Increased depressive behaviour in females and heightened corticosterone release in males to swim stress after adolescent social stress in rats

Available online at www.sciencedirect.com

Behavioural Brain Research 190 (2008) 33–40

Research report

Increased depressive behaviour in females and heightened corticosterone release in males to swim stress after adolescent social stress in rats Iva Z. Mathews a , Aleena Wilton a , Amy Styles b , Cheryl M. McCormick a,b,∗ a b

Psychology Department, Brock University, St. Catharines, ON L2S 3A1, Canada Neuroscience Program, Brock University, St. Catharines, ON L2S 3A1, Canada

Received 12 November 2007; received in revised form 26 January 2008; accepted 1 February 2008 Available online 8 February 2008

Abstract We previously reported that males undergoing chronic social stress (SS) (daily 1 h isolation and new cage partner on days 30–45 of age) in adolescence habituated (decreased corticosterone release) to the homotypic stressor, but females did not. Here, we report that adolescent males exposed to chronic social stress had potentiated corticosterone release to a heterotypic stressor (15 min of swim stress) compared to acutely stressed and control males. The three groups of males did not differ in depressive-like behaviour (time spent immobile) during the swim stress. Corticosterone release in socially stressed females was elevated 45 min after the swim stress compared to acutely stressed and control females, and socially stressed females exhibited more depressive behaviour (longer durations of immobility and shorter durations of climbing) than the other females during the swim stress. Separate groups of rats were tested as adults several weeks after the social stress, and there were no group differences in corticosterone release after the swim stress. The only group difference in behaviour among the adults was more time spent climbing in socially stressed males than in controls. Thus, there are sex-specific effects of social stress in adolescence on endocrine responses and depressive behaviour to a heterotypic stressor, but, unlike for anxiety, substantial recovery is evident in adulthood in the absence of intervening stress exposures. © 2008 Elsevier B.V. All rights reserved. Keywords: Adolescence; Puberty; Swim stress; Corticosterone; Sex differences; Hypothalamic–pituitary–adrenal axis; Depression

1. Introduction Adaptive physiological responses to a stressor involve the activation of the hypothalamic–pituitary–adrenal (HPA) axis. The HPA response to stress is initiated primarily by the release of corticotropin releasing hormone from the paraventricular nucleus (PVN) of the hypothalamus, which induces the release of ACTH from the pituitary gland. ACTH then acts on the adrenal cortex to induce the release of glucocorticoids (cortisol in people, corticosterone in rats). HPA function is regulated by negative feedback, whereby circulating glucocorticoids inhibit their own release by actions at corticosteroid receptors in the brain [23]. Reactivity of the HPA axis is shaped in part by the individual’s own stress history, especially dur∗

Corresponding author at: Canada Research Chair in Behavioural Neuroscience, Centre for Neuroscience and Department of Psychology, Brock University, 500 Glenridge Ave, St. Catharines, ON L2S 3A1, Canada. Tel.: +1 905 688 5550x3700; fax: +1 905 688 6922. E-mail address: [email protected] (C.M. McCormick). 0166-4328/$ – see front matter © 2008 Elsevier B.V. All rights reserved. doi:10.1016/j.bbr.2008.02.004

ing early life periods of rapid brain development (reviewed in [42,48,55]). The neuroendocrine response to many stressors (e.g., footshock, restraint, cold stress) is reduced after repeated or chronic exposure (reviewed in [6]). A dampened corticosterone response to a stressor can be evident after a single previous exposure to the same (homotypic) stressor [2], which illustrates the robust malleability of the HPA axis even in adulthood. Habituation of the release of glucocorticoids is thought to be due in large part to enhanced corticosteroid receptor negative feedback at neural regions upstream from the hypothalamus [12,25]. Habituation is also stressor-specific, such that subsequent exposure to a novel (heterotypic) stressor restores or potentiates the release of glucocorticoids (e.g., [6,31,54,58]). Dysregulation of the HPA axis by chronic exposure to stressors has been implicated in various diseases, including diabetes, heart disease, and psychiatric disorders such as depression (reviewed in [18,22,41]). Adolescence is a transitional time between early life and adulthood and is characterized by developmental changes in the HPA axis, including the emergence of sex differences (e.g.,

