Cholangiocytes actively contribute to the final composition of secreted bile. These cells are exposed to abnormal mechanical stimuli during obstructive cholestasis, which has a deep impact on their function. However, the effects of mechanical insults on cholangiocyte function are not understood. Combining gene silencing and pharmacological assays with live calcium imaging, we probed molecular candidates essential for coupling mechanical force to ATP secretion in mouse cholangiocytes. We show that Piezo1 and Pannexin1 are necessary for eliciting the downstream effects of mechanical stress. By mediating a rise in intracellular Ca2+, Piezo1 acts as a mechanosensor responsible for translating cell swelling into activation of Panx1, which triggers ATP release and subsequent signal amplification through P2X4R. Co-immunoprecipitation and pull-down assays indicated physical interaction between Piezo1 and Panx1, which leads to stable plasma membrane complexes. Piezo1–Panx1–P2X4R ATP release pathway could be reconstituted in HEK Piezo1 KO cells. Thus, our data suggest that Piezo1 and Panx1 can form a functional signaling complex that controls force-induced ATP secretion in cholangiocytes. These findings may foster the development of novel therapeutic strategies for biliary diseases.

Funder(s):
 ERA-Net ERACoSysMed
    Funder(s):
     European Commission
      Funder(s):
       Horizon 2020
        Funder(s):
         Centre National de la Recherche Scientifique
          Funder(s):
           Aix-Marseille University

            Cholangiocytes are epithelial cells that line the intra-hepatic ducts of the biliary tree and serve important functions in bile formation. These cells modify the composition of primary hepatocyte-derived bile through a sequence of secretory and reabsorptive processes, critical for digestion and absorption of aliments (Banales et al., 2019). Although they account for a minor subset of the total hepatic cell population, they generate up to ≈40% of the bile volume formation (Boyer, 2013).

            Cholangiocyte functions are regulated by multiple chemical factors, including hormones, peptides, nucleotides, and bile acids, through a variety of intracellular signaling pathways, ion channels, transporters, and receptors (Tabibian et al., 2013). Besides being targets for chemical mediators, cholangiocytes are also exposed to mechanical forces, including osmotic pressure and shear stress due to changes in bile composition and flow. Pathological conditions, associated with biliary obstruction, are also responsible for abnormal mechanical tension applied onto the cholangiocyte plasma membrane. Primary sclerosing cholangitis is an exemplar case, where fibro-obliteration of the bile ducts is responsible for liver cirrhosis (Pinzani and Luong, 2018). Biliary outflow obstruction leads to increased biliary pressure, triggering pressure-induced maladaptive signaling in cholangiocytes that strengthens the disease progression (Maroni et al., 2015).

            Mechanosensitive signaling has therefore emerged as a regulatory mechanism for the functions of cholangiocytes in both health and disease. However, the initiating steps translating mechanical forces into biochemical messages are not quite understood. Previous in vitro studies have demonstrated that mechanical strain and shear stress induce an elevation of intracellular Ca2+ concentration ([Ca2+]i) through an increase in plasma membrane permeability to Ca2+ ions. These responses were proposed to depend on the primary cilium (Masyuk et al., 2008; Larusso and Masyuk, 2011; Mansini et al., 2018). Primary cilium’s ability to integrate mechanical signals relies on the presence of a variety of ciliary-associated proteins that act as mechanotransducers. Among the various known mechanosensitive proteins, the importance of transient receptor potential (TRP) channels has been extensively studied. Changes in bile flow has been proposed to activate a ciliary complex formed by the association of TRPP1 (polycystin-1) and TRPP2 (polycystin-2; Masyuk et al., 2006), two members of the TRPP (polycystin) protein family thought to hold mechanosensory properties (Nauli et al., 2003). On the other hand, the cilia-associated TRPV4 channel, a well-established osmosensor (Liedtke, 2005), has been shown to transduce changes in bile tonicity in cholangiocytes (Gradilone et al., 2007). Activation of these mechanosensitive channels promotes ATP release, which in turn, through autocrine/paracrine stimulation of one or more purinergic receptors, further increases [Ca2+]i (Feranchak and Fitz, 2002). Purinergic signals promote transepithelial secretion of anions, water, and HCO3, resulting in dilution and alkalinization of bile (Roman et al.,1999; Tabibian et al., 2013).

            Therefore, there has emerged from the existing data a model in which mechanotransducers coordinate ductal bile modification through intracellular Ca2+ signaling pathways and ATP release. However, neither the mechanotransducer apparatus nor the ATP-release pathway has been definitely identified. Accordingly, the present study aims at dissecting the molecular mechanisms that couple mechanosensitive transducers to ATP secretion in cholangiocytes from mouse liver. We found that cholangiocytes from intrahepatic bile duct units (IBDUs) sense mechanical strain through the activation of Piezo1, a Ca2+-permeable cation channel directly gated by mechanical forces (Coste et al., 2010; Ranade et al., 2015; Parpaite and Coste, 2017). The ensuing ATP release requires Pannexin1 (Panx1), a member of the “gap junction” protein family that forms nonjunctional membrane channels permeable to ATP (Dahl, 2015; Taruno, 2018). Both proteins show potential for physical interaction. Thus, Piezo1 and Panx1 constitute a novel pathway to sense and signal mechanical cues in cholangiocytes. These proteins may be important new targets to regulate bile secretion in the treatment of cholestatic liver diseases.

            Animals

            All animals were used in accordance with the European Community guiding in the care and use of animals (2010/63/UE). Experiments were performed in accordance with the French government regulations.

            Osmotic solutions

            The Krebs solution consisted of (in mM) 130 NaCl, 3 KCl, 1 MgCl2, 10 HEPES, 10 glucose, and 2.5 CaCl2 (290 ± 5 mOsm ⋅ liter−1). The isotonic external solution consisted of (in mM) 65 NaCl, 3 KCl, 1 MgCl2, 2.5 CaCl2, 10 HEPES, 10 glucose, and 120 mannitol (290 ± 5 mOsm ⋅ liter−1). The hypotonic solution (160 ± 5 mOsm ⋅ liter−1) was generated by removing mannitol to the isotonic external solution, whereas the hypertonic solution (460 ± 5 mOsm ⋅ liter−1) was prepared by adding mannitol to the Krebs solution. Ca2+-free solutions were prepared by omitting CaCl2, and no EGTA was added.

            Primary cultures from mouse IBDUs

            Primary cultures of cholangiocytes from IBDUs were obtained mainly as described (Mennone et al., 1995). Male C57BL6 mice aged 10–20 wk were used for isolation of the biliary tree. Mice were anesthetized with isoflurane and sacrificed by decapitation. The liver was perfused through the portal vein for 5 min with PBS and then for 20 min with PBS containing 1 mg/ml collagenase D (Sigma-Aldrich; #COLLD-RO) and 100 µM CaCl2. After careful removal of the Glisson’s capsule, the intrahepatic biliary tree was isolated by repeated washes in PBS and thinly minced with scissors in PBS supplemented with 2 mg/ml collagenase D, 0.32 mg/ml pronase (Sigma-Aldrich; #PRON-RO), 0.04 mg/ml deoxyribonuclease (DNase; Sigma-Aldrich; #DN25), 1% FBS (Gibco; #10270), and 1% penicillin/streptomycin (P/S; Gibco; #15140-122). These pieces were digested for 45 min at 37°C and then mechanically dissociated and sifted in a 100-µm cell-strainer. After centrifugation, cells were resuspended in PBS supplemented with 2 mg/ml collagenase D, 0.34 mg/ml hyaluronidase (Sigma-Aldrich; #H3506), 0.04 mg/ml DNase, 1% FBS, and 1% P/S. Cells were incubated for 20 min at 37°C. Finally, cells were seeded on rat-tail type-I collagen–coated Petri dishes (Corning) or coverslips with Dulbecco’s modified Eagle’s medium/F-12 medium (Gibco; #31331-028) supplemented with 3.6% heat-inactivated FBS, 1% P/S, and 1‰ gentamicin (Gibco; #15750-037) and incubated at 37°C in a CO2-equilibrated incubator. A small subset of cells, negative for cytokeratin 9 (CK19) and albumin, was detected in IBDU cultures using phalloidin staining. They had elongated cell bodies with bundles of F-actin terminating at the cell surface typical of fibroblasts (data not shown). Deciliation of cholangiocytes was achieved by incubation with 4 mM of chloral hydrate for 24 h (Masyuk et al., 2006).

