Luteotrophic factors in hyperstimulated pseudopregnant rabbit: I--Evidence for aromatase activity in luteal tissue and luteal cells.

It is well established that the rabbit corpus luteum (CL) function depends upon endogenous oestradiol, the major source of which in the rabbit ovary is considered to be the ovarian follicles. The absence of oestradiol formation by the rabbit CL has been previously reported. In a hyperstimulated pseudopregnant rabbit model used in our laboratory which developed a large number of corpora lutea in response to chorionic gonadotrophin (eCG)/hCG, we observed the survival of corpora lutea in vivo, and normal levels of plasma progesterone throughout pseudopregnancy (PP), despite the scarcity or the absence of follicles as a source of the luteotrophic hormone. Measurement of oestradiol in the plasma indicated that it was at high levels and correlated with the number of corpora lutea. This led us to investigate the luteal origin of oestradiol in this model. PP was induced in rabbits by i.m. injection of 200 IU eCG daily for 2 days followed on day 4 by i.m. injection of 200 IU hCG (day 0 of PP). Luteal tissue obtained at days 5, 9 and 12 of PP and cultured for 24 h synthesized oestradiol and testosterone in addition to progesterone. However, under the same conditions, follicles had limited capacity to secrete oestradiol. The presence of an aromatase activity in luteal tissue was confirmed when exogenous testosterone was added to the culture medium. P450aromatase (P450arom) mRNA was found in luteal tissue at days 5, 9 and 12 of PP. Small or large luteal cells, obtained by enzymatic digestion of the tissue followed by centrifugation in a Percoll density gradient, were cultured during several days with or without gonadotrophin or dibutyryl cAMP (dbcAMP). Both types of cells secreted oestradiol. In small cells and luteal tissue, aromatase activity was stimulated (1.5-2-fold) by hCG and dbcAMP. Large cells exhibited a greater capacity to aromatize testosterone than small cells, but aromatase activity was not modified by hCG or by dbcAMP. FSH had no effect on aromatase activity of either luteal cell type. This intrinsic luteal tissue aromatase capacity and the absence of premature regression of corpora lutea despite the limited support of follicular oestrogen, suggest an autocrine and luteotrophic role for this luteal oestrogen.