Localization, regulation and possible consequences of apoptotic protease-activating factor-1 (Apaf-1) expression in granulosa cells of the mouse ovary.

The recent characterization of apoptotic protease-activating factor-1 (Apaf-1) in vertebrates as a putative homolog of the Caenorhabditis elegans gene, ced-4, indicates that the third major arm of the C. elegans programmed cell death machinery has also been conserved through evolution. Although apoptosis is now known to be important for ovarian follicular atresia in vertebrates, nothing is known of the role of Apaf-1 in ovarian function. Herein we show by immunohistochemical analysis that Apaf-1 is abundant in granulosa cells of early antral follicles whereas in vivo gonadotropin priming completely suppresses Apaf-1 expression and granulosa cell apoptosis. Western blot analysis of fractionated protein extracts prepared from granulosa cells before and after in vitro culture without hormonal support to induce apoptosis indicated that mitochondrial cytochrome c release, a biochemical step required for the activation of Apaf-1, occurs in granulosa cells cultured in vitro. Moreover, Western blot analysis of procaspase-3 processing, a principal downstream event set in motion by activated Apaf-1, indicated that healthy granulosa cells possess almost exclusively the inactive (pro-) form of the enzyme whereas granulosa cells deprived of hormonal support rapidly process procaspase-3 to the active enzyme. Lastly, we show that serum-starved granulosa cells activate caspase-3-like enzymes both prior to and after nuclear pyknosis, as revealed by a single-cell fluorescent caspase activity assay. These data, combined with previous observations regarding the role of homologs of the two other C. elegans cell death regulatory genes, ced-9 (Bcl-2 family members) and ced-3 (caspases), in atresia fully support the hypothesis that granulosa cell apoptosis is precisely coordinated by all three major arms of a cell death program conserved through evolution.