Stretch- and swelling-activated cation (SSAC) channels play essential roles not only in sensing and transducing external mechanical stresses but also in regulating cell volume in living cells. However, the molecular nature of the SSAC channel has not been clarified. In human epithelial HeLa cells, single-channel recordings in cell-attached and inside-out patches revealed expression of a Mg2+- and Gd3+-sensitive nonselective cation channel that is exquisitely sensitive to membrane stretch. Whole cell recordings revealed that the macroscopic cationic currents exhibit transient receptor potential (TRP) melastatin (TRPM)7-like properties such as outward rectification and sensitivity to Mg2+ and Gd3+. The whole cell cation current was augmented by osmotic cell swelling. RT-PCR and Western blotting demonstrated molecular expression of TRPM7 in HeLa cells. Treatment with small interfering RNA (siRNA) targeted against TRPM7 led to abolition of single stretch-activated cation channel currents and of swelling-activated, whole cell cation currents in HeLa cells. The silencing of TRPM7 by siRNA reduced the rate of cell volume recovery after osmotic swelling. A similar inhibition of regulatory volume decrease was also observed when extracellular Ca2+ was removed or Gd3+ was applied. It is thus concluded that TRPM7 represents the SSAC channel endogenously expressed in HeLa cells and that, by serving as a swelling-induced Ca2+ influx pathway, it plays an important role in cell volume regulation.
- regulatory volume decrease
cell volume regulation is essential for the survival and function of animal cells (23, 39, 40, 49). In epithelial cells, volume regulation after osmotic swelling occurs by water efflux mainly driven by parallel operation of volume-sensitive Cl− channels and Ca2+-activated K+ channels (13, 38). Thus an increase in the intracellular free Ca2+ concentration due to Ca2+ influx and Ca2+ release has an essential role in regulatory volume decrease (RVD) (15, 17, 23, 26, 37). Christensen (5) first suggested that the Ca2+ influx is induced by activation of a stretch-activated cation channel on osmotic cell swelling in Necturus epithelial cells, and we (37) provided direct evidence that whole cell Ca2+ currents via Gd3+-sensitive, stretch-activated cation channels are associated with osmotic swelling in human epithelial cells. However, the molecular nature of the stretch- and swelling-activated cation (SSAC) channel, which serves as a route for volume-regulatory Ca2+ influx, has not been clarified.
Since osmotic swelling has been reported to activate a number of members of the transient receptor potential (TRP) channel family such as TRP vanilloid (TRPV)4 (24, 46, 47, 51), TRPV2 (31), TRP canonical (TRPC)1 (4), and TRP melastatin (TRPM)3 (10), there is a possibility that some of these channels play a volume-regulatory role. Actually, TRPV4 was recently shown to play a role in RVD in tracheal epithelial cells (2) and keratinocyte HaCaT cells (3). It is known, however, that TRPV4 is an osmosensitive, not mechanosensitive, cation channel (46). Also, it must be noted that expression of TRPV4 may rather work against cell volume regulation in some conditions, because almost complete downregulation of volume-regulatory Cl− channels was observed in TRPV4-transfected cells (34). Thus the purpose of the present study was to identify the molecule corresponding to the SSAC channel that is endogenously expressed and plays an essential role in cell volume regulation in human epithelial HeLa cells. Here we demonstrate that TRPM7 represents the endogenous SSAC channel and is involved in RVD in HeLa cells.
MATERIALS AND METHODS
Human cervix HeLa and human embryonic kidney HEK293T cells were grown as monolayers in minimum essential medium and Dulbecco's modified Eagle's medium, respectively, supplemented with 10% fetal bovine serum, 40 U/ml penicillin G, and 100 μg/ml streptomycin under a 95% air-5% CO2 atmosphere at 37°C. For cell volume measurements and electrophysiological experiments, cells were detached from a plastic substrate and cultured in suspension with agitation for 15–300 min. HEK293T cells in six-well plates were transfected with pEGFP-N1 and pCI-neo-TRPM7 (a generous gift from Dr. Y. Mori and Dr. Y. Hara, Kyoto University) 24 h after plating, when they were partially confluent (40–60%).
