Characterization of Molecular Changes in Endometrium Associated With Chronic Use of Progesterone Receptor Modulators: Ulipristal Acetate Versus Mifepristone

Ulipristal acetate (UPA) is a selective progesterone receptor modulator (PRM), which is used as an emergency contraceptive in women. Recent studies demonstrated the efficacy of a UPA contraceptive vaginal ring (UPAICVR) as a blocker of ovulation. However the endometrium of women exposed to UPA over a sixImonth period display glandular changes, termed PRMI associated endometrial changes (PAECs). We, therefore, investigated whether UPAIinduced PAECs are associated with altered expression of the transcription factor HAND2 whose down regulation is observed in endometrial epithelial hyperplasia and cancer. Our results showed that while exposure to mifepristone, a wellIknown PRM, leads to suppression of endometrial HAND2 expression, longIterm exposure to UPAICVR did not cause down regulation of this marker. Further studies, using human primary endometrial stromal cells, confirmed that whereas mifepristoneImediated suppression of HAND2 elevated the levels of its downstream target fibroblast growth factor 18, UPA did not significantly alter the expression of this growth factor. A rationale for the differential regulation of HAND2 by these PRMs was provided by our observation that mifepristoneIbound progesterone receptors turn over at a faster rate than those bound to UPA. Collectively, these results support the selective effects of different PRMs and indicate that chronic exposure to UPA does not alter the HAND2 pathway whose dysregulation is linked to complex atypical endometrial hyperplasia and cancer. The results from this study involving a limited number of clinical samples should pave the way for a larger study to determine the safety of UPA for longIterm use. Key words: Ulipristal acetate; Mifepristone; Contraception; Endometrium; HAND2 Page 2 of 28 2 Reproductive Sciences

It is estimated that 225 million women worldwide lack access to effective and acceptable contraceptive methods. Therefore, the development of novel clinically safe and effective

of fertility control remains a necessity. The steroid hormone progesterone (P) acting through its nuclear receptor critically controls the ovulatory process as well as endometrial function in the human. Progesterone receptor modulators (PRM) are synthetic compounds that interact with the progesterone receptor (PR) to suppress ovulation and/or induce endometrial atrophy, resulting in amenorrhea, a condition that is perceived favorably in many cultures around the world 1. Therefore, the development and use of PRMs as contraceptives is of particular interest. Ulipristal acetate (UPA), also referred to as VA/CDBI2914, is a new and promising PRM 2I6. UPA has been approved as an emergency contraceptive 7, 8 in the United States and abroad and as a treatment for heavy menstrual bleeding due to uterine fibroids 9, 10 in Canada and Europe. Successful use of this PRM as an emergency contraceptive has raised the possibility that a simplified continuous delivery of UPA could improve longIterm contraceptive safety and efficacy and compliance. With this goal in mind, a UPA contraceptive vaginal ring (UPAICVR) was designed for longIterm contraceptive use by the Population Council, New York. In a study conducted by the Council, healthy women with normal baseline ovulation were randomized to receive UPAICVR for two consecutive 12Iweek treatment periods, followed by a recovery cycle 11. The results from these studies indicated that the UPAICVR has the potential to become an effective longIacting, userIcontrolled contraceptive. However, endometrial biopsies taken at the end of the treatment period displayed histological glandular changes, described as PRMI associated endometrial changes (PAECs) 11. While these endometrial changes are considered to be benign due to the lack of cytological atypia12 an inIdepth study is needed to confirm the absence of any endometrial abnormality, including hyperplasia, following chronic PRM use. The endometrium, the innermost layer of the uterus, undergoes proliferation and differentiation in a cyclical manner in response to the steroid hormones, 17βIestradiol (E) and P acting via their cognate receptors 13I15. While E acting via ERα f unctions as a mitogen and promotes the growth and proliferation of the endometrial epithelium in a cyclical fashion during the reproductive Page 3 of 28 3 Reproductive Sciences cycle, P acting via PR inhibits EIinduced epithelial proliferation and causes differentiation. Uncontrolled proliferation of the endometrial epithelium results in alterations of glandular architecture (shape and size) and an increase in endometrial glandItoIstroma ratio, leading to endometrial hyperplasia 15I17. The majority of cases of endometrial hyperplasia are associated with compromised P signaling that fails to oppose E signaling 17I19. We have previously shown that the transcription factor HAND2 (HeartI and neural crest derivativesIexpressed protein 2), which is regulated by the PR present in the endometrial stroma, is a key mediator of the wellIknown antiIproliferative effect of P on the endometrial epithelium 