34

I.Z. Mathews et al. / Behavioural Brain Research 190 (2008) 33–40

females have a higher stress-induced corticosterone release than males) and the development of HPA axis regulation by sex hormones (reviewed in [38]). However, very little is known about HPA axis function in adolescence. The available studies typically find that adolescent rodents have a prolonged corticosterone response to an acute stressor relative to adults (e.g., [19,52,59]), which is largely attributed to immature negative feedback systems (e.g., [20]). As for HPA axis function in response to chronic stress in adolescence, there is evidence that unlike adult males, prepubertal (28 days of age) male rats did not habituate to repeated restraint [53]. We have reported evidence of habituation of corticosterone release to repeated stress (daily 1 h isolation for 16 days from day 30 to day 45 of age) in later adolescence, and that habituation in adolescence can be impeded by social instability (when daily 1 h isolation is followed by a change of cage partner) [39]. For rats that experienced the additional social instability in adolescence, females undergoing their 16th episode of isolation on day 45 of age did not differ in corticosterone concentrations (i.e. did not habituate) compared to control females, whereas males did have reduced corticosterone release compared to males undergoing a first isolation. Both adolescent socially stressed males and females had greater CRH mRNA expression in the PVN than controls [39]. Thus, chronic social stress in adolescence appears to increase the central drive of the HPA axis and alter corticosterone release to a homotypic stressor in a sex-specific manner, which may then impact HPA function when exposed to a heterotypic stressor. In a separate experiment, adolescent socially stressed male rats mounted a corticosterone response to a mild heterotypic stressor (15 min confinement to the open arm of an elevated plus maze) that was similar to that of controls with no history of social stress [40]. Unexpectedly, adolescent socially stressed male rats had lower plasma corticosterone concentrations than control rats 45 and 90 min after exposure to the heterotypic stressor. However, the faster return to baseline may reflect a diminished adrenal capacity due to the brief inter-stress interval (e.g., [45]), since rats exposed to an acute stressor a few hours before the heterotypic stressor also showed a faster return of corticosterone concentrations to baseline than controls. Thus, the extent to which chronic stress in adolescence alters neuroendocrine responses to new stressors requires further evaluation, particularly since dysregulated HPA function is associated with vulnerability to physical and mental disorders [10,21]. In the present paper, the interval between the last homotypic stressor exposure and the heterotypic stressor was increased to 24 h, and swim stress (15 min) was used because it is a more intense stressor (e.g., more physically demanding) than an elevated platform. This also allowed us to investigate, first, the extent to which chronic social stress in adolescence affected depressive-like behaviour, and second, the extent to which the risk for depressive behaviour would endure into adulthood. We previously reported that as adults, adolescent socially stressed rats had increased anxiety-like behaviour in both sexes [40], and had heightened psychostimulant-induced locomotor activity in females [35,36]. Thus, we predicted increased depressive behaviour in adolescent socially stressed rats, given the high co-

morbidity among anxiety disorders, depressive disorders, and drug abuse in clinical studies [28]. 2. Materials and methods 2.1. Animals Long-Evans female and male rats (n = 118) were obtained from Charles River, St. Constant, Quebec, at 22 days of age. Rats were housed in pairs, and were identified by tail colouring with a felt marker. The rats were kept on a 12-h light:12-h dark cycle (lights on at 08:00 h), with full access to rat chow and water at all times. The experimental procedures were consistent with National Institutes of Health Guide for Care and Use of Laboratory Animals (Publication No. 85-23, revised 1985), and Canadian Council on Animal Care guidelines and were approved by the Brock University Institutional Animal Care and Use Committee.

2.2. Adolescent stress conditions Rats were randomly assigned to the adolescent chronic social stress condition or one of two control conditions: acute stress (AS) or no stress control (CTL). The SS condition was carried out each of days 30–45 of age and involved 1 h of isolation in ventilated, round plastic containers (approximately 10 cm in diameter and 10 cm in height). After isolation the rats were returned to the colony to be housed with a new SS partner and a new cage (see [39] for a description of behaviour upon return to new cage and partner). The stress regimen occurred at various times during the light cycle to decrease the predictability of the event. After the last isolation on day 45, rats were returned to the original cage partner. CTL and AS rats received only regular cage maintenance until day 45. On day 45, all rats were weighed, and AS rats were isolated for the first and only time for 1 h in the isolation containers and then returned to their original cage and partner. The AS control group allows more interpretation of any difference observed between CTL and SS groups. The AS control group allows us to test whether the pattern of corticosterone release after chronic SS when faced with a new heterotypic stressor differs on the basis of the repeated stress experience or whether any one exposure to a stressor prior to exposure to the new stressor would alter corticosterone release. No differences were expected between CTL and AS on any measure. For experiment 1 (neuroendocrine and behavioural response to a heterotypic stressor after chronic social stress), a subset of rats (N = 12 for each of the three stress condition groups for both sexes except N = 10 for CTL females) underwent behavioural testing on day 46. For experiment 2 (neuroendocrine and behavioural responses to swim stress in adulthood), a subset of rats from the SS and the CTL conditions (N = 12 for each group) were left undisturbed except for regular cage maintenance from day 46 to day 70 of age, after which behavioural testing began.

