Localization of steroidogenic enzymes in macaque luteal tissue during the menstrual cycle and simulated early pregnancy: immunohistochemical evidence supporting the two-cell model for estrogen production in the primate corpus luteum.

Abstract

It is hypothesized that the two-cell model for estrogen production by the ovarian follicle is preserved in the primate corpus luteum, but there is little direct evidence to support this theory. To determine the sites of androgen and estrogen synthesis within the primate corpus luteum and to ascertain whether changes in steroid hormone levels are related to steroidogenic enzyme expression, the enzymes converting progesterone to androgen (cytochrome P450 17alpha-hydroxylase/17,20 lyase; P450(c17)) and then to estrogen (aromatase; P450(arom)), as well as P450 side-chain cleavage (P450(scc)) and 3beta-hydroxysteroid dehydrogenase (3beta HSD), were detected by immunohistochemistry in macaque luteal tissue throughout the menstrual cycle and simulated early pregnancy. Corpora lutea were collected from rhesus monkeys in the early (Days 2-4 post-LH surge), mid (Days 6-8), mid-late (Days 10-12), and late (Days 14-15) luteal phase and after 1, 3, 6, or 9 days of hCG treatment that began on Day 9 of the luteal phase. Specific cytoplasmic staining for P450(c17), P450(arom), P450(scc), and 3beta HSD was present in luteal cells, but not in the microvasculature, within all luteal tissues examined. P450(c17)-stained luteal cells were located along the vascular tracts and around the periphery of the corpus luteum. Intensely stained luteal cells were associated with blood vessels entering from the outer surface of the corpus luteum, but not with blood vessels returning from the connective tissue centrum. In contrast, P450(arom)-stained luteal cells were distributed throughout the luteal parenchyma. P450(c17) staining intensity was similar at all stages of the luteal phase; however, the number and intensity of P450(arom)-stained cells decreased by late luteal phase. In simulated early pregnancy, cells stained for P450(c17) were present near blood vessels, with some positive cells scattered throughout the corpus luteum. P450(arom) immunostaining was heterogeneous within the corpus luteum; many intensely stained cells were interspersed among others that were only lightly stained. Overall, cellular staining for P450(c17) and P450(arom) remained intense through 9 days of simulated early pregnancy. In contrast, P450(scc) and 3beta HSD immunoreactivity were not located in distinct luteal compartments. These results are consistent with a two-cell model for steroid hormone production in the primate corpus luteum, whereby paraluteal (theca-luteal) cells produce androgen substrate that is converted to estrogens by true (granulosa-) luteal cells. The divergence in enzyme detection as the luteal phase progresses, with P450(c17) labeling high and P450(arom) staining having decreased, suggests a shift in the function of the corpus luteum as it ages. Enzyme localization during chorionic gonadotropin exposure simulating early pregnancy demonstrates the continued capacity of the primate corpus luteum to produce steroid hormones.