Proteolytic Imbalance of the MMP/E-Cadherin Axis in Genitourinary Syndrome of Menopause: A Cross-Sectional Study.

With increasing female life expectancy globally, the postmenopausal phase now spans approximately one-third of a woman’s lifespan. During this period, many women experience Genitourinary Syndrome of Menopause (GSM), a term encompassing a range of symptoms and signs associated with decreased estrogen levels which significantly impact quality of life. 1 Several studies have documented a prevalence rate of up to 80% for at least one symptom associated with genitourinary syndrome of menopause (GSM). 2 The hallmark pathological features of these atrophic changes include a measurable thinning of the vaginal epithelium and mucosa, 3 reduced collagen content, and compromised extracellular matrix integrity. However, the precise molecular mechanisms driving these structural changes in the postmenopausal vagina remain poorly understood. The vaginal epithelial barrier serves as the primary defense against external insults. Its integrity is essential not only for physical protection but also for maintaining the vaginal microecological balance, an ecosystem known to be significantly altered during menopause. 4 The integrity of this barrier is critically dependent on intercellular adhesion molecules. While tight junctions (TJs) regulate paracellular permeability, adherens junctions (AJs), anchored by E-cadherin, provide the mechanical cohesion necessary to maintain TJ stability and overall epithelial barrier function. Central to this process is the E-cadherin/β-catenin complex, which serves as a cornerstone of adherens junctions and is essential for epithelial architecture and tissue integrity. 5 Under certain pathological conditions, the ectodomain of E-cadherin can be cleaved, generating soluble fragments that not only disrupt the epithelial barrier but can also actively participate in disease pathogenesis. 6 While animal models have documented weakened vaginal barrier function following ovariectomy, direct evidence elucidating these specific molecular events in humans across the menopausal transition remains limited. Matrix metalloproteinases (MMPs) are a family of zinc-dependent endopeptidases that are key regulators of extracellular matrix (ECM) remodeling. Beyond their role in ECM turnover, members of the metalloproteinase superfamily, including both MMPs and ADAMs (A Disintegrin and Metalloproteinase), are known to be key proteases capable of shedding the E-cadherin ectodomain. 7 Furthermore, evidence from both animal models and human clinical studies indicates that the expression and activity of various MMPs are elevated in the postmenopausal, hypoestrogenic state. We focused on MMP-1, MMP-7, and MMP-9 due to their distinct yet complementary substrate specificities and their established relevance to epithelial biology. 8 , 9 Estrogen, acting primarily through ER-β, normally suppresses MMP expression, and its withdrawal during menopause is therefore expected to de-repress these proteases. 10 We therefore hypothesized that a similar upregulation of MMP activity occurs in the vagina after menopause, resulting in the proteolytic degradation of E-cadherin. This potential interplay represents a critical and underexplored mechanism underlying postmenopausal vaginal atrophy. Therefore, this study was designed to characterize the expression profiles of E-cadherin and key related MMPs (MMP-1, −7, and −9) across the menopausal stages. By employing an integrative approach that simultaneously evaluates tissue protein, mRNA expression, and soluble protein levels, this study provides a comprehensive molecular characterization of the human vaginal epithelial barrier. We aimed to elucidate the molecular relationship between epithelial adhesion and tissue remodeling in the postmenopausal vagina, thereby providing a deeper understanding of the pathophysiological basis of the atrophic changes that define GSM.