2.3. Swim stress test The swim stress test was adapted from the forced swim test of Porsolt et al. [50]. On either day 46 or day 70 of age, rats were taken from the colony room and shuttled in pairs using a cage lined with cotton towels to the testing room. Rats were placed for 15 min in one of two Plexiglas cylinders (45 cm in height and 20 cm in diameter) containing 33 cm of water maintained at 27 ◦ C for 15 min. A Sony Digital Video Camera Recorder (CCD-TRV318) was used to record the session. The rats were taken out of the cylinders, placed back in the shuttle cage with the towels to dry off and taken back to the colony room. Behavioural scoring was conducted blind to experimental condition with the assistance of SMART Version 2.0 software (San Diego Instruments) to record the amount of time rats spent engaged in each behaviour during the first 5 min of the test. Behavioural measures for analysis were derived from the literature (e.g., [47,51]) and included the following four behaviours: swimming (quick movements of forelimbs and/or hindlimbs, including swimming in circles and pedaling), climbing (in a vertical position, pawing at the side of the cylinder with front paws and leveraging the body slightly out of the water), diving (head under water, swimming to the bottom of cylinder, including circling at the bottom) and

I.Z. Mathews et al. / Behavioural Brain Research 190 (2008) 33–40 immobility (reduced movement including floating and slow circling or pedaling). Diving occurred with very low frequency, and was thus not included in statistical analyses. Climbing reflects the most vigorous attempt at escape whereas immobility is the measure considered to represent depressive-like behaviour (e.g., [11,47]).

2.4. Blood sampling and corticosterone measurement Blood samples were obtained from the tail vein for the measurement of plasma concentration of corticosterone before and after (0, 45, and 90 min) 15 min of swim stress. Samples were collected in ice-chilled microcapillary tubes (Sarstedt), which were then centrifuged at 3000 rpm for 10 min to collect plasma. All plasma samples were stored at −80 ◦ C until radioimmunoassay (RIA) using highly specific antiserum (Dr. Greg Brown, University of Toronto), [3 H]-corticosterone (specific activity 70.0 Ci/mM; PerkinElmer), and duplicate aliquots of 5–10 ␮l of plasma. The minimum level of detection was 15 pg per tube with intra- and inter-assay variability less than 5% and 10%, respectively.

2.5. Estrous cycle Estrous cycle phase was determined by vaginal cytology after behavioural testing. Vaginal smears were obtained using a pipette and 10 ␮l of saline, placed on slides, dried and stained with Methylene blue. Using a microscope and low magnification, samples were characterized by estrous (cornified, anucleated cells), diestrous (a mix of leucocytes and nucleated epithelial cells) or proestrous (primarily round nucleated epithelial cells) phase of the Estrous cycle [30,32].

2.6. Statistical analyses Statistical analyses were conducted using repeated-measure analysis of variance (ANOVA) for the corticosterone concentrations and multivariate analysis of variance (MANOVA) for behavioural measures in the swim stress test. Analyses were conducted for males and females separately given the well-established sex differences in HPA function in response to stress [38] and in behaviour during the swim test [8,11]. Post hoc analyses consisted of F-tests for simple effects and Fisher’s LSD.

3. Results 3.1. Exposure to a heterotypic stressor 24 h after chronic social stress in adolescence 3.1.1. Females There were no group differences in baseline concentrations of plasma corticosterone. Stress Group by Timepoint × Estrous cycle ANOVA for corticosterone concentrations after the swim–stress test found a significant interaction between Stress Group and Timepoint (F4,48 = 4.18, p = 0.006). Post hoc analysis indicated that SS had higher corticosterone concentrations than CTL (p = 0.05) and AS (p = 0.06) 45 min after removal from the test (see Fig. 1a). For behaviour during the swim test, there was a betweensubjects effect of Estrous cycle for immobility (F1,28 = 6.32, p = 0.02) whereby rats in diestrus spent more time immobile than rats in estrus, and an effect of Stress Group for both climbing and immobility (F2,28 = 5.65, p = 0.009 and F2,28 = 3.47, p = 0.045, respectively). Post hoc analysis indicated that SS rats climbed less and spent more time immobile than AS (p = 0.003 and p = 0.014) and CTL (p = 0.03 and p = 0.06), which did not differ (see Fig. 2a).

35

3.1.2. Males The Stress Groups did not differ in baseline concentrations of plasma corticosterone. A Stress Group by Timepoint ANOVA for corticosterone levels after the swim–stress test found that corticosterone levels decreased over time (F2,46 = 14.76, p < 0.0001). There also was a main effect of Stress Group (F2,23 = 4.88, p = 0.017), which did not interact with Timepoint: post hoc analysis indicated that corticosterone levels after the swim test were higher in SS than in AS (p = 0.01) and CTL (p = 0.02), which did not differ (see Fig. 1b). There were no significant group differences for behaviours in the swim test (see Fig. 2b). 3.2. Exposure to a heterotypic stressor 25 days after chronic social stress in adolescence 3.2.1. Females There were no group differences in baseline concentrations of plasma corticosterone. Although the interaction of Stress Group by Estrous cycle by Timepoint for corticosterone levels after the swim–stress test was not significant, the possibility of group differences were further explored by analyzing the Estrous cycle groups separately. For rats in diestrus, the interaction of Timepoint and Stress Group was significant (F2,26 = 3.32, p = 0.05), although the group difference was not significant for any one Timepoint. SS and CTL did not differ for rats in estrus (see Fig. 1c). There were no significant group differences for behaviours in the swim test (see Fig. 2c). 3.2.2. Males The Stress Groups did not differ in baseline concentrations of plasma corticosterone. A Stress Group by Timepoint ANOVA for corticosterone levels after the forced swim test found that corticosterone levels decreased over time (F2,46 = 18.90, p < 0.0001), but there was no effect of, or interaction with, Stress Group (see Fig. 1d). During the swim test, SS males climbed more than CTL (F1,21 = 4.69, p = 0.04), but did not differ in time swimming or immobility (see Fig. 2d). 4. Discussion The Porsolt swim test was used to gauge the endocrine response to a heterotypic stressor after chronic social stress in adolescence and to determine if increased depressive behaviour would be evident in the adolescent socially stressed rats compared to controls. 4.1. Effects evident 24 h after adolescent social stress 4.1.1. Corticosterone response in males Chronic social stress in adolescence did not alter baseline plasma concentrations of corticosterone, which is consistent with our previous reports [39,40]. SS resulted in a potentiated corticosterone release to a heterotypic stressor (swim stress) in adolescent male rats that remained elevated compared to control and acutely stressed males over the 90 min period after their return to the home cage. These results, in combination with

