Methods in Drug Abuse Research: Cellular and Circuit Level Analyses (Methods and New Frontiers in Neuroscience)

METHODS & NEW FRONTIERS IN NEUROSCIENCE Series Editors Sidney A. Simon, Ph.D. Miguel A.L. Nicolelis, M.D., Ph.D.

Published Titles Apoptosis in Neurobiology Yusuf A. Hannun, M.D., Professor of Biomedical Research and Chairman/Department of Biochemistry and Molecular Biology, Medical University of South Carolina Rose-Mary Boustany, M.D., tenured Associate Professor of Pediatrics and Neurobiology, Duke University Medical Center Methods for Neural Ensemble Recordings Miguel A.L. Nicolelis, M.D., Ph.D., Professor of Neurobiology and Biomedical Engineering, Duke University Medical Center Methods of Behavioral Analysis in Neuroscience Jerry J. Buccafusco, Ph.D., Alzheimer’s Research Center, Professor of Pharmacology and Toxicology, Professor of Psychiatry and Health Behavior, Medical College of Georgia Neural Prostheses for Restoration of Sensory and Motor Function John K. Chapin, Ph.D., Professor of Physiology and Pharmacology, State University of New York Health Science Center Karen A. Moxon, Ph.D., Assistant Professor/School of Biomedical Engineering, Science, and Health Systems, Drexel University Computational Neuroscience: Realistic Modeling for Experimentalists Eric DeSchutter, M.D., Ph.D., Professor/Department of Medicine, University of Antwerp Methods in Pain Research Lawrence Kruger, Ph.D., Professor or Neurobiology (Emeritus), UCLA School of Medicine and Brain Research Institute Motor Neurobiology of the Spinal Cord Timothy C. Cope, Ph.D., Professor of Physiology, Emory University School of Medicine Nicotinic Receptors in the Nervous System Edward D. Levin, Ph.D., Associate Professor/Department of Psychiatry and Pharmacology and Molecular Cancer Biology and Department of Psychiatry and Behavioral Sciences, Duke University School of Medicine Methods in Genomic Neuroscience Helmin R. Chin, Ph.D., Genetics Research Branch, NIMH, NIH Steven O. Moldin, Ph.D, Genetics Research Branch, NIMH, NIH Methods in Chemosensory Research Sidney A. Simon, Ph.D., Professor of Neurobiology, Biomedical Engineering, and Anesthesiology, Duke University Miguel A.L. Nicolelis, M.D., Ph.D., Professor of Neurobiology and Biomedical Engineering, Duke University

The Somatosensory System: Deciphering the Brain’s Own Body Image Randall J. Nelson, Ph.D., Professor of Anatomy and Neurobiology, University of Tennessee Health Sciences Center New Concepts in Cerebral Ischemia Rick C. S. Lin, Ph.D., Professor of Anatomy, University of Mississippi Medical Center DNA Arrays: Technologies and Experimental Strategies Elena Grigorenko, Ph.D., Technology Development Group, Millennium Pharmaceuticals Methods for Alcohol-Related Neuroscience Research Yuan Liu, Ph.D., National Institute of Neurological Disorders and Stroke, National Institutes of Health David M. Lovinger, Ph.D., Laboratory of Integrative Neuroscience, NIAAA In Vivo Optical Imaging of Brain Function Ron Frostig, Ph.D., Associate Professor/Department of Psychobiology, University of California, Irvine Primate Audition: Behavior and Neurobiology Asif A. Ghazanfar, Ph.D., Primate Cognitive Neuroscience Lab, Harvard University

2345_frame_FM Page 4 Wednesday, October 23, 2002 2:16 PM

Library of Congress Cataloging-in-Publication Data Methods in drug abuse research : cellular and circuit level analyses / edited by Barry D. Waterhouse p. cm. -- (Methods & new frontiers in neuroscience) Includes bibliographical references and index. ISBN 0-8493-2345-2 (alk. paper) 1. Drugs of abuse--Research--Methodology. 2. Drug abuse--Research--Methodology. 3. Neurons. 4. Neural circuitry. I. Waterhouse, Barry D. II. Methods & new frontiers in neuroscience series. RM316 .M48 2002 615¢.78¢072--dc21

2002074127

This book contains information obtained from authentic and highly regarded sources. Reprinted material is quoted with permission, and sources are indicated. A wide variety of references are listed. Reasonable efforts have been made to publish reliable data and information, but the authors and the publisher cannot assume responsibility for the validity of all materials or for the consequences of their use. Neither this book nor any part may be reproduced or transmitted in any form or by any means, electronic or mechanical, including photocopying, microfilming, and recording, or by any information storage or retrieval system, without prior permission in writing from the publisher. All rights reserved. Authorization to photocopy items for internal or personal use, or the personal or internal use of specific clients, may be granted by CRC Press LLC, provided that $1.50 per page photocopied is paid directly to Copyright Clearance Center, 222 Rosewood Drive, Danvers, MA 01923 USA. The fee code for users of the Transactional Reporting Service is ISBN 0-8493-2345-2/03/$0.00+$1.50. The fee is subject to change without notice. For organizations that have been granted a photocopy license by the CCC, a separate system of payment has been arranged. The consent of CRC Press LLC does not extend to copying for general distribution, for promotion, for creating new works, or for resale. Specific permission must be obtained in writing from CRC Press LLC for such copying. Direct all inquiries to CRC Press LLC, 2000 N.W. Corporate Blvd., Boca Raton, Florida 33431. Trademark Notice: Product or corporate names may be trademarks or registered trademarks, and are used only for identification and explanation, without intent to infringe.

Visit the CRC Press Web site at www.crcpress.com © 2003 by CRC Press LLC No claim to original U.S. Government works International Standard Book Number 0-8493-2345-2 Library of Congress Card Number 2002074127 Printed in the United States of America 1 2 3 4 5 6 7 8 9 0 Printed on acid-free paper

2345_frame_FM Page 5 Wednesday, October 23, 2002 3:20 PM

Series Preface Our goal in creating the Methods & New Frontiers in Neuroscience series is to present the insights of experts on emerging experimental techniques and theoretical concepts that are, or will be, at the vanguard of neuroscience. Books in the series cover topics ranging from methods to investigate apoptosis, to modern techniques for neural ensemble recordings in behaving animals. The series also covers new and exciting multidisciplinary areas of brain research, such as computational neuroscience and neuroengineering, and describes breakthroughs in classical fields like behavioral neuroscience. We want these books to be the books every neuroscientist will use in order to get acquainted with new methodologies in brain research. These books can be given to graduate students and postdoctoral fellows when they are looking for guidance to start a new line of research. Each book is edited by an expert and consists of chapters written by the leaders in a particular field. Books are richly illustrated and contain comprehensive bibliographies. Chapters provide substantial background material relevant to the particular subject. Hence, they are not only “ methods books,” but they also contain detailed “tricks of the trade” and information as to where these methods can be safely applied. In addition, they include information about where to buy equipment and about web sites helpful in solving both practical and theoretical problems We hope that as the volumes become available, the effort put in by us, by the publisher, by the book editors, and by individual authors will contribute to the further development of brain research. The extent that we achieve this goal will be determined by the utility of these books. Sidney A. Simon, Ph.D. Miguel A.L. Nicolelis, M.D., Ph.D. Series Editors

2345_frame_FM Page 6 Wednesday, October 23, 2002 2:16 PM

2345_frame_FM Page 7 Wednesday, October 23, 2002 2:16 PM

Dedication To Kathy for her unwavering love, companionship, and support through all my scientific endeavors.

2345_frame_FM Page 8 Wednesday, October 23, 2002 2:16 PM

2345_frame_FM Page 9 Wednesday, October 23, 2002 2:16 PM

About the Editor Barry D. Waterhouse is a professor in the Department of Neurobiology and Anatomy and an associate dean of biomedical graduate studies at Drexel University College of Medicine (formerly MCP-Hahnemann University School of Medicine). After receiving his B.S. degree in biology in 1971 from Muhlenberg College, Dr. Waterhouse completed his Ph.D. in pharmacology at Temple University in 1977. From 1977 through 1987 he worked at Southwestern Medical School, University of Texas at Dallas, rising from postdoctoral fellow, to instructor, and then finally to assistant professor. In 1987 he was recruited to the Department of Physiology and Biophysics as an associate professor at Hahnemann University School of Medicine, where in 1988 he developed and was subsequently appointed director of the university's graduate program in neuroscience, a post he held until 1994. In 1992 he was promoted to professor of physiology and biophysics, and in 1994, when Hahnemann University merged with Medical College of Pennsylvania (MCP), Dr. Waterhouse was invited to join the Department of Neurobiology and Anatomy in the newly formed university. He continued as director of the neuroscience graduate program at MCP-Hahnemann until 2001 and also served as vice-chair of the Department of Neurobiology and Anatomy from 1999 to the present. He was elected to the American College of Neuropsychopharmacology in 1996 and to the College on Problems of Drug Dependence in 1995. Throughout his research career Dr. Waterhouse has focused on the neurobiology of central monoaminergic systems and psychostimulant drug actions.

2345_frame_FM Page 10 Wednesday, October 23, 2002 2:16 PM

2345_frame_FM Page 11 Wednesday, October 23, 2002 2:16 PM

Contributors Michael H. Baumann Medications Discovery Research Branch National Institute on Drug AbuseIntramural Research Program National Institutes of Health Baltimore, Maryland

Alexander F. Hoffman Cellular Neurobiology Branch National Institute on Drug AbuseIntramural Research Program National Institutes of Health Baltimore, Maryland

Craig W. Berridge Department of Psychiatry University of Wisconsin Madison, Wisconsin

Patricia H. Janak Ernest Gallo Clinic and Research Center Department of Neurology The University of California at San Francisco San Francisco, California

Jason J. Burmeister Center for Sensor Technology University of Kentucky Chandler Medical Center Lexington, Kentucky David M. Devilbiss Department of Neurobiology and Anatomy MCP-Hahnemann University Philadelphia, Pennsylvania Steven I. Dworkin Department of Psychology University of North Carolina at Wilmington Wilmington, North Carolina Greg A. Gerhardt Department of Anatomy and Neurobiology University of Kentucky Lexington, Kentucky

Laura L. Peoples Department of Psychology, Neuroscience Graduate Group University of Pennsylvania Philadelphia, Pennsylvania John J. Rutter Department of Biological Sciences Truman State University Kirksville, Missouri Michael F. Salvatore Department of Anatomy and Neurobiology University of Kentucky Lexington, Kentucky

2345_frame_FM Page 12 Wednesday, October 23, 2002 2:16 PM

Dustin J. Stairs Department of Psychology University of North Carolina at Wilmington Wilmington, North Carolina

Barry D. Waterhouse Department of Neurobiology and Anatomy MCP-Hahnemann University Philadelphia, Pennsylvania

2345_frame_FM Page 13 Wednesday, October 23, 2002 2:16 PM

Table of Contents Chapter 1 Overview....................................................................................................................1 Barry D. Waterhouse and Laura L. Peoples Chapter 2 Self-Administration of Drugs of Abuse ..................................................................17 Steven I. Dworkin and Dustin J. Stairs Chapter 3 Application of In Vivo Microdialysis Methods to the Study of Psychomotor Stimulant Drugs .......................................................................................................51 Michael H. Baumann and John J. Rutter Chapter 4 In Vivo Voltammetry in Drug Abuse Research .......................................................87 Michael F. Salvatore, Alexander F. Hoffman, Jason J. Burmeister, and Greg A. Gerhardt Chapter 5 Extracellular Single Unit Recording Strategies for Investigating the Actions of Drugs of Abuse in Anesthetized Animals .............................................................111 Barry D. Waterhouse Chapter 6 Application of Many-Neuron Microelectrode Array Recording to the Study of Reward-Seeking Behavior .....................................................................................143 Patricia H. Janak Chapter 7 Application of Chronic Extracellular Recording to Studies of Drug Self-Administration................................................................................................161 Laura L. Peoples Chapter 8 Determination of Drug Actions on Multiple Simultaneously Recorded Neurons across Functionally Connected Networks .............................................................213 David M. Devilbiss and Barry D. Waterhouse

2345_frame_FM Page 14 Wednesday, October 23, 2002 2:16 PM

Chapter 9 Pharmacological Investigations of Neural Mechanisms Underlying Amphetamine-Like Stimulant-Induced Arousal: Involvement of Noradrenergic Systems ..................................................................................................................239 Craig W. Berridge Index ......................................................................................................................271

2345_frame_MASTER.book Page 1 Wednesday, October 23, 2002 1:03 PM

1 Overview Barry D. Waterhouse and Laura L. Peoples CONTENTS 1.1

Introduction ......................................................................................................1 1.1.1 Theories of Drug Addiction.................................................................2 1.1.1.1 Incentive Motivation Theory of Addiction...........................2 1.1.1.2 Hedonic Dysregulation of Reward and Allostasis ...............3 1.2 Rationale for Cellular and Circuit Levels of Analysis of Drugs of Abuse ....5 1.2.1 Studies of Cells within Circuits...........................................................6 1.2.2 Studies in Awake Animals ...................................................................7 1.2.3 Delineation of Mechanisms That Contribute to Behavior ..................7 1.2.4 Mechanisms That Mediate Drug Effects on Behavior........................8 1.3 Cellular and Circuit Level Analysis of Drugs of Abuse .................................8 1.3.1 Current Methods ..................................................................................8 1.3.2 Future Directions..................................................................................9 Acknowledgements....................................................................................................9 References................................................................................................................10

1.1 INTRODUCTION Drug addiction is a progressive disorder characterized by a transition from controlled to uncontrolled and compulsive drug seeking that continues despite knowledge of adverse consequences (Hoffman and Goldfrank, 1990; Leshner, 1997; McGinnis and Foege, 1999). It is also a chronic relapse disorder. Periods of successful drug abstinence for many individuals end with relapse to compulsive drug use. Drug addiction is a devastating disorder that has severe health costs to both the individual and the public (McLellan et al., 2001). Although environmental variables can influence an individual’s risk for developing addiction, human and animal research show that addiction is fundamentally a disorder of the brain. Application of neuroscience approaches to the study of addiction is thus an integral part of efforts to understand and ultimately to treat the disorder. Two issues that are central to understanding the problem of drug abuse and addiction are 1) identification of drug actions that contribute to an initially positive drug experience and 2) elucidation of the neural mechanisms underlying the progression of addiction and the development of drug craving. Acute self-administration of addictive compounds produces a multitude of

1

2345_frame_MASTER.book Page 2 Wednesday, October 23, 2002 1:03 PM

2

Methods in Drug Abuse Research: Cellular and Circuit Level Analyses

transient effects on the brain, only some of which contribute to a positive drug experience and the desire to repeat that experience in a social setting. However, chronic use of these substances leads to long-lasting changes in nervous system function that are thought to contribute to the development and maintenance of compulsive and uncontrollable drug seeking.

1.1.1 THEORIES

OF

DRUG ADDICTION

Investigation of the acute and chronic effects of addictive drugs has led to the development of a number of formal theories of addiction. Two that are particularly influential today are the incentive motivation theory (DiChiara, 1998; Everitt et al., 1999; Robbins and Everitt, 1999; Robinson and Berridge, 1993; Stewart et al., 1984; Stewart, 1992) and the hedonic dysregulation theory (Koob and LeMoal, 1997). Evidence consistent with each theory can be found in the animal and human literature on addiction; however, important predictions of each have yet to be tested fully. Although the theories differ in a number of aspects, they are not necessarily mutually exclusive and actually share at least two common assumptions. First, it is proposed that repeated exposure to drugs produces long-lasting changes in the brain that contribute to the development of compulsive and uncontrollable drug taking. Second, these changes are proposed to occur in multiple regions and at several levels of organization, from the level of molecular regulation of protein synthesis and cellular function to the level of local neural networks and circuits involving interactions between multiple brain structures. These theories, and an additional hypothesis that is gaining increasing influence, are reviewed and critically evaluated relative to existing data in a number of publications (Robinson and Berridge, 1993; Jentsch and Taylor 1999; Wise, 1999; Koob and LeMoal, 2001). They will thus be only briefly summarized here because they are relevant to understanding the utility of the research methods described in this volume. 1.1.1.1 Incentive Motivation Theory of Addiction Incentive motivation theories of drug addiction propose that the disorder reflects a pathological responsivity of individuals to the influences of drug-associated conditioned stimuli on behavior. The theories further assert that the abnormal responsivity to drug stimuli is caused by acute and long-lasting actions of addictive drugs on brain. Acute drug actions are proposed to amplify mechanisms that contribute to stimulus–reward learning and lead to abnormally powerful conditioning of stimuli associated with drug administration. It is also proposed that long-lasting neuroplasticity induced by drugs facilitates this drug-induced amplification of learning (for review see DiChiara, 1998; Everitt et al., 1999; Robbins and Everitt, 1999; Robinson and Berridge, 1993; Stewart et al., 1984; Stewart, 1992). These proposals are based, in part, on evidence that stimuli associated with drugs undergo conditioning and can facilitate drug seeking in both animals and humans (Arroyo et al., 1998; Davis and Smith, 1987; deWit and Stewart, 1981; Ehrman et al., 1992; Goldberg et al., 1976; Ranaldi and Roberts, 1996; Stewart et al., 1984; Tiffany, 1990). Additionally, animal studies show that acute actions of addictive drugs amplify the gain of behavioral

2345_frame_MASTER.book Page 3 Wednesday, October 23, 2002 1:03 PM

Overview

3

responses to conditioned stimuli and may facilitate conditioning (Harmer and Phillips, 1998; O’Brien et al., 1998; Panililio et al., 1998; Robbins et al., 1989; Taylor and Robbins, 1986; Taylor and Horger, 1999; Weiss et al., 2000). Furthermore, repeated administration of addictive drugs sensitizes animals to various effects of those drugs including acute reinforcing effects and gain-amplifying effects on responsivity to conditioned stimuli (Lorrain and Vezina, 2000; Taylor and Horger, 1999; Wyvell and Berridge, 2001). 1.1.1.2 Hedonic Dysregulation of Reward and Allostasis The hedonic dysregulation theory (Koob and LeMoal, 1997 and 2001) proposes that addiction is a self-regulatory condition. The theory is based on principles of homeostatic self-regulation and allostasis (Sterling and Eyer, 1988) and has roots in the opponent process theory of motivation described by Solomon and Corbit (1974). Homeostasis corresponds to the mechanisms that maintain stability within physiological systems and hold all parameters of an organism’s internal milieu within adaptive limits. Certain parameters are held at a constant set point by local negative feedback responses to deviations from the set point. Other parameters are allowed to vary within a wide range so as to maintain balanced function within particular physiological systems. In contrast, the principle of allostasis involves the stabilization or balancing of function by a resetting of the set point in response to a chronic demand on homeostatic mechanisms that are insufficient to fully compensate for the deviations in the original set point. These allostatic changes can compensate for the demand, but they can also lead to an abnormal state when the demand is removed and are proposed to contribute directly to the development of drug addiction. More specifically, acute actions of addictive drugs are proposed to overactivate reward pathways in the brain. This overactivation leads to a homeostatic response that involves down-regulation of neurochemical systems involved in mediating the drug-induced overactivation. However, the homeostatic changes in reward neurotransmitters are hypothesized to be insufficient to maintain balanced function within the reward system. This insufficiency, in conjunction with a chronic demand on these homeostatic mechanisms, leads to the onset of an allostatic process. This process involves changes that tend toward reestablishing the balanced reward function by changing the reward set point that is normally guarded by homeostatic mechanisms. The increase in reward set point is proposed to be mediated, in part, by recruitment of nonreward systems, specifically brain stress systems that are normally involved in negative emotional states. The recruitment and changes in the brain stress systems tend to counteract the drug-induced overactivation of the reward system but engender a negative mood or state in the absence of drug. This negative mood state is thought to set up a negative reinforcement mechanism. That is, individuals begin to seek drug in order to avoid the negative mood state. It is proposed that these allostatic responses grow in magnitude with repeated drug use. Thus, addiction is proposed to be a feed-forward cycle in which increases in drug intake are followed by increases in the magnitude of the allostatic response and resetting of the reward set point, which leads to further increases in drug intake and the reinitiation of the cycle.

2345_frame_MASTER.book Page 4 Wednesday, October 23, 2002 1:03 PM

4

Methods in Drug Abuse Research: Cellular and Circuit Level Analyses

This theory is based primarily on two observations. First, acute withdrawal from addictive drugs is commonly associated with negative affective states including dysphoria, depression, irritability, and anxiety. Second, after cessation of intravenous self-administration, animals exhibit an increased threshold for intracranial selfstimulation (ICSS) reward (i.e., a higher level of stimulation is required for the stimulation to be reinforcing). Additional evidence consistent with the hypothesis is described by Koob and LeMoal (2001). 1.1.1.2.1 Hypoactivity in Cortical Inhibitory Mechanisms Neuroimaging studies in humans show that individuals addicted to drugs such as cocaine show evidence of reduced activity and lower cell density in cortical regions such as the anterior cingulate and orbitofrontal cortex (Volkow, 1991; Childress et al., 1999; Franklin et al., 2002). These and other cortical structures are involved in willed or executive control of behavioral selection. Executive control of behavior involves dynamic emotional and cognitive analyses of past and expected events and the influence of these analyses on decisions about future actions. This online willed control comes into play when automatic or habitual behaviors controlled by subcortical brain structures are not sufficient to guide adaptive behavior. Moreover, it can be important in the initiation of actions, persistence of adaptive actions in the absence of reward, and inhibition of impulses to engage in alternative but less beneficial behaviors. The anterior cingulate cortex (ACC) and orbitofrontal cortex contribute to the generation of emotion and to executive control of the influence of these emotions on behavior. Abnormalities in these brain regions are associated with a range of disorders involving disturbances in emotion and action. In light of these data, it is hypothesized that the cortical abnormalities observed in addicted individual contribute to addiction (Volkow, 1991; Childress et al., 1999; Rogers et al., 1999; Volkow and Fowler, 2000; London et al., 2000). Consistent with this hypothesis, addicted individuals exhibit symptoms associated with insults to the ACC and orbitofrontal cortices including anhedonia and an inability to make adaptive decisions regarding future actions (e.g., Grant et al., 1996; Rogers et al., 1999). In fact, a hallmark of drug addiction is compulsive and uncontrollable drug use, despite knowledge of adverse consequences. It is thus possible that addicted individuals, like others who suffer from hypoactivity in these brain regions, are unable either to experience normal affective responses to future events or to exert executive control over the adaptive influence of those emotions on selection of beneficial actions. Perhaps most important in the case of addiction, the deficiencies in emotional regulation may limit the ability of addicted individuals to inhibit responses to rewards, including drug, in order to avoid harm (e.g., incarceration or death). There is some evidence that these neural abnormalities are induced by drug exposure, but it is also possible that some or all of the brain abnormalities are present in the individual prior to drug exposure and perhaps enhance vulnerability to addiction. Finally, it has been proposed that addiction may reflect a combination of weakened cortical inhibitory mechanisms and the overactive (sensitized) responses of subcortical mediated responses to drug-associated stimuli (for review see Jentsch and Taylor, 1999).

2345_frame_MASTER.book Page 5 Wednesday, October 23, 2002 1:03 PM

Overview

5

1.1.1.2.2 Experimental Approaches for Studying Drugs of Abuse Research into the problem of drug abuse and drug addiction represents an interdisciplinary enterprise that to date includes social scientists, behavorists, biochemists, pharmacologists, physiologists, anatomists, neuroimaging specialists, and cell and molecular biologists. Within these disciplines many different techniques and experimental preparations have been utilized to address fundamental biological questions regarding the actions of acute and chronic drug administration. Initial studies using behavioral approaches characterized drug-related behaviors and defined assays for evaluating the reinforcing properties of addictive compounds. Biochemical and pharmacological investigations determined the cellular targets and time course of drug actions within brain tissue. Early anatomical and more recent imaging studies have identified specific pathways and regions of the brain that are activated directly or subsequent to acute or chronic drug administration or during drug withdrawal or craving. Experiments employing molecular biological techniques have identified gene products in brain tissue that represent genomic responses to chronic drug administration. In many cases, the appearance of these products has been linked to specific drug-related behaviors. Additional studies have provided evidence of genetic predisposition to compulsive behaviors and susceptibility to drug addiction. The reader is referred to numerous excellent reviews that document progress made with each of the abovementioned approaches (Altman et al., 1996; Amara and Sonders, 1998; Bozarth, 1987; Carroll and Mattox, 1997; Cloninger and Dinwiddie, 1993; Gatley and Volkow, 1998; George, 1997; Hyman et al., 2001; Johanson and Fischman, 1989; Koob et al., 1998; Lukas, 1997; Lyons et al., 1997; Markou et al., 1993; Nestler, 1992; Pickens et al., 1978; Self and Nestler, 1998; Vrana and Vrana, 1997; White and Kalivas, 1998; Wise and Bozarth, 1987;). Despite the advances that have resulted from the use of these varied experimental applications, they do not address the question of how addictive drugs initiate and consolidate system-wide changes in neuronal functions that underlie drug-related behavior. However, combinations of electrophysiological, neurochemical, and behavioral techniques are uniquely capable of providing information that can fill this considerable gap in our understanding.

1.2 RATIONALE FOR CELLULAR AND CIRCUIT LEVELS OF ANALYSIS OF DRUGS OF ABUSE Implicit in any description of drug addiction is the fact that circuits within the brain must be reorganized so as to generate compulsive behaviors for acquisition and selfadministration of rewarding compounds. Such changes are clearly long-lasting and not readily reversible, as evidenced by the chronic relapsing character of drug addiction. The physical and psychological symptoms that emerge during withdrawal from chronic drug use and the sensitization that is evident with many addictive compounds further underscore the likelihood that drug addiction is a behavioral manifestation of fundamental changes in cellular biological processes. Many studies have in fact begun to identify neuron-specific gene products that are synthesized in response to chronic

2345_frame_MASTER.book Page 6 Wednesday, October 23, 2002 1:03 PM

6

Methods in Drug Abuse Research: Cellular and Circuit Level Analyses

drug treatment and, in addition, are correlated with various behavioral dimensions of the addicted state (Self and Nestler, 1998; White and Kalivas, 1998). On this basis one could argue that molecular and cellular biological studies hold the greatest promise for elucidating the underlying mechanisms of drug abuse and addiction and for identifying genetic factors that contribute to susceptibility to substance abuse. Such reductionist approaches have, in fact, yielded many new data and have provided many new insights regarding the intracellular targets and modes of action of abusepotential substances. Moreover, such studies may also be the most likely to identify potential targets for therapeutic intervention and treatment of drug addiction. On the other hand, this level of inquiry leaves many unanswered questions regarding the molecular and cellular actions of drugs of abuse and their impact on whole-animal behavior. For example, it is well to remember that short- and long-term molecular and cellular responses to addictive compounds are embedded within selected local circuits and neural networks that give rise to specific behavioral responses. Drug-associated behaviors develop over time and are most likely the product of drug influences on sensory, motor, and associational circuits in the brain. Such alterations may result from long-lasting neuroadaptations to drug exposure or may represent more fundamental physiological responses to drug administration by membrane receptors or synaptic release mechanisms. To better understand the physiological bases of these dimensions of drug abuse and provide a meaningful framework for more reductionist approaches studies are needed to determine the specific time course and outcome of drug influences on neurotransmitter levels, activity patterns, local circuit operations, and neural network interactions that are directly responsible for whole-animal responses to drug administration.

1.2.1 STUDIES

OF

CELLS

WITHIN

CIRCUITS

The goal of intracellular and extracellular recording studies in tissue slice and intact, anesthetized preparations is to characterize drug actions on individual neurons under reasonably controlled conditions. In many cases, particularly with in vitro tissueslice experiments, dose-response relationships and receptor mechanisms of drug action can be evaluated with great precision using these approaches. Likewise, extracellular single-unit recording methods have been used in intact, anesthetized animals to monitor and measure the responses of individual neurons to local or systemic drug administration. Such experiments have provided a useful means of assessing potential changes in cell function over a range of drug doses. In addition, in many cases neural circuits have been well characterized in terms of cytoarchitecture and connectivity and include neurons that can be classified according to unique morphological and intrinsic electrophysiological properties (Larkman and Mason, 1990; Mason and Larkman, 1990; McCormick et al., 1985). Individual cells may vary substantially in their responsiveness to systemically administered drugs. Such a view is supported by evidence (see reviews by Llinas, 1988; Nicoll et al., 1990; Schofield et al., 1990) that suggests that within a heterogeneous population of cells there are neuronal subtypes that express different complements of membrane receptors and signal transduction mechanisms and ion channels. With both intracellular and extracellular

2345_frame_MASTER.book Page 7 Wednesday, October 23, 2002 1:03 PM

Overview

7

studies the morphological, electrophysiological, and pharmacological (i.e., drug effects) properties of individual cells can be compared and sorted according to recurrent patterns that emerge. This approach begins to assess the potential range of selective responses of neurons within a complex circuit to acute or chronic drug administration. Such information is crucial for the development of future testable hypotheses concerning the impact of an exogenous agent on ensembles of functionally related cells within awake, behaving animals.

1.2.2 STUDIES

IN

AWAKE ANIMALS

Most if not all theories of addiction share a common assumption that addiction is caused by drug-induced changes in reward-related behavioral processes (Koob and Le Moal, 1997; for reviews of various theories see Wise and Bozarth, 1987; Robinson and Berridge, 1993; Koob and Le Moal, 2001). Such processes are expected to be present only in the awake animal and only within certain behavioral contexts. It is thus possible that drug effects and neural events that are critical to the development of addiction and relapse are observable and thus subject to investigation only under these conditions. Appreciation of this fact is an important impetus for the application of in vivo measurement techniques to awake animals exposed to stimulus and behavioral conditions relevant to drug taking and relapse in humans. The intravenous-drug self-administration model, and variations thereof, is widely viewed as the animal model of human drug addiction and relapse that has the greatest face and predictive validity. Extensive use of the paradigm has shown that the drug self-administration behavior of laboratory animals is consistent with patterns of drug taking in humans, including that of drug-addicted individuals studied in laboratory settings (e.g., Griffiths et al., 1980; Johanson and Balster, 1978; Pickens et al., 1978; Stewart et al., 1984). Pharmacological and neuroimaging studies in humans show that the pharmacology- and neurobiology-mediating drug taking in animals and humans are mediated by homologous, if not the same, neuropharmacological mechanisms (Breiter et al., 1999; Childress et al., 1999; Grant et al., 1996; Volkow and Fowler, 1999). The behavioral paradigm is thus the paradigm of choice to be employed in in vivo investigations of mechanisms that contribute to addiction. Within the last 15 years researchers have developed methodologies that allow for the use of microdialysis and in vivo voltammetry recordings to characterize neurochemical events in animals self-administering drug (e.g., Bradberry et al., 2000; Hurd et al., 1989; Gratton and Wise, 1994; Pettit and Justice, 1989; Wise et al., 1995). In parallel, chronic extracellular recording and EEG procedures have been developed to characterize neurophysiological events during drug-related behavioral states (Berridge, this volume; Peoples et al., 1989, 1999; Chang et al., 1990, 1994; Carelli et al., 1993; Peoples and West, 1996).

1.2.3 DELINEATION OF MECHANISMS THAT CONTRIBUTE TO BEHAVIOR Given the fine temporal resolution of electrophysiological recording studies in behaving animals, this approach can be extremely useful in identifying structures that

2345_frame_MASTER.book Page 8 Wednesday, October 23, 2002 1:03 PM

8

Methods in Drug Abuse Research: Cellular and Circuit Level Analyses

potentially contribute to a behavior. However, the method can additionally be used to explore the mechanisms that mediate the contribution. As noted by Rolls (2000), to understand how the contribution is made it is necessary to understand what information is represented in that region and, moreover, what information is received, how that information is integrated, and how the information is transmitted to target sites. An important aspect of characterizing this information processing is to analyze the responses of single neurons and groups of single neurons, for it is at this level that much of the information processing occurs. These analyses of single-neuron responses are most relevant when they can be made during the behavior of interest under relevant test conditions (Rolls, 2000). Delineating the nature of information encoded and the mechanics of information processing within regions important to drug seeking will ultimately contribute to a blueprint of potential targets for therapeutic interventions.

1.2.4 MECHANISMS THAT MEDIATE DRUG EFFECTS

ON

BEHAVIOR

An effect of drug on a given behavior is mediated by drug-induced changes in the activity of neurons that control that behavior. Acute electrophysiological methods are often used in efforts to delineate those changes in neural activity. Recordings in behaving animals can corroborate and complement these studies. In slice and anesthetized recording preparations, neurons can be stripped of normal afferent input. Anesthesia can additionally influence membrane properties. Both the deafferentation and the direct effects of anesthesia on a recorded neuron can alter the impact of drug on the neuron. The potential for this alteration is demonstrated by previous observations of differences in the results of acute and awake animal recordings (Chapin, Waterhouse, and Woodward, 1981; Deadwyler, 1986; Moxon, 1999; West, 1997). It is also emphasized by the observation that the “mere” differential activation of a neural circuit in association with a change in behavioral state can alter the observed effect of drug on brain (e.g., Hemby et al., 1997; Smith et al., 1980, 1982). Given the importance of the state of a neuron on the response of that neuron to drug, recordings in waking animals can be a useful complement to acute recording studies. Analysis of the comparability, as well as the differences, among the different studies can yield a more complete and accurate picture of the drug effects on brain that mediate a drug-induced change in behavior.

1.3 CELLULAR AND CIRCUIT LEVEL ANALYSIS OF DRUGS OF ABUSE 1.3.1 CURRENT METHODS In this volume, we describe the rationale and methodologies for characterizing the operation of single cells, local circuits, and neural networks before, during, and after acute or chronic administration of drugs of abuse. For each technique advantages as well as limitations are given. For example, single-unit intracellular and extracellular recording techniques in tissue slice and anesthetized preparations (see Waterhouse, Chapter 5) offer the opportunity to obtain detailed information about drug effects on cells in intact local circuits under well-controlled conditions. Knowing the identity of

2345_frame_MASTER.book Page 9 Wednesday, October 23, 2002 1:03 PM

Overview

9

recorded cells further enhances the interpretation of results since drug actions can be evaluated in the context of the role of selected cell types in specific neural circuit operations. Despite these advantages, drug effects in these preparations are examined in the absence of the behavioral or physiological state in which they normally occur. By contrast, neurochemical (Baumann and Rutter, Chapter 3; Salvatore et al., Chapter 4), multichannel, multineuron recording (Devilbiss and Waterhouse, Chapter 8; Janak, Chapter 6; Peoples, Chapter 7), and EEG (Berridge, Chapter 9) techniques can be applied in the awake, behaving animal but have the added complication of introducing effects that are secondary to drug-induced behaviors. For the most part, the experimental methods described here are used for many other applications in neuroscience but clearly have unique advantages for studies of drug abuse and addiction. Moreover, they can be applied in combination (i.e., self-administration, Dworkin and Stairs, Chapter 2, and electrophysiological recording, Peoples, Chapter 7) to yield new opportunities for examining physiological functions under relevant behavioral conditions. Finally, they can be applied in a variety of preparations and animal models to address specific questions about drug effects under different physiological or drug-related conditions, e.g., evaluation of cell and circuit function in naïve vs. chronic drug-treated animals and offspring of drug-addicted mothers.

1.3.2 FUTURE DIRECTIONS In their current form, the techniques and experimental strategies described here will continue to provide new information about the impact of drugs of abuse on individual neurons and neural networks and the behaviors that result from these influences. However, a number of future developments that will enhance their utility can be anticipated. For example, implantable probes that combine recording surfaces for multichannel, multineuron recording and in vivo voltammetry are on the horizon. This technology will provide a window on local tissue levels of endogenous transmitter/modulators and local spike train activity in regions of the brain that are influenced by drug actions and subsequently responsible for mediating drug-related behaviors. Other improvements in multineuron recording electrodes and procedures may facilitate long-term recording of spike train activity from the same neurons across extended time periods such that physiological changes associated with chronic drug administration can be routinely studied. Likewise, adaptation of these in vivo voltammetry, microdialysis and multichannel, multineuron recording procedures for the mouse brain would encourage neurochemical and electrophysiological investigations using a broad spectrum of genetically manipulated animals (Drouin et al., 2002). Such approaches will offer new opportunities to bridge the gap between drug actions at the cellular and molecular level in the brain and the behavioral manifestations of substance abuse.

ACKNOWLEDGMENTS The authors would like to thank Ms. Joy Hudson for help in preparation of the manuscript. This work was supported by grant NIDA 05117 to BDW and DA 535404 to LLP.

2345_frame_MASTER.book Page 10 Wednesday, October 23, 2002 1:03 PM

10

Methods in Drug Abuse Research: Cellular and Circuit Level Analyses

REFERENCES 1. Altman, J., Everitt, B.J., Glautier, S., Markou, A., Nutt, D., Oretti, R., Phillips, G.D., and Robbins, T.W., The biological, social and clinical bases of drug addiction, commentary and debate, Psychopharmacology, 125, 285, 1996. 2. Amara, S.G. and Sonders, M.S., Neurotransmitter transporters and molecular targets for addictive drugs, Drug Alcohol Dependence, 51, 87, 1998. 3. Arroyo, M., Markou, A., Robbins, T.W., and Everitt, B.J., Acquisition, maintenance and reinstatement of intravenous cocaine self-administration under a second-order schedule of reinforcement in rats, effects of conditioned cues and continuous access to cocaine, Psychopharmacology, 140, 331, 1998. 4. Bradberry, C.W., Barrett-Larimore, R.L., Jatlow, P., and Rubino, S.R., Impact of selfadministered cocaine and cocaine cues on extracellular dopamine in mesolimbic and sensorimotor striatum in rhesus monkeys, J. Neurosci., 20, 3874, 2000. 5. Breiter, H.C., Gollub, R.L., Weisskoff, R.M., Kennedy, D.N., Makris, N., Berke, J.D., Goodman, J.M., Kantor, H.L., Gastfriend, D.R., Riorden, J.P., Mathew, R.T., Rosen, B.R., and Hyman, S.E., Acute effects of cocaine on human brain activity and emotion, Neuron., 19, 591, 1997. 6. Bozarth, M.A., Methods of Assessing the Reinforcing Properties of Abused Drugs. Springer-Verlag, New York, NY, 1987. 7. Carroll, M.E. and Mattox, A.J., Drug reinforcement in animals, in Drug Addiction and Its Treatment, Nexus of Neuroscience and Behavior, Johnson, B.A. and Roache, J.D., Eds., Lippincott-Raven, New York, NY, 1997, pp. 3–38. 8. Carelli, R.M., King, V.C., Hampson, R.E., and Deadwyler, S.A., Firing patterns of nucleus accumbens neurons during cocaine self-administration in rats, Brain Res., 626, 14, 1993. 9. Chang, J.-Y., Sawyer, S.F., Lee, R.S., Maddux, B.N., and Woodward, D.J., Activity of neurons in nucleus accumbens during cocaine self-administration in freely moving rats, Neurosci. Abstr., 16, 252, 1990. 10. Chang, J.-Y., Sawyer, S.F., Lee, R.S., and Woodward, D.J., Electrophysiological and pharmacological evidence for the role of the nucleus accumbens in cocaine selfadministration in freely moving rats, J. Neurosci. 14, 1224, 1994. 11. Chapin, J.K., Waterhouse, B.D., and Woodward, D.J., Differences in somatic response properties of single cortical neurons in awake and halothane anesthetized rats, Brain Res. Bull., 6, 63, 1981. 12. Childress, A.R., Mozley, D., McElgin, W., Fitzgerald J., Reivich, M., and O’Brien, C.P., Limbic activation during cue-induced cocaine craving, Am. J. Psychiatry, 156, 11, 1999. 13. Cloninger, C.R. and Dinwiddie, S.H., Genetic risk factors in susceptibility to substance abuse in Biological Basis of Substance Abuse, Korenman, S.G. and Barchas, J.D., Eds., Oxford University Press, New York, NY, 1993, pp. 405–412. 14. Davis, W.M. and Smith, S.G., Conditioned reinforcement as a measure of the rewarding properties of drugs, in Methods of Assessing the Reinforcing Properties of Abused Drugs, Bozarth, M.A., Ed., Springer-Verlag, New York, NY, 1987, pp. 199–210. 15. Deadwyler, S.A., Electrophysiological investigations of drug influences in the behaving animal, in Modern Methods in Pharmacology Volume 3 Electrophysiological Techniques in Pharmacology, Geller, H.M., Ed., Alan R. Liss, New York, NY, 1986, pp. 1–16. 16. DeWit, H. and Stewart, J., Reinstatement of cocaine-reinforced responding in the rat, Psychopharmacology, 75, 134, 1981.

2345_frame_MASTER.book Page 11 Wednesday, October 23, 2002 1:03 PM

Overview

11

17. Di Chiara, G., A motivational learning hypothesis of the role of mesolimbic dopamine in compulsive drug use, J. Psychopharmacol., 12, 54, 1998. 18. Drouin, C., Darracq, L., Tovero, F., Blanc, G., Glowinski, J., Cotecchia, S., and Tassin, J-P., Alpha 1b-adrenergic receptors control locomotor and rewarding effects of psychostimulants and opiates, J. Neurosci,. 22, 2873, 2002. 19. Ehrman, R.N., Robbins, S.J., Childress, A.R., and O’Brien, C.P., Conditioned responses to cocaine-related stimuli in cocaine abuse patients, Psychopharmacology, 107, 523, 1992. 20. Everitt, B.J., Parkinson, J.A., Olmstead, M.C., Arroyo, M., Robledo, P., and Robbins, T.W., Associative processes in addiction and reward: The role of amygdala-ventral striatal subsystems, in Advances from the Ventral Striatum to the Extended Amygdala, McGinty, J.F., Ed., New York Academy of Sciences, New York, NY, 1999, pp. 412–438. 21. Franklin, T.R., Acton, P.D., Maldjian, J.A., Gray, J.D., Croft, J.R., Dackis, C.A., O’Brien, C.P., and Childress, A.R., Decreased gray mattern concentration in th insular, orbitofrontal, cingulate, and temporal cortices of cocaine patients, Biol. Psychiatry, 51, 13, 2002. 22. Gatley, S.J. and Volkow, N.D., Addiction and imaging of the living human brain, Drug Alcohol Dependence, 51, 97, 1998. 23. George, F.R., The behavioral genetics of addiction, in Drug Addiction and Its Treatment, Nexus of Neuroscience and Behavior, Johnson, B.A. and Roache, J.D., Eds., Lippincott-Raven, New York, NY, 1997, pp. 187–204. 24. Goldberg, S.R., Morse, W.H., and Goldberg, D.M., Behavior maintained under a second-order schedule by intramuscular injection of morphine or cocaine in rhesus monkeys, J. Pharmacol. Exp. Ther., 199(1), 278, 1976. 25. Grant, S., London, E.D., Newlin, D.B., Villemagne, V.L., Liu, X., Contoreggi, C., Phillips, R.L., Kimes, A.S., and Margolin, A., Activation of memory circuits during cue-elicited cocaine craving, Proc. Natl. Acad. Sci., 93, 12040, 1996. 26. Gratton, A. and Wise R.A., Drug- and behavior-associated changes in dopaminerelated electrochemical signals during intravenous cocaine self-administration in rats, J. Neurosci., 14, 4130, 1994. 27. Griffiths, R.R., Bigelow, G.E., and Henningfield, J.E., Animal and human drug-taking behavior, in Advances in Substance Abuse Behavioral and Biological Research, Mello, N.K., Ed., JAI Press. Greenwich, CT, 1980, pp. 3–90. 28. Harmer, C.J. and Phillips, G.D., Enhanced appetitive conditioning following repeated pre-treatment with d-amphetamine, Behav. Pharmacol., 9, 299, 1998. 29. Hemby, S.E., Co, C., Koves, T.R., Smith, J.E., and Dworkin, S.E., Differences in extracellular dopamine concentrations in the nucleus accumbens during responsedependent and response-independent cocaine administration in the rat, Psychopharmacology, 133, 7, 1997. 30. Hurd, Y.L., Weiss, F., Koob, G.F., And, N.E., and Ungerstedt, U., Cocaine reinforcement and extracellular dopamine overflow in rat nucleus accumbens, an in vivo microdialysis study, Brain Res., 498(1), 199, 1989. 31. Hoffman, R.S. and Goldfrank, L.R., The impact of drug abuse and addiction on society, Emergency Med. Clin. North Am., 8, 467, 1990. 32. Hyman, S.E. and Malenka, R.C., Addiction and the brain, the neurobiology of compulsion and its persistence, Nat. Rev. Neurosci., 2(10), 695, 2001. 33. Jentsch, J.D. and Taylor, J.R., Impulsivity resulting from frontostriatal dysfunction in drug abuse, implications for the control of behavior by reward-related stimuli, Psychopharmacology, 146, 373, 1999.

2345_frame_MASTER.book Page 12 Wednesday, October 23, 2002 1:03 PM

12

Methods in Drug Abuse Research: Cellular and Circuit Level Analyses 34. Johanson, C.E. and Balster, R.L., A summary of the results of a drug self-administration study using substitution procedures in rhesus monkeys, Bull. Narcotics, 30, 43, 1978. 35. Johanson, C.E. and Fischman, M.W., The pharmacology of cocaine related to its abuse, Pharmacol. Rev., 41, 3, 1989. 36. Koob, G.F. and LeMoal, M., Drug addiction, dysregulation of reward, and allostasis. Neuropsychopharmacology, 24, 97, 2001. 37. Koob, G.F. and Le Moal, M., Drug abuse, hedonic homeostatic dysregulation, Science. 278(5335), 52, 1997. 38. Koob, G.F., Sanna, P.P., and Bloom F.E., Neuroscience of addiction, Neuron, 21, 467, 1998. 39. Larkman, A. and Mason, A., Correlations between morphology and electrophysiology of pyramidal tract neurons in slices of rat visual cortex, J. Neurosci., 10, 1407, 1990. 40. Leshner, A.I., Addiction is a brain disease, and it matters, Science, 278, 45, 1997. 41. Llinas, R., The intrinsic electrophysiological properties of mammalian neurons, insights into central nervous system function, Science, 242, 1654, 1988. 42. London, E.D., Ernst, M., Grant, S., Bonson, K., and Weinstein, A., Orbitofrontal cortex and human drug abuse, functional imaging. Cerebral Cortex, 10, 334, 2000. 43. Lorrain, D.S., Arnold, G.M., and Vezina, P., Previous exposure to amphetamine increases incentive to obtain the drug, long-lasting effects revealed by the progressive ratio schedule, Behav. Brain. Res., 107, 9, 2000. 44. Lukas, S.E. Topographical brain mapping during drug-induced behaviors, in Drug Addiction and Its Treatment, Nexus of Neuroscience and Behavior, Johnson, B.A. and Roache, J.D., Eds., Lippincott-Raven, New York, NY, 1997, pp. 259–276. 45. Lyons, D.J., Letchworth, S.R., Daunais, J.B., and Porrino, L.J., Structural and functional brain imaging, in Drug Addiction and Its Treatment, Nexus of Neuroscience and Behavior, Johnson, B.A. and Roache, J.D., Eds., Lippincott-Raven, New York, NY, 1997, pp. 277–294. 46. Markou, A., Weiss, F., Gold, L.H., Caine, S.B., Schulteis, G., and Koob, G.F., Animal models of drug craving, Psychopharmacology, 112, 163, 1993. 47. Mason, A. and Larkman, A., Correlations between morphology and electrophysiology of pyramidal neurons in slices of rat visual cortex, II. Electrophysiology, J. Neurosci., 10, 1415, 1990. 48. McCormick, D.A., Connors, B.W., Lighthall, J.W., and Prince, D.A., Comparative electrophysiology of pyramidal and sparsely spiny stellate neurons of the neocortex, J. Neurophysiol., 54, 782, 1985. 49. McGinnis, J.M. and Foege, W.H., Mortality and morbidity attributable to use of addictive substances in the United States, Proc. Assoc. Am. Phys. 111, 109, 1999. 50. McLellan, A.T., Lewis, D., O’Brien, C.P., and Kleber, H., Is drug dependence a chronic medical illness: Implications for treatment, insurance and outcome evaluation, JAMA, 284, 1689, 2000. 51. Moxon, K.A., Multichannel electrode design, considerations for different applications, in Methods for Neural Ensemble Recordings, Nicolelis, M.A.L., Ed., CRC Press, Boca Raton, FL, 1999, pp. 25–46. 52. Nestler, E.J., Molecular mechanisms of drug addiction, J. Neurosci., 12, 2439, 1992. 53. Nicoll, R.A., Malenka, R.C., and Kauer, J.A., Functional comparison of neurotransmitter receptor subtypes in mammalian central nervous system, Physiol. Rev. 70, 513, 1990. 54. O’Brien, C.P., Childress, A.R., Ehrman, R., and Robbins, S.J., Conditioning factors in drug abuse, can they explain compulsion? J. Psychopharmacol., 12(1), 15, 1998.

2345_frame_MASTER.book Page 13 Wednesday, October 23, 2002 1:03 PM

Overview

13

55. Panlilio, L.V., Weiss, S.J., and Schindler, C.W., Motivational effects of compounding discriminative stimuli associated with food and cocaine, Psychopharmacology, 136(1), 70, 1998. 56. Peoples, L.L., Wolske, M., Dworkin, S.I., Smith, J.E., Deadwyler, S.A., and West, M.O., A method for recording single unit activity during IV self-administration of drugs in the freely moving rat, Soc. Neurosci. Abstr., 15, 1097, 1989. 57. Peoples, L.L., Bibi, R., and West, M.O. Effects of intravenous self-administered cocaine on single cell activity in the nucleus accumbens of the rat, National Institute on Drug Abuse Research Monograph #141, Harris, L., Ed., U.S. Government Printing Office, Washington, D.C., 1994. 58. Peoples, L.L., Uzwiak, A.J., Gee, F., and West, M.O., Tonic inhibition of single nucleus accumbens neurons in the rat: A predominant but not exclusive firing pattern induced by cocaine self-administration sessions, Neuroscience, 86, 13, 1998. 59. Peoples, L.L. and West, M.O., Phasic firing of single neurons in the rat nucleus accumbens correlated with the timing of intravenous cocaine self-administration, J. Neurosci., 16(10), 3459, 1996. 60. Pettit, H.O. and Justice, J.B., Jr., Dopamine in the nucleus accumbens during cocaine self-administration as studied by in vivo microdialysis, Pharmacol. Biochem. Behav., 34, 899, 1989 61. Pickens, R., Meisch, A., and Thompson, T., Drug self-administration, an analysis of the reinforcing effects of drugs, in Handbook of Psychopharmacology, Vol. 12. Iversen, L.L., Iversen, S.D., and Snyder, S.H., Eds. Plenum, New York, NY, 1978, pp. 1–37. 62. Ranaldi R. and Roberts, D.C., Initiation, maintenance and extinction of cocaine selfadministration with and without conditioned reward, Psychopharmacology, 128(1), 89, 1996. 63. Robbins, T.W., Cador, M., Taylor, J.R., and Everitt, B.J., Limbic-striatal interactions in reward-related processes, Neurosci. Biobehav. Rev., 13, 155, 1989. 64. Robbins, T.W. and Everitt, B.J., Drug addiction, bad habits add up, Nature, 398, 567, 1999. 65. Robinson, T.E. and Berridge, K.C., The neural basis of drug craving, an incentivesensitization theory of addiction, Brain Res. Rev.,18, 247, 1993. 66. Rogers, R.D., Everitt, B.J., Baldacchino, A., Blackshaw, A.J., Swainson, R., Wynne, K., Baker, N.B., Hunter, J., Carthey, T., Booker, E., London, M., Deakin, J.F.W., Sahakian, B.J., and Robbins, T.W., Dissociable deficits in the decision-making cognition of chronic amphetamine abusers, opiate abusers, patients with focal damage to prefrontal cortex, and tryptophan-depleted normal volunteers, evidence for monoaminergic mechanisms, Neuropsychopharmacology, 20, 322, 1998. 67. Rolls, E.T., Neurophysiology and functions of the primate amygdala, and the neural basis of emotion, in The Amygdala, A Functional Analysis, 2nd ed., Aggleton, J.P., Ed., Oxford University Press, New York, NY, 2000, pp. 447–478. 68. Schofield, P.R., Shivers, B.D., and Seeburg, P.H., The role of receptor subtype diversity in the CNS, Trends Neurosci., 13, 8, 1990. 69. Self, D.W. and Nestler, E.J., Relapse to drug-seeking, neural and molecular mechanisms, Drug Alcohol Dependence, 51, 49, 1998. 70. Smith, J.E., Co, C., Freeman, M.E., and Lane, J.D., Brain neurotransmitter turnover correlated with morphine-seeking behavior in rats, Pharmacol. Biochem. Behav., 16(3), 509, 1982.

2345_frame_MASTER.book Page 14 Wednesday, October 23, 2002 1:03 PM

14

Methods in Drug Abuse Research: Cellular and Circuit Level Analyses 71. Smith, J.E., Co, C., Freeman, M.E., Sands, M.P., and Lane, J.D., Neurotransmitter turnover in rat striatum is correlated with morphine self-administration, Nature, 287(5778), 152, 1980. 72. Solomon, R.L. and Corbit, J.D., An opponent-process theory of motivation. I. Temporal dynamics of affect, Psychol. Rev., 81, 119, 1974. 73. Sterling, P. and Elyer, J., Allostasis, a new paradigm to explain arousal pathology, in Handbook of Life Stress, Cognition and Health, Fisher, S. and Reason, J., Eds., John Wiley & Sons, Chichester, 1988, pp. 629–649. 74. Stewart, J., Reinstatmenet of heroin and cocaine self-administration behavior in the rat by intracerebral application of morphine in the ventral tegmental area, Pharmacol. Biochem. Behav., 20, 917, 1984. 75. Stewart, J., Neurobiology of conditioning to drugs of abuse. Ann. N.Y. Acad. Sci. 654, 335, 1992. 76. Stewart, J., deWit, H., and Eikelboom, R., Role of unconditioned and conditioned drug effects in the self-administration of opiates and stimulants, Psychol. Rev. 91, 251, 1984. 77. Taylor, J.R. and Horger, B.A., Enhanced responding for conditioned reward produced by intra-accumbens amphetamine is potentiated after cocaine sensitization, Psychopharmacology, 142, 31, 1999. 78. Taylor, J.R. and Robbins, T.W., 6-hydroxydopamine lesions of the nucleus accumbens, but not of the caudate nucleus, attenuate enhanced responding with rewardrelated stimuli produced by intra-accumbens d-amphetamine, Psychopharmacology, 90, 390, 1986. 79. Tiffany, S.T., A cognitive model of drug urges and drug-use behavior, role of automatic and nonautomatic processes, Psychol. Rev., 97, 147, 1990. 80. Volkow, N.D., Fowler J.S., Wolf, A.P., Hitzemann, R., Dewey, S., Bendriem, B., Alpert, R., and Hoff, A., Changes in brain glucose metabolism in cocaine dependence and withdrawal, Am. J. Psychiatry., 148, 621, 1991 81. Volkow, N.D. and Fowler, J.S. Addiction, a disease of compulsion and drive, involvement of the orbitofrontal cortex, Cerebral Cortex. 10, 318, 2000. 82. Vrana, S.L. and Vrana, K.E., Substance abuse and gene expression, in Drug Addiction and Its Treatment, Nexus of Neuroscience and Behavior, Johnson, B.A. and Roache, J.D., Eds., Lippincott-Raven, New York, NY, 1997, pp. 317–338. 83. Weiss, F., Maldonado-Vlaar, C.S., Parsons, L.H., Kerr, T.M., Smith, D.L., and BenSharhar, O., Control of cocaine-seeking behavior by drug-associated stimuli in rats, effects on recovery of extinguished operant-responding and extracellular dopamine levels in amygdala and nucleus accumbens, Proc. Natl. Acad. Sci. U.S.A. 97, 4321, 2000. 84. West, M.O. Anesthetics eliminate somatosensory-evoked discharges of neurons in the somatotopically organized sensorimotor striatum of the rat, J. Neurosci., 18, 9055, 1998. 85. White, F.J. and Kalivas, P.W., Neuroadaptations involved in amphetamine and cocaine addiction, Drug Alcohol Dependence, 51, 141, 1998. 86. Wise, R.A. Cognitive factors in addiction and nucleus accumbens function, some hints from rodent models, Psychobiology, 27, 300, 1999. 87. Wise, R.A. and Bozarth, M.A., A psychomotor stimulant theory of addiction, Psychol. Rev., 94, 469, 1987. 88. Wise, R.A., Newton, P., Leeb, K., Burnette, B., Pocock, D., and Justice, J.B., Jr., Fluctuations in nucleus accumbens dopamine concentration during intravenous cocaine self-administration in rats, Psychopharmacology, 120, 10, 1995.

2345_frame_MASTER.book Page 15 Wednesday, October 23, 2002 1:03 PM

Overview

15

89. Wyvell, C.L. and Berridge, K.C., Incentive sensitization by previous amphetamine exposure, increased cue-triggered “wanting” for sucrose reward, J. Neurosci., 1(19), 7831, 2001.

2345_frame_MASTER.book Page 16 Wednesday, October 23, 2002 1:03 PM

2345_frame_MASTER.book Page 17 Wednesday, October 23, 2002 1:03 PM

2

Self-Administration of Drugs of Abuse Steven I. Dworkin and Dustin J. Stairs

CONTENTS 2.1 2.2

Introduction ....................................................................................................17 Three Phases of Drug Self-Administration ...................................................18 2.2.1 Early Research in the Area ................................................................19 2.3 Effects of Environmental and Behavioral Variables......................................21 2.3.1 Behavioral and Drug History.............................................................21 2.3.2 Differences between Contingent and Noncontingent Drug Administration....................................................................................25 2.3.3 Factors That Attenuate Acquisition or Maintenance of Drug Self-Administration ............................................................................27 2.3.4 Relapse and Reinstatement ................................................................27 2.4 Methodology ..................................................................................................28 2.4.1 Acquisition of Drug Self-Administration ..........................................28 2.4.2 Maintenance of Drug Self-Administration ........................................29 2.4.3 Fixed-Ratio Schedules .......................................................................29 2.4.4 Progressive-Ratio Schedules ..............................................................31 2.4.5 Second-Order Schedules of Drug Reinforcement .............................33 2.4.6 Subjects ..............................................................................................33 2.5 Apparatus .......................................................................................................33 2.5.1 Operant Chambers..............................................................................33 2.5.2 Housing Chambers.............................................................................34 2.5.3 Swivels ...............................................................................................35 2.5.4 Catheters.............................................................................................35 2.6 Detailed Rat Catheterization Procedure ........................................................38 2.7 Summary and Future Directions....................................................................42 Acknowledgments....................................................................................................42 References................................................................................................................42

2.1 INTRODUCTION It has been over 40 years since the technical aspects of automatic intravenous drug delivery to rodents were described.1 The development of self-administration meth17

2345_frame_MASTER.book Page 18 Wednesday, October 23, 2002 1:03 PM

18

Methods in Drug Abuse Research: Cellular and Circuit Level Analyses

odology provided a tremendous technological advance for investigating drug abuse and resulted in clear demonstrations that the behavioral actions of psychoactive substances could be investigated using the same methods used to evaluate the reinforcing effects of nondrug reinforcers. The intravenous route of administration has several methodological advantages over other routes. The drug is discretely presented through an intravenous catheter so that the concentration of drug delivered per unit time can be more accurately assessed. Further, there are no problems associated with drug spillage, the handling of animals, or other confounding variables such as taste. Intravenous drug delivery is an extremely rapid procedure for delivering a pharmacologically active substance. Consequently, the intravenous route of drug delivery has enabled behavioral pharmacologists to determine the behavioral effects of a wide variety of drugs. More importantly, intravenous delivery of drugs has allowed for the characterization of behaviorally active drugs based on their ability to maintain self-administration. Intravenous self-administration procedures allow for direct assessment of the reinforcing or abuse liability of psychoactive compounds. During the past four decades significant advances in our understanding of substance abuse resulted from utilization of drug self-administration procedures. This fueled important developments in several fields including the experimental analysis of behavior, behavioral pharmacology, and behavioral neuroscience. Since there are a large number of reviews of the literature on rodent self-administration, this chapter will provide only a brief history and overview of the current use of the rodent intravenous drug administration procedure. Methodological and procedural considerations for utilizing intravenous drug delivery systems will then be discussed. From its conception continuing through to the present, the intravenous drug delivery system has been utilized for drug self-administration studies. While methodological advances have occurred in the development of sophisticated technological enhancements in the areas of drug self-administration research, the basic drug delivery system remains relatively unchanged.

2.2 THREE PHASES OF DRUG SELF-ADMINISTRATION There are three important components of drug abuse that can be assessed using rodent self-administration procedures. These three aspects are the acquisition or engendering of self-administration, the continued maintenance of drug intake in animals that are self-administering the drug, and the reinitiation of self-administration following protracted withdrawal or extinction. Evaluations of acquisition and maintenance provide an assessment of abuse liability, whereas studies investigating the reacquisition of extinguished behavior are suggested to provide an assessment of drug relapse.2 While initial studies used rodent intravenous drug delivery to assess behavioral and pharmacologic variables that maintained self-administration, more recent research has extended these investigations to include evaluations of the initiation of drug intake and relapse of drug intake. Thus, recent studies in this area have evaluated behavioral, pharmacologic, and neurobiologic aspects of the acquisition, maintenance, and relapse to self-administration.

2345_frame_MASTER.book Page 19 Wednesday, October 23, 2002 1:03 PM

Self-Administration of Drugs of Abuse

2.2.1 EARLY RESEARCH

IN THE

19

AREA

Initial research in the area of rodent drug self-administration was directed towards identification of pharmacological agents that would maintain self-administration.3–6 This early work demonstrated that the rodent self-administration procedure provided an exquisite model of human drug abuse in that all of the compounds self-administered by rats were abused by humans.7 Although there continue to be a few exceptions (e.g., hallucinogens), this research provided ample evidence that most of the drugs abused by humans would establish and maintain drug taking by rodents. Early studies with opiates (e.g., morphine) used experimenter-delivered injections to make rats physically dependent on the compound before self-administration was established. Now many studies with heroin and other opiates have clearly shown that this establishing procedure is not necessary to support self-administration by opiates or most other pharmacologic classes. After clearly establishing the utility of the rodent self-administration procedure for identifying and investigating human substance abuse, research efforts were directed towards investigating both pharmacologic and environmental determinants of drug self-administration. Some of the pharmacologic variables that were investigated were the drug, drug dose, infusion duration, effects of pharmacologic antagonists, role of tolerance, cross-tolerance and dependence, and various drug histories. Many excellent reviews of the effects of these variables on self-administration of several drug classes have been published.8–11 The data presented in Figure 2.1 depict typical dose–response curves for responding maintained by cocaine under fixedratio (FR) schedules of reinforcement. The lowest dose of cocaine investigated in this study was shown to be a threshold dose for maintaining self-administration. Increasing doses of cocaine resulted in dose-related decreases in the number of injections self-administered during a 4-h session. Furthermore, increasing doses of cocaine resulted in dose-related increases in the time it took to complete the FR 10 schedule requirement as indicated by the increase in the mean interinjection interval and dose-related increases in the total amount of drug that was self-administered during the session. One of the most intriguing and perhaps most studied pharmacologic aspect of drug self-administration is the shape of the dose–response curve. The prototypical dose–effect curve consists of a bimodal or inverted U-shaped function indicating that both low and high doses of a drug maintain comparable rates of response or number of infusions. An example of this type of dose–effect curve is presented in Figure 2.2. This figure illustrates the effects of increasing the dose of heroin on responding maintained by an FR 10 schedule of reinforcement. Increasing doses of heroin from 3 to 50 mg/inf resulted in an increased number of infusions selfadministered during the 4-h session. Further increases in heroin dose from 50 to 300 mg/inf resulted in dose-related decreases in the number of infusions self-administered during the session. The total dosage of self-administered heroin increases to an asymptotic value as the dose per infusion increases. In this case, the bitonic function resulted from a difference in the patterning of drug intake maintained by the lower and higher doses. However, several different explanations for the typical

2345_frame_MASTER.book Page 20 Wednesday, October 23, 2002 1:03 PM

Methods in Drug Abuse Research: Cellular and Circuit Level Analyses

MEAN # INFUSIONS/4 h

20

100 80 60 40 20 0 0.1

0.3 0.5

1

COCAINE (mg/infusion) COCAINE (mg/4 h)

MEAN III (min)

20 16 12 8 4 0

20 16 12 8 4 0

0.1

0.3 0.5

1

COCAINE (mg/infusion)

0.1

0.3 0.5

1

COCAINE (mg/infusion)

FIGURE 2.1 Dose–response curves for cocaine self-administration under an FR 10 schedule of cocaine infusions. The data are means and SD from six rats run during 4-h sessions. The top panel shows the mean number of infusions maintained by the four doses investigated. The bottom left and right panels show the effects of cocaine dose on the mean interinjection interval and total drug intake. Increasing doses of cocaine resulted in dose-related decreases in the number of ratios completed and increases in the mean interinjection interval (time between each injection) and amount of cocaine that was delivered during the session. The large variability for the number of infusions maintained by the lowest dose of cocaine resulted from individual rats self-administering at a high rate on some days and not responding at all on others.

bitonic function have been proposed. This issue is discussed in greater detail later in the chapter. In addition to the considerable influence of pharmacologic variables, early research in the field elucidated the pervasive dominance of behavioral or environmental variables on drug taking. Collectively these studies demonstrated the similarity of drug and nondrug reinforcers in their control over behavior (for reviews see References 4 and 12–15). Thus, abused drugs are not unique in their ability to influence characteristics of behavior. These characteristics include the rate and temporal patterns of drug intake and the persistence of drug taking. It became obvious that basic research on substance abuse could utilize the experimental methods and procedures employed to investigate the effects of nondrug reinforcers. Figure 2.3 shows how the rate and temporal patterns of cocaine intake can be controlled by the schedule of reinforcement used to maintain self-administration. Responding was maintained by the same dose of cocaine under the four schedules shown. Cocaine-maintained responding was directly influenced by schedule of reinforcement and was remarkably similar to patterns previously reported for food and other nondrug reinforcers. A considerable amount of research has supported

2345_frame_MASTER.book Page 21 Wednesday, October 23, 2002 1:03 PM

21

60

6

50

5 TOTAL INTAKE (mg)

NUMBER OF INFUSIONS

Self-Administration of Drugs of Abuse

40 30 20 10

4 3 2 1

0

0 S

3

5

10

30 50

100

300

HEROIN (µg/inj)

3

5

10

30 50

100

300

HEROIN (µg/inj)

FIGURE 2.2 Dose–response curves for heroin self-administration maintained under an FR 10 schedule of heroin presentations. The data are means and SD from six rats allowed to selfadminister heroin during 4-h sessions. The number of infusions maintained by increasing doses of heroin resulted in a bitonic function in that the highest rate of drug intake was maintained by a moderate dose of the drug (left panel). Total drug intake followed a monotonic increasing function with an asymptote at the two largest doses. Low doses of the drug maintained responding only during the first 2 h of the session, whereas larger doses maintained consistent self-administration during the 4 h with longer interreinforcement intervals compared to the lower doses. This resulted in the equivalent number of infusions for both higher and lower doses.

the notion that drug taking is influenced by the same variables that control responding maintained by other environmental events such as food, water, and sex. Although the unique aspects of behavior controlled by abused drugs continued to maintain experimental interest, recent studies using sophisticated brain-imaging techniques have provided confirmational support that drug and nondrug reinforcers have similar influences on brain systems proposed to regulate appetitive behaviors.16,17

2.3 EFFECTS OF ENVIRONMENTAL AND BEHAVIORAL VARIABLES 2.3.1 BEHAVIORAL

AND

DRUG HISTORY

Investigations of intravenous cocaine self-administration represent a major portion of the literature on drugs as reinforcers.13 Some of the earliest reports employed unlimited access to the drug.18,19 Under these conditions, rats and rhesus monkeys exhibited cyclic patterns of responding with periods of high levels of responding followed by periods of little or no drug self-administration. These patterns of drugintake would usually continue until the animals died. With restricted access to the drug, animals will self-administer cocaine during daily sessions for weeks or even months without developing observable adverse health effects.20,21 The data from more recent studies evaluating changes in the self-administration of cocaine during extended session durations has demonstrated a dysregulation in the self-administration of the drug resulting in a considerable increase in drug intake.22

2345_frame_MASTER.book Page 22 Wednesday, October 23, 2002 1:03 PM

22

Methods in Drug Abuse Research: Cellular and Circuit Level Analyses

RESPONDING MAINTAINED BY 0.33 mg INFUSIONS OF COCAINE

500 RESPS

FIXED-RATIO 10

PROGRESSIVE-RATIO 25

VARIABLE-INTERVAL 5 MIN

FIXED-INTERVAL 10 MIN

30 MIN

FIGURE 2.3 Representative cumulative response records depicting responding maintained under four different schedules of cocaine administration. The completion of each schedule requirement resulted in a 5-sec injection of a 0.33-mg dose of cocaine. Performance maintained by an FR 10 schedule is shown in the top panel. An injection was delivered following the completion of each ten responses. The second panel displays the performance maintained under a PR 25 schedule. Under this schedule the ratio value was increased by 25 responses after each infusion. The third panel contains a record of responding maintained by a VI 5min schedule. Under this schedule contingency the first response after a mean interval of 5 min resulted in an injection of the drug. The bottom panel depicts performance maintained under a fixed-interval 10-min schedule. A 30-sec TO was scheduled following the completion of each ratio requirement under the two ratio schedules, and a 10-min TO was presented following each injection delivered under the two interval schedules. Deflections of the bottom pen in the second and fourth panels indicate the duration of the TO. The motor on the cumulative recorder was stopped during the TO programmed under the VI schedule.

The development of intravenous self-administration has brought the investigation of the reinforcing effects of drugs into the realm of procedures used to study other reinforcers. Thus, basic and complex operant schedules of reinforcement23 can be used to determine the conditions under which cocaine would be self-administered (for reviews see References 9 and 21). Moreover, the degree to which the results of nonhuman laboratory self-administration studies have generalized to human drug

2345_frame_MASTER.book Page 23 Wednesday, October 23, 2002 1:03 PM

Self-Administration of Drugs of Abuse

23

use and abuse suggests that many of the factors involved in modulating behavior maintained by nonpharmacologic reinforcers may also be important determinants of drug use.7 Several variables have been shown to alter the behavioral effects of nonpharmacologic as well as pharmacologic agents on schedule-controlled behavior. These variables include the schedule of reinforcement, the type of reinforcer used to maintain behavior, the behavioral and drug history of the organism, and the environmental context in which the behavior is studied.8 The behavioral and drug history of an organism can alter the effects of acutely administered drugs on schedule-controlled behavior.8,24,25 Moreover, historical variables can reverse the behavioral effects of chronically administered environmental events. Behavioral and drug history has also been shown to influence the reinforcing effects of drugs. In many situations, researchers have trained their subjects on a schedule of food presentation or engendered responding using an extremely efficacious drug reinforcer such as cocaine before substituting a potentially less efficacious reinforcing drug. In some cases, these less efficacious drugs may not engender selfadministration without this history. Prior food training appears to be necessary for obtaining nicotine self-administration.26,27 Additional studies have demonstrated that nicotine can serve as a punisher,28 negative reinforcer,29 or positive reinforcer30 depending on the historical variables involved. Caffeine can also maintain selfadministration under conditions similar to those that support nicotine self-administration. Figure 2.4 shows responding maintained by FR schedules in rats that were first trained to lever press using food as the reinforcer. Responding was maintained under an FR 1–3 schedule at rates that were considerably higher than when vehicle was substituted for caffeine. These data demonstrate that even relatively weak reinforcers can maintain self-administration under the appropriate conditions. Cocaine can also have reinforcing or aversive effects depending on the history of the organism.31,32 The rate and temporal pattern of behavior resulting in cocaine administration can be dramatically altered in monkeys when they are exposed to different behavioral histories.33 Another important historical variable is previous exposure of the organism to environmental stress. The involvement of stress and the subsequent activation of the hypothalamo-pituitary-adrenal (HPA) axis in the acquisition and maintenance of intravenous cocaine self-administration has been clearly demonstrated.34–37 A subject’s previous exposure to different pharmacologic agents can alter the rate of acquisition of self-administration or affect whether or not a particular compound will be self-administered. Schenk and colleagues have shown that preexposure to cocaine38 or caffeine39 increases the acquisition of responding maintained by threshold doses of cocaine. The N-methyl-D-aspartate (NMDA) receptor antagonist MK-801, which is not self-administered when substituted for cocaine, will maintain responding in the same monkeys following a history of phencyclidine self-administration. Evaluations of the neurobiologic mechanisms involved in the alterations that historical variables can exert on the behavioral effects of cocaine are limited, although caffeine preexposure was reported to increase the extracellular levels of dopamine obtained in the ventral striatum following a cocaine challenge.39 Perhaps the most frequently utilized variable in drug self-administration studies is the level of food restriction. Minimal levels of food restriction increase locomotor

24

FIGURE 2.4 Cumulative number of caffeine (300 mg/kg/inf) infusions self-administered across individual 15-min segments of a 1-h session. The data in the left panel show the effects of substituting saline for the maintenance dose of caffeine for three consecutive sessions and the reacquisition of caffeine self-administration. The right panel depicts the effects of increasing the FR value from 1 to 3. These data indicate that caffeine is self-administered by rats using parameters similar to those used to maintain nicotine self-administration.

2345_frame_MASTER.book Page 24 Wednesday, October 23, 2002 1:03 PM

Methods in Drug Abuse Research: Cellular and Circuit Level Analyses

2345_frame_MASTER.book Page 25 Wednesday, October 23, 2002 1:03 PM

Self-Administration of Drugs of Abuse

25

activity and can enhance the acquisition of schedule-controlled behavior maintained by both nondrug and drug reinforcers. Moderate levels of food restriction increase the acquisition and maintenance of self-administration of several different classes of abused drugs.40 For most drugs decreasing the level of food restriction does not considerably attenuate drug self-administration once responding is engendered. However, several investigators have reported that moderate food restriction is necessary to maintain nicotine self-administration. Another important but often overlooked behavioral variable is the contingency between responding and the delivery of drug reinforcers. This contingent relationship is of primary importance because it provides the empirical validation that an event is reinforcing. In order for an environmental event to be considered a reinforcer the event must increase the probability or frequency of a behavior that results in its presentation or result in schedule-appropriate patterns of responding. When an environmental event increases the behaviors it follows, the event is considered to be a positive reinforcer. When the event is removed following the occurrence of a behavior and the behavior is increased, the event is termed a negative reinforcer. Thus, the term reinforcer is used to indicate that the event has increased behavioral output. Early attempts to determine whether there are fundamental characteristics of all reinforcers led to the suggestion that perhaps the only common feature of reinforcers was their rate-increasing effect.41 However, the reinforcing effects of a particular environmental event are not transsituational and must be empirically demonstrated to reinforce behavior under any new conditions of interest. Environmental events do not have inherent reinforcing or punishing properties. They result in effects that can be altered. It can be very difficult to predict whether or not environmental events as disparate as food, shock, or a drug will function as a reinforcer without information on the current conditions and the organism’s history. The findings that electric shock could have both punishing and reinforcing effects,42,43 that electrical stimulation of the same brain site could serve as a positive or negative reinforcer,44 and that pharmacologic agents (e.g., cocaine and nicotine) could serve as positive reinforcers, negative reinforcers, or punishers28–32,45,46 clearly demonstrated that environmental events do not have intrinsic reinforcing or punishing properties. Thus, it is not always possible to predict what effects an event will have on behavior. The behavioral effects of environmental events can be altered by a number of conditions that can be empirically examined. These include the ongoing rate of responding, the organism’s behavioral and drug history, and the current environmental context (for reviews see References 8 and 47).

2.3.2 DIFFERENCES BETWEEN CONTINGENT DRUG ADMINISTRATION

AND

NONCONTINGENT

Research on potential differences between response-dependent and response-independent presentation of environmental events played a pioneering role in the experimental analysis of behavior. Some of the earliest studies in this regard investigated differences in the effects of electric shocks that could be postponed vs. those that could not.48,49 Pairs of rats were studied under conditions during which responding by one rat postponed the delivery of electric shocks. Responding by the second rat

2345_frame_MASTER.book Page 26 Wednesday, October 23, 2002 1:03 PM

26

Methods in Drug Abuse Research: Cellular and Circuit Level Analyses

had no scheduled consequences, and this rat received an electric shock whenever the first rat failed to postpone a shock. Thus, the schedule of shock delivery for the second rat was yoked to the schedule obtained by the first. These studies demonstrated that rats that were allowed to postpone electric shock had fewer and less severe ulcers than rats exposed to response-independent shock presentation.48,49 These studies also reported differences in stress reactivity in rats exposed to three different conditions. One group was allowed to postpone shock delivery. The second and third groups were exposed to signaled and unsignalled response-independent shocks, respectively. Both physiological and neurochemical indices of stress were greatest in rats exposed to unsignalled shock. Intermediate effects were observed in rats exposed to signaled shock presentation, and there were fewest stress-related physiologic and neurochemical effects when shock delivery or postponement was dependent on behavior. Response-dependent vs. -independent presentation of electrical brain stimulation50 or intravenous morphine delivery51–53 produces different neurochemical effects. Differences in the effects of electrical brain stimulation have been evaluated using 2-deoxyglucose methods. Neurotransmitter turnover has been used to evaluate the effects of morphine (for reviews of these neurochemical procedures see References 11, 54, and 55). These studies indicate that the response-dependent administration of either electrical brain stimulation or intravenous morphine activates specific CNS sites involved in the reinforcing effects of drugs from several pharmaceutical classes to a greater degree than the response-independent presentation of these events.56 Further, response-dependent administration of morphine results in more severe withdrawal responses than response-independent presentation.57 The results of studies comparing the response-dependent vs. -independent delivery of environmental events suggest that while the response-dependent presentation of brain stimulation or pharmacologic agents is reinforcing, the response-independent presentations of these same events have different behavioral and neurochemical effects. Influences of the contingent relationship between responding and cocaine delivery on the behavioral effects of the drug have also been investigated.58 For these studies, three littermates were housed in individual operant conditioning chambers located within the same sound-attenuating enclosure immediately following catheterization. One rat from each triad was allowed to intravenously selfadminister cocaine hydrochloride (0.33 mg/infusion delivered over 5.5 sec). At the same time, the other two littermates received either a noncontingent infusion of the same dose of cocaine or saline according to the schedule provided by the self-administering rat. One of the most profound aspects of these studies was the extreme lethality associated with response-independent intravenous administration of cocaine. Under these conditions, 38% of the rats receiving response-independent injections died, while none of the rats self-administering the drug died.59 Several studies have reported considerable differences between response-dependent and response-independent administration of pharmacological agents. Humans drink more alcohol and experience decreased aversive effects when allowed to self-administer alcohol than when forced to consume alcohol.60 Phencyclidine results in lethal effects following response-independent administration in monkeys, whereas response-dependent

2345_frame_MASTER.book Page 27 Wednesday, October 23, 2002 1:03 PM

Self-Administration of Drugs of Abuse

27

administration of larger doses does not result in lethal effects (see Reference 4). Finally, response-dependent administration of midazolam has been shown to shift a generalization gradient for the drug to the left, while response-independent administration shifts the gradient to the right.61 Neurochemical and neurobiological changes associated with the response-dependent administration of cocaine have also been compared to those observed during or following response-independent presentations of the drug.62,63 The results of these studies suggest that drug delivery method can influence the behavioral and neuropharmacologic effects of psychoactive drugs.

2.3.3 FACTORS THAT ATTENUATE ACQUISITION OR MAINTENANCE OF DRUG SELF-ADMINISTRATION Studies directed towards identification and discovery of potential pharmacotherapies and those investigating the neurobiology of drug abuse have focused either on disrupting the acquisition of self-administration or attenuating the reinforcing effects of abused drugs. Recent reviews of both areas of research have been published63–66 and will only be briefly mentioned in this chapter. The mesolimbic dopamine system is critically involved in the reinforcing effects of self-administered drugs.56,67 Lesions of this dopaminergic pathway attenuate the acquisition or maintenance of responding, resulting in the administration of opiates and stimulants. Moreover, selective dopaminergic compounds may also attenuate the reinforcing effects of these agents.68–71 Research development of potential pharmacotherapies for substance abuse has been pursued along three different objectives. Drugs that antagonize the reinforcing effects of both opiates and stimulants have been developed and assessed using the self-administration paradigm. Problems with this approach include patient compliance, side effects, and the fact that most of these compounds are competitive antagonists. Thus, increased drug intake can override the effects of the pharmacotherapies. A second approach has been the development of replacement compounds that can substitute for the abused compound. This general strategy has been referred to as “harm reduction” because the ideal substitute should have significantly less deleterious consequences than the compound it replaces. Methadone and buprenorphine have served this role for opiates, while bupropion may attenuate the reinforcing effects of nicotine. Several groups of researchers are developing potential substitutes for cocaine.72–74 A relatively new area of research in medications development is the use of antibodies that attach to a drug and keep it from passing through the blood–brain barrier.75,76 Medication development is an extremely active research area that relies heavily on results from self-administration studies.

2.3.4 RELAPSE

AND

REINSTATEMENT

The most recently refined aspect of the self-administration paradigm is the assessment of behavioral, pharmacologic, and neurobiologic factors involved in reacquisition or reinstatement of drug self-administration.2,77–79 These studies involve engendering and maintaining drug self-administration for a period of time before the behavior is extinguished by substituting saline for the maintaining substance. After a period of extinction the effects of noncontingent infusions of the initial drug or different

2345_frame_MASTER.book Page 28 Wednesday, October 23, 2002 1:03 PM

28

Methods in Drug Abuse Research: Cellular and Circuit Level Analyses

compound on reinstating extinguished lever pressing is assessed. More recent studies have evaluated the possible attenuation of reinstatement by medications and the potential role of stress in the reinstatement of drug self-administration.80–84

2.4 METHODOLOGY Typical drug self-administration procedures involve placing the subject into an operant chamber where a quantitative measure of behavior (i.e., lever pressing) or activity leads to the delivery of a substance. An operant chamber is an apparatus in which an animal is placed and some type of behavior is observed. The behavior recorded in these boxes is sometimes termed free-operant responding. The term free-operant refers to a situation in which the organism has the opportunity to make responses, and these responses operate on the environment.85 Chambers hold a waterspout, a food cup, an intelligence panel (levers, lights, etc.), and other instruments important for the particular study. Specific experimental contingencies may be arranged requiring the animal to respond (e.g., lever press) in order to receive a reinforcer (e.g., drug).

2.4.1 ACQUISITION

OF

DRUG SELF-ADMINISTRATION

There are several ways to train an animal to self-administer a drug. Training by successive approximations involves reinforcing behavior that closely approximates the targeted behavior (i.e., lever pressing). For example, standing near the lever may initially be reinforced, then sniffing the lever, and finally touching the lever. This process will eventually shape lever pressing that results in a drug infusion. A second way to train an animal to press the lever is through successive approximations using food as reinforcement. When the animal’s behavior results in the delivery of food, drug may replace the food. The animal will then receive drug as a consequence of responding. This procedure is used with drugs of unknown or limited reinforcing efficacy (i.e., nicotine).86 Another common way to train an animal to self-administer a drug is to place the animal into the experimental chamber using the drug as the reinforcer. Through operant-level behavior patterns the animal will come into contact with the schedule contingency between lever pressing and drug delivery. A moderate amount of food deprivation that increases activity levels can enhance the acquisition of lever pressing. To aid in each of these procedures a small piece of tape may be placed on the lever to promote more frequent contact with the lever by the animal (rats will typically chew on tape). Acquisition of drug self-administration is typically engendered under a fixedratio 1 (FR 1) schedule of reinforcement. This ratio requires the animal to press the lever one time to receive each injection. This ratio may be increased once drug intake has reached a steady state, which can depend on the drug class. The interinjection interval (III), or the period of time between each injection, is dose dependent. Rats whose responding is maintained by cocaine typically show IIIs in the range of 3 to 6 min when using a training dose of 0.33 mg/infusion. Under an FR schedule the animal is given control over the rate of drug delivery by completing the required number of responses in order to receive an injection.

2345_frame_MASTER.book Page 29 Wednesday, October 23, 2002 1:03 PM

Self-Administration of Drugs of Abuse

2.4.2 MAINTENANCE

OF

29

DRUG SELF-ADMINISTRATION

Once self-administration results in a steady state of intake, this behavior can be maintained using appropriate self-administration procedures. Animals typically selfadminister opiates and stimulants under a variety of scheduling conditions. In addition to FR schedules, self-administration can be maintained using fixed-interval (FI) schedules. Under an FI schedule a response is reinforced only if it occurs after some time interval has passed and responses during this interval have no consequences.87 Cocaine self-administration has been maintained under a variable-interval (VI) schedule of reinforcement.88 Under this schedule the length of time that must pass before another injection of cocaine is delivered varies from reinforcer to reinforcer. Under food-deprived conditions animals will maintain more regular patterns of cocaine self-administration.89 Animals will self-administer opiates and stimulants when each injection is followed by a brief period of time during which additional injections cannot be obtained (i.e., time-out or TO). The TO procedure is typically used to eliminate the potential for drug overdose and to bridge the gap from schedule completion to drug infusion. Thus, responses during the injection have no scheduled consequences. A complex stimulus of a tone and light are typically presented during the TO. Research has shown that opiate and stimulant self-administration can be maintained in rats under relatively short (20 sec to 1 min) TO conditions. Opiates and several stimulants can maintain responding in laboratory animals under a variety of different schedules and different scheduling conditions. It does not seem that one particular condition is necessary for the maintenance of self-administration. This information demonstrates the efficacy of drug as reinforcer and attests to the possible compulsive use of cocaine in a variety of different situations by humans.

2.4.3 FIXED-RATIO SCHEDULES Researchers typically use fixed-ratio (FR) schedules to maintain drug self-administration in animal models.90,91 An FR schedule is one of the least complex schedules to arrange, and it typically is easy to establish behavior under this schedule. This schedule also is frequently used because it gives most of the control over III to the organism. Under the FR schedule a high terminal rate of responding follows an initial pause. An FR schedule requires the animal to make a specified number of responses to receive a drug infusion. For example, an FR 5 would require the animal to make five responses in order to receive each injection, whereas an FR 10 schedule would require the animal to make ten responses. This ratio remains constant throughout the duration of the session, and the animal is given the opportunity to selfadminister a range of doses either within or between experimental sessions. For this range of doses a dose–effect curve or dose–response curve can be generated for the number of injections taken or the number of responses made at each dose level. Cocaine self-administration maintained under an FR schedule typically results in an inverted U-shaped dose–effect curve for the number of injections or responses as a function of dose. As the dose increases, the number of self-administered injections increases to a peak. At higher doses the number of injections decreases. Consequently, both low and high doses maintain the same rate of responding or

2345_frame_MASTER.book Page 30 Wednesday, October 23, 2002 1:03 PM

30

Methods in Drug Abuse Research: Cellular and Circuit Level Analyses

result in the administration of the same number of infusions. Similarly, under an FR schedule as the dose per injection increases, the overall rate of responding increases to a peak and then decreases at higher doses. The total amount of drug taken typically increases as the dose increases.92 One of the first studies evaluating the reinforcing properties of cocaine demonstrated that rhesus monkeys maintained a consistent pattern of drug self-administration.93 This study showed that cocaine injections tended to be equally spaced throughout the session. As the unit dose of cocaine was increased the number of responses during the session decreased. A distinctive post-reinforcement pause (PRP) (i.e., the duration of time after receiving a reinforcer until the first response was made) occurred at each dose level. With smaller doses that maintain responding, the animals showed the shortest PRP. The animals showed longer pauses with more moderate doses. Even greater pauses were seen at the highest doses. As the dose increased, the duration of the PRP also increased. Wilson et al.93 proposed several possible explanations for these results. First, subjects may have adjusted their response rates to maintain an optimum blood level of cocaine (no blood level assays were taken). Second, the drug may have become aversive as the concentration increased. Third, the drug may have physically disrupted lever pressing. This conclusion was drawn from the increased response pause seen after each injection. Wilson et al.93 suggested that more time may have been required for disrupting effects to dissipate when the animal was self-administering higher doses of the drug. In addition to the inverted U-shaped function obtained by Wilson et al.,93 other research has shown similar patterns of cocaine self-administration. Sizemore et al.88 used an FR 10 schedule of cocaine delivery and found that the number of infusions increased to a peak at 0.33 mg/infusion. At a higher dose (0.67 mg/infusion) the number of injections decreased to injection levels similar to those obtained with lower doses (0.17 mg/infusion). The animals did self-administer greater total amounts of cocaine when higher doses were available despite fewer numbers of injections. The inverted U-shaped dose–response curve that is typically reported may occur for several reasons.13,88 Katz13 noted that during typical cocaine self-administration sessions animals administer the drug repeatedly. Because of this, several effects may occur that are independent of the reinforcing effects of the drug. These include potential rate-decreasing effects. Evidence for a rate-decrease effect is demonstrated when cocaine is administered to animals whose behavior is maintained by food. This would remove the “motivational” aspects of cocaine-maintained behavior, leaving the change to be described as direct effects rather than reinforcing effects. When cocaine is administered to animals whose behavior is maintained by food, a similar dose-related increase in pause is seen when either food or cocaine is used to reinforce behavior.18 Katz13 suggested that drug satiety might be a cause of the descending limb of the dose–effect curve. This explanation assumes that after a dose of drug is administered, subsequent intake may not be reinforcing (i.e., just as food may not be reinforcing after a sufficient amount is consumed) and that satiety is dose dependent. It is also possible that larger doses may not be as reinforcing as smaller doses, although this appears unlikely. Rhesus monkeys given a choice between cocaine and

2345_frame_MASTER.book Page 31 Wednesday, October 23, 2002 1:03 PM

Self-Administration of Drugs of Abuse

31

a set amount of food will typically choose cocaine with increasing frequency as the dose of cocaine increases.94,95 In this situation, the higher doses of cocaine are more reinforcing than smaller doses, demonstrating that the dose–effect curve as an indication of reinforcing efficacy may be misleading when using an FR schedule. The downward turn of the dose–effect curve may be due to the direct effects of the drug on responding (i.e., blood level titration, aversive effects, rate-decreasing effects, satiation). Direct effects may modify behavior for reasons that are independent of the reinforcing effects. The reinforcing effects or the indirect effects of a drug refer to the extent to which the drug can establish and maintain self-administration.96 The reinforcing efficacy of a drug is determined from a dynamic interplay between the current environmental conditions (i.e., schedule parameters), the pharmacological effects of the drug, the animal’s drug history, and whether or not other drugs are present.97 It is assumed that there is a positive relation between reinforcing efficacy and abuse liability of self-administered drugs. Thus, evaluations of the reinforcing effects of cocaine not contaminated by direct effects are advantageous as an illustration of the dependence potential of cocaine. The aforementioned data indicate that standard self-administration paradigms may involve both the direct and indirect effects of the drug for maintenance of self-administration. Studies that have used schedule or TO manipulations to increase obtainable interinjection intervals have shown that the descending limb of the curve can be eliminated when using these procedures.88,98 Although standard self-administration procedures can be used to determine the potential abuse liability associated with different drugs of abuse, for reasons noted above they may not adequately compare the reinforcing effects of different doses or different drugs. Thus, there are inherent limitations regarding the assessment and interpretation of relative reinforcing efficacy or abuse liability using standard FR self-administration procedures. Recent studies have demonstrated the utility of a behavioral economic assessment of drug self-administration using a series of FR values to maintain responding.99–101 Although these studies used nonhuman primates, the potential for similar studies using rats to provide a rate-independent assessment of reinforcing efficacy is intriguing.

2.4.4 PROGRESSIVE-RATIO SCHEDULES A progressive-ratio (PR) schedule may circumvent some of the problems noted above. Under a PR schedule the ratio requirement is increased after each successive injection until the animal fails to respond for a specified time, at which point the session is ended. The final ratio completed is the break point and may provide a quantitative measure of reinforcing efficacy.102 First introduced in 1961,103 the schedule has since been used as a measure of reinforcing efficacy of various substances (i.e., food, drug, milk/water solution). Under a PR schedule the ratio may be incremented in several different ways. For example, a PR 5 method of increase or step size would require five responses for the delivery of the first reinforcer, ten for the next, and so on. Hodos103 developed a PR schedule using a step size of two (PR 2) following the delivery of each reinforcer to determine the reinforcing properties of different concentrations of a milk/water

2345_frame_MASTER.book Page 32 Wednesday, October 23, 2002 1:03 PM

32

Methods in Drug Abuse Research: Cellular and Circuit Level Analyses

solution. In studies using the PR schedule with drug reinforcers, the increase in step size may follow an arithmetic series104,105 or an exponential series.106 When step size is increased slowly, break point will also be obtained more slowly. There are a variety of ways to implement step size. For example, several studies have increased the step size after each drug infusion,102,106 while others have increased the ratio requirements after each daily session.104,105 Bedford et al.107 maintained responding for cocaine under a PR schedule by increasing the response requirement by one for the first 32 reinforcers and by two for the next 16 reinforcers. For each subsequent set of 16 reinforcers the response requirement was doubled. Although there is limited research on the effects of step size on drug-maintained responding, Stafford and Branch108 investigated the effect of step size on PR performance maintained by food in pigeons. The study showed that break points remained relatively constant independent of step size and that as step size was increased the number of completed ratios decreased. These results indicate that animals will respond until the break point is reached, independent of the step size. Thus, to investigate PR schedules using cocaine a log increment in step size may be necessary to limit session duration to obtain break point in a maximum of 4 h. The use of log increments in the ratio value results in a rapid increase in the response requirement and shorter sessions due to break point being obtained earlier in the session. In addition to saving time, shorter sessions allow for data to be collected more quickly, typically before catheters lose patency or infection occurs. There are a variety of ways to define break point. In one study, break point was defined as the final ratio completed when the animal made no response for 15 min.107 The study also evaluated a more dynamic definition of break point of no responses made for the duration of three times the longest inter-response time. This criterion allows the animal a sufficient amount of time to respond based on the subject’s longest run time. This method is based on the assumption that the animal should not require three times more time than previously required in order to complete the next ratio. Roberts109 defined break point as the final response ratio completed after 60 min of no injections. It was assumed that animals would not resume responding after 60 min. Break-point criteria must be sensitive to patterns of responding maintained by different classes of self-administered drugs. For example, due to low rates of responding maintained by apomorphine or long post-infusion pauses characteristic of dAmphetamine, 60 min of no reinforcement as the determination of break point may yield false break points.110 This is important because break-point determinations need to be sensitive to the properties and magnitude of the reinforcer and the resulting behavior patterns. This sensitivity is needed so as not to artificially inflate the parameters for obtaining break points for smaller rewards and not artificially decrease break points with larger rewards. Another problem with break point is lack of sensitivity to dose manipulations. For example, Roberts109 used a PR schedule to quantify the reinforcing effects of three doses of cocaine in rats. The drug maintained higher break points at high doses, suggesting that high doses were more reinforcing than smaller doses. Under this schedule, however, doubling the dose from 0.3 to 0.6 mg/infusion produced only a modest increase in break point. Additionally, Bedford and colleagues107 obtained a

2345_frame_MASTER.book Page 33 Wednesday, October 23, 2002 1:03 PM

Self-Administration of Drugs of Abuse

33

bitonic dose–response curve in rhesus monkeys responding under a PR schedule of self-administration that was very similar to dose–effect curves commonly seen when FR schedules are used. Under PR and FR schedules both low and high doses can result in similar dose–effect curves. Under the PR schedule, the modest change in break point with increasing dose levels as well as the downward turn in the dose–response curve may be due to similar effects that result in the inverted Ushaped function under the FR schedule (i.e., direct effects of the drug). Although the PR schedule appears to provide a better measure of reinforcing effects compared to fixed-ratio schedules, the break point may be influenced by potential rate-altering effects of the drug.

