{"paper_id":"2640d3dc-ef6e-484b-a735-d1b9361bc8fc","body_text":"Int. J. Med. Sci. 2023, Vol. 20 \n \n \nhttps://www.medsci.org \n415 \nInternational Journal of Medical Sciences\n \n2023; 20(3): 415-428. doi: 10.7150/ijms.79516 \nResearch Paper \nBioinformatics-based analysis of the roles of sex \nhormone receptors in endometriosis development \nXiaoling Zhao1, Weimin Kong1, Chunxiao Zhou2, Boer Deng1,2, He Zhang1, Huimin Guo1, Shuning Chen1, \nZhendong Pan1 \n1. Department of Gynecologic Oncology, Beijing Obstetrics and Gynecology Hospital, Capital Medical University, Beijing Maternal and Child Health Care \nHospital, Beijing, China. \n2. Division of Gynecologic Oncology and Lineberger Comprehensive Cancer Center, University of North Carolina at Chapel Hill, Chapel Hill, NC, USA.  \n Corresponding author: Weimin Kong. E-mail address: kwm1967@ccmu.edu.cn \n© The author(s). This is an open access article distributed under the terms of the Creative Commons Attribution License (https://creativecommons.org/licenses/by/4.0/). \nSee http://ivyspring.com/terms for full terms and conditions. \nReceived: 2022.10.04; Accepted: 2023.01.20; Published: 2023.02.05 \nAbstract \nEndometriosis is a hormone-dependent disease in women of reproductive age and seriously affects \nwomen's health. To analyze the involvement of sex hormone receptors in endometriosis development, \nwe performed bioinformatics analysis using four datasets derived from the Gene Expression Omnibus \n(GEO) database, which may help us understand the mechanisms by which the sex hormones act in vivo in \nendometriosis patients. The enrichment analysis and protein–protein interaction (PPI) analysis of the \ndifferentially expressed genes (DEGs) revealed that there are different key genes and pathways involved \nin eutopic endometrium aberrations of endometriosis patients and endometriotic lesions, and sex \nhormone receptors, including androgen receptor (AR), progesterone receptor (PGR) and estrogen \nreceptor 1 (ESR1), may play important roles in endometriosis development. Androgen receptor (AR), as \nthe hub gene of endometrial aberrations in endometriotic patients, showed positive expression in the \nmain cell types for endometriosis development, and its decreased expression in the endometrium of \nendometriotic patients was also confirmed by immunohistochemistry (IHC). The nomogram model \nestablished based on it displayed good predictive value. \nKey words: Endometriosis; Hormone receptor; Bioinformatics analysis \nIntroduction \nEndometriosis is caused by the presence of \nendometrium-like tissues outside of the uterus [1]. It \nis estimated that endometriosis can affect 10-15% of \nreproductive-age women, resulting in pelvic pain and \ninfertility [1, 2]. The long-term presence of \nendometriosis also carries the risk of cancers [3, 4]. \nSeveral hypotheses, such as retrograde menstruation \ntheory, have been proposed to explain the etiology \nand pathogenesis of endometriosis, however, none of \nthem can fully explain it.  \nIt is now basically clear that endometriosis is a \nhormone-dependent inflammatory disease [5]. In the \nendometrium of patients with endometriosis, \nestrogen, which can promote endometrial cell \nproliferation and inflammation, was dominant while \nprogesterone was resistant and failed to properly \nantagonize the effects of estrogen [5, 6]. The \nprogesterone resistance and estrogen dominance in \nectopic lesions lead to increased lesion growth and \ncontribute to pelvic pain and infertility [5, 6]. \nAndrogen, which can reduce the chronic pain and \ninflammation, can be converted to estrogen by \naromatase in the eutopic and ectopic endometrium of \nwomen with endometriosis, thereby increasing local \nestrogen levels [7]. So far, the exact mechanisms by \nwhich these hormones act in vivo remain unclear, \nmaking prevention and treatment challenging.  \n Currently, with the rapid development of \nsequencing technology and the emergence of \nbioinformatics analysis and public databases, we can \nobtain a massive amount of gene information to \nexplore the underlying molecular mechanisms of the \ndisease [8]. The sex hormones mediate the biological \neffects on endometrium by binding to their receptors, \n \nIvyspring  \nInternational Publisher \n\nInt. J. Med. Sci. 2023, Vol. 20 \n \nhttps://www.medsci.org \n416 \noccurring at cell surface and in the nucleus [5, 9]. \nUnderstanding the roles of these receptors in the \npathogenesis of endometriosis may help us uncover \nthe mechanisms of these sex hormones’ actions. \nTherefore, in this study, we aimed to investigate the \nroles of sex hormone receptors in endometriosis \ndevelopment by bioinformatics analysis, which may \nprovide us with new insights into the disease. The \nworkflow of this study was presented in Figure 1. \nMethods \nData collection and DEGs identification \nThe gene expression profiles associated with \nendometriosis were obtained from the GEO database, \nwhich is searched using ‘endometriosis’ and \n‘endometrioma’ as keywords and restricts the source \nof tissues to ‘homo sapiens’. The information of the \nselected datasets is displayed in Table 1. Then the \ndatasets (GSE51981, GSE120103, GSE37837 and \nGSE7305) with complete clinical information and \nsufficient case numbers were selected for further \nanalysis. GSE51981, based on the GPL570 platform, \nincludes 34 normal endometria and 77 endometria \nfrom endometriosis patients [10]. GSE120103, based \non GPL6480, includes 18 normal endometria and 18 \nendometria from endometriosis patients [11]. The \ndatasets GSE37837 and GSE7305 were produced using \nthe GPL6480 and GPL570 platforms, which contained \n36 samples (18 ectopic lesions and 18 matched control \nendometria from the same patients) and 20 samples \n(10 ectopic lesions and 10 matched control endometria \nfrom the same patients), respectively [12, 13]. \n \nTable 1. The datasets associated with endometriosis. \nDatasets Platform Sample size Sample types \nGSE141549 GPL10558 & \nGPL1336 \nn=408 Ectopicvs normal endometrium \n& peritoneum \nGSE51981 GPL570 33 vs 77 Eutopic vs normal endometrium \nGS135485 GPL21290 52 vs 12 Ectopic vs normal endometrium \nGSE120103 GPL6480 18 vs 18 Eutopic vs normal endometrium \nGSE37837 GPL6480 18 vs 18 Ectopic vs eutopic endometrium \nGSE25628 GPL571 n=22 Ectopic vs eutopic vs normal \nendometrium \nGSE5108 GPL2895 11 vs 11 Ectopic vs eutopic endometrium \nGSE11691 GPL96 9 vs 9 Ectopic vs eutopic endometrium \nGSE99949 GSE17301 4 vs 4 Ectopic vs eutopic endometrium \nGSE153740 GPL18573 4 vs 4 Eutopic vs normal endometrium \nGSE7305 GPL570 10 vs 10 Ectopic vs eutopic endometrium \nGSE58178 GPL6947 6 vs 6 Stromal cells derived from \neutopic vs normal endometrium \nGSE12768 GPL7304 2 vs 2 Ectopic vs eutopic endometrium \n \n \n \n \nFigure 1. Flowchart of the integrated analysis for endometriosis. AR, androgen receptor; ESR1, estrogen receptor 1; ESR2, estrogen receptor 