Estrogen promotes luteolysis by redistributing prostaglandin F2α receptors within primate luteal cells

Abstract

Prostaglandin F2α (PGF2α) has been proposed as a functional luteolysin in primates. However, administration of PGF2α or prostaglandin synthesis inhibitors in vivo both initiate luteolysis. These contradictory findings may reflect changes in PGF2α receptors (PTGFRs) or responsiveness to PGF2α at a critical point during the life span of the corpus luteum. The current study addressed this question using ovarian cells and tissues from female cynomolgus monkeys and luteinizing granulosa cells from healthy women undergoing follicle aspiration. PTGFRs were present in the cytoplasm of monkey granulosa cells, while PTGFRs were localized in the perinuclear region of large, granulosa-derived monkey luteal cells by mid-late luteal phase. A PTGFR agonist decreased progesterone production in luteal cells obtained at mid-late and late luteal phases, but did not decrease progesterone production by granulosa cells or luteal cells from younger corpora lutea. These findings are consistent with a role for perinuclear PTGFRs in functional luteolysis. This concept was explored using human luteinizing granulosa cells maintained in vitro as a model for luteal cell differentiation. In these cells, PTGFRs relocated from the cytoplasm to the perinuclear area in an estrogen- and estrogen receptor-dependent manner. Similar to our findings with monkey luteal cells, human luteinizing granulosa cells with perinuclear PTGFRs responded to a PTGFR agonist with decreased progesterone production. These data support the concept that PTGFR stimulation promotes functional luteolysis only when PTGFRs are located in the perinuclear region. Estrogen receptor-mediated relocation of PTGFRs within luteal cells may be a necessary step in the initiation of luteolysis in primates.

Figure 1

PTGFR expression and function in monkey granulosa cells. Panel A. Levels of PTGFR mRNA in monkey granulosa cells obtained before (0) and 12, 24, and 36 hours after hCG administration in vivo were determined by qPCR and expressed relative to ACTB mRNA. Data are expressed as mean+SEM and were assessed by ANOVA, followed by Duncan’s post hoc test. Groups with no common superscripts are different, p<0.05; n=4–5 animals/group. Panel B. PTGFR protein was detected as a single band of 67 MW in monkey granulosa cells obtained before (0 hour) and 36 hours after hCG, and monkey corpus luteum (luteal day 10); human luteinizing granulosa cells (lgc) are included for size comparison. Pan-actin detection confirms protein loading; positions of MW size standards are shown on left. Panels C–H. PTGFR immunodetection (green) in monkey ovarian tissues obtained after controlled ovarian stimulation and before hCG (0 hour (h) C,D) or 24 h (E,F) and 36 h (G,H) after hCG. In Panels C, E, and G, follicles were imaged by conventional microscopy and are oriented as indicated in Panel C, with stroma (st) in the lower left, granulosa cells (gc) central, and follicle antrum (an) in upper right. Arrows indicate PTGFR detection throughout granulosa cells at all times examined. Panels C, E, and G use bar in Panel C = 25 μm. Panels D, F, and H show PTGFR dispersed throughout granulosa cells as imaged by confocal microscopy and use bar in Panel D = 100 μm. PTGFR detection was reduced when primary antibody was preabsorbed with the peptide used to generate the antibody (insets in Panels F and G). Nuclei are counterstained red. Images shown are representative of 3–4 monkeys. Panel I. Granulosa cells obtained at 0, 24, and 36 hours were cultured for 16 hours in the presence of no treatment, fluprostenol (F), hCG, or F+hCG. Progesterone levels (ng/ml media) were determined by EIA. Within each time of hCG exposure, data were assessed by ANOVA with 1 repeated measure, followed by Duncan’s post hoc test. Groups with no common superscripts are different, p<0.05; n=4–5 animals/time point.

