- Email: [email protected]
Social interaction decreases stress responsiveness during adolescence
Psychoneuroendocrinology (2011) 36, 1370—1377
a v a i l a b l e a t w w w. s c i e n c e d i r e c t . c o m
j o u r n a l h o m e p a g e : w w w. e l s e v i e r. c o m / l o c a t e / p s y n e u e n
Social interaction decreases stress responsiveness during adolescence Stephanie Lu ¨rzel a,b,*, Sylvia Kaiser a,b, Norbert Sachser a,b a b
Department of Behavioural Biology, University of Mu ¨nster, Badestraße 13, 48149 Mu ¨nster, Germany Otto Creutzfeldt Center for Cognitive and Behavioural Neuroscience, Badestraße 13, 48149 Mu ¨nster, Germany
Received 20 December 2010; received in revised form 24 February 2011; accepted 18 March 2011
KEYWORDS HPA axis; Cortisol; Adolescence; Social experience; Testosterone; Organizational effect
Summary Adolescence is the transition from infancy to adulthood and encompasses major changes in the brain, the endocrine systems, and behavior. During late adolescence, male guinea pigs living in mixed-sex colonies exhibit a lower cortisol (C) response to novelty compared with animals in other ages and housing conditions. It was hypothesized that this reduction in stress responsiveness is induced by a high amount of social interactions in the colonies. In a previous ¨rzel et al., 2010), late adolescent colony-housed males (CM) were compared with study (Lu similarly aged males that were housed in heterosexual pairs (PM) as well as with males that were also housed in pairs, but regularly received additional social stimulation by allowing them ten times to interact with unfamiliar adult animals of both sexes for 10 min (SM). CM had a significantly lower stress response than PM, with SM being intermediate and not significantly different from either of the other groups. We assumed that the amount of social stimulation in SM was insufficient in order to achieve a significant reduction of stress responsiveness compared with PM. For the present study, we hypothesized that with a higher amount of social stimulation, a significant difference in stress responsiveness between PM and SM becomes apparent during late adolescence. Thus, PM were again compared with SM that, this time, had received twice as much social stimulation as in the previous study. As a result, stress responsiveness was indeed significantly lower in SM than in PM during late adolescence. Thus, a high amount of social interactions during the course of adolescence leads to a decreased stress responsiveness. Furthermore, SM showed an increase in testosterone (T) levels caused by social stimulation. We hypothesize that the reduction in stress responsiveness is brought about by high T levels that organize central neural structures over the course of adolescence. # 2011 Elsevier Ltd. All rights reserved.
1. Introduction * Corresponding author at: Department of Behavioural Biology, University of Mu ¨nster, Badestraße 13, 48149 Mu ¨nster, Germany. Tel.: +49 251 83 21004; fax: +49 251 83 23896. E-mail address: [email protected] (S. Lu ¨rzel).
Adolescence is the gradual transition from infancy to adulthood and, as such, characterized not only by the attainment of sexual maturity, but also by a variety of changes in anatomy, endocrine systems, neural circuitry, as well as
0306-4530/$ — see front matter # 2011 Elsevier Ltd. All rights reserved. doi:10.1016/j.psyneuen.2011.03.010
Social interaction decreases stress responsiveness during adolescence behavior (Sisk and Zehr, 2005; Yurgelun-Todd, 2007; Blakemore, 2008). Due to these changes, adolescence represents a period of high vulnerability to adverse environmental conditions (Dahl, 2004; McCormick and Mathews, 2010). On the other hand, it is also a window of opportunity for adaptation to the current environment (Williams et al., 2001; Dahl, 2004; Steinberg, 2005; McCormick et al., 2008), similar to the sensitive phases during the pre- and early postnatal period (for a review, see Sachser et al., 2010). Among the alterations in endocrine systems during adolescence, stress responsiveness seems to play an important role. Individuals of various species undergo an upregulation of the hypothalamo—pituitary—adrenal (HPA) axis during adolescence (Spear, 2000; McCormick and Mathews, 2010), but in some species, the contrary is the case. For instance, the adult-typical HPA response to novelty is reduced in periadolescent mice (Adriani and Laviola, 2000), and periadolescent rats have a lower, although prolonged, corticosterone response compared with adult males (McCormick et al., 2008). In a study conducted throughout the life span of male guinea pigs, Hennessy et al. (2006) found a similar cortisol (C) hyporesponsiveness during late adolescence. A follow-up study revealed that this phenomenon does not exist under all environmental conditions, but depends on the amount and nature of social experiences made during this phase of life (Kaiser et al., 2007): Late adolescent male guinea pigs housed in a mixed-sex colony display a significantly lower increase of C levels in a novel environment than equally aged males kept only with a female since weaning. The mechanisms that lead to a reduced C responsiveness are only partially understood. One possible explanation is the inhibiting effect of testosterone (T) on C secretion. In addition to activational effects, organizational effects of Ton HPA activity during the perinatal phase (reviewed in McCormick and Mathews, 2007) are well known. A recent study in rats delivered evidence for organizational effects occurring during adolescence as well (Evuarherhe et al., 2009). A modulation of HPA function by T has also been discussed for guinea pigs, since colony-housed males exhibit a peak in T levels during adolescence (Sachser and Pro ¨ve, 1988) and have higher T levels than pair-housed males in adulthood (Sachser, 1990). This peak and subsequent higher T levels are most likely due to the constant challenge posed by other males and to the presence of females in the colony, resulting in a high amount of social interactions (Wingfield et al., 1990; Goymann et al., 2007). Thus, a very recent study investigated whether the difference in stress responsiveness between colony-housed and pair-housed males is indeed due to differences in the amount of social interactions experienced during adolescence (Lu ¨rzel et al., 2010). For these purposes, a third group was established in addition to pair-housed and colonyhoused males. This group was kept under pair-housing conditions but received additional social stimulation by allowing the males 10 times to interact with unfamiliar conspecifics of both sexes for 10 min. As a result, C responsiveness of these males was intermediate between colony- and pair-housed males. The social stimulation paradigm failed, however, to bring about a significantly reduced stress response compared to pair-housed males. Based on these data, the authors argued that a higher amount of social stimulation might lead to a stress response that is significantly reduced in comparison with pair-housed males without additional stimulation.
1371
Following this reasoning, the amount of social stimulation was doubled in this study, and the socially stimulated males were again compared with pair-housed males without additional social stimulation. Social stimulation was hypothesized to lead to a significantly reduced stress responsiveness during late adolescence. Furthermore, it was predicted that social stimulation triggers an acute elevation of T levels, which would be in accordance with the challenge hypothesis (Wingfield et al., 1990; Goymann et al., 2007).
2. Methods 2.1. Animals and housing conditions The guinea pigs (Cavia aperea f. porcellus) used were descendents of a heterogeneous shorthaired and multicolored stock of 40 animals obtained from a breeder in 1975. Experimental animals were derived from two mixed-sex colonies housed in the same room. The colonies consisted of 8—10 males and 13—15 females as well as their offspring, and showed a graduated age structure ranging from approximately 1 to 19 months. All animals could be individually identified by natural markings and were housed under controlled conditions: 12:12 LD (lights on 0700 h), temperature 22 2 8C, relative humidity about 60%. Commercial guinea pig diet (Ho ¨veler ‘‘Spezialfutter’’ 1070 for guinea pigs, Ho ¨veler Spezialfutterwerke GmbH&C. KG, Langenfeld, Germany) and water were available ad libitum. This diet was supplemented with hay and straw. Vitamin C was provided in the water twice per week. Enclosures were cleaned once per week. All experiments were announced to the competent local authority and were approved by the ‘‘Tierschutzbeauftragter’’ of the University of Muenster. Experiments were carried out in accordance with the European Communities Council Directive of 24 November 1986 (86/609/EEC). All efforts were made to minimize animal suffering and to reduce the number of animals used.
2.2. Experimental design The experimental design is depicted in Figure 1. All experimental animals were born in large mixed-sex colonies (day 0). After weaning, but before sexual maturation, experimental animals were housed with a single female of the same approximate age in 0.5 m2 enclosures (day 30). The females used as partners were colony-born and had been housed in all-female groups since the time of weaning (day 21 of age) to prevent pregnancy. Experimental animals were randomly assigned to one of two groups: pair-housed males (PM; n = 12) remained with only their female partner, whereas the other group received additional social stimulation from day 80 of life onward (SM; n = 13). No more than one male from each litter was assigned to the same group. In both early adolescence (day 55) and late adolescence (day 120), experimental animals of both groups were exposed to a standardized psychological stressor to assess C responsiveness. All procedures were conducted at specified ages of the animals with a tolerance of 1 d.
1372
Figure 1 Experimental design. PM: pair-housed males; SM: pair-housed males with additional social stimulation. From an age of 30 days onwards, PM and SM were kept with one female only. From day 80 to day 118, stimulus animals were added to SM every other day. Blood samples were taken for five social stimulation sessions from SM, and from PM for control procedures on the corresponding days. On days 55 and 120, stress response tests (SRTs) were conducted with all animals.
