Piezo1 mediates hypoxia-induced endometriosis fibrosis via the mtDNA -dependent cGAS-STING pathway

Studies have shown that Piezo1 is involved in the fibrotic processes underlying various diseases [ 43 , 44 ], yet its function in EMs fibrosis is uncertain. To investigate the potential role of Piezo1 in endometriosis-associated fibrogenesis, we collected a spectrum of tissue samples: normal endometrium, eutopic endometrium and ectopic lesions from patients with endometriosis. Histological examination revealed marked architectural distortion in ectopic lesions, characterized by stromal thickening, glandular hyperplasia, and inflammatory cell infiltration (Supplementary Fig. S1 ) Endometriotic lesions exhibited significant fibrosis, as assessed by markedly increased collagen deposition in both Masson’s trichrome and Sirius red staining compared with normal endometrium (Fig.  1 A, B). Immunofluorescence further confirmed the concurrent upregulation of Piezo1, α-SMA, and Collagen I in ectopic lesions versus normal endometrium. These proteins co-localized in active fibrotic foci, with their fluorescence intensities showing a positive correlation (Fig.  1 C, D). Collagen I, a major component of the extracellular matrix, and α-SMA, a marker of myofibroblast activation, are well-established indicators of tissue fibrosis. This supports the involvement of Piezo1 in endometriosis-associated fibrogenesis. Immunohistochemical (IHC) analysis demonstrated that the expression of both HIF-1α and Piezo1 was significantly higher in ectopic lesions compared with matched eutopic endometrium (Fig.  1 D-F). Immunohistochemistry (IHC) revealed elevated Piezo1 expression in ectopic lesions, but not in eutopic endometrium, relative to normal controls (Fig.  1 E, F). In contrast, western blotting (WB) detected significant Piezo1 upregulation in both eutopic and ectopic tissues (Fig.  1 G, H), a discrepancy likely attributable to the higher sensitivity of WB for detecting membrane proteins under denaturing conditions. Importantly, both methods consistently showed robust Piezo1 upregulation in ectopic lesions, supporting its role in endometriosis-associated fibrosis. Western blotting analysis revealed a coordinate upregulation of Piezo1, HIF-1α, cGAS and STING at the protein level in ectopic endometriotic lesions relative to normal endometrial controls (Fig.  1 G, H), while qRT-PCR analysis confirmed a corresponding upregulation at the mRNA level (Supplementary Fig. S2 ). Our findings demonstrate that Piezo1 is specifically upregulated in endometriotic lesions and that its expression positively correlates with both the extent of fibrosis and the hypoxia marker HIF-1α, thus supporting a pivotal role for Piezo1 in the pathogenesis of endometriosis-associated fibrosis. Fig. 1 Piezo1 expression was upregulated in ectopic lesions from patients with endometriosis. (A , B) Masson’s trichrome and Sirius Red staining show collagen deposition (indicative of fibrosis) in normal endometrial (NM), eutopic endometrial (Eu), and ectopic endometrial (Ec) (Scale bar: 100 μm). (C , D) Immunofluorescence analysis of Piezo1, Collagen I (a major ECM component, indicator of collagen deposition), and α-SMA (a myofibroblast activation marker) expression in NM, Eu, and Ec tissues. Representative images are shown (Scale bar: 100 μm). (E , F) Immunohistochemical (IHC) analysis and quantification of HIF-1α and Piezo1 protein expression in NM, Eu, and Ec tissues (Scale bar: 100 μm). (G , H) Western blotting analysis and quantification of HIF-1α, Piezo1, cGAS, and STING protein levels in NM, Eu, and Ec tissues ( n  = 9). *p  <  0.05; ** P  <  0.01; *** P  <  0.001; **** P  < 0.0001 Piezo1 expression was upregulated in ectopic lesions from patients with endometriosis. (A , B) Masson’s trichrome and Sirius Red staining show collagen deposition (indicative of fibrosis) in normal endometrial (NM), eutopic endometrial (Eu), and ectopic endometrial (Ec) (Scale bar: 100 μm). (C , D) Immunofluorescence analysis of Piezo1, Collagen I (a major ECM component, indicator of collagen deposition), and α-SMA (a myofibroblast activation marker) expression in NM, Eu, and Ec tissues. Representative images are shown (Scale bar: 100 μm). (E , F) Immunohistochemical (IHC) analysis and quantification of HIF-1α and Piezo1 protein expression in NM, Eu, and Ec tissues (Scale bar: 100 μm). (G , H) Western blotting analysis and quantification of HIF-1α, Piezo1, cGAS, and STING protein levels in NM, Eu, and Ec tissues ( n  = 9). *p  <  0.05; ** P  <  0.01; *** P  <  0.001; **** P  < 0.0001 The significant upregulation of HIF‑1α, a major transcription factor responsive to hypoxia, was confirmed histologically in EMs lesions (Fig.  1 D‑G). These results suggest that hypoxia plays a crucial role in driving fibrosis in EMs. To model the hypoxic microenvironment of endometriotic lesions, which ranges from 0.5% to 2% O₂ in ovarian endometrioma cysts, HESCs were cultured under 1% O₂, a widely accepted condition for studying chronic hypoxic stress in endometrial stromal cells [ 45 – 48 ]. To establish an in vitro disease model, HESCs were exposed to hypoxia (1% O₂, 5% CO₂) for 0-48 h to identify the critical time point at which hypoxia induces fibrotic transition. Western blotting and qRT-PCR analyses revealed that HIF-1α expression peaked following 24 h of hypoxia, which correlated with a concurrent, significant increase in both Piezo1 protein and mRNA levels at this time point (Fig.  2 A-C). Meanwhile, the expression of the fibrosis markers α‑SMA and Collagen I was significantly upregulated by 24 h and remained elevated through 48 h (Fig.  2 D, E). Consistently, Immunofluorescence staining revealed a significant increase in the expression of the myofibroblast marker α-SMA in HESCs following 24 h of hypoxia (Fig.  2 F, G). Concurrently, flow cytometry analysis revealed no significant increase in apoptosis under hypoxic conditions (Fig.  2 H, I), and EdU cell proliferation assay demonstrated that cell proliferative capacity remained robust under the same 24 h hypoxic culture conditions (Fig.  2 J, K). Given the concurrent peak activation of HIF-1α, marked upregulation of Piezo1, and maintained cellular viability and proliferation, the 24 h hypoxic exposure was selected for subsequent hypoxic induction in our cellular model. To determine whether HIF‑1α directly regulates Piezo1 transcription, we performed ChIP‑qPCR assays. Based on JASPAR prediction, a cluster of hypoxia response elements (HREs) was identified within the Piezo1 promoter (positions - 220 to -180 bp). Chromatin fragmentation was confirmed by agarose gel electrophoresis, with DNA fragments predominantly ranging from 200 to 500 bp (Supplementary Fig. S3 ). ChIP‑qPCR analysis revealed significant enrichment of HIF‑1α at this HRE cluster under hypoxic conditions