Glutamate Receptors: Methods and Protocols [1st ed.] 978-1-4939-9076-4, 978-1-4939-9077-1

Table of contents :
Front Matter ....Pages i-x
Front Matter ....Pages 1-1
Single-Molecule FRET Methods to Study Glutamate Receptors (Douglas B. Litwin, Ryan J. Durham, Vasanthi Jayaraman)....Pages 3-16
Ultrastructural Observation of Glutamatergic Synapses by Focused Ion Beam Scanning Electron Microscopy (FIB/SEM) (Ai Takahashi-Nakazato, Laxmi Kumar Parajuli, Hirohide Iwasaki, Shinji Tanaka, Shigeo Okabe)....Pages 17-27
Considerations for Imaging and Analyzing Neural Structures by STED Microscopy (Martin O. Lenz, Jan Tønnesen)....Pages 29-46
Cell-Based Enzyme-Linked Immunosorbent Assay (Cell-ELISA) Analysis of Native and Recombinant Glutamate Receptors (Elek Molnár)....Pages 47-54
Front Matter ....Pages 55-55
Preparation of Organotypic Slice Cultures for the Study of Glutamate Receptor Function (Andres Barria)....Pages 57-64
Glutamate Receptor Probing with Rapid Application and Solution Exchange (RASE) (Nathanael O’Neill, Sergiy Sylantyev)....Pages 65-78
Electrophysiological Investigation of Metabotropic Glutamate Receptor-Dependent Metaplasticity in the Hippocampus (Regina U. Hegemann, Wickliffe C. Abraham)....Pages 79-91
Induction of Metabotropic Glutamate Receptor-Mediated Long-Term Depression in the Hippocampal Schaffer Collateral Pathway of Aging Rats (Kirstan E. Gimse, Kenneth O’Riordan, Corinna Burger)....Pages 93-105
Whole-Cell Patch-Clamp Electrophysiology to Study Ionotropic Glutamatergic Receptors and Their Roles in Addiction (Jonna M. Leyrer-Jackson, M. Foster Olive, Cassandra D. Gipson)....Pages 107-135
Front Matter ....Pages 137-137
Gene Expression Analysis by Multiplex Single-Cell RT-PCR (Ludovic Tricoire, Bruno Cauli, Bertrand Lambolez)....Pages 139-154
Metabolomics Analysis of Glutamate Receptor Function (Nataliya E. Chorna, Anatoliy P. Chornyy)....Pages 155-165
Locus-Specific DNA Methylation Assays to Study Glutamate Receptor Regulation (Jordan A. Brown, J. David Sweatt, Garrett A. Kaas)....Pages 167-188
Preparation of Synaptoneurosomes for the Study of Glutamate Receptor Function (Cara J. Westmark, Pamela R. Westmark)....Pages 189-197
Front Matter ....Pages 199-199
Fractionation of Subcellular Compartments from Human Brain Tissue (Toni M. Mueller, Pitna Kim, James H. Meador-Woodruff)....Pages 201-223
Glutamate Receptor Antibodies in Autoimmune Central Nervous System Disease: Basic Mechanisms, Clinical Features, and Antibody Detection (William J. Scotton, Abid Karim, Saiju Jacob)....Pages 225-255
Back Matter ....Pages 257-260

Citation preview

Methods in Molecular Biology 1941

Corinna Burger Margaret Jo Velardo Editors

Glutamate Receptors Methods and Protocols

METHODS

IN

MOLECULAR BIOLOGY

Series Editor John M. Walker School of Life and Medical Sciences University of Hertfordshire Hatfield, Hertfordshire, AL10 9AB, UK

For further volumes: http://www.springer.com/series/7651

Glutamate Receptors Methods and Protocols

Edited by

Corinna Burger Department of Neurology, University of Wisconsin, Madison, WI, USA

Margaret Jo Velardo ANSER, Homeland Security Studies and Analysis Institute, Falls Church, VA, USA

Editors Corinna Burger Department of Neurology University of Wisconsin Madison, WI, USA

Margaret Jo Velardo ANSER Homeland Security Studies and Analysis Institute Falls Church, VA, USA

ISSN 1064-3745 ISSN 1940-6029 (electronic) Methods in Molecular Biology ISBN 978-1-4939-9076-4 ISBN 978-1-4939-9077-1 (eBook) https://doi.org/10.1007/978-1-4939-9077-1 Library of Congress Control Number: 2018967452 © Springer Science+Business Media, LLC, part of Springer Nature 2019 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. 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. The publisher, the authors, and the editors are safe to assume that the advice and information in this book are believed to be true and accurate at the date of publication. Neither the publisher nor the authors or the editors give a warranty, express or implied, with respect to the material contained herein or for any errors or omissions that may have been made. The publisher remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. This Humana Press imprint is published by the registered company Springer Science+Business Media, LLC, part of Springer Nature. The registered company address is: 233 Spring Street, New York, NY 10013, U.S.A.

Preface The discovery of glutamate receptors (GluRs) in neurons and glia in the central nervous system and identification of the roles of ionotropic glutamate receptors in mediating fast excitatory synaptic transmission, followed by studies implicating metabotropic glutamate receptors in metaplasticity, are some of the seminal neuroscience findings of the twentieth century. And, as their structure and function has been elucidated, this phenomenal suite of receptor families has only grown in scientific importance and significance. Over the past four decades, a tremendous body of work has provided the details of how glutamate receptors regulate a broad spectrum of critical physiological, developmental, and pathological processes in the brain, spinal cord, retina, and peripheral nervous system. But as the glutamate receptor biology field has matured, new glutamate receptor research challenges have arisen. Recent and exciting work tells us that glutamate receptors are much more than the fundamental engine of excitatory neurotransmission in the nervous system. They are also key players in a wide array of non-nervous system biological processes across diverse species. Each day brings new work that demonstrates their novel involvement in processes as varied as human tumor progression, mammalian immunity and autoimmunity, and GluR homologue signaling in plant growth and defense. In parallel with the increasing recognition of the importance of these receptors in a plethora of biological processes, there has arisen an increasing need and a demand for reliable and/or cutting-edge methods to probe their structure, function, and regulation in a wide variety of experimental paradigms or settings. New technologies have been developed to help us elucidate the rules and roles of glutamate receptor diversity, so we can develop new therapeutic strategies to mitigate dysfunctional glutamatergic transmission in many biological systems and species. In this methods volume, we will add our efforts to those of others in addressing the need for providing a useful and detailed resource for glutamate receptor research methodologies. We have designed the contents so that graduate students, postdoctoral fellows, and seasoned researchers in neuroscience, pharmacology, biochemistry, molecular and structural biology can gain an understanding of the issues surrounding well-established glutamate receptor techniques and also learn new and cutting-edge techniques. To that end, we have attempted to clearly present the appropriate uses, strengths, and pitfalls of these techniques in a way that is useful to both novices and experts. Further, in the face of the growing use of core facilities, we believe that even if researchers are outsourcing techniques, they can benefit greatly from theoretical background information and the technical details of the methods themselves. But most of all, it is our sincere hope that these methods will help researchers develop compelling basic science and clinical data that increase understanding of these important receptor families and catalyze exciting and timely breakthroughs in therapeutic strategies. Madison, WI, USA Falls Church, VA, USA

Corinna Burger Margaret Jo Velardo

v

Contents Preface . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Contributors. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

PART I

SYNAPTIC ULTRASTRUCTURE/RECEPTOR DYNAMICS/RECEPTOR INTERACTIONS AND TRAFFICKING

1 Single-Molecule FRET Methods to Study Glutamate Receptors . . . . . . . . . . . . . . Douglas B. Litwin, Ryan J. Durham, and Vasanthi Jayaraman 2 Ultrastructural Observation of Glutamatergic Synapses by Focused Ion Beam Scanning Electron Microscopy (FIB/SEM) . . . . . . . . . . . . . . . . . . . . . . . Ai Takahashi-Nakazato, Laxmi Kumar Parajuli, Hirohide Iwasaki, Shinji Tanaka, and Shigeo Okabe 3 Considerations for Imaging and Analyzing Neural Structures by STED Microscopy. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Martin O. Lenz and Jan Tønnesen 4 Cell-Based Enzyme-Linked Immunosorbent Assay (Cell-ELISA) Analysis of Native and Recombinant Glutamate Receptors . . . . . . . . . . . . . . . . . . . Elek Molna´r

PART II

v ix

3

17

29

47

FUNCTION/CELLULAR PLASTICITY

5 Preparation of Organotypic Slice Cultures for the Study of Glutamate Receptor Function . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 57 Andres Barria 6 Glutamate Receptor Probing with Rapid Application and Solution Exchange (RASE) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 65 Nathanael O’Neill and Sergiy Sylantyev 7 Electrophysiological Investigation of Metabotropic Glutamate Receptor-Dependent Metaplasticity in the Hippocampus . . . . . . . . . . . . . . . . . . . . 79 Regina U. Hegemann and Wickliffe C. Abraham 8 Induction of Metabotropic Glutamate Receptor-Mediated Long-Term Depression in the Hippocampal Schaffer Collateral Pathway of Aging Rats. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 93 Kirstan E. Gimse, Kenneth O’Riordan, and Corinna Burger 9 Whole-Cell Patch-Clamp Electrophysiology to Study Ionotropic Glutamatergic Receptors and Their Roles in Addiction . . . . . . . . . . . . 107 Jonna M. Leyrer-Jackson, M. Foster Olive, and Cassandra D. Gipson

PART III 10

RECEPTOR GENE REGULATION/EPIGENETICS

Gene Expression Analysis by Multiplex Single-Cell RT-PCR . . . . . . . . . . . . . . . . . 139 Ludovic Tricoire, Bruno Cauli, and Bertrand Lambolez

vii

viii

11 12

13

Contents

Metabolomics Analysis of Glutamate Receptor Function. . . . . . . . . . . . . . . . . . . . . 155 Nataliya E. Chorna and Anatoliy P. Chornyy Locus-Specific DNA Methylation Assays to Study Glutamate Receptor Regulation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 167 Jordan A. Brown, J. David Sweatt, and Garrett A. Kaas Preparation of Synaptoneurosomes for the Study of Glutamate Receptor Function . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 189 Cara J. Westmark and Pamela R. Westmark

PART IV 14

15

CLINICAL APPLICATIONS

Fractionation of Subcellular Compartments from Human Brain Tissue . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 201 Toni M. Mueller, Pitna Kim, and James H. Meador-Woodruff Glutamate Receptor Antibodies in Autoimmune Central Nervous System Disease: Basic Mechanisms, Clinical Features, and Antibody Detection . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 225 William J. Scotton, Abid Karim, and Saiju Jacob

Index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

257

Contributors WICKLIFFE C. ABRAHAM  Department of Psychology, Brain Health Research Centre and Brain Research New Zealand, University of Otago, Dunedin, New Zealand ANDRES BARRIA  Department of Physiology and Biophysics, University of Washington School of Medicine, Seattle, WA, USA JORDAN A. BROWN  Department of Pharmacology, Vanderbilt University, Nashville, TN, USA CORINNA BURGER  Cellular and Molecular Pathology Graduate Program, University of Wisconsin-Madison, Madison, WI, USA; Department of Neurology, University of Wisconsin-Madison, Madison, WI, USA BRUNO CAULI  Sorbonne Universite´, UPMC Univ Paris 06 UM119, Centre National de la Recherche Scientifique (CNRS) UMR8246, Institut National de la Sante´ et de la Recherche Me´dicale (INSERM) UMRS1130, Neuroscience Paris Seine, Institut de Biologie Paris-Seine, Paris, France NATALIYA E. CHORNA  Department of Biochemistry, School of Medicine, University of Puerto Rico, San Juan, PR, USA; PR-INBRE Metabolomics Research Core, School of Medicine, University of Puerto Rico, San Juan, PR, USA ANATOLIY P. CHORNYY  High Performance Computing Facility, Central Administration, University of Puerto Rico, San Juan, PR, USA RYAN J. DURHAM  Department of Biochemistry and Molecular Biology, Center for Membrane Biology, University of Texas Health Science Center at Houston, Houston, TX, USA KIRSTAN E. GIMSE  Cellular and Molecular Pathology Graduate Program, University of Wisconsin-Madison, Madison, WI, USA CASSANDRA D. GIPSON  Department of Psychology, Arizona State University, Tempe, AZ, USA REGINA U. HEGEMANN  Department of Psychology, Brain Health Research Centre and Brain Research New Zealand, University of Otago, Dunedin, New Zealand HIROHIDE IWASAKI  Department of Cellular Neurobiology, Graduate School of Medicine, The University of Tokyo, Tokyo, Japan; CREST, JST, Tokyo, Japan SAIJU JACOB  Department of Neurology, University Hospitals Birmingham NHS Foundation Trust, Birmingham, UK; Department of Neuroimmunology, University of Birmingham, Birmingham, UK VASANTHI JAYARAMAN  Department of Biochemistry and Molecular Biology, Center for Membrane Biology, University of Texas Health Science Center at Houston, Houston, TX, USA GARRETT A. KAAS  Department of Pharmacology, Vanderbilt University, Nashville, TN, USA ABID KARIM  Department of Neuroimmunology, University of Birmingham, Birmingham, UK PITNA KIM  Department of Psychiatry and Behavioral Neurobiology, University of Alabama at Birmingham, Birmingham, AL, USA BERTRAND LAMBOLEZ  Sorbonne Universite´, UPMC Univ Paris 06 UM119, Centre National de la Recherche Scientifique (CNRS) UMR8246, Institut National de la Sante´ et

ix

x

Contributors

de la Recherche Me´dicale (INSERM) UMRS1130, Neuroscience Paris Seine, Institut de Biologie Paris-Seine, Paris, France MARTIN O. LENZ  Cambridge Advanced Imaging Centre, University of Cambridge, Cambridge, UK JONNA M. LEYRER-JACKSON  Department of Psychology, Arizona State University, Tempe, AZ, USA DOUGLAS B. LITWIN  Department of Biochemistry and Molecular Biology, Center for Membrane Biology, University of Texas Health Science Center at Houston, Houston, TX, USA JAMES H. MEADOR-WOODRUFF  Department of Psychiatry and Behavioral Neurobiology, University of Alabama at Birmingham, Birmingham, AL, USA ELEK MOLNA´R  School of Physiology, Pharmacology and Neuroscience, University of Bristol, Bristol, UK TONI M. MUELLER  Department of Psychiatry and Behavioral Neurobiology, University of Alabama at Birmingham, Birmingham, AL, USA SHIGEO OKABE  Department of Cellular Neurobiology, Graduate School of Medicine, The University of Tokyo, Tokyo, Japan; CREST, JST, Tokyo, Japan M. FOSTER OLIVE  Department of Psychology, Arizona State University, Tempe, AZ, USA NATHANAEL O’NEILL  CCBS, University of Edinburgh, Edinburgh, UK KENNETH O’RIORDAN  APC Microbiome Ireland, University College Cork, Cork, UK LAXMI KUMAR PARAJULI  Department of Cellular Neurobiology, Graduate School of Medicine, The University of Tokyo, Tokyo, Japan; CREST, JST, Tokyo, Japan WILLIAM J. SCOTTON  Department of Neurology, University Hospitals Birmingham NHS Foundation Trust, Birmingham, UK J. DAVID SWEATT  Department of Pharmacology, Vanderbilt University, Nashville, TN, USA SERGIY SYLANTYEV  CCBS, University of Edinburgh, Edinburgh, UK AI TAKAHASHI-NAKAZATO  Department of Cellular Neurobiology, Graduate School of Medicine, The University of Tokyo, Tokyo, Japan; CREST, JST, Tokyo, Japan SHINJI TANAKA  Department of Cellular Neurobiology, Graduate School of Medicine, The University of Tokyo, Tokyo, Japan; CREST, JST, Tokyo, Japan JAN TØNNESEN  Achucarro Basque Center for Neuroscience, Leioa, Spain; Department of Neurosciences, Faculty of Medicine and Dentistry, University of the Basque Country (UPV/EHU), Leioa, Spain LUDOVIC TRICOIRE  Sorbonne Universite´, UPMC Univ Paris 06 UM119, Centre National de la Recherche Scientifique (CNRS) UMR8246, Institut National de la Sante´ et de la Recherche Me´dicale (INSERM) UMRS1130, Neuroscience Paris Seine, Institut de Biologie Paris-Seine, Paris, France CARA J. WESTMARK  Department of Neurology, University of Wisconsin, Madison, WI, USA PAMELA R. WESTMARK  Department of Neurology, University of Wisconsin, Madison, WI, USA

Part I Synaptic Ultrastructure/Receptor Dynamics/Receptor Interactions and Trafficking

Chapter 1 Single-Molecule FRET Methods to Study Glutamate Receptors Douglas B. Litwin, Ryan J. Durham, and Vasanthi Jayaraman Abstract Single-molecule fluorescence energy transfer methods allow us to determine the complete structural landscape between the donor and acceptor fluorophores introduced on the protein of interest. This method is particularly attractive to study ion channel proteins as single-molecule current recordings have been used to study the function of these proteins for several decades. Here we describe the smFRET method used to study glutamate receptors. Key words Single-molecule FRET, Fluorescence, NMDA receptor, Glutamate receptor

1

Introduction The first biological X-ray crystallography experiments characterizing myoglobin and hemoglobin not only opened our eyes to the relation of function and form in biology but also revealed the utility of structural biology for studying the evolutionary origins of proteins and for designing useful drugs. X-ray crystallography and cryo-electron microscopy have since emerged as the front-runners in the field of structural biology and have provided new insight into all fields of biological science. These techniques will without question continue to lead the way in structural biology. They both, however, suffer from the limitation that they lack the ability to provide the complete structural landscape dynamics in terms of transitions between states and energetics of transitions between states. This limitation in current methods has created a niche for studying the structure of unresolvable molecules in addition to validating and studying the true dynamics of previously resolved x-ray and cryo-EM models [1, 2]. Fluorescent spectroscopy, in particular single-molecule fluorescent resonance energy transfer (smFRET), has emerged as the leading method for this role [3–5].

Corinna Burger and Margaret Jo Velardo (eds.), Glutamate Receptors: Methods and Protocols, Methods in Molecular Biology, vol. 1941, https://doi.org/10.1007/978-1-4939-9077-1_1, © Springer Science+Business Media, LLC, part of Springer Nature 2019

3

4

Douglas B. Litwin et al.

smFRET has become of particular interest to groups studying highly dynamic proteins and proteins composed of multiple domains acting in concert to perform mechanically complex motions. Initial efforts have focused on characterizing the structure of DNA/RNA molecules, DNA/RNA protein complexes, enzymes, signaling proteins, and ion channels [3–21]. Ionotropic glutamate receptors (iGluRs) are a family of ion channels that serve as a perfect subject for smFRET studies [1–5, 8, 21]. iGluRs are multimeric, have four domains with little functional overlap, operate with millisecond kinetics, and have previously resolved X-ray and cyro-EM models. iGluRs are tetrameric proteins with two extracellular domains, a transmembrane domain (TMD), and an intracellular C-terminal domain. iGluRs are divided by their selective agonists into AMPA, kainite, and NMDA subgroups, and some subgroups can express as homomers or heteromers depending on the subunits used. The extracellular amino terminal domain (NTD) and ligand binding domain (LBD) are of particular interest for smFRET studies due to the complex rearrangements proposed during the operation of the protein. In this chapter we will discuss the approach to and execution of smFRET to study the conformational dynamics of iGluRs. 1.1 Fo¨rster Resonance Energy Transfer

Fo¨rster (or fluorescent) resonance energy transfer (FRET) is a term describing the non-radiative transfer of energy from a donor fluorophore to an acceptor fluorophore. This relay of energy relies on the overlap of the donor and acceptor spectra, specifically that the emission of the donor overlaps with the excitation of the acceptor as seen in Fig. 1. This phenomenon is distance-dependent where the closer the fluorophore pairs are in space, the more likely they are to

Fig. 1 Spectral data for the excitation (dotted) and emission (solid) wavelengths of Alexa 555 (green) and Alexa 647 (red)

smFRET of Glutamate Receptors

5

FRET. The efficiency of energy transfer is related to the inter-dye distance (RDA) and the distance of half-maximal energy transfer for a given fluorophore set (R0), as seen in Eq. 1. The effective distance of FRET for fluorophores ranges between 10 and 100 A˚, making FRET ideal for measuring distances at the protein level. E¼

1 1 þ ðRDA =R0 Þ6

ð1Þ

When choosing a fluorophore pair for a FRET experiment, the most important factors to consider are the distance intended to be measured and the R0 of the fluorophore pair (see Note 1). The factors contributing to the R0 of a fluorophore pair are the orientation factor, k; the quantum yield of the donor, ΦD; the spectral overlap integral, J; and the refractive index of the media, n. The relationship between these values and the R0 is shown in Eq. 2. There are many detailed reviews on the physics of FRET if additional information is needed [22]. !1=6 8:785  105  k2  ΦD  J ð2Þ R0 ¼ n4 1.2 Single-Molecule Fo¨rster Resonance Energy Transfer

In traditional FRET experiments, a measurement is made on a large population of molecules, and the signal obtained is an ensemble of their collective behavior. This is useful when looking at large protein movements and proximity; however, these measurements provide an ensemble average, and the complete structural landscape along with transitions between the states cannot be resolved. To circumvent this, total internal reflection microscopy, 2-photon excitation microscopy, and confocal microscopy have been adopted to bring measurements to the single-molecule level of both immobilized and freely diffusing molecules as well as in living cells. These advantages of smFRET have vastly improved our ability to characterize sub-millisecond dynamics in protein, DNA, and RNA in addition to resolving protein subunit stoichiometry. However, collecting data with such detail does not come without its caveats. Data analysis has become the most difficult part of smFRET, and methods for collecting and analyzing data have had to evolve. The advent of multiparameter fluorescence detection (MFD), time-correlated single-photon counting (TCSPC), pulsed interleaved excitation (PIE), and fluorescence correlation spectroscopy (FCS) has vastly improved the potential quality of smFRET data, discussed further in the data analysis section. Instruments capable of collecting data using MFD, FCS, PIE, and TCSPC simultaneously are now available. The collective use of these techniques has greatly improved our ability to resolve the inherent complications of fluorophore photophysics and movements from true FRET signal [23]. In this chapter we will focus on the smFRET measurements using confocal imaging on slide-attached iGluRs.

6

Douglas B. Litwin et al.

1.3 Experimental Design

When designing a smFRET experiment studying iGluRs, the labeling strategy used will depend on the subunit being studied and the intended measurement. The main design step is site selection, and considerations here will be the subunit arrangement, distance to be measured, and fluorophore pairs. In a heteromeric iGluR such as the NMDA receptor, there will be two sets of the subunits arranged in a defined orientation. By taking this into account, it is possible to pick a surface-exposed amino acid position on either of the subunits used that is 0.5–1 the R0 of the fluorophore pair. In this case labeling with a sitespecific cysteine is recommended. For example, in Fig. 2 the heteromeric NMDA receptor is shown with transmembrane residues selected to measure the change in distance across the axis of the pore. Only the highlighted residues will be labeled and measured. If using a homomeric iGluR such as in homomeric AMPA receptors, choosing sites for study could be more complicated. This is the result of having the site of interest on all four subunits. In this case, it is required that the site of interest has one high FRET distance 0.5–1 the R0 and one distance ideally well above the R0 of the fluorophore pair for a given state. As seen in Fig. 3, the site chosen will have a primary FRET distance of 41 A˚ with an addi˚ across dimer pairs. This ensures that the tional distance of 64 A main FRET component, 41 A˚, is 0.5–1 the R0 and will be easily resolved from the longer component. In this chapter we will focus on the use of heteromeric iGluRs. When studying full-length iGluRs, immobilizing the protein on slide via antibody pulldown is preferred. Any antibody with high specificity will work; it is recommended, however, to introduce an affinity tag at the C-terminus to minimize the effect of antibody

Fig. 2 A structural model of the NMDA receptor showing phenylalanine 554 (red) as the fluorescent labeling site for measurements across the ion pore (PDB:5UP2) [24], side view (left) and top view (right)

smFRET of Glutamate Receptors

7

Fig. 3 A structural model of the AMPA receptor showing aspartic acid 473 (red) as the fluorescent labeling site for measurements between the LBDs (PDB:5VHZ) [25], side view (left) and top view (right)

binding on protein function. When using a new antibody, it is critical to run western blots to confirm specificity of the antibody (see Note 2). This chapter will focus on the use of immobilized molecules. 1.4 Cysteine Labeling

The best option for fluorophore labeling for smFRET is by introducing a cysteine at the site of interest. This allows use of many commonly available thiol-reactive fluorophores and retains high expression levels. This does however require that all the native non-disulfide-bonded cysteines be mutated to serine to provide a clean background. Once this is achieved, the expression and function of the cysteine-less construct must be verified (see Note 3). When approaching the design of an experiment, it is ideal that one choose sites that will show a change in distance and will also have one effective FRET distance. It is not always possible to achieve the ideal labeling strategy in every region of the protein, or the measurement may need to be made within the subunit itself. If either of these is the case, utilizing unnatural amino acids allows for additional site-specific labeling.

1.5 Unnatural Amino Acid Labeling

Unnatural amino acids are advantageous in that one retains the native coding sequence of the protein; however, additional plasmids must be maintained for use, and expression and function of the protein need to be verified. The principle behind utilizing unnatural amino acids is using a tRNA and aminoacyl-tRNA synthetase that have been artificially evolved to suppress the amber stop codon (TAG). The TAG can be inserted by mutagenesis into any position along the protein that, when co-transfected with

8

Douglas B. Litwin et al.

plasmids containing the tRNA and aminoacyl-tRNA synthetase, will be suppressed by the incorporation of the unnatural amino acid into the peptide chain (see Note 4) [26].

2

Materials

2.1 Protein Expression

1. Human embryonic kidney 293 cells (HEK 293) maintained at 37  C and 5% CO2. 2. Dulbecco’s Modified Eagle Media (DMEM) with high glucose, L-glutamine, 1 unit/mL penicillin, 1 ng/mL streptomycin, and 10% fetal bovine serum (FBS), without sodium pyruvate. 3. Phosphate-buffered saline (PBS). 4. Trypsin-EDTA: 0.25% trypsin, 1 mM EDTA. 5. Plasmid DNA at a concentration preferably >1 μg/μL. 6. Transfection buffer. 7. Transfection reagent. 8. If expressing NMDARs, (2R)-amino-5-phosphonovaleric acid (APV) and 5,7-dichlorokynurenic acid (DCKA) are needed in the media to inhibit receptor function.

2.2 Protein Sample Preparation

1. Extracellular buffer (ECB): 135 mM NaCl, 3 mM KCl, 2 mM CaCl2, 20 mM glucose, and 20 mM HEPES. 2. Alexa Fluor 555 C2 maleimide. 3. Alexa Fluor 647 C2 maleimide. 4. Solubilization buffer: 2 mM cholesteryl hydrogen succinate, 1% lauryl maltose neopentyl glycol, and protease inhibitor cocktail in 1 PBS.

2.3

Slide Preparation

1. Glass coverslips, 20 mm  20 mm. 2. Silicone spacer templates. 3. Methanol. 4. Liqui-Nox phosphate-free detergent. 5. Acetone. 6. Tl-1 solution: 4.3% NH4OH and 4.3% H2O2. 7. Compressed nitrogen. 8. Compressed oxygen. 9. Tissue section adhesive such as Vectabond Reagent. 10. Overnight polyethylene glycol (PEG) solution: 10 mM NaHCO3, 0.25% w/w NHS-PEG4-Biotin (biotinylated PEG), 50 mM mPEG-succinimidyl carbonate (mPEG-SC).

smFRET of Glutamate Receptors

9

11. Short-chain PEG solution: 25 mM MS(PEG)4 methyl-PEGNHS-ester reagent (short-chain PEG), 0.1 M NaHCO3. 12. Fluor-“friendly” hybridization system adhesive chambers. 13. Press fit tubing connectors. 14. 10 smFRET imaging buffer: 2 mM cholesteryl hydrogen succinate and 10 mM n-dodecyl-beta-maltoside detergent in 10 PBS. 15. Streptavidin solution: 0.2 mg/mL streptavidin in 1 smFRET imaging buffer. 16. Primary antibody solution: 10 nM anti-iGluR primary antibody in 1 PBS. 17. Secondary antibody solution: 10 nM anti-primary biotinylated secondary antibody in 1 PBS. 18. Bovine serum albumin (BSA) solution: 0.1 mg/mL BSA in 1 PBS. 19. ROXS solution: 3.3% glucose, 0.1 mg/mL pyranose oxidase, 0.01 mg/mL catalase, 1 mM ascorbic acid, and 1 mM methyl viologen in 1 smFRET imaging buffer. Ligands can be added to the ROXS solution, if desired.