Figure 2

PTGFR expression and function in monkey luteal cells. Panel A. Levels of PTGFR mRNA in monkey corpora lutea obtained during the early (days 3–4), mid (days 6–8), mid-late (days 10–11), and late (days 12–15) stages of the luteal phase were determined by qPCR and expressed relative to ACTB mRNA. Data are expressed as mean+SEM and were assessed by ANOVA, followed by Duncan’s post hoc test. Groups with no common superscripts are different, p<0.05; n=3–4 animals/group. Panels B–J. PTGFR immunodetection (green) in monkey luteal tissues obtained at the mid (B–D), mid-late (E–G), and late (H–J) luteal phases. Nuclei are counterstained red. Arrows indicate green fluorescence located in the perinuclear region of large luteal cells. Arrowheads indicate elongated nuclei of non-steroidogenic cells which do not colocalize with PTGFR. For Panels B–J, use bar in Panel G = 25 μm; images are representative of 3–4 monkeys/time point. Panels K–L. Dispersed luteal cells were cultured with no treatment, fluprostenol (F), hCG, or F+hCG for 4 hours; media was harvested for assay of progesterone (ng/ml media) and cAMP (pmol/ml media) by EIA. Within each stage of the luteal phase, data were assessed by ANOVA with 1 repeated measure, followed by Duncan’s post hoc test. Groups with no common superscripts are different, p<0.05; n=4–5 animals/luteal stage.

Figure 3

PTGFR expression and function in human luteinizing granulosa cells differentiating into granulosa-lutein cells in vitro. Panel A. Progesterone (μg/ml media) was assessed in cells after culture for 0, 2, 6, and 10 days. Media were changed on the day of culture indicated, and progesterone accumulation over 4 hours was assessed. Data were expressed as mean ± SEM and assessed by ANOVA with 1 repeated measure, followed by Duncan’s post hoc test. Groups with no common superscripts are different, p<0.05; n=4–6 women. Panels B–E. Immunodetection of PTGFR (green) in cells maintained in vitro 0 days (A), 2 days (B), 6 days (C), or 10 days (D); cells are not counterstained. Arrowheads indicates cytoplasmic detection of PTGFR; arrows indicate perinuclear PTGFR. Inset in Panel C shows reduced immunodetection when the PTGFR antibody was preabsorbed with the peptide used to generate the antibody. For Panels B–E, use bar in Panel B = 25 μm. Images shown are representative of 3–4 patients. Panels F–G. PTGFR (67 MW) detection by western blotting in total human luteinizing granulosa cell lysate (F) and in nuclear (nuc) and cytoplasmic (cyto) fractions of luteinizing granulosa cells (G) on Day 0 and Day 2 of culture (representative of n=3–4 women). Pan-actin detection (37 MW) detection in Panel F was performed on the same blot as PTGFR detection and confirms similar protein loading. Nuclear and cytoplasmic fractions were probed for tubulin (50 MW), Na/K ATPase (112 MW), and histone H3 (17 MW). Panels H–I. On Day 0 (H) or Day 2 (I–K) in vitro, fresh media containing fluprostenol (F), hCG or no treatment (control) were added to each well; media were harvested after 4 hours for assay of progesterone by EIA. Progesterone varied widely between women, so control for each woman was set at 1.0, and progesterone levels after hormone treatment are expressed relative to control levels. Panel J. Media from Day 2 cultures were assayed for cAMP by EIA (expressed as pmol cAMP/ml media). Panel K. On Day 2 in vitro, fresh media containing fluprostenol (F) alone or in combination with U73122 (F+U7), Ro31-3220 (F+Ro), or rapamycin (F+Rap) were added and collected 4 hours later for assay of progesterone. For Panels H–K, data are expressed as mean+SEM, n=4–6 patients per treatment or time point. Within each panel, data were assessed by ANOVA with 1 repeated measure, followed by Duncan’s post hoc test; data with no common superscripts are different, p<0.05.

Figure 4

Estrogen promotes PTGFR localization to the perinuclear region and PTGFR function in cultured human luteinizing granulosa cells. Cells were cultured for 2 days with vehicle (control), ICI 182,780 (ICI), letrozole (Let), or letrozole and estradiol (Let+E2). Panels A–L. PTGFR immunodetection (green), nuclear counterstain (red), and merged images are shown. Arrows indicate perinuclear location of PTGFR; arrowheads indicate PTGFR dispersed throughout cells. Images were obtained with confocal microscopy and are representative of n=3 women. Panel M. Cells were cultured for 2 days with vehicle (control), ICI, Let, and Let+E2 as described above; media were replaced and cells were treated for 4 hours with media containing vehicle or fluprostenol before assay for progesterone by EIA. Progesterone levels after hormone treatment were expressed relative to control levels, which was set at 1.0 for each woman. Fluprostenol reduced media progesterone (asterisks) in control and Let+E2-treated cells as assessed by paired t-test, p<0.05. Data are expressed as mean+SEM, n=4 women.