2.4.5 SECOND-ORDER SCHEDULES

OF

DRUG REINFORCEMENT

Second-order schedules were initially used to demonstrate that extended sequences of behavior could be maintained in nonhuman primates by drug reinforcers and that stimuli paired with drug delivery could serve as discriminative stimuli and conditioned reinforcers.66,72,111–113 Second-order schedules of drug administration have also been established in rats.26,114–117 A second-order schedule can be regarded as a schedule of reinforcement embedded into a second schedule. Both schedule requirements must be completed before the reinforcer is presented. A typical second-order schedule used for rats is a fixed-interval (FI) 15-min (FR 5) schedule. This schedule specifies that a drug-paired stimulus is presented following the completion of each FR 5 and that the first FR 5 completed after 15 min results in the presentation of the drug and drug-paired stimulus. This schedule can maintain significant rates of responding even though each infusion is separated by at least 15 min. The advantages and difficulties in using second-order schedules of drug administration in primates and rodents been reviewed recently.118 Second-order schedules may provide a rateindependent assessment of reinforcing effects, especially when the rate of response before the first drug injection is used to indicate reinforcing efficacy. Further, the procedure can be used to assess the effects of drug-paired stimuli in the reinforcing effects of self-administered compounds.

2.4.6 SUBJECTS A variety of rodent strains have been used in self-administration studies. Although there are reports in the literature that particular strains such as Fischer F344 rats may be less sensitive to the reinforcing effects of drugs compared to Lewis, Sprague, Darley or Wistar strains, all four strains have been reported to self-administer opiates and stimulants. Although a majority of published reports have used male rats, females may be more sensitive to the reinforcing effects of abused drugs.119–122

2.5 APPARATUS 2.5.1 OPERANT CHAMBERS Self-administration chambers similar to the one shown in Figure 2.5 (see Color Figure 2.5 following page 50) are available from several commercial vendors. These are essen-

2345_frame_MASTER.book Page 34 Wednesday, October 23, 2002 1:03 PM

34

Methods in Drug Abuse Research: Cellular and Circuit Level Analyses

FIGURE 2.5 Standard operant chamber for drug self-administration studies. Details are provided in the text.

tially standard rodent operant chambers that have been modified to accommodate drugdelivery paraphernalia. These modifications include a midline narrow slit in the top of the chamber that ends midway into the chamber, terminating in a small-diameter circular hole. This slit allows for a subject connected to a protective spring leash to be easily inserted into the chamber. The circular hole provides for minimal restraint of the subject in all areas of the operant chamber. A counterbalanced arm is attached to the outside of the back wall of the chamber. A gimbal ring holding a fluid swivel is attached to the distal end of the counter balance. The adjustable weight attached to the other end of the arm is used to provide a slight positive pressure on the leash. The spring leash (Figure 2.6, and Color Figure 2.6 following page 50), attached to the bottom of the swivel, houses and protects the polyvinal catheter. The opposite end of the leash is attached to either a head mount or back plate. Typical chambers are equipped with two fixed or retractable levers and a light located above each lever. Responses on one of the levers result in drug administration. Responses on the other are recorded but have no scheduled consequences. Responses on the inactive lever are sometimes used to assess the effects of self-administered drug on nonspecific activity. The standard self-administration chamber also contains a white house light, a tone generator, and a ventilation fan. A motor- or pistondriven syringe pump (see Figure 2.7, and Color Figure 2.7 following page 50) can be placed either inside or outside the sound-attenuating enclosure that encloses the selfadministration chamber. These pumps can be purchased from several vendors including MED Associates, Razel, Instech, and Fisher Scientific.

2.5.2 HOUSING CHAMBERS Standard polycarbonate rodent housing chambers can be modified to provide a suitable housing environment for catheterized rats (see Figure 2.8, and Color Figure 2.8 following page 50). The modification includes attachment of a counterbalanced liquid swivel (the same as the one used in the operant chamber) and the addition of

2345_frame_MASTER.book Page 35 Wednesday, October 23, 2002 1:03 PM

Self-Administration of Drugs of Abuse

35

FIGURE 2.6 A standard leash assembly that is used to house and protect the catheter from the attachment to the liquid swivel (top) to the attachment to the back plate (bottom). This spring and solid metal leash is commercially available from MED Associates, St. Albans, VT (http://www. med-associates.com).

a Plexiglas top that is split down the center to allow for subject movement. A motordriven syringe pump can be programmed to deliver periodic delivery of heparinized saline to maintain catheter patency while the rats are in their home cage. A modified leur lock system can be inserted in the tubing attached to the top of the swivel to allow for the rapid disconnection of the infusion line in the home cage. A syringe can be inserted into the lines during the transfer process to maintain sterility. The subject can remain attached to the liquid swivel while the gimbal joint is removed from the home cage and inserted into the counterbalance located on the operant chamber. The syringe attached to the line can then be removed and replaced by the tubing connected to the syringe containing the drug of interest.

2.5.3 SWIVELS Liquid swivels similar to those shown in Figure 2.9 can be purchased from commercial vendors including Instech and MED Associates or constructed from 3-cc and 1-cc syringes.123 Both disposable and serviceable swivels can be purchased from these companies. It is extremely important to carefully examine these swivels for potential leaks and malfunction on a daily basis. Excessive amounts of drug residue on the leash assembly are a good indication that the leash needs to be replaced.

2.5.4 CATHETERS Catheters can be purchased from several different sources or constructed using the following materials and method. The materials utilized for making intravenous cath-

2345_frame_MASTER.book Page 36 Wednesday, October 23, 2002 1:03 PM

36

Methods in Drug Abuse Research: Cellular and Circuit Level Analyses

(b)

(a)

(c)

FIGURE 2.7 Three different pumps available for drug infusions. The peristaltic pump shown on the top right can be programmed to deliver a wide range of flow rates from 0.02 to1800 ml/h. The motor-driven syringe pump shown on the bottom right can be purchased from either MED Associates in St. Albans, VT or Razel Scientific Instruments, Inc. in Stamford, CT (http://www.razelscientific.com). The flow rates for this pump can be easily adjusted by changing the size of the syringe or the speed of the motor. Both vendors provide a more expensive version that can be programmed to provide over 60 different infusion rates. A 20cc syringe is typically used in these pumps to infuse from 200 to 500 ml over a 5- to 20-sec period. A high-speed microliter syringe pump from MED Associates in St. Albans, VT is shown on the left. A 100-ml syringe can be used is this pump to deliver a 50-ml dose at a rate of 100 ml/sec. A 500-ml syringe will deliver 250-ml dose at a rate of 500 ml/sec.

eters similar to the one presented in Figure 2.10 (see Color Figure 2.10 following page 50) include: (1) a 36-cm length of narrow Tygon Tubing ID 0.010¢¢, OD 0.030¢¢, and wall 0.010¢¢, (2) a 7.5-cm length of large Tygon Tubing ID 0.020¢¢, OD 0.06¢¢, (3) two shoulders (3-mm pieces of large tubing slit down one side), (4) cyclohexanone, two pieces of 24-cm silk thread, (5) glass rod, (6) boiling water, (7) beaker of ice chips, (8) syringe with a 23-gauge needle, and (9) a 55-cm narrow piece of wire that can pass through the small tubing. The first step in making a catheter is to measure and cut the pieces of tubing. Stretch out one end of the big tubing by inserting a pair of forceps into one end. Run a wire through the big tubing and into one end of the small tubing so that one end of the small tubing is inside the big tubing. Make sure that the wire sticks out of the other end of the big tubing. Place the two shoulders on the small end of the catheter. The first shoulder should be 11/2 in. from the small end of the catheter. The second shoulder should be 7.5 cm from the small end of the catheter. Put a drop of cyclohexanone on the joint where the big and small tubing overlap and one drop in the cracks on each shoulder. Hang the catheters up to dry so that they are not touching and let them dry for at least 30 min. When catheters are dry, remove wires and cut two silk threads for each catheter. Tie the catheters using a square knot starting with the shoulder closest to the big tubing. Both ends of the thread should be about the same length. Run both ends of the thread through the crack in the shoulder.

2345_frame_MASTER.book Page 37 Wednesday, October 23, 2002 1:03 PM

Self-Administration of Drugs of Abuse

37

FIGURE 2.8 Standard housing chamber for three rats that have been prepared with chronic in-dwelling intravenous catheters. These subjects have unrestricted access to water and are given 20 g food each day 30 min following the self-administration session. Other details are provided in the text.

FIGURE 2.9 Liquid swivels used to provide relatively unrestricted movement of the subjects in both the home cage and operant chamber. The swivel depicted in the top pane is commercially available from MED Associates, St. Albans, VT, and the two types of swivels shown in the bottom panel can be purchased from Instech Laboratories, Inc., Plymouth Meeting, PA (http://www.instechlabs.com). (From (top) MED Associates, and (bottom) from Instech Laboratories. With permission.)

2345_frame_MASTER.book Page 38 Wednesday, October 23, 2002 1:03 PM

38

Methods in Drug Abuse Research: Cellular and Circuit Level Analyses

FIGURE 2.10 The contents of a sterilized jugular catheter pack, which includes the jugular catheter, Nalgene back plate, and connection tubing. Instructions for preparing these items can be found in the text.

Tie another square knot at the other end. Try the second shoulder in a similar manner starting at the end closest to the big tubing. Boil water and fill a beaker with ice. Wrap the part of the catheter between the two shoulders around a glass rod so that the two shoulders are even and the small end of the tubing curves inward. Place in boiling water for 45 sec. Immediately place in ice to cool. Once the catheter has cooled, verify that the loop is tight and the small tubing curves inward towards the catheter. If anything is wrong, repeat process. Finally, check the catheter to make sure liquid can flow through. To do this place a syringe with a 23-gauge needle in the large tubing and push distilled water through the catheter

2.6 DETAILED RAT CATHETERIZATION PROCEDURE 1. Anesthetize rat a. Use pentobarbital sodium (50 mg/kg, i.p.) and atrophine methyl nitrate (10 mg/kg, i.p.). b. If rat begins to wake up, it may be necessary to use halothane as an inhalation anesthetic. This requires caution because the animal can quickly overdose (marked by severe respiratory and cardiac depression). If the rat stops breathing, it can be revived by artificial respiration through a tube placed over its nose. c. Once the catheter is in the vein the rat may be kept down by injecting 0.050 ml Brevital. 2. Prepare rat for surgery a. Using animal clippers, clip hair from a 4-cm-wide band from midback to behind the ears. Also, clip hair from the right side of the ventral neck, from mandible to sternum, midline to shoulder. b. Inject penicillin (75,000 units, i.m.) into one hind leg (0.25 ml). c. Apply liquid betadine to back and shave with #22 blade the immediate area of the incision (1.5 cm behind scapulae, 3 cm across back, or approximately 2–3 fingers behind ears).

2345_frame_MASTER.book Page 39 Wednesday, October 23, 2002 1:03 PM

Self-Administration of Drugs of Abuse

3.

4.

5.

6.

d. Apply liquid betadine to neck and shave with #22 blade the immediate area of the incision (from mandible diagonally to point between midline and shoulder). e. Change to sterile surgical gloves and open sterilized catheter and instrument packages. Incision in back a. Reapply betadine to back and make an incision using a #11 (or #15) blade on the scalpel. Open incision with hemostats. (Incision should be just through the skin and down to the muscle). The width of the incision should be the same as the distance between the screws on the backplate. b. Use the hemostats — pointed upward — to separate skin from muscle (separate an area large enough to accommodate backplate). c. Irrigate the incision with bacteriostatic saline and cover with gauze. Turn the rat on its back. Incision in neck a. Reapply betadine to neck and make an incision about 2 cm long from the mandible diagonally to a point between midline and shoulder. Open incision with hemostats and separate skin from muscle using hemostats pointed upward. Locate vein a. Locate a white area of tissue and perhaps part of the vein (a purple line just lateral and posterior to the white area); however, it is not always visible. b. Using a spreading motion with the hemostats bluntly dissect the muscle in a line with the incision until the vein is clearly visible. (Two veins come together to form a Y.) c. The correct vein is the left (lateral) branch (the one to the outside of the rat.) The catheter will be inserted into it and run through the vein represented by the base of the Y to the heart. d. Using both the straight and curved forceps isolate the vein by pulling up and holding the vein with the curved forceps and tearing away the muscle and tissue with the straight forceps. The vein needs to be completely cleaned of extra tissue from the junction of Y up about 1 cm. However, the more the vein is handled, the more it will constrict. e. Irrigate the area and let the vein fall back into place, allowing it to recover its natural shape. Run catheter just under rat’s skin a. Check the catheter with heparinized saline (in a 3-cc syringe with a 23-gauge needle) to make sure it is functional (not blocked or leaking). b. Remove syringe and insert polished end of a 34-cm 23-gauge piece of metal tubing into large end of catheter. (When handling catheter, clean it and anything that comes in contact with it with alcohol pads). c. Dampen trocar with saline. Keeping the point just under the skin, bevel side down (away from skin), push the trocar from the neck to just caudal of the foreleg, then turn the trocar dorsally and push it through the incision in the back.

39

2345_frame_MASTER.book Page 40 Wednesday, October 23, 2002 1:03 PM

40

Methods in Drug Abuse Research: Cellular and Circuit Level Analyses

d. Run the metal tubing and catheter through the trocar, then push the trocar through the back and remove the metal tubing, leaving the catheter in place under the skin. (Pull catheter through until the only part of the catheter extending from the neck incision is the two shoulders with black silk attached.) e. Adjust the catheter so that the loop faces laterally and slightly downward. f. Estimate distance between vein and heart and trim small tubing of the catheter accordingly (should be approximately 2.5–2.7 cm). An unbeveled tip is preferred. g. Put 3-cc syringe with heparinized saline back on large end of catheter and flush with saline until no air remains in catheter. 7. Getting catheter in vein a. Fill the 25-gauge and 23-gauge needles with heparinized saline. b. Dampen glass rod with saline and, using forceps to lift vein, place glass rod under vein. c. Once again use forceps to make sure vein is completely clear of all tissue. The better the vein is cleaned, the easier it is to get the catheter inserted. d. Put saline on the vein to let it return to size. e. Approximately 0.5 cm above junction of Y insert 25-gauge needle with bevel down and tip toward heart. Push the needle up and down in the vein to make sure a hold has been made. Be careful not to push needle out the back of the vein! f. Remove the 25-gauge needle and insert the 23-gauge needle into the same hole. If the needle does not push up and down easily, that means the needle is just between the tissue and the vein and not inside the vein. It needs to be cleaned off again. g. With the 23-gauge needle still in the vein, grasp the small tubing 2 mm from the tip of the catheter with the curved forceps. Pull the needle towards you and insert the tubing into the vein behind the bevel of the needle. Pushing saline through the needle with your forefinger will keep the vein large. h. Remove the needle and thread the catheter into the vein until the first joint is reached. i. Carefully remove glass rod. One should be able to infuse saline and withdraw blood freely. Until the vein is anchored, watch to make sure it does not push out of the vein. If the column of blood in catheter pulsates, catheter tip is in right atrium of heart; withdraw catheter until pulsations stop. 8. Anchoring catheter and sewing up neck a. Anchor the catheter by suturing the first two silk ties into the muscle below the vein (this will be a dark area deeper down). Be careful not to suture any veins!!! b. Thread the ties through two different holes in a piece of .5 ¥ .5 cm square Teflon mesh. Tie one knot (not too tight in order to prevent crimping catheter), then secure tightly with a second knot.

2345_frame_MASTER.book Page 41 Wednesday, October 23, 2002 1:03 PM

Self-Administration of Drugs of Abuse

c. Check to see if blood can still be withdrawn. Cut off excess silk. d. Push the loop of catheter beneath the layer of muscle that was torn apart. If catheter loop does not stay below muscle on its own, you will have to stuff it as you sew up the muscle. e. Using 4–0 chromic gut, suture muscle that had been bluntly dissected back together with a running stitch, enveloping the loop. Be careful not to suture catheter. f. Secure silk ties of second joint to this muscle in the same manner as before. g. Check catheter again for infusion of saline and withdrawal of blood. h. Sew up skin with 4–0 chromic gut, tying off each stitch. i. Clean area and apply Neosporin to sutures. Turn rat over. 9. Backplate and sewing up back a. Insert straight forceps into center hold of backplate to stretch it out some. b. Remove syringe from catheter; quickly thread catheter through the large hole of the backplate, and reinsert syringe. c. Fold backplate in half and insert just under skin. Using hemostats point side up between skin and top of backplate and stretch and loosen skin. d. Flatten skin out and sew up with 4–0 chromic gut. Put a stitch on either side of the catheter as close as possible, a stitch to the inside of each screw, and a third one between the two stitched, if possible. e. Clean area and apply Neosporin to sutures. 10. Putting the leash on the rat a. Remove top nuts of backplate. b. Cut a piece of plastic tube (approximately 1 to 1.5 cm) to go in center hole of bottom of leash. c. Remove syringe and insert polished end of metal tubing into large end of catheter. (Clean with alcohol preps first.) d. Run the metal tubing and catheter up through the plastic tube at the base of the leash all the way up until the top of the catheter comes out the top of the leash. e. Remove metal tubing and pull end of catheter through the side hole at the top of the leash. Reinsert syringe. f. Using hemostats to hold screw below remaining nuts of backplate, put leash on the screws and replace top nuts and tighten down. g. Put rat in cage and screw leash into swivel; connect catheter to swivel. 11. Postoperative care of rat a. Postoperative analgesics can be used as directed by your institution veterinarian. b. Tetracycline or other antibiotics can be used as prophylaxis against infections. c. Do not feed until the next day. d. If rat is still being used after approximately 6 weeks, check back to see if it healed and to make sure that there are no holes in the skin. If so, cut away dead skin and sew skin back up.

41

2345_frame_MASTER.book Page 42 Wednesday, October 23, 2002 1:03 PM

42

Methods in Drug Abuse Research: Cellular and Circuit Level Analyses

e. Flush catheter once a week with 0.05 ml heparinized saline to make sure it is not blocked. f. Approximately once every 10 to 14 d, check catheter by infusing 0.05 ml Brevital immediately followed by 0.05 ml heparinized saline. Rat’s head should drop slightly.

2.7 SUMMARY AND FUTURE DIRECTIONS Forty years have passed since the procedures to establish drug self-administration in rodents were first published. Since that time there has been an almost exponential increase in the number of publications concerning some aspect of rodent drug selfadministration and a similar escalation in the number of laboratories utilizing selfadministration methodologies. While in many ways there has been an increase in the sophistication of the methodologies that have been coupled with rodent selfadministration procedures, the answers to important fundamental questions remained illusive. Rodent self-administration procedures have been extended into very diverse areas of research including behavioral genetics, molecular biology, and behavioral economics; however, the definition of reinforcing efficacy is still debated at conferences and in the literature. It is likely that rodent drug self-administration procedures will continue to be utilized for assessing the potential abuse liability of novel compounds and in pursuing the discovery of potential pharmacotherapeutic agents for drug-abuse disorders. Moreover, rodent self-administration procedures have been critical to our understanding of the neurobiology of reinforcement and will continue to be utilized as more sophisticated molecular approaches are pursued. It is also likely that recent refinements in the experimental analysis of behavior, including behavioral economics, delay discounting, and models of impulsivity, will be incorporated into novel procedures to elucidate behavioral mechanisms involved in engendering, maintaining, and treating compulsive drug abuse.

ACKNOWLEDGMENTS Preparation of this chapter was partially supported by a gift from the R.J. Reynolds Tobacco Co. and by grant #P01 DA11470 from the U.S. Public Health Service.

REFERENCES 1. Weeks, J.R., Experimental morphine addiction: method for automatic intravenous injections in unrestrained rats, Science, 138, 143, 1962. 2. Stewart, J., Reinstatement of heroin and cocaine self-administration behavior in the rat by intracerebral application of morphine in the ventral tegmental area, Pharmacol. Biochem. Behav., 20(6), 917, 1984. 3. Griffiths, R., Brady, J.V., and Bradford, L.D., Predicting the abuse liability of drugs with animal drug self-administration procedures: psychomotor stimulants and hallucinogens, in Advances in Behavioral Pharmacology, Academic Press, 1979, pp. 163–208.

2345_frame_MASTER.book Page 43 Wednesday, October 23, 2002 1:03 PM

Self-Administration of Drugs of Abuse

43

4. Johanson, C.E. and Schuster, C.R., A comparison of the behavioral effects of l- and dl-cathinone and d- amphetamine, J. Pharmacol. Exp. Ther., 219 (2), 355, 1981. 5. Thompson T. and Schuster, C.R., Behavioral Pharmacology, Prentice-Hall, Englewood Cliffs, NJ, 1968. 6. Weeks, J.R., Environmental influences affecting the voluntary intake of drugs: an overview, Fed. Proc., 34 (9), 1755, 1975. 7. Griffiths, R.R., Bigelow, G.E., and Henningfeild, J.E., Similarities in animal and human drug-taking behavior, in Advances in Substance Abuse, Mello, N.K., Ed., JAI Press, Greenwich, CT, 1980, pp. 1–90. 8. Barrett, J.E., Nonpharmacological factors determining the behavioral effects of drugs, in Psychopharmacology: The Third Generation of Progress, Meltzer, H.Y., Ed., Raven Press, New York, NY, 1987, pp. 1493–1501. 9. Johanson, C.-E., Fischman, M.W., The pharmacology of cocaine related to its abuse, Pharmacol. Rev., 41 (1), 3, 1989. 10. Hemby, S.E., Johnson, B.A., and Dworkin, S.I., Neurobiology of drug reinforcement, in Drug Addiction and Its Treatment: Nexus of Neuroscience and Behavior, Bankole, A., Johnson, B.A., and Roach, J.D., Eds., Lippincott-Raven, New York, NY, 1997. 11. Dworkin, S.I. and Smith, J.E., Neurobiological aspects of drug-seeking behaviors, in Neurobehavioral Pharmacology Advances in Behavioral Pharmacology, Dews, P. B, Thompson, T., and Barrett J.E., Eds., Lawrence Erlbaum Assoc., Mahwah, New Jersey, 1987, pp. 1–43. 12. Johanson, C.E. and Schuster, C.R., The effects of pharmacological and environmental variables on drug choice in rhesus monkeys, Psychopharmacol. Bull., 13 (1), 39, 1977. 13. Katz, J.L., Drugs as reinforcers; pharmacological and behavioral factors, in The Neuropharmacological Basis of Reward, Leibman, J.M. and Cooper, S.J., Eds., Clarendon Press, Oxford, 1989, pp. 164–212. 14. Kelleher, R.T. and Goldberg, S.R., Control of drug-taking behavior by schedules of reinforcement, Pharmacol. Rev., 27(3), 291, 1975. 15. Morse, W.H., The control of behavior by consequent drug injections, Pharmacol. Rev., 27(3), 301, 1975. 16. Childress, A.R., Mozley, P.D., McElgin, W., Fitzgerald, J., Reivich, M., and O’Brien, C.P., Limbic activation during cue-induced cocaine craving, Am. J. Psychiatry, 156(1), 11, 1999. 17. Garavan, H., Pankiewicz, J., Bloom, A., Cho, J.K., Sperry, L., Ross, T.J., Salmeron, B.J., Risinger, R., Kelley, D., and Stein, E.A., Cue-induced cocaine craving: neuroanatomical specificity for drug users and drug stimuli, Am. J. Psychiatry, 157(11), 1789, 2000. 18. Pickens, R. and Thompson, T., Cocaine-reinforced behavior in rats: effects of reinforcement magnitude and fixed-ratio size, J. Pharmacol. Exp. Ther., 161(1), 122, 1968. 19. Deneau, G., Yanagita, T., and Seevers, M.H., Self-administration of psychoactive substances by the monkey, Psychopharmacologia, 16(1), 30 1969. 20. Johanson, C.E., Effects of intravenous cocaine, diethylpropion, d-amphetamine and perphenazine on responding maintained by food delivery and shock avoidance in rhesus monkeys, J. Pharmacol. Exp. Ther., 204(1), 118, 1978. 21. Young, A.M. and Herling, S., Drugs as reinforcers: studies in laboratory animals, in Behavioral Analysis of Drug Dependence, Goldberg, S.R. and Stolerman, I.P., Eds., Academic Press, London, 1986, pp. 9–67. 22. Ahmed, S.H. and Koob, G.F., Long-lasting increase in the set point for cocaine selfadministration after escalation in rats, Psychopharmacology (Berl), 146(3), 303, 1999.

2345_frame_MASTER.book Page 44 Wednesday, October 23, 2002 1:03 PM

44

Methods in Drug Abuse Research: Cellular and Circuit Level Analyses 23. Fester, C.B. and Skinner, B.F., Schedules of Reinforcement, Prentice-Hall, Englewood Cliffs, NJ, 1957. 24. Barrett, J.E. and Witkin, J.M., The role of behavioral and pharmacological history in determining the effects of abused drugs, in Behavioral Analysis of Drug Dependence, Goldberg, S.R. and Stolerman, I.P., Eds., Academic Press, Orlando, 1986, pp. 195–224. 25. Barrett, J.E., Glowa, J.R., and Nader, M.A., Behavioral and pharmacological history as determinants of tolerance- and sensitization-like phenomena in drug action, in Psychoactive Drugs: Tolerance and Sensitization, Goudie, A.J. and Emmett-Oglesby, M.W., Eds., Humana Press, Clifton, NJ, 1989, pp. 181–219. 26. Corrigall, W.A. and Coen, K.M., Nicotine maintains robust self-administration in rats on a limited-access schedule, Psychopharmacology, 99(4), 473, 1989. 27. Donny, E.C., Caggiula, A.R., Knopf, S., and Brown, C., Nicotine self-administration in rats, Psychopharmacology (Berl.), 122(4), 390, 1995. 28. Goldberg, S.R. and Spealman, R.D., Suppression of behavior by intravenous injections of nicotine or by electric shocks in squirrel monkeys: effects of chlordiazepoxide and mecamylamine, J. Pharmacol. Exp. Ther., 224(2), 334, 1983. 29. Spealman, R.D., Maintenance of behavior by postponement of scheduled injections of nicotine in squirrel monkeys, J. Pharmacol. Exp. Ther., 227(1), 154, 1983. 30. Goldberg, S.R., Spealman, R.D., and Goldberg, D.M., Persistent behavior at high rates maintained by intravenous self-administration of nicotine, Science, 214(4520), 5735, 1981. 31. Ettenberg, A. and Geist, T.D., Animal model for investigating the anxiogenic effects of self- administered cocaine, Psychopharmacology, 103(4), 455, 1991. 32. Spealman, R.D., Behavior maintained by termination of a schedule of self-administered cocaine, Science, 204(4398), 1231, 1979. 33. Nader, M.A. and Bowen, C.A., Effects of different food-reinforcement histories on cocaine self-administration by rhesus monkeys, Psychopharmacology (Berl.), 118(3), 287, 1995. 34. Goeders, N.E., The HPA axis and cocaine reinforcement, Psychoneuroendocrinology, 27(1–2), 13, 2002. 35. Goeders, N.E., A neuroendocrine role in cocaine reinforcement, Psychoneuroendocrinology, 22(4), 237, 1997. 36. Goeders, N.E. and Guerin, G.F., Non-contingent electric footshock facilities the acquisition of intravenous cocaine self-administration in rats, Psychopharmacology (Berl.), 114(1), 63, 1994. 37. Miczek, K.A. and Mutschler, N.H., Activational effects of social stress on IV cocaine self-administration in rats, Psychopharmacology (Berl.), 128(3), 256, 1996. 38. Horger, B.A., Shelton, K., and Schenk, S., Preexposure sensitizes rats to the rewarding effects of cocaine, Pharmacol. Biochem. Behav., 37(4), 707, 1990. 39. Horger, B.A., Wellman, P.J., Morien, A., Davies, B.T., and Schenk, S., Caffeine exposure sensitizes rats to the reinforcing effects of cocaine, Neuroreport, 2(1), 53, 1991. 40. Carroll, M.E. and Meisch, R.A., Increased drug-reinforced behavior due to food deprivation, in Advances in Behavioral Pharmacology, Thompson, T., Dews, P.B., and Barrett J.E., Eds., Academic Press, New York, NY, 1984, pp. 47–88. 41. Herrnstein, R.J. and Hineline, P.N., Negative reinforcement as shock-frequency reduction, J. Exp. Anal. Behav., 9(4), 421, 1966. 42. Kelleher, R.T. and Morse, W.H., Determinants of the specificity of behavioral effects of drugs, Ergeb. Physiol., 60, 1, 1968.

2345_frame_MASTER.book Page 45 Wednesday, October 23, 2002 1:03 PM

Self-Administration of Drugs of Abuse

45

43. McKearney, J.W., Maintenance of responding under a fixed-interval schedule of electric shock-presentation, Science, 160(833), 1249, 1968. 44. Steiner, S.S., Beer, B., and Shaffer, M.M., Escape from self-produced rates of brain stimulation, Science, 163(862), 90, 1969. 45. Goldberg, S.R. and Spealman, R.D., Maintenance and suppression of behavior by intravenous nicotine injections in squirrel monkeys, Fed. Proc., 41(2), 216, 1982. 46. Spealman, R.D. and Goldberg, S.R., Maintenance of schedule-controlled behavior by intravenous injections of nicotine in squirrel monkeys, J. Pharmacol. Exp. Ther., 223(2), 402, 1982. 47. Morse, W.H. and Kelleher, R.T., Determinants of reinforcement and punishment, in Handbook of Operant Behavior, Honig, W. K. and Staddon, J.E.R., Eds., PrenticeHall, Englewood Cliffs, NJ, 1977, pp. 174–200. 48. Weiss, J.M., Somatic effects of predictable and unpredictable shock, Psychosom. Med., 32(4), 397, 1970. 49. Weiss, J.M., Influence of psychological variables on stress-induced pathology, Ciba Found. Symp., 8, 253, 1972. 50. Porrino, L.J., Esposito, R.U., Seeger, T.F., Crane, A.M., Pert, A., and Sokoloff, L., Metabolic mapping of the brain during rewarding self-stimulation, Science, 224(4646), 306, 1984. 51. Smith, J.E., Co, C., Freeman, M.E., and Lane, J. D., Brain neurotransmitter turnover correlated with morphine-seeking behavior of rats, Pharmacol. Biochem. Behav., 16(3), 509, 1982. 52. Smith, J.E., Co, C., and Lane, J.D., Limbic muscarinic cholinergic and benzodiazepine receptor changes with chronic intravenous morphine and self-administration, Pharmacol. Biochem. Behav., 20(3), 443, 1984. 53. Smith, J.E., Co, C., and Lane, J.D., Limbic acetylcholine turnover rates correlated with rat morphine-seeking behaviors, Pharmacol. Biochem. Behav., 20(3), 429, 1984. 54. Dworkin, S.I. and Smith, J.E., Assessment of neurochemical correlates of operant behavior, in Neuromethods: Psychopharmacology I, Boulton, A.A., Bike, G.G., and Greenshaw, A.J., Eds., Humana Press, Totowa, NJ, 1989, pp. 741–785. 55. Barrett, J.E. and Hoffmann, S.M., Neurochemical changes correlated with behavior maintained under fixed-interval and fixed-ratio schedules of reinforcement, J. Exp. Anal. Behav., 56(2), 39, 1991. 56. Koob, G.F. and Bloom, F.E., Cellular and molecular mechanisms of drug dependence, Science, 242(4879), 715, 1988. 57. Siegel, S., State dependent learning and morphine tolerance, Behav. Neurosci., 102(2), 228, 1988. 58. Dworkin, S.I. and Pitts, R.C., Use of rodent self-administration models to develop pharmacotherapies for cocaine abuse, NIDA Res. Monogr., 145, 88, 1994. 59. Dworkin, S.I., Mirkis, S., and Smith, J.E., Response-dependent versus responseindependent presentation of cocaine: differences in the lethal effects of the drug, Psychopharmacology (Berl.), 117(3), 262, 1995. 60. Mello, N.K. and Mendelson, J.H., Experimentally induced intoxication in alcoholics: a comparison between programed and spontaneous drinking, J. Pharmacol. Exp. Ther., 173(1), 101, 1970. 61. Ator, N.A. and Griffiths, R.R., Differential sensitivity to midazolam discriminativestimulus effects following self-administered versus response-independent midazolam, Psychopharmacology, 110(1–2), 1, 1993.

2345_frame_MASTER.book Page 46 Wednesday, October 23, 2002 1:03 PM

46

Methods in Drug Abuse Research: Cellular and Circuit Level Analyses 62. Dworkin, S.I., Co, C., and Smith, J.E., Rat brain neurotransmitter turnover rates altered during withdrawal from chronic cocaine administration, Brain Res., 682(1–2), 116, 1995. 63. Hemby, S.E., Co, C., Koves, T.R., Smith, J.E., and Dworkin, S.I., Differences in extracellular dopamine concentrations in the nucleus accumbens during responsedependent and response-independent cocaine administration in the rat, Psychopharmacology (Berl.), 133(1), 7, 1997. 64. Mello, N.K. and Negus, S.S., Preclinical evaluation of pharmacotherapies for treatment of cocaine and opioid abuse using drug self-administration procedures, Neuropsychopharmacology, 14(6), 375, 1996. 65. Bardo, M.T., Neuropharmacological mechanisms of drug reward: beyond dopamine in the nucleus accumbens, Crit. Rev. Neurobiol., 12(1–2), 37, 1998. 66. Spealman, R.D. and Goldberg S.R., Drug self-administration by laboratory animals: control by schedules of reinforcement, Annu. Rev. Pharmacol. Toxicol., 18, 313, 1978. 67. Wise, R.A., Neurobiology of addiction, Curr. Opin. Neurobiol., 6(2), 243, 1996. 68. Caine, S.B. and Koob, G.F., Effects of dopamine D-1 and D-2 antagonists on cocaine self-administration under different schedules of reinforcement in the rat, J. Pharmacol. Exp. Ther., 270(1), 209, 1994. 69. Caine, S.B., Koob, G.F., Parsons, L.H., Everitt, B.J., Schwartz, J. C., and Sokoloff, P., D3 receptor test in vitro predicts decreased cocaine self-administration in rats, Neuroreport, 8(9–10), 2373, 1997. 70. Self, D.W., Barnhart, W.J., Lehman, D.A., and Nestler, E.J., Opposite modulation of cocaine-seeking behavior by D1- and D2-like dopamine receptor agonists, Science, 271(5255), 1586, 1996. 71. Self, D.W., Karanian, D.A., and Spencer, J.J., Effects of the novel D1 dopamine receptor agonist ABT-431 on cocaine self-administration and reinstatement, Ann. N.Y. Acad. Sci., 909, 133, 2000. 72. Bergman, J., Madras, B.K., Johnson, S.E., and Spealman, R.D., Effects of cocaine and related drugs in nonhuman primates. III. Self-administration by squirrel monkeys, J. Pharmacol. Exp. Ther., 251(1), 150, 1989. 73. Dworkin, S.I., Lambert, P., Sizemore, G.M., Carroll, F.I., and Kuhar, M.J., RTI-113 administration reduces cocaine self-administration at high occupancy of dopamine transporter, Synapse, 30(1), 49, 1998. 74. Katz, J. L., Agoston, G.E., Alling, K.L., Kline, R.H., Forster, M.J., Woolverton, W.L., Kopajtic, T.A., and Newman, A.H., Dopamine transporter binding without cocaine-like behavioral effects: synthesis and evaluation of benztropine analogs alone and in combination with cocaine in rodents, Psychopharmacology (Berl.), 154(4), 362, 2001. 75. Self, D.W., Drug addiction. Cocaine abuse takes a shot, Nature, 378(6558), 666, 1995. 76. Kantak, K.M., Collins, S.L., Bond, J., and Fox, B.S., Time course of changes in cocaine self-administration behavior in rats during immunization with the cocaine vaccine IPC-1010, Psychopharmacology (Berl.), 153(3), 334, 2001. 77. Schenk, S., Worley, C.M., McNamara, C., and Valadez, A., Acute and repeated exposure to caffeine: effects on reinstatement of extinguished cocaine-taking behavior in rats, Psychopharmacology (Berl.), 126(1), 17, 1996. 78. Comer, S.D., Lac, S.T., Wyvell, C.L., Curtis, L.K., and Carroll, M.E., Food deprivation affects extinction and reinstatement of responding in rats, Psychopharmacology (Berl.), 121(2), 150, 1995.

2345_frame_MASTER.book Page 47 Wednesday, October 23, 2002 1:03 PM

Self-Administration of Drugs of Abuse

47

79. Alleweireldt, A.T., Weber, S.M., Kirschner, K.F., Bullock, B.L., and Neisewander, J. L., Blockade or stimulation of D1 dopamine receptors attenuates cue reinstatement of extinguished cocaine-seeking behavior in rats, Psychopharmacology (Berl.), 159(3), 284, 2002. 80. Erb, S., Shaham, Y., and Stewart, J., Stress reinstates cocaine-seeking behavior after prolonged extinction and a drug-free period, Psychopharmacology (Berl.), 128(4), 408, 1996. 81. Shaham, Y. and Stewart, J., Stress reinstates heroin-seeking in drug-free animals: an effect mimicking heroin, not withdrawal, Psychopharmacology (Berl.), 119(3), 334, 1995. 82. Shaham, Y. and Stewart, J., Effects of restraint stress and intra-ventral tegmental area injections of morphine and methyl naltrexone on the discriminative stimulus effects of heroin in the rat, Pharmacol. Biochem. Behav., 51(2–3), 491, 1995. 83. Shaham, Y., Kelsey, J.E., and Stewart, J., Temporal factors in the effect of restraint stress on morphine-induced behavioral sensitization in the rat, Psychopharmacology (Berl), 117(1), 102, 1995. 84. Mantsch, J.R. and Goeders, N.E., Ketoconazole blocks the stress-induced reinstatement of cocaine-seeking behavior in rats: relationship to the discriminative stimulus effects of cocaine, Psychopharmacology (Berl.), 142(4), 399, 1999. 85. Ferster, C.B., The use of the free operant in the analysis of behavior, Psychol. Bull., 50, 263, 1953. 86. Chiamulera, C., Borgo, C., Falchetto, S., Valerio, E., and Tessari, M., Nicotine reinstatement of nicotine self-administration after long-term extinction, Psychopharmacology (Berl.), 127(2), 102, 1996. 87. Goldberg, S.T. and Kelleher, R.T., Behavior controlled by scheduled injections of cocaine in squirrel and rhesus monkeys, J. Exp. Anal. Behav., 25(1), 93, 1976. 88. Sizemore, G.M., Gaspard, T.M., Kim, S.A., Walker, L.E., Vrana, S.L., and Dworkin, S.I., Dose-effect functions for cocaine self-administration: effects of schedule and dosing procedure, Pharmacol. Biochem. Behav., 57(3), 523, 1997. 89. Carroll, M.E. and Meisch, R.A., Determinants of increased drug self-administration due to food deprivation, Psychopharmacology, 74, 197, 1981. 90. Ahmed, S.H. and Koob, G.F., Cocaine- but not food-seeking behavior is reinstated by stress after extinction, Psychopharmacology (Berl.), 132(3), 289, 1997. 91. Worley, C.M., Valadez, A., and Schenk, S., Reinstatement of extinguished cocainetaking behavior by cocaine and caffeine, Pharmacol. Biochem. Behav., 48(1), 217, 1994. 92. Yokel, R.A. and Pickens, R., Drug level of d- and l-amphetamine during intravenous self-administration, Psychopharmacologia, 34(3), 255, 1974. 93. Wilson, M.C., Hitomi, M., and Schuster, C.R., Psychomotor stimulant self administration as a function of dosage per injection in the rhesus monkey, Psychopharmacologia, 22(3), 271, 1971. 94. Nader, M.A. and Woolverton, W.L., Effects of increasing response requirement on choice between cocaine and food in rhesus monkeys, Psychopharmacology, 108(3), 295, 1992. 95. Nader, M.A. and Woolverton, W.L., Choice between cocaine and food by rhesus monkeys: effects of conditions of food availability, Behav. Pharmacol., 3(6), 635, 1992. 96. Katz, J.L., Models of relative reinforcing efficacy of drugs and their predictive utility, Behav. Pharmacol., 1(4), 283, 1990.

2345_frame_MASTER.book Page 48 Wednesday, October 23, 2002 1:03 PM

48

Methods in Drug Abuse Research: Cellular and Circuit Level Analyses

97. Stafford, D., LeSage, M.G., and Glowa, J.R., Progressive-ratio schedules of drug delivery in the analysis of drug self-administration: a review, Psychopharmacology (Berl.), 139(3), 169, 1998. 98. Winger, G., Fixed-ratio and time-out changes on behavior maintained by cocaine or methohexital in rhesus monkeys. I. Comparison of reinforcing strength, Exp. Clin. Psychopharmacol., 1(1), 142, 1993. 99. Hursh, S.R. and Winger, G., Normalized demand for drugs and other reinforcers, J. Exp. Anal. Behav., 64(3), 373, 1995. 100. Ko, M.C., Terner, J., Hursh, S., Woods, J.H., and Winger, G., Relative reinforcing effects of three opioids with different durations of action, J. Pharmacol. Exp. Ther., 301(2), 698, 2002. 101. Winger, G., Hursh, S.R., Casey, K.L., and Woods, J.H., Relative reinforcing strength of three N-methyl-d-aspartate antagonists with different onsets of action, J. Pharmacol. Exp. Ther. 301(2), 690, 2002. 102. Loh, E.A., Fitch, T., Vickers, G., and Roberts, D.C., Clozapine increases breaking points on a progressive-ratio schedule reinforced by intravenous cocaine, Pharmacol. Biochem. Behav. 42(3), 559, 1992. 103. Hodos, W., Progressive-ratio as a measure of reward strength, Science 134, 943, 1961. 104. Griffiths, R.R., Bradford, L.D., and Brady, J. V., Progressive ratio and fixed ratio schedules of cocaine-maintained responding in baboons, Psychopharmacology (Berl.), 65(2), 125, 1979. 105. Hoffmeister, F., Progressive-ratio performance in the rhesus monkey maintained by opiate infusions, Psychopharmacology (Berl.), 62(2), 181, 1979. 106. Roberts, D.C. and Bennett, S.A., Heroin self-administration in rats under a progressive ratio schedule of reinforcement, Psychopharmacology, 111(2), 215, 1993. 107. Bedford, J. A., Bailey, L.P., and Wilson, M.C., Cocaine reinforced progressive ratio performance in the rhesus monkey, Pharmacol. Biochem. Behav., 9(5), 631, 1978. 108. Stafford, D. and Branch, M.N., Effects of step size and break-point criterion on progressive-ratio performance, J. Exp. Anal. Behav., 70(2), 123, 1998. 109. Roberts, D.C., Breaking points on a progressive ratio schedule reinforced by intravenous apomorphine increase daily following 6-hydroxydopamine lesions of the nucleus accumbens, Pharmacol. Biochem. Behav., 32(1), 43, 1989. 110. Richardson, N.R. and Roberts, D.C., Progressive ratio schedules in drug self-administration studies in rats: a method to evaluate reinforcing efficacy, J. Neurosci. Methods, 66(1), 1, 1996. 111. Bergman, J., Kamien, J.B., and Spealman, R.D., Antagonism of cocaine self-administration by selective dopamine D(1) and D(2) antagonists, Behav. Pharmacol., 1(4), 355, 1990. 112. Goldberg, S.R., Comparable behavior maintained under fixed-ratio and second-order schedules of food presentation, cocaine injection or d-amphetamine injection in the squirrel monkey, J. Pharmacol. Exp. Ther., 186(1), 18, 1973. 113. Goldberg, S.R. and Tang, A.H., Behavior maintained under second-order schedules of intravenous morphine injection in squirrel and rhesus monkeys, Psychopharmacology (Berl.), 51(3), 235, 1977. 114. Alderson, H.L., Robbins, T.W., and Everitt, B.J., Heroin self-administration under a second-order schedule of reinforcement: acquisition and maintenance of heroinseeking behaviour in rats, Psychopharmacology (Berl.), 153(1), 120, 2000. 115. Markou, A., Arroyo, M., and Everitt, B.J., Effects of contingent and non-contingent cocaine on drug-seeking behavior measured using a second-order schedule of cocaine reinforcement in rats, Neuropsychopharmacology, 20(6), 542, 1999.

2345_frame_MASTER.book Page 49 Wednesday, October 23, 2002 1:03 PM

Self-Administration of Drugs of Abuse

49

116. Ranaldi, R. and Roberts, D.C., Initiation, maintenance and extinction of cocaine selfadministration with and without conditioned reward, Psychopharmacology (Berl.), 128(1), 89, 1996. 117. Dougherty, J. and Pickens, R., Fixed-interval schedules of intravenous cocaine presentation in rats, J. Exp. Anal. Behav., 20(1), 111, 1973. 118. Everitt, B.J. and Robbins, T.W., Second-order schedules of drug reinforcement in rats and monkeys: measurement of reinforcing efficacy and drug-seeking behaviour, Psychopharmacology (Berl.), 153(1), 17, 2000. 119. Klein, L.C., Popke, E.J., and Grunberg, N.E., Sex differences in effects of predictable and unpredictable footshock on fentanyl self-administration in rats, Exp. Clin. Psychopharmacol., 5(2), 99, 1997. 120. Lynch, W.J., Kushner, M.G., Rawleigh, J. M., Fiszdon, J., and Carroll, M.E., The effects of restraint stress on voluntary ethanol consumption in rats, Exp. Clin. Psychopharmacol., 7(4), 318, 1999. 121. Lynch, W.J. and Carroll, M.E., Regulation of intravenously self-administered nicotine in rats, Exp. Clin. Psychopharmacol., 7(3), 198, 1999. 122. Lynch, W.J. and Carroll, M.E., Sex differences in the acquisition of intravenously self-administered cocaine and heroin in rats, Psychopharmacology (Berl.), 144(1), 77, 1999. 123. Brown, Z.W., Amit, Z., and Weeks, J.R., Simple flow-thru swivel for infusions into unrestrained animals, Pharmacol. Biochem. Behav., 5(3), 363, 1976.

2345_frame_MASTER.book Page 50 Wednesday, October 23, 2002 1:03 PM

2345_frame_MASTER.book Page 51 Wednesday, October 23, 2002 1:03 PM

3

Application of In Vivo Microdialysis Methods to the Study of Psychomotor Stimulant Drugs Michael H. Baumann and John J. Rutter

CONTENTS 3.1 3.2

3.3

3.4

3.5

Introduction ....................................................................................................52 Basic Principles of Microdialysis ..................................................................53 3.2.1 Microdialysis Probes and Diffusion ..................................................53 3.2.2 Perfusion Fluid...................................................................................55 3.2.3 Relative vs. Absolute Recovery .........................................................55 3.2.4 Probe/Tissue Interactions ...................................................................57 Surgical Procedures........................................................................................58 3.3.1 Implantation of an In-Dwelling Jugular Catheter .............................58 3.3.1.1 Preparation of the Catheter.................................................59 3.3.1.2 Surgical Implantation of the Catheter ................................59 3.3.1.3 Post-Surgical Catheter Maintenance ..................................60 3.3.2 Implantation of an Intracranial Guide Cannula.................................60 3.3.2.1 Surgical Implantation of the Cannula ................................61 3.3.2.2 Fixing the Implant to the Skull ..........................................62 Microdialysis Methods...................................................................................62 3.4.1 Microdialysis Probes..........................................................................62 3.4.2 Perfusion Fluid...................................................................................62 3.4.3 In Vitro Microdialysis ........................................................................63 3.4.4 In Vivo Microdialysis in Awake Animals ..........................................64 3.4.5 Running Experiments.........................................................................65 Analytical Methods ........................................................................................66 3.5.1 Instrumentation...................................................................................67 3.5.1.1 HPLC Pump........................................................................68 3.5.1.2 Flow Cell Compartment .....................................................68 3.5.1.3 Electrochemical Detector....................................................70 51

2345_frame_MASTER.book Page 52 Wednesday, October 23, 2002 1:03 PM

52

Methods in Drug Abuse Research: Cellular and Circuit Level Analyses

3.5.2 3.5.3

Mobile Phase Conditions ...................................................................70 Data Acquisition and Analysis...........................................................71 3.5.3.1 Standard Curves..................................................................72 3.5.3.2 Assaying Dialysate Samples ..............................................74 3.6 Applications ...................................................................................................75 3.6.1 Acute Effects of Stimulants ...............................................................75 3.6.2 Chronic Effects of Stimulants............................................................78 3.6.3 Implications for Medications Development ......................................79 3.7 Conclusions ....................................................................................................82 Acknowledgments....................................................................................................82 References................................................................................................................83

3.1 INTRODUCTION In vivo microdialysis has become a standard approach for studying brain neurochemistry in laboratory animal models (reviewed in Reference 1). In vivo microdialysis is a perfusion technique that allows continuous sampling of the extracellular fluid (ECF) from intact living tissue. A small microdialysis probe is implanted into the brain, and the probe is perfused with a salt solution that closely mimics the composition of the ECF. Low-molecular-weight solutes from the ECF, such as neurotransmitters and hormones, move by diffusion across the probe membrane into the perfusion medium and are collected at specific time intervals. While in vivo microdialysis has been best adapted for use in brain tissue, any organ or tissue is amenable to the technique. The theoretical rationale for utilizing a dialysis membrane to sample the ECF was put forth more than 30 years ago.2,3 Practical application of the technique was developed by pioneering investigators who demonstrated that small-diameter dialysis fibers implanted in the brain afforded ready access to the extracellular compartment.4 Once dialysate samples are obtained they can be assayed for a variety of bioactive compounds using sophisticated analytical techniques. For example, high-performance liquid chromatography with electrochemical detection is commonly used for the detection of monoamine neurotransmitters — norepinephrine (NE), dopamine (DA), and serotonin (5-HT).5,6 Radioimmunoassay methods are used for the detection of larger-molecular-weight substances, such as neuropeptides.7 It is important to note that highly sensitive analytical techniques are essential when assaying dialysate samples due to small sample size (<20 ml) and the extremely low basal amounts of neurotransmitters (<1 pg) present in the ECF. The popularity of in vivo microdialysis as a tool to study brain neurochemistry is reflected by the dramatic increase in publications reporting microdialysis data over the past two decades. In particular, in vivo microdialysis has been instrumental in evaluating the neurochemical mechanisms underlying the effects of illicit stimulants like cocaine and amphetamine. Using the Pub Med database we have found that the number of papers utilizing in vivo microdialysis to investigate the neurochemical effects of cocaine and amphetamines has risen dramatically over the past two decades, a trend paralleled by the increase in the general use of microdialysis

2345_frame_MASTER.book Page 53 Wednesday, October 23, 2002 1:03 PM

Application of In Vivo Microdialysis Methods

Total Amphetamines Cocaine

600

40

400

20

200

0

Total Papers

Stimulant Papers

60

53

0 84 86 88 90 92 94 96 98 00 Year

FIGURE 3.1 Number of published papers utilizing in vivo microdialysis to investigate the neurochemical effects of amphetamines and cocaine. The total number of papers reporting the use of in vivo microdialysis in brain tissue has risen steadily since the inception of the technique. A parallel rise has occurred in the number of these studies that investigated the effects of stimulant drugs.

(see Figure 3.1). The growth in the popularity of in vivo microdialysis has occurred for a number of reasons: 1) researchers have realized that in vivo microdialysis offers distinct advantages over other in vivo neurochemical methods,8 2) there has been a rapid dissemination of information about perfecting and implementing the microdialysis technique, and 3) there have been major advances in the analytical methods required for assaying bioactive molecules in dialysate samples. The main goal of this chapter is to describe the use of in vivo microdialysis to study the neurochemical effects of stimulant drugs in the laboratory rat. We will achieve this aim by first summarizing some basic principles of in vivo microdialysis. This will be followed by a detailed account of practical information regarding surgical implantation of guide cannulae and probes, instrumentation and materials needed for microdialysis, and analysis of samples using high-performance liquid chromatography with electrochemical (EC) detection. We will then discuss the application of in vivo microdialysis for determining the acute and long-term effects of stimulant drugs on monoamine neurotransmission (NE, DA, 5-HT) in rat brain. Finally, the application of in vivo microdialysis in the development of medications to combat stimulant addiction will be considered.

3.2 BASIC PRINCIPLES OF MICRODIALYSIS 3.2.1 MICRODIALYSIS PROBES

AND

DIFFUSION

The microdialysis probe serves as the collection device for in vivo sampling. The probe is constructed (either in the laboratory or by the commercial vendor) to provide a means for introducing a section of microdialysis tubing into the brain. Probe design must be optimized to balance the constraints of very small size to minimize tissue damage yet remove sufficient amounts of ECF for sample analysis. In many respects, the microdialysis probe functions as an artificial capillary. Like the flow of blood

2345_frame_MASTER.book Page 54 Wednesday, October 23, 2002 1:03 PM

54

Methods in Drug Abuse Research: Cellular and Circuit Level Analyses

outflow

inflow

steel shaft

dialysis membrane

solute flux across the membrane is bi-directional

dialysis probe tip

FIGURE 3.2 Schematic diagram of a concentric-style microdialysis probe. Vertical arrows indicate path of fluid flow within the probe; horizontal arrows depict the bidirectional exchange, via diffusion, of low molecular weight substances across the membrane at the tip of the probe.

in a capillary, there is normally no net fluid gain or loss across the dialysis membrane; the probe, therefore, represents a closed fluid system, separated from the surrounding tissue by the dialysis membrane. Because of this separation, the tissue around the probe is not subject to the damaging effects of rapid fluid transfer found in earlier perfusion techniques like push–pull perfusion. Slow perfusion of fluid in the probe allows dissolved solutes to enter or leave the perfusate via simple diffusion through the dialysis membrane. Due to the size exclusion characteristics of the dialysis membrane, large-molecular-weight solutes (e.g., enzymes that hydrolyze smaller molecules) do not enter the probe perfusate. Thus, dialysate samples are “clean” and can be analyzed without additional sample preparation. Figure 3.2 shows a schematic diagram of a microdialysis probe that has a typical concentric tube-within-a-tube design. Perfusion fluid is delivered by a microinjection pump and flows into the top of the probe through flexible, narrow-bore inflow tubing. The fluid flows down a central stainless steel shaft and exits the shaft near the probe tip. It is here, at the probe tip, where exchange of materials occurs across the dialysis membrane. Fluid then flows upward in the space between the dialysis membrane and the outer wall of the stainless steel shaft. Ultimately, the fluid flows from the probe into outflow tubing that leads to a collection vial. Dissolved solutes move across the dialysis membrane via the process of diffusion, which dictates that molecules move from an area of high concentration to an area of low concentration. The movement of specific solute molecules will continue until the concentration gradient for that solute is dissipated and equilibrium is reached. During in vivo microdialysis, exchange of materials across the dialysis membrane is bidirectional (Figure 3.2), governed by solute concentration gradients between the perfusion fluid and ECF. For example, endogenous neurotransmitters like DA are present in the ECF but not in the perfusion fluid. Therefore, DA readily diffuses from the ECF into the perfusate along its own concentration gradient. Exogenous drugs that are added to the perfusion fluid will diffuse from the perfusate

2345_frame_MASTER.book Page 55 Wednesday, October 23, 2002 1:03 PM

Application of In Vivo Microdialysis Methods

55

into the local microenvironment surrounding the probe. It is noteworthy that when drugs are administered through the probe, neurotransmitters from the ECF are simultaneously extracted. Thus, the local infusion of drugs and the determination of neurotransmitter responses evoked by those drugs can be accomplished concurrently.

3.2.2 PERFUSION FLUID The composition of the perfusion medium is an important consideration when conducting in vivo microdialysis experiments. Neurons in the brain, like other living cells, are bathed in ECF that is tightly regulated in terms of constituent ions, pH, and osmolarity. Unlike other cells, however, neurons are excitable cells capable of transmitting impulses and are therefore exquisitely sensitive to changes in ionic makeup of the ECF. For example, action potentials in neurons are dependent upon normal Na+ ion concentration in the ECF; decreases in ECF Na+ will render neurons unable to transmit impulses. Indeed, adding a Na+ channel blocker like tetrodotoxin to the perfusate can be used to verify the neuronal origin of transmitter molecules collected in the dialysate.9,10 Likewise, normal exocytotic release of neurotransmitters is dependent upon Ca++. Changes in Ca++ concentration in the ECF can markedly affect basal levels of endogenous neurotransmitters. Boosting Ca++ concentrations in the ECF is one mechanism used to elevate basal levels of neurotransmitters in the dialysate. While this augmentation facilitates detection of the neurotransmitters of interest, it may disrupt both the normal physiological regulation of neurotransmitter release as well as the sensitivity to pharmacological manipulation.10,11 For these reasons, the composition of the microdialysis perfusion fluid should closely mimic the composition of neuronal ECF.11 At the bare minimum, perfusion fluid must contain the following ions at the concentrations given: 145 mM Na+, 2.8 mM K+, 1.2 mM Ca++, and 150 mM Cl–. More complete perfusion fluids might be buffered to neutral pH with phosphate and contain Mg++ or glucose. As with the probes, the perfusion fluid can be made in the laboratory or purchased from a vendor.

3.2.3 RELATIVE

VS.

ABSOLUTE RECOVERY

The main objective when performing in vivo microdialysis experiments is to recover endogenous bioactive molecules from the brain ECF. Therefore, it is of great interest to optimize the conditions of any in vivo microdialysis system to maximize recovery. For low-molecular-weight solutes, such as monoamines, the major factor governing solute recovery is the rate of diffusion through the ECF. At this point, it is important to define two terms that can be used to express recovery of endogenous molecules. Relative recovery is the concentration of a solute recovered in the dialysate sample with respect to the concentration of that solute in the ECF. Absolute recovery, on the other hand, refers to the actual amount of solute recovered in the dialysate sample per unit of time. Figure 3.3 shows the relationship between perfusion flow rate and recovery of DA as determined in an in vitro system. The two measures of recovery and their practical significance will be discussed below. Since relative recovery is a ratio of solute concentration in the dialysate sample relative to the source medium, it is most often expressed as a percentage. Relative

2345_frame_MASTER.book Page 56 Wednesday, October 23, 2002 1:03 PM

56

Methods in Drug Abuse Research: Cellular and Circuit Level Analyses

Relative Absolute

80

6

60

4

40

2

20

0

0 0

2

4

6

8

10

Relative Recovery (%)

Absolute Recovery (pg/20 min)

8

12

Flowrate (µl/min)

FIGURE 3.3 Results of an in vitro probe-recovery experiment. An increase in perfusion flow rate results in an elevation of the total amount (pg) of analyte recovered (absolute recovery); a concomitant decrease occurs in the relative amount (%) of analyte recovered (relative recovery) since decreased time is allowed for diffusional exchange to occur.

recovery is influenced by a number of factors including the concentration gradient for the solute, dimensions of the dialysis probe (particularly those that influence the surface area for exchange), intrinsic properties of the dialysis membrane, and the flow rate of the perfusion fluid. In addition, the characteristics of the environment outside the probe influence recovery of substances; for this reason, relative recovery is most accurately determined in an in vitro experiment as depicted in Figure 3.4. While holding other variables constant, one observes an increase in relative recovery with a decrease in flow rate (see Figure 3.3). As the flow rate is decreased and FEP inflow tubing

tubing adapters

FEP outflow tubing

dialysis probe clip To collection vial

From microsyringe pump dialysis probe

in vitro stand test solution containing analyte of interest

FIGURE 3.4 Schematic diagram of the apparatus used for an in vitro probe-recovery experiment and for preexperimental probe preparation. Probe inflow/outflow tubing, the microdialysis probe mounted in a probe clip, and the vial of test solution are shown.

2345_frame_MASTER.book Page 57 Wednesday, October 23, 2002 1:03 PM

Application of In Vivo Microdialysis Methods

57

approaches zero, the movement of the perfusate through the probe is slow enough to allow nearly complete equilibrium across the membrane, and relative recovery approaches 100%. Although an accurate determination of relative recovery requires an in vitro approach, decreasing the perfusion flow rate to very low levels has been used by some investigators to estimate the “true” concentration of solute in the ECF.11 At the typical flow rates used during in vivo microdialysis experiments (0.5 to 2.5 ml/min), relative recovery of monoamine neurotransmitters ranges from 15 to 25% as determined in vitro. Absolute recovery is the actual amount of total solute, irrespective of neurotransmitter content, recovered over time; one way to increase this measure is to simply increase the perfusion flow rate. While dialysate samples will be more dilute under these fast flow circumstances, a larger sample size will hold more solute. In reality, this strategy only works up to the point where the perfusion flow rate exceeds the rate of diffusion of the solute between the ECF and the perfusate. When deciding what flow rate to use during in vivo microdialysis experiments the investigator must balance the relative and absolute recovery constraints to determine the perfusion flow rate that will yield optimum recovery of solute under a given set of conditions. It is important to note that while in vitro determination of probe recovery allows determination of the best yield of neurotransmitter concentration per unit time (and serves as an excellent training tool), extrapolation of these characteristics for calculation of “true” neurotransmitter concentration in ECF under in vivo conditions is probably not valid. The in vitro test conditions do not accurately mimic the complex matrix of brain tissue, and they do not take into consideration the active processes of reuptake and release.12 Many pharmacological investigations do not require calculation of true extracellular concentrations; establishment of basal neurotransmitter levels for individual animals prior to drug treatment allows comparison of predrug baselines to post-treatment response levels, and this is the comparison made for grouped effects of a compound. Valid determination of actual levels, such as changes in basal neurotransmitter content following chronic treatment of a stimulant, requires application of one of several quantitative approaches including the slow flow experiment described above. The principles behind these applications are described in detail in several excellent reviews, to which the reader is referred.13,14

3.2.4 PROBE/TISSUE INTERACTIONS The probe tip, which represents the “working end” of the system, is quite large compared to the cells surrounding the dialysis membrane (e.g., membranes are typically 200 to 500 mm in diameter). This discrepancy has important consequences for both the protocol of the experiment and the interpretation of the results obtained. First, immediately following implantation of the probe, recovery of bioactive molecules is very high due to release of compounds from the cells displaced or damaged by the probe. While the extent of this en masse release can be reduced by slow implantation of the probe, some damage is inevitable. Analysis of the time course of the implantation-induced neurotransmitter elevation suggests that accurate basal levels return 8 to 12 h after probe insertion.10,15 In keeping with this observation, we generally implant probes on the evening before an experiment, thereby allowing

2345_frame_MASTER.book Page 58 Wednesday, October 23, 2002 1:03 PM

58

Methods in Drug Abuse Research: Cellular and Circuit Level Analyses

sufficient recovery from the damage prior to the determination of baseline levels of the interested compounds. Second, probe size precludes sampling from individual neurons; the dialysate therefore represents “overflow” of transmitter from the surrounding population of cells. More specifically, the neurotransmitter outside the probe is the result of the summed effects of release, reuptake, and metabolism. Interpretation of the data therefore requires consideration of these opposing processes; contributions of individual factors can be discerned by application of pharmacological agents with specificity for particular aspects of neuronal function.

3.3 SURGICAL PROCEDURES One of the primary advantages of in vivo microdialysis has been its adaptation for sampling from conscious, freely moving animals. This allows the experimenter to circumvent the complications of general anesthesia and provides a means for sampling of the ECF concurrent with behavioral observation and testing. Thus, all necessary surgical procedures must be carried out well in advance of the actual microdialysis sampling. In most cases, we surgically implant an in-dwelling intravenous (i.v.) catheter at the time of implantation of the intracranial guide cannula. The i.v. catheter allows us to administer drugs to the animal during an experiment without the stressful effects of handling and also permits simultaneous sampling of blood for hormonal or pharmacokinetic determinations. The intracranial guide cannula serves as a stationary pathway to direct the probe tip to a specific region in the brain. Under these conditions, the catheter and probe guide must be implanted in a manner that will not compromise normal and drug-induced behavior on the day of the experiment. The following account describes in detail the surgical approaches used for the implantation of both i.v. catheters and the intracranial guide cannulae. As these surgeries require a general anesthetic like sodium pentobarbital, the operations are performed at least 1 week prior to the actual experiment to allow the animal sufficient time to recover from the trauma of the surgery. Insertion of the dialysis probe through the previously implanted guide occurs the night before an experiment and requires only a brief reduction in consciousness, usually via an inhalation or short-lived i.v. anesthetic. All of the surgical procedures described below are performed under aseptic conditions with surgeons donning sterile gloves and using sterilized instruments. Animals are maintained in facilities fully accredited by the American Association of the Accreditation of Laboratory Animal Care (AAALAC), and procedures are performed in accordance with the Institutional Care and Use Committee of the National Institute on Drug Abuse (NIDA), Intramural Research Program (IRP).

3.3.1 IMPLANTATION

OF AN IN-DWELLING JUGULAR

CATHETER

The objective is to surgically implant a catheter (i.e., soft flexible tube) into the right jugular vein of a rat for the purpose of drug administration or blood withdrawal. When the surgery is complete, the proximal tip of the tube resides within the right atrium, whereas the distal end of the tube is exteriorized on the nape of the neck

2345_frame_MASTER.book Page 59 Wednesday, October 23, 2002 1:03 PM

Application of In Vivo Microdialysis Methods

59

between the ears. The catheter itself has a hybrid structure that reflects the varied requirements of the device: at one end is a flexible, medical-grade Silastic tubing (OD 0.94 mm ¥ ID 0.51 mm, cat. #508–002, Dow-Corning, Midland, MI) that is inserted into the blood vessel whereas at the other end is a more rigid polyvinyl tubing (OD 1.00 mm ¥ ID 0.50 mm, cat. #111037, Critchley Electrical Products, Silverwater N.S.W., Australia) that emerges from the animal’s back. The flexibility and inert nature of the Silastic discourages clot formation, while the vinyl tubing can withstand the occasional investigation of an overly curious rat. 3.3.1.1 Preparation of the Catheter The Silastic tubing is joined to a 5-cm piece of vinyl tubing via a 1-cm 23-gauge (0.6-mm outer diameter) stainless steel thin-wall connecting tube (cat. #HTX-23TW, Small Parts, Miami, FL). The juncture between the Silastic and vinyl tubing is reinforced with shrink tubing (cat. #SMT-18, Small Parts), and a 3–0 silk ligature is tied tightly around the juncture at the level of the 23-gauge connector. The catheter is soaked in a cold sterilant such as Cidex prior to implantation. The Silastic end of the catheter is cut to a specific length depending on the weight of the rat; adjustment of this length accounts for changes in the jugular to right atrium distance as the rat grows. The appropriate length is represented by the body weight in grams divided by 10, with the resultant number being the length, in mm, of the inserted end of the cannula. For example, a 300-g rat would get a 30-mm insert, a 350-g rat would get a 35-mm insert, and so on. 3.3.1.2 Surgical Implantation of the Catheter The rat is anesthetized with sodium pentobarbital (60 mg/kg i.p.). Once the plane of anesthesia is reached the rat is shaved along the ventral surface of the neck between the chin and the chest and on the top of the head between the ears. The rat is placed belly up with its head facing the surgeon. A small incision is made through the skin above the right collarbone about halfway between the midline and the side of the animal. The right jugular vein is isolated by blunt dissection, and a 1-cm region anterior to the point where the vein enters the thoracic cavity (below the pectoralis muscle) is carefully freed from the surrounding connective tissue. Before implantation into the jugular the distal vinyl end of the i.v. catheter is run through a subcutaneous tunnel to exit through the skin on the back between the shoulders and connected to a syringe filled with saline. After filling the catheter completely with saline, the proximal Silastic end is connected to a 20-gauge guide needle via a fitted 23-gauge insert. The beveled tip of the guide needle is inserted into the jugular vein pointing towards the heart and advanced so that the needle exits through the overlying pectoralis muscle. The needle is removed, and the Silastic portion of the catheter is drawn back into the body through the muscle until it falls into the jugular; the catheter can now be easily advanced into the right atrium. Correct placement of the catheter is verified by flushing a small amount of saline into the catheter and subsequently withdrawing blood into the line. The catheter is anchored to the neighboring hyoid muscle, filled with 0.5 ml of 48 IU/ml heparin saline, and

2345_frame_MASTER.book Page 60 Wednesday, October 23, 2002 1:03 PM

60

Methods in Drug Abuse Research: Cellular and Circuit Level Analyses

the vinyl end plugged with a 23-gauge stylet. The wound is closed with surgical staples, and the rat is placed into the stereotaxic apparatus for surgical implantation of the guide cannula. 3.3.1.3 Post-Surgical Catheter Maintenance Properly maintained catheters can remain patent for several days to several weeks, provided the lines are flushed to prevent buildup of deposits from the bloodstream. In addition, flushing habituates the animals to the procedure, thereby reducing stressrelated confounds in the experiment. Prepare catheter “flush” solution using sterile reagents to prevent infection: to a 20-ml vial of sterile saline, add 1 ml of sterile 1000 IU/ml heparin to yield a solution of 48 IU/ml heparinized saline. Daily flushing with this solution is recommended. If catheter patency is in question, perform a patency test by injecting 5 mg/kg i.v. methohexital (Brevital sodium) into the catheter over a period of about 5 sec. If the catheter is patent, the rat will be sleeping by the end of the injection . When the rat awakens, flush catheter with saline and plug.

3.3.2 IMPLANTATION

OF AN INTRACRANIAL

GUIDE CANNULA

The objective is to surgically implant a guide cannula (i.e., a rigid plastic tube) above the rat nucleus accumbens; the cannula serves as a rigid guide for the subsequent implantation of a microdialysis probe into the brain (Figure 3.5). When the surgery is complete, the proximal cannula tip will reside just above the nucleus accumbens, whereas the distal end will be exteriorized on the top of the head. The following procedure describes implantation of a cannula purchased from the probe manufacturer; slight modification of the approach may be necessary for those using probes

balance arm

collection vial i.v. catheter extension

fluid swivel wire tether

dialysis chamber

motion sensors

FIGURE 3.5 Depiction of a rat undergoing in vivo microdialysis within a testing chamber. The rat is attached to a tethering system that allows free movement within the chamber; perfusion lines for the implanted probe are routed through a fluid swivel, as described in the text.

2345_frame_MASTER.book Page 61 Wednesday, October 23, 2002 1:03 PM

Application of In Vivo Microdialysis Methods

61

and guides constructed in the laboratory. Readers interested in neurochemical analysis within different brain regions can follow this procedure, substituting alternate stereotaxic coordinates for the site of interest.16 3.3.2.1 Surgical Implantation of the Cannula The cannula for implantation is a commercially manufactured product that consists of a CMA/12 plastic guide cannula (cat. #830 9024, CMA/Microdialysis (CMA), Acton, MA) and a tight-fitting dummy probe that serves to plug the cannula. The cannula is removed from its packaging and soaked in a cold sterilant such as Cidex. The cannula is mounted in a probe clip (cat. #830 9102, CMA) that is fixed to a stereotaxic adaptor and connecting rod (cat. #830 9004, CMA). The connecting rod is, in turn, attached to the carrier arm of a stereotaxic apparatus (cat. #51600, Stoelting, Wood Dale, IL). The anesthetized rat is mounted in the stereotaxic frame, with the head immobilized according to the flat-skull position.17 A 2-cm midline sagittal incision is made with a scalpel blade on the top of the head to expose the skull. It is important to carefully clean and dry the skull surface since this will facilitate both accurate positioning of the guide and adherence of the cement. The skull surface is vigorously wiped to remove the connective tissue adhering to the skull. The skin flaps on both sides of the incision are retracted using serrefine clamps. Excessive bleeding during this procedure can be stopped by gently rubbing the dull end of a scalpel blade on the skull surface over the region of the bleeding. The intersection of the sagittal and coronal sutures is identified; this landmark, known as bregma, serves as a reference point on the skull surface. The carrier arm of the stereotaxic apparatus is locked in position over the exposed skull surface. Using the thumbscrews on the stereotaxic carrier arm, position the cannula tip extending down from the connecting rod directly on the skull surface at bregma; the resultant coordinates in the anteroposterior (AP), mediolateral (ML), and dorsoventral (DV) planes are recorded. The carrier arm is raised up above the skull. Using the bregma coordinates as reference “zero,” the micrometers are adjusted to the appropriate position for the nucleus accumbens: +1.6 mm in the AP plane and –1.7 mm in the ML plane. The carrier arm is lowered back down to the skull surface and a small mark, indicating the site for drilling of a burr hole, is made on the skull. The burr hole is drilled using a high-speed microdrill (cat. #18000–17, Fine Science Tools [FST], Foster City, CA) and a 2.1-mm steel burr (cat. #190007–21, FST). The burr hole is made through the skull, just deep enough to expose the meninges covering over the surface of the brain below. The tough dura mater is carefully punctured using a sterile 25-g stainless steel needle tip. Three additional “screw” holes are drilled into the skull in areas flanking the first hole; each of these holes is made on a different skull bone. Three #0–80 stainless steel machine screws (cat. #MX-080–2FL, Small Parts), which will serve to anchor the guide implant, are carefully seated into the holes and screwed in. The stereotaxic carrier arm is lowered into the first hole to touch the surface of the dura mater, and stereotaxic coordinates in the dorsoventral (DV) plane are recorded. The carrier arm is lowered slowly to a

2345_frame_MASTER.book Page 62 Wednesday, October 23, 2002 1:03 PM

62

Methods in Drug Abuse Research: Cellular and Circuit Level Analyses

position 6.2 mm below dura; a gradual descent (1-mm steps) helps to minimize trauma. The cannula is now properly placed just dorsal to the nucleus accumbens. 3.3.2.2 Fixing the Implant to the Skull After the skull has been carefully cleaned and dried, powdered dental acrylic (Coltene/Whaledent, Mahwah, NJ) is piled around the cannula and the three anchor screws. Liquid catalyst is added to the powder, and the dental cement is allowed to harden for a few minutes. Subsequent layers of dental cement are built up to form a solid stage that completely surrounds and covers the implant. The guide cannula is released from the stereotaxic carrier arm, and additional dental cement is added if needed. Stainless steel wound clips are used to close the wound if necessary. The rat is placed on a warmed heating pad until it has fully recovered from the anesthesia. Rats are singly housed post-operatively, and animals can be provided with acetominophen (300 mg/100 ml) in their drinking water for 1–2 d post-operatively as an analgesic.

3.4 MICRODIALYSIS METHODS 3.4.1 MICRODIALYSIS PROBES We use commercially available CMA/12 probes (cat. #830 9562, CMA) in our studies. The CMA/12 probe has an outer stainless steel shaft that connects to a 2mm polycarbonate–polyether membrane with a molecular weight cutoff of 20,000 Da. The membrane offers excellent biocompatibility and superior diffusion characteristics. The outer diameter of the steel shaft is 0.64 mm, whereas the outer diameter of the membrane is 0.55 mm. These probes are reliable, can be ordered with varying membrane lengths and molecular weight cutoffs, and can be reused with proper post-experimental treatment. As usual, the convenience of buying probes comes at a cost; for this reason, those wishing to save money can make their own probes by following one of the approaches found in the literature.1,17

3.4.2 PERFUSION FLUID As discussed above, the composition of the perfusion fluid within the probe is critical to successful exploration of stimulant-induced changes of neurotransmitters in the ECF. Perfusion fluid can be made up in the laboratory or purchased from commercial sources. In either case, great care must be taken to insure that perfusion fluid is absolutely free of debris and contaminants. We typically use commercially available artificial cerebrospinal fluid (aCSF) perfusion medium (cat. #AH 59–7316, Harvard Apparatus, Holliston, MA) that contains 150 mM Na+, 3 mM K+, 1.4 mM Ca++, 0.8 mM Mg++, and 155 mM Cl–. To make up a simple Ringers’ perfusion fluid in the laboratory, 1 M stock solutions of ultrapure reagent grade NaCl, KCl, and CaCl2*2H2O should be prepared every 2 weeks. Stocks should be kept refrigerated; this, along with the high salt concentration of the stock solutions, deters bacterial contamination. A 1.0-M NaCl stock solution is prepared by dissolving 58.44 g NaCl (cat. #S 7653, Sigma Chemical Co. (Sigma), St. Louis, MO) into 1 L distilled water.

2345_frame_MASTER.book Page 63 Wednesday, October 23, 2002 1:03 PM

Application of In Vivo Microdialysis Methods

63

To prepare the 1-M KCl stock solution, dissolve 7.45 g KCl (cat. #P 333, Sigma) into 100 ml distilled water. For the CaCl2*2H2O stock solution, dissolve 14.70 g CaCl2*2H2O (cat. #C 5080, Sigma) into 100 ml distilled water. Fresh Ringers’ solution is prepared daily by adding 75 ml of 1 M NaCl, 1.5 ml of 1 M KCl, and 0.8 ml of 1 M CaCl2*2H2O to a 500-ml volumetric flask, and final volume is achieved with distilled water. This Ringers’ recipe yields a solution of 150 mM Na+, 3 mM K+, 1.6 mM Ca++, and 156 mM Cl–. All perfusion fluids are filtered through 0.22mM filters under vacuum whenever possible. At a bare minimum, perfusion fluid should be filtered through 0.22-mM Millex-GS syringe filters (cat. #SLGS 025 0S, Millipore, Bedford, MA) before use.

3.4.3 In Vitro MICRODIALYSIS Before microdialysis investigations are carried out, assay methods for detecting and quantifying an analyte of interest (e.g., neurotransmitter) must be developed and tested (see Section 3.5, Analytical Methods). Once such assay methods are verified, an easy way to begin performing microdialysis sampling is to run an in vitro dialysis experiment. A schematic diagram for setting up such an experiment is shown in Figure 3.4. A number of instruments, devices, and accessories are required: syringe, microinfusion pump, narrow-bore tubing and connectors, microdialysis probe, probe clip affixed to an in vitro stand, and plastic vial filled with test solution. Prior to running the in vitro experiment, all of the components of the perfusion system must be connected together. A 5.0-ml gas-tight syringe (cat. #830 9021, CMA) is filled with filtered aCSF and mounted in a CMA 100 microinjection pump (cat. #821 0040, CMA). It is critical that the perfusion fluid be at room temperature and free of air bubbles. The pump is set at 5 ml/min and started so that aCSF leaves the tip of the syringe. A 30-cm length of narrow-bore FEP tubing (cat. #840 950, CMA) is connected to the needle tip of the syringe using a special tubing adapter (cat. #340 9500, CMA). Subsequent connections between the FEP tubing and the probe are also made using the tubing adapters. The tubing adapters are soaked in ethanol prior to use in order to slightly expand their internal diameter; when the adapters dry, they shrink to afford a tight connection. A CMA/12 microdialysis probe is mounted in a microdialysis clip that is affixed to a CMA/130 in vitro stand (cat. #830 9102, CMA). The probe is oriented so that the dialysis membrane is immersed in a 1.5-ml Eppendorf vial of perfusion medium. The FEP tubing from the syringe is connected to the metal inlet of the microdialysis probe while aCSF is pumping at 5 ml/min. Once fluid elutes from the probe outlet, a 30-cm length of FEP tubing is connected to the metal outlet of the probe, and this outflow line is suspended in a collection vial. The connections between syringe, tubing, and probe are checked for leaks. It should be noted that once the dialysis membrane is wetted with perfusion fluid the membrane becomes sensitive to drying and must always be perfused if exposed to the air. In order for reliable, stable exchange to occur between the probe and its surroundings, the membrane itself must be cleaned of manufacturing contaminants, and the exchange surface must be purged of air bubbles that would decrease the available surface area for diffusion. The cleaning is accomplished by immersing the probe in

2345_frame_MASTER.book Page 64 Wednesday, October 23, 2002 1:03 PM

64

Methods in Drug Abuse Research: Cellular and Circuit Level Analyses

a vial of ethanol while pumping aCSF through the probe at 5 ml/min for 10 min. During the cleaning procedure the syringe, FEP tubing, and probe are checked for the presence of bubbles. The presence of air in the probe can be discerned by a quick check under magnification; a small “flick” of the probe is usually enough to knock any air pockets from the membrane. Once the probe has been cleaned and checked in this manner, the perfusion rate is slowed to 1 ml/min, and the probe is immersed in a solution aCSF. The probe is now ready for use. With the perfusion system connections set up, a simple probe-recovery experiment can be performed. Prepare a test solution of 4 ng/ml DA (~20 nM) dissolved in aCSF (this solution simulates the DA concentration in brain ECF), and fill a 1.5ml plastic Eppendorf vial with the DA solution. Immerse the microdialysis probe in the DA solution while perfusing aCSF through the probe at 1 ml/min. After a 30min equilibration period, dialysate samples are collected at 20-min intervals. Dialysate samples and aliquots of the DA test solution are immediately assayed for DA concentration using microbore HPLC with EC detection (see Section 3.5, Analytical Methods). The collection of three or four dialysate samples is sufficient. The relative recovery of DA under the in vitro conditions can be calculated by dividing the amount of DA in the dialysate sample by the amount of DA in the source vial solution. Typically, the relative recovery of DA is between 15 and 25%.

3.4.4 In Vivo MICRODIALYSIS

IN

AWAKE ANIMALS

Setting up the microdialysis perfusion system for awake rats is similar to the in vitro experiment described above, but there are some important differences related to the tethering and containment of the rat during the experiment. First, the FEP tubing from the infusion syringe is connected to a quartz-lined fluid swivel (cat. #375/D/22QM, Instech Solomon, Plymouth Meeting, PA) that is attached to a wire tether and collar connector (cat. #240 9090, CMA). The fluid swivel and tether are, in turn, suspended from a balance arm that is mounted on the wall of a Plexiglas dialysis chamber (see Figure 3.5). A length of FEP tubing connects the fluid swivel to the probe inlet. The fluid swivel/tether arrangement allows the rat to move within the container without tangling the FEP inflow and outflow lines. Second, the microdialysis probe is mounted in a microdialysis probe clip affixed to the side of the plastic container that will house the rat during the microdialysis experiment. Finally, the FEP outflow line from the microdialysis probe runs upward along the wire tether to a collection vial mounted atop the tether. As noted previously, when making the connections for the in vivo set up, perfusion fluid is pumped through the system at 5 ml/min. Once the probe is cleaned and checked for bubbles the flow rate is decreased to 1 ml/min, and the probe is ready to be inserted into the guide cannula. We have found that lightly anesthetizing the rat allows for easier and less traumatic insertion of a microdialysis probe into the guide cannula. A rat fitted with a guide cannula is weighed and receives 5 mg/kg i.v. methohexital (Brevital sodium) for anesthesia. Methohexital is an ultrashort-acting barbiturate that renders the rat unconscious for a few minutes. Once the rat is anesthetized, a plastic collar (cat. #743 1059, CMA) is fitted around its neck, and the collar is attached to the wire tether. A length of saline-filled vinyl tubing (i.e., catheter extension) is attached to

2345_frame_MASTER.book Page 65 Wednesday, October 23, 2002 1:03 PM

Application of In Vivo Microdialysis Methods

65

the i.v. catheter, and the catheter is flushed with saline. The dummy cannula is carefully removed from the implanted guide cannula, and a CMA/12 microdialysis probe is lowered into the guide tube. Great care must be taken not to damage the delicate membrane surface when inserting the probe. Furthermore, slowly sinking the probe through the brain tissue will help to minimize damage to the cellular components. During insertion of the probe into the brain, aCSF should be pumping through the probe at 1 ml/min. When the probe insertion is complete, aCSF should be freely flowing from the outflow tubing atop the tether. Each rat, now connected to the wire tether, is placed into its testing chamber, as shown in Figure 3.5. If the perfusion system appears to be functioning properly, the perfusion flow rate is set to the desired flow rate, and the probe is perfused in situ overnight. In some instances, there will be no aCSF flowing from the outflow tubing, and this situation must be rectified. A number of problems can impair fluid flow through the in vivo perfusion system. The most common malfunctions are related to leaks in the system. Carefully check for fluid leaks at all junctions, especially at the connections between FEP tubing and the probe inlet/outlet. If leaks are found, use new tubing adapters soaked in ethanol to correct the problem. Another major cause of flow problems is the ripping of the microdialysis membrane during probe insertion. Under these circumstances, perfusion fluid drains out into the brain ECF and does not flow back out through the probe. This situation is troublesome because severe edema can occur in the brain tissue surrounding the probe. The best way to correct this problem is to carefully remove the damaged probe and insert a new one.

3.4.5 RUNNING EXPERIMENTS On the morning following insertion of the probe, a 150-ml plastic collection vial is mounted on the end of the outflow tubing, and dialysate samples are collected. We perfuse aCSF through the probe at 1 ml/min and collect samples at 20-min intervals to yield a sample volume of 20 ml. In our laboratory, dialysate samples are immediately injected into an HPLC-EC system set up for the detection of monoamine transmitters such as DA and 5-HT (see section 3.5, Analytical Methods). Other investigators have opted to freeze the collected samples and store them at –70°C until the time of analysis. Either strategy for sample handling will work fine. In a typical pharmacological investigation, three or four baseline samples are collected, and a drug is injected systemically into the rat. Samples are collected at 20-min intervals for 2 h after drug injection. When baseline samples are collected, each rat can serve as its own control, and post-drug changes in dialysate transmitter levels are compared to predrug baseline levels. When attempting to correlate the timing of drug injection to the timing of drug effects, knowledge of the dead volume of the perfusion system is essential. Using Figure 3.6 as an example, suppose a drug is infused locally into the nucleus accumbens via the microdialysis probe, and we wish to know when the effect of the drug will be observed in the dialysate sample entering the collection vial. The perfusion flow rate is 1 ml/min, while the inflow and outflow tubing lengths are 300 mm each. It is known that 100 mm of FEP tubing has an inner volume of 1.2 ml, whereas the probe has a dead volume of 3 ml on the outlet side. With this information we calculate

2345_frame_MASTER.book Page 66 Wednesday, October 23, 2002 1:03 PM

66

Methods in Drug Abuse Research: Cellular and Circuit Level Analyses

FEP inflow tubing

From microsyringe pump

FEP outflow tubing

dialysis probe guide cannula

To collection vial

Nucleus accumbens

A10 VTA dialysis membrane

FIGURE 3.6 Schematic diagram of a sagittal section through the rat brain showing a microdialysis probe inserted into an implanted guide cannula. The microdialysis probe is oriented such that the tip has access to the extracellular compartment of the nucleus accumbens.

the volume of the tubing from the infusion syringe tip to the nucleus accumbens to be 3.6 ml (3 ¥ 1.2 ml), whereas the volume from the probe tip to the collecting vial is 6.6 ml [(3 ¥ 1.2 ml) + 3 ml]. Thus, the total volume of the perfusion system is 10.2 ml; at a flow rate of 1 ml/min, it will take approximately 10 min before the effects of the locally administered drug are first detected in the dialysate reaching the collecting vial.

3.5 ANALYTICAL METHODS The implanted dialysis probe affords access to the extracellular compartment for the purpose of collection of low molecular weight substances. In order for the specific neurotransmitter content of the sample to be determined, the chemical components must first be separated from one another and then quantified by suitable means. The separation phase is accomplished via HPLC: the dialysate sample is injected into the path of a fluid (the mobile phase) and forced, at high pressure, across a chromatographic column (the stationary phase). Chemical interaction between the stationary and mobile phases results in a variable elution of components from the injected sample; these components are then sent, in a particular temporal pattern, across an electrode within the EC detector. Oxidation of the compounds occurs as they cross the electrode; background current, as well as the current generated by each compound, is converted to a digital signal for subsequent computer analysis. While the analytical chemistry underlying the “chemical interaction” alluded to above is beyond the scope of this review, the relevant principles are relatively straightforward. Once a particular HPLC column is decided upon, adjustment of the components of the mobile phase allows the investigator to regulate the speed with which particular compounds move across the column relative to others. The goal is

2345_frame_MASTER.book Page 67 Wednesday, October 23, 2002 1:03 PM

Application of In Vivo Microdialysis Methods

67

to isolate the neurotransmitter species of interest so that its oxidation at the electrode is not accompanied by oxidation of concurrently eluting compounds. Once the components have been separated, the sample is sent across the electrode within the detector; the potential applied to the electrode is set to optimize the oxidation of the sample relative to background oxidation of electroactive species in the mobile phase, so that the most favorable signal-to-noise ratio can be obtained. Subsequent quantification of the neurotransmitters of interest is accomplished by the comparison between the amount of oxidation current generated by the unknown sample with that generated by neurotransmitter standards of known quantity. The details that follow describe our current approach to analyzing samples for DA, NE, and 5-HT. We have optimized this process in several important ways: 1) use of a microbore column and a 5-ml sample loop allows small samples to be analyzed, thereby enhancing the temporal resolution of our experiments; 2) use of a pulse-free syringe pump and the highest quality components for the mobile phase decrease the background signal generated at the electrode, thereby increasing the signal-to-noise ratio.

3.5.1 INSTRUMENTATION We have opted to use microbore HPLC with EC detection to assay monoamines in dialysate samples; the microbore methodology is optimally suited for handling the small sample volumes (5 to 20 ml) acquired during microdialysis experiments. Our system consists of an HPLC syringe pump, a sample injector, a microbore HPLC column, an EC detector, and a computerized chromatography workstation. The components of the system are connected together in a manner that minimizes dead space, and no unnecessary fittings or connectors are present. Figure 3.7 shows how the various components of the HPLC-EC system are arranged with respect to the in vivo microdialysis perfusion set up.

syringe

microsyringe activity pump monitor

HPLC pump computer monitor

inflow tubing

Injector & column

dialysis chamber

EC detector

FIGURE 3.7 Depiction of a typical in vivo microdialysis workstation. The components of the work station include: microdialysis instrumentation, dialysis testing chamber, HPLC-EC instrumentation, and the computer used for data acquisition and analysis of the electrochemical signal.

2345_frame_MASTER.book Page 68 Wednesday, October 23, 2002 1:03 PM

68

Methods in Drug Abuse Research: Cellular and Circuit Level Analyses

3.5.1.1 HPLC Pump The HPLC pump is an ISCO 260D syringe pump (cat. #68–1240–021, ISCO Inc., Lincoln, NE) that delivers mobile phase in a pulse-free manner. With classical dualpiston HPLC pumps, reciprocating pump parts result in oscillating background “noise.” The feature of pulse-free solvent delivery afforded by the ISCO 260D dramatically decreases baseline noise in the EC detector output, affording better resolution of sample peaks. Additionally, the 260D is capable of pumping mobile phase at very low flow rates with great accuracy. Several components are needed to effectively use the 260D pump for microbore HPLC applications: a 260D pump module with computer controller (cat. #67–1240–520), an outlet valve package (cat. #68–1247–078), and a manual refill kit (cat. #68–1247–077). All aspects of 260D operation are carried out via a computer control module, and a single module can run up to three 260D pumps. The outlet valve package is needed to connect the 260D to the injector/column of the HPLC system. The refill kit is necessary for connecting the pump to a mobile-phase reservoir. For our HPLC applications, mobile phase is delivered at a flowrate of 60 ml/min, maintaining a backpressure of 1000 to 1500 psi. There are some procedural modifications that must be kept in mind when using the ISCO 260D pump for microbore HPLC. First, because mobile phase cannot be recycled using the syringe pump, the pump must be refilled with new mobile phase every few days. This is accomplished using the hardware supplied with the outlet valve package and manual refill kits. Second, due to the low flow rates used with microbore HPLC (< 100 ml/min), more time must be allowed for the HPLC system to equilibrate when the pump flow rate is decreased, then increased, during refilling of the pump. Third, clogging of the narrow bore columns can be a frequent problem resulting in very high backpressures; the mobile phase and injected samples must therefore be kept absolutely clean and free of debris or impurities. Finally, while the small sample size and low flow rates enhance the temporal resolution of the experiments, great care must be taken to reduce dead space and to avoid fluid leaks in the system, both of which would compromise results. 3.5.1.2 Flow Cell Compartment Figure 3.8 is a schematic diagram of the specific LC components arranged in a commercially available flow cell compartment (cat. #LC44–01000, CC-5, Bioanalytical Systems Inc. (BAS), West Lafayette, IN). The sample injector is a UniJet microbore valve obtained from BAS (cat. #MF-4161), which is a modified Rheodyne-style injector specifically designed for microbore HPLC. The UniJet valve is mounted directly on the CC-5 flowcell compartment. The purpose of the injector is to allow delivery of dialysate samples into the path of the column. While many investigators choose to use automated injector modules to deliver samples onto the HPLC column, we have chosen to manually inject our samples as they are acquired during a microdialysis experiment. The stator of the UniJet injector valve contains threaded openings, or ports, that allow connection of the tubing from the ISCO 260D, a sample loop, and the microbore column, as shown in Figure 3.8. Many of

2345_frame_MASTER.book Page 69 Wednesday, October 23, 2002 1:03 PM

Application of In Vivo Microdialysis Methods

from 260D HPLC pump injector knob

Ag/AgCl reference 5 µL sample loop

1 needle port

ultrathin gasket

2 glassy carbon working electrode

3 4 UniJet Injector valve

69

injector ports

microbore column

auxiliary electrode

outflow to waste

FIGURE 3.8 Schematic diagram of the contents of a flow cell compartment (CC-5 available from BAS). The components include injector valve, microbore HPLC column, auxiliary electrode with reference cell, and working electrode. The dead space of the system has been greatly reduced by the direct connection of the column from the injector to the auxiliary electrode.

the components of the UniJet injector are made of inert polyetheretherketone (PEEK) plastic rather than metal to circumvent grounding problems. The injector also incorporates low-dispersion characteristics that are desirable for microbore applications. Mobile phase from the 260D pump enters the injector at port #2 and exits toward the column from port #3. A 5-ml loop is connected to injector ports #1 and #4. Dialysate samples are injected through a needle port in the front of the valve using Hamilton microsyringes with 22-gauge ¥ 2-in.-long square-tipped needles (Hamilton Co., Reno, NV). Samples are injected when the injector valve is oriented in the “load” position. Under these circumstances the loop is not connected to the path of flow of the mobile phase, and loading of the sample simply fills the 5-ml loop. To ensure accurate filling, the sample volume injected should be at least twice the volume of the loop, such that 10 ml of sample should be injected for a 5-ml loop. Once the loop is filled with dialysate sample, the knob on the UniJet injector valve is rotated 60° to the “inject” position, which opens the loop into the path of mobile phase flowing toward the column. The sample is quickly washed into the column for subsequent separation. The microbore column is a 1-mm ID ¥ 100-mm length UniJet microbore column from BAS (cat. #MF-8902, BAS). It is noteworthy that the column is directly connected to port #3 of the injector to minimize dead space, as depicted in Figure 3.8. A PEEK fitting is used to connect the column to the injector. The reversedphase column is packed with a C-18 matrix composed of 5-mM particles with 100Å pore size. We have found that 5-mm particles are superior to 3-mM particles as a microbore column matrix because the larger particles tend to resist blockage more readily. As noted above, clogging of the microbore columns can become a significant problem if one is not careful about keeping mobile phase and injected samples absolutely clean. Before the UniJet columns can be used with mobile phase, they

2345_frame_MASTER.book Page 70 Wednesday, October 23, 2002 1:03 PM

70

Methods in Drug Abuse Research: Cellular and Circuit Level Analyses

must be “wetted.” To accomplish the wetting procedure, 100% acetonitrile is pumped at 100 ml/min through the column for at least 15 min. This is followed by running a mixture of 40% acetonitrile:60% water through the column at 100 ml/min for an additional 15 min. The column should not be connected to the EC detector until after the wetting procedure has been completed. The UniJet column is connected directly to a radial-flow auxiliary electrode block (part #MF-1091, BAS) that houses a Ag/AgCl reference cell, as shown in Figure 3.8. A glassy carbon working electrode (cat. #MF-1095, BAS) is apposed to the auxiliary electrode, and an ultrathin gasket (cat. #MF-1068, BAS) separates the auxiliary electrode surface from the working electrode surface. The reference cell serves as a floating ground for internal comparison with the working electrode. The radial-flow characteristics of the auxiliary electrode cause the mobile phase and constituents of the injected sample from the HPLC column to spread out radially, which increases the available surface area for oxidation/reduction reactions to occur. Likewise, surface area for reaction between the mobile phase and the working electrode is increased as the liquid phase is forced into a thin film via the presence of the ultra-thin gasket. Electroactive analytes, such as catecholamines (NE and DA) and indoleamines (5-HT), generate small current signals as they are eluted from the column and move across the electrode surface. The amount of current is directly proportional to the number of molecules present at the electrode surface. After passing over the working electrode, the liquid phase travels via an exit port to a waste container. 3.5.1.3 Electrochemical Detector The auxiliary electrode, the reference cell, and the working electrode are connected via cables to an LC-4C amperometric detector made by BAS (cat. #EF-1015). The LC-4C detects the current fluctuations produced at the working electrode surface by molecules of electroactive solutes in the dialysate sample, and these fluctuations are amplified and sent to a recording device. We have found that the optimal settings for the detector are as follows: applied potential = +650 mV, signal filter = 0.1 Hz, and range = 1.0 nA. A detailed discussion of the theory of EC detection is beyond the scope of this review, but the interested reader is referred to BAS technical manuals for further information. The analog output from the LC-4C detector can be routed to either a chart recorder or a computerized data-acquisition system. In our laboratory, the LC-4C output is routed to an analog-to-digital converter (cat. #WAT 073645, SATIN interface, Waters Corporation, Milford, MA) that is connected to an IBMbased PC loaded with Millenium M32 chromatographic software (cat. #WAT 042864, Waters).

