Filamin A C-terminal fragment modulates Orai1 expression by inhibition of protein degradation

RESEARCH ARTICLE Filamin A C-terminal fragment modulates Orai1 expression by inhibition of protein degradation Alvaro Macias-Díaz, 1 Joel Nieto-Felipe, 1 Isaac Jardín, 1 Pedro J. Camello, 1 Eva M. Martinez-Quintana, 4 Gines M. Salido, 1 Tarik Smani, 2,3 Jose J. Lopez, 1 and Juan A. Rosado 1  1 Department of Physiology (Cellular Physiology Research Group), Institute of Molecular Pathology Biomarkers (IMPB), University of Extremadura, Caceres, Spain; 2 Group of Cardiovascular Pathophysiology, Institute of Biomedicine of Seville, University Hospital of Virgen del Rocío/University of Seville/CSIC, Seville, Spain; 3 Department of Medical Physiology and Biophysics, Faculty of Medicine, University of Seville, Seville, Spain; and 4 Pathology Service, University Hospital of Caceres, Caceres, Spain Abstract Filamin A (FLNA) is an actin-binding protein that has been reported to interact with STIM1 modulating the activation of Orai1 channels. Cleaving of FLNA by calpain leads to a C-terminal fragment that is involved in a variety of functional and pathological events, including pro-oncogenic activity in different types of cancer. Here, we show that full-length FLNA is downregulated in samples from patients with colon cancer as well as in the adenocarcinoma cell line HT-29. This is consistent with an increased calpain-dependent FLNA cleaving with enhanced expression of the C-terminal FLNA fragment accompanied by enhanced expression of Orai1 and STIM1, as well as store-operated Ca 2þ entry (SOCE). To further explore the mechanism underlying the enhancement of SOCE by the C-terminal FLNA fragment, we expressed in HEK-293 cells the C-terminal FLNA region encompassing repeats 16–24 (FLNA 16–24 fragment), which enhancedbothOrai1andSTIM1aswellasSOCE.TransfectionoftheFLNA 16–24 fragment attenuates Orai1 and STIM1 protein degradation,and,specifically, abrogates Orai1alysosomal degradation and retains this channel in the plasma membrane. However, the C-terminal FLNA fragment did not induce a detectable modification in Orai1bdegradation. Due to the relevance of SOCE in cell physiology, our results provide evidence of a novel mechanism for the regulation of Ca 2þ influx with relevant pathophysiological implications. NOTE & NOTEWORTHY FLNA cleaving by calpain has been observed in a variety of tumoral, including prostate and colorectal cancer cells, as well as in nontumoral cells, leading to a C-terminal fragment encompassing repeats 16–24. Expression of the FLNA 16–24 fragment in HEK-293 cells enhances Orai1 and STIM1 expression, as well as SOCE, a mechanism mediated by attenuation of Orai1a and STIM1 degradation, providing evidence for a novel mechanism for the regulation of SOCE in normal and malignant cells. FLNA; Orai1a; Orai1 b ; store-operated Ca 2þ entry INTRODUCTION Orai1 is the pore-forming subunit of the CRAC (Ca 2þ release-activated Ca 2þ ) channels, a highly selective Ca 2þ channel, ubiquitously expressed, that mediate store-operated Ca 2þ entry (SOCE), a major pathway for agonist-induced Ca 2þ mobilization (1–3). SOCE plays a relevant role supporting a variety of cellular functions, including gene expression, cell proliferation, differentiation, lactation, and platelet aggregation (4–7). SOCE is activated upon discharge of the intracellular Ca 2þ stores, mainly, but not exclusively (8), the endoplasmic reticulum (ER), leading to dissociation of Ca 2þ from the Ca 2þ sensor proteins, STIM1 and STIM2, two single-pass transmembrane proteins with highly conserved luminal EF-hand motifs (9). Ca 2þ dissociation from the EF-hand motif leads to STIM proteins oligomerization and translocation to regions close to the plasma membrane (PM). This, together with a well-described conformational change, results in the association of STIM proteins with Orai channels in the PM, to initiate Ca 2þ influx (10– 14). Two Orai1 paralogs, Orai2 and Orai3, have been proposed to modulate Ca 2þ influx through the CRAC channels (15). Furthermore, two Orai1 variants have been identified in mammalian cells generated by alternative initiation of translation, giving rise to the canonical full-length Orai1 protein, designated as Orai1a, which comprises 301 amino acids, and a short variant, termed Orai1b, that originates from an alternative translation initiation at either methionine 64 or 71 (16). Although both variants show similar efficacy in supporting the highly Ca 2þ -selective I CRAC currents (17), important functional and biophysical differences have been reported between Orai1aand Orai1b.Forinstance,althoughOrai1a A. Macias-Díaz and J. Nieto-Felipe contributed equally to this work. T. Smani and J. A. Rosado contributed equally as cosenior authors. Correspondence: A. Macias-Diaz ([email protected]); J. J. Lopez ( [email protected]); J. A. Rosado ([email protected]). Submitted 8 October 2024 / Revised 16 December 2024 / Accepted 30 December 2024 http://www.ajpcell.org 0363-6143/25 Copyright ©2025 The Authors. Licensed under Creative Commons Attribution CC-BY 4.0. Published by the American Physiological Society. C657 Am J Physiol Cell Physiol 328: C657–C669, 2025. First published January 7, 2025; doi:10.1152/ajpcell.00745.2024 Downloaded from journals.physiology.org/journal/ajpcell at Univ De Sevilla (193.147.173.206) on June 13, 2025. supports the less Ca 2þ -selective store-operated current I SOC and the arachidonate-regulated current I ARC (17), Orai1bdoes not participate in I ARC ,anditsroleinI SOC is cell-type-specific (18). Furthermore, in contrast to Orai1b,Orai1ais required for NF-κB transcriptional activity (19). Biophysically, Orai1ais more susceptible to rapid Ca 2þ -dependent inactivation than Orai1b(17). Among the proteins involved in the modulation of