Suppression of endometriosis by miRNA-34a via inhibition of matrix metalloproteinase-2: An alternative pathway to impede invasion

Yasmin Begum: Writing – review & editing, Writing – original draft, Visualization, Validation, Methodology, Investigation, Formal analysis, Data curation, Conceptualization. Anuradha Pandit: Writing – original draft, Visualization, Validation, Methodology, Investigation, Formal analysis, Conceptualization. Devendra Shukla: Writing – original draft, Validation, Software, Methodology. Rahul Gupta: Writing – review & editing, Validation. Pramathes DasMahapatra: Resources. Amit Kumar Srivastava: Writing – review & editing, Supervision, Resources, Conceptualization. Snehasikta Swarnakar: Writing – review & editing, Supervision, Resources, Conceptualization.

The Human Ethics Committees of the Indian Institute of Chemical Biology (approval number: IICB/IRB/SS02-2015) and the Spectrum Clinic & Endoscopy Research Institute (approval number: SCERI-IEC015) approved the research protocol.

This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.

All the patients who participated voluntarily have given written consent.

The CancerMIRNome analysis revealed a decreased miR-34a-5p expression in distinctive gynecological malignancies ( Fig. 1 A). Therefore, its downregulation in ovarian cancer was anticipated as well. Additionally, the GEOdataset GSE239685 re-analysis indicated a remarkable decrease in miR-34a-5p expression in ovarian clear cell cancer. The findings suggested markedly reduced miR-34a-5p levels in ovarian clear cell carcinoma (n = 15) in comparison to normal control (n = 4) ( Fig. 1 B). Since endometriosis can also lead to ovarian cancer, therefore, its downregulation in endometriosis was predicted as well. Fig. 1 Gene expression profile of miR-34a and MMP-2. (A) A computational study using CancerMIRNome analysis for the expression of miR-34a-5p in distinctive cancer revealed the downregulation of miR-34a-5p in gynecological malignancies. (B) GEOdataset GSE239685 re-analysis showing the top 50 downregulated miRNAs in ovarian cancer among which miR-34a is one of the highly downregulated miRNAs. (C, D) qPCR analysis also revealed a significant decrease of miR-34a (C) and increased MMP-2 (D) explaining their inverse correlation. (E) The probable site of miR-34a-5p binding on the MMP-2 3′UTR region. (F) The dual luciferase assay revealed a strong inhibition in the activity of renilla luciferase upon the treatment with miR-34a-5p mimic. Each experiment has been carried out thrice. The data is depicted as mean ± SEM. A P-value of <0.05 and  0.05) is represented as ∗, ∗∗, and ns respectively. Fig. 1 Gene expression profile of miR-34a and MMP-2. (A) A computational study using CancerMIRNome analysis for the expression of miR-34a-5p in distinctive cancer revealed the downregulation of miR-34a-5p in gynecological malignancies. (B) GEOdataset GSE239685 re-analysis showing the top 50 downregulated miRNAs in ovarian cancer among which miR-34a is one of the highly downregulated miRNAs. (C, D) qPCR analysis also revealed a significant decrease of miR-34a (C) and increased MMP-2 (D) explaining their inverse correlation. (E) The probable site of miR-34a-5p binding on the MMP-2 3′UTR region. (F) The dual luciferase assay revealed a strong inhibition in the activity of renilla luciferase upon the treatment with miR-34a-5p mimic. Each experiment has been carried out thrice. The data is depicted as mean ± SEM. A P-value of <0.05 and  0.05) is represented as ∗, ∗∗, and ns respectively. The miR-34a and MMP-2 gene expression levels were assessed using RNU6B and 18S rRNA as endogenous controls, respectively, through quantitative PCR analysis. Human endometriotic tissue samples from 22 patients were used, 11 control and 11 endometriosis, for the analysis ( Table 1 ). The mean cycle threshold value (Ct) was determined for each sample. The mean Ct values of miR-34a and MMP-2 were then normalized to RNU6B and 18S rRNA respectively (PROTOCOL DOI). The delta cycle threshold value (ΔCt) was calculated for each sample to compare the control with the diseased samples. The miR-34a gene expression level is noticed to be substantially decreased in the patient tissues compared