3.5.2 MOBILE PHASE CONDITIONS Using the HPLC-EC instrumentation described above, it is possible to optimize conditions for the detection of sub-pg amounts of NE, DA, and 5-HT simply by changing the composition of the mobile phase. All of the mobile-phase mixtures that we use are composed of three basic constituents: a buffering salt (e.g., monochlo-

2345_frame_MASTER.book Page 71 Wednesday, October 23, 2002 1:03 PM

Application of In Vivo Microdialysis Methods

71

roacetic acid), an ion-pairing agent (e.g., sodium octyl sulfate), and an organic modifier (e.g., methanol or acetonitrile). By making simple changes in the mobilephase composition, predictable changes in the elution pattern of monoamines from the HPLC column can be realized. For example, changing the concentration of ionpairing agent in the mobile phase selectively affects the time it takes for monoamines to elute from the HPLC column (i.e., retention time); increasing the amount of ionpairing agent retains monoamines on the column better, thereby slowing elution of these compounds and increasing their retention times. Described below are two types of mobile phases used in our laboratory. All recipes use highest-quality ultrapure salts (Sigma), chromatography-grade organic solvents, and double-distilled highresistance water. For most studies, we employ a mobile phase that is optimized for the detection of DA and 5-HT. With this mobile phase the retention times for DA and 5-HT are approximately 5 min and 12 min, respectively. The mobile phase consists of 215 mM disodium EDTA, 145 mM sodium hydroxide, 150 mM monochloroacetic acid, 1.49 mM sodium octyl sulfate (SOS), 10 mM triethylamine, 6% methanol, 6% acetonitrile, at a final pH of 5.3.18 Prepare the mobile phase as follows: in a 1-L volumetric flask, add 80 mg disodium EDTA, 5.8 g sodium hydroxide, 14.2 g monochloroacetic acid, and 345 mg SOS to 500 ml distilled water with stirring. When salts are dissolved, add 1 ml triethylamine, 60 ml methanol, and 60 ml acetonitrile while stirring. The volume is filled with water to 900 ml, and the pH is adjusted to 5.3 with sodium hydroxide or glacial acetic acid. The volume is adjusted to 1 L. The mobile phase is filtered (0.22 mM) under vacuum and degassed for 30 min. We have invested considerable time and effort into developing a mobile phase that can be used to detect NE, DA, and 5-HT in a single dialysate sample. This task presents significant technical challenges because NE comes off the HPLC column quickly (i.e., < 3 min), whereas 5-HT elutes rather slowly (i.e., > 15 min). In addition, mobile-phase conditions must be adjusted frequently to avoid a number of coeluting peaks. Nevertheless, we have achieved some success using a mobile phase that consists of 27 mM disodium EDTA, 75 mM sodium phosphate monobasic, 347 mM sodium dodecyl sulfate (SDS), 1 mM triethylamine, 11% methanol, 11% acetonitrile, at a final pH of 5.6.19 Prepare the mobile phase as follows: in a 1-L volumetric flask, add 10 mg disodium EDTA, 9.0 grams sodium phosphate monobasic, and 100 mg sodium dodecyl sulfate to 500 ml distilled water with stirring. When salts are dissolved, add 100 ml triethylamine, 110 ml acetonitrile, and 110 ml methanol. The volume is brought to 900 ml with water, and the pH is adjusted to 5.6 with phosphoric acid. The final volume is adjusted to 1 L. The mobile phase is filtered (0.22 mM) under vacuum and degassed for 30 min.

3.5.3 DATA ACQUISITION

AND

ANALYSIS

The electrical signals produced at the working electrode can be sent to a recording device such as a sensitive strip-chart recorder. An output port on the EC detector is connected to the device, and the EC signal is converted to a continuous graphical representation of the baseline signal and the deviations from baseline produced by oxidation of electroactive substances in an injected sample. Thus, the signals gen-

2345_frame_MASTER.book Page 72 Wednesday, October 23, 2002 1:03 PM

72

Methods in Drug Abuse Research: Cellular and Circuit Level Analyses

erated by the EC detector are viewed as a series of peaks on the chart recorder; the graphical representation of the peaks is known as a chromatogram. The height, or area-under-the-curve (AUC), of each peak is proportional to the amount of the corresponding electroactive substance passing across the electrode surface. Determination of the neurotransmitter concentration in a dialysate sample is accomplished by comparison to a neurotransmitter standard; the standards have a known concentration and a well-established retention time. When the peak AUC of a neurotransmitter standard is compared with the peak AUC of an electroactive substance in a sample with the same retention time, the amount of neurotransmitter in the sample can be determined with great accuracy. As is true in other areas of science, the practical aspects of acquiring and analyzing data has benefited greatly from the evolution and development of computer-based data-handling systems. The use of computerized chromatographic data-analysis systems is relatively straightforward, and there are two basic tasks performed by such systems — data acquisition and data processing. For data acquisition, the analog signal generated by the EC detector is converted into a digital signal using some type of analog-todigital (A/D) converter module. The converted digital signal can be processed in real time or stored for later analysis. For data processing, the computer is used to convert peak AUCs, generated by signals from the EC detector, into actual neurotransmitter concentrations. The application of computerized data processing has been an important advance over the earlier chart-based approach; the digital representation affords greater resolution than that offered by mechanical chart recorders. Moreover, the newest chromatographic software systems allow the investigator to process data in a more sophisticated manner. For example, the computer can be used to integrate peaks from previously stored data (i.e., data processing) while simultaneously acquiring incoming data in real time from the EC detector (i.e., data acquisition). Computer software can be used to control instrumentation such as autoinjectors and HPLC pumps. As mentioned previously, we currently use the Millennium Chromatography Manager (Waters) for data handling. This system includes the hardware interface that mediates A/D conversion of the working electrode signal as well as the software package that allows the investigator to customize user-defined criteria for data analysis, file management, and report generation. This package is incredibly flexible and allows the investigator to utilize a number of different approaches to data handling. A full description of the Millennium software package is beyond the scope of this review, and the interested reader is referred to the Waters technical manuals for detailed information. 3.5.3.1 Standard Curves Prior to the actual analysis of dialysate samples from the CNS, the elution profile (i.e., retention time) and detection capacity (i.e., detector sensitivity) for neurotransmitter standards must be ascertained. When initially developing an analytical method for the detection of neurotransmitters in standards or samples, three specific aims must be kept in mind. The first aim is to guarantee, within technical limits, that the retention time for the specific neurotransmitter of interest is isolated from the elution time of any other possible electroactive compounds (i.e., coeluting peaks). The

2345_frame_MASTER.book Page 73 Wednesday, October 23, 2002 1:03 PM

Application of In Vivo Microdialysis Methods

73

second goal is to ensure that the electrochemical signal from the neurotransmitter is detectable in the range expected in a dialysate sample collected under baseline conditions (i.e., sub-pg amounts). Finally, it is essential that the electrochemical signal generated by the neurotransmitter is linear over a range of concentrations expected in the dialysate sample, or about 0.1 to 100 pg. In our laboratory, we are most interested in examining DA and 5-HT neurotransmission in the rat nucleus accumbens. Therefore, we will describe a method for generating standard curves for DA and 5-HT, with both transmitters detected within the same chromatographic run. The HPLC-EC setup and mobile-phase conditions are as described in Section 3.5, Analytical Methods. To establish the retention times and detectability for DA and 5-HT, singlecomponent standard solutions must be prepared, injected, and analyzed. Standards should be prepared for DA and 5-HT as well as for any possible electroactive compounds that might be found in dialysate samples. We prepare standard stock solutions for 3,4-dihydroxyphenylacetic acid (DOPAC), DA, 5-hydroxyindoleacetic acid (5-HIAA), homovanillic acid (HVA) and 5-HT. It should be mentioned that peaks for DOPAC, HVA, and 5-HIAA are not usually visible on the chromatograms using our mobile-phase conditions. However, the standards for these compounds are included during method development to insure that these compounds are not interfering with the peaks for DA and 5-HT in the chromatograms. To prepare the standard stock solutions a few milligrams of standard powder are weighed on an analytical balance and transferred to a borosilicate glass test tube. The powder is dissolved in 0.1 N perchloric acid to yield a final concentration of 1 mg/ml. Standard stocks for DOPAC, DA, 5-HIAA, HVA, and 5-HT are prepared monthly and stored in the refrigerator. Because monoamines and their metabolites are sensitive to light, temperature, and pH, fresh working standards for injection are prepared daily and kept on ice. Each working standard solution is prepared by adding a 10-ml aliquot of stock standard to 990 ml of aCSF perfusion medium to yield a concentration of 10 mg/ml, or 10 ng/ml. Aliquots of this solution are serially diluted to yield standards of 0.1 pg/ml, 0.3 pg/ml, and 1 pg/ml. It is convenient to use the units of pg/ml when expressing transmitter concentration because 5 ml of solution is injected onto the HPLC column (the injection loop is 5 ml) and picogram amounts of transmitter are typically detected in a dialysate sample. The single-component standards are injected and analyzed. The proper method for injecting solutions onto the HPLC column, via the UniJet valve, is described in Section 3.5, Analytical Methods. Once it is verified that the peaks for DA and 5-HT are isolated and detectable, monoamine mix standards are prepared and run. The monoamine mix standard contains all of the analytes mentioned above in the same solution and is prepared from the 1-mg/ml stock solutions. The working mix standards are prepared by adding 10-ml aliquots of stock solutions of DOPAC, DA, 5-HIAA, HVA, and 5-HT to 950 ml of aCSF perfusion medium in the same test tube to yield a final tube concentration of 10 ng/ml. Aliquots of this solution are serially diluted to yield mix standards of 0.1 pg/ml, 0.3 pg/ml, and 1 pg/ml. These standards are injected and analyzed. The power of the online data analysis becomes apparent at this stage of applications development. The Millennium software generates (with the appropriate prompting by the user) a record of retention time and AUC for each peak and will generate a

2345_frame_MASTER.book Page 74 Wednesday, October 23, 2002 1:03 PM

74

Methods in Drug Abuse Research: Cellular and Circuit Level Analyses

best-fit curve for a range of standards of varying concentration to allow determination of linearity. 3.5.3.2 Assaying Dialysate Samples In vivo microdialysis experiments can be carried out once the analytical method for generating standard curves for DA and 5-HT has been tested and verified. In most instances, a microdialysis probe is inserted into a previously implanted guide cannula the night before an experiment, and the probe is perfused with aCSF overnight at 1.0 ml/min. The next morning the FEP outflow line from the probe is placed into a collection vial atop the tether, and dialysate samples are collected at 20-min intervals. Prior to the injection of the first dialysate sample, however, several tasks must be completed. The first task is the creation of a sample set using the Millennium software program. A sample set is a sequential list of the standards and samples to be injected along with the specifications for detecting and quantifying DA and 5-HT. Next, an aliquot of aCSF perfusion medium is injected onto the HPLC column; the aCSF injection is an important way to verify that the baseline signal from the EC detector is stable (i.e., flat baseline) and that no interfering peaks are present in the aCSF chromatogram. Finally, the 0.1-, 0.3-, and 1.0-g/ml mix standards are injected onto the HPLC column and subsequently analyzed; these standards are injected in ascending order. Because 5 ml of each standard is injected, the total picogram amount for the three mix standards is 0.5, 1.5, and 5 pg. The computer is preprogrammed with the amounts of DA and 5-HT present in the standards, and this information is used to generate a standard curve of picogram injected vs. microvolt signal detected. After the last standard is injected, dialysate samples collected from the vial atop the tether can be injected. We typically inject dialysate samples onto the HPLC column immediately after they are collected, and three to four baseline samples are obtained prior to any pharmacological manipulation. Because of the power of the computerized software, the investigator can process the data from previous injections while acquiring data from the current injection. Identification of DA and 5-HT in the dialysate sample is based upon an accurate match with the retention time of the standards (this match can be made manually or automatically); determination of the concentration of DA and 5-HT is based upon extrapolation from the standard curve for that specific neurotransmitter. It should be noted that various problems are routinely encountered when analyzing neurotransmitters in dialysate samples. Some of the most common problems include noisy baseline, decreased sensitivity, and coeluting peaks. Noisy baseline signal from the EC detector can be caused by many factors. Trace impurities or tiny bubbles in the mobile phase are frequent causes. For this reason, it is imperative, as mentioned previously, that ultrapure reagents be used to prepare mobile-phase solutions, and these solutions should be filtered (0.22-mm filters) and degassed prior to use. Another cause of noisy baseline is improper seating of the reference electrode within the auxiliary electrode compartment. This problem can be corrected by simply adjusting the reference electrode to a more desirable position. Decreased sensitivity for a neurotransmitter of interest is usually due to the deposition of contaminants on the surface of the working electrode. This situation can be corrected by polishing

2345_frame_MASTER.book Page 75 Wednesday, October 23, 2002 1:03 PM

Application of In Vivo Microdialysis Methods

75

the glassy carbon working electrode, as described by the manufacturer. Finally, the presence of co-eluting peaks is a significant and serious problem that has many possible causes. The best way to remove coeluting peaks from the chromatogram is to eliminate the source of the electroactive compound creating the peak. In some instances, this is impossible. For example, the interfering peaks could be monoamine metabolites that are present in much greater amounts in the ECF than their active amine precursors. Careful alteration of the mobile-phase conditions can usually be used to “move” the neurotransmitter peaks of interest away from the coeluting peaks.

3.6 APPLICATIONS The use of illicit stimulants, especially cocaine and methamphetamine, is a persistent worldwide health problem, and few treatment options are available for patients suffering from stimulant dependence. In order to develop successful treatments for stimulant addictions, it is imperative to understand the neurobiological underpinnings of stimulant drug action. As described in the Introduction, in vivo microdialysis offers investigators a unique view of the CNS. In vitro preparations (e.g., binding and release assays) can help to elucidate isolated cellular effects of stimulant drugs, but they cannot directly reveal what these compounds do to neurotransmitter levels in the ECF. Likewise, post-mortem tissue analyses provide important information about drug-induced changes in neurotransmitter synthesis, storage, and metabolism, but these alterations often do not reflect the state of affairs in the extracellular compartment. Microdialysis, when used in conjunction with an analytical technique like HPLCEC, allows the investigator to examine the effect of stimulant drugs on basal and evoked neurotransmitter levels in the ECF of discrete regions of the CNS. Because the probe itself can be used as a drug-infusion device, the effects of stimulants on neurotransmission can be determined within the localized microenvironment surrounding the probe. The ability to repeatedly sample from the same animal makes microdialysis a powerful approach to studying the kinetics of drug-induced changes in neurotransmitters; knowledge of the time course of the response to drugs can be particularly useful for investigations of drug interactions. The adaptation of microdialysis to awake animals allows correlations to be made between brain chemistry and behavior. Finally, using sophisticated analytical methods to assess changes in more than one neurotransmitter system allows one to discern the relative contribution of specific neurotransmitters to the response profile of a particular drug.

3.6.1 ACUTE EFFECTS

OF

STIMULANTS

A large body of preclinical evidence shows that DA neurons in the CNS are involved in drug addiction.20 Figure 3.9 depicts the major ascending DA neuron systems in rat brain. The mesolimbic DA system in particular has been implicated in mediating the addictive properties of abused drugs including psychomotor stimulants.21 The mesolimbic DA system in rats consists of cell bodies residing in the ventral tegmental area (VTA) that send axonal projections to numerous limbic forebrain regions, most notably the nucleus accumbens. Due to the importance of the nucleus accumbens

2345_frame_MASTER.book Page 76 Wednesday, October 23, 2002 1:03 PM

76

Methods in Drug Abuse Research: Cellular and Circuit Level Analyses

mPFC Caud N A9 SN N Acc VTA

A10 Olf Tub

FIGURE 3.9 Depiction of a sagittal section through the rat brain showing dopaminergic projections from cell bodies in the midbrain, substantia nigra (SN) and ventral tegmental area (VTA), to forebrain innervation sites including the caudate nucleus (Caud N), medial prefrontal cortex (mPFC), nucleus accumbens (N Acc), and olfactory tubercle (Olf Tub).

in mediating the effects of stimulants, many investigators have focused their attention on performing in vivo microdialysis sampling in this brain region. Early microdialysis studies provided important information about the acute neurochemical effects of stimulants like cocaine and amphetamines. A central theme that arose from this work is the discovery that diverse types of stimulant drugs elevate extracellular concentrations of DA in rat nucleus accumbens.22–24 Recent advances in molecular biology have shown that stimulant drugs exert their effects on DA cells by interacting with DA transporter (DAT) proteins located on nerve cell membranes. In cell lines expressing DAT, cocaine blocks DA uptake into cells by binding to DAT, whereas methamphetamine releases DA from cells by serving as a substrate for DAT.25,26 Thus, both cocaine (a DA uptake blocker) and methamphetamine (a DA releaser) produce elevations in extracellular DA via DAT-dependent mechanisms. The data illustrated in Figure 3.10 show the results of a typical experisaline 0.3 mg/kg 1.0 mg/kg 3.0 mg/kg

DA (% baseline)

800

* 600

*

400

*

*

200 0 –40 –20

0

20

40

60

80

Time (min)

FIGURE 3.10 Effects of acute administration of i.v. cocaine on extracellular DA in the nucleus accumbens of awake rats. Baseline levels of dialysate DA were 1.65 ± 0.16 nM. Data are mean ± S.E.M. for n = 5 rats/group, expressed as percentage of baseline. * = P < 0.05 with respect to saline-injected control group. (Data from Baumann, M.H., Char, G.U., DeCosta, B.R., Rice, K.C., and Rothman, R.B., J. Pharmacol. Exp. Ther., 271, 1216, 1994.)

2345_frame_MASTER.book Page 77 Wednesday, October 23, 2002 1:03 PM

Application of In Vivo Microdialysis Methods

77

DA (% baseline)

* 1200 1.0 mg/kg

900 600

*

0.3 mg/kg

*

0.1 mg/kg

300

*

* *

*

0

5-HT (% baseline)

ment where acute i.v. cocaine administration produces dose-related increases in dialysate DA levels in the nucleus accumbens of rats.27 While many investigators have shown that stimulants increase dialysate DA in the brain, more elegant experiments have combined in vivo microdialysis with cocaine self-administration to show that extracellular DA concentration is an important signal maintaining stimulant selfadministration behavior.28,29 The effect of stimulant drugs is, of course, not confined to the nucleus accumbens; alterations in DA neurotransmission in the caudate nucleus may underlie the behavioral activation that accompanies stimulant use, while drug-induced changes in the frontal cortex are likely to influence the complex interplay between cognitive processes and addiction (see Figure 3.9). Furthermore, stimulant drugs are not selective for DA neurons — these agents affect NE and 5-HT neurotransmission concurrently with their effects on the DA system. A number of in vivo microdialysis studies have shown that administration of cocaine or methamphetamine causes simultaneous elevation of extracellular levels of DA, NE, and 5-HT in diverse brain areas.18,19,30–32 Like the effects of stimulants on DA neurons, these agents increase extracellular NE and 5-HT via interactions with NE transporters (NET) and 5-HT transporters (SERT), respectively.25,33 The importance of non-DA neurotransmitters in mediating the effects of stimulants is attracting more attention as new data are reported in the literature. Development of analytical methods to assay multiple neurotransmitters in a single dialysate sample will undoubtedly reveal critical information about the interaction between the various monoamines and other neurotransmitters. As mentioned in Section 3.5, Analytical Methods, we have developed HPLCEC methods to simultaneously measure DA and 5-HT in dialysate samples. The data depicted in Figure 3.11 demonstrate that acute i.v. administration of methamphetamine increases extracellular levels of both DA and 5-HT in the nucleus accumbens of rats. It is noteworthy that the effects of methamphetamine on DA and 5-HT are nearly equivalent. This observation is interesting because in vitro investigations have reported that methamphetamine is much more potent as a releaser of DA when compared to 5-HT.25,33 Thus, the in vitro profile of activity of a given stimulant may

1200 1.0 mg/kg

900 600

* *

0.3 mg/kg 0.1 mg/kg

*

300

* *

0 –60 –20

20

60

100 140 180

Time (min)

–60 –20

20

60

100 140 180

Time (min)

FIGURE 3.11 Effects of i.v. methamphetamine on extracellular DA (left) and 5-HT (right) in the rat nucleus accumbens. Doses of 0.1, 0.3, and 1.0 mg/kg were injected at 0, 60, and 120 min, respectively. Data are mean ± S.E.M. for n = 6 rats, expressed as percentage of baseline. Baseline levels of dialysate DA and 5-HT were 1.86 ± 0.37 and 0.23 ± 0.02 nM, respectively. * = P < 0.05 with respect to preinjection control levels.

2345_frame_MASTER.book Page 78 Wednesday, October 23, 2002 1:03 PM

78

Methods in Drug Abuse Research: Cellular and Circuit Level Analyses

not necessarily predict the neurochemical effects of the drug in vivo. Another important feature of the results depicted in Figure 3.11 is the use of microdialysis to determine dose–response relationships in the same individual. By collecting samples at baseline and after multiple doses of methamphetamine, a dose–effect relationship between drug dose and dialysate DA and 5-HT is generated for each individual subject. This type of experimental design is quite powerful in terms of statistical analysis and helps to greatly reduce the number of subjects required to determine the acute actions of drugs.

3.6.2 CHRONIC EFFECTS

OF

STIMULANTS

It is well accepted that chronic use of drugs like stimulants, opiates, and alcohol produces neuroadaptive changes in the brain.34 Such alterations may be involved in the appearance of mood disturbances following cessation of drug use (i.e., withdrawal). Deficiencies in the normal functioning of monoamine neurons during withdrawal from stimulant drugs, for example, has been proposed to underlie the major depressive-like symptoms associated with the “crash” phase of stimulant withdrawal — a period of extreme dysphoria that is experienced upon abrupt termination of stimulant use.35,36 In vivo microdialysis methods have provided valuable information about the neurochemistry of stimulant drug withdrawal. Numerous reports in the literature demonstrate that withdrawal from repeated cocaine administration in rats leads to reduced basal levels of extracellular DA in mesolimbic reward pathways.37–39 In one study, Parson et al.38 used quantitative microdialysis methods to estimate the “true” extracellular concentration of DA in rat nucleus accumbens, and this index of DA transmission was significantly decreased during cocaine withdrawal. Other investigators have performed in vivo microdialysis sampling in rats trained to self-administer drugs; these studies reveal that abrupt cessation of cocaine self-administration is accompanied by significant reductions in basal levels of dialysate DA and 5HT.40,41 A dual deficit in synaptic DA and 5-HT may contribute significantly to the broad spectrum of withdrawal symptoms in human stimulant addicts who attempt to stop using drugs. Substantial evidence indicates that repeated high-dose administration of methamphetamine causes prolonged loss of tissue DA and decreased DAT binding, in the brain; these DA deficits have been interpreted to reflect drug-induced neurotoxicity.42 In vivo microdialysis has been used to examine functional correlates of DA neurotoxicity in rats previously exposed to methamphetamine.43–45 In one study, Cass44 showed that 7 d after exposure to high-dose methamphetamine (5 mg/kg, s.c, four doses at 2-h intervals), rats exhibit reduced basal levels of extracellular DA in the caudate nucleus. Moreover, the ability of local amphetamine infusion to evoke caudate DA release is significantly blunted in methamphetamine-treated rats. Interestingly, methamphetamine produced no change in basal dialysate DA levels, or evoked DA release, in the nucleus accumbens. Subsequent microdialysis studies demonstrated that DA neurons in the caudate nucleus slowly recover normal functioning by 1 month after methamphetamine treatment.45 Collectively, such studies show that in vivo microdialysis is a powerful approach for identifying regional

2345_frame_MASTER.book Page 79 Wednesday, October 23, 2002 1:03 PM

Application of In Vivo Microdialysis Methods

200

PRETREATMENT Saline Methamphetamine

5-HT (% baseline)

DA (% baseline)

300

79

d-FEN 1.0 mg/kg

d-FEN 0.3 mg/kg

100

0 –40

0

40

80

120

1800 d-FEN 1.0 mg/kg

1200 d-FEN 0.3 mg/kg

*

600

*

*

*

*

0 –40

Time (min)

0

40

80

120

Time (min)

FIGURE 3.12 Effects of acute administration of i.v. fenfluramine on extracellular DA (left) and 5-HT (right) in rat nucleus accumbens during withdrawal from repeated injections of saline or methamphetamine (10 mg/kg, i.p., b.i.d., 4-d). Data are mean ± S.E.M. for n = 5 rats/group, expressed as percentage of baseline. Baseline dialysate DA levels for saline and methamphetamine groups were 2.03 ± 0.36 and 1.77 ± 0.21 nM. Baseline dialysate 5-HT levels for saline and methamphetamine groups were 0.18 ± 0.02 and 0.26 ± 0.05 nM. * = P < 0.05 with respect to saline-treated group.

differences in vulnerability to neurotoxins and for examining the time course of recovery from neurotoxic insult. In an attempt to examine possible neuroadaptive changes in 5-HT transmission produced by high-dose methamphetamine, we utilized in vivo microdialysis in rat nucleus accumbens to evaluate neurotransmitter changes following repeated methamphetamine administration (Figure 3.12). In this experiment, either saline or methamphetamine (10 mg/kg, i.p., b.i.d.) was administered for 4 d, and rats were subjected to microdialysis testing 3 to 4 d later. Based on the findings of others,46 we hypothesized that high-dose methamphetamine would impair the ability of the 5-HT releaser, fenfluramine, to evoke 5-HT release. Previous experiments like those shown in Figures 3.10 and 3.11 helped us determine the appropriate doses of fenfluramine to use in our challenge paradigm.47,48 The results depicted in Figure 3.12 show that acute i.v. fenfluramine injections produce dose-related increases in extracellular 5HT, but not DA, in rat nucleus accumbens. More importantly, fenfluramine-evoked 5-HT release is severely impaired in rats previously exposed to methamphetamine. It should be mentioned that fenfluramine-induced 5-HT responses return to normal by 2 weeks after methamphetamine exposure, indicating rapid recovery of 5-HT function after methamphetamine treatment.

3.6.3 IMPLICATIONS

FOR

MEDICATIONS DEVELOPMENT

The ultimate goal for many of our in vivo microdialysis investigations is twofold: (1) to contribute to an understanding of the mechanisms by which stimulant drugs exert their effects (as presented above) and (2) to explore the development of potential medications to treat stimulant addictions. The crucial role of DAT proteins in mediating the addictive properties of stimulant drugs has prompted investigators to examine DAT ligands, like the DA uptake blocker GBR12909 (1-[2-[bis(4-fluo-

2345_frame_MASTER.book Page 80 Wednesday, October 23, 2002 1:03 PM

80

Methods in Drug Abuse Research: Cellular and Circuit Level Analyses

rophenyl)methoxy]ethyl]-4-(3-phenylpropyl) piperazine), as potential therapeutic agents and as tools to test the DA hypothesis of addiction.18,27,49 GBR12909 displays a number of unique properties that are consistent with its use as a pharmacotherapy for stimulant dependence. At the molecular level, GBR12909 binds with high affinity and selectivity to DAT proteins, thereby blocking DA uptake.50 GBR12909 exhibits persistent “pseudoirreversible” binding that is characterized by slow dissociation of the drug from its binding site.51 At the organismic level, GBR12909 penetrates into the brain very slowly upon systemic administration.52 In vivo microdialysis studies show that GBR12909 produces elevations in extracellular DA in rat brain that are slow in onset, modest in magnitude, and sustained in duration.27,51 Because GBR12909 binds to DAT sites, pretreatment with the drug antagonizes amphetamineinduced release of DA in vivo.27 Perhaps more importantly, GBR12909 suppresses cocaine self-administration in animals and appears to be less reinforcing than abused stimulants.53,54 Collectively, the preclinical data suggest that GBR12909 might be a useful pharmacological adjunct in the management of stimulant addictions. There are numerous impediments to successful treatment of stimulant-dependent patients, including high dropout rates and lack of compliance with medication regimens. In order to circumvent issues related to patient noncompliance, investigators have developed a long-acting depot formulation of GBR12909, GBR-decanoate (1[2-[bis(4-fluorophenyl)methoxy]ethyl]-4-(3-hydroxy-3-phenylpropyl) piperazyl decanoate), which can be administered as a single i.m. injection.55 In a preliminary report, Glowa et al.56 showed that one dose of GBR-decanoate suppresses cocaine selfadministration behavior in rhesus monkeys for up to 1 month. Like other depot preparations, GBR-decanoate is an oil-soluble prodrug that slowly releases an active hydroxylated derivative, GBR-hydroxy (1-[2-[bis(4-fluorophenyl)methoxy]ethyl]-4(3-hydroxy-3-phenylpropyl) piperazine) into the circulation. Like GBR12909, GBRhydroxy displays high affinity for DAT (Ki = 2.2 ± 0.1 nM) and low affinity for SERT (Ki = 117 ± 7 nM).55 Based on the in vitro findings with GBR-hydroxy, we wished to examine the effects of GBR-hydroxy on extracellular DA and 5-HT levels in vivo. The data in Figure 3.13 show that i.v. administration of GBR-hydroxy produces slow-onset elevations in extracellular DA, but not 5-HT, in the rat nucleus accumbens. The sustained increase in dialysate DA levels produced by GBR-hydroxy is consistent with the pseudoirreversible binding of the drug to DAT sites in the brain. Because GBR-hydroxy is the active compound generated after i.m. injection of GBR-decanoate, it was of interest to test the ability of GBR-hydroxy to influence methamphetamine-induced DA and 5-HT release. In vitro studies have shown that GBR12909 can block methamphetamine-induced DA release, presumably via binding to DAT sites.33 Thus, we postulated that pretreatment of rats with GBR-hydroxy might antagonize methamphetamine-evoked DA release in vivo. Figure 3.14 shows that acute pretreatment with i.v. GBR-hydroxy (0.5 mg/kg) significantly reduces the DA-releasing capability of i.v. methamphetamine (0.5 mg/kg).18 In contrast, GBRhydroxy has no effect on 5-HT release induced by methamphetamine. Based on the promising results with GBR-hydroxy, we sought to evaluate the potential utility of GBR-decanoate as a long-acting medication for METH dependence. To achieve this aim, we examined methamphetamine-evoked release of DA and 5-HT using in vivo microdialysis in rats that had previously received injection

2345_frame_MASTER.book Page 81 Wednesday, October 23, 2002 1:03 PM

400

*

*

*

*

*

300

*

*

200

*

*

*

*

100 0 –40 –20 0

5-HT (% baseline)

DA (% baseline)

Application of In Vivo Microdialysis Methods

81

400

saline 0.3 mg/kg 1.0 mg/kg

300 200 100 0

20 40 60 80 100 120

–40 –20 0

20 40 60 80 100 120

Time (min)

Time (min)

FIGURE 3.13 Effects of acute administration of i.v. GBR-hydroxy on extracellular DA (left) and 5-HT (right) in rat nucleus accumbens. Data are mean ± S.E.M. for n = 5 rats/group, expressed as percentage of baseline. Baseline levels of dialysate DA and 5-HT were 2.50 ± 0.31 and 0.37 ± 0.07 nM, respectively. * = P < 0.05 with respect to saline-treated group. (Modified from Baumann, M.H., Ayestas, M.A., Sharpe, L.G., Lewis, D.B., Rice, K.C., and Rothman, R.B., J. Pharmacol. Exp. Ther., 301, 1190, 2002.)

5-HT (% baseline)

DA (% baseline)

800

PRETREATMENT saline GBR-OH METH 0.5 mg/kg

600 400

* *

200 0 –60

–20

20

60

Time (min)

100

140

400 METH 0.5 mg/kg

300 200 100 0 –60

–20

20

60

100

140

Time (min)

FIGURE 3.14 Effects of GBR-hydroxy (GBR-OH) on METH-induced DA release (left) and 5-HT release (right) in rat nucleus accumbens. Rats were pretreated with i.v. GBR-OH (0.5 mg/kg) or vehicle at time 0, followed by acute i.v. injection of 0.5 mg/kg METH or saline at 60 min. Data are mean ± S.E.M. for n = 5 rats/group, expressed as percentage of baseline. Baseline levels of dialysate DA and 5-HT were 2.22 ± 0.18 and 0.37 ± 0.11 nM, respectively. * = P < 0.05 with respect to saline-treated group that received METH. (Modified from Baumann, M.H., Ayestas, M.A., Sharpe, L.G., Lewis, D.B., Rice, K.C., and Rothman, R.B., J. Pharmacol. Exp. Ther., 301, 1190, 2002.)

of GBR-decanoate. In these experiments, rats received single i.m. injections of GBRdecanoate (1 ml/kg of a 48% solution, or 480 mg/kg) or its sesame oil vehicle, and groups of rats were tested weeks later. When tested 2 weeks after i.m. injection, rats pretreated with GBR-decanoate displayed significant 2.4-fold elevations in basal extracellular DA (3.95 ± 0.64 nM; n = 14) compared to oil-treated controls (1.69 ± 0.27 nM; n = 14). As depicted in Figure 3.15, GBR-decanoate significantly reduced DA release produced by methamphetamine at 0.3-mg/kg and 1.0-mg/kg challenge doses. GBR-decanoate did not alter baseline levels of extracellular 5-HT and failed to alter METH-induced increases in 5-HT. The collective findings with GBRdecanoate show that this agent is capable of producing sustained elevations in baseline extracellular DA and persistent blockade of methampetamine-evoked DA

2345_frame_MASTER.book Page 82 Wednesday, October 23, 2002 1:03 PM

1000

PRETREATMENT vehicle GBR-dec

800 600 400

METH 1.0 mg/kg METH 0.3 mg/kg

*

200

*

*

*

0 –40 –20 0

20 40 60 80 100 120

Time (min)

5-HT (% baseline)

Methods in Drug Abuse Research: Cellular and Circuit Level Analyses

DA (% baseline)

82

1000 800 600 400

METH 1.0 mg/kg METH 0.3 mg/kg

200 0 –40 –20 0

20 40 60 80 100 120

Time (min)

FIGURE 3.15 Effects of GBR-decanoate (GBR-dec) on METH-induced DA release (left) and 5-HT release (right) in rat nucleus accumbens: 2-week test. Rats were treated with single i.m. injection of GBR-dec (1 ml/kg of a 48% solution, or 480 mg/kg) or sesame oil vehicle. Two weeks later, acute i.v. doses of 0.3 and 1.0 mg/kg METH were injected at 0 and 60 min, respectively. Data are mean ± S.E.M. for n = 6 to 7 rats/group, expressed as percentage of baseline. Baseline dialysate DA levels for vehicle and GBR-dec groups were 1.69 ± 0.17 and 3.95 ± 0.64 nM. Baseline dialysate 5-HT levels for vehicle and GBR-dec groups were 0.33 ± 0.05 and 0.41 ± 0.09 nM. * = P < 0.05 with respect to vehicle-pretreated group that received METH. (Modified from Baumann, M.H., Ayestas, M.A., Sharpe, L.G., Lewis, D.B., Rice, K.C., and Rothman, R.B., J. Pharmacol. Exp. Ther., 301, 1190, 2002.)

release. The potential role of GBR-type compounds in the treatment of stimulant addiction will require careful testing of safety and efficacy in human subjects.

3.7 CONCLUSIONS In vivo microdialysis offers a unique view into the intact functioning brain. This chapter has described the basic procedures involved in performing microdialysis sampling and the analytical methods necessary to assay the samples collected. Microdialysis has been used to glean critical information about the actions of abused stimulants on monoamine neurons in the CNS. Creative use of the technique, such as application of quantitative approaches and combination with drug self-administration behavior, has advanced our understanding of the neurobiological effects of cocaine during prolonged exposure and during withdrawal. In vivo microdialysis has been instrumental in establishing functional correlates of methamphetamine neurotoxicity. Despite the great strides in elucidating the mechanisms of stimulant drug action, the scourge of stimulant abuse remains a significant problem. It is hoped that the knowledge acquired using in vivo microdialysis and the other techniques described in this volume will ultimately foster development of effective treatments to alleviate the pain and suffering associated with stimulant addictions.

ACKNOWLEDGMENTS The authors wish to thank Mario A. Ayestas and Robert D. Clark for their expert technical assistance with the microdialysis studies. Special appreciation goes to

2345_frame_MASTER.book Page 83 Wednesday, October 23, 2002 1:03 PM

Application of In Vivo Microdialysis Methods

83

Robert B. Horel for his artistic contributions to this chapter. We are grateful to Richard B. Rothman for his continued guidance and support. Finally, the authors are indebted to the NIDA Intramural Research Program for providing the funds and resources needed to carry out much of the work described herein.

REFERENCES 1. Robinson, T.E. and Justice, J.B., Jr., Eds., Microdialysis in the Neurosciences, Elsevier, New York, NY, 1991. 2. Bito, R., Davson, H., Levin, E.M., Murray, M., and Snider, N., The concentration of free amino acids and other electrolytes in the cerebrospinal fluid, in vivo dialysate of brain and blood plasma of the dog, J. Neurochem., 13, 1057, 1966. 3. Delgado, J.M.R., De Freudis, F.V., Roth, R.H., Ryugo, D.K., and Mitruka, B.M., Dialytrode for long term intracerebral perfusion in awake monkeys, Arch. Int. Pharmacodyn., 198, 9, 1972. 4. Ungerstedt, U., Measurement of neurotransmitter release by intracranial dialysis, in Measurement of Neurotransmitter Release in Vivo, Marsden, C.A., Ed., John Wiley & Sons, New York, NY, 1984, pp. 81–105. 5. Kissinger, P.T. and Shoup, R.E., Optimization of LC apparatus for determinations in neurochemistry with and emphasis on microdialysis samples, J. Neurosci. Methods, 34, 3, 1990. 6. Mefford, I.N., Biomedical uses of high-performance liquid chromatography with electrochemical detection, Methods Biochem. Anal., 31, 221, 1985. 7. Kendrick, K.M., Microdialysis measurement of in vivo neuropeptide release, J. Neurosci. Methods, 34, 35, 1990. 8. Westerink, B.H.C. and Justice, J.B., Jr., Microdialysis compared with other in vivo release models, Microdialysis in the Neurosciences, Robinson, T.E. and Justice, K.B., Eds., Elsevier, New York, NY, 1991, pp. 23–46. 9. Westerink, B.H. and DeVries, J.B., Characterization of in vivo dopamine release as determined by brain microdialysis after acute and subchronic implantations: methodological aspects, J. Neurochem., 51, 683, 1988. 10. Auerbach, S.B., Minzenberg, M.J., and Wilkinson, L.O., Extracellular serotonin and 5-hydroxyindoleacetic acid in hypothalamus of the unanesthetized rat measured by in vivo dialysis coupled to high-performance liquid chromatography with electrochemical detection: dialysate serotonin reflects neuronal release, Brain Res., 499, 281, 1989. 11. Moghaddam, B. and Bunney, B.S., Ionic composition of microdialysis perfusing solution alters the pharmacological responsiveness and basal outflow of striatal dopamine, J. Neurochem., 53, 652, 1989. 12. Benveniste, H., Brain microdialysis, J. Neurochem., 52, 1667, 1989. 13. Justice, J.B., Jr., Quantitative microdialysis of neurotransmitters, J. Neurosci. Methods, 48, 263, 1993. 14. Bungay, P.M., Morrison, P.F., and Dedrick, R.L., Steady-state theory for quantitative microdialysis of solutes and water in vivo and in vitro, Life Sci., 46, 105, 1990. 15. Santiago, M. and Westerink, B.H., Characterization of the in vivo release of dopamine as recorded by different types of intracerebral microdialysis probes, Nauyn Schmiedebergs Arch. Pharmacol., 342, 407, 1990.

2345_frame_MASTER.book Page 84 Wednesday, October 23, 2002 1:03 PM

84

Methods in Drug Abuse Research: Cellular and Circuit Level Analyses 16. Paxinos, G. and Watson, C., The Rat Brain in Stereotaxic Coordinates, 2nd ed., Academic Press, New York, NY, 1986. 17. Hernandez, L., Stanley, B.G., and Hoebel, B.G., A small, removable microdialysis probe, Life Sci., 39, 2629, 1986. 18. Baumann, M.H., Ayestas, M.A., Sharpe, L.G., Lewis, D.B., Rice, K.C., and Rothman, R.B., Persistent antagonism of methamphetamine-induced dopamine release in rats pretreated with GBR12909 deconoate, J. Pharmacol. Exp. Ther., 301, 1190, 2002. 19. Rutter, J.J., Baumann, M.H., and Waterhouse, B.D., Systemically administered cocaine alters stimulus-evoked responses of thalamic somatosensory neurons to perithreshold vibrissae stimulation, Brain Res., 798, 7, 1998. 20. Koob, G.F. and Bloom, F.E., Cellular and molecular mechanisms of drug dependence, Science, 242, 715, 1988. 21. Wise, R.A., Neurobiology of addiction, Curr. Opin. Neurobiol., 6, 243, 1996. 22. Hernandez, L. and Hoebel, B.G., Food reward and cocaine increase extracellular dopamine in the nucleus accumbens as measured by microdialysis, Life Sci., 42, 1705, 1988. 23. DiChiara, G. and Imperato, A., Drugs abused by humans preferentially increase synaptic dopamine concentrations in the mesolimbic system of freely moving rats, Proc. Natl. Acad. Sci., 85, 5274, 1988. 24. Robinson, T.E., Jurson, P.A., Bennett, J.A., and Bentgen, K.M., Persistent sensitization of dopamine neurotransmission in ventral striatum (nucleus accumbens) produced by prior experience with (+)–amphetamine: a microdialysis study in freely moving rats, Brain Res., 462, 211, 1988. 25. Wall, S.C., Gu, H., and Rudnick, G., Biogenic amine flux mediated by cloned transporters stably expressed in cultured cell lines: amphetamine specificity for inhibition and efflux, Mol. Pharmacol., 47, 544, 1995. 26. Sonders, M.S., Zhu, S.J., Zahniser, N.R., Kavanaugh, M.P., and Amara, S.G., Multiple ionic conductances of the human dopamine transporter: the actions of dopamine and psychostimulants, J. Neurosci., 17, 960, 1997. 27. Baumann, M.H., Char, G.U., DeCosta, B.R., Rice, K.C., and Rothman, R.B., GBR12909 attenuates cocaine-induced activation of mesolimbic dopamine neurons in the rat, J. Pharmacol. Exp. Ther., 271, 1216, 1994. 28. Hurd, Y.L., Weiss, F., Koob, G.F., And, N.E., and Ungerstedt, U., Cocaine reinforcement and extracellular dopamine overflow in rat nucleus accumbens: an in vivo microdialysis study, Brain Res., 498, 199, 1989. 29. Pettit, H.O. and Justice, J.B., Jr., Dopamine in the nucleus accumbens during cocaine self-administration as studied by in vivo microdialysis, Pharmacol. Biochem. Behav., 34, 899, 1989. 30. Parsons, L.H. and Justice, J.B., Jr., Serotonin and dopamine sensitization in the nucleus accumbens, ventral tegmental area, and dorsal raphe nucleus following repeated cocaine administration, J. Neurochem., 61, 1611, 1993. 31. Kuczenski, R., Segal, D.S., Cho, A.K., and Melega, W., Hippocampus norepinephrine, caudate dopamine and serotonin, and behavioral responses to the stereoisomers of amphetamine and methamphetamine, J. Neurosci., 15, 1308, 1995. 32. Reith, M.E., Li, M.Y., and Yan, Q.S., Extracellular dopamine, norepinephrine, and serotonin in the ventral tegemental area and nucleus accumbens of freely moving rats during intracerebral dialysis following systemic administration of cocaine and uptake blockers, Psychopharmacology, 134, 309, 1997.

2345_frame_MASTER.book Page 85 Wednesday, October 23, 2002 1:03 PM

Application of In Vivo Microdialysis Methods

85

33. Rothman, R.B., Partilla, J.S., Dersch, C.M., Carroll, F.I., and Baumann, M.H., Methamphetamine dependence: medication development efforts based on a dual deficit model of stimulant addiction, Ann. N.Y. Acad. Sci., 914, 71, 2000. 34. Koob, G.F. and Le Moal, M., Drug addiction, dysregulation of reward, and allostasis, Neuropsychopharmacology, 24, 97, 2001. 35. Baumann, M.H. and Rothman, R.B., Alterations in serotonergic responsiveness during cocaine withdrawal in rats: similarities to major depression in humans, Biol. Psychiatry, 44, 578, 1998. 36. Markou, A., Kosten, T.R., and Koob, G.F., Neurobiological similarities in depression and drug dependence: a self-medication hypothesis, Neuropsychopharmacology, 18, 135, 1998. 37. Robertson, M.W., Leslie, C.A., and Bennett, J.P., Jr., Apparent synaptic dopamine deficiency induced by withdrawal from chronic cocaine treatment, Brain Res., 538, 337, 1991. 38. Parsons, L.H., Smith, A.D., and Justice, J.B., Jr., Basal extracellular dopamine is decreased in the rat nucleus accumbens during abstinence from chronic cocaine, Synapse, 9, 60, 1991. 39. Rossetti, Z.L., Hmaidan, Y., and Gessa, G.L., Marked inhibition of mesolimbic dopamine release: a common feature of ethanol, morphine, cocaine and amphetamine abstinence in rats, Eur. J. Pharmacol., 221, 227, 1992. 40. Weiss, F., Markou, A., Lorang, M.T., and Koob, G.F., Basal extracellular dopamine levels in the nucleus accumbens are decreased during cocaine withdrawal after unlimited-access self-administration, Brain Res., 593, 314, 1992. 41. Parsons, L.H., Koob, G.F., and Weiss, F., Serotonin dysfunction in the nucleus accumbens of rats during withdrawal after unlimited access to intravenous cocaine, J. Pharmacol. Exp. Ther., 274, 1182, 1995. 42. Kleven, M.S. and Seiden, L.S., Methamphetamine-induced neurotoxicity: structure activity relationships, Ann. N.Y. Acad. Sci., 654, 292, 1992. 43. Robinson, T.E., Castaneda, E., and Whishaw, I.Q., Compensatory changes in striatal dopamine neurons following recovery from injury induced by 6-OHDA or methamphetamine: a review of evidence from microdialysis studies, Can. J. Psychol., 44, 253, 1990. 44. Cass, W.A., Decreases in evoked overflow of dopamine in rat striatum after neurotoxic doses of methamphetamine, J. Pharmacol. Exp. Ther., 280, 105, 1997. 45. Cass, W.A. and Manning, M.W., Recovery of presynaptic dopaminergic functioning in rats treated with neurotoxic doses of methamphetamine, J. Neurosci., 19, 7653, 1999. 46. Cass, W.A., Attenuation and recovery of evoked overflow of striatal serotonin in rats treated with neurotoxic doses of methamphetamine, J. Neurochem., 74, 1079, 2000. 47. Rothman, R.B., Ayestas, M.A., Dersch, C.M., and Baumann, M.H., Aminorex, fenfluramine, and chlorphenteramine are serotonin transporter substrates. Implications for primary pulmonary hypertension, Circulation, 100, 869, 1999. 48. Baumann, M.H., Ayestas, M.A., Dersch, C.M., and Rothman, R.B., 1-(m-chlorophenyl)piperazine (mCPP) dissociates in vivo serotonin release from long-term serotonin depletion in rat brain, Neuropsychopharmacology, 24, 492, 2001. 49. Rothman, R.B. and Glowa, J.R., A review of the effects of dopaminergic agents on humans, animals, and drug seeking behavior and its implications for medication development: focus on GBR12909, Mol. Neurobiol., 11, 1, 1995. 50. Andersen, P.H., The dopamine uptake inhibitor GBR12909: selectivity and molecular mechanism of action, Eur. J. Pharmacol., 166, 493, 1989.

2345_frame_MASTER.book Page 86 Wednesday, October 23, 2002 1:03 PM

86

Methods in Drug Abuse Research: Cellular and Circuit Level Analyses 51. Rothman, R.B., Mele, A., Reid, A.A., Akunne, H.C., Greig, N., Thurkauf, A., DeCosta, B.R., Rice, K.C., and Pert, A., GBR12909 antagonizes the ability of cocaine to elevate extracellular levels of dopamine, Pharmacol. Biochem. Behav., 40, 387, 1991. 52. Pogun, S., Scheffel, U., and Kuhar, M.J., Cocaine displaces [3H]WIN 35,428 binding to dopamine uptake sites in vivo more rapidly than mazindol or GBR 12909, Eur. J. Pharmacol., 198, 203, 1991. 53. Tella, S.R., Effects of monoamine reuptake inhibitors on cocaine self-administration in rats, Pharmacol. Biochem. Behav., 51, 687, 1995. 54. Glowa, J.R., Wojnicki, F.H.E., Matecka, D., Rice, K.C., and Rothman, R.B., Effects of dopamine reuptake inhibitors on food and cocaine maintained responding. II. Comparisons with other drugs and repeated administrations, Exp. Clin. Psychopharmacol., 3, 232, 1995. 55. Lewis, D.B., Matecka, D., Zhang, Y., Hsin, L-W., Dersch, C.M., Stafford, D., Glowa, J.R., Rothman, R.B., and Rice, K.C., Oxygenated analogs of 1-[2-(diphenylmethoxy)ethyl]- and 1-[2-[bis-(4-fluorophenyl)methoxy]ethyl] 4-(3-phenylpropyl)piperazines (GBR12935 and GBR 12909) as potential extended-action cocaine-abuse therapeutic agents, J. Med. Chem., 42, 5029, 1999. 56. Glowa, J.R., Fantegrossi, W.E., Lewis, D.B., Matecka, D., Rice, K.C., and Rothman, R.B., Sustained decrease in cocaine-maintained responding in rhesus monkeys with 1-[2-[bis-(4-fluorophenyl)methoxy]ethyl]-4-(3-hydroxy-3-phenylpropyl) piperazinyl decanoate, a long-acting ester derivative of GBR 12909, J. Med. Chem., 39, 4689, 1996.

2345_frame_MASTER.book Page 87 Wednesday, October 23, 2002 1:03 PM

4

In Vivo Voltammetry in Drug Abuse Research Michael F. Salvatore, Alexander F. Hoffman, Jason J. Burmeister, and Greg A. Gerhardt

CONTENTS 4.1 4.2

4.3

4.4

Introduction ....................................................................................................88 Principles of in Vivo Voltammetry .................................................................88 4.2.1 Compounds Must Have Intrinsic Electrochemical Properties ..........89 4.2.2 Working and Reference Electrodes Used in Voltammetry................90 4.2.2.1 Carbon Fiber “Working” Microelectrodes .........................90 4.2.2.2 Electrode Coatings..............................................................91 4.2.2.3 Reference and Auxiliary Electrodes...................................92 4.2.3 Recording Methods ............................................................................93 4.2.3.1 Amperometry ......................................................................93 4.2.3.2 High-Speed Chronoamperometry.......................................93 4.2.3.3 Fast-Scan Cyclic Voltammetry (Triangular).......................94 4.2.4 Quantification and Identification Issues ............................................95 4.2.5 Instrumentation for Voltammetric Studies.........................................95 4.2.5.1 Commercial.........................................................................95 4.2.5.2 Laboratory Designed ..........................................................96 Scope of Voltammetry....................................................................................96 4.3.1 Neurotransmitters Studied Using Voltammetry.................................96 4.3.1.1 L-Glutamate ........................................................................96 4.3.2 Information Gained from Voltammetric Studies ...............................97 4.3.2.1 Release of Neurotransmitters and Neuromodulators .........97 4.3.2.2 Uptake/Reuptake and Diffusion .........................................97 4.3.3 Experimental Paradigms Studied with Voltammetry.........................97 4.3.3.1 Cells in Culture...................................................................97 4.3.3.2 Brain Slices.........................................................................98 4.3.3.3 Whole Animals .................................................................100 Applications of Voltammetry to Drug Abuse Questions.............................101 4.4.1 Ethanol .............................................................................................101 4.4.2 Psychomotor Stimulants ..................................................................101 4.4.2.1 d-Amphetamine.................................................................102 4.4.2.2 Cocaine .............................................................................103 4.4.2.3 THC (Cannabinoids).........................................................104 87

2345_frame_MASTER.book Page 88 Wednesday, October 23, 2002 1:03 PM

88

Methods in Drug Abuse Research: Cellular and Circuit Level Analyses

4.5 Future Applications ......................................................................................104 Acknowledgments..................................................................................................104 Abbreviations and Acronyms ................................................................................105 References..............................................................................................................105

4.1 INTRODUCTION This chapter is intended for those who wish to gain a general knowledge of the principles, materials, and approaches that underlie the current methodologies being used to study drug-abuse issues in central nervous system (CNS) tissues using electrochemical techniques, which are often termed “voltammetric recordings.” We have not tried to be comprehensive in our citation of all literature pertaining to this topic. The interested reader is directed to other reviews and review chapters for a more thorough understanding of the use of these methods.1–6 We hope that this review will be of value to the numerous scientists who wish to use these methods to study the effects of drugs of abuse on the dynamics of neurotransmitters and neurochemicals in the CNS. In addition, we have included summaries of some of the results to date regarding the use of voltammetric recordings on issues of drug abuse.

4.2 PRINCIPLES OF IN VIVO VOLTAMMETRY In the 1970s, Ralph Adams pioneered the concept of using in vivo voltammetric recordings for the analysis of neurotransmitters and neurochemicals, such as dopamine (DA), in the CNS.7 Voltammetric recording techniques have been refined for rapid in vivo electrochemical measures of neurotransmitters in brain slices and in the brains of rodents and monkeys.8–15 In contrast to microdialysis techniques, which permit measurement of basal extracellular levels and generally slow (ca. 10- to 20-min) changes in extracellular monoamines and metabolites, electrochemical recordings using microelectrodes 5 to 30 mm in diameter allow for rapid (1 to 1000 Hz) measurements of neurotransmitters and neuromodulators in discrete brain nuclei.8–15 In addition, such methods can dynamically monitor neurotransmitter release and clearance kinetics.16,17 In vivo voltammetric studies involve the insertion of microelectrodes into the brain regions of interest in the intact brain or brain slices (see Figures 4.1A and 4.1B). In the case of cells in culture, the microelectrodes are positioned as close as possible to individual cells (Figure 4.1C). Voltammetric measurements are then started to achieve a stable background signal. Once stable measurements are recorded, the effects of an event, such as electrical stimulation or chemical stimulation of the neurons or cells, are studied. At the present time, these methods are not sensitive enough to record basal levels of neurotransmitters or neuromodulators; instead they measure concentration changes from the basal level. Once a stable evoked signal is obtained, the effects of drugs of abuse on the neuronal systems are studied by local, systemic, or bath application of the drugs. A more detailed explanation of the principles of the recordings as well as the materials and equipment used for such measures is included in this section.

2345_frame_MASTER.book Page 89 Wednesday, October 23, 2002 1:03 PM

In Vivo Voltammetry in Drug Abuse Research

89

FIGURE 4.1 Schematic diagram illustrating the basic instrumentation needed to perform in vivo voltammetric recordings. Recording preparations such as whole animals (A), brain slices (B), and cells in culture (C) are depicted. Waveforms 1, 2, 3, and 4 show the input potentials and output waveforms for (1) amperometry, (2) high-speed chronoamperometry, (3) fast-scan cyclic voltammetry, and (4) fast cyclic voltammetry.

4.2.1 COMPOUNDS MUST HAVE INTRINSIC ELECTROCHEMICAL PROPERTIES Voltammetry involves the use of microelectrodes (see Section 4.2.2) to directly measure the release and uptake of neurotransmitters and neuromodulators in CNS tissues or cells in culture. To detect and quantify the neurotransmitter or neuromodulators that touch the surface of the microelectrode, the compound of interest must

2345_frame_MASTER.book Page 90 Wednesday, October 23, 2002 1:03 PM

90

Methods in Drug Abuse Research: Cellular and Circuit Level Analyses

have intrinsic electrochemical properties, or the microelectrode must be chemically modified (see Section 4.2.2.2.3) with a reaction layer on its surface to generate a reaction product that has intrinsic electrochemical properties. Voltammetry measurements are based on an analyte coming in contact with the surface of the microelectrode, which has a potential sufficient to produce the oxidation or reduction of the analyte at the surface of the microelectrode in order to measure its concentration. These principles can be reviewed in depth in Bard and Faulkner18 or Adams.19 As a function of the voltammetric method (see Section 4.2.3) and the magnitude of the potentials applied to the microelectrode surface (see below), electrochemical reactions occur at the surface of the microelectrode, which generate a current proportional to the concentration of the electrochemically active species. The current flow due to these chemical reactions (termed the Faradaic current) is proportional to the concentration of the analyte. More than one equation describes the current responses of all voltammetric recordings as they are related to the recording method.18 However, the responses of the microelectrodes are related to the following: (1) the number of electrons for a given chemical reaction (for example, DA oxidizes to a quinone with a two-electron process), (2) the Faraday constant, (3) the total active recording surface area of the microelectrode, (4) the bulk solution concentration of the analyte, and (5) the mass transport of analyte to the microelectrode surface. Microelectrodes have both linear and spherical diffusion of analyte to the microelectrode tip, and an additional set of terms must be applied to completely model the microelectrode response.5 In an unstirred solution, such as the extracellular space of the brain or in a quiet solution next to a cell, a simple equation can be used to describe the response of a microelectrode for a given neurotransmitter. i = X Cb

4.1

In this equation, X — the calibration factor for a given microelectrode — can be determined pre- or post-experimentation by varying Cb (bulk solution concentration of the analyte) and measuring i (due to electrochemical reaction of the compound of interest) under solution conditions that mimic the extracellular space of the brain (artificial cerebrospinal fluid) or the media in which cells are studied.20,21 The analytical foundation for the use of voltammetric recording methods in biological systems is that the microelectrodes respond linearly to changes in concentration of the neurotransmitter or neuromodulator.

4.2.2 WORKING AND REFERENCE ELECTRODES USED IN VOLTAMMETRY 4.2.2.1 Carbon Fiber “Working” Microelectrodes The “working” or recording electrode used for in vivo voltammetric measurements in CNS tissues is generally fabricated to achieve a tip diameter of 5 to 60 µm, depending on the application. Carbon fiber or carbon monofilament working microelectrodes are the most common and widely used forms of microelectrodes for CNS

2345_frame_MASTER.book Page 91 Wednesday, October 23, 2002 1:03 PM

In Vivo Voltammetry in Drug Abuse Research

91

FIGURE 4.2 Photomicrograph of the tip of a carbon monofilament microelectrode (30 mm diameter) used for in vivo voltammetry recordings. Scale bar = 10 mm.

recordings. One or more carbon fibers or carbon monofilaments are sealed in glass capillaries to form the working microelectrode.3,13,22,23 An example of the tip of a 30-mm-diameter carbon monofilament microelectrode is seen in Figure 4.2. Electrical connections to the fibers are made using graphite pastes3,13,22 or salt solutions.24 Manufacturers and part numbers for popular fibers and one monofilament are as follows: 10-mm-diameter (P-55, Amoco, Greenville, SC),24 30-mm-diameter carbon monofilament (#80036–001, Textron Systems Division, Wilmington, MA).3,5,13 Several companies sell prefabricated carbon-fiber- or carbon-monofilament-based microelectrodes including Quanteon, LLC (Lexington, KY), World Precision Instruments (Sarasota, FL), Dagan Corp. (Minneapolis, MN), MPB Electrodes (London) and Axon Instruments (Foster City, CA). The recording properties of carbon fiber microelectrodes can be quite different even for carbon fibers of similar size. The type of fiber or carbon monofilament and the measures taken to prepare the fibers can affect the resulting microelectrode’s recording properties.25 Thus, it is important to use carbon fibers and carbon monofilaments that have been previously characterized. Fabrication methods for a variety of fiber microelectrodes can be found in previously published articles.3,6,26–28 Aside from carbon fibers, several designs of fabricated microelectrodes formed by lithographic methods are emerging as useful microelectrodes for voltammetric recordings. Both silicon-based29,30 and ceramic-based sensors31,32 are being developed. At present they are capable of measuring a variety of neurotransmitters analogous to carbon-fiber-based microelectrodes. 4.2.2.2 Electrode Coatings A variety of coatings are used to enhance the selectivity of microelectrodes for certain neurotransmitters and to make them sensitive for compounds that are not readily oxidized or reduced. Some coatings can protect the surfaces of microelectrodes from fouling. The trade-off is that such layers can increase the response times of the recordings by blocking the analyte from reaching the microelectrode surface. Thus, coatings must be used sparingly.

2345_frame_MASTER.book Page 92 Wednesday, October 23, 2002 1:03 PM

92

Methods in Drug Abuse Research: Cellular and Circuit Level Analyses

4.2.2.2.1 Nafion The Teflon derivative Nafion is used to repel anions and prevent microelectrodes from fouling; it is able to do this because it has sulfonic acid groups substituted into the polymer. Films formed by applying Nafion to the microelectrode surface are one of the most widely used materials for improving the selectivity of voltammetric recordings in CNS tissues.2,3,21,33 Positively charged ions like DA may also be concentrated at the microelectrode surface by Nafion’s negative charge. Nafion is commercially available as suspensions in alcohol (5%, Aldrich Chemical Company). Microelectrodes may simply be dipped into this mixture. Many drying techniques have been investigated including air drying,2,34 use of heat guns,3 and oven baking.21,35 In general, a thicker Nafion coating will enhance the selectivity of the microelectrode for cationic neurotransmitters (such as DA) vs. anions, but at the expense of a slower response time. Thus, the effect upon microelectrode selectivity and sensitivity of various coating and drying methods can vary among labs and experimenters. However, once coating procedures are optimized, Nafion consistently enhances the selectivity of carbon fiber or carbon monofilament microelectrodes by ≥ 100: 1 for DA, NE, and 5-HT as compared to AA, uric acid, DOPAC, and 5-HIAA.21 4.2.2.2.2 Electropolymerized Coatings (Poly)phenylenediamine electropolymerized onto the microelectrode surface has been used to enhance selectivity and prevent fouling on microelectrodes based on a molecule’s size. (Poly)phenylenediamine has holes that allow small molecules like NO and H2O2 to diffuse to the microelectrode surface to be oxidized while blocking larger organic molecules such as DA from the microelectrode.36 We recently developed a microelectrode based on a Nafion-coated, o-phenylenediamine-modified carbon fiber microelectrode to selectively measure NO in the presence of ascorbate, DA, and nitrite. 4.2.2.2.3 Enzymes Enzymes immobilized onto microelectrodes can be used to measure neurotransmitters or neuromodulators that are not by themselves electroactive. This represents another area of major growth in this field. The enzymes are utilized by converting the compound of interest into an electroactive product, which can be quantified at the microelectrode surface. Enzymes are often highly selective for a particular substrate, which contributes to the selectivity of such enzyme-based sensors. To carry out these studies oxidase and dehydrogenase enzymes are commonly attached to the surfaces of the microelectrodes. Some oxidase enzymes produce hydrogen peroxide as one of their products, which may be monitored at the microelectrode. Glucose, lactate, choline, and glutamate oxidase enzymes have been made commercially available.31,32,36–39 Microelectrodes have been reported for the detection of all these molecules. Oxygen dependence is a concern because oxygen is consumed in the reaction. Dehydrogenase enzymes are the second class of enzymes. They oxidize the substrate while reducing NAD+ to NADH that can be monitored. 4.2.2.3 Reference and Auxiliary Electrodes A stable reference electrode is a critical component of electrochemical measurements because the potential of the recording electrode is controlled relative to a reference

2345_frame_MASTER.book Page 93 Wednesday, October 23, 2002 1:03 PM

In Vivo Voltammetry in Drug Abuse Research

93

electrode. Miniature forms of reference electrodes and auxiliary electrodes are used so that they can be directly implanted into CNS tissue. However, standard size reference electrodes are often employed for brain slice and single-cell recordings. The most common reference electrode used for voltammetric recordings in CNS tissues is the miniature Ag/AgCl electrode,3,10,19,21 which can be constructed by using a short length (4 to 6 cm) of Teflon-coated (250-µm O.D.) silver wire. Teflon is removed from 3 to 5 mm on one end to expose the silver wire, which is coated with AgCl. To coat the wire a voltage of +1 V is applied to the Ag wire that is submerged in 1 M HCl for about 5 to 10 min using platinum wire as the cathode. The RE-5 (BAS, Bioanalytical Systems, Lafayette, IN) is a commonly used standard size Ag/AgCl reference for slice and single-cell recordings. Auxiliary electrodes are not used in many laboratories because of the small currents and low impedance associated with the recording solutions. Refer to Bard and Faulkner18 or Adams19 for further discussion of auxiliary electrodes.

4.2.3 RECORDING METHODS The single most confusing aspect of voltammetric recordings in CNS tissues and cells is the choice of recording method. This section explains the different recording methods used for rapid (1–1000Hz) recordings, and the following section discusses how they have been used to study neurotransmitters and neurochemicals in CNS tissues and single cells. The difference among the methods stems from the fact that the potential to record the oxidation or reduction of analytes such as DA is varied at the working microelectrode. 4.2.3.1 Amperometry Amperometry is the simplest of the electrochemical techniques because it involves the measurement of current at a constant fixed potential (see Figure 4.1, inset 1). Current can be monitored continuously; therefore, this technique allows one to record events that occur very fast (200 to 1000 Hz). Because of the high sampling resolution amperometric recordings are typically used to study catecholamines released from single cells.40–42 4.2.3.2 High-Speed Chronoamperometry Based on the work of Cottrell, chronoamperometry is one of the oldest methods used for in vivo electrochemical measurements in CNS tissue.18 High-speed chronoamperometric (HSC) recordings are a rapid version of amperometry.5,22 In a suitable buffer solution to simulate CNS tissue, a carbon fiber microelectrode is held at a resting potential of –0.20 to 0.0 V vs. Ag/AgCl reference. The voltage is then instantaneously stepped to an oxidation voltage chosen to exceed the oxidation potential of the analyte (ca. +0.45 or +0.55 V) (see Figure 4.1, inset 2). The potential step produces a rapid change in the current that rapidly decays as a function of time. This predominant “charging current” is allowed to decay for 4 to 20 msec before the background current of the microelectrode is measured, resulting in a lower background upon which to measure changes in analyte concentration. The square-

2345_frame_MASTER.book Page 94 Wednesday, October 23, 2002 1:03 PM

94

Methods in Drug Abuse Research: Cellular and Circuit Level Analyses

wave voltage step is held at the higher oxidation voltage for a fixed time period of 20, 50, or 100 msec during which the oxidation current is sampled for 15, 40, or 80 msec. The voltage step is then instantaneously returned to the resting potential to complete a square wave. Like the first oxidation step, the return of the potential to the resting voltage produces a reduction reaction at the surface of the microelectrode, which is recorded as per the oxidation current step. The symmetric voltage steps are continued at a set time period to produce signals with a sampling resolution of 1 Hz, 5 Hz (100 msec oxidation, 100 msec reduction; 5/sec), or 10 Hz (50 msec oxidation, 50 msec reduction; 10/sec. The signal-to-noise of such recordings is greatly enhanced because small changes in current due to the detection of an analyte can be masked by large background current. In addition, the lower background current levels produce fewer recording artifacts. Detection limits for DA and 5-HT are typically 25 nM using a 30-mm-diameter carbon monofilament cylinder microelectrode.44,45 Many recordings in CNS tissues and other applications do not require temporal resolutions of 5 to 25 Hz. Therefore, for more standard recordings, the signals are averaged over a 1- or 2-sec time period, which further enhances the signal-to-noise of the recording methods.13,44,45 4.2.3.3 Fast-Scan Cyclic Voltammetry (Triangular) In cyclic voltammetry, the potential is steadily ramped up from a starting potential to the switching potential, where it is ramped down to the starting potential (see Figure 4.1, inset 3). The current is measured while this triangular potential wave is applied. The same potential changes as normal cyclic voltammetry are utilized with fast-scan cyclic voltammetry only at much higher scan rates (typically 100 to 1000 V/sec). Excellent sensitivity, time resolution, and compound identification capabilities make this one of the most popular in vivo voltammetric methods.3,8,10,11,17 Typical resting and switching potentials for a single carbon fiber microelectrode are –0.4 and ±1.0 V vs. Ag/AgCl reference, respectively, for DA. High scan rates are coupled with 10- to 300-Hz repetition rates to track rapid biological events. However, high scan rates cause large background currents, which can increase detection limits. Voltammograms are background-subtracted to make the analyte peaks (Faradaic current) apparent and to obtain an analytically useful signal. A background voltammogram or series of voltammograms is selected and subtracted from subsequent scans during a recording session. Repetitive cyclic voltammograms, often 10 Hz or higher, are sampled to obtain time-dependent concentration profiles. The current is averaged over a desired potential range corresponding to oxidation (or reduction) of the analyte. Other waveforms have been used with cyclic voltammetry including a triphasic ramp voltage (see waveform 4 in Figure 4.1).4,25,46 This technique has been called 11/2 triangle or fast cyclic voltammetry (FCV). In FCV, the potential decreases from the resting potential to a lower potential, followed by an increase to the switching potential, which is higher than the resting potential. The potential is finally ramped down to the lower potential then back to the resting potential, completing the Wshaped waveform. Each scan typically lasts 15 to 20 msec, and the scan rate is 300 V/s or higher, with a frequency 1 to 2 Hz. The resting, lower, and upper potentials

2345_frame_MASTER.book Page 95 Wednesday, October 23, 2002 1:03 PM

In Vivo Voltammetry in Drug Abuse Research

95

are typically 0, –1, and +1, respectively. Using this approach, DA has an oxidation peak at +0.6 V and a reduction peak at –0.2 V using certain carbon fiber microelectrodes with a detection limit of about 200 nM.25,46

4.2.4 QUANTIFICATION

AND IDENTIFICATION ISSUES

Perhaps the greatest controversy surrounding voltammetric measurements in CNS tissues and cells in culture concerns the question of how you know what you’re measuring. In reality there are relatively few electroactive compounds present in brain tissues and cells, and the methods are inherently selective.3,6 However, there is the valid concern that electrochemically active species in the region being measured, other than the one of interest, can contribute to the current at the microelectrode surface and give a false positive. Unlike microdialysis methods that use powerful separation techniques such as HPLC-EC and CE to identify and quantify, the microelectrodes can detect a number of molecules at the same time. This can complicate the interpretation of in vivo voltammetric data. Electrode coatings (see above) have been used to further enhance the selectivity of the microelectrodes for neurotransmitters or neurochemicals of interest, and this has been a major area of research. With fast cyclic voltammetry methods the voltage ranges of the oxidation and reduction peaks on the signals, called voltammograms, or the shapes of the voltammograms can be used to identify compounds that are responsible for a given signal. In addition, the number of peaks can be used to identify the source of the sensor response.3,8,10 In another recent report, three-dimensional plots of voltage vs. time with current plotted in false color were used to visualize the source of the sensor response.47 All of the approaches are useful for distinguishing between DA, 5-HT, and pH changes during fast-scan cyclic voltammetry recordings. High-speed chronoamperometric recordings can also yield information regarding the major contributor to the electrochemical signal.2,15,33 The detection of DA yields reduction to oxidation (red/ox) current ratios with Nafion-coated microelectrodes between 0.5 and 0.9, depending on the thickness of the Nafion film and the drying temperature of the Nafion coating procedure.5,33 In contrast, ascorbic acid (AA) produces red/ox ratios of 0.0, and 5-HT produces ratios of 0.0 to 0.2.2,15 Thus, somewhat analogous to the waveforms seen with fast-scan cyclic voltammetry,3,10 DA, AA, and 5-HT produce distinctive red/ox ratios, which can be used to further identify the detection of these compounds in brain tissues and cells.

4.2.5 INSTRUMENTATION

FOR

VOLTAMMETRIC STUDIES

Specialized instruments (potentiostats) often capable of making fast, repetitive (1 to 1000 Hz), and low-noise current measurements are required for measurements of neurotransmitters and neurochemicals with microelectrodes. These instruments must be able to measure currents that are typically in the low-pico- to mid-nanoampere range. 4.2.5.1 Commercial Several commercial patch-clamp amplifiers can be used for amperometric recordings. Axon Instruments, Inc. (Foster City, CA) produces a potentiostat called the

2345_frame_MASTER.book Page 96 Wednesday, October 23, 2002 1:03 PM

96

Methods in Drug Abuse Research: Cellular and Circuit Level Analyses

Gene Clamp 500 that is ideal for fast amperometric recordings. The Dagan Corporation (Minneapolis, MN) makes the Chem-Clamp, which is also suitable for amperometric recordings. Several companies market specialized systems designed for high-speed in vivo voltammetric measurements. All of these instruments are optimized for rapid recordings using microelectrodes, but they cannot perform all voltammetry methods. Cypress Systems sells the EI-400, which is a low-noise potentiostat generally used for FSCV measurements. The Millar Voltammeter,48 an analog system designed primarily for FCV recordings, is sold by P.D. Systems (West Molesey, Surrey, UK). GMA Technologies, Inc. (Vancouver, BC, Canada) markets a microcomputer-controlled potentiostat called the E-Chempro electrochemical instrument. This instrument can perform chronoamperometric recordings. Quanteon, L.L.C. (Lexington, KY) now markets the FAST-12 instrument, a microcomputer-controlled system for performing high-speed chronoamperometry and amperometry. The availability and capabilities of all of the abovementioned instruments can be obtained from the manufacturers, many of whom maintain Web sites for the interested investigator. 4.2.5.2 Laboratory Designed In vivo electrochemical recording systems are built or modified by many laboratories. These systems typically consist of 2-stage amplification, a general-purpose potentiostat, an external waveform generator (if required), and an output device such as a strip chart recorder, sample and hold amplifier, and microcomputer. The general configuration of such a system is shown in Figure 4.1.

4.3 SCOPE OF VOLTAMMETRY 4.3.1 NEUROTRANSMITTERS STUDIED USING VOLTAMMETRY By far the largest number of voltammetric recordings in CNS tissues have been carried out on the nigrostriatal and mesolimbic DA pathways in the CNS of rodents and monkeys.9–14,16,17,35,44 Voltammetric measures of DA have extensively involved studies of electrical stimulation of the medial forebrain bundle and the effects of drugs that alter DA neurons in anesthetized rats and, more recently, in freely moving animals.11–14 The low extracellular levels of DA, NE, and 5-HT have limited the number of voltammetric studies of resting levels or basal levels of the neurotransmitters; such studies can be carried out with microdialysis methods. This has resulted in the extensive use of electrical and chemical stimuli to cause the release of DA so that it can be reliably measured. In contrast to the extensive number of studies of DA, there is a growing body of data regarding NE and 5-HT release from brain areas of the rat such as the hippocampus, substantia nigra, raphe nucleus, cerebellum, locus coeruleus, and dorsal raphe.2,8,10,15,50 In addition, voltammetric techniques can be used to measure NO in brain tissue.36,51 4.3.1.1 L-Glutamate An exciting area of growth in the area of voltammetric measures involves the development of microelectrodes for the detection of the excitatory amino acid glutamate.52

2345_frame_MASTER.book Page 97 Wednesday, October 23, 2002 1:03 PM

In Vivo Voltammetry in Drug Abuse Research

97

Substantial progress has been made by Adrian Michael and co-workers on the development of a carbon-fiber-based microelectrode using a multilayer process of a crosslinked redox polymer that incorporates horseradish peroxidase glutamate oxidase and ascorbic acid oxidase followed by an overcoating of Nafion.53 The resulting microelectrode is capable of measuring glutamate changes that occur with a temporal resolution of 10 to 30 sec and a spatial resolution of 10 mm. In addition, Hu et al.54 have reported microelectrodes based on a platinum–iridium wire microelectrode coated with Nafion, cellulose acetate, and crosslinked glutamate oxidase with ascorbic acid oxidase, and cellulose. Finally, Burmeister and colleagues have assembled microelectrodes31,32 capable of measuring glutamate based on their ceramic-based multisite microelectrodes. The microelectrodes are coated with Nafion and then crosslinked with glutamate oxidase. This produces microelectrodes that are capable of measuring sensitive changes in glutamate with a temporal resolution of about 1 sec. These microelectrode designs are being implemented for studies of the CNS.

4.3.2 INFORMATION GAINED

FROM

VOLTAMMETRIC STUDIES

4.3.2.1 Release of Neurotransmitters and Neuromodulators In vivo electrochemical recording methods were developed primarily to measure the release of neurotransmitters such as DA, NE, and 5-HT in the intact CNS. The majority of in vivo voltammetric studies to date have relied on ways of activating the release of these compounds from nerve endings, varicosities, and dendrites by using electrical stimulation because of the very low basal release levels of these neurotransmitters.3,10,55 An example of electrically evoked release of DA in rat striatal brain slices is shown in Figure 4.3. Chemical stimuli, such as d-amphetamine, excess potassium, veratridine, and tyramine, have also been extensively used to cause the release of neurotransmitters.2,45,56 In freely moving animals, systemically administered drugs and even selfadministered drugs have been used to activate release processes.11–14 4.3.2.2 Uptake/Reuptake and Diffusion A major mechanism for inactivating released neurotransmitters is through membrane-bound proteins known as transporters. These efficient proteins shuttle the released neurotransmitter back into the neuron and inactivate the signal produced by the neurotransmitter.17,57 The function of the DA, NE, and 5-HT neuronal transporters can be studied using voltammetric methods.2,9,10,15,17,22,24,34,43 An approach based on electrical stimulation of the slice or intact animal that allows for measuring both release and uptake parameters in the same preparation has been developed by Wightman and co-workers.3,11,12,16

4.3.3 EXPERIMENTAL PARADIGMS STUDIED

WITH

VOLTAMMETRY

4.3.3.1 Cells in Culture Nearly a decade ago Wightman and colleagues42 introduced the concept of using fast amperometric recordings to study the release of neurotransmitters such as NE from single cells. This is a rapidly growing area of study that uses £ 5-mm-diameter

2345_frame_MASTER.book Page 98 Wednesday, October 23, 2002 1:03 PM

Methods in Drug Abuse Research: Cellular and Circuit Level Analyses

µM Change from baseline

A

(5 µM)

µM Change from baseline

98

B

FIGURE 4.3 Effect of cocaine on electrically evoked dopamine signals in rat brain slice. (A) High-speed chronoamperometric recording from the nucleus accumbens. Release was elicited every 5 min by delivering a 1-sec stimulus train (100 Hz, 0.1 msec width, 5 V) to a bipolar stimulating electrode positioned near the recording electrode. Cocaine (5 mM) was applied via superfusion, as indicated by the bar. Note the potentiation of the signal amplitude, which eventually recovered to baseline levels following washout of the cocaine. The inset plot shows responses on an expanded time scale in order to highlight the change in the signal time course. (B) Summary of the effects of cocaine (5 or 50 mM, n = 7). Data are expressed as the mean ± S.E.M. of the percentage change from baseline. Both the signal amplitude and time course (T80) were increased significantly (**p < 0.05, student’s paired t-test), whereas the Tc (rate of uptake) parameter was not significantly altered by cocaine.

disk recording microelectrodes coupled with amperometric or FSCV recordings to study properties of quantal release from individual cells in culture. To date a number of studies have been carried out on PC-12 cells, adrenal chromaffin cells, mast cells, and, more recently, engineered cells.40,42,58,59 These studies demonstrate the very rapid recording capabilities of the amperometric recording methods. 4.3.3.2 Brain Slices The use of brain slices from rats and mice is one of the most rapidly growing areas using voltammetric recording techniques. A photograph of a brain-slice recording chamber and the voltammetric system is shown in Figure 4.4A, and measurements of

2345_frame_MASTER.book Page 99 Wednesday, October 23, 2002 1:03 PM

In Vivo Voltammetry in Drug Abuse Research

99

A i

t 1s

Outflow

Inflow

0.1 µM

5s

µM Change from baseline

C

B

0.4

Ox Red

0.3 0.2

0.1 µM 10 s

0.1 0.0

TTX (500 nM)

–0.1 0

10

20

30

Time (min)

FIGURE 4.4 High-speed chronoamperometric recording in brain slices. (A) Schematic illustration of the slice chamber setup. The left panel shows the basic configuration of the recording chamber. The outer chamber is filled with H2O and is heated and maintained at 32°C. The slice is fixed to nylon netting and is continuously superfused with aCSF at a rate of 2 ml/min. The right panel of the figure shows the relative positions of a stimulating electrode and working microelectrode in the dorsolateral striatum. A 1-sec train of 0.1-msec pulses is delivered to the stimulating electrode, and chronoamperometric recordings are performed as described. (B) Representative examples of reproducible signals obtained in the dorsolateral striatum during the stimulus train shown in (A). Signals were obtained every 5 min and were collected and displayed “as is” at 10 Hz. Each point represents 100 msec. Stimulus trains were delivered at the time indicated by the arrow in each trace. (C) Sample record showing the action-potential-dependent origin of the electrochemical signals obtained in the striatum. Signals were collected at 5 Hz and displayed at 1-sec intervals. After three reproducible signals were obtained at 5-min intervals, tetrodotoxin (TTX, 500 nM) was superfused into the slice chamber. The electrochemical responses were rapidly and completely eliminated by TTX. In this and other slices, the effect was partially reversed only after prolonged (> 1 h) washout of TTX (not shown). The inset shows a single signal on an expanded time scale as well as both the oxidation and reduction currents. The redox ratio is consistent with DA.

2345_frame_MASTER.book Page 100 Wednesday, October 23, 2002 1:03 PM

100

Methods in Drug Abuse Research: Cellular and Circuit Level Analyses

DA release in rat striatal brain slices can be seen in Figure 4.3. Voltammetric recordings can be used in parallel with electrophysiological measurements of synaptic activity. This methodology also has a number of advantages over studying the CNS using anesthetized or freely moving animals. First, the slice preparation allows for visual control of all microelectrode placements. This is advantageous when studying layered structures such as the hippocampus and small nuclei such as the substantia nigra.2,8,10,50 Second, the slice preparations can be superfused with known concentrations of drugs for pharmacological studies. Third, the slice preparation makes it easier to study the brains of smaller animals such as mice. Finally, the slice preparation is unanesthetized tissue. Disadvantages include the fact that some synaptic connectivity is invariably lost in preparing brain slices and that the viability of the slices can be a problem. 4.3.3.3 Whole Animals Anesthetized rats have been used extensively to study the dynamics of DA, NE, and 5-HT neuronal systems using voltammetric techniques.1,9,15,16,22 A photograph of an anesthetized animal preparation is shown in Figure 4.5 (see Color Figure 4.5 following page 50). Potential disadvantages of this preparation are the effects of anesthesia on the measurements and the inability to know the exact concentrations of drugs that affect CNS function. However, the anesthetized preparation is still powerful for studies of the intact CNS.

FIGURE 4.5 Photograph of an anesthetized animal preparation and recording system for voltammetric studies in anesthetized rats.

2345_frame_MASTER.book Page 101 Wednesday, October 23, 2002 1:03 PM

In Vivo Voltammetry in Drug Abuse Research

101

The freely moving animal is the other major preparation for voltammetric studies. In these studies, the recording microelectrodes are often implanted days prior to measurements, or a microdrive cannulae system is attached to the skull of the rat.11–14,28,49 This typically involves the use of special miniature potentiostats that are placed close to the animal’s head.11,12,49 Electrically evoked release of DA and potassium-induced release of DA have been measured. In addition, a number of studies in freely moving rats have been carried out to investigate the effects of drugs of abuse such as cocaine and amphetamine. These studies are very time consuming as compared with brain-slice or cell-culture studies, but they hold the promise to understanding more of the underlying properties of neurotransmitter signaling in freely moving and behaving rats.11–14

4.4 APPLICATIONS OF VOLTAMMETRY TO DRUG ABUSE QUESTIONS The involvement of catecholamines (particularly DA and 5-HT) in behaviors influenced by drugs of abuse is well established.60,61 The brain’s motivation and reward center is the mesolimbic system, which is a dopamine-rich brain system originating in the ventral tegmental area. The termination sites of the mesolimbic system that are loci for motivational and reward-directed behavior are the prefrontal cortex, nucleus accumbens, and amygdala. The major drugs of abuse, including ethanol, the psychomotor stimulants cocaine and amphetamine, delta9-tetrahydrocannibinol (THC), and opiates, have been shown to influence catecholamine levels as well as release and clearance properties. Over the last two decades, voltammetry studies have provided insights into how drugs of abuse affect catecholamine function in the CNS.

4.4.1 ETHANOL Electrophysiological and microdialysis studies have shown that ethanol influences the firing rate of midbrain dopamine neurons and levels of dopamine in the nucleus accumbens.62,63 Voltammetry studies have shown that ethanol can attenuate potassium-stimulated dopamine release in the nucleus accumbens or striatum.64,65 Recent studies using FSCV have shown that ethanol dose-dependently decreases electrically stimulated dopamine efflux in the caudate-putamen in freely moving rats and in brain slices. It has recently been shown that this ethanol-mediated reduction in electrically induced dopamine efflux was not related to any effect on dopamine uptake and that the decrease in efflux was not related to a decrease in DA biosynthesis.66,67 However, previous data65 support the hypothesis that ethanol may enhance the activity of the dopamine transporter. Thus, in vivo voltammetry studies have provided insights into how ethanol affects dopamine dynamics in the CNS.

4.4.2 PSYCHOMOTOR STIMULANTS Cocaine and amphetamine are among the most widely studied of all commonly abused drugs. Voltammetric-based studies have revealed insights into the mecha-

2345_frame_MASTER.book Page 102 Wednesday, October 23, 2002 1:03 PM

102

Methods in Drug Abuse Research: Cellular and Circuit Level Analyses

nisms of how these drugs increase catecholamine levels in the CNS by affecting transmitter release and uptake. 4.4.2.1 d-Amphetamine d-amphetamine is thought to increase extracellular DA levels through a reverse transport process mediated by the DAT.68,69 Voltammetric recordings in striatal slices obtained from mice containing a targeted deletion (“knockout”) of the DAT gene demonstrated that d-amphetamine failed to promote DA release.17 However, a subsequent voltammetry study revealed that d-amphetamine could still reduce electrically evoked DA release even in DAT-deficient mice, suggesting that damphetamine’s ability to deplete vesicular DA pools can be dissociated from its effects on DAT activity.70 These studies make clear that d-amphetamine has a complex mechanism of action that includes both depletion of intracellular DA pools and alterations in DAT activity. Consistent with this finding, voltammetric studies have demonstrated that, in contrast to depolarization-induced DA release, d-amphetamine-stimulated release of DA is a slow process, requiring several minutes to attain peak DA levels.45,70 Moreover, voltammetric methods have revealed substantial differences between dendritic and nerve terminal release and clearance of DA produced by either d-amphetamine or K+ depolarization.45,71 For example, in the striatum, d-amphetamine-stimulated peak DA amplitude is about fourfold less than that caused by KCl-induced depolarization, whereas in the substantia nigra there is no difference in the amplitude of these signals.45 This finding is in agreement with previous microdialysis studies44,72 and suggests that Ca2+-dependent and Ca2+-independent release pools may differ between cell bodies and nerve terminals. In support of this hypothesis, we and others have found changes in extracellular Ca2+ affect release in the substantia nigra differently as compared to the striatum.45,71 Although the significance of these differences has yet to be fully explored, especially in relation to the mesolimbic DA system, these studies have been useful in further characterizing the actions of d-amphetamine at dopaminergic synapses. In addition to characterizing the acute effects of d-amphetamine, voltammetric techniques have also been used to assess the effects of chronic d-amphetamine treatment on dopaminergic function. For example, there is evidence that electrically evoked DA release in the ventral tegmental area is enhanced following chronic d-amphetamine treatment.73 Interestingly, these same investigators have also found a reduction in DA release in the striatum and nucleus accumbens (NAc) of rats that have been behaviorally sensitized to d-amphetamine.74 Thus, it appears that, analogous to the differences observed under acute conditions, DA cell bodies and nerve terminals may respond differently to chronic d-amphetamine. In addition, repeated administration of other amphetamine derivatives, such as methamphetamine, reduces both DA release and clearance as measured by high-speed chronoamperometry.75,76 Together these studies demonstrate the power of voltammetric techniques in assessing the effects of acute and chronic administration of d-amphetamine on dopaminergic systems and the regional differences that lie within these neurons.

2345_frame_MASTER.book Page 103 Wednesday, October 23, 2002 1:03 PM

In Vivo Voltammetry in Drug Abuse Research

103

4.4.2.2 Cocaine Although cocaine has similar affinity for all three major catecholamine transporters (DAT, NET, and SERT), many of the behavioral effects of the drug are thought to be due to its inhibition of DA uptake. Voltammetry studies investigating the effects of cocaine and similar uptake inhibitors in the CNS reported results that were not previously seen using microdialysis. For example, it was shown that cocaine increased the amplitude, but not the clearance rate, of KCl-evoked extracellular DA77 (see Figure 4.3B). A similar finding was reported by Justice and colleagues, who found that the DAT inhibitor mazindol increased rather than decreased the clearance rate of DA.78 This paradoxical, “anomalous” effect of uptake inhibition on the dynamics of the DA signal first revealed that the use of voltammetric techniques is still the subject of some debate. However, we have consistently observed this effect in data from both intact animals9 and brain slices (see Reference 43 and Figures 4.3A and 4.3B). We have hypothesized that the failure of cocaine to affect the clearance rate is due, at least in part, to the ability of the DAT to clear DA in a substrate-dependent (e.g., Michaelis-Menten) fashion. Thus, the apparent clearance rate of the voltammetric signal will increase as extracellular DA concentration increases, up to a maximum value.79,80 The spatial resolution permitted by voltammetry has also revealed some interesting differences in the regional effects of cocaine. For example, cocaine appears to have a more pronounced effect on DA uptake in the NAc than in the dorsolateral striatum.81 Similarly, the uptake of DA in the medial dorsal striatum appears to be more sensitive to cocaine than uptake in the dorsolateral striatum.82 More recently, we have found that cocaine may differentially affect DA uptake in the core and shell subregions of the NAc.83 All of these findings suggest that cocaine’s effects on DA uptake are enhanced in regions of lower DAT density and could be explained by a higher fractional occupancy of the DAT by cocaine in brain regions with a lower number of transporter sites. A more rigorous comparison of DA clearance parameters with DAT binding data might yield more definitive information in this regard. Finally, voltammetry-based studies have been used to assess the effects of cocaine on DA release and clearance during paradigms that model drug abuse such as chronic treatment or self-administration. In one study, DA release and clearance in the NAc was transiently enhanced following chronic cocaine administration.84 Another study found no differences in the kinetics of DA uptake in brain slices following chronic cocaine treatment, although the potency of cocaine for inhibiting DA uptake was increased.85 In our laboratory, we found that, following withdrawal from chronic cocaine treatment, DA clearance in response to a challenge dose of cocaine resulted in a decreased clearance rate in the nucleus accumbens but not in the striatum.9 These variable results may reflect differences in the chronic treatment protocols employed or other methodological issues. Nevertheless, it is clear that the use of voltammetric techniques to measure changes in DA dynamics following chronic cocaine treatment or following self-administration of the drug13,14,86 will continue to provide useful information.

