Evidence for a role of the alternatively spliced ED-I sequence of fibronectin during ovarian follicular development

This study was aimed at testing the hypothesis that different forms of fibronectin (FN), produced as a consequence of the alternative splicing of the precursor messenger RNA, play specific roles during development of the ovarian follicle. In particular, we were interested in determining the effect of the ED-I (also termed ED-A) type III repeat, which is absent in the plasma form. Analysis of FN levels in follicular fluids corresponding to different stages of development of bovine follicles revealed marked changes in the concentrations of ED-I+FN, whereas total FN levels remained relatively constant. ED-I+FN levels were higher in small follicles, corresponding to the phase of granulosa cell proliferation. The hypothesis of a physiological role for ED-I+FN was further supported by the finding of a regulation of the alternative splicing of FN in primary cultures of bovine granulosa cells by factors known to control ovarian follicular development. cAMP produced a 10-fold decrease in the relative proportion of the ED-I region. In contrast, transforming growth factor-beta elicited a 2-fold stimulation of overall FN synthesis and a 4-fold increase in the synthesis of ED-I containing FN. This effect was evident at the protein (Western blots) and messenger RNA (Northern blots) levels. Although a negative correlation (P < 0.001) was detected between ED-I+FN and estradiol levels in follicular fluid, this steroid was unable to modulate in vitro the alternative splicing of FN. A possible mitogenic effect of ED-I+FN was suggested by the observation that a recombinant peptide corresponding to the ED-I domain stimulated DNA synthesis in a bovine granulosa cell line (BGC-1), whereas a peptide corresponding to the flanking type III sequences had no effect. The hypothesis of ED-I+FN as a growth regulatory factor was further strengthened by the fact that depletion of FN from BGC-1-conditioned medium, which contained ED-I+FN, abrogated its mitogenic activity, whereas plasma FN was without effect. We propose that changes in the primary structure of FN may mediate some of the effects of gonadotropin and intraovarian factors during follicular development.