Neuroendocrine Basis for Disrupted Ovarian Cyclicity in Females During Chronic Undernutrition: A Mouse Model

Chronic undernutrition is a type of metabolic stress that impairs reproduction across species and, in women, is implicated in the development of functional hypothalamic amenorrhea. Although the tight coupling of energy balance to reproductive capacity is recognized in principle, the neuroendocrine loci and molecular mechanisms that mediate ovarian cycle dysfunction during undernutrition remain poorly understood. Ovarian cyclicity is dependent on a population of kisspeptin (Kiss1) neurons in arcuate nucleus (ARC(Kiss1)) for luteinizing hormone (LH) pulses and in the anteroventral periventricular nucleus (AVPV(Kiss1)) for LH surge secretion. Here, we present a series of studies in which we tested the hypothesis that inhibition of both Kiss1 cell populations underlies the impairment of the cycle by undernutrition. During a baseline period, body weight, feed intake, and ovarian cycle stage (via vaginal cytology) were evaluated in female c57bl6 mice. Then, animals were randomly assigned into one of two groups (n=6-8/grp): 1) ad libitum fed controls or 2) feed restricted (70% of feed consumed during the baseline period). Control animals displayed clear and regular cycles throughout the 4-week treatment period. In contrast, feed restriction caused a significant and rapid cessation of ovarian cyclicity (4.8±0.3 vs. 1.5±0.5 estrus cycles/4 weeks; control vs. restricted, p<0.05), causing all females to enter and remain mostly in diestrus. Based on these results, we conducted two experiments to directly test the hypothesis that undernutrition inhibits both modes of LH secretion (and both Kiss1 cell populations) using two well-defined estradiol (E) replacement paradigms. We first evaluated LH pulses in mice that were ovariectomized and implanted subcutaneously with a pellet containing a diestrus level of E (100 ng, OVX+LowE). Following 3 days of feed restriction or control diet (n=3/grp), serial blood samples were collected every 8 min for 88 min. Undernutrition prevented LH pulses and significantly reduced mean LH (5.2±0.6 vs. 0.6±0.2 ng/mL; control vs. restricted, p<0.05). Fixed neural tissue was evaluated by immunohistochemistry to determine whether undernutrition impairs ARC(Kiss1) neuronal activation, using c-Fos as a marker. The percent of ARC(Kiss1) neurons expressing cFos was reduced by 90% (p<0.05). We next evaluated the LH surge. After 3 days, control or feed restricted mice were OVX and implanted subcutaneously with a surge-inducing estradiol implant (OVX+HighE, 1 µg, n=3-4/grp). Undernutrition completely blocked the E-induced LH surge (1.9±0.3 vs. 0.2±0.02 ng/mL; control vs. restricted, p<0.05) and diminished Kiss1 mRNA abundance in micropunches of the AVPV (42%, p<0.05). Collectively, these studies clearly show that undernutrition impairs both ARC(Kiss1) control of LH pulses and AVPV(Kiss1) induction of the LH surge, via mechanisms that remain to be identified.