Regulation of the Ca2+ Sensitivity of the Nonselective Cation Channel TRPM4
2004; Elsevier BV; Volume: 280; Issue: 8 Linguagem: Inglês
10.1074/jbc.m411089200
ISSN1083-351X
AutoresBernd Nilius, Jean Prenen, Jisen Tang, Chunbo Wang, Grzegorz Owsianik, Annelies Janssens, Thomas Voets, Michael X. Zhu,
Tópico(s)Nicotinic Acetylcholine Receptors Study
ResumoTRPM4, a Ca2+-activated cation channel of the transient receptor potential superfamily, undergoes a fast desensitization to Ca2+. The mechanisms underlying the alterations in Ca2+ sensitivity are unknown. Here we show that cytoplasmic ATP reversed Ca2+ sensitivity after desensitization, whereas mutations to putative ATP binding sites resulted in faster and more complete desensitization. Phorbol ester-induced activation of protein kinase C (PKC) increased the Ca2+ sensitivity of wild-type TRPM4 but not of two mutants mutated at putative PKC phosphorylation sites. Overexpression of a calmodulin mutant unable to bind Ca2+ dramatically reduced TRPM4 activation. We identified five Ca2+-calmodulin binding sites in TRPM4 and showed that deletion of any of the three C-terminal sites strongly impaired current activation by reducing Ca2+ sensitivity and shifting the voltage dependence of activation to very positive potentials. Thus, the Ca2+ sensitivity of TRPM4 is regulated by ATP, PKC-dependent phosphorylation, and calmodulin binding at the C terminus. TRPM4, a Ca2+-activated cation channel of the transient receptor potential superfamily, undergoes a fast desensitization to Ca2+. The mechanisms underlying the alterations in Ca2+ sensitivity are unknown. Here we show that cytoplasmic ATP reversed Ca2+ sensitivity after desensitization, whereas mutations to putative ATP binding sites resulted in faster and more complete desensitization. Phorbol ester-induced activation of protein kinase C (PKC) increased the Ca2+ sensitivity of wild-type TRPM4 but not of two mutants mutated at putative PKC phosphorylation sites. Overexpression of a calmodulin mutant unable to bind Ca2+ dramatically reduced TRPM4 activation. We identified five Ca2+-calmodulin binding sites in TRPM4 and showed that deletion of any of the three C-terminal sites strongly impaired current activation by reducing Ca2+ sensitivity and shifting the voltage dependence of activation to very positive potentials. Thus, the Ca2+ sensitivity of TRPM4 is regulated by ATP, PKC-dependent phosphorylation, and calmodulin binding at the C terminus. TRPM4 1The abbreviations used are: TRPM, transient receptor potential channel, melastatin subfamily; CaM, calmodulin; GFP, green fluorescent protein; MBP, maltose-binding protein; PKC, protein kinase C; PMA, phorbol 12-myristate 13-acetate; SK2, type 2 small conductance Ca2+-activated K+ channel; MOPS, 4-morpholinepropanesulfonic acid. is a Ca2+-activated and voltage-dependent Ca2+-impermeable cation channel with a unitary conductance of 25 picosiemens that belongs to the melastatin subfamily of transient receptor potential membrane proteins (1Launay P. Fleig A. Perraud A.L. Scharenberg A.M. Penner R. Kinet J.P. Cell. 2002; 109: 397-407Abstract Full Text Full Text PDF PubMed Scopus (543) Google Scholar, 2Nilius B. Prenen J. Voets T. Droogmans G. J. Physiol. (Lond.). 2004; 560: 753-765Crossref Scopus (92) Google Scholar, 3Nilius B. Prenen J. Voets T. Droogmans G. Pfluegers Arch. 2004; 448: 70-75Crossref PubMed Scopus (113) Google Scholar, 4Nilius B. Prenen J. Droogmans G. Voets T. Vennekens R. Freichel M. Wissenbach U. Flockerzi V. J. Biol. Chem. 2003; 278: 30813-30820Abstract Full Text Full Text PDF PubMed Scopus (267) Google Scholar). Ca2+-activated, Ca2+-impermeable nonselective cation channels that share functional properties with expressed TRPM4 (or the closest homologue, TRPM5) have been found in many excitable and non-excitable cells (Refs. 5Colquhoun D. Neher E. Reuter H. Stevens C.F. Nature. 