TRPV4 Channel Participates in Receptor-Operated Calcium Entry and Ciliary Beat Frequency Regulation in Mouse Airway Epithelial Cells

TRPV4 channel participates in receptor-operated calcium entry and ciliary beat frequency regulation in mouse airway epithelial cells

Ivan M. Lorenzo*, Wolfgang Liedtke†, Michael J. Sanderson‡, and Miguel A. Valverde*§

*Laboratory of Molecular Physiology and Channelopathies, Department of Experimental and Health Sciences, Universitat Pompeu Fabra, Parc Recerca Biome`dica de Barcelona, Carrer Doctor Aiguader 88, 08003 Barcelona, Spain; †Center for Translational Neuroscience, Duke University Medical Center, Durham, NC 27708; and ‡Department of Physiology, University of Massachusetts Medical School, Worcester, MA 01655

Edited by Ramoń Latorre, Centro de Neurociencias, Universidad de Valparaiso, Valparaiso, Chile, and approved June 25, 2008 (received for review April 24, 2008) The rate of mucociliary clearance in the airways is a function of The molecular identity of the SOCE pathway has just started to ciliary beat frequency (CBF), and this, in turn, is increased by emerge because of the discovery of two key molecules, the Ca2ϩ increases in intracellular calcium. The TRPV4 cation channel medi- sensor within the depleted stores and the plasma membrane ates Ca2؉ influx in response to mechanical and osmotic stimuli in store-operated channel, the STIM and ORAI proteins, respectively ciliated epithelia. With the use of a TRPV4-deficient mouse, we now (reviewed in refs. 5 and 6). However, the role for the family of show that TRPV4 is involved in the airways’ response to physio- transient receptor potential (TRP) cationic channels in the SOCE logically relevant physical and chemical stimuli. Ciliary TRPV4 mechanism remains unknown (7–9). expression in tracheal epithelial cells was confirmed with immu- Less clear is the molecular identity of the ROCE pathway. There nofluorescence in TRPV4؉/؉ mice. Ciliated tracheal cells from is a fundamental difference between ROCE and SOCE. Although TRPV4؊/؊ mice showed no increases in intracellular Ca2؉ and CBF the latter depends on the presence of a Ca2ϩ sensor within the in response to the synthetic activator 4␣-phorbol 12,13-didecano- endoplasmic reticulum (ER), ROCE is independent of the Ca2ϩ ate (4␣PDD) and reduced responses to mild temperature, another content of the ER but involves one or several of the signaling TRPV4-activating stimulus. Autoregulation of CBF in response to molecules resulting from the stimulation of PLC activity, such as ؊/؊ high viscosity solutions is preserved in TRPV4 despite a reduced diacylglycerol, IP3, or arachidonic acid (5). Different TRPC chan- Ca2؉ signal. More interestingly, TRPV4 contributed to an ATP- nels have been shown to respond to these downstream molecules induced increase in CBF, providing a pathway for receptor-oper- (10), thereby implicating them in the ROCE mechanism. Outside ated Ca2؉ entry but not store-operated Ca2؉ entry as the former of the TRPC subfamily of channels, members of the melastatin mechanism is lost in TRPV4؊/؊ cells. Collectively, these results (TRPM) (11) and vanilloid (TRPV) subfamilies (12–14) appear to suggest that TRPV4 is predominantly located in the cilia of tracheal be regulated by phosphatidylinositides, key molecules in the PLC epithelial cells and plays a key role in the transduction of physical signaling pathway, although none of the TRP channels have been .and chemical stimuli into a Ca2؉ signal that regulates CBF and formally implicated in ROCE mucociliary transport. Moreover, these studies implicate the par- The TRPV4 cation channel, a member of the TRP vanilloid ticipation of TRPV4 in receptor-operated Ca2؉ entry. subfamily, responds to a range of stimuli, including osmotic cell swelling, mechanical stress, temperature, endogenous arachidonic ATP ͉ trachea ͉ temperature ͉ transient receptor potential channel ͉ acid metabolites, and phorbol esters (15, 16). As a result, 4␣- store-operated calcium entry phorbol 12,13-didecanoate (4␣PDD) has become a valuable phar- macological tool to functionally test TRPV4 activity, because

4␣PDD interacts directly with transmembrane domains 3 and 4 of PHYSIOLOGY n mammalian airways, ciliated and mucus-secreting epithelial TRPV4 (17). TRPV4 can be also sensitized by coapplication of cells form a structural and functional unit that functions as a I different stimuli (18–20) or participation of different cell signaling conveyor belt system for particle transport. In this analogy, cilia pathways (21). TRPV4 messenger and protein have been identified provide the power, whereas the mucus serves as the viscoelastic belt in both native ciliated epithelial cells of oviducts (21–23) and cell (1). A critical factor regulating the velocity of mucociliary transport lines derived from human ciliated airway cells (24). In these is the ciliary beat frequency (CBF), a mechanism in which cytosolic 2ϩ 2ϩ epithelial cells, the TRPV4 channel plays a key role in cell volume Ca plays a major role (1, 2). Increases in cytosolic Ca are ϩ ϩ homeostasis, by activating Ca2 -dependent K channels (25, 26) associated with increases in CBF (3, 4), although the ultimate ϩ and in the regulation of CBF, by providing a Ca2 entry pathway molecular mechanism explaining CBF regulation by Ca2ϩ remains in response to changes in fluid viscosity or tonicity (21, 22). TRPV4 controversial (3). Both mechanical and chemical (paracrine) stim- splice variants, some of which do not oligomerize and are retained ulation of epithelial ciliated cells can lead to an increase in intracellularly, have also been found in airway epithelial cell lines intracellular Ca2ϩ concentration and the consequent enhancement (27) (for detailed review on TRP splicing, see ref. 28). of CBF (2). ATP-mediated activation of G protein-coupled recep- In this study, we have evaluated the coupling of TRPV4 channel tors, typically P2Y receptors, is one of the strongest signals known activity to the regulation of CBF in tracheal ciliated cells by using to increase CBF (2). As with many other agonists of G protein- coupled receptors, ATP promotes both the release of Ca2ϩ from 2ϩ inositol trisphosphate (IP3)-sensitive intracellular stores and Ca Author contributions: M.A.V. designed research; I.M.L. performed research; W.L. and M.J.S. entry across the plasma membrane (4), and the latter process is contributed new reagents/analytic tools; I.M.L. and M.A.V. analyzed data; and W.L., M.J.S., more evident at micromolar concentrations of ATP where a and M.A.V. wrote the paper. sustained Ca2ϩ influx occurs. The stimulation of Ca2ϩ influx The authors declare no conﬂict of interest. involves signaling from the depleted stores to plasma membrane This article is a PNAS Direct Submission. 2ϩ Ca channels (store-operated calcium entry, SOCE; also known as §To whom correspondence should be addressed. E-mail: [email protected]. 2ϩ capacitative Ca entry) and/or through a phospholipase C (PLC)- This article contains supporting information online at www.pnas.org/cgi/content/full/ dependent mechanism (receptor-operated calcium entry, ROCE; 0803970105/DCSupplemental. ϩ also known as noncapacitative Ca2 entry) (5). © 2008 by The National Academy of Sciences of the USA