3

Methods

3.1 Preparation of Slides for smFRET

1. Submerge the silicone templates in methanol in a beaker. Place the beaker in an ultrasonic bath and sonicate the templates for 30 min. Transfer the templates into a 50 mL tube containing clean methanol and vortex the templates. Finally, store the templates in clean methanol. 2. Place 20  20 mm glass slides in a slide holder, and submerge the holder in a beaker containing a solution of soapy water (use phosphate-free soap). Place the beaker in an ultrasonic bath and sonicate the slides for 10 min. Wash the slides with purified water three times. Repeat the 10-min sonication step with the slides in a beaker of acetone, and wash the slides three times with purified water. 3. Prepare the Tl-1 solution in a beaker and place the beaker in a 60–80  C water bath. Place the slide holder in the beaker and allow the Tl-1 solution to boil for 5 min. Wash the slides with purified water, dry them using a jet of nitrogen, and place them in a metal slide holder. 4. Place the slide holder with the slides in a plasma cleaner such as the Harrick Plasma PDC-32G Plasma Cleaner. Turn on the vacuum pump and allow the chamber pressure to drop below 200 mTorr. Flush the chamber with oxygen 3 times, allowing

10

Douglas B. Litwin et al.

the pressure to stabilize below 200 mTorr after each flush. Conduct plasma cleaning for 2 min, remove the vacuum, and extract the slides. 5. Mix 1 mL tissue section adhesive and 40 mL acetone, and submerge the slide holder with the slides in the adhesive/ acetone solution for 5 min. Wash the slides with purified water for 1 min, dry the slides with nitrogen, and place the slides on a clean delicate task wipe. These slides can be stored under vacuum for up to 5 days. Tissue section adhesive such as Vectabond only lasts 2 weeks after it has been opened and must be stored under nitrogen. The tissue section adhesive/acetone solution should only be used during the day on which it was made. 6. Remove the silicone templates from the methanol in which they are stored and dry them with nitrogen. Place one template on each slide, making an effort to center the template on the slide. Ensure that there are no bubbles trapped under the template. On the back of the slide, mark the location of the oblong area that the template encircles. 7. Make the overnight PEG solution as described above. Add 50 μL of the overnight PEG solution to the well in the middle of each slide. Place the slides on a damp delicate task wipe inside a petri dish. Allow the slides to incubate overnight in the dark at room temperature. 8. Next day wash slides with purified water, dry with gentle nitrogen flow, apply 50 μL short-chain PEG solution to each slide, and incubate at room temperature for 2–3 h. Remove the template from the slide and store the template in methanol. Wash the slide with purified water and dry with nitrogen. 9. Apply the elliptical chambers to the slides such that the chambers align with the treated areas of the slides. Ensure that there are no bubbles trapped underneath the chamber. Apply the tubing connectors to the ports on the chambers. Mark one port as the inlet port. 10. Make the streptavidin solution as described above. Add 36 μL to the inlet port of each slide (see Note 5). Ensure that the solution fills the chamber and there are no bubbles trapped in the chamber. Also ensure the solution exits the outlet port. Allow the slides to incubate at room temperature for 10 min. Wash the chambers by adding 60 μL of PBS three times to each chamber. 11. Apply 60 μL of secondary antibody solution to each chamber twice. Incubate 20 min at 4  C. Wash the chambers by adding 60 μL of PBS two times to each chamber. Apply 60 μL of primary antibody solution to each chamber twice. Incubate 20 min at 4  C. Wash the chambers by adding 60 μL of PBS

smFRET of Glutamate Receptors

11

two times to each chamber. Apply 60 μL of BSA solution to each chamber once. Incubate 20 min at 4  C. Wash the chambers by adding 60 μL of PBS two times to each chamber. Prepare the ROXS solution during these previous incubation steps. 12. Dilute the protein by a factor of 4–5 by adding 1 buffer (made from 10 buffer described above). Apply 90 μL of diluted protein sample (preparation described below) to each chamber twice. Incubate 20 min at 4  C. Wash the chambers by adding 100 μL of ROXS solution one time to each chamber and image. 3.2 Preparation of Protein Sample

1. Use transfection reagent to transfect two 10 cm plates of approximately 50% confluent HEK293 cells with DNA construct of interest. Allow cells to express protein of interest for 24–48 h. 2. Visually inspect the cells to ensure that they look healthy. Scrape the cells off of the plates and transfer them into one 50 mL conical vial. Pellet the cells in a centrifuge at 1100  g for 3 min at 23  C. Decant the media away from the pelleted cells. Wash the cells three times with ECB. During the first wash, transfer the cells to a 15 mL conical vial. After three washes, resuspend the cells in 3 mL of ECB. Add fluorophores such that the final concentration in 3 mL will be 600 nM donor and 2.4 μM acceptor. Rotate the cells for 30–60 min at room temperature (see Notes 6–8). 3. After the 1-h incubation, wash the cells three times with ECB. Resuspend the cells in 2 mL of solubilization buffer (see Note 9). Nutate the cell suspension for 1 h at 4  C. 4. Transfer the crude lysate to an ultracentrifuge tube and make a balance tube. Spin the tubes at 100,000  g for 1 h at 4  C. 5. Collect the supernatant containing the labeled protein of interest. Use this supernatant to make the diluted protein solution that will be applied to the slides (see Note 10).

3.3

Data Collection

A confocal microscope is used for data collection. An example confocal system that is used for such smFRET experiments is the PicoQuant MicroTime 200 Fluorescence Lifetime Microscope. Scan a 20 μm  20 μm area of the slide to identify molecules for imaging; then record the fluorescence intensity of each molecule until the donor and acceptor fluorophores undergo photobleaching (see Note 11). Example traces showing anticorrelation between the donor and acceptor are shown in Fig. 4. To ensure that the fluorescence is from the receptor tagged at the site of interest, one will need to perform control experiments with no protein, protein with a single labeled site, and protein with two labeled sites as shown in Fig. 5.

12

Douglas B. Litwin et al.

Fig. 4 (a) A representative smFRET trace measured from an agonist binding domain of AMPA receptor showing raw trajectories of the donor (blue) and acceptor (red) photons as a function of time. (b) The resulting calculated FRET trajectory (green) and its denoised counterpart (black). This research was originally published in the Journal of Biological Chemistry [3] © the American Society for Biochemistry and Molecular Biology

Fig. 5 smFRET slide imaging showing the specificity of the labeling strategy. Blue pane corresponds to donor channel, and brown pane is FRET channel (acceptor frequency emission with donor frequency excitation). As seen in (a), there is no labeling when protein is not applied to the slide. In (b), the appearance of donor signal is seen when protein with single cysteine labeled with fluorophores is applied to the slide. In (c), the appearance of FRET is seen when protein with two cysteine labeling sites labeled with fluorophores is applied to the slide. This research was originally published in the Journal of Biological Chemistry [27] © the American Society for Biochemistry and Molecular Biology

smFRET of Glutamate Receptors

3.4

Data Analysis

13

For data analysis, export the donor and acceptor traces as ASCII files. Then, calculate the corrected FRET efficiencies (EA) according to Eqs. 3 and 4:   I D 1 ð3Þ EA ¼ 1 þ γ IA γ¼

ηA ΦA ηD ΦD

ð4Þ

where ID is the intensity of donor fluorescence, IA is the intensity of acceptor fluorescence, γ is a correction factor accounting for the efficiencies of the detectors used and for the quantum yield of the fluorophores, η is the efficiency of a given detector, and Φ is the quantum yield of a given fluorophore. Once corrected FRET efficiencies have been calculated, plot the histograms depicting the distribution of various FRET efficiencies, and analyze those histograms using any graphing software. To identify states in the single-molecule trajectory, the hidden Markov modeling (HaMMy) analysis [28] or the step transition and state identification (STaSI) analysis can be used [29].

4

Notes 1. The fluorophore pair that will be used should be selected to have an R0 that is in the linear efficiency range for the distance that will be measured. Care must be taken to ensure that the fluorophores selected have spectral overlap so that FRET can occur. Furthermore, keep in mind that the distance between two given sites can change as a result of conformational rearrangements within the protein; this makes it necessary to ensure that the R0 of the fluorophore pair will allow measurements across the full range of possible distances. 2. When using antibodies to pull down the protein of interest onto the slide, one must ensure that the only species of protein bound by that antibody in the cell lysate is the protein of interest. Nonspecific binding of other proteins to the slide will increase the background fluorescence on the slide and lower the signal-to-noise ratio. Western blotting with that antibody is a good control to determine this. 3. For any new mutant proteins that one uses in a smFRET experiment, the expression and function of that protein must be verified. Western blotting with a previously validated antibody can confirm the expression of the mutant protein. It is also necessary to compare the function of the mutant protein to that of the wild-type protein using electrophysiological measurements. The function of the mutant protein should be at

14

Douglas B. Litwin et al.

least partially preserved; if the mutant has only a partial loss of function, that is acceptable. 4. It is highly recommended to perform a screen of various transfection conditions when using the unnatural amino acid experimental design. Because there are several constructs that will need to be transfected into the cells, it is necessary to empirically determine the ratios of DNA that will result in maximal expression. 5. When applying solution to the slide, ensure that no bubbles enter the chamber. Bubbles in the chamber can cause patchy labeling of the slide. When applying solutions to the slide, the bubbles are often found at the tip of the pipet that are barely discernible by the eye, and these bubbles must not be injected into the slide chamber. 6. The donor and acceptor fluorophores should be mixed with each other in a separate tube before being mixed with the sample. This will prevent one fluorophore from labeling the majority of the sites before the other fluorophore is added. 7. The amount of time that the sample is incubated with the fluorophores can affect the degree of labeling that occurs. When examining homomeric iGluRs, if one sees that the majority of molecules are exhibiting donor-only or acceptoronly labeling, one might consider adjusting the donor-toacceptor ratio in future experiments. The donor-to-acceptor ratio of 1:4 that is listed in the protocol is one that has been used for heteromeric NMDA receptors. Each investigator will need to optimize the donor-to-acceptor ratio that is best for his/her experimental setup. 8. Avoid exposing the fluorophores to light. Exposure to light should be avoided by covering the sample tube with foil. 9. Once the solubilization buffer has been added to the cells, the sample must be kept either on ice or at 4  C to prevent protein degradation. It is crucial to use fresh protease inhibitors in the solubilization buffer because endogenous proteases from the HEK cells will remain in the sample. Without protease inhibitors, the sample will be rapidly degraded. 10. Once the ultracentrifugation step is complete, the sample should be applied to the slide as soon as possible. Therefore one should time the slide preparation to line up with the protein preparation so that the slide and protein will be ready simultaneously. 11. When first imaging a slide, the immersion oil between the objective and the slide will be equilibrating to a stable volume. Additionally, the temperature of the slide will be equilibrating with the surrounding air. These two processes will affect the focus of the scope on the slide and cause that focus to drift. One will need to check the focus prior to imaging.

smFRET of Glutamate Receptors

15

Acknowledgments This project was supported by NIH grants R35GM122528 (VJ) and F31GM130035 (RJD) and by the Houston Area Molecular Biophysics Training Program NIH- 2T32 GM008280-26 (DBL). References 1. Sirrieh RE, MacLean DM, Jayaraman V (2015) A conserved structural mechanism of NMDA receptor inhibition: a comparison of ifenprodil and zinc. J Gen Physiol 146(2):173–181 2. MacLean DM, Ramaswamy SS, Du M, Howe JR, Jayaraman V (2014) Stargazin promotes closure of the AMPA receptor ligand-binding domain. J Gen Physiol 144(6):503–512 3. Ramaswamy S, Cooper D, Poddar N et al (2012) Role of conformational dynamics in α-amino-3-hydroxy-5-methylisoxazole-4-propionic acid (AMPA) receptor partial agonism. J Biol Chem 287(52):43557–43564 4. Shaikh SA, Dolino DM, Lee G et al (2016) Stargazin modulation of AMPA receptors. Cell Rep 17(2):328–335 5. Dolino DM, Chatterjee S, MacLean DM et al (2017) The structure-energy landscape of NMDA receptor gating. Nat Chem Biol 13:1232–1238 6. Baker KA, Lamichhane R, Lamichhane T, Rueda D, Cunningham PR (2016) Protein–RNA dynamics in the central junction control 30S ribosome assembly. J Mol Biol 428(18):3615–3631 7. Bal M, Zaika O, Martin P, Shapiro MS (2008) Calmodulin binding to M-type K+ channels assayed by TIRF/FRET in living cells. J Physiol 586(9):2307–2320 8. Dolino DM, Rezaei Adariani S, Shaikh SA, Jayaraman V, Sanabria H (2016) Conformational selection and submillisecond dynamics of the ligand-binding domain of the N-Methyl-d-aspartate receptor. J Biol Chem 291(31):16175–16185 9. Gomes G-N, Gradinaru CC (2017) Insights into the conformations and dynamics of intrinsically disordered proteins using singlemolecule fluorescence. Biochim Biophys Acta Proteins Proteom 1865(11 Pt B):1696–1706 10. Gouridis G, Schuurman-Wolters GK, Ploetz E et al (2015) Conformational dynamics in substrate-binding domains influences transport in the ABC importer GlnPQ. Nat Struct Mol Biol 22(1):57–64

11. Kempe D, Cerminara M, Poblete S, Scho¨ne A, Gabba M, Fitter J (2017) Single-molecule FRET measurements in additive-enriched aqueous solutions. Anal Chem 89(1):694–702 12. Kim J-Y, Kim C, Lee NK (2015) Real-time submillisecond single-molecule FRET dynamics of freely diffusing molecules with liposome tethering. Nat Commun 6:6992 13. Martinac B (2017) Single-molecule FRET studies of ion channels. Prog Biophys Mol Biol 130(Pt B):192–197 14. McLoughlin SY, Kastantin M (2013) Singlemolecule resolution of protein structure and interfacial dynamics on biomaterial surfaces. Proc Natl Acad Sci U S A 110 (48):19396–19401 15. Song C-X, Diao J, Brunger AT, Quake SR (2016) Simultaneous single-molecule epigenetic imaging of DNA methylation and hydroxymethylation. Proc Natl Acad Sci U S A 113 (16):4338–4343 16. Stockmar F, Kobitski AY, Nienhaus GU (2016) Fast folding dynamics of an intermediate state in RNase H measured by singlemolecule FRET. J Phys Chem B 120 (4):641–649 17. Wang S, Vafabakhsh R, Borschel WF, Ha T, Nichols CG (2016) Structural dynamics of potassium-channel gating revealed by singlemolecule FRET. Nat Struct Mol Biol 23 (1):31–36 18. Wang Y, Liu Y, DeBerg HA et al (2014) Single molecule FRET reveals pore size and opening mechanism of a mechano-sensitive ion channel. elife 3:e01834 19. Warhaut S, Mertinkus KR, Ho¨llthaler P et al (2017) Ligand-modulated folding of the fulllength adenine riboswitch probed by NMR and single-molecule FRET spectroscopy. Nucleic Acids Res 45(9):5512–5522 20. Landes CF, Rambhadran A, Taylor JN, Salatan F, Jayaraman V (2011) Structural landscape of isolated agonist-binding domains from single AMPA receptors. Nat Chem Biol 7 (3):168–173

16

Douglas B. Litwin et al.

21. Cooper DR, Dolino DM, Jaurich H et al (2015) Conformational transitions in the glycine-bound GluN1 NMDA receptor LBD via single-molecule FRET. Biophys J 109 (1):66–75 22. Roy R, Hohng S, Ha T (2008) A practical guide to single-molecule FRET. Nat Methods 5(6):507–516 23. Sisamakis E, Valeri A, Kalinin S, Rothwell PJ, Seidel CAM (2010) Accurate single-molecule FRET studies using multiparameter fluorescence detection. Methods Enzymol 475:455–513 24. Lu¨ W, Du J, Goehring A, Gouaux E (2017) Cryo-EM structures of the triheteromeric NMDA receptor and its allosteric modulation. Science 355(6331):eaal3729 25. Twomey EC, Yelshanskaya MV, Grassucci RA, Frank J, Sobolevsky AI (2017) Structural bases

of desensitization in AMPA receptor-auxiliary subunit complexes. Neuron 94(3):569–580 26. Ye S, Ko¨hrer C, Huber T et al (2008) Sitespecific incorporation of keto amino acids into functional G protein-coupled receptors using unnatural amino acid mutagenesis. J Biol Chem 283(3):1525–1533 27. Dolino DM, Cooper D, Ramaswamy S, Jaurich H, Landes CF, Jayaraman V (2015) Structural dynamics of the glycine-binding domain of the N-methyl-D-aspartate receptor. J Biol Chem 290(2):797–804 28. McKinney SA, Joo C, Ha T (2006) Analysis of single-molecule FRET trajectories using hidden Markov modeling. Biophys J 91 (5):1941–1951 29. Shuang B, Cooper D, Taylor JN et al (2014) Fast step transition and state identification (STaSI) for discrete single-molecule data analysis. J Phys Chem Lett 5(18):3157–3161

Chapter 2 Ultrastructural Observation of Glutamatergic Synapses by Focused Ion Beam Scanning Electron Microscopy (FIB/SEM) Ai Takahashi-Nakazato, Laxmi Kumar Parajuli, Hirohide Iwasaki, Shinji Tanaka, and Shigeo Okabe Abstract A thorough understanding of the synaptic ultrastructure is necessary to bridge our current knowledge gap about the relationship between neuronal structure and function. Recent development of focused ion beam scanning electron microscopy (FIB/SEM) has made it possible to image neuronal structures with high speed and efficiency. Here, we present our routine protocol for correlative two-photon microscopy and FIB/SEM imaging of glutamatergic synapses. Femtosecond-pulsed near-infrared laser was used to create fiducial marks around the dendrite of interest in aldehyde-fixed tissues. Thereafter, samples were subjected to en bloc staining with rOTO (reduced osmium tetroxide-thiocarbohydrazide-osmium tetroxide), followed by lead aspartate and uranyl acetate to enhance tissue contrast. Reliable detection of postsynaptic density (PSD) and plasma membrane contours by the sample preparation protocol optimized for FIB/SEM allows us to precisely evaluate morphological features that shape glutamatergic synaptic transmission. Key words Glutamatergic synapse, Dendritic spines, FIB/SEM, Correlative microscopy, 3D reconstruction

1

Introduction Neurons communicate with each other via specialized structures called synapses. Electron microscopic observation has identified two types of synaptic contacts in the brain [1]. Gray type I synapses (also called as asymmetrical synapses) have characteristic thickenings of postsynaptic membrane and mediate excitatory synaptic transmission in the brain. On the other hand, Gray type II synapses do not have prominent PSD and are therefore called as symmetrical, inhibitory synapses. In cortical neurons, excitatory synapses typically impinge on dendritic spines. Functional in vivo imaging studies have shown that the excitatory synapses are highly plastic during development [2–4] and their structure changes in response

Corinna Burger and Margaret Jo Velardo (eds.), Glutamate Receptors: Methods and Protocols, Methods in Molecular Biology, vol. 1941, https://doi.org/10.1007/978-1-4939-9077-1_2, © Springer Science+Business Media, LLC, part of Springer Nature 2019

17

18

Ai Takahashi-Nakazato et al.

to experience and learning [5, 6]. Furthermore, psychiatric diseases are often associated with altered dynamics and stability of glutamatergic synapses [4, 7–9]. However, despite of their important role in the neuronal network, the heterogeneity of synaptic structures and the structural changes upon plasticity is poorly understood. Due to the diffraction limit, light microscopic techniques are of limited use in characterizing activity-induced ultrastructural changes. Therefore, electron microscopy (EM) is indispensable to faithfully identify subcellular contents of individual synaptic connections. Thus far, ultrastructural analysis of glutamatergic synapses is mainly performed on single-section transmission electron microscopy (TEM) images or a handful of serial section images. However, depending on the plane of sectioning, identification of glutamatergic synapses that are cut obliquely or en face is not reliable on single-section TEM images, and one needs to observe series of serial images to reveal precise ultrastructural changes at dendrites and dendritic spines. Although TEM can produce highresolution images, it is not suitable for extension of an imaging volume because of difficulty in manual handling of a large number of serial ultrathin sections. This is particularly a concern for largescale circuit reconstruction studies aimed at understanding the biological meaning of individual connections in the context of integrated functions of local circuits. Recently, focused ion beam scanning electron microscope (FIB/SEM) has emerged as an attractive, high-throughput, automated tool for volumetric EM imaging of neuropil. FIB/SEM employs gallium ion beam to mill the surface of the plasticembedded tissue and collects either backscattered or secondary electron beams from the newly exposed block of tissue to create SEM images [10]. Milling and imaging processes are repeatedly performed to obtain stacks of serial EM images of any desired thickness from a given tissue volume. As the thickness of the tissue to be milled by gallium ions can be accurately controlled, FIB/ SEM images have higher axial resolution (~10 nm) than those acquired by TEM (~40 nm). Thinner FIB/SEM sections permit the possibility to obtain images at isotropic voxels (10 nm
10 nm
10 nm), which help precise quantification of PSD, synaptic vesicles, and intracellular organelles. Since its first introduction to the field of neurobiology in 2008, many refinements have been made in the protocol for better visualization of neuronal ultrastructure [11]. Here, we report our routine FIB/SEM sample preparation protocol, which results in clear visualization of structures that influence glutamatergic synaptic transmission. Furthermore, we also describe near-infrared branding (NIRB) protocol [12, 13] that facilitates retrospective EM observation of glutamatergic synapses that has been imaged in live preparations.

Observation of Glutamatergic Synapse by FIB/SEM

2

19

Materials

2.1 Fixation (See Note 1)

1. Paraformaldehyde (PFA): 10% in water. Dissolve 1.5 g of paraformaldehyde in 15 ml of water. Add 1 N NaOH (up to 30 μl), vortex to mix thoroughly, and heat at 60  C with water bath. Filtrate the solution through filter paper (No. 1 grade) and store at 4  C. 2. Cacodylate buffer: 1 M in water as a stock solution. Adjust pH to 7.4 with HCl. Store at room temperature (RT) (see Note 2). 3. Ringer’s solution: Dissolve 0.912 g of NaCl in 80 ml of water using a stirrer. Add the reagents as follows: 0.04 g KCl, 0.12 g HEPES, 0.18 g glucose, 0.4 ml of 1 M CaCl2, 0.2 ml of 1 M MgCl2, and 0.5 ml of 0.1 M NaH2PO4. Adjust pH to 7.4 with NaOH. Use ultrapure water to bring the final volume to 100 ml. The solution is stored at 4  C refrigerator for longterm storage. Warm the solution to RT before use. 4. Fixative solution: 2% PFA and 2.5% glutaraldehyde (GA) in 0.1 M cacodylate buffer. 20 ml of 0.2 M cacodylate buffer, 8 ml of 10% PFA, and 8 ml of water. Add 4 ml of 25% GA immediately before perfusion (see Note 3).

2.2 rOTO (Reduced Osmium TetroxideThiocarbohydrazideOsmium) Staining

1. Potassium ferrocyanide: 5% in water (see Note 4). 2. Osmium tetroxide solution: 2% osmium tetroxide (OsO4) and 1.5% potassium ferrocyanide in 0.1 M cacodylate buffer. Mix 0.2 ml of 1 M cacodylate buffer, 0.6 ml of 5% potassium ferrocyanide, and 0.2 ml of water. 1 ml of 4% OsO4 is added immediately before use (see Note 5). 3. Thiocarbohydrazide (TCH) solution: 1% in water (see Note 6). 4. Uranyl acetate solution: 2% in 50% ethanol. Dissolve 0.2 g of uranyl acetate in 5 ml of distilled water. Store at 4  C under lightproof condition. Centrifuge immediately before use and mix the supernatant to the equal volume of 99.5% ethanol (see Note 7). 5. Lead aspartic acid solution: Dissolve 0.01 g of L-aspartic acid in 25 ml of water. Pour 0.164 g of lead nitrate to 25 ml of aspartic acid solution. Use magnetic stirrers to dissolve the solution. Adjust pH to 5.5 with KOH. Place the glass flask containing the solution to 60  C oven for 30 min to achieve complete dissolution. Filtrate with a filter paper (No. 7 grade), and collect the solution in a 50 ml disposable plastic syringe attached to a filter (pore size, 0.22 μm) (see Note 8).

2.3 Resin Embedding and Block Trimming

1. Graded series of ethyl alcohol: 50%, 70%, 80%, 90%, 95%, 99.5%, and 100% (vol/vol). 2. Propylene oxide (see Note 9).

20

Ai Takahashi-Nakazato et al.

3. Durcupan ACM embedding media: 11.4 g of component A (epoxy resin), 10 g of component B (hardener), 0.33 g of component C (accelerator), and 0.1 ml of component D (plasticizer). Sequentially pour components A, B, and D into a disposable plastic cup. Mix these three components thoroughly (>5 min) using plastic Pasteur pipettes. Add component C and continue to mix (>5 min). 4. Silicone-coated glass slide. 5. Polychlorotrifluoroethylene film (ACLAR film). 6. Toluidine blue solution: 1% in water. Dissolve 1.895 g of sodium tetraborate decahydrate in 100 ml of water. Dissolve 1 g of toluidine blue in sodium tetraborate solution, and filtrate with filter paper (No. 1 grade).

3

Methods

3.1 Sample Preparation for FIB/SEM 3.1.1 Fixation

Fixation serves the purpose of preserving the biological structures as close as possible to its natural state. Tissue fixation is achieved by treating biological specimens with reagents that cross-link protein and lipid, such as glutaraldehyde and OsO4, respectively [14]. Fixation results in hardening of soft, fragile brain tissues, thereby making it easier to handle and less susceptible to mechanical damage during tissue processing. 1. Anesthetize mouse by intraperitoneal injection of pentobarbital (100 mg/kg body weight). 2. After ensuring that the animal is completely anesthetized, open abdominal cavity by a long midline incision, followed by removing the frontal part of the ribs and the sternum to expose the heart. With the help of a peristaltic pump and 27-gauge butterfly needle, transcardially perfuse mouse with approximately 10 ml of ringer solution to flush blood cells and clear blood vessel for unobstructed passage of fixatives to the brain. Thereafter, at a flow rate of 5–10 ml/min, continue perfusion with approximately 100 ml of fixative solution. 3. Excise the brain immediately after perfusion, and postfix it by immersing in the same fixative solution for 1 h at RT. 4. Glue the brain block containing the region of interest to the stage of the vibratome, and prepare 100 μm thick sections. Use a paint brush to collect sections in a multi-well plastic plates.

3.1.2 Near-Infrared Branding (NIRB)

Near-infrared laser pulses are used to create a fiducial mark in the vicinity of the synapse of interest. As the laser mark can be easily visualized under EM, NIRB facilitates retrospective EM identification of dendritic spines that were previously imaged by two-photon microscopy in live preparations (Fig. 1).

Observation of Glutamatergic Synapse by FIB/SEM

21

Fig. 1 An example of NIRB mark in a fixed brain section. (a and b) Highmagnification image of the dendrite of interest before (a) and after (b) NIRB. The mark can easily be visualized by autofluorescence. (c) Low-magnification image of the target dendrite with smaller and bigger NIRB marks. (d) Bigger NIRB mark can be identified under the standard light microscopy. Scale bars, 10 μm in (a) and (b), 50 μm in (c) and (d)

1. Put a drop of 0.1 M cacodylate buffer on a clean glass slide. Transfer the brain sections onto the glass slide and place a clean cover glass onto the slide. 2. Place the glass slide on the stage of two-photon microscopy. Image the dendrites with an infrared-pulsed laser tuned to an appropriate wavelength (approximately 920 nm for EGFP), and locate the dendrites of interest. 3. Tune the laser source to 800 nm, and operate it in a line-scan mode (500–1000 scans, 10–12.5 μs duration/pixel, 100–120 mW power, 60
water immersion lens with an N.A. of 0.9) to generate high enough thermal energy that is capable of creating a square-shaped burn mark (20 μm on each side) around the target dendrite. The dimension and shape of the square frame can vary depending on the length of the target dendrite that the experimenter is interested to reconstruct from

22

Ai Takahashi-Nakazato et al.

the EM images. In order to prevent excess heating of the tissue, it is advisable to first start with low power (100 mW) and then gradually increase the power until 2–3 μm deep grooves are formed. Note that the glass slide must be chilled with ice to allow rapid heat dissipation from the irradiated brain slices. 4. To facilitate the identification of the target dendrite at low magnification, create an additional bigger square frame (100 μm on each side), concentric to the inner square, by following the similar procedure as above. Increase the laser intensity and the pulse duration (500–1000 scans, 100–200 μs duration/pixel, 180–200 mW power) to create thick burnt lines. 3.1.3 rOTO Staining

In order to enhance the visibility of the membranes under EM, rOTO staining is performed by treating sections with a battery of contrasting agents such as OsO4, TCH, uranium, and lead [15] (see Note 10). OsO4 is a strong oxidant and works as a fixative by forming intra- and intermolecular cross-links of unsaturated fatty acids in the membranes. TCH increases the absorption of OsO4 into tissue by the formation of OsO4-TCH complex. Heavy metals such as uranium and lead enhance membrane contrast. 1. With the aid of a stereomicroscope and a razor blade, excise the region of interest from sections containing NIRB marks. 2. Place the sections in glass scintillation vials, and postfix with 2% OsO4 in 0.1 M cacodylate buffer/1.5% potassium ferrocyanide for 1 h at RT. 3. Wash the sections with distilled water (dH2O) for 3
5 min at RT. 4. Incubate sections in 1% TCH for 25 min at RT. Wash sections with dH2O for 3
5 min at RT. 5. Stain with 2% OsO4 for 30 min at RT. Wash sections with dH2O for 3
5 min at RT. 6. Stain with 2% uranyl acetate and leave sections overnight at 4  C in dark. Wash sections with dH2O for 3
5 min at RT. 7. Stain with lead aspartic acid solution for 30 min at 60  C. Wash sections with dH2O for 3
5 min at RT.

3.1.4 Resin Embedding

1. Transfer sections to scintillation vials. The sections are submerged for 10 min each in graded ethanol series of 50%, 70%, 80%, 90%, 95%, and 99.5% of ethanol. The sections are finally incubated in two changes of 100% ethanol for 10 min each. 2. Incubate sections in propylene oxide (PO) (2
10 min) to facilitate resin infiltration to the tissue.

Observation of Glutamatergic Synapse by FIB/SEM

23

3. Prepare epoxy resin and infiltrate sections with mixtures of resin and PO as follows: 25% resin/PO (resin: PO ¼ 1: 3) for 3 h, 50% resin/PO (resin: PO ¼ 1: 1) for overnight, and 75% resin/PO (resin: PO ¼ 3: 1) for 3 h (see Note 11). All of these steps are carried out at RT and in a desiccator with silica gels. 4. Place a piece of ACLAR film on a silicone-coated glass slide. The sections are sandwiched between ACLAR films and made flat by placing a glass slide on top. Use filter paper to absorb excess resin that leaks from the sides of the glass slides. Place the flat-embedded sections in 60  C oven for 48 h. 5. After resin curing, remove ACLAR films and glass slides from the flat-embedded sections. Excise the region containing the NIRB marks or other regions of interest from flat-embedded section, and glue it to a resin block. 3.1.5 Tissue Sectioning

1. Mount the block in an ultramicrotome, and trim the resin block around the target object with a razor blade to form a trapezoid cutting surface. 2. By using a glass knife, cut the block surface until the embedded tissue is exposed to the surface (see Note 12). 3. For the NIRB sample, trim the edge of the block until the distance from the target dendrite to the nearest edge of the block is less than 50 μm (Fig. 2a) (see Note 13).

