan before storage in their original cartons at −20 °C until use (see Note 2).

The sectioning is done with a cryostat at −20 °C. In comparison with ordinary autoradiography where brain slices between 10 and 14 μm are usually used the GTPγS slices are thicker. In most research groups 20 μm is selected for the GTPS assay. This is suf- ficient to receive a quality picture of the receptor functionality and still enough slices to get a reasonable number of autoradiograms from each region (see Notes 3–7).

172 Alfhild Grönbladh and Mathias Hallberg

Fig. 2 Representative autoradiograms from rat brain (coronal sections, bregma 1.60 mm) displaying the (a) basal binding, (b) DAMGO-stimulated [35S]GTPγS, and (c) unspecific binding, i.e., basal binding in the pres- ence of cold GTP

. 2.1.3 Assay Buffer

. 2.1.4 Washing Buffer

3 Methods

3.1 Preparation of Brain (Tissue)

3.2 GTPγS Incubations

Mount the slices on the gelatin-coated slides (see Note 8). For a rat brain (with coronal slicing) we usually mounted 3–4 slices/ slide but this of course differs depending on species, aim of the autoradiography or the use of sagittal or coronal frozen sections (an example of coronal slices is shown in Fig. 2).

Dry the glass slides completely before storing at −80 °C (see Note 9). Drying the slides overnight is commonly preferred.

Prepare the assay buffer: 50 mM Tris–HCl, 4 mM MgCl2, 0.3 mM EGTA and 100 mM NaCl (pH 7.4) at room temperature.

Prepare the washing buffer: 50 mM Tris–HCl (pH 7.4 at 4 °C) (see Note 10).

Carry out all procedures at room temperature unless otherwise specified.

After euthanizing (usually by decapitation) of the animals the whole brains (tissue) are rapidly removed. The brains, or tissue, are fresh frozen in isopentane (−35±5 °C) for 30 s (see Note 11). After being frozen, the tissue is stored at −80 °C until further use.

1. Incubate the slides in the assay buffer for 10 min at room temperature.

2. Move the slides into a humid chamber and incubate the slides in the assay buffer containing 10 mU/ml adenosine deaminase (ADA) and 2 mM GDP in room temperature for 15 min (see Note 12).

3. Add 0.04 nM [35S]GTPγS to the assay buffer (containing 10 mU/ml ADA and 2 mM GDP) and stimulate the receptors with the ligands of choice (e.g., DAMGO, DPDPE, and CI-977) for the respective subtypes. Incubate for 2 h in 25 °C (see Notes 13–17).

3.3 Film or Phosphor Imaging

3.4 Analysis of Data

2 min. Repeat the step.

5. Wash the slides with distilled water for 30 s.

6. Dry the slides with a fan until they are completely dry.

7. In order to confirm the selectivity of the stimulation, an antagonist (e.g., naloxone (1 μM)) should be added on the subsequent slides.

8. Under the same conditions as described above, the basal levels are determined in the absence of agonist. The nonspecific binding is determined by incubating the sections in the pres- ence of unlabeled GTPγS. All procedures and incubation steps should be as described above.

After being dried overnight, the sections are exposed to film together with the standards (microscales) for 2–3 days, or longer depending on the ligand. In our laboratory, we recently used the Kodak BioMax MR-1 film. The films are developed either manu- ally or using a film processor, for example the Konica medical film processor SRX-101A (Konica Europe GmbH, Hohenbrunn, Germany) [19].

The slides are digitalized (e.g., using an Epson Perfection 4870 photo scanner) and analyzed using a program suitable for deter- mining the receptor functionality. An example of representative figures is demonstrated in Fig. 2. In our lab and others, we use the Image J (National Institutes of Health, Bethesda, MD, USA) image processing software [19–21]. The optical densities can be converted to nCi/g using a standard curve calculated from the [14C]-microscales, previous described by Sim et al. [20].

1. This protocol is fully devoted to the GTPγS autoradiography on slices. However, the technique is also easily utilized on membrane preparations from cells or different brain homoge- nates. In these cases, some parameters are different. Usually much lower concentrations of GDP can be used.

2. Regarding the coating of slides: There is a risk of getting gelatin edges on the glass slides during and after the coating procedure. Try to slightly shake the racks occasionally when drying the glass slides.

3. The frozen brains or tissues can be stored in different ways. The brains or tissues can either be wrapped in cold foil, placed in small plastic bags and put on dry ice or placed in a plastic container with 2–3 ml of isopentane (<−35 °C).

[35S]GTPγS Autoradiography for Studies of Opioid Receptor Functionality 173 4. Wash the slides with ice-cold 50 mM Tris–HCl, pH 7.4 for

4 Notes

174 Alfhild Grönbladh and Mathias Hallberg

4. Use a sharp knife for the cryostat-cutting. A good motto is to change the knife every third brain, of course depending on the number of slices from each brain.

5. Problems with electrostatics can occur depending on humidity in the laboratory. We usually clean the equipment with ethanol.

6. Gentle handling or movement of the slices can well be done with tiny paintbrushes. Avoid metal parts on the paint brush since the knife usually is attached with a magnet.

7. A proper order and similar distances between the slices on the slide will ease and facilitate the following steps, especially the analysis after digitalization.

8. Use a waterproof and permanent pen when marking the slides.

9. The number of functional receptors will decrease over time in the freezer. We usually try to conduct our experiments within 3 months of the decapitation. There is a marked difference in expression when comparing 1-month-old tissue with for instance 6 months old tissues.

10. If the washing buffer (50 mM Tris–HCl) is prepared in room temperature, adjust the pH to approximately 6.8 and check again when ice-cold; should be pH 7.4 at 4 °C.

11. When lowering the dissected brain into the isopentane (−35 °C) be careful not to drop it since the brain will be deformed just before freezing.

12. High concentrations of GDP are required to reduce the basal binding. The increasing concentration of GDP inactivates the α-units in the G protein. Thus, the GPCR is shifted from an active state to an inactive state. Concentrations are recom- mended to be as high as 1–2 mM for GTPγS autoradiography.

13. Before adding the buffers to the slides in the incubation cham- ber, it can be good to circle the slices with a PAP pen. This will make the solution stay on slides and also decrease the area, i.e., the amount of assay buffer.

14. If an incubation chamber is used, it is important to make sure it is absolutely horizontal. If not, part of the slides can dry during the incubation procedure. For the same reason, a high humidity in the chamber is also recommended.

15. Depending on the receptor, several different ligands can be chosen. For the mu opioid receptor, DAMGO is probably the most characterized ligand in GTPγS studies. Different mu opioid ligands such as DAMGO and endomorphin-1 (EM-1) have, as expected, been reported to display the same anatomi- cal distribution of receptors in the brain. However, different ligands can display variations in the stimulation. For example, EM-1 does only generate approximately 60 % of the stimula- tion in comparison to DAMGO [22].

[35S]GTPγS Autoradiography for Studies of Opioid Receptor Functionality 175

Acknowledgements

References

1. Gloriam DE, Fredriksson R, Schioth HB (2007) The G protein-coupled receptor subset of the rat genome. BMC Genomics 8:338

2. Fredriksson R, Lagerstrom MC, Lundin LG et al (2003) The G-protein-coupled receptors in the human genome form five main families. Phylogenetic analysis, paralogon groups, and fingerprints. Mol Pharmacol 63:1256–1272

3. Civelli O, Reinscheid RK, Zhang Y et al (2013) G protein-coupled receptor deorphanizations. Annu Rev Pharmacol Toxicol 53:127–146

4. Granier S, Manglik A, Kruse AC et al (2012) Structure of the delta-opioid receptor bound to naltrindole. Nature 485:400–404

10. Yasuda K, Raynor K, Kong H et al (1993) Cloning and functional comparison of kappa and delta opioid receptors from mouse brain. Proc Natl Acad Sci U S A 90:6736–6740

11. Asano T, Pedersen SE, Scott CW et al (1984) Reconstitution of catecholamine-stimulated binding of guanosine 5′-O-(3-thiotriphosphate) to the stimulatory GTP-binding protein of ade- nylate cyclase. Biochemistry 23:5460–5467

12. Kurose H, Katada T, Haga T et al (1986) Functional interaction of purified muscarinic receptors with purified inhibitory guanine nucleotide regulatory proteins reconstituted in phospholipid vesicles. J Biol Chem 261: 6423–6428

16. When studying the kappa opioid receptors, U69-593 or CI-977 is commonly used. However, since these receptors have low abundance in the brain (rat and mice) it is more difficult to generate a measurable stimulation in comparison to the delta and mu opioid receptors.

17. When using the GTPγS assay with membranes from cells or tissue, some ligands such as beta-endorphin seem to generate an overshoot in stimulation. In these cases, it is important to evaluate whether the effect is reversible by an antagonist.

This work was supported by the Kjell and Märta Beijer Foundation and Swedish Medical Research Council (grant 9459).

5. Manglik A, Kruse AC, Kobilka TS et al (2012)
Crystal structure of the μ-opioid receptor 13.SimLJ,SelleyDE,ChildersSR(1995)Invitro

bound to a morphinan antagonist. Nature 485:

321–326

6. Wu H, Wacker D, Mileni M et al (2012)
Structure of the human kappa-opioid receptor
in complex with JDTic. Nature 485:327–332

7. Evans CJ, Keith DE Jr, Morrison H et al (1992) Cloning of a delta opioid receptor by func-
tional expression. Science 258:1952–1955

8. Kieffer BL, Befort K, Gaveriaux-Ruff C et al (1992) The delta-opioid receptor: isolation of a cDNA by expression cloning and pharmaco- logical characterization. Proc Natl Acad Sci
U S A 89:12048–12052

9. Chen Y, Mestek A, Liu J et al (1993) Molecular
cloning and functional expression of a mu- opioid receptor from rat brain. Mol Pharmacol 44:8–12

autoradiography of receptor-activated G pro- teins in rat brain by agonist-stimulated guanylyl 5′-[gamma-[35S]thio]-triphosphate binding. Proc Natl Acad Sci U S A 92:7242–7246

14. Sim LJ, Selley DE, Dworkin SI et al (1996) Effects of chronic morphine administration on mu opioid receptor-stimulated [35S]GTP gam- maS autoradiography in rat brain. J Neurosci 16:2684–2692

15. Sovago J, Dupuis DS, Gulyas B et al (2001) An overview on functional receptor autoradiogra- phy using [35S]GTPgammaS. Brain Res Brain Res Rev 38:149–164

16. Harrison C, Traynor JR (2003) The [35S] GTPgammaS binding assay: approaches and applications in pharmacology. Life Sci 74: 489–508

176 Alfhild Grönbladh and Mathias Hallberg

17. Milligan G (2003) Principles: extending the utility of [35S]GTP gamma S binding assays. Trends Pharmacol Sci 24:87–90

18. Strange PG (2010) Use of the GTPgammaS ([35S]GTPgammaS and Eu-GTPgammaS) binding assay for analysis of ligand potency and efficacy at G protein-coupled receptors. Br J Pharmacol 161:1238–1249

19. Gronbladh A, Johansson J, Nyberg F et al (2013) Recombinant human growth hormone affects the density and functionality of GABAB receptors in the male rat brain. Neuroendo- crinology 97:203–211

20. Sim LJ, Selley DE, Childers SR (1997) Autoradiographic visualization in brain of recep- tor-G protein coupling using [35S]GTP gamma S binding. Methods Mol Biol 83:117–132

21. Johansson J, Gronbladh A, Nyberg F (2013) Application of in vitro [(35)S]GTPgamma-S autoradiography in studies of growth hormone effects on opioid receptors in the male rat brain. Brain Res Bull 90:100–106

22. Sim LJ, Liu Q, Childers SR (1998) Endomorphin- stimulated [35S]GTPgammaS binding in rat brain: evidence for partial agonist activity at mu- opioid receptors. J Neurochem 70:1567–1576

Chapter 14

Fluorescence-Based, High-Throughput Assays for μ-Opioid Receptor Activation Using
a Membrane Potential-Sensitive Dye

Alisa Knapman and Mark Connor Abstract

The development of new and improved opioid analgesics requires high-throughput screening (HTS) methods to identify potential therapeutics from large libraries of lead compounds. Here we describe two simple, real-time fluorescence-based assays of μ-opioid receptor activation that may be scaled up for HTS. In AtT-20 cells expressing the μ-opioid receptor (MOPr), opioids activate endogenous G protein gated inwardly rectifying K channels (GIRK channels), leading to membrane hyperpolarization. In Chinese hamster ovary cells expressing MOPr, adenylyl cyclase activation via forskolin results in membrane hyper- polarization, which is inhibited by opioids. Changes in membrane potential can be measured using a proprietary membrane potential-sensitive dye. In contrast to many HTS methods currently available, these assays reflect naturalistic coupling of the receptor to effector molecules.

Key words Adenylyl cyclase, AtT-20, CHO, Fluorescent assay, GIRK, High-throughput screening, Membrane potential, μ-Opioid receptor

1 Introduction

The μ-opioid receptor (MOPr) is a G-protein-coupled receptor (GPCR), and is the primary target for opioid analgesics [1]. There is increasing evidence that different ligands can stabilize GPCRs including MOPr in various active conformations, leading to the preferential activation of distinct signaling pathways via Gα and Gβγ subunits [2, 3]. The development of novel opioid pharmaco- therapies is now beginning to focus on identifying MOPr ligands that can selectively activate a subset of effector pathways associated with analgesia, without activating pathways leading to adverse effects [4–7].

Drug development typically involves the screening of large libraries of lead compounds to identify those capable of binding to and signaling via a receptor. The vast array of lead compounds available requires high-throughput screening (HTS) methods to

Santi M. Spampinato (ed.), Opioid Receptors: Methods and Protocols, Methods in Molecular Biology, vol. 1230, DOI 10.1007/978-1-4939-1708-2_14, © Springer Science+Business Media New York 2015

177

178 Alisa Knapman and Mark Connor

2 Materials

enable identification of potential therapeutic compounds. Many of the HTS methods available for opioid receptors involve the expres- sion of highly engineered proteins, require harvesting or lysing cells, and rely on a single endpoint measurement [8, 9]. Here, we describe two simple, real-time assays reflecting a naturalistic cou- pling of MOPr to Gα and Gβγ proteins, using a proprietary mem- brane potential-sensitive dye.

In mouse pituitary AtT-20 cells, heterologously expressed MOPr activates endogenous G-protein-gated inwardly rectifying potassium channels (GIRKs), via direct coupling of Gβγ subunits to the channels [10]. The resulting membrane hyperpolarization can be measured using a membrane potential-sensitive dye. The extent of hyperpolarization corresponds to the number of activated GIRK channels, and therefore closely reflects the degree of MOPr activation [11].

Inhibition of AC activity leading to a decrease in cAMP production is one of the hallmarks of MOPr activation and mea- surement of cAMP accumulation is frequently used to examine opioid potency and efficacy. Most cAMP accumulation assays rely on endpoint measurements after significant incubation times to allow cAMP accumulation [9, 12, 13], while real-time assays of cAMP accumulation such as the GloSensor assay (Promega) require transfection of sensor constructs with cAMP binding domains and use of specialized reagents [14]. The assay described here is a real-time, robust assay of AC inhibition in CHO cells requiring minimal preparation and reagents. Stimulation of cAMP production Chinese hamster ovary (CHO) cells via the AC activa- tor forskolin results in membrane hyperpolarization, which is inhibited with the simultaneous addition of opioids [15].

Both assays described here have z-factors of 0.7, indicating suf- ficient robustness for HTS [11, 15, 16]. These assays offer an alter- native approach for measuring MOPr activation by targeting naturalistic signaling pathways. The membrane potential assay is a rapid, reliable, and inexpensive method for identifying ligands that modulate Gα- and Gβγ-mediated signaling and may be scaled up to enable HTS for novel opioid drugs.

1. FlexStation3 Plate Reader, running at 37 °C (see Note 1).

2. Incubator with room air, at 37 °C.

3. FLIPR Membrane Potential Dye (Molecular Devices)—blue rather than red (see Note 2).

4. Assay Buffer—consisting of (in mM): NaCl 145, HEPES 22, Na2HPO4 0.338, NaHCO3 4.17, KH2PO4 0.441, MgSO4

3 Methods

0.407, MgCl2 0.493, CaCl2 1.26, Glucose 5.56, pH 7.4, osmolarity 310-320.

5. AtT-20 cells expressing MOPr (GIRK channel activation assay), or CHO-K1 cells expressing MOPr (AC inhibition assay) at 90 % or more confluency in 75 cm2 tissue culture flasks (1 flask per microplate).

6. Leibovitz’s L-15 media supplemented with 1 % FBS, 100 U penicillin/streptomycin per mL and 15 mM glucose (see Note 3).

7. 96-Well black-walled, sterile clear-bottomed microplates.

8. 96-Well clear, v-bottomed microplates.

9. 96-Well Black FlexStation pipette tips (Molecular Devices).

10. 8-Channel multi-pipette.

11. Troughs for holding cells and dye while using multichannel pipette.

12. Opioid ligands.

13. Forskolin (AC inhibition assay).

On the day before the assay, harvest cells from single 75 cm2 flask, re-suspend the cells obtained from each flask in 10 mL L-15 media, and plate in a volume of 90 μL/well in a black-walled 96-well plate using an eight-channel pipettor. Incubate overnight at 37 °C in ambient CO2 (see Note 4).

Prepare the dye according to the manufacturer’s instructions in assay buffer. Prepared dye can be stored at −80 °C for several months (see Note 5).

1. Load the cells with 90 μL/well prepared membrane potential dye. Incubate for a minimum of 45 min at 37 °C in ambient CO2. Ensure the plate is uncovered for the final 15 min of incubation (see Notes 6–8).

2. Prepare drug solutions: Make up drugs in assay buffer at 10× final concentration desired. See Note 9. Load 200 μL/well drug solutions or vehicle into wells in a v-bottom 96-well plate (see Notes 10 and 11).

Insert drug plate into appropriate FlexStation drawer and incu- bate for approx. 10 min to warm solutions to 37 °C (see Note 12).

1. Read Mode: Fluorescence, bottom read. 2. Excitation wavelength: 530.
3. Emission wavelength: 565.

3.1 GIRK Channel Activation Assay (AtT-20 Cells)

3.1.1 Set Up
the Assay Parameters

Fluorescent Membrane Potential Determination 179

180

Alisa Knapman and Mark Connor

3.1.2

Compound Transfer

4. Cutoff: Auto.
5. Readings per well: 6.
6. PMT: Medium.
7. Run time: 300 s.
8. Interval: 2 s.
9. Select appropriate assay plate type and wells to read. See Note 13.

1. Initial volume: 180 μL. 2. Transfers: 1.
3. Pipette height: 190 μL. 4. Volume: 20 μL.

5. Rate: 2.

6. Time point: 120 s—this will allow a stable baseline to be estab- lished before drug addition.

7. Select appropriate compound plate.

8. Triturate: Select Assay plate, volume 20 μL, 3 cycles, pipette height 150 μL. Pipette Tip Layout: Select columns of tips to be used corresponding to columns of wells to be read. Tips may be re-used up to three times for replicates.

9. Select “Read” (see Notes 14–17).
Load the cells with 90 μL/well prepared membrane potential dye.

Incubate for a minimum of 60 min at 37 °C in ambient CO2 (see Notes 6–8).

1. Forskolin is added simultaneously with drug or vehicle. Make up drugs with forskolin in assay buffer at 10× final concentration desired, e.g. 3 μM FSK + 10 nM–10 μM DAMGO for a 1 nM–1 μM DAMGO concentration–response curve. See Note 9. Load 200 μL/well drug solutions or vehicle into wells in a v-bottom 96-well plate (see Notes 10 and 11).

2. The FlexStation 3 reads column by column, so concentration– response curves are best set up within columns rather than across rows. Insert drug plate into appropriate FlexStation drawer and incubate for approx. 10 min to warm solutions to 37 °C (see Note 12).

1. Read Mode: Fluorescence, bottom read. 2. Excitation wavelength: 530.
3. Emission wavelength: 565.
4. Cutoff: Auto.

5. Readings per well: 6.

3.2
Inhibition Assay (CHO Cells)

3.2.1 Drug Solutions Administration

Adenylyl Cyclase

3.2.2 Set Up the Assay Parameters

3.2.3 Compound Transfer

6. PMT: Medium.
7. Run time: 720 s.
8. Interval: 2 s.
9. Select appropriate assay plate type and wells to read (see Note 13).

1. Initial volume: 180 μL. 2. Transfers: 1.
3. Pipette height: 190 μL. 4. Volume: 20 μL

5. Rate: 2

6. Time point: 120 s—this will allow a stable baseline to be estab- lished before drug addition.

7. Select appropriate compound plate.

8. Triturate: Select Assay plate, volume 20 μL, 3 cycles, pipette height 150 μL Pipette Tip Layout: Select columns of tips to be used corresponding to columns of wells to be read. Tips may be re-used up to three times for replicates.

9. Select “Read” (see Notes 13–15 and 17–19).

1. We used a FlexStation 3 for this assay, however a similar plate reader with a fluidic module could be used for the GIRK assay. The AC assay could potentially be performed as an endpoint assay on a static plate reader approximately 5 min after drug treatment, providing a 300 nM FSK control is included in each column.

2. The membrane potential-sensitive dye is available in a blue or red formulation. We found blue dye to give the largest signal window for our cell lines, however, this may vary. To ensure optimal results both blue and red dyes should be trialed, par- ticularly if adapting the assays for other cell types.

3. Serum-starving the cells synchronizes the cell cycle phase and helps to increase reproducibility of the results. Serum starv- ing also prevents cells overgrowing and forming multiple lay- ers in wells.

4. For the cell lines used in these assays, the use of polylysine coated plates was not necessary. We have found that experi- ments with other AtT-20 lines, such as those expressing the cannabinoid-CB1 receptor, require polylysine-coated plates to prevent cells detaching.

Fluorescent Membrane Potential Determination 181

4 Notes

182 Alisa Knapman and Mark Connor

5. Membrane potential dye (blue) can be successfully used at half the concentration suggested by the manufacturer for these assays with no appreciable difference in signal, however we observed decreased stability of half-strength dye with longer assays (>20 min). We have not tested red membrane potential dye at half-concentration.

6. Although the manufacturer’s protocol recommends an incuba- tion time of 30 min with dye, we found that when readings were taken after 30 min incubation the baseline steadily increased over time, indicating incomplete uptake of dye into cells. An incubation of 45 min was sufficient to achieve a flat baseline for AtT-20 cells, a minimum of 60 min was necessary for CHO cells. This may vary between cell lines and assay conditions. Incubate for sufficient time period to achieve a flat baseline.

7. For the GIRK channel activation assay, an entire plate may be read in approximately 1 h, so it is practical to load an entire plate with dye at once with no deterioration of signal. For longer assays such as the AC inhibition assay, dye loading can be stag- gered so that cells are not loaded with dye for an extended period of time. In this case, reading of half a plate takes approxi- mately 80 min, so load half a plate with dye, incubate for 80 min, then immediately prior to reading load the second half of the plate with dye and cover with parafilm, allowing dye to load in the second half of the plate while the first half is being read.

8. Where possible, the assay plate should be incubated in the FlexStation 3 while the dye is loading to minimize temperature changes when transferring between the incubator and Flexstation (in our lab these are located in different rooms). When this is not possible, the assay plate may be incubated in an incubator, and transferred to the FlexStation for the last 15 min of incubation, uncovered.

9. When making up drug solutions, keep the concentration of any solvents used (e.g. DMSO or ethanol) constant. Also include the same concentration of solvent in the vehicle blank.

10. There should be a minimum of 80 μL of compound in the wells of the compound plate to ensure consistency in com- pound transfer volume. Compound may be taken from the same column for replicates until volume in well reaches 60 μL.

11. For relatively short assays, no changes in response over time were detected when reading an entire plate, i.e. there appears to be little effect of evaporation of drug or dye from com- pound and assay plates. For the AC inhibition assay, the com- pound plate should be only be loaded with the drugs required for the portion of the plate being read. Load fresh compounds when beginning a new read for the second half of the plate.

Fluorescent Membrane Potential Determination 183

12. Some compounds, for example somatostatin, may be unstable for extended time periods at 37 °C. In this case, drugs should be kept on ice and loaded into compound plate just prior to reading.

13. For short assays, providing you are not using any unstable drugs, you can select and read the entire plate. Otherwise, select only the columns you will read before loading fresh drug solutions. Note that when using SoftMax Pro with the FlexStation 3, only consecutive columns can be read in a single run.

14. For this assay, on our Flexstation, a baseline of between 600 and 1,200 RFU is optimal. These values may vary between machines, but a low baseline does not give a sufficient window for the decreased fluorescent signal associated with cellular hyperpolarization, and a high baseline can lead to unpredict- able results for a number of reasons including too many cells per well, cells which are excessively depolarized or a double addition of dye. More than a single layer of cells can lead to problems with drug access and also loss of cells when drug is added, which leads to a large and immediate drop in signal.

15. The background signals for these assays are low—cells without dye usually read approximately 20–30 RFU (<5 % of the opti- mal signal), while L-15 plus dye without cells gives a negligible reading. We do not routinely correct for these background sig- nals, but they should be checked occasionally by reading a few wells without dye, with and without cells.

16. GIRK Activation Assay Data Analysis: For this assay, we calcu- lated GIRK activation as a percentage decrease in fluorescence from baseline. We calculated the mean of the readings from 30 s before drug addition, ensuring that the baseline is flat and stable. The peak response occurs rapidly after drug addition, usually within 20 s (see Fig. 1).

17. We took the mean of the lowest reading from the peak response and two readings either side, calculating the percentage differ- ence between this value and the mean baseline. We usually include a buffer/solvent blank in each row and subtract the change produced by the 20 μL addition of buffer from all the experimental data. This change is usually less than 5 %.

18. If you want to normalize data between assays, the hyperpolar- ization produced by activation of endogenous SST receptors in AtT-20 cells can be used, alternatively nigericin (1 μM), a K-selective ionophore, can be used to determine the fluores- cence change associated with hyperpolarization of the cells to EK. Both somatostatin and nigericin produce decreases in fluo- rescence greater than the maximum produced by opioids.

19. AC Inhibition Data Analysis: For this assay, we calculated AC inhibition as a percentage inhibition of the FSK response. The

184 Alisa Knapman and Mark Connor

Fig. 1 Example trace of fluorescent signal in AtT-20-MOPr cells with the addition of vehicle (light trace) or 300 nM DAMGO at 120 s. The rapid drop in fluorescent signal (raw fluorescence units, RFU) peaks 20 s after DAMGO addition, at which point measurements were taken

Fig. 2 Example trace of the fluorescent signal in CHO-MOPr cells with the addi- tion of 300 nM FSK alone (light trace) or 300 nM FSK + 1 μM DAMGO. The maximum FSK-stimulated decrease in fluorescent signal (raw fluorescence units, RFU) occurred 5 min after addition, after which time the signal stabilized. Measurements for FSK alone and FSK + opioid were taken at this time-point

FSK response was calculated as the maximum % decrease in fluorescence from baseline after addition of FSK alone. We cal- culated the mean of the readings from 30 s before drug addi- tion, ensuring that the baseline is flat and stable. The peak response of FSK occurs approximately 5 min after drug addi- tion (see Fig. 2). We took the mean of the lowest reading from

References

1. Matthes HWD, Maldonado R, Simonin F et al (1996) Loss of morphine-induced analgesia, reward effect and withdrawal symptoms in mice lacking the μ-opioid-receptor gene. Nature 383:819–823

2. Kenakin T, Miller LJ (2010) Seven transmem- brane receptors as shapeshifting proteins: the impact of allosteric modulation and functional selectivity on new drug discovery. Pharmacol Rev 62:265–304

3. Pinyero G, Archer-Lahlou E (2007) Ligand- specific receptor states: implications for opiate receptor signaling and regulation. Cell Signal 19:8–19

4. DeWire SM, Yamashita DS, Rominger DH et al (2013) A G protein-biased ligand at the μ-opioid receptor is potently analgesic with reduced gastrointestinal and respiratory dys- function compared with morphine. J Pharmacol Exp Ther 344:708–717

5. Kenakin T, Christopoulos A (2013) Signalling bias in new drug discovery: detection, quanti- fication and therapeutic impact. Nat Rev Drug Discov 12:205–216

6. Pradhan AA, Smith ML, Kieffer BL et al (2012) Ligand-directed signalling within the opioid receptor family. Br J Pharmacol 167:960–969

7. Zheng H, Loh HH, Law PY (2013) Post- translation modification of G protein-copuled receptor in relationship to biased agonism. Methods Enzymol 522:391–408

8. Coward P, Chan SDH, Wada HG et al (1999) Chimeric G-proteins allow a high-throughput signaling assay of Gi-coupled receptors. Anal Biochem 270:242–248

9. Hill SJ, Williams C, May LT (2010) Insights into GPCR pharmacology from the mea- surement of changes in intracellular cyclic AMP: Advantages and pitfalls of differing methodologies. Br J Pharmacol 161: 1266–1275

10. Celver J, Xu M, Jin W et al (2004) Distinct domains of the μ-opioid receptor control uncoupling and internalization. Mol Pharmacol 65:528–537

11. Knapman A, Santiago M, Du YP et al (2013) A continuous, fluorescence-based assay of μ-opioid receptor activation in AtT-20 cells. J Biomol Screen 18:269–276

12. Gabriel D, Vernier M, Pfeifer MJ et al (2003) High throughput screening technologies for direct cyclic AMP measurement. Assay Drug Dev Technol 1:291–303

13. Williams C (2004) cAMP detection methods in HTS: Selecting the best from the rest. Nat Rev Drug Discov 3:125–135

14. Binkowski BF, Fan F, Wood KV (2011) Luminescent biosensors for real time monitor- ing of intracellular cAMP. Methods Mol Biol 756:263–271

15. Knapman A, Abodadie F, McIntyre P et al (2014) A real-time, fluorescence-based assay for measuring μ-opioid receptor modulation of adenylyl cyclase activity in Chinese hamster ovary cells. J Biomol Screen 19:223–231

16. Zhang J, Chung TDY, Oldenburg KR (1999) A simple statistical parameter for use in evaluation and validation of high through- put screening assays. J Biomol Screen 4: 67–73

Fluorescent Membrane Potential Determination 185

the peak response and two readings either side, calculating the percentage difference between this value and the mean base- line for the maximum FSK response. We took readings at the same time point when FSK was added with opioid, and calcu- lated the difference between the FSK response and the FSK+opioid response, and expressed as a percentage inhibi- tion of FSK response. We also included a buffer/solvent blank in each row and subtracted the change produced by the 20 μL addition of buffer containing the solvent for FSK from the experimental data. This change is usually less than 5 %.

Chapter 15

Analysis of Potassium and Calcium Imaging to Assay the Function of Opioid Receptors

Viola Spahn, Dinah Nockemann, and Halina Machelska Abstract

As the activation of opioid receptors leads to the modulation of potassium and calcium channels, the ion imaging represents an attractive method to analyze the function of the receptors. Here, we describe the imaging of potassium using the FluxORTM potassium ion channel assay, and of calcium using Fura-2 ace- toxymethyl ester. Specifically, we (1) characterize the activation of the G-protein-coupled inwardly rectify- ing potassium 2 channel by agonists of μ- and δ-opioid receptors with the aid of the FluxORTM assay in cultured mouse dorsal root ganglion neurons, and (2) describe calcium imaging protocols to measure capsaicin-induced transient receptor potential vanilloid 1 channel activity during opioid withdrawal in transfected human embryonic kidney 293 cells.

Key words Calcium imaging, Capsaicin, DAMGO, DPDPE, G-protein-coupled inwardly rectifying potassium channel (GIRK), Morphine, Opioid receptor, Potassium imaging, TRPV1

1 Introduction

In this chapter, we introduce the measurement of opioid receptor activity using microfluorimetry. In this context, we concentrate on potassium imaging using the FluxORTM potassium ion channel assay and calcium imaging using Fura-2 acetoxymethyl ester (AM; Life Technologies). According to the manufacturer’s instructions, the “FluxORTM assay takes advantage of the well described perme- ability of potassium channels to thallium (Tl+) ions.” The FluxORTM reagent is a thallium-sensitive fluorescent indicator dye with an AM group. The AM group will be cleaved by intracellular esterases once the dye passes the cell membrane. Thallium is added to the extracellular solution and once the potassium channels are opened (due to Gβγ coupling to potassium channels following opioid receptor agonist application), thallium ions diffuse into the cell and activate the dye resulting in a fluorescent cytosolic signal (http:// tools.invitrogen.com/content/sfs/manuals/mp10016.pdf).

Santi M. Spampinato (ed.), Opioid Receptors: Methods and Protocols, Methods in Molecular Biology, vol. 1230, DOI 10.1007/978-1-4939-1708-2_15, © Springer Science+Business Media New York 2015

187

188 Viola Spahn et al.

Calcium imaging is based on the use of calcium-sensitive dyes, which fluorescence intensity changes upon binding to calcium. A widely used calcium-sensitive dye is Fura-2, AM, which is a stilbene fluorophore with characteristics of the calcium chelating substances ethylene glycol-bis(2-aminoethylether)-N,N,N′,N′-tetraacetic acid (EGTA) and 1,2-bis(o-aminophenoxy)ethane-N,N,N′,N′-tetra- acetic acid (BAPTA), firstly synthesized by Grynkiewicz et al. in 1985 [1]. Fura-2 is highly fluorescent and its excitation spectra shift to shorter wavelengths as the intracellular calcium concentra- tion increases. Therefore, the ratio of the fluorescence intensities of Fura-2 measured at excitation wavelength of 340 and 380 nm is directly correlated with the intracellular calcium concentration. The binding of an AM group to Fura-2 allows Fura-2, AM to pass the cell membrane. In the cytosol, the AM will be cleaved by endogenous esterases, while Fura-2 will stay intact [1].

Opioids are widely used to treat severe pain [2] and they act via three classical types of genetically and pharmacologically specified opioid receptors, the μ, δ, and κ [3]. Following activation, opioid receptors couple to inhibitory G-proteins leading to their dissocia- tion into Gα- and Gβγ-subunits. Both subunits have several down- stream effects, for instance, direct activation of potassium channels like the G-protein-coupled inwardly rectifying potassium (GIRK) channel 2 by Gβγ [4, 5] and the modulation of the unselective cation channel transient receptor potential vanilloid 1 (TRPV1) via the cyclic adenosine monophosphate (cAMP) and protein kinase A (PKA) pathway [6–8].

In the following sections of this chapter we analyze (1) the activation of the potassium channel GIRK2 by agonists of μ- ([D-Ala2, N-MePhe4, Gly-ol]-enkephalin; DAMGO) and δ- ([D-Pen(2, 5)]-enkephalin; DPDPE) opioid receptors with the aid of the FluxORTM assay [5], and (2) the calcium imaging pro- tocols to measure capsaicin-induced TRPV1 activity during opi- oid withdrawal compared to untreated control cells in transfected human embryonic kidney (HEK) 293 cells [7]. The underlying mechanisms for this modulation are following: Beside the well known inhibition of adenylyl cyclases (AC) by acute opioid recep- tor activation, the withdrawal of chronically applied opioids induces AC superactivation leading to cAMP overshoot [9]. We have recently shown that the increased cAMP level results in PKA-dependent TRPV1 sensitization, which was reflected by an increased capsaicin-induced TRPV1 activity (represented as increased calcium influx into the cytosol via the activation of TRPV1) during opioid withdrawal compared to control condi- tions in μ-opioid receptor and TRPV1 expressing HEK 293 cells and dorsal root ganglion (DRG) neurons [7].

2 Materials

2.1 Measurement of DAMGO-
and DPDPE-Induced GIRK2 Activity

in Mouse DRG Neurons Using Potassium Imaging

2.1.1 Cells, Media, Buffers, Chemicals

1. Primary culture of mouse DRG neurons.

2. Dulbecco’s modified eagle medium (DMEM)/HAM’s F12 without glutamine.

3. Horse serum (10 %).
4. Penicillin (10,000 U)/streptomycin (10 mg/ml). 5. Glucose (0.8 %).
6. Glutamine (1 mM).
7. Glibenclamide (20 μM).
8. Tetraethylammonium chloride (20 mM).
9. 1× Phosphate buffered saline (PBS).

10. Loading buffer (prepared according to the manufacturer’s instructions): PowerLoadTM concentrate 100×, FluxORTM reagent reconstituted in dimethyl sulfoxide (DMSO), FluxORTM assay buffer, probenecid reconstituted in deionized water.

11. Assay buffer: FluxORTM assay buffer 10×, probenecid reconsti- tuted in deionized water, tetraethylammonium chloride 20 mM, glibenclamide 20 μM.

12. Stimulus buffer (prepared according to the manufacturer’s instructions): FluxORTM chloride free buffer 5×, Tl2SO4 concentrate.

13. Collagenase IV (1 mg/ml). 14. Trypsin (0.05 %).
15. DAMGO (10 μM).
16. DPDPE (10 μM).

17. Poly-l-lysine (0.1 mg/ml). 18. Bidestilled water.

1. Cell culture plates (Ø 34 mm). 2. Cover glasses (Ø 24 mm).
3. Needles (18G, 22G, 25G).
4. Cell strainer (40 μM).

5. CO2 cell incubator.

6. Pump SCI Q 323 (Watson Marlow, Rommerskirchen, Germany).

7. Polychrome V monochromator (Till Photonics, Gräfeling, Germany).

8. CCD camera (Till Photonics).

2.1.2 Expandable Material and Technical Equipment

Potassium and Calcium Imaging 189

190 Viola Spahn et al.

2.2 Measurement
of Capsaicin-Induced TRPV1 Activity During Opioid Withdrawal
in TRPV1 and μ-Opioid Receptor Expressing HEK 293 Cells Using Calcium Imaging

2.2.1 Cells, Plasmids, Media, Buffers, Chemicals

9. Eclipse TE 2000-s microscope (Nikon, Japan).

10. Objective S fluor 40× 1.3 oil (Nikon).

11. Yellow fluorescent protein filter.

12. Microscopy perfusion chamber.

13. HS 18 laminar airflow.

14. Parafilm.

15. Aluminum foil.

1. Cell line: HEK 293 cells (German collection of microorgan- isms and cell cultures, DSMZ, Braunschweig, Germany).

2. Plasmids: (a) rat pcDNA3.1 wild type TRPV1-YFP (provided by Prof. Michael Schaefer, Rudolf Boehm Institute for Phar- macology and Toxicology, University of Leipzig, Germany); (b) pcDNA 3.1 wild type MOR-Flag.

3. DMEM.

4. Fetal bovine serum (FBS).

5. Penicillin (10,000 U)/streptomycin (10 mg/ml).

6. X-tremeGENE DNA transfection reagent (Roche Diagnostics).

7. Calcium imaging buffer (CIB): 2 mM CaCl2, 10 mM glucose, 20 mM HEPES, 5 mM KCl, 140 mM NaCl, adjusted to pH 7.4 with NaOH.

8. Pluronic solution: 20 % pluronic F-127 in DMSO.

9. Fura-2, AM solution: 50 μg Fura-2, AM diluted in 10 μl
pluronic solution and 50 μl DMSO.

10. Capsaicin (see Note 1).

11. Forskolin (FSK).

12. 3-Isobutyl-1-methylxanthin (IBMX).

