OR08-2 Androgen Receptor- Vs Estrogen Receptor-Mediated Actions in the Development of Distinct Clinical Polycystic Ovary Syndrome (PCOS) Phenotypes

Polycystic ovary syndrome (PCOS) is a complex hormonal disorder characterised by reproductive, endocrine and metabolic abnormalities. Currently, as the origins of PCOS remain unknown, mechanism-based treatments are not feasible and management relies on treatment of symptoms. However if the underlying mechanisms involved in the development of PCOS were uncovered then this would allow development of new interventions for PCOS. Hyperandrogenism is a key defining characteristic of PCOS and clinical and animal studies support an important role for androgen driven actions in the development of PCOS. However, while testosterone (T) is consistently elevated in PCOS patients, it can be aromatized into estradiol which can act via the estrogen receptor (ER). To determine if ER-mediated actions play a role in the development of PCOS, we assessed the ability of T vs dihydrotestosterone (DHT, non-aromatizable androgen) excess to induce PCOS traits in wild type (WT, control) and androgen receptor knockout (ARKO) mice, who have a loss of androgen receptor (AR) but not ER actions. WT and ARKO prepubertal mice were implanted with a blank, T or DHT implant and examined after 12 weeks. As expected, T and DHT excess were able to induce the PCOS trait of anovulation in WT mice as no corpora lutea (CL) were observed in their ovaries (CL number: WT+blank: 9.5±2.1; WT+DHT: 0±0; WT+T: 0±0, P<0.01). In contrast, ARKO mice treated with DHT implants ovulated, but those treated with T still exhibited ovulatory disruption (P<0.05) (CL number: ARKO+blank: 2.3±0.3; ARKO+DHT: 3.3±0.4; ARKO+T: 0.4 ±0.4). This finding implies that ER actions in the absence of AR actions can induce ovulatory dysfunction, one reproductive feature of PCOS. In WT mice, DHT but not T excess induced metabolic features of PCOS (e.g. body weight (WT+blank: 23.4g±0.5; WT+DHT: 27.1g±0.6; WT+T: 22.5g±0.67, P<0.01), however neither androgen had an effect in ARKO mice (body weight, ARKO+blank: 23.1g±0.4; ARKO+DHT: 24.1g±0.7; ARKO+T: 23.5g±0.4). This data demonstrates a key role for AR signalling in the establishment of reproductive and metabolic traits of PCOS, but also suggests that ER-mediated actions may also contribute to the development of reproductive features of PCOS. These results suggest that lean PCOS patients lacking hyperandrogenism but with reproductive features of PCOS may have a distinct pathophysiology and prognosis from hyperandrogenic-PCOS patients, requiring a different treatment approach.