{"paper_id":"0bfd47cf-916a-4566-8cb2-ea64644fe621","body_text":"Orai1 is required for Ca2+-dependent plasma membrane repair and \nmechanoadaptation \nHaitao Luan1, Azize Cerci2, Timothy A. Bielecki1, Bhopal C. Mohapatra2,3, Santosh \nShrestha1, William Wu2, Matthew D. Storck1, Lynette M. Smith4, Kenneth A. \nStauderman5, Donald W. Coulter3,6, Anupam Kotwal3,7, Jixin Dong1,3, Jung Yul Lim8, \nVimla Band2,3*, Subramanian P . Ramanathan2,3*, Hamid Band1,2,3* \n1Eppley Institute for Research in Cancer, University of Nebraska Medical Center, \nOmaha, NE; 2Department of Genetics, Cell Biology & Anatomy, College of Medicine, \nUniversity of Nebraska Medical Center, Omaha, NE; 3Fred & Pamela Buffett Cancer \nCenter, University of Nebraska Medical Center, Omaha, NE; 4Department of \nBiostatistics, College of Public Health, University of Nebraska Medical Center, Omaha, \nNE; 5CalciMedica, Inc., La Jolla, CA; 6Division of Pediatric Hematology/Oncology, \nDepartment of Pediatrics, University of Nebraska Medical Center, Omaha, NE; 7Division \nof Endocrinology and Metabolism, Department of Medicine, University of Nebraska \nMedical Center, Omaha, NE; and \n8Department of Mechanical & Materials Engineering, \nCollege of Engineering, University of Nebraska-Lincoln, NE. \nLead Contact: Hamid Band (hband@unmc.edu) \n*Senior Authors/Co-Corresponding Authors: HB (hband@unmc.edu) ; SPR \n( sramanathan@unmc.edu) ; VB (vband@unmc.edu) ; \nRunning Title: EHD2-Orai1 in plasma membrane repair and mechanoadaptation \n  \n(which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. \nThe copyright holder for this preprintthis version posted May 15, 2026. ; https://doi.org/10.64898/2026.05.13.724989doi: bioRxiv preprint \n\nSUMMARY: \nCa2+-dependent repair of plasma membrane breaches is essential for animal cell \nviability. An initial passive influx of extracellular Ca2+ triggers the formation of a protein \nplug that rapidly seals breaches. However, the mechanism of extracellular Ca2+ \nrequirement for subsequent repair remains undefined. EHD2 protein stabilizes the \nplasma membrane caveolae, which sustain membrane repair, and maintains high \nsurface levels of the caveolae-resident Ca\n2+ channel Orai1. We establish the \nrequirement of both Orai1 and EHD2 for repair of plasma membrane lesions induced by \nmechanical injury or by a model bacterial pore-forming toxin. We demonstrate rapid \nEHD2 recruitment and Orai1-mediated Ca\n2+ entry at plasma membrane sites of \nlocalized mechanical stimulus, the latter requiring EHD2 and CAV1. EHD2 and Orai1 \nare necessary for mechanosensitive YAP/TAZ-TEAD activation and positive feedback \nfor CAV1 expression that promotes membrane repair. Our studies establish EHD2 and \nOrai1 as novel components of mammalian plasma membrane repair and \nmechanoadaptation.  \n(which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. \nThe copyright holder for this preprintthis version posted May 15, 2026. ; https://doi.org/10.64898/2026.05.13.724989doi: bioRxiv preprint \n\nIntroduction: \nThe ability to sense and respond to changes in mechanical properties of the \nextracellular microenvironment is a fundamental feature of cellular life during \ndevelopment, homeostasis and diseases (Hannezo and Heisenberg 2019, Narain, \nMuncie-Vasic et al. 2025). Mechanosensing and mechanotransduction are particularly \ncritical for cell types physiologically exposed to dynamic changes in external forces, \nsuch as myofibers in skeletal and cardiac muscle exposed to contraction-relaxation \ncycles, endothelial cells exposed to stretch and shear forces from changing pressure \nand flow, and lung alveolar cells exposed to cyclical compression during respiration \n(Martino, Perestrelo et al. 2018). Metastatic tumor cells must similarly navigate through \nstiff microenvironments during extravasation from the primary tumor, intravasation at \nmetastatic sites and while exposed to flow-associated shear forces in circulation \n(Narain, Muncie-Vasic et al. 2025). Adaptive mechanisms that allow such cell types to \nwithstand harsh mechanical environments and swiftly repair damaged plasma \nmembrane are therefore of significant interest in understanding physiological systems \nand pathological states. \nCell types physiologically exposed to high mechanical stress to the plasma membrane, \nincluding skeletal and cardiac myocytes, endothelial cells and fibroblasts, show an \nabundance of caveolae, 50-80 nm diameter cholesterol- and glycosphingolipid-rich \nplasma membrane invaginations scaffolded on the cytoplasmic side by Caveolin and \nCavin proteins (Parton 2018, Sotodosos-Alonso, Pulgarin-Alfaro et al. 2023). Studies \nusing applied external forces and hypo-osmotic conditions have shown a key role for \ncaveolae in protecting the plasma membrane from mechanical force-induced breaches \n(which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. \nThe copyright holder for this preprintthis version posted May 15, 2026. ; https://doi.org/10.64898/2026.05.13.724989doi: bioRxiv preprint \n\nand their repair (Sinha, Koster et al. 2011). Caveolae-mediated endocytosis also plays a \nkey role in repairing plasma membrane pores formed by bacterial pore-forming toxins, \nby facilitating their lysosomal degradation (Corrotte, Fernandes et al. 2012, Corrotte, \nAlmeida et al. 2013). More recent studies have validated the physiological \nmechanoprotective role of caveolae in vivo. CAV1 knockout studies show that intact \ncaveolae are required for the integrity of microvascular endothelium, acting against \nmechanical rupture induced by increased cardiac output (Cheng, Mendoza-Topaz et al. \n2015). Loss of caveolae upon Caveolin or Cavin knockdown was also found to cause \nthe collapse of vacuolated cells in Zebrafish notochord under mechanical strain of \nlocomotion (Garcia, Bagwell et al. 2017). Besides their mechanoprotective roles, \ncaveolae function as hubs for cell signaling, ion and nutrient transport, and receptor \nendocytic traffic (Shvets, Ludwig et al. 2014). \nCaveolae respond to plasma membrane tension by flattening to relieve strain (Sinha, \nKoster et al. 2011), and the related CAV1-containing dolines respond by activating the \nYAP/TAZ-TEAD pathway (Moreno-Vicente, Pavon et al. 2018, Lolo, Walani et al. 2023). \nThe latter mediate positive feedback through TEAD-dependent induction of CAV1 and \nCAVIN1 gene expression to sustain caveolae (Dupont, Morsut et al. 2011, Rausch, \nBostrom et al. 2019, Lolo, Walani et al. 2023). The contribution of caveolae in the repair \nof plasma membrane injuries has primarily focused on their role as providers of \nmembrane needed to plug the plasma membrane breaches, either through localized \nrearrangements or through endocytic/exocytic processes (Sinha, Koster et al. 2011, \nCorrotte, Fernandes et al. 2012, Corrotte, Almeida et al. 2013, Cheng, Mendoza-Topaz \net al. 2015, Garcia, Bagwell et al. 2017, Stefl, Takamiya et al. 2024). In contrast, any \n(which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. \nThe copyright holder for this preprintthis version posted May 15, 2026. ; https://doi.org/10.64898/2026.05.13.724989doi: bioRxiv preprint \n\nroles of caveolae in regulating signaling mechanisms critical for membrane repair are \nnot well-defined. \nWhile Caveolins and Cavins are the structurally required elements of caveolae, \naccessory proteins localized to caveolae are known to regulate their dynamics at the \nplasma membrane (Parton 2018, Sotodosos-Alonso, Pulgarin-Alfaro et al. 2023) and \nthus are candidates to modulate caveolae function in mechanosensing and \nmechanoprotection. The EPS15 Homology Domain containing protein 2 (EHD2) \nlocalizes to and stabilizes the plasma membrane caveolae (Moren, Shah et al. 2012, \nStoeber, Stoeck et al. 2012, Hoernke, Mohan et al. 2017, Yeow, Howard et al. 2017), \nand has been shown to rapidly accumulate at sites of laser-induced plasma membrane \nlesions in skeletal muscle models, localizing to the shoulder region of the membrane \nrepair cap together with dysferlin (DYSF) (Marg, Schoewel et al. 2012, Demonbreun, \nQuattrocelli et al. 2016). DYSF and its related family members are known to be involved \nin plasma membrane repair (Demonbreun and McNally 2016). EHD2 and DYSF interact \nphysically and were required for myotube fusion (Posey, Pytel et al. 2011). Notably, \nhypotonic stress-induced increase in plasma membrane tension was found to induce \nrapid translocation of EHD2 into the nucleus and transcriptomic analyses identified \nEHD2-dependent changes in gene expression (Torrino, Shen et al. 2018), supporting a \nmechanosensitive role of EHD2. Further, NIH-3T3 cells with combined EHD1, 2 and 4 \nKO, which reduced the cell surface caveolae reservoir, were found to be vulnerable to \nPM rupture upon prolonged cyclical stretch (Yeow, Howard et al. 2017). Together, these \nfindings suggest that EHD2 may play a functional role in plasma membrane repair. \n(which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. \nThe copyright holder for this preprintthis version posted May 15, 2026. ; https://doi.org/10.64898/2026.05.13.724989doi: bioRxiv preprint \n\nWhether EHD2 is indeed involved in plasma membrane repair and the mechanisms of \nsuch a role are currently unknown. \nOur previous studies of the functional role of EHD2 in breast cancer revealed that its \noverexpression, found in ~40% of all patients and a majority of HER2+ and triple-\nnegative (TNBC) subtypes, is associated with shorter patient survival and propensity for \nmetastasis (Luan, Bielecki et al. 2023). Knockdown and knockout analyses in TNBC cell \nmodels established a pro-tumorigenic and pro-metastatic role of EHD2 (Luan, Bielecki \net al. 2023). Mechanistically, we showed that EHD2 was critical to sustain high plasma \nmembrane levels of Orai1 (Luan, Bielecki et al. 2023), a Ca\n2+ channel required for \nstore-operated calcium entry (SOCE).  SOCE is a conserved molecular process in \nwhich the endoplasmic reticulum (ER) Ca2+ depletion induces a conformational change \nin ER Ca2+ sensor STIM1 to promote its translocation to the ER-plasma membrane (ER-\nPM) contact sites where it binds to and activates Orai1 (Ong, Subedi et al. 2019, Lewis \n2020). Orai1-mediated Ca\n2+ entry promotes Ca2+-dependent signaling and helps refill \nthe depleted ER stores to protect against unfolded protein response (Elaib, Saller et al. \n2016, van Vliet, Giordano et al. 2017). Orai1 is known to reside in and functionally \nrequire the cholesterol-rich and CAV1-containing plasma membrane microdomains \n(Sathish, Abcejo et al. 2012, Chantôme, Potier-Cartereau et al. 2013, Jardin and \nRosado 2016, Bohorquez-Hernandez, Gratton et al. 2017). However, whether Orai1-\nmediated Ca\n2+ entry has any role in caveolae-dependent mechanosensing or \nmechanoprotection is unknown. \nA potential role of the EHD2-Orai1 axis in caveolae-dependent mechanoprotection is a \nquestion of broad interest as plasma membrane repair across species is well-\n(which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. \nThe copyright holder for this preprintthis version posted May 15, 2026. ; https://doi.org/10.64898/2026.05.13.724989doi: bioRxiv preprint \n\nestablished to require extracellular Ca2+ entry and to be carried out by Ca2+-dependent \nproteins (Cheng, Zhang et al. 2015, Cooper and McNeil 2015, Demonbreun and \nMcNally 2016, Andrews and Corrotte 2018, Horn and Jaiswal 2018). Based on studies \nin model organisms, it is widely accepted that the repair process is initiated by passive \nflow of Ca\n2+ through the breached plasma membrane down the steep concentration \ngradient from the extracellular space (millimolar Ca2+) to the cytoplasm (sub-micromolar \nCa2+) (Cheng, Zhang et al. 2015, Cooper and McNeil 2015, Demonbreun and McNally \n2016, Andrews and Corrotte 2018, Horn and Jaiswal 2018). It is also well-established \nthat the initial sealing of plasma membrane injuries occurs within seconds, but that \nsubsequent repair, which lasts for an extended duration, continues to be Ca2+-\ndependent even though the passive flow of extracellular Ca2+ has ceased (Cheng, \nZhang et al. 2015, Cooper and McNeil 2015, Demonbreun and McNally 2016, Andrews \nand Corrotte 2018, Horn and Jaiswal 2018). These late repair steps include the \nessential roles of Ca\n2+-dependent proteins and Ca2+-dependent movement of exocytic \nand endocytic membrane vesicles that help restore the plasma membrane. The sources \nof Ca2+ required to complete the membrane repair after initial sealing of the plasma \nmembrane lesions remain unclear. The lysosomal Ca2+ channel MCOLN1 was found to \nbe important for plasma membrane repair, yet effective repair still required extracellular \nCa\n2+, possibly to replenish lysosomal stores (Cheng, Zhang et al. 2014). It is unknown if \nplasma membrane-localized Ca2+ channels contribute to plasma membrane repair. \nRecent studies have shown the importance of proteins identified as critical for plasma \nmembrane repair in other models, such as myoferlin (Leung, Yu et al. 2013), annexins \n(Bouvet, Ros et al. 2020, Gounou, Bouvet et al. 2023) and annexin-associated S100 \n(which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. \nThe copyright holder for this preprintthis version posted May 15, 2026. ; https://doi.org/10.64898/2026.05.13.724989doi: bioRxiv preprint \n\nfamily members (Jaiswal, Lauritzen et al. 