The effects of a long-term psychosocial stress on reproductive indicators in the baboon

Abstract

Psychosocial stress is thought to negatively impact fecundity, but human studies are confounded by variation in nutrition and lifestyle. Baboons offer a useful model to test the effect of prolonged mild stress on reproductive indicators in a controlled setting. Following relocation from social groups to solitary housing, a previously documented stressful event for nonhuman primates, daily urine samples, tumescence, and menstrual bleeding were monitored in twenty baboons (Papio sp.) for 120-150 days. Specimens were assayed for estrone conjugates (E1C), pregnanediol-3-glucuronide (PDG), follicle-stimulating hormone (FSH), and cortisol. Linear mixed effects models examined (1) the effects of stress on frequency of anovulation, hormone levels, tumescence and cycle length, and (2) the relationship of cortisol with reproductive indicators. Despite cortisol levels indicative of stress, anovulation was negligible (1% in 102 cycles). PDG, FSH, cycle length, and tumescence declined during the first four cycles, but began recovery by the fifth. Cortisol was negatively associated with FSH but not associated with PDG, E1C or tumescence. Ovulation, E1C, and luteal phase length were not affected. Tumescence tracked changes in FSH and PDG, and thus may be a useful indicator of stress on the reproductive axis. Elevated cortisol was associated with reduced FSH, supporting a model of cortisol action at the hypothalamus rather than the gonad. After four to five menstrual cycles the reproductive indicators began recovery, suggesting adjustment to new housing conditions. In conclusion, individual housing is stressful for captive baboons, as reflected by cortisol and reproductive indicators, although ovulation, a relatively direct proxy for fecundity, is unaffected.

Fig. 1

Stress can down regulate the hypothalamic-pituitary-gonadal axis at several locations, resulting in suppression of circulating reproductive hormones. GnRH, gonadotropin releasing hormone; FSH, follicle stimulating hormone; LH, luteinizing hormone. A color version of this figure may be viewed in the online issue, which is available at http://onlinelibrary.wiley.com/.

Fig. 2

Illustrative examples of urinary cortisol, estrone conjugates (E1C), pregnanediol-3-glucuronide (PDG), follicle stimulating hormone (FSH), and tumescence index (TI) across the observation period for two luteal phase move (LPM) (panels A and B) and two follicular phase move (FPM) baboons (panels C and D). SG adj., specific gravity adjusted.

Fig. 2

Illustrative examples of urinary cortisol, estrone conjugates (E1C), pregnanediol-3-glucuronide (PDG), follicle stimulating hormone (FSH), and tumescence index (TI) across the observation period for two luteal phase move (LPM) (panels A and B) and two follicular phase move (FPM) baboons (panels C and D). SG adj., specific gravity adjusted.

Fig. 2

Illustrative examples of urinary cortisol, estrone conjugates (E1C), pregnanediol-3-glucuronide (PDG), follicle stimulating hormone (FSH), and tumescence index (TI) across the observation period for two luteal phase move (LPM) (panels A and B) and two follicular phase move (FPM) baboons (panels C and D). SG adj., specific gravity adjusted.

Fig. 2

Illustrative examples of urinary cortisol, estrone conjugates (E1C), pregnanediol-3-glucuronide (PDG), follicle stimulating hormone (FSH), and tumescence index (TI) across the observation period for two luteal phase move (LPM) (panels A and B) and two follicular phase move (FPM) baboons (panels C and D). SG adj., specific gravity adjusted.

Fig. 3

Estimates of average reproductive indicator by cycle number and phase move group for cycle length (panel A), pregnanediol-3-glucuronide (PDG) area under the curve (AUC) (panel B), follicle stimulating hormone (FSH) AUC (panel C), estrone conjugates (E1C) AUC (panel D), and tumescence index (TI) AUC (panel E) by cycle number from linear mixed effects models adjusting for cycle length (panels B through E).

Fig. 4

Estimates of average cortisol area under the curve (AUC) by cycle number and phase move group from a linear mixed effects model adjusting for cycle length (panel A). Cortisol AUC (divided by cycle length) by cycle number: left side follicular phase move (FPM) group, right side luteal phase move (LPM) group (panel B).