{"paper_id":"e7cf6052-6332-47eb-8168-510e9f62716c","body_text":"Endometriosis is a non-cancerous gynecological disorder characterized by the presence of endometrial glands and stromal cells outside the uterine cavity ( 1 ). It can be considered as the obvious cause of disability in the women with the reproductive ages affecting 6-12% of the asymptomatic women, approximately 71-78% of women suffering chronic pelvic pain and up to 5% of the infertile women ( 2 , 4 ).\nPathogenesis of endometriosis includes metaplastic alteration of epithelial cells in the peritoneal surface ( 5 ), retrograde of endometrial cells, immunological insufficiency, genetics and epigenetics ( 6 , 7 ), as well as hormone disruption ( 8 ). Highly embraced hypothesis for the endometriosis development is retrograde menstruation ( 9 ). It was found that women\nwith endometriosis have a considerable amount\nof basalis endometrium in their menstrual\ndebris than those without endometriosis ( 10 ),\nmore likely because of the excessive uterine\nperistaltic contractions in women suffering\nthis disease ( 11 ). The importance of basalis\nlayer of endometrium in the endometriotic\nlesion development can be related to the large\nnumbers of stem cells in this area ( 12 ). In this\nrespect, some studies also revealed that the\nstem cell theory has a significant role in the\nendometriosis pathogenesis ( 13 ,  14 ).\nIn addition, recent medications were found\nto have severe side-effects for treating\nendometriosis. Therefore, topnotch and effective\ntreatments for endometriosis are required. The\nmain mode of action for all of the present\nmedications in treating pains, associated with\nendometriosis, is mostly caused by suppression\nof the implants proliferation ( 15 ,  16 ). The\ntheory of stem cell in endometriosis began the\nlast advanced avenue in the targeting these cells\nas cutting-edge therapy ( 17 ).\nIn this respect, lovastatin shifts the cells from a\nstem-like state to more differentiated condition\nand reduces the stemness ( 18 ). Furthermore,\nlovastatin is effective in the suppression\nof cell proliferation and angiogenesis in an\nexperimental model of endometriosis ( 19 ). In\nthis line, lovastatin function via modulating\nDNA methyltransferase (DNMT) activity,\naltering methylation of gene promoters, and\nconsequently regulating mRNA expression in\nthe various malignancies ( 18 ,  20 ).\nOn the other hand, activity of DNMTs, the\nenzymes that catalyze addition of methyl\ngroups to cytosine residues in DNA, is elevated\nin the ectopic endometrium compared to the\nnormal control ( 21 ). DNMT inhibitors have\nprofoundly been examined as the promising\nnovel drugs for endometriosis treatment ( 22 -\n 24 ). Recently, decitabine and 5-azacytidine\nhave been introduced into the clinical trial\nexperiment ( 25 ), but it was found that DNMT\ninhibitors cause considerable toxicity. In\naddition, they interfere with protein translation\nprocedure through incorporating into RNA ( 26 ,\n 27 ). Because of this reason, drugs like statins,\ndemonstrating DNMT inhibitory function with\nno toxic side-effect, would open up a new\nhorizon regarding the novel advancement in the\ndisease treatment.\nSome investigations revealed that lovastatin\ntreatment leads to demethylation of the  BMP2 \npromoter, up-regulation of the  BMP2  mRNA\nand activation of BMP signaling pathway.\nConsequently, these alterations induce\ncolorectal cancer (CRC) cell differentiation\nand reduce proliferation of the respective cells\n( 18 ,  28 ).\nMoreover, BMP pathway, particularly  BMP2 \nplays a crucial role in the pathogenesis of\nendometriosis ( 29 ).  