Ceratonia siliqua L. pod Effects on Viability Gene Expression of Endometrial Mesenchymal Stromal/Stem Cells Isolated from Women with Endometriosis-Associated Infertility

Endometriosis is a chronic disorder characterised by abnormal growth and adherence of functional endometrial tissue outside the uterine cavity. It is a common cause of chronic pelvic pain and infertility in reproductive-aged women, with estimates suggesting that 20-50% of infertile women suffer from endometriosis and 30-50% of women with endometriosis are sub-fertile or infertile ( 1 ). Chaichian et al. ( 2 ) reported that the prevalence of endometriosis among 441 infertile Iranian women who underwent laparoscopy is 18.6. Genetic and epigenetic factors play important roles in the development of endometriosis ( 3 ). DNA methyltransferases (DNMTs), namely DNMT1, DNMT3A and DNMT3B that maintain CpG methylation, along with histone deacetylases (HDACs), play central roles in epigenetic regulation of cancers ( 4 ). Previous studies found that aberrant expressions of the DNMT1, DNMT3A, DNMT3B, and HDAC1 genes were more prominent in women with endometriosis, as this disease exhibits malignant-like features ( 5 ). Overexpression of cyclooxygenase-2 (COX-2) and the subsequent increase in prostaglandin E2 level are the main causes of dysmenorrhea and recurrence of endometriotic lesions ( 6 ). Matrix metalloproteinase-2 ( MMP2 ) is another well-known factor that plays a role in endometriosis development ( 7 ). MMP2 contributes to the degradation of collagens and other components of the extracellular matrix and basement membrane, which results in disseminated malignancies ( 8 ). A similar mechanism is likely involved in the pathogenesis of endometriosis ( 7 ). Despite various medical and surgical strategies used for the management of endometriosis, current approaches are not completely effective and often have major side effects ( 9 ). Therefore, novel medical agents obtained from traditional medicine may prove effective in targeting the biological pathways involved in endometriosis and improve treatment outcomes. Ceratonia siliqua L ., commonly known as carob, is an evergreen tree that belongs to the Le-guminosae family, and is widely grown in Mediterranean countries. It has been used in folk medicine to treat diabetes mellitus, dyslipidaemia, and various gastrointestinal ailments ( 10 ). Furthermore, indigenous people in Middle Eastern countries have used this natural product since ancient times to improve fertility ( 11 ). The main constituents of Ceratonia siliqua are gum, protein, and polyphenols, of which the latter is mainly responsible for anti-inflammatory, antioxidant, and anti-ulcer activities ( 12 ). Elbouzidi et al. ( 13 ) evaluated the antioxidant and antimicrobial properties of carob to inhibit the proliferation of breast cancer cells (MCF-7, MDA-MB-231, and MDA-MB-436) and studied its genotoxicity. It has been traditionally used for medicinal purposes ( 10 , 12 ); moreover, there is limited research on the potential uses of carob, and they are often considered a by product of carob production. The aim of this study is to investigate the potential therapeutic effect of carob (Ceratonia si-liqua L.) pod extract (CPE) on gene and peptide expressions of DNMT1, DNMT3A, DNMT3B, HDAC1, MMP2 , and COX-2 in endometrial mesenchymal stromal/stem cells (EnMSCs) isolated from women with endometriosis-associated infertility (EAI).

The results showed that the human EnMSCs in all three groups were positive for the standard mesenchymal markers CD73 and CD90, and negative for the hematopoietic marker CD45 and endothelial marker CD31. Surface antigen marker analysis of the endometrial cell cultures revealed relatively homogenous cell populations that were free from contaminant cells ( Fig .S1A , See Supplementary Online Information at www.ijfs.ir). The EnMSCs in the C-EnMSCs were plastic adherent cells that exhibited a spindle-shaped, fibroblast-like morphology with radial or helical growth patterns ( Fig .S1B , See Supplementary Online Information at www.ijfs.ir). EEnMSCs and OMA-EnMSCs isolated from different patients also showed typical spindle-shaped and fibroblastlike morphological features. The C-EnMSCs, E-EnMSCs, and OMA-EnMSCs cultured in osteogenic and adipogenic media for one month and three weeks, respectively, were positive for Alizarin Red S and Oil Red O staining. The red brilliant stain in the culture plate indicated the calcium deposits by EnMSCs that had differentiated into osteoblasts. The lipid droplets appeared in the cytoplasms of the EnMSCs, which confirmed their multipotent capabilities ( Fig .S1B , See Supplementary Online Information at www.ijfs.ir). Polyphenols are one of the important constituents of hydrochloric CPE. Therefore, the TPC was evaluated using the Folin-Ciocalteu assay. The results revealed the presence of 18.13 ± 2.47 mg of pure phenol in 1 g of dry CPE. Several compounds with different biological activities were observed in the GC-MS profile of the CPE ( Fig .S2 , See Supplementary Online Information at www.ijfs.ir). In detail, 60 compounds were