Objective: The mechanisms underlying the adrenal androgen (AA) excess of polycystic ovary syndrome (PCOS) remain unclear, although it is possible that the adrenocortical dysfunction may be a response to other, extraadrenal factors. Consistent with the pathophysiology of PCOS and with in vivo data in normal and PCOS women, we have hypothesized that insulin inhibits and that T stimulates AA secretion in vitro.
Design: In vitro experimental study.
Setting: University medical center.
Patient(s): Normal human adrenals (n = 4 women, ages 25-57 years) were obtained with consent at the time of organ donation.
Intervention(s): Fresh adrenal tissue minces were incubated in serum-free medium with 10-microM pregnenolone substrate and 1-microM ACTH-(1-24). Challenge doses of 0.2, 1, 5, 20, and 100 nM of insulin and 1, 10, 100, 1,000, and 10,000 nM of T were added, and the media were sampled after 8 hours of incubation at 37 degrees C, 4% CO2. Dehydroepiandrosterone (DHEA), DHEA-sulfate (DHEAS), and cortisol (F) were measured by radioimmunoassay (significant effects compared with the case of zero-dose control).
Main outcome measure(s): The production of DHEA, DHEAS, and F in the media of the adrenal minces was compared between different subjects and at different concentrations of T and insulin.
Result(s): Analysis of the combined data from all donors indicated that insulin stimulated DHEAS and suppressed DHEA production but had no consistent effect on F. Similar analyses of the combined data indicated that T had no significant predictable effect on the production of DHEAS, DHEA, or F. When examining donor data individually, insulin and T did elicit significant increases and/or decreases in steroid production within subjects, although no consistent trends were observed.
Conclusion(s): On the basis of these data, it is clear that extra-adrenal factors such as insulin and T have some adrenal regulatory capacity. In general, insulin stimulated DHEAS and decreased DHEA production, suggesting that it increases adrenocortical sulfotransferase activity. However, although in the individual subjects studied, both insulin and T frequently altered the production of DHEAS, DHEA or F, these effects did not appear to be uniform or consistent from subject to subject. Expanded studies are required to confirm these results.