Mood and Anxiety Related Phenotypes in Mice: Characterization Using Behavioral Tests (Neuromethods)

NEUROMETHODS

Series Editor Wolfgang Walz University of Saskatchewan Saskatoon, SK, Canada

For other titles published in this series, go to www.springer.com/series/7657

Mood and Anxiety Related Phenotypes in Mice Characterization Using Behavioral Tests

Edited by

Todd D. Gould University of Maryland School of Medicine, Baltimore, MD, USA

Editor Todd D. Gould Department of Psychiatry MSTF; Rm934D University of Maryland School of Medicine 685 W. Baltimore Street Baltimore MD 21201 USA [email protected]

ISSN 0893-2336 e-ISSN 1940-6045 ISBN 978-1-60761-302-2 e-ISBN 978-1-60761-303-9 DOI 10.1007/978-1-60761-303-9 Library of Congress Control Number: 2009927010 # Humana Press, a part of Springer ScienceþBusiness Media, LLC 2009 All rights reserved. This work may not be translated or copied in whole or in part without the written permission of the publisher (Humana Press, c/o Springer ScienceþBusiness Media, LLC, 233 Spring Street, New York, NY 10013, USA), except for brief excerpts in connection with reviews or scholarly analysis. Use in connection with any form of information storage and retrieval, electronic adaptation, computer software, or by similar or dissimilar methodology now known or hereafter developed is forbidden. The use in this publication of trade names, trademarks, service marks, and similar terms, even if they are not identified as such, is not to be taken as an expression of opinion as to whether or not they are subject to proprietary rights. While the advice and information in this book are believed to be true and accurate at the date of going to press, neither the authors nor the editors nor the publisher can accept any legal responsibility for any errors or omissions that may be made. The publisher makes no warranty, express or implied, with respect to the material contained herein. Printed on acid-free paper springer.com

Series Preface Under the guidance of its founders Alan Boulton and Glen Baker, the Neuromethods series by Humana Press has been very successful since the first volume appeared in 1985. In about 17 years, 37 volumes have been published. In 2006, Springer ScienceþBusiness Media made a renewed commitment to this series. The new program will focus on methods that are either unique to the nervous system and excitable cells or which need special consideration to be applied to the neurosciences. The program will strike a balance between recent and exciting developments like those concerning new animal models of disease, imaging, in vivo methods, and more established techniques. These include immunocytochemistry and electrophysiological technologies. New trainees in neurosciences still need a sound footing in these older methods in order to apply a critical approach to their results. The careful application of methods is probably the most important step in the process of scientific inquiry. In the past, new methodologies led the way in developing new disciplines in the biological and medical sciences. For example, Physiology emerged out of Anatomy in the 19th century by harnessing new methods based on the newly discovered phenomenon of electricity. Nowadays, the relationships between disciplines and methods are more complex. Methods are now widely shared between disciplines and research areas. New developments in electronic publishing also make it possible for scientists to download chapters or protocols selectively within a very short time of encountering them. This new approach has been taken into account in the design of individual volumes and chapters in this series. Wolfgang Walz

v

Preface Mood and anxiety disorders are common brain diseases that affect over 15% of the world population. Included in this group of diseases are major depression and bipolar disorder and generalized anxiety, panic, obsessive compulsive, and posttraumatic stress disorders. Recent years have seen a tremendous increase in our understanding of the pathogenesis and pathophysiology of mood and anxiety disorders. This increased knowledge parallels a remarkable growth in the use of the laboratory mouse as a tool to both understand the biological and genetic basis of diseases, including those of a psychiatric origin, and to develop improved treatments. While it is not possible to reproduce fully human mood or anxiety disorders in mice, the study of behavioral phenotypes modeling aspects of these diseases provides invaluable insights into potential disease and treatment mechanisms. For this reason, the application of mouse models will increase as additional underlying susceptibility genes are discovered, new targets for medications are identified, and clinical studies reveal novel neurobiological markers that may be translated between humans and mice. This book provides an overview of behavioral approaches that are utilized in the characterization of mood and anxiety disorder-related behaviors in mice. Additionally, many of the chapters describe behavioral assays that are commonly used – both in industry and academia – to assess the potential antidepressant and anxiolytic efficacy of novel compounds. The contributing authors to this book are world-renowned scientists with broad experience in the development and application of behavioral tasks in mice. The book is intended first as a resource for scientists actively pursuing or interested in establishing behavioral protocols in their laboratories. It can also serve as a reference for those students, scientists, and practitioners who have an interest in better understanding the preclinical behavioral methods used in mood and anxiety research. As a cautionary note, there are a number of subtleties in mouse husbandry, handling, and testing procedures that cannot be acquired solely from following a book. Thus, those inexperienced with techniques used to test behavior in mice are encouraged to seek collaboration with an experienced behavioral neuroscientist to help address these underappreciated but significant experimental issues. Todd D. Gould

vii

Contents SeriesPreface . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Preface . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Contributors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1. 2.

3.

4. 5.

6.

7.

8.

9.

The Open Field Test . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Todd D. Gould, David T. Dao, and Colleen E. Kovacsics Analysis of Grooming Behavior and Its Utility in Studying Animal Stress, Anxiety, and Depression . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Amanda N. Smolinsky, Carisa L. Bergner, Justin L. LaPorte, and Allan V. Kalueff Digging in Mice: Marble Burying, Burrowing, and Direct Observation Reveal Changes in Mouse Behavior . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Robert M.J. Deacon Circadian and Light Modulation of Behavior . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Cara M. Altimus, Tara A. LeGates, and Samer Hattar Ultrasonic Vocalizations by Infant Mice: An Ethological Expression of Separation Anxiety . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . James T. Winslow The Forced Swimming Test in Mice: A Suitable Model to Study Antidepressants . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Martine Hascoe¨t and Michel Bourin

v vii xi 1

21

37 47

67

85

The Tail-Suspension Test: A Model for Characterizing Antidepressant Activity in Mice . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 119 Olivia F. O’Leary and John F. Cryan Stress-Induced Hyperthermia in the Mouse . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 139 Christiaan H. Vinkers, Ruud van Oorschot, Berend Olivier, and Lucianne Groenink Factors of Reproducibility of Anhedonia Induction in a Chronic Stress Depression Model in Mice . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 153 Tatyana Strekalova and Harry Steinbusch

10. Learned Helplessness in Mice . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 177 Hymie Anisman and Zul Merali 11. The Mouse Light–Dark Box Test . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 197 Martine Hascoe¨t and Michel Bourin 12. Using the Elevated Plus Maze as a Bioassay to Assess the Effects of Naturally Occurring and Exogenously Administered Compounds to Influence Anxiety-Related Behaviors of Mice . . . . . . . . . . . . . . . . . . . . . . . . . . . 225 Alicia A. Walf and Cheryl A. Frye 13. Novelty-Induced Hypophagia. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 247 Stephanie C. Dulawa

ix

x

Contents

14. Acute and Chronic Social Defeat: Stress Protocols and Behavioral Testing. . . . . . . 261 Alessandro Bartolomucci, Eberhard Fuchs, Jaap M. Koolhaas, and Frauke Ohl 15. Reduction of Submissive Behavior Model for Antidepressant Drug Testing in Mice . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 277 Ewa Malatynska, Albert Pinhasov, and Richard J. Knapp 16. Mice Models for the Manic Pole of Bipolar Disorder . . . . . . . . . . . . . . . . . . . . . . . 297 Shlomit Flaisher-Grinberg and Haim Einat Index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 327

Contributors HYMIE ANISMAN l Institute of Neuroscience, Carleton University, Ottawa, ON, Canada CARA M. ALTIMUS l Department of Biology, Johns Hopkins University, Baltimore, MD, USA ALESSANDRO BARTOLOMUCCI l Dipartimento di Biologia Evolutiva e Funzionale, Universita` di Parma, Parma, Italy CARISA L. BERGNER l Department of Physiology and Biophysics, Georgetown University Medical School, Washington, DC, USA MICHEL BOURIN l Faculte´ de Me´decine, Neurobiologie de l’Anxie´te´ et de la De´pression, Nantes, France JOHN F. CRYAN l School of Pharmacy, Department of Pharmacology and Therapeutics, Alimentary Pharmabiotic Centre, University College Cork, Cork, Ireland DAVID T. DAO l Department of Psychiatry, Mood and Anxiety Disorders Program, University of Maryland School of Medicine, Baltimore, MD, USA ROBERT M.J. DEACON l Department of Experimental Psychology, University of Oxford, Oxford, UK STEPHANIE C. DULAWA l Committee on Neurobiology, Department of Psychiatry, University of Chicago, Chicago, IL, USA HAIM EINAT l College of Pharmacy, University of Minnesota, Duluth, MN, USA SHLOMIT FLAISHER-GRINBERG l College of Pharmacy, University of Minnesota, Duluth, MN, USA CHERYL A. FRYE l Departments of Psychology, Biological Sciences and The Centers for Neuroscience and Life Sciences Research, The University at Albany, SUNY, Albany, NY, USA EBERHARD FUCHS l Clinical Neurobiology Laboratory, German Primate Center, Leibniz Institute for Primate Research, G¨ottingen, Germany TODD D. GOULD l Department of Psychiatry, Mood and Anxiety Disorders Program, University of Maryland School of Medicine, Baltimore, MD, USA LUCIANNE GROENINK l Department of Psychopharmacology, Utrecht Institute for Pharmacological Sciences and Rudolf Magnus Institute of Neuroscience, Utrecht University, Utrecht, The Netherlands MARTINE HASCOE¨T l Faculte´ de Me´decine, Neurobiologie de l’Anxie´te´ et de la De´pression, Nantes, France SAMER HATTAR l Department of Biology, Johns Hopkins University, Baltimore, MD, USA ALLAN V. KALUEFF l Department of Physiology and Biophysics as well as the Stress Physiology and Research Center (SPaRC), Georgetown University Medical School, Washington, DC, USA

xi

xii

Contributors

RICHARD J. KNAPP l Sanofi-Aventis, Bridgewater, NJ, USA JAAP M. KOOLHAAS l Department Behavioral Physiology, University of Groningen, Haren, The Netherlands COLLEEN E. KOVACSICS l Department of Psychiatry, Mood and Anxiety Disorders Program, University of Maryland School of Medicine, Baltimore, MD, USA JUSTIN L. LAPORTE l Department of Physiology and Biophysics, Georgetown University Medical School, Washington, DC, USA TARA A. LEGATES l Department of Biology, Johns Hopkins University, Baltimore, MD, USA FRAUKE OHL l Faculty of Veterinary Medicine, Department of Animals, Science & Society, University Utrecht, Utrecht, The Netherlands OLIVIA F. O’LEARY l School of Pharmacy, Department of Pharmacology and Therapeutics, University College Cork, Cork, Ireland EWA MALATYNSKA l Lilly Research Laboratories, Lilly Corporate Center, Indianapolis, IN 46285 ZUL MERALI l Institute of Mental Health Research, Royal Ottawa Hospital and the School of Psychology, and Institute of Cellular and Molecular Medicine, University of Ottawa, Ottawa, ON, Canada BEREND OLIVIER l Department of Psychopharmacology, Utrecht Institute for Pharmacological Sciences and Rudolf Magnus Institute of Neuroscience, Utrecht University, Utrecht, The Netherlands; Department of Psychiatry, Yale University School of Medicine, New Haven, CT, USA RUUD VAN OORSCHOT l Department of Psychopharmacology, Utrecht Institute for Pharmacological Sciences and Rudolf Magnus Institute of Neuroscience, Utrecht University, Utrecht, The Netherlands ALBERT PINHASOV l Department of Molecular Biology, Ariel University, Center of Samaria, Ariel, Israel AMANDA N. SMOLINSKY l Department of Physiology and Biophysics, Georgetown University Medical School, Washington, DC, USA HARRY STEINBUSCH l Department of Neuroscience, Faculty of Health, Medicine, and Life Sciences, Maastricht University, Maastricht, The Netherlands TATYANA STREKALOVA l Department of Neuroscience, Faculty of Health, Medicine, and Life Sciences, Maastricht University, Maastricht, The Netherlands CHRISTIAAN VINKERS l Department of Psychopharmacology, Utrecht Institute for Pharmacological Sciences and Rudolf Magnus Institute of Neuroscience, Utrecht University, Utrecht, The Netherlands ALICIA A. WALF l Department of Psychology Research, The University at Albany, SUNY, Albany, NY, USA JAMES T. WINSLOW l NIMH IRP Neurobiology, Primate Core, National Institutes of Health (NIH), Bethesda, MD, USA

Chapter 1 The Open Field Test Todd D. Gould, David T. Dao, and Colleen E. Kovacsics Abstract The open field test (OFT) is a common measure of exploratory behavior and general activity in both mice and rats, where both the quality and quantity of the activity can be measured. Principally, the open field (OF) is an enclosure, generally square, rectangular, or circular in shape with surrounding walls that prevent escape. The most basic and common outcome of interest is ‘‘movement’’; however, this can be influenced by motor output, exploratory drive, freezing or other fear-related behavior, sickness, relative time in circadian cycle, among many other variables. Distance moved, time spent moving, rearing, and change in activity over time are among many measures that can be tabulated and reported. Some outcomes, particularly defecation, center time, and activity within the first 5 minutes, likely gauge some aspects of emotionality including anxiety. The OFT is also commonly used as a mechanism to assess the sedative, toxic, or stimulant effects of compounds. Thus, the OFT measures a number of facets of behavior beyond simple locomotion. As such, the test has a number of uses and is included in almost every thorough analysis of rodent behavior. Key words: Activity, locomotion, exploratory activity, arena, anxiety, thigmotaxis mouse, rodents.

1. Background and Historical Overview The open field test (OFT) is a common measure of exploratory behavior and general activity in rodents, where both the quality and quantity of the activity can be measured. Hall is widely credited for first introducing the OFT (1, 2). While the original studies were in rats, the OFT has also been extensively used in mice. In addition to total distance moved, many qualities of the movement are also analyzed. These include time spent along the walls (thigmotaxis) compared to time in center, distance moved over different time periods, and rearing. Additionally, the OFT is sometimes used as a means to assess general activity levels as a ‘‘control’’ experiment for other behavioral tests that involve activity (3–5). T.D. Gould (ed.), Mood and Anxiety Related Phenotypes in Mice, Neuromethods 42, DOI 10.1007/978-1-60761-303-9_1, ª Humana Press, a part of Springer Science+Business Media, LLC 2009

1

2

Gould, Dao, and Kovacsics

The OFT is also commonly used to assess the sedative, toxic, or stimulant effects of compounds. As such, the test has a number of uses and is included in almost every thorough analysis of rodent behavior. Principally, the open field (OF) is an enclosure, generally square, rectangular, or circular in shape with surrounding walls that prevent escape (Fig. 1.1). Thus, the OFT does not generally utilize a ‘‘field’’ in the true sense of a wide, expansive area, where an animal can move and behave unfettered, but rather an arena, that for mice, will generally vary from 25 cm2 to over 250 cm2. While there are some exceptions (6), in almost all experimental designs mice are placed in the arena by the investigator and thus forced to interact with a novel environment. This forced entry should be considered when interpreting the results, as animals do not actively seek entry into the arena. The traditional OFT is between 2 and 10 minutes in duration. This short length of time emphasizes exploratory behavior and

Fig. 1.1. Examples of open field arenas. (A) Square open field, 100  100 cm. (B) Square open field, 50  50 cm. (C) Circular open field, 250 cm diameter (photograph contributed by Dr. Greg Elmer, University of Maryland School of Medicine, Maryland Psychiatric Research Center and Department of Psychiatry).

The Open Field Test

3

response to novelty, rather than baseline activity. The short length of time (used in early studies) was due to many reasons including the method of manually acquiring data. More recent approaches, including video tracking and tracking by the number of infrared beam breaks, allow for much higher throughput and longer periods of monitoring. Distance moved, time spent moving, rearing, and change in activity over time are all measures that can be tabulated and reported (Table 1.1). There are many other less commonly reported measures including time spent without moving, sniffing, vocalization, and teeth chattering (see Walsh and Cummins (7) for a thorough review of OFT-dependent measures). More recently, particularly with the advent of higher-throughput methodology and a focus on single outcomes in behavioral studies, there has generally been a decrease in the number of variables that are reported. However, there have been efforts to include many of the variables that can be assessed by computer systems including darting, texture of the path taken, activity density, and principal component analysis including ethological measurements and conventional variables (8–12). Overall, we do not extensively discuss many of these less commonly monitored outcomes in this chapter. Furthermore, we do not extensively discuss approaches to conduct factor analysis or to define relationships between factors (13).

Table 1.1 Partial list of common open-field-dependent parameters that can be assessed. Assessment of these measures can be as total number, grouped into time bins, or measured in terms of latency. See Walsh and Cummins (1976) (7) for an historical perspective Movement – Distance moved (either actual or relative distance) – Time spent moving – Rearing (vertical activity) – Freezing – Grooming – Other stereotypic behaviors Location – Time spent in center – Crosses into center Autonomic nervous system – Defecation (number of fecal boli) – Urination

4

Gould, Dao, and Kovacsics

The most basic and common outcome of interest is ‘‘movement’’; however, this can be influenced by motor output, exploratory drive, freezing (or other fear-related behavior), sickness, relative time in circadian cycle, and a variety of other variables. Importantly, the OFT (especially a single short test) is not an accurate proxy for baseline or spontaneous activity, which can only be measured over longer monitoring in an environment to which mice are acclimated to or in the home cage (14, 15). Likewise, the OFT is not simply a measure of motor activity, but involves other factors such as exploratory drive (curiosity) and fear (or anxiety) (16, 17). Most mice will have a tendency to spend the majority of their time in close proximity to the walls, a phenomenon referred to as thigmotaxis. The OFT initially begins as a test of novelty, thereafter, if the test continues for a long enough period of time, general activity can be measured (5). During the early exposure period, anxiety likely plays a role both in terms of general activity and thigmotaxis, and later more so only in regard to thigmotaxis. Thus, while the OFT is increasingly being used as a proxy for measuring general locomotor activity, this is a trend that disregards a number of different facets that contribute to OF behavior (4, 17). There have been discussions as to what degree the OFT is a reliable measure of emotionality (4, 18). Some outcomes, particularly defecation, center time, and activity within the first 5 minutes, likely measure some aspect of emotionality. Early activity in the OF can measure anxiety, as it may induce separation stress (since the animal is separated from its cage mates) and agoraphobia (exposure to a large arena is quite different from the familiar standard shoebox-style holding cages) (19). The arrangement of a typical OF also contrasts with how rodents usually live in the wild: in social networks, burrows underground, or protected areas. Moreover, center time is generally sensitive to acute administration of gamma-aminobutyric acid-(GABA) acting anxiolytics (in most cases benzodiazepines), but is generally considered insensitive to short- or longterm treatment with SSRIs (19). However, this may be strain and dose dependent, and many of the existing studies have been performed in rats. As one example of data that conflict with this general finding, Dulawa et al. have shown that the long-term administration of fluoxetine in drinking water at doses of 18 and 25 mg/kg/day, but not 10 mg/kg/day, resulted in an increase in the number of center entries in BALB/cJ mice (20). However, the general insensitivity to monoamine-acting anxiolytics, the natural habits and tendencies of rodents, and other evidence, have led to the suggestion that center time may measure ‘‘normal’’ anxiety and be less sensitive to severely ‘‘pathological’’ anxiety as seen in some human diseases (19).

The Open Field Test

5

Movement of mice within an arena can be monitored using a number of different mechanisms. Historically, the first method relies on marking the surface of the OF with parallel lines both horizontal and vertical (or sometimes concentric in the case of a circular arena) forming a grid (Fig. 1.2A). An observer (either live or via video recording) keeps track of the number of times a mouse crosses from one box on the grid to another, and also keeps track of whether the entry is into a center or peripheral square. A second mechanism involves the use of infrared sensors, again generally placed in two directions (Fig. 1.2B). Sensors may also be placed in a third dimension (height) to monitor either rearing or head dips (the latter if the OF includes a hole poke floor). The final common mechanism involves overhead tracking using a video camera and software (Fig. 1.2C) available from any of a number of suppliers including Noldus (Ethovision; Wageningen, The Netherlands), Clever Sys Inc. (TopScan; Reston, VA, USA), and Stoelting Co (ANY-maze; Wood Dale, IL, USA) San Diego Instruments (ANYmaze; San Diego, CA, USA).

Fig. 1.2. Approaches to track open field behavior. (A) Open field with squares marked for manual acquisition of rat locomotion (photograph contributed by Dr. Leonardo Tonelli, University of Maryland School of Medicine, Department of Psychiatry). (B) Digiscan IR photocell automated open field (Omnitech Electronics, Columbus, OH, USA; photograph contributed by Dr. Harry June, University of Maryland School of Medicine, Department of Psychiatry). (C) Automated video tracking by TopScan (Cleversys Inc., Reston, VA, USA).

6

Gould, Dao, and Kovacsics

2. Equipment, Materials, and Setup 2.1. Room

1. Isolated from sound and unintentional interruptions. If there is a concern of variability in ambient sound, then a white noise machine (e.g., White Noise Generator, San Diego Instruments, San Diego, CA, USA or Sound Therapy Machine, Conair Corporation, Glendale, AZ, USA) can be used. 2. Consistent lighting of the OF. As discussed in ‘‘Experimental Variables’’ the amount of lighting used by different investigators varies considerably and can be either red or white. As a rule of thumb, our laboratory typically uses dim lighting of about 30 lux. Brighter light (especially brighter white light) may increase thigmotaxis and/or decrease locomotion, but it is not uncommon to conduct the test in a brightly lit environment. Regardless of the light level chosen, the investigator should use a light meter to record the lighting level and report this level in any publication.

2.2. Open Field

Any number of materials including plastic, metal, or wood can be, and probably has been, used to construct the arena. Plastic or corrosion-resistant metals are preferable, as these materials can be easily cleaned and sterilized. Size can vary dramatically (see Experimental Variables). Both rectangular and circular arenas are used. Rectangular arenas have the advantage of making it easier to fit multiple arenas into a single testing room. Circular arenas remove the option for mice to spend time in the corners. For the simple measurement of activity, most sizes will provide reasonable data. If one is also measuring other parameters, in particular thigmotaxis, then larger-sized fields may provide more accurate data (a size of at least 40  40 cm is recommended). Wall height should generally be of at least 35 cm. This height both limits the ability of mice to see beyond the walls and is generally, in all but extreme circumstances, high enough to prevent mice from jumping out of the arena. It is most common to test mice on a bare floor. However, some laboratories test with a thin layer of bedding. Many protocols call for cleaning thoroughly between mice, though other procedures have bedding remaining in place between animals. Still other protocols only remove feces and urine for the reason that mild, but consistent, olfactory clues are preferable to any lingering smell of disinfectant. We provide a protocol for mild cleaning with a disinfectant between mice; however, regardless of the approach chosen, the most important thing is to remain consistent within and between experiments in your own laboratory. If using an overhead video tracking system consider a dull floor to prevent reflections.

The Open Field Test

7

2.3. Method to Acquiring Data

As mentioned this can be done manually, if the OF is partitioned into different regions (Fig. 1.2A). More commonly the test is performed in arenas equipped with photocells or the animals are monitored with computerized tracking software (Fig. 1.2B,C).

2.4. Timer

This may be automated by the apparatus or digital tracking software or may be a separate handheld timer.

2.5. New Cage for Mice After Test

Generally it is preferable to place the mouse in a new cage rather than back with his or her cage mates, as reintroduction of the mouse may modify behavior of mice not yet tested.

2.6. Material for Clean-Up

Generally includes a trash bin and paper towels to clean up urine and to wipe down the arena. A small and quiet handheld vacuum generally works well for removal of compacted fecal boli.

2.7. Disinfectant

A spray bottle with a disinfectant such as MB-10 (Quip Laboratories Inc.,Wilmington, DE, USA) or 10–70% alcohol can be used for a general wipe down. Note that alcohol should be used with caution since it can cause cracks in many plastics and dissolves glues used to hold plastics and other materials together.

3. Procedure Any type of novelty or stress may modify exploratory behavior. For this reason, it is critical to reduce novelty, for example cage changes, for a period of time prior to the test, and generally for at least 24 hours. It is equally critical that all animals to be compared within a cohort are, and have been, handled in the same manner throughout their lives. All attempts should be made to minimize background stress and to make certain that all treatments are randomized. Procedures will vary between laboratories, and depending upon the experimental design, mice may only be tested once or multiple times to assess acclimation or sensitization to drugs of abuse. Historically, the OFT generally occurs over a period of 5 or 10 minutes. One reason for this time period was that activity was manually scored, one animal at a time. Newer systems are automated and allow animals to be individually tested in multiple arenas at any one time. In our laboratory, we commonly perform the OFT over a time period of 60 minutes. This allows us to assess both exploratory activity over the initial exposure period and general activity that is often seen after the novelty of the arena has abated, typically 30 minutes into the exposure.

8

Gould, Dao, and Kovacsics

– Transport mice to the testing room at least 1 hour prior to testing to allow mice to acclimate to the experimental room. This time period should be extended if the transfer involves excessive changes in the ambient environment or other potential stressors. – Turn on video camera, photocell detection apparatus, or alternative tracking system. Note that regardless of the data acquisition approach utilized, it is good practice to record each session using a video camera. This will allow one to reassess the results of a session at a later time if necessary. – Clean the OF. Even if the OF is not dirty prior to the initial trial, it is important to clean so that any residual smell of the disinfectant is experienced equally by all animals. – Place mouse in the OF. Placing mice in the center is most common. However, placing adjacent to a wall is sometimes used either because it does not immediately expose the mouse to a stressful situation (center of the OF) or because it may be easier to reach (in the case of a large area OF). – Start timer. – The tester should either leave the room or position himself or herself as far as possible away from the arena and field of view of the mouse. The tester should remain still and quiet throughout each trial. – At completion of trial, remove the mouse from the arena and place in a new cage. – Count fecal boli. As described, the number of boli is related to the affective or anxiety state of the animal. An increased number of boli is correlated with other measures of anxiety (21–23). – Clean the OF with disinfectant (and vacuum if used). – Wait until disinfectant has fully dried prior to placing next mouse in the OF. The type of system, and to some extent the size of the OF, will determine the type and number of parameters that can reliably be measured. For manual tracking, this generally includes number of squares crossed, time spent in center squares, number of rears, and fecal boli. For automated systems, a seemingly endless number of parameters can be measured. However, generally the most useful outcomes are distance moved (either in true distance measurements or relative units), time spent moving, time spent in the center, vertical activity, and fecal boli. It is generally helpful to compute these outputs over the entire time period (e.g., total distance moved), and depending upon the measure, into time bins as well. For example, distance moved per 5-minute bin is a common method to present and analyze the data. The primary outcomes of the experiments are generally

The Open Field Test

9

considered as continuous variables and analyzed with a t-test or analysis of variance (ANOVA) depending upon the number of groups or experimental conditions compared. Binned data (for example, distance moved in 5-minute intervals over 1 hour) should be analyzed by two-way ANOVA with time as one factor and treatment (or genotype, etc.) as the other factor.

4. Anticipated Results As mentioned, it is generally preferable to not simply measure total distance moved or total duration of behavior, but also to temporally assess variations of behavior over time. This is most commonly accomplished for ambulatory movements, but may be accomplished for any of the dependent variables mentioned in Table 1.1. For measurements of ambulation, it is generally possible to assess the data in as short as 5-minute intervals. Assessing below 5 minutes is also possible, but this may result in a large degree of variability between measurements as well as large standard error (or standard deviation) for measurements. While the pattern of activity can vary due to any number of factors (including strain variations; Fig. 1.3A,B), mice will generally be highly active (show exploratory activity) when initially placed in the OF or shortly after being placed in the OF. This high level of activity will slowly decrease over time, and usually in about 30–60 minutes will reach a relatively steady state. Mice will generally tend to spend a minority of the time (generally less than 15%) in the center of an OF, but this depends on a number of factors including the size of the OF, strain, length of the test, and lighting conditions (Fig. 1.3C,D).

5. Experimental Variables A number of experimental variables can modify the outcome of the OFT. Many of the more common variables are discussed below, while Table 1.2 lists these as well as less common variables. 5.1. Genetic Background

Evidence of a genetic component to OFT behavior has been demonstrated through selective breeding experiments. One such experiment bred ‘‘high’’-activity and ‘‘low’’-activity male and female mice (24). After 10 generations of selection, subjects in the ‘‘high’’-activity line were more than twice as active as the mice

10

Gould, Dao, and Kovacsics

Fig. 1.3. Examples of results obtained from open field testing. (A) D-Amphetamine significantly increased the distance moved in C57Bl/6 J mice given an i.p. injection at a dose of 2 mg/kg 15 minutes before testing. (B) D-Amphetamine given at the same dose did not affect the distance moved in A/J mice, which overall showed a low level of activity. (C and D) Examples of trace images from TopScan (CleverSys Inc, Reston, VA, USA) showing the path taken by an individual mouse over a 10-minute testing session in the OF. (C) Trace image of a mouse that spent limited time moving in the center of the arena, and that displayed a high level of thigmotaxis. (D) Trace image of a mouse that frequently entered the center of the arena.

Table 1.2 A number of variables have been shown to modify the behavior of mice and rats in the open field test Variable

References (mouse)

References (rat/other rodent)

Age

(27, 39)

(40–48)

Cage position

(54)

Color of OF arena

(45) (continued)

The Open Field Test

11

Table 1.2 (continued) Variable

References (mouse)

Estrus cycle status

(39, 49)

External cues

References (rat/other rodent)

(81)

Sex

(30, 33, 49, 50)

(37, 42, 43, 45, 51–55)

Handling

(34)

(35, 36, 47, 51, 56–59, 65)

Injections

(82)

Housing

(32, 34, 49, 62, 83, 84)

(35, 52, 57, 60, 61, 85, 86)

Lighting

(50, 87, 88)

(37, 46, 55, 65,, 73)

Mating status

(39)

Noise level

(50)

Previous behavioral Experience

(63, 64, 89)

(56, 90, 91)

Re-exposure to OF

(27, 30, 70)

(35, 42, 43, 45–47, 51, 53, 55, 65, 66, 86, 91)

Shape

(70, 89)

(55, 69)

Size

(50, 88)

(68, 92)

Strain

(24, 26–34, 63)

(35–37, 54)

Texture

(72)

Time of day

(75)

Wall/ceiling height

(71, 93)

Weaning age

(48)

(56, 61)

from the ‘‘low’’-activity line. A similar line of experiments selectively bred mice for both high and low levels of thigmotaxis in an OF (25). When inbred strains are compared on OF activity, the C57BL/6 line is often found to show high levels of activity in various versions of the OFT (26–30), although this effect may be age dependent and not evident until mice are older (31). Some studies have not found this strain to be most active, indicating that there is some variability that may be attributable to other experimental factors (32, 33). Many strains of mice have been shown to vary in their OFT behavior, as have several rat strains (34–37). When working with any mouse, especially mice with genetic manipulations, it will be important to consider the background strain the mouse line is on, when designing and conducting OFT experiments (38).

12

Gould, Dao, and Kovacsics

5.2. Age

In mice, the effect of age on OFT behavior is not precisely defined. It has been reported that age is associated with increases in activity or no change in activity (27, 39). Looking at the rat literature, there also seems to be an unclear picture. As a general trend (although there are contradictory studies), younger rats seem to show more activity than older rats (40– 44). However, this pattern has been shown in at least one study to be dependent on sex with no age effect seen in female rats, but younger male rats were more active than their older counterparts (45). Still other studies find a nonlinear relationship with an increase in activity in early to midlife, followed later by a decline in activity (27, 46, 47). Yet another group determined that when rats were tested during the day, no age effects were seen, but when tested at night, the older rats were more active than the younger rats (48). Other factors may impact locomotor activity and interact with the effects of age.

5.3. Sex

Another variable that may affect OF behavior is sex. Studies have found both no differences and significant sex differences (25, 33, 49, 50). In one study males of several strains were more active than females (included D2, C3H, Balb, and B6129F1), and in several other strains no difference was observed between sex (30). For rats, there are more studies available for review. The results of most studies suggest that female rats have significantly higher levels of locomotor activity compared to male rats (37, 42, 45, 51–54). Other studies have found an increased locomotor activity in female rats that is dependent on the age of the animal or the time of testing (43, 55). While there is evidence indicating that female rats may be more active than male rats, the evidence for mice is not as clear. However, it is still advisable to analyze the data obtained from the OFT separately by sex, to determine if your cohort of mice do display sex differences.

5.4. Handling

The handling history of the mice may impact the results of the OFT. Some investigators spend several minutes a day handling the mice in the days leading up to behavioral testing, with the hopes of reducing pre-test stress that may be induced by transferring the animal from the home cage to the testing arena. However, some investigators do not expose their mice to handling before testing. In mice, a handling procedure did not modify OFT behavior compared to control mice that were not exposed to handling (34). In rats, handling before testing generally seems to increase locomotor activity (35, 36, 47, 51, 56, 57). However, other studies in rats have found that instead of increasing activity in the OFT, handling had no effect on locomotor behavior (58, 59). In all of those studies, handling procedures and timing varied somewhat: some placed pups on a

The Open Field Test

13

warm surface daily, while others physically stroked pups daily. In a study with mice, the pups were placed under a glass jar for several minutes a day. The type, daily length, and overall duration of handling may differently affect behavioral results. It is important to handle/not handle mice consistently within and between cohorts in your laboratory. 5.5. Housing Conditions

Another factor to consider is how the mice will be housed prior to OF, or any behavioral, testing. Typically, mice are housed several animals to a cage; this is due mainly to ethological concerns as mice are social animals and show signs of stress when assessed in behavioral tests after being single-housed. In mice, locomotion was increased in socially isolated animals compared to mice that were group-housed before OF testing (32, 34). Another study found that isolated female mice showed decreased locomotion compared to group-housed mice, and in males there was no effect of isolation (49). In rats, studies show that isolated animals show a higher level of locomotor activity than group-housed animals (52, 60, 61). In addition to social isolation, another option for housing is whether to add physical enrichment, which typically consists of rotating novel objects through the animals’ cages. In mice, physical enrichment of the home cage has been reported not to modify OF locomotion (62). In rats, physical enrichment was shown to lead to a decrease in activity when the animals were tested in the OF (52, 60). Both social and physical enrichment may impact results in the OF test.

5.6. Behavioral History

The behavioral testing history of mice can effect the results of the OFT. It is common to run behavioral tests in the order from least to most stressful; so the OFT will typically be run early on in a behavioral testing battery. If behavioral tests are conducted on mice before the OFT, then it is important to keep in mind that these experiences may alter behavior in the OFT. For example, one study subjected mice to no stimulation, moderate stimulation, or a series of electrical shocks before OF testing (63). The undisturbed and shocked groups showed low levels of activity, while the moderately stimulated group showed the highest activity. In this particular study, the intensity of previous exposure had a U-shaped effect on locomotor activity. In a separate study, C57BL/6J mice that had undergone testing on different tasks were reported to show different levels of both locomotion and vertical activity in the OFT when compared with naı¨ve mice (64). Also important to keep in mind is the animal’s familiarity with the OFT. Re-exposure to the same arena may result in different activity levels (generally less) than during the first exposure. One early study examining the effects of repeated OF exposure in mice used naı¨ve age-matched mice as controls (27). Mice that were tested repeatedly showed lower

14

Gould, Dao, and Kovacsics

activity at each time point than naı¨ve age-matched control mice. Another study found that the effects of re-exposure varied by strain of mouse tested, with some strains showing an increase with more exposures, others a decrease, and still other strains showed no change over repeated exposures (30). In rats, most of the data available suggest that repeated exposure to the OFT results in decreased ambulation with each new exposure (35, 45, 55, 65). The effect of re-exposure may vary by sex; females showed an increase in activity over repeated exposures, while males showed no change or a decrease in activity (43, 51, 53). Another study found that rats showed a decrease in activity with repeated exposures, but then with continued testing, an increase in activity was observed (66). 5.7. Open Field Size

Two studies in mice have found that larger OFs (greater than 80 cm) are associated with an increase in total distance moved when compared to a smaller apparatus while total ambulation time and rearing behavior stayed the same (50, 67). A more recent study assessed the effect of OF size in the social vole (Microtus socialis guentheri) and found a trend toward increased activity as OF size increased (68). However, this study found that despite the drastic changes in OF size (40  60 cm–260  400 cm), the behavioral changes were mild suggesting that the overall locomotion and spatial pattern of OF activity are relatively stable between OF sizes. Thus, generally it appears that while mice will travel more in a larger OF, overall activity patterns are relatively stable when using either a small or large OF.

5.8. Open Field Shape, Walls, and Ceiling

The OF apparatus has taken many shapes over the years ranging from circular to square and rectangular. Yet, there are a lack of studies that consider the effect shape has on OF behavior. It is often suggested that the presence of corners in the square and rectangular OF presents an aversive stimuli to mice. The two studies that have investigated differences of OF shape on behavior have shown that activity and temporal organization are similar between square and circular OFs in both mice and rats (69, 70). The height of the OF perimeter walls varies among the different types of OF apparatus and some studies use an elevated OF with no walls (55). However, a walled OF is most commonly used. Some OF apparatuses are enclosed with a ceiling and there is some evidence that the height of the ceiling does have an effect on OF behavior. In one study, the ceiling height of a square OF was varied among 2.5 cm, 4 cm, and 91 cm (71). The study found that male and female BALB/c mice were most active with the lowest ceiling and least active with the highest ceiling. Male C57BL/6 mice also had a tendency for higher activity with the lowest ceiling, while male DBA mice had the highest activity at the 4-cm ceiling height.

The Open Field Test

15

5.9. Open Field Color and Texture

The most common colors of the OF apparatus are white, black, and transparent, though other colors have been used. Considering the poor vision of mice and rats, it is not expected that color has a significant effect on OF behavior. The results of one study found that rats traveled more in a white OF versus a black OF (45). In general, OF color is more of a practical concern in allowing the experimenter, or computerized tracking software, easy visualization of the mouse during testing. For example, a black mouse is easier to see on a white background and vice versa. Texture of the OF apparatus is a behavioral variable that has not been widely reported in the literature. Most OF apparatuses are made of plastic due to the ease of cleaning and durability, but many different materials like wood, metal, and glass have been used. Yet, it is unclear if changes in texture of the OF through the use of a different materials or the addition of bedding have effects on OF behavior. One study compared soil, bedding, metal, and Astroturf on the OF behavior of mice and found that mice traveled least on the metal surface, traveled most on soil, and defecated more on both Astroturf and metal (72). Practically, a bare plastic surface is commonly used due to ease and speed of cleaning.

5.10. Lighting

Mice and rats have a natural, evolutionarily conserved aversion toward open, brightly lit areas and as a result the lighting of the OF must be consistently controlled. Multiple studies have shown that different light levels significantly affect OF behavior. In mice, studies have shown that light levels above 300 lux significantly reduced locomotor activity in the OF when compared to activity at 2–10 lux (50, 67). Rat studies have found similar effects with one study showing that rats had more rears, increased center time, and less fecal boli when tested under a 25-W red light versus a 150-W white light (37). Further support is found in a study that looked at a different species of rodent, the Tristam’s jird (Meriones tristrami), where the investigators found significantly higher locomotion and a more even spatial distribution of exploration in the OF in low, red light illumination versus full white light (73). Generally, high light levels suppress locomotor activity in rodents; however, conditions of high light levels may provoke latent anxiety-related behaviors and thus may be helpful in some experimental designs.

5.11. Ambient Noise

Ambient noise levels can also have effects on OFT behavior; however, the extent of the effect has not been clearly defined. In one study, OF activity with varying intensities of white noise at 65 dB, 78 dB, and 94 dB was assessed (50). The authors found that the highest white noise level increased ambulation, while 65 dB and 78 dB white noise had no effect (50). In general, the use of a white noise machine set at a moderate level (65 dB) during the OFT is useful in keeping ambient noise levels consistent during testing and to mask any additional noises caused by the experimenter.

16

Gould, Dao, and Kovacsics

5.12. Time of Testing

Mice are nocturnal animals; however, at most animal facilities, mice are housed on a normal day–night cycle. Only a few studies were identified that investigated the effect that time of day has on OF behavior. One study tested mice in the OFT during the light or dark cycle and found no difference in total distance traveled (74). Another study controlled the circadian cycle by keeping the mice in constant dim green light, entrained the mice with two 15-minutes bright light Zeitgebers, and assessed the behavior of mice in the OFT during subjective day or night but found no differences in activity (75). Generally, although testing mice during the dark, active phase may be ideal and intuitively the most appropriate time to conduct studies, few studies have compared performance in the OFT between light and dark phases. Overall, regardless of the light/dark cycle, it is good practice for mice to be tested in the OFT around the same time of day within a given study.

6. Troubleshooting Below we mention some common problems one may encounter when conducting the OFT. 6.1. Low (or High) Amount of Time in Center

1. Confirm that mice have not experienced stress. 2. Modify lighting. Generally, mice are expected to spend more time in the center under lower light levels (7). Mice can also be tested under red light, certain wavelengths of which are not perceptible by rodents. 3. Modify size of center zone. Depending upon the size of the OF, the center zone is generally between 25–50% of the total area. 4. Use a different size of OF. Generally, a large OF is more sensitive to thigmotaxis.

6.2. Large Standard Error Bars

1. Confirm that all experimental animals have the same testing history and are within a few weeks of age of each other. There is sometimes a sex effect on activity that should be accounted for. 2. Confirm that the studies are sufficiently powered. While a group size of 8–10 mice is sometimes adequate, it is not uncommon for double this number to be required in behavioral experiments.

6.3. Lack of Exploratory Activity

1. There are well-characterized strain effects both in unmedicated and medicated (for example when treated with the stimulant D-amphetamine) mice. For example, BALB/c and

The Open Field Test

17

A/J mice are known to have low baseline levels of activity and do not show hyperlocomotion in response to D-amphetamine (Fig. 1.3B) (76–80). 2. The time of testing can have effects on activity. Increased activity is sometimes observed during the dark phase (active phase for rodents) of the light–dark cycle. 3. Lighting conditions. Confirm that lighting is equal in all areas of the OF. Lower light levels and switching from white to red light will tend to increase activity (7). 4. Ambient environment. Confirm that there are no environmental stressors such as changes in humidity, temperature, or humans within the field of view of the mice.

References 1. Hall C, Ballachey EL. A study of the rat’s behavior in a field: a contribution to method in comparative psychology. University of California Publications in Psychology 1932;6:1–12. 2. Hall C. Drive and emotionality: factors associated with adjustment in the rat. J Comp Psychol 1934;27:89–108. 3. Stanford SC. The Open Field Test: reinventing the wheel. J Psychopharmacol 2007;21(2):134–5. 4. Rodgers RJ. More haste, considerably less speed. J Psychopharmacol 2007;21(2):141–3. 5. Blizard DA, Takahashi A, Galsworthy MJ, Martin B, Koide T. Test standardization in behavioural neuroscience: a response to Stanford. J Psychopharmacol 2007;2(2):136–9. 6. Kopp C, Misslin R, Vogel E, Rettori MC, Delegrange P, Guardiola-Lematre B. Effects of day-length variations on emotional responses toward unfamiliarity. Behav Proc 1997;41:151–7. 7. Walsh RN, Cummins RA. The Open-Field Test: a critical review. Psychol bull 1976;83 (3):482–504. 8. Takahashi A, Kato K, Makino J, Shiroishi T, Koide T. Multivariate analysis of temporal descriptions of open-field behavior in wild-derived mouse strains. Behav Genet 2006;36(5):763–74. 9. Kafkafi N, Elmer GI. Texture of locomotor path: a replicable characterization of a complex behavioral phenotype. Genes brain behav 2005;4(7):431–43. 10. Kafkafi N, Elmer GI. Activity density in the open field: a measure for differentiating the effect of psychostimulants.

11.

12.

13.

14.

15.

16.

17.

18.

19.

Pharmacol Biochem Behav 2005;80(2): 239–49. Kafkafi N, Pagis M, Lipkind D, et al. Darting behavior: a quantitative movement pattern designed for discrimination and replicability in mouse locomotor behavior. Behav Brain Res 2003;142(1–2):193–205. Lipkind D, Sakov A, Kafkafi N, Elmer GI, Benjamini Y, Golani I. New replicable anxiety-related measures of wall vs center behavior of mice in the open field. J Appl Physiol 2004;97(1):347–59. Royce JR. On the construct validity of openfield measures. Psychological bulletin 1977;84(6):1098–106. Broadhurst PL. Determinants of emotionality in the rat. I. situational factors. British J Psychol 1957;49:12–20. Mill J, Galsworthy MJ, Paya-Cano JL, et al. Home-cage activity in heterogeneous stock (HS) mice as a model of baseline activity. Genes brain behav 2002;1(3):166–73. Belzung C. Measuring rodent exploratory behavior. In: Crusio WE, Gerlai RT, eds. Handbook of molecular-genetic techniques for brain and behavior research (techniques in the behavioral and neural sciences). Amsterdam: Elsevier, 1999;739–49. Russell PA. Relationships between exploratory behaviour and fear: a review. Br J Psychol 1973;64(3):417–33. Archer J. Tests for emotionality in rats and mice: a review. Animal behaviour 1973;21 (2):205–35. Prut L, Belzung C. The open field as a paradigm to measure the effects of drugs on

18

20.

21.

22.

23.

24.

25.

26.

27.

28.

29.

30.

31.

32.

Gould, Dao, and Kovacsics anxiety-like behaviors: a review. Eur J Pharmacol 2003;463(1–3):3–33. Dulawa SC, Holick KA, Gundersen B, Hen R. Effects of chronic fluoxetine in animal models of anxiety and depression. Neuropsychopharmacology 2004;29(7): 1321–30. Hall CS. Emotional behavior in the rat I. defacation and urination as measures of individual differences in emotionality. J Comp Psychol 1934;18:385–403. Whimbey AE, Denenberg VH. Two independent behavioral dimensions in open-field performance. J Comp Physiol Psychol 1967;63(3):500–4. Flint J, Corley R, DeFries JC, et al. A simple genetic basis for a complex psychological trait in laboratory mice. Science (NY) 1995;269(5229):1432–5. DeFries JC, Wilson JR, McClearn GE. Open-field behavior in mice: selection response and situational generality. Behav Genet 1970;1(3):195–211. Leppanen PK, Ravaja N, Ewalds-Kvist SB. Twenty-three generations of mice bidirectionally selected for open-field thigmotaxis: selection response and repeated exposure to the open field. Behav Processes 2006;72(1):23–31. Carola V, D’Olimpio F, Brunamonti E, Mangia F, Renzi P. Evaluation of the elevated plus-maze and open-field tests for the assessment of anxiety-related behaviour in inbred mice. Behav Brain Res 2002; 134(1–2):49–57. Dixon LK, Defries JC. Development of openfield behavior in mice: Effects of age and experience. Dev Psychobiol 1968;1(2):100–7. Tang X, Xiao J, Liu X, Sanford LD. Strain differences in the influence of open field exposure on sleep in mice. Behav Brain Res 2004;154(1):137–47. Crabbe JC, Wahlsten D, Dudek BC. Genetics of mouse behavior: interactions with laboratory environment. Science 1999;284(5420):1670–2. Bolivar VJ, Caldarone BJ, Reilly AA, Flaherty L. Habituation of activity in an open field: A survey of inbred strains and F1 hybrids. Behav Genet 2000;30(4):285–93. Homanics GE, Quinlan JJ, Firestone LL. Pharmacologic and behavioral responses of inbred C57BL/6 J and strain 129/SvJ mouse lines. Pharmacol Biochem Behav 1999;63(1):21–6. Voikar V, Polus A, Vasar E, Rauvala H. Long-term individual housing in C57BL/ 6 J and DBA/2 mice: assessment of

33.

34.

35.

36.

37.

38.

39.

40.

41.

42.

43.

44.

behavioral consequences. Genes Brain Behav 2005;4(4):240–52. Voikar V, Koks S, Vasar E, Rauvala H. Strain and gender differences in the behavior of mouse lines commonly used in transgenic studies. Physiol Behav 2001;72(1–2):271–81. Gariepy JL, Rodriguiz RM, Jones BC. Handling, genetic and housing effects on the mouse stress system, dopamine function, and behavior. Pharmacol Biochem Behav 2002;73(1):7–17. Reboucas RC, Schmidek WR. Handling and isolation in three strains of rats affect open field, exploration, hoarding and predation. Physiol Behav 1997;62(5): 1159–64. Schmitt U, Hiemke C. Strain differences in open-field and elevated plus-maze behavior of rats without and with pretest handling. Pharmacol Biochem Behav 1998;59(4): 807–11. Valle FP. Effects of strain, sex, and illumination on open-field behavior of rats. Am J Psychol 1970;83(1):103–11. Crawley JN, Belknap JK, Collins A, et al. Behavioral phenotypes of inbred mouse strains: implications and recommendations for molecular studies. Psychopharmacology 1997;132(2):107–24. Beatrice S, Kvist M, Selander RK. Openfield thigmotaxis during various phases of the reproductive cycle. Scand J Psychol 1994;35(3):220–9. Bronstein PM. Repeated trials with the albino rat in the open field as a function of age and deprivation. J Comp Physiol Psychol 1972;81(1):84–93. Todorovic C, Dimitrijevic M, Stanojevic S, et al. Correlation between age-related changes in open field behavior and plaque forming cell response in DA female rats. Int J Neurosci 2003;113(9):1259–73. Goodrick CL. Free exploration and adaptation within an open field as a function of trials and between-trial-interval for mature-young, mature-old, and senescent Wistar rats. J Gerontol 1971;26(1):58–62. Masur J, Schutz MT, Boerngen R. Gender differences in open-field behavior as a function of age. Dev Psychobiol 1980;13 (2):107–10. Bronstein PM. Open-field behavior of the rat as a function of age: Cross-sectional and longitudinal investigations. Journal of Comparative and Physiological Psychology 1972;80(2):335–41.

The Open Field Test 45. Seliger DL. Effects of age, sex, and brightness of field on open-field behaviors of rats. Percept Mot Skills 1977;45(3 Pt 2):1059–67. 46. Livesey PJ, Egger GJ. Age as a factor in open-field responsiveness in the white rat. J Comp Physiol Psychol 1970;73(1):93–9. 47. Doty BA, Doty LA. Effects of handling at various ages on later open-field behaviour. Can J Psychol 1967;21(6):463–70. 48. Nowak G, Mogilnicka E, Klimek V. Age-dependent day/night variations of alpha 1- and beta-adrenoceptors in the rat cerebral cortex. Physiol Behav 1986;38(1):53–5. 49. Palanza P, Gioiosa L, Parmigiani S. Social stress in mice: gender differences and effects of estrous cycle and social dominance. Physiol Behav 2001;73(3):411–20. 50. Blizard DA. Situational determinants of open-field behaviour in mus musculus. British Journal Psychology 1971;62(2):245–52. 51. Bronstein PM, Wolkoff FD, Levine WJ. Sexrelated differences in rats open-field activity. Behav Biol 1975;13(1):133–8. 52. Elliott BM, Grunberg NE. Effects of social and physical enrichment on open field activity differ in male and female Sprague-Dawley rats. Behav Brain Res 2005;165(2):187–96. 53. Romero RD, Chen WJ. Gender-related response in open-field activity following developmental nicotine exposure in rats. Pharmacol Biochem Behav 2004;78(4):675–81. 54. Izidio GS, Lopes DM, Spricigo L, Jr., Ramos A. Common variations in the pretest environment influence genotypic comparisons in models of anxiety. Genes Brain Behav 2005;4(7):412–9. 55. Alstott J, Timberlake W. Effects of rat sex differences and lighting on locomotor exploration of a circular open field with free-standing central corners and without peripheral walls. Behav Brain Res 2008. 56. Joffe JM, Levine S. Effects of weaning age and adult handling on avoidance conditioning, open-field behavior, and plasma corticosterone of adult rats. Behav Biol 1973;9 (2):235–44. 57. Genaro G, Schmidek WR. The influence of handling and isolation postweaning on open field, exploratory and maternal behavior of female rats. Physiol Behav 2002;75 (5):681–8. 58. Meerlo P, Horvath KM, Nagy GM, Bohus B, Koolhaas JM. The influence of postnatal handling on adult neuroendocrine and behavioural stress reactivity. J Neuroendocrinol 1999;11(12):925–33.

19

59. Vallee M, Mayo W, Dellu F, Le Moal M, Simon H, Maccari S. Prenatal stress induces high anxiety and postnatal handling induces low anxiety in adult offspring: correlation with stress-induced corticosterone secretion. J Neurosci 1997;17(7):2626–36. 60. Brenes JC, Rodriguez O, Fornaguera J. Differential effect of environment enrichment and social isolation on depressive-like behavior, spontaneous activity and serotonin and norepinephrine concentration in prefrontal cortex and ventral striatum. Pharmacol Biochem Behav 2008;89(1):85–93. 61. Ferdman N, Murmu RP, Bock J, Braun K, Leshem M. Weaning age, social isolation, and gender, interact to determine adult explorative and social behavior, and dendritic and spine morphology in prefrontal cortex of rats. Behav Brain Res 2007;180(2):174–82. 62. Lewejohann L, Reinhard C, Schrewe A, et al. Environmental bias? Effects of housing conditions, laboratory environment and experimenter on behavioral tests. Genes Brain Behav 2006;5(1):64–72. 63. Henderson ND. Prior treatment effects on open field behaviour of mice–a genetic analysis. Anim Behav 1967;15(2):364–76. 64. McIlwain KL, Merriweather MY, Yuva-Paylor LA, Paylor R. The use of behavioral test batteries: effects of training history. Physiology & behavior 2001;73(5):705–17. 65. Igarashi E, Takeshita S. Effects of illumination and handling upon rat open field activity. Physiol Behav 1995;57(4):699–703. 66. Russell PA, Williams DI. Effects of repeated testing on rats’ locomotor activity in the openfield. Anim Behav 1973;21(1):109–11. 67. Krsiak M. Measurement of Pharmacological Depression of Exploratory Activity in Mice: A Contibution to the Problem of TimeEconomy and Sensitivity. Psychopharmacologia 1971;21:118–30. 68. Eilam D. Open-field behavior withstands drastic changes in arena size. Behav Brain Res 2003;142(1–2):53–62. 69. Yaski O, Eilam D. How do global and local geometries shape exploratory behavior in rats? Behav Brain Res 2008;187(2):334–42. 70. Kalueff AV, Keisala T, Minasyan A, Kuuslahti M, Tuohimaa P. Temporal stability of novelty exploration in mice exposed to different open field tests. Behav Processes 2006;72(1):104–12. 71. Whitford FW, Jr., Zipf SG. Open-field activity in mice as a function of ceiling height: A genotype-environment interaction. Behav Genet 1975;5(3):275–80.

20

Gould, Dao, and Kovacsics

72. Dixon LK, Van Mayeda D. Effects of floor textures on open-field behavior in selected lines of mice. Behav Genet 1976;6(1):87–92. 73. Avni R, Zadicario P, Eilam D. Exploration in a dark open field: a shift from directional to positional progression and a proposed model of acquiring spatial information. Behav Brain Res 2006;171(2):313–23. 74. Beeler J, Prendergast B, Zhuang X. Low amplitude entrainment of mice and the impact of circadian phase on behavior tests. Physiology & Behavior 2006;87:870–80. 75. Valentinuzzi VS, Buxton OM, Chang AM, et al. Locomotor response to an open field during C57BL/6 J active and inactive phases: differences dependent on conditions of illumination. Physiol Behav 2000;69(3):269–75. 76. Kitahama K, Valatx JL. Strain differences in amphetamine sensitivity in mice. I. A diallel analysis of open field activity. Psychopharmacology 1979;66(2):189–94. 77. Anisman H, Wahlsten D, Kokkinidis L. Effects of d-amphetamine and scopolamine on activity before and after shock in three mouse strains. Pharmacol Biochem Behav 1975;3(5):819–24. 78. Gould TD, O’Donnell KC, Picchini AM, Manji HK. Strain differences in lithium attenuation of d-amphetamine-induced hyperlocomotion: a mouse model for the genetics of clinical response to lithium. Neuropsychopharmacology 2007;32(6):1321–33. 79. DeFries JC. Pleiotropic effects of albinism on open field behaviour in mice. Nature 1969;221(5175):65–6. 80. DeFries JC, Hegmann JP, Weir MW. Openfield behavior in mice: evidence for a major gene effect mediated by the visual system. Science (NY) 1966;154(756):1577–9. 81. Hughes CW. Observer influence on automated open field activity. Physiol Behav 1978;20(4):481–5. 82. Izumi J, Washizuka M, Hayashi-Kuwabara Y, et al. Evidence for a depressive-like state induced by repeated saline injections in Fischer 344 rats. Pharmacol Biochem Behav 1997;57(4):883–8. 83. Whitaker J, Moy SS, Saville BR, et al. The effect of cage size on reproductive

84.

85.

86.

87.

88.

89.

90.

91.

92.

93.

performance and behavior of C57BL/6 mice. Lab Anim (NY) 2007;36(10):32–9. Jahkel M, Rilke O, Koch R, Oehler J. Open field locomotion and neurotransmission in mice evaluated by principal component factor analysis-effects of housing condition, individual activity disposition and psychotropic drugs. Prog Neuropsychopharmacol Biol Psychiatry 2000;24(1):61–84. Brenes Saenz JC, Villagra OR, Fornaguera Trias J. Factor analysis of Forced Swimming test, Sucrose Preference test and Open Field test on enriched, social and isolated reared rats. Behav Brain Res 2006;169(1):57–65. Bors DA, Forrin B. The effects of postweaning environment, biological dam, and nursing dam on feeding neophobia, open field activity, and learning. Can J Exp Psychol 1996;50(2):197–204. Eilam D. Locomotor activity in common spiny mice (Acomys cahirinuse): the effect of light and environmental complexity. BMC Ecol 2004;4(1):16. Krsiak M, Steinberg H, Stolerman IP. Uses and limitations of photocell activity cages for assessing effects of drugs. Psychopharmacologia 1970;17(3):258–74. Kvist SB, Selander RK. A qualitative aspect of learning-sensitive open field ambulation in mice. Scand J Psychol 1992;33(2):97–107. Roth KA, Katz RJ. Stress, behavioral arousal, and open field activity–a reexamination of emotionality in the rat. Neurosci Biobehav Rev 1979;3(4):247–63. Dishman RK, Dunn AL, Youngstedt SD, et al. Increased open field locomotion and decreased striatal GABAA binding after activity wheel running. Physiol Behav 1996;60(3):699–705. Whishaw IQ, Gharbawie OA, Clark BJ, Lehmann H. The exploratory of home bases in mice (C57BL/6) influenced behavior of rats in an open environment optimizes security. Behav Brain Res 2006;171(2):230–9. Clark BJ, Hamilton DA, Whishaw IQ. Motor activity (exploration) and formation by visual and tactile cues: modification of movement distribution, distance, location, and speed. Physiol Behav 2006;87(4):805–16.

Chapter 2 Analysis of Grooming Behavior and Its Utility in Studying Animal Stress, Anxiety, and Depression Amanda N. Smolinsky, Carisa L. Bergner, Justin L. LaPorte, and Allan V. Kalueff Abstract In rodents, grooming is a complex and ethologically rich behavior, sensitive to stress and various genetic and pharmacological manipulations, all of which may alter its gross activity and patterning. Observational analysis of grooming activity and its microstructure may serve as a useful measure of stress and anxiety in both wild and laboratory animals. Few studies have looked at grooming behavior more than cursorily, though in-depth analysis of the behavior would immensely benefit fields utilizing rodent research. Here, we present a qualitative approach to grooming activity and patterning analysis in mice, which provides insight into the effects of stress, anxiety, and depression on this behavioral domain. The method involves quantification of the transitions between different stages of grooming, the percentages of incorrect or incomplete grooming bouts, as well as the regional distribution of grooming activity. Using grooming patterning as a behavioral endpoint, this approach permits assessment of stress levels of individual animals, allows identification of grooming phenotypes in various mouse strains, and has vast implications in biological psychiatry, including psychopharmacology, genetics, neurophysiology, and experimental modeling of affective disorders. Key words: Grooming behavior, stress, anxiety, depression, behavioral organization (sequencing), animal experimental and genetic models, neuropsychiatric disorders.

1. Background and Historical Overview Grooming is an important and evolutionarily ancient behavior observed across many animal taxa (1–4). Beyond the primary purpose of hygiene and caring for the body surface, grooming serves a variety of other functions, including stimulation of the skin, thermoregulation, chemo-communication, social interaction, de-arousal, and stress reduction (1, 4–7). In both wild and laboratory rodents, this behavior constitutes 15–50% of waking T.D. Gould (ed.), Mood and Anxiety Related Phenotypes in Mice, Neuromethods 42, DOI 10.1007/978-1-60761-303-9_2, ª Humana Press, a part of Springer Science+Business Media, LLC 2009

21

22

Smolinsky et al.

time and may be triggered by novelty, swimming, pain, exposure to predators, or sexual behavior (for review see (8, 9)). Genetic factors play an important role in the regulation of rodent grooming, and various genetic manipulations have been reported to produce robust grooming phenotypes in mice (6, 10–14). Rodent grooming is a complex patterned behavior, which generally proceeds in a cephalocaudal direction (3, 15, 16). The behavioral sequence (Fig. 2.1) usually begins with licking of the paws, followed by washing the nose and face, the head, the body, the legs, and finally washing and licking the tail and genitals (3, 15, 16). Stereotyped grooming behaviors are clearly centrally controlled (rather than driven by peripheral sensory input), since mice with amputated front paws continued to make facial grooming gestures with their stumps (5). Regulation of grooming behavior is mediated by multiple brain regions (especially the basal ganglia and hypothalamus) (15–18), as well as by various endogenous agents (neuromediators (5, 16, 19), hormones (5, 20–23)), and psychotropic drugs (12, 19, 24–27). Given the robust nature of grooming behavior in animal phenotypes (2, 9, 28, 29), it is logical to expect that alterations in this domain will be seen in experimental mouse models of stress, anxiety, and depression.

Fig. 2.1. Prototypical syntactic grooming chain pattern in mice (Prof. K. Berridge, with permission). Phase I: series of ellipse-shaped strokes tightly around the nose (paw, nose grooming). Phase II: series of unilateral strokes (each made by one paw) that reach up the mystacial vibrissae to below the eye (face grooming). Phase III: series of bilateral strokes made by both paws simultaneously. Paws reach back and upwards, ascending usually high enough to pass over the ears (head grooming). Phase IV: body licking, preceded by postural cephalocaudal transition from paw/head grooming to body grooming.

Neurophenotyping Animal Grooming Behavior

23

Despite the complexity and importance of grooming in mice, many studies that include grooming observations have dealt with this behavior only cursorily. For example, some analyses include only cumulative grooming scores, or have lumped grooming into ‘‘overall activity scores’’ (for review see (8, 9, 30)). Furthermore, traditional measures of grooming often include only time to onset and/or the number and duration of bouts (Table 2.1), but ignore the unique, data-dense feature of this behavior – its complex microstructure (27, 29, 30).

Table 2.1 Methodological approaches to mouse grooming phenotyping Global assessment l

Coat state (40–42)

General cumulative measures The latency to onset, the duration, and the number of grooming episodes (bouts) (28, 30) Temporal patterning (e.g., per-minute distribution) of grooming duration and frequency may be recorded to examine habituation of this behavior l The following patterns can be recorded for each bout: paw licking; nose/face grooming; head washing; body and leg grooming/scratching; tail/genitals grooming l Additional cumulative indices: the average duration of a single grooming bout, total number of transitions between grooming stages, and average number of transitions per bout (8, 9) l l

Patterning (sequencing) l

The percentages of incorrect transitions, as well as interrupted and incomplete grooming bouts (8, 9, 30)

Regional distribution of grooming Can be assessed as directed to the following five anatomic areas: forepaws, head, body, hind legs, and tail/genitals l Rostral grooming includes forepaw (preliminary rostral grooming) and head grooming. Body, legs, and tail/genital grooming can be considered as caudal grooming l Each bout can be categorized as being directed to (i) multiple regions or (ii) a single region, and the percentages of grooming bouts and of time spent grooming can be calculated for both categories (6, 8, 9, 28, 30) l

Additional useful indices of grooming l l

Probability of chain initiation (frequency of chain initiation per minute of grooming time) Probability of pattern completion once initiated [these indices were not discussed here, but see (8, 15, 16) for details and useful background information]

Representing a typical displacement behavior, grooming is often seen in animal models of stress and anxiety (19, 28, 31, 32), leading to a long-standing view of grooming as a mere anxiogenic response (25, 33–35). Some data, however, indicate that higher stress or anxiety in animals does not necessarily translate

24

Smolinsky et al.

into their increased grooming activity (27, 36–38). Such oversimplification of complex behavior has also been recently challenged by more detailed analyses of animal grooming phenotypes. Indeed, since grooming activity in rodents is increased under conditions of both high and low stress, the amount of grooming may not be a reliable indicator of animal anxiety (8, 9, 27–30). However, unlike quantitative measures, the ‘‘quality’’ of grooming – its sequencing (Table 2.1) – varies substantially according to the degree of stress experienced (8, 27, 30). Specifically, low-stress ‘‘comfort’’ grooming occurs spontaneously as a transition between rest and activity, and generally proceeds in a ‘‘relaxed’’ uninterrupted manner following the cephalocaudal rule (Fig. 2.1). Conversely, stress-evoked grooming is generally characterized by frequent bouts of interrupted ‘‘chaotic’’ activity that defies the cephalocaudal rule, and may serve as a way to cope with fear or anxiety (3, 30). Additionally, several manipulations (including brain lesions, psychotropic drugs, and genetic mutations) alter the behavioral microstructure of grooming (8), sometimes without affecting the cumulative amount of grooming activity (27). Therefore, traditional observations of grooming that focus only on quantitative measures of its activity (Table 2.1) are insufficient for proper interpretation of stress data, as they may provide ambiguous results (9, 28, 30). Alterations in the rodent depression-like states have also been shown to affect animal grooming (39–42). However, unlike the ‘‘acute’’ nature of anxiety-induced grooming responses, the effects of depression on grooming are delayed and somewhat less obvious. Therefore, the role of grooming as a behavioral marker of depression has been much less studied, compared to the large body of literature on grooming responses to anxiety (see above). Do depressed animals groom more or less? Is the patterning of rodent grooming affected in depressed animals? Do these behavioral alterations in mouse grooming resemble clinical endophenotypes seen in depressed patients? These are the important questions that are currently under investigation, as they are only partially answered by the available literature on this topic, which will be briefly discussed further. Because grooming represents only one domain, other behavioral endpoints and domains should be considered while performing an in-depth ethological analysis. However, the ability of grooming patterning to reflect (and indirectly measure) stress in mice has numerous potential applications. These include gauging the degree of stress induced by various tests, behavioral phenotyping of mutant or transgenic strains, and testing of psychotropic drugs for their ability to alter anxiety or depression levels (9, 19, 30, 43). In addition, it

Neurophenotyping Animal Grooming Behavior

25

may assist in interpreting various non-grooming behaviors, and detect motor/coordination anomalies and age-related behavioral changes. Furthermore, understanding ethological patterning of grooming also has implications for developing better mouse models of human behavioral disorders (such as obsessive-compulsive disorder, Rett or Tourette’s syndrome), and for decoding normal human nervous behaviors elicited by everyday stress (7, 8, 16, 44). This chapter will provide a detailed up-to-date overview of how researchers can assess mouse self-grooming behavior, and apply their findings to understand animal and human affective disorders.

2. Equipment, Materials, and Setup

Although various inbred, selectively bred, and genetically modified (mutant or transgenic) mice may be used to assess grooming (28, 32, 43, 45, 46), in behavioral experiments, it is important to select the appropriate laboratory mouse strain. While some searchable online databases (such as Mouse Genome Informatics, MGI) may provide appropriate genetic models for studying mouse grooming, note that the activity fluctuates between strains and may be confounded by strainspecific phenotypes (28, 30) and other factors alike (see further). In order to analyze animal grooming activity, transparent observation apparatuses (such as small plexiglas or glass boxes and cylinders) are generally utilized. For mouse studies, the dimensions of the apparatus may be 20  20  30cm (although other dimensions may be used, depending on mouse activity and anxiety levels). Between sessions, it is necessary to remove olfactory cues in the apparatuses by thoroughly cleansing the equipment (e.g., with a 30% ethanol solution). Researchers may also use various anxiolytic, anxiogenic, antidepressant, psychostimulant, and other psychotropic drugs to analyze their effects on grooming behaviors in mice. Common routes of injection include systemic [intraperitoneal (i.p.), intramuscular (i.m.), intravenous (i.v.), per oral (p.o.), subcutaneous (s.c.)] and local [intracerebral (i.c.) or intracerebroventricular (i.c.v)]. Route of administration, dose, and pre-treatment time generally vary depending on the drug and strain sensitivity. Importantly, all experimental procedures (including handling, housing, husbandry, and drug treatment) must be conducted in accordance with National and Institutional Guidelines for the Care and Use of Laboratory Animals.

26

Smolinsky et al.

3. Procedures 3.1. Coat State

Coat state assessment is the simplest method to evaluate animal grooming activity (41, 47). After removing animals from their homecages, the state of the coat of eight separate body parts such as head, neck, forepaws, dorsal coat, ventral coat, hindlegs, tail, and genital region of each individual mouse may be inspected visually and recorded systematically (40–42). For example, a score of 0 could be attributed to a coat in good form, and a score of 1 could be given to a dirty or disheveled coat. The resulting score (to be compared between different experimental groups) will represent the average for all body areas. Other similar scales may be used consistently within the laboratory to record the condition of the coat. Although this approach cannot be used to study acute effects of stress and anxiety, it has been shown that mouse coat state generally correlates with the level of experimental depression. For example, chronically stressed depressed mice generally display poor coat status, whereas antidepressant treatments tend to reverse this phenotype (40–42). Thus, the coat state assessment provides a gross method of grooming analysis, and may reveal some very overt differences in animal behavior. Nevertheless, this method may lack ethological sensitivity, and therefore may need to be complemented with more sophisticated analyses of animal grooming that will be discussed further.

3.2. Acute StressEvoked Grooming

It is important to distinguish two forms of self-grooming in rodents: spontaneous (stress-evoked) and artificial grooming. To encourage stress-evoked grooming, a typical experiment may include exposure to novelty, such as a novel observation box, for 5–10 min. To ensure proper acclimation to the experimental room, it is recommended that rodents are transferred to the room at least 1 h before testing. The mouse may then be removed from the cage and presented with an anxiogenic stressor to stimulate grooming activity. In addition to novelty stress, researchers may also use stronger stressors, such as a brief pre-exposing the mouse to a bright light, conspecific, a predator (e.g., rat or cat) or its odor. In general, this procedure enables fast and reliable detection of alterations in mouse grooming related to anxiety domain, and may be a useful tool in basic research of emotionality.

3.3. Chronic StressEvoked Grooming

While chronic mild stress has been shown to grooming (40–42, 48) (but see (39)), stronger as olfactobulbectomy or peripheral anosmia), chronically, produce pronounced activation

reduce animal stressors (such when applied of stereotypic

Neurophenotyping Animal Grooming Behavior

27

grooming activity. This ‘‘pathological’’ grooming is usually focused on a specific body area, and is accompanied by severe depression-like behaviors including anhedonia, hypoactivity, aggression, and self-aggression (49–52). Overall, these procedures may be particularly relevant to modeling severe protracted depression in animals, and are generally in line with clinical data showing overall increases in stereotypic behavior (e.g., grooming disorders, hair-pulling) in depressed patients (53, 54). However, more research is needed to understand whether animal depression produces consistent alterations in grooming patterning. 3.4. Artificial Grooming

Artificially induced grooming can be stimulated by allowing the mouse to swim or by smearing the animal with food (8). The splash test is another method to evoke ‘‘artificial’’ grooming in mice. For this, a sucrose solution (e.g., 10%) may be squirted onto the mice in the dorsal region while they remain in their homecages (40–42), and grooming activity measures (Table 2.1) can be recorded for 5 min after the vaporization of the solution. Misting with water (e.g., using fine water spray) is also an easy and reliable method to evoke artificial grooming behavior (6, 30, 55), and is widely used in neurobehavioral experiments. Since spontaneous and artificial grooming represent two different forms of this behavior, abnormalities in one type do not necessarily imply deficits in another form of grooming. Thus, a parallel assessment of the two types of grooming is necessary for a more careful characterization of animal behavioral phenotypes (6, 30, 56).

3.5. Hybridizing Behavioral Protocols

In addition to the above-mentioned procedures, researchers may consider combining several behavioral tests into a ‘‘smart battery’’ that simultaneously examines anxiety, depression, and grooming domains. For example, an initial 5-min open-field testing (to assess baseline anxiety and spontaneous novelty-induced grooming behavior) may be followed by the Porsolt’s forced-swim test that evaluates depression-related immobility or despair (57). In order to maximize the number of behavioral endpoints and domains per experiment, immediately after the forced-swim test, researchers may place the mice in an observation cylinder (e.g., for 5 min) to investigate artificial, swim-induced grooming (43). Comparing the patterning and activity of the artificial post-swim grooming with the spontaneous (novelty-evoked) pre-swim grooming could lead to interesting findings regarding the animal’s grooming phenotypes. In some instances, mice may also have a fatigability phenotype (43, 57) that should be discriminated from grooming behaviors. Fatigability will often interfere with mouse grooming activities, could confound data, and therefore needs to be carefully dissected from grooming domains (see further).

28

Smolinsky et al.

3.6. Time Required

To minimize initial procedure-related anxiety, researchers may choose to gently handle naı¨ve mice 5 min/mouse/day for 3–4 days prior to the grooming experiments. Acclimation to the procedure room requires at least 1 h. The time required for grooming assessment protocols varies depending on the test battery used (see above), the number of animals per group and the number of experimental groups, and based on mouse grooming activity levels (see troubleshooting). In general, grooming behavior assessment will last 5–10 min per animal. Depending on the amount of grooming and other behavioral data collected, analysis could take between 2 and 4 days. It is advised that researchers maintain a 7-day minimum acclimation period between tests.

3.7. Data Analysis

To analyze the data, researchers may generally use the Mann– Whitney U-test for comparing two groups (parametric Student’s t-test may be used if data are normally distributed) or an analysis of variance (ANOVA) for multiple groups, followed by a post hoc test. More complex designs, such as one-way ANOVA with repeated measures (time) or n-way ANOVA (additional factors: treatment, genotype, stress, sex, etc.), can also be used in grooming studies.

4. Experimental Variables The present protocol, largely based on the method called the Grooming Analysis Algorithm (GAA) (9), provides a highthroughput approach to analyze mouse grooming activity and microstructure. Several indices of grooming can be recorded as generalized measures, including coat state, latency to onset, cumulative duration of grooming, and number of bouts (grooming episodes); see Table 2.1 for details. A shorter latency period to begin grooming, a longer duration of grooming, and more bouts may be behavioral markers for stress in mice (but see the discussion of validity of cumulative measures above). Calculating the average duration of a single bout (total time grooming/number of bouts), the total number of transitions between bouts, and the average number of transitions per bout (total number of transitions per bout/number of bouts) will also help provide necessary data in determining the level of stress of the mice. To accurately evaluate grooming bout patterns, the researchers may develop a standardized scale to represent specific grooming activity and use it consistently within each laboratory. A typical scale may be as follows (Table 2.1): no grooming (0),

Neurophenotyping Animal Grooming Behavior

29

paw licking (1), nose, face, and head wash, characterized by pawing nose and semicircular strokes of the head and ears (2), body grooming, including body fur licking and scratching with hind paws (3), leg licking (4), and tail or genital grooming (5) (8). However, researchers may modify this scale to suit their individual needs by including additional strain-specific grooming behaviors of interest, or by simplifying this scale for better detectability. A ‘‘correct’’ bout is cephalocaudal in direction and follows a (0-1) (1-2) (2-3) (3-4) (4-5) (5-0) pattern of correct transitions (Table 2.1). An ‘‘incorrect’’ transition can vary from the model in one of four ways: an aborted or prematurely terminated bout (2-0, 4-0), a skipped transition (1-3, 2-5), a reversed bout (4-3, 5-2), or an incorrectly initiated bout (0-2, 0-5). A ‘‘complete’’ bout consists of a strict (0-1-2-3-4-5-0) sequence and any other pattern is considered incomplete. Frequently, researchers will notice grooming interruptions. Any sequence that contains at least one interruption is deemed ‘‘interrupted.’’ However, an interruption of 6 s or greater is judged to be an entirely separate bout (8, 9). Again, maintaining a consistent standard of all defined behaviors and criteria used within each laboratory is strongly recommended to avoid confusion and poor validity of data. With this system, researchers may assess the three primary ethological measures of grooming patterning – the percentage of incorrect transitions, interrupted bouts, and incomplete bouts. In addition, researchers may calculate the duration of correct versus incorrect patterns, the number of interruptions during bouts, and the duration of complete versus incomplete bouts. It is also useful to investigate the regional distribution of grooming patterning, as highly stressed mice spend significantly more time grooming rostral areas than caudal (8, 9). For example, data may be collected based on five anatomic areas (forepaws, head, body, hind legs, and tail/genitals) or simply a rostral (forepaw and head) versus caudal (body, legs, and tail/genitals) partition. Again, this distribution criterion may be modified or simplified to fit the individual needs of the researcher; however, maintaining a consistent standard within the laboratory will help prevent inaccuracies. Researchers may also classify each grooming bout as being directed to a single anatomic region or multiple regions, and calculate the percentage of grooming bouts and the percentage of time spent grooming for each category. Furthermore, the percentage of total grooming patterns, the percentage of time spent grooming, and the number of interruptions for each anatomic area may be assessed. Stressed anxious mice generally tend to display a greater number of interruptions, especially in rostral areas, when licking the forepaws or washing the face.

30

Smolinsky et al.

5. Typical/ Anticipated Results A typical experiment assessing mouse grooming sensitivity to different pharmacological manipulations is presented in Fig. 2.2. In this study, anxiolytic drug diazepam normalized grooming patterning by lowering the percentage of incorrect transitions and interrupted bouts. In contrast, an anxiogenic substance pentylenetetrazole typically increased these indices and also increased the duration of grooming (see (27) for details). These data parallel recent data in rats showing that their grooming sequencing is sensitive to different classes of psychotropic drugs (19, 24, 26).

Fig. 2.2. Sensitivity of mouse grooming behaviors to anxiolytic and anxiogenic drugs (27). Anxiolytic diazepam lowers the percentages of incorrect transitions and incorrect bouts, while anxiogenic drug pentylenetetrazole increases duration of grooming, with higher percentages of incorrect transitions and interrupted bouts. (*P < 0.05, U-test).

Another typical experiment examining the regional distribution of mouse grooming is shown in Fig. 2.3, using the vitamin D receptor knockout mice as a model (6). Note the difference between grooming behavior in the wild type and ‘‘anxious’’ mutant mice (i.e., more rostral grooming, less caudal grooming). Also notice the variances between the two different types of grooming: spontaneous (novelty-induced) and artificial (swiminduced) grooming mentioned above. Overall, while spontaneous grooming showed sensitivity to genetic differences, in this experimental model, the ‘‘more rigid’’ swim-induced grooming was not altered between the genotypes. It is expected that analyses of mouse grooming behavior using microstructure-oriented approaches (Table 2.1) may be useful in examining rodent stress levels in experimental conditions (6, 8, 9, 24, 26, 29, 30). Since grooming patterning in mice appears to be sensitive to stressful manipulations and could

Neurophenotyping Animal Grooming Behavior

31

Fig. 2.3. Regional distribution of grooming patterns (of total taken as 100%) in the wild type and the vitamin D receptor knockout mice (6). In the spontaneous novelty induced grooming test, the knockout mice displayed significantly higher percentages of forepaw, head and hind leg grooming, also showing less caudal (tail, genital) grooming than wildtype mice (*P < 0.05, U-test). Artificial swim-induced grooming showed no genotype differences between the groups.

serve as an additional measure of stress and anxiety, emotionality-related behaviors in mice could be investigated and assessed more accurately. Additionally, when paired with in-depth assessment of non-grooming phenotypes, grooming analyses could further confirm or invalidate unclear results. New reliable methods for phenotyping mouse behavior could be formulated based on sensitivity of grooming analysis to alterations in patterning between various strains of mice. Researchers would also have a new useful criterion for choosing appropriate experimental subjects for their studies, since grooming (in addition to other specific phenotypes) could aid in the correct classification of novel strains of mutant or transgenic mice. Finally, mouse grooming behavior may also have a significant application in the study of human brain disorders (10, 13, 44, 46). Likewise, brain lesion studies, particularly those focusing on basal ganglia motor control and patterned behavior regulation, could also lead to interesting neurobehavioral mouse models based on grooming activity and its patterning (8).

6. Troubleshooting Several practical recommendations, summarized here, may help the researchers to obtain more reliable and reproducible behavioral data.

32

Smolinsky et al.

1. If mice display abnormally high or low levels of grooming, it may be a strain-specific phenomenon (28). While it is encouraged to further investigate strain differences, the researchers may need to re-assess the strain’s suitability for their experiment. 2. Ameliorating the environmental and testing conditions would also aid in normalizing mice behaviors. This includes proper handling, a better enrichment, the use of fewer and/or less stressful tests, and improving husbandry (8). If grooming activity remains too low, extending the tests for 5–10 more minutes may be a good practical solution, as it minimizes the initial anxiety and disinhibits grooming activity. 3. Factors such as altered skin/pain sensitivity and motor coordination deficits can be very pronounced in some mice. These factors may non-specifically alter animal behavior in a way that could be misinterpreted as altered grooming phenotype. To address this possibility and rule out non-specific factors, a careful examination of mouse neurological and sensory phenotypes is recommended. 4. When assessing the coat state, note that some mouse strains are poor (e.g., BALB/c mice) or excellent (e.g., A/J mice) groomers regardless of the level of their stress. Therefore, it is important to understand that, due to floor or ceiling effects, not every strain will produce reliable results in this test. Likewise, for socially housed mice, hetero-grooming may compensate for poor self-grooming, so the coat will have a clean appearance. To rule out this possibility, single housing may be employed (but with caution, since isolation itself may also have some behavioral effects). 5. When using novelty-induced grooming protocol, the size of arena (see above) is a very important factor. Since strain differences in anxiety and activity may affect all other behaviors, including grooming, the general rule is that the observation box needs to be relatively small. In a smaller box, the animals become familiar with the novelty faster, and this may help quickly reduce anxiety, enabling the mice to better ‘‘reveal’’ their grooming phenotypes. 6. Since it can be difficult to accurately detect exact grooming behaviors in mice, a frame-by-frame analysis with an event recorder is recommended. For example, without intense scrutiny of the animal’s behavior, a stroke could easily be overlooked and the sequence could be misinterpreted. Video recording of all behavioral experiments is strongly recommended for more accurate grooming phenotyping. 7. Mice often engage in context-specific grooming (e.g., genital licking during mating, wound-inflicted body scratching) and, therefore, separate documentation of these instances may be

Neurophenotyping Animal Grooming Behavior

33

necessary (8). Since mice may partake in both self-grooming and hetero-grooming behaviors, the researchers are advised to analyze these categories carefully. For example, in some mouse strains, hetero-grooming may naturally occur more frequently or for a longer duration, and consequently, selfgrooming will be reciprocally decreased, which could be interpreted incorrectly as a stress-related response. It is useful to consider each occurrence separately, to avoid confounding data (e.g., reciprocal decrease in self-grooming in mice with abnormally increased hetero-grooming). 8. Rare ‘‘atypical’’ forms of grooming may also be difficult to categorize (8). For example, some mice may partake in peculiar ‘‘pre-grooming’’ or ‘‘vertical grooming’’ (28) behaviors that could also lead to data misinterpretation. Thus, a careful analysis of both common and rare grooming activities is a key for accurate data collection and behavioral interpretation. In some other cases, grooming behavior needs to be separated from barbering (behavior-associated hair loss) phenotypes. This interesting rodent behavior will not be discussed here, but readers are encouraged to peruse recent works on this topic [e.g., (10, 58–61)]. Although separating self-grooming from hetero-barbering may be easy in most cases, self-grooming and self-barbering behaviors may sometimes be very similar. 9. In some instances, when using swim-evoked grooming models, the separation of swim test effects on artificial grooming per se and fatigability is necessary. To help differentiate between the two factors, researchers may shorten the swim test. For example, a 5-min swim test could potentially affect both artificial grooming and fatigability, whereas a very short 10-s swim session will only induce artificial grooming. Alternatively, using a different type of inductor that cannot evoke fatigue, such as smearing the animal with food, may be recommended to stimulate the artificial grooming. 10. Since the procedure that induces grooming may represent a stress for the mice, especially for some anxious mouse strains, it may be necessary to separate the procedure stress effects on grooming from those produced by artificial grooming inductors. Although this is a difficult task, some behavioral methods may enable dissection of spontaneous from artificial grooming. For example, while novelty stressevoked grooming will habituate, artificial grooming is unlikely to decrease with repeated exposure. Likewise, artificial grooming microstructure will generally be more rigid and inflexible, compared to the spontaneous stressevoked grooming.

34

Smolinsky et al.

7. Conclusion Overall, there are clear benefits of in-depth analyses of mouse grooming activity and patterning in neurobiological experiments. First, it allows assessment of strain differences in grooming behaviors per se. Second, grooming activity and its sequencing may reflect fine differences in other domains, such as activity, motor patterning, anxiety, and depression. Finally, given the sensitivity of mouse grooming and its sequencing to various pharmacological and physiological manipulations, ethologically oriented analysis of grooming may be used extensively in pharmacogenetics and neurophysiology (e.g., for testing psychotropic drugs in different strains or for dissection of brain substrates involved in the regulation of behaviors). On the whole, behavioral analysis of mouse grooming can be a rich source of information in neuroscience and the biological psychiatry of anxiety and depression. Providing more comprehensive coverage of mouse behavioral phenotypes and offering ideas on their grooming peculiarities may assist researchers in correct data interpretation and selection of appropriate mouse models for their studies.

Acknowledgments This research was supported by the NARSAD YI Award to AVK.

References 1. Sachs BD. The development of grooming and its expression in adult animals. Ann N Y Acad Sci 1988;525:1–17. 2. Bolles RC. Grooming behavior in the rat. J Comp Physiol Psychol 1960;53:306–10. 3. Fentress JC. Expressive contexts, fine structure, and central mediation of rodent grooming. Ann N Y Acad Sci 1988;525:18–26. 4. Terry RL. Primate grooming as a tension reduction mechanism. J Psychol 1970;76:129–36. 5. Spruijt BM, van Hooff JA, Gispen WH. Ethology and neurobiology of grooming behavior. Physiol Rev 1992;72:825–52. 6. Kalueff AV, Lou YR, Laaksi I, Tuohimaa P. Abnormal behavioral organization of grooming in mice lacking the vitamin D receptor gene. J Neurogenet 2005;19:1–24.

7. Colbern DL, Gispen WH. Neural mechanisms and biological significance of grooming behavior. In: Colbern DL, Gispen WH, eds. Ann N Y Acad Sci. New York; 1988:Preface. 8. Kalueff AV, Aldridge JW, LaPorte JL, Murphy DL, Tuohimaa P. Analyzing grooming microstructure in neurobehavioral experiments. Nat Protoc 2007;2: 2538–44. 9. Kalueff AV, Tuohimaa P. Grooming analysis algorithm for neurobehavioural stress research. Brain Res Brain Res Protoc 2004; 13:151–8. 10. Hill RA, McInnes KJ, Gong EC, Jones ME, Simpson ER, Boon WC. Estrogen deficient male mice develop compulsive behavior. Biol Psychiatry 2007;61:359–66. 11. Hyman SE. Neuroscience: obsessed with grooming. Nature 2007;448:871–2.

Neurophenotyping Animal Grooming Behavior 12. Rupniak NM, Carlson EJ, Webb JK, et al. Comparison of the phenotype of NK1R–/– mice with pharmacological blockade of the substance P (NK1 ) receptor in assays for antidepressant and anxiolytic drugs. Behav Pharmacol 2001;12:497–508. 13. Campbell KM, de Lecea L, Severynse DM, et al. OCD-Like behaviors caused by a neuropotentiating transgene targeted to cortical and limbic D1+ neurons. J Neurosci 1999;19:5044–53. 14. Clement Y, Adelbrecht C, Martin B, Chapouthier G. Association of autosomal loci with the grooming activity in mice observed in open-field. Life Sci 1994;55:1725–34. 15. Aldridge JW, Berridge KC, Rosen AR. Basal ganglia neural mechanisms of natural movement sequences. Can J Physiol Pharmacol 2004;82:732–9. 16. Berridge KC, Aldridge JW, Houchard KR, Zhuang X. Sequential super-stereotypy of an instinctive fixed action pattern in hyperdopaminergic mutant mice: a model of obsessive compulsive disorder and Tourette’s. BMC Biol 2005;3:1–16. 17. Roeling TA, Veening JG, Peters JP, Vermelis ME, Nieuwenhuys R. Efferent connections of the hypothalamic ‘‘grooming area’’ in the rat. Neuroscience 1993;56:199–225. 18. Kruk MR, Westphal KG, Van Erp AM, et al. The hypothalamus: cross-roads of endocrine and behavioural regulation in grooming and aggression. Neurosci Biobehav Rev 1998;23:163–77. 19. Barros HM, Tannhauser SL, Tannhauser MA, Tannhauser M. The effects of GABAergic drugs on grooming behaviour in the open field. Pharmacol Toxicol 1994;74:339–44. 20. Bertolini A, Poggioli R, Vergoni AV. Crossspecies comparison of the ACTH-induced behavioral syndrome. Ann N Y Acad Sci 1988;525:114–29. 21. Dunn AJ. Studies on the neurochemical mechanisms and significance of ACTHinduced grooming. Ann N Y Acad Sci 1988;525:150–68. 22. Dunn AJ, Berridge CW, Lai YI, Yachabach TL. CRF-induced excessive grooming behavior in rats and mice. Peptides 1987;8:841–4. 23. Ukai M, Toyoshi T, Kameyama T. Multidimensional analysis of behavior in mice treated with the delta opioid agonists DADL (D-Ala2-D-Leu5-enkephalin) and DPLPE (D-Pen2-L-Pen5-enkephalin). Neuropharmacology 1989;28:1033–9. 24. Audet MC, Goulet S, Dore FY. Repeated subchronic exposure to phencyclidine elicits

25.

26.

27.

28.

29.

30.

31.

32.

33.

34.

35.

36.

35

excessive atypical grooming in rats. Behav Brain Res 2006;167:103–10. Choleris E, Thomas AW, Kavaliers M, Prato FS. A detailed ethological analysis of the mouse open field test: effects of diazepam, chlordiazepoxide and an extremely low frequency pulsed magnetic field. Neurosci Biobehav Rev 2001;25:235–60. Enginar N, Hatipoglu I, Firtina M. Evaluation of the acute effects of amitriptyline and fluoxetine on anxiety using grooming analysis algorithm in rats. Pharmacol Biochem Behav 2008;89:450–5. Kalueff AV, Tuohimaa P. Mouse grooming microstructure is a reliable anxiety marker bidirectionally sensitive to GABAergic drugs. Eur J Pharmacol 2005;508:147–53. Kalueff AV, Tuohimaa P. Contrasting grooming phenotypes in three mouse strains markedly different in anxiety and activity (129S1, BALB/c and NMRI). Behav Brain Res 2005;160:1–10. Kalueff AV, Tuohimaa P. The grooming analysis algorithm discriminates between different levels of anxiety in rats: potential utility for neurobehavioural stress research. J Neurosci Methods 2005;143:169–77. Kalueff AV, Tuohimaa P. Contrasting grooming phenotypes in C57Bl/6 and 129S1/SvImJ mice. Brain Res 2004;1028:75–82. File SE, Mabbutt PS, Walker JH. Comparison of adaptive responses in familiar and novel environments: modulatory factors. Ann N Y Acad Sci 1988;525:69–79. Leppanen PK, Ravaja N, Ewalds-Kvist SB. Twenty-three generations of mice bidirectionally selected for open-field thigmotaxis: selection response and repeated exposure to the open field. Behav Processes 2006;72:23–31. Nosek K, Dennis K, Andrus BM, et al. Context and strain-dependent behavioral response to stress. Behav Brain Funct 2008;4:23. Sousa FC, Melo CT, Monteiro AP, et al. Antianxiety and antidepressant effects of riparin III from Aniba riparia (Nees) Mez (Lauraceae) in mice. Pharmacol Biochem Behav 2004;78:27–33. Ferre P, Fernandez-Teruel A, Escorihuela RM, et al. Behavior of the Roman/Verh high- and low-avoidance rat lines in anxiety tests: relationship with defecation and selfgrooming. Physiol Behav 1995;58:1209–13. Bouwknecht JA, Spiga F, Staub DR, Hale MW, Shekhar A, Lowry CA. Differential

36

37.

38.

39.

40.

41.

42.

43.

44.

45.

46.

47.

48.

Smolinsky et al. effects of exposure to low-light or high-light open-field on anxiety-related behaviors: relationship to c-Fos expression in serotonergic and non-serotonergic neurons in the dorsal raphe nucleus. Brain Res Bull 2007;72:32–43. Kalueff AV, Lou YR, Laaksi I, Tuohimaa P. Increased grooming behavior in mice lacking vitamin D receptors. Physiol Behav 2004;82:405–9. Boccalon S, Scaggiante B, Perissin L. Anxiety stress and nociceptive responses in mice. Life Sci 2006;78:1225–30. Kompagne H, Bardos G, Szenasi G, Gacsalyi I, Harsing LG, Levay G. Chronic mild stress generates clear depressive but ambiguous anxiety-like behaviour in rats. Behav Brain Res 2008;192:311–4. Piato AL, Detanico BC, Jesus JF, Lhullier FL, Nunes DS, Elisabetsky E. Effects of Marapuama in the chronic mild stress model: Further indication of antidepressant properties. J Ethnopharmacol 2008;118:300–4. Yalcin I, Aksu F, Belzung C. Effects of desipramine and tramadol in a chronic mild stress model in mice are altered by yohimbine but not by pindolol. Eur J Pharmacol 2005;514:165–74. Yalcin I, Aksu F, Bodard S, Chalon S, Belzung C. Antidepressant-like effect of tramadol in the unpredictable chronic mild stress procedure: possible involvement of the noradrenergic system. Behav Pharmacol 2007;18:623–31. Burne TH, Johnston AN, McGrath JJ, Mackay-Sim A. Swimming behaviour and post-swimming activity in Vitamin D receptor knockout mice. Brain Res Bull 2006;69:74–8. Welch JM, Lu J, Rodriguiz RM, et al. Cortico-striatal synaptic defects and OCDlike behaviours in Sapap3-mutant mice. Nature 2007;448:894–900. Gorris LG, van Abeelen JH. Behavioural effects of (-)naloxone in mice from four inbred strains. Psychopharmacology (Berl) 1981;74:355–9. McFarlane HG, Kusek GK, Yang M, Phoenix JL, Bolivar VJ, Crawley JN. Autism-like behavioral phenotypes in BTBR T+tf/J mice. Genes Brain Behav 2008;7:152–63. Mineur YS, Prasol DJ, Belzung C, Crusio WE. Agonistic behavior and unpredictable chronic mild stress in mice. Behav Genet 2003;33:513–9. Wang D, An SC, Zhang X. Prevention of chronic stress-induced depression-like

49.

50.

51.

52.

53.

54.

55.

56.

57.

58.

59.

60.

61.

behavior by inducible nitric oxide inhibitor. Neurosci Lett 2008;433:59–64. Kalueff AV, Maisky VA, Pilyavskii AI, Makarchuk NE. Persistent c-fos expression and NADPH-d reactivity in the medulla and the lumbar spinal cord in rat with short-term peripheral anosmia. Neurosci Lett 2001;301:131–4. Makarchuk M. [An electrophysiological evaluation of the role of the olfactory analyzer in brain integrative activity]. Fiziol Zh 1999;45:77–83. Makarchuk M, Zyma IH. [Effect of anosmia on sex-related differences in conditioned avoidance in rats]. Fiziol Zh 2002;48:9–15. Makarchuk NE. [The effect of anosmia on sex dimorphism in the patterns of orientingexploratory, emotional and passive defensive behaviors in rats]. Zh Vyssh Nerv Deiat Im I P Pavlova 1998;48:997–1003. Fineberg NA, Saxena S, Zohar J, Craig KJ. Obsessive-compulsive disorder: boundary issues. CNS Spectr 2007;12:359–64, 67–75. Diefenbach GJ, Tolin DF, Hannan S, Crocetto J, Worhunsky P. Trichotillomania: impact on psychosocial functioning and quality of life. Behav Res Ther 2005;43:869–84. Hartley JE, Montgomery AM. 8-OHDPAT inhibits both prandial and waterspray-induced grooming. J Psychopharmacol 2008. Navarro M, Rubio P, de Fonseca FR. Behavioural consequences of maternal exposure to natural cannabinoids in rats. Psychopharmacology (Berl) 1995;122:1–14. Kalueff AV, Laporte JL, Murphy DL, Sufka K. Hybridizing behavioral models: A possible solution to some problems in neurophenotyping research? Prog Neuropsychopharmacol Biol Psychiatry 2008;32:1172–8. Garner JP, Weisker SM, Dufour B, Mench JA Barbering (fur and whisker trimming) by laboratory mice as a model of human trichotillomania and obsessive-compulsive spectrum disorders. Comp Med 2004;54:216–24. Kalueff AV, Minasyan A, Keisala T, Shah ZH, Tuohimaa P. Hair barbering in mice: implications for neurobehavioural research. Behav Processes 2006;71:8–15. Kurien BT, Gross T, Scofield RH. Barbering in mice: a model for trichotillomania. BMJ 2005;331:1503–5. Sarna JR, Dyck RH, Whishaw IQ. The Dalila effect: C57BL6 mice barber whiskers by plucking. Behav Brain Res 2000;108:39–45.

