Citation
Rats Exhibit Behavioral Despair and Hormonal Alterations after Social Defeat Stress: Implications for Major Depression

Material Information

Title:
Rats Exhibit Behavioral Despair and Hormonal Alterations after Social Defeat Stress: Implications for Major Depression
Creator:
STONE, KRISTEN L. ( Author, Primary )
Copyright Date:
2008

Subjects

Subjects / Keywords:
Animal social behavior ( jstor )
Despair ( jstor )
Freezing ( jstor )
Hormones ( jstor )
Locomotion ( jstor )
Psychological stress ( jstor )
Rats ( jstor )
Social interaction ( jstor )
Stress tests ( jstor )
Traumatic stress disorders ( jstor )

Record Information

Source Institution:
University of Florida
Holding Location:
University of Florida
Rights Management:
Copyright Kristen L. Stone. Permission granted to University of Florida to digitize and display this item for non-profit research and educational purposes. Any reuse of this item in excess of fair use or other copyright exemptions requires permission of the copyright holder.
Embargo Date:
3/1/2007
Resource Identifier:
649814571 ( OCLC )

Downloads

This item has the following downloads:


Full Text











RATS EXHIBIT BEHAVIORAL DESPAIR AND HORMONAL ALTERATIONS
AFTER SOCIAL DEFEAT STRESS: IMPLICATIONS FOR MAJOR DEPRESSION
















By

KRISTEN L. STONE


A THESIS PRESENTED TO THE GRADUATE SCHOOL
OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT
OF THE REQUIREMENTS FOR THE DEGREE OF
MASTER OF SCIENCE

UNIVERSITY OF FLORIDA


2006

































Copyright 2006

by

Kristen L. Stone





























To my loving husband (Justin),
my wonderful parents (Linda, Kenneth, and Donna),
and my brothers and sisters (Kim, Rick, Minet, Roger, and Jay).
Thank you for your uninterrupted love and support.

Also to my extended family (both biological and in-law) and my friends.
Thank you for your continued encouragement and guidance.
















ACKNOWLEDGMENTS

I would like to thank my supervisory committee members (Dr. Mohamed Kabbaj

from Florida State University; Dr. Neil Rowland; and Dr. Linda Hermer-Vazquez) for

their time and patience. I would like to extend special gratitude to my advisor (Dr.

Darragh Devine) for his persistent guidance; and to Andrea Naccarato for her diligent

work with the Porsolt swim tests. I would also like to thank my mentors in Gainesville

(especially Clare Mathes) for their support during this long j ourney of higher education.

Lastly, and most importantly, I would like to thank God for the breadth of knowledge He

has bestowed upon me.




















TABLE OF CONTENTS


page

ACKNOWLEDGMENT S .............. .................... iv


LI ST OF T ABLE S ............ ...... .___ ............... vii...


LIST OF FIGURES ........._.._.._ ....__. ..............viii...


AB STRAC T ................ .............. ix


CHAPTER


1 INTRODUCTION ................. ...............1.......... ......


2 IVETHODS .............. ...............7.....


Animal s............... ...............7.

Surgical Procedures .............. ...............8.....
Experimental Procedures .................. ........... ...............8.......
General Social Defeat Stress Procedure ................. ............. ......... .......8
Social Defeat Stress Regimen: Experiment la............... ...............9...
Social Defeat Stress Regimen: Experiment lb............... .. ...............10
Social Defeat Stress and Porsolt Swim Test: Experiment 2 ............... ... ............11
Behavioral Assays .............. ...............12....
Histological As says .............. ...............12....
Statistical Analyses ................. ...............13.................

3 RE SULT S ................. ...............17.......... .....


Experiment la: Six Daily Sessions of Social Defeat Stress .............. ....................17
Experiment lb: Twelve Intermittent Sessions of Social Defeat Stress .....................23
Experiment 2: Social Defeat Stress Followed by Porsolt Swim Testing...................27
Inter-Ob server Reliability ................. ...............27................

4 DI SCUS SSION ................. ...............29................


Daily Social Defeat Stress .............. ...............29....
Intermittent Social Defeat Stress .............. ...............30....
Relevance to Previous Work .............. ...............3 1....













Porsolt Swim Test............... .... ...............33
Conclusions and Future Directions............... ...............3


LIST OF REFERENCES ............. ...... .__ ...............35..


BIOGRAPHICAL SKETCH .............. ...............38....


















LIST OF TABLES

Table pg

2-1 Experiment la, group assignments............... ..............1

2-2 Experiment lb, group assignments .............. ...............16....

2-3 Experiment 2, group assignments .............. ...............16....


















LIST OF FIGURES


Figure pg

3-1 Social defeats per daily session.. ............. ...............19.....

3-2 Intruder behavior during daily social defeat sessions. ............. .....................2

3-3 Circulating hormones after daily social defeat stress ................. ........__. ........21

3-4 Effects of daily social defeat stress on glandular masses. .............. ...................22

3-5 Social defeats per intermittent session. ............. ...............24.....

3-6 Intruder behaviors during intermittent social defeat sessions. ............. .................25

3-7 Circulating hormones after intermittent social defeat stress. ............. ..................26

3-8 Effects of intermittent social defeat stress on glandular masses. ............. ...... ..........26

3-9 Immobility during the Porsolt swim test. .............. ...............28....

3-10 Immobility measured in 5-minute bins during the Porsolt swim test. .....................28
















Abstract of Thesis Presented to the Graduate School
of the University of Florida in Partial Fulfillment of the
Requirements for the Degree of Master of Science

RATS EXHIBIT BEHAVIORAL DESPAIR AND HORMONAL ALTERATIONS
AFTER SOCIAL DEFEAT STRESS: IMPLICATIONS FOR MAJOR DEPRESSION

By

Kristen L. Stone

December 2006

Chair: Darragh P. Devine
Major Department: Psychology

Chronic emotional stress plays a pivotal role in the origin of many psychiatric

disorders, including major depression. Accordingly, we investigated the behavioral,

hormonal, and glandular effects of both repeated and acute emotional stress using the

social defeat model in rats. We compared one week of daily social defeat exposure to

one month of intermittent exposure to examine the effectiveness of massed versus

intermittent stressors.

In two experiments, naive male intruder rats were each exposed to dominant male

resident rats. Each resident and intruder pair was allowed to interact for 5 minutes or

until the intruder exhibited submissive supine posture three times. The intruder was then

briefly removed, placed into a protective wire mesh cage, and returned to the resident' s

cage until a total of 10 minutes elapsed from initial entry. Additional control rats were

not exposed to social defeat stress. For the first experiment, the intruder rats were killed

10 minutes, 30 minutes, or 24 hours after the last social defeat session and plasma









corticosterone and adrenocorticotropic hormone concentrations were assayed. For the

second experiment, the previously-stressed rats were tested with a 15-minute Porsolt

forced swim test.

The intruder rats exhibited more freezing behavior and less exploratory locomotion

across consecutive social defeat sessions, resembling the loss of interest found in

depressed patients. Exposure to social defeat stress produced significant elevations in

circulating hormones 10 minutes and 30 minutes after the session, when compared with

concentrations in the control rats. The repeatedly stressed rats also exhibited higher basal

concentrations of circulating corticosterone 24 hours later, mimicking the augmented

circulating hormones found in clinically depressed patients. These results were evident

after six daily exposures; however, basal hormone concentrations were not significantly

elevated with the extended regimen of one social defeat session every 72 hours.

Inconsistent thymus involution in the chronically stressed rats in both regimens suggests

that a longer, more intense daily stress regimen may be necessary to alter glandular

masses. Exposure to chronic social defeat stress also produced a significant increase in

total immobility time during the forced swim test when compared with the immobility

times for the rats that were exposed to a single acute social defeat session and with the

immobility times for the rats that were not exposed to social defeat stress, thus

representing behavioral despair in the chronically stressed animals. Overall, the

behavioral, hormonal, and glandular alterations that occurred after repeated social defeat

stress resemble some of the symptoms of maj or depression in humans.















CHAPTER 1
INTTRODUCTION

Annually, approximately 6.6% of the national adult population suffers from

depression (Kessler et al., 2003). It is a wide-spread illness that interferes with the ability

to eat, sleep, work, and enj oy formerly pleasurable activities. The economic impact of

this devastating disorder is high, but the cost in human distress cannot be estimated.

The 4th Edition of the Diagnostic and Statistical Manual of Mental Disorders

(1994) defines depression according to the following criteria, with at least Hyve of the

symptoms present on a daily basis for at least 2 weeks: depressed mood, loss of interest

or pleasure, significant weight loss or gain, insomnia or hypersomnia, psychomotor

agitation or retardation, fatigue, undue guilt and/or feelings of worthlessness,

indecisiveness, recurrent thoughts of death, and significant distress or impairment in

social or occupational functioning. These symptoms must represent a change from the

individual' normal level of interpersonal function.

Maj or depression is also clinically characterized by altered hormonal function

stemming from ongoing elevations in overall organism distress. It has been well

established that chronic emotional stress plays a pivotal role in the genesis of many

psychiatric disorders such as depression (for review, see Agid et al., 2000). Chronic

stress weighs on the physiological systems that maintain homeostasis and produces

changes in the operating limits of those hormonal systems. Allostatic load, or the strain

from the elevated activity of systems under maj or stress, can predispose an animal to

many psychiatric disorders, including depression (McEwen and Stellar, 1993). The









concept of allostatic load suggests that there is a fixed state in which enduring

environmental challenges are balanced by a hormonal response that is raised above

normal, basal levels. Patients with major depression, regardless of age, show higher

24-hour average cortisol levels when compared to normal subj ects (Linkowski et al.,

1985). Additionally, depressed patients reach the nadir of their daily cortisol cycle two to

three hours before control subj ects (Pfohl et al., 1985). This imbalance between

activation and recovery of the stress response is implicated in the inability to maintain

homeostasis, thus leading to neuroendocrine maladjustment and heightened risk for

depression (De Kloet, 2003).

Stressful stimuli are categorized into two descriptive classes. Systemic stressors,

such as exposure to heat or cold, present immediate threats to somatic homeostasis while

processive stressors emphasize higher level cognitive processing (Herman and Cullinan,

1997). Common processive stressors include instability in the social hierarchy and loss

of environmental control. In accordance with the emotional nature of processive stress, it

is particularly implicated in a variety of psychiatric disorders including depression (for

review, see Anisman and Matheson, 2005).

Inputs from the brainstem (if the stressor is systemic in nature) and cortical and

limbic structures (if the stressor is processive in nature) converge at the paraventricular

nucleus (PVN) of the hypothalamus where parvocellular neurons proj ect to the median

eminence. From there, corticotrophin-releasing hormone (CRH) and arginine

vasopressin (AVP) are released into the hypophyseal portal circulation, stimulating

adrenocorticotropic hormone (ACTH) release from the anterior pituitary corticotrope

cells into the general circulation. The release of ACTH then stimulates the synthesis and









release of glucocorticoids from the adrenal cortex. Cortisol is the main glucocorticoid in

humans and corticosterone (CORT) is the main glucocorticoid in rats. The elevated

circulating levels of glucocorticoids decrease the further activity of the HPA axis through

exertion of negative feedback on neurosecretory cells of the hypothalamus and

corticotrope cells of the pituitary (for review, see Whitnall, 1993).

Altered regulation of hormonal activity in depressed patients is thought to be a

result of increased activity of specific CRH-containing neurons in the PVN. The average

total number of such neurons is up to four times higher in depressed patients when

compared to normal control subjects. Also, co-localization of AVP in CRH-expressing

neurons has been indicated as an index for stress-activated neuronal activity. The

average number of neurons co-expressing both CRH and AVP in depressed patients is up

to three times higher than those for normal subj ects. These results suggest that increased

expression of CRH- and AVP-containing neurons in the PVN may cause at least a

fraction of the collective symptomatology of depression (Raadsheer et al., 1994).

When physically or emotionally stressed, non-human animals endure physiological

responses that lead to behavioral and hormonal impairment which may be fundamentally

similar to the impairment seen in human stress-induced psychopathology. Behavioral

responses to stressful stimuli (including increased drug-taking propensity, decreased

performance in learning tasks, sleep disturbances, and unsocial behavior) have been

observed in a variety of species (for review, see Amiel-Tison et al., 2004). Also, there is

significant evolutionary homology in stress-regulating peptides, such as CRH and CRH-

related molecules (Chang and Hsu, 2004). Thus, the use of an animal model for










processive stress that produces both behavioral and hormonal effects is a logical approach

to the study of human stress-induced disorders.

The effects of chronic and acute processive stress have been studied in rats

(Simpkiss and Devine, 2003). Experimentally naive rats were exposed to a chronic

variable stress (CVS) regimen of twice daily stressors for fifteen days. The stressors

included novel environment, switched cage mates, forced swim, light open Hield,

intermittent white noise, and intermittent footshock, administered on a random

intermittent schedule. The CVS regimen was unsuccessful in producing elevations in

basal circulating concentrations of ACTH and CORT, or in hormonal response to an

acute stressor. The rats showed a blunted ACTH response, but no altered CORT

response.

Another model for emotional stress, social defeat, has also produced significant

elevations in circulating CORT during and after acute and repeated exposure to the

stressful stimulus (Sgoifo et al., 1996). The procedure, developed by Miczek (1979), is

designed to model social stress. A male "intruder" rat is exposed to social stress when it

is placed into the home cage of a larger male "resident" rat. The resident rat exhibits

dominant behavior toward the intruder rat by displaying assertive posture, standing over

the intruder. The intruder submits by displaying supine posture, positioned beneath the

resident.

The further effects of repeated social defeat stress on behavioral and hormonal

(HPA axis) responses have been studied in rats (Lopes and Devine, 2004). In a

preliminary study, a repeatedly stressed group of intruder rats showed significantly

elevated circulating CORT concentrations 24 hours after their Einal social defeat session,









when compared to the concentrations in the unstressed control animals. This suggests an

enduring change in the circadian regulation of the HPA axis following chronic social

defeat. Overall, the effects of six days of social defeat stress exceeded the results attained

with the fifteen days of CVS previously described.