BMP2 , a tumor growth\nfactor (TGF) superfamily member, acts down-\nstream of PGR and is essential for the stromal\ncell differentiation and decidualization in\nboth mouse and human endometrium ( 30 ).\nFurthermore, Aghajanova et al. ( 31 ) found that\nBMP-2 can promote osteogenic differentiation\nof the human endometrial stem cells.\nIn this study, we initially set out to determine\n( 1 ) whether lovastatin treatment influences\nmethylation status of the  BMP2  promoter as\nwell as mRNA expression of the respective\ngene and ( 2 ) whether lovastatin can also alter\nthe expression level of other genes playing\npivotal role in differentiation and proliferation\npotential of endometrial mesenchymal stem\ncells (eMSCs), such as  GATA2  and  RUNX2  ( 32 ,\n 33 ). Additionally, we then evaluated the effects\nof lovastatin on the endometrial stem cell\nmarkers derived from the patient and normal\nindividuals.\n\nThis experimental investigation was approved\nby the Institutional Review Board of the Faculty\nof Medicine at Tarbiat Modares University in\nIran. Endometrial and endometriotic tissues\nwere obtained from six patients (endometrial\ntissues from three patients; endometriosis\nsamples from three patients) at Obstetric\nGynecology Department of Sarem Women Hospital (Tehran, Iran). The patients were\nundergone hysterectomy and laparoscopy\nfor benign pathologies and written informed\nconsent was also received from the participants.\nThe surgery was performed irrespective of the\nday of patient’s menstrual cycle. The exclusion\ncriteria were any endometrial abnormality (e.g.\npolyps, hyperplasia or cancer), administration\nof the hormonal treatment and gonadotropin-\nreleasing hormone (GnRH) agonist therapy.\nAdditionally, this study was performed\naccording to the Helsinki declaration.\nFirst, tissue was separated and washed with\nthe phosphate-buffered saline (PBS). It was\nminced into the small pieces measuring 1 mm3\nand digested with 1 mg/ml collagenase type I\n(Sigma, Germany) for 60 minutes at 37˚C and\ncentrifuged for 10 minutes at 500 g. Second,\ncells were plated in the 25 cm 2 tissue culture\nusing Dulbecco’s Modified Eagle’s Medium\n(DMEM, Biowest, France) supplemented with\n20% fetal bovine serum (FBS, Gibco, USA), 50\nmg/ml of streptomycin and 50 U/ml of penicillin\n(Invitrogen, USA) at 37˚C in 95% air and 5%\nCO 2 . After that, when cultures reached at 80 to\n90% confluence, eMSCs were trypsinized using\ntrypsin EDTA 0.25% (Biowest, France) and\nthen the media were replaced. For this study\ncells were treated with lovastatin and decitabin\nat the passage four.\nTo characterize and quantify the expression\nof MSCs markers according to the surface\nmolecular markers ( 34 ), flow cytometry analysis\nwas performed. First, cells were detached with\ntrypsin EDTA 25% at the end of third passage\nand washed with PBS by centrifugation (300 g,\n5 minutes). After that, cells (1×10 6 cells) were\nincubated with the monoclonal antibodies (e.g.\nCD90, CD44, CD146, CD45 and CD34) and\nthe matched-isotype control for 30 minutes at\n4˚C. Finally, cell analysis was performed using\nPartec CyFlow® Space flow cytometer system\n(German Biotechnology Company, Germany)\nand the flowmax Software.\nIn order to perform the osteogenic and\nadipogenic differentiation, eMSCs were seeded\nat the density of 2×10 4 cells/cm 2  in 24-well\ntissue culture plates and incubated in DMEM\novernight at 37˚C and 5% CO 2  until 80%\nconfluency. Differentiation was carried out using\nosteogenic and adipogenic media according\nto the manufacturer’s instructions. In this\nrespect, osteogenic differentiation was induced\nusing DMEM high glucose supplemented with\n10% FBS, 10 nM dexamethasone, 10 mM\nβ-glycerophosphate and 10 μM ascorbic acid\n2-phosphates (both from Sigma) for 21 days.