ARID1A and PGR proteins interact in the endometrium and reveal a positive correlation in endometriosis

In order to examine the co-localization of ARID1A and PGR proteins, we performed double immunofluorescence for ARID1A and PGR in human endometrial tissue from the proliferative, early secretory, and mid secretory phases ( Figure 1A ). The expression of PGR was strongly detected in epithelium (66.43 ± 9.73%) and stroma (76.31 ± 3.58%) at the proliferative phase. At the mid secretory phase, the expression of epithelial PGR expression was significantly reduced (1.08 ± 0.50%) compared to early secretory (p<0.001). ARID1A was strongly detected in the epithelial cells at the proliferative (99.75 ± 0.17%), early secretory (99.85 ± 0.13%), and mid secretory (100.00 ± 0.00%) phases. Stromal cells were 18.03 ± 7.44% and 29.97 ± 8.28% ARID1A-positive at the proliferative phase and early secretory phase, respectively. Interestingly, stromal ARID1A positive cells were most highly detected at the mid secretory phase (74.07 ± 11.32%). We next analyzed the co-localization of PGR and ARID1A during the phases of the menstrual cycle ( Figure 1B ). Co-expression of epithelial ARID1A and PGR was highly detected in proliferative and early secretory phases, 66.18 ± 9.69% and 80.50 ± 7.85% respectively. However, co-expression of PGR and ARID1A at epithelium was significantly reduced at mid secretory phase (1.08 ± 0.50%) compared with proliferative and early secretory phase (p <0.001). In contrast, co-expression of ARID1A and PGR in the stroma was continuously increased, and co-expression was the highest in the mid secretory phase (45.60 ± 14.61%) compared to proliferative and early secretory phases (8.27 ± 6.02% and 13.71 ± 7.38% respectively). These results suggest that the co-localization of ARID1A and PGR is dependent on the menstrual cycle, and their tight regulation is important in the human endometrium. In order to determine whether ARID1A physically interacts with PGR in the endometrium, we performed co-immunoprecipitation for ARID1A in mouse and human endometrium samples. The results of co-immunoprecipitation for ARID1A proteins showed that ARID1A interacts with PGR-A not PGR-B in the wild-type mouse uterus at gestational day (GD) 3.5 ( Figure 2A ). Next, we performed co-immunoprecipitation for ARID1A in endometrial samples from women without endometriosis at secretory phase. Similar to the result from the mouse tissue, ARID1A interacted with PGR-A not PGR-B in human endometrial tissue ( Figure 2A ). To determine the existence of protein interactions and their localization within endometrial compartments, we applied proximity ligation assay (PLA), a sensitive tool to assess protein-protein interaction and localization, using the Duolink in situ PLA kit [ 23 ]. To examine the protein-protein interaction between ARID1A and PGR-A by PLA, we transiently transfected ARID1A or/and PGR-A into HEK293T cells. Red fluorescence marking interaction between the two proteins was strongly detected in the ARID1A plus PGR-A transfected cells compared to non-transfected controls or ARID1A-only or PGR-A only transfected cells ( Figure 2B ). We next examined the interaction between ARID1A and PGR-A in the human endometrium. PLA showed that ARID1A and PGR were highly co-localized in the epithelium and stroma during the proliferative and early secretory phases. Stromal co-localization was increased during the mid-secretory phase ( Figure 2C ). Overall, our results showed that ARID1A directly interacts with PGR-A in both the mouse and human endometrium. We previously found that the expression of ARID1A proteins was significantly reduced in the endometrium of women with endometriosis compared to controls [ 20 ]. To determine whether the level of ARID1A protein correlates with PGR levels in women with endometriosis, we performed Western blot analysis for ARID1A and PGR in human endometrial samples during the early-, mid-, and late-secretory phases ( Figure 3A ). Quantification and correlation analysis of the Western blot results revealed a significant positive correlation of ARID1A expression levels with PGR-A (Spearman correlation coefficient = 0.7178, p < 0.0001) and with PGR-B (Spearman correlation coefficient = 0.6027, p = 0.0001) in women with endometriosis ( Figure 3B ). These results confirmed that ARID1A and PGR protein levels positively correlate in endometriosis patients. Having established the link between endometriosis development and ARID1A attenuation in the eutopic endometrium, we sought to determine if ARID1A depletion is involved in endometriotic lesion development. To assess the effect of ARID1A deficiency in endometriosis development, we surgically induced endometriosis in control ( Pgr cre/+ ), Pgr cre/cre (PRKO), and Pgr cre/+ Arid1a f/f mice using one of their uterine own horns. One month after endometriosis induction, the number of endometriotic lesions was significantly increased in PRKO and Pgr cre/+ Arid1a f/f mice compared to controls (p<0.01 and p<0.05, respectively; Figure 4 ). These results demonstrate that ARID1A and PGR attenuation in ectopic lesions increased endometriosis development.