            Cultures of cell lines and transfection

            Human embryonic kidney (HEK) cells (293T and P1KO) and normal mouse cholangiocytes (NMCs) were cultured in Dulbecco’s modified Eagle’s medium containing 10% heat-inactivated FBS and 1% P/S. Piezo1-deficient (P1KO) HEK293T cells were generated using CRISPR/Cas9 nuclease genome engineering (Lukacs et al., 2015). Plasmid transfection (mPiezo1iresGFP; mPiezo1-GFP; pCMV3-N-GFPspark, mPanx1-flag) was achieved using Lipofectamine 2000 with 1.2 µg/ml of plasmids.

            Calcium imaging

            Cholangiocytes or HEK cells, grown on collagen type I–coated coverslips, were loaded with Fura-2-AM (2 µM) for 45 or 30 min, respectively, at 37°C in Krebs solution (Rodat-Despoix et al., 2013). Fura-2-AM–loaded cells were placed on the stage of an inverted epifluorescence microscope (Olympus IX71) equipped with a 20× UPLSAPO objective and continuously superfused. Fura-2AM was alternately excited at 340 and 380 nm, and the ratios of the resulting images (340 nm/380 nm) were produced every 1 s. The source of excitation light was a xenon arc lamp, and excitation wavelength was selected by a fast excitation filter wheel (Olympus; Illumination Systems MT20). Digital images were sampled at 12-bit resolution by a fast-scan, cooled charge-coupled device (B/W CCD) digital camera (Hamamatsu; Orca-ER). All the images were background-subtracted and controlled by CellˆR software (Olympus). A positive response was defined as a signal at least twice larger than the Δratio deviation determined from a 3-min recording of the standard baseline. The peak ratio value was determined as the difference (Δratio) between the maximum peak value and the baseline during hypotonic shock or drug application. The area under the curve (AUC) above baseline was determined using a trapezoidal method for the duration of drug application. The dose–response curve for Yoda1 was obtained from different experimental sets of coverslips that were prepared from different animals on different days. Dose–response curves were fitted to the data using nonlinear regression. The recording traces in the figures show representative examples of the effect of the drug/test of interest. The histograms/bar charts show pooled data from individual cells (n), collected from independent experiments (N). The numbers of cells analyzed in each of these independent experiments and the corresponding average responses are presented in Table S4.

            Gene silencing by small interfering RNA (siRNA)

            Cholangiocytes at day 1 in vitro (DIV1) were transfected for 7 h with Lipofectamine RNAimax, 10 nM of Piezo1-siRNA or scrambled siRNA, and 2 nM of siGLO Red Transfection Indicator without serum and antibiotic. siRNAs are listed in Table S1.

            Immunostaining

            Cholangiocytes, HEK293T-P1KO, and HEK293T cells, cultured on (collagen I and poly-D-lysin, respectively) coated glass coverslips, were first washed with PBS and fixed using 4% paraformaldehyde for 10 min, permeabilized in PBS containing 0.1% Triton X-100 for 10 min, blocked with 3% (vol/vol) gelatin for 1 h, and incubated with primary antibodies at 4°C overnight. After washes, cells were incubated with secondary antibodies (1:1,000) at room temperature for 1 h, and nuclei were stained using DAPI for 10 min. Primary and secondary antibodies are listed in Table S5.

            Reverse transcription-PCR

            Total RNA was extracted from DIV5 cholangiocytes and NMC with Trizol Reagent according to the manufacturer’s recommendation. RNA concentration was measured using a NanoDrop ND-2000 Spectrophotometer. 4 μg of total RNA were treated with DNaseI, and reverse transcription (RT) reactions with oligo(dT)20 primers were performed with the SuperScriptIII First-Strand Synthesis System kit. PCR was performed with Taq DNA Polymerase. Sequences of PCR primers are listed in Table S2.

            Real-time RT quantitative PCR (qPCR)

            Total RNA extraction and DNase treatment of cholangiocytes were achieved with the NucleoSpin RNA kit according to the manufacturer’s instructions. RT reactions were performed as described above. qPCR reactions were run in duplicate using the Kapa Sybr Fast qPCR kit on an Applied Biosystems 7500 Fast Real-Time PCR thermocycler. Sequences of specific primers are listed in Table S3. The relative quantification (RQ) method was used to compare the relative expression of the Piezo1 gene between scramble-siRNA (siCtr)– and siPiezo1-transfected cholangiocytes.

            Extracellular ATP measurements

            Detection of released ATP was performed using a luciferin/luciferase detection assay using a spectrofluorimeter-luminometer (Tecan Infinite F500). IBDU and HEK cells were cultured in black 96-well Greiner dishes with clear bottoms. ATP secretion was estimated in stimulated and unstimulated cells after a 15-min incubation period with luciferin/luciferase diluted at 1/25 for 5 min. To determine the concentration of ATP, a calibration curve was constructed using known concentrations of ATP (0.02–200 nM).

            Immunoprecipitation from transfected HEK293T-P1KO cells

            Protein extraction is detailed in the supplemental text at the end of the PDF. Protein A magnetic beads were coated with 5 µg of rabbit polyclonal anti-GFP antibody or 10 µg of rabbit polyclonal anti-flag antibody or 10 µg of rabbit polyclonal anti-Ki67 antibody (negative control) by rotating at room temperature for 10 min. Cells were added to the beads and rotated for 1 h. Protein complexes were purified by magnetization of the beads and washed with PBS-Tween 0.1% at room temperature. Proteins were eluted from beads by adding Laemmli denaturation sample buffer and incubated 10 min at 70°C.

            Immunoprecipitation from DIV5 cholangiocytes and NMCs

            DIV5 cholangiocytes were harvested using TrypLE express and pelleted by centrifugation at 300 g for 4 min. Following protein extraction, immunoprecipitation was performed with the Dynabeads coimmunoprecipitation kit according to the manufacturer’s instructions. The primary antibodies, rabbit polyclonal anti-Piezo1 (Thermo Fisher Scientific) and rabbit polyclonal anti-Ki67 (7 µg/mg of beads), were covalently linked to magnetic beads. The extraction buffer was supplemented by adding 150 mM NaCl, 20 mM N-ethylmaleimide, 1.2% Triton X-100, and a cocktail of protease inhibitors without EDTA. Proteins were eluted from beads by adding Laemmli denaturation Sample Buffer and incubated 10 min at 70°C.

            Western blot

            Protein samples were denatured by adding 0.1 M dithiothreitol and 3 M urea, and boiling for 10 min at 99°C. Protein samples were loaded in polyacrylamide gel and transferred on a nitrocellulose membrane (GE Healthcare Life Science). Proteins were controlled by Ponceau staining (0.2% wt/vol Ponceau diluted in 5% glacial acetic acid). Membranes were blocked with 5% milk diluted in TBST (Tris-buffered saline Triton) for 1 h at room temperature and probed with primary antibodies overnight at 4°C. The antibodies used are detailed in the supplemental text and Table S5. After washing with TBST, the membranes were incubated for 1 h with horseradish peroxidase (HRP)–coupled goat anti-mouse (1:1,000) or anti-rabbit secondary antibodies. HRP was revealed by chemiluminescence with blotting substrate (1:100), and membranes were scanned on an iBright CL750 imaging system.