\nAdditionally, adipogenic differentiation was\ncarried out by culturing eMSCs in DMEM\nhigh glucose supplemented with 10% FBS,\n1 µM dexamethasone, 10 µM ascorbic acid\n2-phosphate and 200 μM indomethacin (both\nfrom Sigma) for 21 days. Three weeks later,\nosteogenic and adipogenic differentiations were\nconfirmed by Alizarin Red and Oil Red (both\nfrom Sigma) staining, respectively ( 35 ).\nFirst, eMSCs derived from the endometriotic\ntissues were seeded at the density of 1×10 4 cells/cm 2  in a 24-well plate and cultured for\n24 hours. Second, cells were treated with 1,\n2 and 5 µM lovastatin diluted in dimethyl\nsulfoxide (DMSO), for 72 hours. Then,\neMSCs were incubated with standard medium\ncontaining 3-( 4 , 5 -dimethylthiazol-2-yl)-2,5-\ndiphenyltetrazolium bromide (Sigma) with final\nconcentration of 0.5 mg/ml (stock solution 5\nmg/ml MTT in PBS) for 4 hours at 37˚C. At the\nend of experiment, the medium was removed\nand 500 μl DMSO was added. Absorbance was\nevaluated at 540 nm in a 96-well plate using\nan Anthos 2020 Microplate Readers (Austria).\nExperiments were carried out in triplicate, from\nthree independent experiments ( 36 ,  37 ).\nFirst, MSCs from the human endometrium and\nendometriosis were seeded at an initial density of 60% confluence. They were then allowed to\nbe attached overnight, and after that treated with\nlovastatin and decitabin (both from Sigma).\nAccording to the previous investigations ( 33 ,\n 38 ), MSCs were treated in the 1, 2 and 5 µM\nconcentration of lovastatin for 72 hours, while\nthese cells were treated in DMSO, as vehicle\ngroup. In addition, dose of 2 µM was used for\ndecitabin treatment in the MSCs for 72 hours\n( 39 ). After treatment, the cells were trypsinized\nand used for flow cytometry analysis, real-\ntime polymerase chain reaction (PCR) and\nquantitative methylation specific PCR (qMSP).\nFirst, total RNA was isolated from the\neMSCs with High Pure RNA Isolation\nKit (Roche, Germany) according to the\nmanufacturer’s protocol. Second, the purity of\nRNA was determined, by gel electrophoresis,\nphotospectrometrically (ratio 260/280 nm),\nand by RT-PCR reactions. For each sample, 1\nµg of RNA was used to generate cDNA with\nRevertAid First Strand cDNA Synthesis Kit\n(Thermo Scientific, USA). Then, quantitative\nreverse transcriptase PCR was carried out to\ndetermine the expression of genes encoding\nBone Morphogenetic Protein2 ( BMP2 ), GATA\nbinding protein 2 ( GATA2 ), Runt-related\ntranscription factor ( RUNX2 ), hypoxanthine\nphosphoribosyl transferase 1 ( HPRT1 ) with\nStepOne™ Real-Time PCR system (Applied\nBiosystems, USA). Primers used for SYBR\nGreen assay were:\nBMP2\nF: 5ˊ-CCACCATGAAGAATCTTTGGAAGAAC-3ˊ\nR: 5ˊ-TGATAAACTCCTCCGTGGGGA-3ˊ\nGATA2\nF: 5ˊ-GCTCGTTCCTGTTCAGAAGGC-3ˊ\nR: 5ˊ-CCCATTCATCTTGTGGTAGAGGC-3ˊ\nRUNX2\nF: 5ˊ-CCCCACGACAACCGCACCAT-3ˊ\nR: 5ˊ-CGCTCCGGCCCACAAATCTC -3ˊ ( 40 )\nHPRT\nF: 5ˊ-GGTCCTTTTCACCAGCAAGCT-3ˊ\nR: 5ˊ-TGACACTGGCAAAACAATGCA-3ˊ.\nHPRT  values were used for normalization.\nPCR product length for  BMP2 ,  GATA2 ,  RUNX2 \nand  HPRT  primers was 101, 126, 289, and 94\nbp, respectively. Gene expression was calculated\nusing the ΔΔCt method ( 41 ).\nFirst, DNA was isolated from eMSCs using\nHigh Pure PCR Template Preparation Kit\n(Roche) as recommended by the manufacturer’s\ninstruction. Second, for sodium bisulfite\ntreatment, 300 ng of DNA was denatured by 0.2\nM NaOH for 10 minutes at 37˚C in 50 ml total\nvolume. Then, 30 µl of freshly prepared 10 mM\nhydroquinone (Merck, US) and 520 μl of 3.5 M\nsodium bisulfite (pH=5, Merck, US) were added\nto the samples. After that, each DNA sample\nwas incubated at 50˚C for 16 hours. Samples\nwere also purified with Roche DNA purification\ncolumns based on the manufacturer’s instruction\nand eluted in 200 μl of elution buffer. At last,\nsamples were desulfonated by 0.3 M NaOH\ntreatment for 5 minutes at 20˚C. Finally, after\nethanol precipitation, DNA was dissolved in 30\nμl distilled water ( 42 ).