compared with normoxia (Fig. L, M), indicating direct binding of HIF‑1α to the Piezo1 promoter. Fig. 2 Hypoxia (24 H) induces Piezo1 upregulation and fibrotic transition in HESCs. (A , B , C) The protein and mRNA expression levels of HIF-1α and Piezo1 in HESCs were analyzed by Western blotting and qRT-PCR, following exposure to hypoxia for 0,6, 12, 24, 36, 48 h, and were subsequently quantified. (D , E) The protein levels of the fibrotic markers α-SMA and Collagen I were determined by Western blotting analysis in HESCs following exposure to hypoxia for 0, 6, 12, 24, 36, 48 h. (F , G) Immunofluorescence was employed to compare α-SMA expression in HESCs subjected to different periods of hypoxic culture (Scale bar: 100 μm). (H , I) Flow cytometric analysis was used to determine the apoptosis levels in HESCs under varying hypoxic conditions. (J , K) EdU assays evaluated cell proliferation capacity under different hypoxia conditions (Scale bar: 100 μm). (L) Schematic representation of the human Piezo1 promoter, highlighting the predicted hypoxia response element (HRE) cluster (positions -220 to -180 bp) and the regions amplified by ChIP-qPCR. (M) ChIP-qPCR analysis of HIF-1α binding to the Piezo1 promoter in HESCs exposed to normoxia or hypoxia (1% O₂, 24 h). Enrichment was calculated as percentage of input (%Input). * P < 0.05 ** P < 0.01 *** P < 0.001 **** P < 0.0001 Hypoxia (24 H) induces Piezo1 upregulation and fibrotic transition in HESCs. (A , B , C) The protein and mRNA expression levels of HIF-1α and Piezo1 in HESCs were analyzed by Western blotting and qRT-PCR, following exposure to hypoxia for 0,6, 12, 24, 36, 48 h, and were subsequently quantified. (D , E) The protein levels of the fibrotic markers α-SMA and Collagen I were determined by Western blotting analysis in HESCs following exposure to hypoxia for 0, 6, 12, 24, 36, 48 h. (F , G) Immunofluorescence was employed to compare α-SMA expression in HESCs subjected to different periods of hypoxic culture (Scale bar: 100 μm). (H , I) Flow cytometric analysis was used to determine the apoptosis levels in HESCs under varying hypoxic conditions. (J , K) EdU assays evaluated cell proliferation capacity under different hypoxia conditions (Scale bar: 100 μm). (L) Schematic representation of the human Piezo1 promoter, highlighting the predicted hypoxia response element (HRE) cluster (positions -220 to -180 bp) and the regions amplified by ChIP-qPCR. (M) ChIP-qPCR analysis of HIF-1α binding to the Piezo1 promoter in HESCs exposed to normoxia or hypoxia (1% O₂, 24 h). Enrichment was calculated as percentage of input (%Input). * P < 0.05 ** P < 0.01 *** P < 0.001 **** P < 0.0001 To examine the function of Piezo1 in the pro-fibrotic responses of HESCs under hypoxia, we performed functional interventions by either suppressing Piezo1 expression via shRNA-mediated knockdown or activating it with the specific agonist Yoda1. Based on Western blotting and qRT-PCR analysis, the shRNA achieved a knockdown efficiency of > 70% (Fig.  3 A, B and Supplementary Fig. S4 ), which was considered adequate for subsequent experiments. Western blotting and qRT-PCR analyses confirmed that hypoxia significantly increased Piezo1 expression in HESCs relative to normoxic controls. Furthermore, this upregulation was enhanced by combined treatment with hypoxia and the agonist Yoda1. In contrast, Piezo1 knockdown under hypoxic conditions effectively reversed its induction (Fig.  3 C-E). Western blotting analysis showed that hypoxia upregulated the expression of the fibrosis markers α-SMA and Collagen I. Piezo1 knockdown under hypoxic conditions suppressed this upregulation, while Piezo1 activation with Yoda1 further enhanced the hypoxia-induced expression of both proteins (Fig.  3 F, G). Immunofluorescence (IF) staining further demonstrated that Piezo1 knockdown attenuated α-SMA expression in hypoxic HESCs, whereas Yoda1-mediated activation of Piezo1 enhanced it (Fig.  3 H, I). Flow cytometry revealed no significant alterations in apoptosis across the experimental groups (Supplementary Fig. S5 ), excluding cytotoxicity as a potential confounder and supporting a direct role for Piezo1 in the pro-fibrotic response. The findings demonstrate that hypoxia promotes the pro-fibrotic transformation of HESCs through upregulation of Piezo1, thereby identifying Piezo1 as a key mediator of hypoxia-induced fibrosis in endometriosis. Fig. 3 Hypoxia promotes endometrial fibrosis by upregulating Piezo1. (A , B) HESCs were transfected with lentiviral vectors encoding either a non-targeting control shRNA (sh-NC) or one of two Piezo1-targeting shRNAs (sh-Piezo1-1 and sh-Piezo1-2). Subsequently, Piezo1 knockdown efficiency was analyzed by Western blotting and qRT‑PCR. (C-E) The expression levels of HIF-1α and Piezo1 in HESCs from the NC, Hypoxia, Hypoxia-sh Piezo1, and Hypoxia-Yoda1 groups were quantified using Western blotting and qRT-PCR, followed by quantitative analysis. (F , G) Western blotting was performed to detect and quantify the protein expression of fibrosis markers α-SMA and Collagen I in HESCs from the NC, Hypoxia, Hypoxia-sh Piezo1, and Hypoxia-Yoda1 groups. (H , I) Immunofluorescence staining was performed to detect α-SMA expression levels in HESCs from the NC group, hypoxia group, Hypoxia-sh Piezo1 group, and Hypoxia-Yoda1 group. Representative images were captured and subjected to quantitative analysis (Scale bar: 100 μm). (J , K) The expression of cGAS and STING was assessed by immunohistochemistry in normal endometrial tissue ( n  = 9), eutopic endometrial and ectopic endometrial tissue ( n  = 9). Representative images were acquired and subjected to quantitative analysis (Scale bar: 100 μm). (L-N) The expression levels of cGAS and STING in HESCs were quantified by Western blotting and qRT-PCR across the following experimental groups: NC, Hypoxia, Hypoxia-shPiezo1, and Hypoxia-Yoda1, along with their respective quantification. (O , P) Western blotting was performed to analyze TBK1 and IRF3 phosphorylation in HESCs across the following groups: NC, Hypoxia, Hypoxia-shPiezo1, and Hypoxia-Yoda1. The ratios of phosphorylated to total protein (p-TBK1/TBK1 and p-IRF3/IRF3) were calculated as indicators of pathway activation. *P<0.05; **P<0.01; ***P<0.001; ****P<0.0001 Hypoxia promotes endometrial fibrosis by upregulating Piezo1. (A , B) HESCs were transfected with lentiviral vectors encoding either a non-targeting control shRNA (sh-NC) or one of two Piezo1-targeting shRNAs (sh-Piezo1-1 and sh-Piezo1-2). Subsequently, Piezo1 knockdown efficiency was analyzed by Western blotting and qRT‑PCR. (C-E) The expression levels of HIF-1α and Piezo1 in HESCs from the NC, Hypoxia, Hypoxia-sh Piezo1, and Hypoxia-Yoda1 groups were quantified using Western blotting and qRT-PCR, followed by quantitative analysis. (F , G) Western