Whole cell and single-channel recordings were performed at room temperature (22–26°C). The patch electrodes prepared from borosilicate glass capillaries had a resistance of ∼2 MΩ for whole cell recordings and ∼7 MΩ for single-channel recordings. Currents were recorded with an Axopatch 200B amplifier (Axon Instruments, Union City, CA). Current signals were filtered at 5 kHz with a four-pole Bessel filter and digitized at 20 kHz. pCLAMP software (version 9.0.2; Axon Instruments) was used for command pulse control, data acquisition, and analysis. To minimize K+ and anion currents, all recordings were carried out with low-Cl−, Cs+-rich intracellular solution and Cl−-free, Cs+-rich extracellular solution.
Single-channel recordings were performed in the cell-attached and excised inside-out configurations. The amplitude of single-channel currents was measured as the peak-to-peak distance in Gaussian fits of the amplitude histogram. The open probability (Po) of single channels was calculated by dividing the total time spent in the open state by the total time of continuous recording (30–200 s) in the patches containing one active channel. For cell-attached recordings, cells were exposed to bathing solution containing (in mM) 100 KCl, 2 CaCl2, 1 MgCl2, 5 Na-HEPES, 10 HEPES, 5 glucose, and 100 mannitol (pH 7.4). The pipette solution was composed of (in mM) 100 Cs-aspartate, 10 HEPES, 0.5 CsCl, and 100 mannitol (pH 7.4). For inside-out recordings, the intracellular side was perfused with a Cl−-free bath solution (300 mosmol/kgH2O) consisting of (in mM) 100 Cs-aspartate, 1 EGTA, 10 HEPES, and 100 mannitol (pH 7.4 adjusted with CsOH), and an extracellular (pipette) solution that contained (in mM) 100 Cs-aspartate, 1 EGTA, 10 HEPES, 0.5 CsCl, and 100 or 180 mannitol (pH 7.4) was used. To observe effects of intracellular or extracellular free Mg2+, 1 mM EGTA was replaced with 1 mM EDTA, and an appropriate amount of MgSO4 was added to give various concentrations (0.0001–10 mM) of free Mg2+. The free Mg2+ concentration was calculated with CaBuf software (provided by Dr. G. Droogmans, Katholieke Universiteit Leuven, Leuven, Belgium). To test the effects of mechanical stretch, patch membranes were subjected to a pulse of negative pressure applied to the back of the patch pipette. The pressure level was monitored with a manometer.
In whole cell recordings, series resistance (<5 MΩ) was compensated (to 70–80%) to minimize voltage errors. Step pulses were applied from a prepotential of +80 mV to test potentials (50-ms duration) of +80 to −80 mV in 20-mV decrements after steady currents were attained. Ramp pulses were applied every 4 s from +100 mV to −100 mV or from −100 mV to +100 mV at a speed of 4 mV/ms. Osmotic cell swelling was induced by applying hypertonic low-Cl− intracellular solution. The isotonic (300 mosmol/kgH2O) and hypertonic (370 mosmol/kgH2O) low-Cl− intracellular (pipette) solutions contained (in mM) 100 Cs-aspartate, 1 EGTA, 10 HEPES, 0.5 CsCl, and 100 or 170 mannitol, respectively (pH 7.4). In some experiments, 1 mM MgSO4 was added to the intracellular solution. The isotonic Cl−-free bath solution (320 mosmol/kgH2O) contained (in mM) 100 Cs-aspartate, 1 EGTA, 10 HEPES, and 120 mannitol (pH 7.4). When required, 0.1 mM MgSO4 was added to the bath solution. Experiments were performed after perfusion of bath solution was stopped, except when the effects of perfusion-induced mechanical stress were observed.
RNA isolation and RT-PCR.