www.pnas.org͞cgi͞doi͞10.1073͞pnas.0803970105 PNAS ͉ August 26, 2008 ͉ vol. 105 ͉ no. 34 ͉ 12611–12616 Downloaded by guest on September 27, 2021 Fig. 1. Detection and activity of the TRPV4 channel in mouse ciliated tracheal cells. (A and B) Differential interference contrast (Upper Left), TRPV4 (green, Upper Right), ␣-tubulin (red, Lower Left), and merged (Lower Right) images obtained from TRPV4ϩ/ϩ (A) and TRPV4Ϫ/Ϫ (B) cells. Colocalization of TRPV4 and tubulin appears as yellow. (Scale bar, 10 ␮m) (C) Western blot showing a typical TRPV4 double band of the predicted molecular size (Ϸ100 kDa) in TRPV4ϩ/ϩ but not in TRPV4Ϫ/Ϫ trachea. Tubulin was detected in both TRPV4ϩ/ϩ and TRPV4Ϫ/Ϫ trachea. (D and E) Representative cytosolic Ca2ϩ signals obtained from TRPV4ϩ/ϩ (D) and TRPV4Ϫ/Ϫ (E) ciliated tracheal cells exposed to 10 ␮M 4␣PDD. (F) Average [Ca2ϩ] increases measured after 10 min in 4␣PDD. TRPV4ϩ/ϩ (ﬁlled column, n ϭ 38) and TRPV4Ϫ/Ϫ cells (open column, n ϭ 40). *, P Ͻ 0.05, Student’s unpaired test.

TRPV4-KO mice. We report that TRPV4-deficient (TRPV4Ϫ/Ϫ) signals can also be associated with the activation of CBF, we mice display a significant reduction in both Ca2ϩ entry and CBF measured CBF in TRPV4ϩ/ϩ and TRPV4Ϫ/Ϫ tracheal cells by using activation in response to different stimuli that can activate TRPV4. high-speed digital video microscopy. Basal CBF of ciliated tracheal Furthermore, we have identified a role for TRPV4 in the ATP- cells did not differ between TRPV4ϩ/ϩ (10.7 Ϯ 0.4 Hz; n ϭ 37) and dependent ROCE mechanism and the subsequent increase in CBF. TRPV4Ϫ/Ϫ (10.3 Ϯ 0.3 Hz; n ϭ 37, P Ͼ 0.05 vs. TRPV4ϩ/ϩ, measured at room temperature) and was not affected by removal Results of extracellular Ca2ϩ (Fig. 2A), in accordance with previous studies Expression and Localization of TRPV4 in Mouse Ciliated Tracheal Cells. suggesting that basal Ca2ϩ CBF is not directly under the influence Fig. 1 shows ciliated tracheal cells from wild-type (TRPV4ϩ/ϩ) (Fig. of Ca2ϩ (3). However, TRPV4ϩ/ϩ tracheal cells, unlike their 1A) and KO (TRPV4Ϫ/Ϫ) mice (Fig. 1B). TRPV4 immunofluo- TRPV4Ϫ/Ϫ counterparts, had increased CBF in response to 10 ␮M rescence (green) was clearly identified in the cilia of TRPV4ϩ/ϩ but 4␣PDD. Fig. 2B shows the time course of the relative changes in not in the cilia of TRPV4Ϫ/Ϫ cells. Double staining with anti-tubulin CBF of a TRPV4ϩ/ϩ and TRPV4Ϫ/Ϫ cells exposed to 4␣PDD. antibody (red) to mark the cilia axoneme confirmed the ciliary Mean increases in CBF are shown in Fig. 2C. localization of TRPV4. Additional immunofluorescence images are shown in supporting information (SI) Fig. S1. Cytoplasmic immu- Response of Ciliated Tracheal Cells to Physical Stimuli Activating noreactivity spots [more pronounced than in hamster oviductal TRPV4. We tested the effect of warm temperature and high viscous ciliated cells (21)] were detected in both TRPV4ϩ/ϩ and TRPV4Ϫ/Ϫ solutions on the generation of Ca2ϩ signals and modulation of CBF cells, a clear indication of their nonspecificity. Molecular identifi- in TRPV4ϩ/ϩ and TRPV4Ϫ/Ϫ ciliated tracheal cells. Switching the cation of TRPV4 was also investigated by Western blot. Fig. 1C temperature of the bathing solution from 24°C to 38°C triggered a shows only a double band of the expected size in TRPV4ϩ/ϩ Ca2ϩ response characterized by a peak followed by a slow decline trachea, which also suggests that the cytoplasmic immunoreactivity toward the baseline (Fig. 3A), whereas TRPV4Ϫ/Ϫ cells showed a spots shown in Fig. 1B were TRPV4 nonspecific. Similar nonspecific reduction in both the Ca2ϩ peak and the more sustained component immunoreactivity has been reported in kidney sections of without significant modification of the time constant of the signal TRPV4ϩ/ϩ and TRPV4Ϫ/Ϫ mice probed with an antibody raised relaxation (0.95 Ϯ 0.12 min for TRPV4ϩ/ϩ and 1.12 Ϯ 0.07 min for against the same C-terminal epitope of rat TRPV4 (29). TRPV4Ϫ/Ϫ, P Ͼ 0.05). Accordingly, TRPV4Ϫ/Ϫ cells exposed to Intracellular Ca2ϩ measurements were carried out to test func- warm temperatures responded with a smaller increase in CBF tional expression of TRPV4 in primary cultures of tracheal explants (Fig. 3B). exposed to the relatively specific TRPV4 agonist 4␣PDD (10 ␮M). Ciliated tracheal cells from TRPV4ϩ/ϩ mice (Fig. 3C) responded Monitoring intracellular Ca2ϩ concentration in fura-2 loaded cili- to high viscous solutions (20% dextran solutions) with an oscillatory ated tracheal cells showed significant increases that commenced at Ca2ϩ signal (3.9 Ϯ 0.6 peaks/cell per 10 min) in 19 of 71 cells (26%). different times in many TRPV4ϩ/ϩ cells (Fig. 1D) but were com- However, TRPV4Ϫ/Ϫ ciliated cells (Fig. 3D) typically presented an pletely absent in TRPV4Ϫ/Ϫ ciliated cells (Fig. 1E). A quantitative initial transient peak (18 of 85 cells, 21%) with sporadic Ca2ϩ analysis of the Ca2ϩ signal measured after 10 min in 4␣PDD is oscillations (1.4 Ϯ 0.1 peaks/cell per 10 min; P Ͻ 0.001 vs. shown in Fig. 1F. To check whether the TRPV4-mediated Ca2ϩ TRPV4ϩ/ϩ). We also tested the impact of TRPV4 disruption on

