Dissecting the ontogeny and functional relevance of altered GABAergic circuitry in Polycystic Ovary Syndrome (PCOS)

Abstract

Polycystic ovary syndrome (PCOS) is the most common cause of female infertility worldwide, yet this prevalent endocrine disorder remains poorly understood. Although classically considered an ovarian disease, altered brain wiring may play a central role in the pathogenesis of PCOS. Neuroendocrine derangements in PCOS include elevated luteinizing hormone (LH) pulse secretion, which likely results from high gonadotropin-releasing hormone (GnRH) pulse frequency and mirrors defects within the GnRH neuronal network. Enhanced GABA actions in GnRH neurons have been proposed to be a culprit of altered GnRH/LH secretion and underlie some of the pathological features of the disorder. Pre-clinical and clinical evidence support the idea that prenatal androgen excess may program an abnormal GABA-to-GnRH neuron circuit to develop PCOS during adult life. In this regard, the present study aimed to answer fundamental questions about the ontogeny, rescue and biological function of altered GABAergic innervation to GnRH neurons. This PhD project used a prenatally androgenized (PNA) mouse model that recapitulates the cardinal features of PCOS and mimics the elevated LH pulse frequency of the disease. Initial studies aimed to address whether GnRH neuronal network changes observed in PNA mice were programmed through early androgen exposure or driven by adult androgen excess. Control and PNA GnRH-GFP transgenic mice were evaluated at postnatal day (PND) 25, prior to pubertal onset. Confocal imaging and analysis of GABA inputs onto GnRH neurons revealed that GABAergic contact was significantly increased in prepubertal PNA mice (P < 0.05). In addition, circulating testosterone levels at PND 25 were not different between PNA and control groups, suggesting that brain circuit abnormalities were not dependent upon early manifestation of androgen excess in a PCOS-like condition. Serial blood sampling defined the developmental timing of androgen excess in PNA mice, showing that circulating testosterone levels rise significantly during early adulthood (PND 50 and PND 60) in PNA animals when compared to controls (P < 0.01). All hyperandrogenic PNA mice presented disruption of estrous cyclicity, displaying significant arrest in the metestrous stage and a complete absence of the proestrous stage (P < 0.0001), indicating anovulatory cycles. To determine whether neuroendocrine derangements of LH regulation would persist after the removal of hyperandrogenic ovaries, the same cohort of mice were ovariectomized (OVX) and serial blood sampling was performed to investigate LH pulse dynamics. Although LH pulse frequency was similar between the groups, OVX PNA mice exhibited greater LH pulse amplitude (P < 0.0001) and magnitude of LH release than controls (P < 0.001), implying that defects within the GnRH network remained in the absence of hyperandrogenic ovaries and suggesting that the primary pathology of PCOS is in the brain. Compelling clinical evidence indicates that long-term androgen receptor (AR) blockade with flutamide (Flut) is able to restore both the sensitivity of the GnRH pulse generator and menstrual cyclicity in PCOS women. This PhD project tested the hypothesis that these improvements may be the result from plastic changes in the brain that rescue normal GABAergic wiring to GnRH neurons. Control and PNA mice were treated with Flut (25 mg/kg/day) or an oil vehicle from PND 40 to PND 60. GABA inputs to GnRH neurons were assessed as previously performed for prepubertal animals and confirmed that oil-treated adult PNA mice display enhanced GABAergic contact on GnRH neurons (P < 0.01). Remarkably, Flut treatment was able to decrease and rescue normal GABA-to-GnRH neuron circuit features in PNA mice (F1, 21 treatment = 41.8; P < 0.0001). Results also showed that estrous cyclicity of PNA mice improved considerably during Flut treatment. Evaluation of ovarian morphology treatment showed that AR signaling blockade improved preovulatory follicle recruitment and restored normal features of the granulosa and theca cell layers in these follicles. Previous neuroanatomical work indicates that increased GABAergic wiring to GnRH neurons in PNA mice originates largely from the arcuate nucleus (ARN) of the hypothalamus. In this PhD project, I investigated the functional role of GABA neurons originating in the ARN in regulating LH secretion using in vivo optogenetics. Selective targeting and expression of channelrhodopsin-2 E123T accelerated variant (ChETA) in the ARN GABA neurons was achieved using vesicular GABA transporter (VGAT)-Cre mice. ARN GABA neurons were activated by delivering blue light pulses (5ms) at 2 and 20 Hz during 10 minutes in diestrus female, male and PNA mice. Optogenetic activation at 20 Hz elicited robust LH release similarly in male and diestrus female (P < 0.05), whereas 2-Hz stimulation failed to evoke changes in LH levels. Interestingly, 20-Hz light stimulation in PNA mice induced smaller changes in LH levels when compared to male and diestrus female groups (P < 0.05). These data suggest that altered LH release in PNA mice might reflect a decreased pituitary LH releasable pool due to a formerly high GnRH pulse frequency stimulation in the PCOS-like condition. Together, these findings support the idea that a prenatal androgen insult can program altered GABAergic brain circuits early in development, prior to pubertal onset, and might be the culprit for developing subsequent androgen excess during early adulthood. This PhD thesis also highlights that abnormal GABA-to-GnRH neuron circuit remains plastic in adult PNA mice; and that the specific GABAergic pathway from ARN GABA neuron is biologically relevant to modulate LH secretion. These data support the important role of ARN GABA neurons in the regulation of the GnRH neuron biology in healthy fertility and in the pathophysiology of PCOS.