blotting was performed to detect and quantify the protein expression of fibrosis markers α-SMA and Collagen I in HESCs from the NC, Hypoxia, Hypoxia-sh Piezo1, and Hypoxia-Yoda1 groups. (H , I) Immunofluorescence staining was performed to detect α-SMA expression levels in HESCs from the NC group, hypoxia group, Hypoxia-sh Piezo1 group, and Hypoxia-Yoda1 group. Representative images were captured and subjected to quantitative analysis (Scale bar: 100 μm). (J , K) The expression of cGAS and STING was assessed by immunohistochemistry in normal endometrial tissue ( n  = 9), eutopic endometrial and ectopic endometrial tissue ( n  = 9). Representative images were acquired and subjected to quantitative analysis (Scale bar: 100 μm). (L-N) The expression levels of cGAS and STING in HESCs were quantified by Western blotting and qRT-PCR across the following experimental groups: NC, Hypoxia, Hypoxia-shPiezo1, and Hypoxia-Yoda1, along with their respective quantification. (O , P) Western blotting was performed to analyze TBK1 and IRF3 phosphorylation in HESCs across the following groups: NC, Hypoxia, Hypoxia-shPiezo1, and Hypoxia-Yoda1. The ratios of phosphorylated to total protein (p-TBK1/TBK1 and p-IRF3/IRF3) were calculated as indicators of pathway activation. *P<0.05; **P<0.01; ***P<0.001; ****P<0.0001 The cGAS-STING innate immune pathway, which senses mislocalized double-stranded DNA, has recently gained prominence in fibrosis research [ 49 – 51 ]. Studies demonstrate its activation under hypoxia and a contributory role in fibrotic progression [ 31 ]. This study has observed a significant increase in the expression of both cGAS and STING in ectopic lesions compared to normal endometrial tissue (Fig.  1 F, G). Immunohistochemistry further showed upregulated expression of both cGAS and STING in eutopic and ectopic endometrium from patients with EMs versus normal controls. The elevation was most significant in ectopic lesions (Fig.  3 J, K), indicating activation of the cGAS-STING pathway in EMs. We have previously shown that hypoxia induces Piezo1 upregulation and the ensuing fibrosis in endometriosis. To assess the role of Piezo1 in hypoxia-induced cGAS-STING activation, we performed Western blotting and qRT-PCR analyses. Our results show that hypoxia significantly upregulated the expression of cGAS and STING in HESCs. This upregulation was suppressed by Piezo1 knockdown but further enhanced by the Piezo1 agonist Yoda1 (Fig.  3 L-N). The phosphorylation levels of TBK1 and IRF3, which are critical downstream effectors of the cGAS-STING pathway, were similarly modulated: hypoxia increased p-TBK1/TBK1 and p-IRF3/IRF3 ratios, an effect abolished by Piezo1 knockdown and augmented by Yoda1(Fig.  3 O, P). Thus, these findings demonstrate that hypoxia activates the cGAS-STING pathway in HESCs and establish Piezo1 as a critical regulator of this activation. Mitochondrial dysfunction under hypoxic stress increases membrane permeability, facilitating the release of mitochondrial DNA (mtDNA) into the cytosol [ 52 , 53 ]. This leaked mtDNA serves as a potent endogenous ligand, which is directly recognized by cGAS to activate the cGAS-STING pathway [ 50 , 54 ]. To investigate whether hypoxia triggers this mechanism in HESCs by analyzing mitochondrial morphology via TOM20 immunofluorescence. The results showed that 24 h hypoxia treatment induced significant fragmentation and morphological alteration of the mitochondrial network in HESCs (Fig.  4 A). Mitochondrial function was further assessed by measuring mitochondrial membrane potential using JC-1 staining and flow cytometry. The results showed that hypoxia treatment for 24 h induced a significant depolarization in HESCs (Fig.  4 B-D). Mitochondrial ROS levels, assessed using the MitoSOX probe, were significantly increased under hypoxia conditions (Supplementary Fig. S6 ), further confirming hypoxia‑induced mitochondrial dysfunction. To determine if mitochondrial damage triggers cytosolic mtDNA leakage, we quantified cytosolic mtDNA by immunofluorescence combined with qRT-PCR. Marked leakage of mtDNA into the cytosol was observed in HESCs following 24 h of hypoxia (Fig.  4 E, F). Using ethidium bromide (EtBr) to induce mtDNA-deficient HESCs, it was found that hypoxic activation of the cGAS-STING pathway measured by Western blotting and qRT-PCR was significantly suppressed compared to controls (Fig.  4 G-I), indicating that mtDNA release is critical at the expression level. Supporting this, EtBr pretreatment abrogated the hypoxia-induced increase in p-TBK1/TBK1 and p-IRF3/IRF3 ratios (Fig.  4 J, K), confirming that functional activation of the pathway is dependent on mtDNA. To elucidate the role of mitochondrial reactive oxygen species (ROS) in the fibrotic process, we compared the effects of mitochondrial DNA depletion (EtBr) and antioxidant treatment (NAC) on hypoxia-induced responses. Both interventions significantly reduced mitochondrial ROS levels (Fig.  4 L, M). However, EtBr significantly inhibited hypoxia-induced upregulation of α-SMA, whereas NAC only partially inhibited this process (Fig.  4 N, O). Collectively, these findings indicate that mtDNA integrity is a prerequisite for both hypoxia‑induced ROS production and subsequent fibrosis, whereas mitochondrial ROS serves as a cooperative amplifier rather than an essential driver in this process. In summary, hypoxia promotes EMs fibrosis by inducing mitochondrial damage, cytosolic mtDNA release, and consequent activation of the cGAS-STING signaling pathway. Within this cascade, mtDNA release acts as the essential trigger, whereas mitochondrial ROS functions as a secondary amplifier of the fibrotic response. Fig. 4 Hypoxic activation of the cGAS-STING pathway relies on mtDNA leakage. (A) Mitochondrial morphological alterations in HESCs under varying durations of hypoxia, shown by immunofluorescence (Scale bar: 20 μm). (B) JC-1 staining of mitochondrial membrane potential in HESCs following hypoxic treatment (Scale bar: 25 μm). (C , D) Quantitative flow cytometric analysis of JC-1 staining, showing changes in mitochondrial membrane potential in HESCs under varying durations of hypoxia. (E) Immunofluorescence co-staining for TOM20 (mitochondria) and dsDNA was performed to visualize mtDNA leakage in control versus hypoxic HESCs. Shown are representative images (Scale bar:20 μm). (F) qRT-PCR analysis of cytosolic mtDNA in control and hypoxic HESCs. (G–I) Western blotting and qRT-PCR analysis was performed to detect the expression levels of HIF-1α, cGAS and STING in the NC, EtBr, Hypoxia and Hypoxia-EtBr groups, along with quantitative analysis. (J , K) Western blotting analysis of TBK1 and IRF3 phosphorylation in HESCs from the NC, EtBr, Hypoxia, and Hypoxia-EtBr groups, along with quantitative analysis. (L , M) Mitochondrial ROS levels