Chapter 3 Digging in Mice: Marble Burying, Burrowing, and Direct Observation Reveal Changes in Mouse Behavior Robert M.J. Deacon Abstract Mice spontaneously dig in many substrates in the laboratory. This behavior comes from their ancestry in the wild, where they would forage for seeds, grain, insects, and other food to be found buried in the soil or leaf litter in their natural habitat. The most convenient and sensitive way of measuring digging in mice is the burrowing test. Mice are placed in individual cages, each fitted with a ‘‘burrow,’’ a tube filled with food pellets or other substances. The amount of substrate spontaneously dug out of the burrow after 2 h and subsequently overnight is measured. The test is extremely simple to run; the apparatus is inexpensive and readily constructed. It exploits a common natural rodent behavior, provides quantitative data under controlled laboratory conditions, and has proved extremely sensitive to prion disease, lipopolysaccharide administration, strain differences, and brain lesions. Other ways of measuring digging behavior include direct observation and the marble burying test; full details as to how to run these are also given in this chapter. Key words: Burrowing, mice, scrapie, digging, marble burying, lipopolysaccharide, LPS.

1. Introduction Mice will spontaneously dig when given a suitable substrate such as deep bedding. This behavior can be measured and quantified in at least three ways, as described in this chapter. In their original natural habitat, the dry steppes of central Asia, mice would have spent much of their time foraging for grass seeds driven by the wind into the sandy soil. The seeds would be a valuable source of protein and carbohydrate, and indeed water, as mice can metabolize carbohydrates to water (1). Although they do not thrive on entirely dry food, mice can survive on very little or even no water. A striking example was the discovery of a live mouse T.D. Gould (ed.), Mood and Anxiety Related Phenotypes in Mice, Neuromethods 42, DOI 10.1007/978-1-60761-303-9_3, ª Humana Press, a part of Springer ScienceþBusiness Media, LLC 2009

37

38

Deacon

in a barrel of dry biscuits which had been sealed for 14 months on the ill-fated Franklin expedition (2). Small insects might also be found in the sand and eaten. When mice spread west into forested land, digging in the leaf litter would have revealed insects, nuts, and seeds. Digging, therefore, became an innate behavior in mice. Digging has been well studied in the laboratory. It is sensitive to factors such as species (3–5) and brain lesions (especially of the hippocampus) (6). In 2001, a paper was published describing ‘‘burrowing,’’ a behavior related to digging in cage bedding or other substrates (7). It was found that mice would spontaneously dig out the contents of a tube filled with food pellets. The original motivation behind this discovery had been to devise a simple method of measuring hoarding, without the need to modify the home cage or build suitable apparatus that would comprise a base or source area connected to an external food source; the food was simply placed in a tube or other container (8) in the home cage. Observation of the mice, however, showed that they were simply digging out the contents of the tube, not storing the food pellets in a discrete part of the cage as would be expected to occur if this behavior was true hoarding. The topology of their movements was also similar to those of digging in the bedding; the front paws gathered the pellets and drew them back under the abdomen, then powerful thrusts of the hind limbs propelled the pellets out of the tube. It was subsequently shown that mice would burrow virtually any substrate, even soiled bedding from their own cage if this was put in the tube. Moreover, placing an empty tube next to a full one did not prevent the contents of the full tube being burrowed. Subjectively, the mice appeared to find burrowing a rewarding activity, although this has never been tested in a formal way as done by Mason et al., who showed that animals would invest effort to gain access to valued resources. Fur-farmed mink were shown to favor a water pool above a range of other resources (9). It seems likely that digging in deep bedding and burrowing represent closely related behaviors, although burrowing may be a more sensitive test. The strain differences between C57BL/6 and C57BL/10 were greater for burrowing than digging (10). This may be because the burrow acts as a ‘‘supra-normal’’ stimulus, to instigate the behavior. Since the open end is much larger than a normal mouse burrow, it may act like a giant egg to an oystercatcher (there are many other examples). The Nobel prize-winning ethologist Niko Tinbergen observed that, presented with a normal oystercatcher egg, a larger egg, and a giant egg, oystercatchers generally chose the giant egg (11). (Also, see the later discussion on the mice burrowing in a semi-natural environment.) Over the course of several years, it was shown that burrowing was an extremely sensitive test, detecting scrapie disease in mice long before there were any clinical signs. Parallel tests of open field

Digging in Mice

39

activity, motor co-ordination, and strength showed that this was not due to any physical deficits; indeed the mice were hyperactive when burrowing was impaired (12). A series of experiments on scrapie-infected mice showed that burrowing was a suitable test for reliable repeated monitoring of a disease state (13–15). Burrowing detected scrapie slightly earlier than did performance on a DRL operant schedule (16). Paradoxically, BSE prion infection increased digging in the home cage (17). Unfortunately, this study did not include burrowing per se. In an elegant study, Mallucci et al. (18) showed that burrowing behavior was depressed in prion-infected mice, but this was reversed when production of indigenous prion protein was stopped. Since the scrapie infection in our studies was performed by injecting prion material into the hippocampus, it was an obvious next step to determine if hippocampal lesions also impaired burrowing. This proved to be the case (19). Lesions of the medial prefrontal cortex also impair burrowing, but to a lesser extent than hippocampal lesions (20). Another example of the sensitivity of burrowing is the impairment following administration of lipopolysaccharide (LPS). This occurs at a dose three orders of magnitude less than that necessary to change body temperature (21). Burrowing was also sensitive to interleukin-1-beta overproduction mediated by an adenovirus (22). Burrowing has been shown to be modified by several genetic manipulations. It is lower in mice lacking the KATP channel subunit Kir6.2 (23). It is also lower in the Tg2576 model of amyloid over-expression, although this may be independent of plaque formation as it occurs as early as 3 months of age, before this develops; lower ages are yet to be tested (24). In spite of this wealth of experimental data, until recently, we were unsure what burrowing actually represented. Was it truly reflecting the natural behavior of burrow-making? The answer to this came from a paper which described mice in a natural environment ‘‘spring cleaning’’ their burrows (25). Piles of waste were deposited outside tunnel entrances in late March–April, probably as an adaptive behavior to reduce the risk of pathogens building up in the warm summer weather. Significantly, Schmid-Holmes et al. observed that the appearance of burrows with large entrances (>6 cm) temporally coincided with the ‘‘spring cleaning’’ activity. It may be that natural burrows >6 cm are causally related to burrowing/spring cleaning; the burrows used for mice in the Oxford laboratory are 6.8 cm in diameter, so (albeit somewhat fortuitously) we may have selected a particularly suitable size. Further studies in our own laboratory demonstrated that burrowing is not shown by the Egyptian spiny mouse (Acomys cahirinus), which correspondingly does not make burrows in the wild.

40

Deacon

Hence, burrowing in the laboratory truly does seem to reflect a natural behavior. It is also shown by rats, gerbils, and hamsters, but in these species only terrestrial substrates such as sand or gravel are vigorously burrowed, whereas food pellets are not. Another way of measuring digging behavior is to use the socalled ‘‘marble burying’’ test, first described by Broekkamp et al. (26). In this model, a pattern of marbles is distributed on the surface of a bedding-filled cage, and the number of marbles at least two-thirds buried after a set time is counted. Marble burying is reduced by anxiolytic agents such as the benzodiazepine drugs (27). So digging has sometimes been considered as an index of anxiety. However, antidepressant drugs, especially those acting on serotonin systems, also reduce marble burying. Some of these drugs are also used to treat obsessive compulsive disorder (OCD), and it was once suggested that marble burying was a model of OCD (albeit one without great face validity, as it was the normal mice that were modeling a disorder!) (28). Our present understanding is that digging is a natural spontaneous species-typical behavior that is provoked by a suitable substrate such as deep wood-chip bedding, rather than an expression of anxiety (29). Digging is distinct from defensive burying, which occurs in response to a noxious stimulus, which is generally electric shock in laboratory settings (for a review see (30)). In the wild, defensive burying occurs when mice are confronted by predators such as snakes. Defensive burying differs from digging in its topology: the former mainly consists of forward movements of the forepaws, kicking the substrate over the snake or electrified probe, whereas in the latter the hind limbs are mainly used to kick the substrate backwards. One advantage of using burrowing, digging or marble burying as behavioral tests is that these procedures do not produce ‘‘pain, suffering, distress or lasting harm’’ in normal animals. Therefore, they are not regulated procedures as defined by the UK Animals (Scientific Procedures) Act of 1986. Indeed, the animals appear to be highly motivated to perform these apparently rewarding activities. Sherwin et al. (31) showed that burrowing in peat was a highly motivated behavior; mice would perform an operant task to gain access to this substrate. They even recommended that burrowing opportunities be provided as environmental enrichment to caged mice. Thus, the tests described here conform to the spirit of the ‘‘3Rs’’ of Russell and Burch (32). A vital question for this chapter is ‘‘do digging, marble burying, and burrowing measure changes in mood or affect?’’ There is good evidence that burrowing does, and by extrapolation it seems likely that digging and marble burying do too. A common factor in prion disease, LPS, and cytokine administration is that animals show reduced consumption of glucose, which reflects their anhedonia, lowered mood or reduced positive affect. Lesions of the

Digging in Mice

41

hippocampus or medial prefrontal cortex in mice (19, 20) did not affect glucose consumption, however, even though the latter structure has been associated with reward (33). So burrowing can be, but does not have to be, associated with changes in affect. Impairment in burrowing models the behavior seen in clinical depression well; apathy and reduced activity (especially goaldirected and purposeful) are hallmarks of depression, and contrast markedly with the vigorous activity seen in normal burrowing.

2. Materials and Methods 2.1. Burrowing

Mouse burrows (Fig. 3.1) are made from grey or black 68-mm diameter plastic downpipe, sealed at one end by a round of mdf, and elevated 3 cm at the other end by two 50-mm machine screws 1 cm in from the end, spaced just less than a quadrant of the tube apart.

Fig. 3.1. A mouse performing the burrowing test.

Fill the cylinder with 200 g food pellets (ordinary laboratory chow) and place in a clean cage with a thin layer of bedding, against the long wall of the cage. The closed end of the cylinder should be against the back wall of the cage. No food is necessary in the hopper, and indeed it may distract their attention from the burrow; so this is an important detail of standardization. Put a single

42

Deacon

(non food-deprived) mouse in and after 2 h measure the amount of food displaced from the tube (defined as being on the floor of the cage rather than in the tube). (For all practical purposes, this is 200 g-weight left in the tube.) The test is optimally run approximately 2 h before the start of the dark cycle. Continue the test overnight, supplying a water bottle. The 2-h measurement is generally more sensitive than the overnight one, the latter often suffering from a ceiling effect as almost all the food is displaced, but with sensitivity also comes variability, particularly if this is the first time the mice are exposed to the test. Many different substrates other than food pellets can be used. We have successfully tried the following substrates: soiled cage bedding, new cage bedding (aspen premium 8/20 wood chip bedding; Lillico, Betchworth, Surrey, UK), soil, gravel (pea shingle; small stones 1 cm diameter; B&Q, Chandlers Ford, Hampshire, UK), clay balls, as used to line the surface of the soil in indoor plant pots (HYDROLECA; William Sinclair Horticulture Ltd., Lincoln, UK), and sand (Play Sand; www.BritishPlaySand.co.uk). The heavier substrates like soil, sand, and gravel are useful if burrowing reaches a ceiling in the test time, as they obviously require more effort to displace. 2.2. Digging Measured by Direct Observation

Mice are individually placed in a black wooden alley (27  9  30 cm) filled with approximately 5-cm deep wood chips, lightly tamped down to make a flat, even surface. The substrate can be reused if it is flattened and firmed down again between mice; reuse of bedding does not seem to affect the burying/digging performance of subsequently tested mice. Two alleys can be run simultaneously, side by side. The latency to start digging, the total duration of digging, and the number of individual digging bouts are measured using a timer/event counter. Test duration is 3 min. Recently we have used a square apparatus of the same area as the alley described here and this makes observation of digging easier.

2.3. Marble Burying

Twenty glass marbles are placed, evenly spaced in five rows of four, on a 5-cm layer of sawdust bedding, lightly pressed down to make a flat even surface, in a plastic cage approximately 20  30 cm (Fig. 3.2). (You can use 12 marbles in a smaller cage.) A mouse is placed in each cage and left for 30 min after which the number of marbles buried (to twothirds their depth) with sawdust is counted. The bedding substrate can be reused, if it is flattened and firmed down again; no systematic studies have been done but reuse of bedding does not seem to affect the burying performance of subsequently tested mice.

Digging in Mice

43

Fig. 3.2. Marble burying. The mouse on the right is a control and has buried several marbles. The mouse on the left received a complete hippocampal lesion and has not buried any.

2.4. General Notes

The above parameters are used in our lab in Oxford but are for guidance only; minor variations in apparatus, setup, and procedure are possible. The tests can be run several times, but it is suggested that no more than three are run in a week, otherwise the motivation of the mice may decline. One test a week is probably adequate for monitoring most chronic disease models. These tests are generally robust and easy to run. However, if poor performance is observed on the first test, test them again in a few days and they often improve. Indeed, if a series of tests is planned to monitor a disease state, it is good practice to obtain baseline performance, then sort the results and allocate the mice to the treatment groups based on their baseline performance. Also social facilitation in the home cage can be very useful to trigger the behavior; put in a full burrow, or fill the home cage with deep bedding, and/or put an array of marbles on the bedding. Fuller details on how to perform these tests can be found on the Nature Protocols web site (34, 35).

Acknowledgments This work was supported by grant GR065438MA from the Wellcome Trust to the Oxford OXION group.

44

Deacon

References 1. Fertig DS, Edmonds VW. The physiology of the house mouse. Sci Am 1969;221:103–8. 2. McClintock FL. A Narrative of the Discovery of the Fate of Sir John Franklin and His Companions. London 1859;197. 3. Dudek BC, Adams N, Boice R, et al. Genetic influences on digging behaviours in mice (Mus musculus) in laboratory and seminatural settings. J Comp Psychol 1983;97: 249–59. 4. Solberg L C, Valdar W, Gauguier D, et al. A protocol for high-throughput phenotyping, suitable for quantitative trait analysis in mice. Mamm Genome 2006;17:129–146. 5. Webster DG, Williams MH, Owens RD, et al. (1981). Digging behavior in 12 taxa of muroid rodents. Anim Learn Behav 1981;9:173–177. 6. Deacon RMJ, Rawlins JNPR. Hippocampal lesions, species-typical behaviours and anxiety in mice. Behav Brain Res 2005;156: 241–249. 7. Deacon RMJ, Raley JM, Perry VH, et al. Burrowing into prion disease. Neuroreport 2001;12:2053–7. 8. Contet C, Rawlins JNP, Deacon RMJ. A comparison of 129S2/SvHsd and C57BL/ 6JOlaHsd mice on a test battery assessing sensorimotor, affective and cognitive behaviours: implications for the study of genetically modified mice. Behav Brain Res 2001;124:33–46. 9. Mason GJ, Cooper J, Clarebrough C. Frustrations of fur-farmed mink. Nature 2001;410:35–36. 10. Deacon RMJ, Thomas CL, Rawlins JNP, et al. A comparison of the behavior of C57BL/6 and C57BL/10 mice. Behav Brain Res 2007;179:239–47. 11. Tinbergen N. The Study of Instinct. Oxford: Clarendon Press, 1951. 12. Betmouni S, Deacon RMJ, Rawlins JNP, et al. Behavioral consequences of prion disease targeted to the hippocampus in a mouse model of scrapie. Psychobiology 1999;27: 63–71. 13. Cunningham C, Deacon R, Wells H, et al. Synaptic changes characterize early behavioural signs in the ME7 model of murine prion disease. Eur J Neurosci 2003;17: 2147–55. 14. Cunningham C, Deacon RMJ, Chan K, et al. Neuropathologically distinct prion strains give rise to similar temporal profiles

15.

16.

17.

18.

19.

20.

21.

22.

23.

24.

25.

26.

of behavioural deficits. Neurobiol Dis 2005;18:258–69. Guenther K, Deacon RMJ, Perry, VH, et al. Early behavioural changes in scrapieaffected mice and the influence of dapsone. Eur J Neurosci 2001;14:401–9. Deacon RMJ, Reisel D, Perry VH, et al. Hippocampal scrapie infection impairs operant DRL performance in mice. Behav Brain Res 2005;157:99–105. Kempster S, Collins ME, Deacon R, et al. Impaired motor coordination on static rods in BSE-infected mice. Behav Brain Res 2004;154:291–5. Mallucci GR, White MD, Farmer M, et al. Targeting cellular prion protein reverses early cognitive deficits and neurophysiological dysfunction in prion-infected mice. Neuron 2007;53:325–35. Deacon RMJ, Croucher A, Rawlins JNP. Hippocampal cytotoxic lesion effects on species-typical behaviors in mice. Behav Brain Res 2002;132:203–13. Deacon RMJ, Penny C, Rawlins JNP. Effects of medial prefrontal cortex cytotoxic lesions in mice. Behav Brain Res 2003;139: 139–55. Teeling JL, Felton LM, Deacon RMJ, et al. Sub-pyrogenic systemic inflammation impacts on brain and behavior, independent of cytokines, Brain Behav Immun 2007;21: 836–850. Campbell SJ, Deacon RM, Jiang Y, et al. Overexpression of IL-1beta by adenoviralmediated gene transfer in the rat brain causes a prolonged hepatic chemokine response, axonal injury and the suppression of spontaneous behaviour. Neurobiol Dis 2007;27:151–63. Deacon RMJ, Brook RC, Meyer D, et al. Behavioral phenotyping of mice lacking the KATP channel subunit Kir6.2. Physiol Behav 2006;87:723–33. Deacon RMJ, Cholerton LL, Talbot K, et al. Age-dependent and -independent behavioral deficits in Tg2576 mice. Behav Brain Res 2008;189:126–38. Schmid-Holmes S, Drickamer LC, Robinson AS, et al. Burrows and burrow-cleaning behaviour of house mice. Am Mid Nat 2001;46:53–62. Broekkamp CL, Rijk HW, Joly-Gelouin D, et al. Major tranquillizers can be distinguished from minor tranquillizers on the basis of effects on marble burying and

Digging in Mice

27.

28.

29.

30.

swim-induced grooming in mice. Eur J Pharmacol 1986;126:223–9. Njung’e K. Handley SL. Evaluation of marble-burying behavior as a model of anxiety. Pharmacol Biochem Behav 1991;38:63–67. Takeuchi H, Yatsugi S-I, Yamaguchi T. Effect of YM992, a novel antidepressant with selective serotonin re-uptake inhibitory and 5-HT2A receptor antagonistic activity, on a marble-burying behavior test as an obsessive-compulsive disorder model. Jpn J Pharmacol 2002;90:197–200. Masuda Y, Ishigooka S, Matsuda Y. Digging behavior of ddY mouse. Exp Anim 2000;49: 235–7. De Boer SF, Koolhaas JM. Defensive burying in rodents: ethology, neurobiology and psychopharmacology. Eur J Pharmacol 2003;463:145–161.

45

31. Sherwin CM, Haug E, Terkelsen N, et al. Studies on the motivation for burrowing by laboratory mice. Appl Anim Behav Sci 2004;88:343–58. 32. Russell WMS, Burch RL. The Principles of Humane Experimental Technique. London, UK: Methuen, 1959. 33. Tzschentke TM. The medial prefrontal cortex as a part of the brain reward system. Amino Acids 2000;19:211–9. 34. Deacon RMJ. Burrowing in rodents: a sensitive method for detecting behavioral dysfunction. Nat Protocols 2006a;1: 118–21. 35. Deacon RMJ. Digging and marble burying in mice: simple methods for in vivo identification of biological impacts. Nat Protocols 2006b;1:122–4.

Chapter 4 Circadian and Light Modulation of Behavior Cara M. Altimus, Tara A. LeGates, and Samer Hattar Abstract Nearly all organisms contain a circadian biological clock that is responsible for coordinating the temporal functions of many physiological systems. The circadian clock is synchronized to the earth’s day/night rhythms via changes in the intensity of light throughout the cycle. In mammals, the eyes and specifically the intrinsically photosensitive retinal ganglion cells are essential for transmitting light information to the brain to influence the physiology of the organism. Several biochemical, hormonal, molecular, and behavioral functions are affected by the interaction of the circadian clock with the daily light/dark cycle. Furthermore, many studies have shown an association between circadian biology and mood regulation. Here, we present several behavioral methods in mice and humans for the measurement of the interaction between the endogenous biological clock and light. By incorporating circadian phenomena into mood studies, the link between the clock, light, and mood could be better understood. Further, modification of the light/dark environment should provide tools to control sleep, mood, and cognition via direct light input on behaviors. Key words: Circadian rhythm, depression, biological clock, light, mood, mice, seasonal affective disorder, bipolar disorder.

1. The Importance of Time Time plays a central role in biological processes. An example concerns a seasonal form of depression known as seasonal affective disorder (SAD). The most parsimonious explanation for this form of depression involves the length of day, which varies depending on the time of the year (1). The time necessary for biological processes varies from nanoseconds, as observed in the movement of ions through channels, to yearly hibernation events. Perhaps one of the most undertood time events involves the basis of our daily homeostatic system that is influenced by the earth’s rotation about its axis. This daily cycle is physiologically important as revealed when one travels between T.D. Gould (ed.), Mood and Anxiety Related Phenotypes in Mice, Neuromethods 42, DOI 10.1007/978-1-60761-303-9_4, ª Humana Press, a part of Springer ScienceþBusiness Media, LLC 2009

47

48

Altimus, LeGates, and Hattar

different time zones and experiences symptoms of ‘‘jet-lag’’ that negatively affects cognitive functions and mood. Furthermore, a comorbidity of mood disorders such as bipolar disorder is circadian dysfunction manifesting as sleep problems (2) showing a clear link between circadian function and mood regulation. Understanding how to measure these daily functions and how to manipulate the circadian time within an organism may be important in the development of strategies to help patients with psychiatric diseases. A temporal hierarchical system influences daily physiological functions. Most organisms regulate physiological processes on a 24-hour basis by incorporating daily and temporal seasonal changes in the light/dark cycles to better adapt to their environment. An autonomous internal timekeeping mechanism independent of environmental input produces oscillations that are approximately 24 hour (circadian) in length (circadian – circa: approximate and diem: day). The ability to align these internal timekeeping processes to an exact 24-hour length to match those of the exogenous light/ dark cycle is known as photoentrainment. To enable synchronization to the photic input, light cues are conveyed to a pacemaker located in the ventral region of the hypothalamus above the optic chiasm, known as the suprachiasmatic nucleus (SCN). This pacemaker in turn, conveys light information to several peripherally located oscillators in the rest of the body (3). A detailed understanding of the molecular components for circadian rhythms has been achieved with the work of many labs over the past several years (4). The molecular clock components are expressed in most tissues of organisms. In mammals, the circadian system is organized with the master circadian oscillator residing in the SCN (5, 6), which in turn coordinates the phase of rhythms expressed in peripheral tissues (3). Without light input to the SCN, the intrinsic rhythm is not physiologically relevant because the animal is not able to confine activities to any temporal niche. The SCN, by coordinating peripheral rhythms, produces a temporal order that is valuable to the organism’s survival by optimizing the timing of physiological functions.

2. The Molecular Clockwork The molecular clock is structured with two interlocking feedback loops. The first loop involves the proteins CLOCK and BMAL, which heterodimerize in the cytoplasm and then shuttle into the nucleus where they regulate transcription (Fig. 4.1). Both proteins are transcription factors that contain basic helix–loop–helix domains that allow them to bind to an E-box element within the promoter of target genes. The CLOCK:BMAL complex activates the transcription of the period (Per) and cryptochrome (Cry) genes that contain E-box

Circadian and Light Modulation of Behavior

Nucleus

2 Clock

49

Bmal

Clock

Bmal

Clock

Bmal

Per1/Per2 Cry1/Cry2

Per1/2 Cry1/2

Clock

Clock

Bmal

Clock

Bmal

Bmal

1

Rorα

Bmal

Rora Rev-erbα

Rev-erbα

Bmal

Clock

Per1/2

4

Per1/2 Cry1/2

Cry1/2

3

Rorα

Rev-erbα

Cytoplasm Fig. 4.1. The molecular clock is found in many cells in the body. (1) CLOCK and BMAL form a heterodimer in the cytoplasm of the cell and subsequently translocate to the nucleus. (2) In the nucleus, CLOCK:BMAL activates transcription of the Per and Cry genes through binding in the promoter region of those genes. (3) Once translated, PER and CRY heterodimerize in the cytoplasm, translocate to nucleus, and inhibit CLOCK:BMAL. (4) CLOCK:BMAL also activates Ror and Rev-Erb transcription. (5) ROR and REV-ERB translocate to the nucleus where ROR activates transcription of Bmal and REV-ERB inhibits transcription of Bmal.

elements in their promoter (Fig. 4.1). The Per and Cry transcripts are then translated to proteins in the cytoplasm forming heterodimers that translocate to the nucleus to inhibit the activity of the CLOCK:BMAL complex, thereby inhibiting their own production. PER/ CRY proteins are then subject to phosphorylation by casein kinase I (CKI and CKId), which lead to their degradation through the ubiquitin pathway. The phosphorylation and degradation of the PER/CRY proteins define the time of inhibition of the CLOCK:BMAL complex. Once PER/CRY levels decrease sufficiently, CLOCK:BMAL will again be able to promote transcription of the Per and Cry genes to restart the cycle. The length of the transcription and inhibition of this loop is approximately 24 hours (4). A second regulatory feedback loop is interconnected to the previously mentioned one through the CLOCK:BMAL complex, which upon entering the nucleus promotes the transcription of ROR and Rev-Erb genes, also through association with E-box elements. These genes, when translated to proteins, will shuttle back

50

Altimus, LeGates, and Hattar

to the nucleus and compete for the retinoic acid-related element (RORE) in the Bmal promoter. ROR promotes transcription of Bmal while REV-ERB inhibits transcription of Bmal, thus creating another layer of regulation on the circadian circuit (Fig. 4.1).

3. Importance of the Mouse Model In this book chapter, we will concentrate primarily on mice as an animal model, because they serve as a useful organism in which to study the circadian clock as well as mood-related behaviors. Mice are genetically amenable, but at the same time, many behavioral functions can be carried out in this organism including tests that measure anxiety- and depression-like behaviors. The availability of a welldefined genetic system and a battery of behavioral tests makes this organism the ideal choice for the research presented here.

4. Clock Mutants Correlate with Mood Disorders

Work in both humans and model organisms have shown links between the circadian system and mood disorders. Genetically modified mice have been used for many of these studies allowing researchers to knock out key components of the molecular circadian clock pathway. Research on the molecular components of circadian rhythms such as Clock, Per2, and CKI revealed mood effects in addition to circadian phenotypes (7–9). We will review key human and animal model findings below. By genetically knocking out Clock in mice, two groups found independently that the Clock-knockout mice show decreased despair and anxiety and an enhanced response to reward (7, 8). Similarly, Per2-knockout mice show decreased despair. In the Per2 study, the researchers also found a reduction in the expression of monoamine oxidase-A (MAO-A), an enzyme necessary to breakdown neuromodulators such as dopamine and serotonin. The decreased expression of this degradative enzyme results in a decrease in total MAO-A activity, thus producing an increase in monoamines, which could account for the behavioral changes in despair (9). In human patients, researchers have associated single nucleotide polymorphisms (SNPs) of essential clock proteins with mood disorders. The SNP studies target a disorder and then determine if a common polymorphism in a gene is present in a significant number of individuals with the disorder. SNP studies of bipolar disorder

Circadian and Light Modulation of Behavior

51

have implicated polymorphisms in several clock components including Bmal, Per3, and a recently discovered CLOCK-binding protein, Npas2 (10, 11). The SNPs studies are supported by a reciprocal approach using pharmacological manipulations showing that many psychoactive drugs also affect the molecular clock components of the circadian system. For example, in bipolar patients, a dysregulation of circadian rhythms commonly manifests in sleep-related problems (12) and administration of lithium, a mood stabilizer, consistently lengthens the circadian period (13, 14). The effects of lithium on the period length of the clock might be mediated by Rev-Erb, which contains sites for phosphorylation by GSK-3b. Lithium is known to directly inhibit the kinase activity of GSK-3b and hence could lead to differential phosphorylation levels on Rev-Erb. Since phosphorylation is important for regulating the molecular clock components, this may lead to changes in the period length of the clock. In fact, lithium administered in cell culture studies decreased levels of Rev-Erb and increased Bmal in agreement with direct modulation of clock components (15). Additionally, chronic administration of the antidepressant fluoxetine induces changes in Clock, Bmal1, and Npas2 expression in the hippocampus, a structure greatly affected by depression and antidepressant treatment (16). Recent advances in microarray analysis allow researchers to determine how specific treatments influence global gene expression profiles. Microarray analysis of tissue from regions in the striatum of mice treated with cocaine showed changes in the expression of Clock, Per1-3, Cry2, and Npas2 (16). Microarray studies also found that administration of the mood stabilizer, valproate, decreased expression of Cry2 and CKI in the amygdala (17). Furthermore, co-administration of methamphetamine with valproate blocks the effects on Cry2 and CKI in the amygdala indicating possible roles for these genes in emotion, anxiety, and mania (17). This work indicates that the circadian system regulates (and may be regulated by) regions of the brain, which control mood.

5. Light Detection to Set the Circadian Clock

Light influences the behavior of many organisms: behaviors as simple as phototropism or as complex as image formation and object tracking. In humans, light impacts physiological functions including sleep and mood that are important for the quality of life. Our visual system allows us to form images of the world around us. For this purpose, light is first detected in the retina (a thin layer of cells in the back of the eye) by rods and cones, the classical photoreceptors, and signaled to the retinal ganglion cells via retinal interneurons such as bipolar and amacrine cells (Fig. 4.2). The

52

Altimus, LeGates, and Hattar Back of the Eye

ONL

H INL A

GCL Optic Nerve

Light

Fig. 4.2. Three main cell layers make up the retina. The cells of the retina are oriented in an inverted fashion such that light must pass through cell layers prior to reaching the photoreceptors rods and cones in the outer nuclear layer (ONL). Light information is transduced by rods and cones into an electrical signal that is then conveyed to interneurons such as bipolar, horizontal (H), and amacrine cells (A) in the inner nuclear layer (INL). The information received by these three cell types integrates the light information received from the rods and cones and conveys it to the ganglion cells. An additional photoreceptive cell is found in the ganglion cell layer (GCL). This ganglion cell expresses the photopigment melanopsin allowing it not only to receive rod/cone input but also to be intrinsically photosensitive. Ganglion cell axons bundle together forming the optic nerve and exit the eye through the optic disk to signal the brain.

retinal ganglion cells (RGCs), the only output neurons of the retina, send light information to image-forming centers of the brain. The eyes, in addition to perceiving light signals for image formation, detect light for other important physiological purposes, collectively referred to as non-image-forming (NIF) visual functions. One key

Circadian and Light Modulation of Behavior

53

visual function of NIF is circadian synchronization to light (known as circadian photoentrainment), which allows us to adjust our daily activities to the light–dark cycle produced by the rising and setting of the sun. For many years, rods and cones were assumed to be the only photoreceptors in the mammalian retina. Intriguingly, a group of blind humans who were unable to form images were found capable of detecting light for NIF functions (18). Genetically manipulated mice that lack all rods and cones and are hence image-blind were likewise able to detect NIF light information (19, 20). However, bilateral loss of the eyes in both humans and mice abolishes all light detection including that for NIF functions (21). A small subset of RGCs that project directly to the SCN, the central circadian pacemaker in the brain, was identified (22).These RGCs were shown to be the non-rod/non-cone photoreceptors in the mammalian retina because they are intrinsically photosensitive (i.e., they respond to light directly in the absence of any rod or cone input) (23). Rods and cones express opsin proteins that are seven-pass transmembrane G-protein-coupled receptors. It has been conclusively demonstrated that the intrinsically photosensitive RGCs (ipRGCs), which represent 1–2% of the total ganglion cell population, express an opsin-like protein, melanopsin (24, 25). When melanopsin is knocked out in mice, the cells lose their intrinsic photosensitivity but retain their ability to signal light information from rod/cone input (25–28). To determine if rods, cones, and the ipRGCs are the only photoreceptors in the mammalian retina, triple knockout animals were generated in which the light-signaling pathway of each of these photoreceptors was eliminated resulting in the loss of the ability to detect light for circadian photoentrainment. Thus, mice lacking all three known photoreceptors in the eye were blind for both image- and nonimage-forming visual functions (29). To determine if ipRGCs are necessary for light detection for NIF functions, the cells were genetically ablated. The ablation of ipRGCs does not affect the ability of the animals to form images but renders them incapable of photoentrainment. Importantly, the ablation of ipRGCs did not affect the SCN circadian oscillator, as the ipRGC-ablated mice produced an endogenous rhythm with a period similar to that observed in wild-type animals housed in constant darkness. Therefore, these animals were only defective in sending light information for circadian photoentrainment, but had an intact circadian oscillator and normal image-forming vision (30, 31). One implication of this work is that normally sighted people with sleep problems or seasonal depression could benefit from tests specifically tailored to detect deficiencies in the NIF light detection system. Several tests that could specifically measure NIF functions are explained in Section 8 of this chapter.

54

Altimus, LeGates, and Hattar

6. Light Not Only Regulates the Clock But Also Affects Behavior Directly

7. Sleep, Circadian Rhythms, and Mood Disorders

Perhaps one of the most recognized effects of light on mood is on a seasonal form of depression known as SAD. As explained above, in mammals, light regulates the circadian clock through a direct connection from the eyes to the SCN. However, light also affects behavior directly, and in the case of SAD, the evidence indicates that it is the decreased exposure to light due to the shorter day that causes the disorder and not the effect of light on the clock, although the therapeutic effects of light could be gated by the clock. In mice, the direct effect of light on behavior is termed masking, because light overcomes the propensity for mice to be active at night. Mice are nocturnal, and therefore, a light pulse presented at night inhibits activity as revealed by lower counts on the wheel-running activity output (32). The intensity of light to which the mice are exposed to during the dark phase of the 24-hour day determines the magnitude of the masking response. Surprisingly, it was shown that light induces sleep directly at night, as measured by electroencephalograms (EEG) and electromyograms (EMG) (33, 34). In diurnal and nocturnal mammals, melatonin levels increase in the night and are directly inhibited by the presence of light (35). In humans, the direct light effects on behavior can be assayed by measuring the decrease in melatonin levels in response to exposure to bright light at night. Because most mouse lines used in the laboratory do not have melatonin, analogous studies could only be carried out in C3H mice that express melatonin.

Another consideration for the study of circadian and mood disorders is that sleep is partially controlled by the circadian system and is often disrupted in mood disorders. Many mood disorders are characterized by either insomnia or hypersomnia, leading to the possibility that the circadian system may also be involved. Several examples of sleep disorders implicate the role of the circadian clock in sleep regulation. Familial advanced sleep-phase syndrome (FASPS) and delayed sleep-phase syndrome (DSPS) are two circadian clock disorders that affect the onset of sleep in humans. Those with FASPS tend to sleep early and awaken early, while those with DSPS sleep later and awaken much later. Though the underlying mechanism behind these disorders remains elusive, Per2 and Per3 polymorphisms have been linked to FASPS and

Circadian and Light Modulation of Behavior

55

DSPS, respectively (36, 37). In addition, research has also associated these disorders with physiological and mental symptoms including high incidence of depression (38). Sleep studies are now possible not only in humans and other large animals, but also in mice allowing for the genetic understanding of the mechanisms of sleep. Because mice are small animals, the recordings are technically challenging. However, several systems have been developed which allow simultaneous recording of two EEG and one electromyograms (EMG) channels (39). With these three electrical waves, sleep and wake are reliably distinguished. Wake is characterized by high-amplitude EMG and low-amplitude/high-frequency EEG. Sleep is broadly divided into rapid eye movement (REM) and non-REM (NREM) sleep. REM sleep, which in humans is the stage when dreaming most commonly occurs, is defined as high theta (8–12 Hz) activity in the EEG and no muscle movement. REM sleep is also known as paradoxical sleep because the EEG activity is most similar to wakefulness. NREM sleep is characterized by minimal muscle movement and highamplitude, low-frequency EEG, which makes this stage reliably distinguishable from REM and wakefulness. Genetic manipulations of light input to sleep and circadian centers could be used to differentiate between circadian and direct light effects on sleep. For example, recent work has shown that light input affects sleep both directly and indirectly via circadian photoentrainment. The direct effect of light on sleep requires input from both rod/cone and melanopsin systems, while in the circadian modulation of sleep, either the rod/cone or melanopsin system is sufficient to photoentrain sleep. Since mice are nocturnal animals, light promotes sleep and darkness induces wakefulness (34). These studies are analogous to alertness studies performed in humans showing that light promotes alertness (40). An understanding of how to manipulate important functions such as sleep by light and dark could allow us to develop strategies to help patients with psychological problems related to sleep disorders.

8. How to Measure Clock Function? 8.1. Wheel-Running Activity

Perhaps one of the most utilized behavioral output in circadian biology is wheel-running activity (Fig. 4.3). This output is advantageous because it allows a non-invasive recording of several circadian paradigms with minimal researcher intervention. The number of wheel revolutions is recorded via a computer program, and then the data are plotted using a circadian software system (Fig. 4.1; Clocklab by Actimetrics, Wilmette, IL). The plot of wheel-running activity over a period of time is termed an actogram with the

56

Altimus, LeGates, and Hattar

Fig. 4.3. Wheel-running activity is a robust measurement of circadian behavior in mice. To measure activity, a mouse is placed in a cage with a wheel. A magnet is attached to the wheel such that a probe positioned next to the wheel detects the revolution of the wheel. This probe is connected to a computer that will receive and collect the entire wheel-running activity. These data can be compiled and viewed using a circadian software.

number of days plotted along the y axis and hours along the x axis. The number of wheel revolutions is plotted on a sub-y axis (Fig. 4.3). The actogram is double plotted to allow measurement of activity onsets and offsets. The activity onsets in constant conditions (Fig. 4.3) are used to determine an animal’s circadian period, which is defined as the average time between two consecutive onsets. Circadian biologists commonly use the following two systems to denote time: zeitgeber time and circadian time. Zeitgeber (from German, time-giver) time is defined as the time point in the cycle based on a cyclical external cue such as light and dark cycles that drives the rhythm of the animal. In a 12:12 hour light:dark cycle, zeitgeber time (ZT) 0 is when the lights are turned on and ZT12 is when the lights are switched off (Fig. 4.4). Circadian time (CT) is defined by the organism’s endogenous biological clock. The CT system is used in constant conditions when there are no external cues influencing the clock and hence only the circadian clock is

Circadian and Light Modulation of Behavior

57

Fig. 4.4. Wheel-running activity shows key features of the circadian system. (A) Photoentrainment of mice to a 12:12 light:dark cycle shows that the mice are able to confine activity to dark portion of the day. (B) Free running behavior in constant darkness shows that the clock is intact. The circle indicates where light pulse was administered to produce a phase delay (compare dashed line to solid line). (C) ‘‘Jet-lag’’ paradigm tests the ability of the clock to adjust to changes in the light cycle. shows a phase advance where it takes for this animal several days to readjust to the new light–dark cycle.
shows a phase delay using the offset part of the activity since ! indicates a masking response of light on the wheel-running activity. In (A–C), black bars show wheel-running activity, and the following backgrounds represent light conditions: grey when lights are off, white when lights are on. All actograms are double plotted.

driving the rhythm. In mice, CT12 is arbitrarily defined as the start of activity. To calculate other CT times, you have to first determine the circadian hour, which is calculated by dividing the circadian period by 24. The circadian period is estimated via plotting a regression line in actograms of animals that are kept in constant

58

Altimus, LeGates, and Hattar

conditions (Fig. 4.4B; green line). Once the circadian hour is calculated, other CT times could be estimated. For example CT18 is CT12 time + (6*circadian hour). To measure the endogenous circadian period, mice are placed in constant dark conditions (DD), while humans are placed in constant dim-light conditions. In these two paradigms, there is no cyclical external input to set the clock allowing the clock to ‘‘free-run.’’ When a mouse is free-running, the clock and therefore the activity of the mouse is undergoing a circadian rhythm without synchronization to an external cue revealing the periodicity of the endogenous rhythm and the clock. Mice have an endogenous period that is less than 24 hours, while humans have a period longer than 24 hours (41). It is important to note that, in mice, constant light conditions lead to period lengthening, and high light intensity induces arrhythmic behavior. The ability of the clock to respond (or reset) to light is measured using several tests. The most basic exposes the mice to a 12:12 hour light/dark cycle (LD) and measures the wheel-running activity rhythms to determine if mice confine their activity to the dark portion of the day or photoentrain (Fig. 4.4A). A photoentrained mouse will consistently begin running at the same time each day, creating a phase relationship to the light/dark cycle. This phase relationship is useful because it allows quantification of the strength of the light input to the circadian oscillator. An unstable ‘‘wobbly’’ activity onset indicates a weaker photoentrainment. A complete lack of any consistent association of the activity rhythms and the light–dark cycle indicates that the animal is free-running and cannot detect light for the circadian system. It is important to note that lack of association with the light/dark cycle does not imply arrythmicity. To measure the amount of time required for a mouse to establish a stable phase relationship to the light/dark cycle, advances or delays of the LD cycle to investigate how long it takes a mouse to re-entrain to the new paradigm are used. The advance and delay paradigms are commonly referred to as ‘‘jet-lag’’ schedule, because they mimic the effect of flying across time zones (Fig. 4.4C). On average, a wild-type mouse requires 5–7 days to adjust its activity to a 6-hour phase advance. In the case of a phase delay, the onset of activity readjusts immediately; however, this is not re-entrainment, but a masking light effect on activity. Under this phase-delay circumstance, it is most useful to look at the offset of activity to determine the time required for re-entrainment (Fig. 4.4C). However, offsets are not as reliable as onsets for determining phase relationships between activity and the light– dark cycle. A classical result in the circadian field is that a single brief light pulse presented to animals housed in constant darkness could cause quantifiable changes in the onset of wheel-running activity in subsequent cycles (Fig. 4.4B; green versus

Circadian and Light Modulation of Behavior

59

orange line). This test is known as a phase shift and is used to quantify light input to the clock. This test allows manipulation of many variables including light intensity, time of presentation, and duration of light presentation. For example, in constant darkness the endogenous period of the clock could be calculated and used to determine onsets of activity for subsequent days, which allow estimation of the phase of the rhythm. After the period and onset of activity are determined, the mouse is then exposed to a light pulse of a given intensity and duration at a specific CT time (e.g., 1000 lux for 15 minutes at CT16). This brief pulse of light will advance, delay, or have no effect on the clock (Fig. 4.5). The amount of phase shift could be measured by the change in the time of the onset of activity before and after the light pulse was presented. The degree and direction of activity onset change differ based on the time of the circadian phase at which the light pulse is administered. The curve describing this phenomenon is a phase response curve, which plots change in activity onset versus circadian time of light pulse (Fig. 4.5). Typically, light pulses presented from CT0–12 produce no phase shift, while light pulses presented from CT12–24 produce a phase shift. For a maximal phase delay in wild-type mice, the pulse of light should be presented at CT16, which is approximately 4 hours after the start of activity. Alternatively, a light pulse presented at CT21 induces a maximal phase advance.

2

Phase Shift (hrs)

1

–1

–2

0

4

8

12 Circadian Time

16

20

24

Fig. 4.5. Phase response curve shows the response to a light pulse based on circadian time. Example graph plots the change in activity onset in response to a standard 1000 lux, 15-minute light pulse. Pulses administered during the inactive phase do not illicit change in activity onset. Light pulses in the early active phase (early subjective night) lead to phase delays while those given in the late active phase (late subjective night) cause phase advances.

60

Altimus, LeGates, and Hattar

The ability of a single short-duration light pulse to change the phase of the circadian oscillator allows the animals to entrain to what is known as ‘‘skeleton photoperiods.’’ A skeleton photoperiod involves the application of short-duration light pulses at two times in the day separated by few hours of darkness or red light, in the case of mice. The use of red light in mice is similar to darkness because mice cannot use red light to entrain their endogenous circadian rhythms (mice lack red sensitive cones). The simplest setup for these experiments is to first photoentrain the mice to a 12:12 hour light/ dark cycle followed by the use of two 1-hour light pulses: the first light is presented at ZT0 and the second at ZT 11. In this scenario, the light pulses ‘‘outline’’ the light phase, which will allow the mice to retain the ability to photoentrain and produce a 24-hour rhythm. This cycle is useful because it leads to photoentrainment without the continuous presence of light. This, in turn, allows researchers to investigate the time of day effect in photoentrained animals independent of the confounding direct effects of light. To verify that mice are photoentrained under the skeleton paradigm, a circadian output such as wheel-running activity should be measured. The battery of quantifiable tests that are presented above allow an accurate determination of the strength of the light input to the circadian oscillator. Specifically, the speed of photoentrainment, the intensity and minimum duration of light required to adjust the phase of the circadian oscillator, could be determined for mutant mice. 8.2. Other Measurable Rhythms

A strength of the circadian clock is that wheel-running activity is not the only measurable output. For example in mice, sleep/wake, body temperature, drinking/feeding, and general activity (Pinnacle Technology, Lawrence, KS; Data Sciences International, St. Paul MN; Respironics Minimitter, Bend OR) also provide robust rhythms in constant conditions. To record sleep and temperature rhythms in mice, survival surgery is necessary. The sleep surgery requires implanting an electrode in the skull while the temperature probe is implanted in the peritoneal cavity and sends the data wirelessly to a computer. Some temperature-recording systems also measure general activity; however, a non-surgical procedure for sampling general activity is accomplished by placing a mouse in a cage with infrared beams and measuring the number of beam breaks. Drinking and feeding rhythms can be recorded by changing the feeding apparatus in the cage to either measure weight changes of food and water containers over time or beam breaks to the food-holding containers. These rhythms are also useful because they are affected by different environmental factors and are therefore mediated by other brain nuclei.

Circadian and Light Modulation of Behavior

61

8.3. Per2 Luciferase Mice

Many methods described thus far have utilized measuring a behavioral output; however, the cyclic behavior of a clock protein can also be used as a circadian measure. This technique is particularly useful because many tissues show daily oscillations in function even in constant conditions and thus are also controlled by the circadian clock. These experiments utilize a mouse expressing the firefly bioluminescence enzyme, luciferase, under the control of the Per2 promoter (3). Because PER2 levels naturally fluctuate with the circadian cycle, the expression of the luciferase also fluctuates. By measuring the luminescence of dissected SCN and peripheral tissues of the Per2:Luc mouse in culture, the period of the tissue can be determined (3).

8.4. Pupillary Light Reflex as a Noninvasive Test for Nonimage-Forming Functions

The pupillary light reflex depends on both rod/cone and melanopsin input for optimal response at all light intensities. This method holds a lot of potential in both humans and mice because it is non-invasive and is an immediate response non-image-forming function (Fig. 4.6). In mice, it has been shown that rods/ cones contribute to the pupil response in dim-light conditions, while melanopsin contributes at high-light intensities (28). This

Fig. 4.6. Pupillary light reflex can be used as a test for non-image-forming light response. Both images are from frames of a video taken of the pupil response of a mouse. (A) A representative image of a dark adapted pupil. (B) The response after the opposite eye was exposed to bright light. The pupil constricts in response to bright white light. White dashed circles outline pupil area.

distinction in the response intensity curves allows for the relative determination of rod/cone versus melanopsin input. In rodents, atropine (a muscarinic acetylcholine receptor antagonist) is administered to one eye to control the degree to which the pupil is open. The atropine-treated eye is exposed to a light source and because light input to either eye is reflected in the pupillary response of both eyes, the pupil constriction response is recorded in the opposite eye. Infrared illumination is used such that the

62

Altimus, LeGates, and Hattar

pupil constrictions could be viewed on a camera without interference of the visible light source. The video recording is then used to measure pupil diameter before and after light exposure. To study pupillary light constriction in humans, one eye is dilated and presented with light, while the opposite eye is video recorded and then analyzed for changes in pupil diameter. In humans, the light is administered for 10 seconds and recording continued for up to 30 seconds after the light presentation. This study found that post-stimulus (sustained) pupil constriction was most sensitive to 482 nm, likely a melanopsin-driven response (42). This assay can be used as a high throughput test for nonimage-forming functions because the test is non-invasive and many subjects and wavelengths can be tested. 8.5. Human Circadian Studies

Both circadian and non-image-forming tests are difficult to perform in humans because they require prolonged assessments in the laboratory. Pupillary light reflex studies are more difficult to interpret than those preformed in mice due to cortical input in humans, but still hold potential for future use to test non-image-forming visual functions. Recently, researchers are able to measure circadian rhythms in humans. These experiments are inherently more difficult than rodent studies because humans are affected by social cues as well as environmental cues, which influence circadian rhythms. To prevent confounding factors from affecting physiological rhythms, a constant condition protocol, which requires that the subject spend the entire experiment in a reclining position, awake, not moving, and eating small evenly spaced meals, was used. These experiments allow for the detection of the endogenous clock by using outputs such as body temperature recordings or melatonin rhythms. In order to study the direct effects of light in humans, melatonin levels are measured. Melatonin is produced in the pineal gland and is often described as the ‘‘dark’’ hormone because its levels begin to rise during the night, and it is acutely suppressed by light. To measure circadian rhythms, the dim-light melatonin onset (DLMO) is determined. The use of dim light does not influence the levels of melatonin, and hence, the circadian clock solely drives melatonin rhythms. To measure the sensitivity of a person to light, levels of melatonin are measured during a light pulse (typically 6 hours) presented in the dark portion of the cycle. Because in-lab human studies require isolation of subjects from their environment, less disruptive studies involving telemetry to monitor core body temperature, respiration, heart rate have been developed. These systems can be worn as a watch, attached to the skin as a dermal patch, or in some cases the transmitter is enclosed in a capsule and given in pill form (some products available at Minimitter, Philips Respironics, Bend, OR). Each of these

Circadian and Light Modulation of Behavior

63

rhythms will give readouts of the subject’s daily rhythms in their environment, which will take into account social/environmental cues and the daily cycle. An exciting method is being developed to measure human circadian output by culturing human skin fibroblasts (from a single skin biopsy) with a lentiviral circadian reporter (43). These studies have found that the skin fibroblasts have an average period of 24.5, congruent with other human circadian studies. However, the variability in period between subjects was much greater in these fibroblast measurements making this procedure potentially useful for identifying heterogeneity in circadian phenotypes (44). Future studies will have to validate that this method could be reliably used to identify the circadian period of individual subjects.

9. Concluding Remarks Many physiological processes are controlled by the light/dark cycle or driven endogenously by the circadian oscillator. Understanding the effects of light on the physiology of organisms and their relationship to the endogenous biological clock could help produce better interventions for psychiatric disorders. Furthermore, studies done on a nocturnal animal such as the mouse using bright-light laboratory environment may produce spurious results due to the direct light effects on the mouse behavior. To summarize, there are biochemical, hormonal, molecular, and behavioral changes in organisms throughout the day–night cycle. These daily variations could result in different experimental outcomes depending on when in the cycle tests are administered. Accounting for these variations produces much more consistent outcomes and better understanding of the biological phenomena. References 1. Golden RN, Gaynes BN, Ekstrom RD, et al. The efficacy of light therapy in the treatment of mood disorders: a review and metaanalysis of the evidence. Am J Psychiatry 2005;162(4):656–62. 2. McClung CA. Circadian genes, rhythms and the biology of mood disorders. Pharmacol Ther 2007;114(2):222–32. 3. Yoo SH, Yamazaki S, Lowrey PL, et al. PERIOD2::LUCIFERASE real-time reporting of circadian dynamics reveals persistent circadian oscillations in mouse peripheral

tissues. Proc Natl Acad Sci USA 2004;101(15):5339–46. 4. Reppert SM, Weaver DR. Coordination of circadian timing in mammals. Nature 2002;418(6901):935–41. 5. Davidson AJ, Yamazaki S, Menaker M. SCN: ringmaster of the circadian circus or conductor of the circadian orchestra? Novartis Found Symp 2003;253:110–21; discussion 21–5, 281–4. 6. Siepka SM, Yoo SH, Park J, Lee C, Takahashi JS. Genetics and neurobiology of

64

Altimus, LeGates, and Hattar

7.

8.

9.

10.

11.

12.

13. 14.

15.

16.

17.

18.

19.

20.

circadian clocks in mammals. Cold Spring Harb Symp Quant Biol 2007;72:251–9. Easton A, Arbuzova J, Turek FW. The circadian Clock mutation increases exploratory activity and escape-seeking behavior. Genes Brain Behav 2003;2(1):11–9. Roybal K, Theobold D, Graham A, et al. Mania-like behavior induced by disruption of CLOCK. Proc Natl Acad Sci USA 2007;104(15):6406–11. Hampp G, Ripperger JA, Houben T, et al. Regulation of monoamine oxidase A by circadian-clock components implies clock influence on mood. Curr Biol 2008;18(9):678–83. Johansson C, Willeit M, Smedh C, et al. Circadian clock-related polymorphisms in seasonal affective disorder and their relevance to diurnal preference. Neuropsychopharmacology 2003;28(4):734–9. Mansour HA, Wood J, Logue T, et al. Association study of eight circadian genes with bipolar I disorder, schizoaffective disorder and schizophrenia. Genes Brain Behav 2006;5(2):150–7. Lenox RH, Gould TD, Manji HK. Endophenotypes in bipolar disorder. Am J Med Genet 2002;114(4):391–406. Klemfuss H. Rhythms and the pharmacology of lithium. Pharmacol Ther 1992;56(1):53–78. Healy D, Waterhouse JM. The circadian system and the therapeutics of the affective disorders. Pharmacol Ther 1995;65(2):241–63. Yin L, Wang J, Klein PS, Lazar MA. Nuclear receptor Rev-erbalpha is a critical lithiumsensitive component of the circadian clock. Science 2006;311(5763):1002–5. Uz T, Ahmed R, Akhisaroglu M, et al. Effect of fluoxetine and cocaine on the expression of clock genes in the mouse hippocampus and striatum. Neuroscience 2005;134(4):1309–16. Ogden CA, Rich ME, Schork NJ, et al. Candidate genes, pathways and mechanisms for bipolar (manic-depressive) and related disorders: an expanded convergent functional genomics approach. Mol Psychiatry 2004;9(11):1007–29. Czeisler CA, Shanahan TL, Klerman EB, et al. Suppression of melatonin secretion in some blind patients by exposure to bright light. N Engl J Med 1995;332(1):6–11. Freedman MS, Lucas RJ, Soni B, et al. Regulation of mammalian circadian behavior by non-rod, non-cone, ocular photoreceptors. Science 1999;284(5413):502–4. Lucas RJ, Freedman MS, Munoz M, Garcia-Fernandez JM, Foster RG. Regulation of the mammalian pineal by non-rod,

21.

22.

23.

24.

25.

26.

27.

28.

29.

30.

31.

32.

33.

34.

non-cone, ocular photoreceptors. Science 1999;284(5413):505–7. Nelson RJ, Zucker I. Photoperiodic control of reproduction in olfactory-bulbectomized rats. Neuroendocrinology 1981;32(5):266–71. Moore RY, Lenn NJ. A retinohypothalamic projection in the rat. J Comp Neurol 1972;146(1):1–14. Berson DM, Dunn FA, Takao M. Phototransduction by retinal ganglion cells that set the circadian clock. Science 2002;295(5557):1070–3. Provencio I, Rodriguez IR, Jiang G, Hayes WP, Moreira EF, Rollag MD. A novel human opsin in the inner retina. J Neurosci 2000;20(2):600–5. Hattar S, Liao HW, Takao M, Berson DM, Yau KW. Melanopsin-containing retinal ganglion cells: architecture, projections, and intrinsic photosensitivity. Science 2002;295(5557):1065–70. Panda S, Sato TK, Castrucci AM, et al. Melanopsin (Opn4) requirement for normal light-induced circadian phase shifting. Science 2002;298(5601):2213–6. Ruby NF, Brennan TJ, Xie X, et al. Role of melanopsin in circadian responses to light. Science 2002;298(5601):2211–3. Lucas RJ, Hattar S, Takao M, Berson DM, Foster RG, Yau KW. Diminished pupillary light reflex at high irradiances in melanopsin-knockout mice. Science 2003;299(5604):245–7. Hattar S, Lucas RJ, Mrosovsky N, et al. Melanopsin and rod-cone photoreceptive systems account for all major accessory visual functions in mice. Nature 2003;424(6944):76–81. Guler AD, Ecker JL, Lall GS, et al. Melanopsin cells are the principal conduits for rod-cone input to non-image-forming vision. Nature 2008;453(7191):102–5. Hatori M, Le H, Vollmers C, et al. Inducible ablation of melanopsin-expressing retinal ganglion cells reveals their central role in non-image forming visual responses. PLoS ONE 2008;3(6):e2451. Mrosovsky N. Masking: history, definitions, and measurement. Chronobiol Int 1999;16(4):415–29. Lupi D, Oster H, Thompson S, Foster RG. The acute light-induction of sleep is mediated by OPN4-based photoreception. Nat Neurosci 2008. Altimus CM, Gu ¨ ler, A. D., Villa, K. L., McNeill, D. S., LeGates, T. A., Hattar, S. Rods-cones and melanopsin detect light and

Circadian and Light Modulation of Behavior

35.

36.

37.

38.

39.

40.

dark to modulate sleep independent of image formation. Proc Natl Acad Sci U S A 2008. Wetterberg L. Melatonin in humans physiological and clinical studies. J Neural Transm Suppl 1978(13):289–310. Ebisawa T, Uchiyama M, Kajimura N, et al. Association of structural polymorphisms in the human period3 gene with delayed sleep phase syndrome. EMBO Rep 2001;2(4):342–6. Vanselow K, Vanselow JT, Westermark PO, et al. Differential effects of PER2 phosphorylation: molecular basis for the human familial advanced sleep phase syndrome (FASPS). Genes Dev 2006;20(19):2660–72. Regestein QR, Monk TH. Delayed sleep phase syndrome: a review of its clinical aspects. Am J Psychiatry 1995;152(4):602–8. Naylor E, Bergmann BM, Krauski K, et al. The circadian clock mutation alters sleep homeostasis in the mouse. J Neurosci 2000;20(21):8138–43. Lockley SW, Evans EE, Scheer FA, Brainard GC, Czeisler CA, Aeschbach D. Shortwavelength sensitivity for the direct effects

41.

42.

43.

44.

65

of light on alertness, vigilance, and the waking electroencephalogram in humans. Sleep 2006;29(2):161–8. Wright KP, Jr., Hughes RJ, Kronauer RE, Dijk DJ, Czeisler CA. Intrinsic near-24-h pacemaker period determines limits of circadian entrainment to a weak synchronizer in humans. Proc Natl Acad Sci USA 2001;98(24):14027–32. Gamlin PD, McDougal DH, Pokorny J, Smith VC, Yau KW, Dacey DM. Human and macaque pupil responses driven by melanopsin-containing retinal ganglion cells. Vision Res 2007;47(7):946–54. Nagoshi E, Saini C, Bauer C, Laroche T, Naef F, Schibler U. Circadian gene expression in individual fibroblasts: cellautonomous and self-sustained oscillators pass time to daughter cells. Cell 2004;119 (5):693–705. Brown SA, Fleury-Olela F, Nagoshi E, et al. The period length of fibroblast circadian gene expression varies widely among human individuals. PLoS Biol 2005;3 (10):e338.

Chapter 5 Ultrasonic Vocalizations by Infant Mice: An Ethological Expression of Separation Anxiety James T. Winslow Abstract The ultrasonic vocalization, or isolation calling, of infant rats and mice has been studied as a measure of anxious affective state and as an early communicative behavior between a pup and mother. The protocol described herein is the typical separation testing procedure. Also included are procedures used to modulate crying by providing contact with littermates and/or dam and increased isolation calling response by a prior brief maternal interaction. These procedures provide the basis for experimental research on the early development of emotion and communication in a critically important experimental model species – the mouse. Key words: Ultrasonic vocalization, isolation calling, mice, rats, rodents, crying, infant, separation anxiety, emotion, maternal separation, depression, mood.

1. Background and Historical Overview The production of vocalizations by infant mammals during parental separations has been measured in virtually every known mammalian species (1). The biological and evolutionary significances of such calls have been the subject of some debate reaching back to and past Charles Darwin to Aristotle (e.g., see (2)). More recently this debate has enlivened in the developmental psychobiological research community into a full fledge argument (3–6). The issues captured by this debate range beyond the scope of the current discussion but briefly center on the possibility that rodent ultrasonic ‘‘crying’’ was evolutionarily selected to serve thermoregulatory rather than communicative needs. The study of ultrasonic vocalizations by infant rodents has a somewhat more immediate history dating back to an excellent comparative study by Noirot (7). This study and subsequent T. D. Gould (ed.), Mood and Anxiety Related Phenotypes in Mice, Neuromethods 42, DOI 10.1007/978-1-60761-303-9_5, ª Humana Press, a part of Springer ScienceþBusiness Media, LLC 2009

67

68

Winslow

comparative studies (8, 9) revealed that rodents including mice and rats emit species-typical ultrasonic vocalizations when separated from their dams. The rate and intensity of vocalizations vary with environmental factors including temperature, olfactory and tactile cues (10, 11). Indeed accounting for environmental temperature as well as body temperature represent critical variables in the interpretation of a variety of experimental manipulations including drug treatments.

2. Ultrasonic Vocalization as a Quantifiable, Ethologically Relevant, Species-Typical Measure

3. Relationship of Ultrasonic Vocalization Responses to Affective States

The ultrasonic vocalization response to maternal separation, the subsequent contact quieting response when the pup is reunited with the dam, and a frustration-like ‘‘maternal potentiation’’ phenomenon, represent different aspects of the infant rodents’ social, affective, and communication behavior (12, 13). Infant calling typically consists of a series of discriminable, short ‘‘whistle-like’’ calls (14, 15). The separation response can consequently be characterized by the rate of calling, and the call rate is generally proportional to how different (or novel) the separation environment is compared to the natal nest. Ultrasonic vocalization rate has been used as an easily quantifiable measure of individual differences in sensitivity to change or challenge and has been proposed to model anxiety with some predictive validity for adult temperament (16, 17). The stimuli and the provocations for these responses are highly relevant to the environment in which young mammals develop both proximally during postnatal development and distally relative to species evolution and ethology. These responses can be detected on the first occasion of testing (within the first 2 weeks of life) and independent of prior learning experience with the eliciting conditions. Consequently, they are regarded as innate or evolved responses rather than learned affective responses such as context and cue-elicited fear conditioning.

Young mammals typically make vigorous attempts to reunite with littermates and dam when separated in familiar as well as novel surroundings. That said, it is not unusual for the rodent dam to leave her nest for relatively long intervals (>2–3 hours) as she forages and feeds. Return of the isolated pup to a lactating dam acts as a powerful reinforcer in the formation of positive

Ultrasonic Vocalizations by Infant Mice

69

associations to previously neutral cues (18). Isolation-induced ultrasonic vocalizations are thought to reflect the developmental sensitivity of neural processes to provocative environments, while the contact quieting response is thought to express an early positive or reward state, both dependent on a homeostatic balance maintained by contact with littermates and dam (19, 20). An isolated preweaning rodent is vulnerable to a wide range of threats (e.g., cold, predation, and starvation) but individual ultrasonic vocalizations do not appear to carry information about the specific nature of these threats. A distress state induced by a signal for unspecified dangers is widely accepted as a definition of anxiety, as used clinically in the diagnosis of childhood separation anxiety in humans (21). This is used to distinguish it from fear, the response to a specific danger, such as the unfamiliar male appears to present to infant rodents (22).

4. Test Conditions Regulating the Ultrasonic Vocalization Response to Isolation

Most infant-call protocols use acute social isolation of the pup to elicit ultrasonic vocalization, so that it is important to consider how variation in test conditions may affect this response. Hofer (22) provides an excellent step-by-step-how-to guide for provoking and measuring ultrasonic vocalizations in rats and mice, which we can do little to improve and guides our presentation here. In general, the ultrasonic vocalization rate of an isolated pup will vary according to the extent of the pup’s separation from familiar features of the home cage nest, as well as to the nature and intensity of the cues for risk in the eliciting conditions. For example, if pups are isolated in a chamber with home cage shavings or if the odor of the dam is present, or if the floor of the chamber is warmed to near nest temperature, the pups’ ultrasonic vocalization rate will be reduced proportionately to the number of such home cage features present in the test chamber. Likewise, cues for risk or danger in the isolation environment, such as wet, cold, or an unstable or moving substrate, increase pups’ ultrasonic vocalization rates. All these stimuli can be considered to be regulators of infant ultrasonic vocalization and are test variables that should be comparable across experimental conditions and taken into account when interpreting results.

5. Temperature The ambient temperature of the test chamber is the most easily manipulated regulator of ultrasonic vocalization rate in rat and mouse pups during their first 2 weeks of postnatal life (22).

70

Winslow

Ultrasonic vocalization rates expressed by isolated pups can be systematically varied from 10 or less calls per minute at typical ambient temperatures up to 200 calls per minute in a cooler environment. This means that the range of pups’ ultrasonic vocalization rates in a particular experiment can be systematically raised or lowered by cooling or warming the floor of the test chamber or by using a temperature-controlled chamber. This can be useful. For example, pups can be tested in warmed test chambers so that they will not be near their ceiling ultrasonic call rate, thus limiting sensitivity to a potentially provocative manipulation. Conversely, a warm environment may result in control pups with relatively low ‘‘basement’’ ultrasonic vocalization response levels that would be difficult to further reduce with an experimental manipulation. Ambient temperature during testing is usually the most difficult test condition to maintain constant over days of testing and replications of experiments. It is important, therefore, to always monitor the temperature of the test environment and to make sure that controls are included under identical temperature conditions for each experimental group in the statistical analysis (22). The relationship between ultrasonic vocalization and temperature has led some to hypothesize that ultrasonic vocalization emissions might also play a role in pups’ physiological thermoregulatory capacity (3). According to this view, ultrasonic vocalization could be considered, in part, as a byproduct of thermoregulatory physiology rather than an affective expression in a communicative system (3). Indeed, correlations have been found between ultrasonic vocalization production and thermoregulatory and cardiovascular changes (reviewed in (23)). Nevertheless, a clear physiological role for ultrasonic vocalization in thermoregulation has yet to be demonstrated except perhaps during recovery from severe hypothermia (24). In contrast, there are several lines of evidence that the physiological changes involved in the act of ultrasonic vocalization emission do not play a functionally significant role in the thermoregulation of young rats, at least under typical test conditions (6, 24–26).

6. Age of Pups Ultrasonic vocalization responses of infant rats follow a fairly predictable developmental pattern. The first ultrasonic vocalization response to isolation occurs in a day or two after birth at a relatively high rate at typical ambient temperatures (32–35C), then rises to a peak in the first week at about 100/min, and then finally beginning a gradual decline until at 17–20 days postnatal, at which point no ultrasonic vocalizations may be detected within a 10-min

Ultrasonic Vocalizations by Infant Mice

71

separation test (27). Mouse pup’s ultrasonic vocalization responses follow a similar course, but terminate sooner, at 12–15 days (8). Mouse pups of all ages show contact quieting with littermates, whereas with the dam, rat pups 14 days and older continue to emit some ultrasonic vocalization.

7. Mouse and Rat Pup Differences As noted in the introduction, in mice, only the acute isolation response has been systematically studied, whereas contact quieting and maternal potentiation have, as yet, only been studied in rat pups. The neuropharmacology of isolation-induced ultrasonic vocalization has been studied in both species (13, 28–31). The usefulness of geneknockout strategies in mice has generated numerous studies employing measures of ultrasonic vocalization during early development (e.g. (31–33)). The maternal retrieval response and the maternal sensory, endocrine, and behavioral adaptations to the mouse pup’s ultrasonic vocalization signal have been analyzed more completely in the mouse (34, 35) than in the rat (36) dam. Mouse pups have higherfrequency ultrasonic vocalizations (50–80 kHz versus 30–50 kHz in rats) and emit them for a shorter period following birth (2–14 days in mice, 2–18 days in rats). Both these measures, as well as the characteristic ultrasonic vocalization rates in isolation at the various ages, differ still further among the many available genetically defined strains of mice (37, 38). As a result, we know more about the genetics of isolation-induced ultrasonic vocalization in mice (e.g., (39), but the only selective breeding study for ultrasonic vocalization responses has been done in rats (40–42)).

8. Ultrasonic Vocalization Response to Isolation

In testing the impact of rearing experience, genetic manipulation or background or pharmaceutical treatment on the ultrasonic vocalization response, it is typical to detect an intensity or concentration-related effect as either an increase or a decrease in ultrasonic vocalization response during a brief separation test at a specific postnatal age. Depending on age and environmental conditions, the ultrasonic vocalization isolation response can be reduced 50–70% or increased 200–300% or more providing remarkable sensitivity. The specificity of a manipulation’s effect can be gauged by the degree to which a clear effect on ultrasonic vocalization rate occurs relative to effect on body temperature,

72

Winslow

activity level, or any of the other isolation-induced behaviors such as motor activity, grooming or exploratory rearing. If these behaviors are not scored, alternative neuromotor tests for confounding effects such as the righting reflex, negative geotaxis, and paw withdrawal should be considered (22). An infant mouse’s ultrasonic vocalization isolation response can give information on the early effects of single gene deletions or targeted gain-of-effect mutations in mice (31, 32). An advantage of the infant ultrasonic vocalization response over traditional behavioral tests in adults is that whereas secondary changes in expression of other genes may complicate interpretation of the adult phenotype, the early expression of the targeted genetic alteration in infancy may be more direct and free of later developmental ‘‘compensatory’’ effects. For example, an OT gene knockout may produce only a weak effect in classic adult tests of anxiety, but produce a robust effect on the infant ultrasonic vocalization (32). The principal weakness of this test is that consequences of a manipulation are measured as an absence of behavior which may be the result of a number of possible deficits including loss of motivation, impaired respiration or altered thermoregulation. Consequently it becomes critical to employ convergent measures of contributing systems to reveal underlying mechanisms. Nevertheless, the ease of measurement, the ethological relevance, and species typical features of this response remain very attractive.

9. Contact Quieting Brain lesions, drugs, or genetic differences may also exert an effect by augmenting or interfering with the actions of naturally occurring regulators of ultrasonic vocalization such as temperature, odor, or substrate texture. This possibility can be tested by measuring the contact quieting response in the experimental subjects. A single anesthetized age-matched pup can be expected to reduce the ultrasonic vocalization rate of rat pups by 60–70%. This effect can be compared, for example, with a group given naltrexone, a drug that virtually eliminates this quieting response, but has no significant effect on the ultrasonic vocalization isolation response (43, 44).

10. Maternal Potentiation Some genetic manipulations, such as selective breeding, can eliminate all ultrasonic vocalization emission in response to isolation. The first question should be ‘‘does the gene alteration interfere

Ultrasonic Vocalizations by Infant Mice

73

with the physical act of vocalizing, so that the animal is unable to emit ultrasonic vocalization (e.g., a laryngeal abnormality) or is there a more central alteration in regulation?’’ A test for maternal potentiation can demonstrate a latent capacity for vocalization in a previously silent animal, which should emit ultrasonic vocalization at rates of 30–50/min, after a brief interaction with its dam (10, 45).

11. How to Measure Ultrasonic Vocalizations

Ultrasonic vocalizations, ranging from 100 to 500 ms in duration, are emitted by rodent pups virtually from the day after birth to the time of weaning. Although low rates (1–5/min) of ultrasonic vocalizations can be recorded from rat pups in the home cage nest at the beginning and end of nursing bouts, separation of a single pup from its home cage, littermates, and dam in a novel test chamber (isolation) elicits high rates of ultrasonic vocalization in mouse and rat pups (up to 200/min), and is by far the most frequently used eliciting stimulus in laboratory studies (22). The rate of calling during the first few minutes of isolation is the critical measure of the intensity of the individual pup’s response. Measurements of duration of ultrasonic vocalization, inter-ultrasonic vocalization interval, bout structure, and acoustic analysis of calls have been performed, but have not been systematically studied in relation to eliciting conditions, regulation by sensory cues, or neuromodulator control (46–48). Typically, ultrasonic vocalizations are transduced into the audible range by special electronic instruments because the sound frequency of these calls (30–50 kHz in rat and 50–80 kHz in mouse) is too high for human perception. Repeated vocalizations in the audible range are observed in most other mammalian species, including humans, when infants are separated from their familiar surroundings and social companions. This early vocalization response is considered to have been strongly conserved in evolution as an affective and communicative display, most likely because of its survival value in eliciting maternal search and retrieval responses, nursing, and caretaking. Complementary maternal physiological and behavioral adaptations to vocalizations have been described in rodent species, for example, a perceptual sensitivity and a hormone-dependent retrieval response (49). Restoration of rat and mouse pup contact with its dam or littermates terminates the isolated pup’s vocal response, an effect referred to as ‘‘contact quieting.’’ Potentiation of the ultrasonic calling can be elicited in rat pups following a brief period of interaction between the pup and its dam or another lactating female (45). In this provocative condition, ultrasonic vocalization

74

Winslow

rates are markedly increased in the second isolation period, rising to levels three to four times of those shown on its initial isolation, an effect that is not seen following contact with littermates or a non-lactating female. The separation test described below includes the basic procedure for eliciting ultrasonic vocalization, the methods available for transducing rodent ultrasound to the range audible to the experimenter, for recording the signals, and for measuring the rate of ultrasonic vocalization produced. Contact quieting measures the inhibition of the isolation-induced ultrasonic vocalization response by contact of the pup with littermates or dam in the test chamber. Maternal potentiation measures the marked increase in ultrasonic vocalization rate that is observed when the pup is isolated immediately following a brief period of contact with the dam. In order to study the effects of experimental manipulations on pups during these protocols, at least two experimental subject designs are possible. The usual between-subjects design allocates each condition to a different animal within each litter and then to repeat the experiment with a series of litters, using litter means for each condition, thus ensuring that all conditions are represented by at least one animal in each litter. Alternatively, if the experimental manipulation is short and rapid-acting, each pup can serve as its own control (withinsubjects design). In this design, for example, a baseline isolation response could be recorded for 2–3 min, then an intervention/ manipulation could take place, and the subsequent course of the experimental pup’s ultrasonic vocalization response could be compared with the initial isolation period.

12. How to Conduct an Infant Separation Test

The typical procedure used to examine infant rodent ultrasonic vocalization is to remove a single pup from its natal cage, transport it to a testing area, and place it alone on the floor of a novel test chamber over which is suspended an ultrasonic detector microphone. This reliably elicits ultrasonic vocalization from 2- to 17-day-old rat pups or 2- to 14-day-old mouse pups at rates ranging from 100/min in younger pups to as few as 5–10/min in older ones (22).

13. Typical Equipment 13.1. The Detector

Ultrasonic vocalizations can be detected by an ultrasonic microphone as short pure-tone pulses of sounds (100–200 ms; (50)). They can be directly detected by a suitable (but hideously

Ultrasonic Vocalizations by Infant Mice

75

expensive) high-frequency range microphone and recorded to a high-speed tape recorder. Alternatively, ultrasounds can be transduced by an ultrasound (Bat) detector into a sound frequency output that is audible to the experimenter listening through earphones and can also be output to the sound track of a video or audio recorder or the sound card of a computer for storage. Individual ultrasonic vocalizations can then be identified and counted both real-time and better still – later from recordings using typical observational data detection strategies such as the Noldus Observer system. The advantage of such an approach is to obtain simultaneous measures of motor and other behaviors along with vocalizations. Several companies currently supply the ultrasound detector (e.g., Pettersson Elektronik AB – http://www.bahnhof.se/ pettersson/ and Ultrasound Advice – http://www.ultrasoundadvice.co.uk/) and their websites typically explain the principles of operation and the features of the various models available commercially. Most studies of rodent ultrasonic vocalization use the simplest (and cheapest) form of detector that depends on tuning the instrument to the usual call frequency range of the animal of interest. This is a relatively low-cost, heterodyne system that detects only sounds within 2–3 kHz of the selected frequency. Tuning is often performed by trial-and-error scanning of a sample of individuals. The limitation of this approach is that the dominant frequency may vary among individuals and certainly between genetic strains. These may be further complicated by drug effects, hypothermia, or age, so that ultrasonic vocalizations can be missed if the detector is tuned improperly or narrowly on the wrong frequency. It is often more appropriate to use a detector with a ‘‘broadband’’ mode of operation in which any ultrasound between 20 and 120 kHz will be detected and transduced. This is more expensive, but is far preferable for most research purposes, unless the frequency range of the ultrasonic vocalization is well established or there are other sources of ultrasound in the laboratory (e.g., polygraph pens, electric motors) that need to be filtered out by the restricted range of the heterodyne system. An automated approach has recently become commercially available (Noldus Information Technology). In this system, a series of three or four heterodyne detectors, each tuned to a different portion of the range from 20 to 80 kHz, is used for ultrasonic vocalization input. The output of these detectors is then led to an automated four-channel recorder with an adjustable trigger and computer connections providing an electronic record of the rates and temporal patterns of calling at each frequency. The drawback with all automated systems is that other sources of ultrasound in the detectable frequency range (such as the pups stepping on shavings) may trigger the counter as well as

76

Winslow

vocalizations, whereas to the human ear the call is usually easy to distinguish from other sources by its tonal pattern. For those experienced in this work, the respiratory act involved in ultrasonic vocalization production can often be visually recognized as different from non-vocal breaths simply by observation. The key visual cue is the contraction of lower abdominal wall muscles visible in the skin of the pup’s flanks just anterior to the hind legs (22). 13.2. The Testing Chamber

The testing chamber (15–20 cm on each side) should have transparent walls high enough to prevent pup’s escape and long enough to provide room for the pup in the center, away from wall contact, and to allow for any other test objects (e.g., anesthetized conspecific). An arm or other device for suspending the microphone 10–15 cm over the center of the test area is also necessary. The floor of the chamber should be visually divided into six to nine 1-inch squares for measuring locomotion (e.g., draw lines on the lower surface of the transparent chamber floor). It is also useful to have an automated activity platform (e.g., an electromagnetic field disturbance detector; Stoelting) under the test chamber to quantify the level of general behavioral arousal (22).

14. The Procedure Remove the dam from the natal cage, and place her in a small cage apart from the litter, preferably in a separate room or soundattenuating chamber because dams become agitated if kept with the litter during testing. Agitated dams that remain with a litter are likely to scatter the litter, thus creating unstable pretest conditions. They also may potentially emit ultrasounds which may alter the output of pups or be confused with infant calls. Place the natal cage in a temperature-controlled incubator or on a heating pad, preferably regulated by a thermostat on the underside of the cage floor. Measure the pups’ axillary or core temperatures with a fine flexible lubricated thermistor (e.g., Yellow SpringsTM) and identify individual pups with odor-free ink. Make sure that the pups have at least 1 cm of shavings under and around them. Cover the cage partially to prevent drafts, or place the natal cage in an incubator set at the lower end of the thermoneutral temperature range (ambient temperatures at which the pups’ oxygen consumption is at basal levels) for the age of pups being tested (for 5- to 7-day-old rats, 34.5C; 9- to 11-day-old rats, 34.0C; 13- to 15-day-old rats, 32.2C; and 17- to 19-dayold rats, 32.0C) (22). Sufficient shavings allow the litter group to thermoregulate behaviorally along the gradient between the warmed cage floor and the cooler surface of the shavings. The

Ultrasonic Vocalizations by Infant Mice

77

correct temperature settings for the heating pad should be determined on pilot litters of the same age, so that they produce stable or slightly declining home-cage temperatures, before beginning the actual experiments. It is important that the pups not be warmed so that their core temperatures rise above thermoneutrality while they are in the home cage prior to testing, since this can reduce their subsequent isolation calling rate. Incubators do not allow such a gradient, and therefore ambient temperature must be closely monitored to prevent overwarming of the pups. Small decreases (1–2C) in the pups’ core temperatures in the natal cage prior to testing have no measureable effect on their ultrasonic vocalizations. Make sure that the ultrasonic vocalization microphone is properly placed 10–15 cm above the test chamber floor. Check the volume setting of the ultrasonic vocalization detector by lightly rubbing a thumb against the tips of the index and middle finger (which generates ultrasound and should produce a clearly audible output from the ultrasonic vocalization detector). Allow 10–15 min to elapse after removing the dam from the nest to promote steady-state conditions for the litter. Identify the pup to be tested by its ink markings and slip the flexible thermistor probe under its anterior axilla, exerting a slight upward pressure without disturbing the sleeping pup. Alternatively, insert the lubricated probe approximately 1 cm into the anus of the pup and record this as the pretest body temperature (51). Measurement of pretest axillary or core temperatures allows the detection of pups that may have been under unsuspected thermal stress prior to the experiment (22). The magnitude of the pretest–posttest change in body temperature is an indicator of the pups’ thermoregulatory response to isolation. Both these measures allow assessment of possible side effects of drugs or environmental manipulations on pup’s thermoregulatory mechanisms that need to be taken into account when interpreting group differences. Allowing the pup to re-acclimate (10–30 min) with its littermates, pick up the pup by ‘‘shoveling’’ it up and free of the bedding and carefully transport it to the testing chamber. As with the dam, this should be in a separate room or sound-attenuating chamber to prevent the pups from influencing each other. Careful and consistent handling of the pup will help reduce inter-pup variability in ultrasonic vocalization response. Place the pup gently down in the center of the test area, record the ambient temperature in the test chamber and begin observations of the pup’s behavior, and/or start recording equipment. Record ultrasonic vocalizations for 2–6 min depending on the experimental objective. In addition to ultrasonic vocalization, record the pup’s general activity level (e.g., using the automated

78

Winslow

counter described above) and the specific behavioral responses to isolation such as pivoting within or crossing squares marked on the chamber floor, face washing, rearing (with one paw and head raised), and urination/defecation. At the end of the test, measure the pup’s temperature again, weigh it, and gently replace it in the natal nest, and then proceed to the next pup to be tested. Use a clean test chamber for each animal and maintain with lukewarm water, wipe with a paper towel, and allow it to dry completely (2–3 min), allowing it to attain room temperature. Rodents have exquisitely sensitive olfactory perception; so great care should be taken to maintain olfactory neutrality. Record sessions with a video-recorder or observations directly into a computer keyboard, or written onto a paper checklist for each animal for subsequent analysis.

15. Contact Quieting A variation on the basic protocol assesses the capacity of the isolated pup in the test chamber to respond to the passive (anesthetized) body of its dam or littermates by inhibiting its vocalizations and maintaining contact for the duration of the test. The procedure is essentially identical to the former except that an anesthetized stimulus animal (typically the dam) is presented in the testing chamber (22). The stimulus animals should be anesthetized in order to present a uniform, predictable display of passive cues and to prevent the stimulus animal from also emitting ultrasonic vocalization. A ‘‘quieting’’ response has been shown to depend on the cumulative effect of familiar olfactory, tactile, and thermal cues presented by the contact stimulus (52). Rat pups deprived of both olfactory and trigeminal tactile senses fail to show contact quieting, whereas if one or the other is present, the quieting response is nearly normal. A single anesthetized age-matched stimulus can be expected to partially inhibit test pup’s ultrasonic vocalization, whereas a group of anesthetized pups is as effective as the dam and usually eliminates ultrasonic vocalization in rat pups, particularly during the first 2 weeks postnatal. Older pups show a lower percent reduction and are less likely to remain completely quiet during contact in the novel test chamber. Contact quieting appears to develop at the same age as the ultrasonic vocalization response itself, during the first 24–48 h in the rat, and is observed throughout the pre-weaning period. Lactating females and agematched peers from other litters are just as effective stimuli as the pup’s own dam and littermates for eliciting this response.

Ultrasonic Vocalizations by Infant Mice

79

16. Preparing the Stimulus Animal Anesthetize the dam, a single pup, or a group of three to four pups using any non-inhalant anesthetic (consult with a facility veterinarian for the appropriate anesthetic) and place it (or them) in a separate holding cage containing home cage shavings, on a regulated heating pad. An anesthetic should be selected based on its safety, its long duration, and low respiratory depressant potential. Tape the dam’s nipple line with fabric-backed adhesive tape to prevent nipple attachment. At the end of the initial isolation period, place the stimulus animal(s) in the test chamber in the prone position in a predetermined place on the floor. Pups may be placed in the middle, but it is best to place the dam against one wall and rotate her slightly towards the wall so the outward nipple line is well hidden, so that pups do not spend their time attempting to attach to a teat. Pick up the isolated pup 1–2 cm from the floor and place it in snout contact with the anesthetized stimulus animal(s), to allow a uniform start point for the test and to assure that the pup locates the stimulus animal(s). Record ultrasonic vocalization and pups’ behaviors for a second 2–5 min in 1-min epochs. Measure and record duration of time out of contact with the stimulus animal(s). Include a control group in which no contact (stimulus) animal is present in the second test period. Pick these pups up, replace them on the test chamber floor, and continue to assess any changes in behavior and other measures that may occur simply as a result of the brief handling and the additional time in the test chamber. There is usually no significant change in behavior measures, although a gradual decline over time might be observed in the control group. Certain drugs or manipulations may alter this pattern; consequently the control group becomes essential for analyzing and interpreting the contact quieting effect following isolation in the experimental group. Record pup’s core temperature and weight at the end of each test. Return the stimulus animals to their cages and change or clean the test chamber for the next pup to be tested.

17. Maternal Potentiation This procedure assesses the isolated pup’s capacity to regulate its ultrasonic vocalization rate in response to cues present in its immediate environment. The following protocol assesses the pup’s capacity to regulate ultrasonic vocalization in response to

80

Winslow

its recent past experience. This is a more complex response than either the isolation or the contact quieting response, and it develops later, after the first week postpartum in the rat. It is not known whether it is present in any strain of mouse. Isolated rat pups’ ultrasonic vocalization rate is doubled after brief periods of contact with an anesthetized lactating female and tripled if it has interacted with an active dam (12, 13). The effect of the anesthetized dam (passive maternal potentiation) is mediated primarily by olfactory cues, whereas active maternal potentiation occurs even in anosmic pups and works through behavioral interactions such as retrieving, licking, and stepping on the pup. Littermates have no potentiating effects, even though they inhibit pups’ ultrasonic vocalization rates as much as the dam. Handling and/or transporting the pup has no consistent potentiating effect. Both active and passive maternal potentiations are specific for ultrasonic vocalization and do not increase any other behaviors or affect pup’s body temperature at any age. Short (1–2 min) periods with the dam are more effective than longer (5 min) periods for passive potentiation, and if passive contact is prolonged (e.g., 30–60 min), then the subsequent isolation ultrasonic vocalization response no longer shows potentiation. Active potentiation is more effective after 5 min than 1 min. If pups are not isolated immediately following brief maternal (passive) contact, but instead are replaced in the home cage litter group, then their second isolation response shows no more potentiation after 5–6 min. Perform pretest preparations as described previously except that an active dam can be used instead of an anesthetized one and the period of pup’s interaction with the dam can take place in the dam’s holding cage (often more convenient) rather than in the test chamber. To observe potentiation, simply follow the quieting protocol and then isolate the pup for a second time immediately after dam contact either by removing the anesthetized dam from the test chamber or by transporting the pup to the dam’s holding cage and back to the test chamber for the second isolation. With passive dam contact, potentiation is the same intensity whether it is the dam or pup that is transported back and forth. Active potentiation is better conducted in the dam’s holding cage before the second isolation. Record pup’s axillary or core temperature and weight at the beginning and end of the test. Return the stimulus animals to their cages and change or clean the test chamber for the next pup to be tested. Calculate the difference between a pup’s first and second separation tests to determine the measure of maternal potentiation for that pup. Include control groups of pups picked up and transported to a second novel test chamber between the first and second separation tests. Such control pups do not usually show any significant trend in ultrasonic vocalization rates over the three control

Ultrasonic Vocalizations by Infant Mice

81

test periods (12). As previously, it is possible to assign individual pups either to a potentiated or a standard isolation condition instead of a within-pup design. This can be done by transporting the test pup from the home cage litter group directly to the maternal holding cage, or to the test chamber in which the dam had previously been placed, for 1–5 min. The pup is then isolated, either by transporting it to the test chamber or by removing the dam from it, and ultrasonic vocalization and other behaviors observed.

18. Data Analysis Considerations Sum the ultrasonic vocalization and other behavioral counts for each minute of the experiment and analyze the results by repeated measures analysis of variance (Fig. 5.1). Representative Infant Mouse Ultrasonic Vocalizations Reflect Individual Differences Spectrograms Amplitude Waveforms 100

Frequency (kHz)

50 25 0 100 50 25 0

0

25

50

100 0 25 Time (msecs)

50

100

Fig. 5.1. Acoustic structure of ultrasonic pup and adult calls emitted by CBA/CaJ mice. Spectrograms (upper left) of pup-call bouts recorded from individually isolated mouse pups between postnatal days 5 and 12. Spectrograms of adult-call bouts recorded when a female was placed into the home cage of a male (upper right). The structure of the two types of vocalizations is quite different, despite the fact that both are ultrasound whistles. Pup calls show very little frequency modulation (except for calls like the first one in panel that jump in frequency, producing a small island near 4.5 kHz/ms), whereas adult calls can have much larger frequency sweeps. Modified from (53).

Ultrasonic vocalization rate responses to isolation show several characteristics that call for thoughtful statistical analysis, in addition to the paired comparisons of ambient temperature subgroups as just described. First, the distribution of individual ultrasonic vocalization rates is skewed, with a few high values contributing disproportionately to the mean, and, in some

82

Winslow

experiments, the number of animals at a floor of zero calls adds to the problem. Furthermore, there is always a high degree of inter-individual variability, about half attributable to litter and half occurring between litters. Thus, all experimental conditions should be included in each litter and litter mean values used in statistical comparisons. For small samples (e.g., n = 6–8) and high, skewed variability, medians and non-parametric statistics are necessary to properly represent the data. However for samples above eight, in which there is a more normal distribution of rates, analysis of variance gives much the same estimates of probability as non-parametric measures and permit two- and three-way analyses that are impossible with the latter. There is no evidence for significant differences between male and female rat or mouse pups in ultrasonic vocalization rate responses. However, the duration of individual calls is longer in male rat pups under some conditions (54). References 1. Newman JD. Neural circuits underlying crying and cry responding in mammals. Behavioural Brain Research 2007;182:155–65. 2. Dunn PM. Aristotle (384–322 BC): philosopher and scientist of ancient Greece. Archives of Disease in Childhood 2006;91:F75-7. 3. Blumberg MS, Alberts JR. Ultrasonic vocalizations by rat pups in the cold: an acoustic by-product of laryngeal braking? Behavioral Neuroscience 1990;104:808–17. 4. Blumberg MS, Sokoloff G. Do infant rats cry? Psychological Review 2001;108:83–95. 5. Panksepp J. Can anthropomorphic analyses of separation cries in other animals inform us about the emotional nature of social loss in humans? Comment on Blumberg and Sokoloff (2001). Psychological Review 2003;110:376–88; discussion 89–96. 6. Hofer MA, Shair HN. Ultrasonic vocalization, laryngeal braking, and thermogenesis in rat pups: a reappraisal. Behavioral Neuroscience 1993;107:354–62. 7. Noirot E. Ultrasounds and maternal behavior in small rodents. Developmental Psychobiology 1972;5:371–87. 8. Nitschke W, Bell RW, Zachman T. Distress vocalizations of young in three inbred strains of mice. Developmental Psychobiology 1972;5:363–70. 9. Motomura N, Shimizu K, Shimizu M, et al. A comparative study of isolation-induced ultrasonic vocalization in rodent pups. Experimental Animals 2002;51:187–90. 10. Shair HN, Brunelli SA, Masmela JR, Boone E, Hofer MA. Social, thermal, and temporal

11.

12.

13.

14.

15.

16.

17.

18.

19.

influences on isolation-induced and maternally potentiated ultrasonic vocalizations of rat pups. Developmental Psychobiology 2003;42:206–22. Winslow JT, Insel TR. The infant rat separation paradigm: a novel test for novel anxiolytics. Trends in Pharmacological Sciences 1991;12:402–4. Hofer MA, Masmela JR, Brunelli SA, Shair HN. The ontogeny of maternal potentiation of the infant rats’ isolation call. Developmental Psychobiology 1998;33:189–201. Hofer MA. Multiple regulators of ultrasonic vocalization in the infant rat. Psychoneuroendocrinology 1996;21:203–17. Brudzynski SM. Principles of rat communication: quantitative parameters of ultrasonic calls in rats. Behavior Genetics 2005;35:85–92. Brudzynski SM, Kehoe P, Callahan M. Sonographic structure of isolation-induced ultrasonic calls of rat pups. Developmental Psychobiology 1999;34:195–204. Olivier B, Molewijk E, van Oorschot R, et al. New animal models of anxiety. European Neuropsychopharmacology 1994;4:93–102. Rodgers RJ. Animal models of ’anxiety’: where next? Behavioural Pharmacology 1997;8:477–96; discussion 97–504. Moriceau S, Sullivan RM. Neurobiology of infant attachment. Developmental Psychobiology 2005;47:230–42. Hofer MA. Early relationships as regulators of infant physiology and behavior. Acta Paediatrica Supplementum 1994;397:9–18.

Ultrasonic Vocalizations by Infant Mice 20. Hofer MA. Early social relationships: a psychobiologist’s view. Child Development 1987;58:633–47. 21. Masi G, Mucci M, Millepiedi S. Separation anxiety disorder in children and adolescents: epidemiology, diagnosis and management. CNS Drugs 2001;15:93–104. 22. Hofer MA, Shair HN, Brunelli SA. Ultrasonic vocalizations in rat and mouse pups. Current protocols in neuroscience/editorial board, Jacqueline N Crawley [et al. 2002;Chapter 8 Unit 8 14]. 23. Blumberg MS, Sokoloff G, Kent KJ. Cardiovascular concomitants of ultrasound production during cold exposure in infant rats. Behavioral Neuroscience 1999;113:1274–82. 24. Hofer MA, Shair HN. Ultrasonic vocalization by rat pups during recovery from deep hypothermia. Developmental Psychobiology 1992;25:511–28. 25. Brunelli SA, Hofer MA. Development of ultrasonic vocalization responses in genetically heterogeneous National Institute of Health (N:NIH) rats. II. Associations among variables and behaviors. Developmental Psychobiology 1996;29:517–28. 26. Brunelli SA, Shair HN, Hofer MA. Hypothermic vocalizations of rat pups (Rattus norvegicus) elicit and direct maternal search behavior. Journal of Comparative Psychology 1994;108:298–303. 27. Allin JT, Banks EM. Functional aspects of ultrasound production by infant albino rats (Rattus norvegicus).Animal Behaviour 1972;20:175–85. 28. Benton D, Nastiti K. The influence of psychotropic drugs on the ultrasonic calling of mouse pups. Psychopharmacology (Berl) 1988;95:99–102. 29. Newman JD, Winslow JT, Murphy DL. Modulation of vocal and nonvocal behavior in adult squirrel monkeys by selective MAOA and MAO-B inhibition. Brain Reseach 1991;538:24–8. 30. Brunelli SA, Hofer MA, Weller A. Selective breeding for infant vocal response: a role for postnatal maternal effects? Developmental Psychobiology 2001;38:221–8. 31. Brunner D, Buhot MC, Hen R, Hofer M. Anxiety, motor activation, and maternal-infant interactions in 5HT1B knockout mice.Behavioral Neuroscience 1999;113:587–601. 32. Winslow JT, Hearn EF, Ferguson J, Young LJ, Matzuk MM, Insel TR. Infant vocalization, adult aggression, and fear behavior of an oxytocin null mutant mouse. Hormones and Behavior 2000;37:145–55.

83

33. Crawley JN. Designing mouse behavioral tasks relevant to autistic-like behaviors. Mental retardation and developmental disabilities Research Reviews 2004;10:248–58. 34. Ehret G, Buckenmaier J. Estrogen-receptor occurrence in the female mouse brain: effects of maternal experience, ovariectomy, estrogen and anosmia. Journal of Physiology, Paris 1994;88:315–29. 35. Koch M, Ehret G. Estradiol and parental experience, but not prolactin are necessary for ultrasound recognition and pupretrieving in the mouse. Physiology & Behavior 1989;45:771–6. 36. Smotherman WP, Bell RW, Starzec J, Elias J, Zachman TA. Maternal responses to infant vocalizations and olfactory cues in rats and mice. Behavioral Biology 1974;12:55–66. 37. Hahn ME, Hewitt JK, Adams M, Tully T. Genetic influences on ultrasonic vocalizations in young mice. Behavior Genetics 1987;17:155–66. 38. Thornton LM, Hahn ME, Schanz N. Genetic and developmental influences on infant mouse ultrasonic calling. III. Patterns of inheritance in the calls of mice 3–9 days of age. Behavior Genetics 2005;35:73–83. 39. Roubertoux PL, Martin B, Le Roy I, et al. Vocalizations in newborn mice: genetic analysis. Behavior Genetics 1996;26:427–37. 40. Brunelli SA, Hofer MA. Selective breeding for infant rat separation-induced ultrasonic vocalizations: developmental precursors of passive and active coping styles. Behavioural Brain Research 2007;182:193–207. 41. Brunelli SA, Myers MM, Asekoff SL, Hofer MA. Effects of selective breeding for infant rat ultrasonic vocalization on cardiac responses to isolation. Behavioral Neuroscience 2002;116:612–23. 42. Brunelli SA, Vinocur DD, Soo-Hoo D, Hofer MA. Five generations of selective breeding for ultrasonic vocalization (USV) responses in N:NIH strain rats. Developmental Psychobiology 1997;31:255–65. 43. Winslow JT, Insel TR. Endogenous opioids: do they modulate the rat pup’s response to social isolation? Behavioral Neuroscience 1991;105:253–63. 44. Shair HN, Brunelli SA, Hofer MA. Lack of evidence for mu-opioid regulation of a socially mediated separation response. Physiology & Behavior 2005;83:767–77. 45. Hofer MA, Brunelli SA, Shair HN. Potentiation of isolation-induced vocalization by brief exposure of rat pups to maternal cues. Developmental Psychobiology 1994;27:503–17.

84

Winslow

46. Ricceri L, Moles A, Crawley J. Behavioral phenotyping of mouse models of neurodevelopmental disorders: relevant social behavior patterns across the life span. Behavioural brain research 2007;176:40–52. 47. Scattoni ML, Crawley J, Ricceri L. Ultrasonic vocalizations: A tool for behavioural phenotyping of mouse models of neurodevelopmental disorders. Neuroscience and biobehavioral reviews 2008. 48. cattoni ML, Gandhy SU, Ricceri L, Crawley JN. Unusual repertoire of vocalizations in the BTBR T+tf/J mouse model of autism. PLoS ONE 2008;3:e3067. 49. Gaub S, Ehret G. Grouping in auditory temporal perception and vocal production is mutually adapted: the case of wriggling calls of mice. Journal of Comparative Physiology.