20. HAND2 suppresses the production of several stromal fibroblast growth factors (FGFs), which act in a paracrine manner via the FGF receptors to promote epithelial proliferation. Therefore, in the absence of HAND2, the endometrial epithelium undergoes unbridled FGFIinduced proliferation that leads to complex atypical hyperplasia. It is of interest to note that the HAND2 gene locus is prone to epigenetic alterations. Our recent studies revealed that the HAND2 gene is a hypermethylated and silenced in endometrial hyperplasia and cancer 21. When compared to other frequent DNAIbased alterations in endometrial cancers, such as p53, PTEN, and PIK3CA mutations, HAND2 hypermethylation was found to be the most common 21. Since the down regulation of HAND2 expression is linked to endometrial hyperplasia and cancer, we examined the expression of this factor in endometrial biopsies of women exposed to UPAICVR for 24 weeks. We also compared the endometrial effects of UPA with those of mifepristone, a wellI known PRM. Ma terials and Methods Endometrial biopsies Endometrial biopsy samples were obtained using either a Pipelle (Cooper Surgical, Trumbull, CT, USA) (DR, Chile) or an Explora (Cooper Surgical) (Oregon) device. A portion of wach sample was placed for use in 10% neutral buffered formaldehyde for histology and immunohistochemistry studies. Page 4 of 28 4 Reproductive Sciences In vitro decidualization of human endometrial stromal cells (HESC) Our studies involving primary HESC cultures follow the regulations stated for the protection of human subjects participating in clinical research and are approved by the institutional review boards of Emory University, Wake Forest University (WinstonISalem, North Carolina), and the University of Illinois at UrbanaIChampaign (UIUC). Endometrial samples from the early proliferative stage of the menstrual cycle were obtained by Pipelle biopsy at Emory University and Wake Forest Medical Centers from fertile, regularly cycling volunteers with no sign of uterine abnormality, providing written informed consent as described previously 21, 22. C ells were cultured in DMEM/FI12 medium (Invitrogen) supplemented with 5% (vol/vol) fetal bovine serum (Hyclone), 50 µg/mL penicillin, and 50 µg/mL streptomycin (Invitrogen). For in vitro differentiation, the cells were treated with differentiation cocktail composed of 10 nM E (Sigma), 1 µM progesterone (Sigma), 0.5 mM 8IbromoadenosineIcAMP (Sigma), 10 µM UPA or mifepristone in DMEM/FI12 medium (Invitrogen) supplemented with 2% (vol/vol) charcoal dextranIstripped fetal bovine serum for 0I6 days. At the end of the culture (2 or 6 days), the cells were detached from the plates, counted, and stored at I80°C for RNA extraction. Additionally some cells were fixed for immunocytochemical (ICC) analysis. In some experiments, the cells were treated with differentiation cocktail composed of 10 nM E, 1 µM progesterone, 0.5 mM 8I bromoadenosineIcAMP, 5 µM UPA or mifepristone in DMEM/FI12 medium supplemented with 2% (vol/vol) charcoal dextranIstripped fetal bovine serum for 0I6 days. Cultures were terminated at days 2 to 6 for RNA extraction. Chemicals, reagents, and antibodies Progesterone (P), 17βIestradiol (E), naphthol ASIMX phosphate, Fast Blue RR (4I benzoylaminoI 2,5Idimethoxyaniline diazonium), collagenase, pancreatin, dimethyl sulfoxide (DMSO), 8Ibromoadenosine 3', 5'Icyclic monophosphate salt (cAMP), and Trypan blue were purchased from Sigma. Hanks Balanced Salt Solution (HBSS), dispase, Dulbecco’s modified Eagle mediumIF12 medium HEPES, no phenol red (DMEM/F12), PenicillinIStreptomycin, and Fungizone, were purchased from Life Technologies. Fetal bovine serum (FBS) was purchased from Fisher Scientific. FluoromountIG with DAPI was purchased from eBiosciences. Page 5 of 28 5 Reproductive Sciences Endometrial sections or endometrial stromal cells were incubated with one or more of the following primary antibodies: heartI and neural crest derivativesIexpressed transcript 2 (HAND2, 1:250, Santa Cruz Biotechnology antibody SCI9409), FGF18 (1:100, Santa Cruz Biotechnology antibody SCI393471), PR (1:100, DAKOIA0098), and PRB22 (1:300, Cell s ignaling CSTI31575). The fluorescentItagged secondary antibodies and normal donkey serum were purchased from Jackson ImmunoResearch. The following secondary antibodies were used: rhodamine or Cy3 donkey antiIrabbit, 488 donkey antiIrabbit, 488 donkey antiImouse, 488 donkey antiIgoat, and Cy3 donkey antiIrat. Immunohistochemistry (IHC) and immunocytochemistry (ICC) ParaffinIembedded endometrial biopsy sections were subjected to IHC as described previously. Tissue sections were deparaffinized in xylene, rehydrated through a graded series of ethanol, and washed in tap water. For most of the immunostaining, antigen retrieval was performed in a pressure cooker in 10 mM sodium citrate buffer (pH 6.0) for 20 min and then the slides were cooled to room temperature. The sections were washed between steps (three times for 5 min each) using 1x phosphateIbuffered saline solution containing 0.05% Tween 20 (PBSIT). Nonspecific binding was inhibited by incubating the sections with 10% normal serum for 1 h at room temperature. After the serum block, sections were