The human endometrial samples were collected from Wake Forest University and University of North Carolina. The study has been approved by Institutional Review Committee of each institution. Thirty-two samples (13 proliferative, 13 early secretory, and 6 mid secretory) were used to investigate co-localization of PGR and ARID1A in control human menstrual cycles. To assess the correlation of PGR and ARID1A expression patterns in women with endometriosis, 13 early-secretory, 12 mid-secretory phase, and 10 late-secretory phase samples from women with endometriosis were analyzed by Western blot. An experienced fertility specialist (B.A.L.) confirmed the human menstrual cycles by histological analysis. Animals were maintained in a designated animal care facility according to Michigan State University’s Institutional Guidelines for the care and use of laboratory animals. All animal procedures were approved by the Institutional Animal Care and Use Committee of Michigan State University. For induction of endometriosis, Pgr cre/+ , Pgr cre/cre (PRKO), and Pgr cre/+ Arid1a f/f mice were injected with 1 μg/ml of E2 daily for 3 days. Induction of endometriosis were performed 6 hours after the last E2 injection. One uterine horn was removed under anesthesia. The uterine horn was opened longitudinally with scissors and then cut into small fragments of about 1 mm 3 in a Petri dish containing phosphate-buffered saline (PBS; pH 7.5). Small fragments were injected back into the peritoneum of the same mouse. After one month, the mice were euthanized, and endometriosis-like lesions were collected and counted. HEK293T cells were cultured in Dulbecco’s modified Eagle’s medium (DMEM; Gibco BRL, Gaithersburg, MD) with 10% (v/v) fetal bovine serum (FBS; Gibco BRL, Gaithersburg, MD), and 1% (v/v) penicillin streptomycin (P/S; Gibco BRL, Gaithersburg, MD) at 37°C under 5% CO 2 . HEK293T cells were transfected with 0.5 μg of ARID1A, PR-A, and ARID1A with PR-A using Lipofectamine 2000 reagent (Invitrogen Crop., Carlsbad, CA) according to the manufacturer’s instructions. Endometrial sections were blocked with 10% normal goat serum in phosphate-buffered saline (PBS, pH 7.5) for immunofluorescence analysis. Sections were exposed to appropriate primary antibodies: anti-ARID1A (sc98441, Santa Cruz Biotechnology, Dallas, TX) or anti-PGR (sc810, Santa Cruz Biotechnology, Dallas, TX) in 10% normal goat serum in PBS (pH 7.5) overnight at 4°C. Sections were incubated with the appropriate following secondary antibodies: Alexa Fluor 488-conjugated anti-mouse IgG (Invitrogen, Grand Island, NY) and Alexa Fluor 594-conjugated anti-rabbit IgG (Invitrogen, Grand Island, NY). Nuclei of the cells were counterstained with 4’,6‐diamidino‐2‐phenylindole (DAPI, H-1200, Vector Laboratories, Burlingame, CA). Immunoprecipitation was performed as described previously [ 21 ]. Briefly, 500 μg of lysates were immunoprecipitated with 1 μg of anti-ARID1A antibodies (Santa Cruz Biotechnology, Dallas, TX) with 30 μl of resuspended protein A-agarose (Pierce Biotechnology, Rockford, IL) and incubated overnight at 4°C. Immunocomplexes were subjected via SDS-PAGE and transferred onto polyvinylidene difluoride membrane (PVDF, Millipore, Bedford, MA). The membrane was exposed to anti-ARID1A (Santa Cruz Biotechnology, Dallas, TX) or anti-PGR (Santa Cruz Biotechnology, Dallas, TX) antibodies. After the incubation with a horseradish peroxidase-linked secondary antibody, immunoreactivity was visualized with ECL reagents (GE Healthcare, Pittsburgh, PA). Western blot analysis was performed as previously described [ 22 ]. Samples containing 15 μg of proteins were applied to SDS-PAGE. The separated proteins were transferred onto a PVDF membrane (Millipore, Bedford, MA). Membranes were blocked overnight with 0.5% casein (weight/volume) in PBS with 0.1% Tween 20 (volume/volume) (Sigma Aldrich, St. Louis, MO) and probed with antibodies. Following the incubation with a horseradish peroxidase-linked secondary antibody, immunoreactivity was visualized by treatment with ECL reagents (GE Healthcare, Pittsburgh, PA). For loading control, the membrane was stripped and probed with anti-ACTIN (Santa Cruz Biotechnology, Dallas, TX). Duolink in situ PLA was performed as previously described [ 23 ]. Duolink in situ PLA analysis was performed according to the manufacturer’s instructions (Olink Biosciences, Watertown, MA). Briefly, paraformaldehyde-fixed cells were washed with PBS, incubated for 10 min in 0.1% (volume/volume) Triton X-100, washed, and blocked with blocking solution. Primary antibodies, against ARID1A and PGR, were applied, and the cells were incubated with PLUS and MINUS secondary PLA probes against rabbit and mouse IgGs. The incubation was followed by hybridization and ligation, and then amplification was performed. After mounting with Duolink mounting medium, samples were examined using a Zeiss 510 Meta (NLO) Confocal Microscope (Carl Zeiss Microscopy, Thornwood, NY). Statistical analyses were performed using one-way ANOVA analysis, Tukey’s post hoc multiple range test or Student’s t-tests using the Instat package from GraphPad (San Diego, CA). p<0.05 was considered statistically significant.