The brain sites activated by social defeat have been studied using c-fos

immunohi stochemi stry. The immediate-early gene (IEG) c-fos is expressed in many cells

in the brain, but typically at very low basal levels. C-fos and other IEGs are intracellular

signaling mechanisms that regulate gene transcription and expression of various

neuropeptides and trophic molecules in response to stress (for review, see Sabban and

Kvetnansky, 2001). Various stressful stimuli can initiate increased levels of c-fos mRNA

expression, lasting for minutes to hours. For instance, social defeat produces elevated

c-fos expression in limbic, limbic-associated, and brainstem sites in both hamsters and

rats one hour after a single defeat session (Kollack-Walker et al., 1997; Martinez et al.,

1998). These results point to the brain structures that are important in the processing of

emotionally stressful events. However, after repeated social defeat sessions in rats, the

pattern of neuronal activity was modified, despite the fact that intruder submissive

behavior persisted across trials. C-fos mRNA expression endured for many of these

limbic and brainstem nuclei, while other limbic and brainstem regions exhibited a

decrease in the social defeat-induced c-fos mRNA expression.

The Porsolt swim test, originally described by Porsolt and colleagues (1978), is the

most commonly utilized behavioral test for screening antidepressant treatments in rats

and has been used to evaluate the behavioral effects of stress exposure. Immobility

during the inescapable swim is measured as an indicator of behavioral despair or










depressive-like behavior. Regimens of chronic stress or repeated administration of

gluccocorticoids increase the amount of time an animal spends immobile, decrease the

latency to immobility, and decrease the amount of time engaged in active swimming

(Molina et al., 1994; Johnson et al., 2006). Antidepressant drugs, on the other hand,

ameliorate these effects. For example, desipramine (a tricyclic antidepressant) and

fluoxetine (a selective serotonin reuptake inhibitor) reduce immobility and increase the

amount of time a stressed animal will struggle to escape the forced swim (Molina et al.,

1994; Lucki et al., 2001).

Based on the results of investigations using social defeat stress and the Porsolt

swim test, we have begun to further characterize the social defeat model of emotional

stress in rats. We have evaluated the impact of both acute and repeated social defeat

stress on regulation of ACTH and CORT at various times after the stressor. We have

also compared six daily social defeat sessions to one month of stress every third day to

examine the effectiveness of massed and intermittent stress exposure. In addition, we

exposed socially defeated rats to the Porsolt swim test in order to characterize the

enduring behavioral effects of social defeat stress.















CHAPTER 2
IVETHOD S

Animals

Ninety male Long-Evans (LE) rats and twelve female LE rats were purchased from

Harlan Co. of Indianapolis, INT. Twenty of the male Long Evans (LE) rats (225-250 g)

were used as unhandled controls. Fifty-eight of the male LE rats (225-250 g) were used

as intruders. Twelve of the male LE rats (300-325 g) were used as residents and the

twelve female LE rats (225-250 g) were used as housing mates for the residents. The

weight ranges indicate weights at the time the rats were purchased. The intruders

weighed 275-325 g and the residents weighed 500-800 g at the time of the experiments.

Six of the unhandled control rats were used in Experiment la, six were used in

Experiment lb, and eight were used in Experiment 2. Thirty of the intruder rats were

used in Experiment la, twelve were used in Experiment lb, and sixteen were used in

Experiment 2. The resident males and the females were used in multiple experiments.

The rats were housed in polycarbonate cages (43 x 21.5 x 25.5 cm) with sex-

matched and weight-matched pairings for five to seven days of acclimation to a

12hr/12hr light/dark cycle (lights on at 7:00a.m.). Standard chow (LabDiet 5001) and tap

water were available ad libitum. Temperature and humidity in the housing facility were

controlled to 22.720C & 0.940C and 54.1% + 14.6%, respectively.

The intruder rats were exposed to social defeat stress and remained pair-housed

throughout the experiment. The control rats were not exposed to social defeat stress and

were pair-housed throughout the experiment. The resident rats were vasectomized,










singly-housed, and given 10 days to recover from the surgery. The residents were then

pair-housed in a separate housing room, each with a female, for 2 weeks prior to and then

throughout the experiment. All procedures were pre-approved by the Institutional

Animal Care and Use Committee at the University of Florida and conducted in

accordance with the Guide for the Care and Use of Laboratory Animals.

Surgical Procedures

After a week of acclimation to the housing facility, the resident rats were

anaesthetized with ketamine:xylazine (50 mg/kg:5 mg/kg, i.p.). Sedation was verified by

testing the pedal withdrawal reflex 10 min after injection and every 10 min during the

procedure. After sedation was verified, each rat was shaved from the rostral edge of the

scrotum to the caudal abdomen. The area was then washed with Betadine three times.

Surgery began with a 2 cm ventral midline incision, just above the scrotal area.

Through the incision, the vas deferens was located and externalized. A 0.25 cm section

of each duct was removed with a micro-cautery knife. The muscle walls were sutured

and the external incision was closed using sterilized staples that were removed ten days

after the surgery. Each surgical procedure lasted 15-30 min.

Each rat was given a 1 mL inj section of 0.9% warm saline and ketorolac

subcutaneously and then placed into a post-operative Plexiglas cage, heated by an electric

heating pad. Following observation of locomotion and urination, each rat was returned to

its individual home cage and was monitored daily throughout the recovery period.

Experimental Procedures

General Social Defeat Stress Procedure

At the beginning of the 10-min social defeat session, the female rat paired with one

resident rat was taken out of the cage and placed into an empty, identical cage. The









intruder rat was then placed into the cage of the resident (direct interaction). The session

was then closely monitored for submissive, supine posture expressed by the intruder.

Each time the intruder exhibited a supine posture with the resident physically contacting

the intruder for 2 sec or more, one defeat was counted. After three defeats or a total of

5 min elapsed (whichever came first), the intruder was taken out of the resident' s cage

momentarily and quickly placed into a 10 x 10 x 15 cm double-walled protective wire

mesh cage. The intruder (within the wire mesh cage) was then placed back into the

resident' s cage for the remainder of the 10-min session (indirect interaction). The two

intruder rats in each housing pair were run in simultaneous defeat sessions in the cages of

two resident males that were placed side by side. Each pair of cage-mates received the

same treatment. Each social defeat session was run between 8:00a.m. and 10:00a.m., and

was videotaped for further analysis.

Social Defeat Stress Regimen: Experiment la

Thirty-six naive male LE rats were assigned to six experimental groups (Table 2-1).

The control rats from Group 1 (n = 6) remained in their original cages throughout the

experiment and were rapidly decapitated between 8:00a.m. and 10:00a.m. on the final

experimental day. These rats were exposed to no chronic and no acute stress (NC/NA).

The intruders from Group 2 (n = 6) were exposed to social defeat stress once every 24 h

across six experimental days and were killed by rapid decapitation between 8:00a.m. and

10:00a.m., 24 h after their final social defeat session. These rats were exposed to chronic,

but not acute stress (C/NA). The intruders from Group 3 (n = 6) were only exposed to

social defeat stress on the final experimental day and were then immediately decapitated,

10 min after the start of the social defeat session (t = 10 min). These rats were exposed to

no chronic, just acute stress (NC/A-10). The intruders from Group 4 (n = 6) were only










exposed to social defeat stress on the final experimental day and were rapidly decapitated

30 min after the start of the social defeat session (t = 30 min). These rats were exposed to

no chronic, just acute stress (NC/A-30). The intruders from Group 5 (n = 6) were

exposed to social defeat stress once every 24 h over the course of six experimental days

and were killed by rapid decapitation 10 min after the start of the social defeat session on

the sixth day (t = 10 min). These rats were exposed to chronic and acute stress (C/A-10).

The intruders from Group 6 (n = 6) were exposed to social defeat stress once every 24 h

over the course of six experimental days and were killed by rapid decapitation 30 min

after the start of the social defeat session on the sixth day (t = 30 min). These rats were

exposed to chronic and acute stress (C/A-30). Each intruder from the repeatedly stressed

groups was placed in the cage of a different resident each day of his chronic stress

routine.

Social Defeat Stress Regimen: Experiment lb

Eighteen naive male LE rats were assigned to three experimental groups (Table

2-2). The unhandled control rats from Group 1 (n = 6) remained in their original cages

throughout the experiment and were killed by rapid decapitation between 8:00a.m. and

10:00a.m. on the final experimental day. These rats were exposed to no chronic and no

acute stress (NC/NA). The intruders from Group 2 (n = 6) were exposed to social defeat

stress once every 72 h across thirty-four experimental days and were rapidly decapitated

between 8:00a.m. and 10:00a.m., 24 h after their final social defeat session. These rats

were exposed to chronic, but not acute stress (C/NA). The intruders from Group 3 (n = 6)

were exposed to social defeat stress once every 72 h across thirty-four experimental

days and were killed by rapid decapitation 30 min after the start of the social defeat

session on the final day (t = 30 min). These rats were exposed to chronic and acute stress










(C/A). All intruder rats from Experiment lb were exposed to a total of twelve social

defeat sessions. Each intruder from the repeatedly stressed groups was placed in the cage

of a different resident each day of his chronic stress routine.

Social Defeat Stress and Porsolt Swim Test: Experiment 2

Twenty-four naive male LE rats were assigned to three experimental groups (Table

2-3). The unhandled control rats from Group 1 (n = 8) remained in their original cages

throughout the social defeat portion of the experiment. These rats were exposed to no

chronic and no acute social defeat stress (NC/NA). The intruders from Group 2 (n = 8)

were only exposed to social defeat stress on the final day of the social defeat portion of

the experiment. These rats were exposed to no chronic, just acute stress (NC/A). The

intruders from Group 3 (n = 8) were repeatedly exposed to social defeat stress over the

course of five days, once every 24 h. These rats were exposed to chronic and acute stress

(C/A). Each repeatedly stressed intruder was placed in the cage of a different resident

each day of his chronic stress routine. All of the rats from each of the three groups were

then exposed to the Porsolt swim test.

Twenty-four hours after the final social defeat session, each rat from Experiment

2 was individually removed from its home cage and placed into a plastic cylinder filled

with approximately 25 cm of clean tap water at 24-270C. The depth of water allowed

each rat to reach the bottom of the cylinder with its tail, with enough head room that the

rat was unable to escape from the tank. The water in each cylinder was changed between

trials. Each rat was subjected, one at a time, to a 15-min swim session then carefully

dried and returned to its home cage. Each Porsolt swim session was run between

8:00a.m. and 10:00a.m., and was videotaped for further analysis.









Behavioral Assays

Number of defeats, latency to first defeat, freezing behavior, and exploratory

locomotion were scored from the recorded social defeat sessions. Freezing behavior was

recorded whenever the intruder remained motionless for at least 2 sec, and is reported as

percent time during the direct interaction between the intruder and the resident.

Exploratory locomotion was recorded every time the intruder crossed with all four paws

from one third of the cage into the adj acent third of the cage. Vertical lines were drawn

on the video image of each resident' s home cage and exploratory locomotion was scored

from the videotapes. Exploratory locomotion is reported as lines crossed per minute of

direct interaction time. Two trained observers scored the social defeat stress videos from

Experiments la and lb and inter-observer reliability was assessed.

Immobility was recorded during the Porsolt swim tests whenever the rat balanced

on its tail, completely motionless, or exhibited only slight forepaw movement for a

minimum of 2 sec. This behavior is reported as total time spent immobile during the

swim test. Two trained observers scored the swim test videos from Experiment 2 and

inter-observer reliability was assessed.

Histological Assays

Immediately after decapitation in experiments la and lb, 6 mL of trunk blood from

each rat was collected in polypropylene tubes on ice with 600 Cl1 of Na2EDTA at

20 Cpg/CLL. The blood samples were then immediately centrifuged at 1000 ref for 5 min

and the plasma fraction frozen in 300 Cl1 aliquots at -800C. The brain from each intruder

was quickly removed, frozen in 2-methylbutane at -400C, and stored at -800C for future

analysis of molecular variables involved in HPA axis function. The thymus and adrenal

glands were removed from each rat and frozen separately at -800C. These glands were









later weighed to determine the health and stress status of each rat at time of death.

Plasma ACTH and CORT concentrations from the rats were later analyzed with

radioimmunoassay (RIA). ACTH RIAs were run with kits from Alpco Diagnostics

(Salem, NH) and CORT RIAs were run with kits from Diagnostic Products Corporation

(Los Angeles, CA).

Statistical Analyses

Potential between-groups differences in number of defeats and latency to first

defeat were analyzed using two one-way analyses of variance (ANOVAs) to compare the

five groups of intruders (i.e. three repeatedly stressed, and two just acutely stressed) in

Experiment la during their first exposure to social defeat stress. A 3x6 (group x session)

repeated-measures ANOVA (for Experiment la) and a 2xl2 (group x session) repeated-

measures ANOVA (for Experiment lb) were used to examine potential differences in

number of defeats between groups and across experimental sessions for the repeatedly

stressed groups. A 3x6 (group x session) repeated-measures ANOVA (for Experiment

la) and a 2xl2 (group x session) repeated-measures ANOVA (for Experiment lb) were

used to examine potential differences in latency to first defeat between groups and across

experimental sessions for the repeatedly stressed groups.

Potential between-groups differences in freezing responses were analyzed using a

one-way ANOVA to compare freezing behavior between the five groups in Experiment

la during their first exposure to social defeat stress. A 3x6 (group x session) repeated-

measures ANOVA (for Experiment la) and a 2xl2 (group x session) repeated-measures

ANOVA (for Experiment lb) were used to examine potential freezing behavior

differences between groups and across experimental sessions for the repeatedly stressed

groups.









Potential between-groups differences in exploratory locomotion were analyzed

using a one-way analysis of variance (ANOVA) to compare exploratory locomotion

between the five groups in Experiment la during their first exposure to social defeat

stress. A 3x6 (group x session) repeated-measures ANOVA (for Experiment la) and a

2xl2 (group x session) repeated-measures ANOVA (for Experiment lb) were used to

examine potential exploratory locomotion differences between groups and across

experimental sessions for the repeatedly stressed groups.

Potential between-groups differences in plasma ACTH concentrations and in

plasma CORT concentrations were analyzed with one-way ANOVAs in each of

Experiments la and lb. Also, potential between-groups differences in adrenal weights

and in thymus weights were analyzed with one-way ANOVAs in each of Experiments la

and lb. The results were further analyzed using pre-planned Newman-Keuls multiple

comparison tests for all significant ANOVAs.

Potential between-groups differences in immobility duration scores for the 15-min

Porsolt swim test session in Experiment 2 were analyzed using a one-way ANOVA. The

results were further analyzed using Newman-Keuls multiple comparison tests. The data

were then analyzed in 5-min bins using a 3x3 (group x bin) repeated-measures ANOVA.

The results were further analyzed using Newman-Keuls multiple comparison tests,

making pair-wise comparisons between each stressed group and the control group, for

each bin.