3.2 Observation of Glutamatergic Synapses by DualBeam FIB/SEM

In dual-beam FIB/SEM, the surface of the block is milled by FIB and then imaged by SEM. The advantage of FIB/SEM is that both image acquisition and block milling are performed automatically, thus reducing the need for high-level technical expertise to manually obtain serial images. However, a particular disadvantage of FIB/SEM system is that the field of view which can be imaged in XY plane is relatively limited (typically less than 100 μm
100 μm). Furthermore, the destructive nature of FIB/SEM imaging

(a)

(b)

top view of the trimmed surface of the block milling direction by FIB

< 50 µm

Focused ion beam

direction for image acquisition

Scanning electron microscopy

~50

specimen target dendrite

stage (sample holder) fiducial marker by NIRB

Fig. 2 (a) Block trimming of NIRB sample. The target dendrite should be close to the edge (10,000 μm2) over tens of minutes, without worrying about drift or structural plasticity affecting the two channels differently during extended acquisition times [20]. 9.2 Post-acquisition Registration

10

When channels are acquired consecutively, post-acquisition registration of two-color images can be performed to correct for potential drift, though again, the higher the resolution and the smaller structures that are imaged, the larger will the potential error be from erroneous corrections, and these should be performed with care. For example, if imaging pre- and postsynaptic partners that are separated by few tens of nanometers, then even small offsets between the two channels will have a major impact on the perceived distances. The potentially confounding effect of registration may be even more pronounced in two-color time-lapse images, where correction algorithms may erroneously interpret structural cellular plasticity as drift and seek to counter it.

Automated Analyses Tools Are Lacking Geometric analyses pose a major advantage over semiquantitative procedures and enable new insights into the morphology, plasticity, and biophysics of synaptic compartments. However, there is a lack of options for automated analyses of geometries in microscopy images in general. Most, if not all, commercially available analysis software, including popular choices Imaris and Huygens, rely on a thresholding and binarization step during the analysis process, which distorts the size distribution structures that approach the resolution limit in size (see, e.g., [47]). This is especially true for structures that are spatially resolved by the microscope in the lateral plane but not axially along the z-axis, which is a considerable fraction of synaptic structures. It is therefore not possible to extract accurate geometric data on spine necks and axons in thresholded and binarized images, where weak structures may either disappear or appear overly large depending on the chosen threshold (Fig. 6). Manual evaluations confined to the lateral plane are therefore the current gold standard for STED image analysis of geometries; however, analysis of hundreds of structures quickly becomes rather laborious. Further, commonly only a single measured width is obtained and presented for a given segment of an axon or dendritic spine, though more information could potentially be extracted, such as maximum, minimum, and average width. For dendritic spines, such information could conceivably be retrieved alongside measurements of neck length and spine head size for a more complete spine analysis. To harvest more data from STED images, make

44

Martin O. Lenz and Jan Tønnesen

Fig. 6 Morphological analysis of raw and binarized data. (a) STED image of dendritic spines with arrows indicating the spine necks that were analyzed in this figure (as well as Fig. 4b). Same image as depicted in Figs. 3 and 5, zoomed out and rotated (© Tønnesen & N€agerl, Universite´ de Bordeaux/CNRS). (b) The same image binarized to a minimum threshold beyond which spine necks effectively start to disappear. (c) Measurements of nine neck widths analyzed in the raw and binarized image, respectively. In the raw image, neck widths were measured as the Gaussian fit FWHM, while in the binarized image, the full width of the (binarized) neck was obtained. A paired t-test found that the variability and average width was larger in the binarized analysis ( p < 0.001). (d) When comparing the spine neck widths between the two approaches, the coefficient of determination (R2) is 0.66, meaning that 66% of the variation in the binary data can be explained from the raw image data. In the case of these nine spines, the regression line is described by (y ¼ 1.73x + 3), indicating that spine neck widths in the binarized image will be overestimated 1.7-fold relative to the geometric data from the raw image

analyses easier and faster, and avoid inadvertent bias from the experimenter performing the analysis, automated options are highly warranted and will likely emerge in the near future.

11

Concluding Remarks STED microscopy images are a potential source of new biological information, which has not been previously accessible in live cell images. Quantitative information about true geometries, as opposed to qualitative analyses, is not only superior in terms of scientific value but also provides a measure that is robust and allows comparisons between labs. However, to assure a valuable outcome, analyses must be carefully planned, conducted, and documented. As with many experimental methods, the analyses procedures vary between labs, which may impact the outcome. Transparency in describing the involved procedures for a given experiment is therefore key, and fortunately this is generally the rule in STED imaging studies. Future developments of automated analysis software are warranted, as these will facilitate extraction of accurate geometric data in an efficient, standardized, and unbiased way.

STED Imaging and Analysis

45

Acknowledgments JT is supported by grants from the Spanish Ministry of Economy and Competitiveness (RYC-2014-15994 and SAF2017-83776-R). References 1. Klar TA, Jakobs S, Dyba M et al (2000) Fluorescence microscopy with diffraction resolution barrier broken by stimulated emission. Proc Natl Acad Sci 97:8206–8210 2. Abbe E (1882) The relation of aperture and power in the microscope. J R Microsc Soc 2:300–309 3. Harris KM, Jensen FE, Tsao B (1992) Threedimensional structure of dendritic spines and synapses in rat hippocampus (CA1) at postnatal day 15 and adult ages: implications for the maturation of synaptic physiology and longterm potentiation. J Neurosci 12:2685–2705 4. Kasthuri N, Hayworth KJ, Berger DR et al (2015) Saturated reconstruction of a volume of neocortex. Cell 162:648–661 5. Mishchenko Y, Hu T, Spacek J et al (2010) Ultrastructural analysis of hippocampal neuropil from the connectomics perspective. Neuron 67:1009–1020 6. Betzig E, Patterson GH, Sougrat R et al (2006) Imaging intracellular fluorescent proteins at nanometer resolution. Science 313:1642–1645 7. Rust MJ, Bates M, Zhuang X (2006) Subdiffraction-limit imaging by stochastic optical reconstruction microscopy (STORM). Nat Methods 3:793–795 8. Urban NT, Willig KI, Hell SW et al (2011) STED nanoscopy of actin dynamics in synapses deep inside living brain slices. Biophys J 101:1277–1284 9. Schermelleh L, Heintzmann R, Leonhardt H (2010) A guide to super-resolution fluorescence microscopy. J Cell Biol 190:165–175 10. Wildanger D, Medda R, Kastrup L et al (2009) A compact STED microscope providing 3D nanoscale resolution. J Microsc 236:35–43 11. Matsuzaki M, Ellis-Davies GC, Nemoto T et al (2001) Dendritic spine geometry is critical for AMPA receptor expression in hippocampal CA1 pyramidal neurons. Nat Neurosci 4:1086–1092 12. Noguchi J, Matsuzaki M, Ellis-Davies GCR et al (2005) Spine-neck geometry determines NMDA receptor-dependent Ca2+ signaling in dendrites. Neuron 46:609–622 13. Berning S, Willig KI, Steffens H et al (2012) Nanoscopy in a living mouse brain. Science 335:551

14. Willig KI, Steffens H, Gregor C et al (2014) Nanoscopy of filamentous actin in cortical dendrites of a living mouse. Biophys J 106: L01–L03 15. Whelan DR, Bell TDM (2015) Image artifacts in single molecule localization microscopy: why optimization of sample preparation protocols matters. Sci Rep 5:7924 16. Galbraith CG, Galbraith JA (2011) Superresolution microscopy at a glance. J Cell Sci 124:1607–1611 17. Korogod N, Petersen CCH, Knott GW (2015) Ultrastructural analysis of adult mouse neocortex comparing aldehyde perfusion with cryo fixation. Elife 4. https://doi.org/10.7554/ eLife.05793 18. Rittweger E, Rankin BR, Westphal V et al (2007) Fluorescence depletion mechanisms in super-resolving STED microscopy. Chem Phys Lett 442:483–487 19. Einstein A (1917) Zur Quantentheorie der Strahlung. Phys Z 18:121–128 20. Tønnesen J, Inavalli VVGK, N€agerl UV (2018) Super-resolution imaging of the extracellular space in living brain tissue. Cell 172:1108–1121.e15 21. Westphal V, Hell SW (2005) Nanoscale resolution in the focal plane of an optical microscope. Phys Rev Lett 94:143903 22. Tonnesen J, Nagerl UV (2013) Two-color STED imaging of synapses in living brain slices. Methods Mol Biol 950:65–80 23. Harke B, Keller J, Ullal CK et al (2008) Resolution scaling in STED microscopy. Opt Express 16:4154–4162 24. Tønnesen J, Nadrigny F, Willig KI et al (2011) Two-color STED microscopy of living synapses using a single laser-beam pair. Biophys J 101 (10):2545–2552 25. Donnert G, Keller J, Wurm CA et al (2007) Two-color far-field fluorescence nanoscopy. Biophys J 92:L67–L69 26. Bottanelli F, Kromann EB, Allgeyer ES et al (2016) Two-colour live-cell nanoscale imaging of intracellular targets. Nat Commun 7:10778 27. Pellett PA, Sun X, Gould TJ et al (2011) Two-color STED microscopy in living cells. Biomed Opt Express 2:2364–2371

46

Martin O. Lenz and Jan Tønnesen

28. Bu¨ckers J, Wildanger D, Vicidomini G et al (2011) Simultaneous multi-lifetime multicolor STED imaging for colocalization analyses. Opt Express 19:3130–3143 29. Lenz MO, Brown ACN, Auksorius E, et al 2011 A STED-FLIM microscope applied to imaging the natural killer cell immune synapse. Proc. SPIE 7903, Multiphoton microscopy in the biomedical sciences XI, p. 79032D 30. Zimmermann T, Rietdorf J, Pepperkok R (2003) Spectral imaging and its applications in live cell microscopy. FEBS Lett 546:87–92 31. Cole RW, Jinadasa T, Brown CM (2011) Measuring and interpreting point spread functions to determine confocal microscope resolution and ensure quality control. Nat Protoc 6:1929–1941 32. White JG, Amos WB, Fordham M (1987) An evaluation of confocal versus conventional imaging of biological structures by fluorescence light microscopy. J Cell Biol 105:41–48 33. Che´reau R, Tønnesen J, N€agerl UV (2015) STED microscopy for nanoscale imaging in living brain slices. Methods 88:57–66 34. Zhang B, Zerubia J, Olivo-Marin J-C (2007) Gaussian approximations of fluorescence microscope point-spread function models. Appl Opt 46:1819–1829 35. Bethge P, Che´reau R, Avignone E et al (2013) Two-photon excitation STED microscopy in two colors in acute brain slices. Biophys J 104:778–785 36. Takasaki KT, Ding JB, Sabatini BL (2013) Live-cell superresolution imaging by pulsed STED two-photon excitation microscopy. Biophys J 104:770–777 37. Wegner W, Ilgen P, Gregor C et al (2017) In vivo mouse and live cell STED microscopy of neuronal actin plasticity using far-red emitting fluorescent proteins. Sci Rep 7:11781 38. Wijetunge LS, Angibaud J, Frick A et al (2014) Stimulated emission depletion (STED) microscopy reveals nanoscale defects in the developmental trajectory of dendritic spine morphogenesis in a mouse model of fragile X syndrome. J Neurosci 34:6405–6412

39. Lenz MO, Sinclair HG, Savell A et al (2014) 3-D stimulated emission depletion microscopy with programmable aberration correction. J Biophotonics 7:29–36 40. Zucker RM, Rigby P, Clements I et al (2007) Reliability of confocal microscopy spectral imaging systems: use of multispectral beads. Cytom Part J Int Soc Anal Cytol 71:174–189 41. Boutet de Monvel J, Le Calvez S, Ulfendahl M (2001) Image restoration for confocal microscopy: improving the limits of deconvolution, with application to the visualization of the mammalian hearing organ. Biophys J 80:2455–2470 42. McNally JG, Karpova T, Cooper J et al (1999) Three-dimensional imaging by deconvolution microscopy. Methods 19:373–385 43. Zanella R, Zanghirati G, Cavicchioli R et al (2013) Towards real-time image deconvolution: application to confocal and STED microscopy. Sci Rep 3:2523 44. Che´reau R, Saraceno GE, Angibaud J et al (2017) Superresolution imaging reveals activity-dependent plasticity of axon morphology linked to changes in action potential conduction velocity. Proc Natl Acad Sci U S A 114:1401–1406 45. Neumann D, Bu¨ckers J, Kastrup L et al (2010) Two-color STED microscopy reveals different degrees of colocalization between hexokinase-I and the three human VDAC isoforms. PMC Biophys 3:4 46. Go¨ttfert F, Wurm CA, Mueller V et al (2013) Coaligned dual-channel STED nanoscopy and molecular diffusion analysis at 20 nm resolution. Biophys J 105:L01–L03 47. Swanger SA, Yao X, Gross C et al (2011) Automated 4D analysis of dendritic spine morphology: applications to stimulus-induced spine remodeling and pharmacological rescue in a disease model. Mol Brain 4:38 48. Tønnesen J, Katona G, Ro´zsa B et al (2014) Spine neck plasticity regulates compartmentalization of synapses. Nat Neurosci 17:678–685

Chapter 4 Cell-Based Enzyme-Linked Immunosorbent Assay (Cell-ELISA) Analysis of Native and Recombinant Glutamate Receptors Elek Molna´r Abstract Glutamate receptors (GluRs) located primarily on the membranes of neurons and glial cells are responsible for excitatory synaptic transmission in the central nervous system. The transport of GluRs to the cell surface is a highly regulated dynamic process that determines neuronal excitability and synaptic responses. The molecular and cellular mechanisms of GluR trafficking are often studied in cell cultures. These studies require sensitive techniques that allow the measurement of total and surface-expressed GluRs in cell populations. The cell-based enzyme-linked immunosorbent assay (cell-ELISA) combines steps of direct immunochemical labelling of cell cultures and ELISA. It can be used for quantitative comparisons of surface-expressed and total protein contents of various cell cultures. While several cell-ELISA protocols are available for different cell types, in this chapter we describe the procedure that we have applied for the investigation of quantitative changes in the cell surface expression of recombinant ionotropic glutamate receptors (iGluRs) in adherent human embryonic kidney 293 (HEK293) cells and endogenous iGluR proteins in primary neuronal cultures. Key words Antibodies, Cell-ELISA, Cell surface protein, Glutamate receptor, Neuronal culture, Targeting, Trafficking

1

Introduction L-Glutamate

is the major excitatory neurotransmitter in the mammalian central nervous system. It acts via a number of different ionotropic (ligand-gated ion channels) and metabotropic (G protein-coupled) glutamate receptors (GluRs). In addition to excitatory neurotransmission, GluRs play a key role in brain development, plasticity of synaptic transmission, neurodegeneration, and neuronal cell death (reviewed in [1]). Molecular cloning identified 18 different genes for ionotropic GluRs (iGluRs) and 8 genes for metabotropic GluRs (mGluRs). Wide range of studies revealed that the activity, synaptic distribution, cell surface trafficking, and internalization of various GluRs are dynamically regulated by complex

Corinna Burger and Margaret Jo Velardo (eds.), Glutamate Receptors: Methods and Protocols, Methods in Molecular Biology, vol. 1941, https://doi.org/10.1007/978-1-4939-9077-1_4, © Springer Science+Business Media, LLC, part of Springer Nature 2019

47

48

Elek Molna´r

posttranslational modifications and protein-protein interactions (reviewed in [1]). To investigate the molecular and cellular mechanisms of these processes, GluRs are often studied in cell cultures where surface-expressed neurotransmitter receptors and other membrane proteins are directly accessible for antibody labelling [2–5]. A highly sensitive cell-based enzyme-linked immunosorbent assay (cell-ELISA) is a useful technique for the quantification of total and surface-expressed recombinant proteins in transfected cells [6–9] or native GluRs in neuronal cultures [2–4]. The assay is performed directly in culture plates and represents a combination of enzyme immunocytochemistry and ELISA [9–11]. Cell-ELISA is mainly used to obtain information about the relative abundance of an antigen in a cell population. Following the direct labelling of cells with a primary anti-GluR (or other) antibody, horseradish peroxidase (HRP)-conjugated secondary antibody is used followed by incubation with the soluble HRP substrate. The optical density of the HRP reaction end product can be measured using a photometer. The assay can be performed with either fixed or unfixed (live) cultures (Fig. 1). Changes in the cell surface-expressed GluR population can be studied selectively in non-permeabilized cells using antibodies raised against extracellular epitopes [2–4]. Cell-ELISA

Surface expressed Non-permeabilised

Key steps: 1) Fixation (4% Paraformaldehyde) 2) Blocking and permeabilisation of membranes (± 1% Triton X-100) 3) Immunostaining with extracellular domain specific antibody 4) Incubation with HRP conjugated secondary antibody 5) Incubation with 3,3',5,5'-tetramethylbenzidine substrate

Total Permeabilised (1% Triton X-100)

5) Termination of reaction with HCl and measurement of OD450 6) Washing plates in PBS and solubilising cells using 0.5% SDS 7) Determination of protein concentration 8) Calculation of OD450/µg protein for each sample 9) Calculation of relative changes in total and surface expressed receptor populations

Fig. 1 Quantification of surface-expressed and total iGluR proteins in cell cultures using cell-ELISA. Schematic diagram illustrates the immunolabelling of surface-expressed and total iGluR populations in paraformaldehyde-fixed cells without (surface) and with (total) Triton X-100 permeabilization. The levels of surface and total immunoreactivities obtained using extracellular domain-specific anti-iGluR antibodies are normalized to protein levels. Key steps of the cell-ELISA procedure are listed on the left

Cell-ELISA Analysis of Glutamate Receptors

49

of fixed and permeabilized cells provides information about the total GluR content of cultures. This approach enables the parallel analysis and comparison of several adherent cell cultures. While other widely used cell surface labelling techniques (e.g., cell surface biotinylation, covalent cross-linking of receptor complexes, enzymatic cleavage of surface-expressed proteins using proteases) require the harvesting of cells for electrophoretic separation of membrane proteins followed by immunoblot analysis [12–14], cell-ELISA can be performed directly in cell culture plates. Although fluorescence-activated cell sorting (FACS) is more appropriate for the analysis of heterogenous cell populations in suspension, it is not suitable for the investigation of adherent cells. Our previous cell-ELISA studies of iGluR subtypes indicate that about 60% of α-amino-3-hydroxy-5-methyl-4-isoxazolepropionic acid receptors (AMPARs) and 70% of N-methyl-D-aspartic acid receptors (NMDARs) are surface-expressed in hippocampal neurons after 10 days in culture [3, 4]. This is consistent with the results of cell surface biotinylation, cross-linking, and proteolysis studies in cultured hippocampal neurons [12, 13] and in cerebellar granule cells [15]. Cell-ELISA proved to be a useful tool for the quantification of developmental changes in surface-expressed and total AMPARs and NMDARs [3] and N-ethylmaleimide-sensitive fusion protein (NSF)-dependent surface expression of AMPARs [2] and for the identification of LTP-induced changes in surface-expressed and total AMPAR immunoreactivity on cultured hippocampal neurons [4]. The advantages of cell-ELISA are the wide sensitivity range and the ability to process and analyze a large number of samples and different controls simultaneously. Therefore, it is suitable for initial screening of a wide range of experimental conditions before more targeted and labor-intensive techniques are used. However, this approach is less suitable for heterogenous cell populations and for antibodies that react only with denatured proteins.

2

Materials

2.1 Buffers and Solution

1. Phosphate-buffered saline (PBS), pH 7.4: Dissolve 8 g sodium chloride (137 mM NaCl), 0.2 g potassium chloride (2.7 mM KCl), 0.24 g potassium phosphate monobasic (1.8 mM KH2PO4), and 1.44 g anhydrous sodium phosphate dibasic (10 mM Na2HPO4) in 800 ml deionized water (dH2O). Adjust the pH to 7.4 with HCl and then add dH2O to 1 l. 10
concentrated stock solution can be prepared and diluted before use if larger quantities of PBS are required. 2. 4% (w/v) paraformaldehyde in PBS, pH 7.2: Dissolve 40 g paraformaldehyde in 800 ml PBS. Heat the solution to 80  C while stirring. While checking the pH, add 1 M NaOH dropwise until the opalescent solution becomes clear. Adjust pH to

50

Elek Molna´r

7.2 and add PBS to 1 l and filter the solution. Paraformaldehyde is toxic; handle it with care and follow local regulations. 3. 0.1 M glycine in PBS: Dissolve 7.5 g glycine in PBS and adjust pH to 7.4 with 1 M NaOH. 4. 3% (w/v) H2O2 in PBS: Add 9 ml of 35% (w/v) H2O2 to 91 ml PBS just before use. 5. Blocking solution: 5% (v/v) fetal bovine serum, 1% (w/v) bovine serum albumin (BSA) in PBS: Dissolve 10 g BSA in 800 ml PBS, add 50 ml fetal bovine serum, and bring volume to 1 l by adding PBS. 6. 1% (w/v) Triton X-100 in blocking solution: Add 1 g of Triton X-100 to 80 ml blocking solution, and bring volume to 100 ml with blocking solution. 7. 3,30 ,5,50 -Tetramethylbenzidine enzyme substrate for ELISA. 8. 1 M HCl: Add 8.3 ml of concentrated (37% w/w) HCl to 100 ml dH2O. 9. 0.5% (w/v) SDS in PBS: Dissolve 0.5 g SDS in 100 ml PBS. 10. Bicinchoninic acid assay (BCA) protein assay reagent.

3

Methods

3.1 Preparation of Cell Cultures

This protocol can be used to study changes in total or cell surface expression of proteins in adherent cells (e.g., transfected HEK293 or dissociated primary neuronal cultures) without detaching them. Culture dissociated primary neurons in poly-L-lysine coated 6-well plates [4] (see Note 1). HEK293 cells can be maintained and transfected in 6-, 12-, 24-, or 96-well plates [6–9, 16, 17]. Keep a few wells free of cells but exposed to the cell culture medium and all subsequent steps, because these will be needed to determine the level of nonspecific background immunolabelling (Subheadings 3.3 and 3.4) and protein content (Subheading 3.5) of samples.

3.2 Treatment of Neuronal Cultures

Various induction protocols can be used to manipulate the activity of neurons in cultures that will lead to changes in the cell surface delivery and internalization of GluRs (e.g., [4, 5, 18]). This part of the protocol is project specific and needs to be tailored to the study’s aims. For accurate measurements, use at least three parallel samples for each condition (see Note 2).

3.3 Immunolabelling of Cells for Cell-ELISA

Unless otherwise stated, use 2 ml/well solutions for 6-well plates or 200 μl/well for 96-well plates at room temperature. 1. Aspirate culture medium. 2. Gently rinse cells with PBS.

Cell-ELISA Analysis of Glutamate Receptors

51

3. Add 4% (w/v) paraformaldehyde in PBS for 20 min to fix the cells (see Notes 3–5). 4. Rinse cultures three times with PBS. 5. Incubate cells in 0.1 M glycine containing PBS for 30 min to neutralize any residual fixative. 6. Add 3% (w/v) hydrogen peroxide (H2O2) in PBS for 5 min to minimize endogenous peroxidase activity. 7. Gently wash cells twice in PBS for 5 min. 8. Incubate cultures in blocking solution (5% (v/v) fetal bovine serum and 1% (w/v) BSA in PBS) in the presence or absence of 1% (w/v) Triton X-100 for 30 min (see Note 6). 9. Label cells with primary antibodies (1 μg/ml) in blocking solution (0.5–1 ml for 6-well plates and 50–100 μl for 96-well plates) for 1 h (see Notes 7 and 8). 10. Wash cells three times with blocking solution (3
5 min). 11. Incubate cells with horseradish peroxidase (HRP)-conjugated secondary antibody (1: 3000–10,000) dilution in blocking solution (0.5–1 ml for 6-well plates and 50–100 μl for 96-well plates) for 1 h (see Note 8). 12. Wash cells four times in PBS (4
5 min). 3.4 HRP Reaction and Spectrophotometry

1. Aspirate wash solution, add HRP substrate containing reagent (3,30 ,5,50 -tetramethylbenzidine enzyme substrate; 0.8 ml for 6-well plates or 150 μl/well for 96-well plates), and incubate it for 10 min (see Note 9). 2. Transfer the blue-colored reaction end product containing solution from the plates into microfuge tubes or clean microtiter plates containing 1 M HCl (0.2 ml/well for 6-well plates or 50 μl/well for 96-well plates) to stop the reaction. Gently shake the tube or plate while the solution turns yellow. Keep the cell culture plate for protein measurement (Subheading 3.5). 3. Measure the absorbance of samples at 450 nm in a spectrophotometer or microplate reader.

3.5 Protein Measurement

1. After the incubation with the HRP substrate and transfer of reaction end product containing solution from the plates (Subheading 3.4), wash cells four times (4
5 min) in PBS. 2. Solubilize cells in 0.5% SDS (0.2 ml/well for 6-well plates or 50 μl/well for 96-well plates). 3. Measure protein content of solubilized samples using a BCA protein assay kit. 4. Subtract background immunoreactivity and protein content measured in cell-free wells (Subheading 3.1), and normalize

52

Elek Molna´r

each OD450 value of the cell-ELISA reaction to protein levels to take into account differences in cell seeding density. Calculate the average value for each experimental condition for comparisons.

4

Notes 1. It is difficult to maintain primary neurons in 96-well plates; therefore larger wells are preferable (e.g., 6-well plates). 2. Cell-ELISA provides information about the entire cell population, and it is less suitable for the detection of localized changes if only a relatively few neurons or transfected cells are affected. Therefore, chemical or pharmacological induction protocols that affect a large proportion of neurons and synapses are preferable. For example, long-term potentiation (LTP) can be induced in neuronal cultures by brief transient depolarization using high K+ (3
1 s application of 90 mM KCl), aminophosphonovaleric acid (APV) preconditioning, or stimulation of NMDARs by glycine (100–200 μM glycine, 5–10 min) in Mg2 + free medium (reviewed in [5]). Long-term depression (LTD) of synaptic activity can be induced by the activation of NMDARs (20 μM NMDA, 3 min) or group I mGluRs (100 μM ((RS)-3,5-DHPG, 10 min) [18]. 3. The paraformaldehyde fixation prevents the detachment of cells from the cell culture plate throughout the experiment. The integrity of the plasma membrane in paraformaldehydefixed cells can be confirmed, using antibodies that are specific to intracellular epitopes (e.g., [4]). These antibodies are unable to interact with the intracellular epitopes in paraformaldehydefixed cells without Triton X-100 permeabilization. 4. Some epitopes may be altered by paraformaldehyde fixation, which can lead to poor immunolabelling. In this case a different antibody or alternative fixatives should be considered [19]. Using methanol or acetone as fixative will permeabilize cell membranes, which precludes the selective investigation of surface proteins. 5. Cell-ELISA of surface proteins can be performed using unfixed cell cultures. In this case, steps 3–5 and the addition of 1% Triton X-100 at step 8 need to be omitted (Subheading 3.3). It is preferable to perform the antibody labelling of unfixed cells at low temperature (e.g., at 4  C in a cold room) to minimize the endocytosis of cell surface proteins and antibodies [18]. The compatibility of cell-ELISA with unfixed adherent cell cultures provides an exciting opportunity for the analysis of anti-iGluR autoantibodies that react with conformational epitopes and implicated in various neurological disorders [20, 21].

Cell-ELISA Analysis of Glutamate Receptors

53

6. Without Triton X-100 antibodies will label epitopes on the cell surface. The permeabilization of fixed cell membranes with Triton X-100 detergent will enable antibodies to access both surface and intracellular receptor proteins. 7. Antibodies against extracellular epitopes are required for surface labelling. If suitable antibodies are not available, an epitope tag can be introduced into the extracellular domain of the target protein to facilitate immunochemical detection [22]. 8. A range of primary and secondary antibody concentrations need to be tested in preliminary experiments to optimize labelling. Control experiments with pre-immune serum or wells without cells need to be included routinely to determine background value, which should be subtracted from the OD450 readings (Subheading 3.5). 9. We use K-Blue substrate from Neogen Corporation. The absorbance of the 3,30 ,5,50 tetramethylbenzidine (TMB) and hydrogen peroxide (H2O2) containing K-Blue substrate HRP reaction product can be measured directly (without adding HCl to stop the HRP reaction) at 620 nm. While the K-Blue substrate is more stable and highly sensitive and produces lower background, alternative protocols are available for the preparation of TMB substrate solutions for cell-ELISA using basic ingredients [11].

