Progesterone receptor and Stat5 signaling cross talk through RANKL in mammary epithelial cells

Abstract

Progesterone (P4) stimulates proliferation of the mammary epithelium by a mechanism that involves paracrine signaling mediated from progesterone receptor (PR)-positive to neighboring PR-negative cells. Here we used a primary mouse mammary epithelial cell (MEC) culture system to define the molecular mechanism by which P4 regulates the expression of target gene effectors of proliferation including the paracrine factor receptor and activator of nuclear factor κB ligand (RANKL). MECs from adult virgin mice grown and embedded in three-dimensional basement-membrane medium resemble mammary ducts in vivo structurally and with respect to other properties including a heterogeneous pattern of PR expression, P4 induction of RANKL and other target genes in a PR-dependent manner, and a proliferative response to progestin. RANKL was demonstrated to have multiple functional P4-responsive enhancers that bind PR in a hormone-dependent manner as detected by chromatin immunoprecipitation assay. P4 also stimulated recruitment of signal transducer and activator of transcription (Stat)5a to RANKL enhancers through an apparent tethering with PR. Analysis of primary MECs from Stat5a knockout mice revealed that P4 induction of RANKL and a broad range of other PR target genes required Stat5a, as did P4-stimulated cell proliferation. In the absence of Stat5a, PR binding was lost at selective RANKL enhancers but was retained with others, suggesting that Stat5a acts to facilitate PR DNA binding at selective sites and to function as a coactivator with DNA-bound PR at others. These results show that RANKL is a direct PR target gene and that Stat5a has a novel role as a cofactor in PR-mediated transcriptional signaling in the mammary gland.

Figure 1.

Primary MECs grown in matrigel form acini structures, express PR, and respond to progestin. A, Primary MECs from 12-week adult virgin Balb/c mice were grown for 14 days in Matrigel, fixed with methacarn, processed for sectioning, and stained with hematoxylin and eosin (left panel). Sections were also stained for luminal and myoepithelial cell markers, respectively, by IF for K8 (red) and K5 (green) (right panel). B, Primary mouse MECs grown for 14 days in Matrigel were treated continuously with the progestin R5020 (100 nM) or vehicle (EtOH). PR protein was detected by IHC with a PR-specific antibody. C, Immunoblotting with a PR-specific antibody for detection of PR protein. Lane 1, PR negative parental HC-11 cells; lane 2, recombinant mouse PR-B expressed from adenoviral vectors in HC-11 cells; lane 3, recombinant mouse PR-A expressed in HC-11; lane 4, untreated isolated primary MECs from 12 week virgin mice. β-Tubulin was used as a loading control. D, PR protein quantification as a percent PR-positive cells and PR mRNA measured by real-time qPCR (right panel). E, Cell proliferation as a quantification of percent positive Ki67 cells. F, Segregation of PR-positive and proliferating cells as detected by dual IF. 3D acini of primary MEC cultures were fixed in place in Matrigel and permeabilized with 0.5% Triton-X-100 and incubated with a monoclonal antibody (PRa7) that recognizes mouse PR and an antibody to the proliferation marker ki67. Alexa-fluor-conjugated secondary antibodies were used to detect PR (green) and ki67 (red), and nuclei were labeled with DAPI (blue) at ×40 magnification. Inset is an expanded view of acini with 2 PR-positive cells (green) adjacent to a ki67 (red)-positive cell. Results are mean ± SEM from 3 independent experiments. Student's t test was used to determine statistical significance. ns, not significant (P > .05); *, P < .05; **, P < .01; ***, P < .001; scale bar, 50 μm. GAPDH, glyceraldehyde 3-phosphate dehydrogenase.

Figure 2.

P4 regulation of a broad range of target genes in primary MECs. Primary MECs grown in Matrigel for 14 days were treated continuously with vehicle (EtOH) or R5020 (100 nM), and gene expression was measured with a TaqMan LDA qPCR assay designed with 40 previously identified genes induced acutely by P4 in the mouse mammary gland in vivo (13). A 2-fold P4 induction over vehicle control was used as a cutoff for induced genes. White bars denote four P4 targets previously identified as key mediators of paracrine proliferative signaling in the mammary gland (Areg, Calca, calcitonin [Calca], RANKL, and Wnt4).