Inter-observer agreement was assessed for freezing behavior in Experiments la and

lb by comparing total times recorded by each trained observer for each social defeat

session. Inter-observer agreement was also assessed for exploratory locomotion in










Experiments la and lb by comparing total number of line crossings recorded by each

trained observer for each social defeat session. Inter-observer agreement was assessed

for immobility duration scores for the Porsolt swim test from Experiment 2 by comparing

total immobility times recorded by each trained observer for each test session.











Table 2-1. Experiment la, group assignments
Group Repeated Stress Acute Stress Kill Time
NC/NA
C/NA X t = 24 h
NC/A-10 X t =10 min
NC/A-30 X t = 30 min
C/A-10 X X t =10 min
C/A-30 X X t = 30 min

Table 2-2. Experiment lb, group assignments
Group Repeated Stress Acute Stress Kill Time
NC/NA
C/NA X t = 24 h
C/A X X t = 30 min

Table 2-3. Experiment 2, group assignments
Group Repeated Stress Acute Stress Porsolt Swim Test
NC/NA X
NC/A X X
C/A X X X















CHAPTER 3
RESULTS

Experiment la: Six Daily Sessions of Social Defeat Stress

During their first exposure to social defeat stress, the rats from all Eive of the

groups in Experiment la did not show statistically different numbers of defeats (F (4, 25)

= 0.2419, p < 0.9118; Fig. 3-1A) nor latencies to Birst defeat (F (4, 25) = 0.7308, p <

0.5795; Fig. 3-1B). Each of the three repeatedly stressed groups (C/NA, C/A-10, and

C/A-30), showed no significant between-groups difference, time effect, or group by time

interaction effect in number of defeats (F (2, 15) = 1.603, p < 0.2340; F (5, 15) = 0.3463,

p < 0.8831; F (10, 15) = 0.3139, p < 0.9753; Fig. 3-1A) nor latencies to first defeat across

the six experimental sessions (F (2, 15) = 1.571, p < 0.2401; F (5, 15) = 0.2106, p <

0.9570; F (10, 15) = 0.5255, p < 0.8669; Fig. 3-1B).

During their first exposure to social defeat stress, the rats from all Eive of the

groups did not show statistically different amounts of freezing behavior (F (4, 25) =

0.1757, p < 0.8405; Fig. 3-2A). Each of the three repeatedly stressed groups showed

significantly increased freezing behavior across the six experimental sessions (F (5, 75) =

17.08, p < 0.0001; Fig. 3-2A), reaching asymptote by day 3, with no significant between-

groups differences or group by time interaction effect.

During their first exposure to social defeat stress, the rats from all five groups did

not show statistically different amounts of exploratory locomotion (F (4, 25) = 0.5711i, p

< 0.5767; Fig. 3-2B). Each of the three repeatedly stressed groups showed significantly

decreased exploratory locomotion across the six experimental sessions (F (5, 75) = 11.76,









p < 0.0001; Fig. 3-2B), reaching asymptote by day 4, with no significant between-groups

differences or group by time interaction effect.

Exposure to acute social defeat (NC/A and C/A) produced significant elevations in

circulating ACTH concentrations when compared with the ACTH concentrations in the

control rats (NC/NA) (F (5, 30) = 3.874, p < 0.0079; Fig. 3-3A). Circulating ACTH

concentrations were not elevated in the chronically, but not acutely, stressed rats (C/NA).

Exposure to social defeat stress for all groups (including C/NA) produced significant

elevations in circulating CORT concentrations when compared with the CORT

concentrations in the control rats (F (5, 30) = 20.79, p < 0.0001; Fig. 3-3B).

Exposure to social defeat significantly decreased thymus masses in Groups C/A-10

and C/A-30, but not for the other groups (F (2, 15) = 5.651, p < 0.0148; Fig. 3-4A).

There were no significant differences in adrenal masses between the rats in the stressed

groups and the rats in the control group (F (5, 30) = 0.9299, p < 0.4756; Fig. 3-4B).













,3-


2- -5- C/NA
-9- NCIA-10
.0 -- NCIA-30
E 1
3- C/A-10
-A- C/A-30

1 2 3 4 5 6
Experimental Day




s 125




o 75
-5- C/NA
,, I +-/-- NCIA-10
9 I T-+-- NCIA-30
o"25-1 -6 C/A-10
I I-- C/A-30

1 2 3 4 5 6
Experimental Day


Figure 3-1. Social defeats per daily session. The rats exposed to repeated social defeat
stress (C/NA, C/A-10, C/A-30) exhibited equivalent (A) number of defeats
per session and (B) latency to first defeat per session across the 6
experimental sessions. The rats that received only one acute defeat session
(NCA-10 and NCA-30) were stressed at the same time as the final stress
session for the chronically stressed groups, and so the values for these groups
are illustrated on day 6. Values expressed are group means + SEM (n = 6 rats
per group). Abbreviations: C/NA = chronic stress/no acute stress, NC/A = no
chronic stress/acute stress, C/A = chronic stress/acute stress.












15' 1 0 0

90-
Ss-


S70-

a60-

50-

40-

S30-

20-

10-
.

0-


-5- C/NA
-V NC/A-10
-9 NC/A-30
-6 C/A-10
-A- C/A-30


Exp~erimental Day


56


-5- C/NA
-V NC/A-10
-9- NC/A-30
-6 C/A-10
-A- C/A-30


Exp~erimental Day


I I


Figure 3-2. Intruder behavior during daily social defeat sessions. The rats exposed to
repeated social defeat stress exhibited (A) increases in freezing behavior and
(B) decreases in exploratory locomotion across the 6 experimental sessions.
The rats that received only one acute defeat session were stressed at the same
time as the final stress session for the chronically stressed groups, and so the
values for these groups are illustrated on day 6. Values expressed are group
means + SEM (n = 6 rats per group).











n.s.
n.s. n.s.


A






o
1--


T


n.s.


Je

I


~ ~


n.s.
+ +


B






o


1--i

O


Je
I


~ ~


Figure 3-3. Circulating hormones after daily social defeat stress. The rats exposed to
social defeat stress had elevated circulating concentrations of (A) ACTH and
(B) CORT when compared to basal concentrations in control rats. Values
expressed are group means & SEM (n = 6 rats per group). Significant
differences between the socially defeated rats and the unhandled control rats
(NC/NA) are expressed as [*] p < 0.05. Significant differences in pre-planned
comparisons between the stressed groups of rats are expressed as [+] p < 0.05,
with lines connecting the groups that were compared. Additional
abbreviation: NC/NA = no chronic stress/no acute stress.






















300-`
200- c









1--~
100-O p'




Biur 60-4 fet di sca ee srs gadl mse Sca ee te

50- infcnl dcesdtyu ase o ftereetdysrse
40us(/-0adCA3) u B i ntsgiiatyafc dea ln







Figre -4.Efect gofp) daniyscial deffeatsress oewen glndla mcasss.Soia defeated ras dtres


control rats are expressed as [*] p < 0.05.









Experiment lb: Twelve Intermittent Sessions of Social Defeat Stress

There were no significant between-groups differences, time effect, or group by time

interaction effect in number of defeats (F (1, 10) = 0.0787, p < 0.7847; F (11, 10) =

0.9032, p < 0.5399; F (11, 10) = 0.8203, p < 0.6199; Fig. 3-5A) nor latency to first defeat

(F (1, 10) = 0.4653, p < 0.5107; F (11, 10) = 0.8400, p < 0.6007; F (11, 10) = 0.3414, p <

0.9743; Fig. 3-5B) for the repeatedly stressed groups (C/NA and C/A) in Experiment lb.

Freezing behavior significantly increased for both of the repeatedly stressed groups

across the twelve experimental sessions (F (11, 10) = 6.224, p < 0.0001; Fig. 3-6A), with

no significant between-groups differences or group by time interaction effect.

Exploratory locomotion significantly decreased for both of the repeatedly stressed

groups across the twelve experimental sessions (F (11, 10) = 7.252, p < 0.0001; Fig.

3-6B), with no significant between-groups differences or group by time interaction effect.

Acute exposure to social defeat stress produced significant elevations in circulating

ACTH concentrations when compared with the basal ACTH concentrations in the control

rats (F (2, 15) = 4.525, p < 0.0290; Fig. 3-7A). Acute exposure to social defeat stress also

produced significant elevations in circulating CORT concentrations when compared with

the basal CORT concentrations in the control rats (F (2, 15) = 5.853, p < 0.0132; Fig.

3-7B). These elevations in circulating hormones were limited to the rats that were

stressed acutely before termination (C/A) the basal concentrations in the other

chronically stressed rats (C/NA) did not significantly differ from the basal concentrations

in the control rats (NC/NA).

There were no significant differences in thymus masses (F (2, 15) = 0.04457, p <

0.9565; Fig. 3-8A) or adrenal masses (F (2, 15) = 1.826, p < 0. 1951; Fig. 3-8B) between

the rats in the socially defeated groups and the rats in the control group.
















3-

O 2-
O




-t- C/NA
-A- C/A


0 12 3 4


5 6 7
Exp~erimental Day


8 9 10 11 12


S100-





050-


S25-



1 2 3 4 5 6 7 8 9 10 11 12
Experimental Day


-t- C/NA
-A- C/A


Figure 3-5. Social defeats per intermittent session. The rats exposed to repeated social
defeat stress (C/NA and C/A) exhibited equivalent (A) number of defeats per
session and (B) latency to first defeat per session across the 12 experimental
sessions. Values expressed are group means + SEM (n = 6 rats per group).






















































11 I I f I -5- C/NA- /

0 1 2 3 4 5 6 7 8 9 10 11 12
Defeat Session


Figure 3-6. Intruder behaviors during intermittent social defeat sessions. The rats
exposed to repeated social defeat stress exhibited (A) increases in freezing
behavior and (B) decreases in exploratory locomotion across 12 experimental
sessions. Values expressed are group means + SEM (n = 6 rats per group).


-5- C/NA
-A- C/A


I I I I I I
012345


Ii I
Defeat Session


IIb I' I












200-
-

S150-


S100-
o

S50-
-


-r


Figure 3-7. Circulating hormones after intermittent social defeat stress. Immediately
after, but not 24 h later, the rats exposed to repeated social defeat stress had
elevated circulating concentrations of (A) ACTH and (B) CORT when
compared to basal concentrations in the control rats (NC/NA). Values
expressed are group means & SEM (n = 6 rats per group). Significant
differences between the socially defeated rats and the unhandled control rats
are expressed as [*] p < 0.05.


B60,

501


-r


O~L~LL~L


I

~


Figure 3-8. Effects of intermittent social defeat stress on glandular masses. Repeated
social defeat stress had no effect on (A) thymus masses or (B) adrenal gland
masses. Values expressed are group means & SEM (n = 6 rats per group).


I I









Experiment 2: Social Defeat Stress Followed by Porsolt Swim Testing

Exposure to chronic social defeat stress (C/A) produced a significant increase in

total immobility time during the 15-min forced swim when compared with the immobility

times for the control rats that were not exposed to social defeat stress (NC/NA) (F (2, 21)

= 5.363, p < 0.0131; Fig. 3-9). When the data were analyzed in 5-min bins, there was a

significant between-groups effect as well as a significant time effect (F (2, 42) = 2.424, p

< 0.0150; F (2, 42) = 72.74, p < 0.0001; Fig. 3-10), but no significant group by time

interaction effect (F (4, 42) = 1.270, p < 0.2970; Fig. 3-10).

Inter-Observer Reliability

The two observer's recordings of total time freezing for the social defeat sessions

from Experiments la and lb combined differed by less than 20 sec for 94% of the

sessions and never differed by more than 28 sec. The two observer's recordings of

exploratory locomotion during social defeat sessions were identical in 70% of the

sessions and never differed by more than 3 lines crossed.

The two observer's recordings of total immobility time during the Porsolt swim test

differed by less than 40 sec in 92% of the sessions and never differed by more than 43 sec.














a, 700 1





NC-N 600- C/
Fiur 50-9.Imblt uigteProtsi et h asepsdt eetdsc
deet 400-(hoic xiie ncesdimbiiyfrteduaino h
O osl ocdsi et hncmae oteimblt ie o h





E 300-J J

-200-

E 100-




Figure 3-90. Immobility esr i -nu bs during the Porsolt swim test. The rt xoe o eetdsca
defeat stresscn (Chrnic exhibitedo tinceased immeegobilit dfor teduainc ofhen

intpruder rts xoe oaue socially defeated stress (Acute n hoi) and the control rt
go (No SD). Values expressed are group means & SEM (n = 8 rats pergru)
gop Significant differences between the socinially defeated rats and theunade
controls are expressed as [*] p < 0.05. Addnitioanal abbrevia btioen: No D
nosciael defeated satrss.dtecnrl r xpesda + .5


















CHAPTER 4
DISCUSSION

Overall, repeated social defeat stress produced behavioral, hormonal, and glandular

changes that model some of the symptoms that are seen in depressed humans. Behavioral

despair, HPA axis dysregulation, and thymus involution were found in rats that were

exposed to daily social stress; and these effects were greater overall than the effects that

were seen in the rats that were stressed intermittently.

Daily Social Defeat Stress

The equivalent mean freezing scores and the equivalent mean locomotion scores

during the first social defeat session indicate that all five groups that experienced social

defeat stress were comparable at the beginning of the experiment. The equivalent

progressive increases in freezing behavior and the equivalent decreases in exploratory

locomotion for the three repeatedly stressed groups indicate that these groups underwent

comparable stress-induced behavioral changes across social defeat sessions. These

behaviors, however, did not diminish the persistent displays of dominant behavior

(consistent number of defeats and equivalent latencies to first defeat) by the resident rats.

After 3-4 days of stress exposure, the behavioral curves for both freezing behavior and

exploratory locomotion reached asymptote, suggesting that the behavioral changes

associated with daily social defeat stress do not progress beyond the first few exposures.

The equivalent mean ACTH concentrations and the equivalent mean CORT

concentrations 10 min after the onset of stress between the groups with and without a

history of stress suggest that there is similar activation of the HPA axis for these groups.









However, the lower CORT concentrations 30 min after the onset of stress in the

repeatedly stressed group compared to the CORT concentrations in the acutely stressed

group suggests a sensitization in the negative feedback regulation of CORT.