Acknowledgment This research was supported by grant from the Biotechnology and Biological Sciences Research Council, UK (Grant BB/J015938/1). Conflict of Interest Disclosure: EM is a member of the Scientific Advisory Board of Hello Bio [www.hellobio.com]. References 1. Molna´r E (2018) Glutamate receptors. In: Choi S (ed) Encyclopedia of signaling molecules, 2nd edn. Springer, New York, pp 2138–2146 2. Noel J, Ralph GS, Pickard L, Williams J, Molnar E, Uney JB, Collingridge GL, Henley JM (1999) Surface expression of AMPA receptors in hippocampal neurons is regulated by an NSF-dependent mechanism. Neuron 23:365–376 3. Pickard L, Noe¨l J, Henley JM, Collingridge GL, Molna´r E (2000) Developmental changes in synaptic AMPA and NMDA receptor distribution and AMPA receptor subunit

composition in living hippocampal neurons. J Neurosci 20:7922–7931 4. Pickard L, Noel J, Duckworth JK, Fitzjohn SM, Henley JM, Collingridge GL, Molna´r E (2001) Transient synaptic activation of NMDA receptors leads to the insertion of native AMPA receptors into hippocampal neuronal plasma membrane. Neuropharmacology 41:700–713 5. Molna´r E (2011) Long-term potentiation in cultured hippocampal neurons. Semin Cell Dev Biol 22:506–513 6. Rutter AR, Freeman FM, Stephenson FA (2002) Further characterization of the molecular interaction between PSD-95 and NMDA

54

Elek Molna´r

receptors: the effect of the NR1 splice variant and evidence for modulation of channel gating. J Neurochem 81:1298–1307 7. Papadakis M, Hawkins LM, Stephenson FA (2004) Appropriate NR1-NR1 disulfide-linked homodimer formation is requisite for efficient expression of functional, cell surface N-methylD-aspartate NR1/NR2 receptors. J Biol Chem 279:14703–14712 8. Cousins SL, Innocent N, Stephenson FA (2013) Neto1 associates with the NMDA receptor/amyloid precursor protein complex. J Neurochem 126:554–564 9. Horak M, Suh YH (2015) Counting NMDA receptors at the cell surface. NeuroMethods 106:31–44 10. Parnas D, Linial M (1998) Highly sensitive ELISA-based assay for quantifying protein levels in neuronal cultures. Brain Res Protocol 2:333–338 11. Lourenc¸o EV, Roque-Barreira MC (2010) Immunoenzymatic quantitative analysis of antigens expressed on the cell surface (cellELISA). Methods Mol Biol 588:301–309 12. Hall RA, Soderling TR (1997) Quantitation of AMPA receptor surface expression in cultured hippocampal neurons. Neuroscience 78:361–371 13. Hall RA, Soderling TR (1997) Differential surface expression and phosphorylation of the N-methyl-D-aspartate receptor subunits NR1 and NR2 in cultured hippocampal neurons. J Biol Chem 272:4135–4140 14. Ball SM, Atlason PT, Shitu-Balogun OO, Molna´r E (2010) Assembly and intracellular distribution of kainate receptors is determined by RNA editing and subunit composition. J Neurochem 114:1805–1818 15. Huh K-H, Wenthold RJ (1999) Turnover analysis of glutamate receptors identifies a rapidly degraded pool of the N-methyl-D-aspartate receptor subunit, NR1, in cultured cerebellar granule cells. J Biol Chem 274:151–157 16. Gallyas F, Ball S, Molna´r E (2003) Assembly and cell surface expression of KA-2 subunit-

containing kainate receptors. J Neurochem 86:1414–1427 17. Atlason PT, Scholefield CL, Eaves RJ, MayoMartin MB, Jane DE, Molna´r E (2010) Mapping the ligand binding sites of kainate receptors: molecular determinants of subunitselective binding of the antagonist [3H] UBP310. Mol Pharmacol 78:1036–1045 18. Gladding CM, Collett VJ, Jia Z, Bashir ZI, Collingridge GL, Molna´r E (2009) Tyrosine dephosphorylation regulates AMPAR internalisation in mGluR-LTD. Mol Cell Neurosci 40:267–279 19. Molna´r E (2013) Immunocytochemistry and immunohistochemistry. In: Langton PD (ed) Essential guide to reading biomedical papers: recognising and interpreting best practice. Wiley-Blackwell, Somerset, pp 117–128 20. Levite M (2014) Glutamate receptor antibodies in neurological diseases: anti-AMPAGluR3 antibodies, anti-NMDA-NR1 antibodies, anti-NMDA-NR2A/B antibodies, antimGluR1 antibodies or anti-mGluR5 antibodies are present in subpopulations of patients with either: epilepsy, encephalitis, cerebellar ataxia, systemic lupus erythematosus (SLE) and neuropsychiatric SLE, Sjogren’s syndrome, schizophrenia, mania or stroke. These autoimmune anti-glutamate receptor antibodies can bind neurons in few brain regions, activate glutamate receptors, decrease glutamate receptor’s expression, impair glutamateinduced signaling and function, activate blood brain barrier endothelial cells, kill neurons, damage the brain, induce behavioral/psychiatric/cognitive abnormalities and ataxia in animal models, and can be removed or silenced in some patients by immunotherapy. J Neural Transm 121:1029–1075 21. Venkatesan A, Adatia K (2017) Anti-NMDAreceptor encephalitis: From bench to clinic. ACS Chem Neurosci 8:2586–2595 22. McIlhinney RA (2004) Generation and use of epitope-tagged receptors. Methods Mol Biol 259:81–98

Part II Function/Cellular Plasticity

Chapter 5 Preparation of Organotypic Slice Cultures for the Study of Glutamate Receptor Function Andres Barria Abstract Organotypic slice cultures enable the study of glutamate receptors in an environment closely related to the in vivo situation but with easy access to genetic manipulation of the receptors and regulatory signaling cascades as well as a more precise pharmacological intervention. We describe a method to prepare organotypic hippocampal slices that can be easily adapted to other brain regions. Brain slices are laid on porous membranes, and culture medium is allowed to form an interface. This method preserves the functionality and gross architecture of the hippocampus for up to 2 weeks in culture. Key words Brain slices, Organotypic cultures, Glutamate receptors, Synaptic transmission, Synaptic physiology

1

Introduction Neuronal circuits underlying the processing of sensory inputs, memory formation, and all sorts of behaviors are the result of finely sculpted synaptic connections among neurons. Synaptic connectivity in the brain, in turn, is the result of a delicate balance between synaptogenesis, synaptic plasticity, and removal of synapses. Changes in synaptic connectivity drive the refinement of neuronal circuits during development [1] and are thought to underlie the formation of new memories [2]. Disruption of this balance has been linked to abnormal brain development and neuropsychiatric disorders such as Alzheimer’s disease and schizophrenia [3]. In all these processes, glutamate receptors and synaptic activity play a critical role. While dissociated cultured neurons are a widely used system to interrogate the properties, regulation, and function of glutamate receptors, preparation of such cultures involves de novo formation of dendritic and axonal processes and the establishment of synapses under unpatterned synaptic activity that can affect glutamate receptors in an unsuspected manner. This is particularly important when

Corinna Burger and Margaret Jo Velardo (eds.), Glutamate Receptors: Methods and Protocols, Methods in Molecular Biology, vol. 1941, https://doi.org/10.1007/978-1-4939-9077-1_5, © Springer Science+Business Media, LLC, part of Springer Nature 2019

57

58

Andres Barria

studying issues like protein expression, synaptic insertion, and regulation of glutamate receptors because they are heavily dependent on levels of correlated synaptic activity; therefore when studying the properties, regulation, and function of synaptic glutamate receptors, a system closely related to the in vivo environment where these receptors exist is highly desirable. Organotypic cultures maintain synaptic organization that is critical to understanding synapse function in a more naturalistic context than routine culture dissociated neurons methods. In addition this system has the advantage that genes can be easily manipulated, and a more precise pharmacological intervention is possible than in vivo [4, 5]. Here we present a method to prepare and culture hippocampal slices that can be easily adapted to other brain regions [6]. This method allows easy genetic manipulation of glutamate receptors using different approaches like viral infection [7, 8] or biolistics [9]. In addition, slices can be easily recovered for biochemical assays [10] or transferred to microscopes for imaging [11] or electrophysiological experiments [12–15].

2

Materials All solutions must be prepared with autoclaved deionized water and handled inside a laminar flow hood appropriated for tissue culture. All the surgical equipment must be clean and wiped off with ethanol and placed under a UV lamp in the tissue culture hood at least 30 min before the procedure.

2.1

Solutions

1. Phenol red stock solution: 0.5% (v/v) phenol red in Dulbecco’s phosphate-buffered saline (DPBS). 2. L-Glutamine stock solution: 200 mM L-glutamine. Store at 20  C in aliquots of 2.5 mL. 3. Ascorbic acid stock solution: 25% (w/v) ascorbic acid. Store at 20  C in aliquots of 100 μL. 4. Insulin stock solution: 1 mg/mL insulin dissolved in HCl 0.01 N. 5. Low Na+ artificial cerebrospinal fluid (ACSF)-dissecting solution: 1 mM CaCl2, 10 mM D-glucose, 4 mM KCl, 5 mM MgCl2, 26 mM NaHCO3, 234 mM sucrose, and 0.1% v/v phenol red as a pH indicator. Use deionized and sterile H2O. Stir the solution ~30 min until everything has thoroughly dissolved. Sterilize the solution by filter sterilization using a 0.22 μm filter. Make 50 mL aliquots and store at 4  C no longer than 2 months.

Organotypic Slice Cultures

59

6. Slice culture medium (SCM): 8.4 g/L minimum essential Eagle medium (MEM), 20% horse serum heat inactivated (see Note 1), 1 mM L-glutamine, 1 mM CaCl2, 2 mM MgSO4, 1 mg/L insulin, 0.00125% ascorbic acid, 13 mM D-glucose, 5.2 mM NaHCO3, and 30 mM Hepes. It is critical this solution is prepared in a sterile manner with deionized and sterile H2O and kept that way. Mix solution until thoroughly dissolved and bring to room temperature. Adjust pH to 7.27–7.28 with 1 N NaOH. Measure osmolarity and adjust it to 320 mmol/kg with deionized and sterilized H2O. Expect to add approximately 25–40 mL. Check osmolarity again. The pH may change slightly while adjusting osmolarity, and this is ok; it is more important that the osmolarity is in the correct range (317–323). Correct osmolarity is crucial for healthy slices. Sterilize the solution by filter sterilization using a 0.22 μm filter. Make 20 mL aliquots and store for up to 2–3 weeks at 4  C. 2.2 Dissecting Tools and Culture Materials

1. Tissue slicer. 2. Teflon sheet 0.78 mm thick. 3. Hippocampus dissecting tool. 4. Perfection large utility scissors. 5. Laminectomy forceps Dumont #2. 6. Dissecting needle, single-cutting edge. 7. Iris spatula, curved. 8. Iris spatula, straight. 9. Small dissecting scissors. 10. Rounded/straight spatula. 11. Six-well plates. 12. Cell culture inserts.

3

Methods

3.1 Initial Preparation

1. Prepare the tissue slicer by placing a new piece of Teflon sheet 1/32 inches thick on the dissecting bed and mounting a new standard double-edged razor blade (Fig. 1). Use tape to keep the Teflon sheet in place. It is important the Teflon sheet to be flat. 2. Wipe the tissue culture hood with 70% ethanol, and set the dissecting microscope inside. Sterilize the hood, microscope, tissue slicer, and all dissecting instruments for at least 15 min with UV light (Fig. 2). 3. Prepare six-well tissue culture plates. Add 750 μL slice culture media (SCM) per well, and place cell culture inserts in each

60

Andres Barria

Fig. 1 Tissue slicer with Teflon sheet used to rapidly produce thin slices of living brain

Fig. 2 Tools used for hippocampus dissection and separation of slices after cutting

well. Make sure the membranes are thoroughly wet with no bubbles underneath. Place the plates in the incubator at 35  C gassed with 5% CO2 until needed (Fig. 3). 4. Pour 50 mL low Na+ ACSF (dissecting solution) into a 100 mL beaker, and place it on ice-salt mix. Bubble the low Na+ ACSF with 5% CO2/95% O2 until color changes and ACSF forms a slurry mix of frozen and liquid solution (10–20 min). 5. Get a postnatal age 5 or 7 (P5–P7) rat pup. Up to three pups can be used.

Organotypic Slice Cultures

61

Fig. 3 Tissue culture materials: six-well plate and cell culture inserts

3.2 Hippocampal Slice Preparation (See Note 2)

Use personal protective equipment most importantly gloves and face mask to prevent contamination of the tissue (see Note 3). 1. Cut the head of the animal with sharp utility scissors. Cut the skin and expose the skull. Open the skull by cutting from side to side along the interaural line and then along the sagittal suture with small scissors. An optional cut from side to side in the front can be made to facilitate removal of the bones and exposure of the brain. Scoop out the brain quickly with a rounded spoon micro spatula, and place it in the slurry of dissecting solution to chill for ~1 min. Pour ~10 mL of ice-cold dissecting solution onto a 60 mm dish, and transfer the brain from the beaker to the dish. The brain should be covered with dissecting solution (see Note 4). 2. Move the tissue into the tissue culture hood. Place the brain under the dissecting microscope, and hold it at the midline with the dissecting forceps pressed to the bottom of the 60 mm dish. Use the hippocampus dissecting tool to separate the hemispheres leaving out the midbrain. The hippocampi are then exposed on each hemisphere. Then gently scoop the hippocampus out with the hippocampus dissecting tool. Use the dissecting needle to completely isolate the hippocampus, and clean it as much as possible. 3. Using a snipped tip of a 1000 μL filter pipette tip (P1000), gently aspirate the hippocampus, and transfer it to the Teflon sheet on the tissue slicer. Position the hippocampus on its concave side. 4. Align the hippocampi perpendicular to the blade to obtain coronal sections, and drain excess liquid. 5. Slice the hippocampi every 300–400 μm.

62

Andres Barria

6. Pour ~10 mL cold SCM into a 60 mm dish, and transfer sliced hippocampi from the slicer using another snipped P1000 filter tip and cold SCM. Avoid making bubbles. 7. With the help of an iris spatula and a straight spatula, gently separate the slices from each other. 8. Separate well-defined and undamaged slices from damaged slices. 3.3 Hippocampal Slice Culture

1. Bring the six-well plates with SCM and cell culture inserts from the incubator into the tissue culture hood. With the help of another snipped P1000 filter tip, transfer individual slices onto the membrane. Place 4–5 slices per membrane. Be careful not to place the slices either close to the insert wall or close to each other. When necessary, use iris spatula to separate slices. Remove excess medium. Touch slices as little as possible once they are on the membrane (see Note 5). 2. Move plate back to incubator and culture at 35  C and 5% CO2. 3. Change SCM every 48 h inside the tissue culture hood by aspirating the SCM with a Pasteur pipette. Add 750 μL of fresh pre-warmed SCM per well. Make sure no bubbles are formed under the membrane (see Note 6).

4

Notes 1. Different serum sources can influence the quality of the slices. We recommend testing several batches first. 2. This method can be adapted to other brain regions provided that the density of the tissue allows proper oxygenation and nutrient penetration. 3. This method is based on the method first described by Stoppini et al. [16] and offers a rapid manner to culture hippocampal slices. The most important aspect of this protocol is to maintain slices sterile; therefore it is critical to use appropriate sterile techniques and to properly disinfect and sterilize all the materials in contact with the tissue. If contamination is a recurrent problem, check incubator and tissue culture hood for possible sources of contamination. Proper use of sterile techniques during the whole procedure is essential. Contamination is easily detected by moving black specks in the medium or turbidity of the SCM. When placed under the microscope, the surface of the slice should look clean after 4 days in culture with clear and discernibly cell bodies. If no clear cell bodies are seen and much debris covers the surface after 4 days, then is not a healthy slice.

Organotypic Slice Cultures

63

4. The total time from decapitation to placing the slices on the membrane and in the incubator should be no longer than 1.5 h. If the procedure takes too long, it will compromise the health of the slices. 5. Placing the slices on a porous membrane warranties proper oxygenation and nutrition via a thin layer of SCM that is formed by capillarity. Thus, tissue density limits this method to young tissue. For hippocampus, slices 300–400 μm thick from p6–p7 animals seem to give best results. This type of slices can rapidly be obtained with a tissue slicer diminishing the time the tissue is exposed to the air. 6. Importantly for those studying synaptic physiology, after a few days in culture, all the debris from dead cells has been removed, leaving a clean surface highly suitable for electrophysiological or imaging experiments. In addition, organotypic hippocampal slices continue developing normal connectivity comparable to acute slices [17]. However after 2 weeks in culture, this normal connectivity disappears as neurons start forming too many connections that increases synaptic activity in the slice. References 1. Cline H, Haas K (2008) The regulation of dendritic arbor development and plasticity by glutamatergic synaptic input: a review of the synaptotrophic hypothesis. J Physiol 586 (6):1509–1517 2. Martin SJ, Grimwood PD, Morris RG (2000) Synaptic plasticity and memory: an evaluation of the hypothesis. Annu Rev Neurosci 23:649–711 3. Lau CG, Zukin RS (2007) NMDA receptor trafficking in synaptic plasticity and neuropsychiatric disorders. Nat Rev Neurosci 8 (6):413–426 4. Barria A, Malinow R (2002) Subunit-specific NMDA receptor trafficking to synapses. Neuron 35(2):345–353 5. Barria A, Malinow R (2005) NMDA receptor subunit composition controls synaptic plasticity by regulating binding to CaMKII. Neuron 48(2):289–301 6. Sanchez JT, Seidl AH, Rubel EW, Barria A (2011) Preparation and culture of chicken auditory brainstem slices. J Vis Exp 49:2527 7. Malinow R et al (2010) Introduction of green fluorescent protein (GFP) into hippocampal neurons through viral infection. Cold Spring Harb Protoc 4:pdb.prot5406 8. Shi S, Hayashi Y, Esteban JA, Malinow R (2001) Subunit-specific rules governing AMPA receptor trafficking to synapses in

hippocampal pyramidal neurons. Cell 105 (3):331–343 9. Woods G, Zito K (2008) Preparation of gene gun bullets and biolistic transfection of neurons in slice culture. J Vis Exp 12:e675 10. McQuate A, Latorre-Esteves E, Barria A (2017) A Wnt/calcium signaling cascade regulates neuronal excitability and trafficking of NMDARs. Cell Rep 21(1):60–69 11. Gambrill AC, Barria A (2011) NMDA receptor subunit composition controls synaptogenesis and synapse stabilization. Proc Natl Acad Sci U S A 108(14):5855–5860 12. Cerpa W, Gambrill A, Inestrosa NC, Barria A (2011) Regulation of NMDA-receptor synaptic transmission by Wnt signaling. J Neurosci 31(26):9466–9471 13. Cerpa W, Latorre-Esteves E, Barria A (2015) RoR2 functions as a noncanonical Wnt receptor that regulates NMDAR-mediated synaptic transmission. Proc Natl Acad Sci U S A 112 (15):4797–4802 14. Mattison HA, Hayashi T, Barria A (2012) Palmitoylation at two cysteine clusters on the C-terminus of GluN2A and GluN2B differentially control synaptic targeting of NMDA receptors. PLoS One 7(11):e49089 15. Storey GP, Opitz-Araya X, Barria A (2011) Molecular determinants controlling NMDA

64

Andres Barria

receptor synaptic incorporation. J Neurosci 31 (17):6311–6316 16. Stoppini L, Buchs PA, Muller D (1991) A simple method for organotypic cultures of nervous tissue. J Neurosci Methods 37(2):173–182

17. De Simoni A, Griesinger CB, Edwards FA (2003) Development of rat CA1 neurones in acute versus organotypic slices: role of experience in synaptic morphology and activity. J Physiol 550(Pt 1):135–147

Chapter 6 Glutamate Receptor Probing with Rapid Application and Solution Exchange (RASE) Nathanael O’Neill and Sergiy Sylantyev Abstract Probing of glutamate receptors at sub-millisecond time scale is a key element needed for understanding of their response kinetics, neural signal transduction, and, in a wider context, intercellular cross talk. One of the classical techniques used to obtain this type of data in electrophysiology is placing a recording pipette in front of a double-barrel solution application pipette, which provides a rapid switch between two solutions. Here we describe a modification of this classical technique, which utilizes a solution application pipette with several loading capillaries. Such a system is capable of replacing multiple application solutions within 7–12 s time intervals. This modified protocol enables ultrafast application of several solutions at the same set of receptors, thus allowing powerful paired-sample statistical approaches. In addition, the same experimental equipment fabricated for the ultra-resolution probing of receptor kinetics can also be applied in several other types of electrophysiological experiments. Key words Non-equilibrium pharmacokinetics, Electrophysiology, Rapid solution exchange, Membrane patches

1

Introduction Activation and deactivation of glutamate receptors in the synaptic cleft is a core mechanism of excitatory neurotransmission. Therefore, gaining a precise knowledge of synaptic glutamate receptor kinetics is key to understanding how information is transferred, processed, and stored by brain neural circuits. Following the release of neurotransmitters from the presynaptic bouton, there is a rapid spike of their concentration (~1.1 mM for glutamate) [1] within the synaptic cleft, which decreases rapidly over a 200–300 μs time interval to low levels. Importantly, this transient increase in neurotransmitter concentration takes place over a much shorter period than the characteristic opening time of ionotropic glutamate receptors; thus, activation of the receptors in living neural tissue occurs under substantially non-equilibrium conditions [1, 2]. Consequently, classical pharmacokinetic models

Corinna Burger and Margaret Jo Velardo (eds.), Glutamate Receptors: Methods and Protocols, Methods in Molecular Biology, vol. 1941, https://doi.org/10.1007/978-1-4939-9077-1_6, © Springer Science+Business Media, LLC, part of Springer Nature 2019

65

66

Nathanael O’Neill and Sergiy Sylantyev

and experimental approaches implying equilibrated conditions (such as measurements of dissociation constant Kd) are unlikely to accurately depict receptor activation/deactivation processes as they occur inside the synaptic cleft. Instead, physiologically relevant measures of receptor activation/deactivation require the sub-millisecond application of neurotransmitter, so as to most closely reproduce intrasynaptic conditions. To achieve sub-millisecond application times in excised outside-out and nucleated cell membrane patches, a system based on piezo-driven double-barrel (θ-glass) micropipette was developed by Colquhoun and coauthors [3], termed rapid solution application. This system allows for the ultrarapid application of one experimental solution (with washout solution in other θ-glass pipette channel). Over the last 20 plus years, numerous studies have demonstrated rapid solution application to be highly effective, particularly at ionotropic glutamate receptors [4–6]. Using this experimental approach, it is possible to achieve a minimum response limit of ~20 μs [7]—well below the characteristic time course of neurotransmitter in a synaptic cleft (200–300 μs). However, a notable shortcoming with the aforementioned approach is the inability to rapidly apply more than one pharmacological agent on the same pool of receptors; the experimenter is limited to two θ-glass channels: one for the pharmaceutical agent and the other for the washout solution. In practice, this means that an experimenter is unable to apply a paired-statistical approach when studying the action of more than one compound at a receptor of interest, thus severely hindering the ability to detect subtle changes in receptor function. A superior approach would be to combine an ultrafast application time course with the ability to apply different solutions to the same patch of membrane. Repeated pharmacological manipulation on the same receptor pool (i.e., receptors comprised in one membrane patch) is of great advantage to experimental design, permitting the use of paired-sample statistics, which substantially reduce the number of individual experiments sufficient for credible conclusion. To meet these requirements, we added several important modifications to the rapid solution application technique introduced by Colquhoun and coauthors and termed it thereafter a rapid application and solution exchange (RASE) protocol. First, we added several solution supply capillaries to each channel of θ-glass pipette (Fig. 1a), enabling channel solutions to be changed multiple times during the same experiment. Second, we reduced a dead volume of the pipette, in order to shorten the time interval needed for full replacement of solutions in the θ-glass pipette channels, thereby allowing us to work easily within the lifetime of cell membrane patch (1–5 min) (Fig. 1a, b). Finally, we developed a closed pressure control system, which allows separate adjustment for each θ-glass barrel during an experiment (Fig. 2).

Sub-millisecond Switch of Multiple Activating Solutions

67

Fig. 1 Preparation of the rapid solution application pipette for RASE. (a) Schematic of the solution application θ-glass pipette prepared for RASE (not to scale; the outer pipette diameter ~1.5 mm, the open tip diameter ~200 μm). (1) θ-glass septum; (2) section of pipette barrels filled with glue; (3) solution supply microfilaments; (4) dead volume near the pipette tip. (b) Pipette preparation steps. (1) the pipette is placed under the microscope; (2) three supply microfilaments are inserted into the right channel; (3) an additional microfilament is connected to the syringe with glue introduced (arrow); (4–6) stages depicting the right channel being filled with glue; (7) three supply microfilaments inserted into the left channel; (8) both channels are partially filled with glue. (c) Pipette with the solution supply microfilaments fixed inside; three filaments and three connectors are marked black and white (unmarked) to distinguish between the two channels. (d) Solution application pipette mounted on a piezo-actuator, which is connected to the rod-shaped microelectrode holder. Insets 1–3 (also the dotted circle in the top image): the plastic pipette holder made of a three-way tube connector, with one branch cut off to provide a channel for the tightening screw, as shown (a small piece of silicone inserted in the cutoff channel can help to dampen vibration). Reprinted by permission from Macmillan Publishers Ltd: Nature Protocols. Sylantyev, S. and D. A. Rusakov. “Sub-millisecond ligand probing of cell receptors with multiple solution exchange.” Nat Protoc 8(7): 1299–1306, © 2013

Fig. 2 The arrangement of multiple solution supply and exchange. (a, b) Two sets of three liquid containers (plastic cylinders) connected to the pressure supply at one end (3) and to the two θ-glass pipette channels at the other end (shown below in c). A connector inlet (1) is connected to the solution-supplying syringe in a and closed in b. An air outlet (a thick syringe needle embedded into the container plastic; (2)) is open during solution supply in a and closed in b. Gas pressure lines (3) connect the two container groups to the pneumatic ejector or micropump outlets (not shown). (c) A rapid solution application assembly mounted near the recording chamber under a microscope objective including the θ-glass pipette (1), the piezo-actuator (2), and an electrode holder (3), which is set in a position above the recording chamber; six solution supply filaments are connected to the solution supply lines (4) coming from the six corresponding containers (shown in a, b). Inset: (5) a 2-μm microfilter inserted into the solution supply line to buffer pressure jumps in the system while providing additional solution filtration. Reprinted by permission from Macmillan Publishers Ltd: Nature Protocols. Sylantyev, S. and D. A. Rusakov. “Sub-millisecond ligand probing of cell receptors with multiple solution exchange.” Nat Protoc 8(7): 1299–1306, © 2013

Sub-millisecond Switch of Multiple Activating Solutions

69

Full RASE protocol was then validated on outside-out and nucleated membrane patches [8, 9] (Fig. 3). Further, rapid solution exchange with RASE hardware can also be used in experimental protocols where rapid solution application is not required. For instance, this approach has proved useful to establish sensitivity of the single-channel receptor currents to endogenous ligand and exogenous antagonists on the same receptor pool [11] (Fig. 4b). In addition, rapid solution application without rapid solution exchange has allowed researchers to investigate the summation and coincidence effects of neural responses, such as evoked postsynaptic potentials (EPSPs) at precisely controlled time intervals, as illustrated below (Fig. 5). Moreover, combining recordings from outside-out patches with recordings from nucleated patches, RASE allows for the experimental separation of the effects pertinent to the intracellular machinery of receptor action (Fig. 4a). However, there are also several potential limitations of RASE. First, in the case of nerve cells, this and similar rapid application techniques are designed to probe predominantly extrasynaptic receptors, which functional profile may differ from that of synaptic receptor population [12, 13]; nevertheless, probing of synaptic receptors with RASE protocol is still possible in some specific preparations [8]. Second, because of the variable characteristics of θ-glass pipettes (profile of the tip, size of the dead space inside, widths of the two barrels, etc.), these pipettes should be individually calibrated to find a suitable pressure in each channel, the amplitude and duration of the electric switching pulse, and the recording pipette position. Third, in nucleated patches or cultured cells attached to substrate, the turbulence at the side opposite to the application side could have a variable (albeit relatively small) effect on the kinetics of recorded receptor responses. In our experience, however, this concomitant effect could be satisfactorily averaged out with an increased number of recorded individual traces. Finally, the RASE protocol is quite technically demanding, requiring substantial hands-on experience in standard electrophysiological patchclamp techniques, especially in work with outside-out and nucleated patches.

2

Materials

2.1 In Vitro Preparation

The present protocol can be used with a variety of in vitro preparations suitable for whole-cell recordings, from acute brain slices to cell cultures. It is important to note that when membrane patches are pulled from acute tissue, usually only one outside-out or nucleated patch is recorded in each individual brain slice with the RASE protocol to avoid both contamination of the perfusion solution and accumulation of applied drugs in the tissue.

70

Nathanael O’Neill and Sergiy Sylantyev

Fig. 3 Glutamate receptors recorded under full RASE protocol (rapid solution application + rapid solution exchange). (a) AMPA-receptor currents evoked in outside-out patch excised from CA1 hippocampal cell by 100 μs application of 1 mM glutamate. Increasing concentrations of AMPAR antagonist γ-DGG are applied in successive trials as indicated. Four filaments per θ-glass pipette barrel were used in this experiment (each containing ACSF only, ACSF + 1 mM glutamate and ACSF + 1 mM glutamate + one concentration of γ-DGG), and the direction of the rapid solution switch was changed accordingly (reflected in the mirrored piezo-switch deflections) in order to optimize the use of solution supply channels; further details and explanations of the corresponding experimental paradigm were published in earlier work [10]. To avoid any residual receptor desensitization, intertrial intervals in these and similar experiments were 10–15 s. (b) NMDA-receptor currents evoked in nucleated patch by 100 μs application of 100 μM glutamate + 100 μM glycine mixture at different holding voltages, as indicated (Mg2+-free bath solution used)

Sub-millisecond Switch of Multiple Activating Solutions

71

Fig. 4 Use of RASE toolset in experimental protocol with rapid solution exchange only. (a) Metabotropic glutamate receptors group I (mGluRI) potentiate effect of NMDARs via cytoplasmic signalling chains. Top: nucleated patch. Bottom: outside-out patch. Patches were pulled from the same cell to decrease inter-sample variance. 5 s application of 50 μM NMDA + 50 μM glycine activates NMDARs only; application of 50 μM glutamate + 50 μM glycine activates both NMDARs and mGluRIs. mGluRI blockers MPEP (200 nM) and LY-456236 (1 μM) downregulate mGluRI effect in nucleated patch with preserved cytoplasmic milieu whereas have no effect in outside-out patch where cytoplasmic signalling chains are destroyed. (b) Single-channel NMDARs openings evoked by increasing concentration of agonists applied on the same nucleated patch. From top to bottom: 10 nM glutamate + 10 nM glycine; 50 nM glutamate + 50 nM glycine; 500 nM glutamate + 500 nM glycine; 1 μM glutamate + 1 μM glycine 2.2

Reagents

1. Standard salts and reagents for preparing electrophysiological solutions: CaCl2, glucose, HEPES, KCH3O3S, KCl, KOH, MgCl2, MgSO4, Na2-ATP, NaCl, Na-GTP, NaHCO3, NaH2PO4, Na2-Phosphocreatine. 2. Deionized water, at least 15 MΩ/cm at 25  C; note that water quality is important for recording pipette solutions. Receptor ligands and drugs should be obtained from reliable sources in accordance with the experimental design.