Total cellular RNA was extracted from HeLa cells by using Sepasol RNA-I reagent (Nacalai Tesque, Kyoto, Japan) according to the protocol supplied by the manufacturer. RNA samples were reverse-transcribed at 42°C for 50 min with Maloney murine leukemia virus reverse transcriptase (Invitrogen) and oligo(dT) primers (Invitrogen), according to the manufacturer's protocols. Gene-specific primers used for PCR were designed with Primer3 software (www.genome.wi.mit.edu/cgi-bin/primer/primer3_www.cgi) and NCBI BLAST (www.ncbi.nlm.nih.gov/blast/) to identify complementary sequences in the human genome. Two sets of primers were used for TRPM7 (GenBank accession no. NM_017672). The sequences of the forward and reverse primers were 5′-AAGTGGCTTTTGCAACTTGG-3′ and 5′-ACACTCACTGCCCAGAAAGC-3′ (product size 503 bp), respectively, and 5′-CCACCCCCAATCAATATTAAACAC-3′ and 5′-GGCAGTCAAAATTTTCCTCAGT-3′ (867 bp), respectively. PCR was performed with 1.25 U of Ex Taq (Takara, Shiga, Japan) in 10× Ex Taq buffer (Takara) containing 0.2 mM dNTPs. Amplification was carried out in a thermal cycler (Gene Amp PCR System 9600, Perkin Elmer Life Sciences, Boston, MA) under the following conditions: initial heating at 94°C for 4 min, followed by 35 cycles of denaturation at 94°C for 1 min, annealing at 55°C for 1 min, and then final extension at 72°C for 1 min. Negative control experiments were performed with RNA that had not been reverse-transcribed. As a positive control, we amplified the glyceraldehyde-3-phosphate-dehydrogenase (GAPDH) sequence with a specific set of primers (5′-ACCACAGTCCATGCCATCAC-3′ and 5′-TCCACCACCACCCTGTTGCTGTA-3′). The products of RT-PCR were electrophoresed on a 2% agarose gel and, after purification with the Wizard SV Gel and PCR Clean-Up System (Promega), were cloned into pGEM-T Easy vector (Promega). Plasmids were purified with the Wizard Plus Minipreps DNA Purification System (Promega) and used as templates for sequencing with the ABI PRISM 310 genetic analyzer (Applied Biosystems, Foster City, CA).
Small interfering RNA transfection.
HeLa cells were transfected with 5 μg of small interfering RNA (siRNA) purchased from Qiagen (Chatsworth, CA) and RNAiFect Transfection Reagent (Qiagen), following the manufacturer's instructions. In brief, siRNA and RNAiFect Transfection Reagent were each diluted to 100 μl with EC-R buffer, mixed together, and incubated at room temperature for 15 min. Cells on a six-well plate were fed with 1.5 ml of fresh culture medium and overlaid with the transfection mixture. After 48–72 h, transfected cells were used for experiments. To determine transfection efficiency, we used mock siRNA and siRNA against TRPM7, which were both Alexa 488 conjugated. The mock siRNA was a commercially available nonsilencing sequence (Qiagen). The order-made TRPM7 siRNA corresponded to nucleotides 1318–1338 in the coding region and has been reported to suppress the level of TRPM7 mRNA (1). No other known genes, including those of other TRP channels, exhibit sequence homology to this target sequence. For patch-clamp experiments and single-cell size measurements, cells exhibiting Alexa 488 fluorescence were selected.
Crude membranes were prepared from control and siRNA-transfected HeLa cells as well as from TRPM7-transfected HEK293T (HEK/TRPM7) cells by Dounce homogenization followed by differential centrifugation. In brief, cells were washed twice with phosphate-buffered saline (PBS), collected by a cell scraper, and homogenized with 20 strokes of a Dounce homogenizer in 10 vols of homogenization buffer, which contained (mM) 250 sucrose, 5 EDTA, 5 EGTA, and 10 HEPES (pH 7.4). The resulting homogenate was centrifuged at 9,100 g for 10 min. The supernatant was subsequently centrifuged at 100,000 g for 1 h to pellet the microsomal fraction, and the pellet was solubilized with sample buffer containing 2% sodium dodecyl sulfate (SDS), 10% glycerol, 0.063 mM Tris·HCl (pH 6.8), 0.002% bromophenol blue, and 5% 2-mercaptoethanol. All these steps were carried out at 4°C. An aliquot of membrane protein (150 or 30 μg for HeLa or HEK293T cells, respectively) was subjected to one-dimensional SDS-PAGE using a 7% polyacrylamide gel (Daiichikagaku, Tokyo, Japan). The proteins on the gel were electrophoretically transferred to a polyvinylidene difluoride sheet, which was subsequently incubated for 1 h with a 1:1,000 dilution of an affinity-purified polyclonal rabbit antibody (a gift from Dr. Y. Mori and Dr. Y. Hara, Kyoto University) raised against a peptide corresponding to amino acids 1816–1835 of human TRPM7 (12). A purified mouse monoclonal antibody recognizing human actin (Chemicon, Temecula, CA) was used for positive control experiments. Antibody detection was performed with an Amersham biotin-streptavidin system with biotinylated anti-rabbit or anti-mouse IgG and the nitro blue tetrazolium-5-bromo-4-chloro-3-indolyl phosphate system (Amersham Life Science, Little Chalfont, UK).