50. 51.

52.

53.

54.

A, Neuroethology, Sensory, Neural, and Behavioral Physiology 2005;191:1131–5. Roberts LH. The rodent ultrasound production mechanism.Ultrasonics 1975;13:83–8. Goodrich CA. Measurement of body temperature in neonatal mice. Journal of Applied Physiology 1977;43:1102–5. Hofer MA, Shair H. Sensory processes in the control of isolation-induced ultrasonic vocalization by 2-week-old rats. Journal of Comparative and Physiological Psychology 1980;94:271–9. Liu RC. Prospective contributions of transgenic mouse models to central auditory research. Brain Research 2006;1091:217–23. Naito H, Tonoue T. Sex difference in ultrasound distress call by rat pups. Behavioural Brain Research 1987;25:13–21.

Chapter 6 The Forced Swimming Test in Mice: A Suitable Model to Study Antidepressants Martine Hascoe¨t and Michel Bourin Abstract Among all animal models, the forced swimming test (FST) remains one of the mostly used tools for screening antidepressants with different mechanisms of action. This chapter reviews the main aspects of the FST in mice. Most of the sensitivity and variability factors that were assessed on the FST are summarized, as well as the most relevant data found in the literature of antidepressant effects on the FST in mice. From this data set, we have extrapolated some information about baseline levels of strain, and sensitivity against antidepressants. We have shown that many parameters have to be considered in this test to gain good reliability. Moreover, there was a fundamental inter-strain difference of response in the FST. The FST is a good screening tool with good reliability and predictive validity. Strain is one of the most important parameters to consider, for example Swiss and NMRI mice can be used to discriminate the mechanism of action of antidepressants; the CD-1 strain seems to be the most useful strain for screening purposes, but all results need to be arbitrated with spontaneous locomotor activity studies. Key words: Antidepressants, animal model, forced swimming test, mice.

1. Introduction Porsolt et al. (1) described ‘‘a new behavioural method for inducing a depressed state in mice’’. The idea arose out from some learning experiments which were released with rats in a water maze. Porsolt et al. (1) not only observed that most of the rats were finding the exit within 10 min, but also noticed that other rats ceased struggling altogether and remained floating passively. To describe this new behavioural model in mice (CD Charles River male of 20–25 g), the following procedure was adopted: ‘‘1 h after a single i.p. injection mice were dropped into the cylinder [height 25 cm, diameter 10 cm, 6 cm of water at 21–23C] and left T. D. Gould (ed.), Mood and Anxiety Related Phenotypes in Mice, Neuromethods 42, DOI 10.1007/978-1-60761-303-9_6, ª Humana Press, a part of Springer ScienceþBusiness Media, LLC 2009

85

86

Hascoe¨t and Bourin

for 6 min. Because little immobility is observed during first 2 min, only that occurring during the last 4 min was counted. The duration of immobility occurring in each minute was scored. A mouse was judged to be immobile when it ceased struggling and remained floating motionless in the water making only movements necessary to keep its head above water’’ (1). In the same paper, Porsolt et al. used the FST to test a large range of antidepressants and noticed a reduction in the immobility of mice for the entire products tested. This reduction in immobility was identified with an antidepressive-like effect. The results obtained with this test were comparative with the other usual clinical therapies which are also effective (e.g. electroconvulsive shock or selective deprivation of REM sleep) (1, 2). The aims of this chapter are mainly to review the characteristics of this model: the forced swimming test (FST) in mice (see Fig. 6.1), to discuss about the main parameters that influence the sensibility of this model and to summarize the advantages and drawbacks of this model in mice, as well as the factors of variability of the test through an extensive review of the literature. Pre-clinical data of drugs with various mechanism of action, obtained from literature and from our laboratory, using the FST are summarized.

Fig. 6.1. Mice in the forced swimming test (FST).

1.1. FST Validity

To evaluate the validity of an animal model, many criteria have to be explored. For example, we could consider reliability and different types of validity such as predictive, face, construct,

The Forced Swimming Test in Mice

87

aetiological, concurrent and discriminate. It was argued that there are only two criteria that a model must satisfy to establish its value in basic neurobiological research: reliability and predictive validity. Undoubtedly, the more types of validity a model satisfies, the greater is its value, utility and relevance to the human condition. FST has a strong predictive validity, a good reliability, some face validity and a poor construct validity (for an extensive analysis see (3)). In a comparative review of drug effects on immobility time in mice, Borsini and Meli (4) adopted a limit of 20% reduction in immobility to consider an antidepressant effect on the FST. In that case, 94% of antidepressants decrease the immobility time in mice (4). In the same paper, the authors also discussed the lack of specificity of the test, as 83% of class of drugs decrease immobility, but it may be mainly explained by methodological considerations. Some authors changed the scoring method, and other authors recorded animal movements by using automated devices. Nevertheless, false-positive effect of motor activity-enhancing drugs would have to be detected with an independent animal model of spontaneous locomotor activity like the actimeter, where psychostimulant drugs could reduce immobility without antidepressant effect (5). However, the FST is a suitable model to detect antidepressants due to the fact that it detects the majority of antidepressants and discriminates antidepressants from neuroleptics and anxiolytics (4) validating the predictive validity. Concerning the reliability criteria, the FST is currently a popular model due to the low cost of the experiments and because it is arguably the most reliable model available. Moreover, it has been reported to be reliable across laboratories. The passive behaviour observed in the FST could be considered as unwillingness to maintain effort in this inescapable situation. Immobility may be seen as an adaptative response to an inescapable situation. This strategy could be perceived as a successful coping rather than a failure of coping (6). As face validity for the FST with mice is not strong, chronic administration remains to be fully studied in order to raise this face validity. In a review of the causes of immobility in the FST, West (6) concluded that FST ‘‘no longer appears to be a valid model of depression. Nonetheless the FST is still likely to be useful in understanding antidepressants treatments’’. This point of view should be moderated by a consideration on ‘‘what is a valid model of depression?’’ When pre-clinical tests were created to study the depressive state, the first role for models of depression was to predict antidepressant potency. Moreover, the validity of these tests was largely based on an empirical observation, namely that the two major groups of antidepressants, MAO-I and tricyclic drugs (TCAs), are active (7).

88

Hascoe¨t and Bourin

The FST, as described by Porsolt et al. (1), has been designed to be ‘‘a primary screening test for antidepressants’’. For this purpose, FST is a good model for screening antidepressants, maybe the best one. FST shows a strong sensitivity to monoamine alterations, but it should not be forgotten that other antidepressant treatments, such as electroconvulsive shock, are efficient (1). To summarize these ideas, we can consider the FST model as a very specific model, where behaviours are induced by stress with no direct relation to symptoms in humans, but which are extremely sensitive to monoaminergic manipulations. Additional possibilities for the FST should be considered on a more neuropharmacological point of view. This test also provides a useful model to study neurobiological and genetic mechanisms underlying stress and antidepressant responses (8, 9, 10). Moreover, new approaches of research for antidepressant treatments continue to use the FST as a preliminary test. For example, some authors work on neurotrophic factor that could potentially be used in the treatment of depression. They used the FST and showed that brain-derived neurotrophic factor (BDNF) infusion in the ventral tegmental area resulted in 57% shorter latency to immobility relative to control animals, in the FST in rats (11). This use of the FST had already been described previously with a 70% decrease in the immobility time compared to vehicle-infused controls after BDNF infusion (12). Other ways of investigation for depression use the FST as a model of depression. Acute antidepressant treatment attenuates swim stress-induced corticosterone release in the rat (13). NK2-receptor antagonists, Kþ channel openers and Kþ channel blockers were considered for their antidepressant-like properties in the FST (14, 15, 16). Nitric oxide synthase (NOS) or neurosteroids have been tested in the FST with mice to look for an antidepressant-like effect (17, 18). Many studies keep using the FST, not only for screening for antidepressant effects, but also for a more neuropsychological purpose. This utilization of the FST differs from the monoaminergic purpose it is often used. Nevertheless, this model of depression is not only linked to monoamines. The uncontrollable stress involved during the test may implicate many mechanisms of reaction that could be considered as possible investigation ways. The fact that electroconvulsive seizures are effective in the test argues for its ability to pick up broader mechanisms of action (10). The relevance of using the FST for this new ways of research needs clinical correlations to validate also the FST for this utilization. The development of clinically effective antidepressant drugs with novel mechanisms should give answers to this question. Another point is the utilization of the FST for genetically modified animals that is applicable to study mechanisms of action of antidepressants on the test. For example, the decrease in immobility observed after paroxetine administration in wild-type mice is absent in 5-HT1B-knockout mice in the test (19). Other data with

The Forced Swimming Test in Mice

89

knockout mice can be useful to determine the role of NA or 5-HT in the test, for example with mice lacking serotonin transporter (20) or with dopamine-beta-hydroxylase-deficient mice (21). This new employment of the test permits to better know the mechanisms of action of drugs on the FST involving or not the monoamine, i.e. for new possible therapies, for example, BDNF –þ/– mice (22) or inducible BDNF-knockout mice (23). 1.2. Modifications of the FST

There have been many modifications of the FST, but improvements of the test are often poorly validated (24). Many parameters have been assessed in order to increase the sensitivity, specificity and reliability of detection of antidepressant activity (Table 6.1).

Table 6.1 Summary of some modifications tested on the FST in mice Factor

Variability

References

Acute versus chronic administration

X

(25)

Age of mice

X

(26, 27)

Automated device/water waves

Sensitivity

X

Circadian rhythm

(28, 29) X

(30)

Cylinder diameter

X

(31)

Depth of water

X

(32)

Environment of the laboratory

X

(33)

Food restriction

X

(34, 35)

Gender

X

(36–38)

Housing of animals

X

(39)

Isolation of animals Interval of observation

(40, 41) X

Observer

(31–42) X

–

Revised scoring

X

(42)

Scoring on categorized behaviour

X

(43)

Side preference in rotation

X

(44)

Strains

X

(9, 38, 45, 46)

Test/retest

X

(47)

Time between treatment and FST

X

–

Water temperature

X

(48, 49)

90

Hascoe¨t and Bourin

1.2.1. Sensitivity Factors

Automated device/water waves: Different procedures have been elaborated to automate the FST. Video-tracking, computer analysis or wave analysis were used to score the immobility of rodents. From the data set, one can extract full or partial turns, clockwise or counter-clockwise rotations, total activity, and speed of swimming clockwise and counter-clockwise (29). Another author used an apparatus consisting of a transparent plastic cylinder (10  20 cm) containing 7 cm of water (23C). Movement by the animal created a wave form in the water resulting in a converted digital signal (28). The ease of use of these systems appears not to counterbalance the cost of the equipment. Few studies use an automated video-tracking device and mainly as a confirmation tool (11). Nevertheless, some automated devices employed in FST studies were reported to be reliable for antidepressant screening (50). Cylinder diameter: To test this parameter, mice were forced to swim for 15 min in tanks of 10 (the original diameter of the Porsolt’s forced swimming chamber), 20, 30, and 50 cm diameter in 20-cm deep water. Modifications of this parameter provide a way to distinguish the antidepressant drugs from caffeine, anticholinergics and antihistaminics, which gave a false-positive response in 10-cm diameter cylinders. The selective effect of antidepressants, namely, the rotatory locomotor activity during swimming can also be studied (31). In our laboratory, we use a cylinder with the closest available diameter to the original test’s diameter, associated with a check of variation of locomotor activity that can discriminate false-positive effects (5). Depth of water: This parameter had to be considered, as mice should not sense a limit under the level of water. Their tails should not touch the bottom of the cylinder or the behaviour of the mice would be altered. Increased depth of water decreases the time spent immobile. No paper clearly described this process in mice; this parameter was shown to alter the behaviour of the rat (4, 51). The original description of the FST by Porsolt et al. (1) explains that 6 cm of water is sufficient. But mice can sense the bottom of the cylinder with this level of water. In our laboratory, the water level is at least 10 cm. Some modifications of Porsolt’s paradigm have often been used; one of the most quoted is the method of Alley and Kulkarni (32). They measure immobility in a glass jar (21  12 cm) containing 12 cm of water maintained at 22–1C, during a 6-min period. It is important to consider that the only main modification of the original test is the increased depth of water. This procedure is consistent with the one we use, and should be considered as the actual standard method. Interval of observation/scoring: Porsolt’s paradigm has been modified by some researchers in order to increase the sensitivity or the specificity of the FST. Some authors have created a totally new analysis procedure for scoring immobility. The observation interval can be separated into 5-s parts in which the main behaviour is

The Forced Swimming Test in Mice

91

scored (42). Analysis of the behaviour of the mice can be totally different with categorization of a specific behaviour (43). Some authors made a series of observations at 30-s intervals, and the mouse was rated as immobile (score 0) or not (score 1) for each observation period (4). Time between treatment and FST: This factor is not often considered, but may explain some of the differences between FST results. Two possibilities seem to be available: acute injection 1 h before the FST as described by Porsolt et al. (1) or acute injection after the FST when the maximal effect is intended. This requires a time-course study of the drug effect. Water temperature: The influence of water temperature on immobility time of the mice was studied. An effect of water temperature was revealed; a higher temperature (35C) resulted in shorter immobility time, after 10 min of forced swimming (48). Other data suggest that immobility, which develops rapidly during forced swimming in cold water, may result from dramatic inhibition of neural functions because of severe brain hypothermia (49). Currently, most studies use warmer water between 23 and 28C. In our laboratory, we choose a temperature between 23 and 25C. Wheel water tank: Some authors have tried to measure immobility time in another way. A wheel was immerged in the water tank. Mice placed on this apparatus keep turning the wheel vigorously; when they abandon their attempts to escape from the water, the wheel stops turning. The number of rotations of the water wheel is counted. All antidepressants tested increased the number of rotations. As tranquillizers, anticholinergics and antihistaminics were not effective. It was suggested that this water wheel test was more appropriate as screening test for antidepressants than Porsolt’s test with regard to both objectivity and specificity (52). 1.2.2. Variability Factors

Acute versus chronic administration: The effectiveness of acute treatment is a particularity of the FST. Useful for a screening test, it appears to decrease the face validity of this model, as the clinical time course requires chronic administration to be active. Experiments were made to find out the effects of chronic administration on the FST. Subchronic or acute effects were increased by chronic administration (25). Age of the mice: This parameter should be considered in parallel with weight. Our team has already shown a strong difference between younger and older mice groups. Sensitivity to some antidepressants is profoundly altered. Tricyclic, noradrenalin reuptake inhibitors and serotonin reuptake inhibitors were more active in 4-week-old mice than 40-week-old Swiss mice (26, 27). In our laboratory experiments, we choose mice weighing 20–25 g. Circadian rhythm: An effect of circadian rhythm was shown in response to antidepressants in the FST. FST was carried with three strains of mice: C3H, C57BL/6 J and ND/4. Immobility time was

92

Hascoe¨t and Bourin

scored at noon (12:00–14:00 h) and midnight (00:00–02:00 h). For C3H/Hen mice, duration of immobility was greater at midnight (30). Another study did not show any difference between the FST made at noon (11:00–12:00 h), early dark (20:00–21:00 h) and at midnight (1:00–2:00 h) for BALB/c and C57BL/6 J mice (53). Genetic studies on Clock gene, implicated in circadian rhythm, revealed an effect of this parameter on immobility time (54). Studies in our laboratory are only made between 8:00 and 12:00 h to avoid any risk of behavioural modification throughout the experiments. Environment of the laboratory: Interactions with laboratory environment have been studied in several strains of mice on few behavioural tests (open field, elevated plus maze, water maze, alcohol preference) (33). Despite standardization, there were systematic differences in behaviour across three different labs. In our opinion, FST is less sensitive to variation of laboratory environment (noise, air temperature, light, atmosphere pressure). Food restriction: Food restriction can strongly modify behavioural responses, as shown with amphetamine or the FST. The authors used FST for two sessions with two groups of DBA/2 mice. One group was isolated and food-restricted, the other group was isolated but had free access to food. Immobility time was significantly decreased in the food-restricted group compared to the other group (34, 35). Gender: Differences in sensitivity between male and female mice were revealed by some studies depending on the strain used. David et al. (37) described a different sensitivity to antidepressants in the FST related to gender. Imipramine and paroxetine were active on CD1 male and female mice, but at different doses. Another study showed a difference between male and female mice, but only in some strains; FVB females, for example, had a smaller floating time than males (38). Sexual differences have also been described in another study of immobility, which was higher in males than in females (36). Housing of animals/isolation of animals: All studies have shown that housing was a critical parameter. In the abovementioned study of Cabib (see Food restriction section; (34), (35)), a group of DBA/2 mice was isolated for 13 days and compared with group-housed mice in the FST. They showed a significant increase in the immobility time in the isolated group (35). Yates et al. (41) linked this difference with the age of the mice. After having isolated the mice for 24 h prior to a 15-min FST, they showed an increase in immobility time in 17- to 21-day-old Swiss Webster mice but not in 26- to 30day-old mice. In another study, the immobility time in the FST was shortened in NIH Swiss mice isolated for 2 or 5 days, suggesting an improved ability to cope with stressful situations (40, 41). Isolation seems to have strain-dependent effects on the FST, but none of these studies had the same isolation time. If isolated for a longer period (8 weeks), mice displayed

The Forced Swimming Test in Mice

93

lower levels of immobility time when exposed to this test (39). Nevertheless, isolation, e.g. for a surgery, had to be specified in methods of a paper, as it may modify dramatically immobility time of the FST. Observer: The most important source of variability (and the best way to consider in order to increase the sensitivity of the FST), with identical environmental parameters, is the observation. Like all behavioural studies, the observer is the main actor of the test, and reproducibility between laboratories is a matter that affects all these tests. The scoring of the immobility time should be strongly considered and assessed by all teams. The mouse is judged to be immobile when it makes only movements necessary to keep its head above water. It can move in the cylinder but without struggling movements. The analysis of active behaviours in the FST has strengthened the possibility of replicating the experiments. Side preference in rotation: A study was made on side rotational preference of mice during the FST. Krahe et al. (44) concluded that side preferences of spontaneous rotational behaviour may account for inter-individual differences. Strains: Strain is one of the most important parameters to deal with (9). Important differences exist between strains in both immobility observed and effects of imipramine (5). Genetic background could modify response by providing an inappropriate baseline level of behaviour (55). There is a maximal 10-fold difference in baseline immobility scores in control animals between strains and baseline level that does not correlate with antidepressant sensitivity (9). Several gender dissociations suggest the strain and task specificity (38). Intra- and inter-strain comparisons indicate that the biological substrates mediating performance in the FST and the tail suspension test (TST) are not identical. For example, in NIH-Swiss mice, a sevenfold difference in baseline immobility was observed between the FST and TST. In contrast, the baseline immobility in C57BL/6 mice was similar in both procedures (45). There is a continuum of variation in basal responses from almost no time spent immobile by DBA/2 J mice to more than 210 s of immobility in a 360-s test session with Balb/cJ mice (56). In one of our studies, we have shown that drug sensitivity is genotype-dependent. FST results have shown that Swiss mice were the most sensitive strain to detect serotonin (5-HT) and/or noradrenalin (NA) treatment. The use of DBA/2-inbred mice may be limited, as an absence of antidepressant-like response was observed in the FST (46). Control mice from the same breeders with comparable housing conditions should have the same immobility time in all laboratories. However, a gene-environment interaction is possible and may account for some difference between laboratories (57). For example, in our data set, animals of the same strain that received no treatments do not have the same immobility time (for CD-1 from 135 s to 223 s of immobility time).

94

Hascoe¨t and Bourin

Test/retest: This method is used normally for rats. In a first session, the animal is able to discover the test; rat usually explores the water surface and dives. In a second session where they will be scored, rats are familiarized to the test and do not try to dive. Mice do not have this behaviour, and this explains the easy use of mice that do not need a second session. This second session has been assessed for the construct validity of the FST. Memory process was involved to explain immobility of the rat. The absence of second session with mice removes this problem and simplifies the test. In their experiments, Alcaro et al. (47) evaluated behavioural responses to FST in naive animals and in animals pre-exposed to the FST 14 days before the test session. They showed a major effect of the presession FST in mice on immobility time with a dramatic increase after pre-exposure. For Andreatini and Bacellar (58), ‘‘this test showed a very low intra-class correlation coefficient in the test-retest design, which suggests a poor reliability of these measures’’. These results suggest that the behavioural parameters of the behavioural despair are not stable. Therefore, they are possibly more related to state than trait characteristics, this test is not appropriate to evaluate trait characteristics which are supposed to be stable over time without treatment. Some authors use the test/retest paradigm to avoid variations and to maintain consistency in the immobility time between different groups (59). 1.3. Screening Purpose Decisional Tree

FST and TST (60) are the mostly common animal models of depression used for antidepressant screening. Both tests placed mice in an inescapable situation, and the antidepressant-like activity is expressed by the decrease in immobility duration. During the last decade, we have routinely used these models in our research laboratory not only to predict antidepressant-like activity of various compounds, but also to investigate their mechanism of action. Using various ligands, we demonstrated the important implication of 5-HT1A and 5-HT1B receptors in the mechanism of action of selective serotonin re-uptake inhibitors (SSRI). Recently, we published two studies establishing the impact of genetic factors in the efficiency of various antidepressants in both tests (46, 61). We have summarized all data previously obtained, in order to propose a strategy that could be used for the development of new potent antidepressants via the determination of the potent antidepressant-like activity and investigation of the mechanism of action. Our objectives were to detect the antidepressant-like effect of each compound using low doses (better specificity of action), and secondly, to obtain the greater effect-size (response amplitude) for each type of antidepressant regarding their mechanism of action. This last point is crucial, as the greater the effect-size, the

The Forced Swimming Test in Mice

95

easier the possibility of antagonizing the antidepressant effect and determining the receptor subtype, or transporter, implicated (see Table 6.2).

Table 6.2 The values represent the maximal significant percentage of decrease in immobility time for each drug in each test for the optimal dose. 0 indicates an absence of significant effect. Drugs were administered intraperitoneally 30 min prior to the test Imipramine

Desipramine

Paroxetine

Citalopram

Bupropion

FST (%)

TST (%)

FST (%)

TST (%)

FST (%)

TST (%)

FST (%)

TST (%)

FST (%)

TST (%)

20

49

22

55

23

65

27

57

0

47

NMRI

0

66

0

0

16

56

0

47

0

0

C57Bl/ 6J Rj

0

45

0

30

0

38

0

68

26

44

DBA/2

0

35

0

0

0

47

0

45

0

0

Swiss

The majority of the drugs were efficient in the FST in Swiss mice. To investigate the mechanism of action of a new potent antidepressant drugs, the FST is a more powerful tool in Swiss mice as the first step. TST utilizing Swiss mice can consistently illustrate antidepressant-like activity. The use of Swiss mice is of greater interest due to thee greater effect-size obtained. As the C57BL/6 and the DBA/2 mice attempted to redress their position (i.e. climbing up their tails previously reported by 62, 61), it was difficult to conclude on their activity in the TST. Swiss mice is the most sensitive strain to detect serotonin and/ or noradrenalin antidepressants, whereas C57BL/6 Rj was the only strain sensitive to bupropion (dopaminergic agent) using the FST. In the TST, all antidepressants studied decreased the immobility time in Swiss and C57BL/6 Rj strains. To investigate the mechanism of action of a potential antidepressant, the use of both tests is required with only three strains of mice (Swiss, NMRI and C57Bl/6 Rj). Some compounds with variable mechanisms of action (like TCAs and SSRIs), induce a similar response regardless of the test and the strain of mouse used. For these drugs, the mechanism of action may be investigated using additive compounds to potentiate or antagonize the response (63). According to all results, a decision tree was established (Fig. 6.2).

96

Hascoe¨t and Bourin

PUTATIVE ANTIDEPRESSANT

FST(Swiss mice)

–

+

TST(NMRImice)

–

+

NRI norepinephrine reuptake inhibitors e.g desipramine

FST(Swiss mice) + 5-HT1A agonist (8-OH-DPAT or antgonist (pindolol)

Antagonist

FST(C57BL/6

+ DRI dopaminergic reuptake inhibitors e.g bupropion

– Antidepressant Lacking monoaminergic reuptake inhibition

Antagonist Agonist

Agonist

SSRI Selective serotonin reuptake inhibitors e.g paroxetine ; citalopram or SNRI Serotonin and noradrenaline reuptake inhibitors e.g. venlafaxine

TCA Like drugs imipramine e.g desipramine

Fig. 6.2. A decision tree.

2. Effect of Some Antidepressants in the FST

2.1. Role of the FST in Evaluating Mood Stabilizers

Many antidepressants have been tested with the FST on mice. Some results available for all classes of antidepressants with different strains of mice are reviewed here. The most frequently used strains, CD1, NMRI and Swiss, have positive results with most of the antidepressants in the test. FST results coupled with a spontaneous locomotor activity test like the actimeter are validated (Tables 6.3–6.9). Bipolar disorder animal models are challenging to develop because of the complex alteration of mania, depression, euthymia and mixed stated. Several animal models are yet available, but they only reflect one part of the illness, depression or manic behaviour (87). Despite many years of research, no valid and satisfactory animal model of bipolar depression has been developed to evaluate the mechanism of action of mood stabilizers. These drugs have been evaluated in the FST, in order to understand their antidepressant activity.

190

Mianserin 44 51

Mianserin

47

Mianserin

89

88

210

Mianserin

185

56

190

Mianserin

106

44

Medifoxamine

100

Iprindole

156

103

Iprindole

156

<50

Iprindole

39

207

Iprindole

81

45

94

207

% Control

Iprindole

Treated (s) 110

Control (s)

Cor 32-24

AD

50

þ

32 128 20 20 40 10 32

0 þ þþ þ þþ þ þ

64

80

þþ

0

40

8

Dose (mg/kg)

þþ

0

Effect

NO

NO

NO

NO

NO

NO

YES

YES

NO

NO

NO

Yes

Actimetry

M

M

M

M

M

M

M

M

M

M

M

M

Gender

25–30

20–30

20–25

30–35

20–25

22–26

20–32

22–26

20–25

20–25

22–26

Weight (g)

CD-1/Charles River

CD-1/Charles River

CD-1/Charles River

CD-1/Charles River

CD-1/Charles River

Swiss/Janvier

Swiss/Janvier

Swiss/Janvier

HaM/ICR (Broekman inst.)

CD-1/Charles River

CD-1/Charles River

Swiss/Janvier

Strain

(continued)

(28)

(68)

(1)

(67)

(1)

(64)

(66)

(64)

(65)

(1)

(1)

(64)

References

Table 6.3 Atypical antidepressants; 0 no effect in the FST, þ antidepressant-like effect, þþ important antidepressant-like effect in the FST

The Forced Swimming Test in Mice 97

148

148

Mianserin

Mianserin

Bold: significant effect; Italic: no effect.

16

101

Trazodone

0

13

þ

50

5

þþ

Trazodone

32

þ

1 to >30

32

þ

0

32

20

þþ

þ

8

þþ

15

5

þ

0

56

Dose (mg/kg)

þþ

Effect

Trazodone

68

123

Minaprine

65

65

68

0

96

96

97

Mianserin

180

142

Mianserin

58

194

Mianserin

112

69

180

Mianserin

124

<50

% Control

Mianserin

Treated (s) 19

Control (s)

Mianserin

AD

Table 6.3 (continued)

YES

NO

NO

YES

YES

YES

YES

YES

YES

YES

NO

NO

Actimetry

M

M

M

F

M

M

M

M

F

F

M

M

Gender

22–26

20–30

18–23

22–26

20–27

20–24

18–33

24–30

24–30

25–30

Weight (g)

Swiss/Janvier

HaM/ICR (Broekman inst.)

CD-1/Charles River

CD-1/Charles River

Swiss/Janvier

Swiss/Janvier

Swiss/Janvier

Swiss

NMRI/Evic-Ceba

NMRI/Evic-Ceba

HaM/ICR (Broekman inst.)

CD-1/Charles River

Strain

(64)

(65)

(68)

(72)

(64)

(66)

(71)

(70)

(69)

(69)

(65)

(28)

References

98 Hascoe¨t and Bourin

210

118

88

Desipramine

Desipramine

50

13

Desipramine

Desipramine

55

100

Desipramine

55

100

68

55

79

Desipramine

60

65

165

77

Desipramine

110

143

Desipramine

90

63

90

65

29

143

72

Desipramine

30

33

Desipramine

102

Desipramine

30

35

30

32

72

% Control

100

90

Desipramine

30

62

66

111

Treated (s)

Desipramine

85

204

Desimipramine

Desipramine

204

155.2

Oxaprotiline

Desimipramine

Control (s)

AD

15 30 20 20 20

þþ þþ þþ þþ þþ

20 30 20

þ þ þ

25

20

þþ

þ

1–16

0

20

10

þ

0

1–16

0

20

5

þþ

0

Dose (mg/ kg)

Effect

NO

NO

NO

YES

NO

NO

NO

NO

NO

YES

NO

NO

NO

NO

NO

NO

YES

Actimetry

M

M

M

M

M

M

M

M

M

M

M

M

M

M

M

M

M

Gender

20–24

30–35

20–24

20–25

20–25

20–25

Weight (g)

HaM/ICR (Broekman inst.)

FVB/NJ(Jackson)

DBA/2 J(Jackson)

DBA/2

CF-1/Charles River

CD-1/Charles River

CD-1/Charles River

C57BL/6 J(Jackson)

C57BL/6 J(Jackson)

C57 BL/6Rj

C3H/HeJ Swiss Webster(Jackson)

BALC/cJ(Jackson)

A/J(Jackson)

129/SvemJ(Jackson)

CD-1/Charles River

CD-1/Charles River

Swiss

Strain

(continued)

(65)

(9)

(9)

(46)

(9)

(9)

(67)

(9)

(9)

(46)

(9)

(9)

(9)

(9)

(1)

(1)

(75)

References

Table 6.4 Noradrenalin reuptake inhibitors (SNRI); 0 no effect in the FST, þ antidepressant-like effect, þþ important antidepressantlike effect in the FST

The Forced Swimming Test in Mice 99

218

200

219.3

81

176

Desipramine

Desipramine

Desipramine

Desipramine

Viloxazine

Bold: significant effect; Italic: no effect.

Viloxazine

37

37

174

Viloxazine

64

37

174

Viloxazine

64

50

41

68

70

41

39

Viloxazine

72

55

152.5

81

86

88

219

Desipramine

193

82

221

Desipramine

181

78

38

% Control

Desipramine

20

Treated (s)

100

52

Control (s)

Desipramine

Desipramine

AD

Table 6.4 (continued)

45 16 16 16

þþ þþ þþ

32

þþ

þ

16

þþ

30

16

þþ

þ

4

þ

20

4

þ

0

4–16

0

1–16

20

þ 0

Dose (mg/ kg)

Effect

NO

?

YES

NO

NO

NO

YES

NO

NO

NO

NO

YES

YES

NO

Actimetry

M

M

M

M

M

M

M

M

M

M

M

M

M

M

Gender

22–26

20–28

20–24

20–25

20–24

41–51

41–51

18–22

35–45

20–24

20–24

Weight (g)

Swiss/Janvier

Swiss/Janvier

Swiss/Janvier

HaM/ICR (Broekman inst.)

CD-1/Charles River

Swiss-Webster/Charles River

Swiss/Janvier

Swiss/Janvier

Swiss/Janvier

Swiss/Janvier

Swiss/Janvier

Swiss/Janvier

NMRI

NIH-Swiss Harlan

Strain

(64)

(66)

(71)

(65)

(1)

(9)

(76)

(26)

(26)

(26)

(26)

(46)

(46)

(9)

References

100 Hascoe¨t and Bourin

100 78

230

200

215

230

200

200

210

Bupropion

Bupropion

Bupropion

Bupropion

Bupropion

Bupropion

Bupropion

Nomifensine 7 50

Nomifensine

33

47

44

37

23

Nomifensine

70

93

88

86

50

85

100

Bupropion

169

50

Bupropion

179

100

Bupropion

10

20

Bupropion

200

74

% Control

Bupropion

Treated (s) ?

Control (sec)

Amineptine

AD

1–16 16 16 32 32 32 32 10 10 10

þþ þ þþ þþ þþ þþ þ þþ þ

1–16

0

0

10

30

þþ

þ

2–4

þ

1–8

3

þ

0

Dose (mg/ kg)

Effect

NO

NO

NO

NO

NO

NO

NO

NO

NO

YES

YES

NO

YES

NO

YES

YES

Actimetry

M

M

M

M

M

M

M

M

M

M

M

M

M

M

F

Gender

25–30

30–35

41–51

18–22

35–45

41–51

18–22

35–45

20–24

20–24

20–24

20–26

20–24

18–23

Weight (g)

HaM/ICR (Broekman inst.)

CD-1/Charles River

CD-1/Charles River

Swiss/Janvier

Swiss/Janvier

Swiss/Janvier

Swiss/Janvier

Swiss/Janvier

Swiss/Janvier

Swiss/Janvier

NMRI

HaM/ICR (Broekman inst.)

DBA/2

CD-1/Charles River

C57 BL/6Rj

CD-1/Charles River

Strain

(65)

(28)

(67)

(26)

(26)

(26)

(26)

(26)

(26)

(46)

(46)

(65)

(46)

(74)

(46)

(73)

References

Table 6.5 Dopamine reuptake inhibitors; 0 no effect in the FST, þ antidepressant-like effect, þþ important antidepressant-like effect in the FST

The Forced Swimming Test in Mice 101

78

240

Venlafaxine

186

69

158

Minalcipran

109

71

% Control

Minalcipran

Treated (s) ?

Control (s)

Midalcipram

AD

Dose (mg/kg) 10 20 30 8

Effect þ þ þ þþ

YES

YES

NO

NO

Actimetry

M

M

M

M

Gender

20–24

25–30

22–26

20–26

Weight (g)

Swiss/Janvier

Swiss

Swiss/Janvier

Swiss

Strain

Table 6.6 Serotonin noradrenalin reuptake inhibitors; 0 no effect in the FST, þ antidepressant-like effect, þþ important antidepressant-like effect in the FST

(79)

(77)

(64)

(76)

References

102 Hascoe¨t and Bourin

210

Clomipramine

32

72

Dosulepine

86 50

180

Clomipramine

227

215

Clomipramine

71

64

180

Clomipramine 153

50

Ciclazindol 115

43

48

Chlorimipramine

Chlorimipramine

?

Amoxapine 84

50

Amitryptiline

175

1

Amitriptyline

76

179

Amitriptyline

136

70

90

27

129

45

36

Amitriptyline

166

Amitriptyline

60

% Control

25

166

Amitriptyline

Treated (s)

Amitriptyline

Control (s)

AD

Dose (mg/kg) 15 30 32 10 4 4 4 6 20 32 12 20 20 30 20 32

Effect þþ þþ þþ þþ þ þ þ þ þþ þþ þ þ þ þ þ þþ

YES

NO

NO

NO

NO

NO

NO

NO

NO

NO

YES

YES

YES

NO

NO

NO

Actimetry

M

M

M

M

M

M

M

M

M

M

M

M

M

M

M

M

Gender

20–25

30–35

30–35

35–45

25–30

20–25

22–26

20–25

18–31

25–30

20–25

20–25

Weight (g)

Swiss/Janvier

HaM/ICR (Broekman inst.)

CD-1/Charles River

CD-1/Charles River

CD-1/Charles River

HaM/ICR (Broekman inst.)

CD-1/Charles River

CD-1/Charles River

HaM/ICR (Broekman inst.)

HaM/ICR (Broekman inst.)

Swiss/Janvier

Swiss/Janvier

Swiss

CD-1/Charles River

CD-1/Charles River

CD-1/Charles River

Strain

(continued)

(71)

(65)

(67)

(85)

(84)

(65)

(28)

(1)

(65)

(65)

(64)

(66)

(70)

(28)

(1)

(1)

References

Table 6.7 Tricyclic antidepressants; 0 no effect in the FST, þ antidepressant-like effect, þþ important antidepressant-like effect in the FST

The Forced Swimming Test in Mice 103

120

149

Imipramine HCl

Imipramine HCl

180

Imipramine HCl

18

Imipramine HCl 100

50

Imipramine HCl

Imipramine HCl

74

Imipramine HCl

25

100

Imipramine HCl

44

21

39

80

40

Imipramine HCl

137.9

189

Imipramine HCl

21

189.2

Imipramine HCl

39.5

55

152.7

Imipramine HCl

84

100

65

58

Imipramine HCl

97

70

50

182

Doxepine

90.3

80

182

Doxepine

146

<50

32

Doxepine

72

% Control

44

227

Dothiepin

Treated (s)

Doxepine

Control (s)

AD

Table 6.7 (continued)

1–16

45

þ 0

3

þ

10

þþ

10

1

þþ

þ

30

þþ

1–16

30

þþ

0

5

þ

30

þ 1–16

15

þ

0

10

10

þ

þ

10

þ

2

32

þþ

0

Dose (mg/kg)

Effect

YES

NO

NO

NO

YES

NO

YES

NO

NO

N

YES

NO

NO

YES

YES

NO

NO

YES

Actimetry

M

M

M

M

M

M

F

M

M

M

M

F

M

F

F

M

M

M

Gender

20–24

25

5 week

20–24

25–30

18–23

20–25

25

20–24

20–25

20–30

24–30

24–30

25–30

20–24

Weight (g)

NMRI

NIH-Swiss Harlan

HaM/ICR (Broekman inst.)

ddY/SLC

DBA/2

CD-1/Charles River

CD-1/Charles River

CD-1/Charles River

CD-1/Charles River

C57 Harlan S-D

C57 BL/6Rj

BABL/cki/Kirschbaum Memorial

Laka/Panjab University

NMRI/Evic-Ceba

NMRI/Evic-Ceba

HaM/ICR (Broekman inst.)

CD-1/Charles River

Swiss/Janvier

Strain

(46)

(44)

(65)

(83)

(46)

(28)

(72)

(1)

(5)

(45)

(46)

(86)

(80)

(69)

(69)

(65)

(28)

(71)

References

104 Hascoe¨t and Bourin

224

139

218

224

Imipramine HCl

Imipramine HCl

Imipramine HCl

Imipramine HCl

150

150

Maprotiline

Maprotiline

197

Nortriptyline

33

Pizotifen

Italic: no effect.

50

7

Nortriptyline

13

66

197

Nortriptyline

130

86

226.1

Maprotiline

195

26

35

36

Maprotiline

53

54

78

139

Maprotiline

109

57

230.8

Imipramine HCl

131.7

25

66

54

25

21

Imipramine HCl

147

117

35

48

49

215

Imipramine HCl

106

32

64

200

Imipramine HCl

66

66

48

52

83

80

91

62.8

42.5

100

151

Imipramine HCl

138

94.6

Imipramine HCl

Imipramine HCl

89.1

192

Imipramine HCl

Imipramine HCl

182

Imipramine HCl

20 16 16 16 32 32 32 32 32 32 10 16 32 32 32 15 30 5 3

þþ þþ þþ þþ þþ þ þþ þþ þþ þþ þþ þþ þþ þ þ þþ þ þ

30

0

0

30

30

þþ 0

5

0

NO

NO

NO

NO

YES

NO

YES

YES

YES

YES

NO

YES

NO

YES

NO

NO

NO

YES

YES

NO

NO

YES

YES

M

M

M

M

M

M

M

M

M

M

M

M

M

M

M

M

M

M

M

M

M

F

F

20–30

20–25

20–25

20–24

22–26

20–24

20–26

18–32

20–24

22–26

20–24

41–51

20–24

18–22

35–45

41–51

20–24

18–30

24–30

24–30

CD-1/Charles River

HaM/ICR (Broekman inst.)

CD-1/Charles River

CD-1/Charles River

Swiss/Janvier

Swiss/Janvier

Swiss/Janvier

Swiss/Janvier

Swiss

Swiss/Janvier

Swiss/Janvier

Swiss/Janvier

Swiss/Janvier

Swiss/Janvier

Swiss/Janvier

Swiss/Janvier

Swiss/Janvier

Swiss/Janvier

Swiss

OF-1/Iffa Credo

NMRI/Evic Ceba

NMRI/Evic-Ceba

NMRI/Evic-Ceba

(68)

(65)

(1)

(1)

(76)

(64)

(71)

(66)

(70)

(76)

(64)

(71)

(26)

(66)

(26)

(26)

(26)

(46)

(70)

(5)

(5)

(69)

(69)

The Forced Swimming Test in Mice 105

300 30 200 400 20

þþ þ þþ þþ þþ

Italic: no effect.

193

Tranylcypromine

58

30

14

202

Toloxatone

29

50

202

Toloxatone

100

60

T-794

35

172

Pargyline

61

300

þ

34

210

Pargyline

72

4–32

150

10

þþ

þ

60

þþ

0

72

60

þ

8

60

þ

0

Dose (mg/ kg)

Effect

Nialamide

138

192

Nialamide

31

97

223

70

57

230

223

Lilly 51641

120

48

Moclobemide

210

Clorgyline

65

% Control

50

135

Clorgyline

Treated (sec)

Moclobemide

Control (sec)

AD

NO

NO

NO

NO

NO

NO

YES

NO

YES

NO

NO

NO

NO

Actimetry

M

M

M

M

M

M

M

M

M

M

M

M

M

Gender

20–25

20–25

20–25

5 weeks

20–25

30–35

20–33

20–25

18–22

20–25

20–25

30–35

20–25

Weight (g)

CD-1/Charles River

CD-1/Charles River

CD-1/Charles River

ddY/SLC

CD-1/Charles River

CD-1/Charles River

Swiss/Janvier

CD-1/Charles River

Swiss/Janvier

Balb/ca/SPF

CD-1/Charles River

CD-1/Charles River

CD-1/Charles River

Strain

(1)

(1)

(1)

(83)

(1)

(67)

(66)

(1)

(26)

(82)

(1)

(67)

(1)

References

Table 6.8 Monoamine reuptake inhibitors (IMAO); 0 no effect in the FST, þ antidepressant-like effect, þþ important antidepressantlike effect in the FST

106 Hascoe¨t and Bourin

226

Citalopram

186

Fluoxetine

174

94

84

204

Fluoxetine

172

58

120

Fluoxetine

70

21

Clovoxamine

77

178

Citalopram

69

53

34

155

113

34

Citalopram

231.1

215

Citalopram

57

32 16 16 32 5

þþ þþ þþ þþ þ

0

8

8

16

þþ

0

16

0

16

þ

170

27

Citalopram

58

211

Citalopram

4

0

76

179

Citalopram

136

8

þ

73

Citalopram

1 1–8

0

4

1–16

Dose (mg/ kg)

0

100

0

0

Effect

Citalopram

Citalopram

141

87

162

% Control

Citalopram

Treated (s) 100

Control (s)

Citalopram

AD

NO

YES

NO

NO

YES

NO

NO

NO

YES

YES

YES

YES

YES

YES

NO

YES

Actimetry

M

M

M

M

M

M

M

M

M

M

M

M

M

M

M

M

Gender

22–25

22–25

20–30

22–26

20–24

22–26

35–45

41–51

20–29

18–22

41–51

20–24

20–24

20–24

22–25

20–24

Weight (g)

CD-1/Charles River

CD-1/Charles River

Laka/Panjab University

Swiss/Janvier

Swiss/Janvier

Swiss/Janvier

Swiss/Janvier

Swiss/Janvier

Swiss/Janvier

Swiss/Janvier

Swiss/Janvier

Swiss/Janvier

NMRI

DBA/2

CD-1/Charles River

C57 BL/6Rj

Strain

(continued)

(80)

(80)

(81)

(64)

(76)

(64)

(26)

(26)

(66)

(26)

(26)

(46)

(46)

(46)

(80)

(46)

References

Table 6.9 Selective serotonin reuptake inhibitors (SSRIs); 0 no effect in the FST, þ antidepressant-like effect, þþ important antidepressant-like effect in the FST

The Forced Swimming Test in Mice 107

171

177

177

Fluvoxamine

Fluvoxamine

Fluvoxamine

85

199.9

Fluvoxamine

235.2

44

Fluvoxamine

44

181

Fluvoxamine

79

72

129

Fluvoxamine

93

<50

79

95

81

81

Fluvoxamine

140

169

138

187.8

65

231.7

150

Fluoxetine

65

232

99

110

48

Fluoxetine

170

Fluoxetine

90

36

71

189

Fluoxetine

72

89

139

198

Fluoxetine

172

98

Fluoxetine

194

Fluoxetine

190

% Control

<50

194

Fluoxetine

Treated (s)

Fluoxetine

Control (s)

AD

Table 6.9 (continued)

20 32 32 32

þþ þþ þþ

20

þ 0

16

4

4

0

0

0

32

þþ

20

þ

16

30

þþ

þþ

32

þþ

40

32

þþ

0

16

8

Dose (mg/ kg)

0

0

Effect

YES

NO

YES

YES

NO

YES

YES

YES

YES

YES

YES

NO

NO

YES

YES

YES

YES

Actimetry

M

M

M

M

M

M

M

M

M

M

M

M

M

M

M

M

Gender

20–24

22–26

20–31

18–35

22–25

22–25

22–25

20–24

20–24

18–34

20–26

22–25

22–25

22–25

22–25

Weight (g)

Swiss/Janvier

Swiss/Janvier

Swiss/Janvier

Swiss

HaM/ICR (Broekman inst.)

CD-1/Charles River

CD-1/Charles River

CD-1/Charles River

Swiss/Janvier

Swiss/Janvier

Swiss

HaM/ICR (Broekman inst.)

CD-1/Charles River

CD-1/Charles River

CD-1/Charles River

CD-1/Charles River

CD-1/Charles River

Strain

(76)

(64)

(66)

(70)

(65)

(80)

(8)

(80)

(76)

(71)

(70)

(65)

(74)

(80)

(8)

(80)

(80)

References

108 Hascoe¨t and Bourin

168

Paroxetine

231.7

Sertraline

Italic: no effect.

158

Sertraline

170.6

57 74

36

26

185

Sertraline

49

72

224.7

Paroxetine

162.8

71

59

Paroxetine

131

8 8 16 16 8 8 32

þþ þ þþ þþ þþ þþ

222

51

þþ

Paroxetine

110

8

þþ

215

45

Paroxetine

100

224

Paroxetine

2

65

179

Paroxetine 0

16

þ

77

Paroxetine

116

8

þ

84

Paroxetine

8

þ 1–2

8

þþ

0

1

0

100

66

32

þþ 1–2

32

þþ

0

16

þþ

Paroxetine

111

47

164

Paroxetine

77

86

170

Paroxetine

146

100

45

Paroxetine

93

59

44

207

Indalpine

100

Indalpine

170

Indalpine

YES

YES

YES

YES

YES

YES

YES

YES

YES

YES

YES

YES

YES

YES

NO

YES

YES

YES

NO

M

M

M

M

M

M

M

M

M

M

M

M

M

M

M

M

M

M

M

20–24

22–25

22–25

20–24

22–26

18–22

41–51

35–45

41–51

20–24

20–24

20–24

22–25

22–25

22–25

20–24

22–26

20–30

22–25

Swiss/Janvier

CD-1/Charles River

CD-1/Charles River

Swiss/Janvier

Swiss/Janvier

Swiss/Janvier

Swiss/Janvier

Swiss/Janvier

Swiss/Janvier

Swiss/Janvier

NMRI

DBA/2

CD-1/Charles River

CD-1/Charles River

CD-1/Charles River

C57 BL/6Rj

Swiss/Janvier

Swiss/Janvier

CD-1/Charles River

(76)

(80)

(80)

(76)

(64)

(26)

(26)

(26)

(26)

(46)

(46)

(46)

(80)

(80)

(80)

(46)

(64)

(66)

(80)

The Forced Swimming Test in Mice 109

110

Hascoe¨t and Bourin

Lithium: In mice, lithium reportedly has few effects of its own in the FST, although it may potentiate the antidepressant-like effects of 5-HT-reuptake inhibitors (88, 89) and 5HT1B-receptor agonists (90). In rats, lithium has been reported to have no effect in the FST (91, 92, 93, 94) or opposite effects to those typically seen with antidepressant drugs, suggesting pro-depressant-like effects (95, 96, 97). It has been previously shown that, combined with lithium, sub-active doses of antidepressants (tricyclic antidepressants, 5-HT reuptake inhibitors, atypical antidepressants) produced significant anti-immobility effects in the mouse forced swimming test (88, 89). Carbamazepine and sodium valproate: Neither carbamazepine nor sodium valproate demonstrate antidepressant-like effect in the mice FST, when administered alone (90); these authors have also tested the hypothesis of the action of carmabazepine and valproate on some 5-HT receptor subtypes. The results of the study suggest that the action of carbamazepine and sodium valproate involved 5-HT1A receptors in the FST. However, considering the complexity of the actions of these compounds, it is possible that other neurotransmitter systems/receptors may be involved. Indeed, carbamazepine affects a multiplicity of neurotransmitter systems implicated in the pathophysiology of mood disorders. The primary neurotransmitters thus far shown to be altered by carbamazepine include serotonin (5-hydroxytryptamine, 5-HT), noradrenalin (NA), dopamine (DA) and gamma-aminobutyric acid (GABA) (98, 99, 100). Lamotrigine, topiramate and phenytoine: The most interesting results obtained with mood stabilizers were brought by the study of lamotrigine, an atypical antiepileptic (101). Lamotrigine, at doses of 8 and 16 mg/kg, i.p., decreased immobility time in the FST. In the same way, topiramate and phenytoine were shown to dramatically decrease immobility time in the FST following i.p. administration (unpublished data). Among all mood stabilizers evaluated in the mice FST, lamotrigine, topiramate and phenytoin seem to be the most efficient in decreasing immobility time, while other mood stabilizers are only able to potentiate the antidepressant-like activity. Lamotrigine is structurally unrelated to the existing antiepileptic drugs, and its anticonvulsant action has been suggested to be due to the inhibition of glutamate release by an effect on voltagesensitive sodium channels (102, 103). This fact has determined our interest to study the role of ion channels in the mechanism of action of drugs in the FST using veratrine, a Naþ channel activator that increases glutamate release (104). In mice, we have investigated the effect of the co-administration of veratrine, with lamotrigine, topiramate and phenytoine as compared with paroxetine, imipramine and desipramine.

The Forced Swimming Test in Mice

111

The decrease in immobility time elicited by lamotrigine, topiramate and phenytoine was reversed by veratrine suggesting that sodium channels underlie their action in the forced swimming test. In contrast, veratrine did not affect the usual antidepressants activity (105, unpublished data). The FST seems able to evaluate the antidepressant potential of mood stabilizers. The main idea is that the mechanism underlying the anti-immobility effect of mood stabilizers is related to sodium channel and that FST is sensible to sodium channel mechanism and in a way to glutamatergic mechanism. In addition, lamotrigine, topiramate and phenytoine have shown more important amplitude of change in immobility time than classical antidepressants in the test.

3. FST and Dopamine It has been proven that FST provoked a significant increase in the concentration of DA in the mice brain, which confirms the interest of this test to investigate the relationship between the DA concentration in brain and antidepressant effect (106). Antidepressants were chosen in function of their affinity to each neurotransmitter system and their magnitude of antidepressant-like effect in the FST. A trend for an inverse correlation between DA level and antidepressant-like effect was found particularly for paroxetine (high magnitude of antidepressant-like effect and low DA concentration) and tranylcypromine (small magnitude of antidepressantlike effect and high concentration of DA). This result is confirmed by a recent study (46) in which influence of mice strain on antidepressant-like results was examined. In this study, results showed that C57 and DBA/2 mice strains did not respond to antidepressant administration in the FST and that DA whole brain concentration of these two strains are high in comparison with Swiss mice strain. Likewise, aged Swiss mice that have high DA whole brain concentrations did not respond to selective serotonin reuptake inhibitor treatment in the FST (27). However, this correlation is limited since other elements than the concentration of DA should be considered and in particular the intrinsic antidepressant-like activity of drugs. The antidepressant activity of bupropion did not seem to be linked with DA concentration in the FST. Bupropion is commonly presented as a selective inhibitor of DA uptake by blocking the DA transporter (107). However, this pharmacological mechanism remains questionable since the occupation of DA transporter sites by bupropion is low during clinical treatment (108) and no

112

Hascoe¨t and Bourin

modification of DA neuron firing by bupropion was put into evidence in the brain of rat (109) suggesting that another mechanism could take part in the antidepressant effect of bupropion. The measurement of the DA tissue concentration is another element which could explain this limited correlation. Indeed, this measurement is adapted to reflect a modification of DA concentration for drugs acting on the neuronal stock concentration but not for drugs acting on the effective concentration of DA in the neuronal synapse. Thus, this method is adapted for tranylcypromine that acts by inhibition of monoamine oxidase, which affects neuronal stock concentration but not for drugs acting only on the DA concentration in the synapse. A high concentration of DA in the whole brain could be a limiting factor for the antidepressant-like effect of some antidepressants (tranylcypromine, paroxetine) and seems to play a minor role in the antidepressant-like activity of other antidepressants (bupropion) in the FST.

4. General Conclusion FST was designed by Porsolt as a primary screening test for antidepressants. It is still one of the best models for this purpose; it is a low-cost, fast and reliable model to test potential antidepressant treatments with a strong predictive validity. However, the low face and construct validities should not forbid the use of this model for neurophysiological studies. It has a great sensitivity with all the antidepressant classes, as well as mood stabilizers, and all the mechanisms of action of treatments could be determined, but clinical correlations should be considered very carefully. Studying the method of action of an antidepressant is different from studying aetiology and how to cure depression. For this reason, some authors decided to abandon the term ‘‘model’’ of depression. They prefer the word ‘‘test’’, which corresponds to an examination of a critically key aspect of the response either to stress or to antidepressant drug action. It could help to reconsider their true role in the process of discovery of novel antidepressants (56). Care must be taken on the strain used for the test and all the experimental parameters involved. For a screening test, CD-1 can be a good strain to use to find out if a treatment has an antidepressant-like activity. Inbred strains have been found very defective in the FST with antidepressants. Only one type of antidepressant (DRI), bupropion, was significantly effective in the FST with C57BL/6Rj. For C57BL6j and DBA/2, no positive result coupled with a locomotor test was found in papers we analyzed. Outbred strains of mice are more responsive to antidepressants in

The Forced Swimming Test in Mice

113

the FST than inbred strains. These four outbred strains may be used for at least three classes of treatments. The most frequently used strains, CD1, NMRI and Swiss, have positive results with most of the antidepressants in the FST. HaM/ICR seems to be very responsive to drugs in the FST, but it is a rare strain. Only one paper was found to use this strain in the FST (65). There are many differences between strains; DBA/2, for example, does not have an appropriate response to the FST. This strain should not be used for behavioural studies with the FST. Despite their intrinsic limitations, the full potential of animal models of depression has not yet been realized and they represent an underexplored opportunity for drug development. Such opportunities arise from the molecular dissection of the biological features of the models (110). FST is also sensible to ionic canal manipulation, and more precisely sodium channel manipulation suggesting that not only neurotransmitter but also more complex mechanism can be studied in the mice FST.

References 1. Porsolt RD, Bertin A, Jalfre M. Behavioral despair in mice: a primary screening test for antidepressants. Arch Int Pharmacodyn Ther 1977;229:327–36. 2. Porsolt RD, Bertin A, Blavet N, Deniel M, Jalfre M. Immobility induced by forced swimming in rats: effects of agents which modify central catecholamine and serotonin activity. Eur J Pharmacol 1979;57:201–10. 3. Petit-Demoulie`re B, Chenu F, Bourin M. Forced swimming test in mice: a review of antidepressant activity. Psychopharmacology (Berl), 2005;177:245–55. 4. Borsini F, Meli A. Is the forced swimming test a suitable model for revealing antidepressant activity? Psychopharmacology (Berl) 1988;94:147–60. 5. Porsolt RD, Bertin A, Jalfre M. ‘‘Behavioural despair’’ in rats and mice: strain differences and the effects of imipramine. Eur J Pharmacol 1978;51:291–94. 6. West AP. Neurobehavioral studies of forced swimming: the role of learning and memory in the forced swim test. Prog Neuropsychopharmacol Biol Psychiatry 1990;14:863–77. 7. Bourin M. Is it possible to predict the activity of a new antidepressant in animals with simple psychopharmacological tests? Fundam Clin Pharmacol 1990;4:49–64.

8. Porsolt RD. Animal models of depression: utility for transgenic research. Rev Neurosci 2000;11:53–8. 9. Lucki I, Dalvi A, Mayorga AJ. Sensitivity to the effects of pharmacologically selective antidepressants in different strains of mice. Psychopharmacology (Berl) 2001;155:315–22. 10. Nestler EJ, Gould E, Manji H, Buncan M, Duman RS, Greshenfeld HK, Hen R, Koester S, Lederhendler I, Meaney M, Robbins T, Winsky L, Zalcman S. Preclinical models: status of basic research in depression. Biol Psychiatry 2002;52:503–28. 11. Eisch AJ, Bolanos CA, de Wit J, Simonak RD, Pudiak CM, Barrot M, Verhaagen J, Nestler EJ. Brain-derived neurotrophic factor in the ventral midbrain-nucleus accumbens pathway: a role in depression. Biol Psychiatry 2003;54:994–1005. 12. Siuciak JA, Lewis DR, Wiegand SJ, Lindsay RM. Antidepressant-like effect of brainderived neurotrophic factor (BDNF). Pharmacol Biochem Behav 1997;56:131–7. 13. Baez M, Volosin M. Corticosterone influences forced swim-induced immobility. Pharmacol Biochem Behav 1994;49:729–36. 14. GuoW,ToddK,BourinM,HascoetM,Kouadio F.Additiveeffectsofglyburideandantidepressants in the forced swimming test: evidence for the involvement of potassium channel blockade. Pharmacol Biochem Behav 1996;54:725–30.

114

Hascoe¨t and Bourin

15. Redrobe JP, Pinot P, Bourin M. The effect of the potassium channel activator, cromakalim, on antidepressant drugs in the forced swimming test in mice. Fundam Clin Pharmacol 1996;10:524–8. 16. Slattery DA, Hudson AL, Nutt DJ. Invited review: the evolution of antidepressant mechanisms. Fundam Clin Pharmacol 2004;18:1–21. 17. Harkin AJ, Bruce KH, Craft B, Paul IA. Nitric oxide synthase inhibitors have antidepressant-like properties in mice. 1. Acute treatments are active in the forced swim test. Eur J Pharmacol 1999;372:207–13. 18. Khisti RT, Chopde CT, Jain SP. Antidepressant-like effect of the neurosteroid 3alphahydroxy-5alpha-pregnan-20-one in mice forced swim test. Pharmacol Biochem Behav 2000;67:137–43. 19. Gardier AM, Trillat AC, Malagie´ I, David D, Hascoe¨t M, Colombel MC, Jolliet P, Jacquot C, Hen R, Bourin M. Re´cepteurs 5HT1B de la se´rotonine et effets antide´presseurs des inhibiteurs de recapture se´lectifs de la se´rotonine. C.R.Acad.Sci.Paris Life 2001;324:433–41. 20. Holmes A, Yang RJ, Murphy DL, Crawley JN. Evaluation of antidepressant-related behavioral responses in mice lacking the serotonin transporter. Neuropsychopharmacology 2002;27:914–23. 21. Cryan JF, Dalvi A, Jin SH, Hirsch BR, Lucki I, Thomas SA. Use of dopamine-betahydroxylase-deficient mice to determine the role of norepinephrine in the mechanism of action of antidepressant drugs. J Pharmacol Exp Ther 2001;298:651–57. 22. MacQueen GM, Ramakrishnan K, Croll SD, Siuciak JA, Yu G, Young LT, Fahnestock M. Performance of heterozygous brain-derived neurotrophic factor knockout mice on behavioral analogues of anxiety, nociception, and depression. Behav Neurosci 2001;115:1145–53. 23. Monteggia LM, Barrot M, Powell CM, Berton O, Galanis V, Gemelli T, Meuth S, Nagy A, Greene RW, Nestler EJ. Essential role of brain-derived neurotrophic factor in adult hippocampal function. Proc Natl Acad Sci USA 2004; 20:10827–32. 24. Bourin M, Fiocco AJ, Clenet F. How valuable are animal models in defining antidepressant activity? Hum Psychopharmacol 2001;16:9–21. 25. Dulawa SC, Holick KA, Gundersen B, Hen R. Effects of chronic fluoxetine in animal

26.

27.

28.

29.

30.

31.

32.

33.

34.

35.

36.

37.

models of anxiety and depression. Neuropsychopharmacology 2004;29:1321–30. Bourin M, Colombel MC, Redrobe JP, Nizard J, Hascoe¨t M, Baker GB. Evaluation of efficacies of different classes of antidepressants in the forced swimming test in mice at different ages. Prog Neuropsychopharmacol Biol Psychiatry 1998;22:343–51. David DJ, Bourin M, Hascoe¨t M, Colombel MC, Baker GB, Jolliet P. Comparison of antidepressant activity in 4- and 40-weekold male mice in the forced swimming test: involvement of 5-HT1A and 5-HT1B receptors in old mice. Psychopharmacology (Berl) 2001;153:443–49. Browne RG. Effects of antidepressants and anticholinergics in a mouse ‘‘behavioral despair’’ test. Eur J Pharmacol 1979;58:331–34. Denenberg VH, Talgo NW, Waters NS, Kenner GH. A computer-aided procedure for measuring swim rotation. Physiol Behav 1990;47:1023–5. Dubocovich ML, Mogilnicka E, Areso PM. Antidepressant-like activity of the melatonin receptor antagonist, luzindole (N-0774), in the mouse behavioral despair test. Eur J Pharmacol 1990;182:313–25. Sunal R, Gumusel B, Kayaalp SO. Effect of changes in swimming area on results of ‘‘behavioral despair test’’. Pharmacol Biochem Behav 1995;49:891–6. Aley KO, Kulkarni SK. GABA-mediated modification of despair behavior in mice. Naunyn Schmiedebergs Arch Pharmacol 1989;339: 306–11. Crabbe JC, Wahlsten D, Dudek BC. Genetics of mouse behavior: interactions with laboratory environment. Science 1999;284:1670–2. Cabib S, Orsini C, Le Moal M, Piazza PV. Abolition and reversal of strain differences in behavioral responses to drugs of abuse after a brief experience. Science 2000;289:463–5. Cabib S, Puglisi-Allegra S, Ventura R. The contribution of comparative studies in inbred strains of mice to the understanding of the hyperactive phenotype. Behav Brain Res 2002;130:103–9. Alonso SJ, Castellano MA, Afonso D, Rodriguez M. Sex differences in behavioral despair: relationships between behavioral despair and open field activity. Physiol Behav 1991;49:69–72. David DJ, Nic Dhonnchadha BA, Jolliet P, Hascoe¨t M, Bourin M. Are there gender differences in the temperature profile of mice after acute antidepressant administration and exposure to two animal

The Forced Swimming Test in Mice

38.

39.

40.

41.

42.

43.

44.

45.

46.

47.

48.

49.

models of depression? Behav Brain Res 2001;119: 203–11. Voikar V, Koks S, Vasar E, Rauvala H. Strain and gender differences in the behavior of mouse lines commonly used in transgenic studies. Physiol Behav 2001;72:271–81. Karolewicz B, Paul IA. Group housing of mice increases immobility and antidepressant sensitivity in the forced swim and tail suspension tests. Eur J Pharmacol 2001;415:197–201. Hilakivi LA, Ota M, Lister RG. Effect of isolation on brain monoamines and the behavior of mice in tests of exploration, locomotion, anxiety and behavioral ’despair’. Pharmacol Biochem Behav 1989;33:371–4. Yates G, Panksepp J, Ikemoto S, Nelson E, Conner R. Social isolation effects on the ‘‘behavioral despair’’ forced swimming test: effect of age and duration of testing. Physiol Behav 1991;49:347–53. Lucki I. The forced swimming test as a model for core and component behavioral effects of antidepressant drugs. Behav Pharmacol 1997;8:523–32. Schramm NL, McDonald MP, Limbird LE. The alpha(2a)-adrenergic receptor plays a protective role in mouse behavioral models of depression and anxiety. J Neurosci 2001;21:4875–82. Krahe TE, Filgueiras CC, Schmidt SL. Effects of rotational side preferences on immobile behavior of normal mice in the forced swimming test. Prog Neuropsychopharmacol Biol Psychiatry 2002;26:169–76. Bai F, Li X, Clay M, Lindstrom T, Skolnick P. Intra- and interstrain differences in models of ‘‘behavioral despair’’. Pharmacol Biochem Behav 2001;70:187–92. David DJ, Renard CE, Jolliet P, Hascoe¨t M, Bourin M. Antidepressant-like effects in various mice strains in the forced swimming test. Psychopharmacology (Berl) 2003;166:373–82. Alcaro A, Cabib S, Ventura R, Puglisi-Allegra S. Genotype- and experience-dependent susceptibility to depressive-like responses in the forced-swimming test. Psychopharmacology (Berl) 2002;164:138–43. Arai I, Tsuyuki Y, Shiomoto H, Satoh M, Otomo S. Decreased body temperature dependent appearance of behavioral despair in the forced swimming test in mice. Pharmacol Res 2000;42:171–6. Taltavull JF, Chefer VI, Shippenberg TS, Kiyatkin EA. Severe brain hypothermia as a factor underlying behavioral immobility

50.

51.

52.

53.

54.

55.

56.

57.

58.

59.

60.

115

during cold-water forced swim. Brain Res 2003;975:244–7. Yoshikawa T, Watanabe A, Ishitsuka Y, Nakaya A, Nakatani N. Identification of multiple genetic loci linked to the propensity for ‘‘behavioral despair’’ in mice. Genome Res 2002;12:357–66. Detke MJ, Lucki I. Detection of serotonergic and noradrenergic antidepressants in the rat forced swimming test: the effects of water depth. Behav Brain Res 1996;73:43–6. Nomura S, Shimizu J, Kinjo M, Kametani H, Nakazawa T. A new behavioral test for antidepressant drugs. Eur J Pharmacol 1982;83:171–5. Raghavendra V, Kaur G, Kulkarni SK. Antidepressant action of melatonin in chronic forced swimming-induced behavioral despair in mice, role of peripheral benzodiazepine receptor modulation. Eur Neuropsychopharmacol 2000;10:473–81. Easton A, Arbuzova J, Turek FW. The circadian Clock mutation increases exploratory activity and escape-seeking behavior. Genes Brain Behav 2003;2:11–9. Holmes A. Mouse Behavioral Models of Anxiety and Depression. In: Crawley JN ed. (ed) Mouse Behavioral Phenotyping. Society for Neuroscience, Washington DC, 2003; 43–7. O’Neil MF, Moore NA. Animal models of depression: are there any? Hum Psychopharmacol 2003;18:239–54. Wahlsten D, Metten P, Phillips TJ, Boehm SL, 2nd, Burkhart-Kasch S, Dorow J, Doerksen S, Downing C, Fogarty J, RoddHenricks K, Hen R, McKinnon CS, Merrill CM, Nolte C, Schalomon M, Schlumbohm JP, Sibert JR, Wenger CD, Dudek BC, Crabbe JC. Different data from different labs: lessons from studies of geneenvironment interaction. J Neurobiol 2003;54:283–311. Andreatini R, Bacellar LF. Animal models: trait or state measure? The test-retest reliability of the elevated plus-maze and behavioral despair. Prog Neuropsychopharmacol Biol Psychiatry 2000;24:549–60. Hirani K, Khisti RT, Chopde CT. Behavioral action of ethanol in Porsolt’s forced swim test: modulation by 3 alpha-hydroxy-5 alpha-pregnan-20-one. Neuropharmacology 2002;43:1339–50. Steru L, Chermat R, Thierry B, Simon P. The automated Tail Suspension Test: a computerized device which differentiates

116

61.

62.

63.

64.

65.

66.

67.

68.

69.

70.

71.

72.

Hascoe¨t and Bourin psychotropic drugs. Neuropsychopharmacol Biol Psychiatry 1987;11:659–71. Ripoll N, David DJP, Dailly E, Hascoe¨ t M, Bourin M. Antidepressant-like effects in various mice strains in the tail suspension test. Behav Brain Res 2003;143:193–200. Mayorga AJ, Lucki I. Limitations on the use of the C57BL/6 mouse in the tail suspension test. Psychopharmacology 2001;155:110–2. Redrobe JP, Bourin M. Augmentation of antidepressant pharmacotherapy: a preclinical approach using the mouse forced swimming test. CNS Spectrums 1999;4:73–81. Bourin M, Colombel MC, Malinge M, Bradwejn J. Clonidine as a sensitizing agent in the forced swimming test for revealing antidepressant activity. J Psychiatry Neurosci 1991;16:199–203. De Graaf JS, Van Riezen H, Berendsen HHG, Van Delft AML. A set of behavioural tests predicting antidepressant activity. Current Trends Review 1985;5:291–301. Malinge M, Bourin M, Colombel MC, Larousse C. Additive effects of clonidine and antidepressant drugs in the mouse forcedswimming test. Psychopharmacology (Berl) 1988;96:104–9. Devoize JL, Rigal F, Eschalier A, Trolese JF, Renoux M. Influence of naloxone on antidepressant drug effects in the forced swimming test in mice. Psychopharmacology (Berl) 1984;84:71–5. Luttinger D, Freedman M, Hamel L, Ward SJ, Perrone M. The effects of serotonin antagonists in a behavioral despair procedure in mice. Eur J Pharmacol 1985;107:53–8. Scotto di Tella AM, Mercier J. [Influence of the procedure of administration in the activity of some antidepressant or disinhibiting drugs upon behavioural despair (author’s transl)]. J Pharmacol 1981;12:179–88. Szymczyk G, Zebrowska-Lupina I. Influence of antiepileptics on efficacy of antidepressant drugs in forced swimming test. Pol J Pharmacol 2000;52:337–44. Bourin M, Redrobe JP, Hascoe¨t M, Baker GB, Colombel MC. A schematic representation of the psychopharmacological profile of antidepressants. Prog Neuropsychopharmacol Biol Psychiatry 2000;20:1389–402. Biziere K, Kan JP, Souilhac J, Muyard JP, Roncucci R. Pharmacological evaluation of minaprine dihydrochloride, a new psychotropic drug. Arzneimittelforschung 1982;32:824–31.

73. Biziere K, Worms P, Kan JP, Mandel P, Garattini S, Roncucci R. Minaprine, a new drug with antidepressant properties. Drugs Exp Clin Res 1985;11:831–40. 74. Zocchi A, Varnier G, Arban R, Griffante C, Zanetti L, Bettelini L, Marchi M, Gerrard PA, Corsi M. Effects of antidepressant drugs and GR 205171, an neurokinin-1 (NK1) receptor antagonist, on the response in the forced swim test and on monoamine extracellular levels in the frontal cortex of the mouse. Neurosci Lett 2003;345:73–6. 75. Mogilnicka E, Czyrak A, Maj J. Dihydropyridine calcium channel antagonists reduce immobility in the mouse behavioral despair test; antidepressants facilitate nifedipine action. Eur J Pharmacol 1987;138:413–6. 76. Clenet F, De Vos A, Bourin M. Involvement of 5-HT(2C) receptors in the anti-immobi lity effects of antidepressants in the forced swimming test in mice. Eur Neuropsychopharmacol 2001;11:145–52. 77. Stenger A, Couzinier JP, Briley M. Psychopharmacology of midalcipran, 1-phenyl-1diethyl-amino-carbonyl-2-aminomethylcyclopropane hydrochloride (F 2207), a new potential antidepressant. Psychopharmacology (Berl) 1987;91:147–53. 78. Rogoz Z, Skuza G, Maj J. Pharmacological profile of milnacipran, a new antidepressant, given acutely. Pol J Pharmacol 1999;51:317–22. 79. Redrobe JP, Bourin M, Colombel MC, Baker GB. Dose-dependent noradrenergic and serotonergic properties of venlafaxine in animal models indicative of antidepressant activity. Psychopharmacology (Berl) 1998;138:1–8. 80. Da-Rocha MA, Jr., Puech AJ, Thie´bot MH. Influence of anxiolytic drugs on the effects of specific serotonin reuptake inhibitors in the forced swimming test in mice. J Psychopharmacol 1997;11:211–8. 81. Anjaneyulu M, Chopra K, Kaur I. Antidepressant activity of quercetin, a bioflavonoid, in streptozotocin-induced diabetic mice. J Med Food 2003;6:391–5. 82. Miura H, Naoi M, Nakahara D, Ohta T, Nagatsu T. Effects of moclobemide on forced-swimming stress and brain monoamine levels in mice. Pharmacol Biochem Behav 1996;53:469–75. 83. Kato M, Katayama T, Iwata H, Yamamura M, Matsuoka Y, Narita H. In vivo characterization of T-794, a novel reversible inhibitor of monoamine oxidase-A, as an antidepressant with a wide safety margin. J Pharmacol Exp Ther 1998;284:983–90.

The Forced Swimming Test in Mice 84. Eschalier A, Rigal F, Devoize JL, Trolese JF, Grillon C. Morphine pretreatment reduces clomipramine effect in mouse forced-swimming test. Eur J Pharmacol 1983;91:505–7. 85. Devoize JL, Rigal F, Eschalier A, Trolese JF. Naloxone inhibits clomipramine in mouse forced swimming test. Eur J Pharmacol 1982;78:229–31. 86. Schechter MD, Chance WT. Non-specificity of ‘‘behavioral despair’’ as an animal model of depression. Eur J Pharmacol 1979;60:139–42. 87. Machado-Viera R, Kapczinski F, Soares JC. Perspective for the development of animal models of bipolar disorders. Prog Neuropsychopharmacol Biol Psychiatry 2004;28:209–24. 88. Nixon M, Bourin M., K., Hascoe¨t M, Colombel MC. Additive effects of the lithium and antidepressants in the forced swimming test: further evidence for involvement of the serotonergic system. Psychopharmacology. 1994;115:59–64. 89. Bourin M, Hascoe¨t M, Colombel MC, Redrobe JP, Baker GB. Differential effects of clonidine, lithium and quinine in the forced swimming test in mice for antidepressants: possible roles of serotonergic system. Eur Neuropsychopharmacol 1996; 6: 231–6. 90. Redrobe JP, Bourin M. Evidence of the activity of lithium on the 5-HT1B receptors in the mouse forced swimming test: comparison with carbamazepine and sodium valproate. Psychopharmacology 1999;141:370–7. 91. Hata T, Itoh E, Nishikawa H. Behavioral characteristics of SART-stressed mice in the forced swim test and drug action. Pharmacol Biochem Behav 1995;51:849–53. 92. Overstreet DH, Pucilowski O, Rezvani AH, Janowsky DS. Administration of antidepressants, diazepam and psychomotor stimulants further confirms the utility of Flinders Sensitive Line rats as an animal model of depression. Psychopharmacology (Berl). 1995;121:27–37. 93. Kitamura Y, Araki H, Gomita Y. Influence of ACTH on the effects of imipramine, desipramine and lithium on duration of immobility of rats in the forced swim test. Pharmacol Biochem Behav 2002;71:63–9. 94. Wegener G, Bandpey Z, Heiberg IL, Mørk A, Rosenberg R. Increased extracellular serotonin level in rat hippocampus induced by chronic citalopram is augmented by subchronic lithium: neurochemical and behavioural studies

95.

96.

97.

98.

99.

100.

101.

102.

103.

104.

105.

106.

117

in the rat. Psychopharmacology (Berl) 2003;166:188–94. Mague SM, Pliakas AM, Todtenkopf MS, Tomasiewicz HC, Zhang Y, Stevens WC, Jones RM, Portoghese PS, Carlezon Jr. WA. Antidepressant-like effects of -opioid receptor antagonists in the forced swim test in rats. J Pharmacol ExpTher 2003;305:1–8. Carlezon Jr WA, Beguin C, DiNieri J, Baumann MH, Richards M, Todtenkopf MS, Rothman RB, Ma Z, Lee DY-L, Cohen BM. Depressive-like effects of the -opioid receptor agonist Salvinorin A on behavior and neurochemistry in rats. J Pharmacol Exp Ther 2006;314:440–7. O’Donnell KC, Gould TD. The behavioral actions of lithium in rodent models: leads to develop novel therapeutics. Neurosci Biobehav Rev 2007;31:932–62. Mac Donald R L, Rogers C J, Twyman R E. Barbiturate regulation of kinetic properties of the GABAA receptor channel of mouse spinal neurons in culture. J Physiol 1989;417:483–500. Post RM, Uhde TW, Roy-Byrne PP, Joffe RT. Antidepressant effects of carbamazepine. Am J Psychiatry 1986;143:29–34. Post RM, Weiss SR, Chuang DM. Mechanisms of action of anticonvulsants in affective disorders: comparisons with lithium. J Clin Psychopharmacol. 1992;12:23S–35S. Bourin M, Masse F, Hascoe¨ t M. Evidence for the activity of lamotrigine at 5-HT1A receptors in the mouse forced swimming test. J PsychiatryNeurosci 2005;30:275–82. Leach MJ, Marden CM, Miller AA. Pharmacological studies of lamotrigine, a novel potential antipsychotic drug: neurochemical and clinical studies of the mechanism of action. Epilepsia 1986;27:490–7. Kuo CC, Lu L. Characterization of lamotrigine inhibition of Naþ channels in rat hippocampal neurons. Br J Pharmacol 1997;121:1231–38. Lizasoain I, Knowles RG, Moncada S. Inhibition by lamotrigine of the generation of nitric oxide in rat forebrain slices. J Neurochem 1995;64:636–42. Prica C, Hascoe¨t M, Bourin M. Antidepressant-like effect of lamotrigine is reversed by veratrine: a possible role of sodium channels in bipolar depression. Behav Brain Res. 2008;191:49–54. Renard CE, Dailly E, David DJ, Hascoe¨t M, Bourin M. Neurochemical changes

118

Hascoe¨t and Bourin

following the mouse forced swimming test but not the tail suspension test. Fundam Clin Pharmacol 2003;17:449–55. 107. Cooper BR, Hester TJ, Maxwell RA. Behavioral and biochemical effects of the antidepressant bupropion (Wellbutrin): evidence for selective blockade of dopamine uptake in vivo. J Pharmacol Exp Ther 1980;215:127–34. 108. Meyer JH, Goulding VS, Wilson AA, Hussey D, Christensen B.K, Houle S. Bupropion occupancy of the dopamine transporter is

low during clinical treatment. Psychopharmacology (Berl) 2002;163:102–05. 109. Dong J and Blier P. Modification of norepinephrine and serotonin, but not dopamine, neuron firing by sustained bupropion treatment. Psychopharmacology (Berl) 2001;155:52–7. 110. Wong ML, Licinio J. From monoamines to genomic targets: a paradigm shift for drug discovery in depression. Nat Rev Drug Discov 2002;3:136–51.

Chapter 7 The Tail-Suspension Test: A Model for Characterizing Antidepressant Activity in Mice Olivia F. O’Leary and John F. Cryan Abstract The tail-suspension test (TST) is a widely used assay for screening potential antidepressant drugs. Its advantages include being a rapid, inexpensive, highly predictive and high-throughput screening test for the acute behavioral effects of antidepressants. The test is based on the principle that mice subjected to the short-term, inescapable stress of being suspended by their tail, will develop an immobile posture. A variety of antidepressant treatments irrespective of their primary mechanism of action decrease the time mice spend immobile by promoting escape-oriented behaviors. In recent years, the TST has been used for the behavioral characterization of genetically modified mice in an effort to identify novel targets of antidepressant activity for which pharmacological tools may not yet exist. Additionally, such mice are also tested in the TST for in vivo evaluation of the validity of current molecular theories of depression and antidepressant action. In this chapter, we provide a historical overview of the TST and discuss the utility and the validity of the TST as an animal model of antidepressant activity. This is followed by a detailed practical protocol on how to conduct a TST experiment in the laboratory as well as some troubleshooting tips. We describe important experimental variables that can interfere with the outcome of a TST experiment, and discuss how some of these factors have been manipulated in an effort to identify neural substrates that contribute to both baseline and antidepressant-induced changes in immobility. Finally, we also provide some examples of how such findings have been extrapolated to clinical depression. We conclude that the TST has great utility as a simple model to rapidly evaluate antidepressant-induced behavior and under some circumstances may also prove useful in determining genetic contributions to depression. Key words: Tail-suspension test, antidepressant, immobility, strain, gender, high-throughput, automation.

1. Background and Historical Overview The tail-suspension test (TST) is a rapid, inexpensive, highly predictive and high-throughput screening test for the acute behavioral effects of antidepressant drugs.(1–3) Although the TST has been used in different rodent species including gerbils and T.D. Gould (ed.), Mood and Anxiety Related Phenotypes in Mice, Neuromethods 42, DOI 10.1007/978-1-60761-303-9_7, ª Humana Press, a part of Springer ScienceþBusiness Media, LLC 2009

119

120

O’Leary and Cryan

rats (4, 5), its use has been largely restricted to mice (6). The TST is based on principles similar to those of the forced swimming test (FST), whereby mice or rats forced to swim in an inescapable water-filled cylinder initially engage in escape-oriented behaviors that are rapidly followed by increased episodes of immobile postures which are reduced or delayed by prior treatment with antidepressant drugs.(7, 8) In the TST, the inescapable stressor involves suspending the mouse by the tail from an elevated bar for several minutes (1). The mouse initially engages in escapeoriented behaviors such as running movements, body jerks and body torsions with attempts to catch the suspending apparatus, followed by increasing bouts of immobility (1, 6). When an antidepressant drug is administered prior to the test, mice will actively persist in escape-oriented behaviors, and thus the time the animal spends immobile is reduced when compared with vehicle treatment (1, 6, 9, 10) The time the mouse spends immobile is recorded either manually or with an automated device (1, 2, 6). The TST demonstrates strong predictive validity as it is sensitive to a variety of antidepressant drugs that differ in their primary neurobiological targets (see Ref. (6) for review) including tricyclic antidepressants (1, 2, 6, 9–11), selective serotonin reuptake inhibitors (SSRI) (6, 9–12), selective norepinephrine reuptake inhibitors (9, 11), monoamine oxidase inhibitors (2, 6, 11), atypical antidepressants such as bupropion (2, 11) as well as electroconvulsive shock (13). Moreover, the TST has been used to study genetic manipulations in mice that are relevant to depression and antidepressant action (6, 11, 14, 15). Such genetically modified mice may aid in the identification of novel targets for antidepressant activity for which pharmacological tools may not yet exist (16). Additionally, such mice can be used for in vivo evaluation of the validity of current molecular theories of depression and antidepressant action. Several theories of what FST- and TST-induced immobility represents have been put forward (see (6, 16)). The duration of immobility in the FST was originally described as an index of ‘‘behavioral despair’’ (7, 17), whereby the animal has given up hope of escaping and therefore no longer persists in escapeoriented behaviors. Others have hypothesized that immobility may be an evolutionary preserved measure of coping or adaptation (18), whereby immobility disengages the animal from active forms of coping with a stressful situation (19). This strategy has been suggested to be analogous to the psychological concept of entrapment described in clinical depression (19–21) and also parallels the clinical observation that depressed patients often exhibit psychomotor retardation (22). Immobility in the TST is unlikely to be a consequence of general hypoactivity because antidepressants reduce immobility at doses that can also induce sedation (6, 10). It is most likely that antidepressant-induced reductions in

Tail-Suspension Test

121

immobility are mediated by chemical changes in the brain because administration of a modified form of the antidepressant, imipramine, which has a reduced capacity to cross the blood brain barrier, does not affect the duration of immobility in the TST (1, 2). Finally, several studies suggest that the duration of immobility in the mouse TST is under genetic control (14, 23, 24). 1.1. Validity of the TST: A Model of Depression or a Model of Antidepressant Activity?

To be considered a valid model of depression, several investigators have proposed that the animal model in question should fulfil a variety of criteria. Those proposed by McKinney and Bunney (25) nearly 40 years ago remain the most cited. They propose that the validity of an animal model can be determined by the extent to which (i) it is ‘‘reasonably analogous’’ to the human disorder in its manifestations, (ii) it induces a behavioral change that can be objectively monitored, (iii) the behavioral changes observed should be reversed by treatments that are therapeutically effective in humans, and (iv) the model should be reproducible between investigators. However, there are many opinions on what criteria an animal model should fulfil (26–31). Perhaps one of the most practical of these propositions was put forward by Geyer and Markou, who propose that the only criteria that are necessary and sufficient for initial use of an animal model are that the paradigm has strong predictive validity and that the behavioral readout be reliable and robust within and between laboratories (30). The authors suggest that while the satisfaction of other criteria such as construct or discriminant validity is highly desirable, it is not essential for the initial use of a model in both basic neurobiological research and drug discovery. The following sections outline examples of how the TST fulfils some of the criteria proposed by McKinney and Bunney (25).

1.1.1. Is the TST ‘‘Reasonably Analogous’’ to Clinical Depression?

The TST was validated retrospectively based on the effects of clinically effective antidepressant treatments on immobility. For this reason, it is perhaps more appropriate to consider the TST as a behavioral test of antidepressant drug-like activity rather than model of depression. However, a number of factors known to be involved in susceptibility to major depression in humans induce a depression-like effect (increased immobility) in the TST. These manipulations include a genetic predisposition (14, 23, 32), stress (33), experimental diabetes (34, 35) and immunological activation (36, 37). In addition, withdrawal from psychostimulants, which is associated with depressive-like behavior in humans, can increase immobility in the TST (38). Several investigators have taken advantage of natural variations in TST-induced immobility scores across different mouse populations in order to elucidate the underlying genetic components of immobility in the TST. An elegant example of this is an ongoing project by Vaugeois et al. who have selectively bred two lines of

122

O’Leary and Cryan

mice for either high immobility (HelpLess (HL) mice; >115 s) or low immobility in the TST (NonLearnedHelpless (NLH) mice; < 35 s) (14, 23). From the original group of CD-1 mice tested in the TST, two pairs of high and low immobility mice were chosen and bred to produce the first generation (S1). Pairs of high (>155 s) and low immobility (<35 s) mice from each subsequent generation were then bred. At least 14 generations have since been produced, and baseline immobility scores and the percentage of animals reaching selection criteria for the helpless category have increased with each consecutive generation with almost all animals now fulfilling the selection criteria (14). The heritability of this trait of high immobility suggests that baseline performance in the TST is at least in part determined by genetic factors. Moreover, various behavioral, neurochemical and endocrine measures of depression and antidepressant action have been investigated in these mice (14, 23). In addition to increased immobility in the TST, later generations of helpless mice also exhibit increased immobility in the FST (14). Helpless mice also display anhedonia, lighter and more fragmented sleep patterns, elevated basal corticosterone levels and various dysfunctions of the serotonergic system (14, 39). Such alterations are similar to those often reported in depression (40, 41). Moreover, chronic treatment with the antidepressant, fluoxetine, reverses deficits in serotonergic cell firing in helpless mice and reduces immobility in helpless but not in NHL mice (14), a finding that closely parallels the absence of a moodaltering effect by antidepressant treatments in healthy humans. Taken together, such findings suggest that TST-induced immobility may have utility in modelling clinical depression. Baseline immobility and antidepressant-induced changes in immobility in the TST vary across mouse strains (see Section 5 for further discussion). Several investigators have taken advantage of these strain differences and used quantitative trait loci (QTL) technology to map genes underlying strain variations in baseline and antidepressant-induced changes in immobility in the TST. Such studies could aid in the identification of genes that contribute to depression or predict drug efficacy, respectively (6, 15, 24, 42–44). A quantitative trait is a specific phenotype that can vary in a measurable manner in the population. A QTL is a single gene in a multiple gene system underlying the specific phenotype being measured. A QTL analysis of immobility behavior in mice in both the FST and the TST has revealed both independent and overlapping QTLs for immobility in these tests (24). Composite interval mapping of 560 F2 mice from an intercross of the C57BL/6 mouse strain, which demonstrates high baseline immobility values, and the C3H/He mouse strain, which exhibits low baseline immobility values, revealed an overlapping QTL for TST and FST immobility in a region of chromosome 11 which contains multiple GABAA receptor subunit genes. Subsequent sequence

Tail-Suspension Test

123

and expression analyses of these genes from the two parental strains revealed a lower expression of the 1 subunit gene in the frontal cortex of C57BL/6 mice compared to the C3H/He mice (24). Remarkably, human genetic studies have reported evidence of linkage to mood disorders with markers from the chromosomal area 5q32-35, which is syntenic with the immobility-associated QTL identified on mouse chromosome 11 (24, 45, 46). Such congruent findings in preclinical and clinical studies further support the utility of the TST in dissecting the pathophysiology of depression. Several studies using intercrosses of other mouse strains have also reported loci associated with baseline immobility in the TST including loci on chromosomes 4 and 15 (from an intercross between C57BL/6 and DBA2/J) (47), loci on chromosomes 4, 5, 12 and 18 (from an intercross of the NMRI and 129S6 mouse strains) (15) and loci on chromosomes 4, 8, 11 and 14 (from an intercross of C57BL/6 and C3H/He strains) (24). The loci found on chromosome 4 from the latter two studies are in agreement with each other (15, 24) The absence of concurring results for the other reported loci is likely a consequence of the use of mouse populations which differ in their parental strain. Similar QTL studies have been conducted to identify loci underlying the acute behavioral response to antidepressants in the TST. Such studies have been conducted by crossing a mouse strain sensitive to a specific antidepressant in the TST with another strain that is relatively insensitive. Responsivity to the tricyclic antidepressant, imipramine, in the TST has been mapped to loci on chromosomes 1, 4 and 5, and the identified locus on chromosome 5 also influences baseline immobility (15). Another study has mapped the behavioral response to the SSRI, citalopram, in the TST to loci on chromosomes 7, 12 and 19 (43). In the latter study, three candidate genes residing on the chromosome 19 locus were suggested to be for the vesicular momoamine transporter 2 (VMAT2), alpha 2a adrenergic receptor and the beta1 adrenergic receptor (43). Subsequent sequencing identified two polymorphisms of VMAT2 (Leu117Pro and Ser505Pro) (43). In conclusion, the TST while best described as a model of antidepressant drug-like activity may also be a useful tool in the elucidation of the contribution of genetic factors to depression. 1.1.2. It Induces a Behavioral Change that can be Objectively Monitored

TST-induced immobility is relatively easy to measure and acute treatment with clinically effective antidepressant drugs consistently reduces immobility in this test (6). The mouse is considered immobile when there is an absence of initiated movement. In general, an immobile mouse will appear to hang passively and will be completely motionless (unless it has retained some momentum from the movement immediately prior to the current bout of immobility). Similarly to other behavioral tests, it is important that

124

O’Leary and Cryan

the rater be blind to experimental treatments so as to avoid biased scoring. This will likely require extra personnel for the experiment. However, scoring bias may be substantially reduced through the use of an automated TST apparatus. The TST is highly amenable to automation and several automated devices have been validated for use in the TST including the IMETATIC-TST apparatus from I.T.E.M-LABO, Paris (2, 3, 48), the TST apparatus from Med Associates Inc., St. Albans, VT (44, 49), the NS-TS/4 apparatus from Neuroscience, Tokyo, Japan (24) and a device from BIOSEB, Chaville, France (6). In many of these devices, the mouse is suspended by taping the tail to an aluminum bar which is connected to a strain gauge that detects movements of the mouse. Signals from the strain gauge are then transmitted to computer software which measures the continuous digital output converted from the strain gauge. Force of movement below a preset threshold is designated as immobility and is automatically calculated and accumulated for the duration of the test. Automation also enables the assessment of additional parameters such as power of movement and energy of movement (2, 3, 6). Although automation is a useful tool to improve scoring objectivity, it should be remembered that automated methods are not entirely free of subjective scoring because settings for automated measurement of mobility are selected by the experimenter. However, a recent study reported that this type of bias was limited because immobility scores from three individual raters demonstrated a high correlation with scores from an automated device that was set up by just one of these raters (49). Together, such studies suggest that the behavioral output of the TST can be measured in an objective manner. 1.1.3. The Behavioral Changes Observed Should be Reversed by Treatments that are Therapeutically Effective in Humans

The third criteria of validity proposed by McKinney and Bunney is that the behavioral changes observed should be reversed by treatments that are therapeutically effective in humans. In other words, is TST-induced immobility reduced by the same manipulations that alter behavioral symptoms of clinical depression? Of all the screens of antidepressant drug activity, the TST demonstrates one of the highest rates of predictive validity for known antidepressants compared to antipsychotics and anxiolytics (6). The TST is sensitive to a variety of antidepressant drugs that differ in their primary neurobiological targets such as tricyclic antidepressants (1, 2, 6, 9–11), selective serotonin reuptake inhibitors (6, 9–12), selective norepinephrine reuptake inhibitors (9, 11), monoamine oxidase inhibitors (2, 6, 11) and atypical antidepressants such as bupropion (2, 11). The TST has also been shown to be sensitive to non-pharmacological interventions such as electroconvulsive shock (13). Moreover, the TST has been commonly used to study genetic manipulations in mice that are relevant to depression and antidepressant action (6, 11, 14, 15). However, the utility of the TST as well as other models sensitive to monoamine-selective antidepressants is frequently

Tail-Suspension Test

125

criticized because they often fail to detect antidepressants with novel mechanisms of action, and this remains to be one of the main challenges in antidepressant drug discovery. For example, whilst the TST detects potential antidepressant-like effects of NK1 receptor antagonists (albeit in the gerbil; as this species has Neurokinin (NK) 1 receptor receptors more similar to humans than mice) (5), the effects of the CRF1 receptor antagonists are largely negative (36, 50, 51). The possible exception is the preliminary data suggesting antidepressant-like effect of the Corticotropin-releasing factor receptor subtype 1 (CRF1) receptor antagonist R121919 in the TST (51). Although both Corticotropin-releasing factor receptor subtype 1 (CRF1) and NK1 receptor antagonists have shown efficacy in promising small clinical trials (52, 53), negative findings in larger trials have also emerged (54, 55) and there are currently no antidepressants marketed with these mechanisms of action (6). One of the main criticisms of the TST is that it is used to reveal acute behavioral effects of antidepressant drugs, while chronic treatment with antidepressants is usually required for complete recovery in humans. Nevertheless, its ability to detect acute behavioral effects is particularly advantageous in industry, as it allows rapid in vivo screening of compounds, and high-throughput screening can be achieved through the use of automated devices (6, 49). Moreover, it has been reported that antidepressant drugs can induce behavioral changes early in the course of treatment and such changes may be relevant to a positive therapeutic outcome (56–58). It is also possible that the reductions in immobility in the TST induced by acute antidepressant treatments may represent some of the early emerging effects of antidepressants reported in the clinic (see ref. (6)). 1.1.4. The Model Should be Reproducible Between Investigators

Ideally, a reliable behavioral model will be reproducible among different laboratories. However, laboratories frequently make small modifications to behavioral tests in order to adapt them to the facilities of their own laboratory. Such modifications may include variations in the testing apparatus, protocol, time of day of behavioral testing as well as variations in environmental conditions (e.g., rodent housing and handling conditions). Such variations can lead to discrepancies in the results of the same experiment conducted by different laboratories or different experimenters. Furthermore, despite the best efforts to keep experimental conditions identical in different laboratories, inter-laboratory differences in the outcome of the experiment are still reported (59, 60). Therefore, the results of any behavioral experiment should be replicated by multiple laboratories before firm conclusions are drawn (59). In general, the behavioral effects of clinically effective antidepressant drugs in the TST are reproducible across laboratories because most studies report antidepressant-induced reductions in immobility in sensitive mouse strains (6). Moreover, a recent study

126

O’Leary and Cryan

reported little inter-rater variability within one laboratory for TST immobility scores (49), but direct comparisons of TST experiments across different laboratories have not yet been conducted. However, different strain survey studies report different rankings for mouse strain differences in immobility values (42, 44, 61). Such variations may be a result of inconsistent scoring techniques and the reliance on subjective ratings (49). Nevertheless, the TST is readily automated (2, 3, 6, 24, 44, 49) and automation offers the advantage of removing inter-rater subjective bias and assures consistent immobility ratings within a laboratory (49). Moreover, automated devices could be employed to standardize methods across different laboratories. In summary, the utility of the TST is supported by several of its features. First of all, immobility in the TST is increased by factors known to induce a depression-like effect in humans. Second, some of the genetic loci underlying variations in baseline immobility and the response to antidepressant drugs have also been associated with mood disorders and drug efficacy in humans. Third, TSTinduced immobility is relatively easy to measure and objective analysis can be facilitated through the use of automated devices. Fourth, the TST has one of the highest rates of predictive validity as it is responsive to many clinically effective antidepressant treatments irrespective of their primary neurobiological action. Finally, the procedural simplicity of the TST as well as its ease of automation has allowed it to be relatively reproducible across laboratories.

2. Equipment and Materials 1. Male mice, 8–12 weeks old. 2. A quiet room with standard room lighting for behavioral testing. 3. Horizontal bar raised 30 cm above bench top surface and a video camera OR an automated tail-suspension test apparatus (Med Associates Inc., St. Albans, VT (44, 49); NS-TS/4, Neuroscience, Tokyo, Japan (24); ID-TECH-BIOSEB, Chaville, France (6)). 4. Metric weigh scales. 5. 1-ml syringes. 6. 30-G needles. 7. Drug treatments: Test compound, reference compound/ positive control (imipramine–HCl; 30 mg/kg) and vehicle solution (negative control). Drug solutions should be freshly prepared on the day of the experiment. Imipramine makes an

Tail-Suspension Test

127

excellent reference compound because it has been repeatedly demonstrated to dose-dependently decrease immobility in the TST, with maximal responses 30 min after a single intraperitoneal (IP) injection of 30 mg/kg (2, 48, 62).