2; GPER1, G- protein coupled \nestrogen receptor 1; PGR, prog esterone receptor; PGRMC1, progesterone receptor membrane component 1; PGRMC2, progesterone receptor membrane component  2; \nGEO, Gene Expression Omnibus; DEGs, the differentially expressed genes; PPI, protein–protein interaction. \n \n\n\nInt. J. Med. Sci. 2023, Vol. 20 \n \n \nhttps://www.medsci.org \n417 \nThe DEGs were identified using GEO2R, an \ninteractive tool that can compare two or more groups \nof samples with the limma package [14]. After \nnormalization and log2 transformation, the genes \nwith a fold change >1 and a P value <0.05 were \nselected. \nEnrichment analysis \nEnrichment analyses were performed using the \nDatabase for Annotation, Visualization and \nIntegrated Discovery (DAVID) (https://david.ncifcrf \n.gov) [15]. Biological processes (BP) were selected for \nfurther analysis to identify the biological attributes of \nthe DEGs. A P value <0.05 was set as the cutoff for \nstatistical significance. \nPPI network analysis \nTo explore the interactions among the identified \nDEGs, we mapped them to the Search Tool for the \nRetrieval of Interacting Genes/Proteins (STRING) \ndatabase (https://cn.string-db.org) to assess the \nprotein–protein interaction information [16]. To ensure \nreliable interactions, only experiments were selected \nas the active interaction sources, with the minimum \nrequired interaction scores at 0.150. The other \nindicated network properties consist of organism \n(Homo sapiens); network type (full string network); \nand meaning of network edges (evidence). Then, \nCytoscape software was used to visualize the PPI \nnetwork [16]. CytoHubba, a plug-in of Cytoscape \nsoftware, was used to rank nodes and screen the hub \ngenes [17]. The top 10 genes calculated by topological \nalgorithms were considered hub genes.  \nThe analysis of androgen receptor (AR) using \nthe Human Protein Atlas (HPA) database \nHPA (https://www.proteinatlas.org/) is a \ncomprehensive database covering the protein \nexpression of many cancerous and normal tissues, \nwith millions of images for human tissue samples \nincluded [18, 19]. The expression level of AR in \nhuman normal tissues was evaluated in the HPA \ndatabase and presented as a histogram. The single-cell \nprofiles of the endometrium are also pictured in the \nHPA database, and the expression level of AR in each \ncell type is presented.  \nAR expression and location detection \nEndometrial tissues and ectopic lesions were \ncollected from patients with indications for \nhysterectomy in Beijing Obstetrics and Gynecology \nHospital from January 2018 to December 2021. \nNormal endometria, including 3 postmenopausal \nendometria and 6 premenopausal endometria were \ncollected from patients with grade III cervical \nintraepithelial neoplasia or stage IA1 cervical cancer. \nEctopic lesions and matched eutopic endometrial \nsamples were collected from 6 patients with \nendometriosis combined with grade III cervical \nintraepithelial neoplasia or cervical cancer stage IA1. \nAll the patients had normal menstrual cycles and \ndidn’t receive any hormone therapy. This study was \napproved by the Ethics Committee of Beijing \nObstetrics and Gynecology Hospital affiliated with \nCapital Medical University, and written informed \nconsent was obtained from all patients.  \nImmunohistochemistry (IHC) was used to detect \nAR expression using an AR antibody (#DF6783, \nAffinity, Japan) at a dilution of 1:200. The H-score \n(H-score = [1*(% of cells 1+) + 2*(% of cells 2+) + 3*(% \nof cells 3+)], where 1 = weak expression, 2 = moderate \nexpression, and 3 = strong expression) was applied to \nquantify the IHC images [20]. Immunofluorescent (IF) \nstaining was performed to determine the location of \nAR in the cells using an AR antibody (#DF6783, \nAffinity, Japan) at a dilution of 1:100.  \nIdentification of proteins that interact with \nhormone receptors \nThe interactors of hormone receptors were \nidentified from two databases: GPS-Prot \n(http://www.gpsprot.org) and Biogrid \n(https://thebiogrid.org). GPS-Prot is a web-based \nvisualization platform for PPIs that allows new \nuser-generated data to be uploaded [21]. Biogrid is a \nbiomedical interaction repository with data compiled \nthrough comprehensive curation efforts [22]. The \ndifferentially expressed interactors of AR between \neutopic endometria and normal endometria with a \nfold change >1 and a P value <0.05 were selected for \nthe establishment of the models. \nDiagnostic model establishment \nA nomogram was established using the rms \npackage in R software with GSE51981 as the training \nset [23]. Genes included in the diagnostic model \nanalysis were selected by least absolute shrinkage and \nselection operator (LASSO) regression using the \nglmnet package [24]. Then, GSE120103, containing 36 \nsamples, was used as the test set to verify the model. \nThe receiver operating characteristic (ROC) curve \ncalculated by the pROC package was used to test the \nefficacy of the diagnostic model [23]. \nResults \nIdentification of differentially expressed genes \n(DEGs) and sex hormone receptors expression  \nIn the present study, a total of 6073 and 6633 \nDEGs between normal endometria and those of \npatients with endometriosis were identified in \nGSE120103 and GSE51981, respectively. 1787 DEGs \n\nInt. J. Med. Sci. 2023, Vol. 20 \n \nhttps://www.medsci.org \n418 \nwere obtained from their intersection. Among them, \nthe numbers of up-regulated genes and \ndown-regulated genes in both two datasets were 359 \nand 459, respectively. A total of 2455 and 1835 DEGs \nbetween eutopic and ectopic endometrial tissues of \nthe same patients from GSE37837 and GSE7305, \nrespectively, were also selected. A total of 384 DEGs \nwere found after intersection, comprising 134 \nup-regulated genes and 220 down-regulated genes in \nboth two datasets. The genes were listed in \nSupplementary Table S1. \nThen the expression of sex hormone receptors in \neutopic and ectopic endometria was analyzed in \nGSE51981, GSE120103, GSE7305 and GSE37837 \n(Figure 2A-G). In the eutopic endometria of \nendometriosis patients, AR, progesterone receptor \n(PGR) and progesterone receptor membrane \ncomponent 1 (PGRMC1) showed decreased \nexpression compared with normal endometria, with \nlog2FC<1 in both the GSE51981 and GSE120103 \ndatasets. In the ectopic endometria of endometriotic \npatients, only ESR1, one of the estrogen receptors, \nshowed markedly decreased expression in both the \nGSE7305 and GSE37837 datasets, with log2FC<1. \nEnrichment analysis of the DEGs \nTo identify the biological functions of the DEGs \nin the development of endometriosis, enrichment \nanalyses were conducted using DAVID, and the top \n20 enriched biological processes were represented in \nTable 2-3. The DEGs between eutopic endometrium \nand normal endometrium mainly enriched in \n‘positive regulation of macromolecule metabolic \nprocess’, ‘positive regulation of metabolic process’ \nand ‘positive regulation of macromolecule \nbiosynthetic process’. And the up-regulated genes in \neutopic endometrium were mainly enriched in ‘cell \nactivation’, ‘anatomical structure development’ and \n‘leukocyte activation’ while the down-regulated genes \nwere mainly enriched in ‘chromosome organization’, \n‘cellular component organization or biogenesis’ and \n‘organelle organization’. For the DEGs between \nectopic endometrium and eutopic endometrium, the \ngenes were primarily enriched in ‘anatomical \nstructure morphogenesis’, ‘anatomical structure \ndevelopment’ and ‘single-multicellular organism \nprocess’. Moreover, the up-regulated genes in ectopic \nendometrium were significantly enriched in \n‘regulation of response to stimulus’, ‘regulation of \nmulticellular organismal process’ and ‘positive \nregulation of response to stimulus’, while the \ndown-regulated genes were significantly enriched in \n‘regulation of cell cycle’, ‘cell cycle’ and ‘cell cycle \nprocess’. \n \n \n \nFigure 2. The expression of sex hormone receptors in the datasets. (A-G) The expression of sex hormone receptors, including AR (A), ESR1 (B), ESR2 (C), GPER1 (D), PGR \n(E), PGRMC1 (F) and PGRMC2 (G), in the datasets. AR, androgen receptor; ESR1, estrogen receptor 1; ESR2, estrogen receptor 2; GPER1, G-protein coupled estrogen receptor \n1; PGR, progesterone receptor; PGRMC1, progesterone receptor membrane component 1; PGRMC2, progesterone receptor membrane component 2. FC of GSE51981 and \nGSE120103= the value of endometria from endometriosis patients/ the value of normal endometria; FC of GSE37837 and GSE7305= the value of ectopic endometria from the \nsame patients/the value of eutopic endometria from endometriosis patients. FC, fold change. \n\n\nInt. J. Med. Sci. 2023, Vol. 20 \n \n \nhttps://www.medsci.org \n419 \nTable 2. The enriched biological processes of the DEGs between \neutopic and normal endometrium.  \n All DEGs Up-regulated DEGs Down-regulated \nDEGs \n \n \n \n \n \n \nBiological \nprocess \nPositive regulation of \nmacromolecule \nmetabolic process; \nPositive regulation of \nmetabolic process; \nPositive regulation of \nmacromolecule \nbiosynthetic process; \nDevelopmental \nprocess; \nCellular component \norganization; \nCellular component \norganization or \nbiogenesis; Response \nto organic substance; \nPositive regulation of \ncellular biosynthetic \nprocess; Positive \nregulation of \nbiosynthetic process; \nPositive regulation of \ncellular metabolic \nprocess; Response to \noxygen-containing \ncompound; \nSingle-organism \ncellular process; \nPositive regulation of \nnitrogen compound \nmetabolic process; \nResponse to \nendogenous stimulus; \nPositive regulation of \ngene expression; \nCellular response to \norganic substance; \nCellular response to \nchemical stimulus; \nRegulation of cellular \ncomponent \norganization; \nSingle-organism \ndevelopmental \nprocess; Anatomical \nstructure \ndevelopment. \nCell activation; \nAnatomical structure \ndevelopment; \nLeukocyte activation; \nRegulation of \ndevelopmental \nprocess; \nDevelopmental \nprocess; \nResponse to \noxygen-containing \ncompound; \nSingle-organism \ndevelopmental \nprocess; Response to \ncytokine; \nAnatomical structure \nmorphogenesis; \nPositive regulation of \ndevelopmental \nprocess; Cell surface \nreceptor signaling \npathway; Positive \nregulation of \nmetabolic process; \nImmune system \nprocess; \nCellular response to \norganic substance; \nLymphocyte \nactivation; \nCellular response to \nchemical stimulus; \nLeukocyte \ndifferentiation; \nPositive regulation of \nmacromolecule \nmetabolic process; \nMulticellular \norganism \ndevelopment; \nSingle-organism \nprocess. \nChromosome \norganization; \nCellular component \norganization or \nbiogenesis; Organelle \norganization; \nCellular component \norganization; \nEstablishment of \nprotein localization; \nIntracellular protein \ntransport; Cellular \nresponse to stress; \nCellular protein \nlocalization; \nMacromolecule \nlocalization; \nRegulation of mRNA \nmetabolic process; \nCellular \nmacromolecule \nlocalization; RNA \nlocalization; \nProtein localization; \nNitrogen compound \ntransport; \nRNA splicing, via \ntransesterification \nreactions with bulged; \nAdenosine as \nnucleophile; \nmRNA splicing, via \nspliceosome; \nRNA splicing, via \ntransesterification \nreactions; Protein \ntransport; \nRNA splicing. \n \nTable 3. The enriched biological processes of the DEGs between \nectopic and eutopic endometrium.  \n All DEGs Upregulated DEGs Downregulated DEGs \n \n \n \n \n \nBiological \nprocess \nAnatomical \nstructure \nmorphogenesis; \nAnatomical \nstructure \ndevelopment; \nSingle-multicellular \norganism process; \nMulticellular \norganism \ndevelopment; \nSystem \ndevelopment; \nDevelopmental \nprocess; \nSingle-organism \ndevelopmental \nprocess; Animal \norgan development; \nTissue development; \nMulticellular \norganismal process; \nCell differentiation; \nSingle-organism \nRegulation of response \nto stimulus; \nRegulation of \nmulticellular \norganismal process; \nPositive regulation of \nresponse to stimulus; \nAnatomical structure \nmorphogenesis; \nSystem development; \nRegulation of \ndevelopmental process; \nResponse to stress; \nSingle-multicellular \norganism process; \nResponse to stimulus; \nMulticellular organism \ndevelopment; \nMulticellular \norganismal process; \nDevelopmental process;\n \nAnatomical structure \ndevelopment; Response \nto wounding; \nRegulation of cell cycle; \nCell cycle; \nCell cycle process; \nRegulation of cell cycle \nprocess; Cell division; \nCell development; \nSingle-organism \ncellular process; \nSingle-organism \nprocess; Anatomical \nstructure development; \nTissue development; \nNuclear division; \nRegulation of nuclear \ndivision; Animal organ \ndevelopment; Mitotic \ncell cycle; \nSingle-organism \ndevelopmental process; \nMicrotubule-based \nprocess; Developmental \nprocess; \nOrganelle fission; \nAnatomical structure \n All DEGs Upregulated DEGs Downregulated DEGs \ncellular process; \nCellular \ndevelopmental \nprocess; Cell \ndevelopment; \nCell proliferation; \nSingle-organism \nprocess; Regulation \nof cell cycle; \nRegulation of \nmulticellular \norganismal process; \nResponse to \nstimulus; \nRegulation of \ndevelopmental \nprocess. \nRegulation of immune \nresponse; \nPositive regulation of \ndevelopmental process; \nSingle-organism \ndevelopmental process; \nRegulation of \nmulticellular \norganismal \ndevelopment; Wound \nhealing; \nSignal transduction. \nmorphogenesis; \nMulticellular organism \ndevelopment. \n \n \nConsidering that endometriosis is a \nhormone-dependent gynecological disease, we \nscreened the hormone-related biological processes. \nFor eutopic endometrium derived from endometriotic \npatients and normal endometrium, the processes, \nincluding ‘response to hormone’, ‘cellular