Decreased thymus mass and/or increased adrenal mass has been shown to result

from chronic stress regimens involving physical and/or psychological stressors (for

examples, see Blanchard et al., 1998; Bryant et al., 1991; Simpkiss and Devine, 2003) as

well as from major depression in victims of suicide (Szigethy et al., 1994). Given that

two of the three groups that were stressed daily (C/A-10 and C/A-30) exhibited thymus

involution while one (C/NA) did not, it cannot be concluded that the daily stress regimen

consistently caused a decrease in thymus mass. These findings, along with adrenal mass

equivalency across groups regardless of number of stress exposures, suggest that the

chronic social defeat stress regimen of six daily exposures may not have been adequate to

consistently alter glandular masses. A longer or unpredictable social defeat stress

regimen could possibly generate consistent thymus involution and adrenal hypertrophy.

Perhaps an additional stressor for the intruders, such as isolation housing to ensure more

timid behavior (Kabbaj et al., 2000), could contribute to the effectiveness of the regimen.

Also, using a restraint stress harness instead of the protective wire mesh cage could

elevate the efficiency of the regimen by producing an inescapable condition for the

intruders .

Intermittent Social Defeat Stress

The equivalent progressive increases in freezing behavior and the equivalent

progressive decreases in exploratory locomotion in the intermittent stress regimen

indicate that the repeatedly stressed groups underwent comparable stress-induced

behavioral changes. However, the escalation of freezing behavior and the decline of










exploratory locomotion occurred more gradually when the social defeat sessions were

72 hours apart than when the social defeat sessions occurred daily. This suggests that the

intermittent stress regimen was not as efficient at inducing these behavioral changes,

although the overall magnitudes of the behavioral effects were eventually roughly

equivalent. Once again, the intruders' behavioral adaptations did not decrease the

displays of agonistic behavior from the residents (consistent number of defeats and

equivalent latencies to first defeat across groups and days).

The equivalence in basal CORT concentrations as well as the equivalence in

glandular masses between the chronically stressed animals and the control animals

indicates that the temporally spaced regimen was not potent enough to sufficiently alter

hormonal or glandular basal states. Even though the more temporally spaced chronic

stress regimen was able to produce equivalent effects with regards to behavioral changes

(albeit, more slowly), it was less effective than the daily regimen in producing the

hormonal and glandular changes that correlate with some of the symptoms of maj or

depression. The results indicate, therefore, that there may be a critical window for

vulnerability to an additional stressor--it is possible that the rats were able to partially

recover from the initial stressor by the next exposure in the more temporally spaced

chronic regimen.

Relevance to Previous Work

Overall, these results confirm and extend a previous report that daily social defeat

is an effective emotional stressor for male rats. In a study by Haller and colleagues

(1999), intruder rats were exposed to resident rats for 4 hours on four consecutive days,

producing increased basal CORT concentrations in the intruders. The daily social defeat

stress regimen used in our study effectively increased basal CORT concentrations as










well, with briefer stressful interactions than previously reported. However, in the same

study, Haller and colleagues found increased adrenal mass with no decrease in thymus

mass, results opposite to those seen after our similar daily chronic stress regimen.

Perhaps it is necessary for a daily chronic stress regimen to extend past four (or six) days

to generate reliable glandular changes.

The HPA axis response to an acute stressor largely depends upon whether the acute

stressor is heterotypic or homotypic. In a study by Armario and colleagues (2004),

investigators reported that when previously-stressed rats (via immobilization) were

presented with a heterotypic stressor (forced swimming), a minor sensitization was

observed. Also, following exposure to a severe stressor (such as shock, restraint,

immobilization, or large doses of endotoxin), there was habituation of the HPA axis

response to a homotypic stressor. These results differ from the results obtained after

daily social defeat stress. Following our repeated daily social defeat stress regimen, there

was a robust activation but rapid shutdown of CORT release. Most probably, the

difference is due to the fact that Armario and colleagues utilized a single stressor

followed by either homotypic or heterotypic challenges. Our model of social defeat does

not appear to function as a homotypic stress regimen. There are several variables

involved in the direct interaction phase of social defeat stress (physical contact, olfactory

cues, and ultra-sonic vocalizations) as well as the introduction of a different resident rat

with each exposure. Conceivably, the use of a different resident for each stress exposure

and the potentially different threats imposed by these different residents creates a novel,

therefore heterotypic, situation with each session.










In general, the behavioral changes in the intruder rats after repeated social defeat

stress model some of the symptoms expressed in clinically depressed patients, including

impairment in social or occupational functioning and loss of interest (DSM IV, 1994).

Also, the elevated basal CORT concentrations after daily social defeat stress during the

nadir of the daily cycle confirm previous findings from our laboratory (Lopes and

Devine, 2004) and mimic the augmented circulating hormones found in clinically

depressed patients (Linkowski et al., 1985; Pfohl et al., 1985).

Porsolt Swim Test

Based on the combined results from the daily and the intermittent stress regimens, a

daily social defeat stress regimen was used to for the Porsolt study to optimize the

potential to produce behavioral despair. The fact that acute social defeat stress did not

produce as much immobility as repeated stress indicates that the neuronal plasticity

caused by repeated exposure to social stress (demonstrated through hormonal and

behavioral plasticity) is a determining factor in the expression of behavioral despair.

A 5-min swim test is commonly used for detection of behavioral despair (Gavioli et

al., 2003; Hinojosa et al., 2006; Porsolt et al., 1978; Rygula et al., 2005). Interestingly,

when analyzed in 5-min bins, the data suggest that in all the bins the repeatedly stressed

rats exhibited more immobility than the controls did, and in one bin (5-10 min) the

acutely defeated rats also exhibited more immobility than the controls did. Therefore, a

10-min session may be necessary to closely examine the subtle differences between

groups that have a less severe stress history and non-stressed control rats.

Overall, these results model the behavioral despair, or low mood and anhedonia,

found in clinically depressed patients (for review, see Harrison, 2002). Also, the

significantly elevated immobility times for the repeatedly stressed rats confirm and









extend the results from a previous report. In a study by Rygula and colleagues (2005),

investigators used a more extensive stress regimen of hour-long daily sessions of social

defeat for five straight weeks to elicit behavioral despair in subsequent Porsolt swim

testing. The daily social defeat stress regimen used in our study effectively increased

immobility in the repeatedly stressed rats as well, with briefer stressful interactions and a

shorter regimen than previously reported.

Conclusions and Future Directions

In conclusion, the behavioral, hormonal, and glandular changes produced by

repeated social defeat closely resemble many of the psychopathological symptoms in

patients with major depression. The daily social defeat stress regimen provides an

interesting and effective model for stress-induced psychopathology.

The effect of repeated social defeat on the overall circadian rhythm was not

evaluated, nor was the persistence of the altered regulation across days, weeks, or

months. Both of these topics may be interesting issues for future studies. Also, potential

adaptations in stress-related molecules (CRH and AVP in the hypothalamus, CRH1 and

Vlb in the pituitary and amygdala, and mineralocorticoid receptor and gluccocorticoid

receptor in the hippocampus) after social defeat should be investigated. These studies

will enhance our growing knowledge of the neurobiological basis for stress-induced

psychopathology (e.g., maj or depression).
















LIST OF REFERENCES


Agid O, Kohn Y, Lerer B (2000). Environmental stress and psychiatric illness.
Bionzedicine andPharnacotherapy,, 54: 135-141.

Amiel-Tison C, Cabrol D, Denver R, Jarreau P, Papiernik E, Piazza PV (2004). Fetal
adaptation to stress part II. Evolutionary aspects; stress-induced hippocampal
damage; long-term effects on behavior; consequences on adult health. Early
Human Development, 78: 81-94.

Anisman H, Matheson K (2005). Stress, depression, and anhedonia: Caveats concerning
animal model s. Neuroscience and Biobehavioral Reviews, 29: 525-546.

Armario A, Valles A, Dal-Zotto S, Marquez C, Belda X (2004). A single exposure to
severe stressors causes long-term desensitization of the physiological response to
the homotypic stressor. Stress, 7 (3): 157-172.

Blanchard RJ, Nikulina JN, Sakai RR, McKittrick C, McEwen B, Blanchard DC (1998).
Behavioral and endocrine change following chronic predatory stress. Physiology
andBehavior, 63 (4): 561-569.

Bryant HU, Bernton EW, Kenner JR, Holaday JW (1991). Role of adrenal cortical
activation in the immunosuppressive effects of chronic morphine treatment.
Endocrinology, 128 (6): 3253-3258.

Chang CL, Hsu SYT (2004). Ancient evolution of stress-regulating peptides in
vertebrates. Peptides, 25: 1681-1688.

De Kloet ER (2003). Hormones, brain and stress. Endocrine Regulations, 37: 51-68.

Diagnostic and statistical manual of mental disorders, 4th ed. (1994). Washington, DC:
American Psychiatry Association.

Gavioli EC, Marzola G, Guerrini R, Bertorelli R, Zucchini S, De Lima TCM, Rae GA,
Salvadori S, Regoli D, Calo G (2003). Blockade of nociceptin/orphanin FQ-NOP
receptor signaling produces antidepressant-like effects: Pharmacological and
genetic evidences from the mouse forced swimming test. European Journal of
Neuroscience, 17: 1987-1990.

Haller J, Fuchs E, Halasz J, Makara GB (1999). Defeat is a maj or stressor in males while
social instability is mainly stressful for females: Towards the development of a
social stress model in female rats. Brain Research Bulletin, 50 (1): 33-39.










Harrison PJ (2002). The neuropathology of primary mood disorder. Brain, 125: 1428-
1449.

Herman JP, Cullinan WE (1997). Neurocircuitry of stress: Central control of the
hypothalamo-pituitary-adrenocortical axis. Trends in Neuroscience, 20 (2): 78-84.

Hinojosa FR, Spricigo L, Izidio G, Bruske GR, Lopes DM, Ramos A (2006). Evaluation
of two genetic animal models in behavioral tests of anxiety and depression.
Behavioural Brain Research, 168: 127-136.

Johnson SA, Fournier NM, Kalynchuk LE (2006). Effect of different doses of
corticosterone on depressive-like behavior and HPA axis responses to a novel
stressor. Behavioural Brain Research, 168: 280-288.

Kabbaj M, Devine DP, Savage VR, Akil H (2000). Neurobiological correlates of
individual differences in novelty-seeking behavior in the rat: Differential
expression of stress-related molecules. The Journal ofNeuroscience, 20 (18):
6983-6988.

Kessler RC, Berglund P, Demler O, Jin R, Koretz Z, Merikangas KR, Rush AJ, Walters
EE, Wang PS (2003). The epidemiology of major depressive disorder. Journal of
the American M\~edical Association, 289 (23): 3095-3 105.

Kollack-Walker S, Watson SJ, Akil H (1997). Social stress in hamsters: Defeat activates
specific neurocircuits within the brain. The Journal ofNeuroscience, 17 (22):
8842-8855.

Linkowski P, Mendlewicz J, Leclercq R, Brasseur M, Hubain P, Golstein J, Copinschi G,
Van Cauter E (1985). The 24-hour profile of adrenocorticotropin and cortisol in
major depressive illness. Journal of Clinical Endocrinology and Metabolism, 61
(3): 429-438.

Lopes KO, Devine DP (2004). Elevation of HPA axis activity in response to repeated
social defeat stress in rats. Society for Neuroscience Abstracts, 30: 426.12.

Lucki I, Dalvi A, Mayorga AJ (2001). Sensitivity to the effects of pharmacologically
selective antidepressants in different strains of mice. Psychopharmacology, 155:
315-322.

Martinez M, Phillips PJ, Herbert J (1998). Adaptation in patterns of c-fos expression in
the brain associated with exposure to either single or repeated social stress in male
rats. European Journal ofNeuroscience, 10: 20-33.

McEwen BS, Stellar E (1993). Stress and the individual. Archives oflnternal2\~edicine,
153: 2093-2101.










Miczek KA (1979). A new test for aggression in rats without aversive stimulation:
Differential effects of d-amphetamine and cocaine. Psychopharmacology, 60:
253-259.

Molina VA, Heyser CJ, Spear LP (1994). Chronic variable stress enhances the
stimulatory action of a low dose of morphine: Reversal by desipramine. European
Journal ofPharmacology, 260: 57-64.

Pfohl B, Sherman B, Schlechte J, Stone R (1985). Pituitary-adrenal axis rhythm
disturbances in psychiatric depression. Archives of GeneralIPsychiatry, 42: 897-
903.

Porsolt RD, Anton G, Blavet N, Jalfre M (1978). Behavioural despair in rats: A new
model sensitive to antidepressant treatments. European Journal ofPharmacology,
47: 379-391.

Raadsheer FC, Hoogendijk WJG, Stam FC, Tilders FJH, Swaab DF (1994). Increased
numbers of corticotrophin-releasing hormone expressing neurons in the
hypothalamic paraventricular nucleus of depressed patients. Clinical
Neuroendocrinology, 60: 436-444.

Rygula R, Abumaria N, Flugge G, Fuchs E, Ruther E, Havemann-Reinecke U (2005).
Anhedonia and motivational deficits in rats: Impact of chronic social stress.
Behavioural Brain Research, 162: 127-134.

Sabban EL, Kvetnansky R (2001). Stress-triggered activation of gene expression in
catecholaminergic systems: Dynamics of transcriptional events. 7RENDS in
Neurosciences, 24 (2): 91-98.

Sgoifo A, De Boer SF, Haller J, Koolhaas JM (1996). Individual differences in plasma
catecholamine and corticosterone stress responses of wild-type rats: Relationship
with aggression. Physiology and Beehavior, 60 (6): 1403-1407.

Simpkiss JL, Devine DP (2003). Responses of the HPA axis after chronic variable stress:
Effects of novel and familiar stressors. Neuroendocrinology Letters Nos. 1/2, 24:
97-103.

Szigethy E, Conwell Y, Forbes NT, Cox C, and Caine ED (1994). Adrenal weight and
morphology in victims of completed suicide. BiologicalPsychiatry, 36 (6): 374-
380.

Whitnall MH (1993). Regulation of the hypothalamic corticotrophin-releasing hormone
neurosecretory system. Progress in Neurobiology, 40: 573-629.
















BIOGRAPHICAL SKETCH

Kristen L. Stone graduated in May 2004 from the University of Central Florida

(Orlando) with her Bachelor of Science degree in psychology. She began her graduate

education in the Psychology Department at the University of Florida (Gainesville) in

August 2004, working toward her Master of Science degree, in the Behavioral

Neuroscience program.