to the control group ( Fig. 1 C). Conversely, the gene expression of MMP-2 was markedly elevated in the patient samples compared to the control ( Fig. 1 D). These findings suggest that the correlation between the gene expressions of miR-34a and MMP-2 is inverse. Table 1 Demography of the study sample for control and endometriosis patients. The table is showing the number of samples taken for the study, their age, gender, infertility status, the origin of the tissue, exclusion, and inclusion criteria. Table 1 Control Endometriosis Case numbers N = 11 N = 11 Age (Mean ± SD) 33.09 ± 5.70 32.18 ± 6.615 Gender Female Female Infertile 0/11 8/11 Origin Eutopic endometrial tissue Ectopic endometrial tissue Exclusion criteria Endometriosis, or other gynecological or oncological illness and without any hormonal therapy for at least 3 months Any other gynecological or oncological illness except endometriosis and without any hormonal therapy for at least 3 months Inclusion criteria Pain, Recurrent Pregnancy Loss, Amenorrhea, Dysmenorrhea Endometriosis, Pelvic pain, Amenorrhea, Dysmenorrhea Demography of the study sample for control and endometriosis patients. The table is showing the number of samples taken for the study, their age, gender, infertility status, the origin of the tissue, exclusion, and inclusion criteria. From literature studies a miR-34a-5p's binding site is found in the MMP-2 promoter's 3′UTR region ( Fig. 1 E) [ 26 ]. The in vitro validation of miR-34a binding with the MMP-2 gene was done by the co-transfection of promoter-reporter plasmids carrying MMP-2-3′UTR with either the mimic of miR-34a-5p alone or both the mimic and its inhibitor. The mimic and the MMP-2 promoter with the constructs co-transfected group exhibited decreased MMP-2 activity compared to the inhibitor co-transfected group. Furthermore, the dual luciferase assay revealed a strong inhibition in the activity of renilla luciferase in the miR-34a-5p mimic's presence (∼1.8-fold, p < 0.005), which is reversed by the inhibitor (∼2-fold, p < 0.005) indicating a direct targeting of 3′UTR of MMP-2 by miR-34a-5p ( Fig. 1 F). To reassert miR-34a's functional implications on MMPs, gelatin-containing SDS gel was implicated in Zymography. Both the activity of MMP-2 and MMP-9 levels were observed to drop in End1/E6E7 cells (∼2–3 fold, p < 0.0001) and SK-OV-3 cells (∼1.5-fold, p < 0.01) upon treatment with miR-34a mimic, leading to a fainter band, which were then increased upon treatment with miR-34-5p inhibitor ( Fig. 2 A–D). Immunoblotting was performed in the miR-34a mimic and inhibitor-transfected SK-OV-3 cells. The protein expression was normalized using β-actin as a normalization control. MMP-2 expression levels are found to be considerably low (∼3-fold, p < 0.005) in the mimic group of miR-34a, and further accompanied by the incorporation of its inhibitor, MMP-2 level increases significantly ( Fig. 2 E and F). These results show the miR-34a tumor suppressor activity of the mir-34a mimic, which suppressed the MMP-2/9 activities. This effect was reverted to negative control levels of MMP after miR-34a-5p inhibitor treatment. Fig. 2 Activity and expression profile of MMPs. (A–D) Zymographic analysis showing the MMP-2/-9 activities in End1/E6E7 and SK-OV-3 cells. Treatment with miR-34a reduces both MMP-2 and MMP-9 activities while enhancing when miR-34a inhibitor is present in both (A) End1/E6E7 cells and (C) SK-OV-3 cells, and (B, D) showing quantitative analysis of MMP-2 activity of End1/E6E7 and SK-OV-3 cells respectively. (E, F) Immunoblotting of MMP-2 shows a lowered MMP-2 expression in the mimic group which reversed back after the incorporation of miR-34a-5p inhibitor in SK-OV-3 cells. Each experiment has been carried out thrice. The data is depicted as mean ± SEM. ∗∗, ∗∗∗, and ∗∗∗∗ /#### are representations of P-values <0.01, <0.001, and <0.0001 respectively. Fig. 2 Activity and expression profile of MMPs. (A–D) Zymographic analysis showing the MMP-2/-9 activities in End1/E6E7 and SK-OV-3 cells. Treatment with miR-34a reduces both MMP-2 and MMP-9 activities while enhancing when miR-34a inhibitor is present in both (A) End1/E6E7 cells and (C) SK-OV-3 cells, and (B, D) showing quantitative analysis of