response to \nhormone stimulus’, ‘response to steroid hormone’, \n‘cellular response to peptide hormone stimulus’, \n‘response to peptide hormone’ and ‘cellular response \nto steroid hormone stimulus’, were screened. Both AR \nand PGR were involved in the biological processes \nnamed ‘response to hormone’, ‘cellular response to \nhormone stimulus’, ‘response to steroid hormone’ and \n‘cellular response to steroid hormone stimulus’ (Table \n4). For ectopic endometrium and eutopic \nendometrium of endometriotic patients, the biological \nprocesses, including ‘cellular response to luteinizing \nhormone stimulus’, ‘response to luteinizing \nhormone’, ‘hormone catabolic process’, ‘response to \npeptide hormone’, ‘response to hormone’, ‘response \nto growth hormone’, ‘cellular response to peptide \nhormone stimulus’ and ‘hormone metabolic process’, \nwere screened. ESR1 involved in ‘response to \nhormone’ and ‘hormone metabolic process’ (Table 5). \nIn addition, we screened the biological processes \nin which sex hormone receptors were involved. The \ntop 20 enriched biological processes were listed in \nTable 6. AR and PGR were involved in most of the \nprocesses in which the DEGs between eutopic \nendometrium and normal endometrium were \nsignificantly enriched. ‘Positive regulation of \nmacromolecule metabolic process’, ‘positive \nregulation of metabolic process’ and ‘positive \nregulation of macromolecule biosynthetic process’ \nwere the top 3 enriched processes in which both AR \nand PGR were involved. Only 15 significantly \nenriched biological processes in which PGRMC1 was \ninvolved. ‘Single-organism cellular process’, \n‘single-organism process’ and ‘cellular biosynthetic \nprocess’ ranked top 3. ESR1 was also involved in \nalmost all the enriched processes of the DEGs between \n\nInt. J. Med. Sci. 2023, Vol. 20 \n \nhttps://www.medsci.org \n420 \nectopic endometrium and eutopic endometrium, and \n‘anatomical structure morphogenesis’, ‘anatomical \nstructure development’ and ‘single-multicellular \norganism process’ ranked the top 3. \n \nTable 4. The hormone-related biological processes of the DEGs between eutopic and normal endometrium \nTerm P Value Genes \nResponse to hormone 1.2421475110305147E-8 CDKN1A, CALCOCO1, AHCYL1, FAM107A, NUCKS1, IRS2, CTSV, SOGA1, AQP1, YY1, RBM4, ZFP36, EDNRA, \nAIFM1, HADH, JAK3, APPL1, SMARCC1, PRKCI, IGFBP5, NCOA4, SORD, CACYBP, FOS, KLF15, SFRP4, AR, SFRP1, \nCEACAM1, NCOR1, RAB31, NCL, PGR, PLCB1, RHOQ, SLC29A2, RAMP2, RAMP3, PCNA, WBP2, SRC, SRF, GATA6, \nPTN, PIK3R1, TYMS, HSPD1, LRP6, SOCS3, SLIT3, FYN, RBBP7, ZBTB7B, EIF4E, SH2B2, ABCA2, EGR1, HSPA8, \nCCL21, IGF2, CDC6, ATP2B1, SMARCA4, NR4A1, REST, SST, GNB1, FOSB, PAM, SLC26A6. \nCellular response to \nhormone stimulus \n1.0762613753458778E-6 CALCOCO1, AHCYL1, FAM107A, NUCKS1, IRS2, SOGA1, AQP1, RBM4, ZFP36, EDNRA, AIFM1, JAK3, APPL1, \nSMARCC1, PRKCI, NCOA4, FOS, AR, SFRP1, CEACAM1, NCOR1, RAB31, NCL, PGR, PLCB1, RHOQ, SLC29A2, \nRAMP2, RAMP3, WBP2, SRC, GATA6, PIK3R1, SOCS3, SLIT3, FYN, ZBTB7B, EIF4E, SH2B2, HSPA8, IGF2, CDC6, \nATP2B1, SMARCA4, NR4A1, REST, SST, GNB1, FOSB, SLC26A6. \nResponse to steroid \nhormone \n8.59418442951083E-5 CDKN1A, RAMP2, CALCOCO1, PCNA, WBP2, SRC, FAM107A, PTN, CTSV, TYMS, HSPD1, AQP1, ZFP36, AIFM1, \nSLIT3, RBBP7, EIF4E, ABCA2, HSPA8, NCOA4, ATP2B1, FOS, SMARCA4, AR, SFRP1, REST, NCOR1, SST, FOSB, PGR. \nCellular response to \npeptide hormone stimulus \n3.59243815958365E-4 AHCYL1, SRC, NUCKS1, IRS2, PIK3R1, SOGA1, RBM4, SOCS3, FYN, ZBTB7B, JAK3, SH2B2, APPL1, SMARCC1, \nPRKCI, IGF2, CDC6, FOS, NR4A1, CEACAM1, RAB31, NCL, PLCB1, RHOQ, SLC26A6, SLC29A2. \nResponse to peptide \nhormone \n0.0010350945953294228 AHCYL1, SRC, NUCKS1, IRS2, PIK3R1, SOGA1, LRP6, RBM4, SOCS3, FYN, HADH, ZBTB7B, JAK3, SH2B2, APPL1, \nEGR1, SMARCC1, PRKCI, IGFBP5, IGF2, CACYBP, CDC6, FOS, KLF15, NR4A1, CEACAM1, RAB31, NCL, PLCB1, \nRHOQ, SLC26A6, SLC29A2. \nCellular response to \nsteroid hormone stimulus \n0.011514297795136757 HSPA8, CALCOCO1, WBP2, SRC, NCOA4, FAM107A, ATP2B1, SMARCA4, AQP1, AR, ZFP36, SFRP1, REST, NCOR1, \nAIFM1, PGR. \n \nTable 5. The hormone-related biological processes of the DEGs between ectopic and eutopic endometrium.  \nTerm P Value Genes \nCellular response to luteinizing \nhormone stimulus \n0.005148208184733331 CCNA2, EDNRA. \nResponse to luteinizing hormone 0.006791515814867501 CCNA2, EDNRA, STAR. \nHormone catabolic process 0.015342719212104462 MME, DIO2, HSD17B11. \nResponse to peptide hormone 0.016996690262721416 XBP1, LEPROT, IGFBP5, CAV1, RARRES2, GCNT1, CCNA2, SCNN1G, BRIP1, \nCXCL12, STAR, SCNN1A, PDK4, TIMP1. JAK3 \nResponse to hormone 0.021250037906878627 XBP1, LEPROT, IGFBP5, CAV1, RARRES2, GCNT1, FHL2, TRH, FBXO32, ESR1, \nRXFP1, TGFBR2, FOXP1, CCNA2, SCNN1G, EDNRA, BRIP1, CXCL12, STAR, \nSCNN1A, TIMP2, PDK4, TIMP1, JAK3. \nResponse to growth hormone 0.02992832174632198 LEPROT, IGFBP5, STAR, JAK3. \nCellular response to peptide \nhormone stimulus \n0.03338644004761159 CCNA2, SCNN1G, LEPROT, XBP1, BRIP1, STAR, CAV1, RARRES2, SCNN1A, \nPDK4, JAK3. \nHormone metabolic process 0.03874287787577624 SCARB1, STAR, MME, ALDH1A2, DIO2, UGT2B28, HSD17B11, ESR1, PAPSS2. \n \n \nTable 6. The enriched biological processes that sex hormone receptors involved in. \n AR PGR PGRMC1 ESR1 \n \n \n \n \n \n \nBiological \nprocess \nPositive regulation of macromolecule \nmetabolic process; Positive regulation \nof metabolic process; \nPositive regulation of macromolecule \nbiosynthetic process; \nDevelopmental process; \nCellular component organization; \nCellular component organization or \nbiogenesis; Response to organic \nsubstance; \nPositive regulation of cellular \nbiosynthetic process; Positive \nregulation of biosynthetic process; \nPositive regulation of cellular \nmetabolic process; Response to \noxygen-containing compound; \nSingle-organism cellular process; \nPositive regulation of nitrogen \ncompound metabolic process; \nResponse to endogenous stimulus; \nPositive regulation of gene expression; \nCellular response to organic substance; \nCellular response to chemical stimulus;\n \nRegulation of cellular component \norganization; Single-organism \ndevelopmental process; Anatomical \nstructure development. \nPositive regulation of macromolecule \nmetabolic process; Positive regulation \nof metabolic process; \nPositive regulation of macromolecule \nbiosynthetic process; \nDevelopmental process; Response to \norganic substance; \nPositive regulation of cellular \nbiosynthetic process; Positive \nregulation of biosynthetic process; \nPositive regulation of cellular \nmetabolic process; Single-organism \ncellular process; \nPositive regulation of nitrogen \ncompound metabolic process; \nResponse to endogenous stimulus; \nPositive regulation of gene expression; \nCellular response to organic substance; \nCellular response to chemical stimulus; \nSingle-organism developmental \nprocess; Anatomical structure \ndevelopment; Positive regulation