Abstract
Polycystic ovary syndrome (PCOS) is associated with impaired gonadal steroid hormone feedback and a hyperactive central reproductive axis, implicating the brain in PCOS pathophysiology. Excess prenatal androgen (PNA) exposure is linked with the development of PCOS, and preclinical evidence suggests that PNA can program altered connections within the gonadotropin-releasing hormone (GnRH) neuronal network. Kisspeptin neurons are critical upstream regulators of GnRH neurons, relaying hormone feedback and potently activating GnRH secretion, but their role in PCOS remains incompletely understood. This study tested the hypothesis that prenatal androgen excess programmes altered neuroanatomical connections between kisspeptin neurons and GnRH neurons and modified neuropeptide and steroid hormone receptor expression in kisspeptin neurons. To investigate this, we labelled, imaged and assessed kisspeptin fibers, GnRH neurons, and gonadal steroid hormone receptors in adult female transgenic kisspeptin reporter mice (Kiss1(Cre-GFP)Coll; ROSA26tdTomato) exposed prenatally to di-hydrotestosterone or an oil vehicle (VEH) on gestational days 16-18 (n=7-10/group). PNA mice were acyclic and all tissue was collected in diestrus. Fixed coronal brain sections were immunofluorescently labelled for GnRH and kisspeptin reporter tdTomato and imaged on a confocal microscope to assess close fiber contacts to the soma and proximal dendrite of GnRH neurons. Neither the proportion of GnRH neurons with kisspeptin fiber contacts (VEH: 86.3±3.4%, PNA: 87.3±4.3%), nor the mean number of kisspeptin fiber close appositions per GnRH neuron (VEH: 6.1±0.6, PNA: 5.7±0.8) was different between groups. Kisspeptin neuron number and kisspeptin neuron co-expression with estrogen and progesterone receptors was evaluated in immunolabelled sections throughout the rostral periventricular nucleus of the 3rd ventricle (RP3V) and the arcuate nucleus (ARN). PNA had no significant impact on kisspeptin neuron number or the proportion of neurons with gonadal hormone co-expression in either population. Expression of neuropeptides was quantified by RT-qPCR in micro-dissected RP3V and ARN tissue from a separate cohort of adult PNA and VEH mice (n=5-8/group). PNA did not affect Kiss1, Tac2 or Pdyn expression in the ARN or Kiss1 expression in the RP3V. These data indicate that PNA does not modify anatomical connections between RP3V kisspeptin neurons and GnRH neurons, nor does it have a robust impact on steroid hormone responsiveness or neuropeptide expression of kisspeptin neurons. While the neuroanatomical relationship between ARN kisspeptin neurons and the GnRH neuron distal dendron in PNA mice remains to be determined, these data suggest that PNA-mediated neuroendocrine dysfunction is not due to programmed changes in the neuroanatomical projections or gonadal hormone sensitivity of kisspeptin neurons.