MMP-2 activity of End1/E6E7 and SK-OV-3 cells respectively. (E, F) Immunoblotting of MMP-2 shows a lowered MMP-2 expression in the mimic group which reversed back after the incorporation of miR-34a-5p inhibitor in SK-OV-3 cells. Each experiment has been carried out thrice. The data is depicted as mean ± SEM. ∗∗, ∗∗∗, and ∗∗∗∗ /#### are representations of P-values <0.01, <0.001, and <0.0001 respectively. To understand if miR-34a has any functions on the invasive potential of End1/E6E7 and SK-OV-3 cells, a Transwell assay for invasion was accomplished on cultured cells. Cells treated with mimic showed decreased invasion in comparison to the negative control, and a significantly greater number of invaded cells (∼5.5-fold in End1/E6E7, p < 0.005, and ∼9-fold in SK-OV-3, p < 0.0001) were found crossing the matrix to the lower side of the chambers in the inhibitor transfected category ( Fig. 3 A and B). This observation points to the miR-34a's impact in inhibiting the invasive property of cells. Fig. 3 Impact of miR-34a on invasion and migration of the cells. (A–D) The cells show a marked increase in invasive (A, B) as well as migratory (C, D) potential upon transfection with miR-34a inhibitor in comparison to its mimic in both End1/E6E7 and SK-OV-3 cells. Magnification 10×, scale bar = 100 μm. (E) Wound-healing (scratch) assay in SK-OV-3 cells at different time intervals. Upon performing the scratch assay, the cells show reduced migratory capabilities when transfected with miR-34a, which is reversed by miR-34a inhibitor treatment. Each experiment has been carried out thrice. The data is depicted as mean ± SEM. ∗, ∗∗, and ∗∗∗ are representations of P-values <0.05, <0.01, and <0.001 respectively. Fig. 3 Impact of miR-34a on invasion and migration of the cells. (A–D) The cells show a marked increase in invasive (A, B) as well as migratory (C, D) potential upon transfection with miR-34a inhibitor in comparison to its mimic in both End1/E6E7 and SK-OV-3 cells. Magnification 10×, scale bar = 100 μm. (E) Wound-healing (scratch) assay in SK-OV-3 cells at different time intervals. Upon performing the scratch assay, the cells show reduced migratory capabilities when transfected with miR-34a, which is reversed by miR-34a inhibitor treatment. Each experiment has been carried out thrice. The data is depicted as mean ± SEM. ∗, ∗∗, and ∗∗∗ are representations of P-values <0.05, <0.01, and <0.001 respectively. Likewise, the migratory properties of miR-34a were investigated through transwell migration assays and it was determined that cells treated with the inhibitor showed a significantly higher migratory rate towards the lower chamber as compared to the mimic group (∼5-fold in End1/E6E7, p < 0.001, and ∼3.8-fold in SK-OV-3, p < 0.005) which clearly shows the functions of miR-34a on cellular migration ( Fig. 3 C and D). The wound-healing experiment was also conducted on SK-OV-3 cells for further validation of the migratory potential of the miR-34a. After scratch formation on the culture dishes, cell transfection with the mimic and inhibitor of miR-34a was done and cellular migration was monitored at regular intervals of 0, 18, 24, and 42 h. The cells in miR-34a mimic show decreased migration which reversed back with the incorporation of the inhibitor ( Fig. 3 E). For determining the influence of miR-34a in epithelial-to-mesenchymal transition, markers of EMT like n-cadherin, slug, vimentin, and e-cadherin expression levels were checked in SK-OV-3 cells. 50 μg of protein was used from the whole cell lysate for immunoblotting to check the expression of various markers of EMT. The protein expression was normalized using either β-actin or GAPDH as a normalization control. Results showed that miR-34a inhibits n-cadherin (∼4-fold, p < 0.05), slug (∼5-fold, p < 0.0005), and vimentin (∼2-fold, p < 0.05), expression whereas activates e-cadherin levels (∼4-fold, p < 0.005). Their effects were reversed back by the inhibitor ( Fig. 4 A–F). These results explained that miR-34a suppresses EMT and the decrease in miR-34a promotes endometriosis via the activation of EMT. Fig. 4 Effects of miR-34a on EMT and tumorigenesis. (A–F) Immunoblotting of the EMT markers in SK-OV-3 cells shows a marked reduction in