Sex differences in the serotonergic influence on the hypothalamic-pituitary-adrenal stress axis

Abstract

Appropriate interactions between serotonin (5-HT) and stress pathways are critical for maintaining homeostasis. Dysregulation of hypothalamic-pituitary-adrenal (HPA) stress axis is a common feature in affective disorders in which an involvement of 5-HT neurocircuitry has been implicated in disease vulnerability and treatment responsiveness. Because there is a greater prevalence of affective disorders in women, sex differences in the 5-HTergic influence on stress pathways may contribute to disease disparity. Therefore, our studies compared stress or citalopram-induced corticosterone levels in male and female mice. To determine whether sex-dependent HPA axis responsiveness was mediated by the difference in testosterone levels, testosterone-treated females were also examined. Gene expression patterns in 5-HTergic and stress neurocircuitry were analyzed to determine sites of potential sex differences and mechanisms of testosterone action. As expected, restraint stress corticosterone levels were higher in intact females and were masculinized by testosterone. Interestingly, citalopram administration independent of stress resulted in a greater corticosterone response in females, which was also masculinized by testosterone. Analyses along the 5-HT-HPA axis revealed sex differences including greater pituitary 5-HT receptors and adrenal weights in females. Moreover, in stress-regulatory regions, we found sex differences in glucocorticoid receptor and glutamic acid decarboxylase expression supportive of greater inhibitory modulation and feedback potential in males. Taken together, these data suggest that multiple sites related to 5-HTergic stimulation, corticosterone production, and negative feedback of HPA neurocircuitry combine to produce higher female stress responsiveness. These studies support a potential for sex-specific involvement of 5-HT and stress pathways in the etiology of affective disorders.

Figure 1

Sex differences in corticosterone levels after restraint stress were masculinized by testosterone. A, Intact females exhibited greater corticosterone basally (at time 0) as well as after restraint (at 15 min) and during recovery (at 120 min) compared with males. *, P < 0.05. OVX+V females were not different from intact females, but corticosterone in OVX+TP females was significantly reduced to male levels at times 0 and 120. #, P < 0.05. Gray bar represents period of restraint stress. B, AUC analysis showed an overall sex difference in restraint-induced corticosterone that was masculinized by TP exposure. *, P < 0.05, different from males and OVX+TP females. C, The rise in corticosterone from 0 to 30 min due to restraint stress was not different among groups. D, The rate of recovery from 30 to 120 min was lower in OVX+V females compared with males. *, P < 0.05.

Figure 2

Sex differences in corticosterone levels in response to citalopram (15 mg/kg) were masculinized by testosterone. Intact females exhibited greater corticosterone than males after ip injection of vehicle (A) and citalopram (B). *, P < 0.05. Corticosterone was not affected by OVX+V but was masculinized by TP. #, P < 0.05, compared with intact and OVX+V females. Citalopram delayed recovery at time 120 in OVX+V females. §, P < 0.05, compared with vehicle. The ip injection occurred immediately after time 0. C, Analysis of AUC revealed overall sex difference in corticosterone that was masculinized by TP in response to vehicle or citalopram injections. *, P < 0.05, different from males and OVX+TP females. Citalopram significantly increased overall levels of corticosterone. #, P < 0.05, main effect of drug. D, The rise of corticosterone from 0 to 30 min in response to vehicle or citalopram injections was greater in intact females than males. *, P < 0.05. Citalopram caused a greater rise than vehicle. #, P < 0.0001. E, The rate of recovery from 30 to 120 min was higher with citalopram treatment than vehicle. #, P < 0.05.

Figure 3

Effects of sex and gonadal hormones on stress-related gene expression in primary tissues of the HPA axis. A, TP exposure in females reduced CRF mRNA in the PVN. #, P < 0.05, different from OVX+V. B, There were no differences in expression of vasopressin in the PVN. C, Intact and TP-treated females displayed lower 5-HT 2A receptor than males in the PVN. *, P < 0.05, compared with males. Expression was increased in OVX+V females. §, P < 0.05, different from intact and TP-treated females. D and F, There were no differences in POMC or CRF receptor-1 in the pituitary. E, There is a dramatic sex difference in CRF binding protein. *, P < 0.05, compared with males. OVX+V females also displayed greater expression than males but reduced from intact females. §, P < 0.05, different from all other groups. TP exposure in females increased CRF binding protein mRNA. #, P < 0.05, different from all other groups. G, TP treatment in females reduced melanocortin 2 receptor mRNA in the adrenal gland. #, P < 0.05, different from intact females. H, 11βHSD-1 mRNA was reduced in intact and OVX+V females (*, P < 0.05, different from males) and was increased by TP treatment. #, P < 0.05, different from OVX+V females. I, Intact and OVX+TP females displayed greater expression of TH than males. *, P < 0.05, different from males. J, Sex differences in adrenal weight were masculinized by TP treatment. Both the right and left adrenal glands were heavier in intact and OVX+V females compared with males and OVX+TP females. *, P < 0.05.

Figure 4

There were no sex differences in TPH2 mRNA in the dorsal raphe (DR) nucleus. A, Representative images of TPH2 silver grains in the middle vmDR. B, There was a trend for group differences in TPH2 in the middle vmDR. C, There were no differences in the caudal dmDR. Diagram of middle DR adapted from mouse atlas (34). aq, Aqueduct; dmDR, dorsomedial DR; vmDR, ventromedial DR; xscp, decussation of the superior cerebellar peduncle.

Figure 5

Effects of sex and gonadal hormones on stress-related gene expression in feedback regions of the HPA axis. A, Intact and OVX+V females had lower GR mRNA expression in the dorsal hippocampus CA1 region than males. *, P < 0.05. TP treatment decreased GR further. #, P < 0.05, compared with OVX+V females. B, There were no group differences in GR mRNA in the ventral hippocampus. C and E, Intact and OVX+TP females had lower expression of GAD65 mRNA in the BNST and lateral septum than males. *, P < 0.05. Both OVX groups had higher expression than intact females in the lateral septum. #, P < 0.05. D and F, There was a trend for group differences in GAD67 in the BNST and no differences in the lateral septum.