A comprehensive comparative transcriptional and translational analyses of the impact of ovarian response type, stimulation protocol and mode of trigger on the luteal function

Abstract

Objective: We aimed to compare molecular characteristics of the luteal granulosa cells between natural vs. stimulated IVF cycles in good and poor-responders. Design: Translational research study. Materials and Methods: Luteinized granulosa cells were obtained from good (n=154) and poor responder (n=64) IVF patients comparable for age, type and dose of gonadotropin and IVF etiology. Good-responders (4-15 oocytes) underwent natural (n=22), GnRH agonist (long protocol n=44) and antagonist IVF cycles triggered with rec-hCG (n=46) or GnRH agonist leuprolide acetate (n=42). Poor-responders fulfilling the Bologna criteria consisted of 64 patients undergoing GnRH antagonist protocol triggered with hCG (n=36) or hCG+GnRH agonist (n=28). Results: In the good-responders, natural cycle (NC) granulosa cells were significantly more viable (88%) compared to the stimulated IVF cycles (66%, 64% and 37% for agonist and antagonist cycles triggered with hCG and agonist respectively, p<0.05). The mRNA expression of steroidogenic enzymes (SCC, stAR, 3B-HSD, 17B-HSD and aromatase), LH receptor and VEGF and in vitro E2 and P productions were comparable between hCG-triggered agonist and antagonist cycles, but significantly higher than NC in the first days of culture. However, on the following days their hormone productions and viability began to decline very rapidly with the most drastic decrease being observed in the agonist triggered cycles. By contrast, NC granulosa cells maintained their viability and produced E 2 and P in increasing amounts in culture up to six days. The expression of anti-apoptotic genes (AKT-1, BCL2-L2) were significantly lower, and pro-apoptotic genes (BAD, BID, BAX, Cas3) were significantly higher in the stimulated cycles particularly in the agonist triggered ones compared to NC granulosa cells. Pulse exposure to cisplatin induced apoptosis only in a small fraction of the cells from the NCs whereas the same exposure caused massive apoptosis in the cells of the stimulated cycles (27% vs. 78% respectively, p<0.01). In the poor-responders both viability and steroidogenic activity of the cells were more severely reduced compared to the antagonist cycles of the good-responders. There were no significant differences between hCG and hCG+agonist triggered cycles in terms of viability, hormone production, VEGF and LH receptor expressions in the luteal granulosa cells.

Conclusions Reduced survival and increased apoptosis of luteal granulosa cells leading to defective steroid production in stimulated cycles in comparison to natural ones may at least in part explain why luteal phase is defective and requires exogenous P supplementation for support in these cycles. Also dual trigger does not appear to improve luteal function in the poor-responders.