SHORT COMMUNICATION

The influence of puberty on stress reactivity andforebrain glucocorticoid receptor levels in inbred andoutbred strains of male and female mice

Russell D. Romeo *, Elianna T. Kaplowitz, Amy Ho, Daly Franco

Barnard College of Columbia University, Department of Psychology and Neuroscience and Behavior Program, Milbank Hall, 3009Broadway, New York, NY 10027, United States

Received 8 March 2012; received in revised form 28 June 2012; accepted 30 July 2012

Psychoneuroendocrinology (2013) 38, 592—596

KEYWORDSPuberty;Sex differences;HPA axis;Restraint stress

Summary

Objective: To examine stress-induced corticosterone responses and forebrain glucocorticoidreceptor (GR) levels in prepubertal and adult, male and female mice of three commonly usedinbred (C57BL/6, BALB/c) and outbred (Swiss Webster) strains.Methods: Prepubertal (30 days of age) and adult (75 days of age), male and female C57BL/6,BALB/c, and Swiss Webster mice were exposed to a 30 min session of restraint stress. Plasmacorticosterone was measured before (basal), or 0, 30, or 60 min after termination of the stressor.GR protein levels of the medial prefrontal cortex, paraventricular nucleus of the hypothalamus,and hippocampus were also measured via tissue punches and western blots in the prepubertal andadult males and females at the basal time point.Results: In response to acute stress, prepubertal males of both inbred strains showed greaterhormonal responsiveness than their adult counterparts, while females of these strains displayedsimilar stress-induced corticosterone responses, independent of age. Conversely, only thefemales of the outbred Swiss Webster strain showed pubertal-related changes, with adult femalesshowing greater hormonal reactivity compared to prepubertal females. Despite these significantdifferences in hormonal reactivity, we found little difference in GR protein levels in the brainregions examined.Conclusions: These data indicate that pubertal-dependent differences in stress reactivity can besignificantly influenced by sex and genetic background. Moreover, these data provide points ofdeparture for future studies investigating how puberty, sex, and genetic background interact toshape both short- and long-term effects of stress on mental and physical health.# 2012 Elsevier Ltd. All rights reserved.

* Corresponding author. Tel.: +1 212 854 5903; fax: +1 212 854 3601.E-mail address: rromeo@barnard.edu (R.D. Romeo).

Available online at www.sciencedirect.com

jou rn a l home pag e : ww w. el sev ie r. com/ loca te /psyn eu en

0306-4530/$ — see front matter # 2012 Elsevier Ltd. All rights reserved.

http://dx.doi.org/10.1016/j.psyneuen.2012.07.019

Pubertal-, sex-, and strain-differences in stress reactivity 593

1. Introduction

Pubertal development can be marked by significant physio-logical and psychological vulnerabilities, including obesity,depression, anxiety, and drug abuse (Patton and Viner, 2007).It is presently unclear what mediates the occurrence of thesedysfunctions, but mounting evidence suggests exposure tostressors during puberty may play a fundamental role (Dahland Gunnar, 2009). Major shifts in stress reactivity have beennoted during puberty in a variety of species, includinghumans (Romeo, 2010). In rats, for instance, prepubertaland adolescent individuals display more prolonged stress-induced hormonal adrenocortical responses compared toadults (Romeo, 2010). In addition to pubertal development,other parameters, such as a sex and genetic background haveconsiderable influence on stress reactivity (Mormede et al.,2002; Becker et al., 2007). These factors, therefore, mayinteract and contribute to stress-related vulnerabilities dur-ing adolescence. Nevertheless, little is known about theinteractions between puberty, sex, and genetic backgroundand their effect on stress responsiveness, and ultimatelyhealth and well-being.

Due to their rich strain differences and genetic modifia-bility (Anisman et al., 2001; Kolber et al., 2008), mousemodels are becoming increasingly more common in stressresearch. Thus, the purpose of the present experiments wasto examine stress-induced corticosterone responses of pre-pubertal and adult, male and female mice in both inbred(C57BL/6 and BALB/c) and outbred (Swiss Webster; SW)strains. Along with our hormonal analysis, we also measuredglucocorticoid receptor (GR) protein levels in key forebrainregions implicated in modulating hormonal stress responsiv-ity, namely the medial prefrontal cortex (mPFC), paraven-tricular nucleus (PVN), and hippocampus (Herman et al.,2003). Together, these studies represent an important firststep in characterizing the effects of puberty, sex, and geneticbackground on stress reactivity.