2.3. Social stimulation and assessment of testosterone concentrations For social stimulation, SM were given the opportunity to interact with unfamiliar animals of both sexes every second day 1 d (SM never experienced more than one social stimulation session per day). The time of day at which social stimulation occurred was varied in order to avoid possible habituation effects by introducing unpredictability. Social stimulation took place in the home cage of SM in the presence of their female partner. Since female guinea pigs attain sexual maturity at about 30 days of age, and males at about 60 days of age (Kemme et al., 2009), it is safe to assume that the females were pregnant during the time of social stimulation, preventing a confounding influence of oestrus. The period of social stimulation lasted from day 80 to day 118, which adds up to 20 stimulation sessions, 15 sessions with a male and 5 sessions with a female stimulus animal that were evenly distributed throughout the social stimulation period. Altogether, six stimulus males were used according to a pseudo-randomized schedule that had been living in the colonies for 11—19 months and were then rehoused with a younger male. Each stimulus male was used a maximum of three times with each experimental animal, with an interval of at least 14 days between encounters of the same animals. Thus, in the second and possible third encounter, the stimulus males were not entirely unfamiliar, but we can reasonably assume that they were not familiar in the sense of a stable dominance relationship, especially since agonistic behavior was not notably diminished. The 34 stimulus females drawn from the colonies were always pregnant at the time of testing to preclude any influence of oestrus. They were never used more than once with the same experimental animal. The stimulus animal remained in the enclosure for 10 min. During the stimulations, the experimenter observed the animals in order to separate them if a fight escalated. Out of a total of 165 stimulation sessions using males as stimulus animals, six had to be cut short because aggression escalated. Behavior during social stimulations was not recorded quantitatively since it has been shown in numerous studies that male guinea pigs usually display agonistic behavior towards
S. Lu ¨rzel et al. each other and courtship and sexual behavior towards females (e.g., Sachser et al., 1994; Kaiser et al., 2007; Lu ¨rzel et al., 2010). On the other hand, the cases in which they show agonistic behavior towards females as well as sociopositive behaviors towards males are very rare. These findings also hold true for the present study (qualitative analysis). On days 80, 86, 98, 110, and 114, two blood samples were taken, one directly before the stimulation (0 h) and another one 1 h later, in order to determine basal T values as well as changes in T concentration. Social stimulation sessions involving blood sampling were started at 1300 h 15 min to minimize confounding effects caused by diurnal variation. Blood samples were also taken from PM at corresponding times.
2.4. Assessment of cortisol responsiveness PM and SM were tested for their stress responsiveness alone in a novel enclosure in another guinea pig housing room for 2 h on days 55 and 120. The enclosure was a 1 m2 wooden box that contained food and water as well as clean wood shavings for bedding. A novel environment has been shown to act as a psychological stressor in guinea pigs, causing an increase in C levels (e.g., Hennessy et al., 2006). All tests were started at 1300 h 15 min. At the beginning of each test, blood samples were taken to assess basal Tand C concentrations. After that, the animal was introduced into the unfamiliar environment. At 60 min and 120 min, subsequent blood samples were taken to determine changes in C concentrations. After the third blood sample, the animal was placed back into his home enclosure.
2.5. Blood sample collection and hormone determination Blood samples were collected from the blood vessels of the ears. A muscle salve (Finalgon, Boehringer Ingelheim Pharma KG, Ingelheim am Rhein, Germany) was applied to the ear to stimulate circulation and the vessels were illuminated with a cold-point lamp. Vessels were pricked with an injection needle and about 0.4 ml of blood was collected in heparinized capillary tubes. One experimenter held the male in his/her lap, while a second collected the sample. Guinea pigs show little struggling during the collection procedure, and no elevation of plasma C or T levels occurs for about 6 min (Sachser, 1994). Accordingly, all samples for determination of C levels were collected within 3 min, all samples for determination of T levels within 6 min of entering the room. Because no anesthesia is required, hormone levels in the second and third samples were not influenced by previous exposure to anesthesia. Plasma was separated by centrifugation (11,700 g for 5 min) and deep frozen ( 20 8C) until assayed. Plasma C concentrations were determined in duplicate by radioimmunoassay without chromatography using specific antibodies against C (Biotrend, Ko ¨ln, Germany) as in earlier work (e.g., Kaiser et al., 2003). The antibodies used cross-reacted with relevant steroids as follows: C 100%, prednisolone 36%, 11-deoxyC 10%, corticosterone 3.2%, cortisone 0.9%. The intra- and interassay coefficients of variation (%CV) were 4.8 and 8.7%, respectively.
Social interaction decreases stress responsiveness during adolescence
1373
Figure 2 Plasma testosterone [ng/ml] before (A) and after (B) social stimulation. Blood samples were taken immediately before 10 min of social stimulation and 1 h later. Indicated are mean and SEM. PM: pair-housed males; SM: pair-housed males with additional social stimulation. Stimulus animals were males on days 80, 98, and 114, and females on days 86 and 110. Statistics: mixed factorial ANOVA (A: main effect of age, p = 0.002; interaction of age and group, p = 0.002; B: main effect of group, p = 0.002; interaction of age and group, p = 0.006), post hoc testing: independent-samples t-tests. nPM = 11; nSM = 8—10. *p 0.05; **p 0.01; ***p 0.001.
Plasma T concentrations were determined using a solid phase enzyme-linked immunosorbent assay (ELISA; Testosterone ELISA Kit, Demeditec Diagnostics GmbH, Kiel, Germany). The antibody used cross-reacted with relevant steroids as follows: T 100%, 11b-hydroxytestosterone 3.3%, 19-nortestosterone 3.3%, androstenedione 0.9%, 5a-dihydrotestosterone 0.8%, 17a-methyltestosterone 0.1%, epitestosterone, oestradiol, progesterone, C, oestrone and danazol <0.1% each. The intraassay %CV was 5.7%, the interassay %CV was 7.2%.
2.6. Data analysis After confirming that none of the data sets deviated from a normal distribution with a one-sample Kolmogorov—Smirnov test, mixed factorial analyses of variance (ANOVAs) were calculated, since differences between dependent as well as independent samples had to be evaluated. Greenhouse—Geisser corrections were applied if appropriate. If ANOVAs yielded significant interactions, they were followed by pairwise comparisons using independent-samples and paired-samples t-tests, respectively. In cases of multiple comparisons, the sequential Bonferroni—Holm method (Holm, 1979) was applied. On the basis of our previous work (Kaiser et al., 2007; Lu ¨rzel et al., 2010), t-tests were 2-tailed for basal hormone concentrations and 1-tailed for increases in hormone concentrations. Data were analyzed using PASW Statistics 18 (IBM, New York, USA). The significance level is 0.05. Graphs show mean and standard error of the mean and were created using SigmaPlot 11 (Systat Software GmbH, Erkrath, Germany). Several animals were dropped from analysis of hormones due to incomplete blood samples. Final sample sizes are given in Section 3.
3. Results 3.1. Testosterone response to social stimulation For T basal values (Figure 2A) taken immediately before 10 min of social stimulation, ANOVA revealed a main effect of age (nPM = 11; nSM = 10; F(4, 76) = 4.677, p = 0.002) as well
as an interaction of age and group (F(4, 76) = 4.784, p = 0.002). The subsequent t-tests revealed significantly higher basal T values in socially stimulated males (SM) than in pair-housed males (PM) on day 86 (n = 11; t = 3.564, p = 0.002). Most importantly, there was a significant main effect of group on T response values (Figure 2B; nPM = 11; nSM = 8; ANOVA: F(1, 17) = 13.147, p = 0.002) and an interaction of age and group (F(2.263, 38.006) = 5.561, p = 0.006). t-tests found higher response values in SM than in PM on days 80, 86, and 110 (n = 11; day 80: t = 2.513, p = 0.013; day 86: t = 3.930, p = 0.001; day 110: t = 3.623, p = 0.001). Basal T levels on the days of stress response tests — that is, before and after the period of social stimulation — differed neither between groups (n = 11; ANOVA: F(1, 20) = 0.089, n.s.) nor between day 55 and day 120 (F(1, 20) = 3.374, n.s.), nor was there any interaction of age and group (F(1, 20) = 0.106, n.s.).
3.2. Consequences of social stimulation on cortisol responsiveness On days of stress response tests, basal C concentrations did not differ between PM and SM (Figure 3; n = 11; ANOVA: F(1, 20) = 3.541, n.s.), and there was no interaction of age and group (F(1,20) = 0.117, n.s.). However, ANOVA revealed a main effect of age, with values being slightly, but significantly higher on day 120 than on day 55 (F(1, 20) = 5.322, p = 0.032). Concerning C responsiveness, there is a main effect of group as well as an interaction of age and group for the increase in C levels 1 h after the onset of the stressor (Figure 4A; n = 11; ANOVA: main effect on group: F(1, 20) = 5.987, p = 0.024, interaction age group: F(1, 20) = 4.382, p = 0.049). SM have a significantly lower stress response on day 120 than on day 55 (t-test: t = 2.185, p = 0.027), whereas values of PM do not differ between the different points in time (t-test: t = 0.888, n.s.). As a consequence, SM have a lower stress response than PM on day 120 (t-test: t = 3.556, p = 0.001), but not on day 55 (t-test: t = 0.932, n.s.). Analysis of the increase in C levels 2 h after
1374
Figure 3 Basal plasma cortisol [ng/ml]. Indicated are mean and SEM. PM: pair-housed males; SM: pair-housed males with additional social stimulation. Statistics: mixed factorial ANOVA. n = 11. *p 0.05.
the onset of the stressor yields similar results: there is a main effect of group as well as an interaction of age and group (Figure 4B; n = 11; ANOVA: main effect on group: F(1, 20) = 6.011, p = 0.024, interaction age group: F(1, 20) = 8.994, p = 0.007). Again, the stress response of SM is significantly decreased on day 120 compared with day 55 (ttest: t = 2.534, p = 0.015), whereas it does not differ in PM between the two testing days (t-test: t = 1.646, n.s.), leading to SM having a lower stress response than PM on day 120 (t-test: t = 3.843, p < 0.001), but not on day 55 (ttest: t = 0.320, n.s.).