identified in the CPE, four of which had anticancer biological activities based on previous studies. Moreover, four compounds had cytotoxic effects, and another four showed antioxidant properties in accordance with previous studies. In addition, six of the CPE compounds had anti-inflammatory biological effects. Other biological activities were also observed in the CPE compounds and included antimicrobial, antiseptic, and muscle relaxant effects based on previous studies ( Table S2 , See Supplementary Online Information at www. ijfs.ir ). The antioxidant potency and reduction capability of the DPPH radicals were determined by a reduction in absorbance at 517 nm that was induced by the antioxidants. A high antioxidant capacity was reported for the CPE, with an IC50 value of 0.63 ± 0.11 mg/mL, compared to quercetin (positive control), which showed an IC 50 value of 1.40 ± 2.89 mg/mL. The cell viability percentages of EnMSCs exposed to various concentrations of CPE after one, three, and seven days were repeated three times, and the mean values were calculated. CPE concentrations higher than 2 μg/ mL did not result in acceptable EnMSC viability after seven days and were excluded. EnMSC viability in CPE concentrations of less than 2 μg/mL were not significantly different. Thus, we chose the 2 and 0.8 μg/mL concentrations based on the proliferation and viability results for the following gene studies. One-way ANOVA and the Duncan test showed a statistically significant decrease in proliferation and viability between the 2 and 0.8 μg/mL concentrations in C-EnMSCs compared to the untreated (0 μg/mL) group ( Fig .1 ). Endometrial mesenchymal stromal/stem cell (EnMSCs) viability ac- cording to the MTT assay results with different Ceratonia siliqua L . pod extract (CPE) concentrations. Lines above the columns indicate statisti- cally significant differences. Data are analysed by one-way ANOVA and the Duncan test (mean ± SD). *; P<0.05, C-EnMSCs; Control-EnMSCs, E-EnM- SCs; Endometriosis-EnMSCs, and OMA-EnMSCs; Endometrioma-EnMSCs. Total RNA was extracted from E-EnMSCs, OMAEnMSCs, and C-EnMSCs after seven days of exposure to CPE. cDNA synthesis was performed, and quantitative PCR was used to determine gene expression levels. Gene expression levels at CPE concentrations of 0.8 μg/mL and 2 μg/mL were compared to the C-EnMSCs ( Fig .2 ). Treatment with 0.8 μg/mL of CPE significantly reduced MMP-2 gene expression in OMA-EnMSCs (P=0.022, Fig .2A ). However, the differences were not statistically significant (P>0.05) for the E-EnMSCs and C-EnMSCs ( Fig .2A ). COX-2 gene expression significantly downregulated in the E-EnMSCs (P=0.022) and upregulated in the C-EnMSCs (P=0.022) after treatment with 0.8 and 2 μg/ mL of CPE, respectively ( Fig .2B ). HDAC1 had significantly lower expression in the EEnMSCs after treatment with 2 μg/mL of CPE (P=0.022). HDAC1 gene expression showed a nonsignificant decrease after CPE treat-ment with both concentrations in the OMA-EnMSCs compared to the C-EnMSCs (P>0.05, Fig .2C ). When compared to the C-EnMSCs, DNMT1 gene expression decreased after treat-ment with 0.8 and 2 μg/mL CPE in the E-EnMSCs and OMA-EnMSCs. However, the dif-ferences did not show any statistical significance (P>0.05, Fig .2D ). DNMT3A gene expression significantly increased with the 2 μg/mL concentration of CPE in the C-EnMSCs (P=0.034). However, its expression after treatment with E-EnMSCs and OMA-EnMSCs was comparable to the C-EnMSCs ( Fig .2E ). Treatment with 0.8 μg/mL of CPE reduced DNMT3B gene expression in OMA-EnMSCs (P=0.034). Treatment with 2 μg/mL of CPE had no significant effect on DNMT3B gene expression ( Fig .2F ). Different Ceratonia siliqua L . pod extract (CPE) concentrations were used to determine the relative mRNA expression levels (mean ± SD) of gene expressions. A. MMP2 , B. COX-2 , C. HDAC1 , D. DNMT1 , E. DNMT3A and F. DNMT3B , and in the three study groups: i. Endometrial mesenchymal stromal/ stem cells (EnMSCs) derived from the endometrium of endometrioma patients (OMA-EnMSCs), ii. Deep infiltrative en-dometriosis (DIE) samples of women with endometriosis associated infertility (E-EnMSCs), and iii. EnMSCs derived from the endometrium of an endometriosis free, normal woman (C-EnMSCs). The lines above the columns show significant differences (P<0.05). Data are analysed by one-way ANOVA and the Duncan test. Western blot analysis demonstrated that the related antibodies decreased in the E-EnMSCs, OMA-EnMSCs, and C-EnMSCs after treatment with CPE. COX-2, HDAC1, DNMT1, DNMT3A, and DNMT3B expressions decreased in the E-EnMSCs, OMA-EnMSCs, and C-EnMSCs after treatment with 0.8 and 2 μg/mL of CPE (Fig. S3, See Supplementary Online Information at www.ijfs.ir, Fig .3 ). Although there was a decrease in MMP2 expression in the OMA-EnMSCs, this result was not observed in E-EnMSCs and C-EnMSCs. Rather, MMP2 expression increased in the E-EnMSCs and C-EnMSCs ( Fig .3 ). GC-MS analysis of CPE identified 14 bioactive molecules that were investigated for their interactions with six target proteins through the docking process. The docking process produced multiple possible conformations for each ligand-protein complex, and we selected the most stable complex with the lowest binding affinity, ΔG [U total in kcal/mol]. The results of the docking process are presented in Table 2 . The binding affinity of the selected confor-mation ranged from -3.6 to -7.8 Kcal/mol. Diphenyl sulfone, 3,4-dihydro-2(1H)-isoquinoline carboximidamide, methanone (1-hydroxycyclohexyl)phenyl, and 4-imidazolidinone,1-benzoyl-2-(1-methylethyl)- 3-methyl exhibited the highest affinity to COX-2 with ΔG values of -7.4, -7.3, -7.2, and -7.1 Kcal/mol, respectively. 