Full Text

PAGE 1

RATS EXHIBIT BEHAVIORAL DESPA IR AND HORMONAL ALTERATIONS AFTER SOCIAL DEFEAT STRESS: IM PLICATIONS FOR MAJOR DEPRESSION By KRISTEN L. STONE A THESIS PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLOR IDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF SCIENCE UNIVERSITY OF FLORIDA 2006

PAGE 2

Copyright 2006 by Kristen L. Stone

PAGE 3

To my loving husband (Justin), my wonderful parents (Linda, Kenneth, and Donna), and my brothers and sisters (Kim, Rick, Minet, Roger, and Jay). Thank you for your uninterrupted love and support. Also to my extended family (both bi ological and in-law ) and my friends. Thank you for your continued encouragement and guidance.

PAGE 4

iv ACKNOWLEDGMENTS I would like to thank my supervisory committee members (Dr. Mohamed Kabbaj from Florida State University; Dr. Neil Ro wland; and Dr. Linda Hermer-Vazquez) for their time and patience. I would like to ex tend special gratitude to my advisor (Dr. Darragh Devine) for his persistent guidance ; and to Andrea Naccarato for her diligent work with the Porsolt swim test s. I would also like to th ank my mentors in Gainesville (especially Clare Mathes) for their support during this long journey of higher education. Lastly, and most importantly, I would like to thank God for the breadth of knowledge He has bestowed upon me.

PAGE 5

v TABLE OF CONTENTS page ACKNOWLEDGMENTS.................................................................................................iv LIST OF TABLES............................................................................................................vii LIST OF FIGURES.........................................................................................................viii ABSTRACT....................................................................................................................... ix CHAPTER 1 INTRODUCTION........................................................................................................1 2 METHODS...................................................................................................................7 Animals........................................................................................................................ .7 Surgical Procedures......................................................................................................8 Experimental Procedures..............................................................................................8 General Social Defeat Stress Procedure................................................................8 Social Defeat Stress Regi men: Experiment 1a.....................................................9 Social Defeat Stress Regi men: Experiment 1b...................................................10 Social Defeat Stress and Porsolt Swim Test: Experiment 2...............................11 Behavioral Assays......................................................................................................12 Histological Assays....................................................................................................12 Statistical Analyses.....................................................................................................13 3 RESULTS...................................................................................................................17 Experiment 1a: Six Daily Sessi ons of Social Defeat Stress......................................17 Experiment 1b: Twelve Intermittent Sessions of Social Defeat Stress.....................23 Experiment 2: Social Defeat St ress Followed by Porsolt Swim Testing...................27 Inter-Observer Reliability...........................................................................................27 4 DISCUSSION.............................................................................................................29 Daily Social Defeat Stress..........................................................................................29 Intermittent Social Defeat Stress................................................................................30 Relevance to Previous Work......................................................................................31

PAGE 6

vi Porsolt Swim Test.......................................................................................................33 Conclusions and Future Directions.............................................................................34 LIST OF REFERENCES...................................................................................................35 BIOGRAPHICAL SKETCH.............................................................................................38

PAGE 7

vii LIST OF TABLES Table page 2-1 Experiment 1a, group assignments...........................................................................16 2-2 Experiment 1b, group assignments..........................................................................16 2-3 Experiment 2, group assignments............................................................................16

PAGE 8

viii LIST OF FIGURES Figure page 3-1 Social defeats per daily session................................................................................19 3-2 Intruder behavior during daily social defeat sessions..............................................20 3-3 Circulating hormones after daily social defeat stress...............................................21 3-4 Effects of daily social def eat stress on glandular masses.........................................22 3-5 Social defeats per intermittent session.....................................................................24 3-6 Intruder behaviors during intermittent social defeat sessions..................................25 3-7 Circulating hormones after inte rmittent social defeat stress....................................26 3-8 Effects of intermittent social defeat stress on glandular masses..............................26 3-9 Immobility during the Porsolt swim test..................................................................28 3-10 Immobility measured in 5-minute bi ns during the Porsolt swim test......................28

PAGE 9

ix Abstract of Thesis Presen ted to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Master of Science RATS EXHIBIT BEHAVIORAL DESPA IR AND HORMONAL ALTERATIONS AFTER SOCIAL DEFEAT STRESS: IM PLICATIONS FOR MAJOR DEPRESSION By Kristen L. Stone December 2006 Chair: Darragh P. Devine Major Department: Psychology Chronic emotional stress plays a pivotal role in the origin of many psychiatric disorders, including major depression. Acco rdingly, we investigat ed the behavioral, hormonal, and glandular effects of both repe ated and acute emoti onal stress using the social defeat model in rats. We compared one week of da ily social defeat exposure to one month of intermittent exposure to exam ine the effectiveness of massed versus intermittent stressors. In two experiments, nave male intruder rats were each exposed to dominant male resident rats. Each resident and intruder pair was allowed to interact for 5 minutes or until the intruder exhibited submissive supine posture three times. The intruder was then briefly removed, placed into a protective wire mesh cage, and returned to the residents cage until a total of 10 minutes elapsed from initial entry. Additional control rats were not exposed to social defeat stress. For the first experiment, the in truder rats were killed 10 minutes, 30 minutes, or 24 hours after the la st social defeat session and plasma

PAGE 10

x corticosterone and adrenocorticotropic horm one concentrations were assayed. For the second experiment, the previously-stressed ra ts were tested with a 15-minute Porsolt forced swim test. The intruder rats exhibited more freezing behavior and less exploratory locomotion across consecutive social defeat sessions resembling the loss of interest found in depressed patients. Exposure to social def eat stress produced signifi cant elevations in circulating hormones 10 minutes and 30 minutes after the session, when compared with concentrations in the control rats. The repeat edly stressed rats also exhibited higher basal concentrations of circulati ng corticosterone 24 hours late r, mimicking the augmented circulating hormones found in clinically depr essed patients. These results were evident after six daily exposures; however, basal horm one concentrations were not significantly elevated with the extended regimen of one social defeat session every 72 hours. Inconsistent thymus involution in the chronically stressed ra ts in both regimens suggests that a longer, more intense daily stress re gimen may be necessary to alter glandular masses. Exposure to chronic so cial defeat stress also produc ed a significan t increase in total immobility time during the forced swim test when compared with the immobility times for the rats that were exposed to a si ngle acute social defeat session and with the immobility times for the rats that were not exposed to social defeat stress, thus representing behavioral desp air in the chronically stressed animals. Overall, the behavioral, hormonal, and glandul ar alterations that occurred af ter repeated social defeat stress resemble some of the sympto ms of major depression in humans.

PAGE 11

1 CHAPTER 1 INTRODUCTION Annually, approximately 6.6% of the na tional adult population suffers from depression (Kessler et al., 2003). It is a wide-spread illness that interferes w ith the ability to eat, sleep, work, and enjoy formerly pleasur able activities. The economic impact of this devastating disorder is high, but the co st in human distress cannot be estimated. The 4th Edition of the Diagnostic and St atistical Manual of Mental Disorders (1994) defines depression according to the follo wing criteria, with at least five of the symptoms present on a daily basis for at leas t 2 weeks: depressed mood, loss of interest or pleasure, significant wei ght loss or gain, insomnia or hypersomnia, psychomotor agitation or retardation, fati gue, undue guilt and/or fee lings of worthlessness, indecisiveness, recurrent thoughts of death, and significant distress or impairment in social or occupational functioning. These sy mptoms must represent a change from the individuals normal level of interpersonal function. Major depression is also clinically characterized by altered hormonal function stemming from ongoing elevations in overall organism distress. It has been well established that chronic emotional stress pl ays a pivotal role in the genesis of many psychiatric disorders such as depression (for review, see Agid et al., 2000). Chronic stress weighs on the physiological systems that maintain homeostasis and produces changes in the operating limits of those hormonal systems. Allostatic load, or the strain from the elevated activity of systems under major stress, can predispose an animal to many psychiatric disorders, including depr ession (McEwen and Stellar, 1993). The

PAGE 12

2 concept of allostatic load suggests that there is a fixed state in which enduring environmental challenges are balanced by a hormonal response that is raised above normal, basal levels. Patients with majo r depression, regardless of age, show higher 24-hour average cortisol levels when compar ed to normal subjects (Linkowski et al., 1985). Additionally, depressed patients reach the nadir of their daily cortisol cycle two to three hours before control subjects (Pfohl et al., 1985) This imbalance between activation and recovery of the stress response is implicated in the inability to maintain homeostasis, thus leading to neuroendocri ne maladjustment and heightened risk for depression (De Kloet, 2003). Stressful stimuli are categorized into two de scriptive classes. Systemic stressors, such as exposure to heat or cold, present immediate threat s to somatic homeostasis while processive stressors emphasize higher leve l cognitive processing (Herman and Cullinan, 1997). Common processive stresso rs include instability in th e social hierarchy and loss of environmental control. In accordance with the emotional nature of processive stress, it is particularly implicated in a variety of psychiatric diso rders including depression (for review, see Anisman and Matheson, 2005). Inputs from the brainstem (if the stressor is systemic in nature) and cortical and limbic structures (if the stressor is processive in nature) converge at the paraventricular nucleus (PVN) of the hypothalamus where parv ocellular neurons project to the median eminence. From there, corticotrophin -releasing hormone (CRH) and arginine vasopressin (AVP) are released into th e hypophyseal portal circulation, stimulating adrenocorticotropic hormone (ACTH) release from the anterior pituitary corticotrope cells into the general circulat ion. The release of ACTH then stimulates the synthesis and

PAGE 13

3 release of glucocorticoids from the adrenal cortex. Cortisol is the main glucocorticoid in humans and corticosterone (CORT) is the ma in glucocorticoid in rats. The elevated circulating levels of glucocor ticoids decrease the further ac tivity of the HPA axis through exertion of negative feedback on neuros ecretory cells of the hypothalamus and corticotrope cells of the pituitar y (for review, see Whitnall, 1993). Altered regulation of hormona l activity in depressed pati ents is thought to be a result of increased activity of specific CRH-c ontaining neurons in the PVN. The average total number of such neurons is up to four times higher in depressed patients when compared to normal control subjects. Also, co-localization of AVP in CRH-expressing neurons has been indicated as an index for stress-activated neuronal activity. The average number of neurons co-expressing both CRH and AVP in depressed patients is up to three times higher than those for normal s ubjects. These results suggest that increased expression of CRHand AVP-containing neur ons in the PVN may cause at least a fraction of the collective symptomatology of depression (Raadsheer et al., 1994). When physically or emotionally stresse d, non-human animals endure physiological responses that lead to beha vioral and hormonal impairment which may be fundamentally similar to the impairment seen in human stress-induced psychopathology. Behavioral responses to stressful stimuli (including in creased drug-taking propensity, decreased performance in learning tasks, sleep distur bances, and unsocial behavior) have been observed in a variety of species (for review, see Amiel-Tison et al., 2004). Also, there is significant evolutionary homol ogy in stress-regulating pep tides, such as CRH and CRHrelated molecules (Chang and Hsu, 2004). Thus, the use of an animal model for

PAGE 14

4 processive stress that produces both behavioral and hormonal e ffects is a logical approach to the study of human st ress-induced disorders. The effects of chronic and acute processi ve stress have been studied in rats (Simpkiss and Devine, 2003). Experimentally nave rats were exposed to a chronic variable stress (CVS) regimen of twice daily stressors for fifteen days. The stressors included novel environment, switched cage mates, forced swim, light open field, intermittent white noise, and intermittent footshock, administered on a random intermittent schedule. The CVS regimen wa s unsuccessful in producing elevations in basal circulating concentrations of ACTH and CORT, or in hormonal response to an acute stressor. The rats showed a blunt ed ACTH response, but no altered CORT response. Another model for emotional stress, social defeat, has also produced significant elevations in circulating CO RT during and after acute and repeated exposure to the stressful stimulus (Sgoifo et al., 1996). The procedure, developed by Miczek (1979), is designed to model social stress. A male intruder rat is expo sed to social stress when it is placed into the home cage of a larger male resident rat. The resident rat exhibits dominant behavior toward the intruder rat by displaying assertive posture, standing over the intruder. The intruder submits by displaying supine posture, positioned beneath the resident. The further effects of repeated social defeat stress on behavioral and hormonal (HPA axis) responses have been studied in rats (Lopes and Devine, 2004). In a preliminary study, a repeatedly stressed group of intruder rats s howed significantly elevated circulating CORT concentrations 24 hours after their final so cial defeat session,

PAGE 15

5 when compared to the concentr ations in the unstressed contro l animals. This suggests an enduring change in the circadian regulation of the HPA axis following chronic social defeat. Overall, the effects of six days of so cial defeat stress exceed ed the results attained with the fifteen days of CVS previously described. The brain sites activated by social defeat have been studied using c-fos immunohistochemistry. The immediate-early gene (IEG) c-fos is expressed in many cells in the brain, but typically at very low basal levels. C-fos and other IEGs are intracellular signaling mechanisms that regulate gene transcription and expression of various neuropeptides and trophic molecules in res ponse to stress (for re view, see Sabban and Kvetnansky, 2001). Various stressful stim uli can initiate increased levels of c-fos mRNA expression, lasting for minutes to hours. For instance, social defeat produces elevated c-fos expression in limbic, limbic-associated, a nd brainstem sites in both hamsters and rats one hour after a single de feat session (Kollack-Walker et al., 1997; Martinez et al., 1998). These results point to the brain structur es that are important in the processing of emotionally stressful events. However, after re peated social defeat sessions in rats, the pattern of neuronal activity was modified, de spite the fact that intruder submissive behavior persisted across trials. C-fos mRNA expression endured for many of these limbic and brainstem nuclei, while other limbic and brainstem regions exhibited a decrease in the social defeat-induced c-fos mRNA expression. The Porsolt swim test, originally describe d by Porsolt and collea gues (1978), is the most commonly utilized behavioral test for screening antidepressant treatments in rats and has been used to evaluate the behavior al effects of stress exposure. Immobility during the inescapable swim is measured as an indicator of be havioral despair or

PAGE 16

6 depressive-like behavior. Re gimens of chronic stress or repeated administration of gluccocorticoids increase the amount of time an animal spends immobile, decrease the latency to immobility, and decrease the am ount of time engaged in active swimming (Molina et al., 1994; Johnson et al., 2006) Antidepressant drugs, on the other hand, ameliorate these effects. For example, de sipramine (a tricyclic antidepressant) and fluoxetine (a selective serotoni n reuptake inhibitor) reduce immobility and increase the amount of time a stressed animal will struggle to escape the forced swim (Molina et al., 1994; Lucki et al., 2001). Based on the results of investigations usi ng social defeat stress and the Porsolt swim test, we have begun to further charac terize the social defeat model of emotional stress in rats. We have evaluated the impact of both acute and re peated social defeat stress on regulation of ACTH and CORT at va rious times after the stressor. We have also compared six daily social defeat sessions to one month of stress every third day to examine the effectiveness of massed and inte rmittent stress exposure. In addition, we exposed socially defeated rats to the Pors olt swim test in order to characterize the enduring behavioral effects of social defeat stress.