STIM1Orai1 interaction and the activation of SOCE, filamin A (FLNA) has been reported to interact with STIM1 thus modulating the activation of Orai1 channels (20). FLNA is an 280 kDa actin-binding protein that supports orthogonal branching of actin microfilaments and stabilizes the cortical actin network (21). Human FLNA comprises an amino-terminal actin-binding domain and 24 immunoglobulin (Ig)-like repeats (22). FLNA is cleaved by calpain into a 190 kDa fragment, containing the actin-binding domain and the first 15 Ig-like repeats and a second fragment of 110 kDa, comprising the remaining Ig-like repeats; the latter is further cleaved to a 90 kDa fragment (23,24). The short C-terminal FLNA fragment has been reported to be involved in a variety of functional and pathological events. This fragment translocate to the nucleus and modulates androgen receptor transcriptional activity (23,24). Furthermore, FLNA cleaving promote angiogenesis by facilitating the nuclear translocation of several transcription factors (25). In macrophages, FLNA C-terminal fragment interacts with STAT3 and enhances its phosphorylation and nuclear translocation (26). FLNA C-terminal fragment has been shown to support intrahepatic cholangiocarcinoma progression and inhibition of calpain by calpeptin impairs cell growth in a variety of human and mouse tumors, including human melanoma and prostate cancer and mouse fibrosarcoma (27,28). Colorectal adenocarcinoma is among the most common cancer types in men and women. Colorectal adenocarcinoma cells exhibit enhanced SOCE associated with abnormal expression of the Orai and STIM isoforms and TRPC1 (29), which play a relevant role in the development of different cancer hallmarks, including cell migration and survival and apoptosis resistance (29,30). The objective of this study is to investigate the functional role of the FLNA C-terminal fragment in the regulation of SOCE. Here, we show that FLNA is downregulated in patients with colon adenocarcinoma and is cleaved both in colorectal adenocarcinoma HT-29 cells and normal colon mucosa cells. Inhibition of FLNA cleaving by calpeptin in HT-29 cells and expression of the FLNA 16–24 fragment in HEK-293 cells strongly suggest that the FLNA C-terminal fragment is involved in the modulation of Orai1 and STIM1 expression and the activation of SOCE. Enhancement of Orai1 expression by the FLNA 16–24 fragment involves inhibition of Orai1adegradation. These observations provide the firstevidencefortheroleofFLNAC-terminalfragment in the regulation of Orai1 expression and function. MATERIALS AND METHODS Reagents and Antibodies Fura-2 acetoxymethyl ester (fura-2/AM) was from Molecular Probes (Leiden, The Netherlands). High-glucose Dulbecco’s modified Eagle’s medium (DMEM), fetal bovine serum, trypsin, penicillin/streptomycin, Trizma base, rabbit polyclonal anti-STIM2 antibody (catalog number STIM2-201AP, epitope: amino acids 600–650 of human STIM2), mouse monoclonal anti-FLNA antibody [clone: FLMN01 (PM6/317), catalog number MA5-11705], mouse monoclonal anti-PMCA antibody (clone 5F10; catalog number MA3-914, epitope: amino acids 724–783 of human PMCA; RRID: AB_2061566), SuperSignal West Dura extended duration substrate reagent and Pierce BCA protein assay kit, high-capacity streptavidin agarose resin, EZ-Link Sulfo-NHS-LC-Biotin, and Live/Dead viability/ cytotoxicity kit were purchased from Thermo Fisher Scientific (Waltham, MA). Complete EDTA-free protease inhibitor cocktail tablets (Reference name: COEDTAF-RO) were from Roche Diagnostics GmbH (Mannheim, Germany). DharmaFECT kb transfection reagent was obtained from Horizon Discovery (Waterbeach, UK). Thapsigargin (TG), cycloheximide, calpeptin [inhibitor of CAPN1 (calpain1)], HEPES [4-(2-hydroxyethyl) piperazine-1-ethanesulfonic acid], EGTA [ethylene glycol-bis (2-aminoethylether)-N,N,N0,N0-tetraacetic acid], EDTA (ethylenedinitrilotetraacetic acid), bovine serum albumin (BSA), sodium azide, dimethyl-BAPTA, sodium ascorbate, bafilomycin A1, MG132, rabbit polyclonal anti-Orai1 antibody (catalog number O8264, epitope: amino acids 288–301 of human Orai1), rabbit polyclonal anti-Orai1 (AB-1) antibody (catalog number AV50117, epitope: amino acids 2–61 of human Orai1), mouse monoclonal anti-phospho-Filamin-A (Ser2152) antibody (Clone PS2,catalog number MABN1834), and rabbit polyclonal anti-b-actin antibody (catalog number A2066, epitope: amino acids 365–375 of human b-actin; RRID: AB_2816311) were obtained from MilliporeSigma (Burlington, MA). Rabbit polyclonal anti-Orai2 antibody (catalog number TA306419, epitope: sequence localized in the C-terminal region; RRID: AB_2040046) was from Origene (Rockville, MD). Mouse monoclonal anti-Orai3 antibody (Clone EPR22575-17; catalog number ab254260; RRID: AB_2530307) was obtained from Abcam (Cambridge, UK). Mouse monoclonal anti-GOK/STIM1 antibody(Clone44/GOK;catalognumber610954,epitope:amino acids: 25–139 of human STIM1; RRID: AB_398267) was purchased from BD Biosciences (San Jose, CA). Horseradish peroxidase-conjugated goat anti-mouse immunoglobulin G (IgG) antibody (RRID: AB_10015289) and goat anti-rabbit IgG antibody (RRID: AB_2337913) were from Jackson Laboratories (West Grove, PA). N-glycosidase F (PNGase F) from Elizabethkingia miricola was from New England Biolabs (Ipswich, MA). Fluorescent goat anti-rabbit IgG StarBright Blue 700 (RRID: AB_2721073; Catalog number 12004161) and goat anti-mouse StarBright Bue 700 (RRID: AB_2884948; Catalog number 12004158) antibodies were from Bio-Rad Laboratories, Inc. (Hercules, CA). pEGFP-N1Orai1a-eGFP and pEGFP-N1-Orai1b-eGFP plasmids were kindly provided by Mohamed Trebak (Department of Pharmacology and Chemical Biology, University of Pittsburgh, Pittsburgh, PA). pDsRed-Monomer-C1-FLNA 16–24 encoding C-terminal FLNA repeats 16–24 was