4-Imidazolidinone,1-benzoyl-2-(1-methylethyl)-3-methyl; ben-zene, 1, 3, 5-tris(1-methylpropyl)-; 3.4-dihydro2(1H)-isoquinoline carboximidamide; naphthalene, 1, 2, 3, 4-tetrahydro-1-isopropyl-1, 2, 4, 4, 7-pentamethyl-; naphtha-lene, decahydro-2, 3-dimethyl-; benzoic acid, 4-heptyl-, 4-cyanophenyl ester; and methanone, (1-hydroxycyclohexyl)phenyl showed the highest affinity to DNMT3A with ΔG values of -5.8, -5.6, -5.5, -5.4, -5.4, -5.4, and -5.4 Kcal/mol. Diphenyl sulfone; methanone, (1-hydroxycyclohexyl)phenyl-; 3.4-dihydro-2(1H)-isoquinoline carboximidamide; and 4-imidazolidinone,1- benzoyl-2-(1-methylethyl)-3-methyl exhibited the highest affinity to MMP2 with ΔG values of -6.8, -6.7, -6.6, and -6.6 Kcal/mol. The best affinity to HDAC1 was observed for benzoic acid, 4-heptyl-,4-cyanophenyl ester with ΔG of -6 Kcal/mol. The highest affinity to DNMT1 was related to naphtha-lene, 1, 2, 3, 4-tetrahydro-1-isopropyl-1, 2, 4, 4, 7-pentamethyl-with ΔG of -7.8 Kcal/mol. Penta-noic acid, octyl ester; hexadecane; 2-methyl-5-(1- methylethenyl)-2-cyclohexen-1-ol; diethy-leneglycol dimethacrylate; naphthalene; 9-octadecenoic acid (Z)-, methyl ester; and naphthalene, decahydro-2,3-dimethyl-; showed the highest affinity to DNMT3B with a ΔG of -7.5, -7.4, -7.4, -7.2, -7.2, -7.1, and -7.1 Kcal/mol, respectively. The intermolecular interactions of these compounds are depicted in Figure 4A-X. Relative peptide expression levels (mean ± SD). A. MMP-2, B. COX-2, C. HDAC1, D. DNMT1, E. DNMT3A, and F. DNMT3B expressions and G. Sodium dodecyl sulfate–polyacrylamide gel electrophoresis (SDS-PAGE) of analyzed peptides in endometrial mesenchymal stromal/stem cells (EnMSCs) derived from endometrium of endometrioma patients (OMA-EnMSCs) and deep infiltrative endometriosis (DIE) samples of women with endometriosis associated infertility (E-EnMSCs), and EnMSCs derived from endometrium of an endometriosis-free, normal woman (C-EnMSCs) at different Ceratonia siliqua L . pod extract (CPE) concentrations. Binding affinity of ligands in Ceratonia siliqua L . pod extract (CPE) to receptors DNMT; DNA methyltransferase, HDAC1; Histone deacetylase 1, COX-2; Cyclooxygenase-2, and MMP2; Matrix metalloproteinase-2. The 3D plot of the binding sites and 2D plot of interactions of DNA methyltransferase 3A ( DNMT3A ), cyclooxygenase-2 (COX-2), histone deacetylase 1 ( HDAC1 ), matrix metalloproteinase-2 ( MMP2 ), DMNT1 , and DNMT3B with different molecules of the Ceratonia siliqua L . pod extract (CPE). Interaction of DNMT3A with: A. 4-imidazolidinone, 1-benzoyl-2-(1-methylethyl)-3-methyl, B. Benzene, 1,3,5-tris(1-methylpropyl)-, C. 3, 4-dihydro2(1H)-isoquinoline carboximidamide, D. Naphthalene,1,2,3,4-tetrahydro-1-isopropyl-1, 2, 4, 4, 7-pentamethyl-, E. naphthalene, decahydro-2, 3-dimethyl-, F. Benzoic acid, 4-heptyl-, 4-cyanophenyl ester, and G. Methanone,(1-hydroxycyclohexyl)phenyl. Interaction of COX-2 with: H. Diphenyl sulfone; I) 3,4-dihydro-2(1H)-isoquinoline carboximidamide, J. Methanone,(1-hydroxycyclohexyl)phenyl, and K. 4-imidazolidinone, 1-benzoyl-2-(1-methylethyl)- 3-methyl. Interaction of HDAC1 with: L. Benzoic acid, 4-heptyl-,4-cyanophenyl ester. Interaction of MMP2 with: M. Diphenyl sulfone, N. Methanone, (1-hydroxycyclohexyl) phenyl-, O. 3,4-dihydro-2(1H)-isoquinoline carboximidamide, and P. 4-imidazolidinone,1-benzoyl-2-(1-methylethyl)-3-methyl. Interaction of DMNT1 with: Q. Naphthalene, 1,2,3,4-tetrahydro-1-isopropyl-1, 2, 4, 4, 7-pentamethyl. Interaction of DNMT3B with: R. Pentanoic acid, octyl ester, S. Hexadecane, T. 2-methyl-5-(1-methylethenyl)-2-cyclohexen-1-ol, U. Diethylene glycol dimethacrylate, V. Naphthalene, and W. 9-octadecenoic acid (Z)-, methyl ester, and X. Naphthalene, decahydro-2, 3-dimethyl. Diphenyl sulfone, 3,4-dihydro-2(1H)-isoquinoline carboximidamide, methanone (1-hydroxycyclohexyl)phenyl, and 4-imidazolidinone,1-benzoyl-2-(1- methylethyl)-3-methyl exhibited the highest affinity to COX-2 with ΔG values of -7.4, -7.3, -7.2, and -7.1 Kcal/mol, respectively. 