PAGE 17

7 CHAPTER 2 METHODS Animals Ninety male Long-Evans (LE) rats and twel ve female LE rats were purchased from Harlan Co. of Indianapolis, IN. Twenty of the male Long Evans (LE) rats (225-250 g) were used as unhandled controls. Fifty-eight of the male LE rats (225-250 g) were used as intruders. Twelve of the male LE rats (300-325 g) were used as residents and the twelve female LE rats (225-250 g) were used as housing mates for the residents. The weight ranges indicate weights at the time the rats were purchased. The intruders weighed 275-325 g and the residents weighed 500800 g at the time of the experiments. Six of the unhandled control rats were used in Experiment 1a, six were used in Experiment 1b, and eight were used in Experime nt 2. Thirty of the intruder rats were used in Experiment 1a, twelve were used in Experiment 1b, and sixteen were used in Experiment 2. The resident males and the females were used in multiple experiments. The rats were housed in polycarbonat e cages (43 x 21.5 x 25.5 cm) with sexmatched and weight-matched pairings for five to seven days of acclimation to a 12hr/12hr light/dark cycle (light s on at 7:00a.m.). Standard chow (LabDiet 5001) and tap water were available ad libitum Temperature and humidity in the housing facility were controlled to 22.72C 0.94C and 54.1% 14.6%, respectively. The intruder rats were exposed to social defeat stress and remained pair-housed throughout the experiment. The control rats we re not exposed to social defeat stress and were pair-housed throughout the experiment. The resident rats were vasectomized,

PAGE 18

8 singly-housed, and given 10 days to recover fr om the surgery. The residents were then pair-housed in a separate housing room, each wi th a female, for 2 weeks prior to and then throughout the experiment. All procedures were pre-approved by the Institutional Animal Care and Use Committee at the Un iversity of Florida and conducted in accordance with the Guide for the Care and Use of Laboratory Animals. Surgical Procedures After a week of acclimation to the hous ing facility, the resident rats were anaesthetized with ketamine:xylazine (50 mg/kg:5 mg/kg, i.p.). Seda tion was verified by testing the pedal withdrawal reflex 10 mi n after injection and every 10 min during the procedure. After sedation was verified, each ra t was shaved from the rostral edge of the scrotum to the caudal abdomen. The area was then washed with Betadine three times. Surgery began with a 2 cm ventral midline incision, just above the scrotal area. Through the incision, the vas deferens was lo cated and externalized. A 0.25 cm section of each duct was removed with a micro-cautery knife. The muscle walls were sutured and the external incision was closed using st erilized staples that were removed ten days after the surgery. Each surgic al procedure lasted 15-30 min. Each rat was given a 1 mL injection of 0.9% warm saline and ketorolac subcutaneously and then placed into a post-ope rative Plexiglas cage, heated by an electric heating pad. Following observation of locomoti on and urination, each rat was returned to its individual home cage and was monito red daily throughout the recovery period. Experimental Procedures General Social Defeat Stress Procedure At the beginning of the 10-min social def eat session, the female rat paired with one resident rat was taken out of the cage and pl aced into an empty, identical cage. The

PAGE 19

9 intruder rat was then placed into the cage of the resident (direct interaction). The session was then closely monitored for submissive, supine posture expressed by the intruder. Each time the intruder exhibited a supine postu re with the resident physically contacting the intruder for 2 sec or more, one defeat was counted. After three defeats or a total of 5 min elapsed (whichever came first), the intr uder was taken out of the residents cage momentarily and quickly placed into a 10 x 10 x 15 cm double-walled protective wire mesh cage. The intruder (within the wire mesh cage) was then placed back into the residents cage for the remainder of the 10min session (indirect interaction). The two intruder rats in each housing pair were run in simultaneous defeat sessions in the cages of two resident males that were placed side by side. Each pair of cage-mates received the same treatment. Each social defeat sessi on was run between 8:00a .m. and 10:00a.m., and was videotaped for further analysis. Social Defeat Stress Regi men: Experiment 1a Thirty-six nave male LE rats were assigned to six experimental groups (Table 2-1). The control rats from Group 1 (n = 6) rema ined in their original cages throughout the experiment and were rapidly decapitated between 8:00a.m. and 10:00a.m. on the final experimental day. These rats were exposed to no chronic and no acute stress (NC/NA). The intruders from Group 2 (n = 6) were expos ed to social defeat stress once every 24 h across six experimental days and were kill ed by rapid decapitation between 8:00a.m. and 10:00a.m., 24 h after their final social defeat se ssion. These rats were exposed to chronic, but not acute stress (C/NA). The intruders fr om Group 3 (n = 6) were only exposed to social defeat stress on the final experiment al day and were then immediately decapitated, 10 min after the start of the social defeat sess ion (t = 10 min). These rats were exposed to no chronic, just acute stress (NC/A-10). Th e intruders from Group 4 (n = 6) were only

PAGE 20

10 exposed to social defeat stress on the final experimental day and were rapidly decapitated 30 min after the start of the social defeat sess ion (t = 30 min). These rats were exposed to no chronic, just acute stress (NC/A-30). The intruders from Group 5 (n = 6) were exposed to social defeat stress once every 24 h over the course of six experimental days and were killed by rapid decapitation 10 min af ter the start of the social defeat session on the sixth day (t = 10 min). These rats were exposed to chronic and acute stress (C/A-10). The intruders from Group 6 (n = 6) were expos ed to social defeat stress once every 24 h over the course of six experimental days and were killed by rapid decapitation 30 min after the start of the social de feat session on the sixth day (t = 30 min). These rats were exposed to chronic and acute st ress (C/A-30). Each intruder fr om the repeatedly stressed groups was placed in the cage of a different resident each day of his chronic stress routine. Social Defeat Stress Regi men: Experiment 1b Eighteen nave male LE rats were assigne d to three experimental groups (Table 2-2). The unhandled control rats from Group 1 (n = 6) remained in their original cages throughout the experiment and were killed by rapid decapitation between 8:00a.m. and 10:00a.m. on the final experimental day. Th ese rats were exposed to no chronic and no acute stress (NC/NA). The intrud ers from Group 2 (n = 6) were exposed to social defeat stress once every 72 h across thirty-four expe rimental days and were rapidly decapitated between 8:00a.m. and 10:00a.m., 24 h after thei r final social defeat session. These rats were exposed to chronic, but not acute stress (C/NA). The intruders from Group 3 (n = 6) were exposed to social defeat stress on ce every 72 h across thirty-four experimental days and were killed by rapid decapitation 30 min after the start of the social defeat session on the final day (t = 30 min). These ra ts were exposed to ch ronic and acute stress

PAGE 21

11 (C/A). All intruder rats from Experiment 1b were exposed to a total of twelve social defeat sessions. Each intruder from the repe atedly stressed groups was placed in the cage of a different resident each da y of his chronic stress routine. Social Defeat Stress and Porsolt Swim Test: Experiment 2 Twenty-four nave male LE rats were assi gned to three experimental groups (Table 2-3). The unhandled control rats from Group 1 (n = 8) remained in their original cages throughout the social defeat portion of the expe riment. These rats were exposed to no chronic and no acute social de feat stress (NC/NA). The intr uders from Group 2 (n = 8) were only exposed to social defeat stress on th e final day of the social defeat portion of the experiment. These rats were exposed to no chronic, just acut e stress (NC/A). The intruders from Group 3 (n = 8) were repeatedly exposed to social defeat stress over the course of five days, once every 24 h. These ra ts were exposed to ch ronic and acute stress (C/A). Each repeatedly stressed intruder wa s placed in the cage of a different resident each day of his chronic stress routine. All of the rats from each of the three groups were then exposed to the Porsolt swim test. Twenty-four hours after the final social defeat session, each rat from Experiment 2 was individually removed from its home cag e and placed into a plastic cylinder filled with approximately 25 cm of clean tap water at 24-27 C. The depth of water allowed each rat to reach the bottom of the cylinder with its tail, wi th enough head room that the rat was unable to escape from the tank. The water in each cylinder was changed between trials. Each rat was subjected, one at a tim e, to a 15-min swim session then carefully dried and returned to its home cage. Each Porsolt swim session was run between 8:00a.m. and 10:00a.m., and was vi deotaped for further analysis.

PAGE 22

12 Behavioral Assays Number of defeats, latency to first de feat, freezing behavior, and exploratory locomotion were scored from the recorded soci al defeat sessions. Freezing behavior was recorded whenever the intruder remained motionl ess for at least 2 sec, and is reported as percent time during the direct interaction between the intruder and the resident. Exploratory locomotion was recorded every time the intruder crossed with all four paws from one third of the cage into the adjacent th ird of the cage. Vertical lines were drawn on the video image of each residents home cage and exploratory locomotion was scored from the videotapes. Exploratory locomotion is reported as lines crossed per minute of direct interaction time. Two tr ained observers scored the soci al defeat stress videos from Experiments 1a and 1b and inter-observer reliability was assessed. Immobility was recorded during the Porsolt swim tests whenever the rat balanced on its tail, completely motionless, or exhi bited only slight forepaw movement for a minimum of 2 sec. This behavior is repor ted as total time spent immobile during the swim test. Two trained observers scored th e swim test videos from Experiment 2 and inter-observer reliability was assessed. Histological Assays Immediately after decapitation in experi ments 1a and 1b, 6 mL of trunk blood from each rat was collected in pol ypropylene tubes on ice with 600 l of Na2EDTA at 20 g/ L. The blood samples were then immedi ately centrifuged at 1000 rcf for 5 min and the plasma fraction frozen in 300 l aliquots at C. The brain from each intruder was quickly removed, frozen in 2-methylbutane at -40C, and stored at -80C for future analysis of molecular variable s involved in HPA axis function. The thymus and adrenal glands were removed from each rat and frozen separately at 80C. These glands were

PAGE 23

13 later weighed to determine the health and st ress status of each rat at time of death. Plasma ACTH and CORT concentrations fr om the rats were later analyzed with radioimmunoassay (RIA). ACTH RIAs were run with kits from Alpco Diagnostics (Salem, NH) and CORT RIAs were run with kits from Diagnostic Products Corporation (Los Angeles, CA). Statistical Analyses Potential between-groups differences in number of defeats and latency to first defeat were analyzed using two one-way anal yses of variance (ANOVAs) to compare the five groups of intruders (i.e. three repeatedly stre ssed, and two just acutely stressed) in Experiment 1a during their firs t exposure to social defeat stress. A 3x6 (group x session) repeated-measures ANOVA (for Experiment 1a ) and a 2x12 (group x session) repeatedmeasures ANOVA (for Experiment 1b) were used to examine potential differences in number of defeats between groups and across experimental sessions for the repeatedly stressed groups. A 3x6 (group x session) repeated-measures ANOVA (for Experiment 1a) and a 2x12 (group x session ) repeated-measures ANOVA (for Experiment 1b) were used to examine potential differences in late ncy to first defeat between groups and across experimental sessions for th e repeatedly stressed groups. Potential between-groups differences in freezing responses were analyzed using a one-way ANOVA to compare freezing behavior between the five groups in Experiment 1a during their first exposure to social defeat stress. A 3x6 (group x session) repeatedmeasures ANOVA (for Experiment 1a) and a 2x12 (group x session ) repeated-measures ANOVA (for Experiment 1b) were used to examine potential freezing behavior differences between groups and across experime ntal sessions for the repeatedly stressed groups.

PAGE 24

14 Potential between-groups differences in exploratory locomotion were analyzed using a one-way analysis of variance ( ANOVA) to compare exploratory locomotion between the five groups in Experiment 1a dur ing their first exposur e to social defeat stress. A 3x6 (group x session) repeated-m easures ANOVA (for Experiment 1a) and a 2x12 (group x session) repeated-measures ANOVA (for Experiment 1b) were used to examine potential exploratory locomotion differences between groups and across experimental sessions for th e repeatedly stressed groups. Potential between-groups differences in plasma ACTH concentrations and in plasma CORT concentrations were anal yzed with one-way ANOVAs in each of Experiments 1a and 1b. Also, potential betw een-groups differences in adrenal weights and in thymus weights were analyzed with one-way ANOVAs in each of Experiments 1a and 1b. The results were further analyzed using pre-planned Newman-Keuls multiple comparison tests for all significant ANOVAs. Potential between-groups differences in immobility duration scores for the 15-min Porsolt swim test session in Experiment 2 were analyzed using a one-way ANOVA. The results were further analyzed using Newman-K euls multiple comparison tests. The data were then analyzed in 5-min bins using a 3x3 (group x bin) repeated-measures ANOVA. The results were further analyzed using Newman-Keuls multiple comparison tests, making pair-wise comparisons between each stressed group and the control group, for each bin. Inter-observer agreement was assessed for freezing behavior in Experiments 1a and 1b by comparing total times recorded by each trained observer for each social defeat session. Inter-observer agreement was also assessed for exploratory locomotion in

PAGE 25

15 Experiments 1a and 1b by comparing total number of line crossings recorded by each trained observer for each social defeat se ssion. Inter-observer agreement was assessed for immobility duration scores for the Porso lt swim test from Experiment 2 by comparing total immobility times recorded by each trained observer for each test session.