13. Morphine sulfate.

14. Poly-l-lysine (0.1 mg/ml).

15. Bidestilled water.

1. Cell culture flasks (75 cm2 growth area).

2. Cell culture plates (Ø 34 mm).

3. Cover glasses (Ø 24 mm).

4. CO2 cell incubator.

5. Pump SCI Q 323 (Watson Marlow).

6. Polychrome V monochromator (Till Photonics).

7. CCD camera (Till Photonics).

8. Eclipse TE 2000-s microscope (Nikon).

9. Objective S Fluor 40× 1.3 oil (Nikon).

2.2.2 Expandable Material and Technical Equipment

3 Methods

3.1 Measurement
of DAMGO- and DPDPE- Induced GIRK2 Activity in Mouse DRG Neurons Using Potassium Imaging

3.1.1 Preparation
of the Cell Culture Medium

3.1.2 Preparation
of Poly-l-Lysine-Coated cover Glasses

10. Yellow fluorescent protein filter. 11. Microscopy perfusion chamber. 12. HS 18 laminar airflow.
13. Parafilm.

1. Take off 50 ml of the DMEM/HAM’s F12 medium without glutamine and put it in a 50 ml tube for later use (working medium).

2. Add 10 % horse serum, 1 mM glutamine, 0.8 % glucose, and 1 % penicillin/streptomycin to the 450 ml of DMEM/HAM’s F12 (cell culture medium).

Coat the cover glasses with 0.1 mg/ml poly-l-lysine (for instance, 20 pieces at once). All steps have to be performed under the lami- nar air flow to avoid contamination.

1. Place one cover glass per cell culture plate (Ø 34 mm).

2. Add 1 ml of 0.1 mg/ml poly-l-lysine solution on top of each cover glass.

3. Incubate the mixture for 30 min.

4. Remove the poly-l-lysine solution (for example, with a clean pump or pipette) and wash the cover glasses in the culture plates three times with bidestilled water. Remove all the fluid and let the cover glass/culture plate to completely dry under the laminar air flow (approximately 5–10 min).

5. Wrap the cover glass/culture plate in parafilm.
6. Store the coated cover glass/culture plate at 4 °C until use.

1. Kill the mouse by placing it in a glass chamber on a ceramic perforated plate located above paper tissues soaked with isoflurane.

2. To isolate DRG (see Note 2), remove the fur and skin on the back of the mouse. Cut off the spine (with surrounding mus- cles) from the level of the pelvis until the ribs. Remove the spinal cord by laminectomy and dissect as many as possible lumbar and thoracic DRG from both sides. Collect the dis- sected DRG in a 1.5 ml eppendorf tube filled with 1× PBS.

3. After isolation of all DRG, remove the PBS and add 1 mg/ml collagenase IV diluted in working medium to the isolated DRG.

3.1.3 Preparation of the Mouse DRG Neuron Culture

Potassium and Calcium Imaging 191

192 Viola Spahn et al.

3.1.4 Performing
the Potassium Imaging Experiment

4. Incubate the DRG in a water bath at 37 °C for 28 min.

5. Remove the collagenase, add 0.05 % trypsin, and again incu-
bate the DRG for 28 min at 37 °C (see Note 3).

6. Suspend the DRG in cell culture medium and dissociate the cells by passing them through 18G, 22G, and 25G needles (four times through each needle). To separate the cells from the debris, pass the DRG through a 40 μm cell strainer.

7. Take the flow-through and centrifuge at 500×g for 5 min at room temperature.

8. Remove the supernatant and resuspend the cell pellet in 300– 500 μl of cell culture medium.

9. Seed 30–50 μl of the cell suspension on poly-lysine (100 μg/ ml) coated glass coverslips retained in cell culture plates (Ø 34 mm) and leave the cells to adhere for 30 min in the incubator.

10. Add 2 ml of the cell culture medium to each culture plate and incubate them overnight. On the next day, the cells can be used for potassium imaging experiments.

1. Prepare 10 ml (for ten plates) of loading buffer containing the thallium-sensitive dye. Keep the solution in the dark (see Notes 4 and 5).

2. Wash the DRG neurons once with 1× PBS (see Note 6).

3. Incubate the neurons with 1 ml of the loading buffer for
90 min at room temperature in the dark.

4. Prepare the assay buffer containing tetraethylammonium chlo- ride (20 mM) and glibenclamide (20 μM) to block voltage- gated and adenosine triphosphate-sensitive potassium channels (see Note 5).

5. Prepare the stimulus buffer containing 20 μM DAMGO or 20 μM DPDPE (the stimulus buffer will be 1:1 diluted during the measurement so that the cells will be stimulated with 10 μM DAMGO or 10 μM DPDPE) (see Note 5).

6. Prepare the stimulus buffer without opioids as a control (see Note 5).

7. Wash the DRG neurons three times with PBS (see Note 6).

8. Incubate the cells for 30 min in assay buffer at room tempera-
ture in the dark.

9. Switch on the polychromator, wait 10 min before you switch on the CCD camera and the computer.

10. Start the Till vision program. Use the following settings: exposure time, 100 ms; image acquiring, every 2 s; exposure (excitation) wavelengths, 488 nm; recording duration, 300 s.

3.2 Measurement
of Capsaicin-Induced TRPV1 Activity During Opioid Withdrawal
in TRPV1 and μ-Opioid Receptor Expressing HEK 293 Cells Using Calcium Imaging

3.2.1 Preparation
of the Cell Culture Media

3.2.2 Cell Culture, Sample Preparation,
and Transient Transfection

11. Take the cover glass out of the plate, transfer it to the microscopy perfusion chamber, and add 200 μl of assay buffer (see Note 6).

12. Place the chamber under the microscope. 13. Switch off the light.
14. Use the 20× objective to focus the cells.
15. Turn in the YFP filter and start the program.

16. Start the measurement: 20 s after the start, carefully add 200 μl of the stimulus buffer containing 20 μM DAMGO or DPDPE (see Note 6).

17. As a control, treat the cells with stimulus buffer without opioids (see Note 6).

18. Make sure that the cells are always surrounded with buffer and do not get “dry.”

19. After the measurement circle all cells and calculate their kinetics with the appropriate software tool in the Till vision program.

20. Convert the data into an excel format and save them for offline analysis.

1. Prepare HEK 293 cell culture medium by adding 10 % FBS and 1 % penicillin/streptomycin to DMEM, and store at 4 °C.

2. Provide unchanged DMEM needed for transient transfection (transfection medium) and store at 4 °C.

3. For preparation of poly-l-lysine coated cover glasses follow steps described in Subheading 3.1.2.

1. Maintain HEK 293 cells in 20 ml of DMEM supplemented with FBS and penicillin/streptomycin at 37 °C and 5 % CO2 in a cell incubator.

2. Passage the cells at the ratio 1:3–1:10 every second to third day, depending on the confluence. To do this, rinse the adher- ent cells from the culture flask bottom with DMEM, transfer them to a 50 ml cell culture tube, and centrifuge for 10 min at 400 × g and room temperature.

3. Remove the supernatant and resuspend the cell pellet in 20 ml of fresh media (cell suspension mixture). Take an appropriate part to maintain the cells: For example, transfer 5 ml of the cell suspension mixture into a fresh culture flask and add 15 ml of medium to a final volume of 20 ml; after approximately 2 days the cells will be confluent again. Use the rest of the cell suspen- sion to prepare the plates for the calcium imaging experiments.

Potassium and Calcium Imaging 193

194 Viola Spahn et al.

3.2.3 Performing the Calcium Imaging Experiment

4. Two days prior to the calcium imaging experiment, an appropriate amount of cells (see step 5 below) has to be plated on culture plates. Depending on the confluence and growth abilities of the cells use 250–500 μl of the cell suspension mix- ture and pipette it on the poly-l-lysine coated cover glass/ culture plate. Add fresh medium to the cell suspension/cover glass/culture plate to a final volume of 2 ml.

5. Twenty-four hours prior the experiment, the seeded cells have to be transfected. For one cover glass/culture plate, prepare the following transfection mixture: Place 95 μl of transfection medium into a 1.5 ml eppendorf tube, add 0.5 μg of pcDNA3.1 TRPV1-YFP and 2.0 μg of pcDNA3.1 MOR-flag. Vortex the X-treme GENE transfection reagent and add 5 μl of it to medium/plasmid mixture. Incubate the mixture for 15 min at room temperature under the hood. Afterwards, slowly pipette the mixture onto the seeded cells (cultured on the cover glasses the day before). For the first calcium imaging experiments, for a beginner, it is suggested to prepare 6–8 plates for one experi- mental day. Later on, 14–16 plates can be measured on an experimental day.

6. Half of the plates will be assigned as a control group (not treated with opioids, but with normal medium) and the other as the opioid withdrawal group. The opioid withdrawal group will be treated with 10 μM morphine sulfate at least 16 h before the calcium imaging experiments starts.

1. Prepare the Fura-2, AM solution: 50 μg of Fura-2, AM diluted in 10 μl of pluronic solution and 50 μl of DMSO (see Note 7).

2. Keep the dissolved solution in the dark and freeze it after the experiments at the end of the day.

3. Prepare 10 ml (for ten plates) of 1 μM capsaicin solution (dissolved in CIB).

4. Prepare 30 ml of CIB/morphine solution (final morphine concentration of 10 μM) (see Note 8).

5. Prepare 50 ml of CIB supplemented with 10 μM FSK and 2 mM IBMX (see Note 9).

6. Switch on the polychromator, wait 10 min before you switch on the CCD camera, computer, and the pump.

7. Start the Till vision program. Use the following settings: expo- sure time, 50 ms; image acquiring, every 2 s; alternating exposure wavelengths of 340 and 380 nm; recording duration, 70 s.

8. Place 700 μl of CIB (for the control group) or CIB/morphine (for the morphine withdrawal group) in a 1.5 ml eppendorf tube and add 3 μl of Fura-2, AM solution (loading solution).

9. Bring a plate with transfected cells and wash it carefully two times with 1 ml of CIB or CIB/morphine.

4 Notes

10. Remove the CIB (or CIB/morphine) and load the cells with the loading solution.

11. Incubate at 37 °C for 30 min in the dark.

12. Remove the loading solution and wash three times with CIB/ FSK/IBMX.

13. Add 1 ml of CIB/FSK/IBMX and incubate the cells for 15 min at 37 °C.

14. Take the cover glass out of the plate and transfer it to the microscopy perfusion chamber, and add 1 ml of CIB/FSK/ IBMX.

15. Place the chamber under the microscope, connect it to the aspiration tube (which is connected with the pump), switch off the light, use the 40× oil objective to focus the cells, turn on the YFP filter, change the excitation wavelength in the pro- gram to 495 nm and press “Live record” to identify cells expressing TRPV1-YFP, select an area where many cells express TRPV1-YFP, remove the YFP-filter, and change the excitation wavelength to 380 nm.

16. Start the measurement: immediately after the start, turn on the pump; the CIB/FSK/IBMX solution in the chamber will be aspirated.

17. As soon as the solution is almost removed add 200 μl of CIB/ capsaicin solution.

18. Immediately afterward wash the cells carefully five times with 1 ml of CIB.

19. Make sure that the cells are always surrounded with buffer and do not get “dry.”

20. After the measurement, define and subtract the background and circle the cells which responded to capsaicin with the appropriate software tool in the Till Vision program.

21. Calculate the ratio of the fluorescence of the circled cells at 340 and 380 nm using Till Vision software by pressing the button “kinetic selection.”

22. Convert the data into an excel format and save them for offline analysis.

1. Prepare the capsaicin and FSK/IBMX solution always fresh, on each experimental day.

2. To optimize the neuronal culture, the DRG need to be dissected fast and tidy.

3. Avoid too long incubation of the cells with collagenase IV or tr ypsin.

Potassium and Calcium Imaging 195

196 Viola Spahn et al.

4. It is recommended to perform all steps after loading of the cells with the thallium-sensitive dye in the dark to avoid bleach- ing of the dye.

5. Prepare all solutions for the potassium imaging always fresh, on each experimental day.

6. Treat the cells very carefully in order to keep them attached to the cover glass.

7. It is recommended to perform all steps after the loading of the cells with Fura-2 in the dark to avoid bleaching of the dye.

8. The CIB/morphine solution can be used until it is expended, but store it at 4 °C after you finish your experiment.

9. Prepare the CIB/FSK/IBMX solution always fresh, on each experimental day.

This work was supported by grants from the Deutsche Forschun- gsgemeinschaft (KFO 100/2), Bundesministerium für Bildung und Forschung (MedSys 0101-31P5783), and the European Society of Anesthesiology.

Acknowledgements

References

1. Grynkiewicz G, Poenie M, Tsien RY (1985) A new generation of Ca2+ indicators with greatly improved fluorescence properties. J Biol Chem 260:3440–3450

2. Zollner C, Stein C (2007) Opioids. Handb Exp Pharmacol 177:31–63

3. Dhawan BN, Cesselin F, Raghubir R et al (1996) International Union of Pharmacology. XII. Classification of opioid receptors. Pharmacol Rev 48:567–592

4. Akins PT, McCleskey EW (1993) Charac- terization of potassium currents in adult rat sensory neurons and modulation by opioids and cyclic AMP. Neuroscience 56:759–769

5. Nockemann D, Rouault M, Labuz D et al (2013) The K channel GIRK2 is both necessary and sufficient for peripheral opioid-mediated analgesia. EMBO Mol Med 5:1263–1277

6. Endres-Becker J, Heppenstall PA, Mousa SA et al (2007) Mu-opioid receptor activation mod- ulates transient receptor potential vanilloid 1 (TRPV1) currents in sensory neurons in a model of inflammatory pain. Mol Pharmacol 71:12–18

7. Spahn V, Fischer O, Endres-Becker J et al (2013) Opioid withdrawal increases transient receptor potential vanilloid 1 activity in a protein kinase A-dependent manner. Pain 154:598–608

8. Vetter I, Wyse BD, Monteith GR et al (2006) The mu opioid agonist morphine modulates potentiation of capsaicin-evoked TRPV1 responses through a cyclic AMP-dependent protein kinase A pathway. Mol Pain 2:22

9. Bie B, Peng Y, Zhang Y et al (2005) cAMP- mediated mechanisms for pain sensitization during opioid withdrawal. J Neurosci 25: 3824–3832

Chapter 16

Electrophysiological Patch Clamp Assay to Monitor the Action of Opioid Receptors

Viola Spahn, Dinah Nockemann, and Halina Machelska Abstract

The patch clamp is a valuable electrophysiological technique, which allows the study of single or multiple ion channels in cells, and it is particularly useful in testing the excitable cells such as neurons. Activation of neuronal opioid receptors results in the modulation of various ion channels, which enables to examine the receptors’ action with the patch clamp. In this chapter, we analyze the activation of the G-protein-coupled inwardly rectifying potassium channel 2 by opioids, and the capsaicin-induced transient receptor potential vanilloid 1 channel currents during opioid withdrawal, using the whole cell patch clamp in transfected human embryonic kidney 293 cells as well as in mouse and rat primary dorsal root ganglion neurons.

Key words Capsaicin, DAMGO, DRG neurons, GIRK currents, HEK 293 cells, Morphine, Whole cell patch clamp, TRPV1 currents

1 Introduction

In this chapter, we depict the measurement of opioid receptor activity using the whole cell patch clamp recordings. In this con- text, we focus on the activation of the G-protein-coupled inwardly rectifying potassium (GIRK) channel 2 by opioids and the mea- surements of increased capsaicin-induced transient receptor poten- tial vanilloid 1 (TRPV1) currents during opioid withdrawal.

The patch clamp technique was developed by Neher and Sakmann in the late 1970s [1]. They were the first to record single- channel currents from the membrane of denervated frog muscle fibers. This was not possible until that time due to extraordinary background noise. Neher and Sakmann circumvented this issue by “applying closely the tip of a glass pipette, 3–5 μm in diameter, on to the muscle surface, thus isolating electrically a small patch of mem- brane” [1]. Later on, this technique was considerably improved by Hamill and colleagues [2]. Their refinements (due to patch pipettes, which can form a giga-seal with the membrane) resulted in a higher current resolution and physical isolation of membrane patches.

Santi M. Spampinato (ed.), Opioid Receptors: Methods and Protocols, Methods in Molecular Biology, vol. 1230, DOI 10.1007/978-1-4939-1708-2_16, © Springer Science+Business Media New York 2015

197

198 Viola Spahn et al.

2 Materials

2.1 Measurement of Opioid-Induced GIRK2 Currents

. 2.1.1 Cell Cultures

. 2.1.2 Cell Culture Media

and Patch Clamp Solutions

This was the basis for different patch configurations, such as whole cell, inside-out, or outside-out patch recordings [2].

Opioid receptors are coupled to inhibitory G-proteins, and their activation results in a dissociation of the G-protein, having several downstream effects, including the activation of potassium channels directly by the Gβγ-subunit and the modulation of the cyclic adenosine monophosphate (cAMP) production via the Gα-subunit. The latter effect influences the activity of adenylyl cyclases (ACs) (which synthesize cAMP from adenosine triphos- phate) in two ways: acute opioid receptor stimulation results in an inhibition of ACs, whereas opioid withdrawal is correlated with increased AC activity and subsequent cAMP overshoot [3–6] (see also chapter “Analysis of potassium and calcium imaging to assay the function of opioid receptors” in this book). This dual regulation affects the unselective cation channel TRPV1 in dual manner: acute morphine application inhibits TRPV1, while morphine with- drawal (following prior prolonged exposure) causes sensitization of the channel [7, 8].

In the following sections of this chapter, we provide detailed protocols to measure acute opioid-induced GIRK2 activation and increased TRPV1 activity during opioid withdrawal.

These experiments are performed in GIRK2 and μ-opioid receptor expressing human embryonic kidney (HEK) 293 cells, DRG neu- rons of Nav1.8-GIRK2 transgenic mice (for more details see [9], and DRG neurons of wild type rats.

1. Cell line: HEK 293 cells (German collection of microorgan- isms and cell cultures, DSMZ, Braunschweig, Germany).

2. Plasmids: (a) pFLAG-CMV2 wild type Flag-GIRK2 (mouse); (b) pcDNA 3.1 wild type MOR-Flag; (c) pEYFP-C1.

3. Primary culture of DRG neurons from mice or rats.

1. HEK 293 cell culture medium containing 10 % FBS and 1 % penicillin/streptomycin to DMEM, and store at 4 °C.

2. Opti-MEM for transient transfection (transfection medium), store at 4 °C.

3. Rat DRG neuron working medium (MEM Earl’s supple- mented with 1 % penicillin/streptomycin) and culture medium (MEM Earl’s supplemented with 1 % penicillin/streptomycin and 10 % horse serum).

2.1.3 Poly-L-Lysine- Coated Culture Plates

4. Mouse DRG neuron medium by taking out 50 ml of the DMEM/HAM’sF12 medium without glutamine and transfer it to a 50 ml tube for later use (working medium). Add horse serum (final concentration 10 %), glutamine (final concentration 1 mM), glucose (final concentration 0.8 %), and 1 % penicillin/ streptomycin to 450 ml of DMEM/HAM’sF12 (cell culture medium).

5. 1 l of extracellular high K+ solution and 1 l of extracellular low K+ solution.

6. 100 ml of intracellular solution; prepare 1 ml aliquots and freeze them at −20 °C until use.

7. 50 ml of extracellular high K+ solution containing 10 μM DAMGO.

8. 50 ml of extracellular high K+ solution containing 20 μM naloxone hydrochloride.

9. 50 ml of extracellular high K+ solution containing 3 mM barium chloride.

10. 1 mg/ml Collagenase IV.
11. 3 mg/1 ml of MEM Collagenase type II.
12. 0.05 % Trypsin.
13. 1 mg/ml of MEM Trypsin type I.
14. 10 μM (D-Ala2,N-Me-Phe4,Glycinol5)-Enkephalin (DAMGO). 15. 0.1 mg/ml Poly-L-lysine.

1. Coat the culture plates with 0.1 mg/ml poly-L-lysine (for instance, 20 pieces at once).

2. All steps have to be performed under the laminar air flow to avoid contamination.

3. Add 1 ml of 0.1 mg/ml poly-L-lysine solution on top of each culture plate.

4. Incubate the mixture for 30 min.

5. Remove the poly-L-lysine solution (e.g., with a clean pump or pipette) and wash the culture plates three times with bidestilled water. Remove all the fluid and let the culture plates to com- pletely dry under the laminar flow (approximately 5–10 min).

6. Wrap the culture plates (e.g., 6 at once) in parafilm. 7. Store the coated culture plates at 4 °C until use.

1. Dulbecco’s modified eagle medium (DMEM), 10 % fetal bovine serum (FBS), penicillin (10,000 U)/streptomycin (10 mg/ml), FuGENE®HD transfection reagent, Opti-MEM® transfection medium.

2.1.4 Media
and Transfection Reagents for HEK 293 Cells

Whole Cell Patch Clamp 199

200 Viola Spahn et al.

2.1.5 Media for the DRG Neuron Cultures

2.1.6 Buffers and Solutions

1. DMEM/HAM’sF12 without glutamine, minimum essential medium (MEM) with Earle’s salts, horse serum, penicillin (10,000 U)/streptomycin (10 mg/ml), glucose, glutamine.

1. 1× phosphate buffered saline (PBS).

2. Extracellular low K+ solution: 140 mM NaCl, 5.6 mM KCl, 2.6 mM CaCl2, 1.2 mM MgCl2, 10 mM hydroxyethyl pipera- zineethanesulfonic acid (HEPES), 2.6 mM D-glucose, adjusted to pH 7.4 with NaOH.

3. Extracellular high K+ solution: 140 mM KCl, 2.6 mM CaCl2, 1.2 mM MgCl2, 5 mM HEPES, adjusted to pH 7.4 with NaOH.

4. Intracellular solution: 122 mM KCl, 11 mM ethylene glycol tetraacetic acid (EGTA), 1 mM CaCl2, 2 mM MgCl2, 10 mM HEPES, 4 mM MgATP, 0.25 mM NaGTP, 5 mM NaCl, adjusted to pH 7.4 with KOH.

1. Cell culture flasks (75 cm2 growth area).

2. Cell culture plates (Ø 34 mm).

3. 18G, 22G, 25G needles.

4. Cell strainer (40 μM).

5. Parafilm.

6. CO2 cell incubator.

7. HS 18 laminar airflow (Heraeus, Kleinostheim, Germany).

8. Borosilicate glass capillaries (Hilgenberg, Malsfeld, Germany).

9. Silver wire (Ø 0.25 mm).

10. Amplifier EPC-10 (HEKA Elektronik, Lambrecht, Germany).

11. Objective A Plan 10× 7.025 Ph1 (Zeiss, Göttingen, Germany).

12. Microscope Axiovert (Zeiss).

13. Micromanipulator 5171 (Eppendorf, Hamburg, Germany).

14. Micropipette puller P-97 (Sutter Instrument, Novato, USA).

15. Six-channel valve controller (Warner Instruments, Hamden, USA).

These experiments are performed in TRPV1 and μ-opioid receptor expressing HEK 293 cells, and primary rat DRG neurons.

1. Cell line: HEK 293 cells.

2. Plasmids: (a) rat pcDNA3.1 wild type TRPV1-YFP (provided by Prof. Michael Schaefer, Rudolf Boehm Institute for Pharma- cology and Toxicology, University of Leipzig, Germany); (b) pcDNA 3.1 wild type MOR-Flag.

3. Primary culture of DRG neurons from rats.

2.1.7 Materials
and Technical Equipment

2.2 Measurement
of Capsaicin-Induced TRPV1 Currents During Opioid Withdrawal

2.2.1 Cell Cultures

2.2.2 Cell Culture Media and Poly-L-Lysine-Coated Culture Plates

1. HEK 293 cell culture medium by adding 10 % FBS and 1 % penicillin/streptomycin to DMEM, and store at 4 °C.

2. Provide unchanged DMEM needed for transient transfection (transfection medium) and store at 4 °C.

3. Rat DRG working medium (MEM Earl’s supplemented with 1 % penicillin/streptomycin) and culture medium (MEM Earl’s supplemented with 1 % penicillin/streptomycin and 10 % horse serum).

4. 1 l of extracellular solution.

5. 100 ml of intracellular solution; make 1 ml aliquots and freeze them at −20 °C until use.

6. Poly-L-lysine-coated culture plates.

1. Penicillin (10,000 U)/streptomycin (10 mg/ml).
2. X-tremeGENE DNA transfection reagent (Roche Diagnostics).

1. MEM with Earle’s salts.
2. Horse serum.
3. Penicillin (10,000 U)/streptomycin (10 mg/ml).

1. Extracellular solution: 2 mM CaCl2, 2 mM MgCl2, 10 mM glucose, 10 mM HEPES, 5 mM KCl, 140 mM NaCl, adjusted to pH 7.4 with NaOH.

2. Intracellular solution: 140 mM KCl, 2 mM MgCl2, 5 mM EGTA, 10 mM HEPES, adjusted to pH 7.4 with KOH.

1. Cell culture flasks (75 cm2 growth area). 2. Cell culture plates (Ø 34 mm).
3. CO2 cell incubator.
4. HS 18 laminar airflow (Heraeus).

5. Parafilm.
6. Cell strainer (40 μm).
7. Borosilicate glass capillaries (Hilgenberg). 8. Silver wire (Ø 0.25 mm).
9. Amplifier EPC-10 (HEKA Elektronik).

10. Objective A Plan 10× 7.025 Ph1 (Zeiss).
11. Microscope Axiovert (Zeiss).
12. Micromanipulator 5171 (Eppendorf).
13. Micropipette puller P-97 (Sutter Instrument).
14. Six-channel valve controller (Warner Instruments).

2.2.3 Media
and Transfection Reagents for HEK 293 Cells

2.2.4 Media for the DRG Neuron Cultures

2.2.5 Buffers and Solutions

2.2.6 Material
and Technical Equipment

Whole Cell Patch Clamp 201

202 Viola Spahn et al.

3 Methods

3.1 Measurement of Opioid-Induced GIRK2 Currents

3.1.1 HEK 293 Cell Culture and Transfection

1. Maintain HEK 293 cells in 20 ml of DMEM supplemented with FBS and penicillin/streptomycin at 37 °C in a cell incuba- tor (5 % CO2).

2. Passage the cells at a ratio of 1:3–1:10 every second to third day, depending on the confluence. To do this, rinse the adher- ent cells with DMEM from the culture flask ground, transfer them to a 50 ml cell culture tube, and centrifuge for 10 min at 400 × g at room temperature.

3. Remove the supernatant and resuspend the cell pellet in 20 ml of fresh medium (cell suspension mixture). Take an appropri- ate part (depending on the cell number ratio you decide on) to maintain the cells. Pipette 500–1,000 μl (depending on the confluence and growing abilities of your cells) of the remaining cell suspension mixture on an uncoated culture plate 2 days prior to the patch clamp experiment. Add fresh media to the cell suspension/culture plate to a final volume of 2 ml.

4. Twenty-four hours prior to the patch clamp experiment, pre- pare the transfection reaction mix for a 35 mm cell culture plate containing 1.5 μg DNA (ratio 1:4 of plasmid DNA encoding GIRK2 and μ-opioid receptor, respectively), and 6 μl of FuGENE®HD in 100 μl OptiMEM transfection medium. Incubate the reaction mix for 15 min at room temperature, and dropwise add the mix to the cells. Use a plasmid that codes for yellow fluorescent protein (0.1 μg of pEYFP-C1) as a con- trol to determine transfected cells.

5. Separate the transfected cells 12 h before electrophysiological recordings. For this, discard the cell culture medium and wash the cells once with PBS. Incubate the cells with 500 μl of tryp- sin for 3 min at 37 °C. Stop the dissociation of the cells by adding 1 ml of culture medium. Centrifuge the cells for 10 min at 400 × g at room temperature and resuspend the pellet in 1 ml of fresh medium. Pipette 30–50 μl of the resuspended cells in poly-L-lysine coated culture plate and fill up with the medium to a final volume of 2 ml. Repeat this step depending on how many plates you wish to study on the experimental day. For beginners it is suggested to start with six plates; later on one can measure 10–16 plates per experimental day.

1. Kill the mouse by placing it in a glass chamber on a ceramic perforated plate located above paper tissues soaked with isoflu- rane (e.g., Abbott). We used C57BL6 Nav1.8-GIRK2 male mice (8–10 weeks old).

2. To isolate DRG remove the fur and skin on the back of the mouse. Cut off the spine (with surrounding muscles) from the

3.1.2 Cultures of Mouse DRG Neurons

3.1.3 Cultures of Rat DRG Neurons

level of the pelvis until the ribs. Remove the spinal cord by laminectomy and dissect as many as possible lumbar and tho- racic DRG from both sides (see Note 1). Collect the dissected DRG in a 1.5 ml eppendorf tube filled with 1× PBS.

3. After dissection of all DRG, remove the PBS and add 1 mg/ml collagenase IV diluted in working medium to the DRG.

4. Incubate the tissue in a water bath at 37 °C for 28 min.
5. Remove the collagenase, add 0.05 % trypsin to the DRG, and

again incubate the DRG for 28 min at 37 °C (see Note 1).

6. Suspend the DRG in cell culture medium and dissociate the cells by carefully passing them four times through each 18G, 22G, and 25G needle. To separate the cells from the debris, pass the DRG through a 40 μm cell strainer.

7. Take the flow-through and centrifuge at 500×g for 5 min at room temperature.

8. Remove the supernatant and resuspend the cell pellet in approximately 300–500 μl of cell culture medium.

9. Seed 30–50 μl of the cell suspension (see Note 2) to the center of the poly-L-lysine coated cell culture plates (Ø 34 mm) (see Note 3). Let the cells adhere for 30 min in the incubator and fill up with culture medium until a final volume of 2 ml. On the next day, the cells can be used for patch clamp experiments.

1. Kill the rat by placing it in a glass chamber on a ceramic perfo- rated plate located above paper tissues soaked with isoflurane (e.g., Abbott). We use male Wistar rats (6–8 weeks old; 180–200 g).

2. Isolate DRG (see Notes 1 and 4); you can use all lumbar and thoracic DRG from both sides. Pool the DRG in 1 ml of ice- cold sterile MEM working medium.

3. All further steps (digestions, dissociation, and plating) have to be performed under the laminar air flow to avoid contamination.

4. Add 1 ml of MEM containing 3 mg of collagenase type II and let it digest for 50 min at 37 °C.

5. Add 1 ml of MEM containing 1 mg of trypsin type I for 10 min at 37 °C.

6. After digestion, carefully dissociate the DRG by pipetting up and down 20 times, and filter them through a 40 μm filter.

7. Collect the filtrate in a 50 ml falcon tube and append 4 ml of working medium supplemented with 80 mg of BSA.

8.Centrifuge the mixture at 500×g for 5 min at room temperature.

Whole Cell Patch Clamp 203

204 Viola Spahn et al.

3.1.4 Performing the Patch Clamp Experiment

9. Decant the supernatant very carefully (see Note 5). Resuspend the pellet in 5 ml of fresh working medium via gentle handshaking.

10. Centrifuge the cells for 5 min at 300×g.

11. Carefully remove the supernatant (see Note 5). Add 3 ml of fresh working medium and dissociate the pellet by pipetting the cells up and down 4–5 times.

12. Transfer 300 μl of the cell suspension to the center of the pre- viously prepared poly-L-lysine coated culture plates (see Note 6); avoid the leaking of the cell drop and incubate the cells for 1 h at 37 °C in the cell incubator (5 % CO2).

13. Add 1.7 ml of the culture medium to the cells and maintain them in the cell incubator overnight. On the next day, use the cells for patch clamp.

These patch clamp experiments are performed as described earlier [9, 10].

1. Thaw 1 aliquot of intracellular solution.

2. Switch on the micromanipulator, the six barrel valve perfusion system, the pipette puller, and the computer.

3. Start the pulse software (HEKA Elektronik) and select the “GIRK” protocol (for detailed protocol information see Fig. 1).

4. Fill one barrel with extracellular low K+ solution, one with extracellular high K+ solution, and the others with the solu- tions containing DAMGO, naloxone hydrochloride, or barium chloride. Make sure the tubes are not blocked and the fluid is running constantly (see Note 7).

5. Bring the plate with cells and wash it carefully once with 1 ml of extracellular low K+ solution before filling the plate with 1 ml of extracellular low K+ solution.

6. Fill the tip of a pulled glass pipette with intracellular solution and fix it, including the electrode which is surrounded with intracellular solution, with the holder of the micromanipulator.

7. Place the plate with the cells under the microscope and focus.

8. For the DRG neurons: Look for round cells without any pro- cesses, with an even surface.

9. For transfected HEK 293 cells: Look for preferentially single standing cells (see Note 8).

10. Place the tip of the glass pipette (including electrode) in the extracellular low K+ solution of the culture plate with the cells, and turn on the perfusion system (the barrel filled with extra- cellular low K+ solution).

11. Press the “SET-UP” button in the pulse program and check the resistance of the pipette (it should be between 3 and 5 MΩ).

Whole Cell Patch Clamp 205

Fig. 1 Example of pulse generation software (HEKA Elektronik)

12. Place the electrode directly above the cell, go down with the pipette (using the micromanipulator), and patch it. At the moment the pipette touches the membrane of the cell, the resis- tance will increase (you will see it in the program). At this time (when the resistance reaches 7–15 MΩ) you have to apply a slight but constant suction pressure via the suction tube (connected to the pipette) until the resistance does not increase anymore (this increases the sealing success).

13. Change to the “whole cell” mode in the program and hold the voltage at −60 mV, if the resistance of the pipette is increased to at least 60 MΩ.

14. Try to get a giga-seal; support the sealing process via slight suction. However, be careful because sometimes the resistance starts to decrease after a while. If this happens, stop the suction immediately.

15. Once you got the giga-seal, press the “CFast–Auto” button in the program.

16. Try to break in the cell via a short but strong suction of the suction tube.

206

Viola Spahn et al.

3.2
of Capsaicin-Induced TRPV1 Currents During Opioid Withdrawal

3.2.1 HEK 293 Cell Culture and Transfection

17. Once you broke into the cell, press the “CSlow–Auto” button. If the “E” in the “Auto” button is transparent and the resis- tance of the pipette is not below 100 MΩ, you broke in successfully.

18. Change the voltage from −60 to −80 mV.

19. Continuously monitor the capacitance by using the auto func- tion of pulse.

20. Turn off the barrel with the extracellular low K+ solution and switch on the barrel with extracellular high K+ solution. The resistance of the membrane of the cell will drop down; wait until the cell has recovered (see Note 9).

21. Start the “GIRK”-protocol (monitor the current at −80 mV). Switch of the extracellular high K+ solution barrel and at the same time switch on the barrel with the DAMGO solution in the perfusion system for 20 s. Afterwards, stop the perfusion with the DAMGO solution and start again the perfusion of the extracellular high K+ solution. Wait until the current recovers and turn on again the DAMGO solution. In parallel, turn on the solution with naloxone or barium chloride to block the DAMGO-induced current.

22. To classify DRG neurons in nociceptors and mechanorecep- tors, you can record the action potentials of the neurons. Nociceptors display broad action potentials with inflections on the falling phase, while mechanoreceptors have narrow action potentials. To induce action potentials, switch the amplifier to current-clamp mode and then inject currents from 0.02 to 10 nA for 80 ms.

23. After the measurement, turn off the perfusion system, replace the electrode, and remove the glass pipette and culture plate.

1. Twenty-four hours prior to the patch clamp experiment, trans- fect the seeded cells. For one cover culture plate prepare the following transfection mixture: Place 95 μl of transfection medium into an eppendorf tube, add 0.5 μg of pcDNA3.1 TRPV1-YFP and 2 μg of pcDNA3.1 MOR-flag. Vortex the X-treme GENE transfection reagent and add 5 μl of it to medium/plasmid mixture. Incubate the mixture for 15 min at room temperature under the hood. Afterwards, slowly pipette the mixture onto the seeded cells.

2. Six to 8 h after the transfection, the transfected cells have to be singularized: Rinse the transfected cells with the medium from the culture plate ground and transfer them into clean 2 ml eppendorf tube. Centrifuge the cells for 10 min at 400×g at room temperature. Remove the supernatant and resuspend the pellet in 1 ml of fresh culture medium. Pipette 30 μl of the resuspended cells in a previously poly-lysine coated culture

Measurement

3.2.2 Cultures of Rat DRG Neurons

3.2.3 Performing the Patch Clamp Experiment

plate and fill up with a medium to a final volume of 2 ml. Repeat this step depending on how many plates you wish to study on the experimental day; for beginners it is suggested to start with six plates, and with more experience one can mea- sure 10–16 plates per experimental day.

3. Half of the plates will be assigned as control group (not treated with opioids) and the other half as the opioid withdrawal group. The opioid withdrawal group will be treated with 10 μM morphine sulfate at least 16 h before the patch clamp experiments (usually directly after the cells are seeded on the poly-L-lysine coated culture plates).

1. Follow steps 1–12 described in Subheading 3.1.3.

2. Add 1.7 ml of the culture medium to the cells and maintain them in the cell incubator overnight. The opioid withdrawal group will be treated with 10 μM morphine sulfate at least 16 h before the patch clamp experiments.

1. Prepare 100 ml of extracellular solution containing 10 μM FSK and 2 mM IBMX.

2. Prepare 10 ml of extracellular solution containing 1 μM capsaicin.

3. Thaw 1 aliquot of intracellular solution.

4. Switch on the micromanipulator, the six barrel valve perfusion system, the pipette puller, and the computer.

5. Start the pulse software (HEKA Elektronik) and select the “cap- saicin” protocol (for detailed protocol information see Fig. 2).

6. Fill one of the barrels with extracellular solution containing FSK/IBMX and a second one with the extracellular solution containing capsaicin. Make sure the tubes are not blocked and the fluid is running constantly (see Note 7).

7. Bring a plate with cells and wash it carefully twice with 1 ml of extracellular solution containing FSK/IBMX.

8. Add 1 ml of extracellular solution containing FSK/IBMX and incubate it for 20 min. During the incubation, pull a glass pipette using the pipette puller, fill the tip with intracellular solution, and fix it, including the electrode (which is sur- rounded with intracellular solution), at the holder of the micromanipulator. Then, place the plate with the cells under the microscope and focus them.

9. For the DRG neurons: Look for round cells, without any pro- cesses, with an even surface, and a diameter less than 26 μm.

10. For transfected HEK 293 cells: Look for preferentially single standing cells (see Note 10).

Whole Cell Patch Clamp 207

208 Viola Spahn et al.

Fig. 2 Pulse software (HEKA Elektronik) image with the “capsaicin” protocol

11. Place the tip of the glass pipette (including electrode) in the extracellular solution containing FSK/IBMX of the culture plate with the cells and turn on the perfusion system (the bar- rel filled with extracellular solution containing FSK/IBMX).

12. Press the “SET-UP” button in the pulse program and check the resistance of the pipette (it should be between 3 and 5 MΩ).

13. Place the electrode directly above the cell, go down with the pipette (using the micromanipulator) and patch it. At the moment the pipette touches the membrane of the cell, the resistance will increase (you will see it in the program). At this time (when the resistance reaches 7–15 MΩ) you have to apply a slight but constant suction pressure via the suction tube (connected to the pipette) until the resistance does not increase anymore (this increases the sealing success).

14. Change to the “whole cell” mode in the program and hold the voltage at −60 mV, if the resistance of the pipette is increased to at least 60 MΩ.

15. Try to get a giga-seal, support the sealing process via slight suction. However, be careful, sometimes the resistance starts to decreases after a while. If this happens, stop the suction immediately.