a gift from Xiaowei Zheng (Department of Molecular Medicine and Surgery of Karolinska Institutet, Solna, Sweden). pcDNA3-myc FLNA WT (Addgene No. 8982; http:// n2t.net/addgene:8982; RRID: Addgene 8982) and pcDNA3-myc FLNA S2152A (Addgene No. 8983; http://n2t.net/addgene: 89823; RRID: Addgene 8983) plasmids were a gift from John Blenis. STIM1-mCherry was a gift from Christoph Romanin (Institute of Biophysics, Johannes Kepler University Linz). All other reagents were of an analytical grade. FLNA CLEAVING MODULATES Orai1 EXPRESSION C658 AJP-Cell Physiol doi:10.1152/ajpcell.00745.2024 www.ajpcell.org Downloaded from journals.physiology.org/journal/ajpcell at Univ De Sevilla (193.147.173.206) on June 13, 2025. Cell Culture and Transfection Nontumoral NCM460 (RRID: CVCL_0460) and colorectal adenocarcinoma HT-29 cells (RRID: CVCL_0320) were kindly provided by Carlos Villalobos [Institute of Molecular Biology and Genetics (IBGM), Valladolid, Spain]. Colorectal adenocarcinoma Caco-2 cells (RRID: CVCL_0025) were provided by Mario Estevez (University of Extremadura, Cáceres, Spain). Colorectal carcinoma HCT116 cells (RRID: CVCL_B7PT) were purchased in ATCC (LGC Standards S.L.U.—Spain Office, Barcelona, Spain). Cells were regularly checked for contamination. CRISPR-generated Orai1-knockout HEK293 cells (RRID: CVCL_0045) (HEK-293 O1-KO), STIM1, and STIM2 doubleknockout HEK-293 cells (RRID: CVCL_0045) (HEK-293 DKO) and parental HEK-293 cells (RRID: CVCL_0045) were kindly supplied by Mohamed Trebak (Department of Pharmacology and Chemical Biology, University of Pittsburgh, Pittsburgh, PA). Cells were cultured at 37Cwitha5%CO 2 in high-glucose Dulbecco’smodified Eagle’s medium (DMEM) supplemented with 10% (vol/vol) fetal bovine serum and 100 U/mL penicillin and streptomycin, as previously described (30). Mycoplasmafree cultures were checked using confocal microscopy. For Western blotting assay, around 5 10 6 cells were plated on 100-mm petri dish, whereas for Ca 2þ imaging, eGFP fluorescent reporter assay and confocal microscopy cells (4 10 5 ) were seeded in a 35-mm six-well multidish. For transient transfections, cells were grown to 60%–80% confluency and transfected with indicated plasmids using DharmaFECT kb transfection reagent and were used 24 h after transfection. Cell viability after transfection throughout the study, estimated using the Live/Dead viability/cytotoxicity kit, was in the range between 92 and 95%. All procedures were approved by the Ethics Committee of University of Extremadura and Servicio Extreme~ no de Salud. Western Blotting Western blotting was performed as described previously (31). Briefly, cells were lysed with ice-cold RIPA buffer (150mMNaCl,25mMTris,5mMEDTA,1%TritonX-100, 1% sodium deoxycholate, 0.1% SDS; pH 7.6) supplemented with complete EDTA-free protease inhibitor cocktail. Cell lysates were homogenized using an ultrasonic homogenizer Sonoplus HD 2200.2 (Bandelin electronic GmbH & Co, Berlin, Germany) and subsequently centrifuged for 15 min at 16,000 gand 4C. Later, Laemmli sample buffer [20% (vol/vol) glycerol, 4% (wt/vol) SDS, 160 mM Tris-HCl pH 6.8, 10% (vol/vol) 2-mercaptoethanol] were added. Cell lysates were resolved by 12% SDS-PAGE, and separated proteins were electrophoretically transferred onto nitrocellulose membranes for subsequent probing. After blocking residual protein binding site with overnight incubation of the blots with 10% (wt/vol) BSA in Tris-buffered saline with 0.1% Tween-20 (TBST). Immunodetection of Orai1, Orai2, Orai3, STIM1, STIM2, FLNA, FLNA phosphorylated at Ser2152, and b-actin was achieved by incubation for 1 h with anti-Orai1, anti-Orai3 or anti-phospho-FLNA (Ser2152) antibody-diluted 1:1,000 in Tris-buffered saline with Tween (TBST)-, or by incubation for 1 h with anti-Orai2, anti-STIM1, anti-STIM2 or antiFLNA antibody, -diluted 1:500 in TBSTor by incubation for 1 hwithanti-b-actin antibody-diluted 1:2,000 in TBST-. To detect the primary antibody, blots were incubated for 1 h with fluorescentgoatanti-rabbitIgGStarBrightBlue700antibody (RRID: AB_2721073) or goat anti-mouse IgG StarBright Blue 700 diluted 1:3,000 in TBST. In addition, primary antibodies were also detected using horseradish peroxidase-conjugated goat anti-rabbit IgG antibody or horseradish peroxidase-conjugated goat anti-mouse IgG antibody diluted 1:10,000 in TBST and then, in this case, blots were exposed to enhanced chemiluminescence reagents for 5 min. In both cases, the antibody binding was detected with a ChemiDoc MP Imaging System (Bio-Rad Laboratories, Inc., Hercules, CA) and the density of bands was measured using Image Lab 6.1 Software (Bio-Rad Laboratories, Hercules, CA). Data were normalized to the amount of b-actin from the same gel. eGFP Fluorescent Reporter Assay HEK-293 Orai1-KO nontransfected or transfected with pEGFP-N1-Orai1a-eGFP or pEGFP-N1-Orai1b-eGFP plasmids were stimulated with 100 μg/mL cycloheximide (CHX) for 0, 6, and 9 h in the absence or presence of 1 μMbafilomycin A1 (BFA) or 10 lM MG132. Later, cells were lysed with ice-cold NP-40 buffer (137 mM NaCl, 20 mM Tris, 2 mM EDTA, 10% glycerol, 1% Nonidet P-40, 1 mM Na 3 VO 4 ; pH 8) supplemented with complete EDTA-free protease inhibitor cocktail. eGFP fluorescence was measured using the Varioskan LUX microplate multimode reader (Thermo Fisher Scientific, Waltham, MA). The excitation/emission wavelengths in the fluorescence assay were 485 nm/518 nm. eGFP fluorescence was normalized with the total amount of proteins measured with BCA protein assay kit. Determination of Cytosolic Free-Ca 21 Concentration Cells were loaded with the Ca 2þ fluorescent probe fura-2 by incubation with 2 μM fura-2/AM for 30 min at 37C. Cultured cells on coverslips mounted on a perfusion chamber were placed on the stage of an epifluorescence inverted