2. Methods

2.1. Animals and housing

All of the male and female C57BL/6, BALB/c (inbred), and SW(outbred) mice were obtained from our breeding colony atBarnard College. All mice were weaned at 21 days of age andhoused (3—5 per cage) with non-sibling, same-sex and -straincagemates. Mice were housed in polycarbonate cages(28 cm � 17 cm � 12 cm) with wood chip bedding, and main-tained on a 12 h light—dark schedule (lights on 09:00 h). Allanimals had free access to food and water. All procedureswere approved by the Institutional Animal Care and UseCommittee of Columbia University.

2.2. Experimental design and tissue collection

At either 30 (prepubertal) or 75 (adult) days of age, male andfemale C57BL/6, BALB/c and SW mice were weighed andrapidly decapitated. Tissues were collected before (basal),immediately after a 30 min session of restraint stress (Time0), or 30 or 60 min following termination of the stressor (n = 5per age, sex, strain, and time point). Restraint stress was

administered by placing the mice in a prone position in wiremesh restrainers, sized so that both sexes in the different agegroups were equally restrained. Animals in the 30 and 60 minpost-stress groups were returned to their home cages in thecolony room until tissues were collected. All animals werekilled between 11:00 and 13:00 h during their circadian nadirof corticosterone release to minimize variations in basalhormone levels. Trunk blood samples were collected inheparinized tubes, spun down, and plasma stored at�20 8C until radioimmunoassays were performed. Testesand uteri were collected, adipose tissue was removed, andtissue wet weights were obtained to verify the maturationalstage of the animals.

2.3. Radioimmunoassay

Plasma corticosterone concentrations were measured using aradioimmunoassay and performed as indicated by the sup-plier (Coat-A-Count; Siemens; Los Angeles, CA). All sampleswere run in duplicate and values were averaged. The intra-assay coefficient of variation and lower limit of detectabilitywere 8.2% and 8.52 ng/ml, respectively.

2.4. Tissue punches, western blots, anddensitometry

Tissue punches were acquired from the brains of prepubertaland adult male and female C57BL/6, BALB/c, and SW mice atthe basal time point (n = 5 per sex and strain). Using a 0.5 mmdiameter sample corer tool (Fine Science Tools; Foster City,CA), bilateral tissue punches were obtained from the mPFC,PVN and hippocampus (Bregma 1.70 mm, �0.46 mm, and�1.94 mm, respectively; Franklin and Paxinos, 2008). Theseareas were examined because they have been implicated inglucocorticoid-dependent modulation of the hypothalamic-pituitary-adrenal (HPA) axis and have a relatively high con-centration of GR (Herman et al., 2003).

Tissue punches were homogenized in lysis buffer (1% SDS indH2O with Roche complete, Mini, EDTA-free protease inhi-bitor cocktail; Roche, Diagnostics, Basal, Switzerland) andthen incubated in boiling water for 10 min. Lysates werecentrifuged (13,000 rpm for 5 min) to remove insolublematerial. The protein concentration of the cleared super-natants was determined by the BCA method (Pierce, Rock-ford, IL). Lysates (3 mg/lane) were subjected to SDS gelelectrophoresis, blotted to a nitrocellulose membrane (Invi-trogen; Carlsbad, CA) and probed with anti-GR (1:10,000;rabbit, M-20; Santa Cruz Biotechnology, Inc., Santa Cruz, CA)or anti-actin (1:5000; mouse; Sigma) as a loading control.Using Kodak XAR film, proteins were visualized by chemilu-minescence (Pierce) according to the manufacturer’s instruc-tions. Films were scanned and the relative optical densities(RODs) of the immunoreactive bands were measured usingopen-access computerized image analysis software (ImageJ).

2.5. Statistical analysis

All data are presented as the mean � S.E.M. Within a sex andstrain, corticosterone levels were analyzed with two-wayANOVAs (age � time). GR RODs within a strain at the basaltime point were also analyzed with two-way ANOVAs

594 R.D. Romeo et al.

(age � sex). Significant main effects and interactions werefurther analyzed with Tukey’s Honestly Significant Differencetests. Differences were considered significant when P < 0.05.

3. Results

3.1. Somatic measures

In each of the strains, there were significant main effects ofage for body, paired testis, and uterine weights, such thatprepubertal animals weighed less and had significantlylighter paired testis and uterine weights compared to adults(Table 1). These somatic data indicate that these 30 and 75day old mice are at disparate stages of pubertal develop-ment. Moreover, despite the variability in pubertal onsetbetween different strains of mice (Nelson et al., 1990), thesesimilar tissue weights at each age across the strains suggeststhat we sampled our different strains at approximately thesame stages of development.

