DOES THE LACK OF A CYCLICAL LUTEINIZING HORMONE-SURGE CONTRIBUTE TO THE POLYCYSTIC OVARIAN PHENOTYPE IN ESTROGEN RECEPTOR ALPHA-NULL MICE?

Adult estrogen receptor-α (ERα) null mice (αERKO) are anovulatory and invariably exhibit ovaries containing multiple, often hemorrhagic cystic follicles, and lack corpora lutea (CL). This phenotype does not appear until after approximately 35 d of age and worsens thereafter. In addition, αERKO ovaries exhibit aberrantly high levels of luteinizing hormone (LH)-receptor, steroidogenesis, and inhibin-A synthesis. To date, these phenotypes have been attributed largely to overstimulation of the ovaries by persistently high LH levels that result from the lack of ERαmediated negative feedback in the hypothalamopituitary axis. Indeed, treatment of young αERKO females with a GnRH antagonist prevents the ovarian phenotype. Still, the positive feedback mechanisms that provide for cyclical LH-surges are also interrupted in αERKO mice. Hence, the ovaries are never exposed to the stimuli necessary to induce ovulation and terminal differentiation of granulosa cells. Therefore, we hypothesized that a lack of cyclical LH-surges in αERKO females allows for the survival of large antral follicles that ultimately degenerate, hemorrhage and form the characteristic cysts. To test this hypothesis, we sought to prevent the ovarian phenotype by treating animals with human chorionic gonadotropin (hCG) at periodic intervals that mimic the natural LH-surge. αERKO females were injected with 10 IU hCG (n=10) or saline (n=9) every 4–5 d over a period of one month, beginning at 30 d of age. Age-matched wild type (WT; n=9) females were injected with saline as controls. Blood and ovaries were collected 3 days after the last injection or, in the case of the WT, on diestrus. Ovaries were used for histology and gene expression analyses, and plasma was prepared for assessing hormone levels. Surprisingly, regular hCG injections did not visibly abate the phenotype of hemorrhagic, cystic follicles in αERKO females. Furthermore, no discernible CL were observed in ovaries of hCG-treated αERKO females. Circulating levels of estradiol and testosterone remained elevated in hCGtreated αERKO females, suggesting further that the treatments did not induce terminal differentiation of the granulosa cells toward a luteal cell phenotype. Real-time PCR analyses revealed no effect of the hCG treatment on the abnormally high expression of LH receptor (Lhcgr), cytochrome P450 17α-hydroxylase/C17–20 lyase (Cyp17a1), cytochrome P450 aromatase (Cyp19a1), and 17β-hydroxysteroid dehydrogenase types I and III, (Hsd17b1, Hsd17b3) that is typical of αERKO ovaries. The failure of “cyclical” hCG-treatments to trigger the formation of CL was unexpected as we have previously shown that αERKO females at 28 and 42 d of age respond to a regimen of exogenous gonadotropins with ovulation and CL formation. Our current findings suggest that prior PMSG treatment may be necessary to the hCG response in αERKO females. Alternatively, the hCG-treatments may have been effective in the beginning of the experiment in the young αERKO females, but no longer one month later in the adult αERKO females that were used for tissue collection. Nonetheless, from the current data we must conclude that exogenous LH-“surges” superimposed upon αERKO females for a period of one month, do not prevent the polycystic ovarian phenotype. Supported by the Intramural Research Program of the NIH and NIEHS. (poster)