1981; 294: 752-754Crossref PubMed Scopus (431) Google Scholar, 6Siemen D. Reuhl T. Pfluegers Arch. 1987; 408: 534-536Crossref PubMed Scopus (20) Google Scholar, 7Liman E.R. J. Physiol. (Lond.). 2003; 548: 777-787Crossref Scopus (69) Google Scholar, 8Ullrich N.D. Voets T. Prenen J. Vennekens R. Talavera K. Droogmans G. Nilius B. Cell Calcium. 2005; (in press)PubMed Google Scholar; for reviews, see Refs. 9Siemen W. Hescheler J. Nonselective Cation Channels: Pharmacology, Physiology & Biophysics. Birkha ̈user Verlag, Basel, Switzerland1994Google Scholar and 10Petersen O.H. Fedirko N.V. Curr. Biol. 2001; 11: R520-R523Abstract Full Text Full Text PDF PubMed Scopus (24) Google Scholar). These nonselective channels may regulate important processes including cardiac rhythmicity and neural bursting activity, and their Ca2+-dependent activation has been suggested to form a general feedback control mechanism for Ca2+ influx in non-excitable cells. The functional analysis of TRPM4 and TRPM5 and their comparison with native nonselective cation channels are complicated by a peculiar property of these channels: when activated by an increase in free intracellular Ca2+ concentration ([Ca2+]i), the currents decay rapidly due to a decrease in the sensitivity of the channels to Ca2+ (1Launay P. Fleig A. Perraud A.L. Scharenberg A.M. Penner R. Kinet J.P. Cell. 2002; 109: 397-407Abstract Full Text Full Text PDF PubMed Scopus (543) Google Scholar, 3Nilius B. Prenen J. Voets T. Droogmans G. Pfluegers Arch. 2004; 448: 70-75Crossref PubMed Scopus (113) Google Scholar, 4Nilius B. Prenen J. Droogmans G. Voets T. Vennekens R. Freichel M. Wissenbach U. Flockerzi V. J. Biol. Chem. 2003; 278: 30813-30820Abstract Full Text Full Text PDF PubMed Scopus (267) Google Scholar, 11Hofmann T. Chubanov V. Gudermann T. Montell C. Curr. Biol. 2003; 13: 1153-1158Abstract Full Text Full Text PDF PubMed Scopus (324) Google Scholar). Moreover, recent studies on TRPM4 exhibit an unusually large variability in reported values for Ca2+ sensitivity and activation time courses (1Launay P. Fleig A. Perraud A.L. Scharenberg A.M. Penner R. Kinet J.P. Cell. 2002; 109: 397-407Abstract Full Text Full Text PDF PubMed Scopus (543) Google Scholar, 2Nilius B. Prenen J. Voets T. Droogmans G. J. Physiol. (Lond.). 2004; 560: 753-765Crossref Scopus (92) Google Scholar, 3Nilius B. Prenen J. Voets T. Droogmans G. Pfluegers Arch. 2004; 448: 70-75Crossref PubMed Scopus (113) Google Scholar, 4Nilius B. Prenen J. Droogmans G. Voets T. Vennekens R. Freichel M. Wissenbach U. Flockerzi V. J. Biol. Chem. 2003; 278: 30813-30820Abstract Full Text Full Text PDF PubMed Scopus (267) Google Scholar). Most likely, these discrepancies reflect a highly regulated Ca2+ affinity of TRPM4, which may be of physiological relevance. Another puzzling property of TRPM4 is its sensitivity to ATP. We have shown recently that cytosolic ATP4– acts as a potent inhibitor of TRPM4 currents in inside-out patches, with half-maximal inhibition at ∼2 μm (3Nilius B. Prenen J. Voets T. Droogmans G. Pfluegers Arch. 2004; 448: 70-75Crossref PubMed Scopus (113) Google Scholar). However, robust TRPM4 currents can be measured in the whole-cell mode, even under conditions in which the free cytosolic ATP4– concentration exceeds 100 μm. One possible explanation of this apparent paradox could be that ATP has both an inhibitory and a stimulatory effect on TRPM4, but experimental data to support this idea are currently lacking. In the present study, we investigated potential cellular factors that influence Ca2+-dependent activation of TRPM4. We report that the Ca2+ sensitivity of TRPM4 is regulated by cytosolic ATP, protein kinase C (PKC)-dependent phosphorylation, and calmodulin (CaM), and we identify distinct residues in the channel protein that are crucial for the modulation by these cellular factors. Cell Culture, Expression of TRPM4, and Mutagenesis—Human embryonic kidney HEK293 cells were grown in Dulbecco's modified Eagle's medium containing 10% (v/v) human serum, 2 mml-glutamine, 2 units/ml penicillin, and 2 mg/ml streptomycin at 37 °C in a humidity-controlled incubator with 10% CO2. For transient expression, we used the recombinant bicistronic expression plasmid pdiTRPM4b, which carries the entire protein-coding region for human TRPM4b (GenBank™ accession number AX443227; kindly provided by Drs. V. Flockerzi and U. Wissenbach) and for the green fluorescent protein (GFP) coupled by an internal ribosomal entry site sequence. HEK293 cells were transiently transfected with the pdiTRPM4b vector using methods described previously, and successfully transfected cells were visually identified by their green fluorescence in the patch clamp set up (for details, see ref. 4Nilius B. Prenen J. Droogmans G. Voets T. Vennekens R. Freichel M. Wissenbach U. Flockerzi V. J. Biol. Chem. 2003; 278: 30813-30820Abstract Full Text Full Text PDF PubMed Scopus (267) Google Scholar). For experiments with the dominant negative CaM mutant, CaM1,2,3,4, we used tetracycline-inducible HEK293 cells expressing FLAG-TRPM4, which were cultured at 37 °C with 5% CO2 in Dulbecco's modified Eagle's medium supplemented with 10% fetal bovine serum and 2 mm glutamine, as described previously (1Launay P. Fleig A. Perraud A.L. Scharenberg A.M. Penner R. Kinet J.P. Cell. 2002; 109: 397-407Abstract Full Text Full Text PDF PubMed Scopus (543) Google Scholar). The medium was supplemented with blasticidin (5 μg/ml; Invitrogen) and zeocin (0.4 mg/ml; Invitrogen). For all experiments, cells were treated with 1 μg/ml tetracycline (Invitrogen) for 16–24 h and used after 16–20 h. Cells were kindly provided by Drs. J.-P. Kinnet and P. Launay (Department of Pathology, Harvard Medical School, Boston, MA). The CaM1,2,3,4 construct with the mutations of all four EF-hand Ca2+ binding sites was kindly provided by Dr. J. Adelman (Vollum Institute, Oregon Health & Science University, Portland, OR). Single amino acid mutations were done by using the standard PCR overlap extension technique (18Prawitt D. Monteilh-Zoller M.K. Brixel L. Spangenberg C. Zabel B. Fleig A. Penner R. Proc. Natl. Acad. Sci. U. S. A. 2003; 100: 15166-15171Crossref PubMed Scopus (289) Google Scholar). A chimera containing the first intracellular linker of TRPM5 in the TRPM4 backbone was obtained using the overlap extension technique with human TrpM4 cDNA and mouse TrpM5 cDNA as templates; both were constructed in the pCAGGSM2/IresGFP vector (12Ho S.N. Hunt H.D. Horton R.M. Pullen J.K. Pease L.R. Gene (Amst.). 1989; 77: 51-59Crossref PubMed Scopus (6833) Google Scholar). The chimerical overlap PCR fragments were replaced in pCAGGSM2/TrpM4/IresGFP using AscI/ClaI restriction enzymes. The nucleotide sequences of all mutants were verified by sequencing of the corresponding cDNAs. Fig. 1 provides an overview of the mutants and constructs used in this study. CaM Binding Assay—Constructs for TRPM4 fragments fused to the C terminus of maltose-binding protein (MBP), in vitro synthesis of 35S-labeled MBP fusion proteins, and assay conditions for their binding to CaM were essentially the same as those described previously (13Zhang Z. Tang J. Tikunova S. Johnson J.D. Chen Z. Qin N. Dietrich A. Stefani E. Birnbaumer L. Zhu M.X. Proc. Natl. Acad. Sci. U. S. A. 2001; 98: 3168-3173Crossref PubMed Scopus (208) Google Scholar, 14Tang J. Lin Y. Zhang Z. Tikunova S. Birnbaumer L. Zhu M.X. J. Biol. Chem. 2001; 276: 21303-21310Abstract Full Text Full Text PDF