Previous studies have established significant roles for both ARID1A and PGR individually in endometrial function and in endometriosis [ 12 , 20 , 24 ]. However, the relationship between the two proteins was not clear. In this study, we comprehensively examined the physical relationship between ARID1A and PGR proteins in the endometrium. Our co-immunoprecipitation experiments revealed that ARID1A directly interacts with PGR in the human and mouse endometrium. Furthermore, our immunostaining and PLA assay results showed that ARID1A and PGR colocalize and interact in the human endometrium in different tissue compartments during the different phases of the menstrual cycle. ARID1A, a chromatin remodeling complex protein that has a critical role in endometriosis, infertility and steroid hormonal signaling, has been reported as an important transcriptional regulator, particularly for nuclear hormone-induced transcription and expression of cell cycle regulators. Mice with conditional ablation of Arid1a in Pgr positive cells ( Pgr cre+ Arid1a f/f ) were infertile due to loss of epithelial PGR, which led to enhanced E2-induced epithelial cell proliferation during embryo implantation Study of epithelial cell specific PRKO mice has shown that P4 is unable to stimulate expression of its epithelial target genes without PGR in the uterine epithelium [ 25 ]. The mutant mice are also infertile due to defects in embryo attachment, decidualization, and inability to cease E2-induced epithelial cell proliferation [ 25 ]. Therefore, our results together with previous data suggest that interaction between ARID1A with PGR proteins plays an important role in regulating P4 and E2 for normal uterine function. ARID1A has been functionally linked to a number of transcription factors [ 26 , 27 ]. Steroid hormone receptors depend on SWI/SNF chromatin remodeling activity for gene transactivation [ 28 , 29 ]. ARID1A contains several putative nuclear hormone receptor-binding sites (LXXLL motifs), and SWI/SNF complexes interact with several nuclear receptors, including glucocorticoid receptors and vitamin D3 receptors, to activate transcription of specific target genes [ 30 , 31 ]. Our results suggest ARID1A is pivotal in regulating transcription of PGR target genes. However, it is possible that PGR is a target gene of ARID1A. Therefore, further study is needed to identify possible PGR binding sites for ARID1A to understand how ARID1A regulates PGR activity for transcriptional activation using in silico analysis such as the CisMols Analyzer or ChIP-seq analyses. We observed that ARID1A physically interacted with PGR-A but not with PGR-B in the mouse and human endometrium, consistent with our previous in vitro data [ 20 ]. PGR-A and PGR-B are transcribed from distinct, estrogen-inducible promoters within a single gene [ 8 ]. The network between the epithelial and stromal compartments of uterine endometrium undergoes dynamic molecular and morphological changes to prepare for embryo implantation and development. Estrogen stimulates the proliferation of uterine epithelial cells, progesterone inhibits this proliferation, and both mediate these changes by activating transcription of target genes through binding their cognate receptors [ 12 ]. The sequential actions of E2 and P4 on uterine cells is critical for successful embryo apposition, attachment, implantation, and pregnancy maintenance [ 20 ]. PGR in humans mediates the role of P4 in reproductive physiology including endometrial differentiation, ovulation, implantation, and successful embryo development [ 7 ]. ARID1A loss is uniquely associated with endometriosis-related ovarian neoplasms [ 32 , 33 ], but the effects of ARID1A loss have not been fully investigated in infertility and endometriosis. To lay a foundation to overcome the effects of ARID1A loss therapeutically, it will be important to examine down-stream effectors that mediate ARID1A gene actions and, ultimately, the aberrant endometrial epithelial proliferation associated with infertility. ARID1A is pivotal in promoting implantation and decidualization, and preventing endometriosis through enhancing PGR activities [ 20 ]. Previously, we found that ARID1A levels are remarkably lower in endometrium of infertile women with endometriosis compared to controls [ 20 ]. The results from Pgr cre/+ Arid1a f/f mice and infertile women with endometriosis suggest that ARID1A loss is one of the causes of infertility and endometriosis by aberrant activation of E2 signaling and P4 resistance. Our study showed a strong positive correlation between ARID1A and PGR proteins in women with endometriosis along with enhanced endometriosis development in endometriosis model mice with ARID1A or PGR deficiency. This finding suggests that the action of both ARID1A and PGR in tandem is necessary to suppress development of endometriosis. Symptomatic endometriosis significantly affects a woman’s quality of life. Characteristically the condition causes severe pain, and half of women with endometriosis are infertile [ 34 – 37 ]. The current therapies, surgical removal of lesions and hormonal suppression, have side effects and a high incidence of relapse [ 38 , 39 ]. By identifying the molecular mechanisms of ARID1A and PGR in the early pathogenesis of endometriosis, we will be in a much better position to develop new therapies against the pain of endometriosis and heartbreak of infertility. Our results demonstrate co-localization and protein-protein interactions of ARID1A and PGR in both the epithelial and stromal cells of human endometrium. The results together suggest that ARID1A and PGR play a critical role in endometrial function and that their expression is important for suppressing the development of endometriosis. Our results will allow better targeted experiments in the present and better targeted therapeutic approaches for endometriosis in the future.

