Expression of steroidogenic acute regulatory protein in the human corpus luteum throughout the luteal phase.

Abstract

The expression of the steroidogenic acute regulatory protein (StAR) in the human corpus luteum (CL) was examined throughout the luteal phase. The primary 1.6-kb StAR transcript was in greater abundance in early (3.1-fold) and mid (2.2-fold) luteal phase CL compared with late luteal phase CL. The larger StAR transcript (4.4 kb) was found in early and midluteal phase CL, but was not detected in late luteal phase specimens. Mature StAR protein (30 kDa) was present in lower amounts within late CL compared with early and midluteal phase CL. The StAR preprotein (37 kDa) was also detected in greater abundance in early and midluteal CL. Immunohistochemistry revealed that StAR staining was most prominent in thecal-lutein cells throughout the luteal phase. The intensity of the signal for StAR exhibited significant changes throughout the luteal phase, being most intense during the midluteal phase and least during the late luteal phase. Plasma progesterone concentrations were highly correlated (r = 0.73 and r = 0.79) with luteal expression of the preprotein and mature StAR isoforms, respectively, throughout the luteal phase. To examine the LH dependency of StAR expression, the GnRH antagonist, Cetrorelix, was administered during the midluteal phase. Cetrorelix caused a decline in serum LH levels within 2 h, which, in turn, caused a pronounced decline in plasma progesterone within 6 h. The StAR 4.4-kb transcript was not detectable, and the 1.6-kb transcript was reduced by approximately 50% within 24 h of Cetrorelix treatment. The mature 30-kDa StAR protein level declined approximately 30% after Cetrorelix treatment. We conclude that 1) StAR mRNA and protein are highly expressed in early and midluteal phase CL; 2) StAR protein is present in both thecal-lutein and granulosa-lutein cells throughout the luteal phase; 3) StAR protein levels in the CL are highly correlated with plasma progesterone levels; 4) declining StAR mRNA and protein levels are characteristic of late luteal phase CL; and 5) suppression of LH levels during the midluteal phase results in a marked decline in plasma progesterone and a diminished abundance of StAR transcripts in the CL without a corresponding significant decline in StAR protein. Collectively, these data are consistent with the idea that StAR gene expression is a key determinant of luteal progesterone during the normal menstrual cycle. However, the pharmacologically induced withdrawal in the midluteal phase of LH support diminishes luteal progesterone output by mechanisms others than reduced StAR protein levels.