4-Imidazolidinone,1-benzoyl-2- (1-methylethyl)-3-methyl; benzene,1,3,5-tris (1-methylpropyl)-; 3.4-dihydro-2 (1H)-isoquinoline carboximidamide; naphtha-lene,1, 2, 3, 4-tetrahydro-1-isopropyl-1, 2, 4, 4, 7-pentamethyl-; naphthalene,decahydro-2, 3-dimethyl-; benzoic acid, 4-heptyl-, 4-cyanophenyl ester; and methanone, (1-hydroxycyclohexyl) phenyl showed the highest affinity to DNMT3A with ΔG values of -5.8, -5.6, -5.5, -5.4, -5.4, -5.4, and -5.4 Kcal/ mol. Diphenyl sulfone; methanone, (1-hydroxycyclohexyl)phenyl-; 3.4-dihydro-2(1H)-isoquinoline carboximidamide; and 4-imidazolidinone,1-benzoyl-2-(1- methylethyl)-3-methyl exhibited the highest affinity to MMP2 with ΔG values of -6.8, -6.7, -6.6, and -6.6 Kcal/ mol. The best affinity to HDAC1 was observed for benzoic acid, 4-heptyl-,4-cyanophenyl ester with ΔG of -6 Kcal/mol. The highest affinity to DNMT1 was related to naphthalene,1,2,3,4-tetrahydro-1-isopropyl-1,2,4,4,7- pentamethyl-with ΔG of -7.8 Kcal/mol. Penta-noic acid, octyl ester; hexadecane; 2-methyl-5-(1-methylethenyl)- 2-cyclohexen-1-ol; diethy-leneglycol dimethacrylate; naphthalene; 9-octadecenoic acid (Z)-, methyl ester; and naphtha-lene, decahydro-2,3-dimethyl-; showed the highest affinity to DNMT3B with a ΔG of -7.5, -7.4, -7.4, -7.2, -7.2, -7.1, and -7.1 Kcal/mol, respectively. The intermolecular interactions of these compounds are depicted in Figure 4A-X.

GC-MS analysis of the CPE identified 60 bioactive compounds with various properties, some of which had anticancer (4 compounds), cytotoxic (4 compounds), and anti-inflammatory (6 compounds) effects. The focus of this study was to evaluate the compounds that had anti-cancer and anti-inflammatory properties. Previous studies have demonstrated the potent biological anticancer effects of octacosane, 9-octadecenoic acid (Z)-, methyl ester, phenol, 2,4-bis( 1 , 1 -dimethylethyl)-, and naphthalene ( 25 - 27 ). Additionally, the phytochemical profile of the CPE showed high levels of phenolic content. Overall, both GC-MS and phytochemical analyses suggest that CPE has potential anticancer effects. This finding is consistent with previous studies that reported high amounts of phenolic content and anticancer properties in CPEs ( 28 , 29 ). Furthermore, the CPE also exhibited anti-inflammatory effects in six of the identified compounds, which is consistent with previous studies on CPEs that reported antiinflammatory properties. The antioxidant, cell cytotoxic, and antimicrobial properties of CPE have also been reported in previous studies ( 12 , 19 ). The ideal medical treatment for endometriosis has yet to be identified, and current medications often have relatively short-term effects. Additionally, hormonal drugs have many undesirable side effects and can induce an amenorrhoeic state by creating a hypo-oestrogenic environment ( 30 ). Therefore, finding novel treatment agents is necessary to improve fertility and other complications of endometriosis. The present study evaluated the impact of CPE on genes related to invasion and attachment ( MMP2 ), inflammation (COX-2), and epigenetic mechanisms (DNMTs and HDAC1) involved in endometriosis. Uncontrolled regeneration of free radicals is a critical contributor to aging because it attacks various biomolecules, membranes, lipids, proteins, enzymes, and DNA, and leads to oxidative stress and cell death. In cases of infertility, overproduction of free radicals and oxidative damage can lead to compromised follicles in endometriosis-associated fecundity impairment ( 31 ). Antioxidants are stable components that donate an electron or hydrogen to counteract the effects of free radicals and terminate the chain reaction before molecules are damaged. Also, the radical scavenging property of antioxidants delays or inhibits cellular damage. Phenolic compounds are regarded as powerful molecules with increased potential to neutralise free radicals ( 32 ). In this study, TPC analysis confirmed the presence of phenolic antioxidants in Ceratonia siliqua , which may be responsible for its therapeutic effects. Also, the hydroalcoholic CPE was screened for its in vitro antioxidant activity and the results showed promising antioxidant potency. Phytochemical analysis suggests natural antioxidants are a promising novel therapy to reduce the EAI rate ( 31 ). The biological activities and pharmacological properties of Ceratonia siliqua may be linked to its high phenolic content, especially gallic acid, catechin, epicatechin gallate, epigallocatechin gallate, and quercetin glycosides ( 12 ). Biometal cofactors such as Cu, Fe, Mn and Zn, which are essential for antioxidant enzymes, are found in Ceratonia siliqua ( 33 ). In the present study, total RNA was extracted from EEnMSCs, OMA-EnMSCs, and C-EnMSCs after exposure to CPE. cDNA synthesis and quantitative-PCR, gene expression levels were calculated. The gene expression levels at CPE concentrations of 0.8 μg/mL and 2 μg/mL were compared to those of the C-EnMSCs. Our data demonstrated that treatment of E-EnMSCs with CPE suppressed COX-2 expression. Previous studies have suggested that COX-2 overexpression plays a crucial role in endometriosis-associated pain and in the pathogenesis and development of this disease through marked enhancement of inflammatory mediators ( 6 ). Moreover, the protein expressions of EEnMSCs, OMA-EnMSCs, and C-EnMSCs were evaluated after treatment with 0.8 μg/mL and 2 μg/mL of CPE. There was a decrease in COX-2 protein expression after treatment with both concentrations compared to the C-EnMSCs. Endometriosis is an inflammatory disease that is associated with various alterations in immune cells and inflammatory