Abstract
Primary monolayer cultures of bovine theca cells isolated from pooled ovarian follicles (3–10 mm diameter) were used to examine the effects of various granulosa cell-derived substances on basal and luteinizing hormone (LH)-induced androgen and progesterone secretion. After an overnight pretreatment period, cells were incubated with a range of treatments including LH, oestradiol-17β, inhibin, activin and follistatin. Media were collected after 48 h and assessment of androstenedione and progesterone secretion made by radioimmunoassay.
Addition of LH (5–50 ng/ml) to the cells resulted in a dose-dependent stimulation of both androstenedione (2·5-to 3-fold rise; P<0·01) and progesterone (∼ 1·6-fold rise; P<0·001) production. Secretion of androstenedione was also raised (up to 5-fold; P<0·001) by addition of oestradiol-17β (0·3–300 ng/ml), whilst levels of the androgen in the presence of both LH (20 ng/ml) and oestradiol (300 ng/ml) were up to 12-fold higher (P<0·001) than control values. In contrast, oestradiol treatment inhibited by up to 50% both basal (P<0·001) and LH-stimulated (P<0·001) secretion of progesterone. Exposure of cells to purified bovine inhibin (5–125 ng/ml) consistently raised androstenedione secretion by up to 42% over basal levels (P<0·001). Inhibin also enhanced both LH-stimulated (∼20%; P<0·001) and oestradiol-stimulated (∼20%; P<0·05) secretion of androstenedione. In direct contrast, treatment of theca cells with human recombinant activin-A (1–50 ng/ml) inhibited both LH-stimulated (∼50%; P<0·001) and oestradiol-stimulated (∼30%; P<0·005) androstenedione secretion. Activin also reversed the positive effect of inhibin on basal (P<0·01), LH-stimulated (P<0·001) and oestradiol-stimulated (P<0·001) androstenedione secretion, though activin alone did not affect basal steroid output. Simultaneous addition of human recombinant follistatin reversed the inhibitory effects of activin on LH- and oestradiol-induced androstenedione secretion but did not modify the effects of inhibin. Follistatin alone did not alter either basal or LH-stimulated androstenedione output. Neither basal nor LH-stimulated secretion of progesterone were consistently affected by inhibin, activin or follistatin.
As well as confirming the stimulatory effects of both LH and oestradiol on bovine thecal cell androgen production, these observations are indicative of opposing intrafollicular paracrine roles for granulosa cell-derived inhibin and activin in modulating thecal cell responses to gonadotrophins and steroids in the bovine ovary. Though inhibin and oestradiol had qualitatively similar effects in promoting thecal androgen secretion, the magnitude of the response to oestradiol was much greater. The results also support an intrafollicular role of follistatin as a binding protein capable of neutralizing the effect of activin, but not inhibin, on thecal androgen production.
Journal of Endocrinology (1995) 145, 491–500