of \nbiological process; Positive regulation \nof \ncellular process; Cellular response to \nendogenous stimulus; \nPositive regulation of \nnucleobase-containing compound \nmetabolic process. \nSingle-organism cellular process; \nSingle-organism process; Cellular \nbiosynthetic process; \nOrganic substance biosynthetic \nprocess; Biosynthetic process; \nCellular metabolic process; \nHeterocycle biosynthetic process \nAromatic compound biosynthetic \nprocess; Cellular process; \nCellular nitrogen compound \nbiosynthetic process; Organic cyclic \ncompound biosynthetic process; \nOrganic substance metabolic \nprocess; \nMetabolic process; \nSingle-organism metabolic process; \nCellular aromatic compound \nmetabolic process. \nAnatomical structure morphogenesis; \nAnatomical structure development; \nSingle-multicellular organism process; \nMulticellular organism development; \nSystem development; \nDevelopmental process; \nSingle-organism developmental process; \nAnimal organ development; \nTissue development; Multicellular \norganismal process; Cell differentiation; \nSingle-organism cellular process; Cellular \ndevelopmental process; Cell \ndevelopment; \nCell proliferation; \nSingle-organism process; \nRegulation of multicellular organismal \nprocess; Response to stimulus; \nRegulation of developmental process; \nRegulation of multicellular organismal \ndevelopment. \n \n\nInt. J. Med. Sci. 2023, Vol. 20 \n \n \nhttps://www.medsci.org \n421 \nPPI network analysis of the DEGs \nTo investigate the associations among the \nscreened DEGs, PPI networks were constructed using \nCytoscape software. And then the top 10 genes were \nselected from each method using CytoHubba. The \nnetwork of the DEGs between normal endometrium \nand those of patients with endometriosis contained \n683 nodes and 5105 edges, with AR, PGR and \nPGRMC1 involved in the network construction \n(Figure 3A). AR interacts with 9 up-regulated genes \n(KDM6B, SRC, FGR, NR4A1, FOS, CEBPB, TNK2, \nZBTB16 and KAT5) and 25 down-regulated genes \n(NCOA4, SPOP, CTNNB1, SMARCA4, HSP90AB1, \nNCOR1, HSPA5, TCF4, HSP90AA1, KPNA3, \nTXNDC5, SMARCC1, DEPDC1, NUP107, APPL1, \nPIK3R1, ATRX, MAPK1, KDM4A, RAH, KDM5B, \nBECN1, KPNB1 and MYLIP) directly. PGR interacts \nwith 3 up-regulated genes (KDM6B, SRC and NR4A1) \nand 6 down-regulated genes (MAPK1, NUP107, \nHSP90AA1, NCOR1, HSP90AB1 and SPOP) directly. \nPGRMC1 interacts with 1 up-regulated gene \n(HSD11B1L) and 5 down-regulated genes (PTPLAD1, \nMPRIP, HNRNPH, CANX and SRSI3) directly. \nBesides, the hub genes of the DEGs between eutopic \nendometrium of endometriosis patients and normal \nendometrium were calculated and displayed in \nSupplementary Table S2. Both AR and PGR were \nidentified as the hub genes. \nThe network of the DEGs between ectopic \nendometrium and eutopic endometrium contained \n213 nodes and 571 edges (Figure 3B). ESR1 interacts \nwith 5 up-regulated genes (FHK2, ST13, CAV1, JUNB \nand EPAS1) and 8 down-regulated genes (SOX9, \nWHSC1, AURKA, XBP1, SMC2, MAP3K1, RAD51 \nand ZMYNDB) directly. What’s more, ESR1 was also \nidentified as one of the hub genes (Supplementary \nTable S2). \nThe potential roles of AR in endometriosis \ndevelopment \nThe roles of estrogen and progesterone in \nendometriosis development and the relevance of their \nreceptors to endometriosis have been discussed and \nvalidated in many studies [5, 6]. Considering the lack \nof research on AR in endometriosis, we next explored \nthe expression of AR in human tissues and cells to \nexplore the possibility of AR involvement in the \nformation of endometriosis. \n The expression of AR in normal human tissues \nwas detected at both RNA and protein levels (Figure \n4A-B). The results showed that the RNA expression of \nAR can be detected in all tissues except bone marrow, \nwhile its protein expression was found only in kidney, \ntestis, epididymis, seminal vesicle, fallopian tube, \nendometrium, cervix and breast. What’s more, AR is \nexpressed in all cell types of endometria, with the \ngreatest expression in endometrial stromal cells \n(Figure 4C-D). The expression of AR in endometrium \ntissues was also detected in this study using the IHC \nmethod (Figure 5A-E). We found that AR expression \nis retained in the post-menopausal endometrium and \npositive AR staining can be detected in both \nendometrial epithelial cells and stromal cells using the \nIHC method. The expression level of AR in normal \npremenopausal endometria was significantly higher \nthan that in endometria derived from endometriotic \npatients (P<0.05). No significant difference in AR \nexpression was found between eutopic endometria \nand their matched ectopic lesions.  \nEndometrial stromal cells are the main cell type \nexpressing AR in endometrial tissues, and they are \nalso an important cell type for endometriosis \ndevelopment [25]. So, we examined the location of AR \nin endometrial stromal cells using the IF method. The \nresults showed that AR localizes to the cytosol and \nnucleus (Figure 5F). \nConsidering the role of AR in the endometrial \naberrations of endometriosis patients, we identified 94 \ninteractors of AR using GPS-Prot and Biogrid. Then, \nAR and its interactors, including 95 proteins, were \nfurther screened by LASSO regression analysis. Seven \ngenes, namely, APPL1, CSNK2A1, ERG, KDM4A, \nSMARCC1, SUZ12 and TRIM25, were selected to \nestablish the diagnostic model for endometriosis, and \na nomogram was constructed based on them (Figure \n6A-C). To test the diagnostic efficacy of the \nnomogram, the ROC curves of the training set \nGSE51981 and test set GSE120103 were plotted, \nyielding an AUC of 0.984 for the training set and an \nAUC of 0.948 for the test set (Figure 6D-E). \nDiscussion \nEstrogen, progesterone and androgen are \nwell-known hormones that play important roles in \nfemale reproductive disease [6, 26-29]. The effects of \nestrogen can be mediated by three types of receptors, \nincluding ESR1, estrogen receptor 2 (ESR2) and \nG-protein coupled estrogen receptor 1 (GPER1/ \nGPR30) [30, 31]. Progesterone exerts its biological \neffects by binding to progesterone receptors, \nincluding the nuclear receptor PGR and membrane \nreceptors PGRMC1 and progesterone receptor \nmembrane component 2 (PGRMC2) [32]. Upon \nbinding to an androgen, AR can translocate into the \ncell nucleus and then activate the transcription of its \ntarget genes [33]. \n\nInt. J. Med. Sci. 2023, Vol. 20 \n \n \nhttps://www.medsci.org \n422 \n \nFigure 3. PPI network analysis of the DEGs. In the network, the red circles represent up-regulated genes, while the blue circles represent down-regulated genes. The size of the \ncircle is positively correlated with |Log2FC|. (A) PPI network analysis of the DEGs between normal endometria and those of endometriosis patients. (B) PPI network analysis of \nthe DEGs between eutopic endometria and ectopic endometria from the same patients. PPI, protein-protein interaction.  \n\n\nInt. J. Med. Sci. 2023, Vol. 20 \n \nhttps://www.medsci.org \n423 \n \nFigure 4. The expression of AR in human tissues obtained from HPA database. (A) The RNA expression level of AR in human tissues. (B) The protein expression level of AR \nin human tissues. (C-D) The single-cell profile of human endometrium (C) and AR expression in each cell type (D). AR, androgen receptor; HPA, the Human Protein Atlas. \n\n\nInt. J. Med. Sci. 2023, Vol. 20 \n \nhttps://www.medsci.org \n424 \n \n \n \nFigure 5. The expression of AR in human endometrial tissues. (A-D) Representative images of AR staining in postmenopausal endometrium (A), premenopausal endometrium \n(B), eutopic endometrium from endometriotic patients (C) and their matched ectopic lesion (D). (E) T he corresponding histograms of positive AR staining level in different \ngroups. (F) Representative images of AR staining in endometrial stromal cells. Scale bar 100 µm. AR, androgen receptor. \n\n\nInt. J. Med. Sci. 2023, Vol. 20 \n \nhttps://www.medsci.org \n425 \n \nFigure 6. Establishment of a diagnostic model for endometriosis. (A) LASSO coefficient profiles of the genes in the normal endometrium and eutopic endometrium tissue from \nendometriosis patients. (B) Selection of the optimal parameter (lambda) in the LASSO model for the normal endometrium and eutopic endometrium tissue from endometriosis \npatients. (C) A nomogram model established based on the LASSO results. (D) ROC curve of the diagnostic nomogram model for the training set (GSE51981). (E) ROC curve of \nthe diagnostic nomogram model for GSE120103. LASSO, least absolute shrinkage and selection operator; ROC curve, receiver operating characteristic curve. \n\n\nInt. J. Med. Sci. 2023, Vol. 20 \n \n \nhttps://www.medsci.org \n426 \nPrevious studies have reported that the \ndecreased ESR1/ESR2 in ectopic lesions leads to the \ndecreased expression of PGR, which can exacerbate \nthe inflammatory response, thereby contributing to \nendometriosis. In our study, we found that, compared \nwith eutopic endometria, the expression of ESR1 in \nectopic lesions was decreased. However, no \nsignificant changes in PGR and ESR2 were \ndiscovered. The most widely accepted hypothesis for \nthe occurrence of endometriosis is Sampson’s \nretrograde menstruation theory, which postulates that \nit is retrograde menstruation, which enters the cavity, \nthat results in endometriosis [34]. The phenomenon \nthat the prevalence of retrograde menstruation is \nmore than 90% while endometriosis affects only 10% \nof the female population further reflects that there \nremain different characteristics between normal \nendometria and the eutopic endometria of \nendometriosis patients [34, 35]. So, in this study, we \nalso analyzed the expression of the receptors between \neutopic endometria of endometriotic patients and \nnormal endometria of healthy women. The results \nshowed that, compared with normal endometria, the \nexpression of AR, PGR, and PGRMC1 in eutopic \nendometria derived from patients with endometriosis \nwas decreased. Of course, strictly speaking, PGRMC1 \ndoesn’t belong to the steroid receptors [36]. As a \nmember of a multi-protein progesterone-binding \ncomplex, PGRMC1 cannot bind directly to \nprogesterone [36]. \nNext, the enrichment and PPI analysis were \nperformed. The results showed that the DEGs \ninvolved in the eutopic endometrium aberrations of \nendometriotic patients and ectopic lesions functioned \ndifferently. However, both contained genes that \nparticipate in the hormone response, in which the \nnuclear receptors (AR, PGR and ESR1) were included. \nThese nuclear receptors (AR, PGR and ESR1) were \nalso involved in almost all the top 20 enriched \nbiological processes of the DEGs. What’s more, both \nAR and PGR were identified as the hub genes \nbetween normal endometria and those of \nendometriosis patients, and ESR1 was selected as the \nhub gene between eutopic and ectopic endometria \nfrom the same patients. These results imply that sex \nsteroid hormones and their receptors may play \nimportant roles in endometriosis development. \nBesides steroid hormone response, we found that \npeptide-related processes were also involved in \nendometriosis development. The synthesis of sex \nsteroid hormones begins with the secretion of \ngonadotropin-releasing hormone (GnRH), which \nbelongs to peptide hormones [37]. In addition, \nluteinizing hormone (LH), which can stimulate the \nproduction of sex hormones, also contributes to the \nformation of ectopic lesions. Both ‘cellular response to \nluteinizing hormone stimulus’ and ‘response to \nluteinizing hormone’ were enriched.  \n Several studies have previously reported the \naberrations of estrogen and progesterone receptor \npathways in endometriosis [6, 31, 38, 39]. As for AR, it \nis reported that the positive staining of AR can be \ndetected in the stroma and glandular epithelium of \neutopic endometrium and ectopic lesions [40], and \ncytosine, adenine, and guanine (CAG) repeat variants \nof AR gene were associated with the increased risk of \nendometriosis [41, 42]. However, the aberrations of \nAR expression in eutopic and ectopic endometrium \nwas uncertain [40]. In this study, the IHC results \ndisplayed the significantly decreased expression of \nAR in the eutopic endometrium of endometriotic \npatients compared with normal endometrium. This \nstudy also found a high expression level of AR in the \norgans of the male and female reproductive systems, \nsuch as testis, endometrium and breast. And the \npositive expression of AR in the main cell types for \nendometriosis development can also be detected, \nespecially in endometrial stromal cells. However, no \nsignificant difference was found between ectopic \nlesions and their matched eutopic tissues, although \nthe expression of AR in ectopic lesions seemed to be \nhigher. For postmenopausal endometrium, AR \nexpression seemed to be decreased, however, a \nsignificant difference has not been detected, possibly \ndue to the limited sample numbers. \n Then we performed disease prediction using AR \nand its interactors. Seven independent factors, inclu-\nding APPL1, CSNK2A1, ERG, KDM4A, SMARCC1, \nSUZ12 and TRIM25, were filtered. The results showed \nthat the AUC of the nomogram model for the training \nset (GSE51981) was 0.984; this finding was further \nverified on the test set (GS120103), with an AUC of \n0.948. APPL1 may function as an adaptor protein in \nmany pathways, including the insulin and \nadiponectin signaling pathways, and suppresses \nandrogen receptor transactivation by potentiating Akt \nactivity [43, 44]. APPL1, Akt, and AR form a complex \nin which Akt serves as the bridge factor for the \nassociation of APPL with AR [43]. CSNK2A1 is the \ngene encoding CK2 alpha, the catalytic subunit of \nprotein kinase casein kinase 2 (CK2) [45]. CK2 can \nincrease AR protein stability and promote AR-depen-\ndent