cytokines. This subclinical inflammation is mainly responsible for the pain and infertility associated with this disease ( 34 ). Flavonoids and tannins, which are the main phenolic compounds found in Ceratonia siliqua , have anti-inflammatory activities because they inhibit the expressions of inflammatory mediators such as cytokines, inducible nitric oxide synthase, and COX-2 ( 35 ). The elevated expression of COX-2 in both endometrial and endometriotic cells of women with endometriosis plays a critical role in the pathogenesis of pain and infertility ( 36 ), and down-regulation of COX-2 by CPE may reduce dis-ease-associated pain and improve fertility outcomes. Further in vivo studies are required to investigate the clinical applications of CPE in endometriosis. A growing body of evidence has suggested that endometriosis is an epigenetic disorder ( 36 ). Two predominant epigenetic mechanisms are DNA methylation, which occurs through DNMTs , and histone modification by HDACs ( 37 ). Consistent with previous studies ( 38 ), we observed an increased level of DNMT3A expression in the E-EnMSCs from women with endometriosis compared to those in the C-EnMSCs. However, the role of other DNMTs in the pathogenesis of endometriosis remains controversial ( 38 ). The results of the current study reveal that DNMT3B gene expression levels can be significantly reduced after treatment with 0.8 μg/mL of CPE. However, HDAC1 expression only reduced in E-EnMSCs exposed to 2 μg/mL of CPE; no significant efficacy of CPE was observed in terms of reducing HDAC1 expression in OMA-EnMSCs. Western blot analysis showed that the DNMT3A and -3B pro-tein expressions decreased in all three cell types after treatment with 0.8 and 2 μg/mL of CPE. Additionally, HDAC1 protein expression also showed a decline after treatment at both concentrations. In contrast with gene expression, a reduction in protein expression of DNMT3B and HDAC1 was higher after exposure to 2 μg/mL of CPE than 0.8 μg/mL. In this regard, HDAC inhibitors have been suggested as attractive treatment targets for endometriosis. In vitro and in vivo studies of human endometrial stromal and epithelial cells and animal models of endometriosis, respectively, that were treated with different HDAC inhibitors have shown anti-proliferative activities and reduced impacts on lesion size ( 39 ). MMP2 acts as a key enzyme associated with tumour metastasis through the degradation of extracellular matrix components ( 7 ). MMPs are essential enzymes in the ectopic implantation of endometrial tissue, and our study revealed that the MMP2 gene expression level in OMAEnMSCs could be modulated by 0.8 μg/mL of CPE. Moreover, the protein expression of MMP2 was also decreased in OMA-EnMSCs. However, this downregulation of MMP2 protein expression was not observed in CEnMSCs and E-EnMSCs. The effectiveness of Ceratonia siliqua in the current study was based on in vitro treatment of endometriosis. Therefore, well-designed animal and human trials are recommended to achieve conclusive re-sults on the efficacy and safety of this natural product for endometriosis. Two noteworthy limitations of our study were the limited sample size and lack of budget to carry out further in vivo evaluations. Furthermore, while the antioxidative and antiinflammatory effects of Ceratonia siliqua have been evaluated, further biochemical studies are necessary to assess the parameters that induced these effects in our study. A computational study showed that not only one specific compound had the highest affinity to different proteins, which made it impossible to identify the most effective specific compound. The combination of several compounds can be involved in the inhibition of proteins. This difference can be attributed to the role of conformation and intermolecular interactions in the formation of the ligand-protein complex ( 40 ). According to the docking scores, 14 of the 60 studied compounds showed good affinity to different proteins. However, among them, three compounds, 3,4-dihydro-2(1H)-isoquinoline carboximidamide, 4-imidazolidinone,1-benzoyl-2-(1-methylethyl)- 3-methyl, and methanone (1-hydroxycyclohexyl)phenyl, showed good affinity to three proteins. Hydrogen bonds, van der Waals interactions, and hydrophobic interactions play a role in the binding of compounds to proteins. Therefore, it can be inferred that the inhibitory effect of CPE is mostly related to these compounds. We briefly touched upon the limitations of this study; however, further elaboration could provide a clearer picture of potential areas for improvement. More detailed discussion on the specific pathways and potential interactions involved in the mechanism of action of Ceratonia siliqua could provide more insight into its therapeutic effects. Further exploration of potential clinical implications that include dosage regimens and considerations for patient populations could be beneficial. Finally, a comprehensive discussion on the current state of endometriosis treatment and how Ceratonia siliqua compares to other treatments may provide more context for its potential impact.