IHC analysis revealed that E-cadherin was predominantly localized to the plasma membrane of vaginal epithelial cells ( Figure 1A–C ). Quantitative H-score analysis showed a significant, stepwise decrease in E-cadherin protein expression across the menopausal transition ( Figure 1D ). The H-scores were significantly lower in both the early postmenopausal (6.07 ± 3.02) and late postmenopausal (5.67 ± 2.23) groups compared to the premenopausal group (8.73 ± 2.70; P < 0.01 for both). Figure 1 Decreased E-cadherin and Increased Matrix Metalloproteinase (MMP) Expression in Vaginal Tissue Across the Menopausal Transition. Immunohistochemical staining shows the expression and localization of E-cadherin in premenopausal ( A ), early postmenopausal ( B ), and late postmenopausal ( C ) vaginal tissue, with corresponding quantitative H-score analysis ( D ). MMP-1 expression in premenopausal ( E ), early postmenopausal ( F ), and late postmenopausal ( G ) groups with quantitative analysis ( H ). MMP-7 expression in premenopausal ( I ), early postmenopausal ( J ), and late postmenopausal ( K ) groups with quantitative analysis ( L ). MMP-9 expression in premenopausal ( M ), early postmenopausal ( N ), and late postmenopausal ( O ) groups with quantitative analysis ( P ). Scale bars = 100 μm. Data are presented as mean ± SD. **** P < 0.0001, ** P < 0.01, * P < 0.05. Decreased E-cadherin and Increased Matrix Metalloproteinase (MMP) Expression in Vaginal Tissue Across the Menopausal Transition. Immunohistochemical staining shows the expression and localization of E-cadherin in premenopausal ( A ), early postmenopausal ( B ), and late postmenopausal ( C ) vaginal tissue, with corresponding quantitative H-score analysis ( D ). MMP-1 expression in premenopausal ( E ), early postmenopausal ( F ), and late postmenopausal ( G ) groups with quantitative analysis ( H ). MMP-7 expression in premenopausal ( I ), early postmenopausal ( J ), and late postmenopausal ( K ) groups with quantitative analysis ( L ). MMP-9 expression in premenopausal ( M ), early postmenopausal ( N ), and late postmenopausal ( O ) groups with quantitative analysis ( P ). Scale bars = 100 μm. Data are presented as mean ± SD. **** P < 0.0001, ** P < 0.01, * P < 0.05. In contrast to E-cadherin, MMP-1, MMP-7, and MMP-9 were primarily detected in the cytoplasm of vaginal epithelial and stromal cells ( Figure 1E–P ). The expression levels of all three MMPs were significantly elevated in the late postmenopausal group compared to the premenopausal group ( P < 0.05 for all). Specifically, MMP-9 expression was already significantly higher in the early postmenopausal group than in the premenopausal group ( P < 0.05). Furthermore, both MMP-7 and MMP-9 levels were significantly higher in the late postmenopausal group compared to the early postmenopausal group ( P < 0.05). To determine if these protein changes were due to transcriptional regulation, we quantified mRNA levels via RT-qPCR ( Figure 2 ). We found no significant differences in CDH1 mRNA expression among the three groups. In contrast, the mRNA levels of MMP1 and MMP9 were significantly elevated in the early postmenopausal group and further increased in the late postmenopausal group ( P < 0.05). MMP7 mRNA was significantly upregulated only in the late postmenopausal group compared to the premenopausal group ( P < 0.05). Figure 2 Upregulation of MMP mRNA but Stable E-cadherin mRNA Expression in the Vagina Across the Menopausal Transition. Relative mRNA expression levels of CDH1 (E-cadherin, ( A ), MMP1 ( B ), MMP7 ( C ), and MMP9 ( D ) were quantified by RT-qPCR in vaginal tissue from the three groups (n=30 per group). While CDH1 mRNA remained unchanged, the mRNA levels of MMP1, MMP7 , and MMP9 were significantly upregulated in postmenopausal groups. Data are presented as mean ± SD. **** P < 0.0001, *** P < 0.001, * P < 0.05. Upregulation of MMP mRNA but Stable E-cadherin mRNA Expression in the Vagina Across the Menopausal Transition. Relative mRNA expression levels of CDH1 (E-cadherin, ( A ), MMP1 ( B ), MMP7 ( C ), and MMP9 ( D ) were quantified by RT-qPCR in vaginal tissue from the three groups (n=30 per group). While CDH1 mRNA remained unchanged, the mRNA levels of MMP1, MMP7 , and MMP9 were significantly upregulated in postmenopausal groups. Data are presented as mean ± SD. **** P < 0.0001, *** P < 0.001, * P < 0.05. ELISA of vaginal lavage fluid revealed that the concentration of sE-cad was significantly higher in both the early and late postmenopausal groups compared to the premenopausal group ( P < 0.05) ( Figure 3 ). Consistent with the tissue expression data, concentrations of soluble MMP-7 and MMP-9 were also significantly elevated in both postmenopausal groups compared to the premenopausal group ( P < 0.05). However, no significant differences in soluble MMP-1 concentration were observed among the three groups. Figure 3 Elevated Levels of Soluble E-cadherin (sE-cad) and MMPs in Vaginal Lavage Fluid from Postmenopausal Women. Concentrations of sE-cad ( A ), MMP-1 ( B ), MMP-7 ( C ), and MMP-9 ( D ) in vaginal lavage fluid (n=30 per group) were quantified by ELISA. Levels of sE-cad, MMP-7, and MMP-9 were significantly elevated in both postmenopausal groups compared to the premenopausal group. Data are presented as mean ± SD. **** P < 0.0001, *** P < 0.001, ** P < 0.01. Elevated Levels of Soluble E-cadherin (sE-cad) and MMPs in Vaginal Lavage Fluid from Postmenopausal Women. Concentrations of sE-cad ( A ), MMP-1 ( B ), MMP-7 ( C ), and MMP-9 ( D ) in vaginal lavage fluid (n=30 per group) were quantified by ELISA. Levels of sE-cad, MMP-7, and MMP-9 were significantly elevated in both postmenopausal groups compared to the premenopausal group. Data are presented as mean ± SD. **** P < 0.0001, *** P < 0.001, ** P < 0.01. To explore the interplay between these molecules, we performed correlation analyses ( Figure 4 ). Spearman correlation analysis demonstrated a significant inverse correlation between E-cadherin protein and MMP-9 protein expression in vaginal tissue (r = −0.552, P < 0.001). In the vaginal lavage fluid, there were significant positive correlations between sE-cad and both MMP-9 (r = 0.448, P = 0.009) and MMP-7 (r = 0.219, P = 0.016). Figure 4 Correlation Analyses of E-cadherin and MMPs. ( A ) Spearman correlation analysis shows a significant inverse correlation between E-cadherin and MMP-9 protein expression in vaginal tissue. ( B ) In vaginal lavage fluid, there are significant positive correlations between sE-cad and both MMP-9 and MMP-7. *** P < 0.001, ** P < 0.01, * P < 0.05. Correlation Analyses of E-cadherin and MMPs. ( A ) Spearman correlation analysis shows a significant inverse correlation between E-cadherin and MMP-9 protein expression in vaginal tissue. ( B ) In vaginal lavage fluid, there are significant positive correlations between sE-cad and both MMP-9 and MMP-7. *** P < 0.001, ** P < 0.01, * P < 0.05.

This cross-sectional study recruited participants were recruited using a consecutive sampling strategy from the Department of Gynecology at the Women’s Hospital, Zhejiang University School of Medicine between December 2019 and May 2021. Inclusion criteria were: (1) age 40–65 years; and (2) having undergone total hysterectomy for benign conditions, including uterine fibroids or adenomyosis, with postoperative pathological confirmation. Exclusion criteria were as follows: (1) diagnosis of acute vaginitis within the past month; (2) lower reproductive tract infections within the past month; (3) history of malignant tumors; (4) human papillomavirus (HPV) infection within the past 6 months; (5) no history of sexual intercourse, making gynecological examination unfeasible; (6) use of menopausal hormone therapy (MHT) within 6 months prior to enrollment; and (7) diagnosis of autoimmune diseases. Participants were stratified into three groups (n=30 per group): the premenopausal group, the early postmenopausal group (1–5 years since the last menstrual period), and the late postmenopausal group (>5 years since the last menstrual period). The baseline clinical characteristics of the participants are summarized in Supplementary Table 1 . Premenopausal participants underwent surgery during the early proliferative phase whenever feasible, based on self-reported menstrual history. All surgeries were performed during daytime hours, between 8 AM and 5 PM. The study protocol was in accordance with the Declaration of Helsinki and received approval from the Human Ethics Committee of Women’s Hospital, Zhejiang University School of Medicine (No. 20190071). Written informed consent was obtained from all participants. Vaginal Lavage Fluid: Prior to any gynecological examination or surgical procedure, vaginal lavage was performed. Briefly, the cervix was exposed using a disposable speculum, and after clearing visible secretions, the vaginal fornix was irrigated with 2 mL of sterile phosphate-buffered saline (PBS). The fluid was immediately aspirated, transferred to a sterile tube, and maintained at 4°C. All samples were transported to the laboratory within 30 minutes. Upon arrival, samples were centrifuged at 1500 × g for 5 minutes at 4°C. The supernatant was then aliquoted and stored at −80°C. Vaginal Wall Tissue: Following the hysterectomy, two full-thickness tissue specimens (approx. 