2345_frame_MASTER.book Page 104 Wednesday, October 23, 2002 1:03 PM

104

Methods in Drug Abuse Research: Cellular and Circuit Level Analyses

4.4.2.3 THC (Cannabinoids) Like other drugs of abuse, marijuana and its active constituents (cannabinoids) have been demonstrated to enhance the firing rate of midbrain DA neurons.87,88 This increased firing rate is thought to result in enhanced release of DA from terminals in the NAc and may explain some of the rewarding effects of marijuana.89 Consistent with this hypothesis, both microdialysis90,91 and voltammetry studies92 performed in anesthetized animals have demonstrated an enhancement of DA release from striatal and NAc terminals following acute cannabinoid administration. In brain slices, the effects of cannabinoids on DA release have been more variable. An earlier study reported that [3H]-DA release in striatal slices is reduced by cannabinoids,93 suggesting that cannabinoids may presynaptically affect DA release. However, a subsequent FSCV study found that the cannabinoid agonist WIN 55,212–2 had no effect on electrically evoked release in brain slices.94 We have recently observed similar results using high-speed chronoamperometry in striatal slices (Hoffman, unpublished observations). The findings from voltammetry studies are thus consistent with the known distribution of cannabinoid receptors, which are not localized to DAergic neurons or DA nerve terminals.95 Since cannabinoid receptors are known to presynaptically inhibit glutamate release in the dorsolateral striatum,96 it is possible that the earlier report of inhibition of DA release by cannabinoids in striatal slices reflects an indirect effect of cannabinoids on glutamate-driven DA release. Future voltammetric studies should help to address the effects of chronic cannabinoid treatment on dopaminergic function as well as the effects of cannabinoids on other catecholamine systems.

4.5 FUTURE APPLICATIONS The uses of voltammetry for understanding neurotransmitter release and uptake have grown since its inception in the early 1970s. It is now used in the study of neurotransmitter dynamics in single cells, slice preparations, and whole animals. Advances in instrumentation, microelectrode technology, and surface treatments of microelectrodes have expanded the number of neurotransmitters that can be measured by voltammetry. The quantification of the monoamines is well established. The ability to routinely measure other major CNS neurotransmitters is on the horizon. Together with existing methodologies these voltammetric techniques should provide powerful insight into the role of neurotransmitter systems in drug abuse.

ACKNOWLEDGMENTS This work was supported by United States Public Health Service grants NS39787, AG06434 DA14944, and a Level II Research Scientist Development Award (MH01245) to Greg Gerhardt from the National Institute of Mental Health. In addition, we thank the National Science Foundation for generous support through grant #DBI-9730899.

2345_frame_MASTER.book Page 105 Wednesday, October 23, 2002 1:03 PM

In Vivo Voltammetry in Drug Abuse Research

105

ABBREVIATIONS AND ACRONYMS 5-HT Serotonin or 5-Hydroxytryptamine AA Ascorbic acid CE Capillary electrophoresis CFE Carbon fiber electrode CNS Central nervous system CV Cyclic voltammetry DA Dopamine DAT Dopamine transporter DOPAC 3,4-Dihydroxyphenylacetic Acid FCV Fast cyclic voltammetry FSCV Fast-scan cyclic voltammetry HPLC-EC High-performance liquid chromatography coupled with electrochemical detection HSC High-speed chronoamperometry NAD(P)H Nicotinamide adenine dinucleotide phosphate (reduced form) NAD± Nicotinamide adenine dinucleotide (oxidized form) NADH Nicotinamide adenine dinucleotide (reduced form) NE Norepinephrine NET Norepinephrine transporter NO Nitric oxide SERT Serotonin transporter

REFERENCES 1. Adams, R.N., In vivo electrochemical measurements in the CNS, Prog. Neurobiol., 35(4), 297, 1990. 2. Gerhardt, G.A., Rapid chronocoulometric measurements of norepinephrine overflow and clearance in CNS tissues, in Neuromethods — Voltammetric Methods in Brain Systems, Vol. 27, Humana Press, Totowa, NJ, 1995, pp. 117. 3. Kawagoe, K.T., Zimmerman, J.B., and Wightman, R.M. Principles of voltammetry and microelectrode surface states, J. Neurosci. Methods, 48, 225, 1993. 4. Millar, J., Simultaneous in vivo voltammetric and electrophysiological recording with carbon fiber microelectrodes, in Methods in Neurosciences, Vol. 4, Conn, M., Ed., Academic Press, New York, NY, 1991. 5. Schenk, J.O. and Adams, R.N., Chronoamperometric measurements in the central nervous system, IBRO handbooks series, in Methods in the Neurosciences, Vol. 6, Marsden, C.A., Ed., John Wiley & Sons, New York, NY, 1984, pp. 193. 6. Gerhardt, G.A. and Burmeister, J., In vivo voltammetry for chemical analysis of the nervous system, in Encyclopedia of Analytical Chemistry: Instrumentation and Applications, Meyers, R.A., Ed., John Wiley & Sons, Chichester, 2000. 7. Kissinger, P.T., Hart, J.B., and Adams, R.N., Voltammetry in brain tissue: a new neurophysiological measurement, Brain Res., 55(1), 209, 1973. 8. Bunin, M.A. and Wightman, R.M. Quantitative evaluation of 5-hydroxytryptamine (serotonin) neuronal release and uptake: an investigation of extrasynaptic transmission, J. Neurosci., 18(13), 4854, 1998.

2345_frame_MASTER.book Page 106 Wednesday, October 23, 2002 1:03 PM

106

Methods in Drug Abuse Research: Cellular and Circuit Level Analyses

9. Cass, W.A. et al., Reduced clearance of exogenous dopamine in rat nucleus accumbens, but not in dorsal striatum, following cocaine challenge in rats withdrawn from repeated cocaine administration, J. Neurochem., 61(1), 273, 1993. 10. Cragg, S.J., Hawkey, C.R., and Greenfield, S.A., Comparison of serotonin and dopamine release in substantia nigra and ventral tegmental area: region and species differences, J. Neurochem., 69(6), 2378, 1997. 11. Garris, P.A. et al., Real-time measurement of electrically evoked extracellular dopamine in the striatum of freely moving rats, J. Neurochem., 68(1), 152, 1997. 12. Garris, P.A. et al., Dissociation of dopamine release in the nucleus accumbens from intracranial self-stimulation, Nature, 398(6722), 67, 1999. 13. Gratton, A. and Wise, R.A., Drug- and behavior-associated changes in dopaminerelated electrochemical signals during intravenous cocaine self-administration in rats, J. Neurosci., 14(7), 4130, 1994. 14. Gratton, A., In vivo analysis of the role of dopamine in stimulant and opiate selfadministration, J. Psychiatry Neurosci., 21(4), 264, 1996. 15. Daws, L.C. et al., In vivo chronoamperometric measurements of the clearance of exogenously applied serotonin in the rat dentate gyrus, J. Neurosci. Methods, 78(1–2), 139, 1997. 16. May, L.J. and Wightman, R.M., Heterogeneity of stimulated dopamine overflow within rat striatum as observed with in vivo voltammetry, Brain Res., 487(2), 311, 1989. 17. Giros, B. et al., Hyperlocomotion and indifference to cocaine and amphetamine in mice lacking the dopamine transporter, Nature, 379, 606, 1996. 18. Bard, A.J. and Falkner, L.R., Electrochemical Methods: Fundamentals and Applications, John Wiley & Sons, New York, 1980. 19. Adams, R.N., Electrochemistry at Solid Electrodes, Marcel Dekker, New York, 1969. 20. Kume-Kick, J. and Rice, M.E., Dependence of dopamine calibration factors on media ca2+ and mg2+ at carbon-fiber microelectrodes used with fast-scan cyclic voltammetry, J. Neurosci. Methods, 84, 55,1998. 21. Gerhardt, G.A. and Hoffman, A.F., Effects of recording media composition on the response of Nafion-coated carbon fiber microelectrodes measured using high-speed chronoamperometry, J. Neurosci. Methods, 109(1):13, 2001. 22. Gerhardt, G.A. et al., In vivo electrochemical studies of dopamine overflow and clearance in the striatum of normal and MPTP-treated rhesus monkeys, J. Neurochem., 66(2), 579, 1996. 23. Stamford, J.A. et al., Simultaneous “real-time” electrochemical and electrophysiological recording in brain slices with a single carbon-fibre microelectrode, J. Neurosci. Methods, 50, 279, 1993. 24. Rice, M.E. and Nicholson, C., Measurement of nanomolar dopamine diffusion using low-noise perfluorinated ionomer coated carbon fiber microelectrodes and high-speed cyclic voltammetry, Anal. Chem., 61, 1805–1810, 1989. 25. McCreery, R.L., Carbon electrode surface chemistry, in Voltammetric Methods in Brain Systems, Boulton, A.A., Baker, G.B., and Adams, R.N., Eds., Humana Press, Totowa, NJ, 1995, 1–26. 26. Boulton, A.A., Baker, G.B., and Adams, R.N., Neuromethods — Voltammetric Methods in Brain Systems,Vol. 27, Boulton, A.A., Baker, G.B., and Adams, R.N., Eds., Humana Press, Totowa, NJ, 1995. 27. Millar, J., Simultaneous in vivo voltammetric and electrophysiological recording with carbon fiber microelectrodes, in Methods in Neurosciences, Vol. 4, Conn, M., Ed., Academic Press, New York, NY, 1991.

2345_frame_MASTER.book Page 107 Wednesday, October 23, 2002 1:03 PM

In Vivo Voltammetry in Drug Abuse Research

107

28. Gonon, F., Monitoring dopamine and noradrenaline (norepinephrine) release in central and peripheral nervous systems with treated and untreated carbon-fiber electrodes, in Neuromethods — Voltammetric Methods in Brain Systems, Vol. 27, Boulton, A.A., Baker, G.B., and Adams, R.N., Eds., Humana Press, Totowa, NJ, 1995, pp. 117–151. 29. Van Horne, C. et al., Multichannel semiconductor-based electrodes for in vivo electrochemical and electrophysiological studies in rat cns, Neurosci. Lett., 120, 249, 1990. 30. Sreenivas, G. et al., Fabrication and characterization of sputter-carbon microelectrode arrays, Anal. Chem., 68, 1858, 1996. 31. Burmeister, J.J., Moxon, K., and Gerhardt, G.A., Ceramic-based multisite microelectrodes for electrochemical recordings, Anal. Chem., 72(1):187, 2000. 32. Burmeister, J.J. and Gerhardt, G.A., Self-referencing ceramic-based multisite microelectrodes for the detection and elimination of interferents from the measurement of L-glutamate and other analytes, Anal. Chem., 73(5):1037, 2001. 33. Garris, P.A. and Wightman, R.M., Diffusional differences in dopamine release, uptake, and diffusion measured by fast-scan cyclic voltammetry, in Voltammetric Methods in Brain Systems, Boulton, A.A., Baker, G.B., and Adams, R.N., Eds., Humana Press, NJ, 1995, pp. 179. 34. Luthman, J., Friedemann, M.N., Hoffer, B.J., and Gerhardt, G.A., In vivo electrochemical measurements of serotonin clearance in rat striatum: effects of neonatal 6hydroxydopamine-induced serotonin hyperinnervation and serotonin uptake inhibitors, J. Neural Transm., 104(4–5), 379, 1997. 35. Hafizi, S., Kruk, Z.L., and Stamford, J.A., Fast cyclic voltammetry: improved sensitivity to dopamine with extended oxidation scan limits, J. Neurosci. Methods, 33, 41, 1990. 36. Friedemann, M.N., Robinson, S.W., and Gerhardt, G.A., O-Phenylenediamine-modified carbon fiber electrodes for the detection of nitric oxide, Anal. Chem., 68(15), 2621, 1996. 37. Hu, Y.G. and Wilson, S., Rapid changes in local extracellular rat brain glucose observed with an in vivo glucose sensor, J. Neurochem., 68(4):1745, 1997. 38. Garguilo, M.G. et al., Amperometric, sensors for peroxide, choline, and, acetylcholine on electron transfer between horseradish peroxidase and a redox polymer, Anal. Chem., 65, 523, 1993. 39. Garguilo, M.G. and Michael, A.C., Quantification of choline in the extracellular fluid of brain tissue with amperometric microsensors, Anal. Chem., 66, 2621, 1994. 40. Clark, R.A. and Ewing, A.G., Quantitative measurements of released amines from individual exocytosis events, Mol. Neurobiol., 15(1), 1, 1997. 41. Gutierrez, L.M., Gil, A., and Viniegra, S., Preferential localization of exocytotic active zones in the terminals of neurite-emitting chromaffin cells, Eur. J. Cell Biol., 76(4), 274, 1998. 42. Travis, E.R. and Wightman, R.M., Spatio-temporal resolution of exocytosis from individual cells, Annu. Rev. Biophys. Biomol. Struct., 27, 77, 1998. 43. Hoffman, A.F. and Gerhardt, G.A., In vivo electrochemical studies of dopamine clearance in the rat substantia nigra: effects of locally applied uptake inhibitors and unilateral 6-OHDA lesions, J. Neurochem., 70, 179–189, 1998. 44. Hebert, M.A. et al., Functional effects of GDNF in normal rat striatum: presynaptic studies using in vivo electrochemistry and microdialysis, J. Pharmacol. Exp. Ther., 279, 1181, 1996. 45. Hoffman, A.F. and Gerhardt, G.A., Differences in pharmacological properties of dopamine release between the substantia nigra and striatum: an in vivo electrochemical study, J. Pharmacol. Exp. Ther., 289, 455, 1999.

2345_frame_MASTER.book Page 108 Wednesday, October 23, 2002 1:03 PM

108

Methods in Drug Abuse Research: Cellular and Circuit Level Analyses

46. Stamford, J.A. et al., Diffusion and uptake of dopamine in rat caudate and nucleus accumbens compared using fast cyclic voltammetry, Brain Res., 448(2), 381, 1988. 47. Michael, D., Travis, E.R., and Wightman, R.M., Color images for fast-scan cv measurements in biological systems, Anal. Chem., 70(17), 586A, 1998. 48. Millar, J.T. and Barnett, G., Basic instrumentation for fast cyclic voltammetry, J. Neurosci. Meth., 25, 91, 1988. 49. Gerhardt, G.A. et al., Methodology for coupling local application of dopamine and other chemicals with rapid in vivo electrochemical recordings in freely-moving rats, J. Neurosci. Methods, 87, 67, 1999. 50. Rice, M.E. et al., Direct monitoring of dopamine and 5-ht release in substantia nigra and ventral tegmental area in vitro, Exp. Brain Res., 100(3), 395, 1994. 51. Malinski, T. and Taha, Z., Nitric oxide release from a single cell measured in situ by a porphyrinic-based microsensor, Nature, 358, 676, 1992. 52. Ryan, M.R., Lowry, J.P., and O’Neill, R.O., Biosensor for neurotransmitter l-glutamic acid designed for efficient use of l-glutamate oxidase and effective rejection of interference, Analyst, 122, 1419, 1997. 53. Kulagina, N.V., Shankar, L., and Michael A.C., Monitoring glutamate and ascorbate in the extracellular space of brain tissue with electrochemical microsensors, Anal. Chem., 71, 5093, 1999. 54. Hu, Y. et al., Direct measurement of glutamate release in the brain using a dual enzyme-based electrochemical sensor, Brain Res., 659, 117, 1994. 55. Stamford, J.A. and Justice, J.B., Jr., Probing brain chemistry, Anal. Chem., 68(11), 359A, 1996. 56. Su, M.-T., Dunwiddie, T.V., and Gerhardt, G.A. Combined electrochemical and electrophysiological studies of monoamine overflow in rat hippocampal slices, Brain Res., 518, 149–158, 1990. 57. Rudnick, G. and Clark, J., From synapse to vesicle: the reuptake and storage of biogenic amine neurotransmitters, Chim. Biophys. Acta, 1144, 249, 1993. 58. Kawagoe, K.T., Jankowski, J.A., and Wightman, R.M., Etched carbon-fiber electrodes as amperometric detectors of catecholamine secretion from isolated biological cells, Anal. Chem., 63(15), 1589, 1991. 59. Elhamdani, A., Zhou, Z., and Artalejo, C.R., Timing of dense-core vesicle exocytosis depends on the facilitation l-type ca channel in adrenal chromaffin cells, J. Neurosci., 18(16), 6230–6240, 1998. 60. Wise, R.A., Addiction becomes a brain disease, Neuron, 26, 27, 2000. 61. Weiss, F., Parsons, L.H., and Markow, A., Neurochemistry of cocaine withdrawal, in The Neurobiology of Cocaine: Cellular and Molecular Mechanisms, Hammer, R.P., Jr., Ed., CRC Press, Boca Raton, FL, pp. 163–180. 62. Brodie, M.S., Pesold, C., and Appel, S.B., Ethanol directly excites dopaminergic ventral tegmental area reward neurons, Alcoholism, 23(11), 1848, 1999. 63. Weiss, F. et al., Ethanol self-administration restores withdrawal-associated deficiencies in accumbal dopamine and 5-hydroxytrypamine release in dependent rats, J. Neurosci., 16(10), 3474, 1996. 64. Samson, H.H. et al., The effects of local application of ethanol in the N accumbens on dopamine overflow and clearance, Alcohol, 14(5), 485, 1997. 65. Wang, Y. et al., Effects of ethanol on striatal dopamine overflow and clearance: an in vivo electrochemical study, Alcohol, 14(6), 593, 1997. 66. Budygin, E.A. et al., Terminal effects of ethanol on dopamine dynamics in rat nucleus accumbens: an in vitro voltammetric study, Synapse, 42, 77, 2001.

2345_frame_MASTER.book Page 109 Wednesday, October 23, 2002 1:03 PM

In Vivo Voltammetry in Drug Abuse Research

109

67. Budygin, E.A. et al., Effect of acute ethanol on striatal dopamine neurotransmission in ambulatory rats, J. Pharmacol. Exp. Ther., 297(1), 27, 2001. 68. Jones, S.R. et al., Dopamine neuronal transport kinetics and effects of amphetamine, J. Neurochem., 73, 2406, 1999. 69. Sulzer, D., Maidment, N.T., and Rayport, S., Amphetamine and other weak bases act to promote reverse transport of dopamine in ventral midbrain neurons, J. Neurochem., 60, 527, 1993. 70. Jones, S.R. et al., Mechanisms of amphetamine action revealed in mice lacking the dopamine transporter, J. Neurosci., 18(6), 1979, 1998. 71. Chen, B.T. and Rice, M.E., Novel Ca2+ dependence and time course of somatodendritic dopamine release: substantia nigra versus striatum, J. Neurosci., 21(19), 7841, 2001. 72. Hoffman, A.F. et al., In vivo microdialysis studies of somatodendritic dopamine release in the rat substantia nigra: effects of unilateral 6-OHDA lesions and GDNF, Exp. Neurol., 147, 130, 1997. 73. Muscat, R., Iravani, M.M., and Kruk, Z.L., The role of ventral tegmental dopamine neurons in locomotor sensitization following quinpirole or (+)–amphetamine: ex vivo voltammetric evidence, Neuroscience, 75, 1175, 1996. 74. Muscat, R.R. et al., Dissociation of the effects of amphetamine and quinpirole on dopamine release in the nucleus accumbens following behavioural sensitization: an ex vivo voltammetric study, Behav. Pharmacol., 4, 411, 1993. 75. Cass, W.A., Decreases in evoked overflow of dopamine in rat striatum after neurotoxic doses of methamphetamine, J. Pharmacol. Exp. Ther., 280, 105, 1997. 76. Cass, W.A., Manning, M.W., and Dugan, M.T., Effects of neurotoxic doses of methamphetamine on potassium and amphetamine evoked overflow of dopamine in the striatum of awake rats, Neurosci. Lett., 248, 175, 1998. 77. Gerhardt, G.A., Gratton, A., and Rose, G.M., In vivo electrochemical studies of the effects of cocaine on dopamine nerve terminals in the rat neostriatum, Physiol. Biochemoslovaca, 37, 249, 1988. 78. Ng, J.P. et al., Anomalous effect of mazindol on dopamine uptake as measured by in vivo voltammetry and microdialysis, Neurosci. Lett., 134, 229, 1992. 79. Hebert, M.A. and Gerhardt, G.A., Age-related changes in the capacity, rat, and modulation of dopamine uptake within the striatum and nucleus accumbens of Fischer 344 rats: an in vivo electrochemical study, J. Pharmacol. Exp. Ther., 288, 879, 1999. 80. Zahniser, N.R., Larson, G.A., and Gerhardt, G.A., In vivo dopamine clearance rate in rat striatum: regulation by extracellular dopamine concentration and dopamine transporter inhibitors, J. Pharmacol. Exp. Ther., 289, 266, 1999. 81. Cass, W.A. et al., Differences in dopamine clearance and diffusion in rat striatum and nucleus accumbens following systemic cocaine administration, J. Neurochem., 59(1), 259, 1992. 82. Cline, E.J. et al., Medial dorsal striatum is more sensitive than lateral dorsal striatum to cocaine inhibition of exogenous dopamine clearance: relation to [3H]mazindol binding, but not striosome/matrix, Exp. Neurol., 134, 135, 1995. 83. David, D.J. et al., In vivo electrochemical studies of dopamine clearance in subregions of rat nucleus accumbens: differential properties of the core and shell, Exp. Neurol., 153, 277, 1998. 84. Ng, J.P., Hubert, G.W., and Justice, J.B., Jr., Increased stimulated release and uptake of dopamine in nucleus accumbens after repeated cocaine administration as measured by in vivo voltammetry, J. Neurochem., 56(5), 1485, 1991.

2345_frame_MASTER.book Page 110 Wednesday, October 23, 2002 1:03 PM

110

Methods in Drug Abuse Research: Cellular and Circuit Level Analyses

85. Lee, T.H. et al., Altered cocaine potency in the nucleus accumbens following 7-day withdrawal from intermittent but not continuous treatment: voltammetric assessment of dopamine uptake in the rat, Psychopharmacology (Berl.), 137, 303, 1998. 86. Kiyatkin, E.A. and Stein, E.A., Fluctuations in nucleus accumbens dopamine during cocaine self-administration behavior: an in vivo electrochemical study, Neuroscience, 64, 599, 1995. 87. French, E.D., Delta9-Tetrahydrocannabinol excites rat VTA dopamine neurons through activation of cannabinoid CB1 but not opioid receptors, Neurosci. Lett., 226, 159, 1997. 88. French, E.D., Dillon, K., and Wu, X.F., Cannabinoids excite dopamine neurons in the ventral tegmentum and substantia nigra, Neuroreport, 8, 649, 1997. 89. Gardner, E.L. et al., Facilitation of brain stimulation reward by delta 9-tetrahydrocannabinol, Psychopharmacology (Berl.), 96, 142, 1988. 90. Chen, J.P. et al., Delta 9-tetrahydrocannabinol produces naloxone-blockable enhancement of presynaptic basal dopamine efflux in nucleus accumbens of conscious, freelymoving rats as measured by intracerebral microdialysis, Psychopharmacology (Berl.), 102, 156, 1990. 91. Tanda, G., Pontieri, F.E., and DiChiara, G., Cannabinoid and heroin activation of mesolimbic dopamine transmission by a common m1 opioid receptor mechanism, Science, 276, 2048, 1997. 92. Ng, J.C.T. et al., The effects of delta 9-tetrahydrocannabinol on potassium-evoked release of dopamine in the rat caudate nucleus: an in vivo electrochemical and in vivo microdialysis study, Brain Res., 451, 59, 1988. 93. Cadogan, A.K. et al., Influence of cannabinoids on electrically evoked dopamine release and cyclic AMP generation in the rat striatum, J. Neurochem., 69, 1131, 1997. 94. Szabo, B. et al., Effects of cannabinoids on dopamine release in the corpus striatum and the nucleus accumbens in vitro, J. Neurochem., 73, 1084, 1999. 95. Herkenham, M. et al., Neuronal localization of cannabinoid receptors in the basal ganglia of the rat, Brain Res., 547, 267, 1991. 96. Gerdeman, G. and Lovinger, D.M., CB1 cannabinoid receptor inhibits synaptic release of glutamate in rat dorsolateral striatum, J. Neurophysiol., 85, 468, 2001.

2345_frame_MASTER.book Page 111 Wednesday, October 23, 2002 1:03 PM

5

Extracellular Single Unit Recording Strategies for Investigating the Actions of Drugs of Abuse in Anesthetized Animals Barry D. Waterhouse

CONTENTS 5.1 5.2

5.3

5.4

Introduction ..................................................................................................112 Single-Unit Studies in Intact, Anesthetized Preparations ...........................112 5.2.1 Advantages of Intact, Anesthetized Preparations ............................112 5.2.2 Single-Unit Recording and Drug Application .................................114 5.2.3 Experimental Protocols ....................................................................115 5.2.4 Data Analysis ...................................................................................116 5.2.4.1 Quantifying Stimulus-Evoked Responses ........................116 5.2.4.2 Quantifying Transmitter-Induced Responses ...................117 5.2.4.3 Examining Drug-Induced Changes in Spontaneous and Evoked Discharge .............................................................117 5.2.4.4 Measuring Drug-Induced Changes in Response Latency..............................................................................118 5.2.4.5 Characterizing Drug-Induced Changes in Response Threshold and Receptive Field Properties .......................118 5.2.4.6 Interpretation of Results ...................................................119 Using Cellular Electrophysiological Techniques to Study Psychostimulant Drug Action ......................................................................119 5.3.1 Background ......................................................................................119 5.3.2 Psychostimulant Effects on Central Monoaminergic Systems .......120 5.3.3 Monoamine Influences on Sensory-Signal Processing ...................121 Electrophysiological Assays for Evaluating Psychostimulant Actions .......122 5.4.1 Psychostimulant-Induced Suppression of Monoamine Cell Discharge..........................................................................................122 5.4.2 Psychostimulant Actions in Monoaminergically Innervated Brain Circuits ...................................................................................122 111

2345_frame_MASTER.book Page 112 Wednesday, October 23, 2002 1:03 PM

112

Methods in Drug Abuse Research: Cellular and Circuit Level Analyses

5.4.2.1 5.4.2.2 5.4.3.3

Cerebellum........................................................................122 Somatosensory Thalamus and Cortex ..............................125 Cocaine Augments Sensory Cortical Neuron Responses to Afferent Pathway Electrical Stimulation or Glutamate Application........................................................................126 5.4.3.4 Cocaine Selectively Enhances Long-Latency Responses of Cortical Neurons to Sensory Inputs.............................127 5.4.3.5 Cocaine Effects on Somatosensory Thalamic Neurons ...129 5.5 Implications of Single-Unit Drug Studies in Intact, Anesthetized Preparations ..................................................................................................132 Acknowledgments..................................................................................................134 References..............................................................................................................134

5.1 INTRODUCTION Despite the extensive literature on drug-related behaviors and theories of reward and reinforcement, the physiological bases of many other aspects of the actions of abusepotential substances, drug dependence, and development of chronic patterns of substance abuse are not well understood. These include drug craving, tolerance, sensitization, and withdrawal. A better delineation of the physiological mechanisms influenced by recreational drugs and the neural network functions that are biased by acute and chronic patterns of substance abuse may lead to more rational and effective therapies for drug addiction. In order to obtain information that is relevant to behavioral circumstances, i.e., when all potential sites of drug action in the CNS are accessed simultaneously via the blood stream, drug effects on single neurons and local circuits must be evaluated in situ and not inferred from membrane or molecular actions of drugs. The ideal test situation is one in which the spike train activity of single neurons can be monitored in intact behaving animals as drug-induced changes occur. Such a preparation allows for examination of the time course of drug-induced modifications in single-cell and neural-circuit response properties under otherwise normal physiological conditions. A significant drawback of this approach is that injected compounds may alter cognitive states or produce motor behaviors that can indirectly affect the area or areas of the CNS being studied. For this reason such experiments should be accompanied by comparable studies in intact, anesthetized preparations. The advantages and disadvantages of this latter approach are discussed here as a means of emphasizing its role in studies of drugs of abuse.

5.2 SINGLE-UNIT STUDIES IN INTACT, ANESTHETIZED PREPARATIONS 5.2.1 ADVANTAGES

OF INTACT,

ANESTHETIZED PREPARATIONS

There are a number of advantages to using whole-animal preparations for evaluating drug effects on brain circuits. First, cell activity is examined in intact

2345_frame_MASTER.book Page 113 Wednesday, October 23, 2002 1:03 PM

Extracellular Single Unit Recording Strategies

113

neuronal networks that are complete with all normal inputs and intrinsic neuronal and nonneuronal (glial) circuit elements. Second, cells are recorded in the presence of endogenous circulating hormones and other blood- or CSF-borne factors that may regulate their normal function. Finally, with systemic application there is the added advantage that behaviorally relevant doses of drugs can gain access to the brain by normal physiologic channels and, likewise, can be inactivated by intact intrinsic mechanisms. Thus, all areas of the brain are simultaneously exposed to drug for a period of time that is defined by the normal pharmacokinetics for that particular agent. In addition to these general advantages of whole animal preparations for drug abuse studies, there are a number of more specific reasons for utilizing this approach. For example, throughout the duration of an experimental protocol the brain is maintained at a constant, albeit anesthetized, state. This insures that drug-induced changes in motor activity or cognitive state, such as might occur in awake animals, are not factors that could influence the dynamics of the cells or circuits being examined. The possibility that drug injection could alter the state of anesthesia and, thus, indirectly influence cell or circuit physiology can be ascertained by monitoring EEG activity throughout the experiment. It is important to emphasize that studies in anesthetized animals can provide important baseline information regarding drug effects on neuronal and network operations under controlled physiologic conditions. From this work it then becomes possible to make inferences about the impact of drugs on cells and circuits in awake animals as a function of changing behavioral conditions. Without prior investigations in anesthetized preparations there are obvious limitations regarding the interpretation of observed electrophysiological effects in awake animals insofar as changes in cell activity may be purely drug-related or secondary to drug-induced changes in behavioral state. Intact, anesthetized preparations also allow for single neurons to be activated by natural, electrical, or chemical stimulation of afferent pathways. Thus, drug actions can be measured within the context of the normal range of signal processing operations for the cell in question. Many initial electrophysiological studies in intact animals ask whether a drug can increase or decrease the spontaneous firing rate of selected neurons. While this is a first step in demonstrating that a drug can act on specific populations of cells, it does not fully delineate all of the parameters of neuronal function that can be influenced by drug actions. Further studies must be undertaken to determine how the cells receive and process afferent synaptic information while the brain is under the influence of a systemically administered compound. An additional advantage of the intact, anesthetized preparation is the ability to determine whether drug effects result from interactions with local circuit mechanisms or circuits that are afferent to, but remote from, the sites being examined. For example, cocaine blocks reuptake of synaptically released dopamine, norepinephrine, and serotonin and therefore can affect neuronal function by actions on monoamine-containing cells, which project broadly throughout the CNS, or by influencing target cells in monoaminergic terminal fields. Systemic drug administration provides the advantage of measuring the effects of system-wide drug actions on local electrophysiological phenomena, whereas with local drug application it is possible to isolate drug influences to the immediate vicinity of the cells or cell being recorded.

2345_frame_MASTER.book Page 114 Wednesday, October 23, 2002 1:03 PM

114

Methods in Drug Abuse Research: Cellular and Circuit Level Analyses

Once the range of a drug’s action on individual cells or circuits has been identified in a whole-animal preparation, numerous pharmacological tools can be used either through systemic administration or local application to identify the transmitter-specific substrates of drug-induced effects. This is particularly useful since drugs of abuse not only have multiple potential anatomical sites of action but also can influence brain activity through interactions with many different transmitter–receptor mechanisms. Finally, intact, anesthetized animal preparations allow for identification and characterization of multiple electrophysiological measures of cell function following acute drug administration. Once these parameters of drug action are firmly established it is then possible to look for alterations in neuronal function before and after chronic drug treatments. Comparisons of cellular function in naive, chronically treated or drug-withdrawn animals can provide insights into the physiological bases for the dependence, craving, tolerance, and sensitization phenomena that are observed with abuse-potential compounds.

5.2.2 SINGLE-UNIT RECORDING

AND

DRUG APPLICATION

Extracellular activity can be recorded from single cells in almost any area of the brain after induction of surgical anesthesia and fixation of the animal in a stereotaxic device. After removal of the skull and dura, recording electrodes (glass or metal) are driven by microdrive devices to specific sites in the brain according to a standardized coordinate system referenced to a brain atlas and the stereotaxic frame. Animals are respirated or allowed to breathe spontaneously over the course of the experiment while body temperature is maintained at 36 to 37°C with a heating lamp or heating pad. Under these conditions viable preparations can be maintained for 8 h or longer. Spike train activity from individual neurons can be electrically isolated and monitored for minutes to hours depending upon the stability of the preparation. Spontaneous and stimulus-evoked patterns of discharge can be characterized for individual neurons before, during, and after systemic, intracerebroventricular, or local (microiontophoresis, micropressure) drug administration in order to identify drug-induced changes in cell function. With microiontophoresis or micropressure-ejection techniques multibarrel glass micropipets are used not only to record the extracellular activity of central neurons but also for applying minute quantities of drug directly within their immediate environment by current or pressure ejection.14,44,45,99,105 Multibarrel micropipets are commercially available or can be fabricated in the laboratory.80,101 Chemical substances are applied by iontophoresis or micropressure to eliminate many of the diffusional and enzymatic barriers restricting the access of drugs to neuronal receptors and, most importantly, to restrict the site of drug action to the neuron from which responses are being recorded. The center barrel of the micropipet is filled with 4 M NaCl and used for recording extracellularly from single neurons. Three side barrels are filled by gravity with drug solutions appropriate to the design of each neuropharmacological test. In the case of microiontophoresis, a constant-current source provides ejection and retaining currents for the drug barrels and automatically passes an equal current

2345_frame_MASTER.book Page 115 Wednesday, October 23, 2002 1:03 PM

Extracellular Single Unit Recording Strategies

115

of opposite polarity through a balance barrel containing 3 M NaCl. Ejection-currentbalancing techniques are used to neutralize tip potentials that can directly influence cell discharge.98 Positive and negative currents are also passed through this “balance” barrel independently to check for possible current artifacts. Artifacts of drug pH and local anesthetic actions are controlled for by conventional techniques described in the literature.46 The microiontophoresis pump (i.e., constant-current source), based upon the design of Geller and Woodward,37 is commercially available (BH-2, Medical Systems) and can be controlled by outputs from a digital computer so that uniform pulses of drug can be passed at regular intervals through the drug barrels.72,111 One or more channels of the microiontophoresis machine are designed to deliver constant pressure to individual barrels of the pipet for micropressure drug application. This method of drug delivery is advantageous for compounds such as neuropeptides that do not readily ionize in solution.

5.2.3 EXPERIMENTAL PROTOCOLS Extracellular activity is recorded from single cells in halothane-anesthetized rats. Once a cell has been electrically “isolated” from background noise, the general features of its response to afferent synaptic input (usually stimulation of peripheral receptive fields or electrical activation of input pathways) or putative transmitter application can be evaluated. Following this initial characterization drug administration (systemic or local) can begin. Unit responses to a fixed number of stimulus presentations are recorded in the form of peri-event or post-stimulus time histograms (PSTHs) and cumulative raster records of spike train activity. The primary goal of this approach is to measure the effects of drug administration on spontaneous and stimulus-evoked activity of a single neuron. Where appropriate, a more thorough characterization of drug effects on other parameters of cell function (e.g., receptive field structure, sensitivity to stimulus intensity) is possible. Drug is administered systemically by intravenous injection through an indwelling venous catheter or intraperitoneally. Once the drug has been injected, spontaneous and evoked activities are measured continuously until the cell demonstrates “recovery” from drug effects as judged from inspection of online rasters and PSTHs. Recovery means that the cell has returned to its predrug level of spontaneous discharge or response to afferent input. In cases where responses to a range of stimulus parameters are measured, all post-drug response histograms are collected within a time frame that permits observation of cell activity at a time when drug levels are peaking and fairly stable. In the event that recovery is not observed, the possibility must be considered that the drug in question produces long-lasting or irreversible changes in cell responsiveness. Such changes are difficult to confirm under our experimental conditions. Nevertheless, throughout each experimental session, spike height, spike waveform and frequency of discharge are monitored to ensure stable and well-isolated, extracellular recordings of individual units. If these conditions are maintained throughout the pre- and post-drug-injection periods, we are inclined to attribute long-lasting changes in cell responsiveness to drug administration rather than spurious fluctuations in cell excitability.

2345_frame_MASTER.book Page 116 Wednesday, October 23, 2002 1:03 PM

116

Methods in Drug Abuse Research: Cellular and Circuit Level Analyses

To determine the linearity and lowest threshold of drug effects, drugs can be tested across a range of cumulative systemic doses. In such studies, drug is administered in a cumulative dose manner (vol. = 0.01 to 0.24 ml/dose) such that each injection doubles the previously administered dose until drug effects are noted in rasters and PSTHs. Additional doses of drug are injected with the requirement that the cell has recovered from the effects of any previous drug injection or that a minimum interval of 30 min has elapsed. As a control, systemic drug administration is preceded by a volume-equivalent injection of 0.9% saline. Also, EEG, blood pressure, and heart rate can be monitored routinely to control for nonspecific drug effects. As with all experiments employing this protocol, only one cell per animal can be studied. For microiontophoretic protocols drug is applied continuously from the pipet once a stable baseline of cell activity has been achieved. This includes not only spontaneous firing rate but also discharge patterns elicited by iontophoretically applied pulses of putative transmitter substance or activation of an afferent synaptic input pathway. Pharmacological antagonists may be administered in a continuous mode from additional barrels of the micropipet as a means of blocking drug actions and identifying the receptors responsible for mediating drug effects. In most cases, following drug interactions, cell activity is continuously monitored until a control pattern of response is reestablished and “recovery” of the cell from drug effects is declared.

5.2.4 DATA ANALYSIS Cumulative raster and histogram records are quantitatively examined to assess drug-induced changes in spontaneous vs. evoked activity of recorded neurons. A simplistic prediction is that drug administration will result in an increase or decrease of spontaneous cell firing and corresponding linear change in the magnitude of evoked discharge. On the other hand, as we have shown previously, norepinephrine,72,111 serotonin,109 and cocaine49 actions in intact neuronal circuits are complex, and simple linear relationships between effects on spontaneous and evoked firing do not exist. For example, some of the most robust effects that we have seen in sensory cortical and thalamic neurons have been marked enhancements of stimulus-evoked discharge at doses of cocaine subthreshold for producing direct suppression or elevation of cell firing. Moreover, in many cases, cocaine has revealed prominent responses to otherwise subthreshold synaptic inputs while having a lesser effect on responses to threshold stimulation of the same afferent pathway. Thus, by necessity the analysis must be highly descriptive of cellular responses to drug over a range of dosages. 5.2.4.1 Quantifying Stimulus-Evoked Responses Computer-generated (Spike 2 note company) cumulative rasters and post-stimulus time histograms (PSTHs) are used to characterize stimulus-related responses and quantitate evoked activity as 1) spikes/stimulus (excitations only) or 2) percentage of baseline spontaneous firing rate (excitations and inhibitions). Rasters and histograms are computed before, during, and after systemic or local drug application. Equal numbers of stimuli are used to generate each histogram.

2345_frame_MASTER.book Page 117 Wednesday, October 23, 2002 1:03 PM

Extracellular Single Unit Recording Strategies

117

For each cell, changes in evoked and spontaneous activity are computed by comparing discharges in identical portions of histograms computed during control and drug-interaction periods. The period of spontaneous activity is selected to be at least 100 msec in duration, 250 msec after stimulus presentation and free of obvious evoked responses. Discharge rates in each spontaneous and evoked period are calculated as described above. These rates computed from control (predrug) and drug(post-injection) interaction histograms are then compared and the difference expressed as percentage suppression or potentiation from control response. Spikesper-stimulus comparisons are made between histograms in a similar fashion. Changes in stimulus-bound activity between control and drug periods are statistically assessed using a one-way analysis of variance test. 5.2.4.2 Quantifying Transmitter-Induced Responses Computer-generated (Spike2, CED, Cambridge, UK) perievent histograms are used to compute the average agonist response to repeated applications of putative transmitter agent or transmitter analog. Histograms are computed before, during, and at intervals after drug administration. To quantify the agonist drug response, the discharge rate during putative transmitter application is compared with the rate between putative transmitter pulses and the differences expressed as percent inhibition or excitation of background activity, accordingly. Differential changes in putative transmitter-induced and spontaneous firing that result from systemic or local drug administration are assessed by comparing discharge rates during identical epochs of spontaneous and transmitter-induced activity in control and drug-interaction histograms. To facilitate comparisons between histograms equal numbers of agonist drug applications are routinely used for each. One-way analysis of variance tests are used to statistically assess the influence of drug administration on neuronal responsiveness to transmitter application. Quantitative changes in spontaneous firing rate and transmitter or synaptically evoked responses can be plotted independently as a function of dose or time after drug administration. The time course of changes in cell activity would be expected to follow the time course of biochemical and behavioral actions of the drug. The potency of a drug for suppressing spontaneous firing rate can be expressed as the half-maximal (50%, SD 50) and maximal (100%, SD 100) systemic doses for producing this effect. Likewise, if a drug causes increases in cell firing, potency for this effect can be expressed in terms of dosage necessary for 50% and 100% elevation of firing rate over control levels. 5.2.4.3 Examining Drug-Induced Changes in Spontaneous and Evoked Discharge In our previous studies of monoamine actions in rat cerebral cortex, we have also found quantitative comparisons of firing rate during equivalent periods of spontaneous and evoked discharge before and after drug administration to be very useful in revealing consistent trends in drug influences on cell responsiveness to synaptic inputs. Accordingly, for each cell tested drug-induced changes in evoked discharge

2345_frame_MASTER.book Page 118 Wednesday, October 23, 2002 1:03 PM

118

Methods in Drug Abuse Research: Cellular and Circuit Level Analyses

can be plotted as a function of change in spontaneous firing rate. The values plotted in this graphic representation are calculated from histograms generated during control periods and at peak time of the drug’s observable effect on cell responsiveness. This analysis makes no assumptions concerning the linearity or nonlinearity of observed effects but rather provides a convenient means for graphic representation of data from many individual cells recorded under the same experimental conditions. In our previous work, we determined that a 15% or greater change in spontaneous vs. stimulus-evoked discharge constituted a statistically significant alteration in cell responsiveness. In practice, we have been able to categorize drug influences on evoked responses as facilitating, suppressive, or no-net-change based on these criteria. In this way, the effects of a range of doses of drug on spontaneous vs. evoked discharge can be compared for a number of individual cells in a sample population. 5.2.4.4 Measuring Drug-Induced Changes in Response Latency Based on results obtained in rat ventral posterior medial thalamus97 we might also expect exogenous compounds to elicit changes in response latency. The demonstration that such changes occur would imply a drug-induced alteration in the speed of signal transmission throughout the neural network being studied. Individual and cumulative raster records and PSTHs are examined for drug-induced shifts in the latency of cell responses. In PSTHs, response latency is defined as the time after stimulus presentation at which cell discharge equals one half its maximal rate during the response epoch. Results from 15 to 20 cells per dose of drug are compared by Student’s t-test or chi-square analysis to establish statistical significance for all drug-induced actions. 5.2.4.5 Characterizing Drug-Induced Changes in Response Threshold and Receptive Field Properties Increases in the magnitude or probability of discharge of neurons in response to stimulus presentation suggest a more reliable or secure transmission of signals along afferent pathways. A combination of these effects suggests a net facilitation of signal transfer along an input pathway. Yet another experimental protocol can determine the extent to which a drug alters neuronal responses to optimal vs. nonoptimal activation of synaptic input pathways. In this paradigm, PSTHs of cell responses to varying levels of stimulus strength are quantitatively compared to assess changes in stimulus detection by the neuron. By definition extracellular recording measures responses that are necessarily above threshold for action-potential generation. However, a paradigm where drug administration is interacted with neuronal responses to varying stimulus strengths (i.e., subthreshold, threshold, suprathreshold) is capable of demonstrating shifts in the threshold of detection of afferent signals. Comparisons of averaged responses to multiple stimulus presentations reduces the chance that spurious fluctuations in the strength of synaptic transmission might be mistaken for drug-induced changes in neuronal responsiveness. In previous studies where prominent excitatory responses to otherwise subthreshold stimulation of the thalamocortical pathway have

2345_frame_MASTER.book Page 119 Wednesday, October 23, 2002 1:03 PM

Extracellular Single Unit Recording Strategies

119

been revealed during norepinephrine administration,108,113 the effect was striking, requiring little quantitative confirmation that a significant change had occurred. Depending on the relative strength of a drug’s effect on excitatory and inhibitory processes along afferent sensory pathways, receptive fields may increase or decrease in size after drug administration. Other changes in receptive field properties of sensory neurons may be manifested as changes in direction or velocity preference. 5.2.4.6 Interpretation of Results In previous studies,10,49,96 we based our evaluation of specific hypotheses regarding drug effects on sensory signal processing on good-quality recordings from 20 to 25 cells/experimental protocol. This expectation derives from our previous experience with endogenous monoaminergic compounds110–113,115 and is consistent with our ability to obtain stable recordings from either thalamic or cortical neurons in anesthetized animals. The major drawback of single-unit recording and systemic drug administration in anesthetized animals is also a major strength, i.e., the potentially broad-reaching effects of a systemically administered agent on signal processing at multiple levels of a functional neural network. For example, with systemic drug administration it is unlikely that the exact level of a sensory pathway (e.g., brainstem, thalamus, neocortex) that is affected by drug administration can be pinpointed. However, by measuring the response properties of cells at the cortical level we can evaluate the net impact of a drug on the feature-extraction properties of the entire sensory-signal-processing network. An additional problem with systemic drug administration is that effects observed at single cells may result from actions occurring at sites remote from, but afferent to, the recording location. For this reason experiments employing local iontophoretic or pressure ejection of the drug in question can provide a critical link in determining if drug actions on cell responsiveness are mediated by local or remote influences.

5.3 USING CELLULAR ELECTROPHYSIOLOGICAL TECHNIQUES TO STUDY PSYCHOSTIMULANT DRUG ACTION 5.3.1 BACKGROUND For reasons that are not entirely clear, psychostimulants such as cocaine and amphetamine are particularly notorious for their abuse liability. These drugs are widely acknowledged as the most potent behavior-reinforcing compounds known.13,36,39,40,50 Though they are not physically addicting in the same sense as opiates, the drugseeking and drug-taking activities associated with these agents can become allconsuming for the addicted individual once the psychostimulant “habit” has been established. Furthermore, the success rate for rehabilitating cocaine and amphetamine addicts even after detoxification and intensive drug counseling programs is alarmingly low.39,40,51,64 Drug-free graduates of such treatment programs report an overwhelming desire to re-initiate cocaine self-administration after they have returned to their previous drug-taking “home” environment or experience drug-

2345_frame_MASTER.book Page 120 Wednesday, October 23, 2002 1:03 PM

120

Methods in Drug Abuse Research: Cellular and Circuit Level Analyses

taking “cues” that activate craving.24,39,40,90,120 These observations suggest a reciprocal, reinforcing link between sensory/behavioral stimuli and drug ingestion.93,103 A central issue that must be addressed in relation to psychostimulant abuse is the neurobiological basis of drug craving. Acutely, cocaine and amphetamine produce an array of CNS effects some or all of which may contribute to a positive drug experience and the desire to repeat that experience. Concerted efforts have been mounted to identify the neural substrates underlying cocaine’s and amphetamine’s “desirable” effects and the processes linking them to the development of drug craving. Clarification of the broad spectrum of central physiological actions accompanying cocaine and amphetamine self-administration can lead to a better understanding of the reasons for their recreational use as well as help explain the development of drug craving that accompanies chronic use of these compounds. Individuals who self-administer psychostimulants describe a sense of well being and euphoria accompanied by arousal, enhanced sensory perception, hyperactivity, and increased capacity for mental and physical work.15,24,32,35,39,40 In laboratory animals, cocaine and amphetamine cause increased locomotion, arousal, and stereotyped behavior89,90,121 as well as increased sensitivity (as measured by motor responsiveness) to environmental changes including auditory, visual, and tactile stimuli,100,118 electrical shock,59 and temperature changes.85

5.3.2 PSYCHOSTIMULANT EFFECTS MONOAMINERGIC SYSTEMS

ON

CENTRAL

At the cellular level cocaine blocks reuptake of the endogenous monoamine transmitters dopamine (DA), norepinephrine (NE), and serotonin (5-HT). Amphetamine causes spontaneous release of central monoamines from storage sites in nerve terminals.38 Regardless of the specific mechanism, the net effect of either of these actions is increased synaptic levels of NE, DA, and 5-HT in neural networks where monoamine-containing fibers terminate. Thus, a predicted outcome of cocaine’s neurochemical actions would be enhanced monoaminergic function in areas of the brain innervated by NE, DA, and 5-HT fibers. A number of studies31,55,58,91,92,119 have concentrated on identifying the “reward” pathways, “pleasure” centers, and neurotransmitters that mediate reinforcement of drug-seeking behavior. In general, the central monoaminergic pathways and their efferent targets have been the focus of this effort; and, in the case of cocaine and amphetamine, the preponderance of evidence119,120 favors involvement of dopaminergic mechanisms in conveying drug-induced “reward signals” to appropriate “reward centers” in the brain. Activation of this intrinsic reward system presumably accounts for much of the euphoria associated with psychostimulant self-administration. Nevertheless, the specific dimensions of reward circuit function that are influenced by acute administration of psychostimulant drugs and the physiological processes that can be altered and ultimately changed to promote and support persistent drug seeking behaviors have not been identified. Moreover, the fact remains that cocaine and amphetamine can potently elevate levels of NE31,41,54 and 5-HT95 at central synapses. Accordingly, one of the most conspicuous gaps in our knowledge is the lack of information concerning the physiological consequences of cocaine and

2345_frame_MASTER.book Page 121 Wednesday, October 23, 2002 1:03 PM

Extracellular Single Unit Recording Strategies

121

amphetamine actions in noradrenergically and serotonergically innervated circuits of mammalian brain. Related to this issue is the fact that the neurobiological substrates underlying other major CNS effects of cocaine and amphetamine, such as enhanced sensitivity to sensory stimuli, have not yet been identified. Such an effect on perceptual processes is by itself a “recreationally desirable” end point of psychostimulant selfadministration and as such must contribute to the overall “pleasurable experience” derived from these compounds. Previous reviews50,51,119 of cocaine’s abuse liability have noted that any positive effects associated with cocaine ingestion may facilitate the reinforcing actions of the drug and, thus, add to its addictive potential. Furthermore, it has recently been argued93,103 that psychostimulant-induced euphoria confers “incentive salience” to stimuli or events associated with drug taking. Thus, over time drug-associated activities and environmental cues become salient and contribute to drug craving. Despite this hypothesis, a survey of the literature reveals a paucity of studies aimed at identifying cocaine’s or amphetamine’s effects on sensory-information processing within the mammalian brain. In fact, up until 15 years ago there was little information available regarding the impact of drugs of abuse on electrophysiological parameters of cell function in any circuit of the brain.

5.3.3 MONOAMINE INFLUENCES

ON

SENSORY-SIGNAL PROCESSING

In the past 25 years, much has been learned regarding the neurobiology of central monoaminergic systems,3,4,33,48,66,67,107,122 thus providing a new conceptual framework for evaluating the actions of agents that are known to interact with these systems. For example, based on numerous electrophysiological studies employing single-unit extracellular recording and microiontophoretic drug-application techniques in intact animal preparations, a low-threshold, modulatory role for NE in complex neuronal network operations has been proposed.34,72,94,99,111,115 The NEinduced modulatory effects that have been reported are distinct from the classic high-threshold, direct-depressant action of this monoamine on spontaneous firing rate.46,57,77,104 Specifically, iontophoretically or synaptically released NE at doses subthreshold for producing direct suppression of spontaneous discharge can facilitate neuronal responses to both excitatory and inhibitory synaptic inputs.72,94,102,115 Moreover, these effects exhibit pharmacological specificity with respect to adrenergic receptor subtypes and putative neurotransmitters.76,110–112 These findings, in combination with other data from anatomical and physiological studies of locus coeruleus,6,7,11,33,61,75 have led to the suggestion that a primary role of the central noradrenergic system is to facilitate nonmonoamine synaptic transmission in target neuronal circuits during periods of increased behavioral arousal and sustained attention. A less extensive series of studies concentrating on the physiology of 5-HT in sensory neocortical circuits has revealed a net depressant action of this monoamine on cell responses to synaptic stimuli.109 In addition, we68,108 and others53,56,99 have determined that both 5-HT and NE can produce selective changes in the receptive field properties of individual sensory cortical neurons. Thus, a tenable working hypothesis is that synaptically released NE and 5-HT may not only up- and down-

2345_frame_MASTER.book Page 122 Wednesday, October 23, 2002 1:03 PM

122

Methods in Drug Abuse Research: Cellular and Circuit Level Analyses

regulate, respectively, the responsiveness of cortical cells to incoming sensory information but also fine-tune the feature-extraction properties of these sensory neurons.

5.4 ELECTROPHYSIOLOGICAL ASSAYS FOR EVALUATING PSYCHOSTIMULANT ACTIONS In conjunction with the abovementioned studies, a sophisticated set of electrophysiological assays for evaluating single neuron function in intact circuits of anesthetized animals was established. Many of these experimental approaches have been used in our laboratory to identify the dimensions of cell and neural network function that are altered by psychostimulant (cocaine or d-amphetamine) actions.10,49,69,70,114 Similar approaches have been used with great success by White and colleagues1,2,30,42,43,65,86,123 and Cunningham and colleagues26,82 to characterize psychostimulant actions in dopaminergic and serotonergic circuits, respectively.

5.4.1 PSYCHOSTIMULANT-INDUCED SUPPRESSION OF MONOAMINE CELL DISCHARGE Because of their ability to elevate central synaptic levels of NE, 5-HT, and DA and activate autoceptors on monoamine-containing cell bodies, cocaine and amphetamine have the potential to regulate the output of monoaminergic nuclei. In fact, single-unit extracellular studies by Cunningham and Lakoski,25,26 Pitts and Marwah,87,88 and Einhorn et al.30 have demonstrated depressant effects of low doses of cocaine on rat dorsal raphe (5-HT), locus coeruleus (NE), and ventral tegmental area (DA) neurons, respectively. Since NE-, DA- and 5-HT-containing axons from these brainstem nuclei distribute broadly throughout forebrain, brainstem, and spinal cord,5,48,52,60–63,73–75,106 any suppression of the otherwise tonic firing rates of these monoaminergic cells could have a broad impact on neural circuit operations throughout the CNS.

5.4.2 PSYCHOSTIMULANT ACTIONS IN MONOAMINERGICALLY INNERVATED BRAIN CIRCUITS 5.4.2.1 Cerebellum A second potential site of psychostimulant action is within monoaminergic terminal field regions themselves. By elevating extracellular levels of NE, DA, and 5-HT, amphetamine or cocaine would be expected to elicit monoamine-like modulatory actions (see above) in monoaminergically innervated reasons of the forebrain, brainstem, and spinal cord. It is also possible that increased synaptic levels of NE, DA, or 5-HT could exert high-threshold direct depressant effects on the spontaneous firing rate of target neurons in these areas.45–47,77,79,81, 104 Figure 5.1 illustrates the results of one experiment designed to test these predictions. Ratemeter and perievent histogram records (Figure 5.1) illustrate the responses of a single cerebellar Purkinje neuron to regularly spaced, uniform iontophoretic pulses of the putative inhibitory transmitter gamma-amino butyric acid (GABA). Cell activity was monitored in the anesthetized preparation before and after systemic administration of cocaine (1.0 mg/kg, i.p.).

2345_frame_MASTER.book Page 123 Wednesday, October 23, 2002 1:03 PM

Extracellular Single Unit Recording Strategies

123

GABA 7nA 35

–19%

35

SA 1

R1

NaCl

–14%

COCAINE HYDROCHLORIDE (1 mg/kg. ip)

0 - 8 MINUTES POST INJECTION –67%

SA 2

8 - 16

R2 –56%

16 - 20

60 sec –15%

10 sec

FIGURE 5.1 Low-dose cocaine augments cerebellar Purkinje cell response to GABA. Continuous ratemeter (left) and corresponding perievent histogram (right) records illustrate the response of a Purkinje neuron to iontophoretic pulses of GABA 7 nA (solid bars) before and after i.p. injections of NaCl and cocaine HCl (1.0 mg/kg). Each perievent histogram sums cell activity during five consecutive GABA administrations beginning with the last five before NaCl administration and then at 4 min post-NaCl and 2,8, and 16 min post-cocaine injection. Quantitative analysis of these histograms indicated that GABA-induced depression of spontaneous discharge was minimally affected by NaCl but was markedly enhanced over control levels within 2 min after cocaine administration, from 19 to 67% inhibition of firing rate. Gradual recovery to the control condition was observed over a period of 16 min post-drug injection. (From Waterhouse, B. et al., Brain Res., 546, 297–309, 1991. With permission.).

In essence, the iontophoretic delivery of GABA was substituted for synaptic release of GABA from axon terminals. As such, the experiment tests for a primary action of cocaine on presynaptic release mechanisms and evaluates cocaine’s interaction with a specific putative inhibitory transmitter. The results show that GABA-induced suppression of Purkinje cell discharge is markedly augmented following cocaine injection. This effect occurs with minimal alteration in Purkinje cell spontaneous firing rate and persists for approximately 15 min after cocaine administration. This result indicates that systemically administered cocaine produces an NE-like enhancement of neuronal responsiveness to inhibitory synaptic stimuli. Further, these findings argue for a postsynaptic site of drug action in enhancing neuronal responsiveness to synaptic input since GABA was applied directly to the cell via microiontophoresis. In a second experiment, the responses of a single Purkinje neuron to iontophoretic pulses of GABA were monitored before, during, and after microiontophoretic application of cocaine. Ratemeter and perievent histogram records again illustrate the facilitating effect of cocaine on GABA-induced suppression of Purkinje

2345_frame_MASTER.book Page 124 Wednesday, October 23, 2002 1:03 PM

124

Methods in Drug Abuse Research: Cellular and Circuit Level Analyses

Cocaine 10nA GABA 7nA

40

Spikes per Second

20

40 sec

Control -45%

Cocaine -56%

Recovery -47%

40

10 sec

FIGURE 5.2 Effects of iontophoretically applied cocaine on cerebellar Purkinje neuron responses to GABA. Continuous ratemeter (top) and corresponding perievent histograms (bottom) show the responses of a single Purkinje cell to iontophoretic pulses of GABA 7 nA (solid bars) before, during, and after continuous microiontophoresis of cocaine HCl 1 nA (broken bar). Each histogram sums unit activity during five consecutive GABA applications. Solid lines and numbers above histograms indicate the duration of the GABA ejection pulse and percentage decrease in spontaneous discharge induced by GABA application, respectively. During cocaine iontophoresis background firing rate remained constant while activity during the GABA response period was reduced, thereby yielding an enhancement (from 45 to 56%) of GABA’s inhibitory action. Recovery to the control pattern of response was observed following termination of the cocaine ejection current. (From Waterhouse, B. et al., Brain Res., 546, 297–309, 1991. With permission.).

cell discharge. As before, the sponstaneous firing rate of the cell was only minimally affected by cocaine administration. This result indicates that locally applied cocaine can enhance responsiveness of individual Purkinje neurons to inhibitory synaptic stimuli. Thus, although systemic effects of cocaine on sites afferent to but remote from the cerebellum cannot be ruled out entirely as potential sites of drug action, it is clear that at least a component of cocaine’s modulatory actions are mediated by mechanisms within the immediate vicinity of the recorded neuron since the drug was applied locally via microiontophoresis. Similar results have been obtained with d-amphetamine.69 Moreover, d-amphetamine has been shown to enhance Purkinje neuron responses to stimulus-evoked excitation. Overall, such drug-induced actions at synapses throughout the cerebellum could facilitate the operation of this circuit and thus provide a physiological basis for the enhancement of motor performance observed with low doses of cocaine and d-amphetamine. Furthermore, insofar as the cerebellum serves as a model system for testing the actions of neuropharmacologic agents,12 these results may be extrapolated to other monoaminergic target regions of the brain where enhancement of synaptic efficacy would likewise have functional implications for neural circuit operations and behavioral responses. In order to further characterize the physiological actions of psychostimulants on individual neurons in central circuits, several additional questions can be

2345_frame_MASTER.book Page 125 Wednesday, October 23, 2002 1:03 PM

Extracellular Single Unit Recording Strategies

125

addressed using combinations of extracellular single-unit recording, microiontophoresis, and systemic drug administration. For example, are changes in neuronal responsiveness to synaptic inputs or putative amino acid transmitters evident with systemic administration of cocaine at doses that are known to elicit behavioral responses? What is the dose/response relationship and time course of such changes? Are the cocaine actions observed thus far dependent upon endogenous stores of DA, NE, or 5-HT or a unique combination of these monoamines? How does cocaine influence response threshold of individual Purkinje neurons? Many of these issues were addressed in subsequent studies for cocaine using the rat trigeminal somatosensory system as a model.8,10,49,114 5.4.2.2 Somatosensory Thalamus and Cortex There are three compelling reasons to investigate the actions of cocaine in sensory systems of the mammalian brain. First, studies in both humans and animals indicate that acutely administered cocaine can produce transient alterations in sensory perception. Second, cocaine has prominent physiological interactions with monoaminergic systems that heavily innervate sensory pathways. Third, these monoaminergic systems have been shown to modulate response threshold and receptive field properties of sensory neurons. Although it is reasonably well established that cocaine produces intense euphoria and reinforces drug-taking behavior via influences on dopaminergic circuits, it may also act through noradrenergic and serotonergic systems to significantly enhance specific dimensions of sensory signal transmission. Over time such concomitant influences on reward circuits and sensory pathways may be mutually reinforcing to the extent that drug-associated sensory experiences become conditioned stimuli that trigger or intensify drug craving. The fundamental question that has been addressed by us in recent studies is how cocaine affects the transmission of afferent signals through primary thalamocortical sensory circuits in rat brain. The specific hypothesis tested was that cocaine should produce monoamine-like modulation of cortical and thalamic neuronal responsiveness to afferent sensory signals because of its well-established interactions with monoaminergic systems. The first goal of this work was to characterize and quantify the effects of acute cocaine on the transmission of sensory signals to cerebrocortical neurons. The rodent somatosensory pathway served as a model for these studies. Experiments were conducted in anesthetized rats and employed combinations of single-cell extracellular recording, systemic drug administration, electrical activation of thalamic input pathways, mechanical stimulation of the mystacial vibrissae, direct iontophoretic application of putative transmitter substances, and computer-assisted analysis of cumulative raster and perievent histogram records to quantitatively assess the responsiveness of individual cortical neurons to synaptic inputs before and after injection of behaviorally relevant doses of cocaine. A second goal was to address similar questions in the thalamic circuits that transform and relay afferent information to primary somatosensory cortical areas. The general aim of these studies was to identify the parameters of sensory thalamic neuron function that can be altered by acute systemic administration of cocaine.

2345_frame_MASTER.book Page 126 Wednesday, October 23, 2002 1:03 PM

126

Methods in Drug Abuse Research: Cellular and Circuit Level Analyses

GLU 60 nA

COCAINE 20 nA

FIRING RATE (Hz)

10

0 40.0

138.0

CONTROL 10

232.0

328.0

Cocaine 20 nA

424.0

RECOVERY

GLU 60 nA 957%

3000%

1150%

5

FIGURE 5.3 Local application of cocaine potentiates glutamate-evoked excitation in sensory cortical neurons. Ratemeter (above) and corresponding histogram (below) records illustrate the response of a single somatosensory cortical neuron to uniform iontophoretic pulses of glutamate (GLU), 60 nA (solid bars) before, during, and after a period of continuous cocaine microiontophoresis (20 nA). During drug administration (broken bar above ratemeter and histogram) glutamate-induced excitatory discharges were increased 85% over control levels, accompanied by a 35% reduction in spontaneous firing rate. Recovery to the control level of glutamate-evoked response and background firing rate was observed after cessation of cocaine application. Each histogram sums unit activity during four glutamate applications. (From Jimenez-Rivera, C.A. and Waterhouse, B., Brain Res., 546, 287–296, 1991. With permission.)

5.4.3.3 Cocaine Augments Sensory Cortical Neuron Responses to Afferent Pathway Electrical Stimulation or Glutamate Application In an initial study in rat barrel field cortex, single-cell responses to electrical stimulation of the thalamocortical pathway or direct iontophoretic application of the putative excitatory transmitter glutamate were monitored before, during, and after systemic or local application of cocaine.49 The results indicated that systemically applied cocaine can augment both short and long latency and excitatory and inhibitory components of cortical neuronal responses to afferent synaptic inputs. The dose of cocaine used in these studies was within the range that a rat would self-administer under experimental conditions.13 Furthermore, glutamate-evoked excitatory discharges of single neurons were enhanced during microiontophoretic application of cocaine (Figure 5.3). This work provided the first evidence that cocaine administration could affect both excitatory and inhibitory components of physiological responses of cells in sensory circuits. In addition, the demonstration that locally applied cocaine can facilitate neuronal responses to iontophoretically applied glutamate suggests that at least a component of the drug’s action on excitatory synaptic activation is mediated through local mechanisms and that these interactions occur postsynaptically. Overall, these results begin to reveal cellular electrophysiological actions of cocaine that could account for the oft-described changes in sensory perception that accompany acute self-administration of the drug or the persistent

2345_frame_MASTER.book Page 127 Wednesday, October 23, 2002 1:03 PM

Extracellular Single Unit Recording Strategies

127

changes in the salience of drug-related environmental stimuli that follow chronic administration regimens. 5.4.3.4 Cocaine Selectively Enhances Long-latency Responses of Cortical Neurons to Sensory Inputs A subsequent study10 investigated the effects of cocaine on responses of individual somatosensory (“barrel field”) cortical neurons to a more natural form of afferent pathway activation, i.e., tactile stimulation of peripheral somatosensory receptive fields via suprathreshold deflection of individual whiskers (mystacial vibrissae) on the contralateral face. This work was conducted in halothane-anesthetized rats. In many barrel field cortical cells, control responses to brief deflection of the “central” whisker (i.e., the one that is optimal for driving the cell) consist of short latency (6 to 12 msec) excitatory discharges (E1) followed by a post-excitatory inhibition of firing (I) and a second longer latency (20 to 100 msec) excitatory discharge (E2). The E1 response component is thought to be mediated by a direct pathway from the periphery to the cortex that includes relays in the trigeminal nucleus and VPM thalamus.18,19,22,23,27–29 The post-excitatory suppression of activity involves local inhibitory interneurons, and the long-latency E2 response is mediated by a reverberating circuit formed by reciprocal connections between the sensory cortex and posterior medial (POm) thalamus.22,23,28 As shown in Figure 5.4 systemic administration of cocaine at 0.25 to 1.0 mg/kg i.v. differentially affected these responses to suprathreshold stimulation of the central whisker. Within 1 to 2 min of cocaine injection long-latency (E2) responses were enhanced, sometimes as much as 600%, above control levels. By contrast, short-latency E1 responses were only moderately (20%) increased or unchanged by cocaine administration. When evident, inhibitory responses were augmented by cocaine. Spontaneous discharge of cells was not affected by cocaine over the range of doses tested. Changes in stimulus-evoked responses peaked at 5 to 6 min post-drug injection and lasted for 25 to 30 min post-injection, thus following the time course of cocaine’s actions at central synapses.38,51 Procaine, a compound similar to cocaine in chemical composition but lacking its abuse potential, did not produce any of the sensory neuron facilitating effects observed with cocaine. The prominent effect of cocaine on long-latency responses of cortical neurons is significant because such an action may have a unique impact on sensory signal processing and, ultimately, sensory perception. For example, it is widely believed that although the long-latency E2 responses are more labile and less robust than the short-latency E1 responses, they nevertheless convey important qualitative information regarding stimulus attributes. Moreover, several additional questions were raised in light of these findings. First, since short- and long-latency responses of cortical neurons to peripheral somatosensory stimulation are believed to be mediated by different thalamocortical circuits, does cocaine have different effects on neurons in different thalamic relay nuclei? Second, does cocaine produce equivalent effects on cortical neuronal responses to subliminal, perithreshold, and suprathreshold peripheral stimuli? Third, how does cocaine administration affect the receptive field prop-

2345_frame_MASTER.book Page 128 Wednesday, October 23, 2002 1:03 PM

128

Methods in Drug Abuse Research: Cellular and Circuit Level Analyses

FIGURE 5.4 Facilitating action of cocaine on short and long latency, excitatory and inhibitory components of sensory cortical neuron response to afferent synaptic input. Illustrated here are perievent histograms depicting the response of a barrel field cortical neuron to displacement (up arrow) of a single whisker on the contralateral face before and after incremental doses of cocaine (0.25 to 2.0 mg/kg, i.v.). Under control conditions, the response consisted characteristically of short E1 (solid bar) and long E2 (broken bar) latency excitatory discharges. Each histogram sums activity during an equal number of stimulus presentations. Numbers above excitatory peaks indicate spikes per 100 stimuli. Unless otherwise indicated, histograms were computed 5 min after drug or saline injection. Incremental drug injections were given at 40min intervals. After cocaine administration at all doses tested, both E1 and E2 components of the response were increased above control levels. At doses of 0.5 to 1.0 mg/kg a prominent postexcitatory inhibition (I – dash-dotted line) was revealed. (From Bekavac, I. and Waterhouse, B., J. Pharmacol. Exp. Ther., 272, 333–342, 1994. With permission.)

erties of sensory cortical neurons? Finally, which of the candidate monoamine systems are likely to be responsible for the observed effects of cocaine on cortical neuron responsiveness? Preliminary studies suggest that cocaine enhances the response of a cortical neuron to stimulation at the center of its receptive field yet suppresses responses of the same cell to stimulation of its receptive field periphery (Figure 5.5). Such an

2345_frame_MASTER.book Page 129 Wednesday, October 23, 2002 1:03 PM

Extracellular Single Unit Recording Strategies

129

CONTROL CENTRAL WHISKER

PERIPHERAL WHISKER

Hz 400

Hz 400

300

300

200

200

100

100

0

TIME (ms)

250

0

TIME (ms)

250

0.25 mg/kg COCAINE

110.2%

74.5%

0.5 mg/kg COCAINE

93.2%

51.0%

1 mg/kg COCAINE

84.4%

10.3%

2 mg/kg COCAINE

70.5%

3.4%

FIGURE 5.5 Post-stimulus-time histogram records (PSTHs) illustrate responses of a single barrel field cortical neuron to uniform displacement (at arrow) of its central (left column) or peripheral (right column) whisker before and after (5 min post-injection) cumulative i.v. doses of cocaine. Drug injection intervals = 40 min. Each histogram sums unit activity during 50 stimulus presentations. Numbers above response peaks indicate percent of control response.

action by a systemically administered compound would yield a net reduction in receptive field size for cells throughout a sensory circuit and, as such, could sharpen the feature-detection properties of a sensory network. 5.4.3.5 Cocaine Effects on Somatosensory Thalamic Neurons The goal of the next series of experiments9,97 was to characterize the effects of systemically administered cocaine on somatosensory thalamic neuron responsive-

2345_frame_MASTER.book Page 130 Wednesday, October 23, 2002 1:03 PM

130

Methods in Drug Abuse Research: Cellular and Circuit Level Analyses

ness to peripheral activation of afferent synaptic pathways. Extracellular recordings were obtained from single units in POm and VPM thalamic nuclei of halothaneanesthetized rats. For each neuron spontaneous firing rate and response to graded mechanical displacement of a single whisker were monitored before and after systemic administration of cocaine. VPM neurons respond to whisker stimulation with a short-latency (4 to 10 msec) burst of action potentials, whereas responses of POm neurons to whisker stimulation occur at longer latencies (16 to 85 msec) and appear as a graded, slightly elevated discharge occurring over a period of 30 to 100 msec. Initial studies using suprathreshold stimuli indicated that systemically administered cocaine had its most prominent and consistent facilitating effect on POm neurons (Figure 5.6). Since POm inputs to cortex are believed to be responsible for long-latency E2 cortical unit responses, these findings are consistent with our previous observation that systemic cocaine had its most robust effect on E2 cortical neuron responses. Thus, these results suggest that short-latency somatosensory relays through VPM thalamus and subsequent E1 responses in barrel field cortex represent a fast, nonlabile path from periphery to cortex that provides for reliable transmission of sensory information but that is less likely to be affected by cocaine administration. On the other hand, longer-latency interactions between POm thalamus and cortex represent a more labile and qualitative component of somatosensory signals that can be significantly enhanced by cocaine. Since POm confers a qualitative dimension to somatosensory stimulus coding, drug actions at this site may underlie subtle influences of cocaine on sensory perception. More recent work96 using a state-of-the-art piezoelectric bimorph stimulator to produce small deflections of individual whiskers across a linear range have revealed multiple actions of cocaine on VPM thalamic neuron responses to subliminal and perithreshold activation of the vibrissae system (Figure 5.7). While cocaine at 0.75 mg/kg consistently decreased response time (Figure 5.7, right) and reduced the trialto-trial variability of response latency, changes in response magnitude occurred in either direction (increased or decreased). Likewise, cocaine-induced shifts of input/output response curves were such that cells were either more (Figure 5.7, left) or less responsive to a range of inputs after cocaine administration. In some instances, stimuli that were otherwise subthreshold for producing responses in individual VPM neurons elicited strong discharges in these same neurons following cocaine administration (see for example neuronal response to 42-mm whisker deflection, Figure 5.7, left). In cumulative dose paradigms, cocaine had minimal effect on the magnitude of stimulus-evoked discharge at the lowest dose tested, enhanced stimulus-evoked responses at intermediate doses, and caused a decrease in the response to centralwhisker stimulation at the highest dose tested. Thus, depending on the dose administered cocaine exerted variable effects on response magnitude in VPM thalamic neurons (Figure 5.8a). In recent experiments (Rutter, Bauman, and Waterhouse, unpublished), we used microdialysis to determine that injections of cocaine over a range of doses, 0.25 to 2.0 mg/kg i.p., produce increased levels of NE and 5-HT in the VPM thalamus at times that are coincident with cocaine-induced changes in VPM thalamic neuron responsiveness to afferent synaptic inputs (Figure 5.8b).

2345_frame_MASTER.book Page 131 Wednesday, October 23, 2002 1:03 PM

Extracellular Single Unit Recording Strategies

131

POm

VPM CONTROL Hz

Hz E–189.6

300 200

200

100

100

0

TIME (ms)

E–20.1

300

500

0

TIME (ms)

500

1 mg/kg cocaine E–39.5

E–195.2

20 min after cocaine E–183.8

E–18.7

FIGURE 5.6 Perievent histograms illustrate the responses of a single VPM (left column) thalamic neuron and a single POm (right column) thalamic neuron to suprathreshold whisker stimulation (arrow) before and after (5 and 20 min) cocaine (1 mg/kg i.v.) administration. Discharge frequency for each neuron during the excitatory response epoch (solid bar or broken bar beneath histogram) is indicated (E = ) above each histogram. Each histogram sums unit activity during 50 stimulus presentations.

Overall, these results indicate that acute systemic administration of cocaine causes increased extracellular tissue levels of NE and 5-HT in VPM thalamus and can reduce the latency and increase the magnitude of sensory thalamic neuron responses to peripheral receptive field stimulation. As evidenced by responses to otherwise subthreshold synaptic stimuli, the drug can also reduce the threshold of detection for sensory stimuli. Thus, immediately following and for approximately 10 to 20 min after i.v. cocaine administration the sensitivity, speed, and reliability of signal transmission through this somatosensory thalamic relay nucleus can be increased. As such, these facilitating actions of the drug on VPM thalamic neuron

2345_frame_MASTER.book Page 132 Wednesday, October 23, 2002 1:03 PM

132

Methods in Drug Abuse Research: Cellular and Circuit Level Analyses

Amplitude response 220

control cocaine

200

18

180

16

160

14 latency (ms)

frequency (Hz)

Latency response

140 120 100 80

control cocaine

12 10 8

60 40

6

20 4

0 20

40 60 80 100 120 140 160 whisker movement (microns)

20

40 60 80 100 120 140 160 whisker movement (microns)

FIGURE 5.7 Graphs illustrate stimulus/response relationships (amplitude at left, latency at right) for a single VPM thalamic neuron before (solid line) and after (dashed line) systemic cocaine administration, 0.75 mg/kg i.v. Each point represents the averaged response of the cell to 40 stimulus presentations. At all stimulus intensities (whisker movement in microns) tested cocaine increased the magnitude (frequency) of the cell’s evoked discharge. Likewise, the latency of evoked response for each intensity tested was decreased following drug administration. Minimal control response occurs at 42-mm whisker deflection.

responsiveness could underlie some of the alterations in sensory experience reported by humans following cocaine self-administration.

5.5 IMPLICATIONS OF SINGLE-UNIT DRUG STUDIES IN INTACT, ANESTHETIZED PREPARATIONS Once the acute effects of a drug on single neuron response properties have been determined, additional extracellular single-unit studies can be conducted to measure cellular functions and neural circuit operations in animals that have been chronically maintained on drug or withdrawn from chronic drug treatment. Such investigations are crucial to understanding the physiological bases for tolerance, sensitization, and persistence of drug-induced changes in neuronal function that accompany chronic drug self-administration. In addition, the results of single-unit studies in intact, anesthetized preparations have historically provided a conceptual framework for experiments utilizing more reductionist approaches, e.g., patch clamp, molecular biology, to study neuronal function. For example, the demonstration in the late 1970s that microiontophoretic application of NE72 or stimulation of the noradrenergic pathway from locus coeruleus to cerebellum71 could augment GABA-induced suppression of Purkinje cell discharge in anesthetized rats led to a series of whole-cell patch-clamp studies in acutely dissociated Purkinje neurons.21 This work provided convincing evidence that the facilitating actions of NE on GABA responsiveness involved increases in chloride conductance subsequent to activation of the catalytic subunit of protein kinase A via the beta-receptor-linked cyclic AMP second mes-

Time (sec)

1200

1600

2000

2800

3200

1.0 mg/kg

Time (sec)

2400

Saline 0.25 mg/kg 0.5 mg/kg

0.16 800

0.17

0.18

0.19

0.20

0.21

0.22

0.23

0.24

0.25

3600

2.0 mg/kg

60 40

80

100

120

140

B.

0.16 800

0.17

0.18

0.19

0.20

0.21

0.22

0.23

0.24

0.25

1200

1600

2000

2800

3200

1.0 mg/kg

Time (sec)

2400

Saline 0.25 mg/kg 0.5 mg/kg

3600

5-HT NE

2.0 mg/kg

200

120 100

140

160

180

% BASELINE

Frequency (Hz)

Time (sec)

FIGURE 5.8 (A.) Vertically oriented cumulative raster illustrates the effects of incremental doses of cocaine on the spike train activity of a single VPM thalamic neuron during threshold level deflection (open arrowhead) of a cell’s primary whisker on the contralateral face. Each vertical line of dots represents the cell’s activity during one stimulus trial. Times of cocaine injections (0.25, 0.5, 1.0 and 2.0 mg/kg i.v.) are indicated by the vertical dashed lines and solid arrowheads. Horizontal axis = running time of experiment, vertical axis = time relative to stimulus presentation (open arrowhead). Dashed line connecting filled circles below cumulative raster indicates the frequency of stimulus-evoked discharge for each dose of cocaine tested. In this case, the magnitude of the whisker-evoked response was increased to a maximum after cocaine injection of 1.0 mg/kg. (B.) Same cumulative raster as in (A.) but with data from a separate microdialysis experiment superimposed. Solid and dashed lines indicate the changes in VPM thalamic tissue levels of 5HT and NE, respectively, as a function of incremental doses of cocaine (0.25, 0.5, 1.0, and 2.0 mg/kg i.v.). Microdialysis data were collected from a single site in the VPM thalamus of a halothane anesthetized rat. Dialysis samples were taken every 10 min for 30-min periods before and after cocaine administration. Note that even though incremental doses of cocaine resulted in progressively higher tissue levels of NE and 5-HT, the same regimen of cocaine-increased stimulus evoked discharge of the thalamic cell to an optimum at 1.0 mg/kg but then markedly suppressed the response at 2.0 mg/kg.

A.

2345_frame_MASTER.book Page 133 Wednesday, October 23, 2002 1:03 PM

Extracellular Single Unit Recording Strategies 133

2345_frame_MASTER.book Page 134 Wednesday, October 23, 2002 1:03 PM

134

Methods in Drug Abuse Research: Cellular and Circuit Level Analyses

senger system. The implication of this work is that GABA A receptor function in Purkinje cells is subject to modulation by synaptically released NE and proteinkinase-A-mediated phosphorylation of the GABA receptor protein. Cocaine or damphetamine-induced enhancement of Purkinje cell responsiveness to GABA (see Figures 5.1 and 5.2) could involve similar mechanisms, many of which could be subject to long-term modification under conditions of chronic drug administration. Finally, the results obtained in single-unit studies in intact, anesthetized preparations constitute the foundation for electrophysiological investigations of drug actions in awake, behaving animals. Techniques involving microiontophoresis in awake animals70,117 and simultaneous recording of multiple single cells16,17,20,78,83,84,116,117 in awake, freely moving animals can provide significant new insights into the processes underlying the rewarding and reinforcing properties of drugs, the anticipatory states of neurons and circuits prior to drug self-administration, and the progression of changes in neural function that accompany chronic administration of a drug of abuse. However, these experimental approaches must allow for and deal with the many constraints and complexities of data interpretation that are related to normal and druginduced changes in behavioral state. Such complications can be dealt with more readily if a base of information regarding drug effects on cell and circuit function in intact, stable (i.e., anesthetized) preparations is available.

ACKNOWLEDGMENTS The original work described in this chapter was supported by federal grant NIDA DA05117 to BDW and done in collaboration with Dr. Bekavac, Dr. Bauman, Dr. Devilbiss, Dr. Gould, Dr. Jimenez-Rivera, Dr. Rutter, and Dr. Stowe.

REFERENCES 1. Ackerman, J.M. and White, F.J., Decreased activity of rat A10 dopamine neurons following withdrawal from repeated cocaine, Eur. J. Pharmacol., 218, 171, 1992. 2. Ackerman, J.M. and White, F.J., A10 somatodendritic dopamine autoreceptor sensitivity following withdrawal from repeated cocaine treatment, Neurosci. Lett., 117, 181, 1990. 3. Aghajanian, G.K., The modulatory role of serotonin at multiple receptors in brain, in Serotonin Neurotransmission and Behavior, Jacobs, B.L. and Gelperin, A., Eds., MIT Press, Cambridge, MA, 1981, pp. 156–185. 4. Amaral, D.G. and Sinnamon, H.M., The locus coeruleus: neurobiology of a central noradrenergic nucleus, Prog. Neurobiol., 9, 147, 1977. 5. Anden, N.E., Dahlstrom, A., Fuxe, K., Larsson, K., Olson, L., and Ungerstedt, U., Ascending monoamine neurons to the telencephalon and diencephalon, Acta. Physiol. Scand., 67, 313, 1971. 6. Aston-Jones, G. and Bloom, F.E., Activity of norepinephrine-containing locus coeruleus neurons in behaving rats anticipates fluctuations in the sleep-waking cycle, J. Neurosci., 1, 876, 1981a. 7. Aston-Jones, G. and Bloom, F.E., Norepinephrine- containing locus coeruleus neurons in behaving rat exhibit pronounced responses to non-noxious environmental stimuli, J. Neurosci., 1, 887, 1981b.

2345_frame_MASTER.book Page 135 Wednesday, October 23, 2002 1:03 PM

Extracellular Single Unit Recording Strategies

135

8. Bekavac, I., Gould, E.M., and Waterhouse, B.D., Monoaminergic substrates underlying cocaine-induced enhancement of somatosensory evoked discharges in rat barrel field cortical neurons, J. Pharmacol. Exp. Ther., 279, 582, 1996. 9. Bekavac, I., Rutter, J.J., and Waterhouse, B.D., Physiological actions of cocaine in sensory circuits: drug influences on signal transmission through rat POm and VPM thalamic nuclei, Soc. Neurosci. Abstr., 20, 1632, 1994. 10. Bekavac, I. and Waterhouse, B.D., Systemically administered cocaine selectively enhances long-latency responses of rat barrel field cortical neurons to vibrissae stimulation, J. Pharmacol. Exp. Ther., 272, 333, 1994. 11. Berridge, C.W. and Foote, S.L., Effects of locus coeruleus activation on electroencephalographic activity in neocortex and hippocampus, J. Neurosci., 11, 3135, 1991. 12. Bloom, F.E., Hoffer, B.J., and Siggins, G.R., Norepinephrine mediated synapses: a model system for neuropharmacology, Biol. Psychiatry., 4, 157, 1972. 13. Bozarth, M.A. and Wise, R.A., Toxicity associated with long term intravenous heroin and cocaine self-administration in the rat, JAMA, 254, 81, 1985. 14. Bradshaw, C.M., Roberts, M.H.T., and Szabadi, E., Kinetics of the release of noradrenaline from micropipettes: interaction between ejecting and retaining currents, Br. J. Pharmacol., 49, 667, 1973. 15. Byck, R. and van Dyke, C., What are the effects of cocaine in man?, in NIDA Research Monograph #13, Petersen, R.E. and Stillman, R.C. (Eds.), U.S. Government Printing Office, Washington, D.C., 1977, pp. 97–118. 16. Carelli, R.M., Activation of accumbens cell firing by stimuli associated with cocaine delivery during self-administration, Synapse, 35,238, 2000. 17. Carelli, R.M., King, V.C., Hampson, R.E., and Deadwyler, S.A., Firing patterns of nucleus accumbens neurons during cocaine self-administration in rats, Brain Res., 626, 14, 1993. 18. Chang, J-Y., Sawyer, S.F., Lee, R-S., and Woodward, D.J., Electrophysiological and pharmacological evidence for the role of the nucleus accumbens in cocaine selfadministration in freely moving rats, J. Neurosci., 14, 1224, 1994. 19. Chapin, J. and Lin, C.-S., The somatic sensory cortex of rat, in The Neocortex of Rat, Kolb, B. and Tees, R., Eds., Academic Press, San Diego, 1990, pp. 341–380. 20. Chapin, J.K., Waterhouse, B.D., and Woodward, D.J., Differences in somatic response properties of single cortical neurons in awake and halothane anesthetized rats, Brain Res. Bull., 6, 63, 1981. 21. Cheun, J.E. and Yeh, H.H., Noradrenergic potentiation of cerebellar Purkinje cell responses to GABA: cyclic AMP as intracellular intermediary, Neuroscience, 74, 835, 1996. 22. Chiaia, N.L., Rhoades, R.W., Bennett-Clarke, C.A., Fish, S.E., and Killackey, H.P., Thalamic processing of vibrissal information in the rat. I. Afferent input to the medial ventral posterior and posterior nuclei, J. Comp. Neurol., 314, 201, 1991a. 23. Chiaia, N.L., Rhoades, R.W., Fish, S.E., and Killackey, H.P., Thalamic processing of vibrissal information in the rat. II. Morphological and functional properties of medial ventral posterior nucleus and posterior nucleus neurons, J. Comp. Neurol. 314, 217, 1991b. 24. Cohen, S., Cocaine: acute medical and psychiatric complications, Psychiatr. Ann., 14, 747, 1984. 25. Cunningham, K.A. and Lakoski, J.M., The interactions of cocaine with serotonin dorsal raphe neurons: single unit extracellular recording studies, Neuropsychopharmacology, 3, 20, 1990.

2345_frame_MASTER.book Page 136 Wednesday, October 23, 2002 1:03 PM

136

Methods in Drug Abuse Research: Cellular and Circuit Level Analyses

26. Cunningham, K.A. and Lakoski, J.M., Electrophysiological effects of cocaine and procaine on dorsal raphe serotonin neurons, Eur. J. Pharmacol., 148, 457, 1988. 27. Diamond, M.E., Armstrong-James, M., and Ebner, F.F., Somatic sensory responses in the rostral sector of the posterior group (POm) and in the ventral posterior medial nucleus of the rat thalamus, J. Comp. Neurol., 318, 462, 1993. 28. Diamond, M.E., Armstrong-James, M., Budway, M.J., and Ebner, F.F., Somatic sensory responses in the rostral sector of the posterior group (POm) and in the ventral posterior medial nucleus (VPM) of the rat thalamus: dependence on the barrel field cortex, J. Comp. Neurol., 319, 66, 1993. 29. Dykes, R.W., The anatomy and physiology of the somatic sensory cortical regions, Prog. Neurobiol., 10, 33, 1978. 30. Einhorn, L.C., Johansen, P.A., and White, F.J., Electrophysiological effects of cocaine in the mesoaccumbens dopamine system: studies in the ventral tegmental area, J. Neurosci., 8, 100, 1988. 31. Esposito, R.U., Motola, A.H.D., and Kornetsky, C., Cocaine: acute effects on reinforcement thresholds for self-stimulation behavior to the medial forebrain bundle, Pharmacol. Biochem. Behav., 8, 437, 1977. 32. Fischman, M.W., Schuster, C.R., and Krasnegor, N.A., Physiological and behavioral effects of intravenous cocaine in man, in Cocaine and Other Stimulants, Ellinwood, E.H. and Kilbey, M.M., Eds., Plenum Press, New York, NY, 1977, pp. 647–664. 33. Foote, S.L., Bloom, F.E., and Aston-Jones, G., Nucleus locus coeruleus: new evidence of anatomical and physiological specificity, Physiol. Rev., 63, 844, 1983. 34. Foote, S.L., Freedman, R., and Oliver, A.P., Effects of putative neurotransmitters on neuronal activity in monkey auditory cortex, Brain Res., 86, 229, 1975. 35. Freud, S., Uber coca, in Cocaine Papers, Byck, R., Ed., Stonehill Publishing, New York, NY, 1974, pp. 47–74. 36. Gawin, F.H., Cocaine addiction: psychology and neurophysiology, Science, 25, 1580, 1991. 37. Geller, H.M. and Woodward, D.J., An improved constant current source for microiontophoretic drug application studies, EEG Clin. Neurophysiol., 31, 430, 1972. 38. Glowinski, J. and Axelrod, J., Effects of drugs on the uptake, release and metabolism of 3H-norepinephrine, J. Pharmacol. Exp. Ther., 149, 43, 1965. 39. Gold, M.S., Cocaine (and crack): clinical aspects, in Substance Abuse: A Comprehensive Textbook, Lowinson, J.H., Ruiz, P., Millman, R.B., and Langrod, J.G., Eds., Williams and Wilkins, Baltimore, 1992, pp. 205–221. 40. Gold, M.S., Miller, N.S., and Jonas, J.M., Cocaine (and crack): neurobiology in Substance Abuse: A Comprehensive Textbook, Lowinson, J.H., Ruiz, P., Millman, R.B., and Langrod, J.G., Eds., Williams and Wilkins, Baltimore, 1992, pp. 222–235. 41. Hadfield, M.G., Mott, D.E.W., and Ismay, J.A., Cocaine: effects of in vivo administration on synaptosomal uptake of norepinephrine, Biochem. Pharmacol., 29, 1861, 1980. 42. Henry, D.J., Greene, M.A., and White, F.J., Electrophysiological effects of cocaine in the mesoaccumbens dopamine system: repeated administration, J. Pharmacol. Exp. Ther., 251, 833, 1989. 43. Henry, D.J. and White, F.J., Repeated cocaine administration causes persistent enhancement of D1 dopamine receptor sensitivity within the rat nucleus accumbens, J. Pharmacol. Exp. Ther., 258, 882, 1991. 44. Hicks, T.P., The history and development of microiontophoresis experimental neurobiology, Prog. Neurobiol., 22, 185, 1984.

2345_frame_MASTER.book Page 137 Wednesday, October 23, 2002 1:03 PM

Extracellular Single Unit Recording Strategies

137

45. Hoffer, B.J., Neff, N.H., and Siggins, G.R., Microiontophoretic release of norepinephrine from micropipettes, Neuropharmacology, 10, 175, 1971. 46. Hoffer, B.J., Siggins, G.R., and Bloom, F.E., Studies on norepinephrine-containing afferents to Purkinje cells of rat cerebellum. II. Sensitivity of Purkinje cells to norepinephrine and related substances administered by microiontophoresis, Brain Res., 25, 522, 1971. 47. Hoffer, B.J., Siggins, G.R., Oliver, A.P., and Bloom, F.E., Activation of the pathway from locus coeruleus to rat cerebellar Purkinje neurons: pharmacological evidence of noradrenergic central inhibition, J. Pharmacol. Exp. Ther., 184, 553, 1973. 48. Jacobs, B.L. and Azmitia, E.C., Structure and function of the brain serotonin system, Physiol. Rev., 72, 165, 1992. 49. Jimenez-Rivera, C.A. and Waterhouse, B.D., Effects of systemically and locally applied cocaine on cerebrocortical neuron responsiveness to afferent synaptic inputs and glutamate, Brain Res., 546, 287, 1991. 50. Johanson, C.E., Balster, R.L., and Bonese, K., Self-administration of psychostimulant drugs: the effects of unlimited access, Pharmacol. Biochem. Behav., 4, 45, 1976. 51. Johanson, C.-E. and Fischman, M.W., The pharmacology of cocaine related to its abuse, Pharmacol. Rev., 41, 3, 1989. 52. Jones, B.E. and Moore, R.Y., Ascending projections of the locus coeruleus in the rat. II. Autoradiography study, Brain Res., 127, 23, 1977. 53. Kasamatsu, T. and Heggelund, P., Single cell responses in cat visual cortex to visual stimulation during iontophoresis of noradrenaline, Exp. Brain Res., 45, 317, 1982. 54. Koe, B.K., Molecular geometry of inhibition of the uptake of catecholamines and serotonin in synaptosomal preparations of rat brain, J. Pharmacol. Exp. Ther., 199, 649, 1976. 55. Kornetsky, C. and Esposito, R.U., Reward and detection thresholds for brain stimulation: dissociative effects of cocaine, Brain Res., 209, 496, 1981. 56. Kossl, M. and Vater, M., Noradrenaline enhances temporal auditory contrast and neuronal timing precision in the cochlear nucleus of the mustached bat, J. Neurosci., 9, 4169, 1989. 57. Krnjevic, K., Vertabrate synaptic transmission, Physiol. Rev., 54, 418, 1974. 58. Kuhar, M.J., Ritz, M.C., and Boja, J.W., The dopamine hypothesis of the reinforcing properties of cocaine, Trends Neurosci., 14, 299, 1991. 59. Lal, H. and Chessick, R.D., Lethal effects of aggregation and electric shock in mice treated with cocaine, Nature, 208, 295, 1965. 60. Levitt, P. and Moore, R.Y., Origin and organization of brainstem catecholamine innervation in the rat, J. Comp. Neurol., 186, 505, 1979. 61. Levitt, P. and Moore, R.Y., Noradrenaline neuron innervation of the neocortex in the rat, Brain Res., 139, 219, 1978. 62. Lidov, H.G.W., Rice, F.L., and Molliver, M.E., The organization of brainstem catecholamine innervation tin the rat, Brain Res., 153, 577, 1978. 63. Lindvall, O., Bjorklund, A., Nobin, A., and Stenevi, U., The adrenergic innervation of the rat thalamus as revealed by the glyoxylic acid fluorescence method, J. Comp. Neurol., 154, 317, 1974. 64. Lowenstein, D.H., Massa, S.M., Rowbotham, M.C., Collins, S.D., McKinney, H.E., and Simon, R.P., Acute neurologic and psychiatric complications associated with cocaine abuse, Am. J. Med., 83, 841, 1987. 65. Marinelli, M. and White, F.J., Enhanced vulnerability to cocaine self-administration is associated with elevated impulse activity of midbrain dopamine neurons, J. Neurosci., 20, 8876, 2000.