3. Setup and Procedure Animal Housing Conditions: Group-house mice according to the size of the cage. Provide free access to rodent diet and water. Maintain animals in a temperature-controlled environment (21C) under a 12-h light/dark cycle with illumination from 07:00 h to 19:00 h. If mice are delivered from an external source, then they should be allowed to acclimate to their new housing conditions for at least 1 week prior to behavioral testing. Experimental Design: A standard experiment for assessing a test compound should include the following treatment groups: (i) vehicle control, (ii) a reference compound/positive control (e.g., the antidepressant, imipramine, dissolved in sterile saline or sterile deionized water, 30 mg/kg, IP) and (iii) at least three doses of the test compound. Behavioral testing should be conducted during the light phase. For experiments split over a number of days, behavioral testing should be carried out during the same time period each day. Procedure: (i) Place mice in a quiet behavioral testing room and leave undisturbed for at least 1 h prior to commencing the experiment. (ii) Weigh the mice and administer freshly prepared drugs in a volume of 10 ml/kg. Different drug treatments should be given to different animals within the same cage to ensure an equal distribution of home-cage variation across the different treatment groups. Many biological factors (e.g., the stress hormone, corticosterone) and environmental factors (e.g., noise levels) change over the course of the day; therefore, drug treatments should be administered in a fixed rotation to ensure an equal distribution of different treatments over time. Placement of an identifying mark with a permanent ink pen on each animal’s tail prior to drug administration will allow drugtreated mice within the same cage to be distinguished from each other. (iii) Begin the behavioral test 30 min after intraperitoneal or subcutaneous drug administration OR 60 min following oral administration. Gently wrap adhesive tape around the tail approximately 2 cm from the end of the tail and place the

128

O’Leary and Cryan

tape on the horizontal bar ensuring that the tail is straight and the abdomen of the mouse is directly facing the video camera. Two animals can be tested simultaneously provided that they are placed on separate bars and shielded from the view of the other animal. Several automated systems are also available which allow the simultaneous testing of up to six to eight mice (e.g., Med Associates Inc., St. Albans, VT; ID-TECHBIOSEB, Chaville, France). (iv) The duration of the test is 6 min. (v) After each testing session, the apparatus should be cleaned to remove any potential olfactory cues. (vi) The time spent immobile during the test is calculated either by manually scoring video tapes or by the automated device. Escape-oriented behaviors can be classified into three types and are as follows: (i) running movements, forward or backwards, (ii) body torsions with attempts to catch the suspending apparatus, and (iii) body jerks (1). The mouse is considered immobile when there is an absence of initiated movement. In general, an immobile mouse will appear to hang passively and will be completely motionless (unless it has retained some momentum from the movement immediately prior to the current bout of immobility). Manual raters should be blind to treatment to promote objective rating. See Fig. 7.1 for a photograph of the test.

A

B

Fig. 7.1. The tail-suspension test. (A) Immobile posture; (B) active escape-oriented behavior.

4. Anticipated Results Animals treated with the reference antidepressant compound (e.g., imipramine) should spend significantly less time immobile when

Tail-Suspension Test

129

compared to the vehicle-treated group. If the test compound also alters the duration of immobility, the following points should be considered to aid in the identification of false positive and false negative findings in the TST: (i) the tail-suspension test relies on motor activity and, therefore, the effect of the test compound on general locomotor activity (e.g., home-cage activity) and motor co-ordination (e.g., the rotarod test) should also be examined; (ii) similarly to any other behavioral screen for antidepressant-like drug activity, drug-induced behavioral changes in the TST should be subsequently followed up using other established tests of antidepressant activity such as the forced swimming test (7, 17, 63), learned helplessness paradigm (64–66), chronic stress and anhedonia models (67) or olfactory bulbectomy (68), among others (16, 64).

5. Experimental Variables Environmental conditions: Mice are sensitive to changes in their environment. For this reason, it is important that a designated quiet room is utilized for behavioral testing and that noise levels in the vicinity of this room are kept to a minimum during the test. Any noise during behavioral testing could easily startle the animal and thus induce either movement or a freezing posture which could subsequently interfere with measurements of immobility. Noise levels in the surrounding area of the room may vary over time (e.g., a nearby refrigerator cooling) and should be minimized; therefore, it is also important that drug treatments are administered in a fixed rotation to ensure an equal distribution of different drug treatments over time. Finally, it is not advisable to conduct behavioral testing following a cage change as this stress may interfere with the outcome of the experiment. Gender of animals: Epidemiological studies have consistently reported that the incidence of major depression is higher in women than men (69). Although limited in number, parallel behavioral phenotyping studies in mice have also revealed sex differences in behavioral responses in the TST. Female mice from several mouse strains including the A/J, C57BL/6 J, NMRI and CD-1 have been reported to have higher baseline immobility values in the TST when compared to their male counterparts (44, 70), although other studies have also reported no difference in baseline immobility in the C57BL/6 J and Swiss mouse strains (71, 72). Gender effects may be even more apparent following other physiological manipulations. For example, female 5-HT1Breceptor-knockout mice behave as if they were administered

130

O’Leary and Cryan

antidepressants in the TST, whereas male mice show no such phenotype (73). The 5-HT1B receptor is a terminal presynaptic autoreceptor linked to the mechanism of action of antidepressants (40, 74). Several studies have demonstrated that ovarian hormones can directly contribute to baseline immobility values in the TST. Performance of female BALB/cByJ and C57BL/6 J mice in the TST varies across the estrous cycle with baseline immobility of BALB/cByJ mice being the highest during the estrous phase, while that of C57BL/6 J mice was the highest during the metestrous phase of the estrous cycle (75). Moreover, ovariectomy of female mice increased baseline immobility values in the TST and this effect was reversed by treatment with estradiol and progesterone (76). Taken together, it is clear that gender and ovarian hormones can influence immobility levels in the TST. Therefore, it is imperative that experiments analyze both male and female mice separately to ensure that gender contributions to behavior can be revealed. This is particularly important in the behavioral analysis of genetically modified mice where breeding constraints may require the use of both male and female animals in behavioral experiments that require large group numbers. For practical purposes, however, investigators tend to use male animals to avoid potential variability associated with the estrus cycle. Strain of animals: The C57BL/6 J mouse strain is one of the most widely used mouse strains in behavioral research and is the most common background strain onto which many genetically modified mice are bred (77, 78). However, the use of this strain in the tail-suspension test can be problematic due to a tendency for these animals to climb their tails during testing (78). Animals that climb their tails during the test should be eliminated from the experiment because they may have learned a potential escape route and thus may be more likely to repeat this behavior. There is a risk that such climbing behavior may go unnoticed when investigators use automated measuring devices that do not include video recording of the animal. However, automated video tracking systems are available that may allow the automatic detection of the disappearance of the animal from a predefined area of the computer screen and thus could automatically eliminate the climbing animal from the experiment (e.g., EthoVision) (79). Several strategies have been adopted to circumvent this problem of tail climbing including increasing the group size to generate more non-climbers (6), taping the base of the tail rather than the end of the tail to the strain gauge (42), taping the tail to a hook which is connected to a perpendicular wire and a strain gauge rather than taping directly to a bar (24), or attaching the tail to fabric rather than taping to a hard support (79).

Tail-Suspension Test

131

When designing a TST experiment, it is important to bear in mind that mouse strains differ in both their baseline and antidepressant-induced immobility values, although strain rankings do vary across different studies (42, 44, 48, 61, 80, 81). Generally, the C57BL/6J mouse strain demonstrates the highest level of baseline immobility when compared to other mouse strains (42, 44, 61). Liu and Gershenfeld (44) examined the baseline and imipramine-induced behaviors of 11 strains of mice in the TST including the 129S6/SvEvTac, A/J, AKR/J, Balb/cJ, C3H/ HeJ, C57BL/6J, DBA/2J, FVB/NJ, NMRI, SencarA/PtJ and SWR/J strains. Clear inter-strain differences were observed for baseline scores in the TST, and only three strains (DBA/2J, NMRI and FVB/NJ) responded to the antidepressant imipramine. A factor analysis of anxiety-like and stress-related behaviors in these strains and in an F2 cross of two strains that differed in both baseline immobility and response to imipramine in the TST (NMRI and 129S6) suggests that neither baseline immobility nor imipramine-induced changes in immobility in the TST is related to anxiety-like behavior or stress responsivity (82). More recently, Crowley et al. (42) examined baseline and citalopram-induced behaviors of eight inbred strains of mice in the TST including the C57BL/6J, DBA/2J, C3H/HeJ, BALB/cJ, A/J, 129/ SvEmsJ, 129/SvImJ and BTBR strains. In short, this strain survey reported that the DBA2, BALB/c and BTBR strains were the most sensitive, while the C57BL/6J and A/J strains were largely insensitive to the antidepressant effects of citalopram in the TST. Interestingly, the poor response of the C57BL/6J and A/J strains to citalopram (42) echoes their reportedly poor response to imipramine in the TST (44), while DBA/2J mice appear to be sensitive to both imipramine and citalopram in the TST (42, 44). Possible causes underlying the variation across strains in their behavioral responses to the TST include genetic factors (14, 23, 24, 43) and pharmacokinetic and pharmacodynamic differences (83, 84). Taken together, it is clear that mice exhibit variations in both baseline and antidepressant-induced immobility values across strains and this should be considered when designing TST experiments. At the outset, the mouse strain should be responsive to the reference compound or positive control. Secondly, baseline immobility values should be at a level so that increases or decreases in immobility can be readily detected (e.g., 120 s). Finally, as with other behavioral tests multi-strain comparisons may be needed to prevent false-negative screening of compounds in a given paradigm. Locomotor activity: The tail-suspension test relies on motor activity. Therefore, the effect of the test compound on general locomotor activity (e.g., home-cage activity) and motor

132

O’Leary and Cryan

co-ordination (e.g., the rotarod test) should also be examined to distinguish anti-immobility effects from stimulant effects and thus aid in the elimination of false-positive findings (1). Generally, antidepressant drugs reduce immobility in the tail-suspension test either at doses that have no effect on general locomotor activity or at doses that have sedative effects (6, 10).

6. Troubleshooting In the case of gene-knockout (and overexpressing) mice, the absence (or overexpression) of a specific protein during development may result in unexpected compensations by effects on other gene products. In addition, knockout mice may display a disproportionate loss of the gene of interest across different brain regions. Such effects rather than the immediate loss of the gene itself may also influence behavior. For these reasons, it is important that findings with knockout mice be validated with pharmacological tools as soon they become available. Pharmacological validation would determine whether alterations in behavior are a direct result of the loss of the gene itself or whether the gene of interest is simply part of a biological pathway that determines the behavioral response. In addition, it is useful to determine the pharmacokinetic profile of drugs in knockout mice to ensure that the behavioral effects are not a result of pharmacokinetic changes.

7. Concluding Remarks In conclusion, the TST has great utility as a model to evaluate antidepressant-induced behavior and under some circumstances may also prove useful in determining genetic contributions to depression. The TST has several advantages over other antidepressant screening tests and some of these are summarized in Table 7.1. Like any other behavioral model, the TST is not without its disadvantages (see Table 7.1), but many of these challenges can be overcome through careful experimental design. Finally, no animal model of antidepressant activity including the TST is without its flaws and therefore should be used in conjunction with other depression models such as the FST, learned helplessness, anhedonia models, stress models or olfactory bulbectomy.

Tail-Suspension Test

133

Table 7.1 Advantages and disadvantages of the TST Advantages

Disadvantages

Short in duration: Facilitates the rapid screening of acute effects of antidepressant drugs in drug discovery screening programs and allows rapid screening in in vivo mutagenesis screens and QTL analysis.

The rapid response to acute antidepressant treatment does not correlate with the long duration of treatment required to elicit clinical effects.

Procedurally simple.

Some behavioral responses are dependent upon the mouse strain employed.

Relatively inexpensive. Utility for studying genetically modified mice.

Dependent on motor activity; therefore, it must be ensured that the result is not due to changes in general locomotor function.

High predictive validity; sensitive to clinically effective antidepressants irrespective of their mechanism of action

Some agents reduce immobility in the FST but are without effect in the TST (85, 86).

Minimum stress to animal compared with other models. Easily Automated: Labor-saving; Objective scoring; Standardizes within lab and inter-lab scoring; Allows assessment of other parameters of such as power of movement and energy of movement; Facilitates high throughput screening.

Unusual behaviors may be missed if tests are not visually recorded by automated systems.

References 1. Steru L, Chermat R, Thierry B, Simon P. The tail suspension test: a new method for screening antidepressants in mice. Psychopharmacology (Berl) 1985;85(3):367–70. 2. Steru L, Chermat R, Thierry B, et al. The automated Tail Suspension Test: a computerized device which differentiates psychotropic drugs. Progress in Neuro-Psychopharmacology & Biological Psychiatry 1987;11(6):659–71. 3. Porsolt RD, Chermat R, Lenegre A, Avril I, Janvier S, Steru L. Use of the automated tail suspension test for the primary screening of psychotropic agents. Archives Internationales de Pharmacodynamie et de Therapie 1987;288(1):11–30.

4. Chermat R, Thierry B, Mico JA, Steru L, Simon P. Adaptation of the tail suspension test to the rat. Journal de Pharmacologie 1986;17(3):348–50. 5. Varty GB, Cohen-Williams ME, Hunter JC. The antidepressant-like effects of neurokinin NK1 receptor antagonists in a gerbil tail suspension test. Behavioural Pharmacology 2003;14(1):87–95. 6. Cryan JF, Mombereau C, Vassout A. The tail suspension test as a model for assessing antidepressant activity: review of pharmacological and genetic studies in mice. Neuroscience and Biobehavioral Reviews 2005;29(4–5):571–625.

134

O’Leary and Cryan

7. Porsolt RD, Bertin A, Jalfre M. Behavioral despair in mice: a primary screening test for antidepressants. Archives Internationales de Pharmacodynamie et de Therapie 1977;229(2):327–36. 8. Porsolt RD, Le Pichon M, Jalfre M. Depression: a new animal model sensitive to antidepressant treatments. Nature 1977;266(5604):730–2. 9. O’Leary OF, Bechtholt AJ, Crowley JJ, Hill TE, Page ME, Lucki I. Depletion of serotonin and catecholamines block the acute behavioral response to different classes of antidepressant drugs in the mouse tail suspension test. Psychopharmacology (Berl) 2007;192(3):357–71. 10. O’Leary OF, Bechtholt AJ, Crowley JJ, Valentino RJ, Lucki I. The role of noradrenergic tone in the dorsal raphe nucleus of the mouse in the acute behavioral effects of antidepressant drugs. European Neuropsychopharmacology 2007;17(3):215–26. 11. Cryan JF, O’Leary OF, Jin SH, et al. Norepinephrine-deficient mice lack responses to antidepressant drugs, including selective serotonin reuptake inhibitors. Proceedings of the National Academy of Sciences of the United States of America 2004;101(21):8186–91. 12. Perrault G, Morel E, Zivkovic B, Sanger DJ. Activity of litoxetine and other serotonin uptake inhibitors in the tail suspension test in mice. Pharmacology, biochemistry, and behavior 1992;42(1):45–7. 13. Teste JF, Martin I, Rinjard P. Electrotherapy in mice: dopaminergic and noradrenergic effects in the Tail Suspension Test. Fundamental & clinical pharmacology 1990;4(1):39–47. 14. El Yacoubi M, Bouali S, Popa D, et al. Behavioral, neurochemical, and electrophysiological characterization of a genetic mouse model of depression. Proceedings of the National Academy of Sciences of the United States of America 2003;100(10):6227–32. 15. Liu X, Stancliffe D, Lee S, Mathur S, Gershenfeld HK. Genetic dissection of the tail suspension test: a mouse model of stress vulnerability and antidepressant response. Biological psychiatry 2007;62(1):81–91. 16. Cryan JF, Mombereau C. In search of a depressed mouse: utility of models for studying depression-related behavior in genetically modified mice. Molecular Psychiatry 2004;9(4):326–57. 17. Porsolt RD, Anton G, Blavet N, Jalfre M. Behavioural despair in rats: a new model sensitive to antidepressant treatments. European Journal of Pharmacology 1978;47(4):379–91.

18. Thierry B, Steru L, Chermat R, Simon P. Searching-waiting strategy: a candidate for an evolutionary model of depression? Behavioral and Neural Biology 1984;41(2):180–9. 19. Lucki I. A prescription to resist proscriptions for murine models of depression. Psychopharmacology (Berl) 2001;153(3):395–8. 20. Dixon AK. Ethological strategies for defence in animals and humans: their role in some psychiatric disorders. The British Journal of Medical Psychology 1998;71 (Pt 4):417–45. 21. Gilbert P, Allan S. The role of defeat and entrapment (arrested flight) in depression: an exploration of an evolutionary view. Psychological Medicine 1998;28(3):585–98. 22. Weingartner H, Silberman E. Models of cognitive impairment: cognitive changes in depression. Psychopharmacology Bulletin 1982;18(2):27–42. 23. Vaugeois JM, Odievre C, Loisel L, Costentin J. A genetic mouse model of helplessness sensitive to imipramine. European Journal of Pharmacology 1996; 316(2–3):R1–2. 24. Yoshikawa T, Watanabe A, Ishitsuka Y, Nakaya A, Nakatani N. Identification of multiple genetic loci linked to the propensity for ‘‘behavioral despair’’ in mice. Genome research 2002;12(3):357–66. 25. McKinney WT, Jr., Bunney WE, Jr. Animal model of depression. I. Review of evidence: implications for research. Archives of General Psychiatry 1969;21(2):240–8. 26. Willner P, Mitchell PJ. The validity of animal models of predisposition to depression. Behavioural Pharmacology 2002;13(3):169–88. 27. Geyer M, Markou A. The role of preclinical models in the development of psychotropic drugs. In: Bloom F, Kupfer D, eds. Psychopharmacology: The Fifth Generation of Progress New York; Raven Press. 2000; 445–55. 28. McKinney WT. Overview of the past contributions of animal models and their changing place in psychiatry. Seminars in Clinical Neuropsychiatry 2001;6(1):68–78. 29. Willner P. Animal models of depression: an overview. Pharmacology & Therapeutics 1990;45(3):425–55. 30. Geyer M, Markou A. Animal models of psychiatric disorders. In: Bloom F, Kupfer D, eds. Psychopharmacology: The Fourth Generation of Progress Raven Press. 1995;787–98. 31. Sarter M, Bruno J. Animal models in biological psychiatry. In: D’haenen H, Boer Jd,

Tail-Suspension Test

32.

33.

34.

35.

36.

37.

38.

39.

40.

41.

42.

Westenberg H, P PW, eds. Textbook of Biological Psychiatry: John Wiley & Sons. 2002:37–44. Vaugeois JM, Passera G, Zuccaro F, Costentin J. Individual differences in response to imipramine in the mouse tail suspension test. Psychopharmacology (Berl) 1997;134(4):387–91. Swiergiel AH, Leskov IL, Dunn AJ. Effects of chronic and acute stressors and CRF on depression-like behavior in mice. Behavioural Brain Research 2008;186(1):32–40. Kamei J, Miyata S, Morita K, Saitoh A, Takeda H. Effects of selective serotonin reuptake inhibitors on immobility time in the tail suspension test in streptozotocin-induced diabetic mice. Pharmacology, Biochemistry, and Behavior 2003;75(2):247–54. Miyata S, Hirano S, Kamei J. Diabetes attenuates the antidepressant-like effect mediated by the activation of 5-HT1A receptor in the mouse tail suspension test. Neuropsychopharmacology 2004;29(3):461–9. Yamano M, Yuki H, Yasuda S, Miyata K. Corticotropin-releasing hormone receptors mediate consensus interferonalpha YM643-induced depression-like behavior in mice. The Journal of Pharmacology and Experimental Therapeutics 2000;292(1):181–7. Dunn AJ, Swiergiel AH. Effects of interleukin-1 and endotoxin in the forced swim and tail suspension tests in mice. Pharmacology, Biochemistry, and Behavior 2005;81(3):688–93. Cryan JF, Hoyer D, Markou A. Withdrawal from chronic amphetamine induces depressive-like behavioral effects in rodents. Biological Psychiatry 2003;54(1):49–58. Naudon L, El Yacoubi M, Vaugeois JM, Leroux-Nicollet I, Costentin J. A chronic treatment with fluoxetine decreases 5HT(1A) receptors labeling in mice selected as a genetic model of helplessness. Brain Research 2002;936(1–2):68–75. Cryan JF, Leonard BE. 5-HT1A and beyond: the role of serotonin and its receptors in depression and the antidepressant response. Human Psychopharmacology 2000;15(2):113–35. Thomson F, Craighead M. Innovative approaches for the treatment of depression: targeting the HPA axis. Neurochemical Research 2008;33(4):691–707. Crowley JJ, Blendy JA, Lucki I. Strain-dependent antidepressant-like effects of citalopram

43.

44.

45.

46.

47.

48.

49.

50.

51.

52.

53.

135

in the mouse tail suspension test. Psychopharmacology (Berl) 2005;183(2):257–64. Crowley JJ, Brodkin ES, Blendy JA, Berrettini WH, Lucki I. Pharmacogenomic evaluation of the antidepressant citalopram in the mouse tail suspension test. Neuropsychopharmacology 2006;31(11):2433–42. Liu X, Gershenfeld HK. Genetic differences in the tail-suspension test and its relationship to imipramine response among 11 inbred strains of mice. Biological Psychiatry 2001;49(7):575–81. Yamada K, Watanabe A, Iwayama-Shigeno Y, Yoshikawa T. Evidence of association between gamma-aminobutyric acid type A receptor genes located on 5q34 and female patients with mood disorders. Neuroscience Letters 2003;349(1):9–12. Edenberg HJ, Foroud T, Conneally PM, et al. Initial genomic scan of the NIMH genetics initiative bipolar pedigrees: chromosomes 3, 5, 15, 16, 17, and 22. American Journal of Medical Genetics 1997;74(3):238–46. Lad HV, Liu L, Paya-Cano JL, Fernandes C, Schalkwyk LC. Quantitative traits for the tail suspension test: automation, optimization, and BXD RI mapping. Mammalian Genome 2007;18(6–7):482–91. van der Heyden JA, Molewijk E, Olivier B. Strain differences in response to drugs in the tail suspension test for antidepressant activity. Psychopharmacology (Berl) 1987;92(1):127–30. Crowley JJ, Jones MD, O’Leary OF, Lucki I. Automated tests for measuring the effects of antidepressants in mice. Pharmacology, Biochemistry, and Behavior 2004;78(2):269–74. Chaki S, Nakazato A, Kennis L, et al. Anxiolytic- and antidepressant-like profile of a new CRF1 receptor antagonist, R278995/ CRA0450. European Journal of Pharmacology 2004;485(1–3):145–58. Nielsen DM, Carey GJ, Gold LH. Antidepressant-like activity of corticotropin-releasing factor type-1 receptor antagonists in mice. European Journal of Pharmacology 2004;499(1–2):135–46. Zobel AW, Nickel T, Kunzel HE, et al. Effects of the high-affinity corticotropinreleasing hormone receptor 1 antagonist R121919 in major depression: the first 20 patients treated. Journal of Psychiatric Research 2000;34(3):171–81. Nielsen DM. Corticotropin-releasing factor type-1 receptor antagonists: the next

136

54.

55.

56.

57.

58.

59.

60.

61.

62.

63.

64.

O’Leary and Cryan class of antidepressants? Life Sciences 2006;78(9):909–19. Czeh B, Fuchs E, Simon M. NK1 receptor antagonists under investigation for the treatment of affective disorders. Expert Opinion on Investigational Drugs 2006;15(5):479–86. Binneman B, Feltner D, Kolluri S, Shi Y, Qiu R, Stiger T. A 6-week randomized, placebo-controlled trial of CP-316,311 (a selective CRH1 antagonist) in the treatment of major depression. The American Journal of Psychiatry 2008;165(5):617–20. Katz MM, Koslow SH, Maas JW, et al. The timing, specificity and clinical prediction of tricyclic drug effects in depression. Psychological Medicine 1987;17(2):297–309. Harmer CJ, Bhagwagar Z, Perrett DI, Vollm BA, Cowen PJ, Goodwin GM. Acute SSRI administration affects the processing of social cues in healthy volunteers. Neuropsychopharmacology 2003;28(1): 148–52. Harmer CJ, Hill SA, Taylor MJ, Cowen PJ, Goodwin GM. Toward a neuropsychological theory of antidepressant drug action: increase in positive emotional bias after potentiation of norepinephrine activity. The American Journal of Psychiatry 2003;160(5):990–2. Crabbe JC, Wahlsten D, Dudek BC. Genetics of mouse behavior: interactions with laboratory environment. Science (New York) 1999;284(5420):1670–2. Wahlsten D, Metten P, Phillips TJ, et al. Different data from different labs: lessons from studies of gene-environment interaction. Journal of Neurobiology 2003;54(1): 283–311. Ripoll N, David DJ, Dailly E, Hascoet M, Bourin M. Antidepressant-like effects in various mice strains in the tail suspension test. Behavioural Brain Research 2003;143(2):193–200. Teste JF, Pelsy-Johann I, Decelle T, Boulu RG. Anti-immobility activity of different antidepressant drugs using the tail suspension test in normal or reserpinized mice. Fundamental & Clinical Pharmacology 1993;7(5):219–26. Cryan JF, Valentino RJ, Lucki I. Assessing substrates underlying the behavioral effects of antidepressants using the modified rat forced swimming test. Neuroscience and Biobehavioral Reviews 2005; 29(4–5):547–69. Cryan JF, Markou A, Lucki I. Assessing antidepressant activity in rodents: recent developments

65.

66.

67.

68.

69.

70.

71.

72.

73.

74. 75.

76.

and future needs. Trends in Pharmacological Sciences 2002;23(5): 238–45. Overmier JB, Seligman ME. Effects of inescapable shock upon subsequent escape and avoidance responding. Journal of Comparative and Physiological Psychology 1967; 63(1):28–33. Vollmayr B, Henn FA. Learned helplessness in the rat: improvements in validity and reliability. Brain Research Brain Research Protocols 2001;8(1):1–7. Willner P. Validity, reliability and utility of the chronic mild stress model of depression: a 10-year review and evaluation. Psychopharmacology (Berl) 1997;134(4):319–29. Kelly JP, Wrynn AS, Leonard BE. The olfactory bulbectomized rat as a model of depression: an update. Pharmacology & Therapeutics 1997;74(3):299–316. Weissman MM, Olfson M. Depression in women: implications for health care research. Science (New York) 1995;269(5225):799–801. Pelloux Y, Hagues G, Costentin J, DuterteBoucher D. Helplessness in the tail suspension test is associated with an increase in ethanol intake and its rewarding effect in female mice. Alcoholism, Clinical and Experimental Research 2005;29(3):378–88. David DJ, Nic Dhonnchadha BA, Jolliet P, Hascoet M, Bourin M. Are there gender differences in the temperature profile of mice after acute antidepressant administration and exposure to two animal models of depression? Behavioural Brain Research 2001;119(2):203–11. Caldarone BJ, Karthigeyan K, Harrist A, et al. Sex differences in response to oral amitriptyline in three animal models of depression in C57BL/6 J mice. Psychopharmacology (Berl) 2003;170(1):94–101. Jones MD, Lucki I. Sex differences in the regulation of serotonergic transmission and behavior in 5-HT receptor knockout mice. Neuropsychopharmacology 2005; 30(6):1039–47. Nichols DE, Nichols CD. Serotonin receptors. Chemical Reviews 2008;108(5):1614–41. Meziane H, Ouagazzal AM, Aubert L, Wietrzych M, Krezel W. Estrous cycle effects on behavior of C57BL/6 J and BALB/cByJ female mice: implications for phenotyping strategies. Genes, Brain, and Behavior 2007;6(2):192–200. Bernardi M, Vergoni AV, Sandrini M, Tagliavini S, Bertolini A. Influence of ovariectomy, estradiol and progesterone on the behavior of mice in an experimental model

Tail-Suspension Test

77.

78.

79.

80.

81.

of depression. Physiology & Behavior 1989;45(5):1067–8. Crawley JN, Belknap JK, Collins A, et al. Behavioral phenotypes of inbred mouse strains: implications and recommendations for molecular studies. Psychopharmacology (Berl) 1997;132(2):107–24. Mayorga AJ, Lucki I. Limitations on the use of the C57BL/6 mouse in the tail suspension test. Psychopharmacology (Berl) 2001; 155(1):110–2. Juszczak GR, Sliwa AT, Wolak P, TymosiakZielinska A, Lisowski P, Swiergiel AH. The usage of video analysis system for detection of immobility in the tail suspension test in mice. Pharmacology, Biochemistry, and Behavior 2006;85(2):332–8. Bai F, Li X, Clay M, Lindstrom T, Skolnick P. Intra- and interstrain differences in models of ‘‘behavioral despair’’. Pharmacology, Biochemistry, and Behavior 2001;70(2–3):187–92. Trullas R, Jackson B, Skolnick P. Genetic differences in a tail suspension test for evaluating antidepressant activity. Psychopharmacology (Berl) 1989;99(2):287–8.

137

82. Liu X, Gershenfeld HK. An exploratory factor analysis of the Tail Suspension Test in 12 inbred strains of mice and an F2 intercross. Brain Research Bulletin 2003;60(3):223–31. 83. Sallee FR, Pollock BG. Clinical pharmacokinetics of imipramine and desipramine. Clinical Pharmacokinetics 1990;18(5):346–64. 84. Popoli M, Brunello N, Perez J, Racagni G. Second messenger-regulated protein kinases in the brain: their functional role and the action of antidepressant drugs. Journal of Neurochemistry 2000;74(1):21–33. 85. Mombereau C, Kaupmann K, Froestl W, Sansig G, van der Putten H, Cryan JF. Genetic and pharmacological evidence of a role for GABA(B) receptors in the modulation of anxiety- and antidepressant-like behavior. Neuropsychopharmacology 2004;29(6):1050–62. 86. Porsolt R, Lenegre A. Behavioral models of depression. In: Elliott J, Heal D, Marsden C, eds. Experimental approaches to anxiety and depression. London: Wiley; 1992:73–85

Chapter 8 Stress-Induced Hyperthermia in the Mouse Christiaan H. Vinkers, Ruud van Oorschot, Berend Olivier, and Lucianne Groenink Abstract In anxiety research, the search for a model with sufficient clinical predictive validity to support the translation of animal studies on anxiolytic drugs to clinical research is challenging. The stress-induced hyperthermia (SIH) model studies the body temperature increase in response to acute stress which is mediated by the autonomic nervous system. This SIH response occurs in all mammals including humans, is easy to measure, reproducible and stable over time. The SIH paradigm has proven to possess excellent predictive validity, and various anxiolytic drugs have been shown to dose-dependently reduce the SIH response, including GABAA-receptor (subunit) agonists, 5-HT1A receptor agonists, mGlu5 receptor antagonists and CRF receptor antagonists. Therefore, the SIH model is a simple and attractive paradigm to study putative anxiolytic drug properties as well as the effects of genetic or brain manipulations. In the SIH procedure, drugs are injected 60 minutes before the actual stressor, consisting of a manual rectal temperature measurement (T1). After 10 minutes, a second manual rectal temperature measurement is taken (T2), which represents the stress-induced body temperature. The SIH (
T) response is the difference between T2 and T1 (
T=T2–T1). As drugs might exert intrinsic effects on the basal body temperature (measured via T1), the SIH response has to be interpreted in the context of absolute body temperatures. Mice can be repeatedly used (up to a year) if tested once a week because the SIH response remains very stable over such elongated periods. Animals can be socially housed except for the actual testing day (starting the day before until immediately after the procedure) when mice have to be singly housed. Key words: Stress, anxiety, stress-induced hyperthermia, SIH, emotional fever, body temperature, autonomic nervous system, thermoregulation, anxiolytics, GABAA receptor, 5-HT1A receptor, strain.

1. Background and Historical Overview 1.1. Introduction

All currently used animal models in anxiety have proven to be of great value, yet none of them is without limitations. Any animal model that uses a different approach would therefore be of additional value in anxiety and stress research. In this chapter, we propose that the stress-induced hyperthermia (SIH) model

T.D. Gould (ed.), Mood and Anxiety Related Phenotypes in Mice, Neuromethods 42, DOI 10.1007/978-1-60761-303-9_8, ª Humana Press, a part of Springer ScienceþBusiness Media, LLC 2009

139

140

H. Vinkers et al.

might be such a different approach (1). The SIH model uses the stress-induced activation of the autonomic nervous system by measuring the body temperature before and after stress exposure. The stress-induced increase in body temperature (SIH response, also called emotional or psychogenic fever) is a very consistent and stable physiological stress response with excellent translational properties. In mice, the SIH response remains intact for over a year when animals are used once a week (2–4) Furthermore, anxiolytic drugs have been shown to selectively and dosedependently reduce the SIH response (5). Besides, the SIH test is optimally suited to test the effects of various brain and genetic manipulations, e.g. null mutations or overexpression of genes (6, 7). Altogether, the SIH procedure is simple and robust, does not require time-consuming training, and drug effects on motor behaviour, feeding, drinking, and nociception do not affect test outcome. 1.2. The SIH Response

Exposure to both physiological and psychological stress increases body temperature in order to prepare for a fight-or-flight reaction. In humans, perceived stressful activities among which attending movies, boxing contests and taking an exam all increase body temperature (8–11). In rats and mice, exposure to any stressor (including novelty, heat or pain) also induces an increase in body temperature (Fig. 8.1). Within 15 minutes after stress exposure, body temperature rapidly rises up to 1.5C which usually returns to basal levels in 2 hours (12). Besides humans and rodents, the

Fig. 8.1. Telemetric measurement of the temperature course in a standard SIH test in 129S6 mice (n=7). At t=60 minutes, a vehicle is injected, causing a transient SIH response peaking at around 10–15 minutes post-injection and returning to baseline 45 minutes later. At t=120 minutes, a rectal temperature measurement (T1) is applied as a stressor and measures basal body temperature, and leads to an SIH response. In the SIH procedure in singly-housed mice, a second rectal measurement is carried out at t=130 minutes to assess the stress-induced temperature (T2). The SIH response (T2–T1) is around 1.4C. The error bar represents the SEM.

Stress-Induced Hyperthermia in the Mouse

141

SIH response has been found to be present in any mammal that has been tested so far, including impalas, silver foxes, sheep, pigs, rabbits and even cold-blooded animals such as turtles (12). The SIH curve parallels hypothalamic-pituitary-adrenal (HPA)-axis activity (13–15). Moreover, the SIH response appears relatively independent of locomotor activity (16, 17) which is a unique characteristic of an anxiety paradigm because it enables determination of anxiolytic activity of a drug independent from the often associated sedative properties that interfere with the primary parameter measured in the anxiety paradigm (e.g. movements on an elevated plus maze or in a light-dark box). Generally, the SIH procedure is not able to detect anxiogenic effects of drugs, probably due to a ceiling effect in the stress-induced temperature rise. 1.3. SIH and Anxiety: Development of a Model

The first notice of a temperature effect in response to stress was when a gradual temperature increase was noted after removing mice one by one from a group-housed cage and measuring their rectal body temperature. This SIH response was thought to represent anticipatory anxiety (18). Later on, the SIH model was refined to a singly housed version in which the rectal temperature of a single mouse was measured twice with an interval of 10 minutes (2). This reduced the number of animals needed per experiment. In this experimental setup, the first rectal temperature measurement (T1) not only represents the basal unstressed core temperature but also functions as an adequate stressor. The second rectal temperature measurement (T2) measured 10 minutes later represents the peak temperature after stress. The SIH response is calculated by subtracting T2 from T1 (SIH=
T=T2–T1). The singly and group-housed versions correlate very well with each other, indicating that both approaches are measuring the same neural processes (14). Telemetric systems measuring body temperature and locomotor activity are increasingly used in the SIH model, although the general principle of measuring an increase in body temperature in response to stress remains identical. However, because telemetric systems are able to measure locomotor activity, simultaneous application and comparison of body temperature and locomotor activity levels in response to stress are possible. As far as pharmacological studies have been performed in these SIH studies using telemetry equipment in rat and mice, results are very similar to those obtained in the standard SIH test in mice (1, 19).

1.4. Drug Testing in the SIH Model

The reproducible and robust SIH response combined with the ease of testing make the SIH paradigm ideal for detecting anxiolytic properties of drugs. The general principle is that anxiolytic drugs are able to reduce or even ablate the SIH response. Anxiolytic drugs including various GABAA receptor agonists, 5-HT receptor agonists and CRF receptor antagonists have been shown to indeed reduce the SIH response, whereas non-anxiolytic

142

H. Vinkers et al.

drugs including dopaminergic and noradrenergic compounds do not influence the SIH response (5). However, anxiolytics can also exert hypothermic effects and decrease basal core body temperature even before stress exposure occurs (20). This is probably due to an activation of the GABAA receptor 1 subunit, leading to a disturbed homeostatic regulation of the basal regulation of core body temperature (21). It is therefore important to not only regard temperature differences, but rather to study the absolute temperature values when assessing the effects of a compound (Fig. 8.2).

Fig. 8.2. Theoretical outcome of the effects of a psychoactive drug (dose 0-1-2 on the x axis) in the SIH paradigm. Depending on whether a drug decreases (left column), has no effect on (middle column), or enhances (right column) the basal temperature (T1) the effects on SIH (
T) can be interpreted as either no effect (top row), anxiolytic effect (middle row) or anxiogenic effect (bottom row). The difference between the rectal temperatures measured with a 10-minutes interval (T2–T1) is the SIH response (
T).

The SIH model possesses good predictive validity for anxiolytic drug properties. Clinically effective anxiolytic compounds such as benzodiazepines (including alprazolam, oxazepam, diazepam and chlordiazepoxide) and 5-HT1A receptor agonists such as buspirone decrease the SIH response (20). Most research has been carried out in rodents (for a review on the pharmacological SIH evidence see ref. (12)) and a wide range of different drug classes have been tested thus far (1). Acute effects of selective serotonin reuptake inhibitors (SSRIs) are absent in the SIH (20), whereas

Stress-Induced Hyperthermia in the Mouse

143

chronic SSRI treatment generates inconsistent findings; chronic fluoxetine treatment either had no influence (22) or reduced the acute SIH response (23). Although there is a linear relationship between stressor intensity and magnitude of the SIH response (19), so far no anxiogenic effects have been observed in the SIH test (1, 5). Anxiogenic compounds such as FG7142 and mCPP did not alter basal body temperature or the SIH response, whereas a high dose of pentylenetetrazol significantly reduced basal body temperature but was also without effect on the SIH response (20). In the standard manual SIH procedure, drugs are administered 60 minutes before stress is applied. This period has been experimentally validated (20), since drug administration in itself is a stressful event involving handling and manipulating the animal, causing an interfering SIH response. Because the maximum body temperature is limited up to certain levels, a newly applied stressor within the first 15 minutes does not cause any additional SIH response, although it prolongs the SIH response (2). Thus, an injection-stressor interval of 60 minutes ensures that body temperature has sufficiently declined to approximately pre-stress baseline levels. In support, injections 60 minutes before assessing the SIH response in mice lead to an identical SIH response as compared to the mice that had received no injection at all (Fig. 8.3C)

a

b

c

Fig. 8.3. Typical output from a stress-induced hyperthermia test in singly-housed male mice. The effects of a dose range of (A) the benzodiazepine receptor agonist chlordiazepoxide (a standard anxiolytic drug) and (B) the 5-HT1A receptor agonist flesinoxan and (C) different types of vehicles (none, gelatine–mannitol, methylcellulose and saline) on SIH (
T), basal temperature (T1) and T2. Drugs (or vehicle) were administered 60 minutes before the first rectal temperature measurement (T1). T2 was measured 10 minutes later. The difference between T2 and T1,
T, is indicated at each dose (inset: *
T significantly different from vehicle treatment, indicating an anxiolytic effect). An asterisk (*) indicates a significant difference (P < 0.05) from the corresponding vehicle treatment (0 mg/kg).

144

H. Vinkers et al.

2. Factors Influencing the SIH Response 2.1. Strain Differences

The SIH response has proven to be robust and reproducible between various labs. The absolute temperature increase after stress exposure depends on various factors. First of all, mouse strains generally differ in magnitude regarding their stress response which can be attributed to their genetic background as well as epigenetic factors (3, 17, 24). Nine different mouse strains (including Swiss, NMRI, C57BL6, 129Sv, FVB, and DBA/2) showed SIH responses between 0.6 and 1.9C (3). Differences in locomotor activity and body weight alone cannot account for the differences in SIH responses between strains, indicating the necessity to assess animal and strain stress responsiveness before any experiment is initiated (17). However, when comparing three different mouse strains (C57BL/6 J, Swiss-Webster and 129 Sv/ Ev mice), none of the three strains was consistently more sensitive to anxiolytic-like (SIH) or intrinsic drug effects on basal body temperature (1).

2.2. Type of Stressor

An increase in body temperature in response to stress can be induced up to certain levels above which no further temperature rise is possible. This ceiling effect limits the maximum SIH response, and any stressor seems to induce sufficient stress to reach a reliable SIH response (2). However, minor stressors such as entering the room where the animal is housed induce smaller SIH amplitudes compared to placing animals in a novel cage, which is in turn exceeded by intense stressors such as repeated social defeat (16, 25, 26). More subtle differences in stress intensity are generally not easily distinguishable, although the duration of the stress effect – the time needed to return to baseline level – correlates well with stress intensity (17).

2.3. Fever State, High and Low Environment Temperature

In infection-induced fever, the core body temperature is drastically enhanced by prostaglandins, lasting from hours to days long, as prostaglandins activate the hypothalamic pre-optic area. In contrast, the duration of a typical SIH response is maximally 2–3 hours. Furthermore, fever states do not eliminate the SIH response, although the SIH amplitude is decreased due to ceiling effects (1). In addition, the SIH response is not sensitive to prostaglandin-blocking drugs, whereas infection-induced fever is hardly responsive to the effect of benzodiazepines (27). This indicates that, even though infection and stress both increase body temperature, they are mediated by different neurotransmitters and brain mechanisms and areas. In general, environmental temperature has direct effects on body temperature homeostasis and resting body temperature (28),

Stress-Induced Hyperthermia in the Mouse

145

not only affecting SIH but also influencing infection-induced fever (29, 30). A high basal body temperature interferes with the SIH response (3, 31), and housing mice at 35C rather than 23C increases body core temperature from around 37C to around 39.5C, and thus likely decreases the SIH response (28). In contrast, housing animals at 11C instead of 24C does not interfere with SIH amplitude (32). Body temperature displays circadian rhythmicity with a 1–2C temperature increase during the dark period. This can influence the SIH amplitude (33, 34), although no consistent studies have been carried to test whether this affects drug sensitivity. Generally, SIH testing is performed during the light period when basal body temperature is lower. In support, one study showed that the SIH amplitude was indeed smaller during the dark phase of the light–dark cycle (35). 2.4. Habituation

3. Equipment, Materials and Setup

Repeated daily stress exposure results in habituation in many anxiety paradigms including the light–dark box (36), the open field (37) and to some – but very limited – degree in the SIH model (2). Because the SIH response partially depends on perceived stress intensity, repeated daily testing with moderate stressors results in a decreasing SIH amplitude, even though a robust SIH response is still discernable (2, 25, 26, 38). Similarly, daily injection stress for six consecutive days decreases the SIH response in rats (own unpublished data). Surprisingly, exposure to a very severe stressor like repeated social defeat does not lead to a habituated SIH response (16, 25, 26). Generally, a 1-week interval is sufficient to prevent any habituation, and testing once weekly with moderate stressors has been shown not to interfere with the SIH response for over a year (2–4, 20).

1. Male mice, weighing 18–20 g upon arrival in the lab (e.g. NMRI mice, Charles River, strain and body weight are not critical). In general 10–12 animals per dose group are used. Mice are group-housed between experiments. 2. Experimental cages for the animals: 27  16  12.5 cm cages with sawdust for group housing between experiments, and 12  18  13 cm cages with sawdust for individual housing. 3. Balances with accuracy of 0.5 g (e.g. Mettler PG5000) 4. Test compound solutions: prepare all suspensions and solutions fresh daily. For solutions, prepare compounds in physiological saline or in a 1% (w/v) methylcellulose/5% (w/v) mannitol mixture. For suspensions, prepare test

146

H. Vinkers et al.

compounds in 1% (w/v) tragacanth, 1% (w/v) methylcellulose or gelatin/mannitol. Other vehicles might also be used but have to be tested in advance to exclude intrinsic temperature effects or worse, act as a stressor (e.g. high concentrations of DMSO). 5. 1-ml syringes, sterile, non-toxic, and pyrogen-free; sterile 25-G  5/8-inch (16-mm) needles. 6. Silicon oil, peanut oil, or K-Y jelly, at room temperature. 7. Digital thermometer, NiCr-NiAl thermocoupled with accuracy of 0.1C (e.g. Keithley 871A, Cleveland, OH). 8. Data analysis software for statistical analysis (e.g. SAS, SPSS).

4. Procedure 4.1. Acclimatization of Mice

Mice should be ordered in time so that they can acclimatize to constant laboratory conditions at least 1 week before starting the experiment. Group-house the animals (five per cage) at constant room temperature (21 – 2C) and relative humidity (60 – 10%) under non-reversed 12 hours light/12 hours dark cycle (light on at 06:00). Provide cages with sawdust and free access to standard rodent diet and water.

4.2. Set Up and Run Experiment

In the afternoon before the test, house each mouse individually in an experimental cage in the experimental room. While isolating the mice, weigh and tail-mark each animal and write down its weight on a run sheet. Make sure that the housing conditions in the experimental room (day/night rhythm, temperature and humidity) are identical to those in the acclimatization room, and that water and food are freely available. On the experimental day, prepare a balance, a digital thermometer, a tray with silicon oil (peanut oil or K-Y jelly), and sufficient cleaning tissues, 1-ml syringes and 25-G  5/8-inch needles. Prepare test compound solutions such that all substances are administered at 10 ml/kg. Use a coding system (A, B, C, etc.) so that treatments are administered blindly. Ensure an even distribution of the different treatments over time (e.g. by using a Latinsquare design).

4.3. Time Schedule

The time schedule below is used for testing 80 mice on 1 day, (40 mice in the morning, 40 mice in the afternoon, taking approximately 5 hours in total), with an injection test interval of 60 minutes. We assume starting the tests in the morning at 9:00 AM, but adjustments can easily be made if starting at a different time.

Stress-Induced Hyperthermia in the Mouse

147

– At 09:00, inject the first mouse (10 ml/kg body weight) with the test compound solution indicated by a previously prepared run sheet and return the mouse to its own cage. Then inject nine other mice, using a 1-minute interval between successive injections. – Between 09:20 and 09:30, inject the second cohort of 10 mice, as described above. – Between 09:40 and 9:55, inject the third and fourth cohort of 10 mice, as described above. Injections of the last 10 mice are performed a bit early so that they do not conflict with temperature measurements to be taken at 10:00. – From 10:00 to 10:10 (i.e. 60 minutes after the injection), measure the rectal temperature (T1) of the first 10 mice using a 1-minutes interval between successive measurements. – For temperature measurements, fixate the mouse horizontally. While also fixating its tail at the base, dip the probe into silicon oil, insert it 2 cm into the rectum, and hold until a stable rectal temperature is measured for approximately 20 seconds. Write down the temperature readout with 0.1C accuracy, and return the mouse to its individual cage. – From 10:10 to 10:20, measure the rectal temperature (T2) of the first 10 mice again using a 1-minutes interval between successive measurements. – At 10:20, start rectal temperature measurement (T1) of the second cohort of 10 mice (injected between 09:20 and 09:30) with 1-minutes intervals as described before. – From 10:30 to 10:40, measure the rectal temperature (T2) of the second cohort of 10 mice again using a 1-minutes interval between successive measurements as described previously. – At 10:40, start rectal temperature measurement (T1) of the third and fourth cohort of 10 mice (injected between 09:40 and 09:55) with 1-minutes intervals. – From 10:55 to 11:05, measure the rectal temperature (T2) of the third and fourth cohort of 10 mice again using a 1minutes interval between successive measurements as described previously. – Repeat injections and measurements for another 40 mice, with another compound in the afternoon. – Return the mice to their group cages (keep always the same animals per cage) after completion of the whole test. 4.4. Perform Statistical Analysis

Use standard software packages to perform graphical presentations (e.g. Excel or Sigmaplot) and statistical analyses (e.g. SAS or SPSS).

148

H. Vinkers et al.

Enter the raw data (two temperature measurements per mouse) in an empty raw data file. Calculate
T (the SIH response) for each mouse as the difference between T2 and T1. Then calculate mean T1, mean T2 and mean
T for each treatment condition. Also calculate the standard error of the mean. Check homogeneity of variance for T1, T2 and
T between treatments, and evaluate the effects of treatment by a two-way analysis of variance (ANOVA) with treatment (four levels if vehicle plus three doses of a drug are tested) as between-subject’s factor and stress level (T1, T2) as within-subject’s factor. For analysing drug effects on
T, perform a one-way ANOVA with treatment condition as between-subject’s factor (four levels). Subsequently, use a post hoc analysis (Dunnett’s multiple comparison test) to determine which treatment dose significantly alters SIH (
T) and/or basal core temperature (T1) relative to vehicle treatment.

5. Experimental Variables and Troubleshooting

This SIH procedure is based on measuring stress-induced changes in rectal temperature. Therefore, it is important that the environmental noise in the experimental room is stable during the test day and over days (5). Furthermore, room climate conditions should be kept constant to eliminate variation in results to prevent possibly interfering effects of environmental temperature on the SIH response (30, 32). Most SIH studies have been performed during the inactive light period (3–9 hours after lights on), when body temperature is 1–2C lower compared to the dark period, and relatively stable (5). In the standard SIH set-up, mice are group-housed and isolated in the afternoon before the test day and returned to their original group cage after the test. It is not absolutely necessary, however, to group-house the mice first. Several studies have shown that also prolonged individual housing does not necessarily influence the SIH response or the pharmacological sensitivity (17). From an ethical point of view, however, it is advised to group-house the mice, especially if they are tested several times in the SIH procedure. Drugs may be administered orally, subcutaneously, or intraperitoneally. However, the injection-test interval should always be at least 45, but preferably 60 minutes. Drug administration injection itself induces a SIH response, and it takes approximately 60 minutes for the body temperature to return to basal levels (2). Using a shorter injection-test interval, the resulting SIH response may be too small to detect significant drug effects in mice (2). A drawback of this rather long

Stress-Induced Hyperthermia in the Mouse

149

injection-stress interval is that drugs need to possess sufficient long half-lives to still be present in effective blood concentrations at the time of stress application 60 minutes later (e.g. nicotine:(5)). Stress-free administration involving subcutaneous drug administration connected to flexible injection lines, in which handling and even disturbing the animal are no longer be necessary, is used in our laboratory to overcome the troubles associated with rapidly metabolized drugs with very short half-lives. In contrast to the results in mice, we have found that in rats, a 10-minute injection-test interval did not lead to an immediate increase in body temperature (39). This indicates that, at least in rats, a shorter injection-stressor interval may be applied. It is also important to control for inter-experimenter effects on the SIH response. Different experimenters may obtain different mean temperature measurements, due to variations in the fixation techniques applied by investigators (e.g. head of the mouse up or down) (40). Although these interexperimenter differences in mean temperature measurements not necessarily affect the SIH response (40), investigators should be trained in the technique holding the mice horizontally.

6. Anticipated Results Theoretically, a compound can either influence basal body temperature, the SIH response, both or none. Figure 8.2 graphically shows these possible drug effects on basal temperature (T1), the stressinduced temperature (T2) and the SIH response (
T). Statistically, an effect on basal body temperature is indicated by a main treatment drug effect in a two-way ANOVA, whereas an anxiolytic effect is indicated by a significant interaction of stress level (T1/T2)  treatment in the two-way ANOVA. A compound that does not influence basal body temperature (T1) (middle column), has three theoretical effects: no influence on SIH (no anxiolytic activity), a decrease in SIH (anxiolytic effect) or an increase in SIH (anxiogenic effect). A compound that decreases basal body temperature (T1) (left column) results in hypothermia. Subsequently, the hypothermia leads to no influence on SIH (no anxiolytic activity), a decreased SIH response (with a steeper decline in T2 compared to T1, indicating an anxiolytic effect) or an increased SIH response (anxiogenic effect). Similar effects can be found with compounds that increase basal body temperature (right column).

150

H. Vinkers et al.

7. Notes – Anxiogenic effects have never been reported in the SIH test because of an apparent ceiling effect of the stress-induced temperature. – Increased basal body temperature with unaltered stressinduced temperature levels results in a decreased SIH response (and a significant stress level  treatment interaction). This is however not an anxiolytic drug effect. It is therefore vital to assess both the absolute temperature values as well as the SIH response. Figure 8.3A shows the results of a typical SIH experiment using three doses of the classical benzodiazepine chlordiazepoxide, an anxiolytic GABAA receptor agonist. In drug-treated mice, the SIH response is significantly decreased compared to
T in vehicle-treated animals (one-way ANOVA F(3,46)= 4.96, p=0.005 with Dunnett’s multiple comparison as post hoc test). Moreover, the stronger reduction in T2 than in T1 following treatment with chlordiazepoxide is reflected in a significant stress  treatment interaction; F(1,43)= 5.6, p<0.01. Together these data indicate an anxiolytic-like effect of chlordiazepoxide. Figure 8.3B shows the effects of flesinoxan, a 5-HT1A receptor agonist in the standard SIH test. In drug-treated mice
T is significantly decreased compared to
T in vehicle-treated animals (one-way ANOVA F(3,43)= 15.3, p<0.001, with Dunnett’s multiple comparison as post-hoc test), Moreover, the stronger reduction in T2 than in T1 following treatment with flesinoxan is reflected in a significant stress * treatment interaction F(3,43)= 15.2, p<0.001. Together these data indicate an anxiolytic-like effect of flesinoxan. Figure 8.3C shows that different vehicle types (gelatin–mannitol, methycellulose, saline) administered 60 minutes before measuring the rectal temperature lead to similar subsequent SIH responses, even compared when no vehicle at all is administered. Together, these data indicate that a 60-minute injection-stressor interval is sufficiently long that injection stress itself does not longer interfere with the stress procedure.

8. Conclusion Altogether, the SIH procedure is simple to perform without extensive or intrusive procedures. Moreover, the stress-induced increase in rectal temperature is a robust and reproducible parameter found

Stress-Induced Hyperthermia in the Mouse

151

in all mammals, including all strains of mice tested thus far (3, 24). So far, available anxiolytic drugs have been shown to reduce the SIH response in rodents. The SIH response is therefore an autonomic stress response that can be successfully studied at the level of its physiology, pharmacology, neurobiology and genetics and possesses excellent animal-to-human translational properties. References 1. Vinkers CH, van Bogaert MJ, Klanker M, et al. Translational aspects of pharmacological research into anxiety disorders: the stressinduced hyperthermia (SIH) paradigm. Eur J Pharmacol 2008;585(2–3):407–25. 2. Van der Heyden JA, Zethof TJ, Olivier B. Stress-induced hyperthermia in singly housed mice. Physiol Behav 1997; 62(3):463–70. 3. Bouwknecht JA, Paylor R. Behavioral and physiological mouse assays for anxiety: a survey in nine mouse strains. Behav Brain Res 2002;136(2):489–501. 4. Bouwknecht JA, van der Gugten J, Groenink L, Olivier B, Paylor RE. Effects of repeated testing in two inbred strains on flesinoxan dose-response curves in three mouse models for anxiety. Eur J Pharmacol 2004;494(1):35–44. 5. Bouwknecht JA, Olivier B, Paylor RE. The stress-induced hyperthermia paradigm as a physiological animal model for anxiety: a review of pharmacological and genetic studies in the mouse. Neurosci Biobehav Rev 2007;31(1):41–59. 6. Groenink L, van Bogaert MJ, van der Gugten J, Oosting RS, Olivier B. 5-HT1A receptor and 5-HT1B receptor knockout mice in stress and anxiety paradigms. Behav Pharmacol 2003;14(5–6):369–83. 7. Pattij T, Hijzen TH, Groenink L, et al. Stress-induced hyperthermia in the 5HT(1A) receptor knockout mouse is normal. Biol Psychiatry 2001;49(7):569–74. 8. Briese E. Emotional hyperthermia and performance in humans. Physiol Behav 1995;58(3):615–8. 9. Kleitman N, Jackson DP. Body temperature and performance under different routines. J Appl Physiol 1950;3(6):309–28. 10. Marazziti D, Di Muro A, Castrogiovanni P. Psychological stress and body temperature changes in humans. Physiol Behav 1992;52(2):393–5. 11. Renbourn ET. Body temperature and pulse rate in boys and young men prior to sporting contests. A study of emotional hyperthermia:

with a review of the literature. J Psychosom Res 1960;4:149–75. 12. Adriaan BJ, Olivier B, Paylor RE. The stress-induced hyperthermia paradigm as a physiological animal model for anxiety: a review of pharmacological and genetic studies in the mouse. Neurosci Biobehav Rev 2007;31(1):41–59. 13. Groenink L, van der Gugten J, Zethof T, van der Heyden J, Olivier B. Stress-induced hyperthermia in mice: hormonal correlates. Physiol Behav 1994;56(4):747–9. 14. Spooren WP, Schoeffter P, Gasparini F, Kuhn R, Gentsch C. Pharmacological and endocrinological characterisation of stressinduced hyperthermia in singly housed mice using classical and candidate anxiolytics (LY314582, MPEP and NKP608). Eur J Pharmacol 2002;435(2–3):161–70. 15. Veening JG, Bouwknecht JA, Joosten HJ, et al. Stress-induced hyperthermia in the mouse: c-fos expression, corticosterone and temperature changes. Prog Neuropsychopharmacol Biol Psychiatry 2004; 28(4):699–707. 16. Pardon MC, Kendall DA, Perez-Diaz F, Duxon MS, Marsden CA. Repeated sensory contact with aggressive mice rapidly leads to an anticipatory increase in core body temperature and physical activity that precedes the onset of aversive responding. Eur J Neurosci 2004;20(4):1033–50. 17. Van Bogaert M, Oosting R, Toth M, Groenink L, van Oorschot R, Olivier B. Effects of genetic background and null mutation of 5-HT1A receptors on basal and stress-induced body temperature: modulation by serotonergic and GABAA-ergic drugs. Eur J Pharmacol 2006;550(1–3):84–90. 18. Borsini F, Lecci A, Volterra G, Meli A. A model to measure anticipatory anxiety in mice? Psychopharmacology (Berl) 1989;98(2):207–11. 19. van Bogaert MJ, Groenink L, Oosting RS, Westphal KG, van der Gugten J, Olivier B. Mouse strain differences in autonomic

152

20.

21.

22.

23.

24.

25.

26.

27.

28.

29.

H. Vinkers et al. responses to stress. Genes Brain Behav 2006;5(2):139–49. Olivier B, Zethof T, Pattij T, et al. Stressinduced hyperthermia and anxiety: pharmacological validation. Eur J Pharmacol 2003;463(1–3):117–32. Vinkers CH, Klanker M, Groenink L, et al. Dissociating anxiolytic and sedative effects of GABAAergic drugs using temperature and locomotor responses to acute stress. Psychopharmacology (Berl) 2009;204:299–311. Roche M, Harkin A, Kelly JP. Chronic fluoxetine treatment attenuates stressorinduced changes in temperature, heart rate, and neuronal activation in the olfactory bulbectomized rat. Neuropsychopharmacology 2007;32(6):1312–20. Conley RK, Hutson PH. Effects of acute and chronic treatment with fluoxetine on stress-induced hyperthermia in telemetered rats and mice. Eur J Pharmacol 2007;564(1–3):138–45. Rodgers RJ, Boullier E, Chatzimichalaki P, Cooper GD, Shorten A. Contrasting phenotypes of C57BL/6JOlaHsd, 129S2/SvHsd and 129/SvEv mice in two explorationbased tests of anxiety-related behaviour. Physiol Behav 2002;77(2–3): 301–10. Barnum CJ, Blandino P, Jr., Deak T. Adaptation in the corticosterone and hyperthermic responses to stress following repeated stressor exposure. J Neuroendocrinol 2007;19(8):632–42. Bhatnagar S, Vining C, Iyer V, Kinni V. Changes in hypothalamic-pituitary-adrenal function, body temperature, body weight and food intake with repeated social stress exposure in rats. J Neuroendocrinol 2006;18(1):13–24. Vinkers CH, Groenink L, Bogaert Mv, et al. Stress-induced hyperthermia and infectioninduced fever: two of a kind? Physiol Behav (2009), doi:10.1016/j.physbeh.2009.04.004 Jiang Q, Cross AS, Singh IS, Chen TT, Viscardi RM, Hasday JD. Febrile core temperature is essential for optimal host defense in bacterial peritonitis. Infect Immun 2000;68(3):1265–70. Buchanan JB, Peloso E, Satinoff E. Influence of ambient temperature on peripherally induced interleukin-1 beta fever in young and old rats. Physiol Behav 2006;88(4–5):453–8.

30. Peloso ED, Florez-Duquet M, Buchanan JB, Satinoff E. LPS fever in old rats depends on the ambient temperature. Physiol Behav 2003;78(4–5):651–4. 31. Dymond KE, Fewell JE. Gender influences the core temperature response to a simulated open field in adult guinea pigs. Physiol Behav 1999;65(4–5):889–92. 32. Long NC, Vander AJ, Kluger MJ. Stressinduced rise of body temperature in rats is the same in warm and cool environments. Physiol Behav 1990;47(4):773–5. 33. Olivier B, van Bogaert, M., van Oorschot, R., Oosting, R., Groenink, L. Stress-induced hyperthermia. Handbook of Stress and the Brain, Steckler, T, Kalin, NH, Reul, JM (Eds), vol 15 Elsevier, Amsterdam 2005;135–55. 34. Peloso E, Wachulec M, Satinoff E. Stressinduced hyperthermia depends on both time of day and light condition. J Biol Rhythms 2002;17(2):164–70. 35. Caramaschi D, de Boer SF, Koolhaas JM. Differential role of the 5-HT1A receptor in aggressive and non-aggressive mice: an across-strain comparison. Physiol Behav 2007;90(4):590–601. 36. Onaivi ES, Martin BR. Neuropharmacological and physiological validation of a computer-controlled two-compartment black and white box for the assessment of anxiety. Prog Neuropsychopharmacol Biol Psychiatry 1989;13(6):963–76. 37. Cook MN, Bolivar VJ, McFadyen MP, Flaherty L. Behavioral differences among 129 substrains: implications for knockout and transgenic mice. Behav Neurosci 2002;116(4):600–11. 38. Thompson CI, Brannon AJ, Heck AL. Emotional fever after habituation to the temperature-recording procedure. Physiol Behav 2003;80(1):103–8. 39. Vinkers CH, de Jong NM, Kalkman CJ, et al. Stress-induced hyperthermia is reduced by rapid-acting anxiolytic drugs independent of injection stress in rates, Pharmacology, Biochemistry and Behavior (2009), doi:10.1016/j.pbb.2009.05.017 40. Zethof TJ, Van der Heyden JA, Tolboom JT, Olivier B. Stress-induced hyperthermia in mice: a methodological study. Physiol Behav 1994;55(1):109–15.

Chapter 9 Factors of Reproducibility of Anhedonia Induction in a Chronic Stress Depression Model in Mice Tatyana Strekalova and Harry Steinbusch Abstract Despite overall extensive use of various chronic stress models in mice during the past decades, the reproducibility of induction of anhedonia and a depressive-like syndrome with this method remains to be dissatisfying. Generally, this is related to problems of stable induction of a depressive-like state (according to a selected criterion) and limitations of behavioral methods detecting depressive-like state in animals. Here, we focus on the first part of the problem. A number of evidences suggest that in mice, the ability of different chronic stress protocols to evoke a depressive-like syndrome and anhedonia, defined by a decrease in sucrose preference, depends on the stress impact of the stress procedure. Data obtained in C57BL/6 strain show that anhedonia could be induced with chronic stress procedures of such intensity and duration, which result in a reduction of body weight. Though, chronic stress protocols, which do not cause a loss of body mass, evoke other behavioral effects in mice, as for instance, increased scores of anxiety and locomotion. In C57BL/6 mice, these behavioral changes occur also with anhedonia-inducing stress regimens. They appear in stressed individuals both susceptible and resilient to depressive-like changes, i.e., without relation to a presence of anhedonia and depressive-like syndrome. Together, these data, first, suggest that behavioral effects induced by chronic stress should be interpreted with a caution while attempting to model depressive symptoms in mice. Chronic stress can evoke general ‘‘non-specific’’ behavioral changes that do not imply an induction of a depressive-like phenotype in rodents. Second, obtained findings led to consider the control over a stress load during the chronic stress experiment as a reasonable approach of ensuring the reproducibility with induction of anhedonia and depressive-like syndrome in mice. The data on inter-batch fluctuation of behavioral traits and stress-response, generally described in laboratory mice, further suggest an importance of a control/moderation of the stress load, with either above-proposed or other validated criteria. We propose a reduction of weekly evaluated body weight and immediate after-stress hypoactivity as criteria of stress load, optimal for anhedonia induction, in C57BL/6 mice. Thus, adjustment of a stress procedure to the characteristics of behavioral stress response observed in a tested batch of animals, instead of using once-defined stress protocol could be a reasonable strategy while aiming at higher reproducibility with chronic stress depression paradigms in mice. Key words: Chronic stress depression model, reproducibility, stress load, individual variability, sucrose test, anhedonia, mouse.

T.D. Gould (ed.), Mood and Anxiety Related Phenotypes in Mice, Neuromethods 42, DOI 10.1007/978-1-60761-303-9_9, ª Humana Press, a part of Springer ScienceþBusiness Media, LLC 2009

153

154

Strekalova and Steinbusch

1. Background and Theoretical Overview 1.1. Chronic Stress as a Method of Induction of Anhedonia and Depressive-Like State in Rodents

Analysis of reports gathered over a 40-year period evidences a prevalent use of experimental stress paradigms over the application of other experimental approaches in modeling human psychopathology. Particularly, chronic stress became one of the most established methods in modeling depressive disorder in animals (1, 2). It is regarded as probably the most etiologically relevant paradigm of induction of a depressive-like state in rodents, since stress is a major environmental factor of the pathogenesis of depression (3, 4). With the chronic stress depression paradigms, originally elaborated by Katz and Willner on the rat, several stressors are applied in rotating manner to evoke a depressive-like state (5–7). Typically, it is defined by a presence of anhedonia, a diminished sensitivity to reward stimuli, which putative behavioral indicator in rodents is a decreased intake of and preference for palatable solutions, such as sucrose and saccharine. Application of hedonic deficit as a criterion of a depressive-like state in the chronic stress models responds the face-validity criterion with modeling depression, since according to a DSMIV classification, presence of anhedonia, a key symptom of depression, is a basis of depressive disorder diagnostics (8, 9, 10). Moreover, because anhedonia is a core symptom of depression and can be mimicked in rodents, a hedonic deficit can be generally regarded as a primary feature to be addressed while modeling depression in animal paradigms (11, 12). Initially, Katz et al. (13) applied a 3-week-long stress procedure in rats, comprising quite intense stressors: electric foot shock, food and water deprivation, exposure to heat, swimming in cold water and others. This protocol caused a decrease in intake of sucrose solution that was interpreted as a sign of the anhedonia. Hedonic deficit in rats was reversible by antidepressants, but not by other compounds, such as anxiolitics and neuroleptics (14). Probably, the most elaborate and broadly used animal paradigm of depression, which mimics anhedonia in rodents, is the chronic mild-stress model of Willner (15, 16, 5). To develop a more realistic model of depression, in comparison to Katz’ paradigm, Willner modified the proposed protocol tending to simulate more closely the human stressors. For this propose, the strength of stressors was essentially blunted. Mild stressors, such as soiled cage, presence of a foreign object, restricted access to food, constant lighting and others were used; exposure to the stress was prolonged from 3 weeks to 3 months. With this protocol, a lasting state of anhedonia, as

Factors of Reproducibility of Anhedonia Induction

155

shown by a decrease in sucrose intake and preference and an increase in the thresholds of intracranial self-stimulation in rats, was observed. Hedonic deficit was again found to be selectively reversible by antidepressants. Based on the initially established theoretical and practical principles of modeling depression with a chronic stress method, i.e., application of a hedonic deficit as a criterion of depressivelike status and the use of a prolonged exposure to rotating stressors of mild intensity for anhedonia induction, a number of stress procedures were developed during the following decades in rats and mice (17–26, 7). Majority of the chronic stress paradigms were developed in young male rats; relatively little attention was paid so far to a role of age, gender, and speciesspecific differences, including differences between rats and mice. The advent of genetically modified mice, providing unique tool for addressing the neurobiological mechanisms of depressive-like phenotype, significantly shifted the researchers’ interest from a rat to a mouse (27, 28). Therefore, efforts were made to translate a rat chronic stress model to similar mouse paradigm. Meanwhile, in comparison to the chronic stress depression paradigms in rats, mouse models are considered to be more difficult because of the problems with reproducibility. By now, many groups reported inconsistencies in the induction of the hedonic deficit and other depressive-like effects with mouse chronic stress procedures (29–32). Obviously, these problems stimulated the use of other behavioral parallels of depressive syndrome, besides accepted parameters of anhedonic state, e.g., deterioration of the coat state (22) and social avoidance (26). Recently, we have established a mouse model of stressinduced anhedonia with internal control for chronic stress, in which hedonic deficit, defined by a decrease in sucrose preference, is used as a parameter of a depressive-like state (33). With this model, we developed and applied criteria of optimal stress load, so that the chronic stress procedure reproducibly leads to a decrease in sucrose preference, a behavioral parallel of human anhedonia, and other features of a depressive-like syndrome. Application of a resilient subgroup of mice, comprised 30–50% of stressed mice, which does not develop a hedonic deficit, as a control for the effects of stress alone, and its comparison to the anhedonic group in pharmacological, physiological, and other assays suggested high validity of a proposed chronic stress model in mimicking human depression (12). With a proposed model, we modified behavioral testing and stress procedure attempting to resolve a problem of reliability in anhedonia induction. Because in general, the reproducibility in modeling of depressive-like state is determined by (1) a reproducibility of induction of a depressive-like state in animals,

156

Strekalova and Steinbusch

according to a selected criterion and (2) resolution of the methods detecting this state, a refining of anhedonia-inducing stress regimen and a sucrose test protocol was a priority with elaboration of our paradigm. The optimization of a sucrose test was described earlier (33, 12). Here, we focus on the first factor of reproducibility of anhedonia induction. 1.2. Stress Intensity/ Duration, Body Weight Changes, and Anhedonia Induction

Using the data obtained on male C57 BL6 mice, we analyzed how differences between the chronic stress protocols in duration and stress impact, assessed by changes in body weight, affect the induction of anhedonia, determined in a sucrose test with identical testing conditions. In one of the studies, we compared prolonged stress of weaker intensity (‘‘weak’’ 6-week chronic stress) and shorter stress of stronger intensity (‘‘standard’’ 4-week chronic stress) by their ability to induce anhedonia, defined by a decrease in sucrose preference below 65%. The arbitrarily taken criterion for anhedonia is based on our results, which indicated that mice with a sucrose preference  65% demonstrate a depressive-like syndrome shown by elevated floating in the forced swim test, increased immobilization in the tail suspension test, decreased novelty exploration, and other depressive-like changes, while stressed mice with a sucrose preference above this value did not display the above behavioral phenotype (33). A validity of the selected criterion was additionally supported by the data with antidepressant treatment applied in our model (34). Reduction in body weight was used as an indirect indicator of stress load. In addition to the sucrose preference, we evaluated the anxiety-like behavior in the O-maze test, which was shown to be changed in chronically stressed animals irrespective of the presence of anhedonia (33). Mice were subjected either to a ‘‘standard’’ stress procedure adapted from (33) or to a ‘‘weak’’ chronic stress regimen. Animals from a control group were not exposed to any stressors. Mice were weighed weekly in a course of experiment and tested for sucrose preference before, during, and after the application of stress procedure (for details of sucrose test protocol, see (34)). Three days after the end of stress procedures, mice were studied in the O-maze test. Experimental conditions were applied as described previously (35). Briefly, during the sucrose test, mice were given, for 8 h (between 9.00 and 17.00), a free access to two bottles, one with 1%-sucrose solution and the other with tap water. The consumption of water, sucrose solution, and total intake of liquids were estimated by weighing the bottles. The preference for sucrose was calculated as a percentage of consumed sucrose solution of the total amount of liquid drunk. The O-maze

Factors of Reproducibility of Anhedonia Induction

157

consisting of a circular path (runway width 5.5 cm, diameter 46 cm) was placed 20 cm above the floor. Two opposing arms were protected by walls (height 10 cm), illumination strength was 5 lux. Latency of the first exit to the open arms and total duration of time spent therein during a 5-min period were recorded. The ‘‘standard’’ 4-week chronic stress procedure included exposure to a rat, restraint stress, and tail suspension. The ‘‘weak’’ chronic stress consisted of the same type of stressors; they were applied for shorter time, frequency, or in less stressful version, as compared to the ‘‘standard’’ stress protocol. During the ‘‘standard’’ stress procedure, stressors were applied in the following sequence: days 1–7 – continuous exposure to a rat in small cages (housing of a mouse in a cage placed to a rat home cage); days 8–10 – 2-h restraint stress in plastic tubes; days 11–17 – continuous exposure to rat in small cages; 18–22 – continuous exposure to a rat in small cages and 40-min tail suspension; days 23–28 – continuous exposure to rat in small cages, 2-h restraint stress in plastic tubes, and 40-min tail suspension twice a day. With the ‘‘weak’’ stress procedure, stressors were applied in the following sequence: days 1–7 – continuous exposure to a rat by placing a mouse cage on a top of a rat cage; days 8–10 – 1-h restraint stress in plastic tubes; days 11–17 – continuous exposure to a rat by placing a mouse cage on a top of a rat cage; days 18–22 – 6-min tail suspension; days 23–28 – 1-h restraint stress in plastic tubes and 6-min tail suspension; days 29–36 – continuous exposure to a rat by placing a mouse cage on a top of a rat cage and 6min tail suspension; days 37–42 – continuous exposure to rat by placing a mouse cage on a top of a rat cage and 1-h restraint stress. The 4-week ‘‘standard’’ chronic stress procedure led to a significant decrease in sucrose preference in the stress group as compared to the non-stressed control animals. According to a previously validated 65%-sucrose preference criterion of anhedonia, 50% of stressed mice were classified as anhedonic (Fig. 9.1A). None of the animals subjected to the 6-week ‘‘weak’’ stress protocol could be assigned to the anhedonic group by employed criterion of anhedonia. Sucrose preference did not change in this group of mice at no time point of stress application (data not shown), including the day of termination of stress (Fig. 9.1A). Starting from the first week of stress, mean body weight was significantly lower in mice subjected to a ‘‘standard’’ stress protocol than in control animals; this difference persisted over the entire duration of the stress procedure (data not shown) and by its termination (Fig. 9.1B). The ‘‘weak’’ stress procedure did not induce significant changes in body weight in comparison to control group at none of the time points of the experiment.

158

Strekalova and Steinbusch

Fig. 9.1. Changes in sucrose preference, body weight and anxiety scores with ‘‘standard’’ and ‘‘weak’’ chronic stress protocols. (A, B) 4-week ‘‘standard’’ chronic stress regimen causes a significant drop of sucrose preference and body weight, as compared to control group (*p < 0.05 vs. control group). There were no such changes detected in mice subjected to a 6-week ‘‘weak’’ chronic stress. Bars indicate medians of the groups. (C, D) In both stressed groups of mice, latency of exit to open arms of the O-maze is significantly elevated, and time spent in open arms is significantly decreased in comparison to control group (*p < 0.05 and # p < 0.05). Data are expressed as mean – standard error of measurement (SEM).

Importantly, anhedonic and non-anhedonic mice from a group of animals subjected to a ‘‘standard’’ stress procedure did not differ in body weight (data not shown). Luck of this difference was observed in every single experiment (of a total number over 12 studies) performed with a 4-week stress procedures, similar to the one employed in the experiment, described here (12). Thus, obviously, stress-induced changes in sucrose preference cannot be attributed to the altered caloric needs of stressed mice (which may be diminished with body weight loss). Moreover, lasting increase in water intake and home-cage activity observed selectively in mice from the anhedonic group in our model led to

Factors of Reproducibility of Anhedonia Induction

159

suggest a rather enhanced sympathetic activation and elevated metabolic demands in these animals (12). Generally, changes in sucrose intake and preference do not correlate with alterations in body weight in our paradigm, neither with groups’ comparison in a course of stress, nor on the individual level. In our model, sucrose intake/preference, first, increase, then, decrease in a course of stress, while the mean body weight gradually declines throughout the stress experiment (33, 34). In previous studies, sucrose intake and preference were found to be not correlating with the body weight neither in control nor in stressed mice (33). Therefore, we strongly believe that in our chronic stress and sucrose test models, decreased sucrose preference reflects impaired sensitivity for reward. Significant elevation in the latency of exit to open arms and decrease in time spent in open arms were observed in mice submitted to either protocols of chronic stress. Thus, both ‘‘standard’’ and ‘‘weak’’ chronic stress exposures increase scores of anxiety-like behaviors in mice in applied experimental conditions (Fig. 9.1C,D). Increase in anxiety-related behavior validated stressful effects of both chronic stress procedures employed in the study. The fact that the values of the body weight and behavioral parameters of anxiety and locomotion are close in nonanhedonic and anhedonic mice suggests a similar impact of the stress procedure in these animals. Hence, it is very unlikely that the differences described above in sucrose test and other behaviors in the non-anhedonic and anhedonic groups can be due to a distinct ‘‘amount’’ of stress and general fatigue perceived by the animals, but suggests rather qualitative differences between non-anhedonic and anhedonic individuals. These data also led to consider the non-anhedonic mice to be an adequate control for the effects of chronic stress per se. Increased scores of anxiety-like behavior in chronically stressed C57BL/6 mice were reported in many experimental setups, including our model of stress-induced anhedonia (33) and recently proposed social defeat paradigm (26). In the two latter studies, elevated anxiety-like behavior was observed both in animals resilient and susceptible to a development of stress-induced anhedonia and other depressive-like features. That is, these behavioral alterations were occurring independently from a development of the depressive-like syndrome. Thus, with respect to induction of a depressive-like state, in previous and the present studies, elevated anxiety in mice submitted to the ‘‘weak’’ and ‘‘standard’’ chronic stress procedures can be regarded as ‘‘nonspecific’’ effect of chronic stress. In a different experiment, we employed rat exposure, restraint stress, exposure to sawdust from rat home cage, isolated island stress, and exposure to the audio recordings of owl singing, for 8 weeks in C57BL/6 mice (Strekalova, unpublished results).

160

Strekalova and Steinbusch

Similarly, in this study, we observed no significant changes in sucrose preference and body weight resulting from the stress application in mice, but we found enhanced scores of anxiety in the dark/light box and of horizontal activity in the open field. Because stress-induced hyperlocomotion in the open field was detected in a 4-week stress paradigm of a ‘‘standard’’-type stress procedure in stressed animals both with and without hedonic deficit (33), hyperactivity in chronically stressed mice can be regarded as another ‘‘non-specific’’ consequence of the chronic stress exposure, which is not associated with stress-induced anhedonia. Thus, prolonged stress procedures of various characteristics induce behavioral effects, which are unrelated to the depressive-like syndrome, as for instance, elevated anxiety and hyperlocomotion. Among various chronic stress paradigms, only those procedures, which led to a reduction in body weight, were effective in anhedonia induction. These results are in line with the data obtained in a model of social defeat stress. In this paradigm, a subgroup of stressed C57BL/6 mice, which were classified as susceptible to the development of depressive-like syndrome by a criterion of social avoidance, showed both anhedonia determined in sucrose preference test, and had decreased body weight. A subgroup of stressed animals, resilient to a development of deficits in social interaction and sensitivity to reward, showed no changes in body weight (26). Because a reduction in body weight depends on the impact of stress effects in animals, these data suggest that defined stress load is necessary for an induction of a hedonic deficit and depressive-like syndrome in C57BL/6 mice. Hence, identification of elevated anxiety and hyperlocomotion in chronically stressed mice does not imply a presence of anhedonia and depressive-like syndrome in C57BL6/N strain. Interestingly, our recent psychopharmacological studies additionally support a lack of relation between anhedonia, on one hand, and anxiety and hyperlocomotion, as ‘‘non-specific’’ consequences of stress, on another. Chronic application of the SSRI citalopram (15 mg/kg with drinking water) in C57BL6/N mice does counteract a development of stress-induced anhedonia, defined by a decreased sucrose preference, but not enhanced anxiety in the elevated O-maze nor hyperactivity in the open field test (Strekalova, unpublished results). Next, accumulated data with our model of stress-induced anhedonia suggest that for induction of a hedonic deficit in mice, stress procedure of sufficient load has to be applied for a certain period of time. With a proposed model of stress-induced anhedonia, this period varied between 2 and 3.5 weeks (33, 34). Altogether, the above-discussed data led to conclude that in mice, anhedonia and depressive-like syndrome can be only induced by a stress procedure of defined intensity and duration.

Factors of Reproducibility of Anhedonia Induction

161

Stress procedure of insufficient stress intensity and duration does not evoke a hedonic deficit, but leads to other behavioral alterations. Together, these data suggest that behavioral effects induced in chronic stress models should be interpreted with caution while attempting to model the depressive-like features in mice. Meanwhile, a number of chronic stress paradigms, in which, for example, anxiety-like features constitute the core behavioral changes induced by stress, are considered as valid animal models of depression. In humans, however, anxiety symptoms are well documented to be related to depressive features and considered as valuable prognostic parameters of dynamics of the disorder (36, 10). Apparently, a discrepancy in relationship between behavioral parallels in anxiety and depression symptoms in mice and humans delineate limitations in modeling depression in rodents and may reflect fundamental species-specific differences between humans and mice. 1.3. Characteristics of Employed Stressors and Their Application in a Chronic Procedure

Since unpredictable uncontrollable chronic stress, presumably, mimics the human life stress, one of the major pathogenic factors of depressive disorders, it represents itself as the only type of chronic stress procedures, which fits the construct validity criteria of animal models of depression (37, 1). With depression models in rodents, importance of stress uncontrollability and unpredictability, the two factors preventing the animals’ adaptation to stressors (11), is suggested by the above-discussed results, which point to a critical role of stress load in induction of hedonic state in mice. Repeated stress of limited duration of application used in our model might be of advantage over the procedures based on the unpredictable uncontrollable stress, as an application of the same stressor during several days in a row is likely to induce a state of learned helplessness, a pathogenetic factor of depression. Besides, in comparison to the models of unpredictable chronic stress, our stress paradigm with repeated stressors application might be closer to the modelling of life stress in human, which typically consists in recurrent repetition of several stress situations, rather than in daily experiencing of new stressors. Studies with chronic stress, in which an application of unpredictable uncontrollable stress was reported, employ the same stressors more than once, that is similar to the chronic stress characteristics of our model. Another important feature of the construct validity criteria of the depression models is the stressor uncontrollability, a factor of development of learned helplessness in animals (38, 39, 40). In our 4-week procedure, we use exclusively uncontrollable stressors. Generally, in contrast to previously elaborated chronic stress models, we aimed to establish an anhedonia-evoking stress regimen of such intensity and duration, upon which not all individuals would develop a hedonic deficit. Therefore, a stress load of our

162

Strekalova and Steinbusch

procedure is presumably lower than in other chronic stress models, which report 100% rate of anhedonia induction in a stress group (16, 21, 23, 22). Employment of the stressed non-anhedonic group as an internal control for the effects of chronic stress per se was expected to provide a possibility to re-evaluate the validity of the chronic stress depression paradigm in a more refined way. Elaboration of this approach was encouraged by numerous evidences of remarkable inter-individual variability in animals’ response to stress observed in our own and others’ studies (41–45). Besides a factor of uncontrollability, following principles were applied in selection of stressors to be used in our model. (1) Ethologically relevant stressors with emotional component are of preference over those with predominant physical stress impact (rat exposure and social defeat are considered to be such stressors; in our model they constitute a basis of the stress procedure). (2) Stressors, which interfere with sucrose test measurement, are to be avoided (e.g., food and water deprivation, high air temperature; these stressors affect total liquid consumption in mice). (3) Stressors, on which important behavioral assays are based, are to be avoided as well, to leave an option of using these assays in the characterization of the model (e.g., foot shock, which is applied in memory paradigms, is not employed in our model; this enabled the investigation of memory scores in fear conditioning and step down avoidance tasks in the model). (4) Stressors whose intensity can be easily assessed and regulated are of preference (e.g., it is difficult to estimate stress impact of continuous lighting and several other procedures, causing a discomfort and applied in most of the CMS paradigms; it is even more difficult to regulate their intensity when it is needed for achievement of optimal stress load). (5) Stressors which act equally on individual experimental animals are of high preference (social defeat therefore, is not very optimal stressor, as its impact is determined by individual scores of intruder’s aggressive behavior; rotating of mouse pairs can diminish this effect). 1.4. Inter-batch Variability in the Mouse Stress Response and Social Behavior: Implications for Reproducibility in the Chronic Stress Depression Paradigm

As it was observed with routinely used tests over a long period of time, laboratory animals display a significant inter-individual and inter-batch variability in various behavioral parameters and characteristics of stress response (12, 42–45). This issue represents itself a significant difficulty for fundamental and industrial research. In our experiments, we found a remarkable fluctuation in a sensitivity of male C57BL6/N mice to the tail suspension stress. At some period of observations, 6-min application of this stressor had high stress impact in these animals causing a lasting immediate after-stress hypoactivity. The same batches of mice were characterized by relative resistance to a restrain stress, which led a limited suppression of the after-stress activity in a home cage (33, 35, 12). Later, mice of

Factors of Reproducibility of Anhedonia Induction

163

the same strain were found to exhibit low sensitivity to a tail suspension and high reaction to a restraint stress procedure, accordingly to the scores of the after-stress hypoactivity. The variability in behavioral characteristics of stress-response, observed over a period of 6 years during our studies carried out in identical experimental conditions, was paralleled by changes in parameters of social behavior in tested batches of C57BL6/N mice. In this strain, male social behavior was shown to predict an individual susceptibility to stress-induced anhedonia in mice. Individuals with submissive social traits were found to be more vulnerable to stress-induced anhedonia (33, 34). In populations with initially low percentage of submissive animals (15–20%), all submissive mice develop anhedonia (Fig. 9.2A). In populations with high percentage of submissive individuals (over 50%), the percentage of non-submissive (dominant or aggressive) animals susceptible to anhedonia is significantly lower in comparison to a submissive cohort of mice (Fig. 9.2B). In a course of 4-week

Fig. 9.2. Variants of the 4-week chronic stress protocol of anhedonia induction. (A) Protocol of chronic stress comprised of a periodic 15 h rat-exposure, 2 h restraint and 6-min tail suspension stress (Strekalova et al., 2004). (B) A variant of chronic stress comprised of continuous a 24 h rat-exposure, water emergency combined with restraint stress, 6-min tail suspension stress and 1 or 2 h restraint stress (Strekalova et al., 2005). (C) A modification of the chronic stress procedure comprised of continuous 24-h and periodic 3–12 h rat-exposure, water emergency combined with a restraint stress, 20-min tail suspension stress and 2 h restraint stress (Strekalova et al., 2006).

164

Strekalova and Steinbusch

Factors of Reproducibility of Anhedonia Induction

165

stress, anhedonia was found to occur earlier in submissive animals. In one of the studies, at the time point of 3½ weeks of stress, 100% of submissive mice had sucrose preference below 65% and matched a taken criterion of anhedonia; only 16.6% of aggressive mice from the anhedonic group had anhedonia at this stage of stress (33). Interestingly, social characteristics of animal batches, such as a percentage of aggressive mice, were related to a behavioral pattern during stress, as was shown by differential changes in parameters of sucrose intake and preference, and total liquid intake in a course of stress exposure. As for instance, in populations with initially low percentage of submissive animals, statistically significant decrease in sucrose preference in the stressed group occurs at 3.5 weeks of stress time point. Stress exposure causes an increase in sucrose intake and preference at the beginning of the stress procedure and leads to a slight elevation of water intake. In experimental groups containing high percentage of submissive individuals (over 50%), the drop in sucrose preference is detected already after 2.5 weeks of stress. In theses animals, chronic stress induces, on its initial phase, an increase in sucrose intake and pronounced elevation of water consumption; sucrose preference is not affected (12). Together, one can hypothesize that the above-discussed interbatch fluctuations in response to particular stressors may be related to distinct characteristics of social behavior in tested mouse populations. Thus, the diversity in social traits of tested cohorts of mice may be a very likely source of unstable reproducibility with the chronic stress models in mice. Analysis of social behavior of C57BL/6 male mice determined in our studies in identical testing conditions evidenced a remarkable fluctuation in percentage of individuals with submissive social traits, which percentage varied over 6 years in a range from 15 to 85% (Fig. 9.3C). The variability in social behavior, which is related to the animal’s individual style of coping with environmental factors, can be considered as a biological factor of species’ adaptation and survival. It can be even of higher relevance when animals are placed in stressful conditions and, therefore, of evolutionary advantage in general. Moreover, the fact that a large inter-individual variability in biological parameters is commonly observed in genetically

Fig. 9.3. Social behaviour in male C57BL 6 N mice as a factor of predisposition to stress-induced anehdonia and variability with the model employment over the years. (A, B) Individual vulnerability to stress-induced anhedonia and initial social traits. In two sets of experiments: in 2001 (Strekalova et al., 2004) and 2005 (Strekalova et al., 2006), percentage of individuals with submissive, non-submissive and socially neutral initial social status was calculated from a total number of animals in anhedonic or non-anhedonic groups. In all experiments, percentage of non-submissive animals was significantly lower in anhedonic than in the non-anhedonic groups (Fisher’ exact test). (C) Fluctuations in percentage of individuals with submissive social behaviour in male C57 BL6 mice scored in 2000–2006. Under identical testing and housing conditions, significant fluctuations in percentage of male C57BL 6 mice defined as subdominant in a resident– intruder paradigm were observed over the years.

166

Strekalova and Steinbusch

homogeneous inbred strains of laboratory animals led to suggest that it is preserved by some mechanisms. Other sources of behavioral variability in the chronic stress experiments can be the environmental and seasonal factors (46, 47). Thus, the factor of inter-batch fluctuation of behavioral traits and stress–response suggests an importance of a control/moderation of the stress load with the chronic stress depression paradigms, so that the optimal conditions of stress, sufficient for induction of depressive-like syndrome and anhedonia, can be applied in each experimental cohort of animals. Adjustment of original protocol with defined stress intensity to the characteristics of habituation and stress response in different batches of C57BL/6 mice resulted in several procedural variations of the 4-week chronic stress model. This allowed to achieve a stable induction of hedonic deficit, despite a large diversity in behavioral stress response of mice from different batches observed over the years. With all variants of the 4-week chronic stress, significant reduction in body weight and sucrose preference was reliably induced in the stress group. Here, we present several examples of the 4-week chronic stress protocols (Fig. 9.2). We employed a tail suspension and a restraint stress of various durations, water emergency stress (as a variant of the restraint stress), and three variants of the rat-exposure stress (for experimental details, see (33–35, 12)). In one of the protocols, tail suspension was omitted from a stress procedure. As it was mentioned above, duration, frequency, and type of stressors were adjusted to specific characteristics of tested population. In summary, presented data suggest that in mice, stressinduced anhedonia can be induced with chronic stress procedures of certain intensity and duration. Chronic stress of insufficient intensity and/or duration can evoke a number of behavioral changes, which also accompany a development of depressive-like syndrome during more intense/long procedures, but not a hedonic deficit. Application of the two parameters, a reduction in body weight and immediate after-stress hypoactivity, as criteria of the optimal stress load in C57BL/6 mice, can allow the adjustment of a chronic stress procedure to the tight characteristics of anhedoniainducing stress regimen. Importance of a control/moderation of the stress load, either with above-proposed or other validated criteria is additionally suggested by the inter-batch fluctuation of behavioral traits and stress–response, generally described in laboratory mice. Together, our results show that adjustment of stress procedure to the characteristics of behavioral stress–response observed in a tested batch of animals, instead of using oncedefined stress protocol, could be a strategy to consider while aiming at higher reproducibility of chronic stress depression models in mice.

Factors of Reproducibility of Anhedonia Induction

167

2. Procedure and Variables 2.1. Animals and Housing

C57BL/6 male mice (age 3.5 months) are used. Starting from 10 to 14 days before the behavioral experiments, mice are housed single-caged under a reverse 12 h: 12 h light–dark cycle in standard laboratory conditions (22 – 1C; 55% humidity, food, and water ad libitum). Male CD1 mice, used as intruders in a social interaction test (see below) and in a social stress paradigm, are kept in the same conditions. Rats of any gender and age can be used for a rat exposure stress. In our studies, we used Wistar rats of both genders (age 2–5 months). During stress period, stress and control groups are housed separately. The group size recommended is 10 for control and 20–25 for stress groups.

2.2. General Conditions of Experiment

Since social status has earlier been found to be a significant factor in animals’ predisposition to stress-induced anhedonia (33), before the chronic stress procedure, parameters of social behavior in a social interaction test has to be determined (see below). Body weight and baseline preference to a 1%-sucrose solution (see sucrose test) are evaluated as well. Stress and control groups are balanced upon the animals’ body weight and initial sucrose preference. All control and stress groups are formed with a similar percentage of aggressive, non-aggressive, and socially neutral (non-definable) individuals (33, 34). Additionally to baseline measurements, sucrose consumption tests are performed after 2.5, 3.5, and 4 weeks of stress; by its end, mice are classified as anhedonic or non-anhedonic (see below). Animals are weighed weekly.

2.3. Resident–Intruder Test

The mice from experimental groups are placed individually as a resident in an observation cage (30 cm  60 cm  30 cm). Thereafter, a male CD1 mouse is introduced as an intruder to the same cage and left with the resident mouse for 4 min. During observation period, resident and intruder mice display either aggressive, submissive or neutral social exploratory behavior. Complete lack of attacks towards the partner, accompanied by specific ‘‘submissive’’ postures, escape, and defence is regarded as submissive type of social behavior. Initiation of attacks toward the partner and fighting back in response to attacks are categorized as non-submissive behavior. Some pairs of animals do not exhibit behavioral confrontation, but instead show no social interaction or social exploratory behavior; these mice are regarded as neutral for their social status.

168

Strekalova and Steinbusch

2.4. Chronic Stress Procedure

We propose the following chronic stress procedure as a variant of basic protocol, which supposively can ensure induction of anhedonic state in about 70% of mice. Stressors are applied in the following sequence: days 1–7 – continuous exposure to rat in small cages; days 8–10 – restraint stress in plastic tubes for 2 h and social defeat stress; days 11–17 – repeated exposure to rat in small containers; days 18–28 – repeated exposure to rat in small containers over-night and social defeat stress twice per day. When stress-induced changes in animals’ activity and body weight do not reach the criteria of optimal stress load (see below), supplementary stressors can be used in replacement of/in addition to a proposed scheme of chronic stress, in order to ensure a stress load, sufficient for anhedonia induction. For these proposes, tail suspension and water emergency can be applied. Replacement of repeatedly used stressors, to which mice show signs of habituation, with those, for which animals are naive, let to increase stress load and factor of unpredictability with the stress procedure. Exposure to rat in small cages: Mice are housed together with a rat. Therefore, they are introduced into small cages, which are then placed into a rat home cage. Customized metallic, mesh cage covers are used to protect the mouse from a rat. Exposure to rat in small containers: Mice are introduced into the transparent glass cylindrical containers and placed into the rat cage in increasing increments starting for 3–16 h. Restraint stress in plastic tube: Animals are placed inside plastic tubes (internal diameter is approximately 26 mm) during the dark phase of the light cycle. Small standard pieces of facial tissue are inserted in the tubes to restrict animals’ activity. Social defeat stress: Mice from stress group are placed inside of a home cage containing an aggressive CD1 mouse (as determined in a resident–intruder test) for 15–20 min. Animals are carefully observed in a course of test; in case of excessive aggressive attacks (detected in appoximately 5–10% of cases), the procedure is interrupted. Tail-suspension stress: Mice are submitted to the tail suspension procedure by hanging them by their tails for about 20 min daily or longer. The procedure is done during the dark phase of the animals’ light cycle. Water emergency stress: The protocol of this stress is based on previously described methods (48, 49, 50). Animals are placed inside plastic tubes (internal diameter 26 mm), in which they could move their bodies. These plastic containers are placed in a vertical position into empty tank (35 cm  22 cm  15 cm), which is then slowly filled with water (temperature 21C) for 10 min, so that their bodies are submerged in the water up to the sternum (half-way). Particular care has been taken in pouring water slowly into the container. Duration of exposure and water temperature can be regulated dependently of desirable stress load

Factors of Reproducibility of Anhedonia Induction

169

of this stress. During the very first stress session, filling with water lasted at least 5 min, the duration of the stress is 5 min. After the termination of the procedure, animals were carefully observed for 30 min. Since there was no hyperlocomotion or agitation detected during the observation period, and signs of habituation to the stress appeared after already two repeated sessions, this procedure is considered as a stress paradigm, free of traumatic components, in which stress impact is due to the physical cooling and temporary inability to escape the situation. 2.5. Sucrose Test

During this test, mice are given for 8 or 10 h, a free choice between two bottles, one with sucrose solution and another with tap water. The beginning of the test should start with the onset of the dark (active) phase of animals’ cycle. To prevent possible effects of side-preference in drinking behavior, the position of the bottles in the cage is switched in a mid-way of testing, i.e., after 4 or 5 h. No previous food or water deprivation was applied before the test. To minimize the spillage of liquids during sucrose test, bottles are filled in advance (during preceding day or evening) and kept in the up-side-down position for at least 12 h prior to testing. In order to balance the air temperature between the room and the drinking bottles, they have to be kept in the same room where the testing takes place. This measure prevents the physical effect of liquid leakage resulting from growing air temperature and pressure inside the bottles, when they are filled with liquids, which are cooler than the room air. With this method proposed, the error of measurement of liquid intake does not exceed 0.1 ml. In order to decrease variability in sucrose consumption during the very first sucrose test (baseline measurement), a day before, animals are allowed to drink 2.5%-sucrose solution in a one-bottle paradigm for 2 h during the dark phase of the day cycle. The1%-sucrose solution is used in tests performed during baseline and chronic stress application. The consumption of water, sucrose solution, and total intake of liquids is estimated simultaneously in the control and experimental groups by weighing the bottles. The preference for sucrose is calculated as a percentage of the consumed sucrose solution from the total amount of liquid drunk by the following formula: Sucrose Preference = V (Sucrose solution)/V (Sucrose solution) + V (Water)  100% A decrease in sucrose preference to a level at or below 65%, measured at the fourth week of continuous stress application, is taken as a criterion for anhedonia. This criterion was validated with previous behavioral, pharmacological, and other assays (12). With a criterion of 65%-sucrose preference, mice are assigned to the anhedonic or non-anhedonic groups.