incubated overnight at 4°C with the diluted antibody solution in PBSIT containing 1% normal serum. Labeling was visualized by incubation with a fluorescentItagged secondary antibody for 1 h at room temperature. All incubations were done using a humidified chamber protected from light. Slides were mounted using a mounting solution containing DAPI. Pictures were taken using the Olympus BX51 microscope equipped for fluorescent imaging and connected to a Jenoptik ProgRes C14 digital camera with cImount interface containing a 1.4 Megapixel CCD sensor. Fluorescent images were processed and merged using Adobe Photoshop Extended CS6 (Adobe Systems). HSCOREs were determined as described previously 23. F or ICC analysis of HESC, cells were fixed in 10% NBF for 10 min, and then washed with PBS. Cells were then permeabilized using PBS containing 0.1% Triton X for 10 min at room temperature. Nonspecific binding was inhibited by incubating the sections with 10% normal Page 6 of 28 6 Reproductive Sciences serum for 1 h at room temperature. After the serum block, the cells were incubated overnight at 4°C with the diluted antibody solution in PBS containing 1% normal serum. Labeling was visualized by incubation with a fluorescentItagged secondary antibody for 1 h at room temperature. One drop of mounting solution containing DAPI was added to each well to stain the nucleus. Pictures were taken using the Olympus Ix70 inverted microscope adapted to a Diagnostic Instrument digital camera containing a 2.0 Megapixel CCD sensor. Fluorescent images were merged and processed using Adobe Photoshop Extended CS6. Quantitative real time PCR analysis (qPCR) Total RNA was isolated from endometrial cells using a standard TRIzolIbased protocol. The RNA concentration of each sample was determined at 260 nm using a Nanodrop ND1000 UVI Vis spectrophotometer (Nanodrop Technologies). RNA samples were reverse transcribed using the High Capacity cDNA Reverse Transcription kit (Applied Biosystems) according to the manufacturer's instructions. Real time quantitative PCR (qPCR) reactions were carried out using SYBRIgreen master mix (Applied Biosystems) in a 7500 Applied Biosystems RealItime PCR machine (Applied Biosystems). For each sample, the mean threshold cycle (Ct) was calculated from Ct values obtained from three replicates. The normalized ∆Ct in each sample was calculated as mean Ct of target gene subtracted by the mean Ct of the reference gene. The fold change of gene expression in each sample relative to a control was generated using the 2−∆∆Ct mathematical model for relative quantification of quantitative PCR. The mean fold induction and SEM were calculated from at least three or more independent experiments. The housekeeping gene RPLP0 (36B4), which encodes a ribosomal protein, was used as a reference gene. Statistical analyses Experimental data for studies related to UPAICVR were collected from 12 independent subjects. For each subject, 4 endometrial biopsy samples were obtained. Biopsy 1 was an endometrial specimen obtained before administration of UPAICVR, biopsies 2 and 3 were endometrial specimens obtained after each 12Iweek period in which UPAICVR released UPA daily, and biopsy 4 was obtained following a 4Iweek postItreatment recovery period. Results from mifepristone studies were obtained from 6 independent clinical samples. Data related to primary HESCs were collected from 3 independent clinical samples, which were subjected to the same Page 7 of 28 7 Reproductive Sciences experimental conditions. All numerical data are expressed as mean ± SEM. When experimental samples were compared with control samples, statistical significance between the control and experimental sample was determined using the Student t test. A P value of ≤.05 was considered to be significant.

Expression of HAND2 is unaltered in human endometrial biopsies exposed to UPA-CVR Human endometrial biopsies were obtained from three different clinics located in the United States, Dominican Republic, and Chile. We have analyzed a total of 12 independent subjects. For each subject, 4 endometrial biopsy samples (biopsies 1I4) were obtained. Biopsy 1 is an endometrial specimen obtained before administration of UPAICVR during the luteal phase based on urine LH determinations. Biopsies 2 and 3 are endometrial specimens obtained after each 12I week period in which UPAICVR released 1.5 mg or 2.5mg UPA daily. Biopsy 4 was obtained following a 4Iweek postItreatment recovery period in the luteal phase, determined as above. Figure 1 shows representative endometrial samples at baseline, before administration of UPAI CVR (panel A), after exposure to UPAICVR (panel B), and in the recovery phase (panel C). Baseline samples show normal midIsecretory phase endometrium. Upon exposure to UPAICVR, the glands show variable cystic dilatation, mildly disordered architecture, nonIphysiological secretory appearances, and coexistent mitoses and apoptotic bodies. The stroma is compact, nonI decidualized and contains occasional thickIwalled vessels. These features are characteristic of PRMIassociated endometrial changes or PAECs. In the recovery phase, the endometrium exhibits normal early secretory phase appearances. To examine the