2014), in plasma membrane repair in tumor \ncells. Notably, the acquisition of a more robust plasma membrane repair capacity was \nidentified as a response to the higher propensity of invasive breast cancer cells to \nundergo increased plasma membrane damage (Jaiswal, Lauritzen et al. 2014), \nsupporting the idea that plasma membrane repair in cancer cells represents a \nfunctionally important mechanoprotective adaptation. As we linked the EHD2-Orai1 axis \nto the stability of CAV1-containing plasma membrane domains and invasive/metastatic \nbehavior of TNBC cells (Luan, Bielecki et al. 2023), we utilized these cell models to \nexamine the role of the EHD2-Orai1 axis in caveolae-dependent mechanoprotection. \nOur findings establish that Orai1 and its ability to import the extracellular Ca\n2+ into \ncytoplasm are required for the repair of plasma membrane injuries induced \nmechanically or by a model bacterial pore-forming toxin, streptolysin O (SLO). We \ndemonstrate that mechanical force applied to the plasma membrane elicits rapid, highly \nlocalized, EHD2 recruitment and Orai1-mediated Ca\n2+ influx, identifying the \nmechanosensitive nature of the EHD2-Orai1 axis. Finally, we show that stiff extracellular \nmatrix activation of YAP/TAZ-TEAD signaling requires the EHD2-Orai1 axis-dependent \nCa2+ import, and in turn helps sustain Orai1-mediated mechanosensing and \nmechanoprotection. Thus, our studies establish a new paradigm for plasma-membrane-\nchannel-mediated import of extracellular Ca2+ as an essential component of mammalian \nplasma membrane repair and identify a novel mechanosensitive role for EHD2 and \nOrai1 in signaling to the YAP/TAZ pathway. \n \n(which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. \nThe copyright holder for this preprintthis version posted May 15, 2026. ; https://doi.org/10.64898/2026.05.13.724989doi: bioRxiv preprint \n\nResults: \nEHD2 is required for efficient repair of plasma membrane injuries induced by \nmechanical rupture or streptolysin O \nIn view of the established role of EHD2 to stabilize the plasma membrane pool of \ncaveolae (Moren, Shah et al. 2012, Stoeber, Stoeck et al. 2012, Hoernke, Mohan et al. \n2017, Yeow, Howard et al. 2017), recruitment of EHD2 to the shoulder region of plasma \nmembrane repair cap in skeletal muscle injury models (Marg, Schoewel et al. 2012, \nDemonbreun, Quattrocelli et al. 2016), mechanical perturbation-induced nuclear \nshuttling of EHD2 and its involvement in gene expression (Torrino, Shen et al. 2018), we \nposited that EHD2 may be required for plasma membrane repair. We adapted the \nscratch wounding protocol commonly used to assess tumor cell migration to examine \nmechanically induced plasma membrane injury repair since a large proportion of cells \nnear the scratch wound border showed the uptake of membrane impermeant \nfluorescent dyes, indicative of cells with plasma membrane damage (Fig. 1A, Fig. S1A \n& S1B). Incubation of wildtype (WT) MDA-MB231 or Hs578T TNBC cell lines for various \ntime points in Ca\n2+-containing medium demonstrated that most cells that incorporated \nthe membrane-impermeant fluorescent dye FITC-dextran (i.e., cells with plasma \nmembrane injury) became impermeant to the subsequently added propidium iodide (PI) \nwithin 1-5 minutes with slower recovery after that (Fig.1B), indicating successful plasma \nmembrane repair. In contrast, the injured cells incubated in medium without Ca\n2+ \nshowed significantly impaired repair (70% vs. 40% cells with repair at 40 min in +Ca2+ \nvs. -Ca2+ media; p<0.05) (Fig. 1B). Thus, as expected, the TNBC cell models we use \nexhibit robust Ca2+-dependent plasma membrane repair. Compared to WT TNBC cells, \n(which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. \nThe copyright holder for this preprintthis version posted May 15, 2026. ; https://doi.org/10.64898/2026.05.13.724989doi: bioRxiv preprint \n\ntheir EHD2-KO versions exhibited a marked and significant reduction in the levels of \nplasma membrane repair (25%  vs. 70% repair at 40 min, p<0.01), close to that \nobserved in WT cells in the absence of Ca2+; absence of Ca2+ further reduced the \nplasma membrane repair of EHD2-KO cells, but the difference was smaller (20% vs. \n15% in +Ca2+ vs. -Ca2+, p>0.05) (Fig. 1B). To further explore the role of EHD2 in plasma \nmembrane repair, we used streptolysin O (SLO) to induce plasma membrane pores. \nSLO is a prototype bacterial toxin that forms smaller and more uniform plasma \nmembrane pores that are also repaired in a Ca\n2+-dependent process (Cheng, Zhang et \nal. 2015, Cooper and McNeil 2015, Demonbreun and McNally 2016, Andrews and \nCorrotte 2018, Horn and Jaiswal 2018). Repair was assessed by analyzing the \nproportion of cells permeable to PI using FACS analysis. In contrast to WT TNBC cells, \nEHD2-KO cells exhibited a significantly higher percentage of PI-high cells (40% vs. 16% \nPI+ cells in MDA-MB-231 and 60% vs. 35% PI+ cells in Hs578T; p<0.001), indicating \nless efficient repair (Fig. 1C &1D). Notably, EHD2-KO MDA-MB231 cells reconstituted \nwith mouse EHD2 (Luan, Bielecki et al. 2023) showed repair comparable to that in WT \ncells (20% vs. 16% PI+ cells; p>0.05) (Fig. 1C &1D). Together, these results led us to \nconclude that EHD2 is required for the repair of plasma membrane injuries induced by \nmechanical force or a prototype pore-forming bacterial toxin. \nPlasma membrane repair is mediated by the activity of Orai1 Ca\n2+ channel \nPreviously, we showed that loss of EHD2 expression leads to lower plasma membrane \nlevels of Orai1, and the functional impact of the EHD2 loss on cell migration and \ntumorigenesis was recapitulated by Orai1 inhibition while overexpression of STIM1 \npartially rescued the SOCE and cell migration defects in EHD2-KO MDA-MB-231 cells \n(which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. \nThe copyright holder for this preprintthis version posted May 15, 2026. ; https://doi.org/10.64898/2026.05.13.724989doi: bioRxiv preprint \n\n(Luan, Bielecki et al. 2023). We therefore tested the possibility that Orai1 may be \nrequired for EHD2-dependent plasma membrane repair. First, we generated Orai1-KO \nderivatives of MDA-MB-231 and Hs578T TNBC cell lines and confirmed the absence of \nOrai1 protein expression (Fig. 2A). Similar to our previous findings upon EHD2-KO in \nthese cell models (Luan, Bielecki et al. 2023), the extent of initial Ca2+ release as a \nmeasure of ER Ca2+ stores (50 % reduction in Orai1-KO in MDA-MB-231, p<0.001; 43 \n% reduction in Orai1-KO in Hs578T, p<0.005) and their SOCE response to Ca2+ store \ndepletion induced by thapsigargin (~65% reduction in Orai1-KO, p<0.001) (Fig. 2B& \n2C), as well as their trans-well cell migration towards serum-containing medium (~60% \nreduction in Orai1-KO, p<0.001) (Fig. S2A), were markedly and significantly reduced. \nThe reduction in ER release reflects a deficit of ER Ca2+ store filling because of \nimpaired SOCE (Luan, Bielecki et al. 2023). Notably, Orai1-KO led to a significant \nimpairment in the repair of mechanically-induced plasma membrane injuries (65% in WT \nvs. 38% in KO MDA-MB-231; 62% in in WT vs. 40% KO Hs578T at 40 min, p<0.01) \n(Fig. 2D& 2E, Fig. S2B) and SLO-induced membrane pores (14% PI\n+ cells in WT vs. \n36% in KO  in MDA-MB-231 and 19% PI+ cells  in WT vs. 48% in KO in Hs578T; \np<0.001) (Fig. 2F& 2G). Complementing the genetic approach, we also assessed the \nimpact of Orai1 inhibition. As the tool inhibitors used in previous studies, such as SKF-\n96365, lack selectivity (Ramsey, Delling et al. 2006, Ding, Zhang et al. 2012), we \nutilized a more recently developed Orai1-selective inhibitor CM4620 which functions by \ninhibiting the activated state of Orai1 (Stauderman 2018, Waldron, Chen et al. 2019) \nand has progressed through phase 2 clinical trials against acute pancreatitis and \nCOVID-19 pneumonia (Miller, Bruen et al. 2020, Bruen, Miller et al. 2021, Bruen, Al-\n(which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. \nThe copyright holder for this preprintthis version posted May 15, 2026. ; https://doi.org/10.64898/2026.05.13.724989doi: bioRxiv preprint \n\nSaadi et al. 2022).  First, we established the Orai1 dependence of the CM4620 effect on \nfunctional readouts of Orai1 activity in TNBC cells. Indeed, CM4620 robustly inhibited \nthe SOCE (~67% reduction in CM4620-treated vs. control; p<0.001) and cell migration \n(~55% reduction with CM4620 vs. control; p<0.001) in WT TNBC cells but had little \nimpact on the residual cell migration in Orai1-KO cells (185 cells per field with DMSO \nvs. 180 cells per field with CM4620,  not significant) (Fig. 2H & Fig. S2D). Importantly, \ntreatment with CM4620 impaired the repair of mechanical (42% in CM4620-treated cells \nvs. 74% in control cells, p<0.01) (Fig. 2I & Fig. S2E) as well as SLO-induced (38% PI+ \ncells in CM4620 vs. 12% PI+ cells in control, p<0.001) (Fig. 2J) plasma membrane \ndamage in TNBC cells. Together, these results support the conclusion that plasma \nmembrane Ca\n2+ channel Orai1 is required for efficient plasma membrane repair. \nEHD2 and Orai1 are required for mechanosensitive spatiotemporally regulated \nimport of calcium into the cytoplasm \nThe requirement of EHD2 and Orai1 for plasma membrane repair suggested that these \nproteins orchestrate a novel pathway of mechanosensitive entry of Ca2+ from the \nextracellular space into the cytoplasm. To test this possibility, we used Atomic Force \nMicroscopy (AFM) to assess the impact of a mechanical stimulus applied to the plasma \nmembrane. MDA-MB-231 cells transfected with fluorescent EHD2 were subjected to \nnano-indentation with cantilevers and localization of EHD2 over time was monitored by \nconfocal imaging (Fig. 3A). We observed rapid (within seconds) accumulation of \nfluorescent EHD2 precisely at the indentation site (indentation point indicated), which \ndissipated quickly once the mechanical force was removed (Fig. 3B & 3C). The focal \naccumulation of EHD2 signals was significantly higher compared to the pre-induction \n(which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. \nThe copyright holder for this preprintthis version posted May 15, 2026. ; https://doi.org/10.64898/2026.05.13.724989doi: bioRxiv preprint \n\nsignals (~2 fold higher at the peak, p<0.01) (Fig. 3D & 3E). Next, we used MDA-MB-\n231 cells transfected with a red fluorescent reporter of cytoplasmic Ca2+ (R-GECO1.2) \n(Wu, Liu et al. 2013) to assess if the localized application of force to the plasma \nmembrane induced Ca2+ import into cytoplasm. We observed rapid Ca2+ entry that \nstarted near the site of the plasma membrane indentation and spread to rest of the cell; \nthe Ca\n2+ entry dissipated quickly when the mechanical stimulus was removed (Fig. 4A). \nThe mechanosensitive Ca2+ entry was abrogated by genetic KO of EHD2 (~45% \nreduction; p<0.01), Orai1 (~40% reduction ; p<0.01) or CAV1 (~55% reduction; p<0.01), \nand the defect in EHD2-KO cells was partially rescued by ectopic expression of mouse \nEhd2 (~45% reduction in EHD2-KO vs. 20% reduction in mEhd2-reconstituted cells; \np<0.05) (Fig. 4B). Further, an Orai1 inhibitor CM5480 (Stauderman 2018, Pallagi, \nGörög et al. 2022, Szabó, Csákány-Papp et al. 2023), which also exhibited Orai1-\ndependent activity in TNBC cells (Fig. S3A), effectively inhibited the mechanosensitive \nCa2+ entry (~85% reduction in CM5480 vs. control, p<0.01) (Fig. 4C). To more directly \ninterrogate if the mechanosensitive, Orai1-dependent, Ca2+ import observed above \nindeed reported the Orai1-mediated Ca2+ entry, we transfected MDA-MB231 with a \ngenetically encoded fluorescent biosensor, G-GECO1-Orai1 (Dynes, Amcheslavsky et \nal. 2016).  In this biosensor, the green-fluorescent Ca2+ indicator fused to the N-\nterminus of Orai1 itself detects the Orai1-associated Ca2+ influx locally in the \ncytoplasmic nanodomain adjacent to the plasma membrane (Dynes, Amcheslavsky et \nal. 2016). Transiently transfected G-GECO1-Orai1 showed plasma membrane \nlocalization as expected (Fig. 4D). The G-GECO1-Orai1 also accurately reported only \nthe SOCE phase of the Ca\n2+ influx in response to thapsigargin treatment of MDA-MB-\n(which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. \nThe copyright holder for this preprintthis version posted May 15, 2026. ; https://doi.org/10.64898/2026.05.13.724989doi: bioRxiv preprint \n\n231 cells, which was completely abolished by the selective Orai1 inhibitor CM5480 (Fig. \nS3B). Indentation of the plasma membrane led to marked, and statistically-significant, \nincrease in Ca2+ influx reported by G-GECO1-Orai1 (6.2-fold increase in signal over \nunstimulated cells; p<0.001) (Fig. 4D). Pretreatment of cells with CM5480 led to a \nhighly significant inhibition of G-GECO1-Orai1 fluorescence upon indentation (1-fold-\nchange in CM5480-treated cells vs. 6.2-fold change in control; p<0.001) (Fig. 4D). \nThese results conclusively establish that mechanical force applied to the plasma \nmembrane leads to Orai1 activation. Since Orai Ca\n2+ channels are gated by STIM \nproteins (Ong, Subedi et al. 2019, Lewis 2020), we asked if STIM proteins are required \nfor mechano-sensitive Orai1 activation. In MDA-MB231 cells, both STIM1 and STIM2 \nwere robustly expressed, and siRNA KD of STIM2 led to a substantial upregulation of \nSTIM1 expression (Fig. S3C). Concurrent STIM1 and STIM2 siRNA transfection in \nMDA-MB231 cells expressing R-GECO1.2 led to efficient knockdown of both STIM1 and \n2 (Fig. S3D) and effectively abolished the plasma membrane indentation induced Ca\n2+ \ninflux (Fig. 4E).  