n-cadherin, slug, and vimentin expression while upregulated e-cadherin expression when transfected with miR-34a, which reverses by its inhibitor treatment. (G–I) miR-34a mimic transfected cells showing a marked reduction in spheroid forming ability which reverses back upon the treatment with its inhibitor. Magnification 10×, scale bar = 100 μm, and 20X, scale bar = 50 μm. (J, K) miR-34a mimic group also shows a decrease in the stemness properties of the cells. Each experiment has been carried out thrice. Each experiment has been carried out thrice. The data is depicted as mean ± SEM. P-value <0.05, <0.01, and <0.001 referred to as ∗, ∗∗, and ∗∗∗ respectively. Fig. 4 Effects of miR-34a on EMT and tumorigenesis. (A–F) Immunoblotting of the EMT markers in SK-OV-3 cells shows a marked reduction in n-cadherin, slug, and vimentin expression while upregulated e-cadherin expression when transfected with miR-34a, which reverses by its inhibitor treatment. (G–I) miR-34a mimic transfected cells showing a marked reduction in spheroid forming ability which reverses back upon the treatment with its inhibitor. Magnification 10×, scale bar = 100 μm, and 20X, scale bar = 50 μm. (J, K) miR-34a mimic group also shows a decrease in the stemness properties of the cells. Each experiment has been carried out thrice. Each experiment has been carried out thrice. The data is depicted as mean ± SEM. P-value <0.05, <0.01, and <0.001 referred to as ∗, ∗∗, and ∗∗∗ respectively. To assess miR-34a's impact in inducing tumor growth, a colony formation assay was performed. miR-34a mimic transfection led to considerably less colony formation in SKO-V-3 cells, compared to both blank and negative control experiments. This effect was reversed upon transfection with an inhibitor for miR-34a, which increased the colony formation approximately two-fold (p < 0.05) ( Supplementary Figs. 1A and B ). Since tumor initiation and growth are the major events in oncogenesis, miR-34a is shown to have a potential contribution to tumor growth in endometriosis as well. In vitro research of tumorigenesis though performed by traditional two-dimensional (2D) colony-forming assay does not mimic the actual tumor microenvironment. Therefore, the oncogenic potential of miR-34a on the cells was further validated by performing a three-dimensional (3D) multicellular tumor spheroid culture technology to provide a better mimicry of conventional solid tumors and miR-34a's impact on its formation and the increased in the spheroids covered area were observed over time. According to our findings, the numbers of spheroids in the mimic group of miR-34a were considerably fewer than its inhibitor group (∼17-fold, p < 0.005) ( Fig. 4 G–I). Compared to the mimic group the spheroid's size also greatly increased upon the treatment of miR-34a-5p inhibitor (∼5-fold, p < 0.0001) proving an inhibitory role of miR-34a on tumorigenesis. To examine miR-34a's impact on the stemness of the cells FACS analysis was performed using CD44 + and CD117 + stem cell markers. The transfected cells were incubated with CD44 + and CD117 + markers after 72 h, and the results were checked using flow cytometry. The numbers of the double positive cells were notably reduced in the mimic group of miR-34a (∼1.8-fold, p < 0.05) ( Fig. 4 J and K). This observation suggests that miR-34a has the potential to reduce the stemness of the cells and therefore has a tumor-suppressing function in endometriosis.

In conclusion, this work is the first to reveal miR-34a-5p's involvement in endometriosis and is inversely correlated to MMP-2 activity. In addition, miR-34a-5p has a direct impact on the MMP-2 gene transcription in vitro. Furthermore, our studies unveiled that the increased MMP-2 gene expression and activity is due to the lack of its inhibition by miR-34a-5p. Exogenous miR-34a treatment markedly reduced colony formation, spheroid formation, stemness, invasion, and migration of the cells by inducing e-cadherin while reducing n-cadherin, slug, and vimentin expressions. Altogether, our study signifies that the loss of the microRNA-34a binding to the promoter region of MMP-2 increases its activity thus resulting in EMT and tumorigenesis thereby aggravating endometriosis progression.

