Store-operated Ca2+ entry in primary murine lung fibroblasts is independent of classical transient receptor potential (TRPC) channels and contributes to cell migration.

Stromal interaction molecules (STIM1, 2) are acting as sensors for Ca2+ in intracellular stores and activate Orai channels at the plasma membrane for store-operated Ca2+ entry (SOCE), while classical transient receptor potential (TRPC) channel mediate receptor-operated Ca2+ entry (ROCE). Several reports, however, indicate a role for TRPC in SOCE in certain cell types. Here, we analyzed Ca2+ influx and cell function in TRPC1/6-deficient (TRPC1/6-/-) and STIM1/2- deficient (STIM1/2ΔpmLF) primary murine lung fibroblasts (pmLF). As expected, SOCE was decreased in STIM1/2- deficient pmLF and ROCE was decreased in TRPC1/6-/- pmLF compared to control cells. By contrast, SOCE was not significantly different in TRPC1/6-/- pmLF and ROCE was similar in STIM1/2-deficient pmLF compared to Wt cells. Most interestingly, cell proliferation, migration and nuclear localization of nuclear factor of activated T-cells (NFATc1 and c3) were decreased after ablation of STIM1/2 proteins in pmLF. In conclusion, TRPC1/6 channels are not involved in SOCE and STIM1/2 deficiency resulted in decreased cell proliferation and migration in pmLF.

In heteromeric TRPC channels TRPC1 appears to work as an ion channel regulator rather than an ion channel per se, because it modifies currents of homotetrameric TRPC5 24 and reduces Ca 2+ permeation of TRPC4/5/6/7 channels 25 . Therefore, the exact function of TRPC channels for SOCE or ROCE needs to be analyzed in each cell type independently.
In here, we set out to study the role of SOCE in primary murine lung fibroblasts (pmLF) using TRPC1/6and STIM1/2-deficient fibroblasts in comparison to Wt control cells. SOCE was independent from TRPC1 and TRPC6 expression in pmLF but clearly dependent on STIM1/2 proteins. STIM1/2-deficiency reduced cell proliferation and migration as well as decreased nuclear levels of nuclear factor of activated T cells (NFATc1 and NFATc3) compared to control cells. Our data suggest an essential role of TRPC-independent SOCE in pmLF survival and cell migration.
Isolation and culture of primary murine lung fibroblasts. Primary murine lung fibroblasts (pmLF) were isolated as previously described for human lung fibroblasts 28 . Briefly, lungs of C57BL/6 mice were flushed through the right heart with sterile, cold PBS and excised. The lungs were dissected into pieces of 1-2 cm 2 in size and digested by 1 mg/ml of Collagenase I (Biochrom, Cambridge, UK) at 37 °C for 2 h. Digested lung pieces were filtered through a nylon filter (pore size 70 μm; BD Falcon, Franklin Lakes, NJ, USA) and centrifuged for 5 min. Subsequently, the pellet was re-suspended in DMEM/F12 fibroblast culture medium (Lonza, Basel, Switzerland) supplemented with 20% fetal bovine serum (Invitrogen, Carlsbad, USA) as well as penicillin/streptomycin (Lonza, Basel, Switzerland) and normocin (InvivoGen, San Diego, USA) before finally plated on 10 cm cell culture dishes. Medium was changed after 2 days and cells were split after reaching a confluence of 80-90%. Only pmLF from passage 3-4 were used for the studies.

Lentiviral infection of pmLf. PmLF from Stim1/2 flox/flox mice were infected by lentiviruses expressing
Cre recombinase to obtain STIM1/2-deficient fibroblasts. Lentiviruses were produced as previously described 29 based on the protocol for the amplification of second generation lentiviruses from the Tronolab (tronolab.epfl.ch). Lenti-X 293T cells (Clontec/Takara, Mountain View, USA) grown in DMEM medium (Lonza, Basel, Switzerland) supplemented with 10% fetal bovine serum (Invitrogen, Carlsbad, USA) as well as penicillin/streptomycin (Lonza, Basel, Switzerland) were transfected with pWPXL (carrying the gene of interest), pMD2G (encoding VSV G envelope protein) and pSPAX (encoding HIV-1 Gag, Pol, Tat and Revprotein) by calcium phosphate transfection. Supernatant containing virus was collected for two days. Virus solution was concentrated by using Peg-it solution (SBI, Mountain View, USA) and the pellet was re-suspended in cold PBS, aliquoted and stored at −80 °C. Successful virus production was verified by LentiX Go-stix (Clontec/Takara, Mountain View, USA). PmLF of the second passage were seeded at 1.5 × 10 5 cells per well of a 6-well plate and infected by lentiviruses expressing Cre recombinase on the next day. Medium was changed the next morning and infected pmLF were used for experiments after 4-5 days. Excision of exons from Stim1 and Stim2 genes was monitored by genomic PCR.