These results establish that EHD2-, CAV1- and Orai1-dependent \nmechano-sensitive Ca2+ entry is indeed mediated by Orai1 and dependent on STIM \nproteins. Altogether, these results identify EHD2 and Orai1 as components of a novel \naxis that mediates mechanosensitive Ca2+ import from the extracellular space into the \ncytoplasm with high spatial and temporal control.  \nEHD2-Orai1 axis is required for mechanosensitive activation of the YAP/TAZ-\nTEAD pathway \nMechanical cell strain, the primary driver of plasma membrane damage, is typically \nelevated in stiff extracellular microenvironments, including in tumors (Lachowski, \n(which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. \nThe copyright holder for this preprintthis version posted May 15, 2026. ; https://doi.org/10.64898/2026.05.13.724989doi: bioRxiv preprint \n\nMatellan et al. 2022, Lee, Yun et al. 2025). Recent work has established that application \nof mechanical force to plasma membrane induces CAV1-dependent YAP/TAZ-TEAD \npathway activation (Dupont, Morsut et al. 2011, Rausch, Bostrom et al. 2019, Lolo, \nWalani et al. 2023), and there is emerging support for Ca2+ as a potential intermediate \nto positively or negatively modulate such mechanosensitive YAP/TAZ activation (Wei \nand Li 2021). Importantly, the downstream targets of mechanosensitive YAP/TAZ-TEAD \nactivation include CAV1 and CAVIN1 in positive feedback that was found to be essential \nto sustain high levels of plasma membrane caveolae (Dupont, Morsut et al. 2011, \nRausch, Bostrom et al. 2019, Lolo, Walani et al. 2023). Our findings that EHD2 and \nOrai1 are required for mechanosensitive Ca\n2+ entry from the extracellular space into \ncytoplasm raised the possibility that EHD2-Orai1 axis serves as a mechanosensitive \nactivator of YAP/TAZ signaling. As reported (Moreno-Vicente, Pavon et al. 2018), culture \nof WT MDA-MB231 cells on stiff matrix (64 kPa) induced the nuclear translocation of \nYAP compared to cells cultured on soft hydrogel (0.2 kPa) (~6 fold higher \nnuclear/cytoplasmic ratio of YAP staining on 64 kPa vs. 0.2 kPa hydrogel, p<0.001) \n(Fig. 5A& 5B, Fig. S5). Analysis of KO cell lines revealed that while YAP nuclear \ntranslocation was still significantly higher on stiff compared to soft matrix \n(nuclear/cytoplasmic YAP ratio  ~1.5 fold higher in EHD2-KO, ~3 fold higher in Orai1-KO \nand ~4 fold higher in Cav1-KO;p<0.001) (Fig. 5A& 5B, Fig. S5), the extent of YAP \nnuclear translocation in EHD2-KO, Orai1-KO and CAV1-KO MDA-MB-231 cells was \nsignificantly reduced compared to that in WT cells (~83% reduction in EHD2-KO vs. WT, \n~ 86% reduction in Orai1-KO vs. WT, ~ 78% reduction in CAV1-KO vs. WT on 64 kPa \nmatrix; p<0.001) (Fig. 5A& 5B, Fig. S5 ). Notably, mouse Ehd2 expression in EHD2-KO \n(which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. \nThe copyright holder for this preprintthis version posted May 15, 2026. ; https://doi.org/10.64898/2026.05.13.724989doi: bioRxiv preprint \n\ncells partially restored the YAP nuclear translocation (~3.5-fold increase in mouse Ehd2-\nrescued vs. 1.5-fold increase in EHD2-KO cells on 64 kPa matrix; p<0.001) (Fig. 5B). \nTreatment of MDA-MB-231 cells with Orai1 inhibitors CM4620 or CM5480 also \nabrogated the nuclear translocation of YAP (~50% reduction with CM4620 vs. control, \n~46% reduction with CM5480 vs. control, p<0.001) (Fig. 5C), validating the results of \ngenetic KOs.  To assess the impact of EHD2-KO or Orai1-KO on YAP/TAZ-TEAD \npathway activity, we determined the activity of a transiently transfected TEAD pathway \nluciferase reporter in cells grown on stiff matrix. Both EHD2-KO and Orai1-KO \nsignificantly reduced the reporter activity (~40% reduction compared to WT, p <0.01; \nFig. 5D). We further examined the induction of established TEAD target genes, \nincluding CAV1 and CAVIN1, the structurally essential components of caveolae, as \nindicators of the positive feedback loop between caveolae and YAP/TZ-TEAD (Dupont, \nMorsut et al. 2011, Rausch, Bostrom et al. 2019, Lolo, Walani et al. 2023), using qPCR. \nThe stiff ECM-dependent TEAD target gene expression was significantly impaired by \nKO of EHD2, or Orai1 (~52% decrease of CTGF, ~53% decrease of CYR61, ~85% \ndecrease of ANKRD1, ~50% decrease of CAV1 and ~40% decrease of CAVIN1 in \nEHD2-KO vs. WT; ~82% decrease of CTGF, ~86% decrease of CYR61, ~90% \ndecrease of ANKRD1, 45% decrease of CAV1 and 40% decrease of CAVIN1 in Orai1-\nKO vs. WT; p<0.001) (Fig. 5E). \nTo further confirm the positive feedback between EHD2/Orai1 regulated YAP/TAZ-TEAD \npathway activity and the expression of caveolar proteins CAV1 and CAVIN1 seen in \nTNBC cells with gene KOs, we examined the impact of pharmacological inhibition of \nTEAD. Treatment of MDA-MB-231 and Hs578T TNBC cells with an allosteric pan-TEAD \n(which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. \nThe copyright holder for this preprintthis version posted May 15, 2026. ; https://doi.org/10.64898/2026.05.13.724989doi: bioRxiv preprint \n\ninhibitor GNE-7883 (Hagenbeek, Zbieg et al. 2023) for 72 hours resulted in a significant \nreduction in the levels of CAV1 (50% reduction compared to control; p<0.01) (Fig. 6A). \nGNE-7883 treatment in both MDA-MB-231 and Hs578T cell lines significantly reduced \nthe thapsigargin-induced SOCE (28% reduction compared to control in MDA-MB-231 \ncells and 23% reduction in Hs578T cells; p<0.01) (Fig. 6B-C). GNE-7883 treatment of \nMDA-MB-231 cells for 24 hours also significantly impaired the Ca2+ uptake induced by \nAFM-induced mechanical stimulus to the plasma membrane (6.5-fold Ca2+ influx \nincrease in DMSO vs. 2.5-fold Ca2+ influx increase in GNE-7883 treatment, p<0.001) \n(Fig. 6D). Finally, short-term GNE-7883 treatment significantly impaired the ability of \nMDA-MB-231 and Hs578T cells to repair the mechanically-induced (65% repair in \ncontrol vs. 56% in treated MDA-MB-231 cells at 40 min; 65% repair in control vs. 50% in \ntreated Hs578T cells at 40 min; p <0.05) (Fig. 6E, Fig. S6) and SLO-induced (13% PI+ \ncells in control vs. 17% in treated MDA-MB-231 cells; 35% PI+ cells in control vs. 58% \nin treated Hs578T cells; p <0.05) (Fig. 6F) plasma membrane injuries.   \nAltogether, we find that in addition to an essential role of EHD2-Orai1 axis in acute \nresponse to plasma membrane strain, long-term exposure of cells to stiff \nmicroenvironment induces EHD2-Orai1 and YAP/TAZ-TEAD axis dependent \nmechanoadaptation to counteract plasma membrane damage. \nDiscussion: \nThe ability to promptly repair plasma membrane breaches is essential for the life of \norganisms without a cell wall (Cooper and McNeil 2015, Demonbreun and McNally \n2016, Andrews and Corrotte 2018). Ca\n2+ inflow from the extracellular space is required \n(which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. \nThe copyright holder for this preprintthis version posted May 15, 2026. ; https://doi.org/10.64898/2026.05.13.724989doi: bioRxiv preprint \n\nto initiate and complete the plasma membrane repair processes (Cheng, Zhang et al. \n2015, Cooper and McNeil 2015, Demonbreun and McNally 2016, Andrews and Corrotte \n2018, Horn and Jaiswal 2018). The passive inflow of Ca2+ from the extracellular space \ninto cytoplasm is thought to be the initial trigger to initiate the repair. Whether plasma \nmembrane-localized Ca2+ channels have any role in plasma membrane repair is not \nknown. Studies presented here establish that the plasma membrane Ca2+ channel \nOrai1, a mediator of store-operated Ca2+ entry, is essential for mammalian plasma \nmembrane repair. We also establish that EHD2, a caveolae-associated protein required \nto sustain high plasma membrane levels of Orai1 (Luan, Bielecki et al. 2023), is also \nessential for mammalian plasma membrane repair. Importantly, we show that EHD2 and \nOrai1 function as required upstream components of a mechanotransduction cascade to \nactivate YAP/TAZ-TEAD signaling. Collectively, our findings identify novel roles of Orai1-\nmediated Ca\n2+ transport to sustain Ca2+-dependent mammalian plasma membrane \nrepair (Cheng, Zhang et al. 2015, Cooper and McNeil 2015, Demonbreun and McNally \n2016, Andrews and Corrotte 2018, Horn and Jaiswal 2018) and for activation of \nYAP/TAZ-TEAD pathway of mechanoadaptation (Dupont, Morsut et al. 2011, Rausch, \nBostrom et al. 2019, Lolo, Walani et al. 2023). \nThe role of Orai1 (or its family members Orai2/3), in partnership with STIM1 (or STIM2) \nin SOCE is well-established, but this role has almost exclusively been investigated in \nthe context of responses to biochemical signaling, such as cell surface receptor \nactivation (Ong, Subedi et al. 2019, Lewis 2020). In contrast, there is little evidence for a \nprimary role for Orai channels in cellular responses to mechanical stimuli. However, \nprior studies have suggested the role of Orai1 secondary to activation of other \n(which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. \nThe copyright holder for this preprintthis version posted May 15, 2026. ; https://doi.org/10.64898/2026.05.13.724989doi: bioRxiv preprint \n\nmechanosensitive Ca2+ channels. For example, Ca2+ entry induced by \nmechanosensitive Piezo1 channel agonist Yoda-2 in endometrial mesenchymal stem \ncells was shown to be dampened by 2-APB, an inhibitor of Orai channel-mediated \nSOCE (Chubinskiy-Nadezhdin, Semenova et al. 2022). In a study using an in vivo \nparadigm of flow shear-dependent mechanotransduction, Orai1-KO embryos displayed \nreduced lymphatic density and impaired lymphatic development (Choi, Park et al. 2017). \nFurther studies demonstrated that Piezo1 channel served in the mechanosensory role, \nwith Orai1 mediating the subsequent SOCE (Choi, Park et al. 2017). We show that \ndirect application of mechanical force to the plasma membrane initiates rapid Ca\n2+ \nimport that is completely abrogated by Orai1-KO as well as by Orai1 channel-selective \ninhibitors (Fig. 4A-C). We have previously established that loss of EHD2 leads to \nreduced SOCE, with a reduction in the plasma membrane pool of Orai1 while the total \nOrai1 levels were unchanged (Luan, Bielecki et al. 2023). Indeed, the mechanosensitive \nCa2+ entry was also lost upon EHD2-KO (Fig. 4B). Consistent with Orai1 localization to \nCAV1-containing plasma membrane domains (Sathish, Abcejo et al. 2012, Chantôme, \nPotier-Cartereau et al. 2013, Jardin and Rosado 2016, Bohorquez-Hernandez, Gratton \net al. 2017), we observed loss of mechanosensitive Ca2+ entry upon CAV1-KO (Fig. \n4B). Mechanosensitivity of Ca2+ import through G-GECO1-Orai1, a genetically-encoded \nCa2+ biosensor that reports direct Ca2+ influx through Orai1 (Dynes, Amcheslavsky et al. \n2016), (Fig. 4D) further substantiates the role of Orai1 in mechanosensitive Ca2+ entry \nacross the plasma membrane. We further demonstrate the mechano-sensitive, Orai1-\nmediated Ca2+ entry to require STIM proteins, indicating that such Ca2+ transport occurs \nat ER-PM contacts, where STIM proteins are known to interact with and activate Orai1 \n(which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. \nThe copyright holder for this preprintthis version posted May 15, 2026. ; https://doi.org/10.64898/2026.05.13.724989doi: bioRxiv preprint \n\n(Ong, Subedi et al. 2019, Lewis 2020).  