Investigating steroid hormone feedback in a mouse model of polycystic ovarian syndrome

Abstract

Polycystic ovarian syndrome (PCOS) is the most common cause of infertility among women of reproductive age worldwide. Although the aetiology is unclear, there are a number of neuroendocrine disruptions which identify PCOS as a state of impaired steroid hormone feedback. Steroid hormones act through a neuronal network to regulate the activity of gonadotropin-releasing hormone (GnRH) neurons, the key cells responsible for controlling fertility. We employed a murine model of PCOS by prenatal androgen (PNA) exposure to investigate steroid hormone feedback to the GnRH neuronal network in adult female offspring. Initial investigation of ovarian morphology was conducted to characterise the model. PNA treatment significantly reduced the area of the adult ovary containing corpora lutea (p<0.01). Investigation of antral follicles revealed the area of the granulosa cell layer was significantly reduced (p<0.001) and the area of the thecal cell layer significantly increased (p<0.001) in PNA-treated mice. This data suggests impaired ovulation and altered hormone synthesis, which matches the cardinal symptoms of PCOS in women.

The suppression of gonadotropin secretion by estrogen was investigated in a negative feedback trial. PNA-treated mice exhibited a blunted post-castration rise in luteinising hormone (LH) (p<0.001) and a lack of LH suppression following estrogen treatment (p<0.001). This provides strong support for impaired feedback regulation of gonadotropin secretion by estrogen. Additionally, expression of the progesterone receptor (PR) was significantly decreased in the anteroventral periventricular nucleus (AVPV) (p<0.001) and the periventricular nucleus (PeN) (p<0.001) of PNA-treated mice, suggesting progesterone feedback in the brain is impaired.

Pilot studies investigating events facilitating the preovulatory GnRH/LH surge downstream of estrogen signalling detected that a critical increase in PR expression was maintained in the AVPV (p<0.001) and arcuate nucleus (p<0.001) of control and PNA-treated animals. Activation of GnRH neurons was visualised by employing transgenic GnRH-GFP mice, but revealed no difference between the percentages of activated (cFos positive) GnRH neurons in vehicle- and PNA-treated animals at the time of the surge. Interestingly, PNA- treated mice did not display the expected increase in spine density in activated GnRH neurons seen in vehicle-treated mice (p<0.001). This indicates an increase in excitatory input did not occur. However, plasma LH levels showed no difference between vehicle- and PNA-treated mice, thus, further work is still required to show whether PNA treatment impairs the LH surge required for ovulation.

Ultimately, this study supports that PNA treatment in the mouse results in a PCOS- like phenotype and that PCOS, a state of impaired steroid hormone feedback by gonadal steroid hormones, involves alterations to central brain circuits important to fertility.