3.2. Hormonal stress reactivity

A two-way ANOVA on corticosterone levels of male C57BL/6mice revealed a significant interaction of age and time(P < 0.05). Specifically, though stress lead to increased cor-ticosterone levels in both prepubertal and adult C57BL/6mice, only the prepubertal mice continued to show a sig-nificant corticosterone response 30 min after termination ofrestraint. In females, there was only a significant main effectof time, with animals immediately after the stressor showinghigher corticosterone levels than the females at all othertimes, independent of age (Fig. 1).

In BALB/c males, there was also a significant interaction ofage and time (P < 0.05) such that restraint stress induced ahigher corticosterone response in prepubertal compared toadult males immediately following termination of the stress.A significant main effect of time was found for the BALB/c

Table 1 Mean (�S.E.M.) body, paired testis, and uterine weightreceptor immunoreactivity bands of the medial prefrontal cortex (mhippocampus (HIPP) in prepubertal (30d) and adult (75d day), masomatic measures, asterisks indicate a significant difference betwethat these weights are collapsed across time points). For GR RODsfemale within a strain (note that these RODs are derived from an

Strain Age/sex Body weight (g) Testes (g)

C57BL/6 30d/male 14.9 � 0.93 0.07 � 0.01

C57BL/6 75d/male 23.5 � 0.44 * 0.17 � 0.01 *

C57BL/6 30d/female 13.1 � 0.62 —

C57BL/6 75d/female 20.9 � 0.16 * —

BALB/c 30d/male 13.9 � 0.42 0.08 � 0.01

BALB/c 75d/male 22.2 � 0.77 * 0.14 � 0.01 *

BALB/c 30d/female 13.0 � 0.35 —

BALB/c 75d/female 18.5 � 0.78 * —

Swiss Web 30d/male 18.7 � 0.97 0.09 � 0.01

Swiss Web 75d/male 30.7 � 0.59 * 0.17 � 0.01 *

Swiss Web 30d/female 15.7 � 0.23 —

Swiss Web 75d/female 26.2 � 0.86 * —

* P < 0.05.

females (P < 0.05) with a post hoc test indicating thatimmediately upon termination of the restraint both prepu-bertal and adult females had higher corticosterone levelsthan the prepubertal and adult BALB/c females at all othertime points (Fig. 1).

The corticosterone response in SW males showed a sig-nificant main effect of time in that independent of age malemice showed higher responses immediately after the stressorthan the males at all other time points. In female SWs, asignificant interaction was found with restraint leading to ahigher corticosterone response in adult compared to prepu-bertal females immediately following termination of thestressor (Fig. 1).

3.3. GR protein levels

Though no significant differences in the GR ROD levels of themPFC or hippocampus in prepubertal and adult male andfemale C57BL/6 mice were found, a sex difference in GRlevels in the PVN was observed where females had greater GRROD values in the PVN compared to males (Table 1). Therewere no significant interactions or main effects of age or timeon GR RODs in the mPFC, PVN, or hippocampus in the BALB/cor SW mice (Table 1).

4. Discussion

In this brief report, we show there are pubertal-relateddifferences in hormonal stress reactivity in mice. Impor-tantly, however, these effects of age are highly dependentupon the sex and strain of mouse examined. For example, inboth inbred strains, prepubertal males showed greater hor-monal responsiveness, as manifested by a more prolonged(C57BL/6) or higher (BALB/c) corticosterone response, whencompared to adult males, while the prepubertal andadult females showed similar hormonal reactivity. In ouroutbred SW mice, we find pubertal-related changes in stress

s (g) and relative optical densities (ROD) of the glucocorticoidPFC), paraventricular nucleus of the hypothalamus (PVN), and

le and female, C57BL/6, BALB/c and Swiss Webster mice. Foren prepubertal and adult animals within a strain and sex (note, asterisks indicate a significant difference between males andimals under basal conditions).