4. Discussion 4.1. Social stimulation determines HPA responsiveness The major aim of this study was to investigate the effects of social interactions on stress responsiveness during late adolescence. To this purpose, male guinea pigs kept under pair-
S. Lu ¨rzel et al. housing conditions with one female only (PM) were compared with males that lived with one female and received additional social stimulation by interactions with unfamiliar adult males and females (SM). In a previous study with a similar experimental design (Lu ¨rzel et al., 2010), ten occasions of social stimulation had not sufficed to reduce stress responsiveness to a significant extent in SM, but the results had allowed for the assumption that with higher amounts of social stimulation, a significant difference in stress responsiveness between the two groups might be induced. In the present study, the amount of social stimulation was doubled, and, as predicted, SM had a significantly reduced stress response compared with PM during late adolescence. Furthermore, C basal values increased slightly but significantly over the course of adolescence, which most likely reflects developmental changes, as there was neither a main effect of group nor an interaction of group and age. From our study, it cannot be concluded whether the reduction of stress responsiveness is mainly due to stimulation with males (eliciting agonistic behavior) or with females (triggering courtship and sexual behavior), or whether the interaction with stimulus animals of both sexes is necessary in order to reduce HPA reactivity. Nevertheless, this study shows that the causal factor underlying the reduced stress responsiveness is indeed a high amount of social interactions with adult conspecifics, since the two groups experienced otherwise identical housing conditions. In other species, effects of social interactions during adolescence on HPA responsiveness have been shown as well: Male hamsters that underwent a social subjugation paradigm during adolescence showed reduced C response values after an aggressive encounter in adulthood compared with control males (Ferris et al., 2005). In another study, adolescent male rats lived in groups of four animals but experienced various social conditions from isolation to crowding for short amounts of time between day 28 and day 56 of age. These males had a prolonged corticosterone response compared with controls 24 h after the last social change, but not three weeks later (Isgor et al., 2004). Male rats that had experienced 1 h isolation daily followed by change of cage partner from day 30 to day 45 of age (mid-adolescence) showed an enhanced corticosterone release after swim stress on day
Figure 4 Increase in plasma cortisol [ng/ml] 1 h (A) and 2 h (B) after the onset of a psychological stressor (novel environment). Indicated are mean and SEM. PM: pair-housed males; SM: pair-housed males with additional social stimulation. Statistics: mixed factorial ANOVA (increase to 1 h: main effect of group, p = 0.024; interaction of age group, p = 0.049; increase to 2 h: main effect of group, p = 0.024; interaction of age group, p = 0.007), post hoc testing: Bonferroni—Holm corrected independent-samples as well as paired-samples t-tests. n = 11. *p 0.05; **p 0.01; ***p 0.001.
Social interaction decreases stress responsiveness during adolescence 46. This effect dissipated over time as well and there was no difference between the experimental and the control group during adulthood (Mathews et al., 2008). In another study using the same paradigm (McCormick et al., 2008), adolescent male rats that had experienced isolation and social instability had a lower stress response to confinement to the open arms of an elevated plus maze compared with controls 2—4 h after the last isolation. Still, it is likely that this effect was not due to the history of social instability, but to the acute influence of the last isolation procedure, since rats that had experienced 1 h of isolation only before the confinement also exhibited this reduction in corticosterone responsiveness. One might argue that this could also be the case in the present study; however, there are two facts that speak against such an interpretation: (1) There was a time interval of at least 24 h between the last stimulation and the stress response test (as in Mathews et al., 2008). (2) In our ¨rzel et al., 2010), the last stimulation took previous study (Lu place at about the same time before the stress response test, and no significant difference between the groups was found. In summary, no consistent conclusions can be drawn from the few studies available up to now. For a comprehensive understanding of the relationship between social interactions during adolescence and stress responsiveness, more studies in different species are urgently needed.
4.2. Testosterone as a mediator of reduced stress responsiveness An enhancing effect of social interactions on T release has been described repeatedly in a variety of species (Harding, 1981; Goymann et al., 2003; Hirschenhauser and Oliveira, 2006; van der Meij et al., 2008) including guinea pigs (Sachser and Pro ¨rzel et al., 2010). In the present study, T ¨ve, 1984; Lu response values were significantly higher in SM than in PM on three out of five occasions that were monitored. Elevation of T levels due to social stimulation is somewhat inconsistent, which may be due to habituation effects or to individual differences between stimulus as well as experimental animals. It is possible that a longer stimulation time might have led to more regular elevations in T concentrations. However, the overall effect as seen in the highly significant main effect of group in the ANOVA indicates that T increased in SM after social stimulation as expected. In addition, SM had higher basal levels than PM on day 86. This difference might be explained by an influence of the preceding three social stimulation sessions. Note, however, that social stimulation did not lead to a difference between groups in basal T levels on the other days; thus, the influence of social stimulation on basal T values does not appear to be substantial. As to the reduction in C secretion found in SM, one might argue that social stimulation acts as a stressor and that the decrease in stress responsiveness is an effect of habituation or desensitization. This, however, seems unlikely since social encounters and non-social novelty are heterotypic stressors, and no habituation should occur (Armario et al., 2004). We favor an alternative explanation: the reduction in C secretion found in SM may be due to an inhibiting effect of T. Studies in rats indicate that high circulating levels of T reduce basal corticosterone levels as well as HPA responsiveness (Seale et al., 2004a,b). In guinea pigs, inhibiting effects of T on HPA
1375
function have been long known as well: castration of juvenile males leads to increased circulating C levels 30 days later (El Hani et al., 1980). However, an activational effect of T on stress responsiveness seems very unlikely in the present study, since no significant difference in basal T levels existed between PM and SM on the days of stress response tests. Rather, we hypothesize an organizational effect: different T levels over the course of adolescence may have acted at central nervous sites to organize HPA responsiveness. The mechanism of such an organization during adolescence is unknown. There are, however, similar organizational effects of T on the HPA axis during the pre- and early postnatal phase, which have been thoroughly investigated. They seem to be mediated mainly via androgen receptors (ARs) (McCormick and Mahoney, 1999), but aromatization and involvement of estrogen receptors (ERs) appear to play an important role as well (McCormick et al., 1998; Seale et al., 2005b). Potential sites of action include the PVN due to Tdependent increases in glucocorticoid receptor mRNA levels (Seale et al., 2005a), which enhance glucocorticoid negative feedback. Since the neurons of the PVN that project to the anterior pituitary do not express ARs (Bingham et al., 2006), other brain regions must control activation of the PVN. Prime candidates are the bed nuclei of the stria terminalis and the medial nucleus of the amygdala, since these areas express ARs and ERs, are highly sensitive to adult T levels, and project to the PVN (Bingham and Viau, 2008). A very recent study (Evuarherhe et al., 2009) found organizational effects of T on HPA function during adolescence as well: the corticosterone response to an immunological challenge with lipopolysaccharide was higher in prepubertally gonadectomized male rats than in rats that had been gonadectomized during adulthood, irrespective of whether or not T was replaced five days before testing. Furthermore, HPA responsiveness was decreased by T replacement in males gonadectomized during adulthood, but not in males gonadectomized before puberty, which may indicate a reduced sensitivity to T caused by prepubertal gonadectomy. These findings might actually explain the differences between guinea pigs in the present study. Although basal T levels are not significantly different during stress response tests, PM might react with higher C levels than SM because they are not as sensitive to the acute inhibiting effects of Ton the day of testing due to low levels of T during the organization of the HPA axis in adolescence. However, since the evidence presented in the present study is merely correlative, experiments involving manipulation of T levels are currently being undertaken to test for a causal relationship.
4.3. Biological function of reduced cortisol responsiveness Late adolescence is a crucial life stage for male guinea pigs, since it is an important time window for behavioral development and acquisition of social skills. If kept in large, mixed-sex colonies, male guinea pigs experience a distinct period of late adolescence: They attain sexual maturity at about 2—3 months of age but are prevented from mating by older, more dominant males that already have monopolized the present females. It is during this time that they will learn to assess a potential opponent’s fighting abilities, to show
1376 submissive behavior when appropriate, and to respect other males’ ‘ownership’ of females (Sachser, 1993; Sachser et al., 1994). In this scenario, young males will basically adapt a ‘queuing strategy’ and wait for an opportunity to monopolize females and mate, which usually occurs at an age of 6—7 months. On the other hand, males that lived in pair-housing with one female during adolescence had no opportunity to learn these social rules, since only mild agonistic behavior occurs between the sexes. Thus, they will react with high amounts of aggressive behavior towards other males as well as courtship and sexual behavior towards unfamiliar females (Sachser and Lick, 1991), making integration into a mixed-sex colony nearly impossible. Nevertheless, they are most likely well adapted for a life with few social encounters, defending their female against potential rivals, which could be described in terms of a ‘resource defense’ strategy (Sachser et al., 2010). It has been argued that the social modulation of the stress response in male guinea pigs may represent an effective mechanism to facilitate the development of the aforementioned behavioral phenotypes via an influence on aggressive behavior (Sachser et al., 2010). A sudden increase in glucocorticoid levels enhances aggressive behavior in rats (Mikics et al., 2004) and hamsters (Hayden-Hixson and Ferris, 1991) via non-genomic actions. Possibly, colony-housed males’ low stress responsiveness might predispose them to interact peacefully and submissively with other males in order to wait and gain a higher rank later without the cost of escalated aggression, whereas the high stress response of pairhoused males provides them with a behavioral strategy to aggressively defend their females. Whether acute C responsiveness indeed influences aggressive behavior in guinea pigs is currently under investigation.
5. Conclusion This study delivers conclusive evidence that the reduced stress responsiveness observed in late adolescent male guinea pigs living in large mixed-sex colonies is causally related to a high amount of social interactions. Furthermore, social interactions with adult animals of both sexes increase T secretion, which agrees with the assumption that the reduced stress responsiveness may be mediated via elevated T levels over the course of adolescence.
Role of the funding source This research was funded by the German Research Foundation with a grant to N.S. (FOR 1232/Sa389/11-1) and by a stipend of the Otto Creutzfeld Center for Cognitive and Behavioral Neuroscience to S.L. Funding sources had no further role in study design; in the collection, analysis and interpretation of data; in the writing of the report; and in the decision to submit the paper for publication.
Conflict of interest The authors declare that they have no conflicts of interest.
S. Lu ¨rzel et al.
Acknowledgement We would like to thank Sabine Kruse and Tanja Mo ¨llers for conducting the endocrine analyses.