12612 ͉ www.pnas.org͞cgi͞doi͞10.1073͞pnas.0803970105 Lorenzo et al. Downloaded by guest on September 27, 2021 Fig. 2. Basal and 4␣PDD-stimulated CBF. (A) Mean CBF before and after removal of extracellular Ca2ϩ in TRPV4ϩ/ϩ (Left) and TRPV4Ϫ/Ϫ (Right) cells (n ϭ 15 for each condition). P Ͼ 0.05, one way ANOVA and Bonferroni post hoc.(B) Representative traces of changes in CBF (% of control) with respect to time of TRPV4ϩ/ϩ (ﬁlled circles) and TRPV4Ϫ/Ϫ (open circles) cells in response to 10 ␮M4␣PDD. (C) Mean normalized CBF response (% control) measured after 10 min in 10 ␮M4␣PDD. TRPV4ϩ/ϩ (ﬁlled Ϫ Ϫ squares, n ϭ 12) and TRPV4 / cells (open squares, n ϭ 9). *, P Ͻ 0.05, one-way ANOVA and Bonferroni post hoc.

CBF response to high viscous loads. After the addition of solutions the Ca2ϩ signal pattern between the two genotypes (Fig. 3 C and D), containing 5% or 20% dextran, CBF declined to a new stable value no differences were detected in the maintenance of a steady-state within 5 min (Fig. 3E); however, despite the apparent difference in CBF in response to either 5% (4.8 cP) or 20% (73 cP) dextran solutions (Fig. 3E).