transcriptional activity [45]. Additionally, a \nsignificant positive correlation was observed between \nCSNK2A1 and AR mRNA levels in prostate cancer \n[46]. ERG is a member of the E-26 transformation- \nspecific (ETS) family, which has been extensively \nstudied in the field of prostate cancer in recent years \n[47]. ERG can disrupt AR signaling by inhibiting AR \nexpression or by binding to AR at gene-specific loci \n\nInt. J. Med. Sci. 2023, Vol. 20 \n \nhttps://www.medsci.org \n427 \nand inhibiting its activity [48]. KDM4A is a histone \ndemethylase related to AR [49]. KDM4A can enhance \nAR-activated gene transcription by forming \ncomplexes with ligand-bound AR, thereby mediating \nmultiple processes, including cell proliferation, \ndifferentiation, development, and metabolism [50]. \nSMARCC1 is a core subunit of the SWI/SNF complex \nand has been found to play important roles in the \ndevelopment of several cancers [51, 52]. The \nSWI/SNF complex, containing 5 core subunits and 7–\n15 accessory subunits, functions by interfering with \nhistone-DNA contacts [53]. Almost 25% of all cancers \nharbor mutations in one or more of these subunits \n[54]. The interaction of SMARCC1 and AR has been \nshown by affinity capture-MS and affinity \ncapture-western experimental techniques [55, 56]. \nSUZ12 is the core subunit of polycomb repressive \ncomplex 2 (PRC2), the epigenetic repressor complex \n[57]. It was reported that PRC2 can regulate the \nAR-associated signaling pathway [58]. The expression \nof SUZ12 was also correlated with the transcriptional \nfunction of AR [59]. TRIM25 has been defined as the \ndownstream target of ESR1 and has been shown by \naffinity capture-MS to interact with AR [59, 60]. \nAlthough we identified the gene sets that appear to \nhave predictive value for endometriosis development, \ntheir use for clinical prediction still needs substantial \nclinical validation. Instead, the gene sets that can \nmodulate AR signaling were involved in endometri-\nosis development and displayed good predictive \nvalue, which also indicates the importance of AR \nsignaling on disease occurrence and provides new \ntargets for the disease. Androgen can inhibit \nendometrial growth, reduce the chronic pain and \ninflammation [7, 61, 62]. The administration of the \nsynthetic androgen Danazol is effective in treating \npain and reducing lesions in endometriosis, but its \nsignificant androgenic side-effects limit its use [61]. \nThe search for the specific targets of AR signaling \nregulation in endometriosis may provide the new \ninsight for the development of treatment options. Of \ncourse, the further research on the roles of AR in \nendometriosis development and how these genes \ninfluence AR signaling in endometriosis still needs to \nbe further explored.  \n In summary, this study explored the importance \nof sex hormone receptors in endometriosis \ndevelopment and improved our understandings of \nthe pathogenesis of endometriosis. Furthermore, the \npotential roles of AR in endometriosis development \nprovide us new insights into the disease, which may \nlead to the development of novel treatment strategies.  \nSupplementary Material \nSupplementary tables. \nhttps://www.medsci.org/v20p0415s1.pdf \nAcknowledgments \nThis work was supported by the National \nNatural Science Foundation of China (Grant no. \n81871142). Additionally, we thank the personnel of \nthe Gene Expression Omnibus (GEO) database for \nproviding their platforms and contributors for \nuploading their meaningful datasets. \nCompeting Interests \nThe authors have declared that no competing \ninterest exists. \nReferences \n1.  Osuchowska-Grochowska I, Blicharska E, Gogacz M, et al. Brief Review of \nEndometriosis and the Role of Trace Elements. Int J Mol Sci. 2021; 22. \n2.  Fukunaga M. Uterus-like mass in the uterine cervix: superficial cervical \nendometriosis with florid smooth muscle metaplasia? Virchows Arch. 2001; \n438: 302-5. \n3.  Murakami K, Kotani Y, Nakai H, et al. Endometriosis-Associated Ovarian \nCancer: The Origin and Targeted Therapy. Cancers (Basel). 2020; 12. \n4.  Kralickova M, Lagana AS, Ghezzi F, et al. Endometriosis and risk of ovarian \ncancer: what do we know? Arch Gynecol Obstet. 2020; 301: 1-10. \n5.  Saunders PTK, Horne AW. Endometriosis: Etiology, pathobiology, and \ntherapeutic prospects. Cell. 2021; 184: 2807-24. \n6.  Marquardt RM, Kim TH, Shin JH, et al. Progesterone and Estrogen Signaling \nin the Endometrium: What Goes Wrong in Endometriosis? Int J Mol Sci. 2019; \n20. \n7.  Franca PRC, Lontra ACP, Fernandes PD. Endometriosis: A Disease with Few \nDirect Treatment Options. Molecules. 2022; 27. \n8.  Wu J, Fang X, Xia X. Identification of Key Genes and Pathways associated with \nEndometriosis by Weighted Gene Co-expression Network Analysis. Int J Med \nSci. 2021; 18: 3425-36. \n9.  Yilmaz BD, Bulun SE. Endometriosis and nuclear receptors. Hum Reprod \nUpdate. 2019; 25: 473-85. \n10.  Tamaresis JS, Irwin JC, Goldfien GA, et al. Molecular classification of \nendometriosis and disease stage using high-dimensional genomic data. \nEndocrinology. 2014; 155: 4986-99. \n11.  Bhat MA, Sharma JB, Roy KK, et al. Genomic evidence of Y chromosome \nmicrochimerism in the endometrium during endometriosis and in cases of \ninfertility. Reprod Biol Endocrinol. 2019; 17: 22. \n12.  Khan MA, Sengupta J, Mittal S, et al. Genome-wide expressions in autologous \neutopic and ectopic endometrium of fertile women with endometriosis. \nReprod Biol Endocrinol. 2012; 10: 84. \n13.  Hever A, Roth RB, Hevezi P, et al. Human endometriosis is associated with \nplasma cells and overexpression of B lymphocyte stimulator. Proc Natl Acad \nSci U S A. 2007; 104: 12451-6. \n14.  Barrett T, Wilhite SE, Ledoux P, et al. NCBI GEO: archive for functional \ngenomics data sets--update. Nucleic Acids Res. 2013; 41: D991-5. \n15.  Dennis G, Jr., Sherman BT, et al. DAVID: Database for Annotation, \nVisualization, and Integrated Discovery. Genome Biol. 2003; 4: P3. \n16.  Shannon P, Markiel A, Ozier O, et al. Cytoscape: a software environment for \nintegrated models of biomolecular interaction networks. Genome Res. 2003; \n13: 2498-504. \n17.  Chin CH, Chen SH, Wu HH, et al. cytoHubba: identifying hub objects and \nsub-networks from complex interactome. BMC Syst Biol. 2014; 8 Suppl 4: S11. \n18.  Uhlen M, Fagerberg L, Hallstrom BM, et al. Proteomics. Tissue-based map of \nthe human proteome. Science. 2015; 347: 1260419. \n19.  Uhlen M, Oksvold P, Fagerberg L, et al. Towards a knowledge-based Human \nProtein Atlas. Nat Biotechnol. 2010; 28: 1248-50. \n20.  Detre S, Saclani Jotti G, Dowsett M. A \"quickscore\" method for \nimmunohistochemical semiquantitation: validation for oestrogen receptor in \nbreast carcinomas. J Clin Pathol. 1995; 48: 876-8. \n21.  Fahey ME, Bennett MJ, Mahon C, et al. GPS-Prot: a web-based visualization \nplatform for integrating host-pathogen interaction data. BMC Bioinformatics. \n2011; 12: 298. \n22.  