5 mm × 5 mm) were obtained from the posterior vaginal fornix. One specimen was fixed for immunohistochemistry, and the other was placed in RNAlater (Qiagen, Valencia, CA, USA), snap-frozen, and stored at −80°C. Vaginal wall tissue specimens were fixed in 4% paraformaldehyde, paraffin-embedded, and sectioned at 5 μm. Immunostaining was performed using the avidin–biotin complex method. Sections were incubated with primary antibodies from Thermo Fisher Scientific: MMP-1 (1:50, overnight at room temperature), MMP-7 (1:50, 60 min at room temperature), MMP-9 (1:100, 60 min at room temperature), and E-cadherin (1:100, 60 min at room temperature). Sections were visualized using a labeled streptavidin–biotin (LSAB) kit (Dako, Carpinteria, CA, USA) with diaminobenzidine. For evaluation, five high-power fields were randomly selected per slide. A final histochemical score (H-score) was calculated by multiplying the intensity score (0–3) by the proportion score (0–4). Two independent pathologists, blinded to group allocation, evaluated all slides. Inter-observer agreement was assessed using intraclass correlation coefficients (ICC), demonstrating good agreement (ICC > 0.85). Total RNA was extracted from vaginal tissues using TRIzol reagent (Invitrogen, Carlsbad, CA, USA). 100 ng of total RNA was reverse-transcribed into cDNA using oligo(dT) primers. Expression of MMP1, MMP7, MMP9 , and CDH1 (E-cadherin) mRNA was quantified using SYBR Green–based qPCR. Primers used were: CDH1 (F: 5′-CGAGAGCTACACGTTCACGG-3′; R: 5′- GGGTGTCGAGGGAAAAATAGG −3′), MMP1 (F: 5′-CTCTGGAGTAATGTCACACCTCT-3′; R: 5′- TGTTGGTCCACCTTTCATCTTC-3′), MMP7 (F: 5′-GAGTGAGCTACAGTGGGAACA-3′; R: 5′-CTATGACGCGGGAGTTTAACAT-3′), MMP9 (F: 5′-AGACCTGGGCAGATTCCAAAC-3′; R: 5′- CGGCAAGTCTTCCGAGTAGT-3′), and ACTB (β-actin) (F: 5′-AGCAGTTGTAGCTACCCGCCA-3′; R: 5′-GGCGGGCACGTGAAGTCT-3′). β-actin served as the internal control. Primer amplification efficiencies were confirmed to be between 95% and 105% using standard curve analysis. Melting curve analysis was performed for all reactions to confirm single-product amplification.All samples were run in triplicate technical replicates. RNA purity and concentration were assessed using a NanoDrop 2000 spectrophotometer (Thermo Fisher Scientific, USA), with A260/A280 ratios between 1.8 and 2.0. RNA integrity was also verified prior to reverse transcription. Relative expression was calculated using the 2 −ΔΔC t method. E-cadherin, total MMP-1, total MMP-7, and MMP-9 levels in vaginal lavage fluid were measured using the DuoSet ELISA Development Systems for Human E-Cadherin (R&D Systems, Minneapolis, MN, USA; Cat. No. DY648), Human Total MMP-1 (Cat. No. DY901B), Human Total MMP-7 (Cat. No. DY907), and Human MMP-9 (Cat. No. DY911), respectively, according to the manufacturers’ general ELISA protocols. Briefly, capture antibodies were diluted in PBS and used to coat 96-well microplates overnight at room temperature. After washing, plates were blocked with 1% BSA in PBS for at least 1 h and then incubated with standards and samples for 2 h at room temperature. Detection antibodies were subsequently applied for 2 h, followed by streptavidin-HRP for 20 min and TMB substrate for 20 min. Optical density was measured immediately at 450 nm with wavelength correction at 540 or 570 nm. All standards and samples were assayed in duplicate, and concentrations were calculated using four-parameter logistic (4-PL) curve fitting according to the manufacturers’ instructions. Data are presented as mean ± standard deviation (SD). The normality of data distribution was verified using the Shapiro–Wilk test. Comparisons between two groups were performed using an unpaired Student’s t -test. One-way ANOVA followed by Tukey’s post-hoc test was used for multiple comparisons. Correlation analyses were performed using Spearman’s rank correlation coefficient. A P -value < 0.05 was considered statistically significant. All analyses were conducted using GraphPad Prism version 10 (GraphPad Software, San Diego, CA, USA) and SPSS version 25.0 (IBM SPSS, Armonk, NY, USA). During the preparation of this manuscript, the authors utilized the AI language model Gemini (Google) for assistance. The AI tool was used primarily for language editing, improving clarity, and refining sentence structure. The authors reviewed, edited, and take full responsibility for all content, ensuring the scientific accuracy and integrity of the final manuscript.