The emergence of actively functional uterine endometrial glands and stroma in the ectopic site is the hallmark of endometriosis, an estrogen-dependent benign disease causing infertility [ 27 ]. The refluxed endometrial cell's potential to migrate and develop is impaired by ECM cleavage along with other factors, thereby becoming major contributors to the progression of endometriosis [ 28 ]. Despite being benign, endometriosis still exhibits a prime cause of developing ovarian cancer [ 29 ]. Studies have shown that several matrix metalloproteinases, such as MMP-9, MMP-2, MMP-7, and MMP-3, are linked with endometriosis [ [30] , [31] , [32] , [33] , [34] ]. Endometrial cells are susceptible to EMT-mediated reversion to their mesenchymal origin because they still bear certain imprints of their mesenchymal lineage. Few recent investigations have also shown that EMT contributes to deep-infiltrating endometriosis [ 34 ]. Our previous study has demonstrated the involvement of MMP-7 in EMT during endometriosis progression [ 27 ]. Additionally, several diseases, including endometriosis, have been linked to aberrant expression of microRNAs, which results in the perturbation of several cellular processes [ [2] , [3] , [4] , [5] , 35 , 36 ]. MicroRNA-34a is one of the most aberrantly expressed microRNAs in different diseases, including endometriosis [ 20 , [37] , [38] , [39] ]. miR-34a acts as a master regulator of EMT, stemness, tumor growth, tumor suppression, and drug resistance [ [40] , [41] , [42] ]. However, no study to date has been done to identify miR-34a's functional aspects in correlation with MMP activity and EMT in endometriosis. Database analysis showed a downregulated miR-34a-5p in various gynecological malignancies, including ovarian clear cell carcinoma. The present study documents decreased miR-34a-5p gene expression in case-control cohorts which is inversely correlated with increased MMP-2 gene expression in endometriosis. Previous reports in cervical cancer have shown that the 3′UTR in the MMP-2 gene's promoter region possesses a site for direct binding of miR-34a-5p where it harbors its anti-proliferative properties [ 26 ]. Therefore, miR-34a-5p's direct binding on the MMP-2-3′UTR was assessed by dual luciferase assay. Furthermore, we reported a considerable reduction in both the MMP-2 expression and activity levels upon the introduction of an exogenous miR-34a mimic in the endometrial cells which was reversed back by the incorporation of the inhibitor. In support, previous reports in esophageal squamous cell carcinoma and bladder cancer also documented the regulation of MMP-2 activity by miR-34a [ 23 , 24 ]. Since miR-34a is the key regulator of EMT, the profiles of various markers of EMT were also evaluated upon treatment of miR-34a mimic and its inhibitor in cultured cells. A notable change in the levels of n-cadherin, slug, vimentin, and e-cadherin was found. Moreover, miR-34a mimic showed a substantial reduction in the migratory as well as invaded cell numbers which further increased with the addition of its inhibitor. Additionally, miR-34a also reduced the colony and spheroid forming ability of the cells thereby proving the decrease in the oncogenic potential of the cells by miR-34a suggesting its tumor suppressor function. Reports also suggest that miR-34a has the potential to decrease the stemness of cancer stem cells [ 43 ]. Hence, miR-34a's impact on the stemness properties was also checked, and found that the miR-34a mimic transfected group showed considerably fewer numbers of CD44 + and CD117 + double-positive cells. Our results corroborated with the reports that describe the high expression of CD117 and CD44 in the cells making them highly tumorigenic [ 44 , 45 ] .