            Mass spectrometry on DIV5 cholangiocyte immunocomplexes

            See supplemental text at the end of the PDF for further information.

            Chemicals

            Stocks of chemicals were reconstructed in either DMSO (1 mM 5-BDBD; 100 mM probenecid [PBC]; 10 mM A-804598; and 10 mM A-740003) or PBS (50 mM ATP; 1 M chloral hydrate; 1 mM gadolinium; 100 U/ml apyrase; 50 mM suramin; 250 mM carbenoxolone [CBX]; and 100 µM GsMTx4) and stored at 4°C or −20°C. Yoda1 was dissolved in 75% DMSO–25% PBS (vol/vol) at 10 mM and stored at −20°C.

            Statistical analysis

            Data were expressed as mean ± SEM. Tests for differences between two normally distributed populations were performed using two-tailed Student’s t test, and differences in percentages were analyzed by Fisher exact test (GraphPad; Prism 7.00). To compare more than two conditions, data were analyzed by ANOVA and Dunnett’s post-test. P values ≤ 0.05 were considered significant.

            Online supplemental material

            Fig. S1 shows that mouse cholangiocytes in primary culture are not sensitive to hypertonic stress. Fig. S2 shows that the primary cilium is not involved in osmosensation. Fig. S3 shows expression of putative mechanosensitive channels in cholangiocytes. Fig. S4 shows that Yoda1 does not induce ATP secretion by itself. Fig. S5 shows that Panx1 colocalizes with Piezo1 in cholangiocyte plasma membrane. Fig. S6 shows reconstitution of Piezo1/Panx1/P2X4R mechanosecretory pathway in HEK cells. Table S1 shows sequences of siRNA. Table S2 shows sequences of RT-PCR primers. Table S3 shows sequences of qPCR primers. Table S4 shows detailed information depicting the numbers of independent experiments, the number of cells analyzed in each individual experiment, and the corresponding averaged values. Units are not reported and can be found in the figures. Table S5 shows reagents and tools. Supplemental text, available at the end of the PDF, shows further information on qPCR, immunostaining, Western blot, and mass spectrometry protocols.

            Isolated IBDU cells express cholangiocyte-specific markers

            Primary cultures of cholangiocytes were made from mouse IBDUs. Quantitation of the proportion of cells expressing the cholangiocyte-specific marker CK19 revealed that ∼80% of IBDU cells were positive for this marker at DIV5 and DIV8, indicating an enriched population of intrahepatic cholangiocytes (Fig. S1, A and B). Mouse cholangiocytes showed typical polygonal body shape with relatively regular dimensions (Tabibian et al., 2013). The absence of hepatocytes was demonstrated by the lack of staining for albumin, an hepatocyte-specific marker (≤1/1,000 cells; Fig. S1 A). Other cholangiocyte specific markers, including cytokeratin 7 (CK7), cystic fibrosis transmembrane conductance regulator (CFTR), the anion exchanger 2 (AE2), and the apical sodium-dependent bile acid transporter (ASBT), were also detected using RT-PCR (Fig. S1 C).

            Hypotonic stress increases membrane permeability to Ca2+ ions

            We examined the response of cholangiocytes to hypotonic stress (160 mOsm ⋅ liter−1) using Fura-2-AM ratio live imaging and measure of Ca2+ signals (Fig. 1 A). Cholangiocytes maintained a stable [Ca2+]i when in isotonic buffer but showed a slow rise in intracellular Ca2+ when bathed with the hypotonic solution (Fig. 1 B). The onset of hypotonic responses followed a variable lag period such that individual cell responses often occurred asynchronously. On average, 74 ± 6% (range, 63–92%) of cholangiocytes were responsive to hypotonic stress (Fig. 1, B and E). Both responsive and nonresponsive cholangiocytes to hypotonic shock showed similar responses to ATP application (150 µM), indicating that nonresponsive cells were healthy (Fig. 1 C). In contrast, cholangiocytes showed no propensity to respond to hypertonic stress (Fig. S1, D and E).

            Hypotonic solution–induced Ca2+ responses were markedly reduced in Ca2+-free solution (Fig. 1 D). Only 27% of cholangiocytes still responded to hypotonic solution under this condition (Fig. 1 E), with a mean peak Δratio reduced by 30% (Fig. 1 F). In addition, most cholangiocytes gave a single Ca2+ spike in response to hypotonic stress in Ca2+-free conditions (Fig. 1 D). As a result, there was a 60% reduction in the AUC of the hypotonic stress–induced Ca2+ responses (Fig. 1 G). Interestingly, purinergic responses also displayed spike-like waveforms in Ca2+-free solution, reminiscent of hypotonic Ca2+ responses recorded likewise (Fig. 1, D, H, and I). Collectively, these data indicate that hypotonic stress induces slow Ca2+ signals in cholangiocytes, which mainly depend on external calcium influx.

            Primary cilium has no role in sensing hypotonic stress

            At DIV5, 28% of CK19-positive cells exhibited a primary cilium, as evidenced by acetylated tubulin staining (Fig. S2, A and B). Because primary cilia may sense modifications in osmolarity (Masyuk et al., 2008; Larusso and Masyuk, 2011), we tested whether deletion of the primary cilium could alter hypotonic responses. Incubation of cells with 4 mM chloral hydrate for 24 h resulted in the almost complete loss of primary cilia in CK19-expressing cells (Fig. S2, C and E). This treatment had no deleterious effects on the proportion of CK19-expressing cells (Fig. S2 D). Chloral hydrate incubation had no effect on the proportion of cholangiocytes responding to hypotonic shock, or on the amplitude of Ca2+ responses (Fig. S2, F–H). However, we observed a significant reduction of the amplitude of purinergic Ca2+ signals (Fig. S2 I), perhaps indicative of the presence of a subset of purinergic receptors located on the cholangiocyte cilium (Larusso and Masyuk, 2011). Collectively, these data indicate that the primary cilium is not required for signaling hypotonic stress.

            Hypotonic stress–induced Ca2+ signals depend on ATP release acting primarily on P2X4 receptors

            To probe the molecular mechanism that mediates plasmalemmal Ca2+ influx under hypotonic conditions, we tested the effects of apyrase, an ATP-diphosphohydrolase that catalyzes the hydrolysis of ATP. Apyrase (5 U/ml) added to hypotonic buffer reduced the proportion of hypotonic solution responding cells from 73.5% to 56.5% (Fig. 2, A and B) and the mean amplitude/AUC of the responses by ∼30% (Fig. 2, C and D). We confirmed that apyrase, at the concentration used, abolished the response to exogenous ATP (150 µM; Fig. 2, E and F).

            We found that mouse cholangiocytes at DIV5 express transcripts for P2X1, P2X2, P2X4, P2X5, and P2X7 receptor subtypes (Fig. 2 G). However the nonselective P2R antagonist suramin at 20 µM, a concentration that should affect most subtypes, except P2X7R and P2X4R subtypes (half maximal inhibitory concentration [IC50] is >100 µM; North and Jarvis, 2013), had no significant inhibitory effects on hypotonic Ca2+ responses (Fig. S2, J and K). Suramin was efficient at relatively high concentrations (200 µM), suggesting the involvement of P2X7R and P2X4R subtypes (Fig. S2, J and K). Therefore, we studied in detail the effects of the selective P2X7R antagonists A804598 and A740003, which have IC50 values of 10 and 50 nM for rodent P2X7R, respectively (Donnelly-Roberts et al., 2009). Both A-compounds, used at 10-fold the IC50, had no inhibitory effects on hypotonic Ca2+ responses (Fig. 2, H–K). Interestingly, they also had modest effects on ATP-induced Ca2+ responses, providing further support that P2X7R does not contribute a major component of purinergic signaling (Fig. 2 L).