2.3

Equipment

1. A standard setup for electrophysiology, including a submergedtype recording chamber (for acute slices or cell cultures) equipped with 20–60 IR differential interference contrast (DIC) visualization. 2. A patch-clamp amplifier capable of current and voltage clamp, multichannel digitizer, constant-voltage stimulus isolator, and

72

Nathanael O’Neill and Sergiy Sylantyev

Fig. 5 The use of RASE toolset in experimental protocol with rapid solution application only. Summation of EPSPs leads to action potential (AP) generation. EPSPs evoked in cerebellar granule cell in disperse culture by application of 50 μM glutamate + 50 μM glycine. Left panel, from top to bottom: ligand mixture applied with 20, 15, and 10 ms intervals. Right panel: ligands mixture applied with 7 ms interval (top trace) and 5 ms interval (bottom trace, AP peaks are truncated)

double-channel pneumatic drug ejector or micropump supplied by compressed inert gas (nitrogen or helium). 3. Patch electrodes pulled from borosilicate glass. Pipettes of 5–6 MΩ resistance appear to be well suited for experiments on outside-out patches, whereas for nucleated patches the optimal pipette resistance is somewhat lower (4–5 MΩ); fire polishing of pipette ends substantially increases the lifespan of a patch. 4. Borosilicate θ-capillary glass. 5. Solution loading microfilaments. 6. Bender piezoelectric actuators with stranded wires. 7. Rapidly hardening glue (e.g., cyanoacrylate glue).

3

Methods

3.1 Assembling Rapid Application Pipette

1. Pull the solution application pipette with a ~200 μm tip from θ-glass capillary using a standard pipette puller, and fix it under a low-magnification binocular microscope, e.g., using blue tack. 2. Insert the supply microfilaments into one barrel, and fix them in a position, e.g., with a piece of blue tack placed near the pipette blunt end (Fig. 1b).

Sub-millisecond Switch of Multiple Activating Solutions

73

3. Insert another freely moving microfilament connected to a syringe filled with glue (Fig. 1b). 4. Gently eject the glue by applying pressure with the syringe; stop ejection when the glue approaches the ends of the supply microfilaments (Fig. 1b). The stop point should be at least 1.5–2 mm away from the ends of the supply microfilaments. This is to prevent drawing of the glue inside the filaments by capillary forces (see Note 1). 5. Slowly remove the filament connected to the syringe with glue; if you experience difficulty, leave it inside the barrel, and later cut at a pipette end. 6. Repeat steps 2–5 for the other barrel of the θ-glass pipette (Fig. 1b). 7. Leave the pipette for several hours (overnight) to let the glue set inside the barrels (Fig. 1c). 8. Glue the piezo-actuator to the holding rod, and glue a pipette holder of an appropriate size to the side of the actuator (Fig. 1d). 3.2 Assembling the Pressurized Solution Supply System

1. Prepare containers for the application solutions (e.g., use syringes for gravity-driven perfusion systems or plastic containers from a standard blood dialysis kit for closed pump-operated systems) in accordance with the number of supply microfilaments in the θ-glass pipette (Fig. 2a, b). Equip each such container with a tap connector and an outlet with an airtight cap, e.g., by using thick syringe needles glued or melted into the container body plastic (Fig. 2a, b). Place containers in a suitable position near the recording setup. 2. Connect two groups of containers (each to supply one of the θ-glass pipette channels) to the gas supply lines using thin tubing (0.8 mm tubing, but see Note 2). 3. In accordance with the available micropump hardware, connect the pneumatic ejector inlets to the source of pressurized inert gas (nitrogen), and connect the outlets to the lines supplying pressure for the two groups of containers (Fig. 2a, b). 4. Clamp the piezo-actuator rod in microelectrode holder; connect wires of the piezo-actuator to the output sockets of a constant-voltage stimulus isolator. Insert the solution application pipette into the pipette holder; connect loading microfilaments of the two θ-glass barrels to the corresponding groups of containers using thin tubing (Fig. 2c, see Note 3). Insert a microfilter connector between the container and the loading capillary to buffer excessive pressure jumps in the system and provide additional solution filtering (Fig. 1d, see Note 4).

74

Nathanael O’Neill and Sergiy Sylantyev

3.3 System Calibration

1. To test solution exchange times, fill containers of the solution supply system with artificial cerebrospinal fluid (ACSF) and with distilled water using syringe inlets (Fig. 2a, b) such that each θ-glass pipette barrel can be supplied with the alternating liquids. Apply pressure to the system to fully fill the supply microfilaments and the pipette dead volume. Because of the different optical characteristics of distilled water and ACSF, the corresponding ejection streams show a visible interface and the outer boundaries; use this indicator to adjust the recording pipette position (Fig. 6c). Alternatively, use different watersoluble inert inks or dyes added to each solution. 2. Submerge the solution application pipette into the recording chamber, and focus the microscope on its tip (see Note 5). 3. Place the patch-clamp recording electrode filled with a standard intracellular solution under the microscope close to the θ-glass pipette tip; open the solution supply lines so that different liquids come from the barrels. Trigger the stimulus isolator to rapidly shift the streams near the tip of the open patch pipette (see Note 6). 4. Adjust the settings of the stimulus isolator (applied voltage, voltage step length), the XYZ patch pipette position, and ejection pressure at both θ-glass channels in order to maximize the solution switch speed reflected in the recorded current pulse (Fig. 6a, b; see Note 7). Locate the optimum XYZ recording pipette tip position (using edge and/or sept of solution application pipette as a reference) by placing it near the visible stream interface (Fig. 6c; see Note 8). 5. Test the system for pulse lengths and solution replacement periods inside θ-glass barrels in accordance with the required protocol. If the ejection stream fails because of the unlikely event of one or more clogged channels, use other channels or replace the application pipette.

3.4 Protocol Implementation in Experiment

1. In experiment, θ-glass pipette should be placed in recording chamber first: at a point, ensuring it does not move tissue slice or cell culture substrate when piezo switch is applied. Then recording pipette tip position should be found referencing to θ-glass pipette sept/edge (see Subheading 3.3, step 4). Cell patching and pulling of outside-out or nucleated patches should be performed according to well-established electrophysiological procedures [14, 15].

3.5 System Maintenance

1. After each experiment, thoroughly rinse the whole setup (pressurized solution supply system, θ-glass pipette) with ~20% (vol/vol) ethanol and distilled water, and then blow it dry with an inert gas at high pressure (see Note 9).

Fig. 6 Calibration and adjustment of rapid solution application. (a) Red and blue traces illustrate responses to a rapid switch between the distilled water and the ACSF streams in θ-glass pipette channels, before and after full solution replacement (swap over) between the two channels. The panel shows a characteristic response with suboptimal settings (pipette positioning, ejector pressure, piezoactuator switch amplitude), leading to characteristic concomitant relaxation waves in the system. For presentation purposes, the currents represented by the red and blue traces were zeroed individually (applies to a and b). (b) Rapid response under optimized (correct) settings, ensuring a sub-millisecond application time constant (normally 150–200 μs). With the correct positioning of the recording pipette, a time lag between the piezo-switch artifact (arrow) and the response onset is clearly seen. (c) Recording pipette positioning. Left: water test, distilled water ejected from top channel of θ-glass pipette forms a stream with clearly visible deflection border. Recording pipette tip placed close to the border. Right: nucleated patch recorded under RASE experimental protocol. No deflection border is seen between solution streams since both barrels are filled with ACSF. Adapted by permission from Macmillan Publishers Ltd: Nature Protocols. Sylantyev, S. and D. A. Rusakov. “Sub-millisecond ligand probing of cell receptors with multiple solution exchange.” Nat Protoc 8(7): 1299–1306, © 2013

76

4

Nathanael O’Neill and Sergiy Sylantyev

Notes 1. The most crucial step in this preparation is the filling of the θ-glass pipette with glue: capillary forces tend to draw the glue into the loading microfilaments and also toward θ-glass pipette tip. To overcome this, a drop of water can be placed at the sharp end of the pipette: on being pulled into the pipette tip by capillary forces, it forms a meniscus. Because the cyanoacrylate glue hardens rapidly upon contact with water, the water meniscus stops its progress toward the pipette end. If the minimal dead volume is not a crucial requirement (e.g., for experiments in which tested patches survive for longer than several minutes or reduced RASE protocol without rapid solution exchange required), filling with the glue can be stopped near the blunt end of the θ-glass pipette. 2. Silicon tubing—especially when long—dumps pressure generated by drug ejector due to elasticity of walls; this dumping is a particularly serious obstacle when short time intervals between pressure steps are required. Thus to avoid dump effect, nonelastic Teflon tubing might be in use with silicon tubing pieces inserted only at connection points. 3. When the θ-glass pipette septum is not parallel to the piezobending actuator and/or the piezo-bending actuator is not perpendicular to the line of sight, it makes the system calibration much harder. Under these conditions, it becomes difficult to optimize stimulus isolator settings and find the recording pipette position that provides a rapid enough switch between solutions. Therefore, one should pay attention when inserting the solution application pipette into the pipette holder and clamping the actuator rod in the micromanipulator holder. 4. Switching between solution supply channels under continuous pressure produces pulses of excessive pressure which can easily damage patch at recording pipette. Microfilter (2 μM pore size in our preparation) inserted between solution container and loading capillary is vital for buffering of such pulses. 5. Maintaining the correct temperature of the applied solution is important for reliable characterization of tested receptors. Therefore, in order to ensure temperature equilibration, the tip of the θ-glass pipette should be submerged into perfusion solution for at least 4–5 mm. Temperature control in the application streams can be performed by placing a microthermocouple (straight-shaft microprobe, tip diameter ~100 μm, precision  1  C) at the position of the patch recording electrode tip. 6. If series of applications with short time intervals are required (as in experiment shown in Fig. 5), constant-voltage stimulus

Sub-millisecond Switch of Multiple Activating Solutions

77

isolator should be triggered via the TTL (transistor-transistor logic) input channel by digital output from digitizer with the corresponding triggering algorithm set in the data acquisition software. 7. The speed at which the application solution switches depends on the voltage amplitude applied at piezo-actuator and the length of the application θ-glass pipette attached to it. This speed could be increased, at least theoretically, either by strengthening the electrical pulse applied to the piezo-actuator or by using longer pipettes (and thus speeding up their tip movement). In practice, however, increasing the pipette length leads to concomitant problems such as increased tip vibration and increased viscous resistance to solution flow due to longer loading capillaries required. 8. Correct recording pipette positioning is the main means to avoid concomitant solution waves due to θ-glass pipette vibration. Another option to reduce vibration is placing of small piece of elastic material (e.g., silicone) under tightening screw in pipette holder (Fig. 1d). 9. The θ-glass pipette tip and, especially, loading microfilaments are highly sensitive to blocking with dust and precipitated crystals. Therefore, it is necessary to treat all used solutions with 4–10 μm filters and to keep them in vessels thereafter. Extensive rinsing after each experiment is also required. However, small amounts of solutions tend to accumulate toward the pipette tip because of the capillary force. It is sometimes virtually impossible to dry this remaining liquid even with gas blowing, and precipitated crystals can sometimes block the pipette tip, thus rendering the pipette unusable. To avoid this, after rinsing the pipette, dry it by touching its tip with a soft (lens) paper tissue, which absorbs the remaining solution. References 1. Clements JD, Lester RA, Tong G, Jahr CE, Westbrook GL (1992) The time course of glutamate in the synaptic cleft. Science 258:1498–1501 2. Lisman JE, Raghavachari S, Tsien RW (2007) The sequence of events that underlie quantal transmission at central glutamatergic synapses. Nat Rev Neurosci 8:597–609 3. Colquhoun D, Jonas P, Sakmann B (1992) Action of brief pulses of glutamate on AMPA/kainate receptors in patches from different neurones of rat hippocampal slices. J Physiol 458:261–287 4. Lester RA, Jahr CE (1992) NMDA channel behavior depends on agonist affinity. J Neurosci 12:635–643

5. Tong G, Jahr CE (1994) Multivesicular release from excitatory synapses of cultured hippocampal neurons. Neuron 12:51–59 6. Raman IM, Zhang S, Trussell LO (1994) Pathway-specific variants of AMPA receptors and their contribution to neuronal signaling. J Neurosci 14:4998–5010 7. Sachs F (1999) Practical limits on the maximal speed of solution exchange for patch clamp experiments. Biophys J 77:682–690 8. Sylantyev S, Savtchenko LP, Ermolyuk Y, Michaluk P, Rusakov DA (2013) Spike-driven glutamate electrodiffusion triggers synaptic potentiation via a homer-dependent mGluRNMDAR link. Neuron 77:528–541

78

Nathanael O’Neill and Sergiy Sylantyev

9. Sylantyev S et al (2008) Electric fields due to synaptic currents sharpen excitatory transmission. Science 319:1845–1849 10. Savtchenko LP, Sylantyev S, Rusakov DA (2013) Central synapses release a resourceefficient amount of glutamate. Nat Neurosci 16:10–12 11. Wlodarczyk AI et al (2013) GABAindependent GABAA receptor openings maintain tonic currents. J Neurosci 33:3905–3914 12. Papouin T et al (2012) Synaptic and extrasynaptic NMDA receptors are gated by different endogenous coagonists. Cell 150:633–646

13. Hardingham GE, Fukunaga Y, Bading H (2002) Extrasynaptic NMDARs oppose synaptic NMDARs by triggering CREB shut-off and cell death pathways. Nat Neurosci 5:405–414 14. Hamill OP, Marty A, Neher E, Sakmann B, Sigworth F (1981) Improved patch-clamp techniques for high-resolution current recording from cells and cell-free membrane patches. Pflu¨gers Arch 391:85–100 15. Molleman A (2003) Patch clamping: an introductory guide to patch clamp electrophysiology. John Wiley & Sons, Hoboken, New Jersey, USA

Chapter 7 Electrophysiological Investigation of Metabotropic Glutamate Receptor-Dependent Metaplasticity in the Hippocampus Regina U. Hegemann and Wickliffe C. Abraham Abstract Metabotropic glutamate receptors (mGluRs) are one of the major types of glutamatergic receptors contributing to synaptic plasticity mechanisms such as long-term potentiation (LTP) and long-term depression. Interestingly, activation of mGluRs alone can engage metaplastic mechanisms that create a new neuronal state, facilitating the induction and maintenance of future LTP. Here we describe typical methods used to investigate mGluR-induced metaplasticity in acute hippocampal slices. While this chapter focuses on in vitro field electrophysiological investigations, many of the principles can be applied to single-cell recordings as well as in vivo electrophysiology and indeed many types of metaplasticity phenomena. Key words Metaplasticity, Synaptic plasticity, Metabotropic glutamate receptors, Long-term potentiation, Long-term depression, Hippocampus

1

Introduction The mechanisms of synaptic plasticity have long been a focus of neuroscientific research, and it is now well established that repeated electrical stimulation can induce long-lasting changes in synaptic strength, such as long-term potentiation (LTP) and long-term depression (LTD). Beyond the classical forms of synaptic plasticity, neural activity at one point in time (often referred to as “priming” activity) can modify the state of synapses or neurons such that there is an altered ability to express synaptic plasticity later on. This phenomenon is known as metaplasticity [1]. Metaplastic regulation of synaptic plasticity comes in multiple forms: for example, it can be either specific to a set of recently activated synapses (homosynaptic) or extend to other synapses on the same postsynaptic neuron or group of neurons (heterosynaptic). Moreover, LTP or LTD can be either upregulated or downregulated, depending on the protocols used for priming and then inducing plasticity. The key distinction

Corinna Burger and Margaret Jo Velardo (eds.), Glutamate Receptors: Methods and Protocols, Methods in Molecular Biology, vol. 1941, https://doi.org/10.1007/978-1-4939-9077-1_7, © Springer Science+Business Media, LLC, part of Springer Nature 2019

79

80

Regina U. Hegemann and Wickliffe C. Abraham

between metaplasticity and other kinds of plasticity modulation is that the priming activity occurs prior to the plasticity-inducing event rather than simultaneously [2]. There is no single criterion for what is considered metaplasticity, but previous research has shown that its effects can last from minutes to hours after the priming activity in vitro and for days in vivo. Thus, the timing between priming and plasticity induction needs to be carefully considered when investigating metaplastic mechanisms. While many different types of metaplasticity have been described, the focus here will be on the regulation of future synaptic plasticity by the activation of metabotropic glutamate receptors (mGluRs). These receptors are G-protein-coupled receptors that initiate a range of signalling cascades depending on the receptor subtype and have been implicated in the induction of LTP, LTD, and learning [3–5]. Activation of group I mGluRs, located on the postsynaptic density and which traditionally couple to Gq signalling cascades, lowers the threshold for subsequent LTP induction in part through a downregulation of the slow after-hyperpolarization (sAHP) resulting in a long-lasting increase of excitability [6–8]. This decrease in the sAHP involves a nonclassical signalling cascade regulating the degree of tyrosine phosphorylation of a yet unknown target protein for which a low phosphorylation level permits the sAHP suppression [8]. Additionally, group I and group II mGluR activation can promote subsequent LTP induction by increasing α-amino-3-hydroxy-5-methyl-4-isoxazolepropionic acid receptor (AMPAR) and N-methyl-D-aspartate receptor (NMDAR) phosphorylation, respectively [9, 10]. Pharmacological or electrical activation of mGluRs has also been shown to facilitate the persistence of LTP by driving transcription-independent protein synthesis [11]. The signalling cascade driving this effect involves phospholipase C and calcium released from internal stores through ryanodine receptors. Calcium influx through storeoperated Ca2+ channels also supplements the Ca2+ response associated with this effect [12, 13]. Similarly, mGluR5 activation mobilizes a “molecular switch” that enables the conversion of short-term potentiation to LTP via CAMKIIα and protein kinase C (PKC) [14, 15]. One important consideration for conducting metaplasticity experiments is the region of interest. Due to its cytoarchitecture and role in learning and memory, much metaplasticity research has focused on the hippocampus. However, even within the hippocampal formation, different regions can respond diversely to priming stimulation. For example, while group I mGluR activation facilitates subsequent LTP and group II mGluR activation inhibits LTD in area CA1, in the dentate gyrus, priming stimulation of group I and II mGluRs inhibits subsequent LTP as well as LTD in a PKC-dependent manner [16–18]. Furthermore, the ventral CA1 appears to be more responsive to plasticity-inducing stimuli than dorsal CA1, making this region a potentially preferred target

Electrophysiological Investigation of Metaplasticity

81

for investigating metaplasticity [19, 20]. Metaplasticity experiments are frequently conducted using extracellular field recordings in acute hippocampal slices. Isolated brain slices have structural integrity while providing relatively easy access to the region of interest and allow precise electrode placement and thus have important advantages over both cell cultures and in vivo electrophysiology. Additionally, it allows for easy application of drugs of a known concentration, while many drugs that are applied during the priming stimulus can be readily washed out prior to the plasticity induction stimulation. This is an important consideration as metaplasticity mechanisms often overlap with those mediating synaptic plasticity, and thus it is important to distinguish whether plasticity or metaplasticity is targeted by any drug treatment. This consideration makes pharmacological in vivo studies harder to interpret, as it can be unclear whether a bolus injection of a drug, given to influence metaplasticity induction, has washed out by the time of the plasticity induction. Similarly, intracellular or patch pipettes filled with a blocking drug will likely be affecting both metaplasticity and plasticity inductions unless a very difficult patch perfusion system is used. Whole-cell recordings are particularly problematic for undertaking metaplasticity experiments, given that the ability to generate LTP washes out after 10–15 min of cell dialysis [21]. LTD metaplasticity experiments should not be similarly affected as LTD survives cell dialysis by a patch pipette for at least 40 min [22]. For these reasons, if single-cell recordings are desired for LTP metaplasticity experiments, these can be undertaken using sharp electrode methodologies which do not suffer the LTP washout effect. While this technique is generally more challenging than extracellular recording, it can be useful to understand how priming affects individual cells and enables tracking of cellular properties such as intrinsic excitability, which might change following priming. The intracellular recording technique using sharp electrodes has been reviewed in detail elsewhere [23]. Alternatively, whole-cell recordings with a patch pipette can be done after priming but before plasticity induction. While the focus here is on methods used to study mGluR-mediated metaplasticity in acute hippocampal slices, many of the principles and considerations can be applied to experiments investigating other types of metaplasticity and in other brain regions.

2

Materials 1. Sucrose cutting solution: 210 mM sucrose, 26 mM NaHCO3, 2.5 mM KCl, 1.25 mM NaH2PO4, 0.5 mM CaCl2, 3 mM MgCl2, 20 mM D-glucose, oxygenated with carbogen (95% O2–5% CO2). Store for no more than 1 month at
20 ∘C.

82

Regina U. Hegemann and Wickliffe C. Abraham

2. Artificial cerebrospinal fluid (aCSF): 124 mM NaCl, 3.2 mM KCl, 1.25 mM NaH2PO4, 26 mM NaHCO3, 2.5 mM CaCl2, 1.3 mM MgCl2, 10 mM D-glucose, oxygenated with carbogen (95% O2–5% CO2). Store at 4 ∘C for no more than 2 days (see Note 1). 3. Recording chamber with either a flow-through perfusion system or a continuous perfusion system recirculating the aCSF (the latter can be very useful when using expensive drugs). 4. Temperature control unit maintaining the aCSF in the recording chamber at the desired temperature. Note that LTP, LTD, and associated metabolic pathways are temperature-dependent, and so temperatures near body temperature are usually preferred (e.g., 32–35 ∘C). 5. Programmable stimulators, preferably constant current with control of pulse amplitude and duration (e.g., diphasic pulse half-wave duration 0.1 ms). 6. AC amplifiers for field potential recordings, 0.1–3 kHz halfamplitude cutoffs typically, and DC amplifiers for intracellular recordings. 7. Software for acquisition and analysis of data. 8. Stimulation electrodes (using fine wires, e.g., Teflon-insulated tungsten or large bore glass electrodes). 9. Glass recording electrode(s) prepared from glass capillary micropipettes with filament (e.g., OD 1 mm, ID ¼ 0.58 mm) using a micropipette puller, filled with aCSF or 2 M NaCl for a final resistance of 2–3 MΩ.

3

Methods 1. Cut 400 μm transverse isolated rat hippocampal slices or coronal whole brain mouse slices of the same thickness (see Note 2). High-quality slices are critical for successful electrophysiological experiments, and slices need to be prepared in a consistent manner (see Note 3). 2. Let slices recover at interface at the appropriate temperature at least 2–3 h before transferring them to the recording chamber (see Note 4). 3. Transfer slices to the recording chamber where they can be kept either at interface or submerged (see Note 5). 4. Place electrodes appropriately to record from the pathway of interest. Figure 1 shows a typical hippocampal slice with electrode placements for field electrophysiological recordings from CA1 (see Note 6).

Electrophysiological Investigation of Metaplasticity

83

Fig. 1 A superfused acute rat hippocampal slice with CA3 removed. A 50 μm Teflon-insulated tungsten monopolar stimulating electrode (S) and a glass pipette recording (R) electrode were placed in stratum radiatum of CA1 midway between stratum lacunosum-moleculare and stratum pyramidale to stimulate Schaffer collateral axons running in CA1 and to record evoked field potentials from their synaptic termination zones

5. Check to ensure that recordings are free of line frequency (50–60 Hz) noise (see Note 7). 6. Stimulate slices at a constant stimulation intensity and frequency (e.g., one pulse every 30 s) for at least 20 min to collect baseline recordings. Choose the baseline stimulus intensity so that the response is ~50% of the maximum slope (see Note 8). 7. Deliver the desired priming protocol which can be done either pharmacologically or via electrical stimulation of synapses or neurons: Pharmacological priming: A variety of mGluR agonists such as 1-amino-cyclopentane-1S,3R-dicarboxylic acid (ACPD), (R, S)-3,5-dihydroxyphenylglycine (DHPG), or LY379268 can be used as a priming stimulus to activate selective mGluRs (see Note 9). Electrical priming: Priming is conducted via electrical stimulation of afferents, either homosynaptically or heterosynaptically to the test pathway (see Note 10). The appropriate priming protocol should be carefully considered (see Note 11). 8. Continue to stimulate slices at baseline intensity for the desired period of time before LTP induction (see Note 12). Any drugs applied during the priming stimulation, for example, for mechanistic studies, need to be washed out completely before the plasticity-inducing event. Include appropriate control groups (see Note 13). 9. Induce LTP, using the desired LTP induction paradigm (see Note 14).

84

Regina U. Hegemann and Wickliffe C. Abraham

10. Analyze LTP expression using the appropriate measure of interest at least 30 min after LTP induction (see Note 15), compared to the average baseline response (see Note 8).

4

Notes 1. To save time and effort, stock aCSF can be prepared in advance and stored for up to 1 month at 4  C. Stock aCSF should contain all salts except NaHCO3 and Glucose at 20 the final concentration. Add 50 mL/L of stock aCSF to make up the final solution. 2. When preparing whole brain slices, slices can be prepared from different directions, but note that slice orientation can affect synaptic plasticity [24, 25]. Because activation of mGluRs induces rhythmic oscillatory activity in CA3 neurons [26] and a slow-onset LTP at CA3-CA1 synapses when CA3 is intact [27], CA3 should be removed when investigating mGluRmediated plasticity mechanisms. 3. If the evoked responses are hard to obtain, small or unstable, or show multiple population spikes, the slice health may not be adequate. Unless there is a clear sign of injury, such as a rip, it can be hard to tell whether the tissue is healthy, but a spread body layer and “stringy” appearing tissue may be an indication of poor slice health. There are many reasons why slice health may have been compromised. For example, small changes in temperature, pH, and oxygenation can also make a significant difference to the electrophysiological properties of a slice, so it is important to optimize slice preparation procedures before conducting electrophysiological experiments. Handle the tissue with upmost care, and be sure not to stretch the tissue during the dissection, and keep all external variables constant. Conditions might differ between laboratories, and slice preparation procedures have been reviewed elsewhere [28–30]. If poor slice health persists, ensure the cutting solution and aCSF have been prepared correctly and the equipment has not been contaminated. Glassware should be cleaned thoroughly with boiling distilled water every day and with ~2% HCL once a month. 4. In our laboratory, slices are left to recover at 32 ∘C for 30 min and then for 2 h at room temperature and are then maintained at 32.5 ∘C during the course of the experiment. Even small differences in temperature can have a major impact on cellular physiology and changes the amount of oxygen the aCSF can hold. It is advisable to routinely check the bath temperature with a separate thermometer as thermistor probes can be sensitive and break easily.