2345_frame_MASTER.book Page 138 Wednesday, October 23, 2002 1:03 PM

138

Methods in Drug Abuse Research: Cellular and Circuit Level Analyses

66. McCormick, D.A., Cholinergic and noradrenergic modulation of thalamocortical processing, Trends Neurosci., 12, 215, 1989. 67. McCormick, D.A. and Prince, D.A., Noradrenergic modulation of firing pattern in guinea pig and cat thalamic neurons, in vitro, J. Neurophysiol., 59, 978, 1988. 68. McLean, J. and Waterhouse, B.D., Noradrenergic modulation of cat area 17 neuronal responses to moving visual stimuli, Brain Res., 667, 83, 1994. 69. Michael, A.J., Waterhouse, B.D., and Woodward, D.J., d-Amphetamine produces a long-lasting potentiation of Purkinje cell response to iontophoretically applied gamma-aminobutyric acid, Soc. Neurosci. Abstr., 9, 1145, 1983. 70. Michael, A.J., West, M.O., Chapin, J.K., Waterhouse, B.D., and Woodward, D.J., Actions of d-amphetamine on cerebellar Purkinje cells in freely moving rats, Soc. Neurosci. Abstr., 11, 550, 1985. 71. Moises, H.C. and Woodward, D.J., Potentiation of GABA inhibitory action in cerebellum by locus coeruleus stimulation, Brain Res., 182, 327, 1980. 72. Moises, H.C., Woodward, D.J., Hoffer, B.J., and Freedman, R., Interactions of norepinephrine with Purkinje cell responses to putative amino acid transmitters applied by microiontophoresis, Exp. Neurol., 64, 489, 1979. 73. Moore, R.Y., The anatomy of central serotonin neuron systems in the rat brain, in Serotonin Neurotransmission and Behavior, Jacobs, B.L. and Gelperin, A., Eds., MIT Press, Cambridge, MA, 1981, pp. 35–71. 74. Morrison, J.H. and Foote, S.L., Noradrenergic and serotonergic innervation of cortical, thalamic and tectal visual structures in old and new world monkeys, J. Comp. Neurol., 243, 117, 1986. 75. Morrison, J.H., Grzanna, R., Molliver, M.E., and Coyle, J.T., The distribution and orientation of noradrenergic fibers in neocortex of the rat: an immunofluorescence study, J. Comp. Neurol., 181, 17, 1978. 76. Mouradian, R.D., Sessler, F.M., and Waterhouse, B.D., Noradrenergic potentiation of excitatory transmitter action in cerebrocortical slices: evidence for mediation by an alpha-1 receptor-linked second messenger pathway, Brain Res., 546, 83, 1991. 77. Nelson, C.M., Hoffer, B.J., Chu, N.-S., and Bloom, F.E., Cytochemical and pharmacological studies on polysensory neurons in the primate frontal cortex, Brain Res., 62, 115, 1973. 78. Nicolelis, M.A.L., Ed., Methods for Neural Ensemble Recordings, CRC Press, Boca Raton, FL, 1999. 79. Olpe, H.-R., The cortical projections of the dorsal raphe nucleus: some electrophysiological and pharmacological properties, Brain Res., 216, 61, 1981. 80. Palmer, M.R., Wuerthele, S.M., and Hoffer, B.J., Physical and physiological characteristics of micropressure ejection of drugs from multibarrel pipettes, Neuropharmacology, 14, 931, 1980. 81. Parfitt, K.D., Freedman, R., and Bickford-Wimer, P.C., Electrophysiological effects of locally applied noradrenergic agents at cerebellar Purkinje neurons: receptor specificity, Brain Res., 462, 242, 1988. 82. Paris, J.M. and Cunningham, K.A., Habenula lesions decrease the responsiveness of dorsal raphe serotonin neurons to cocaine, Pharmacol. Biochem. Behav., 49, 555, 1994. 83. Peoples, L.L., Uzwiak, A.J., Gee, F., and West, M.O., Tonic inhibition of single nucleus accumbens neurons in the rat: A predominant but not exclusive firing pattern induced by cocaine self-administration sessions, J. Neurosci., 86, 13, 1998. 84. Peoples, L.L. and West, M.O., Phasic firing of single neurons in the rat nucleus accumbens correlated with the timing of intravenous cocaine self-administration, J. Neurosci., 16, 3459, 1996.

2345_frame_MASTER.book Page 139 Wednesday, October 23, 2002 1:03 PM

Extracellular Single Unit Recording Strategies

139

85. Peterson, D. and Hardinge, J., The effect of various environmental factors on cocaine and ephedrine toxicity, J. Pharmacol., 19, 810, 1967. 86. Peterson, J.D., Wolf, M.E., and White, F.J., Altered responsiveness of medial prefrontal cortex neurons to glutamate and dopamine after withdrawal from repeated amphetamine treatment, Synapse, 36, 342, 2000. 87. Pitts, D.K. and Marwah, J., Reciprocal pre- and postsynaptic actions of cocaine at a central noradrenergic synapse, Exp. Neurol., 98, 518, 1987. 88. Pitts, D.K. and Marwah, J., Electrophysiological effects of cocaine on central monoaminergic neurons, Eur. J. Pharmacol., 131, 95, 1986. 89. Post, R.M. and Rose, H., Increasing effects of repetitive cocaine administration in the rat, Nature, 260, 731, 1976. 90. Post, R.M., Weiss, S.R.B., and Pert, A., Sensitization and kindling effects of chronic cocaine administration, in Cocaine: Pharmacology, Physiology and Clinical Strategies, Lakoski, J.M., Galloway, M.P., and White, F.J., Eds., CRC Press, Boca Raton, FL, 1992, pp. 115–162. 91. Risner, M.E. and Jones, B.E., Role of noradrenergic and dopaminergic processes in amphetamine self-administration, Pharmacol. Biochem. Behav., 5, 477, 1976. 92. Roberts, D.C.S. and Koob, G.F., Description of cocaine self-administration following 6-hydroxydopamine lesions of the ventral tegmental area in rats, Pharmacol. Biochem. Behav., 17, 901, 1982. 93. Robinson, T.E. and Berridge, K.C., The neural basis of drug craving: an incentivesensitization theory of addiction, Brain Res. Rev., 18, 247, 1993. 94. Rogawski, M.A. and Aghajanian, G.K., Norepinephrine and serotonin: opposite effects on the activity of lateral geniculate neurons evoked by optic pathway stimulation, Exp. Neurol., 69, 678, 1980. 95. Ross, S.B. and Renyi, A.L., Uptake of some tritiated sympathomimetic amines by mouse brain cortex slices in vitro, Acta. Pharmacol. Toxicol., 24, 56, 1966. 96. Rutter, J.J., Bauman, M.H., and Waterhouse, B.D., Systemically administered cocaine alters stimulus-evoked responses of thalamic somatosensory neurons to perithreshold vibrissae stimulation, Brain Res., 798, 7, 1998. 97. Rutter, J.J., Simpson, K.L., and Waterhouse, B.D., Acute systemic cocaine alters the processing of sensory information from the rat vibrissae, Soc. Neurosci. Abstr., 22, 928, 1996. 98. Salmoiraghi, G.C. and Weight, F., Micromethods in neuropharmacology: an approach to the study of anesthetics, Anesthesiology, 328, 54, 1967. 99. Sato, H. and Kayama, Y., Effects of noradrenaline applied iontophoretically on rat superior collicular neurons, Brain Res. Bull., 10, 453, 1983. 100. Scheel-Kruger, J., Braestrup, C., Nielsen, M., Golembiowski, R., and Mogilnicka, E., Cocaine: discussion of the role of dopamine in the biochemical mechanism of action, in Cocaine and Other Stimulants, Ellinwood, E.H. and Kilbey, M.M., Eds., Plenum Press, New York, NY, 1977, pp. 373–407. 101. Schurr, A. and Rigor, B.M., A special holder allows replacement of the recording barrel of a ‘piggy-back’ multibarrel microelectrode, EEG Clin. Neurophysiol., 51, 571, 1981. 102. Sessler, F.M., Cheng, J.-T., and Waterhouse, B.D., Electrophysiological actions of norepinephrine in rat lateral hypothalamus. I. Norepinephrine-induced modulation of LH neuronal responsiveness to afferent synaptic inputs and putative neurotransmitters, Brain Res., 446, 77, 1988.

2345_frame_MASTER.book Page 140 Wednesday, October 23, 2002 1:03 PM

140

Methods in Drug Abuse Research: Cellular and Circuit Level Analyses

103. Stewart, J., deWit, H., and Eikelboom, R., Role of unconditioned and conditioned drug effects in the self-administration of opiates and stimulants, Psychol. Rev., 91, 251, 1984. 104. Stone, T.W., Pharmacology of pyramidal tract cells in the cerebral cortex, Arch. Pharmacol., 278, 333, 1973. 105. Trubatch, J. and van Harreveld, A., Spread of iontophoretically injected ions in a tissue, J. Theor. Biol., 36, 355, 1972. 106. Ungerstedt, U., Stereotaxic mapping of the monoamine pathways in the rat brain, Acta Physiol. Scand. Suppl., 367, 1, 1971. 107. Valentino, R.J. and Aston-Jones, G., Physiological and anatomical determinants of locus coeruleus discharge: behavioral and clinical implications, in Psychopharmacology: The Fourth Generation of Progress, Bloom, F.E. and Kupfer, D.J., Eds., Raven Press, New York, NY, 1995, pp. 373–386. 108. Waterhouse, B.D., Azizi, S.A., Burne, R.A., and Woodward, D.J., Modulation of rat cortical area 17 neuronal responses to moving visual stimuli during norepinephrine and serotonin microiontophoresis, Brain Res., 514, 276, 1990. 109. Waterhouse, B.D., Moises, H.C., and Woodward, D.J., Interaction of serotonin with somatosensory cortical neuronal responses to afferent synaptic inputs and putative transmitter substances, Brain Res. Bull., 17, 507, 1986. 110. Waterhouse, B.D., Moises, H.C., and Woodward, D.J., Alpha receptor mediated facilitation of somatosensory cortical neuronal responses to excitatory synaptic inputs and iontophoretically applied acetylcholine, Neuropharmacology, 20, 907, 1981. 111. Waterhouse, B.D., Moises, H.C., and Woodward, D.J., Noradrenergic modulation of somatosensory cortical neuronal responses to iontophoretically applied putative neurotransmitters, Exp. Neurol., 69, 30, 1980. 112. Waterhouse, B.D., Moises, H.C., Yeh, H.H., and Woodward, D.J., Norepinephrine enhancement of inhibitory synaptic mechanisms in cerebellum and cerebral cortex: mediation by beta adrenergic receptors, J. Pharmacol. Exp. Ther., 221, 495, 1982. 113. Waterhouse, B.D., Sessler, F.M., Cheng, J-T., Woodward, D.J., Azizi, S.A., and Moises, H.C., New evidence for a gating action of norepinephrine in central neuronal circuits of mammalian brain, Brain Res. Bull., 21, 425, 1988. 114. Waterhouse, B.D., Stowe, Z.N., Jimenez-Rivera, C.A., Sessler, F.M., and Woodward, D.J., Cocaine actions in a central noradrenergic circuit: enhancement of cerebellar Purkinje neuron responses to iontophoretically applied GABA, Brain Res., 546, 297, 1991. 115. Waterhouse, B.D. and Woodward, D.J., Interaction of norepinephrine with cerebrocortical activity evoked by stimulation of somatosensory afferent pathways in the rat, Exp. Neurol., 67, 11, 1980. 116. West, M.O., Peoples, L.L., Chapin, M.A., and Woodward, D.J., Low-dose amphetamine elevates movement-related firing of rat striatal neurons, Brain Res., 745, 331, 1997. 117. West, M.D. and Woodward, D.J., A technique for microiontophoretic study of single neurons in the freely moving rat, J. Neurosci. Methods, 11, 179, 1984. 118. Wilson, M.C., Bedford, J.A., Buelke, J. and Kibbe, A.H., Acute pharmacological activity of intravenous cocaine in the rhesus monkey, Psychopharmacol. Commun., 2, 251, 1976. 119. Wise, R.A. Neural Mechanisms of the Reinforcing Actions of Cocaine, NIDA Research Monograph #50, DHHS, Washington, D.C., 1984, pp. 15–35. 120. Wise, R.A. and Bozarth, M.A., A psychomotor stimulant theory of addiction, Psychol. Rev., 94, 469, 1987.

2345_frame_MASTER.book Page 141 Wednesday, October 23, 2002 1:03 PM

Extracellular Single Unit Recording Strategies

141

121. Woods, J., Behavioral Effects of Cocaine in Animals, NIDA Research Monograph #13, Washington, D.C., 1977, pp. 63–96. 122. Woodward, D.J., Moises, H.C., Waterhouse, B.D., Hoffer, B.J., and Freedman, R., Modulatory actions of norepinephrine in the central nervous system, Fed. Proc., 38, 2109, 1979. 123. Zhang, X.F., Hu, X.T., White, F.J., and Wolf, M.E., Increased responsiveness of ventral tegmental area dopamine neurons to glutamate after repeated administration of cocaine or amphetamine is transient and selectively involves AMPA receptors, J. Pharmacol. Exp. Ther., 281, 699, 1997.

2345_frame_MASTER.book Page 142 Wednesday, October 23, 2002 1:03 PM

2345_frame_MASTER.book Page 143 Wednesday, October 23, 2002 1:03 PM

6

Application of ManyNeuron Microelectrode Array Recording and the Study of Reward-Seeking Behavior Patricia H. Janak

CONTENTS 6.1 6.2

Introduction ..................................................................................................143 Description of the Technique.......................................................................144 6.2.1 Electrodes .........................................................................................144 6.2.2 Chronic Implants of Microelectrode Arrays....................................145 6.2.3 Recording Procedures ......................................................................146 6.3 Yields, Stability, and Longevity of Recordings ..........................................148 6.4 Analysis of Spike Train Data Recorded during Reward-Seeking Behavior .......................................................................................................151 6.4.1 Single-Neuron Spike Trains.............................................................151 6.4.2 Correlations between Neuronal Pairs ..............................................153 6.4.3 Ensemble Analyses ..........................................................................154 6.5 Future Directions..........................................................................................157 6.6 Conclusions ..................................................................................................158 Acknowledgments..................................................................................................158 References..............................................................................................................158

6.1 INTRODUCTION This chapter describes the application of multichannel electrophysiological recording in the awake behaving rat to the study of the neurobiology of drug abuse and reward. The in vivo many-neuron recording technique described here allows for the simultaneous measurement of the spike activity of dozens of individual neurons in behaving rodents. Its purpose is to provide a real-time window onto neuronal function so

143

2345_frame_MASTER.book Page 144 Wednesday, October 23, 2002 1:03 PM

144

Methods in Drug Abuse Research: Cellular and Circuit Level Analyses

that we may better understand the relationship between neuronal activity and behavior. Many-neuron electrophysiology therefore can be used to ask questions regarding the contribution of specific neural circuits to reward-seeking behavior. We can ask: how do neuronal networks acquire, process, and act upon information about rewards in the environment? The advantages of the chronic many-neuron recording technique are threefold. First, because many neurons are recorded at the same time, the data-collection rate is faster than previous one-neuron-at-a-time methods of recording neurons during behavior. This makes neuron recording practical for use in conjunction with rat behavioral models. Second, because neurons within the same and across different brain regions are recorded simultaneously, we can begin to understand how neurons work together, within and across brain regions, to accomplish complex behaviors. Third, we can begin to define with greater precision the contribution of a particular brain region to a particular behavior on the actual millisecond time scale used by the brain for interneuronal communication. This type of temporal precision has contributed to research on the role of the nucleus accumbens in intravenous drug self-administration as described by Peoples (this volume). This chapter continues the discussion of the technique as applied in our laboratory to other, orally delivered, reinforcers and will consider some of the issues relevant to the analysis of manyneuron spike train data.

6.2 DESCRIPTION OF THE TECHNIQUE 6.2.1 ELECTRODES The in vivo many-neuron recording technique described here is designed to allow for the simultaneous detection of the spike activity of dozens of individual neurons in behaving rodents. The first important component of this technique is the electrode, which measures local changes in the extracellular potential field produced when a neuron whose cell body is near the tip of an electrode emits an action potential. The amplitude and shape of the recorded waveform depend upon a variety of factors, such as the size and shape of the neuron, including its dendritic tree, and the relative distance between the electrode tip and the cell body.15 Because we are recording extracellular potentials, this technique does not allow exact identification of the recorded cell type in the manner of intracellular techniques that allow one to fill the cell with dye to facilitate later identification. Hence, the first issue to be aware of is that cell-type identification can only be made using characteristic electrophysiological features that have been identified from intracellular recording experiments such as firing rates and bursting patterns. Many types of electrodes have been used for recording multiple units including microwires and glass micropipettes. Electrodes made of metal microwires are obviously much less delicate than glass electrodes filled with conducting solutions and therefore are more convenient for use in behaving rats. Recently, silicon and ceramic electrodes with multisite recording circuits have been etched using thin-film photofabrication technology.2 In theory, the silicon or ceramic probe technology should allow for chronic in vivo recordings with an increase in the number of neurons one

2345_frame_MASTER.book Page 145 Wednesday, October 23, 2002 1:03 PM

Many-Neuron Microelectrode Array Recording and Reward-Seeking Behavior 145

can record and less concomitant disruption of neural tissue. So far, successful longterm recordings in behaving animals with this technology have remained elusive.17 (http://www.engin.umich.edu/facility/cnct/backind.html). Hence, the best choice to date for electrodes for chronic implant appears to be microwire arrays. These arrays provide reliable and long-lasting signal acquisition in behaving subjects. The arrays or bundles of wire are inserted into the brain regions of interest and are then repeatedly attached to preamplifiers, digitizers, and computers for acquisition of neural data. Microwire electrode arrays are typically made of insulated stainless steel, tungsten, nichrome (nickel-chrome alloy), or platinum. Microwire diameters in the range of 25 to 80 mm work well for detecting extracellular potentials of one or more nearby cell bodies. Fine beveled tips on these wires are sometimes used, although it is now clear that blunt cut wires are easier to prepare and are sufficient for recording extracellular spikes. Electrode bundles or arrays may be homemade or bought commercially. The electrode arrays we use are available from NB Labs (Dennison, TX). They are constructed from Teflon-insulated 50-mm stainless steel wire soldered to plastic female connectors. These electrode arrays are arranged in groups of 8 or 16 wires, with one stainless steel or silver ground wire per eight recording electrodes. An indepth discussion of electrodes used in chronic ensemble recording in behaving animals can be found in a recent review by Moxon.17 The insertion of microwire arrays into brain is described below, followed by a description of the data-acquisition hardware and software.

6.2.2 CHRONIC IMPLANTS

OF

MICROELECTRODE ARRAYS

The quality and longevity of many-neuron chronic recordings appear to depend upon the creation of a secure headstage.12,31 Here we describe in some detail the surgical procedure for implanting the microwire electrode arrays, from NB Labs or elsewhere. The surgery can be lengthy (3 to 5 h). Therefore, it is important to keep the subject warm using a heating pad and to choose an anesthetic that is appropriate for long surgeries. We use a ketamine (80 mg/kg) and xylazine (10 mg/kg) mixture or isoflurane. Aseptic surgical techniques are used. The scalp is shaved and cleaned with alcohol pads and disinfectant soap. The subject is mounted within a standard stereotaxic apparatus (David Kopf, Tujunga, CA), and antibiotic optical ointment is gently applied to the surface of each eye with a sterile cottontipped applicator. This ointment is reapplied every hour or so to keep the membranes of the eye moist and to protect the eye from accidental contamination with dental acrylic, etc. during the procedure. Next, a midline scalp incision is made with a sterile scalpel, and the skull surface is cleaned of blood and tissue. A dissecting microscope, or other means to magnify the surgical field, aids greatly in the remaining steps of this procedure. Hand-held or mounted drills are used to create openings in the skull large enough to accommodate the wires of the electrode arrays. The electrode arrays are secured to the stereotaxic device using holders made of male connectors that complement the female connectors of the electrode arrays. NB Labs coats their electrode arrays with an inert agent (polyethylene glycol) along the length of the array, leaving a customer-

2345_frame_MASTER.book Page 146 Wednesday, October 23, 2002 1:03 PM

146

Methods in Drug Abuse Research: Cellular and Circuit Level Analyses

specified bare length of wire, to enable these fine wires to penetrate the brain. Care must be taken to keep the insulation intact along the length of the wires. Because the 50-mm wires sometimes have difficulty piercing the dura, we gently rip an opening in this layer of the meninges using fine forceps. If bleeding occurs during the removal of the skull or the tearing of the dura, the region is carefully rinsed and cleaned and the electrodes are not lowered until the bleeding has ceased. The electrodes can be lowered to the desired depth over a period of minutes using micromanipulators; it is thought that slow penetration of the brain tissue produces less tissue damage and, hence, better recordings. Once the electrodes are in place very small fragments of gel foam (Pharmacia and Upjohn, Kalamazoo, MI) are placed gently around the electrodes to protect the surface of the brain; the electrodes are then secured to the skull using a small amount of dental acrylic (Jet dental acrylic, Lang Dental Mfg. Co., Inc., Wheeling, IL). Before any dental acrylic is applied, the skull is carefully cleaned and dried. A very clean and very dry skull surface is crucial for the close and tight bonding of the dental acrylic with the skull. All initial applications of dental acrylic are runny, allowing the acrylic to flow easily between the individual electrodes and increasing the likelihood that the wires will be well anchored to the cement. After the acrylic dries the stiffening agent can be rinsed from the distal section of the array with sterile saline or PBS if desired. This process is repeated if multiple arrays are to be implanted. At least four skull screws are used to anchor the cement headstage to the skull. The use of many skull screws around the perimeter of the surgical area adds to the stability of the headset. Additional holes are drilled for ground wires that are inserted 1 to 2 mm into the cortex and are secured using a small amount of dental acrylic. After all electrodes and ground wires are secured and all screws placed, the connectors of the electrodes are arranged above the rat’s skull in the desired final position using homemade holders secured to the stereotaxic device. Again, the skull is carefully cleaned and dried. Dental acrylic is then applied in stages over the entire assemblage such that all metal and the bottom half of the connectors are covered. Care is taken to remove any dental acrylic that may have seeped over the subject’s skin and to ensure that no sharp edges remain that might irritate the skin. If the original skin incision is longer than the final headstage requires, then sutures are used to close the tissue. Topical anesthetics (5% lidocaine ointment) are applied to the wound margin to reduce post-surgical discomfort. Topical antibiotics are used to combat infection. Injection of systemic antibiotics may also aid greatly in reducing the incidence of infection that contributes to premature loss of the dental cement headstage.

6.2.3 RECORDING PROCEDURES Subjects are allowed to recover from the surgery for 1 week before recording sessions commence. Extracellular signals from the implanted wires are detected by attaching a headset cable to the connector on top of the rat’s head. Commercial headset (recording) cables are available from NB Labs and Plexon Inc. (Dallas, TX); cables can also be made in the laboratory.13 Headset cables contain one channel for each

2345_frame_MASTER.book Page 147 Wednesday, October 23, 2002 1:03 PM

Many-Neuron Microelectrode Array Recording and Reward-Seeking Behavior 147

microwire as well as one or more channels for ground wires and an input for delivering power to the headset amplifiers. An important function of these cables is to boost the current of the relatively low-voltage signals detected by microwires; this is accomplished with Field Effect Transistors (FETs) or Operational Amplifiers (Op-Amps) embedded in the headset as close to the connector as possible. The FETs or Op-Amps within the headset cables also serve to reduce the impedance of the signal output and, hence, movement artifacts generated from cable movements.13 The opposite end of the headset/recording cable is then connected to a commutator (Dragonfly Inc., Ridgeley, WV) to allow continuous signal transmission from each channel during the subject’s movement. Both the acquisition and the analysis of many-neuron ensemble recordings have been aided immensely by advances in electronics including computing power and data-storage capabilities. Several companies supply recording hardware and software (see Reference 26 for a review of basic attributes of multichannel extracellular spike recording equipment). The commercial products used to collect and analyze data in our laboratory are produced by Plexon, Inc. and by Biographics, Inc. (WinstonSalem, NC). The specific attributes of each of these systems are available from the manufacturers; however, the basic requirements for any multichannel recording system are similar. First, after passing through the commutator at the top of the behavioral chamber, the signals from each channel must be amplified, filtered, and digitized. Some recording systems, such as the Plexon recording system, allow for differential recording of the signals from each wire relative to a user-selected wire (one with little or no spike activity) to further reduce the background noise measured on the wires of interest. The ability to sample and digitize information from many independent channels simultaneously at a high rate allows us to observe a faithful picture of the actual spike output of large distributed networks of neurons throughout the brain. This is a major advantage of multichannel recording systems, and rates from 25 to 40 kHz can be obtained with currently available commercial devices. Data-acquisition software allows for spike sorting to be carried out by the user online. Typically, one sorts the spikes from each channel by separating those waveforms thought to represent firing from individual neurons from each other and from the noise, and then one begins recording the neural data. Online spike-sorting routines apply statistical algorithms to assist in classifying waveforms as belonging to one or more distinct neurons. These techniques include template matching and principal-component analysis, among others, as described by Wheeler.30 We use these techniques as implemented in the Plexon data-acquisition software as well as the intuitively simpler method of sorting using time-voltage parameters, which in this system is accomplished visually by the placement of two adjustable windows on segments of the waveform display. The digitized waveforms can then be saved to the hard drive of the host computer. In addition, all times of occurrence of each individual neuron’s extracellular action potential, or spike, are saved. These times are referred to as timestamps, and the progression of these spikes through time is commonly referred to as the spike train. At the same time behavioral data can be collected. Many systems are now

2345_frame_MASTER.book Page 148 Wednesday, October 23, 2002 1:03 PM

148

Methods in Drug Abuse Research: Cellular and Circuit Level Analyses

designed to interface with any behavioral control system that can emit TTL pulses for each behavioral event such as lever presses, tones, etc. Hence, the times of each spike from each user-identified neuron and the times of each behavioral event are all saved together in the same data file as separate variables, greatly simplifying offline analysis. The raw data files can be analyzed using custom programs. However, software programs designed to import data files from various data-acquisition systems are available including Stranger (Biographics, Inc.) and NeuroExplorer (Plexon, Inc.). These programs rapidly prepare numerical and graphical results for many-neuron data including overall firing rates, autocorrelograms, crosscorrelograms, perievent histograms, and more. The numerical results can be exported into other software programs as needed for statistical analyses and other manipulations. Following the final recording session the subject is deeply anesthetized, and the location of the tips of the electrodes is marked by passing a 10- to 20-mA current for 10 sec through one or more wires per array to deposit iron ions into the surrounding tissue. Subjects are then transcardially perfused with PBS followed by a 4% paraformaldehyde/3% potassium ferrocyanide solution. The potassium ferrous cyanide forms a blue reaction product with the deposited iron. Standard histological procedures are then used to slice and stain the tissue to visualize electrode location.

6.3 YIELDS, STABILITY, AND LONGEVITY OF RECORDINGS The many-neuron microwire array recording technique discussed here allows for the spike activity of groups of individual neurons to be recorded simultaneously in the awake behaving rat.32 The number of neurons one records depends upon the number of wires implanted into the brain, which determines the number of data channels required. Some wires detect the waveforms from one or more nearby neurons, while other wires are silent. Typical yields from a 16-wire implant into the dorsal or ventral striatum can vary from 5 to 20 clearly discernable units. Once subjects have recovered from the surgery, daily recording sessions can commence. A purchased or handmade recording headset cable, described previously, is easily attached to an awake rat, and neural activity is then gathered during the behavior of interest such as the operant self-administration of drug. The practical advantages offered by this technique include the ability to obtain data from more than one neuron at a time and the ability to obtain data for many days. Both of these points are likely to depend upon the quality of the surgery. Placement of the electrodes in regions rich with neurons, rather than in fiber bundles, is critical. This problem can be mitigated by recording during the implanting process. The microwire electrode arrays are fixed following surgery; their position cannot be changed. Hence, this technique can emphasize the type of information to be gained from observing the same neuron over time. The appearance of similar-shaped waveforms on a given wire from session to session suggests that the waveform represents the extracellular action potential emitted by one neuron that we are able to follow day after day. Conclusive proof that one is recording the same neuron over

2345_frame_MASTER.book Page 149 Wednesday, October 23, 2002 1:03 PM

Many-Neuron Microelectrode Array Recording and Reward-Seeking Behavior 149

TABLE 6.1 Typical Accumbens Neuronal Yields from Bilateral Eight-Wire Microarray Electrodes: A Representative Study Number of Neurons Recorded Subject

Totals Mean SD SEM

Number Sessions Analyzed

In Only One Session

In More Than One Session

1 2 3 4 5 6 7 8

1 5 5 13 18 14 14 15

12 11 12 16 0 0 3 1

0 9 12 9 19 9 12 9

8

85 10.63 6.07 2.15

55 6.85 6.51 2.30

79 9.88 5.25 1.86

Total Units 12 20 24 25 19 9 15 10 134 16.75 6.18 2.19

prolonged periods is not possible. However, the use of a combination of observations increases confidence that the neuron recorded across two distinct sessions is the same. These factors include the waveform shape, the firing pattern of the unit as indicated by the autocorrelogram, and the average firing rate. Unfortunately, any or all of these parameters may change over time as a result of plasticity, perhaps especially if a subject’s behavior has changed or as a result of a pharmacological agent, and so conclusions must be drawn with caution. The longevity of quality recordings within our individual experiments has varied. An example experiment is described in Table 6.1, which gives the number of sessions for which neuronal recordings were obtained and the number of neurons considered to be observed during more than one session. For this study eight rats were implanted with eight-wire microelectrode arrays into the nucleus accumbens bilaterally (total of 16 wires). The number of sessions for each subject for which good-quality recordings were obtained varied from 1 to 18. Therefore, it is possible to obtain a considerable amount of data from one subject. Moreover, long-term studies for the determination of within-subject dose–effect functions or studies with experimental variables that change over time, such as examinations of the acquisition or extinction of a drug habit, become possible. Over 60% of the accumbens units listed in Table 6.1 were judged to be the same cells recorded during more than one session. The possibility of testing the same unit under different circumstances over time allows us to ask questions about the stability and specificity of the coding properties of each unit or of networks of continuously observed units. Figure 6.1A depicts the activity of an accumbens neuron recorded during performance of an operant task reinforced by 0.1-ml drops

2345_frame_MASTER.book Page 150 Wednesday, October 23, 2002 1:03 PM

150

Methods in Drug Abuse Research: Cellular and Circuit Level Analyses

Perievent Histograms, bin = 100 ms

A

R3 4/14 Reward Self-Administration

Autocorrelograms, bin = 1 ms

C

10

dsp015b

R3 4/14

Hz

6 8

4 2

6 0 0 Extinction

5

Hz

-5

4

Hz

6 4

2

2 0

0 -5

0 5 Reward Reinstatement

D

Hz

6

-0.2

10

-0.1

0 ime (sec)

0.1

0.2

dsp015b

R3 4/15

4 8 2

-5 R3 4/15

0

5

Reward Self-Administration

6

Hz

0

B

4

Hz

6 2

4 2 0

0 -5

0 Time (sec)

5

-0.2

-0.1

0 Time (sec)

0.1

0.2

FIGURE 6.1 Stability of recording in behaving rats. Accumbens neurons were recorded during the performance of a nosepoke response reinforced by a 5% sucrose solution. (A) Three perievent histograms that depict the average firing rate of one unit recorded from the nucleus accumbens during the operant self-administration of sucrose. In this session, the subject performed 35 trials that were reinforced (“Reward Self-Administration”), an additional 23 trials that were not reinforced (“Extinction”), and a further 35 trials that again were reinforced (“Reward Reinstatement”). Histograms are aligned at the time at which the 0.1-ml drop of sucrose was delivered (t = 0 on the x-axis). A reward-related decrease in spike activity is apparent during Reward Self-Administration that is absent during Extinction. When the sucrose is again delivered following an operant response, the decrease in spike activity is once more observed (“Reward Reinstatement”). (B) A perievent histogram depicting the average response of the same neuron as in ‘A’ recorded during the next day’s session in which the subject performed 47 reinforced trials. A decrease in spike activity following reward receipt is again observed. (C) Autocorrelogram and waveform (inset) for same neuron as in (A) The autocorrelogram depicts the spike activity of the unit during the entire behavioral session as a function of the {1st order, 2nd order…nth order} time intervals between each reference spike and the remainder of the spikes in the spike train. Hence, the autocorrelogram depicts the firing pattern (tonic vs. phasic) of the unit. (D) Autocorrelogram and waveform for same unit as in (B.) The similarity between the autocorrelograms and waveforms in (C) and (D) suggest that they are the same unit.

of 5% sucrose solution. During the first session depicted here the solenoid mechanism that delivered the reinforcer was shut off during the middle of the session. The reinforcer was made available once again later in the session after the subject had ceased to respond. This within-session extinction-reinstatement procedure can be used to examine the sensitivity of some neurons to the presence and absence of the reinforcer during the same session, allowing great confidence that we are

2345_frame_MASTER.book Page 151 Wednesday, October 23, 2002 1:03 PM

Many-Neuron Microelectrode Array Recording and Reward-Seeking Behavior 151

observing coding changes for the same neuron. In this case, this unit showed a typical decrease in spike activity at the time of receipt of the reinforcer. This decrease was eliminated when the reinforcer was no longer available and returned when the reinforcer was again available. Hence, this behavioral correlate depends upon the presence of the reinforcer. The stability of the correlate over time can be investigated by looking at the activity of this neuron the following day (Figure 6.1B) during the performance of the same sucrose-reinforced operant task. We can see that the same decrease in activity during the reinforcer period is present. The autocorrelograms and waveforms for the neuron from each of the two consecutive recording sessions are also shown in Figure 6.1C and D. Their similarity in appearance suggests that the neuron is the same.

6.4 ANALYSIS OF SPIKE TRAIN DATA RECORDED DURING REWARD-SEEKING BEHAVIOR Once the spike trains have been recorded along with the relevant behavioral events the next formidable task for the electrophysiologist is to determine the meaning of the spike activity in terms of both information processing in the brain and behavior. Issues of neural coding have long been considered by physiologists, psychologists, and neuroscientists. Some decades ago, Perkel and Bullock22 provided an exhaustive list of “candidate neural codes or forms of representation of information in the nervous system” that is still relevant today. Their list is comprised of three categories: neuronal events other than impulses, such as post-synaptic potentials, impulses in unit neurons, and ensemble activity. We will be concerned with the latter two categories.

6.4.1 SINGLE-NEURON SPIKE TRAINS Examination of spike trains recorded in the awake animal reveals a variety of changes related to behavior. When we examine the spike activity recorded from a single neuron in a behaving animal, we typically look at changes in firing rate during the behavioral session and changes in firing rate relative to specific behavioral events such as lever-presses (Figure 6.2A) (see Color Figure 6.2 following page 50). Both of these methods have been used to ascribe functions to specific neurons in selected brain regions. In-depth interpretation of these types of neuronal responses is described in Chapter 7. Determining the type of information encoded in these responses is facilitated by manipulating behavioral variables, as described previously for comparisons of neural coding in the presence and absence of reinforcement. Quantitative evaluations of the relationship between single-neuron firing and specific perceptual and behavioral events have been made and have provided important insights in a number a species, including nonhuman primates (please see an early highly influential series of papers from Richmond and colleagues20,23,24 and the book Spikes.25 An example of this type of analysis as applied to drug abuse research can be found in Bowman et al.1 These authors trained rhesus monkeys to respond on a multiple-ratio reaction time task for both juice and i.v. cocaine reinforcement during the same session. Principal-component and information analyses revealed reinforcerdependent differences in the way in which accumbens neurons encoded the proximity

2345_frame_MASTER.book Page 152 Wednesday, October 23, 2002 1:03 PM

152

Methods in Drug Abuse Research: Cellular and Circuit Level Analyses

A

Single-Neuron Analysis

Hz

1. Rate Histograms

experimental manipulation

Condition 2

Condition 1

12 8 4 0 0

250

500

750 Time (sec)

1000

1250

1500

3. Perievent Histograms and Rasters

2. Rasters

10 1100

1105

1110

1115

1100

1105

1110 1115 Session Time (sec)

1120

1125

1120

1125

5

Hz

Spike train Event

B

0 -5

0 Time (sec)

5

Neuronal Pairs 2. Crosscorrelogram

1. Two Simultaneously Recorded Spike Trains

Hz

Reference Target

Time (sec)

t1a

ix

i2

t1b-t1a = i1

t1b

t2b

txb

4 3.5 3 2.5 2 1.5 -1

-0.5

0 Time (sec)

0.5

1

FIGURE 6.2 Spike train analysis. (A) Single-neuron analysis. 1. Rate histograms. These histograms allow observation of changes in the average firing rate or frequency (Hz) of one neuron as conditions change during the recording session. In the example presented here, the experimental manipulation at time = 700 sec was followed by an overall increase in firing frequency by this unit. 2. Rasters. Rasters, or spike trains, depict the time of occurrence of each spike recorded from a given neuron. Patterns in the spike train may be apparent. In this case, a comparison of the time of occurrences of an experimental event, such as a lever-press, with the spike train indicates that the unit bursts prior to the emission of the operant response. 3. Perievent histograms and rasters. To better determine the relationship between a spike train and a repeated behavioral event, a perievent histogram is created. The x-axis is time-relative to each behavioral event, such as a lever-press, and the y-axis is firing frequency (Hz). The average firing frequency of one neuron relative to all instances of the experimental event is depicted. In this example, the unit increased spiking just prior to the occurrence of the event. The raster, above, depicts the spike activity (spike train) of the neuron for each individual instance (trial) of the behavioral event, aligned on the same time scale as the perievent histogram. Thus, the perievent histogram is the average of the spike activity across all of the rasters. (B) Neuronal Pairs. 1. Two simultaneously recorded spike trains. To detect the presence of nonrandom relationships between the firing of two simultaneously recorded neurons using crosscorrelograms, the cross-neuron interspike intervals are tabulated. For each spike in the spike train for the reference neuron all possible interspike intervals between that reference spike and each spike in the spike train of the target neuron are determined. 2. Crosscorrelogram. The cumulated intervals determined for all possible interspike intervals between the reference and target neurons are displayed in a histogram following conversion of the units to either firing probability or frequency; t = 0 on the x-axis corresponds to the time of occurrence of each spike in the reference neuron. In this example, the y-axis shows the firing frequency of the target unit relative to a reference neuron spike. The horizontal dotted line is the 95% confidence interval; bars above this dotted line represent significantly correlated firing between the two units. For this neuronal pair, the target amygdala neuron tends to fire at 4 Hz within the 50-msec interval preceding a spike in the reference accumbens neuron.

2345_frame_MASTER.book Page 153 Wednesday, October 23, 2002 1:03 PM

Many-Neuron Microelectrode Array Recording and Reward-Seeking Behavior 153

within the schedule to delivery of reinforcer, indicating either that subjects differed in their motivation to work for juice vs. cocaine or that natural and drug rewards are encoded differently by accumbens neurons.

6.4.2 CORRELATIONS

BETWEEN

NEURONAL PAIRS

Within a given brain region the patterns of activity across networks of neurons are thought to determine the output of that nucleus. The presence of synchronized activity among neurons can be examined by looking at correlated firing between pairs of simultaneously recorded neurons. In addition, correlated activity can be studied in pairs of neurons from two different anatomically connected regions to characterize the functional relationship of neurons within larger circuits comprised of multiple brain regions. Crosscorrelograms are used to quantify the relationship between the firing of two neurons.16 A crosscorrelogram is a histogram that depicts the firing probability, or likelihood, of one neuron relative to a first. The histogram is created by compiling the time intervals between a spike emitted by the reference neuron and each spike emitted from a second target neuron; this process is repeated for all spikes in the reference neuron.21,22 Hence, one obtains a histogram depicting the relative likelihood of firing in the second neuron given a spike in the reference neuron (Figure 6.2B). A flat histogram indicates no relationship between the firing of two neurons. Peaks indicate temporally related firing; troughs indicate that the second unit is not firing when the first fires. Neurons might fire in a correlated fashion if they are directly or indirectly synaptically connected, or if they receive common input. With extracellular recording in the mammalian brain it is very difficult to state with certainty that a functional relationship between two units reflects a monosynaptic interaction between the units. Rather, these techniques can be used in the behaving animal to determine functional interactions within and between brain regions to further our knowledge of the dynamics of the neural activity that occurs during behaviors such as drug selfadministration. For example, crosscorrelations among neurons in the mesocorticolimbic circuit during i.v. cocaine self-administration have been reported. Chang et al.5 found that 18% of neuronal pairs within the nucleus accumbens and the medial prefrontal cortex were significantly correlated if those neurons that made up the pairs also demonstrated significant phasic activity just before the lever-press made to receive the cocaine infusion (termed, anticipatory activity). In contrast, only 4% of neuronal pairs comprised of neurons without anticipatory activity showed significant correlated activity. This type of finding suggests that interactions among pairs of simultaneously recorded neurons contribute to drug-seeking behavior. Figure 6.2B is an example of correlated activity observed within a pair of neurons recorded simultaneously during the operant self-administration of ethanol. The reference neuron in this example was recorded from the medial shell of the accumbens while the target neuron in this pair was recorded from the ipsilateral basolateral amygdala, a region that sends excitatory projections to the medial accumbens. The peak in activity within 200 msec prior to zero indicates that the amygdala neuron tended to fire 200 msec before the accumbens neuron, with the greatest likelihood within 50 msec. This finding demonstrates that spike firing within the basolateral

2345_frame_MASTER.book Page 154 Wednesday, October 23, 2002 1:03 PM

154

Methods in Drug Abuse Research: Cellular and Circuit Level Analyses

amygdala precedes that of accumbens neurons, suggesting that neuronal activity within this region of the amygdala may drive at least a portion of the neuronal activity observed within the accumbens during ethanol self-administration. Further specificity in the relationship of correlated activity to ongoing behavior can be defined by calculating the crosscorrelation during selected behaviorally relevant time epochs. For example, Tabuchi, Mulder, and Wiener28 calculated crosscorrelations between simultaneously recorded hippocampal and nucleus accumbens neuronal pairs, using only the 1-sec time period prior to reward delivery within a water-reinforced-plus maze. These authors found that spiking within hippocampal–accumbens neuronal pairs tended to be tightly correlated as the subject approached a reward but not after reward receipt. In this example, crosscorrelation analysis suggests that the anatomical projections from the hippocampus to the accumbens may direct the subject to the location of the reward. It is possible that correlated activity within a neuronal pair actually reflects coincident responding of both neurons to an external stimulus or event, for instance, to the onset of a conditioned stimulus, rather than an actual physiological relationship between the two neurons. In behavioral paradigms with repeated trials/stimulus presentations, such as operant drug self-administration, this confound can be examined by using a shuffling procedure. This procedure mismatches the real-time correlations among two spike trains; instead it looks at the correlated spike firing across all different trials or stimulus presentations and subtracts the average correlated firing under that condition from the original crosscorrelation statistics. For example, the correlation would be calculated between stimulus presentation 1 for spike train 1 and stimulus presentation 2, 3, 4, and so on, for spike train 2; the same is repeated for all other stimulus presentations for spike train 1. The average of all possible shuffles (called the average shift predictor) is subtracted from the original correlation. Any remaining correlation reflects a functional, physiological relationship for the neuronal pair rather than coincident activity driven by the stimulus alone. This procedure was used by Tabuchi et al.28 to examine the hippocampal–accumbens neuronal pairs described above, and the reader is referred to their paper. We can apply this type of analysis to our same neuronal pair from Figure 6.2D; these results are illustrated in Figure 6.3A and B. This procedure is important for disentangling the contribution of coincident activity driven by external stimuli from true functional interactions among neuronal pairs.

6.4.3 ENSEMBLE ANALYSES As our discussion of correlated firing acknowledged, neurons do not work alone; they are embedded in a matrix with thousands of other neurons of the same and different types, and the connections among these neurons shape their individual firing patterns. Scientists long ago arrived at the conclusion that individual representations (of memories, perceptual events, etc.) must be represented within the brain as a specific pattern of activity distributed across many neurons in space and perhaps time. Quantitative characterization of these patterns in relation to ongoing drug-seeking behavior is a challenge. Researchers have applied traditional multivariate statistical techniques to examine neuronal ensemble data4,6–8,12,14,18,19,29 as well

2345_frame_MASTER.book Page 155 Wednesday, October 23, 2002 1:03 PM

Many-Neuron Microelectrode Array Recording and Reward-Seeking Behavior 155

A.

B. Crosscorrelogram during 2-sec Reward Perio d

Shift Predictor 2

Hz

Hz

4 3 2 1

1 0

-1

-0.5

0 Time (sec)

0.5

1

-1 -1

-0.5

0 Time (sec)

0.5

1

FIGURE 6.3 Event-related crosscorrelograms. (A) A crosscorrelation between the same neurons in Figure 6.2.B2. calculated only during the 2-sec time interval following delivery of the reward. No correlation is present between the two units in the 2-sec interval prior to reward delivery (data not shown). (B) Shift predictor histogram produced following shuffling procedure on same data as in (A). A significant correlation between the two neurons still exists following subtraction of shuffled data, as revealed by peaks above the 95% confidence interval line, suggesting that the correlated firing is physiological.

as statistical techniques specifically tailored for neuronal ensemble analysis such as neuronal network models,19 gravity analysis,11 vector reconstruction techniques,9,10 and others.27 Much more detail on ensemble analyses can be found in Methods for Neural Ensemble Recordings18 or in related publications mentioned above. A simple example is described here to give the reader a sense of the goals, tactics, and problems associated with this approach. The application of linear discriminant analysis (DA) to ensemble neuronal data is perhaps the simplest of these techniques to understand because it is based upon concepts we are familiar with from more common statistical procedures such as analysis of variance and multiple regression. DA is a classification technique that can be used in the context of neuronal ensemble analysis to define different brain states (spatiotemporal patterns of activity) that relate to specific behavioral states (events). DA attempts to find the best linear combination of weighted variables that will separate experimental categories by maximizing the intergroup variance relative to the intragroup variance.3 This kind of analysis works for experimental designs based upon repeated occurrences such as trials found within operant self-administration paradigms. To conduct this type of analysis, each variable is an array of values that represents the spike firing of a given neuron, usually within a given time bin, with each value representing the spike firing for one trial. Each case, then, represents the spike firing across all variables (neurons) for a given trial. An illustration of a question we might ask is whether accumbens ensemble neural activity that occurs around the time of the operant response is sensitive to the presence or absence of the reinforcer. As an example, let us say we obtained simultaneous recordings from ten accumbens neurons from a subject whose operant responding was reinforced for the first half, but not for the second half, of the experimental session (intrasession extinction). We’ve recorded ensemble spike activity from trials for which the operant response was and was not followed by reinforcer. DA will allow us to test whether we can discriminate between the pattern of activity

2345_frame_MASTER.book Page 156 Wednesday, October 23, 2002 1:03 PM

156

Methods in Drug Abuse Research: Cellular and Circuit Level Analyses

A.

B. Bin spike counts 500 msec bin size Perievent Histograms

Create trial-by-trial spreadsheet Variables are columns of spike activity for one neuron during one time bin for every trial. The classification variable indicates the conditions (trial type). neurons 1 2 4 3 1 3

1 1 0 2 1 0

1 2 0 1 0 1

0 1 1 1 2 2

0 0 0 1 0 0

2 0 3 0 2 0

1 0 3 1 0 1 …

1 0 2 1 1 1

…

2

2 1 0 2 1 1

…

1

1 2 1 3 1 3

…

-2 -1 0 Time (sec)

1 1 1 2 2 2

…

trials

…

2

…

0 N r_2

…

-2

…

Condition 1_bin1 1_bin2 1_bin3 1_bin4 2_bin1 2_bin2 2_bin3 2_bin4 3_bin1 3_bin2 …

…

1.2 0.8 0.4 0

Operant response

…

Hz

Hz

Nr_1

1.5 1 0.5 0

… … … … … …

analysis period

C. Conduct stepwise linear discriminant analysis. 40 Variables entered. Variables remaining in analysis along with their discriminant function coefficients (cx): Variables c Discriminant Function: N02A06 -.783 For each case, N03A04 -.609 Discriminant score = c1x1 + c2 x2…. c9 x9 .576 .572 1.111 1.105 .567 1.016 .694

Discriminant Scores by Condition No of obs

N04A03 N04A05 N05C06 N05C07 N06A06 N08A02 N08A06

26 24 22 20 18 16 14 12 10 8 6 4 2 0

<= -5

(-5,-4] (-4,-3] (-3,-2] (-2,-1] (-1,0]

No Reinforcer

(0,1]

(1,2]

(2,3]

(3,4]

>4

Reinforcer

FIGURE 6.4 Steps in discriminant analysis of neuronal ensemble data. (A.) Spike train data are tabulated into time bins relative to a behavioral event. In this example, the average firing rate relative to the performance of the operant response (t = 0 on the x-axis) is depicted, with a 500-msec bin size. A 2-sec time period around the operant response is chosen for analysis. (B.) The spike activity of each neuron in each of the four 500-msec time bins that surround the operant response is tabulated for each trial. Thus, each vertical column in the spreadsheet represents one time bin for one neuron (variable). Each horizontal row in the spreadsheet represents all the data from one trial (case). The categorization variable on the far left, Condition, indicates the type of trial by using a numerical code. (C.) The spreadsheet is imported into an analysis program such as SPSS for step-wise discriminant analysis. This procedure determines if our variables can be used to discriminate between our two trial types and, if so, which variables contribute to the model. In this example, nine variables from six neurons were included in the model. The coefficient calculated for each variable is shown. A discriminant score for each trial case (trial) can be calculated by multiplying the value of the included variables for that trial by their respective coefficients. The frequency histogram of the discriminant scores obtained for each trial finds little overlap between the two trial types, indicating good discrimination. In this example, the model correctly classified the trial type for 91.8% of the trials (50% is chance).

within the ensemble on trials that were or were not reinforced. The data preparation and anlaysis steps are described briefly here and in Figure 6.4 (see Color Figure 6.4 following page 50). Let us say we recorded from these ten neurons across 100 trials (50 trials were reinforced; 50 trials were not reinforced), and we want to examine the 2-sec time period around the operant response. We now need to choose a relevant

2345_frame_MASTER.book Page 157 Wednesday, October 23, 2002 1:03 PM

Many-Neuron Microelectrode Array Recording and Reward-Seeking Behavior 157

time bin for analysis within this 2-sec time period. One might do this by trial and error or by examination of the single-neuron perievent histograms. If we chose 500msec time bins, we would have 40 variables in our analysis (4 bins ¥ 10 neurons). A DA would then find the weighted combination of all variables that would maximally discriminate between cases (trials) that were reinforced (Condition 1) and that were not reinforced (Condition 2). Some of the cases are used to build the model, and other cases are used to test the model, to control for overfitting the data. If the analysis is not able to classify the individual test trials with any likelihood greater than chance, then it is possible that the ensemble spike firing that occurs when the subject performs the operant response is insensitive to the outcome of the operant response. Classification better than chance might indicate that our ensemble does contain information about the outcome. The analysis can be used to determine the relative importance of individual neurons to the performance of the entire ensemble. Likewise, if an ensemble of neurons performs better than any one neuron, then we have strengthened the notion that it is networks of neurons, not individual neurons, that are the functional units that carry information in the nervous system. Of course, like many statistical techniques, the appropriate application of DA to any data set is based upon certain assumptions, and therefore it is critical that attention be paid to these aspects of the analysis to avoid false positive outcomes (see References 3 and 6 for more information). Analyses like DA are informative for these early stages of interpreting spatiotemporal patterns of neural activity but of course are necessarily based upon assumptions about the underlying biology. For example, by choosing a certain bin width for the data we are assuming that we have a relevant unit of measure to capture the phenomenon in which we are interested. If our time bins are too big, we might lose the fine grain necessary to sensitively detect certain interactions among neurons relative to ongoing behavior. However, if our time bins are very small, we increase the total number of variables in the analysis, contributing to a decrease in statistical power and a possible increase in highly correlated (redundant) variables that contribute to spurious results.3 In addition, by defining a small number of behavioral states a priori we are undoubtedly oversimplifying the explanation of patterns of neural activity. Analysis techniques that are unsupervised may provide more biological relevance by allowing structure to be determined within the data empirically.

6.5 FUTURE DIRECTIONS Advances in the technique of many-neuron recording are likely to continue and result in increased numbers of neurons that can be simultaneously recorded. Such advances will likely include changes in microelectrode design, in the online data-sampling capabilities of our hardware, and in the computational power for storing and analyzing acquired data. These technical advances will allow us to continue to refine the precision of our view of real-time brain function in the awake behaving animal. In the future, it is likely that increasing numbers of experiments will combine current techniques to provide a powerful means to address specific issues within the study of the neurobiological basis of drug self-administration. Pharmacology, neurochemistry, and molecular biology can all be combined with neurophysiology. For

2345_frame_MASTER.book Page 158 Wednesday, October 23, 2002 1:03 PM

158

Methods in Drug Abuse Research: Cellular and Circuit Level Analyses

example, many-neuron recording can be conducted in genetically altered mice to understand better the means by which specific genes may alter the neuronal activity that drives drug self-administration. The most important obstacle for many-neuron recording is not a difficulty but a challenge — in fact, it is the essential scientific problem in behavioral neurophysiology. This problem is one of interpretation of the neural data we record. We now have access to the actual millisecond-by-millisecond activity of networks of neurons in the awake functioning brain — but what does it all mean? Perhaps we are able to read the letters, but we are still struggling to understand the meaning of the words. There is much work to be done to decipher the information that we are now able to record from behaving animals. Understanding the means by which neuronal networks function to control addictive behavior, and all behavior in general, will be an exciting undertaking for years to come.

6.6 CONCLUSIONS Advances in our ability to acquire and store digital information now allow us to sample from large numbers of neurons simultaneously throughout the brain. It appears that the sampling rate is sufficient to permit a faithful record of the spike output of large ensembles of neurons, allowing us to examine in detail the relationship among individual neurons within an ensemble and between these neurons and the behavior of the animal. The pursuit of the nature of neural encoding that controls drug seeking will provide fundamental understanding of the neurobiological basis of addictive behavior.

ACKNOWLEDGMENTS This chapter was supported by award #DAMD17–01–1–0739 to PHJ awarded and administered by the U.S. Army Medical Research Acquisition Activity, 820 Chandler St., Fort Detrick, MD 21702. The content of information herein does not reflect the position or policy of the U.S. government, and no official endorsement should be inferred. The chapter was also supported by funds from the State of California for Medical Research on Alcohol and Substance Abuse through the University of California at San Francisco. Many thanks to M. Laubach for statistical guidance and formats for the visual presentation of LDA.

REFERENCES 1. Bowman, E.M., Aigner, T.G., and Richmond, B.J., Neural signals in the monkey ventral striatum related to motivation for juice and cocaine rewards, J. Neurophysiol,. 75, 1061, 1996. 2. Bragin, A., Hetke, J., Wilson, C.L., Anderson, D.J., Engel, J., Jr., and Buzsaki, G., Multiple site silicon-based probes for chronic recordings in freely moving rats: implantation, recording and histological verification, J. Neurosci. Methods, 98, 77, 2000.

2345_frame_MASTER.book Page 159 Wednesday, October 23, 2002 1:03 PM

Many-Neuron Microelectrode Array Recording and Reward-Seeking Behavior 159 3. Bray, J. H. and Maxwell, S.E., Multivariate Analysis of Variance, Sage Publications, Newbury Park, CA, 1985. 4. Chang, J.-Y., Chen, L., Luo, F., Shi, L.-H., and Woodward, D.J., Neuronal responses in the frontal cortico-basal ganglia system during delayed matching-to-sample task: ensemble recording in freely moving rats, Exp. Brain Res., 142, 67, 2002. 5. Chang, J. Y., Janak, P.H., and Woodward, D.J., Neuronal and behavioral correlations in the medial prefrontal cortex and nucleus accumbens during cocaine self-administration by rats, Neuroscience, 99, 433, 2000. 6. Chapin, J. K., Population-level analysis of multi-single neuron recording data: multivariate statistical methods, in Methods for Neural Ensemble Recordings, Nicolelis, M.A., Ed., CRC Press, Boca Raton, FL, 1999, pp. 193–228. 7. Chapin, J. K., Moxon, K.A., Markowitz, R.S., and Nicolelis, M.A., Real-time control of a robot arm using simultaneously recorded neurons in the motor cortex, Nat. Neurosci., 2, 664, 1999. 8. Deadwyler, S.A., Bunn, T., and Hampson, R.E., Hippocampal ensemble activity during spatial delayed-nonmatch-to-sample performance in rats, J. Neurosci., 16, 354, 1996. 9. Georgopoulos, A.P., Crutcher, M.D., and Schwartz, A.B., Cognitive spatial-motor processes. 3. Motor cortical prediction of movement direction during an instructed delay period, Exp. Brain Res., 75, 183, 1989. 10. Georgopoulos, A.P. and Massey, J.T., Cognitive spatial-motor processes. 2. Information transmitted by the direction of two-dimensional arm movements and by neuronal populations in primate motor cortex and area 5, Exp. Brain Res., 69, 315, 1988. 11. Gerstein, G.L., Perkel, D.H., and Dayhoff, J. E., Cooperative firing activity in simultaneously recorded populations of neurons: detection and measurement, J. Neurosci., 5, 881, 1985. 12. Janak, P.H., Multichannel neural ensemble recording during alcohol self-administration, in Liu, Y. and Lovinger, D.M., Eds., Methods for Alcohol-Related Neuroscience Research, CRC Press, Boca Raton, FL, 2002, pp. 243–259. 13. Kubie, J.L., Muller, R.U., and Hawley, E.S., Single-cell recording in awake behaving animals, in Neuroscience Labfax, Lynch, M.A. and O’Mara, S.M., Eds., Academic Press, San Diego, 1997. 14. Laubach, M., Wessberg, J., and Nicolelis, M.A., Cortical ensemble activity increasingly predicts behaviour outcomes during learning of a motor task, Nature, 405, 567, 2000. 15. Lemon, R. and Prochazka, A., Methods for Neuronal Recording in Conscious Animals, John Wiley & Sons, Chichester, 1984. 16. Moore, G.P., Perkel, D.H., and Segundo, J.P., Statistical analysis and functional interpretation of neuronal spike data, Annu. Rev. Physiol., 28, 493, 1966. 17. Moxon, K.A., Multichannel electrode design: considerations for different applications, in Methods for Neural Ensemble Recordings, Nicolelis, M.A.L., Ed., CRC Press, Boca Raton, FL, 1999, pp. 25–45. 18. Nicolelis, M.A., Methods for Neural Ensemble Recordings, CRC Press, Boca Raton, FL, 1999. 19. Nicolelis, M.A.L., Stambaugh, C.R., Brisben, A., and Laubach, M., Methods for simultaneous multisite neural ensemble recordings in behaving primates, in Methods for Neural Ensemble Recordings, Nicolelis, M.A.L., Ed., CRC Press, Boca Raton, FL, 1999, pp. 121–156.

2345_frame_MASTER.book Page 160 Wednesday, October 23, 2002 1:03 PM

160

Methods in Drug Abuse Research: Cellular and Circuit Level Analyses

20. Optican, L.M. and Richmond, B.J., Temporal encoding of two-dimensional patterns by single units in primate inferior temporal cortex. III. Information theoretic analysis, J. Neurophysiol., 57, 16, 1987. 21. Perkel, D.J., Gerstein, G.L., and Moore, G.P., Neuronal spike trains and stochastic point processes. I. The single spike train, Biophys. J,. 7, 391, 1967. 22. Perkel, D.H. and Bullock, T.H., Neural coding, Neurosci. Res. Program Bull., 6, 221, 1968). 23. Richmond, B.J. and Optican, L.M., Temporal encoding of two-dimensional patterns by single units in primate inferior temporal cortex. II. Quantification of response waveform, J. Neurophysiol., 57, 147, 1987. 24. Richmond, B.J., Optican, L.M., Podell, M., and Spitzer, H., Temporal encoding of two-dimensional patterns by single units in primate inferior temporal cortex. I. Response characteristics, J. Neurophysiol,. 57, 132, 1987. 25. Rieke, F., Warland, D., de Ruyter van Steveninck, R., and Bialek, W., Spikes: Exploring the Neural Code, MIT Press, Cambridge, MA, 1997. 26. Sameshima, K. and Baccala, L.A., Trends in multichannel neural ensemble recording instrumentation, in Methods for Neural Ensemble Recordings, Nicolelis, M.A.L., Ed., CRC Press, Boca Raton, FL, 1999, pp. 47–60. 27. Seidemann, E., Meilijson, I., Abeles, M., Bergman, H., and Vaadia, E., Simultaneously recorded single units in the frontal cortex go through sequences of discrete and stable states in monkeys performing a delayed localization task, J. Neurosci., 16, 752, 1996. 28. Tabuchi, E.T., Mulder, A.B., and Wiener, S.I., Position and behavioral modulation of synchronization of hippocampal and accumbens neuronal discharges in freely moving rats, Hippocampus, 10, 717, 2000. 29. Wessberg, J., Stambaugh, C.R., Kralik, J.D., Beck, P.D., Laubach, M., Chapin, J.K., Kim, J., Biggs, S.J., Srinivasan, M.A., and Nicolelis, M.A., Real-time prediction of hand trajectory by ensembles of cortical neurons in primates, Nature, 408, 361, 2000. 30. Wheeler, B.C. and Heetderks, W.J., A comparison of techniques for classification of multiple neural signals, IEEE Trans. Biomed. Eng., 29, 752, 1982. 31. Williams, J.C., Rennaker, R.L., and Kipke, D.R., Long-term neural recording characteristics of wire microelectrode arrays implanted in cerebral cortex, Brain Res. Protocol, 4, 303, 1999. 32. Woodward, D.J., Janak, P.H., and Chang, J.Y., Ethanol action on neural networks studied with multineuron recording in freely moving animals, Alcohol Clin. Exp. Res., 22, 10, 1998.

2345_frame_MASTER.book Page 161 Wednesday, October 23, 2002 1:03 PM

7

Application of Chronic Extracellular Recording to Studies of Drug Self-Administration Laura L. Peoples

CONTENTS 7.1 7.2

7.3

7.4

Chapter Overview ........................................................................................162 Advantages and Limitations ........................................................................162 7.2.1 Rationale and Advantages................................................................162 7.2.1.1 General..............................................................................162 7.2.1.2 Intravenous Drug Self-Administration .............................163 7.2.1.3 Chronically Implanted vs. Movable Electrodes ...............163 7.2.1.4 Anatomical Resolution of the Recording Technique.......164 7.2.1.5 Incentive Motivation Theories..........................................164 7.2.2 Challenges and Limitations .............................................................165 7.2.2.1 Drug Effect vs. Behavioral Feedback ..............................165 7.2.2.2 Multiple Drug Effects.......................................................166 7.2.2.3 Mechanisms of Drug Action ............................................166 7.2.3 Overview of Utility ..........................................................................167 Methods ........................................................................................................167 7.3.1 Overview of Experimental Procedures............................................167 7.3.2 Electrophysiological Recording Session .........................................169 7.3.3 Analysis of Electrophysiological Data ............................................169 7.3.3.1 Waveform Analysis...........................................................169 7.3.3.2 ISI Analysis.......................................................................171 7.3.4 Quantitative and Statistical Analyses of Firing Patterns.................175 7.3.4.1 Individual Neuron Data ....................................................175 7.3.4.2 Group Mean Neural Data .................................................178 Example Investigations ................................................................................178 7.4.1 Behavior ...........................................................................................179 7.4.2 Incentive-Related Information Encoding.........................................179 7.4.2.1 Phasic Firing Patterns Time-Locked to Cocaine Self-Infusion .....................................................................179 7.4.2.2 Information Encoded by the Lever-Press Firing Patterns..............................................................................180 161

2345_frame_MASTER.book Page 162 Wednesday, October 23, 2002 1:03 PM

162

Methods in Drug Abuse Research: Cellular and Circuit Level Analyses

7.4.3

Acute Drug Effects ..........................................................................184 7.4.3.1 Firing Patterns That Mirror Changes in Drug Level .......184 7.4.3.2 Dose-Dependent Changes in Firing Rate.........................184 7.4.3.3 Changes in Firing Rate: Drug Effect vs. Behavioral Feedback ...........................................................................185 7.4.3.4 Drug Effects and Accumbal Information Processing ......188 7.4.3.5 Drug Effect That Mediates Accumbal Role in Cocaine Reward ..............................................................................189 7.4.4 Effects of Repeated Self-Administration Sessions..........................190 7.4.5 Histological Analyses.......................................................................192 7.5 Future Directions..........................................................................................194 Acknowledgments..................................................................................................196 Appendices.............................................................................................................196 A.1 Instrumentation.............................................................................................196 A.1.1 Intravenous Catheter and Swivel .....................................................196 A.1.2 Microwire Array Headset and Electronic Harness..........................197 A.1.3 Tethering System..............................................................................198 A.1.4 Operant Chambers............................................................................200 A.1.5 Electrophysiological Equipment ......................................................201 A.2 Surgical Procedures......................................................................................202 A.3 Post-Operative Care .....................................................................................203 A.4 Histology ......................................................................................................204 References..............................................................................................................205

7.1 CHAPTER OVERVIEW To further the investigation of the neurobiology of drug reward and addiction, researchers have integrated chronic extracellular recording and intravenous drug selfadministration procedures. The combined technique is the focus of the present chapter. The chapter is divided into four main sections. The first section reviews advantages and limitations of the method. The second section reviews the particular procedures that our laboratory has used to make and analyze chronic extracellular recordings of accumbal neural activity in rats self-administering cocaine. The third section describes a few studies that demonstrate the application and utility of the technique. The chapter closes with a look at future directions.

7.2 ADVANTAGES AND LIMITATIONS 7.2.1 RATIONALE

AND

ADVANTAGES

7.2.1.1 General Most if not all theories of addiction propose that interactions between drug effects and reward-related processes contribute importantly to the development of addiction.39,73,95 Reward-related processes involve activation of particular neural circuits that mediate and encode interactions between animals and the environment. These

2345_frame_MASTER.book Page 163 Wednesday, October 23, 2002 1:03 PM

Chronic Extracellular Recording to Studies of Drug Self-Administration

163

neural and experiential events are expected to be wholly present in only the awake animal and only under certain behavioral conditions. It is thus possible that drug effects that are integral to the development of addiction are observable and subject to investigation in only behaving animals exposed to behavioral conditions similar to those associated with human drug taking. Appreciation of this fact is an important impetus for the application of in vivo measurement techniques to awake animals exposed to conditions relevant to drug seeking and drug taking in humans (Waterhouse and Peoples, Chap. 1, this volume). Historically, the emphasis of these in vivo measurements has been the characterization of changes in neurochemical activity. Although an understanding of presynaptic mechanisms is important in delineating a complete picture of the neurobiology of addictive drugs, it is unlikely to be sufficient. It is not changes in neurochemical activity alone but rather the influences of such changes on the activity of post-synaptic neurons that transduce most drug effects on behavior. Thus, a full understanding of the mechanisms that mediate addictive effects of drugs will most likely depend on characterization of post-synaptic neurophysiological responses. Application of the chronic extracellular recording technique to the intravenous selfadministration paradigm provides a method that can be used to study these postsynaptic events. 7.2.1.2 Intravenous Drug Self-Administration A variety of behavioral paradigms are used to investigate neuropharmacological mechanisms that contribute to addiction (for review see Reference 7). Among them, intravenous drug self-administration is thought to offer the greatest face validity and predictive power. In the simplest intravenous drug self-administration procedure, an instrumental behavior, such as pressing a lever, is reinforced by the intravenous infusion of an addictive drug. However, the procedure can be more complex and structured so as to establish the various relations among stimuli, behavior, and drug exposure that are encountered by human addicts.2,4,25,89 Animals exposed to the intravenous drug self-administration procedure acquire patterns of drug seeking and drug taking that are quite similar to those exhibited by humans24,28,35,52,70,84 (see also Dworkin and Stairs, Chap. 2, this volume). Moreover, brain areas and neurochemicals implicated in mediating drug self-administration in animals are consistent with regions implicated in drug reward and addiction in humans.8,20,27,88 Given the predictive power and face validity of drug self-administration it is the behavioral paradigm of choice for in vivo investigations of the neurobiology of addiction. 7.2.1.3 Chronically Implanted vs. Movable Electrodes Extracellular recordings in freely moving animals can be conducted with either chronically implanted microwires or movable electrodes. The chronic-implant method is the only recording method applied thus far to drug self-administration studies. This is in part because stable recordings of single neurons can be more readily maintained for long periods of time if one uses the chronic recording procedures rather than more traditional acute methods. Maintenance of long-duration recordings is particularly important in studies of drug self-administration. For exam-

2345_frame_MASTER.book Page 164 Wednesday, October 23, 2002 1:03 PM

164

Methods in Drug Abuse Research: Cellular and Circuit Level Analyses

ple, in most circumstances, successive occurrences of the self-infusion behavior are separated by long intervals (min). It is thus often necessary to record the activity of an individual neuron for hours to acquire the data needed to make a reliable assessment of the firing pattern of a neuron in relation to the behavior. Additionally, some of the drug effects thought to contribute to drug addiction occur across repeated days and weeks of drug self-administration. Characterization of neurophysiological changes over periods of days and weeks is more feasible when chronic rather than the acute recording procedures are used. 7.2.1.4 Anatomical Resolution of the Recording Technique In chronic extracellular recording studies, the location of a recorded neuron can be identified with a spatial resolution of approximately 100 mm. This resolution exceeds that of numerous other neuroscience techniques. A number of structures implicated in drug reward and addiction are relatively small and heterogenous in neurochemical innervation and connectivity. Analysis of the anatomical distribution of single neurons that exhibit particular firing patterns during drug self-administration sessions could therefore contribute importantly to understanding the contribution of different regions and subregions of the brain to drug reward and addiction. 7.2.1.5 Incentive Motivation Theories Although the chronic recording technique is applicable to addressing a wide variety of hypotheses related to drug addiction, much of the research that has thus far been conducted has focused on the incentive motivation theories of drug addiction. These theories hypothesize that the disorder reflects a pathological responsivity of individuals to the influences of drug-associated conditioned stimuli on behavior. It is further proposed that the abnormal responsivity to drug stimuli is caused by acute actions of addictive drugs on the brain. The acute actions are believed to amplify mechanisms that contribute to stimulus–reward learning and lead to abnormally powerful conditioning of stimuli associated with the drug. It is also proposed that long-lasting changes in the brain induced by the drug, specifically those associated with sensitization, facilitates this drug-induced amplification of learning.23,26,72,73,83,84 The incentive motivation theories of addiction make a number of predictions. First, acute drug actions will amplify neural signals related to either of the following: (1) the encoding of stimulus–reward associations, (2) the influence of those associations on pavlovian and instrumental behavior, or (3) both. Second, repeated exposure to these acute drug actions will lead to a lasting enhancement in neural signals related to stimulus–reward associations; moreover, this enhancement of neural signals will be correlated with an increase in cue-controlled drug seeking. Third, neurons involved in encoding stimulus–reward associations (or the influence of those associations on behavior) will show sensitization to acute drug actions after repeated exposure to drug (Incentive Sensitization Theory73,83). The chronic extracellular recording technique is particularly well suited to investigate these predictions. The temporal base of the measurements that can be made with the extracellular recording technique is unusually flexible, spanning a range from 1 msec to weeks. It is thus possible to measure neural responses to both of the following:

2345_frame_MASTER.book Page 165 Wednesday, October 23, 2002 1:03 PM

Chronic Extracellular Recording to Studies of Drug Self-Administration

165

(1) discrete stimulus and reward events that can occur in a time frame of millisecond or seconds and (2) drug actions that can occur over periods of either minutes, hours, days, or weeks. Given this temporal range of the recording measurements, the technique can be used to test each of the three aforementioned predictions. Specifically, the technique can be used to test for interactions between acute drug actions and neural encoding of discrete stimulus and reward events (i.e., first prediction) (see Section 7.4.2). Moreover, it can be used to test for changes across repeated drug exposures in either the neural encoding of drugassociated stimuli or the neural response to drug itself (i.e., the second and third predictions) (see Section 4.3). Thus, application of the recording technique provides a tool that is “tailor-made” for investigation of drug addiction within the context of the incentive motivation theories.

7.2.2 CHALLENGES

AND

LIMITATIONS

7.2.2.1 Drug Effect vs. Behavioral Feedback An important challenge associated with investigating the neurophysiological mechanisms that mediate a particular behavioral effect of a drug is to discriminate between the following: (1) changes in neural activity that are due to actions of the drug and (2) changes in neural activity that reflect nonpharmacological encoding of the execution or monitoring of behavior. A control used to differentiate between these two types of changes is called the “clamping” procedure. The goal of the control is to make comparisons of different drug conditions (e.g., drug vs. no drug) across periods in which behavior is maintained constant by the experimenter. If the drug-correlated changes in neural activity disappear when behavior is held constant, it is possible that the neural change is related to the behavior rather than to an effect of the drug on the neuron. On the other hand, if the drug-correlated changes in firing are unaltered by the behavioral clamp, it is likely that the neural changes are unrelated to changes in behavior and hence may be pharmacological in origin.71 One can clamp behavior by forcing animals to engage in an unconditioned behavior such as walking on a treadmill17,91 or by using behavioral conditioning methods that motivate animals to engage in particular patterns of behavior. For example, it is possible to reinforce an animal to engage in a behavior under nondrug conditions that is comparable to that which is typically induced by the drug.57,91 Another clamping strategy that we have employed is to allow animals to engage in normal (self-administration) behavior and to then limit comparisons of neural firing to select periods of time in which behavior is topographically the same.61 This latter approach is difficult to apply to drug vs. nondrug comparisons but can be used in “between-dose” comparisons. As a cautionary note, in designing a behavioral clamp control, one should keep in mind that drug actions can vary as a function of the state of a neural circuit. It may thus be important that the clamping conditions are as comparable as possible to the conditions that are normally associated with the drug effect of interest. The utility and relevance of the clamping procedure to neurophysiological investigations of drug self-administration have already been demonstrated in numerous experiments. Examples are described in Section 7.4 of this chapter.

2345_frame_MASTER.book Page 166 Wednesday, October 23, 2002 1:03 PM

166

Methods in Drug Abuse Research: Cellular and Circuit Level Analyses

7.2.2.2 Multiple Drug Effects Psychoactive drugs, including addictive drugs, have multiple effects on behavior, typically mediated by drug actions at multiple sites. This pharmacological profile poses another challenge to interpreting changes in neural activity recorded in behaving animals exposed to drug. Specifically, it is necessary to discriminate changes in firing that mediate the particular drug effect of interest from changes in firing that may mediate other drug effects on behavior. One method that can be used to make this discrimination is to evaluate the tightness of the relationship between the time course of the neural changes and the time course of the various drug effects on behavior. Additionally, one can evaluate the tightness of the dose response curves for the neural change and the various drug effects.22,49,63,67 The time courses and the dose-response curves of different drug effects are typically distinct. Thus, if a particular neural change mediates the drug-induced behavior of interest, the time course and the dose-response curve of the neural change are expected to be tightly related to those for the target behavior but not for other drug-induced behaviors. The extent to which the presence of multiple drug effects confounds interpretations of firing patterns can be minimized by limiting neural recordings to regions already implicated as necessary substrates of the target drug effect. Given the appropriate behavioral and pharmacological conditions, firing patterns exhibited by neurons in that brain region are highly likely to mediate the drug effect of interest. 7.2.2.3 Mechanisms of Drug Action In evaluating the post-synaptic events that contribute to a drug effect on behavior it is ultimately advantageous to delineate the specific drug actions that mediate the changes in neural activity. Characterizations of these mechanisms bridge studies of presynaptic and post-synaptic mechanisms and represent a final step in developing a complete picture of the transduction of the drug effect. There are a number of procedures that are typically combined with acute electrophysiological recording techniques to elegantly address these types of mechanistic questions. Unfortunately, some of those methods are not readily applied to a chronic recording study. In acute intracellular recording preparations, dye and tracing techniques can be used to identify the specific type of neuron from which neural recordings are made; moreover, the techniques can delineate circuitry that may either contribute to or be impacted by drug-induced changes in neural activity. The use of these methods in chronic extracellular recording studies is precluded by the chronic and extracellular placement of solid microwires. Another method for investigating the mechanisms of drug action is local application (i.e., iontophoretic or microinjection) of specific receptor agonists and antagonists. This experimental strategy is effectively employed in electrophysiological studies conducted in the anesthetized animal but has not yet been applied to chronic extracellular recordings. The barrier in applying the technique to the chronic recording studies is the development of a drug-delivery system that will not interfere with the unique stability of the recordings provided by the microwires. This stability is thought to depend on the low mechanical resistance of the fine and flexible wires that prevents the wires from displacing or otherwise disturbing

2345_frame_MASTER.book Page 167 Wednesday, October 23, 2002 1:03 PM

Chronic Extracellular Recording to Studies of Drug Self-Administration

167

tissue around the wire. It is possible that the stability of the recordings would be disturbed by local placement of additional instrumentation near the recording wires. Nevertheless, there are ongoing efforts to develop this technology. Despite these limitations, it should be possible to make some progress in addressing mechanistic questions in chronic recording studies. For example, electrical stimulation techniques could aid in the identification of the afferent and efferent circuitry that contributes to or is impacted by drug-induced changes in the activity of the recorded neuron.22 Microinjection of pharmacological probes into regions that project to the recorded neurons may similarly be useful. Finally, it may be informative to determine whether changes in neural firing associated with systemic drug administration in behaving animals are consistent with effects of region-specific drug administration in acute recording studies. Similar findings obtained in the two types of studies may be indicative of the mechanisms delineated in the acute recording studies to the behaving animal.

7.2.3 OVERVIEW

OF

UTILITY

To summarize, researchers must contend with technical challenges and limitations when using the chronic extracellular recording technique to address questions about the mechanisms of drug action that mediate changes in neural activity. Despite these difficulties, use of the combined method in studies of neural mechanisms that mediate drug reward and drug addiction offers a number of advantages. Chronic recordings in animals self-administering drug offer powerful tests of hypotheses regarding the neurophysiological mechanisms that mediate drug reward and addiction. This is particularly true under the following conditions: (1) recordings are conducted in a brain area that is known to be necessary for the drug effect to occur; (2) a contribution of behavioral feedback on neural firing is controlled for; and (3) the neural change is tightly correlated with the time course and dose–response curve of drug effects on self-administration behavior. It is also important to note that the chronic extracellular recording technique has good anatomical resolution that should be useful in refining our understanding of sub-regional differences in the contribution of heterogenous structures to drug seeking, reward, and addiction. Perhaps most important, given the temporal characteristics of the combined method, it is unique in allowing investigations of interactions between either acute or chronic drug actions and the neurophysiological events that mediate the complex behavioral processes, such as stimulus–reward learning, that are hypothesized to contribute to drug addiction.

7.3 METHODS 7.3.1 OVERVIEW

OF

EXPERIMENTAL PROCEDURES

Long-Evans rats are chronically implanted with a catheter in the jugular vein and an array of microwires in the nucleus accumbens (Appendices A.1.1, A.1.2, A.2). Following surgical recovery, animals are placed in Plexiglas operant chambers that are used henceforth to house the animals and to conduct behavioral and electrophysiological sessions (Appendix A.1.4; Figures 7.1 and 7.2).

2345_frame_MASTER.book Page 168 Wednesday, October 23, 2002 1:03 PM

168

Methods in Drug Abuse Research: Cellular and Circuit Level Analyses

FIGURE 7.1 Rat and operant chamber. Photograph shows a subject pressing the lever associated with cocaine reward. The electrode cap and headstage of the harness are visible at the top of the animal’s head. The spring leash encasing the catheter is just posterior to the cap and harness.

FIGURE 7.2 Full view of operant chamber. Photograph provides a full view of the Plexiglas operant chamber that was used to house the animals and conduct the self-administration and recording sessions. The spring leash encasing the catheter is connected to a fluid swivel that is mounted on a counterbalanced arm. For more details see Appendices A.1.3 and 7.1.4.

2345_frame_MASTER.book Page 169 Wednesday, October 23, 2002 1:03 PM

Chronic Extracellular Recording to Studies of Drug Self-Administration

169

Animals are then trained to intravenously self-administer cocaine. The onset of the session is signaled by manual insertion of a response lever into the chamber and the illumination of a stimulus light. During the remainder of the session each press of the lever is followed by an intravenous infusion of cocaine (0.7 mg/kg/0.2 ml). Each infusion is paired with the sounding of a 7.5-sec tone that corresponds to the operation of the syringe pump. A stimulus light above the lever additionally turns off for 40 sec. Training sessions are conducted 6 hours a day, 7 days per week.

7.3.2 ELECTROPHYSIOLOGICAL RECORDING SESSION The extracellular recording session typically consists of three successive phases: (1) a predrug baseline recording period in which animals are awake but resting quietly (20 to 40 min), (2) an intravenous drug self-administration session (6 hours), and (3) a post-drug-recovery period (40 to 60 min). During the recording session neural signals are led through an electrical cable (referred to as a harness) and a modified fluid and electronic swivel to a preamplifier that differentially amplifies the signal on the recording wire against another microwire. The signal then is led through a bandpass filter and amplifier. With the use of software and hardware from DataWave Technologies Corp., Longmont, CO, electrical signals are sampled and digitized, time-stamped, and stored for offline analysis. The DataWave system additionally controls and records the occurrence of all stimulus and behavioral events during the recording session (see Figure 7.1, Appendices A.1.1–A.1.5).

7.3.3 ANALYSIS

OF

ELECTROPHYSIOLOGICAL DATA

7.3.3.1 Waveform Analysis The primary objective of the extracellular recording studies is to characterize the activity of individual neurons and networks of individual neurons in relation to behavioral and pharmacological variables.74 (see also Janak, Chap. 6, this volume, and Waterhouse and Peoples, Chap. 1, this volume). Signals recorded with an electrode can include non-neural signals (electrical artifact) and signals that correspond to discharges of either one neuron or multiple neurons. To meet the objectives of the “single-neuron” recording studies it is thus necessary to isolate the neural signals from the non-neural signals and to discriminate between the signals of different neurons. Each electrical signal recorded during the experiment consists of a sequence of changes in electrical potential that is referred to as the “waveform.” Neural discharges (i.e., action potentials) are associated with characteristic waveforms. Moreover, assuming that recording conditions are held constant, individual neurons tend to exhibit waveforms that show limited variation in shape and amplitude. Often, although not always, the waveforms corresponding to one neuron are different from those of other neurons recorded with the same electrode. Given the above, discrimination among waveforms on the basis of parameters that index shape and amplitude of the waveforms is used to differentiate neural and nonneural data and to discriminate between the discharges of distinct neurons.

2345_frame_MASTER.book Page 170 Wednesday, October 23, 2002 1:03 PM

170

Methods in Drug Abuse Research: Cellular and Circuit Level Analyses

To conduct recordings and discriminate single neuron data we have thus far used products of DataWave Technologies Corp. The system is a digital storage system that records both a digitized version of the waveform for each electrical signal and a time stamp marking the time at which the signal occurred during the course of the recording session. Given the digitized record, it is possible for the original analog signals to be reconstructed and for the researcher to isolate and discriminate neural waveforms offline at anytime after the recording session. Other systems that perform online spike sorting93 sometimes transmit only the time stamp and sorted identity (i.e., neuron 1 vs. neuron 2) to computer storage (referred to as an event timing system).77 In this case, the researcher is committed to the sorting, or discrimination, criteria established online during the experiment. When we conduct the offline discrimination, we isolate waveform populations corresponding to single neurons on the basis of eight parameters (see Figure 7.3). Measurements are made of each waveform for all eight parameters. The parameter values for all the recorded signals are displayed in scattergrams that plot waveform parameters, one against another. Waveforms that correspond to a given neuron tend

Valley

Amplitude

Peak

1000 mV

FIGURE 7.3 Waveform parameters and discrimination of single-neuron data. Left: An overlay of successively recorded waveforms corresponding to a single neuron. Positive voltage is up. Each waveform trace spans 0.64 msec. Vertical calibration bar = 1000 mV. Voltages correspond to the amplified signal. Total amplification of original neural signal is approximately 31,200. Four of the eight waveform parameters that we apply in our characterization of waveform shape and amplitude are highlighted and include the following: peak voltage, valley voltage, total waveform amplitude (i.e., sum of the absolute values of peak and valley voltages) (referred to as spike height in the figure), and voltage at a midpoint on the ascending limb of the waveform (i.e., vertical line through ascending limb of waveform). The values for these parameters are determined for every electrical signal recorded by the electrode. These values are then plotted in scattergrams. Right: A scattergram shows a plot of one waveform parameter, voltage on the ascending limb, against another parameter, spike height. Each point in the scattergram corresponds to a single electrical signal. Any of the signals successively recorded in a session can be displayed in such a scattergram, although the maximum number of signals that can be displayed in a single scattergram is 4096. Waveforms that correspond to discharges of a single neuron tend to cluster together. Such a cluster of waveforms is shown in the center of the scattergram and is surrounded by a box. The box around the cluster is drawn by the experimenter and sets maximum and minimum values against which all recorded signals are indexed. Similar scattergrams are drawn for additional combinations of parameters, and maximum and minimum settings for those parameters are established. The cluster shown in this scattergram and the waveforms on the left correspond to the same neuron.

2345_frame_MASTER.book Page 171 Wednesday, October 23, 2002 1:03 PM

Chronic Extracellular Recording to Studies of Drug Self-Administration

171

to cluster together in these scattergrams. On the basis of the distribution of the clusters in the scattergram, minimum and maximum values can be set for the parameters (see Figure 7.3). Waveforms that fall within the minimum and maximum values for ALL the parameters are classified as likely to correspond to a single neuron. Any waveform that falls outside the limits set for even a single parameter is excluded and considered likely to correspond to either electrical artifact or discharges of another neuron. Conducting the discrimination analysis offline can be laborious and time consuming. However, it offers the opportunity to review all recorded signals before setting the discrimination parameters. This capacity can be advantageous. Although recordings obtained with the chronically implanted microwires are highly stable, shifts in the amplitude of the neural waveforms can occur from day to day, and on occasion within a behavioral session. Offline discrimination procedures give the researcher the opportunity to detect these changes and to adjust discrimination parameters so as to account for them. In the absence of this capacity, for example, in the case of an event timing system, analysis of firing rates and firing patterns can be confounded by changes in the accuracy with which discharges of the same single neurons are detected, recorded, and thus counted over the course of the session. 7.3.3.2 ISI Analysis Although waveform discrimination is the gold standard for discriminating between signals that correspond to different neurons, it is possible for multiple neurons proximal to a given electrode to produce overlapping distributions of waveform shape and amplitude. Additionally, waveforms corresponding to a single neuron can differ enough so as to appear to correspond to more than one neuron. Thus, basing discrimination of neural data solely on the basis of waveform analysis can lead to multiple neurons being treated as a single neuron and vice versa. Both errors can produce erroneous conclusions regarding the activity of individual neurons. To minimize the risk of these errors and to increase the rigor with which we discriminate single neuron data we supplement the waveform discrimination procedure with an analysis of interspike-interval (ISI) histograms. An ISI histogram is a frequency distribution of the intervals between all the successive signals included in the discriminated population (i.e., cluster) of waveforms. A cleanly discriminated single neuron is expected to show a minimum ISI consistent with a physiological refractory period. In most circumstances, a similar minimum ISI is unlikely to be observed when the data of two or more independent neurons are combined into one cluster. The timing of the discharges of the two neurons, relative to one another, is not restricted and thus may occur within the intervals shorter than the refractory period of a single neuron. Evidence of a minimum ISI can thus provide confirmatory evidence that a discriminated population of waveforms corresponds to a single neuron.42,46,68 We have observed that populations of waveforms that appear to correspond to single accumbal neurons typically show a minimum ISI between 2 and 5 msec. Minimum ISIs as long as 2 to 5 msec are not typically observed in cases where we have recorded two clearly distinguishable populations of waveforms and deliberately treated them as a single neuron (Figure 7.4). Given these observations,

2345_frame_MASTER.book Page 172 Wednesday, October 23, 2002 1:03 PM

172

Methods in Drug Abuse Research: Cellular and Circuit Level Analyses

FIGURE 7.4 Interspike interval histograms and discrimination between discharges of different neurons. This figure provides a demonstration of our use of the ISI histogram to evaluate whether discriminated populations of waveforms correspond to distinct neurons. The ISI histograms display all ISIs £ 5 msec. Number of intervals (counts, y-axis) is plotted as a function of 0.1-msec increments in interval length (x-axis). Above each of the top two histograms is a sample of the corresponding neural waveform. The sample consists of the first 50 consecutively recorded waveforms of the recording session. Positive voltage is up. Each waveform trace spans 0.64 msec. Vertical calibration bar = 1000 mV. The top and middle rows of the figure show the waveform and ISI for two discriminated populations of waveforms (clusters) that were nonoverlapping in a number of scattergram displays of waveform parameters (not shown) and were thus likely to correspond to discharges of two neurons. To further evaluate this possibility we compared the ISI histograms of the individual clusters to the ISI histogram obtained when the clusters were combined and treated as one. The bottom row shows the ISI histogram obtained when the two clusters were combined. The ISI histogram for each of the individual clusters, but not for the combined cluster, showed evidence of a ≥2-msec minimum ISI. The loss of the long-duration minimum ISI upon combining the clusters was consistent with the waveform analysis in indicating that the two individual clusters correspond to two, rather than one, neurons. The clusters were therefore treated as two neurons.

populations of waveforms that do not yield evidence of the typical minimum ISI are presumed to include either signals of multiple neurons or a combination of neural and nonneural data. In our laboratory, such populations of waveforms are not treated as corresponding to single neurons (i.e., are not included in our analyses). We thus have greater confidence that the firing patterns that we observe reflect the activity of individual neurons rather than an average pattern exhibited by a group of neurons.

2345_frame_MASTER.book Page 173 Wednesday, October 23, 2002 1:03 PM

Chronic Extracellular Recording to Studies of Drug Self-Administration

173

In addition to facilitating discrimination among neurons, the ISI analysis can also help to insure that waveforms that correspond to a single neuron are not mistakenly subdivided into subpopulations and treated as corresponding to multiple neurons. During the waveform-discrimination procedure, differences in waveforms, particularly amplitude, can be large enough to create uncertainty about whether the waveforms correspond to a single neuron or multiple neurons. This is particularly true when discrimination is based on only waveform overlays (i.e., Figure 7.3, left) rather than plots of waveform clusters or distributions. Analysis of the ISI histograms can inform on the likelihood that the subpopulations should be treated as one or more neurons. Specifically, if the minimum ISI obtained when the subclusters are combined is consistent with that of a single neuron and comparable to the minimum ISIs observed when the two subpopulations are evaluated separately, it is likely that the two clusters correspond to the same neuron. On the other hand, if the ISI analysis shows that evidence of a minimum ISI is apparent when the two samples are analyzed separately but not when they are combined, it is likely that the two samples correspond to distinct neurons. Although subdivision of single-neuron data is less often discussed as a potential concern than is inadequate discrimination between two neurons, it is an issue that deserves consideration. We have observed that subdividing populations of waveforms that appear to correspond to the activity of a well-discriminated single neuron can confound analyses of neural activity by “creating” firing patterns that are not actually exhibited by a single neuron. This distortion of firing patterns can occur in two circumstances. One situation that can lead to misrepresentative firing patterns involves neurons that show an interaction between firing rate and waveform amplitude. We have observed that some accumbal neurons show decreases in waveform amplitude as the firing rate of the neuron increases. This relation between firing rate and waveform amplitude has been observed in other structures. If one discriminates between small and large amplitude waveforms that correspond to such a neuron, the two subclusters differentially include discharges that occur at particular time points relative to events around which the neuron fires phasically. The sample of small waveforms will include those discharges that occur at the time at which the peak firing rate occurs. The sample of large waveforms will, on average, include discharges that occur at time points removed from the point of peak firing. Thus, plots of firing rate relative to the event will be different for the two clusters and will not correspond to the real pattern of firing exhibited by any one neuron (e.g., Figure 7.5). A second circumstance in which subdividing single neuron data can lead to misrepresentative firing patterns involves neurons that show changes in waveform amplitude in response to drug exposure. For example, we have found that for at least some accumbal neurons the amplitude of the neural waveform decreases during the drug self-administration session relative to the pre- and post-session non-drug recording periods. It is likely that these changes in waveform amplitude reflect an effect of drug on the recorded neuron.53,58 In such a case, if one discriminates between the large and small amplitude waveforms, the two subpopulations of waveforms differentially sample discharges during non-drug and drug periods. The cluster of larger waveforms tends to include discharges that occur during the non-drug period but

FIGURE 7.5. Effect of discriminating between discharges of a single neuron on phasic firing patterns. The data in this figure demonstrate the potential effect of subdividing the discharges of a single neuron into two waveform subpopulations (i.e., subclusters). In particular, this figure demonstrates the effect of the subdivision on phasic firing patterns. In this example, a cluster corresponding to a single neuron was subdivided into subclusters on the basis of waveform amplitude during the nondrug baseline period. The two subclusters consisted of the largest and smallest 30% of waveforms. The top row of the figure shows histograms that were constructed using the complete cluster; the middle and bottom rows show histograms that were constructed using the subclusters of the largest and smallest waveforms. The histograms on the left show ISI histograms. Overlays of the waveforms included in the cluster and subclusters are shown with the ISI histogram (c.f., Figures 7.3 and 7.4). After the clusters were defined, we constructed perievent histograms (right column of figure) displaying average firing rate (per 0.1-sec bin) during the seconds before and after all of the cocaine-reinforced lever presses in a single recording self-administration session. The histogram constructed with the complete cluster shows an overall long-duration increase in firing post-press relative to pre-press. Moreover, there is evidence of a peak increase in firing during the first several seconds after the press. The same histogram constructed using only the discharges associated with large waveform amplitudes shows the overall increase in firing but not the peak increase around the press. On the other hand, the histogram constructed using only the discharges associated with the smallest waveforms shows a sharp peak increase around the lever press and less evidence of the overall post-press increase in firing rate. These data are consistent with the conclusion that waveform amplitude of the neuron may be affected by the firing rate of the neuron. Specifically, the absence of the peak increase in firing when the histogram was constructed using only the large waveforms suggests that waveform amplitude of this neuron decreased as the firing rate of the neuron increased. The data also demonstrate that subdivision of discharges corresponding to a single neuron into two subpopulations can “create” firing patterns that may not accurately represent phasic firing patterns that are exhibited by individual accumbal neurons.