            We further tested the effects of the P2X4R antagonist 5-BDBD, which exhibits no significant antagonist effects at other P2XRs (Coddou et al., 2019). 5-BDBD reduced by ∼50% both the amplitude and AUC of hypotonic responses (Fig. 2, M–P). Interestingly, 5-BDBD also inhibited Ca2+ responses evoked by exogenous ATP by the same amount (Fig. 2 Q). Altogether, these findings suggest that the P2X4R contributes to the ATP-dependent hypotonic calcium response.

            The mechanosensitive ion channel Piezo1 is functionally active in mouse cholangiocytes

            As a first step toward identifying the plasmalemmal channel(s) that senses hypoosmotic challenge and primes ATP release, we applied Gd3+ (50 µM), a broad-spectrum inhibitor of stretch-activated ion channels. Under Gd3+ application, 85% of cholangiocytes were unresponsive to hypotonic shock (Fig. 3, A and B). The few remaining responsive cells had reduced peak responses with features reminiscent of those seen in Ca2+-free solution (Fig. 3, A, C, and D; compare with Fig. 1 D).

            Multiple putative mechanosensitive ion channels are known to be expressed in cholangiocytes; however, with the notable exception of TRPV4, there is little evidence supporting any of these as a component of the osmosensitive transduction complex. With the use of RT-PCR assay from DIV5 primary cultures, we detected RNA for the vanilloid subtypes TRPV2 and TRPV4, the members of the polycystin-like groups TRPP1 and TRPP2 (Delmas, 2004), and the mechanosensitive ion channel Piezo1 (Coste et al., 2010; Fig. 3 F and Fig. S3 A). mPiezo1 was detected with different sets of primers (see Materials and methods and Table S2), whereas Piezo2 was barely detectable (Fig. S3 A). Immunostaining using a polyclonal Piezo1 antibody (Thermo Fisher Scientific; see Materials and methods and Table S5) further showed labeling of cholangiocytes (Fig. 3 E). To verify that the antibody interacts with the intended target, we transfected mpiezo-GFP in HEK293T-P1KO cells. Strong immunoreactivity for Piezo1 was found in HEK293T-P1KO cells that have been effectively transfected (Fig. 3 E). Unfortunately, no specific banding was observed in Western blots from mpiezo1-transfected HEK293T-P1KO cells with this anti-Piezo1 antibody. The same holds for the five other commercially available antibodies tested (see Materials and methods). However, using immunoprecipitation with the Thermo Fisher Scientific anti-Piezo1 antibody coupled to mass spectrometry, we identified mPiezo1 peptide sequences from lysates of DIV5 cholangiocytes (Fig. S3 B).

            To stimulate Piezo1 channels, we used the Piezo1 agonist, chemical compound Yoda1, which is selective for Piezo1 over Piezo2 (Botello-Smith et al., 2019; Syeda et al., 2015). Bath application of Yoda1 in DIV5 cholangiocytes evoked robust Ca2+ signals (Fig. 3 G). Approximate half-maximal effective concentration (EC50)for Yoda1 in cholangiocytes was 29.4 ± 1.25 µM (Fig. 3 H), which compares well with EC50 of 17.1 and 26.6 µM for murine and hpiezo1-transfected HEK293T cells, respectively (Syeda et al., 2015). Yoda1 signal was dependent on extracellular Ca2+ in cholangiocytes (Fig. 3 I) and exclusively related to the expression of Piezo1, since Yoda1 had no effects on HEK293T-P1KO cells (Fig. 3, J and K).

            Piezo1 contributes to hypotonic stress–induced Ca2+ signals in cholangiocytes

            To test whether Piezo1 contributes to hypotonicity-induced Ca2+ responses, we inhibited endogenous Piezo1 by using specific piezo1-siRNAs targeting four distinct sequences of piezo1 mRNA. Quantification of mRNA transcripts and data normalization were carried out by qPCR against the invariant endogenous housekeeping genes GAPDH, β-actin, and CK19. As shown in Fig. 4 A, piezo1 mRNA level decreased twofold in piezo1-siRNA–transfected cholangiocytes compared with those transfected with scrambled siRNAs. Accordingly, Piezo1-siRNA significantly reduced the responses of cholangiocytes to Yoda1 exposure (Fig. 4, B–D). Piezo1-siRNA also reduced the proportion of cholangiocytes responding to hypotonic stress and the peak/AUC of the responses (Fig. 4, E–H). Collectively, these data suggest that Piezo1 activation contributes to hypotonic calcium responses. Consistent with a role in osmosensation, overexpression of mPiezo1 promoted hypotonic responses in characteristically unresponsive HEK293T-P1KO cells to hypotonic stimulation (Fig. 4, I–K).

            Hypotonic stress–induced activation of Piezo1 promotes ATP release

            Whether ATP release induced by hypotonic stimulation was subsequent to Piezo1 activation was first assessed by pharmacological assays. Using the luciferase-ATP bioluminescent detection assay, we detected a sixfold increase in ATP release, from 2.85 ± 0.43 to 17.58 ± 2.28 nM (n = 10), following hypotonic stimulation (Fig. 5 A). ATP release was inhibited by ∼30% by Gd3+ (100 µM) and GsMTx4 (5 µM), a peptide from tarantula venom effective against Piezo1 at low micromolar concentrations (Suchyna et al., 2000; Fig. 5 A). Because none of these inhibitors are piezo-specific, we tested whether piezo1-siRNAs could inhibit hypotonic stress–induced ATP release. ATP secretion was decreased by 46% in piezo1-siRNA–treated cholangiocytes compared with control-siRNA–treated cells (Fig. 5 B).

            To further investigate Piezo1-dependent ATP release, we examined Ca2+ mobilization upon exposure to Yoda1. One trait of Yoda1 responses was their biphasic shape composed of an initial rise in [Ca2+]i, concurrent with Yoda1 application, followed by a delayed, slow [Ca2+]i increase (Fig. 5 C). The delayed “rebound” response often outperformed the initial Yoda1 response. Importantly, when apyrase was applied concurrently with Yoda1, only the initial phase was significantly diminished (Fig. 5, D–F). However, the delayed response was impaired when apyrase was maintained beyond Yoda1 exposure, but resumed as soon as apyrase was washed out (Fig. 5 G). Consequently, there was a clear relationship between the occurrence of the delayed phase and the duration of apyrase application (Fig. 5 H). Another manifestation of sustained ATP release induced by Yoda1 was the observation that cholangiocytes became refractory to ATP stimulation following exposure to Yoda1 (Fig. 5 I). Consequently, the amplitude of ATP responses was inversely correlated with the concentration of Yoda1 (Fig. 5 J). Collectively, these data indicate that Piezo1 activation by Yoda1 promotes long-lasting ATP release.

            Piezo1 is not the ATP secretion pathway

            Does Piezo1 by itself qualify as a permeation pathway for ATP? To answer this question, we took advantage of the fact that Yoda1 did not generate Ca2+ responses in HEK293T-P1KO cells. However, Yoda1 caused large Ca2+ signals in HEK293T-P1KO cells overexpressing mPiezo1 (see Fig. 3 J). Importantly, in these cells, Yoda1-induced Ca2+ signals were not inhibited by apyrase (Fig. S4, A and B), suggesting that ATP does not amplify Yoda1-induced Ca2+ responses in mPiezo1-expressing HEK293T-P1KO cells, although they express P2X4R (Fig. S4 C). As a corollary, no secretion of ATP was detected in response to Yoda1 application in HEK293T cells, HEK293T-P1KO cells, and HEK293T-P1KO cells overexpressing mPiezo1 (Fig. S4, D and E). These observations are all suggestive of a channel-mediated ATP release mechanism distinct from Piezo1.