References Adriani, W., Laviola, G., 2000. A unique hormonal and behavioral hyporesponsivity to both forced novelty and D-amphetamine in periadolescent mice. Neuropharmacology 39, 334—346. Armario, A., Valles, A., Dal-Zotto, S., Marquez, C., Belda, X., 2004. A single exposure to severe stressors causes long-term desensitisation of the physiological response to the homotypic stressor. Stress 7 (3), 157—172. Bingham, B., Viau, V., 2008. Neonatal gonadectomy and adult testosterone replacement suggest an involvement of limbic arginine vasopressin and androgen receptors in the organization of the hypothalamic—pituitary—adrenal axis. Endocrinology 149 (7), 3581—3591. Bingham, B., Williamson, M., Viau, V., 2006. Androgen and estrogen receptor-beta distribution within spinal-projecting and neurosecretory neurons in the paraventricular nucleus of the male rat. J. Comp. Neurol. 499 (6), 911—923. Blakemore, S.-J., 2008. The social brain in adolescence. Nat. Rev. Neurosci. 9, 267—277. Dahl, R.E., 2004. Adolescent brain development: a period of vulnerabilities and opportunities. Ann. N.Y. Acad. Sci. 1021, 1—22. El Hani, A., Dalle, M., Delost, P., 1980. Role of testosterone in the sexual dimorphism of adrenal activity at puberty in the guineapig. J. Endocrinol. 87, 455—461. Evuarherhe, O., Leggett, J.D., Waite, E.J., Kershaw, Y.M., Atkinson, H.C., Lightman, S.L., 2009. Organizational role for pubertal androgens on adult hypothalamic—pituitary—adrenal sensitivity to testosterone in the male rat. J. Physiol. 587 (12), 2977—2985. Ferris, C.F., Messenger, T., Sullivan, R., 2005. Behavioral and neuroendocrine consequences of social subjugation across adolescence and adulthood. Front. Zool. 2, 7. Goymann, W., East, M.L., Hofer, H., 2003. Defense of females, but not social status, predicts plasma androgen levels in male spotted hyenas. Physiol. Biochem. Zool. 76 (4), 586—593. Goymann, W., Landys, M.M., Wingfield, J.C., 2007. Distinguishing seasonal androgen responses from male—male androgen responsiveness — revisiting the challenge hypothesis. Horm. Behav. 51 (4), 463—476. Harding, C.F., 1981. Social modulaton of circulating hormone levels in the male. Am. Zool. 21 (1), 223—231. Hayden-Hixson, D.M., Ferris, C.F., 1991. Steroid-specific regulation of agonistic responding in the anterior hypothalamus of male hamsters. Physiol. Behav. 50, 793—799. Hennessy, M.B., Hornschuh, G., Kaiser, S., Sachser, N., 2006. Cortisol responses and social buffering: a study throughout the life span. Horm. Behav. 49, 383—390. Hirschenhauser, K., Oliveira, R.F., 2006. Social modulation of androgens in male vertebrates: meta-analyses of the challenge hypothesis. Anim. Behav. 71, 265—277. Holm, S., 1979. A simple sequentially rejective multiple test procedure. Scand. J. Stat. 6 (2), 65—70. Isgor, C., Kabbaj, M., Akil, H., Watson, S.J., 2004. Delayed effects of chronic variable stress during peripubertal—juvenile period on hippocampal morphology and on cognitive and stress axis functions in rats. Hippocampus 14 (5), 636—648. Kaiser, S., Haderthauer, S., Sachser, N., Hennessy, M.B., 2007. Social housing conditions around puberty determine later changes in plasma cortisol levels and behavior. Physiol. Behav. 90, 405—411. Kaiser, S., Kirtzeck, M., Hornschuh, G., Sachser, N., 2003. Sexspecific difference in social support — a study in female guinea pigs. Physiol. Behav. 79, 297—303.
Social interaction decreases stress responsiveness during adolescence Kemme, K., Kaiser, S., von Engelhardt, N., Wewers, D., Groothuis, T., Sachser, N., 2009. An unstable social environment affects sex ratio in guinea pigs: an adaptive maternal effect? Behaviour 146, 1513—1529. Lu ¨rzel, S., Kaiser, S., Sachser, N., 2010. Social interaction, testosterone, and stress responsiveness during adolescence. Physiol. Behav. 99, 40—46. Mathews, I.Z., Wilton, A., Styles, A., McCormick, C.M., 2008. Increased depressive behaviour in females and heightened corticosterone release in males to swim stress after adolescent social stress in rats. Behav. Brain Res. 190 (1), 33—40. McCormick, C.M., Furey, B.F., Child, M., Sawyer, M.J., Donohue, S.M., 1998. Neonatal sex hormones have ‘organizational’ effects on the hypothalamic—pituitary—adrenal axis of male rats. Dev. Brain Res. 105, 295—307. McCormick, C.M., Mahoney, E., 1999. Persistent effects of prenatal, neonatal, or adult treatment with flutamide on the hypothalamic—pituitary—adrenal stress response of adult male rats. Horm. Behav. 35 (1), 90—101. McCormick, C.M., Mathews, I.Z., 2007. HPA function in adolescence: role of sex hormones in its regulation and the enduring consequences of exposure to stressors. Pharmacol. Biochem. Behav. 86, 220—233. McCormick, C.M., Mathews, I.Z., 2010. Adolescent development, hypothalamic—pituitary—adrenal function, and programming of adult learning and memory. Prog. Neuropsychopharmacol. Biol. Psychiatry 34, 756—765. McCormick, C.M., Smith, C., Mathews, I.Z., 2008. Effects of chronic social stress in adolescence on anxiety and neuroendocrine response to mild stress in male and female rats. Behav. Brain Res. 187, 228—238. ´., Kruk, M.R., Haller, J., 2004. Genomic and non-genomic Mikics, E effects of glucocorticoids on aggressive behavior in male rats. Psychoneuroendocrinology 29, 618—635. Sachser, N., 1990. Social organization, social status, behavioural strategies and endocrine responses in male guinea pigs. In: Balthazart, J. (Ed.), Hormones, Brain and Behaviour in Vertebrates. 2. Behavioural Activation in Males and Females — Social Interaction and Reproductive Endocrinology. Comp. Physiol., vol. 9. Karger, Basel, pp. 176—187. Sachser, N., 1993. The ability to arrange with conspecifics depends on social experiences around puberty. Physiol. Behav. 53, 539—544. Sachser, N., 1994. Sozialphysiologische Untersuchungen an Hausmeerschweinchen. Paul Parey, Berlin. Sachser, N., Hennessy, M.B., Kaiser, S., 2010. Adaptive modulation of behavioural profiles by social stress during early phases of life and adolescence. Neurosci. Biobehav. Rev., doi:10.1016/ j.neubiorev.2010.09.002.
1377
Sachser, N., Lick, C., 1991. Social experience, behavior, and stress in guinea pigs. Physiol. Behav. 50, 83—90. Sachser, N., Lick, C., Stanzel, K., 1994. The environment, hormones, and aggressive behavior — a 5-year-study in guinea pigs. Psychoneuroendocrinology 19 (5—7), 697—707. Sachser, N., Pro ¨ve, E., 1984. Short-term effects of residence on the testosterone responses to fighting in alpha male guinea pigs. Aggressive Behav. 10, 285—292. Sachser, N., Pro ¨ve, E., 1988. Plasma-testosterone development in colony and individually housed male guinea pigs. Ethology 79, 62—70. Seale, J.V., Wood, S.A., Atkinson, H.C., Bate, E., Lightman, S.L., Ingram, C.D., Jessop, D.S., Harbuz, M.S., 2004a. Gonadectomy reverses the sexually diergic patterns of circadian and stressinduced hypothalamic—pituitary—adrenal axis activity in male and female rats. J. Neuroendocrinol. 16 (6), 516—524. Seale, J.V., Wood, S.A., Atkinson, H.C., Harbuz, M.S., Lightman, S.L., 2004b. Gonadal steroid replacement reverses gonadectomyinduced changes in the corticosterone pulse profile and stressinduced hypothalamic—pituitary—adrenal axis activity of male and female rats. J. Neuroendocrinol. 16, 989—998. Seale, J.V., Wood, S.A., Atkinson, H.C., Harbuz, M.S., Lightman, S.L., 2005a. Postnatal masculinization alters the HPA axis phenotype in the adult female rat. J. Physiol. 563 (1), 265—274. Seale, J.V., Wood, S.A., Atkinson, H.C., Lightman, S.L., Harbuz, M.S., 2005b. Organizational role for testosterone and estrogen on adult hypothalamic—pituitary—adrenal axis activity in the male rat. Endocrinology 146 (4), 1973—1982. Sisk, C.L., Zehr, J.L., 2005. Pubertal hormones organize the adolescent brain and behavior. Front. Neuroendocrinol. 26, 163—174. Spear, L.P., 2000. The adolescent brain and age-related behavioral manifestations. Neurosci. Biobehav. Rev. 24, 417—463. Steinberg, L., 2005. Cognitive and affective development in adolescence. Trends Cognit. Sci. 9 (2), 69—74. van der Meij, L., Buunk, A.P., van de Sande, J.P., Salvador, A., 2008. The presence of a woman increases testosterone in aggressive dominant men. Horm. Behav. 54, 640—644. Williams, B.M., Luo, Y., Ward, C., Redd, K., Gibson, R., Kuczaj, S.A., McCoy, J.G., 2001. Environmental enrichment: effects on spatial memory and hippocampal CREB immunoreactivity. Physiol. Behav. 73, 649—658. Wingfield, J.C., Hegner, R.E., Dufty Jr., A.M., Ball, G.F., 1990. The ‘‘challenge hypothesis’’: theoretical implications for patterns of testosterone secretion, mating systems, and breeding strategies. Am. Nat. 136 (6), 829—846. Yurgelun-Todd, D., 2007. Emotional and cognitive changes during adolescence. Curr. Opin. Neurobiol. 17, 251—257.