PAGE 26

16 Table 2-1. Experiment 1a, group assignments Group Repeated Stress Acute Stress Kill Time NC/NA C/NA X t = 24 h NC/A-10 X t = 10 min NC/A-30 X t = 30 min C/A-10 X X t = 10 min C/A-30 X X t = 30 min Table 2-2. Experiment 1b, group assignments Group Repeated Stress Acute Stress Kill Time NC/NA C/NA X t = 24 h C/A X X t = 30 min Table 2-3. Experiment 2, group assignments Group Repeated Stress Acute Stress Porsolt Swim Test NC/NA X NC/A X X C/A X X X

PAGE 27

17 CHAPTER 3 RESULTS Experiment 1a: Six Daily Sess ions of Social Defeat Stress During their first exposure to social defeat stress, the ra ts from all five of the groups in Experiment 1a did not show statisti cally different numbers of defeats (F (4, 25) = 0.2419, p < 0.9118; Fig. 3-1A) nor latencies to first defeat (F (4, 25) = 0.7308, p < 0.5795; Fig. 3-1B). Each of the three repeatedly stressed groups (C/NA, C/A-10, and C/A-30), showed no significant between-group s difference, time effect, or group by time interaction effect in numbe r of defeats (F (2, 15) = 1.603, p < 0.2340; F (5, 15) = 0.3463, p < 0.8831; F (10, 15) = 0.3139, p < 0.9753; Fig. 3-1A) nor latencies to first defeat across the six experimental sessions (F ( 2, 15) = 1.571, p < 0.2401; F (5, 15) = 0.2106, p < 0.9570; F (10, 15) = 0.5255, p < 0.8669; Fig. 3-1B). During their first exposure to social defeat stress, the ra ts from all five of the groups did not show statistically different amounts of freezing behavior (F (4, 25) = 0.1757, p < 0.8405; Fig. 3-2A). Each of the th ree repeatedly stressed groups showed significantly increased freezing behavior across the six experimental sessions (F (5, 75) = 17.08, p < 0.0001; Fig. 32A), reaching asymptote by day 3, with no significant betweengroups differences or group by time interaction effect. During their first exposure to social defeat stress, the ra ts from all five groups did not show statistically different amounts of exploratory locomotion (F (4, 25) = 0.5711, p < 0.5767; Fig. 3-2B). Each of the three repeat edly stressed groups showed significantly decreased exploratory locomotion across the si x experimental sessions (F (5, 75) = 11.76,

PAGE 28

18 p < 0.0001; Fig. 3-2B), reaching asymptote by da y 4, with no significant between-groups differences or group by time interaction effect. Exposure to acute social de feat (NC/A and C/A) produced significant elevations in circulating ACTH concentrations when compar ed with the ACTH concentrations in the control rats (NC/NA) (F (5, 30) = 3.874, p < 0.0079; Fig. 3-3A). Circulating ACTH concentrations were not elevated in the chroni cally, but not acutely, st ressed rats (C/NA). Exposure to social defeat stress for al l groups (including C/NA) produced significant elevations in circulating CORT concentrations when compared with the CORT concentrations in the control rats (F (5, 30) = 20.79, p < 0.0001; Fig. 3-3B). Exposure to social defeat significantly d ecreased thymus masses in Groups C/A-10 and C/A-30, but not for the other groups (F (2, 15) = 5.651, p < 0.0148; Fig. 3-4A). There were no significant diffe rences in adrenal masses betw een the rats in the stressed groups and the rats in the control gr oup (F (5, 30) = 0.9299, p < 0.4756; Fig. 3-4B).

PAGE 29

19 1 2 3 4 5 6 0 1 2 3 C/NA NC/A-10 NC/A-30 C/A-10 C/A-30 Experimental DayNumber of Defeats 1 2 3 4 5 6 0 25 50 75 100 125 C/NA NC/A-10 NC/A-30 C/A-10 C/A-30 Experimental DayLatency to First Defeat (sec)A B Figure 3-1. Social defeats per daily session. The rats exposed to repeated social defeat stress (C/NA, C/A-10, C/A-30) exhibite d equivalent (A) number of defeats per session and (B) latency to firs t defeat per sess ion across the 6 experimental sessions. The rats that received only one acu te defeat session (NCA-10 and NCA-30) were stressed at the same time as the final stress session for the chronically stressed groups, and so the values for these groups are illustrated on day 6. Values expre ssed are group means SEM (n = 6 rats per group). Abbreviations: C/NA = ch ronic stress/no acute stress, NC/A = no chronic stress/acute stress, C/A = chronic stress/acute stress.

PAGE 30

20 0 1 2 3 4 5 6 0 10 20 30 40 50 60 70 80 90 100 C/NA NC/A-10 NC/A-30 C/A-10 C/A-30 Experimental DayTotal Time Freezing (% direct interaction time) 0 1 2 3 4 5 6 0 1 2 3 4 5 6 7 8 9 10 11 C/NA NC/A-10 NC/A-30 C/A-10 C/A-30 Experimental DayLine Crossings per MinuteA B Figure 3-2. Intruder behavior during daily soci al defeat sessions. The rats exposed to repeated social defeat stress exhibited (A) increases in fr eezing behavior and (B) decreases in exploratory locomotion across the 6 experimental sessions. The rats that received only one acute de feat session were stressed at the same time as the final stress session for the ch ronically stressed groups, and so the values for these groups are illustrate d on day 6. Values expressed are group means SEM (n = 6 rats per group).

PAGE 31

21 NC /NA C/N A N C/A-1 0 NC / A -3 0 C/A 10 C/A-3 0 0 50 100 150 200* * n.s. n.s.n.s. n.s.ACTH Concentration (pg/ml) NC/NA C / N A N C /A-10 N C /A-30 C / A -1 0 C / A -3 0 0 100 200 300 400 500* * * ++ +n.s.CORT Concentration (ng/ml)A B Figure 3-3. Circulating hormones after daily social defeat st ress. The rats exposed to social defeat stress had elevated circ ulating concentrations of (A) ACTH and (B) CORT when compared to basal con centrations in control rats. Values expressed are group means SEM (n = 6 rats per group). Significant differences between the socially defeat ed rats and the unhandled control rats (NC/NA) are expressed as [* ] p < 0.05. Significant di fferences in pre-planned comparisons between the stressed groups of rats are expressed as [+] p < 0.05, with lines connecting the groups th at were compared. Additional abbreviation: NC/NA = no chr onic stress/no acute stress.

PAGE 32

22 N C /N A C/ N A N C /A1 0 N C /A3 0 C /A1 0 C /A3 0 0 100 200 300 400 500 600* *Thymus Weight (mg) NC/ NA C/ NA NC/A-10 NC/A-30 C/ A1 0 C/ A3 0 0 10 20 30 40 50 60Adrenal Weight (mg)A B Figure 3-4. Effects of daily soci al defeat stress on glandular masses. Social defeat stress (A) significantly decreased thymus masse s for 2 of the repeatedly stressed groups (C/A-10 and C/A-30), but (B) did not significantly affect adrenal gland masses for any group. Values expressed are group means SEM (n = 6 rats per group). Significant differences between the socially defeated rats and the control rats are expressed as [*] p < 0.05.

PAGE 33

23 Experiment 1b: Twelve Intermittent Sessions of Social Defeat Stress There were no significant between-groups differences, time effect, or group by time interaction effect in num ber of defeats (F (1, 10) = 0.0787, p < 0.7847; F (11, 10) = 0.9032, p < 0.5399; F (11, 10) = 0.8203, p < 0.6199; Fig. 3-5A) nor latency to first defeat (F (1, 10) = 0.4653, p < 0.5107; F (11, 10) = 0.8400, p < 0.6007; F (11, 10) = 0.3414, p < 0.9743; Fig. 3-5B) for the repeatedly stressed groups (C/NA and C/A) in Experiment 1b. Freezing behavior significantly increased for both of the repeatedly stressed groups across the twelve experimental sessions (F (11, 10) = 6.224, p < 0.0001; Fig. 3-6A), with no significant between-groups differences or group by time interaction effect. Exploratory locomotion signifi cantly decreased for both of the repeatedly stressed groups across the twelve experimental se ssions (F (11, 10) = 7.252, p < 0.0001; Fig. 3-6B), with no significant between-groups differences or group by time interaction effect. Acute exposure to social defeat stress pr oduced significant elevations in circulating ACTH concentrations when compared with th e basal ACTH concentra tions in the control rats (F (2, 15) = 4.525, p < 0.0290; Fig. 3-7A). Acut e exposure to social defeat stress also produced significant elevations in circulating CORT concentrations when compared with the basal CORT concentrati ons in the control rats (F (2, 15) = 5.853, p < 0.0132; Fig. 3-7B). These elevations in circulating hormones were limite d to the rats that were stressed acutely before termination (C/A) the basal concentrations in the other chronically stressed rats (C/ NA) did not significantly differ from the basal concentrations in the control rats (NC/NA). There were no significant differences in thymus masses (F (2, 15) = 0.04457, p < 0.9565; Fig. 3-8A) or adrenal masses (F ( 2, 15) = 1.826, p < 0.1951; Fig. 3-8B) between the rats in the socially defeated groups and the rats in the control group.

PAGE 34

24 0 1 2 3 4 5 6 7 8 9 10 11 12 0 1 2 3 C/NA C/A Experimental DayNumber of Defeats 1 2 3 4 5 6 7 8 9 10 11 12 0 25 50 75 100 125 C/NA C/A Experimental DayLatency to First Defeat (sec)A B Figure 3-5. Social defeats per intermittent session. The rats exposed to repeated social defeat stress (C/NA and C/A) exhibite d equivalent (A) number of defeats per session and (B) latency to first defeat per session across the 12 experimental sessions. Values expressed are group means SEM (n = 6 rats per group).

PAGE 35

25 0 1 2 3 4 5 6 7 8 9 10 11 12 0 10 20 30 40 50 60 70 80 90 100 C/NA C/A Defeat SessionTotal Time Freezing (% direct interaction time) 0 1 2 3 4 5 6 7 8 9 10 11 12 0 1 2 3 4 5 6 7 8 9 10 11 C/NA C/A Defeat SessionLine Crossings per MinuteA B Figure 3-6. Intruder behaviors during intermittent social defeat sessions. The rats exposed to repeated social defeat stress exhibited ( A ) increases in freezing behavior and ( B ) decreases in exploratory lo comotion across 12 experimental sessions. Values expressed are group means SEM (n = 6 rats per group).

PAGE 36

26 NC/ NA C/NA C /A 0 50 100 150 200*ACTH Concentration (pg/ml) NC/ NA C/ NA C/ A 0 100 200 300 400 500*CORT Concentration (ng/ml)A B Figure 3-7. Circulating hormones after interm ittent social defeat stress. Immediately after, but not 24 h later, th e rats exposed to repeated social defeat stress had elevated circulating concentrations of (A) ACTH and (B) CORT when compared to basal concentrations in the control rats (NC/NA). Values expressed are group means SEM (n = 6 rats per group). Significant differences between the socially defeat ed rats and the unhandled control rats are expressed as [*] p < 0.05. NC/NA C /N A C/A 0 100 200 300 400 500 600Thymus Weight (mg) NC/NA C / NA C / A 0 10 20 30 40 50 60Adrenal Weight (mg)A B Figure 3-8. Effects of intermittent social de feat stress on glandular masses. Repeated social defeat stress had no effect on (A) thymus masses or (B) adrenal gland masses. Values expressed are group m eans SEM (n = 6 rats per group).

PAGE 37

27 Experiment 2: Social Defeat Stre ss Followed by Porsolt Swim Testing Exposure to chronic social de feat stress (C/A) produced a significant increase in total immobility time during the 15-min forced swim when compared with the immobility times for the control rats that were not exposed to social defe at stress (NC/NA) (F (2, 21) = 5.363, p < 0.0131; Fig. 3-9). When the data we re analyzed in 5-min bins, there was a significant between-groups effect as well as a significant tim e effect (F (2, 42) = 2.424, p < 0.0150; F (2, 42) = 72.74, p < 0.0001; Fig. 3-10), but no significant group by time interaction effect (F ( 4, 42) = 1.270, p < 0.2970; Fig. 3-10). Inter-Observer Reliability The two observer's recordings of total time freezing for the social defeat sessions from Experiments 1a and 1b combined differed by less than 20 sec for 94% of the sessions and never differed by more than 28 sec. The two observer's recordings of exploratory locomotion during social defeat sessions were identical in 70% of the sessions and never differed by more than 3 lines crossed. The two observer's recordings of total immobility time during the Porsolt swim test differed by less than 40 sec in 92% of the sessions and never differed by more than 43 sec.

PAGE 38

28 NC/NANC/AC/A 0 100 200 300 400 500 600 700 800 900*Total Immobility Time (sec) Figure 3-9. Immobility during the Porsolt swim test. The rats exposed to repeated social defeats stress (Chronic) e xhibited increased immobility for the duration of the Porsolt forced swim test, when comp ared to the immobility times for the intruder rats exposed to acute social de feat stress (Acute) a nd the control rats (No SD). Values expressed are group means SEM (n = 8 rats per group). Significant differences between the soci ally defeated rats and the unhandled control rats are expressed as [*] p < 0.05. Additional abbreviation: No SD = no social defeat stress. 1 2 3 0 100 200 300 NC/NA NC/A C/A * *+Total Immobility Time (sec) Figure 3-10. Immobility measured in 5-minute bins during the Porsolt swim test. There were significant main effects for time and between-groups difference when comparing the socially defeated groups (Acute and Chronic) to the control group (No SD). Values expressed ar e group means SEM (n = 8 rats per group). Significant differences between the chronically def eated rats and the controls are expressed as [*] p < 0.05. Significant differences between the acutely defeated rats and the controls are expressed as [+] p < 0.05.

PAGE 39

30 CHAPTER 4 DISCUSSION Overall, repeated social de feat stress produced behavioral, hormonal, and glandular changes that model some of th e symptoms that are seen in depressed humans. Behavioral despair, HPA axis dysregulation, and thymus involution were found in rats that were exposed to daily social stress; and these effect s were greater overall than the effects that were seen in the rats that were stressed intermittently. Daily Social Defeat Stress The equivalent mean freezing scores and the equivalent mean locomotion scores during the first social defeat se ssion indicate that all five gr oups that experienced social defeat stress were comparable at the be ginning of the experiment. The equivalent progressive increases in freezing behavior a nd the equivalent decreases in exploratory locomotion for the three repeatedly stressed groups indicate that these groups underwent comparable stress-induced behavioral change s across social defeat sessions. These behaviors, however, did not diminish the pe rsistent displays of dominant behavior (consistent number of defeats and equivalent la tencies to first defeat) by the resident rats. After 3-4 days of stress exposure, the behavi oral curves for both freezing behavior and exploratory locomotion reached asymptote, suggesting that the behavioral changes associated with daily social defeat stress do not progress beyond the first few exposures. The equivalent mean ACTH concentra tions and the equivalent mean CORT concentrations 10 min after the onset of st ress between the groups with and without a history of stress suggest that there is simila r activation of the HPA axis for these groups.