assessed by MitoSOX staining in HESCs from Normal, Hypoxia, Hypoxia-EtBr and Hypoxia-NAC groups. Shown are representative images and quantification (Scale bar: 20 μm). (N , O) Immunofluorescence showing α-SMA expression in HESCs from the Normal, Hypoxia, Hypoxia-EtBr and Hypoxia-NAC groups. Representative images and quantification are shown (Scale bar: 100 μm). *P<0.05; **P<0.01; ***P<0.001; ****P<0.0001 Hypoxic activation of the cGAS-STING pathway relies on mtDNA leakage. (A) Mitochondrial morphological alterations in HESCs under varying durations of hypoxia, shown by immunofluorescence (Scale bar: 20 μm). (B) JC-1 staining of mitochondrial membrane potential in HESCs following hypoxic treatment (Scale bar: 25 μm). (C , D) Quantitative flow cytometric analysis of JC-1 staining, showing changes in mitochondrial membrane potential in HESCs under varying durations of hypoxia. (E) Immunofluorescence co-staining for TOM20 (mitochondria) and dsDNA was performed to visualize mtDNA leakage in control versus hypoxic HESCs. Shown are representative images (Scale bar:20 μm). (F) qRT-PCR analysis of cytosolic mtDNA in control and hypoxic HESCs. (G–I) Western blotting and qRT-PCR analysis was performed to detect the expression levels of HIF-1α, cGAS and STING in the NC, EtBr, Hypoxia and Hypoxia-EtBr groups, along with quantitative analysis. (J , K) Western blotting analysis of TBK1 and IRF3 phosphorylation in HESCs from the NC, EtBr, Hypoxia, and Hypoxia-EtBr groups, along with quantitative analysis. (L , M) Mitochondrial ROS levels assessed by MitoSOX staining in HESCs from Normal, Hypoxia, Hypoxia-EtBr and Hypoxia-NAC groups. Shown are representative images and quantification (Scale bar: 20 μm). (N , O) Immunofluorescence showing α-SMA expression in HESCs from the Normal, Hypoxia, Hypoxia-EtBr and Hypoxia-NAC groups. Representative images and quantification are shown (Scale bar: 100 μm). *P<0.05; **P<0.01; ***P<0.001; ****P<0.0001 The Piezo1 channel mediates extracellular Ca²⁺ influx, which can induce mitochondrial Ca²⁺ overload, resulting in mitochondrial damage and subsequent mtDNA leakage, which is a well-characterised activator of the cGAS-STING pathway [ 55 , 56 ]. To determine whether Piezo1 promotes fibrosis in HESCs specifically through this “Ca²⁺ overload-mtDNA leakage” axis, we performed a series of mechanistic investigations. To monitor intracellular calcium homeostasis, cytosolic and mitochondrial Ca²⁺ levels were assessed using Fluo-4 AM and Rhod-2 AM, respectively. Hypoxia treatment significantly elevated Ca²⁺ concentrations in both the cytosol and mitochondria. This increase was substantially attenuated by either the Piezo1-specific inhibitor GsMTx4 or the intracellular Ca²⁺ chelator BAPTA-AM (Fig.  5 A), indicating that hypoxia induces Ca²⁺ influx and subsequent mitochondrial Ca²⁺ overload predominantly through Piezo1 activation. To further validate the specificity of GsMTx4, we performed complementary genetic knockdown experiments. Transfection with Piezo1-specific shRNA (shPiezo1) similarly abolished hypoxia-induced Ca²⁺ elevation (Supplementary Fig. S7 A), confirming that the effects of GsMTx4 are indeed mediated through Piezo1 rather than off-target actions. Consistent with these findings, both GsMTx4 and shPiezo1 similarly prevented hypoxia-induced mitochondrial fragmentation, ROS production, mtDNA release, and cGAS-STING pathway activation (Supplementary Fig. S7 B-E). Immunofluorescence analysis of mitochondrial morphology revealed that hypoxia induced marked mitochondrial fragmentation, which was effectively prevented by GsMTx4 treatment (Fig.  5 B), suggesting that Piezo1-mediated Ca²⁺ overload leads to mitochondrial structural damage. To establish the dependence of mtDNA leakage on Piezo1 activation, we assessed cytosolic mtDNA using immunofluorescence and qRT-PCR. Hypoxia-induced mtDNA leakage into the cytosol was effectively abolished by the Piezo1 inhibitor GsMTx4 (Fig.  5 C, D), demonstrating that Piezo1-mediated Ca²⁺ signaling is a critical upstream trigger for mtDNA release in HESCs. To further explore the role of oxidative stress in this process, we assessed mitochondrial ROS production using MitoSOX staining. Hypoxia markedly increased mitochondrial ROS levels, an effect significantly reduced by both GsMTx4 and BAPTA-AM (Fig.  5 E; Supplementary Fig. S8 ). These results indicate that ROS acts merely as an amplifier rather than an initiator. To delineate the relationship between Piezo1 and the cGAS-STING pathway, hypoxic HESCs were treated with the Piezo1 agonist Yoda1, Piezo1 inhibitor GsMTx4, the STING inhibitor C-176 or the cGAS inhibitor RU.521. Western blotting analysis demonstrated that hypoxia significantly upregulated Piezo1, cGAS, and STING protein levels, and increased the phosphorylation of TBK1 and IRF3 (Fig.  5 F), quantitative analysis is shown in the Supplementary figure. S9 . Consistent with these protein changes, qRT-PCR analysis revealed that hypoxia also upregulated Piezo1, cGAS, and STING mRNA levels (Fig.  5 G). Furthermore, immunofluorescence analysis revealed that, consistent with the western blot findings, Yoda1 significantly increased α-SMA expression, whereas GsMTx4, C-176, and RU.521 each markedly suppressed it (Fig.  5 H, I). To distinguish the contribution of inflammation from direct cGAS-STING-mediated fibrosis, we neutralized IL-6, a major pro-inflammatory cytokine induced by STING activation. IL-6 neutralization only partially reversed hypoxia-induced α-SMA upregulation (Fig.  5 H, I), indicating that while the inflammatory response contributes to fibrosis, the cGAS-STING pathway plays a dominant role. Collectively, our findings demonstrate that Piezo1 channel activation acts as a master regulator initiating fibrotic signaling under hypoxia. This is achieved by mediating extracellular Ca²⁺ influx, which triggers mitochondrial Ca²⁺ overload and mtDNA release, thereby activating the cGAS-STING pathway and ultimately drives the fibrotic phenotypic shift in HESCs. Notably, while mtDNA serves as the essential trigger for pathway activation, mitochondrial ROS functions as a downstream amplifier. These results provide a novel mechanistic perspective and establish potential therapeutic targets for fibrosis in endometriosis (EMs). Fig. 5 Piezo1-mediated Ca²⁺ influx promotes endometrial fibrosis via activation of the cGAS-STING pathway. (A) Immunofluorescence images showing cytosolic and mitochondrial Ca²⁺ levels in HESCs under the indicated conditions: Normoxic control (NC), Hypoxia, Hypoxia-GsMTx4, and Hypoxia-BAPTA AM (Scale bar: 100 μm). (B) Immunofluorescence analysis of mitochondrial morphology in HESCs from the NC, Hypoxia, and Hypoxia with GsMTx4 groups (Scale bar: 20 μm). (C) qRT-PCR analysis quantifying the levels of cytoplasmic mtDNA in NC and Hypoxic HESCs. (D) Representative