MicroRNAs (miRNAs or miRs), highly conserved single-stranded non-coding RNAs, interact with mRNA's 3′ untranslated regions (UTRs) to prevent their translation or cause their destruction at the post-transcriptional stage. About 30 % of the fundamental genes of humans, required for healthy growth and survival, are regulated by miRNAs [ 1 ]. Furthermore, miRNAs are dysregulated in many types of clinical disorders, encompassing cancer, cardiovascular disease, and neurological illnesses [ [2] , [3] , [4] , [5] ]. The family of miR-34 which involves miR-34a, 34b, and 34c, is dysregulated in many types of malignancies and is considered a suppressive miRNA for tumors due to its complementary effect with p53, which can also directly modulate miR-34. miR-34 is also a key regulator of epithelial-to-mesenchymal transition (EMT), metastasis, and cancer survival [ 6 ]. For instance, increasing miR-34a levels in bladder cancer cells can reduce their capacity to invade [ [7] , [8] , [9] ]. In the mice system Apc Min/+ where miR-34a/b/c was eradicated, the incidence of tumors and the chances of mortality were both considerably elevated. Moreover, colorectal cancer cells may be hampered in their ability to invade and migrate by miR-34a [ 10 ]. Furthermore, miR-34a can limit EMT-associated invasion and migration by negatively affecting the Wnt signaling pathway in primary prostate cancer [ 11 ]. The significance of miR-34 is gaining tremendous acknowledgment after the use of MRX34, the first microRNA medication targeting tumors, based on miR-34a mimics, in a phase-I clinical study ( NCT01829971 ) [ 12 ]. MiR-34 is in a dormant condition in various cancer cell types due to the absence of a 5′-phosphate which promptly activates by phosphorylation of 5′ end, as a result of DNA damage stimulation [ 13 ]. Zeng et al. [ 14 ] noted that patients with triple-negative breast cancer have downregulated miR-34a/c and an overall diminished survival rate. In vitro research revealed that miR-34a/c downregulation halts the invasion and migration of cells of breast cancer [ 15 ]. Endometriosis, a nonmalignant gynecological condition that leads to infertility in women, is marked by the deposition of uterine endometrium tissue in the ectopic sites and shows cancerous properties like invasion, migration, EMT, and tumor formation. Matrix metalloproteinases (MMPs) have been documented to advance endometriotic ailments [ 16 ]. Additionally, research indicates that endometriosis promotes the risk of ovarian cancer [ [17] , [18] , [19] ]. The involvement of miR-34a in endometriosis remains uncertain, as is the level of how deep the signaling cascade of miR-34a is embroiled in the advancement of the disease. However, one study found that endometrial tissue from endometriosis patients showed considerably suppressed expressions of p53, miR-34a, FoxO-1, Bax, and Bcl-2, and elevated Bcl-xL and SIRT-1 expressions suggesting miR-34a's impact on p53 in the pathophysiology of endometriosis [ 20 ]. MMPs are auxiliary with endometriosis and have a crucial impact on the disease's pathogenesis [ 21 ]. miR-34a has been shown to modulate specific MMP activity in cancers. For instance, in glioma patients, miR-34a expression was reduced while elevated MMP-9 expression was found. Overexpression of miR-34a dropped MMP-9 and MMP-2 protein levels and decreased invasion and migration in glioma and ESCC cells [ 22 , 23 ]. Accrued protein cleavage action of MMPs, including MMP-9 and MMP-7 affects ECM remodeling, invasion, and EMT process [ 24 ]. In this current research, our goal was to look into the functions of miR-34a in endometriosis disease advancement and the involvement of the MMP-2/9 axis thereon. Firstly, we assessed the levels of MMP-2 and miR-34a in a hospital-based case-control cohort, and the direct interaction between MMP-2 and miR-34a gene was studied by a gene reporter assay in cultured endometrial epithelial cells. We also did cell transfection with miR-34a mimic and its inhibitor to study various biochemical assays. Lastly, we studied EMT signaling molecules and tumorigenicity towards the disease pathogenesis.