Uterus (g) mPFC (ROD) PVN (ROD) HIPP (ROD)

— 0.21 � 0.01 0.05 � 0.01 0.37 � 0.03— 0.16 � 0.04 0.10 � 0.03 0.33 � 0.060.01 � 0.001 0.14 � 0.02 0.16 � 0.03 * 0.34 � 0.020.09 � 0.002 * 0.16 � 0.09 0.22 � 0.02 * 0.36 � 0.07— 0.18 � 0.06 0.20 � 0.05 0.17 � 0.09— 0.13 � 0.04 0.19 � 0.02 0.38 � 0.060.01 � 0.001 0.31 � 0.05 0.14 � 0.02 0.30 � 0.030.09 � 0.004 * 0.13 � 0.04 0.16 � 0.03 0.26 � 0.04— 0.06 � 0.04 0.17 � 0.05 0.13 � 0.06— 0.05 � 0.04 0.35 � 0.07 0.20 � 0.070.02 � 0.001 0.11 � 0.01 0.34 � 0.02 0.14 � 0.030.09 � 0.001 * 0.10 � 0.01 0.27 � 0.09 0.31 � 0.06

Figure 1 Mean (�S.E.M.) plasma corticosterone (ng/ml) concentrations in prepubertal (30d) and adult (75d), male and female,C57BL/6, BALB/c, and SW mice. Asterisks indicate a significant difference between the ages within that sex and strain. Groups of barsthat share letters within a sex and strain are not significantly different from each other. The horizontal black lines on the x-axes markthe 30 min session of restraint stress.

Pubertal-, sex-, and strain-differences in stress reactivity 595

responsiveness only in the females, with adult femalesdisplaying a heightened corticosterone response comparedto prepubertal females.

Previous reports in both male and female rats, particularlyin the inbred Sprague-Dawley strain, have shown that pre-pubertal animals display more prolonged or larger adreno-corticotropin hormone and corticosterone responsescompared to adults (Romeo, 2010). Thus, our current datain the male C57BL/6 and BALB/c mice are in agreement withthese earlier findings in the rat. However, the absence of anage effect in the female C57BL/6 and BALB/c mice and maleSW mice, and the greater hormonal reactivity in the adultcompared to prepubertal SW females, indicates importantsimilarities and differences in the way different species(e.g., rats and mice) respond to acute stressors. Althoughit is currently unknown what mediates these age-, sex-, andstrain-dependent effects on stress reactivity, gonadal hor-mones such as testosterone, estradiol, and progesteronehave been shown to modulate stress responsiveness in adultmales and females (Viau, 2002). Thus, future studies shouldinvestigate the role that pubertal changes in gonadal hor-mone levels play in these age-dependent effects.

Despite the differential stress-induced hormonalresponses, we found little difference in GR protein levels,except for a sex difference in C57BL/6 mice in which femaleshad relatively higher levels of GR in the PVN than males,regardless of age. Though we cannot rule out possible differ-ences in GR function, or differences outside of the areas we

measured, there appears to be no association between basalGR protein levels and our observed corticosterone responses,similar to the lack of correlation between hippocampal GRmRNA and strain-related differences in stress reactivity(Priebe et al., 2005). These data suggest that perhaps differ-ences in central drive to the HPA axis and/or adrenocorticalsensitivity to its releasing hormone may be responsible forthese differences in hormonal responsiveness.

Though not analyzed directly, the relative differencesbetween the males and females and the strains used inthe current set of experiments are in agreement with pre-viously published studies. In particular, BALB/c mice havebeen shown to be more hormonally responsive than C57BL/6sfollowing exposure to restraint stress or an immune challenge(Priebe et al., 2005; Harizi et al., 2007), while adult femaleBALB/c mice have also been reported to show greater hor-monal reactivity than adult male BALB/c mice (Harizi et al.,2007). Sex differences have also been reported in adultC57BL/6 mice, with females demonstrating higherrestraint-induced corticosterone responses than males (Goeland Bale, 2008). Though it appears we did not observe thissex difference in our C57BL/6 mice, not all studies havefound this difference in this strain of mice (Jones et al., 1998;Harizi et al., 2007).

It is important to note that the females used in ourexperiments were not cycled to determine the phase of theirestrous cycle when tissues were collected. Thus, it is possiblethat the normal fluctuations in estradiol (E) and progesterone

596 R.D. Romeo et al.

(P) levels associated with the cycle may have affected thecorticosterone responses in our adult females (Viau, 2002).However, given the relative low wet weight and variability ofthe adult uteri, it would appear the vast majority of our adultfemales were collected during estrus or diestrus, when E andP levels are low (Becker et al., 2005).