36

I.Z. Mathews et al. / Behavioural Brain Research 190 (2008) 33–40

Fig. 1. Mean (S.E.M.) plasma corticosterone concentrations before and after swim stress in rats that underwent chronic social stress (SS), acute stress (AS), or no stress (controls, CTL) as adolescents and were stress tested either in adolescence (a and b) or in adulthood (c and d). N = 10–12/group. (a) Adolescent females at 45 min: * SS > CTL, p = 0.05; # SS > AS, p = 0.06; (b) adolescent males across timepoints: * SS > CTL, p = 0.02; # SS > AS, p = 0.01.

our previous finding of habituation of corticosterone release to a homotypic stressor in chronically stressed adolescent males [39], suggest that chronic stress has similar effects on the HPA axis in adolescents as it does in adult males, which have been found to show restored or potentiated corticosterone release in response to a heterotypic stressor after chronic stress (e.g., [5,11,31,58]). We previously reported a greater expression of CRH mRNA in the PVN under baseline conditions, and a greater increase in expression of CRH mRNA in the central nucleus of the amygdala to a 16th isolation in SS male rats at 45 days of age compared to controls (baseline and after a first isolation) [39]. Thus, the potentiated and prolonged corticosterone release of SS males to a heterotypic stressor may be due to an increased central drive of the HPA axis. When exploring the open arm of an elevated plus maze (a mild stressor, [13,56]), adolescent SS males had a blunted corticosterone response 45 and 90 min after open arm stress compared to controls [40]. It is possible that the faster return to baseline after open arm stress in that experiment was the result of a diminished adrenal capacity due to the brief inter-stress interval (e.g., [45]), since rats exposed to an acute stressor a few hours before the heterotypic stressor also showed a faster return of corticosterone concentrations to baseline than controls. Thus, it is possible that SS males in our previous experiment would have shown potentiated release of corticosterone in the open arm maze as did SS males in the present experiment to the forced swim test had the interval between the last isolation and the heterotypic stressor

been as long. Another possibility is that the different patterns of corticosterone release to the open arm maze than to the forced swim test after adolescent social stress in males might be due to differences in the intensity of, or nature of, the stressor [1,23,34]. The evidence that socially stressed males had a blunted corticosterone response to a mild stressor and an enhanced corticosterone release to a more intense stressor (e.g., more physically demanding) parallels a recent report in which male rats given voluntary exercise (a stressor for which there also is evidence of habituation of corticosterone levels with time [15]) for 4 weeks had a lower corticosterone response to a novel environment and a potentiated corticosterone response to swim stress [14]. Inputs from sensory and cognitive regions upstream of the PVN are important in determining the extent of activation of the HPA axis, and the latter results suggest that facilitation of the HPA axis depends in part on how the stressor is perceived. 4.1.2. Corticosterone response in females Although there are fewer studies using females, exaggerated corticosterone release in response to a heterotypic stressor after chronic stress also has been reported for adult females (e.g., [11]), which is notable considering that adult females also show less habituation of corticosterone release to a homotypic stressor than do males (e.g., [17]). We found that adolescent SS females had a more prolonged heterotypic stressor-induced elevation of corticosterone than did control females, in that their concentrations of plasma corticosterone were higher 45 min after stress

I.Z. Mathews et al. / Behavioural Brain Research 190 (2008) 33–40

37

Fig. 2. Mean (S.E.M.) time spent climbing or immobile during 5 min of swim stress in rats that underwent chronic social stress (SS), acute stress (AS), or no stress (controls, CTL) as adolescents and were tested either in adolescence (a and b) or in adulthood (c and d). N = 10–12/group. (a) Adolescent females: climbing, SS < AS (p = 0.003), SS < CTL (p = 0.03); immobility, SS > AS (p = 0.014), SS > CTL (p = 0.06). For immobility: estrus < diestrus, p = 0.02; (d) adult males: climbing, * SS > CTL (p = 0.04).