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Samples of tissue and blood were procured from the patients with endometriosis-related symptoms at Spectrum Clinic & Endoscopy Research Institute's gynecology department (Kolkata, India). The Committee for Human Ethics of the Indian Institute of Chemical Biology (approval number: IICB/IRB/SS02-2015) and the Spectrum Clinic & Endoscopy Research Institute's Human Ethics Committee (approval number: SCERI-IEC015) approved the research procedure and the participants handed prior consent in writing. The endometriosis diagnosis was evidenced via laparoscopy. All samples of tissue and blood were taken in the menstrual cycle's proliferative stage. Samples of 11 women, aged within the range of 18–40, who showed pelvic pain symptoms and diagnosed endometriosis during laparoscopy, without the presence of any other gynecological or oncological problems and were not under any hormonal treatments for at least three months, and 11 healthy fertile controls of the same age range without any signs of endometriosis or other gynecologic illnesses, and undergone laparoscopy due to pain, recurrent pregnancy loss, amenorrhea or dysmenorrhea like symptoms, met the inclusion criteria. The tissue samples from the eutopic endometrium were collected from the control participants whereas the endometriotic tissues from the ectopic sites were collected from the endometriosis patients. A comprehensive computational analysis was meticulously conducted to thoroughly investigate and examine the miR-34a-5p expression in various types of malignancies using the CancerMIRNome database [ 25 ]. Additionally, the publicly available GEO dataset GSE239685 was re-analyzed to identify differential microRNA expression between normal controls and ovarian clear cell carcinoma. Human endometriotic tissues were utilized to extract total RNA through the RNeasy Mini Kit (Cat#74104, Qiagen, Germany). cDNAs were constructed from the RNA utilizing a High-Capacity cDNA Reverse Transcription Kit (Applied Biosystems, Foster City, CA). Comparative analysis of gene expression was done on a real-time thermocycler (7500 Real-Time PCR System, Applied Biosystems) implementing protocol of manufacturer using either TaqMan Universal PCR Master Mix No AmpErase UNG (Cat#4324018, Applied Biosystems) for microRNA-34a (Cat#4427975, Applied Biosystems) and RNu6b (Cat#4427975, Applied Biosystems), or TB Green Premix Ex Taq (Tli RNaseH Plus) (Takara) for MMP-2 (Sigma) and 18S rRNA (IDT, Integrated DNA Technology, Kolkata). The following primers were used: (a) MMP-2: (F: ACCAGAACACCATCGAGACC; R: TACTTTTAAGGCCCGAGCAA), (b) 18S: (F: GTAACCCGTTGAACCCCATT; R: CCATCCAATCGGTAGTAGCG). MMP-2: (F: ACCAGAACACCATCGAGACC; R: TACTTTTAAGGCCCGAGCAA), 18S: (F: GTAACCCGTTGAACCCCATT; R: CCATCCAATCGGTAGTAGCG). End1/E6E7, an epithelial cell line of endocervical origin, and SK-OV-3, an epithelial cell line of ovarian adenocarcinoma origin, were bought from the American Type Culture Collection (Manassas, VA, USA) and RPMI-1640 was used for culturing with a supplement of 10 % FBS (fetal bovine serum), streptomycin-penicillin (100 U/mL each) (Gibco, Thermo Fisher Scientific, Waltham, MA, USA) maintaining 5 % CO 2 . At 80 % confluency, transfection with 10 nM of either miR-34a mimic, miR-34a-5p inhibitor, or negative control (Sigma Aldrich, St. Louis, MO, USA) was done using Lipofectamine 2000 (Invitrogen). The cell supernatant was accumulated 48 h post-transfection for gelatin zymography. The cells were trypsinized, and counted and were used for invasion, migration, colony formation, and spheroid formation assays. Another set of transfection experiments with either negative control or mimic or inhibitor of miR-34a was done for wound healing assays and FACS. The transfection experiment for each analysis was replicated at least three times for accuracy. Imaging was done under an Olympus microscope with Olympus Camedia software (Chicago, MI, USA) (E-20P 5.0 megapixel). Densitometry was done for quantification using the software LabImage (Kapelan Bio-Imaging, Leipzig, Germany), and image processing was done on Adobe Photoshop 2021 v22.5.1.441 (San Jose, CA). MMP-2 promoter amplification from genomic DNA, isolated from the blood samples of the endometriosis patients, was carried out using the primers, 5′-TT GCGGCCGC GCCATGGGTATGCCGCTGGCCCTGGCTCCCACAGGCCC-3′ (forward), and 5′-CCC CTCGAG CACCTGGCCATGCCCTGTCCC-3′ (reverse), carrying restriction sites for Not I and Xho I respectively (NEB, Massachusetts, US) (the restriction sites are depicted in boldface) was done and the MMP-2 3′-UTR was incorporated in the renilla luciferase reporter vector psiCHECK2 (Promega, Madison, USA), and the presence of miR-34a-5p binding region for MMP-2 3′UTR in the plasmid vector was confirmed by DNA sequencing (BBS, Barcode BioSciences, Bangalore). 