Genomic pcR. Genotyping of Trpc1/6 −/− and Stim1/2 flox/flox mice as well as STIM1/2 ΔpmLF fibroblasts was done as described 21,26,27 . ca 2+ imaging of intracellular ca 2+ . STIM1/2 ΔpmLF , TRPC1/6 −/− , and TRPC1/6 −/− STIM1/2 ΔpmLF as well as control cells (Wt and Wt infected Cre recombinase expressing lentiviruses) were grown on 25 mm coverslips and loaded with Fura-2-AM (2 µM, Sigma, Taufkirchen, Germany) in 0.1% BSA in HEPES/HBSS buffer at 37 °C for 30 min. Coverslips were washed with HEPES/HBSS buffer and placed on a microscope in a low-volume recording chamber. To measure receptor-operated Ca 2+ entry (ROCE) endothelin-1 (4 µM, Merck, Darmstadt, Germany) was applied in HBSS buffer with (2 mM) Ca 2+ or in nominal Ca 2+ free (0.5 mM EGTA) buffer after adding Ca 2+ (2 mM). Store-operated Ca 2+ entry (SOCE) was analyzed after depletion of internal Ca 2+ stores by 1 µM thapsigargin (Sigma, Taufkirchen, Germany) in Ca 2+ free HBSS solution containing 0.5 mM EGTA by adding extracellular Ca 2+ (2 mM) 30 . An increase in intracellular Ca 2+ ([Ca 2+ ] i ) was recorded using a Polychrome V monochromator (Till Photonics, Martinsried, Germany) and a 14-bit EMCCD camera (iXON3 885, Andor, Belfast, UK) coupled to an inverted microscope (IX71with an UPlanSApo 20×/0.85 oil immersion objective, Olympus, Hamburg, Germany) at 340 and 380 nm as described 30 . Quantitative reverse transcription (qRT)-PCR analysis. Total RNA from primary lung fibroblasts was isolated using the Invitrap Spin Universal RNA Mini Kit (Stratec, Berlin, Germany) according to the manufacturer's protocol. First-strand cDNA was synthesized from the isolated total RNA using RevertAid RT containing reverse transcription polymerase (ThermoScientific, St. Leon-Rot, Germany) and random primer. MRNA expression of targeted genes in pmLF was analyzed by real time PCR as previously described 30 . Briefly, 10 pmol of each primer pair and 2 μl from the first strand synthesis were added to the reaction mixture consisting of 2x ABsolute QPCR SYBR Green Mix (ThermoScientific, St. Leon-Rot, Germany) and water. PCR was carried out in a light-cycler apparatus (Roche, Mannheim, Germany) using the following conditions: 15 min initial activation and 45 cycles of 12 s at 94 °C, 30 s at 50 °C, 30 s at 72 °C. Primer pairs (see Table 1) were used for the amplification of specific DNA-fragments from the first strand synthesis. Fluorescence intensities were recorded after an extension step at 72 °C after each cycle. Samples containing primer dimers were excluded by melting curve analysis and identification of the products were done by agarose gel electrophoresis. Crossing points were determined by the software program provided by the manufacturer. Relative gene expression was quantified using the formula: (2e(Crossing point reference gene − Crossing point X)) × 100 = % of the reference gene expression of the housekeeping gene (β-actin).