a v a i l a b l e a t w w w. s c i e n c e d i r e c t . c o m
j o u r n a l h o m e p a g e : w w w. e l s e v i e r. c o m / l o c a t e / p s y n e u e n
Social interaction decreases stress responsiveness during adolescence Stephanie Lu ¨rzel a,b,*, Sylvia Kaiser a,b, Norbert Sachser a,b a b
Department of Behavioural Biology, University of Mu ¨nster, Badestraße 13, 48149 Mu ¨nster, Germany Otto Creutzfeldt Center for Cognitive and Behavioural Neuroscience, Badestraße 13, 48149 Mu ¨nster, Germany
Received 20 December 2010; received in revised form 24 February 2011; accepted 18 March 2011
KEYWORDS HPA axis; Cortisol; Adolescence; Social experience; Testosterone; Organizational effect
Summary Adolescence is the transition from infancy to adulthood and encompasses major changes in the brain, the endocrine systems, and behavior. During late adolescence, male guinea pigs living in mixed-sex colonies exhibit a lower cortisol (C) response to novelty compared with animals in other ages and housing conditions. It was hypothesized that this reduction in stress responsiveness is induced by a high amount of social interactions in the colonies. In a previous ¨rzel et al., 2010), late adolescent colony-housed males (CM) were compared with study (Lu similarly aged males that were housed in heterosexual pairs (PM) as well as with males that were also housed in pairs, but regularly received additional social stimulation by allowing them ten times to interact with unfamiliar adult animals of both sexes for 10 min (SM). CM had a significantly lower stress response than PM, with SM being intermediate and not significantly different from either of the other groups. We assumed that the amount of social stimulation in SM was insufficient in order to achieve a significant reduction of stress responsiveness compared with PM. For the present study, we hypothesized that with a higher amount of social stimulation, a significant difference in stress responsiveness between PM and SM becomes apparent during late adolescence. Thus, PM were again compared with SM that, this time, had received twice as much social stimulation as in the previous study. As a result, stress responsiveness was indeed significantly lower in SM than in PM during late adolescence. Thus, a high amount of social interactions during the course of adolescence leads to a decreased stress responsiveness. Furthermore, SM showed an increase in testosterone (T) levels caused by social stimulation. We hypothesize that the reduction in stress responsiveness is brought about by high T levels that organize central neural structures over the course of adolescence. # 2011 Elsevier Ltd. All rights reserved.
1. Introduction * Corresponding author at: Department of Behavioural Biology, University of Mu ¨nster, Badestraße 13, 48149 Mu ¨nster, Germany. Tel.: +49 251 83 21004; fax: +49 251 83 23896. E-mail address: [email protected] (S. Lu ¨rzel).
Adolescence is the gradual transition from infancy to adulthood and, as such, characterized not only by the attainment of sexual maturity, but also by a variety of changes in anatomy, endocrine systems, neural circuitry, as well as
0306-4530/$ — see front matter # 2011 Elsevier Ltd. All rights reserved. doi:10.1016/j.psyneuen.2011.03.010
Social interaction decreases stress responsiveness during adolescence behavior (Sisk and Zehr, 2005; Yurgelun-Todd, 2007; Blakemore, 2008). Due to these changes, adolescence represents a period of high vulnerability to adverse environmental conditions (Dahl, 2004; McCormick and Mathews, 2010). On the other hand, it is also a window of opportunity for adaptation to the current environment (Williams et al., 2001; Dahl, 2004; Steinberg, 2005; McCormick et al., 2008), similar to the sensitive phases during the pre- and early postnatal period (for a review, see Sachser et al., 2010). Among the alterations in endocrine systems during adolescence, stress responsiveness seems to play an important role. Individuals of various species undergo an upregulation of the hypothalamo—pituitary—adrenal (HPA) axis during adolescence (Spear, 2000; McCormick and Mathews, 2010), but in some species, the contrary is the case. For instance, the adult-typical HPA response to novelty is reduced in periadolescent mice (Adriani and Laviola, 2000), and periadolescent rats have a lower, although prolonged, corticosterone response compared with adult males (McCormick et al., 2008). In a study conducted throughout the life span of male guinea pigs, Hennessy et al. (2006) found a similar cortisol (C) hyporesponsiveness during late adolescence. A follow-up study revealed that this phenomenon does not exist under all environmental conditions, but depends on the amount and nature of social experiences made during this phase of life (Kaiser et al., 2007): Late adolescent male guinea pigs housed in a mixed-sex colony display a significantly lower increase of C levels in a novel environment than equally aged males kept only with a female since weaning. The mechanisms that lead to a reduced C responsiveness are only partially understood. One possible explanation is the inhibiting effect of testosterone (T) on C secretion. In addition to activational effects, organizational effects of Ton HPA activity during the perinatal phase (reviewed in McCormick and Mathews, 2007) are well known. A recent study in rats delivered evidence for organizational effects occurring during adolescence as well (Evuarherhe et al., 2009). A modulation of HPA function by T has also been discussed for guinea pigs, since colony-housed males exhibit a peak in T levels during adolescence (Sachser and Pro ¨ve, 1988) and have higher T levels than pair-housed males in adulthood (Sachser, 1990). This peak and subsequent higher T levels are most likely due to the constant challenge posed by other males and to the presence of females in the colony, resulting in a high amount of social interactions (Wingfield et al., 1990; Goymann et al., 2007). Thus, a very recent study investigated whether the difference in stress responsiveness between colony-housed and pair-housed males is indeed due to differences in the amount of social interactions experienced during adolescence (Lu ¨rzel et al., 2010). For these purposes, a third group was established in addition to pair-housed and colonyhoused males. This group was kept under pair-housing conditions but received additional social stimulation by allowing the males 10 times to interact with unfamiliar conspecifics of both sexes for 10 min. As a result, C responsiveness of these males was intermediate between colony- and pair-housed males. The social stimulation paradigm failed, however, to bring about a significantly reduced stress response compared to pair-housed males. Based on these data, the authors argued that a higher amount of social stimulation might lead to a stress response that is significantly reduced in comparison with pair-housed males without additional stimulation.
1371
Following this reasoning, the amount of social stimulation was doubled in this study, and the socially stimulated males were again compared with pair-housed males without additional social stimulation. Social stimulation was hypothesized to lead to a significantly reduced stress responsiveness during late adolescence. Furthermore, it was predicted that social stimulation triggers an acute elevation of T levels, which would be in accordance with the challenge hypothesis (Wingfield et al., 1990; Goymann et al., 2007).
2. Methods 2.1. Animals and housing conditions The guinea pigs (Cavia aperea f. porcellus) used were descendents of a heterogeneous shorthaired and multicolored stock of 40 animals obtained from a breeder in 1975. Experimental animals were derived from two mixed-sex colonies housed in the same room. The colonies consisted of 8—10 males and 13—15 females as well as their offspring, and showed a graduated age structure ranging from approximately 1 to 19 months. All animals could be individually identified by natural markings and were housed under controlled conditions: 12:12 LD (lights on 0700 h), temperature 22 2 8C, relative humidity about 60%. Commercial guinea pig diet (Ho ¨veler ‘‘Spezialfutter’’ 1070 for guinea pigs, Ho ¨veler Spezialfutterwerke GmbH&C. KG, Langenfeld, Germany) and water were available ad libitum. This diet was supplemented with hay and straw. Vitamin C was provided in the water twice per week. Enclosures were cleaned once per week. All experiments were announced to the competent local authority and were approved by the ‘‘Tierschutzbeauftragter’’ of the University of Muenster. Experiments were carried out in accordance with the European Communities Council Directive of 24 November 1986 (86/609/EEC). All efforts were made to minimize animal suffering and to reduce the number of animals used.
2.2. Experimental design The experimental design is depicted in Figure 1. All experimental animals were born in large mixed-sex colonies (day 0). After weaning, but before sexual maturation, experimental animals were housed with a single female of the same approximate age in 0.5 m2 enclosures (day 30). The females used as partners were colony-born and had been housed in all-female groups since the time of weaning (day 21 of age) to prevent pregnancy. Experimental animals were randomly assigned to one of two groups: pair-housed males (PM; n = 12) remained with only their female partner, whereas the other group received additional social stimulation from day 80 of life onward (SM; n = 13). No more than one male from each litter was assigned to the same group. In both early adolescence (day 55) and late adolescence (day 120), experimental animals of both groups were exposed to a standardized psychological stressor to assess C responsiveness. All procedures were conducted at specified ages of the animals with a tolerance of 1 d.
1372
Figure 1 Experimental design. PM: pair-housed males; SM: pair-housed males with additional social stimulation. From an age of 30 days onwards, PM and SM were kept with one female only. From day 80 to day 118, stimulus animals were added to SM every other day. Blood samples were taken for five social stimulation sessions from SM, and from PM for control procedures on the corresponding days. On days 55 and 120, stress response tests (SRTs) were conducted with all animals.