2345_frame_MASTER.book Page 174 Wednesday, October 23, 2002 1:03 PM

174 Methods in Drug Abuse Research: Cellular and Circuit Level Analyses

2345_frame_MASTER.book Page 175 Wednesday, October 23, 2002 1:03 PM

Chronic Extracellular Recording to Studies of Drug Self-Administration

175

not the drug period. Plots of firing based on this cluster can “create” either a firing pattern that is indicative of a drug-induced inhibition in firing when no such drug effect exists or a firing pattern that underestimates a real excitatory effect. The converse is true for the subcluster of small waveforms. (see Figures 7.6 and 7.7). The ISI analysis, as we currently use it, does not provide definitive evidence of the singularity or pluralism of neurons represented in a discriminated population of waveforms in all cases. In particular, the utility of the ISI in testing whether a cluster includes more than one neuron may be limited when neurons exhibit very low firing rates, and hence the probability of two neurons firing within short intervals of one another is low. However, the use of the ISI analysis in combination with the standard waveform analysis is a much stronger approach to accurate discrimination of singleneuron data than is the use of the waveform analysis by itself. The examples described above highlight the potential importance of the issue. It should be noted that the risk of making errors like those described above is much greater when an event timing system is used, and it is thus not possible to review the waveforms at different points in the experiment and to make post-hoc adjustments in the discrimination parameters.

7.3.4 QUANTITATIVE AND STATISTICAL ANALYSES OF FIRING PATTERNS 7.3.4.1 Individual Neuron Data As has always been the case, quantitative and statistical analysis of changes in the firing rate of a single neuron is wrought with difficulties. Although the development of the procedures for analyzing our electrophysiological data is still a work in progress, we have found that the use of a number of nonparametric statistical procedures has a number of advantages. These procedures are described below. Events that may be associated with changes in firing can occur either singularly or repeatedly within a self-administration session. In the case of an event that occurs repeatedly during the session, firing rate (number of discharges) is calculated during two intervals per occurrence of the event, one interval preceding the onset of the event and one interval following the onset of the event. Discharges during the preand post-event intervals are then compared using a Wilcoxon Matched Pairs Test. Analyses of neural responses to events that occur once per session are carried out in one of two ways. First, in some cases, the total number of discharges during an interval before and after the event onset are simply counted and used to calculate a percent change in firing rate. Second, discharges before and after the event are additionally compared in a Mann-Whitney test.78,80 The nonparametric statistical tests are designed for comparisons between populations of subjects. However, the tests can be appropriately applied to “single-subject” data that are not autocorrelated36,94 across the time periods being employed in the analysis. There is no physiological basis for expecting such an autocorrelation in the firing of accumbal neurons on the time bases of our comparisons (e.g., no evidence of cycles or rhythms on the time scale of our pre- and post-press comparisons). It is thus appropriate to apply the tests to our data. Experience in evaluating the utility of

FIGURE 7.6. Effect of discriminating between discharges of a single neuron on evaluations of drug effect on firing rate. The data in this figure demonstrate the potential effect of subdividing a waveform population corresponding to a single neuron into two subpopulations. In particular, these data exemplify the effect of subdividing single neuron data on the evaluation of the effect of cocaine administration on tonic firing rate. A complete cluster was subdivided into subclusters on the basis of waveform amplitude during the pre-drug baseline period (see Figure 7.7). The two subclusters consist of the largest and smallest 30% of waveforms. The top row of the figure shows histograms that were constructed using the complete cluster; the middle and bottom rows show histograms that were constructed using the subclusters of the largest and smallest waveforms. On the left of the figure are ISI histograms. Overlays of the waveforms included in the cluster and subclusters are shown with the ISI histogram (see Figures 7.3 and 7.4). On the right are histograms that display average firing rate (per 0.5-min bin) plotted as a function of time during the entire recording session. At the top of each of these histograms, the horizontal bars indicate the nondrug periods that preceded (left bar) and followed (right bar) the intravenous cocaine-self-administration session. The period between the bars corresponds to the drug-self-administration session. The histogram constructed with the complete cluster shows a robust increase in the tonic firing rate during the drug period relative to the presession non-drug period (top row). This increase is also apparent in the histogram that was constructed using the subcluster of the smallest waveform amplitudes (bottom row). On the other hand, the histogram made using the subcluster of the largest waveforms shows evidence of a possible inhibitory effect of drug on the firing rate of the neuron (middle row). These data are consistent with the conclusion that waveform amplitude of some accumbal neurons may be impacted by the drug (see Figure 7.7). The data also demonstrate that subdivision of discharges corresponding to a single neuron into two samples can “create” firing patterns that may not accurately represent the response of a recorded accumbal neuron to drug administration.

2345_frame_MASTER.book Page 176 Wednesday, October 23, 2002 1:03 PM

176 Methods in Drug Abuse Research: Cellular and Circuit Level Analyses

FIGURE 7.7. Change in waveform amplitude during the self-administration session. All the data shown in this figure correspond to the same neuron (i.e., neuron represented in Figure 7.6). Each of the six graphs shown in this figure is a scattergram that plots values for one waveform parameter, voltage, on the ascending limb relative to values for another waveform parameter, spike height (see Figure 7.3). The scattergrams in the left column of the figure show a single cluster of waveforms that correspond to a single neuron during the pre-session non-drug period (top row), the self-administration session (middle row), and the postsession non-drug-recovery period (bottom row) of a single recording session. The scattergrams in the right column show the same waveform population during the exact same periods of the self-administration session. However, the population of waveforms was subdivided into two subclusters on the basis of waveform amplitudes observed during the non-drug baseline-recording period. The two subclusters include the largest (right box inside the scattergram) and smallest (left box inside the scattergram) amplitude waveforms. Comparisons between the scattergrams showing firing during the different phases of the session show that there was a shift in the density of points included in the subclusters. Specifically, there was a decrease in large amplitude discharges and an increase in small amplitude discharges during the self-administration session relative to the two non-drug periods (compare the middle scattergram to the top and bottom scattergrams). These data show that the waveform amplitude may have been decreased by cocaine.

2345_frame_MASTER.book Page 177 Wednesday, October 23, 2002 1:03 PM

Chronic Extracellular Recording to Studies of Drug Self-Administration 177

2345_frame_MASTER.book Page 178 Wednesday, October 23, 2002 1:03 PM

178

Methods in Drug Abuse Research: Cellular and Circuit Level Analyses

various statistical approaches has shown us that the nonparametric tests, in conjunction with visual inspection methods,36 offer a number of advantages. The tests are sensitive to both the magnitude of the difference in firing rate between two comparison periods and the reliability of the difference. However, because the tests compare ranked data, they are not biased by the extreme low and high firing rates of an individual neuron. Given these two properties, the tests are insensitive to absolute changes in firing that are relatively large but do not occur reliably across repeated trials or bins. On the other hand, the tests are still sensitive enough to detect changes in firing rate that are relatively small in absolute magnitude but that occur reliably across repeated occurrences of an event. Changes in firing rate detected by the statistical tests are readily observed by independent observers on visual inspection of histograms and raster displays. Moreover, they are almost always associated with a percent change in firing rate that exceeds criteria typically employed in electrophysiological studies using a simple percent change criterion for defining neural responses. 7.3.4.2 Group Mean Neural Data It is of interest in some cases to characterize and compare the activity of groups of neurons. In such cases, it is necessary to control for interneuron differences in firing rate. Neurons that have higher tonic firing rates than other neurons in a group can carry a disproportionate weight in determining the average firing rate of the entire group and can therefore lead to inaccurate assessments of the central tendency of the group. We control for interneuron differences in firing rate by transforming the data of each neuron prior to calculating a group average. We have typically done this by dividing the firing rate of a given neuron at all relevant time points by the highest firing rate exhibited by that neuron during all of those time points. In this case, firing rate for each neuron varies between 0 and 1.0, and each neuron thus contributes proportionately to the average firing rate. The one difficulty that we have encountered with this method of normalization is that the transformed data do not fit a normal distribution and can thus pose a problem for parametric statistical analyses. In some cases, the data can be analyzed with a nonparametric statistic that does not assume normality. Alternatively, we have found that other transformations can be useful. For example, a log transformation minimizes disproportionate contributions of high-firing-rate neurons and often transforms the data to fit a normal distribution so that they can be subjected to parametric statistics. In some cases, it is necessary to add 1.0 to the measures of firing rate to eliminate zero values for which log values are undefined.

7.4 EXAMPLE INVESTIGATIONS The first applications of the chronic extracellular recording technique to the self-administration model were designed to characterize the activity of single nucleus accumbens neurons.13,16,17,59,60,67,92 This ongoing focus is based on evidence that the accumbens is importantly involved in mediating the rewarding properties of all addictive drugs and, moreover, may contribute to the etiology of addiction and the occurrence of relapse (for review see References 41 and 96). However, the technique is being employed

2345_frame_MASTER.book Page 179 Wednesday, October 23, 2002 1:03 PM

Chronic Extracellular Recording to Studies of Drug Self-Administration

179

successfully to record the activity of neurons in other brain structures and in animals self-administering a variety of drugs including heroin, alcohol, and nicotine.18,34 A number of the accumbal recording studies are reviewed in this section. The studies demonstrate the unique advantages of the technique. For example, behavioral data demonstrate that the recordings can be made during periods in which drug selfadministration occurs and, hence, when the neural and pharmacological conditions typically associated with drug seeking, drug reward, and, perhaps, drug addiction are present and subject to investigation (Section 7.4.1). The reviewed studies also demonstrate the utility of the recording technique for investigating the neural events that are predicted by the incentive motivation theories to mediate addictive effects of drugs. For example, the temporal properties of the technique have enabled researchers to identify the following: (1) changes in firing to discrete events that occur within a time frame of seconds and that are likely to reflect incentive-related information processing (Section 7.4.2), (2) changes in firing that occur on a time base of the minutes and hours of the drug self-administration session and that are likely to reflect acute pharmacological effects of the self-administered drug (Section 7.4.3), (3) a potential interaction between incentive-related and pharmacological firing patterns (Sections 7.4.4), and (4) changes in firing that occur across weeks of repeated drugself-administration and that could potentially correspond to drug sensitization (Section 7.4.5). The reviewed research additionally demonstrates the analytical and experimental procedures that can be used to discriminate between changes in neural activity that are due to actions of the drug and changes in firing that reflect nonpharmacological aspects of the execution or monitoring of behavior (Section 7.4.3).

7.4.1 BEHAVIOR The patterns of drug taking and associated changes in drug level during the recording studies are reliable and consistent with those described in many previous reports.67,70,97,101 In our cocaine studies, the pattern of drug intake includes an initial loading phase and a subsequent maintenance phase. During the loading phase (i.e., first 5 to 10 infusions), animals self-infuse cocaine rapidly. Thereafter, during the maintenance phase, animals level press at an intermediate and regular rate (e.g., once every 6 to 8 min). Consistent with this pattern of responding, drug level rises rapidly during the loading phase to a peak level. As response rate slows, drug level at the time of the press first decreases slightly from the peak that is attained during loading and then remains within stable narrow limits for the duration of the session. Drug level also shows regular oscillations between lever presses. Specifically, drug rapidly increases to a stable maximum shortly after each self-infusion and then slowly decreases to a stable minimum that is attained shortly before the next self-infusion (Figure 7.8).

7.4.2 INCENTIVE-RELATED INFORMATION ENCODING 7.4.2.1 Phasic Firing Patterns Time-Locked to Cocaine Self-Infusion During the simple fixed-ratio 1 (FR1) intravenous self-administration sessions, one of the events that is most likely to be associated with incentive-related neural

2345_frame_MASTER.book Page 180 Wednesday, October 23, 2002 1:03 PM

180

Methods in Drug Abuse Research: Cellular and Circuit Level Analyses

FIGURE 7.8 Pattern of intravenous self-administration and calculated drug level for a single subject. Interinfusion interval and calculated drug level at the time of each lever press (within 0.125 min before each press) are plotted as a function of the first 50 cocaine-reinforced lever presses. Insert shows the change in drug level during the interinfusion interval before and after a single self-infusion (dashed line in center of insert = lever press). (Drug level is calculated according to equations described by Pan et al., J. Neurochem., 46(4), 1299–1306, 1991.)

encoding is the instrumental response. Researchers have thus tested for phasic firing patterns in relation to this event. Numerous studies have shown that accumbal neurons do in fact exhibit phasic firing time-locked to the cocaine reinforced lever press.6,12,13,16,17,44,64,67,85 The patterns are of several types and vary in direction (i.e., increases or decrease in firing) and time course (Figure 7.9). We have classified these responses on the basis of time course as either exclusively prepress, predominantly prepress, symmetrical with respect to the press, predominantly post-press, or exclusively post-press. We refer to these firing patterns as lever-press firing patterns. The patterns that we have observed are for the most part comparable to those reported by others (for further discussion see Reference 85). 7.4.2.2 Information Encoded by the Lever-Press Firing Patterns Given that accumbal neurons are necessary for intravenous self-administration behavior, it is probable that the phasic firing time-locked to the cocaine self-infusion is associated with that behavior. However, there are numerous nonincentive-related events that occur within the second that lead up to and follow a lever press. To differentiate between firing patterns associated with the instrumental behavior vs.

2345_frame_MASTER.book Page 181 Wednesday, October 23, 2002 1:03 PM

Chronic Extracellular Recording to Studies of Drug Self-Administration

181

FIGURE 7.9 Lever-press firing patterns. For each of four neurons, average firing rate (Hz per 0.1-sec bin) is plotted during the 12 sec before and after all cocaine-reinforced lever presses during the maintenance phase of the selfadministration session. In each histogram, the time of the cocaine reinforced lever press is indicated on the abscissa at time zero. Top-down, the histograms show examples of the following lever-press firing patterns: (1) exclusively prepress, (2) predominantly prepress, (3) predominantly post-press, and (4) exclusively post-press. (Data from Uzwiak, A.J. et al., Brain Res., 767, 363, 1997.)

other events associated with the cocaine infusion we and others have conducted a test similar to the clamping procedure. In one study,64 we tested for responses that were associated with the instrumental behavior by varying the occurrence of the behavior as we held many other factors constant. At the beginning of the recording session subjects self-administered cocaine infusions until response rates stabilized. Thereafter, the session consisted of alternate phases in which infusions were either response-contingent or response-noncontingent. During each contingent phase, infusions (5 or 15) occurred only when the rat depressed the lever. The infusions were paired with a tone and light stimulus. In each noncontingent phase, infusions were activated by the computer, according to the schedule of infusions self-administered by the rat during the preceding contingent phase, and were paired with the same tone and light stimulus event that had been paired with contingent infusions. Lever presses during the noncontingent phase were nonreinforced and were not paired with the tone and light stimulus events.

2345_frame_MASTER.book Page 182 Wednesday, October 23, 2002 1:03 PM

Methods in Drug Abuse Research: Cellular and Circuit Level Analyses

Drug Level (µM/kg)

182

FIGURE 7.10 Pattern of lever presses and calculated drug levels during alternating phases of response-contingent and response-noncontingent infusions. At the top of the figure, horizontal bars demarcate phases of the experiment. Horizontal bars labeled with As and Bs correspond to periods of response-contingent and response-noncontingent delivery of cocaine infusions, respectively. The unlabeled horizontal bars at the far left shows the loading phase of the session. Below these horizontal bars is shown the pattern of lever presses (each vertical bar corresponds to a single press of the lever). Underneath the lever press display a calculated drug level at the time of each self-infusion is plotted as a function of successive presses. Data used to construct the figure were taken from Peoples et al.64 Drug level calculated for this figure using the equations described by Pan et al., J. Neurochem., 46(4), 1299–1306, 1991.)

The intrasession alternation between phases of contingent and noncontingent infusions systematically varied the amount of operant behavior that occurred before cocaine infusions while the recorded neuron, biological subject variables, and experiential subject variables were held constant. Because the schedules of contingent and noncontingent infusions were identical, drug level at the time of infusion was also held constant. Hence, the behavioral and motivational state of the animal would also be expected to have been stable; this is especially true given that the schedule of infusions was based on the animals’ pattern of self-administration behavior during the response-contingent phases. Consistent with these expectations, nonoperant behavior was comparable during the contingent and noncontingent phases. Moreover, during the noncontingent phase, animals occasionally made lever presses during the seconds before the scheduled noncontingent infusion and were therefore still motivated to self-infuse cocaine (Figure 7.10). Given the selectivity of the experimental manipulation that was made between the contingent and noncontingent phases, differences in firing patterns between the phases were likely to be specifically related to the presence vs. the absence of operant behavior. Of the 70 neurons recorded in the study, 29 showed a lever-press firing pattern during the contingent phases of the session. During the noncontingent infusion phase, all predominantly prepress patterns, almost all symmetrical patterns, and half of the predominantly post-press patterns were completely lost when the operant was absent (Figure 7.11, columns A and B). For the remaining neurons, the prepress firing was

2345_frame_MASTER.book Page 183 Wednesday, October 23, 2002 1:03 PM

Chronic Extracellular Recording to Studies of Drug Self-Administration

(1)

(2)

183

(3)

FIGURE 7.11 Firing patterns exhibited by neurons time-locked to response-contingent and response-noncontingent cocaine infusions. The three columns show lever-press firing patterns exhibited by three neurons. The lever-press patterns, from left to right, are of the following types: symmetrical, exclusively post-press, and predominantly prepress. In each column, topdown, firing rate is plotted during the 12 sec before and after the following events (time zero): (1) response-contingent infusions (a,d,f), (2) response-noncontingent infusions (b,e,g), and (3) nonreinforced lever presses (c and h). (Data from Peoples, L.L. et al., J. Neurochem., 72, 85, 1999b. With permission.)

lost, but the post-press firing either did not change or was only diminished (Figure 7.11, column C). These data show that all prepress firing and, in most cases, post-press firing that accompanied it were related to the occurrence of the operant. Consistent with that conclusion, all three types of firing patterns were present when the rat made nonreinforced presses during the noncontingent phases (i.e., when the press was present but the infusion and associated cues were absent). On the other hand, about half of the neurons that showed exclusively post-press firing during the contingent phase showed firing, albeit usually diminished, when the cues and drug were administered in the absence of the lever press operant. Moreover the same firing patterns were diminished when animals made nonreinforced presses. These data show that some exclusively post-press firing patterns were probably associated primarily with cues that signal the onset of reward rather than with the operant behavior. Additional evidence of the relationships between the lever press firing patterns and the instrumental behavior was reported by Carelli.11

2345_frame_MASTER.book Page 184 Wednesday, October 23, 2002 1:03 PM

184

Methods in Drug Abuse Research: Cellular and Circuit Level Analyses

There has been some characterization of the time course of the lever-press firing patterns and the time course of particular movements that lead up to and follow the lever press. These analyses show that the lever-press firing patterns are temporally dissociable from specific movements13,17 and thus do not encode the movements involved in the instrumental behavior per se. Instead, it is possible that the firing associated with the instrumental behavior encodes information related to reward expectation or some aspect of the influence of that expectation on the motivational (e.g., directional) properties of behavior. Such a relation between accumbal phasic firing and drug reward expectation would be consistent with the proposed contribution of the accumbens to behavior directed toward natural rewards.6,26,43,76,78,96 Nevertheless, additional studies are required to further test and refine this interpretation.

7.4.3 ACUTE DRUG EFFECTS 7.4.3.1 Firing Patterns That Mirror Changes in Drug Level Accumbal neurons exhibit two additional categories of firing patterns, both of which occur on a much slower time base than do the lever-press firing patterns. First, almost all neurons show a stable change in average firing rate during the cocaine selfadministration session relative to nondrug periods that precede and follow the drug session60,92 (Figures 7.12 and 7.13). For approximately two thirds of the neurons this “session” change in firing rate is a decrease. Similar findings have been described by other investigators and for other drugs.15,34 Second, approximately 40% of all neurons show a change in firing that occurs during the minutes that elapse between successive self-infusions. For most neurons this change in firing rate follows the previously described decrease + progressive reversal pattern.60,67 Firing rate decreases during the first minute post-press and then progressively increases until the time of the next lever press when the neural cycle begins again (Figures 7.12 and 7.13). About half of the neurons that show a session decrease in firing rate also show the decrease + progressive reversal firing pattern. The tonic decrease in firing and the decrease + progressive reversal firing pattern mirror changes in drug and dopamine level during the self-administration session. It is thus possible that the changes in firing are pharmacological in nature.49,60–63,65,67 If this were true, one would expect that the firing patterns would change in a predictable dose-dependent manner with changes in drug level. Experimental observations are consistent with this prediction. 7.4.3.2 Dose-Dependent Changes in Firing Rate To evaluate the dose-dependent nature of the changes in tonic firing we compared firing rates across the initial self-infusions of the self-administration session. This analysis showed that most, but not all, of the neurons that show a decrease in tonic firing rate during the drug session (i.e., session decrease neurons) show a progressive decrease in firing rate in relation to the gradual increase in cumulative drug level that occurs across the initial successive self-infusions (Figure 7.13, bottom row). These data indicate that the majority of the decreases in tonic firing and the associated decrease + progressive reversal patterns are likely to be, at least in part, pharmacological in nature. In contrast,

2345_frame_MASTER.book Page 185 Wednesday, October 23, 2002 1:03 PM

Chronic Extracellular Recording to Studies of Drug Self-Administration

185

FIGURE 7.12 Changes in tonic firing rate and decrease + progressive reversal firing patterns. On the left, histograms show average firing rate (Hz per 0.5-min bin) plotted as a function of session time for each of three neurons. At the top of each of these histograms, the horizontal bars indicate the non-drug periods that preceded (left bar) and followed (right bar) the intravenous cocaine selfadministration session. The period between the bars corresponds to the drug-self-administration session. Comparison of overall tonic firing rate during the self-administration session relative to the nondrug periods show that neuron A exhibited an increase in firing and neurons B-C showed a decrease in firing. These same neurons exhibited decrease + progressive reversal firing patterns, which are shown in the perievent histograms in the right column of the figure. In the histograms displaying the decrease + progressive reversal firing pattern, average firing rate (Hz per 0.1-min bin) (ordinate) is plotted as a function of the 4-min pre- and post-press (abscissa).

the increases in tonic firing rate exhibited by some neurons during the drug-selfadministration session developed in a more abrupt and non-dose-dependent manner (not shown). This lack of dose dependency is consistent with the conclusion that the increases in tonic firing are not pharmacological in origin. 7.4.3.3 Changes in Firing Rate: Drug Effect vs. Behavioral Feedback Observation of animals’ behavior during the interinfusion intervals showed that the animals exhibited a pattern of change in the frequency with which they locomoted.

2345_frame_MASTER.book Page 186 Wednesday, October 23, 2002 1:03 PM

186

Methods in Drug Abuse Research: Cellular and Circuit Level Analyses

FIGURE 7.13 Changes in firing rate and drug level. In the top row, firing rate (Hz per 0.5min bin, left ordinate) and drug level (right ordinate) are plotted as a function of 1-min bins during a 30-min period of the maintenance phase of the self-administration session. The arrows at the top of this graph show the occurrence of individual cocaine-reinforced lever presses. The regular oscillations in drug level that occurred between successive self-infusions are shown just below the arrows. In the middle and bottom rows of the figure, drug level at the time of the lever press is plotted as a function of time during the session. In these same graphs, firing rate (Hz per 0.5-min bin, left ordinate) is shown immediately below the plot of drug level. Comparisons of drug level and firing rate in each of the three graphs show that changes in firing tended to mirror changes in the drug level.

This pattern of locomotion paralleled the decrease + progressive reversal firing pattern. Locomotion decreased during the minute after each self-infusion and then progressively increased during the remainder of the interinfusion interval. A priori it was possible that the average decrease + progressive reversal change in firing rate reflected changes in the average percent time that animals spent locomoting. To test this interpretation we determined whether neurons that show the decrease + progressive reversal firing pattern also showed phasic changes in firing time-locked to specific locomotion events. Moreover, we used the clamping-control procedure to test whether the decrease + progressive reversal firing pattern depended on the variations in locomotion over the course of the interinfusion interval. We carried out these tests with the aid of video analysis methods. During each recording session behavior was videotaped. Each video frame (30 frames/sec) was sequentially time-stamped by a computer coupled with a video frame counter. Frames were time-stamped according to the same computer clock that timestamped each neural discharge. After the recording sessions, in offline frame-byframe analysis, time stamps associated with the onsets and offsets of particular

2345_frame_MASTER.book Page 187 Wednesday, October 23, 2002 1:03 PM

Chronic Extracellular Recording to Studies of Drug Self-Administration

187

FIGURE 7.14 Analysis of the relationship between the decrease + progressive reversal firing pattern and locomotive behavior. Left: For eight animals, average percent time spent in stereotypy (top) and locomotion (bottom) is plotted during the 3 min before and after all cocaine-reinforced lever presses during the maintenance phase of a single self-administration session. Right: Average normalized firing rate of all neurons that showed a decrease + progressive reversal firing pattern (n = 18) is shown during the same time periods. In the top right histogram, average normalized firing rate was calculated using all discharges during the minute before and after the lever press. In the bottom right histogram, only discharges that occurred during periods of stereotypy were used to calculate firing rate (see Reference 61). A comparison of the plot of time spent in locomotion (bottom left) and the histogram of average neural firing during all behaviors (top right) shows a similar pattern of variation in these two variables during the minutes before and after the lever press. Comparison of the two neural histograms shows that the pattern of change in firing rate was not affected when all periods of locomotion were excluded from the calculation of firing rate (i.e., the bottom neural histogram looks comparable to the top histogram).

behaviors (33-msec temporal resolution) were compiled and input as nodes into the computer19,91 for subsequent histogram analysis of neural firing in relation to specific behaviors. The video analysis confirmed that animals exhibited a regular pattern of locomotion during the interinfusion interval (Figure 7.14). However, additional analyses showed that the decrease + progressive reversal firing pattern was dissociable from locomotion. First, when we used histogram analyses to directly compare the firing rate of individual neurons before and after the onset of individual locomotor events, only half of the neurons showed any change in firing in relation to those events. Changes in locomotion cannot contribute to the decrease + progressive reversal firing pattern if neurons do not show differential firing in response to locomotion. The absence of firing correlated to locomotion for half the neurons indicated that the decrease + progressive reversal firing pattern for those neurons were unrelated to that behavior. The clamping control provided additional evidence against the locomotor interpretation of the decrease + progressive reversal firing pattern. Specifically,

2345_frame_MASTER.book Page 188 Wednesday, October 23, 2002 1:03 PM

188

Methods in Drug Abuse Research: Cellular and Circuit Level Analyses

when histograms displaying firing during the minute before and after cocaine selfinfusion were reconstructed with all periods of locomotion excluded from the calculation of firing rate, none of the decrease + progressive reversal patterns were lost. Thus, none of the decrease + progressive reversal firing patterns could be attributed to variations in locomotor behavior. Given that the firing pattern could be observed when behavior was held constant (i.e., clamped) it is difficult to attribute the firing pattern to the execution or monitoring of any specific movements. It is thus plausible that the decrease + progressive reversal firing pattern is at least in part pharmacologically determined. That is, the firing pattern may reflect an inhibition in firing caused by the rapid increase in drug level after each reinforced press followed by a progressive recovery from that inhibition in association with the gradual metabolism of the drug during the remainder of the interinfusion interval. One question that remains to be clarified is the relationship between the firing pattern and the motivational processing that contributes to drug seeking. It is possible that the decrease + progressive reversal firing pattern mediates a drug-induced change in motivational state that contributes directly to the regulation of drug seeking.67 Alternatively, it is possible that the firing reflects motivational processing that is only secondarily related to drug actions on accumbal neurons (i.e., feedback about drug-induced motivational changes). For example, one could propose that the progressive increase in firing leading up to the press corresponds, in part, to a progressive increase in excitatory incentive-related inputs that anticipate the next drug reward and that the post-press decrease reflects a cessation of that excitatory firing in association with the execution of the operant and delivery of the reward. Our current hypothesis is that the decrease + progressive reversal firing pattern and the associated tonic decrease in firing are related at least in part to inhibitory effects of cocaine, but that the decrease + progressive reversal, at least for some neurons, may additionally reflect incentive-related firing that is only secondarily related to drug level.63,67 It is hoped that we will discriminate further between direct pharmacological and secondary motivational contributions to the firing patterns in future studies designed to dissociate incentive-related processing (e.g., reward expectation) and drug level. 7.4.3.4 Drug Effects and Accumbal Information Processing As described in Section 7.2.1.3, the incentive motivation theories propose that addictive drugs modulate incentive-related information processing in the accumbens. It is further hypothesized that these acute drug actions amplify accumbal mechanisms that contribute to stimulus–reward learning and the influences thereof. The capacity to investigate these predictions is one of the most important and unique advantages of the chronic extracellular recording method. Specifically, with the recording technique it is possible to test whether rapid phasic firing patterns that encode incentiverelated information are affected by drug actions that occur over a time course of minutes, hours, and days. To date there has been little work aimed at addressing this question. However, as a first approximation we have begun to evaluate the relationship between the

2345_frame_MASTER.book Page 189 Wednesday, October 23, 2002 1:03 PM

Chronic Extracellular Recording to Studies of Drug Self-Administration

189

lever press firing patterns, which are believed to encode incentive-related information, and the firing patterns that appear pharmacological in nature (i.e., the tonic decrease in firing and the decrease + progressive reversal firing pattern). These analyses show that the majority of neurons that exhibit the rapid phasic firing patterns also exhibit the decrease + progressive reversal firing pattern. Moreover, these lever-press patterns are rarely exhibited by neurons that do not show the decrease + progressive reversal.63 Given the current understanding of the nature of the lever press and decrease + progressive reversal firing patterns the relationship between the firing patterns is potentially consistent with the conclusion that neurons that encode incentive-related information during the drug-self-administration session are modulated by drug. The tendency for the lever-press patterns to be selectively expressed by neurons that exhibit the decrease + progressive reversal pattern additionally suggests that the two firing patterns are related mechanistically. These recording data provide novel information about the neurophysiological mechanisms that could mediate interactions between addictive drugs and incentive-related processing in the accumbens. Additional work in our laboratory is testing the hypothesis that the interaction between the lever-press and decrease + progressive reversal firing patterns reflects a net amplification of incentive-related signals.60a 7.4.3.5 Drug Effect That Mediates Accumbal Role in Cocaine Reward Cocaine reward has incentive, satiating, and reinforcing effects. Administration of low and intermediate doses of cocaine elicits and enhances drug seeking in animals with a history of drug self-administration (incentive effects). High doses of drug tend to delay drug seeking during drug-self-administration sessions. Control studies have shown that the delay is related to drug satiety (for review see References 84, 95, and 96). Within limits, low, intermediate, and high doses of drug have reinforcing effects, although high doses of drug generally exert greater reinforcing efficacy than do lower doses. Dopamine-mediated actions of drug in the accumbens contributes to all of these effects of cocaine. The neurophysiological changes that might transduce these dopamine-mediated drug actions are not fully understood. Thus far, electrophysiological recordings suggest that the primary pharmacological effect of cocaine and other addictive drugs on accumbal firing is inhibition. Recovery from inhibition to a minimum firing rate is predictive of the timing of successive self-infusions during an ongoing drug self-administration session.50,60,67 It is thus possible that the inhibitory effects of self-administered cocaine contribute to the satiating, and perhaps reinforcing, effects of drug.49,67 It is difficult to understand how the same inhibitory effects might also contribute to the incentive effects of addictive drugs. However, evidence of the relationship between the decrease + progressive reversal firing pattern and the lever-press firing patterns suggests that the inhibitory effects of cocaine may interact with incentive-related processes in the accumbens.62,63 Moreover, we have hypothesized that drug has smaller inhibitory effects on the drug incentive-related signals than on other, background, neural activity. Such differential changes in firing would increase the signal-to-background

2345_frame_MASTER.book Page 190 Wednesday, October 23, 2002 1:03 PM

190

Methods in Drug Abuse Research: Cellular and Circuit Level Analyses

ratio of the drug-related signals and potentially translate into an increased impact of drug cues on instrumental behavior. Recent findings in our laboratory are consistent with this proposal (Peoples and Cavanaugh, 2002). Additional research is required to differentiate the extent to which the inhibitory effects of drug on accumbal activity contribute to the incentive, satiating, and reinforcing effects of cocaine. It may be possible to dissociate these drug effects by manipulating the behavioral paradigm and to then evaluate the relationship between the dose–response curves for the various effects of drug on behavior and the inhibitory effect of drug on accumbal firing rates.

7.4.4 EFFECTS

OF

REPEATED SELF-ADMINISTRATION SESSIONS

Numerous theories of addiction propose that long-lasting drug-induced changes in brain contribute to the development of drug addiction.38,39,73,83 For example, it is proposed that long-lasting changes in the meso-accumbal DA pathway that mediate drug sensitization contribute to the development of compulsive drug seeking.73,83 Consistent with this hypothesis, repeated exposure to cocaine and other addictive drugs enhances accumbal neurochemical responses to drug (for review see Reference 86). There is also evidence of long-lasting changes in signal transduction in accumbal neurons.31,48 Finally, acute electrophysiological recording studies show that repeated drug exposure can lead to lasting, although not necessarily permanent, changes in the firing patterns of accumbal neurons.103 If these presynaptic and post-synaptic changes within the accumbens contribute to an increased tendency to seek and take drug, they would be expected to be measurable in awake animals and to develop in conjunction with increases in drug seeking and drug taking across repeated drugself-administration sessions. To evaluate this and other hypotheses, researchers have begun to use the chronic extracellular recording technique to characterize the firing patterns of single accumbal neurons during repeated drug-self-administration sessions17,66 (see also Janak, Chap. 6, this volume). The most elegant method of evaluating the effects of repeated drug exposure on single-neuron activity is to characterize the firing of the same single neurons across the repeated drug sessions. There are a number of observations that suggest that this is feasible, albeit difficult. Characterization of recordings made from day to day are consistent with the conclusion that, in many cases, the chronically implanted microwires allow one to record the activity of the same neurons over a period of weeks. Although there are changes from one day to the next, typically decreases, in waveform amplitude across days, wires that show a neural signal on day 1 of recording tend to do so for many days. Conversely, and perhaps even more importantly, wires on which neural signals are absent on early recording days rarely show neural activity during later sessions. These observations suggest that although wires can show small shifts in brain that are sufficient to lead to changes in the amplitude of the signal associated with activity of a particular neuron, the wires are rarely displaced sufficiently to move away from one neuron and close enough to another so as to record the activity of different neurons across repeated recording sessions. Thus, it is possible to evaluate the effects of repeated drug-self-administration sessions on the firing patterns of single accumbal neurons.

2345_frame_MASTER.book Page 191 Wednesday, October 23, 2002 1:03 PM

Chronic Extracellular Recording to Studies of Drug Self-Administration

191

We have attempted to make such comparisons in one study. Specifically, we compared the firing of accumbal neurons recorded on day 2 and day 15 of cocaine self-administration. The comparisons of neural activity were designed to be as conservative as possible in identifying between-day differences in neural activity as corresponding to changes in the activity of the same single neurons. Comparisons of firing were limited to neural recordings that met a number of criteria. First, the population of waveforms recorded by a single microwire had to show evidence of a minimum ISI consistent with the refractory period of a single neuron during both sessions. Second, the waveforms recorded by the same microwire had to be comparable between sessions (stability criterion). Specifically, for both days it was necessary for neural waveforms to have been defined by the same eight discrimination parameters and by a comparable range of variation in each of those eight parameters. Comparable waveforms recorded with the same microwires were likely to correspond to the same neuron. To be included in the between-session comparisons, neural recordings had to be consistent with an additional criterion. Specifically, the neural waveforms had to be sufficient in amplitude on each day for the smallest discriminated waveforms to exceed the amplitude of the typical maximal fluctuations in the noiseband (completeness criterion). This requirement for a minimum waveform amplitude allowed us to verify that our ability to detect discharges did not change between day 2 and day 15. Given the foregoing, we interpreted a betweensession difference in neural activity recorded by a given microwire to be a betweensession difference in the activity of a single neuron. On day 2 of cocaine self-administration, 83 microwires recorded activity of single neurons. Of the 83 wires, 53 recorded neural activity on both day 2 and day 15. Of those 53 wires, 13 (24%) yielded neural records that met the criteria of stability and completeness (e.g., Figure 7.15). Comparisons of these 13 neurons on the two days showed the following: 1) the number of neurons that showed a significant decrease in firing during the self-administration session was greater on day 15 than on day 2, 2) for most neurons the magnitude of inhibition during the selfadministration session was significantly greater on day 15 than on day 2, and 3) average firing rate during the self-administration session, as well as during the predrug baseline recording period that preceded each self-administration session, was also lower on day 15 than on day 2. The evidence of increased inhibition during the drug-self-administration session on day 15, relative to day 2, is consistent with changes in accumbal firing measured in acute recording studies and could reflect a sensitized response of accumbal neurons to the inhibitory effects of cocaine. The between-session decrease in basal firing rates during nondrug periods could potentially reflect a long-lasting change caused by repeated drug exposure. There are, however, additional interpretations of the data that remain to be evaluated. Between-day comparisons of self-administration behavior showed that the rate and amount of drug intake was greater on day 15 than it was on day 2. This increase in drug intake across repeated self-administration is consistent with that observed in other studies and is thought to potentially model the increase in drug intake associated with the development of addiction.1 The concurrent observation of a between-session increase in drug intake with a between-session increase in “druginduced” inhibition of accumbal firing and depressed basal firing rates suggests that

2345_frame_MASTER.book Page 192 Wednesday, October 23, 2002 1:03 PM

192

Methods in Drug Abuse Research: Cellular and Circuit Level Analyses

FIGURE 7.15 Individual neuron recorded on day 2 and day 15 of cocaine self-administration. Data are shown for day 2 (top row) and day 15 (bottom row) of cocaine self-administration. The left column shows scattergrams and waveform overlays corresponding to an individual neuron on day 2 and day 15. The waveform overlays (positive voltage = up) for each day consist of the first 500 occurrences of the discriminated waveform. Vertical calibration bar = 1000 mV (amplified signal, see Figure 7.4). In the scattergrams, voltage on the ascending limb of the waveform (see vertical line through waveform overlays) is plotted as a function of peak amplitude (i.e., voltage, c.f., Figure 7.3). The right column shows ISI histograms display all ISIs £ 25 msec. Number of intervals (counts, y-axis) is plotted as a function of 0.1-msec increments in interval length.

the observed changes in firing could potentially contribute to the increase in drug intake and is thus of great interest. However, the change in drug intake also confounds the analysis of possible drug-induced neuroplasticity with a change in drug dose during the early and later sessions. For example, it is possible that the greater inhibition of firing during the self-administration session on day 15 relative to day 2 reflects the effect of a higher dose of drug rather than a sensitized response of accumbal neurons to the inhibitory effects of a given dose of drug. This study points out the need to incorporate certain controls in future studies. For example, it will be important for future studies to control the rate of drug intake during the recording sessions. Another issue to be considered in the design of future studies is that learning could contribute to changes in neural firing across repeated drug-self-administration sessions.

7.4.5 HISTOLOGICAL ANALYSES The nucleus accumbens is a heterogenous structure. On the basis of differentiations in afferent and efferent connectivity, neurochemical levels, neurochemical responses

2345_frame_MASTER.book Page 193 Wednesday, October 23, 2002 1:03 PM

Chronic Extracellular Recording to Studies of Drug Self-Administration

193

FIGURE 7.16 Changes in tonic firing rate exhibited by an individual neuron on day 2 and day 15 of cocaine self-administration. At the top of each column is shown the pattern of drug intake of an individual subject on day 2 (left) and day 15 (right) of cocaine self-administration. In these graphs, calculated drug intake is plotted as a function of lever press for the entire self-administration session. Below the drug curves are histograms displaying average firing rate (Hz per 0.5 min) plotted as a function of time during the day 2 and day 15 selfadministration sessions. The horizontal bars at the top of each histogram demarcate the nondrug periods that preceded and followed each cocaine self-administration session.

to addictive drugs, and receptor densities (for review see References 30, 32, and 102), two major subdivisions or subterritories have been identified along the mediallateral axis (referred to as core and shell). In addition, the core and shell and transitions along the rostral–caudal axis of the accumbens contain subcompartments and zones. Consistent with what would be expected on the basis of the differential connectivity of the core and shell, lesion and microinjection studies indicate that these regions contribute differentially to behavioral processes including stimulus–reward learning and the acute rewarding effects of addictive drugs.3,14,26,37,55,102 One of the potential benefits of the recording technique is that it will enable researchers to refine the understanding of the differential contribution that the subregions of the accumbens and other structures make to drug taking. However, there are few subregional differences in firing patterns that have been observed thus far.63,85 This general lack of differentiation is, at least at first glance, surprising, especially given that evidence of functional differences between subterritories has been obtained using lesion and microinjection techniques that have a lower anatomical resolution than does the recording method. However, there are a number of possible explanations. In a significant portion of studies, there has been no real attempt to make anatomical distinctions. In fact, some researchers have more often than not included nonaccumbal neurons in their samples and characterizations of “accumbal” neural activity. This approach is based on evidence that the broader ventral striatal area of the brain contributes to motivated behavior. It is also possible that the paucity of anatomical distinctions is attributable to certain technical parameters of the previous experiments. The histological localization of multiple wire tips can be challenging. This is particularly the case when a bundle of wires is used rather than an array of well-spaced wires. In the case of the bundle, researchers typically generate one or

2345_frame_MASTER.book Page 194 Wednesday, October 23, 2002 1:03 PM

194

Methods in Drug Abuse Research: Cellular and Circuit Level Analyses

two lesion markers. These markers are used to identify the location of the corresponding one or two wire tips; the locations of all other wire tips are estimated on the basis of wire tracks only. This method can lead to misidentification of the location of particular wire tips and thus of the location of particular recorded neurons. The risk of this misidentification is greatly reduced when an array of geometrically organized and well-spaced wires are implanted and a lesion marker is generated for each recording wire.63,85 Perhaps one of the most important factors contributing to the lack of subregional differentiations is the behavioral procedures that have been thus far employed in most studies, including our own. Most studies have used a simple FR1 schedule of self-administration. This procedure does not provide temporal isolation between multiple motivational and behavioral processes as well as between nonpharmacological and pharmacological events. Thus, it is possible that neurons of diverse functional and anatomical types are not discriminable on the basis of firing patterns time-locked to the self-infusion behavior. There are at least two pieces of evidence supportive of this proposal. First, we have observed that some neurons exhibiting comparable lever-press firing patterns during the simple FR1 self-administration session responded differentially to an alternative behavioral test. A number of examples of this phenomenon were observed in the experiment in which we compared response-contingent and response-noncontingent cocaine infusions on the lever-press firing patterns. For example, when the drug and associated cues were administered response-noncontingently, some symmetrical and predominantly post-press firing patterns were completely abolished, whereas others showed a loss of prepress but not post-press firing. Additionally, some exclusively post-press firing patterns were abolished and others were not. The distinct responses of neurons that showed comparable lever-press firing patterns suggest that the original firing patterns did not encode the same type of information. It is thus possible that these neurons correspond to anatomically distinct groups that have thus far gone undifferentiated in our FR1 experiments. We have observed a second related example of the potential “insensitivity” of the simple behavioral procedure. Neurons in regions surrounding the accumbens, including those not expected to play an important role in the motivational aspects of drug self-administration behavior, exhibit patterns of phasic activity time-locked to self-infusion that are not unlike those exhibited by accumbal neurons (unpublished observation). Behavioral procedures that separate in time different events of appetitive behavioral sequences have allowed researchers to differentiate between neurons that contribute to various aspects of behavior directed toward natural rewards.78 It is likely that applying similar behavioral procedures to studies of neurons that contribute to drug-directed behavior would increase our ability to discriminate between neurons that encode different types of information and that may therefore be differentially located within the accumbens.

7.5 FUTURE DIRECTIONS The initial studies using the chronic recording procedure have shown that the research not only is feasible but is useful for delineating mechanisms that contribute

2345_frame_MASTER.book Page 195 Wednesday, October 23, 2002 1:03 PM

Chronic Extracellular Recording to Studies of Drug Self-Administration

195

to drug seeking and drug reward. However, if we are going to take full advantage of the chronic recording procedures it will be important to advance the methodology in a number of domains. First, it is important that we enhance the behavioral sophistication of the experiments so that specific behavioral and pharmacological processes can be better isolated than is possible with the simple FR1 self-administration procedure. This will allow us to more readily attribute observed changes in firing to particular nonpharmacological and pharmacological events and also to better understand the nature of interactions between addictive drugs and rewardrelated information processing. Addiction is thought to include multiple stages, including controlled drug use, drug abuse, compulsive and uncontrollable drug seeking, abstinence, and relapse. An effort to extend the recording technique to self-administration paradigms that model these different phases is ongoing in several laboratories and will help to delineate the mechanisms that contribute to the progression of the disease. In addressing pharmacological questions it will be useful for researchers to make better use of systematic dose-response-curve analyses. These analyses will not only facilitate the differentiation of pharmacological and nonpharmacological firing patterns but will help to better define the nature of pharmacological effects, on neural activity and to relate the changes in neural firing to particular effects of drug on motivation and behavior. These paradigmatic changes are technically feasible, albeit labor intensive, and will better position researchers to identify “the” firing patterns that are relevant to drug-seeking behavior, drug reward, and addiction. As discussed by Dr. Janak (Chap. 6, this volume), identification of neural patterns of activity that mediate these behavioral and pharmacological phenomena will also have to extend beyond the level of the single neuron. Specifically, there may be critical aspects of information processing and drug effects that are observable at the level of groups of neurons rather than at the level of the individual neuron. The instrumentation required to conduct recordings of many neurons within individual structures as well as in multiple interconnected brain regions is now available. However, there are many questions about the analysis of these data that remain to be addressed in the future. The identification of activity patterns of single neurons and groups of neurons that are related to drug seeking and drug reward will be an important contribution of the chronic recording studies. However, the power of the research will be greatly enhanced if we can develop the methodology to more readily investigate the mechanisms that mediate those neural firing patterns. To do so it will be necessary to integrate the recording technique with additional neurophysiological and neuropharmacological techniques such as electrical stimulation and microinjection procedures. Some of this integration can be accomplished with little additional technical instrumentation development (e.g., stimulation experiments); other combinations will be more challenging. Integration of the chronic recordings with any additional neuroscience technique will add to the complexity and difficulty of the experiments. However, it will also provide a unique opportunity to bridge studies of presynaptic and post-synaptic mechanisms and contribute to developing a complete picture of the mechanisms that transduce drug effects that mediate reward and addiction.

2345_frame_MASTER.book Page 196 Wednesday, October 23, 2002 1:03 PM

196

Methods in Drug Abuse Research: Cellular and Circuit Level Analyses

ACKNOWLEDGMENTS Much of the intravenous self-administration instrumentation and procedures that we use were transferred from the laboratories of Dr. James Smith (Wake Forest University School of Medicine) and Dr. Steven Dworkin (University of North Carolina at Wilmington). Modifications to the procedures were made to integrate the self-administration and electrophysiological recording techniques. Our initial efforts to develop a microwire array and procedures for implanting the arrays were aided by helpful conversations with Dr. Steven Sawyer in the laboratory of Dr. Donald Woodward. Dr. Martin Wolske contributed to the electronic and computer system that we used to conduct recording sessions. Mr. Patrick Grace, Ms. Linda King, Mr. Fred Gee, Dr. David Shapiro, and numerous students made important contributions to the development of the integrated technique described herein. The methods were developed and the experiments were conducted in the laboratory of Dr. Mark O. West (Rutgers, The State University of New Jersey) under the direction of Dr. Laura Peoples. Data analyses were also conducted at the University of Pennsylvania with the expert assistance of Mr. Jason Collison, Ms. Jamie Lesnock, and Mr. Daniel Cavanaugh. Dr. Kevin Lynch and Dr. Jon Baron (University of Pennsylvania) provided helpful conversations regarding statistical analyses. Calculations of drug level were carried out using programs provided by Dr. Joseph Justice (Emory University) and Dr. Mark S. Kleven (Center de Recherche Pierre Fabre). Photographs shown in Figures 7.1, 7.2, and 7.17 were taken by Mr. Steven J. DeSalvo. Mr. Patrick Grace and Dr. Mark West contributed to text included in the Appendix. Research was supported by DA06886 (PI, MOW), DA 535405 (PI, LLP), and the University of Pennsylvania (startup funds for LLP).

APPENDICES A.1 INSTRUMENTATION A.1.1 INTRAVENOUS CATHETER

AND

SWIVEL

The catheter is made of a narrow-diameter length of microbore tubing inserted into a wider-diameter length of tubing. Details regarding the catheter are provided by Dworkin and Stairs (Chap. 2, this volume). The swivel used during recording sessions is the Airflyte CAY-675–24 electronic and fluid swivel (Airflyte Electronics, Bayonne, NJ). This swivel can be purchased with either 27 or 32 electrical channels. It has low contact noise, low rotational torque, and O-ring seals that can withstand both alcohol sterilization and salt solutions. This swivel is additionally light and compact enough to be counterbalanced (see below). Adaptors and connectors are added to the Airflyte swivel in order to interface the swivel with the tether system. (see Figure 7.2 and below). A simple fluid swivel (not shown) similar to that described by Brown et al.9 is used when experiments are not in progress (see Figure 7.2).

2345_frame_MASTER.book Page 197 Wednesday, October 23, 2002 1:03 PM

Chronic Extracellular Recording to Studies of Drug Self-Administration

197

FIGURE 7.17 Counterbalanced Airflyte swivel with leash adapter. Near swivel base and proceeding from left to right are: acrylic swivel clamp made in journal and cap style (I), 1in. hinge with fixed pin (L), lever arm end block (M), lever arm (N), inertial dampening springs attached at tower and end block (O), adjustable height fulcrum tower (P), fulcrum pin (Q), and counterbalance weight (not shown) on lever arm that extends to right of fulcrum tower. Proceeding from top to bottom are: Tygon microbore tubing (F) connecting infusion pump (not shown) to input tube of fluid channel of Airflyte swivel, electrical leads of Airflyte swivel (K) leading up to preamplifier (not shown), chassis of Airflyte swivel (J), tail tube of Airflyte swivel (H), leash adapter,© Airflyte wire harness with connector block (G) on leash adapter,© catheter attached to fluid swivel output tube (F), electrical harness plugged into connector block (E), electrical harness shielded with wire braid (now replaced by the spring leash) (D), leash coupling (B), and upper leash (A).

A.1.2 MICROWIRE ARRAY HEADSET

AND

ELECTRONIC HARNESS

Descriptions of microwires, and various microwire arrays, along with photographs and diagrams have been provided by other authors.47,81,98 The array that we use is shown in Figure 7.18 and briefly described herein. Microwires are soldered into two Microtech (Boothwyn, PA) miniature connector strips (GF-8). The solder connections are coated with epoxy. The microwire array (Figure 7.18) is arranged as desired by use of a jig. Wires are held in the desired configuration with polyethylene glycol (PEG). In our case, the array is rectangular in shape and consists of two rows of 6 to 8 quad-teflon-coated stainless steel microwires (California Fine Wire, Grover City, CA). The diameter of each wire in the array is 50 mm with no insulation. The two rows of microwires are 1.80 mm long and separated from one another by 0.45 to 0.50 mm. Adjacent wires within each row are approximately 0.35 mm apart (wire center to wire center). This array configuration yields reliable success in recording the activity of single cells. The arrays can be difficult and time-consuming to construct. Most laboratories thus purchase the arrays from one of several sources. The present author worked extensively with

2345_frame_MASTER.book Page 198 Wednesday, October 23, 2002 1:03 PM

198

Methods in Drug Abuse Research: Cellular and Circuit Level Analyses

FIGURE 7.18 Microwire array. Left: saggital view of microwire array headset showing one row of six microwires (A), PEG-coated microwires (B), Microtech miniature connector strips (c), and epoxy-coated solder connections (D). Middle: coronal view of microwire array headset showing two rows of microwires (E) and microwire array epoxied to side of connector strip (F). Right: saggital view of microwire array, ground wire (G), and stimulating electrode (H) attached to a microwire array headset holder that is inserted into a stereotaxic adaptor (Figure 7.2) during surgery. Ground wire and stimulating electrode are each made of a twisted pair of teflon coated 0.05-mm wires soldered into a two-pin Microtech connector (GF-2). Array holder is a modified NB Labs protection cap (CD101).

Dr. David Shapiro in developing and testing the arrays that we currently use. Dr. Shapiro now sells the arrays (Microwire Technologies, Inc., East Windor, NJ) and is particularly adept not only at the construction of the arrays but in customizing them. Another source is NB Labs (Denison, TX). The harness consists of thin wires that connect the headstage of the animal to the Airflyte swivel. At the head of the animal (i.e., at the headstage of the harness), the harness wires are soldered to FETs that are in turn soldered either via additional wire or directly to the pins of male Microtech connector strips. The FETs and connector strips are encased in epoxy. The wires of the harness are passed through a protective spring leash (PS95, Instech Laboratories Inc., Plymouth Meeting, PA). The wires at the swivel end of the harness are then soldered to strip connectors that plug into corresponding connectors mounted on the Airflyte swivel. The spring leash is connected via dental cement to the headstage and swivel ends of the harness. Small hooks made from molded dental cement are cemented on either side of the rat-end of the harness (Figure 7.19). When the harness is plugged into the electrode cap on the animal, small rubber bands are placed into the hooks on the harness and then pulled across hooks mounted on the electrode cap of the animal. The use of the hook-and-rubber-band connection system prevents the harness from working free of the headstage over the course of the experiment

A.1.3 TETHERING SYSTEM The tethering system protects the catheter and electronic harness, transfers rotational force to the swivel, and provides strain relief for both the harness and the catheter. The tethering system consists of a J-shaped catheter cannula, a leash with coupling, and a swivel with leash adaptor. The tethering system is suspended by a counter-

2345_frame_MASTER.book Page 199 Wednesday, October 23, 2002 1:03 PM

Chronic Extracellular Recording to Studies of Drug Self-Administration

199

FIGURE 7.19 Electrical harness. (A) Protective spring leash encasing wires that isolate signals from each implanted wire, (B) animal or “headstage” end of the harness, (C) Cannon connector at “swivel” end of harness, (D) dental acrylic hooks embedded in the headstage of the harness, and (E) orthodontic rubber band used to connect (D) to (F), which are dental acrylic hooks embedded in the electrode cap at the time of surgery.

balance. The counterbalance consists of an acrylic swivel clamp, hinge with fixed pin, lever arm end block, lever arm, variable-height fulcrum tower, counterbalance weight, and inertial dampening springs. The J-shaped cannula (Figure 7.20) is constructed in the following manner. At one end in a long length of 13-gauge thin-wall stainless steel hypodermic tubing

FIGURE 7.20 Spring leash and J-shaped cannula. Left: J-shaped catheter cannula (A) and spring holder with set screw (B). Right: the leash, made of spring (H) and hypodermic tubing (I) is shown attached to the J-shaped cannula (c), which, in turn, is attached to stereotaxic adaptor (D) by way of cannula holder (E). Rotating and angling of the adaptor’s ball joint (G) allowed optimal positioning of cannula at the back of the head. Stereotaxic adaptor (shown attached to stereotaxic arm, [F]) consists of components taken from a Radio Shack Project Holder (cat. no. 64–2093).

2345_frame_MASTER.book Page 200 Wednesday, October 23, 2002 1:03 PM

200

Methods in Drug Abuse Research: Cellular and Circuit Level Analyses

(Small Parts, Miami Lakes, FL; cat. no. K-htx-13TW), a slight “j” bend is made. This cannula is then cut to the proper length (18 mm) and at the open ends the inside walls are beveled with a countersink drill bit. The leash (Figures 7.1, 7.2, and 7.20) is made in four parts from three materials: spring, tubing, and brass stock The leash length is a critical variable to be attended to closely. It influences the freedom of movement of the animal and the proper working of the tether system. The leash adaptor (5.50 cm in length, Figure 7.2) on the Airflyte swivel and the system used to counterbalance the swivel are not commercially available. It is fabricated from 2.54 cm round brass stock. Using a threaded die, the final 7 mm of one end (rat end) of the adaptor is threaded (outside) to mate with the inside threads of the leash coupling (9.52 mm ¥ 9.45 threads/cm). At the opposite end (swivel end) a round hole is drilled into the axis to a diameter (approx. 5 mm) and depth (approx. 2 cm) that allows a press-fit onto the stainless steel tail tube of the Airflyte swivel. At the upper end two longitudinal access slots (2 cm long ¥ 3 mm wide) are cut on each side to join the longitudinal hole and to allow exit of both the fluid channel output tube and the electrical harness of the Airflyte swivel. A set screw opposite the fluid swivel pipe access slot allows attachment of the adaptor to the swivel tail tube. The adaptor is attached to the swivel by Airflyte Electronics. The Airflyte counterbalance is shown in Figure 7.17. It consists of a swivel holder or clamp that is connected via a hinge to a lever arm. The swivel clamp is made from a 4-cm cubical piece of 6.25-mm-thick acrylic. A hole the same diameter as the swivel base (approx. 1.80 cm) is drilled through the center of the cube. Two holes are drilled into one edge of the acrylic, centered 5 mm from each of the adjacent edges. These holes are tapped for inserting 3.175-cm ¥ 3.5-mm screws used to tighten the clamp around the swivel. The hinge is attached to the acrylic swivel clamp on one end and to the lever arm end block on the other. Before the hinge is attached, however, the end block (aluminum stock) is attached to the lever arm with a screw (18.75 mm ¥ 3.5 mm) that passes through a countersink hole in the block and into a hole tapped into the axis of the lever arm. A set screw in the end block is tightened against the lever arm to prohibit rotation of the block. The lever arm is a 30-cm length of 6.35-mm stainless steel rod with one end threaded inside to accept the end block attachment screw and one transverse hole in the center of its length for the fulcrum pin (3.175 cm ¥ 3.5 mm screw) that is passed through the fulcrum tower. A 200-g brass counterweight is attached to the far end of the lever arm with a thumbscrew. Two lengths of spring curtain rod are attached with the end block to the fulcrum tower. As the subject moves up and down quickly the counterbalanced tethering system will react and then rebound if left unchecked. The spring dampeners oppose the inertia of the system and a smooth “floating” motion is achieved.

A.1.4 OPERANT CHAMBERS Our operant chambers (Figures 7.1 and 7.2) have traditionally been made in-house, but chambers are commercially available (MED Associates Inc., St. Albans, VT). The chamber itself is made largely of Plexiglas. The inner walls and floor of the chamber are free of any metal or protrusions, except for a water bottle sipper tube.

2345_frame_MASTER.book Page 201 Wednesday, October 23, 2002 1:03 PM

Chronic Extracellular Recording to Studies of Drug Self-Administration

201

All devices (lights and speakers) are mounted on the outside of the chamber. Response levers are also made of Plexiglas and are manually inserted and secured before the start of each self-administration session and removed at the end of the session. The floors are removable and made of Plexiglas square rods. The operant chambers are housed in sound-attenuating outer chambers. In designing, purchasing and ultimately placing operant chambers and devices such as lights, pumps, and levers it is important to keep AC-operated devices and motors that can generate electrical noise distal from the animal.

A.1.5 ELECTROPHYSIOLOGICAL EQUIPMENT Instrumentation required to conduct extracellular recordings has been described by Moxon47 and Sameshima and Baccala.77 Our particular equipment and procedures are outlined here. The signals on individual wires are very small and ideally would be amplified at the level of the headstage to avoid degradation. Until recently this technology has not been available (see Neuralynx Inc. and Plexon Inc.). As a result, in all of our experiments thus far we have used the following procedures. The signal on each implanted wire is passed through a field effect transistor (FET), which boosts current without affecting voltage and allows the signal to be transmitted with minimal loss in strength. From the FET, the signal passes up the harness and through the swivel to a preamplifier that differentially amplifies the signal on each recording electrode relative to a local differential electrode. This electrode is usual one of the microwires within the bundle or array that is functional but not located near a recordable neuron. The differential electrode is proximal in space to the recording wires and identical in properties so that it transmits the same electrical noise, including locally generated changes in potential, as do the recording wires. The signal on the differential electrode is subtracted from the signal on each recording electrode at the preamplifier stage, thereby eliminating any common source of noise and providing a better conditioned and more readily discriminable neuron signal on the recording wires. The differential potential is amplified and then transmitted to an amplifier and filter. This equipment selectively amplifies signals (waveforms) that are within a range of frequencies consistent with those potentially generated by an action potential. Our filter is a bandpass filter (450 Hz to 10 kHz). The amplified signal is then input into an analog/digital circuit. At this stage the analog signal is digitized and the time stamp marking the time of its occurrence is stored by the computer storage. We digitize waveforms at a sampling rate of 50 kHz and timestamp the occurrence of signals with a temporal resolution of 0.1 msec. DataWave hardware and software, along with user-written C programs (clock sequences) compiled with the purchased software, control and record experimental events in the operant chamber at the same time that the system records neural data. The system controls and records experimental events via a digital input/output card that, in turn, interfaces with various relays and circuits connected to devices in the operant chamber. In such a system, the same computer clock is used to time-stamp the occurrence of all neural events as well as all stimulus and behavior events of interest to the experimenter. The temporal pattern of firing can then be readily related to that of stimulus and behavior events of interest. Other data-acquisition systems

2345_frame_MASTER.book Page 202 Wednesday, October 23, 2002 1:03 PM

202

Methods in Drug Abuse Research: Cellular and Circuit Level Analyses

are not self-contained in this way but require that the neural data-acquisition system be integrated with a separate behavioral control system such as that of MED Associates Inc. Although the necessity of integrating the two systems may seem disadvantageous, it has been done successfully by a number of laboratories and is not particularly difficult (c.f., Janak, 2002). Thus, it is well worth considering. We are in fact transitioning to this type of integrated system.

A.2 SURGICAL PROCEDURES We have typically carried out the intravenous catheterization and microwire implant during the same surgery and prior to any behavioral training. This has worked well given the simple behavioral procedures and short acquisition periods required to establish stable self-administration behavior. However, if one wishes to train animals in a more sophisticated behavioral paradigm that requires a relatively long training period and one is not interested in studying acquisition processes, it would probably be better if the catheterization and behavioral training were completed prior to the stereotaxic implant. Although the catheter and electrode cap remain intact for months, the number of wires from which neural recordings can be made tends to decrease over the same time period. Thus, delaying the wire implant will potentially increase the number of neurons that can be successfully recorded in relation to the behavior of interest. Several procedures can be used to carry out the jugular catheterization. The surgical procedures that we use are described by Dworkin and Stairs (Chap. 2, this volume) and will be discussed only briefly here. The reader may also wish to consult VanDongen et al.87 and Caine et al.10 The procedures used to stereotaxically implant the arrays are described in another chapter in this book (Janak, Chap. 6, this volume). Other helpful descriptions of procedures for stereotaxic surgery and array implants are provided by Cooley and Vanderwolf,21 Lemon (1983), and Nicolelis et al.51 It should be noted that we have thus far implanted only a single array unilaterally in the brain. Other researchers regularly implant multiple arrays in the hopes of increasing the number of neurons that can be characterized in individual animals and experiments. This is a strategy that offers several advantages (see Janak, Chap. 6, this volume) and is one that we plan to employ in the future. Although we do not have experience with this approach, it seem likely that there will be a limit to the number of wires that can be implanted before extensive mechanical damage is imposed on the brain and a primary benefit of the technique is compromised. Aseptic procedures are used throughout the catheterization and stereotaxic implant. All surgical tools and supplies are autoclaved prior to surgery. The surgeon wears a clean lab coat, face mask, and surgical gloves. Surgery is carried out in a dedicated and aseptic surgical suite. Before the start of surgery rats are anesthetized with an injection of sodium pentobarbital (50 mg/kg, i.p.). Subjects are also injected with atropine methyl nitrate (10 mg/kg, i.p.) and with penicillin G (Bicillin, Wyeth Laboratories, Philadelphia, PA, 75,000 units, into hind leg muscle). Over the course of surgery anesthesia is maintained by alternate injections of sodium pentobarbital (10 mg/kg, I.P.) and ketamine hydrochloride (60 mg/kg, i.p.). When the catheterization is conducted in

2345_frame_MASTER.book Page 203 Wednesday, October 23, 2002 1:03 PM

Chronic Extracellular Recording to Studies of Drug Self-Administration

203

conjunction with the stereotaxic implant, the surgery can take 4 to 6 h. This period exceeds that which is appropriate for the commonly employed ketamine and xylazine mixture of anesthetic. The pentobarbital and ketamine mixture is a more appropriate injectable anesthetic combination for maintaining adequate anesthesia for such a protracted surgery; however, it can also lead to respiratory difficulties and death. Careful evaluation of the level of anesthesia and avoidance of overdose is critical. Although we do not have direct experience with gaseous anesthetics, it is possible that isoflurane anesthesia administered via a face mask may be the optimal anesthetic. The scalp and neck are shaved and washed with betadine. A 4-mm incision is made in the scalp. The incision is placed such that it is centered over the midline of the skull; the most posterior end of the incision meets with the back ridge of the occipital skull plate. A second incision is made through the skin of the neck diagonally from the mandible to a point midway between midline and shoulder. Procedures described by Dworkin and Stair (Chap. 2, this volume) are then used to implant the jugular catheter. The nonimplanted (swivel) end of the catheter is passed subcutaneously to the back of the animal and exited through the incision made at the back of the skull. The first 20 mm of exited catheter is coated with a thin layer of medical silicone (Silastic Brand Silicone Type A cat. no. 81, Dow Corning, Midland, MI), and then the J-shaped cannula is gently slid down over this same portion of the catheter (convex side of j-bend facing anterior). This step closes off the space between the catheter and the inner walls of the cannula through which contaminants can pass into the animal and eliminates the occurrence of infection. A 5-mm length of the leash is wound onto the cannula. Rats are then placed into the stereotaxic apparatus and prepared for implant of the microwire array. The electrodes are then implanted using procedures similar to those described by Janak (Chap. 6, this volume). Once the microwire array is cemented into place, the J-shaped cannula is mounted in the “j-holder” (Figure 7.20) and positioned so that the cannula and leash-covered catheter are vertical and centered along the medial-lateral axis of the cemented array. The curved end of the J-shaped cannula is placed so that it rests just above the muscle behind the bony ridge of the occipital skull plate. Care is taken to avoid crimping or stretching the catheter and to avoid pressing against the animal with the cannula. A bridge of cement is placed at skull level between the cannula and the cemented array. The cement is then built up to encase the cannula and to fill in any space between the cannula and the array (Figure 7.1). The cemented headstage is smoothed out with wet cement so that there are no rough edges.

A.3 POST-OPERATIVE CARE Animals are administered hourly infusions of heparinized saline and have free access to water. Food is restricted so as to maintain animals at a body weight of 350 g. Animals are weighed regularly, and food supplied to the animal is adjusted accordingly. Catheter patency is verified occasionally with an infusion of methohexital. Animals are placed in operant chambers at least 3 days prior to the onset of self-administration training. Animals remain in the chambers for the duration of the study. The chamber is cleaned regularly. The floor is removable and can be

2345_frame_MASTER.book Page 204 Wednesday, October 23, 2002 1:03 PM

204

Methods in Drug Abuse Research: Cellular and Circuit Level Analyses

FIGURE 7.21 Histology. Shown is a coronal view of a rat brain with two lesion markers corresponding to the tips of microwires implanted in the accumbens. The lateral lesion is just medial to the anterior commissure and located in the core of the accumbens. The more medial lesion marker is ventral to the lateral ventricle and is along, or near, the border between the core and shell of the accumbens. Lesion markers typically emerge, expand, and then disappear across four to five coronal sections. The medial lesion marker shows the typical maximal circumference at the lesion center.

easily exchanged with a clean replacement as needed. The animal and tether system are inspected twice daily to insure that the following are true: (1) the animal is free of infection and in good health and (2) the tether system is fully functional and free of any leaks. Aseptic procedures are maintained as much as possible in the regular handling and inspection of the animal and tether system. Animals remain active and healthy for the duration of our studies, which have typically lasted 3 to 4 months.

A.4 HISTOLOGY At the end of the experiment histological procedures are used to confirm the location of the tip of the recording electrodes. Subjects are injected with a lethal dose of sodium pentobarbital. Anodal current (50 mA for 4 sec) is passed through each microwire. Animals are perfused with formalin-saline. Coronal sections (50 mm) are mounted on slides and incubated in a solution of 5% potassium ferricyanide and 10% HCl to stain the iron deposits left by the recording tip.29 The tissue is counterstained with 0.2% solution of Neutral Red. Examples of the stained lesion markers are shown in Figure 7.20. The location of each wire tip is plotted on the coronal plate56 that most closely corresponds to its anterior–posterior position. The microwire array is rectangular in configuration and made of two parallel rows of wires. The configuration and geometric relationship of the various wires to each other is usually

2345_frame_MASTER.book Page 205 Wednesday, October 23, 2002 1:03 PM

Chronic Extracellular Recording to Studies of Drug Self-Administration

205

maintained as the wires are lowered into brain so that it is possible to identify the location of each of the 12 to 16 wire tips during histology. Two or more individuals carry out histological analysis of wire locations, for there is a certain level of subjectivity in these evaluations. We have employed a number of strategies in our histological analyses and find that the greatest inter-rater reliability is obtained when we follow tracks left by the electrode from the most dorsal entry point in the brain to the blue-stained lesion corresponding to the wire tip (a procedure originally suggested by Hendrik J. Groenewegen, Vrije Universiteit, Amsterdam).

REFERENCES 1. Ahmed, S.H. and Koob, G.F., Transition from moderate to excessive drug intake: change in hedonic set point, Science, 282(5387), 298, 1998. 2. Ahmed, S.H., Walker, J. R., and Koob, G., Persistent increase in the motivation to take heroin in rats with a history of drug escalation, Neuropsychopharmacology, 22, 413, 2000. 3. Alderson, H.L., Parkinson, J. A., Robbins, T.W., and Everitt, B.J., The effects of excitotoxic lesions of the nucleus accumbens core or shell regions on intravenous heroin self-administration in rats, Psychopharmacology, 153, 455, 2001. 4. Arroyo, M., Markou, A., Robbins, T.W., and Everitt, B.J., Acquisition, maintenance and reinstatement of intravenous cocaine self-administration under a second-order schedule of reinforcement in rats: effects of conditioned cues and continuous access to cocaine, Psychopharmacology, 140, 331, 1998. 5. Benet, L.Z., Droetz, D.L., and Sheiner, L.B., Pharmacokinetics: the dynamics of drug absorption, distribution and elimination, in Goodman and Gilman’s The Pharmacological Basis of Therapeutics, Hardman, J.G., Limbird, L., Molinoff, P.B., Ruddon, R.W., and Gilman, A.G., Eds., McGraw-Hill, New York, NY, 1996. 6. Bowman, E.M., Aigner, T.G., and Richmond, B.J., Neural signals in the monkey ventral striatum related to motivation for juice and cocaine rewards, J. Neurophysiol., 75, 1061, 1996. 7. Bozarth, M.A., Methods of Assessing the Reinforcing Properties of Abused Drugs, Springer-Verlag, New York, NY, 1987. 8. Breiter, H.C. et al., Acute effects of cocaine on human brain activity and emotion, Neuron, 19, 591–611, 1997. 9. Brown, Z.W., Amit, Z., and Weeks, J.R., Simple flow thru swivel for infusions into unrestrained animals, Pharmacol. Biochem. Behav., 5, 363, 1976. 10. Caine, S.B., Lintz, R., and Koob, G.F., Intravenous drug self-administration techniques in animals, in Behavioral Neuroscience: A Practical Approach, Vol. II, Oxford University Press, New York, NY, 1993, 117. 11. Carelli, R.M., Activation of accumbens cell firing by stimuli associated with cocaine delivery during self-administration, Synapse, 35, 238, 2000. 12. Carelli, R.M. and Deadwyler, S.A., Dual factors controlling activity of nucleus accumbens cell-firing during cocaine self-administration, Synapse, 24, 308, 1996a. 13. Carelli, R.M., King, V.C., Hampson, R.E., and Deadwyler, S.A., Firing patterns of nucleus accumbens neurons during cocaine self-administration in rats, Brain Res., 626, 14, 1993. 14. Carlezon, A., Jr., Devine, D.P., and Wise, R.A., Habit-forming actions of nomifensine in nucleus accumbens, Psychopharmacology, 122, 194, 1995.

2345_frame_MASTER.book Page 206 Wednesday, October 23, 2002 1:03 PM

206

Methods in Drug Abuse Research: Cellular and Circuit Level Analyses

15. Chang, J.-Y., Janak, P.H., and Woodward, D.J., Comparison of mesocorticolimbic neuronal responses during cocaine and heroin self-administration in freely moving rats, J. Neurosci., 18(8), 3098, 1998. 16. Chang, J.-Y., Sawyer, S.F., Lee, R.S., Maddux, B.N., and Woodward, D.J., Activity of neurons in nucleus accumbens during cocaine self-administration in freely moving rats, Neurosci. Abstr., 16, 252, 1990. 17. Chang, J.-Y., Sawyer, S.F., Lee, R.S., and Woodward, D.J., Electrophysiological and pharmacological evidence for the role of the nucleus accumbens in cocaine selfadministration in freely moving rats, J. Neurosci., 14, 1224, 1994. 18. Chang, J.-Y., Zhang, L.L., Janak, P.H., and Woodward, D.J., Neuronal responses in prefrontal cortex and nucleus accumbens during heroin self-administration in freely moving rats, Brain Res., 754, 12, 1997. 19. Chapin, J.K., Loeb, G.E., and Woodward, D.J., A simple technique for determination of footfall patterns of animals during treadmill locomotion, J. Neurosci. Methods, 2, 97, 1980. 20. Childress, A.R., Mozley, P.D., McElgin, W., Fitzgerald, J., Reivich, M., and O’Brien, C.P., Limbic activation during cue-induced cocaine craving, Am. J. Psychiatry, 156(1), 11, 1999. 21. Cooley, R.K. and Vanderwolf, C.H., Stereotaxic Surgery in the Rat: A Photographic Series, AJ Kirby, New York, NY, 1990. 22. Deadwyler, S.A., Electrophysiological investigations of drug influences in the behaving animal, in Modern Methods in Pharmacology. Vol. 3: Electrophysiological Techniques in Pharmacology, Geller, H.M., Ed., 1986, Chap.1. 23. DiChiara, G., A motivational learning hypothesis of the role of dopamine in compulsive drug use, J. Psychopharmacol., 12, 54, 1998. 24. DeWit, H., Priming effects with drugs and other reinforcers, Exp. Clin. Psychopharmacol., 4(1), 5, 1996. 25. DeWit, H. and Stewart, J., Reinstatement of cocaine-reinforced responding in the rat, Psychopharmacology, 75, 134, 1981. 26. Everitt, B.J., Parkinson, J. A., Olmstead, M.C., Arroyo, M., Robledo, P., and Robbins, T.W., Associative processes in addiction and reward, in the role of amygdala-ventral striatal subsystems, in Advances from the Ventral Striatum to the Extended Amygdala, McGinty, J.F., Ed., Annals of the New York Academy of Sciences, New York, 877, 412, 1999. 27. Grant, S. et al., Activation of memory circuits during cue-induced cocaine craving, Proc. Natl. Acad. Sci. U.S.A., 93(21), 12040, 1996. 28. Griffiths, R.R., Bigelow, G.E., and Henningfield, J.E., Animal and human drug-taking behavior, in Advances in Substance Abuse Behavioral and Biological Research, Mello, N.K., Ed., JAI Press, Greenwich, CT, 1980, Chap. 3. 29. Green, J.D., A simple microelectrode for recording from the central nervous system, Nature, 182, 1958. 30. Groenewegen, H.J., Mulder, A.B., Beijer, A.V.J., Write, C.I., Lopes Da Silva, F.H., and Pennartz, C.M.A., Hippocampal and amygdaloid interactions in the nucleus accumbens, Psychobiology, 27(2), 149, 1999. 31. Hyman, S.E. and Malenka, R.C., Addiction and the brain: the neurobiology of compulsion and its persistence, Nat. Rev. Neurosci., 2(10), 695, 2001. 32. Ito, R., Dalley, J.W., Howes, S.R., Robbins, T.W., and Everitt, B.J., Dissociations in conditioned dopamine release in the nucleus accumbens core and shell in response to cocaine cues and during cocaine-seeking behavior in rats, J. Neurosci., 20(19), 7589, 2000.

2345_frame_MASTER.book Page 207 Wednesday, October 23, 2002 1:03 PM

Chronic Extracellular Recording to Studies of Drug Self-Administration

207

33. Janak, P.H., Multichannel neural ensemble recording during alcohol self-administration, in Methods for Alcohol-Related Neuroscience Research, Liu, Y. and Lovinger, D.M., Eds., CRC Press, Boca Raton, FL, 2002, pp. 243–259. 34. Janak, P.H., Chang, J.-Y., and Woodward, D.J., Neuronal spike activity in the nucleus accumbens of behaving rats during ethanol self-administration, Brain Res., 817, 172, 1999. 35. Johanson, C.E. and Balster, R.L., A summary of the results of a drug self-administration study using substitution procedures in rhesus monkeys, Bull. Narcotics, 30(3), 55, 1978. 36. Kazdin, A.E., Statistical analyses for single-case experimental designs, in Single Case Experimental Designs: Strategies for Studying Behavior Change, 2nd ed., Barlow, D.H. and Hersen, M., Eds., Allyn and Bacon, Needham Heights, MA, 1984. 37. Kelley, A.E., Neural integrative activities of nucleus accumbens subregions in relation to learning and motivation, Psychobiology, 27(2), 198, 1999. 38. Koob, G.F. and Bloom, F.E., Cellular and molecular mechanisms of drug dependence, Science, 242(4879), 715–723, 1988. 39. Koob, G.F. and Le Moal, M., Drug abuse: hedonic homeostatic dysregulation, Science, 52, 1997. 40. Koob, G.F. and Le Moal, M., Drug addiction, dysregulation of reward, and allostasis, Neuropsychopharmacology, 24(2), 97, 2001. 41. Koob, G.F., Sanna, P.P., and Bloom, F.E., Neuroscience of addiction, Neuron, 21, 467, 1998. 42. Kosobud, A.E.K., Harris, G.C., and Chapin, J.K., Behavioral associations of neuronal activity in the ventral tegmental area of the rat, J. Neurosci., 14, 7117, 1994. 43. Lavoi, A.M. and Mizumori, S.J.Y., Spatial-, movement, and reward-sensitive discharge by medial ventral striatum neurons of rats, Brain Res., 638, 157, 1994. 44. Lee, R.-S., Criado, J. R., Koob, G.F., and Henriksen, S.J., Cellular responses of nucleus accumbens neurons to opiate-seeking behavior: sustained responding during heroin self-administration, Synapse, 33(1), 49, 1999. 45. Lemon, R., The development and use of single neuron recording techniques in conscious animals, IBRO Handbook Series: Methods in the Neurosciences Vol. 4, Methods for Neuronal Recording in Conscious Animals, Smith, A.D., Ed., John Wiley & Sons, New York, NY, 1983. 46. Moore, G.P., Perkel, D., and Segundo, J. P., Statistical analysis and functional interpretation of neuronal spike data, Annu. Rev. Physiol., 28, 493, 1966. 47. Moxon, K.A., Multichannel electrode design: considerations for different applications, in Methods for Neural Ensemble Recordings, Nicolelis, M.A.L., Ed., CRC Press, Boca Raton, FL, 1999, 25. 48. Nestler, E.J., Molecular neurobiology of addiction, Am. J. Addictions, 10(3), 201–217, 2001. 49. Nicola, S.M. and Deadwyler, S.A., Firing rate of nucleus accumbens neurons is dopamine-dependent and reflects the timing of cocaine-seeking behavior in rats on a progressive ratio schedule of reinforcement, J. Neurosci., 20(14), 5526, 2000. 50. Nicola, S.M., Surmeier, D.J., and Malenka, R.C., Dopaminergic modulation of neuronal excitability in the striatum and nucleus accumbens, Annu. Rev. Neurosci., 23(1), 185, 2000. 51. Nicolelis, M.A.L., Stambaugh, C.R., Brisben, A., and Laubach, M., Methods for simultaneous multisite neural ensemble recordings in behaving primates, in Methods for Neural Ensemble Recordings, Nicolelis, M.A.L., Ed., CRC Press, Boca Raton, FL, 1999, 121.

2345_frame_MASTER.book Page 208 Wednesday, October 23, 2002 1:03 PM

208

Methods in Drug Abuse Research: Cellular and Circuit Level Analyses

52. O’Brien, C.P., Childress, A.R., Ehrman, R., and Robbins, S., J., Conditioning factors in drug abuse: can they explain compulsion, J. Psychopharmacol., 12(1), 15, 1998. 53. O’Donnell, P. and Grace, A.A., Dopaminergic reduction of excitability in nucleus accumbens neurons recorded in vitro, Neuropsychopharmacology, 15(1), 87, 1996. 54. Pan, H.-T., Menacherry, S., and Justice, J. B., Jr., Differences in the pharmacokinetics of cocaine in naïve and cocaine-experienced rats, J. Neurochem., 46(4), 1299, 1991. 55. Parkinson, J. A., Olmstead, M.C., Burns, L.H., Robbins, T.W., and Everitt, B.J., Dissociation in effects of lesions of the nucleus accumbens core and shell on appetitive pavlovian approach behavior and the potentiation of conditioned reinforcement and locomotor activity by d-amphetamine, J. Neurosci., 19(6), 2401, 1999. 56. Paxinos, G. and Watson, C., The Rat Brain in Stereotaxic Coordinates, Academic Press, San Diego, 1998. 57. Pederson, C. L, Wolske, M., Peoples, L.L., and West, M.O., Firing rate dependent effect of cocaine on single neurons of the rat dorsolateral striatum, Brain Res., 760, 261–265, 1997. 58. Pennartz, C.M.A., Dollerman-van der Weel, M.J.L., and Lopes da Silva, F.H., Differential membrane properties and dopamine effects in the shell and core of the rat nucleus accumbens studied in vitro, Neurosci. Lett., 136, 109, 1992. 59. Peoples, L.L., Wolske, M., Dworkin, S.I., Smith, J.E., Deadwyler, S.A., and West, M.O., A method for recording single unit activity during IV self-administration of drugs in the freely moving rat, Soc. Neurosci. Abstr., 15, 1097, 1989. 60. Peoples, L.L., Bibi, R., and West, M.O., Effects of intravenous self-administered cocaine on single cell activity in the nucleus accumbens of the rat, in Problems of Drug Dependence, 1993: Proc. 55th Annu. Sci. Meeting College Problems Drug Dependence, Vol. II, National Institute on Drug Abuse Research Monograph #141, Harris, L., Ed., US GPO, Washington, D.C., 1994, 326. 60a. Peoples, L.L. and Cavanaugh, D.J., Drug-induced amplification of neural signals related to drug cues during cocaine self-administration: possible neurophysiological mechanism, Soc. Neurosci. 32nd Annu. Mtg., 2002. 61. Peoples, L.L., Gee, F., Bibi, R., and West, M.O., Phasic firing time locked to cocaine self-infusion and locomotion: dissociable firing patterns of single nucleus accumbens neurons in the rat, J. Neurosci., 18, 7588, 1998a. 62. Peoples, L.L., Uzwiak, A.J., Gee, F., and West, M.O., Tonic inhibition of single nucleus accumbens neurons in the rat: a predominant but not exclusive firing pattern induced by cocaine self-administration sessions, J. Neurosci., 86, 13, 1998b. 63. Peoples, L.L., Uzwiak, A.J., Gee, F., Fabbricatore, A.T., Muccino, K.J., Mohta, B.D., and West, M.O., Phasic accumbal firing may contribute to the regulation of drug taking during intravenous cocaine self-administration sessions, in Advances from the Ventral Striatum to the Extended Amygdala, McGinty, J.F., Ed., Ann. N.Y. Acad. Sci., 877, 781, 1999a. 64. Peoples, L.L., Uzwiak, A.J., Gee, F., and West, M.O., Operant behavior during sessions of intravenous cocaine infusion is necessary and sufficient for phasic firing of single nucleus accumbens neurons, Brain Res., 757, 280, 1997. 65. Peoples, L.L., Uzwiak, A.J., Gee, F., and West, M.O., The predominant changes in the firing rate of single nucleus accumbens neurons during intravenous cocaine selfadministration sessions mirror changes in accumbal dopamine, Am. Soc. Neurochem. 30th Annu. Meeting New Orleans, LA., J. Neurochem., 72, 85, 1999b. 66. Peoples, L.L., Uzwiak, A.J., Gee, F., and West, M.O., Tonic firing of rat nucleus accumbens neurons: changes during the first 2 weeks of daily cocaine self-administration sessions, Brain Res., 822, 231, 1999c.

2345_frame_MASTER.book Page 209 Wednesday, October 23, 2002 1:03 PM

Chronic Extracellular Recording to Studies of Drug Self-Administration

209

67. Peoples, L.L. and West, M.O., Phasic firing of single neurons in the rat nucleus accumbens correlated with the timing of intravenous cocaine self-administration, J. Neurosci., 16(10), 3459, 1996. 68. Perkel, D.H., Gerstein, G.L., and Moore, G.P., Neuronal spike trains and stochastic point processes. I. The single spike train, J. Biophys., 7, 391, 1967. 69. Pettit, H.O. and Justice, J.B., Jr., Dopamine in the nucleus accumbens during cocaine self-administration as studied by in vivo microdialysis, Pharmacol. Biochem. Behav., 34, 899, 1989. 70. Pickens, R., Meisch, A., and Thompson, T., Drug self-administration: an analysis of the reinforcing effects of drugs, in Handbook of Psychopharmacology, Vol. 12, Iversen, L.L., Iversen, S.D., and Snyder, S.H., Eds., Plenum, New York, NY, 1978, Chap. 1. 71. Ranck, J.B., Jr., Kubie, J.L., Fox, S.E., Wolfson, S., and Muller, R.U., Single neuron recording in behaving mammal: bridging the gap between neuronal events and sensorbehavioral variables, in Behavioral Approaches to Brain Research, Robinson, T.E., Ed., Oxford University Press, New York, NY, 1983, Chap. 5. 72. Robbins, T.W. and Everitt, B.J., Drug addiction: bad habits add up, Nature, 398(6728), 567, 1999. 73. Robinson, T.E. and Berridge, K.C., The neural basis of drug craving: an incentivesensitization theory of addiction, Brain. Res. Rev., 18, 247, 1993. 74. Rolls, E.T., Neurophysiology and functions of the primate amygdala, and the neural basis of emotion, in The Amygdala: A Functional Analysis, 2nd ed., Aggleton, J.P., Ed., Oxford University Press, New York, NY, 2000, p. 447. 75. Ross, E.M., Pharmacodynamics: mechanisms of drug action and the relationship between drug concentration and effect, in Goodman and Gilman’s The Pharmacological Basis of Therapeutics, Hardman, J.G., Limbird, L., Molinoff, P.B., Ruddon, R.W., and Gilman, A.G., Eds., McGraw-Hill, New York, NY, 1996. 76. Salamone, J.D., Complex motor and sensorimotor functions of striatal and accumbens dopamine: involvement in instrumental behavior processes, Psychopharmacology, 107, 160, 1992. 77. Sameshima, K. and Baccala, L.A., Trends in multichannel neural ensemble recording instrumentation, in Methods for Neural Ensemble Recordings, Nicolelis, M.A.L., Ed., CRC Press, Boca Raton, 1999, 47. 78. Schultz, W., Apicella, P., Scarnati, E., and Ljungberg, T., Neuronal activity in monkey ventral striatum related to the expectation of reward, J. Neurosci., 12(12), 4595, 1992. 79. Schultz, W., Multiple reward signals in the brain, Nat. Rev., 1, 199, 2000. 80. Siegel, S. and Castellan, N.J., Nonparametric Statistics for the Behavioral Sciences, 2nd ed., McGraw-Hill, New York, 1988. 81. Schmidt, E.M., Electrodes for many single neuron recordings, in Methods for Neural Ensemble Recordings, Nicolelis, M.A.L. and Simon, S.A., Eds., CRC Press, Boca Raton, FL, 1999. 82. Stewart, J., Reinstatement of heroin and cocaine self-administration behavior in the rat by intracerebral application of morphine in the ventral tegmental area, Pharmacol. Biohem. Behav., 20, 917, 1984. 83. Stewart, J., Neurobiology of conditioning to drugs of abuse, Ann. N.Y. Acad. Sci., 654, 335, 1992. 84. Stewart, J., deWit, H., and Eikelboom, R., Role of unconditioned and conditioned drug effects in the self-administration of opiates and stimulants, Psychol. Rev., 91(2), 251, 1984.

2345_frame_MASTER.book Page 210 Wednesday, October 23, 2002 1:03 PM

210

Methods in Drug Abuse Research: Cellular and Circuit Level Analyses

85. Uzwiak, A.J., Guyette, F., West, M.O., and Peoples, L.L., Neurons in accumbens subterritories of the rat: phasic firing time-locked within seconds of intravenous cocaine self-infusion, Brain Res., 767, 363, 1997. 86. Vanderschuren, L.J.M.J. and Kalivas, P.W., Alterations in dopaminergic and glutamatergic transmission in the induction and expression of behavioral sensitization: a critical review of preclinical studies, Psychopharmacology, 151, 99–120, 2000. 87. Van Dongen, J.J., Remie, R., Rensema, J.W., and van Wunnik, G.H.J., Manual of Microsurgery on the Laboratory Rat, Techniques in the Behavioral and Neural Sciences, Vol. 4, Huston, J.P., Series Ed., Elsevier, New York, 1990. 88. Volkow, N.D. et al., Association of methylphenidate-induced craving with changes in right striato-orbitofrontal metabolism in cocaine abusers: implications in addiction, Am. J. Psychiatry, 156(1), 19, 1999. 89. Weiss, F., Maldonado-Vlaar, C.S., Parsons, L.H., Kerr, T.M., Smith, D.L., and BenSharhar, O., Control of cocaine-seeking behavior by drug-associated stimuli in rats: effects on recovery of extinguished operant-responding and extracellular dopamine levels in amygdala and nucleus accumbens, Proc. Natl. Acad. Sci. U.S.A., 97, 4321, 2000. 90. West, M.O., Anesthetics eliminate somatosensory-evoked discharges of neurons in the somatotopically organized sensorimotor striatum of the rat, J. Neurosci., 18(21), 9055, 1998. 91. West, M.O., Peoples, L.L, Michael, A.J., Chapin, J.K., and Woodward, D.J., Lowdose amphetamine elevated movement-related firing of rat striatal neurons, Brain Res., 745, 331, 1997. 92. West, M.O., Peoples, L.L., Wolske, M., and Dworkin, S.I., Psychomotor stimulant effects on single neurons in awake, behaving rats, in Neurobiological Approaches to Brain-Behavior Interaction, National Institute on Drug Abuse Research monograph series #124, Brown, R. and Frascella, J., Eds., US GPO, Washington, D.C., 1992, pp. 57–71. 93. Wheeler, B.C., Automatic discrimination of single units, in Methods for Neural Ensemble Recordings, Nicolelis, M.A.L., Ed., CRC Press, Boca Raton, FL, 1999, Chap. 4. 94. Wiener, S.I., Spatial and behavioral correlates of striatal neurons in rats performing a self-initiated navigation task, J. Neurosci., 13(9), 3802, 1993. 95. Wise, R.A. and Bozarth, M.A., A psychomotor stimulant theory of addiction, Psychol. Rev., 94, 469, 1987. 96. Wise, R.A., Cognitive factors in addiction and nucleus accumbens function: some hints from rodent models, Psychobiology, 27(2), 300, 1999. 97. Wise, R.A., Newton, P., Leeb, K., Burnette, B., Pocock, D., and Justice, J.B., Jr., Fluctuations in nucleus accumbens dopamine concentrations during intravenous cocaine self-administration in rats, Psychopharmacology, 120, 10, 1995. 98. Woodward, D.J., Chang, J.-Y., Janak, P., Azarov, A., and Anstrom, K., Mesolimbic neuronal activity across behavioral states, in Advancing from the Ventral Striatum to the Extended Amygdala, McGinty, J.F., Ed., Ann. N.Y. Acad. Sci., New York, NY, 1999, p. 91. 99. Wyvell, C.L. and Berridge, K.C., Intra-accumbens amphetamine increases the pure incentive salience of a pavlovian cue for food reward: enhancement of “wanting” without either “liking” or reinforcement, J. Neurosci., 21(19), 7831, 2000. 100. Wyvell, C.L. and Berridge, K.C., Incentive sensitization by previous amphetamine exposure: increased cue-triggered “wanting” for sucrose reward, J. Neurosci., 21(19), 7831, 2001.

2345_frame_MASTER.book Page 211 Wednesday, October 23, 2002 1:03 PM

Chronic Extracellular Recording to Studies of Drug Self-Administration

211

101. Yokel, R.A., Intravenous self-administration: response rates, the effects of pharmacological challenges, and drug preferences, in Methods of Assessing the Reinforcing Properties of Abused Drugs, Bozarth, M.A., Ed., Springer-Verlag, New York, NY, 1989, Chap. 1. 102. Zahm, D.S., An integrative neuroanatomical perspective on some subcortical substrates of adaptive responding with emphasis on the nucleus accumbens, Neurosci. Biobehav. Rev., 24, 85, 2000. 103. Zhang, X.-F., Hu, X.-T., and White, F.J., Whole-cell plasticity in cocaine withdrawal: reduced sodium currents in nucleus accumbens neurons, J. Neurosci., 18(1), 488, 1998.