molecular changes in the endometrium following prolonged exposure to UPAI CVR, we investigated the expression of HAND2 in the biopsy specimens. An intense nuclear staining specific to HAND2 was observed in the endometrial stromal cells of preItreatment biopsyI1 specimen (Figure 2). This expression of stromal HAND2 remained unaltered in the biopsies exposed to UPA (biopsyI2, and I3, Figs. 2B and C) and in the postItreatment biopsy Page 8 of 28 8 Reproductive Sciences specimen (biopsyI4, Fig. 2D). To more accurately quantify the immunohistochemical findings, HSCOREs were analyzed. HSCORES of endometrial HAND2 immunostaining revealed no significant changes across the treatment period (Figure 3). These results indicate that a CVR releasing 1.5 or 2.5mg/day of UPA for 24 weeks does not affect HAND2 expression. Expression of HAND2 is reduced in human endometrial biopsies exposed to mifepristone We also analyzed the expression of HAND2 in endometrial biopsies collected from women exposed to mifepristone, a wellIknown PRM. In this study, 50 mg of oral mifepristone was administered every other day for 12 weeks. Endometrial biopsies were taken in the secretory phase of the last week of mifepristone treatment. Luteal phase biopsies from unexposed women demonstrated robust expression of HAND2 in the nuclei of stromal cells, as expected. In contrast, endometrial biopsies of women treated with mifepristone showed a significant decline in the expression of HAND2 (Figure 4). Quantification of HAND2 immunoIpositive cells in the stroma revealed greater than 80% reduction in HAND2 expression in mifepristoneIexposed biopsies when compared to unexposed controls. Collectively, these results suggest that UPAI CVR and mifepristone have differential effects on endometrial HAND2 expression. It is possible that the differences are due to the pharmacology of the PRM compounds, their doses, duration or route of administration. UPA and mifepristone differentially regulate HAND2 and FGF18 expression in cultured human endometrial stromal cells To directly examine the pharmacological effects of UPA and mifepristone on HAND2 expression in the endometrial stroma under identical study conditions, we utilized a wellI established human endometrial stromal cell culture system. In this system, undifferentiated stromal cells isolated from human endometrial biopsies (HESC) obtained from normal women in the proliferative stage of the menstrual cycle were placed in culture and subjected to decidualization in response to a hormonal mixture containing 10 nM E, 1 µM P, and 0.5 mM 8I bromoIcAMP 24, 25. Under the treatment conditions, cells were treated with the hormonal mixture w ith or without 10 µM UPA or mifepristone. HESCs were cultured in the presence of hormones Page 9 of 28 9 Reproductive Sciences with or without PRMs for up to six days. HAND2 mRNA expression was reduced when endometrial stromal cells were exposed to UPA or mifepristone for two days, compared to cells not treated with PRMs (Figure 5). While UPA exposure reduced HAND2 expression by 20%, treatment with mifepristone resulted in almost 40% reduction in HAND2 expression (P<0.05). Further, the inhibitory effect of mifepristone on HAND2 expression increased in severity with longer duration of treatment. Stromal cells exposed to mifepristone for six days displayed more than 80% reduction in HAND2 expression when compared to untreated control cells. In contrast, treatment with UPA for six days had milder effects, resulting in a 30% reduction in HAND2 expression (P<0.05). Consistent with the RNA profile, immunocytochemical (ICC) analysis revealed similar reductions in HAND2 in the UPA (Figure 6, panel B)I and mifepristone (Figure 6, panel C) Iexposed endometrial stromal cells compared to UPAI or vehicleItreated stromal cells (Figure 6, panel A). To further investigate the differential effects of UPA and mifepristone on HAND2 expression, we reduced the levels of PRMs from 10Ifold to 5Ifold molar excess of P. Human endometrial stromal cells were cultured in the presence of hormones with or without 5 µM UPA or mifepristone for up to six days. HAND2 mRNA expression was monitored on day 2, day 3, day 4, day 5, and day 6 after initiation of the culture. As shown in Figure 7, treatment of HESC with 5 µM UPA did not affect the expression of HAND2 on days 2 to 6 upon initiation of the culture. By contrast, administration of 5 µM mifepristone led to a significant down regulation of HAND2 expression in HESCs. The decline in HAND2 expression was evident on day 2 and continued up to day 6 of culture. Our previous studies have shown that Hand2 expression in the stroma suppresses the production of fibroblast growth factors (FGFs) and inhibits cell proliferation 20. In the absence of Ha nd2, continued induction of FGFs in the stroma activates FGF receptor (FGFR) signaling in the epithelium to promote cell proliferation 20. Consistent with this observation, a recent study r eported a decrease in HAND2 expression and marked increase in the levels of FGF18 in human endometrial adenocarcinoma 26. MifepristoneItreated endometrial stromal cells demonstrated a m arked increase in the levels of FGF18 mRNA (Figure 8, upper panel) and protein (Figure 8, lower panel) compared to vehicleItreated controls. In contrast, endometrial stromal cells exposed Page 10 of 28 10 Reproductive Sciences to UPA did not exhibit alterations in FGF18 expression (Figure 8). Collectively, these results support our in vivo findings and indicate that endometrial stromal cells cultured with mifepristone or UPA under identical in vitro conditions exhibit differential effects on HAND2 and FGF18 expression. Mifepristone and UPA differentially affect PR stability in human endometrial stromal cells Both UPA and mifepristone are known to regulate the function of a tissue by modulating the activity of PR, so it is interesting that endometrial stromal cells display differential gene expression when exposed to the same concentrations of these two PRMs. We considered the possibility that the stability of endometrial PR might be regulated differentially by mifepristone and UPA. To investigate this possibility, we determined the expression of total PR protein in progesteroneI, UPAI, or mifepristoneItreated HESC by ICC. Cells exposed to progesterone or UPA for 6 days displayed prominent nuclear PR staining, while those treated with mifepristone showed markedly reduced levels of PR (Figure 9). Our recent studies revealed that the PR isoform PRIB plays a predominant functional role during human endometrial stromal differentiation by controlling the expression of a large number of target genes, including HAND2 21. We noted distinct expression of PRIB 22 in the nuclei of p rogesteroneI or UPAItreated stromal cells (Figure 10). In contrast, nuclei of stromal cells exposed to mifepristone were mostly devoid of PRIB expression. Taken together, these results are consistent with our view that UPA and mifepristone differentially affect PR stability in human endometrial stromal cells and this is reflected in altered expression of PR target genes, such as HAND2, in response to these ligands in the endometrial stroma. Di scussion A critical balance of E and P drives proper endometrial stromalIepithelial crosstalk and maintains normal uterine physiology. Disruption of PR function results in unopposed E action, causing epithelial hyperplasia and potentially carcinoma 17I19. HAND2, a PRIregulated gene in t he stromal cells, mediates the antiproliferative action of P to regulate endometrial epithelial Page 11 of 28 11 Reproductive Sciences function. Loss of the antiproliferative actions of P in the uterus has been linked to EIdependent endometrial cancer 27. Indeed, our recent study showed that the H AND2 gene is hypermethylated in premalignant endometrial lesions compared to normal endometrium and its expression is suppressed in endometrial hyperplasia and cancer 21. HAND2 has therefore emerged as a key molecular alteration in endometrial cancer that could potentially be employed as a biomarker for early detection of endometrial cancer. To determine the clinical utility of UPA as a longIterm contraceptive, it is critical to assess whether this compound, which effectively blocks PR action and ovulation, also alters the critical balance of E and P in the endometrium. Evaluation of endometrial histology following chronic UPA treatment revealed the presence glandular changes, known as PAECs, which did not show any cytological atypia, a characteristic feature of hyperplasia and cancer. However, routine histological examination of the endometrium may not provide molecular information related to a subtle imbalance of EI and PIdependent signaling that may arise due to PRM exposure. In this study, we show that the expression of HAND2, which critically regulates the balance of PI and EI dependent signaling in the endometrium, is unaffected in women exposed to UPAICVR continuously for 24 weeks. Since downregulation of endometrial HAND2 has been linked to complex atypical hyperplasia and cancer, unaltered expression of this factor gives us confidence that exposure to the studied dose of UPA by the vaginal route of administration does not disrupt the critical balance of EI and PI dependent signaling necessary for normal endometrial physiology. In contrast, we found that endometrial biopsies from women treated with mifepristone for 12 weeks displayed a dramatic downregulation of HAND2. Differential effects of UPA and mifepristone on HAND2 expression were confirmed in endometrial stromal cells cultured under identical conditions, suggesting distinct mechanisms underlie the actions of these PRMs. Analysis of PR in endometrial stromal cells following in vitro exposure to PRMs demonstrated that mifepristone down regulates the PR levels, whereas equivalent molar concentrations of UPA did not have these effects. This suppression of cellular PR levels by mifepristone is consistent with previous reports that addition of mifepristone to a progestogenIonly regimen of contraception leads to downregulation of PRIB 28. Additionally, recent studies have Page 12 of 28 12 Reproductive Sciences demonstrated that administration of mifepristone to an endometrial coIculture system causes suppression of PR expression compared to vehicleItreated controls 29. While the mechanism by w hich mifepristone causes PR turnover remains unclear, we believe that this downregulation