Figure 3.

P4 regulation of paracrine growth factor genes in primary 3D MEC cultures. Primary MECs grown for 14 days in Matrigel were treated continuously with EtOH or R5020 (100 nM), and expression of RANKL (panel A), Wnt4 (panel B), and Areg (panel C) was measured by gene-specific qPCR. D, Primary MECs were grown and treated, and expression of RANKL, Areg, and Wnt4 was determined by qPCR as in panels A–C, except R5020 treatment times were for 6 hours, 24 hours, 48 hours prior to harvest or R5020 treatment was continuous for the entire 14 days. Results represent mean ± SEM from 3 independent experiments. Student's t test was used to determine statistical significance between 2 hormone treatments. One-way ANOVA with Bonferroni's post test was used to determine statistical significance between multiple hormone treatments. ns, not significant (P > .05); *, P < .05; **, P < .01; ***, P < .001. GAPDH, glyceraldehydes 3-phosphate dehydrogenase; GOI, gene of interest.

Figure 4.

Progestin induction of paracrine growth factor genes is PR dependent. Primary MEC cultures in Matrigel were grown and treated for 14 days with either EtOH, R5020 (100 nM), RU486 (100 nM), or R5020 and RU486 (100 nM each). RANKL, Wnt4, and Areg were analyzed by qPCR as in Figure 3. GAPDH, glyceraldehydes 3-phosphate dehydrogenase; GOI, gene of interest.

Figure 5.

RANKL is a direct target of PR. A, Schematic representation of distal enhancers of RANKL. Potential PREs and GAS sequences were identified by the Transcription Element Search System (TESS) (54), in which black boxes denote enhancers with both PRE and GAS sequences and white boxes denote enhancers with PREs only. B, RANKL enhancer activation in response to progestin. NMuMg cells transfected with individual RANKL enhancer-luciferase reporters (100 ng) and mouse PR-B (100 ng) were treated with either EtOH or R5020 (100 nM) for 24 hours. Luciferase activity was measured by a dual luciferase reporter assay, normalized to Renilla activity, and displayed as fold R5020 induction over vehicle (EtOH) setting the value of vehicle-treated NMuMg cells to 1. C, ChIP assay of PR binding to the RANKL distal enhancers in response to R5020 in HC-11 cells. Cells were transduced with adenovirus encoding mouse PR-B and treated for 1 hour with EtOH or R5020 (100 nM). Cells were processed for ChIP with an antibody to PR (sc-7208X). Immunoprecipitated DNA was amplified by qPCR using primers flanking RANKL enhancers (Supplemental Figure 1). Amplified ChIP DNA was normalized to the input and displayed as fold R5020-induced PR recruitment over vehicle (EtOH), setting the value of vehicle treated HC-11 cells to 1. D, ChIP assay of PR binding to RANKL distal enhancers in primary MECs isolated from mouse mammary glands treated in vitro (4.5 hours) with EtOH or P4 (100 nM) and processed for ChIP with an antibody to PR (sc-7208X) as in panel C. Results are mean ± SEM from 3 independent experiments. Student's t test was used to determine statistical significance. ns, not significant (P > .05); *, P < .05; **, P < .01; ***, P < .001; TSS, transcription start site.

Figure 6.

PRL induces binding of Stat5a to selected enhancers but does not induce RANKL expression. Primary MECs grown for 14 days in Matrigel were treated continuously with vehicle or PRL (3 μg/mL) and expression of RANKL (panel A) or β-casein (panel B) were measured by gene-specific qPCR. C, Primary mouse MECs were treated with vehicle or PRL (3 μg/mL) for 4.5 hours and were processed for ChIP similar to Figure 5D, but with a Stat5a antibody (sc-1081). DNA was amplified by qPCR using primers flanking each RANKL enhancer (Supplemental Figure 1), and amplified DNA was normalized to the input. Data are presented as fold PRL-induced Stat5a recruitment (PRL/vehicle), setting the vehicle-treated MECs to 1. Results are mean ± SEM from 3 independent experiments. Student's t test was used to determine statistical significance. ns, not significant (P > .05); *, P < .05; **, P < .01; ***, P < .001. GAPDH, glyceraldehydes 3-phosphate dehydrogenase.

