Search

Excess androgens are associated with a characteristic polyfollicular ovarian morphology; however, it is not known to what extent this problem is due to direct androgen action on follicular development vs. interference with gonadotropin release at the level of the pituitary or hypothalamus. To elucidate potential androgen effects on the ovary, we investigated the cellular localization of androgen receptor (AR) messenger ribonucleic acid (mRNA) in rhesus monkey using in situ hybridization. To investigate the regulation of ovarian AR gene expression, we compared the relative abundance of AR transcripts in monkeys during follicular and luteal phases of the menstrual cycle and in monkeys treated with testosterone. To assess potential functional consequences of AR expression in the primate ovary, we compared AR mRNA levels with indexes of follicular cell proliferation and apoptosis in serial sections from individual follicles. AR mRNA expression was most abundant in granulosa cells of healthy preantral and antral follicles in the primate ovary. Theca interna and stromal cells also expressed AR mRNA, but to a lesser degree than granulosa cells. No significant cycle stage effects were noted in AR mRNA levels; however, larger numbers of animals would be necessary to definitively establish a cycle stage effect. AR mRNA level was significantly increased in granulosa cells and was decreased in theca interna and stromal cells of testosterone-treated monkeys. Importantly, granulosa cell AR mRNA abundance was positively correlated with expression of the proliferation-specific antigen Ki-67 (r = 0.91; P < 0.001) and negatively correlated with granulosa cell apoptosis (r = -0.64; P < 0.001). In summary, these data show that primate ovary AR gene expression is most abundant in granulosa cells of healthy growing follicles, where its expression is up-regulated by testosterone. The positive correlation between granulosa AR gene expression and cell proliferation and negative correlation with programmed cell death suggests that androgens stimulate early primate follicle development.

Insulin-like growth factors (IGF)-I and -II are abundant in the primate myometrium and are implicated in the sex steroid-induced growth of this tissue. To evaluate the potential for modulation of IGF action in the primate myometrium by locally produced IGF binding proteins (IGFBPs), we examined the cellular localization and sex steroid regulation of IGFBPs 1-5 in the nonhuman primate uterus. Ovariectomized rhesus monkeys were treated with placebo, estradiol (E2), or E2 and progesterone (P4) (E2 and P4) for 2 weeks, after which their uteri were removed and cut into thin serial sections for analysis by in situ hybridization. IGFBP-1 messenger RNA (mRNA) was not detected in control or E2-treated uteri but was found in a few unidentified cells in the E2 and P4-treated group. IGFBP-2, -3, -4, and -5 mRNAs were all expressed by myometrial smooth muscle cells but each displayed distinctive patterns of regulation by sex steroids. IGFBP-2 was barely detectable in control myometrium, was significantly increased by E2 and even more significantly by E2 and P4. IGFBP-4 and 5 mRNAs were readily detectable in control myometrium and significantly increased by E2 treatment. The addition of P4 to E2 treatment did not produce a significantly greater augmentation in IGFBP-4 or 5mRNA level compared with E2 alone. IGFBP-3 mRNA was abundant in the control myometrium, but in contrast to other IGFBPs was significantly reduced by approximately 75% in smooth muscle cells by E2 and by E2 and P4 treatment. Interestingly, however, IGFBP-3 mRNA was increased in the uterine vascular endothelium by E2 and by E2 and P4 treatment. In summary, this study has shown that four of the six known IGFBPs are highly expressed in the primate myometrium, and that their expression is differentially regulated by sex steroids. The cellular and sex steroid-regulated pattern of IGFBP-2 gene expression is very similar to that of IGF-I, as previously determined in these same myometrial samples. Both IGF-I and IGFBP-2 are dependent on E2 for significant myometrial expression, and both are further augmented by the addition of P4 to E2 treatment. Uterine smooth muscle IGFBP-3 mRNA levels are negatively regulated, whereas IGFBP-4 and -5 mRNA levels are positively regulated by E2: none of these myometrial IGFBPs appears sensitive to the effects of P4. The present observations, together with our previous data from the same animals, demonstrate that the primate myometrial smooth muscle cell expresses mRNAs for IGF-I and -II, IGF-I and -II receptors, as well as expresses mRNAs for IGFBP-2, -3, -4, and -5. These data provide evidence for complex local interactions between IGF system components regulated by estrogen and progesterone.