            Piezo1-driven activation of Panx1 mediates ATP release

            We found strong expression of Panx1 in DIV5 cholangiocytes using RT-PCR, immunocytochemistry, and Western blots (Fig. S5, A–C). We investigated the role of Panx1 using CBX and PBC, which are well-known to inhibit Panx1-mediated ATP release in a variety of systems (Locovei et al., 2006; Silverman et al., 2008). Pretreatment of cholangiocytes with CBX or PBC decreased the secretion of ATP induced by hypotonic stimulation by ∼50% (Fig. 6 A). PBC also decreased the proportion of cholangiocytes that responded to hypotonic stress along with the amplitude of the Ca2+ responses (Fig. 6, B–E). Likewise, CBX also decreased significantly the proportion of cholangiocytes responsive to hypotonic stimulation from 89% in control to 51% with CBX (Fig. 6 F). In addition, PBC reduced Yoda1-induced Ca2+ responses, acting on both the initial and the delayed components of Yoda1 responses (Fig. 6, G–I). Because PBC may have off-targets, we confirmed that PBC had no direct effects on Piezo1 since Yoda1-induced responses in HEK293T-P1KO cells overexpressing Piezo1 were not impaired (Fig. 6, J and K). Collectively, these findings identify a role for Panx1 in ATP secretion secondary to hypotonic stimulation of Piezo1.

            Piezo1 and Panx1 can assemble within a protein complex

            We investigated potential interactions between Piezo1 and Panx1 by coimmunoprecipitation in HEK293T-P1KO cells expressing GFP-tagged mPiezo1 and flag-tagged mPanx1. Expression of epitope tagged proteins was confirmed by Western blotting using antibodies raised against the appropriate epitope tag. Piezo1 was detected as a monomeric ≈290-kD immunoreactive band, whereas Panx1 was detected as a double band of 42–44 and 44–47 kD (Fig. 7 A and Fig. S5 C). A similar band pattern has been observed on Western blotting for Panx1 in nine independent experiments, and may reflect differential S-nitrosylation and glycosylation states (Lohman et al., 2012). When an anti-flag antibody was used to immunoprecipitate flag-tagged Panx1, GFP-tagged mPiezo1 was found to be coimmunoprecipitated (Fig. 7 A). If Piezo1 is truly able to coassociate with Panx1, then it should also be possible to demonstrate the presence of Panx1 in samples immunoprecipitated using an anti-GFP antibody. To directly test this prediction, we performed a reciprocal immunoprecipitation protocol, using the same transfected HEK293T-P1KO cell lysates described above. When the anti-GFP antibody was used to immunoprecipitate GFP-Piezo1, flag-tagged mPanx1 was coimmunoprecipitated, consistent with an interaction (Fig. 7 B). Neither Piezo1 nor Panx1 immunoreactivity was detected in control experiments with a rabbit polyclonal antibody recognizing an unrelated epitope (anti-Ki67 antibody; Fig. 7 C, left). As additional controls, Panx1 immunoreactivity was not detected in anti-GFP immunoprecipitates from lysates of cells cotransfected with GFP and flag-tagged mPanx1 cDNAs (Fig. 7 C, right).

            We further attempted to isolate Piezo1 immunocomplexes from native cholangiocytes, but this turned to be a challenge due to the low level of Piezo1 protein under natural expression levels. Therefore, we took advantage of the use of an immortalized cell line of murine cholangiocytes that retains characteristics similar to those of freshly isolated cholangiocytes (Ueno et al., 2003). When the Thermo Fisher Scientific anti-Piezo1 antibody was used to immunoprecipitate Piezo1, Panx1 was found to be present in Piezo1 immunocomplexes (Fig. 7 D).

            Our findings therefore suggest that Piezo1 and Panx1 can be isolated in a stable complex by coimmunoprecipitation. To assess whether these complexes may localize in plasma membrane domains, we dually transfected Piezo1-GFP and Panx1-flag in HEK293T-P1KO cells. Subsequent analysis of the immunostained tags using confocal microscopy showed patterns of membrane colocalization (Fig. S5 D). Profile analysis across cotransfected cells showed dense fluorescence intensity at the cell periphery (Fig. S5 E). We confirmed the absence of detectable immunostaining in cells not expressing the cognate protein, demonstrating the specificity of both the primary and secondary antibodies (data not shown). The observed overlapping distribution is consistent with our prediction that these two proteins may interact in plasma membrane domains.

            Coexpression of mPiezo1, Panx1, and P2X4R reconstitutes a functional ATP-secreting protein complex

            We cotransfected mPiezo1-GFP, mPanx1-HA, and P2X4R-myc cDNAs in HEK293T-P1KO cells to reconstitute functional complexes. Subsequent analysis of immunostained tags using confocal microscopy showed successful coexpression of the three proteins at the cell periphery (Fig. S6 A). Yoda1-induced Ca2+ responses in these cells were markedly augmented compared with HEK293T-P1KO cells expressing only mPiezo1-GFP or mPiezo1-GFP and mPanx1-HA (Fig. S6 B), suggesting stimulation of P2X4Rs by released ATP. Consistently, application of exogenous ATP caused a massive calcium signal (+218% on average), confirming the effective expression of P2X4Rs. On average, Yoda1 responses in cells expressing the three proteins were significantly increased by 119% and 45% compared with responses in cells expressing mPiezo1-GFP alone or mPiezo1-GFP and mPanx1-HA (Fig. S6 C). Moreover, apyrase reduced Yoda1 signals in cells coexpressing the three proteins, and produced overshooting Ca2+ responses on washout, indicative of ATP release and stimulation of P2X4Rs (Fig. S6 D).

            Using molecular, pharmacological, and functional approaches, the present study identifies a new pathway for osmosensation and ATP secretion in cholangiocytes. We show that Piezo1 serves as an osmosensor triggering [Ca2+]i increase and subsequent activation of Panx1, which acts as ATP-conducting pathway. This chain of events leads to signal amplification through stimulation of autocrine/paracrine P2X4R. Collectively, these data provide a new working model for mechanosensitive secretion of ATP in bile ducts (Fig. 8).

            Hypotonic Ca2+ responses occur through cilium-independent pathways and P2X4R stimulation

            Our data indicate that mouse cholangiocytes are sensitive to hypotonic but not hypertonic challenges. Hypotonic responses are based on membrane deformation following the swelling of the cell by water influx (Wang et al., 1996). At variance with previous work (Gradilone et al., 2007), we found that hypotonic Ca2+ responses do not require the primary cilium. This discrepancy may be related to the difference in methodological approaches or preparations. Nevertheless, there are many precedents for cilium-independent osmotic responses and ATP release in many cell types (Rodat-Despoix et al., 2013; Iomini et al., 2004), including cholangiocyte-derived cells (Woo et al., 2008). While our data suggest that hypotonic response occurs through a cilium-independent pathway, they do not discount a contribution of the primary cilium in the response to shear stress (Gradilone et al., 2007).

            We found that hypotonic Ca2+ responses were essentially dependent on extracellular calcium. In a few instances, residual Ca2+ responses were observed in free [Ca2+]o, suggesting that P2YRs, linked to inositol triphosphate-mediated release of Ca2+ from intracellular endoplasmic reticulum stores, may contribute a minor component of the hypotonic Ca2+ response (Hirata et al., 2002; Dutta et al., 2008). Yet we found that propagation of Ca2+ waves from stimulated cholangiocytes to the neighboring cells prominently relies on the autocrine/paracrine action of ATP on P2X4R, the dominant subtype in rodent cholangiocytes (Doctor et al., 2005; Woo et al., 2010). P2R stimulation is known to increase transepithelial Cl secretion and to contribute to transport of water and HCO3 in both rodent and human biliary epithelial models (Roman et al., 1999; Zsembery et al., 1998). Thus, stimulation of P2X4R may be an important regulatory step linking membrane-directed forces to coordinated cholangiocyte responses regulating bile formation and secretion.