Mean cell volume measurements.
Mean cell volume was measured at room temperature by electronic sizing with a Coulter-type cell size analyzer (CDA-500; Sysmex, Kobe, Japan), as previously described (13). The mean volume of the cell population was calculated from the cell volume distribution measured after the machine was calibrated with latex beads of known volume. Isotonic (310 mosmol/kgH2O) and hypotonic (200 mosmol/kgH2O) solutions contained (in mM) 95 NaCl, 4.5 KCl, 1 MgCl2, 1 CaCl2, 110 or 0 mannitol, respectively, and 5 HEPES (pH 7.3). Ca2+-free and Mg2+-free solutions were made by removing CaCl2 or MgCl2 and adding 1 mM EGTA or EDTA, respectively.
Single-cell size measurements.
For siRNA-transfected cells, application of the electronic sizing technique was found to be unsuitable because of limited transfection efficiency. Single-cell size measurements were therefore performed at room temperature, as previously reported (22). Spherical cells exhibiting Alexa 488 fluorescence were selected and held under gigaseal in the cell-attached mode by a patch pipette. Cell images were recorded through a charge-coupled device camera (C2400, Hamamatsu Photonics, Hamamatsu, Japan) and analyzed with ImageJ software (version 1.25 s; freeware at http://rsb.info.nih.gov/ij/ provided by Dr. W. Rasband, National Institutes of Health, Bethesda, MD). The relative cell size was calculated as the cube of the ratio of diameters ([d/d0]3 where d and d0 represent the cell diameter at a given time and the initial cell diameter, respectively.). The isotonic and hypotonic solutions used were the same as those used in mean cell volume measurements.
Data are presented as means ± SE of n observations. Statistical differences of the data were evaluated by paired or unpaired Student's t-test and were considered significant at P < 0.05.
Stretch-induced activation of endogenous cation channels at single-channel level.
Single-channel openings were consistently observed only when negative pressure was applied to the back of the pipette in the cell-attached mode (data not shown; n = 32) or in the inside-out mode (Fig. 1A, top and middle; 3 or 4 channels were activated in the patch). The Po observed in excised patches that contained only one channel increased with increasing negative pressure at +80 mV (Fig. 1A, bottom). The half-maximum activation pressure was ∼3 cmH2O. As shown in Fig. 1B, the single-channel current exhibited ohmic dependence on voltage (Fig. 1B, top), with a slope conductance of 22.7 ± 0.01 pS (n = 22–34) (Fig. 1B, middle) in the voltage range of −40 to +80 mV and ∼37 pS at −80 to −60 mV. The steady-state Po increased with depolarization at a constant negative pressure of 8 cmH2O (Fig. 1B, bottom).
Sensitivity of stretch-activated single-channel cation currents to Mg2+, Gd3+, and ruthenium red.
Single-channel recordings revealed concentration-dependent blocking effects of intracellular and extracellular Mg2+ on the single-channel amplitude of Cs+ current activated by membrane stretch (8 cmH2O suction), as summarized in Fig. 2, A and B. The concentration-inhibition curve for blocking by intracellular Mg2+ was little affected by voltage, and the concentration for half-maximal inhibition (IC50) values were 0.65 mM at −40 mV and 0.66 mM at +80 mV (Fig. 2A). These results are in good agreement with a value of ∼0.6 mM for Mg2+-inhibited cation (MIC) channels, which are believed to be endogenous TRPM7, in Jurkat T cells (41). The concentration-inhibition curve for blocking by extracellular Mg2+ at a negative potential of −40 mV was shifted toward lower concentrations compared with the curve at a positive potential of +80 mV (Fig. 2B), suggesting open channel block by extracellular Mg2+. The IC50 for extracellular Mg2+ was 3.2 μM at −40 mV and 0.11 mM at +80 mV, showing that block was sharply dependent on voltage. These IC50 values are comparable to those previously observed for MIC channels in Jurkat T cells (1 μM at −40 mV and 30 μM at +40 mV; Ref. 20).