4 Notes

16. Once you got the giga-seal, press the “CFast–Auto” button in the program.

17. Try to break in the cell via a short but strong suction of the suction tube.

18. Once you broke into the cells press the “CSlow–Auto” button. If the “E” in the “Auto” button is transparent and the resis- tance of the pipette is not below 100 MΩ, you broke in successfully.

19. Start the “capsaicin”-protocol and switch off the barrel with the extracellular solution containing FSK/IBMX, and at the same time switch on the barrel with the extracellular solution containing capsaicin in the perfusion system for 5 s. Afterwards, turn off the perfusion with extracellular solution containing capsaicin and turn on the perfusion with extracellular solution containing FSK/IBMX.

20. Measure the capsaicin-induced current.

21. After the measurement, turn off the perfusion system, replace the electrode, and remove the glass pipette and culture plate.

1. To successfully patch DRG neurons, it is important to culture healthy DRG neurons. Therefore, try to dissect the DRG neu- rons as tidy and fast as possible, immediately after the animal is killed, and avoid too long incubation of the cells with collage- nase IV or trypsin.

2. The density of your cell culture depends on the number of DRG you collect. The dilution of the cells in 300–500 μl is just a suggestion. Since it is easier to patch DRG neurons which are not seeded too dense, you will need to find the optimal dilution for your cell preparation.

3. The advantage of pipetting the cell suspension to the center of the plate is that the neurons attach in a concentrated area. This makes it easier to find the right cell for recording during the patch clamp experiments.

4. Remove the spine by cutting it cranially and caudally as well as by cutting the ribs and the surrounding tissue. Afterwards, open the spinal column by cutting the lamina of the vertebral arch, lift the dorsal part of the column, and put it beside. Remove the spinal cord from the ventral part of the column. Now, you should see the foramen intervertebralia including the DRG. Do not squeeze the DRG when you remove them from the foramen intervertebralia. Try to place the forceps below them and carefully pull them out from the foramen.

Whole Cell Patch Clamp 209

210 Viola Spahn et al.

Acknowledgements

Once you take out the DRG, cut the distal and proximal axons directly at the DRG, and transfer it to the tube with the work- ing medium.

5. Be very careful performing this step. The cell pellet is very often flimsy and it can happen that you loose many cells. Also, avoid major agitation while you transfer the tube from the cen- trifuge to the laminar air flow. If you realize that the pellet is loosened, centrifuge again for 5 min at the corresponding speed. Collect the supernatant in a separate clean 50 ml tube just in case you loose the pellet.

6. It is recommended to use maximum of 300 μl of the cell sus- pension because the neurons attach in a concentrated area. This makes it easier to find the right cell for recording during the patch clamp experiments.

7. It is very important that the cells are well perfused. However, do not do it too fast because the cells will be washed away (you have to establish the right speed for your set-up and your cells) in the sealing process. Especially DRG neurons need some time to form a giga-seal, and the constant perfusion is essential for the success.

8. Try to discriminate GIRK-positive cells through their optical appearance; they do not look that “healthy” as untransfected cells do.

9. It can take a couple of minutes until the cell recovers, there- fore, make sure that you have enough extracellular high K+ solution in your perfusion system to ensure a stable perfusion of the cells. If the cell does not recover after 5 min, you should wash the plate thoroughly, but carefully, with the extracellular low K+ solution, and start to patch a new cell. If the next cell also dies during the change to the extracellular high K+ solu- tion, you should start with a new plate.

10. You can increase the chance to patch a transfected cell using the YFP-filter and an excitation wavelength of 495 nm to select TRPV1-positive cells (if your microscope is equipped with it), since the TRPV1 sequence is tagged with the YFP sequence. Otherwise, separate the cells by their optical appear- ance; TRPV1 expressing cells do not look as “healthy” as untransfected cells do.

References

1. Neher E, Sakmann B (1976) Single-channel currents recorded from membrane of dener- vated frog muscle fibres. Nature 260:799–802

2. Hamill OP, Marty A, Neher E et al (1981) Improved patch-clamp techniques for high- resolution current recording from cells and cell-free membrane patches. Pflugers Arch 391: 85–100

3. Childers SR (1991) Opioid receptor-coupled sec- ond messenger systems. Life Sci 48:1991–2003

4. Nestler EJ (1992) Molecular mechanisms of drug addiction. J Neurosci 12:2439–2450

5. Sharma SK, Klee WA, Nirenberg M (1975) Dual regulation of adenylate cyclase accounts for narcotic dependence and tolerance. Proc Natl Acad Sci U S A 72:3092–3096

6. Sharma SK, Nirenberg M, Klee WA (1975) Morphine receptors as regulators of adenylate cyclase activity. Proc Natl Acad Sci U S A 72: 590–594

7. Endres-Becker J, Heppenstall PA, Mousa SA et al (2007) Mu-opioid receptor activation modulates transient receptor potential vanil- loid 1 (TRPV1) currents in sensory neurons in a model of inflammatory pain. Mol Pharmacol 71:12–18

8. Spahn V, Fischer O, Endres-Becker J et al (2013) Opioid withdrawal increases transient receptor potential vanilloid 1 activity in a pro- tein kinase A-dependent manner. Pain 154: 598–608

9. Nockemann D, Rouault M, Labuz D et al (2013) The K channel GIRK2 is both neces- sary and sufficient for peripheral opioid- mediated analgesia. EMBO Mol Med 5: 1263–1277

10. Raveh A, Cooper A, Guy-David L et al (2010) Nonenzymatic rapid control of GIRK channel function by a G protein-coupled receptor kinase. Cell 143:750–760

Whole Cell Patch Clamp 211

Part IV

Model Systems to Studying Opioid Receptor-Mediated Functions

Chapter 17

Skin–Nerve Preparation to Assay the Function of Opioid Receptors in Peripheral Endings of Sensory Neurons

Rabih Moshourab, Yvonne Schmidt, and Halina Machelska Abstract

This chapter describes the methodology of the in vitro skin–saphenous nerve preparation and its application to test for the modulatory effects of opioids on the function of cutaneous sensory neurons in experimental models of pain. We detail the skin–nerve setup requirements and the technique to record action potentials from single sensory fibers. We address how to test for inhibitory effects of opioid receptor activation on mechanical and thermal sensitivity of nociceptors and mechanoreceptors in the complete Freund’s adju- vant-induced inflammation and the chronic constriction injury model of neuropathic pain.

Key words Inflammation, In vitro skin–nerve preparation, Neuropathy, Nociceptor, Opioid receptor

1 Introduction

The in vitro skin–saphenous nerve preparation is a method for extracellular recording of action potentials from peripheral sensory nerve endings [1]. Mechanical, thermal, or chemical stimulation of the receptive field in distinct subgroups of isolated primary sensory neurons evokes action potentials. Mechanoreceptors transduce mechanical stimuli of low intensity to mediate touch sensation, while nociceptors transduce potentially damaging, high intensity stimuli to mediate pain sensation [2]. The skin–nerve preparation is a popular technique in the field of experimental pain medicine. Tissue inflammation and nerve injury are associated with patho- logical processes in primary sensory nerve endings and axons in addition to pathology of the central nervous system. Precise phar- macological manipulation facilitates the examination of the modu- latory effects of drugs and provides an approach to screen for new therapeutic targets. Being a pure sensory nerve, the saphenous nerve is suitable for studies requiring larger samples of sensory neurons.

Santi M. Spampinato (ed.), Opioid Receptors: Methods and Protocols, Methods in Molecular Biology, vol. 1230, DOI 10.1007/978-1-4939-1708-2_17, © Springer Science+Business Media New York 2015

215

216 Rabih Moshourab et al.

2 Materials

2.1 Oxygenated Synthetic Interstitial Fluid

2.2 Setup Skeleton and Perfusion System

Peripheral tissue inflammation is regarded as a regulatory stimulus of opioid receptor plasticity in sensory neurons [3]. Inflammation induces upregulation of opioid receptors in dorsal root ganglion (DRG) neuronal cell bodies and their peripheral endings innervating inflamed tissue [4, 5]. The opioid receptor agonists, fentanyl and morphine, inhibit nociceptor excitability in a rat model of acute and chronic inflammation induced with complete Freund’s adjuvant (CFA) [6, 7]. At the behavioral level, application of mu-, delta-, and kappa-opioid receptor agonists to inflamed tissue produces analgesia mediated by local opioid recep- tors in animals and humans [3]. Peripheral nerve injury is another pathological condition which alters the function of opioid recep- tors. The most studied animal model of neuropathic pain is the chronic constriction injury (CCI) of the sciatic nerve. Following CCI, the opioid receptor immunoreactivity is enhanced in the DRG and at the nerve injury site [8, 9]. Several studies have provided evidence for peripheral analgesic effects of immune cell- derived opioid peptides and exogenous opioids [8, 10–12]. The skin–nerve preparation can be employed to assay the activity of opioid receptors on single primary sensory fibers, test the drug efficacy, and screen for selective, peripherally acting opioid recep- tor agonists as a novel analgesic target.

In this chapter, we describe the basic requirements of the in vitro skin–saphenous nerve setup, the classification of single sen- sory fibers, the stimulation protocols, and the drug application. The procedure will yield an electrophysiological characterization of the effects of opioid receptor activation on single sensory fiber function in animal models of inflammatory and neuropathic pain (Fig. 1a).

1. Synthetic interstitial fluid (SIF): 107.8 mM NaCl, 3.5 mM KCl, 0.69 mM MgSO4, 1.67 mM NaH2PO4, 26.2 mM NaHCO3, 9.64 mM gluconic acid (sodium salt), 5.55 mM glu- cose CaCl2, 7.6 mM sucrose, and 1.53 mM CaCl2. CaCl2 is added after all other components are dissolved in ultrapure distilled water and gassed with carbogen (5 % CO2, 95 % O2) for at least 5 min [1].

2. Cylinder with compressed carbogen gas (5 % CO2, 95 % O2), including pressure-reducing and flow-adjusting valves.

1. Tubings and connectors (e.g., Tygon).

2. Flow regulation pump (e.g., Gilson).

3. Water bath (e.g., Julabo) and a thermometer to control the bath temperature.

Assay of Opioid Sensitive Primary Sensory Nerve Endings 217

Fig. 1 Assessing opioid sensitivity of cutaneous sensory fibers in pathological painful conditions. (a) Scheme showing the induction of the CFA hind paw inflammation (rat) and CCI of the saphenous nerve (mouse). (b) Typical arrangement of the skin–nerve setup. The recording electrode feeds the signals from sensory fibers in the saphenous nerve to a preamplifier connected to the amplifier with a bandpass filter (100–1,000 Hz). The signal is then fed to an oscilloscope, a loudspeaker, and data acquisition system. The system also features software controlled outputs for stimulation and pulse generation. (c) Example of the opioid receptor agonist- induced inhibition of a C-fiber nociceptor. The discharges to the nanomotor mechanical stimulation were recorded at baseline, following opioid application, and after its washout

4. Fiberoptic light source (e.g., Leica, Fiber-Lite).

5. Silicone elastomer (Sylgard 184), a two-component silicone elastomer.

6. The skin–nerve compartment consists of two chambers (Fig. 1b): the nerve recording chamber connected to an adja- cent skin–nerve bath chamber. The whole compartment should be designed to fit the size of the nerve and the skin of the mouse or rat hind limb. The surface of the skin–nerve bath

218 Rabih Moshourab et al.

2.3 Electronic Equipment
for Recording

chamber should be covered with silicone elastomer. The skin– nerve bath chamber should have an integrated perfusion sys- tem by an inlet and outlet. SIF is pumped from a separate container through tubings integrated within a water bath. The adjacent nerve recording chamber needs to be equipped with (a) a mirror on the bottom, onto which the nerve is placed (see Subheading 2.5), (b) one pair of recording gold (or silver) wire electrodes, and (c) one grounding electrode. Both cham- bers need to be connected by a small hole at the bottom (1.5–2 mm diameter). The hole is required to thread the nerve through. A binocular microscope (40×) is mounted above the nerve recording chamber and aids in teasing nerve filaments. The nerve recording chamber is filled with mineral oil for elec- trical isolation of the recording electrode.

1. Amplifier: NeuroLog system (e.g., Digitimer) provides a modular way of carrying out recordings in one apparatus (Fig. 1b). It allows installation of up to 13 modules for signal amplification, filtering, and pulse generation. Alternatively, amplifiers can be obtained from World Precision Instruments (ISO-80) and Warner instruments (DP-301).

2. Oscilloscope: Two-channel digital storage oscilloscope (e.g., Tektronix; Fig. 1b).

3. Loudspeaker (Fig. 1b).
4. Data acquisition system (e.g., Powerlab; ADInstruments) and

standard PC (Fig. 1b). 5. Cables and connectors.

1. Glass rods. Custom-designed glass rods with blunt tips (~1 mm diameter) are useful for mechanical probing of sensory fiber receptive fields in the skin.

2. Set of calibrated von Frey hairs (e.g., Stoelting) for the deter- mination of sensory fiber mechanical thresholds.

3. Metal microelectrodes. They can be custom-made or pur- chased (e.g., Frederick Haer or World Precision Instruments). According to the requirements of the experiment, they have high (e.g., 9–12 MΩ) or low (e.g., 50–100 kΩ) impedance. Materials are epoxy-insulated steel or tungsten.

4. Electrical stimulation device; constant current stimulus isolator (e.g., Digitimer NL800A).

5. Mechanical stimulator (e.g., nanomotor from Kleindieck; Fig. 1b). The stimulator should apply standardized indenta- tions of known magnitude and velocity to the receptive field at regular intervals. The probe is driven by a motor and is in direct contact with the skin. The nanomotor device is equipped

2.4 Devices
for Stimulation
of Sensory Afferents

2.5 Other Tools

with a controller device and software. Optional accessory devices include a force transducer that measures the force of the displacements.

6. A device for thermal stimulation (e.g., Peltier device; Fig. 1b).

7. For chemical stimulation, use custom-designed rings (stainless steel, 1 cm height and ~0.6 cm width) for application of chem- icals on the skin (Fig. 1b).

1. Electric razor.

2. Surgical dissection tools: surgical scissors and forceps (e.g., Fine Science Tools).

3. Watchmaker forceps: several pairs of very fine forceps to tease fibers from the nerve (e.g., Dumont #5SF Forceps from Fine Science Tools).

4. Student microscope (e.g., Leica MS 5, 15×).

5. A mirror glass of about 1 × 2 cm. It is placed in the nerve record- ing chamber. Teasing neuronal fibers under microscopy is much easier when the saphenous nerve is placed on the mirror.

6. Metal plate (size ~60×60 cm) for grounding of the whole setup.

7. Micromanipulators with magnetic base to hold thermal and mechanical stimulators, and microelectrodes in place (e.g., Word Precision Instruments).

1. Before starting the experiments, all animal procedures should be approved by the ethical committees.

2. Care should be taken not to injure the saphenous nerve. Try to keep the dissection time to less than 45 min.

3. Kill the animal with an overdose of isoflurane (e.g., Abbott), by placing it in a glass chamber on a ceramic perforated plate located above paper tissues soaked with isoflurane.

4. Remove the hair on the hind paw and leg with an electric razor.

5. Use small needles to fix the hind paw onto a silicon rubber bottom of a Petri dish.

6. Make a circular incision approximately 2 mm above the knee joint and remove the skin proximal to the incision up to the inguinal ligament to make the saphenous nerve visible (see Note 1).

7. To obtain the paw skin (the part distal to the incision) with the attached saphenous nerve, cut the most distal toe phalan- ges and the skin on the plantar side of the paw and leg toward

Assay of Opioid Sensitive Primary Sensory Nerve Endings 219

3 Methods

3.1 Dissection
of the Saphenous Nerve with the Skin of the Hind Paw

220 Rabih Moshourab et al.

3.2 Preparing for Recording

the knee joint. Then, starting at the remaining phalanges and using very fine scissors, separate the skin from the subcutane- ous tissue and pull it very slowly backwards until the point where the saphenous nerve enters the skin (see Note 2).

8. Free the remaining part of the nerve from the surrounding muscles and blood vessels, and place a knot around the nerve— at its most proximal position within the inguinal ligament—with surgical silk.

1. Perfuse the skin with 15 ml/min of oxygenated SIF. Make sure that the temperature of the water in the bath is between 30 and 32 °C.

2. Mount the skin corium side-up (which facilitates diffusion of compounds to sensory afferents) in the skin–nerve chamber of the skin–nerve compartment. Gently pull the nerve through the hole into the nerve recording chamber and place it on the mirror.

3. Reduce the electrical noise by grounding. This needs to be done for fluids in both chambers, all electrical devices near the recording chamber, the stimulator devices, the microscope, and lamps containing metal parts. Avoid grounding loops by double grounding, as this increases electrical noise.

1. Under the microscope, gently pull off the saphenous nerve sheath to expose the nerve, and dissect fine strands from the nerve with sharpened or very fine watchmaker forceps. Try to obtain fine strands of fibers (see Note 3) and place one of them on the recording electrode. Keep doing this until you are able to identify action potentials from only few separate recep- tive fields in the skin representing the corresponding few single sensory fibers. To induce action potentials, you can electrically stimulate the saphenous nerve. Another way is to mechanically probe the skin with a blunt glass rod to evoke discharges from mechanically sensitive receptive fields.

2. Select a single sensory fiber for further study. This is best done if the software can discriminate the action potential waveform of this fiber from the other waveforms. If this is not possible, select a fiber whose receptive field does not overlap with the receptive fields of other fibers. Adjust the gain to obtain the optimal waveform amplification. If all of this does not help, try to subdivide the fiber strand into thinner ones (but see Note 3).

3. Determine the conduction velocity (CV) for each isolated single fiber. Use the stimulus isolator device to apply constant current electrical stimulations through the microelectrode. Place the microelectrode onto the most sensitive spot of the receptive field.

3.3 Recording form Single Sensory Fibers

Assay of Opioid Sensitive Primary Sensory Nerve Endings 221

Deliver suprathreshold electrical current pulses (see Note 4). Capture the signal, which usually consists of the electrical stimulus artifact followed by the waveform of the identified unit, on the oscilloscope or the data acquisition system. To calculate the CV divide the distance between the receptive field and the recording electrode by the latency of the action potential. The latency is the time delay between the onset of the electrical stimulation artifact and the first action potential of the fiber.

4. Determine the mechanical threshold by using calibrated von Frey hairs. Start to test by applying always the same von Frey hair (we started with the 0.13 g hair) onto the fiber’s receptive field. If the fiber does not respond, test the next stronger hair. If the fiber responds to the stimulation by reliably firing more than one action potential, use the next weaker von Frey hair. The threshold can be defined as the force of the smallest von Frey hair necessary to evoke at least one action potential. Try to keep an interval of at least 10–20 s between individual von Frey hair applications.

5. Characterize the response properties of a single sensory fiber to standardized mechanical stimulation. We use a computer- controlled nanomotor that delivers ramp-and-hold displace- ment stimuli (see Subheading 2.4). Different diameter probes can be attached to the nanomotor. We use a stainless-steel rod with a flat circular contact area of 0.5 mm2. The nanomotor can also be equipped with a force transducer to measure the applied force. Before starting the stimulation protocol, posi- tion the nanomotor just above the receptive field. Use small movements (e.g., 100 μm) to advance onto the receptive field until one action potential is evoked. Then, move the probe the same length (100 μm) upwards again. Apply the same approach while stepwise reducing the movement to the smallest stimulus (e.g., 10 μm) used in the nanomotor stimulation protocol. The position of the probe is therefore just above the threshold. Apply a standardized protocol of increasing displacement stimuli at regular intervals to the receptive field (see Note 5). The response characteristic to constant mechanical stimulation of different displacements (forces) allows to distinguish slowly from rapidly adapting fibers (see Subheading 3.4).

6. Determine the heat and cold sensitivity. We apply heat stimula- tion of up to 52 °C, and cold stimuli of down to 12 °C with a Peltier-based device. Peltier device with contact dimensions of 5×5 mm can gradually heat or cool the skin with a rate of 1 °C/s. Place the device directly over the receptive field but be careful not to mechanically stimulate the receptive field.

222 Rabih Moshourab et al. Table 1

Classification of cutaneous sensory fibers in hairy skin of mouse and rat

CV conduction velocity, RAM rapidly adapting mechanoreceptors, SAM slowly adapting mechanoreceptors, AM A-mechanonociceptor, CM C-mechanonociceptor, CMH C mechano-heat nociceptors. The sample recordings show typical patterns of discharge of mouse mechanoreceptors and nociceptors from the saphenous nerve to a standardized ramp and hold indentation stimulus. A typical CMH response to a heat stimulus is also shown

Type

CV (m/s)

Subtype

Response properties

Sample recording

Mechanoreceptors

Aβ

>10, mouse >13, rat

RAM

Rapidly adapting
Low mechanical threshold

SAM

Slowly adapting
Low mechanical threshold

Aδ

1–10, mouse 1.6–13, rat

D-hair

Rapidly adapting
Low mechanical threshold

Nociceptors

Aδ

1–10, mouse 1.6–13 rat

High mechanical threshold

C fibers

<1, mouse <1.6, rat

High mechanical threshold, heat-insensitive

CMH

High mechanical threshold, heat-sensitive

3.4 Classification of Single Sensory Fibers

1. Classify single units based on their CV into Aβ, Aδ, or C fibers (see Table 1).

2. Aβ and Aδ fibers can be further classified into subclasses based on their mechanical threshold and response properties to a mechanical stimulus. Aβ fibers subclasses are low threshold slowly (SAM) or rapidly (RAM) adapting mechanoreceptors. Aδ fibers contain low threshold, rapidly adapting D-hair fibers, and

3.5 Opioid Application

3.6 Assaying Opioid Sensitivity of Sensory Fibers Under Pathological Conditions

high threshold mechanonociceptors, the Aδ mechanonociceptors (AM). AM characteristically increase their discharge to increas- ing stimulus intensities [13, 14]. Receptive fields excited by von Frey hairs of more than 6.1 bars (80 mN hair with 0.406 mm diameter) are very-high threshold or mechanically insensitive units [15].

3. Classify C fibers based on thermal sensitivity. C fibers respond- ing to thermal—heat or cold—and mechanical stimulation belong to the “polymodal” C fiber group, designated as C mechano-heat nociceptors (CMH), C mechano-heat and cold (CMHC), or C mechano-cold (CMC). C-mechanonociceptors (CM) are mechanosensitive but insensitive to thermal stimula- tion. Low threshold C fibers (CLT) have non-nociceptive tactile function, von Frey mechanical threshold <6 mN, and afterdischarges to mechanical stimulus removal [16].

Use the metal ring to isolate the receptive field of a sensory fiber (Fig. 1b). Dissolve and apply the opioid agonist in SIF for 2–3 min (see also Subheading 3.7). It is suggested to use concentrations of opioid agonists in the nM to μM concentration range [6, 7, 17]. Nevertheless, for a proper analysis, a concentration–response test- ing is advisable (see Subheading 3.9).

1. To study somatic inflammatory pain, induce inflammation using an inflammatory agent; we used CFA injected into one hind paw of Wistar rats [7]. Deeply anesthetize the rat, e.g., by placing it in a glass chamber on a ceramic perforated plate located above paper tissues soaked with isoflurane, and subcu- taneously inject 150 μl of CFA (e.g., Calbiochem) into the ventromedial area of one hind paw (Fig. 1a). The time after which you isolate the skin and nerve following CFA injection depends on the stage of inflammation you are interested in. We assayed sensory fibers for opioid sensitivity on days 3–4 follow- ing CFA application [7]. Other investigators have performed their assays after 18–24 h [6].

2. To study neuropathic pain in vivo, a common model used is the CCI of the sciatic nerve [18]. This model was also adapted to the saphenous nerve [19], and we used it in our study [17]. To deeply anaesthetize the mouse, place it in a glass chamber on a ceramic perforated plate located above paper tissues soaked isoflurane, until the animal looses consciousness. Subsequently, cover the animal’s nose by a tube attached to an anesthesia machine delivering a gaseous mixture of isoflurane (3–4 %) and oxygen. To induce CCI, expose the saphenous nerve at the level of the right thigh by making a small incision on the anterior surface of the thigh. Under the microscope, free the saphenous nerve (running parallel to the saphenous

Assay of Opioid Sensitive Primary Sensory Nerve Endings 223

224

Rabih Moshourab et al.

3.7
of Opioid Effects on Evoked Activity in Sensory Fibers

vein) from the surrounding muscle, and loosely tighten three nylon sutures (8-0) around the nerve (see Note 6 and Fig. 1a). Close the wound with silk sutures. We assayed opioid sensi- tivity of single nerve fibers on days 10–14 following CCI (see Note 7).

1. Following identification of the sensory fiber type and deter- mination of the mechanical threshold with von Frey hairs (see Subheadings 3.3 and 3.4), stimulate the receptive field of a single mechanosensitive fiber to a first series of graded mechanical stimuli (applied, e.g., by a nanomotor; see Subheading 3.3) under SIF/vehicle exposure. This serves as the baseline recording. Record the total number of evoked action potentials during each 10 s mechanical stimulus. We used the LabChart (ADInstruments) spike histogram extension software program for offline analysis. If desired, record the mechanical latency (time between the onset of the ramp and the first recorded action potential corrected for conduction delay).

2. At least 5 min after the nanomotor-induced mechanical stimu- lation, isolate the receptive field with a metal ring (diameter 3–4 mm) to separate it from the SIF in the organ bath. Remove the SIF within the metal ring.

3. Dilute the opioid receptor agonist (from a stock solution, the best kept on ice) in fresh, pre-warmed SIF (a total volume of 100 μl should be sufficient). Apply the opioid to the metal ring for 2–3 min.

4. Apply a second series of mechanical stimulations while the metal ring is in place. The dimensions of the nanomotor probe should be small (preferably <1 mm diameter). If this is not possible, apply the second series of nanomotor stimulation immediately after removal of the drug and the ring (see step 6 below).

5. Remove the contents of the ring and remove the metal ring. Measure the mechanical threshold a second time, to see whether opioid application induced any changes in the threshold. We defined a change by one von Frey hair force as an opioid-sensitive response (see also Subheading 3.8 for selection of opioid- sensitive fibers).

6. Run a second mechanical stimulation series with the nanomotor.

7. Depending on drug affinity and lipid solubility, drug washout can take between 15 and 30 min (or longer). Assess the recov- ery with a third determination of the mechanical threshold and a third series of mechanical nanomotor stimulation.

8. The same procedure can be applied for heat and cold stimulation. The difficulty is the thermal stimulation under simultaneous drug

Analysis

3.8 Identification of Opioid-Sensitive Sensory Fibers

application, which is technically challenging because of the size of the Peltier element. Apply thermal stimulation immediately after removal or metal ring and drug.

1. Sensory fiber responses to successive mechanical stimulations of the same magnitude can vary substantially [20]. To distin- guish a normal change in a fiber response to repeated stimula- tion from a potential drug effect, determine the upper and lower range of changes by conducting pilot experiments with repeated stimulations using a drug vehicle. The cutoff ranges are usually two standard deviations from the mean response. Alternatively, one can express the cutoff as the percentage of baseline response [6, 7].

2. Categorize sensory fibers as opioid-sensitive, if their response to mechanical stimulation with the nanomotor deviates by more than two standard deviations below or above the mean. Using this criterion, drug-induced effects are distinguished from normal variability due to repeated stimulation [7].

3. Alternatively, identify opioid-sensitive fibers by assaying changes in the von Frey mechanical threshold. We defined a change by one von Frey hair force as an opioid-sensitive response (pro- viding that the change could be reversed by washout and prevented by opioid-receptor antagonist application) [17].

In the hind paw skin of animals with peripheral inflammation or nerve injury, opioid receptor agonists, morphine, fentanyl, or [D-Ala2,N-Me-Phe4,Gly5]-ol-enkephalin (DAMGO), reduce dis- charges to thermal and/or mechanical stimulation in about 40–50 % of C and Aδ nociceptors [6, 7, 17]. Fig. 1c depicts an example of opioid-induced decrease of C-mechanonociceptor dis- charges, and the recovery following the opioid washout, to the nanomotor mechanical stimulation, in the CFA inflamed rat skin. The criterion for opioid sensitivity has been introduced above (see Subheading 3.8). The reduction in discharges is dose-dependent. Apply at least three different concentrations (e.g., 10 nM to 10 μM) of the agonist. For each concentration, a new set of sensory fibers should be sampled, i.e., a single fiber is subjected to only one opioid concentration application.

1. To check whether the effect of the agonist is mediated via activation of opioid receptors, use specific opioid receptor antagonist following the washout of the agonist.

2. Isolate the receptive field of an opioid-sensitive sensory fiber again with the metal ring. Remove the SIF in the ring and apply the opioid receptor antagonist for 3 min. Depending on the selectivity and affinity of the antagonist at hand, equimolar and up to five times higher concentration relative to the ago- nist concentration can be applied.

3.9 Dose–Response of Opioid Effects

3.10 Testing Opioid Receptor Specificity

Assay of Opioid Sensitive Primary Sensory Nerve Endings 225

226 Rabih Moshourab et al.

3.11 Controls

3. Remove the opioid receptor antagonist solution from the metal ring and immediately apply the respective opioid recep- tor agonist (see Subheading 3.7) before measuring the mech- anical threshold and running another nanomotor series (see Note 8).

1. To control for the opioid receptor agonist action, correspond- ing experiments need to be performed following application of the vehicle.

2. To control for tissue inflammation, corresponding experiments need to be performed in skin–nerve preparations obtained from naïve animals (or animals injected with 0.9 % NaCl, instead of CFA).

3. To control for nerve injury, corresponding experiments need to be performed in skin–nerve preparations obtained from naïve and sham-operated animals. For sham surgery, follow step 2 in Subheading 3.6, but do not ligate the nerve.

1. Keep the exposed nerve and muscle slightly covered with oxy- genated SIF throughout the duration of the preparation. We do this by repetitive rinsing of the exposed tissue with SIF using a small syringe.

2. Pulling the skin backwards is, to our experience, possible in a few week old mice; the older the mice the more difficult the skin detaches from the underlying tissue. In this case, use fine scissors to cut the skin carefully. The later case also applies for rats as they have much thicker skin.

3. The mirror on which the cut end of the nerve is placed should be almost free of SIF to allow easier teasing of the nerve and avoid short circuits when SIF contacts the recording elec- trode. Identifying a single robust fiber (i.e., clearly responding by firing action potentials upon stimulation of its receptive field in the skin) can take some time and depends on the qual- ity of the skin–nerve preparation. Moreover, nerve injury (e.g., a CCI of the saphenous nerve) decreases the chance of getting a robust single fiber recording because of de- and regeneration processes taking place predominantly in the dis- tal (recording) part of the nerve. Moreover, too fine fiber strands may hinder robust recording, because very fine strands are more vulnerable to detachment from the recording elec- trode and to rupture.

4 Notes

Acknowledgements

4. Stimulate the receptive field using a sharp tungsten metal electrode with suprathreshold current pulses of 50, 150, or 500 μs. To evoke action potentials in Aβ fibers, pulse durations of 50 μs are often sufficient, but longer pulse durations might be needed to excite Aδ and C fibers.

5. Mechanical stimuli of magnitude between approximately 10–600 μm displacements can be delivered. We applied dis- placements of 10 s duration with an interval of at least 30 s. The nanomotor (Kleindieck) displacements are in linear relationship with the force delivered to the skin [13]. The nanomotor is computer-controlled and the stimulation programs can be designed as needed. We usually use an ascending series of up to seven mechanical stimulations.

6. We use an anesthesia machine to maintain the anesthesia, because CCI of the saphenous nerve is relatively time- consuming (approximately 15–20 min for an experienced researcher). Immediately after opening of the skin, place a drop of 0.9 % NaCl on the wound. If it is placed appropriately, it stays there throughout the whole operation; if this is not the case, apply 0.9 % NaCl as often as needed. Drying of the nerve resulted in recording difficulties in our experiments. The 0.9 % NaCl solution can also be used if a blood vessel is accidently disrupted and blood has to be rinsed out of the wound.

7. Single sensory fiber activities are best recorded 10–14 days following CCI. Tremendous difficulties were encountered on days 1 or 2 [17, 19]. The problem might be related to the extensive Wallerian degeneration of sensory fibers during the first 2 days following nerve injury [21, 22].

8. Difficulties might occur when using multiple series of mechan- ical stimulations (e.g., baseline, following opioid application, following washout, and following antagonist application). Not all sensory fibers might survive the corresponding longer stim- ulation period because often signal rundown (decrease in sig- nal amplitude with time) and fiber fatigue (decreased sensitivity and discharge) occurs. To avoid this problem, keep the num- ber of stimulations within a nanomotor stimulation series as low as possible and try to complete the stimulation protocol within 45 min.

This work was supported by the Deutsche Forschungsgemeinschaft grant (MA 2437/1-4; H.M) and the Charité-Universitätsmedizin Berlin funds (R.M.).

Assay of Opioid Sensitive Primary Sensory Nerve Endings 227

228 Rabih Moshourab et al.

References

1. Zimmermann K, Hein A, Hager U et al (2009) Phenotyping sensory nerve endings in vitro in the mouse. Nat Protoc 4:174–196

2. Lewin GR, Moshourab R (2004) Mechano- sensation and pain. J Neurobiol 61:30–44

3. Stein C, Schafer M, Machelska H (2003) Attacking pain at its source: new perspectives on opioids. Nat Med 9:1003–1008

4. Puehler W, Rittner HL, Mousa SA et al (2006) Interleukin-1 beta contributes to the upregula- tion of kappa opioid receptor mrna in dorsal root ganglia in response to peripheral inflam- mation. Neuroscience 141:989–998

5. Zollner C, Shaqura MA, Bopaiah CP et al (2003) Painful inflammation-induced increase in mu-opioid receptor binding and G-protein coupling in primary afferent neurons. Mol Pharmacol 64:202–210

6. Wenk HN, Brederson JD, Honda CN (2006) Morphine directly inhibits nociceptors in inflamed skin. J Neurophysiol 95:2083–2097

7. Moshourab R, Stein C (2012) Fentanyl decreases discharges of C and A nociceptors to suprathreshold mechanical stimulation in chronic inflammation. J Neurophysiol 108: 2827–2836

8. Truong W, Cheng C, Xu QG et al (2003) Mu opioid receptors and analgesia at the site of a peripheral nerve injury. Ann Neurol 53: 366–375

9. Schmidt Y, Gavériaux-Ruff C, Machelska M (2013) μ-Opioid receptor antibody reveals tissue-dependent specific staining and increased neuronal μ-receptor immunoreactivity at the injured nerve trunk in mice. PLoS One 8:e79099

10. Obara I, Przewlocki R, Przewlocka B (2004) Local peripheral effects of mu-opioid receptor agonists in neuropathic pain in rats. Neurosci Lett 360:85–89

11. Obara I, Parkitna JR, Korostynski M et al (2009) Local peripheral opioid effects and expression of opioid genes in the spinal cord and dorsal root ganglia in neuropathic and inflammatory pain. Pain 141:283–291

12. Kolesnikov Y, El-Maarouf A, Rutishauser U et al (2007) Reorganization of dorsal root

ganglion neurons following chronic sciatic nerve constriction injury: correlation with morphine and lidocaine analgesia. Eur J Phar- macol 568:124–133

13. Milenkovic N, Wetzel C, Moshourab R et al (2008) Speed and temperature dependences of mechanotransduction in afferent fibers recorded from the mouse saphenous nerve. J Neuro- physiol 100:2771–2783

14. Treede RD, Meyer RA, Campbell JN (1998) Myelinated mechanically insensitive afferents from monkey hairy skin: heat-response proper- ties. J Neurophysiol 80:1082–1093

15. Meyer RA, Davis KD, Cohen RH et al (1991) Mechanically insensitive afferents (MIAs) in cutaneous nerves of monkey. Brain Res 561: 252–261

16. Olausson H, Wessberg J, Morrison I et al (2010) The neurophysiology of unmyelinated tactile afferents. Neurosci Biobehav Rev 34: 185–191

17. Schmidt Y, Labuz D, Heppenstall PA et al (2012) Cutaneous nociceptors lack sensitisa- tion, but reveal mu-opioid receptor-mediated reduction in excitability to mechanical stimula- tion in neuropathy. Mol Pain 8:81

18. Bennett GJ, Xie YK (1988) A peripheral mono- neuropathy in rat that produces disorders of pain sensation like those seen in man. Pain 33:87–107

19. Walczak JS, Pichette V, Leblond F et al (2006) Characterization of chronic constriction of the saphenous nerve, a model of neuropathic pain in mice showing rapid molecular and elec- trophysiological changes. J Neurosci Res 83: 1310–1322

20. Slugg RM, Meyer RA, Campbell JN (2000) Response of cutaneous A- and C-fiber nocicep- tors in the monkey to controlled-force stimuli. J Neurophysiol 83:2179–2191

21. Koeppen AH (2004) Wallerian degeneration: history and clinical significance. J Neurol Sci 220:115–117

22. Lin YW, Tseng TJ, Lin WM et al (2001) Cutaneous nerve terminal degeneration in painful mononeuropathy. Exp Neurol 170: 290–296

Chapter 18

Mechanical Nociception Measurement in Mice and Rats with Automated Von Frey Equipment

Gabriele Campana and Roberto Rimondini Abstract

Von Frey hairs are important tools for the study of mechanisms of cutaneous stimulation-induced sensory input. Mechanical force is exerted via application of a particular hair to the cutaneous receptive field until buckling of the hair occurs. The most commonly used Von Frey filaments are productive in evaluating behavioral responses of neuropathic pain in preclinical and clinical research. To reduce the potential experi- menter bias, automated instruments are being developed for behavioral assessment.

Key words Animal model, Dynamic plantar aesthesiometer, Nociception, Opioid receptor, Von Frey hairs

1 Introduction

Several nociception assays have been developed over the years. One of these is represented by the Von Frey test that involves mechani- cal nociception or mechanical sensibility [1]. The Von Frey test consists of thin filaments (hairs) applied to the plantar surface of the hindpaw. The force applied to the hairs that evokes hindpaw withdrawal represent the threshold. The test could be automatized with the Dynamic Plantar Aesthesiometer (Fig. 1) that has been designed to automate the assessment of “touch sensitivity” on the plantar surface of rats or mice [2]. With this device, latency and actual force at the time of paw withdrawal reflex are automatically detected and recorded. The device consists of transparent plastic cages over a mesh platform. Under the platform the device is free to be moved in order to reach with the calibrated hair the hindpaw of the animal and once is in position a force is applied until the withdraw. Latency and actual force at the time of paw withdrawal reflex are automatically detected and recorded by the unit.

Santi M. Spampinato (ed.), Opioid Receptors: Methods and Protocols, Methods in Molecular Biology, vol. 1230, DOI 10.1007/978-1-4939-1708-2_18, © Springer Science+Business Media New York 2015

229

230 Gabriele Campana and Roberto Rimondini

2 Materials

2.1 Apparatus

2.2 Reagents

2.3 Animals

3 Methods

3.1 Calibration of the Plantar Aesthesiometer

Fig. 1 Dynamic plantar aesthesiometer apparatus

Dynamic Plantar Aesthesiometer (Ugo Basile, Varese, Italy) (see Note 1).

Prepare all solution using distillated water and store all reagents at room temperature. Prepare 1 l of ethanol solution at 70 % v/v to clean the test grid metal cage at the end of each animal trial.