\nFor analyzing  BMP2  promoter methylation,\nMethySYBR Method was performed with\nStepOne™ Real-Time PCR System (Applied\nBiosystems). In this study, according to the\none-step MethySYBR method ( 43 ), the primers\n(BMP2-EXT-F and BMP2-EXT-R; product\nlength=308) were used in the externally nested\nreal time PCR amplified the target gene regardless\nof their methylation status. This was used as a\nreference control to normalize the proportion of\nmethylated target alleles which were detected by\nthe methylation specific primer pair (BMP2-FM\nand BMP2-RM, product length=113) between\nthe samples. Each reaction contained 20-25 ng\nof bisulfite-treated DNA as a template, 10 ml\n2x RealQ Master Mix ampliqon and 500 nM of\neach forward and reverse primer ( Table 1 ) in a\ntotal volume of 20 µl. For  BMP2 , real-time PCR\nthermocyclic conditions included an initial step\nof 10 minutes at 95˚C, followed by 40 cycles of\n95˚C for 15 seconds, and 60˚C for 30 seconds.\nIn this method, plasmid template was included\nas the control for calculation of methylation\npercentage of each sample. Methylated DNA\nlevel was calculated with 2 -ΔΔCt \nin which ΔΔC t  equals to ΔC t sample -ΔC t plasmid \n( 43 ). Furthermore,\nto evaluate the methylation status of  GATA2 ,\nwe performed qMSP using the primers directed\nagainst differentially methylated regions\nof exon 4 of  GATA2 . Briefly, primers were\ndesigned to determine either the methylated\nor unmethylated form of the sequence after\nthe bisulfite converted sequences of the sense\nstrand. Primer information is provided in Table 1. For  GATA2 , the thermocyclic conditions of\nreal-time PCR included an initial denaturation\nstep of 10 minute at 95˚C, followed by 40\ncycles of 95˚C for 15 seconds and 57˚C for 30\nseconds. Additionally, the product length for\nthe  GATA2 -Meth and  GATA2 -Unmeth primers\nwas 139 bp.\nComparison of gene expressions, methylation\nvalues as well as cell viability tests between\nsamples were assessed with a two tailed student’s\nt test using GraphPad Prism 6 software. Results\nwere statistically significant at a P<0.05.\n\nMSCs from human endometrium were\nisolated and cultured, while they predominantly\nhad fibroblastic shape as expected ( Fig.1A ). To\nevaluate differentiation potential of eMSCs,\ninduction to adipogenic and osteogenic lineage\nwas performed  in vitro . A potential for the\ndifferentiation to adipogenic lineage was\nconfirmed through staining of lipid vacuoles\nby oil red ( Fig.1B ). Furthermore, osteogenic\ndifferentiation was detected through alizarin\nred staining of calcium deposits ( Fig.1C ). Flow\ncytometer analysis indicated that cells expressed\nthe mesenchymal markers CD44 (94.60%),\nCD90 (94.33%) and CD146 (94.83%), but they\nlacked hematopoietic markers including CD45\n(3.77%) and CD34 (5.40%) ( Fig.1D-H ).\nPrimer sequences for qMSP analysis of  GATA2  and  BMP2\nMesenchymal stem cells (MSCs) characterization. Human endometrial MSCs (eMSCs) exhibited, A. A fibroblast-like cell\nshape. These cells also represented successful, B. Adipogenic, C. Osteogenic differentiation potential, showing a positive\nsignal for D. CD44 (94.60%), E. CD90 (94.33%), F. CD146 (94.83%) and no signal for G. CD45 (3.77%), and H. CD34 (5.40%)\nmarkers (n=3).