Western blot analysis. Protein expression levels for STIM1 and STIM2 were determined by Western Blot analysis as previously described 30 . PmLF from cell culture dishes of 20 cm diameter were washed two times with cold PBS before 250 μl of lysis buffer (20 mM Tris-HCL, pH 7.5, 150 mM NaCl, 1% Nonidet P40, 0.5% sodium deoxycholate, 1% SDS, 5 mM EDTA) containing phosphatase und protease inhibitors (Roche, Mannheim, Germany) was applied for 60 min on ice. After centrifugation of the protein lysates at 5500 × g for 30 min at 4 °C protein concentration was quantified using a BCA-Assay (Pierce, Thermo Fisher, Schwerte, Germany) according to the manufacturer's instructions. 6× Laemmli buffer (375 mM 4 × Tris/SDS buffer, pH 6.8, 48% glycerin, 6% SDS, 0,03% bromophenol blue and 9% β-mercaptoethanol) was added, the mixture incubated at 90 °C for 10 min and sonicated for 15 s. 10 μg protein of each sample was loaded on a 10% SDS gel. Protein separation was performed at room temperature using a current of 20 mA for 3-4 h. To transfer the proteins to a PVDF membrane a current of 20 mA was applied for 20 h at 4 °C. After transfer, the membrane was rinsed with 10 ml TBST for 5 min at room temperature. Transfer was checked using Ponceau solution (A2935 0500, AppliChem, Darmstadt, Germany). Blocking was performed for 1 h at room temperature using 10 ml blocking buffer (5% low fat milk in TBST). Each primary antibody was diluted in TBST containing 5% of blocking solution and applied over night at 4 °C. After washing with TBST three times for 10 min each, HRP-conjugated secondary antibody was applied for 2 h at room temperature. The membrane was washed with TBST three times, for 10 min each and incubated in SuperSignal West Femto chemiluminescent substrate (Thermo Scientific, Waltham, MA, USA). Chemiluminescence was detected by exposure of the filter in an Odyssey-Fc-unit (Licor, Lincoln, NE, USA). Used antibodies and dilutions: HRP-conjugated anti-β-actin antibody (Sigma A3854HRP, 1:10,000), anti-STIM1 (CellSignaling, #4916S, 1:1000), anti-STIM2 (CellSignaling, #4917S, 1:1000) and secondary anti-rabbit IgG peroxidase (POX)-antibody (Sigma A6154, 1:10000).
Viability assay. Viability assays were performed by using the WST-1 Reagent (Roche, Mannheim, Germany) according to the manufacturer's instructions. Tetrazolium salts like WST-1 are cleaved to colored formazan in viable cells, which can be measured by spectrophotometry. Cells were plated at a density of 2 × 10 4 cells per ml per well of a 24-well plate and incubated at 37 °C and 5% CO 2 overnight. WST-1 reagent was diluted 1:10 in pmLF medium before it was added to each well. After 3 hours of incubation absorbance of formazan was measured at a wavelength of 450 nm by spectrophotometry (Tecan, Switzerland). Cell free wells containing pmLF Medium plus WST-1 reagent served as background values.
Proliferation assay. DNA synthesis of lung fibroblasts was assessed using the Click-iT 5-ethynyl-2′-deoxyuridine (EdU) Imaging Kit (ThermoScientific, St. Leon-Rot, Germany). In brief, 1.5 × 10 5 pmLF per well of a 6 well plate were plated on coverslips overnight and were treated with 10 µM EdU for 3 hours on the next day. After washing cells with PBS and fixation in 3.7% formaldehyde for 10 min pmLF were treated according to the manufacturer's protocol as previously described 29 . EdU is a thymidine analogue which gets incorporated into DNA during active DNA synthesis, if added to the culture medium 31 . After incorporation the ethynyl group of EdU covalently couples to a small fluorescent azide in a copper-dependent click reaction, which can be detected under a fluorescence microscope. To detect all cell nuclei, an additional staining with Hoechst 33342 (Life Technologies, Darmstadt, Germany) was performed. Stained cells were visualized by confocal imaging (LSM 880, Carl Zeiss) and stained nuclei were analyzed by the ImageJ software.