Electrophysiological Investigation of Metaplasticity

85

5. For mGluR-dependent metaplasticity experiments, it is advisable to keep the slices at interface in a humidified atmosphere and perfused with pre-warmed, oxygenated aCSF at a rate of 2 mL/min. Slices can also be kept submerged, but submerged slices have a lower pH and reduced expression of mGluRdependent LTP, which can complicate such experiments [27]. Superfusion does, however, have the significant advantage of faster washin and washout of drugs [11, 12]. It is important to keep the perfusion rates and the fluid level in the recording chamber constant as variation in fluid level can introduce instability in the recordings. 6. For recording Schaffer collateral-CA1-evoked responses, place the stimulating and recording electrodes ~500 μm apart in stratum radiatum, about midway between the stratum lacunosum-moleculare and the stratum pyramidale. Keeping the distance between stimulating and recording electrode as constant as possible is crucial as variations in electrode placements can affect the electrophysiological profile of the recorded response and change the amount of LTP obtainable [31]. 7. Electrical noise (50–60 Hz) can be a major source of problems when conducting electrophysiological recordings, and it can reflect a range of issues with the recording setup and equipment. To limit noise on your rig, work inside a faraday cage and ensure that all equipment (even fluid lines) is sufficiently grounded, fluid lines are away from electrical cables, and all connections in the electrical circuit are intact. Be sure to check points that are prone to break and keep all connection points rust-free. Blocked electrodes or air bubbles in the recording electrode also increase noise. 8. A stable baseline is essential when conducting plasticity experiments, and there should be no obvious drift with variation >10%. It is impossible to make a meaningful conclusion about synaptic plasticity effects if responses were drifting before the induction of LTP as you cannot be sure if the change is due to plasticity mechanisms or a continuously increasing/decreasing baseline. Online tracking of the relevant measurements is a helpful way to determine baseline stability. For some slices, experiments might have to be restarted multiple times. A drifting baseline might indicate instability in the system (e.g., moving electrode or an unstable fluid level) or declining slice health. When conducting two-pathway experiments or recording from more than one slice at a time, the relative changes in the responses might provide some information of the source of the problem. Additionally, it can be difficult to measure a change of synaptic plasticity if the priming stimulation alters baseline responses. Baseline measurements should always be taken directly before the plasticity-inducing event, but any

86

Regina U. Hegemann and Wickliffe C. Abraham

change in baseline responses after priming needs to be taken into consideration. A control group only receiving priming stimulation should be included in the analysis to assess whether responses will return to baseline over time or remain stable after priming. 9. It is important to understand the drugs used, including their non-specific effects, washout periods, and concentrationdependent changes of their effect (desired and non-desired). Consult the literature for details, and conduct control experiments to understand their effects on plasticity. We have successfully used 20 μM ACPD or DHPG in aCSF for 10 min starting 30 min before high-frequency stimulation in order to facilitate the induction and persistence of LTP [8, 12]. However, the choice of agonist depends on the aim of the experiment. ACPD is a broad-spectrum mGluR agonist, DHPG is a selective agonist for group I mGluRs, and LY37268 is a selective group II mGluR agonist. 10. Priming with electrical stimulation is an important part of the metaplasticity toolkit as it has more physiological credence than pharmacological approaches. Priming stimulation can be delivered homosynaptically to the test pathway in which the plasticity of interest is to be induced or heterosynaptically to an independent pathway. For heterosynaptic experiments involving two input pathways to the same dendritic arbor, for example, place one stimulating electrode on each side of the recording electrode. Ensure that the pathways are in fact independent of each other, for example, by using a crossed pairedpulse facilitation paradigm where a conditioning pulse to one pathway is shown not to facilitate the response to a test pulse in the other pathway (50 ms interpulse interval). 11. Priming stimulation can be chosen for either theoretical or empirical reasons. We have used two trains of theta-burst stimulation (TBS) to activate group I mGluRs and thereby facilitate subsequent LTP (Fig. 2) [11]. This protocol was chosen because we knew that it was capable by itself of causing a protein synthesis-dependent LTP and we were searching for a protein synthesis-dependent metaplasticity effect. In our terminology, each train of TBS is made up of ten bursts of four or five pulses at 100 Hz, with 200 ms between bursts (start-tostart) and 30 s between trains. Each train of HFS consists of 100 pulses, delivered at 100 Hz (Fig. 3). When priming cells electrically, it is important to consider any other signalling cascades that might be activated by the priming stimulus and could interfere with the process of interest. For example, to isolate the effect of priming on mGluRs, an NMDAR inhibitor such as D-AP5 (50 μM) should be added to the solution to

Electrophysiological Investigation of Metaplasticity

87

Fig. 2 Homosynaptically delivered priming stimulation (two trains of TBS), in the presence of the NMDA receptor antagonist AP5, facilitated LTP induced 20 min later with one train of TBS. Copyright © 2000 Society for Neuroscience. Adapted from ref. 11 permission from the Society for Neuroscience

Fig. 3 Schematic representation of the pulse sequences delivered during one train of TBS or HFS to induce LTP. One train of TBS consists of ten bursts delivered at 5 Hz (200 ms interburst interval), with each burst made up of four to five pulses, delivered at 100 Hz (10 ms interpulse interval). For simplicity, the illustration shows only two bursts of pulses. One train of HFS is typically composed of 100 pulses, delivered at 100 Hz

prevent LTP induction and NMDAR-mediated mechanisms in the primed pathway. 12. The timing of the priming activity and plasticity-inducing event needs to be carefully considered. Precise timing might depend on the signalling cascades involved. While most studies investigating mGluR-mediated metaplasticity have studied delays of 20–60 min, time delays of 5–20 min between priming and plasticity induction might be appropriate, depending on the type of priming used. If a priming effect is observed with a particular protocol, understanding the time course of the effect (onset and duration) may provide clues to the mechanisms

88

Regina U. Hegemann and Wickliffe C. Abraham

involved. Drug washout periods need to be taken into consideration when priming is conducted pharmacologically. 13. When investigating specific signalling cascades involved in mGluR-dependent metaplasticity, additional pharmacological manipulations such as antagonists of a molecule might be required. It is important to carefully consider the concentration and timing of drug application. For example, effectiveness of a protein synthesis inhibitor at different time points can aid the understanding of the signalling cascades (priming translation vs de novo protein synthesis) [11]. It is also crucial to isolate the effect to metaplastic mechanisms rather than plasticity itself by including sufficient control groups. Control groups are essential for being able to make sound conclusions about an observed effect. When describing a metaplasticity effect, primed slices are generally compared to unprimed control slices. The treatment of control and primed slices should be as similar as possible, and priming stimulation periods should be substituted with the equivalent time of baseline stimulation to avoid confounding effects of stimulation periods. When using pharmacological priming, control slices should be perfused with the vehicle used to deliver the drug of choice. Additionally, it can be useful to include a control group receiving priming treatment without delivering the plasticity-inducing stimulation to assess any potential longterm effects on baseline responses which might mask effects on LTP. If other pharmacological treatments (e.g., antagonist for molecule of interest) are used in conjunction with priming, two control groups should be included: one receiving the drug without priming and one receiving the priming without the presence of the drug. The timing of drug application is also crucial when investigating mechanisms. It may be of interest, for example, to include drug treatments during the time between priming and plasticity, or even just during plasticity induction (i.e., shortly before and after). These control groups give extra information as to exactly when the metaplasticity mechanism becomes manifest. 14. It is important to consider which paradigm should be used as different means of LTP induction can engage distinct intracellular signalling cascades [32]. However, both HFS and TBS LTP induction protocols can be facilitated by mGluR activation (Fig. 4). Furthermore, it is useful to choose an induction protocol that allows observation of inhibition as well as facilitation of LTP to avoid metaplastic effects being masked by ceiling or floor effects. This is particularly important if it is uncertain which direction of effect will be manifested by priming. While one train of TBS or two trains of HFS, at

Electrophysiological Investigation of Metaplasticity

89

Fig. 4 Example of metabotropic glutamate receptor-dependent metaplasticity. Activation of group 1 mGluRs with DHPG facilitated LTP induced by 0.5 TBS (i.e., 5 bursts, a) or 0.5 HFS (i.e., 100 Hz for 500 ms, b). Copyright © 1998 by John Wiley Sons, Inc. Reproduced from ref. 12 with permission from Wiley & Sons Inc

baseline intensity, are generally sufficient to induce a stable LTP which can be up- and downregulated by previous neuronal activity, the amount of LTP obtained with a given paradigm might differ between laboratories. 15. The standard measure of LTP is the change in the field excitatory postsynaptic potential (fEPSP) which is most accurately reflected by the slope of a given potential. Additionally, the population spike can provide information about a change in cell firing. However, this measure is more useful to determine changes in the EPSP-spike relationship rather than LTP per se. LTP should be assessed at least 30 min post-LTP induction to rule out short-tetanic potentiation but can be up to 3 h or even longer post-induction, depending on the type of LTP of interest [33].

90

Regina U. Hegemann and Wickliffe C. Abraham

Acknowledgment We wish to thank Dr. Bruce Mockett for his contribution to Fig. 1 and the University of Otago for financial support of R.H. Hegemann. References 1. Abraham WC, Bear MF (1996) Metaplasticity: the plasticity of synaptic plasticity. Trends Neurosci 19:126–130 2. Abraham WC (2008) Metaplasticity: tuning synapses and networks for plasticity. Nat Rev Neurosci 9:387. https://doi.org/10.1038/ nrn2356 3. Manahan-vaughan D, Braunewell K (2005) The metabotropic glutamate receptor, mGluR5, is a key determinant of good and bad spatial learning performance and hippocampal synaptic plasticity. Cereb Cortex 15 (11):1703–1713. https://doi.org/10.1093/ cercor/bhi047 4. Bashir ZI, Jane DE, Sunter DC, Watkins JC, Collingridge GL (1993) Metabotropic glutamate receptors contribute to the induction of long-term depression in the CA1 region of the hippocampus. Eur J Pharmacol 239:265–266. https://doi.org/10.1016/0014-2999(93) 91009-C 5. Fitzjohn SM, Palmer MJ, May JER, Neeson A, Morris SAC, Collingridge GL (2001) A characterisation of long-term depression induced by metabotropic glutamate receptor activation in the rat hippocampus in vitro. J Physiol 537:421–430. https://doi.org/10.1111/j. 1469-7793.2001.00421.x 6. Cohen AS, Abraham WC (1996) Facilitation of long-term potentiation by prior activation of metabotropic glutamate receptors. J Neurophysiol 76:953–962 7. Ireland DR, Abraham WC (2002) Group I mGluRs increase excitability of hippocampal CA1 pyramidal neurons by a PLC-independent mechanism. J Neurophysiol 88:107–116 8. Ireland DR, Guevremont D, Williams JM, Abraham WC (2004) Metabotropic glutamate receptor-mediated depression of the slow after hyperpolarization is gated by tyrosine phosphatases in hippocampal CA1 pyramidal neurons. J Neurophysiol 92:2811–2819. https://doi. org/10.1152/jn.01236.2003 9. Oh MC, Derkach VA, Guire ES, Soderling TR (2006) Extrasynaptic membrane trafficking regulated by GluR1 serine 845 phosphorylation primes AMPA receptors for long-term

potentiation. J Biol Chem 281:752–758. https://doi.org/10.1074/jbc.M509677200 10. Rosenberg N, Gerber U, Ster J (2016) Activation of group II metabotropic glutamate receptors promotes LTP induction at schaffer collateral-CA1 pyramidal cell synapses by priming NMDA receptors. J Neurosci 36 (45):11521–11531 11. Raymond CR, Thompson VL, Tate WP, Abraham WC (2000) Metabotropic glutamate receptors trigger homosynaptic protein synthesis to prolong long-term potentiation. J Neurosci 20:969–976 12. Cohen AS, Raymond CR, Abraham WC (1998) Priming of long-term potentiation induced by activation of metabotropic glutamate receptors coupled to phospholipase C. Hippocampus 8:160–170. https://doi. org/10.1002/(SICI)1098-1063(1998) 8:23.0.CO;2-P 13. Mellentin C, Jahnsen H, Abraham WC (2007) Priming of long-term potentiation mediated by ryanodine receptor activation in rat hippocampal slices. Neuropharmacology 52:118–125. https://doi.org/10.1016/j.neuropharm. 2006.07.009 14. Bortolotto ZA, Collingridge GL (1998) Involvement of calcium/calmodulindependent protein kinases in the setting of a molecular switch involved in hippocampal LTP. Neuropharmacology 37:535–544 15. Bortolotto ZA, Bashir ZI, Davies CH, Collingridge GL (1994) A molecular switch activated by metabotropic glutamate receptors regulates induction of long-term potentiation. Nature 368:740–743. https://doi.org/10.1038/ 368740a0 16. Rush AM, Wu J, Rowan MJ, Anwyl R (2002) Group I metabotropic glutamate receptor (mGluR)-dependent long-term depression mediated via p38 mitogen-activated protein kinase is inhibited by previous high-frequency stimulation and activation of mGluRs and protein kinase C in the rat dentate gyrus in vitro. J Neurosci 22(14):6121–6128 17. Wu J, Rowan M, Anwyl R (2004) Synaptically stimulated induction of group i metabotropic glutamate receptor-dependent long-term

Electrophysiological Investigation of Metaplasticity depression and depotentiation is inhibited by prior activation of metabotropic glutamate receptors and PROTEIN KINASE C. Neuroscience 123:507–514. https://doi. org/10.1016/j.neuroscience.2003.09.013 18. Gisabella B, Rowan MJ, Anwyl R (2003) Mechanisms underlying the inhibition of long-term potentiation by preconditioning stimulation in the hippocampus in vitro. Neuroscience 121:297–305 19. Maggio N, Segal M (2006) Unique regulation of long term potentiation in the rat ventral hippocampus. Hippocampus 17(1):10–25 20. Tidball P, Burn HV, Teh KL, Volianskis A, Collingridge GL, Fitzjohn SM (2017) Differential ability of the dorsal and ventral rat hippocampus to exhibit group I metabotropic glutamate receptor–dependent synaptic and intrinsic plasticity. Brain Neurosci Adv 1:1–13. https://doi. org/10.1177/2398212816689792 21. Kato K, Clifford DB, Zorumski CF (1993) Long-term potentiation during whole-cell recording in rat hippocampal slices. Neuroscience 53:39–47. https://doi.org/10.1016/ 0306-4522(93)90282-K 22. Jo J, Heon S, Kim MJ, Son GH, Park Y, Henley JM, Weiss JL, Sheng M, Collingridge GL, Cho K (2008) Metabotropic glutamate receptormediated LTD involves two interacting Ca2+ sensors, NCS-1 and PICK1. Neuron 60:1095–1111. https://doi.org/10.1016/j. neuron.2008.10.050 23. Parkington HC, Coleman HA (2012) Intracellular “sharp” microelectrode recording. In: Essential guide to reading biomedical papers. John Wiley & Sons, Ltd, Chichester, UK, pp 77–84 24. Bartlett TE, Lu J, Wang YT (2011) Slice orientation and muscarinic acetylcholine receptor activation determine the involvement of N-methyl D-aspartate receptor subunit GluN2B in hippocampal area CA1 long-term depression. Mol Brain 4:41. https://doi.org/ 10.1186/1756-6606-4-41

91

25. Sherwood JL, Amici M, Dargan SL, Culley GR, Fitzjohn SM, Jane DE, Collingridge GL, Lodge D, Bortolotto ZA (2012) Differences in kainate receptor involvement in hippocampal mossy fibre long-term potentiation depending on slice orientation. Neurochem Int 61:482–489. https://doi.org/10.1016/j. neuint.2012.04.021 26. Taylor G, Merlin L, Wong R (1995) Synchronized oscillations in hippocampal CA3 neurons induced by metabotropic glutamate receptor activation. J Neurosci 15 27. Bortolotto ZA, Collingridge GL (1995) On the mechanism of long-term potentiation induced by (1S,3R)-1-aminocyclopentane1,3-dicarboxylic acid (ACPD) in rat hippocampal slices. Neuropharmacology 34:1003–1014 28. Suter KJ, Smith BN, Dudek FE (1999) Electrophysiological recording from brain slices. Methods 18:86–90. https://doi.org/ 10.1006/meth.1999.0761 29. Villers A, Ris L (2013) Improved preparation and preservation of hippocampal mouse slices for a very stable and reproducible recording of long-term potentiation. J Vis Exp. https://doi. org/10.3791/50483 30. Lein PJ, Barnhart CD, Pessah IN (2011) Acute hippocampal slice preparation and hippocampal slice cultures. Humana Press, Totowa, NJ, pp 115–134 31. Kauer JA (1999) Blockade of hippocampal long-term potentiation by sustained tetanic stimulation near the recording site. J Neurophysiol 81(2):940–944 32. Zhu G, Liu Y, Wang Y, Bi X, Baudry M (2015) Different patterns of electrical activity lead to long-term potentiation by activating different intracellular pathways. J Neurosci 35:621–633. https://doi.org/10.1523/JNEUROSCI. 2193-14.2015 33. Raymond CR (2007) LTP forms 1, 2 and 3: different mechanisms for the “long” in longterm potentiation. Trends Neurosci 30:167–175. https://doi.org/10.1016/j.tins. 2007.01.007

Chapter 8 Induction of Metabotropic Glutamate Receptor-Mediated Long-Term Depression in the Hippocampal Schaffer Collateral Pathway of Aging Rats Kirstan E. Gimse, Kenneth O’Riordan, and Corinna Burger Abstract Scientific progress in the understanding of the molecular mechanisms of aging in the brain is essential for the identification of novel targets for the treatment and prevention of age-associated cognitive disorders. Electrophysiological analysis of synaptic plasticity using extracellular field recordings in rodent hippocampal slices is a well-established method for investigating molecular mechanisms of learning and memory. These methods can be applied to the study of aging in the brain by utilizing hippocampal slices from aged animals. However, this application can be challenging as it is difficult to ensure and maintain the health of slices originating from aged animals. The technique described in this chapter outlines the procedure for measuring metabotropic glutamate receptor-mediated long-term depression in hippocampal slices using extracellular field recordings and includes specific details for the application of this technique in the study of neuronal aging. Key words Metabotropic glutamate receptors, Long-term depression, Aging, Hippocampus, Synaptic plasticity, Field recordings

1

Introduction As the world population ages, age-associated neurological disorders are rapidly increasing in incidence [1]. Due to the prevalence and severity of these disorders, scientific progress into understanding the mechanisms underlying cognitive changes in the brain with aging is essential for identifying novel targets for their treatment and prevention. In addition to understanding how signaling mechanisms in the brain are altered with aging, identifying the molecular mechanisms facilitating successful cognitive aging may offer insight to further treatment and prevention strategies. Aged rats exhibit similar variability in cognitive ability with aging and are often used as a model organism for learning and memory studies [2–6].

Corinna Burger and Margaret Jo Velardo (eds.), Glutamate Receptors: Methods and Protocols, Methods in Molecular Biology, vol. 1941, https://doi.org/10.1007/978-1-4939-9077-1_8, © Springer Science+Business Media, LLC, part of Springer Nature 2019

93

94

Kirstan E. Gimse et al.

Spatial learning and memory deficits are common symptoms of age-associated cognitive decline observed both in aged humans and aged rats [2, 7–11]. These processes are largely hippocampal dependent and are mediated by hippocampal synaptic plasticity. Along the Schaffer collateral pathway, both long-lasting increases (long-term potentiation, LTP) and decreases (long-term depression, LTD) in the strength of signaling between synapses have been shown to be important for spatial learning [12–21]. Hippocampal LTD and LTP can be analyzed in vitro utilizing extracellular field recordings in acute hippocampal slices. However, these recordings can be more difficult in hippocampal slices from aged animals, and inadequate preservation of slice health during the dissection and preparation process can confound experimental findings. The technique outlined in this protocol focuses on the induction of a particular type of LTD, group I metabotropic glutamate receptordependent LTD (mGluR-LTD); however, the strategies described for the maintenance of slice health can be applied to other types of synaptic plasticity studies in hippocampal slices from aged rats or mice.

2

Materials 1. Ultrapure water [dH2O (18.2 MΩ·cm at 25
C)]. 2. Sucrose cutting solution (CS): 212 mM sucrose, 2.6 mM KCl, 1.25 mM NaH2PO3, 26 mM NaHCO3, 0.5 mM CaCl2, 5 mM MgCl2, 10 mM glucose. 3. Artificial cerebral spinal fluid (ACSF): 124 mM NaCl, 5 mM KCl, 1.25 mM NaH2PO3, 26 mM NaHCO3, 2 mM CaCl2, 1 mM MgCl2, 10 mM glucose. 4. Carboxygen tanks (95%O2/5%CO2) with regulators. 5. Tabletop research anesthesia apparatus with O2 flush. 6. Isoflurane (99%). 7. O2 tank with regulator. 8. Large sharpened guillotine (see Note 1). 9. Tissue chopper equipped with a Vernier micrometer and razor blades. 10. Dissection tools: scalpel and blades (we use #20 or # 21blade, #4 handle), mini Blumenthal rongeurs, large scissors, small scissors, and two double-ended stainless steel spatulas (rounded on one end and tapered on the other), paintbrush, small plastic spoon to move brains between containers (see Note 2). 11. Glassware: 10 cm petri dish, four 250 mL beakers (see Note 2).

mGluR-LTD Induction in the Hippocampus of Aged Rats

95

12. Plastic transfer pipettes: one 8 mL pipette and one 3 mL narrow stem pipette (cut tip of stem at ~45
angle to facilitate delicate transfer of slices). 13. (S)-3,5-Dihydroxyphenylglycine (DHPG) (see Note 3). 14. Recording chamber and perfusion system. 15. Temperature control unit. 16. Stimulus isolator. 17. AC amplifier and recording headstages. 18. Digitizer. 19. Software for acquisition and analysis of data. 20. Stimulation electrodes made with 0.0279 mm platinum/ tungsten wire. 21. Glass recording electrodes prepared from glass capillary micropipettes with filament (1.5 mm outer diameter, 0.86 mm inner diameter, 10.16 cm length) using a micropipette puller filled with fresh carboxygenated ACSF. 22. Filling needle.

3

Methods

3.1 Preparation of Solutions and Dissection Area

1. Retrieve animal and let acclimate to the room where the decapitation will be performed for at least 30 min, avoiding startling noises and movements. 2. Prepare CS and ACSF: For both solutions, dissolve all reagents except MgCl2 and CaCl2 and dilute to ~90% final volume with dH2O. Then add MgCl2 and CaCl2 and dilute to final volume. These solutions should be carboxygenated (bubbled with 95% O2 and 5% CO2) while being prepared or immediately after preparation. CS can be stored at 4
C for 1 week; ACSF can be stored at 4
C for not longer than 24 h. 3. Fill two 250 mL beakers with ~150 mL of CS, and place at 80
C for approximately one-half hour until the CS is partially frozen. 4. While CS is cooling, prepare the dissection area (Fig. 1): Gather ice bucket and dissection tools, and transfer pipettes, petri dish, and paintbrush. Next to the ice bucket, place a large disposable surface protecting mat—this surface will be used during brain extraction (Fig. 1a) (see Note 4). 5. Retrieve beakers of partially frozen CS, and break up ice with a metal spatula to create an icy slurry. 6. Scoop out an area of ice for each beaker; place the beakers in the ice such that the fluid/slurry level in the beaker is roughly equal

96

Kirstan E. Gimse et al.

Fig. 1 Surgical setup for hippocampal dissection. (a) Necessary dissection tools. (b) Ice bucket with chilled cutting solution, dissection surface, and beaker for collecting slices

to the level of ice in the bucket. Pack the ice around the beakers and begin bubbling one or both beakers with carboxygen (Fig. 1b). 7. Fill a third beaker with ~150 mL of cutting solution, and position in the ice bucket in the same manner as described above. This beaker will store the slices as they are being cut (Fig. 1b). 8. Place a 10 cm glass petri dish upside down in the ice bucket, sinking it so that the ice is touching the inside surface of the dish. This will be the foundation upon which the dissection is performed (Fig. 1b). 9. Place the scissors and the rongeurs on the disposable mat next to the ice bucket. Place all remaining dissection tools and transfer pipettes into the bubbling beaker of ACSF. 3.2 Preparation of Hippocampal Slices

1. Using isoflurane, induce a surgical plane of anesthesia in the rat: place the animal in anesthesia box and turn the oxygen flow to 2 mL/min with 4% isoflurane.

mGluR-LTD Induction in the Hippocampus of Aged Rats

97

Fig. 2 Brain extraction steps. (a) (Subheading 3.2, step 4) Removal of the occipital and interparietal plates to expose the cerebellum and pulling temporal plates away from brain. (b) (Subheading 3.2, step 6) Cutting through the midline suture of the parietal plates. (c) (Subheading 3.2, step 7) Pulling of the parietal plates away from the brain. (d) (Subheading 3.2, step 9) Scooping brain into chilled CS

2. As soon at the animal reaches a surgical plane (indicated by an inability to self-right and a lack of response to foot pinch), decapitate the animal immediately rostral to the first cervical vertebra using a small animal guillotine (see Note 5). 3. Using the scalpel, with firm pressure, make an incision in the middle of the skull starting near the caudal portion of the nasal bone cutting all the way back to the occipital bone. 4. Using the rongeurs, quickly remove the occipital and interparietal bones exposing the cerebellum (Fig. 2a). 5. Next, using the rongeurs, grip the left temporal plate below the cerebellum. Twist the rongeurs to pull the temporal plate away from the brain. Repeat on the left side. 6. Next, using a very sharp pair of small scissors (we use straighttipped, non-serrated Wagner scissors with a 30 mm cutting edge), cut through the midline suture of the parietal plates continuing through the frontal skull plates. Make sure to keep the cutting edge of the lower sheer pressed firmly against the skull’s inner surface and away from the brain (Fig. 2b). 7. Grip the right parietal plate with the rongeurs. With an upward and outward motion, pull the plate away from the brain. It is important to pull up and out rather than to just rotate the rongeurs upward, and this can cause the outer edge of the plate to twist inward and damage the brain. Repeat with the left side (Fig. 2c). 8. Use the rongeurs to remove the frontal plates. 9. Position the rounded end of a spatula just rostral to the frontal lobe. Use a scooping motion to slide the spatula under the brain, slide the spatula from left to right to release the brain, and scoop directly into an ice-cold slurry of carboxygenated CS

98

Kirstan E. Gimse et al.

(Fig. 2d). Make sure the brain is submerged in the cutting solution; allow the brain to cool in the solution for 1–2 min (see Notes 6 and 7). 10. Place a piece of #1 Whatman filter paper on the flat surface of the large petri dish; moisten with ice-cold CS; this is your dissection surface. 11. Remove the brain from its beaker and place on the dissection surface. Using the scalpel, remove the cerebellum (Fig. 3a, left panel) and cut along the longitudinal fissure (Fig. 3a, middle panel). Separate the hemispheres using the rounded end of a spatula. 12. Place both hemispheres in the clean beaker of partially frozen carboxygenated CS slurry (Fig. 3a, right panel). Remove and replace the filter paper. Leaving one hippocampus submerged in the partially frozen CS, remove the other and place on the dissection surface. Begin bubbling the third beaker containing only liquid CS with carboxygen. Make sure that the carboxygen bubbling is very gentle to prevent the bouncing of slices as they are added to the beaker. 13. Using the scalpel, remove the foremost region of the frontal lobe, creating a flat surface on which the hemisphere can stand upright (Fig. 3b, left panel). 14. Slide the rounded end of one spatula between the brainstem and the cortex. Use this spatula to protect the hippocampus while using the angled spatula to pull the brainstem and midbrain away from the cortex. Using the angled spatula, pull the brainstem forward while pushing back gently on the cortex with the rounded spatula, working to remove any connective tissue or vasculature with the brain stem (Fig. 3b, middle panel). 15. Once the white fimbria which forms a hyperbola at the bottom of the hippocampus is visible, firmly press downward with the rounded spatula, and use the angled spatula to completely remove the brainstem, midbrain, and thalamus from the cortex (Fig. 3b, right panel). 16. Using a transfer pipette, gently squirt ice-cold CS into the gap underneath the hippocampus to help separate it from the cortex. Then, keeping the angled spatula parallel to the hippocampus, slide this spatula between the hippocampus and cortex. Gently roll the hippocampus up and away from the cortex (Fig. 3c, left and middle panels). 17. Carefully trim away the cortex and any vasculature from the hippocampus (Fig. 3c, right panel). 18. Using a tissue chopper equipped with a Vernier micrometer, slice the hippocampus into 400 μm transverse slices. Use a soft

mGluR-LTD Induction in the Hippocampus of Aged Rats

99

Fig. 3 Hippocampal dissection. (a) (Subheading 3.2, steps 11 and 12) Removal of the cerebellum, separation of hemispheres, and submersion in CS. (b) (Subheading 3.2, steps 13 and 14) Removal of the frontal lobe (left panel), brainstem (middle panel), and midbrain away from cortex. (c) (Subheading 3.2, steps 15–17) Separation of the hippocampus from the cortex

paintbrush to carefully transfer each slice from the blade to the beaker containing liquid ice-cold CS as they are cut (Fig. 4). 19. Repeat the hemisphere.

procedure

with

the

second

hippocampal

20. Remove the beaker containing slices in ice-cold CS from the ice bucket and place at room temperature, continuing to bubble gently with carboxygen. Very slowly add room temperature ACSF to the beaker, once again ensuring that the slices remain

100

Kirstan E. Gimse et al.

Fig. 4 Hippocampal slice preparation. (a) Isolated hippocampus. (b) Hippocampus shown on tissue chopper. (c) Transfer of slices to ice-cold CS using paintbrush. (d) Slices in beaker gently bubbled with carboxygen

undisturbed, until the beaker is filled with a roughly 50:50 mixture of ACSF and CS. 21. Let slices incubate at room temperature for 45 min, and then carefully transfer to gently carboxygenated pure ACSF at room temperature, and incubate for an additional 45 min. 22. Carefully transfer slices to the recording chamber perfused with carboxygenated ACSF; allow slices to equilibrate in the recording chamber for 2 h. Slices should be incubated at 30
C during this period and retained at this temperature throughout the duration of the experiment (see Note 8). 3.3 Establishment of a Steady Baseline Response

1. Prepare recording carboxygenated ACSF.

electrodes

filled

with

fresh

2. Place stimulating electrode in the Schaffer collaterals of the CA3 region and the recording electrode in the stratum radiatum of the CA1 region. It is important to place electrodes gently, so that they are just touching the surface of the hippocampus. You should not be able to see a hole or depression in the tissue where your electrode was positioned when it is removed. 3. To ensure slice health, deliver a single-pulse stimulus of 15 V and record the slope of the response. Repeat slice test with the same input stimulus three times, waiting at least 3 min between stimuli. A healthy slice will not exhibit more than a 10% fluctuation between response slopes. 4. Determine the stimulus intensity which evokes half that of the maximum response: Using the data acquisition software, deliver single-pulse stimuli ranging from 5 to 25 V (if using mouse hippocampal slices, this input/output relationship can be evaluated with stimuli ranging from 5 to 15 V) with 1-min intervals between pulses, and record the slope of the change in field potential in response to each applied stimulus. Calculate the stimulus which evokes a response equal to half that of the maximum response.

mGluR-LTD Induction in the Hippocampus of Aged Rats

101

5. Program the software to deliver a single-pulse stimulus with this intensity every 40 s. In our lab, responses are collected and averaged in sets of three, producing one averaged response every 2 min. Record the baseline for at least 30 min, ensuring that responses do not fluctuate more than 10% from the average baseline response (see Notes 9 and 10). 3.4 Induction of mGluR-LTD

1. Prepare 100 μM DHPG: Prepare 10 mM aliquots of DHPG in dH2O. These can be stored at 20
C for no more than 10 days. Remove 10 mM DHPG from freezer and allow to come to room temperature while protected from light. Dilute to 100 μM with ACSF (see Note 11). 2. Once a steady baseline has been recorded (typically 30–45 min), replace the ACSF being perfused into the recording well with 100 μM DHPG in ACSF. Continue bath application of 100 μM DHPG for 10 min, and then replace and continue perfusion with ACSF (see Note 12). 3. Continue recording for at least 60–90 min post-LTD induction (see Note 13). 4. When recording is finished, ensure maintenance of slice health by again assessing basal responses to single-pulse stimuli ranging from 5 to 25 mA. 5. Use the data acquisition and analysis software to analyze the data by measuring the slope of the change in field potential in response to the applied stimulus. The slope of each response is calculated as a percentage of the average response during baseline and plotted again the time of acquisition. Typically, timepoint “0” is designated at the begining of DHPG incubation (Fig. 5).

4

Notes 1. To facilitate rapid and humane decapitation, it is important to keep the small animal guillotine clean, sharp, and free of rust. To this end the guillotine should be thoroughly cleaned with deionized water and dried after each use. The sharpness of the blades should be evaluated monthly, and blades should be sharpened at least once yearly depending on frequency of use. 2. To ensure good recordings and healthy slices, it is important to keep glassware, perfusion and carboxygen tubing, recording electrodes, as well as chamber wells, slice supports, and well covers clean and free from salt buildup, bacteria, and detergents.

102

Kirstan E. Gimse et al.