Endometriosis, a disorder in which the endometrial cells that normally line the uterus proliferate outside the uterine cavity, is a major cause of pelvic pain and infertility. The disease affects 5–10% of women at reproductive age and 176 million women worldwide [ 1 ]. Symptomatic endometriosis causes both social and psychological effects mostly because of the inability to consistently work and decreased quality of life. Estimated costs to diagnose and treat endometriosis exceed $22 billion in the United States [ 2 ]. Endometriosis prevalence increases to as much as 50% in women with infertility [ 3 , 4 ]. Several mechanisms have been proposed to explain the link between endometriosis and infertility [ 5 , 6 ], but a direct cause-effect relationship has not been established. Therefore, identifying the mechanisms contributing to the early pathogenesis of endometriosis will enhance the development of new therapies against endometriosis-related infertility. The progesterone receptor (PGR) mediates the role of progesterone (P4) in reproductive physiology including the processes of endometrial differentiation, ovulation, implantation, successful development of the embryo, development of the mammary gland, and the regulation of central signals from the hypothalamic–pituitary axis [ 7 ]. The PGR belongs to the nuclear receptor family and exists as two main isoforms, PGR-A (94 kDa) and PGR-B (114 kDa), which are transcribed from a single-copy progesterone receptor (PGR) gene [ 8 ]. Establishment of endometrial receptivity by the sequential actions of estrogen (E2) and P4 on uterine cells is critical for successful embryo apposition, attachment, implantation, and pregnancy maintenance [ 9 , 10 ]. P4 inhibits endometrial epithelial proliferation through PGR regulation. P4 and its receptor isoforms, PGR-A and -B, have established, important roles in endometriosis [ 11 , 12 ]. The endometrium of women with endometriosis frequently exhibits an attenuated response to P4 suggesting the presence of an endometrial progesterone resistance phenotype [ 13 ]. The AT-rich interaction domain 1A (ARID1A) protein is a SWI/SNF nucleosome remodeling complex family protein thought to play a tumor-suppression role in endometrioid and clear cell ovarian carcinomas and other endometriosis-associated cancers [ 14 ]. Like other SWI/SNF-related proteins, ARID1A regulates the transcription of genes normally inactive due to closed chromatin conformation, and its specific role may be to provide gene target specificity [ 15 ]. By loosening chromatin binding to histones in an ATP-dependent manner, chromatin remodeling complexes like this one restructure nucleosomes to change the expression pattern of the surrounding DNA [ 16 ]. Several studies have linked ARID1A and SWI/SNF to transcriptional regulation, particularly nuclear hormone-induced transcription and expression of cell cycle regulators. The loss of ARID1A protein expression because of ARID1A mutations [ 17 ] has been found in 46% of ovarian clear-cell carcinomas and 30% of endometrioid ovarian carcinomas [ 18 , 19 ]. Previously, we found that ARID1A levels are remarkably lower in endometrium from women with endometriosis compared to healthy women [ 20 ]. Mice with conditional ablation of Arid1a in Pgr -positive cells ( Pgr cre+ Arid1a f/f ; Ard1a d/d ) were infertile due to loss of epithelial PGR, which led to enhanced E2-induced epithelial cell proliferation during embryo implantation [ 20 ]. Despite this link between ARID1A and PGR in the endometrium, there are currently no studies investigating the relationship between ARID1A and PGR in the context of endometriosis. In the present study, we examined their physical protein interactions in human endometrial tissue using immunoprecipitation and PLA analysis. Our immunostaining analysis revealed colocalization and strong correlation of ARID1A and PGR proteins in the endometrium. Furthermore, deletion of these proteins in the uteri of mice induced with endometriosis leads to similar increases in lesion numbers. These results suggest that the interaction of ARID1A and PGR plays a role in endometrial function and is important to consider in seeking to understand the development of endometriosis.