of PR is in part responsible for the dramatic suppression of HAND2 expression observed in response to mifepristone compared to UPA. We have previously shown that HAND2 mediates the antiproliferative effects of P by suppressing the production of the FGF growth factors that mediate the growthIinducing effects of E on the endometrial epithelium. In the EIdominant proliferative endometrium, FGFs secreted from the stroma act on the FGFR(s) in the epithelium to promote proliferation 20. Following o vulation and in response to P production and signaling, HAND2 is induced in stromal cells, causing inhibition of FGF synthesis and attenuation of epithelial proliferation. Disruption of PR function in the endometrium therefore runs the risk of increasing FGF signaling, leading to inappropriate uterine epithelial growth, hyperplasia and cancer. Similar findings were noted in the epithelial glands of rhesus macaques treated with mifepristone 30 . Indeed, a recent study has s hown downregulation of HAND2 and upregulation of FGF18 in human endometrial adenocarcinoma 26. We demonstrate that administration of mifepristone to cultured endometrial s tromal cells caused inhibition of HAND2 expression and a concomitant enhancement of FGF18 expression. However, treatment of endometrial stromal cells with UPA did not significantly affect the expression of either HAND2 or FGF18, further confirming that UPA does not significantly alter the PIdependent antiproliferative pathways in the endometrium. In summary, this study shows that UPA and mifepristone exhibit differential effects on endometrial gene expression in vivo and in vitro, apparently due to differences in stability of PRs in response to these PRMs. It also confirms that chronic exposure to UPAICVR over a 24Iweek period does not lead to adverse effects, such as suppression of the expression of HAND2, which is reported to occur in endometrial hyperplasia and cancer. The results from this study involving a limited number of clinical samples should pave the way for a larger study to determine the safety of UPA for longIterm use. Page 13 of 28 13 Reproductive Sciences

The study was supported in part by a grant from the NICHD/NIH U54 HD 29990. Figure legends Fig. 1: Endometria of women exposed to UPA-CVR display PRM-associated endometrial changes (PAECs). Haematoxylin & eosin staining of endometrial sections obtained from normal midIsecretory phase endometrium (panel A), UPAICVR releasing 2.5mg UPA daily for two consecutive 12Iweek treatment periods (panel B), and postItreatment recovery period in the luteal phase (panel C). Note cystic dilatation, mildly disordered architecture, and nonI physiological secretory appearances in the endometria of women with UPAICVR. These endometrial samples are part of the large clinical trial. Representative images are shown. Fig. 2: Expression of HAND2 in human endometrial biopsies exposed to UPA-CVR for 24 weeks. Immunohsitochemical localization of HAND2 in endometrial sections before and after exposure to UPAICVR. A total of 12 independent subjects were analyzed and for each subject, 4 endometrial biopsy samples were obtained (N=48). Panel A represents endometrial specimen obtained during the luteal phase before administration of UPAICVR. Panels B and C indicate endometrial specimens obtained after each 12Iweek period with UPAICVR releasing 1.5mg or 2.5mg UPA daily. Panel D represents endometrial specimen collected during the luteal phase following a postItreatment recovery period. Panel E shows endometrial sections from a biopsy sample after a 12Iweek exposure to UPAICVR and subjected to IHC protocol omitting the primary antibody. Red staining indicates positive staining for HAND2 in endometrial sections. Representative images are shown. S and E indicate stroma and epithelium respectively. Fig. 3: HAND2 expression is unaltered in human endometrial biopsies exposed to UPA- CVR. The percentages of the immunostaining positive cells for HAND2 were analyzed by ImageJ software. The values represent mean ± SEM of twelve independent samples (N=5 for UPAICVR releasing 1.5 mg and N=7 for UPAICVR releasing 2.5mg UPA daily) with a total of N=48 clinical samples. No obvious doseIresponse effects were noted between the two doses. Page 14 of 28 14 Reproductive Sciences Fig. 4: Expression of HAND2 is reduced in human endometrial biopsies exposed to mifepristone. Upper: Immunohistochemical analysis of HAND2 in human endometrium before (panel A) and after administration of 50 mg of oral mifepristone every other day for 12 weeks of the menstrual cycle (panel B). Representative images are shown. Lower: HSCORES of HAND2Ipositive cells in the endometrium revealed a significant reduction in HAND2 expression in mifepristoneIexposed biopsies when compared to unexposed controls (N=6). B indicates baseline (0 wks) and after 12 weeks (12 wks) indicates end of treatment, respectively, and shows a significant decrease (P<0.02). Fig. 5: UPA and mifepristone differentially regulate HAND2 mRNA expression in human endometrial stromal cells. Primary cultures of human stromal cells were grown in Dulbecco's modified Eagle's medium/FI12 medium containing 5% charcoalIstripped fetal bovine serum. The cells