PubMed Scopus (268) Google Scholar). Cell Surface Biotinylation Assay—In order to facilitate immunodetection of the expressed proteins, coding regions of TRPM4 and its C-terminal CaM site mutants were subcloned into pEGFP-C3 (Clontech). The resulting plasmids express TRPM4 and its mutants with GFP fused at the N termini. Transfection, biotinylation, and streptavidin precipitation were performed as described previously (15Wang C. Hu H.Z. Colton C.K. Wood J.D. Zhu M.X. J. Biol. Chem. 2004; 279: 37423-37430Abstract Full Text Full Text PDF PubMed Scopus (94) Google Scholar). Immunoblotting was performed using anti-GFP antibodies (Clontech). Sample loading for the crude cell lysate and streptavidin-precipitated portion represents 12 and 355 μg, respectively, of total proteins in cell lysates. Identical exposure time was used to reveal the chemiluminescent signals for the GFP fusion proteins in crude lysates and streptavidin-precipitated samples. Solutions—The extracellular solution for cell attached measurements and the pipette solution for inside-out patch clamp measurements contained 156 mm NaCl, 5 mm CaCl2,10mm glucose, and 10 mm HEPES, buffered at pH 7.4 with NaOH. Before patch excision, the extracellular bath solution was changed to an "internal solution" for inside-out patch clamp, which contained 156 mm CsCl, 1 mm MgCl2, and 10 mm HEPES, pH 7.2 with CsOH. The Ca2+ concentration at the inner side of the membrane was adjusted between 100 nm and 10 μm by adding the appropriate amounts of CaCl2 calculated by the CaBuf program (ftp://ftp.cc.kuleuven.ac.be/pub/droogmans/cabuf.zip) to the 10 mm EGTA-containing solution. For Ca2+ concentrations between 100 μm and 10 mm, CaCl2 was added to an EGTA-free solution. In ATP-containing pipette solutions, we always used 4 mm Na2ATP. Mg2+ was added to obtain a free Mg2+ concentration of 500 μm, in which case the Mg-ATP concentration amounts to 3.6 mm. The extracellular solution for whole-cell measurements contained 156 mm NaCl, 5 mm CaCl2,10mm glucose, and 10 mm HEPES, buffered at pH 7.4 with NaOH. The pipette solution contained 156 mm CsCl, 1 mm MgCl2,10mm HEPES, and 10 mm EGTA, pH 7.2 with CsOH. Ca2+ concentration was adjusted between 100 nm and 100 μm by adding appropriate amounts of CaCl2. Recombinant CaM was added to the internal solution (bath solution) in inside-out patch clamp experiments at a concentration of 10 μm (diluted from a stock solution of 0.5 mm in 10 mm MOPS, pH 7.5, and 5 mm KCl). Electrophysiology—Currents were monitored with an EPC-9 (HEKA Elektronik, Lambrecht, Germany). Patch electrodes had a DC resistance between 2 and 4 megaohms. An Ag-AgCl wire was used as a reference electrode. For whole-cell measurements, we used 400-ms ramps from –150 to +100 mV (sample interval, 800 μs; 512 points; 2-s interval between ramps; holding potential, 0 mV; series resistance compensation, 60–80%). For inside-out patches, the sampling interval was 500 μs, the interval between steps was 2 s (2048 points; filter setting, 1 kHz). Step protocols for single channel measurements were applied from holding potentials of 0 mV and consisted of 400-ms steps to –100 mV followed by a step to +100 mV. The interval between the pulses was 5 s. Experiments were performed at room temperature (22–25 °C). Data Analysis—Electrophysiological data were analyzed using WinASCD software (G. Droogmans, Leuven, Belgium). Pooled data are given as the mean ± S.E. of n cells. Significance was tested using Student's unpaired t test (p < 0.05). ATP Restores the Ca2+ Sensitivity of TRPM4 after Desensitization—In all patch-clamp configurations, an increase in [Ca2+]i induced a fast activating, transient current in HEK293 cells that expressed TRPM4. Fig. 2A shows a typical time course of the whole-cell current activated