170

Strekalova and Steinbusch

2.6. Control Over the Optimal Stress Load

In order to adjust stress load to specific characteristics of stress response of used batches of mice, it is necessary to monitor animals’ habituation to the chronic stress. Therefore, in a course of the experiment, weekly measured body weight has to be analyzed. Statistically significant reduction in body weight in stressed nontreated group in comparison to control mice is interpreted as an indicator of adequate stress load. To control habituation to single stress sessions during chronic stress on a daily basis, immediate post-stress locomotor activity of mice has to be evaluated; a decrease in these characteristics in more than 50% of mice for at least 30 min of a post-stress period detected by subjective rating is taken as a sign of optimal stress intensity. When changes in animals’ behavior and body weight do not correspond the above-described criteria, stress impact of applied procedures has to be increased.

2.7. Statistical Analysis

Data are analyzed with a statistical software package (Prizm 3, Chicago, IL). Independent data sets are analyzed by ANOVA, followed by the post-hoc Tukey test. Qualitative data are analyzed by the Fischer’s exact test (exact probability for contingency table). The level of confidence was set at 95% (p < 0.05).

3. Equipments, Materials, and Setup 1. Cages for rat exposure: Small cages for a mouse of approximate size 20 cm  8 cm  15 cm, fitting in the rat cage of approximate size 40 cm  25 cm  15 cm, covered with customized metallic, mesh cage tops are used. High-density mesh material is used for covers. It is important for good mechanical protection of a mouse from a rat. 2. Small containers: Customized transparent glass cylindrical containers of approximal size 15 cm  Ø 8 cm with small holes in metallic covers (Ø <0.5 cm), which can ensure protection of a mouse from a rat on one hand, and visual and odor contact, on another, are used. 3. Plastic tubes for restraint stress: Any plastic transparent tubes of internal diameter of approximately 26 mm, with holes for air can suit this test; e.g., falcon tubes of a 50-ml volume can be used for this procedure. Facial tissue of a size, which is sufficient for a restriction of animals’ body movement, is used. The size of the tissue determines a degree of restraint in this setup.

Factors of Reproducibility of Anhedonia Induction

171

4. Bottles for sucrose test: To provide unrestricted access to liquid consumption from one hand, and to minimize the spillage of liquids during the test, bottles with rubber stoppers and glass tips of internal diameter of 2 mm are used. Commercially available rubber stoppers of external diameter 26 and 20 mm with wholes for glass tips (Ø 6 mm) made out of odor-free material are recommended for use. Falcon tubes of a size of 50 ml are used as the bottles. Weighing of the bottles has to be done with balances, whose resolution is 0.1 g. 5. Preparation of sucrose solution: Regular sugar stored in paper bags well protected from sources of flavors can be used for a sucrose test. It is recommended to avoid any contact of sugar used in the experiment with plastic material in order to prevent any transfer of plastic odor, which is aversive for mice. It is recommended to prepare sucrose solution with a fresh tap water approximately 12 h prior the test and place the filled bottles in an upside-down position in a room, in which the sucrose test will take place. This will ensure a temperature balance between the solutions in the bottles and air in the testing room and prevent spillage of the bottles. 6. Manipulation of bottles during sucrose test: Bottles are supposed to be placed gently in the mouse cage, a position close to a horizontal is recommended. The placement of the bottles should be finished with the onset of the dark phase of the animals’ light cycle. Position of the bottles has to be changed in the midway of testing. Testing room has to be protected from noise. Control and stress groups of mice have to be tested in the identical temperature/humidity conditions.

4. Anticipated Results A 4-week chronic stress paradigm, which can be applied as proposed here and other procedural variations (see above), results in a decrease in sucrose preference  65% in 50–70% of mice (12). Before the onset of stress, all three groups, control, non-anhedonic, and anhedonic mice, have a similar preference for sucrose over water and absolute intake of water and sucrose solution. Figure 9.4 represents changes in sucrose preference in one of the typical chronic stress experiments. Based on the chosen criterion of 65%-sucrose preference, mice are assigned to the anhedonic and non-anhedonic groups. A selection of a criterion for anhedonia is determined above in the text. After stress, mice from the anhdonic group are characterized by significantly decreased consumption of sucrose solution, whereas in non-anhedonic animals, these parameters are not different from the control values (data not shown).

172

Strekalova and Steinbusch

Fig. 9.4. Chronic stress leads to a decrease in sucrose preference in a subgroup of mice. Individual data show that chronic stress causes a drop of sucrose preference, measured after the termination of a 4-week stress, in a subgroup of mice. Stressed mice split into anhedonic and non-anhedonic subgroups, according to the criterion of 65% preference for sucrose solution (see the text). Bars indicate medians of the groups.

5. Troubleshooting Animals display a baseline sucrose preference below 65%: It is not uncommon that in mice, the variability in sucrose values assessed with the very first test (baseline measurement) is high and some individuals show a sucrose preference below a criterion of anhedonia, used at the end of stress procedure, after several sucrose test assays. This phenomenon is typical for mice and though to be due to neophobia, a feature, characteristic for mice (22). We recommend ensuring that not more than 10% of the population is characterized by sucrose preference below 65% in a baseline conditions. If such individuals are detected with a baseline test, then they have to be assigned to a control group. If the percentage of mice showing low-baseline sucrose preference is higher than 10 and lower than 30%, then we recommend to repeat the sucrose test on the next day after a baseline measurement solely in mice with low sucrose preference. If this percentage is about 50%, then our recommendation is to repeat the sucrose test in all animals. When low sucrose preference is detected in over 50% of tested mice, it is necessary to check for a source of confounding factors, which are related to stress (vibration, noise, etc.), interfere with measurement of drinking behavior (leakage of the bottles), drinking itself (presence of foreign orders on the bottles or in drinking solutions, such as soap, rubber etc.), and taste of sucrose (stock of sugar is stored next to a source of flavors, e.g., plastic). When no obvious sources of artifacts are detected, it is worthwhile to change the stock of sucrose used in the test.

Factors of Reproducibility of Anhedonia Induction

173

Animals display mean sucrose preference above 80%: In case of increased sensitivity to sucrose solution (because of seasonal factors, repeated access to sucrose, specific characteristics of the batch of animals used), mean sucrose preference in the group may exceed 80%. This decreases a sensitivity of the method due to ceiling effect. Therefore, sucrose solution of lower concentration has to be used in subsequent tests (e.g., 0.8% instead of 1%). Experimental data then have to be normalized to a control values. Low values of absolute amounts of liquid intake in sucrose test: Normally, adult C57 mice show mean values of total liquid consumption of about 3–3.5 ml. Mean values of a total liquid intake below 2 ml, measured in mice over 8–12 h period, are very likely to point to a presence of confounding factors, such as stressors in a laboratory environment. Inhibition of drinking behavior does affect the evaluation of a sucrose test and therefore has to be prevented. Another reason of low values of the liquid drunk could be too tight tips of drinking bottles. High values of absolute amounts of liquid intake in sucrose test. Mean values of total intake of liquid above 5 ml measured over 8–12 h indicate that the testing conditions are suboptimal; this can be due to the leakage of the drinking bottles and presence of stressors in a laboratory environment. Proper experimental conditions, which exclude this abnormality, have to be provided. Bottles have to be checked for leakage by placing them in a cage in the absence of a mouse. Lack of significant changes in body weight and after-stress activity in stressed group: Stress impact of applied procedures has to be increased, if changes in animals’ behavior and body weight do not correspond to the described criteria of the optimal stress load. Following methods can be applied: (1) prolongation of rat exposure for 2–4 additional hours; (2) employment of more stressful version of the stress procedure (exposure to a rat in small cages can be replaced by rat exposure in small containers); and (3) application of additional stress session during a day time. Too profound decrease in immediate after-stress activity: This effect, most likely, indicates high sensitivity of animals to a particular stressor used. Stress load has to be reduced by skipping a stress session and following reduction in intensity of the stressors using methods opposite to those described in the above paragraph.

Acknowledgments We would like to thank Profs. Raymond Cespuglio and Peter Gruss, Drs. Oleg Dolgov, Barbara Maier, Careen Schroeter, and Martti Valilla for their contribution. Material used in the

174

Strekalova and Steinbusch

introduction and Figs. 9.2 and 9.3 (adapted from Strekalova T. Optimization of the chronic stress depression model in C57 BL/6 mice: evidences for improved validity. 2008 In: ‘‘Behavioral models in stress research. Volume I’’. Eds. Kalueff A. and LaPorte J. pp. 111–157) is reproduced accordingly to the permission of Nova Science Publishers, NY, USA. Material shown in Fig. 9.4 is presented as significantly modified graphic representation of the data, partly published by Lippincott Williams & Wilkins, Behavioral Pharmacology, Strekalova T, Gorenkova N, Schunk E, Dolgov O, Bartsch D. ‘‘Selective effects of citalopram in the mouse model of stress-induced anhedonia with control effects for chronic stress’’. 2006;17:271–287. References 1. Auriacombe M, Reneric JP, Le Moal M. Animal models of anhedonia. Psychopharmacol 1997;134:337–48. 2. McArthur R, Borsini F. Animal models of depression in drug discovery: a historical perspective. Pharmacol Biochem Behav 2006;84:436–52. 3. Newport DJ, Stowe ZN, Nemeroff CB. Parental depression: animal models of an adverse life event. Am J Psychiatry 2002;159:1265–83. 4. Kessler RC, Akiskal HS, Ames M, et al. Considering the costs of bipolar depression. Behav Health 2007;27:45–7. 5. Willner P. Chronic mild stress (CMS) revisited: consistency and behavioural-neurobiological concordance in the effects of CMS. Neuropsych 2005;52:90–110. 6. Van den Hove D, Blanco C, Aendekerk B, et al. Prenatal restraint stress and long-term affective consequences. Dev Neurosci 2005;27:313–20. 7. Michelsen KA, van den Hove DL, Schmitz C, et al. Prenatal stress and subsequent exposure to chronic mild stress influence dendritic spine density and morphology in the rat medial prefrontal cortex. BMC Neurosci 2007;8:107. 8. Hamilton M. Development of a rating scale for primary depressive illness. Br J Soc Clin Psychol 1967;6: 278–96. 9. Klein DF. Endogenomorphic depression. A conceptual and terminological revision. Arch Gen Psychiatry 1974;31:447–54. 10. Kessler RC, Chiu WT, Demler O, et al. Prevalence, severity and comorbidity of 12month DSM-IV disorders in the National Comorbidity Survey Replication. Arch Gen Psychiatry 2005;62:617–27.

11. Anisman H and Matheson K. Stress, depression and anhedonia: caveats concerning animal models. Neurosci Biochem Rev 2005;29:525–36. 12. Strekalova T. Optimization of the chronic stress depression model in C57 BL/6 mice: evidences for improved validity. In: ‘‘Behavioral models in stress research’’. Eds. Kalueff A. & LaPorte J. Nova Science Publishers, NY, USA. 2008;111–57. 13. Katz RJ. Animal models and human depressive disorders. Neurosci Biobehav Rev 1981;5:231–46. 14. Katz RJ. Animal model of depression: pharmacological sensitivity of a hedonic deficit. Pharmacol Biochem Behav 1982;16: 965–8. 15. Willner P, Towell A, Sampson D et al. Reduction of sucrose preference by chronic unpredictable mild stress, and its restoration by a tricyclic antidepressant. Psychopharmacology (Berl) 1987;93:358–64. 16. Willner P. Validity, reliability and utility of the chronic mild stress model of depression: a 10-year review and evaluation. Psychopharmacology 1997;134:319–29. 17. Moreau JL, Jenck F, Martin JR, et al. Antidepressant treatment prevents chronic unpredictable mild stress-induced anhedonia as assessed by ventral tegmentum selfstimulation behavior in rats. Eur Neuropsychopharm 1992;2: 43–9. 18. Pucilowski O, Overstreet DH, Rezvani AH, et al. Chronic mild stress-induced anhedonia: greater effect in a genetic rat model of depression. Physiol Behav 1993;54:1215–20. 19. Papp M, Nalepa I, Vetulani J. Reversal by imipramine on serotonergic and beta-

Factors of Reproducibility of Anhedonia Induction

20.

21.

22.

23.

24.

25.

26.

27.

28.

29.

30. 31.

adrenergic receptor binding in a chronic mild stress model of depression. Eur J Pharmacol 1994;261:141–7. Von Frijtag JC, Reijmers LG, Van der Harst JE, et al. Defeat followed by individual housing results in long-term impaired reward- and cognition-related behaviours in rats. Behav Brain Res 2000;117:137–46. Harkin A, Houlihan DD, Kelly JP. Reduction in preference for saccharin by repeated unpredictable stress in mice and its prevention by imipramine. J Psychopharm 2002;16:115–23. Ducottet C and Belzung C. Correlations between behaviours in the elevated plusmaze and sensitivity to unpredictable subchronic mild stress: evidence from inbred strains of mice. Behav Brain Res 2005;156:153–62. Gronli J, Murison R, Fiske E, et al. Effects of chronic mild stress on sexual behavior, locomotor activity and consumption of sucrose and saccharine solutions. Physiol Behav 2005;84:571–7. Baker SL, Kentner AC, Konkle AT, et al. Behavioral and physiological effects of chronic mild stress in female rats. Physiol Behav 2006;87:314–22. Jayatissa MN, Bisgaard C, Tingstrom A, et al. Hippocampal cytogenesis correlates to escitalopram-mediated recovery in a chronic mild stress rat model of depression. Neuropsychopharm 2006;31:2395–404. Krishnan V, Han MH, Graham DL, et al. Molecular adaptations underlying susceptibility and resistance to social defeat in brain reward regions. Cell 2007;131:391–404. Gould TD and Gottesman II. Psychiatric endophenotypes and the development of valid animal models. Genes Brain Behav 2006;2:113–9. Jacobson LH and Cryan JF. Feeling strained? Influence of genetic background on depression-related behavior in mice: a review. Behav Genet. 2007;37:171–213. Weiss JM and Simson PG. Depression in an animal model: focus on the locus ceruleus. Ciba Found Symp 1986;123:191–215. Cabib S. What is mild in mild stress? Psychopharmacology (Berl) 1997;134:344–46. Harris RB, Zhou J, Youngblood BD, et al. Failure to change exploration or saccharin preference in rats exposed to chronic mild stress. Physiol Behav 1997;63:91–100.

175

32. Nestler EJ, Gould E, Manji H, et al. Preclinical models: Status of basic research in depression. Biol Psychiatry 2002;52:503–8. 33. Strekalova T, Spanagel R, Bartsch D, et al. Stress-induced anhedonia in mice is associated with deficits in forced swimming and exploration. Neuropsychopharm 2004;11:2007–17. 34. Strekalova T, Gorenkova N, Schunk E, et al. Selective effects of citalopram in the mouse model of stress-induced anhedonia with control effects for chronic stress. Behav Pharm 2006;17:271–87. 35. Strekalova T, Spanagel R, Dolgov O, et al. Stress-induced hyperlocomotion as a confounding factor in anxiety and depression models in mice. Behav Pharm 2005; 16:171–80. 36. Freeman MP, Freeman SA, McElroy SL. The comorbidity of bipolar and anxiety disorders: prevalence, psychobiology, and treatment issues. J Affect Disord 2002;68:1–23. 37. Cryan JF, Markou A, Lucki I. Assessing antidepressant activity in rodents: recent developments and future needs. Trends Pharmacol Sci 2002;23:238–45. 38. De Boer SF, Van der Gugten J, Slangen JL. Pharmacol Biochem Behav 1991;1:13–9. Effects of chlordiazepoxide, flumazenil and DMCM on plasma catecholamine and corticosterone concentrations in rats. 39. Meerlo P, Sgoifo A, De Boer SF, et al. Longlasting consequences of a social conflict in rats: behavior during the interaction predicts subsequent changes in daily rhythms of heart rate, temperature, and activity. Behav Neurosci 1999;6:1283–90. 40. El Yacoubi M, Vaugeois JM. Genetic rodent models of depression. Curr Opin Pharmacol 2007;7:3–7. 41. Strekalova TV. The characteristics of the defensive behavior of rats in accordance with their resistance to emotional stress. Zh Vyssh Nerv Deiat Im I P Pavlova 1995;45:420–422. 42. Holmes A, le Guisquet AM, Vogel E, et al. Early life genetic, epigenetic and environmental factors shaping emotionality in rodents Neurosci Biobehav Rev 2005;29:1335–1346. 43. Hyman SE. How mice cope with stressful social situations. Cell 2007;131:232–4. 44. Macrı` S, Pasquali P, Bonsignore LT, et al. Moderate neonatal stress decreases withingroup variation in behavioral, immune and

176

Strekalova and Steinbusch

HPA responses in adult mice. PLoS ONE 2007;2:e1015. 45. Jakovcevski M, Schachner M, Morellini F. Individual variability in the stress response of C57BL/6J male mice correlates with trait anxiety. Genes Brain Behav 2008;2:35–43. 46. de la Iglesia HO, Meyer J, Schwartz WJ. Using Per gene expression to search for photoperiodic oscillators in the hamster suprachiasmatic nucleus.Brain Res Mol Brain Res 2004;127:121–7. 47. Bozinovic F, Mun ˜ oz JL, Cruz-Neto AP. Intraspecific variability in the basal metabolic rate: testing the food habits hypothesis. Physiol Biochem Zool 2007;80:452–60.

48. Retana-Marquez S, Bonilla-Jaime H, Vazquez-Palacios G, et al. Changes in masculine sexual behavior, corticosterone and testosterone in response to acute and chronic stress in male rats. Horm Behav 2003; 44:327–37. 49. Klenerova V, Jurcovicova J, Kaminsky O, et al. Combined restraint and cold stress in rats: effects on memory processing in passive avoidance task and on plasma levels of ACTH and corticosterone. Behav Brain Res 2003;142:143–9. 50. Qui BS, Mei QB, Liu L, Tchou-Wong KM. Effects of nitric oxide on gastric ulceration induced by nicotine and cold-restraint stress. World J Gastroenterol 2004;10:594–7.

Chapter 10 Learned Helplessness in Mice Hymie Anisman and Zul Merali Abstract Exposure to inescapable shock provokes behavioral disturbances in subsequent shock-escape tests, as well as in other behavioral paradigms, including those that reflect anhedonia. The interference induced by inescapable shock using a yoked (triadic) paradigm has frequently been referred to as a ‘‘learned helplessness’’ paradigm. The interference effect, although attributed to cognitive factors by several investigators, has also been explained on the basis of neurochemical changes induced by the uncontrollable stressor. In the present report, we briefly describe the various theoretical positions concerning the interference effect induced by inescapable hock, describe procedures that can be used to investigate this phenomenon in mice, and provide caveats that might be considered in conducting these experiments. Key words: Coping, depression, helplessness, triadic paradigm, uncontrollable stressor.

1. Learned Helplessness in Mice

Stressful events likely contribute to the emergence of several psychological and physical disturbances, including anxiety, posttraumatic stress disorder (PTSD) and depression, through to cardiovascular and immunologically related illnesses and even those that involve neurodegenerative processes (1). Although there are considerable data from human patients supporting this view, it is difficult to deduce on the basis of studies in humans what biological mechanisms might underlie the processes by which stressors come to engender or exacerbate these disorders. In contrast, animal studies permit the identification of specific brain neurochemical changes and hormonal variations that are associated with stressful events, and in conjunction with studies that involve pharmacological manipulations or those that involve genetic engineering, it is possible to determine whether these

T. D. Gould (ed.), Mood and Anxiety Related Phenotypes in Mice, Neuromethods 42, DOI 10.1007/978-1-60761-303-9_10, ª Humana Press, a part of Springer ScienceþBusiness Media, LLC 2009

177

178

Anisman and Merali

mechanisms might play a provocative role in pathological outcomes. Of course, animal studies are limited in several respects. Not only do they not provide the richness of data obtained from human studies, but also one is never certain whether the symptom profile exhibited by humans are actually recapitulated in animals. Moreover, the influence of some very powerful stressors in humans (e.g., shame and humiliation, guilt, grief) cannot be deduced in animals, and the influence of important coping factors, such as social support, and the influence of unsupport (the latter referring to inappropriate or inadequate support, or not receiving support when it was reasonably expected) cannot be assessed in rodent models (2).

2. Background and Historical Overview One of the best known phenomena in stress research is the ‘‘learned helplessness’’ effect, which essentially demonstrated that stressors could provoke behavioral disturbances in animals, and that the animal’s ability to exert control over the stressor’s termination was fundamental in promoting these effects (3, 4). In the initial studies, it was shown that dogs that had been exposed to an escapable stress, or that had not been stressed, subsequently exhibited proficient performance in a test where they could escape from a stressor (footshock). However, dogs that had received an identical amount of uncontrollable shock as animals in the original escape condition (applied in a yoked paradigm) later exhibited profound behavioral impairments in a shock-escape test where an active response would have terminated the stressor. These dogs apparently did not make overt attempts to avoid or escape the footshock, but instead seemed to passively accept the stressor. In fact, when an animal made a successful escape response, this was not predictive of further successful escape responses. However, dogs initially trained to emit a shuttle response were immunized against the effects of later uncontrollable shock (5). Based on these data and related findings, it was proposed that dogs exposed to a series of uncontrollable, inescapable shock presentations ‘‘learned’’ to be helpless. Essentially, these animals learned that their responses and outcomes were independent of one another (‘‘nothing I do matters’’) and thus gave up further attempts to escape from the stressor. In their review, Maier and Watkins (6) make it clear that the original studies were not focused on modeling human pathologies, such as depression, but rather addressed the issue of whether and why controllable and uncontrollable events differentially

Learned Helplessness in Mice

179

influenced later behavioral outcomes. In fact, researchers frequently confuse the ‘‘learned helplessness paradigm’’ (i.e., the escape deficits introduced by inescapable shock) with the theoretical construct of learned helplessness (i.e., the cognitive changes that are hypothesized to lead to behavioral disturbances). This misconception is curious, as the presence of a behavioral interference in a shuttle-escape test can hardly be taken to mean that the animal ‘‘feels’’ more helpless, and elimination of the interference likewise does not suggest attenuation of cognitive disturbances. In fact, as will be seen, alternative explanations for the interference effect have been offered that do not resort to cognitive interpretations. Following the introduction of the learned helplessness paradigm, it was several years before it was suggested that humans who experience uncontrollable stressful events develop feelings of helplessness and hence ‘‘fall’’ into depression (7). By the mid-1970 s, several papers had been published showing that nondepressed university students exposed to unsolvable problems subsequently displayed impaired performance in a problem solving task, much like that evident in depressed college students who had not been exposed to the unsolvable test (8). It appeared that the incongruity between the participants expectancy regarding their performance and their failure to meet this expectancy was fundamental in provoking the impaired performance (9). Perhaps because depressed patients often describe themselves as feeling helpless and their situation as hopeless (10), the learned helplessness model was widely adopted in the psychological and psychiatric literature as being fundamental to the emergence of depression (11). The possibility that feelings of helplessness were related to depression, but not necessarily a causal agent in this regard, seemed to receive only very limited attention (12). To be sure, it is certainly possible that stressful events of an uncontrollable nature could lead to helplessness and hence promote depressive illness, but it is equally likely that feelings of helplessness represent one of the symptoms associated with depression or that that helplessness was more readily induced by stressors among depressed individuals. Although the data derived from studies in dogs were extended to college students and to depressed patients, there seemed to be a lengthy period before research on learned helplessness was conducted using rodents in an effort to determine the biological changes that accompany the interference effect (although there were several studies that assessed avoidance, rather than escape deficits in rats). This was likely due to an interference effect not being readily induced in rodents that had been exposed to uncontrollable shock, certainly not as readily as such an effect could be induced in dogs. However, if the operant required of the animal entailed a motor response that was relatively difficult to

180

Anisman and Merali

accomplish or where an active response had to be maintained for several seconds, then performance deficits could be provoked. In this regard, Seligman and Beagley (13) and Maier et al. (reviewed in (6)) demonstrated escape disturbances in several motorically demanding tests following inescapable shock. For example, disturbances were evident in a task in which successful escape entailed running from one compartment of a shuttle box to a second compartment and then back again (this is termed an FR-2 procedure) or having rats make repeated lever press response (FR-3) to escape, or in a test that required animals jumping onto a high platform. Likewise, in the tests conducted by Weiss and Glazer (14), the hurdle separating the two compartments in the shuttle box were relatively high, and hence reflexive running responses did not get the animal out easily. Finally, the procedure we used entailed briefly preventing the escape response (by keeping the gate separating the two compartments closed), thereby requiring animals to sustain active responding for a somewhat longer period (15, 16). 2.1. Does Performance in an Escape Test Following Exposure to a Learned Helplessness Paradigm Actually Represent the Induction of Helplessness and Does This Reflect Depressive Illness?

As indicated by Maier and Watkins (6), the learned helplessness paradigm has been used to evaluate the effects of an uncontrollable stressor on later performance, and was not initially developed to model depression. As the behavioral disturbances in the shuttle-escape test are modifiable by repeated administration of antidepressants (17, 18), it has frequently been assumed that performance in a shuttle test after exposure to inescapable shock actually represents a behavior that might reflect depression. Yet, it has also been shown that manipulations that increase the animals’ ability to maintain an active response predictably influenced expression of the interference effect (56). Moreover, drugs that are not typically thought of as antidepressants (e.g., the anticholinergic, scopolamine, as well as dopamine precursor, l-DOPA) also attenuated the effects of inescapable footshock, whereas treatments that reduced NE and/or DA functioning mimicked the behavioral effects of inescapable shock (16, 19, 20). It has since been shown that scopolamine may engender potent antidepressant effects (21), but comparable data for l-DOPA have not been reported (although the DA agonist pramipexole was found to diminish the symptoms of bipolar II disorder (22)). Together, these data suggest that performance in a shuttleescape test after inescapable footshock may well be an effective screen for antidepressant agents, just as the forced-swim test is useful in this regard (23). However, it is less certain whether the learned helplessness paradigm and testing animals in a shockescape test can actually be used to model depressive illness or feelings of helplessness. This should not be misconstrued to suggest that uncontrollable stressors do not lead to pathological

Learned Helplessness in Mice

181

outcomes, but rather that the shuttle-escape (or lever press) test may simply not be the best one in which to model depression. As will be discussed later, there are alternatives. Several explanations for the behavioral interference were offered as alternatives to the learned helplessness hypothesis. According to one of these, when animals are exposed to shocks of long duration, but moderate intensity, they might learn ‘‘unauthorized responses’’ that eliminate or diminish the aversiveness of the stressor (24, 25). Similarly, because shock-elicited activity almost invariably declines during a given shock trial (15), animals may be adventitiously reinforced for immobility. These responses may be inconsistent with those necessary for successful escape, and hence will have long-lasting disruptive effects on performance. The second explanation for the interference effect comes from studies showing that stressful events provoke several neurochemical changes, some of which are aligned with the processes that are thought to subserve depression. It was suggested that when animals are confronted by a stressor, behavioral responses are made to contend with this challenge. Concurrently, a constellation of neurochemical changes is provoked (e.g., elevated utilization of norepinephrine, dopamine and serotonin, as well as hypothalamic–pituitary–adrenal activation) that facilitate adaptive behavioral responses, or are essential to maintain the integrity of biological processes necessary for survival. When behavioral methods of contending with the stressor are unavailable, then the burden of coping rests more fully on neurochemical processes as reflected by still greater monoamine functioning. Among other things, however, under these conditions amine utilization may be sufficiently great so that the synthesis is exceeded, leading to a decline in amine levels, thereby rendering the animal less able to deal with the stressor (26, 27). Since the time of this formulation, other factors have been found to vary in response to stressors, and some of these have been implicated in the interference effect as well as in depressive illness. For instance, it is well established that stressors increase the release of corticotropin releasing hormone (CRH) from the paraventricular nucleus of the hypothalamus and influence CRH receptors within limbic brain regions (1). Moreover, consistent with the suggestion that CRH, together with serotonin (5-HT) might be involved in depressive disorder (28, 29), it was shown that injection of CRH into the dorsal raphe nucleus (DRN), the site of 5HT cell bodies of neurons that project to the frontal cortex, provoked an effect akin to the interference effect produced by inescapable footshock. Conversely, a CRH2 receptor antagonist attenuated the effects ordinarily induced by inescapable footshock (6, 30, 31), although it did not seem that CRH1 manipulations influenced performance appreciably. Yet, given that CRH1 has

182

Anisman and Merali

more commonly been associated with anxiety and depressive disorder (1), and depression has been associated with stressor controllability, it is difficult to understand the absence of CRH1 effects (and the presence of effects of CRH2) in the context of the helplessness effect. However, it may be significant that a sexually dichotomous anxiety response was reported in (CRH1 and CRH2) double knockout mice, and that this effect was most apparent in females (32). In addition to the potential of CRH involvement, there have been reports implicating stressors-provoked changes of growth factor, including that of brain-derived neurotrophic factor (BDNF), in the provocation of depression (33, 34). Consistent with this, several investigators demonstrated that BDNF manipulations influenced the behavioral interference in the learned helplessness paradigm. For instance, it was reported that BDNF administration to the hippocampus attenuated the interference introduced by inescapable shock (35, 36), and the interference effect provoked by inescapable shock was augmented in BDNFknockout mice (37). Likewise, BDNF (or yet another growth factor, neurotrophin-3) administered directly to the dorsal raphe attenuated the performance deficit in a learned helplessness paradigm (38). However, others have indicated that the BDNF changes associated with a stressor were unrelated to stressor controllability (39, 40), suggesting that although BDNF reductions might be instigated by a stressor, the conditions necessary to induce the interference of performance were not directly related to the BDNF changes. Interestingly, however, it was found that changes of another growth factor, namely fibroblast growth factors (FGF2), were dependent on stressor controllability, making this a viable mediator of the uncontrollable stressor effects (39). 2.2. Sensitization of Neuronal Processes Influence Long-Term Effects of Uncontrollable Stressors

The monoamine changes introduced by uncontrollable stressors are evident for a limited time, typically lasting for a matter of hours. Yet, the behavioral disturbances evident in some paradigms are fairly persistent lasting for weeks after the stressor experience. To accommodate this ‘‘apparent’’ dissociation, Anisman and Sklar (41) suggested that the processes governing the neurochemical disturbances associated with a stressor are subject to sensitization, so that upon subsequent re-exposure to the stressor, the neurochemical alterations are readily reestablished, thus leading to the prolonged behavioral disturbances. Indeed, in the behavioral paradigm used by Anisman (i.e., delaying the availability of an escape response), the interference effect is typically minimal over the first few trials and becomes progressively more pronounced, possibly reflecting the re-induction of the neurochemical changes. Likewise, in the paradigm used by Maier et al. (i.e., the FR-2 procedure), during the initial five trials, rats are tested in a simple FR-1 paradigm (one passage across the shuttle apparatus will result in

Learned Helplessness in Mice

183

shock termination), after which the FR-2 requirement is introduced (42). The initial five trials may be sufficient to activate the neurochemical changes that favor impaired performance, which then emerge on the FR-2 trials. Sensitization-like effects are not limited to monoamines, but have been observed with respect to several neurobiological processes, and have been implicated in not only depressive illness but also reinstatement of drug self administration (43). For instance, Tilders and Schmidt (44) reported that with the passage of time following a stressor, phenotypic changes occur with the terminal regions of paraventricular nucleus neurons (located at the external zone of the median eminence). These neurons primarily contain CRH, but following a stressor experiences, the co-expression of CRH and arginine–vasopressin (AVP) increases in a time-dependent fashion. As CRH and AVP synergistically stimulate the release of ACTH from the anterior pituitary, it was suggested that such a mechanism was responsible for increased HPA functioning evident upon later stressor re-exposure. Thus, the long-term behavioral effects of an uncontrollable stressor experiences may stem from long-lasting changes in the characteristics of the neurons themselves. Since the introduction of the proposition that sensitized neurochemical changes may be important in promoting behavioral disturbances in animal models, this phenomenon has been applied to the development of depression in humans (45). In this regard, it seems that with successive stressor experiences (or with repeated bouts of depression), the degree of distress necessary to promote a further episode declines, so that ultimately even minor events may be sufficient to precipitate a depressive episode (45, 46). Moreover, it has been suggested that sensitization processes might also contribute (acting as a risk factor) for the frequent recurrence of depressive illness (47). 2.3. Chronic Stressor Exposure

If, as argued by Maier and Seligman, the interference induced by inescapable shock was due to learned helplessness, then it would be expected that repeated sessions of inescapable shock would promote more profound effects than a single session of this treatment. Contrary to this fundamental view, however, it was reported that with repeated stressor exposure, the interference effect was less pronounced than after a single stressor session, and this was paralleled by the NE reduction associated with an acute stressor no longer being evident (48). Subsequent studies have likewise shown that with repeated stressor exposure, an adaptation-like effect develops with respect to 5-HT levels. In this regard, it was suggested that the increased amine synthesis and utilization elicited by an acute stressor ordinarily wanes when the stressor terminates. Likewise, the increased amine utilization also diminishes with the termination of a chronic stressor; however, with repeated

184

Anisman and Merali

insults, the enhanced synthesis persists for a more protracted period following stressor offset. As a result, amine levels are elevated, presumably as an adaptive so that animals might be prepared to deal with impending stressors (26). Although an adaptation-like effect may develop with a chronic stressor, this should not be misinterpreted as suggesting that all is well. Indeed, it has been suggested that with chronic stressor experiences, the demands placed on adaptive biological systems may become excessively taxed, and ultimately the effectiveness of several neurobiological systems may be compromised (allostatic overload). Such an overload of biological systems may culminate in pathological outcomes, including those related to mood states, neurodegenerative processes, immune-related pathologies, diabetes, and cardiovascular disease (1, 49). 2.4. Cross-Situational Effects of Uncontrollable Stressors

In view of the wide array of neurobiological effects of stressors, it should come as no surprise that the effects of inescapable shock are not restricted to shock-escape performance. In fact, controllable and uncontrollable stressors have been found to differentially influence performance in other aversive situations, such as water-escape tasks and in a Porsolt forced-swim test (50) or in water-escape tests (27, 51, 52). Interestingly, when animals are assessed in a forced-swim test soon after an uncontrollable stressor, increased swimming may be evident (depending on the severity of the stressor), possibly reflecting arousal stemming from the stressor experience. However, at longer intervals (e.g., 24 h), when the initial arousal/anxiety has subsided, animals exhibit reduced active efforts and instead display increased floating (53). It has also been shown that uncontrollable aversive events can influence the acquisition of response contingencies in appetitive paradigms (54), novelty reward (55), and in responding for rewarding brain stimulation (56). Thus, although shuttle-escape testing following inescapable shock may be of limited usefulness in modeling depression, the fact that behavioral disturbances are evident across a variety of situations following exposure to an uncontrollable stressor speaks to the potential value of using uncontrollable stressors in eliciting depressive-like symptoms given appropriate test situations. Despite the transituational effects of stressors, it seems that contextual and explicit cues associated with the inescapable shock situation are important in determining subsequent escape performance. For instance, Maier (57) reported that although the effects of inescapable shock on subsequent escape performance dissipate over time, periodic reminders, comprising exposure to the stressor environment, limited the time-related attenuation of the deficits. Likewise, it was reported that when shuttle testing was conducted in the same environment in which inescapable shock had been

Learned Helplessness in Mice

185

delivered, the escape interference was long lasting, and the effects of SSRI were in line with the effects seen in depressed humans. However, the parallel to the human condition was not apparent when the initial stressor and later test situations differed from one another (58). 2.5. Genetic Factors in Relation to the Interference Effect

Although uncontrollable stressors reliably induce disturbances in later shock-escape tests, marked strain differences have been observed in this regard. For instance, in a shuttle-escape test, BALB/cByJ mice display exceptional performance disturbances that are apparent within the first few trials. The C57BL/ 6 J strain generally appears to be much hardier and only modest behavioral deficits are typically evident. In still other strains (e.g., A/J), performance deficits are typically seen during the initial part of a testing session, after which proficient performance may emerge (59). As will be seen in the procedural sections of this chapter, the specific test parameters ought to consider the characteristics of the mouse strain being examined, as inappropriate parameters my obfuscate possible interference effects. The relevance of these strain differences to the current presentations is threefold. First, these genetic differences lend themselves to the analysis of endophenotypes of depression (i.e., identifying specific genes that are related to discrete symptoms or specific neurochemical changes related to illness) (60, 61). Likewise, having these genetic differences present favors the analysis of potential pleiotropic effects of particular genes (i.e., a gene or set of genes having more than a single phenotypic outcome), and to determine whether a neurochemical and behavioral effect stemming from a gene (or set of genes) are causally related to one another. Finally, genetically engineered mice (knockout and transgenic strains) often use C57BL/6 and BALB/c as background strains. Knowing that these strains differ with respect to the emergence of stressor-induced behavioral disturbances allows the researcher to establish behavioral disturbances in the wild-type mice. Specifically, when a difference is not observed between control and stressed animals, it is often the case that the strain of mice used is too vulnerable or too resilient to the effects of stressor. If mice (experimental and control) show rapid escape latencies, then it is likely that the strain is relatively hardy one or that the task demands are too simple. Conversely, if all animals, including controls, show very slow latencies, then it may be that the strain is exceptionally vulnerable or that the task demands are too stringent. In the former instance, increasing the motoric difficulty of the test can ameliorate this, whereas in the latter instance the problem can be remedied by relaxing the task demands.

186

Anisman and Merali

3. Basic Protocol The procedure we have used to assess the effects of controllable and uncontrollable shock on subsequent shock-escape performance in mice is predicated on the supposition that motorically demanding tasks are more likely to permit the appearance of an interference effect. This paradigm, which involves as escape-delay procedure, is modifiable by a variety of treatments thought to affect processes that might subserve depressive illness. At the same time, performance in this test is also modifiable by treatments that are not commonly used as antidepressants. Thus, as indicated earlier, it may be that the paradigm might be useful as a screen for antidepressants, but might not be most practical for modeling depressive illness. 3.1. Equipment, Materials, and Setup

Typically, when the effects of controllable vs. uncontrollable stressor are assessed, a yoked triadic paradigm is used where three animals are tested concurrently. In one condition, mice receive a series of escape trials in which footshock is terminated when the animal makes an appropriate instrumental response (e.g., running from one compartment of a test chamber to another). Mice in a second condition are placed in an identical chamber in which the grid floor is connected in series with the chamber containing the first animal. Thus, whenever shock is delivered to the first chamber, it is concurrently delivered to the mouse in the second chamber. When the mouse in the first chamber emits an appropriate operant response that results in the shock being terminated, the shock also terminates in the second chamber. Thus, identical amounts of shock are delivered to the two mice, but only the mouse in the escape condition can control shock termination. For the mouse in the second chamber, stressor offset is independent of the animal’s responses. Mice in the third condition serve as a no stressor control; mice are placed into an identical apparatus, but shock is not delivered. Escapable and inescapable shock (and no shock) are administered in three identical shuttle boxes. These (sized for the mouse) can be commercially obtained or produced in-house. These chambers should comprise materials (e.g., dark Plexiglas that will not absorb odors and are readily washable). A shuttle box for mice typically measures about 30.0  10.0  15.0 cm and is covered by a red translucent Plexiglas roof. The floor of each chamber is composed of 0.32-cm stainless steel rods, spaced 1.0 cm apart (center to center) to permit feces to drop through. It is important that the grid floor be wired so that the mice are unable to stand on bars of the same polarity, thereby avoiding shock. To this end, the

Learned Helplessness in Mice

187

grid floor is wired through a commercially obtainable scrambler (companies that manufacture avoidance/escape equipment usually also provide scramblers) where the polarity of bars changes rapidly. As well, the end walls of each chamber are of stainless steel (or lined with stainless steel) and connected to the grid floor, thus preventing the mice from escaping shock by standing on a single grid rod and leaning on a back wall. Footshock (usually about 150–300 mA; AC, 60 Hz) should be delivered to the grid floor through a high-voltage source (approximately 3000 V), thereby limiting the current changes attributable to the resistance of the mouse. The boxes are divided into two compartments by a stainlesssteel wall containing an opening of approximately 5.0  6.0 cm. This opening is covered by a movable light-weight gate that moves horizontally. When the gate is open (by having it slide through a slot in the shuttle box wall) access between the compartments is possible. Several mechanical devices are available that allow for the gate to be opened at predetermined intervals. Photocells located on either side of the gate (1.5 cm), situated approximately 2.5 cm and 4.5 cm. above the grid floor, detect when a shuttle crossing was made. When cells on both sides of the gate are broken concurrently (as occurs when the mouse was halfway through the opening), the cells do not trigger. Only when the beam on the safe side is crossed, without simultaneously crossing a beam on the shock side, is the cell triggered and the trial ends. Through this method, trials are not terminated by events such as a tail flick that cuts the photobeam or if a mouse standing in one chamber only sticks its head into the adjacent chamber. A computer controls the operation of the trials and records response latencies. Whether shuttle boxes are constructed in-house or obtained commercially, care needs to be taken to avoid several potential pitfalls. Mice are exceptionally talented in finding unauthorized methods of avoiding shock. Among other things, mice will find the smallest orifice or protrusions from the walls from which they can hang, thereby avoiding contact with the grid floor. Care also needs to be taken to assure that the chamber is thoroughly cleaned (e.g., using 70% alcohol or an alternative cleaning substance), particularly as urine or feces, as well as fur, could provide danger pheromones, thereby affecting the behavior of the next animal that is tested. As well, urine, feces, or fur may short-circuit specific grid rods allowing animals a way of avoiding shock. Shuttle boxes are housed individually in ventilated, sound attenuating chambers. The shuttle boxes are strapped down to ensure that movements, such as the gate opening, do not disturb or shake them, thereby stimulating responses on the part of the mouse.

188

Anisman and Merali

3.2. Methods 3.2.1. Subjects

3.2.2. Procedure 3.2.2.1. Exposure to Controllable, Uncontrollable, or NoShock Treatments in a Triadic Design

In our studies, we use male or female mice, over 60 days of age. As transport from the supplier may prove to be a stressor, we typically breed mice in house; however, if they are purchased, then at least 2 weeks are permitted for acclimation to the laboratory. To avoid the influence of isolation (which could potentially be a stressor in this socially ordered species), mice are housed in groups of four. However, as the impact of the stressor procedures could interact with social interactions (e.g., could influence fighting) that might affect behavioral or neurochemical changes, mice are individually housed several days prior to their use for experimental purposes. The interference effect is evident across a wide number of mouse strains. However, it ought to be noted that BALB/c mice exhibit particularly marked disturbances in a shock-escape task (and hence to avoid ceiling effects it is best to use relatively short escape delays, e.g., 2 s). In contrast, C57BL/6 mice tend to be relatively resilient and the interference effect is much less pronounced (it is suggested that longer escape delays be used with this strain, e.g., 6 s). The strain differences are especially important as mice with engineered gene deletions often use one or the other of these as background strains. The protocol consists of two sessions. The time between the two is dependent on the specific questions being addressed. l To avoid potential effects of transporting mice from the vivaria to the test area, as well as the potential influence of odors that might be present in hallways, mice are brought to the test area on the day prior to the procedures being initiated. l

All testing is conducted over a fairly narrow time window (typically between 08:00 and 12:00 h in our studies) to limit the influence of diurnal neurochemical variations.

l

Mice are initially exposed to either controllable shock, uncontrollable shock, or no shock.

l

The treatment is typically applied over approximately a 1-h period. Evaluation of the stressor-provoked behavioral impairments is conducted during the second session. Depending on the procedures used (e.g., time between the two sessions), it is possible to assess the effects of pharmacological treatments applied prior to testing (therapeutic effects of the drug), as well as prior to the inescapable shock session (prophylactic actions).

l

Mice are individually placed into the shuttle boxes relatively quickly to assure that all the three animals in a triplet receive comparable treatments. After 1 min, the training procedure commences. Mice in the escapable condition receive 60 shockescape trials (150 mA; AC, 60 Hz) at an inter-trial interval of 60 s between trials. Alternatively, trials can occur at random intervals ranging, say, from 30 to 90 s.

Learned Helplessness in Mice l

189

A trial consists of footshock being presented, but only after a brief period (1.0–2.0 s) does the gate separating the compartments open, thus permitting escape. In the absence of this procedure, mice will often escape in a fraction of a second, and hence mice in the uncontrollable condition will receive little stressor exposure. When the gate opens, it is important that this occurs gently and quietly, as loud sounds or the shuttle-box being shaken, will elicit a startle response that will facilitate motor responses and hence improved escape performance.

Upon crossing the photobeams on the safe side of the shuttle box, the trial terminates and the gate closes. As well, if an escape response does not occur within 30 s, the trial terminates. In the latter instance, the trial will commence on the same side of the shuttle box as it did on the preceding trial. As indicated earlier, the grid floors of the first (escapable shock) and second (inescapable shock) chambers are connected to a common shock controller. Thus, when the mouse in the escapable condition makes a response that results in a trial ending, the shock in the second chamber terminates as well. The mouse in the third chamber is not shocked, and is left undisturbed throughout the session. Latencies (sec) to escape from the shock, measured from the time that the gate opens, are recorded to provide a record of the amount of shock the mice received. Mice are returned to their holding cages until subsequent behavioral testing. In our studies, as indicated earlier, the shock and test apparatus are similar thus promoting relatively long-lasting interference effects. However, the uncontrollable stressor session can be delivered in an entirely different apparatus, and may involve an operant (in the escapable condition) other than a shuttle response. It is best to use an operant that is consistent with the animals’ defensive style, as opposed to one such as lever pressing. In general, the yoked paradigm is used to establish the influence of stressor controllability on performance in subsequent tests. In some studies dealing with stressor uncontrollability, the focus might not be on the specific influence of controllable vs. uncontrollable stressors, but on factors that influence the effects of an uncontrollable stressor experience (e.g., attenuation of the interference following inescapable shock by pharmacological interventions, or provocation of the interference by other drugs). In such instances, it may not be necessary to include the triadic design in each of a series of studies. Having demonstrated that the procedure, in fact, yields outcomes consistent with the learned helplessness paradigm (i.e., controllable and uncontrollable stressors differentially influence later performance), later studies need l

190

Anisman and Merali

only include the uncontrollable and nonshock conditions. In this case, the initial shock treatment can be administered in chambers that do not include a movable gate or photocells to capture response latencies. Shock delivery is presented at predetermined durations (e.g., 2 s), or of variable duration averaging 2 s (0.5 s on some trials, 2 s on others, etc) and these can be administered at fixed or variable intervals. 3.2.2.2. Testing in a Shuttle Escape Task

3.2.2.3. Behavioral Tests Other than Shuttle Escape

Performance in an escape test may vary as a function of the time between the inescapable shock session and subsequent testing. In fact, we found that the performance deficit is less pronounced if testing occurs soon after the initial session, but becomes progressively more pronounced over the 72 h following this session. The performance deficits typically persist for weeks afterward. In most of our studies, we test animals 24 h after inescapable shock, but if investigators are assessing the effects of chronic intervening treatments (e.g., repeated administration of an antidepressant), then lengthy intervals between inescapable shock and testing can be used. Testing animals involves essentially the same procedure as that of the escape training used in the initial session, with a few exceptions. l Mice receive 30 test trials. l

During the initial five test trials, shock onset is accompanied by the gate separating the compartments opening immediately without an escape delay being imposed. These five trials have two purposes. Specifically, they may act as reinstating stimuli so that neurochemical changes will again be engendered in mice that had previously been exposed to inescapable shock. Moreover, as the task can be a difficult one for some animals, it facilitates performance in mice that had not been previously stressed.

l

On the ensuing 25 trials, a delay is introduced between shock onset and the gate being opened. In most of our studies, the delay is 4 s in duration, but in strains of mice that are particularly vulnerable to the effects of inescapable shock, the gate delay is shortened (2 s) whereas in strains that are relatively resilient, a longer delay (4 or 6 s) is used.

l

As in the training session, the trial terminates when the mouse crosses from one chamber to the next, or if an escape response is not emitted within 30 s. The interval between trials can range from 30 to 90 s. On each trial, latencies to escape are recorded from the time of the gate being opened.

As already indicated, there is ample reason to believe that testing animals in a shuttle-escape test likely does not provide a model for depression. Indeed, we argued that even if the shuttle-escape test

Learned Helplessness in Mice

191

bore face validity, evaluation of any single behavior might not be sufficient as an index of depression. To be sure, diagnoses of depression in humans are based on the presence of a set of behaviors, and it would be reasonable to expect that this would be true of mice as well. Short of this, it may be appropriate to examine those behaviors that are most closely aligned with major depressive disorder. In our studies, we chose to assess the potential anhedonic effects of controllable vs. uncontrollable stressors, as anhedonia is a key symptom of major depression. To this end we assessed mice in a test in which they respond for rewarding brain stimulation from the lateral hypothalamus, ventral tegmentum, or the nucleus accumbens. Behavioral disturbances are readily induced in a self-stimulation paradigm, provided that animals are tested soon after the inescapable shock session. Once the interference of performance is apparent, it will still be evident days later. However, if the initial test is first conducted several days after the uncontrollable shock session, then performance deficits will not be evident (see review in (56)). Other alternatives for measuring anhedonia include determination of operant responding (nose-poke) on a progressive ratio (PR) schedule to obtain a palatable snack (0.05 ml of 20% sucrose) in nondeprived animals, while lab chow is freely available (through an aperture that allows for food consumption to be determined by weight). Animals are trained to receive rewards based on a schedule where the number of operant responses required for reward increases according to a predetermined progression (1, 2, 4, 6, 9, 12, 15, 20, 25, 32, 40, 50, 62, 77, 95). Motivation is defined as the last response ratio achieved before responding ceased for 10 min (break point). Once the break point is reached, a time-out is instituted, and then the procedure commences again (mice live in the test cage, and in each 24 h period, PR performance regimens are 3 h apart). In this test, which measures the relationship between cost and reward, we measure both snack and lab chow consumption concurrently. Thus, we can determine whether anhedonic effects (decreased responding for sucrose) are present even in the absence of disturbed chow intake reflecting sickness or anorexia that could be elicited by stressors (62). Inasmuch as anxiety and depression are frequently comorbid, it might be advisable to assess animals in various tests of anxiety (e.g., open-field emergence, light/dark exploration, social interaction tests, elevated plus-maze, defensive burying, approaching a favored snack in a novel environment, and acoustic startle response. In general, a fear-potentiated startle response is not readily elicited in mice (at least not in our hands, and there have been few demonstrations of this in the literature), and thus we have not included this procedure in our protocols. 3.3. Anticipated Results

Investigators frequently define the interference effect in terms of the number of escape failures animals’ display. However, we found that the course of the interference over a session may

192

Anisman and Merali

vary across strains of mice, and thus providing the ‘‘total’’ number of escape failures may not be an appropriate index of what the inescapable stressor actually provoked. For instance, as described earlier, we found that in the vulnerable BALB/c mice, the interference effect is evident after only a few trials, and then persists throughout the session, whereas in C57BL/6 mice the escape failures begin to appear later in the session, and often improves thereafter. Mice of the A/J strain often display behavioral disturbances relatively early in a session (depending on the severity of the stressor) followed by marked improvement towards the later trials. Thus, we typically averaged escape latencies on successive blocks of five trials, and these values were then subjected to a repeated measures analysis of variance. This said, it ought to be recognized that a given treatment (e.g., a pharmacological manipulation) may influence response latencies, but without altering failures to escape. Thus, in addition to measuring escape latencies, it is important to specify the frequency of escape failures or the number of animals that display escape failures. One might assume that the interference effect would be most apparent soon after the inescapable shock session, and then wane with the passage of time. Indeed, in rats this has been reported with some paradigms (27). However, we have found that this was not necessarily the case. To the contrary, we have frequently found the interference to be more pronounced 24–72 h after the inescapable shock session than immediately afterward. Likewise, as described earlier, we have seen a similar effect when mice were assessed in a forced-swim paradigm (53). This time-dependent effect for the interference may vary with the severity of the stressor used, although we have not systematically assessed the parameters that are best in this regard. 3.4. Experimental Variables

The experimental variables that have been assessed to date are fairly broad. These have included a wide range of pharmacological manipulations, including those that act as antidepressants and anti-anxieties, as well as those used in the treatment of PTSD (e.g., propranolol; although this was done before propranolol was suggested as a potential agent to treat PTSD). In addition, other treatments were assessed that were thought to influence processes important for the emergence of depression (e.g., BDNF, CRH, 5-HT). With the increasing focus on genetic contributions to pathology, coupled with the development of methods to manipulate the expression of particular genes that code for brain neurotransmitters, hormones, and growth factors, the mouse has become the species of choice in many areas of research. Although this has not been the case with respect to animal models of depression, the advent of knockout and transgenic lines, as well as strategies to

Learned Helplessness in Mice

193

knock down certain genes for brief periods of time, the use of mice in these paradigms can be expected to increase. Yet, there are certain limitations to using strains of mice relatively indiscriminately. Strains of mice may display behavioral differences for any number of reasons; these may include differential neurochemical activity in response to stressors, differences in pain sensitivity or in the specific behavioral styles (jumping vs. running) adopted to deal with stressors. Still other strains (C3H/He and related substrains) suffer from retinal degeneration, and of course will be incompetent in dealing with visual discriminations. Importantly as well, mice from different suppliers may display very marked performance differences, and likewise, mice obtained from a given supplier but tested in different laboratories (despite the same testing procedures) may exhibit behavioral responses that are unique to each laboratory (63). 3.5. Troubleshooting

Whether or not an interference effect is elicited by inescapable shock may be influenced by any number of variables. If the task is too difficult, or if the strain of mouse is one that cannot readily deal with the task demands, then latencies of control mice will be relatively high, thus obfuscating potential effects of the previous stressor experience. Conversely, if the test is too simple (or not motorically demanding), then the very rapid escape latencies will be apparent in both the inescapably shocked mice as in the control conditions. Once the problem has been diagnosed, then it is a matter of finding the appropriate escape delay to be used. As described earlier, using too long an escape delay will hamper performance even in control animals, whereas too brief a delay will not allow an interference to become manifest. Generally, about a 2–4-s delay of an escape response will be sufficient, but this will vary with the strain of mouse used. In most studies, not all animals show an interference effect after inescapable shock (70% show the effect), and some previously nonstressed mice do (20%). The investigator should initially (five trials) use no delay in the shuttle test. This has the effect of reinstating the neurochemical disturbances and the interference effect in mice that had previously been exposed to inescapable shock, and importantly, also has the effect of minimizing the interference in the control mice. Mice are exceptionally talented in finding ways to avoid or escape shock (standing on a single grid bar and leaning against a wall; finding slight cracks from which to hang). Thus, care must be taken to avoid unauthorized responses from being established. Furthermore, mice may be adventitiously reinforced for certain behaviors. Should this occur an interference effect may be evident, but is not one that will readily be modifiable by antidepressant agents, whereas an interference related to neurochemical disturbances would be more readily modifiable.

194

Anisman and Merali

References 1. Anisman H, Merali Z, Hayley S. Neurotransmitter, peptide and cytokine processes in relation to depressive disorder: Comorbidity of depression with neurodegenerative disorders. Prog Neurobiol 2008; 85:1–74. 2. Anisman H, Matheson K. Anhedonia and Depression: Caveats Concerning Animal Models. Neurosci Biobehav Rev 2005; 29:525–546. 3. Overmier JB, Seligman ME. Effects of inescapable shock upon subsequent escape and avoidance responding. J Comp Physiol Psychol 1967;63:28–33. 4. Seligman ME, Maier SF. Failure to escape traumatic shock. J Exp Psychol 1967; 74:1–9. 5. Maier SF, Seligman MEP. Learned helplessness: Theory and Evidence. J Exp Psychol Gen 1976;105:3–46. 6. Maier SF, Watkins LR. Stressor controllability and learned helplessness: the roles of the dorsal raphe nucleus, serotonin, and corticotropin-releasing factor. Neurosci Biobehav Rev 2005;29:829–841. 7. Seligman MEP. 1975. Helplessness: On depression, development and death. Freeman, San Francisco; 1975. 8. Klein DC, Fencil-Morse E, Seligman ME. Learned helplessness, depression, and the attribution of failure. J Pers Soc Psychol 1976;33:508–516. 9. Douglas D, Anisman H. Expectancy incongruency or learned helplessness: Effects of inescapable stress on subsequent performance. J Exper Psychology 1975;1:411–417. 10. Beck AT, Kovacs M, Weissman A. Hopelessness and suicidal behavior. An overview. Behav Brain Res 2005;164:222–230. 11. Alloy LB, Abramson LY, Metalsky GI, et al. The hopelessness theory of depression: attributional aspects. Br J Clin Psychol 1988;27:5–21. 12. Schwartz B. Does helplessness cause depression, or do only depressed people become helpless? Comment on Alloy and Abramson. J Exp Psychol Gen 1981;110:429–435. 13. Seligman ME, Beagley G. Learned helplessness in the rat. J. Comp Physiol Psychol 1975;88:534–541. 14. Weiss JM, Glazer HI. Effects of acute exposure to stressors on subsequent avoidanceescape behavior. Psychosom Med 1975; 37:499–521.

15. Anisman H, deCatanzaro D, Remington G. Escape performance following exposure to inescapable shock: Deficits in motor response maintenance. J Exp Psychol Anim Behav Proc 1978;4:197–218. 16. Anisman H, Remington G, Sklar LS. Effects of inescapable shock on subsequent escape performance: Catecholaminergic and cholinergic mediation of response initiation and maintenance. Psychopharmacology 1979; 61:107–124. 17. Petty F, Sherman A. A neurochemical differentiation between exposure to stress and the development of learned helplessness. Drug Devel Res 1982;2:43–45. 18. Sherman AD, Petty F. Neurochemical basis of the action of antidepressants on learned helplessness. Behav Neur Biol 1988; 30:119–134. 19. Anisman H, Suissa A, Sklar LS. Escape deficits induced by uncontrollable stress: Antagonism by dopamine and norepinephrine agonists. Behav Neur Biol 1980;28:34–47. 20. Anisman H, Glazier S, Sklar LS. Modification by cholinergic manipulation of escape deficits produced by inescapable shock. Psychopharmacology 1981;74:81–87. 21. Furey ML, Drevets WC. Antidepressant efficacy of the antimuscarinic drug scopolamine: a randomized, placebo-controlled clinical trial. Arch Gen Psychiat 2006;63:1121–1129. 22. Zarate CA Jr, Payne JL, Singh J, Quiroz JA, Luckenbaugh DA, Denicoff KD, Charney DS, Manji HK. Pramipexole for bipolar II depression: a placebo-controlled proof of concept study. Biol Psychiat 2004;56:54–60 23. Porsolt RD. Animal model of depression. Biomedicine 1979;30:139–140. 24. Glazer HI, Weiss JM. Long-term and transitory interference effects. J Exper Psychol Anim Behav Proc 1976;2:191–201. 25. Glazer HI, Weiss JM. Long-term interference effect: An alternative to learned helplessness. J Exper Psychol Anim Behav Proc 1976;2:202–213. 26. Anisman H, Zacharko RM. Depression: The predisposing influence of stress. Behav Brain Sci 1982;5:89–137. 27. Weiss JM, Goodman PA, Losito BG, et al. Behavioral depression produced by an uncontrollable stressor: Relationship to norepinephrine, dopamine and serotonin levels in various regions of rat brain. Brain Res Rev 1981;3:161–191.

Learned Helplessness in Mice 28. Holsboer F. Corticotropin-releasing hormone modulators and depression. Curr Opin Investig Drugs 2003;4:46–50. 29. Nemeroff CB, Vale W. The neurobiology of depression: inroads to treatment and new drug discovery. J Clin Psychiatry 2005;66: 5–13. 30. Hammack SE, Richey KJ, Schmid MJ, et al. The role of corticotropin-releasing hormone in the dorsal raphe nucleus in mediating the behavioral consequences of uncontrollable stress. J Neurosci 2002; 22:1020–1026. 31. Hammack SE, Schmid MJ, LoPresti ML, et al. Corticotropin releasing hormone type 2 receptors in the dorsal raphe nucleus mediate the behavioral consequences of uncontrollable stress. J Neurosci 2003; 23:1019–1025. 32. Bale TL, Picetti R, Contarino A, et al. Mice Deficient for Both Corticotropin-Releasing Factor Receptor 1 (CRFR1) and CRFR2 Have an Impaired Stress Response and Display Sexually Dichotomous Anxiety-Like Behavior. J Neurosci 2002; 22: 193–199. 33. Duman RS, Monteggia LM. A neurotrophic model for stress-related mood disorders. Biol Psychiatry 2006;59:1116–1127. 34. Manji HK, Duman RS. Impairments of neuroplasticity and cellular resilience in severe mood disorders: implications for the development of novel therapeutics. Psychopharmacol Bull 2001;35:5–49. 35. Shirayama Y, Chen AC, Nakagawa S, et al. Brain-derived neurotrophic factor produces antidepressant effects in behavioral models of depression. J Neurosci 2002; 22:3251–3261. 36. Siuciak JA, Lewis DR, Wiegand SJ, et al. Antidepressant-like effect of brain-derived neurotrophic factor (BDNF). Pharmacol Biochem Behav 1997;56:131–137. 37. McQueen GM, Ramakrishnan K, Croll SD, et al. Performance of heterozygous brainderived neurotrophic factor knockout mice on behavioral analogues of anxiety, nociception, and depression. Behav Neurosci 2001; 115:1145–1153. 38. Altar CA. Neurotrophins and depression. Trends Pharmacol Sci 1999; 20:59–61. 39. Bland ST, Tamlyn JP, Barrientos RM, et al. Expression of fibroblast growth factor-2 and brain-derived neurotrophic factor mRNA in the medial prefrontal cortex and hippocampus after uncontrollable or controllable stress. Neuroscience 2007;144:1219–1228.

195

40. Greenwood BN, Strong PV, Foley TE, et al. Learned helplessness is independent of levels of brain-derived neurotrophic factor in the hippocampus. Neuroscience 2007; 144:1193–1208. 41. Anisman H, Sklar LS. Catecholamine depletion upon reexposure to stress: Mediation of the escape deficits produced by inescapable shock. J Comp Physiol Psychol 1979; 93:610–625. 42. Maier SF. Learned helplessness and animal models of depression. Prog Neuropsychopharmacol Biol Psychiatry 1984; 8:435–446. 43. Flores C, Stewart J. Basic fibroblast growth factor as a mediator of the effects of glutamate in the development of longlasting sensitization to stimulant drugs: studies in the rat. Psychopharmacology 2000; 151:152–165. 44. Tilders FJH, Schmidt ED. Cross-sensitization between immune and non-immune stressors. A role in the etiology of depression? Adv Exp Med Biol 1999;461:179–197. 45. Post RM. Transduction of psychosocial stress into the neurobiology of recurrent affective disorder. Amer J Psychiat 1992; 149:999–1010. 46. Harkness, K.L. 2008. Life events and Hassles. In K.S. Dobson, & D. Dozois Eds. Risk Factors in Depression. Elsevier Science, NY; 317–342. 47. Anisman H, Matheson K, Hayley S. Neurochemical and transmitter models of depression. In: Dobson KS, Dozois DJA, eds. Risk Factors for Depression. N.Y., Elsevier Science; 2008:63–90. 48. Weiss JM, Glazer HI, Pohorecky LA, et al. Effects of chronic exposure to stressors on avoidance-escape behavior and on brain norepinephrine. Psychosom Med 1975; 37:522–534. 49. McEwen BS. Physiology and neurobiology of stress and adaptation: central role of the brain. Physiol Rev 2007;87:873–904. 50. Valentino DA, Dufresne RL, Riccitelli AJ. Effects of a single inescapable swim on longterm brain stimulation reward thresholds. Physiol Behav 1990;48:215–219. 51. Irwin J, Suissa A, Anisman H. Differential effects of inescapable shock on escape performance and discrimination learning in a water escape task. J Exp Psychol Anim Behav Proc 1980;6:21–40. 52. Szostak C, Anisman H. Stimulus perseveration in a water-maze following exposure to

196

53.

54.

55.

56.

57.

58.

Anisman and Merali controllable and uncontrollable shock. Behav Neur Biol 1985;43:178–198. Prince CR, Anisman H. Acute and chronic stress effects on performance in a forced swim task. Behav Neur Biol 1984;42:99–119. Rosellini RA, DeCola JP, Plonsky M, et al. Uncontrollable shock proactively increases sensitivity to response-reinforcer independence in rats. J Exp Psychol Anim Behav Process 1984;10:346–359. Bevins RA, Besheer J. Novelty reward as a measure of anhedonia. Neurosci Biobehav Rev 2005;29:707–714. Zacharko RM, Anisman H. 1991 Stressor-induced anhedonia in the mesocorticolimbic system. Neurosci Biobehav Rev 1991;15:391–405. Maier SF. Exposure to the stressor environment prevents the temporal dissipation of behavioral depression/learned helplessness. Biol Psychiatry 2001;49:763–773. Valentine G, Dow A, Banasr M, et al. Differential effects of chronic antidepressant

59.

60.

61.

62.

63.

treatment on shuttle box escape deficits induced by uncontrollable stress. Psychopharmacology 2008; 200: 585–96. Shanks N, Anisman H. 1988. Stressorprovoked behavioral changes in six strains of mice. Behav. Neurosci 1988;102: 894–905. Cryan JF, Slattery DA. Animal models of mood disorders: Recent developments. Curr Opin Psychiatry 2007;20:1–7. Gottesman II, Gould TD. The endophenotype concept in psychiatry: etymology and strategic intentions. Am J Psychiatry 2003;160:636–645. Merali Z, Brennan K, Brau P et al. Dissociating Anorexia and Anhedonia Elicited by Interleukin-1b: Antidepressant and Gender Effects on Responding for ‘‘Free Chow’’ and ‘‘Earned’’ Sucrose Intake. Psychopharmacology 2002;165:413–418. Crabbe JC, Wahlsten D, Dudek BC. Genetics of mouse behavior: interactions with laboratory environment. Science 1999;284:1670–1672.

Chapter 11 The Mouse Light–Dark Box Test Martine Hascoe¨t and Michel Bourin Abstract The light/dark test is based on the innate aversion of rodents to brightly illuminated areas and on the spontaneous exploratory behaviour of the animals, applying mild stressors, i.e. novel environment and light. The test apparatus consists of a small dark secure compartment (one-third) and a large illuminated aversive compartment (two-thirds). The test was developed with male mice. The strain, weight and age may be crucial factors. The extent to which an anxiolytic compound can facilitate the exploratory activity depends on the baseline level in the control group. Differences between the type and severity of external stressors might account for variable results reported by different laboratories. The light–dark test may be useful to predict anxiolytic-like or anxiogenic-like activity in mice. Transitions have been reported to be an index of activity exploration because of habituation over time and the time spent in each compartment to be a reflection of aversion. Classic anxiolytics (benzodiazepines) as well as the newer anxiolytic-like compounds (e.g. serotonergic drugs or drugs acting on neuropeptides receptors) can be detected using this paradigm. It has the advantages of being quick and easy to use, without requiring the prior training of animals. Key words: Antidepressants, antipsychotics, anxiolytics, mouse light/dark test, neuropeptides receptor ligands.

1. Introduction Classification of animal models of anxiety has often been achieved according to the nature of the aversive stimulus and the response elicited, suggesting that the neuronal control of anxiety may differ according to whether the interpretation of an aversive signal is innate or learned (1, 2) and whether it causes the emission of a response or conversely inhibits an ongoing, rewarded behaviour. Animal models of anxiety can be grouped into two main subclasses. Animal’s conditioned responses involve stressful and T.D. Gould (ed.), Mood and Anxiety Related Phenotypes in Mice, Neuromethods 42, DOI 10.1007/978-1-60761-303-9_11, ª Humana Press, a part of Springer ScienceþBusiness Media, LLC 2009

197

198

Hascoe¨t and Bourin

often painful events (e.g. exposure to electric foot shock). Ethologically based paradigms involve the animal’s spontaneous or natural reactions (e.g. flight, avoidance, freezing) to stress stimuli that do not explicitly involve pain or discomfort (e.g. exposure to a novel highly illuminated test chamber or to a predator). Ethologically based animal models of fear and anxiety attempt to approximate the natural conditions under which such emotional states are elicited. By employing non-painful aversive stimuli to induce fear and anxiety, ethological tests are thought to minimise possible confounding effects of motivational or perceptual states arising from interference with learning/memory, hunger/thirst or nociceptive mechanisms (3). The light/dark test is based on the innate aversion of rodents to brightly illuminated areas and on the spontaneous exploratory behaviour of the animals, applying mild stressors, i.e. novel environment and light (4). A natural conflict situation occurs when an animal is exposed to an unfamiliar environment or novel objects. The conflict is between the tendency to explore and the initial tendency to avoid the unfamiliar (neophobia). The exploratory activity reflects the combined result of these tendencies in novel situations. Thus, in the light/dark test, drug-induced increases in behaviours in the white part of a two-compartment box, in which a large white compartment is illuminated and a small black compartment is darkened, are suggested as an index of anxiolytic activity. As well an increase in transitions without an increase in spontaneous locomotion is considered as anxiolytic activity. It is interesting to note that this effect is only observed in certain strains of mice or with certain drugs. This model differs with other supposedly models of anxiety which are not equivalent in terms of elicited/induced emotional state (5, 6, 7, 8). The light–dark paradigm does not involve pathological anxietyrelated behaviours. Lister (9) has described this kind of model as animal models of ‘‘state’’ anxiety. State anxiety is that seen in a response to the level of stress and to the way that stress is (10). In such procedures, subjects experience anxiety at a particular moment in time and it is increased by the presence of anxiogenic stimulus. Ethological stimuli are diverse in nature. Furthermore, ethological models present individual differences, variable behavioural baseline levels. The goal of this chapter is to give a review of the main data obtained with the light/dark test and point out the inconsistent findings linked with the various modifications of the apparatus as well as the methodology.

The Mouse Light–Dark Box Test

199

2. Test Apparatus Although the light/dark test was based on the initial model described by Crawley and Goodwin (4), many authors have used it with several structural modifications (Table 11.1). Typical dimensions of the compartment are generally one-third for the dark and two-thirds for the light compartment with an exterior size of 46  27  30 cm (Fig. 11.1). Nevertheless Costall et al. (11) have differently distributed the compartments with two-thirds for the dark compartment. The model is based on the observation that, although nocturnal rodents such as mice will naturally tend to explore a novel environment, open fields appear to have aversive properties which inhibit exploratory behaviour. Here, the secure area is achieved by the small dark compartment (one-third) and the aversive area by the large illuminated compartment (two-thirds). The opening between the two compartments is no more than 7 cm. One variation in the model was the size of the compartment with an increase to 42  42  30 cm high for the light compartment and 40  23  20 cm high for the dark one. Gao and Cutler (12) used an apparatus with high dimensions (31  46 cm for the dark area and 60  46 for the light one). Imaizumi and Onodera (13) have used a modified model based on the automated one of Young and Johnson (14) with slight modifications. The test chamber consisted of two compartments of equal size (15  15  15 cm) with two partitions between the compartments. Another modified model is the one used by Belzung et al. (15). The apparatus consisted of two polyvinylchloride boxes of the same size (20  20  14 cm). One was darkened with cardboard and the other was highly illuminated; an opaque plastic tunnel (5  7  10 cm) separated the two compartments. The more modified light/dark transition test is the one of Shimada et al. (16), as the apparatus was a corridor-type runway with outer (40  12 cm) and inner (20  12 cm) walls. Two rectangular chambers made of black Plexiglas were diagonally placed on the corner of the corridor. The centre of the apparatus was illuminated by a fluorescent lamp. In the first descriptions of the light/dark test that used anxiolytics in mice, photocells across the partition were used to detect transitions between the two areas (4, 17, 18). Later studies used human observers to record transitions of the mice and the amount of time mice spent in each area (15). Video recordings of mice behaviour in the chamber have also been used to detect increases in locomotor and rearing activities in the lit area that were accompanied by decreases in these behaviours in the dark area (17, 19). An apparatus was developed that allows automatic measurement of locomotion, rearing and time spent in light and dark zones and shuttle crossings between zones using a computer-controlled detection equipped with infrared beam sensors, where

45  27  27 dark 3/5 light

44  21  21 # dark  light

Normal

In the light

10 min

Size of boxes (cm)

Light/dark cycle

Compartment where mice are placed

Duration of the test 5 min

In the light

Inversed

00

11, 24, 25, 26, 27

4, 21, 22, 23

Costall et al.

References

Crawley et al.

Table 11.1 The main different light/dark procedures used

5 min

In the light

5 min

In the light

Inversed

20  20  14 ½ dark ½ light Tunnel between the 2 compartments

31  46 (dark) 60  46 (light)

Inversed

15, 30

Belzung et al. 29

Gao and Cutler

5 min

In the light

3 min

In the corner

Normal

Corridor-type runway 2 compartments diagonally Placed on corner

46  27  30 # dark  light

Normal

16

Shimada et al.

20, 31, 32, 33

Hascoe¨t and Bourin

10 min

in the light

4 hours of dark before testing

40  33  20 (dark) 42  42  20 (light)

14, 34

Young and Johnson

200 Hascoe¨t and Bourin

+

+

+

+

Movements in the light

Movements in the dark

Time in the light

Time in the dark

+Parameters used by authors considered

Rears

+

+

Transitions

+

+

+

+

+

+

Latency time

Parameters used

+

+

+

+

+

+

+

+

+

+

+

+

+

+

+

+

+

+

+

+

+

+

+

+

+

+

+

The Mouse Light–Dark Box Test 201

202

Hascoe¨t and Bourin

Fig. 11.1. Image of a light–dark box. The test apparatus is typically composed of two compartments (area ration one-third dark and two-thirds light), with a door that allows movement between the two compartments.

interruptions of the infrared beams in the chamber are automatically recorded by an analyser and then transmitted to a computer (14). The computerised apparatus is highly desirable, as the animals are subjected to minimal external perturbations. The apparatus used by Hascoe¨t and Bourin (20) consisted of a fully automated box monitored by a computer. An open-topped rectangular box (46  27  30 cm high) was divided into a small (18 cm  27 cm) area and a large (27 cm  27) area with an opening door (7.5 cm  7.5 cm) located in the centre of the partition at floor level. The small compartment was painted black and illuminated under a dim red light (60 W; 4 lux), whereas the large compartment was painted white and brightly illuminated with a 60-W (400 lux) light source. The compartments were equipped with infrared beam sensors (four in the white area, three in the black one) enabling the detection of locomotion in each zone, time spent in each zone, latency of the first crossing from one compartment to the other and shuttle crossings between both compartments. The data from these four parameters were directly collected by computer.

3. Animals Mouse is far more studied as a genetic organism, because it is more easily housed (many more mice can be housed in a given space), it breeds more quickly, homologue recombination

The Mouse Light–Dark Box Test

203

techniques are now standard for the mouse (and not yet generally available for the rat) and the mouse genome is more completely characterised. Further, genetic modifications to create knockout mice are easier than that in rats. The developmental impairment elicited by the mutation remains a source of discussion, but this barrier is going to disappear with the design of inducible knockout mice. This test was first developed with male mice. As inbred mice displayed substantial variability in spontaneous behaviour, the choice of the strain may be a crucial parameter. Early studies with this paradigm have concluded that highly active strains of mice show consistently larger percentage increases in exploratory behaviour after diazepam treatment. The C57Bl/6 J and the SW-NIH strains seemed the strains of choice for anti-anxiety testing in the light/dark test (21). The C57Bl/6 J has been reported to show a robust effect as it demonstrated a maximum diazepam response of 129% in mean exploratory behaviour (21). Strains with a low number of baseline transitions generally show weak responses to anxiolytics (23). However, the results of a recent study call this theory into question (20). C57Bl/6 J mice were obtained from two different sources and tested in comparison with the Swiss mouse strain. All strains showed the same baseline transition activity. Swiss strain showed a decrease in the time spent in the dark area at a dose of 1 mg/kg of diazepam (46%), as did the C57Bl/6 J from IFFA CREDO (L’Arbresle, France); however, this effect was less significant. On the other hand, C57Bl/6 J from Janvier breeding farm (Le Genest, France) did not show any significant anxiolytic behaviour at any dose. All strains showed a similar baseline activity in movements in each compartment. More recently, Bouwknecht and Paylor (35), analysed the behaviour of nine mouse strains using three models of anxiety, including the light–dark paradigm. Some mice strains, such as A/J, 129S6/ SvEvTac, 129S1/SvImJ and BALB/cByJ, demonstrate a high anxiety-related response as seen by a very low number of transition and high percentage of time spent in the dark after administration of saline control. On the other hand, other strains were more active in crossing between the two compartments and spent more time in the light compartment, showing nevertheless high preference for the dark one; this is the case for DBA/2 J, C3H/HeJ, C57BL/6 J, FVB/NJ and the outbred CD1-ICR mice. Other reports have found that transitions were low with BALB/c mice, intermediate with DBA/2 mice and high with C57BL/6 mice (21, 36). The S129 mice display low transition level and low time spent in the light area (37, 38). In vehicle-treated mice, Griebel et al. (39), divided mice strain from IFFA CREDO (L’Arbresle, France) into three distinct groups using the light–dark model, one group for non-reactive

204

Hascoe¨t and Bourin

mice strain (NZB and SJL), an intermediate reactive group (C3H, CBA, DBA/2, NMRI, C57BL/6 and Swiss) and a highly reactive group for BALB/c strain. The authors point out that there exists marked difference between a number of strain in their sensitivity to the anxiolytic-like effect of the benzodiazepine diazepam both in the light–dark and the elevated plus maze tests. Mice of the BALB/c and Swiss strain were reactive to diazepam treatment in both tests, while C57BL/6 (as we found previously), DBA/2, NMRI and NZB mice demonstrated anxiolytic-like effect only in the elevated plus maze. We should notice that BALB/c and C57BL/6 mice differ in affinity and/or density of the benzodiazepine receptors (40). All these results demonstrate that the Swiss strain does display anxiolytic-like behaviour and is highly suitable to be used in the light–dark test. We need to be careful with the origin of the strain, and the genetic, neurochemical or neuroanatomical background of the mice. The light/dark exploration test has not been validated for female mice and the influence of the oestral cycle has not been yet investigated. Only one study, revealed no influence of gender in the light–dark test (38) Transgenic and knockout mice are increasingly popular tools in the study of gene function in emotional behaviour (41, 42). As we saw previously the importance of mice strain in the test, we would like to highlight the importance of the genetic background for the behavioural pharmacological studies. Negative data have been found regarding the sensitivity to the anxiolyticlike effect of diazepam for the knockout generated on different genetic backgrounds, as for example with mice with C57BL/6 backgrounds. The pharmacological response between gene-targeted mice and their wild-type controls have more to do with genetic background rather than with the targeted mutation (43). Recent studies reported the effect of some genetic-modified mice in the light–dark test. There is now growing body of evidence that those GABAB receptors play a critical role in anxiety. Using mice lacking functional GABAB receptor, (GABA(B(1)))–/– mice, Mombereau et al. (44) have shown that those mice were more anxious than their wild-type littermates using a light/dark box, spending less time in the light area and decreasing number of transitions. More controversial results have been found in innate model of anxiety like the light–dark model (45). Genetic invalidation of CCK (2) receptors (cholecystokinin) induced an anxiolytic-like effect in the light–dark test (46). Another parameter that must be discussed in interpreting the results is the weight, or more precisely, the age and the neuronal maturation of the animals. Some authors (47), using the test of habituation of acoustic startle response (ASR) and freezing responses in rats of different ages, found that rats were

The Mouse Light–Dark Box Test

205

not able to express long-term habituation of the ASR before 30 days of age. The question of age seemed therefore to be of particular importance in the experimental conditions. Male Swiss mice of different ages (from 3.5 weeks to 8 weeks; 14– 32 g) were used in the light/dark test, receiving saline only and their performance was compared. The optimal age at which control values were optimum was that of 4 weeks in our experience (32). Mice at this age spent 58% of the total test duration in the dark compartment. The oldest mice (i.e. 8 weeks old) exhibited an increase in total activity, characterised by increase in movements in each compartment, together with an increase in the number of transitions. An age-related effect was found, the best period being that of 4 weeks, suggesting caution when interpreting the results of mice in the light/dark paradigm, and also it is now well established that diminished central nervous system function that occurs with ageing is accompanied by discrete changes in various neurotransmitter networks and/or receptors. Problems of methodology are also highlighted in that increasing weight and age should be taken into consideration in chronic studies or repeated administration studies where animals are retained for several week.

4. Test Conditions The extent to which an anxiolytic compound can facilitate the exploratory activity depends on the baseline level in the control group. There are a number of non-genetic, non-pharmacological manipulations that lead to modulate the general stress levels of the animal, which, when performed before testing, have profound effects on behaviour. Deliberate or accidental manipulation of these influential factors can also dramatically alter the effects of drugs (48). Differences between the type and severity of external stressors (housing conditions, handling procedures during test) might account for variable results reported by different laboratories, because stress affects motor activity, and different types of stressors might affect the behavioural response differently (49). Each new drug is first screened for non-specific increases or decreases in general locomotion in a small bare undifferentiated arena in an activity meter which measures all forms of horizontal and vertical locomotor activities. Any drug yielding increases in both transitions and locomotion is considered a general motor stimulant and eliminated as a putative anxiolytic. Studies suggest that acute stress immediately before testing mice in a black-and-white twocompartment test box can enhance the anxiolytic-like response and thereby increase the robustness of the test model. A study by

206

Hascoe¨t and Bourin

Belzung and Le Pape (30) suggests that tail suspension test (TST) compared with the forced swimming test (FST) and foot shock is an appropriate stressor, producing a general behavioural suppression in both compartments, and the administration of diazepam producing a selective disinhibition of behaviour in the white compartment. TST acute stress immediately before the test can increase the sensitivity to anxiolytic-like responses significantly (30). In the procedures of Crawley and Goodwin (4) and Costall et al. (11), after each trial, any faeces are removed, urine wiped up and the box wiped with water. In order to reduce any neophobic response to the test situation in Hascoe¨t and Bourin (20) procedure, the light/dark compartments are previously dirtied by mice other than those used during the test. Mice are always tested in a soiled apparatus and there is no cleaning between trials. Results have demonstrated that cleaning between trials masked the anxiolytic behaviour observed with 1 mg/kg of diazepam in soiled apparatus. It is suggested that measures would reflect more purely the influence of dark or bright areas on exploration activity rather than the influence of a clean new area smelling of detergent. A soiled apparatus removed, or at least reduced, the neophobia factor; rather the aversive stimulus is the novel environment. In the majority of experiments, mice are taken from a dark holding room in a dark container to the dark testing room where after a 1-h period of adaptation to the new environment, they are placed in the test box. In most experiments, mice are placed into the centre of the white, brightly lit area, and the operator withdraws from the room. If mice are removed from the holding room during their dark cycle in a dark container to the dark test room and placed into the test box where both sections are illuminated with a red light, they demonstrate a random activity within the two compartments. Thus, mice spent approximately 60–70% of their time in the white area and 30% in the black area, which appear to reflect the relative sizes of the two chambers. The pattern of exploratory behaviour continues, if red illumination in the white section was replaced with a relatively low level of white illumination of 10–240 lux. However, 400-lux illumination in the white area was sufficiently aversive to significantly reduce the time spent and the rears and line crossings in the white section, with corresponding increases in the black. Further the time taken for the initial movement of mice from the white section to the black was reduced by 50%; conversely animals placed initially into the black section delay their movement into the white section (11). It has been documented that many types of behaviour in nocturnal rodents show circadian variations with a higher activity during the dark period and a lower activity during the light one (50). The diurnal rhythm of the animals is thus an important factor. Therefore, the effects of cycle reversal and exposure to the apparatus on selected physiological measures have been investigated. Results (18, 51) suggest that reversal of the L:D cycle is

The Mouse Light–Dark Box Test

207

critical for exploratory behaviour in this test because behavioural changes are affected by plasma corticosterone levels, with the greatest sensitivity occurring during the dark cycle. Animals are maintained on a reversed 12:12 L:D cycle and are adapted to the reversed L:D cycle for at least 2–3 weeks prior to the commencement of the drug administration and anxiety testing. In the mouse black-and-white habituation test (11), with repeated daily testing, mice habituate to the paradigm such that they ‘‘learn’’ to locate the position of the opening, which allows movement from the aversive white section to the dimly lit dark section in which they spend more time and exhibit most exploratory behaviour. Consequently, they display a reduced latency to move from the white to the black section of the test box (52). The crucial factor is whether the environment sufficiently novel to elicit exploration and this may to some extent be strain and apparatus specific. Naive mice were tested in six trials at 2–3-days interval over a 2-week period by Blumstein and Crawley (19) to determine the effects of multiple uses of animals and to ascertain the test– retest reliability of the protocol and also to determine whether or not a learning phenomenon developed. They found that grouphoused mice can be routinely used repeatedly up to three times and that the interval between treatments and time of day did not appear to be critical in designing the test protocol. Repeated testing of 12 naive mice placed individually in the centre of the white chamber and exploratory measures of anxiety assessed were undertaken by Onaivi and Martin, (51) and they found that daily repeated testing was possible with a maximum of up to four times a week using naive mice. Barry et al. (53) observed habituation to the test environment, and Costall et al. (11) also found that on daily testing naive young adult animals habituated to the test system when compared to naive aged mice.

5. Scoring of Behaviour Crawley and Goodwin (4) described a model where benzodiazepines produced a facilitation of exploratory behaviour between a lighted open field and a dark enclosure. Mice placed in the white area (which they found aversive) would generally move around the periphery until they found an opening, at floor level, to enable access to the black compartment; this usually occurs within 7–12 s. The essential feature was the measurement of increased transitions between the light and dark chambers, the time spent in each compartment remaining the same. Costall et al. (11) found that increased exploratory behaviour was associated with an increased time spent in the light area, transitions between the two compartments remaining

208

Hascoe¨t and Bourin

unchanged. The exploratory activity depends on the baseline level in the control group and the behavioural data suggest that naive mice prefer the dark chamber, where they spend approximately 60% of their time. The entry into a chamber is defined as the placement of all four paws in the chamber. A correlation test performed between light and dark transitions and exploratory behaviours of mice in the light side of the two-chambered apparatus showed a significant correlation between the number of exploratory rearing, defined as directed sniffing with the forepaws extended vertically upon the sides of the chamber, and the number of light–dark transitions (22). Anxiolytics have been found to increase the parameters locomotion and time spent in the light zone, whereas anxiogenics decrease them (54), and mice placed into the brightly lit white area show a reduced latency in moving into the black section and an increase in time spent there, with markedly increased rears and line crossings in that area, with all such measures markedly decreased in the white section (16). A parameter suggested by Lapin (55) as an index of the effect of anxiogenics is the leanings out (or peeking-out) of the dark chamber by the mice, where a decrease in the rate of leanings out appeared to be a constant effect of standard anxiety-inducing drugs. A disinhibition of the suppression of behaviour is shown by the time taken for mice to move from or to the white section. Thus, control mice placed into the brightly lit white section would move rapidly into the black area: mice placed in the black area would show a delay in moving into the white. After the drug treatment, the apparent apprehension of remaining in or moving to the white area was abolished and indeed the delay for the animals to move from the white section or the speed of moving from the black section showed that aversion was reduced below that of normal mice. It has been suggested that putative ‘‘exploratory’’ activity measured in this paradigm is not simply a generalised motor effect, but rather is a function of the novelty of converting environments having different characteristics. A study by Griebel et al. (56) included the parameters such as time spent in the lit box, attempt at entry into the lit box followed by avoidance responses, which include stretched attend postures (the mouse stretches forward and retreats to original position), and total number of tunnel crossings. Diazepam increased the time spent and decreased the number of attempts. Decreasing aborted attempts at entry in the aversive areas is a profile which is consistent with an anxiolytic-like action (57). Increases in behavioural activity in the lit area are not due to a generalised increase in motor behaviour since total activity remains unchanged. Interesting results were obtained with the pure psychostimulant adrafinil (28, 54) and the stimulant antidepressant amineptine (11). No change in latency time was noticed.

The Mouse Light–Dark Box Test

209

Enhanced transitions and movements in both compartments were noted. The psychostimulant effect did not induce any increase in the time spent in the dark, showing that this parameter is specific for anxiolytic activity. The stimulant profile of action of adrafinil and amineptine in the black-and-white test is different from that of amphetamine, which is an anxiogenic psychostimulant. From data collected in the literature, it was difficult to compare the effects of movements in each compartment. Indeed, movements were expressed whatever be the time spent in the compartment under consideration. However, it seems obvious that mice that spend less time in one compartment demonstrate few movements and vice versa. It was therefore surprising to see false sedative effects or false psychostimulant effects. To avoid this problem, in a recent study (20), results of movements/exploratory behaviour in each area were expressed as a function of time spent in the compartment under consideration. This approach resulted in a more reliable idea of the indices of exploration and made the comparison between treatments easier. Another parameter, the latency time for the first passage from the light compartment to the dark one, has been used by some other authors (11, 26, 28). The real sense of this parameter is difficult to appreciate and is rarely discussed in literature. Two hypotheses of definition could be advanced. Increase in latency time could be the result of disinhibitory behaviour and decreased anxiolysis, where animals spend more time in exploring the white area. The other explanation is the influence of sedation, where animals are unable to move quickly to the dark compartment (58). Data suggest that the time mice spent in the lit area and behavioural activities such as locomotor and rearing behaviours may be more useful measures of the anxiolytic potential of a compound than transitions between the two compartments. In fact, the measurement found to be most consistent and useful for assessing anxiolytic-like action was the time mice spent in the lit area, this parameter providing the most consistent dose–effect results with the drugs (28). In the light/dark test, a range of active or passive behaviours like freezing, thimotaxis and risk assessment are almost invariably ignored in favour of a simple spatiotemporal index (59).

6. Drugs In the present section, the mentioned results were obtained after systemic administration of the drugs intraperitoneally or subcutaneously (Tables 11.2–11.8).

210

Hascoe¨t and Bourin

6.1. Effects of Ligands Acting on Benzodiazepines Sites

Many animal models of anxiety have been developed on the basis of their sensitivity to the benzodiazepines. Benzodiazepines are reliably detected in this paradigm (Table 11.2) as in many other animal models of anxiety; however, at high doses they can induce sedation detectable by a significant decrease in transitions (17). Some partial agonists are as well active on the test but not the antagonist flumazenil. The inverse agonists are mainly anxiogenic.

Table 11.2 Effects of ligands acting on the BZD sites in the light/dark test in mice: acute administration Drugs

Effects

References

Diazepam

++

(17, 20, 26, 51, 54, 60)

Flurazepam

++

(17)

Lorazepam

++

(61)

Alprazolam

++

(16, 20)

Chlordiazepoxide

++

(17, 62)

RO-4864 (peripheral)

0

(17)

Y 23684

++

(63)

RO16-6028

++

(64)

CGS 9896

++

(65)

Flumazenil

0

(15, 16, 18); Hascoe¨t and Bourin, unpublished

CGS-8216

0

(18, 66)

FG 7142

0

(20, 22, 62)

B CCM

–

(22, 54)

B CCE

–

(16, 54)

B CCP

–

(51)

Benzodiazepines (BZD)

BZD Partial agonists

BZD antagonists

BZD inverse agonists

+ anxiolytic-like effects; – anxiogenic-like effects; 0 no effect.

6.2. Effects of Drugs Acting on 5-HT1 Receptors

The recent introduction to clinical practice of the non-benzodiazepine anxiolytic buspirone and the discovery of new selective compounds have resulted in resurgence in the study of the

The Mouse Light–Dark Box Test

211

involvement of serotonin in anxiety. The classic 5-HT hypothesis of anxiety suggests that decreased anxiety is related to decreased activity in control 5-HT neurons and vice versa; however, paradoxical drug effects have often been found. A great number of studies found no evidence for anxiolytic or anxiogenic-like effects of drugs modulating 5-HT neurotransmission (67) (Table 11.3). The reasons for this variability in drug effect remain in part

Table 11.3 Effects of ligands acting on the serotonin system in he light/dark test in mice: acute administration Drugs

Effects

References

Buspirone

++

(11, 17, 20, 51, 54, 60)

8-OHDPAT

++, 0

(71); Hascoe¨t and Bourin, unpublished data

Gepirone

0

(72)

Ipsapirone

++

(72)

MDL 73005EF

++

(72, 73)

Flesinoxan

++

(74)

Nan 190

0

Hascoe¨t and Bourin, unpublished data

Pindolol

0

(71, 75)

++

(76); Hascoe¨t and Bourin, unpublished data

mCPP

0

(77, 78)

DOI

0–

(14, 51, 78)

Ro 60-0175

0

(78)

BW 723C86

0

(78)

0

(78)

Ritanserin

+0–

(24, 29, 79)

Ketanserin

–

(78)

5-HT1A receptor agonists

5-HT1A receptor antagonists

5-HT1B receptor agonists Anpirtoline 5-HT2 receptor agonists

5-HT2A receptor antagonists SR 46349B 5-HT2a/c receptor antagonists

(continued)

212

Hascoe¨t and Bourin

Table 11.3 (continued) Drugs

Effects

References

RS 10-2221

0

(78)

SDZ SER082

0

(78)

SB 206553

0

(78)

0

(25); Hascoe¨t and Bourin unpublished data

Ondansetron

+, 0

(25, 34, 80); Hascoe¨t et al. unpublished data

MDL 72 222

+, 0

(11, 51, 81)

Zacopride

+

(60)

DAU 6215

+

(82)

Granisetron

+

(11, 79)

5-HT2c receptor antagonists

5-HT3 receptor agonists 2-methyl-5-HT 5-HT3 receptor antagonists

+ anxiolytic-like effects; – anxiogenic-like effects; 0 no effect

unknown, but certainly include some factors (species difference, gender of animals, environment in which a test is conducted). Variations in effects might also reflect differences in the degree to which the models themselves represent fear or anxiety (1, 68, 69). The most convincing explanation of these discrepancies has recently emerged from several papers of Handley and McBlane (1, 69, 70) who suggested that there is more than one 5-HT mechanism involved in anxiety models. It is obvious that all models are not equivalent. Thus, models based on spontaneous responses, such as exploration tests like the light/dark test, may reflect a type of anxiety linked with uncontrollable stress (‘‘depressive anxiety’’) as animals are exposed by force to a novel and/or aversive environment from which they cannot escape, while those based on conditioning (e.g. Vogel’s conflict test) may reflect a type of anxiety associated with controllable aversive events (‘‘anticipatory anxiety’’). 6.3. Effect of Acute Administration of 5-HT2 Receptor Agonists

DOI significantly reduced the movements in the dark compartment at doses of 2, 4 and 8 mg/kg after i.p. acute administration, and the latency time for passing from the lit compartment to the dark compartment at doses of 1 and 8 mg/kg. BW 723C86 had no effect on mouse behaviour in the light/dark paradigm for the doses tested. RO 60-0175 significantly reduced transitions between compartments and movement in the lit compartment at a dose of 4 mg/kg.

The Mouse Light–Dark Box Test

213

mCPP induced a significant decrease in the movements in the lit compartment at a dose of 0.5 mg/kg. 6.4. Effect of Acute Administration of 5-HT2 Receptor Antagonists

On the indices of anxiety measured (transitions and per cent time in the dark compartment) only ketanserin administration induced a reduction in the number of transitions between compartments along with a significant reduction in the movements in the dark compartment. Neither RS 10-2221, SDZ SER082 nor SB 206553 administration had an effect on mouse behaviour for all doses tested in the light/dark paradigm. SR 46349B administration significantly increased movement in the dark compartment. So ketanserin administration induced an anxiogenic-like effect, while the specific 5-HT2A receptor antagonist SR 46349B was without an effect in this paradigm. The effects observed with ketanserin may potentially involve adrenergic or histaminergic receptor activity (83) or the combined antagonism of both 5-.HT2A and 5-HT2C receptors. The literature has provided little evidence of anxiolytic-like effects of 5-HT2 receptor ligands in the mouse L/D choice (84, 85, 86, 87, 88), perhaps indicating that the 5-HT2 receptor is not implicated in that test and thus the model is incapable of detecting their potential anxiolytic-like effects. It has been suggested that the 5-HT3 receptor rather than the 5-HT2 receptor may be involved in the fear provoked by the L/D paradigm and that the nucleus accumbens (89) or the amygdala (90) may be involved in mediating the disinhibitory effects of 5-HT3 receptor antagonists.

6.5. Effects of Acute Administration on 5-HT3 Receptors

Anxiolytic-like effects of 5-HT3 receptor antagonists have been established in selected test procedures (91). In these, only the light–dark test choice paradigm in mice constantly showed anxiolytic-like action of these agents (67).