\nTo investigate the effect of different\nlovastatin concentrations on  BMP2  and  RUNX2 \nmRNA expression, eMSCs were isolated\nfrom the patient and normal individuals and\nthen were incubated for 72 hours with 2 and\n5 µM concentrations of lovastatin. Relative\nexpression amounts of  BMP2  in plates treated\nwith the different concentrations of lovastatin,\nvehicle (DMSO) and 2 µM decitabin, as the\npositive control, are shown in  Figure 2 . In\ncomparison with untreated control of both\ngroups, relative expressions of  BMP2  at 2 µM\nlovastatin-treated eMSCs were 1.69 ± 0.26 and\n2.22 ± 0.1 fold further than those of control\neMSCs in the patients and normal groups,\nrespectively. Statistical analysis showed a\nsignificant difference between 2 μM lovastatin\nand control in both groups (P<0.05, Student’s\nt test). Furthermore,  RUNX2  expression was\nmarkedly up-regulated in the plates treated\nwith 2 and 5 µM of lovastatin as well as 2 µM\ndecitabin in comparison with the untreated\ncontrol of patient group (2.58 ± 0.32 fold, 1.86\n± 0.22 fold and 2.26 ± 0.18 fold, respectively,\nP<0.05 for 2 and 5 µM of lovastatin and P<0.01\nfor 2 µM decitabin).  RUNX2  expression was\nalso up-regulated in plates treated with 2 and\n5 µM of lovastatin as well as 2 µM decitabin\ncompared to the untreated control of normal\ngroup (3.35 ± 0.21 fold, 2.02 ± 0.10 fold and\n2.12 ± 0.10 fold respectively, P<0.05).\nFollowing 2 μM lovastatin treatment,  GATA2 \nexpression was slightly down-regulated, while the\nexpression of  GATA2  was significantly decreased\nin response to 5 μM lovastatin treatment in\ncomparison with the untreated control of patient\ngroup ( Fig.3 , 0.57 ± 0.14 fold, P<0.05). On the\nother hand,  GATA2  expression at 2 μM statin-\ntreated eMSCs were 1.75 ± 0.07 fold higher than\nthose of the control eMSCs in normal ( Fig.3 ,\nP<0.05). There was also no significant difference\nin the expression levels of  GATA2  in response to 2\nµM decitabin, compared to the untreated control\nof both groups.\nBMP2  and  RUNX2  expressions following the lovastatin treatment in endometrial mesenchymal stem cells (eMSCs) cultures of three\npatients and three normal individuals, detected by quantitative real-time polymerase chain reaction (RT-PCR). A. Relative expression of\n BMP2  at 2 μM statin-treated eMSCs was 1.69 ± 0.26 and 2.22 ± 0.1 fold higher than those of control eMSCs in patient and normal groups,\nrespectively (P<0.05, Student’s t test) and B.  RUNX2  expression was up-regulated in plates treated with 2 and 5 µM of lovastatin as well as\n2 µM decitabin in comparison with untreated control of patient group (2.58 ± 0.32 fold, 1.86 ± 0.22 fold and 2.26 ± 0.18 fold respectively,\nP<0.05 for 2 and 5 µM of lovastatin and P<0.01 for 2 µM decitabin) and normal group (3.35 ± 0.21 fold, 2.02 ± 0.10 fold and 2.12 ± 0.10\nfold respectively.\n*; P<0.05 and **; P<0.01 in comparison to untreated control in each groups.\nGATA2  expression following the lovastatin treatment in endometrial mesenchymal stem cells (eMSCs) cultures of three patients and three normal individuals, detected by quantitative real-time polymerase chain reaction (RT-PCR). Relative expression\nof  GATA2  was decreased in response to 5 μM lovastatin treatment in comparison with untreated control of patient gro0.57 ±\n0.14 fold (P<0.05). On the other hand,  GATA2  expression at 2 μM\nlovastatin-treated eMSCs was 1.75 ± 0.07 fold higher than those\nof control eMSCs in normal.\n*; P<0.05 in comparison to untreated control in each groups.\nTo determine involvement of DNA methylation in\nthe  BMP2  gene down-regulation of eMSCs treated\nwith the different concentration of lovastatin, we\nused quantitative Methylation Specific PCR (qMSP)\nfor the respective promoter status, starting 214 bp\nupstream of exon 1. This region contains a CpG\nisland that methylated in the gastric and colorectal\ncancers ( 18 ,  44 ). As Figure 4A shows, lovastatin\ntreatment induced demethylation of the  BMP2 \npromoter in eMSCs treated with 2 μM lovastatin\nfor 72 hours. The qMSP results showed that  BMP2 \npromoter methylation was decreased from 28.2 to\n7.6% in eMSCs after treatment with 2 μM lovastatin\nfor 72 hours (P<0.05, Student’s t test).