2345_frame_MASTER.book Page 212 Wednesday, October 23, 2002 1:03 PM

2345_frame_MASTER.book Page 213 Wednesday, October 23, 2002 1:03 PM

8

Determination of Drug Actions on Multiple Simultaneously Recorded Neurons across Functionally Connected Networks David M. Devilbiss and Barry D. Waterhouse

CONTENTS 8.1 8.2 8.3 8.4

Introduction .................................................................................................214 Model Sensory System ...............................................................................215 Surgical Procedures, Recording and Stimulating Electrodes......................217 Components of the Multineuron Recording System.................................. 218 8.4.1 Instrumentation ...............................................................................218 8.4.2 Experimental Session...................................................................... 221 8.5 Waveform Discrimination ...........................................................................221 8.5.1 Online Spike Sorting ......................................................................221 8.5.2 Offline Validation Single-Neuron Recording .................................222 8.6 Strategies for Analyzing Multiunit Spike Train Activity ............................224 8.6.1 Data Preprocessing .........................................................................224 8.6.2 Single-Unit Data .............................................................................224 8.6.3 Population Data................................................................................228 8.6.3.1 Functional Connectivity....................................................228 8.6.3.2 Factor Analysis ................................................................229 8.7 Discussion ...................................................................................................231 8.7.1 Advantages and Limitations ...........................................................231 8.7.2 Future Directions............................................................................. 233 Acknowledgments ................................................................................................233 References ............................................................................................................234

213

2345_frame_MASTER.book Page 214 Wednesday, October 23, 2002 1:03 PM

214

Methods in Drug Abuse Research: Cellular and Circuit Level Analyses

8.1 INTRODUCTION Humans have used psychoactive substances to alter cognitive processes and behavior since the time of ancient civilizations. However, we still do not fully understand the linkage between the impact of these substances on neurons and the resulting behavior. As discussed by other authors in this book, many methods exist for determining how drugs of abuse or any pharmacological agent affects brain function at the molecular, cellular, or systemic level. This continuum stretches from effects on individual ion channels to effects on individual neuronal signaling properties to global indices of cortical function. Our approach utilizes simultaneous, multiple single-cell recordings to determine the actions of psychoactive substances on local circuit and larger neural network functions. These studies describe how information encoded within a group of neurons is altered following administration of drugs of abuse in the awake, freely moving animal. Targeting circuits or networks known to code and convey reward signals throughout the brain is one strategy for understanding the neural basis of drug-seeking and drug-taking behavior. A second strategy is to examine drug actions on circuits or networks that encode information about the drug-related experience. An example of this alternative approach is to determine how cells and circuits in primary sensory pathways of the brain process and relay information after drug administration. In our studies, we have used the rat trigeminal somatosensory system as a model to address these questions. This approach offers advantages over other techniques for examining the larger question of how drugs with abuse potential alter sensory perceptual processes. For example, the physiology and anatomy of the rodent trigeminal somatosensory system have been studied in depth; thus, the techniques to decipher information coded within neurons along this pathway and techniques to obtain that information are available to us. Moreover, drug actions on sensory neurons and circuits are more readily interpreted in terms of neural network function. This is in contrast to reward circuits of the brain, where precise descriptions of the information carried by those neurons are still emerging. Moreover, within rewardlimbic networks it is difficult to unequivocally interpret what changes in neuronal activity mean in relationship to circuit function. As with other drugs with abuse potential, cocaine has substantial impact on sensory signal processing (see Waterhouse, Chap. 5, this volume) and perception18 in laboratory animals and humans. To understand how cocaine affects sensory perceptual processes our recent studies have employed multichannel recording techniques to investigate the actions of cocaine on sensory signal processing in the mammalian brain. During the development of this research a number of pertinent issues have been considered, including: 1. Selection of a model sensory network 2. Consistent presentation of well-defined sensory information to be encoded by the network 3. Methods to record neural activity in the network and properly discriminate individual neural recordings

2345_frame_MASTER.book Page 215 Wednesday, October 23, 2002 1:03 PM

Drug Actions on Multiple Simultaneously Recorded Neurons

215

4. Decoding the neuronal spike trains from the network of neurons to recreate information carried by those neurons 5. Interpreting changes in neural spike-train activity following administration of psychoactive substances In this chapter, we will discuss each of these points in addition to potential new directions of analysis to determine how drugs of abuse affect sensory neural network functions.

8.2 MODEL SENSORY SYSTEM There are many good model systems, from the cricket29,48 to the nonhuman primate, for examining the flow of information through a network of neurons. We have chosen the rat trigeminal somatosensory system as a model for several reasons. First, the rat has been widely used in studies of drugs of abuse. Second, neuronal response properties and the coding of tactile stimuli by the trigeminal somatosensory system have been well characterized both anatomically and physiologically. Third, we can take advantage of techniques that have been developed in rats for delivering somatosensory stimuli and recording spike-train activity of many single neurons at multiple brain sites under waking conditions. The trigeminal somatosensory system of the rat encodes velocity, direction, and magnitude of displacement of each whisker (mystacial vibrissae) on the animal’s face. Rodents primarily rely on this sophisticated tactile system to navigate their environment. As suggested above, this ascending sensory pathway provides several unique advantages for studying cocaine actions on sensory functions. The anatomy of this sensory pathway has been well characterized by numerous investigators over the past 20 years5 (Figure 8.1). This sensory pathway decussates completely, includes relays in the ipsilateral trigeminal complex (principal nucleus of V–PrV and spinal nucleus of V–SpV) and contralateral thalamus (ventrobasal–VB and posteromedial–POm nuclei), and projects to a cytoarchitectonically specialized region of the contralateral primary somatosensory cortex referred to as the “barrel field” (BF) as well as a vibrissae-related region of the secondary somatosensory cortex (SII). In addition, a high degree of topographic ordering with respect to individual vibrissae is maintained throughout its extent.57 For example, cellular aggregates termed barrels in cortex,56 barreloids in thalamus,49 and barrelettes in the brainstem trigeminal complex25 represent individual vibrissae on the whisker pad at each level along the pathway. The physiology of the trigeminal somatosensory pathway is also well characterized. Neurons located in PrV have small receptive fields40 and give rise to highly topographic fast-conducting, trigeminothalamic lemniscal projections that mediate fine tactile discrimination. By contrast, cells located in the SpV have larger receptive fields and give rise to less topographic and slower-conducting, paralemniscal projections that are believed to convey high-threshold cutaneous and noxious information. From the brainstem somatosensory information reaches the cortex by way of parallel pathways that relay through two major nuclei in the thalamus: the VPM

2345_frame_C08 Page 216 Wednesday, October 23, 2002 3:22 PM

216

Methods in Drug Abuse Research: Cellular and Circuit Level Analyses

BARREL BFC/SII

POm

VB nRT

SpV

PrV

DR

LC

FIGURE 8.1 Schematic diagram indicates the flow of information from the mystacial vibrissae to multiple subcortical and cortical sites along the trigeminal somatosensory pathway in rat. Tactile information coded by the whiskers is relayed to the spinal nucleus of V (SpV) and principal nucleus of V (PrV) within the brainstem trigeminal complex. Information is then transmitted to the contralateral ventrobasal (VB), posteriormedial (POm), and reticular (nRT) thalamic nuclei. Output from POm is directed to secondary (SII) cortex and interbarrel/septal regions of barrel field cortex (BFC), while signals from VB are conveyed to the barrel hollows in layer IV of BFC. Output from locus coeruleus (LC) and dorsal raphe (DR) nuclei to brainstem, thalamic, and cortical levels of the pathway is also indicated.

(also known as ventrobasal, VB) and posteromedial, PoM.8,9,13 The lemniscal pathway (VPM–cortex) conducts impulses rapidly and with a high degree of fidelity. The paralemniscal pathway (POm–cortex) on the other hand is slow-conducting and adds a dynamical dimension to the coding of stimulus properties. The response of somatosensory cortical neurons to cutaneous stimuli represents an integration of inputs from each of these input pathways.5,12,13 The VB thalamus also maintains a reciprocal, topographic relationship with the nRT.20,21 NRT cells receive inputs from primary somatosensory cortex and exert a GABAergic inhibitory influence over sensory driven activity in the VB thalamus.41 Essentially output from nRT selectively regulates the excitability of VPM relay neurons and, by so doing, modulates the flow of information from the periphery through the thalamus and onto the cortex. Thus, as signals reverberate through cortico-thalamo-cortical circuitry (cortex-nRT-VPM-cortex and cortex-POm-cortex) sensory signals are encoded.31 Traditionally, the trigeminal somatosensory pathway has been studied by eliciting electrical discharge patterns from cells with electromechanical deflections of individual vibrissae.43,44,58 Precise control and uniformity of whisker movements can be achieved with solenoid- or piezoelectric-driven stimulators. The precision of these

2345_frame_MASTER.book Page 217 Wednesday, October 23, 2002 1:03 PM

Drug Actions on Multiple Simultaneously Recorded Neurons

217

Implanted Electrode Array

Subcutaneous Whisker Pad Stimulator

4 3 5 A B C D E

2

1

D2

FIGURE 8.2 Schematic diagram indicating the placement of the subcutaneous whisker-pad stimulating electrode and contralateral recording microelectrodes. Whisker C3 is the most frequently targeted whisker; however, in this case, the stimulator was anchored at the base of whisker follicle D2. Electrical connections for the whisker-pad stimulating electrode, VPM, and SI recording arrays are embedded in dental cement and anchored to the skull.

stimulators is decreased in experiments using awake, freely moving animals since the position of the whiskers relative to stimulator and the parameters of stimulation are altered following every movement of the animal. One alternative is to directly activate the infraorbital nerve with electrical stimulation via a nerve cuff.17 However, such implants cannot provide spatially or anatomically selective activation of the whisker pathway. To address the question of how psychoactive substances alter sensory perceptual processes in the awake, freely moving animal we have developed an indwelling subcutaneous electrode that is capable of delivering repetitive, focal electrical stimulation to the whisker pad in the behaving rat over periods of days to weeks (Figure 8.2).10,26,39,557

8.3 SURGICAL PROCEDURES, RECORDING AND STIMULATING ELECTRODES Extracellular recordings are made from multiple single units within the VPM thalamus and BF cortex using chronically implanted microwire bundles (SB103, NB

2345_frame_MASTER.book Page 218 Wednesday, October 23, 2002 1:03 PM

218

Methods in Drug Abuse Research: Cellular and Circuit Level Analyses

Labs, Dennison, TX). Illustrations of these electrodes, descriptions of surgical procedures, and typical recordings can be found in several previously published reports.17,32–34 This strategy permits us to monitor the spike-train activity of dozens of single neurons as punctate sensory signals are encoded and ascend through the trigeminal somatosensory network. Sensory input is generated via a subcutaneous stimulation electrode placed within the whisker pad adjacent to a specifically targeted vibrissae follicle (Figure 8.2). Detailed descriptions of the whisker stimulator assembly and placement have been described previously.11 Briefly, implantation of the electrode occurs in three steps: (1) preparation of the stimulation site, (2) insertion of the electrode, and (3) assembly of the electrode components. The electrode is comprised of a twisted pair of seven-stranded stainless steel Teflon-coated wires (0.001 mm bare, 0.0055 mm coated; A-M Systems Inc., Sequim, WA) inserted under the skin. This is accomplished by first tunneling a 26-gauge stainless steel cannula (U-HTX-26, Small Parts Inc., Miami Lakes, FL) into the whisker pad from an initial incision in the scalp to the targeted vibrissa. Second, the twisted pair electrode is placed within a 20-gauge needle that is inserted along the path created by the 26-gauge cannula. The electrode remains hooked around the targeted whisker after the needle is removed from the whisker pad. The electrode is assembled by crimping the free ends of the electrode wires into a conductive grease-filled (Chemtronics) connector (MS303–120, Plastics One, Inc., Roanoke, VA). After a week of recovery, electrical stimulation of the whisker pad activates the somatosensory afferent pathway and elicits responses in thalamic and cortical neurons that are similar in magnitude, latency, and duration to responses evoked by mechanical deflection of individual whiskers (Figure 8.3). Thus, this preparation provides the opportunity to study neural representations of sensory information in the awake animal, before and after systemic administration of psychoactive compounds.

8.4 COMPONENTS OF THE MULTINEURON RECORDING SYSTEM 8.4.1 INSTRUMENTATION Within the last decade several commercial solutions have emerged to accomplish multichannel, many-neuron recordings in the intact animal. However, a number of issues were considered as we assembled a system to monitor the effects of drugs of abuse on functionally connected networks in awake rats. Reliability, ease of use, and the ability to chronically record for days to weeks from a single animal were the highest priorities of this system. These requirements prompted us to integrate several commercially available and custom devices for recording multiple single neurons, stimulating the whisker pad and monitoring animal behavior as diagramed in Figure 8.4 (see Color Figure 8.4 following page 50). The Realtime Acquisition System Programs for Unit Timing In Neuroscience (RASPUTIN) software and Multichannel Acquisition Processor (MAP) hardware (Plexon Inc., Dallas, TX) are configured to simultaneously record 32 channels of neuronal activity in up to four different anatomical locations in the

0.05 T2 0.04 0.03 0.02 0.01 0 -0.1 -0.05 0.05 T3 0.04 0.03 0.02 0.01 0 -0.1 -0.05 0.05 T4 0.04 0.03 0.02 0.01 0 -0.1 -0.05 0.05 0.04 T 5 0.03 0.02 0.01 0 -0.1 -0.05 0.05 T6 0.04 0.03 0.02 0.01 0 -0.1 -0.05 0.05 T7 0.04 0.03 0.02 0.01 0 -0.1 -0.05

0.05 T1 0.04 0.03 0.02 0.01 0 -0.1 -0.05

0.05

0.05

0.05

0.05

0.05

0

0

0

0

0.05

0.05

0

0

0

0.1

0.1

0.1

0.1

0.1

0.1

0.1

0.15

0.15

0.15

0.15

0.15

0.15

0.15

0.05 0.04 T 8 0.03 0.02 0.01 0 0.2 -0.1 -0.05 0.05 0.04 T 9 0.03 0.02 0.01 0 0.2 -0.1 -0.05 0.05 0.04 C 1 0.03 0.02 0.01 0 0.2 -0.1 -0.05 0.05 0.04 C 2 0.03 0.02 0.01 0 0.2 -0.1 -0.05 0.05 0.04 C 3 0.03 0.02 0.01 0 0.2 -0.1 -0.05 0.05 0.04 C 4 0.03 0.02 0.01 0 0.2 -0.1 -0.05 0.05 0.04 C 5 0.03 0.02 0.01 0 0.2 -0.1 -0.05 0.05

0.05

0.05

0.05

0.05

0

0

0

0

0.05

0.05

0

0

0

0.1

0.1

0.1

0.1

0.1

0.1

0.1

0.15

0.15

0.15

0.15

0.15

0.15

0.15

Time (sec.)

0.2

0.2

0.2

0.2

0.2

0.2

0.2

0.05 0.04 T 1 0.03 0.02 0.01 0 -0.1 -0.05 0.05 0.04 T 2 0.03 0.02 0.01 0 -0.1 -0.05 0.05 0.04 T 3 0.03 0.02 0.01 0 -0.1 -0.05 0.05 0.04 T 4 0.03 0.02 0.01 0 -0.1 -0.05 0.05 0.04 T 5 0.03 0.02 0.01 0 -0.1 -0.05 0.05 0.04 T 6 0.03 0.02 0.01 0 -0.1 -0.05 0.05 0.04 T 7 0.03 0.02 0.01 0 -0.1 -0.05 0.05

0.05

0.05

0.05

0.05

0

0

0

0

0.05

0.05

0

0

0

0.1

0.1

0.1

0.1

0.1

0.1

0.1

0.15

0.15

0.15

0.15

0.15

0.15

0.15

0.05 0.04 T 8 0.03 0.02 0.01 0 0.2 -0.1 -0.05 0.05 0.04 T 9 0.03 0.02 0.01 0 0.2 -0.1 -0.05 0.05 0.04 C 1 0.03 0.02 0.01 0 0.2 -0.1 -0.05 0.05 0.04 C 2 0.03 0.02 0.01 0 0.2 -0.1 -0.05 0.05 0.04 C 3 0.03 0.02 0.01 0 0.2 -0.1 -0.05 0.05 0.04 C 4 0.03 0.02 0.01 0 0.2 -0.1 -0.05 0.05 0.04 C 5 0.03 0.02 0.01 0 0.2 -0.1 -0.05

B. Whisker Mechanical Stimulation

0.05

0.05

0.05

0.05

0.05

0

0

0

0

0.05

0.05

0

0

0

0.1

0.1

0.1

0.1

0.1

0.1

0.1

0.15

0.15

0.15

0.15

0.15

0.15

0.15

0.2

0.2

0.2

0.2

0.2

0.2

0.2

Drug Actions on Multiple Simultaneously Recorded Neurons

FIGURE 8.3 Perievent stimulus histograms from nine thalamic (T1–9) and five cortical neurons (C1–5) recorded simultaneously during whisker-pad electrical stimulation (2.5 mA) (A) or single-whisker mechanical stimulation (400 mm) (B). In all histograms, the x-axis represents time before and after stimulus presentation (bin = 1 msec); the y-axis represents the probability of the neuron discharging in a given bin. Each cell demonstrates a response to both whisker-pad stimulation and whisker deflection with approximately the same latency and intensity. Slight differences may occur depending upon the spread of electrical stimulation to surrounding whiskers.

Probability of Discharge

A. Whisker Pad Electrical Stimulation

2345_frame_MASTER.book Page 219 Wednesday, October 23, 2002 1:03 PM

219

2345_frame_MASTER.book Page 220 Wednesday, October 23, 2002 1:03 PM

220

Methods in Drug Abuse Research: Cellular and Circuit Level Analyses

Parallel Control Line Rasputin Sort Client Nex

Behavioral Interface 1401plus

Spike2

OUTPUT

Plexon Inc. Stimulator 1 (Programmable

CD Writer Se L i ri al ne C on tro l

Stimulator 2

Pentium II 350 mHz

MXI Bus

TIM

SIG

Neural Signal Input

SIG

Plexon Inc. PreAmp

Synchronization Line

Oscilloscope 00.00.00

BETA VCR

Video

Video Re-formatter

Audio Monitor

Video timer

FIGURE 8.4 Schematic of the multineuronal recording apparatus for the awake, freely moving rat preparation. The rat in the behavioral chamber (bottom left) is connected to preamplifiers and stimulators via a detachable lightweight cable. A behavioral interface (tone generator, etc.) controls the behavioral environment as well as the programmable stimulators that are, in turn, regulated by the 1401plus (CED) and Spike2 software (CED) scripts on the host computer. Additionally, the preamp box feeds the Plexon mainframe that performs online waveform discriminations but is controlled by programs on the host computer. Furthermore, the 1401plus and Plexon discrimination mainframe are synchronized with a video timer so that the animal’s behavior, the time at which it occurred, and a sample of neuronal activity can be recorded onto videotape. The neuronal electrical activity can also be monitored via audio monitor and digital oscilloscope. The behavioral control program and multiunit acquisition program run simultaneously on the host computer.

brain. Usage of these devices for recording the extracellular electrical activity of multiple single units have been described in detail elsewhere.32,33 A custom behavioral control system was developed to monitor ongoing behavior of the subject, deliver different patterns and intensities of whisker-pad stimulation, and allow timing synchronization to the MAP and RASPUTIN unit. The behavioral control system written in a scripting language for the 1401plus hardware and accompanying Spike2 software (CED, Cambridge, U.K.) is available at http://neurobio.mcphu.edu/sng/software.html. The behavioral system controls the timing and current intensity of a modified bipolar, constant current stimulus isolation unit (WPI Sarasota, FL) that delivers square wave pulses to the subject’s whisker pad. Several behavioral control systems are commercially available; however, we chose to design

2345_frame_MASTER.book Page 221 Wednesday, October 23, 2002 1:03 PM

Drug Actions on Multiple Simultaneously Recorded Neurons

221

one de novo to accommodate the transport of timing signals and other experimental information to and from the MAP system. During recording sessions the animals are housed in a testing chamber consisting of a clear Plexiglas box (18 cm wide ¥ 43 cm long ¥ 40 cm high) supported by a solid textured floor. The animal’s activity is time-stamped with a video counter (Thalner Electronic Products, Ann Arbor MI; resolution = 0.1 sec) synchronized to the MAP system and behavioral control system and videotaped (Sony sl-hf900, Panasonic wv-3260).

8.4.2 EXPERIMENTAL SESSION Following recovery from surgical procedures animals are habituated to the testing chamber, the experimenter, and the experimental procedures (whisker pad stimulation). During this habituation period single neuronal units were discriminated (see below) from each wire of the chronically implanted electrode bundles. To ensure a good yield of neurons for study, cells are discriminated during periods of spontaneous and evoke neuronal activity. Online analysis confirms that discriminated units respond to peripheral sensory stimuli. Once animals have habituated to the testing chamber and whisker stimulation, neuronal activity is recorded for one or more 20min control periods that include 400 randomly presented whisker-pad stimulations (mean inter-trial interval 2 sec, range 1.5 to 2.5 sec) of varying intensity (0.1 to 3.0 mA). The slow rate and random presentation of punctate stimuli is chosen so as to minimize habituation of the somatosensory system. When a stable baseline of neuronal responses is achieved for both control and post-saline injection periods, drug is administered and neuronal activity recorded continuously for two or more hours.

8.5 WAVEFORM DISCRIMINATION The ability to maintain stable, well-isolated recordings of single cells is a central requirement for studying spike-train activity across drug-induced states. In practice, recording from well-isolated cells with this multichannel recording system is different from traditional high-impedance glass pipette recording since the low impedance microwires can detect and discriminate spike-train activity from several individual neurons simultaneously. To study the actions of drugs of abuse on multiple individual neurons for periods of 2 to 6 h we have adopted a set of stringent criteria to confirm that spike-train activity originates from single neurons. These criteria include online spike sorting in addition to offline verification of the accuracy of the recordings and discrimination.

8.5.1 ONLINE SPIKE SORTING During adaptation of the animal to the recording chamber action potentials from putative single neurons are discriminated from electrical activity recorded from each microwire of the bundle. The MAP system provides several online discrimination options including (1) time-voltage window discrimination, (2) template matching, or (3) cluster cutting in principal component space using manual selection or an

2345_frame_MASTER.book Page 222 Wednesday, October 23, 2002 1:03 PM

222

Methods in Drug Abuse Research: Cellular and Circuit Level Analyses

automated k-means algorithm. (For a review of the spike-sorting algorithms mentioned here, see Reference 53.) In our experience, template matching produces the most accurate and reproducible discrimination of somatosensory thalamic and cortical neurons that are activated by electrical whisker-pad stimulation. Briefly, template matching creates an average, i.e., “the template,” of a series of waveforms then compares that average to new action potentials as they are recorded. This method not only accounts for the peak amplitude of the action potential, but also takes into account the entire shape of the waveform. These discriminated waveforms are digitized and time-stamped to permit offline validation that the discriminated waveforms originate from single neurons.

8.5.2 OFFLINE VALIDATION SINGLE-NEURON RECORDING Following conclusion of the experiment, the accuracy of online waveform discrimination is verified using several criteria based on waveform characteristics and spiketrain discharge patterns. These criteria include (1) peak voltage of the waveform, (2) waveform slopes, (3) scattergram of the first two principal components, and (4) spiketrain autocorrelogram. Online recordings of discriminated waveforms are referred to as “units,” whereas the designation of “single neuron” is applied only after confirmation by these established criteria. The digitized waveforms recorded for each online discriminated unit are first inspected to determine the variability of the waveform peak voltage. Just as traditional single-unit discrimination techniques apply a voltage “window,” we allow, post-hoc, the peak voltage from waveforms of a discriminated unit to vary 10% of the total waveform voltage. Second, the peak-to-peak slope of unit waveforms are permitted variability of less than 10% from the average slope. Although determined offline, these two criteria establish traditional offline time-voltage limits for unit waveform discrimination. The digitized waveforms are then plotted as points in a scattergram comprised of principal component space.53 Distinct clusters of points represent recorded waveforms that possess similar features (peak voltage, slope, etc.) that differ from other units recorded from the same electrode. In addition to the first three criteria based upon the action-potential waveform shape, spike trains from each discriminated unit are used to calculate autocorrelograms. This criterion requires that there be no activity during the 2-msec refractory period27 of the cell in order for the waveform units to qualify as potential single neurons. These four criteria were used in conjunction with one another to determine whether the waveforms discriminated as a “unit” originated from a single neuron. The failure of the autocorrelogram test or two or more of the first three validation measures (peak voltage, slope, scattergram) eliminated a set of waveforms from further consideration as a single neuron. The potential variability associated with extracellular waveform shape19 was the motivating factor for weighing these criteria unequally in qualifying “units” as waveforms originating from single neurons. A typical extracellular recording of neural activity is illustrated in Figure 8.5 (see Color Figure 8.5 following page 50) with the application of the above criteria. Four single units (Figure 8.5 B–E) were discriminated online from the electrical activity recorded from a single microwire (channel). The remaining waveforms (8.5 A) were ignored during the online discrimination procedure and remained

2345_frame_MASTER.book Page 223 Wednesday, October 23, 2002 1:03 PM

Drug Actions on Multiple Simultaneously Recorded Neurons

A

B

DSP08i

223

DSP08

F

DSP08a 2 1.5 1 0.5

C

0 –0.02

DSP08b

–0.01

0

0. 01

0. 02

–0.01

0

0. 01

0. 02

–0.01

0

0. 01

0. 02

–0.01

0

0. 01

0. 02

15

D

DSP08c

Counts/Bin

10 5 0 –0.02

40 30 20 10

E

DSP08d

0 –0.02

100 80 60 40 20 0 –0.02

Time(sec.)

FIGURE 8.5 Offline validation of waveform discrimination and spike sorting. Extracellularly recorded electrical activity was sorted as nondiscriminated waveforms (A) and single-unit discharges (B–E) with associated interspike interval autocorrelograms (right). The autocorrelogram plots interspike intervals between action potentials with preceding spikes left of zero and each subsequent spike to the right of zero. The normal refractory period for these neurons equals 2 msec; thus, if all waveforms in a recorded spike train are from the same neuron, the first two bins to the left of zero and the first two bins to the right of zero should be empty. In addition, the waveforms were subjected to a principal component analysis and plotted as a scattergram of component 1 vs. component 2 (F). Note that the cluster of points representing cell DSP08a (red) is spatially segregated from the representation of other recorded units. The visualization of the waveforms (B–E), their representation in component space (F), and generation of corresponding autocorrelograms provided objective criteria for validating the accuracy of single-unit discrimination.

2345_frame_MASTER.book Page 224 Wednesday, October 23, 2002 1:03 PM

224

Methods in Drug Abuse Research: Cellular and Circuit Level Analyses

undiscriminated. The first two verification measures (peak voltage and slope variability) were applied to the online discriminated units, resulting in the elimination of the third and fourth unit (Figure 8.5 D and E) from further analysis. Application of the autocorrelation analysis (histograms adjacent to 8.5 B–E) revealed the absence of an adequate refractory period for the second unit (Figure 8.5 C), thus eliminating it from consideration as a single neuron. Furthermore, the autocorrelation analysis confirmed the accurate elimination of the third and fourth units by the previous criteria. Lastly, the scattergram of the waveforms plotted in principal component space (Figure 8.5 F) suggest that the second, third, and fourth units were composed of mixtures of waveforms emitted from many individual neurons, again confirming that these last three units (Figure 8.5 C–E) were poorly discriminated units. Following this sequence of cross-validation tests only the first unit in this group was identified as a set of waveforms accurately discriminated from a single neuron.

8.6 STRATEGIES FOR ANALYZING MULTIUNIT SPIKE TRAIN ACTIVITY 8.6.1 DATA PREPROCESSING Videotape record analysis defined the subject’s behavioral state during all phases of the experiment, initiating the analysis of neural data from awake, freely moving animals. This additional step is important for eliminating potential confounds relating to shifts in behavioral state that alone could influence sensory neuronal responsiveness.7,17 Accordingly, drug effects are evaluated within the context of a single behavioral condition. The video data were segmented into four specified, overt spontaneous behaviors: (1) quiet resting — lying down, (2) quiet resting — sitting, (3) grooming, or (4) ambulating. The initiation and termination of each behavior was time-stamped and merged with the recorded spike-train data and other experimental information provided by the Spike2–1401plus behavioral control program. The data sets could then be analyzed for changes in spontaneous or stimulus-evoked activity of single cells following the administration of drugs of abuse or other pharmacological agents. Moreover, changes in the information contained within neuronal interactions of a functional circuit can be determined following drug administration. The scripts and functions for NeuroExplorer (Nex; Nex Technologies, Winston-Salem, NC) and Matlab (The MathWorks, Inc., Natick, MA) used in our laboratory to perform these single-unit and neuronal population analysis, as well as a sample data set, are available at http://neurobio.mcphu.edu/sng/software.html.

8.6.2 SINGLE-UNIT DATA Spontaneous and sensory-evoked discharge rates are quantified from peristimulus time histograms (PSTH) by a series of analysis scripts and functions for Nex and Matlab. Whisker stimulation-evoked discharge for each neuron is calculated by averaging the neuronal activity within an identified time window. The window is produced from Poisson (imported from Nex) or Gaussian 99% confidence intervals calculated by the Matlab analysis function EPOCH that defined the onset and

2345_frame_MASTER.book Page 225 Wednesday, October 23, 2002 1:03 PM

Drug Actions on Multiple Simultaneously Recorded Neurons

225

Probability of Discharge

0.482

0.03

0.02

0.01

0 –0.2

0.003

–0.1

0

Onset 4 ms

0.1

0.2

Termination 26 ms

Time (sec) FIGURE 8.6 Quantitative analysis of peristimulus time histograms. This histogram illustrates the excitatory response of a single thalamic neuron to electrical stimulation (0 msec. Onset, n = 400 trials) of the contralateral whisker pad. The x-axis depicts time (in 1-msec bins) before and after the stimulus presentation; the y-axis is the probability of neuronal discharge for a given bin. The response of this thalamic neuron began 4 msec post-stimulus and terminated at 26 msec post-stimulus as measured by activity that exceeded the 99% confidence interval (dotted line) for spontaneous firing rate (mean probability = 0.003). The cell exhibited a 0.48 probability of discharge during the 22-msec response window and reached the maximum discharge rate at 7 msec. Spontaneous firing rate was calculated for 100 msec and indicated by the solid bar above the histogram.

termination of the evoked response. A 3-msec minimum onset latency of the window is imposed to eliminate any potential artifact created by the whisker-pad stimulator. The window defined during control conditions is subsequently applied to histograms generated after drug administration. During each experimental condition the probability of an action potential occurrence within the response window is then calculated by a second Matlab function, RESPONSE. RESPONSE additionally calculates the average spontaneous firing rate (for 100 msec preceding the onset of the whisker stimulation) and the statistical mode of the evoked response, which is used to determine the latency to peak discharge. Figure 8.6 illustrates the result of the calculations performed by EPOCH and RESPONSE on a sample PSTH from a VPM thalamic cell. In this case, the time window generated by EPOCH was 22 msec in duration that began at 4 msec following whisker stimulation and terminated at 26 msec. RESPONSE calculated that the neuron would discharge an action potential

2345_frame_MASTER.book Page 226 Wednesday, October 23, 2002 1:03 PM

226

Methods in Drug Abuse Research: Cellular and Circuit Level Analyses

within a response window with a probability of 0.48. Furthermore, spontaneous discharge was determined to occur with a probability of 0.003 during the 100 msec preceding the whisker stimulus presentation. Additionally, the latency to peak discharge (statistical mode) occurs 7 msec after the whisker stimulus presentation. The trial-to-trial variability of neuronal discharge evoked by whisker-pad stimulation is calculated by quantifying peristimulus time rasters for each experimental condition. The Matlab function TRIAL calculates the average discharge rate and variability for each trial during the response window generated by EPOCH in a similar manner to RESPONSE. The function TRIAL also calculates additional descriptors of the trial-by-trial response including the statistical mean, median, and standard deviation of the latency of the first spike of the response to whisker pad stimulation. Additionally, TRIAL reports both the probability that a given neuron discharges following stimulus presentation and the average number of spikes generated by a stimulus presentation. These measures provide a means to determine the stability of neuronal responsiveness over control conditions and to determine whether drug administration significantly altered the response discharge rate on a cell-bycell basis. Moreover, these measures were used to determine if drug administration altered the temporal properties of the sensory-evoked response. Although other methods have been proposed to determine when a neuronal response begins and terminates,3,30,47 we have developed these analysis measures to quantify changes in the structure of stimulus-evoked response during and after the administration of drugs of abuse. The effects of cocaine on sensory signal processing and perception are likely mediated by interactions with the central noradrenergic and serotonergic systems. These monoaminergic pathways provide substantial innervation to sensory pathways in the brain and are implicated in the regulation of neuronal responsiveness to afferent sensory information.28,50–52 Thus, a predicted outcome of cocaine’s well known ability to elevate central levels of norepinephrine and serotonin would be enhanced monoaminergic function (i.e., modulation of synaptic efficacy) in sensory areas of the brain. An experimental example using these analysis routines to investigate the actions of the locus coeruleus-norepinephrine efferent system on thalamocortical neuron function is shown in Figure 8.7. Neural discharge evoked by whisker-pad stimulation from multiple, single BF cortical and VPM thalamic cells was compared before and after activation of the locus coeruleus-noradrenergic system by analyzing peristimulus time rasters and PSTHs (Figure 8.7A). The peristimulus time rasters and PSTHs were analyzed with the Nex analysis scripts and Matlab functions and stored in Matlab as a three-dimensional matrix for easy access and data manipulation (Figure 8.7B). Tables of single-neuron-discharge properties are transferred from the Matlab data matrix to a spreadsheet (Figure 8.7C). Comparison of these responses demonstrates the ability of the locus coeruleus-norepinephrine system to modulate the probability of neuronal discharge and temporal properties of the response to whisker-pad stimulation. These results indicate that multineuron recording strategies and appropriate analysis methods can describe the simultaneous impact of neuroactive substances on the discharge patterns of individual, functionally related neurons.

2345_frame_MASTER.book Page 227 Wednesday, October 23, 2002 1:03 PM

Drug Actions on Multiple Simultaneously Recorded Neurons

A.

LC Stimulation

si g 004 a

Control

227

–0.1

0

0.1

0.2

0.06 0.04 0.02 0 –0.2

–0.1

0

0.1

0.2

–0.1

0

0.1

0.2

0 0.1 Time (sec)

0.2

si g 007 a

Probability

0.06 0.04 0.02 0 –0.2

0.08

0.08

0.04

0.04 –0.1

0

0.1

0.2

0 –0.2

si g00 9c

0 –0.2

0.08

0.08

0.04 0 –0.2

B.

0.04 –0.1

0 0.1 Time (sec)

0.2

0 –0.2

–0.1

Matlab data matricies Response rate Response latency

Response magnitude Latency to peak discharge Background 1st spike latency mean 1st spike latency median 1st spike latency std. dev.

. Cell 1, Cell 2, Cell 3 . Condition 1 . Condition 2 . Condition 3 . . Etc.

Continued. FIGURE 8.7 Modulatory actions of the locus coeruleus-noradrenergic pathway activation on neuronal responses to whisker-pad stimulation. (A) Peristimulus rasters and corresponding PSTHs are shown for each neuron (row) during control conditions (left column) and during phasic electrical stimulation (30 mA; 10 Hz train of three pulses every 3 sec) of the ipsilateral locus coeruleus (LC, right column). Each dot of the raster represents the occurrence of an action potential; each horizontal line of dots depicts cell activity during one trial. (B) Cell responses to afferent pathway stimulation were quantified by Nex scripts and the Matlab functions EPOCH, TRIAL, and RESPONSE and stored in a three-dimensional matrix (represented by overlapping layers of information). (c) Information about the neural activity throughout the experimental session could be analyzed further in Matlab or transferred to a spreadsheet. Comparison of responses before and in conjunction with phasic LC stimulation reveals a consistent decrease in the latency and variability of the occurrence of the first spike of the response for all three cells. Additionally, the response magnitude (probability of discharge) for the first neuron was suppressed, the second was increased, and the third was unchanged.

2345_frame_MASTER.book Page 228 Wednesday, October 23, 2002 1:03 PM

228

Methods in Drug Abuse Research: Cellular and Circuit Level Analyses

C. Response Magnitude Latency Spontaneous

1st Spike Latency Mean Std. Dev.

Cell 1

Control LC Stim.

0.3095 0.2619

7 7

0.0007 0.0005

10.90952 5.07811 9.37647 4.75809

Cell 2

Control LC Stim.

0.3333 0.381

5 7

0 0

10.05789 4.8608 9.32381 4.01058

Cell 3

Control

0.2619

7

0.0014

10.28

3.1726

FIGURE 8.7 Continued.

8.6.3 POPULATION DATA The representation of sensory information by ensembles of neurons is another parameter of neural function that may be altered by systemically administered compounds. This expectation is based on the theory that neuronal computations are not performed by individual cells but rather are distributed across the discharge patterns of ensembles of neurons.2,6,16 Cross-correlation analysis and factor analysis are two approaches capable of identifying and evaluating changes in these neuron interactions. 8.6.3.1 Functional Connectivity Functional connectivity, derived from patterns of synchronous or near-synchronous neuronal discharge, can be extracted from a neural population by several methods.1,14,15,32 However, we have chosen the technique of cross-correlation analysis to describe the functional connectivity of assemblies of neurons. The structure of functional connections derived from spike-train correlations describes the functional relationships of neurons as signals travel within a network of cells but may not represent direct anatomical or synaptic connections between cells. This technique, developed by Gerstein and colleagues,22,35 has been used to demonstrate changes in functional connectivity that are dependent on pharmacological and physiological manipulations.1,2,14,42 In our studies, cross-correlation histograms (CCH) are generated for each pair of recorded neurons for both periods of spontaneous activity or periods in which the cells responded to whisker-pad stimulation. The CCHs generated contain appropriate corrections for stimulus-induced activity effects on the CCHs to permit accurate evaluation of the underlying correlation between the spike-train activity of the neuron pair. Issues concerning the usage of appropriate CCH corrections (i.e., poststimulus time predictor) can be found in several papers.14,15 Nevertheless, among BF cortical and VPM neuronal populations, CCHs generated during spontaneous or stimulus-evoked activity are examined for significant (p < 0.001) differences before and after drug administration. To test for differences in spike-train correlations between different experimental conditions, CCHs were transferred to Matlab and compared using the function XcorrStat. XcorrStat performs a two-sample (before and after drug administration) Kolmogorov-Smirnov test45 for each of the neuron pairs. Based upon the shape of the CCH this nonparametric statistic determines whether the cells exhibit significant changes in spike-train correlations after drug

2345_frame_MASTER.book Page 229 Wednesday, October 23, 2002 1:03 PM

Drug Actions on Multiple Simultaneously Recorded Neurons

229

administration. Therefore, experimental manipulation effects on the CCH intensity or changes in the time at which the CCH peak occurs can be detected and represent putative changes in the functional connectivity structure between recorded neurons in a neural circuit. 8.6.3.2 Factor Analysis Factors underlying a large number of variables (multiple simultaneously recorded spike trains) can be extracted by multivariate statistical procedures such as principal component (PC) analysis. This technique has been used to describe the function of many neuronal circuits including those in sensory systems.6,24,36,37,54 A detailed description and mathematical underpinnings of principal component analysis for use in neurophysiological data has been described elsewhere.4,6,23 Briefly, a correlation matrix is generated from the firing rates of simultaneous recordings of individual VPM thalamic or BF cortical neurons. This matrix is used to generate the PCs that are orthogonal and rotated; thus, each PC reflects condensed and independent representations of the information carried among the collective firing patterns of the recorded population of neurons. More simply stated, sensory signal properties represented within the collective firing patterns of ensembles of neurons can be described with PC analysis. For example, information about the direction of movement of a rat’s vibrissae is represented by PCs derived from neural activity of the rat VPM thalamus.6 The PC representation of neuronal activity from the ensemble of neurons can be quantified and analyzed like spike-train data from a single neuron. In our studies, PC representations were examined for changes following the administration of drugs of abuse as before with the analysis routines EPOCH, RESPONSE, and TRIAL. Figure 8.8A illustrates the first PC representation of an ensemble of VPM thalamic neurons as a ratemeter before, during, and after systemic administration of incremental doses of cocaine (0.5, 1.0, 2.0 mg/kg I.V.). PSTHs of the first (Figure 8.8B1–3) and second (Figure 8.8C1–3) PC were then generated from specific time periods (i.e., saline, post 1.0 mg/kg, and post 2.0 mg/kg). Inspection of these histograms reveals an increase in the first and second eigenfunction’s response to the whisker stimulation following systemic administration of cocaine. However, the first PC representation is monotonically increased from control levels for both doses of cocaine, whereas the second PC exhibits a magnitude increase only at the highest dosage of cocaine. The first PC, analyzed in Figure 8.8, is a general factor6,23 related to the overall activity of the neurons. However, information about the direction of the rat’s whisker movement in a dorsal to ventral or rostral to caudal direction is represented in the second and third higher order components.6 Therefore, a drug that alters the function of this sensory circuitry should elicit changes in the first as well as higher-order PCs. The increase in the responsiveness of the first PC representation in Figure 8.8B can be interpreted as an increase in the generalized activity of VPM neurons. Furthermore, the second PC only demonstrated increases in the magnitude of response to the whisker stimulation following the highest dose of cocaine administration. The augmentation of the first PC representation without a change in the

2345_frame_MASTER.book Page 230 Wednesday, October 23, 2002 1:03 PM

230

Methods in Drug Abuse Research: Cellular and Circuit Level Analyses

Control

0.5 mg/kg

Saline

1.0 mg/kg

2.0 mg/kg

Recovery

A.

Counts/bin

8 6 4 2 0 0

200

40 0

600

800

1000

1200

3500 Time (sec)

B.

Counts/bin

1.

1.

25

20

15

15

15

10

10

10

5

5 0.1 0.15 Time (sec)

0.2

1.

0 0.05

10

0.2

2.

0 0.05

8

6

6

6

4

4

4

2

2

2

0

0 0.2

-2 0.05

0.1 0.15 Time (sec)

10

8

0.1 0.15 Time (sec)

5500

5 0.1 0.15 Time (sec)

8

-2 0.05

50 00

3.

25

20

10

Counts/bin

2.

25

4500

3.

20

0 0.05

C.

4000

2.

0.2

3.

0 0.1 0.15 Time (sec)

0.2

-2 0.05

0.1 0.15 Time (sec)

0.2

FIGURE 8.8 Rate-meter and peristimulus histograms of the first and second eigenfunctions (principal component) before and during two doses of cocaine administration. (A) The ratemeter of principal component 1 is shown for the first 20 min and 40 min following the initial cocaine injection, where x-axis is accumulated time and the y-axis represents weighted activity of the population of recorded cells (PC1; counts/bin). The magnitude of the function fluctuates over the time course of the experiment but is comparable between the control period and following incremental (0.5, 0.5, 1.0 mg/kg) injections of cocaine I.V. This eigenfunction can also be viewed as a peristimulus time histogram (PSTH) for different experimental conditions. The PSTH in (B). illustrates the first principal component representation of whisker stimulation during the control period (B1), after 1.0 mg/kg accumulative dosage of cocaine (B2) and after 2.0 mg/kg cocaine. The series of PSTHs in C illustrates the second principal component during the same experimental periods.

second PC suggests that at low doses of cocaine the overall responsiveness of thalamic neurons may be increased, but information about movement direction may be unaltered. Furthermore, at the higher doses (i.e., 2.0 mg/kg) of cocaine the overall activity of the thalamic neurons were not increased from that observed following lower doses of cocaine, but the representation of whisker movement by the population of VPM neurons is augmented. Thus, the differential effects of cocaine on the first and second PCs suggest that this compound has the potential to increase the neural representation of the direction of whisker movement independent of its action on global neuronal activity.

2345_frame_MASTER.book Page 231 Wednesday, October 23, 2002 1:03 PM

Drug Actions on Multiple Simultaneously Recorded Neurons

231

8.7 DISCUSSION This chapter describes methods to determine and quantify the impact of psychoactive substances on single-cell, local-circuit, and neural network sensory signal processing in the awake, freely moving animal. The reliable, physiologic-like activation of neurons along the ascending trigeminal somatosensory circuitry from a chronic, indwelling stimulating electrode within the whisker pad provides an excellent model system to examine the actions of pharmacological agents on information encoded by the neural network. These investigations into the actions of drug effects on sensory signal processing have prompted us to establish a series of criteria to identify and isolate discharge activity from single neurons during these lengthy experiments. Moreover, standardized quantitative analysis routines have been developed to describe, assess, and interpret the impact of drugs of abuse and other pharmacological agents on the function of large assemblies of individual, simultaneously recorded neurons.

8.7.1 ADVANTAGES

AND

LIMITATIONS

Monitoring drug actions on spike train activity of functionally related ensembles of single cells in the awake, behaving animal provides considerable advantages over conventional serial recordings of single-neuron activity. Data collected in multiunit recording experiments can be analyzed in a traditional manner (i.e., as single independent recordings) or examined as simultaneously recorded effects. Observations from recent single-unit recording studies have emphasized the need to determine how drug effects impact ensembles of simultaneously recorded neurons.39,42,55 Systemic administration of cocaine in the anesthetized rat was shown to either enhance or reduce sensory neuron responses of single cells.39 The reason for these variable results was not clear but could have been due to level of anesthesia, differences between animals, variances in drug metabolism, or any other facet of the experiment that could not be held constant. Replication of this study using multiunit recording strategies demonstrated that ensembles (>20) of simultaneously recorded single cells exhibit the same cell-to-cell variability of sensory responses following cocaine administration.41 These effects were observed from ensemble recordings in several animals indicating that the cell-specific modulation of sensory responses is an inherent property of cocaine’s actions rather than an outcome of day-to-day variations in experimental conditions. Moreover, the observation that cocaine produces variable effects on sensory responses of “homogeneous” ensembles of simultaneously recorded thalamic relay neurons (i.e., cells within the rat VPM thalamus22) emphasizes the importance of multiunit recording as a means of confirming results observed in single-cell recording studies. Another advantage of the multichannel recording approach is that data collected to determine the effects of pharmacological manipulations can be used to address a broad spectrum of questions at the individual neuronal level. These experiments produce large quantities of single-unit data (20 to 40 or more neurons) from a single animal, whereas conventional single-unit recording techniques in intact, anesthetized animals produce approximately 4 to 6 cells per animal. Moreover, neural recordings

2345_frame_MASTER.book Page 232 Wednesday, October 23, 2002 1:03 PM

232

Methods in Drug Abuse Research: Cellular and Circuit Level Analyses

from chronically implanted animals can be collected over many subsequent days or weeks.48 This ability to examine the effects of drugs of abuse or other psychoactive agents over time is of significant advantage for investigating the effects of chronic administration of those compounds on neural function. For example, chronically implanted animals may yield stable neural recordings for days to weeks. In addition to reducing experimental variability and the ability to examine the actions of psychoactive compounds on a number of neurons simultaneously, the effect of these drugs on information contained in neuron discharge patterns may also be examined. Data collected using multichannel recording strategies may be analyzed with traditional single-neuron spike-train analyses (i.e., magnitude/latency/duration of response to synaptic input) or analyses focusing on the contributions of ensembles of neurons to perform neural computations before and after administration of drugs of abuse. These analyses are aimed at interactions among neurons within a network and include the mapping of functional connectivity between cells and describing aspects of information distributed across the activity patterns of those neurons. Mapping functional connections between neurons before and after administration of drugs of abuse or other pharmacological agents provides a significant advantage over single-cell recording experiments. Cross-correlation analysis of singleneuron spike trains can reveal the strength of connections between neurons within well-defined anatomical structures or across multiple sites within a network.1,37,44 Moreover, another advantage of this technique is the ability to reveal important temporal relationships between sets of neurons. In a recent study,10 the activity between VPM thalamic and BF cortical neurons was correlated before and after systemic administration of the a-2A adrenergic agonist clonidine, which suppresses the output of the locus coeruleus-norepinephrine projection system. A thalamocortical flow of information in the vibrissa somatosensory system of the quietly resting but awake rat was replaced with a cortical-thalamic flow following systemic clonidine administration. Our interpretation of these results is incomplete; however, the altered temporal relationships of neuronal activity in thalamus and cortex suggest that mechanisms of neuronal computation are substantially altered by a drug that suppresses noradrenergic synaptic transmission. A further significant advantage of multichannel recording studies is the application of multivariate statistics to describe sensory stimulus properties distributed among ensembles of recorded neurons. The analysis of neuronal spike-train activity by principal component analysis is capable of revealing changes in population information coding strategies following administration of psychoactive compounds.6 In contrast, serial single-cell recordings are limited to describing drug-mediated changes in coding of sensory signal properties by individual neurons. In our view, the relationship between a drug effect on animal behavior and the changes in singleneuron activity is more difficult to correlate than changes in population information coding strategies. Moreover, eigenfunctions generated by PC analysis can be used to sum the activity of many simultaneously recorded neurons, weighted according to their covariance structure. This approach provides us with a tool to examine the effects

2345_frame_MASTER.book Page 233 Wednesday, October 23, 2002 1:03 PM

Drug Actions on Multiple Simultaneously Recorded Neurons

233

of drug administration on ensembles of neurons processing behaviorally relevant information. As such, this approach of summing activity across neurons (i.e., spatially) may be used to examine the actions of drugs of abuse on a trial-by-trial basis. Spatial summation on a trial-by-trial basis can be done within a functional group of cells (i.e., VPM thalamus) or across multiple structures within a neural network (i.e., thalamus and cortex). The idea of spatial summation contrasts traditional approaches, whereby spike-train activity from a single neuron has been summed over multiple stimulus presentations before and after administration of psychoactive substances. Summation of neuronal responses over multiple trials decreases intertrial variability of the neuronal response and has revealed much about the parameters of neuronal function that are subject to alteration by drug actions. However, trial-by-trial analysis of pharmacological actions on a functional ensemble of neurons is a significant advantage over previous serial recording methods since natural stimuli are more likely to present themselves in single instances rather than repetitively. By investigating how drugs of abuse impact network function on a trial-by-trial basis, this method may provide a better representation of the mechanisms underlying drug actions on brain function.

8.7.2 FUTURE DIRECTIONS Strategies to analyze data from multichannel recordings of single neurons responding to sensory stimuli in the awake, behaving animal extend further than those presented here. In contrast to single punctate stimuli utilized in the current studies, a future direction of the current work is to analyze drug effects on network functions responsible for processing and coding complex sensory stimuli. The transition from simple stimulus to more enriched or complex stimuli, such as movement of individual or multiple whiskers in various directions, will be useful to bridge the gap between behavioral and neurobiological actions of drugs of abuse. For example, the study of drug effects on complex sensory events, such as object edges and textures, will provide insight into how contextual cues from these complex sensory stimuli can initiate drug-seeking behaviors in models of craving. Additionally, data collected during movements of multiple whiskers, or even the delivery of a continuous random movement of a single whisker, provide data needed for information theory. In the context of these experiments, information theory38 can quantify the ability of a cell or network to describe information about the whisker stimulus. Increases in neuronal or network information content predict higher-fidelity sensory stimulus representations that may lead to better behavioral performance on sensory perception tasks.

ACKNOWLEDGMENTS The authors would like to thank the organizers of the Workshop on the Analysis of Neural Data sponsored by NIMH for inspiring this work, Dr. John Chapin for his insightful comments during the developments of these methodologies, and L. Andrews, H. Wiggins, and J.J. Rutter for their technical assistance. These experiments were supported by NIH grants MH14602, NS32461, NS34808, and DA05117 to BDW.

2345_frame_MASTER.book Page 234 Wednesday, October 23, 2002 1:03 PM

234

Methods in Drug Abuse Research: Cellular and Circuit Level Analyses

REFERENCES 1. Aertsen, A.M., Gerstein, G.L., Habib, M.K., and Palm, G., Dynamics of neuronal firing correlation: modulation of “effective connectivity,” J. Neurophysiol., 61, 900, 1989. 2. Ahissar, E., Haidarliu, S., and Shulz, D.E., Possible involvement of neuromodulatory systems in cortical Hebbian-like plasticity, J. Physiol., 90, 353, 1996. 3. Brillinger, D.R., A note on the estimation of evoked response, Biol. Cybern., 31, 141, 1978) 4. Chapin, J.K., Population-level analysis of multi-single neuron recording data: multivariate statistical methods, in Methods for Neural Ensemble Recordings, Nicolelis, M.A.L., Ed., CRC Press, Boca Raton, FL, 1999, pp. 193–228. 5. Chapin, J.K. and Lin, C-S., The somatic sensory cortex of rat, in The Neocortex of Rat, Tees, R. and Kolb, B., Eds., Academic Press, San Diego, 1990, pp. 341–380. 6. Chapin, J.K. and Nicolelis, M.A., Principal component analysis of neuronal ensemble activity reveals multidimensional somatosensory representations, J. Neurosci. Methods, 94, 121, 1999. 7. Chapin, J.K. and Woodward, D.J., Modulation of sensory responsiveness of single somatosensory cortical cells during movement and arousal behaviors, Exp. Neurol., 72, 164, 1981. 8. Chiaia, N.L., Rhoades, R.W., Bennett-Clarke, C.A., Fish, S.E., and Killackey, H.P., Thalamic processing of vibrissal information in the rat. I. Afferent input to the medial ventral posterior and posterior nuclei, J. Comp. Neurol,. 314, 201, 1991a. 9. Chiaia, N.L., Rhoades, R.W., Fish, S.E., and Killackey, H.P., Thalamic processing of vibrissal information in the rat. II. Morphological and functional properties of medial ventral posterior nucleus and posterior nucleus neurons, J. Comp. Neurol., 314, 217, 1991b. 10. Devilbiss, D.M. and Waterhouse, B.D., Pharmacological manipulation of locus coeruleus tonic output alters information transfer through forebrain sensory circuits, Soc. Neurosci. Abstr., 26(2), 1762, 2000. 11. Devilbiss, D.M. and Waterhouse, B.D., Determination of neurotransmitter and drug actions on multiple simultaneously recorded neurons within functional networks, J. Neurosci. Methods, (in press). 12. Diamond, M.E., Armstrong-James, M., Budway, M.J., and Ebner, F.F., Somatic sensory responses in the rostral sector of the posterior group (POm) and in the ventral posterior medial nucleus (VPM) of the rat thalamus: dependence on the barrel field cortex, J. Comp. Neurol., 319, 66, 1992. 13. Diamond, M.E., Armstrong-James, M., and Ebner, F.F., Somatic sensory responses in the rostral sector of the posterior group (POm) and in the ventral posterior medial nucleus (VPM) of the rat thalamus, J. Comp. Neurol., 318, 462, 1992b. 14. Eggermont, J.J., Neural interaction in cat primary auditory cortex: dependence on recording depth, electrode separation, and age, J. Neurophysiol., 68, 1216, 1992. 15. Eggermont, J.J., Neural interaction in cat primary auditory cortex II. Effects of sound stimulation, J. Neurophysiol., 71, 246, 1994. 16. Erickson, R.P., Stimulus coding in topographic and nontopographic afferent modalities: on the significance of the activity of individual sensory neurons, Psychol. Rev., 75, 447, 1968. 17. Fanselow, E.E. and Nicolelis, M.A., Behavioral modulation of tactile responses in the rat somatosensory system, J. Neurosci., 19, 7603, 1999.

2345_frame_MASTER.book Page 235 Wednesday, October 23, 2002 1:03 PM

Drug Actions on Multiple Simultaneously Recorded Neurons

235

18. Gutierrez-Noriega, C., Mental alterations produced by coca, in The Coca Leaf and Cocaine Papers, Andrews, G. and Solomon, D., Eds., Harcourt Brace Jovanovich, New York, NY, 1975, pp. 262–263. 19. Harris, K.D., Henze, D.A., Csicsvari, J., Hirase, H., and Buzsaki, G., Accuracy of tetrode spike separation as determined by simultaneous intracellular and extracellular measurements, J. Neurophysiol., 84, 401, 2000. 20. Harris, R.M., Morphology of physiologically identified thalamocortical relay neurons in the rat ventrobasal thalamus, J. Comp. Neurol., 251, 491,1986. 21. Harris, R.M., Axon collaterals in the thalamic reticular nucleus from thalamocortical neurons of the rat ventrobasal thalamus, J. Comp. Neurol,. 258, 397, 1987. 22. Kirkland, K., Crosscorrelation, http://mulab.physiol.upenn.edu/crosscorrelation.html, 2001. 23. Kline, P., An Easy Guide to Factor Analysis, Routledge, London, 1994. 24. Laubach, M., Shuler, M., and Nicolelis, M.A., Independent component analyses for quantifying neuronal ensemble interactions, J. Neurosci. Methods, 94, 141, 1999. 25. Ma, P.M., The barrelettes — architectonic vibrissal representations in the brainstem trigeminal complex of the mouse. I. Normal structural organization, J. Comp. Neurol., 309, 161, 1991. 26. Markowitz, R.S., Devilbiss, D.M., and Waterhouse, B.D., Effects of tonic LC activation on thalamic and cortical neuronal responsiveness to afferent synaptic inputs, Soc. Neurosci. Abstr., 26(2), 1762, 2000. 27. McCormick, D.A., Connors, B.W., Lighthall, J.W., and Prince, D.A., Comparative electrophysiology of pyramidal and sparsely spiny stellate neurons of the neocortex, J. Neurophysiol., 54, 782, 1985. 28. McLean, J. and Waterhouse, B.D., Noradrenergic modulation of cat area 17 neuronal responses to moving visual stimuli, Brain Res., 667, 83, 1994. 29. Miller, J.P., Jacobs, G.A., and Theunissen, F.E., Representation of sensory information in the cricket cercal sensory system. I. Response properties of the primary interneurons, J. Neurophysiol., 66, 1680, 1991. 30. Nawrot, M., Aertsen, A., and Rotter, S., Single-trial estimation of neuronal firing rates: from single-neuron spike trains to population activity, J. Neurosci. Methods, 94, 81, 1999. 31. Nicolelis, M.A., Baccala, L.A., Lin, R.C., and Chapin, J.K., Sensorimotor encoding by synchronous neural ensemble activity at multiple levels of the somatosensory system, Science, 268, 1353, 1995. 32. Nicolelis, M.A.L., Methods for neural ensemble recordings, in Methods in Neuroscience Series, Simon, S.A. and Nicolelis, M.A.L., Eds., CRC Press, Boca Raton, FL, 1999, p. 257. 33. Nicolelis, M.A.L. and Chapin, J.K., Spatiotemporal structure of somatosensory responses of many-neuron ensembles in the rat ventral posterior medial nucleus of the thalamus, J. Neurosci., 14, 3511, 1994. 34. Nicolelis, M.A.L., Ghazanfar, A.A., Faggin, B.M., Votaw, S., and Oliveira, L.M., Reconstructing the engram: simultaneous, multisite, many single neuron recordings, Neuron., 18, 529, 1997. 35. Perkel, D.H., Gerstein, G.L., and Moore, G.P., Neuronal spike trains and stochastic point processes. II. Simultaneous spike trains, Biophys. J., 7, 419, 1967. 36. Richmond, B.J. and Optican, L.M., Temporal encoding of two-dimensional patterns by single units in primate primary visual cortex. II. Information transmission, J. Neurophysiol, 64, 370, 1990.

2345_frame_MASTER.book Page 236 Wednesday, October 23, 2002 1:03 PM

236

Methods in Drug Abuse Research: Cellular and Circuit Level Analyses

37. Richmond, B.J., Optican, L.M., and Spitzer, H., Temporal encoding of two-dimensional patterns by single units in primate primary visual cortex. I. Stimulus-response relations, J. Neurophysiol,. 64, 351, 1990. 38. Rieke, F., Warland, D., de Ruyter van Steveninck, R., and Bialek, W., Spikes: Exploring the Neural Code, MIT Press, Cambridge, MA., 1997. 39. Rutter, J.J., Devilbiss, D.M., Waterhouse, B.D., and Moxon, K., Systemically administered cocaine and signal transfer through somatosensory cortical circuits: multineuron recording studies, Soc. Neurosci. Abstr., 25(1), 560, 1999. 40. Shipley, M.T., Response characteristics of single units in the rat's trigeminal nuclei to vibrissa displacements, J. Neurophysiol., 37, 73, 1974. 41. Shosaku, A., Kayama, Y., Sumitomo, I., Sugitani, M., and Iwama, K., Analysis of recurrent inhibitory circuit in rat thalamus: neurophysiology of the thalamic reticular nucleus, Prog. Neurobiol., 32, 7, 1989. 42. Shulz, D.E., Cohen, S., Haidarliu, S., and Ahissar, E., Differential effects of acetylcholine on neuronal activity and interactions in the auditory cortex of the guinea-pig, Eur. J. Neurosci., 9, 396, 1997. 43. Simons, D.J., Response properties of vibrissa units in rat SI somatosensory neocortex, J. Neurophysiol., 41, 798, 1978. 44. Simons. D.J., Multi-whisker stimulation and its effects on vibrissa units in rat SmI barrel cortex, Brain Res., 276, 178, 1983. 45. Smallwood, R.H., A two-dimensional Kolmorgov-Smirnov test for binned data, Phys. Med. Biol., 41, 12, 1996. 46. Smith, S.S. and Chapin, J.K., A paradigm for determination of direct drug-induced modulatory alterations in Purkinje cell activity during treadmill locomotion, J. Neurosci. Methods, 21, 335–344, 1987. 47. Szucs, A., Applications of the spike density function in analysis of neuronal firing patterns, J. Neurosci. Methods, 81, 159, 1998. 48. Theunissen, F.E. and Miller, J.P., Representation of sensory information in the cricket cercal sensory system. II. Information theoretic calculation of system accuracy and optimal tuning-curve widths of four primary interneurons, J. Neurophysiol., 66, 1690, 1991. 49. Van der Loos, H., Barreloids in mouse somatosensory thalamus, Neurosci. Lett., 2, 1, 1976. 50. Waterhouse, B.D., Azizi, S.A., Burne, R.A., and Woodward, D.J., Modulation of rat cortical area 17 neuronal responses to moving visual stimuli during norepinephrine and serotonin microiontophoresis, Brain Res., 514, 276, 1990. 51. Waterhouse, B.D., Moises, H.C., and Woodward, D.J., Interaction of serotonin with somatosensory cortical neuronal responses to afferent synaptic inputs and putative neurotransmitters, Brain Res. Bull., 17, 507, 1986. 52. Waterhouse, B.D. and Woodward, D.J., Interaction of norepinephrine with cerebrocortical activity evoked by stimulation of somatosensory afferent pathways in the rat, Exp. Neurol,. 67, 11, 1980. 53. Wheeler, B.C., Automatic discrimination of single units, in Methods for Neural Ensemble Recordings, Nicolelis, M.A.L., Ed., CRC Press, Boca Raton, FL, 1999, pp. 61–77. 54. Wiener, M.C., Oram, M.W., Liu, Z., and Richmond, B.J., Consistency of encoding in monkey visual cortex, J. Neurosci., 21, 8210, 2001.

2345_frame_MASTER.book Page 237 Wednesday, October 23, 2002 1:03 PM

Drug Actions on Multiple Simultaneously Recorded Neurons

237

55. Woodward, D.J., Waterhouse, B.D., Chang, J., Paris, J.M., Rutter, J., and Gould, E., Single and ensemble neuron spike train analysis in studies of drugs of abuse, in Drug Addiction and Its Treatment: Nexus of Neuroscience and Behavior, Johnson, B.A. and Roache, J.D., Eds., Lippincott-Raven, Philadelphia, 1997, pp. 339–363. 56. Woolsey, T.A. and van der Loos, H., The structural organization of layer IV in the somatosensory region (SI) of mouse cerebral cortex. The description of a cortical field composed of discrete cytoarchitectonic units, Brain Res., 17, 205, 1970. 57. Yuste, R. and Simons, D., Barrels in the desert: the Sde Boker workshop on neocortical circuits, Neuron, 19, 231, 1997. 58. Zucker, E. and Welker, W.I., Coding of somatic sensory input by vibrissae neurons in the rat’s trigeminal ganglion, Brain Res., 12, 138, 1969.

2345_frame_MASTER.book Page 238 Wednesday, October 23, 2002 1:03 PM

2345_frame_MASTER.book Page 239 Wednesday, October 23, 2002 1:03 PM

9

Pharmacological Investigations of Neural Mechanisms Underlying Stimulant-Induced Arousal: Involvement of Noradrenergic Systems Craig W. Berridge

CONTENTS 9.1 9.2 9.3 9.4

9.5

Overview .....................................................................................................240 What Is Arousal? ........................................................................................241 Neurochemical Actions of Stimulants Are Dependent on Both the Dose and Identity of the Drug .............................................................242 Approaches for the Study of the Neurobiology of Stimulant-Induced Arousal ........................................................................................................ 243 9.4.1 Lesions ............................................................................................244 9.4.2 Systemically Administered Pharmacological Manipulations .........244 9.4.3 Intratissue Infusions of Drugs to Identify Sites of Action .............245 9.4.4 Neurochemical Measures of Stimulant-Induced Alterations in Rates of Neurotransmitter Release.................................................................246 Procedures for Assessment of the Arousal-Enhancing Actions of Pharmacological Treatments .......................................................................247 9.5.1 EEG-Modulating Actions of Pharmacological Manipulations in Halothane-Anesthetized Animals ...................................................247 9.5.2 EEG/EMG-Based Assessment of Time Spent in Sleep-Wake States ...............................................................................................249 9.5.3 Behavior-Based Assessment of Time Spent in Sleep-Wake States ...............................................................................................250 9.5.4 Assessment of the Wake-Promoting Actions of Intracerebral Infusions ..........................................................................................251

0-8493-2345-2/03/$0.00+$1.50 © 2003 by CRC Press LLC

239

2345_frame_MASTER.book Page 240 Wednesday, October 23, 2002 1:03 PM

240

Methods in Drug Abuse Research: Cellular and Circuit Level Analyses

9.5.5

Assessment of the Wake-Promoting Actions of Systemically Administered Drugs ........................................................................252 9.5.6 Assessment of the Necessity of a Neurotransmitter System for the Maintenance of Alert Waking .............................................252 9.6 Case Study: Involvement of the LC-NA System in AMPH-Induced Arousal ........................................................................................................253 9.6.1 The LC-NA System and Arousal ...................................................253 9.6.1.1 The LC Modulates the Activity State of the Forebrain .................253 9.6.1.2 NE Acts within the Basal Forebrain to Increase Arousal............................................................................. 255 9.6.1.3 NE Is Necessary for the Maintenance of Alert Waking: Synergistic Actions of a1- and b-Receptors ...................258 9.6.2 Does NE Contribute to AMPH-Induced Arousal? ......................... 258 9.6.2.1 AMPH-Induced Increases in Arousal and NE Release Are Highly Correlated ............................................................ 259 9.6.2.2 AMPH Acts within the Basal Forebrain to Increase Arousal............................................................................. 260 9.6.2.3 Involvement of NA b-Receptors in AMPH-Induced Arousal............................................................................. 262 9.7 Summary ..................................................................................................... 263 Reference .............................................................................................................. 264

9.1 OVERVIEW Amphetamine (AMPH)-like stimulants are potently addictive drugs that in addition to their reinforcing properties target a variety of cognitive, affective, and motor processes. The neurobiological mechanisms underlying the reinforcing, locomotor-activating, and stereotypy-inducing actions of AMPH-like stimulants have been studied intensely over the past few decades. This work has provided substantial insight into the neurochemical and neuroanatomical bases of the reinforcing and motor effects of these drugs. The multitude of behavioral effects of AMPH-like stimulants are superimposed upon potent arousal-enhancing actions of these drugs. Given the prominent nature of these arousal-enhancing actions and their contribution to the widespread use and abuse of these drugs it is surprising that little is known regarding the neural mechanisms underlying stimulant-induced arousal. These drugs target dopaminergic, noradrenergic (NA), and, to a varying degree, serotonergic neurotransmission. The early observation that AMPH increased rates of dopamine (DA) and norepinephrine (NE) neurotransmission stimulated intense inquiry into the role of these neurotransmitters in the behavioral actions of these drugs (see Reference 69). Through the use of selective DA receptor agonists and antagonists it was demonstrated that DA is critically involved in the reinforcing,56,77,79,97 locomotor-activating,48,53 and stereotypy-inducing52 effects of AMPHlike stimulants. Subsequent studies demonstrated that the locomotor-activating and

2345_frame_MASTER.book Page 241 Wednesday, October 23, 2002 1:03 PM

Pharmacological Investigations of Neural Mechanisms

241

reinforcing effects of stimulants are dependent on enhanced release of DA in the nucleus accumbens, whereas the stereotypy-inducing effects involve DA release in the ventrolateral striatum.24,31,51,75 The degree to which NE participates in the behavioral effects of stimulants remains unclear, with limited evidence supporting a limited role of NE in the locomotor- and startle-enhancing, but not the stereotypy-inducing or rewarding (except see Reference 25), effects of these agents.26,54,55,64,72,77,84,92 However, it should be noted that recent studies indicate pronounced behavioral consequences of combined NA a1- and b-receptor blockade.15 These effects are substantially more potent than those observed with blockade of either receptor subtype alone or following NA lesions (as reviewed below, lesions are not well suited for the study of NA function). Most previous studies on the role of NE in the behavioral actions of stimulants have not examined the effects of simultaneous blockade of a1/breceptors. These observations raise the possibility that the combined actions of a1/breceptors may contribute to stimulant-induced behavior traditionally viewed as being relatively independent of NE. A large body of work indicates that the locus coeruleus (LC)-NA system exerts arousal-enhancing actions via actions at multiple receptor subtypes located in a variety of brain structures. This, combined with the targeting of NE neurotransmission by AMPH-like stimulants, suggests the hypothesis that NE participates in the arousal-enhancing actions of these drugs. This chapter will review a variety of issues and methodology relevant to the study of the neural substrates of the arousal-enhancing effects of these drugs. Following this, a case study is presented in which work that examines both the extent to which NE modulates arousal and the extent to which NE participates in the arousal-enhancing actions of AMPH is reviewed. Information gained in these studies suggests a strong case can be made that NE participates in the arousal-modulating actions of these drugs. Further, this work provides a guide for future studies examining the participation of other neurotransmitter systems in the arousal-enhancing actions of AMPH-like stimulants, particularly DA.

9.2 WHAT IS AROUSAL? The current study of the neurobiological mechanisms of arousal traces its roots back to the work of Bremer18 and Moruzzi and Magoun.71 This work utilized EEG recordings to document an excitatory drive originating from the brainstem that is essential for the activation of the forebrain. During the subsequent decades a large body of work emerged, one goal of which has been the identification of the subcomponents of this reticular-activating system. Typically, the study of arousal has largely focused on the regulation of waking vs. sleeping. However, one can examine a multitude of behaviors and behavioral processes that are sensitive to alterations in arousal level (e.g., startle, attention). Given this, it is important that when discussing general terms such as arousal that an operational definition of the term be provided. Arousal, as discussed in this chapter, refers to the extent to which an animal is sensitive/responsive to information arising from the environment. Responsivity to sensory stimuli changes greatly across the sleep-wake cycle, with the ability to attend to, process, and respond to information arising from the environment increasing as an animal

2345_frame_MASTER.book Page 242 Wednesday, October 23, 2002 1:03 PM

242

Methods in Drug Abuse Research: Cellular and Circuit Level Analyses

transitions from deep sleep to waking.19,36,74 Within waking, sensitivity to the environment also fluctuates, ranging from relative inattentiveness during epochs of drowsiness to intense sensitivity/responsivity to environmental stimuli (e.g., enhanced startle response27) and decreased ability to focus on a single stimulus (e.g., scanning6). Associated with these changes in arousal state are changes in activity patterns of the forebrain as measured by cortical EEG,91,94 which, in turn, reflects the activity patterns of large populations of cortical neurons. Thus, during non-REM sleep (e.g., slow-wave sleep), cortical EEG is characterized by the presence of large-amplitude, slow-wave activity, reflecting the synchronous and slow burst-type firing of cortical and thalamic neurons.65 During active waking the forebrain is in what is typically referred to as an activated state, characterized by a depolarization of cortical and thalamic neurons and an asynchronous, single-spike mode of firing of cortical and thalamic neurons.65 This neuronal activity pattern is reflected in cortical EEG recordings by the presence of high-frequency, low-amplitude activity (e.g., desynchronized activity91). Within waking, fluctuations in the degree to which an animal attends to the environment are associated with concomitant changes in EEG activity. For example, during self-directed behaviors (such as grooming), lower-frequency and larger-amplitude activity is observed in cortical EEG, whereas when an animal’s attention is directed to the environment (walking, sniffing), higher-frequency and smaller-amplitude activity is observed (for review, see Reference 94). In humans, vigilance, a measure of sustained selective attention, is highly correlated with cortical EEG activity patterns: increased-amplitude and slower-frequency EEG activity is associated with decreased performance in tests of vigilance.62 EEG activation, associated with both waking and REM sleep, has been defined as a state of enhanced readiness in which neural systems are prepared for reception of information and primed for rapid response to received information.86–88 At the neuronal level, waking and REM sleep (in which forebrain EEG is activated) are associated with increased excitability of thalamic and neocortical neurons.65,86 Thus, EEG activation reflects enhanced ability of the forebrain to process ascending signals. The most obvious difference between the two sleep–wake states in which EEG activation is observed (waking, REM sleep) is the level of behavioral arousal, or the degree to which the animal is prepared for acquisition of sensory information from the environment. Distinction between EEG measures of REM sleep, slow-wave sleep, quiet waking, and active waking can be made on the basis of electromyographic (EMG) recordings: from REM to active waking progressively larger-amplitude EMG activity is observed, corresponding to progressively greater muscle tone.91 In addition, recording of ponto-geniculo-occipital waves can also be used to facilitate identification of REM sleep (see below).

9.3 NEUROCHEMICAL ACTIONS OF STIMULANTS ARE DEPENDENT ON BOTH THE DOSE AND IDENTITY OF THE DRUG The arousal-enhancing actions of stimulants are observed with all members of this class of drug, and these actions predominate at low doses. Thus, the arousal-enhanc-

2345_frame_MASTER.book Page 243 Wednesday, October 23, 2002 1:03 PM

Pharmacological Investigations of Neural Mechanisms

243

ing actions of these drugs should involve at least a subset of neurochemical actions common to low doses of all AMPH-like stimulants. In this regard, it is important to note that the qualitative and quantitative neurochemical actions of these drugs are influenced strongly by (1) identity of the drug and (2) dose of drug (for review see Reference 57). Thus, although AMPH-like stimulants target monoaminergic neurotransmission in general, the extent to which a particular monoamine is targeted can vary greatly across different members of this class of drug. For example, serotonergic systems are relatively insensitive to even moderate-to-high doses (10 to 30 mg/kg, subcutaneously [SC]) of methylphenidate.58,59 This is in contrast to the particularly potent NE-releasing actions of methylphenidate at these and substantially lower doses.59,60 Additionally, there are a variety of dose-dependent effects of these drugs. For example, across a wide range of doses AMPH both stimulates DA efflux from the terminal and inhibits DA reuptake (for review see Reference 57). In contrast, at lower doses (0.5 to 2.0 mg/kg, SC) AMPH acts primarily as an inhibitor of NE reuptake, stimulating NE efflux only at higher doses.39,57 Further, although AMPH increases serotonin efflux, this occurs only at high doses (greater than 2.0 mg/kg). This is in contrast to potent effects of cocaine on serotonin neurotransmission across a range of doses. These observations provide two important pieces of information pertinent to the study of AMPH-like stimulant-induced arousal. First, because substantial qualitative and quantitative differences in molecular action of these drugs can occur with varying dose, choice of dose is an important consideration for experimental design. In the case of AMPH-like stimulant-induced arousal, it is important to use doses at which increased arousal predominates (e.g., lower doses). Second, although different AMPH-like stimulants display varying activity profiles across monoaminergic systems, all of these drugs possess arousal-increasing properties. This suggests that not all monoamines are essential to the arousal-increasing actions of these drugs. Thus, although serotonergic systems likely influence behavioral state,34 enhanced serotonergic neurotransmission is probably not an essential mechanism underlying the arousal-increasing actions of AMPH-like stimulants, given certain AMPH-like stimulants increase arousal but do not appear to target serotonin neurotransmission, particularly at lower doses (e.g., methylphenidate, AMPH).

9.4 APPROACHES FOR THE STUDY OF THE NEUROBIOLOGY OF STIMULANT-INDUCED AROUSAL It is assumed that the arousal-enhancing properties of AMPH-like stimulants involve actions of one or more of the monoamines. Thus, one question of relevance concerns the extent to which the monoamine transmitters exert wake-promoting actions. In examining the extent to which a particular neurotransmitter system exerts wakepromoting actions, there are two general approaches. In the first, the effects of enhancing neurotransmission of a given neurotransmitter can be examined. In the second, the effects of blocking neurotransmission of that transmitter can be examined. However, it is important to keep in mind that each approach addresses funda-

2345_frame_MASTER.book Page 244 Wednesday, October 23, 2002 1:03 PM

244

Methods in Drug Abuse Research: Cellular and Circuit Level Analyses

mentally different questions. In the first, the extent to which a transmitter is sufficient to promote waking is examined. In the second, the extent to which a transmitter is necessary for the maintenance of the waking state is examined. Given that a variety of systems contributes to spontaneous, normal waking, inactivation of any one system may have minimal effects. In contrast, from a baseline state of low-level arousal (e.g., sleep) activation of a single transmitter system may be sufficient to elicit prominent increases in arousal level (e.g., waking). As in most areas of neuroscience, it is important to note that any individual methodological approach will have associated limitations. Understanding the nature and extent of these limitations is essential for accurate interpretation of data collected with a given method. In the study of the neuropharmacology of stimulantinduced behavior, we are limited to a set of methods involving manipulations of monoaminergic neurotransmission. As described above, in the case of AMPH-like stimulant-induced arousal, it seems appropriate to initially focus on the catecholamines (DA, NE).

9.4.1 LESIONS Lesions of dopaminergic and NA axons have been used widely to examine the extent to which AMPH-induced behavior is dependent on either of these transmitter systems. This work has contributed to our understanding of the neurobiology of AMPHinduced behavior. In particular, lesion studies have provided information concerning the extent to which dopaminergic systems contribute to the reinforcing, locomotoractivating, and stereotypy-inducing actions of AMPH. However, the interpretation of lesion studies is problematic due to the well-documented lesion-induced compensatory responses that occur within central catecholaminergic systems. For example, within NA systems compensatory responses are observed at the levels of release, post-synaptic receptor number, and second-messenger systems.32,44,45,61,82,85 In terms of release, substantial reduction of extracellular levels of catecholamines typically does not occur unless tissue levels are reduced by at least 90%.1,2,78 Compensation can occur to such a degree as to produce a system that appears hyperactive relative to the prelesion state.7,30,33,68 Even if a 90% or greater depletion can be obtained, upregulation of post-synaptic receptors and second-messenger systems may negate even substantial reductions in extracellular levels. Thus, failure of lesions to alter stimulant-induced behavior may not reflect a lack of participation of a given neurotransmitter system. Based on these observations, it is concluded that lesions of dopaminergic or NA systems are not the best approach for examination of the extent to which either transmitter participates in stimulant-induced arousal. At the very least, results obtained with lesions need to be confirmed with acute pharmacological blockade of NA or dopaminergic neurotransmission.

9.4.2 SYSTEMICALLY ADMINISTERED PHARMACOLOGICAL MANIPULATIONS The development of catecholamine receptor-selective drugs was a boon to the field of neuroscience, in general, and has provided the majority of our understanding of

2345_frame_MASTER.book Page 245 Wednesday, October 23, 2002 1:03 PM

Pharmacological Investigations of Neural Mechanisms

245

the neuropharmacological bases of AMPH-induced behavior. Nonetheless, it is important to note that no single drug displays absolute specificity for a given receptor subtype. Thus, it is necessary to consider whether a given effect of a single drug is due to the intended action or additional pharmacological activities of the drug. The comparison of the effects of multiple drugs can provide useful information in this regard. In the case of receptor-specific drugs, it is perhaps ideal to compare the effects of an antagonist with those of an agonist. Although neither drug will display absolute selectivity for the receptor of interest, it is unlikely that they will have diametrically opposed actions at other receptors or other cellular targets. Further, when discussing systemic administration peripheral actions of a drug can have profound influence on behavior that, if not taken into account, could lead to erroneous conclusions regarding the behavioral actions of centrally located neurotransmitters. This is particularly true in the case of NE, given the critical role of NE in the regulation of cardiovascular function. Alterations in blood pressure can have a profound effect on cognitive, affective, and motor processes, both nonspecifically, as in the case of diminished blood flow to brain tissue, and via peripherally driven alterations in specific CNS pathways, as in the case of activation of LC neurons in response to decreased blood pressure.93 When using receptor-selective drugs it is also important to keep in mind that most neurotransmitters utilize a variety of postsynaptic receptor subtypes. The selective targeting of a given receptor subtype will not mimic the physiological consequence of alterations in rates of release of a neurotransmitter. For example, early studies examined the effects of blockade of either NA b- or a1-receptors on AMPHinduced stereotypy and locomotor activation. However, given AMPH increases synaptic NE concentrations and thus increases rates of neurotransmission at b-, a1- and a2-receptors, blockade of a single receptor subtype may not be sufficient to prevent NE-dependent behavioral actions of AMPH. Consistent with this are recent observations that indicate synergistic actions of a1- and b-receptors in the modulation of arousal15 (see below) and memory.37

9.4.3 INTRATISSUE INFUSIONS OF ACTION

OF

DRUGS

TO IDENTIFY

SITES

To avoid peripheral actions and to identify sites within the brain involved in stimulant-induced behavior one can use microinfusion techniques. This permits making small infusions of either AMPH-like stimulants or receptor-selective drugs into specific structures. This approach has provided critical insight into both the neurochemistry and site of action underlying stimulant-induced reinforcement, locomotor activation, and stereotypy. For example, infusion of selective DA and NE receptor antagonists and agonists has demonstrated a critical role of DA in these behavioral actions of AMPH-like stimulants. Critical to this approach is the issue of site of action of the infusate. Accurate placement of an infusion needle into a structure of interest is not sufficient to conclude that behavioral effects observed following an infusion result from actions of the drug in that structure. Clearly, the volume and concentration of the infusate are critical variables affecting diffusion of the drug both throughout, as well as beyond, the site of interest. In most cases, smaller is

2345_frame_MASTER.book Page 246 Wednesday, October 23, 2002 1:03 PM

246

Methods in Drug Abuse Research: Cellular and Circuit Level Analyses

better. However, it is essential that sufficiently large volumes are used to provide diffusion throughout a majority of a given structure. The extent to which a drug will diffuse through tissue and extracellular fluid is highly dependent on the chemical structure of the drug. For example, in previous studies we demonstrated that at a concentration of 1 ng/nl, 150 nl infusions (5.0 nmol) of the cholinergic agonist, bethanechol (highly water-soluble), exerted physiologically significant actions on LC neurons (increased discharge rates) when infused within a radius of 500 to 600 mm.8 Infusions beyond this radius did not alter LC discharge rates. Thus, mapping studies indicated that this drug has an effective radius of action, when infused at this concentration and volume, of approximately 500 to 600 mm. In contrast, 150 nl infusions of 1 ng/nl of the a2-agonist clonidine (3.7 nmol), a substantially more lipophilic drug, exerted robust physiological actions on LC neurons when infused at a radius greater than 1000 mm from the recording electrode.9 To achieve an effective radius of action of 500 to 600 mm it was necessary to infuse only 35 nl of the drug at this concentration. Given the extent to which drug diffusion is dependent on the chemical composition, infusion of a dye, which may or may not be structurally similar to the drug being infused, is insufficient for determining radius of action of a drug. Some studies have attempted to assess diffusion of a drug using radiolabeled drug combined with autoradiography. On the surface, this appears to be a reasonable way to measure diffusion of a drug. However, the sensitivity of autoradiography is dependent on the length of time the film is exposed to the tissue. With a sufficiently long exposure period, the sensitivity can far exceed physiologically active concentrations of drug. For example, one study demonstrated that when using autoradiography with a 3-week exposure time a 100-nl infusion of radiolabeled carbachol, a cholinergic agonist, diffused over an area with a radius of approximately 1 mm.43 However, the exposure time used in this study produced a sensitivity capable of detecting radioactivity at a concentration of 1/10,000 that of the infusate. In our studies mentioned above, dilution of drug 1/10,000 would eliminate the physiological actions of the drug (e.g., LC activation/inhibition). Thus, measurement of diffusion of this concentration of drug does not determine the diffusion range over which the drug exerts physiologically significant effects. Therefore, the only accurate method for determination of the radius of action of a locally infused drug appears to be the performance of mapping studies in which the region within which a drug exerts behavioral/physiological actions is explicitly determined.

9.4.4 NEUROCHEMICAL MEASURES OF STIMULANT-INDUCED ALTERATIONS IN RATES OF NEUROTRANSMITTER RELEASE Given the fact that AMPH-like stimulants target monoamine neurotransmission the correlation of stimulant-induced behavior with stimulant-induced alterations in rates of release can provide important information for the assessment of the extent to which a given monoamine may participate in a specific behavior influenced by these drugs. Typically, alterations in monoamine neurotransmission is assessed either by the use of postmortem measures or in vivo microdialysis. In postmortem measures, tissue levels of parent amines and their catabolites are measured. With increased rates of

2345_frame_MASTER.book Page 247 Wednesday, October 23, 2002 1:03 PM

Pharmacological Investigations of Neural Mechanisms

247

release, greater levels of catabolites are produced. Thus, increased tissue levels of catabolites (assessed with between-group comparisons) is indicative of greater rates of release. Although an indirect measure of release, this method provides a good estimation of alterations in rates of release when such alterations are driven by alterations in neuronal discharge activity (e.g., impulse-dependent35,80). However, in the case of AMPH-like stimulants, catabolite or catabolite/amine ratios do not accurately reflect alterations in monoamine release because these drugs bypass normal physiological mechanisms that regulate neurotransmitter release and metabolism (e.g., interfere with reuptake and stimulate release in an impulse-independent manner). The advent of in vivo microdialysis substantially expanded the opportunity for the study of the in vivo neurochemistry of AMPH-like stimulants. In this method, a semipermeable dialysis membrane is used to construct a probe through which artificial extracellular fluid is passed. Artificial extracellular fluid flows in contact with the membrane prior to passing through a small-diameter fused-silica tubing and into sample collection vials (for review see Reference 3). These probes can be made in the laboratory or obtained commercially. Small molecules, including monoamines, travel down a concentration gradient from extracellular fluid into the dialysis probe. The levels of monoamines can be measured, typically through the use of HPLC combined with electrochemical detection (for review see Reference 3). Through the use of this method, the extent to which alterations in extracellular concentrations of DA, NE, and serotonin correlate with AMPH-induced locomotor activity and stereotypy has been characterized extensively. This work has provided strong evidence that despite the necessity of DA for AMPH-induced locomotion and stereotypy, alterations in DA release do not fully explain certain aspects of AMPHinduced locomotion and stereotypy or sensitization of these behaviors upon repeated exposure to these drugs.58 It is important to note that in vivo microdialysis provides only an indirect measure of neurotransmitter release, collecting neurotransmitter that escapes the synaptic cleft and reuptake sites. Further, the temporal resolution of microdialysis is within the minute range and thus exceeds millisecond-to-millisecond alterations in release associated with phasic/brief fluctuations in neuronal discharge rates. However, given the time course of the pharmacological mechanisms underlying stimulant-induced alterations in monoamine release and cognitive and behavioral processes microdialysis provides sufficient temporal resolution to address a variety of questions regarding the neurochemistry of stimulant-induced behavior, including stimulant-induced arousal.

9.5 PROCEDURES FOR ASSESSMENT OF THE AROUSAL-ENHANCING ACTIONS OF PHARMACOLOGICAL TREATMENTS 9.5.1 EEG-MODULATING ACTIONS OF PHARMACOLOGICAL MANIPULATIONS IN HALOTHANE-ANESTHETIZED ANIMALS Potential behavioral state-modulating actions of pharmacological manipulations can be assessed in the halothane-anesthetized rat in which the state of the forebrain is

2345_frame_MASTER.book Page 248 Wednesday, October 23, 2002 1:03 PM

248

Methods in Drug Abuse Research: Cellular and Circuit Level Analyses

measured by EEG recordings. This anesthetized preparation facilitates obtaining a stable baseline EEG from which experimentally induced EEG responses can be assessed. If intratissue infusions are used, this preparation permits the performance of mapping studies using a within-subject design. In this case, an infusion needle can be moved readily from one site to another in the same subject without disturbing baseline EEG measures. The use of inhalant anesthesia permits very fine control over anesthetic level, which facilitates achieving either stable slow-wave or desynchronized (activated) EEG baseline activity patterns. Thus, either sedative or activating effects of manipulations can be examined using this preparation. When combined with electrophysiological recordings of LC NA neurons, this paradigm has permitted the explicit examination of the relationship between LC neuronal discharge activity and forebrain activity state.8,9 We have used this paradigm to demonstrate robust actions of the LC–NA system on forebrain activity state and to explore potential NA terminal fields within which NE and AMPH might act to regulate behavioral state.8,9,11,12 To observe NA-dependent alterations in EEG activity state under these conditions it is essential that the appropriate level of anesthesia be obtained. For example, an EEG-activating influence of LC neurons is observed only when the level of anesthesia is adjusted such that noxious stimuli (e.g., tail pinch) are capable of eliciting a robust EEG activation against a background of synchronized cortical EEG activity.8 At anesthetic levels that prevent tail-pinch-induced EEG activation, NA manipulations do not alter EEG activity patterns. In these studies, animals are placed in a stereotaxic instrument and body temperature maintained at 36 to 38°C with a water-circulating heating blanket. Bipolar surface-to-depth EEG electrodes are implanted into frontal cortical (A +3.0; L ± 1.5) and hippocampal (A –4.8; L ± 2.5; V –2.8) sites (for these studies, the nose bar was positioned at –11.5 below ear-bar zero to facilitate recording from LC). EEG electrodes are manufactured from double-stranded, insulated stainless steel wire (California Fine Wire, Grover Beach, CA), with one wire trimmed 1.5 to 2.0 mm shorter than the other on the proximal end of the electrode (the portion inserted into tissue). Male Amphenol connectors are attached to the distal end of the electrode. An oval-shaped hole is drilled that permits clear visualization of both wires of the electrode during electrode implantation. The electrode is inserted until the shorter wire is touching the surface of the brain and then lowered an additional 100 to 200 mm. Bone wax is used to fill the hole around the electrode, being careful not to move the tissue. The electrode is glued to a jeweler’s screw inserted adjacent to the EEG electrode hole using a two-stage cyanoacrylate-based cement (Loctite, Loctite Corp, Hartford, CT). For intracerebral infusions, 26-gauge guide cannulae (Plastics One, Roanoke, VA) are positioned over areas of interest and cemented into place with dental acrylic. In some cases (e.g., mapping studies) a cannula is cemented to a plastic holder held in a micromanipulator, which facilitates making infusions into multiple sites within a given subject. Infusions are made via a 33-gauge needle, which extends 3 mm beyond the ventral tip of the cannula. This needle is slightly beveled. In our experience, when making small infusions with this gauge tubing, the slight bevel substantially reduces the travel of infusate dorsally along the needle. The infusion needle is attached to PE20 tubing containing water via a 26-gauge stainless steel sleeve glued to the needle. The other end of the tubing is attached to a 10-ml syringe, the

2345_frame_MASTER.book Page 249 Wednesday, October 23, 2002 1:03 PM

Pharmacological Investigations of Neural Mechanisms

249

plunger of which is controlled by a microprocessor-controlled infusion pump (Harvard Apparatus, South Natick, MA). For intratissue infusions, typically 150-nl infusions are performed over a 60-sec period. For intracerebral ventricular (ICV) infusions, infusion volume is typically 2 to 4 ml. An approximately 100-nl air bubble is inserted between the water and infusate (drug). During the infusion the travel of the air bubble is assessed visually, using marks made on the PE20 tubing. With 150-nl intratissue infusions, if no effect of drug or vehicle is observed within 10 min of the first infusion, a second infusion is given. This strategy is based on previous studies utilizing peri-LC infusions to alter LC neuronal discharge activity.8 In these studies, it was observed that with protocols similar to those just described, the first infusion was often less effective than subsequent infusions, presumably due to dilution of drug within the needle tip over time. Typically, we dissolve drug in vehicle containing 0.5 to 2.0% Pontamine Sky Blue dye to visualize the extent to which the infusate diffused radially from the tip of the infusion needle or traveled dorsally within the track created by the infusion needle. In addition to intracerebral infusions, systemic administration of AMPH-like stimulants or other drugs can be performed with either an intravenous catheter14 or subcutaneous or intraperitoneal injections.8 The level of anesthesia can be adjusted to permit either a stably activated EEG (lightly anesthetized preparation9) or a baseline characterized by large-amplitude, low-frequency (e.g., slow-wave) EEG activity. In the latter conditions, a 2-sec tail pinch applied 1 in. from the tip of the tail elicits a robust activation of the EEG that persists for 10 to 120 sec beyond termination of the pinch. Once a stable EEG baseline is obtained, defined as present for at least 20 min, pharmacological treatments can be initiated. EEG signals are amplified, filtered (0.1 to 50.0 Hz bandpass), and recorded on a polygraph and videorecording tape. Power-spectrum analyses (PSA) are later conducted on selected epochs of recorded EEG to better characterize alterations in EEG activity.8,11,12 In our experiments, we select varying-duration EEG epochs (typically 1 to 5 min) from pre-infusion, post-infusion recovery portions of the experiments that are stored on videotape. Each segment is digitized at a sampling frequency of 300 Hz and tapered at the ends as a cosine function. The preinfusion segment is defined as ending immediately prior to the start of the infusion. The post-infusion epoch is selected on the basis of visual inspection as the initial portion of the greatest change in EEG activity from baseline. The recovery period is defined as the point at which EEG activity first appears to have returned to preinfusion activity levels. Each segment is subjected to Fast Fourier Transform and PSA. The mean absolute and mean relative power (percentage of total power) are calculated for the frequency bands 0.3–2.3 Hz, 2.3–6.9 Hz, 6.9–13.0 Hz, 13.0–20.0 Hz, 20.0–30.0 Hz, 30.0–40.0 Hz, and 40.0–50.0 Hz. These frequency bands were selected on the basis of being sensitive to changes induced by sensory stimulation, such as tail pinch.8

9.5.2 EEG/EMG-BASED ASSESSMENT IN SLEEP- WAKE STATES

OF

TIME SPENT

When combined, EEG and EMG recordings permit identification of distinct sleepwake states (e.g., REM sleep, slow-wave sleep, quiet awake, active awake). Bipolar

2345_frame_MASTER.book Page 250 Wednesday, October 23, 2002 1:03 PM

250

Methods in Drug Abuse Research: Cellular and Circuit Level Analyses

surface-to-depth electrodes are cemented into place and used to record cortical and hippocampal EEG, as described in Section 9.5.1. EMG is recorded using two 3-cm lengths of insulated, flexible wire (Cooner Wire, Chatsworth, CA) threaded into the neck muscle and positioned such that an approximately 3-mm length of exposed wire is in direct contact with muscle. The wire is tied in a knot to hold the exposed length of wire in contact with the muscle and the skin sutured. A screw electrode (Plastics One) is placed over the cerebellum and serves as ground. Male Amphenol pins are attached to the free ends of all electrodes and inserted into a 5-pin plastic connector. The 5-pin connector is cemented in place, along with the cannulae and four additional jeweler’s screws to act as anchors, using acrylic cement (Plastics One). Like the description for anesthetized animals, EEG signals are amplified, filtered (0.1 to 50.0 Hz bandpass), and recorded on a polygraph and videorecording tape along with behavioral recordings. In the unanesthetized animal, EEG and EMG recordings are facilitated by the use of a four-channel headstage FET amplifier in addition to standard EEG amplifiers. EEG and EMG are scored for the following behavioral state categories: 1) slow-wave sleep (high-voltage EEG, low-voltage EMG); 2) REM sleep (low-voltage EEG combined with EMG activity of approximately 50% lower amplitude than that observed in slow-wave sleep, with occasional short-duration, large-amplitude deflections due to muscle twitches); 3) quiet-waking (low-voltage EEG with EMG activity of an average amplitude twice that observed in slow-wave sleep); 4) active-waking (low-voltage EEG, sustained high-voltage EMG of approximately twice that observed in quiet waking, with frequent movement deflections). In our studies, to be scored as a distinct epoch, the appropriate EEG and EMG activity patterns needs to persist for a minimum of 15-sec. Typically, we assess time spent in each state for the five 30-min epochs that comprise the observation period. However, this will depend on the goals of a given study. EEG and EMG recordings can be combined with recordings from either pontine, geniculate, or occipital sites to better document episodes of REM sleep.47,70 In some cases, it may be appropriate to perform PSA on the EEG recordings.15 As described in Section 9.5.1, selected preinfusion and post-infusion epochs of the experiment are stored on videotape and later subjected to PSA. Alterations in absolute and relative power of selected frequency bands are calculated.