2.3. Social stimulation and assessment of testosterone concentrations For social stimulation, SM were given the opportunity to interact with unfamiliar animals of both sexes every second day 1 d (SM never experienced more than one social stimulation session per day). The time of day at which social stimulation occurred was varied in order to avoid possible habituation effects by introducing unpredictability. Social stimulation took place in the home cage of SM in the presence of their female partner. Since female guinea pigs attain sexual maturity at about 30 days of age, and males at about 60 days of age (Kemme et al., 2009), it is safe to assume that the females were pregnant during the time of social stimulation, preventing a confounding influence of oestrus. The period of social stimulation lasted from day 80 to day 118, which adds up to 20 stimulation sessions, 15 sessions with a male and 5 sessions with a female stimulus animal that were evenly distributed throughout the social stimulation period. Altogether, six stimulus males were used according to a pseudo-randomized schedule that had been living in the colonies for 11—19 months and were then rehoused with a younger male. Each stimulus male was used a maximum of three times with each experimental animal, with an interval of at least 14 days between encounters of the same animals. Thus, in the second and possible third encounter, the stimulus males were not entirely unfamiliar, but we can reasonably assume that they were not familiar in the sense of a stable dominance relationship, especially since agonistic behavior was not notably diminished. The 34 stimulus females drawn from the colonies were always pregnant at the time of testing to preclude any influence of oestrus. They were never used more than once with the same experimental animal. The stimulus animal remained in the enclosure for 10 min. During the stimulations, the experimenter observed the animals in order to separate them if a fight escalated. Out of a total of 165 stimulation sessions using males as stimulus animals, six had to be cut short because aggression escalated. Behavior during social stimulations was not recorded quantitatively since it has been shown in numerous studies that male guinea pigs usually display agonistic behavior towards
S. Lu ¨rzel et al. each other and courtship and sexual behavior towards females (e.g., Sachser et al., 1994; Kaiser et al., 2007; Lu ¨rzel et al., 2010). On the other hand, the cases in which they show agonistic behavior towards females as well as sociopositive behaviors towards males are very rare. These findings also hold true for the present study (qualitative analysis). On days 80, 86, 98, 110, and 114, two blood samples were taken, one directly before the stimulation (0 h) and another one 1 h later, in order to determine basal T values as well as changes in T concentration. Social stimulation sessions involving blood sampling were started at 1300 h 15 min to minimize confounding effects caused by diurnal variation. Blood samples were also taken from PM at corresponding times.
2.4. Assessment of cortisol responsiveness PM and SM were tested for their stress responsiveness alone in a novel enclosure in another guinea pig housing room for 2 h on days 55 and 120. The enclosure was a 1 m2 wooden box that contained food and water as well as clean wood shavings for bedding. A novel environment has been shown to act as a psychological stressor in guinea pigs, causing an increase in C levels (e.g., Hennessy et al., 2006). All tests were started at 1300 h 15 min. At the beginning of each test, blood samples were taken to assess basal Tand C concentrations. After that, the animal was introduced into the unfamiliar environment. At 60 min and 120 min, subsequent blood samples were taken to determine changes in C concentrations. After the third blood sample, the animal was placed back into his home enclosure.
2.5. Blood sample collection and hormone determination Blood samples were collected from the blood vessels of the ears. A muscle salve (Finalgon, Boehringer Ingelheim Pharma KG, Ingelheim am Rhein, Germany) was applied to the ear to stimulate circulation and the vessels were illuminated with a cold-point lamp. Vessels were pricked with an injection needle and about 0.4 ml of blood was collected in heparinized capillary tubes. One experimenter held the male in his/her lap, while a second collected the sample. Guinea pigs show little struggling during the collection procedure, and no elevation of plasma C or T levels occurs for about 6 min (Sachser, 1994). Accordingly, all samples for determination of C levels were collected within 3 min, all samples for determination of T levels within 6 min of entering the room. Because no anesthesia is required, hormone levels in the second and third samples were not influenced by previous exposure to anesthesia. Plasma was separated by centrifugation (11,700 g for 5 min) and deep frozen ( 20 8C) until assayed. Plasma C concentrations were determined in duplicate by radioimmunoassay without chromatography using specific antibodies against C (Biotrend, Ko ¨ln, Germany) as in earlier work (e.g., Kaiser et al., 2003). The antibodies used cross-reacted with relevant steroids as follows: C 100%, prednisolone 36%, 11-deoxyC 10%, corticosterone 3.2%, cortisone 0.9%. The intra- and interassay coefficients of variation (%CV) were 4.8 and 8.7%, respectively.
Social interaction decreases stress responsiveness during adolescence
1373
Figure 2 Plasma testosterone [ng/ml] before (A) and after (B) social stimulation. Blood samples were taken immediately before 10 min of social stimulation and 1 h later. Indicated are mean and SEM. PM: pair-housed males; SM: pair-housed males with additional social stimulation. Stimulus animals were males on days 80, 98, and 114, and females on days 86 and 110. Statistics: mixed factorial ANOVA (A: main effect of age, p = 0.002; interaction of age and group, p = 0.002; B: main effect of group, p = 0.002; interaction of age and group, p = 0.006), post hoc testing: independent-samples t-tests. nPM = 11; nSM = 8—10. *p 0.05; **p 0.01; ***p 0.001.
Plasma T concentrations were determined using a solid phase enzyme-linked immunosorbent assay (ELISA; Testosterone ELISA Kit, Demeditec Diagnostics GmbH, Kiel, Germany). The antibody used cross-reacted with relevant steroids as follows: T 100%, 11b-hydroxytestosterone 3.3%, 19-nortestosterone 3.3%, androstenedione 0.9%, 5a-dihydrotestosterone 0.8%, 17a-methyltestosterone 0.1%, epitestosterone, oestradiol, progesterone, C, oestrone and danazol <0.1% each. The intraassay %CV was 5.7%, the interassay %CV was 7.2%.
2.6. Data analysis After confirming that none of the data sets deviated from a normal distribution with a one-sample Kolmogorov—Smirnov test, mixed factorial analyses of variance (ANOVAs) were calculated, since differences between dependent as well as independent samples had to be evaluated. Greenhouse—Geisser corrections were applied if appropriate. If ANOVAs yielded significant interactions, they were followed by pairwise comparisons using independent-samples and paired-samples t-tests, respectively. In cases of multiple comparisons, the sequential Bonferroni—Holm method (Holm, 1979) was applied. On the basis of our previous work (Kaiser et al., 2007; Lu ¨rzel et al., 2010), t-tests were 2-tailed for basal hormone concentrations and 1-tailed for increases in hormone concentrations. Data were analyzed using PASW Statistics 18 (IBM, New York, USA). The significance level is 0.05. Graphs show mean and standard error of the mean and were created using SigmaPlot 11 (Systat Software GmbH, Erkrath, Germany). Several animals were dropped from analysis of hormones due to incomplete blood samples. Final sample sizes are given in Section 3.
3. Results 3.1. Testosterone response to social stimulation For T basal values (Figure 2A) taken immediately before 10 min of social stimulation, ANOVA revealed a main effect of age (nPM = 11; nSM = 10; F(4, 76) = 4.677, p = 0.002) as well
as an interaction of age and group (F(4, 76) = 4.784, p = 0.002). The subsequent t-tests revealed significantly higher basal T values in socially stimulated males (SM) than in pair-housed males (PM) on day 86 (n = 11; t = 3.564, p = 0.002). Most importantly, there was a significant main effect of group on T response values (Figure 2B; nPM = 11; nSM = 8; ANOVA: F(1, 17) = 13.147, p = 0.002) and an interaction of age and group (F(2.263, 38.006) = 5.561, p = 0.006). t-tests found higher response values in SM than in PM on days 80, 86, and 110 (n = 11; day 80: t = 2.513, p = 0.013; day 86: t = 3.930, p = 0.001; day 110: t = 3.623, p = 0.001). Basal T levels on the days of stress response tests — that is, before and after the period of social stimulation — differed neither between groups (n = 11; ANOVA: F(1, 20) = 0.089, n.s.) nor between day 55 and day 120 (F(1, 20) = 3.374, n.s.), nor was there any interaction of age and group (F(1, 20) = 0.106, n.s.).
3.2. Consequences of social stimulation on cortisol responsiveness On days of stress response tests, basal C concentrations did not differ between PM and SM (Figure 3; n = 11; ANOVA: F(1, 20) = 3.541, n.s.), and there was no interaction of age and group (F(1,20) = 0.117, n.s.). However, ANOVA revealed a main effect of age, with values being slightly, but significantly higher on day 120 than on day 55 (F(1, 20) = 5.322, p = 0.032). Concerning C responsiveness, there is a main effect of group as well as an interaction of age and group for the increase in C levels 1 h after the onset of the stressor (Figure 4A; n = 11; ANOVA: main effect on group: F(1, 20) = 5.987, p = 0.024, interaction age group: F(1, 20) = 4.382, p = 0.049). SM have a significantly lower stress response on day 120 than on day 55 (t-test: t = 2.185, p = 0.027), whereas values of PM do not differ between the different points in time (t-test: t = 0.888, n.s.). As a consequence, SM have a lower stress response than PM on day 120 (t-test: t = 3.556, p = 0.001), but not on day 55 (t-test: t = 0.932, n.s.). Analysis of the increase in C levels 2 h after
1374
Figure 3 Basal plasma cortisol [ng/ml]. Indicated are mean and SEM. PM: pair-housed males; SM: pair-housed males with additional social stimulation. Statistics: mixed factorial ANOVA. n = 11. *p 0.05.
the onset of the stressor yields similar results: there is a main effect of group as well as an interaction of age and group (Figure 4B; n = 11; ANOVA: main effect on group: F(1, 20) = 6.011, p = 0.024, interaction age group: F(1, 20) = 8.994, p = 0.007). Again, the stress response of SM is significantly decreased on day 120 compared with day 55 (ttest: t = 2.534, p = 0.015), whereas it does not differ in PM between the two testing days (t-test: t = 1.646, n.s.), leading to SM having a lower stress response than PM on day 120 (t-test: t = 3.843, p < 0.001), but not on day 55 (ttest: t = 0.320, n.s.).