immunofluorescence images of HESCs co-immunostained for the mitochondrial marker TOM20 and dsDNA, showing mtDNA leakage in the NC, Hypoxia, and Hypoxia-GsMTx4 groups (Scale bar: 20 μm). (E) Immunofluorescence staining analysis of the mitochondrial ROS levels in HESC cell (Scale bar: 20 μm). (F , G) Western blotting and qRT-PCR analysis of HIF-1α, Piezo1, cGAS and STING expression in HESCs Quantitative data are presented as mean ± SD. (H , I) Representative immunofluorescence images and quantification of the fibrosis marker α-SMA in HESCs (Scale bar: 100 μm). *P<0.05; **P<0.01; ***P<0.001; ****P<0.0001 Piezo1-mediated Ca²⁺ influx promotes endometrial fibrosis via activation of the cGAS-STING pathway. (A) Immunofluorescence images showing cytosolic and mitochondrial Ca²⁺ levels in HESCs under the indicated conditions: Normoxic control (NC), Hypoxia, Hypoxia-GsMTx4, and Hypoxia-BAPTA AM (Scale bar: 100 μm). (B) Immunofluorescence analysis of mitochondrial morphology in HESCs from the NC, Hypoxia, and Hypoxia with GsMTx4 groups (Scale bar: 20 μm). (C) qRT-PCR analysis quantifying the levels of cytoplasmic mtDNA in NC and Hypoxic HESCs. (D) Representative immunofluorescence images of HESCs co-immunostained for the mitochondrial marker TOM20 and dsDNA, showing mtDNA leakage in the NC, Hypoxia, and Hypoxia-GsMTx4 groups (Scale bar: 20 μm). (E) Immunofluorescence staining analysis of the mitochondrial ROS levels in HESC cell (Scale bar: 20 μm). (F , G) Western blotting and qRT-PCR analysis of HIF-1α, Piezo1, cGAS and STING expression in HESCs Quantitative data are presented as mean ± SD. (H , I) Representative immunofluorescence images and quantification of the fibrosis marker α-SMA in HESCs (Scale bar: 100 μm). *P<0.05; **P<0.01; ***P<0.001; ****P<0.0001

This study included 18 patients who underwent surgical treatment at Wuhan University People’s Hospital. Paired samples of eutopic endometrium and ovarian endometrioma cyst walls were collected from 9 patients undergoing surgery for endometriosis. Control samples ( n  = 9) of normal endometrium were obtained from patients undergoing hysteroscopy for benign conditions such as abnormal uterine bleeding, endometrial thickening, or polyps. All participants met the following inclusion criteria: No exposure to hormonal agents (including oral contraceptives, GnRH agonists, or progestins) for at least three months prior to surgery; Age between 18 and 50 years. Exclusion criteria included: Concurrent benign or malignant gynecological pathologies (e.g., cervical intraepithelial neoplasia or ovarian malignancy); Systemic fibrotic or autoimmune disorders (e.g., systemic lupus erythematosus, rheumatoid arthritis); Age below 18 or above 50 years. The phases of the menstrual cycle were determined based on the date of the last menstrual period, followed by histological confirmation via HE staining of endometrial samples (Supplementary Fig. S1 ). This study was approved by the Ethics Committee of Renmin Hospital of Wuhan University (Approval No: WDRY2026-K011) and was granted a waiver of informed consent. All relevant clinicopathological data, with all personally identifiable information removed to protect patient privacy. We are grateful to Professor Qingzhen Xie at Renmin Hospital of Wuhan University for generously providing the immortalized human endometrial stromal cell line (HESC). HESCs were grown in DMEM/F12 (Hyclone) containing 10% charcoal-stripped FBS (SERANA) and 1% penicillin/streptomycin/gentamicin (Biosharp), at 37 °C in a humidified 5% CO₂ atmosphere. When cells reached 60 ~ 70% confluency, HESCs were placed in a hypoxia chamber (1% O₂, 5% CO₂, balanced N₂) and cultured for the indicated durations (0, 6, 12, 24, 36, 48 h). Based on preliminary experiments, a 24 h hypoxia exposure was selected for subsequent studies. Cells were harvested at respective time points for downstream analysis. At 60-70% confluency, cells were treated for 24 h with the specified compounds: the Piezo1 agonist Yoda1 (5 µM; MCE, HY-18723); the Piezo1 inhibitor GsMTx4 (5 µM; MCE, HY-P1410); Ethidium bromide (0.2 µg/ml; Macklin, 1239-45-8) to deplete mitochondrial DNA; the calcium chelator BAPTA-AM (4 µM; MCE, HY-100168); the cGAS inhibitor RU.521(2.5µM, MCE, HY-114180 ); the STING inhibitor C-176 (2.5 µM; MCE, HY-112906); the anti-IL6 antibody (2.5 µg/ml, Sigma-Aldrich, I7901); or the ROS inhibitor NAC (2.0mM, MCE, HY-B0215). Piezo1 was knocked down in HESC using short hairpin RNA (shRNA). The targeting sequences were as follows: shRNA1, 5′-GCGTCATCATCGTGTGTAAGA-3′; shRNA2, 5′-GCTACGAGAACAAGCCCTACT-3′, and the non-targeting shRNA was 5′-TTCTCCGAACGTGTCACGT-3′ as control. Lentiviral particles for Piezo1‑targeting shRNAs (shRNA1 and shRNA2) and a non‑targeting control were generated using the GV493 vector (obtained from Shanghai GeneChem Co.,Ltd), with titers exceeding 1 × 10⁷ TU/mL. HESCs in 35-mm dishes were transduced with 500 µl of lentivirus and selected with 1 µg/ml puromycin to establish stable Piezo1-knockdown (shRNA1 and shRNA2) and control shRNA cell lines. Tissue samples of normal endometrium, eutopic endometrium and ectopic endometrium from patients with endometriosis were collected. Tissue samples were fixed in 4% paraformaldehyde, embedded in paraffin, and sectioned at 5 μm. The sections were then dewaxed and hydrated through a graded ethanol series. For histological evaluation, sections were stained with hematoxylin and eosin (HE) to assess overall tissue morphology, including glandular architecture, stromal cellularity, and inflammatory cell infiltration. Collagen content was evaluated using Sirius red (Solarbio, G1473) and Masson’s trichrome staining (Solarbio, G1340) according to the manufacturers’ instructions, for subtype differentiation and total collagen visualization, respectively. Immunohistochemistry (IHC) was performed on deparaffinized and hydrated tissue sections. After antigen retrieval in 10 mM sodium citrate buffer (pH 6.0) and quenching of endogenous peroxidase, sections were blocked with 5% BSA for 30 min. They were then incubated overnight at 4 °C with antibodies against HIF‑1α (1:200; Wanleibio, WL01607), Piezo1 (1:300; Proteintech, 15939‑1‑AP), cGAS (1:500; Proteintech, 26416‑1‑AP), and STING (1:4000; Proteintech, 19851‑1‑AP). Subsequently, sections were incubated with an HRP-conjugated secondary antibody (1 h, RT), developed with DAB (ZSGB-BIO), and counterstained with hematoxylin (Biosharp, BL702A) to visualize the nucleus. The stained areas were quantified from digitized images using Image J. Immunofluorescence (IF) was performed on both paraffin-embedded tissues and cultured cells. For tissue Sect.  (5 μm), standard processing included deparaffinization, hydration, and antigen retrieval in citrate buffer (pH 6.0). For cells, the procedure involved PBS washing, fixation in 4% paraformaldehyde, and three subsequent PBS washes. To reduce nonspecific binding and facilitate antibody penetration, samples were permeabilized with 0.3% Triton X-100 for 30 min, followed by incubation in blocking buffer (1% BSA in PBS) for 30 min, with both steps carried out at room temperature. Tissue and cell samples were incubated separately with specific primary antibodies overnight at 4 °C. Tissue sections were labeled with antibodies against Piezo1 (1:300; Proteintech, 15939-1-AP), α-SMA (1:400; Proteintech, 67735-1-Ig), and Collagen I (1:300; Proteintech, 67288-1-Ig). Cultured cells were labeled with antibodies against α-SMA (1:400; Proteintech), TOM20 (1:1000; HuaBio, ET1609-25), and dsDNA (1:50; Sigma-Aldrich, ZMS1047). Samples were then incubated with fluorescent secondary antibodies (1 h, RT) and DAPI for nuclear counterstaining. Fluorescence in tissue sections and cells was visualized using an Olympus BX51 microscope with a 20× objective. For higher-resolution imaging of mitochondrial structures, a super-resolution confocal laser scanning microscope (Olympus FV3000) equipped with a 60× oil-immersion objective was employed. Subsequently, Image J software was used to perform quantitative analysis of fluorescence intensity or area for specific markers. The proliferation of HESCs was evaluated using the E-Click™ EdU Imaging Kit (Elabscience, E-CK-A377) according to the manufacturer’s protocol. Briefly, immortalized HESCs were cultured in DMEM/F12 medium supplemented with 10% FBS and 10 µM EdU (5-ethynyl-2′-deoxyuridine) for 24 h. Following the EdU incorporation period, cells were fixed, permeabilized, and subjected to the click reaction for fluorescence detection. EdU-positive cells were visualized and quantified using an Olympus BX5 optical microscope (20×), with analysis performed using Image J. To assess apoptosis and mitochondrial membrane potential, hypoxia-exposed HESCs were collected using EDTA-free trypsin (Biosharp, BL527A). For apoptosis, cells were stained with Annexin V-FITC and PI (Solarbio, CA1020) per the manual. For membrane potential, cells were stained with JC-1 (Beyotime, C2005; 30 min at 37 °C in the dark). All samples were analyzed on a Beckman Coulter CytoFlex flow cytometer using CytExpert software. Mitochondrial membrane potential (ΔΨm) was evaluated using JC-1 fluorescence. Hypoxia-treated HESCs were incubated with JC-1 working solution for 30 min at 37 °C in the dark, followed by nuclear counterstaining with Hoechst. Cells were then imaged using a super‑resolution confocal microscope (Olympus FV3000, 60× oil objective). The shift from red JC-1 aggregates (indicative of high ΔΨm) to green monomers (indicative of low ΔΨm) was quantified to evaluate mitochondrial health. For detection of mitochondrial reactive oxygen species, cells were subjected to the indicated treatment and incubated with MitoSOX™ Red (2 µM) for 30 min at 37 °C in the dark according to the manufacturer’s protocol (Mitochondrial Superoxide Assay Kit). After washing twice with PBS, cells were counterstained with Hoechst and imaged using a super-resolution confocal laser scanning microscope (Olympus FV3000) equipped with a 60× oil-immersion objective. Fluorescence intensity and area were quantified using Image J software. Intracellular and mitochondrial Ca²⁺ were assessed in parallel via dual-probe staining: one set of cells with Fluo‑4 AM (5 µM; 40 min, RT) for cytosolic signals; another set sequentially with Rhod‑2 AM (2 µM; 30 min, RT) and MitoTracker Green (100 nM; 15 min, 37 °C) for mitochondrial signals. Live-cell imaging (Olympus IX71, 20×) was performed immediately post-staining for fluorescence quantification. To quantify the release of mitochondrial DNA (mtDNA) into the cytosol, cytoplasmic fractions were first isolated using a Mitochondria Isolation Kit (Proteintech, PK10016). Total nucleic acids were then extracted from these fractions using an RNA/DNA Extraction Kit (Beyotime). Finally, relative mtDNA (ND1) and nDNA (18S rRNA) levels in the cytosol were quantified by qRT-PCR with specific primers (Supplementary Table 1 ). Cells and tissues were lysed in RIPA buffer. Total protein concentration was quantified using a bicinchoninic acid (BCA) assay kit (Elabscience, E‑BC‑K318‑M), and concentrations were normalized before further analysis. Following centrifugation, 30 µg of protein lysate was separated by 8% or 10% sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) and subsequently transferred onto a polyvinylidene difluoride (PVDF) membrane. The membrane was blocked with 5% non-fat milk for 2 h at room temperature and then incubated overnight at 4 °C with primary antibodies against HIF-1α, Piezo1, cGAS, STING, α-SMA, and COL1. GAPDH served as the control. After overnight incubation, the membrane was washed with TBST and then incubated for 1 h at room temperature with an HRP-conjugated goat anti-rabbit IgG secondary antibody (Servicebio, GB23303) diluted in the blocking buffer (5% non-fat milk). Immunoreactive bands were visualized using enhanced chemiluminescence (ECL) and imaged with a ChemiDoc™ Touch Imaging System (Bio-Rad). Following total RNA extraction using TRIzol reagent (Vazyme) and complementary DNA (cDNA) synthesis with a commercial kit (Yeasen, 11141ES60), quantitative reverse transcription PCR (qRT‑PCR) was performed. Target genes (HIF‑1α, Piezo1, cGAS, STING) and the reference gene (GAPDH) were amplified using SYBR Green Master Mix (Yeasen, 11185ES08) and gene‑specific primers (Sangon Biotech; sequences provided in Supplementary Table 1 ). Expression data were analyzed using the 2^(-ΔΔCt) method. Chromatin immunoprecipitation (ChIP) was performed using the ChIP Assay Kit (Beyotime, P2078). HESCs were cross-linked with 1% formaldehyde, and chromatin was sheared to 200-500 bp fragments. Lysates were incubated overnight at 4 °C with 2 µg of anti-HIF-1α antibody (Proteintech, 20960-1-AP) or normal rabbit IgG (Cell Signaling Technology, 2729) as a negative control. DNA was purified and analyzed by qPCR using primers specific for the predicted hypoxia response element (HRE) cluster within the Piezo1 promoter (positions - 220 to -180 bp) and a non-HRE region as a negative control. Enrichment was calculated as percentage of input (%Input). Primer sequences are provided in Supplementary Table 1 . All data are presented as the mean ± standard deviation (SD) from at least three independent experiments. Differences between two groups were analyzed by t-test, while comparisons among multiple groups were evaluated using one‑way analysis of variance (ANOVA) followed by Tukey’s post hoc test (analyzed with SPSS, version 21.0). Figures were processed and assembled using Adobe Photoshop 2019, GraphPad Prism (version 9.0), and Adobe Illustrator 2024. P  < 0.05 was considered statistically significant.