6.6. Effects of Acute Administration of Antidepressants

A growing interest for the use of antidepressant drugs in the treatment of anxiety disorders has led to studying their potential anxiolytic-like effects in various animal models. Data suggest that moclobemide (MOC), a reversible inhibitor of type-A monamine-oxidase (MAO) enzyme, significantly reduced anxiogenic-like behaviour in the light/dark test, whereas selegiline, an irreversible and selective MAO-B inhibitor, showed a lack of anxiolytic-like effect (92). Animal studies and clinical findings suggest that SSRIs, when given acutely do not reduce experimental anxiety as the symptoms of GAD and panic disorders (93, 94, 95), which are in contrast to results found by Hascoe¨t et al. (33) (Table 11.4). Indeed studies have frequently reported that acute administration of SSRIs elicits anxiogenic-like responses (93, 96). Sanchez and Meier (97) studying the profile of five SSRIs found that citalopram produced a mixed anxiogenic/ anxiolytic-like response in the light/dark test in rats and

214

Hascoe¨t and Bourin

Table 11.4 Effects of antidepressants in the light/dark test in mice (acute administration) Drugs

Effects

References

Paroxetine

+

(33)

Amitriptyline

–

(11)

Citalopram

–

(98)

Imipramine

+

(14, 16, 31)

0

(51)

Clorgyline

0

(17)

Dothiepin

+

(31)

Mianserin

0

(31, 67)

Fluoxetine

0

(31)

Maprotoline

0

(31)

Viloxazine

0

(31)

Moclobemide

+

(92)

Selegiline

0

(92)

+ anxiolytic-like effects; – anxiogenic-like effects; 0 no effect

paroxetine induced an anxiogenic-like response at low doses. The mechanism of action of SSRIs in anxiety is yet not understood, but as many of these molecules possess affinity for the 5-HT2 receptors, the various receptor subtypes may be implicated in the anxiolytic properties. 6.7. Effects of Acute Administration of Antipsychotics

The administration of cyamemazine at doses of 0.375 and 0.5 mg/kg significantly decreased the percentage time spent in the dark (99) (Table 11.5). However, the dose of 0.5 mg/kg

Table 11.5 Effects of antipsychotics in the light/dark test in mice (acute administration) Drugs

Effects

References

Clozapine

+

Bourin et al., unpublished data

Cyamemazine

+

(99) ‘‘acute but not repeated doses’’

Risperidone

0

Bourin et al., unpublished data

+ anxiolytic-like effects; – anxiogenic-like effects; 0 no effect

The Mouse Light–Dark Box Test

215

reduced the movements in the light compartment, the number of transitions between the compartments and the latency time to leave lit compartment. As well as dramatically decreasing the movements in the dark compartment, however not in a significant manner. Clozapine administration decreased the percentage time spent in the dark compartment at the dose of 0.015 mg/kg. The acute administration of both antipsychotics, cyamemazine and clozapine showed a clear anxiolytic-like effect in the L/D paradigm (risperidone failed to do it because of its high sedation potential in mice). Both these compounds have antagonistic effects at the 5-HT3 receptor level and the L/D test is extremely reliable in detecting positive effects of 5-HT3 antagonists (100, 101, 102). 6.8. Effects of Acute Administration of Psychostimulants

In a study by Hascoe¨t et al. (20) (Table 11.6), amphetamine was found to be stimulant and anxiogenic producing a significant dramatic increase in the time spent in the dark area, in accordance with the results of Pellow et al. (103) who found an anxiogenic action in rats using the elevated plus maze. On the other hand, Young and Johnson (14) demonstrated that in the light/dark test, amphetamine did not significantly change the time spent in the dark area.

Table 11.6 Effects of psychostimulant drugs in the light/dark test in mice Drugs

Effects

References

Adrafinil

0

(20)

Amineptine

0

(20)

Caffeine

0 –

(16, 20) (104)

Amphetamine

–

(20, 51)

+ anxiolytic-like effects; – anxiogenic-like effects; 0 no effect

6.9. Effects of Acute Administration of Neuropeptides Receptor Ligands

Cholecystokinin 2 receptor antagonists are a novel group of agents with a potential for use in the treatment of anxiety (56). Experiments with CCK-2 receptor agonists, such as CCK-4, have shown them to possess anxiogenic-like effects (105, 106) (Table 11.7). The development of highly specific and potent antagonists for CCK receptors have been discovered and were found to abolish the anxiogenic-like effects of CCK-2 receptor stimulation in rodents (56). The selective CCK-2 receptor antagonists PD

216

Hascoe¨t and Bourin

Table 11.7 Effects of neuropeptides receptors ligands in the light/dark test in mice Drugs

Effects

References

CCK 4

0–

(20, 113)

BC 197

–

(114)

CCK8 s

–

(113)

PD 134308

+

(107)

PD 135138

+

(27)

CI-988 (L-365,260)

+

(87, 115, 116)

CP-154,526

+

(110)

SSR125543A

0

(117)

Antalarmin

0

(117)

CP-96,345

–

(111)

RP 67580

0

(111)

+

(97)

CCK2 receptor agonist

CCK 2 receptor antagonists

CRF1 receptor antagonists

NK1 receptor antagonists

Opioid receptor ligands Sigma2 ligand Lu 28-179

+ anxiolytic-like effects; – anxiogenic-like effects; 0 no effect

134308 and PD 135138 were found to be as effective as diazepam in inhibiting aversive responding in the absence of sedation, muscle relaxation or withdrawal phenomenon (27, 107). Yet any of these compounds were found to be anxiolytic in human (108). Corticotrophin-releasing factor, CRF, is a neuropeptide that plays a prominent role in the endocrine, autonomic, behavioural and immune responses to stress, through its action on the major physiological regulator of the hypothalamic–pituitary–adrenal (HPA) axis (109). Treatment with antagonists that selectively block CRF1 receptor action was shown to promote anxiolytic responses in the mouse light–dark box test (110). The non-peptide NK1 antagonists, RP 67580, and (2S, 3S)CP-96,345, the NK1 receptor-selective enantiomer of the racemic compound, were tested in Swiss albino mice in the black-and-white

The Mouse Light–Dark Box Test

217

box behavioural paradigm. Both qualitatively and quantitatively, (2S, 3S)-CP-96,345 produced the same behavioural effects as the racemic compound. In contrast, RP 67580 decreased exploratory behaviour only in the white section, whereas crossings and rearings in the black section were not changed. In addition, RP 67580 decreased transitions. While the observed changes induced by CP96,345 are caused by sedation and motor impairment, the effects of RP 67580 might be due to sedation plus an additional anxiogenic effect (111). The anxiolytic potential of the selective sigma 2 ligand 1-[4-[1-(4-Fluorophenyl)-1H-indol-3-yl]-1-butyl] spiro [isobenzofuran-1(3H), 4-piperidine] (Lu 28-179) was assessed in various animal models of anxiety in rodents. Lu 28-179 facilitated the exploratory behaviour of mice in the black-and-white two-compartment box over a large dose range (112). 6.10. Effect of Cannabinoids

There is growing evidence that the endocannabinoid system is implicated in the control of emotional behaviour via CB1 receptors. Rutkowska et al. (118) have investigated the influence of the cannabinoid system modulation produced by CB receptors (Table 11.8). The CB1 CB2 agonist induced anxiogenic-like effect in the mouse light–dark test using BALB/c mice. The CB1 receptor agonist, AM281, and the anandamide transporter inhibitor, AM 404, were without effect, while the inhibitor of anandamide hydrolysis, AACOCF3, induced anxiolytic-like effect.

Table 11.8 Effects of cannabinoid drugs in the light/dark test in mice Drugs

Effects

References

WIN 55,212-2

–

(118)

AM 281

0

(118)

AACOCF3

+

(118)

AM 404

0

(118)

+ anxiolytic-like effects; – anxiogenic-like effects; 0 no effect

7. Conclusion In conclusion, the black-and-white test may be useful to predict anxiolytic-like or anxiogenic-like activity in mice. It has the advantages of being quick and easy to use, without the prior training of animals; food and water deprivation is unnecessary and natural

218

Hascoe¨t and Bourin

stimuli are used. Transitions have been reported to be an index of activity exploration because of habituation over time and the time spent in each compartment to be a reflection of aversion (15), but the best measures seem to be the percentage of time spent and the movements/exploratory behaviour in each compartment (20). The light/dark transition test is limited by its ability to yield false-positive results due to a drug’s ability to increase general activity. As with many experimental protocols, drugs that effect general motor function will affect light–dark performance. Preliminary screening of locomotor activity (such as an open field or an actimeter test) appears to be necessary and sufficient for eliminating false positives. It remains difficult to estimate the value of the test in terms of effect size compared to the other tests, which predictability regarding the human anxiety disorders. It has been suggested that some animal models based on spontaneous behaviour or ethologically based models (9) (like the light/dark test) may be more sensitive to the behavioural responses than conditioned paradigms (119). Ethological models, however, present individual differences, variable behavioural baseline levels. Ethological stimuli are diverse in nature. It is interesting to notice that there exist differences in strain sensitivity between the elevated plus maze (EPM) and the light–dark test (39). We must emphasise the importance of choosing the appropriate strain to study the pharmacological effects of putative anxiolytic drugs (35). Differences also exist concerning drugs responding. For example 5-HT3 antagonist rather than 5HT2 ligands demonstrated anxiolytic-like effect in the light–dark test, contrary to the EPM (78). Often faced with the lack of reproducibility and sensibility of non-BZD drugs in animals models of anxiety, tests have been subjected to an abundance of variation and a plethora of parameters of measurement. This would also suggest problems with the test’s sensitivity and reproducibility to anxiolytic and potential anxiolytic compounds in other laboratories also. Standardisation represents a way to ensure the reproducibility of both qualitative and quantitative aspects of a measure and is suggested to lead to more reproducible or interpretable results of complex experiments performed in different laboratories or within the same laboratory. References 1. Handley SL. 5-Hydroxytryptamine pathways in anxiety and its treatment. Pharmacol Ther 1995; 66:103–48. 2. Gray J. The neuropsychology of anxiety. Oxford: Oxford University Press, 1982. 3. Rodgers RJ, Cao BJ, Dalvi A, Holmes A. Animal models of anxiety: an ethological

perspective. Braz J Med Biol Res 1997; 30:289–304. 4. Crawley JN, Goodwin FK. Preliminary report of a simple animal behaviour for the anxiolytic effects of benzodiazepines. Pharmacol Biochem Behav 1980; 13:167–70.

The Mouse Light–Dark Box Test 5. File SE. Usefulness of animal models with newer anxiolytics. Clin Neuropharmacol 1992; 15(suppl 1):525A–526A. 6. Njung’e K, Handley SL. Evaluation of marble-burying behaviour as a model of anxiety. 1991; 38:63–7. 7. Treit D. A comparison of anxiolytic and nonanxiolytic agents in the shock probe buring test for anxiolytics. Pharmacol Biochem Behav 1990; 36:203–5. 8. De Vry J, Benz U, Schreiber R, Traber J. Shock-induced ultrasonic vocalization in young adults rats : a model for testing putative antianxiety drugs. Eur J Pharmacol 1993; 249:331–39 9. Lister RG. Ethologically-based models of anxiety disorders. Pharmac Ther 1990; 46:321–40. 10. Belzung C, Griebel G. Measuring normal and pathological anxiety-like behaviour in mice: a review. Behav Brain Res 2001; 125:141–9. 11. Costall B, Jones BJ., Kelly ME, Naylor RJ, Tomkins DM. Exploration of mice in a black and white box: validation as a model of anxiety. Pharmacol Biochem Behav 1989; 32:777–85. 12. Gao B, Cutler MG. Effect of acute administration of the 5-HT3 receptor antagonist, BRL 46470A, on the behavior of mice in a two compartment light-dark box and during social interactions in their home cage and an unfamiliar neutral cage. Neuropharmacology 1992; 31:743–8. 13. Imaizumi M, Onodera K. The behavioural and biochemical effects of thioperamide, a histamine H3-receptor antagonist, in a light/dark test measuring anxiety in mice. Life Sciences 1993; 53:1675–83. 14. Young R, Johnson DN. A fully automated light/dark apparatus useful for comparing anxiolytic agents. Pharmacol Biochem Behav 1991; 40:739–43. 15. Belzung C, Misslin R, Vogel E, Dodd RH , Chapouthier G. Anxiogenic effects of methyl-b-carboline-carboxylate in a light/ dark choice situation. Pharmacol Biochem Behav 1987; 28:29–33. 16. Shimada T, Matsumoto K, Osanai M, Matsuda H, Terasawa K, Watanabe H. The modified light/dark transition test in mice: evaluation of classic and putative anxiolytic and anxiogenic drugs. Gen Pharmac 1995; 26:205–10. 17. Crawley JN. Neuropharmacologic specificity of a simple model for the behavioural

18.

19.

20.

21.

22.

23.

24.

25.

26.

27.

28.

29.

219

actions of benzodiazepines. Pharmacol Biochem Behav 1981; 15:695–99. Crawley JN. Exploratory behaviour models of anxiety in mice. Neurosci Biobehav Rev 1985; 9:37–44. Blumstein LK, Crawley JN. Further characterisation of a simple, automated exploratory model for the anxiolytic effects of benzodiazepines. Pharmacol Biochem Behav 1983; 18:37–40. Hascoe¨t M, Bourin M. A new approach to the light/dark procedure in mice. Pharmacol Biochem Behav 1998; 60:645–53. Crawley JN, Davis LG. Base line exploratory activity predicts anxiolytics responsiveness to diazepam in five mouse strains. Brain Res Bull 1982; 8:609–12. Crawley JN, Skolnick P, Paul SM. Absence of intrinsic antagonist actions of benzodiazepine antagonist on an exploratory model of anxiety in the mouse. Neuropharmacology 1984; 5:531–7. Crawley JN, Belknap JK, Collins A, Crabbe JC, Frankel W, Henderson N, Hitzeman RJ, Maxson SC, Miner LL, Silva AJ, Wehner JM, Wynshaw-Boris A , Paylor R. Behavioural phenotypes of inbred mouse strains: implication and recommendations for molecular studies. Psychopharmacology 1997; 132:107–24. Costall B, Kelly ME, Naylor RJ,Onaivi ES. Actions of buspirone in a putative model of anxiety in the mouse. J Pharm Pharmacol 1988a; 40:494–500. Costall B, Domeney AM, Gerrard PA, Kelly ME, Naylor RJ. Effects of the 5-HT receptors antagonist GR38032F, ICS 205-930 and BRL 43694 in tests for anxiolytic activity. Br J Pharmacol 1988b; 93(Suppl):195P. Costall B, Jones BJ, Kelly ME, Naylor RJ, Onaivi ES, Tyers MB. Sites of action of ondansetron to inhibit withdrawal from drugs of abuse. Pharmacol Biochem Behav 1990; 36:97–104. Costall B, Domeney AM, Hughes J, Kelly ME, Naylor RJ, Woodruff GN. Anxiolytic effects of CCK-B antagonists. Neuropeptides 1991; 19:65–73. Costall B, Domeney AM, Kelly ME, Tomkins DM, Naylor RJ, Wong EHF, Smith WL, Whiting RL, Eglen R. The effect of the 5-HT3 receptor antagonist, RS 42358197, in animal models of anxiety. Eur J Pharmacol 1993; 234:91–9. Gao B and Cutler MG. Effect of acute and subchronic administration of ritanserin on

220

30.

31.

32.

33.

34.

35.

36.

37.

38.

39.

40.

41.

Hascoe¨t and Bourin the social behaviour of mice. Neuropharmacology 1993; 32:265–72. Belzung C, Le Pape G. Comparison of different behavioural test situations used in Psychopharmacology for measurement of anxiety. Physiol Behav 1994; 3:623–28. Bourin M, Redrobe JP, Hascoe¨t M, Baker GB, Colombel MC. A schematic representation of the psychopharmacological profile of antidepressants. Prog Psychopharmacol Biol Psychiat 1996; 20:1389–402. Hascoe¨t M, Colombel MC, Bourin M. Influence of age on behavioural response in the light-dark paradigm. Physiol Behav 1999; 66:567–70. Hascoe¨t M, Bourin M, Nic Dhonnchadha. ´ . The influence of buspirone and its metaA bolite1-PP on the activity of paroxetine in the mouse light dark paradigm and four plates test. Pharmacol Biochem Behav 2000b; 67:45–53. Young R, Johnson DN. Comparative effects of zacopride, GR 38032F, Buspirone and diazepam in the mouse light/dark exploratory model. Soc Neurosci Abstr 1988;14:207. Bouwknecht J.A., Paylor R. Behavioral and physiological mouse assays for anxiety: a survey in nine mouse strains. Behav Br Res 2002; 136:489–501. Gullet P.V., Chapouthier G. Intermale aggression and dark/light preference in ten inbred mouse strains. Behav Br Res 1996; 77:211–13. Mc Ilwain KL, Merriweather MY, Yuva-Paylor L.A, Paylor R. The use of behavioural test batteries: effects of training history. Physiol Behav 2001; 73:705–17. Voikar V, Koks S, Vasar E, Rauvala H. Strain and gender differences in the behaviour of mouse lines commonly used in transgenic studies. Physiol Behav 2001; 72:271–81. Griebel G., Belzung C., Perrault G. and Sanger D.J. Differences in anxiety-related behaviours and in sensitivity to diazepam in inbred and outbred strains of mice. Psychopharmacology 2000; 148:164–70. Chapouthier G, Bondoux D, Martin B, Desforges C and Launay J.M. Genetic difference in sensitivity to beta carboline: evidence for the involvement of brain benzodiazepine receptors. Brain Res 1991; 553:342–46. Singer JB, Hill AE, Nadeau JH, Lander ES. Mapping quantitative trait loci for anxiety in chromosome substitution strain of mice. Genetics 2005; 169:855–62.

42. Takahashi A, Nishi A, Ishii A, Shiroishi T, Koide T. Systematic analysis of emotionality in consomic mouse strains establish from C57BL/6 J and wild-derived MSM/Ms. Genes Brain Behavior. Science direct on line (juil 2008). 43. Rodgers RJ, Boullier E, Chatzimichalaki P, Cooper GD, Shorten A. Contrasting phenotypes of C57BL/6JOlaHsd, 129SvHsd and 129/SvEv mice in two explorationbased test of anxiety-related behavior. Physiol Behav 2002; 77:301–10. 44. Mombereau C, Kaupmann K, Froesti W, Sansig G, Van der Putten H, Cryan JF. Genetic and pharmacologic evidence of a role for GABA (B) receptors in the modulation of anxiety and antidepressant-like behaviour. Neuropsychopharmacol 2004; 29:1050–62. 45. Jacobson H., Bettler B., Kaupmann K., Cryan J.F. Behavioral evaluation of mice in GABA(B1) receptor isoform in test of unconditioned anxiety. Psychopharmacol 2007; 190:541–3. 46. Raud S, Innos J, Abramov U, Reimets A, Soosaar A, Matsui T, Vasar E. Targeted invalidation of CCK2 receptors gene induces anxiolytic-like action in light-dark exploration, but not in fear conditioning test. Psychopharmacoly 2005; 181:347–57. 47. Pletnikov MV, Storozheva ZI, Sherstnev VV. Relationship between memory and fear: Developmental and pharmacological studies. Pharmacol Biochem Behav 1996; 54:93–8. 48. Hoggs S. A review of the validity and variability of the elevated plus maze as an animal model of anxiety. Pharmacol Biochem Behav 1996; 54:21–30. 49. Sanchez C. 5HT1A receptors play an important role in modulation of behaviour of rats in a two-compartment black and white box. Behav Pharmacol 1996; 7:788–97. ´ 50. Gorka Z, Maj J. Effects of repeated treatment with antidepressant drugs on the 24 hour behaviour in the light-dark synchronised mice. Pol J Pharmacol Pharm 1986; 38:493–9. 51. Onaivi ES, Martin BR. Neuropharmacological and physiological validation of a computer-controlled two compartment black and white box for the assessment of anxiety. Prog Neuropsychopharmacol Biol Psychiat 1989; 13:963–76. 52. Barnes NM, Costall B, Kelly ME, Onaivi ES, Naylor RJ. Ketotifen and its analogues reduce aversive responding in the rodent.

The Mouse Light–Dark Box Test

53.

54.

55.

56.

57.

58.

59. 60.

61.

62.

63.

Pharmacol Biochem Behav 1990; 37:785–93. Barry JM, Costall B, Kelly ME, Naylor RJ, Onaivi E.S. A simple habituation test in the mouse. Br J Pharmacol 1987; 92:651P. Imaizumi M, Suzuki T, Machida H, Onodera K. A fully automated apparatus for a light/dark test measuring anxiolytic or anxiogenic effects of drugs in mice. Jpn. J Psychopharmacol 1994; 14:83–91. Lapin IP. A decreased frequency of peeking out from the dark compartment-the only constant index of the effect of anxiogenes on the behaviour of mice in light-darkness chamber. Zh Vyssh Nerv Deiat IM. I. P. Pavlova 1999; 49:521–6. Griebel G, Perrault GH, Sanger DJ. CCK receptor antagonists in animal models of anxiety: comparison between exploration tests, conflict procedures and a model based on defensive behaviours. Behav Pharmacol 1997a; 8:549–60. Griebel G, Lanfumey L, Blanchard DC, Rettori MC, Guaardiola-Lemaitre B, Hamon M, Blanchard RJ. Preclinical profile of the mixed 5-HT 1A/5-HT2A receptor antagonist S21357. Pharmacol. Biochem Behav 1996; 54:509–16. Rodgers RJ, Shepherd JH. Influence of prior maze experience on behaviour response to diazepam in the elevated plus maze and light/dark test of anxiety in mice. Psychopharmacology 1993; 113:237–42. Rodgers RJ. Animal models of ‘anxiety’: where next? Behav Pharmacol 1997; 8:477–96. Young R, Johnson DN. Comparison of routes of administration and time course effects of zacopride and buspirone in mice using an automated light/dark test. Pharmacol Biochem Behav 1991b; 40:733–37. De Angelis L. The anxiogenic-like effects of pentylenetetrazole in mice treated chronically with carbamazepine or valproate. Meth Find Exp Clin Pharmacol 1992; 14:767–71. Kilfoil T, Michel A, Montgomery D, Whiting RL. Effects of anxiolytic and anxiogenic drugs on exploratory activity in a simple model of anxiety in mice. Neuropharmacology 1989; 28:901–905. Yasumatsu H, Morimoto Y, Yamamoto Y, Takehara S, Fukuda TF, Nako T, Setoguchi M. The pharmacological properties of Y-23684, a benzodiazepine receptor partial agonist. Br. J. Pharmacol 1994; 11:1170–8.

221

64. Belzung C, Misslin R, Vogel E. Behavioural effects of the benzodiazepine receptor partial agonist RO 16-6028 in mice. Psychopharmacology 1989; 97:388–91. 65. Smith CB, Crawley JN. Anxiolytic action of CGS 9896 on mouse explorator behaviour. Eur J Pharmacol 1986; 132:259–262. 66. Wieland S, Lan NC, Mirasedeghi S, Gee KW. Anxiolytic activity of the progesterone metabolite 5-pregnan-3-ol-20-one. Br Res 1991; 565:263–68. 67. Griebel G. 5-Hydroxytryptamine-interacting drugs in animal models of anxiety disorders: more then 30 years of research. Pharmacol Ther 1995; 65:319–95. 68. Barret JE, Vanover KE. 5-HT receptors as targets for the development of novel anxiolytic drugs/ models, mechanisms and future directions. Psychopharmacology 1993; 112:1–12. 69. Handley SL, McBlane JW. Serotonin mechanisms in animal models of anxiety. Braz J Med Biol Res 1993; 26:1–13. 70. Handley SL, McBlane JW, Critchley MA, Njung’e K. Multiple serotonin mechanisms in animal models of anxiety: environmental, emotional and cognitive factors. Behav Brain Res 1993; 58:203–10. 71. Lopez-Rubalcava C, Saldivar A, FernandezGuasti A. Interaction of GABA and serotonin in the anxiolytic action of diazepam and serotonergic anxiolytics. Pharmacol Biochem Behav 1992; 43:433–40. 72. Bill DJ, Fletcher A, Knight M. Actions of 5HT1A ligands and standard anxiolytics on mouse exploratory behaviour in a two compartment light: dark arena. Br. J. Pharmacol 1989; 98 (Suppl):679P. 73. Misslin R, Griebel G, Saffroy-Spittler M, Vogel E. Anxiolytic and sedative effects of 5-HT1A ligands, 8-OHDPAT and MDL 73005EF, in mice. Neuroreport 1990; 1:267–70. 74. Schipper J, Van Der Poel AM, Mos J, Van Der Heyden JAM, Olivier B. Flesinoxan: anxiolytic activity in animal models. In SEROTONIN 1991, 5-hydroxytryptamineCNS receptors and brain function, p. 138, Birmingham, 14–17th July 1991. 75. Fernandez-Guasti A, Lopez-Rubalcava C. Evidence for the Involvement of the 5HT1A receptor in the anxiolytic action of indorate and ipsapirone. Psychopharmacology 1990; 107:354–358. 76. Metzenauer P, Barnes NM, Costall B, Gozlan H, Hamon M, Kelly ME, Murphy DA, Naylor RJ. Anxiolytic like action of

222

77.

78.

79.

80.

81.

82.

83.

84.

85.

86.

87.

88.

Hascoe¨t and Bourin anpirtoline in a mouse light dark aversion paradigm. NeuroReport 1992; 3:527–29. Griebel G, Misslin JR, Pawlowski M, Vogel E. m- Chlorophenylpiperazine enhances neophobic and anxious behaviour in mice. NeuroReport 1991; 2:627–9. Nic Dhonnchadha BA, Bourin M, Hascoe¨t M. Anxiolytic-like effects of 5-HT2 ligands on three mouse models of anxiety Behav Brain Res 2003; 140:203–14. Barnes NM, Cheng CHK, Costall B, Ge J, Kelly ME, Naylor RJ. Profiles of interaction of R (+)/S (–)-zacopride and anxiolytic agents in a mouse model. Eur J Pharmacol 1992; 218:91–100. Tyers MB, Costall B, Domeney A, Jones BJ, Kelly M.E, Naylor RJ, Oakeley N. The anxiolytic activities of 5-HT3 antagonists in laboratory animals. Neurosci Lett 1987; 29:S68. Bill DJ, Fletcher A, Glenn BD , Knight M.. Behavioural studies of WAY 100289, a novel 5-HT3 receptor antagonist, in two animal models of anxiety. Eur J Pharmacol 1992; 218:327–34. Borsini F, Brambilla A, Cesana R, Donetti A. The effects of DAU 6215, a novel 5HT3 antagonist in animal models of anxiety. Pharmacol Res 1993;27:151–64. Kennett GA, Pittaway K, Blackburn TP. Evidence that 5-HT2C receptor antagonists are anxiolytic in the rat Geller-Seifter model of anxiety. Psychopharmacology 1994; 114:90–96. Cheng CHK, Costall B, Kelly ME, Naylor RJ. Actions of 5-hydroxytryptophan to inhibit and disinhibit mouse behaviour in the light/dark test. Eur J Pharmacol 1994; 255:39–49. Sanchez C. Serotonergic mechanisms involved in the exploratory behaviour of mice in a fully automated two-compartment black and white test box. Pharmacol Toxicol 1995; 77:71–8. Costall B, Naylor RJ. Behavioural interactions between 5-hydroxytryptophan neuroleptic agents and 5-HT receptor antagonists in modifying rodent responding to aversive situations. Br J Pharmacol 1995; 116:2989–99. Costall B, Naylor RJ. The influence of 5HT2 and 5-HT4 receptor antagonists to modify drug induced disinhibitory effects in the mouse light/dark test. Br J Pharmacol 1997; 122:1105–8. Griebel G, Perrault G, Sanger DJ. A comparative study of the effects of selective and

89.

90.

91. 92.

93.

94.

95.

96.

97.

98.

99.

non-selective 5-HT2C receptor subtype antagonists in rat and mouse models of anxiety. Neuropharmacology 1997b; 36:793–802. Higgins GA, Jones BJ, Oakley NR, Tyers MB. Evidence that the amygdala is involved in the disinhibitory effects of 5-HT3 receptor antagonists. Psychopharmacology 1991; 104:545–51. Stefanski R, Palejo W, Bidzinski A, Kostowski W, Plaznik A. Serotonergic innervation of the hippocampus and nucleus accumbens septi and the anxiolytic-like action of the 5-HT3 receptor antagonists. Neuropharmacology 1993; 32:987–93. Kilpatrick GJ, Bunce KT, Tyers MB. 5-HT3 receptors. Med Res Rev. 1990; 10:441–75. De Angelis L, Furlan C. The anxiolytic like properties of two selective MAOIs, moclobemide and selegiline, in a standard and an enhanced light/dark aversion test. Pharmacol Biochem Behav 2000; 65:649–53. Bodnoff SR, Suranyi-Cadotte B, Quirion R, Meaney MJ. A comparison of the effects of diazepam versus several typical and atypical antidepressant drugs in an animal model of anxiety. Psychopharmacology 1989; 97:277–79. Lightowler S, Kennett GA, Williamson IJ, Blackburn TP, Tulloch IF. Anxiolytic like effect of paroxetine in a rat social interaction test. Pharmacol Biochem Behav 1994; 49:281–5. Matto V, Allikmets L, Harro J. The mechanism of anxiogenic-like effect of antidepressants on exploratory behaviour in rats. Pharmacol Toxicol 1995; 76(suppl.3):53. File SE, Pellow S, Chopin P. Can animal test of anxiety detect anti-panic compounds? Neurosci Abstr 1985; 11:273. Sanchez C, Meier E. Behavioural profiles of SSRIs in animal models of depression, anxiety and aggression: are they all alike? Psychopharmacology 1997; 129:197–205. Griebel G, Moreau JL, Jenck F, Martin R, Misslin JR. Acute and chronic treatment with 5-HT reuptake inhibitors differentially modulate emotional responses in anxiety models in rodents. Psychopharmacology 1994; 113:463–70. Bourin M, Nic Dhonnchadha BA, Colombel MC, Dib M, Hascoe¨t M. Cyamemazine as an anxiolytic drug on the elevated plus maze and light/dark paradigm in mice. Behav Br Res 2001; 124:87–95.

The Mouse Light–Dark Box Test 100. Costall B, Naylor R. Anxiolytic potential of 5-HT3 receptor antagonists. Pharmacol Toxicol 1992; 70:157–62. 101. File SE, Andrews N, Hoggs S. New developments in animal tests of anxiety. Advances in the Neurobiology of Anxiety disorders.Ed. Westenberg H.G.M., Den Boer J.A. and Murphy D.L., 1996, pp.61–79.John Wiley and Sons Ltd. 102. Olivier B, Van Wijngaarden I, Soudijn W. 5HT3 receptor antagonists and anxiety; a preclinical and clinical review. Eur Neuropsychopharm 2000; 10:77–95. 103. Pellow S, Chopin P, File SE, Briley M. Validation of open/closed arm entries in an elevated plus-maze as a measure of anxiety in the rat. J Neurosci Meth 1985; 14:145–67. 104. Imaizumi M, Miyazaki S, Onodera K. Effects of xanthine derivatives in a light/ dark test in mice and contribution of adenosine receptors. Meth Find Exp Clin Pharmacol 1994; 16:639–44. 105. Van Megen HJGM, Westenberg HGM, Den Boer JA, Kahn RS. Cholecystokinin in anxiety. Eur Neuropsychopharm 1996; 6:263–80. 106. Bourin M, Baker GB, Bradwejn J. Neurobiology of panic disorder. J Psychosis Res 1998a; 44:161–80. 107. Belzoni C, Pineal N, Bautzen A, Muslin R. PD135158, a CCK-B antagonist, reduces ‘‘state,’’ but not ‘‘trait’’ anxiety in mice. Pharmacology Brioche Behave 1994; 49:433–36. 108. Bourin M. Cholecystokinin as a target for neuropsychiatric drugs. Drug News Perspect 1998b; 11:342–49. 109. Arborelius L, Owens MJ, Bissette G, Plotsky PM, Nemeroff CB. The role of corticotrophin-releasing factor in depression and anxiety. J Endocrinol 1999; 160:1–12 110. Griebel G, Perrault G, Sanger DJ. Characterization of the behavioral profile of the non-peptide CRF receptor antagonist CP154,526 in anxiety models in rodents. Comparison with diazepam and buspirone. Psychopharmacology 1998; 138:55–66. 111. Zernig G, Troger J, Saria A. Different behavioral profiles of the non-peptide substance P (NK1) antagonists CP-96,345 and RP 67580 in Swiss albino mice in the black-

112.

113.

114.

115.

116.

117.

118.

119.

223

and-white box. Neurosci Lett 1993; 151:64–66. Sanchez C, Arnt J, Costall B, Kelly ME, Meier E, Naylor RJ, Perregaard J. The selective sigma 2-ligand Lu 28-179 has potent anxiolytic-like effects in rodents. J Pharmacol Exp Ther 1997; 283:1323–32. Rex A, Barth T, Voigt JP, Domeney AM, Fink H. Effects of cholecystokinin tetrapeptide and sulfated cholecystokinin octapeptide in rat models of anxiety. Neurosci Lett. 1994; 172:139–42 Dauge´ V, Roques BP. Opioid and CCK systems in anxiety and reward. In: Cholecystokinin and Anxiety: from Neuron to Behavior, 1995. pp 152–171, Bradwejn, J. and Vasar, E. (eds.) R. G. Landes Company, Georgetown. Hughes J, Boden P, Costall B, Domeney A, Kelly E, Horwell DC, Hunter JC, Pinnock RD and Woodruff GN. Development of a class of selective cholecystokinin type B receptor antagonists having potent anxiolytic activity. Proc. Natl. Acad. Sci. USA 1990; 87:6728–32. Singh L, Field MJ, Hughes J, Menzies R, Oles RJ, Vass CA and Woodruff GN. The behavioural properties of CI-988, a selective cholecystokininB receptor antagonist. Br. J. Pharmacol. 1991; 104:239–45. Griebel G, Simiand J, Steinberg R, Jung M, Gully D, Roger P, Geslin M, Scatton B, Malfrand JP, Soubrie´ P. 4-(2-Chloro-4-methoxy5-methylphenyl)-N-[(1S)-2cyclopropyl1-(-3-fluoro-4-methylphenyl)ethyl]5-methylN–(2-propynyl)-1,3-thiazol-2-amine Hydrochloride (SSR125543A), a potent and selective corticotrophin-releasing factor1 receptor antagonist. II characterization in rodent models of stress-related disorders. J Pharmacol Exp Ther 2002; 301:333–45. Rutkowska M., Jamontt J., Gliniak H. Effect of cannabinoids on the anxiety-like response in mice. Pharmacological report. 2006; 58:200–6. Griebel G. Variability in the effects of-5HT related compounds in experimental models of anxiety: evidence for multiple mechanisms of 5-HT in anxiety or never ending story? Pol J Pharmacol 1996; 48:129–36.

Chapter 12 Using the Elevated Plus Maze as a Bioassay to Assess the Effects of Naturally Occurring and Exogenously Administered Compounds to Influence Anxiety-Related Behaviors of Mice Alicia A. Walf and Cheryl A. Frye Abstract To assess a construct, such as anxiety, and to determine potential neurobiological underpinnings of this construct, it is often necessary to utilize an animal model. For example, the elevated plus maze is a widely used behavioral assay that has been validated to assess the anxiety-related behavior of rodents. There is great value in using a whole systems approach to assess the many potential targets for a complex disorder, such as anxiety. The elevated plus maze produces reliable results, can be fully automated, is easy and economical to use, and valid results are obtained in a single 5-minute testing session. Briefly, mice are placed at the intersection of the four arms of the maze (two open, two closed), facing an open arm. The number of entries and time spent in each arm is recorded. An increase in the open-arm time is an index of anti-anxiety behavior of mice. Importantly, the patterns of effects that we have observed investigating the role of steroids for effects on anxiety-related behavior has been replicable across cohorts, experiments, species, laboratories, and other anxiety models. Thus, the elevated plus maze is an indispensable tool for investigating the neurobiological substrates of anxiety disorders, using murine models. Key words: Anxiety, affect, mood, murine, knockout, steroid.

1. Background and Historical Overview 1.1. Background

Using a whole animal model can be fruitful for the investigation of the neurobiological mechanisms underlying complex human behaviors, such as those related to disorders in mood and/or affect. To date, basic research and clinical studies have generally supported the notion that affective processes likely involve modulation by several neurotransmitters, neuromodulators, and hormones, which have actions in central nervous system targets, such as the hippocampus, amygdala, and other limbic regions.

T.D. Gould (ed.), Mood and Anxiety Related Phenotypes in Mice, Neuromethods 42, DOI 10.1007/978-1-60761-303-9_12, ª Humana Press, a part of Springer ScienceþBusiness Media, LLC 2009

225

226

Walf and Frye

However, there is great heterogeneity in the behavioral and emotional changes that are manifested in neuropsychiatric disorders, such as anxiety, and these symptoms may be related to specific changes in any one or a number of these substrates, and individual differences among affected individuals. By using a systems approach involving animal models to investigate these fundamental questions, one can elucidate potential targets in the context of other processes, which more similarly mimic the typical human situation. Because of vast individual differences among people with anxiety disorders, it is important to have a valid measure to be able to accurately diagnose, and effectively treat, the anxiety disorder. Indeed, a challenge in the laboratory is to be able to use a straightforward approach to investigate the many potential mechanisms and brain targets of a complex disorder, while retaining a semblance of the typical situation of the patient that is encountered outside the laboratory. A typical approach in the clinical setting is to use an objective behavioral endpoint for psychological evaluations, i.e., use diagnostic evaluations and paperand-pencil self-report inventories to assess specific concerns regarding a neurobiological process that cannot be seen (i.e., anxiety). An example of a psychological measure that is used for anxiety is the Spielberger state-trait anxiety inventory (STAI) test. It is critical that these tests can be used to determine variations from norms, and are reliable and valid. Moreover, the same criteria are important for tasks that are generated to model these behavior using animals. What follows is a discussion of the elevated plus maze as a valid measure of anxiety-like behavior in mice, a background on the use of this task, a description of a typical protocol to use for this task, and a discussion of some of the potential confounds that need to be addressed when using this task. Also included are typical data values that we have obtained using this task in mice in our laboratory to investigate how naturally occurring and exogenously administered steroids may alter anxiety-like behavior of mice in the elevated plus maze. This review is focused on the use of the elevated plus maze in mice, despite its even more extensive use in rats. Although there are clear similarities in the behavior of rats and mice in the elevated plus maze (1), there are also some species-typical responses of rats and mice in this task to consider. As such, for the purpose of this report, to the extent possible, the background literature discussed as follows will be based upon findings in mice, but critical findings in rats will also be included as appropriate and necessary. 1.2. Validity of the Elevated Plus Maze

The elevated plus maze has been used for over two decades as a model of state, unconditioned anxiety, of rodents and is one of the most widely used behavioral assay for anxiety-related behavior in mice. A lengthy discussion of these key findings is beyond the

Using the Elevated Plus Maze as a Bioassay

227

scope of this protocol; readers are referred to several excellent and instructive reviews (2, 3, 4, 5, 6, 7, 8, 9, 10). The elevated plus maze consists of four arms that are elevated from the ground; two opposing arms are enclosed by walls while the other two arms are left open. Responses of mice in the elevated responses are easily assessed and quantified by an experimenter. In brief, mice are placed at the junction of the four arms of the maze, and the time spent in the open arms, or entries made to these arms, is recorded as an objective index of the anti-anxiety-like behavior of the experimental mouse. The reliability and validity of the elevated plus maze, which supports its use as one of the most popular behavioral assays of anxiety-related behavior in animals, are as follows. An important consideration to make regarding psychological measures, including animal models, such as the elevated plus maze, is whether the measures are reliable. Reliable measures are repeatable and consistent. Some of the types of reliability that may be most important to consider in terms of the elevated plus maze are inter-rater reliability and test–retest reliability. Inter-rater reliability for the elevated plus maze is high in our laboratory. Experimenters are trained by others who have expertise in conducting the task until there is little variability in the scores obtained by each experimenter. The elevated plus maze uses well-defined and easily observable and recordable indices (e.g., time spent on the open arms, entries to arms, etc.). The test is recorded and tracked by an automated system as well as an observer, and concordance between these approaches and experimenters is greater than 95%. The case for test–retest reliability in the elevated plus maze can be more complicated because there is some evidence for a test decay effect in this task. Although some early reports did not suggest that prior exposure to the elevated plus maze produced test decay and altered subsequent responding in the task (11, 12, 13), recent reports suggest that there is some evidence of test decay effects, such that there are differences in elevated plus maze behavior when rodents are exposed to the plus maze on more than one occasion. For instance, decreased activity on the open arms of the maze is typical on the second exposure to this task compared to the first exposure (14, 15, 16, 17, 18, 19, 20). Another report has shown no great differences between mice that were tested on three occasions in the elevated plus maze, whether it was once per week or on three consecutive days (21). However, greater reliability has been shown when rodents are tested on the elevated plus maze on two occasions, which are 3 weeks apart, if the maze itself is moved to a novel room (22, 23). One of the first measures of validity that is considered in assessing the utility of a behavioral task is its face validity, which refers to whether the task measures what it is supposed to measure. Mice are prey for many other larger animals, which may underlie their natural tendency to avoid open and, thus, unprotected,

228

Walf and Frye

spaces (and, to a lesser extent, heights). As such, the elevated plus maze appears to measure anxiety in mice because mice typically spend more time on the open arms of the plus maze than the closed arms. Another species-typical anxiety/fear response of mice (and other rodents, such as rats) is immobility/freezing and defecation. These behaviors are increased on the open arms, compared to the closed arms, of the maze (11). A related topic is whether the elevated plus maze has construct validity, which refers to the ability of an objective, and observable measure (e.g., time spent by the experimental mouse on the open arms of the plus maze) reflects a state that cannot be observed (i.e., anxiety). One approach that is typically utilized to assess the construct validity of the elevated plus maze is to compare the effects of anxiogenic (i.e., negative control) and anxiolytic (i.e., positive control) drugs to, respectively, decrease and increase time spent on the open arms of the elevated plus maze (11). In addition to face and construct validity, the elevated plus maze has predictive validity. Predictive validity refers to the extent to which a dependent variable to measure one construct predicts behavior on a related measure. For example, rats or mice that spend more time on the open arms of elevated plus maze also spend more time engaging in anti-anxiety behavior in other tasks that utilize anxiogenic or noxious stimuli that are similar or not (stimuli in parentheses): open field (bright, open spaces), light dark transition (bright, open spaces), social interaction (presence of novel conspecific), Vogel punished drinking (shock stimulus), defensive freezing (shock stimulus), etc. (24, 25, 26, 27, 28). 1.3. Historical Overview

The elevated plus maze is considered a straightforward behavioral assay to assess the anxiety-related responses of rodents, such as rats and mice (and has been used with some success in other species, such as wild mice, voles, guinea pigs, and gerbils) (29, 30, 31, 32, 33). The historical roots of the elevated plus maze are a Y-shaped apparatus that included an elevated open alley, which was first described by Montgomery (34). Rats that were placed in this apparatus demonstrated an approach (enclosed alley) and avoidance (open alleys) responses, and this task was since modified into a maze with two open and closed arms at 90 angles that together form a plus shape (35). Handley and Mithani described how antianxiety responses of rats could be elucidated by measuring the ratio of time spent by rats on the open and closed arms of the maze. As such, the measures obtained in the elevated plus maze are observable and objective. Thus, behavior of mice in this task reflects a conflict between their innate motivations to explore a novel context and stay safe in a protected area. Indeed, unlike other conditioned responses to noxious stimuli (e.g., foot or tail shock, food or water deprivation, loud noises, exposure to predator stimuli, physical restraint, etc.), behavior in the elevated plus

Using the Elevated Plus Maze as a Bioassay

229

maze is considered to reflect an unconditioned response (i.e., proclivity toward dark and enclosed environments and avoidance of open spaces) (36).

2. Equipment, Materials, and Setup 2.1. Elevated Plus Maze – Equipment Specifications

The elevated plus maze consists of four arms that are 40–50 cm above the ground (see Fig. 12.1). There are two arms without walls that are considered ‘‘open’’ and two arms with 15.25-cm high walls. The arms are 30-cm long and 5-cm wide with a 5  5 cm intersection, where all of the arms meet. The elevated plus maze that is currently used in our laboratory was purchased from Columbus Instruments (Columbus, OH) and is made of stainless steel, which is painted matte black, and contains four small holes in each closed arm to allow for setup of an automated system using infrared beams to track movement. We have used mazes with slightly different construction, i.e., one made entirely of wood that was painted matte black, or made from Plexiglas with clear walls, no holes in the arms, and with the same dimensions in the past. We have found similar responses of ovariectomized, placebo-administered mice (a typical negative control

Fig. 12.1. Photograph of elevated plus maze testing of mouse in our laboratory. See text for additional details about the construction of the maze.

230

Walf and Frye

condition in studies in our laboratory) in the elevated plus maze we currently use and those other mazes that we have used (25, 28, 37). Color can be an important consideration in choosing the construction of the maze, if a video-tracking system that relies on contrast of the mouse to the maze will be used. 2.2. Elevated Plus Maze – Materials

In our laboratory, the following items are needed to test mice in the elevated plus maze. Before testing, temporary housing cages are needed. To collect data, an elevated plus maze, a video-tracking system, computer with Microsoft Excel, web-camera, datasheets, data binder, pen, and stopwatch are needed. To analyze data, a statistical program is needed.

2.3. Elevated Plus Maze – Setting Up

The elevated plus maze is setup in a brightly lit room with six, 32-Watt fluorescent overhead lights that produce consistent illumination of 2,800 lumens each. There is consistent illumination of the open and closed arms (ranging from 300 to 750 lux, similar to reports in the literature). The maze is perpendicular to the ground and each arm is level. Although effects have not been found in all studies, lighting can be an important consideration to be made when setting up the maze because it is likely to influence behavior in the maze.

2.4. Elevated Plus Maze – Using a Computerized VideoTracking System

In our laboratory, behavior in the elevated plus is simultaneously collected by a trained experimenter, and with a computerized video-tracking system (e.g., Any-Maze, Stoelting Inc., Wood Dale, IL) installed on a PC using a compact web-camera mounted on the ceiling above the maze. We typically have the system track the majority (i.e., 75þ%) of the body of the mouse to avoid head dips being considered open-arm activity when the head is tracked, or open-arm activity being considered when the tail is being tracked and the mouse rests it on an open arm while sitting in the intersection of the arms. Video-tracking is also enhanced by zooming the camera as close to the maze as possible, and adjusting the brightness and contrast so that the data are collected by the observer and automated tracking system.

2.5. Elevated Plus Maze – Data Collection by a Trained Experimenter

Data of mice in the elevated plus maze are also collected by a trained experimenter, who makes hatchmarks on a datasheet for each arm entry of the mouse and uses a stopwatch to be able to record the amount of time spent on the open arms, or in the intersection of the arms. The trained observer also focuses on the entire body and legs of the mouse, given that the operational definition of an arm entry is when all four legs of the mouse have crossed into the arm. Comparisons in our

Using the Elevated Plus Maze as a Bioassay

231

laboratory demonstrate that the data collected using these two different methods are very similar (i.e., greater than 95% concordance). Data are collected with these two methods for the following reasons. First, by using two methods of data collection, there are multiple ways that the data are stored. The videos are stored on an internal server, which could be re-reviewed as necessary, as are the data output (in a Microsoft Excel file) from the tracking system. The trained experimenter collects data on datasheets that are securely stored in the laboratory. Second, using these two methods of data collection further ensures the accuracy of the results. For example, in rare cases, there are problems with the contrast of the maze, which can disrupt the automated collection of the data from the video-tracking system. This is typically caused by the mouse excreting a large amount of urine, or if the contrast is inadequate due to similar fur color of mouse and maze. As such, the experimenter is present in the room and will notice if there are problems with contrast and obtaining accurate results. A consideration to make is that the presence of the experimenter may alter behavior in this task. This is obviated by making sure that the same protocol, with the experimenter, being in the room is utilized for all experimental subjects. Furthermore, the experimenter is trained to be quiet and make minimal movements when testing the mice. Third, by having the experimenter hand-collect data in the elevated plus maze, other ethological measures, which typically cannot be automatically tracked by the video-tracking system, can be collected simultaneously (see Procedure, Dependent Variables, below). 2.6. Elevated Plus Maze – Calibration/ Validation

Given that many variables regarding the maze or experimental protocol have the potential to confound interpretation of the results in the elevated plus maze, it is necessary for the elevated plus maze protocol be calibrated and validated when it is setup. For example, the experimenter should run mice that have been administered positive and negative control compounds, e.g., anxiolytics or anxiogenics, respectively, in the elevated plus maze. These data can be compared to results reported in the literature. In our laboratory, we have validated the plus maze by comparing the acute effects of the positive control (the anxiolytic, diazepam) and the negative controls (placebo vehicle, and the anti-depressant fluoxetine). The results show that diazepam increases the duration spent by mice on the open arms, compared to vehicle or fluoxetine (Fig. 12.2). Note that in our laboratory, at a single, acute dosing of 20 mg/kg IP fluoxetine does not alter performance in the elevated plus maze or likewise produce anxiolytic-like effects in other anxiety behavior measures. However, the authors acknowledge that in practice, patients with anxiety disorders often respond favorably with chronic SSRI treatment, and this course of treatment if common.

232

Walf and Frye

Fig. 12.2. Validation of elevated plus maze – open-arm duration of mice that were treated with placebo vehicle, a positive control (diazepam, 10 mg/kg IP), or a negative control (fluoxetine, 20 mg/kg IP). Diazepam increased open-arm duration compared to other conditions.

3. Procedure

Step 1: Experimental mice are transported in their home cages from their housing to the hallway outside the behavioral testing room in which they will be assessed in the elevated plus maze. Step 2: In the hallway, mice are singly-caged in temporary housing cages that have their subject number written on it. 3.1. Temporary Housing

To avoid any possibility of different experience or stressor exposure altering behavior of rodents in the elevated plus maze, we ensure that mice have similar experiences before testing. As such, it is critical that all experimental mice consistently spend a brief time in temporary housing before testing or remain in their home cages before testing. Mice are typically housed for a consistent duration, but few differences have been observed if mice have been housed for 0.5–2 h in our laboratory. Step 3: Fill out datasheets (with subject numbers, dates, coded conditions, experimenter’s initials) and make sure that videotracking system is working and ready to use. Step 4: Transport the mouse in its temporary cage into the behavioral testing room.

Using the Elevated Plus Maze as a Bioassay

233

Step 5: Pick up the mouse by the tail and place at the intersection of the closed and open arms, facing the open arm opposite to where the experimenter is standing. It is important that each experimental mouse be placed in the plus maze the same way. There are differences in the time spent in the open and closed arms depending on whether the mice are placed facing toward the open or closed arm (11). Step 6: Start the video-tracking system (with built-in 5-minute timer). The behavior of mice must be consistently recorded for 5 minutes, so it is important that data collection is immediately made from the time the mouse is placed in the maze. Step 7: Data are collected simultaneously by the video-tracking system and experimenter. 3.2. Dependent Variables

The typical measures of anxiety-like behavior of mice in the elevated plus maze are reflected by the mean entries made into, and the time (seconds) spent on, the open arms of the plus maze. An increase in the time spent, or entries made onto, the open arms is considered to reflect anti-anxiety behavior of mice in this task. The time spent in the intersection of the maze is recorded. There are other ethological measures that can be collected coincident with the most commonly used and validated dependent variables described above. Spontaneous motor behavior is determined by simultaneously recording the total arm entries made as well as the closed arm entries made. The ratio of open or closed arm entries/time to the total arm entries/time can be calculated and analyzed, which may be necessary to be done, particularly when there are differences between groups in the closed or total arm entries made. However, in our laboratory, we typically only consider the elevated plus maze behavior of a subject to be valid when there are few differences in spontaneous motor activity in this task found, which would limit interpretation of the specificity of the behavioral effects observed to anxiety. In addition to spontaneous motor activity, the other ethological measures that are typically collected in mice are the rears (standing on two back feet), head dips (orientation of head over the edge of arms of maze toward the ground), stretched-attend posture (stance in which the back feet of the mice remain stationary, while the mouse stretches the front legs and head away and often toward a stimuli during exploration), freezing (cessation of movement), and fecal boli. An increase in rears, head dips, or stretch-attend postures made is thought to reflect anti-anxiety behavior, while an increase in the other two measures reflects the opposite. Note that these other ethological measures must be observed by the experimenter and are not all automatically recorded by the video-tracking system.

234

Walf and Frye

Step 8: The mouse is removed from the maze by its tail and placed inside its temporary housing cage. The experimenter carries the temporary housing cage to the hallway and places the mouse back into its housing cage. Step 9: The elevated plus maze is cleaned with Quatricide and thoroughly dried with paper towels in between the time each mouse is tested. Step 10: Data are analyzed. 3.3. Data Analysis

The typical approach that is taken to determine the effects of independent variables for behavior in the elevated plus maze is to determine differences in mean values of dependent variables using analyses of variance. We use a commercially available statistical analyses program, such as Statview or SPSS, but freeware statistics programs could also be used, to analyze results. An appropriate p-value to use in experiments using the elevated plus maze is one that is less than 0.05 to demonstrate statistically significant effects of independent variables on the dependent variables in the plus maze. In the event of significant main effects, post hoc tests, such as Tukey’s HSD, which corrects for multiple comparisons, or Fisher’s LSD tests, can be utilized to determine differences between groups.

4. Anticipated Results The following results exemplify the data we have obtained using the elevated plus maze protocol in mice described in this report to investigate the anti-anxiety effects of estrogens (e.g., 17b-estradiol; E2), which are hormones secreted by the ovaries. To date, the results that we have obtained using the elevated plus maze to assess the potential of estrogens to determine the effects of E2 have been robust and replicable in the elevated plus maze. These effects are also valid because they are similar to those of the anxiolytic, diazepam (see Fig. 12.2), and in other tasks of anxiety-like behavior (28). For example, we find that ovariectomized mice subcutaneously administered E2 (0.09 mg/kg, which produces proestrus-like E2 levels) spend more time (240% increased versus control) on the open arms of the elevated plus maze than do their placebo vehicle-administered comparison group. These effects are similar to what is observed among proestrus mice, which are naturally receptive and have increased circulating E2 levels and increased time (i.e., typically 300 increased versus control) on the open arms compared to diestrus, low-E2

Using the Elevated Plus Maze as a Bioassay

235

control group). Thus, these data suggest that E2 to ovariectomized mice has anti-anxiety-like effects in the elevated plus maze.

5. Experimental Variables 5.1. Experimental Subjects

Laboratory-bred mice that are typically tested for anxiety-related behaviors in the elevated plus maze are adults (55 days of age or older). Several animal subjects’ variables need to be taken into account when designing experiments with the elevated plus maze, such as mouse strain (38, 39, 40, 41, 42), sex (5, 43), estrus cycle (28), and age (43, 44). It is critical to decide upon the strain, sex, and reproductive status of the mice to be utilized before initiating a study using the elevated plus maze. Historically, many studies have used male rodents to avoid the influence of circulating estrus cycle hormones for elevated plus maze behavior. However, in attempting to model anxiety disorders of human, in which more women than men are diagnosed (45), it is important to also consider the behavior of female mice in the elevated plus maze. Reference data in support of these effects in female mice from our laboratory are as follows. Strain: C57BL/6 mice are the most commonly used inbred strain of mice. Overall, they show normative responses in behavioral measures (46, 47). We have three knockout mouse colonies in our laboratory to assess the effects of estrogens (estrogen receptor bERb, knockout mice), progestins and neurosteroids (5-reductaseknockout mice), and androgens (testicular feminized mutation (tfm) or androgen receptor-insensitive mice). The ERb knockout and tfm mice have been bred on a C57BL/6 J background. On the other hand, 5-reductase-deficient mice and their wildtype littermates were raised on a 129  SVEVBrd background, which has been backcrossed to a C57BL/6 J background, such that they are congenic. We compared the open arm time of ovariectomized mice that were homozygous for the ERb knockout or the 5-reductaseknockout and their wildtype littermates, and mice that were wildtype or heterozygous for tfm (Note that tfm is a dominant spontaneous mutation on the X chromosome, such that there are no homozygous tfm female mice). Figure 12.3 shows differences among those three strains of mice. There were subtle differences between ovariectomized wildtype and homozygous ERb-knockout mice. Ovariectomized homozygous mice had increased open arm time than did their wildtype counterparts. Ovariectomized heterozygous tfm mice spent more time than did their wildtype counterparts on the open arms of the elevated plus maze; however, tfm mice demonstrated great variability in their behavior in this task.

236

Walf and Frye

Fig. 12.3. Effects of strain – knockout (KO) mice and their wildtype (WT) controls, all bred onto a C57BL6/J background, were compared. Few differences were noted among ovariectomized ERb-knockout mice, but larger differences were noted in 5-reductase and tfm mice, and their respective controls, for anti-anxiety behavior in the elevated plus maze.

Sex: We compared the behavior of young adult, approximately 60 days of age, male and female mice for behavior in the elevated plus maze. As Fig. 12.4 shows, intact male mice typically have greater anti-anxiety-like behavior in the elevated plus maze compared to female mice.

Fig. 12.4. Sex differences – adult, gonadally intact male and female mice were compared for duration on the open arms of the elevated plus maze. Male mice had more anti-anxiety behavior than did female mice.

Using the Elevated Plus Maze as a Bioassay

237

Estrus cycle: Although sex differences in the elevated plus maze of mice typically favor males, the role of ovarian steroid cyclicity for those effects is of interest. We compared the behavior of mice that were in the high estradiol and progestin phase of the estrus cycle, proestrus, to those in the low-ovarian steroid phase of the cycle, diestrus. As Fig. 12.5 shows, mice that were in proestrus spent more time in the open arms of the elevated plus maze.

Fig. 12.5. Estrus cycle differences – Adult female mice in proestrus or diestrus were compared for time spent on the open arms of the elevated plus maze. Mice in proestrus spent more time on the open arms of the plus maze compared to diestrus mice.

Given that the potential for differences based upon these variables, it is critical that these subjects’, and the following procedural, variables be seriously considered (3). Note that all experiments using animal subjects must be performed in accordance with relevant guidelines and regulations regarding animal usage and with approval of a governing body. 5.2. Housing of Mice

We typically obtain adult mice from the breeding colonies that we have and maintain in the Life Sciences Research Buildings in the Laboratory Animal Care Facility at The University at Albany – SUNY. The original stock of mice used in the studies described in this report was from Jackson Laboratory (Bar Harbor, ME) or Taconic Farms (Germantown, NY). Mice are always grouphoused (three to five per cage) in our laboratory to obviate potential for social isolation stress to alter anxiety responses (48, 49). They are housed in standard polycarbonate cages with one Nestlet, so that species-typical nest-building activities can

238

Walf and Frye

occur, in a temperature-controlled room (21 – 1C). Mice are maintained on a 12/12 h reversed light cycle (lights off at 8:00 AM) with continuous access to typical rodent chow and tap water.

5.3. Timing of Testing

In our laboratory, mice are housed on a reversed-light cycle. They are always tested in the early phase of their active, dark-cycle. This is important to consider because anxiety, as well as other behaviors that could influence results in the elevated plus maze (alertness, arousal, motor behavior, etc.), are influenced by circadian rhythms/light cycle (3, 50, 51). Furthermore, by testing at this same time of day, there will be consistency in levels of stress as well as gonadal steroids.

5.4. General Health and Normative Response Screening Procedure

To verify that all experimental mice show normative responses before initiation in behavioral studies, mice will be evaluated for their general health and normative responses to stimuli, using modified methods (as per those reported by Crawley) (52). To determine health status of mice, daily observations of their general appearance (cleanliness of fur, typical whisker movement, good posture, normal gait, good muscle tone) and normative behavior (i.e., presence of fur grooming, building nests in home cages with Nestlets, huddling/sleeping with other mice in their home cages, ability to cage climb, paw withdrawal when gently pulled) were done. To determine normative sensory responses, whether there was a presence of a blink reflex to a cotton swab placed close to their eyes and an ear twitch reflex when a swab was lightly placed against ears was determined. To determine sensitivity to heat stimuli, mice were placed on a hot-plate (Fisher Scientific) warmed to 55C, and latency to lick front and back paws was recorded (maximum latency 60 s). A tailflick measure (latency to move tail from heat source; 50C; San Diego Instruments, San Diego, CA) was recorded (maximum latency 10 s for each of three trials). Evaluations were done by observers blind to the conditions of mice so that potential differences in health, reflexes, and/or motor behavior/coordination could be ruled out as contributors to impairments in endpoints in the present study. Only the mice that did not demonstrate deviations from norms for all these measures are included in the study. Current studies in the laboratory assessing functional effects of steroids in ERb-knockout, 5-reductaseknockout, and tfm mice show that the majority of mice (95þ%) have normal responses before behavioral testing and post-surgery, if they have undergone surgical modifications.

5.5. Handling/ Habituation Procedure

Prior experience with handling, stress, injections, or other experimental manipulations can alter responses of rodents in the elevated plus maze (53, 54, 55, 56, 57, 58, 59, 60, 61). Furthermore,

Using the Elevated Plus Maze as a Bioassay

239

behavior of mice, as compared to rats, may be more sensitive to arousal (62, 63, 64). To minimize potential effects of arousal and habituate mice to steroid injections, handling by the investigator, and behavioral testing, mice typically receive a week of handling and habituation prior to behavioral testing (37). Briefly, mice are picked up from their home cage, handled for 15 s, and returned to their home cage (Day 1). Mice are then transferred from their home cage to a novel cage (Day 2) and weighed and then replaced to their home cage (Day 3), transferred to another room via a cart (Day 4), and transferred to another room via a cart, injected with 0.2-cc vegetable oil vehicle subcutaneously, and placed in novel environment for 5 minute (Day 5). 5.6. Effects of Stress

Stressor exposure, whether experimentally induced or from the environment, can alter performance of rodents in the elevated plus maze. We have shown that restraint stress obviates the anti-anxiety-like effects of E2 administration to ovariectomized adult female rats, if they are stressed as adults immediately before testing, or if they are exposed to gestational stress, such that their mothers are restraint-stressed (27, 65). Preliminary ongoing studies in our laboratory have also shown a similar pattern for female or male C57BL/6J mice that are intact or have steroid extirpation and replacement to have attenuated anti-anxiety responses with 20 minutes of pre-testing restraint stress, compared to their nonstressed counterparts. We have typically obtained mice in our experiments from our own colonies because of concerns that shipping stress may obviate some of the normative anti-anxiety effects of steroid treatments in our laboratory, despite the requisite 1-week undisturbed habituation period post-shipment. Recently, mice had to be purchased for an experiment in our laboratory. We were finding different baseline patterns of responding to E2 in these mice on several measures (cognitive behavior, lordosis, open field), including the elevated plus maze. We investigated this effect further to determine if differences may be due to shipping stress or other potential factors. To this end, data from C57BL/6 mice that were purchased from Taconic Farms (Germantown, NY) and those that are bred from our wildtype mice (ERb-knockout mice controls) that have been backcrossed and bred on a C57BL/6J background for over 20 generations in our laboratory, such that at this point they can be considered congenic, were compared. All mice were group-housed and age-matched when tested. When mice were tested in the plus maze task, mice from Taconic had been living in the housing rooms at The University at Albany for 4 weeks, whereas mice from our colony had been living in these housing conditions since conception. Another difference in mice was their exposure to environmental stressors (i.e., a loud fire alarm in

240

Walf and Frye

the animal-care facility). Because of facilities issues, these alarms sounded without warning, often several times in a year. As such, mice that were purchased from Taconic Farms had not been exposed to the alarm before elevated plus maze testing (but were exposed once to the alarm later in the experimental protocol). On the other hand, mice from in-house colonies had the potential to be exposed throughout their lifespan. In this study, mice were administered placebo vehicle or 0.1 mg/kg E2 subcutaneously 44 h before testing in the elevated plus maze, using our standard protocol described above. We found differences in baseline performance of mice that were purchased and those from in-house breeding colonies following treatments (Fig. 12.6). Of note, the facilities deficiencies are being corrected, particularly because of the differences in baseline stress-mediated responses. These data are provided in support of the basic principle of behavioral analyses that have ethological relevance; that it is necessary to ensure that all experimental subjects have similar experiences and consistent treatment behavioral analyses. In humans, it is not possible to systematically parse out the role of their external and internal environment for anxiety or other related behaviors. Indeed, this is one strength of using animal models to probe constructs, such as anxiety, to gain insight about their neurobiological bases, which can be applied to the human condition.

Fig. 12.6. Effects of housing and source – Adult, ovariectomized mice that were obtained from our breeding colonies (wildtype controls of ERb-knockout mice on a C57BL/6J background and exposed to fire alarm disturbances) or purchased from a vendor (C57BL/6Tac and not exposed to fire alarm disturbances) were compared for responses to placebo vehicle (vehicle; SC) or estradiol (E2; 0.1 mg/kg SC). Differences were noted between these groups.

Using the Elevated Plus Maze as a Bioassay

241

5.7. Length of Test

Mice are tested in the elevated plus maze for 5 minutes. This is the typical procedure that is utilized in mice and rats, which is based upon the observations of Montgomery that rats demonstrated the greatest avoidance responses in the first 5 minutes of the Y-maze after they were placed in the elevated open alleys (34). Reports of using a longer than 5-minute test period in this task are uncommon, but often show different baseline behaviors of rodents, and can be difficult to interpret in relationship to the vast literature using the plus maze for anxiety-related behavior of rodents.

5.8. Exposure to Other Novel Environments Before Testing

There is some indication that pre-exposure of rodents to a novel environment, such as an open field or holeboard, immediately before testing in the elevated plus maze may enhance the amount of time spent on the open arms (11, 66, 67). In our laboratory, mice are often run through a battery of tasks, which include testing in the open field immediately followed by the elevated plus maze and then subsequent assessment in other measures of affect (e.g., the forcedswim test) and/or socio-sexual behavior (social interaction and/or paced mating). Another common protocol that is run in our laboratory is to test mice in a single task once per week, for several weeks. Comparing effects on the elevated plus maze of mice that have been tested in a battery of tasks in a single session versus over several testing session has not revealed robust differences. However, some of this may be due to mice in our laboratory being habituated before initiation in either type of behavioral testing protocols.

5.9. Test–Retest Effects

As discussed above in the section on the reliability of the elevated plus maze, the effects of repeated testing in this task must be considered. Although for the purposes of investigating novel pharmacological compounds, mice are typically pre-exposed to the plus maze (in much the same way that the forced-swim test is used to assess the effects of potential anti-depressants). However, we and others typically use the elevated plus maze to assess the anxiety-like responses of mice in a slightly different way. That is, mice are tested in a single session to assess their behavioral responses to the novel environment of the elevated plus maze, which would more likely produce unconditioned avoidance responses toward the open arms. Clearly, careful consideration of this experimental permutation should be made when designing experiments utilizing the elevated plus maze.

6. Troubleshooting 6.1. Mice Fall Off of the Maze

Although it is not a common occurrence and only happens in less than 1% of the mice tested, on occasion mice do run or slip off the open arms. This is more likely to happen if there is a disturbance

242

Walf and Frye

during testing, so care must be taken to avoid this during testing. In our laboratory, we hang signs outside the testing room to inform others in the laboratory that elevated plus maze testing is in progress and loud noises and other disturbances should be minimized. Another possibility for this type of response may be related to the neurological effects of experimental manipulations (i.e., those that may cause sedation, dizziness, hyperactivity, etc.). As such, it is important to screen the mice for normative behavior (see the above procedure). In the event of the mouse slipping from the open arms of the maze, the experimenter must rapidly pick the mouse up by the tail and place it back onto the open arms of the maze. This behavior must be recorded on the datasheet and taken into consideration when the behavioral data are analyzed, but it is most prudent to not include the behavioral data in this type of situation. However, it is a practice in our laboratory that the experimenter continues to test the mouse to ensure that experimental mice have consistent treatment and exposure to tasks, such as the plus maze. This is particularly important when mice are tested in a battery of tasks. 6.2. Mice are Immobile/ Freeze on Open Arms

Another rare occurrence, but one that must be addressed, is when mice become immobile, or freeze, on the open arms. This happens in few mice (less than 1% of those tested), and is defined as spending more than 30% of the total test time on the open arms in an immobile stance. This is most likely to occur in the event of a disturbance (loud noise, etc., which are considered exclusionary criteria) when the mouse is on the open arms. As such, the experimenter must be careful to avoid this potential problem while testing. In the event that the mouse freezes on one of the open arms, the experimenter must record these results and continue testing the animal, so that the experience of each experimental mouse is kept as consistent as possible. However, the data from this mouse are typically not considered in final data analyses.

6.3. Too Low (or High) Baseline Levels of Open-Arm Activity

As with many behavioral tasks, there are many variables that can influence the baseline responding of mice in the elevated plus maze. Especially with consideration of different arousal/reactivity levels among strains of mice, it may be important to alter the maze conditions (lighting of room and/or maze arms, color/brightness of maze, construction) so that differences between experimental conditions can be parsed out. As well, different conditions may be necessary if the hypothesis to be tested is related to whether a condition will produce anxiolytic versus anxiogenic effects. Several modifications can be considered. Illumination could be changed. For example, testing mice in a dim room, or under a red light, which produces 20 lux illumination on each, may increase time

Using the Elevated Plus Maze as a Bioassay

243

on the open arms (68). The color of the maze, and therefore the brightness of the arms, could be changed as to increase or decrease time spent on the open or closed arms (69). Changing the construction of the maze (material, height, arm width, etc.) can be considered in designing experiments using the elevated plus maze. For example, time spent on the open arms is increased when a very low railing (0.25 cm) is added to the edge of the open arms (68, 70).

7. Conclusions The elevated plus maze is a widely used, reliable, and valid measure of anxiety-related behavior of mice that can be easily implemented as an animal model of anxiety. It has great value as a model because it allows for systematic investigations in a whole animal system on the role of genes, neurotransmitters, neuromodulators, and hormones in several potential brain targets important for anxiety.

Acknowledgments The research described in this paper from our laboratory was supported in part by grants from the National Institute of Mental Health (MH06769801), National Science Foundation (IBN0316083), and U.S. Army Department of Defense (BC051001). Technical assistance provided by Fabiola Estrada, Carolyn Koonce, and Danielle Osborne is greatly appreciated. References 1. Walf AA, Frye CA. The use of the elevated plus maze as an assay of anxietyrelated behavior in rodents. Nat Protoc. 2007;2:322–8. 2. Belzung C, Griebel G. Measuring normal and pathological anxiety-like behaviour in mice: a review. Behav Brain Res. 2001;125:141–9. 3. Carobrez AP, Bertoglio LJ. Ethological and temporal analyses of anxiety-like behavior: the elevated plus-maze model 20 years on. Neurosci Biobehav Rev. 2005;29:1193–205. 4. Dawson GR, Tricklebank MD. Use of the elevated plus maze in the search for novel anxiolytic agents. Trends Pharmacol Sci. 1995;16:33–6.

5. File SE. Factors controlling measures of anxiety and responses to novelty in the mouse. Behav Brain Res. 2001;125:151–7. 6. Hogg S. A review of the validity and variability of the elevated plus-maze as an animal model of anxiety. Pharmacol Biochem Behav. 1996;54:21–30. 7. Kulkarni SK, Sharma AC. Elevated plusmaze: a novel psychobehavioral tool to measure anxiety in rodents. Methods Find Exp Clin Pharmacol. 1991;13:573–7. 8. Rodgers RJ, Dalvi A. Anxiety, defence and the elevated plus-maze. Neurosci Biobehav Rev. 1997;21:801–810. 9. Wahlsten D, Metten P, Phillips TJ, Boehm SL 2nd, Burkhart-Kasch S, Dorow J, Doerksen S,

244

10.

11.

12.

13.

14.

15.

16.

17.

18.

19.

20.

Walf and Frye Downing C, Fogarty J, Rodd-Henricks K, Hen R, McKinnon CS, Merrill CM, Nolte C, Schalomon M, Schlumbohm JP, Sibert JR, Wenger CD, Dudek BC, Crabbe JC. Different data from different labs: lessons from studies of gene-environment interaction. J Neurobiol. 2003;54:283–311. Wall PM, Messier C. Methodological and conceptual issues in the use of the elevated plus-maze as a psychological measurement instrument of animal anxiety-like behavior. Neurosci Biobehav Rev. 2001;25:275–86. Pellow S, Chopin P, File SE, Briley M. Validation of open:closed arm entries in an elevated plus-maze as a measure of anxiety in the rat. J Neurosci Methods. 1985;14:149–67. Lister RG. The use of a plus-maze to measure anxiety in the mouse. Psychopharmacology. 1987;92:180–5. File SE, Mabbutt PS, Hitchcott PK. Characterisation of the phenomenon of ‘‘one-trial tolerance’’ to the anxiolytic effect of chlordiazepoxide in the elevated plus-maze. Psychopharmacology. 1990;102:98–101. Almeida SS, Garcia RA, de Oliveira LM. Effects of early protein malnutrition and repeated testing upon locomotor and exploratory behaviors in the elevated plusmaze. Physiol Behav. 1993;54:749–52. Bertoglio LJ, Carobrez AP. Previous maze experience required to increase open arms avoidance in rats submitted to the elevated plus-maze model of anxiety. Behav Brain Res. 2000;108:197–203. Bertoglio LJ, Carobrez AP. Anxiolytic effects of ethanol and phenobarbital are abolished in test-experienced rats submitted to the elevated plus maze. Pharmacol Biochem Behav. 2002;73:963–9. Bertoglio LJ, Carobrez AP. Behavioral profile of rats submitted to session 1-session 2 in the elevated plus-maze during diurnal/nocturnal phases and under different illumination conditions. Behav Brain Res. 2002;132:135–43. Fernandes C, File SE. The influence of open arm ledges and maze experience in the elevated plus-maze. Pharmacol Biochem Behav. 1996; 54:31–40. Lee C, Rodgers RJ. Antinociceptive effects of elevated plus-maze exposure: influence of opiate receptor manipulations. Psychopharmacology. 1990;102:507–13. Treit D, Menard J, Royan C. Anxiogenic stimuli in the elevated plus-maze. Pharmacol Biochem Behav. 1993;44:463–9.

21. Espejo EF. Effects of weekly or daily exposure to the elevated plus-maze in male mice. Behav Brain Res 1991;87:233–238. 22. Adamec R, Strasser K, Blundell J, Burton P, McKay DW. Protein synthesis and the mechanisms of lasting change in anxiety induced by severe stress. Behav Brain Res 2006;167:270–86. 23. Adamec R, Blundell J, Burton P. Anxiolytic effects of kindling role of anatomical location of the kindling electrode in response to kindling of the right basolateral amygdala. Brain Res. 2004;1024:44–5. 24. Frye CA, Petralia SM, Rhodes ME. Estrous cycle and sex differences in performance on anxiety tasks coincide with increases in hippocampal progesterone and 3,5-THP. Pharmacol Biochem Behav. 2000;67:587–596. 25. Frye CA, Walf AA, Rhodes ME, Harney JP. Progesterone enhances motor, anxiolytic, analgesic, and antidepressive behavior of wild-type mice, but not those deficient in type 1 5 alpha-reductase. Brain Res. 2004;1004:116–24. 26. Frye CA, Koonce CJ, Edinger KL, Osborne DM, Walf AA. Androgens with activity at estrogen receptor beta have anxiolytic and cognitive-enhancing effects in male rats and mice. Horm Behav. 2008;54:726–734. 27. Walf AA, Frye CA. Estradiol’s effects to reduce anxiety and depressive behavior may be mediated by estradiol dose and restraint stress. Neuropsychopharmacology. 2005; 30:1288–301. 28. Walf AA, Frye CA. Estradiol or diarylpropionitrile decrease anxiety-like behavior of wildtype, but not estrogen receptor b knockout, mice. Behav Neurosci. 2008;122:974–81. 29. Hendrie CA, Eilam D, Weiss SM. Effects of diazepam and buspirone on the behaviour of wild voles (Microtus socialis) in two models of anxiety. Pharmacol. Biochem Behav. 1997;58:573–6. 30. Holmes A, Parmigiani S, Ferrari PF, Palanza P, Rodgers RJ. Behavioral profile of wild mice in the elevated plus-maze test for anxiety. Physiol Behav. 2000;71:509–16. 31. Rex A, Marsden CA, Fink H. Effect of diazepam on cortical 5-HT release and behaviour in the guinea-pig on exposure to the elevated plus maze. Psychopharmacology. 1993;110:490–6. 32. Stowe JR, Liu Y, Curtis JT, Freeman ME, Wang Z. Species differences in anxietyrelated responses in male prairie and meadow voles: the effects of social isolation. Physiol Behav. 2005;86:369–78.

Using the Elevated Plus Maze as a Bioassay 33. Rodgers RJ, Cole JC. Influence of social isolation, gender, strain, and prior novelty on plus-maze behaviour in mice. Physiol Behav. 1993;54:729–36. 34. Montgomery KC. The relation between fear induced by novel stimulation and exploratory behavior. J Comp Physiol Psychol. 1958;48:254–260. 35. Handley SL, Mithani S. Effects of -adrenoreceptor agonists and antagonists in a maze-exploration model of ‘fear’-motivated behaviour. Naunyn-Schmeideberg’s Arch Pharmacol. 1984;327:1–5. 36. Barnett SA. The Rat- A Study in Behavior. University Chicago Press. (1975). 37. Frye CA, Sumida K, Dudek BC, Harney JP, Lydon JP, O’Malley BW, Pfaff DW, Rhodes ME. Progesterone’s effects to reduce anxiety behavior of aged mice do not require actions via intracellular progestin receptors. Psychopharmacology. 2006;186:312–22. 38. Augustsson H, Dahlborn K, Meyerson BJ. Exploration and risk assessment in female wild house mice (Mus musculus musculus) and two laboratory strains. Physiol Behav. 2005;84:265–77. 39. Carola V, D’Olimpio F, Brunamonti E, Mangia F, Renzi P. Evaluation of the elevated plusmaze and open-field tests for the assessment of anxiety-related behaviour in inbred mice. Behav Brain Res. 2002;134:49–57. 40. Ramos A, Berton O, Mormede P, Chaouloff F. A multiple-test study of anxiety-related behaviours in six inbred rat strains. Behav Brain Res. 1997;85:57–69. 41. Trullas R, Skolnick P. Differences in fear motivated behaviors among inbred mouse strains. Psychopharmacology. 1993;111:323–31. 42. Voikar V, Koks S, Vasar E, Rauvala H. Strain and gender differences in the behavior of mouse lines commonly used in transgenic studies. Physiol Behav. 2001;72:271–81. 43. Frick KM, Burlingame LA, Arters JA, Berger-Sweeney J. Reference memory, anxiety and estrous cyclicity in C57BL/6NIA mice are affected by age and sex. Neuroscience. 2000;95:293–307. 44. Frye CA, Edinger K, Sumida K. Androgen administration to aged male mice increases anti-anxiety behavior and enhances cognitive performance. Neuropsychopharmacology. 2008;33:1049–61 45. Seeman MV. Psychopathology in women and men: focus on female hormones. Am J Psychiatry. 1997;154:1641–7. 46. Bothe GW, Bolivar VJ, Vedder MJ, Geistfeld JG. Behavioral differences among fourteen

47.

48.

49.

50.

51.

52.

53.

54.

55.

56.

57.

58.

245

inbred mouse strains commonly used as disease models. Comp Med. 2005;55:326–34. Bothe GW, Bolivar VJ, Vedder MJ, Geistfeld JG. Genetic and behavioral differences among five inbred mouse strains commonly used in the production of transgenic and knockout mice. Genes Brain Behav. 2004;3:149–57. Hunt C, Hambly C. Faecal corticosterone concentrations indicate that separately housed male mice are not more stressed than group housed males. Physiol Behav. 2006;87:519–26. Zhu SW, Yee BK, Nyffeler M, Winblad B, Feldon J, Mohammed AH. Influence of differential housing on emotional behaviour and neurotrophin levels in mice. Behav Brain Res. 2006;169:10–20. Andrade MM, Tome MF, Santiago ES, Lucia-Santos A, de Andrade, TG. Longitudinal study of daily variation of rats’ behavior in the elevated plus-maze. Physiol Behav. 2003;78:125–33. Jones N, King SM. Influence of circadian phase and test illumination on pre-clinical models of anxiety. Physiol Behav. 2001;72:99–106. Crawley JN, Chen T, Puri A, Washburn R, Sullivan TL, Hill JM, Young NB, Nadler JJ, Moy SS, Young LJ, Caldwell HK, Young WS. Social approach behaviors in oxytocin knockout mice: comparison of two independent lines tested in different laboratory environments. Neuropeptides. 2007;41:145–63. Mechiel Korte S, De Boer SF. A robust animal model of state anxiety: fear-potentiated behaviour in the elevated plus-maze. Eur J Pharmacol. 2003;463:163–75. Adamec RE, Shallow T. Lasting effects on rodent anxiety of a single exposure to a cat. Physiol Behav. 1993;54:101–9. Andrews N, File SE. Handling history of rats modifies behavioural effects of drugs in the elevated plus-maze test of anxiety. Eur J Pharmacol. 1993;235:109–12. Brett RR, Pratt JA. Chronic handling modifies the anxiolytic effect of diazepam in the elevated plus-maze. Eur J Pharmacol. 1990;178:135–8. File SE, Andrews N, Wu PY, Zharkovsky A, Zangrossi, H Jr. Modification of chlordiazepoxide’s behavioural and neurochemical effects by handling and plusmaze experience. Eur J Pharmacol 1992;218:9–14. Lapin IP. Only controls: effect of handling, sham injection, and intraperitoneal injection

246

59.

60.

61.

62.

63.

64.

Walf and Frye of saline on behavior of mice in an elevated plus-maze. J Pharmacol Toxicol Methods. 1995;34:73–7. Padovan CM, Guimaraes FS. Restraintinduced hypoactivity in an elevated plusmaze. Braz J Med Biol Res. 2000;33:79–83. Schmitt U, Hiemke C. Strain differences in open-field and elevated plus-maze behavior of rats without and with pretest handling. Pharmacol Biochem Behav. 1998;59:807–11. Steenbergen HL, Heinsbroek RP, Van Hest A, Van de Poll NE. Sex-dependent effects of inescapable shock administration on shuttleboxescape performance and elevated plus-maze behavior. Physiol Behav. 1990;48:571–6. Mong JA, Pfaff DW. Hormonal and genetic influences underlying arousal as it drives sex and aggression in animal and human brains. Neurobiol Aging. 2003;24:S83–8. Morgan MA, Pfaff DW. Estrogen’s effects on activity, anxiety, and fear in two mouse strains. Behav Brain Res. 2002;132:85–93. Morgan MA, Pfaff DW. Effects of estrogen on activity and fear-related behaviors in mice. Horm Behav. 2001;40:472–82.

65. Walf AA, Frye CA. Estradiol decreases anxiety behavior and enhances inhibitory avoidance and gestational stress produces opposite effects. Stress. 2007;10:251–60. 66. File SE, Wardill AG. Validity of head-dipping as a measure of exploration in a modified hole-board. Psychopharmacologia. 1975;44:53–9. 67. File SE, Wardill AG. The reliability of the hole-board apparatus. Psychopharmacologia. 1975;44:47–51. 68. Cao BJ, Rodgers, RJ. Dopamine D4 receptor and anxiety: behavioural profiles of clozapine, L-745,870 and L-741,742 in the mouse plus-maze. Eur J Pharmacol. 1997;335:117–125. 69. Lamberty Y, Gower AJ. Arm width and brightness modulation of spontaneous behaviour of two strains of mice tested in the elevated plus-maze. Physiol Behav. 1996;59:439–44. 70. Holmes A, Rodgers RJ. Influence of spatial and temporal manipulations on the anxiolytic efficacy of chlordiazepoxide in mice previously exposed to the elevated plus-maze. Neurosci Biobehav Revs. 1999;23:971–980.

Chapter 13 Novelty-Induced Hypophagia Stephanie C. Dulawa Abstract The inhibition of feeding produced by novelty, termed ‘‘hyponeophagia,’’ provides a measure of anxietyrelated behavior in rodents that is sensitive to the effects of multiple classes of anxiolytic and anxiogenic treatments and their time-course of action. This chapter provides detailed protocols for assessing hyponeophagia using the novelty-induced hypophagia (NIH) test. The NIH paradigm was developed to improve the practicality, sensitivity, and reliability of measuring hyponeophagia in laboratory mice. Historical background, as well as recommendations regarding materials, setup, experimental design, procedure, analysis, and interpretation are presented. Key words: Hyponeophagia, novelty-suppressed feeding, neophobia, anxiety, food intake.

1. Background and Historical Significance

The reduction in feeding in response to novelty, termed ‘‘hyponeophagia,’’ was described by Hall (1934) who reported an inverse relationship between feeding and defecation in laboratory animals exposed to a novel environment. Hyponeophagia has been documented in both wild and laboratory rodents (1). Hyponeophagia can be evoked by various novel features of the environment, including novel food (2), a novel testing environment (3), and novel food containers (4). Hyponeophagia-based tests are conflict tests in which animals face a choice between engaging in feeding and neophobic behaviors. Such tests are ethologically relevant, not confounded by painful stimuli, simple to conduct, and cost-effective (5, 6). Experimental paradigms assessing hyponeophagia have been used most frequently to measure anxiety in rats. More recently, the ability to directly manipulate genes in mice using transgenic and

T.D. Gould (ed.), Mood and Anxiety Related Phenotypes in Mice, Neuromethods 42, DOI 10.1007/978-1-60761-303-9_13, ª Humana Press, a part of Springer ScienceþBusiness Media, LLC 2009

247

248

Dulawa

gene targeting strategies has led to a dramatic increase in the use of this rodent species for studying mechanisms underlying behavior and central nervous system function. As a consequence, many behavioral paradigms originally developed for rats have required modification for use with mice. The present chapter provides detailed and up-to-date methods for assessing murine hyponeophagia. Extensive work in rodents indicates that hyponeophagiabased tests exhibit strong predictive validity as anxiety measures (7). Specifically, pharmacological and environmental manipulations that reduce or increase anxiety in humans also reduce or increase hyponeophagia in rodents, respectively. For example, hyponeophagia is reduced in rodents by diverse classes of compounds that are anxiolytic in humans including benzodiazepines (8–12), barbiturates (13–15), azapirones (10, 11), antidepressants (16), and b-adrenergic antagonists (17, 18–22). Anticonvulsants (12, 23) also reduce anxiety in humans and reduce hyponeophagia in rodents, although they are not typically prescribed for the treatment of anxiety. Furthermore, the GABA antagonist picrotoxin, which has anxiogenic effects in other rodent models (24, 25), increases hyponeophagia (23). Exercise has been suggested to reduce anxiety and depression in humans in controlled trials (26), and access to running wheels as a part of an enrichment paradigm has been reported to reduce hyponeophagia in mice (27). Other environmental manipulations that reduce emotionality, such as postnatal handling, also reduce hyponeophagia (28, 29) in rodents. Agents that do not alter anxiety in humans typically do not alter hyponeophagia in animals. Hyponeophagia-based paradigms also predict the timecourse of action of many therapeutic compounds including azapirones and antidepressants, suggesting that these tests can be used to study the mechanisms underlying their therapeutic effects as well as for drug screening. Reductions in hyponeophagia induced by the azapirones gepirone and buspirone are observed after chronic, but not acute, treatment (10, 11) in rodents, as in anxiety-disordered patients. Similarly, chronic treatment with antidepressants is required to reduce anxiety in anxiety-disordered patients and hyponeophagia in rodents. For example, the selective serotonin reuptake inhibitor (SSRI) fluoxetine (10, 30–32), the tricyclics desipramine and amitriptyline (9–11, 31), and the atypical antidepressant tetracyclic mianserin (10) only reduce hyponeophagia when administered to animals chronically. Furthermore, acute treatment with fluoxetine, desipramine, amitriptyline, and nomifensine (10) has been shown to increase hyponeophagia in rodents, consistent with reports which showed that initial treatment with SSRIs (33–36) and tricyclics (37) exacerbate anxiety, agitation, and nervousness in patients.

Novelty-Induced Hypophagia

249

Genetic factors also influence the expression of hyponeophagia. Inbred mouse strains and rat lines selected for emotionality (38) show baseline differences in hyponeophagia (39). Genetically altered mice exhibiting anxious phenotypes including 5-HT1A receptor knockout (KO) mice (40), type-II glucocorticoid receptor antisense-expressing mice (41), substance P receptor KO mice (42), and neuropeptide Y1 receptor KO mice (43) all show increases in hyponeophagia. Many variants of hyponeophagia-based paradigms have been performed in rodents (2, 4, 7, 9, 11). All paradigms measure the inhibition of consumption of a nutritive substance evoked by exposure to some form of novelty. Tests of hyponeophagia for rats have frequently used food deprivations to instigate feeding behavior (9, 31, 40, 44, 45), and large open field arenas as the test environment (9, 46, 47). This approach has frequently, but not always, been referred to as the novelty-suppressed feeding test (9, 10, 31, 40) and has more recently been used with mice by several investigators. We have found that with mice, the use of food deprivations is problematic because of the narrow time-window in which hunger is optimal for evaluating feeding behaviors. We, therefore, use a highly palatable snack, rather than food deprivation, to instigate feeding using the NIH test (7). Furthermore, the NIH test uses an empty cage, rather than an open field, for the novel condition to permit testing of a greater number of mice simultaneously, and to reduce the potential for locomotor activity or exploratory behavior to act as confounding variables. The NIH test also circumvents other common methodological problems, including a lack of control for potential effects of the independent variable on other variables, such as appetite and locomotor activity, besides anxiety. For example, many studies assessing hyponeophagia have not assessed the same dependent measures in the home and the novel environments, providing inadequate controls. The NIH test employs the same dependent measures, latency, and consumption, in an identical fashion in both the home and novel cages. Because the home cage should not elicit anxiety in animals, any changes in latency or consumption elicited by the independent variable should not reflect anxiety.

2. Equipment, Materials, and Setup 2.1. Animals

Hyponeophagia is ubiquitous, and has been reported in many inbred and outbred mouse strains, including CD1 (11), C57BL/ 6ByJ (11), BALB/cByJ (11), BALB/cJ (16, 30), 129/Sv (30, 31), and mixed C57BL/6  129 Sv strains (43, 48). The vast majority of NIH testing has been performed in mice aged 7 weeks to

250

Dulawa

6 months. Whether hyponeophagia can be measured in younger or older animals has not been determined to our knowledge. Hyponeophagia is observed in both sexes. 2.2. Test Environment

Mice are tested in two test environments: the home cage and a novel cage. The home cage is the same cage mice that are singly housed in during training (see procedure). Home-cage bedding should not be changed for approximately 4 days before testing begins; although if this is unavoidable, a half-cage change should be used. The novel cage consists of a standard home cage without bedding. The novel-cage environment consists of the novel cage, a white surface (such as paper) underneath empty cages, and increased overhead illumination, which can be provided by positioning overhead lamps close to cages (Fig. 13.1)

Fig. 13.1. Setup for assessing latency and consumption in the novel cage.

2.3. Pipettes and Milk Solution

Sweetened condensed milk is diluted (1:3; milk:water) and stored at 4C. Fresh batches of milk solution are used for (1) training days and (2) testing days. Milk solution is drawn up into pipettes using a pipettman immediately before use, as milk solution may form a residue. Pipettes are prepared by sawing ends off of 10-ml plastic serological pipettes, and placing sippers (6 cm with ball bearings,

Novelty-Induced Hypophagia

251

Allentown Inc., Allentown, NJ) flush with one end. Sippers are then secured to pipettes by wrapping with parafilm. Once milk solution is drawn into pipettes, the open end is closed using rubber stoppers (red-rubber micro stoppers, 12.7–3 mm diameter; Sigma, St. Louis, MO). 2.4. Setup

For both novel- and home-cage testing days, test cages should be lined up on a countertop in a sound-attenuated room (Fig. 13.1). Lamps should be positioned overhead and white paper placed under cages to increase illumination for novel-cage testing. Cages should be arranged such that mice can be placed into cages quickly, with sippers oriented toward the front so that latency can be assessed easily. Mice can be observed quietly from a chair several feet in front of cages.

2.5. Dependent Measures

To maximize the sensitivity and reliability of any behavioral paradigm, more than one behavioral measure should be assessed. Both latency and consumption are measured in the home and novel cages in the NIH paradigm. Many studies have assessed only latency scores (9, 11, 44), which are very small in the home cage and thus preclude detection of increased appetite induced by independent variables. Home- and novel-cage testings last for 30 min in the NIH test, with consumption values recorded every 5 min from the start of the timer. Some studies assessing hyponeophagia have assessed behavior in the novel cage for as little as 5 min (47). Although the anxiolytic effects of some compounds were identified using this paradigm, the 5-min cut-off prevented the detection of any potential anxiogenic effects. Thus, the NIH test uses a 30-min test period to allow detection of both anxiolytic and anxiogenic effects.

3. Procedure Procedures for using the NIH test to assess the effects of an acute drug treatment using a between-subjects design are presented below. This procedure can be readily adapted to evaluate the effects of chronic drug treatments, genetic manipulations, or other independent variables. 1. Sixty male 8-week-old Balb/cJ mice are weighed and singly housed following 1 week of adaptation to the vivarium. After being housed singly for at least 24 h, mice are then trained to drink sweetened condensed milk for three consecutive days (days 1–3). Mice are presented with milk solution in pipettes in their home cage for 30 min each day, and the total amount

252

Dulawa

consumed each day is recorded. The time of day of training and testing should be held relatively constant. Mice that never drink during the 3 days of training are eliminated from the study. 2. Mice receive acute injection with control or drug before both home- and novel-cage testing. Because of potential carry-over effects of drug, the order of testing (home vs. novel cage) should be counterbalanced within each treatment group. For home-cage testing (day 4 or 5), mice are briefly removed from their cages to position pipettes in wire lids and initial readings are taken. Mice are then quickly returned to their cages and a timer is started. The latency to drink and the volume consumed are recorded every 5 min for 30 min. Latency to drink is defined as the time taken for the mouse to first lick the sipper. To be counted as a lick, the tongue should make contact with the sipper; merely sniffing the sipper is not counted. The sipper should be positioned such that a mouse resting on the floor of the cage can drink comfortably. Homecage testing should be conducted under relatively dim lighting (approx. 50 lux). One experimenter may test approximately four mice at a time in both cage conditions, provided they can place mice into cages rapidly and observe them simultaneously for latency measures. If a mouse drinks before the timer is started, a latency value of 0 is recorded. Mice that never drink during the 30 min of home-cage testing are eliminated from the experiment due to failure to train. 3. For novel-cage testing (day 4 or 5), clean cages identical to the home cage but without bedding are prepared, and pipettes containing the milk solution are positioned. Any drops of milk solution that collect on the floor of the novel cage during pipette positioning should be cleaned thoroughly. Mice are then quickly placed into the novel cage and a timer is started. The latency to drink and the volume consumed are recorded every 5 min for 30 min. Novel-cage testing is performed under bright lighting (approx. 1200 lux), with white paper placed under cages to increase aversiveness of the cage. Lighting conditions can be optimized as needed.

4. Data Analysis Here we describe data analysis for the above example. This approach can be easily adapted for more complex experimental designs, including additional factors such as genotype, sex, or strain.

Novelty-Induced Hypophagia

253

Training data are analyzed using a two-way ANOVA with day as a within-subjects factor and drug as a between-subjects factor. Total consumption over 30 min is the dependent variable. As with analyses for all measures of the NIH test, significant main effects or interactions including within-subjects variables are resolved using post-hoc ANOVAs with alpha levels adjusted according to the Bonferroni procedure. Main effects or interactions for between-subjects variables are resolved using Newman–Keuls post hoc tests or similar procedures. Latency data are analyzed using a two-way ANOVA with cage condition as a within-subjects factor and drug as a betweensubjects factor. Significant main effects or interactions are resolved as described above, although other statistical approaches can also be used as appropriate. For example, latency data often do not conform to a normal distribution. This problem can be circumvented by applying cut-offs for latency, transformations, or non-parametric statistics (49). Outliers may be removed following transformation to a normal distribution. For example, home- vs. novel-cage latency difference scores often conform more closely to a normal distribution and can be used to identify outliers. Consumption data are analyzed using a three-way ANOVA with time (5-min blocks) and cage condition as within-subjects factors, and drug as a between-subjects factor. Significant main effects or interactions are resolved as described above. The power to detect a significant three-way interaction in this design can be suboptimal. Thus, if a strong a priori hypothesis justifies this approach, then planned comparisons can be used to analyze the effects of drug and time using separate ANOVAs for each cage condition.

5. Anticipated Results Many strains of mice learn to drink the sweetened condensed milk solution within 3 days, and can be tested in the home- vs. novelcage conditions on days 4 and 5. If more than approximately 5– 10% of mice never drink the milk solution during any of the three training days, then training can be extended (see Troubleshooting). Many compounds and genetic alterations altering anxiety do not alter training, such that all experimental groups reach similar levels of consumption by day 3 of training. If groups show significantly different learning curves and consumption on day 3, then any differences observed between groups during testing could reflect effects of the independent variable on learning,

Dulawa

rather than anxiety (see Troubleshooting). Thus, home- and novel-cage testing should not begin until different groups have attained the same level of training. In the control group, the latency values should be significantly larger and consumption values should be significantly smaller in the novel cage compared to the home-cage condition (Figs. 13.2 and 13.3). If this relationship is not observed, it is likely that the novel-cage condition is not sufficiently anxiogenic (see Troubleshooting). In such circumstances, conclusions cannot be drawn regarding any effects of the independent variable on anxiety. Another possibility is that some disturbance in the test environment such as rough transport or noise increased anxiety in mice in the home-cage condition. Average home-cage latencies of controls should be approximately < 250 s.

Consumption (mls)

A

Home .6

B

Novel

*

.5 .4

*

*

.3

#

.2 .1 0

300

0

10 18 25

0

10 18 25

#

250

Latency (sec)

254

200 150 100 50 0

#

* * * 0 10 18 25 0 10 18 25 Chronic fluoxetine (mg/kg/day)

Fig. 13.2. Chronic fluoxetine treatment reduces hyponeophagia. The latency to consume, and the amount consumed in the home and novel cage are shown for Balb/cJ mice receiving 0 (n = 13), 10 (n = 13), 18 (n = 12), and 25 (n = 14) mg/kg/day fluoxetine for 4 weeks. (A) The amount consumed in the first 5 min is shown. (B) The latency to consume is shown. Values are mean – SEM. An asterisk (*) indicates a significant difference from controls within the same cage condition. A pound sign (#) indicates a significant difference between cage conditions within the same drug treatment group (p <0.05). A trend for the novel environment to reduce consumption in controls was also observed.

Novelty-Induced Hypophagia

Consumption (mls)

A

Home

Novel

.16 .14 .12 .1 .08 .06 .04 .02 0

* #

* #

0

B

255

1

3

0

1

3

#

700

Latency (sec)

600 500 400

* *

300 200 100 0 0

1

3

0

1

3

Diazepam (mg/kg)

Fig. 13.3. Acute diazepam treatment reduces hyponeophagia. The latency to consume, and the amount consumed in the home and novel cage are shown for Balb/cJ mice in following acute treatment with 0 (n = 20), 1 (n = 19), or 3 (n = 19) mg/kg diazepam. (A) The amount consumed over 30 min is shown. (B) The latency to consume is shown. Values are mean – SEM. An asterisk (*) indicates a significant difference from controls within the same cage condition. A pound sign (#) indicates a significant difference between cage conditions within the same drug treatment group (p <0.05).