            Piezo1 serves as an osmosensor in cholangiocytes

            Piezo1 has been involved in mechanosensitive ATP release pathways in different cell systems (Cinar et al., 2015; Miyamoto et al., 2014; Wang et al., 2016; Diem et al., 2020). Here we showed that Piezo1 is an important determinant of mechanosensation in mouse cholangiocytes. Piezo1 is endowed with properties that make it a likely candidate for a cholangiocyte osmosensor (Coste et al., 2010; Ranade et al., 2015). Down-regulation of Piezo1 by siRNAs resulted in substantial inhibition of both hypotonic Ca2+ responses and ATP secretion. We obtained similar inhibitory effects with the spider toxin blocker GsMTx4, and gadolinium, although both might have off-targets. Additionally, activation of Piezo1 by Yoda1 induced long-lasting Ca2+ responses in cholangiocytes, which require ATP secretion and P2XR signaling. Collectively, these data indicate that Piezo1 contributes to hypotonic Ca2+ responses and ATP release in cholangiocytes.

            A previous search for osmosensitive ion channels in cholangiocytes pointed to the involvement of TRPV4, located on the primary cilium (Gradilone et al., 2007). More recently, TRPV4 has also been shown to translate fluid flow into intracellular signaling and biliary secretion (Li et al., 2020). It is therefore plausible that both Piezo1 and TRPV4 serve as mechanosensors in cholangiocytes as neither Piezo1 (this study) nor TRPV4 (Gradilone et al., 2007) knock-down abolished hypotonic Ca2+ responses. Thus, the presence of different mechanosensors endows cholangiocytes with the aptitude to distinguish different mechanical stimuli, and eventually to signal them differentially.

            Piezo1–Panx1 complex couples force detection to ATP secretion

            Our data indicate that Piezo1 is necessary but not sufficient for ATP secretion. Previous studies have proposed two nonexclusive mechanisms for ATP secretion in cholangiocytes, that is, exocytosis of ATP enriched vesicles and release through ATP-permeable CFTR channels (Fitz, 2007). Our investigation is suggestive of an additional mechanism involving Panx1-mediated ATP release. Supporting this idea, we found that pharmacological inhibition of Panx1 strongly reduced ATP secretion in response to both hypotonic stress and Yoda1 exposure. A broad set of experimental evidence indicates that Panx1 can mediate ATP release under physiological and pathological conditions in a variety of cell types (Dahl, 2015). Panx1 can form a transmembrane channel called pannexon, permitting ATP efflux from cells down its concentration gradient.

            How might Piezo1 regulate the function of Panx1 channels? Prior studies have shown that an increase in cytoplasmic Ca2+ concentration can activate Panx1 (Locovei et al., 2006). A tantalizing hypothesis is therefore that calcium influx mediated by Piezo1 provides a triggering mechanism for Panx1 activation. This functional relationship may be enhanced by physical promiscuity. By a combination of immunocytochemistry and coimmunoprecipitation techniques, we were able to provide evidence that both proteins can colocalize and interact physically in recombinant systems and NMCs. Although these findings suggest an ability to interact, directly or through intervening proteins, we could not demonstrate whether the Piezo1–Panx1 complex exists in native cholangiocytes. Thus, whether Piezo1 physically interacts with Panx1 in native cholangiocytes will have to be resolved in further studies.

            In conclusion, the present study was initiated to explore stretch-sensitive mechanisms and ATP release in cholangiocytes. Our data are consistent with the idea that both Piezo1 and Panx1 molecules are necessary for eliciting downstream effects of hypotonic stress. By mediating a rise in intracellular Ca2+, Piezo1 acts as a mechanosensor responsible for translating cell swelling into activation of Panx1, which is critical for ATP release. These findings provide new insight into the molecular mechanisms controlling ATP secretion and may facilitate the development of novel therapeutic strategies for liver diseases.

            Jeanne M. Nerbonne served as editor.

            The authors thank Shannon Glaser (Texas A&M University System Health Science Center, Bryan, TX) for sharing the NMC cell line, Ardem Patapoutian (Department of Neuroscience, Scripps Research, La Jolla, CA) for sharing the mpiezo1 clone and HEK293T-P1KO cell line, Nancy Osorio for critical discussion, Axel Fernandez and Ophélie Picard for technical assistance, and Maya Belghazi for proteomic analysis.

            This work was supported by the ERA-Net ERACoSysMed, DYNAFLOW collaborative project (European Commission/Horizon 2020) to Patrick Delmas and by recurrent operating grants from Centre National de la Recherche Scientifique and Aix-Marseille University.

            The authors declare no competing financial interests.

            Author contributions: A. Desplat, Conceptualization: Equal; Data curation: Lead; Formal analysis: Lead; Writing – original draft: Supporting; Writing – review & editing: Supporting; E. Gros, Investigation: Supporting; Data curation: Supporting; Formal analysis: Supporting; V. Penalba, Investigation: Supporting; Data curation: Supporting; Formal analysis: Supporting; T. Parpaite, Investigation: Supporting; Writing – review & editing: Supporting; B. Coste, Investigation: Supporting; Writing – review & editing: Supporting. P. Delmas, Conceptualization: Lead; Data curation: Lead; Formal analysis: Lead; Investigation: Lead; Writing-original draft: Lead; Writing-review & editing: Lead.