Participation of TRPV4 in the ATP-Induced Ca2؉ Signal. We have recently reported a cross-talk between the ATP-PLC-inositol trisphosphate receptor (IP3R) pathway and TRPV4 to initiate and maintain the oscillatory Ca2ϩ signal triggered by mechanical and osmotic stimuli (21). In the present study, the role of TRPV4 in the generation of ATP-mediated Ca2ϩ signals and regulation of CBF was addressed. The Ca2ϩ signal obtained after the addition of 20 ␮M ATP showed clear differences between the two genotypes. Although ciliated TRPV4ϩ/ϩ and TRPV4Ϫ/Ϫ cells showed no statistical difference in peak increases in [Ca2ϩ], the sustained component was significantly reduced (by Ϸ30%) in the latter (Fig. 4 A and B) without significant differences in the time constant of the signal relaxation (1.45 Ϯ 0.2 and 0.9 6 Ϯ 0.09 min for TRPV4ϩ/ϩ and TRPV4Ϫ/Ϫ, P Ͼ 0.05). TRPV4Ϫ/Ϫ cells also showed a diminished increase in the CBF when exposed to 20 ␮M ATP (Fig. 4C). Altogether, the data suggested the contribution of TRPV4 to the sustained component of the Ca2ϩ signal and its coupling to the acceleration of CBF induced by 20 ␮M ATP. As with many other G protein-coupled receptor agonists, the Ca2ϩ signal generated by low ATP concentrations (200 nM) usually presents a more oscil- PHYSIOLOGY latory pattern instead of a large transient peak followed by sustained elevation (Fig. 4D). Neither the pattern (Fig. 4D) nor the amplitude of the Ca2ϩ signal (Fig. 4E) generated by 200 nM ATP was significantly different in ciliated tracheal cells from TRPV4ϩ/ϩ and TRPV4Ϫ/Ϫ mice. Accordingly, no differences in the CBF response to 200 nM ATP were detected between the two genotypes (Fig. 4F). The sustained component of the Ca2ϩ signal recorded with 20 ␮M ATP can also be evaluated with a Ca2ϩ-free protocol. Fig. 5 shows representative Ca2ϩ traces obtained from TRPV4ϩ/ϩ (Fig. 5A) and TRPV4Ϫ/Ϫ (Fig. 5B) ciliated tracheal cells exposed to 20 ␮M ATP in the absence of extracellular Ca2ϩ followed by replacement of external Ca2ϩ. The Ca2ϩ entry component after Ca2ϩ Fig. 3. Effect of temperature and high viscous solutions on Ca2ϩ and CBF replacement in the presence of ATP was clearly reduced in Ϫ/Ϫ responses in tracheal cells. (A) Mean [Ca2ϩ] increases in response to a change in TRPV4 cells (Fig. 5C). To distinguish whether TRPV4 partic- the bathing solution temperature from 24°C to 38°C in TRPV4ϩ/ϩ (filled circles, nϭ ipates in the ROCE or SOCE mechanism, we use the sarcoplasmic 153) and TRPV4Ϫ/Ϫ cells (open circles, n ϭ 100). (B) Mean normalized CBF response (endoplasmic) reticulum Ca2ϩ pump inhibitor thapsigargin (TG). (% control) measured after 10 min at 38 °C in TRPV4ϩ/ϩ (filled sqaures, n ϭ 27) and The ability of TG to empty the intracellular stores and the subse- Ϫ Ϫ ϩ ϩ TRPV4 / cells (open squares, n ϭ 17). *, P Ͻ 0.05, for TRPV4 / (38°C) versus all quent activation of SOCE (without the involvement of receptor other conditions, one way ANOVA and Bonferroni post hoc. (C and D) Different activation) provides a well known protocol to functionally differ- intracellular Ca2ϩ signals (⌬ ratio 340/380) obtained from TRPV4ϩ/ϩ (C) and TRPV4Ϫ/Ϫ (D) primary cultures stimulated with 20% dextran solutions. (E) Time entiate SOCE from ROCE, which is defined to require seven course of CBF changes in TRPV4ϩ/ϩ (filled symbols; n ϭ 9) and TRPV4Ϫ/Ϫ (empty transmembrane receptors, G proteins, and PLC activation. The 2ϩ 2ϩ symbols; n ϭ 7) tracheal ciliated cells exposed to 5% dextran (triangles) and 20% Ca entry component after Ca replacement after TG (1 ␮M) ϩ ϩ dextran (circles) solutions. stimulation (SOCE) was not different between TRPV4 / and

Lorenzo et al. PNAS ͉ August 26, 2008 ͉ vol. 105 ͉ no. 34 ͉ 12613 Downloaded by guest on September 27, 2021 Fig. 4. Ca2ϩ and CBF response to ATP in ciliated tracheal cells. (A) Time course of mean Ca2ϩ responses to 20 ␮M ATP in TRPV4ϩ/ϩ (filled circles) and TRPV4Ϫ/Ϫ (open circles) ciliated tracheal cells. (B) Comparison of mean Ca2ϩ responses measured 3 min after the addition of 20 ␮M ATP. Number of cells for A and B are: ϩ ϩ Ϫ Ϫ TRPV4 / (filled squares), n ϭ 172 cells and TRPV4 / cells (open sqaures), n ϭ 135. *, P Ͻ 0.05, Student’s unpaired test. (C) Mean normalized CBF response (% ϩ ϩ Ϫ Ϫ control) measured after 3 min in the presence of 20 ␮M ATP in TRPV4 / (filled squares, n ϭ 7) and TRPV4 / cells (open squares, n ϭ 13). *, P Ͻ 0.05, for the TRPV4ϩ/ϩ versus TRPV4Ϫ/Ϫ response to ATP, one-way ANOVA, and Bonferroni post hoc. Although not marked, the response of both genotypes to ATP is statistically different versus the control conditions (P Ͻ 0.05), one-way ANOVA, and Bonferroni post hoc. (D) Representative time course of Ca2ϩ responses to 200 nM ATP in TRPV4ϩ/ϩ (filled circles) and TRPV4Ϫ/Ϫ (open circles) ciliated tracheal cells. (E) Mean responses after 3 min in 200 nM ATP of TRPV4ϩ/ϩ (filled squares, n ϭ 70) and TRPV4Ϫ/Ϫ (open squares, n ϭ 83) cells. P Ͼ 0.05, Student’s unpaired test. (F) Mean normalized CBF response (% control) measured after 3 min in the presence of 200 nM ATP in TRPV4ϩ/ϩ (filled squares, n ϭ 5) and TRPV4Ϫ/Ϫ (open squares, n ϭ 5) cells. The response of both genotypes to 200 nM ATP is statistically different versus the control conditions (P Ͻ 0.05) but not versus each other (P Ͼ 0.05), one-way ANOVA, and Bonferroni post hoc.