microscope (Nikon Eclipse Ti2, Amsterdam, The Netherlands) with an image acquisition and analysis system for videomicroscopy (NIS-Elements Imaging Software v.5.02.00, Nikon, Amsterdam, The Netherlands). Cells superfusion was carried out at room temperature with HEPES-buffered saline (HBS) containing (in mM) 125 NaCl, 5 KCl, 1 MgCl 2 , 5 glucose, and 25 HEPES, pH 7.4, supplemented with 0.1% (wt/vol) BSA. Cells were evaluated at 40 magnification (Nikon CFI S FLUOR 40Oil, Amsterdam, The Netherlands) and were alternatively excited with light from a xenon lamp passed through a high-speed monochromator Optoscan ELE 450 (Cairn Research, Faversham, UK) at 340/380 nm. Fluorescence emission was detected at 510 nm using a cooled digital sCMOS camera PCO Panda 4.2 (Excelitas PCO GmbH, Germany) and recorded using NIS-Elements AR software (Nikon, Amsterdam, The Netherlands). Fluorescence ratio (F340/F380) was calculated pixel by pixel, and the data were presented as DF 340 /F 380 as described previously (18). TG-evoked Ca 2þ release and Ca 2þ entry were estimated as the area under the curve measured as the integral of the rise in fura-2 fluorescence ratio 4 min after the addition of TG (for Ca 2þ release) or Ca 2þ (for Ca 2þ entry), respectively, and taking a sample every second. Alternatively, Ca 2þ entry was determined as the initial peak in fura-2 340/380 fluorescence ratio above basal levels after re-addition of Ca 2þ to the extracellular medium. FLNA CLEAVING MODULATES Orai1 EXPRESSION AJP-Cell Physiol doi:10.1152/ajpcell.00745.2024 www.ajpcell.org C659 Downloaded from journals.physiology.org/journal/ajpcell at Univ De Sevilla (193.147.173.206) on June 13, 2025. Confocal Microscopy The subcellular location of DsRed-FLNA 16–24 ,STIM1mCherry, or Orai1-eGFP was determined by imaging cells 24 h post transfection, upon 30-min incubation at 37C with or without Hoechst 33258 (1 μg/mL) for nuclear staining. The imaging was performed using a confocal microscope (LSM900, Carl Zeiss, Germany) equipped with a 63 FLNA CLEAVING MODULATES Orai1 EXPRESSION C660 AJP-Cell Physiol doi:10.1152/ajpcell.00745.2024 www.ajpcell.org Downloaded from journals.physiology.org/journal/ajpcell at Univ De Sevilla (193.147.173.206) on June 13, 2025. oil immersion objective and an image acquisition and analysis system for video microscopy (ZenBlue 3.4 Software, Carl Zeiss, Germany). To assess the expression of DsRedFLNA 16–24 , seven to ten random regions of interest (ROIs), each measuring 2.5 lm 2 , were selected for fluorescence measurement, corrected by background subtraction and thresholding methods, and conducted using a custom script integrated into the FIJI ImageJ 1.54f software platform (RRID:SCR_003070). Biotinylation of Cell Surface Proteins Labeling and isolation of plasma membrane proteins were performed by surface biotinylation assay, as described previously (18). HEK-293 and Orai1-KO HEK-293 cells were washed three times with phosphate-buffered saline (PBS, NaCl 137 mM, KCl 2.7 mM, KH 2 PO 4 , 1.5 mM, Na 2 HPO 4 ·2H 2 O 8 mM, pH 8). Cells were then incubated at 4C for 1 h with biotynilation buffer (PBS supplemented with 1 mg/mL EZLink sulfo-NHS-LC-biotin). The reaction was terminated by addition of Tris base (final concentration 50 mM). Following biotinylation, cells were washed twice in PBS, disrupted using Nonidet P-40 buffer and sonicated. Cell lysates were centrifuged (16,000 gfor 5 min at 4C), and protein concentration was measured using BCA assay. Samples were incubated with 50 lL streptavidin beads at 4C for 2 h and resuspended in Laemmli buffer for subsequent analysis by Western blotting. Statistical Analysis All experiments were performed and analyzed using strategies to avoid bias. Data are presented as the means ± SE. Analysis of statistical significance was performed using GraphPad Prism v.8.4.3 (RRID:SCR_002798, GraphPad Software, San Diego, CA). Kruskal–Wallis test combined with Dunn’s post hoc test were used to compare the different experimental groups. For comparison between two groups, the Mann–Whitney Utest was used. All data with P<0.05 was deemed significant. RESULTS Expression of Orai and STIM Members and Filamin A in Colorectal Adenocarcinoma Cell Lines and Normal Colon Mucosa NCM460 Cells Several studies have revealed that the expression of Orai and STIM proteins is significantly altered in colorectal adenocarcinoma HT-29 cells as compared with normal colon mucosa NCM460 cells, leading to enhanced SOCE accompanied by a reduced ability to accumulate Ca 2þ into the intracellular stores (29,30,32). According to this, first of all, we have further analyzed SOCE as well as the expression of the key SOCE molecular players in HT-29 cells, the most widely studied colorectal adenocarcinoma cell line, in comparison with normal colon mucosa cells. As shown in Fig. 1A,treatment of NCM460 cells with TG in the absence of extracellular Ca 2þ results in a transient increase in cytosolic free-Ca 2þ concentration [Ca 2þ ] i , as a result of passive Ca 2þ efflux from the intracellular stores as a result of SERCA inhibition. Subsequent addition of 1.8 mM Ca 2þ to the extracellular medium led to a more sustained increase in [Ca 2þ ] i , which is indicative of SOCE. In colorectal adenocarcinoma HT-29 cells, TG-evoked Ca 2þ efflux from the intracellular Ca 2þ stores was significantly attenuated and SOCE was found to be significantly enhanced as compared with that observed NCM460 (Fig. 1, A–D,P<0.0001). The analysis of the expression of Orai and STIM proteins revealed that, in agreement with previous studies (29), HT-29 cells exhibit a significantly greater expression of Orai1, particularly Orai1a,Orai3,andSTIM1(Fig. 1, E– H;P<0.05). Furthermore, we found that STIM2 is overexpressed in HT-29 cells as compared with NCM460 cells (Fig. 1I; P<0.001), an