We demonstrated that the CPE could reduce the viability and proliferation of OMA-EnMSCs and C-EnMSCs, as well as decrease the expressions of key enzymes and epigenetic factors involved in endometriosis pathogenesis, including MMP2, DNMT3A, DNMT3B , and HDAC1 . Our study also identified several compounds in the CPE that showed good affinity to different proteins involved in endometriosis pathogenesis, which suggests that a combination of these compounds could be responsible for the inhibitory effects of CPE.

In this experimental study, EnMSCs were obtained from the endometrium, endometrioma, and deep infiltrative endometriosis (DIE) samples of 10 women diagnosed with EAI and compared to EnMSCs derived from the endometrium of one endometriosis-free normal woman as the control group (C-EnMSCs). DIE samples were obtained from the rectovaginal septum and deep retroperitoneal portions of the women with EAI, and ectopic endometrial cells were derived from their ovarian endometrioma (OMA). These 10 women were candidates for laparoscopic endometriosis surgery who referred to the Laparoscopy Clinic of the Obstetrics and Gynaecology Department at Shiraz University of Medical Sciences (Shiraz, Iran) from 2018 to 2019. Women who received gonadotropin-releasing hormone (GnRH) agonist, progesterone, or other hormones three months before the study were excluded. All included cases had surgical and histological evidence of stage III or IV endometriosis according to the American Society of Reproductive Medicine classification ( 14 ). Endometrial samples were obtained by dilatation and curettage from the endometrial cavity as ectopic endometrial cells. Samples of eutopic endometrial cells were taken from OMA and DIE lesions by laparoscopic intervention. Eutopic and ectopic endometrial samples were taken by the laparoscopic surgical team of Shiraz University of Medical Sciences. Eutopic endometrial cells were also obtained from a normal, endometriosis-free woman who was scheduled for benign gynaecologic surgery by dilatation and curettage to serve as the C-EnMSCs. The cellular samples were classified into three groups ( Table 1 ). Group 1 consisted of eutopic endometrial cells derived from the EAI women (E-EnMSCs); group 2 included ectopic endometrial cells derived from OMA and DIE of the EAI women (OMA-EnMSCs.); and group 3 included eutopic endometrial cells taken from a normal, endometriosis-free woman (C-EnMSCs.). Grouping of different samples of endometrial mesenchymal stromal/stem cells (EnMSCs) for evaluation of different Ceratonia siliqua L . pod extract (CPE) concentrations CPE was purchased from an herbal market in Shiraz, Iran. The plant material was authenticated by a botanist and assigned voucher specimen number (PM-1292), and deposited in the herbarium of the Phytopharmaceutical Department, Shiraz University of Medical Sciences, Shiraz, Iran. The CPE was prepared as follows: Ceratonia siliqua pods were ground into a powder and sieved through a 50 µm mesh. Then, 100 g of the prepared CPE was added to 500 mL of 70% ethanol and the suspension was shaken at room temperature for 24 hours. After filtration, the supernatant, at a concentration of 200 000 μg/mL, was stored at -20°C for the next steps of the study. Gregoriou et al. ( 15 ) showed that the total phenolic content (TPC) varies according to the maturity of the plant. The total phenolic content (TPC) of the CPE was evaluated by the Folin-Ciocalteu method ( 16 ). Briefly, a 5 μL aliquot of the CPE (at different concentrations in methanol) was mixed with 158 μL of dH 2 O followed by the addition of 10 μL of Folin-Ciocalteu reagent. Then, 30 μL of 15% sodium carbonate was added. The mixture was incubated at room temperature for 45 minutes in the dark and the absorbance was measured at 765 nm using a microplate reader (POLARstar Omega, BMG Labtech, Germany). A standard curve was generated based on a prepared standard gallic acid solution (0, 10, 50, 100, 150, 250, and 500 mg/mL), which revealed a linear regression of R2 >0.99. The phenolic level was expressed as milligrams of gallic acid equivalents (GAE) per gram of dry plant extract (mg GAE/g dry extract). Each sample was assessed in triplicate ( 17 ). The CPE was lyophilised and analysed with an Agilent 7890B Gas chromatography-mass spectroscopy (GC-MS) system. Electron ionisation mass spectra were obtained with a scan range of m/z 50-500 using electrons with an energy of 70 eV and a filament emission of 0.5 mA. Separation was carried out using an HP-5MS UI column with dimensions of 30 m×0.25 mm i.d. and a film thickness of 0.5 μm. Helium was used as a carrier gas with a flow rate of 0.8 mL/minutes. The GC oven temperature was increased at a rate of 5°C/minutes from 80°C after 3 minutes of sample injection and held at 250°C for 10 minutes. The transfer line, ion source, and injection port of the gas chromatograph of 5977MSD were maintained at 250°C, 220°C and 240°C, respectively. The separated compounds were identified by comparing them with the compound data in the National Institute of Standards and Technology (NIST MS database) library ( 18 ). The in vitro antioxidant activity of the CPE was measured based on the decolourisation of the 2,2-Diphenyl1-picrylhydrazyl (DPPH) as described previously ( 19 ). In brief, 20 μL of different concentrations of the crude CPE or quercetin, as the positive control, were added to 180 μL of 110 µM DPPH radical dissolved in methanol. The reagent was then incubated in the dark for 30 minutes. Methanol was used as