In conclusion, this study provides evidence that the pathophysiology underlying postmenopausal vaginal atrophy involves a proteolytic imbalance associated with the loss of estrogen. This state is characterized by the upregulation of MMPs which, in turn, may contribute to the post-transcriptional degradation of E-cadherin, compromising the integrity of the vaginal epithelial barrier. This MMP/E-cadherin axis may represent a key mechanism of these atrophic changes, offering a deeper understanding of the molecular events in the postmenopausal vagina and identifying sE-cad and MMP-9 as promising avenues for novel biomarkers and therapeutic strategies. While the link to clinical symptoms remains inferential due to the absence of symptom stratification, our findings provide a novel framework for GSM. Future longitudinal and mechanistic studies are needed to clarify causal relationships and evaluate their clinical applicability.

As a cornerstone of epithelial health, E-cadherin is essential for tissue homeostasis; it forms the adherens junctions (AJs) that seal intercellular spaces, thereby creating a selective, semi-permeable barrier. 5 Against this backdrop, our primary finding of a significant post-transcriptional downregulation of E-cadherin protein in the vaginal epithelium of postmenopausal women is particularly impactful. This reduction provides a direct molecular basis for the clinical signs of epithelial fragility and thinning characteristic of GSM. This result is consistent with foundational work in animal models of menopause; for instance, recent studies have confirmed that E-cadherin expression is significantly reduced in the atrophic vaginal epithelium of ovariectomized rats. 11 Furthermore, our molecular data align perfectly with and provide a mechanistic explanation for earlier functional studies in women. These seminal studies by Gorodeski established that the vaginal epithelial barrier becomes significantly more permeable in the postmenopausal state, a functional impairment directly linked to the loss of estrogen’s modulatory effects, likely mediated through pathways such as the estrogen receptor-β signaling axis. 10 , 12 , 13 Indeed, the protective role of estrogen on epithelial surfaces appears to be a systemic phenomenon, not confined to the reproductive tract. For example, the menopause transition has been directly linked to increased gut permeability, which promotes systemic inflammation and may contribute to other postmenopausal comorbidities like bone loss. 14 A parallel process occurs in the skin, where estrogen withdrawal leads to a decline in collagen production, impaired barrier function, and accelerated aging. 15 Therefore, our observation of E-cadherin downregulation in the vagina aligns with a broader biological principle of estrogen-mediated mucosal and epithelial homeostasis throughout the body. Our study advances this understanding by demonstrating that this protein loss occurs despite stable mRNA levels and is accompanied by a significant increase in sE-cad fragments in the vaginal fluid. This strongly suggests that E-cadherin depletion is not due to reduced synthesis but to accelerated post-transcriptional degradation. Among the known post-transcriptional mechanisms, a key pathway is the proteolytic shedding of the E-cadherin ectodomain from the cell surface. This process is mediated by families of zinc-dependent metalloproteinases, most notably the ADAM family and, critically for our study, various MMPs. 7 , 16 , 17 An alternative, and potentially complementary, pathway involves the enhanced internalization and subsequent lysosomal degradation of the entire protein complex. 18 While multiple mechanisms may be at play, our finding of elevated sE-cad levels provides strong support for the proteolytic shedding pathway. Indeed, sE-cad is now recognized as a direct biomarker of genital epithelial disruption resulting from this enzymatic cleavage, 19 pointing to an active proteolytic process on the cell surface as a key event in the pathogenesis of vaginal atrophy. Complementing the loss of epithelial adhesion, our study demonstrates a marked and progressive upregulation of MMP-1, MMP-7, and MMP-9 at both the protein and mRNA levels in the vaginal tissue of postmenopausal women. The increased transcription of these genes is consistent with the established regulatory role of estrogen; indeed, studies in various tissues have shown that estrogen signaling can directly suppress MMP transcription and maintain a favorable balance between MMPs and their endogenous inhibitors. 