The amplitude of stretch-activated single-channel currents in inside-out patches was suppressed by extracellular application of 30 μM Gd3+ in a manner independent of voltage (Fig. 2C). In contrast, 100 μM ruthenium red (RR) added to the extracellular (pipette) solution blocked the mechanosensitive channel currents in a manner dependent on voltage. Whereas the inward currents were largely inhibited by RR, the outward currents were little affected (Fig. 2C). The unitary events of stretch-activated cation channel were not affected by 100 μM amiloride, a known blocker of epithelial Na+ channel (ENaC) and some types of nonselective cation channels (data not shown; n = 5).
Abolition of stretch-activated single-cation channel currents by siRNA-induced knockdown of TRPM7.
Since sensitivity of stretch-activated cation channels in HeLa cells to Mg2+ and Gd3+ is essentially the same as that of TRPM7 to Mg2+ (32, 44) and to Gd3+ (1, 16), we next conducted RT-PCR studies to check for the expression of TRPM7. Robust amplification of PCR products of expected size (503 bp) from reverse-transcribed RNA was seen with two separate sets of TRPM7-specific primers, as shown in Fig. 3A, top left, for one set of primers. Also, no PCR product was amplified when reverse transcriptase was omitted from the reaction (data not shown). The nucleotide sequences of the PCR products obtained with TRPM7-specific primers were completely identical to the corresponding sequences in human TRPM7 (42). Treatment of HeLa cells with TRPM7 siRNA for 48–72 h eliminated expression of TRPM7 mRNA almost completely, whereas mock siRNA had no effect. Treatment not only with mock siRNA but also with TRPM7 siRNA did not affect expression of mRNA for the housekeeping gene GAPDH (Fig. 3A, top right), indicating that the siRNA-induced silencing was specific.
Western blotting of cell membrane proteins prepared from HeLa cells without (control) and with treatment with mock siRNA or TRPM7 siRNA was performed with a polyclonal anti-TRPM7 antibody, as shown in Fig. 3A, bottom. The immunoreactive bands detected in Control and Mock lanes (at arrowhead) had a molecular mass of ∼220 kDa, and this band (but not other nonspecific bands) was almost completely abolished when HeLa cells were treated with siRNA for TRPM7. In HEK/TRPM7 cells, the band reactive to anti-TRPM7 antibody was also detected at ∼220 kDa (Fig. 3A, bottom), and this band disappeared when antibodies were preabsorbed with antigen peptides (blocking peptides). In these experiments, actin bands of a molecular mass of ∼42 kDa were detected at equal levels (data not shown).
The siRNA-induced inhibition of TRPM7 expression led to the abolition of single-channel activity of the mechanosensitive cation channel in HeLa cells, as shown in Fig. 3B. In contrast, in HeLa cells treated with mock siRNA, we consistently observed typical single-channel events of the stretch-activated cation channel. We thus conclude that TRPM7 is the channel corresponding to the stretch-activated cation channel in HeLa cells.
Sensitivity of swelling-activated whole cell cation currents to Mg2+, Gd3+, and siRNA-induced knockdown of TRPM7.
As shown in Fig. 4A, top, intracellular dialysis with Mg2+- and ATP-free, Cs+-rich solution induced spontaneous activation of whole cell currents in HeLa cells. The current-voltage (I-V) relationship for steady-state currents showed strong outward rectification under ramp-clamp (Fig. 4A, trace 1). When osmotic cell swelling was induced by intracellular hypertonicity, spontaneously activated whole cell currents became larger (Fig. 4A, top). Since swelling-induced augmentation was more marked at negative potentials, the I-V relationship became less outwardly rectified and almost linear (Fig. 4A, trace 2).