exposure than control females. SS females also had higher corticosterone concentrations at 45 min than acute-stress controls, but this difference missed statistical significance (p = 0.06). The differences between SS and control females in response to a heterotypic stressor were more subtle than those between SS and control males, which may be related to our finding that SS females, unlike SS males, did not habituate to the homotypic stressor (repeated isolation) [39]. Further, because plasma corticosteroid binding globulin levels are lower in SS females and males compared to controls [39], the inference can be made that the free, bioactive fraction of corticosterone would be higher in SS females than in control females. Together, these results suggest that chronic social stress may impair the ability of adolescent females to effectively ‘shut down’ the stress response to a heterotypic stressor (i.e., impair negative feedback). 4.1.3. Behaviour during the swim stress The differences in corticosterone release in response to swim stress in SS compared to controls did not map on to group differences in behaviour during the swim test. SS males did not

differ from control groups in duration of climbing or duration of immobility in the swim test. Irrespective of Estrous cycle group, SS females had shorter durations of climbing and longer durations of immobility than control groups. It is unlikely that the differences between SS and control females reflect a generalized reduction in locomotor function in SS females because no differences were observed between SS and controls in distance traveled during habituation to a locomotor test box at 46 days of age or in distance traveled after injection of saline on subsequent test days [37]. Exposure of rats to forced swimming is a pharmacologically validated model of depression in rats, with immobility as the measure of depressive-like behaviour [44,46]. Time spent struggling and passively floating (i.e., immobile) is thought to reflect active and passive coping, respectively. Thus, SS female rats exhibited more depressive-like behaviour, which is consistent with a large literature that draws a causal link between chronic stress exposure and depressive behaviour (e.g., [10,43]). That SS females showed depressive-like behaviour in the swim test and that SS males did not may reflect more their exposure to glucocorticoids over the course of the chronic

38

I.Z. Mathews et al. / Behavioural Brain Research 190 (2008) 33–40

stress procedure as opposed to their glucocorticoid response to the swim stress. Dissociations between the glucocorticoid response to the swim test and behaviour in the swim test have been reported [60]. Further, repeated administration of exogenous glucocorticoids increased immobility in a swim test in a dose-dependent manner whereas acute administration of glucocorticoids was without effect in male rats [26], and repeated administration of a high dose of corticosterone (40 mg/kg) reliably induced depressive behaviour in both male and female rats [27]. Thus, the sex-specific effects of SS in adolescence on behaviour in the swim test may be due to the habituation to the chronic stress exposure in males and not in females [39], such that only females received sufficiently high exposure to glucocorticoids over the course of the adolescent social stress period to increase depressive behaviour. In addition, we found that females in diestrus spent more time immobile than females in estrus, which is consistent with the literature, and may reflect either the higher generalized locomotor activity that rats exhibit during proestrus/estrus and/or anti-depressive effects of estrogens and progestins (e.g., [16,33,51]). The effect of Estrous cycle group on immobility was not observed in adult females, although the direction of the difference (diestrus > estrus) was the same as in adolescents. 4.2. Effects evident 25 days after adolescent social stress 4.2.1. Corticosterone response When tested as adults several weeks after the adolescent social stress, SS rats did not differ from controls in corticosterone release in response to swim stress, which is consistent with our previous report of no difference in adulthood in corticosterone release between SS and control rats to restraint stress [36]. Thus, any effects of chronic social stress in adolescence on HPA function may dissipate over time in the absence of intervening exposure to stressors. Our results suggest that, in adulthood, differences between SS and control females may depend on the phase of the Estrous cycle, in that group differences were found for females in the diestrous group that was similar to the difference observed between SS and control females tested in adolescence. However, because the difference between SS and control adult females was only observed in exploratory post hoc comparisons, more thorough investigation of a moderating effect of Estrous cycle on the relationship between SS and HPA function is warranted before a conclusion can be made. There are reports of altered estrous cyclicity after chronic stress [3,29], and the development of gonadal hormone regulation of the HPA axis occurs over adolescence (reviewed in [38]). However, the extent to which our chronic social stress procedure affects gonadal function is unknown. 4.2.2. Behaviour during the swim stress In contrast to the behavioural differences observed when tested in adolescence soon after the SS exposure, when tested as adults, SS and control females did not differ in behaviour during the swim test, whereas SS and control males did. The greater duration of climbing in SS males compared to control males may reflect a stress-induced bias toward active coping that

