Isolation and Characterization of Cell Subpopulations from the Monkey Corpus Luteum of the Menstrual Cycle1

Abstract

This study was designed to test the hypothesis that the corpus luteum of primate species consists of cell subpopulations that differ in physical characteristics, function, and regulation by endocrine and paracrine factors. The corpus luteum (n = 25) was removed from rhesus monkeys at the mid-luteal phase of the menstrual cycle (Days 7–8 after the surge of luteinizing hormone, LH) and enzymatically dispersed. Freshly dispersed cells were analyzed and sorted on the basis of their forward and 90° light scatter (FLS and 90LS, respectively) properties using an EPICS C flow cytometer. Freshly dispersed and sorted cells were fixed, stained histochemically for the presence of 3β-hydroxysteroid dehydrogenase (3β-HSD), and measured to determine their diameters. Freshly dispersed (MIX) and sorted cells from corpora lutea during the early (Days 4–5 after the LH surge; n = 4) and mid-luteal phases of the cycle were incubated in vitro and steroid production was assessed. The size distribution of dispersed cells revealed four peaks that corresponded to small (10–15 μm in diameter) 3β-HSD-negative, and small, medium (16–20 μm), and large (>20 μm) 3β-HSD-positive cells. Analysis of dispersed cells for FLS and 90LS demonstrated two continua (Cα and Cβ). Cα contained single cells and cell clusters; 99.7 ± 0.3% (n = 3) of the cells were ≤15 μm in diameter and 96.7 ± 0.3% were 3β-HSD-negative. Cα cells produced low levels of progesterone (0.2 ± 0.1 ng/ml per 5 × 104 cells; n = 3) in vitro under basal conditions. Cβ consisted of single cells from 10 μm to 40 μm in diameter and contained the lipid-filled and 3β-HSD-positive cells. Two regions (R1 and R3) of Cβ were defined and their cells separated. In R1, 96 ± 2% (n = 3) of the cells had diameters of ≤15 μm, whereas 82 ± 4% (n = 3) of those in R3 were ≥20 μm. Basal progesterone production by R3 cells from early luteal phase of the cycle was 12 times >that by R1 cells (n = 3 per group). Relative stimulation of progesterone production by human chorionic gonadotropin (hCG) was not different (p>0.05) between Mix, R1, and R3 cells (2.4 ± 0.4, 2.3 ± 0.6, 3.3 ± 0.6 × basal, respectively). Likewise the relative stimulation by N, O′-dibutyryl cyclic adenosine monophosphate (dbcAMP; 3.0 ± 0.5, 3.3 ± 1.6, 3.6 ± 0.6 × basal) or prostaglandin (PG) E2 (1.8 ± 0.2, 3.0 ± 1.9, 1.8 ± 0.2 × basal) was not dıfferent (p>0.05) between MIX, R1, and R3 cells. Whereas hCG and dbcAMP stimulated MIX and R3 cells to a greater extent than did PGE2 (p<O.O5), the response to these three agonists did not differ in R1 cells. The addition of androstenedione increased (p<O.O5) estrogen production by MIX, R1, and R3 cells from the early luteal phase, with all three cell groups producing more estrone than estradiol. R3 cells produced 12-fold more estrone than R1 cells during androgen exposure. Basal steroid production by MIX, R1, and R3 cells obtained from the mid-luteal phase was not dıfferent (p>0.05) from that in the early luteal phase. However, the relative stimulation of progesterone production by hCG, dbcAMP, and PGE2 was less (p<O.05) in all cell groups at the mid-luteal phase than in the early luteal phase. Also, androstenedione increased estrogen production by MIX and R3 cells, but not R1 cells, at the mid-luteal phase. These results support the hypothesis that the primate corpus luteum consists of distinct cell subpopulations that differ in size and steroidogenic capacity. However, the different-sized cells that secrete progesterone during the early to mid-luteal phase are typically responsive to gonadotropin and PGE2, possible via a cAMP-mediated pathway.