4. Discussion 4.1. Social stimulation determines HPA responsiveness The major aim of this study was to investigate the effects of social interactions on stress responsiveness during late adolescence. To this purpose, male guinea pigs kept under pair-
S. Lu ¨rzel et al. housing conditions with one female only (PM) were compared with males that lived with one female and received additional social stimulation by interactions with unfamiliar adult males and females (SM). In a previous study with a similar experimental design (Lu ¨rzel et al., 2010), ten occasions of social stimulation had not sufficed to reduce stress responsiveness to a significant extent in SM, but the results had allowed for the assumption that with higher amounts of social stimulation, a significant difference in stress responsiveness between the two groups might be induced. In the present study, the amount of social stimulation was doubled, and, as predicted, SM had a significantly reduced stress response compared with PM during late adolescence. Furthermore, C basal values increased slightly but significantly over the course of adolescence, which most likely reflects developmental changes, as there was neither a main effect of group nor an interaction of group and age. From our study, it cannot be concluded whether the reduction of stress responsiveness is mainly due to stimulation with males (eliciting agonistic behavior) or with females (triggering courtship and sexual behavior), or whether the interaction with stimulus animals of both sexes is necessary in order to reduce HPA reactivity. Nevertheless, this study shows that the causal factor underlying the reduced stress responsiveness is indeed a high amount of social interactions with adult conspecifics, since the two groups experienced otherwise identical housing conditions. In other species, effects of social interactions during adolescence on HPA responsiveness have been shown as well: Male hamsters that underwent a social subjugation paradigm during adolescence showed reduced C response values after an aggressive encounter in adulthood compared with control males (Ferris et al., 2005). In another study, adolescent male rats lived in groups of four animals but experienced various social conditions from isolation to crowding for short amounts of time between day 28 and day 56 of age. These males had a prolonged corticosterone response compared with controls 24 h after the last social change, but not three weeks later (Isgor et al., 2004). Male rats that had experienced 1 h isolation daily followed by change of cage partner from day 30 to day 45 of age (mid-adolescence) showed an enhanced corticosterone release after swim stress on day
Figure 4 Increase in plasma cortisol [ng/ml] 1 h (A) and 2 h (B) after the onset of a psychological stressor (novel environment). Indicated are mean and SEM. PM: pair-housed males; SM: pair-housed males with additional social stimulation. Statistics: mixed factorial ANOVA (increase to 1 h: main effect of group, p = 0.024; interaction of age group, p = 0.049; increase to 2 h: main effect of group, p = 0.024; interaction of age group, p = 0.007), post hoc testing: Bonferroni—Holm corrected independent-samples as well as paired-samples t-tests. n = 11. *p 0.05; **p 0.01; ***p 0.001.
Social interaction decreases stress responsiveness during adolescence 46. This effect dissipated over time as well and there was no difference between the experimental and the control group during adulthood (Mathews et al., 2008). In another study using the same paradigm (McCormick et al., 2008), adolescent male rats that had experienced isolation and social instability had a lower stress response to confinement to the open arms of an elevated plus maze compared with controls 2—4 h after the last isolation. Still, it is likely that this effect was not due to the history of social instability, but to the acute influence of the last isolation procedure, since rats that had experienced 1 h of isolation only before the confinement also exhibited this reduction in corticosterone responsiveness. One might argue that this could also be the case in the present study; however, there are two facts that speak against such an interpretation: (1) There was a time interval of at least 24 h between the last stimulation and the stress response test (as in Mathews et al., 2008). (2) In our ¨rzel et al., 2010), the last stimulation took previous study (Lu place at about the same time before the stress response test, and no significant difference between the groups was found. In summary, no consistent conclusions can be drawn from the few studies available up to now. For a comprehensive understanding of the relationship between social interactions during adolescence and stress responsiveness, more studies in different species are urgently needed.
4.2. Testosterone as a mediator of reduced stress responsiveness An enhancing effect of social interactions on T release has been described repeatedly in a variety of species (Harding, 1981; Goymann et al., 2003; Hirschenhauser and Oliveira, 2006; van der Meij et al., 2008) including guinea pigs (Sachser and Pro ¨rzel et al., 2010). In the present study, T ¨ve, 1984; Lu response values were significantly higher in SM than in PM on three out of five occasions that were monitored. Elevation of T levels due to social stimulation is somewhat inconsistent, which may be due to habituation effects or to individual differences between stimulus as well as experimental animals. It is possible that a longer stimulation time might have led to more regular elevations in T concentrations. However, the overall effect as seen in the highly significant main effect of group in the ANOVA indicates that T increased in SM after social stimulation as expected. In addition, SM had higher basal levels than PM on day 86. This difference might be explained by an influence of the preceding three social stimulation sessions. Note, however, that social stimulation did not lead to a difference between groups in basal T levels on the other days; thus, the influence of social stimulation on basal T values does not appear to be substantial. As to the reduction in C secretion found in SM, one might argue that social stimulation acts as a stressor and that the decrease in stress responsiveness is an effect of habituation or desensitization. This, however, seems unlikely since social encounters and non-social novelty are heterotypic stressors, and no habituation should occur (Armario et al., 2004). We favor an alternative explanation: the reduction in C secretion found in SM may be due to an inhibiting effect of T. Studies in rats indicate that high circulating levels of T reduce basal corticosterone levels as well as HPA responsiveness (Seale et al., 2004a,b). In guinea pigs, inhibiting effects of T on HPA
1375
function have been long known as well: castration of juvenile males leads to increased circulating C levels 30 days later (El Hani et al., 1980). However, an activational effect of T on stress responsiveness seems very unlikely in the present study, since no significant difference in basal T levels existed between PM and SM on the days of stress response tests. Rather, we hypothesize an organizational effect: different T levels over the course of adolescence may have acted at central nervous sites to organize HPA responsiveness. The mechanism of such an organization during adolescence is unknown. There are, however, similar organizational effects of T on the HPA axis during the pre- and early postnatal phase, which have been thoroughly investigated. They seem to be mediated mainly via androgen receptors (ARs) (McCormick and Mahoney, 1999), but aromatization and involvement of estrogen receptors (ERs) appear to play an important role as well (McCormick et al., 1998; Seale et al., 2005b). Potential sites of action include the PVN due to Tdependent increases in glucocorticoid receptor mRNA levels (Seale et al., 2005a), which enhance glucocorticoid negative feedback. Since the neurons of the PVN that project to the anterior pituitary do not express ARs (Bingham et al., 2006), other brain regions must control activation of the PVN. Prime candidates are the bed nuclei of the stria terminalis and the medial nucleus of the amygdala, since these areas express ARs and ERs, are highly sensitive to adult T levels, and project to the PVN (Bingham and Viau, 2008). A very recent study (Evuarherhe et al., 2009) found organizational effects of T on HPA function during adolescence as well: the corticosterone response to an immunological challenge with lipopolysaccharide was higher in prepubertally gonadectomized male rats than in rats that had been gonadectomized during adulthood, irrespective of whether or not T was replaced five days before testing. Furthermore, HPA responsiveness was decreased by T replacement in males gonadectomized during adulthood, but not in males gonadectomized before puberty, which may indicate a reduced sensitivity to T caused by prepubertal gonadectomy. These findings might actually explain the differences between guinea pigs in the present study. Although basal T levels are not significantly different during stress response tests, PM might react with higher C levels than SM because they are not as sensitive to the acute inhibiting effects of Ton the day of testing due to low levels of T during the organization of the HPA axis in adolescence. However, since the evidence presented in the present study is merely correlative, experiments involving manipulation of T levels are currently being undertaken to test for a causal relationship.
4.3. Biological function of reduced cortisol responsiveness Late adolescence is a crucial life stage for male guinea pigs, since it is an important time window for behavioral development and acquisition of social skills. If kept in large, mixed-sex colonies, male guinea pigs experience a distinct period of late adolescence: They attain sexual maturity at about 2—3 months of age but are prevented from mating by older, more dominant males that already have monopolized the present females. It is during this time that they will learn to assess a potential opponent’s fighting abilities, to show
1376 submissive behavior when appropriate, and to respect other males’ ‘ownership’ of females (Sachser, 1993; Sachser et al., 1994). In this scenario, young males will basically adapt a ‘queuing strategy’ and wait for an opportunity to monopolize females and mate, which usually occurs at an age of 6—7 months. On the other hand, males that lived in pair-housing with one female during adolescence had no opportunity to learn these social rules, since only mild agonistic behavior occurs between the sexes. Thus, they will react with high amounts of aggressive behavior towards other males as well as courtship and sexual behavior towards unfamiliar females (Sachser and Lick, 1991), making integration into a mixed-sex colony nearly impossible. Nevertheless, they are most likely well adapted for a life with few social encounters, defending their female against potential rivals, which could be described in terms of a ‘resource defense’ strategy (Sachser et al., 2010). It has been argued that the social modulation of the stress response in male guinea pigs may represent an effective mechanism to facilitate the development of the aforementioned behavioral phenotypes via an influence on aggressive behavior (Sachser et al., 2010). A sudden increase in glucocorticoid levels enhances aggressive behavior in rats (Mikics et al., 2004) and hamsters (Hayden-Hixson and Ferris, 1991) via non-genomic actions. Possibly, colony-housed males’ low stress responsiveness might predispose them to interact peacefully and submissively with other males in order to wait and gain a higher rank later without the cost of escalated aggression, whereas the high stress response of pairhoused males provides them with a behavioral strategy to aggressively defend their females. Whether acute C responsiveness indeed influences aggressive behavior in guinea pigs is currently under investigation.
5. Conclusion This study delivers conclusive evidence that the reduced stress responsiveness observed in late adolescent male guinea pigs living in large mixed-sex colonies is causally related to a high amount of social interactions. Furthermore, social interactions with adult animals of both sexes increase T secretion, which agrees with the assumption that the reduced stress responsiveness may be mediated via elevated T levels over the course of adolescence.
Role of the funding source This research was funded by the German Research Foundation with a grant to N.S. (FOR 1232/Sa389/11-1) and by a stipend of the Otto Creutzfeld Center for Cognitive and Behavioral Neuroscience to S.L. Funding sources had no further role in study design; in the collection, analysis and interpretation of data; in the writing of the report; and in the decision to submit the paper for publication.
Conflict of interest The authors declare that they have no conflicts of interest.
S. Lu ¨rzel et al.
Acknowledgement We would like to thank Sabine Kruse and Tanja Mo ¨llers for conducting the endocrine analyses.