TRPV4Ϫ/Ϫ cells (Fig. 5 D–F). Also, the pattern and amplitude of We found no differences in the basal CBF or the CBF in the the peak Ca2ϩ signals generated by ATP and TG in Ca2ϩ-free absence of extracellular Ca2ϩ between TRPV4ϩ/ϩ and TRPV4Ϫ/Ϫ media (reflecting intracellular Ca2ϩ release) were not significantly cells, in agreement with previous studies suggesting the Ca2ϩ- different between TRPV4ϩ/ϩ and TRPV4Ϫ/Ϫ cells (2 Ϯ 0.13 and independent nature of basal CBF and the Ca2ϩ dependency of the 1.98 Ϯ 0.09 for ATP; 1.83 Ϯ 0.2 and 1.8 Ϯ 0.2 for TG, respectively; stimulated CBF (3). We have shown that TRPV4Ϫ/Ϫ cells, unlike P Ͼ 0.05). TRPV4ϩ/ϩ cells and cultured human airway epithelial cells (20, 24, 27), are unresponsive to 4␣PDD when measuring intracellular Discussion [Ca2ϩ]. The CBF response to 4␣PDD was also abrogated in This study presents different lines of evidence supporting the role TRPV4Ϫ/Ϫ ciliated cells, providing evidence for the coupling of of TRPV4 in the Ca2ϩ signaling of ciliated tracheal cells and its TRPV4 activity to the regulation of CBF. Interestingly, other coupling to the regulation of CBF. Our study opens three main 4␣-phorbol isomers positively modulate CBF through PKC phos- points for discussion: (i) the coupling of TRPV4 activity to the phorylation (32). The facts that 4␣PDD binds and activates TRPV4 regulation of CBF, (ii) the minor role of TRPV4 in the autoregu- (17) without the involvement of PKC (16) and that the response to lation of CBF in response to increased viscous load, and (iii) the 4␣PDD is totally lost in different TRPV4Ϫ/Ϫ cells (ref. 29 and this participation of TRPV4 in the ROCE mechanism. study) strongly suggest that the 4␣PDD effects are mainly TRPV4- The main task of ciliated cells is the transport of mucus and mediated, although we cannot completely rule out the participation trapped particles. A primary determinant of mucus transport is of other pathways. the CBF, which can be regulated by different signals (2) with Both native and heterologously expressed TRPV4 channels increases in intracellular Ca2ϩ being particularly relevant (1). respond to warm temperatures, in the range of 30°C to 40°C, with Although elevations of intracellular [Ca2ϩ] accelerate CBF, transient increases in [Ca2ϩ] followed by a slow decay toward the Ca2ϩ signals are most efficient in regulating CBF when produced baseline (33, 34). TRPV4Ϫ/Ϫ ciliated tracheal cells show a reduced at the base of the cilia (30). Therefore, Ca2ϩ entry pathways Ca2ϩ signal and CBF response to changes in temperature from 24°C designed to modulate CBF should be localized close to the base to 38°C, reinforcing the observation that TRPV4 activity can be of the cilia within the apical membrane of the cell. Previously, coupled to CBF regulation. Interestingly, considering that maximal TRPV4 has been principally localized at the base of the cilia of increase in CBF is reached within 33°C to 43°C in human and hamster oviduct cells (21), and its activation by the synthetic bovine airways cells (35, 36), our data suggest that TRPV4 may play agonist 4␣PDD increased the CBF in these cells (22). In the an important role in the control of CBF under physiological present study, we show that TRPV4-specific immunoreactivity is temperatures. restricted mostly to cilia of TRPV4ϩ/ϩ tracheal epithelial cells, The second point for discussion is the role of TRPV4 in the similar to hamster oviduct ciliated cells (21) and rat ciliated autoregulation of CBF in mouse tracheal cells. Mucus-transporting cholangiocytes (31), whereas the TRPV4 signal is absent in the ciliated cells are capable of maintaining their CBF under high cilia of epithelial cells obtained from TRPV4Ϫ/Ϫ mice. viscosity conditions without reducing mucus transport, a process