Fig. 5 LTD induction in young and aged rats. mGluR-LTD was induced with bath application of DHPG (100 μM) for 10 min. Aged and young rats display similar LTD function. LTD is blocked in young rats treated with a small hairpin RNA targeting Homer1b/c [shH1c; adapted from: [25]. Inset: Field potentials (average of six waveforms) from a representative experiment taken at the times indicated by the numbers on the graph from aged slices. fEPSP ¼ field excitatory postsynaptic potential. n ¼ 7(4), i.e., 7 slices from 4 animals

Therefore, all glassware used for electrophysiology experiments should be kept separate from other lab glassware. This glassware should be thoroughly rinsed with pressurized deionized (DI) water, followed by a dH2O rinse, and autoclaved prior to the first usage. After each use glassware should be thoroughly rinsed with pressurized DI water and placed upside down to air-dry. When setting up an experiment, glassware should again be rinsed with DI and dH2O prior to usage. Rinsing followed by autoclaving should be repeated monthly. Dissection tools should be cleaned and autoclaved in the same fashion as glassware. Well covers and slice supports should be rinsed with pressurized DI water and dH2O before and after each experiment. All tubing should be replaced roughly every 2 months depending on the degree of usage. Perfusion tubing should be flushed with dH2O for at least 20 min at the end of every day. Carboxygen tubing which has been submerged in ACSF should also be rinsed at the end of each day. Wells in the recording chamber should also be rinsed and flushed with dH2O prior to experimental setup. If a water bath is used for temperature regulation, it should be filled with dH2O only and

mGluR-LTD Induction in the Hippocampus of Aged Rats

103

emptied at the end of each experiment. The bath should be thoroughly cleaned quarterly. 3. For our experiments we use pure samples of the active (S)DHPG enantiomer; however, the racemic mixture (RS)DHPG is also available. Aliquots of 10 mM DHPG in dH2O should be prepared and stored at 20
C for no more than 10 days. Both lyophilized and aliquoted DHPG are light sensitive. Lyophilized DHPG should be brought to room temperature before dissolution in dH2O. 4. We find dissection the easiest when using a large low-walled rectangular ice bucket, but this may vary on the preferences of the experimenter. 5. Sometimes the decapitation does not occur immediately rostral to the first vertebra and excess tissue, and the spinal cord may block access to the occipital bone. In this event use the large scissors to sever this tissue as close as possible to the occipital bone. 6. Minimizing time of brain extraction is essential for hippocampal slice health, especially in aged animals. Steps 4–10 should be completed as quickly as possible. The time from decapitation to brain submersion in CS slush should not exceed one and a half minutes. 7. We find it is of vital importance to dissect brains in CS instead of ACSF. It significantly improves the health of the aged rat slices. 8. It is essential to keep the temperature regulation finely tuned. Temperature should not fluctuate more than 1
C during incubation or recording. Temperature fluctuations during incubation can have a detrimental impact on slice health, and even small fluctuations during recording can impact response intensity. 9. During the running of your experiment, keep watch on the carboxygenation levels: too high and it will disturb the stimulating wire in the slice and mess up the run; too low and the slices will dry out and die off. 10. Keep an eye out for big condensation water droplets growing around electrodes and mop up as necessary touching gently with a Q-tip (usually not a problem on runs lasting less than 2 h). Alternatively, you can coat the bottom of the glass electrode with petroleum jelly to prevent water drops from falling into the recording chamber. 11. It is not necessary to keep diluted DHPG covered during bath application; however, if any remaining DHPG in ACSF will be used later that day, it should be stored on ice away from light. Do not store DHPG in ACSF overnight.

104

Kirstan E. Gimse et al.

12. A stable baseline not fluctuating more than 10% of the average should be observed for at least 30 min to ensure slice health. It may take up to an hour for the baseline to stabilize. If the baseline does not stabilize within an hour, it is likely that the slice is not healthy. 13. mGluR-LTD consists of two-phases: an induction phase and a protein synthesis-dependent late phase. Recordings less than 60 min may fail to show differences between groups in the expression of late-phase LTD [22–24].

Acknowledgments This work was supported by NIH R01AG048172-03 and NIH T32 GM081061 to Kirstan Gimse. References 1. Aarli JA, Dua T, Janca A, Muscetta A (2006) Neurological disorders: public health challenges. World Health Organization, Geneva, pp 177–182 2. Gage FH, Dunnett SB, Bjorklund A (1984) Spatial learning and motor deficits in aged rats. Neurobiol Aging 5:43–48 3. Gage FH, Bjorklund A (1986) Cholinergic septal grafts into the hippocampal formation improve spatial learning and memory in aged rats by an atropine-sensitive mechanism. J Neurosci 6:2837–2847 4. Markowska AL et al (1989) Individual differences in aging: behavioral and neurobiological correlates. Neurobiol Aging 10:31–43 5. Gallagher M, Burwell R, Burchinal M (1993) Severity of spatial learning impairment in aging: development of a learning index for performance in the Morris water maze. Behav Neurosci 107:618–626 6. Gerstein H, Hullinger R, Lindstrom MJ, Burger C (2013) A behavioral paradigm to evaluate hippocampal performance in aged rodents for pharmacological and genetic target validation. PLoS One 8:e62360. https://doi.org/ 10.1371/journal.pone.0062360 7. Flicker C, Bartus RT, Crook TH, Ferris SH (1984) Effects of aging and dementia upon recent visuospatial memory. Neurobiol Aging 5:275–283 8. Perlmutter M, Metzger R, Nezworski T, Miller K (1981) Spatial and temporal memory in 20 to 60 year olds. J Gerontol 36:59–65

9. Sharps MJ, Gollin ES (1987) Memory for object locations in young and elderly adults. J Gerontol 42:336–341 10. Barnes CA, Nadel L, Honig WK (1980) Spatial memory deficit in senescent rats. Can J Psychol 34:29–39 11. Wallace JE, Krauter EE, Campbell BA (1980) Animal models of declining memory in the aged: short-term and spatial memory in the aged rat. J Gerontol 35:355–363 12. Rapp PR, Rosenberg RA, Gallagher M (1987) An evaluation of spatial information processing in aged rats. Behav Neurosci 101:3–12 13. Morris RG, Garrud P, Rawlins JN, O’Keefe J (1982) Place navigation impaired in rats with hippocampal lesions. Nature 297:681–683 14. Hullinger R, O’Riordan K, Burger C (2015) Environmental enrichment improves learning and memory and long-term potentiation in young adult rats through a mechanism requiring mGluR5 signaling and sustained activation of p70s6k. Neurobiol Learn Mem 125:126–134. https://doi.org/10.1016/j. nlm.2015.08.006 15. Cortese GP, Olin A, O’Riordan K, Hullinger R, Burger C (2018) Environmental enrichment improves hippocampal function in aged rats by enhancing learning and memory, LTP, and mGluR5-Homer1c activity. Neurobiol Aging 63:1–11. https://doi.org/10. 1016/j.neurobiolaging.2017.11.004 16. Ayala JE et al (2009) mGluR5 positive allosteric modulators facilitate both hippocampal LTP and LTD and enhance spatial learning.

mGluR-LTD Induction in the Hippocampus of Aged Rats Neuropsychopharmacology 34:2057–2071. https://doi.org/10.1038/npp.2009.30 17. Balschun D et al (1999) A specific role for group I mGluRs in hippocampal LTP and hippocampus-dependent spatial learning. Learn Mem 6:138–152 18. Goh JJ, Manahan-Vaughan D (2013) Spatial object recognition enables endogenous LTD that curtails LTP in the mouse hippocampus. Cereb Cortex 23:1118–1125. https://doi. org/10.1093/cercor/bhs089 19. Yang S et al (2013) Integrity of mGluR-LTD in the associative/commissural inputs to CA3 correlates with successful aging in rats. J Neurosci 33:12670–12678. https://doi.org/10. 1523/JNEUROSCI.1086-13.2013 20. Kumar A (2011) Long-term potentiation at CA3-CA1 hippocampal synapses with special emphasis on aging, disease, and stress. Front Aging Neurosci 3:7. https://doi.org/10. 3389/fnagi.2011.00007 21. Bachevalier J, Nemanic S (2008) Memory for spatial location and object-place associations are differently processed by the hippocampal formation, parahippocampal areas TH/TF

105

and perirhinal cortex. Hippocampus 18:64–80. https://doi.org/10.1002/hipo. 20369 22. Neyman S, Manahan-Vaughan D (2008) Metabotropic glutamate receptor 1 (mGluR1) and 5 (mGluR5) regulate late phases of LTP and LTD in the hippocampal CA1 region in vitro. Eur J Neurosci 27:1345–1352. https://doi.org/10.1111/j.1460-9568.2008. 06109.x 23. Huber KM, Kayser MS, Bear MF (2000) Role for rapid dendritic protein synthesis in hippocampal mGluR-dependent long-term depression. Science 288:1254–1257 24. Luscher C, Huber KM (2010) Group 1 mGluR-dependent synaptic long-term depression: mechanisms and implications for circuitry and disease. Neuron 65:445–459. https://doi. org/10.1016/j.neuron.2010.01.016 25. Gimse K et al (2018) Hippocampal Homer1b/ c is necessary for contextual fear conditioning and group I metabotropic glutamate receptor mediated long-term depression. Neurobiol Learn Mem 156:17–23.

Chapter 9 Whole-Cell Patch-Clamp Electrophysiology to Study Ionotropic Glutamatergic Receptors and Their Roles in Addiction Jonna M. Leyrer-Jackson, M. Foster Olive, and Cassandra D. Gipson Abstract Development of the whole-cell patch-clamp electrophysiology technique has allowed for enhanced visualization and experimentation of ionic currents in neurons of mammalian tissue with high spatial and temporal resolution. Electrophysiology has become an exceptional tool for identifying single cellular mechanisms underlying behavior. Specifically, the role of glutamatergic signaling through α-amino-3-hydroxy-5methyl-4-isoxazolepropionic acid (AMPA) and N-methyl-D-aspartate (NMDA) receptors underlying behavior has been extensively studied. Here we will discuss commonly used protocols and techniques for performing whole-cell patch-clamp recordings and exploring AMPA and NMDA receptor-mediated glutamatergic responses and alterations in the context of substance abuse. Key words Electrophysiology, Addiction, Patch clamp, AMPA, NMDA, Substance abuse, Glutamatergic receptors

1

Introduction

1.1 Overview of Patch-Clamp Electrophysiology

The use of patch-clamp electrophysiology began in the 1970s, where Neher and Sakmann developed a technique that allowed for the visualization of the opening and closing of single ion channels within the plasma membrane of muscle cells, termed cellattached electrophysiology (Fig. 1a). Further development of this technique resulted in whole-cell patch-clamp electrophysiology, which reduced the influence of background noise throughout recordings, and allowed for enhanced visualization and further experimentation of ionic currents with higher spatial and temporal resolution. A whole-cell patch is obtained by rupturing the membrane patch without breaking the micropipette-obtained seal, allowing for electrical access to the entire cell, not just ion channels on the cell surface as developed previously (Fig. 1b). These techniques were the first to identify whole-cell characteristics including cell firing and all ionic conductance across the plasma membrane,

Corinna Burger and Margaret Jo Velardo (eds.), Glutamate Receptors: Methods and Protocols, Methods in Molecular Biology, vol. 1941, https://doi.org/10.1007/978-1-4939-9077-1_9, © Springer Science+Business Media, LLC, part of Springer Nature 2019

107

108

Jonna M. Leyrer-Jackson et al.

A.

B.

Recording Pipette

Recording Pipette

Plasma Membrane

Plasma Membrane

Ion Channel

Cell Body

Cell Body

Fig. 1 (a) Cell-attached electrophysiology is performed by placing a recording pipette against the plasma membrane, without rupturing through the membrane. This technique allows for the recording of single ion channel conductance. (b) Whole-cell electrophysiology is conducted when a seal is made between the recording pipette and plasma membrane. Following seal formation, negative pressure is applied to the recording pipette rupturing the plasma membrane, allowing the recording pipette electrical access to the entire cell

thus revolutionizing the field of neurophysiology. Although still a widely used technique to observe physiological changes in a single cell, over the last few decades, single-cell patch-clamp recordings have been further combined with multielectrode paired recordings, optogenetic approaches, and various other techniques, each allowing for further isolation and study of individual neurons and their behavior in neural circuits. Here we will discuss commonly used protocols and techniques for performing whole-cell patch-clamp recordings and exploring α-amino-3-hydroxy-5-methyl-4-isoxazolepropionic acid (AMPA) and N-methyl-D-aspartate (NMDA) receptor-mediated glutamatergic responses and alterations in the context of substance abuse. 1.2 Overview of Glutamatergic Receptors 1.2.1 AMPA Receptor Physiology and Function

One category of ionotropic glutamatergic receptors, primarily responsible for fast synaptic transmission, are AMPA (α-amino-3hydroxy-5-methyl-4-isoxazolepropionic acid) receptors and are known to mediate a vast majority of communication between neurons within the central nervous system. AMPA receptors (AMPARs) are a hetero-oligomeric protein complexes consisting of four subunits (GluA1-GluA4, also known as GluRA-GluRD) which form a tetrameric channel most commonly composed of GluA2-GluA3 and GluA2-GluA1 subunit combinations [1, 2],

Patch-Clamp Isolation of Glutamatergic Receptors in Addiction

109

although there is evidence of other subunit compositions across various brain regions [3, 4]. The subunit composition of each receptor determines its conduction kinetics, ionic permeability, and subcellular trafficking properties. Functional studies have shown that variations in the GluA2 subunit (modifications of the Q/R site in pre-mRNA) lead to alterations in calcium permeability, where GluA2 subunits lacking or containing the Q/R site show high and low calcium permeability, respectively [5, 6]. Additionally, variations in receptor subunit compositions have been observed across brain regions and give rise to differences in AMPAR-mediated currents within different cell types [7]. In combination, AMPAR permeability and kinetics can be altered based on subunit composition. It is interesting to note that AMPA receptors have unique properties that allow them to change functionally in an activitydependent manner. For example, AMPA receptors play an important role in long-term potentiation (LTP), which is defined as use-dependent strengthening of an individual synapse and is believed to underlie learning and memory. LTP can be induced in slices with brief trains of high-frequency stimulation of presynaptic fibers [8, 9]. Following this stimulus, the strength of these synaptic connections (i.e., postsynaptic responses to the same stimulus intensity) can be increased as much as fivefold [10]. It is well established that LTP induction protocols increase the surface expression of AMPA receptors through exocytosis. However, it has recently been discovered that AMPA receptors containing GluA1 subunits are exocytosed more readily than AMPA receptors consisting of GluA2, GluA3, and GluA4 subunit compositions [11]. Due to their kinetics, it is thought that during LTP induction protocols, GluA1 subunit-containing receptors may promote/ facilitate AMPA responses by increasing their expression within the plasma membrane. However, with long-term depression (LTD) protocols, GluA2 containing subunits become more prominent within the plasma membrane and may play a role in dampening AMPA receptor-mediated responses [12]. Although not yet completely understood, it is clear that trafficking of AMPA receptors can be subunit specific and may play a prominent role in shaping neuronal communication, especially under the influence of addictive substances [13]. Although LTP and LTD induction protocols can induce exocytosis and endocytosis of AMPA receptors [14–16], the activation of these receptors is reliant on presynaptic release of glutamate. Upon presynaptic release, glutamate can then bind to the AMPAR causing it to undergo a conformational change resulting in channel opening. Upon opening, sodium can flow through the channel and into the postsynaptic cell, causing postsynaptic membrane depolarization [17, 18] (Fig. 2). Postsynaptic depolarization not only aides in fast synaptic transmission (heavily mediated by

110

Jonna M. Leyrer-Jackson et al.

Action Potential

1. AMPA Receptor

Ca2+

2.

NMDA Receptor Pre-Synaptic Vesicle

3. Glutamate Voltage Gated Calcium Channel

4.

Na+

6.

Na+

8.

Magnesium Ion

Ca2+

7.

5. 6.

+ Post-Synaptic

+

9.

Fig. 2 Overview of AMPA and NMDA receptor activation. (1) An action potential enters the synaptic bouton of the presynaptic cell, causing voltage-gated calcium channels to open (2). A rise in cytosolic calcium causes glutamate-filled vesicles to fuse with the plasma membrane (3) and release glutamate into the synaptic cleft. Glutamate within the cleft binds to both AMPA and NMDA receptors (4) causing AMPA channels to undergo a conformational change and open, allowing sodium to flow into the postsynaptic cell (5). The flow of sodium causes a postsynaptic depolarization, alleviating the magnesium block of NMDA channels (6). In combination, the alleviation of magnesium and binding of glutamate to the NMDA receptor allow them to open (7) and both sodium and calcium to enter the postsynaptic cell (8) causing further depolarization of the postsynaptic cell (9)

AMPA receptors) but also facilitates activation of NMDA receptors (discussed below). AMPA receptor physiology can be studied using single-cell electrophysiology. Specifically, experiments can be conducted using various AMPAR agonists and NMDAR antagonists, alterations in membrane potentials, and/or changes in solution composition. However, regardless of isolation protocol (specific details and options discussed below), AMPAR currents/potentials are characterized by a short rise and decay times, nearing 1 millisecond and less than 50 milliseconds, respectively. 1.2.2 NMDA Receptor Physiology and Function

Another type of ionotropic glutamatergic receptor primarily responsible for synaptic plasticity are NMDA (N-methyl-D-aspartate) receptors (NMDARs). These receptors consist of various

Patch-Clamp Isolation of Glutamatergic Receptors in Addiction

111

subunit combinations, including NR1, NR2, and NR3, and each subtype has many isoforms (i.e., GluN2A, GluN2B, etc.). Different subunit combinations determine the activation and conduction time of the receptor. It has been well established that NMDARs are hetero-oligomeric and composed of multiple NR1 subunits in combination with at least one GluN2 subunit. Interestingly, NR3 subunits must co-assemble with NR1/NR2 complexes in order to be functional; alone, these subunits cannot form functional receptors [19]. The activation and inactivation times of NMDARs are much slower than AMPAR, yet these receptors also facilitate changes in synaptic plasticity. Like AMPARs, the subunit composition of these receptors determines their activation and deactivation times. For example, GluN1/GluN2A subunit assemblies have a deactivation constant ranging in tens of milliseconds when activated by glutamate, whereas NR1/NR2D combinations shift the deactivation time to several seconds [19]. Interestingly, the composition of NMDAR is known to change developmentally to promote and modify neuronal connectivity during maturation of the nervous system. During early development, it is essential for NMDAR currents to be prolonged in order to promote learning and memory and strengthen the formation of synapses. However, after the initial period of peak plasticity modification, NMDAR currents decrease in duration (i.e., making neurons less plastic). It has been reported that changes in NMDAR containing different subunits throughout development lead to these physiological changes [20, 21], where GluN2B and GluN2D expression is elevated in young animals and supplemented with or replaced by GluN2A subunits over the course of development [19]. Interestingly, alterations in NMDAR subunit composition have been reported with chronic use of drugs of abuse, which will be discussed in more detail below. Interestingly, NMDA receptors are expressed both synaptically and extrasynaptically. While synaptic NMDA receptors play a role in synaptic transmission and plasticity, extrasynaptic NMDA receptors are thought to play a role in promoting cell death [22] and neuronal synchronization [23, 24]. Due to their differences in function, it is not surprising that these receptor pools are composed of different subunit compositions, with GluN2B and GluN2A subunits showing higher expression levels extrasynaptically and synaptically, respectively [25]. Although incompletely understood, these differences in subunit compositions may promote the vastly different functionality observed between these location-dependent NMDA receptor pools. These differences in NMDA receptor locations may prove important for the study of addictive substances and their downstream signaling effects associated with synaptic plasticity and cell death.

112

Jonna M. Leyrer-Jackson et al.

Unlike AMPARs, the pore of a synaptic NMDAR is blocked by magnesium ions at hyperpolarized potentials (near resting membrane potential). For activation of these receptors to occur, two events must happen simultaneously, making this subtype of glutamatergic receptors “coincidence receptors.” Postsynaptic depolarization, due to cation influx through AMPA receptors, leads to alleviation of the voltage-dependent magnesium ion blockade. In addition to alleviation of the magnesium block, the receptor must also be externally activated by presynaptic release of glutamate. NMDA receptors have the unique activation property that requires detection and pairing of both presynaptic glutamate release and binding to the receptor and postsynaptic depolarization causing alleviation of the magnesium block (Fig. 2). The activation and opening of NMDA channels have been shown to increase postsynaptic calcium influx, which has been linked to activation of multiple downstream pathways and changes in synaptic plasticity. To characterize these responses with single-cell electrophysiology, it is extremely common to alter magnesium concentrations of extracellular solutions. Additionally, recordings are routinely conducted in the presence of NMDAR agonists and AMPAR antagonists and at various membrane potentials, dependent on solution composition. However, regardless of isolation protocol (specific details are discussed below), NMDAR currents/potentials are characterized by long rise and decay times, nearing tens of milliseconds and hundreds of milliseconds, respectively, compared to AMPA receptors. 1.3 Overview of Addiction

Addiction is defined as a progressing neurobehavioral disorder characterized by repetitive and compulsive substance seeking and use, despite its detrimental consequences, and repeated unsuccessful attempts at abstinence. Addiction is thought to be caused, in part, by long-lasting maladaptive memories related to drug experiences [26]. Abuse of tobacco, alcohol, and illicit drugs cost the United States over $740 billion annually (NIDA trends and statistics) and is characterized by repeated use of one or more substances often accompanied by physiologically harmful and detrimental effects. Individuals suffering from addiction often display symptoms including narrowing of the behavioral repertoire in favor of drug seeking and drug taking, neglected appearance, physical health issues, and difficulties performing everyday activities (i.e., difficulties at work or school). Furthermore, addiction is also characterized by difficulty quitting the substance of abuse, relapse following attempts at abstinence, and intense urges (craving) for the drug. It is unclear how use of drugs of abuse can lead to addictive traits in some individuals and not others. However, there is evidence to support that drugs of abuse can alter synaptic plasticity in

Patch-Clamp Isolation of Glutamatergic Receptors in Addiction

113

specific brain regions that may give rise to common traits observed with addiction. All commonly abused drugs of addiction are known to target the mesolimbic dopamine system, either directly or indirectly. The mesolimbic dopamine system originates from dopaminergic cell bodies in the ventral tegmental area (VTA) of the midbrain. This region sends dense dopaminergic projections to the prefrontal cortex (PFC) and nucleus accumbens (NAc). All of these areas have been heavily associated with addiction and are thought to undergo neurobiological modifications in the presence of substances of abuse [26], which some have shown to have rapid onset and can be long lasting [27]. Arguably the most well-studied cellular modifications due to addictive substances have been characterized in dopaminergic cells of the VTA and have revealed that drug-induced modifications in dopaminergic cell plasticity and firing patterns may play a substantial role in mediating behavioral changes by modifying rewardrelated neuronal circuitry. While a large body of literature focuses on drug-induced changes in mesolimbic dopaminergic neurons, an additional subset of literature supports the role of glutamate in learning and other adaptive mechanisms in animal models of drug addiction [28–30]. For this reason, the study of AMPA and NMDA receptors in addiction has become extremely important and has gained much attention for determining synaptic modifications that may underlie the disease. Thus, in order to properly study the mechanisms of addiction, it is essential to study the effects of abused substances on AMPA and NMDA receptor function in addiction-related brain regions. This chapter aims to provide a general protocol for using slice electrophysiology for studying AMPA and NMDA receptors, especially in the context of addiction research. Overall, these protocols will step through solution preparation (Subheading 3.1), slice preparation (Subheading 3.2), whole-cell recordings (Subheading 3.3), isolating glutamatergic receptor-mediated responses (Subheading 3.4), various whole-cell induction protocols (Subheading 3.5), the relevance of in vitro to in vivo studies (Subheading 3.6), and common analysis conducted with patch-clamp data (Subheading 3.7). We will also discuss approaches for finding healthy cell types, drugs used for isolating AMPA and NMDA receptormediated responses, the relationship and validation of in vitro studies in vivo, as well as new techniques used within the field of neurophysiology and patch-clamp physiology. These protocols provide insight into how slice physiology can aide in the study of neuronal circuits and cellular mechanisms related to and/or caused by addiction.

114

2

Jonna M. Leyrer-Jackson et al.

Materials Regardless of brain region of interest, slice physiology relies heavily on the health of the tissue and cell viability. Slice electrophysiology generally uses three separate solutions, each pertinent in ensuring healthy tissue. The first solution necessary is a cutting solution, a solution generally containing sucrose. The second solution necessary for slice physiology is the recording buffer, often a modification of a “Krebs” or “Ringer Solution” and referred to as artificial cerebrospinal fluid (aCSF). This solution is used throughout recording. The last solution used is the intracellular solution, which mimics neuronal intracellular fluid. 1. Cutting solution: 206 mM sucrose; 25 mM NaHCO3; 10 mM dextrose; 3.3 mM KCl; 1.23 mM NaH2PO4; 1.0 mM CaCl2; 4.0 mM MgCl2, osmolarity adjusted to 295  5 mOsm and pH adjusted to 7.40  0.03. 2. aCSF: 120 mM NaCl; 25 mM NaHCO3; 3.3 mM KCl; 1.23 mM NaH2PO4; 0.9 mM CaCl2; 2.0 mM MgCl2; 10 mM dextrose, osmolarity adjusted to 295  5 mOsm and pH adjusted to 7.40  0.03. 3. Intracellular solution: 135 mM potassium gluconate; 10 mM KCl; 1.0 mM EGTA; 10 mM HEPES; 2 mM Mg-ATP; 0.38 mM Tris-GTP, osmolarity adjusted to 285  5 mOsm and pH adjusted to 7.30  0.01. For optimal recordings, intracellular solutions should have roughly a 10 mOsm difference (with intracellular being less) from the recording solution, in order to maintain a proper osmotic gradient.

3

Methods

3.1 Solution Preparation 3.1.1 General Steps for Making Recording and Cutting Solutions

1. Fill graduated cylinder with distilled (ultrafiltered) water to approximately ¾ of the total desired volume. 2. Add all solvents (excluding MgCl2 and CaCl2 stock solutions) (Fig. 3a). 3. Saturate the solution with 95% O2/5% CO2 gas using a gas dispersion stone. Depending on volume, the time it takes for complete saturation will be different. With a 2-liter solution, saturation takes approximately 15 min (Fig. 3b). 4. Add MgCl2 and CaCl2 solutions to the saturated solution. 5. Adjust the pH of the solution to 7.40  0.03 using NaOH or HCl. Ensure that the solution is being constantly stirred to ensure an accurate pH reading (Fig. 3c). 6. Transfer the solution to a volumetric flask and bring to full volume with distilled water.

Patch-Clamp Isolation of Glutamatergic Receptors in Addiction

115

Fig. 3 General steps of solution preparation include (a) adding solvents and (b) saturating with 95% O2/5% CO2 gas followed by adjusting the pH (c) and the osmolarity (d) of the solution to proper physiological ranges

7. Obtain an osmolarity reading (using a calibrated osmometer) and adjust the solution to appropriate osmolarity. To decrease and increase osmolarity in recording solutions, add water or NaCl/sucrose, respectively (Fig. 3d). 3.1.2 General Steps for Making Intracellular Solutions

1. Start by adding ½ of the total desired volume of distilled (ultrafiltrated) water to the beaker/container. 2. Place a stir bar into the beaker and spin at low to moderate speed throughout entire solution-making process. 3. Add dry potassium gluconate and KCl and allow them to dissolve into solution. 4. Add EGTA and HEPES (these are generally stock solutions kept at 80  C). 5. Add Mg-ATP. 6. Add Tris-GTP. Because the amount of GTP is very small, it is easiest to order 10-mg pre-weighed aliquots and use the entire volume. Be sure to rinse the aliquot container with water to ensure total amount was added (see Note 1). 7. Bring the volume just short of the desired final volume using distilled water. 8. pH the solution to 7.2–7.3 using KOH or HCl. 9. Adjust the osmolarity to 285–290 mOsm by adding water or potassium gluconate. 10. Aliquot solution into 1-mL aliquots in 1.7-mL centrifuge tubes.

116

Jonna M. Leyrer-Jackson et al.

11. Place centrifuges tubes into a liquid nitrogen bath to rapidly freeze them. Once they are frozen, store at 80  C if possible. For solution alterations, see Note 2. 3.2 In Vitro Tissue Slice Preparation

1. The animal should be sacrificed through rapid decapitation and the brain rapidly removed (i.e., in less than 2 min). Although any age animal can be used, tissue from younger animals (2–6 weeks) is often more viable and easier to perform patchclamp recordings. 2. Once removed, the brain should be placed in a beaker filled with pre-chilled and well-oxygenated cutting buffer and placed left on ice for 3 min. 3. After 3 min, remove the brain from the beaker, place it on the vibratome cutting block, and mount using a thin layer of cyanoacrylate glue. 4. Using an automated tissue vibratome, make tissue slices of the desired region. Slices should be kept in the range of 200–300 μm for optimal viability and oxygenation purposes. It is common to use a blade angle of 10–20 , with a slow cutting speed and fast oscillation. We frequently use an automated Leica Microsystems vibratome (Fig. 4) which features an oscillating tissue microtome and digital slice thickness controller. 5. As slices are made, remove them from the brain block using a pair of spring-tension microscissors and transfer each slice to an incubation chamber filled with aCSF (recording buffer) using a truncated disposable pipet. The incubation chamber should be placed in a water bath incubator at 34  C, and the aCSF should be continuously oxygenated with 95% O2/5% CO2. See Fig. 5 for an example of an incubation chamber and water bath. 6. After all slices have been transferred to the incubation chamber, allow them to incubate at 34  C for 45 min. After 45 min, the incubation chamber should be removed from the water bath and left to cool to room temperature for a minimum of 10 min before transferring slices to a recording chamber.