takes time to develop (since it was not observed in adolescent SS males), and may indicate a reduced vulnerability to depressive behaviour. However, we previously found that SS males as adults, and not as adolescents, exhibited more anxiety-like behaviour than controls [40], suggesting that increased climbing may reflect a panic state that only emerges as the animal matures. Others also have reported that some effects of stress in adolescence are delayed likely because further brain maturation is required for the effects to emerge (e.g., [24]), and that, depending on the nature and timing of the stressor, greater effects may be evident for anxiety measures than for measures of depression [49,57]. 4.3. Conclusion In summary, the effects of social stress in adolescence were sex-specific and depended on whether the rats were tested soon after or long after the stress exposures. Acutely stressed rats did not differ from control rats on any measure, which reinforces that chronic or repeated exposure to stressors is necessary to alter behaviour and neuroendocrine responses to the swim test. The present results indicate that our model of chronic social stress potentiates in males, and prolongs in females, corticosterone release in response to a heterotypic stressor. The neural mechanisms underlying facilitated corticosterone release to a heterotypic stressor have been investigated in adult male rats (e.g., [6,7,25,45]). However, the extent to which these same mechanisms operate similarly at other stages of development in both sexes remains to be determined. The results for corticosterone response to swim stress further validate our model of adolescent social instability as ‘chronic stress’. However, the behavioural consequences of social stress in adolescence for depressive behaviour were far more evident in females than in males when tested as adolescents. Greater immediate and/or enduring effects of adolescent social stress were found in females than in males for anxietylike behaviour [40] and for some behavioural responses to drugs of abuse [35–37]. Another possibility is that the blood sampling that occurred before the swim test served to alter behaviour in the swim test and thereby minimized or maximized any group differences. That chronic social stress in adolescence appears to increase the propensity for depressive behaviour in females parallels findings in the clinical literature, such as that the prevalence of depression is higher for women than men, and that the gender-bias in prevalence emerges in adolescence [4,9]. Lastly, adults exhibited recovery from the effects of adolescent social stress. Although the stress effects dissipated over time in the absence of intervening exposure to stressors, continued exposure to intermittent stressors into adulthood may increase risk for depressive behaviour in adolescent-stressed rats based on their exaggerated corticosterone release to a heterotypic stressor evident in the present data as adolescents and their increased anxiety-like behaviour as adults reported previously [40]. The results highlight the importance of investigating sex-specific and stressor-specific developmental trajectories in preclinical research on risk factors for psychopathology.

I.Z. Mathews et al. / Behavioural Brain Research 190 (2008) 33–40

Acknowledgements We thank Alanna Harrison and Lisa Tourounstzas for their assistance. The research was supported by the Natural Sciences and Engineering Research Council (B-USRA fellowship, AS; PGM fellowship, IZM; NSERC Discovery Grant, CMM), the Canadian Foundation for Innovation (CMM), and the Ontario Innovation Trust (CMM). References [1] Anisman H, Matheson K. Stress, depression, and anhedonia: caveats concerning animal models. Neurosci Biobehav Rev 2005;29:525– 46. [2] Armario A, Valles A, Dal-Zotto S, Marquez C, Belda X. A single exposure to severe stressors causes long-term desensitisation of the physiological response to the homotypic stressor. Stress 2004;7:157– 72. [3] Baker S, Bialajew C. Influence of housing on the consequences of chronic mild stress in female rats. Stress 2007;10:283–93. [4] Bebbington PE, Dunn G, Jenkins R, Lewis G, Brugha T, Farrell M, et al. The influence of age and sex on the prevalence of depressive conditions: report from the National Survey of Psychiatric Morbidity. Psychol Med 1998;28:9–19. [5] Bhatnagar S, Meaney M. Hypothalamic–pituitary adrenal function in chronic intermittently cold-stressed neonatally handled and non-handled rats. J Neuroendocrinol 1995;7:97–108. [6] Bhatnagar S, Dallman MF. Neuroanatomical basis for facilitation of hypothalamic–pituitary–adrenal responses to a novel stressor after chronic stress. Neuroscience 1998;84:1025–39. [7] Bhatnagar S, Viau V, Chu A, Soriano L, Meijer OC, Dallman MF. A cholecystokinin-mediated pathway to the paraventricular thalamus is recruited in chronically stressed rats and regulates hypothalamic– pituitary–adrenal function. J Neurosci 2000;20:5564–73. [8] Bielajew C, Konkle AT, Kentner AC, Baker SL, Stewart A, Hutchins AA, et al. Strain and gender specific effects in the forced swim test: effects of previous stress exposure. Stress 2003;6:269–80. [9] Born L, Shea A, Steiner M. The roots of depression in adolescent girls: is menarche the key? Curr Psychiatry Rep 2002;4:449–60. [10] Checkley S. The neuroendocrinology of depression and chronic stress. Brit Med Bull 1996;52(3):597–617. [11] Dalla C, Antoniou K, Drossopoulou G, Xagoraris M, Kojras N, Sfikakis A, et al. Chronic mild stress impact: are females more vulnerable? Neuroscience 2005;135:703–14. [12] Dallman M, Akana S, Cascio C, Darlington D, Jacobson L, Levin N. Regulation of ACTH secretion: variations on a theme of B. Recent Prog Horm Res 1987;43:113–73. [13] Degroot A, Wade M, Salhoff C, Davis RJ, Tzavara ET, Nomikos GG. Exposure to an elevated platform increases plasma corticosterone and hippocampal acetylcholine in the rat: reversal by chlordiazepoxide. Eur J Pharmacol 2004;493:103–9. [14] Droste SK, Chandramohan Y, Hill LE, Linthorst AC, Reul JM. Voluntary exercise impacts on the rat hypothalamic–pituitary–adrenocortical axis mainly at the adrenal level. Neuroendocrinology 2007;86:26–37. [15] Fediuc S, Campbell JE, Riddell MC. Effect of voluntary wheel running on circadian corticosterone release and on HPA axis responsiveness to restraint stress in Sprague–Dawley rats. J Appl Physiol 2006;100:1867–75. [16] Frye CA, Walf AA. Changes in progesterone metabolites in the hippocampus can modulate open field and forced swim test behavior of proestrous rats. Horm Behav 2002;41:306–15. [17] Galea LAM, McEwen BS, Tanapat P, Spencer RL, Dhanbhar FS. Sex differences in dendritic atrophy of CA3 pyramidal neurons in response to chronic restraint stress. Neuroscience 1997;81:689–97. [18] Gold PW, Drevets WC, Charney DS. New insights into the role of cortisol and the glucocorticoid receptor in severe depression. Biol Psychiatry 2002;52:381–5.