Collectively, our results define a new \nmechanosensitive role for plasma membrane caveolae-localized Orai1. The absence of \na retained intracellular store Ca2+ release component upon mechanical indentation of \nOrai1-, EHD2- or CAV1-KO cell models argues against the likelihood of a distinct \nmechanosensitive Ca2+ channel mediating the initial Ca2+ release with Orai1 functioning \nin a classical role as an SOCE channel in our cell system. However, given the findings \nin Orai1-KO mice discussed above, it remains possible that mechanosensitive proteins, \nincluding the known mechanosensitive Ca\n2+ channels, are involved in Orai1 activation in \nresponse to mechanical stimuli. A systematic analysis of candidate mechanosensitive \nCa2+ channels and accessory proteins will be needed to clarify this further. \nGiven the accumulation of EHD2 at plasma membrane repair sites in skeletal muscle \nmodels (Marg, Schoewel et al. 2012, Demonbreun, Quattrocelli et al. 2016), our findings \nin cancer cells suggest a comparable role of EHD2 in other cell types with known roles \nof caveolae in plasma membrane repair (Sinha, Koster et al. 2011, Corrotte, Fernandes \net al. 2012, Corrotte, Almeida et al. 2013, Shvets, Ludwig et al. 2014, Cheng, Mendoza-\nTopaz et al. 2015, Garcia, Bagwell et al. 2017). More importantly, our results raise the \npossibility of a broader and essential role of Orai1 in caveolae-dependent repair across \ncell types and organisms. Whether EHD2 functions as an obligate partner in such a role \nremains to be determined; while EHD2 and CAV1 proteins are coordinately expressed \nin breast cancer cell lines (Luan, Bielecki et al. 2023) and CAV1/CAVIN1 and EHD2 \nmRNAs show strong co-expression across cell types, EHD2 and Orai1 mRNA \nexpression is not similarly correlated (based on single cell portal and mRNA co-\nexpression databases). Consistent with EHD2-independent Orai1 function in membrane \n(which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. \nThe copyright holder for this preprintthis version posted May 15, 2026. ; https://doi.org/10.64898/2026.05.13.724989doi: bioRxiv preprint \n\nrepair, the residual repair in EHD2-KO cells was still Ca2+ dependent (Fig. 1B), likely \nreflecting the reduction but not a complete absence of cell surface Orai1 in these cells \n(Luan, Bielecki et al. 2023). Further analyses in naturally EHD2-low/non-expressing cell \nsystems will be required to explore this further. \nPrior work has established the YAP/TAZ-TEAD pathway as a major mechanosensitive \nsignaling axis in response to ECM stiffness, shear stress and cell stretching in a manner \nindependent of the upstream Hippo pathway kinases (Dupont, Morsut et al. 2011, \nRausch, Bostrom et al. 2019, Lolo, Walani et al. 2023). CAV1 was identified as a critical \nupstream positive regulator of such mechanosensitive YAP/TAZ-TEAD pathway \nactivation (Moreno-Vicente, Pavon et al. 2018). Consistent with the required role of \nEHD2 to stabilize plasma membrane caveolae (Sathish, Abcejo et al. 2012, Chantôme, \nPotier-Cartereau et al. 2013, Jardin and Rosado 2016, Bohorquez-Hernandez, Gratton \net al. 2017), we found that deletion of EHD2 impaired the stiff ECM dependent YAP \nnuclear translocation, phenocopying the impact of CAV1-KO (Fig. 5A-B). Importantly, \nwe found that Orai1-KO or its inhibition also impaired the YAP translocation induced by \nstiff ECM (Fig. 5C). Further, EHD2-KO or Orai1-KO reduced the stiff ECM induced \nTEAD target gene expression (Fig. 5D). These findings establish a novel requirement of \nEHD2 and Orai1 in mechanosensitive YAP/TAZ-TEAD activation. Combining insights \nfrom our earlier study linking EHD2 and Orai1 (Luan, Bielecki et al. 2023), and our \nfindings here, we propose that EHD2-mediated stabilization of CAV1-containing plasma \nmembrane domains places Orai1 at mechanosensitive plasma membrane domains and \nthat mechanical stimuli activate Orai1 as a required step in YAP/TAZ-TEAD pathway \nactivation. Orai1-mediated Ca\n2+ import in the context of cell surface receptor activation \n(which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. \nThe copyright holder for this preprintthis version posted May 15, 2026. ; https://doi.org/10.64898/2026.05.13.724989doi: bioRxiv preprint \n\nas a trigger to regulate multiple cellular signaling pathways is well established, including \nseveral transcriptional regulatory pathways (Nieto-Felipe, Macias-Diaz et al. 2023). The \ncritical role of Orai1 as an upstream positive regulator of YAP/TAZ-TEAD pathway \nactivation, as we identify here, therefore provides a novel paradigm to understand \nCAV1-dependent mechanosensitive signaling. Consistent with this suggestion, \nmechanosensitive YAP/TAZ-TEAD activation was found to require Rho GTPase activity \nand actomyosin cytoskeletal contractility (Dupont, Morsut et al. 2011), which in turn are \nknown be regulated by mechanosensitive Ca\n2+ fluxes (Higashida, Kiuchi et al. 2013, \nPardo-Pastor, Rubio-Moscardo et al. 2018, Lakk and Krizaj 2021, Miroshnikova, Manet \net al. 2021, Varadarajan, Chumki et al. 2022, Fu, Wang et al. 2024). \nPrior studies have shown that YAP/TAZ-TEAD pathway is required for the expression of \nstructural components of caveolae, CAV1 and CAVIN1, and inhibition of YAP/TAZ-TEAD \naxis led to loss of plasma membrane caveolae (Dupont, Morsut et al. 2011, Rausch, \nBostrom et al. 2019, Lolo, Walani et al. 2023). More recently, mild to moderate \nmechanical plasma membrane stress was shown to activate YAP/TAZ-TEAD pathway \nthrough CAV1-containing but CAVIN1-negative plasma membrane dolines as a \nmechanoadaptation mechanism through feedback upregulation of CAV1 and CAVIN1 \ngene expression to increase the CAV1/CAVIN1-containing plasma membrane caveolae, \nwhich are required for mechanical protection against more severe plasma membrane \nstress  (Dupont, Morsut et al. 2011, Rausch, Bostrom et al. 2019, Lolo, Walani et al. \n2023). Consistent with EHD2-Orai1 axis as an intermediary in such mechanoadaptive \npositive feedback, stiff matrix-induced CAV1 and CAVIN1 gene expression was reduced \nby EHD2 or Orai1 KO (Fig. 5E). Further supporting this mechanism downstream of the \n(which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. \nThe copyright holder for this preprintthis version posted May 15, 2026. ; https://doi.org/10.64898/2026.05.13.724989doi: bioRxiv preprint \n\nEHD2-Orai1 axis, even short-term pharmacological TEAD inhibition reduced the levels \nof CAV1/CAVIN1 proteins  (Fig. 6A) and the SOCE response elicited upon ER Ca2+ \nstore depletion (using thapsigargin) (Fig. 6B-C ), Remarkably, TEAD inhibition \nsignificantly impaired the Ca2+ import in response to plasma membrane force \napplication (Fig. 6E-F), expanding the YAP/TAZ-TEAD mediated feedback to the EHD2-\nOrai1-dependent mechanosensitive Ca2+ import, which in turn our findings establish as \nessential for YAP/TAZ-TEAD pathway activation.   \nOur findings that EHD2 is required for rapid Orai1 activation in response to mechanical \nforce applied to the plasma membrane (Fig. 4A) are consistent with the requirement of \nEHD2 to sustain the plasma membrane pool of caveolae (Sathish, Abcejo et al. 2012, \nChantôme, Potier-Cartereau et al. 2013, Jardin and Rosado 2016, Bohorquez-\nHernandez, Gratton et al. 2017) and our previous work that EHD2 is required to sustain \nhigh plasma membrane levels of Orai1 (Luan, Bielecki et al. 2023). However, our finding \nof rapid EHD2 recruitment to plasma membrane sites of applied mechanical force (Fig. \n3B) differs from previous findings of CAV1-dependent modest or substantial release of \nEHD2 from the plasma membrane in response to cyclical stretch or hypotonic stress, \nrespectively, with sumo modification of EHD2 leading to its nuclear localization (Torrino, \nShen et al. 2018). The discordant results may reflect our use of transiently applied \nlocalized mechanical force as opposed to more prolonged cell-wide mechanical \nstimulation in prior studies and will need further investigation. That mechanosensitive \nOrai1 activation, which occurs at the plasma membrane, requires EHD2, strongly \nargues for the observed mechanosensitive role of EHD2 at the plasma membrane \nrather than through nuclear localization. In prior work, we found that high nuclear \n(which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. \nThe copyright holder for this preprintthis version posted May 15, 2026. ; https://doi.org/10.64898/2026.05.13.724989doi: bioRxiv preprint \n\nstaining of EHD2 in breast cancer tissues was associated with longer patient survival, \ndiametrically opposite to the association of high non-nuclear EHD2 overexpression with \nshorter patient survival (Luan, Bielecki et al. 2023). Thus, the nuclear localization of \nEHD2 in response to mechanical stress may reflect its sequestration for later utilization \nin plasma membrane-associated functions. Indeed, the prior study discussed above \n(Torrino, Shen et al. 2018) found rapid exit of EHD2 from the nucleus and its re-\nlocalization to plasma membrane during recovery from hypotonic stress. Further, NIH-\n3T3 cells rendered caveolae-deficient by the combined EHD1, 2 and 4 KO exhibited \nsusceptibility or membrane ruptures upon prolonged cyclical stretching (Yeow, Howard \net al. 2017).  \nAn unresolved question is how the mechanical force applied to plasma membrane might \nactivate Orai1 in an EHD2-dependent and STIM-dependent manner. Changes in \nmembrane curvature sensed by curvature sensing proteins are well known to affect \ncellular responses (McMahon and Boucrot 2015). For example, mechanosensitive \nPiezo1 Ca\n2+ channel was found enriched at plasma membrane invaginations and \ndepleted at filopodia (Yang, Miao et al. 2022). Recently, use of vertical pillars to induce \nplasma membrane curvature changes mimicking the cardiomyocyte plasma membrane \ntransverse tubules, sites enriched for ER-PM contacts, was found to induce ER-PM \ncontacts through junctophilin proteins; interaction of junctophilin-2 with EHD proteins \nwas identified as a mechanism for curvature sensing to promote ER-PM contact \nenrichment (Yang, Valencia et al. 2024). Thus, it is plausible that mechanical force \ninduced curvature on the plasma membrane recruits EHD2 to promote rapid ER-PM \ncontact formation or stabilization to promote STIM1-Orai1 interaction and \n(which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. \nThe copyright holder for this preprintthis version posted May 15, 2026. ; https://doi.org/10.64898/2026.05.13.724989doi: bioRxiv preprint \n\nmechanosensitive Ca2+ entry. While studies in cardiomyocytes showed junctophilin-2 \ninteraction with multiple EHD proteins, and EHD4 was functionally implicated, these \nother family members are expressed in parental as well as EHD2-KO breast cancer cell \nmodels used here (Luan, Bielecki et al. 2023) and do not appear to compensate for the \nrole of EHD2. It is possible that the requirement of EHD2 for high PM Orai1 expression \n(Luan, Bielecki et al. 2023) contributes to this relative specificity. \nIn conclusion, our findings establish the EHD2-Orai1 duo as a novel regulator of \nmechanosensitive import of extracellular Ca\n2+ essential for efficient mammalian cell \nplasma membrane repair. We also establish the EHD2-Orai11 axis as a critical \nupstream element required for the activation of mechanosensitive YAP/TAZ-TEAD \ndependent gene expression for mechanoadaptation. Further studies of this novel \nplasma membrane mechanosensory apparatus are likely to reveal key new insights into \nmechanotransduction and regulation of plasma membrane homeostasis under \nphysiological states and in diseases, such as cancer.  \n ACKNOWLEDGEMENTS \nWe thank the Band Lab members for their many discussions, and the staff of \nUNMC Core facilities for their invaluable a ssistance. We thank Drs. Robert Campbell \n(CMV-R-GECO1.2), Michael Cahalan (G-GECO1-Orai1) and Ari Helenius (EHD2-\nmCherry) for plasmids obtained through Addgene. This work was supported by grants \nfrom Department of Defense breast cancer research program (W81XWH-17-1-0616 and \nW81XWH-20-1-0058 to HB, and W81XWH-20-1-0546, HT94252410337 and\n HT9425-\n23-1-0052 to VB), the NIH (R21CA297629 to HB, R21CA241055 and R03CA253193 to \nVB, and P20 GM121316 to SPR), a UNMC startup grant (to SPR), the Fred & Pamela \n(which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. \nThe copyright holder for this preprintthis version posted May 15, 2026. ; https://doi.org/10.64898/2026.05.13.724989doi: bioRxiv preprint \n\nBuffett Cancer Center pilot grants (to HB & VB), a Nebraska Research Initiative seed \ngrant (to HB and JYL), a Nebraska DHHS LB506 pilot award (18123-Y3 to HB), a \nChildren's Hospital Research Institute pilot grant (to HB), and the Raphael Bonita \nMemorial Fund (to HB). Research reported in this publication was supported by the \nAdvanced Microscopy, Flow Cytometry, and Data Science Shared Core Facilities which \nare partially funded by the National Cancer Institute of the National Institutes of Health \nCancer Center Support Grant to the Fred & Pamela Buffett Cancer Center \n(P30CA036727) and the Nebraska Research Initiative. TAB was a trainee under the \nNational Cancer Institute Cancer Biology Training Grant T32CA009476.  \n \nAuthor Contributions: HL (Investigation, data curation, formal analysis, methodology, \nvisualization, writing – original draft); AC, TAB, BCM, SS, WW, MDS (Investigation, data \ncuration, formal analysis, methodology); LMS (Formal analysis); KAS (Resources); \nDWC & AK (Funding acquisition), JD (Resources, methodology), JTL \n(Conceptualization, funding acquisition, writing – review and editing), VB \n(Conceptualization, funding acquisition, supervision, writing – original draft, review & \nediting), SPR (Conceptualization, supervision, investigation, formal analysis, \nvisualization, writing – original draft; review & editing), HB (Conceptualization, funding \nacquisition, supervision, resources, formal analysis, writing – original draft, review & \nediting). \n \nDECLARATION OF INTERESTS: KAS is an employee of CalciMedica Inc., which is \nadvancing Orai1 inhibitors into clinic. Other authors have nothing to declare. \n(which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. \nThe copyright holder for this preprintthis version posted May 15, 2026. ; https://doi.org/10.64898/2026.05.13.724989doi: bioRxiv preprint \n\n \nMaterials availability: All unique/stable reagents generated in this study are available \nfrom the lead contact with a completed materials transfer agreement. The recipient will \nincur shipping costs.  \n(which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. \nThe copyright holder for this preprintthis version posted May 15, 2026. ; https://doi.org/10.64898/2026.05.13.724989doi: bioRxiv preprint \n\nReferences: \nAndrews, N. W. and M. Corrotte (2018). \"Plasma membrane repair.\" Curr Biol 28(8): \nR392–R397. \nBohorquez-Hernandez, A., E. Gratton, J. Pacheco, A. Asanov and L. Vaca (2017). \n\"Cholesterol modulates the cellular localization of Orai1 channels and its disposition \namong membrane domains.\" Biochim Biophys Acta Mol Cell Biol Lipids 1862(12): \n1481–1490. \nBouvet, F., M. Ros, E. Bonedeau, C. Croissant, L. Frelin, F. Saltel, V. Moreau and A. \nBouter (2020). \"Defective membrane repair machinery impairs survival of invasive \ncancer cells.