3.3 Whole-Cell Patch-Clamp Recordings in Slices

1. After slices have been allowed to cool to room temperature for at least 10 min (last step of slice preparation), use a truncated disposable plastic pipette and transfer the one slice to the recording chamber located on the microscope (Fig. 6a). Depending on recording location, slices should be positioned properly for optimal cell approach (i.e., largest portion of the cell body positioned toward electrode approach location). 2. Once positioned, slices should be secured by a net/harp (an example harp can be seen in Fig. 6b) to avoid fluid induced slice movement. It is essential to remove most of the fluid in the

Patch-Clamp Isolation of Glutamatergic Receptors in Addiction

117

Fig. 4 A commonly used motorized, oscillating tissue microtome (vibratome) used for preparing thin (between 100 and 300 μm thick) tissue slices for electrophysiology (Leica Microsystems, Wetzlar, Germany)

chamber surrounding the slice before harp placement, as excess liquid will cause the slice to shift during placement. 3. Upon placement of the harp, start a continuous flow of aCSF (recording buffer) at a flow rate of 1–2 mL/min. The aCSF source should be continuously bubbled to ensure oxygen saturation. Some perfusion systems/options are shown in Fig. 7. 4. After ensuring proper flow of the aCSF perfusion, the tissue is ready for visualization. Slice quality and viability can be visually observed by focusing on the tissue at high power and looking for the following characteristics of healthy and unhealthy cells (e.g., see Notes 3 and 4). See Fig. 8 for a pictorial representation of healthy and unhealthy cells. 5. After determining the slice tissue is healthy, locate a single cell on high power using infrared/differential interference contrast (IR/DIC) to approach for recording. See Note 5 for additional details. 6. Before approaching the cell, the recording pipette must be filled with intracellular solution using a MicroFil needle. Once filled with intracellular solution, the pipette should be placed on the micromanipulator (see Fig. 9). Apply positive

118

Jonna M. Leyrer-Jackson et al.

Fig. 5 Instruments commonly used for slice incubation. (a) Water incubation bath (Fisher Scientific, Pittsburgh, PA) commonly used to keep slices maintained at 34  C throughout the incubation period. (b) Incubation chamber, equipped with four distinct chambers and a perfusion stone placed in a walled off chamber between the four slice chambers (Automate Scientific, Berkeley, CA). This system is engineered for optimal oxygen saturation (vigorous bubbling) with minimal slice movement, preventing tissue damage

Fig. 6 (a) A typical slice recording chamber containing both an inlet (a) and outlet (b) dam for continuous solution flow. This chamber is also equipped with an inline heater (not shown) and bath temperature sensor (c). The microelectrode ground pellet is also shown (d). (b) Slices are positioned and held in place with a metal harp containing 1.5 mm spaced harp strings

Patch-Clamp Isolation of Glutamatergic Receptors in Addiction

119

Fig. 7 Common perfusion options used in tissue slice electrophysiological studies. (a) Gravity-fed, manually operated perfusion option which is equipped with a drip chamber to prevent air bubbles within the inlet line and a roller clamp to modify flow rate. (b) A 4-solution pinch-valve automated perfusion system (Automate Scientific, Berkeley, CA) which can be started and stopped electronically (c). This system is not gravity fed; therefore, flow rate is maintained regardless of location and fluid level. This system also prevents air bubbles from entering the recording chamber inlet line. Automated pinch-valve systems also come in 8- and 16-solution options

Fig. 8 Healthy and unhealthy cells are shown for layer V cells of the medial prefrontal cortex. The solid circle surrounds a healthy cell with preferred pyramidal shape. An unhealthy cell is depicted by the arrow, where the arrowhead shows nuclear swelling encompassing nearly 2/3 of the cell body. The dashed circle depicts an extremely unhealthy cell displaying a shrunken cell body with a “wrinkled” appearance

120

Jonna M. Leyrer-Jackson et al.

Fig. 9 Recording pipette placed in a pipette holder (Molecular Devices, San Jose, CA) located on a micromanipulator (Siskiyou, Grants Pass, OR)

Fig. 10 Positive pressure through the recording pipette elicits a “firehose effect” when placed in the bathing solution

pressure to the pipette using a 10-mL syringe and place the recording pipette in the bath just above the slice. 7. Verify that the resistance of the pipette tip is between 4 and 6 mΩ (this will vary by cell). For older animals, resistance may need to be modified in order to obtain a proper seal. 8. After checking the resistance of the recording pipette, use your amplifier or amplifier software to compensate for the resistance in order to prevent an offset in recording data. 9. While maintaining positive pressure through the recording pipette (which can be observed by the “fire hose effect” protruding from the recording pipette; see Fig. 10), begin the approach toward the cell. Maintaining positive pressure serves to push any debris out of the way for easier visualization of the cell and to keep the end of the recording pipette free from debris. 10. As the cell is approached, a “dimpling” of the cell membrane will occur due to the positive pressure. See Note 6 for additional details.

Patch-Clamp Isolation of Glutamatergic Receptors in Addiction

121

11. Before removing the positive pressure, begin monitoring pipette resistance. Most recording softwares have this option built into program. If obtaining recordings in voltage clamp mode, it is also recommended to set a holding potential near cellular resting membrane potential. In the event that the cell has a spontaneous membrane rupture, this holding potential will prevent cell death. This holding potential will vary depending on resting membrane potential of the cell type. 12. Once cell dimpling is visualized and the cell does not move from its location (slipping in any direction indicates an unhealthy cell), rapidly release the positive pressure. The release of pressure will bring the cell body toward the recording pipette tip where a seal between the pipette tip and cell membrane will begin to form. This seal will be apparent by an abrupt increase in pipette resistance. For a giga-seal (a seal over 1 GΩ in resistance) to be reached, suction (either mouth or syringe induced) is often necessary. 13. Once a giga-seal is obtained, additional suction should be applied (again, either mouth or syringe induced) to rupture the plasma membrane. This is evident by a rapid reduction in pipette resistance. Membrane resistance, membrane capacitance, and resting membrane potentials can now be visualized through the recording software (see Note 7 for additional information). 14. In addition to the cellular characteristics listed above, additional measurements can be used to test for cell health and viability. Some of these measurements include spiking characteristics (inter-spike interval, spike frequency, spike amplitude, and spike patterns (i.e., doublets/bursting)) and ion channel conductance (sodium and calcium channel conductances are commonly used). 15. Optional: Many protocols use retrobead labeled and/or genetically encoded fluorescent proteins to label neuronal subtypes. For these, the process of approaching a cell, establishing a gigaseal and breaking into a cell is identical to the steps described above. However, finding fluorescently labeled cells requires the use of additional microscope equipment, including (but not limited to) LED systems and appropriate wavelength filters. These systems allow for visual observation of fluorescent cells within slices. The easiest way to find a healthy and fluorescently labeled cell is to find the brain area of interest and identify a healthy cell for approach (as described above). After finding a healthy cell turn on the LED/equivalent system to verify if that cell is fluorescent. If the cell shows fluorescence, turn off the illumination system and proceed as described above. If the cell does not contain fluorescence, return back to IR/DIC and

122

Jonna M. Leyrer-Jackson et al.

Fig. 11 (a) IR/DIC visualization of a layer V pyramidal cell visualized with white light and (b) under blue light to identify GFP labeled retrobeads located within the soma

continue looking for another healthy cell in the area and repeat. As opposed to looking for fluorescent cells and then determining if they are healthy, identifying healthy cells first will quicken the searching process. Figure 11 depicts a retrobead labeled fluorescent cell body under bright-field (A) and blue light illumination with an LED (to visualize GFP retrobeads). For additional information regarding Adult vs. Young Animals, see Note 8. 3.4 Isolating NMDA and AMPA ReceptorMediated Potentials/ Currents 3.4.1 Solution Alterations

3.4.2 Pharmacological Isolation

Alterations in recording solutions have proven most beneficial for isolating NMDA receptor-mediated currents. As discussed above, NMDA receptors are blocked by magnesium ions at resting membrane (more hyperpolarized) potentials. By decreasing the concentration of magnesium in the recording solution, magnesium is prevented from blocking the NMDA receptor pore; therefore, NMDA receptors can be robustly activated at more hyperpolarized potentials. It is common to completely remove magnesium from the recording solution; however, many protocols and solution preparations only remove roughly 50 to 75 percent of the normal magnesium concentration to keep the solution physiological. The most common way for isolating NMDA and AMPA receptormediated currents is by using pharmacological agents. For isolating AMPAR-mediated and NMDAR-mediated responses, AMPAR agonists and NMDAR antagonists or NMDAR agonists and AMPAR antagonists are used, respectively. Most commonly, these pharmacological agents are added to bathing solutions (recording solutions/aCSF) that are continuously flowing across the tissue (as discussed above). This method is referred to as “bath applied.” An additional method for applying these compounds is through puff application, which allows for localized application of the compounds with the accuracy of targeting a single neuron, even to the level of a single synapse. This technique allows researchers to

Patch-Clamp Isolation of Glutamatergic Receptors in Addiction

123

Table 1 Commonly used pharmacological agents for isolating AMPA and NMDA receptor-mediated responses and application routes Concentration used (μM)

Common application methods

APV/AP-5 (DL-2-amino- NMDA receptor antagonist 5-phosphonopentanoic acid)

30–50

Bath and puff applied

DNQX AMPA receptor (6,7-dinitroquinoxalineantagonist 2,3-dione)

10–20

Bath and puff applied

CNQX (6-cyano-7nitroquinoxaline-2,3dione)

AMPA receptor antagonist

10–40

Bath and puff applied

NMDA (N-methyl-Daspartate)

NMDA receptor agonist

5–100

Bath and puff applied

Spermine

AMPA receptor (Ca2+ permeable only) antagonist

25

Intracellular solution through recording micropipette

Other polyamines

AMPA receptor (Ca2+ permeable only) antagonists

Varies, depending Intracellular solution on the polyamine through recording used micropipette

Pharmacological agent

Receptor action

examine individualized responses and prevent influences from downstream and indirect effects, which could potentially be observed with bath application. Lastly, inclusion of pharmacological compounds within the intracellular solution allows for manipulation of receptors from the inside of the cell. Although less common, this technique allows for single-cell isolation of receptors and proves beneficial for blocking currents with a higher efficiency with less equipment (as used for puffing experiments) and is costefficient (less drug needed than bath application experiments). Table 1 outlines commonly used drugs for isolating AMPA and NMDA receptor-mediated responses, the concentration used, and the most common application route(s). Commonly used concentrations are optimal for activating the receptor stated in Table 1; alterations in this concentration may nonspecifically activate other receptor subtypes, especially in instances of greater concentrations. 3.4.3 Membrane Potential AMPA and NMDA Current Isolation

Due to the kinetics of these channels and their vastly different activation potentials, it is common to isolate AMPA and NMDA receptor-mediated currents by holding the recording cell at different membrane potentials. AMPA receptors can be readily activated at hyperpolarized potentials; thus, recording at membrane potentials between 80 mV and 70 mV is ideal for measuring maximal

124

Jonna M. Leyrer-Jackson et al.

AMPAR-mediated currents. However, because NMDA receptors are blocked by magnesium ions at hyperpolarized potentials, it is readily difficult to obtain NMDAR-mediated currents at hyperpolarized potentials. Shifting the holding potential between 25 mV and 35 mV alleviates the magnesium block and allows for maximal NMDAR-mediated currents to be measured. If this manipulation is performed, it is not necessary to lower magnesium concentrations in the recording buffer. It is necessary to keep in mind that there can be overlap of AMPA and NMDA receptor activation; therefore, it is highly recommended to use pharmacological isolation in combination with alterations in holding membrane potential to ensure complete isolation of each AMPA and NMDA receptor-mediated currents. 3.5 Various Induction/Recording Protocols 3.5.1 Somatic Current Injections

3.5.2 External Stimulation

One of the many ways to induce excitatory postsynaptic potentials (EPSPs) is by using somatic current injections. These protocols can be conducted without the use of external stimulation, transgenic animals, or viral vectors. While performing a somatic whole-cell patch-clamp recording, current can be injected through the recording pipet to induce cellular depolarization and therefore obtain EPSPs. After following the protocol above for obtaining a somatic recording, a somatic depolarizing current (typically around 50 pA) can be injected into the soma of the recorded neuron. These depolarizations will elicit membrane depolarization in the recorded cell closely resembling an externally (presynaptic transmitter release) stimulated EPSP. For example studies using this technique and these parameters, see [31, 32]. Using bath application of NMDA and AMPA receptor agonists and antagonists (see Table 1), the currents mediated by these receptors can be pharmacologically isolated. Additionally, alterations in membrane potentials (as discussed above) can aid in isolation of these receptorinduced currents. Although less commonly used, somatic current injections can be very beneficial for identifying the effects of drugs and neurotransmitters on AMPA and NMDA receptor-mediated currents. One benefit of slice electrophysiology is the study of functional synapses, which are preserved in slice preparations. Therefore, stimulation of afferent fibers can induce excitatory and inhibitory synaptic responses in a postsynaptic cell. One common induction method of excitatory and inhibitory currents is by using external stimulation. External stimulation is conducted when a stimulating electrode is placed within 50–100 μm from the recorded cell and triggered using an electrical stimulation. Most commonly, this method uses concentric bipolar stimulating electrodes, tungsten wire, or stainless steel bipolar electrodes or micropipettes containing high (1–5 M) NaCl concentrations (Fig. 12). Once a healthy/ approachable cell is found (using the protocol above), move the

Patch-Clamp Isolation of Glutamatergic Receptors in Addiction

125

Fig. 12 Commonly used external stimulating electrodes in slice electrophysiology. (a) A recording pipette made from 10-mm length, 1.5 mm O.D., and 0.85 mm I.D. capillary glass (Sutter Instruments, Novato, CA) filled with 3 M sodium chloride. (b) A bipolar stimulating electrode (FHC, Bowdoin, ME) used for stimulating groups of cells

objective to low power and place the stimulating device within 50–100 μm from the cell body. Once placed, an approach (as outlined above) should be conducted. After establishing a seal, breaking into the cell and establishing cell health, a “stimulus current— evoked response/output” curve should be conducted by increasing stimulus intensity while measuring the evoked response until maximal response is reached. The stimulus intensity should then be adjusted to establish an unsaturated response near the midrange of this curve. If necessary, the stimulating electrode can be removed and relocated either closer to or farther away from the cell to establish an optimal sigmoidal-shaped curve. However, removing and relocating a stimulating electrode while already recording from a single cell increase the probability of disrupting the seal and losing the recording. Moving the electrode should be a last resort and deemed absolutely necessary for the experiment to continue. 3.5.3 Optogenetics

Optogenetic techniques have revolutionized the field of neurobiology allowing scientists the ability to manipulate neuronal cell populations both in vitro and in vivo with unsurpassed cell type and spatial and temporal resolution. Using genetically encoded

126

Jonna M. Leyrer-Jackson et al.

light-sensitive proteins originally isolated from microalgae or other organisms, neurons can be selectively activated using different wavelengths of light. Recent advances in the field of optogenetics allow for both activation and inhibition of specific populations using different opsins that allow for the passage of sodium and chloride/protons, respectively. The diversity of optogenetic proteins currently available are powerful for ion selection and allow for high levels of spatial and temporal resolution for studying synaptic connectivity. One method of optogenetics uses a nonspecific cell-type approach. This approach allows labeling and opsin expression in axonal projections/afferents from specific brain areas without celltype specificity via the use of pan-neuronal promoter elements such as synapsin. Opsins are most commonly delivered to a brain area via packaging into viral vectors such as adeno-associated viruses (AAVs) and retroviruses, although genetically engineered animals expressing opsins in various cell types are also available as discussed below. Once delivered to the brain region of interest, the viral vector containing the opsin-coding sequence transfects host neurons where it either becomes incorporated into the host genome or exists in an episomal state in the nucleus. Generally, protein expression occurs within 2–3 weeks postinjection, though more rapid expression can be obtained with herpes-type viruses. Most laboratories perform recordings from animals 2 weeks postinjection in order to keep animals within the optimal age range for patch-clamp recordings as well as to allow for recovery from surgical procedures [33, 34]. After 2 weeks postinjection, cell bodies and their axonal projections will contain light-activated channels tagged with a fluorescent molecule (i.e., green fluorescent protein, GFP) as a reporter. With this approach, experimenters should expect to see prominent labeling in every area that neurons of the injected area project to, with greater concentration/intensity in areas where there is stronger innervation. This technique has proved beneficial for studying projection patterns and downstream effects/influences of specific brain regions, regardless of cell type. Additionally, if studies do not require neuronal subtype specificity and focus on regional projections, this methodology does not require transgenic animals and is therefore extremely cost-effective. Viral vectors for these experiments are commonly obtained from various academic vector core facilities (i.e., University of Pennsylvania and University of North Carolina at Chapel Hill) as well as commercial vendors (i.e., Addgene, Cambridge, MA). A major question within the field of neuroscience is how brain functions are driven by specific neuronal cell types. For this reason, Cre-dependent animals have become a widely used tool for expression of opsins in specific neuronal populations with exceptional precision. Cell-type-specific targeting of these proteins is generally

Patch-Clamp Isolation of Glutamatergic Receptors in Addiction

127

achieved through the Cre/lox recombinase system where viral vectors are used to deliver opsin proteins in a Cre-dependent manner, resulting in expression only in Cre-positive cells. Cre-dependent mouse and rat lines have been developed for hundreds of neuronal subtypes, and many others are currently in the process of being developed. These animals have greatly enhanced the field of optogenetics and have proven extremely beneficial by allowing scientists the ability to target specific neuronal populations, and determine cell-specific roles in neural networks. Although these animals are excellent for labeling neuronal population subtypes, they pose some limitations. Most commonly, incomplete coverage of neurons (i.e., not every cell within that population expresses the opsin) prevents experiments requiring complete labeling of that entire cell population. Additionally, each neuron can vary in the intensity of opsin (also true with non-Credependent expression) which can lead to inconsistencies between animals and cells. Additionally, Cre-dependent animals are often expensive, require breeding at the user’s institution, and are specific for only one neuronal cell type, limiting the possible number of experiments. Furthermore, a major drawback of transgenic animals is the continuous expression of Cre throughout development, which may have detrimental effects on physiological function. Although relatively unexplored, some have suggested that Cre activity can have dramatic developmental deficits as well as promote apoptosis in embryonic tissues [35]. While these animals remain on the forefront of technological advances in neurophysiology, it is important to acknowledge that like other techniques, the use of these animals comes with limitations and potential drawbacks. The variety of light-activated proteins available for optogenetic techniques is growing daily. However, there are proteins that are more commonly used for excitation and inhibition with their kinetics and expression capabilities already extensively characterized. The most prominently used excitatory opsin is channelrhodopsin2 (ChR2), a cation-selective ion channel that allows primarily sodium to pass into the cell upon activation. This channel has the ability to be expressed in a wide variety of tissue types with prominent expression capabilities in mammalian brain tissue and is widely used in both nonspecific and cell-specific neuronal expression. Although there are known light-induced desensitization properties of ChR2, it is primarily used for rapid manipulation studies. Other ChR2 variants and opsin proteins, with varied kinetics and activation properties, have been developed and are used for studying specific neuronal properties. Optogenetics can also be used for inhibition of neuronal firing. Most commonly, halorhodopsin (NpHR) is a light-activated chloride channel that leads to hyperpolarization of the cell upon opening. Known for its step-like and stable photocurrents, NpHR

128

Jonna M. Leyrer-Jackson et al.

achieves neuronal inhibition with greater efficiency than other inhibitory opsins. Membrane hyperpolarizations of greater than 100 mV are routinely recorded using NpHR [36], making it one of the most robust optogenetic inhibitors in the field of optogenetics. Other inhibitory proteins such as archaerhodopsin can also be utilized. 3.6 In Vivo Relevance of In Vitro Recording Protocols

Although this chapter is primarily focused on single-cell whole-cell patch-clamp recordings conducted in mammalian brain slices, it is essential to ensure that results observed in vitro translate to in vivo studies. Thus, studies identifying in vivo effects of drug use on glutamatergic receptors have proven beneficial for validating results observed in whole-cell patch-clamp slice physiology. Changes in NMDA and AMPA receptor-mediated responses due to addictive substances in vivo have been heavily studied in animal models and have proven fruitful for validating in vitro findings, for contributing to the understanding the mechanisms underlying addiction, and beneficial for linking behavior with alterations in NMDA and AMPA receptors due to drug intake. The use of whole-cell patch-clamp electrophysiology in brain slices allows for the study of NMDA and AMPA receptors with controlled parameters yielding less variability between animals and cells. As discussed above, in vitro experiments using slice physiology have shown that commonly abused substances can elicit changes in synaptic plasticity. These results have been partially validated in vivo in dopaminergic neurons of the VTA, where cocaine exposure induces AMPAR-mediated LTP within 5 days of a single exposure and was no longer observed after 10 days. These results indicate that changes in LTP from a single cocaine exposure are transient effect that is often difficult to observe in slices [37]. In slice physiology, LTP induction protocols have been reported to promote exocytosis of AMPA receptors to the plasma membrane [1, 12], a phenomenon more difficult to observe in vivo. Interestingly, drug consumption (especially cocaine) also induces AMPA receptor recruitment to the plasma membrane, specifically those lacking GluN2A and GluN2B subunits. A major pitfall of slice physiology alone is the inability to directly link behavior with subcellular changes. The use of in vivo studies has shown that drug abuse induces cellular alterations, which have been further characterized using pharmacological isolation in slices. Recent technological advances have even linked behavior with cellular alterations. Interestingly, optogenetic inhibition of prelimbic afferents of the medial prefrontal cortex, targeting the core of the nucleus accumbens, prevents reinstatement of cocaine in rats and decreases dendritic spine head diameter as well as AMPA/NMDA ratios [38]. Recently, low-frequency deep brain stimulation within the nucleus accumbens in combination with a dopaminergic D1 receptor antagonist has been shown to abolish

Patch-Clamp Isolation of Glutamatergic Receptors in Addiction

129

drug-seeking behavior in mice. This protocol induces mGluR1dependent LTD thought to depotentiate synapses on D1 receptor-containing medium spiny neurons, reversing druginduced plasticity [39]. Additionally, NMDA receptor antagonists administered in vivo have been shown to impair lever pressing for cocaine self-administration, while AMPA receptor antagonists impair lever pressing for cocaine, alcohol, and amphetamine selfadministration [28], directly linking glutamatergic receptors to addictive behaviors. NMDA receptor antagonists have also been shown to block locomotor effects induced by cocaine [40], further indicating the role of NMDA receptors in drug-induced behavior. Interestingly, an increase in spine head density, AMPA/NMDA ratios, and GluA1 (AMPA) as well as GluN2A and GluN2B (NMDA) receptor subunits were observed in spiny neurons of the nucleus accumbens following self-administration of nicotine in rats [41]. In combination, these studies have contributed significantly to linking behavioral and cellular aspects of addiction while also suggesting the importance in studying glutamatergic receptors in addiction phenomena. One recent theory is that cocaine addiction increases plasticity in the brain through alterations of NMDA receptor subunits, reverting the neural environment back to a developmental state within the nucleus accumbens. Silent synapses are extremely abundant within the developing brain and contain a large number NMDA receptors consisting of GluN2B subunits and GluA2containing AMPA receptors. However, as discussed above, over the course of development, NMDA receptors containing GluN2B subunits are downregulated, making the synapse less “plastic.” In vivo exposure to cocaine has been shown to generate silent synapses through insertion of new GluN2B-containing NMDARs within the membrane [13, 42], thus making the synapse more prone to synaptic modifications in the presence of cocaine. Enhanced plasticity of these synapses is thought to promote long-lasting memory modifications induced by the drug itself. Furthermore, GluA2containing AMPA receptors have also been shown to be upregulated due to cocaine withdrawal. The enhanced conductance of GluA2-containing AMPA receptors and an increase in number of AMPARs are thought to increase plasticity and facilitate cocaineinduced memory formation, increasing drug craving and relapse in the presence of certain cocaine-related cues [43]. Together, these studies indicate the importance of studying AMPA and NMDA receptor subunits in addiction, as they may lead to new drug targets for preventing relapse. For a review of drug-induced glutamatergic subunit alterations, see Russo et al. [44]. While it has been established in vivo and in vitro that AMPA and NMDA receptors play a role in drug-seeking and addictionrelated physiological changes, a complete understanding of the mechanisms remains unknown. Using a combination of in vivo

130

Jonna M. Leyrer-Jackson et al.

and in vitro approaches remains most beneficial for studying unknown aspects of these disorders. Importantly, establishing the effects of addictive substances on currents mediated by glutamatergic receptors and subcellular mechanisms for synaptic modifications that may underlie addictive behaviors remains an important aspect in understanding addiction. 3.7 Analysis of Electrophysiological Currents

Analysis of AMPA and NMDA receptor-mediated currents in vitro can be used to: 1. Estimate the ratio of AMPA to NMDA receptors within a postsynaptic cell and how pharmacological agents and stimulation protocols can alter it. 2. Determine how pharmacological agents can alter the kinetics of these receptors. 3. Determine if pharmacological agents preferentially alter one current over another (i.e., effects on AMPA versus NMDA receptors) providing insight to their overall effects on cellular function (synaptic transmission versus plasticity). 4. Lend insight into glutamatergic subunit alterations. 5. Lend insight into overall behavioral changes due to receptor number and kinetic alterations.

3.7.1 Analysis of AMPA/ NMDA Ratios

As discussed above, AMPA and NMDA receptors are responsible for very different physiological aspects (fast synaptic transmission and synaptic plasticity, respectively). For this reason, analysis of the AMPA/NMDA ratio has yielded beneficial in determining the relative expression of AMPA and NMDA receptors at the synapse. Because it is difficult to determine the strength of excitatory synapses between different cell types and connections, the AMPA/NMDA ratio provides a normalized measurement for such comparisons due to its independence from positioning electrodes or a number of stimulated synapses. It is common to measure an intact (“whole”) EPSP, containing both AMPA and NMDA receptor-mediated components, and applying either an AMPA or NMDA receptor antagonist to isolate the NMDA or AMPA receptor-mediated current, respectively. Using digital subtraction, the isolated current can then be subtracted from the “whole” EPSP, where the remainder is mediated by the other receptor subtype. For example, if an AMPA antagonist was used, the second measurement should contain only the NMDA receptor-mediated current. When subtracted from the “whole” EPSP, the remainder would be current mediated by the AMPA receptor. Once isolated, the amplitude is then taken for each current, and a ratio is taken (Fig. 13). Alterations of this ratio can lend insight into drug-induced behavioral effects.

Patch-Clamp Isolation of Glutamatergic Receptors in Addiction

131

Fig. 13 Commonly measured characteristics of glutamatergic responses. (a) Decay time is measured as the time it takes for the response to reach 37% of the initial peak amplitude. (b) Rise time is calculated as the time it takes from initial deflection for the response to reach peak amplitude. (c) Response amplitude is measured from baseline to peak. In current clamp, this measurement is recorded in millivolts, and in voltage clamp, this measurement is recorded in amps (pA or nA). (d) The area under the curve is measured as the total area under the curve and above baseline (dotted line) 3.7.2 Analysis of Channel Kinetics

4

Most commonly, measurements of NMDA and AMPA receptor kinetics are indicated by changes in rise time, decay time, amplitude, and area under the curve. All of these measurements can be taken from the same current trace (Fig. 13). However, each measurement can indicate very different aspects of the current. Changes in rise time suggest alterations in activation kinetics of the receptor making its activation time quicker (decrease in rise time) or slower (increase in rise time). An alteration in decay time suggests a change in inactivation kinetics of the channel, while changes in current amplitude and area under the curve indicate changes in channel permeability. Additionally, changes in decay time can be indicative of GluN2B subunit composition, where an increase in NMDA decay time can indicate an increase in extrasynaptic GluN2B subunit expression, where these subunits have been shown to have preferred distribution patterns [25, 45]. In combination, these analyses can lend insight into how channels may be undergoing subunit and kinetic alterations. These measurements are commonly used in experiments identifying pharmacological effects on AMPA and NMDA receptor-mediated currents and prove beneficial in determining where pharmacological agents may be having their effects on a synaptic and subcellular level.

Notes 1. ATP and GTP readily degrade at room temperature; therefore, once they have been added to solution, the process should be conducted as quickly as possible. For this reason, it is also common to add ATP and GTP just before use to ensure they are not degraded. 2. There are many different iterations of these solutions. For example, many researchers use a cutting solution completely free of sucrose, as it is known to alter endogenous dopamine release in slices. It is essential to make solutions based on slice

132

Jonna M. Leyrer-Jackson et al.

preparation and physiological measurements. Intracellular solutions can vary widely depending on the cell type and the measurement to be recorded. Additionally, alterations in recording solution are also common (e.g., removing magnesium for the study of NMDA receptors—to be discussed below). It is always important to review previous literature regarding an area of interest before choosing the correct solutions. 3. Healthy cells: Cell bodies are round; red blood cells have normal “donut”-shaped appearance (suggesting osmolarity is acceptable); lack of debris floating or coming off from the tissue. 4. Unhealthy cells: Cell bodies are swollen or crinkled; nuclei of cells are visually apparent; red blood cells are oddly shaped (e.g., cell bodies swollen or crinkled); abundance of debris floating or coming off from tissue. 5. Generally, healthiest cells are 50–100 μm below the slice surface. These cells are deep enough in the slice that they remain relatively unaffected and undamaged throughout the slicing process. When looking for a healthy cell to approach, the deeper the cell is located, the higher likelihood that it is viable and healthy for longer recordings. 6. If the cell moves away from the pipette, it is either unhealthy or the approach is too close to the edge of the tissue. If this occurs, slightly retract the pipette and retry the approach. If the cell no longer slips away from the pipette, it can be assumed that the cell is healthy. However, if the cell still slips, it is likely unhealthy and should be disregarded. The degree of cell dimpling will vary across cell types and animal age. However, the approach of the cell should always be directed toward the largest part of the cell body. For example, if recording from a pyramidal (tear drop shaped) cell within the PFC, the approach should be aimed toward the large end of the tear drop. This idea holds true for all cells—the larger the surface area, the easier it is to obtain a seal. 7. There are many options for recording software (e.g., Axon, Axograph, etc.) which provide an interface between the amplifier, digitizer, and any other patch-clamp electronics. Digitizers allow for compensation of capacitance and resistance through way of manual control (i.e., nobs and switches present on the digitizer itself) or digital control through the computer software. It is essential to explore the capabilities of the recording software, as some software is only compatible with certain digitizers and amplifiers. For healthy cells, Table 2 outlines normal pipet resistance (after break-in), membrane resistance, and membrane capacitance, and resting membrane potentials

Patch-Clamp Isolation of Glutamatergic Receptors in Addiction

133

Table 2 Electrophysiological characteristics of commonly studied cell types in addiction

Brain region

mPFC

Hypothalamus

Nucleus accumbens

Pipette resistance (mΩ)

Membrane resistance (mΩ)

Layer II/III pyramidal cell