Rats or mice can be used.

Animals should be placed in the experimental room 1 h before the test.

1. First, make sure the probe is mounted on its pushing rod. Then, without any weight on the probe, perform the zeroing following the instruction on the screen.

2. Remove the probe and run the calibration with an additional weight. Select the calibration weight between the two avail- able: (a) 5 g calibration weight for trials which involve forces from 0 to 5 g (typical for Mice but used for rats); (b) 50 g cali- bration weight for trials which involve forces from 5 to 50 g (for Rats).

3. If you select the 5 g weight, start the calibration by depressing the 5 g key. If the 50 g weight is involved, start the calibration by depressing the 50 g key.

4. After weight selection, perform the calibration following the instruction on the screen.

3.2 Carrying Out the Assay

At the end of the 1-h room habituation, animals are placed in plastic cages with a metal grid bottom at least 45 min (rat) or 2 h (mice) prior to testing to allow accommodation (see Note 2).

Paw withdrawal latency to mechanical stimulation is assessed with the automated testing device consisting of a steel rod (diam- eter 0.5 mm) that is pushed against the plantar surface of the paw with increasing force until the paw was withdrawn (see Note 3).

When the animal withdraws its hindpaw, mechanical stimulus automatically stops, and the force at which the animal withdraws its paw is recorded to the nearest 0.1 g.

The paw withdrawal latency and actual force at the time of paw withdrawal reflex are calculated as the mean of 6 consecutive trials (three alternated trials per paw with a 2-min intertrail) (see Note 4).

1. Equipment should be place in a quite experimental room with controlled light and temperature. Direct daylight should be avoided.

2. This step is really important to avoid uncorrected value due to the spontaneous locomotor activity of the no-habituated animals.

3. The maximum force is set to 5 g to prevent tissue damage and the ramp speed is 0.5 g/s.

4. The main advantages of the Dynamic Plantar Aesthesiometer over the classic manual filaments are: (a) the measurement is continuous, i.e., you apply the increasing force with the same rigid filament until you see a response, rather than having to change filaments; (b) the operator does not personally apply the force, thus avoiding human error, as it is the device that applies the force at the rate and intensity that the operator sets; (c) The individual animal plexiglas enclosure allows the animal to be in an enclosed area, but unrestrained and untouched by the operator.

4 Notes

References

1. Zuker K (1951) Sensory examination with 2. Nirogi R, Goura V, Shanmuganathan D et al

von Frey’s hairs, and the problem of a basic function of the brain. Dtsch Z Nervenheilkd 165: 109–126

(2012)Comparisonofmanualandautomatedfila- ments for evaluation of neuropathic pain behavior in rats. J Pharmacol Toxicol Methods 66:8–13

Mechanical Stimulation and Nociception 231

Chapter 19

Detecting Zinc Release Induced by Mu-Opioid Receptor Agonists in Brain Slices

María Rodríguez-Muñoz, Pilar Sánchez-Blázquez, Concha Bailón, and Javier Garzón

Abstract

After iron, zinc is the second most abundant transition metal in living organisms and it is known to be a contributory factor in a series of neurological disorders. In biological systems zinc exists as either bound Zn2+, representing the majority of the total zinc in tissues, or free (chelatable) Zn2+. Several fluorescents dyes have been developed to detect free zinc when it is released from zinc-binding proteins, which occurs via redox mechanisms in response to the stimulation of a number of neurotransmitter receptors, including the μ opioid receptor. Here we describe a detailed protocol to detect drug stimulated intracellular zinc release in rodent brain slices using time-lapse microscopy and fluorescence imaging.

Key words Brain slices, Fluorescence, Imaging, Opioid receptors, Redox signaling, Time lapse, Zinc

1 Introduction

Opioids are probably the most effective drugs in attenuating nociceptive perception and accordingly, they are the analgesics of choice to relieve intense chronic pain. These substances exert these effects by binding to specific μ (MOR), delta, and kappa receptors in the nervous system. A series of studies have revealed that the activity of the NMDAR/NO/CaMKII cascade in neural tissue is mostly responsible for the development of tolerance to morphine antinociception [1]. This negative regulation of the effects of MORs by NMDARs provides the basis for the clinical efficacy of NMDAR antagonists in preventing, and even rescuing, mor- phine analgesia from tolerance [2]. A series of signaling proteins participate in this MOR-NMDAR cross-talk [3], and this process is mediated by redox mechanisms and zinc metabolism. In fact, the free zinc ions required in these events are initially provided by the morphine-induced activation of the nNOS pathway [4] and the subsequent action of NO on intracellular zinc stores [5], in

Santi M. Spampinato (ed.), Opioid Receptors: Methods and Protocols, Methods in Molecular Biology, vol. 1230, DOI 10.1007/978-1-4939-1708-2_19, © Springer Science+Business Media New York 2015

233

234 María Rodríguez-Muñoz et al.

2 Materials

2.1 Animals

2.2 Dissection Tools and Brain Slicers

particular on RGS-Rz [6, 7]. The protocol presented here describes a method to image zinc in brain slices, with the ultimate objective of assessing its influence on morphine-induced antinociception and on the development of analgesic tolerance. This technique can be used to detect Zn2+ influxes in rodent brain slices but also in cultured cells.

The animals used in the experiments (see Note 1) were housed and used strictly in accordance to NIH, EC (Directive 2010/63/EU), or IBRO guidelines. All solutions were prepared using ultrapure water and analytical grade reagents. The surgical material must have been sterilized and disposable gloves should be used throughout the procedure.

Finest grade stainless steel scissors, scrapers, needles, razor blades, and petri dishes should be used (Fig. 1), and we also used a Rodent Brain Matrix (see Note 2) to prepare the slices. Each rodent brain slicer matrix is designed with anatomically proportioned cavity dimensions and the slice channels were spaced at precise intervals. Depending on the matrix model, coronal or sagittal sections of a rodent brain can be obtained. It is important to follow the appro- priate regulations when disposing of animal tissues.

Fig. 1 Dissection tools and brain slicer setup. Fine grade stainless steel surgical materials should be used. The dissection should be performed rapidly, having introduced the materials into an ice container. The brain should be kept at 4 °C to reduce cell damage and to improve its texture for slicing

2.3 Working Solutions

1. Cutting solution: prepare the desired volume of this solution in oxygenated pure water from concentrated stocks (10×) stored at 4 °C. The ionic composition of this is often close to (in mM): 124 NaCl, 1.75 KCl, 2.4 CaCl2, 1.3 MgSO4, 1.25 KH2PO4, 26 NaHCO3 (see Note 3). Keep the solution ice- cold during the procedure.

2. Artificial Cerebrospinal Fluid (ACSF): prepare the solution as indicated above and add glucose (10 mM). Mix for about 10 min at room temperature and maintain this solution at a physiological temperature (approximately 32 °C) and glass with 5 % CO2 in O2 (see Note 4).

Brain slices were mounted onto an eight well borosilicate cham- bered coverglass (Fig. 1), a plate that is perfectly designed for confocal image analysis. The nonremovable glass is thin, allowing high magnification examination on an inverted microscope. The polystyrene upper structure is attached to the glass with a nontoxic medical grade silicone adhesive. Pre-coated products offer many advantages to researchers and they include collagen, fibronectin, and poly-d-lysine coatings. Different suppliers offer such pre-coated multiwell plates in different formats, and these products avoid the lengthy coating process; they ensure that the plates are coated in a consistent manner and that they are always available when needed.

A variety of Zn2+ indicators have been developed to help elucidate the role of Zn2+ release and to localize free or chelatable Zn2+ in cells. Of these, the Newport GreenTM DCF Diacetate is a cell per- meant indicator (Invitrogen N-7991) that displays moderate zinc- binding affinity (Kd for Zn2+ ~1 μM) and that is essentially insensitive to Ca2+ (Kd for Ca2+ >100 μM), making this a valuable probe to detect Zn2+ influx into neurons through voltage- or glutamate- gated channels. The diacetate ester form of Newport Green enters cells where endogenous esterases hydrolyze the ester to form the membrane impermeable free acid form, effectively trapping the dye inside the cell [8]. Stock solutions (1–5 mM) are prepared fresh in anhydrous dimethylsulfoxide (DMSO) for single use, and they are stored desiccated at −20 °C and protected from light. In these con- ditions the compound is stable for about 6 months.

Live brain tissue experiments were performed using a LEICA AF 6500/7000 Time Lapse advanced fluorescence imaging system, equipped with a CO2 incubator chamber, a high resolution mono- chrome camera (ANDOR DU8285), an automatic wheel system, and different emission-excitation filters that allow high-speed acquisition of sets of images. Green fluorescence was monitored with excitation and emission settings of 480/40 nm and 527/30 nm, respectively. A HCX PL FLUOTAR 10.0×0.30 DRY objective was used, and the images were obtained automatically (see Note 5).

2.4 Recording Chambers

2.5 Indicators for Zinc

2.6 Fluorescence Microscopy

Free Zinc-Imaging 235

236 María Rodríguez-Muñoz et al.

3 Methods

3.1 Tissue Slice

1. This procedure should be carried out rapidly and at 4 °C to reduce cell damage, and to improve the texture for slicing.

2. Swiftly decapitate the animal with surgical scissors and remove the brain from the skull (see Note 6). Place the brain into a dish of ice-cold cutting medium gassed with 95 % O2 and 5 % CO2 to eliminate blood.

3. Place the brain rapidly in the previously chilled Rodent Brain Matrix (ventral side up, 1–4 °C) such that when properly set in the matrix, the brain’s ventral surface should be parallel to the top of the mould. Coronal 1 mm brain slices (between 1.70 and 0.14 mm to bregma) are then obtained (Fig. 2).

4. To obtain the slices, the first blade should be placed in the channel that lies at the caudal most boundary of the tissue and then two blades should be placed in the channel that

Fig. 2 Slice preparation. Expose the brain by cutting the skull shallowly up the midline from the foramen magnum to the level of the eye sockets. Gently slide the spatula under the brain. Flip the brain out of the skull into the ice-cold Rodent Brain Matrix. After slicing, the sections were divided into two groups for dupli- cate, vehicle and morphine treatment, and they were immediately transferred to the eight-well chamber

3.2 Slide Treatment and Image Acquisition

corresponds to the rostral boundary. Each channel utilized adds 1 mm to the thickness of the slice. Once all three blades are in place, the first and second blade can be removed while leaving the third (rostral most blade) in place, and the desired slice will be lifted out between these first two blades. The third blade, which is left in place, serves two purposes: (a) it prevents any movement of the brain within the matrix; and (b) it now serves as the first blade which will be utilized to remove the next slice (see Note 7).

5. Once all the slices have been obtained, the sections are divided into two to obtain duplicates and with the help of a glass pipette, they are immediately transferred to an eight-well chambered slide containing 300 μl of ACSF preheated to 32 °C (see Note 8). The plate is placed in an incubator cham- ber (32 °C, 5 % CO2) integrated on a microscope for 30 min, positioning the chambered slide in the adapter and ensuring that it is held firmly in place. The cover should be left on the chambered slide so that the ACSF medium does not dry out.

6. Prior to loading the dye, fix the preferred image acquisition conditions in the Time Lapse advanced fluorescence imaging system (see Note 9). When mounting free-floating tissue slices, it is essential to prevent their movement during imaging.

7. Select the frequency and length of the time lapse, the exposure time for each color (if using multiple settings), the number and location of the stage positions, and the number of z-sections (if desired). Also, mark an appropriate number of stage posi- tions so that adequate sampling of the cell population can be achieved. Finally, select where to save the data files.

8. Dilute an aliquot of the Newport Green DCF diacetate (NG) stock solution (prepared in DMSO) in the buffered ACSF medium to a final concentration of 50 μM. Adding the non- ionic detergent Pluronic F127 (0.1 %) can help disperse the dye in the physiological media (see Note 10). The slices can then be incubated with NG for 45–60 min at 32 °C.

9. Before taking fluorescence measurements, wash the slices with indicator free ACSF medium (3×300 μl), which should be taken as the zero, basal fluorescence.

1. Using our protocol, we investigated whether the redox-induced mobilization of endogenous zinc was a signaling property shared by a range of neural G protein coupled receptors (GPCRs), including the MOR [6]. For sustained stimulation, saline, agonists and antagonists (1–10 μM in 15 μl) were added to the wells from concentrated stock solutions.

2. After the drugs were added, images were acquired according to the preestablished parameters. To monitor ligand-induced Zn2+, we acquired images at two wavelengths every 10 or

Free Zinc-Imaging 237

238 María Rodríguez-Muñoz et al.

3.3 Processing and Image Quantification

15 min over 1 h and at 6–8 stage positions. Incubating mouse brain slices with the opioid agonist morphine increased Newport Green fluorescence (Fig. 3), indicating that the zinc ions are being released from endogenous stores in response to MOR activation by the agonist (see Note 11). The assay was typically repeated at least twice on brain coronal slices from different mice.

1. After acquisition, the image data must be converted into a form that is easier to handle. Images acquired with the time- lapse station are usually saved as digital computer files in a format that allows them to be easily manipulated with the pro- prietary software provided as part of the LEICA Microsystems setup, or with specific software such as Image J ver. 1.47t.

2. The intracellular Zn2+ released in response to the addition of the opioid agonist was determined as the difference in fluores- cence intensity between the baseline image and that registered after treatment of the slices (see Note 12). The size and resolu- tion of the images captured (a pixel depth of 8 bits) was identi- cal for each treatment (vehicle or compounds). The grayscale images were resolved into 256 shades of gray, where 0 is black and 255 corresponds to white. For each image, histograms of the number of pixels with a given shade of gray were plotted (Fig. 3), and the mean luminosity of the control and treated images was then computed (AlphaEase FC Software). The fluo- rescence signal can be expressed as the percentage %A = (Ati×100/A0) or as an increased ∆A=(Ati/A0) in the area detected in each image with respect to the baseline.

3. The fluorescent emission intensity can also be quantified by defining a region of interest (ROI) in each frame, and the ∆F or ∆F/F0 was calculated for the selected ROI, where “F0” refers to the baseline fluorescent emission intensity: [9]. The data were analyzed to determine the mean of the luminosity and the 95 % confidence intervals using Sigmaplot/Sigmastat v12 [10]. Alternatively, the green, blue, and red channels can be quantified independently [6]. Representative color indexed images that are presented in pseudocolor can be found in recent publications [6, 10, 11].

Fig. 3 (continued) Leica LEICA AF 6500/7000 Time Lapse advanced fluorescence imaging system. The data shown were obtained 15, 30, and 60 min posttreatment with morphine (10 μM). Scale bar = 100 μm. To see similar illustrations in color, the reader is referred to the publications (see refs. [6, 9, 10]). Right side: Histograms of the number of pixels with a given shade of gray are shown. Each register is expressed as the mean luminos- ity (grayscale of 256 channels; 0 black and 255 white) for both the control (vehicle treated, a) or morphine- treated slices. The minimum and maximal luminosity in the image are also indicated

12000

10000

8000

6000

4000

2000

12000

10000

8000

6000

4000

2000

12000

10000

8000

6000

4000

2000

12000

10000

8000

6000

4000

2000

0 50 100 150 200 250

Pixels

Min. Max. Mean

125316

80 170

0 50 100 150 200 250

Fig. 3 Fluorescence images showing the effect of morphine on zinc mobilization from endogenous stores. Left side: Coronal slices of the mouse frontal cortex were oxygenated and preloaded for 1 h with the cell-permeable Newport Green diacetate (50 μM), assaying the preparations after its removal. The evolution of spontaneous endogenous zinc release was determined in the Control sections (untreated, a), obtaining fluorescent images at various intervals (b, c, d). The images were obtained through a HCX PLfluotar 10.0 × 0.30 dry objective on a

Min. 30 Max. 127 Mean 97.66

Std. Desviation 21.34

142.23 Std. Desviation 18.42

Pixels

Min. Max. Mean

Std. Desviation

Pixels

Min. Max. Mean

Std. Desviation Pixels 125316

125316

110 207 174.60

19.85 125316

140 235 203.41

16.83

240 María Rodríguez-Muñoz et al.

4 Notes

1. As an alternative to naïve animals, knock-out strains with tar- geted disruption of a particular protein, as well as acute or chronic drug-treated mice, can also be used.

2. A vibrating microtome (“vibratome”) may be also used to obtain thinner slices (200–500 μm). However, it may take considerably longer to cut a series of brain slices using a vibratome, as the slices are usually obtained with the blade moving forward slowly to reduce mechanical damage of the superficial cell layers.

3. To avoid the CaCl2 precipitating out of the solution it should be carefully added to the solution by bubbling it with 5 % CO2 in O2.

4. Under appropriate in vitro conditions, brain slices remain functional for periods of up to 4 h. One important factor determining their viability is the composition of the media in which they are prepared, stored, and in which the recordings are obtained. A glucose concentrations of 10–30 mM should be used for such preparations, in particular when the experi- ments are done at physiological temperatures.

5. There are three main parts to the image acquisition setup for “in vivo” studies: (a) a cell incubator that fits onto the micro- scope stage and maintains a constant humid environment of 32 °C and 5 % CO2; (b) an inverted microscope equipped with an automated stage, automated fluorescence and brightfield shutters, and with automated filter wheels; and (c) a software package that integrates and controls the stage, shutters, and filter wheels. In our imaging setup, we use a customized cell incubator from LEICA, and the stage, shutters, and filter wheels are controlled by the LEICA LAS AF ver. 1.8.2. software.

6. The agent chosen for anesthesia should be appropriate to the activities to be studied. Volatile agents such as isoflurane may produce mild interference as these agents are rapidly washed out from brain slices.

7. The dissection should be performed rapidly, ideally within 1–3 min, and once recovered, the brain should be kept at 4 °C in a carbogenated solution to reduce cell damage and to improve its texture for slicing.

8. In order to “fish” out the brain slices, the glass pipette should be cut obliquely in its thin side and a rubber teat should be attached.

9. Theparametersusedinourprotocolare:(a)imagesize,512×512 pixels; (b), brightfield and fluorescence excitation and emission filters, 480/40 and 527/30 nm, respectively; (c) location of the stage positions if using multiple settings; and (d) frequency and length of the acquisition, every 10 or 15 min for 1 h.

10. Slices containing DMSO and pluronic F127 have properties comparable to those of physiological membranes [12]. Also,

slices may be incubated with different ion indicators or labeled with fluorescent dyes.

11. As reported elsewhere [6, 10, 11], incubation of mouse brain slices with morphine increased Newport Green fluorescence. The selectivity of morphine for the MOR was reflected by the attenuation of zinc mobilization in the presence of the opioid antagonist, naloxone.

12. In this procedure it is important that there is no increase in the release of intracellular Zn2+ in the control slices (vehicle treated). An increase of over 10–15 % might indicate cellular damage that compromises receptor functionality.

This research was supported by MSC “Plan de Drogas 2011-014” & Ministerio de Economía y Competividad (MINECO), SAF 2012- 34991.

Free Zinc-Imaging 241

Acknowledgements

References

1. Garzón J, Rodríguez-Muñoz M, Sánchez- Blázquez P (2008) Do pharmacological approaches that prevent opioid tolerance target different elements in the same regulatory machinery? Curr Drug Abuse Rev 1:222–238

2. Kissin I, Bright CA, Bradley EL Jr (2000) The effect of ketamine on opioid-induced acute tol- erance: can it explain reduction of opioid con- sumption with ketamine-opioid analgesic combinations? Anesth Analg 91:1483–1488

3. Garzón J, Rodríguez-Muñoz M, Sánchez- Blázquez P (2012) Direct association of Mu-opioid and NMDA glutamate receptors supports their cross-regulation: molecular impli- cations for opioid tolerance. Curr Drug Abuse Rev 5:199–226

4. Sánchez-Blázquez P, Rodríguez-Muñoz M, Garzón J (2010) Mu-opioid receptors tran- siently activate the Akt-nNOS pathway to pro- duce sustained potentiation of PKC-mediated NMDAR-CaMKII signaling. PLoS One 5:e11278

5. Kröncke KD (2001) Zinc finger proteins as molecular targets for nitric oxide-mediated gene regulation. Antioxid Redox Signal 3: 565–575

6. Sánchez-Blázquez P, Rodríguez-Muñoz M, Bailón C et al (2012) GPCRs promote the release of zinc ions mediated by nNOS/NO and the redox transducer RGSZ2 protein. Antioxid Redox Signal 17:1163–1177

7. Rodríguez-Muñoz M, Garzón J (2013) Nitric oxide and zinc-mediated protein assemblies involved in mu opioid receptor signaling. Mol Neurobiol 48:769–782. doi:10.1007/ s12035-013-8465-z

8. Haugland R (1996) Handbook of fluorescent probes and research chemicals. Molecular Probes Inc., San Diego

9. Wei G, Hough CJ, Li Y et al (2004) Characterization of extracellular accumulation of Zn2+ during ischemia and reperfusion of hippocampus slices in rat. Neuroscience 125: 867–877

10. Rodríguez-Muñoz M, de la Torre-Madrid E, Sánchez-Blázquez P et al (2011) NO-released zinc supports the simultaneous binding of Raf-1 and PKCγ cysteine-rich domains to HINT1 protein at the mu-opioid receptor. Antioxid Redox Signal 14:2413–2425

11. Sánchez-Blázquez P, Rodríguez-Muñoz M, Vicente-Sánchez A et al (2013) Cannabinoid receptors couple to NMDA receptors to reduce the production of NO and the mobili- zation of zinc induced by glutamate. Antioxid Redox Signal 19:1766–1782. doi:10.1089/ ars.2012.5100

12. Li Y, Hough CJ, Suh SW, Sarvey JM et al (2001) Rapid translocation of Zn(2+) from pre- synaptic terminals into postsynaptic hippo- campal neurons after physiological stimulation. J Neurophysiol 86:2597–2604

Opioid Receptors: Methods for Detection and Their Modes of Actions in the Eye

Shahid Husain Abstract

This study provides evidence for the presence of opioid-receptors in the retina, optic nerve, and optic nerve head astrocytes. These receptors were measured by more than one technique including Western blotting, immunohistochemistry, and functional assays such as scotopic electroretinogram (ERG) and Pattern ERG. I also have provided evidence in recently published work from my laboratory that opioid receptors, more specifically δ-opioid receptors, play crucial roles in retina neuroprotection against ischemic and glau- comatous injuries. This chapter provides detailed procedures to measure opioid receptor activation and their roles in retina neuroprotection using functional assays such as scotopic ERG and pattern ERG.

Key words Electroretinogram, Glaucoma, Neuroprotection, Pattern electroretinogram, Retina

1 Introduction

The diverse biological effects of opioids including cytoprotection, analgesia, neuroendocrine regulation, immunomodulation, and beha- vioral modifications are manifested through specific opioid receptors that are distributed throughout the body. The opioid receptors, upon the binding of opioid drugs (or endogenous opioid peptides), regu- late a multitude of intracellular signaling pathways. Endogenous opi- oid receptors are biologically relevant ligand in various parts of the human body and facilitate signaling pathways among the immune, endocrine, and nervous system. Numerous studies have shown a mul- titude of signaling pathways that are regulated by three classes of seven transmembrane opioid receptors through a variety of heterotrimeric G-protein-coupled receptors (GCPRs). Sustained activation of opioid receptors also induces desensitization/tolerance (defined as the neces- sity to increase the dose to achieve a specified magnitude of effect). Tolerance to repeated administration of opioids involves alterations of receptor dynamics and of the signal transduction systems linked to opioid receptors. The mechanisms by which tolerance to opioids develops are complex and ongoing research is continuing to define

Santi M. Spampinato (ed.), Opioid Receptors: Methods and Protocols, Methods in Molecular Biology, vol. 1230, DOI 10.1007/978-1-4939-1708-2_20, © Springer Science+Business Media New York 2015

Chapter 20

243

244 Shahid Husain

1.1 Classification of Opioid Receptors

the intricacies of this phenomenon. Under the influence of cognitive and noncognitive stimuli, activation of opioid receptors may occur by the release of endogenous peptides (e.g., endorphin, dynorphin, and enkephalins) [1]. This chapter provides an overview of opioid recep- tors-mediated beneficial effects in ocular diseases (e.g., glaucoma and ischemia/reperfusion-induced retinopathy) and it also provides func- tional methods to measure opioid receptors activity in the eye.

Opioid-binding sites were discovered in the early 1950s [2] and 1960s [3] in mammalian gut and subsequently described in brain tissues [4, 5]. In the succeeding three decades, extensive pharma- cological studies have delineated three classes of opioid receptor subtypes: δ-opioid (DOR), κ-opioid (KOR), and μ-opioid (MOR) receptors. Based upon additional pharmacological evidence DOR, KOR, and MOR have been further subclassified into δ1, and δ2; κ1, κ2, and κ3; and μ1, μ2, and μ3 opioid receptors. Identification and characterization of new ligands at the level of the main opioid receptor subtypes typically utilizes competitive binding assays, employing a high affinity radioligand to one receptor subtype that is displaced by a test ligand. However, functional assays, such as inhibition of electrically induced contraction of the guinea pig ileum or mouse vas deferens, can be used extensively to quantify the potency and efficacy of ligands for opioid receptors. Alternatively, one can characterize opioid receptor ligands by utilizing a signal transduction-based assay such as the aequorin luminescence-based calcium assay [6]. Recently we have employed several biochemical and electrophysiological techniques to iden- tify opioid receptors in ocular tissues, which are discussed in detail in this chapter.

In 1992, the δ-opioid receptor was first cloned from a NG108-15 glioma hybridoma cell line [7, 8]. To date, δ-opioid receptor genes have been cloned from rat [9] and human [10] tissues. Studies have found mRNA expression of δ-opioid receptors in the heart (most abundant), lungs, adrenal glands, stomach, small and large intes- tine, kidney, spleen, brain, and sex organs [11, 12]. The κ-opioid receptor has been cloned from rat [13], mouse [14], human [15], and guinea pig [16] tissues. The μ-opioid receptor genes have been cloned from nervous tissues of rat [1], mouse [17], human [18], pig [19], cow [20], and zebra fish [21]. Additionally, these recep- tors are widely expressed in peripheral tissues, such as the lungs, small intestine, large intestine, adrenal glands, kidney, and spleen [12]. Recently, we have shown the expression of opioid receptors in the retina, optic nerve, and optic nerve head astrocytes using biochemical and functional assays [22–24].

1.2 Opioid Receptors

1.3 Opioid Receptors Dynamics

Even though opioids, such as morphine, remain the analgesic of choice in many cases, a major limitation to their long-term use is the development of physiological tolerance. The development of opioid tolerance in humans varies depending on the route of administration and on the disease state for which the opioids are prescribed. For example, tolerance is usually not a problem with short-term (acute) postoperative epidural or intrathecal opioids, but rather presents itself following chronic parenteral, epidural, or intrathecal opioid usage. Physical dependence appears after dis- continuation of opioid uptake and results in withdrawal symp- toms, such as hyper-analgesia, gastrointestinal cramping, and joint and muscle aches. Despite considerable progress, the molecular and cellular mechanisms mediating the development of tolerance and withdrawal to morphine remain controversial. A substantial amount of work has been dedicated to identify the molecular mechanisms of tolerance and now it is believed that opioid recep- tor desensitization (defined as a decrease of receptor signaling after sustained agonist activation) is closely connected to the phenomenon of tolerance. Interference with the function of G-proteins has been shown to reduce the response to opioid agonists and to interfere with the development of tolerance and dependence [25, 26]. The available data also suggest that it is unlikely that receptor downregulation is solely responsible for the development of opioid tolerance and indicate that multiple mech- anisms (phosphorylation, uncoupling, internalization) could be involved in this process. A better understanding of opioid recep- tor dynamics and of interactions with other mediator systems will be necessary if opioid ligands are to become therapeutic entities in the treatment of ocular disorders.

Opioid receptors have been demonstrated to be present in tissues of the anterior segment as well as in various layers of the retina [11, 22]. Experimental and/or clinical evidence has suggested potential roles for endogenous opioids and their receptors in the regulation of iris function, accommodative power, aqueous humor dynamics, cor- neal wound healing, and retinal development. Various endogenous opioid peptides, as well as their receptors, have been identified in the eye of humans [27], rabbits [28], and rats [22, 24, 29]. Opioid- containing (enkephalinergic) neurons have been reported to be present in the iris-ciliary body, choroid membranes, cornea, lacrimal glands, and adnexa of various species [29, 30]. Although opioids have been demonstrated to alter a variety of ocular functions, the detailed mechanisms by which opioidergic ligands elicit their phar- macological actions in the eye have not been clearly defined [31]. Opioid receptor ligands are known to alter intraocular pressure (IOP) in humans [27, 28, 32]. Originally, morphine was thought

1.4 Opioid Receptors in the Eye

Opioids and Retinal Elettroretinogram 245

246 Shahid Husain

1.5 Retina Neuroprotection
by Opioid Receptors Activation

to raise intraocular pressure; subsequently, it was demonstrated to lower intraocular pressure. Interestingly, the intraocular pressure of opium and heroin addicts has been demonstrated to be lower than that of non-addicted patients [33]. Moreover, conjunctival instilla- tion of opioid antagonist, naloxone, increases IOP in morphine- addicted patients, and this phenomenon has been proposed as a possible test for detecting misuse of morphine [32, 33]. In the eye, opioid agonists have been demonstrated to lower IOP and enhance outflow facility in both human and animals [27, 34]. Additionally, relatively selective δ- and κ-opioid receptor agonists such as DPDPE, bremazocine, spiradoline, and dynorphin have been shown to reduce IOP via reducing aqueous flow rates in New Zealand rabbits [34–36]. Kappa-opioid agonists also have other neuroendocrine effects, such as the release of atrial natriuretic peptide (ANP), which has been associated with enhanced outflow facility in rabbits [29]. Moreover, activation of the endogenous opioidergic system by elec- troacupuncture which elevates levels of endomophin-1, beta-endor- phin, and enkephalins has been demonstrated to induce ocular hypotension in rabbits [37]. There have also been claims of improved visual acuity and reduced IOP have been reported in response to acupuncture in glaucomatous patients [38]. These clin- ical and basic research findings suggest that opioid receptors can have a modulatory role on aqueous humor dynamics and possibly influence the decline of visual field defects in glaucoma.

More recently we have shown the neuroprotective activity of opi- oid receptors in the eye [22–24, 39, 40]. We also have identified the expression and location of the three opioid receptor subtypes (δ, κ, and μ) in the adult rat retina, and demonstrated that opioid receptor activation is required for the development of ischemic preconditioning in the rat model [22]. Subsequently we have dem- onstrated that the opioids play key roles in the retina neuroprotec- tion against glaucomatous injury because: (1) RGC function was preserved by exogenous morphine treatment in chronic ocular hypertensive rat eyes, as determined by pattern ERG; (2) loss of RGCs was reduced in morphine-treated ocular hypertensive eyes, as determined by retrograde labeling of RGCs. Moreover, we have shown a neuroprotective role of δ-opioid receptors in retina against glaucomatous injury [22–24] via inhibition of tumor necrosis factor-α (TNF-α). Mechanistically, we have also shown that δ-opioid receptors activation by a selective agonist (SNC-121) reduces neu- roinflammation and caspase activation that subsequently provide retina neuroprotection against glaucomatous injury [23, 24]. Overall, our findings provided strong clues that δ-opioid receptors activation plays a key role in the retina neuroprotection against glaucomatous injury.

2.2 Animals and Anesthesia

Opioids and Retinal Elettroretinogram 247

2 Materials

2.1 Pattern Electroretinograms (Pattern ERG)
in the Rat

2.1.1 Recording and Electrodes

2.1.2 Recording and Equipments

Generally, thin conductive fibers and foils are being used without topical anesthesia. Usually, a reference electrode is placed near the nose (see Note 1). The forehead locations should be avoided as these locations may result in contamination of the Pattern ERG from cortical potentials or from responses of the other eye.

AC-coupled amplifier with a minimum input impedance of 10 MX is required. Signal averaging is necessary because of the small amplitude of the Pattern ERG. The analysis period (sweep time) for the standard, Pattern ERG should be 150 ms or greater; with a 4 rps stimulation rate, the full 250 ms between reversals is recom- mended. Computerized artifact rejection is essential and should be set at no higher than ±100 μV. The amplifier must return to base- line rapidly following signals containing artifacts to avoid inadver- tent storage of nonphysiological data. A minimum sampling rate of 1,000 Hz (1 ms per point) is recommended. Even with a comput- erized artifact rejection system, it is important that the input signal should be continuously monitored for baseline stability and for the absence of amplifier saturation.

1. Adult male or female Brown Norway rats (3–5 months of age; 150–200 g) are used in these studies. The rats are kept under a cycle of 12-h light and 12-h dark. Animal handling is performed in accordance with the ARVO Statement for the Use of Animals in Ophthalmic and Vision Research, and the study protocol should be approved by the local Animal Care and Use Committee.

2. Rats are dark adapted overnight, and the following day they are anesthetized with ketamine (75 mg/kg) and xylazine (8 mg/kg) and body temperature should be maintained at 37 °C with a heating pad for scotopic ERG.

3. Rat pupils should be dilated with a 10 μL drop of a solution containing phenylephrine HCl (2.5 %) and tropicamide (1 %) (Akorn, Inc., Buffalo Grove, IL) for scotopic ERG.

Pattern ERG has been used to assess the functional deficits of retinal ganglion cells (RGCs) function in response to retinal injury. Pattern ERG is obtained in response to contrast reversal of patterned visual stimuli (gratings, checkerboards), rather than uni- form flashes of light. Pattern ERG can reflect direct damage to the RGCs. Studies in humans, primates, and rodents have shown that Pattern ERG is an indicator of RGC function [23, 24, 41]. In glaucomatous conditions, Pattern ERG overall response is

3 Methods

3.1 Pattern ERG
to Detect Opioid Receptors in the Eye

248 Shahid Husain

3.1.1 Waveform Measurements

reduced [23, 24, 42]. Pattern ERG, therefore, has clinical value in both neurological and ophthalmological practice. More specifi- cally, it is a promising technique to measure early damage of RGCs in glaucoma. Following should be considered for an optimal mea- surement of Pattern ERG.

The waveform evoked by pattern reversal stimuli depends on the temporal frequency of the stimulus (transient versus steady state). The Pattern ERG waveform in normal subjects usually consists of a small initial negative component with a peak time of appro- ximately 35 ms (N35), followed at 45–60 ms by a much larger positive component (P50). This positive component is followed by a large negative component at 90–100 ms (N95). The amplitudes of the standard Pattern ERG components are normally measured between peaks and troughs. The trough of N35 to the peak of P50 will be the P50 amplitude.

The amplitude of PERG depends on the stimulus parameters such as spatial frequency, temporal frequency, contrast, and luminance. The mean of the width and the height of the stimulus field should be between 9 and 15 ± 3 cm. The mean luminance of the stimulus screen must be constant during checkerboard/bars reversals (i.e., no transient luminance change). A photopic luminance for the white areas of greater than 80 cd/m2 is required. The contrast between black and white squares or bars should be maximal (close to 99–100 %) for the standard Pattern ERG and not less than 80 %. The contrast and mean luminance should be included in reports.

1. Rats should be anesthetized with ketamine and xylazine and body temperature should be maintained at 37 °C with a heat- ing pad. Pattern ERG recordings (without dark adaptation) should be conducted in both eyes (sequentially) 3 days prior to any retinal injury as described earlier [23, 24]. The eyes of the rats should be wide open, with the undilated pupils pointing towards the center of the stimulus.

2. The Pattern ERG electrode should be placed on the corneal surface by means of a micromanipulator and positioned in such a way as to encircle the undilated pupil without limiting the field of view (see Note 2).

3. A visual stimulus generated by black and white alternating contrast reversing bars (mean luminance, 50 cd/m2; spatial frequency, 0.033 cycle/deg; contrast, 100 %; and temporal fre- quency, 1 Hz) should be aligned with the projection of the undilated pupil at an 11 cm distance using UTAS-2000 (LKC Technologies). Each Pattern ERG should be an average of 300 sweeps at an interval of 1 s.

4. For the Pattern ERG amplitudes, measurements should be made between a peak and adjacent trough of the waveform.

3.1.2 Stimulus Parameters

3.1.3 Measurement of Pattern ERG in Rats

3.2 Scotopic Electroretinograms to Detect Opioid Receptors
in the Rat Eye

To quantitate the overall health of the retina, scotopic electroreti- nograms (ERGs) should be performed as previously described [22]. To perform scotopic ERGs the following steps are required.

1. Rats should be dark-adapted overnight, and the following day they should be anesthetized with ketamine and xylazine as described earlier [22]. Pupils should be dilated with a 10 μL drop of a solution containing phenylephrine HCl (2.5 %) and tropicamide (1 %) (Akorn, Inc., Buffalo Grove, IL). A needle ground electrode should be placed subcutaneously in the back of the animal and a reference electrode on the tongue.

2. The settings of the ERG: average number of sweeping = 3; time between stimulus: 15 s; sweep before update: 0; artifact reject: 0; display raw waves: yes; store individual waves: no. The set- tings of the amplifier: display should be 500 μV; low cutoff 0.3 Hz; high cutoff 500 Hz; notch filter off. The settings of the time base: number of sample 512; sample rate 2,000 Hz; pre- stimulus base 20 Hz. The settings of the sunbrush flash: test type, single flash; stimulus source, LED; intensity, −40 to 0 dB. Keep the fixation light off and the infrared light on.

3. A stimulus intensity series of ERGs should be recorded in response to single-flash intensities using 40 dB attenuation (low-intensity flash) through no attenuation (high-intensity flash). The single-flash responses should be an average of 2–3 flashes with an inter-stimulus interval of 2 min to ensure that ERG amplitudes at a given intensity were identical between the first and the last flash. Each single-flash ERG response should be measured using a contact lens containing a gold ring electrode held in place by a drop of methylcellulose. ERGs should be recorded by means of a UTAS-2000 system (LKC Technologies, Gaithersburg, MD).

4. Corneal electrical responses to a single, 10 μs white-light flash should be delivered by a Ganzfeld stimulator. Amplitudes of ERG a-wave (a measure of photoreceptors activity) and b-waves (a measure of inner retina activity) should be mea- sured and compared to contralateral control eye responses and corresponding ipsilateral responses from other treatment groups.