In conclusion, this study innovatively reveals the key role of the mechanosensitive ion channel Piezo1 within the hypoxic microenvironment of EMs. We define a novel Piezo1‑mediated signaling axis, which mechanistically links hypoxic sensing to mitochondrial stress, activation of the cGAS-STING innate immune pathway, and subsequent extracellular matrix deposition, thereby providing a critical molecular explanation for how hypoxia drives fibrogenesis in EMs. Notably, mitochondrial ROS acts as an amplifier rather than an essential driver, and cGAS‑STING promotes fibrosis through both inflammatory and direct mechanisms. However, this study has certain limitations. The precise transcriptional co-regulators of HIF-1α-mediated Piezo1 upregulation remain to be explored. The interplay between hypoxia and other microenvironmental factors, such as mechanical stress or inflammation, warrants further investigation. In addition, while sustained hypoxia was employed in the present in vitro model to ensure experimental consistency, endometriotic lesions may experience intermittent hypoxia in vivo, for instance during cyclic bleeding and tissue repair. Future studies comparing continuous versus intermittent hypoxia will be valuable for a more comprehensive understanding of the dynamic hypoxic microenvironment. The current study primarily relies on in vitro models; future in vivo validation is required. Finally, whether this signaling axis operates in other cell types within the endometriotic microenvironment remains uncertain. Despite these limitations, we propose that targeting Piezo1, mitochondrial calcium homeostasis or the cGAS-STING pathway may offer a promising therapeutic strategy for intervening in the fibrotic process of EMs.

Endometriosis (EMs) is a common gynecological disorder characterized primarily by chronic pelvic pain and infertility. Its high recurrence rate poses a substantial burden on the physical and psychological health of women worldwide [ 57 , 58 , 16 ]. Although considerable progress has been made in understanding its etiology and improving clinical management, the precise pathogenesis of EMs remains incompletely elucidated, thereby hindering the development of curative therapies. Fibrosis is a hallmark pathological feature of endometriosis, characterized by the aberrant activation of myofibroblasts and excessive deposition of extracellular matrix (ECM) within ectopic lesions. This fibrotic process thereby critically contributes to lesion progression, adhesion formation, and chronic pain [ 59 , 60 ]. Recent studies have confirmed that ectopic lesions in endometriosis exist within a chronic hypoxic microenvironment, which serves as a critical driver of fibrotic progression [ 26 , 61 – 63 ]. Yet, the precise molecular mechanisms that enable cells to sense hypoxia and transduced into pro-fibrotic responses remain elusive. This study delineates a complete Piezo1/Ca²⁺/mtDNA/cGAS-STING axis that links hypoxia sensing to fibrosis in endometriosis. We show that hypoxia upregulates Piezo1 expression in a HIF-1α-dependent manner, mediating Ca²⁺ influx, which triggers mitochondrial Ca²⁺ overload and dysfunction characterized by fragmentation, depolarization, and increased reactive oxygen species (ROS) production, thereby causing mtDNA leakage into the cytosol. Cytosolic mtDNA then acts as an endogenous danger signal, activating the cGAS-STING pathway, ultimately promoting fibrotic transdifferentiation in HESCs. Piezo1 is a non-selective cation channel that can be activated by mechanical forces and chemical signals, with broad expression across numerous tissues and cell types [ 36 , 44 , 64 ]. It has attracted considerable interest for its roles in various physiological and pathological processes. Emerging evidence positions Piezo1 as a key mediator of fibrotic signaling, contributing to pathological tissue remodeling in multiple organs including the skin [ 65 , 66 ], lungs [ 67 , 68 ], heart [ 69 ], and kidneys [ 70 , 71 ], thereby underscoring its broad relevance in fibrosis pathogenesis. Furthermore, in diverse tumor types (e.g., hepatocellular carcinoma [ 72 ], breast cancer [ 73 ], and glioblastoma [ 74 ]), hypoxia has been documented to induce the up-regulation of Piezo1 expression in an HIF-1α-dependent manner and to be involved in processes such as tumor cell invasion, immune regulation, and matrix remodeling. Nevertheless, whether the mechanism of Piezo1 up-regulation is consistent with the findings of this study remains to be explored. Analysis of patient tissues showed that Piezo1 was significantly upregulated in ectopic lesions, and its expression level correlated with fibrotic severity, suggesting its involvement in EMs progression. Mechanistic studies revealed that 24 h of hypoxia induced both Piezo1 and the fibrotic marker α-SMA in HESCs, thereby defining a functional “hypoxia-Piezo1-fibrosis” axis in EMs pathogenesis. Notably, genetic knockdown (shPiezo1) and pharmacological inhibition (GsMTx4) of Piezo1 yielded comparable effects, validating the specificity of the inhibitor and reinforcing the central role of Piezo1 in this cascade. The cGAS-STING pathway serves as a critical immune surveillance system that detects aberrant cytosolic DNA. Previous studies have shown it to be upregulated in EMs, where it enhances HESC invasiveness and migration by upregulating autophagy [ 41 , 42 ]. However, the mechanisms underlying the activation of the cGAS-STING signaling pathway in HESCs warrant further investigation. Aberrant activation of the cGAS-STING pathway was observed in endometriotic lesions, which correlated with Piezo1 expression. Piezo1 was identified as the essential transducer of hypoxic signaling to cGAS-STING. Activation of Piezo1 recapitulated the hypoxic effect by stimulating cGAS-STING and downstream pro-fibrotic factors, while knockdown of Piezo1 or inhibition of its channel function with GsMTx4 abolished the response. Notably, inhibition of STING or cGAS had no impact on Piezo1 expression, but reversed its pro-fibrotic outcomes, definitively confirming Piezo1 as the dominant upstream regulator. Given that mitochondrial damage bridges Piezo1 activation and cGAS-STING signaling, we focused on mtDNA leakage. Hypoxia caused mitochondrial injury, depolarization, and ROS production in HESCs, triggering the release of mtDNA into the cytosol. To substantiate the role of mtDNA, we employed EtBr to induce mtDNA deficiency. This intervention prevented both cGAS-STING activation and subsequent fibrosis, confirming mtDNA as an indispensable signal for this