            Banales
            ,
            J.M.
            ,
            R.C.
            Huebert
            ,
            T.
            Karlsen
            ,
            M.
            Strazzabosco
            ,
            N.F.
            LaRusso
            , and
            G.J.
            Gores
            .
            2019
            .
            Cholangiocyte pathobiology
            .
            Nat. Rev. Gastroenterol. Hepatol.
            16
            :
            269
            281
            .
            Botello-Smith
            ,
            W.M.
            ,
            W.
            Jiang
            ,
            H.
            Zhang
            ,
            A.D.
            Ozkan
            ,
            Y.C.
            Lin
            ,
            C.N.
            Pham
            ,
            J.J.
            Lacroix
            , and
            Y.
            Luo
            .
            2019
            .
            A mechanism for the activation of the mechanosensitive Piezo1 channel by the small molecule Yoda1
            .
            Nat. Commun.
            10
            :
            4503
            .
            Boyer
            ,
            J.L.
            2013
            .
            Bile formation and secretion
            .
            Compr. Physiol.
            3
            :
            1035
            1078
            .
            Cinar
            ,
            E.
            ,
            S.
            Zhou
            ,
            J.
            DeCourcey
            ,
            Y.
            Wang
            ,
            R.E.
            Waugh
            , and
            J.
            Wan
            .
            2015
            .
            Piezo1 regulates mechanotransductive release of ATP from human RBCs
            .
            Proc. Natl. Acad. Sci. USA.
            112
            :
            11783
            11788
            .
            Coddou
            ,
            C.
            ,
            R.
            Sandoval
            ,
            M.J.
            Hevia
            , and
            S.S.
            Stojilkovic
            .
            2019
            .
            Characterization of the antagonist actions of 5-BDBD at the rat P2X4 receptor
            .
            Neurosci. Lett.
            690
            :
            219
            224
            .
            Coste
            ,
            B.
            ,
            J.
            Mathur
            ,
            M.
            Schmidt
            ,
            T.J.
            Earley
            ,
            S.
            Ranade
            ,
            M.J.
            Petrus
            ,
            A.E.
            Dubin
            , and
            A.
            Patapoutian
            .
            2010
            .
            Piezo1 and Piezo2 are essential components of distinct mechanically activated cation channels
            .
            Science.
            330
            :
            55
            60
            .
            Dahl
            ,
            G.
            2015
            .
            ATP release through pannexon channels
            .
            Philos. Trans. R. Soc. Lond. B Biol. Sci.
            370
            :20140191.
            Delmas
            ,
            P.
            2004
            .
            Polycystins: from mechanosensation to gene regulation
            .
            Cell.
            118
            :
            145
            148
            .
            Diem
            ,
            K.
            ,
            M.
            Fauler
            ,
            G.
            Fois
            ,
            A.
            Hellmann
            ,
            N.
            Winokurow
            ,
            S.
            Schumacher
            ,
            C.
            Kranz
            , and
            M.
            Frick
            .
            2020
            .
            Mechanical stretch activates piezo1 in caveolae of alveolar type I cells to trigger ATP release and paracrine stimulation of surfactant secretion from alveolar type II cells
            .
            FASEB J.
            34
            :
            12785
            12804
            .
            Doctor
            ,
            R.B.
            ,
            T.
            Matzakos
            ,
            R.
            McWilliams
            ,
            S.
            Johnson
            ,
            A.P.
            Feranchak
            , and
            J.G.
            Fitz
            .
            2005
            .
            Purinergic regulation of cholangiocyte secretion: identification of a novel role for P2X receptors
            .
            Am. J. Physiol. Gastrointest. Liver Physiol.
            288
            :
            G779
            G786
            .
            Donnelly-Roberts
            ,
            D.L.
            ,
            M.T.
            Namovic
            ,
            B.
            Surber
            ,
            S.X.
            Vaidyanathan
            ,
            A.
            Perez-Medrano
            ,
            Y.
            Wang
            ,
            W.A.
            Carroll
            , and
            M.F.
            Jarvis
            .
            2009
            .
            [3H]A-804598 ([3H]2-cyano-1-[(1S)-1-phenylethyl]-3-quinolin-5-ylguanidine) is a novel, potent, and selective antagonist radioligand for P2X7 receptors
            .
            Neuropharmacology.
            56
            :
            223
            229
            .
            Dutta
            ,
            A.K.
            ,
            K.
            Woo
            ,
            R.B.
            Doctor
            ,
            J.G.
            Fitz
            , and
            A.P.
            Feranchak
            .
            2008
            .
            Extracellular nucleotides stimulate Cl currents in biliary epithelia through receptor-mediated IP3 and Ca2+ release
            .
            Am. J. Physiol. Gastrointest. Liver Physiol.
            295
            :
            G1004
            G1015
            .
            Feranchak
            ,
            A.P.
            , and
            J.G.
            Fitz
            .
            2002
            .
            Adenosine triphosphate release and purinergic regulation of cholangiocyte transport
            .
            Semin. Liver Dis.
            22
            :
            251
            262
            .
            Fitz
            ,
            J.G.
            2007
            .
            Regulation of cellular ATP release
            .
            Trans. Am. Clin. Climatol. Assoc.
            118
            :
            199
            208
            .
            Gradilone
            ,
            S.A.
            ,
            A.I.
            Masyuk
            ,
            P.L.
            Splinter
            ,
            J.M.
            Banales
            ,
            B.Q.
            Huang
            ,
            P.S.
            Tietz
            ,
            T.V.
            Masyuk
            , and
            N.F.
            Larusso
            .
            2007
            .
            Cholangiocyte cilia express TRPV4 and detect changes in luminal tonicity inducing bicarbonate secretion
            .
            Proc. Natl. Acad. Sci. USA.
            104
            :
            19138
            19143
            .
            Hirata
            ,
            K.
            ,
            J.F.
            Dufour
            ,
            K.
            Shibao
            ,
            R.
            Knickelbein
            ,
            A.F.
            O’Neill
            ,
            H.P.
            Bode
            ,
            D.
            Cassio
            ,
            M.V.
            St-Pierre
            ,
            N.F.
            Larusso
            ,
            M.F.
            Leite
            , and
            M.H.
            Nathanson
            .
            2002
            .
            Regulation of Ca(2+) signaling in rat bile duct epithelia by inositol 1,4,5-trisphosphate receptor isoforms
            .
            Hepatology.
            36
            :
            284
            296
            .
            Iomini
            ,
            C.
            ,
            K.
            Tejada
            ,
            W.
            Mo
            ,
            H.
            Vaananen
            , and
            G.
            Piperno
            .
            2004
            .
            Primary cilia of human endothelial cells disassemble under laminar shear stress
            .
            J. Cell Biol.
            164
            :
            811
            817
            .
            Larusso
            ,
            N.F.
            , and
            T.V.
            Masyuk
            .
            2011
            .
            The role of cilia in the regulation of bile flow
            .
            Dig. Dis.
            29
            :
            6
            12
            .
            Li
            ,
            Q.
            ,
            C.
            Kresge
            ,
            K.
            Boggs
            ,
            J.
            Scott
            , and
            A.
            Feranchak
            .
            2020
            .
            Mechanosensor transient receptor potential vanilloid member 4 (TRPV4) regulates mouse cholangiocyte secretion and bile formation
            .
            Am. J. Physiol. Gastrointest. Liver Physiol.
            318
            :
            G277
            G287
            .
            Liedtke
            ,
            W.
            2005
            .
            TRPV4 as osmosensor: a transgenic approach
            .
            Pflugers Arch.
            451
            :
            176
            180
            .
            Locovei
            ,
            S.
            ,
            L.
            Bao
            , and
            G.
            Dahl
            .
            2006
            .
            Pannexin 1 in erythrocytes: function without a gap
            .
            Proc. Natl. Acad. Sci. USA.
            103
            :
            7655
            7659
            .
            Lohman
            ,
            A.W.
            ,
            J.L.
            Weaver
            ,
            M.
            Billaud
            ,
            J.K.
            Sandilos
            ,
            R.
            Griffiths
            ,
            A.C.
            Straub
            ,
            S.
            Penuela
            ,
            N.
            Leitinger
            ,
            D.W.
            Laird
            ,
            D.A.
            Bayliss
            , and
            B.E.
            Isakson
            .
            2012
            .
            S-nitrosylation inhibits pannexin 1 channel function
            .
            J. Biol. Chem.
            287
            :
            39602
            39612
            .
            Lukacs
            ,
            V.
            ,
            J.
            Mathur
            ,
            R.
            Mao
            ,
            P.
            Bayrak-Toydemir
            ,
            M.
            Procter
            ,
            S.M.
            Cahalan
            ,
            H.J.
            Kim
            ,
            M.
            Bandell
            ,
            N.
            Longo
            ,
            R.W.
            Day
            , et al
            .
            2015
            .
            Impaired PIEZO1 function in patients with a novel autosomal recessive congenital lymphatic dysplasia
            .
            Nat. Commun.
            6
            :
            8329
            .
            Mansini
            ,
            A.P.
            ,
            E.
            Peixoto
            ,
            K.M.
            Thelen
            ,
            C.
            Gaspari
            ,
            S.
            Jin
            , and
            S.A.
            Gradilone
            .
            2018
            .
            The cholangiocyte primary cilium in health and disease