\" Sci Rep\n 10(1): 21821. \nBruen, C., M. Al-Saadi, E. A. Michelson, M. Tanios, R. Mendoza-Ayala, J. Miller, J. \nZhang, K. Stauderman, S. Hebbar and P. C. Hou (2022). \"Auxora vs. placebo for the \ntreatment of patients with severe COVID-19 pneumonia: a randomized-controlled \nclinical trial.\" Crit Care\n 26(1): 101. \nBruen, C., J. Miller, J. Wilburn, C. Mackey, T. L. Bollen, K. Stauderman and S. Hebbar \n(2021). \"Auxora for the Treatment of Patients With Acute Pancreatitis and \nAccompanying Systemic Inflammatory Response Syndrome: Clinical Development of a \nCalcium Release-Activated Calcium Channel Inhibitor.\" Pancreas\n 50(4): 537–543. \nChantôme, A., M. Potier-Cartereau, L. Clarysse, G. Fromont, S. Marionneau-Lambot, M. \nGuéguinou, J. C. Pagès, C. Collin, T. Oullier, A. Girault, F. Arbion, J. P. Haelters, P. A. \nJaffrès, M. Pinault, P. Besson, V. Joulin, P . Bougnoux and C. Vandier (2013). \"Pivotal \nrole of the lipid Raft SK3-Orai1 complex in human cancer cell migration and bone \nmetastases.\" Cancer Res 73(15): 4852–4861. \nCheng, J. P ., C. Mendoza-Topaz, G. Howard, J. Chadwick, E. Shvets, A. S. Cowburn, B. \nJ. Dunmore, A. Crosby, N. W. Morrell and B. J. Nichols (2015). \"Caveolae protect \nendothelial cells from membrane rupture during increased cardiac output.\" J Cell Biol\n \n211(1): 53–61. \nCheng, X., X. Zhang, Q. Gao, M. Ali Samie, M. Azar, W. L. Tsang, L. Dong, N. Sahoo, X. \nLi, Y . Zhuo, A. G. Garrity, X. Wang, M. Ferrer, J. Dowling, L. Xu, R. Han and H. Xu \n(2014). \"The intracellular Ca(2)(+) channel MCOLN1 is required for sarcolemma repair \nto prevent muscular dystrophy.\" Nat Med\n 20(10): 1187–1192. \nCheng, X., X. Zhang, L. Yu and H. Xu (2015). \"Calcium signaling in membrane repair.\" \nSemin Cell Dev Biol 45: 24–31. \nChoi, D., E. Park, E. Jung, Y. J. Seong, M. Hong, S. Lee, J. Burford, G. Gyarmati, J. \nPeti-Peterdi, S. Srikanth, Y. Gwack, C. J. Koh, E. Boriushkin, A. Hamik, A. K. Wong and \nY. K. Hong (2017). \"ORAI1 Activates Proliferation of Lymphatic Endothelial Cells in \nResponse to Laminar Flow Through Kruppel-Like Factors 2 and 4.\" Circ Res\n 120(9): \n1426–1439. \nChubinskiy-Nadezhdin, V., S. Semenova, V. Vasileva, A. Shatrova, N. Pugovkina and Y. \nNegulyaev (2022). \"Store-Operated Ca(2+) Entry Contributes to Piezo1-Induced Ca(2+) \nIncrease in Human Endometrial Stem Cells.\" Int J Mol Sci\n 23(7). \nCooper, S. T. and P. L. McNeil (2015). \"Membrane Repair: Mechanisms and \nPathophysiology.\" Physiol Rev\n 95(4): 1205–1240. \n(which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. \nThe copyright holder for this preprintthis version posted May 15, 2026. ; https://doi.org/10.64898/2026.05.13.724989doi: bioRxiv preprint \n\nCorrotte, M., P . E. Almeida, C. Tam, T. Castro-Gomes, M. C. Fernandes, B. A. Millis, M. \nCortez, H. Miller, W. Song, T. K. Maugel and N. W. Andrews (2013). \"Caveolae \ninternalization repairs wounded cells and muscle fibers.\" Elife 2: e00926. \nCorrotte, M., M. C. Fernandes, C. Tam and N. W. Andrews (2012). \"Toxin pores \nendocytosed during plasma membrane repair traffic into the lumen of MVBs for \ndegradation.\" Traffic 13(3): 483–494. \nDemonbreun, A. R. and E. M. McNally (2016). \"Plasma Membrane Repair in Health and \nDisease.\" Curr Top Membr 77: 67–96. \nDemonbreun, A. R., M. Quattrocelli, D. Y . Barefield, M. V. Allen, K. E. Swanson and E. \nM. McNally (2016). \"An actin-dependent annexin complex mediates plasma membrane \nrepair in muscle.\" J Cell Biol 213(6): 705–718. \nDing, J., J. R. Zhang, Y. Wang, C. L. Li, D. Lu, S. M. Guan and J. Chen (2012). \"Effects \nof a non-selective TRPC channel blocker, SKF-96365, on melittin-induced spontaneous \npersistent nociception and inflammatory pain hypersensitivity.\" Neurosci Bull 28(2): \n173–181. \nDupont, S., L. Morsut, M. Aragona, E. Enzo, S. Giulitti, M. Cordenonsi, F. Zanconato, J. \nLe Digabel, M. Forcato, S. Bicciato, N. Elvassore and S. Piccolo (2011). \"Role of \nYAP/TAZ in mechanotransduction.\" Nature\n 474(7350): 179–183. \nDynes, J. L., A. Amcheslavsky and M. D. Cahalan (2016). \"Genetically targeted single-\nchannel optical recording reveals multiple Orai1 gating states and oscillations in calcium \ninflux.\" Proc Natl Acad Sci U S A 113(2): 440–445. \nElaib, Z., F. Saller and R. Bobe (2016). \"The Calcium Entry-Calcium Refilling Coupling.\" \nAdv Exp Med Biol 898: 333–352. \nFu, R., W. Wang, Y. Huo, L. Li, R. Chen, Z. Lin, Y . Tao, X. Peng, W. Huang and C. Guo \n(2024). \"The mechanosensitive ion channel Piezo1 contributes to podocyte cytoskeleton \nremodeling and development of proteinuria in lupus nephritis.\" Kidney Int 106(4): 625–\n639. \nGarcia, J., J. Bagwell, B. Njaine, J. Norman, D. S. Levic, S. Wopat, S. E. Miller, X. Liu, J. \nW. Locasale, D. Y . R. Stainier and M. Bagnat (2017). \"Sheath Cell Invasion and Trans-\ndifferentiation Repair Mechanical Damage Caused by Loss of Caveolae in the Zebrafish \nNotochord.\" Curr Biol 27(13): 1982–1989 e1983. \nGounou, C., F. Bouvet, B. Liet, V. Prouzet-Mauleon, L. d'Agata, E. Harte, F. Argoul, G. \nSiegfried, R. Iggo, A. M. Khatib and A. Bouter (2023). \"Annexin-A5 and annexin-A6 \nsilencing prevents metastasis of breast cancer cells in zebrafish.\" Biol Cell 115(6): \ne202200110. \nHagenbeek, T. J., J. R. Zbieg, M. Hafner, R. Mroue, J. A. Lacap, N. M. Sodir, C. L. \nNoland, S. Afghani, A. Kishore, K. P. Bhat, X. Yao, S. Schmidt, S. Clausen, M. Steffek, \nW. Lee, P. Beroza, S. Martin, E. Lin, R. Fong, P. Di Lello, M. H. Kubala, M. N. Yang, J. T. \nLau, E. Chan, A. Arrazate, L. An, E. Levy, M. N. Lorenzo, H. J. Lee, T. H. Pham, Z. \nModrusan, R. Zang, Y. C. Chen, M. Kabza, M. Ahmed, J. Li, M. T. Chang, D. Maddalo, \nM. Evangelista, X. Ye, J. J. Crawford and A. Dey (2023). \"An allosteric pan-TEAD \ninhibitor blocks oncogenic YAP/TAZ signaling and overcomes KRAS G12C inhibitor \nresistance.\" Nat Cancer\n 4(6): 812–828. \nHannezo, E. and C. P. Heisenberg (2019). \"Mechanochemical Feedback Loops in \nDevelopment and Disease.\" Cell\n 178(1): 12–25. \n(which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. \nThe copyright holder for this preprintthis version posted May 15, 2026. ; https://doi.org/10.64898/2026.05.13.724989doi: bioRxiv preprint \n\nHigashida, C., T. Kiuchi, Y . Akiba, H. Mizuno, M. Maruoka, S. Narumiya, K. Mizuno and \nN. Watanabe (2013). \"F- and G-actin homeostasis regulates mechanosensitive actin \nnucleation by formins.\" Nat Cell Biol 15(4): 395–405. \nHoernke, M., J. Mohan, E. Larsson, J. Blomberg, D. Kahra, S. Westenhoff, C. \nSchwieger and R. Lundmark (2017). \"EHD2 restrains dynamics of caveolae by an ATP-\ndependent, membrane-bound, open conformation.\" Proc Natl Acad Sci U S A 114(22): \nE4360–e4369. \nHorn, A. and J. K. Jaiswal (2018). \"Cellular mechanisms and signals that coordinate \nplasma membrane repair.\" Cell Mol Life Sci 75(20): 3751–3770. \nJaiswal, J. K., S. P. Lauritzen, L. Scheffer, M. Sakaguchi, J. Bunkenborg, S. M. Simon, \nT. Kallunki, M. Jaattela and J. Nylandsted (2014). \"S100A11 is required for efficient \nplasma membrane repair and survival of invasive cancer cells.\" Nat Commun 5: 3795. \nJardin, I. and J. A. Rosado (2016). \"STIM and calcium channel complexes in cancer.\" \nBiochim Biophys Acta 1863(6 Pt B): 1418–1426. \nLachowski, D., C. Matellan, S. Gopal, E. Cortes, B. K. Robinson, A. Saiani, A. F. Miller, \nM. M. Stevens and A. E. Del Rio Hernandez (2022). \"Substrate Stiffness-Driven \nMembrane Tension Modulates Vesicular Trafficking via Caveolin-1.\" ACS Nano 16(3): \n4322–4337. \nLakk, M. and D. Krizaj (2021). \"TRPV4-Rho signaling drives cytoskeletal and focal \nadhesion remodeling in trabecular meshwork cells.\" Am J Physiol Cell Physiol 320(6): \nC1013–C1030. \nLee, J. H., Y . G. Yun and H. W. Kim (2025). \"Matrix-induced nuclear remodeling and \nmechano-therapeutics.\" Cell Rep 44(9): 116176. \nLeung, C., C. Yu, M. I. Lin, C. Tognon and P. Bernatchez (2013). \"Expression of \nmyoferlin in human and murine carcinoma tumors: role in membrane repair, cell \nproliferation, and tumorigenesis.\" Am J Pathol 182(5): 1900–1909. \nLewis, R. S. (2020). \"Store-Operated Calcium Channels: From Function to Structure \nand Back Again.\" Cold Spring Harb Perspect Biol\n 12(5). \nLolo, F. N., N. Walani, E. Seemann, D. Zalvidea, D. M. Pavon, G. Cojoc, M. Zamai, C. \nViaris de Lesegno, F. Martinez de Benito, M. Sanchez-Alvarez, J. J. Uriarte, A. Echarri, \nD. Jimenez-Carretero, J. C. Escolano, S. A. Sanchez, V. R. Caiolfa, D. Navajas, X. \nTrepat, J. Guck, C. Lamaze, P. Roca-Cusachs, M. M. Kessels, B. Qualmann, M. Arroyo \nand M. A. Del Pozo (2023). \"Caveolin-1 dolines form a distinct and rapid caveolae-\nindependent mechanoadaptation system.\" Nat Cell Biol\n 25(1): 120–133. \nLuan, H., T. A. Bielecki, B. C. Mohapatra, N. Islam, I. Mushtaq, A. M. Bhat, S. Mirza, S. \nChakraborty, M. Raza, M. D. Storck, M. S. Toss, J. L. Meza, W. B. Thoreson, D. W. \nCoulter, E. A. Rakha, V. Band and H. Band (2023). \"EHD2 overexpression promotes \ntumorigenesis and metastasis in triple-negative breast cancer by regulating store-\noperated calcium entry.\" Elife\n 12. \nMarg, A., V. Schoewel, T. Timmel, A. Schulze, C. Shah, O. Daumke and S. Spuler \n(2012). \"Sarcolemmal repair is a slow process and includes EHD2.\" Traffic 13(9): 1286–\n1294. \nMartino, F., A. R. Perestrelo, V. Vinarsky, S. Pagliari and G. Forte (2018). \"Cellular \nMechanotransduction: From Tension to Function.\" Front Physiol 9: 824. \nMcMahon, H. T. and E. Boucrot (2015). \"Membrane curvature at a glance.\" J Cell Sci \n128(6): 1065–1070. \n(which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. \nThe copyright holder for this preprintthis version posted May 15, 2026. ; https://doi.org/10.64898/2026.05.13.724989doi: bioRxiv preprint \n\nMiller, J., C. Bruen, M. Schnaus, J. Zhang, S. Ali, A. Lind, Z. Stoecker, K. Stauderman \nand S. Hebbar (2020). \"Auxora versus standard of care for the treatment of severe or \ncritical COVID-19 pneumonia: results from a randomized controlled trial.\" Crit Care \n24(1): 502. \nMiroshnikova, Y. A., S. Manet, X. Li, S. A. Wickstrom, E. Faurobert and C. Albiges-Rizo \n(2021). \"Calcium signaling mediates a biphasic mechanoadaptive response of \nendothelial cells to cyclic mechanical stretch.\" Mol Biol Cell\n 32(18): 1724–1736. \nMoren, B., C. Shah, M. T. Howes, N. L. Schieber, H. T. McMahon, R. G. Parton, O. \nDaumke and R. Lundmark (2012). \"EHD2 regulates caveolar dynamics via ATP-driven \ntargeting and oligomerization.\" Mol Biol Cell\n 23(7): 1316–1329. \nMoreno-Vicente, R., D. M. Pavon, I. Martin-Padura, M. Catala-Montoro, A. Diez-\nSanchez, A. Quilez-Alvarez, J. A. Lopez, M. Sanchez-Alvarez, J. Vazquez, R. Strippoli \nand M. A. Del Pozo (2018). \"Caveolin-1 Modulates Mechanotransduction Responses to \nSubstrate Stiffness through Actin-Dependent Control of YAP .\" Cell Rep\n 25(6): 1622–\n1635 e1626. \nNarain, R., J. M. Muncie-Vasic and V. M. Weaver (2025). \"Forcing the code: tension \nmodulates signaling to drive morphogenesis and malignancy.\" Genes Dev 39(1-2): 163–\n181. \nNieto-Felipe, J., A. Macias-Diaz, J. Sanchez-Collado, A. Berna-Erro, I. Jardin, G. M. \nSalido, J. J. Lopez and J. A. Rosado (2023). \"Role of Orai-family channels in the \nactivation and regulation of transcriptional activity.\" J Cell Physiol\n 238(4): 714–726. \nOng, H. L., K. P. Subedi, G. Y. Son, X. Liu and I. S. Ambudkar (2019). \"Tuning store-\noperated calcium entry to modulate Ca(2+)-dependent physiological processes.\" \nBiochim Biophys Acta Mol Cell Res 1866(7): 1037–1045. \nPallagi, P., M. Görög, N. Papp, T. Madácsy, Á. Varga, T. Crul, V. Szabó, M. Molnár, K. \nDudás, A. Grassalkovich, E. Szederkényi, G. Lázár, V. Venglovecz, P. Hegyi and J. \nMaléth (2022). \"Bile acid- and ethanol-mediated activation of Orai1 damages pancreatic \nductal secretion in acute pancreatitis.\" J Physiol\n 600(7): 1631–1650. \nPardo-Pastor, C., F. Rubio-Moscardo, M. Vogel-Gonzalez, S. A. Serra, A. Afthinos, S. \nMrkonjic, O. Destaing, J. F. Abenza, J. M. Fernandez-Fernandez, X. Trepat, C. Albiges-\nRizo, K. Konstantopoulos and M. A. Valverde (2018). \"Piezo2 channel regulates RhoA \nand actin cytoskeleton to promote cell mechanobiological responses.\" Proc Natl Acad\n \nSci U S A 115(8): 1925–1930. \nParsonage, G., K. Cuthbertson, N. Endesh, N. Murciano, A. J. Hyman, C. H. Revill, O. \nV. Povstyan, E. Chuntharpursat-Bon, M. Debant, M. J. Ludlow, T. S. Futers, L. \nLichtenstein, J. A. Kinsella, F. Bartoli, M. G. Rotordam, N. Becker, A. Bruggemann, R. \nFoster and D. J. Beech (2023). \"Improved PIEZO1 agonism through 4-benzoic acid \nmodification of Yoda1.\" Br J Pharmacol\n 180(16): 2039–2063. \nParton, R. G. (2018). \"Caveolae: Structure, Function, and Relationship to Disease.\" \nAnnu Rev Cell Dev Biol 34: 111–136. \nPosey, A. D., Jr., P. Pytel, K. Gardikiotes, A. R. Demonbreun, M. Rainey, M. George, H. \nBand and E. M. McNally (2011). \"Endocytic recycling proteins EHD1 and EHD2 interact \nwith fer-1-like-5 (Fer1L5) and mediate myoblast fusion.\" J Biol Chem 286(9): 7379–\n7388. \nRamsey, I. S., M. Delling and D. E. Clapham (2006). \"An introduction to TRP channels.