\nWe also performed qMSP for  GATA2  before and\nafter lovastatin and decitabin treatments using the\nmethylated and unmethylated primers directed\nagainst differentially methylated region of  GATA2 \nexon 4 ( 45 ). As Figure 4B shows, lovastatin treatment\ninduced methylation of the differentially methylated\nregion of  GATA2  exon 4 in eMSCs treated for 72\nhours with 2 and 5 μM lovastatin. The qMSP results\nshowed that the  GATA2  exon 4 methylation was\nincreased in eMSCs from 12.0 to 26.95 and 70.49%\nafter treatment with 2 and 5 μM lovastatin for 72\nhours, respectively (P<0.05, Student’s t test).\nQuantitative methylation specific PCR (qMSP) analysis of\n BMP2  promoter region and  GATA2  exon 4 in endometrial mesenchymal stem cells (eMSCs) treated with different concentration of\nlovastatin. A. qMSP results showed that  BMP2  promoter methylation was decreased from 28.2 to 7.6% after treatment with 2 μM of\nlovastatin for 72 hours (P<0.05, Student’s t test) and B. qMSP results\nshowed that  GATA2  exon 4 methylation was increased from 12.0 to\n26.95 and 70.49% after treatment with 2 and 5 μM of lovastatin for\n72 hours, respectively.\n*; P<0.05 in comparison to untreated control in each groups.\nChanges in the cell viability could lead to a\ndecrease in cell number and a low influence on\ncell therapy. In order to determine the effects of\nlovastatin (diluted in DMSO) and DMSO on eMSCs\nviability, cells were incubated with 1, 2 and 5 µM\nlovastatin for 72 hours, and then, mitochondrial\ndehydrogenase activity was evaluated in the living\ncells by the MTT test ( 46 ). The cell viability in\nvehicle as well as treatment group was observed\nin approximately 80% of cells in comparison to\nthe untreated group. However, statistical analysis\nshowed a significant difference at the enzyme level\nunder 2 and 5 µM lovastatin and DMSO treatment\n( Fig.5 , P<0.05, Student’s t test).\nTo investigate the effect of lovastatin on\neMSCs markers, cells were treated with 1 μM\nlovastatin for 72 hours and then, analyzed by\nflow cytometer. The results indicated that\nCD146 cell marker was down-regulated to 53%\nin response to 1 µM lovastatin, compared to the\nuntreated group ( Fig.6 , P<0.05, Student’s t test).\nIn this respect, recent studies have shown that\nMelanoma Cell Adhesion Molecule (MCAM/\nCD146) was a key marker of endometrial stem/\nprogenitor cells involved in the inflammation\nand angiogenesis procedures ( 47 ,  48 ).\nMesenchymal stem cells (MSCs) were plated in 24 well plates and either no treated, or treated with dimethyl sulfoxide (DMSO) or\n1, 2 and 5 μM lovastatin (diluted in DMSO) for 72 hours, followed by MTT test. Values are shown as living cells percentage relative to the\ncontrol untreated cells with set at 100% in control values. Results expressed the mean ± SD (n=3).\n*; P<0.05 in comparison to untreated control in each groups.\nFlow cytometry analysis of eMSCs markers treated with 1 µM lovastatin. The results indicated that CD146 cell marker was down-regulated to 53% in response to 1 μM lovastatin in comparison with untreated group. Data are from three experiments ± SD (n=3).\n*; P<0.05 in comparison to untreated control.\n\nStem cell theory began the final advanced avenue for the etiology of endometriosis. A great number of studies demonstrated presence of the endometrial stem cells, not only from residing cells in the endometrium but also from reprograming bone marrow MSCs ( 17 ). Feasibility of targeting stem cells was suggested to be as of the remarkable advancement to eliminate endometriosis ( 49 ).