by dialysis of a pipette solution that contained 10 μm [Ca2+]i but no ATP. The I-V curves obtained from the voltage ramps, measured at the times indicated by the solid points, display the typical rectification behavior of TRPM4 currents (1Launay P. Fleig A. Perraud A.L. Scharenberg A.M. Penner R. Kinet J.P. Cell. 2002; 109: 397-407Abstract Full Text Full Text PDF PubMed Scopus (543) Google Scholar, 4Nilius B. Prenen J. Droogmans G. Voets T. Vennekens R. Freichel M. Wissenbach U. Flockerzi V. J. Biol. Chem. 2003; 278: 30813-30820Abstract Full Text Full Text PDF PubMed Scopus (267) Google Scholar). The current was almost completely abolished within 2 min of cell dialysis (Fig. 2, A and B). The same results were obtained when dialyzing the cells with 4 mm ATP and 0.5 mm free Mg2+ (Fig. 2B). In the inside-out configuration, currents measured in the absence of ATP also decayed rapidly to a stationary level. We have shown previously that this decay can be explained by an increased KdCa for Ca2+ binding to the channel (2Nilius B. Prenen J. Voets T. Droogmans G. J. Physiol. (Lond.). 2004; 560: 753-765Crossref Scopus (92) Google Scholar). Exposure to a Ca2+-free solution completely abolished the currents. Subsequent application of Ca2+ restored the currents to a level that was nearly identical to that before exposure of the patch to the Ca2+-free solution (Fig. 2C). However, this recovery, as estimated from the current amplitude during reapplication of Ca2+ in the absence of ATP, was nearly complete if the Ca2+-free solution contained Mg-ATP. The latter currents were again followed by desensitization (Fig. 2, D and E). Thus, ATP is able to restore Ca2+ sensitivity after desensitization. We have shown previously that free ATP4–, but not Mg-ATP, is an efficient blocker of TRPM4 channels (3Nilius B. Prenen J. Voets T. Droogmans G. Pfluegers Arch. 2004; 448: 70-75Crossref PubMed Scopus (113) Google Scholar). Therefore, ATP has a dual effect on TRPM4: its ionized form blocks the channel, whereas binding of ATP might have an effect on stabilizing or restoring the Ca2+ sensitivity of the channel. We have therefore explored the latter hypothesis in more detail. Direct Binding of ATP Is Important for Maintaining the Ca2+ Sensitivity of TRPM4—Multiple ATP binding sites, including Walker B motifs and ABC transporter signature motifs (16Orelle C. Dalmas O. Gros P. Di Pietro A. Jault J.M. J. Biol. Chem. 2003; 278: 47002-47008Abstract Full Text Full Text PDF PubMed Scopus (137) Google Scholar, 17Ren X.Q. Furukawa T. Haraguchi M. Sumizawa T. Aoki S. Kobayashi M. Akiyama S. Mol. Pharmacol. 2004; 65: 1536-1542Crossref PubMed Scopus (32) Google Scholar), can be predicted from the TRPM4 amino acid sequence (Fig. 1). In the N terminus, 231AFFLVD (WB1) and 273VLLL-LID (WB2) represent two Walker B motifs. An ABC transporter signature motif, 322GSSGGARQGEARDR, is located in the vicinity of WB1 and WB2. Two other predicted Walker B motifs are present in the first intracellular loop between TM2 and TM3, 800VLLVD (WB3) and 858LYLAD (WB4). They are again coupled to an ABC transporter motif, 833LSGGGGSLASSGG. Mutations of the WB2 site resulted in an extremely fast and more complete decay of the currents in inside-out patches, indicating that desensitization is accelerated. An example for the L275C mutation located in WB2 is given in Fig. 3A. The time course and the extent of the decay were quantified by the time constant (τ) and the apparent open probability in the "stationary" phase (Popen,ss), respectively. In comparison with the wild-type channel, τ was reduced by 50%, whereas Popen,ss was reduced by 95% for L275C (Fig. 3, A and B). The same result was obtained for L275A (Fig. 3B). Mutation of a neighboring negatively charged residue (Asp279) or a nonpolar