FSH/LH-Dependent Upregulation of Ahr in Murine Granulosa Cells Is Controlled by PKA Signaling and Involves Epigenetic Regulation

Abstract

The aryl hydrocarbon receptor (Ahr) is a ligand-activated transcription factor primarily known for its toxicological functions. Recent studies have established its importance in many physiological processes including female reproduction, although there is limited data about the precise mechanisms how Ahr itself is regulated during ovarian follicle maturation. This study describes the expression of Ahr in ovarian granulosa cells (GCs) of immature mice in a gonadotropin-dependent manner. We show that Ahr upregulation in vivo requires both follicle stimulating hormone (FSH) and luteinizing hormone (LH) activities. FSH alone increased Ahr mRNA, but had no effect on Ahr protein level, implicating a possible LH-dependent post-transcriptional regulation. Also, the increase in Ahr protein is specific to large antral follicles in induced follicle maturation. We show that Ahr expression in GCs of mid-phase follicular maturation is downregulated by protein kinase A (PKA) signaling and activation of Ahr promoter is regulated by chromatin remodeling.

Keywords: aryl hydrocarbon receptor (AhR); chromatin remodeling; follicle-stimulating hormone (FSH); luteinizing hormone (LH); post-transcriptional regulation; protein kinase A (PKA).

Figure 1

The effect of gonadotropins on Ahr expression in ovarian GCs in vivo. (a) Mice were injected once with 5 IU of PMSG, 5 IU of FSH or vehicle (NT) and lysates were collected from granulosa cells (GCs) isolated before (0 h) or 48 h later. Representative Western blot shows Ahr and actin protein levels. Specificity of Ahr antibody is shown in Supplementary Figure S1d). (b) Densitometry analysis of three independent experiments (mean ± SD) representing Ahr protein levels normalized to actin. (c) qPCR analysis of Ahr mRNA levels. (d) Mice were injected in total 4 times (every 12 h) with FSH (1.5 IU) or FSH (1.5 IU) + LH (1.25 IU) and lysates were collected from GCs isolated before (NT) or 48 h after initial injection. Representative Western blot shows Ahr and actin protein levels. (e) Densitometry analysis of three independent experiments (mean ± SD) representing Ahr protein levels normalized to actin. (f) qPCR analysis of Ahr mRNA levels. Bars with no common superscripts are significantly different (p < 0.05).

Figure 2

The effect of PMSG on expression dynamics of Ahr and follicle maturation marker genes in GCs in vivo. (a) Mice were injected with 5 IU of PMSG or vehicle (NT) and lysates were collected from GCs isolated before (0 h) or up to 48 h later. Representative Western blot shows Ahr and actin protein levels. (b) Densitometry analysis of five independent experiments (mean ± SD) representing Ahr protein levels normalized to actin. qPCR analysis of Ahr (c), Fshr (d), Cyp19a1 (e) and Lhcgr (f) mRNA in GCs isolated before (0 h) or up to 48 h after PMSG (5 IU) or vehicle (NT) treatment. * p < 0.05 vs. NT; # p < 0.05 vs. 0 h.

Figure 3

The effect of PMSG treatment on Ahr localization in ovary in vivo. Mice were injected with 5 IU of PMSG or vehicle (NT) and ovaries were isolated 48 h later. Representative images of immunofluorescence analysis of vehicle-treated (a) and PMSG-treated (b) ovaries from three independent experiments using Ahr-specific antibody (green). Nuclei were stained with DAPI (blue). Normal rabbit IgG as primary antibody was used for isotype control (Supplementary Figure S2). C—cumulus granulosa cells. M—mural granulosa cells. Scale bar, 100 μm.

Figure 4

The effect of PKA signaling on Ahr expression. (a) Mice were treated with 5 IU of PMSG for 24 h in vivo, GCs were isolated and cultured in vitro. Following attachment (2–3 h later), GCs were treated with vehicle (NT) or Fsk (10 μM). Representative Western blot shows Ahr, p-CREB and actin levels in GCs before (0 h) and 4 h after vehicle (NT) or forskolin (Fsk, 10 μM) treatment. (b) Densitometry analysis of three independent experiments (mean ± SD) representing Ahr protein levels normalized to actin. (c) Densitometry analysis of three independent experiments (mean ± SD) representing p-CREB protein levels normalized to actin. (d) Mice were treated with 5 IU of PMSG for 24 h in vivo, GCs were isolated and cultured in vitro. Following attachment (2–3 h later), GCs were treated with vehicle (NT) or Fsk (10 μM) alone or in combination with PKA inhibitor H89 (10 μM). Ahr mRNA levels were measured 2 h and 4 h later by qPCR. (e) Representative Western blot shows p-CREB and actin protein levels measured in GCs isolated before (0 h) or up to 48 h after PMSG (5 IU) or vehicle treatment in vivo. (f) Densitometry analysis of p-CREB protein levels. The data are presented as ratio of PMSG vs, NT for different timepoints from three independent experiments (mean ± SD). Bars with no common superscripts are significantly different (p < 0.05).

Figure 5

The effect of pregnant mare’s serum gonadotropin (PMSG) on Ahr gene expression mechanisms. (a) Ahr heteronuclear RNA (hnRNA) levels were measured by qPCR in GCs isolated before (0 h) or up to 48 h after PMSG (5 IU) or vehicle (NT) treatment in vivo. Data are presented as means ± SD from five independent experiments. * p < 0.05 vs NT; # p < 0.05 vs. 0 h. (b) Mice were injected with 5 IU of PMSG, GCs were isolated 48 h later and cultured in vitro. Following attachment (2–3 h later), cells were incubated with vehicle (NT) or PMSG (5 IU/mL) alone or in combination with transcription inhibitor actinomycin D (ActD, 1 μg/mL). Ahr mRNA levels were measured 2 h and 4 h later by qPCR. Data are presented as means ± SD from three independent experiments. (c) GCs isolated from immature mice were cultured in vitro, transfected with empty vector (pGL3-basic) or Ahr-luciferase reporter constructs and treated with vehicle (NT) or PMSG (5 IU/mL) for 48 h followed by analysis of luciferase activity. Data are presented as means ± SD from three independent experiments. (d) Mice were injected with 5 IU of PMSG or vehicle (NT) and GCs were isolated 48 h later. DNA from nuclei treated with increasing concentrations of DNase I was analyzed by qPCR. The recovered DNA at promoter sites of interest were normalized to recovered genomic DNA from Pax7 promoter, the data are presented as ratio of PMSG vs. NT from five independent experiments (mean ± SD). Bars with no common superscripts are significantly different (p < 0.05).