Migration assay. Approximately 1.5 × 10 4 cells per insert were seeded on a 3 well silicone insert with a 500 µm cell-free gap (ibidi GmbH, Martinsried, Germany) and grown at 37 °C and 5% CO 2 overnight. Insert detachment created a defined cell-free gap. Images were taken 0, 4, 8, 12 and 24 h after releasing inserts. Migration was analyzed by measuring the remaining gap width by the ImageJ software in 3 pictures per time point and replicate.  TCC AGT TGT   STIM2  mouse  GAG GGC GCA GAG TGT GAG  TTT AGA GCC ATG CGG ACC T   Orai1  mouse  TAC TTA AGC CGC GCC AAG  ACT TCC ACC ATC GCT ACC A   Orai2  mouse  GGA CCT CAG CCC TCC TGT  GGG TAC TGG TAC TTG GTC TCC A   Orai3  mouse  CAC ATC TGC TCT GCT GTC G  GGT GGG TAT TCA TGA TCG TTC T   TRPC1  mouse  TGA ACT TAG TGC TGA CTT AAA GGA AC  CGG GCT AGC TCT TCA TAA TCA   TRPC3  mouse  TGG ATT GCA CCT TGT AGC   www.nature.com/scientificreports www.nature.com/scientificreports/ Isolation of nuclear fractions. Isolation of nuclear protein extracts from pmLF was performed with a Nuclear Extract Kit according to the manufacturer's instructions (Active Motif, 40010, La Hulpe, Belgium) as described 30 . In brief, cells were first washed with PBS containing phosphatase inhibitors. Cytoplasmic protein fractions were collected by adding hypotonic lysis buffer and detergent, causing leakage of cytoplasmic proteins into the supernatant. After centrifugation (14.000 × g for 30 s) nuclear protein fractions were obtained by re-suspending pellets in detergent-free lysis buffer containing protease inhibitors. NFAT proteins were analyzed by Western Blotting as described above with the following modifications: 20 μl of each protein sample was loaded on a 7.5% SDS gel. Transfer of proteins to PVDF membrane was performed by an applied current of 360 mA at 4 °C for 1 h and 15 min. Unspecific binding sites were blocked in 5%BSA in TBST for one hour prior to incubation with the first antibody incubation overnight. All other steps were performed as described before. Antibodies used: Anti NFATc1 (mouse, SantaCruz Biotechnology, sc-7294, 1:500), anti-NFATc3 (mouse, SantaCruz Biotechnology, sc-7294, 1:500), anti-mouse IgG HRP-linked antibody (CellSignaling, #7076, Danvers, USA) as secondary antibody. Anti Lamin B1 (rabbit, ThermoScientific, PA5-19468, 1:5000) and secondary anti-rabbit IgG peroxidase-linked antibodies (goat, Sigma A6154, 1:10000) served as loading control. Protein bands were normalized to loading controls and quantified by an Odyssey-Fc unit (Licor, Lincoln, USA).
Statistics. All statistical tests were performed using GraphPad Prism 7 (GraphPad Software, San Diego, USA). All Data were first tested for Gaussian distribution using the Shapiro-Wilk test. Gaussian distributed data were analyzed by t-tests or ordinary one-way ANOVA test. If Gaussian distribution was not assumed, nonparametric tests (Wilcoxon matched-pairs signed-rank test, Mann-Whitney U test or Kruskal-Wallis test) were used. Data are shown in means ± SEM. Significant differences are indicated by asterisks for P < 0.05 (*), 0.01 (**), 0.001 (***) and 0.0001 (****).

Results
Cre recombinase induced excision of exons in Stim1 and Stim2 genes in primary murine fibroblasts (pmLF) and resulting changes in mRNA and protein levels. To investigate the role of SOCE in pmLF we set out to delete exons in genes, which might be essential for SOCE. We first bred mice deficient for TRPC1 and TRPC6 to obtain Trpc1/6 −/− -double-deficient mice after a crossing over event on chromosome 9, where both genes are located. These mice are viable, fertile and have a normal life span. In clear contrast, STIM1/2 deficient mice die within a few weeks after birth 26 . To examine the effect of STIM deficiency in primary murine lung fibroblasts (pmLF), we therefore isolated these cells from mice with loxP flanked Stim1 and Stim2 genes 26 and infected them with recombinant lentiviruses expressing Cre-recombinase. By genomic PCR, we detected DNA fragments corresponding to the deleted Stim1 and Stim2 genes after lentiviral infection, respectively (Fig. 1a,b).

Receptor-operated ca 2+ entry (ROCE) is decreased in TRPC1/6-deficient, but not significant different in STIM1/2-deficient fibroblasts.