residue (Ile278) to asparagine (D279N or I278N) did not affect the rate of decay or the stationary value (Fig. 3B). These results indicate that the loss of the WB2 ATP binding site causes a more pronounced desensitization. Note that it was difficult to quantify the effects in many of these mutants (whole-cell as well as insideout conditions) due to the extremely fast current decay (∼10 s). In fact, whole-cell currents were hardly measurable. We have therefore focused experiments on inside-out patches. Next, we mutated the first ABC transporter signature motif, which is located in the vicinity of WB2. A single amino acid substitution in this region, G325A, again accelerated the decay and completely abolished currents within ∼10 s (Fig. 3, C and D). Mutation of neighboring residues, G324A and R327A, had no significant effect. To investigate the role of WB3 and WB4, which are co-localized with the second ABC transporter signature motif in the first intracellular linker between TM2 and TM3, we exchanged the TM2-TM3 of TRPM4 with that of TRPM5, which lacks these motifs. Note that TRPM5 is not sensitive to ATP and is not blocked by ATP4– (8Ullrich N.D. Voets T. Prenen J. Vennekens R. Talavera K. Droogmans G. Nilius B. Cell Calcium. 2005; (in press)PubMed Google Scholar). Fig. 3E shows that for wild-type TRPM4, the currents decayed to a stationary value after patch excision in 100 μm Ca2+. This decay is faster and more complete at negative potentials. The TRPM4/5 chimera that contained the TM2-TM3 linker of TRPM5 displayed an extremely fast and complete decay (Fig. 3F). Data are summarized in Fig. 3G. Thus, exchanging the TM2-TM3 linker of TRPM4 with a sequence that does not contain the Walker B and ABC motifs resulted in a highly significant acceleration of the decay, indicating a much faster loss of Ca2+ sensitivity (2Nilius B. Prenen J. Voets T. Droogmans G. J. Physiol. (Lond.). 2004; 560: 753-765Crossref Scopus (92) Google Scholar). Obviously, all mutations predicted to affect ATP binding to the channel caused a very fast and complete decay of the Ca2+-activated currents of TRPM4. Interestingly, as shown in Fig. 3, H and I, all mutants and the TRPM4/5 chimera showed a strongly reduced recovery after Ca2+-free perfusion with 4 mm ATP and 0.5 mm free Mg2+ as compared with that of the wild type shown in Fig. 2D. PKC Phosphorylation Enhances the Ca2+ Sensitivity of TRPM4—The experiments described above indicate that ATP binding might be essential for channel functioning. Next, we tested whether phosphorylation could affect channel activation. TRPM4 contains multiple putative PKC phosphorylation sites. The predicted sites with the highest scores (NetPhos 2.0; www.cbs.dtu.dk/services/NetPhos/) were Ser1145 and Ser1152 in the C terminus and Thr68 and Thr356 in the N terminus (Fig. 1). First, we tested whether phosphorylation could modify channel activation. For this experiment, we dialyzed the cells with 1 μm Ca2+, which only caused activation of whole-cell currents through TRPM4 channels in a small fraction of cells both in the absence and presence of 4 mm ATP (3 of 12 cells). Fig. 4A shows an example of a responding cell, in which current activation was induced by 1 μm [Ca2+]i in the absence of ATP. The activated currents are small and lack the typical outward rectification pattern of TRPM4 currents. Fig. 4D shows the averaged current amplitudes. Next, we tested the role of PKC by applying phorbol 12-myristate 13-acetate (PMA). Preincubation with 1 μm PMA for 1 h significantly increased the incidence of TRPM4 current activation by 1 μm Ca2+ (7 of 10 cells). The whole-cell current densities resulting from activation by 1 μm Ca2+ were also significantly increased by the PMA pretreatment (Fig. 4, B and D). The EC50 value for Ca2+, as