9.5.3 BEHAVIOR-BASED ASSESSMENT OF TIME SPENT IN SLEEP-WAKE STATES In addition to EEG/EMG recordings, behavioral measures can provide useful information concerning time spent in distinct behavioral states. In addition, these measures can provide information concerning the qualitative nature of drug-induced waking (e.g., whether it resembles spontaneous waking). In our studies, behavior is videotaped using a low-level illumination video camera (Panasonic WV-BL2000) positioned in the hole in the outer chamber door. Behavioral and EEG data are recorded simultaneously onto videorecording tape using a modified VCR (Vetter, Model 620). The output of the camera is sent to a time and date imprinter, a monitor, and the VCR. Behavior is recorded continuously starting 60 to 120 min following insertion of the infusion needle, at a point when the animals are no longer behav-

2345_frame_MASTER.book Page 251 Wednesday, October 23, 2002 1:03 PM

Pharmacological Investigations of Neural Mechanisms

251

iorally active. Behavior is scored from videotape by a trained observer using a computer-based event recorder (Noldus Information, Wageningen, The Netherlands). The following behaviors are typically scored: (1) asleep: body resting on floor, head resting on floor; (2) quiet awake: head raised off floor, body resting on floor; (3) body up: any point in which the body was raised off the floor and the animal was not engaged in any of the scored behaviors (grooming, rears, eating, or drinking) other than horizontal locomotion;( 4) grooming; (5) rears (both free and wall); (6) quadrant entries: a measure of horizontal locomotion defined by hind legs crossing into a new quadrant; (7) eating; (8) drinking; (9) total time spent awake, defined as the total observation period minus the time spent asleep. In our studies, the frequency and duration of all behaviors, except quadrant entries (frequency only), are scored for each of the five consecutive 30-min epochs of the experiment. Our previous studies indicate that when compared to EEG/EMG measures these behavioral measures largely provide an accurate estimation of time spent asleep and awake.11,12 The only exception to this is that these measures consistently underestimate slightly total time spent awake by missing periods of quiet waking characterized by a prone animal with head on the floor but an activated EEG and EMG. Of course, behavioral measures alone cannot be used to assess time spent in REM sleep. Nonetheless, they provide a relatively accurate and easy method to measure time spent awake.

9.5.4 ASSESSMENT OF THE WAKE-PROMOTING ACTIONS INTRACEREBRAL INFUSIONS

OF

To accurately assess wake-promoting actions of a given neurotransmitter system or a given drug it is important that baseline behavioral state be characterized by a lowlevel of arousal (e.g., sleep). Once an animal is in the alert waking state, it is difficult to objectively measure additional increases in arousal. To avoid waking animals at the time of the infusion, we have developed methodology that permits making intracerebral infusions in sleeping rats without the need to handle them at the time of this infusion.10,12 Animals are housed individually in Plexiglas chambers on the night prior to testing. The Plexiglas testing chamber (32 ¥ 32 ¥ 40 cm) is housed in a wooden, sound-attenuated outer chamber containing a 15-W light bulb on a 12–12 hour cycle, a speaker through which white noise (80 dB) is played, and a 12V fan, running at reduced speed, attached to the rear of the chamber. There are two 10-cm holes in the outer chamber: one in the center of the top panel of the chamber to permit infusion lines and EEG cables to exit the chamber, one in the front of the chamber to permit videotaping of the animal. At the time the animal is placed within the testing chamber the stylet, inserted in the cannula at the time of surgery, is removed and a stainless steel coil spring is threaded onto the cannula via a plastic connector (Plastics One). The other end of the spring is attached to a liquid swivel (Instech Laboratories Inc., Plymouth Meeting, PA) held in a counterbalance outside the outer chamber. For animals implanted with EEG/EMG electrodes, the FET headstage and cable are attached at this time and connected to the counterbalance. The animals have free access to food and water. On the following morning, a 33-gauge needle is inserted

2345_frame_MASTER.book Page 252 Wednesday, October 23, 2002 1:03 PM

252

Methods in Drug Abuse Research: Cellular and Circuit Level Analyses

into the cannula, which extends 3 mm beyond the ventral tip of the cannula. This needle is slightly beveled to minimize travel of fluid up the length of the needle. The infusion needle is attached to PE20 tubing. The other end of the tubing is attached to the outlet of a liquid swivel, the inlet of which is connected to a 10 ml syringe. The plunger of the syringe is advanced using a microprocessor-controlled infusion pump (Harvard Apparatus). The entire length of tubing, liquid swivel, and needle is flushed/filled with water. An air bubble is placed in the tubing above the liquid swivel to permit visualization of fluid displacement during advancement of the syringe plunger. The PE20 tubing connected to the infusion needle is contained in a stainless steel coil spring that is tightly secured to the cannula via plastic threaded sleeves. Usually, needle insertion is performed without handling the animal. Following needle insertion the doors to the testing chamber and sound-attenuation chamber are closed and the animal left undisturbed. Recording of behavioral and EEG data is initiated approximately 60 to 120 min following needle insertion, always at a point when the animals are in slow-wave sleep. Infusions are made after collection of 60 min of baseline data. Infusions of 150 nl are made over a 1-min period. The travel of the air bubble is assessed visually, using marks made on the PE20 tubing above the liquid swivel.

9.5.5 ASSESSMENT OF THE WAKE-PROMOTING ACTIONS OF SYSTEMICALLY ADMINISTERED DRUGS The methods described above work well for intracerebral infusions where a cannula exists for attaching a needle enclosed in a stainless steel wire spring. A similar strategy can be employed for systemic administration of AMPH-like stimulants or other drugs. In these cases, tubing would run from a cannula to either a subcutaneous or intraperitoneal site. However, in our previous work with systemic administration of drugs that elicit wake-promoting actions, we have demonstrated that when animals are housed overnight in the testing chamber, as described above, subcutaneous or intraperitoneal injections can be given without the necessity to pick up or otherwise disturb the animal excessively. Under these conditions, vehicle injections result in only brief periods of waking (5 to 15 min) and low-dose AMPH administration results in a substantially longer period of waking (60 to 120 min) than that observed with vehicle treatment.

9.5.6 ASSESSMENT SYSTEM FOR

NECESSITY OF A NEUROTRANSMITTER MAINTENANCE OF ALERT WAKING

OF THE THE

The methodology described above permits examination of wake-promoting actions from a low arousal baseline. These conditions are optimal for determining whether a particular treatment is sufficient to elicit sustained epochs of alert waking. To determine the extent to which a particular neurotransmitter system is necessary for, or contributes to, the maintenance of alert waking, different testing conditions are required. In this case, we want the baseline conditions to be characterized by a moderate amount of spontaneous (non-drug-induced) alert waking. In our laboratory, we achieve this by exposing rats to a novel testing chamber.15 Typically, we house

2345_frame_C09 Page 253 Wednesday, October 23, 2002 3:23 PM

Pharmacological Investigations of Neural Mechanisms

253

animals in the colony room in pairs. Animals receive either a central or systemic injection of a particular drug (for example, an NA receptor antagonist) and are placed back in their home cage. At some point following this injection (typically 20 to 30 min, depending on the drug being tested) the animals are removed from the home cage and placed in the testing chambers described above (Plexiglas chambers contained in a wooden sound-attenuated chamber). The novelty of this chamber is sufficient to sustain alert, active waking for 30 to 60 min.

9.6 CASE STUDY: INVOLVEMENT OF THE LC-NA SYSTEM IN AMPH-INDUCED AROUSAL 9.6.1 THE LC-NA SYSTEM

AND

AROUSAL

9.6.1.1 The LC Modulates the Activity State of the Forebrain The LC is a small nucleus that, via a massive efferent projection system, innervates broad segments of the neuraxis from spinal cord to neocortex. Synaptically released NE produces its effects in LC terminal fields via interactions with three major NA receptor subtypes: a1, a2, and b. a1- and b-receptors are thought to exist primarily postsynaptically, whereas a2-receptors are present both pre- and postsynaptically. This classification system has been expanded greatly with the more recent identification of multiple subtypes each of the b-, a1-, and a2-receptors.17,23,50 Rates of NE release are linearly related to LC neuronal discharge rates across the range of LC discharge rates typically observed across the sleep–wake cycle.13 These observations indicate that relatively small alterations in absolute LC discharge rates within the activity range typically observed in normal sleep and waking result in pronounced alterations in NE levels within the terminal field: increased discharge rates from 1.5 Hz to 3.0 Hz (100%) results in an approximately 100% increase in extracellular levels of NE within cortex. Combined, these observations indicate that the LC–NA system constitutes a system in which alterations in activity of a small number of neurons are broadcast to immense neuronal populations. These characteristics suggest possible global actions of this system such as modulation of behavioral state and state-dependent processes. This hypothesis is buttressed by recordings from LC neurons in unanesthetized animals. LC neurons display two distinct firing modes: tonic and phasic. Phasic discharge is characterized by a brief burst of 2 to 3 action potentials followed by a sustained period of suppression of discharge activity (200 to 500 msec5,40). This activity is typically observed in the awake animal and is driven by novel or salient sensory stimuli. This activity has been linked to alerting/orienting reactions and sustained attention, as determined in tests of vigilance.6 Phasic activity is superimposed on alterations in tonic discharge activity. Tonic discharge activity refers to relatively slow and highly regular state-dependent levels of activity, with highest firing rates observed during waking, very slow rates during slow-wave sleep, and virtually no activity during REM sleep. Of particular relevance is that changes in tonic discharge rates anticipate changes in behavioral state.4,40,46 Within waking, LC neurons discharge most rapidly in anticipation of EEG and behavioral changes that

2345_frame_MASTER.book Page 254 Wednesday, October 23, 2002 1:03 PM

254

Methods in Drug Abuse Research: Cellular and Circuit Level Analyses

ECoG LC TRIGGER LC 200COUNTS 010 SEC

BETHANECHOL 135NG/135NL

10 SECONDS

FIGURE 9.1 Relationship of LC neuronal discharge activity to cortical EEG activity before, during, and after peri-LC bethanechol infusions in the halothane-anesthetized rat. In this experiment, the cholinergic agonist bethanechol was infused 300 to 400 mm lateral to LC, while LC neuronal activity was measured via a multiunit recording. Simultaneously, cortical EEG was recorded from the frontal cortex using a bipolar surface-to-depth recording electrode. EEG activity is shown in the top trace, the raw trigger output from LC activity in the middle trace, and the integrated trigger output (10-sec intervals) in the bottom trace. Bethanechol infusion elicits an increase in LC discharge rate that is readily apparent approximately two thirds of the way through the 60-sec infusion. Several seconds following LC activation an abrupt change in cortical EEG activity is observed as the large-amplitude, low-frequency (slow-wave) activity is replaced by reduced amplitude and increased frequency (EEG activation). As LC activity begins to decrease following the infusion, ECoG amplitude begins to increase, and its frequency decreases. A similar activation of EEG is observed in the hippocampus. Thus, in the halothane-anesthetized rat, increases in LC discharge activity elicit the robust activation of the forebrain, as measured by EEG (see Reference 8).

signal enhanced arousal or attentiveness.4,40 Consistent with these observations is the ability of NE to induce, in vitro, a shift in activity patterns of cortical/thalamic neurons from a burst mode typically associated with slow-wave sleep to a single spike mode typically associated with waking.66 Further, numerous studies demonstrate a modulatory effect of NE on information processing by individual cortical and hippocampal neurons (see References 95 and 96). Combined, these properties and actions of the LC–NA system suggest a causal role for the LC in the regulation of arousal (waking) state. This hypothesis is supported by work in anesthetized rat that demonstrates a causal relationship between alterations in forebrain EEG and alterations in LC neuronal discharge rates.8,9,73 In these studies, electrophysiological recordings were used to guide placement of an infusion needle in close proximity of the LC. Drugs were then infused through this needle to elicit alterations in LC neuronal discharge. Under these conditions, the selective unilateral enhancement of LC neuronal activity results in a robust bilateral activation of forebrain EEG8 (Figure 9.1). The activation of NA b-receptors appears to be essential for LC-dependent EEG activation because EEG activation is blocked by pretreatment with the NA b-antagonist, propranolol.8 Conversely, bilateral inhibition (via infusions of the a2-agonist, clonidine), but not unilateral inhibition, of LC neuronal discharge activity decreases forebrain EEG activation.9 This series of studies demonstrates that minimal LC neuronal discharge activity within one hemisphere is sufficient to bilaterally maintain an activated forebrain. This observation is consistent with the above-described observations indicating substantial levels of NE within cortex even with minimal LC neuronal discharge activity.13 Both observations are consistent with previous observations demonstrating diminished, yet not

2345_frame_MASTER.book Page 255 Wednesday, October 23, 2002 1:03 PM

Pharmacological Investigations of Neural Mechanisms

255

absent, LC discharge activity in animals that are clearly awake yet engaged in consummatory or grooming behavior.4 9.6.1.2 NE Acts within the Basal Forebrain to Increase Arousal The complete array of sites within which NA efferents act to modulate behavioral state remain to be elucidated. Potential sites include cortical, thalamic, basal forebrain, and brainstem regions. Basal forebrain structures implicated in the regulation of cortical and hippocampal activity state include the general region of the basal forebrain encompassing the medial septal area/diagonal band of Broca (MS10,11,20,83), the general region of the anterior medial hypothalamus, encompassing the medial preoptic area (MPOA12,38,63,81), and the substantia innominata (SI22,67). Each of these regions receives a relatively dense NA innervation, the preponderance of which arises from LC and thus could be involved in LC-dependent alterations in forebrain activity state.42,89,98 Results from a series of studies conducted in our laboratory demonstrate potent EEG-activating and wake-promoting actions of NE via actions at both b- and a1receptor subtypes located within MS and MPOA, but not SI. In these studies, we examined the EEG effects of small (150-nl) unilateral infusions of the b-agonist, isoproterenol (15 nmol), or the a1-agonist, phenylephrine (40 nmol), into either MS or MPOA in halothane-anesthetized and sleeping, unanesthetized rat.10,11,16 In the halothane-anesthetized rat, in contrast to that of vehicle, unilateral infusions of either the b-agonist or the a1-agonist elicit a robust and sustained bilateral activation of cortical and hippocampal EEG (Figure 9.2). These EEG responses are observed with a latency of approximately 3 to 8 min. Conversely, bilateral, but not unilateral, infusions of the b-antagonist timolol increase cortical and hippocampal slow-wave activity.11 In the sleeping, unanesthetized rat in which remote-controlled infusions are used to avoid waking of the animal, as described above, infusions of either a b- or a1agonist into MS or MPOA produce a robust and sustained increase in time spent awake10,16 (Figures 9.3 and 9.4). REM sleep is nearly completely suppressed following these infusions (Figure 9.4). In neither the anesthetized nor the unanesthetized animal are EEG-activating/wake-promoting actions observed when infusions are outside MS or MPOA. Both b- and a1-agonist-induced waking resembles spontaneous waking in that behaviors observed were typical of those observed during spontaneous waking, and behaviors not typically observed during spontaneous waking were not observed (e.g., stereotypy, excessive locomotion). SI, situated immediately lateral to MPOA, exerts a potent modulatory influence on cortical activity state as measured by EEG. For example, chemical stimulation of SI elicits activation of cortical EEG.11,67 In contrast, lesions of SI decrease EEG activation within waking.22 To test the hypothesis that NE acts within SI to modulate cortical EEG, the EEG/EMG and behavioral effects of infusions of NE, a b-agonist, and an a1-agonist into SI have been examined in the sleeping, unanesthetized rat. In contrast to the behavioral and EEG effects of infusions of noradrenergic agonists into MS and MPOA, infusions of either NE, a b-agonist, or an a1-agonist into SI do not elicit changes in EEG or behavioral indices of behavioral state11,16 (Figure 9.2). The only exception to this is observed with the highest dose of NE tested, in

2345_frame_MASTER.book Page 256 Wednesday, October 23, 2002 1:03 PM

256

Methods in Drug Abuse Research: Cellular and Circuit Level Analyses

CC

LV

CC LV CC CC

LV

LV

CP

CP

CP

MS MS

CP

AC

NA AC

MS NA AC NA AC

(a) CP LV

CP

LV

CP

LV

I

I

GP

AC ms

I AC

GP

AC

SI

SI SI

(b)

FIGURE 9.2 Schematic diagram depicting sites where 150-nl infusions of the b-agonist isoproterenol were effective or ineffective in eliciting an activation of cortical/hippocampal EEG in the halothane-anesthetized rat. Panel (a) depicts sites at which isoproterenol infusions made within and around the region of MS were either effective (squares) or ineffective (circles) at inducing activation of forebrain EEG. When infusions were placed within a relatively circumscribed region that largely includes portions of the medial septum/diagonal band of Broca (MS), a robust activation of forebrain EEG was observed, similar to that shown in Figure 9.1. Infusions placed immediately outside this region did not alter EEG activity. Panel (b) indicates sites at which isoproterenol infusions (either 150 or 450 nl) were made into the substantia innominata (SI). In no case did isoproterenol infusions into SI elicit EEG activation. AC = anterior commissure; CC = corpus callosum; CP = caudate-putamen; GP = globus pallidus; IC = internal capsule; LS = lateral septum; LV = lateral ventricle; MS = medial septum; NA = nucleus accumbens; SI = substantia innominata. (From Berridge, C.W. et al., J. Neurosci., 16(25), 7010–7020, 1996. With permission.)

which case the magnitude and duration of waking is substantially less than that observed with MPOA infusions.16 Based on these and other observations it is concluded that waking observed following high-dose NE infusion into SI likely results from diffusion of NE into MPOA. In vitro, NE depolarizes cholinergic basal forebrain neurons,41 indicating a neuromodulatory role of NE within SI. Combined with previous observations, the current observations indicate that although NE acts within SI to modulate cortical neuronal activity state,22,67 these actions do not elicit a transition from sleep to waking. To date, the behavioral/cognitive functions of NE within SI remain to be elucidated.

2345_frame_MASTER.book Page 257 Wednesday, October 23, 2002 1:03 PM

Pharmacological Investigations of Neural Mechanisms

257

1-min

PRE-INFUSION ECoG

EMG

POST-INFUSION ECoG

EMG

FIGURE 9.3 Effects of isoproterenol infusions into the general region of the medial septal area (MS) on cortical EEG and EMG. Shown are 10-min traces of cortical EEG and EMG recorded immediately prior to or 15 min following an infusion of isoproterenol into MS. Prior to the infusion the animal spent the majority of time in slow-wave sleep (large-amplitude, slow-wave activity present in cortical EEG and low-amplitude activity present in EMG indicative of minimal muscle tone/activity). The most striking post-infusion changes are the decrease in large-amplitude, slow-wave EEG activity and the increase in EMG amplitude (see Reference 10).

SWS SLEEP

1000 500 0

1500 1000 500 0

PRE1

PRE2

POST1 POST2 POST3

PRE1

REM SLEEP

400

PRE2

POST1 POST2 POST3

ACTIVE AWAKE 1200 SECONDS

SECONDS

TOTAL AWAKE

2000 SECONDS

SECONDS

1500

300 200 100 0

900 600 300 0

PRE1

PRE2

POST1 POST2 POST3

PRE1

PRE2

POST1 POST2 POST3

FIGURE 9.4 Effects of isoproterenol infusions into the medial septal area (MS) on REM sleep, slow-wave sleep, quiet waking, active waking, and total time spent awake (quiet waking + active waking) as defined by EEG and EMG measures. Symbols represent means (± SEM) of time (seconds) spent in the five different behavioral state categories per 30-min epoch. Infusions were made using remote-controlled procedures that avoid awakening the animal. PRE1 and PRE2 represent preinfusion portions of the experiment. POST1-POST3 represent post-infusion epochs. Isoproterenol infusions into MS elicit a large increase in time spent awake and decreases in time spent in slow-wave and REM sleep. The lack of visible error bars indicates that the magnitude of the SEM fell within the range corresponding to the dimensions of the symbol. Vehicle infusions have no apparent effect on time spent awake/asleep. +P < 0.05, *P < 0.01 compared to PRE1 (see Reference 10).

2345_frame_MASTER.book Page 258 Wednesday, October 23, 2002 1:03 PM

258

Methods in Drug Abuse Research: Cellular and Circuit Level Analyses

9.6.1.3 NE Is Necessary for the Maintenance of Alert Waking: Synergistic Actions of a1- and b-Receptors Pharmacological suppression of LC–NA neurotransmission via systemic, ICV, or intrabrainstem administration of a2-agonists elicits substantial increases in behavioral and EEG indices of sedation.25,29 Although these observations are consistent with the hypothesis that NA systems contribute to the maintenance of alert waking, issues of receptor specificity, and potential hypotensive actions of these drugs limit interpretation of these observations. In contrast to these observations, lesions of NA systems have been reported to have minimal effects on EEG and behavioral indices of waking.49,94 However, as described above, it is unlikely that these lesions resulted in a largemagnitude decrease in rates of NE neurotransmission. In the above-described studies with LC inactivation, EEG effects were observed only when complete bilateral suppression of LC discharge activity was obtained. Therefore, given that lesions typically do not result in a complete cessation of NE neurotransmission it is not surprising that lesions have had minimal effects on EEG measures of arousal. In contrast to a2-agonistinduced increases in sedation, blockade of postsynaptic b-receptors does not appear to alter forebrain EEG activity in alert waking.15,94 Further, although blockade of a1receptors increases sleep-spindle activity,21 increased large-amplitude, slow-wave activity is not observed. Thus, blockade of either a1- or b-receptors does not elicit EEG indices of sedation comparable to that observed following a2-agonist administration. The observations described above indicate that stimulation of either a1- or breceptors elicits an activation of forebrain and behavioral activity states. Given this, it was proposed that combined blockade of a1- and b-receptors may be necessary to substantially impair alert waking. In support of this hypothesis, combined administration of a b-antagonist (timolol, ICV) and an a1-antagonist (prazosin, intraperitoneally) results in a profound and dose-dependent increase in large-amplitude, slowwave activity in cortical EEG in animals exposed to an arousal-increasing novel environment15 (Figure 9.5). This increase in slow-wave activity is in contrast to the minimal EEG effects observed following administration of either a b-antagonist alone or the increase in sleep spindles observed following administration of an a1antagonist alone (Figure 9.5). In the anesthetized rat, b-antagonists administered either ICV8 or bilaterally within MS11 increase cortical and hippocampal EEG indices of sedation. Thus, in the unanesthetized animal, a1-receptor blockade more closely mimics the sensitivity of forebrain EEG to NA b-receptor blockade observed in anesthetized animals. It remains to be determined why b-receptor blockade is sufficient to prevent forebrain activation in the presence, but not in the absence, of anesthesia. Possible explanations include lower rates of neurotransmission at a1-receptors during anesthesia compared to waking or lower rates of neurotransmission within other activational systems during anesthesia compared to waking.

9.6.2 DOES NE CONTRIBUTE

TO

AMPH-INDUCED AROUSAL?

The activational influence of the LC–NE system on forebrain activity state combined with the potent excitatory actions of AMPH-like stimulants on NA neurotransmission

2345_frame_MASTER.book Page 259 Wednesday, October 23, 2002 1:03 PM

Pharmacological Investigations of Neural Mechanisms

259

10-sec

VEH/SAL

TIM/SAL

VEH/PRAZ

TIM/PRAZ FIGURE 9.5 Effects of a1-, b-, and combined a1/b-receptor blockade on cortical EEG activity. In these studies, the b-antagonist, timolol, was administered ICV and the a1-antagonist, prazosin, was administered IP. The following four groups of animals were treated 30 min prior to testing: (1) ICV vehicle + IP saline (VEH/SAL), (2) 150 mg ICV timolol + IP saline (TIM/SAL), (3) ICV vehicle + 500 mg/kg prazosin (VEH/PRAZ,), and (4) combined timolol + prazosin (TIM/PRAZ). Shown are EEG traces from the second 5-min epoch of exposure to the novel environment. Vehicle-treated controls displayed behavioral and EEG indices of alert waking throughout much of the recording session. This is reflected in sustained EEG desynchronization (low-amplitude, high-frequency). b-receptor blockade alone (TIM/SAL) had no effects on ECoG activity. a1-receptor blockade alone (VEH/PRAZ) elicited substantial increases in the frequency, amplitude, and duration of sleep spindles (high-voltage spindles). In the presence of a1-receptor blockade, b-receptor blockade elicited substantial increases in large-amplitude, slow-wave activity (see Reference 15).

suggests a possible role of NE in the arousal-enhancing actions of these drugs. Data from a variety of studies support this hypothesis. 9.6.2.1 AMPH-Induced Increases in Arousal and NE Release Are Highly Correlated In recently completed studies, we examined the relationship between AMPH-induced increases in extracellular NE levels within PFC and increases in behavioral and EEG/EMG indices of arousal to better assess the extent to which NE might contribute to AMPH-induced arousal. Animals were injected subcutaneously with vehicle or a low dose of AMPH (0.15, 0.25 mg/kg) and behavioral/EEG, and NE responses were measured. Both doses of AMPH elicited sustained increases in quiet waking relative to controls with only minimal locomotor activation. At these doses robust increases in NE levels occur: an approximately 250% increase in NE at the 0.15 mg/kg dose and a 350% increase at the 0.25 mg/kg dose (Figure 9.6). AMPH-induced increases in time spent awake assessed with EEG/EMG measures are highly correlated with AMPH-induced increases in NE levels across the entire duration of AMPH-induced waking (Figure 9.6). These studies demonstrate a robust effect of low-dose AMPH on NE efflux and support the hypothesis that the arousal-enhancing actions of lowdose AMPH derive, in part, from AMPH-induced increases in NE release.

2345_frame_MASTER.book Page 260 Wednesday, October 23, 2002 1:03 PM

Methods in Drug Abuse Research: Cellular and Circuit Level Analyses

NE (pg/20 ul)

5.0 4.0 3.0 2.0 1.0 0.0 -4 -3

-2 -1 1

2

3

4

5

6

Sample Number

(a)

7

8

9

10

Quiet Awake NE

800 700 600 500 400 300 200 100 0

5.0 4.0 3.0 2.0

NE (pg/20 ul)

Vehicle AMPH 0.15 AMPH 0.15

6.0

Time Awake (sec)

260

1.0 0.0 -4 -3

-2 -1

1

2

3

4

5

6

7

8

9

10

Sample Number

(b)

FIGURE 9.6 Panel (a): effects of varying doses of AMPH on extracellular levels of NE within the prefrontal cortex of unanesthetized rat. Shown are NE levels (pg/20 ml) from 15-min dialysate samples before and following a SC injection of either vehicle (open diamonds), 0.15 mg/kg (solid circles), or 0.25 mg/kg (solid squares) AMPH. Four baseline samples were collected prior to drug treatment. Collection of sample +1 began immediately following the injection. AMPH produced a dose-dependent increase in NE levels returning to baseline over the subsequent 60- to 90-min period, depending on dose. Panel (b): relationship between 0.15 mg/kg AMPH-induced increases in extracellular levels of NE (as shown in Panel (a); open boxes) and time spent in quiet waking, determined from EEG/EMG recordings (solid diamonds). It is evident that there is a close relationship between AMPH-induced waking and AMPH-induced increases in NE levels.

9.6.2.2 AMPH Acts within the Basal Forebrain to Increase Arousal Based on the observations described above, it was proposed that the basal forebrain region encompassing MS and MPOA was a site within which AMPH acts to increase waking (via actions of NE at a1- and b-receptors). To assess this hypothesis small unilateral or bilateral infusions of varying doses of AMPH were made into MS or MPOA in both halothane-anesthetized and unanesthetized rat.12 AMPH infusions into these regions elicited a robust activation of forebrain EEG in both anesthetized and unanesthetized animals. In unanesthetized animals, AMPH infusions elicited sustained increases in EEG/EMG and behavioral indices of waking (Figures 9.7 and 9.8). The effective diameter of these infusions was approximately 500 to 600 mm, with infusions placed anterior to MS, posterior to MPOA, or lateral or dorsal to MS/MPOA having no effects on forebrain EEG. Infusions into SI did not alter either behavioral or EEG/EMG indices of behavioral state (Figure 9.7). Thus, the pattern of sites within which AMPH infusions increase waking is identical of that observed with infusions of NA receptor agonists. In both MS and MPOA, the arousal-enhancing actions of AMPH were dosedependent, with larger and more sustained increases in waking observed with higher doses (Figure 9.8). Although unilateral infusions of AMPH elicited bilateral and robust EEG activation and waking, bilateral infusions produced greater behavioral activation. At the lower dose of AMPH tested, AMPH-induced waking appeared to resemble normal waking in that behaviors typically observed in spontaneous waking were observed (e.g., eating, grooming) and behaviors not typically observed during spontaneous waking were not observed (e.g., stereotypy, intense locomotor activity).

2345_frame_MASTER.book Page 261 Wednesday, October 23, 2002 1:03 PM

Pharmacological Investigations of Neural Mechanisms

261

CC LV

CC CC CC

LS

LV CP

CP

LV

CP

CP

LS MS NA

MS

AC

LS NA AC

NA LS

AC

NA AC

CC CC LV

CP

LV CP

LS

CC LS CP

CC

BST

LV

LS

LV

MS

AC

CP

LS

MS

NA

LPO

AC

LPO BST AC

MS

I

MS

MPO

SI

SI

MPO

AC CC

LV CP

CC

LV CP

FX

CC

LV

FX

I

I

V3 FX

BST

BST BST

SI

MPN

LHA

MPN

SI

SI

MPN MPO

GP

AC

AC BST AC

MPO

I

V3 FX

I

LS

LS

LS

CP

CP

LV

CC

MPO

LPO

LPO

SI LPO

CC

LV CP

CC LV

V3

I V3

GP SI

PVH

AHN LHA

I AHN

OPT

SI

AM

LHA OPT

AM

FIGURE 9.7 Schematic depiction of the location of each 3.75-mg AMPH infusion in the unanesthetized rat, with a numeral indicating the magnitude of the effect of each infusion on total time spent awake. Numerals specify the time range (1 = 0 to 400 sec; 2 = 401 to 800 sec; 3 = 801 to 1200 sec; 4 = 1201 to 1600 sec; 5 = 1601 to 2000 sec; 6 > 2000) for total time spent awake during the 60-min post-infusion interval for each animal. Vehicle-treated animals displayed a mean total time awake in this period of 351 ± 46 sec. These infusions identify a region within which AMPH infusions increase waking that encompasses the medial septum, the vertical limb of the diagonal band of Broca, the posterior portions of the shell region of the nucleus accumbens, and the preoptic area of the hypothalamus. This region is identified by the dotted-line border. Infusions outside this region were substantially less effective at increasing waking, including within the substantia innominata. Panels are arranged anterior–posterior with the anterior-most panel shown in the upper left and the posterior-most panel shown in the bottom right position. AC = anterior commissure, BST = bed nucleus of the stria terminalis, CC = corpus callosum, CP = caudate-putamen, GP = globus pallidus, I = internal capsule, LS = lateral septum, LC = lateral ventricle, MS = medial septum, NA = nucleus accumbens, SI = substantia innominata. (From Berridge, C.W. et al., Neuroscience, 93(13), 885–896, 1999. With permission.)

2345_frame_MASTER.book Page 262 Wednesday, October 23, 2002 1:03 PM

262

Methods in Drug Abuse Research: Cellular and Circuit Level Analyses

ACTIVE AWAKE

TOTAL AWAKE-EEG

1750

1750

1500

SECONDS

1500

SECONDS

UNI VEH BI VEH UNI AMPH 3.75 UNI AMPH 15 BI AMPH 15

1250 1000 750

1250 1000 750

500

500

250

250 0

0 PRE1

PRE2

POST1

POST2

PRE1

POST3

TOTAL AWAKE-BEHAVIOR

PRE2

POST1

POST2

POST3

POST2

POST3

BODY UP

1500

1750 1250

1250

SECONDS

SECONDS

1500

1000 750 500

1000 750 500 250

250

0

0 PRE1

PRE2

POST1

POST2

POST3

PRE1

PRE2

POST1

FIGURE 9.8 Effects of 150-nl AMPH infusions into the medial basal forebrain on time spent awake as determined either by EEG/EMG (top row) or behavioral (bottom row) measures. Animals received either unilateral vehicle (UNI VEH), bilateral vehicle (BI VEH), unilateral 3.75-mg AMPH (UNI AMPH 3.75), unilateral 15.0-mg AMPH (UNI AMPH 15), or bilateral 15-mg AMPH (BI AMPH 15) infusions. Symbols represent mean (± SEM) time (seconds) spent in the five different behavioral-state categories per 30-min epoch. PRE1 and PRE2 represent preinfusion portions of the experiment. POST1–POST3 represent post-infusions epochs. The lack of visible error bars indicates that the magnitude of the SEM fell within the range corresponding to the dimensions of the symbol. Top row: total time spent awake (left panel) and time spent in active waking (right panel) as determined by combined EEG/EMG measures. +P < 0.05, *P < 0.01 compared to PRE1. Bottom row: total time spent awake or with body up (a subset of active waking) as determined from videotaped records of behavior. +P < 0.05, *P < 0.01 compared to PRE1 (see Reference 12).

With bilateral infusions of the highest dose tested, substantial increases in locomotor activity were observed. 9.6.2.3 Involvement of NA b-Receptors in AMPH-Induced Arousal As an initial step in testing the hypothesis that NA receptors participate in AMPHinduced arousal, the effects of ICV pretreatment with the b-antagonist, timolol, on intravenous (IV) AMPH-induced EEG activation were examined in halothane-anesthetized rat.14 Animals were implanted with EEG electrodes, a femoral vein catheter, and a 26-gauge cannula aimed at the lateral ventricle. Following collection of baseline EEG data an ICV injection of either vehicle or varying doses of timolol (10, 50, or 100 µg) was made. Thirty minutes after this infusion an IV AMPH (0.15 mg/kg) injection was made. The study showed that in vehicle-pretreated animals, AMPH

2345_frame_MASTER.book Page 263 Wednesday, October 23, 2002 1:03 PM

Pharmacological Investigations of Neural Mechanisms

263

VEHICLE PRE-AMPH

10-sec

POST-AMPH

TIMOLOL PRE-AMPH

POST-AMPH

FIGURE 9.9 Effects of ICV vehicle and b-antagonist timolol (100 mg) pretreatment on 0.15 mg/kg IV AMPH-induced cortical EEG activation. Shown are 60-sec traces of EEG recordings taken immediately prior to (PRE-AMPH) or 3-min following (POST-AMPH) AMPH infusion into the femoral vein. Data from two separate cases are shown: vehicle-pretreated (top pair of traces) and timolol-pretreated (bottom pair of traces). AMPH was infused 30 min following ICV infusion of either vehicle or timolol. In vehicle-pretreated animals, AMPH induces a robust activation of ECoG characterized primarily by a decrease in large-amplitude, slowwave activity. In contrast, following ICV timolol pretreatment AMPH does not elicit an observable alteration in ECoG activity patterns. (From Berridge, C.W. and Morris, M.F., Psychopharmacology, 148(3), 307–313, 2000. With permission.)

elicits a potent activation of cortical and hippocampal EEG within 120 sec of the infusion. Timolol pretreatment dose-dependently blocks/attenuates AMPH-induced EEG activation14 (Figures 9.9 and 9.10). Thus, in the anesthetized animal, blockade of central b-receptors is sufficient to block the EEG-activating effects of systemic AMPH. The effects of b-antagonist on AMPH-induced arousal in the unanesthetized animal remains to be examined. However, based on the above-described observations it is predicted that combined blockade of a1- and b-receptors is likely to be necessary to attenuate AMPH-induced arousal in the unanesthetized animal.

9.7 SUMMARY This chapter reviews issues and methods associated with the study of the arousalenhancing actions of AMPH-like stimulants. Paramount to the study of arousal is an understanding of the limitations of a given methodology and the careful design of experimental conditions to permit the appropriate baseline level of arousal from which the effects of pharmacological manipulations are assessed. Work to date suggests that AMPH-induced increases in NA neurotransmission likely contribute to the arousal-enhancing properties of these drugs. Further, at least some of the arousal-enhancing properties of AMPH-like stimulants involve actions of these drugs

2345_frame_MASTER.book Page 264 Wednesday, October 23, 2002 1:03 PM

264

Methods in Drug Abuse Research: Cellular and Circuit Level Analyses

140

% Pre-Infusion

120

Vehicle Timolol 10 Timolol 50 Timolol 100

100 80 60 40 20 0 0.3–2.3

2.3–7.5

7.5–20.0

20.0–30.0

30.0–40.0

Frequency (Hz)

FIGURE 9.10 Effects of varying dose of b-antagonist timolol on AMPH-induced alterations in power of selected frequency bands in cortical EEG as determined by power spectral analyses. Absolute power, expressed as percentage of preinfusion mean (± SEM), is shown for the specified frequency bands. When administered 30 min following ICV vehicle pretreatment, IV AMPH decreased the absolute power of the 0.3- to 2.3- (slow-wave), 2.3- to 7.5-, and 7.5to 20.0-Hz frequency bands. ICV pretreatment with timolol produced a dose-dependent blockade of the AMPH induced decreases in absolute power of these frequency bands. *P < 0.05, **P < 0.01 significantly different from preinfusion epochs. (From Berridge, C.W. and Morris, M.F., Psychopharmacology, 148(3), 307–313, 2000. With permission.)

within a circumscribed region of the medial basal forebrain that extends from the anterior medial septal area to the posterior aspects of the medial preoptic area. Future studies will need to assess the extent to which dopaminergic systems also contribute to the arousal-enhancing actions of AMPH-like stimulants and explore the extent to which these drugs act outside the medial basal forebrain to elicit sustained epochs of alert waking.

REFERENCES 1. Abercrombie, E.D. and Zigmond, M.J., Partial injury to central noradrenergic neurons: reduction of tissue norepinephrine content is greater than reduction of extracellular norepinephrine measured by microdialysis, J. Neurosci., 9, 4062, 1989. 2. Abercrombie, E.D., Bonatz, A.E., and Zigmond, M.J., Effects of L-dopa on extracellular dopamine in striatum of normal and 6-hydroxydopamine-treated rats, Brain Res., 525, 36,1990. 3. Abercrombie, E.D. and Finlay, J.M., Monitoring extracellular norepinephrine in brain using in vivo microdialysis and HPLC-EC, in Microdialysis in the Neurosciences, Robinson, T.E. and Justice, J.B., Jr., Eds., Elsevier, Amsterdam, 1991, pp. 253–274. 4. Aston-Jones, G. and Bloom, F.E., Activity of norepinephrine-containing locus coeruleus neurons in behaving rats anticipates fluctuations in the sleep-waking cycle, J. Neurosci., 1:876, 1981a. 5. Aston-Jones, G. and Bloom, F.E., Norepinephrine-containing locus coeruleus neurons in behaving rats exhibit pronounced responses to non-noxious environmental stimuli, J. Neurosci., 1, 887, 1981b.

2345_frame_MASTER.book Page 265 Wednesday, October 23, 2002 1:03 PM

Pharmacological Investigations of Neural Mechanisms

265

6. Aston-Jones, G., Rajkowski, J., Ivanova, S., Usher, M., and Cohen, J., Neuromodulation and cognitive performance: recent studies of noradrenergic locus ceruleus neurons in behaving monkeys, in Catecholamines: Bridging Basic Science with Clinical Medicine, Goldstein, D.S., Eisenhofer, G., and McCarty, R., Eds., Academic Press, San Diego, 1998, pp. 755–759. 7. Berridge, C.W. and Dunn, A.J., DSP-4-induced depletion of brain norepinephrine produces opposite effects on exploratory behavior 3 and 14 days after treatment, Psychopharmacology, 100, 504, 1990. 8. Berridge, C.W. and Foote, S.L., Effects of locus coeruleus activation on electroencephalographic activity in neocortex and hippocampus, J. Neurosci., 11, 3135, 1991. 9. Berridge, C.W., Page, M., Valentino, R.J., and Foote, S.L., Effects of locus coeruleus inactivation on forebrain electroencephalogragaphic activity, Neuroscience, 55, 381, 1993. 10. Berridge, C.W. and Foote, S.L., Enhancement of behavioral electroencephalographic (EEG), and electromyographic (EMG) indices of waking following stimulation of noradrenergic b-receptors located within the medial septal region of the basal forebrain in the unanesthetized rat, J. Neurosci., 16, 6999, 1996. 11. Berridge, C.W., Bolen, S.J., Manley, M.S., and Foote, S.L., Modulation of forebrain electroencephalographic (EEG) activity in the halothane-anesthetized rat via actions of noradrenergic b-receptors located within the medial septal region of the basal forebrain, J. Neurosci., 16, 7010, 1996. 12. Berridge, C.W., O’Neil, J., and Wifler, K., Amphetamine acts within the basal forebrain to initiate and maintain alert waking, Neuroscience, 93, 885, 1999. 13. Berridge, C.W. and Abercrombie, E.D., Relationship between locus coeruleus neuronal discharge rate and rates of norepinephrine efflux in cortex, Neuroscience, 93, 1263, 1999. 14. Berridge, C.W. and Morris, M.F., Amphetamine-induced activation of forebrain EEG is prevented by noradrenergic b-receptor blockade in the halothane-anesthetized rat, Psychopharmacology, 148, 307, 2000. 15. Berridge, C.W. and España, R., Synergistic actions of b- and a1-noradrenergic receptors in the maintenance of alert waking, Neuroscience, 99, 495, 2000. 16. Berridge, C.W. and O’Neill, J., Differential sensitivity to the wake-promoting actions of norepinephrine within the medial preoptic area and the substantia innominata, Behav. Neurosci., 115,165, 2001. 17. Boyajian, C.L. and Leslie, F.M., Pharmacological evidence for alpha-2 adrenoceptor heterogeneity: differential binding properties of [3H]rauwolscine and [3H]idazoxan in rat brain, J. Pharmacol. Exp. Ther., 241, 1092, 1987. 18. Bremer, F., L’activité cérébrale au cours de sommeil et de la narcose. Contribution a l’étude du mécanisme du sommeil, Bull. l’Acad. Royale Belgique, 4, 68, 1937. 19. Bushnell, M.C., Goldberg, M.E., and Robinson, D.L., Behavioral enhancement of visual responses in monkey cerebral cortex, J. Neurophysiol., 46, 755, 1981. 20. Buzsaki, G., Leung, L.W., and Vanderwolf, C.H., Cellular bases of hippocampal EEG in the behaving rat, Brain Res. Rev., 287, 139, 1983. 21. Buzsaki, G., Kennedy, B., Solt, B.V., and Ziegler, M., Noradrenergic control of thalamic oscillation: the role of a-2 receptors, Eur. J. Neurosci., 3, 222, 1991. 22. Buzsaki, G., Bickford, R.G., Ponomareff, G., Thal, L.J., Mandel, R., and Gage, F.H., Nucleus basalis and thalamic control of neocortical activity in the freely moving rat, J. Neurosci., 8, 4007, 1988.

2345_frame_MASTER.book Page 266 Wednesday, October 23, 2002 1:03 PM

266

Methods in Drug Abuse Research: Cellular and Circuit Level Analyses

23. Bylund, D.B., Blaxall, H.S., Iversen, L.J., Caron, M.G., Lefkowitz, R.J., and Lomasney, J.W., Pharmacological characteristics of a-2 adrenergic receptors: comparison of pharmacologically defined subtypes with subtypes identified by molecular cloning, Mol. Pharmacol., 42, 1, 1992. 24. Carr, G.D. and White, N.M., Anatomical disassociation of amphetamine’s rewarding and aversive effects: an intracranial microinjection study, Psychopharmacology, 89, 340, 1986. 25. Correa-Sales, C., Rabin, B.C., and Maze, M.. A hypnotic response to dexmedetomidine, an alpha 2 agonist, is mediated in the locus coeruleus in rats, Anesthesiology, 76, 948, 1992. 26. Creese, I. and Iversen, S.D., The pharmacological and anatomical substrates of the amphetamine response in the rat, Brain Res., 83, 419, 1975. 27. Davis, M., Neurobiology of fear responses: the role of the amygdala, J. Neuropsychiatr. Clin. Neurosci., 9, 382, 1997. 28. Davis, W.M., Smithe, S.G., and Khalsa, J.H., Noradrenergic role in the self-administration of morphine or amphetamine, Pharmacol. Biochem. Behav., 3, 477, 1975. 29. De Sarro, G.B., Ascioti, C., Froio, F., Libri, V., and Nistico, G., Evidence that locus coeruleus is the site where clonidine and drugs acting at alpha-1- and alpha-2adrenoceptors affect sleep and arousal mechanisms, Br. J. Pharmacol., 90, 675, 1987. 30. Diaz, J., Elison, G., and Masouka, D., Stages of recovery from central norepinephrine lesions in enriched and impoverished environments: a behavioral and biochemical study, Exp. Brain Res., 31, 117, 1978. 31. Dickson, P.R., Lang, C.G., Hinton, S.C., and Kelley, A.E., Oral stereotypy induced by amphetamine microinjection into the striatum: an anatomical mapping study, Neuroscience, 61, 81, 1994. 32. Dismukes, R.K., Ghosh, P., Creveling, C.R., and Daly, J.W., Altered responsiveness of adenosine 3',5'-monophosphate-generating systems in rat cortical slices after lesions of the medial forebrain bundle, Exp. Neurol., 49, 725, 1975. 33. Dooley, D.J., Mogilnicka, E., Delini-Stula, A., Waechter, F., Truog, A., and Wood, J., Functional supersensitivity to adrenergic agonists in the rat after DSP-4, a selective noradrenergic neurotoxin, Psychopharmacology, 81, 1, 1983. 34. Dringenberg, H.C. and Vanderwolf, C.H., Involvement of direct and indirect pathways in electrocorticographic activation, Neurosci. Biobehav. Rev., 22, 243, 1998. 35. Elchisak, M.A., Murrin, C.L., Roth, R.H., and Maas, J.W., Free and conjugated dihydroxyphenylacetic acid: effect of alterations in impulse flow in rat neostriatum and frontal cortex, Psychopharmacol. Commun., 2, 411, 1976. 36. Evarts, I.U.Effects of sleep and waking on spontaneous and evoked discharge of single units in visual cortex, Fed. Proc., 19, 828, 1960. 37. Ferry, B., Roozendaal, B., and McGaugh, J.L., Basolateral amygdala noradrenergic influences on memory storage are mediated by an interaction between beta- and alpha1-adrenoceptors, J. Neurosci., 19, 5119, 1999. 38. Findlay, A.L.R. and Hayward, J.N., Spontaneous activity of single neurones in the hypothalamus of rabbits during sleep and waking, J. Physiol., 201, 237, 1969. 39. Florin, S.M., Kuczenski, R., and Segal, D.S., Regional extracellular norepinephrine responses to amphetamine and cocaine and effects of clonidine pretreatment, Brain Res., 654, 53, 1994. 40. Foote, S.L., Aston-Jones, G., and Bloom, F.E., Impulse activity of locus coeruleus neurons in awake rats and monkeys is a function of sensory stimulation and arousal, Proc. Natl. Acad. Sci. U.S.A., 77, 3033, 1980.

2345_frame_MASTER.book Page 267 Wednesday, October 23, 2002 1:03 PM

Pharmacological Investigations of Neural Mechanisms

267

41. Fort, P., Khateb, A., Pegna, A., Muhlethaler, M., and Jones, B.E., Noradrenergic modulation of cholinergic nucleus basalis neurons demonstrated by in vitro pharmacological and immunohistochemical evidence in the guinea pig brain, Eur. J. Neurosci., 7,1502, 1995. 42. Gaspar, P., Berger, B., Alvarez, C., Vigny, A., and Henry, J.P., Catecholaminergic innervation of the septal area in man: immunocytochemical study using TH and DBH antibodies, J. Comp. Neurol., 241, 12, 1985. 43. Gnadt, J.W. and Pegram, G.V., Cholinergic brainstem mechanism of REM sleep in the rat, Brain Res., 384, 29, 1986. 44. Hallman, H., Sundstrom, E., and Jonsson, G., Effects of the noradrenaline neurotoxin DSP-4 on monoamine neurons and their transmitter turnover in rat CNS, J. Neural Transm., 60, 89, 1984. 45. Harik, S.I., Duckrow, R.B., LaManna, J.C., Rosenthal, M., Sharma, V.K., and Banerjee, S.P., Cerebral compensation for chronic noradrenergic denervation induced by locus ceruleus lesion: recovery of receptor binding, isoproterenol-induced adenylate cyclase activity, and oxidative metabolism, J. Neurosci., 1, 641, 1981. 46. Hobson, J.A., McCarley, R.W., and Wyzinski, P.W., Sleep cycle oscillation: reciprocal discharge by two brainstem neuronal groups, Science, 189, 55, 1975. 47. Horner, R.L., Sanford, L.D., Pack, A.I., and Morrison, A.R., Activation of a distinct arousal state immediately after spontaneous awakening from sleep, Brain Res., 778, 127, 1997. 48. Iversen, S.D., Kelly, P.H., Miller, R.J., and Seviour, P.W., Amphetamine and apomorphine responses in the rat after lesion of mesolimbic or striatal dopamine neurones, Br. J. Pharmacol., 54, 224P, 1975. 49. Jones, B.E., Harper, S.T., and Halaris, A.E., Effect of locus coeruleus lesions upon cerebral monoamine content, sleep-wakefulness states and the response to amphetamine in the rat, Brain Res., 124, 473, 1977. 50. Jones, C.R. and Palacios, J.M., Autoradiography of adrenoceptors in rat and human brain: alpha-adrenoceptor and idazoxan binding sites, Prog. Brain Res,. 88, 271, 1991. 51. Kelley, A.E., Gauthier, A.M., and Lang, C.G., Amphetamine microinjections into distinct striatal subregions cause dissociable effects on motor and ingestive behavior, Behav. Brain Res., 35, 27, 1989. 52. Kelly, P.H., Seviour, P.W., and Iversen, S.D., Amphetamine and apomorphine responses in the rat following 6-OHDA lesions of the nucleus accumbens septi and corpus striatum, Brain Res., 94, 507, 1975. 53. Kelly, P.H. and Iversen, S.D., Selective 6-OHDA destruction of mesolimbic dopamine neurons: abolition of psychostimulant-induced locomotor activity in rats, Eur. J. Pharmacol., 40, 45, 1976. 54. Kokkinidis, L. and Anisman, H., Involvement of norepinephrine in startle arousal after acute and chronic d-amphetamine administration, Psychopharmacology, 59, 285, 1978. 55. Kokkinidis, L. and Anisman, H., Circling behavior following systemic d-amphetamine administration: potential noradrenergic and dopaminergic involvement, Psychopharmacology, 64, 45, 1979. 56. Koob, G.F. and Bloom, F.E., Cellular and molecular mechanisms of drug dependence, Science, 242, 715, 1988. 57. Kuczenski, R. and Segal, D.S., Neurochemistry of amphetamine, in Amphetamine and Its Analogues: Psychopharmacology, Toxicology and Abuse, Cho, A.K. and Segal, D.S., Eds., Academic Press, San Diego, 1994, pp. 81–113.

2345_frame_MASTER.book Page 268 Wednesday, October 23, 2002 1:03 PM

268

Methods in Drug Abuse Research: Cellular and Circuit Level Analyses

58. Kuczenski, R., Segal, D.S., Cho, A.K., and Melega, W., Hippocampus norepinephrine, caudate dopamine and serotonin, and behavioral responses to the stereoisomers of amphetamine and methamphetamine, J. Neurosci., 15, 1308, 1995. 59. Kuczenski, R. and Segal, D.S., Effects of methylphenidate on extracellular dopamine, serotonin, and norepinephrine: comparison with amphetamine, J. Neurochem., 68, 2032, 1997. 60. Kuczenski, R. and Segal, D.S., Locomotor effects of acute and repeated threshold doses of amphetamine and methylphenidate: relative roles of dopamine and norepinephrine, J. Pharmacol. Exp. Ther., 296, 876, 2001. 61. Logue, M.P., Growdon, J.H., Coviella, I.L.G., and Wurtman, R.J., Differential effects of DSP-4 administration on regional brain norepinephrine turnover in rats, Life Sci., 37, 403, 1985. 62. Makeig, S. and Inlow, M., Lapses in alertness: coherence of fluctuations in performance and EEG spectrum, Electroencephalogr. Clin. Neurophysiol., 86, 23, 1993. 63. Mallick, B.N. and Alam, M.N., Different types of norepinephrinergic receptors are involved in preoptic area mediated independent modulation of sleep-wakefulness and body temperature, Brain Res., 591, 8, 1992. 64. Mavridis, M., Colpaert, F.C., and Millan, M.J., Differential modulation of (+)-amphetamine-induced rotation in unilateral substantia nigra-lesioned rats by a1 as compared to a2 agonists and antagonists, Brain Res., 562, 216, 1991. 65. McCormick, D.A. and Bal, T., Sleep and arousal: thalamocortical mechanisms, Annu. Rev. Neurosci., 20, 185, 1997. 66. McCormick, D.A., Pape, H.C., and Williamson, A., Actions of norepinephrine in the cerebral cortex and thalamus: implications for function of the central noradrenergic system, Prog. Brain Res., 88, 293, 1991. 67. Metherate, R., Cox, C.L., and Ashe, J.H., Cellular bases of neocortical activation: modulation of neural oscillations by the nucleus basalis and endogenous acetylcholine, J. Neurosci., 12, 4701, 1992. 68. Mogilnicka, E., Increase in b- and a1-adrenoceptor binding sites in the rat brain and in the a1-adrenoceptor functional sensitivity after the DSP-4-induced noradrenergic denervation, Pharmacol. Biochem. Behav., 25, 743, 1986. 69. Moore, K., Amphetamines: biochemical and behavioral actions in animals, in Handbook of Psychopharmacology, Vol. 11, Iversen, L.L., Iversen, S.D., and Snyder, S.H., Eds., Plenum Press, New York, NY, 1978, pp. 41–98. 70. Morrison, A.R. and Bowker, R.M., The biological significance of PGO spikes in the sleeping cat, Acta Neurobiol. Exp., 35, 821, 1975. 71. Moruzzi, G. and Magoun, H.W., Brain stem reticular formation and activation of the EEG, Electroencephalogr. Clin. Neurophysiol., 1, 455, 1949. 72. Ogren, S.O., Archer, T., and Johansson, C., Evidence for a selective brain noradrenergic involvement in the locomotor stimulant effects of amphetamine in the rat, Neurosci. Lett., 43, 327, 1983. 73. Page, M.E., Berridge, C.W., Foote, S.L., and Valentino, R.J., Corticotropin-releasing factor in the locus coeruleus mediates EEG activation associated with hypotensive stress, Neurosci. Lett., 164, 81, 1993. 74. Pfingst, B.E., O’Connor, T.A., and Miller, J.M., Response plasticity of neurons in auditory cortex of the rhesus monkey, Exp. Brain Res., 29, 393, 1977. 75. Pijnenburg, A.J.J., Honig, W.M.M., and van Rossum, J.M., Inhibition of d-amphetamine-induced locomotor activity by injection of haloperidol into the nucleus accumbens of the rat, Psychopharmacologia, 41, 87, 1975.

2345_frame_MASTER.book Page 269 Wednesday, October 23, 2002 1:03 PM

Pharmacological Investigations of Neural Mechanisms

269

76. Randrup, A. and Munkvad, I., Role of catecholamines in the amphetamine excitation response, Nature, 211, 540,1966. 77. Risner, M.E. and Jones, B.E., Role of noradrenergic and dopaminergic processes in amphetamine self-administration, Pharmacol. Biochem. Behav., 5, 477, 1976. 78. Robinson, T.E., Castañeda, E., and Whishaw, I.Q., Compensatory changes in striatal dopamine neurons following recovery from injury induced by 6-OHDA or methamphetamine: a review of evidence from microdialysis studies, Can. J. Psychol., 44, 253, 1990. 79. Rolinski, Z. and Scheel-Kruger, J., The effect of dopamine and noradrenaline antagonists on amphetamine induced locomotor activity in mice and rats, Acta Pharmacol. Toxicol., 33, 385, 1973. 80. Roth, R.H., Murrin, C.L., and Walters, J.R., Central dopaminergic neurons: effects of alterations in impulse flow on the accumulation of dihydroxyphenylacetic acid, Eur. J. Pharmacol., 36, 163, 1976. 81. Sherin, J.E., Shiromani, P.J., McCarley, R.W., and Saper, C.B., Activation of ventrolateral preoptic neurons during sleep, Science, 271, 216, 1996. 82. Skolnick, P., Stalvey, L.P., Daly, J.W., Hoyler, E., and Davis, J.N., Binding of alpha and beta adrenergic ligands to cerebral cortical membranes. Effect of 6-hydroxydopamine treatment and relationship to the responsiveness of cyclic AMP generating systems in two rat strains, Eur. J. Pharmacol., 47, 201, 1978. 83. Smythe, J.W., Christie, B.R., Colom, L.V., Lawson, V.H., and Bland, B.H., Hippocampal theta field activity and theta-on/theta-off cell discharges are controlled by an ascending hypothalamo-septal pathway, J. Neurosci., 11, 2241, 1991. 84. Snoddy, A.M. and Tessel, R.E., Prazosin: effect on psychomotor-stimulant cues and locomotor activity in mice, Eur. J. Pharmacol., 116, 221, 1985. 85. Sporn, J.R., Wolfe, B.B., Harden, T.K., Kendall, T., and Molinoff, P.B., Supersensitivity in rat cerebral cortex: pre-and post-synaptic effects of 6-hydroxydopamine at noradrenergic synapses, Mol. Pharmacol., 13, 1170, 1977. 86. Steriade, M. and Buzsaki, G., Parallel activation of thalamic and cortical neurons by brainstem and basal forebrain cholinergic systems, in Brain Cholinergic Systems, Steriade, D. and Biesold, M., Eds., Oxford University Press, Oxford, 1990, pp. 3–62. 87. Steriade, M. and McCarley, R.W., Brainstem Control of Wakefulness and Sleep, Plenum Press, New York, NY, 1990. 88. Steriade, M., Jones, E.G., and Llinas, R.R., Thalamic Oscillations and Signaling, Wiley-Interscience, New York, NY, 1990. 89. Swanson, L.W. and Hartman, B.K., The central adrenergic system: an immunoflourescence study of the location of cell bodies and their efferent connections in the rat utilizing dopamine-b-hydroxylase as a marker, J. Comp. Neuroanat.. 163, 467, 1975. 90. Swanson, L.W., Brain Maps: Structure of the Rat Brain, Elsevier, Amsterdam, 1992. 91. Timo-Iaria, C., Negrao, N., Schmidek, W.R., Hoshino, K., de Menezes Lobato, C.E., and Da Rocha, T.L., Phases and states of sleep in the rat, Physiol. Behav., 5, 1057, 1970. 92. Tyler, T.D. and Tesel, R.E., Norepinephrine uptake inhibitors as biochemically and behaviorally selective antagonists of the locomotor stimulation induced by indirectly acting sympathomimetic amines in mice, Psychopharmacology, 69, 27, 1980. 93. Valentino, R.J., Page, M.E., and Curtis, A.L., Activation of noradrenergic locus coeruleus neurons by hemodynamic stress is due to local release of corticotropinreleasing factor, Brain Res., 555, 25, 1991. 94. Vanderwolf, C.H. and Robinson, T.E., Reticulo-cortical activity and behavior: a critique of the arousal theory and a new synthesis, Behav. Brain Sci., 4, 459, 1981.

2345_frame_MASTER.book Page 270 Wednesday, October 23, 2002 1:03 PM

270

Methods in Drug Abuse Research: Cellular and Circuit Level Analyses

95. Waterhouse, B.D., Moises, H.C., Yeh, H.H., and Woodward, D.J., Norepinephrine enhancement of inhibitory synaptic mechanisms in cerebellum and cerebral cortex: mediation by beta adrenergic receptors, J. Pharmacol. Exp. Ther., 221, 495, 1982. 96. Waterhouse, B.D., Sessler, F.M., Cheng, J.T., Woodward, D.J., Azizi, S.A., and Moises, H.C., New evidence for a gating action of norepinephrine in central neuronal circuits of mammalian brain, Brain Res. Bull., 21, 425, 1988. 97. Wise, R.A. and Bozarth, M.A., Brain mechanisms of drug reward and euphoria, Psychiatr. Med., 94, 469, 1987. 98. Zaborsky, L., Afferent connections of the forebrain cholinergic projection neurons, with special reference to monoaminergic and peptidergic fibers, in Central Cholinergic Synaptic Transmission, Frotscher, M. and Misgeld, U., Eds., Birkhauser Verlag, Boston, 1989.

2345_frame_IDX Page 271 Tuesday, October 29, 2002 10:58 AM

Index A Accumbal activity chronic extracellular recordings of, 162 firing patterns of, 184 inhibitory effects of cocaine on, 190 in stimulus-reward learning, 188 Accumbal recording studies, 179. See also Chronic extracellular recording Acquisition of drug self-administration. See also Self-administration factors attenuating, 27 studies, 18, 28 Addiction, drug. See also Reinforcement cellular aspects of, 5–6 characteristics of, 1 and cortical abnormalities, 4 genetic predisposition to, 5 incentive motivation theories of, 164–165 neural mechanisms of, 1 stages of, 195 theories of, 2–5 Addictive behavior, neurobiological basis for, 158. See also Behavior Airflyte counterbalance, 197, 200 Alcohol, neuroadaptive changes in brain produced by, 78 Allostasis, principle of, 3 American Association of the Accreditation of Laboratory Animal Care (AAALAC), 58 Amperometric detector, LC–4C, 70 Amperometry, 93. See also High-speed chronoamperometry d-amphetamine and Purkinje neuron responses, 124 voltammetric study of, 102 Amphetamine (AMPH)-like stimulant drugs, 240 arousal-enhancing actions of, 243, 263 arousal induced by, 258–263 basal forebrain and, 260–262 in vivo neurochemistry, 247 NE-dependent behavioral actions of, 245 wake-promoting actions of, 252 Amphetamines behavior reinforcing aspects of, 119, 120 in vivo microdialysis study of, 52–53

monoaminergic effects of, 120 sensory-information processing affected by, 121 Anesthesia halothane, 247–249 for in vivo microdialysis, 58, 59, 64 for jugular catheterization, 202–203 membrane properties influenced by, 8 Anesthetics, topical, 146 Animal models catheterization procedure in, 38–42 in chronic extracellular recording technique, 168 cricket, 215 for drug seeking, 163 halothane-anesthetized, 247–249, 255, 262–263, 264 of human drug addiction, 7 Long-Evans rats, 167 nonhuman primate, 215 rodent trigeminal somatosensory system, 125, 229 voltammetric techniques used with, 100–101 whole-animal preparations, 112–114 Anterior cingulate cortex (ACC), 4 Antibodies, in self-administration studies, 27 Apparatus. See also Instrumentation for in vivo probe-recovery experiment, 56 for self-administration studies, 33–38 Area-under-the-curve (AUC), in microdialysis analysis, 72 Arousal, AMPH-induced involvement of NA ß-receptors in, 262–263, 264 and LC-NA system, 253–263 and role of norepinephrine, 258–259 Arousal, defined, 241 Arousal, stimulant-induced, 240 neurobiology of, 243–247 procedures for assessment of, 247–253 Artificial cerebrospinal fluid (aCSF), 65 Artificial cerebrospinal fluid (aCSF) perfusion medium, 62, 63, 74 Autoinjectors, 72 Autoradiography, to assess diffusion, 244 Axon Instruments, Inc., 95–96

271

2345_frame_IDX Page 272 Tuesday, October 29, 2002 10:58 AM

272

Methods in Drug Abuse Research: Cellular and Circuit Level Analyses

B “Barrel field,” 215 Behavior. See also Addiction cocaine-maintained, 30 experimental analysis of, 25 neuronal activity and, 144 neuropharmacology of, 244, 245 and specific drug actions, 166 Behavior, drug-related, 6 animal models for, 7 mechanisms mediating, 8 patterns of, 179 responding, 20–21 variables affecting, 23 Bethanechol, radius of action for, 246 Biographics, Inc., 147 Brain drug effects on, 8 extracellular fluid (ECF) of, 55 neurochemistry of, 52 in whole animal preparations, 113 Brain slices, voltammetric recording techniques for, 98–100 Brain stress systems, and drug addiction, 3 Break point, in self-administration studies, 32–33 Buprenorphine, as replacement compound, 27 Bupropion, as replacement compound, 27

C Caffeine, in self-administration studies, 23, 24 Calcium, endogenous neurotransmitters affected by, 55 Cannabinoids, voltammetric study of, 104 Cannula, intracranial guide, 60–62 Carbon fiber “working” microelectrodes, 90–91 Catecholamines and AMPH-induced behavior, 244–246 in dialysate sample, 70 lesion-induced responses in, 244 voltammetric study of, 101 Catheter in chronic extracellular recording, 196 in self-administration studies, 35–38 Catheter, in-dwelling jugular, 58–59 post-surgical maintenance of, 60 preparation of, 59 surgical implantation of, 59–60 Catheterization procedure, for rats, 38–42 Cells in culture, voltammetric study of, 97–98. See also Voltammetry Ceramic-based sensors, 91

Cerebellum, effect of psychostimulants on, 122–125 Cerebrospinal fluid, artificial (aCSF), 65 Chem-Clamp, 96 Cholinergic agonist, radius of action for, 246 Chromatogram, in microdialysis analysis, 72 Chronic extracellular recording anatomical resolution of, 164 chronically implanted vs. movable electrodes in, 163–164 drug effect vs. behavioral feedback, 165 incentive motivation theories in, 164–165 i.v. drug self-administration in, 163 of mechanisms of drug action, 166–167 of multiple drug effects, 166 rationale for, 162–163 temporal range of, 164–165 Chronic extracellular recording technique for acute drug effects, 184–189 applications of, 178–179 electrophysiological recording session, 169 firing patterns in, 175–178 histological analyses in, 192–194, 204–205 incentive-related information encoding, 179–184 instrumentation for, 195, 196–202 methods, 167–169 post-operative care for, 203–204 for repeated self-administration, 190–192 research in, 194–195 surgical procedures in, 202–203 utility of, 167 Chronoamperometry, high-speed, 93–94, 99 Clamping procedure in chronic extracellular recording, 186, 187–188 in neurophysiological investigation, 165 Clearance kinetics, 88 Clonidine, neuronal computation altered by, 232 Coatings electrode, 91–92 electropolymerized, 92 Cocaine, 120 alterations in sensory experience with, 132 behavior reinforcing aspects of, 119 differential effects of, 230 drug delivery study of, 26, 27 effect on cortical neurons of, 127–129 effect on dopamine of, 76 effects on neuronal function, 113 effects on thalamic neurons, 129–132 and GABA iontophoretic pulses, 123–124 lethality associated with, 26 neurobiological effects of, 82 perceptual processes affected by, 214–215

2345_frame_IDX Page 273 Tuesday, October 29, 2002 10:58 AM

Index

pharmacological effects of, 189 reinforcing properties of, 25, 30, 31 in self-administration studies, 19, 23, 29–31 sensory cortical neuron responses augmented by, 126 sensory-information processing affected by, 121 Cocaine studies accumbal role in, 189–190 in vivo microdialysis, 52–53 incentive-related information-encoding in, 179–184 systemic administration in, 231 voltammetric, 98, 103 Co-eluting peaks, problem of, 75 Computerization, of microdialysis, 72. See also Software Connectivity in dose-response studies, 6 functional, 228–229, 232 Contingent relationship, in self-administration studies, 25–27 Cortex abnormalities in addicted individuals, 4 effect of cocaine on, 127–129 effects of psychostimulants on, 125 orbitofrontal, 4 Crash phase, of stimulant withdrawal, 78 Cross-correlation analysis, of single-neuron spike trains, 228–229, 232 Cross-correlation histograms (CCH), in multineuron recording system, 228 Crosscorrelograms, event-related, 155 Cyclic voltammetry, 94–95 Cypress Systems, 96 Cytoarchitecture, in dose-response studies, 6

D Dagan Corp., 96 Data, electrophysiological, 169–175. See also Electrophysiological studies Data-acquisition software, for many-neuron recording technique, 147–148. See also Software Data-analysis systems, computerized chromatographic, 72 DataWave Technologies Corp., 169, 170 Delay discounting, 42 Detector, electrochemical, 70 Dialysate sample assaying, 74–75 neurotransmitter concentration in, 72 obtaining, 69

273

recovery in, 55–56 Dialysis membrane action of, 63 properties of, 54 Diffusion drug, 246 microdialysis, 53–55 voltammetric study of, 97–98 3,4-dihydroxyphenylacetic acid (DOPAC), standards for, 73 Discriminant analysis, ensemble neuronal data for, 155 Discrimination procedures, in waveform analysis, 171 Dopamine (DA), 27 and actions of AMPH-like stimulants, 245 and AMPH-induced locomotor activity and stereotypy, 247 detection of, 52, 53 detection of sub-pg amounts of, 70–71 in dialysate sample, 77 in drug addiction, 75 effects of cocaine on, 103 effects of methamphetamine on, 77, 78 effects of stimulant drugs on, 77, 120 electrically evoked release of, 97, 98 impact of AMPH on, 240–241 methamphetamine-evoked release of, 80 in microdialysis, 54 standard curves for, 73 voltammetric measure of, 96, 101 Dopamine transporter (DAT) proteins, 102 in medication of addiction, 79–80 and stimulant drugs, 76 Dopaminergic systems and arousal-enhancing actions of AMPH-like stimulants, 264 effects of chronic d-amphetamine treatment on, 102 Dose, in experimental design, 243 Dose-effect curve, downward turn of, 31 Dose-response curves for cocaine self-administration, 19, 20 for heroin self-administration, 21 and multiple drug effects, 166 Dose-response relationships, cellular experiments in, 6 Drug abuse and liability, 42 over time, 232 voltammetric study of, 101–104 Drug action, receptor mechanisms of, 6 Drug administration, contingent vs. noncontingent, 25–27. See also Selfadministration

2345_frame_IDX Page 274 Tuesday, October 29, 2002 10:58 AM

274

Methods in Drug Abuse Research: Cellular and Circuit Level Analyses

Drug craving, neurobiological basis of, 120 Drug delivery, response-dependent vs. -independent, 26 Drug effects and accumbal information processing, 188–189 vs. behavioral feedback, 165, 185–188 dose-dependent firing rate, 184–185, 186 mediated by nucleus accumbens, 75–76 mirroring changes in drug level, 184, 185, 186 multiple, 166 of repeated self-administration, 190–192 on sensory signal processing, 119 (See also Amphetamine-like stimulant drugs; Stimulant drugs) wake-promoting actions, 251 Drug seeking, 2 animal models in study of, 163 in chronic recording studies, 195 and motivational processing, 188 neural basis of, 212 neural encoding controlling Drugs of abuse, cellular techniques in study of, 8–9

E E-Chempro electrochemical instrument, 96 Economics, behavioral, 42 EI–400, 96 Electrical stimulation effect of cocaine on, 126–127 as reinforcer, 25 Electrochemical detector, in microdialysis, 70, 71. See also High-performance liquid chromatography Electrodes. See also Microelectrodes auxiliary, 93 for many-neuron recording technique, 144–145 reference, 92–93 stimulating, 217–218 used in voltammetry, 90–93 Electroencephalogram (EEG) recordings in arousal studies, 241–242 of drug-related behavior, 9 of forebrain, 247–248 NA-dependent alterations in, 246 of sleep-wake states, 249–250 Electromyogram (EMG) recordings, of sleepwake states, 249–250 Electronic harness, in chronic extracellular recording, 198, 199

Electrophysiological recording, multichannel, 143. See also Many-neuron recording technique Electrophysiological studies equipment for, 201–202 for evaluating psychostimulant actions, 122–132 in intact animals, 113 ISI analysis of, 171–175 specific receptor agonists and antagonists in, 166 waveform analysis, 169–171 Electrophysiological techniques background of, 119 for central monoaminergic systems, 120–121 for sensory-signal processing, 121 Environment and drug-related behavior, 20–21 as reinforcer, 25 Enzymes, as electrode coatings, 92 EPOCH analysis function, 224–227, 229 Ethanol, voltammetric study of, 101 Evoked discharges, drug-induced changes in, 117–118 Extracellular fluid (ECF). See also Dialysate sample continuous sampling of, 52 in microdialysis, 54 neuronal, 55 neurotransmitter concentration in, 57 Extracellular recording, 118

F Factor analysis, in spike train activity, 229–230 FAST–12 instrument, 96 Fast-scan cyclic voltammetry (FCV), 94 Feed-forward cycle, drug addiction as, 3 FEP tubing, 63, 64 Field effect transistors (FETs), 147, 201 Firing patterns and dose-dependent changes in, 184–185, 186 and drug effect vs. behavioral feedback, 185–188 and drug reward expectation, 184 and drug taking, 193 group mean neural data for, 178 individual neuron data in, 175–178 information encoding with lever-press, 180 in ISI analysis, 172, 173 mirroring changes in drug level, 184, 185, 186 misrepresentative, 173 and motivational processing, 188 and multiple drug effects, 166

2345_frame_IDX Page 275 Tuesday, October 29, 2002 10:58 AM

Index

time-locked to cocaine, 179–180, 181 Firing rates, in factor analysis, 227 Fixed-interval (FI) schedule, in drug selfadministration studies, 33 Fixed-ratio (FR) schedules, in self-administration studies, 22, 29–31 Flow cell compartment, in microdialysis, 68–70 Flush procedures, for in-dwelling jugular catheter, 60 Food restriction, in self-administration studies, 23, 25 Forebrain activity state of, 253–255 during arousal, 242, 260–262 EEG measurement of, 247–248 Functional connectivity, in spike train activity, 228–229, 232

275

High-performance liquid chromatography (HPLC), 52, 66 High-performance liquid chromatography with EC detector (HPLC-EC system), 66, 67, 77 for detection of monoamine transmitters, 65 and dialysate sample, 74–75 High-speed chronoamperometry (HSC) recordings, 93–94, 95, 99 Hippocampus, 100 History, behavioral and drug, 21–25 Homovanillic acid (HVA), standards for, 73 Housing chambers, in self-administration studies, 34–35, 37. See also Testing chamber HPLC pumps, 68, 72 5-hydroxyindoleacetic acid (5-HIAA), standards for, 73 Hypothalamo-pituitary-adrenal (HPA) axis, in self-administration studies, 23

G Gain-amplifying effects, of drug addiction, 3 Gamma-amino butyric acid (GABA), 122 cerebellar Purkinje cell response to, 122 in chronic drug administration, 134 iontophoretic delivery of, 123 GBR 12909 (dopamine (DA) uptake blocker), 79–80 GBR-decanoate, 80–82 GBR-hydroxy, 80, 81 Gene Clamp 500, 96 Genetic predisposition, to drug addiction, 5 Genetics, behavioral, 42 L-glutamate, voltammetry study of, 96–97 Glutamate application, effect of cocaine on, 126–127 GMA Technologies, Inc., 96 Guide cannula, for in vivo microdialysis, 60–62

H Halothane-anesthetized animals AMPH-induced EEG activation in, 262–263, 264 isoproterenol infusion of, 255 pharmacological manipulations in, 247–249 Harm reduction, 27 Harness, electronic, 198, 199 Headset, microwire array, 197–198 Hedonic dysregulation theory, 3 experimental approaches in, 5 weakened cortical inhibitory mechanisms in, 4 Heroin, in self-administration studies, 19–20

I Imaging studies, of brain, 5 Implantation of jugular catheter, 58–60 of microelectrode arrays, 145 Impulsivity, models of, 42 In vivo measurements, 163 Incentive motivation theories, of addiction, 2–3, 164–165 Incentive-related information encoding, 179–184 Indoleamines, in dialysate sample, 70 Information, digital, 158. See also Software Information encoding incentive-related, 179 with lever-press patterns, 180 Inhibition, effect of addictive drugs on, 189 Instrumentation for chronic extracellular recording, 195, 196–202 for in vivo microdialysis, 67–70 for in vivo voltammetric recordings, 88, 89 for multineuron recording system, 218, 220–221 for voltammetric studies, 95–96 Interspike-interval (ISI) histograms defined, 171 examples, 172, 174 limitations of, 175 minimum, 171–172 single-neuron data on, 173, 174 Intracerebral infusions, wake-promoting actions of, 251–252 Intracerebral ventricular (ICV) infusions, 249

2345_frame_IDX Page 276 Tuesday, October 29, 2002 10:58 AM

276

Methods in Drug Abuse Research: Cellular and Circuit Level Analyses

Intracranial guide cannula, for in vivo microdialysis, 60–62 Intracranial self-stimulation (ICSS), and withdrawal, 4 Intramural Research Program (IRP), 58 Intratissue infusions, 249 Intravenous drug delivery automatic, 17–18 methodological advantages of, 18 Intravenous-drug self-administration model, 7 ISCO 260D pump, 68, 69 Isoproterenol, EEG effects of, 256, 257

J Jugular catheter, implantation of, 58–60 Jugular catheterization, in chronic extracellular recording, 202

L Learning drug effects of, 193 drug-induced amplification of, 2–3 Lemniscal pathway, 217 Lesion studies, of AMPH-induced behavior, 244 Liability, abuse, 42 Liquid chromatography, high-performance, 52, 66 Locomotion, in chronic extracellular recording, 186, 187 Locus coeruleus (LC), properties of, 253 Locus coeruleus (LC)-NA system and arousal, 253–256 and arousal-enhancing actions, 241 Locus coeruleus-norepinephrine (LC-NE) system and forebrain activity, 258–259 and thalamocortical neuron function, 226, 227, 232 Long-Evans rats, 167

M Maintenance of drug self-administration factors attenuating, 27 studies, 18, 29 Mann-Whitney test, 175 Many-neuron electrophysiology, 144 Many-neuron recording technique advances in, 157 advantages of, 144 chronic implants in, 145–146 electrodes in, 144–145 in vivo, 143

microwire array, 148 procedures, 146–148 stability of, 150, 151 yields from, 149–150 Mapping studies, 246, 248 Marijuana, voltammetric study of, 104 Matlab, 224, 226, 227, 228 MED Associates Inc., 202 Medial preoptic area (MPOA), in sleep/wake studies, 255, 256, 260 Medications, development of, 79–82 Methadone, as replacement compound, 27 Methamphetamine effects on dopamine of, 76, 78 effects on serotonin of, 79 5-HT release induced by, 80 neurotoxicity of, 82 Methohexital (Brevital sodium), for anesthesia, 64 Methylphenidate, NE-releasing actions of, 243 Microdialysis, 7 in alterations in monoamine neurotransmission, 246–247 chemical interaction in, 66 compared with voltammetric technique, 88 methods, 62–66 quantitative methods to, 78 in research, 82 running experiments, 65 testing chamber, 60 voltammetry compared with, 95 Microdialysis, in vivo, 52 advantages of, 58 analytical methods, 66-75 in awake animals, 64-65 basic principles of, 53-58 catheter implantation for, 58-60 data acquisition and analysis in, 71-75 intracranial guide cannula implantation for, 60-62 mobile phase conditions, 70-71 probe/tissue interactions in, 57-58 technique, 63-64 workstation, 67 Microelectrode arrays, chronic implants of, 145146 Microelectrodes. See also Electrodes carbon fiber "working," 90-91 coatings for, 91-92 for voltammetry, 89-90, 91 Microinfusion techniques, 245-246 Microinjection pump, CMA 100, 63 Microiontophoresis, 114-115 Micropressure-ejection techniques, 114 Microwire array headset, 197-198 Microwire electrode arrays, 145

2345_frame_IDX Page 277 Tuesday, October 29, 2002 10:58 AM

Index

Midazolam, drug delivery study of, 27 Millar Voltammeter, 96 Millenium Chromatography Manager, 72 MK-801, 23 Molecular biology, 5, 42, 76 Monoamine mix standard, 73 Monoamine neurotransmitters effects of stimulant drugs on, 120 in vivo microdialysis assessment of, 246-247 in solute recovery, 55 wake-promoting actions of, 243 Morphine, response-dependent administration of, 26. See also Opiates Mothers, studies in offspring of drug-addicted, 9 Motivation, 188 and drug-seeking, 188 opponent process theory of, 3 Multichannel Acquisition Processor (MAP) hardware, 218, 220, 221 Multichannel recording approach advantages of, 231, 232 research in, 233 Multineuron recording system apparatus for, 220 components of, 218, 220-221 experimental session in, 221 multiunit spike train activity in, 224-230 offline validation single-neuron recording in, 222-224 online spike sorting in, 221-222 perievent stimulus histograms in, 219 waveform discrimination in, 221-224 Multiunit recording experiments, 231 single-cell recording studies confirmed by, 231

N Nafion coating, 92, 95, 97 National Institute on Drug Abuse (NIDA), 58 NB Labs, 145 Nervous system, effect of chronic drug addiction on, 2 Neural activity, extracellular recording of, 222 Neural firing, and drug administration, 167. See also Firing patterns Neurobiology of drug abuse, 143 of stimulant-induced arousal, 243-247 Neurochemistry brain, 52 of drug-related behavior, 9 of stimulant drug withdrawal, 78

277

of stimulant drugs, 53 NeuroExplorer software, 148, 224 Neuromodulators, voltammetric study of, 97 Neuron discharge patterns, multichannel recording strategies for, 232 Neuronal activity, PC representation of, 229-230, 232 Neuronal ensemble data, 154-157, 158 Neuronal pairs, in spike train analysis, 152-154 Neurons. See also Firing patterns; Single neurons in dose-response studies, 6-7 in drug-related behavior, 8 functional connectivity of, 228-229, 232 Neurophysiology behavioral, 158 chronic vs. acute approach to, 164 of drug effect vs. behavioral feedback, 165 of drug-related behavior, 7 Neuroplasticity, drug-induced, 2 Neuroscience, experimental methods in, 9 Neurotoxins, vulnerability to, 79 Neurotransmission effects of enhancing, 243 effects of stimulants on, 75 monoamine, 53 Neurotransmitter release, stimulant-induced alterations in, 246-247 Neurotransmitters detection of, 52, 53, 72-73 in dialysate sample, 74 exocytotic release of, 55 for maintenance of alert waking, 252-253 microelectrode response for, 90 release, 88 voltammetric study of, 96-97, 104 wake-promoting actions of, 251 Nex analysis scripts, 226, 227 Nicotine as reinforcer, 25 in self-administration studies, 23 NO, voltammetry study of, 96 Norepinephrine (NE) and alterations in blood pressure, 245 and AMPH-induced increases in arousal, 259, 260 and AMPH-induced locomotor activity and stereotypy, 247 and arousal-enhancing action of drugs, 258263 and arousal state, 255-256 behavioral/cognitive functions within SI of, 256 in chronic drug administration, 134 detection of, 52, 53 detection of sub-pg amounts of, 70-71

2345_frame_IDX Page 278 Tuesday, October 29, 2002 10:58 AM

278

Methods in Drug Abuse Research: Cellular and Circuit Level Analyses

effect of cocaine on, 131 effects of stimulant drugs on, 77, 120 impact of AMPH on transmission of, 240-241 and increased arousal, 255-257 and maintenance of alert waking, 258 and sensory-signal processing, 121-122 voltammetric measures of, 96 Nucleus accumbens. See also Accumbal activity drug effects mediated by, 75-76, 77 histological analyses of, 192-194

O Operant chambers. See also Testing chamber in chronic extracellular recording studies, 168, 200-201 Plexiglas, 167, 168 in self-administration studies, 33-34 Operational Amplifiers (Op-Amps), 147 Opiates neuroadaptive changes in brain produced by, 78 self-administration studies of, 19 Opponent process theory, of motivation, 3 Orbitofrontal cortex, 4

P Paralemniscal pathway, 216 Patch clamp studies, 132 P.D. Systems, 96 Perception, effects of cocaine on, 226 Perfusion fluid, in microdialysis, 55, 62-63 Perfusion system dead volume of, 65 total volume of, 66 Peristimulus time histogram (PSTH) in cocaine studies, 229, 230 in spike train activity, 224 Persistence of drug-induced changes and drug-related behavior, 20 physiological bases for, 132 Pharmacological agents, as reinforcers, 25 Pharmacological probes, microinjection of, 167 Phencyclidine lethal effects associated with, 26 self-administration of, 23 Physiology, of drug abuse, 6. See also Neurophysiology Piezoelectric bimorph stimulator, 130 Plexon, Inc., 147 Polyethylene glycol coating, 145-146 Population data, in spike train activity, 228

Post-reinforcement pause (PRP), 30 Post-stimulus time histograms (PSTHs), 115, 116, 118, 129 Potentiostats, 95-96, 101 Power-spectrum analyses (PSA), 249, 250 Prazosin, impact on arousal of, 258, 259 Principal component (PC) analysis, 229-230, 232 eigenfunctions generated by, 232-233 of spike-train activity, 229-230, 232 Probe-recovery experiment, 56-57 Probes, in microdialysis, 53-55, 62 Probe/tissue interactions, in microdialysis, 57-58 Progressive-ratio schedules, in self-administration studies, 22, 31-33 Psychoactive substances actions on neural network functions of, 214 and sensory signal processing, 231 Psychomotor stimulants. See Stimulant drugs Psychostimulant drugs. See also Stimulant drugs behavior-reinforcing aspects of, 119 effects on central monoaminergic systems of, 120-121 monoamine cell discharge suppressed by, 122 in monoaminergically innervated brain circuits, 122-126 physiological actions of, 124-125 Pub Med database, 52 Pumps for drug infusions, 36 HPLC, 68 for in vivo microdialysis, 68 microinjection, 63 Purkinje cells in chronic drug administration, 134 in patch-clamp studies, 132

Q Quanteon, L.L.D., 96 Quantifying stimulus-evoked responses, in singleunit studies, 116-117 Quantifying transmitter-induced responses, in single-unit studies, 117 Quiet-waking, EEG/EMG scoring of, 248

R Radioimmunoassay methods, 52 Rasters, 152. See also Spike trains Rat. See also Trigeminal somatosensory system, rodent as animal model, 215-216

2345_frame_IDX Page 279 Tuesday, October 29, 2002 10:58 AM

Index

in-dwelling subcutaneous electrode in, 217, 218 Reacquisition, of drug self-administration, 27-28 Realtime Acquisition System Programs for Unit Timing in Neuroscience (RASPUTIN), 218, 220 Rehabilitation treatment programs, lack of success of, 119 Reinforcement of drug addiction, 3 and behavioral and drug history, 23 in cocaine self-administration, 22 cocaine vs. food for, 30-31 fixed-ratio schedules for, 22, 28, 29-31 neurobiology of, 42 progressive-ratio schedules for, 31-33 second-order schedules of, 33 Reinforcers drug vs. nondrug, 21 rate-increasing effect of, 25 Reinitiation, in self-administration, 18 Reinstatement, of drug self-administration, 27-28 Relapse animal models of, 7 disorder, 1 and reinstatement of drug self-administration, 27-28 REM (rapid eye movement) sleep EEG activation associated with, 242 EEG/EMG scoring of, 250 effects of isoproterenol on, 255, 257 and tonic discharge activity, 253 Responding in animal models, 21-22 cocaine-maintained, 20 RESPONSE analysis function, 225, 227, 229 Response latency, drug-induced changes in, 118 Response threshold, drug-induced changes in, 118-119 Responsivity to conditioned stimuli, effect of addictive drugs on, 3 Reuptake, voltammetric study of, 97-98 Reward, drug in chronic recording studies, 195 in drug addiction, 3 Reward-limbic networks, 214 Reward-seeking behavior in cocaine studies, 189-190 and firing patterns, 184 spike train data recorded during, 151-157 Rheodyne-style injector, 68 Ringers' perfusion fluid, 62 Rodents. See Animal models; Rat

279

S Scattergram, in waveform analysis, 170, 224 Second-order schedules, in self-administration studies, 33 Self-administration of addictive compounds, 1-2 acquisition of, 28 chronic extracellular recording of, 163 (See also chronic extracellular recording) effects of repeated, 190-192 in extracellular recording studies, 162 long-term effects of, 134 maintenance of, 29 neurobiological basis of, 157 operant chambers for study of, 168 three phases of, 18 Self-administration procedures dose-response curve in, 19, 21 early research in, 19-21 methodology, 28-33 potential abuse liability and, 31 rodent, 17-18 Self-administration studies apparatus for, 33-38 in drug use determination, 22-23 subjects for, 33 Sensitization, physiological bases for, 132 Sensory-signal processing effects of cocaine on, 226 impact of psychoactive substances on, 231 monoamine influences on, 121-122 Sensory system, model, 215-216. See also trigeminal somatosensory system, rodent Serial recording methods, limitations of, 233 Serotonin (5-HT) and AMPH-induced locomotor activity and stereotypy, 247 detection of, 52, 53 detection of sub-pg amounts of, 70-71 in dialysate sample, 77 effect of cocaine on, 131 effects of methamphetamine on, 77, 79 effects of stimulant drugs on, 77, 120 METH-induced increases in, 81 in sensory neocortical circuits, 121 standard curves for, 73 voltammetric measures of, 96 Serotonin (5-HT) transporters (SERT), 77 Shock, electric, as reinforcer, 25 Silicon-based sensors, 91 Single-neuron studies in drug-related behaviors, 8 and waveform analysis, 169, 170 Single-unit studies

2345_frame_IDX Page 280 Tuesday, October 29, 2002 10:58 AM

280

Methods in Drug Abuse Research: Cellular and Circuit Level Analyses

data analysis in, 116-119 drawback of, 119 drug application in, 114-115 drug-induced changes in, 117-119 experimental protocols in, 115-116 in intact, anesthetized preparations, 112-119 recording techniques, 231 results in, 119 in spike train activity, 224-228 Skull, rat, 62 Sleep. See REM sleep; slow-wave sleep Sleep-wake states behavior-based assessment of, 250-251 EEG/EMG-based assessment of, 249-250 Slow-wave sleep EEG/EMG scoring of, 250 and tonic discharge activity, 253 Software chromatographic, 72 for chronic extracellular recordings, 201-202 for many-neuron recording technique, 147148 NeuroExplorer, 148, 224 RASPUTIN, 218, 220 Solute recovery in dialysate sample, 55-56 relative vs. absolute, 55-57 Spike-train activity across drug-induced states, 221 drug actions on, 231 functional connectivity in, 228-229, 232 population data for, 228-230 single-unit data in, 224-228 Spike-train analyses, 232 Spike trains ensemble analyses of, 154-157, 158 from individual neurons, 114 neuronal pairs and, 152-154 single-neuron, 151-153 Spontaneous discharge, drug-induced changes in, 117-118 Standard curves, 72-74 Statistical approaches, to firing patterns analysis, 175, 178 Stereotaxic surgery, 202-203, 248 Stimulant-dependency, treatment of, 80 Stimulant drugs acute effects of, 75-78 chronic effects of, 78-79 electrophysiological study of, 119-120 illicit, 75 in vivo microdialysis in study of, 52-53 neuroadaptive changes in brain produced by, 78 neurochemical actions of, 242-243

neurotransmission and, 75 voltammetric study of, 101-104 Stimulus-evoked responses, in single-unit studies, 116-117 Stranger software, 148 Stress, environmental, in self-administration studies, 23 Substance abuse, physiological mechanisms in, 112 Substantia innominata (SI), in sleep/wake studies, 255, 256 Substantia nigra, 100 Swivels in chronic extracellular recording, 196, 197 in self-administration studies, 35, 37

T Temporal patterns, and drug-related behavior, 20 Testing chamber. See also Operant chambers for alert waking study, 252-253 in vivo microdialysis within, 60 Plexiglas, 251 Tethering system, in chronic extracellular recording studies, 198-200 Thalamic neurons, effect of cocaine on, 129-132 Thalamocortical neuron function, and LC-NE system, 226, 227, 232 Thalamus effect of cocaine on VPM, 130, 133 effects of psychostimulants on somatosensory, 125 THC (cannabinoids), voltammetric study of, 104 Time-out (TO) procedure, 29 Timing, and drug-related behavior, 20 Timolol EEG effects of, 255 effects on AMPH-induced cortical EEG activation, 263 impact on arousal of, 258, 259 Tolerance, physiological bases for, 132 Transmitter-induced responses, in single-unit studies, 117 TRIAL analysis function, 226, 227, 229 Trigeminal somatosensory system, rodent, 214, 215 equipment in study of, 217, 218 physiology of, 215-216 recording procedures with, 217-218, 219

U UniJet microbore valve, 68, 69

2345_frame_IDX Page 281 Tuesday, October 29, 2002 10:58 AM

Index

Uptake/reuptake, voltammetric study of, 97-98

V Ventral tegmental area (VTA), 75, 76 Videotape analysis in chronic extracellular recording, 186-187 of multineuron recording system, 224 of sleep-wake states, 251 Voltammetric recordings, 88 Voltammetry drug abuse applications of, 101-104 fast-scan cyclic, 94-95 Voltammetry, in vivo brain slices in, 98-100 electrochemical properties of chemicals in, 89 electrodes used in, 90-93 experimental paradigms studied with, 97-101 instrumentation for, 95-96 principles of, 88, 89 quantification and identification in, 95 recording methods for, 7, 93-95 scope of, 96-101 with whole animals, 100-101 VPM pathway, 215-216

281

W Waking EEG activation associated with, 242 and role of neurotransmission, 252-253 Waveform analysis, of electrophysiological data, 169-171 Waveform discrimination, in multineuron recording system, 221-224 Whisker pad, individual vibrissae on, 215, 217 Wilcoxon Matched Pairs Test, 175 Withdrawal from addictive drugs negative affective states associated with, 4 symptoms of, 5 Workstation, in vivo microdialysis, 67. See also Testing chamber

X XcorrStat function, 226

2345_frame_IDX Page 282 Tuesday, October 29, 2002 10:58 AM