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

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