Figure 7.

P4 induces recruitment of Stat5a to RANKL enhancers. A, HC-11 cells transduced to express PR-B were treated with EtOH or R5020 (100 nM) for 1 hour and were processed for ChIP with an antibody to Stat5a (sc-1081). Immunoprecipitated DNA was amplified by qPCR using primers flanking each RANKL enhancer (Supplemental Figure 1), and amplified DNA was normalized to the input. Amount of Stat5a binding is displayed as fold R5020-induced Stat5a recruitment over EtOH, setting the value of vehicle-treated HC-11 cells to 1. B, Primary mouse MECs were treated with EtOH or P4 (100 nM) for 4.5 hours and processed for ChIP assay as in Figure 5D except DNA was immunoprecipitated with a Stat5a antibody (sc-1081). Data are presented as fold P4-induced Stat5a recruitment (P4/EtOH), setting the vehicle-treated MECs to 1. C, Immunoblot (IB) of Y694 phosphorylation of Stat5a. Primary MECs from adult virgin WT mice grown in Matrigel were analyzed by immunoblot for phosphorylated (Y694) Stat5a and total Stat5a after 1 hour treatment with vehicle, R5020 (100 nM), or PRL (3 μg/mL). D, P4-induced nuclear localization of Stat5a in HC11 cells transduced with recombinant adenovirus to ectopically express mouse PR (B isoform). Cells were plated on coverslips and analyzed for intracellular localization of Stat5a and PR by IF after 2 hours of treatment with vehicle, R5020 (100 nM), or PRL (3 μg/mL). Images were captured at ×40 magnification and are representative for each hormone treatment. Results are mean ± SEM from 3 independent experiments. Student's t test was used to determine statistical significance. ns, not significant (P > .05); *, P < .05; **, P < .01; ***, P < .001.

Figure 8.

Deletion of Stat5a abrogates progestin induction of RANKL and other PR target genes. MECs isolated from WT or Stat5a KO mice (FVB) were cultured for 14 days in Matrigel and treated continuously with EtOH or R5020 (100 nM). RNA was extracted and analyzed either by gene-specific qPCR for expression of RANKL, Wnt4, and Areg (panel A) or by a TaqMan LDA qPCR assay (panel B) for a subset of known P4-induced target genes (GOI, gene of interest). PR protein was detected by IHC (panel C) and cell proliferation was quantified as percent positive Ki67 cells by IHC (panel D). Images were captured at ×40 magnification (bar, 50 μm). Results are mean ± SEM from 3 independent experiments. Two-way ANOVA was used to determine statistical significance. ns, not significant (P > .05); *, P < .05; **, P < .01; ***, P < .001. GAPDH, glyceraldehyde 3-phosphate dehydrogenase.

Figure 9.

Stat5a is required for PR recruitment to selective RANKL enhancers. A, Primary MECs from WT or Stat5a KO mice (FVB) were treated in vitro with EtOH or P4 (100 nM) for 4.5 hours and processed for ChIP assay with a PR antibody, and immunoprecipitated DNA was amplified by qPCR with primers flanking each RANKL enhancer (D1–D6) as in Figures 5 and 7. Data were displayed as fold P4 recruitment of PR to each enhancer over EtOH, setting the vehicle to 1. Results are mean ± SEM from 3 independent experiments. A two-way ANOVA was used to determine statistical significance. ns, not significant (P > .05); *, P < .05; **, P < .01; ***, P < .001. B, Mechanistic model for coregulation of RANKL by PR and Stat5a. PR and Stat5a convergence at enhancer regions is required for P4 induction of RANKL gene expression. In response to P4, Stat5a is recruited with PR to bind to enhancer regions of RANKL and may have distinct roles at different enhancers. Enhancers D4, D5, and D6 are proposed to contain composite elements that require both PR and Stat5a for stable DNA binding, or that Stat5a through protein tethering stabilizes PR-DNA binding. With enhancers D1 and D3 in which Stat5a is not required for PR DNA binding, Stat5a corecruited to these sites may act as a PR coactivator. TSS, translation start site.