\" \nAnnu Rev Physiol 68: 619–647. \n(which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. \nThe copyright holder for this preprintthis version posted May 15, 2026. ; https://doi.org/10.64898/2026.05.13.724989doi: bioRxiv preprint \n\nRausch, V., J. R. Bostrom, J. Park, I. R. Bravo, Y . Feng, D. C. Hay, B. A. Link and C. G. \nHansen (2019). \"The Hippo Pathway Regulates Caveolae Expression and Mediates \nFlow Response via Caveolae.\" Curr Biol 29(2): 242–255 e246. \nSathish, V., A. J. Abcejo, M. A. Thompson, G. C. Sieck, Y . S. Prakash and C. M. \nPabelick (2012). \"Caveolin-1 regulation of store-operated Ca(2+) influx in human airway \nsmooth muscle.\" Eur Respir J 40(2): 470–478. \nShvets, E., A. Ludwig and B. J. Nichols (2014). \"News from the caves: update on the \nstructure and function of caveolae.\" Curr Opin Cell Biol 29: 99–106. \nSinha, B., D. Koster, R. Ruez, P. Gonnord, M. Bastiani, D. Abankwa, R. V. Stan, G. \nButler-Browne, B. Vedie, L. Johannes, N. Morone, R. G. Parton, G. Raposo, P . Sens, C. \nLamaze and P. Nassoy (2011). \"Cells respond to mechanical stress by rapid \ndisassembly of caveolae.\" Cell\n 144(3): 402–413. \nSotodosos-Alonso, L., M. Pulgarin-Alfaro and M. A. Del Pozo (2023). \"Caveolae \nMechanotransduction at the Interface between Cytoskeleton and Extracellular Matrix.\" \nCells 12(6). \nStauderman, K. A. (2018). \"CRAC channels as targets for drug discovery and \ndevelopment.\" Cell Calcium 74: 147–159. \nStefl, M., M. Takamiya, V. Middel, M. Tekpinar, K. Nienhaus, T. Beil, S. Rastegar, U. \nStrahle and G. U. Nienhaus (2024). \"Caveolae disassemble upon membrane lesioning \nand foster cell survival.\" iScience 27(2): 108849. \nStoeber, M., I. K. Stoeck, C. Hanni, C. K. Bleck, G. Balistreri and A. Helenius (2012). \n\"Oligomers of the ATPase EHD2 confine caveolae to the plasma membrane through \nassociation with actin.\" EMBO J 31(10): 2350–2364. \nSuchyna, T. M., J. H. Johnson, K. Hamer, J. F. Leykam, D. A. Gage, H. F. Clemo, C. M. \nBaumgarten and F. Sachs (2000). \"Identification of a peptide toxin from Grammostola \nspatulata spider venom that blocks cation-selective stretch-activated channels.\" J Gen \nPhysiol 115(5): 583–598. \nSzabo, V., N. Csakany-Papp, M. Gorog, T. Madacsy, A. Varga, A. Kiss, B. Tel, B. Jojart, \nT. Crul, K. Dudas, M. Bagyanszki, N. Bodi, F. Ayaydin, S. Jee, L. Tiszlavicz, K. A. \nStauderman, S. Hebbar, P . Pallagi and J. Maleth (2023). \"Orai1 calcium channel \ninhibition prevents progression of chronic pancreatitis.\" JCI Insight 8(13). \nSzabó, V., N. Csákány-Papp, M. Görög, T. Madacsy, Á. Varga, A. Kiss, B. Tel, B. Jójárt, \nT. Crul, K. Dudás, M. Bagyánszki, N. Bódi, F. Ayaydin, S. Jee, L. Tiszlavicz, K. A. \nStauderman, S. Hebbar, P . Pallagi and J. Maléth (2023). \"Orai1 calcium channel \ninhibition prevents progression of chronic pancreatitis.\" JCI Insight\n. \nTorrino, S., W. W. Shen, C. M. Blouin, S. K. Mani, C. Viaris de Lesegno, P. Bost, A. \nGrassart, D. Koster, C. A. Valades-Cruz, V. Chambon, L. Johannes, P. Pierobon, V. \nSoumelis, C. Coirault, S. Vassilopoulos and C. Lamaze (2018). \"EHD2 is a \nmechanotransducer connecting caveolae dynamics with gene transcription.\" J Cell Biol \n217(12): 4092–4105. \nvan Vliet, A. R., F. Giordano, S. Gerlo, I. Segura, S. Van Eygen, G. Molenberghs, S. \nRocha, A. Houcine, R. Derua, T. Verfaillie, J. Vangindertael, H. De Keersmaecker, E. \nWaelkens, J. Tavernier, J. Hofkens, W. Annaert, P. Carmeliet, A. Samali, H. Mizuno and \nP. Agostinis (2017). \"The ER Stress Sensor PERK Coordinates ER-Plasma Membrane \nContact Site Formation through Interaction with Filamin-A and F-Actin Remodeling.\" Mol \nCell 65(5): 885–899.e886. \n(which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. \nThe copyright holder for this preprintthis version posted May 15, 2026. ; https://doi.org/10.64898/2026.05.13.724989doi: bioRxiv preprint \n\nVaradarajan, S., S. A. Chumki, R. E. Stephenson, E. R. Misterovich, J. L. Wu, C. E. \nDudley, I. S. Erofeev, A. B. Goryachev and A. L. Miller (2022). \"Mechanosensitive \ncalcium flashes promote sustained RhoA activation during tight junction remodeling.\" J \nCell Biol 221(4). \nWaldron, R. T., Y. Chen, H. Pham, A. Go, H. Y. Su, C. Hu, L. Wen, S. Z. Husain, C. A. \nSugar, J. Roos, S. Ramos, A. Lugea, M. Dunn, K. Stauderman and S. J. Pandol (2019). \n\"The Orai Ca(2+) channel inhibitor CM4620 targets both parenchymal and immune cells \nto reduce inflammation in experimental acute pancreatitis.\" J Physiol\n 597(12): 3085–\n3105. \nWei, Y. and W. Li (2021). \"Calcium, an Emerging Intracellular Messenger for the Hippo \nPathway Regulation.\" Front Cell Dev Biol 9: 694828. \nWu, J., L. Liu, T. Matsuda, Y. Zhao, A. Rebane, M. Drobizhev, Y . F. Chang, S. Araki, Y . \nArai, K. March, T. E. Hughes, K. Sagou, T. Miyata, T. Nagai, W. H. Li and R. E. \nCampbell (2013). \"Improved orange and red Ca(2)+/- indicators and photophysical \nconsiderations for optogenetic applications.\" ACS Chem Neurosci\n 4(6): 963–972. \nYang, S., X. Miao, S. Arnold, B. Li, A. T. Ly, H. Wang, M. Wang, X. Guo, M. M. Pathak, \nW. Zhao, C. D. Cox and Z. Shi (2022). \"Membrane curvature governs the distribution of \nPiezo1 in live cells.\" Nat Commun 13(1): 7467. \nYang, Y., L. A. Valencia, C. H. Lu, M. L. Nakamoto, C. T. Tsai, C. Liu, H. Yang, W. \nZhang, Z. Jahed, W. R. Lee, F. Santoro, J. Liou, J. C. Wu and B. Cui (2024). \"Plasma \nmembrane curvature regulates the formation of contacts with the endoplasmic \nreticulum.\" Nat Cell Biol\n 26(11): 1878–1891. \nYeow, I., G. Howard, J. Chadwick, C. Mendoza-Topaz, C. G. Hansen, B. J. Nichols and \nE. Shvets (2017). \"EHD Proteins Cooperate to Generate Caveolar Clusters and to \nMaintain Caveolae during Repeated Mechanical Stress.\" Curr Biol\n 27(19): 2951–2962 \ne2955. \nYeow, I., G. Howard, J. Chadwick, C. Mendoza-Topaz, C. G. Hansen, B. J. Nichols and \nE. Shvets (2017). \"EHD Proteins Cooperate to Generate Caveolar Clusters and to \nMaintain Caveolae during Repeated Mechanical Stress.\" Curr Biol\n 27(19): 2951–\n2962.e2955. \n \n  \n(which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. \nThe copyright holder for this preprintthis version posted May 15, 2026. ; https://doi.org/10.64898/2026.05.13.724989doi: bioRxiv preprint \n\nFigure Legends \nFigure 1.  EHD2 is required for plasma membrane repair. A-B. Demonstration of \nEHD2 requirement for mechanically-induced Ca2+-dependent plasma membrane repair \nin TNBC cell lines. The indicated cell lines were subjected to cell scraping to induce \nmechanical injury to plasma membrane in media containing FITC-dextran (to label cell \nwith damaged plasma membrane) with or without Ca\n2+. At the indicated times after \nwounding at room temperature, cells were rinsed and incubated with propidium iodide to \nlabel cells that had failed to repair. The number of wounded cells with successful repair \n(only FITC-Dextran-labeled) are shown as a percentage of total cells (green cells plus \ngreen and red cells). A. Schematic diagrams of plasma membrane repair assays \ninduced by mechanical injury or Streptolysin O toxin.  B. Quantification of cells with \nsuccessful plasma membrane repair over time. Data represents mean +/- SEM of three \nexperiments, two-way ANOVA, *,p<0.05, **,p<0.01. C-D. Demonstration of EHD2 \nrequirement for the repair of streptolysin O (SLO) induced plasma membrane pores. \nThe indicated cell lines were treated with SLO for 5 min in Ca\n2+ free Tyrode’s buffer on \nice, followed by incubation for 10 min in Ca2+-containing Tyrode’s buffer. Cells were \nincubated with propidium iodide-containing medium and analyzed by FACS. C. \nRepresentative FACS analyses of propidium Iodide (PI) staining after streptolysin O \n(SLO)-induced membrane damage and repair. Cells to the right of the main peak on left \nrepresent those that failed to repair their plasma membrane. D. Quantification of PI \npositive cell population shown in (C). EHD2 KO-mEhd2 represents EHD2-KO MDA-MB-\n231 cells rescued by stable expression of mouse Ehd2. Quantified data shown are from \nthree independent experiments. Welch’s t-test. *,p<0.05, **,p<0.01, ***, p<0.001. \n(which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. \nThe copyright holder for this preprintthis version posted May 15, 2026. ; https://doi.org/10.64898/2026.05.13.724989doi: bioRxiv preprint \n\nFigure 2. Orai1 is required for plasma membrane repair. A. Western blot \nconfirmation of loss of Orai1 expression upon CRISPR/Cas9-mediated KO in MDA-MB-\n231 and Hs578T cells; shown are pools of three clones maintained individually. B-C. \nImpairment of SOCE upon Orai1 KO in TNBC cells. SOCE was measured using \nthapsigargin-induced ER Ca\n2+ depletion. Typical profiles are shown on left and \nquantification of fold-change in peak fluorescence intensity from 3 independent \nexperiments is shown on right. Welch’s t test, ***p<0.001. D-E. Orai1-KO impairs the \nrepair of mechanically induced plasma membrane injury. WT vs. Orai1-KO cells were \nsubjected to mechanical injury to plasma membrane and repair assay was performed as \nin Fig. 1A-B. Representative confocal images are shown in D and quantified data are \nshown in E. Scale bar, 200 µm. Data represents mean +/- SEM of three experiments, \ntwo-way ANOVA, **,p<0.01. F-G. Orai1-KO impairs the repair of SLO-induced \nmembrane injury. Plasma membrane damage using SLO and the repair assay on MDA-\nMB-231 (F) and Hs578T (G) cell lines were as in Fig. 1C-D. Representative confocal \nimages are shown on left. Quantification of PI-stained cells from 3 independent \nexperiments is shown on right. Welch’s t test, ***p<0.001. H-J. Orai1 inhibitors impair \nplasma membrane repair. Panel H shows Thapsigargin (TG)-induced SOCE \nmeasurements (initial peak, Ca\n2+ store release in the absence of extracellular Ca2+; \nsecond peak, SOCE in the presence of extracellular Ca2+) performed on the indicated \ncell lines cultured with or without CM4620 (10 μM, 4 hours pretreatment). Note \nsignificant SOCE inhibition by CM4620 in WT cells but not in Orai1-KO cells, supporting \nOrai1-selelctive effect of CM4620. Data represents mean +/- SEM of three experiments, \nWelch’s t test, ***p<0.001. The I panel shows the impairment of mechanically induced \n(which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. \nThe copyright holder for this preprintthis version posted May 15, 2026. ; https://doi.org/10.64898/2026.05.13.724989doi: bioRxiv preprint \n\nplasma membrane repair by CM4620 in MDA-MB-231. Cells were cultured without or \nwith pretreatment with CM4620 (10 μM) and analyzed for repair as in Fig. 1A-B. Scale \nbar, 200 µm. Data represents mean +/- SEM of three experiments, two-way ANOVA, \n**,p<0.01. The J panel shows the impairment of SLO-induced membrane damage in \nMDA-MB-231. Left panel shows representative FACS analysis of membrane repair in \ncells without or with CM4620 treatment. Right panel shows quantification of cells that \nfailed to repair (PI staining) from 3 independent experiments. Data represents mean +/- \nSEM of three experiments. Welch’s t test, ***, p<0.001. \nFigure 3. Rapid recruitment of EHD2 to plasma membrane at sites of mechanical \nforce application. A. Schematic diagram of force application to plasma membrane \nusing indentation using an atomic force microscope (AFM). MDA-MB231 cells plated on \nglass-bottom dishes were transiently transfected with EHD2-mCherry (Stoeber, Stoeck \net al. 2012) and imaged with a confocal microscope while AFM micro-cantilevers were \nused to apply force at specific locations on the plasma membrane (indentation). Pre-\nindentation, green; indentation, red; post-indentation, blue. B. Representative time-\nlapse image of EHD2-mCherry transfected MDA-MB-231 cells at the indicated times \nbefore, during and after indentation. The yellow box indicates the area of membrane \nindentation. Lower panels show higher-magnification images highlighting the area \naround the indentation. Scale bar, 5 µm. C. Measurement of the force applied to the cell \n(μN; blue) in relation to cantilever indentation (depth in μM, red). D. Quantification of \nEHD2 fluorescence intensity at various time points during plasma membrane \nindentation. E. Fold change of EHD2 fluorescence intensity before and after indentation \n(which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. \nThe copyright holder for this preprintthis version posted May 15, 2026. ; https://doi.org/10.64898/2026.05.13.724989doi: bioRxiv preprint \n\n(1 min). Data points represent cells analyzed through three independent experiments \n(n= 12). Two-way ANOVA; **p<0.01.  \nFigure 4. EHD2 and Orai1 are required for mechanosensitive extracellular calcium \nimport into cytoplasm. MDA-MB231 cells plated on glass-bottom dishes were \ntransiently transfected with the Ca2+ reporter R-GECO1.2. Cells were imaged with a \nconfocal microscope while micro-cantilevers were employed to indent specific locations \non the plasma membrane. A. Representative time-lapse images wherein white circles \nindicate the location of membrane indentation (Left panel). Quantification of Ca2+ \nintensity fold change (Right panel). B. Abrogation of mechanical force-induced Ca2+ \nimport by EHD2-, Orai1- or CAV1-KO and rescue of EHD2-KO cell response with stable \nmouse EHD2 (mEHD2) expression. Left, representative images at baseline and peak of \nAFM indentation. Right, quantification of Ca2+ signals. Shown are fold change in Ca2+ \nreporter (R-GECO1.2) signals in MDA-MB-231 cell lines without (WT) or with the \nindicated genetic perturbations.  