a blank solution. The pale-yellow colour of the reduced DPPH radical was measured at an optical density of 517 nm at room temperature using a microplate reader (POLARstar Omega, BMG Labtech, Germany). The volume of an-tioxidant required to produce a 50% reduction of the DPPH was recorded as the half maximal inhibitory concentration (IC 50 ) value. A lower IC 50 value indicated a stronger DPPH scavenging capacity ( 20 ). Proliferation capacity and cell viability of the E-EnMSCs, OMA-EnMSCs, and C-EnMSCs were determined using the MTT assay (Sigma-Aldrich, USA). Briefly, cells were seeded in 96-well plates at a density of 5000 cells/well. After 24 hours of incubation, different concentrations of CPE (0- 20 000 μg/mL) were added to the wells and incubated for one, three, and seven days. At the end of each incubation period, MTT solution (5 mg/mL) was added to the medium and the cells were further incubated for another four hours at 37°C. The MTT solution was then replaced with 200 μL of dimethyl sulfoxide for 30 minutes. The absorbance was measured at 570 nm using an ELISA plate reader (Bio‐Rad, Hercules, CA, USA). The cell viability percentage was calculated using the following formula: (%)=[100×(sample absorbance)/(control absorbance)]. Each test was repeated three times. The means of the data were calculated, and CPE concentrations of 0.8 and 2 μg/mL were selected based on the viability results and compared to the C-EnMSCs ( 21 ). The EnMSCs were isolated according to modifications to previously described methods ( 22 ). Briefly, tissues were minced into small pieces of about 1-2 mm 3 and digested with 1 mg/mL of collagenase type 1 (Cat. no. 17100-017, Gibco, USA) in Dulbecco’s modified Eagle medium (DMEM, Biovet, Bulgaria) that contained 10% foetal bovine serum (FBS, Biovet, Bulgaria) for one hour with constant stirring. The resultant cell suspension was passed through a 40 µm nylon sieve to remove undigested aggregates. Red blood cells were removed using Ficoll (Ref. 002041600, Gibco, USA). The cell suspension was then seeded in T25 culture flasks and incubated in DMEM/F12 medium (Sigma‐Aldrich, UK) that contained 10% FBS, 100 μg/mL penicillin, and 100 U/mL streptomycin (Sigma, USA) until the cells reached 80% confluency. These cells were used for the subsequent experiments. Interestingly, the DIE cells did not show any growth and proliferation in the culture medium, and no stem cells could be obtained from those samples. However, the ESC isolation procedure was successfully performed on eutopic endometrial cells from the EAI, normal control women, and ectopic OMA. Next, we assessed cell purity in the three groups (E-EnMSCs, OMA-EnMSCs, and C-EnMSCs). Flow cytometry was used to categorise the passage-3 cells based on their expressions of cell surface markers. The resultant cell suspension was washed in blocking solution, cold phosphate-buffered saline (PBS) that contained 10% FBS, for 20 minutes. Then, the cells were labelled with FITC-conjugated anti-CD45, anti-CD90, anti-CD73, and an-ti-CD31 antibodies (all from Abcam, Cambridge, UK). Subsequently, the cells were washed twice and resuspended in cold PBS. The percentage of positive or negative cells was evaluated using a calibrated FACS device (FACSCalibur™, BD Biosciences) and analysed by FlowJo software (BD Biosciences). Next, we evaluated the multipotency of the isolated cells by differentiating them into two mesenchymal lineagesosteocytes and adipocytes. Briefly, 1×10 4 passage-3 cells/ cm 2 were cultured in 24-well plates. Once the cells reached approximately 80% confluency, the growth medium was changed to osteogenic medium (DMEM-LG supplemented with 10% FBS, 2 mM L-glutamine, 100 U/mL penicillin, 100 μg/mL streptomycin, 100 nM dexamethasone, 0.2 mM L-ascorbate, and 10 mM β-glycerophosphate) and adipogenic medium (DMEM-LG supplemented with 10% FBS, 2 mM L-glutamine, 100 U/mL penicillin, 100 μg/mL streptomycin, 60 μM indomethacin, 1 μM dexamethasone, 0.5 mM IBMX, and 5 μg/mL insulin solution). The osteogenic and adipogenic cells were incubated for 28 and 21 days, respectively, with half of the medium replaced every three days. The differentiated cell lines were fixed in 4% paraformaldehyde and stained with Alizarin Red S and Oil Red O to observe differentiated osteoblasts and adipocytes, respectively ( 23 ). Total RNA was extracted from the E-EnMSCs, OMAEnMSCs, and C-EnMSCs after seven days of exposure to CPE using a FavorPrep™ Blood/Cultured Cell Total RNA kit (Favorgen, Ping-Tung, Taiwan) according to the manufacturer’s instructions. The quantity and quality of the obtained RNA were measured using a Nanodrop™ spectrophotometer (Thermo Fisher Scientific, Wilmington, DE, USA) based on a 260/280 nm optical density ratio, and then stored at -80˚C until cDNA synthesis. cDNA synthesis was performed using a RevertAid™ First Strand cDNA Synthesis kit (Thermo Fisher Scientific, Inc., Waltham, MA, USA) according to the manufacturer’s protocols. The synthesized cDNA was used as a template for PCR amplification. The expression levels of MMP2, COX-2, HDAC1, DNMT1, DNMT3A , and DNMT3B were quantified by realtime PCR, which was performed using an ABI Biosystems StepOne and RealQ Plus 2x Master Mix Green (Ampliqon A/S, Odense, Denmark). Primers were designed based on human DNA sequences obtained from the gene bank Primer-Blast online program ( Table S1 , See Supplementary Online Information at www.ijfs.ir ). The housekeeping gene TATA-binding protein (TBP) was used as a reference to normalise the expression