1. The use of the contact lens electrodes should be avoided because they degrade image quality on the retina.

2. A small drop of saline was applied to keep the cornea and lens moist during each recording during pattern ERG.

Opioids and Retinal Elettroretinogram 249

4 Notes

250 Shahid Husain

Acknowledgements

References

1. Chen Y, Mestek A, Liu J et al (1993) Molecular cloning and functional expression of a mu- opioid receptor from rat brain. Mol Pharmacol 44:8–12

2. Beckett AH, Casy AF (1954) Synthetic analge- sics: stereochemical considerations. J Pharm Pharmacol 6:986–1001

3. Portoghese PS (1965) A new concept on the mode of interaction of narcotic analgesics with receptors. J Med Chem 8:609–616

4. Terenius L (1973) Stereospecific interaction between narcotic analgesics and a synaptic plasma membrane fraction of rat cerebral cor- tex. Acta Pharmacol Toxicol 32:317–320

5. Simon EJ, Hiller JM, Edelman I (1973) Stereospecific binding of the potent narcotic analgesic (3H) Etorphine to rat-brain homoge- nate. Proc Natl Acad Sci U S A 70:1947–1949

6. Fichna J, Gach K, Piestrzeniewicz M et al (2006) Functional characterization of opioid receptor ligands by aequorin luminescence- based calcium assay. J Pharmacol Exp Ther 317:1150–1154

7. Kieffer BL, Befort K, Gaveriaux-Ruff C et al (1992) The delta-opioid receptor: isolation of a cDNA by expression cloning and pharmaco- logical characterization. Proc Natl Acad Sci U S A 89:12048–12052

8. Evans CJ, Keith DE Jr, Morrison H et al (1992) Cloning of a delta opioid receptor by func- tional expression. Science 258:1952–1955

9. Fukuda K, Kato S, Mori K (1993) Primary structures and expression from cDNAs of rat opioid receptor delta- and mu-subtypes. FEBS Lett 327:311–314

10. Knapp RJ, Malatynska E, Fang L et al (1994) Identification of a human delta opioid recep- tor: cloning and expression. Life Sci 54: PL463–PL469

11. Zhu Y, Hsu MS, Pintar JE (1998) Develop- mental expression of the mu, kappa, and delta opioid receptor mRNAs in mouse. J Neurosci 18:2538–2549

12. Wittert G, Hope P, Pyle D (1996) Tissue dis- tribution of opioid receptor gene expression in the rat. Biochem Biophys Res Commun 218: 877–881

Supported in part by NIH/NEI grant EY019081 and an unre- stricted grant to MUSC-SEI from Research to Prevent Blindness, New York, NY.

13. Meng F, Xie GX, Thompson RC (1993) Cloning and pharmacological characterization of a rat kappa opioid receptor. Proc Natl Acad Sci U S A 90:9954–9958

14. Yasuda K, Raynor K, Kong H et al (1993) Cloning and functional comparison of kappa and delta opioid receptors from mouse brain. Proc Natl Acad Sci U S A 90:6736–6740

15. Simonin F, Gaveriaux-Ruff C, Befort K et al (1995) kappa-Opioid receptor in humans: cDNA and genomic cloning, chromosomal assignment, functional expression, pharmacol- ogy, and expression pattern in the central nervous system. Proc Natl Acad Sci U S A 92: 7006–7010

16. Xie GX, Meng F, Mansour A et al (1994) Primary structure and functional expression of a guinea pig kappa opioid (dynorphin) recep- tor. Proc Natl Acad Sci U S A 91:3779–3783

17. Min BH, Augustin LB, Felsheim RF et al (1994) Genomic structure analysis of promoter sequence of a mouse mu opioid receptor gene. Proc Natl Acad Sci U S A 91:9081–9085

18. Wang JB, Johnson PS, Persico AM et al (1994) Human mu opiate receptor. cDNA and genomic clones, pharmacologic characterization and chro- mosomal assignment. FEBS Lett 338:217–222

19. Pampusch MS, Osinski MA, Brown DR et al (1998) The porcine mu opioid receptor: molecular cloning and mRNA distribution in lymphoid tissues. J Neuroimmunol 90: 192–198

20. Onoprishvili I, Andria ML, Vilim FS et al (1999) The bovine mu-opioid receptor: clon- ing of cDNA and pharmacological character- ization of the receptor expressed in mammalian cells. Brain Res Mol Brain Res 73:129–137

21. Barrallo A, Gonzalez-Sarmiento R, Alvar F et al (2000) ZFOR2, a new opioid receptor- like gene from the teleost zebrafish (Danio rerio). Brain Res Mol Brain Res 84:1–6

22. Husain S, Potter DE, Crosson CE (2009) Opioid receptor-activation: retina protected from ischemic injury. Invest Ophthalmol Vis Sci 50:3853–3859

23. Husain S, Abdul Y, Crosson CE (2012) Preservation of retina ganglion cell function by

morphine in a chronic ocular-hypertensive rat model. Invest Ophthalmol Vis Sci 53: 4289–4298

24. Abdul Y, Akhter N, Husain S (2013) Delta- opioid agonist SNC-121 protects retinal gan- glion cell function in a chronic ocular hypertensive rat model. Invest Ophthalmol Vis Sci 54:1816–1828

25. Parolaro D, Patrini G, Giagnoni G et al (1990) Pertussis toxin inhibits morphine analgesia and prevents opiate dependence. Pharmacol Biochem Behav 35:137–141

26. Gomes BA, Shen J, Stafford K et al (2002) Mu-opioid receptor down-regulation and toler- ance are not equally dependent upon G-protein signaling. Pharmacol Biochem Behav 72: 273–278

27. Drago F, Panissidi G, Bellomio F et al (1985) Effects of opiates and opioids on intraocular pressure of rabbits and humans. Clin Exp Pharmacol Physiol 12:107–113

28. Drago F, Gorgone G, Spina F et al (1980) Opiate receptors in the rabbit iris. Naunyn Schmiedebergs Arch Pharmacol 315:1–4

29. Selbach JM, Buschnack SH, Steuhl KP et al (2005) Substance P and opioid peptidergic innervation of the anterior eye segment of the rat: an immunohistochemical study. J Anat 206:237–242

30. Stone RA, Kuwayama Y, Laties AM (1987) Regulatory peptides in the eye. Experientia 43:791–800

31. Murray RB, Adler MW, Korczyn AD (1983) The pupillary effects of opioids. Life Sci 33: 495–509

32. Drago F, Aguglia E, Dal Bello A et al (1985) Ocular instillation of naloxone increases intra-

ocular pressure in morphine-addicted patients: a possible test for detecting misuse of morphine. Experientia 41:266–267

33. Fanciullacci M, Boccuni M, Pietrini U et al (1980) The naloxone conjunctival test in morphine addiction. Eur J Pharmacol 61: 319–320

34. Russell KR, Wang DR, Potter DE (2000) Modulation of ocular hydrodynamics and iris function by bremazocine, a kappa opioid recep- tor agonist. Exp Eye Res 70:675–682

35. Wang D, Potter DE (1996) Ocular action of an opioid peptide, DPDPE. J Ocul Pharmacol Ther 12:131–139

36. Moore TT, Potter DE (2001) Kappa opioid agonist-induced changes in IOP: correlation with 3H-NE release and cAMP accumulation. Exp Eye Res 73:167–178

37. Chu TC, Potter DE (2002) Ocular hypoten- sion induced by electroacupuncture. J Ocul Pharmacol Ther 18:293–305

38. Kurusu M, Watanabe K, Nakazawa T et al (2005) Acupuncture for patients with glau- coma. Explore 1:372–376

39. Husain S, Abdul Y, Potter DE (2012) Non- analgesic effects of opioids: neuroprotection in the retina. Curr Pharm Des 18:6101–6108

40. Husain S, Potter DE (2008) The opioidergic system: potential roles and therapeutic indica- tions in the eye. J Ocul Pharmacol Ther 24: 117–140

41. Porciatti V (2007) The mouse pattern electro- retinogram. Doc Ophthalmol 115:145–153

42. Wanger P, Persson HE (1983) Pattern-reversal electroretinograms in unilateral glaucoma. Invest Ophthalmol Vis Sci 24:749–753

Opioids and Retinal Elettroretinogram 251

Chapter 21

Evaluation of Murine Macrophage Cytokine Production After In Vivo Morphine Treatment

Silvia Franchi, Mara Castelli, Sarah Moretti, Alberto Panerai, and Paola Sacerdote

Abstract

The discovery of opioid receptors expression on immune cells has originated a large research activity on the possible modulation by opioid drugs of immune system responses. In the present chapter we describe an easy methodology useful to obtain information about the potential immunomodulatory activity of opioid drugs. An in vivo treatment schedule is used, and macrophages are studied for their ability to release different cytokines.

Key words Cytokines, Lipopolysaccharide, Murine macrophages, Morphine, Opioid-induced immunosuppression

1 Introduction

Solid evidence accumulated in both animal and human studies in the last 15 years has demonstrated that morphine and other opioid drugs currently used for treating pain in patients can have an impact on several functions of the immune system [1–3]. The study of these pharmacological aspects of morphine can be of interest for the better understanding and identification of new reg- ulatory pathways in the immune system, but it can have an impor- tant clinical relevance, in consideration of the many patients treated with opioids for acute and chronic pain [2, 3].

As well known, the immune system is a complex well-regulated and orchestrated system, and the final functional outcome depends on interaction of several cells as well as soluble mediators and receptors. When studying the impact of a drug or a treatment on immune responses, an enormous number of parameters can be studied, using a large numbers of methodological approaches.

Our laboratory over the years has reached a comprehensive expertise in evaluating the effects of opioid drugs on cytokine

Santi M. Spampinato (ed.), Opioid Receptors: Methods and Protocols, Methods in Molecular Biology, vol. 1230, DOI 10.1007/978-1-4939-1708-2_21, © Springer Science+Business Media New York 2015

253

254 Silvia Franchi et al.

2 Materials

2.1 Animals

production by different cell types, such as lymphocytes and macrophages [4–6].

The study of the immune effect of opioids is further compli- cated by the fact that the immunomodulation is exerted both peripherally and centrally. Peripheral immune cells express func- tional μ opioid receptors (MOR) that can be directly activated by opioid ligands [7–9]. On the other hand, by binding to MOR in the brain, opioids activate descending pathways such as the hypothalamus-pituitary adrenal axis (HPA) and the sympathetic nervous system. Both descending pathways have been involved in opioid-induced immunomodulation [10, 11]. These findings limit the significance of studying the effects of opioids in vitro experi- ments only, since the central component is obviously missing. In our experience, and also scrolling the literature, it emerges that the most interesting results obtained in the fields of opioids and immunity have been obtained by treating animals in vivo, or, when possible, collecting cells from patients under opioid treatment for pain therapy. As the possibility to measure immune responses in vivo is often limited, the easiest and more reliable approach is to evaluate immune responses ex vivo, i.e., to treat experimental animals in vivo, collect immune cells, prepare in vitro primary cul- tures, and measure responses.

In the present chapter we describe the methodology that we use to obtain, purify, and stimulate peritoneal macrophages obtained from mice treated in vivo with morphine, in order to assess the release of the prototypical pro- and anti-inflammatory cytokines.

Balb C or C57BL/6J male mice, young adults, approximately 9 weeks old (see Note 1).

Animal care must be in compliance with the official guidelines on the use and protection of animals in experimental research. The guidelines may vary for the different countries. All efforts must be made to minimize animal suffering and to reduce the number of animals used. Animals are housed in cages (4–5/cage) under con- trolled laboratory conditions (12-h alternate light/dark cycles, room temperature 20–22 °C, humidity 55±10 %, food and water ad libitum) and are normally acclimatized to the environment for 5–7 days before experiments.

1. Morphine-HCl, 2, 5, 5, 10, 20 mg/kg or other opioid drugs under investigation. Obviously opioid drugs purchase is restricted and needs special permissions. Time for obtaining permission may be quite long.

2.2 Drugs
and Substances

2.3 Macrophage Cultures

2. Lipopolysaccharide (LPS) is phenol extracted from E. Coli serotype O111:B4. The product is soluble in water at a con- centration of 5 mg/ml or in cell culture medium at 1 μg/ml. The solution is stable for approximately 1 month at +2–8 °C; frozen aliquots (−20 °C) can be stored up to 2 years.

1. Thyoglicollate: suspend 3 % of Brewer thioglycollate medium (Fluka, Sigma-Aldrich) in distilled water (e.g., 9 g/300 ml H2O) into a pyrex bottle. Boil to completely dissolve the medium using a magnetic heating. Sterilize by autoclaving for 20 min at 121 °C. Once prepared, the solution can be stored at RT for several weeks. If, by a visual analysis, the medium appears green prior to use, reheat once to drive off absorbed oxygen (see Note 2).

2. Collection medium: cold RPMI 1640 (Sigma-Aldrich)+10 % of fetal bovine serum FBS.

3. 10 ml sterile syringe with 18G needles. 4. Conic 15 ml sterile tubes.
5. Sterile scissors.
6. Sterile forceps.

7. Centrifuge.
8. Burker chamber.
9. Trypan blue, Turk liquid.

10. 0.5 ml Eppendorf tubes.

11. 24 Well plates for cell culture.

12. Phosphate buffered saline (prepare stock solution 10×: 500 ml dH2O, 14.6 g Na2HPO4⋅2H2O, 1.04 g NaH2PO4⋅H2O, 43.85 g NaCl). Before using it for cell collection, prepare a final buffer 1× by diluting 10× in dH2O; pH must be at 7.4 and the final bottle must be autoclaved.

13. Culture media: RPMI 1640, 10% FBS, 2 % pen-strep, 1 % glutamine, 0.1 % 2ME.

14. Disposable sterile tips/pipettes. 15. Pasteur pipettes.
16. 96 Well plates.
17. Micropipettes (p20, p100, p1000). 18. Disposable sterile tips.

19. Multichannel/repetitive pipette. 20. Plate washer, if available.

Multiple commercial kits are available for the determination of murine cytokines in culture media. The assay should be prepared according to the protocol instructions for each ELISA kit.

2.4 ELISA
for Cytokine Measurements

Morphine Modulation of Macrophage Cytokines 255

256 Silvia Franchi et al.

3 Methods

3.1 Treatment

The reagents here described are representative of the R&D system kit for measuring IL-1beta (DY401, DuoSet ELISA Development System).

1. Coating buffer: PBS (137 mM NaCl, 2.7 mM KCl, 8.1 mM Na2HPO4, 1.5 mM KH2PO4) pH 7.2–7.4, 0.2 mm filtered.

2. Block buffer: 1 % BSA in PBS.

3. Reagent diluent: 0.1 % BSA, 0.05 % Tween 20 in Tris-buffered Saline (20 mM Trizma base, 150 mM NaCl) pH 7.2–7.4, 0.2 μm filtered.

4. Wash buffer: 0.05 % Tween 20 in PBS, pH 7.2–7.4.

5. Capture antibody: dilute to a working concentration of 4 μg/ ml in coating buffer.

6. Detection antibody: dilute to a working concentration of 600 ng/ml in reagent diluents.

7. Standard: range 15.625–1,000 pg/ml (serial 1:2 dilutions in reagent diluent).

8. Streptavidin-HRP: dilute 1:200 in reagent diluents.

9. Substrate solution: 1:1 mixture H2O2 and TMB.

10. Stop solution: 2 N H2SO4.

Macrophage elicitation is achieved by injecting 2 ml of 3 % of Brewer thioglycollate medium intraperitoneally (i.p.) 3 days before treating the animal with the opioid drugs (see Note 3).

Morphine, as opioid golden standard, or other opioid drugs are injected subcutaneously (s.c.) to mice. The s.c. route of adminis- tration is generally the most frequently used, together with the intraperitoneal one, for opioid drugs. However, since we need to withdraw macrophage from peritoneum, the i.p. injection is not indicated.

The immunomodulation induced by opioids is often dose related, and it is therefore important to perform a dose finding if a new opioid is investigated. We normally prepare a solution of mor- phine made in saline with the mg/kg chosen dissolved in 10 ml, and we inject a volume proportional to mice weight (100 ml/10 g mice).

According to our experience an optimal time point in order to assess the acute effects of opioids on cytokine ranges between 2 and 3 h after s.c. administration [12]. At this time point animals must be sacrificed as required by the local animal guideline (CO2 overdose).

We do not shave mice abdomens, but, after sacrifice, the fur is wet by quickly dipping the animals in ethanol/water solution.

Morphine Modulation of Macrophage Cytokines 257

3.2 Harvest of Peritoneal Macrophages

Fig. 1 Typical peritoneal macrophage layer obtained after adherence to plastic wells.1 × 106 macrophages are incubated for 2 h, and nonadherent cells washed out. Phase-contrast microphotograph (200× magnification)

The macrophage collection is performed under a laminar flood hood, using sterilized forceps and scissors. 10 ml of cold collection medium is gently injected and the abdomen is massaged without withdrawing the needle (see Note 4). The fluid is then collected and put in 15 ml conical tubes (see Note 5). After centrifugation (2,000×g for 7 min at 4 °C), peritoneal cells (PEC) are resus- pended in 4 ml of collection medium and counted, using Trypan blue for checking cell viability and Turk solution (dilute 1:10), in order to discriminate nuclei (see Note 6).

The cells are recognized on the basis of their morphology and macrophages counted.

PEC are diluted in collection medium at the final concentra- tion of 1×106 macrophages/ml and seeded (1×106/well) in 24 well plastic plate for 2 h, in order to separate and purify adherent macrophages.

Nonadherent cells are removed using vacuum and adherent cells are washed twice with PBS solution. A monolayer of macro- phage cells is now present on the plastic surface and is ready for LPS stimulation (see Note 7). Following this procedure the per- centage of macrophages in the adherent cells ranges between 85 and 95 % [4, 5, 13]. A typical macrophage layer obtained with this methodology is shown in Fig. 1.

Either culture medium alone or culture medium+LPS (1 μg/ ml) is added to the wells reaching a final volume of 1 ml/well.

We normally prepare samples in duplicate. The plates are incu- bated at 5 % CO2, 37 °C for 24 h (see Note 8). At the end of incubation, the supernatant is collected with a micropipette, using 1 ml tips and transferred into 1.5 ml Eppendorf tubes, paying attention not to touch the plate bottom (see Note 9). To avoid

258

Silvia Franchi et al.

3.3

IL-1beta ELISA

the presence of residue cells the sample tubes are centrifuged (10,000 × g for 7 min, RT) and the supernatant collected into new tubes that are then put in a sample storage box and immediately frozen at −80 °C for ELISA assay (see Note 10).

The ELISA protocol is generally well described in manufacturer’s instruction. However we briefly describe the procedure that we follow for IL-1beta evaluation (R&D system).

1. Coating: dilute the Capture antibody to the working concentration in coating buffer (100 μl/well); seal the plate and incubate overnight at RT.

2. Wash the plate with wash buffer 3 times (see Note 11).

3. Block plate by adding block buffer and incubate 1 h at RT.

4. Wash the plate with wash buffer 3 times.

5. Add 100 μl/well of sample/standard and incubate the plate 2 h at RT (see Note 12).

6. Wash the plate with wash buffer 3 times.

7. Add 100 μl/well of the Detection antibody and incubate the plate 2 h at RT.

8. Wash the plate with wash buffer 3 times.

9. Add 100 μl/well of the Streptavidin-HRP and incubate the plate 20 min at RT avoiding direct light.

10. Wash the plate with wash buffer 3 times.

11. Add 100 μl/well of Substrate Solution and incubate the plate
for 20 min at RT avoiding direct light (see Note 13).

12. Add 50 μl/well of Stop solution; gently tap the plate to ensure
thorough mixing.

13. Read immediately the plate at λ 450 nm using a microplate reader.
A typical dose response of the effect of morphine treatment

on IL-1beta production by peritoneal macrophages is reported in Fig. 2.

1. The mouse strain does not affect the sensibility of macrophages to morphine. Therefore the choice of one strain rather than another may depend upon specific experiments requirements: for example genetically modified mice are often on a C57 background.

2. The efficacy of thioglycollate elicitation is variable. The highest elicitation is reached when the solution is of brownish color. If the solution remains greenish it can be autoclaved again.

4 Notes

Morphine Modulation of Macrophage Cytokines 259

Fig. 2 Dose curve inhibition of in vivo acute morphine administration on the production of IL-1 by murine macrophages. Cells are obtained after 2 h of morphine administration. Cultures are stimulated in vitro for 24 h in presence of LPS. Mean ± SD of 6 animals-one way ANOVA. *p < 0.051 versus saline; **p < 0.01 versus saline; **p < 0.001 versus saline; °p < 0.05 versus 5 mg/kg

If a stock of thioglycollate is working well, it is convenient to finish it up before preparing a new one.

3. The time elapsing from thioglycollate injection and cell collec- tion is important. After 1–2 days from thioglycollate treatment the cells collected are mainly polymorphonuclear cells and only between the second and third days the PEC is enriched in macrophages.

4. When inserting the needle for peritoneal lavage, it is important to avoid puncturing the gut. Moreover, the right angle of entrance must be found (30–45°) in order not to be stacked against the gut walls. If it is possible maintain visible the groove of the needle to avoid its clogging.

5. The PEC should be transparent/yellowish, with no evident blood contamination. In case of clear presence of red blood cells, it is better to discard the sample or perform a further step for red blood cell lysis.

6. The amount of cells that can be recovered from a single mouse ranges between 6 and 15×106. In general we prefer to plate cells obtained from single animals. However if the amount obtained is not enough, the cells obtained from mice belong- ing to the same treatment groups can be pooled, without significantly affecting the results.

260 Silvia Franchi et al.

7. After discarding nonadherent cells, it is useful to check with a microscope whether macrophages are evenly attached to plas- tic before continuing the experiments.

8. In general a good stimulation with LPS is evident when culture medium became yellowish after 24 h culturing.

9. Since macrophages are generally strongly attached to plastic, it is not necessary to spin the plates before collecting the supernatants.

10. After collecting media for cytokine determination, a layer of the adherent cells can eventually be used also for other tasks, such as obtaining RNA for PCR assays.

11. Be sure that no bubbles remain in wells after washing. In case, the single bubbles have to be blown up using a needle tip. The washing step may be performed either manually or using an automated plate washer. Obviously the automated plate washer should be preferred when a large number of plates are pro- cessed. In an average research laboratory, manual washing is often convenient.

12. ELISA sample dilution. The right sample dilution in order to fall in the middle rectified part of the standard curve may be verified depending on the treatment and the kit used. By using an LPS concentration of 1 μg/ml, as reported above, a good sample dilution would be:

13. Even if not specifically reported in kit manufacturer’ instruc- tion, it may be useful to put the ELISA plate on an orbital shaker for better uniforming the color development, after the addition of both TMB and the stop solution.

Sample dilution

Standard curve range

IL-1beta

Undiluted

7.8–1,000 pg/ml

TNF-alpha

1: 10

7.8–1,000 pg/ml

IL-10

Undiluted

31.25–4,000 pg/ml

References

1. Sharp BM (2006) Multiple opioid receptors on immune cells modulate intracellular signaling. Brain Behav Immun 20:9–14

2. Sacerdote P, Franchi S, Moretti S et al (2012) Cytokine modulation is necessary for effica- cious treatment of experimental neuropathic pain. J Neuroimmune Pharmacol 8:202–211

3. Vallejo R, de Leon-Casasola O, Benyamin R (2004) Opioid therapy and immunosuppres- sion: a review. Am J Ther 11:354–365

4. Martucci C, Franchi S, Lattuada D et al (2007) Differential involvement of RelB in morphine- induced modulation of chemotaxis, NO, and cytokine production in murine macrophages and lymphocytes. J Leukoc Biol 81:344–354

5. Franchi S, Moretti S, Castelli M et al (2012) Sacerdote P Mu opioid receptor activation modulates Toll like receptor 4 in murine macrophage. Brain Behavior Immunity 26: 480–488

6. Sacerdote P, Martucci C, Vaccani A et al (2005) The non-psychoactive component of marijuana cannabidiol modulates chemotaxis and IL-10 and IL-12 production of murine macrophages both in vivo and in vitro. J Neuroimmunol 159:97–105

7. Gaveriaux-Ruff C, Matthes HW, Peluso J et al (1998) Abolition of morphine- immunosuppression in mice lacking the mu- opioid receptor gene. Proc Natl Acad Sci U S A 95:6326–6330

8. Bidlack JM, Khimich M, Parkhill AL et al (2006) Opioid receptors and signaling on cells from the immune system. J Neuroimmune Pharmacol 1:260–269

9. Roy S, Barke RA, Loh HH (1998) Mu-opioid receptor-knockout mice: role of mu-opioid receptor in morphine mediated immune func- tions. Brain Res Mol Brain Res 61:190–194

10. Wang J, Charboneau R, Balasubramanian S et al (2002) The immunosuppressive effects of chronic morphine treatment are partially depen- dent on corticosterone and mediated by the mu-opioid receptor. J Leukoc Biol 71:782–790

11. Fecho K, Maslonek KA, Dykstra LA et al (1996) Evidence for sympathetic and adrenal involve- ment in the immunomodulatory effects of acute morphine treatment in rats. J Pharmacol Exp Ther 277:633–645

12. Limiroli E, Gaspani L, Panerai AE et al (2002) Differential morphine tolerance development in the modulation of macrophage cytokine production in mice. J Leukoc Biol 72:43–48

13. Martucci C, Franchi S, Giannini E et al (2006) Bv8, the amphibian homologue of the mamma- lian prokineticins, induces a pro-inflammatory phenotype of mouse macrophages. Br J Phar- macol 147:225–234

Morphine Modulation of Macrophage Cytokines 261

Chapter 22

Measurement of Macrophage Toll-Like Receptor 4 Expression After Morphine Treatment

Mara Castelli, Alberto Panerai, Paola Sacerdote, and Silvia Franchi Abstract

The immune system is a complex and finely orchestrated system, and many soluble molecules and receptors contribute to its regulation.

Recent studies have suggested that many of the modulatory effects induced by morphine on innate immunity, and in particular the effects on macrophage activation and function, can be due to the modula- tion of an important macrophage surface receptor, the toll-like receptor (TLR), that is primarily involved in early regulatory steps. In the present chapter we describe a Reverse transcription (RT)-real time PCR method for assessing TLR expression in macrophage after in vivo morphine treatment.

Key words Macrophage, Morphine, Opioid receptor, RT-real time PCR, TLR4

1 Introduction

TLRs represent the early barrier against infectious diseases and are the first line of innate immune defense [1–3]. TLRs are expressed at high levels on Antigen Presenting Cells such as macrophages and dendritic cells [4, 5] and recognize conserved structures of microbes or ligands of exogenous and endogenous (host-derived) sources [6, 7]. These can include bacterial or viral nucleic acids, the unmethylated CpG islands of pathogen DNA or proteins unique to microbes such as flagellin15 [2]. Other TLR ligands include lipids and carbohydrates synthesized by bacteria such as LPS and lipotei- choic acid (LTA). In particular TLR4 recognizes LPS on gram- negative bacteria and with the assistance of LPS-binding protein (LBP) this recognition is enhanced. LBP carriers LPS to the CD14 molecule where it is then presented to the MD2-TLR4 complex [5–9].

The activation of the TLR4 signaling triggers an inflammatory and defensive response necessary for the correct activation of immune signaling, such as the induction of TNF and IL-1 [7, 8].

Santi M. Spampinato (ed.), Opioid Receptors: Methods and Protocols, Methods in Molecular Biology, vol. 1230, DOI 10.1007/978-1-4939-1708-2_22, © Springer Science+Business Media New York 2015

263

264 Mara Castelli et al.

2 Materials

2.1 Animals

Morphine and other opioids are able to reduce the LPS-induced cytokine production both in vitro and in vivo, and TLR4 has been demonstrated to be a target for the immunomodulatory effect of morphine [10–13].

In this chapter, that is linked to Chapter 21, we describe the procedure to evaluate TLR4 expression in peritoneal macrophages, obtained from morphine-treated animals.

The identification of the effect of opioid drugs on this recogni- tion receptor can be of great importance, since TLR4 is clearly involved in the first line of defense against microbes and viruses. When an opioid drug is able to decrease or modify the expression of TLRs, the capability of the patient to respond to infection might be altered. This aspect could be particularly relevant when consid- ering that opioid treatment is correctly the first choice for postop- erative pain, and the staying in the hospital is often a risk for getting infections [14].

Balb C or C57BL/6J male mice, young adults, approximately 9 weeks old (see Note 1).

Animal care must be in compliance with the official guidelines on the use and protection of animals in experimental research. The guidelines may vary for different countries. All efforts must be made to minimize animal suffering and to reduce the number of animals used. Animals are housed in cages (4–5/cage) under con- trolled laboratory conditions (12-h alternate light/dark cycles, room temperature 20–22 °C, humidity 55±10 %, food and water ad libitum) and are normally acclimatized to the environment for 5–7 days before experiments.

2. Lipopolysaccharide (LPS) is phenol extracted from E. Coli serotype O111:B4. The product is soluble in water at a concen- tration of 5 mg/ml or in cell culture medium at 1 μg/ml. The solution is stable for approximately 1 month at +2–8 °C; frozen aliquots (−20 °C) can be stored up to 2 years.

2.2 Drugs
and Substances

2.3 Setup
of Macrophage Cultures

2.4 RNA Expression Analysis

2.4.1 RNA Extraction

20 min at 121 °C. Once prepared, the solution can be stored at room temperature for several weeks. If, by a visual analysis, the medium appears green prior to use, reheat once to drive off absorbed oxygen (see Note 2).

2. Collection medium: cold RPMI 1640 (Sigma-Aldrich)+10 % of fetal bovine serum FBS.

3. 10 ml sterile syringe with 18G needles. 4. Conical 15 ml sterile tubes.
5. Sterile scissors.
6. Sterile forceps.

7. Centrifuge.
8. Burker chamber.
9. Trypan blue, Turk liquid.

10. 0.5 ml Eppendorf tubes.

11. 24 Well plates for cell culture.

13. Culture media: RPMI 1640, 10%FBS, 2 % pen-strep, 1 % glu- tamine, 0.1 % 2ME.

14. Disposable sterile tips/pipettes. 15. Pasteur pipettes.
16. 96 Well plates.
17. Micropipettes (p20, p100, p1000). 18. Disposable sterile tips.

19. Multichannel/repetitive pipette.

1. High-performance dispersing instrument (for instance Ultra- turrax®).

2. Ready to use reagent for the isolation of total RNA from cells or tissues, consisting of a mono-phasic solution of phenol and guanidine isothiocyanate (for instance Trizol® from Invitrogen or Trifast Eurogold® from Euroclone).

3. Chloroform.

4. Isopropyl alcohol.

5. RNase-free/molecular biology grade water: it can be either purchased or prepared by autoclaving distilled water at 121 °C for 1 h into RNase-free bottle (see Note 3).

Morphine Affects Macrophage TLR4 Expression 265

266

Mara Castelli et al.

2.4.2 Digestion

6. Ethanol 75 %: prepare diluting ethanol 96 % in RNA-free water: for example 35 ml of ethanol 96 %+9.8 ml of RNase- free water.

7. Refrigerated microfuge.

1. Commercially available DNA-free DNase treatment and removal reagent.

1. 55–60 °C water bath.

2. Biophotometer.

3. Plastic cuvettes for reading at wavelengths of 220–1,600 nm; volumes of 50 μl or more.

4. Commercially available reverse transcription supermix for RT-qPCR. We use iScriptTM from Bio-Rad.

5. Real Master Mix Probe ROX (Eppendorf).

6.Taqman probes for mouse TLR4 (Mm00445274_m1) and glyceraldehydes-3-phosphate dehydrogenase, GAPDH (Mm99999915_g1) are purchased from Life Technologies.

7. Thermal cycler.

8. Real-Time PCR system; we use ABI PRISM 7000 from Applied Biosystems.

9. Disposable RNase-free plastic (see Note 4).

1. Macrophage elicitation, morphine treatment, and harvest of peritoneal macrophages (PECs) are also described in Chapter 21.

2. The cells are recognized on the basis of their morphology and macrophages counted.

3. For PCR analysis it is better to dilute cells at a final concentra- tion of 4×106 cells in 2 ml.

4. PECs are seeded in six well plastic plates for 2 h, in order to separate and purify adherent macrophages. Nonadherent cells are removed using vacuum and adherent cells are washed twice with PBS solution. A monolayer of macrophage cells is now present on the plastic surface.

5. For TLR4 mRNA evaluation, macrophage monolayer can be either immediately collected or further stimulated with LPS, as described in Chapter 21 depending on the aim of the experiment.

6. After removing (with a vacuum pump) media from plate, mac- rophages are washed once with PBS and total RNA extracted

Genomic DNA

2.4.3 RNA Quantification

3 Methods

3.1 RNA Extraction

from purified macrophages using Trizol® or similar reagent. Trizol® (1 ml/well) is added directly on plates on adherent cells, and then recovered into 5 ml tubes and immediately fro- zen at −80 °C (see Note 5).

1. After thawing, Trizol® samples are homogenized on ice, using a high-performance dispersing instrument, collected with a 1 ml syringe and transferred into 1.5 ml Eppendorf tube (see Note 6). Start the RNA extraction procedure following Trizol® or similar reagent instruction (see Note 7).

2. Phase separation. Add 0.2 ml of chloroform to each Trizol® sample, shake tubes vigorously, and centrifuge (22,000×g for 15 min at 4 °C). Following centrifugation sample solution separates into 3 distinct layers; collect into fresh 1.5 ml Eppendorf tube the colorless upper aqueous phase containing RNA (see Note 8).

3. RNA precipitation. Precipitate the RNA adding to the aque- ous phase 0.5 ml of isopropyl alcohol, shake and leave tubes at RT for 10 min, then centrifuge (22,000×g for 15 min at 4 °C). RNA precipitate appears now as gel-like pellet.

4. RNA wash. Remove the supernatant, add 1 ml of 75 % ethanol to the pellet, and centrifuge (10,000 × g for 10 min at 4 °C).

5. Re-dissolving the RNA. Remove ethanol and dry the RNA pellet (air-dry); the final pellet obtained must be resuspended in 10/20 μl of RNase-free water and frozen at −80 °C for further processing.

1. Before continuing to genomic DNA digestion, it is necessary to heat the samples at 55/60 °C in a water bath for 10 min. This step can be performed either before or after freezing.

2. Transfer RNA sample into fresh 0.2 ml Eppendorf tubes.

3. Use commercially available DNA-freeTM DNase kit in order to remove any genomic DNA contamination from samples. If you follow the instructions, a good result is guaranteed. Briefly, add 1 μl DNase (2 U) and DNase buffer (10×) to RNA samples and incubate for 40 min at 37 °C using a thermal cycler. At the end of the incubation time, add 2 μl or 0.1 vol- ume of DNAse inactivation reagent, vortex, spin, and leave 3 min at RT: centrifuge (16,000 × g for 3 min). Remove super- natant and transfer into fresh 0.2 ml Eppendorf tubes.

4. After removing DNA contaminations, mRNA is quantified (see Note 9). Add 1 μl of RNA solution obtained after digestion into 1.5 ml Eppendorf tube containing 49 μl of RNase-free water; vortex, centrifuge (quick spin), and transfer volume into cuvettes for biophotometer reader. Read at 260 nm (see Note 10). The samples are now ready for reverse transcription reaction.

3.2 Genomic DNA Digestion and RNA Quantification

Morphine Affects Macrophage TLR4 Expression 267

268 Mara Castelli et al.

3.3 Reverse Transcription

1. Start from 1,000 ng of RNA. In 0.2 ml Eppendorf tubes add: a volume of RNA solution obtained after digestion corre- sponding to 1,000 ng of RNA+4 μl of reverse transcription supermix (5×)+RNase-free water to reach a final volume of 20 μl.

2. Vortex, spin, and incubate the samples in a thermal cycler as follows: 5 min at 25 °C (Priming), 30 min at 42 °C (Reverse transcription, RT), and the last 5 min at 85 °C (RT inactiva- tion). The cDNA samples obtained can be either frozen or immediately used for the amplification reaction.

1. Put a 96 well PCR plate on PCR cooler. In each well dispense: 11.75 ml of RNase-free water, 2 μl of cDNA (corresponding to 100 ng cDNA), 10 μl of Real Master Mix Probe ROX (2.5×), and 1.25 μl of Taqman probe (20×) to reach a final volume of 25 μl. This procedure must be performed both for the gene of interest (TLR4) and for the housekeeping control gene (GAPDH) (see Note 11). Normally all PCR assays are done in duplicate or triplicate.

2. The reaction mixture without the cDNA is used as control of TLR4 and GAPDH gene. Add 13.75 μl of RNase-free water, 10 μl of Real Master Mix Probe ROX, and 1.25 μl of specific Taqman probe (TLR4 and GAPDH) into appropriate wells.

3. Perform the amplification reaction using a Real-Time PCR sys- tem as follows: 95 °C for 2 min (Initial denaturation), followed by 40 cycles at 95 °C for 15 s (cycled template denaturation) and at 60 °C for 60 s (annealing and extension).

4. The results are quantified using the comparative threshold method. The Ct (threshold cycle) value of the specific gene of interest (TLR4) was normalized to the Ct value of the endog- enous control, GAPDH, and the comparative Ct method (2−ΔΔCt) was then applied using control group as calibrator.

1. Morphine (10 mg/kg) is administered subcutaneously to mice. Peritoneal cells are collected 2 h after morphine administration, and plated for macrophage purification. After macrophage purification (see text), Trizol® is added into plates and cells recovered.

2. RNA is extracted, quantified, and an equal amount of mRNA undergoes RT to obtain cDNA. cDNA is used as a template in Real-Time PCR for TLR4 mRNA expression analysis (Fig. 1). The results are quantified using the comparative threshold method (Table 1).

3.4 Real-Time PCR

3.5 Example of TLR4 mRNA Expression Analysis

Morphine Affects Macrophage TLR4 Expression 269

Fig. 1 Effect of in vivo morphine treatment on TLR4 mRNA expression in perito- neal macrophages. The data are reported as mean±SD of the 2−ΔΔCt values calculated for the 3 control samples and the 3 morphine samples as reported in Table 1. ***p < 0.001 versus control group; Student’s t-test

Table 1
Example of TLR4 mRNA expression analysis

Sample/treatment

Ct TLR4

Ct GAPDH

ΔCt

ΔΔCt

2−ΔΔCt

1 Saline

21.32

14.99

6.33

0.16

0.895

2 Saline

21.42

15.37

6.05

−0.12

1.087

3 Saline

21.36

15.22

6.14

−0.03

1.021

1 Morphine

23.34

15.9

7.44

1.27

0.415

2 Morphine

23.36

15.69

7.67

1.5

0.354

3 Morphine

22.5

15.02

7.48

1.31

0.403

4 Notes

ΔCt = CtTLR4 − CtGAPDH
ΔΔCt = ΔCt of all samples (both saline and morphine) − control group ΔCT mean (i.e., saline). In the example reported ΔCT mean is 6.17

2. The efficacy of thioglycollate elicitation is variable. The highest elicitation is reached when the solution is of brownish color. If the solution remains greenish it can be autoclaved again. If a stock of thioglycollate is working well, it is convenient to finish it up before preparing a new one.

270 Mara Castelli et al.

References

1. Akira S, Uematsu S, Takeuchi O (2006) Pathogen recognition and innate immunity. Cell 124:783–801

2. Kawai T, Akira S (2005) Pathogen recognition with Toll-like receptors. Curr Opin Immunol 17:338–344

3. Medzhitov R (2001) Toll-like receptors and innate immunity. Nat Rev Immunol 1: 135–145

4. McCoy CE, O’Neill LA (2008) The role of toll-like receptors in macrophages. Front Biosci 13:62–70

5. Laird MH, Rhee SH, Perkins DJ et al (2009) TLR4/MyD88/PI3K interactions regulate TLR4 signaling. J Leukoc Biol 85:966–977

6. Hughes AL, Piontkivska H (2008) Functional diversification of the toll-like receptor gene family. Immunogenetics 60:249–256

7. Lu YC, Yeh WC, Ohashi PS (2008) LPS/ TLR4 signal transduction pathway. Cytokine 42:145–151

8. Bowie A, O’Neill LA (2000) The interleukin-1 receptor/Toll-like receptor superfamily: signal generators for pro-inflammatory interleukins

3. All the RNase-free water and plastic ware may be either purchased or prepared in lab by autoclaving at 121 °C for 1 h.

