Stress and the menstrual cycle: short- and long-term response to a five-day endotoxin challenge during the luteal phase in the rhesus monkey.

Abstract

Previously, we reported that in the rhesus monkey a 5-day inflammatory-like stress during the early-mid follicular phase acutely stimulates the hypothalamic-pituitary-adrenal axis and exerts effects on the hypothalamic-pituitary-gonadal axis, delays folliculogenesis and in some animals decreases luteal function in the post-treatment cycle. Because the endocrine environment at the time of the stress may influence the response to the stress, we now investigate the acute and long-term responses to a similar stress challenge during the luteal phase of the menstrual cycle, at a time of progesterone dominance. Nine monkeys with normal cycles were injected with endotoxin (lipopolysaccharide; LPS, 150 microg i.v.) twice a day for 5 days starting on days 4-8 after the LH peak. Blood samples were taken at hour 3 and hour 8 after each morning LPS injection to monitor the acute gonadotropin and cortisol responses. To verify cyclicity, menses were checked every day, and daily blood samples were taken for estradiol and progesterone measurement. Two control cycles, the LPS treatment cycle, and two post-treatment cycles were documented. Endotoxin activated the adrenal axis: mean (+/-SE) cortisol secretion was significantly increased at hour 3 after the first morning LPS injection (74.1 +/- 4.9 vs. 24.1 +/- 1.8 microg/dL in the control; P < 0.05) and remained elevated at hour 8. This response decreased progressively with time: on day 5 of LPS treatment, the cortisol level was still significantly higher than control at hour 3 (38.5 +/- 5.0 microg/dL; P < 0.05) but had returned to the control concentration by hour 8 (days 3-5 of LPS). Mean integrated progesterone through the luteal phase of the LPS treatment cycle was significantly decreased (33.5 +/- 3.3 ng/ml vs. 48.9 +/- 3.7 and 54.0 +/- 4.9 in the two control cycles; P < 0.05), but luteal phase length remained unchanged. When compared with control levels on the same day of the luteal phase, about one third of LH and FSH values were lower than one SD below mean control levels. LPS administration had no effect on the two post-treatment cycles, except that integrated luteal progesterone in 3 out of 9 monkeys was still reduced in post-treatment cycle 1. There were no differences in follicular phase length and preovulatory estradiol peaks between control cycles and post-treatment cycles. When compared with our previous study, the results illustrate specific responses to stress at different phases of the menstrual cycle and support the notion that a moderate short-term inflammatory-like stress episode has the potential to subtly alter critical aspects of cyclicity.