values. The following quantitativePCR conditions were used: 10 minutes at 94˚C, followed by 40 cycles of 15 seconds at 94˚C, 60 seconds at 60˚C, and finally a melting curve stage was performed to determine the specificity of the product. Additionally, to confirm the results, real-time PCR products were analysed on 2% ultrapure agarose gel electrophoresis (Sigma, USA) according to the manufacturer’s instructions ( 24 ). Based on the results of the MTT cell viability and proliferation assay, the most effective concentrations of the CPE (0.8 and 2 μg/mL) were selected and used for Western blot analysis. Specifically, the E-EnMSCs, OMA-EnMSCs, and C-EnMSCs were treated with CPE; after 72 hours, the cells were lysed with RIPA buffer that consisted of 50 mM trisHCl (pH=8.0), 0.4% Nonidet P-40, 120 mM NaCl, 1.5 mM MgCl2, 2 mM phenylmethylsulfonyl fluoride, 80 μg/mL leupeptin, 3 mM NaF, and 1 mM DTT, at 4°C for 20 minutes. The lysed components were then centrifuged at 12 000 ×g for 20 minutes at 4°C. The protein content of the resultant solution was measured by the Bradford protein assay. The proteins were then transferred to microporous polyvinylidene difluoride membranes (Millipore, France), which were incubated with 5% BSA (Sigma, USA) as a blocking buffer for one hour at room temperature. Subsequently, the membranes were incubated overnight at 4°C with primary antibodies. Samples were immunoblotted (1:200) with anti-beta-actin, anti-MMP2, anti-COX-2, anti-HDAC1, anti-DNMT1, antiDNMT3A, and anti-DNMT3B antibodies (Cell Signaling Technology, Danvers, MA, USA). The membranes were washed three times (10 minutes each) with Tween buffer, and then incubated with horseradish peroxidase (HRP)- conjugated goat anti-mouse or rabbit secondary antibodies. After four washes, the excess antibodies were removed from the membranes, and HRP activities were detected using ECL Plus Chemiluminescence Reagent (Amersham, Chalfont, UK) according to the manufacturer’s protocol. We selected 62 compounds detected in the CPE and six proteins for the docking process. The 3D structure of the ligands and receptors were obtained from the PubChem databases of the Protein Data Bank (PDB). The downloaded receptors included COX-2, MMP2, DNMT3B, DNMT1, DNMT3A , and HDAC1 with PDB codes 3nt1, 3ayu, 3flg, 3os5, 4qbs, and 4bkx. The geometry of the ligands was optimised using HyperChem software version 8.0.10. The receptors were prepared using Chimera 1.15 by removing all non-standard residues, water, and original hydrogens, and by adding polar hydrogen, charges, and bond orders. Lastly, the outputs were suitably formatted for the docking process. The generation of a grid box is a crucial step in the docking process. Grid boxes with spacing of 0.375 Å were generated using AutoGrid at the active site of each receptor selected based on CASTp analysis. AutoDock Vina 1.1.2 was used to investigate the interactions between the target receptors and selected ligands. After completing the docking process, the conformation with the lowest binding affinity and RMSD ≤2 Å was selected as the best. The following critical parameters were considered in the sample size calculation. Standard deviation (σ): The standard deviation reflects the variability within each treatment group. We obtained estimates of the standard deviation for each specific gene expression of interest from preliminary experiments. Desired effect size (δ): The effect size represents the minimum difference in gene expression that we considered to have biological and clinical significance. It was determined based on prior research findings and expected biological significance. Level of confidence (α): We selected a 95% confidence level, which corresponded to a sig-nificance level (α) of 0.05, and indicated our willingness to accept a 5% chance of a type I error. Power (1-β): We aimed for a statistical power (1-β) of 80% or higher, which represented our ability to correctly detect true effects with a type II error rate of 20% or less. The sample size for each treatment group was calculated using the following formula: n = (2(Z α/2 +Z β ) 2 ×(σ 2 ))/(δ 2 ) Where: n=Required sample size per treatment group. Z α/2 =Z-score for the chosen level of confidence (e.g., 1.96 for a 95% confidence level). Zβ =Z-score for the desired statistical power (e.g., 0.84 for an 80% power). σ=Estimated standard deviation. δ=Desired effect size. Final sample size: We calculated the required sample size for each treatment group after plug-ging in the specific values for each gene of interest and using the above formula. All statistical analyses were performed using IBM SPSS Statistics 26 software (SPSS Inc., Chicago, IL, USA). Results are presented as mean ± standard deviation. Differences between groups were evaluated by one-way ANOVA followed by the Duncan test. In order to ensure the validity of our one-way ANOVA analyses, we rigorously assessed the key assumptions, including independence of observations, normality of residuals, homogeneity of variances, and independence of groups, as detailed in the Assumption Checks subsection. P<0.05 indicated statistical significance. Graphs were prepared using GraphPad Prism software (v7.0a, GraphPad Software, Inc., San Diego, CA, USA). The study was approved by the Institutional Review Board and the Ethics Committee of Shiraz University of Medical Sciences, Shiraz, Iran (IR.SUMS.REC.1397.555). Informed consent was obtained from all donors after a thorough explanation of the study design prior to sample collection.