Fold change of Ca2+ peak fluorescence intensity after \nindentation relative to basal fluorescence intensity prior to indentation are computed \nfrom three independent experiments (n=14). One-way ANOVA with Dunnett’s multiple \ncomparisons test, ***p<0.001, **p<0.01; ns, not significant. C. Inhibition of mechanical \nforce-induced Ca\n2+ import by Orai1 inhibitor CM5480. Ca2+ reporter R-GECO1.2-\ntransfected MDA-MB-231 cells cultured without or with CM5480 (10 μM; 24 h \npretreatment) were subjected to AFM cantilever indentation and fluorescence intensity \nrecorded over time. Data points represent cells (n= 14) analyzed through three \nindependent experiments. Welch’s t test, ***p<0.001. D. Mechanical force induced Ca\n2+ \nimport recorded by Orai1-linked Ca2+ biosensor and its inhibition by CM5480. Orai1-\n(which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. \nThe copyright holder for this preprintthis version posted May 15, 2026. ; https://doi.org/10.64898/2026.05.13.724989doi: bioRxiv preprint \n\nlinked Ca2+ biosensor (G-GECO1-Orai1)-transfected MDA-MB-231 cells cultured \nwithout or with CM5480 (10 μM) were subjected to indentation with AFM cantilever and \nfluorescence intensity recorded over time. Left, representative images at baseline and \npeak of indentation. Right, quantification of data. Data points represent cells (n= 21) \nanalyzed through three independent experiments. Welch’s t test, ***p<0.001. E. \nInhibition of mechanical force-induced Ca2+ import upon combined STIM1 and STIM2 \nsiRNA knockdown. MDA-MB-231 cells transfected with the Ca2+ reporter (R-GECO1.2) \nand co-transfected with control or STIM1 and STIM2 siRNAs (knockdown verified in Fig. \nS3D) were subjected to AFM cantilever indentation and fluorescence intensity recorded \nover time. Left, representative images at baseline and peak of indentation. Right, \nquantification of data. Data points represent cells (n=18) analyzed through three \nindependent experiments. Welch’s t test, *p<0.05. Scale bar, 5 µm.Figure 5. EHD2 and \nOrai1 are required for mechanosensitive YAP/TAZ pathway activation. A-B. Impact \nof EHD2, Orai1, or CAV1 KO on stiff matrix-induced YAP translocation. The indicated \nMDA-MB231 wildtype (WT), EHD2-KO, EHD2-KO/mEhd2, Orai1-KO, and CAV1-KO cell \nlines were cultured on collagen-coated 24-well Cytosoft Rigidity plates layered with 0.2 \nKPa (soft) or 64 Kpa (stiff) hydrogels, fixed, stained with AF488 conjugated anti-YAP \nantibody and imaged by confocal imaging. Representative images are shown in A \n(magnified images of a single cell shown under each panel to highlight \nnuclea/cytoplasmic localization of YAP) . Quantification of data is shown in B. Image J \nwas used to put masks around the nucleus to quantify fluorescence signals (pixels) \nwithin (nuclear) and outside (cytoplasmic) the mask and the data are shown as the ratio \nof nuclear/cytoplasmic YAP staining signals. Each data point represents cells analyzed \n(which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. \nThe copyright holder for this preprintthis version posted May 15, 2026. ; https://doi.org/10.64898/2026.05.13.724989doi: bioRxiv preprint \n\nin 63 X fields from three independent experiments (at least 39 cells analyzed in each \ngroup). One-way ANOVA with Dunnett’s multiple comparisons test, **p<0.01, \n***p<0.001. C. Impact of Orai1 inhibitors on stiff matrix-induced nuclear localization of \nYAP . MDAM-MB-231 cells plated on stiff matrix, as in A, were treated for last 4 hours \nwith vehicle (DMSO) or Orai1 inhibitors (CM4620 or CM5480; 10 μM) and cells \nprocessed for YAP staining. Nuclear/cytoplasmic ratios of YAP staining were determined \nas in A/B. Representative images are shown on left. Quantified data of nuclear to \ncytoplasmic YAP staining ratio are shown on right. Each data point represents cells \nanalyzed from three experiments (n=24). Welch’s t test, ***p<0.001.  D-E. Impact of \nEHD2 or Orai1-KO on YAP/TAZ-TEAD pathway gene targets. D shows the luciferase \nactivity of transiently transfected YAP/TAZ reporter (pRP-hRluc-8X GTIIC-Luc) in the \nindicated MDA-MB-231 cell lines cultured on stiff matrix as in A. Data represents mean \n+/- SEM of three experiments, each with 6 replicates. Welch’s t test, ** p<0.01. E shows \nRT-qPCR analyses of YAP/TAZ downstream genes, CTGF, CYR61, ANKRD1, CAV1 \nand CAVIN1 in the indicated MDA-MB-231 cell lines cultures on 0.2 or 64 Kpa \nhydrogels. The gene expression values were normalized to GAPDH. Data represents \nmean +/- SEM of three experiments. One-way ANOVA with Dunnett’s multiple \ncomparisons test, ***p<0.001. \nFigure 6. YAP/TAZ-TEAD pathway activity is required for mechanosensitive Ca\n2+ \nimport and membrane repair. A. TEAD inhibition reduces CAV1 expression. MDA-MB-\n231 or Hs578T cells cultured without or with the pan-TEAD inhibitor GNE-7883 (10 μM) \nfor 48 hours were analyzed by WB for CAV1 expression; Hsc70, loading control. Fold-\nchange in signals quantified by densitometry and analyzed using Image J are shown on \n(which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. \nThe copyright holder for this preprintthis version posted May 15, 2026. ; https://doi.org/10.64898/2026.05.13.724989doi: bioRxiv preprint \n\nright. Welch’s t test, *, p<0.05; **, p<0.01. B-C. TEAD inhibition reduces the \nThapsigargin (TG)-induced SOCE. Cells were pretreated with GNE-7883 (10 μM) for 24 \nhours. Left, representative plots; right, quantified data from three experiments. Welch’s t \ntest; **p<0.001. D. TEAD inhibition impairs the mechanical force-induced Ca2+ import. \nAFM cantilevers were used to apply force and measure Ca2+ import from the \nextracellular space using fluorescence intensity of transfected Ca2+ reporter (R-\nRECO1.2) as the readout. Cells were pretreated with GNE-7883 (10 μM) for 24 hours \nwhere indicated. Data points represent cells analyzed through three independent \nexperiments (n=33). Welch’s t test, ***， p<0.001. E-F. TEAD inhibition impairs the repair \nof induced plasma membrane injuries. Shown is the impact of GNE-7883 on the repair \nof mechanically induced (E) and SLO-induced (F) plasma membrane damage, \nassessed as in Fig. 1A-B and Fig. 1C-D, respectively. GNE-7883 pre-treatment (10 μM), \nwhere indicated, was for 72 hours. Representative confocal images (E; top panel) or \nFACS plots (F; right panel) are shown. Quantified data shown are from 3 independent \nexperiments and presented as mean +/- SEM of 3 experiments. Two-way ANOVA or \nWelch’s t test applied to data in E and F, respectively, *p<0.05. Scale bar, 20 µm. \n  \n(which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. \nThe copyright holder for this preprintthis version posted May 15, 2026. ; https://doi.org/10.64898/2026.05.13.724989doi: bioRxiv preprint \n\nSTAR METHODS \nKey resources table \nREAGENT OR RESOURCE SOURCE IDENTIFIER \nAntibodies \nOrai1 Millipore-Sigma Cat# O8264; (Luan, \nBielecki et al. 2023) \nBeta-actin Millipore-Sigma Cat# SAB11305567 \nHsc70 Santa Cruz Cat# sc-7298 \nCaveolin1  BD Biosciences Cat# 610057 \nEHD2 (Luan, Bielecki et al. 2023) N/A \nYAP (IF) Cell Signaling Technology Cat# 14729 \nHRP-conjugated Protein A antibody ThemoFisher Scientific Cat# 101023 \nHRP-conjugated goat anti-mouse antibody ThemoFisher Scientific Cat# 31431 \nChemicals, peptides, and recombinant proteins \nThapsigargin ThermoFisher Scientific Cat# T7459 \nFluo 4 AM ThemoFisher Scientific Cat# 14201 \nCM4620 SelleckChem Cat# S6834 \nCM5480 CalciMedica N/A \nGsMTx4 SelleckChem Cat# P1205 \nYoda2 TOCRIS Cat# 8051 \nTRIzol ThemoFisher Scientific Cat# 15596026 \nCritical commercial assays \nDual-Luciferase® Reporter assay Kit Promega Cat# E1910 \nRT-qPCR kit Qiagen Cat# 204141 \nExperimental models: Cell lines \nMDA-MB-231 ATCC Cat# HTB-26 \nHs578T ATCC Cat# HTB-126 \nMDA-MB-231 EHD2 KO (Luan, Bielecki et al. 2023) N/A \nHs578T EHD2 KO (Luan, Bielecki et al. 2023) N/A \nMDA-MB-231 CAV1 KO (Luan, Bielecki et al. 2023) N/A \nMDA-MB-231 EHD2 KO mEhd2 (Luan, Bielecki et al. 2023) N/A \nRecombinant DNA \nCMV-R-GECO1.2 (Wu, Liu et al. 2013) Addgene# 45494 \n(which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. \nThe copyright holder for this preprintthis version posted May 15, 2026. ; https://doi.org/10.64898/2026.05.13.724989doi: bioRxiv preprint \n\nG-GECO1-Orai1 Joseph L Dynes et al. 2015 \n(Dynes, Amcheslavsky et al. \n2016) \nAddgene# 73561 \nEHD2-mChery GeneCopoeia Cat# EX-A3485-Lv155 \npLenti-U6-sgRNA-SFFV-Cas9-2A-Puro \n(Orai1 KO generation) \nAbm Cat# 35720125 \nSoftware and algorithms \nImageJ https://imagej.net/software/imag\nej/index \nN/A \nFlowJo 10 FlowJo N/A \nZen Zeiss N/A \nPrism 9 GraphPad N/A \nOligonucleotides \nSee Table S1 for RT-qPCR primers N/A N/A \n \nEXPERIMENTAL MODEL AND SUBJECT DETAILS \nCell lines and medium \nMDA-MB-231 cell line (obtained from ATCC) was cultured in complete α -MEM medium \nwith 5% fetal bovine serum, 10 mM HEPES, 1 mM each of sodium pyruvate, \nnonessential amino acids, and L-glutamine, 50 μ M 2-ME, and 1% penicillin/ \nstreptomycin (Life Technologies, Carlsbad, CA). Hs578T cell line (ATCC) was cultured \nin \nα -MEM medium supplemented as above plus 1 μ g/mL hydrocortisone and 12.5 \nng/mL epidermal growth factor (Millipore Sigma, St. Louis, MO). Generation and \nmaintenance of EHD2-KO TNBC cell lines, EHD2-KO cell lines reconstituted with \nmouse EHD2 (EHD2-KO-mEHD2) and CAV1-KO cell lines have been described \npreviously (Luan, Bielecki et al. 2023). \nAntibodies and reagents \n(which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. \nThe copyright holder for this preprintthis version posted May 15, 2026. ; https://doi.org/10.64898/2026.05.13.724989doi: bioRxiv preprint \n\nAntibodies used for immunoblotting were as follows: Orai1 (# O8264) and beta-actin (# \nSAB1305567) from Millipore-Sigma; HSC70 (# sc-7298) from Santa Cruz \nBiotechnology; Caveolain-1 (#610057) from BD Biosciences; Cavin 1(#46379) from Cell \nSignaling Technology. Horseradish peroxidase (HRP)-conjugated Protein A or HRP-\nconjugated goat anti-mouse secondary antibody for immunoblotting were from \nInvitrogen. YAP antibody (Alexa Fluor® 488 Conjugate, #14729) for \nimmunofluorescence (IF) staining was from Cell Signaling Technology. Thapsigargin (# \nT7459) and Fluo 4AM (#14201) were from ThermoFisher Scientific. Orai1 inhibitor \nCM4620 (Waldron, Chen et al. 2019) was from SelleckChem (#S6834); Orai1 inhibitor \nCM5480 (Szabo, Csakany-Papp et al. 2023) was provided by CalciMedica Inc. (La \nJolla, CA).  \nTransfection reagents and plasmids \nXtremeGENE 9 transfection reagent was from Roche Applied Science (Indianapolis, \nIN); CMV-R-GECO1.2 (Wu, Liu et al. 2013) was a gift from Robert Campbell (Addgene \nplasmid # 45494 ; http://n2t.net/addgene:45494 ; RRID:Addgene_45494).\n G-GECO1-\nOrai1 (Dynes, Amcheslavsky et al. 2016) was a gift from Michael Cahalan (Addgene \nplasmid # 73561; http://n2t.net/addgene:73561; RRID:Addgene_73561). EHD2-mChery \nplasmid (# EX-A3485-Lv155) was from GeneCopoeia. \nGeneration of CRISPR-Cas9 knockout cell lines \nAll-in-One sgRNA CRISPR/Cas9 Lentivectors from Applied Biological Materials \n(Richmond, BC, Canada) were used to derive Orai1 (pLenti-U6-sgRNA-SFFV-Cas9-2A-\nPuro, #35720125) KO cell lines. \n(which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. \nThe copyright holder for this preprintthis version posted May 15, 2026. ; https://doi.org/10.64898/2026.05.13.724989doi: bioRxiv preprint \n\nSOCE assay \nCells were seeded in 35 mm glass-bottom dishes (cat. #FD35-100, WPI Inc) and loaded \nwith Fluo4-AM in modified Tyrode’s solution (2 mM calcium chloride, 1 mM magnesium \nchloride, 137 mM sodium chloride, 2.7 mM potassium chloride, 12 mM sodium \nbicarbonate, 0.2 mM sodium dihydrogen phosphate, 5.5 mM glucose, pH 7.4) for 1 hour \nat 37\noC. After washing with Ca2+-free Tyrode’s solution, live cells were imaged under a \nconfocal microscope (LSM710; Carl Zeiss), with fluorescence excitation at 488 nm and \nemission at 490–540 nm. To initiate Ca2+ release from intracellular stores, 2.5 μ M \nthapsigargin was added in the absence of extracellular Ca2+. Once the Ca2+ signals \napproached the baseline, calcium chloride was added to 2 mM final concentration to \nrecord the SOCE. Data is presented as fold change in fluorescence emission relative to \nbaseline.  \nMembrane repair assay \nFor mechanical injury, confluent cell monolayers in 48-well plates were incubated in \nTyrode’s solution containing membrane-impermeable FITC-Dextran (500 µg/mL). A \nstandardized mechanical injury was introduced via a single scratch wound using a 200 \nµL pipette tip, allowing dye entry into cells with compromised plasma membranes. At \ndesignated time points post-injury, the extracellular dye was removed followed by a PBS \nwash. Cells were then briefly incubated with propidium iodide (PI; 50 µg/mL) to label the \nnuclei of cells with unrepaired plasma membrane defects. Imaging was performed via \nfluorescence microscopy. FITC+ cells represented the total population of injured cells. \nFITC+PI+ (double-positive) cells were scored as those with failure to repair their \n(which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. \nThe copyright holder for this preprintthis version posted May 15, 2026. ; https://doi.org/10.64898/2026.05.13.724989doi: bioRxiv preprint \n\nmembrane defects. The repair efficiency was calculated as the percentage of injured \n(FITC+) cells that excluded PI (FITC+PI-) at each time point. For Streptolysin O (SLO) \ninduced membrane injury, 106 trypsin/EDTA-released and washed cells were incubated \nwith SLO (25 U/mL) in suspension for 5 min at 4°C in 250 μ l of Ca2+-free Tyrode’s \nsolution followed by resuspension in 37°C Tyrode’s solution for 10 min and PI staining. \nAfter flow cytometry (FACSCalibur; Becton Dickinson) of at least 10,000 cells, the data \nwere analyzed using the FlowJo software (Tree Star, Inc.). \nPlasma membrane indentation using Atomic Force Microscopy (AFM) \nCells were cultured on 35 mm glass bottom dishes and transfected with R-GECO1.2 \n(Wu, Liu et al. 2013) or EHD2-mCherry (Stoeber, Stoeck et al. 2012) plasmid. Live cell \nimaging was conducted with cells were in CO\n₂ -independent medium supplemented with \n10% FBS and 1% penicillin–streptomycin and maintained at 37 °C with a JPK Petri Dish \nHeater to preserve the physiological conditions. Indentation was carried out on a Bruker \nCellHesion 200 AFM using a pyramidal-tipped microcantilever (FM-10, NanoAndMore). \nA loading rate of 3 µm/s and a maximum contact force of 100 nN were applied, with \neach indentation held for 20 s. Simultaneous high-resolution imaging was performed \nusing a Zeiss LSM 900 confocal microscope. AFM force–distance data were processed \nin JPK Data Processing (JPK-DP), while confocal images were analyzed using \nFIJI/ImageJ. Statistical analyses and plotting were conducted in OriginLab Pro. \nImmunofluorescence microscopy \nCells were cultured on collagen (#5005, Advanced Biomatrix) pre-coated glass bottom \nCytoSoft® Imaging 24-Well Plate of different stiffness (#5183 for 0.2 kPa, #5189 for 64k \n(which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. \nThe copyright holder for this preprintthis version posted May 15, 2026. ; https://doi.org/10.64898/2026.05.13.724989doi: bioRxiv preprint \n\nPa, Advanced Biomatrix) to about 50% confluency, fixed with 4% PFA/PBS (10 min), \nblocked with 5% BSA/PBS (60 min), and incubated with Fluorescent conjugated \nantibodies in 5% BSA/PBS overnight at 4 °C. Nuclei were visualized with Hoechst \n33342 (#62249, ThermoFisher Scientific) staining. Fluorescence images were captured \non a Zeiss LSM-800 confocal microscope (63X objective) and analyzed using the ZEN \nsoftware (Zeiss). \nWestern blotting \nCells were lysed in Triton-X-100 lysis buffer (50 mM Tris pH 7.5, 150 mM NaCl, 0.5% \nTriton-X-100, 1 mM PMSF, 10 mM NaF, and 1 mM sodium orthovanadate). Lysates \nwere rocked at 4 °C for at least 1 hr, spun in a microfuge at 13,000 rpm for 20 min at 4 \n°C and supernatant protein concentration determined using the BCA assay kit (Thermo \nFisher Scientific, Rockford, IL). 50\n/i3μ g aliquots of lysate proteins were resolved on \nsodium dodecyl sulfate-7.5% or 12% polyacrylamide gel electrophoresis (SDS-PAGE), \ntransferred to polyvinylidene fluoride (PVDF) membrane, and immunoblotted with the \nindicated antibodies. \nTrans-well migration assay \nCells grown in 0.5% FBS-containing starvation medium for 24 h were trypsinized and \nseeded at 10\n4 on top chambers of 24-well plate trans-wells (# 353097, Corning) in 200 \nμ L of growth factor deprived medium. After 3 h, medium containing 10% FBS was \nadded to lower chambers and trans-wells incubated at 37oC for 16 h. The non-migrated \ncells on the upper surface of membranes were removed with cotton swabs, and the \nmigrated cells on the lower surface methanol-fixed and stained in 0.5% crystal violet in \n(which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. \nThe copyright holder for this preprintthis version posted May 15, 2026. ; https://doi.org/10.64898/2026.05.13.724989doi: bioRxiv preprint \n\nmethanol. Six random 10× fields per insert were photographed, and cells counted using \nthe ImageJ software. Each experiment was run in triplicates and repeated three times. \nYAP/TAZ-TEAD pathway luciferase reporter assay \nCells were transfected with a synthetic TEAD dual luciferase reporter (pRP-hRluc-8X \nGTIIC-Luc, cat# VB250204-1311 from VectorBuilder). All luciferase emission \nmeasurements were performed using a Dual-Luciferase® Reporter assay Kit (DLR™ \nassay, Promega). Luminescence was recorded using a GloMax® luminometer \n(Promega). \nQuantitative real-time PCR \nTotal RNA was extracted using TRIzol reagent (#15596026, Invitrogen), reverse \ntranscribed using a real-time Quantitative PCR kit (#204141, Qiagen) and used for real-\ntime QPCR with primers listed in Table 1.   \nTable 1. Primer sequences for RT-qPCR \n Forward Reverse \nCYR61 ATGGTCCCAGTGCTCAAAGA GGGCCGGTATTTCTTCACAC \nCTGF CAGCATGGACGTTCGTCTG AACCACGGTTTGGTCCTTGG \nANKRD1 ACGCCAAAGACAGAGAAGGA TTCTGCCAGTGTAGCACCAG \nCAV1 TTCTGGGCTTCATCTGGCAAC GCTCAGCCCTATTGGTCCACTTTA \nCAVIN1 ATCAAGAAGCTGGAGGTCAACGAG TCTCAGGTTTTCCTTGGTCTTGA \nGAPDH AACTGCTTAGCACCCCTGGC ATGACCTTGCCCACAGCCTT \n \nStatistical analysis  \nGraphPad Prism software (version 9) was used to perform statistical analyses. \nStatistical analysis of cell biological data was performed by comparing groups using \nunpaired Welch’s t-test (two groups), one-way ANOVA with Dunnett’s multiple \n(which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. \nThe copyright holder for this preprintthis version posted May 15, 2026. ; https://doi.org/10.64898/2026.05.13.724989doi: bioRxiv preprint \n\ncomparisons test (more than two groups) and two-way ANOVA test (two factors). p \nvalues of <0.05 were considered significant. \n \n(which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. \nThe copyright holder for this preprintthis version posted May 15, 2026. ; https://doi.org/10.64898/2026.05.13.724989doi: bioRxiv preprint \n\nA\n*\n**\n*\n**\nFig 1 \nB MDA-MB-231\nHs578T\nWT                         \nEHD2 KO               \nEHD2 KO mEhd2 \nMDA-MB-231 Hs578T\nC\nNormalized to Mode\nNormalized to Mode\nWT                         \nEHD2 KO               \nD\n***\nns\n***\n0\n10\n20\n30\n40\n50\nWT EHD2 KO EHD2  KO\nmEhd2\nPI+ cells (%)\nMDA-MB-231\n0\n20\n40\n60\n80\nWT EHD2 KO\nPI+ cells (%)\nHs578T\nAssay of mechanically-induced PM breaches\nAssay of pore-forming toxin induced PM breaches\nSLO\n+ PI\nPIScratch woundDextran\nPropidium Iodide-APropidium Iodide-A\nPlasma membrane repair (%)Plasma membrane repair (%)\n(which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. \nThe copyright holder for this preprintthis version posted May 15, 2026. ; https://doi.org/10.64898/2026.05.13.724989doi: bioRxiv preprint \n\nFig S1 \nWT EHD2 KO WT EHD2 KO\nWithout Calcium\nDextranMerge\nWith Calcium\nPropidium Iodide\nMDA-MB-231A\nB\nDextranMerge Propidium Iodide\nWT EHD2 KO WT EHD2 KO\nWithout Calcium With Calcium\nHs578T\n(which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. \nThe copyright holder for this preprintthis version posted May 15, 2026. ; https://doi.org/10.64898/2026.05.13.724989doi: bioRxiv preprint \n\nFig 2 \n0\n1\n2\n3\n4\n0 2 4 6 8 10\nFt/F0\nTime (min)\nMDA-MB-231\nWT\nOrai1 KO\nTg 2.5 μM\nCa2+ 2 mM\nA B\nMDA-MB-231\nOrai1\nβ-actin\nWT KO\nHs578T\nWT KO\n0\n1\n2\n3\n4\n5\n0 2 4 6 8 10\nFt/F0\nTime (min)\nHs578T\nWT\nOrai1 KO\n***\n***\n***\n**\nMDA-MB-231Hs578T\nWT                       Orai1 KO\n** **\nC D\nTg 2.5 μM\nCa2+ 2 mM\nDextran PI\nMDA-MB-231\nHs578T\nE\nWT\nOrai1 KO\nPropidium Iodide-A\nPropidium Iodide-A\nNormalized to Mode Normalized to Mode 0\n10\n20\n30\n40\n50\nWT Orai1 KO\nPI+ cells (%)\nMDA-MB-231\n0\n10\n20\n30\n40\n50\n60\nWT Orai1 KO\nPI+ cells (%)\nHs578T\n***\n***\nF\nMDA-MB-231\nHs578T\nH\n0\n1\n2\n3\n4\n0 5 10\nFt/F0\nTime (min)\nMDA-MB-231\nWT\nOrai1 KO\nWT + CM4620\nOrai1 KO +\nCM4620\nTg 2.5μM\nCa2+ 2mM\n*\nns\n***\nns\n***\n***\nMDA-MB-231\n**\nCM4620 Control\nDextran PI MDA-MB-231\nI\n0\n10\n20\n30\n40\n50\nControl CM4620\nPI positive (%)\nMDA-MB-231\n***Control\nCM4620\nMDA-MB-231\nWT\nOrai1 KO\nG\nJ\nNormalized to Mode\nPropidium Iodide-A\nMDA-MB-231\nHs578T\nPlasma membrane repair (%)Plasma membrane repair (%)\nPlasma membrane repair (%)\n(which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. \nThe copyright holder for this preprintthis version posted May 15, 2026. ; https://doi.org/10.64898/2026.05.13.724989doi: bioRxiv preprint \n\nFig S2 \nOrai1 KO WT\nDextran Merge\nOrai1 KO WT\nB\nMDA-MB-231Hs578T\nA\n0\n500\n1000\n1500\nWT Orai1 KO\nMigrated cells\nHs578T\n0\n500\n1000\n1500\nWT Orai1 KO\nMigrated cells\nMDA-MB-231\n*** ***\nC\n0\n200\n400\n600\n800\nDMSO CM4620\nMigrated cells\nMDA-MB-231\nWT\nOrai1 KO\n******\nns\nCM4620 Control\nDextran Merge\nD\nPropidium Iodide\nPropidium Iodide\n(which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. \nThe copyright holder for this preprintthis version posted May 15, 2026. ; https://doi.org/10.64898/2026.05.13.724989doi: bioRxiv preprint \n\nFig 3 \n(which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. \nThe copyright holder for this preprintthis version posted May 15, 2026. ; https://doi.org/10.64898/2026.05.13.724989doi: bioRxiv preprint \n\nFig 4 \nA\nB\nC\nMDA-MB-231 R-GECO-1\nMD-MB-231 R-GECO-1\n***\n**\n**\n***\nns\nMD-MB-231\nD\n*\nMDA-MB-231 R-GECO-1\nE\n***\nMDA-MB-231  G-GECO1-Orai1\nMD-MB-231\nBaseline Peak\nWTEHD2 KOORAI1 KOCAV1 KOmEhd2ControlCM5480\nBaseline Peak\nMD-MB-231\nNTCSTIM1/2 KD\nMD-MB-231\nBase line Peak\nControlCM5480\nMD-MB-231\nBase line Peak\nCa2+ intensity (R-GECO-1)\nCa2+ intensity (G-GECO1-Orai1)\nCa2+ intensity (R-GECO-1)\nCa2+ intensity (R-GECO-1)\nBright field R-GECO-1\n(which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. \nThe copyright holder for this preprintthis version posted May 15, 2026. ; https://doi.org/10.64898/2026.05.13.724989doi: bioRxiv preprint \n\nFig S3\nNTC  STIM1/2 KD\nSTIM1\nSTIM2\nβ-actin\nSTIM1\nβ-actin\n-70\n-100\nSTIM2\nMDA-MB-231\nSTIM1 siRNA              -          50        100     200       -         -        -           -         pmol\nSTIM2 siRNA              -           -             -         -         -        50     100      200      pmol\n-100\n-130\n-70\n-100\n-130\nC\n0\n1\n2\n3\n4\n5\n0 100 200 300\nFt/F0\nTime (s)\nMDA-MB-231 G-GECO1-Orai1\nControl\nCM5480\nTg 2.5 μM\nCa2+ 2 mM\nB\nA\n0\n1\n2\n3\n4\n5\n6\n0 2 4 6 8 10\nFt/F0\nTime (min)\nMDA-MB-231\nWT\nOrai1 KO\nWT + CM5480\nOrai1 KO + CM5480\nTg 2.5 μM\nCa2+ 2 mM\n0\n200\n400\n600\n800\n1000\n1200\nDMSO CM5480\nMigrated cells\nMDA-MB-231\nWT\nOrai1 KO\n***\nns\nD\n(which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. \nThe copyright holder for this preprintthis version posted May 15, 2026. ; https://doi.org/10.64898/2026.05.13.724989doi: bioRxiv preprint \n\nEHD2 KO \nmEhd2\nMDA-MB-231    YAP/DAPI\nWT EHD2 KO Orai1 KO Cav1 KO\n0.2 kPa\n64 kPa\nFig 5 \nA\nC\nB\n***\n***\n***\n***\n***\n***\n***\n***\n 0.2kPa\n 64kPa\n**\n0\n0.2\n0.4\n0.6\n0.8\n1\n1.2\nWT EHD2 KO Orai1 KO\nReoaltive lucifearse activity\nMDA-MB-231\n**\n**\nD\n0\n0.5\n1\n1.5\n2\nWT EHD2  KO Orai1 KO\nRelative  expression\nCTGF\n0\n0.5\n1\n1.5\n2\nWT EHD2  KO Orai1 KO\nRelative  expression\nCYR61\n0\n0.5\n1\n1.5\n2\n2.5\nWT EHD2  KO Orai1 KO\nRelative  expression\nANKRD1\n0\n1\n2\n3\n4\nWT EHD2 KO Orai1 KO\nRelative expression \nCAV1\n0\n1\n2\n3\nWT EHD2 KO Orai1 KO\nRelative expression \nCAVIN1\n***\n***\n***\n***\n***\n***\n***\n***\n***\n***\n***\n***\n***\n***\n***\n0.2kPa\n64kPa\n0.2kPa\n64kPa\n0.2kPa\n64kPa\n0.2kPa\n64kPa\n0.2kPa\n64kPa\nE\nCM4620CM5480 DMSO\nMDA-MB-231    YAP DAPI\n***\n***\nMDA-MB-231\nNuclear/Cytoplasmic YAP intensity\nNuclear/Cytoplasmic YAP intensity\n(which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. \nThe copyright holder for this preprintthis version posted May 15, 2026. ; https://doi.org/10.64898/2026.05.13.724989doi: bioRxiv preprint \n\nFig S4 \nEHD2 KO \nmEhd2 WTEHD2 KOOrai1 KOCav1 KO\nYAP DAPI MERGE\n64 kPa\nYAP DAPI MERGE\n0.2 kPa\nCM4620CM5480 DMSO\nMDA-MB-231\nA\nB\nYAP DAPI MERGE\nMDA-MB-231\n(which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. \nThe copyright holder for this preprintthis version posted May 15, 2026. ; https://doi.org/10.64898/2026.05.13.724989doi: bioRxiv preprint \n\nFig 6 \nA\nB\nC\nGNE-7883\nCav1\nHsc70\nMDA-MB-231 Hs578T\n-       +       -       +\n0\n2\n4\n6\n8\n10\n0 100 200 300 400 500\nFt/F0\nTime (s)\nMDA-MB-231\nControl\nGNE-7883\n0\n1\n2\n3\n4\n5\n0 100 200 300 400 500\nFt/F0\nTime (s)\nHs578T\nControl\nGNE-7883\nTg 2.5 μM\nCa2+ 2 mM\nTg 2.5 μM\nCa2+ 2 mM\n**\n**\nF\n0\n5\n10\n15\n20\nControl GNE-7883\nPI+ cells (%)\nMDA-MB-231\n0\n20\n40\n60\n80\nControl GNE-7883\nPI+ cells (%)\nHs578T\nMDA-MB-231\nControl\nGNE-7883\nControl\nGNE-7883\nHs578T\n*\n*\nE\nGNE-7883Control\nMDA-MB-231Hs578T\nDextran PI\n**\nMDA-MB-231\n***\nD\nMDA-MB-231\nHs578T\nPropidium Iodide-A\nNormalized to Mode\nPropidium Iodide-A\nNormalized to Mode\nns\nns\nCav1\n** *\nMDA-MB-231\nHs578T\nBase line Peak\nControlGNE7883\nMD-MB-231\nCa2+ intensity (R-GECO-1)\nPlasma membrane repair (%)Plasma membrane repair (%)\n(which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. \nThe copyright holder for this preprintthis version posted May 15, 2026. ; https://doi.org/10.64898/2026.05.13.724989doi: bioRxiv preprint \n\nFig S5 \nGNE-7883 Control\nDextran Merge\nGNE-7883 Control\nMDA-MB-231Hs578T\nPropidium Iodide\n(which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. \nThe copyright holder for this preprintthis version posted May 15, 2026. ; https://doi.org/10.64898/2026.05.13.724989doi: bioRxiv preprint","source_license":"CC-BY-4.0","license_restricted":false}