measured from the peak currents in the presence of 4 mm ATP and at various Ca2+ concentrations, was 15 μm in control cells (Fig. 4C) compared with 4 μm in cells preincubated with PMA (Fig. 4, B and C). The effect of PMA was absent in cells expressing the mutants S1145A or S1152A (Fig. 4, C and D). These findings indicate that PKC-dependent phosphorylation increases the Ca2+ sensitivity of TRPM4. The nearly negligible effect of the presence and absence of ATP in the patch pipette may hint at the non-complete washout of ATP during ATP-free dialysis during the fast first peak of current activation. CaM Is Involved in Conferring the Ca2+ Sensitivity of TRPM4—Next, we tried to find the molecular determinant(s) that transfers Ca2+ sensitivity to TRPM4. We reasoned that either Ca2+ could bind and activate the channels directly or the activation could proceed via Ca2+ binding to an accessory Ca2+-binding protein such as CaM. To validate the second possibility, we have first studied on whole-cell currents the effect of overexpression of a dominant negative CaM mutant, CaM1,2,3,4, in which all four EF-hand Ca2+ binding sites are mutated. These experiments were performed in an inducible, hTRPM4-expressing HEK293 cell line (1Launay P. Fleig A. Perraud A.L. Scharenberg A.M. Penner R. Kinet J.P. Cell. 2002; 109: 397-407Abstract Full Text Full Text PDF PubMed Scopus (543) Google Scholar, 3Nilius B. Prenen J. Voets T. Droogmans G. Pfluegers Arch. 2004; 448: 70-75Crossref PubMed Scopus (113) Google Scholar). Transient expression of CaM1,2,3,4 in these cells resulted in a drastic decrease of the current amplitude (Fig. 5, A and B), although some TRPM4 activity was still detectable (Fig. 5C). These data suggest that CaM may be involved in regulating TRPM4 activation. In a second approach, we added recombinant CaM to the Ca2+-containing solution in which the inside-out patches were excised (100 μm Ca2+ and 10 μm CaM). Desensitization was also seen in the presence of CaM (Fig. 5, D and E). However, the ratio between inward current at –100 mV and +100 mV was increased, and current decay was less pronounced than in the absence of CaM (Fig. 5, F–H). These data indicate that in cell-free patches, CaM reduced desensitization of TRPM4. We then searched for CaM-binding sites on TRPM4 using in vitro binding assays. Fragments of TRPM4 were fused to MBP and prepared by in vitro synthesis using transcription and translation-coupled rabbit reticulocyte lysates in the presence of [35S]Met and [35S]Cys. The 35S-labeled proteins were incubated with CaM-Sepharose in a binding solution that contained 1 mm Ca2+. The MBP fusion proteins retained by CaM-Sepharose were considered positive and were further divided into smaller segments until the minimal binding region was obtained. The diagram in Fig. 6A summarizes the results, with the filled, hatched, and open bars indicating positive, weakly positive, and negative binding to Ca2+-CaM, respectively. Examples of the binding results are shown in Fig. 6B. Semiquantitatively, the darker the band, the stronger the binding of the fragment to CaM. Because many fragments showed positive binding to Ca2+-CaM in the initial screening, we have focused on the regions that displayed the relatively strongest binding. These experiments led to the identification of five CaM-binding sites on TRPM4: two sites located in the N terminus (E5/V25 and M186/R212), and three clustered in the C terminus (A1076/S1098, R1095/T1151, and E1146/E1176). Among them, fragment v (R1095/T1151) is rather unique in that it is much longer than most commonly found CaM-binding domains, which are typically α-helices of about 20–30 amino acids. Shorter fragments were made from R1095/T1151, but they all failed to show a strong binding to Ca2+-CaM (Fig. 6B, right