4. In order to minimize the risk of RNase contaminations it is advisable to keep an appropriate set of dispensing pipettes for RNA handling only. It is also useful to wash gloved hands, laboratory benches, pipettes, and Ultraturrax® tips with a solu- tion of NaOH 0.1 N.

5. Although macrophages strongly adhere to plastic surface, the addition of Trizol® is sufficient to detach them from it. Once Trizol® is added it is enough to vigorously pipette before col- lecting the liquid.

6. It may be useful to pass the Trizol® homogenate through the syringe needle once, to be sure that cells are correctly lysed.

7. All extraction procedures MUST be performed in a chemical hood, since the reagents used are dangerous for health. Cap sample tubes securely.

8. To avoid RNA degradation, keep RNA samples always on ice, unless otherwise specified.

9. If the procedures are well conducted from 4×106 macro- phages, it can be expected to obtain an amount of RNA rang- ing from 400 to 600 μg/ml. However it is not uncommon to find lower values. In this case it is possible to pool RNA com- ing from similar cultures.

10. In order to be sure to have a good quality RNA sample, the ratio 260/280 nm must be comprised between 1.8 and 2.1. Abnormal 260/280 ratio usually indicates sample contamina- tion by protein or a reagent such as phenol or solvent.

11. Obviously the choice of the endogenous housekeeping gene is important, since it must not be modified by treatment. In our experience, we have clearly demonstrated that GAPDH can be used when assessing the immunomodulatory effect of morphine.

and microbial products. J Leukoc Biol 67:

508–514

9. Triantafilou M, Triantafilou K (2002) Lipopoly-
saccharide recognition: CD14, TLRs and the LPS-activation cluster. Trends Immunol 23: 301–304

10. Martucci C, Franchi S, Lattuada D et al (2007) Differential involvement of RelB in morphine- induced modulation of chemotaxis, NO, and cytokine production in murine macrophages and lymphocytes. J Leukoc Biol 81:344–354

11. Wang J, Barke RA, Charboneau R et al (2008) Morphine induces defects in early response of alveolar macrophages to Streptococcus

pneumoniae by modulating TLR9-NF-kappa

B signaling. J Immunol 180:3594–3600
12. Wang J, Ma J, Charboneau R et al (2011) Morphine inhibits murine dendritic cell IL-23 production by modulating Toll-like receptor 2 and Nod2 signaling. J Biol Chem 286:

10225–10232
13. Franchi S, Moretti S, Castelli M et al (2012)

Mu opioid receptor activation modulates toll like receptor 4 in murine macrophage. Brain Behav Immun 26:480–488

14. Risdahl JM, Khanna KV, Peterson PK et al (1998) Opiates and infection. J Neuroimmunol 83:4–18

Morphine Affects Macrophage TLR4 Expression 271

Chapter 23

The Role of Opioid Receptors in Migration and Wound Recovery In Vitro in Cultured Human Keratinocytes and Fibroblasts

Mei Bigliardi-Qi and Paul L. Bigliardi Abstract

We have previously described significant changes in skin differentiation and the delay in wound healing from delta-opioid receptor knockout mice. In addition, we have shown that opioid receptor ligands and their receptor systems affect wound healing in vitro and the migration pattern of human skin cells, such as keratinocytes and fibroblasts (Bigliardi-Qi et al., Differentiation 74:174–185, 2006; Bigliardi et al., Exp Dermatol 18:424–430, 2009; Bigliardi et al., J Recept Signal Transduct Res 22:191–199, 2002). This observation is true for both primary keratinocytes and fibroblasts derived from foreskin or normal human skin as well as for immortalized cell lines such as HaCaT cells.

Key words Delta-opioid receptor, Fibroblast, Keratinocyte, Wound scratch assay

1 Introduction

Wound scratch assays have been widely used to study cellular function in wound healing and regeneration. There are two ways of conducting the scratch assay: using a scratcher such as a pipette tip to scrape the cell monolayer, or using an insert to create a gap within the cell monolayer in order to simulate a wound gap (Fig. 1a, b). The scratch generated by a tip is less standardized, often having a gap of variable size and a risk of damage to the plastic chamber from the pipette tip, which could inhibit migration. This description uses culture inserts as they provide a more consistent and defined cell-free gap without damaging the plastic.

Santi M. Spampinato (ed.), Opioid Receptors: Methods and Protocols, Methods in Molecular Biology, vol. 1230, DOI 10.1007/978-1-4939-1708-2_23, © Springer Science+Business Media New York 2015

273

274 Mei Bigliardi-Qi and Paul L. Bigliardi

Fig. 1 (a) Cell culture inserts used to study wound healing and cell migration. Excerpt from http://www.ibidi.com. (b) Ibidi 8-well insert chamber slide. The total volume contained in each well is 400 μl. Excerpt from http://www.ibidi.com

2 Materials

1. Culture inserts (Ibidi, cat. no. 80209).

2. μ-Slide 8 well (Ibidi, 80826).

3. 10 mm culture dishes and T-75 culture flasks.

4. Dulbecco’s phosphate buffered saline (DPBS).

5. Dulbecco’s modified eagle medium (DMEM) plus 10 % FBS for fibroblasts.

6. Keratinocyte serum-free medium (KSFM) plus supplements for keratinocytes.

7. Gentamicin.

3 Methods

3.1 Isolation
and Culture of Primary Keratinocytes

8. Foetal bovine serum (FBS).

9. Dispase.

10. Trypsin-EDTA.

11. Phosphate-buffered saline (PBS).

12. Microscope equipped for live-cell imaging (i.e., stage incubator, CO2 control, camera, software).

1. Human foreskins obtained from elective circumcisions are placed in ice-cold complete KSFM containing 5 μg/ml gentamicin.

2. The samples can be stored at 2–8 °C for up to 5 days before use. 3. Wash the tissue in DPBS (without calcium and magnesium)

containing 20 μg/ml gentamicin for approximately 1 h.
4. Cut the tissue into 2–4 small pieces and transfer to a sterile

15 ml centrifuge tube.

5. Submerge the sections in dispase solution (25 caseinolytic units/ml, with 5 μg/ml gentamicin in DMEM) and incubate for 18 h at 2–8 °C.

6. Separate the epidermal layer of human keratinocytes from the dermis using forceps and place into a sterile 15 ml tube contain- ing 2 ml 0.25 % Trypsin-EDTA, and incubate at 37 °C for 30 min. Retain the dermis for isolation of fibroblasts (see below).

7. Centrifuge the cells at 180×g for 7 min at room temperature and resuspend the pellet in 5 ml complete KSFM medium.

8. Seed into T-75 flasks and change the medium every 2–3 days until the cells reach a confluency of 60–70 % when they are passaged.

1. Follow the procedure for primary keratinocyte isolation up to step 6 to obtain a section of dermis.

2. Cut the dermis into 2 × 2 mm pieces and place in a sterile 15 ml tube containing 2.5 ml trypsin-EDTA. Incubate at 37 °C for 40 min.

3. Centrifuge at 500 rpm for 10 min at room temp.

4. Aspirate the medium and wash once with DMEM before resuspending the pellet in 5 ml DMEM with 10 % FBS.

5. Seed cells in a 10 cm dish in 10 ml of medium and incubate at 37 °C, changing the medium on the following day.

6. Grow cells until they are 80 % confluent before passaging.

3.2 Isolation
and Culture of Primary Fibroblasts

Opioid Receptors in Migration and Wound Recovery 275

276

Mei Bigliardi-Qi and Paul L. Bigliardi

3.3

Cell Seeding

1. Pre-warm DMEM and KSFM medium to 37 °C in a water bath. Attach individual culture inserts to the Ibidi chamber slide (see Note 1).

2. Wash the cells once with 1× DPBS and trypsinize using 2 ml of TrypLETM for a T-75 flask. Leave the flask for approximately 5 min in an incubator at 37 °C until the cells are observed to have detached.

3. Resuspend the cells in medium (DMEM for fibroblasts and KSFM for keratinocytes) and centrifuge at 450 × g for 5 min at room temperature.

4. Aspirate the supernatant, resuspend cells in the appropriate fresh medium, and perform cell counting.

5. Seed 30,000-40,000 cells in 70 μl of the appropriate medium on each side of the insert (see Note 2).

6. Leave the cells overnight in an incubator at 37 °C to settle.

1. Carefully remove the inserts from the individual wells after overnight incubation.

2. Wash the cells twice with 1× PBS (at 37 °C) and add 400 μl of the appropriate fresh medium to each individual well. The medium may contain any required drug/ligand treatment.

A microscope equipped with live-cell imaging components (tem- perature and CO2 control system) is used to allow imaging of migrating cells over prolonged periods of time without any detri- mental effect on the cells (Fig. 2). Phase-contrast (and fluores- cence) images can be acquired every 15 min for a period of 24 h or more until the end of the experiment (see Note 3). The field of view for each experimental condition is determined manually using MetaMorph software, and up to six different fields can be sequen- tially recorded using an automated stage.

3.4 Sample Treatments

3.5 Live-Cell Imaging

Fig. 2 Wound healing assay. Wound gap migration of primary human fibroblasts imaged at 0 time (left panel ) and after 24 h (right panel )

3.6 Data Analysis

ImageJ software is used to analyze the acquired image files [4]. The area of wound recovery at a fixed time point and the percentage area of wound recovery over the total time course can be obtained via various macros within the software. In summary, the scratch wound macro automates the task of finding the edge of the wound, thresholding, binary operations, identifying the wound area, and calculating the wound areas at each time point (Fig. 2). After the wound areas are determined, the macro will then calculate the recovery ratio, plot the results, and save the measurements.

1. To help create a well-defined gap within the cells, it is essential to check that the inserts are well stuck—especially the central part of the inserts.

2. Seeding the cells at the right density is very important. This varies between cell types and needs to be optimized for each cell line.

3. The duration of the imaging will be determined based on the cells used and the experimental conditions; it is generally desir- able to observe complete closure of the wound gap.

This work was supported by Swiss National Fund 31003A-116811 and Agency for Science, Technology & Research (A*STAR) Singapore.

4 Notes

Opioid Receptors in Migration and Wound Recovery 277

Acknowledgements

References

1. Bigliardi-Qi M, Gavériaux-Ruff C, Zhou H et al (2006) Deletion of delta-opioid receptor in mice alters skin differentiation and delays wound heal- ing. Differentiation 74:174–185

2. Bigliardi PL, Tobin DJ, Gavériaux-Ruff C et al (2009) Opioids and the skin – where do we stand? Exp Dermatol 18:424–430

3. Bigliardi PL, Büchner S, Rufli T et al (2002) Specific stimulation of migration of human kera- tinocytes by m-opiate receptor agonists. J Recept Signal Transduct Res 22:191–199

4. Rasband WS (1997–2014) ImageJ. U.S. National Institutes of Health, Bethesda, MD, USA, http://imagej.nih.gov/ij/

Part V Behavioral Effects Mediated by Opioid Receptors

Chapter 24

Role of Opioid Receptors in the Reinstatement of Opioid-Seeking Behavior: An Overview

Liana Fattore, Paola Fadda, Silvia Antinori, and Walter Fratta

Abstract

Opioid abuse in humans is characterized by discontinuous periods of drug use and abstinence. With time, the probability of falling into renewed drug consumption becomes particularly high and constitutes a considerable problem in the management of heroin addicts. The major problem in the treatment of opioid dependence still remains the occurrence of relapse, to which stressful life events, renewed use of heroin, and exposure to drug-associated environmental cues are all positively correlated. To study the neurobiol- ogy of relapse, many research groups currently use the reinstatement animal model, which greatly contrib- uted to disentangle the mechanisms underlying relapse to drug-seeking in laboratory animals. The use of this model is becoming increasingly popular worldwide, and new versions have been recently developed to better appreciate the differential contribution of each opioid receptor subtype to the relapse phenomenon. In this chapter we review the state of the art of our knowledge on the specific role of the opioid receptors as unrevealed by the reinstatement animal model of opioid-seeking behavior.

Key words Addiction, Opioid abuse, Opioid receptors, Opioid-seeking behavior, Relapse

1 Opioid Abuse: From Abuse to Addiction to Withdrawal and Relapse

Addiction is defined as obsessive thinking and compulsive need for something, like drugs, food, or sex, despite the resulting negative consequences. Drug addiction is a chronically relapsing disorder characterized by (a) compulsion to seek and take the drug, (b) loss of control in limiting intake, (c) emergence of a withdrawal syn- drome and related negative emotional states (e.g., dysphoria, anxi- ety, irritability) when access to the drug is prevented, and (d) development of tolerance. Clinically, the occasional but limited use of a drug is different from escalated drug use and the emergence of chronic drug dependence. Yet, for some individuals drug addiction develops over time and usually begins with misuse, moving toward abuse and resulting in addiction (Fig. 1).

Santi M. Spampinato (ed.), Opioid Receptors: Methods and Protocols, Methods in Molecular Biology, vol. 1230, DOI 10.1007/978-1-4939-1708-2_24, © Springer Science+Business Media New York 2015

281

282

Liana Fattore et al.

Fig. 1 The addiction cycle. Different phases of the addiction cycle, from the initial occasional, recreational use of the drug, toward chronic consumption and loss of control over use, until the occurrence of withdrawal upon cessation and relapse to drug-use after periods, even prolonged, of drug abstinence

Opioids are powerful painkillers that cause sedation and euphoria and are commonly abused. Persistent use of opioids induces addiction. According to the 2013 World Drug Report [1], the use of opioids (heroin, opium, and prescription opioids) has increased in Asia since 2009, although a prevalence of opioid use higher than the global average was found in North America (3.9 %), Oceania (3 %), and Europe (1.2 %). Once addicted, many heroin users feel completely powerless and continue to use the drug despite potentially dangerous or life-threatening conse- quences. Even if it proves possible for heroin addicts to refrain from drug use, maintaining abstinence is difficult. “Opioid-seeking behavior” is a common expression that refers to those behavioral patterns aimed to search and acquire opioids when these are not readily available. At the clinical level, craving and relapse are now considered major challenges in opioid addiction treatment, and to prevent relapse when an abstinent patient is exposed to the drug or a drug-related stimulus is still demanding.

Modeling Drug-Seeking and Relapse in Animal Models

Animal studies have been crucial in understanding the biology and pathophysiology of drug addiction, and much of the progress in our comprehension of the mechanisms regulating opioid-seeking

Opioid Receptors in Relapse 283

Fig. 2 Most common animal models or test used to study the different aspects of the addiction cycle. Behavioral animal models commonly used in preclinical research to study in detail every phase of the addiction cycle

behavior has derived from studies employing animal models of extinction/reinstatement. In contrast to clinical studies, variables can be controlled more easily in animal studies. Animal studies have demonstrated that the rewarding effect is not dependent on preex- isting conditions; that is, exposure to opioid agonists (generally heroin or morphine) is sufficient to motivate drug-taking behavior. While no animal model of addiction fully imitates the human condi- tion, animal models do permit investigation of specific elements of the process of drug addiction, such as positive and negative rein- forcement. Many models have been developed and validated so far in both rodents and primates, each of them specifically designed to study a definite stage of the addiction cycle (Fig. 2).

Although according to negative reinforcement theories with- drawal seems a logical trigger of relapse, such a link has not been confirmed yet [2]. According to this, clinical concerns are not limited to the detoxification and initial abstinence periods but rather extend to the occurrence of relapse after a long pause in drug consumption. Therefore, it has been necessary to develop new methods to model drug-seeking and relapse that appear with- out a withdrawal syndrome or long after it has passed. At present, the most frequently used animal models of drug-seeking reinstate- ment stem from the conditioned place preference (CPP) and self-administration (SA) procedures.

284 Liana Fattore et al.

Fig. 3 Conditioned place preference (CPP) apparatus. An example of CPP box, which is divided into two compartments differing each other for visual (walls) and tactile (floors) cues

The CPP paradigm is a preclinical behavioral model extensively used to study the rewarding/aversive effect of drugs and to assess incentive motivation and drug-motivated behavior in animals. In this paradigm, the primary motivational properties of a drug, e.g., mor- phine, serve as unconditioned stimulus that is repeatedly paired with a previously neutral set of environmental stimuli which acquire, throughout conditioning, secondary motivational properties such that they can act as conditioned stimuli (and, hence, elicit approach) when the animal is subsequently re-exposed to these stimuli. Briefly, the CPP apparatus consists in a box divided into two (or more) com- partments provided with different visual cues (e.g., present in the walls, which can be either brown, black, or white striped) and tactile cues (e.g., present in the floor, being either grid or chequered) so that the animal can distinguish between them (Fig. 3).

Due to the contiguous association between the context and the drug stimulus (unconditioned stimulus), which are alternated over days to associations of a different context to the absence of the drug stimulus (vehicle), CPP measures a learning process where the animal acquires a preference for a specific environmental con- text (conditioned stimulus). As shown in Fig. 4, in a first phase, often referred to as “preconditioning,” animals are allowed to freely explore both compartments, and the time spent in each compartment is measured to evaluate the animal’s innate prefer- ence for each side. Then, the two compartments are separated (usually by a guillotine door) and animal confined in an alternate manner in one compartment following repeated drug or vehicle injections (conditioning). At the end of the conditioning phase, on the test day the animals do not receive any injection and are allowed to freely explore again both compartments under a drug-free state.

Opioid Receptors in Relapse 285

Fig. 4 Cpp experimental procedure. An example of the different phases (I–IV) and corresponding test (pretest for Phase I) of morphine CPP. Length of each phase (days) is indicated below each phase. For a detailed and critical description of CPP protocols, Chapter Tzschentke [3]

The time spent in the drug-associated environment is considered a critical measurement of the rewarding effects induced by the drug. Usually, CPP is extinguished by placing the animal in the drug- associated compartment without administering the drug to the animal, i.e., pairing both compartments with vehicle. Once CPP has been extinguished, the reinstatement test can be conducted by priming the animal with different types of stimulus [3]. CPP is reinstated when animals show a preference for the compartment that during conditioning was associated to the drug.

Another traditional model of opioid-seeking reinstatement entails training an animal to self-administer heroin, i.e., to engage in a particular behavior, such as lever pressing or nose poking, in order to receive a small amount of the drug, in a short daily session, typically 1 or 2 h. Self-administration training occurs in operant chambers which are equipped with two levers or holes, one paired with reward delivery (the active lever/hole) and one resulting in no outcome (Fig. 5). A stimulus light (conditioned stimulus) is located above the active lever/hole and is turned on for 5–10 s when the animal makes an active response.

Opioids are usually self-administered intravenously, as humans do, and although rodents are most often used in these studies, this model has been used with a variety of species includ- ing nonhuman primates, dogs, cats, and pigeons. In rodents, a low-ratio requirement typically is used, such as a continuous fixed-ratio 1 (FR-1), where each operant response produces a

286 Liana Fattore et al.

Fig. 5 Operant box for intravenous self-administration in rats. A representative rat within a self-administration box with its intravenous catheter connected through a swivel system to a syringe pump that delivers a drug infusion contin- gently to the appropriate response (i.e., lever pressing)

drug delivery, although higher fixed-ratio protocols (e.g., FR-5) are also commonly used. The ability of opioids to sustain self- administration behavior under fixed-ratio schedules of reinforce- ment is now widely documented [4].

Briefly, during the initial phase of self-administration training, often referred to as “acquisition,” animals learn how to operate within the box in order to obtain the reward. Once reliable drug self-administration is acquired, animals typically stabilize respond- ing for the drug to a certain level, and make a rather constant number of responses day after day in order to self-administer over time similar amounts of the drug. This phase is called “mainte- nance,” and is characterized by stable drug-taking behavior. Then, drug-reinforced instrumental responding is extinguished by replac- ing the drug with a neutral stimulus, such as saline solution, i.e., by ceasing to reinforce responses by drug delivery and allowing the animal to perform the operant response without programmed con- sequences. During extinction training, behavioral responses decrease (i.e., extinguish) as the animal learns that drug reward is no longer achieved by pressing the lever or nose poking.

Finally, the re-emergence of this behavior in response to a spe- cific trigger is conducted under a drug-free condition. This allows for studying the recovery of the extinguished behavior without the interference produced by the psychoactive effects of the drug (e.g., drugs, like opioids, that typically decrease locomotor behavior

Fig. 6 Phases of drug-seeking reinstatement procedures. An example of the different phases of the extinction/ reinstatement model. Length of each phase may vary according to the specific protocol adopted. For a detailed and critical description of the opioid-seeking protocol in rat see the dedicated chapter in this book

could produce spurious decreases in lever pressing), and the increase in the number of operant responses (compared with that observed during extinction) is assumed as an enhancement of the subject’s drug-seeking behavior. Figure 6 illustrates the main phases of the extinction/reinstatement animal model.

The reinstatement test is conducted only once animal has com- pletely extinguished operant responses (e.g., lever pressing activity). Three different types of stimuli have been shown to trigger relapse to heroin in human addicts and to reinstate heroin-seeking behavior in rats [5]. A stimulus is said to reinstate responding if it causes an increase in responding that was reinforced formerly by the drug. Reinstatement refers to the return to drug-seeking behavior (e.g., lever pressing) even though the behavior is not rewarded with the drug. Thus, reinstatement serves as index of the animal’s drug- seeking (not drug-taking) behavior and presumably reflects the animal’s desire and drive to receive the drug. The degree of recovery of responding at the previously drug-paired lever is operationally defined as a measure of craving or relapse. This model has been pharmacologically validated with drugs that reduce opioid craving and relapse in opioid-dependent patients. A detailed description of the extinction/reinstatement animal model is provided in a separate chapter of this book (see Chapter 25).

3 Role of Opioid Receptors in Regulating Opioid-Seeking Behavior

Opioids bind to three main types of opioid receptors: mu-opioid receptor (MOR), kappa-opioid receptor (KOR), and delta-opioid receptor (DOR). MOR agonists and antagonists sustain and reduce, respectively, opioid self-administration [6, 7], and constitutive dele- tion of the MOR attenuates opioid-induced CPP [8]. Selective MOR blockade is sufficient to induce conditioned aversion in morphine-dependent animals, presumably because of unopposed activation of KOR. Based on these and many other evidence accu- mulated so far, it is now well established that MOR and DOR are

Opioid Receptors in Relapse 287

288 Liana Fattore et al.

3.1 Mu-Opioid Receptors (MORs)

implicated in reward for heroin and morphine, whereas KOR is implicated in aversion [9]. MORs are fundamental regulators of the reinforcing positive effects induced by opioids and other drugs of abuse but also by natural rewards such as sex or food [10]. In more general terms, opioid receptors not only regulate drug reward but also largely contribute to emotional and cognitive processes whose dysfunction favors the development of addictive behaviors [11]. It is therefore not surprising that they are also key modulators in reinstating opioid-seeking behavior.

More than 30 years ago, Stewart and colleagues showed that priming injections with heroin effectively reinstated heroin-seeking behavior in rats [12, 13]. Few years ago we found that non-contingent non- reinforced priming injections of heroin produces a reliable reinstate- ment of drug-seeking even following long-term (3-weeks) extinction of heroin-seeking behavior [14–16]. The self-administration data have been successfully confirmed and extended using the CPP ver- sion of the reinstatement model. In fact morphine CPP is promptly reinstated by a morphine priming injection in rats and mice [17, 18]. It was also demonstrated that morphine CPP may persist for weeks without re-exposing animals to the drug compartment showing that the elapsing of time alone is not sufficient to cancel an established CPP. On the contrary, planned daily extinction sessions abolish mor- phine CPP while a priming injection of morphine is able to renew the salience of the environment previously paired with the drug resulting in the reinstatement of the CPP [19, 20].

When administered as drug priming, MOR agonists mimic the effect of heroin on heroin-seeking reinstatement when given sys- temically [12, 21] or infused into the ventral tegmental area (VTA) [22], an effect attenuated by prior administration of naltrexone [22, 23]. Chronic MOR activation by heroin given via Alzet osmotic minipumps also attenuates heroin-induced reinstatement [24]. Importantly, priming infusions of pharmacologically related drugs also reinstate responding for heroin and morphine CPP in rodents [23–25].

The cue-induced reinstatement model is based on the assump- tion that opioid neurotransmission, in particular via MORs, strongly impacts on the incentive salience attribution to condi- tioned stimuli that predict reward. Thus, a list of MOR agonists/ antagonists has been tested for their ability to influence the rein- statement of heroin-seeking induced by drug-associated cues. Similarly to heroin primings, drug-associated cues are able to rein- state responding for heroin even after prolonged withdrawal in rats [26]. Moreover, heroin-associated conditioned stimuli could induce robust heroin-seeking behavior that is associated with increased c-Fos immunoreactivity in the medial part of the lateral habenula [27], which suggests the involvement of this brain region in mediating cue-induced heroin-seeking behavior after drug

Opioid Receptors in Relapse 289

abstinence. As expected, classical MOR antagonists, such as naltrexone and naloxone, have been shown to reduce cue-elicited reinstatement after extinction of heroin-seeking [23]. More recently, the cue-induced reinstatement model of opioid-seeking was used to test knock-in mice expressing a mutant recycling mu- opioid receptor (R-MOR) that desensitizes, internalizes, and recycles in response to morphine [28]. Authors first showed that R-MOR mice are less likely than their wild-type (WT) littermates to become addicted to morphine, although they are initially more sensitive to its rewarding effects. More interestingly, facilitating morphine-induced MOR trafficking reduces the development of addiction-like behaviors despite the fact that morphine was more acutely rewarding. That is, when measuring parameters like moti- vation to obtain morphine, persistent drug-seeking behavior in the face of adverse consequences, or reduced preference for alter- native rewards, WT mice showed a progressive increase over time whereas R-MOR mice showed no significant change [28]. Intriguingly, following a period of abstinence, WT mice also demonstrated significantly greater reinstatement of morphine- seeking than R-MOR mice. Taken together with previous find- ings of the same group [29], these data demonstrate that enhancing agonist-induced MOR trafficking reduces the devel- opment of tolerance, dependence, and addiction while preserving analgesia, and suggests a strategy for identifying novel opioid drugs with increased utility for treating chronic pain but devoid of the addictive effects of opioids.

In the last few years, the group of Barry Everitt investigated the effects of GSK1521498, a novel selective μ-opioid receptor antagonist in clinical development for behavioral and drug addic- tive disorders [30, 31] on heroin-seeking behavior measured under a second-order schedule of reinforcement, in which prolonged periods of drug-seeking behavior are maintained by contingent presentation of a drug-associated conditioned reinforcer [32]. They showed that reducing opioid receptor signaling at the μ-opioid receptor by GSK1521498, but not naltrexone, may have therapeutic potential by decreasing drug stimulus-maintained heroin-seeking with the additional effect of promoting abstinence. In addition, GSK1521498 decreased heroin-seeking, but not cocaine-seeking behavior, after the first drug infusion had been self-administered, suggesting that the effect of MOR antagonists on the conditioned control over heroin-seeking may be more effective therapeutically than naltrexone, especially in the promo- tion of abstinence and in the prevention of heroin relapse in addicted individuals seeking treatment.

Currently, most studies that investigate opioid-seeking rein- statement in animals employ extinction/reinstatement animal mod- els, while only few studies have been conducted so far on cue-induced reinstatement of extinguished CPP. Interestingly, they reported

290 Liana Fattore et al.

3.2 Kappa-Opioid Receptors (KORs)

that extinction of a drug-cue association is context-dependent, since an extinguished heroin CPP can be reinstated by a contextual tactual stimulus that is part of the conditioning context, phenome- non that requires an intact basolateral amygdala [33]. Furthermore, morphine withdrawal associated cues significantly elicited the rein- statement of morphine CPP and increased corticosterone levels, effects attenuated by pretreatment with an antagonist of the corti- cotrophin-releasing factor [34].

As concerns stress-induced reinstatement of opioid-seeking behavior, one study demonstrated that naltrexone is able to attenu- ate foot-shock-induced reinstatement of heroin-seeking in rats [23].

Unlike MOR agonists, KOR agonists were reported to produce dysphoric and psychotomimetic effects, and to function as negative reinforcers, producing conditioned aversive effects [35]. KOR stimulation has anti-addictive effects, as it generally antagonizes the acute rewarding effects of drugs, whereas KOR blockade has no consistent effect. According to this, initial hypotheses regarding the role of opioid receptors in drug addiction suggested that opposing endogenous opioid systems regulate emotional and per- ceptual experience [35]. However, more recent studies suggest that KOR antagonists may reverse motivational aspects of depen- dence and that, in a drug dependent-like state, KOR blockade is effective in reducing increased drug intake. Evidence has been pro- vided that manipulations that decrease the activity of KORs may be effective in the treatment of depression and drug addiction. Remarkably, in animal models of reinstatement, KOR stimulation can induce reinstatement via a stress-like mechanism [36–38], while blocking KOR with the selective antagonist nor-BNI pre- vents stress-induced reinstatement of heroin-seeking in trained rats [39]. Thus, enhanced tonic KOR function has been suggested to represent one mechanism underlying the pro-addictive effect of stress in promoting relapse [40]. According to this, it is now rec- ognized that KORs are key players in stress-induced addictive-like behaviors, and that blocking KORs represents a promising strategy to prevent relapse and alleviate negative affect in addiction [11].

Notably, a non-rewarding and non-aversive buprenorphine/ naltrexone doses combination, acting as a functional MOR/KOR antagonist, has been proposed to be protective against relapse, since it attenuates drug-primed reinstatement of cocaine and mor- phine CPP [41].

DOR antagonists prevent or attenuate the development of toler- ance and dependence to morphine [42], and tolerance to mor- phine does not develop in mice lacking the DOR [43]. While MORs seem to play a modest role in context-induced renewal, DORs are crucial for cue-induced reinstatement of ethanol- seeking, as the DOR antagonist, naltrindole, reduces context- and

3.3 Delta-Opioid (DORs) Receptors

4 Conclusions

cue-induced reinstatement of ethanol-seeking behavior in rodents more effectively than the MOR selective antagonists, naloxonazine or CTOP [44, 45].

In line with this, DORs have been shown to play a dominant role also in the reinstatement of cocaine-seeking behavior in the CPP test [46]. As concerns opioids, Le Merrer and collaborators have found that in contrast to spatial cues, presentation of circa- dian, drug, or auditory cues predicting drug association reinstates morphine CPP in mice lacking DORs. Delta-opioid receptors, therefore, appear to play a crucial role in modulating spatial contextual cue-related responses [47]. Moreover, bivalent ligands consisting of a MOR agonist coupled to a DOR antagonist by variable-length spacers fail to reinstate morphine-induced CPP, an effect that authors ascribed to the dissimilar internal cues gener- ated by the bivalent ligands compared to those generated by mor- phine or heroin [48]. DOR knockout animals exhibit an augmentation of context-dependent sensitization and a significant reduction in tolerance to both the locomotor-activating and con- ditioned rewarding effects of morphine [49]. Although other stud- ies suggested that DORs are not essential for morphine-induced reward, their involvement in the facilitation of the association of morphine effects with the context has been confirmed [47, 50], with obvious implication for context-induced reinstatement of opioid-seeking.

Relapse prevention has been a major target in the development of pharmacological treatments for opioid abuse. Nevertheless, despite considerable progress in understanding the underlying behavioral indices of opioid dependence, as well as molecular and cellular mechanisms of sensitization (increased sensitivity to drug effects), tolerance (decreased sensitivity to drug effects), and withdrawal (the aversive experience taking place when the drug is no longer present), there is still a large amount of work to be done in understanding the behavioral and neural substrates of compul- sive opioid use.

Animal models of extinction/reinstatement have greatly con- tributed to our understanding of neurobiological processes under- pinning the rewarding properties of opioids as well as the neuronal substrates implicated in relapse to opioid-seeking. Importantly, studies conducted using these animal models of relapse have iden- tified the MOR and DOR as crucially implicated in goal-directed actions and inhibitory controls. Moreover, animal models of rein- statement point to KOR as key players in stress-induced addictive- like behaviors, since KOR stimulation can induce reinstatement via a stress-like mechanism.

Opioid Receptors in Relapse 291

292 Liana Fattore et al.

References

1. United Nation Office on Drugs and Crime (UNODC) (2013) World drug report. p. 18

2. Epstein DH, Preston KL, Stewart J et al (2006)
Toward a model of drug relapse: an assessment of the validity of the reinstatement procedure. Psychopharmacology 189:1–16

3. Tzschentke TM (2007) Measuring reward with the conditioned place preference (CPP) paradigm: update of the last decade. Addict Biol 12:227–462

4. Arnold JM, Roberts DC (1997) A critique of fixed and progressive ratio schedules used to examine the neural substrates of drug reinforce- ment. Pharmacol Biochem Behav 57:441–447

5. Shalev U, Grimm JW, Shaham Y (2002) Neurobiology of relapse to heroin and cocaine seeking: a review. Pharmacol Rev 54:1–42

6. Mello NK, Mendelson JH, Bree MP (1981) Naltrexone effects on morphine and food self- administration in morphine-dependent rhesus monkeys. J Pharmacol Exp Ther 218:550–557

7. Mello NK, Bree MP, Mendelson JH (1983) Comparison of buprenorphine and methadone effects on opiate self-administration in pri- mates. J Pharmacol Exp Ther 225:378–386

8. Matthes HW, Maldonado R, Simonin F et al (1996) Loss of morphine-induced analgesia, reward effect and withdrawal symptoms in mice lacking the mu-opioid-receptor gene. Nature 383:819–823

9. Liang J, Li Y, Ping X et al (2006) The possible involvement of endogenous ligands for mu-, delta- and kappa-opioid receptors in modulat- ing morphine-induced CPP expression in rats. Peptides 27:3307–3314

10. Le Merrer J, Becker JA, Befort K et al (2009) Reward processing by the opioid system in the brain. Physiol Rev 89:1379–1412

11. Lutz PE, Kieffer BL (2013) The multiple fac- ets of opioid receptor function: implications for addiction. Curr Opin Neurobiol 23:473–479

12. De Wit H, Stewart J (1983) Drug reinstate- ment of heroin-reinforced responding in the rat. Psychopharmacology 79:29–31

13. Stewart J (1983) Conditioned and uncondi- tioned drug effects in relapse to opiate and stimulant drug self-administration. Prog Neuropsychopharmacol Biol Psychiatry 7: 591–597

14. Fattore L, Spano MS, Cossu G et al (2003) Cannabinoid mechanism in reinstatement of heroin-seeking after a long period of absti- nence in rats. Eur J Neurosci 17:1723–1726

15. Fattore L, Spano S, Cossu G et al (2005) Cannabinoid CB(1) antagonist SR 141716A

attenuates reinstatement of heroin self- administration in heroin-abstinent rats. Neuropharmacology 48:1097–1104

16. Spano MS, Fattore L, Fratta W et al (2007) The GABAB receptor agonist baclofen pre- vents heroin-induced reinstatement of heroin- seeking behavior in rats. Neuropharmacology 52:1555–1562

17. Parker LA, Mcdonald RV (2000) Reinstatement of both a conditioned place preference and a conditioned place aversion with drug primes. Pharmacol Biochem Behav 66:559–561

18. Ribeiro Do Couto B, Aguilar MA et al (2003) Reinstatement of morphine-induced condi- tioned place preference in mice by priming injections. Neural Plast 10:279–290

19. Mueller D, Perdikaris D, Stewart J (2002) Persistence and drug-induced reinstatement of a morphine-induced conditioned place prefer- ence. Behav Brain Res 136:389–397

20. Ribeiro Do Couto B, Aguilar MA et al (2005) Long-lasting rewarding effects of morphine induced by drug primings. Brain Res 1050:53–63

21. Stewart J, Wise RA (1992) Reinstatement of heroin self-administration habits: morphine prompts and naltrexone discourages renewed responding after extinction. Psychopharmacology 108:79–84

22. Stewart J (1984) Reinstatement of heroin and cocaine self-administration behavior in the rat by intracerebral application of morphine in the ventral tegmental area. Pharmacol Biochem Behav 20:917–923

23. Shaham Y, Stewart J (1996) Effects of opioid and dopamine receptor antagonists on relapse induced by stress and re-exposure to heroin in rats. Psychopharmacology 125: 385–391

24. Shaham Y, Rajabi H, Stewart J (1996) Relapse to heroin-seeking in rats under opioid mainte- nance: the effects of stress, heroin priming, and withdrawal. J Neurosci 16:1957–1963

25. Leri F, Rizos Z (2005) Reconditioning of drug-related cues: a potential contributor to relapse after drug reexposure. Pharmacol Biochem Behav 80:621–630

26. Liu H, Lai M, Zhou X et al (2012) Galantamine attenuates the heroin seeking behaviors induced by cues after prolonged withdrawal in rats. Neuropharmacology 62:2515–2521

27. Zhang F, Zhou W, Liu H et al (2005) Increased c-Fos expression in the medial part of the lat- eral habenula during cue-evoked heroin- seeking in rats. Neurosci Lett 386:133–137

28. Berger AC, Whistler JL (2011) Morphine- induced mu opioid receptor trafficking enhances reward yet prevents compulsive drug use. EMBO Mol Med 3:385–397

29. Finn AK, Whistler JL (2001) Endocytosis of the mu opioid receptor reduces tolerance and a cellular hallmark of opiate withdrawal. Neuron 32:829–839

30. Nathan PJ, Bush MA, Tao WX et al (2012) Multiple-dose safety, pharmacokinetics, and pharmacodynamics of the μ-opioid receptor inverse agonist GSK1521498. J Clin Pharmacol 52:1456–1467

31. Cambridge VC, Ziauddeen H, Nathan PJ et al (2013) Neural and behavioral effects of a novel mu opioid receptor antagonist in binge-eating obese people. Biol Psychiatry 73:887–894

32. Giuliano C, Robbins TW, Wille DR et al (2013) Attenuation of cocaine and heroin seeking by μ-opioid receptor antagonism. Psychopharmacology 227:137–147

33. Cummins E, Boughner E, Leri F (2014) Cue- induced renewal of heroin place preference: involvement of the basolateral amygdala. Neuroreport 25:297–302

34. Lu L, Chen H, Su W et al (2005) Role of with- drawal in reinstatement of morphine-conditioned place preference. Psychopharmaocology 181: 90–100

35. Pfeiffer A, Brantl V, Herz A et al (1986) Psychotomimesis mediated by kappa opiate receptors. Science 233:774–776

36. Carey AN, Borozny K, Aldrich JV et al (2007) Reinstatement of cocaine place-conditioning prevented by the peptide kappa-opioid recep- tor antagonist arodyn. Eur J Pharmacol 569: 84–89

37. Redila VA, Chavkin C (2008) Stress-induced reinstatement of cocaine seeking is mediated by the kappa opioid system. Psychopharmacology 200:59–70

38. Land BB, Bruchas MR, Schattauer S et al (2009) Activation of the kappa opioid receptor in the dorsal raphe nucleus mediates the aver- sive effects of stress and reinstates drug seeking. Proc Natl Acad Sci U S A 106:19168–19173

39. Zhou Y, Leri F, Grella SL et al (2013) Involvement of dynorphin and kappa opioid receptor in yohimbine-induced reinstatement of heroin seeking in rats. Synapse 67:358–361

40. Bruchas MR, Land BB, Chavkin C (2010) The dynorphin/kappa opioid system as a modulator

of stress-induced and pro-addictive behaviors.