0.5 × 10 5 SK-OV-3 cells co-transfection with 200 ng psiCHECK2-MMP-2-3′-UTR vector and either mimic of miR-34a or both mimic and inhibitor in a plate of 96-well was done. Post 24 h the renilla luciferase activity and the activity of the firefly were determined from the collected cells utilizing the Dual-Luciferase Reporter Assay System (Cat#E1910, Promega, Madison, USA). Further, the normalization of the activity of renilla to firefly was done. The study conducted an MMP-2 activity assay by collecting transfected cell supernatant, electrophoresing it on an 8 % SDS-PAGE gel having gelatin (1 mg/mL), washing the gels, incubating in a calcium assay buffer, staining with Coomassie blue, and quantifying the gelatinolytic activity. The assays for the invasion were executed on End1 E6/E7 and SK-OV-3 cells utilizing the Transwell Boyden Chamber Assay Kit (BD Biosciences, Franklin Lakes, NJ, USA). 25 × 10 3 transfected cells (with either mimic or inhibitor of miR-34a) were given on the well's upper chamber (pre-coated with Matrigel) in serum-free conditions while the chamber on the lower side was supplied with complete medium (having 10 % FBS). Upper chamber cells were removed after 24 h and cell fixation and staining of the upper chamber's lower side was done using paraformaldehyde and crystal violet. Images were taken from 4 random sites. The migration assays were carried out on End1 E6/E7 and SK-OV-3 cells exploiting the Transwell Boydon Chamber Assay Kit similarly to invasion assay but without matrigel coating, and the images were taken for each well from 4 random sites. Wound-healing assays were executed on SK-OV-3 cells by scratch method. 10 6  cells were seeded, and cultured till a monolayer formed, then a scratch was made with the tip of a pipette, following transfection with mimic or inhibitor. The images were captured in multiple intervals of time to study the migration of the cells. Whole-cell extracts (50 μg/Lane) were electrophoresed using 8 % reducing SDS-PAGE and the transfer was done onto a PVDF membrane and overnight incubation with primary antibodies. Primary antibodies to e-cadherin (R&D#AF748) from R&D Systems (Minneapolis, USA), n-cadherin (CST#13116), slug (CST#9585), vimentin (CST#5741), and β-actin (CST#4967) (Cell Signaling Technology, Danvers, MA, USA) and GAPDH (SC#47724) (Santa Cruz Biotechnology, Dallas, TX, USA) were used in 1:1000 ratio (for CST) and 1:500 ratio (for Santa Cruz). Incubation of the membranes with secondary antibody (HRP-conjugated) (CST#7074 & CST#7076), was done following washing. The band visualization was done using Immobilon HRP substrate (Millipore). A colony formation experiment on SK-OV-3 cells was done by seeding 10 3 transfected cells (either with negative control miRNA, miR-34a mimic, or its inhibitor) onto a 100 mm cell culture dish. Post 10 days of transfection, the media was discarded from the cells followed by washing with PBS and fixation and visualization with paraformaldehyde and methylene blue respectively. The number of colonies was counted from each dish. Spheroid formation assays on SK-OV-3 cells was done by seeding 10 3 transfected cells (either with negative control miRNA or mimic or inhibitor of miR-34a) onto an ultra-low attachment 6-well plate in cancer stem cell-specific media which is composed of 80 mL of DMEM F12/KO media and 20 mL of knockout serum along with 10 ng/mL EGF and 10 ng/mL bFGF (Gibco, Thermo Fisher Scientific, Waltham, MA, USA). The formation of spheroids was then checked followed by imaging, first after 24 h and again after 7 days. The size of the spheroids was measured and the total spheroid numbers were counted from each well. The stemness properties of the cells were checked in either control miRNA or miR-34a mimic transfected SK-OV-3 cells. The transfected cell pellets were accumulated after 72 h and processed with the following antibodies. Control was treated with either CD44 + (Clone REA690), or CD117 + (Clone REA787) (Miltenyi Biotech, Auburn, CA, USA), both in combination or unstained, and mimics were treated with either CD44 + , CD117 + , or both in combination and were incubated in the dark. The data was acquired utilizing flow cytometry. The proportional levels of miR-34a and MMP-2 mRNAs normalized to RNU6B and 18S mRNAs were calculated respectively. The statistical analysis was executed on GraphPad InStat3 (Graph Pad Software, Inc) and the comparison between the groups was done by performing either Student's t -test or a one-way or two-way ANOVA test, and Tukey's test for multiple-comparison. Each experiment has been carried out thrice. All values were shown as mean ± SEM. The level chosen to be significant is at a P-value of <0.05.

All data presented in the article is available with the author and can be presented upon reasonable request.