As summarized in Fig. 4B, swelling-activated whole cell currents were sensitive to Mg2+, Gd3+, and RR. Mg2+ (1 mM) added to the intracellular (pipette) solution strongly blocked both outward and inward currents. In contrast, extracellular Mg2+ (0.1 mM) added to the bath solution preferentially suppressed inward currents. Swelling-activated whole cell currents were blocked in a voltage-independent manner by extracellular application of 30 μM Gd3+ but in a voltage-dependent manner by extracellular application of 100 μM RR. In contrast, 100 μM amiloride failed to affect swelling-activated whole cell cation currents (data not shown; n = 10). These data show that osmotic volume expansion of HeLa cells causes augmentation of whole cell cation currents, the properties of which are essentially the same as those of TRPM7 channel currents in regard to their sensitivity to extracellular Mg2+ (32), intracellular Mg2+ (44), and extracellular Gd3+ (1, 16). Furthermore, swelling-activated whole cell cation currents were inhibited almost completely by treatment with TRPM7 siRNA but not by treatment with mock siRNA (Fig. 4C). Thus it appears that swelling-activated whole cell cation currents are conveyed by TRPM7 endogenously expressed in HeLa cells.
Sensitivity of RVD to Mg2+, Gd3+, and siRNA-induced knockdown of TRPM7.
An increase in the intracellular free Ca2+ concentration is known to be involved in RVD in many cell types (26), including human epithelial cells (14, 38). Because a SSAC channel has been reported to serve as the volume-regulatory Ca2+ influx pathway in epithelial cells (5, 37), we tested the possible involvement of the TRPM7-like channel in RVD by measuring the mean cell volume in HeLa cells. As shown in Fig. 5A, top, control HeLa cells responded quickly to a hypotonic challenge (65% osmolality) with osmotic swelling and then with an RVD that resulted in an ∼70% recovery toward the original volume within 30 min after exposure to the hypotonic solution (containing 1 mM Ca2+ and 1 mM Mg2+). The rate of RVD was significantly diminished when swelling-activated cation channel currents were blocked by Gd3+ (30 μM) or when Ca2+ influx through the channel was prevented by removal of extracellular Ca2+ ions. In contrast, the rate of RVD was significantly facilitated when Ca2+ influx mediated by swelling-activated cation channels was augmented by elimination of extracellular Mg2+. These data are summarized in Fig. 5A, bottom. RR (100 μM) also prominently inhibited the RVD in HeLa cells (Fig. 5A, bottom).
The rate of RVD in HeLa cells treated with siRNA was assessed by single-cell size measurements using transfected cells selected based on their Alexa 488 fluorescence. The cells treated with TRPM7 siRNA exhibited a significantly slower rate of RVD than cells treated with mock siRNA, as shown in Fig. 5B. It thus appears that the TRPM7 channel is involved in the RVD process, presumably serving as the pathway for volume-regulatory Ca2+ influx in HeLa cells.
Cells can convert a mechanical stimulus into an electrical signal. The mechanotransduction mechanism, which involves a stretch-activated cation channel, exists not only in specialized mechanosensory cells such as hair cells and skin mechanoreceptor neurons but also in other cell types. Mechanosensing cation channels have essential roles in sensing touch, sound, and gravity, as well as in detecting cardiovascular tone and membrane tension during muscle contraction and cell motility or movement (8, 11, 43). The role of a stretch-activated cation channel in cell volume regulation is also well known (17, 39, 49). However, the actual molecular entity of SSAC has not yet been defined.
In prokaryotes and invertebrates, a number of molecules such as mechanosensitive channel large (MscL), Mid1, NOMPC, and members of the degenerin (DEG) family have been reported to be stretch-activated cation channels (11). No homologs of MscL or Mid1 have ever been identified in vertebrates. It has been suggested that the mammalian homolog of DEG, ENaC, serves as a stretch-activated, amiloride-sensitive cation channel in vertebrates (21). However, HeLa cells lack expression of ENaC (48), and in the present study the stretch-activated cation channel in HeLa cells was shown to be amiloride insensitive. The heterodimer of TRP polycystin (TRPP)1 and TRPP2 was proposed to be a mechanosensitive cation channel expressed in the primary cilia of mouse kidney epithelium (33). In HeLa cells, however, our preliminary studies showed that treatment with siRNA targeted against TRPP2 abolished expression of TRPP2 mRNA but failed to affect swelling-activated whole cell cation currents (T. Numata and Y. Okada, unpublished observations). Also, TRP ankyrin (TRPA)1 has recently been suggested as a candidate for the mechanosensitive transduction channel in mouse and zebra fish hair cells (6). However, the mechanosensitivity of TRPA1 has never been directly tested by electrophysiological experiments. Similarly, direct evidence is still lacking for the mechanosensitivity of TRPC6, TRPM4, and NOMPC. TRPC6 and TRPM4 were suggested to be mechanosensitive channels involved in the myogenic tone of arteries (7, 50), and NOMPC was reported to be a mechanosensory channel in Drosophila (9) and zebra fish larvae (45). Evidence that TRPC1 forms a component of a frog mechanosensitive cation channel has been obtained recently by electrophysiological and molecular biological experiments (25). However, molecular expression of TRPC1 was not observed by RT-PCR in human epithelial HeLa cells (T. Numata, unpublished observations).