were treated with a hormone mixture containing 10 nM E, 1 µM P, 0.5 mM 8IbromoI cAMP, and 10 µM UPA, mifepristone or vehicle for 6 days. Cells were harvested 2 days (left panel) or 6 days (right panel) after addition of hormone mixture. Total RNA was isolated and subjected to qPCR using primers for HAND2. The level of Rplp0 was used as an internal control to normalize gene expression. The values are presented as the mean fold induction ± SEM, P<0.05. Fig. 6: UPA and mifepristone differentially regulate HAND2 protein expression in human endometrial stromal cells. Immunocytochemical analysis of HAND2 in stromal cells during in vitro decidualization. Panels represent primary cultures of human endometrial stromal cells cultured in the absence of UPA or mifepristone (A), in the presence of UPA (B), in the presence of mifepristone (C) for 6 days. Representative images from three independent experiments are shown. Fig. 7: UPA and mifepristone differentially regulate HAND2 mRNA expression in human endometrial stromal cells. Primary cultures of human stromal cells were treated with a hormone mixture containing 10 nM E, 1 µM P, 0.5 mM 8IbromoIcAMP, and 5 µM UPA, mifepristone or vehicle for 6 days. Cells were harvested 2, 3, 4, 5, and 6 days after addition of hormone mixture. Total RNA was isolated and subjected to qPCR using primers for HAND2. The level of Rplp0 Page 15 of 28 15 Reproductive Sciences was used as an internal control to normalize gene expression. The values are presented as the mean fold induction ± SEM, *P<0.05, **P<0.01. Fig. 8: Downregulation of FGF18 expression in response to mifepristone in human endometrial stromal cells. Human endometrial stromal cells were subjected to differentiation in response to 0.5 mM 8IbromoIcAMP, 1 µM P, 10 nM E, and 10 µM UPA or mifepristone for 6 days. Upper. Total RNA was isolated and subjected to qPCR using primer for FGF18. YIaxis indicates fold induction. The level of Rplp0 was used as an internal control to normalize gene expression. The data are represented as the mean fold induction ± SEM from three separate samples. Lower. Immunocytochemical analysis of FGF18 expression in endometrial stromal in the absence of UPA or mifepristone (left panel), in the presence of UPA (middle panel) and in the presence of mifepristone (right panel). Representative images are shown. Fig. 9: PR stability in response to UPA or mifepristone in human endometrial stromal cells. Primary cultures of human stromal cells were treated with a hormone mixture containing 10 nM E, 1 µM P, 0.5 mM 8IbromoIcAMP, and 10 µM UPA, mifepristone or vehicle for 6 days. Immunocytochemical analysis of PR in endometrial stromal in the absence of UPA or mifepristone (left panel), in the presence of UPA (middle panel) and in the presence of mifepristone (right panel) are shown. Representative images are shown. Fig. 10: PR-B stability in response to UPA or mifepristone in human endometrial stromal cells. Primary cultures of human stromal cells were treated with a hormone mixture containing 10 nM E, 1 µM P, 0.5 mM 8IbromoIcAMP, and 10 µM UPA, mifepristone or vehicle for 6 days. PRIB expression in endometrial stromal in the absence of UPA or mifepristone (left panel), in the presence of UPA (middle panel) and in the presence of mifepristone (right panel) are shown. Representative images are shown. Page 16 of 28 16 Reproductive Sciences For Peer Review

1 Wagenfeld A, Saunders PT, Whitaker L, Critchley HO: Selective progesterone receptor m odulators (SPRMs): progesterone receptor action, mode of action on the endometrium and treatment options in gynecological therapies. Expert Opin Ther Targets 2016:1-10. 2 Glasier A: The rationale for use of Ulipristal Acetate as first line in emergency contraception: biological and clinical evidence. Gynecol Endocrinol 2014;30:688-690. 3 Glasier AF, Cameron ST, Fine PM, Logan SJ, Casale W, Van Horn J, Sogor L, Blithe DL, Scherrer B, Mathe H, Jaspart A, Ulmann A, Gainer E: Ulipristal acetate versus levonorgestrel for emergency contraception: a randomised non-inferiority trial and meta-analysis. Lancet 2010;375:555-562. 4 Gainer EE, Ulmann A: Pharmacologic properties of CDB(VA)-2914. Steroids 2003;68:1005-1011. 5 Nichols MI: Ulisprisal acetate: a novel molecule and 5-day emergency contraceptive. Obstet Gynecol 2010;116:1252-1253. 6 Brache V, Cochon L, Jesam C, Maldonado R, Salvatierra AM, Levy DP, Gainer E, Croxatto HB: Immediate pre-ovulatory administration of 30 mg ulipristal acetate significantly delays follicular rupture. Hum Reprod 2010;25:2256-2263. 7 Blithe DL, Nieman LK, Blye RP, Stratton P, Passaro M: Development of the selective progesterone receptor modulator CDB-2914 for clinical indications. Steroids 2003;68:1013-1017. 8 Creinin MD, Schlaff W, Archer DF, Wan L, Frezieres R, Thomas M, Rosenberg M, Higgins J: Progesterone receptor modulator for emergency contraception: a randomized controlled trial. Obstet Gynecol 2006;108:1089-1097. 9 Donnez J, Tomaszewski J, Vazquez F, Bouchard P, Lemieszczuk B, Baro F, Nouri K, Selvaggi L, Sodowski K, Bestel E, Terrill P, Osterloh I, Loumaye E, Group PIS: Ulipristal acetate versus leuprolide acetate for uterine fibroids. N Engl J Med 2012;366:421-432. 