Oughtred R, Stark C, Breitkreutz BJ, et al. The BioGRID interaction database: \n2019 update. Nucleic Acids Res. 2019; 47: D529-D41. \n23.  Balachandran VP, Gonen M, Smith JJ, et al. Nomograms in oncology: more \nthan meets the eye. The Lancet Oncology. 2015; 16: e173-e80. \n\nInt. J. Med. Sci. 2023, Vol. 20 \n \nhttps://www.medsci.org \n428 \n24.  Waldmann P, Ferencakovic M, Meszaros G, et al. AUTALASSO: an automatic \nadaptive LASSO for genome-wide prediction. BMC Bioinformatics. 2019; 20: \n167. \n25.  Bulun SE, Yilmaz BD, Sison C, et al. Endometriosis. Endocr Rev. 2019; 40: \n1048-79. \n26.  Falcone T, Flyckt R. Clinical Management of Endometriosis. Obstet Gynecol. \n2018; 131: 557-71. \n27.  Kim JJ, Kurita T, Bulun SE. Progesterone action in endometrial cancer, \nendometriosis, uterine fibroids, and breast cancer. Endocr Rev. 2013; 34: \n130-62. \n28.  Gompel A. Progesterone and endometrial cancer. Best Pract Res Clin Obstet \nGynaecol. 2020; 69: 95-107. \n29.  Berger MH, Messore M, Pastuszak AW, et al. Association Between Infertility \nand Sexual Dysfunction in Men and Women. Sex Med Rev. 2016; 4: 353-65. \n30.  Eyster KM. The Estrogen Receptors: An Overview from Different \nPerspectives. Methods Mol Biol. 2016; 1366: 1-10. \n31.  Chantalat E, Valera MC, Vaysse C, et al. Estrogen Receptors and \nEndometriosis. Int J Mol Sci. 2020; 21. \n32.  Grimm SL, Hartig SM, Edwards DP. Progesterone Receptor Signaling \nMechanisms. J Mol Biol. 2016; 428: 3831-49. \n33.  Matsumoto T, Shiina H, Kawano H, et al. Androgen receptor functions in male \nand female physiology. J Steroid Biochem Mol Biol. 2008; 109: 236-41. \n34.  Koninckx PR, Ussia A, Adamyan L, et al. Pathogenesis of endometriosis: the \ngenetic/epigenetic theory. Fertil Steril. 2019; 111: 327-40. \n35.  Alnafakh R, Choi F, Bradfield A, et al. Endometriosis Is Associated with a \nSignificant Increase in hTERC and Altered Telomere/Telomerase Associated \nGenes in the Eutopic Endometrium, an Ex-Vivo and In Silico Study. \nBiomedicines. 2020; 8. \n36.  Rohe HJ, Ahmed IS, Twist KE, et al. PGRMC1 (progesterone receptor \nmembrane component 1): a targetable protein with multiple functions in \nsteroid signaling, P450 activation and drug binding. Pharmacol Ther. 2009; \n121: 14-9. \n37.  Lovejoy DA. Peptide hormone evolution: functional heterogeneity within \nGnRH and CRF families. Biochem Cell Biol. 1996; 74: 1-7. \n38.  Tang ZR, Zhang R, Lian ZX, et al. Estrogen-Receptor Expression and Function \nin Female Reproductive Disease. Cells. 2019; 8. \n39.  Reis FM, Coutinho LM, Vannuccini S, et al. Progesterone receptor ligands for \nthe treatment of endometriosis: the mechanisms behind therapeutic success \nand failure. Hum Reprod Update. 2020; 26: 565-85. \n40.  Carneiro MM, Morsch DM, Camargos AF, et al. Androgen receptor and \n5alpha-reductase are expressed in pelvic endometriosis. BJOG. 2008; 115: \n113-7. \n41.  Fujimoto J, Hirose R, Sakaguchi H, et al. Expression of size-polymorphic \nandrogen receptor (AR) gene in ovarian endometriosis according to the \nnumber of cytosine, adenine, and guanine (CAG) repeats in AR alleles. \nSteroids. 1999; 64: 526-9. \n42.  Shin JJ, Choi YM, Choi HY, et al. Androgen receptor cytosine, adenine, and \nguanine trinucleotide repeat polymorphism in Korean patients with \nendometriosis: A case-control study. Eur J Obstet Gynecol Reprod Biol. 2017; \n218: 1-4. \n43.  Yang L, Lin HK, Altuwaijri S, et al. APPL suppresses androgen receptor \ntransactivation via potentiating Akt activity. J Biol Chem. 2003; 278: 16820-7. \n44.  Artimani T, Najafi R. APPL1 as an important regulator of insulin and \nadiponectin-signaling pathways in the PCOS: A narrative review. Cell Biol Int. \n2020; 44: 1577-87. \n45.  Borgo C, D'Amore C, Sarno S, et al. Protein kinase CK2: a potential therapeutic \ntarget for diverse human diseases. Signal Transduct Target Ther. 2021; 6: 183. \n46.  Trembley JH, Kren BT, Abedin MJ, et al. CK2 Pro-Survival Role in Prostate \nCancer Is Mediated via Maintenance and Promotion of Androgen Receptor \nand NFkappaB p65 Expression. Pharmaceuticals (Basel). 2019; 12. \n47.  Adamo P, Ladomery MR. The oncogene ERG: a key factor in prostate cancer. \nOncogene. 2016; 35: 403-14. \n48.  Yu J, Yu J, Mani RS, et al. An integrated network of androgen receptor, \npolycomb, and TMPRSS2-ERG gene fusions in prostate cancer progression. \nCancer Cell. 2010; 17: 443-54. \n49.  Morozov VM, Li Y, Clowers MM, et al. Inhibitor of H3K27 demethylase \nJMJD3/UTX GSK-J4 is a potential therapeutic option for castration resistant \nprostate cancer. Oncotarget. 2017; 8: 62131-42. \n50.  Shin S, Janknecht R. Activation of androgen receptor by histone demethylases \nJMJD2A and JMJD2D. Biochem Biophys Res Commun. 2007; 359: 742-6. \n51.  Cai X, Zhou J, Deng J, et al. Prognostic biomarker SMARCC1 and its \nassociation with immune infiltrates in hepatocellular carcinoma. Cancer Cell \nInt. 2021; 21: 701. \n52.  Xiao ZM, Lv DJ, Yu YZ, et al. SMARCC1 Suppresses Tumor Progression by \nInhibiting the PI3K/AKT Signaling Pathway in Prostate Cancer. Front Cell \nDev Biol. 2021; 9: 678967. \n53.  Carmel-Gross I, Levy E, Armon L, et al. Human Pluripotent Stem Cell Fate \nRegulation by SMARCB1. Stem Cell Reports. 2020; 15: 1037-46. \n54.  Mittal P, Roberts CWM. The SWI/SNF complex in cancer - biology, \nbiomarkers and therapy. Nat Rev Clin Oncol. 2020; 17: 435-48. \n55.  Wang Q, Udayakumar TS, Vasaitis TS, et al. Mechanistic relationship between \nandrogen receptor polyglutamine tract truncation and androgen-dependent \ntranscriptional hyperactivity in prostate cancer cells. J Biol Chem. 2004; 279: \n17319-28. \n56.  Lempiainen JK, Niskanen EA, Vuoti KM, et al. Agonist-specific Protein \nInteractomes of Glucocorticoid and Androgen Receptor as Revealed by \nProximity Mapping. Mol Cell Proteomics. 2017; 16: 1462-74. \n57.  Long Y, Hwang T, Gooding AR, et al. RNA is essential for PRC2 chromatin \noccupancy and function in human pluripotent stem cells. Nat Genet. 2020; 52: \n931-8. \n58.  Liu Q, Wang G, Li Q, et al. Polycomb group proteins EZH2 and EED directly \nregulate androgen receptor in advanced prostate cancer. Int J Cancer. 2019; \n145: 415-26. \n59.  Chuang CH, Wang WJ, Li CF, et al. The combination of the prodrugs \nperforin-CEBPD and perforin-granzyme B efficiently enhances the activation \nof caspase signaling and kills prostate cancer. Cell Death Dis. 2014; 5: e1220. \n60.  Vatapalli R, Sagar V, Rodriguez Y, et al. Histone methyltransferase DOT1L \ncoordinates AR and MYC stability in prostate cancer. Nat Commun. 2020; 11: \n4153. \n61.  Vannuccini S, Clemenza S, Rossi M, et al. Hormonal treatments for \nendometriosis: The endocrine background. Rev Endocr Metab Disord. 2022; \n23: 333-55. \n62.  Simitsidellis I, Saunders PTK, Gibson DA. Androgens and endometrium: New \ninsights and new targets. Mol Cell Endocrinol. 2018; 465: 48-60.","source_license":"CC0","license_restricted":false}