39

[19] Goldman L, Winget C, Hollingshead GW, Levine S. Postweaning development of negative feedback in the pituitary–adrenal system of the rat. Neuroendocrinology 1973;12:199–211. [20] Gomez F, Houshyar H, Dallman MF. Marked regulatory shifts in gonadal, adrenal, and metabolic system responses to repeated restraint stress occur within a 3-week period in pubertal male rats. Endocrinology 2002;143:2852–62. [21] Heim C, Ehlert U, Hellhammer DH. The potential role of hypocortisolism in the pathophysiology of sterss-related bodily disorders. Psychoneuroendocrinology 2000;25:1–35. [22] Herbert J, Goodyer IM, Grossman AB, Hastings MH, de Kloet ER, Lightman SL, et al. Do corticosteroids damage the brain. J Neuroendocrinol 2006;18:393–411. [23] Herman JP, Ostrander MM, Mueller NK, Figueiredo H. Limbic system mechanisms of stress regulation: hypothalamo–pituitary–adrenocortical axis. Prog Neuro Psychopharm Biol Psychiat 2005;29:1201–13. [24] Isgor C, Kabbaj M, Akil H, Watson SJ. Delayed effects of chronic variable stress during peripubertal-juvenile period on hippocampal morphology and on cognitive and stress axis functions in rats. Hippocampus 2004;14:636–48. [25] Jaferi A, Bhatnagar S. Corticosterone can act at the posterior paraventricular thalamus to inhibit hypothalamic–pituitary–adrenal activity in animals that habituate to repeated stress. Endocrinology 2006;147:4917–30. [26] Johnson SA, Fournier NM, Kalynchuk LE. Effect of different doses of corticosterone on depression-like behavior and HPA axis responses to a novel stressor. Behav Brain Res 2006;168:280–8. [27] Kalynchuk LE, Gregus A, Boudreau D, Perrot-Sinal TS. Corticosterone increases depression-like behavior, with some effects on predator odorinduced defensive behavior, in male and female rats. Behav Neurosci 2004;118:1365–77. [28] Kessler RC. The global burden of anxiety and mood disorders: putting ESEMeD findings into perspective. J Clin Psychiatry 2007;68:10–9. [29] Konkle ATM, Baker SL, Kentner AC, Barbagallo LS, Merali Z, Bielajew C. Evaluation of the effects of chronic mild stressors on hedonic and physiological responses: sex and strain compared. Brain Res 2003;992:227–38. [30] LeFevre JA, McClintock MK. Social modulation of behavioral reproductive senescence in female rats. Physiol Behav 1992;52:603–8. [31] Ma XM, Lightman SL, Aguilera G. Vasopressin and corticotropin-releasing hormone gene responses to novel stress in rats adapted to repeated restraint. Endocrinology 1999;140:3623–32. [32] Marcondes FK, Bianchi FJ, Tanno AP. Determination of the estrous cycle phases of rats: some helpful considerations. Braz J Biol 2002;62:609–14. [33] Marvan ML, Santana S, Chavez-Chavez L, Bertran M. Inescapable shocks accentuate fluctuations of forced swimming immobility in different phases of the rat estrous cycle. Arch Med Res 1997;28:369–72. [34] McCarty R, Horwatt K, Konarska M. Chronic stress and sympatheticadrenal medullary responsiveness. Soc Sci Med 1988;26:333–41. [35] McCormick CM, Robarts D, Gleason E, Kelsey JE. Stress during adolescence enhances locomotor sensitization to nicotine in adulthood in female, but not male, rats. Horm Behav 2004;46:458–66. [36] McCormick CM, Robarts D, Kopeikina K, Kelsey JE. Long-lasting, sexand age-specific effects of social stress on corticosterone responses to restraint and locomotor responses to psychostimulants in rats. Horm Behav 2005;48:64–74. [37] McCormick CM, Ibrahim FN. Locomotor sensitization to nicotine and Fos immunoreactivity in the paraventricular nucleus of the hypothalamus in adolescent socially-stressed rats. Pharmacol Biochem Behav 2007;86:92–102. [38] McCormick CM, Mathews IZ. HPA function in adolescence: role of sex hormones in its regulation and the enduring consequences of exposure to stressors. Pharmacol Biochem Behav 2007;86:220–33. [39] McCormick CM, Merrick A, Secen J, Helmreich DL. Social instability in adolescence alters the central and peripheral hypothalamic–pituitary– adrenal responses to a repeated homotypic stressor in male and female rats. J Neuroendocrinol 2007;19:116–26. [40] McCormick CM, Smith C, Mathews IZ. Effects of chronic social stress in adolescence on anxiety and neuroendocrine response to mild stress in male and female rats. Behav Brain Res 2008;187:228–38.