Brain Res 1314:44–55
41. Cordery SF, Taverner A, Ridzwan IE et al (2012)

A non-rewarding, non-aversive buprenorphine/ naltrexone combination attenuates drug-primed reinstatement to cocaine and morphine in rats in a conditioned place preference paradigm. Addict Biol. doi:10.1111/adb.12020

42. Abdelhamid EE, Sultana M, Portoghese PS et al (1991) Selective blockage of delta opioid receptors prevents the development of mor- phine tolerance and dependence in mice. J Pharmacol Exp Ther 258:299–303

43. Zhu Y, King MA, Schuller AG et al (1999) Retention of supraspinal delta-like analgesia and loss of morphine tolerance in delta opioid receptor knockout mice. Neuron 24:243–252

44. Ciccocioppo R, Martin-Fardon R, Weiss F (2002) Effect of selective blockade of mu(1) or delta opioid receptors on reinstatement of alcohol-seeking behavior by drug-associated stimuli in rats. Neuropsychopharmacology 27:391–399

45. Marinelli PW, Funk D, Harding S et al (2009) Roles of opioid receptor subtypes in mediating alcohol-seeking induced by discrete cues and context. Eur J Neurosci 30:671–678

46. Kotlinska JH, Gibula-Bruzda E, Pachuta A et al (2010) Influence of new deltorphin ana- logues on reinstatement of cocaine-induced conditioned place preference in rats. Behav Pharmacol 21:638–648

47. Le Merrer J, Faget L, Matifas A et al (2012) Cues predicting drug or food reward restore morphine-induced place conditioning in mice lacking delta opioid receptors. Psycho- pharmacology 223:99–106

48. Lenard NR, Daniels DJ, Portoghese PS et al (2007) Absence of conditioned place prefer- ence or reinstatement with bivalent ligands containing mu-opioid receptor agonist and delta-opioid receptor antagonist pharmacoph- ores. Eur J Pharmacol 566:75–82

49.Chefer VI, Shippenberg TS (2009) Augmentation of morphine-induced sensitiza- tion but reduction in morphine tolerance and reward in delta-opioid receptor knockout mice. Neuropsychopharmacology 34:887–898

50. Le Merrer J, Plaza-Zabala A, Del Boca C et al (2011) Deletion of the delta opioid receptor gene impairs place conditioning but preserves morphine reinforcement. Biol Psychiatry 69: 700–703

Opioid Receptors in Relapse 293

Chapter 25 Analysis of Opioid-Seeking Reinstatement in the Rat

Liana Fattore, Paola Fadda, Mary Tresa Zanda, and Walter Fratta

Abstract

The inability to maintain drug abstinence is often referred to as relapse and consists of a process by which an abstaining individual slips back into old behavioral patterns and substance use. Animal models of relapse have been developed and validated over the last decades, and significantly contributed to shed light on the neurobiological mechanisms underlying vulnerability to relapse. The most common proce- dure to study drug-seeking and relapse-like behavior in animals is the “reinstatement model.” Originally elaborated by Pavlov and Skinner, the concepts of reinforced operant responding and conditioned behav- ior were applied to addiction research not before 1971 (Stretch et al., Can J Physiol Pharmacol 49:581– 589, 1971), and the first report of a reinstatement animal model as it is now used worldwide was published only 10 years later (De Wit and Stewart, Psychopharmacology 75:134–143, 1981). According to the proposed model, opioids are typically self-administered intravenously, as humans do, and although rodents are most often employed in these studies, this model has been used with a variety of species including nonhuman primates, dogs, cats, and pigeons. A variety of operant responses are available, depending on the species studied. For example, a lever press or a nose poke response typically is used for rodents, whereas a panel press response typically is used for nonhuman primates. Here, we describe a simple and easily reproducible protocol of heroin-seeking reinstatement in rats, which proved useful to study the neurobiological mechanisms underlying relapse to heroin and vulnerability factors enhancing the resumption of heroin-seeking behavior.

Key words Drug-associated cues, Drug priming, Heroin-seeking, Relapse, Reinstatement, Stress

1 Introduction

Under the definition of “drug-seeking reinstatement” different pro- cedures are currently included, all of them assessing the resumption of a previously extinguished drug-reinforced behavior in response to a trigger of different nature [1, 2]. The reinstatement model is cur- rently used in many laboratories to investigate mechanisms under- lying “relapse” or relapse-like behavior. However, it should be noted that, contrary to the human situation in which relapse is defined as renewed drug consumption following a period of abstinence, the reinstatement test is performed in animals under drug-free

Santi M. Spampinato (ed.), Opioid Receptors: Methods and Protocols, Methods in Molecular Biology, vol. 1230, DOI 10.1007/978-1-4939-1708-2_25, © Springer Science+Business Media New York 2015

295

296 Liana Fattore et al.

2 Materials

2.1 Animals

conditions. Yet, this model proved to be useful to study the brain mechanisms underlying craving and increased drug-seeking behavior. From a procedural point of view, a reinstatement procedure needs to first “instate” the drug-seeking behavior to an adequate level, then to introduce a drug-free period and extinguish drug-seeking, and finally to test the resumption of the extinguished behavior in response to a specific trigger.

In humans, a variety of events or stimuli can precipitate drug craving and elicit the urge to use the drug, which ultimately result in relapse in abstinent individuals. The presentation of drugs them- selves, or stimuli previously associated with drug delivery (e.g., place, people, paraphernalia), or stressful events (loss of beloved person or employment, divorce) all increase the motivation to engage in drug-taking and the likelihood of relapse. Notably, the conditions that reinstate drug-seeking in laboratory animals are similar to those that trigger relapse in humans, and include small doses of the drug itself, environmental stimuli previously associ- ated with the drug delivery (i.e., cues), and exposure to stressors (electrical footshock, food deprivation), thereby demonstrating the predictive validity of this model. These three kinds of stimuli classify a wide number of procedural alternatives.

This protocol allowed the recognition of the specific roles played by the different opioid receptor subtypes in the resumption of heroin-seeking, which have been discussed in details in a dedi- cated chapter of this book (see Chapter 24). Importantly, it has been demonstrated that acute priming with drugs displaying high affinity for μ-opioid receptors dose-dependently reinstates extin- guished heroin-seeking behavior, while opioids with high affinity for κ-opioid receptors typically do not.

Use adult rats weighing 260 ± 280 g at the beginning of the experi- ments, housed 4/cage, handled daily for approximately 10 min for at least 6 days after arrival and maintained at a temperature of 21 ± 2 °C under a reversed 12:12 h light/dark cycle with food and water freely available. Both males and females of different strains can be used, although significant sex- and strain-dependent differ- ences have been described in both heroin self-administration and heroin-seeking behavior (see Note 1). Perform surgery for intrave- nous catheter implantation 7 days after arrival, as describe in details below (Subheading 2.3). After surgery, rats should be housed indi- vidually for avoiding damages to the external part of the catheter assembly. Starting from 1 week of recovery, maintain animals on a restricted diet (about 20 g laboratory chow per day), given after each daily testing session, to maintain body weight and growth rates. Perform self-administration training during the dark phase of the cycle, at least 6 days/week.

2.2 Drugs

2.3 Surgery
for Implantation
of Venous Catheters

For self-administration training, dissolve heroin in heparinized (1 %) saline solution at a dose of 0.03 mg/inf. For priming tests, dissolve heroin in 0.9 % sterile saline solution at a dose of 0.1 mg/kg and administer it intravenously (i.v.) immediately before starting the session in a volume of 100 ml followed by 0.2 ml saline solution to flush the drug solution through the cath- eter. Heroin primings could also be administered subcutaneously at a dose of 0.25 mg/kg 10 min before starting the session.

Surgical instruments have to be sterilized using liquid sterilizers (NOT alcohol) or autoclave; if the first option is preferred, instru- ments have to be rinsed in sterile saline before using. Catheters can be sterilized in liquid sterilizers by immersion overnight, and then rinsed with sterile saline. Cotton swabs, drapes, and gauze can be autoclaved.

1. Under deep anesthesia, animals are implanted with silastic cath- eters inserted into the right external jugular vein. The rat should preferably be anesthetized with isoflurane 2 % or other inhalants (halothane), which allow a more rapid recovery, or with an intraperitoneal (i.p.) administration of injectable anesthetics, like a ketamine/xylazine combination or Equithesin (0.97 g pentobarbital, 2.1 g magnesium-sulfate, 4.25 g chloral hydrate, 42.8 ml propylene glycol, 11.5 ml ethanol 90 %, 5 ml/kg).

2. The sterile catheter should be connected to a 1 ml syringe filled with a heparinized (1 %) sterile saline solution. Pass the catheter under the skin and thread it out of an incision made on the back. Insert one end of the catheter into the right atrium via the right jugular vein, whereas pass the distal end subcutaneously to exit in the mid-scapular region. At the end of the surgical procedure, 10–15 ml/kg physiologic saline should be administered subcutaneously for hydration.

3. Allow rats to recover individually for a minimum of 6 days with food and water freely available, and give them daily subcutane- ous administration of 0.1 ml of enrofloxacin (Baytril, Bayer©) as post-surgery antibiotic therapy. Once animals recover from surgery, maintain a food supply of 20 g/day rat chow given at the end of the session with ad libitum access to water. Flush the catheter every second or third day with 0.2 ml heparinized (1 %) physiological saline. During recovery, changes in general behavior and body weight should be monitored.

4. Before starting each self-administration session, the catheter is connected to a single-guide cannula (Plastics One, Roanoke, VA, USA) and attached to an infusion harness (Instech Laboratories, Plymouth Melting, PA, USA). To ensure patency, after each daily session flush 0.1 ml heparinized sterile saline (20 USP/ml 9 % sterile saline) through the catheter and seal it with a stainless steel cap when not in use. When a catheter is

Heroin-Seeking Reinstatement 297

298 Liana Fattore et al.

2.4 Apparatus

obstructed or damaged, implant a new one into the left jugular vein, and resume testing 6 days after the animal recovered from surger y.

5. At the end of the study, confirm catheter patency by intrave- nous infusion of an short-acting anesthetic, such as barbiturate methohexital sodium (Brevital®, 10 mg/ml, 0.2 ml/rat); a positive test is indicated by loss of righting reflex within 5 s after injection.

1. The intravenous self-administration apparatus consists of Plexiglas operant chambers (29.5×32.5×23.5 cm), each encased in a sound- and light-attenuating cubicle equipped with a ventilation fan (Med Associates). The sound-proofing prevents the effects of unintentional environmental stimuli, such as ultrasonic vocalizations of other animals, while fan pro- vides a background white noise. The front panel of each cham- ber contains two retractable levers (each 4 cm wide), positioned 12 cm apart and 8 cm from the grid, and extending 1.5 cm into the box.

2. A white stimulus light (cue) is placed between the levers, and a dim red house light located on the opposite wall is kept on throughout the entire session. Helpful additional devices include lights or tones to be used as discriminative or condi- tioning stimuli, and food hoppers to assist in operant training (see Note 2) or for use in multiple schedules (see Note 3). A grid floor allows waste to collect in a tray easily removable for systematic cleaning. Chambers are equipped with syringe pump systems that consist of an infusion pump with a 5 ml syringe connected by a single channel 22 gauge swivel and an extra length of Teflon tubing enclosed in a metal spring to the catheter fitting on the animal’s back.

3. The infusion pump apparatus can be located within or outside the sound-attenuating cubicle that housed the operant cham- ber. Intravenous infusion of heroin is delivered by a software operated infusion pump (Med Associates). The swivel system should be extremely easy to turn; resistance to movement jeop- ardizes the delicate catheter assembly and affects the animal’s performance. The cylindrical swivel is anchored in an assembly that is counterbalanced with an adjustable weight so that the lead remains gently stretched tight to prevent twisting.

4. The lead that delivers the drug solution and attaches to the catheter must be flexible but coated or wrapped with a chew- resistant material such as wire. An internal cannula at the tip of the lead inserts into the external cannula guide of the implanted catheter. An IBM-compatible computer with Med-PC inter- face (Med Associates) is used for programming, data collection, and storage.

Heroin-Seeking Reinstatement 299

3 Methods

3.1 Heroin Self- Administration Training

1. Drug self-administration involves operant behavior that is reinforced and maintained by drug delivery. There are more than one approach possible to train animals to self-administer heroin: (1) to directly train animals with intravenous heroin contingently available upon active responses; (2) to initially train food-restricted animals with food as reinforcement, and then replace food with intravenous heroin as the reinforcer, keeping animals in a food-restricted state; (3) to initially train food- deprived animals with food reinforcement, and then replace food with intravenous heroin as the reinforcer, with food freely available. One objection to the two latter approaches is that the animals may self-administer heroin as a result of prior experience with food reinforcement generally, rather than of the reinforcing effects of the drug specifically. This concern may be obviated by studying extinction of responding in the absence of food or her- oin, and resumption of responding in the presence of heroin.

2. The most basic schedule of reinforcement is the fixed-ratio (FR) schedule, which defines the number of responses required per drug injection. Typically, responding is initiated using a continuous FR-1 schedule, so that each response in the pres- ence of a stimulus light will result in the intravenous injection of a drug solution. Each self-administration session starts with the extension of the two levers into the operant chamber. According to a continuous (FR-1) schedule of reinforcement, depression of one lever, defined as active, results in: (1) extinc- tion of the house light and the illumination of the stimulus light above the active lever, which remains on for 20 s, (2) retraction of both levers, and (3) activation of the infusion pump for 5.8 s delivering 0.1 ml intravenous infusion of 0.03 mg/kg heroin solution. On completion of the 20 s time-out period, levers are re-extended into the chamber, the stimulus light is out, and the house light is switched on. The drug delivery reinforces the behavior, making it more likely the rat will press the active lever again. Each response producing the drug reinforcer is contigu- ously paired with a brief presentation of an environmental stimulus (e.g., a tone or cue light) that, by virtue of repeated pairings, comes to serve as a conditioned stimulus.

3. Depressions on the other lever, defined as inactive, have no programmed consequences but were always recorded, thus providing an index of basal activity levels. The assignment of the active (drug-paired) and the inactive (not drug-paired) levers is counterbalanced between rats and remains constant for each subject throughout all experiments.

4. Under these circumstances, the experimenter controls the drug concentration and volume of each infusion, while animals control the timing of infusions. Similarly, humans also control

300

Liana Fattore et al.

3.2

Acquisition

their own drug intake when they have free access to the drug. The patterns of intake exhibited by animals in the self- administration paradigm are quite similar to those of humans.

5. The ability to self-administer drugs for short periods of time daily (1–2 h per session) results in a stable drug intake over time, a so-called “titration” process that is suggested to reflect individual control of intake responding to optimal dosing around which blood levels fluctuate in the course of the self- administration session.

6. The primary dependent measures are number of drug injec- tions and rate of responding during each session. However, different protocols may involve longer access to heroin, if not unlimited access to the drug (see Note 4).

1. Acquisition of drug self-administration behavior is defined as a transition process characterized by a relatively quick shift from low or no drug-maintained responding to high, stable levels of responding. Since this process is difficult to study in drug-naive humans, this particular phase of the self- administration training is critical, and allows studying how rapidly this process takes place or what percentage of ani- mals acquire drug self-administration, which environmen- tal conditions (feeding, stress, enriched environment) can accelerate or decrease acquisition of drug self-administra- tion, and which biological (age, sex, genetics), behavioral (impulsive traits, depression-like phenotype), and pharma- cological approaches (previous drug history, pretreatment drugs) may prevent or reverse acquisition.

2. Some protocols involve the use of 1–2 heroin priming infu- sions given by the experimenter at the beginning of the session in an attempt to facilitate acquisition. Carry out the acquisition sessions until steady baseline of heroin intake is reached (typi- cally within 8–10 days). Heroin self-administration is consid- ered to be acquired if an animal display accurate discrimination between the active and the inactive lever, with at least 15 active lever presses per session not differing by more than 20 % for five consecutive days, and less five inactive lever presses per session. Rats not meeting the acquisition criterion have to be excluded from the experiment.

3. As also for other drugs, particular attention should be paid to avoid excessive heroin intake and toxicity during training ses- sions. Drug intake can be limited by either increasing the scheduled timeout periods following each injection (see train- ing general description above) or limiting the total number of drug injections per session, with session automatically ceasing upon reaching the maximum number of injections allowed.

3.3 Maintenance

1. Continue the daily training until the total number of heroin infusions per session stabilizes with no more than 10 % varia- tion over 7 consecutive days. Once the animal has acquired the operant self-administration and is responding at stable levels over sessions, the maintenance of heroin self-administration can be examined. This is particularly used to determine whether the behavior is maintained or altered in response to a pharma- cological challenge or behavioral manipulations. For example, a common behavioral challenge is to change the schedule of heroin reinforcement, and to employ a progressive ratio (PR) schedule of reinforcement in which the number of lever presses required to obtain heroin delivery is increased after each reinforcement.

2. Under a progressive ratio schedule of reinforcement, a rat needs to work harder and harder to receive a single heroin infusion: the point at which animal ceases responding is called “breaking point.” This procedure is frequently used to deter- mine the degree of motivation to obtain heroin and proved to be a useful tool for investigating potential pharmacological treatments, e.g., to test whether the breaking point is reduced with a particular drug pretreatment. For a detailed description of heroin self-administration protocols under a progressive schedule of reinforcement see [3] and [4].

1. After stable and reliable responding level has been reached, drug-reinforced instrumental responding is extinguished by replacing heroin with saline solution allowing the animal to perform the operant response without the drug reward.

2. Under these conditions, each lever press of the active lever results in an infusion of 100 μl of saline. During extinction train- ing, behavioral responses eventually decrease (extinguish) as the animal learns that heroin is no longer achieved by pressing the lever. Conduct extinction sessions until responding is substan- tially lower than that maintained by delivery of heroin and deemed incidental rather than motivated by expectation of obtaining the drug. Heroin-reinforced behavior can be consid- ered extinguished when responding on the active lever is decreased by at least 85 % for 3 consecutive days, which typically occurs within 10 days. Yet, a “long-term extinction” paradigm (typically 3 weeks) can also be used, as it more closely resembles the human situation, where drug addicts may be in a “drug-free state” for weeks or months before relapse occurs [5–7].

3. It is important to note that extinction does not cause “forget- ting” of the original stimulus–reward relation [8], but promotes new learning that counters the motivating impact of drug- related cues [9], i.e., it is a process of new and active learning.

3.4 Extinction

Heroin-Seeking Reinstatement 301

302 Liana Fattore et al.

3.5 Heroin-Seeking Reinstatement

4. To habituate animals to subsequent drug priming administra- tions for the reinstatement test sessions, on both the third from last day and the last day of extinction training, prime rats with an injection of saline (2 ml). Route and timing of saline priming administrations should be the same of the drug prim- ings planned for the subsequent reinstatement testing.

1. Reinstatement refers to the return to drug-seeking behavior (e.g., lever-pressing activity) even though the behavior is not rewarded with the drug. The reemergence of heroin-seeking behavior in response to a specific trigger is conducted under a drug-free condition. This allows studying the recovery of the extinguished behavior without the interference produced by the psychoactive effects of the drug.

2. After reaching extinction criterion, give each rat a noncontin- gent non-reinforced priming—a drug injection, exposure to a drug-associated cue or stress—to test its effect on the resump- tion of heroin-seeking behavior. Assign counterbalanced treat- ments on the basis of a Latin square design. Catheter life permitting, test the same animals with different primings; however, do not test more than two drug primings (plus one saline/vehicle priming) per animal, with a minimum of three extinction days separating each drug treatment to assess car- ryover effects.

3. The increase in the number of operant responses as compared with that observed during extinction is assumed as a heighten- ing of the subject’s heroin-seeking behavior. Thus, reinstate- ment serves as index of the animal’s drug-seeking, not drug-taking, behavior and presumably reflects the animal’s desire and drive to receive the drug of abuse. In the reinstate- ment test, a stimulus is said to reinstate responding if it causes an increase in responding that was previously reinforced by the drug. This model has been pharmacologically validated with drugs that reduce opioid craving and relapse in opioid- dependent patients.

4. It is important to note that heroin self-administration, extinc- tion, and the reinstatement test sessions here described are performed over several experimental sessions, i.e., using a “between-session” design, although some groups prefer using a “within-session” protocol, i.e., they implement all phases in a single session [10].

1. A typical trigger to relapse to heroin use occurs when people consume a small amount of the drug itself, believing that they can “handle it.” In heroin addicts, this small amount of the drug actually serves as a kind of reminder that induces a full return to drug-taking and relapse.

3.6 Drug Primings

3.7 Cue Priming

2. In animal models of drug-induced reinstatement, priming injections are usually higher than the unit injection dose used in the self-administration procedure. In heroin-seeking reinstate- ment testing, drug primings usually range between 0.1 and 0.3 mg/kg, and are administered preferentially by intravenous route but also subcutaneously or intraperitoneally, respectively, immediately before or 10 min before starting the session.

3. Drug priming injections could be considered as a particular case of cue-induced reinstatement (in which drug administra- tion acts as an interoceptive cue), but it is important to con- sider that when priming injections are used this procedure cannot be longer considered as a “drug-free” test.

1. Human heroin addicts frequently return to drug use after being exposed to the cues or places associated with the drug consumption.

2. The cue-induced model of heroin-seeking reinstatement involves the use of a heroin-predictive discriminative stimulus, usually a cue light associated with the response-contingent presentation of heroin during the self-administration training. Indeed, in most of the cue-induced reinstatement protocols of heroin-seeking behavior, cues (light, tone) are contingently presented upon the completion of the operant requirements and act as “conditioned reinforcers” which boost lever-pressing behavior. Learned Pavlovian associations between heroin- induced positive effects and the cue that predicts them endow these drug cues with the ability to trigger appetitive heroin- directed responses [11].

3. In this protocol of heroin-seeking reinstatement, it is necessary to suppress the contingent presentation of the cue upon active responses during extinction training, i.e., to avoid its association with no-reinforced responses [12]. During the reinstatement test session, animal is re-exposed once to the heroin-associated cue at the beginning of the session.

1. Stressful events are well known to activate the hypothalamic- pituitary-adrenal axis and to enhance vulnerability to relapse to heroin use in abstinent patients.

2. Stress-induced reinstatement of heroin-seeking is usually imple- mented by inescapable, intermittent footshock (0.5–0.8 mA, 0.5 s on with a mean off period of 40 s) administered for 10–15 min just before starting the session (for a review, see [13]). Yet, other methodological possibilities have also been explored.

3. Acute (21 h) or prolonged (48 h) food deprivation, for exam- ple, can be used to trigger reinstatement of opioid-seeking [14, 15], and it is of special interest because its effects on

3.8 Stress Primings

Heroin-Seeking Reinstatement 303

304 Liana Fattore et al.

3.9 Factors Modulating Opioid- Seeking Reinstatement

reinstatement can be abolished by leptin administration [16]. Moreover, an injection of the alpha2-adrenergic antagonist yohimbine is often used to elicit opioid-seeking reinstatement in heroin-experienced rats [17, 18], in line with the reported ability of yohimbine-induced stress to enhance craving in opioid-dependent human subjects [19].

4. It should be noted that the neurobiological mechanisms underly- ing the ability of these different kinds of stimuli (stress, cue, or priming injections) to initiate and sustain lever-pressing behavior after extinction show only a partial overlap and indeed they can show additive efficacy when combined [20, 21] (see Note 5).

1. The wide availability of reliable animal models for studying relapse to opioid-seeking in humans led to the identification of a number of factors that may significantly affect heroin-taking and heroin-seeking behaviors. For example, increasing either the amount of heroin self-administered or the duration of the withdrawal period amplifies stress-induced reinstatement of heroin-seeking [22, 16], while spontaneous withdrawal from heroin reinstated drug-seeking [23].

2. Curiously, heroin-seeking is positively associated with both the total amount of heroin taken during self-administration train- ing [4] and the number of daily heroin self-administration training sessions [24]. Increasing the duration of either heroin self-administration or the withdrawal periods from heroin self- administration augments the reinstatement induced by cues that were associated previously with heroin reinforcement [24]. In other terms, more extended the heroin self-administration training period, the greater the reinstatement of drug-seeking elicited by a conditioned cue.

1. Female rats acquired heroin self-administration more rapidly than males [25]. Besides sex, opioid self-administration may considerably vary also depending upon the strain and the age of the animals used. For example, Lewis rats showed higher acquisition rate of morphine self-administration than Fisher 344 rat strain [26], while adolescent rats self-administer more heroin but exhibit less heroin-seeking than adults [27].

2. Initial training with a schedule of food reinforcement yields rapid and reliable acquisition of operant behavior and may facilitate acquisition of heroin self-administration, both in terms of time to meet baseline criteria for self-administration and in the percentage of animals that ultimately do so. This approach implies the disadvantage to complicate interpretation

4 Notes

Heroin-Seeking Reinstatement 305

of subsequent self-administration behavior. Yet, for studies focused on the reinstatement of drug-seeking behavior, initial food training has the advantage to shorten the acquisition of a stable drug-taking and, more generally, to accelerate the com- pletion of the entire training procedural phases.

3. Heroin-seeking behavior can be measured under different, multiple schedules of reinforcement, in which prolonged periods of drug-seeking behavior are maintained by contin- gent presentation of a drug-associated conditioned rein- forcer. Specifically, rats are first trained to self-administer heroin and then to seek the drug over prolonged periods of time under a second-order schedule of reinforcement, in which responding is maintained by contingent presentation of a drug-associated conditioned reinforcer. This procedure allows the experimenter to test the effect of a pharmacologi- cal compound on drug-seeking both before and after heroin is self-administered. For a detailed description of heroin self- administration protocols under second-order schedule of reinforcement see [28] and [29].

4. If exposed to short daily sessions and moderate drug doses, rats exhibit stable rates of intake across many days. Conversely, if animals are exposed to high doses of drug and/or long daily sessions, animals exhibit escalation of drug intake across repeated self-administration sessions, as well as other behaviors that are considered addiction-like. For example, when rats are given long access to heroin during 11-h daily sessions, the total intake of the drug gradually escalates. After escalation, rats exposed to long-access heroin are slower to extinguish heroin- seeking behavior and more responsive to the reinstating effect of stress during reinstatement test sessions conducted after extinction [22].

5. The neural mechanisms underlying relapse to heroin-seeking induced by exposure to the self-administered drug (drug priming), conditioned drug cues, and stressors are dissociable although recent studies revealed a large degree of overlap among them. Indeed, earlier studies demonstrated that drug- induced reinstatement of heroin-seeking involves activation of the mesolimbic dopaminergic pathways, while cue-induced reinstatement mostly involves the basolateral amygdala. Conversely, stress-induced reinstatement involves brain nor- adrenergic (NE) and corticotrophin-releasing factor (CRF) systems, and their interaction at the level of the bed nucleus of the stria terminalis [30]. However, these neuroanatomical distinctions are not sharp-cutting as, for example, decreasing glutamate transmission in the ventral tegmental area was found to attenuate cue-induced reinstatement of heroin- seeking [31].

306 Liana Fattore et al.

References

1. Stretch R, Gerber GJ, Wood SM (1971) Factors affecting behavior maintained by response-contingent intravenous infusions of amphetamine in squirrel monkeys. Can J Physiol Pharmacol 49:581–589

2. De Wit H, Stewart J (1981) Reinstatement of cocaine-reinforced responding in the rat. Psychopharmacology 75:134–143

3. Roberts DC, Bennett SA (1993) Heroin self- administration in rats under a progressive ratio schedule of reinforcement. Psychopharmacology 111:215–218

4. Zhang F, Zhou W, Tang S et al (2004) Motivation of heroin-seeking elicited by drug- associated cues is related to total amount of heroin exposure during self-administration in rats. Pharmacol Biochem Behav 79:291–298

5. Fattore L, Spano MS, Cossu G et al (2003) Cannabinoid mechanism in reinstatement of heroin-seeking after a long period of absti- nence in rats. Eur J Neurosci 17:1723–1726

6. Fattore L, Spano S, Cossu G et al (2005) Cannabinoid CB(1) antagonist SR 141716A attenuates reinstatement of heroin self- administration in heroin-abstinent rats. Neuropharmacology 48:1097–1104

7. Spano MS, Fattore L, Fratta W et al (2007) The GABAB receptor agonist baclofen pre- vents heroin-induced reinstatement of heroin- seeking behavior in rats. Neuropharmacology 52:1555–1562

8. Bouton ME (2004) Context and behavioral processesinextinction.LearnMem11:485–494

9. Havermans RC, Jansen AT (2003) Increasing the efficacy of cue exposure treatment in pre- venting relapse of addictive behavior. Addict Behav 28:989–994

10. Martin TJ, Walker LE, Sizemore GM et al (1996) Within-session determination of dose- response curves for heroin self-administration in rats: comparison with between-session determination and effects of naltrexone. Drug Alcohol Depend 41:93–100

11. Stewart J (1984) Reinstatement of heroin and cocaine self-administration behavior in the rat by intracerebral application of morphine in the ventral tegmental area. Pharmacol Biochem Behav 20:917–923

12. Fattore L, Spano MS, Altea S et al (2010) Drug- and cue-induced reinstatement of cannabinoid-seeking behavior in male and female rats: influence of ovarian hormones. Br J Pharmacol 160:724–735

13. Shaham Y, Shalev U, Lu L et al (2003) The reinstatement model of drug relapse:

histor y, methodology and major findings. Psychopharmacology 168:3–20

14. Tobin S, Newman AH, Quinn T et al (2009) A role for dopamine D1-like receptors in acute food deprivation-induced reinstatement of her- oin seeking in rats. Int J Neuropsychopharmacol 12:217–226

15. Shalev U (2011) Chronic food restriction aug- ments the reinstatement of extinguished heroin-seeking behavior in rats. Addict Biol 17:691–693

16. Shalev U, Yap J, Shaham Y (2001) Leptin attenuates acute food deprivation-induced relapse to heroin seeking. J Neurosci 21:RC129

17. Banna KM, Back SE, Do P et al (2010) Yohimbine stress potentiates conditioned cue- induced reinstatement of heroin-seeking in rats. Behav Brain Res 208:144–148

18. Zhou Y, Leri F, Grella SL et al (2013) Involvement of dynorphin and kappa opioid receptor in yohimbine-induced reinstatement of heroin seeking in rats. Synapse 67:358–361

19. Stine SM, Southwick SM, Petrakis IL et al (2002) Yohimbine-induced withdrawal and anxiety symptoms in opioid-dependent patients. Biol Psychiatry 51:642–651

20. Liu X, Weiss F (2002) Additive effect of stress and drug cues on reinstatement of ethanol seeking: exacerbation by history of dependence and role of concurrent activation of corticotropin-releasing factor and opioid mechanisms. J Neurosci 22:7856–7861

21.Shalev U, Grimm JW, Shaham Y (2002) Neurobiology of relapse to heroin and cocaine seeking: a review. Pharmacol Rev 54:1–42

22. Ahmed SH, Walker JR, Koob GF (2000) Persistent increase in the motivation to take heroin in rats with a history of drug escalation. Neuropsychopharmacology 22:413–421

23. Shaham Y, Stewart J (1996) Effects of opioid and dopamine receptor antagonists on relapse induced by stress and re-exposure to heroin in rats. Psychopharmacology 125:385–391

24. Zhou W, Zhang F, Liu H et al (2009) Effects of training and withdrawal periods on heroin seeking induced by conditioned cue in an ani- mal of model of relapse. Psychopharmacology 203:677–684

25. Lynch WJ, Carroll ME (1999) Sex differences in the acquisition of intravenously self- administered cocaine and heroin in rats. Psychopharmacology 144:77–82

26. Ambrosio E, Goldberg SR, Elmer GI (1995) Behavior genetic investigation of the relationship between spontaneous locomotor activity and the

acquisition of morphine self-administration

behavior. Behav Pharmacol 6:229–237

27. Doherty JM, Frantz KJ (2012) Heroin self- administration and reinstatement of heroin- seeking in adolescent vs adult male rats.
Psychopharmacology 219:763–773

28. Corrigall WA, Coen KM (1989) Fixed-interval schedules for drug self-administration in the
rat. Psychopharmacology 99:136–139

29. Everitt BJ, Robbins TW (2000) Second-order schedules of drug reinforcement in rats and

monkeys: measurement of reinforcing efficacy and drug-seeking behavior. Psychopharmaco- logy 153:17–30

30. Stewart J (2000) Pathways to relapse: the neurobiology of drug- and stress-induced relapse to drug-taking. J Psychiatry Neurosci 25:125–136

31. Bossert JM, Liu SY, Lu L et al (2004) A role of ventral tegmental area glutamate in contex- tual cue-induced relapse to heroin seeking. J Neurosci 24:10726–10730

Heroin-Seeking Reinstatement 307

Induction of a High Alcohol Consumption in Rats and Mice: Role of Opioid Receptors

Roberto Rimondini and Gabriele Campana Abstract

Alcohol dependence continues to be an important health concern and animal models are critical to furthering our understanding of this complex disease. A hallmark feature of alcoholism is a significant increase in alcohol drinking over time. While several different animal models of excessive alcohol (ethanol) drinking exist for mice and rats, a growing number of laboratories are using a model that combines chronic ethanol exposure procedures with voluntary ethanol drinking with mice as experimental subjects. In the last years several experimental evidences have shown an involvement of opioid system in alcoholism.

Key words Alcoholism, Animal models, Limited access, Opioid receptor, Two-bottle free choice

1 Introduction

Since the mid-90s, several strain of mice lacking mu-opioid receptors (MOR), delta-opioid receptors (DOR) and kappa- opioid receptors (KOR) have been produced, without any anatomical defects. The breeding MOR/KOR or MOR/DOR- deficient mice [1] produced MOR/DOR double mutants. Triple mutants were obtained from MOR/KOR×MOR/DOR mating [2]. To investigate the role for the opioid system in alcoholism, single or combinatorial MOR, DOR, and KOR knockout mice could be use since they are viable [3–5].

Here we describe different methods to induce a high alcohol consumption depending on use of selected animals lines with high innate alcohol intake (>4 g/kg/day) i.e. C57B6J and 129X1/SvJ mice, AA rats or unselected line.

All the described paradigms may include fading of sweeteners from the ethanol solutions or an ascending ethanol concentration series.

It is recommended to initiate the acquisition phase before or, at the latest, at the age of 3 months. Both male and female animals can be used.

Santi M. Spampinato (ed.), Opioid Receptors: Methods and Protocols, Methods in Molecular Biology, vol. 1230, DOI 10.1007/978-1-4939-1708-2_26, © Springer Science+Business Media New York 2015

309

Chapter 26

310 Roberto Rimondini and Gabriele Campana

2 Materials

2.1 Housing Conditions

2.2 Equipment

The animal rooms should have controlled temperature (20–24 °C) and humidity (55±10 %). Reversed light/dark cycle is recom- mended, but is not necessary, especially if 24-h continual alcohol drinking is measured. Limited access drinking sessions can occur also during the light phase.

1. This relatively simple procedure can be adopted in most labo- ratory animal units with standard equipment. The free-choice drinking protocols require individual animal housing during the experiments. The cage types employed include the wire mesh hanging cages and the standard Macrolon 3-type poly- carbonate cages.

2. For accurate measurement of fluid consumption, the drinking bottles should allow readings to be made to the nearest 0.1 ml or 0.1 g. Depending on the cage type, both graduated Richter tubes and polycarbonate drinking bottles equipped with ball bearing caps can be used. Spillage and evaporation should be minimized, and if possible, quantified.

3. Measurement of concomitant food consumption, if desired, requires special food containers that minimize spillage.

Animals should be habituated to the individual cages for 1 week at least and food is freely available at all times.

10 % Ethanol solution is presented in both drinking bottles (see Notes 1 and 2).

1. Animals are allowed to trained to drink 10 % ethanol in a free- choice two-bottle choice situation in their home cages.

2. Animals are presented with two bottles: one containing the ethanol solution, the other tap water (see Note 3).

3. Measurements from the drinking bottles are taken daily at the same time.

Measuring intervals of 24 h may not allow assessment of the effects of pharmacological agents on alcohol drinking, especially if the agents are short acting (see Notes 4 and 5).

1. Limit daily alcohol access first to 4 h (five sessions), then 2 h (five sessions), and finally 1 h, until stable 1-h drinking is achieved. Also shorter drinking sessions are possible. This phase may take 2–3 weeks.

3 Methods

3.1 Protocol
for Animals with High Alcohol Consumption

3.1.1 Forced Alcohol Drinking (4–7 Days)

3.1.2 Free-Choice Alcohol Drinking

3.1.3 Limited Access Drinking

3.2 Protocol for Unselected Heterogeneous Animals

3.2.1 Bottle Free-Choices

3.2.2 Limited Access

Opioid Receptor and Alcoholism 311 2. Water and food are continuously available.

3. Limited access sessions can occur daily, or only every 2nd day (see Notes 6 and 7).

4. Limited access sessions can occur also during the light phase. However, placing the limited access period at the beginning of the dark phase (the naturally occurring drinking peak) may enhance acquisition (see Note 8).

1. Animals are placed in single cages for 1-week habituation before to start the experiment.

2. Ethanol concentration is increased as follows:

Check bottles Monday, Wednesday, Fridays. Weight on Monday.

1. Limit daily alcohol access during the first week is lasting 4 h (five sessions Day 1–5). In other words, during the first week (day 1–5) animals are placed in single cages and have access to two bottle for a period of 4 h (bottle 1 EtOH concentration 2 %+0.2 % saccharin; bottle 2 0.2 % saccharin). After 4 h animals return to home cages.

2. After the first week the two bottles are presented for 2 h (see Notes 9 and 10).

3. Ethanol concentration is increased as follows:

Day 1–5

Bottle 1 2 %
EtOH+0.2 % saccharin

Bottle 2 0.2 % saccharin

Day 6–7

Bottle 1 4 % EtOH +0.2 % saccharin

Bottle 2 0.2 % saccharin

Day 8 until the end of exp

Bottle 1 6 %
EtOH+0.2 % saccharin

Bottle 2 0.2 % saccharin

First week for 4 h

Bottle 1 2 %
EtOH+0.2 % saccharin

Bottle 2 0.2 % saccharin

Monday–Tuesday– Wednesday

Bottle 1 4 % EtOH +0.2 % saccharin

Bottle 2 0.2 % saccharin

Thursday until the end of exp

Bottle 1 6 %
EtOH+0.2 % saccharin

Bottle 2 0.2 % saccharin

3.3 Outcome Measure for Induction of High Alcohol Consumption

The outcome measure reported is usually ethanol intake, expressed as g/kg body weight/day. In order to calculate the intake, animals have to be weighed at regular intervals. Stable intake levels are achieved after 3–4 weeks of free-choice drinking. Also ethanol preference, i.e., the ratio of ethanol and total fluid intake (in ml) has often been used.

312 Roberto Rimondini and Gabriele Campana

4 Notes

References

1. Filliol D, Ghozland S, Chluba J et al (2000) The δ- and μ-opioid receptor-deficient mice exhibit opposing alterations of emotional responses. Nat Genet 25:195–200

2. Kieffer BL, Gavériaux-Ruff C (2002) Exploring the opioid system by gene knockout. Prog Neurobiol 66:285–306

3. Kovacs KM, Szakall I, O’Brien D et al (2005) Decreased oral self-administration of alcohol in kappa-opioid receptor knock-out mice. Alcohol Clin Exp Res 29:730–738

4. Roberts AJ, Gold LH, Polis I et al (2001) Increased ethanol self-administration in delta- opioid receptor knockout mice. Alcohol Clin Exp Res 25:1249–1256

5. Kang-Park MH, Kieffer BL, Roberts AJ et al (2009) Mu-opioid receptors selectively regu- late basal inhibitory transmission in the central amygdala: lack of ethanol interactions. J Pharmacol Exp Ther 328:284–293

1. This procedure may speed the acquisition of alcohol drinking in alcohol-preferring animals, because it exposes them immedi- ately to high blood alcohol levels, but it is not necessary for acquisition.