The independent variable may alter home-cage measures by affecting appetite, motor activity, or some other confounding variable aside from anxiety. For example, if a drug treatment increases consumption in both the home and novel cages (see Fig. 13.2; 25 mg/kg dose), then drug may be increasing appetite rather than reducing anxiety. However, if other drug dose increases feeding in the novel but not in the home cage, then this dose is likely anxiolytic (see Fig. 13.2; 18 mg/kg dose). If opposite effects of an independent variable are found in the home and novel cages, then novel cage results may still reflect anxiety. For example, if treatment increases latency in the home cage (perhaps reflecting sedation) but still decreases latency in the novel cage, then treatment is likely reducing anxiety in the novel cage. In the home cage, consumption values are typically largest in the first 5 min, and then decrease rapidly over time blocks. In the novel cage, consumption values are often smaller initially and more similar across time blocks. Significant effects of the independent variable in one block or across all blocks in the novel cage indicate

256

Dulawa

an alteration in anxiety levels. Furthermore, effects of manipulations on either latency or consumption are sufficient to indicate changes in anxiety.

6. Experimental Variables and Troubleshooting 6.1. A Large Percentage of Mice Never Drank After 3 Days of Training

If a sufficient number (> 5–10%) of mice have not learned to drink the milk solution after three training days, more days of training may be added. Three days of training is typically sufficient for BALB/cJ (16) and 129Sv (50) mice, but may vary for other strains. If greater than 5–10% of mice do not train after many days (i.e., 1 week), then NIH testing may proceed, but only if equivalent percentages of mice trained per group to avoid introducing bias.

6.2. No Difference Was Found Between Homeand Novel-Cage Latencies

If latency values were low in both cage conditions (< 250 s), then the novel cage condition was not sufficiently anxiogenic. Overhead lamps should be positioned closer to the cages to increase illumination.

6.3. Latency Data Are Highly Variable

High variability in latency data is often the consequence of data distributions that are skewed to the right. Data can be normalized by applying transformations. Cut-offs for latency can also be applied. Before removing outliers, data should be converted to a normal distribution. Besides applying transformations, data can be converted to a more normal distribution by subtracting novel and home latency values to create a latency difference score.

6.4. Drug Treatment Is Increasing Consumption in the Home and Novel Cages

Such a finding may suggest that drug treatment is increasing appetite, or altering some other variable. If chronic drug treatment or genotype were the independent variable, then increases in appetite should also be observed on training days. Ideally, the effects of the drug and doses being assessed in the NIH test should be first assessed for their potential effects on other variables, like locomotor activity.

6.5. Can I Use a WithinSubjects Design in the NIH Test

Although within-subjects designs have not been reported for the NIH test, a similar test of hyponeophagia has evaluated the effects of multiple drug treatments in the same mice (11). In this design, home- and novel-cage testing is alternated each day for a total of 7 days. Thus, the effects of three different treatments (including control) can be evaluated. This strategy is also likely to be successful using the NIH test.

Novelty-Induced Hypophagia

6.6. It Takes a Long Time to Run Experiments Testing Only Four Mice at a Time

257

Two experimenters can easily test groups of mice side by side, doubling productivity. Similarly, once all animals in a group have taken the first lick of milk solution, another group can be started on a separate timer.

7. Concluding Remarks In summary, work to date assessing the effects of experimental manipulations on hyponeophagia indicate that hyponeophagiabased paradigms provide a useful measure of anxiety in rodents. The NIH test provides an improved method for assessing hyponeophagia in mice by incorporating rigorous controls, more behavioral measures, and avoiding food deprivation. In conclusion, the NIH paradigm represents a powerful anxiety model with strong predictive validity for the effects of a wide variety of manipulations and their time course of action. References 1. Mitchell D. Experiments on neophobia in wild and laboratory rats: a reevaluation. J Comp Physiol Psychol 1976;90(2):190–7. 2. Poschel BP. A simple and specific screen for benzodiazepine-like drugs. Psychopharmacologia 1971;19(2):193–8. 3. Hall CS. Emotional behavior in the rat. 1.Defecation and urination as measures of individual differences in emotionality. J Comp Physiol Psychol 1934;18:385–403. 4. Tye NC, Nicholas DJ, Morgan MJ. Chlordiazepoxide and preference for free food in rats. Pharmacol Biochem Behav 1975;3(6):1149–51. 5. Belzung C, El Hage W, Moindrot N, Griebel G. Behavioral and neurochemical changes following predatory stress in mice. Neuropharmacology 2001;41(3):400–8. 6. Blanchard RJ, Hebert MA, Ferrari P, et al. Defensive behaviors in wild and laboratory (Swiss) mice: the mouse defense test battery. Physiol Behav 1998;65(2):201–9. 7. Dulawa SC, Hen R. Recent advances in animal models of chronic antidepressant effects: The novelty-induced hypophagia test. Neurosci Biobehav Rev 2005;29(4–5):771–83. 8. Shephard RA, Broadhurst PL. Effects of diazepam and of serotonin agonists on

9.

10.

11.

12.

13.

14.

hyponeophagia in rats. Neuropharmacology 1982;21(4):337–40. Bodnoff SR, Suranyi-Cadotte B, Aitken DH, Quirion R, Meaney MJ. The effects of chronic antidepressant treatment in an animal model of anxiety. Psychopharmacology (Berl) 1988;95(3):298–302. Bodnoff SR, Suranyi-Cadotte B, Quirion R, Meaney MJ. A comparison of the effects of diazepam versus several typical and atypical anti-depressant drugs in an animal model of anxiety. Psychopharmacology (Berl) 1989;97(2):277–9. Merali Z, Levac C, Anisman H. Validation of a simple, ethologically relevant paradigm for assessing anxiety in mice. Biol Psychiatry 2003;54(5):552–65. Shephard RA, Estall LB. Effects of chlordiazepoxide and of valproate on hyponeophagia in rats. Evidence for a mutual antagonism between their anxiolytic properties. Neuropharmacology 1984;23(6):677–81. Soubrie P, Kulkarni S, Simon P, Boissier JR. [Effects of antianxiety drugs on the food intake in trained and untrained rats and mice (author’s transl)]. Psychopharmacologia 1975;45(2):203–10. Stephens RJ. Proceedings: The influence of mild stress on food consumption in

258

15.

16.

17.

18.

19.

20.

21.

22.

23.

24.

25.

26.

27.

28.

Dulawa untrained mice and the effect of drugs. Br J Pharmacol 1973;49(1):146P. Poschel BP, McCarthy DA, Chen G, Ensor CR. Pyrazapon (CI-683): a new antianxiety agent. Psychopharmacologia 1974;35(3):257–71. Dulawa SC, Holick KA, Gundersen B, Hen R. Effects of chronic fluoxetine in animal models of anxiety and depression. Neuropsychopharmacology 2004;29(7):1321–30. Harris GC, Aston-Jones G. Beta-adrenergic antagonists attenuate withdrawal anxiety in cocaine- and morphine-dependent rats. Psychopharmacology (Berl) 1993;113(1):131–6. Liu HH, Milgrom P, Fiset L. Effect of a beta-adrenergic blocking agent on dental anxiety. J Dent Res 1991;70(9):1306–8. Dyck JB, Chung F. A comparison of propranolol and diazepam for preoperative anxiolysis. Can J Anaesth 1991;38(6):704–9. Laverdure B, Boulenger JP. [Beta-blocking drugs and anxiety. A proven therapeutic value]. Encephale 1991;17(5):481–92. Bailly D, Servant D, Blandin N, Beuscart R, Parquet PJ. Effects of beta-blocking drugs in alcohol withdrawal: a double-blind comparative study with propranolol and diazepam. Biomed Pharmacother 1992;46(9):419–24. Tyrer P. Current status of beta-blocking drugs in the treatment of anxiety disorders. Drugs 1988;36(6):773–83. Shephard RA, Stevenson D, Jenkinson S. Effects of valproate on hyponeophagia in rats: competitive antagonism with picrotoxin and non-competitive antagonism with RO 15-1788. Psychopharmacology (Berl) 1985;86(3):313–7. File SE, Lister RG. Do the reductions in social interaction produced by picrotoxin and pentylenetetrazole indicate anxiogenic actions? Neuropharmacology 1984;23(7A):793–6. Spinosa Hde S, Stilck SR, Bernardi MM. Possible anxiolytic effects of ivermectin in rats. Vet Res Commun 2002;26(4):309–21. Manger TA, Motta RW. The impact of an exercise program on posttraumatic stress disorder, anxiety, and depression. Int J Emerg Ment Health 2005;7(1):49–57. Meshi D, Drew MR, Saxe M, et al. Hippocampal neurogenesis is not required for behavioral effects of environmental enrichment. Nat Neurosci 2006;9(6):729–31. Ferre P, Nunez JF, Garcia E, Tobena A, Escorihuela RM, Fernandez-Teruel A. Postnatal handling reduces anxiety as measured by emotionality rating and hyponeophagia

29.

30.

31.

32.

33.

34.

35.

36.

37.

38.

39.

40.

tests in female rats. Pharmacol Biochem Behav 1995; 51(2–3):199–203. Steimer T, Escorihuela RM, FernandezTeruel A, Driscoll P. Long-term behavioural and neuroendocrine changes in Roman high-(RHA/Verh) and low-(RLA-Verh) avoidance rats following neonatal handling. Int J Dev Neurosci 1998;16(3–4):165–74. Holick KA, Lee DC, Hen R, Dulawa SC. Effects of chronic fluoxetine in BALB/cJ mice do not require adult hippocampal neurogenesis or the serotonin 1A receptor Neuropsychopharmacology 2007;33(2):406–17. Santarelli L, Saxe M, Gross C, et al. Requirement of hippocampal neurogenesis for the behavioral effects of antidepressants. Science 2003;301(5634):805–9. Dulawa SC, Holick KA, Gundersen B, Gingrich JA, Hen R. Effects of chronic fluoxetine in animal models of anxiety, depression, and sensorimotor gating. Neuropsychopharmacology 2004;in press. Salchner P, Singewald N. Neuroanatomical substrates involved in the anxiogenic-like effect of acute fluoxetine treatment. Neuropharmacology 2002;43(8):1238–48. Lipinski JF, Jr., Mallya G, Zimmerman P, Pope HG, Jr. Fluoxetine-induced akathisia: clinical and theoretical implications. J Clin Psychiatry 1989;50(9):339–42. Amsterdam JD, Hornig-Rohan M, Maislin G. Efficacy of alprazolam in reducing fluoxetine-induced jitteriness in patients with major depression. J Clin Psychiatry 1994;55(9):394–400. Gorman JM. The use of newer antidepressants for panic disorder. J Clin Psychiatry 1997;58 Suppl 14:54–8; discussion 9. Bystritsky A, Ackerman DL, Pasnau RO. Low dose desipramine treatment of cocainerelated panic attacks. J Nerv Ment Dis 1991;179(12):755–8. Ferre P, Fernandez-Teruel A, Escorihuela RM, et al. Behavior of the Roman/Verh high- and low-avoidance rat lines in anxiety tests: relationship with defecation and selfgrooming. Physiol Behav 1995;58(6):1209–13. Trullas R, Skolnick P. Differences in fear motivated behaviors among inbred mouse strains. Psychopharmacology (Berl) 1993; 111(3):323–31. Gross C, Zhuang X, Stark K, et al. Serotonin1A receptor acts during development to establish normal anxiety-like behaviour in the adult. Nature 2002;416(6879):396–400.

Novelty-Induced Hypophagia 41. Rochford J, Beaulieu S, Rousse I, Glowa JR, Barden N. Behavioral reactivity to aversive stimuli in a transgenic mouse model of impaired glucocorticoid (type II) receptor function: effects of diazepam and FG- 7142. Psychopharmacology (Berl) 1997;132(2):145–52. 42. Santarelli L, Gobbi G, Blier P, Hen R. Behavioral and physiologic effects of genetic or pharmacologic inactivation of the substance P receptor (NK1). J Clin Psychiatry 2002;63 Suppl 11:11–7. 43. Karlsson RM, Choe JS, Cameron HA, et al. The neuropeptide Y Y1 receptor subtype is necessary for the anxiolytic-like effects of neuropeptide Y, but not the antidepressant-like effects of fluoxetine, in mice. Psychopharmacology (Berl) 2008;195(4):547–57. 44. Bilkei-Gorzo A, Racz I, Michel K, Zimmer A. Diminished anxiety- and depressionrelated behaviors in mice with selective deletion of the Tac1 gene. J Neurosci 2002; 22(22):10046–52. 45. Shephard RA, Broadhurst PL. Hyponeophagia and arousal in rats: effects of diazepam, 5-meth-

46.

47.

48.

49. 50.

259

oxy-N,N-dimethyltryptamine, d-amphetamine and food deprivation. Psychopharmacology (Berl) 1982;78(4):368–72. Britton DR, Koob GF, Rivier J, Vale W. Intraventricular corticotropin-releasing factor enhances behavioral effects of novelty. Life Sci 1982;31(4):363–7. Rex A, Voigt JP, Voits M, Fink H. Pharmacological evaluation of a modified open-field test sensitive to anxiolytic drugs. Pharmacol Biochem Behav 1998;59(3):677–83. Chen ZY, Jing D, Bath KG, et al. Genetic variant BDNF (Val66Met) polymorphism alters anxiety-related behavior. Science 2006;314(5796):140–3. Zar J. Biostatistical Analysis. IV ed. New York: Prentice Hall; 1998. Holick KA, Lee DC, Hen R, Dulawa SC. Behavioral effects of chronic fluoxetine in BALB/cJ mice do not require adult hippocampal neurogenesis or the serotonin 1A receptor. Neuropsychopharmacology 2008; 33(2):406–17.

Chapter 14 Acute and Chronic Social Defeat: Stress Protocols and Behavioral Testing Alessandro Bartolomucci, Eberhard Fuchs, Jaap M. Koolhaas, and Frauke Ohl Abstract This chapter provides an overview of experimental approaches for acute and chronic social defeat in male mice and how to test for the behavioral consequences of social stress using the modified hole board. The first part refers to acute social defeat describing the classical resident/intruder test. It summarizes the experimental protocols including behavioral characterizations of the animals. The second part describes various models of chronic defeat and chronic stress such as the sensory contact model, chronic psychosocial stress, social disruption, and repetitive social defeat. The third part is devoted to behavioral testing during/ after stress using the modified hole board. Besides an up-to-date collection of protocols, troubleshooting approaches are provided. Key words: Social defeat, resident/intruder test, sensory contact model, chronic psychosocial stress, social disruption, repetitive social defeat, modified hole board.

1. Acute Social Defeat in Male Mice: The Classical Resident/Intruder Test 1.1. Background

Adult male mice have a strong motivation to aggressively exclude unfamiliar males from their own territory. Accordingly, the classical resident/intruder test has been scheduled according to basic investigations on mouse aggressive behavior conducted on wild populations and laboratory strains (1–3). In modern terms, the classical resident/ intruder test is aimed at establishing an experimental context in which an intruder mouse is defeated by a resident territorial male (4–6). To ensure this outcome, the experimenter should properly choose the age, strain, and social expertise (among the other characteristics) of both the resident and the intruder as discussed below. Female mice usually do not display territorial aggression and current protocols have been established with males only (7).

T.D. Gould (ed.), Mood and Anxiety Related Phenotypes in Mice, Neuromethods 42, DOI 10.1007/978-1-60761-303-9_14, ª Humana Press, a part of Springer ScienceþBusiness Media, LLC 2009

261

262

Bartolomucci et al.

1.2. The Ideal Resident: The Winner

In general, the ideal resident male having high probability to defeat an intruder should be at least 3-months old, should have been individually housed for at least 5–7 days, and should belong to a highly aggressive strain (such as outbred CD1 or NMRI mice). It has been reported that the latency to attack an intruder male can be strongly reduced if the resident is a ‘‘trained fighter,’’ which means that he was exposed to intruder males (having similar characteristics than the intruder described below) the days preceding the test with the experimental animal. In this case, the experimenter may select as residents only males who reliably attack the intruder with a low latency (conventionally set at <10–15 s). To increase the territorial motivation, the resident can also be pair-housed with a female. This procedure, however, requires extra procedural steps (removal of the female before placement of the intruder; surgical sterilization); it may not be necessary to induce proper aggressive motivation in mice, but single housing for a too long peiod of time will lead to isolation induced aggression. This may be a different form of aggression both in its form and function.

1.3. The Ideal Intruder: The Loser

To increase the probability that an experimental intruder mouse is defeated by the resident, the intruder should have the following characteristics: (a) should be derived from inbred strains which are generally less aggressive than outbred strains, (b) should be younger than the resident but not younger than 2 months to elicit territorial aggression by the resident, (c) in general should weigh less than the resident, (d) should not be housed with a female (breeding pair), and (e) should be housed in a sibling group, although individually housed males can be used as well particularly if they belong to an inbred strain. However, one has to realize that strain and housing conditions are important variables that may strongly determine the outcome of the resident/intruder test.

1.4. Methods and Experimental Protocol

The resident is individually housed for at least 7 days preferably in a large cage such as a commercial standard cage (38  20  18 cm) or larger. The intruder is placed into the resident’s cage and an observation period of 5–10 min starts. After the defeat, the intruder is returned into his home cage. For animal welfare reason, the fight should be carefully followed by the experimenter and interrupted if the intruder is injured by the resident. The interaction can be scored live to determine dominance and subordination and for a rough estimate the level of aggression, but videotaping of the entire confrontation is mandatory for a detailed behavioral analysis (see below). In order to prevent injuries and to reduce the amount of bites received, the intruder can be placed within a wire-mesh box after the first attack by the resident or after the first behavioral sign of subordination (see below) while keeping the animal in sensory contact within the resident’s home cage for the rest of the test.

Acute and Chronic Social Defeat

263

In general, the time of the day for the defeat experiment is not crucial. As mice are nocturnal animals and to ensure maximal aggression level, the defeat can be performed under red light during the first hours after lights off (usually with mice housed under a reverse light:dark cycle). 1.4.1. Behavioral Definition of Winner and Loser

A full description of mice offensive behavior was given by Miczek et al. (1). The attack bite is the most salient and conspicuous element of aggression in mice; it is directed preferentially at the back when the social status is fully defined while bites on the flanks of the opponent often occur in the early phase of the social interaction when the social status is not yet defined. The attack bite is usually preceded by a sideways threat, which is a lateral rotation of the body, accompanied by piloerection and short steps, directed toward the opponent. Although sideways threats and attack bites follow each other often during an aggressive encounter, sideways threats may also be displayed without subsequent bites. Defeated mice are pursued in a rapid chase by the aggressive opponent. When cornered, mice rear up on their hindlegs and assume a defensive upright posture, rotating the upper torso toward the aggressive opponent. Defeated mice also emit squeaking vocalizations. Eventually, frequently attacked mice display a defeat posture with limp forelimbs, upward angled head, and retracted ears. Basic measurement of aggressive behavior allowing to discriminate between a winner (dominant) or a loser (subordinate) can be determined through the following live observations: (a) the latency for the first attack bite, (b) the duration and frequency of attack bouts or flurries, and (c) the frequency and duration of fights without clear winner/loser outcome; in this case usually animals are silent and display flank attacks.

1.4.2. Expected Outcome of the Resident/Intruder Test

The expected outcome of a well-performed resident/intruder test, aimed at having reliable social defeat in the intruder male, is that this animal will show clear behavioral signs of subordination (upright defensive posture, escape behavior, squeaking vocalization, and immobility) upon first attacks while the resident showing repeated attacks during the test phase.

1.4.3. Troubleshooting

1. It is of main importance that the resident’s cage and in particular the sawdust is not changed before the introduction of the intruder. In an optimal way, cage and sawdust are not changed for 7 days. 2. It can be useful to replace the normal wire-mesh cage top with a plastic transparent cover to allow unrestricted view of the interactions in the cage. However, this may disturb the

264

Bartolomucci et al.

resident and consequently increase attack latency. If researchers choose this option, then it is important to familiarize the resident with a new cover, the days preceding the test. 3. Before starting the core experiment, it is always preferable to run a few pilot tests with animals having the same characteristics as the intruders to ensure that the procedure will result in the expected rapid and robust defeat of the intruder.

2. Chronic Defeat and Animal Models of Chronic Stress

2.1. The Sensory Contact Model

The resident/intruder or social defeat test is a very useful tool to investigate the acute stress response or the acute physiological and behavioral consequences of a single social defeat. As repeatedly discussed (1, 6, 8), the resident/intruder test cannot be considered as a model of ‘‘social subordination’’ as this usually concerns a stable hierarchical social relationship between two or more mice in a given context and for a given time. Accordingly, mouse models of social subordination are based on a continuous dyadic interaction or a certain degree of repetition of the resident/intruder test when compared to a single test in the resident/intruder paradigm. The development of these paradigms is also stimulated by the growing appreciation that mouse models of chronic social subordination can be considered as valid models for human psychopathologies and associated co-morbidities and modern gene technology techniques which employ mice as the preferred mammalian experimental tool to knock in/out or overexpress specific genes of interest. Among the currently available protocols, we will discuss below four different paradigms and provide a rationale for their choice. The basic procedure and recommendation follow the same criteria exposed in the previous section. This model has been developed by Kudryavtseva et al. (9, 10) and was later adapted by several other groups with minor procedural changes (e.g., (11–13)). Pairs of weight-matched male mice are placed in cages bisected by perforated transparent partitions allowing the animals to see, hear, and smell each other, but preventing physical contact. The animals are left undisturbed for 2–3 days to adapt to the new housing conditions and sensory contact, before they are exposed to encounters. Subsequently, the lid is replaced by a transparent one, and 5 min later, the partition is removed for 10 min to allow agonistic interactions. The superiority of one of the mice is firmly established within two or three encounters with the same opponent. The superior mouse (winner) attacks, bites, and chases the loser which displays defensive behavior only. The

Acute and Chronic Social Defeat

265

duration of each fight is usually kept to 3 min, at which point the partition is closed. Each loser is exposed to the same winner for 3 days, while afterwards each loser is placed, once a day after the fight, in an unfamiliar cage with an unfamiliar winner behind the partition. Each winner remains in its home cage. This procedure can be performed once a day for 20 days and yields an equal number of winners and losers. In the standard protocol, the controls are animals that have been housed individually for 5 days. In other protocols, groups of siblings were the preferred control group (12). The basic setup allows several variations of the protocol: 1. Duration of the procedure: generally, the procedure is continued for 3–4 weeks, but may be extended for longer period. 2. Frequency of direct social interaction: the separation may be removed once a day or at least twice a week. 3. Duration of the direct social interaction: usually the direct social interaction is terminated when one of the males show clear signs of subordination. In less aggressive strains, the interaction may be standardized for time. 4. Type of the winner: sometimes trained winner males or males from a more aggressive strain are used as partners of the experimental animals. 2.1.1. Significance of the Protocol

This protocol is among the more widely adopted and has been therefore validated by several laboratories; it is procedurally complex but behavioral and physiological effects are stable and reproducible.

2.2. Chronic Psychosocial Stress

The procedure has been developed by Bartolomucci et al. (8, 14). Three-months-old male mice to be used later as residents or intruders are individually housed in Plexiglas cages (38  20  18 cm) under standard lightning conditions (12 h lights on:12 h lights off) for 7 days. During this period, various parameters (body weight, food intake, locomotor activity, etc.) can be measured. This baseline period is followed by a stress period lasting 21–42 days. Each resident receives an unfamiliar weight-matched intruder mouse (same gender) and the two animals are allowed to freely interact for 10 min. After the interaction, the two animals are separated by means of a wire-mesh partition, which allows continuous sensory contact but no physical interaction. Daily (usually 3–4 h after lights on) the partition is removed for a maximum of 10 min to allow social interaction and social status confirmation (see Figs. 14.1 and 14.2). The same dyad is maintained for the entire duration of the stress phase. In order to have a homogenous sample, only dyads are investigated in which the following three criteria are fulfilled: (a) the hierarchy is established within the first 3–4 days, (b) a stable dominant/subordinate relationship is

266

Bartolomucci et al.

15-42 DAYS

Basic protocol: CD1 strain Residents and intruders are 3 months old CD1 males

Control Group of 3 male CD1 siblings of the same age

Modified protocol: Inbred strains Resident: Male of an inbred or transgenic strain Intruder: CD1 male having same age as the resident

Control Group of 3 same age siblings of the inbred or transgenic strain of interest

Example of an experimental protocol Baseline - 7 days

Chronic psychosocial stress – 3 weeks

Post mortem

Body weight, activity

Body weight, activity, aggression, behavioral tests

Hormone analysis, neurobiological parameters, etc.

Fig. 14.1. How to perform the chronic psychosocial stress procedure in mice. Schematic presentation of how to perform psychosocial stress in mice including an example of the basic experimental procedure. The main procedural difference between the basic procedure with the CD1 strain and a modified protocol to be used with inbred or transgenic mice is also presented.

maintained through the entire stress phase, and (c) the subordinate animal never counterattacks after the social status has been established. During the social interaction, offensive behaviors of the animals is manually recorded and the social status of the mice is determined as follows: the chasing and biting animal is defined as ‘‘dominant,’’ while the mouse displaying upright posture flight behavior and squeaking vocalization is the ‘‘subordinate.’’ The numbers of attack bouts performed by each animal can be quantified live or remotely on videotapes for a detailed investigation. The following four behavioral categories can be identified: (a) resident dominant (RD), (b) resident subordinate (RS), (c) intruder dominant (InD), and (d) intruder subordinate (InS). In general, RS and InS can be lumped in the general category ‘‘subordinate’’ and InD and RD in the category ‘‘dominant’’ (see (8, 15) for details). To prevent injuries, the social interaction is interrupted if fighting escalates, for example, if the dominant persistently bites the opponent. Throughout the study, food and water are always available ad libitum to all experimental animals; body weight, food intake etc. are monitored at constant intervals depending on the specific experimental requirements.

Acute and Chronic Social Defeat

267

Fig. 14.2. Chronic psychosocial stress in mice. Top (A) and side (B, C) view of the experimental apparatus consisting of a standard cage divided by a Plexiglas wire-mesh partition bisecting the cage in two symmetrical compartments. (D) A resident bites the subordinate on the back.

Age-matched male mice housed in groups of three siblings serve as controls (non-stress). This choice is based on previous observations showing no metabolic, immune-endocrine, and behavioral evidence of stress activation or anxiety in group-housed siblings (see (8) for discussion). It is recommended that within each control group, the hierarchical status of the animals is determined at least once per week on the occasion of the cage cleaning, a procedure which elicits territorial aggression. The standard procedure has been developed with outbred CD1 males and results in approximately 50% RS and RD (8). The procedure can be easily adapted to investigate the impact of chronic subordination in inbred mice as well (see Fig. 14.1). The procedural change requires housing inbred males as residents and same-aged CD1 males as intruders. Because of a clear competitive disadvantage (16), the expected result of this

268

Bartolomucci et al.

procedure is that most, or all, inbred mice will become subordinates and CD1 dominants (Bartolomucci et al., unpublished observations). 2.2.1. Significance of the Protocol

This protocol – which is procedurally complex but the behavioral and physiological effects are stable and reproducible – may be useful to investigate behavioral and physiological consequences of chronic stress activation. When compared with the sensory contact model, the main difference of the chronic psychosocial stress paradigm is the stability of the dyad which allows a clearer investigation of the stress effects on the social status. The physiological effects of animals exposed to the sensory contact model and to the chronic psychosocial stress model are not fully overlapping (for references see (8, 15, 17)).

2.3. Social Disruption

In this protocol established by Padgett et al. (18) mice are grouped in groups of unfamiliar animals. The stress procedure consists of six sessions of social disruption stress (SDR), administered from 7 days. During each SDR session, cage mates are exposed together to a new isolated aggressive male. SDR sessions last for 2 h and have been scheduled both during the light and the dark phase (18, 19). The intruder is replaced if it does not attack or if it is attacked by any of the residents. Control mice remain undisturbed in groups within their home cage. Each group is observed for 3  15 min at arbitrary intervals during a 3-h period. During observation, the number of social investigatory (sniffing), aggressive (chase, bite, tail rattle, allogrooming, and aggressive upright and aggressive sideways postures) and defensive behaviors (flee and submissive upright or sideways postures) are assessed for each animal. In addition, a fur score is assigned ranging from 1 (no bald, damaged, or disheveled patches, fur well groomed) to 5 (reflecting increasing incidence of damage to, or deterioration in, the apparent condition of the fur). Individuals within groups are ranked according to the ratio of the number of investigatory aggressive interactions initiated and the number of defensive interactions. Top-ranked aggressor males show the highest attack ratio.

2.3.1. Significance of the Protocol

This protocol may be useful when a larger number of defeated mice is required; it is procedurally easier than the two previous protocols, although open to a certain degree of individual variability depending on the amount of aggression displayed by each resident and the amount of intra-cage agonistic interactions elicited in each experimental cage by the different intruders.

2.4. Repetitive Social Defeat

The protocol has been described first for adult male CD1 mice by Maestripieri et al. (20). Individually housed animals are weight-matched in fighting pairs. Established pairs are introduced daily into a clean box of the same type as the home cage for a

Acute and Chronic Social Defeat

269

20-min interaction session. Afterwards, the animals are returned to their home cage. The same individuals are paired for 10 consecutive days always in a clean cage. Dominant and subordinate mice can be reliably identified after two to three session and hierarchy remains stable afterwards. Only dyads, in which the status is reliably defined, should be considered for experimental analyses. Behavioral observations can be conducted by either live observation or detailed behavioral analysis performed after remote scoring of videotapes. 2.4.1. Significance of the Protocol

3. Behavioral Testing During/ After Social Stress

3.1. The Modified Hole Board: Equipment, Materials, and Setup 3.1.1. Modified Hole-Board (mHB) Apparatus

This protocol may be useful to generate a large number of subordinate mice which do not face a continuous sensory contact and, accordingly, a severe chronic stress response; this procedure is easier to perform than the previous ones.

Social stress will induce changes in behavior ranging from food intake to sexual behavior or exploration. Hence, a wide range of tests can be used to assess the behavioral consequences of social stress. It is beyond the scope of this chapter to describe widely used testing procedures such as the open field test, the Porsolt forced swim test, the elevated plus maze, learning tasks, etc. As an example, we here describe a multidimensional approach, the modified hole-board test. When testing for the behavioral effects of stress in animals, it is of importance to exclude possible stressful and confounding factors by the behavioral testing method itself. Further, social stress can result in different behavioral alterations, depending on the duration of the stressful period and on the characteristics of the stress-responder. It is therefore important to evaluate the effects of social stress on different motivational systems and/or cognitive processes in mice. The modified hole-board test allows for evaluating a variety of behavioral parameters in parallel under nonaversive conditions (21, 22). In addition, cognitive performance can be investigated without the need for stressful food or water restriction (23, 24). The mHB test is of opaque grey PVC and consists of a box (100  50  40 cm) with a hole board (80  25  1 cm) with 23 holes positioned in the center of this box (21, 22). The area surrounding the board is divided into 12 squares of equal size by grey lines (see Fig. 14.3). For cognitive testing in mice, the experimental box is reduced in size to 50  50 cm, and a

270

Bartolomucci et al.

smaller board (35  22  1 cm) with 10 cylinders (2.5  3.0 cm, for details see Fig. 14.3) is used (23, 24). Each cylinder contains a piece of almond, fixed beneath a grid, which cannot be removed by the mice. The bottom of each cylinder is scented with flavor (e.g., vanilla-flavor dissolved in water 0.02%; Micro-Plus, Stadtoldendorf, Germany), as pilot mice are attracted by this flavor. For visuo-spatial task, three randomly selected cylinders are marked with a white (i.e., highly contrasting to the grey background) PVC ring and baited with a removable piece of almond as food reward. This design enables the performance of a visuo-spatial task by changing the positions of the baited cylinders and, in doing so, to investigate flexible cognitive processes (23, 24).

novel/unknown food

b)

3 cm

baited cylinder

a)

colored ring

grid

cylinder

Fig. 14.3. Behavioral testing using the modified hole board. (A) Schematic overview over the modified hole board setup. (B) Details of the cognitive version of the modified hole board.

3.1.2. Procedure 3.1.2.1. Animals and Testing Conditions

The mHB has successfully been used in different strains of mice, different ages (10 weeks to 24 month), and both genders (21, 22, 25–27). In general, both single behavioral and repeated cognitive testing in the mHB seem to be the least stressful if testing is performed under red-light conditions (red light: 5 lux, Spot R63, General Electric, 40 W, E27, Ø63 mm) during the activity phase of the animals (24). In case of testing socially housed groups of animals, the groupmates of the respective testing animal can be placed in a group-compartment, which is separated from the adjacent testing compartment by a transparent partition (Fig. 14.3). It should be considered that testing of social groups can result in testing-order effects (27). The group compartment can as well be used to perform testing under social-stress condition by placing a dominant conspecific in the group compartment while testing the socially stressed/defeated individual.

Acute and Chronic Social Defeat

271

3.1.2.2. Modified HoleBoard Testing Procedure

For 2 days before the experiment, each animal receives one piece of almond (about 0.05 g) daily for habituation, as almonds are used as familiar food in the mHB test (and as food reward for cognitive testing). The test is performed in the housing room of the animals after 2 weeks of habituation in the testing/housing room to avoid transport stress. For behavioral testing a piece of almond (familiar food) and, for example, a food pellet (unfamiliar food; e.g., Precision Pellets F0021-J; Bioserv, Holton Industries, Frenchtown, NJ, USA) are placed in 2 cm distance from each other in the corner across the starting point. In case of group testing, all individuals from one group are first placed into the group compartment for about 10 min. Thereafter, all individuals are successively placed into the mHB for 5 min. In case of individual testing, each animal is transferred directly from its home cage to the experimental box and placed at the start position in always the same corner of the compartment. For cognitive testing, each mouse performs four trials daily, with an inter-trial interval of about 30 min. In young untreated adult mice and under red light conditions, baseline (= constant total time trial) usually is reached within 5 days (20 trials) in unstressed mice (23, 24). All tests are videotaped and are directly monitored by a trained observer.

3.1.3. Parameter

The parameters to be measured, assigned to different behavioral categories, according to previous studies (21–24), are listed in Table 14.1.

3.1.4. Anticipated Results

Table 14.1 summarizes the effects of acute and chronic stress, respectively, which can be expected for mHB parameters. Notably, specific effects are depending on the behavioral profile of the mice used (26), e.g., avoidance behavior in mice with a high innate anxiety profile may remain unaffected by stress due to a floor effect.

3.1.5. Experimental Variables

When using the mHB for cognitive testing in mice, the test schedule may vary depending on the specific demands. If one is interested in long-term effects of social stress, then the sequence of baited holes can repeatedly be changed; breaks, spatial trials or reversals can be introduced.

3.1.6. Troubleshooting

1. Highly stressed mice may initially respond to a novel environment with strong behavioral inhibition. In this case it may be useful to prolong the testing time. 2. The time interval between the social defeat and the testing of a (defeated) mouse influences the behavioral response of the latter. When repeatedly testing the animal during a chronic

Behavioral Processes

"

"/$

Chronic # " # " " " " " " # " #

Acute $ " $ $ $ " $ " # $ # $

Description of behavior At last two front paws and head on board Not moving at all but not asleep Sniffing at familiar or unfamiliar food Eating at familiar or unfamiliar food

Dipping inside a cylinder Sniffing at the cylinder without visiting

Parameter Percent time spent on the board Latency board visits Frequency board visits Immobility Latency to first exploration of familiar food Latency to first exploration of unfamiliar food Latency to first intake of familiar food Latency to first intake of unfamiliar food Latency to first cylinder visit Total number of cylinders visited Latency to first cylinder exploration Total number of cylinders explored

Avoidance behavior (anxiety)

Directed exploration

Time until all baited cylinders have been visited

Dipping into non-baited cylinder; nose below the rim

Motivational system

Total time trial

Overall performance

$

"

Repeated choice

Short-term memory

"

$

Omission of a baited cylinder

Omission error

"

Dipping into non-baited cylinder; nose below the rim

$

Wrong choice

Long-term memory

Chronic

Description of behavior

Acute

Parameter

Cognitive Processes

Memory system

Stress effects on parameter

Modified hole board parameters

Table 14.1 Description of modified hole-board parameters and indication of the effects of acute/chronic stress on behavior/cognition. Notably, the specific effects can vary depending on the behavioral phenotype of the mouse tested

272 Bartolomucci et al.

Rearing Stretched attends Defecation Latency grooming Frequency grooming Duration grooming Time spent in group contact (if group mates present in group compartment) Latency line crossings Frequency line crossings

Undirected exploration

Risk assessment

Arousal

Social affinity

Locomotor activity

$ " " " # # " " #

$ " " # " $ " $ "

Body completely stretched forward Cleaning fur

All four paws crossed a line

Animal at the separating wall, attention toward group compartment

Sitting up, both front paws raised, sniffing

Acute and Chronic Social Defeat 273

274

Bartolomucci et al.

stress procedure, the test should not be performed directly before or after social defeat to avoid an association between the two procedures. 3. The presence of a dominant conspecific can result in avoidance of the area next to the partition. In this case the dominant animal either should be removed from the group compartment or a second partition should be introduced to increase the distance to the experimental compartment. 4. In case of active stress responders, parameters indicating avoidance behavior may show contradicting/unexpected effects (e.g., decreased latency board). Usually, such an effect is restricted to latencies and is paralleled by increased locomotor activity. 5. If mice are used which are characterized by a ‘‘home base building’’ strategy instead of thigmotaxis, parameters indicating avoidance are of limited use (22). 6. The high number of parameters measured weakens the statistical power when analyzing the results (28). 7. Careful calculation of the number of animals needed strongly is advised. 8. Live-scoring is necessary to exactly score all parameters. Careful definition of criteria and extensive training of the observer are prerequisites. 9. Cognitive testing can result in ‘‘therapeutic effects’’ in chronically stressed animals, probably due to the positive experience of finding/receiving rewards. References 1. Miczek KA, Maxson SC, Fish EW, et al. Aggressive behavioral phenotypes in mice. Behav Brain Res 2001;125:167–81. 2. Parmigiani S, Ferrari PF, Palanza P. An evolutionary approach to behavioral pharmacology: using drugs to understand proximate and ultimate mechanisms of different forms of aggression in mice. Neurosci Biobehav Rev 1998;23:143–53. 3. Benus RF, Bohus B, Koolhaas JM, et al. Heritable variation for aggression as a reflection of individual coping strategies. Experientia 1991;47:1008–19. 4. Ginsburg B, Allee WC. Some effects of conditioning on social dominance and subordination in inbred strains of mice, Physiol Zool 1942;15:485–506. 5. Roches KE, Leshner AI. ACTH and vospressin treatments immediately after a defeat increase future submissiveness in male mice. Science 1979;204:1343–1344.

6. Miczek KA, Yap JJ, Covington HE 3rd. Social stress, therapeutics and drug abuse: preclinical models of escalated and depressed intake. Pharmacol Ther 2008;120:102–28. 7. Palanza P. Animal models of anxiety and depression: how are females different? Neurosci Biobehav Rev 2001;25:219–33. 8. Bartolomucci A, Palanza P, Sacerdote P, et al. Social factors and individual vulnerability to chronic stress exposure. Neurosci Biobehav Rev 2005; 29:67–81. 9. Kudryavtseva NN. The sensory contact model for the study of aggressive and submissive behaviors in male mice. Aggress Behav 1991a;17:285–291. 10. Kudryavtseva NN, Bakshtanovskaya IV, Koryakina LA. Social model of depression in mice of C57BL/6J strain. Pharmacol Biochem Behav 1991b;38:315–20. 11. Veenema AH, Meijer OC, de Kloet ER, et al. Genetic selection for coping style

Acute and Chronic Social Defeat

12.

13.

14.

15.

16.

17.

18.

19.

20.

predicts stressor susceptibility. J Neuroendocrinol 2003;15:256–67. Moles A, Bartolomucci A, Garbugino L, et al. Psychosocial stress affects energy balance in mice: modulation by social status. Psychoneuroendocrinol 2006; 31:623–33. Berton O, McClung CA, Dileone RJ, et al. Essential role of BDNF in the mesolimbic dopamine pathway in social defeat stress. Science 2006;311:864–8. Bartolomucci A, Pederzani T, Sacerdote P, et al. Behavioral and physiological characterization of male mice under chronic psychosocial stress. Psychoneuroendocrinol 2004;29:899–910. Bartolomucci A, Cabassi A, Govoni P, et al. Metabolic consequences and vulnerability to diet-induced obesity in male mice under chronic social stress. PLoS ONE 2009;4:e4331. Parmigiani S, Palanza P, Rogers J, et al. Selection, evolution of behavior and animal models in behavioral neuroscience. Neurosci Biobehav Rev 1999;23:957–69. Kudryavtseva NN. Use of the ‘‘partition’’ test in behavioral and pharmacological experiments. Neurosci Behav Physiol 2003;33:461–71. Padgett DA, Sheridan JF, Dorne J, et al. Social stress and the reactivation of latent herpes simplex virus type 1. Proc Natl Acad Sci U S A. 1998;95:7231–5. Merlot E, Moze E, Bartolomucci A, et al. The rank assessed in a food competition test influences subsequent reactivity to immune and social challenges in mice. Brain Behav Immun 2004;18:468–75. Maestripieri D, De Simone R, Aloe L, et al. Social status and nerve growth factor serum

21.

22.

23.

24.

25.

26.

27.

28.

275

levels after agonistic encounters in mice. Physiol Behav 1990;47:161–4. Ohl F, Holsboer F, Landgraf R. The modified hole board as differential screen for behavior in rodents. Behav Res Meth Instr Comp 2001;33:392–397. Ohl F, Sillaber I, Binder E, et al. Differential analysis of basal behavior and diazepaminduced alterations in C57BL/6 and BALB/c mice using the modified hole board. J Psychiatr Res 2001;35:147–154. Ohl F, Roedel A, Binder E, et al. Impact of high and low anxiety on cognitive performance in a modified hole board test in inbred mice strains C57BL/6 and DBA/2. Eur J Neurosci 2003;17:128–136. Roedel A, Storch C, Holsboer F, et al. Effects of light- and dark-phase testing on behavioural and cognitive performance in DBA mice. Lab Anim 2006;40:371–381. Binder E, Droste SK, Ohl F, et al. Regular voluntary exercise reduces anxiety-related behaviour and impulsiveness in mice. Behav Brain Res 2004;115:197–206. Erhardt A, Mu¨ller MB, R¨odel A, et al. Consequences of chronic social stress on behaviour and vasopressin gene expression in the PVN of DBA/2OlaHsd mice – influence of treatment with the CRHR1-receptor antagonist R121919/NBI, J Psychopharmacol 2009;23:31–39. Arndt SS, Laarakker MC, van Lith HA, et al. Individual housing of mice – impact on behaviour and stress-responses, Physiol Behav 2009;97:385–393. Laarakker MC, Ohl F, van Lith HA. Reducing the number of animals used in behavioural genetic experiments using chromosome substitution strains. Animal Welfare 2006;15:49–54.

Chapter 15 Reduction of Submissive Behavior Model for Antidepressant Drug Testing in Mice Ewa Malatynska, Albert Pinhasov, and Richard J. Knapp Abstract Dominant submissive relationship (DSR)-based models, one for antidepressant testing, the reduction of submissive behavior model (RSBM), and another for antimanic drug testing, the reduction of dominant behavior model (RDBM), were first established in rats. This chapter discusses development of the RSBM in mice. Mouse strains involved in this task were the inbred strains C57BL/6J and Balb/c and the outbred SABRA mouse. The predictive validity of the RSBM was studied for the C57BL/6J mouse by testing several antidepressants (desipramine, imipramine, amitriptyline, fluoxetine, amoxapine) and non-antidepressants (amphetamine, yohimbine, thiotixen and diazepam) for reduction submissive mouse behavior as determined in a competition for palatable food test. The face validity was studied for the SABRA mouse by determining if there is a genetic component for dominant and submissive traits. Technical details of the RSBM and data analysis that need to be considered for successful application of this methodology in research are also discussed. Key words: C57BL/6 J mouse, SABRA mouse, inbred strain, outbred strain, dominant-submissive relationships, submissive behavior, dominant behavior, subordinate animals, animal model, depression, antidepressants.

1. Background and Historical Overview Social hierarchy is observed throughout the animal kingdom. It was shown that mice of several strains, similar to other mammals, form social hierarchies (1–3). There are many ways of detecting and measuring social hierarchy by observation of animal behavior. For example, the pattern of territory marking and the preference for a sexual mate depend on the social status of an animal as it was shown in house mice (4, 5).

T.D. Gould (ed.), Mood and Anxiety Related Phenotypes in Mice, Neuromethods 42, DOI 10.1007/978-1-60761-303-9_15, ª Humana Press, a part of Springer ScienceþBusiness Media, LLC 2009

277

278

Malatynska, Pinhasov, and Knapp

We have described and systematized methods used to define animal social status in our two recent reviews (6, 7). In summary, there are two main approaches to define animal dominant-submissive interactions (1) by group observation in semi-natural habitats and (2) by facilitation of competition over scarce resources such as food, territory or sexual partners. Using the latter approaches, techniques for modeling mood disorders and testing antidepressant or antimanic drugs were developed. They are based on animal competition for territory, such as resident–intruder (RI) test (8–14), or for food, such as the reduction of submissive behavior model (RSBM) and the reduction of dominant behavior model (RDBM) (6, 7, 15, 16). In many ways, the RI model resembles the RSBM and RDBM. An investigator performing the resident–intruder test often preselects animals for the role of resident or intruder, judging their aggressiveness level by agonistic or defensive postures. In fact it is better when animals are preselected because both types of resident mice, dominant or subordinate, can be identified and the same is true for intruder mice as determined in semi-naturalistic settings (17). Different variants of the behavior and experimental endpoints were used for antidepressant drug testing in the RI model (for example see (8, 18–20)). We have compared distinct versions of the RI test with the RSBM and RDBM in our two recent reviews (6, 7). In this chapter, we will exclusively focus on dominant-submissive relationships (DSR)-based models, the RSBM and the RDBM that use researcher-facilitated competition between paired animals for palatable food (sweetened milk). DSR is established if one animal (dominant) spends significantly more time at the feeder than the paired one (submissive). We have shown that submissive behavior was sensitive to and selectively reduced by antidepressants (RSBM), and dominant behavior was sensitive to a range of drugs used to treat mania in humans (RDBM). The RSBM and RDBM were first established in rats (for review see (6, 15)) and then the RSBM was transferred to mice (16). There are several mouse strains, so we have to decide which strain we would like to choose to start experiments with. We have tested the predictive validity of the mouse RSBM using the C57BL/6J strain for three reasons. First, we compared literature on the performance of several strains in different behavioral models. The C57BL/6J mouse appeared to be most reactive across all behavioral tests studied. For example, C57BL/6J mouse are popular subjects in behavioral models for antidepressant drug testing including tail suspension test (TST), forced swim tests (FST), RI test, unpredictable chronic mild stress (UCMS), learned helplessness (LH), and RSBM. Current representation and performance of C57BL/6J mouse in comparison with a few other mouse strains in depression models are

Reduction of Submissive Behavior Model for Antidepressant Drug Testing in Mice

279

presented in Table 15.1 (see also (21) for a comprehensive review on this subject). It should be stressed that C57BL/6J mice when repeatedly defeated for 3 weeks developed a depressive phenotype (8, 22). Second, the C57BL/6J strain is very often used as the genetic background to produce knockout mice including those for genes related to depression or antidepressant drug activity such as norepinephrine transporter (NET) (23–25), serotonin transporter (SERT) (26), or m or d receptor (27) knockouts. To study their effect, it is good to have the background strain characterized. For full behavioral characteristics of transgenic and knockout mice see Bolivar et al. (28). Third, we have performed preliminary studies with mice in the RSBM using C57BL/6J and BALB/c strains. We found that mice of the C57BL/6J strain formed DSR more easily than the BALB/c strain. In addition, mice of the C57BL/6J strain consistently dominated BALB/c mice when tested in combined pairs (6).

Table 15.1 Behavioral characteristic of different mouse strains in depression models and tests for antidepressant drug activity Strain performance Behavioral test

C57BL/6 J

DBA/2

CBA/J

C3H/HeJ

BALB/c

References

Tail suspension immobility

++++

–

++

+

+++

(61, 62, 63)

Porsolt’s test immobility

++++

+

ns

++

+++

(61, 64, 65, 66)

Resident–Intruder

++++

+++

ns

ns

ns

(8, 67, 68, 69)

UCMS

++++

++

+

++

++++

(70, 71, 72)

Learned helplessness

+/–

–

ns

+++

+++

(73, 74, 75)

RSBM

++++

ns

ns

ns

+

(16)

ns, not studied; +, positive effect; –, no effect; number of (+) or (–), persistence of finding; UCMS, unpredictable chronic mild stress; RSBM, reduction of submissive behavior model. For comprehensive review comparing performance of different mouse strains in commonly used depression models see Jacobson and Cryan (21).

For testing of the presence of the genetic component for dominant-submissive behavior, outbred SABRA mice were used. SABRA mice were originally developed at the Hebrew University and are commonly used in neuroscience research (29–35). We have shown that SABRA mice formed DSR in a competition for palatable food test and their submissiveness was reduced after treatment with imipramine for 3 weeks (36). Using selective breeding technique, we have developed mice with dominant and submissive features (36).

280

Malatynska, Pinhasov, and Knapp

2. Equipment, Materials, and Setup 2.1. Apparatus

Dominant-submissive relationships (DSR) were established and observed in a specially designed apparatus that originally was constructed to study those behavioral traits at the Institute of Psychiatry and Neurology (Warsaw, Poland) as part of Dr. Malatynska’s graduate work (37, 38). Different versions of the rat DSR apparatus were described in our recent publication (15). An apparatus for the study of DSR in mice was first constructed at Indiana University and is a scaled-down version of the apparatus used for rats (16). This apparatus was made of Plexiglas and consisted of two identical chambers (12  8.5  7 cm) joined by a 2.5  2.5  27 cm passage. This passage had narrow slits cut on both sides of the feeder for easy gate insertion and removal. In this way, the paired mice had an equal starting position at the beginning of the session (Fig. 15.1A).

Fig. 15.1. The design of the RSBM apparatus (A) and schematic of a self-refilling feeder (B). The apparatus consisted of two Plexiglas chambers (12  8.5  7 cm), connected by a passage (2.5  2.5  27 cm) having a small feeder dish with milk in the center from which only one mouse can comfortably drink at a time. Note that the gates are inserted on each side of the feeder for equal start of paired animals and small beaker that serves as a feeder. The diameter of both beaker feeder and self-refilling feeder was 0.5 cm.

2.2. Feeder

In the center of the passage, there was a 0.5-cm diameter hole cut in the floor. A beaker filled with palatable food (sweetened milk, 10 g sucrose/250 ml) was placed in the hole before the start of the behavioral testing procedure. In a recent version of the apparatus, we have used a self-refilling feeder that fitted the mouse apparatus

Reduction of Submissive Behavior Model for Antidepressant Drug Testing in Mice

281

(Fig. 15.1B). The feeder was connected to the tunnel before the start of the behavioral test sessions and allowed a constant supply of sweetened milk through the hole. This feeder was a scaleddown version of a feeder that fitted the rat DSR apparatus described in our recent publication (15). Self-refilling feeders were designed for our studies at Johnson & Johnson Pharmaceutical Research and Development, and both the apparatus and feeders for rats and for mice were produced at Arkay Tool and Die (Trevose, PA, USA). It is worth noting that the construction of the apparatus and feeder facilitated competition in paired mice or rats. First, there was only one access point to the feeder for an animal at a time since the width of the tunnel allowed only one animal’s passage and prevented turns. When an animal entered the tunnel, he could only leave by moving forward to the opposite site of the tunnel or backup. The latter happens almost exclusively when the competing partner pushed an animal to the back. Second, the size of a feeder hole allowed only one mouse to drink comfortably at a time. Sharing the feeder was possible but difficult for two animals. 2.3. Scoring System

The endpoint measured to quantitate the DSR is the length of the time spent on the feeder by each member of a pair. Currently for mouse DSR experiments, the feeder time is manually recorded by a human observer, who uses two stopwatches, one for each member of a competing pair of animals. The observers are instructed to measure the actual milk ingestion time. All mouse data discussed in this chapter were recorded using this manual method. With this method during 5-minute experimental session, one observer can record the feeder time for only one pair of animals. To increase experimental throughput for rats, we have automated the scoring of animal feeder time using video-tracking software. We have validated this automatic scoring system for rats and described it in our two recent publications (15, 39). We have only started automation of mouse DSR. However, we think that for such an easily measured endpoint as feeder time, there should be no major obstacles to use the same video-tracking system for mice as we did for rats. In automatic scoring either for rats or for mice, the basic testing apparatus is the same as described on the beginning of this section. For rats, four tunnel units were arranged on each of two tables. This arrangement allows four pairs of rats to be tested simultaneously on a table while the second set is being prepared. The units were separated by opaque plastic partitions to prevent animals from different pairs seeing each other during a test. For more stability, end chambers were sealed to the tables using Velcro tape. The colors of the table and partitions should be adjusted depending upon

282

Malatynska, Pinhasov, and Knapp

what software is going to be used. For mice, because the apparatus is smaller, it is possible to put together 8–10 units on one table. An example of an arrangement for mice that was prepared for use at JNJPRD is shown in Fig. 15.2. However, this arrangement was not tested.

Fig. 15.2. DSR test apparatus for mice, one table experimental set-up prepared for video tracking, and automatic scoring by the computer software. Ten units are placed in close proximity. They are separated by thin, non-transparent plastic dividers to decrease animals’ distraction. Self-refilling feeders are attached to the tunnels.

2.4. Video-Tracking System

There are several video-tracking systems available on the market that can be used to score DSR. We have used two systems. At first, we used PanLab software (PanLab, San Diego Instruments, CA, USA) (15, 39) and later the Cleversys Inc. customized Topscan software (unpublished observations). Both systems did basic scoring of paired animals’ feeder time well. However, the PanLab system required the coloring of both animals’ heads to be able to distinguish them and independently score their presence in the feeder zone. This coloring required a considerable amount of time. Second, the last version of the PanLab software we were using had very rigid data input and output requirements and did not allow for video review of the experiment. Thus, we switched to Cleversys Inc. software. This software required coloring of only one animal and permitted post-session video analysis of the experiment already analyzed in the real time. This feature allowed review and additional analysis of the data for validity. In addition, the Cleversys software has several options giving greater flexibility in

Reduction of Submissive Behavior Model for Antidepressant Drug Testing in Mice

283

regard to data collection and processing that allow more precise definitions of animal behavior. It would be the system of choice to automate mouse DSR scoring.

3. Procedure 3.1. Animals

All mouse strains tested thus far have formed dominant-submissive relationships. This includes C57BL/6J, BALB/c and SABRA mice. All mouse strains used in our experiments required a reversed 12 h day/12 h night cycle and they need to be adapted to such cycle for at least 1 week prior to testing. If animals were not maintained on reverse light/dark cycle, then their feeder time scores were much lower and they did not feed properly on the regular food given after the experimental session, which usually resulted with rapid weight loss. Before testing, mice need to be food-restricted to enhance their motivation to compete. However, in order to maintain weight that is not significantly different from that indicated by developmental charts for a given mouse strain, the food restriction time for mice should be shorter than that for rats (23 h). We have shown that the number of mouse pairs forming a DSR depends on the food restriction time. C57BL/6J mice formed DSR even without food restriction, but the number of pairs that formed significant DSR was lower. Increased food restriction time from 0 to 8 h, initially, resulted in larger number of mouse pairs forming DSRs. The number of DSR pairs formed after 12 and 16 h of food restriction time decreased (6). This nonlinear function of the number of DSR pairs formed with food restriction time may indicate the involvement of a motivational component in the formation of DSR relationships and led us to speculate that to model depression, we need to choose animals that are still submissive under a strong stimulus (long food-restriction time). Following these premises for validation of the RSBM, we have fed C57BL/6J mouse just after experimental session for 5 h so that the food restriction time was about 18–19 h on the weekdays. On weekends (from Friday to Sunday afternoon) mice were fed ad libitum. The opposite could be true for selecting dominant animals for the RDBM. It seems that for validity in the RDBM, a shorter food restriction time should be chosen since animals still dominant under the condition of a weak stimulus should model mania better.

3.2. Experimental Flow and Data Collection

Shipments of mice were received at 2-week intervals for several months. Each shipment was assigned a number as a separate DSR entity (DSR1, DSR2, and so on). Groups of eight mice were

284

Malatynska, Pinhasov, and Knapp

housed four per cage. Each group was assigned a letter, so that a shipment contains several groups (e.g., A–D for 32 mice or 16 pairs). The mice in each cage were tail-marked one through four. On the beginning of the experiments, mice from each group (two cages) were randomly assigned to fixed pairs. These pairs were brought together only once a day during a testing period. Otherwise the mice from pairs were separated to different home cages. For daily 5-min experimental sessions, each member of a pair was placed in an opposite chamber of the testing apparatus. The feeder time for each animal was recorded during the 5-min testing period. Mice from each shipment went through a 1-week habituation to the apparatus and a partner and a 1-week selection process, followed by 3–5 weeks of drug (or vehicle) treatment to those pairs selected (Fig. 15.3). Initially, we tested drug-treated rats and mice for 3 weeks. However, some antidepressant drugs produce effects that can be delayed for more than 3 weeks, so in later experiments, we extended the treatment period to 5 weeks. Behavioral testing was conducted each weekday and suspended on weekends, but drug treatments were given every day including weekends.

Fig. 15.3. Timetable for the execution of DSR (RSBM/RDBM) experiments. Two first weeks of experiment, habituation and selection, are training weeks. Initially, we used to select animals only on the statistical difference in feeder time (FT) of paired animals during second week (selection week). Currently the results from the first week are also taken into account (see Section 3.3). Training weeks are followed by five treatment weeks, where in the RSBM submissive mice are treated with a drug and dominant partner with vehicle. Data are collected daily. They can be analyzed as daily input or as weekly averages (see Sections 3.5 and 3.6).

3.3. Selection Process and Criteria

The selection process is the foundation for all DSR models and it is based on the observation that animals in the population dominate or submit to others in a varying degree and that only fraction of a population form strong DSRs. The selection goal is to choose animals that form well-defined DSRs. The reasons for selection criteria and details of their evolution are described elsewhere (6, 15). The following summary of current selection criteria was applied across our experiments. However, we think that these selection criteria are still not ideal and will undergo further definition changes with increasing knowledge of underlying bases of these two behavioral traits.

3.4. Selection Criteria for Mice

Mice were tested for 10 days during 2 weeks before pairs that formed DSR were selected. During the first week (5 days) of testing, the animals habituate to the new environment. Dominance was assigned to the animal with the higher score during

Reduction of Submissive Behavior Model for Antidepressant Drug Testing in Mice

285

the second week of testing, if there was a significant difference (two-tailed t-test, P<0.05) between the average daily feeder-time (FT) of both animals. Initially, selection was based only on this one criterion and the data for the first week were not taken into account. These selection criteria were applied in all experiments with C57BL/6J and BALB/c mice. In later experiments with SABRA mice additional selection criteria deduced from advanced experiments with rats were applied. This includes two other criteria, the dominant animal FT must be at least 25% greater than the submissive animal’s FT, and there must not be any ‘‘reversals’’ during the pair selection week, where the putative submissive animal outscores his/her dominant partner on isolated occasions. Ideally there are minimal reversals during the acclimation week as well. Addition of these criteria increased the stability of DSR in control animals through the duration of the experiment. Under such condition, about 25– 33% of the initial animal pairs achieved DSR status (15). 3.5. Data Analysis – Selection Confirmation

After selection, the weekly feeding time (FT) data were plotted for a group of dominant and submissive animals, and the statistical difference between these two groups of animals during selection week was assessed using ANOVA with a post hoc t-test. If this difference is not statistically significant for each day, then it could mean that there is some flow in selection, e.g., one of the criterion was overlooked or missed or that the statistical power of the experiment is too low. It is always a good idea to test the variation obtained in a pilot experiment against required sample size by power analysis to achieve adequate statistical power of the data, which usually is at least 80%. Currently we have also realized that it is advantageous to take data from the first week (habituation week) into account and test statistical significance of difference between selected dominant and submissive animal groups during 2 weeks of testing.

3.6. Data Analysis – Treatment Effect Verification

There are several comparisons that can be made to look for treatment effect. This includes intra-pair control, internal control within a behavioral group, and external control between treatment groups. Intra-pair control indicates a change in relation to the partner: comparisons were made between dominant and submissive animals during each week of the study. Any loss of significance in FT between paired mice indicated a change in their relationship. Internal control: comparison within a group of submissive mice of the difference between initial (before treatment) week and subsequent weeks after treatment. A significant difference indicated a change in the submissive behavior of animals. These two analyses do not allow for comparisons of two independent groups of animals, where in one group the submissive partner of a pair was treated with a drug, while dominant one with vehicle, and in

286

Malatynska, Pinhasov, and Knapp

another group where both partners were treated with vehicle. To compare these groups efficiently, we have summarized the relation of paired animals by calculating dominance level (DL) value for the pair, a difference between feeder time of dominant mouse (FTD) and submissive mouse (FTS). This value is sensitive to the behavior of both animals from a pair that may change under the influence of a drug or other conditions. In order to eliminate the influence of other conditions, the difference in DL between the control group (pairs of mice were both dominant and submissive mice were treated with vehicle) and the treatment group (submissive mice were treated with drug and dominant mice with vehicle) was compared using ANOVA, followed by a post hoc t-test on the level of each week. The normalization to the selection week DL value was necessary to account for individual differences in the initial score level in pairs under different treatment and was calculated according the following formula. DLweek n ð%Þ ¼

DLweek n DLweek 2

This calculation is an external control between treatment groups, and if significant differences are found, they provide strongest evidence for drug effect.

4. Anticipated Results From the experimental layout described above, we would first expect that it will be possible to select animals with firmly formed DSR and differentiate them from other competing pairs that do not develop a DSR. On average, 20–30% of population formed DSR. Thus, starting an experiment with 32 animals (16 pairs), one should be able to select 4 pairs having a strong DSR. It should be noted that at any given experiment, the number of DSR pairs could be lower or higher than 4 for the total population of 16 pairs. An additional expectation would be that selected pairs would hold their relationships through the duration of the experiment. This is not always true and the quality of the selection is important to keep the DL value of control group close to 100% of the DL value of selection week (15). When a submissive animal is dosed with a drug that reduces submissiveness, his assertiveness should increase gradually which should be reflected by his increased feeder time. After few days to few weeks of dosing, the significant differences between dominant and treated submissive animals feeder time should disappear and a significant difference in DL between control group of pairs and group of mice pairs with treated submissive animals should become evident.

Reduction of Submissive Behavior Model for Antidepressant Drug Testing in Mice

287

We have examined the predictive validity of submissive mouse behavior as a model of depression (16). We expected that antidepressant drugs would be differentiated from other psychotropic drugs that are not used in the treatment of depression if the mouse DSR is a valid model. We have dosed animals for 3 weeks with antidepressant drugs from different classes such as the serotonin reuptake inhibitor, fluoxetine (5 and 10 mg/kg), tricyclic antidepressants like imipramine (IMI), desipramine (DMI), and amitriptyline (AMI), all at 10 mg/kg, and the atypical tricyclic antidepressants from the dibenzoxazepine class, amoxapine (AMX, 1 and 10 mg/kg). We have also dosed animals with nonantidepressant drugs such as psychostimulants including amphetamine (AMPH) and yohimbine (YOH), the neuroleptic such as thiotixen (THTX), and the anxiolytic such as diazepam (DIAZ), all at 1 mg/kg. In addition, diazepam was also dosed at 2 mg/kg.

Fig. 15.4. Effect of antidepressants (A & C) or non-antidepressants (B & D) on dominance level (DL) in pair of mice, where submissive animals were treated with the drug, while dominant animals with vehicle. On panels A and B, weekly average of dominance level is plotted in seconds for the experiment duration including habituation, selection, and treatment weeks. DL normalized to the selection week is shown on panel C and D. (see Sections 3.5 and 3.6 for details). Abreviations: FLX, – fluoxetine, IMI – imipramine, DMI – desipramine, AMI – amitriptyline, AMX – amoxapine, AMPH – amphetamine, YOH – yohimbine, THTX – thiotixen, DIAZ – diazepam. Number next to the abbreviation indicates dose in mg/kg. The statistical significance between control and treatment groups of animals was assessed using two-way ANOVA with post hoc t-test. Data points marked as (*) are significant at p <0.05, (**) at p<0.01.

288

Malatynska, Pinhasov, and Knapp

All antidepressants reduced submissiveness of C57BL/6J mice during 3 weeks of administration depending on dose, except amitriptyline (Table 15.2 and Fig. 15.4A,C). Amitriptyline was one false negative under our experimental conditions. A similar lack of effect for AMI (10 mg/kg ip for 21 days) was obtained in SpragueDawley rats (40). It is possible, that it would have an effect after longer administration or at a different dose. However, we did not test these possibilities so far. We do recommend, though to extend the period of treatment in this model to 5 weeks at least, since the clinic onset of antidepressant activity is known to be delayed 2–6 weeks (for review see (7)). In light of this review, it is difficult to say whether the reported effect of amitriptyline in the clinic is slower than that of other antidepressants or not (41, 42). Amitriptyline is known to have anxiolytic and sedative effects in the clinic and was shown to be more effective in anxious depression than other antidepressants (43). Perhaps the dose used in our studies had too large of a sedative component to reduce submissive behavior. We have shown that the anxiolytic drug diazepam did not reduce submissive mouse behavior (16). We have previously shown in a different model that AMI at 2 mg/kg reversed clonidine-induced submissive behavior of Wistar rats (38). The activity of AMI in the RSBM should be studied further.

Table 15.2 Effect of different antidepressant drugs in mice and rats RSBM Effect in the RSBM Rats

Mice

Drug class

Drug name

Dose (mg/kg)

Sprague-Dawley

C57BL/6J

SABRA

Tricyclics

Desipramine

10.0

+

+

ns

Imipramine

10.0

+/–

+

+

+

ns

ns

10.0

–

–

ns

1.0

ns

+

ns

ns

–

ns

+

ns

ns

+

ns

ns

+/–

–

ns

+

+

ns

20.0 Amitriptyline Tetracyclics

Amoxapine

10.0 Maprotyline

10.0 20.0

SRIs

Fluoxetine

5.0 10.0

ns, not studied; +, positive effect; –, no effect.

Reduction of Submissive Behavior Model for Antidepressant Drug Testing in Mice

289

Of all non-antidepressant drugs tested, only amphetamine had an effect similar to that of fluoxetine, imipramine, and amoxepine in the mouse (Table 15.3 and Fig. 15.4BD,). This effect contrasts to the lack of a reduction of submissive behavior previously observed in Sprague-Dawley rats (6). It needs to be mentioned that AMPH can have mood elevating effects in the clinic (44). Overall, the predictive validity of the RSBM looks reasonably promising for studies done so far. Other laboratories should repeat these studies to demonstrate inter-laboratory reproducibility, and also to study the drugs effects under an improved definition of DSR.

Table 15.3 Effect of different non-antidepressant drugs in mice and rats RSBM Effect in the RSBM Rats

Mice

Drug class

Drug name

Dose (mg/kg)

Sprague-Dawley

C57BL/6J

SABRA

Psychostimulants

Amphetamine

1.0

–

+

ns

Yohimbine

1.0

ns

–

ns

Neuroleptics

Thiotixen

1.0

ns

–

ns

Anxiolytics

Diazepam

1.0

–

–

ns

ns

–

ns

2.0 ns, not studied; +, positive effect; –, no effect.

There are two other validities of animal models of depression that should be established before the model can be considered fully validated: construct and face validity. First, construct validity is defined as the presence of critical symptoms of human disease, e.g., helplessness or anhedonia in depression. These forms of validity are not established for the RSBM using either rats or mice. The performance of submissive and dominant mice in other depression models should also be assessed. These could include acute models such as forced-swim test (FST) (45, 46), tail-suspension test (TST) (47, 48), or chronic models such as learned helplessness (LH), (49–52) or chronic mild stress (CMS) (53–56) models. We would anticipate that submissive animals, in contrast to dominant ones, should show features of depression-like symptoms as assessed by these models. On the other hand, animals that show anhedonia in CMS or animals from other models like the genetically created Flinders Sensitive Line (FSL) of rats (57, 58) should show submissiveness in the RSBM. It was demonstrated by Pucilowski et al. (59) in a water competition test that FSL rats were

290

Malatynska, Pinhasov, and Knapp

submissive in contrast to Flinders Resistant Line (FRL) of rats, which adds to the validity of models based on social competition paradigm. Second, face validity is based on the presence of neurochemical changes that can be observed in the model and also during the course of human illness, e.g., increased adrenal hormone levels in depression. The link to several loci on different chromosomes is well documented for bipolar disorders (for review, see (60)). Thus, genetic control of a trait would fall in the face validity category since this is a neurochemical feature observed for depression and mania in humans. Recently, we have shown that SABRA mice form DSR and that imipramine reduces submissive behavior of SABRA mice (Table 15.2). When we bred dominant animals (males and females) and submissive animals (males and females) and tested them for DSR formation, we obtained increasing number of pairs forming such relationship in each consecutive generation of males and females (Fig. 15.5A). We have also shown that time necessary to form significant DSR shortened (Fig. 15.5B). These data support face validity of the mouse RSBM since they are consistent with a genetic component. Other biochemical measures such as corticosterone levels in plasma still need to be investigated.

Fig. 15.5. Evidence for genetic control of dominant and submissive behavioral traits as expressed in pairs competition test in SABRA mice. (A) Percent of SABRA males and SABRA females pairs forming dominant-submissive relationships in parental generation (P1) and offspring generations (F1–F4). Note that this value increases with consecutive generations. (B) Time when dominance level in DSR SABRA pairs reached 50% of its maximal value across generations. It should be noted that this value can be considered as a measure of DSR onset time and it is significantly shorter in offspring generations than in parental generation. The statistical significance between groups of dominant and submissive animals was assessed using two-way ANOVA with post hoc t-test. Data points marked as (*) are significant at p <0.05, (**) at p<0.01, and (***) at p<0.001.

Reduction of Submissive Behavior Model for Antidepressant Drug Testing in Mice

5. Experimental Variables and Troubleshooting

291

Identification, definition, and quantification of many behaviors are always on the list of variables in behavioral tests. Many behaviors do not lent themselves for easy quantification, and behavioral tests modeling disease symptoms, or searching for underlying biochemical entities controlling a behavior, always face the necessity of strict definition and limit to be able to distinguish between the presence and absence of the behavior. For example, in behavioral testing for antidepressant-like effects in FST and TST, the immobility of animal is contrasted with animal activity to escape an unpleasant or inconvenient situation. The question always arises in such research, does the animal have to be completely immobile to call it immobile or can he move a little bit? If we agree that not all animal movements mean escape and that an animal can move slightly and still be counted as immobile, then the question arises as to how slightly? So it is obvious that the threshold of what we call immobility and activity needs to be established either for human observers or for computer software if the observation is to be automated. This can be one source of variability in the results obtained by different laboratories. The same is true about DSR-based tests including definition of agonistic and defensive postures in the resident–intruder tests used to detect antidepressant-like activity. The endpoint measured in the RSBM and RDBM is feeding time of paired food-restricted animals at a feeder to which access is limited in a specially designed apparatus. It is a good idea to establish the definition of feeding activity because it can also be a source of variation. When a human observer scores animal performance in the apparatus, usually the real drinking time with the animal actively ingesting sweetened milk is measured. However, there are moments when animals struggle for access to the feeder, where they rapidly exchange drinking status. Should they be timed at the feeder together in such situation? Usually observers are instructed to score with as much resolution between the animals as possible. When actual milk ingestion time is recorded, the reversals of social status within a pair are more easily achieved since the dominant animal’s score goes down in direct relationship to the increase in the submissive animal’s score. When feeder time for paired animals was automated, the video-tracking software recorded the animal’s presence in the feeder zone and not an actual milk ingestion time. In the automated experiment setting, the presence of dominant animals in the feeder zone did not have that much tendency to decrease, while the presence of submissive animals increased.

292

Malatynska, Pinhasov, and Knapp

The details of results from both experimental settings were discussed in our previous publications (15, 16). The question that the researcher should ask is which condition describes the submissive animal’s status better: the inability to gain access to the feeder or the effort to gain access even if unsuccessful? This question still awaits answer. However, we would rather give a preference to the interpretation that struggle time over an access to the feeder is not evidence for submission, so the counting of a presence in the feeder as equal for both animals would better reflect their social status. The selection process, used to distinguish pairs with a dominant-submissive relationship from pairs that do not, is another source of variation. What criteria are applied and how closely they are followed decides which pairs will be selected and which rejected. To see this process in action, it is advisable to conduct first experiments without a treatment and follow groups of selected and rejected animal pairs for planned duration of an experiment, e.g., 7 weeks. Such an experiment is a good control for experimental conditions and the criteria applied to the selection. We are aware that this is a labor and time-consuming experiment, so it is more likely conducted using automated animal scoring which has higher throughput. It is expected that animal FT in selected pairs will stay significantly different throughout the duration of experiment. It does not mean that FT for all animal pairs will be well separated all the time. It is enough that a statistically significant difference should be maintained between groups. FT for animals in the rejected group even if somewhat different during first weeks has usually tendency to get less different. If one would calculate DL values for these two groups, then the rejected group of animals may look like groups where submissive mice were treated with antidepressant drugs. Thus, it is important to have control groups for each series of experiments. Figure 15.4 illustrates the importance of having an external control group and also DL calculations and normalization. A necessity for DL normalization is especially evident when the variation of DL values during the selection week is large (Fig. 15.4A). An example of control group importance is given in Fig. 15.4C,D. For these particular groups of experiments, a significant difference in FT for the paired animals was the only selection criterion. This is reflected in the control group where the DL values went down about 40% from the initial DL value. It definitely affected the interpretation of the results for the drug-treated groups. Even if in drug-treated groups, the DL values had a tendency to drop more than control almost immediately, they are only becoming significant after these values are close to zero or below (Fig. 15.4C,D).

Reduction of Submissive Behavior Model for Antidepressant Drug Testing in Mice

293

6. Concluding Remarks The mouse RSBM has potential utility for modeling depression. It is a mirroring paradigm to the RDBM with the potential utility to model mania. The RSBM is only partially characterized for mice at this time. More work on the characterization of the RSBM was done on rats and limited studies with the RDBM were exclusively done on rats. There is a lot of advantage to fully establish RSBM and RDBM for mice. This includes mouse body size and its consequences; mice are smaller than rats, so larger groups of animals take less space, which facilitates manipulation and decreases cost. In addition, when used for drug testing, they require less compound. There are many genetic strains of mice including transgenic animals that facilitate the search for drug targets and testing hypotheses for underlining mechanism of mood disorders. However, even if they did not have the potential to serve as a model of mania and depression, the rigorous experimental settings and behavioral definitions for inclusions and exclusions of animals as dominant and submissive facilitate studies of biochemical and genetic underlying of these two behavioral traits. We hope that more work will be done in the future to fully characterize these two promising models.

References 1. Bakshtanovskaia IV, Kudriavtseva NN. [The strategy of submissive behavior in male mice: the effect of the genotype and the experience of preceding agonistic encounters]. Nauchnye Doki Vyss Shkoly Biol Nauki 1991:73–9 (Russian). 2. Bysygina TV, Osadchuk AV. [Effect of genotype and social stress on cAMP- and substrate-dependent mechanisms of regulating hormonal function of testis in mice]. Genetika 2001;37:649–56 (Russian). 3. Turney TH, Hunt EF, Money VM. Systolic blood pressure during the formation of a social dominance hierarchy in C57BL/6j mice. Physiol Behav 1983;31:299–301. 4. Desjardins C, Maruniak JA, Bronson FH. Social rank in house mice: differentiation revealed by ultraviolet visualization of urinary marking patterns. Science 1973;182:939–41. 5. Mossman CA, Drickamer LC. Odor preferences of female house mice (Mus domesticus) in seminatural enclosures. J Comp Psychol 1996;110:131–8. 6. Malatynska E, Knapp RJ. Dominant-submissive behavior as models of mania and

7.

8.

9.

10.

11.

depression. Neurosci Biobehav Rev 2005;29:715–37. Malatynska E, Pinhasov A, Creighton CJ, et al. Assessing activity onset time and efficacy for clinically effective antidepressant and antimanic drugs in animal models based on dominant-submissive relationships. Neurosci Biobehav Rev 2007;31:904–19. Kudryavtseva NN, Bakshtanovskaya IV, Koryakina LA. Social model of depression in mice of C57BL/6 J strain. Pharmacol Biochem Behav 1991;38:315–20. Miczek KA, O’Donnell JM. Intruderevoked aggression in isolated and nonisolated mice: effects of psychomotor stimulants and L-dopa. Psychopharmacology (Berl) 1978;57:47–55. Miczek KA, Thompson ML, Shuster L. Opioid-like analgesia in defeated mice. Science 1982;215:1520–2. Mitchell PJ, Fletcher A. Venlafaxine exhibits pre-clinical antidepressant activity in the resident-intruder social interaction paradigm. Neuropharmacology 1993;32:1001–9.

294

Malatynska, Pinhasov, and Knapp

12. Mitchell PJ, Fletcher A, Redfern PH. Is antidepressant efficacy revealed by druginduced changes in rat behaviour exhibited during social interaction? Neurosci Biobehav Rev 1991;15:539–44. 13. Olivier B, van Aken H, Jaarsma I, van Oorschot R, Zethof T, Bradford D. Behavioural effects of psychoactive drugs on agonistic behaviour of male territorial rats (resident-intruder model). Prog Clin Biol Res 1984;167:137–56. 14. Thurmond JB. Technique for producing and measuring territorial aggression using laboratory mice. Physiol Behav 1975;14:879–81. 15. Malatynska E, Pinhasov A, Crooke JJ, Smith-Swintosky VL, Brenneman DE. Reduction of dominant or submissive behaviors as models for antimanic or antidepressant drug testing: technical considerations. J Neurosci Methods 2007;165:175–82. 16. Malatynska E, Rapp R, Harrawood D, Tunnicliff G. Submissive behavior in mice as a test for antidepressant drug activity. Pharmacol Biochem Behav 2005;82:306–13. 17. Bartolomucci A. Resource loss and stressrelated disease: is there a link? Med Sci Monit 2005;11:RA147–54. 18. Mitchell PJ. Antidepressant treatment and rodent aggressive behaviour. Eur J Pharmacol 2005;526:147–62. 19. Rygula R, Abumaria N, Flugge G, Fuchs E, Ruther E, Havemann-Reinecke U. Anhedonia and motivational deficits in rats: impact of chronic social stress. Behav Brain Res 2005;162:127–34. 20. Rygula R, Abumaria N, Flugge G, et al. Citalopram counteracts depressive-like symptoms evoked by chronic social stress in rats. Behav Pharmacol 2006;17:19–29. 21. Jacobson LH, Cryan JF. Feeling strained? Influence of genetic background on depression-related behavior in mice: a review. Behav Genet 2007;37:171–213. 22. Kudryavtseva NN, Bakshtanovskaia IV, Popova NK. [The development of pathological forms of behavior in submissive male C57BL/6 J mice during agonistic zoosocial interactions. A possible model of depression?]. Zh Vyssh Nerv Deiat Im I P Pavlova 1989;39:1134–41. 23. Amara SG, Sonders MS. Neurotransmitter transporters as molecular targets for addictive drugs. Drug Alcohol Depend 1998;51:87–96. 24. Dziedzicka-Wasylewska M, Faron-Gorecka A, Kusmider M, et al. Effect of antidepressant drugs in mice lacking the norepinephrine

25.

26.

27.

28.

29.

30.

31.

32.

33.

34.

35.

36.

37.

transporter. Neuropsychopharmacology 2006;31:2424–32. Xu F, Gainetdinov RR, Wetsel WC, et al. Mice lacking the norepinephrine transporter are supersensitive to psychostimulants. Nat Neurosci 2000;3:465–71. Rioux A, Fabre V, Lesch KP, et al. Adaptive changes of serotonin 5-HT2A receptors in mice lacking the serotonin transporter. Neurosci Lett 1999;262:113–6. Filliol D, Ghozland S, Chluba J, et al. Mice deficient for delta- and mu-opioid receptors exhibit opposing alterations of emotional responses. Nat Genet 2000;25:195–200. Bolivar V, Cook M, Flaherty L. List of transgenic and knockout mice: behavioral profiles. Mamm Genome 2000;11:260–74. Avraham Y, Menachem AB, Okun A, et al. Effects of the endocannabinoid noladin ether on body weight, food consumption, locomotor activity, and cognitive index in mice. Brain Res Bull 2005;65:117–23. Fride E, Foox A, Rosenberg E, et al. Milk intake and survival in newborn cannabinoid CB1 receptor knockout mice: evidence for a ‘‘CB3’’ receptor. Eur J Pharmacol 2003;461:27–34. Gobshtis N, Ben-Shabat S, Fride E. Antidepressant-induced undesirable weight gain: prevention with rimonabant without interference with behavioral effectiveness. Eur J Pharmacol 2007;554:155–63. Pick CG, Yanai J. Studies into the mechanisms of strain differences in hippocampus-related behaviors. Behav Genet 1989;19:315–25. Sumariwalla PF, Gallily R, Tchilibon S, Fride E, Mechoulam R, Feldmann M. A novel synthetic, nonpsychoactive cannabinoid acid (HU-320) with antiinflammatory properties in murine collagen-induced arthritis. Arthritis Rheum 2004;50:985–98. Yanai J, Bergman A, Shafer R, Yedwab G, Tabakoff B. Audiogenic seizures and neuronal deficits following early exposure to barbiturate. Dev Neurosci 1981;4:345–50. Yanai J, Woolf M, Feigenbaum JJ. Morphological alterations in the medial preoptic area after prenatal administration of phenobarbital. Acta Anat (Basel) 1982;114:347–54. Feder Y, Ogran A, Yadid G, Malatynska E, Pinhasov A. Evidence that Dominant-Submissive Relationships Formation in Mice is Increased by Selective Breeding. Behav Brain Res submitted. Malatynska E. Clonidine model of depression in laboratory animals [Ph.D.]. Warsaw:

Reduction of Submissive Behavior Model for Antidepressant Drug Testing in Mice

38.

39.

40.

41.

42.

43.

44.

45.

46.

47.

48.

49.

Institut of Psychiatry and Neurology, College of Medicine; 1985. Malatynska E, Kostowski W. The effect of antidepressant drugs on dominance behavior in rats competing for food. Pol J Pharmacol Pharm 1984;36:531–40. Pinhasov A, Crooke J, Rosenthal D, Brenneman D, Malatynska E. Reduction of Submissive Behavior Model for antidepressant drug activity testing: study using a videotracking system. Behav Pharmacol 2005;16:657–64. Malatynska E, De Leon I, Allen D, Yamamura HI. Effects of amitriptyline on GABAstimulated 36CI– uptake in relation to a behavioral model of depression. Brain Res Bull 1995;37:53–9. Lopez-Ibor Alino JJ, Ayuso Gutierez JL, Montejo Iglesias ML, Ramons JL. A double-blind clinical comparison between nomifensine and amitriptyline in the treatment of endogenous depressions. Int Pharmacopsychiatry 1982;17(Suppl 1):97–105. Weissman MM, Lieb J, Prusoff B, Bothwell S. A double-blind trial of maprotiline (Ludiomil) and amitriptyline in depressed outpatients. Acta Psychiatr Scand 1975;52:225–36. Rampello L, Nicoletti G, Raffaele R, Drago F. Comparative effects of amitriptyline and amineptine in patients affected by anxious depression. Neuropsychobiology 1995;31:130–4. Sabelli H, Fink P, Fawcett J, Tom C. Sustained antidepressant effect of PEA replacement. J Neuropsychiatry Clin Neurosci 1996;8:168–71. Porsolt RD, Anton G, Blavet N, Jalfre M. Behavioural despair in rats: a new model sensitive to antidepressant treatments. Eur J Pharmacol 1978;47:379–91. Porsolt RD, Le Pichon M, Jalfre M. Depression, a new animal model sensitive to antidepressant treatment. Nature 1977;266:730. Steru L, Chermat R, Thierry B, et al. The automated Tail Suspension Test: a computerized device which differentiates psychotropic drugs. Prog Neuropsychopharmacol Biol Psychiatry 1987;11:659–71. Steru L, Chermat R, Thierry B, Simon P. The tail suspension test: a new method for screening antidepressants in mice. Psychopharmacology (Berl) 1985;85:367–70. Seligman ME, Maier SF. Failure to escape traumatic shock. J Exp Psychol 1967;74:1–9.

295

50. Seligman ME, Maier SF, Geer JH. Alleviation of learned helplessness in the dog. J Abnorm Psychol 1968;73:256–62. 51. Seligman ME, Rosellini RA, Kozak MJ. Learned helplessness in the rat: time course, immunization, and reversibility. J Comp Physiol Psychol 1975;88:542–7. 52. Seligman ME, Weiss J, Weinraub M, Schulman A. Coping behavior: learned helplessness, physiological change and learned inactivity. Behav Res Ther 1980;18:459–512. 53. Katz RJ. Animal models and human depressive disorders. Neurosci Biobehav Rev 1981;5:231–46. 54. Katz RJ. Animal model of depression: pharmacological sensitivity of a hedonic deficit. Pharmacol Biochem Behav 1982;16:965–8. 55. Willner P, Muscat R, Papp M. An animal model of anhedonia. Clin Neuropharmacol 1992;15(Suppl 1 Pt A):550A–1A. 56. Willner P, Towell A, Sampson D, Sophocleous S, Muscat R. Reduction of sucrose preference by chronic mild stress and its restoration by a tricyclic antidepressant. Psychopharmacology 1987;93:358–64. 57. Overstreet DH, Janowski DS. A cholinergic supersensitivity model of depression. In: Boulton AA, Baker GB, Martin-Iverson MT, eds. Animal Models in Psychiatry II. Clifton, NJ: Humana Press; 1991:81–114. 58. Overstreet DH, Pucilowski O, Rezvani AH, Janowsky DS. Administration of antidepressants, diazepam and psychomotor stimulants further confirms the utility of Flinders Sensitive Line rats as an animal model of depression. Psychopharmacology (Berl) 1995;121:27–37. 59. Pucilowski O, Overstreet DH, Rezvani AH, Janowsky DS. Effect of verapamil on submissive behavior in genetically bred hypercholinergic rats in a water competition test. Eur J Pharmacol 1990;187:507–11. 60. Mathews CA, Reus VI. Genetic linkage in bipolar disorder. CNS Spectr 2003;8:891–904. 61. Porsolt RD, Brossard G, Hautbois C, Roux S. Rodent models of depression: forced swimming and tail suspension behavioral despair tests in rats and mice. Curr Protoc Neurosci 2001;Chapter 8:Unit 8 10A. 62. Bourin M, Chenu F, Ripoll N, David DJ. A proposal of decision tree to screen putative antidepressants using forced swim and tail suspension tests. Behav Brain Res 2005;164:266–9. 63. Trullas R, Jackson B, Skolnick P. Genetic differences in a tail suspension test for

296

64.

65.

66.

67.

68.

Malatynska, Pinhasov, and Knapp evaluating antidepressant activity. Psychopharmacology 1989;99:287–8. Sugimoto Y, Kajiwara Y, Hirano K, et al. Mouse strain differences in immobility and sensitivity to fluvoxamine and desipramine in the forced swimming test: analysis of serotonin and noradrenaline transporter binding. Eur J Pharmacol 2008;592:116–22. Bachli H, Steiner MA, Habersetzer U, Wotjak CT. Increased water temperature renders single-housed C57BL/6 J mice susceptible to antidepressant treatment in the forced swim test. Behav Brain Res 2008;187:67–71. Kurtuncu M, Luka LJ, Dimitrijevic N, Uz T, Manev H. Reliability assessment of an automated forced swim test device using two mouse strains. J Neurosci Methods 2005;149:26–30. Lumley LA, Charles RF, Charles RC, Hebert MA, Morton DM, Meyerhoff JL. Effects of social defeat and of diazepam on behavior in a resident-intruder test in male DBA/2 mice. Pharmacol Biochem Behav 2000;67:433–47. Avgustinovich DF, Lipina TV, Kudriavtseva NN. [Response of the serotoninergic brain system to social stress of various duration in male mice C57BL/6 J and CBA/Lac]. Ross Fiziol Zh Im I M Sechenova 2001;87:532–42.

69. Siegfried B, Frischknecht HR, Waser PG. Defeat, learned submissiveness, and analgesia in mice: effect of genotype. Behav Neural Biol 1984;42:91–7. 70. Ibarguen-Vargas Y, Surget A, Touma C, Palme R, Belzung C. Multifaceted strainspecific effects in a mouse model of depression and of antidepressant reversal. Psychoneuroendocrinology 2008. 71. Yalcin I, Aksu F, Belzung C. Effects of desipramine and tramadol in a chronic mild stress model in mice are altered by yohimbine but not by pindolol. Eur J Pharmacol 2005;514:165–74. 72. Mineur YS, Belzung C, Crusio WE. Effects of unpredictable chronic mild stress on anxiety and depression-like behavior in mice. Behav Brain Res 2006;175:43–50. 73. Anisman H, Merali Z. Rodent models of depression: learned helplessness induced in mice. Curr Protoc Neurosci 2001;Chapter 8:Unit 8 10C. 74. Shanks N, Anisman H. Stressor-provoked behavioral changes in six strains of mice. Behav Neurosci 1988;102:894–905. 75. Shanks N, Anisman H. Escape deficits induced by uncontrollable foot-shock in recombinant inbred strains of mice. Pharmacol Biochem Behav 1993;46:511–7.

Chapter 16 Mice Models for the Manic Pole of Bipolar Disorder Shlomit Flaisher-Grinberg and Haim Einat Abstract The lack of appropriate animal models for bipolar disorder (BPD) is a major factor hindering the research of its pathophysiology and the development of new drug treatments. One approach for the development of better models for the disorder is to separately model a number of its critical behavioral domains. In line with this approach, the current chapter describes a number of tests, which have been validated in the context of mania in Black Swiss mice. These tests include sweet-solution preference test, representing reward seeking (see Protocol 2), resident–intruder test, representing aggression/intrusion (see Protocol 3), and a variation of the forced-swim test, representing increased vigor and resistance to despair (see Protocol 4). The chapter also include a protocol for assessing spontaneous activity (see Protocol 1), since this test is critical for the interpretation of results from the other tests. It is suggested that these tests can be used independently for the study of different domains of the manic pole of BPD, and that pending further validation, they could be integrated into a coherent and continuous test battery that may also include tests for additional domains of BPD. The use of tests for distinct domains of BPD, either separately or as a continuous battery, can potentially be utilized to screen new drug treatments, to distinguish between specific effects of drugs and to explore the mechanisms underlying mania and BPD. Key words: Bipolar disorder, mood, depression, mania, behavioral battery, test battery, animal model, mice.

1. Background and Historical Overview The development of models for bipolar disorder (BPD) is hindered by many factors, including the cycling nature of the disorder, its heterogeneous clinical phenotype, and the lack of knowledge about its underlying pathophysiology (1–3). Previous attempts to develop a single comprehensive model for BPD had initial encouraging results but encountered significant validation and practicality impediments (e.g., (4–8)). Although recent mutant mice studies offer some progress, confounds related to the effects of targeted T.D. Gould (ed.), Mood and Anxiety Related Phenotypes in Mice, Neuromethods 42, DOI 10.1007/978-1-60761-303-9_16, ª Humana Press, a part of Springer ScienceþBusiness Media, LLC 2009

297

298

Flaisher-Grinberg and Einat

mutations prevent them from supplying a complete answer (9). For this reason, the common practice today is the use of separate models, developed in the context of depression, to explore the depressive pole of BPD and most frequently just a single model, the psychostimulant-induced hyperactivity test, to explore the manic pole of the disorder (10, 11). The psychostimulant-induced hyperactivity test holds significant validity as an animal model of mania. Specifically, psychostimulants such as amphetamine were found to induce measurable hyperactivity when administered at appropriate doses, and mood stabilizers treatment was found to inhibit psychostimulantsinduced hyperactivity (12). These effects are in line with human studies, showing that psychostimulants induce mania in susceptible individuals (13, 14) and that the mood stabilizer lithium can prevent the behavioral effects of stimulants (15, 16). Yet hyperactivity is only one of the behavioral domains of mania and is not unique to BPD, but is apparent in a variety of other disorders (e.g., attention deficit/hyperactivity disorder, ADHD (17)). Moreover, the use of this model is complicated by the fact that high doses of amphetamine cause stereotypy rather than elevated ambulatory activity (10, 18), which is reduced by dopamine receptor blockers such as antipsychotic drugs (18) and is suggested to model some of the domains of schizophrenia (19, 20). Given that the shortage of animal models for BPD and especially for its manic-pole obstructs the research related to the disorder (1, 21, 22), a novel strategy, which includes an attempt to develop and validate specific models representing separate domains of the disorder, has been recently suggested (11, 23– 27). In recent reviews, Einat (23, 25) examines symptoms frequently used in the diagnosis of mania in patients and points to the ones which are behavioral (or can be extrapolated to behavior), and therefore can be objectively measured and consequently modeled. These components include increased vigor, energy, activity or restlessness; abuse of drugs (measures of drug response and measures of hedonistic, reward seeking behavior); provocative, intrusive or aggressive behavior; extreme irritability; unrealistic beliefs in one’s abilities and powers and poor judgment (measures of risk-taking behavior or resilience to despair); reduced sleep; increased sexual drive; and distractibility and reduced ability to concentrate (measures of distractibility and switching) (modified from NIMH disorders site at http://www.nimh.nih.gov/health/ topics/bipolar-disorder/index.shtml). Other symptoms frequently used in the diagnosis of mania in patients are based solely on the patients’ verbalization of their feelings or mood and hence cannot be modeled. These components include excessively ‘‘high’’, overly good, euphoric mood; denial that anything is wrong; racing thoughts and talking very fast, jumping from one idea to the other; spending sprees and a lasting period of behavior

Mice Models for the Manic Pole of Bipolar Disorder

299

that is different from usual (modified from NIMH disorders site at http://www.nimh.nih.gov/publicat/bipolar). For the behavioral facets of the disease which might be modeled, Einat (23) reviews several existing tests, which were developed in different contexts, and suggests that they might be validated as models for these phenotypes. Clearly, none of these behaviors can be used by itself to diagnose mania/BPD in patients, and therefore none of the proposed animal tests that may model these behaviors can by itself represent the entire disorder. However, the same logic that combines several behaviors to a unified diagnosis asserts that an assemble of animal models can serve to explore the underlying biology of the disorder and to screen for new mood stabilizers (24, 26–28). The multifaceted approach to the modeling of PBD is in line with the present perception of the disorder, recognizing it as a heterogeneous disease which encompasses a range of symptoms, variable pathologies, and possible relationships with a number of endophenotypes (29). Importantly, such models for the different facets of the disorder can later be integrated into a coherent test battery of mania and/or of BPD (25, 29, 30). Although a test battery may not be an ideal solution compared to a theoretical single comprehensive model, it is possible that a comprehensive model will not be attainable at all (31) or at least not until further knowledge regarding the underlying biology of the disorder is obtained. It is suggested that a test battery could offer an intermediate tool in the attempt to gain such additional knowledge. Moreover, the drugs currently used to treat BPD (lithium, anticonvulsant mood stabilizers, atypical antipsychotics and to some extent, antidepressants) exert a variety of undifferentiated, nonspecific, therapeutic effects on the different components of BPD (32). Given that no preclinical in vivo model enables discrimination between the spectrum of effects of existing and newly developed drugs (1, 29), a battery of tests for specific domains of the disorder might prompt clarification and assist in defining the appropriateness of specific drugs for specific symptoms. Two critical steps in the development of animal models for the manic pole of BPD and in combining them into a battery are (1) to identify ways to induce manic-like behaviors in animals and (2) to establish the validity of these models as separate units and as part of a continuous battery. Whereas the initial selection of optional models can be based on face validity (a similarity between a behavior observed in manic patients and a behavior of the animal in the model), there is a need to test these tentative models for predictive validity (effects of drugs) and to search for possible construct validity (shared mechanisms). Manic-like behaviors can be induced by a variety of methods (genetic, pharmacological, and environmental (e.g., (33)). However, it had been suggested that there are significant advantages in

300

Flaisher-Grinberg and Einat

the identification of model animals for psychiatric disorders (27, 34). This approach suggests that it might be possible to use the basic tendencies of certain species or strains to behave in ways that are similar to a disease phenotype, in order to explore the underlying mechanisms of the disease and to screen for treatments (27, 34). In line with this viewpoint, a number of standard mice strains, which were found to be reactive to amphetamine and lithium administration in the psychostimulant-induced hyperactivity test (12), were screened for behaviors that might reflect different domains of mania. It was found that compared with C57Bl/6, CBA/J and A/J mice, Black Swiss mice demonstrated an aggregate of behaviors strongly associated with a manic-like phenotype including elevated risk taking, reward seeking/hedonism, aggression, resistance to despair, and a heightened response to amphetamine. These behaviors were tested using the black/ white box (BWB) and the elevated plus maze (EPM), the sweetsolution preference test, the resident–intruder test, the forcedswim test (FST), and the psychostimulant-induced hyperactivity test, respectively (27, 28, 35). Although none of these data were taken to imply that Black Swiss mice are ‘‘manic,’’ it did suggest that this strain may have advantages in attempts to model the domains of the disorder (27, 28, 35). Hence, the effects of the prototypic, structurally dissimilar, mood stabilizers lithium and valproate in the above-mentioned tests were assessed using Black Swiss mice. Both lithium and valproate are non-specific in their biological effects, the therapeutic mechanisms through which they act are not yet determined, and their efficacy is often augmented by other drugs as the clinical effect in not equable in all patients (32). However, the two drugs have been demonstrated to share a number of molecular targets (36), are efficacious as mood stabilizers in the treatment of BPD, and are considered to be more effective for the manic, rather than for the depressed, pole of the disorder (32). Results showed that both lithium and valproate inhibited the high sweet-solution preference (26) aggression, and amphetamine-induced hyperactivity demonstrated by Black Swiss mice, at doses that had no effect on spontaneous activity. In contrast, only valproate clearly and reliably reduced resistance to despair. Finally, the high risk-taking behavior demonstrated by Black Swiss mice tested in the BWB and the EPM was not affected by any of the mood stabilizers. In line with the principles of model validation (31, 37), tests that did not respond to any of the mood stabilizing drugs were not included in the current protocols. However, it is suggested that tests that were found to respond to one of the drugs but not the other could be a valuable tool in the attempt to distinguish the specific effects of drugs on separate facets of the disorder and to explore particular mechanisms related to its different domains. Therefore, the current chapter presents the

Mice Models for the Manic Pole of Bipolar Disorder

301

detailed protocols of models which were found to be responsive to one of the mood stabilizers: lithium, valproate, or both. The chapter also includes a protocol for assessing spontaneous activity since this test is critical for the interpretation of results from the other tests.