40

I.Z. Mathews et al. / Behavioural Brain Research 190 (2008) 33–40

[41] McEwen BS. Physiology and neurobiology of stress and adaptation: central role of the brain. Physiol Rev 2007;87:873–904. [42] Meaney MJ, Szyf M. Environmental programming of stress responses through DNA methylation: life at the interface between a dynamic environment and a fixed genome. Dialogues Clin Neurosci 2005;7:103– 23. [43] Molina VA, Heyser CJ, Spear LP. Chronic variable stress or chronic morphine facilitates immobility in a forced swim test: reversal by naloxone. Psychopharmacology 1994;114:433–40. [44] O’Neil MF, Moore NA. Animal models of depression: are there any? Hum Psychopharmacol Clin Exp 2003;18:239–54. [45] Ostrander MM, Ulrich-Lai YM, Choi DC, Richtand NM, Herman JP. Hypoactivity of the hypothalamo–pituitary–adrenocortical axis during recovery from chronic variable stress. Endocrinology 2006;147:2008– 17. [46] Palanza P, Gioiosa L, Parmigiani S. Social stress in mice: gender differences and effects of estrous cycle and social dominance. Physiol Behav 2001;73:411–20. [47] Perrot-Sinal TS, Gregus S, Boudreau D, Kalynchuk LE. Sex and repeated restraint interact to affect cat dor-induced defensive behavior in adult rats. Brain Res Bull 2004;1027:161–72. [48] Phillips DI, Jones A. Fetal programming of autonomic and HPA function: do people who were small babies have enhanced stress responses? J Physiol 2006;572:45–50. [49] Pohl J, Olmstead MC, Wynne-Edwards KE, Menard JL. Repeated exposure to stress across the childhood-adolescent period alters rats’ anxiety- and depression-like behaviors in adulthood: the importance of stressor type and gender. Behav Neurosci 2007;121:462–74. [50] Porsolt RD, Anton G, Blavet N, Jalfre M. Behavioral despair in rats: a new model sensitive to antidepressant treatments. Eur J Pharmacol 1978;47:379–91.

[51] Rachman IM, Unnerstall JR, Pfaff DW, Cohen RS. Estrogen alters behavior and forebrain c-fos expression in ovariectomized rats subjected to the forced swim test. Proc Natl Acad Sci 1998;95:13941–6. [52] Romeo RD, Lee SJ, Chhua N, McPherson CR, McEwen BS. Testosterone cannot activate an adult-like stress response in prepubertal male rats. Neuroendocrinology 2004;79:125–32. [53] Romeo RD, Bellani R, Karatsoreos IN, Chhua N, Vernov M, Conrad CD, et al. Stress history and pubertal development interact to shape hypothalamic–pituitary–adrenal axis plasticity. Endocrinology 2006;147: 1664–74. [54] Sabban EL, Serova LI. Influence of prior experience with homotypic or heterotypic stressor on stress reactivity in catecholaminergic systems. Stress 2007;10:137–43. [55] Sanchez MM. The impact of early adverse care on HPA axis development: nonhuman primate models. Horm Behav 2006;50:623–31. [56] Storey JD, Robertson DA, Beattie JE, Reid IC, Mitchell SN, Balfour DJ. Behavioural and neurochemical responses evoked by repeated exposure to an elevated open platform. Behav Brain Res 2006;166:220–9. [57] Tsoory M, Cohen H, Richter-Levin G. Juvenile stress induces a predisposition to either anxiety or depressive-like symptoms following stress in adulthood. Eur Neuropsychopharm 2007;17:245–56. [58] van Raaij MT, Dobbe CJ, Elvers B, Timmerman A, Schenk E, Oortigiesen M, et al. Hormonal status and the neuroendocrine response to a novel heterotypic stressor involving subchronic noise exposure. Neuroendocrinology 1997;65:200–9. [59] Vazquez DM, Akil H. Pituitary–adrenal response to ether vapor in the weanling animal: characterization of the inhibitory effect of glucocorticoids on adrenocorticotropin secretion. Pediatr Res 1993;34:646–53. [60] Walker CD, Trottier G, Rochford J, Lavallee D. Dissociation between behavioral and hormonal responses to the forced swim stress in lactating rats. J Neuroendocrinol 1995;7:615–22.