To quantify ROCE in primary murine lung fibroblasts (pmLF), endothelin 1 (Et-1) was used to activate Gq protein-coupled endothelin receptors. Stimulation of Phospholipases C-β by Gα q -subunits resulted in cleavage of phosphatidylinositol 4,5-bisphosphate (PIP2) and generation of diacylglycerol (DAG), which activates TRPC6 channels 32 . ROCE was quantified by analyzing Ca 2+ transients measured fluorometrically at wavelengths of 340 and 380 nm (Fig. 4a,b) and calculating areas under the curve (AUC) (Fig. 4c) after adding Et-1 to pmLF. Both values were significantly decreased in TRPC1/6-deficient but not in STIM1/2-deficient pmLF compared to control cells (Fig. 4a-c). We also performed recalcification experiments after application of Et-1 in Ca 2+ free buffer (Fig. S2 in Supplementary Information). While there is no difference www.nature.com/scientificreports www.nature.com/scientificreports/ in cytoplasmic Ca 2+ levels after the release of Ca 2+ from the internal stores following IP 3 production, Ca 2+ influx from the extracellular medium is reduced in TRPC1/6 −/− fibroblasts compared to Wt cells.

Store-operated ca 2+ entry (SOCE) is reduced in STIM1/2-deficient, but not in TRPC1/6-deficient primary murine lung fibroblasts (pmLF). SOCE was induced in pmLF by emptying internal Ca 2+ stores
after application of thapsigargin in Ca 2+ free buffer containing the Ca 2+ chelator EGTA and subsequent readdition of extracellular Ca 2+ . While TRPC1/6-deficient fibroblasts showed no differences, SOCE in STIM1/2-deficient cells was significantly reduced comparing peak levels and areas under the curve (AUC) (Fig. 5a-c). Ablation of all four proteins STIM1/2 and TRPC1/6 did not further reduce SOCE. Therefore, SOCE is exclusively regulated by STIM1/2 proteins and Orai channels in primary murine fibroblasts and not dependent on TRPC1 and TRPC6.

STIM1/2 deficiency reduces cell proliferation, migration and nuclear localization of NFAT transcription factors in fibroblasts.
To understand the general role of SOCE in cell function of pmLF, we quantified cell viability using a WST assay in STIM1/2-deficient fibroblasts in comparison to control cells. Cell viability was not impaired by STIM1/2 deficiency 5 to 8 days after infection with recombinant lentiviruses expressing Cre recombinase in comparison to infected and non-infected Wt pmLF (Fig. 6a). In contrast to these results. DNA synthesis as a measure of cell proliferation was significantly reduced in STIM1/2-deficient fibroblasts in comparison to Wt cells infected with recombinant lentiviruses expressing Cre recombinase (Fig. 6b,c).
To analyze SOCE on a molecular level in pmLF, we quantified nuclear levels of the Ca 2+ -dependent transcription factor nuclear factor of activated T cells (NFAT). Both isoforms NFATc1 and NFATc3 were significantly reduced in nuclear extracts of STIM1/2-deficient pmLF compared to Wt pmLF infected with recombinant lentiviruses expressing Cre recombinase as control cells (Fig. 7).
An important function during repair processes by pmLF is cell migration. To ask whether SOCE may have a role in migration of pmLF, we quantified gap closure times of migrating STIM1/2-deficient and control pmLF. It took STIM1/2-deficient pmLF significantly longer to close a defined gap in a confluent cell layer compared to control cells (Wt and Wt Cre) (Fig. 8).  Fig. 4a,b was used to quantify ROCE (c). One single dot represents the mean of at least 20 cells from one cell isolation. Asterisks mark significant differences from left to right (n > 3 mice, **P < 0.01, ***P < 0.001, ****P < 0.0001) between ratios of deficient cells compared to control cells.

Discussion
To dissect the molecular correlate of ROCE and SOCE in pmLF we deleted essential genes involved in one or presumably both processes in pmLF. In a former publication, we identified TRPC1 as the predominantly expressed TRPC channel in pmLF, while TRPC6 is up-regulated in TGF-β1 induced fibroblast to myofibroblast differentiation 30 . Three members of this TRPC family namely TRPC3, TRPC6 and TRPC7 are activated by diacylglycerol (DAG), which is produced after ligand binding to G protein-coupled receptors (GPCR) and subsequent cleavage of phosphatidylinositol 4,5-bisphosphate (PIP2) by phospholipase-C isozymes 32,33 . These so called DAG-sensitive channels are mainly expressed in brain, endothelial and smooth muscle cells of the vasculature and mediate ROCE (reviewed in 34 ). Recent findings indicate that Na + /H + exchanger regulatory factors dynamically determine the DAG sensitivity of TRPC4 and TRPC5 channels 35 and that TRPC1 proteins work as channel regulators in heteromeric complexes with all other six channels of the TRPC family 25 . Therefore, all TRPC channels are responsible for ROCE and global ablation of TRPC1 and TRPC6 proteins is sufficient to reduce intracellular Ca 2+ levels during ROCE in pmLF from TRPC1/6-deficient mice (Fig. 4).