            .
            Biochim. Biophys. Acta Mol. Basis Dis.
            1864
            :
            1245
            1253
            .
            Maroni
            ,
            L.
            ,
            B.
            Haibo
            ,
            D.
            Ray
            ,
            T.
            Zhou
            ,
            Y.
            Wan
            ,
            F.
            Meng
            ,
            M.
            Marzioni
            , and
            G.
            Alpini
            .
            2015
            .
            Functional and structural features of cholangiocytes in health and disease
            .
            Cell. Mol. Gastroenterol. Hepatol.
            1
            :
            368
            380
            .
            Masyuk
            ,
            A.I.
            ,
            T.V.
            Masyuk
            ,
            P.L.
            Splinter
            ,
            B.Q.
            Huang
            ,
            A.J.
            Stroope
            , and
            N.F.
            LaRusso
            .
            2006
            .
            Cholangiocyte cilia detect changes in luminal fluid flow and transmit them into intracellular Ca2+ and cAMP signaling
            .
            Gastroenterology.
            131
            :
            911
            920
            .
            Masyuk
            ,
            A.I.
            ,
            T.V.
            Masyuk
            , and
            N.F.
            LaRusso
            .
            2008
            .
            Cholangiocyte primary cilia in liver health and disease
            .
            Dev. Dyn.
            237
            :
            2007
            2012
            .
            Mennone
            ,
            A.
            ,
            D.
            Alvaro
            ,
            W.
            Cho
            , and
            J.L.
            Boyer
            .
            1995
            .
            Isolation of small polarized bile duct units
            .
            Proc. Natl. Acad. Sci. USA.
            92
            :
            6527
            6531
            .
            Miyamoto
            ,
            T.
            ,
            T.
            Mochizuki
            ,
            H.
            Nakagomi
            ,
            S.
            Kira
            ,
            M.
            Watanabe
            ,
            Y.
            Takayama
            ,
            Y.
            Suzuki
            ,
            S.
            Koizumi
            ,
            M.
            Takeda
            , and
            M.
            Tominaga
            .
            2014
            .
            Functional role for Piezo1 in stretch-evoked Ca2+ influx and ATP release in urothelial cell cultures
            .
            J. Biol. Chem.
            289
            :
            16565
            16575
            .
            Nauli
            ,
            S.M.
            ,
            F.J.
            Alenghat
            ,
            Y.
            Luo
            ,
            E.
            Williams
            ,
            P.
            Vassilev
            ,
            X.
            Li
            ,
            A.E.
            Elia
            ,
            W.
            Lu
            ,
            E.M.
            Brown
            ,
            S.J.
            Quinn
            , et al
            .
            2003
            .
            Polycystins 1 and 2 mediate mechanosensation in the primary cilium of kidney cells
            .
            Nat. Genet.
            33
            :
            129
            137
            .
            North
            ,
            R.A.
            , and
            M.F.
            Jarvis
            .
            2013
            .
            P2X receptors as drug targets
            .
            Mol. Pharmacol.
            83
            :
            759
            769
            .
            Parpaite
            ,
            T.
            , and
            B.
            Coste
            .
            2017
            .
            Piezo channels
            .
            Curr. Biol.
            27
            :
            R250
            R252
            .
            Pinzani
            ,
            M.
            , and
            T.V.
            Luong
            .
            2018
            .
            Pathogenesis of biliary fibrosis
            .
            Biochim. Biophys. Acta Mol. Basis Dis.
            1864
            :
            1279
            1283
            .
            Ranade
            ,
            S.S.
            ,
            R.
            Syeda
            , and
            A.
            Patapoutian
            .
            2015
            .
            Mechanically Activated Ion Channels
            .
            Neuron.
            87
            :
            1162
            1179
            .
            Rodat-Despoix
            ,
            L.
            ,
            J.
            Hao
            ,
            M.
            Dandonneau
            , and
            P.
            Delmas
            .
            2013
            .
            Shear stress-induced Ca2+ mobilization in MDCK cells is ATP dependent, no matter the primary cilium
            .
            Cell Calcium.
            53
            :
            327
            337
            .
            Roman
            ,
            R.M.
            ,
            A.P.
            Feranchak
            ,
            K.D.
            Salter
            ,
            Y.
            Wang
            , and
            J.G.
            Fitz
            .
            1999
            .
            Endogenous ATP release regulates Cl secretion in cultured human and rat biliary epithelial cells
            .
            Am. J. Physiol.
            276
            :
            G1391
            G1400
            .
            Silverman
            ,
            W.
            ,
            S.
            Locovei
            , and
            G.
            Dahl
            .
            2008
            .
            Probenecid, a gout remedy, inhibits pannexin 1 channels
            .
            Am. J. Physiol. Cell Physiol.
            295
            :
            C761
            C767
            .
            Suchyna
            ,
            T.M.
            ,
            J.H.
            Johnson
            ,
            K.
            Hamer
            ,
            J.F.
            Leykam
            ,
            D.A.
            Gage
            ,
            H.F.
            Clemo
            ,
            C.M.
            Baumgarten
            , and
            F.
            Sachs
            .
            2000
            .
            Identification of a peptide toxin from Grammostola spatulata spider venom that blocks cation-selective stretch-activated channels
            .
            J. Gen. Physiol.
            115
            :
            583
            598
            .
            Syeda
            ,
            R.
            ,
            J.
            Xu
            ,
            A.E.
            Dubin
            ,
            B.
            Coste
            ,
            J.
            Mathur
            ,
            T.
            Huynh
            ,
            J.
            Matzen
            ,
            J.
            Lao
            ,
            D.C.
            Tully
            ,
            I.H.
            Engels
            , et al
            .
            2015
            .
            Chemical activation of the mechanotransduction channel Piezo1
            .
            eLife.
            4
            :e07369.
            Tabibian
            ,
            J.H.
            ,
            A.I.
            Masyuk
            ,
            T.V.
            Masyuk
            ,
            S.P.
            O’Hara
            , and
            N.F.
            LaRusso
            .
            2013
            .
            Physiology of cholangiocytes
            .
            Compr. Physiol.
            3
            :
            541
            565
            .
            Taruno
            ,
            A.
            2018
            .
            ATP Release Channels
            .
            Int. J. Mol. Sci.
            19
            :
            808
            .
            Ueno
            ,
            Y.
            ,
            G.
            Alpini
            ,
            K.
            Yahagi
            ,
            N.
            Kanno
            ,
            Y.
            Moritoki
            ,
            K.
            Fukushima
            ,
            S.
            Glaser
            ,
            G.
            LeSage
            , and
            T.
            Shimosegawa
            .
            2003
            .
            Evaluation of differential gene expression by microarray analysis in small and large cholangiocytes isolated from normal mice
            .
            Liver Int.
            23
            :
            449
            459
            .
            Wang
            ,
            Y.
            ,
            R.
            Roman
            ,
            S.D.
            Lidofsky
            , and
            J.G.
            Fitz
            .
            1996
            .
            Autocrine signaling through ATP release represents a novel mechanism for cell volume regulation
            .
            Proc. Natl. Acad. Sci. USA.
            93
            :
            12020
            12025
            .
            Wang
            ,
            S.
            ,
            R.
            Chennupati
            ,
            H.
            Kaur
            ,
            A.
            Iring
            ,
            N.
            Wettschureck
            , and
            S.
            Offermanns
            .
            2016
            .
            Endothelial cation channel PIEZO1 controls blood pressure by mediating flow-induced ATP release
            .
            J. Clin. Invest.
            126
            :
            4527
            4536
            .
            Woo
            ,
            K.
            ,
            A.K.
            Dutta
            ,
            V.
            Patel
            ,
            C.
            Kresge
            , and
            A.P.
            Feranchak
            .
            2008
            .
            Fluid flow induces mechanosensitive ATP release, calcium signalling and Cl transport in biliary epithelial cells through a PKCzeta-dependent pathway
            .
            J. Physiol.
            586
            :
            2779
            2798
            .
            Woo
            ,
            K.
            ,
            M.
            Sathe
            ,
            C.
            Kresge
            ,
            V.
            Esser
            ,
            Y.
            Ueno
            ,
            J.
            Venter
            ,
            S.S.
            Glaser
            ,
            G.
            Alpini
            , and
            A.P.
            Feranchak
            .
            2010
            .
            Adenosine triphosphate release and purinergic (P2) receptor-mediated secretion in small and large mouse cholangiocytes
            .
            Hepatology.
            52
            :
            1819
            1828
            .
            Zsembery
            ,
            A.
            ,
            C.
            Spirlì
            ,
            A.
            Granato
            ,
            N.F.
            LaRusso
            ,
            L.
            Okolicsanyi
            ,
            G.
            Crepaldi
            , and
            M.
            Strazzabosco
            .
            1998
            .
            Purinergic regulation of acid/base transport in human and rat biliary epithelial cell lines
            .
            Hepatology.
            28
            :
            914
            920
            .
            This article is available under a Creative Commons License (Attribution 4.0 International, as described at https://creativecommons.org/licenses/by/4.0/).