2. Mice Models for the Manic Pole of BPD – Protocols

2.1. Spontaneous Activity Test (Protocol 1) 2.1.1. Background Information

The protocols described in this section are essentially variations on basic protocols that are used in many laboratories for different purposes and include spontaneous activity evaluation (Protocol 1), sweet-solution preference evaluation (Protocol 2), resident–intruder test (Protocol 3), and forced-swim test (Protocol 4). The chapter does not include a specific protocol for the amphetamine-induced hyperactivity test as it is similar to protocols already described by others (11, 12). The specific descriptions included in the current chapter represent the methods used and validated in our laboratory with Black Swiss mice. It is possible that some of these tests can also be used with other strains, as it had been repeatedly shown that different behaviors and response patterns can be obtained with different mice strains. However, at this time no data are available to support this possibility. NOTE: All protocols using live animals must first be reviewed and approved by an Institutional Animal Care and Use Committee (IACUC) and must conform to governmental regulations regarding the care and use of laboratory animals. In the spontaneous activity test, mice are tested for locomotor activity in transparent plastic automated activity monitors for a 60-minutes session. This test is often used not only as an initial screening test for drug-induced alternation of neuronal functioning (38), but also to assess the influence of other manipulations, as well as baseline locomotion, novelty exploration and vigilance in non-treated animal. It may also be important to use this test when the effects seen in a given experiment are suspected to be generalized (e.g., sedation) or to confound other behaviors demonstrated by the animal. Given that the effects of many manipulations are time-dependent, it is important that the spontaneous activity test will be long enough to detect such changes. Moreover, it is advised to examine not just the total activity level for the entire session, but also the changes in behavior across time, as some manipulation may induce time-dependant changes (39, 40).

302

Flaisher-Grinberg and Einat

2.1.2. Animals and Housing

Single house 9–12-weeks-old male mice, weighing 20 to 40 g, in transparent plastic cages (30  19  13 cm) with wood shavings bedding, ad libitum standard rodent chow and water. Mice are to be housed in a colony room at 22–1C with a standard (non-reversed) light/dark cycle (lights on 07:30–19:30). Permit at least 7 days of habituation prior to behavioral testing. Conduct the test during the light phase of the cycle, starting at least 1 hour after ‘‘lights on’’ and ending at least an hour before ‘‘lights off.’’ Keep lights in the testing room similar to the lights in the colony room (standard fluorescent lights). 1. Activitymonitors:Infraredphotocell-basedactivitymonitors(transparent plastic automated activity monitors, 50  25  20 cm, with infrared beam crossings spaced at 2.54-cm interval on the x plane [horizontal arena] and 4 cm above the floor, covered with a perforated plastic lid) (Opto-M3, Columbus Instruments, Columbus, OH or equivalent), interfaced with a computer running C-I MultiDevice Software v. 1.30 or equivalent.

2.1.3. Equipment and Materials

2. White-noise machine such as ‘‘sound therapy machine,’’ model SU1, programmed to ‘‘white noise’’ (Conair, Conair Corporation, Glendale, AZ, or equivalent). 3. Paper towels and 10% ethanol solution. 1. Set up the computer and interface according to the manufacturer’s instructions and attach the appropriate cables to the activity monitors. Calibrate the system before placing the test subject in the monitors. * No bedding is to be placed in the activity monitors during the test and no food or water is to be provided.

2.1.4. Setting

2. Set the computer to record the locomotor response for 60 minutes and to collect activity for each 10-minutes interval within the session separately. 3. Using paper towels, wipe the activity monitors with 10% alcohol solution. Allow the alcohol to evaporate fully before using the apparatus. Clean the activity monitors thoroughly before each use to eliminate any stimuli left in the chamber by the previous subject. 4. Activate white-noise machine. 5. At least 15 minutes prior to testing, bring the animals in their home cages into the experimental room. 6. If drugs or specific treatments are to be administered, randomly assign animals to the different treatment groups. 7. Administer drugs or treatments and place mice in the activity monitors, or, if testing should be delayed for a set period of time after treatments, place animals back in their home cages until the appropriate time has elapsed. *

is significant to the text.

Mice Models for the Manic Pole of Bipolar Disorder

303

8. Repeat the testing with additional groups of mice as needed. 9. When all sessions are completed, turn off all experimental equipment, save the data to the computer, and return the animals to the home cages and the colony room. 2.1.5. Data Analysis

Analyze the data for total activity or for total ambulatory activity (depends on the measures provided by the manufacturer/specific equipment; see Critical parameters and Troubleshooting section) in 10-minutes time sessions. This analysis will provide you with six data points for each animal, representing 10-minutes time-bins within the test. Analyze the data using repeated measures ANOVA with time-bins as a repeated measure factor.

2.1.6. Anticipated Results

Black Swiss mice tested for spontaneous activity in a 60-minutes session demonstrate activity averaging at about 3,000 ambulation counts. This level of activity, which is similar to that demonstrated by the commonly used C57Bl/6 strain, is also far from either floor or ceiling levels and is suitable for evaluating the effects of a variety of manipulations. As with many other strains, Black Swiss mice show habituation to the test condition, and the activity levels are reduced across the session. It is important to recognize that the amount and duration of activity observed in a particular experiment depend on a number of factors, including noise and illumination in the experimental room. Therefore, significant efforts should be invested in unifying the conditions across settings.

2.1.6.1. Baseline Behavior

2.1.6.2. Drugs and Test Compounds Effects

Drugs effects on locomotion can be expressed as elevated or reduced spontaneous activity. For instance, with few exceptions, antipsychotic drugs decrease spontaneous locomotor activity in rodents (e.g., (41)). Other drugs, such as amphetamine, increase this measure (12). Neither lithium nor valproate, in the doses and administration schedules used for validation of the protocols described in this chapter, was found to have a significant effect on the spontaneous activity of Black Swiss mice (for example, see Fig. 16.1). Importantly, some drugs may increase locally oriented activity (such as stereotypy) rather than environment-directed locomotion. The activity of animals treated with such drugs (e.g., apomorphine (42)) may not be clearly detected with simple activity monitors, and it is important therefore to observe animals during testing and to verify that the equipment is appropriate for the type of behavior performed. Most activity monitors have a measure for locally oriented activity, usually defined as repetitive breaks of the same set of photocell beams (38).

Flaisher-Grinberg and Einat

Abmulation counts

304

1200 900

Vehicle Lithium 100 mg/kg Lithium 200 mg/kg

600 300 0 0–10 10–20 20–30 30–40 40–50 50–60

Time (min)

Fig. 16.1. Ambulation scores (mean+SE) of Black Swiss mice, treated twice daily with 100 or 200 mg/kg lithium, administered I.P., with last injection administered 30 minutes prior to testing. Data are presented for each 10-minutes time-bin of a 60-minutes session. ANOVA: Treatment effect, F(2, 45)=0.24, p=0.78; Time-bin effect, F(5, 225)=144.39, p<0.0001, Treatment  Time-bin effect, F(10, 225)=1.00, p=0.44.

2.1.6.3. Time Considerations

The total time needed for testing spontaneous activity depends on the number of available activity monitors and the number of animals tested. For many of the experiments performed in our laboratory, which included four treatment groups, n=12/group, and eight activity monitors, the total time was approximately 6.5 hours (not including drugs preparation time).

2.1.7. Critical Parameters and Troubleshooting

Some mice strains can be extremely hypoactive or hyperactive. Similarly, some drugs and treatments can induce extreme hyperactivity or hypoactivity. In such cases, there is a clear possibility for a floor or ceiling effects in this test, and the effects of additional manipulations can be overshadowed. It has been found in our laboratory that the level of spontaneous activity demonstrated by Black Swiss mice can be manipulated both up and down by different interventions (see also (12)).

2.1.7.1. Floor and Ceiling Effects

2.1.7.2. Test Session Duration

A 60-minute-long test session is suggested to offer a good balance between the need to follow the effects of different manipulations over time and the need to avoid overshadowing of effects by habituation across time. However, given that some manipulations might have an extremely prolonged effect on mice’s behavior, the length of the session can be extended. Moreover, if there is a reason to expect that the manipulation will result in very short term changes (minutes), then the length of the time-bins within the session can be shortened and the activity can be recorded for time-bins shorter than 10 minutes (most activity monitors will allow sub-sessions of 1 minute or more).

2.1.7.3. Environmental Conditions and Baseline Behavior

Considering the sensitivity of mice to environmental stimuli, it is important to maintain constant levels of sound and light across the testing session and avoid movement in the room. A ‘‘white-noise’’

Mice Models for the Manic Pole of Bipolar Disorder

305

machine is therefore strongly recommended as well as making sure that no other activity is conducted in the testing room. It is also important to verify that no object blocks the infrared beams and that the activity monitors themselves are well attached to the surface and will not move as a result of the mouse activity. 2.1.7.4. Specific Measurements by Different Activity-Monitoring Systems

Many activity-monitoring systems (including the Opto-M3) will record more than one measure during the session. Most frequently the systems will have some measures that differentiate between exploratory (environment-oriented) locomotion and locally oriented (stereotypical) activity. For the Opto-M3, these measures are ‘‘total activity’’ and ‘‘ambulation score.’’ Under these definitions (or similar names), the ‘‘total activity’’ score represents the total number of beams crossed during the session or parts of the session. The ‘‘ambulation’’ score represents the total activity score minus the repetitive and consecutive crossing of the same infrared beam. Therefore, the ambulation score might have a better representation of the spatial exploratory behavior as it eliminates locally oriented, stereotypic-like activity. The difference between these scores might be important for many manipulations in which locally oriented movements are induced in laboratory animals (43). However, in experiments testing mood stabilizers, we found a high correlation between these measures (unpublished data), indicating that at the doses and administration schedules tested, these drugs had no effect on either exploratory or locally oriented activity.

2.2. Sweet-Solution Preference Test (Protocol 2)

Reduced sweet-solution preference after depressenogenic manipulations had been well validated in the context of depression research (44–47). Recently, high-baseline saccharin preference was validated as a measurable mouse model for elevated reward seeking (26, 27), an important domain of manic behavior (48, 49). It has been demonstrated that the structurally dissimilar mood stabilizers lithium and valproate reduce saccharin preference in mice with high-preference baseline, while the prototypic antidepressant imipramine had no effects (26). The protocol presented here includes 48 hours of continuous exposure to the saccharin solution, but it can be extended to 4 days if needed. It is, however, suggested not to shorten the protocol, so that the mice will have enough time to become familiarized with the new taste and acquire a preference for it. The 2-days length protocol has been demonstrated to be long enough to distinguish the effects of mood-stabilizing drugs in this procedure (26), but it is possible that with other drugs a longer procedure will be needed.

2.2.1. Background Information

2.2.2. Animals and Housing

The animals and housing conditions are as described above for the spontaneous activity protocol (see Protocol 1).

306

Flaisher-Grinberg and Einat

2.2.3. Equipment and Materials

1. Tight-lid bottles for both saccharin and water. 2. A source of boiling water (to ease dissolving the saccharin). 3. Saccharin, 98+%, SIGMA, St. Lewis, MI. 4. An accurate scale (at least 0.1 g accuracy). 5. Permanent markers.

2.2.4. Setting

Preparation of saccharin solution: There is a specific preferred range of saccharin concentrations with variations between different mice strains. In Black Swiss mice, the data show that concentrations between 0.1% and 1% saccharin are suitable and result in a significant preference (26, 35). The protocol described here utilizes a 1% saccharin solution that was found in our studies to provide the best distinction between the effects of the drugs used for validation. 1. Per day, drug-naive Black Swiss mice will consume an amount of saccharin solution which approximately equals their body weight. Therefore, in order to avoid replenishing the bottle within the 48 hours of the experiment, it is recommended to fill each saccharin bottle with approximately 100–150 ml. If a longer procedure is used, then it is recommended to fill the bottles with enough liquid for the entire length of the experiment. 2. To prepare the saccharin solution, mix the amount of saccharin needed for all mice in a large container with a measured amount of boiling water (to ease dissolving), and add tap water to reach the needed concentration. For example, for an experiment in which four groups of 12 mice per group are tested, there is a need for approximately 7200 ml saccharin solution, calculated for 48 mice and for 2 days (48  150 ml = 7200 ml). To reach 1% saccharin solution, 72-g saccharin powder are mixed with at least 2000 ml of boiling water and stirred well. Than, tap water is added to attain a total volume of 7200 ml. 3. Mark the saccharin bottles clearly using a permanent marker, so that it will be easy to distinguish them from the similar water bottles. 4. Fill bottles with the appropriate amount of saccharin solution. 5. Let saccharin bottles chill to room temperature. 6. Fill water bottles to a level that will be enough for the test days, to avoid the need to add water during test days. 7. If drugs are to be administered, randomly assign animals to the different treatment groups. Administer drugs and place mice in the home cages. 8. In the colony room, for each mouse, weigh the mouse, saccharin, and water bottle, and document the data. Adjust saccharin and water bottles in place and verify that the lids are

Mice Models for the Manic Pole of Bipolar Disorder

307

tight and there is no leakage. Most commonly, the bottles will be placed through the food grid of the mice’s home cage. Make sure that the tips of the bottles are reachable for the mice. 9. Twenty four and 48 hours after the start of the experiment, weigh saccharin and water bottles and document. While weighing the saccharin bottles, plug the tip with your fingers and shake well to verify that the solution is still homogenous and that the tips are not plugged (although this is very rare and not as probable as when using sugar). Take care not to spill any liquids from the bottles during weigh times. 10. At the end of the test (48 hours time point), weigh mice in addition to weighing the bottles. 2.2.5. Data Analysis

Calculate daily saccharin and water consumption by subtracting bottles’ weight from their weight on the previous day (e.g., saccharin consumption Day 1 = saccharine bottle’s weigh at baseline minus saccharine bottle’s weigh at 24 hours). Use these numbers to compute daily saccharin preference by dividing saccharin consumption by total liquid consumption (saccharin solution + water). Hence, a computed score of 0.5 demonstrate no preference for any of the liquids, a ratio higher than 0.5 demonstrate preference for the saccharin solution, while a ratio lower than 0.5 demonstrate preference for tap water. Analyze the data using repeated measures ANOVA with day as a repeated measure factor. Body weight measurements can be used to explore additional relationship between mice’s weight and total liquid consumption or saccharin preference. When drugs are administered, their effects on mice’s weight, total liquid consumption, and saccharin preference can be analyzed using mixed ANOVA with treatment as a main factor and day as a repeated measure factor.

2.2.6. Anticipated Results

Black Swiss mice supplied with 1% saccharin solution display relatively high-saccharin preference, ranging between 0.6 and 0.8 preference scores across the days of the experiment (26). This behavior is different compared to the commonly used C57Bl/6 strain, which demonstrate such preference ratio only on the first day of the protocol, and to other strains (e.g., A/J and CBA/J mice, which demonstrate a lower preference ratio (28)). This level of saccharin preference provides a suitable baseline for evaluating the effects of a variety of manipulations, especially manipulations that might reduce preference.

2.2.6.1. Baseline Behavior

2.2.6.2. Drug and Test Compound Effects

The high preference for saccharin solution demonstrated by Black Swiss mice was recently validated as an animal model for the elevated reward-seeking domain of mania (26, 35). Specifically, the high preference for saccharin in these mice was reduced following treatment with the mood stabilizers lithium and valproate but

308

Flaisher-Grinberg and Einat

Saccharin preference

was not affected by the antidepressant imipramine (26). To the best of our knowledge, the effects of mood stabilizers on highbaseline sweet-solution preference were not evaluated yet in other strains. Figure 16.2 is an example of the effects of lithium administration (two injections/day, 200 mg/kg) on saccharin preference in Black Swiss mice.

0.9

0.7

Vehicle Lithium 200mg/kg

*

*

0.5 Day 1

Day 2

Day

Fig. 16.2. Saccharin preference ratio (mean+SE) of Black Swiss mice, treated twice daily with 200 mg/kg lithium, administered I.P., with last injection administered 120 minutes prior to bottles’ weighing, across the 2 days of the test. Preference is computed as daily consumption of saccharin solution divided by daily consumption of liquids (saccharin solution + water). *Significantly different from the vehicle-treated group on each day of the test (p<0.05). ANOVA: Treatment effect, F(1,29)=10.15, p=0.003. Post hoc LSD, lithium different than vehicle on each day.

2.2.6.3. Time Considerations

For many of the experiments performed in our laboratory and included 48 mice, the time was approximately 1 hour to prepare saccharin solution and bottles on the first day of the procedure, and approximately 1 hour for weighing mice and bottles on each day of the procedure (not including drugs preparation time).

2.2.7. Critical Parameters and Troubleshooting

As mentioned above, different mice strains have different baseline preference for saccharin solution as well as differences in preference for the concentration of the solution. At this time, the validation of the test using mood-stabilizing drugs was conducted only with Black Swiss mice (26). However, it is clearly possible that this procedure will be equally effective with other strains as long as they have a high-baseline preference for saccharin solution, and that the concentration they prefer can be identified.

2.2.7.1. Strain of Animals

2.2.7.2. Test Length

The validation of this test was conducted with 2-days and 4-days protocols. It has been found that in Black Swiss mice, preference for saccharin remains pronounced across all days of both protocols. Also, the mood stabilizers lithium and valproate were found to exert similar effects in both protocols, suggesting that a 2-days

Mice Models for the Manic Pole of Bipolar Disorder

309

protocol is sufficient for drug-induced attenuation of saccharin preference to be established (26). This shortened protocol is advised when the drugs being tested are known to have specific side effects. For example, both lithium and valproate induce polydipsia, and hence electrolyte imbalances (50–52), and require supplementation of salts if administrated for a longer period. However, it is possible that with different mice strains or different compounds, the longer procedure might have an advantage. Specifically, if certain mice strains are found to demonstrate neophobia toward the saccharin bottles, then it is advised to employ longer habituation periods. Importantly, as some drugs might be found to exert an effect only after long administration regimen, the length of the protocol can potentially be adapted according to necessity or drug administration should start days before testing. 2.2.7.3. Environmental Conditions and Baseline Behavior

Since consumption of fluids is the primary measure in the procedure, it is critical to avoid any spillage or leakage from the bottles. Any movement of the cages should therefore be done very carefully, and it is recommended that the standard care of animals (changing bedding, adding food etc.) will not be done during the days of the test. Clean bedding and sufficient food should be supplied before the start of the procedure. Given that the location of the saccharin and water bottles in the cage can also have an effect on mice behavior, it is recommended to counterbalance locations of saccharin and water bottles across test groups.

2.2.7.4. Housing of Animal

In order to be able to accurately assess liquid consumption patterns, mice have to be housed singly during the procedure. As single housing has some effects on mice behavior (e.g., elevated aggression in later sessions (53)), this factor has to be taken under consideration.

2.2.7.5. Drugs Effects on Saccharin Consumption

When drug effects are evaluated, it is not always easy to determine whether changes in sweet-solution preference result from alterations in reward seeking or in the solution’s stimulus intensity. In order to distinguish between these two possible effects, there is a need for a multi-concentrations experimental design which allows the analysis of the effects of several drug doses on shifting the concentration/preference curve.

2.2.7.6. Mice Weight and Weight Loss

Various differences in sweet-solution preference may be secondary to factors related to consummatory behaviors. One such factor is the baseline body weight of the mice or changes in body weight induced by drugs administration or other manipulations (e.g., stress (46)). Given that it has been previously demonstrated that reward-seeking behavior might be connected with body weight (46), the weights of mice should be measured at the start and end

310

Flaisher-Grinberg and Einat

of the procedure. If significant differences are apparent between groups, either for baseline weight or for weight change during the procedure, these findings have to be addressed. 2.3. Simplified Resident–Intruder Test (Protocol 3) 2.3.1. Background Information

The resident–intruder test is frequently used in the context of aggression research (e.g., (54–56)), and a simplified version of this test was recently demonstrated to be a measurable animal model for the aggression/intrusion facet of manic behavior (24). In this procedure, an ‘‘intruder’’ mouse (i.e., intruder) is placed in the home cage of a ‘‘resident’’ mouse (i.e., resident), and their behavior is monitored for 10 minutes. As repeatedly demonstrated (54–56), the residents’ interaction with the intruders includes both aggressive and non-aggressive behaviors. In the procedure described in this chapter, in order to increase residents’ baseline-aggressive behaviors, intruder mice are approximately 2 weeks younger than residents, about 5 g lighter, and are group-housed, whereas residents are singly housed for at least 2 weeks. In contrast to the standard resident–intruder test, but in similar to the previous version demonstrated in the context of BPD (24), mice in the simplified protocol are not allowed to fight freely during the session. When (and if) an attack becomes vicious and significant biting starts, the experimenter uses a probe to separate the mice from each other for a few seconds, after which the probe is withdrawn and the mice are allowed to continue their interactions. This protocol was established in order to avoid serious harm and injury and minimize pain and distress. Because of this design, measures of total time spent in aggressive interactions cannot be scored as done in other protocols of the resident– intruder test (57). Accordingly, the primary measures in the current protocol are the number of aggressive and non-aggressive interactions during the session. It has been previously demonstrated that these measures are sensitive enough to detect the effects of mood stabilizers (24). The comparison of manipulation’s effects on both behavior types is critical for the differentiation of general changes in alertness/vigilance/sociability of animals from a specific effect on aggression (see Drug and Test Compound Effects section). Importantly, the protocol includes a 10-minutes session, but given that some strains of mice will gradually switch from social behaviors to more aggressive behaviors (58), this time interval can be divided into shorter sessions. Such division (to at least two, 5minutes intervals) can assist in detecting time-induced differences in interactive patterns (e.g., escalation of aggression) and avoiding overshadowing of the later aggressive behavior by the earlier, frequent social behaviors. Importantly, this division is not always necessary as some species and strains were found to immediately demonstrate aggressive behaviors (59).

Mice Models for the Manic Pole of Bipolar Disorder 2.3.2. Animals and Housing

2.3.3. Equipment and Materials

311

The animals and housing conditions are as described above for previous protocols (see Protocol 1). Additionally, intruder mice, 2 weeks younger and approximately 5 g lighter than residents are to be housed in group-caging. As mentioned above, resident mice should be singly housed for at least 2 weeks before testing. 1. A video camera interfaced with a computer and placed in a way that the recording of all mice’s behavior is possible. 2. White-noise machine such as ‘‘sound therapy machine’’; model SU1, programmed to ‘‘white noise’’ (Conair, Conair Corporation, Glendale, AZ, or equivalent). 3. A probe to separate mice, in case vicious fighting begins.

2.3.4. Setting

1. Set the video camera in a position that will allow proper recording of a resident’s cage (depending on the video camera and lens, more than one cage can be recorded at the same time, if placed next to each other. However, an opaque partition should be placed to block sight between the cages, and the noise factor is to be controlled). 2. Set the computer to record the mice’s behavior for 10 minutes. 3. Transfer the resident animal in its own home cage into the experimental room. 4. If drugs are to be administered, randomly assign animals to the different treatment groups. Administer drugs and place mice in the home cage. If testing should be delayed for a set period of time after treatments, place animals back in their home cages until the appropriate time has elapsed. 5. Remove the cage lid and wait for 5 minutes to allow the resident mice to acclimate and adapt to the open cage (the removal of the cage cover is done in order to allow easy recoding as well as fast access in case of severe aggression). 6. Mark the intruder mice in a way that will permit differentiation between a resident and an intruder while scoring the recordings. Marking can be done by coloring the fur/tail or by taping a piece of tape to the tail. 7. Activate white-noise machine. 8. Introduce the intruder mouse into the resident’s home cage and record behavior for 10 minutes. The experimenter must be present throughout the entire session, and make sure that if any of the mice attempts to climb out of the cage, it will be gently pushed back using the probe. Also, if a severe attack starts, the experimenter should separate the mice for a few seconds using the probe.

312

Flaisher-Grinberg and Einat

9. At the end of the session, move the resident mouse aside using the probe, remove the intruder mouse and place it back in its home cage, cover the resident’s cage, and return all mice to the colony room. If a tape was used to mark intruders, remove it gently before returning the animal back to the colony room. 2.3.5. Scoring and Data Analysis

Use digital recordings to score the number of aggressive and nonaggressive interactions for the 10-minutes session. If there is a reason to expect time-dependent differences in the pattern of aggressive behavior between groups, then it is possible to score the behaviors in shorter sub-sessions (for instance, two 5-minute sub-sessions). Aggressive interactions are defined as attempts to bite, actual bite, boxing postures, and wrestling postures. Nonaggressive interactions are defined as other types of body contact including sniffing, allogrooming, and body contact (24). In this protocol, the behaviors that are performed by the resident mice when they are not interacting with the intruders are not scored. Moreover, the procedure described here is designed to test the aggression of the resident mice only and hence, behaviors performed by the intruder mice are not scored. It is, however, possible that in different contexts, the scoring of non-social behaviors and of behaviors performed by the intruder mouse can be valuable (e.g., (60)). As mentioned above, in this test it is important to differentiate a generalized effect of a manipulation on the alertness/vigilance/ sociability of animals from a specific effect on its aggression. A specific effect of a manipulation on aggression is expected to influence aggressive, but not non-aggressive, behaviors in the test. To demonstrate a clear picture of a specific effect on aggression, it is recommended to compute an ‘‘aggression ratio’’: the number of aggressive behaviors divided by the total number of interactions (aggressive + non-aggressive). Hence, a computed score higher than 0.5 demonstrate higher rate of aggressive behaviors compared with non-aggressive behaviors, while a computed score lower than 0.5 demonstrate the opposite behavioral pattern. Alternatively, these behaviors can be analyzed using ANOVA with type of behavior (aggressive/non-aggressive) as a main factor. If the 10-minutes session was divided into sub-sessions, the results could be analyzed with an additional repeated measures factor of time intervals.

2.3.6. Anticipated Results

The ‘‘aggression ratio’’ of Black Swiss mice, introduced with an unfamiliar, younger intruder mouse for a 10-minute session range between 0.6 and 0.7, meaning that they show more aggressive than non-aggressive interactions. This aggression rate is different for the commonly used C57Bl/6 strain, which were found in our laboratory to demonstrate an ‘‘aggression ratio’’ of about 0.3–0.5

2.3.6.1. Baseline Behavior

Mice Models for the Manic Pole of Bipolar Disorder

313

(unpublished data; (24)). The aggressive rate of Black Swiss mice can therefore provide a suitable baseline for evaluating the effects of a variety of manipulations suggested to reduce aggression. 2.3.6.2. Drug and Test Compound Effects

Drugs effects on aggression in this test will be expressed as a change in the number of aggressive interactions with no, or less pronounced, change in non-aggressive behaviors. A non-specific drug effect can be expressed as a similar change in both parameters. For instance, it has been demonstrated that chronic treatment with the mood stabilizers lithium and valproate reduces aggressive interactions but has no effects on non-aggressive interactions in C57Bl/6 mice (24). Similar effects have been found in Black Swiss mice (unpublished data). These effects are in line with studies showing that mood stabilizers reduce aggressive behaviors in different protocols, strains and species, including humans (61–67). Figure 16.3 is an example of the effects of lithium administration (two injections/day, 200 mg/kg) on ‘‘aggression ratio’’ in Black Swiss mice.

Aggression ratio

0.8

0.6

* 0.4

0.2

0.0 Vehicle

Lithium 200mg/kg Treatment

Fig. 16.3. Aggression ratio (mean+SE) of Black Swiss mice, treated twice daily with 200 mg/kg lithium, administered I.P., with last injection administered 30 minutes prior to testing. Aggression ratio is computed as the number of aggressive behaviors divided by the number of all interactions (aggressive + non-aggressive). *Significantly different from the vehicle-treated group (P<0.05). Treatment effect, t(13)=2.45, p=0.03.

2.3.6.3. Time Considerations

The total time needed for the simplified resident–intruder protocol depends on the number of cages which can be recorded at the same time. For many of the experiments performed in our laboratory, which included 48 mice and a recording system of two cages at the time, the total time was approximately 8 hours (not including drugs preparation time). It is important to note that significant time is also needed for the scoring of the behavior from digital recording. Scoring time could be twice as much as the actual recording time depending on the experience of the scorer.

314

Flaisher-Grinberg and Einat

2.3.7. Critical Parameters and Troubleshooting 2.3.7.1. Strain of Animals

So far, the specific procedure described in this protocol was tested for the effects of mood stabilizers with Black Swiss and C57Bl/6 mice as both residents and intruders (resident and intruder of the same strain). Clearly, the strain of the resident and the strain of the intruder can have significant effects on the patterns of interactions in this test.

2.3.7.2. Age, Weight, and Housing of Animals

As discussed above, the age and weight of the mice can significantly affect their interaction patterns. To increase the likelihood of aggression in resident mice, it is recommended to use intruders that are younger and a few grams lighter than the residents. Yet very young mice (less than 8-weeks old) are not recommended because their behavior is different than the behavior of adult mice and because they might invoke a different set of behaviors by the adult residents.

2.3.7.3. Test Session Duration

A 10-minutes session is recommended to ensure the occurrence of aggressive interactions, as it might develop gradually. The division of the 10-minute session into two or more sub-sessions is an option that had been discussed above.

2.3.7.4. Environmental Conditions and Baseline Behavior

As with the previous tests described in this chapter, the sensitivity of mice to environmental stimuli should be considered, and the test room should be kept with constant lights, sounds, and minimal movement by the experimenter.

2.3.7.5. Reusing of Intruder Mice

In order to reduce the number of animals used in the procedure, it is possible to use each intruder twice. It is clear that a repeated exposure of a mouse to the resident–intruder situation has significant and long-lasting effects on its behavior and on underlying stress mechanisms (e.g., (60, 68)), and that these changes might in turn affect the mice’s interaction with the resident. However, in experiments performed in our laboratory, no significant differences were identified in the behavior of the residents when intruders were used twice (unpublished data). If intruders are to be used twice, then there must be a recovery period between the first and second introductions to the resident–intruder situation (at least 1 hour), and the number of the introduction (first/second) should be counterbalanced across the test groups of the residents.

2.4. The Forced-Swim Test as a Model for Resilience to Despair (Protocol 4)

The forced-swim test (FST) is frequently used in the context of depression and antidepressants research. Specifically, when mice are placed in a relatively narrow water container from which there is no escape, they demonstrate a common behavioral pattern. This pattern consist an initial period of vigorous activity, followed by an increased amount of immobility, characterized by performance of only the minimal movements necessary to keep their heads above water. This increased immobility was

2.4.1. Background Information

Mice Models for the Manic Pole of Bipolar Disorder

315

suggested to represent ‘‘behavioral despair,’’ a common domain of major depression, and is significantly affected by depressogenic manipulations and antidepressant drugs (e.g., (69–77)). As a mirror image of the despair-like immobility, it is possible that high-baseline activity and minimal immobility in the FST may represent resistance to despair, similar to the increased vigor commonly diagnosed in manic patients (78, 79). Accordingly, mice that show baseline low-immobility (high activity) score could potentially be considered as a model for this domain of mania. When such mice are tested in the FST, it is also expected that mood stabilizers will increase their immobility time. This is in contrast to mice that show baseline high-immobility, where mood stabilizers can be expected to reduce immobility time (80). 2.4.2. Animals and Housing

The animals and housing conditions are as described above for previous protocols (see Protocol 1). 1. A transparent water-sealed cylindrical container (plastic or glass) measuring about 18–20 cm in diameter and 25 cm in height.

2.4.3. Equipment and Materials

2. A video camera interfaced with a computer and placed in a way that the recording of all mice’s behavior is possible. 3. A water source with control over water temperature. 4. A thermometer that can be used in water. 5. White-noise machine such as ‘‘sound therapy machine’’; model SU1, programmed to ‘‘white noise’’ (Conair, Conair Corporation, Glendale, AZ, or equivalent). 6. Paper towels. 2.4.4. Setting

*

As for the previous protocols, mice should have at least 1-week habituation in their colony rooms before the start of the experiment. Experiments are conducted during the light phase of the light/dark cycle and under similar light conditions as in the colony room. 1. Set the video camera in a position that will allow proper recording of the cylindrical container (depending on the video camera and lens, more than one container can be recorded at the same time, if placed next to each other. However, an opaque partition should be placed to block sight between the containers, and the noise factor is to be controlled). Recording from the side is recommended as it allows the best view of mice behavior in the cylinders. *

is not significant to the text.

316

Flaisher-Grinberg and Einat

2. Set the computer to record the mice’s behavior for 6 minute. 3. If drugs are to be administered, randomly assign animals to the different treatment groups. Administer drugs and place mice in the home cage. If testing should be delayed for a set period of time after treatments, place animals back in their home cages until the appropriate time has elapsed. 4. Set up cylindrical container(s), filled to a depth of 15+1 cm with tap water at 22–1C. The depth of the water is to be sufficient to ensure that the mice could not escape or touch the floor of the container. 5. Activate white-noise machine. 6. Place a mouse into the cylinder and record behavior for 6 minutes. 7. When session is completed, remove the mouse from the water, dry it with a paper towel, and place it back in its home cage. Replace the water in the container before inserting the next mouse. 2.4.5. Scoring and Data Analysis

The last 4 minutes of the 6-minutes recorded session are considered the test session and should be scored continuously (no sampling). The scoring parameters are (1) total time of activity including swimming (defined as correlated movements of all four limbs) and struggling/climbing (defined as intensive movements of all paws, mostly against the container wall) and (2) immobility (floating with only minimal movements needed to keep head above water). Different ways to score the behavior in the FST were introduced over the years in the context of depression research and might also be appropriate, including differentiation between swimming and climbing behaviors (81). Also, automated scoring systems (82) can be used to speed up the process. The current protocol was validated using both manual and automated scoring. Analyze the data using one-way ANOVA with manipulation or treatment as a main factor.

2.4.6. Anticipated Results

Immobility time of Black Swiss mice during the last 4 minutes of the FST ranges between 2 and 20%. This immobility time is remarkably low compared with most commonly used mice strains (see Table 16.1). It is therefore suggested that the Black Swiss strain might be appropriate for modeling increased vigor and resilience to despair.

2.4.6.1. Baseline Behavior

2.4.6.2. Drug and Test Compound Effects

In order to validate this model in the context of manic-like behavior, the effects of the mood stabilizers lithium and valproate were tested as well as the prototypic antidepressant imipramine. The data of these preliminary validation experiments show that at doses that had no significant effect on spontaneous activity, valproate

Mice Models for the Manic Pole of Bipolar Disorder

317

Table 16.1 Immobility time for a number of the commonly used male mice strains tested in the standard FST procedure, as reported in literature Strain

Immobility time mean across studies and range

References

C57Bl/6

175 (125–220)

(57, 70, 107–112)

DBA/2

112 (100–124)

(110, 112)

129 Sv (emJ or Ev)

135 (85–180)

(57, 110, 113)

BALB/C

127 (110–144)

(110, 112)

NIH or Albino Swiss

145 (52–225)

(72, 108, 110, 114, 115)

Swiss Webster

123 (80–160)

(110, 116, 117)

significantly increases immobility time in Black Swiss mice whereas lithium shows a similar, but less stable, trend. In contrast, impiramine had no effects on the immobility time of Black Swiss mice. Whereas these data should first be replicated, it is suggested that mood stabilizers might increase immobility time in the FST in mice with a low-immobility baseline (an anti-manic effect of mood stabilizers). These findings are unique because mood stabilizers were previously found to decrease immobility in mice that have high-baseline immobility score (80). Importantly, although imipramine had been repeatedly demonstrated to increase activity time in strains with high-baseline immobility (83), in the experiments performed in our laboratory, it could be that the lack of effect for imipramine resulted from a ceiling effect related to the high-activity level of the Black Swiss mice. Figure 16.4 is an

Immobility time (sec)

160

*

120

80

40

0 Vehicle

Valproate 100mg/kg Treatment

Fig. 16.4. Immobility time in the FST (mean+SE) of Black Swiss mice, treated twice daily with 100 mg/kg valproate with last injection administered I.P., 30 minutes prior to testing. *Significantly different from the vehicle-treated group (p<0.05). Treatment effect, t(12)=7.45, p<0.001.

318

Flaisher-Grinberg and Einat

example of the effects of valproate administration (two injections/ day, 100 mg/kg) on the ‘‘immobility time’’ of Black Swiss mice in the FST. 2.4.6.3. Time Considerations

The total time needed for the described FST protocol depends on the number of cylinders which can be recorded at the same time. For many of the experiments performed in our laboratory, which included 48 mice and a recording system of two cylinders at the time, the total time was approximately 6 hours (not including drugs preparation time). It is important to note that significant time is also needed for the scoring of the behaviors from digital recording. Scoring time could be twice as much as the actual recording time depending on the experience of the scorer.

2.4.7. Critical Parameters and Troubleshooting

As mentioned above, different strains perform significantly different in the FST. The Black Swiss strain demonstrated relatively low amount of despair-like behaviors in the test and was therefore selected as an appropriate strain for the evaluation of anti-manic interventions.

2.4.7.1. Strain of Animals

2.4.7.2. Environmental Conditions and Baseline Behavior

As explained above, constant lighting, low-noise levels, and minimal activity in the testing room should be kept. When testing more than one mouse at a time, it is important that the mice should not be able to see each other. Also, the water temperature in the cylinder is important and should be maintained at approximately 22–1C.

2.5. Discussion

The protocols included in this chapter are all adaptations of tests that were developed in different contexts and had been recently validated in the context of the different domains of mania. Each of these protocols was found applicable for testing the effects of a variety of manipulations on a specific domain of the disease, and as such, they introduce an advantage in the attempt to study the disorder and possibly its specific endophenotypes. The next step in this line of research will be the aggregation of these tests into a valid, unified protocol in which one batch of mice could be tested sequentially in a number of tests. A sequential battery of tests, utilizing the same mice, will allow the achievement of a fast-operated, cost-effective screening method for current and new drug treatments. The development of an expeditious test battery is important, since it has been repeatedly claimed that behavioral testing is a bottleneck for the development of new medications and that the behavioral field is lagging behind the speedy and high-throughput methods that were developed for the molecular and cellular stages of drug development (22). Moreover, pending the validation of a sequential battery, it will significantly reduce the number of animals needed for the screening of compounds with mood-stabilizing effects. Therefore, it will

Mice Models for the Manic Pole of Bipolar Disorder

319

support the efforts of the scientific community to reduce the number of animals used for experimentations. Data collected in our laboratory so far suggest that it may already be possible to combine the tests presented above into a comprehensive tests battery (28, 35). However, additional research must focus on the validation of these tests in a coherent battery for mania and on the addition of tests for the depression pole of BPD, in an attempt to form a standard battery of models for the critical domains of BPD. Such a battery could be used as a tool to explore the underlying mechanisms of BPD and to distinguish the effects of existing and new drugs on the diverse facets of the disorder (23–27). Although not described in this chapter, it has been recognized that the amphetamine-induced hyperactivity test, for the hyperactivity and psychostimulant sensitivity domains of mania, should also be integrated into the battery (28). The amphetamineinduced hyperactivity test is one of the most common and validated animal models for mania, and as such it was found to be a useful tool in the exploration of the mechanism of the therapeutic action of lithium and in the search for novel lithium-mimetic compounds (12, 84–90). Importantly, like most of the models described in this chapter (with the exception of the spontaneous activity test), this model is also reactive to mood-stabilizing treatments. Specifically, lithium was repeatedly demonstrated to attenuate stimulant-induced hyperlocomotion in rats (91) and mice (10, 12, 92). Some additional data imply that other mood stabilizers might also be active in this model (84, 88, 93). When tested in this model, Black Swiss mice, which were used in most of the models described in this chapter due to their unique behavioral profile, were found to be reactive to both the stimulating effects of amphetamine and the attenuating effects of lithium (12). In experiments performed in our laboratory, Black Swiss mice were also found to demonstrate high sensitivity to amphetamine administration compared with other strains (94). This property not only makes Black Swiss mice a good strain for the evaluation of treatments that are expected to reduce amphetamine-induced hyperactivity, but also requires cautious dosing to avoid the induction of a behavioral switch from hyperactivity to stereotypical, locally oriented behavior (95). The attempt to integrate a number of tests into one continuous battery raises the need for a number of important considerations. First, combining the different tests into a unified protocol requires that they will be executed sequentially. It is well known that behavioral testing can influence later behavior in consecutive tests. While not ignoring this general rule, it is still possible to minimize these effects by arranging tests in an order from less to more intrusive, as is frequently recommended in studies where the number of animals is limited (96, 97). For the tests described above, it is premised that the best order should be (1) spontaneous

320

Flaisher-Grinberg and Einat

activity test, (2) sweet-solution preference test, (3) resident–intruder test, (4) forced-swim test, and (5) amphetamine-induced hyperactivity test. The spontaneous activity test is advised to be administered first in a battery since it allows the identification of changes in baseline-activity levels which can potentially confound the results collected in other tests. Hence, treatments that increase or decrease activity levels in the spontaneous activity test are suspected to induce generalized effects in the following tests. Moreover, the spontaneous activity and the sweet-solution preference tests are considered to be relatively unintrusive and to inflict only minor long-lasting consequences on behavior. In contrast, the other tests had been demonstrated to have profound effects on rodent’s behavior. Specifically, both the resident–intruder test and the FST had been demonstrated to induce significant stressinduced behavioral and biochemical consequences (53, 98, 99). However, it is important to note that a lot of the stress-related effects described for the resident–intruder test affect the intruder mice, whereas the mice tested in this protocol are the residents. Therefore, it might be appropriate to first execute the resident– intruder test and later the FST. Last, the amphetamine-induced hyperactivity is induced by a pharmacological intervention and amphetamine was documented to have significant and long-lasting effects on behavior (100–102). Hence, it seems best to execute this test last. Beyond the issue of order, some tests require special housing conditions that can affect mice’s behavior in the other tests. For example, the sweet-solution preference and the resident–intruder tests demand that mice be singly housed, a condition which has been demonstrated to affect many behaviors (103–106). The incorporation of all tests into a continuous battery therefore introduces single-housing-induced stress, a matter that should be considered. A different issue involves the subject of drug administration within a test battery. Specifically, if a drug is to be evaluated in the battery, treatment should be continuously administered across all tests. Given that tests are sequentially conducted (one test per day, with the exception of the sweet-solution preference test, which requires at least 3 days), the length of the battery proposed here will be a minimum of 7 days, meaning that in the first tests the length of drug treatment is shorter compared with later tests. Naturally, the incorporation of intervals (rest days) between the tests will prolong this duration. This factor introduces variability in the length of drug treatment in the different tests. Although the effects of acute treatment are not always comparable with the effects of sub-chronic or chronic treatments, this factor might be overcome by starting treatments a few days prior to the behavioral testing. Importantly, most mood stabilizers are expected to have better effects after chronic treatment compared with acute

Mice Models for the Manic Pole of Bipolar Disorder

321

treatment, whereas other drugs may have effects that tolerate across time. Additionally, the chronicity of treatment may also induce a variety of side effects (such as lithium- and valproateinduced electrolytes imbalance due to polydypsia and polyurea (50–52)) (see Critical Parameters and Troubleshooting section for Protocol 2). Whereas the issue of differential chronicity of treatment across tests should be seriously considered, it is still suggested that with precaution and careful interpretation of results, a continuous test battery might offer significant and important information within the field of BPD research. In summary, the protocols presented in this chapter have face validity for a number of the critical domains of the manic pole of BPD. These tests were successfully evaluated for predictive validity with Black Swiss mice and may be utilized at this time to explore the effects of treatments on these domains of the disorder. It is suggested that pending further research, it might become possible to combine these tests into a coherent test battery for mania. Also, given the integration of valid tests for the depression pole, it might culminate to form a coherent battery of models for the critical domains of BPD.

Acknowledgements The studies described in this chapter were supported by a NARSAD Independent Investigator Award to HE. References 1. Gould TD, Einat H. Animal models of bipolar disorder and mood stabilizer efficacy: a critical need for improvement. Neurosci Biobehav Rev 2007;31(6):825–31. 2. Hasler G, Drevets WC, Gould TD, Gottesman, II, Manji HK. Toward constructing an endophenotype strategy for bipolar disorders. Biol Psychiatry 2006;60(2):93–105. Epub 2006 Jan 9. 3. Machado-Vieira R, Kapczinski F, Soares JC. Perspectives for the development of animal models of bipolar disorder. Prog Neuropsychopharmacol Biol Psychiatry 2004;28(2):209–24. 4. Antelman SM, Caggiula AR, Kucinski BJ, et al. The effects of lithium on a potential cycling model of bipolar disorder. Prog Neuropsychopharmacol Biol Psychiatry 1998;22(3):495–510.

5. Antelman SM, Caggiula AR, Kiss S, Edwards DJ, Kocan D, Stiller R. Neurochemical and physiological effects of cocaine oscillate with sequential drug treatment: possibly a major factor in drug variability. Neuropsychopharmacology 1995;12(4):297–306. 6. Caggiula AR, Antelman SM, Kucinski BJ, et al. Oscillatory-sensitization model of repeated drug exposure: cocaine’s effects on shock-induced hypoalgesia. Prog Neuropsychopharmacol Biol Psychiatry 1998;22(3):511–21. 7. Kucinski BJ, Antelman SM, Caggiula AR, Fowler H, Gershon S, Edwards DJ. Cocaine-induced oscillation is conditionable. Pharmacol Biochem Behav 1999;63(3):449–55. 8. Malatynska E, Knapp RJ. Dominantsubmissive behavior as models of mania

322

9.

10.

11.

12.

13.

14.

15.

16.

17.

18.

19.

Flaisher-Grinberg and Einat and depression. Neurosci Biobehav Rev 2005;29(4–5):715–37. Le-Niculescu H, McFarland MJ, Ogden CA, et al. Phenomic, Convergent Functional Genomic, and biomarker studies in a stress-reactive genetic animal model of bipolar disorder and co-morbid alcoholism. Am J Med Genet B Neuropsychiatr Genet 2008;147B(2):134–66. Borison RL, Sabelli HC, Maple PJ, Havdala HS, Diamond BI. Lithium prevention of amphetamine-induced ’manic’ excitement and of reserpine-induced ’depression’ in mice: possible role of 2-phenylethylamine. Psychopharmacology (Berl) 1978;59(3):259–62. Einat H, Shaldubina A, Bersudskey Y, Belmaker RH. Prospects for the Development of Animal Models for the Study of Bipolar Disorder. In: Soares JC, Young A, eds. Bipolar disorders: Basic Mechanisms and Therapeutic Implications. 2nd ed. New York: Taylor & Francis; 2007. Gould TD, O’Donnell KC, Picchini AM, Manji HK. Strain differences in lithium attenuation of d-amphetamine-induced hyperlocomotion: a mouse model for the genetics of clinical response to lithium. Neuropsychopharmacology 2007;32(6):1321–33. Anand A, Verhoeff P, Seneca N, et al. Brain SPECT imaging of amphetamineinduced dopamine release in euthymic bipolar disorder patients. Am J Psychiatry 2000;157(7):1108–14. Murphy DL, Brodie HK, Goodwin FK, Bunney WE, Jr. Regular induction of hypomania by L-dopa in ‘‘bipolar’’ manic-depressive patients. Nature 1971;229(5280):135–6. Huey LY, Janowsky DS, Judd LL, Abrams A, Parker D, Clopton P. Effects of lithium carbonate on methylphenidate-induced mood, behavior, and cognitive processes. Psychopharmacology (Berl) 1981; 73(2):161–4. Van Kammen DP, Murphy DL. Attenuation of the euphoriant and activating effects of d- and l-amphetamine by lithium carbonate treatment. Psychopharmacologia 1975;44(3):215–24. Cormier E. Attention deficit/hyperactivity disorder: a review and update. J Pediatr Nurs 2008;23(5):345–57. Epub 2008 Jun 20. Ebstein RP, Eliashar S, Belmaker RH, Ben-Uriah Y, Yehuda S. Chronic lithium treatment and dopamine-mediated behavior. Biol Psychiatry 1980;15(3):459–67. Geyer MA, Krebs-Thomson K, Braff DL, Swerdlow NR. Pharmacological studies of prepulse inhibition models of

20.

21.

22.

23.

24.

25.

26.

27.

28.

29.

30.

31.

32.

sensorimotor gating deficits in schizophrenia: a decade in review. Psychopharmacology (Berl) 2001;156(2–3):117–54. Peleg-Raibstein D, Knuesel I, Feldon J. Amphetamine sensitization in rats as an animal model of schizophrenia. Behav Brain Res 2008;191(2):190–201. Epub 2008 Apr 8. Nestler EJ, Gould E, Manji H, et al. Preclinical models: status of basic research in depression. Biol Psychiatry 2002;52(6):503–28. Tecott LH, Nestler EJ. Neurobehavioral assessment in the information age. Nat Neurosci 2004;7(5):462–6. Einat H. Modelling facets of mania–new directions related to the notion of endophenotypes. J Psychopharmacol 2006;20(5):714–22. Epub 2006 Jan 9. Einat H. Establishment of a battery of simple models for facets of bipolar disorder: a practical approach to achieve increased validity, better screening and possible insights into endophenotypes of disease. Behav Genet 2007;37(1):244–55. Epub 2006 Jul 22. Einat H. Different behaviors and different strains: potential new ways to model bipolar disorder. Neurosci Biobehav Rev 2007;31(6):850–7. Flaisher-Grinberg S, Overgaard S, Einat H. Attenuation of high sweet solution preference by mood stabilizers: a possible mouse model for the increased reward-seeking domain of mania. Journal of Neuroscience Methods 2009;177:44–50. Flaisher-Grinberg S, Kronfeld-Schor N, Einat H. Models of mania: from facets to domains and from animal models to model animals. J Psychopharmacol 2008;6:6. Flaisher-Grinberg S, Overgaard S, Einat H. Strain-specific battery of tests for maniclike behavior in mice: implications for model development. Biological Psychiatry 2008;63(Supp 7):64S. Gould TD, Gottesman, II. Psychiatric endophenotypes and the development of valid animal models. Genes Brain Behav 2006;5(2):113–9. Cryan JF, Slattery DA. Animal models of mood disorders: Recent developments. Curr Opin Psychiatry 2007;20(1):1–7. McKinney WT. Overview of the past contributions of animal models and their changing place in psychiatry. Semin Clin Neuropsychiatry 2001;6(1):68–78. Cousins DA, Young AH. The armamentarium of treatments for bipolar disorder: a

Mice Models for the Manic Pole of Bipolar Disorder

33.

34. 35.

36.

37.

38.

39.

40.

41.

42.

43.

44.

45.

46.

review of the literature. Int J Neuropsychopharmacol 2007;10(3):411–31. Einat H, Manji HK, Belmaker RH. New approaches to modeling bipolar disorder. Psychopharmacol Bull 2003;37(1): 47–63. Insel TR. From animal models to model animals. Biol Psychiatry 2007;62(12):1337–9. Hiscock K, Linde J, Einat H. Black Swiss mice as a new animal model for mania: a preliminary study. Journal of Medical and Biological Sciences 2007;1(2). Mathew SJ, Manji HK, Charney DS. Novel drugs and therapeutic targets for severe mood disorders. Neuropsychopharmacology 2008;33(9):2080–92. Willner P. The validity of animal models of depression. Psychopharmacology (Berl) 1984;83(1):1–16. Pierce RC, Kalivas PW. Locomotor behavior. Curr Protoc Neurosci 2007;Chapter(8):Unit 8.1. Eilam D, Szechtman H. Biphasic effect of D-2 agonist quinpirole on locomotion and movements. Eur J Pharmacol 1989; 161(2–3):151–7. Shaldubina A, Einat H, Szechtman H, Shimon H, Belmaker RH. Preliminary evaluation of oral anticonvulsant treatment in the quinpirole model of bipolar disorder. J Neural Transm 2002;109(3):433–40. Heffner TG, Downs DA, Meltzer LT, Wiley JN, Williams AE. CI-943, a potential antipsychotic agent. I. Preclinical behavioral effects. J Pharmacol Exp Ther 1989;251(1):105–12. Belmaker RH, Elami A, Bannet J. Intermittent treatment with droperidol, a short-acting neuroleptic, increases behavioral dopamine receptor sensitivity. Psychopharmacology Suppl 1985;2:194–9. Szechtman H, Ornstein K, Teitelbaum P, Golani I. Snout contact fixation, climbing and gnawing during apomorphine stereotypy in rats from two substrains. Eur J Pharmacol 1982;80(4):385–92. Willner P, Towell A, Sampson D, Sophokleous S, Muscat R. Reduction of sucrose preference by chronic unpredictable mild stress, and its restoration by a tricyclic antidepressant. Psychopharmacology (Berl) 1987;93(3):358–64. Papp M, Moryl E, Willner P. Pharmacological validation of the chronic mild stress model of depression. Eur J Pharmacol 1996;296(2):129–36. Willner P, Moreau JL, Nielsen CK, Papp M, Sluzewska A. Decreased hedonic

47.

48.

49. 50.

51.

52.

53.

54.

55.

56.

57.

58.

323

responsiveness following chronic mild stress is not secondary to loss of body weight. Physiol Behav 1996;60(1):129–34. Willner P. Validity, reliability and utility of the chronic mild stress model of depression: a 10-year review and evaluation. Psychopharmacology (Berl) 1997;134(4):319–29. Molander A, Soderpalm B. Glycine receptors regulate dopamine release in the rat nucleus accumbens. Alcohol Clin Exp Res 2005;29(1):17–26. Sadock J, Kaplan H. Synopsis of Psychiatry. 9 ed: Lippincott & Williams; 2002. Brightwell DR, Halmi KA, Finn R. Lithium-induced polydipsia and polyuria: mechanism of action? Biol Psychiatry 1973;7(2):167–71. Guaiana G, Eyrek C, Mariconda G. Polydipsia with normal natremia induced by Valproic Acid. Aust N Z J Psychiatry 2006; 40(9):815–6. Mailman RB. Lithium-induced polydipsia: dependence on nigrostriatal dopamine pathway and relationship to changes in the renin-angiotensin system. Psychopharmacology (Berl) 1983;80(2):143–9. Takeda A, Tamano H, Kan F, Hanajima T, Yamada K, Oku N. Enhancement of social isolation-induced aggressive behavior of young mice by zinc deficiency. Life Sci 2008; 82(17–18):909–14. Epub 2008 Feb 23. Frye CA, Rhodes ME, Walf A, Harney JP. Testosterone enhances aggression of wild-type mice but not those deficient in type I 5alpha-reductase. Brain Res 2002; 948(1–2):165–70. Miczek KA, O’Donnell JM. Intruderevoked aggression in isolated and nonisolated mice: effects of psychomotor stimulants and L-dopa. Psychopharmacology (Berl) 1978;57(1):47–55. Miczek KA, Maxson SC, Fish EW, Faccidomo S. Aggressive behavioral phenotypes in mice. Behav Brain Res 2001; 125(1–2):167–81. Abramov U, Puussaar T, Raud S, Kurrikoff K, Vasar E. Behavioural differences between C57BL/6 and 129S6/SvEv strains are reinforced by environmental enrichment. Neurosci Lett 2008;443(3):223–7. Epub 2008 Aug 5. Vishnivetskaya GB, Skrinskaya JA, Seif I, Popova NK. Effect of MAO A deficiency on different kinds of aggression and social investigation in mice. Aggress Behav 2007;33(1):1–6.

324

Flaisher-Grinberg and Einat

59. Ashkenazy T, Einat H, Kronfeld-Schor N. We are in the dark here: induction of depression- and anxiety-like behaviours in the diurnal fat sand rat, by short daylight or melatonin injections. Int J Neuropsychopharmacol 2008;17:1–11. 60. Erhardt A, Muller MB, Rodel A, et al. Consequences of chronic social stress on behaviour and vasopressin gene expression in the PVN of DBA/2OlaHsd mice–influence of treatment with the CRHR1-antagonist R121919/NBI 30775. J Psychopharmacol 2008;30:30. 61. Haw C, Stubbs J. A survey of the off-label use of mood stabilizers in a large psychiatric hospital. J Psychopharmacol 2005;19(4):402–7. 62. Hellings JA, Weckbaugh M, Nickel EJ, et al. A double-blind, placebo-controlled study of valproate for aggression in youth with pervasive developmental disorders. J Child Adolesc Psychopharmacol 2005;15(4): 682–92. 63. Krsiak M, Sulcova A, Tomasikova Z, Dlohozkova N, Kosar E, Masek K. Drug effects on attack defense and escape in mice. Pharmacol Biochem Behav 1981;14(Suppl 1):47–52. 64. Oehler J, Jahkel M, Schmidt J. The influence of chronic treatment with psychotropic drugs on behavioral changes by social isolation. Pol J Pharmacol Pharm 1985;37(6):841–9. 65. Sheard MH. Aggressive behavior: Modification by amphetamine, p-chlorophenylalanine and lithium in rats. Agressologie 1973;14(5):327–30. 66. Sheard MH. Lithium in the treatment of aggression. J Nerv Ment Dis 1975;160(2–1):108-18. 67. Simler S, Puglisi-Allegra S, Mandel P. Effects of n-di-propylacetate on aggressive behavior and brain GABA level in isolated mice. Pharmacol Biochem Behav 1983;18(5):717–20. 68. Bhatnagar S, Vining C. Facilitation of hypothalamic-pituitary-adrenal responses to novel stress following repeated social stress using the resident/intruder paradigm. Horm Behav 2003;43(1):158–65. 69. Bourin M, Redrobe JP, Hascoet M, Baker GB, Colombel MC. A schematic representation of the psychopharmacological profile of antidepressants. Prog Neuropsychopharmacol Biol Psychiatry 1996;20(8):1389–402. 70. Cleary C, Linde JA, Hiscock KM, et al. Antidepressive-like effects of rapamycin in animal models: Implications for mTOR inhibition as a new target for treatment of

71.

72.

73.

74.

75.

76.

77.

78.

79.

80.

81.

82.

affective disorders. Brain Res Bull 2008;76(5):469–73. Epub 2008 Apr 3. Einat H, Karbovski H, Korik J, Tsalah D, Belmaker RH. Inositol reduces depressivelike behaviors in two different animal models of depression. Psychopharmacology (Berl) 1999;144(2):158–62. Einat H, Clenet F, Shaldubina A, Belmaker RH, Bourin M. The antidepressant activity of inositol in the forced swim test involves 5HT(2) receptors. Behav Brain Res 2001;118(1):77–83. O’Brien WT, Harper AD, Jove F, et al. Glycogen synthase kinase-3beta haploinsufficiency mimics the behavioral and molecular effects of lithium. J Neurosci 2004;24(30):6791–8. Porsolt RD, Deniel M, Jalfre M. Forced swimming in rats: hypothermia, immobility and the effects of imipramine. Eur J Pharmacol 1979;57(4):431–6. Porsolt RD, Bertin A, Blavet N, Deniel M, Jalfre M. Immobility induced by forced swimming in rats: effects of agents which modify central catecholamine and serotonin activity. Eur J Pharmacol 1979;57(2–3):201–10. Porsolt RD, Bertin A, Jalfre M. ‘‘Behavioural despair’’ in rats and mice: strain differences and the effects of imipramine. Eur J Pharmacol 1978;51(3):291–4. Porsolt RD. Animal models of depression: utility for transgenic research. Rev Neurosci 2000;11(1):53–8. Akiskal HS, Kilzieh N, Maser JD, et al. The distinct temperament profiles of bipolar I, bipolar II and unipolar patients. J Affect Disord 2006;92(1):19–33. Epub 2006 Apr 25. Maremmani I, Akiskal HS, Signoretta S, Liguori A, Perugi G, Cloninger R. The relationship of Kraepelian affective temperaments (as measured by TEMPS-I) to the tridimensional personality questionnaire (TPQ). J Affect Disord 2005;85(1–2):17–27. Bersudsky Y, Shaldubina A, Belmaker RH. Lithium’s effect in forced-swim test is blood level dependent but not dependent on weight loss. Behav Pharmacol 2007;18(1):77–80. Cryan JF, Valentino RJ, Lucki I. Assessing substrates underlying the behavioral effects of antidepressants using the modified rat forced swimming test. Neurosci Biobehav Rev 2005;29(4–5):547–69. Shimamura M, Kobayashi T, Kuratani K, Kinoshita M. Optimized analysis of the forced swim test using an automated

Mice Models for the Manic Pole of Bipolar Disorder

83.

84.

85.

86.

87.

88.

89.

90.

91.

92.

93.

94.

experimental system: Detailed time course study in mice. J Pharmacol Toxicol Methods 2007;11:11. Porsolt RD, Bertin A, Jalfre M. Behavioral despair in mice: a primary screening test for antidepressants. Arch Int Pharmacodyn Ther 1977;229(2):327–36. Frey BN, Valvassori SS, Reus GZ, et al. Effects of lithium and valproate on amphetamine-induced oxidative stress generation in an animal model of mania. J Psychiatry Neurosci 2006;31(5):326–32. Engel SR, Creson TK, Hao Y, et al. The extracellular signal-regulated kinase pathway contributes to the control of behavioral excitement. Mol Psychiatry 2008;29:29. Beaulieu JM, Sotnikova TD, Yao WD, et al. Lithium antagonizes dopamine-dependent behaviors mediated by an AKT/glycogen synthase kinase 3 signaling cascade. Proc Natl Acad Sci USA 2004;101(14):5099–104. Epub 2004 Mar 24. Einat H, Yuan P, Szabo ST, Dogra S, Manji HK. Protein Kinase C Inhibition by Tamoxifen Antagonizes Manic-Like Behavior in Rats: Implications for the Development of Novel Therapeutics for Bipolar Disorder. Neuropsychobiology 2007;55(3–4):123–31. Maj J, Chojnacka-Wojcik E, Lewandowska A, Tatarczynska E, Wiczynska B. The central action of carbamazepine as a potential antidepressant drug. Pol J Pharmacol Pharm 1985;37(1):47–56. Renwart N, Frances H, Simon P. The calcium entry blockers: anti-manic drugs? Prog Neuropsychopharmacol Biol Psychiatry 1986;10(6):717–22. Vanover KE. Effects of AMPA receptor antagonists on dopamine-mediated behaviors in mice. Psychopharmacology (Berl) 1998;136(2):123–31. Cox C, Harrison-Read PE, Steinberg H, Tomkiewicz M. Lithium attenuates druginduced hyperactivity in rats. Nature 1971;232(5309):336–8. Berggren U, Tallstedt L, Ahlenius S, Engel J. The effect of lithium on amphetamineinduced locomotor stimulation. Psychopharmacology (Berl) 1978;59(1):41–5. Agmo A, Medrano A, Garrido N, Alonso P. GABAergic drugs inhibit amphetamineinduced distractibility in the rat. Pharmacol Biochem Behav 1997;58(1):119–26. Hiscock KM, Linde JA, Einat H. Black Swiss mice as a new animal model for mania: a preliminary study. Journal of Medical and Biological Sciences 2007;1(2).

325

95. Antoniou K, Kafetzopoulos E, Papadopoulou-Daifoti Z, Hyphantis T, Marselos M. D-amphetamine, cocaine and caffeine: a comparative study of acute effects on locomotor activity and behavioural patterns in rats. Neurosci Biobehav Rev 1998;23(2):189–96. 96. Crawley JN. Behavioral phenotyping of transgenic and knockout mice: experimental design and evaluation of general health, sensory functions, motor abilities, and specific behavioral tests. Brain Res 1999;835(1):18–26. 97. Crawley JN. What’s wrong with my mouse? behavioral phenotyping of transgenic and knockout Mice. 1 ed. New York: WileyLiss; 2000. 98. Connor TJ, Kelly JP, Leonard BE. Forced swim test-induced endocrine and immune changes in the rat: effect of subacute desipramine treatment. Pharmacol Biochem Behav 1998;59(1):171–7. 99. Porsolt RD, Brossard G, Hautbois C, Roux S. Rodent models of depression: forced swimming and tail suspension behavioral despair tests in rats and mice. Curr Protoc Neurosci 2001;Chapter(8):Unit 8.10A. 100. Cabib S. Strain-dependent behavioural sensitization to amphetamine: role of environmental influences. Behav Pharmacol 1993;4(4):367–74. 101. Kalivas PW, Stewart J. Dopamine transmission in the initiation and expression of drugand stress-induced sensitization of motor activity. Brain Res Brain Res Rev 1991;16(3):223–44. 102. Steckler T, Holsboer F. Conditioned activity to amphetamine in transgenic mice expressing an antisense RNA against the glucocorticoid receptor. Behav Neurosci 2001;115(1):207–19. 103. Bowling SL, Bardo MT. Locomotor and rewarding effects of amphetamine in enriched, social, and isolate reared rats. Pharmacol Biochem Behav 1994;48(2):459–64. 104. Hall FS, Huang S, Fong GW, Pert A, Linnoila M. Effects of isolation-rearing on voluntary consumption of ethanol, sucrose and saccharin solutions in Fawn Hooded and Wistar rats. Psychopharmacology (Berl) 1998;139(3):210–6. 105. Lister RG, Hilakivi LA. The effects of novelty, isolation, light and ethanol on the social behavior of mice. Psychopharmacology (Berl) 1988;96(2):181–7. 106. Malkesman O, Maayan R, Weizman A, Weller A. Aggressive behavior and HPA

326

107.

108.

109.

110.

111.

112.

Flaisher-Grinberg and Einat axis hormones after social isolation in adult rats of two different genetic animal models for depression. Behav Brain Res 2006; 175(2):408–14. Epub 2006 Oct 27. Lutter M, Krishnan V, Russo SJ, Jung S, McClung CA, Nestler EJ. Orexin signaling mediates the antidepressant-like effect of calorie restriction. J Neurosci 2008; 28(12):3071–5. Bai F, Li X, Clay M, Lindstrom T, Skolnick P. Intra- and interstrain differences in models of ‘‘behavioral despair’’. Pharmacol Biochem Behav 2001;70(2–3):187–92. Einat H. Chronic oral administration of ginseng extract results in behavioral change but has no effects in mice models of affective and anxiety disorders. Phytother Res 2007;21(1):62–6. Lucki I, Dalvi A, Mayorga AJ. Sensitivity to the effects of pharmacologically selective antidepressants in different strains of mice. Psychopharmacology (Berl) 2001;155(3):315–22. Whirley BK, Einat H. Taurine trials in animal models offer no support for anxiolytic, antidepressant or stimulant effects. Isr J Psychiatry Relat Sci 2008;45(1):11–8. Yoshikawa T, Watanabe A, Ishitsuka Y, Nakaya A, Nakatani N. Identification of

113.

114.

115.

116.

117.

multiple genetic loci linked to the propensity for ‘‘behavioral despair’’ in mice. Genome Res 2002;12(3):357–66. Einat H, Yuan P, Manji HK. Increased anxiety-like behaviors and mitochondrial dysfunction in mice with targeted mutation of the Bcl-2 gene: Further support for the involvement of mitochondrial function in anxiety disorders. Behav Brain Res 2005;165(2):172–80. Kroczka B, Branski P, Palucha A, Pilc A, Nowak G. Antidepressant-like properties of zinc in rodent forced swim test. Brain Res Bull 2001;55(2):297–300. Popik P, Kos T, Sowa-Kucma M, Nowak G. Lack of persistent effects of ketamine in rodent models of depression. Psychopharmacology (Berl) 2008;198(3):421–30. Epub 2008 May 7. Karolewicz B, Paul IA, Antkiewicz-Michaluk L. Effect of NOS inhibitor on forced swim test and neurotransmitters turnover in the mouse brain. Pol J Pharmacol 2001;53(6):587–96. Kochanowska AJ, Rao KV, Childress S, et al. Secondary metabolites from three Florida sponges with antidepressant activity. J Nat Prod 2008;71(2):186–9. Epub 2008 Jan 25.

INDEX A Acclimatization ............................................................... 146 Acetylcholine .................................................................... 61 Acomys cahirinus see Egyptian Spiny Mouse Acoustic startle................................................................ 191 acoustic startle response (ASR) .............................. 204, 205 Actogram .............................................................. 55, 56, 57 b-adrenergic antagonists................................................. 248 Age, of mice..........................10, 11, 12, 13, 14, 16, 25, 39, 70–71, 72, 75, 76, 77, 78, 80, 89, 91, 92, 167, 188, 204, 205, 235, 236, 239, 261, 267, 314 Aggression................................................ 27, 261, 262, 263, 266, 267, 268, 300, 309, 310, 311, 312, 313, 314 ratio ................................................................... 312, 313 A/J mouse strain ..................................10, 17, 32, 129, 131, 185, 192, 203, 300, 307 AKR/J mouse strain........................................................ 131 Ambient sound ................................................................... 6 Amphetamine ..................................10, 16, 17, 51, 92, 209, 215, 287, 289, 298, 300, 301, 303, 319, 320 Amygdala .......................................................... 51, 213, 225 Amyloid ............................................................................ 39 Androgen receptor-insensitive mice ............................... 235 Anhedonia or anhedonic ........................... 27, 40, 122, 129, 132, 153–173, 191, 289 Anosmic ............................................................................ 80 Anticholinergics.................................................. 90, 91, 180 Anticonvulsants or Antiepileptic.................... 110, 248, 299 Antidepressant .................................25, 26, 40, 51, 85–113, 119–133, 154, 155, 156, 180, 186, 190, 192, 193, 208, 213–214, 248, 277–293, 299, 305, 308, 314, 315, 316 Antihistaminics........................................................... 90, 91 Antimanic ....................................................................... 278 Antipsychotics.........................124, 214–215, 298, 299, 303 Anxiety-induced grooming............................................... 24 Anxiogenic .....................................23, 25, 26, 30, 141, 142, 143, 149, 150, 198, 208, 209, 210, 211, 212, 213, 214, 215, 216, 217, 228, 231, 242, 248, 251, 254, 256 Anxiolytic...........................4, 25, 30, 40, 87, 124, 140, 141, 142, 143, 144, 149, 150, 151, 198, 199, 203, 204, 205, 206, 208, 209, 210, 211, 212, 213, 214, 215,

216, 217, 218, 228, 231, 234, 242, 248, 251, 255, 287, 288, 289 Arena...............2, 4, 5, 6, 7, 8, 10, 12, 13, 32, 205, 249, 302 Arginine–vasopressin (AVP) .......................................... 183 Atropine............................................................................ 61 Attention deficit/hyperactivity disorder ......................... 298 Autonomic nervous system......................................... 3, 140 Avoidance ..............................155, 160, 162, 179, 187, 198, 208, 228, 229, 241, 271, 272, 274 Azapirones ...................................................................... 248

B BALB/c mouse strain ....................14, 16, 32, 92, 131, 185, 188, 192, 203, 204, 279, 283, 317 Basal ganglia ............................................................... 22, 31 BDNF-knockout mice.................................................... 182 Bedding....................................6, 15, 37, 38, 40, 41, 42, 43, 77, 250, 252, 302, 309 Behavioral battery ............................13, 27, 28, 50, 60, 241, 242, 299, 318, 319, 320, 321 Behavioral despair................................................... 120, 315 Behavioral history ....................................................... 13–14 Benzodiazepine ............................4, 40, 142, 143, 144, 150, 204, 207, 210, 248 Biological clock ........................................................... 56, 63 BIOSEB ......................................................... 124, 126, 128 Bipolar disorder ............................48, 50, 96, 290, 297–321 Black Swiss mouse strain ............... 300, 301, 303, 304, 306, 307, 308, 312, 313, 314, 316, 317, 318, 319 Blood brain barrier.......................................................... 121 BMAL protein................................................ 48, 49, 50, 51 Body temperature...............................39, 60, 62, 68, 71, 77, 80, 140, 141, 142, 143, 144, 145, 148, 149, 150 Bottles for sucrose test .................................................... 171 Brain derived neurotrophic factor (BDNF) .............. 88, 89, 182, 192 Brain lesions hippocampal.......................................................... 39, 43 medial prefrontal cortex........................................ 39, 41 Breeding..................9, 71, 72, 130, 203, 237, 240, 262, 279 Bupropion ...........................95, 96, 101, 111, 112, 120, 124 Burrowing ................................................................... 37–43 Buspirone ................................................ 142, 210, 211, 248

T.D. Gould (ed.), Mood and Anxiety Related Phenotypes in Mice, Neuromethods 42, DOI 10.1007/978-1-60761-303-9, ª Humana Press, a part of Springer Science+Business Media, LLC 2009

327

MOOD AND ANXIETY RELATED PHENOTYPES IN MICE

328 Index C

C57BL/6 mouse strain ................11, 14, 38, 91, 92, 93, 95, 96, 99, 122, 123, 129, 130, 144, 159, 160, 162, 163, 165, 166, 167, 185, 188, 192, 203, 204, 235, 239, 279 C57BL/10 mouse strain ................................................... 38 C3H/He or C3H/HeJ mouse strain ............... 99, 122, 123, 131, 193, 203, 279 Caffeine..................................................................... 90, 215 Cannabinoids .................................................................. 217 Carbamazepine ............................................................... 110 Casein Kinase I ..................................................... 49, 50, 51 CCK(2) knockout mouse ............................... 204, 215, 216 CD-1 mouse strain ............................92, 93, 96, 97, 98, 99, 100, 101, 103, 104, 105, 106, 107, 108, 109, 112, 113, 122, 129, 167, 168, 203, 249, 262, 266, 267, 268 Center time, see Thigmotaxis Cephalocaudal rule ........................................................... 24 Chlordiazepoxide .................................... 142, 143, 150, 210 Chronic psychosocial stress..................................... 265–268 Chronic stress .................................... 26–27, 129, 153–173, 183–184, 264–269, 271, 272 Circadian.................4, 16, 47–63, 89, 91–92, 145, 206, 238 Citalopram ..........95, 96, 107, 123, 131, 160, 174, 213, 214 CKI, see Casein Kinase I Cleversys Inc. .............................................................. 5, 282 Clock knockout mouse ..................................................... 50 Clock protein .............................................................. 50, 61 Clocklab ............................................................................ 55 Coat state ................................................23, 26, 28, 32, 155 Controllable shock.......................................................... 188 Construct validity .......................87, 94, 161, 228, 289, 299 Contact quieting ...............................68, 69, 71, 72, 73, 74, 78, 79, 80 Controllable versus uncontrollable stressors.................. 186, 189, 191 Corticosterone...................................88, 122, 127, 207, 290 Corticotrophin releasing factor (CRF)/Corticotrophin releasing hormone (CRH)......................... 181, 182, 183, 192, 216 Corticotropin releasing hormone 1 (CRH1) .......... 181, 182 Corticotropin releasing hormone 2 (CRH2) .......... 181, 182 Crawley, J.N....................................199, 200, 206, 207, 238 CRF receptor agonist...................................................... 141 CRF1 receptor antagonists ..................................... 125, 216 Cry, see Cryptochrome gene Cryptochrome gene .................................................... 48, 49 Cytokine............................................................................ 40

D DA, see Dopamine DBA mouse strain ............................................................ 14

DBA/2 or DBA/2 J mouse strain.................. 92, 93, 95, 99, 101, 104, 107, 109, 111, 112, 113, 131, 144, 203, 204, 279, 317 Defecation.........................................3, 4, 78, 228, 247, 273 Defensive freezing .......................................................... 228 Delayed sleep-phase syndrome (DSPS) ..................... 54, 55 Diazepam .................................30, 142, 203, 204, 206, 208, 210, 216, 231, 232, 234, 255, 287, 288, 289 Diestrous (diestrus)................................................. 234, 237 Digging ....................................................................... 37–43 Digital thermometer ....................................................... 146 Dominant/Dominance ............................ 75, 163, 235, 262, 263, 265, 266, 268, 269, 270, 274, 278, 279, 280, 283, 284, 285, 286, 287, 289, 290, 291, 292, 293 Dopamine or dopaminergic ........................... 50, 89, 95, 96, 101, 110, 111–112, 142, 180, 181, 298 Dopamine-beta-hydroxylase-deficient mice .................... 89 Dorsal raphe nucleus....................................................... 181 Drinking................4, 60, 140, 160, 169, 172, 173, 228, 291 DRN, see Dorsal raphe nucleus

E Egyptian spiny mouse....................................................... 39 Electroconvulsive shock or Electroconvulsive seizures ............................................ 86, 88, 120, 124 Electroencephalograms (EEG) .................................. 54, 55 Electromyogram (EMG)............................................ 54, 55 Elevated plus maze (EPM)...................... 92, 141, 191, 204, 215, 218, 225–243, 269, 300 Emergence test................................177, 179, 185, 191, 192 Endophenotype......................................... 24, 185, 299, 318 Enrichment ................................................... 13, 32, 40, 248 Environmental conditions ............................... 71, 125, 129, 304–305, 309, 314, 318 Escape failure .......................................................... 191, 192 Estrogen.................................................................. 234, 235 Estrogen receptor b (ER b) knockout mice ................... 235 Estrus cycle or estrous cycle...................... 11, 130, 235, 237 Ethological analysis........................................................... 24 Euthymia........................................................................... 96 Exploratory activity.....................7, 9, 16–17, 198, 205, 208 Exploratory drive ................................................................ 4

F Face validity .................................40, 87, 91, 154, 191, 227, 289, 290, 299, 321 Familial advanced sleep-phase syndrome (FASPS) ......... 54 Fear ...................4, 24, 68, 69, 162, 191, 198, 212, 213, 228 Fecal boli.......................................................3, 7, 8, 15, 233 Female................................9, 12, 13, 14, 73, 74, 78, 80, 81, 82, 92, 129, 130, 182, 188, 204, 235, 236, 237, 239, 261, 262, 290 Fibroblast growth factors (FGFs)................................... 182

MOOD AND ANXIETY RELATED PHENOTYPES IN MICE

Index 329

Fluoxetine ....................................4, 51, 107, 108, 122, 143, 214, 231, 232, 248, 254, 287, 288, 289 Food deprivation or food restriction.................. 89, 92, 249, 257, 283, 291 Footshock................................178, 180, 181, 186, 187, 189 Forced swim test (FST) cylinder diameter .................................................. 89, 90 depth of water ....................................................... 89, 90 scoring interval...................................................... 90–91 screening .........................88, 90, 91, 112, 119, 132, 301 test/retest............................27, 156, 180, 184, 241, 269, 278, 289, 301, 314–318, 320 water temperature .........................89, 91, 168, 315, 318 Freezing ..........................3, 4, 129, 198, 205, 209, 228, 233 Fur state, see Coat state FVB mouse strain ...............................92, 99, 131, 144, 203

G GABAA receptor agonist........................................ 141, 150 GABAB1 knockout mice ................................................ 204 Gamma-aminobutyric acid (GABA) ......................... 4, 110 Gelatine–mannitol .......................................................... 143 Gender .........................................11, 12, 89, 92, 93, 97, 98, 99, 100, 129, 130, 155, 167, 177, 180, 190, 204, 212, 265, 270 Generalized Anxiety Disorder (GAD)........................... 213 Genetic background.............................. 9–11, 144, 204, 279 Gerbil ........................................................40, 119, 125, 228 Glutamate ............................................................... 110, 111 Grooming ......................................... 3, 21–34, 72, 238, 273 Grooming Analysis Algorithm (GAA)............................ 28 Group housing................................................................ 145

H Habituation...................................... 23, 145, 166, 168, 169, 170, 204, 207, 218, 238–239, 271, 284, 285, 287, 302, 303, 304, 309, 315 Hall, C. ........................................................................... 247 Hamster ............................................................................ 40 Handling ......................................11, 12–13, 25, 32, 77, 79, 80, 125, 143, 149, 205, 238–239, 248 Hedonia ..........................154, 155, 160, 161, 166, 298, 300 Helplessness ............................129, 132, 161, 177–193, 289 Hippocampus..................................38, 39, 41, 51, 182, 225 Hole board, modified, see Modified hole board Home cage ...............................4, 12, 13, 38, 39, 43, 69, 73, 77, 79, 80, 81, 127, 157, 159, 162, 168, 232, 238, 239, 249, 250, 251, 252, 254, 255, 262, 265, 268, 269, 271, 284, 302, 303, 306, 307, 310, 311, 312, 316 Home cage activity ......................................... 129, 131, 158 Homeostasis.................................................................... 144 Housing conditions............13, 93, 127, 146, 165, 205, 239, 262, 264, 305, 311, 315, 320

HPA axis, see Hypothalamic–pituitary–adrenal axis 5-HT, see Serotonin 5-HT receptor agonist.................................................... 141 5-HT1 receptor ....................................................... 210–212 5-HT1B-knockout mice................................................... 88 5-HT2 receptor ....................................... 211, 212–213, 214 5-HT3 receptor ....................................... 212, 213, 215, 218 Husbandry................................................................... 25, 32 Hyperactivity/Hyperactive....................... 39, 160, 242, 298, 300, 301, 304, 319, 320 Hyperlocomotion...................................... 17, 160, 169, 319 Hypoactivity/Hypoactive ........................27, 120, 153, 162, 163, 166, 304 Hyponeophagia.......247, 248, 249, 251, 254, 255, 256, 257 Hypophagia............................................................. 247–257 Hypothalamic–pituitary–adrenal (HPA)-axis .............. 141, 181, 183, 216 Hypothalamic pre-optic area .......................................... 144 Hypothalamus............................................. 22, 48, 181, 191 Hypothermia................................................. 70, 75, 91, 149

I Imipramine ....................................92, 93, 95, 96, 104, 105, 110, 121, 123, 126, 127, 128, 131, 214, 279, 287, 288, 289, 290, 305, 308, 316, 317 Immobility .............................27, 86, 87, 88, 90, 91, 92, 93, 94, 95, 110, 111, 120, 121, 122, 123, 124, 125, 126, 128, 130, 131, 132, 181, 228, 263, 272, 279, 291, 314, 315, 316, 317 Inbred mice ............................................... 93, 203, 249, 267 Infant mouse ............................................................... 67–82 Infrared ...........................3, 60, 61, 199, 202, 229, 302, 305 sensors ........................................................................... 5 Interference effect .................179, 180, 181, 182, 183, 185, 186, 188, 189, 191, 192, 193 Interleukin-1-beta............................................................. 39 Isolation .......................................13, 32, 62, 69, 70, 71–72, 74, 77, 78, 79, 80, 89, 92, 188, 237, 262 I.T.E.M-LABO ............................................................. 124

J Jackson Laboratory ......................................................... 237 Jet-lag.................................................................... 48, 57, 58

K KATP channel .................................................................. 39 K+ Channel blockers ......................................................... 88 K+ Channel openers.......................................................... 88 Kir6.2, see KATP Channel Knockout mice ...................................30, 31, 50, 72, 88, 89, 129, 132, 182, 203, 204, 235, 236, 239, 240, 279 Kudryavtseva, N.............................................................. 264

MOOD AND ANXIETY RELATED PHENOTYPES IN MICE

330 Index L

Lamotrigine ............................................................ 110, 111 L-DOPA ........................................................................ 180 Learned helplessness...................................... 129, 132, 161, 177–193, 278, 279, 289 Lesions, see Brain lesions Light/dark cycle or Light cycle...................... 16, 48, 57, 58, 60, 63, 127, 168, 171, 200, 238, 283, 302, 315 Light/dark test (black/white box or light/dark box) ........................... 198, 199, 203, 204, 205, 209, 210, 211, 212, 213, 214, 216, 217, 218 Lighting conditions .............................................. 9, 17, 252 Line crossing................................................... 206, 208, 273 Lipopolysaccharide (LPS) .......................................... 39, 40 Lithium ....................................51, 110, 298, 299, 300, 301, 303, 304, 305, 307, 308, 313, 316, 317, 319, 321 Littermates......................68, 69, 71, 73, 74, 77, 78, 80, 235 Locomotion..................................................5, 6, 13, 14, 15, 76, 159, 160, 169, 198, 199, 202, 205, 208, 301, 303, 305, 319 Lux ..................6, 15, 59, 157, 202, 206, 230, 242, 252, 270

M Mania .........................................51, 96, 278, 283, 290, 293, 298, 299, 300, 307, 315, 318, 319, 321 Manic ................................................................ 96, 297–321 MAO-A, see Monoamine oxidase-A MAO-I, see Monoamine oxidase, inhibitors Marble burying ........................................................... 37–43 Maternal potentiation................... 68, 71, 72–73, 74, 79–81 Maternal separation .......................................................... 68 Mating ........................................................................ 11, 32 paced mating............................................................. 241 Med Associates Inc......................................... 124, 126, 128 Medial prefrontal cortex ............................................. 39, 41 Melanopsin ...............................................52, 53, 55, 61, 62 Melatonin.................................................................... 54, 62 Methamphetamine ........................................................... 51 Methylcellulose ............................................... 143, 145, 146 Milk/Sweetened condensed milk .................. 250, 251, 252, 253, 256, 278, 280, 281, 291 Minimitter .................................................................. 60, 62 Modified hole board ............................................... 269–274 Monoamine................................4, 50, 88, 89, 96, 106, 106, 112, 120, 124, 181, 182, 183 Monoamine oxidase-A ............................................. 50, 112 oxidase inhibitors .............................................. 120, 124 Mood stabilizer .................................. 51, 96–111, 112, 298, 299, 300, 301, 305, 307, 308, 310, 313, 314, 315, 316, 317, 319, 320 Motor activity ......................................4, 12, 13, 15, 72, 87, 90, 96, 129, 131, 132, 133, 141, 144, 170, 205, 218, 233, 249, 255, 256, 265, 273, 274, 301, 303

Motivation ...........................................38, 43, 72, 191, 198, 228, 261, 262, 269, 272, 283 129 mouse strain ............................................................. 123 Mutation .................................................... 24, 72, 140, 203, 204, 235, 241, 298

N NA, see Norepinephrine/Noradrenaline Na+ channel activator...................................................... 110 Naltrexone......................................................................... 72 Neophobia............................................... 172, 198, 206, 309 Neurokinin...................................................................... 125 Neuroleptic ............................................... 87, 154, 287, 289 Neuropeptide .................................................. 215–217, 249 Neuropeptide Y1 receptor KO mice............................... 249 Neurosteroids............................................................ 88, 235 NIH-Swiss mouse strain .................................. 93, 100, 104 Nitric oxide synthase (NOS) ............................................ 88 NK1 receptor antagonists ....................................... 125, 216 NK2-receptor antagonists................................................. 88 NMRI mouse strain........................................................ 145 Nocturnal ....................................................... 16, 54, 55, 63, 199, 206, 263 Noirot, E........................................................................... 67 Noise, ambient .........................6, 11, 15, 92, 127, 129, 148, 171, 172, 228, 242, 254, 302, 303, 304, 311, 315, 316, 318 Noldus........................................................................... 5, 75 Norepinephrine/Noradrenaline/Noradrenalin/ Noradrenergic ..............................91, 93, 95, 96, 99, 102, 110, 120, 124, 181, 279 Norepinephrine reuptake inhibitors ................. 96, 120, 124 NOS, see Nitric oxide synthase Nose poke ....................................................................... 191 Novel cage...................................... 144, 239, 249, 250, 251, 252, 254, 255, 256 Novelty.........................................3, 4, 7, 22, 26, 27, 30, 31, 32, 33, 140, 156, 184, 208 -induced hypophagia ........................................ 247–257 Novelty suppressed feeding............................................. 249 Npas2 ................................................................................ 51 NZB mouse strain .......................................................... 204

O Obsessive compulsive disorder (OCD) ...................... 25, 40 OCD, see Obsessive compulsive disorder Olfactobulbectomy (olfactory bulbectomy) ...... 26, 129, 132 Olfactory .......................................................... 6, 25, 68, 80, 128, 129, 132 Olfactory tactile senses...................................................... 78 Open field test ceiling height ........................................................ 11, 14 center time, see Thigmotaxis color .................................................................... 15, 242

MOOD AND ANXIETY RELATED PHENOTYPES IN MICE

Index 331

shape ..................................................................... 11, 14 size......................................................................... 11, 14 texture ................................................................... 11, 15 walls....................................................................... 11, 14 O-maze test .................................................................... 156 Operant .......................................39, 40, 179, 186, 189, 191 Oscillation................................................................... 48, 61 OT knockout, see Oxytocin knockout Outbred mice .................................................................. 249 Ovariectomized............................................... 229, 236, 239 mice........................................................... 234, 235, 240 Oxytocin knockout ........................................................... 72

P Pain sensitivity .......................................................... 32, 193 Panic disorder ................................................................. 213 PanLab............................................................................ 282 Parental separation............................................................ 67 Paroxetine ...............88, 92, 95, 96, 109, 110, 111, 112, 214 Per, see Period gene Per2 knockout................................................................... 50 Per2 luciferase mouse........................................................ 61 Period gene ....................................................................... 48 Phenytoin................................................................ 110, 111 Photocell activity monitoring ......................................... 305 Photoentrainment...............................48, 53, 55, 57, 58, 60 Picrotoxin........................................................................ 248 Placebo ............................................229, 231, 232, 234, 240 Plastic tubes for restraint stress....................................... 170 Polydipsia........................................................................ 309 Polymorphism............................................... 50, 51, 54, 123 Polyurea .......................................................................... 321 Porsolt forced swim test, see Forced swim test Porsolt, R. .................................................85, 86, 88, 90, 91 Postnatal......................................68, 69, 70, 71, 78, 81, 248 Postpartum........................................................................ 80 Posttraumatic stress disorder (PTSD).................... 177, 192 Predictive validity.......................................68, 87, 112, 120, 121, 124, 126, 133, 142, 228, 248, 278, 287, 289, 299, 321 Pre-weaning...................................................................... 78 Prion infection .................................................................. 39 Proestrous or Proestrus ........................................... 234, 237 Progestin ................................................................. 235, 237 Propranolol ..................................................................... 192 Prostaglandins................................................................. 144 Psychostimulant......................................... 25, 87, 121, 208, 209, 215, 287, 289, 298, 300, 319 PTSD, see Posttraumatic stress disorder Pupillary light reflex.................................................... 61–62

Q Quantitative trait loci (QTL) ......................... 122, 123, 133

R Rapid eye movement (REM) ........................................... 55 Rat.......................................................15, 71, 78, 167, 228, 288, 289 Rat exposure....................159, 162, 163, 166, 167, 170, 173 Rearing......................................................1, 3, 5, 8, 14, 15, 71, 72, 78, 199, 201, 206, 208, 209, 217, 233, 263, 273 Red light illumination ...................................................... 15 5-reductase-knockout mice .......................... 235, 236, 238 Reduction of submissive behavior model (RDBM) ..................................................... 277–293 Resident intruder test ............................ 167, 168, 261–264, 278, 291, 300, 301, 310–314, 320 Restraint stress ................157, 159, 163, 166, 168, 170, 239 Retina............................................................ 51, 52, 53, 193 Retinoic acid-related element........................................... 50 Rev-Erb .............................................................. 49, 50, 51 Reverse light cycle........................................................... 238 Reward .....................................38, 40, 41, 50, 69, 154, 159, 160, 184, 191, 197, 270, 271, 274, 298, 300, 305, 307, 309 Reward seeking behavior ........................................ 298, 309 RORE, see Reitnoic acid-related element ROR ......................................................................... 49, 50 Rotarod ................................................................... 129, 132

S SABRA mouse strain .............279, 283, 285, 288, 289, 290 Saccharin.................................154, 305, 306, 307, 308, 309 San Diego Instruments...................................5, 6, 238, 282 Schizophrenia ................................................................. 298 SCN, see Suprachiasmatic Nucleus Scopolamine.................................................................... 180 Scrapie......................................................................... 38, 39 Seasonal affective disorder (SAD).............................. 47, 54 Sedative ...........................................1, 2, 132, 141, 209, 288 Selective serotonin reuptake inhibitor (SSRI) ...................96, 107, 111, 120, 124, 142, 248 SencarA/PtJ mouse strain............................................... 131 Sensory contact model .................................... 264–265, 268 Separation stress.................................................................. 4 Separation test ...................................................... 71, 74, 80 Serotonin (5-HT) .....................89, 93, 110, 141, 181, 183, 192, 211, 212 Serotonin reuptake inhibitors ............91, 96, 107, 111, 120, 124, 142, 148, 287 Sexual sexual behavior ............................................ 22, 241, 269 sexual differences (gender differences) ................. 12, 92 Shock......................................13, 40, 86, 88, 120, 124, 154, 162, 178, 179, 180, 181, 182, 183, 184, 185, 186, 186, 187, 188–190, 191, 192, 193, 198, 206, 228

MOOD AND ANXIETY RELATED PHENOTYPES IN MICE

332 Index

Shock-escape test.................................................... 178, 185 Shuttle box ...................................... 180, 186, 187, 188, 189 Shuttle-escape test .......................... 179, 180, 184, 185, 190 Single-housed ................................................................... 13 SJL mouse strain............................................................. 204 Smell ...............................................................6, 8, 206, 264 Sniffing ...............................3, 208, 252, 268, 272, 273, 312 Social avoidance ...................................................... 155, 160 Social defeat ..................................144, 145, 159, 160, 162, 168, 261–274 Social disruption ............................................................. 268 Social interaction test.............................................. 167, 191 Software ...............................................5, 7, 15, 55, 56, 124, 146, 147, 170, 281, 282, 291, 302 Spielberger state-trait anxiety inventory (STAI) test..... 226 Spontaneous activity .................................................. 4, 300, 301–305, 316, 319, 320 Stereotypy (or stereotypical or stereotypic) ............ 298, 303 Steroids ................................................... 226, 237, 238, 239 Stoelting Inc. .................................................................. 230 Strain differences .............................32, 34, 37, 38, 85, 122, 126, 131, 144, 185, 188 Stress ....................................4, 7, 12, 13, 16, 21–34, 77, 88, 112, 119, 121, 127, 129, 131, 132, 133, 139–151, 153–174, 177, 178, 198, 205, 206, 212, 216, 237, 238, 239, 240, 261–274, 278, 279, 289 Stress induced hypothermia.................................... 139–151 Stressor, type of........................................ 26, 120, 139, 140, 141, 143, 144, 145, 146, 149, 150, 161, 162, 173, 177, 178, 180, 181, 182, 183, 184, 185, 186, 188, 189, 192, 193, 206, 232, 239 Submissive behavior................................................ 277–293 Submissive behavior model (RSBM) ..................... 277–293 Substance P receptor KO mice....................................... 249 Sucrose .....................................27, 154, 155, 156, 157, 158, 159, 160, 162, 165, 166, 167, 169, 171, 172, 173, 191, 280 Sucrose preference test/Sweet-solution consumption test.......................160, 167, 300, 301, 305–310, 320 Sucrose test .............156, 159, 162, 167, 169, 171, 172, 173 Suprachiasmatic nucleus (SCN) ....................................... 48 129 Sv/Ev, 129S6 or 129S6/SvEvTac mouse strain....................................123, 131, 140, 144, 203 Swimming.......................................... 22, 85–113, 120, 129, 154, 184, 206, 316 Swiss mouse strain .................................................. 129, 203 Swiss-Webster mouse strain...............92, 99, 100, 144, 317

T Taconic Farms ................................................ 237, 239, 240 TCA, see Tricyclic antidepressants or tricyclics Tail suspension test (TST) ............................ 93, 94, 95, 96, 119–133, 156, 206, 278, 289, 291 Telemetry/Telemetric ....................................... 62, 140, 141

Temperature ambient temperature...........................69, 70, 76, 77, 81 basal body temperature ............................ 140, 143, 144, 145, 149, 150 body temperature ....................................39, 60, 62, 68, 71, 77, 80, 140, 141, 142, 143, 144, 145, 148, 149, 150 core body temperature ................................ 62, 142, 144 rectal temperature ............................ 140, 141, 142, 143, 147, 148, 150 water temperature .........................89, 91, 168, 315, 318 Tetracyclic............................................................... 248, 288 Tg2576 model................................................................... 39 Thiotixen................................................................. 287, 289 Thermoregulation......................................... 21, 70, 72, 139 Thigmotaxis ......................................1, 4, 6, 10, 11, 16, 274 Time bins ........................................................3, 8, 303, 304 Time of testing ..................................................... 12, 16, 17 Tinbergen, Niko ............................................................... 38 Topiramate.............................................................. 110–111 Tranquillizers .................................................................... 91 Transgenic mice.................................................. 25, 31, 266 Tranylcypromine............................................. 106, 111, 112 Triadic paradigm..................................................... 177, 186 Tricyclic antidepressants or tricyclics..................... 103, 110, 120, 123, 124, 248, 287, 288 Trigeminal tactile senses................................................... 78 Type-II glucocorticoid receptor antisense-expressing mice ..................................................................... 249

U Ultrasonic vocalization................................................ 67–82 Uncontrollable stress/Uncontrollable shock ............ 88, 161, 177, 178, 179, 180, 182, 183, 184–185, 186, 188, 189, 191, 212 Unpredictable chronic mild stress........................... 278, 279 Urination....................................................................... 3, 78

V Valproate/Sodium valproate ................................... 51, 110, 300, 301, 303, 305, 307, 308, 309, 313, 316, 317, 318, 321 Ventral tegmental area.............................................. 88, 191 Veratrine ................................................................. 110, 111 Vesicular monoamine transporter 2................................ 123 Video tracking system................................ 6, 130, 230, 231, 233, 281, 282–283 VMAT2, see Vesicular monoamine transporter 2 Vocalization ............................................ 3, 67–82, 263, 266 Vogel(s)................................................................... 212, 228 Vogel punished drinking ................................................ 228 VTA, see Ventral tegmental area

MOOD AND ANXIETY RELATED PHENOTYPES IN MICE

Index 333

W Water deprivation or water restriction .................. 154, 162, 169, 217, 228, 269 Water emergency stress .......................................... 166, 168 Weaning.......................................................... 11, 69, 73, 78 Wheel running...................................................... 54, 55–60 Wheel water tank.............................................................. 91 White noise.................................................. 6, 15, 302, 304, 311, 315, 316 Wildtype ...........................................31, 235, 236, 239, 240

Willner, P. ...................................................................... 154 Wistar rats............................................................... 167, 288

Y Yohimbine .............................................................. 287, 289 Yoked paradigm...................................................... 178, 189

Z Zeitgeber..................................................................... 16, 56