20 , 21 Its withdrawal during menopause logically leads to a de-repression of MMP gene expression, creating a proteolytic-rich environment. The consequences of this upregulation are likely twofold. First, as a fundamental function of this enzyme family, these MMPs contribute to the direct degradation of ECM components like collagen, leading to the characteristic tissue thinning and loss of structural support seen in vaginal atrophy. 22 Second, and central to our hypothesis, is their role in actively dismantling the epithelial barrier itself. The strong inverse correlation we observed between E-cadherin and MMP-9 protein expression in tissue, coupled with the robust positive correlation between their soluble forms (sE-cad and MMP-9) in lavage fluid, provides compelling evidence for this direct mechanistic link. This strongly suggests that in the postmenopausal vagina, upregulated MMPs—particularly MMP-9—are key proteases responsible for the cleavage of E-cadherin. The coordinated molecular changes observed in this study—diminished E-cadherin and elevated MMPs—are clearly linked to the hormonal shifts of the menopausal transition. We propose these findings can be integrated into a cohesive model that describes the molecular underpinnings of postmenopausal vaginal tissue alterations. This model provides a pathophysiological basis for the epithelial fragility and tissue atrophy that are the hallmarks of GSM, a highly prevalent condition with significant impact on women’s health. 23 , 24 In this model, the hypoestrogenic state may contribute to the deterioration of vaginal tissue integrity through a “dual-hit” mechanism: first, it removes the suppressive brakes on the expression of matrix-degrading enzymes, and second, it fosters a proteolytic environment that promotes the degradation of key adhesion molecules essential for epithelial integrity. The progressive nature of MMP upregulation from early to late postmenopause further suggests that the duration of hypoestrogenism is a critical factor that exacerbates this proteolytic imbalance over time. Our findings have significant clinical implications and open new avenues for research. From a diagnostic perspective, the quantification of sE-cad and MMP-9 in vaginal lavage fluid could represent a potential non-invasive biomarker strategy to objectively assess the severity of vaginal epithelial disruption and perhaps monitor the efficacy of treatments, such as MHT or other local therapies. From a therapeutic standpoint, this work highlights the MMP/E-cadherin axis as a potential non-hormonal target. Interventions aimed at selectively inhibiting MMP-9 activity could offer a novel approach for restoring vaginal epithelial health in women for whom hormone therapy is contraindicated or not desired. This study had several limitations. Its cross-sectional design identifies strong associations but cannot definitively establish causality. Additionally, the study population was comprised of women undergoing hysterectomy for benign indications, which may not be fully representative of the general postmenopausal population. Variations in indications for hysterectomy may introduce unmeasured biological heterogeneity. The ELISA assays were DuoSet Development Systems, not validated for vaginal lavage fluid, and lot-specific detection limits and assay variability were unavailable; therefore, these results should be interpreted with caution. Although the age difference between groups is an inherent biological feature of the menopausal transition, the potential confounding influence of chronological aging cannot be entirely excluded, particularly as no multivariate analysis was performed to control for this variable. Crucially, as our cohort was not stratified by the presence or absence of clinical GSM, this study describes the molecular events following menopause but cannot directly conclude a definitive causal link to the syndrome itself. Future research should prioritize longitudinal studies that investigate the relationship between these molecular markers and the presence, severity, and treatment response of clinical GSM. Furthermore, functional experiments using in vitro vaginal epithelial cell cultures or ex vivo tissue models are warranted to confirm the causal link between MMP activity and E-cadherin cleavage and to test potential therapeutic inhibitors.