In conclusion, these data indicate significant pubertaldifferences in the way that male and female C57BL/6,BALB/c and SW mice respond to acute stress. Moreover,these data highlight important similarities and differencesin stress responsiveness both within strains and betweenspecies (i.e., mice vs. rats), and set the stage for futureprojects investigating how puberty, sex, and genetic back-ground impact physiological and behavioral reactivity tostress.

Conflict of interest

All authors declare that they have no conflicts of interest.

Funding

This work was supported in part from a grant to BarnardCollege from the Undergraduate Science Education Programof the Howard Hughes Medical Institute.

Acknowledgements

We would like to thank Page Buchanan for excellent animalcare.

References

Anisman, H., Hayley, S., Kelly, O., Borowski, T., Merali, Z., 2001.Psychogenic, neurogenic, and systemic stressor effects on plasmacorticosterone and behavior: mouse strain-dependent outcomes.Behav. Neurosci. 115, 443—454.

Becker, J.B., Arnold, A.P., Berkley, K.J., Blaustein, J.D., Eckel, L.A.,Hampson, E., Herman, J.P., Marts, S., Sadee, W., Steiner, M.,et al., 2005. Strategies and methods for research on sexdifferences in brain and behavior. Endocrinology 146,1650—1673.

Becker, J.B., Monteggia, L.M., Perrot-Sinal, T.S., Romeo, R.D., Tay-lor, J.R., Yehuda, R., Bale, T.L., 2007. Stress and disease: is beingfemale a predisposing factor? J. Neurosci. 27, 11851—11855.

Dahl, R.E., Gunnar, M.R., 2009. Heightened stress responsiveness andemotional reactivity during pubertal maturation: implications forpsychopathology. Dev. Psychopathol. 21, 1—6.

Franklin, K.B.J., Paxinos, G., 2008. The mouse brain in stereotaxiccoordinates. Elsevier, New York.

Goel, N., Bale, T.L., 2008. Organizational and activational effects oftestosterone on masculinization of female physiological and be-havioral stress responses. Endocrinology 149, 6399—6405.

Harizi, H., Homo-Delarche, F., Amrani, A., Coulaud, J., Mormede, P.,2007. Marked genetic differences in the regulation of bloodglucose under immune and restraint stress in mice reveals a widerange of coticosensitivity. J. Neuroimmunol. 189, 59—68.

Herman, J.P., Figueiredo, H., Mueller, N.K., Ulrich-Lai, Y., Ostander,M.M., Choi, D.C., Cullinan, W.E., 2003. Central mechanisms ofstress integration: hierarchical circuitry controlling hypothalam-ic-pituitary-adrenocortical responsiveness. Front. Neuroendocri-nol. 24, 151—180.

Jones, B.C., Sarrieau, A., Reed, C.L., Azar, M.R., Mormede, P., 1998.Contribution of sex and genetics to neuroendocrine adaption tostress in mice. Psychoneuroendocrinology 23, 505—517.

Kolber, B.J., Wieczorek, L., Muglia, L.J., 2008. Hypothalamic-pitui-tary-adrenal axis dysregulation and behavioral analysis of mousemutants with altered glucocorticoid or mineralocorticoid recep-tor function. Stress 11, 321—338.

Mormede, P., Courvoisier, H., Ramos, A., Marissal-Arvy, N., Ousova,O., Desautes, C., Duclos, M., Chaouloff, F., Moisan, M.-P., 2002.Molecular genetic approaches to investigate individual variationsin behavioral and neuroendocrine stress responses. Psychoneur-oendocrinology 27, 563—583.

Nelson, J.F., Karelus, K., Felicio, L.S., Johnson, T.E., 1990. Geneticinfluences on the timing of puberty in mice. Biol. Reprod. 42,649—655.

Patton, G.C., Viner, R., 2007. Pubertal transitions in health. Lancet369, 1130—1139.

Priebe, K., Romeo, R.D., Francis, D.D., Sisti, H.M., Mueller, A.,McEwen, B.S., Brake, W.G., 2005. Maternal influences of adultstress and anxiety-like behavior in C57BL/6J and BALB/cJ mcie: across fostering study. Dev. Psychobiol. 47, 398—407.

Romeo, R.D., 2010. Pubertal maturation and programming of hypo-thalamic-pituitary-adrenal reactivity. Front. Neuroendocrinol.31, 232—240.

Viau, V., 2002. Functional cross-talk between the hypothalamic-pituitary-gonadal and -adrenal axes. J. Neuroendocrinol. 14,506—513.