PAGE 40

31 However, the lower CORT concentrations 30 min after the onset of stress in the repeatedly stressed group compared to the CO RT concentrations in the acutely stressed group suggests a sensitization in the ne gative feedback regulation of CORT. Decreased thymus mass and/or increased adrenal mass has been shown to result from chronic stress regimens involving physic al and/or psychologi cal stressors (for examples, see Blanchard et al., 1998; Bryant et al., 1991; Simpkiss and Devine, 2003) as well as from major depression in victims of suicide (Szigethy et al ., 1994). Given that two of the three groups that were stressed daily (C/A-10 and C/A-30) exhibited thymus involution while one (C/NA) did not, it cannot be concluded that the daily stress regimen consistently caused a decrease in thymus ma ss. These findings, along with adrenal mass equivalency across groups regard less of number of stress e xposures, suggest that the chronic social defeat stress regimen of six da ily exposures may not have been adequate to consistently alter glandular masses. A l onger or unpredictable social defeat stress regimen could possibly generate consistent thymus involution and adrenal hypertrophy. Perhaps an additional stressor for the intruders, such as isolation housing to ensure more timid behavior (Kabbaj et al., 2000), could cont ribute to the effectiven ess of the regimen. Also, using a restraint stress harness instead of the prot ective wire mesh cage could elevate the efficiency of the regimen by producing an inescapable condition for the intruders. Intermittent Social Defeat Stress The equivalent progressive increases in freezing behavior and the equivalent progressive decreases in exploratory loco motion in the intermittent stress regimen indicate that the repeated ly stressed groups underwent comparable stress-induced behavioral changes. However, the escala tion of freezing behavior and the decline of

PAGE 41

32 exploratory locomotion occurred more gradua lly when the social defeat sessions were 72 hours apart than when the social defeat sess ions occurred daily. This suggests that the intermittent stress regimen was not as effici ent at inducing these behavioral changes, although the overall magnitudes of the beha vioral effects were eventually roughly equivalent. Once again, the intruders' behavioral adapta tions did not decrease the displays of agonistic behavior from the re sidents (consistent nu mber of defeats and equivalent latencies to first defeat across groups and days). The equivalence in basal CORT concentr ations as well as the equivalence in glandular masses between the chronically st ressed animals and the control animals indicates that the temporally spaced regime n was not potent enough to sufficiently alter hormonal or glandular basal states. Even though the more temporally spaced chronic stress regimen was able to pr oduce equivalent effects with re gards to behavioral changes (albeit, more slowly), it was less effec tive than the daily regimen in producing the hormonal and glandular changes that correlate with some of the symptoms of major depression. The results indicate, therefore, that there may be a critical window for vulnerability to an additional stressorit is possible that the rats were able to partially recover from the initia l stressor by the next exposure in the more temporally spaced chronic regimen. Relevance to Previous Work Overall, these results confirm and extend a previous report that daily social defeat is an effective emotional stressor for male rats. In a study by Haller and colleagues (1999), intruder rats were exposed to reside nt rats for 4 hours on four consecutive days, producing increased basal CORT concentrations in the intruders. The daily social defeat stress regimen used in our study effectivel y increased basal CORT concentrations as

PAGE 42

33 well, with briefer stressful interactions than previously reported. However, in the same study, Haller and colleagues found increased ad renal mass with no decrease in thymus mass, results opposite to those seen after our similar daily chr onic stress regimen. Perhaps it is necessary for a daily chronic stre ss regimen to extend past four (or six) days to generate reliable glandular changes. The HPA axis response to an acute stre ssor largely depends upon whether the acute stressor is heterotypic or homotypic. In a study by Armario and colleagues (2004), investigators reported that when previously -stressed rats (via immobilization) were presented with a heterotypic stressor (for ced swimming), a minor sensitization was observed. Also, following exposure to a se vere stressor (such as shock, restraint, immobilization, or large doses of endotoxin) there was habituation of the HPA axis response to a homotypic stressor. These resu lts differ from the results obtained after daily social defeat stress. Fo llowing our repeated daily social defeat stress regimen, there was a robust activation but rapid shutdown of CORT release. Most probably, the difference is due to the fact that Armari o and colleagues utiliz ed a single stressor followed by either homotypic or heterotypic chal lenges. Our model of social defeat does not appear to function as a homotypic st ress regimen. There are several variables involved in the direct interacti on phase of social defeat stre ss (physical contact, olfactory cues, and ultra-sonic vocalizations) as well as the introduction of a di fferent resident rat with each exposure. Conceivably, the use of a different resident for each stress exposure and the potentially different threats imposed by these different reside nts creates a novel, therefore heterotypic, situation with each session.

PAGE 43

34 In general, the behavioral changes in the intruder rats after rep eated social defeat stress model some of the symptoms expresse d in clinically depre ssed patients, including impairment in social or occupational func tioning and loss of interest (DSM IV, 1994). Also, the elevated basal CORT concentrations after daily social de feat stress during the nadir of the daily cycle c onfirm previous findings from our laboratory (Lopes and Devine, 2004) and mimic the augmented ci rculating hormones found in clinically depressed patients (Linkowski et al., 1985; Pfohl et al., 1985). Porsolt Swim Test Based on the combined results from the daily and the intermittent stress regimens, a daily social defeat stress regimen was us ed to for the Porsolt study to optimize the potential to produce behavi oral despair. The fact that ac ute social defeat stress did not produce as much immobility as repeated st ress indicates that th e neuronal plasticity caused by repeated exposure to social stress (demonstrated through hormonal and behavioral plasticity) is a determining factor in the expression of behavioral despair. A 5-min swim test is commonly used for det ection of behavioral despair (Gavioli et al., 2003; Hinojosa et al., 2006; Porsolt et al ., 1978; Rygula et al., 2005). Interestingly, when analyzed in 5-min bins, the data suggest that in all the bins th e repeatedly stressed rats exhibited more immobility than the c ontrols did, and in one bin (5-10 min) the acutely defeated rats also exhibited more immo bility than the contro ls did. Therefore, a 10-min session may be necessary to closely examine the subtle differences between groups that have a less severe stress history and non-stre ssed control rats. Overall, these results model the behavior al despair, or low mood and anhedonia, found in clinically depresse d patients (for review, s ee Harrison, 2002). Also, the significantly elevated immobility times for the repeatedly stressed rats confirm and

PAGE 44

35 extend the results from a previous report. In a study by Rygula a nd colleagues (2005), investigators used a more extensive stress re gimen of hour-long daily sessions of social defeat for five straight weeks to elicit be havioral despair in s ubsequent Porsolt swim testing. The daily social def eat stress regimen used in our study effectively increased immobility in the repeatedly stressed rats as we ll, with briefer stressful interactions and a shorter regimen than previously reported. Conclusions and Future Directions In conclusion, the behavioral, hormona l, and glandular changes produced by repeated social defeat closely resemble many of the psychopathol ogical symptoms in patients with major depression. The daily so cial defeat stress regimen provides an interesting and effec tive model for stressinduced psychopathology. The effect of repeated social defeat on the overall circadian rhythm was not evaluated, nor was the persistence of the altered regulation across days, weeks, or months. Both of these topics may be interes ting issues for future studies. Also, potential adaptations in stress-related molecules (CRH and AVP in the hypothalamus, CRH1 and V1b in the pituitary and amygdala, and minera locorticoid receptor and gluccocorticoid receptor in the hippocampus) after social def eat should be investigated. These studies will enhance our growing knowledge of the neurobiological basis for stress-induced psychopathology (e.g., major depression).

PAGE 45

36 LIST OF REFERENCES Agid O, Kohn Y, Lerer B (2000). Environm ental stress and psychiatric illness. Biomedicine and Pharmacotherapy, 54: 135-141. Amiel-Tison C, Cabrol D, Denver R, Jarreau P, Papiernik E, Piazza PV (2004). Fetal adaptation to stress part II. Evolutio nary aspects; stress-induced hippocampal damage; long-term effects on behavior; consequences on adult health. Early Human Development, 78: 81-94. Anisman H, Matheson K (2005). Stress, depr ession, and anhedonia: Caveats concerning animal models. Neuroscience and Biobehavioral Reviews 29: 525-546. Armario A, Valles A, Dal-Zotto S, Marqu ez C, Belda X (2004). A single exposure to severe stressors causes long-term desensit ization of the physio logical response to the homotypic stressor. Stress 7 (3): 157-172. Blanchard RJ, Nikulina JN, Sakai RR, McKittri ck C, McEwen B, Blanchard DC (1998). Behavioral and endocrine change fo llowing chronic predatory stress. Physiology and Behavior 63 (4): 561-569. Bryant HU, Bernton EW, Kenner JR, Holaday JW (1991). Role of adrenal cortical activation in the immunos uppressive effects of chr onic morphine treatment. Endocrinology, 128 (6): 3253-3258. Chang CL, Hsu SYT (2004). Ancient evolu tion of stress-regulating peptides in vertebrates. Peptides 25: 1681-1688. De Kloet ER (2003). Hormones, brain and stress. Endocrine Regulations, 37: 51-68. Diagnostic and statistical manua l of mental disorders, 4th ed. (1994). Washington, DC: American Psychiatry Association. Gavioli EC, Marzola G, Guerri ni R, Bertorelli R, Zucchin i S, De Lima TCM, Rae GA, Salvadori S, Regoli D, Calo G (2003). Blockade of nocicep tin/orphanin FQ-NOP receptor signaling produces antidepressant -like effects: Pharmacological and genetic evidences from the m ouse forced swimming test. European Journal of Neuroscience 17: 1987-1990. Haller J, Fuchs E, Halasz J, Makara GB (1999). Defeat is a major stressor in males while social instability is mainly stressful for females: Towards the development of a social stress model in female rats. Brain Research Bulletin, 50 (1): 33-39.

PAGE 46

37 Harrison PJ (2002). The neuropath ology of primary mood disorder. Brain, 125: 14281449. Herman JP, Cullinan WE (1997). Neurocircuit ry of stress: Central control of the hypothalamo-pituitary-adrenocortical axis. Trends in Neuroscience 20 (2): 78-84. Hinojosa FR, Spricigo L, Izidio G, Bruske GR, Lopes DM, Ramos A (2006). Evaluation of two genetic animal models in behavi oral tests of anxiet y and depression. Behavioural Brain Research, 168: 127-136. Johnson SA, Fournier NM, Kalynchuk LE ( 2006). Effect of different doses of corticosterone on depressive-like behavi or and HPA axis responses to a novel stressor. Behavioural Brain Research 168: 280-288. Kabbaj M, Devine DP, Savage VR, Akil H (2000). Neurobiological correlates of individual differences in novelty-seeking behavior in the rat: Differential expression of stress-related molecules. The Journal of Neuroscience 20 (18): 6983-6988. Kessler RC, Berglund P, Demler O, Jin R, Ko retz Z, Merikangas KR, Rush AJ, Walters EE, Wang PS (2003). The epidemiology of major depressive disorder. Journal of the American Medical Association 289 (23): 3095-3105. Kollack-Walker S, Watson SJ, Akil H (1997). Soci al stress in hamsters: Defeat activates specific neurocircuits within the brain. The Journal of Neuroscience 17 (22): 8842-8855. Linkowski P, Mendlewicz J, Leclercq R, Brasseur M, Hubain P, Golstein J, Copinschi G, Van Cauter E (1985). The 24hour profile of adrenocorti cotropin and cortisol in major depressive illness. Journal of Clinical E ndocrinology and Metabolism 61 (3): 429-438. Lopes KO, Devine DP (2004). Elevation of HPA axis activity in response to repeated social defeat stress in rats. Society for Neuroscience Abstracts 30: 426.12. Lucki I, Dalvi A, Mayorga AJ (2001). Sens itivity to the effect s of pharmacologically selective antidepressants in different strains of mice. Psychopharmacology 155: 315-322. Martinez M, Phillips PJ, Herbert J (1998). Adaptation in patterns of c-fos expression in the brain associated with exposure to either single or repeated social stress in male rats. European Journal of Neuroscience 10: 20-33. McEwen BS, Stellar E (1993). Stress and the individual. Archives of Internal Medicine 153: 2093-2101.

PAGE 47

38 Miczek KA (1979). A new test for aggression in rats without aver sive stimulation: Differential effects of d -amphetamine and cocaine. Psychopharmacology 60: 253-259. Molina VA, Heyser CJ, Spear LP (1994). Chronic variable stress enhances the stimulatory action of a low dose of mo rphine: Reversal by desipramine. European Journal of Pharmacology 260: 57-64. Pfohl B, Sherman B, Schlechte J, Stone R (1985). Pituitary-ad renal axis rhythm disturbances in psychiatric depression. Archives of General Psychiatry 42: 897903. Porsolt RD, Anton G, Blavet N, Jalfre M ( 1978). Behavioural desp air in rats: A new model sensitive to antidepressant treatments. European Journal of Pharmacology 47: 379-391. Raadsheer FC, Hoogendijk WJG, Stam FC, T ilders FJH, Swaab DF (1994). Increased numbers of corticotrophin-releasing hormone expressing neurons in the hypothalamic paraventricular nucl eus of depressed patients. Clinical Neuroendocrinology 60: 436-444. Rygula R, Abumaria N, Flugge G, Fuchs E, Ruther E, Havemann-Reinecke U (2005). Anhedonia and motivational defi cits in rats: Impact of chronic social stress. Behavioural Brain Research, 162: 127-134. Sabban EL, Kvetnansky R (2001). Stress-tri ggered activation of gene expression in catecholaminergic systems: Dynamic s of transcriptional events. TRENDS in Neurosciences 24 (2): 91-98. Sgoifo A, De Boer SF, Haller J, Koolhaas JM (1996). Individual differences in plasma catecholamine and corticosterone stress res ponses of wild-type rats: Relationship with aggression. Physiology and Behavior, 60 (6): 1403-1407. Simpkiss JL, Devine DP (2003). Responses of th e HPA axis after chronic variable stress: Effects of novel and familiar stressors. Neuroendocrinology Letters Nos. 1/2 24: 97-103. Szigethy E, Conwell Y, Forbes NT, Cox C, and Caine ED (1994). Adrenal weight and morphology in victims of completed suicide. Biological Psychiatry, 36 (6): 374380. Whitnall MH (1993). Regulation of the hypot halamic corticotrophin-releasing hormone neurosecretory system. Progress in Neurobiology 40: 573-629.

PAGE 48

39 BIOGRAPHICAL SKETCH Kristen L. Stone graduated in May 2004 fr om the University of Central Florida (Orlando) with her Bachelor of Science degree in psychol ogy. She began her graduate education in the Psychology Department at th e University of Florida (Gainesville) in August 2004, working toward her Master of Science degree, in the Behavioral Neuroscience program.