10 Fernandez H, Schmidt T, Powell M, Costa AP, Arriagada P, Thaler C: Real world data of 1473 patients treated with ulipristal acetate for uterine fibroids: Premya study results. Eur J Obstet Gynecol Reprod Biol 2017;208:91-96. 11 Huang Y, Jensen JT, Brache V, Cochon L, Williams A, Miranda MJ, Croxatto H, Kumar N, Sussman H, Hoskin E, Plagianos M, Roberts K, Merkatz R, Blithe D, Sitruk-Ware R: A randomized study on pharmacodynamic effects of vaginal rings delivering the progesterone receptor modulator ulipristal acetate: research for a novel estrogen-free, method of contraception. Contraception 2014;90:565-574. 12 Mutter GL, Bergeron C, Deligdisch L, Ferenczy A, Glant M, Merino M, Williams AR, Blithe DL: The spectrum of endometrial pathology induced by progesterone receptor modulators. Mod Pathol 2008;21:591-598. 13 Pawar S, Hantak AM, Bagchi IC, Bagchi MK: Minireview: Steroid-regulated paracrine mechanisms controlling implantation. Mol Endocrinol 2014;28:1408-1422. 14 Hantak AM, Bagchi IC, Bagchi MK: Role of uterine stromal-epithelial crosstalk in embryo implantation. Int J Dev Biol 2014;58:139-146. 15 Chandra V, Kim JJ, Benbrook DM, Dwivedi A, Rai R: Therapeutic options for management of endometrial hyperplasia. J Gynecol Oncol 2016;27:e8. 16 Horn LC, Schnurrbusch U, Bilek K, Hentschel B, Einenkel J: Risk of progression in complex and atypical endometrial hyperplasia: clinicopathologic analysis in cases with and without progestogen treatment. Int J Gynecol Cancer 2004; 14:348-353. 17 Daud S, Jalil SS, Griffin M, Ewies AA: Endometrial hyperplasia - the dilemma of management remains: a retrospective observational study of 280 women. Eur J Obstet Gynecol Reprod Biol 2011;159:172-175. Page 17 of 28 17 Reproductive Sciences For Peer Review 18 Yang S, Thiel KW, Leslie KK: Progesterone: the ultimate endometrial tumor suppressor. Trends Endocrinol Metab 2011;22:145-152. 19 Amant F, Moerman P, Neven P, Timmerman D, Van Limbergen E, Vergote I: Endometrial cancer. Lancet 2005;366:491-505. 20 Li Q, Kannan A, DeMayo FJ, Lydon JP, Cooke PS, Yamagishi H, Srivastava D, Bagchi MK, Bagchi IC: The antiproliferative action of progesterone in uterine epithelium is mediated by Hand2. Science 2011;331:912-916. 21 Jones A, Teschendorff AE, Li Q, Hayward JD, Kannan A, Mould T, West J, Zikan M, Cibula D, Fiegl H, Lee SH, Wik E, Hadwin R, Arora R, Lemech C, Turunen H, Pakarinen P, Jacobs IJ, Salvesen HB, Bagchi MK, Bagchi IC, Widschwendter M: Role of DNA methylation and epigenetic silencing of HAND2 in endometrial cancer development. PLoS Med 2013;10:e1001551. 22 Whitaker LH, Murray AA, Matthews R, Shaw G, Williams AR, Saunders PT, Critchley HO: Selective progesterone receptor modulator (SPRM) ulipristal acetate (UPA) and its effects on the human endometrium. Hum Reprod 2017;32:531-543. 23 Pritts EA, Ryan IP, Mueller MD, Lebovic DI, Shifren JL, Zaloudek CJ, Korn AP, Darney PD, Taylor RN: Angiogenic effects of norplant contraception on endometrial histology and uterine bleeding. J Clin Endocrinol Metab 2005;90:2142-2147. 24 Kaya HS, Hantak AM, Stubbs LJ, Taylor RN, Bagchi IC, Bagchi MK: Roles of progesterone receptor A and B isoforms during human endometrial decidualization. Mol Endocrinol 2015;29:882-895. 25 Li Q, Kannan A, Das A, Demayo FJ, Hornsby PJ, Young SL, Taylor RN, Bagchi MK, Bagchi IC: WNT4 acts downstream of BMP2 and functions via beta-catenin signaling pathway to regulate human endometrial stromal cell differentiation. Endocrinology 2013;154:446-457. 26 Flannery CA, Fleming AG, Choe GH, Naqvi H, Zhang M, Sharma A, Taylor HS: Endometrial Cancer- Associated FGF18 Expression Is Reduced by Bazedoxifene in Human Endometrial Stromal Cells In Vitro and in Murine Endometrium. Endocrinology 2016;157:3699-3708. 27 Kim JJ, Chapman-Davis E: Role of progesterone in endometrial cancer. Semin Reprod Med 2010;28:81-90. 28 Glasier AF, Wang H, Davie JE, Kelly RW, Critchley HO: Administration of an antiprogesterone up- regulates estrogen receptors in the endometrium of women using Norplant: a pilot study. Fertil Steril 2002;77:366-372. 29 Boggavarapu NR, Berger C, von Grothusen C, Menezes J, Gemzell-Danielsson K, Lalitkumar PG: Effects of low doses of mifepristone on human embryo implantation process in a three- dimensional human endometrial in vitro co-culture system. Contraception 2016;94:143-151. 30 Greb RR, Kiesel L, Selbmann AK, Wehrmann M, Hodgen GD, Goodman AL, Wallwiener D: Disparate actions of mifepristone (RU 486) on glands and stroma in the primate endometrium. Hum Reprod 1999;14:198-206. Page 18 of 28 18 Reproductive Sciences Fig.1 254x190mm (72 x 72 DPI) Page 19 of 28 Reproductive Sciences Fig.2 254x190mm (72 x 72 DPI) Page 20 of 28Reproductive Sciences Fig.3 254x190mm (72 x 72 DPI) Page 21 of 28 Reproductive Sciences Fig.4 254x190mm (72 x 72 DPI) Page 22 of 28Reproductive Sciences Fig.5 254x190mm (72 x 72 DPI) Page 23 of 28 Reproductive Sciences Fig.6 254x190mm (72 x 72 DPI) Page 24 of 28Reproductive Sciences Fig.7 254x190mm (72 x 72 DPI) Page 25 of 28 Reproductive Sciences Fig.8 254x190mm (72 x 72 DPI) Page 26 of 28Reproductive Sciences Fig.9 254x190mm (72 x 72 DPI) Page 27 of 28 Reproductive Sciences Fig.10 254x190mm (72 x 72 DPI) Page 28 of 28Reproductive Sciences