\nThis study evaluated the effect of lovastatin on eMSCs properties including differentiation and proliferation potential. In the current experiment,  BMP2  activity was significantly augmented in eMSCs within three days after treatment with 2 µM concentration of lovastatin.  BMP2  activity was proved to be a marker of osteogenesis differentiation ( 38 ). Previous studies have demonstrated that lovastatin increased the level of  BMP2  gene expression ( 50 , 51 ). Moreover,  BMP2  reportedly was downregulated in endometriosis ( 29 ). In this respect, there was remarkable evidence showing that  BMP2  signaling pathway plays a pivotal role in the decidualization ( 52 , 53 ). The study carried out by van Kaam et al. ( 54 ) revealed that both ectopic and eutopic endometrium of patients suffering from endometriosis demonstrated a decreased capacity for differentiation, as well as decidualization and implantation.\nIn addition,  RUNX2  expression level was significantly increased in the treatment of 2 and 5 μM lovastatin, compared to the control group.  RUNX2  is a major downstream mediator of  BMP2 , functioning and playing a critical role in the stromal differentiation and decidualization ( 55 ). Furthermore, MSCs differentiation towards osteogenic lineage was determined by definite group of elements ( 56 ). Among these factors, the first and highly specific marker was  RUNX2 . In this line,  RUNX2  activated osteogenic differentiation by signaling pathways including  BMP2  and TGF-β1 ( 57 , 58 ).\nIn the present study, expression of GATA2 was investigated in eMSCs after lovastatin treatment. In this case,  GATA2  mRNA level was significantly decreased in response to the 5 μM lovastatin treatment, compared to the untreated control of patient group.\nIncreasingly, it was found that  GATA2  expression, a member of the six zinc-finger family transcription factors, was essential for various tissues including urogenital and hematopoietic system and adipose maturation ( 59 ). Moreover, Kamata et al. ( 32 ) demonstrated that  GATA2  could be one of the significant factors regulating differentiation of bone marrow MSCs toward adipocytes. Given the results of previous investigations ( 59 , 32 ), reduction of  GATA2  in response to lovastatin treatment might be in favor of decreased proliferation and increased differentiation potentials. However, this observation should be confirmed by other studies.\nFurthermore, SYBR Green-based quantitative real time PCR method was performed to analyze DNA methylation level in eMSCs. The MethySYBR assay is a very sensitive, precise and less vulnerable to false positives ( 60 ). In this study, lovastatin treatment induced DNA demethylation and reactivation of  BMP2  gene expression, which was suppressed by hypermethylation in the endometriosis. More importantly, we found demethylation of other methylated genes including  RUNX2  in the endometriosis after treatment with lovastatin, implying more general effect on gene hypermethylation. Given the results of this study, it is not obvious how lovastatin inhibits DNMTs. Kodach et al. ( 18 ) showed that lovastatin has either little or no effect on DNMTs expression levels. Therefore, further investigations are required to evaluate the mechanism(s) by which lovastatin inhibit DNMTs. On the other hand, we found that lovastatin induced methylation of  GATA2  factor. This result was in consistent with the latest data reported by MacLeod et al. ( 20 ), showing that lovastatin therapy is related to higher MTHFR methylation levels in a stroke group implying that statins can induce DNA methylation.