12614 ͉ www.pnas.org͞cgi͞doi͞10.1073͞pnas.0803970105 Lorenzo et al. Downloaded by guest on September 27, 2021 basal concentration of ATP in airway surface liquid can increase from the low nanomolar range (39) to the low micromolar range in response to certain stimuli (40). ATP-elicited cellular responses in mouse tracheal epithelia are linked mainly to P2Y2 receptors with minor contributions of other receptors (41). Low micromolar ATP 2ϩ concentrations induce a peak release of Ca from IP3-sensitive stores and a more sustained Ca2ϩ influx, whereas lower concentrations of ATP typically generate oscillatory Ca2ϩ signals in ciliated cells (4). Ultimately, the rise in intracellular [Ca2ϩ] triggers activation of CBF (4). We have shown that targeted disruption of TRPV4 reduced (by 30%) the sustained component of the Ca2ϩ signal generated by ATP (at micromolar concentrations), which should have an important impact on mucociliary transport as small increments in CBF (16%) result in a large increases (56%) in surface liquid velocity and mucus clearance (42). This effect appeared to be related to the participation of TRPV4 in the ROCE but not in the SOCE mechanism, because ciliated tracheal cells from TRPV4Ϫ/Ϫ mice showed no deficiency in the Ca2ϩ influx elicited by store depletion using TG. The reduced Ca2ϩ influx in TRPV4Ϫ/Ϫ is also accompanied by a diminished response of the CBF to micromolar concentrations of ATP, confirming again the coupling of TRPV4 to CBF regulation. Interestingly, the lack of TRPV4 does not affect the Ca2ϩ signal [as reported for TRPV4Ϫ/Ϫ urothelial cells (29)] or CBF acceleration induced by 200 nM ATP. At this low ATP concentration, the Ca2ϩ signal in tracheal cells is mainly oscillatory, a pattern that, at least for HEK cells, has been associated with the activity of store-operated pathways involving STIM1 and ORAI1 proteins without major contribution of TRP channels (43). Participation of TRPV4 in ciliated tracheal cells ROCE contrasts with the impact of TRPC1 in the Ca2ϩ homeostasis of salivary gland epithelia, where it clearly affects both ROCE and SOCE (44). TRPC1, because all TRPCs except TRPC7, required STIM1 for its activation by receptor stimulation (45). We suspect that this is not the case for TRPV4 as no involvement of TRPV4 in SOCE was found. The molecular mechanism linking Fig. 5. ATP- and thapsigargin-stimulated Ca2ϩ entry in ciliated tracheal cells. TRPV4 to ROCE in ciliated tracheal cells is unknown at present Ca2ϩ signals measured in ciliated tracheal cells stimulated with 20 ␮M ATP and is an interesting focus for future studies. In conclusion, we (A–C)or1␮MTG(D–F)inCa2ϩ-free solutions (white box, reflecting intracel- provide molecular evidence for the physiological function of lular Ca2ϩ release) followed by addition of 1.2 mM Ca2ϩ to the bathing TRPV4 in ROCE and regulation of CBF in mouse tracheal solution (black box) to detect Ca2ϩ influx. (C and F), Mean Ca2ϩ entry from epithelial cells. TRPV4ϩ/ϩ (filled circles, n ϭ 59) and TRPV4Ϫ/Ϫ (open circles, n ϭ 77) cells. Materials and Methods PHYSIOLOGY 2ϩ Chemicals and Solutions. All chemicals were purchased from Sigma–Aldrich known as CBF autoregulation (37) that depends on Ca entry and except dextran T-500 (500,000 Daltons; Amersham Pharmacia), fura2-AM (Mo- subsequent activation of cilia (22). TRPV4 has been proposed to lecular Probes), Hank’s balanced salt solution (HBSS; Gibco), and collagen type I 2ϩ participate in the generation of the oscillatory Ca signal required from rat tail (Upstate). Isotonic bathing solutions used for imaging experiments to activate this autoregulation in hamster oviduct ciliated cells (21, contained (in mM): 140 NaCl; 5 KCl; 1.2 CaCl2; 0.5 MgCl2; 5 glucose; 10 Hepes, pH 22). TRPV4ϩ/ϩ mouse tracheal cells also respond to high viscosity 7.4; and 305 mosmol/liter. Ca2ϩ-free extracellular solutions were obtained by ϩ solutions with oscillatory Ca2 signals, although the amplitude of replacing CaCl2 with MgCl2 and adding 0.5 mM EGTA. CBF measurements were the Ca2ϩ peaks is smaller and the percentage of cells responding carried out in phenol red-free HBSS supplemented with 25 mM Hepes (pH 7.4). ␣ with oscillatory signals (26%) is lower than in hamster oviduct cells ATP and 4 PDD were dissolved in milli-Q distilled water and ethanol, respectively. (76%) (21). The transient Ca2ϩ signals seen in TRPV4Ϫ/Ϫ rarely The viscosity of the isotonic solution was increased by adding 5% or 20% dextran T-500, which does not change the osmolality (305 mosmol/liter) of the solution. oscillate and resemble the response observed in hamster oviduct 2ϩ ciliated cells in the absence of Ca influx (21). Together, these data Primary Cultures of Mouse Tracheal Cells. Adult (10–14 weeks old) TRPV4 2ϩ confirm the involvement of TRPV4 in the maintenance of Ca wild-type mice (TRPV4ϩ/ϩ) and null (TRPV4Ϫ/Ϫ) mice generated in a C57Bl6/J oscillations under conditions of high viscosity. Mouse tracheal cells background (46) were used for these studies. Primary cultures of airway tracheal in this study maintained higher CBF in the presence of dextran- epithelial cells were prepared as described in ref. 47. Briefly, the trachea was containing solutions than hamster oviduct cells (22) without sig- opened and cut into rings, placed onto 1 mg/ml collagen-coated coverslips, and nificant differences between TRPV4ϩ/ϩ and TRPV4Ϫ/Ϫ cells. The cultured in DMEM containing 1 mg/liter glucose and supplemented with 10% FBS reason for the relative small contribution of TRPV4 to the auto- and 1% penicillin-streptomycin at 37°C in 5% CO2 for 3–4 days. All experiments regulation of CBF in mouse airways, compared with hamster were carried out with beating ciliated cells. Animals were maintained and ex- oviduct cells, is unknown at present, although it may be related to periments were performed according to the guidelines issued by both the Insti- 2ϩ tutional Ethics Committees of the Institut Municipal d’Investigacio´Me` dica (Uni- the smaller Ca increases generated in the mouse tracheal cells versitat Pompeu Fabra) and the Institutional Animal Care and Use Committee of when exposed to high viscous solutions. the University of Massachusetts Medical School. The third point of our study refers to the role of TRPV4 in the ATP-triggered response of mouse tracheal ciliated cells. Airway Immunodetection. Epithelial cell isolation, immunofluorescence, and laser con- epithelia release ATP in response to a myriad of stimuli, including focal immunolocalization were performed as described in ref. 21. Mouse tracheas mechanical stimulation induced by tidal breathing (38–40). The were fixed with 4% paraformaldehyde in a 3.7% (wt/vol) sucrose solution before