observation that we have recently reported (30). Therefore, our current results confirm previous observations. FLNA is a cytoskeletal protein that plays a relevant role in the modulation of SOCE (20). Data from the Clinical Proteomic Tumor Analysis Consortium (33) (CPTAC) (https:// ualcan.path.uab.edu) have shown that FLNA expression at the protein level is reduced in a number of cancer types Figure 1. Store-operated Ca 2þ entry and expression of Orai, STIM, and FLNA in colorectal cancer and normal mucosa cells. A: fura-2-loaded normal mucosa NCM460 cells and colorectal cancer HT-29 cells were suspended in a Ca 2þ -free (100 lM EGTA) HBS and then stimulated with 2 lMTGfollowed by reintroduction of external Ca 2þ (final concentration 1.8 mM) to initiate Ca 2þ entry. B–D: scatter plots represent quantification of TG-evoked Ca 2þ release (B) and entry determined as the area under the curve (AUC; C)andCa 2þ entry determined as the initial peak in fura-2 340/380 fluorescence ratio above basal levels after addition of Ca 2þ to the extracellular medium (D), determined as described in MATERIALS AND METHODS. Data are presented as means ± SE and are statistically analyzed using Mann–Whitney Utest. P<0.0001 as compared with NCM460 cells. E–I: NCM460 and HT-29 cells were lysed. Cell lysates were treated in the absence (E,G–I) or the presence of PNGase F (F) and were then subjected to 10% SDS-PAGE and Western blotting with the anti-Orai1 (Eand F), anti-Orai3 (G), anti-STIM1 (H) or anti-STIM2 (I) antibody. J: NCM460, HT-29, Caco-2 and HCT116 cells were lysed and then subjected to 10% SDS-PAGE and Western blotting with the anti-FLNA, anti-STIM1, and anti-Orai1 (J) antibody, as described in MATERIALS AND METHODS.E–J: membranes were reprobed with the anti-b-actin antibody for protein loading control. Molecular masses indicated on the right were determined using molecular-mass markers run in the same gel. Blots are representative of three to four separate experiments. Bar graphs represent the quantification of the protein expression as fold increase over the level in NCM460 cells, presented as means ± SE. Data were statistically analyzed using Mann–Whitney Utest [except for J, where Kruskal–Wallis test with multiple comparisons (Dunn’stest)wasused].P<0.05, P<0.01, P<0.001, and P<0.0001 as compared with NCM460 cells. #P<0.05, ###P<0.001, and ####P<0.0001 as compared with HT-29 cells. K–O: HT-29 cells were treated with 1 lM calpeptin for 24 h or the vehicle as control. Cells were then lysed and subjected to 10% SDS-PAGE and Western blotting with the anti-FLNA (K), anti-Orai1 (L), anti-Orai3 (M), anti-STIM1 (N), and anti-STIM2 (O) antibody, as described in MATERIALS AND METHODS. Membranes were reprobed with the anti-b-actin antibody for protein loading control. Molecular masses indicated on the right were determined using molecular-mass markers run in the same gel. Blots are representative of four separate experiments. Bar graphs represent the quantification of protein expression as fold increase over the level in the absence of calpeptin and presented as means ± SE. Data were statistically analyzed using Mann–Whitney Utest. P< 0.001 and P<0.0001. P–S: HT-29 cells were treated with 1 lM calpeptin for 24 h or the vehicle as control. Cells were loaded with fura-2, suspended in a Ca 2þ -free (100 lM EGTA) HBS, and then stimulated with 2 lMTGfollowedbyreintroductionofexternalCa 2þ (final concentration 1.8 mM) to initiate Ca 2þ entry. Q–S: scatter plots represent quantification of TG-evoked Ca 2þ release (Q)Ca 2þ entry determined as the area under the curve (AUC; R) and Ca 2þ entry determined as the initial peak in fura-2 340/380 fluorescence ratio above basal levels after addition of Ca 2þ to the extracellular medium (S), as described in MATERIALS AND METHODS. Data are presented as means ± SE and are statistically analyzed using Mann–Whitney Utest. P<0.01 and P<0.0001. FLNA, Filamin A; HBS, HEPES-buffered saline; TG, Thapsigargin. FLNA CLEAVING MODULATES Orai1 EXPRESSION AJP-Cell Physiol doi:10.1152/ajpcell.00745.2024 www.ajpcell.org C661 Downloaded from journals.physiology.org/journal/ajpcell at Univ De Sevilla (193.147.173.206) on June 13, 2025. including colon, ovarian and lung cancer, which exhibit greater SOCE and Orai1 and STIM1 expression (29,34–36). Concerning colon cancer, a significantly reduced FLNA protein level has been detected in patients samples compared with normal colon tissue (Supplemental Fig. S1). Hence, we have explored the possible differences in the expression of this protein in the adenocarcinoma cell lines HT-29, Caco-2, and HCT116, as well as in normal colon mucosa NCM460 cells. As shown in Fig. 1J, Western blotting of whole cell lysates with a specific anti-FLNA antibody revealed a band that corresponds to the predicted size of full-length FLNA (280 kDa). In addition to the 280-kDa band, we detected an 190 kDa fragment corresponding to the larger product of FLNA cleavage by calpain (37). Interestingly, we found that the expression of full-length FLNA was significantly reduced in HT-29 cells as compared with the colorectal cell lines Caco-2 and HCT116 or NCM460 cells (Fig. 1J;P<0.0001). Reciprocally, the expression of the 190 kDa FLNA fragment was enhanced in HT-29 cells and, to a lesser extent, in NCM460 cells (Fig. 1J;P<0.0001). These observations strongly suggest that FLNA is mostly cleaved in colorectal adenocarcinoma HT-29 cells as compared with Caco-2 or HCT116 cells or the normal colon mucosa cells. As shown in Fig. 1J, HT-29 cells show enhanced Orai1 and STIM1 expression as compared with the colorectal cell lines Caco-2 and HCT116 or NCM460 cells; however, the mechanism underlying the increase in Orai1 and STIM1 protein level remains unknown. Our results indicate that FLNA is mostly cleaved in HT-29 