\nWe also evaluated the effect of lovastatin on expression of MSCs markers. Some recent investigations have indicated that CD146 could be considered as a highly specific marker of endometrial stem/progenitor cells ( 17 , 61 ). In addition, Figueira et al. ( 13 ) used CD146 marker to identify mesenchymal stem cells for the first time. The eMSCs expressed typical MSC surface markers including CD44, CD90 and CD105.\nIn this study, flow cytometer data displayed that CD146 was reduced in response to lovastatin treatment in eMSCs, suggesting that CD146 could effectively be implicated in the endometriosis pathogenesis by activating the angiogenesis and inflammation ( 47 ). Additionally, CD146 is an endothe¬lial cell adhesion molecule that is upregulated in different types of malignant cell, such as ovarian cancer ( 62 , 65 ). A great number of experiments have suggested that CD146 induced angiogenesis, tumor growth and metastasis ( 66 ). Moreover, Flanagan et al. ( 67 ) showed that laminin-411 attached to CD146 enabling TH17 cell penetrate into the tissues and induce inflammation. Studies have revealed that lovastatin, which is a potent inhibitor for the expression of VEGF, plays a pivotal role in diminishing blood-vessel formation ( 68 ). Similarly, in the recent study, Jiang et al. ( 69 ) reported that CD146 interacts with VEGFR-2 in a tumor angiogenesis mechanism. In line with previous investigations, our research presents a new target of action for lovastatin, in inhibition of angiogenesis via suppressing CD146.\nBased on the previous  in vitro  studies, doses of 1 to 5 µM were used for lovastatin treatment in MSCs ( 33 , 38 ). In these experiments, mild growth stimulatory effects in eMSCs were derived from human endometrium, and endometriosis was observed at dose of 1 and 5 µM. In addition, Kupcsik et al. ( 38 ) revealed that lovastatin concentration at 10 µM is associated with cytotoxic effects and leads to detachment of eMSC from culture plate.\nZhou and Hu ( 23 ) showed that stem cell differentiation could be augmented by DNA demethylation, starting advancement for studying the induction of stem cell fate through epigenetic reprograming. In this study, for the first time, we demonstrated that aberrant demethylation of CpG island promoter of  BMP2  occurred in endometriosis tissues. We also provided a facet of molecular basis of the  BMP2  down-regulation in these tissues from the viewpoint of epigenetic disease. It is hoped that epigenetic reprograming of  BMP2  becomes a helpful cue for the further research in the pathogenesis of endometriosis.\n\nThe proposed mechanisms of statins action on the endometriosis tissues are suppression of endometrial cells proliferation and apoptosis, reduction of oxidative stress and inflammation, and inhibition of the angiogenesis. Our study indicated that lovastatin treatment could increase osteogenic differentiation through up-regulation of  BMP2  and  RUNX2  mRNA expression. In addition, reduction of  GATA2  in response to lovastatin treatment might be in favor of increased adipogenic differentiation potentials. Expression of stem cell markers and subsequently stemness was also reduced in the eMSCs after lovastatin treatment.\nFurthermore, consistent with the previous studies, our investigation revealed that lovastatin decreased angiogenesis and increased implantation and decidualization.\nSeveral recent investigations have suggested that statins could have a pivotal role in the medical management of women suffering from endometriosis. They also offer clinical benefits without interfering in estrogen. Despite this fact, more clinical trials are needed to confirm the safety and effectiveness of this kind of treatment in endometriosis.","source_license":"CC0","license_restricted":false}