Lorenzo et al. PNAS ͉ August 26, 2008 ͉ vol. 105 ͉ no. 34 ͉ 12615 Downloaded by guest on September 27, 2021 cell isolation using 0.005% Protease XIV dissolved in Ca2ϩ-free isotonic solution. Measurement of CBF. CBF of cultured ciliated cells was detected and quantified Isolated cells were attached to 1.5% gelatin-coated coverslips by spinning at 500 with a high-speed digital imaging system as described in ref. 4. In general, rpm for 3 min using a cytospin (Shandon; Thermo Fisher Scientific) and fixation phase-contrast images (512 ϫ 512 pixels) were collected at 120–135 frames sϪ1 procedure for 10 min more at room temperature. Single cells were permeabilized (fps) with a high speed CCD camera using a frame grabber and recording with Tween 20 (0.05%) in PBS (15 min at room temperature), and nonspecific software from Video Savant (IO Industries). CBF was determined from the vari- interactions were blocked with PBS containing 1.5% BSA, 5% FBS, and 0.05% ation in the light intensity of the image that resulted from the repetitive motion Tween 20. Isolated epithelial cells were incubated overnight at 4 °C with the of cilia. Video recordings of beating cilia lasting 1-2 s were analyzed, and the primary antibodies diluted in the same blocking solution. frequency of each ciliary beat cycle was determined from the period of each cycle The anti-TRPV4 polyclonal antibody (21, 27) was used at 6.4 ␮g/ml. A commer- of the gray-intensity waveform. cial anti-␣-tubulin (Sigma-Aldrich) was diluted to 1:500. For immunodetection, we used a goat anti-rabbit IgG Alexa Fluor 488 (Molecular Probes) and a goat Ϯ anti-mouse IgG Alexa Fluor 555 (Molecular Probes) diluted 1:750 in the same Statistics. All data were expressed as means SEM. Statistical analysis was solution used with the primary antibodies. Images were taken at room temper- performed with the Student’s paired or unpaired tests or one-way ANOVA using ature with an inverted Leica SP2 confocal microscope using a Leica HCX Pl APO SigmaPlot or OriginPro software. Bonferroni’s test was used for post hoc com- 63 ϫ 1.32 NA Oil Ph3 CS objective. parison of means. The criterion for a significant difference was a final value of Proteins from TRPV4ϩ/ϩ and TRPV4Ϫ/Ϫ tracheas were also detected by Western P Ͻ 0.05. blot technique using the anti-TRPV4 antibody (1:100) as described in refs. 21, 22, and 24. ACKNOWLEDGMENTS. We thank G. Horvath for training with cell cultures; S.A. Serra, A. Garcia-Elias, and X. Sanjuan for help with immunodetection Measurement of Intracellular [Ca2؉]. Cytosolic Ca2ϩ signals were determined at protocols; B Penãlba and C. Plata for help with animal care; and A. Lindy for room temperature (Ϸ24°C unless otherwise indicated) in ciliated cells loaded comments and proofreading. This work was supported by Spanish Ministries ␮ ⅐ with 4.5 M fura-2 AM (45 min) as described in detail in refs. 24 and 25. Cytosolic of Education and Science Grant SAF2006-4973, Health, (Red HERACLES, Fondo 2ϩ [Ca ] increases are presented as the ratio of emitted fluorescence (510 nm) after de Investigacioń Sanitaria) and Generalitat de Catalunya Grant 2005SGR266. excitation at 340 nm and 380 nm relative to the ratio measured before cell M.J.S. was supported by National Institutes of Health Grant HL71930. stimulation (ratio 340/380).