cells leading to a fragment that resembles the previously mentioned product of calpain proteolysis, which is involved in transcriptional regulation (38). To figure out whether FLNA is cleaved by calpain in HT-29 cells and the possibleroleofFLNAcleavageinOrai1andSTIM1proteincontent, cells were treated with the cell-permeant calpain inhibitor calpeptin. As shown in Fig. 1K,treatmentofHT29 cells with 1 lMcalpeptinfor24hsignificantly reduced FLNA cleavage as detected by the enhancement of the expression of full-length FLNA and attenuation of the FLNA fragment, which strongly suggests that FLNA is cleaved by calpain. Furthermore, treatment with calpeptin significantly reduced Orai1 and STIM1 expression without having any effect on the protein content of Orai3 and STIM2 (Fig. 1, L–O;P<0.001), consequently, SOCE was significantly reduced in cells treated with calpeptin (Fig. 1, P–S;P<0.01). We further found that calpeptin did not induce a detectable reduction in the ability of STIM1 to form clusters (Supplemental Fig. S2). These findings indicate that calpain is essential for FLNA cleavage and that this event modulates Orai1 and STIM1 protein expression. FLNA phosphorylation at Ser2152 has been reported to be essential for its biological functions (23,39), including the modulation of SOCE (20). Western blotting of whole cell lysates with a specific anti-phospho-FLNA (FLNA P-Ser2152) antibody detected a single band at 280 kDa corresponding to the full-length FLNA that was significantly reduced in HT-29 cell lysates, probably as a result of the reduced expression of full-length FLNA in these cells (Supplemental Fig. S3, Aand B;P<0.05). Furthermore, normalization of the amount of pFLNA to the total FLNA revealed that relative FLNA phosphorylation at Ser2152 is significantly greater in HT-29 cells than in NCM460 cells. We were unable to detect phosphorylation of the FLNA fragments. We have further investigated the functional role of FLNA serine phosphorylation in the regulation of SOCE in HT-29 by transiently expressing wild-type FLNA or the nonphosphorylatable FLNA S2152A mutant. As shown in Supplemental Fig. S3, expression of wild-type FLNA in HT-29 cells significantly reduced SOCE and enhanced TG-evoked Ca 2þ release from the intracellular stores (Supplemental Fig. S3, C–F;P<0.0001). By contrast, expression of the nonphosphorylatable FLNA mutant was without effect on TG-induced Ca 2þ release or SOCE (Supplemental Fig. S3, C–F), which supports a role for FLNA Ser2152 phosphorylation in the modulation of these events. In addition, our results indicate that expression of wild-type FLNA significantly attenuated Orai1 and STIM1 expression (Supplemental Fig. S3, G–J; P<0.05). Meanwhile, the expression of the FLNA S2152A mutant reduced STIM1 expression without having any significant effect on the protein content of Orai1, Orai3, or STIM2 as compared with nontransfected HT-29 cells (Supplemental Fig. S3, G–J;P<0.05). However, it is worth mentioning that Orai1 expression in cells expressing wild-type FLNA or the FLNA S2152A mutant was not significantly different (Supplemental Fig. S3G;P¼ 0.537). These observations indicate that FLNA phosphorylation at Ser2152 plays a significant functional role in the modulation of SOCE and the accumulation of Ca 2þ into the intracellular stores in the colorectal adenocarcinoma cell line HT-29, as previously reported (20). The FLNA 16–24 Fragment Enhances Orai1 and STIM1 Protein Content Cleavage of FLNA at the first cleaving site leads to the formation of 190and 90-kDa fragments, corresponding to the N-terminal 15 repeats of FLNA and the C-terminal region encompassing repeats 16–24 (FLNA 16–24 )(38,40). The FLNA 16–24 fragment has been reported to modulate gene transcription by repressing androgen receptor activity (38,41). Hence, we have further investigated the possible role of the FLNA 16–24 fragment on the expression of Orai and STIM proteins by extending our studies in HEK-293 cells, a commonly used cell model for cell biology, where FLNA is not naturally cleaved (Fig. 2A). To explore whether the FLNA 16–24 fragment is able to regulate Orai1 and STIM1 protein content, HEK-293 cells were transfected with DsRedFLNA 16–24 expression plasmid and the expression of Orai and STIM proteins, as well as SOCE, was analyzed. As shown in Fig. 2A, Western blotting of HEK-293 lysates with anti-FLNA antibody revealed a single band of 280-kDa corresponding to the native, full-length, FLNA in HEK-293 cells transfected with empty vector and two bands of 280and 120-kDa in HEK-293 cells transfected with DsRed-FLNA 16–24 expression plasmid, the later corresponding with the predicted size of DsRed-FLNA 16–24 fragment. Expression of DsRed-FLNA 16–24 was further confirmed by confocal microscopy, which revealed a cytoplasmic location of the FLNA fragment (Fig. 2A). Expression of the FLNA 16–24 fragment in HEK-293 cells significantly enhanced the protein FLNA CLEAVING MODULATES Orai1 EXPRESSION C662 AJP-Cell Physiol doi:10.1152/ajpcell.00745.2024 www.ajpcell.org Downloaded from journals.physiology.org/journal/ajpcell at Univ De Sevilla (193.147.173.206) on June 13, 2025. Figure 2. Expression of the FLNA 16–24 construct enhances the protein level of Orai1 and STIM1 in HEK-293 cells. A: HEK-293 cells were transfected with DsRed-FLNA 16–24 fragment or empty vector and 48 h later cells were either lysed or visualized by confocal microscopy. Top: cell lysates were subjected to 10% SDS-PAGE and Western blotting with the anti-FLNA antibody, as described in MATERIALS AND METHODS. Molecular masses indicated on the right were determined using molecular-mass markers run in the same gel. Blots are representative of four separate experiments. Bottom:DsRedfluorescence was detected using an LSM900 confocal microscope. The images show