7869 Conditional deletion of androgen receptors from kisspeptin neurons only partially rescues the reproductive phenotype of a prenatal androgen-induced mouse model of polycystic ovary syndrome

Abstract

Abstract Disclosure: A.G. Novak: None. A.V. Petriv: None. D.B. Lohr: None. K.M. Dillon: None. N.T. Neifert: None. A.M. Moore: None. Polycystic ovary syndrome (PCOS) is a prevalent cause of infertility in people with ovaries of reproductive age. Although the etiology of PCOS remains unclear, research in the clinic and with animal models suggests that hyperandrogenism (the presence of elevated circulating androgens) plays a significant role in the development of reproductive dysfunction, including irregular menstrual/estrous cycles and altered ovarian morphology. Although the site at which androgens disrupt the reproductive axis is unclear, neurons expressing kisspeptin are crucial regulators of the hypothalamic-pituitary-gonadal axis and express androgen receptors. Therefore, we hypothesize that androgen signaling at kisspeptin cells is a critical mediator of reproductive dysfunction in PCOS. To test this, we utilized Cre-lox technology to conditionally delete androgen receptors from kisspeptin cells (KARKO) in a commonly used mouse model of PCOS induced by using prenatal dihydrotestosterone treatment (prenatal androgen-treated (PNA) mice). To assess estrous cyclicity, vaginal cytology was examined daily for three weeks beginning at 50 days of age. Prenatal vehicle-treated (PNV) wild-type (WT) and PNV KARKO mice (n=6/group) exhibited regular estrous cycles with no significant differences in the percentage of time spent in proestrus, estrus, or metestrus/diestrus, indicating androgen receptors are not required to maintain fertility. As expected, PNA WT mice (n=6) rarely entered proestrus, spent significantly less time in estrus (p<0.05), and spent significantly more time in metestrus and diestrus (p<0.001) compared to the PNV groups (n=6/group, two-way ANOVA), replicating the previously reported acyclic phenotype of PNA mice. No significant differences in the percentage of time spent in cycle stages were detected between PNA WT and PNA KARKO (n=5) mice in this analysis, suggesting androgen signaling at kisspeptin cells is not required to induce acyclicity in PNA mice. Despite this acyclic phenotype, preliminary data using ovarian sections (5μm thickness) stained with hematoxylin and eosin revealed that the percentage of ovarian area composed of corpora lutea was significantly elevated in PNA KARKO mice (n=3) compared to PNA WT mice (n=4) (p<0.05, one-way ANOVA). Further, in this analysis, the area composed of corpora lutea in PNA KARKO mice was not significantly different from PNV WT and KARKO controls (n=6/group), suggesting PNA KARKO mice exhibit a comparable ovulation rate to fertile controls. Together, these data show that androgen signaling at kisspeptin cells partially contributes to the development of anovulation in PNA mice. However, additional mechanisms likely contribute to developing the full PCOS-like reproductive phenotype. Presentation: 6/1/2024