1. Satir P, Sleigh MA (1990) The physiology of cilia and mucociliary interactions. Annu Rev 25. Fernandez-Fernandez JM, Nobles M, Currid A, Vazquez E, Valverde MA (2002) Maxi Kϩ Physiol 52:137–155. channel mediates regulatory volume decrease response in a human bronchial epithe- 2. Salathe M (2006) Regulation of mammalian ciliary beating. Annu Rev Physiol 69:401–422. lial cell line. Am J Physiol 283:C1705–C1714. 3. Ma W, Silberberg SD, Priel Z (2002) Distinct axonemal processes underlie spontaneous 26. Vazquez E, Nobles M, Valverde MA (2001) Defective regulatory volume decrease in human and stimulated airway ciliary activity. J Gen Physiol 120:875–885. cystic fibrosis tracheal cells because of altered regulation of intermediate conductance 4. Zhang L, Sanderson MJ (2003) Oscillations in ciliary beat frequency and intracellular Ca2ϩ-dependent potassium channels. Proc Natl Acad Sci USA 98:5329–5334. calcium concentration in rabbit tracheal epithelial cells induced by ATP. J Physiol 27. Arniges M, Fernandez-Fernandez JM, Albrecht N, Schaefer M, Valverde MA (2006) 546:733–749. Human TRPV4 channel splice variants revealed a key role of ankyrin domains in 5. Putney JW, Jr (2007) Recent breakthroughs in the molecular mechanism of capacitative multimerization and trafficking. J Biol Chem 281:1580–1586. calcium entry (with thoughts on how we got here). Cell Calcium 42:103–110. 28. Vazquez E, Valverde MA (2006) A review of TRP channels splicing. Semin Cell Dev Biol 6. Lewis RS (2007) The molecular choreography of a store-operated calcium channel. 17:607–617. Nature 446:284–287. 29. Gevaert T, et al. (2007) Deletion of the transient receptor potential cation channel 7. Lopez JJ, Salido GM, Pariente JA, Rosado JA (2006) Interaction of STIM1 with endog- TRPV4 impairs murine bladder voiding. J Clin Invest 117:3453–3462. enously expressed human canonical TRP1 upon depletion of intracellular Ca2ϩ stores. 30. Tamm S (1994) Ca2ϩ channels and signalling in cilia and flagella. Trends Cell Biol 4:305–310. J Biol Chem 281:28254–28264. 31. Gradilone SA, et al. (2007) Cholangiocyte cilia express TRPV4 and detect changes in 8. Huang GN, et al. (2006) STIM1 carboxyl-terminus activates native SOC, I(crac) and TRPC1 channels. Nat Cell Biol 8:1003–1010. luminal tonicity inducing bicarbonate secretion. Proc Natl Acad Sci USA 104:19138– 9. Liao Y et al. (2008) Functional interactions among Orai1, TRPCs, and STIM1 suggest a 19143. STIM-regulated heteromeric Orai/TRPC model for SOCE/Icrac channels. Proc Natl Acad 32. Mwimbi XK, et al. (2002) 4alpha-Phorbol negates the inhibitory effects of phorbol- Sci USA 105:2895–2900. 12-myristate-13-acetate on human cilia and alters the phosphorylation of PKC. FEBS 10. Trebak M, Lemonnier L, Smyth JT, Vazquez G, Putney JW, Jr (2007) Phospholipase C- Lett 530:31–36. coupled receptors and activation of TRPC channels. Handb. Exp. Pharmacol. 593–614. 33. Guler AD, et al. (2002) Heat-evoked activation of the ion channel, TRPV4. J Neurosci 11. Zhang Z, Okawa H, Wang Y, Liman ER (2005) Phosphatidylinositol 4,5-bisphosphate 22:6408–6414. rescues TRPM4 channels from desensitization. J Biol Chem 280:39185–39192. 34. Watanabe H, et al. (2002)Heat-evoked activation of TRPV4 channels in a HEK293 cell expres- 12. Chuang HH, et al. (2001)Bradykinin and nerve growth factor release the capsaicin sion system and in native mouse aorta endothelial cells. J Biol Chem 277:47044–47051. receptor from PtdIns(4,5)P2-mediated inhibition. Nature 411:957–962. 35. Sisson JH, Stoner JA, Ammons BA, Wyatt TA (2003) All-digital image capture and 13. Lukacs V, et al. (2007) Dual regulation of TRPV1 by phosphoinositides. J Neurosci whole-field analysis of ciliary beat frequency. J Microsc 211:103–111. 27:7070–7080. 36. Mwimbi XK, Muimo R, Green MW, Mehta A (2003) Making human nasal cilia beat in 14. Lee J, Cha SK, Sun TJ, Huang CL (2005) PIP2 activates TRPV5 and releases its inhibition the cold: A real time assay for cell signalling. Cell Signal 15:395–402. by intracellular Mg2ϩ. J Gen Physiol 126:439–451. 37. Johnson NT, Villalon M, Royce FH, Hard R, Verdugo P (1991) Autoregulation of beat 15. Nilius B, Vriens J, Prenen J, Droogmans G, Voets T (2004) TRPV4 calcium entry channel: frequency in respiratory ciliated cells. Demonstration by viscous loading. Am Rev Respir A paradigm for gating diversity. Am J Physiol 286:C195–C205. Dis 144:1091–1094. ϩ 16. Xu F, Satoh E, Iijima T (2003) Protein kinase C-mediated Ca2 entry in HEK 293 cells 38. Schwiebert EM, Zsembery A (2003) Extracellular ATP as a signaling molecule for transiently expressing human TRPV4. Br J Pharmacol 140:413–421. epithelial cells. Biochim Biophys Acta 1615:7–32. 17. Vriens J, Owsianik G, Janssens A, Voets T, Nilius B (2007) Determinants of 4 alpha- 39. Tarran R, et al. (2005) Normal and cystic fibrosis airway surface liquid homeostasis The phorbol sensitivity in transmembrane domains 3 and 4 of the cation channel TRPV4. effects of phasic shear stress and viral infections. J Biol Chem 280:35751–35759. J Biol Chem 282:12796–12803. 40. Okada SF, Nicholas RA, Kreda SM, Lazarowski ER, Boucher RC (2006) Physiological 18. Gao X, Wu L, O’Neil RG (2003) Temperature-modulated diversity of TRPV4 channel regulation of ATP release at the apical surface of human airway epithelia. J Biol Chem gating: activation by physical stresses and phorbol ester derivatives through protein 281:22992–23002. kinase C-dependent and -independent pathways. J Biol Chem 278:27129–27137. 41. Homolya L, Watt WC, Lazarowski ER, Koller BH, Boucher RC (1999) Nucleotide- 19. Alessandri-Haber N, Dina OA, Joseph EK, Reichling D, Levine JD (2006) A transient regulated calcium signaling in lung fibroblasts and epithelial cells from normal and receptor potential vanilloid 4-dependent mechanism of hyperalgesia is engaged by P2Y(2) receptor (Ϫ/Ϫ) mice. J Biol Chem 274:26454–26460. concerted action of inflammatory mediators. J Neurosci 26:3864–3874. 42. Seybold ZV, et al. (1990) Mucociliary interaction in vitro: Effects of physiological and 20. Grant AD, et al. (2007) Protease-activated receptor 2 sensitizes the transient receptor inflammatory stimuli. J Appl Physiol 68:1421–1426. potential vanilloid 4 ion channel to cause mechanical hyperalgesia in mice. J Physiol 578:715–733. 43. Wedel B, Boyles RR, Putney JW, Jr., Bird GS (2007) Role of the store-operated calcium entry proteins Stim1 and Orai1 in muscarinic cholinergic receptor-stimulated calcium 21. Fernandes J, et al. (2008) IP3 sensitizes TRPV4 channel to the mechano- and osmotrans- ducing messenger 5Ј-6Ј-epoxyeicosatrienoic acid. J Cell Biol 181:143–155. oscillations in human embryonic kidney cells. J Physiol 579:679–689. 2ϩ 22. Andrade YN, et al. (2005) TRPV4 channel is involved in the coupling of fluid viscosity 44. Liu X, et al. (2007) Attenuation of store-operated Ca current impairs salivary gland Ϫ Ϫ changes to epithelial ciliary activity. J Cell Biol 168:869–874. fluid secretion in TRPC1( / ) mice. Proc Natl Acad Sci USA 104:17542–17547. 23. Teilmann SC, et al. (2005) Localization of transient receptor potential ion channels in 45. Yuan JP, Zeng W, Huang GN, Worley PF, Muallem S (2007) STIM1 heteromultimerizes TRPC primary and motile cilia of the female murine reproductive organs. Mol Reprod Dev channels to determine their function as store-operated channels. Nat Cell Biol 9:636–645. 71:444–452. 46. Liedtke W, Friedman JM (2003) Abnormal osmotic regulation in trpv4Ϫ/Ϫ mice. Proc 24. Arniges M, Vazquez E, Fernandez-Fernandez JM, Valverde MA (2004) Swelling- Natl Acad Sci USA 100:13698–13703. activated Ca2ϩ entry via TRPV4 channel is defective in cystic fibrosis airway epithelia. 47. Lansley AB, Sanderson MJ (1999) Regulation of airway ciliary activity by Ca2ϩ: Simul- J Biol Chem 279:54062–54068. taneous measurement of beat frequency and intracellular Ca2ϩ. Biophys J 77:629–638.

12616 ͉ www.pnas.org͞cgi͞doi͞10.1073͞pnas.0803970105 Lorenzo et al. Downloaded by guest on September 27, 2021