representative confocal images of DsRed-FLNA 16–24 and nuclear staining with Hoechst 33342. The scale bar represents 10 μm. B–E: HEK-293 cells were transfected with DsRed-FLNA 16–24 fragment or empty vector, as described. Forty-eight hours later, cells were lysed and subjected to 10% SDS-PAGE and Western blotting with the anti-Orai1 (B), anti-Orai3 (C), anti-STIM1 (D), and anti-STIM2 (E) antibody. Membranes were reprobed with the anti-b-actin antibody for protein loading control. Molecular masses indicated on the right were determined using molecular-mass markers run in the same gel. Blots are representative of four separate experiments. Bar graphs represent the quantification of protein expression as fold increase over the level in mock-transfected cells and presented as means ± SE. Data were statistically analyzed using Mann–Whitney Utest. P<0.05 and P<0.01. F–I: HEK-293 cells were transfected with DsRed-FLNA 16–24 fragment or empty vector. Fortyeight hours later, cells were loaded with fura-2. Cells were then suspended in a Ca 2þ -free (100 lM EGTA) HBS and stimulated with 2 lM TG followed by reintroduction of external Ca 2þ (final concentration 1.8 mM) to initiate Ca 2þ entry. G–I: scatter plots represent quantification of TG-evoked Ca 2þ release (G)Ca 2þ entry determined as the area under the curve (AUC; H)andCa 2þ entry determined as the initial peak in fura-2 340/380 fluorescence ratio above basal levels after addition of Ca 2þ to the extracellular medium (I), as described in MATERIALS AND METHODS.Dataarepresentedasmeans±SEand are statistically analyzed using Mann–Whitney Utest. P<0.05 and P<0.01. FLNA, Filamin A; HBS, HEPES-buffered saline; TG, Thapsigargin. FLNA CLEAVING MODULATES Orai1 EXPRESSION AJP-Cell Physiol doi:10.1152/ajpcell.00745.2024 www.ajpcell.org C663 Downloaded from journals.physiology.org/journal/ajpcell at Univ De Sevilla (193.147.173.206) on June 13, 2025. content of Orai1 and STIM1, without having any effect on the expression of Orai3 and STIM2 (Fig. 2, B–E;P<0.05). As a result, the expression of the FLNA 16–24 fragment significantly enhanced SOCE and reduced the ability of HEK293 cells to accumulate Ca 2þ into the intracellular stores (Fig. 2, F–I;P<0.05). These findings are consistent with the effect of calpeptin on Orai1 and STIM1 expression in HT-29 cells and resembles the phenotype (enhanced Orai1 and STIM1 protein content and SOCE) of the HT-29 cell line. The FLNA 16–24 Fragment Enhances Orai1 and STIM1 Protein Content by Attenuating Degradation The protein content is the result of a balance between protein synthesis and degradation. We have further explored the mechanism underlying the regulation of Orai1 and STIM1 protein content by the FLNA 16–24 fragment by using the protein synthesis inhibitor cycloheximide (CHX). HEK-293 cells were transfected with DsRedFLNA 16–24 fragment expression plasmid or empty vector and were further treated with CHX or the vehicle for 9 h. As depicted in Fig. 3, Orai1 and STIM1 protein expression increases by 40 and 30%, respectively, after transfection of the FLNA fragment (P<0.01). CHX by itself attenuated the expression of Orai1 and STIM1 proteins by 38 and 35%, respectively (Fig. 3). In the presence of CHX, the expression of Orai1 and STIM1 increases after the expression of the FLNA fragment by 40 and 37%, respectively, (Fig. 3;P<0.01). Therefore, our results indicate that the FLNA 16–24 fragment was able to enhance Orai1 and STIM1 expression in the absence of protein synthesis, which strongly suggest that the effect is likely to be mediated by an increase in protein half-life induced by inhibition of protein degradation. The Orai1 Variants Orai1aand Orai1bExhibit Different Degradation Rates As the FLNA 16–24 fragment modulates the degradation of Orai1 and two Orai1 variants have been identified, we have further analyzed the effect of the FLNA 16–24 fragment on the degradation of Orai1aand Orai1b.First,wehaveassessedthe degradation of both Orai1 variants in Orai1-KO HEK-293 transfected with Orai1aor Orai1bby estimating their protein content in the presence of the protein synthesis inhibitor CHX. Orai1-KO HEK-293 was transfected with cytomegalovirus (CMV)-driven Orai1a-eGFP or Orai1b-eGFP, and the protein content of the Orai1 variants was determined by Western blotting using an anti-Orai1 antibody. As depicted in Fig. 4A,Orai1aprotein content decreases by 20 and 30% after cell exposure to CHX to 6 and 9 h, respectively (P< 0.01). By contrast, the protein content of Orai1bwas unaffected by inhibition of protein synthesis at least during 9 h. Similar results were obtained when the protein content of Orai1a-eGFP or Orai1b-eGFP was determined by eGFP fluorescence quantification (Fig. 4B). These findings provide for the first time evidence supporting that the stability of Orai1b is greater than that of Orai1a. In mammalian cells, protein degradation mainly occurs through the ubiquitin-proteasome pathway and the lysosomal-dependent proteolysis. To elucidate the pathway involved in Orai1 degradation, cells were treated with bafilomycin A1 (BFA), a selective inhibitor of vacuolar H þ -ATPase that prevents lysosomal degradation (42)orMG132,aninhibitor of proteasome (43), and Orai1 variant protein content was Figure 3. Expression of the FLNA 16–24 construct attenuates Orai1 and STIM1 degradation. HEK-293 cells were transfected with DsRed-FLNA 16–24 fragment or empty vector, as indicated. Forty-eight hours later, cells were either treated with cycloheximide (CHX; 100 lg/mL) or the vehicle for 9 h and lysed. Cell lysates were subjected to 10% SDS-PAGE and Western blotting with the anti-FLNA antibody and either anti-Orai1 (A)oranti-STIM1(B)antibody. Membranes were reprobed with the anti-b-actin antibody for protein loading control. Molecular masses indicated on the