References Adriani, W., Laviola, G., 2000. A unique hormonal and behavioral hyporesponsivity to both forced novelty and D-amphetamine in periadolescent mice. Neuropharmacology 39, 334—346. Armario, A., Valles, A., Dal-Zotto, S., Marquez, C., Belda, X., 2004. A single exposure to severe stressors causes long-term desensitisation of the physiological response to the homotypic stressor. Stress 7 (3), 157—172. Bingham, B., Viau, V., 2008. Neonatal gonadectomy and adult testosterone replacement suggest an involvement of limbic arginine vasopressin and androgen receptors in the organization of the hypothalamic—pituitary—adrenal axis. Endocrinology 149 (7), 3581—3591. Bingham, B., Williamson, M., Viau, V., 2006. Androgen and estrogen receptor-beta distribution within spinal-projecting and neurosecretory neurons in the paraventricular nucleus of the male rat. J. Comp. Neurol. 499 (6), 911—923. Blakemore, S.-J., 2008. The social brain in adolescence. Nat. Rev. Neurosci. 9, 267—277. Dahl, R.E., 2004. Adolescent brain development: a period of vulnerabilities and opportunities. Ann. N.Y. Acad. Sci. 1021, 1—22. El Hani, A., Dalle, M., Delost, P., 1980. Role of testosterone in the sexual dimorphism of adrenal activity at puberty in the guineapig. J. Endocrinol. 87, 455—461. Evuarherhe, O., Leggett, J.D., Waite, E.J., Kershaw, Y.M., Atkinson, H.C., Lightman, S.L., 2009. Organizational role for pubertal androgens on adult hypothalamic—pituitary—adrenal sensitivity to testosterone in the male rat. J. Physiol. 587 (12), 2977—2985. Ferris, C.F., Messenger, T., Sullivan, R., 2005. Behavioral and neuroendocrine consequences of social subjugation across adolescence and adulthood. Front. Zool. 2, 7. Goymann, W., East, M.L., Hofer, H., 2003. Defense of females, but not social status, predicts plasma androgen levels in male spotted hyenas. Physiol. Biochem. Zool. 76 (4), 586—593. Goymann, W., Landys, M.M., Wingfield, J.C., 2007. Distinguishing seasonal androgen responses from male—male androgen responsiveness — revisiting the challenge hypothesis. Horm. Behav. 51 (4), 463—476. Harding, C.F., 1981. Social modulaton of circulating hormone levels in the male. Am. Zool. 21 (1), 223—231. Hayden-Hixson, D.M., Ferris, C.F., 1991. Steroid-specific regulation of agonistic responding in the anterior hypothalamus of male hamsters. Physiol. Behav. 50, 793—799. Hennessy, M.B., Hornschuh, G., Kaiser, S., Sachser, N., 2006. Cortisol responses and social buffering: a study throughout the life span. Horm. Behav. 49, 383—390. Hirschenhauser, K., Oliveira, R.F., 2006. Social modulation of androgens in male vertebrates: meta-analyses of the challenge hypothesis. Anim. Behav. 71, 265—277. Holm, S., 1979. A simple sequentially rejective multiple test procedure. Scand. J. Stat. 6 (2), 65—70. Isgor, C., Kabbaj, M., Akil, H., Watson, S.J., 2004. Delayed effects of chronic variable stress during peripubertal—juvenile period on hippocampal morphology and on cognitive and stress axis functions in rats. Hippocampus 14 (5), 636—648. Kaiser, S., Haderthauer, S., Sachser, N., Hennessy, M.B., 2007. Social housing conditions around puberty determine later changes in plasma cortisol levels and behavior. Physiol. Behav. 90, 405—411. Kaiser, S., Kirtzeck, M., Hornschuh, G., Sachser, N., 2003. Sexspecific difference in social support — a study in female guinea pigs. Physiol. Behav. 79, 297—303.
Social interaction decreases stress responsiveness during adolescence Kemme, K., Kaiser, S., von Engelhardt, N., Wewers, D., Groothuis, T., Sachser, N., 2009. An unstable social environment affects sex ratio in guinea pigs: an adaptive maternal effect? Behaviour 146, 1513—1529. Lu ¨rzel, S., Kaiser, S., Sachser, N., 2010. Social interaction, testosterone, and stress responsiveness during adolescence. Physiol. Behav. 99, 40—46. Mathews, I.Z., Wilton, A., Styles, A., McCormick, C.M., 2008. Increased depressive behaviour in females and heightened corticosterone release in males to swim stress after adolescent social stress in rats. Behav. Brain Res. 190 (1), 33—40. McCormick, C.M., Furey, B.F., Child, M., Sawyer, M.J., Donohue, S.M., 1998. Neonatal sex hormones have ‘organizational’ effects on the hypothalamic—pituitary—adrenal axis of male rats. Dev. Brain Res. 105, 295—307. McCormick, C.M., Mahoney, E., 1999. Persistent effects of prenatal, neonatal, or adult treatment with flutamide on the hypothalamic—pituitary—adrenal stress response of adult male rats. Horm. Behav. 35 (1), 90—101. McCormick, C.M., Mathews, I.Z., 2007. HPA function in adolescence: role of sex hormones in its regulation and the enduring consequences of exposure to stressors. Pharmacol. Biochem. Behav. 86, 220—233. McCormick, C.M., Mathews, I.Z., 2010. Adolescent development, hypothalamic—pituitary—adrenal function, and programming of adult learning and memory. Prog. Neuropsychopharmacol. Biol. Psychiatry 34, 756—765. McCormick, C.M., Smith, C., Mathews, I.Z., 2008. Effects of chronic social stress in adolescence on anxiety and neuroendocrine response to mild stress in male and female rats. Behav. Brain Res. 187, 228—238. ´., Kruk, M.R., Haller, J., 2004. Genomic and non-genomic Mikics, E effects of glucocorticoids on aggressive behavior in male rats. Psychoneuroendocrinology 29, 618—635. Sachser, N., 1990. Social organization, social status, behavioural strategies and endocrine responses in male guinea pigs. In: Balthazart, J. (Ed.), Hormones, Brain and Behaviour in Vertebrates. 2. Behavioural Activation in Males and Females — Social Interaction and Reproductive Endocrinology. Comp. Physiol., vol. 9. Karger, Basel, pp. 176—187. Sachser, N., 1993. The ability to arrange with conspecifics depends on social experiences around puberty. Physiol. Behav. 53, 539—544. Sachser, N., 1994. Sozialphysiologische Untersuchungen an Hausmeerschweinchen. Paul Parey, Berlin. Sachser, N., Hennessy, M.B., Kaiser, S., 2010. Adaptive modulation of behavioural profiles by social stress during early phases of life and adolescence. Neurosci. Biobehav. Rev., doi:10.1016/ j.neubiorev.2010.09.002.
1377
Sachser, N., Lick, C., 1991. Social experience, behavior, and stress in guinea pigs. Physiol. Behav. 50, 83—90. Sachser, N., Lick, C., Stanzel, K., 1994. The environment, hormones, and aggressive behavior — a 5-year-study in guinea pigs. Psychoneuroendocrinology 19 (5—7), 697—707. Sachser, N., Pro ¨ve, E., 1984. Short-term effects of residence on the testosterone responses to fighting in alpha male guinea pigs. Aggressive Behav. 10, 285—292. Sachser, N., Pro ¨ve, E., 1988. Plasma-testosterone development in colony and individually housed male guinea pigs. Ethology 79, 62—70. Seale, J.V., Wood, S.A., Atkinson, H.C., Bate, E., Lightman, S.L., Ingram, C.D., Jessop, D.S., Harbuz, M.S., 2004a. Gonadectomy reverses the sexually diergic patterns of circadian and stressinduced hypothalamic—pituitary—adrenal axis activity in male and female rats. J. Neuroendocrinol. 16 (6), 516—524. Seale, J.V., Wood, S.A., Atkinson, H.C., Harbuz, M.S., Lightman, S.L., 2004b. Gonadal steroid replacement reverses gonadectomyinduced changes in the corticosterone pulse profile and stressinduced hypothalamic—pituitary—adrenal axis activity of male and female rats. J. Neuroendocrinol. 16, 989—998. Seale, J.V., Wood, S.A., Atkinson, H.C., Harbuz, M.S., Lightman, S.L., 2005a. Postnatal masculinization alters the HPA axis phenotype in the adult female rat. J. Physiol. 563 (1), 265—274. Seale, J.V., Wood, S.A., Atkinson, H.C., Lightman, S.L., Harbuz, M.S., 2005b. Organizational role for testosterone and estrogen on adult hypothalamic—pituitary—adrenal axis activity in the male rat. Endocrinology 146 (4), 1973—1982. Sisk, C.L., Zehr, J.L., 2005. Pubertal hormones organize the adolescent brain and behavior. Front. Neuroendocrinol. 26, 163—174. Spear, L.P., 2000. The adolescent brain and age-related behavioral manifestations. Neurosci. Biobehav. Rev. 24, 417—463. Steinberg, L., 2005. Cognitive and affective development in adolescence. Trends Cognit. Sci. 9 (2), 69—74. van der Meij, L., Buunk, A.P., van de Sande, J.P., Salvador, A., 2008. The presence of a woman increases testosterone in aggressive dominant men. Horm. Behav. 54, 640—644. Williams, B.M., Luo, Y., Ward, C., Redd, K., Gibson, R., Kuczaj, S.A., McCoy, J.G., 2001. Environmental enrichment: effects on spatial memory and hippocampal CREB immunoreactivity. Physiol. Behav. 73, 649—658. Wingfield, J.C., Hegner, R.E., Dufty Jr., A.M., Ball, G.F., 1990. The ‘‘challenge hypothesis’’: theoretical implications for patterns of testosterone secretion, mating systems, and breeding strategies. Am. Nat. 136 (6), 829—846. Yurgelun-Todd, D., 2007. Emotional and cognitive changes during adolescence. Curr. Opin. Neurobiol. 17, 251—257.