pathway. Notably, while the antioxidant NAC effectively scavenged hypoxia-induced ROS, it only partially attenuated fibrosis; in contrast, EtBr-mediated mtDNA depletion completely abrogated both ROS production and fibrosis. These findings indicate that mtDNA integrity serves as an upstream prerequisite for both ROS generation and fibrotic signaling, whereas ROS functions as a cooperative amplifier rather than an essential driver. How does Piezo1 provoke mtDNA leakage? Pinpointing Ca²⁺ influx as the key mechanism, we found that Piezo1 activation rapidly elevates cytosolic and mitochondrial Ca²⁺ levels. Chelation of this Ca²⁺ not only reversed the resulting mitochondrial overload and depolarization but also blocked all subsequent events: mtDNA leakage, cGAS-STING pathway activation, and fibrosis. Finally, to distinguish the contribution of inflammation from direct fibrotic signaling, we neutralized IL 6 signaling. While this intervention partially alleviated hypoxia induced fibrosis, it did not affect upstream cGAS-STING activation, indicating that inflammatory mediators contribute to the fibrotic outcome but are not required for pathway initiation. This observation suggests the existence of direct, inflammation independent fibrotic signaling downstream of cGAS-STING. Overall, our study highlights the important role of Piezo1 in the pathogenesis of EMs fibrosis, suggesting that the mechanism by which Piezo1/Ca²⁺ influx/mitochondrial damage/mtDNA leakage/cGAS-STING activation contributes to enhance EMs fibrosis.

Endometriosis is a common chronic gynecological disorder characterized by viable endometrial-like tissue outside the uterine cavity, and clinically associated with chronic pelvic pain and infertility [ 1 ]. The ovaries, uterosacral ligament and pelvic peritoneum are the most frequently affected anatomical locations in this condition [ 2 , 3 ]. Endometriosis affects an estimated 10% of reproductive-aged women globally; however, its precise etiology and pathophysiology remain incompletely understood [ 2 , 4 – 6 ]. The current treatment for endometriosis primarily involves hormonal modulation (such as GnRH agonists and oral contraceptives) and surgical intervention [ 7 – 10 ]. Hormonal therapies suppress ectopic endometrial growth by inducing a hypoestrogenic state. However, prolonged use can cause adverse effects related to estrogen deficiency, including hot flashes and osteoporosis [ 11 – 13 ]. Surgical intervention seeks to directly excise or ablate endometriotic lesions. Nevertheless, high postoperative recurrence rates and the potential for iatrogenic ovarian damage remain significant limitations of this approach [ 14 – 16 ]. Consequently, elucidating the core pathogenic mechanisms of endometriosis and discovering novel therapeutic targets constitute pressing clinical priorities. The pathogenesis of endometriosis involves complex, interconnected pathophysiological processes, among which fibrosis, angiogenesis and chronic inflammation are established inherent characteristics of EMs [ 17 – 20 ]. Fibrosis, a pathological process marked by excessive extracellular matrix (ECM) deposition, is a common feature across all endometriosis (EMs) subtypes [ 21 , 22 ]. Accumulating evidence demonstrates that fibrosis in endometriosis (EMs) is associated not only with chronic pelvic pain but also with impaired ovarian function, which can lead to clinical sequelae such as infertility [ 17 , 23 ]. Endometriosis lesions, which are estrogen-dependent and inflammatory in nature, often reside in a chronically hypoxic microenvironment. This condition is largely sustained by aberrant angiogenesis and increased aerobic metabolism within the lesions [ 24 ]. During the initial establishment of endometriosis, retrograde endometrial tissue is confronted with a hypoxic microenvironment, a condition that may paradoxically promote its survival and implantation [ 25 ]. Hypoxia serves as a pivotal mediator in endometriosis, fostering cell adhesion, disrupting hormone synthesis, promoting inflammation, and stimulating angiogenesis-processes that collectively propel disease progression and associated clinical manifestations [ 26 , 27 , 25 ]. While hypoxia is an established mediator of fibrosis in organs such as the heart [ 28 ], lungs [ 29 ], liver [ 30 ] and kidneys [ 31 ], its potential role and regulatory mechanisms in driving fibrosis in endometriosis remain to be elucidated. Recent research identifies mechanosensitive ion channels, notably Piezo1, as critical contributors to tissue fibrosis [ 32 , 33 ]. As a pivotal molecular sensor, Piezo1 translates extracellular mechanical and biochemical signals into cellular responses, a mechanism implicated in fibrotic diseases across multiple organs such as the liver [ 34 ], lungs [ 35 ] and kidneys [ 36 , 37 ]. The activation of piezo1 has been shown in previous studies to cause an influx of Ca²⁺, which then leads to the deposition of abnormal extracellular matrix (ECM) and promotes fibrotic progression [ 38 , 39 ]. However, the function of Piezo1 in the fibrotic process of EMs remains unclear at present. The cGAS-STING signaling pathway, functioning as a cytosolic DNA sensor and a central hub in immune responses, upregulates the expression of pro-fibrotic factors upon sensing cellular stress or damage [ 31 , 40 ]. While emerging evidence indicates that the cGAS-STING pathway is activated in endometriotic lesions and promotes disease progression through inflammation, senescence and invasion [ 41 , 42 ], its contribution to fibrosis remains unclear. This study suggests that Piezo1 might contribute to the fibrotic process in EMs. We found that Piezo1 is upregulated in endometriotic lesions. Hypoxia increases Piezo1 expression in HESCs, which triggers Ca²⁺ influx and causes mitochondrial calcium overload and damage. This leads to leakage of mitochondrial DNA (mtDNA) into the cytosol, activating the cGAS-STING pathway and resulting in fibroblast activation and pathological extracellular matrix accumulation. Knocking down Piezo1 or pharmacological inhibition of Piezo1, along with inhibiting the cGAS-STING signaling pathway significantly suppresses fibroblast activation and extracellular matrix accumulation, thereby alleviating the fibrotic process. These results suggest that Piezo1 and cGAS-STING are promising therapeutic targets for anti-fibrotic strategies in the treatment of endometriosis.

Below is the link to the electronic supplementary material. Supplementary Material 1 Supplementary Material 1 Supplementary Material 2 Supplementary Material 2