TRPM7 is a member of the recently emerging TRPM subfamily of TRP proteins and has been characterized as a bifunctional protein consisting of a channel and protein kinase (29, 42). TRPM7 is expressed in a wide variety of tissues including brain and hematopoietic tissues (27), as well as in kidney and heart (30). TRPM7 has been implicated in cell proliferation and survival (12, 44), cell Mg homeostasis (44), regulation of anoxic neuronal cell death (1), and the uptake of trace metal ions (28). Oancea et al. (35) showed that shear stress upregulates TRPM7 currents by augmenting exocytotic insertion of TRPM7 into the plasma membrane in vascular smooth muscle A7R5 cells and HEK/TRPM7 cells. However, it has not been shown that the TRPM7 channel is a cation channel that can be directly activated by membrane stretch. The present study indicates that TRPM7 is the mechanosensitive cation channel endogenously expressed in human epithelial HeLa cells. First, the endogenous activity in HeLa cells of the stretch-activated cation channel that had exquisite mechanosensitivity exhibited sensitivity to Mg2+ and Gd3+ typical of TRPM7 (1, 16, 32, 44). Second, the endogenous stretch-activated single-cation channel current was abolished by treatment of HeLa cells with siRNA targeted against TRPM7.
In epithelial cells, the stretch-activated cation channel was shown to be activated by osmotic cell swelling, thereby serving as a route of volume-regulatory Ca2+ influx (5, 37). However, the molecular identity of the SSAC channel has not been known. The present study demonstrated that swelling-activated whole cell cation currents exhibit sensitivity to Mg2+, Gd3+, and treatment with TRPM7 siRNA in HeLa cells. Also, all these properties of swelling-activated whole cell current were found, in the present study, to match well with those of stretch-activated single-channel current in HeLa cells. Thus it appears that TRPM7 represents the SSAC channel in HeLa cells.
In the present study, it was found that RVD in HeLa cells was largely inhibited by extracellular Ca2+ removal and extracellular application of Gd3+. Also, it was found that RVD was facilitated by extracellular Mg2+ removal. These effects could be explained well by the suppression or augmentation, resulting from these manipulations, of Ca2+ influx mediated by TRPM7 or MIC channels, which are known to be permeable to Ca2+ (1, 12, 19, 28, 32, 42). It was found, in fact, that siRNA-induced knockdown of the TRPM7 channel largely inhibited RVD. Thus it appears that TRPM7-mediated Ca2+ influx plays an essential role in RVD, which is known to require both Ca2+ release and Ca2+ influx. Also, there is a possibility that TRPM7-mediated Mg2+ influx might somehow affect the RVD process.
In conclusion, TRPM7 represents the SSAC channel endogenously expressed in human epithelial HeLa cells, and it is essentially involved in RVD by serving as a swelling-induced, volume-regulatory Ca2+ influx pathway.
This work was supported by Grants-in-Aid for Scientific Research from the Ministry of Education, Science, Sports and Culture of Japan (MEXT) and from the Japan Society for the Promotion of Science (JSPS) and by the Foundation for Promotion of Material Science and Technology of Japan (MST Foundation).
We are grateful to E. L. Lee for discussion and for reading the manuscript, to K. Shigemoto and M. Ohara for technical assistance, and to T. Okayasu for secretarial assistance. We also thank Y. Mori and Y. Hara for helpful suggestions and for providing us with the human TRPM7 cDNA and anti-TRPM7 antibody.
The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
- Copyright © 2007 the American Physiological Society