right were determined using molecular-mass markers run in the same gel. Blots are representative of four separate experiments. Bar graphs represent the quantification of protein expression as fold increase over the level in mock-transfected cells not treated with CHX. Data are presented as means ± SE and statistically analyzed using Kruskal–Wallis test with multiple comparisons (Dunn’stest).P<0.01 and P<0.001 as compared to mock-transfected cells not treated with CHX. ##P<0.01 as compared with DsRed-FLNA 16–24 fragment-transfected cells not treated with CHX. CHX, cycloheximide; FLNA, Filamin A. FLNA CLEAVING MODULATES Orai1 EXPRESSION C664 AJP-Cell Physiol doi:10.1152/ajpcell.00745.2024 www.ajpcell.org Downloaded from journals.physiology.org/journal/ajpcell at Univ De Sevilla (193.147.173.206) on June 13, 2025. determined by the determination of eGFP fluorescence. As reported above, treatment for 9 h with CHX by itself significantly reduced Orai1aexpression (Fig. 4C;P<0.001). In the presence of CHX and BFA, the Orai1aprotein content was similar to that in cells not treated with CHX (Fig. 4C), indicating that BFA impairs Orai1aprotein degradation. By contrast, MG132 was unable to reverse the drop in Orai1aprotein content observed upon blockade of protein synthesis with CHX (Fig. 4C;P<0.01). The expression of Orai1bwas unaffected by treatment with CHX in the absence or presence of BFA and MG132 (Fig. 4C). These findings indicate that Orai1adegradation occurs through lysosomal proteolysis, which is consistent with previous studies by Yeh et al. (44). The FLNA 16–24 Fragment Attenuates Orai1aProtein Degradation Retaining the Channel in the Plasma Membrane We have specifically explored whether the FLNA 16–24 fragment modifies the protein degradation of Orai1aor Orai1b. Orai1-KO HEK-293 cells were transfected with Orai1a-eGFP or Orai1b-eGFP in the absence or presence of DsRed-FLNA 16–24 fragment expression plasmid or empty vector and were further treated with CHX or the vehicle for 9 h. As depicted in Supplemental Fig. S4 for native Orai1, detection of Orai1 variants fused to eGFP leads to several diffuse bands, as previously described (16), likely due to N-linked glycosylation of Orai1 (Fig. 5). As shown in Fig. 5,Orai1aproteinexpressionsignificantly increases after transfection of the FLNA fragment (P<0.01). CHX by itself significantly attenuated the expression of Orai1a(P<0.001). In the presence of CHX, the expression of Orai1asignificantly increases after expression of the FLNA fragment (Fig. 5;P<0.01) reaching a value that was comparable to that observed in untreated Orai1-KO HEK-293 cells expressing Orai1a, which indicates that the FLNA 16–24 fragment impairs Orai1aprotein degradation. Concerning Orai1b, neither treatment with CHX nor transfection of the DsRed-FLNA 16–24 fragment altered the protein expression at least at the time investigated, which is consistent with the low rate of Orai1bprotein degradation. We further explored whether the C-terminal fragment of FLNA impairs Orai1adegradation by impairing protein endocytosis by analyzing the expression of Orai1ain the plasma membrane by surface protein biotinylation. HEK293 cells were transfected with DsRed-FLNA 16–24 fragment expression plasmid or empty vector and, after biotinylation, Figure 4. Analysis of Orai1aand Orai1bdegradation. A:Orai1-KOHEK-293 cells were transfected with CMV-driven Orai1a-eGFP (lanes 1–3)orOrai1beGFP (lanes 4–6). Forty-eight hours later, cells were then treated in the absence or presence of cycloheximide (CHX; 100 lg/mL) for 6 or 9 h, as indicated, and lysed. Cell lysates were then subjected to 10% SDS-PAGE and Western blotting with anti-Orai1 antibody, as described in MATERIALS AND METHODS. Membranes were reprobed with the anti-b-actin antibody for protein loading control. Molecular masses indicated on the right were determined using molecular-mass markers run in the same gel. Blots are representative of three separate experiments. Scatter plots represent Orai1a-eGFP and Orai1b-eGFP expression at the different experimental conditions. Data are presented as means ± SE and statistically analyzed using Kruskal–Wallis test with multiple comparisons (Dunn’stest).P< 0.01 as compared with cells not treated with CHX. B:Orai1-KOHEK-293 cells were transfected with CMV-driven Orai1a-eGFP or Orai1b-eGFP. Forty-eight hours later, cells were then treated in the absence or presence of cycloheximide (CHX; 100 lg/mL) for 6 or 9 h, as indicated and GFP fluorescence was determined as described in MATERIALS AND METHODS.Scatter plots represent Orai1a-eGFP and Orai1b-eGFP fluorescence at the different experimental conditions. Data are presented as means ± SE and statistically analyzed using Kruskal–Wallis test with multiple comparisons (Dunn’stest).P<0.01 as compared with cells not treated with CHX. C: Orai1-KO HEK-293 cells were transfected with CMV-driven Orai1a-eGFP or Orai1b-eGFP, as indicated. Forty-eight hours later, cells were then treated in the absence or presence of cycloheximide (CHX; 100 lg/mL) for 9 h, alone or in combination with 1 lMbafilomycin A1 (BFA) or 10 lMMG132,as indicated, and GFP fluorescence was determined as described in MATERIALS AND METHODS. Scatter plots represent eGFP fluorescence at the different experimental conditions. Data are presented as means ± SE and statistically analyzed using Kruskal–Wallis test with multiple comparisons (Dunn’stest).P<0.01 and P<0.001 as compared with cells not treated with CHX. ##P<0.01 and ###P<0.001 as compared with cells treated with CHX and BFA. CHX, cycloheximide; KO, knockout. FLNA CLEAVING MODULATES Orai1 EXPRESSION AJP-Cell Physiol doi:10.1152/ajpcell.00745.2024 www.ajpcell.org C665 Downloaded from journals.physiology.org/journal/ajpcell at Univ De Sevilla (193.147.173.206) on June 13, 2025.