Activation of ubiquitously expressed Orai channels in the plasma membrane is exclusively dependent on the multimerization of the ER Ca 2+ sensors STIM1 and 2 after detecting reduced Ca 2+ levels after store depletion (summarized in 36 ). As global deletion of STIM1 and 2 proteins induces early death in the corresponding gene-deficient mouse models 26 , we choose a different approach and isolated pmLF from STIM1/2 floxed mouse models, which carry loxP sites downstream and upstream of exons essential for protein function. Infection of these cells with recombinant lentiviruses expressing Cre recombinase efficiently deletes floxed exons in both genes ( Fig. 1) resulting in an almost complete absence of both proteins (Fig. 3). No compensatory up-regulation of mRNAs for Orai and TRPC channels (Fig. 2) was detected. On a protein level, TRPC1/6-deficient pmLF expressed similar amounts of STIM1/2 proteins (Fig. 3) in comparison to Wt cells and no changes in SOCE were observed (Fig. 5a). In contrast, complete ablation of STIM1/2 proteins in STIM1/2 ΔpmLF significantly reduced SOCE (Fig. 5b), but had no significant effect on ROCE induced by activation of endothelin receptors by endothelin 1 (Et-1) (Fig. 4b). ROCE, however, was significantly reduced in TRPC1/6-deficient cells (Fig. 4a) confirming the hypothesis that TRPC1 and -6 channels are responsible for receptor-dependent Ca 2+ influx in these cells.
Next, we analyzed the role of SOCE in basal cell functions of pmLF. While metabolic activity indicative of cell viability was not changed in STIM1/2-deficient compared to lentivirus infected Wt cells (Fig. 6a), quantification of DNA synthesis as a marker for cell proliferation was decreased in STIM1/2-deficient cells compared to Wt cells infected with lentiviruses (Fig. 6b,c). The role of SOCE in cell proliferation has already been demonstrated in many other cell types (reviewed in 44 ) and especially in cancer cells 45 . The nuclear translocation of NFATc transcription factors depends on increases in the intracellular Ca 2+ concentration, which makes NFATc a preferred target for SOCE induced changes in cell function 46 . Accordingly, we identified lower levels of nuclear NFATc1 and 3 levels in STIM1/2-deficient pmLF compared to lentivirus infected Wt control cells emphasizing an important role of SOCE in Ca 2+ -induced mRNA transcription of pmLF (Fig. 7). Similar results were obtained for NFATc3 in arterial smooth muscle cells 47 . We also quantified cell migration as an essential function of pmLF during repair processes in the lung and detected significant longer gap closure times in STIM1/2-deficient pmLF compared to lentivirus infected control cells (Fig. 8). The role of SOCE in cell migration was intensively studied in cancer cells (reviewed in 48 ), where intracellular Ca 2+ influx through STIM-Orai interaction affects focal adhesion turnover as a critical step in the migration mechanism 49 . A similar mechanism may be important in migration of pmLF and needs to be further analyzed.
In summary, we were able to show for the first time that SOCE, which is exclusively induced by STIM1/2 proteins in the ER of pmLF, is not dependent on TRPC1 and TRPC6, the predominantly expressed TRP channels in pmLF. Therefore, an interaction of STIM proteins and/or Orai channels with TRPC channels in these cells to mediate SOCE is unlikely. SOCE contributes to cell proliferation and migration as well as nuclear localization of nuclear factor of activated T-cells (NFATc1 and c3) in pmLF. Therefore, TRPC6 channels, which are also important for cellular functions of pmLF when differentiated to myofibroblasts by TGF-β1 30 , work independently of STIM1/2 proteins and Orai channels in this cell type.