panels). Nevertheless, the N-terminal half, R1095/E1130 (fragment w), retained partial binding to CaM and was considered weakly positive. With the exception of the A1076/S1098 fragment, binding of CaM only occurred in the presence of Ca2+ (Fig. 6C). A1076/S1098 also binds CaM in the absence of Ca2+, but binding is strongly enhanced by addition of Ca2+. Next, we tested whether deletion of the CaM-binding sites affects TRPM4 activity. For both N-terminal deletion mutants (ΔE5/V25 and ΔM186/R211), whole-cell currents upon dialysis of 10 μm Ca2+ showed a similar activation time course as wild-type TRPM4. The shape of the I-V curves was also similar to that of the wild-type channels, and current densities were not significantly different (wild type, 165 ± 47 pA/picofarad, n = 6; ΔE5/V25, 120 ± 29 pA/picofarad, n = 6; ΔM186/R211, 99 ± 39 pA/picofarad, n = 5). In inside-out patches, the kinetics and Ca2+ sensitivity of the ΔE5/V25 and ΔM186/R211 mutants were also similar (data not shown). In contrast, deletion of the C-terminal CaM-binding sites strongly affected TRPM4 channel function (Fig. 7). Truncation of the C terminus after Ala1068 (A1068X), which eliminates all three sites, reduced the amplitude of Ca2+-activated currents to that of non-transfected cells. Truncation of the C terminus after Glu1146 (E1146X), which only eliminated the last CaM-binding site, reduced current amplitudes by ∼90% compared with wild-type TRPM4, without altering the biophysical properties of the channel (Fig. 7, A and B). Disruption of individual sites (ΔA1076/S1098, ΔR1095/E1130, or ΔV1127/T1151) resulted in very small currents (Fig. 7C). The ΔE1146/E1176 deletion mutant had similar currents as the E1146X truncation (data not shown). The double mutant ΔA1076/S1098-R1095/E1130 was completely silent. The most straightforward explanation of these results is that CaM is involved in Ca2+ sensing of TRPM4. However, because the mutant channels were still activated by Ca2+, albeit to a lesser extent, it cannot be excluded that Ca2+ activates TRPM4 directly and that CaM only plays a modulatory role. To test this possibility, we stimulated excised cell-free patches from all CaM mutants with Ca2+. Although the whole-cell currents were not resolvable from the background, we were still able to activate these TRPM4 deletion mutants in excised patches, albeit only at extremely positive potentials and very high Ca2+ concentrations. Fig. 7, D and E, shows an example for the deletion mutant ΔE1146/E1176. Currents were activated in inside-out patches by 1000 μm Ca2+. In control patches, a very large current can be measured at +100 mV. For all deletion mutants of the three identified Ca2+-CaM binding sites, however, the current at +100 mV was very small, and larger currents could only be activated at more positive potentials. This pattern was measured in all three CaM-deletion mutants (n = 6 for ΔA1076/S1098, ΔR1095/E1130, and ΔE1146/E1167) but absent in non-transfected cells (n = 7). Also, among all C-terminal mutants, the largest currents were measured with the most distal ΔE1146/E1176 mutant in the inside-out patches, consistent with the whole-cell data. However, these currents show a similar pattern as described for the deletion of putative ATP binding site, namely, an accelerated and complete decay (data not shown; five patches). All these findings indicate that Ca2+ can still activate TRPM4, but its sensitivity is dramatically reduced by mutating the CaM-binding site or by structural changes in the C-terminal part of the channel that confers CaM binding. In order to confirm that the mutant TRPM4 proteins were expressed on the plasma membrane, we performed a cell surface biotinylation assay using GFP-tagged TRPM4 constructs. As shown in Fig. 7F, simi
Referência(s)