2. This procedure is not recommended for normal heterogeneous strains, because it may cause aversion to alcohol in most sub- jects with no inborn alcohol preference.

3. In order to avoid position preferences, the positions of the bottles are changed once a week.

4. For pharmacological experiments, it is possible to limit alcohol access to short daily periods that can range from 30 min to 4 h.

5. This can be achieved after the subjects have reached stable 24-h alcohol consumption.

6. It may be necessary to include special drinking bottles or drink- ing spouts for accurate quantification of alcohol intake.

7. For best results, animals have to be weighed frequently; if weighing takes place immediately before drinking sessions, the procedure signals the animals the start of the session.

8. Typically, the level of alcohol intake during daily 1-h access in animals (mice and rats) is approximately 20 % of their total 24-h intake.

9. Water and food are continuously available.

10. Limited access sessions can occur daily and bottles have to be presented every day.

Chapter 27

Evaluation of Social and Nonsocial Behaviors Mediated by Opioids in Mouse Pups

Francesca R. D’Amato Abstract

The experimental approach to carry out a behavioral study involving opioids in mouse pups needs equipments and procedures different from those used for adult animals. Pups are immature at birth and only slowly acquire all the potentialities that characterize adult con-specifics. The standard and abnormal development of behav- ioral systems and their neural correlates can be followed during the first postnatal weeks, using appropriate methodologies that exploit characteristic pups’ capabilities. Behavioral tests designed for pups to evaluate the activity and involvement of the opioid system, according to the well-known role of the system in adult animals, are described in this chapter.

Key words Postnatal development, Opioid system, Behavioral tests, Ultrasonic vocalizations, Homing, Tail flick

1 Introduction

Rats and mice are altricial mammals, and this implies that pups are immature at birth and completely depend on the mother for their survival. At birth they are naked and cannot thermoregulate, their eyes and ears are closed, and they have poor sensory-motor capa- bilities. During the first days of life they are only able to orient in their environment according to their olfactory and proprioceptor capabilities. These capacities allow them to survive if they remain in close proximity with the mother to gain food, warmth, and protec- tion from inter and intraspecific environmental threats. Altricial pups complete their maturation during their postnatal life, facilitat- ing/allowing environmental factors to play a relevant role in pro- gramming the individual phenotype [1]. These pups cannot be simply considered “small adults,” rather they are different from adult con-specifics, and for this reason, the behavioral phenotyping assessment needs specifically designed tests, techniques, and appara- tuses. This is important because tested pups should have the capac- ity both to perceive and respond to the experimental stimuli [2].

Santi M. Spampinato (ed.), Opioid Receptors: Methods and Protocols, Methods in Molecular Biology, vol. 1230, DOI 10.1007/978-1-4939-1708-2_27, © Springer Science+Business Media New York 2015

313

314 Francesca R. D’Amato

To complicate the study of developing rodents, time to reach maturation of specific reflexes or capacity to respond to some sen- sory stimulations does not simply depend on the age (days) of pups. Rodent inbred strains can differ among each other for their developmental time schedules, and each strain can present a differ- ent mature/adult behavioral and neurobiological maturation tim- ing [3]. In addition, also environmental factors affecting the dimension of the litter, the amount of maternal care received by pups, the environmental temperature, and the quality of food eaten by the mother can modulate the maturation trajectory of each pup. These factors complicate the experimental design when comparing pups belonging from different strains, but also pups belonging from different litters.

Studies on developing pups explore (1) the development of a neurobiological system or of a behavioral response in the individual, or (2) the effects of a gene, a treatment, a drug at different ages during development, between individuals. In both cases the experi- mental subjects should be as much as possible similar in terms of litter size and sex ratio, body weight at birth and body weight dur- ing development, maternal care received, and so on. Taking into consideration all these factors, it is possible to set up a series of behavioral tests that allow measuring the neurobiological and behavioral development of pups. Exposing pups to a specific condi- tion, such as separation from the mother or low temperature, it is possible to evaluate the response of a complex adaptive system evolved to promote survival in these immature animals. Many neu- robiological agents are probably recruited, according to the matu- rational stage of the pup, in response to threatening situations to promote adaptive responses. Due to the relevance of the these experimental situations and the behavioral systems involved, prob- ably none of the behavioral responses measured in the tests subse- quently described can be specifically attributed to a single agent only. This sort of general reasoning can be partially eluded when knocking-out techniques are adopted to silence specific genes, allowing identifying their role in the presence of a negative response.

The opioid system is already mature at birth, even if receptor expression and binding is known to undergo considerable postna- tal reorganization, reaching the adult distribution and functioning by postnatal day 21 [4, 5]. To explore these changes, behavioral tests are conducted on pups also during the first postnatal days, using pharmacological and environmental treatments aimed at interact with this system [6]. The opioid system is mainly impli- cated in response to stress, reward processes, and pain sensitivity [7]. According to these functions, three behavioral tests will be described in this paper that can measure the activity of these behav- ioral systems in developing rodents. Response to stress is measured in pups by the amount (and characteristics) of ultrasonic calls emit- ted during isolation, the functionality of reward process can be

2 Materials

2.1 Animals

evaluated by comparing the behavioral responses, ultrasonic calls or homing behavior, displayed in differently aversive/attractive conditions, and pain sensitivity is measured by tail withdrawal in response to high/low temperature. Validity of these tests derives from pharmacological treatments and from the use of genetically modified mice. Studies demonstrating the involvement of the opi- oid system in the behavioral tests below described, will be reported together with the description of the experimental procedures.

The subjects of the experiments are developing pups and for this reason is important to control as much as possible, not only post- natal factors as previously stated, but also prenatal factors that can affect body weight gain and maturation timing and could represent confounding factors in the experimental design. For this reason is important to avoid stressing females during pregnancy: animal transfer as well as changes in the animal house conditions may rep- resent risk factors for pregnancy. Pregnant and lactating females are less responsive to stressors, but long-term effects have been clearly demonstrated in the adult offspring [8, 9]. It is suggested to start the experiments from mating, in order to maintain stable environ- mental conditions, control pregnancy duration, litter size at birth, and homogeneity (both in terms of genetic relatedness and shared environment) both within and between experimental groups.

In addition, precautions should be considered to control for the developmental stage of the experimental and control animals as pups change in terms of new sensory-motor capabilities from 1 day to another one [10, 11]. Removal of the mate is suggested to pre- vent mating during the post partum estrous that would result in mothers being concomitantly lactating and pregnant. This phe- nomenon would affect the mother’s physiology and, consequently, the amount of energies and care given to the newborns.

Usually one male is housed with two or three females and left undisturbed for 15 days. Females are isolated when visibly preg- nant and according to a consistent increase in body weight (more than 30 %): the day of birth is considered as postnatal day 0 (PND0). It is better not to disturb females immediately after par- tum and inspect cages for live pups after a few hours. The number of pups, as well as the sex/ratio at birth, cannot be a priori con- trolled. These variables are affected by rodent strain, age and parity of the dam, some characteristic of the environment (e.g. the qual- ity of food eaten) and, if it is the case, the experimental treatment the animals have been exposed. One possibility is to include in the experimental groups only pups belonging from litters within a defined litter size range (e.g. between 4 and 8). Otherwise, it is

2.2 Mating and Control
of Litter Effect

Opioid-Mediated Behaviors in Mouse Pups 315

316 Francesca R. D’Amato

2.3 Response to Stress

2.3.1 Preparation of the Experimental Subjects

possible to modify both the litter size and the sex/ratio within (24–48) hours from birth with no disruption of maternal behavior. Litters can be adjusted to a fixed number by reducing or increasing the number of pups to obtain homogeneous sample of litters. Of course increasing the number of pups in a litter means that pups from different dams are pulled together, sometimes given in adop- tion to a third mother. However, even if dams display adequate maternal care towards adoptive pups, recent evidences are suggest- ing that this cross fostering procedure has indeed long term effects in the pups [12–14]. Moreover, giving a mother a number of pups larger than the number of fetuses she carried during pregnancy, may conflict with her lactating expected requests and result in pups’ behavioral and/or nutritional deficits.

Finally it should be considered that because of genetic and environmental homogeneity within each single litter, experimental groups should contain no more than one or two pups per litter.

Before staring testing pups, it is necessary that mother and pups habituate to the novel experimental environment, for about 1 h (see Note 1). The mother should be removed from the home cage, separated from pups and housed in a clean cage, far from pups. The home cage with pups huddled in the nest should be kept warm to prevent pups’ cooling, by placing it on an hotplate, set at a tem- perature around 35 °C, that will maintain body temperature of pups not able to thermo-regulate, as a the mother should do. It is possible to start testing pups 5 min after mother’s removal. This procedure should provide all litters a similar basic physiological, nutritional, and behavioral background. It is however also impor- tant that pups do not remain without the mother for a long time (to avoid starvation) and to minimize difference in separation time from the mother among pups. When in the presence of large lit- ters, two experimenters, working simultaneously on different pups are necessary to reduce within and between litters’ differences in behavior due to time of separation from the mother.

The suggested time interval of habituation to the experimen- tal room, as well as time before starting testing pups after moth- er’s removal from the home cage are indicative; what is important is to maintain them as constant as possible to reduce inter-litter variability.

Pups emit ultrasonic vocalizations (USVs) during isolation and sepa- ration from the mother, mainly during the first 2 weeks of life, and these calls are used as an index of their emotional condition. Mothers easily locate isolated pups, also thanks to USVs, and retrieve them rapidly into the nest protected area. These calls present a develop- mental trend that usually reached the peak around postnatal day 6–8, depending on the mouse strain. As pups mature and start to move and orient by themselves, the number of USVs emitted during

2.3.2 Ultrasonic Vocalizations

2.3.3 Apparatus

2.3.4 Experimental Conditions

isolation decreases and these vocalizations, still emitted in adult life, will be associated to other social contexts [15, 16]. There are several equipments (hardware and software) designed to record, measure, and analyze ultrasonic vocalizations (e.g. Avisoft Bioacoustics and Binary Acoustic Technology).

These vocalizations, being considered correlated to emotionality in pups, are affected by benzodiazepines, as well as by serotoninergic, dopaminergic, and other neurobiological agents (e.g. [17, 18]).

The involvement of the opioid system, and in particular of the mu component of the system has been demonstrated by studies using agonist and antagonist agents [19–23]: mu-opioid agonists decrease the number of calls, antagonists tend to increase it or have no effects (discordant results). In addition, silencing the mu-opioid system through knocking out techniques, results in a reduction of calls emitted during separation from the mother (but not in other contexts). This reduction in USVs during separation from the mother stresses the role of the opioid system in the reward circuit behind infant-mother bond (see Note 2).

The microphone should be sensitive to frequencies from 15 to 180 kHz because different mouse strains and different environ- mental conditions can affect the frequency of calls emission. The microphone should be connected to a personal computer, where acoustic data will be recorded and stored for subsequent analysis, according to the software used.

Pup’s age, strain, body and environmental temperature, odors and tactile stimulation affect ultrasonic emission. All conditions that can be considered as dangerous for pup’s survival, elicit ultrasound emission that should stimulate maternal care and retrieving of the pup in the nest. These are some experimental conditions that elicit USVs and quantitative/qualitative differences in vocalizations sug- gest differences in intensity of emotional distress. Mouse pups emit calls when isolated:

. (a) in the absence of familiar cues (clean bedding or empty container);

. (b) in the presence of unfamiliar possibly dangerous olfactory cues (unknown con-specifics, predators);

. (c) under unpleasant temperature;

. (d) while exposed to rough tactile stimulation.
The choice of the appropriate control condition depends on

the ongoing project. It is suggested to use, as control condition, USVs emitted during exposure to home cage bedding during isolation.

Opioid-Mediated Behaviors in Mouse Pups 317

318 Francesca R. D’Amato

2.4 Functionality of the Reward System

2.4.1 Preparation of the Experimental Subjects

2.4.2 Homing Behavior

Before staring testing pups, it is necessary that mother and pups have been acclimatized to the novel experimental environment for about 1 h. Then, the mother is removed from the home cage, sepa- rated from pups, and housed in a clean cage, far from pups for about 20 min. It is suggested to test developing pups when they are able to move around, but are still blind, to privilege environ- mental information coming from olfactory cues. Pups from post- natal day 8 and 14 usually have these characteristics.

Blind pups tend to orient in an open area on the basis of olfactory cues. When in an open space, 10 day-old pups are able to move and tend to approach familiar/positive stimuli, rather than unknown or aversive ones. This capability can be used to evaluate the reward system functioning, once the integrity of the olfactory system has been ascertained. In fact, it has been demonstrated that pups associ- ate mother/nest cues to the rewarding value of receiving maternal care (suckling, being warming, licked, and groomed). The involve- ment of the opioid system on this behavioral response has been demonstrated treating pups with agonist and antagonists (generally decreasing and increasing the response, respectively) [24, 25]. The malfunctioning of the reward system in mu-KO mouse pups results in reduced approach to the home-cage bedding and loss of capac- ity to recognize their own mother’ olfactory cues [22, 23]. Environmental factors and drugs affecting locomotion and modu- lating learning processes can modify and disrupt the homing response [26–28]. This spontaneous response of approach to famil- iar and avoidance of unfamiliar or aversive cues change as pups grow up and they become more explorative and interested in novelty.

A Plexiglas 35×5×10 h cm apparatus is used. The central 5×5 area is separated from the two lateral arms by sliding doors. The two lateral arms (15×5 cm) should have each one its own top to confer each arm its own olfactory environment. To shorten the experimental testing of the entire litter, more pups can be tested together, using different apparatuses at the same time. A video camera, situated above the apparatuses, is necessary to record the behavior of pups into the mazes.

The floors of the two arms of the apparatus are covered with bed- dings characterized by different olfactory cues to evaluate the capabil- ity of pups to orient and prefer one of the two arms. In a first situation, preference for home cage versus clean bedding can be measured allowing to evaluate olfactory capability. A second, more difficult task could consist in evaluating preference for own home cage versus home cage of an alien mother-infant cage. In this case the capability of recognizing and prefer own (mother-littermates) cues informs about the rewarding system functioning whether the pup is able to associate own home cage stimuli and the pleasurable state deriving from receiving maternal care (food, warmth, tactile stimulation, etc).

2.4.3 Apparatus

2.4.4 Experimental Conditions

2.5 Pain Sensitivity

2.5.1 Preparation of the Experimental Subjects

2.5.2 Tail Withdrawal

Before staring testing pups, it is necessary that mother and pups have been acclimatized to the novel experimental environment for about 1 h. Then, the mother is removed from the home cage, separated from pups, and housed in a clean cage. The home cage with pups huddled in the nest should be warmed to prevent pups’ cooling, by placing it on a hotplate, set at a temperature around 35 °C, that will maintain body temperature of pups not able to thermo-regulate. Pups can be tested about 5–10 min from mother removal.

The tail-flick test is commonly used to examine the behavioral responses to noninjurious noxious thermal stimuli in mice and rats and the implication of the opioid system in the modulation of this response during development has been demonstrated [6, 29]. Adaptations of this simple response of avoidance of a thermal potentially noxious stimulation to mouse pups are presented. In fact, the common available commercial apparatuses available for adult mice cannot be used for mouse pups. Two different proce- dures to evaluate pain in pups from postnatal day 8 (PND8) to PND20 are subsequently described [30].

According to the age of the pups, a hot place test apparatus (PND 8-14) or a beaker containing hot water (PND16-20) is used.

Temperature of the hot plate and water in the beaker is set and maintained at 48–49 °C. Since tail-flick is affected by tail skin tem- perature, careful attention should be paid to maintain the ambient temperature at 22–23 °C.

Ultrasonic calls are usually measure during 2–5 min of isolation. Each pup is individually placed into a small container (e.g. a bea- ker, a glass) situated 1–5 cm under an ultrasound microphone. The experimental pup and the microphone should be in a sound proof- attenuated environment to avoid interference with other sounds. Attention should be paid to keep other animals, as well as elec- tronic apparatuses such as computers far from the microphone.

After recording USVs, the subject is returned to the home cage and, the next littermate is tested. Between each trial, the con- tainer is washed and dried.

The simpler analysis embraces temporal parameters including the latency to call and the mean and total number and duration of vocalizations. More sophisticated analysis can include parameters, such as peak frequency and peak amplitude, which derive from the average spectrum of the entire element. Peak amplitude is defined

. 2.5.3 Apparatuses

. 2.5.4 Experimental

Conditions

3 Methods

3.1 Evaluation of the Response to Stress

3.1.1 Behavioral Testing Procedure

3.1.2 USVs Analysis

Opioid-Mediated Behaviors in Mouse Pups 319

320

Francesca R. D’Amato

3.2
Functionality of the Reward System

3.2.1 Behavioral Testing Procedure and Parameters Considered for the Analysis

3.3 Evaluation of Pain Sensitivity

3.3.1 Behavioral Testing Procedure and Parameters Considered for the Analysis

4 Notes

as the highest energy within the spectrum of the element, while the peak frequency is defined as the frequency at the location of the peak amplitude. Peak amplitude and peak frequency at the start and the end of the calls can also be measured along with frequency modulation, which is the difference between the highest and the lowest peak frequency within the element. For detailed USVs description and interpretation of emotional and motivational back- grounds it is suggested to read specific papers (e.g. [31]).

The pup is introduced in the central part of the apparatus, while the two lateral doors are closed and arms covered with selected beddings. After 1 min, doors are simultaneously opened and the pup is free to move in the entire apparatus for 5 min. The entire session is video-recorded and number of entries and time spent in the different arms during the 5 min test are subsequently measured by an expert observer or by a video-tracking software.

The pups are placed on a piece of cardboard with the tail extend- ing beyond the cardboard’s edge. The tail is then lowered onto the 48 °C hot-plate by setting the cardboard support down next to the heated surface. Latency either to lift the tail or to curl it away from the heated surface is recorded. From PND16 onward, two-thirds of the tail is immersed in heated water (48–49 °C), with the mouse lightly restrained, and the latency to tail-flick is recorded. The cutoff time is 15 s. For both versions of tail-flick test, three separate withdrawal latency determinations (20-min interval) are averaged. Mean latency times are the parameters used for the statistical analysis.

1. It should be stressed that great attention should be paid to control environmental factors, to respect time for animals to habituate to the experimental context, to control postnatal fac- tors, inter-litter and intra-litter variability that can alter the behavioral response of mouse pups, having nothing to do with the opioid system functioning.

2. It should not be excluded, in theory, that the opioid system exerts other specific, not already detected, functions during development. No differentiation of the implications of the three components (mu, delta, and kappa) of the opioid sys- tem has been made, but the majority of the results and papers cited refers to the effects of mu—opioid agonists and antagonists.

Evaluation of

References

1. Weaver IC (2009) Shaping adult phenotypes through early life environments. Birth Defects Res C Embryo Today 87:314–326

2. Oddi D, Luchetti A, D’amato FR (2014) Impact of postnatal manipulations on offspring development in rodents. In: Tucci V (ed) Neuro-phenome. Cutting-edge approaches and technologies in neurobehavioral genetics. Wiley, New York

3. Roubertoux P, Semal C, Ragueneau S (1985) Early development in mice: II. Sensory motor behavior and genetic analysis. Physiol Behav 35:659–666

4. Rahman W, Dashwood MR, Fitzgerald M et al (1998) Postnatal development of multiple opi- oid receptors in the spinal cord and develop- ment of spinal morphine analgesia. Dev Brain Res 108:239–254

5. Beland B, Fitzgerald M (2001) Mu- and delta opioid receptors are downregulated in large primary sensory neurons during postnatal development in rats. Pain 90:143–150

6. Nandi R, Fitzgerald M (2005) Opioid analge- sia in the newborn. Eur J Pain 9:105–108

7. Kieffer BL, Gavériaux-Ruff C (2002) Exploring the opioid system by gene knockout. Prog Neurobiol 66:285–306

8. Windle RJ, Wood SA, Kershaw YM et al (2010) Reduced stress responsiveness in pregnancy: relationship with pattern of forebrain c-fos mRNA expression. Brain Res 1358:102–109

9. Weinstock M (2008) The long-term behav- ioural consequences of prenatal stress. Neurosci Biobehav Rev 32:1073–1086

10. Calamandrei G, Venerosi A, Branchi I et al (1999) Effects of prenatal AZT on mouse neu- robehavioral development and passive avoidance learning. Neurotoxicolog Teratol 21:29–40

11. Crawley JN (2007) What’s wrong with my mouse? Behavioral phenotyping of transgenic and knockout mice. Wiley, Hoboken, NJ

12. Barbazanges A, Vallée M, Mayo W et al (1996) Early and later adoptions have different long- term effects on male rat offspring. J Neurosci 16:7783–7790

13. Bartolomucci A, Gioiosa L, Chirieleison A et al (2004) Cross fostering in mice: behavioral and physiological carry-over effects in adulthood. Genes Brain Behav 3:115–122

14. D’Amato FR, Zanettini C, Lampis V et al (2011) Unstable maternal environment, sepa- ration anxiety, and heightened CO2 sensitivity induced by gene-by-environment interplay. PLoS One 6:e18637

15. Branchi I, Santucci D, Alleva E (2001) Ultrasonic vocalisation emitted by infant rodents: a tool for assessment of neurobehavioural development. Behav Brain Res 125:49–56

16. Ehret G (2005) Infant rodent ultrasounds – a gate to the understanding of sound communi- cation. Behav Genet 35:19–29

17. Winslow JT, Insel TR (1991) Infant rat separa- tion is a sensitive test for novel anxiolytics. Prog Neuropsychopharmacol Biol Psychiatry 15:745–757

18. Groenink L, Verdouw PM, van Oorschot R et al (2008) Models of anxiety: ultrasonic vocalizations of isolated rat pups. Curr Protoc Pharmacol 5:5.18. doi:10.1002/0471141755

19. Panksepp J, Siviy SM, Normansell LA (1985) Brain opioids and social emotion. In: Field T, Reite M (eds) The psychobiology of attach- ment and separation. Academic, New York, pp 3–49

20. Carden SE, Barr GA, Hofer MA (1991) Differential effects of specific opioid receptor agonists on rat pup isolation calls. Brain Res Dev Brain Res 62:17–22

21. Carden SE, Davachi L, Hofer MA (1994) U50,488 increases ultrasonic vocalizations in 3-, 10-, and 18-day-old rat pups in isolation and the home cage. Dev Psychobiol 27:65–83

22. Moles A, Kieffer B, D’Amato FR (2004) Deficit in attachment behavior in mice lacking the μ-opioid receptor gene. Science 304:1983–1986

23. Cinque C, Pondiki S, Oddi D et al (2012) Translational psychiatr y modeling socially anhedonic syndromes: genetic and pharmaco- logical manipulation of opioid neurotransmis- sion in mice. Transl Psychiatry. doi:10.1038/ tp.2012.83

24. Panksepp J, DeEskinazi FG (1980) Opiates and homing. J Comp Physiol Psychol 94:650–663

25. Najam N, Panksepp J (1989) Effect of chronic neonatal morphine and naloxone on sensorim- otor and social development of young rats. Pharmacol Biochem Behav 33:539–544

26. Ricceri L, Markina N, Valanzano A et al (2003) Developmental exposure to chlorpyrifos alters reactivity to environmental and social cues in adolescent mice. Toxicol Appl Pharmacol 191: 189–201

27. Honeycutt H, Alberts JR (2005) Housing pregnant mice (Mus musculus) in small groups facilitates the development of odor- based homing in offspring. J Comp Psychol 119:418–429

Opioid-Mediated Behaviors in Mouse Pups 321

322 Francesca R. D’Amato

28. Krüger HS, Hanganu-Opatz IL (2013) Neonatal cholinergic lesion alters the acoustic structure of infant rat vocalization but not the early cognitive development. Dev Psychobiol 55:294–308

29. Marsh DF, Hatch DJ, Fitzgerald M (1997) Opioid systems and the newborn. Brit J Anaesth 79:787–795

30. Luvisetto S, Marinelli S, Panasiti MS et al (2006) Pain sensitivity in mice lacking the Ca(v)2.1alpha1 subunit of P/Q-type Ca2+ channels. Neuroscience 142:823–832

31. Lahvis GP, Alleva E, Scattoni ML (2011) Translating mouse vocalizations: prosody and frequency modulation. Genes Brain Behav 10:4–16

Addiction……….10, 14, 88, 151, 156, 281–283, 289, 290, 305 Adenylyl cyclase (AC) …………………..155, 180–181, 188, 198 Alcohol dependence …………………………………………………….6 Anesthesia ………………… 90, 99, 158, 223, 227, 240, 247, 297 Antiserum against mu-opioid receptor…………………145–146 Arrestin ………………………………………….. 14, 17, 90, 115–127 AtT-20 cells………………………………………. 178–180, 182, 183 Autoradiography……………………………………………….169–175

Chronic constriction injury (CCI)………………. 156, 158, 159, 216, 217, 223, 224, 226, 227

Co-immunoprecipitation……………………………………130–136 Complementary DNA (cDNA) ………………………. 66, 68–70,

81, 84, 93, 107, 136, 137, 162, 268
Computational analysis ……………………………… 17, 18, 20, 25 Confocal microscopy ………………………………………………..235 Crystal ………………………………….. 13–18, 28, 30, 31, 139, 169 Cyclic adenosine monophosphate

(cAMP) …………………………………….. 5, 178, 188, 198

[D-Ala2, N-Me-Phe4, Gly5]-ol-enkephalin (DAMGO)………………………….5, 79, 81, 82, 85, 172, 174, 180, 184, 188–193, 199, 204, 206, 225

Deglycosylation ………………………………………… 146–147, 150 Delta-opioid receptor (DOR) ……………………….. 3, 4, 15, 16,

88–99, 107–110, 244, 287–291, 309 Deoxyribonucleic acid (DNA)

extraction …………………………………………………………. 48 methylation……………………………………………..39, 40, 46 purification ………………………………………………………. 49 sequencing …………………………………………..7, 9, 46, 109

DOR. See Delta-opioid receptor (DOR)
Dorsal root ganglia (DRG) …………………………….. 95, 97–99,

155–157, 159, 160, 163, 164, 188–193, 195,

198–204, 206, 207, 209, 210, 216
Dot blot………………………………………………………………40, 44 [D-Pen(2, 5)]-enkephalin (DPDPE) ……. 172, 188–193, 246 DRG. See Dorsal root ganglia (DRG)
DRG neuron culture ………………………….. 191–192, 200, 201 Drug-associated conditioned reinforcer………………..289, 305 Drug-associated cues…………………………………………288, 302 Drug-priming……………………………………. 288, 302–303, 305 Drug withdrawal ………………………….281, 282, 288, 291, 304 Dynorphin…………………………………………………… 4, 244, 246 Dysphoria…………………………………………………….4, 117, 281

Electrical stimulation…………………………………. 218, 220, 221 Electropherogram ………………………………………………………68 Electroretinogram (ERG) …………………………………. 246–249 ELISA. See Enzyme linked immunosorbent assay (ELISA)

[3H] beta-funaltrexamine ………………………………………….142 Beta-galactosidase ………………………………………….. 54, 60, 62 Binding pocket ………………………. 3, 16, 18, 21, 22, 24, 30, 31 Binding simulations………………………………………………19–24 Bioluminescent resonance energy transfer

(BRET) ……………………………. 14, 105–112, 115–127 Biotinylated secondary antibody…………………………………158 Biotynylation……………………………69, 70, 136, 146, 151, 158 Brain slices……………………………………………….. 171, 233–241 BRET. See Bioluminescent resonance energy transfer

(BRET)

Calcitonin gene-related peptide (CGRP)……………. 155, 157, 160, 163

Calcium imaging ………………………………………. 187–196, 198 cAMP. See Cyclic adenosine monophosphate (cAMP)
CCI. See Chronic constriction injury (CCI)
cDNA. See Complementary DNA (cDNA)

Cell harvesting …………………………………………………178, 179 Cell lysate ……………………………………………………….. 135, 139 Cell transfection …………………………55, 57–58, 118, 120–121 c-Fos immunoreactivity …………………………………………….288 CGRP. See Calcitonin gene-related peptide (CGRP) Chemiluminescence imaging……………………………………..144 Chinese hamster ovary (CHO) cell ……………………. 142, 143,

148, 149, 178–182, 184
ChIP. See Chromatin immunoprecipitation (ChIP)
CHO. See Chinese hamster ovary (CHO) cell
Chromatin ………………………………………….. 40, 41, 46, 47, 49 Chromatin immunoprecipitation (ChIP) ……… 41, 42, 46–50 Chromosome …………………………………………………….. 4, 6, 40

Santi M. Spampinato (ed.), Opioid Receptors: Methods and Protocols, Methods in Molecular Biology, vol. 1230, DOI 10.1007/978-1-4939-1708-2, © Springer Science+Business Media New York 2015

INDEX

323

324 OPIOID RECEPTORS: METHODS AND PROTOCOLS Index

Endoh,T…………………………………………………………………..79 Endomorphin………………………………………………………4, 174 Enhanced yellow fluorescent protein (EYFP) ………. 106–112 Enkephalin ………………………………………………… 88, 244, 246 Enzyme linked immunosorbent assay

(ELISA) ………….. 40–44, 49, 145, 255–256, 258, 260 Epigenetic ……………………………………………………..39–50, 53 ERG. See Electroretinogram (ERG)
Ethanol drinking ……………………………………………… 310, 311 EYFP. See Enhanced yellow fluorescent protein (EYFP)

Fibroblasts ……………………………………………………….273–277 FITC. See Fluorescein isothiocyanate (FITC)
FLAG-tagged kappa opioid receptor …………… 131, 133–135 Fluorescein isothiocyanate (FITC)……………………… 157, 160 Fluorescence energy transfer (FRET), 14

Fluorescence microscopy …………………………. 79–85, 94, 130, 132, 138, 139, 158, 235

FluxORTM assay………………………………………………..187–189 Forskolin………………………………………………….. 178–180, 190 FRET. See Fluorescence energy transfer (FRET)
Fusion protein …………………………….91, 92, 94, 95, 107–109,

111, 117, 136

Homing ………………………………………………………….. 315, 318 Human embryonic kidney cells (HEK293) …………….. 80, 81,

84, 91, 92, 108–110
Human mu-opioid receptor gene (OPRM1) …….. 3–6, 8–10,

53–63, 151

IBMX. See 3-Isobutyl-1-methylxanthin (IBMX) Immunoblot……………………………………………… 135, 137, 140 Immunohistochemical analysis……………………………155–164 Immunosuppression …………………………………………………254 Inflammatory pain…………………………………… 95, 97, 98, 223 Interleukin-10 (IL-10) ……………………………………………..260 Interleukin-1beta (IL-1beta) ………………………. 256, 258, 260 Intraocular pressure (IOP)…………………………………. 245, 246 3-Isobutyl-1-methylxanthin (IBMX)………….. 190, 194–196,

207–209

Kappa-opioid receptor (KOR)……………………….. 3, 4, 15, 16, 107–110, 115–127, 129–140, 170, 175, 216, 287, 288, 290, 291, 309

Keratinocytes……………………………………………………273–277 Knockin strategies ………………………………………………..90, 91 Knockout mice ……………….. 94, 143, 148, 157, 162–164, 309 KOR. See Kappa-opioid receptor (KOR)

Lectin IB4 ………………………………………………………………..97 Lipofectamine delivery system ……61–62, 125, 131, 135, 140 Lipopolysaccharide (LPS) …………………………. 255, 257, 259,

260, 263, 264, 266

Mass spectrometry (MS) …………………………………… 130, 134 Mechanical nociception……………………………………..229–231 Microarray analysis……………………………………………….69–70 Microelectrodes ………………………………………………..218–220 Microfluorimetry……………………………………………………..187 MicroRNA (miRNA)………………………………….. 5, 41, 53–54 MOR. See Mu-opioid receptor (MOR)

Morphine …… 5, 30, 40, 53, 67, 79–81, 88, 89, 151, 170, 190, 194–198, 207, 216, 225, 233, 236, 238, 239, 241, 245, 246, 253–260, 263–270, 283–285, 288–291, 304

Mouse
perfusion ………………………………………………………… 159 pup…………………………………………………………. 313–320

mRNA………………………………………… 5, 6, 10, 41, 53, 54, 59, 60, 65, 66, 68, 69, 74, 92, 93, 95, 97, 244, 266–269

Mu-opioid receptor (MOR)………………………3–6, 15, 16, 31, 40, 41, 46, 53–55, 79–85, 88, 89, 107, 141–152, 170, 174, 175, 233–241, 244, 254, 287–291, 309

Murine macrophage ………………………………………….253–260

Galactosidase …………………………………………………………….62 GDP. See Guanosine diphosphate (GDP)
Gene reporter assay ………………………………………………54, 55 Genomic signature …………………………………………………….67 Genotyping …………………………………………………………..8, 10 GFP. See Green fluorescent protein (GFP)

GIRKs. See G-protein-gated inwardly rectifying potassium channels (GIRKs)

Glycosylation……………………………………………………141–152 GPCR. See G-protein coupled receptor (GPCR)
G-protein coupled receptor (GPCR) …………………. 3, 13–16,

19–21, 24, 26, 31, 32, 87, 88, 90, 91, 105, 107, 116,

125, 141, 142, 163, 169, 170, 174, 177, 237, 243 G-protein-gated inwardly rectifying potassium channels

(GIRKs)………………………………… 178–183, 188–193,

197–200, 202, 204, 206, 210
Green fluorescent protein (GFP) …………80, 83, 91, 105–107 GST pull-down ………………………………………………..136–137 Guanosine diphosphate (GDP)…………………………..170–174 Guanosine triphosphate (GTP)…………………………..170–172

HEK293. See Human embryonic kidney cells (HEK293) Hemagglutinin-tagged mu-opioid receptor………………….142 High-throughput assay………………………………………177–185 High-throughput profiling…………………………………….65–75 Histone ……………………………………………………. 40, 41, 47, 49

Neuropathy……………………………………………………………..156 Neuroprotection ………………………………………………………246 Next generation sequencing (NGS) ……..7–10, 41, 66, 70–74 NGS. See Next generation sequencing (NGS)

Nociceptin receptor (NOR)…………………………………. 15, 107 Nociceptor …………………………………. 206, 215–217, 222, 225

Oligomerization ……………………………………… 17, 24, 25, 116 Opioid antagonist …………………………………………….. 241, 246 Opioid receptor …………………….. 3, 13, 39, 87, 105, 115, 148,

155, 169, 178, 187, 197, 215, 243, 273, 281, 296, 309 Opioid-seeking behavior ……………………………………281–291 Opioid self-administration ………………………………… 287, 304 Opioid-sensitive sensory fiber ……………………………………225 Opioid sensitivity ……………………………………… 217, 223–225 OPRM1. See Human mu-opioid receptor gene (OPRM1) Overexpression …………………………………………………..93, 129

PCR. See Polymerase chain reaction (PCR) Peptide…………………………………… 4, 5, 15, 16, 21, 89, 92, 94,

131, 133, 134, 142, 145–146, 148, 149, 152,

155–157, 162–164, 170, 216, 243–245
Peptide N-glycosidase F (PNGase F)…………………. 142, 146,

147, 150, 151
Phenotype …………………………………………………… 4, 300, 313 Plantar aesthesiometer……………………………………….229–231 Plasmid……………………………………………… 41, 54–57, 59–62,

107, 108, 111, 117–121, 125, 126, 131, 193, 194, 198,

200, 202, 206
PNGase F. See Peptide N-glycosidase F (PNGase F) Polymerase chain reaction (PCR)……………….. 40, 43, 46–49,

70, 74, 260, 266, 268
Postnatal development………………………… 314–316, 318–320 Potassium imaging…………………187, 189–190, 192–193, 196 PPIs. See Protein-protein interactions (PPIs)
Protein electro-transfer……………………………………………..135 Protein kinase A (PKA) ……………………………………………188 Protein-protein interactions (PPIs) ……………………. 105, 111,

115, 116, 130
Pull-down assay……………………………. 45, 130, 132, 136–137

Rat brain…………………………………………… 143, 146, 169, 172 Real-time fluorescence-based assay …………………………….178 Real-time imaging………………………………………………..79–85 Receptor

endocytosis ………………………………………………………. 79 heterodimer ………………………………………………4, 14, 24 heteromer …………………………………………………. 14, 107

OPIOID RECEPTORS: METHODS AND PROTOCOLS 325 Index

homomer ……………………………………………………….. 107 recycling…………………………………………………………… 81 trafficking …………………………………….80, 87–90, 97–99

Relapse prevention …………………………………………………..291 Renilla luciferase……………………………………. 53–63, 105, 117 Response to stress ……………………….. 314, 316–317, 319–320 Restriction enzymes ………………………………………..40, 45–47 Retina ……………………………………………………… 244–246, 249 Reverse transcription (RT)……….70, 147, 255, 258, 266–268 Ribonucleic acid (RNA)

extraction ………………………………………………… 265–267 isolation …………………………………………………………… 68 quantification…………………………………………… 266, 267 stabilization ……………………………………………………… 73

RNA complementary (cRNA) ………………………………. 68, 69

Sampling simulations ……………………………………………27–28 Sciatic nerve ……………………………………… 156–160, 216, 223 Scotopic Electroretinogram (ERG)…………………………….249 SDS-PAGE. See Sodium dodecyl sulfate polyacrylamide gel

electrophoresis (SDS-PAGE)
Selection marker ………………………………………………………224 [35S]GTPγS……………………………………………………..169–175 Single nucleotide polymorphism (SNP)…………………. 4–6, 8,

10, 151
Skin-saphenous nerve preparation………………………………215 SNP. See Single nucleotide polymorphism (SNP)
Sodium dodecyl sulfate polyacrylamide gel

electrophoresis (SDS-PAGE)………………….130–139,

142–144, 147, 150–152
Stress …………………………………… 88, 170, 290, 300, 303–305,

314, 316–317, 319–320

Tail flick assay…………………………………………………..319, 320 TaqMan probes ………………………………………………..266, 268 Template preparation…………………………………………….70, 74 Time lapse microscopy …………………………. 80, 235, 237, 238 TIRF. See Total internal reflection fluorescence (TIRF) Tolerance …………………………………. 79, 80, 98–100, 107, 233,

234, 243, 245, 281, 289–291
Toll like receptor 4…………………………………………….263–270 Total internal reflection fluorescence (TIRF), 79–85 Transcardial perfusion …………………………………………91, 100 Transcriptome ………………………………………….. 65, 66, 70–72 Transient receptor potential vanilloid 1

(TRPV1) ………………………………… 97, 188, 190–191,

193–195, 197, 198, 200–201, 206–210
Tumor necrosis factor-alpha (TNF-α) ………………… 246, 260

Ultrasonic vocalization (USV )………. 298, 316–317, 319–320

326 OPIOID RECEPTORS: METHODS AND PROTOCOLS Index

von Frey hairs …………………………..90, 98, 218, 221, 223–225

Wound healing ……………………………………………….. 245, 273, 274, 276

Wound scratch assay…………………………………………………273

Zamboni’s fixative …………………………………….. 157, 160, 162 Zinc release………………………………………………………233–241

Wallerian degeneration……………………………………………..227 Wheat germ agglutinin (WGA)-affinity

purification ……………………………………………146, 152