Aberrant GDF9 expression and activation are associated with common human ovarian disorders.

CONTEXT: Growth differentiation factor 9 (GDF9) is a central regulator of folliculogenesis and ovulation rate. Fourteen mutations in human (h) GDF9 have been reported in women with premature ovarian failure or polycystic ovarian syndrome as well as in mothers of dizygotic twins, implicating GDF9 in the etiology of these conditions. We sought to determine how these mutations alter the biological activity of hGDF9. OBJECTIVE: The objective of the study was to determine whether aberrant GDF9 expression or activation is associated with common ovarian disorders. DESIGN: Homology modeling was used to predict the location of individual mutations within structurally important regions of the pro domains and mature domains of hGDF9. Each hGDF9 variant was generated by site-directed mutagenesis, expressed from human embryonic kidney 293T cells and assessed as to whether it resulted in defective production or the enhanced activation of mature hGDF9 in an in vitro granulosa cell proliferation bioassay. RESULTS: Mutations observed in mothers of dizygotic twins (P103S and P374L) completely abrogated GDF9 expression, suggesting that women heterozygous for these mutations would have a 50% reduction in GDF9 levels. Comparable declines in GDF9 in ewes result in a 2-fold increase in ovulation rate and fecundity. Remarkably, three prodomain mutations associated with premature ovarian failure (S186Y, V216M, and T238A) all resulted in the activation of hGDF9. Mechanistically, these mutations reduced the affinity of the prodomain for mature hGDF9, allowing the growth factor to more readily access its signaling receptors. CONCLUSIONS: Together these findings indicate that alterations to hGDF9 synthesis and activity can contribute to the most common ovarian pathologies.

Species differences in the expression and activity of bone morphogenetic protein 15.

Oocyte-derived bone morphogenetic protein 15 (BMP15) regulates ovulation rate and female fertility in a species-specific manner, being important in humans and sheep and largely superfluous in mice. To understand these species differences, we have compared the expression and activity of human, murine, and ovine BMP15. In HEK293F cells, human BMP15 is highly expressed (120 ng/ml), ovine BMP15 is poorly expressed (15 ng/ml), and murine BMP15 is undetectable. Because BMP15 synthesis is dependent upon interactions between the N-terminal prodomain and the C-terminal mature domain, we used site-directed mutagenesis to identify four prodomain residues (Glu(46), Glu(47), Leu(49), and Glu(50)) that mediate the high expression of human BMP15. Substituting these residues into the prodomains of murine and ovine BMP15 led to significant increases in growth factor expression; however, maximal expression was achieved only when the entire human prodomain was linked to the mature domains of the other species. Using these chimeric constructs, we produced and purified murine and ovine BMP15 and showed that in a COV434 granulosa cell bioassay, these molecules displayed little activity relative to human BMP15 (EC(50) 0.2nM). Sequence analysis suggested that the disparity in activity could be due to species differences at the type I receptor binding interface. Indeed, murine BMP15 activity was restored when specific residues through this region (Pro(329)/Tyr(330)) were replaced with the corresponding residues (Arg(329)/Asp(330)) from human BMP15. The identified differences in the expression and activity of BMP15 likely underlie the relative importance of this growth factor between species.

Activation of latent human GDF9 by a single residue change (Gly 391 Arg) in the mature domain.

Growth differentiation factor 9 (GDF9) controls granulosa cell growth and differentiation during early ovarian folliculogenesis and regulates cumulus cell function and ovulation rate in the later stages of this process. Similar to other TGF-ß superfamily ligands, GDF9 is secreted from the oocyte in a noncovalent complex with its prodomain. In this study, we show that prodomain interactions differentially regulate the activity of GDF9 across species, such that murine (m) GDF9 is secreted in an active form, whereas human (h) GDF9 is latent. To understand this distinction, we used site-directed mutagenesis to introduce nonconserved mGDF9 residues into the pro- and mature domains of hGDF9. Activity-based screens of the resultant mutants indicated that a single mature domain residue (Gly(391)) confers latency to hGDF9. Gly(391) forms part of the type I receptor binding site on hGDF9, and this residue is present in all species except mouse, rat, hamster, galago, and possum, in which it is substituted with an arginine. In an adrenocortical cell luciferase assay, hGDF9 (Gly(391)Arg) had similar activity to mGDF9 (EC(50) 55 ng/ml vs. 28 ng/ml, respectively), whereas wild-type hGDF9 was inactive. hGDF9 (Gly(391)Arg) was also a potent stimulator of murine granulosa cell proliferation (EC(50) 52 ng/ml). An arginine at position 391 increases the affinity of GDF9 for its signaling receptors, enabling it to be secreted in an active form. This important species difference in the activation status of GDF9 may contribute to the variation observed in follicular development, ovulation rate, and fecundity between mammals.