Syntaxin 1A Binds to the Cytoplasmic C Terminus of Kv2.1 to Regulate Channel Gating and Trafficking
2003; Elsevier BV; Volume: 278; Issue: 19 Linguagem: Inglês
10.1074/jbc.m213088200
ISSN1083-351X
AutoresYuk‐Man Leung, Youhou Kang, Xiao‐Dong Gao, Fuzhen Xia, Huanli Xie, Laura Sheu, Sharon Tsuk, Ilana Lotan, Robert G. Tsushima, Herbert Y. Gaisano,
Tópico(s)Erythrocyte Function and Pathophysiology
ResumoVoltage-gated K+(Kv) 2.1 is the dominant Kv channel that controls membrane repolarization in rat islet औ-cells and downstream insulin exocytosis. We recently showed that exocytotic SNARE protein SNAP-25 directly binds and modulates rat islet औ-cell Kv 2.1 channel protein at the cytoplasmic N terminus. We now show that SNARE protein syntaxin 1A (Syn-1A) binds and modulates rat islet औ-cell Kv2.1 at its cytoplasmic C terminus (Kv2.1C). In HEK293 cells overexpressing Kv2.1, we observed identical effects of channel inhibition by dialyzed GST-Syn-1A, which could be blocked by Kv2.1C domain proteins (C1: amino acids 412–633, C2: amino acids 634–853), but not the Kv2.1 cytoplasmic N terminus (amino acids 1–182). This was confirmed by direct binding of GST-Syn-1A to the Kv2.1C1 and C2 domains proteins. These findings are in contrast to our recent report showing that Syn-1A binds and modulates the cytoplasmic N terminus of neuronal Kv1.1 and not by its C terminus. Co-expression of Syn-1A in Kv2.1-expressing HEK293 cells inhibited Kv2.1 surfacing, which caused a reduction of Kv2.1 current density. In addition, Syn-1A caused a slowing of Kv2.1 current activation and reduction in the slope factor of steady-state inactivation, but had no affect on inactivation kinetics or voltage dependence of activation. Taken together, SNAP-25 and Syn-1A mediate secretion not only through its participation in the exocytotic SNARE complex, but also by regulating membrane potential and calcium entry through their interaction with Kv and Ca2+ channels. In contrast to Ca2+ channels, where these SNARE proteins act on a common synprint site, the SNARE proteins act not only on distinct sites within a Kv channel, but also on distinct sites between different Kv channel families. Voltage-gated K+(Kv) 2.1 is the dominant Kv channel that controls membrane repolarization in rat islet औ-cells and downstream insulin exocytosis. We recently showed that exocytotic SNARE protein SNAP-25 directly binds and modulates rat islet औ-cell Kv 2.1 channel protein at the cytoplasmic N terminus. We now show that SNARE protein syntaxin 1A (Syn-1A) binds and modulates rat islet औ-cell Kv2.1 at its cytoplasmic C terminus (Kv2.1C). In HEK293 cells overexpressing Kv2.1, we observed identical effects of channel inhibition by dialyzed GST-Syn-1A, which could be blocked by Kv2.1C domain proteins (C1: amino acids 412–633, C2: amino acids 634–853), but not the Kv2.1 cytoplasmic N terminus (amino acids 1–182). This was confirmed by direct binding of GST-Syn-1A to the Kv2.1C1 and C2 domains proteins. These findings are in contrast to our recent report showing that Syn-1A binds and modulates the cytoplasmic N terminus of neuronal Kv1.1 and not by its C terminus. Co-expression of Syn-1A in Kv2.1-expressing HEK293 cells inhibited Kv2.1 surfacing, which caused a reduction of Kv2.1 current density. In addition, Syn-1A caused a slowing of Kv2.1 current activation and reduction in the slope factor of steady-state inactivation, but had no affect on inactivation kinetics or voltage dependence of activation. Taken together, SNAP-25 and Syn-1A mediate secretion not only through its participation in the exocytotic SNARE complex, but also by regulating membrane potential and calcium entry through their interaction with Kv and Ca2+ channels. In contrast to Ca2+ channels, where these SNARE proteins act on a common synprint site, the SNARE proteins act not only on distinct sites within a Kv channel, but also on distinct sites between different Kv channel families. voltage-dependent Ca2+ channels syntaxin 1A synaptosome-associated protein of 25 kDa voltage-dependent K+ channels glutathioneS-transferase green fluorescent protein soluble N-ethylmaleimide-sensitive factor attachment protein receptors fluorescein isothiocyanate human embryonic kidney In neurons and neuroendocrine cells, depolarization opens voltage-dependent Ca2+ channels (VDCC),1 and the subsequent Ca2+ influx triggers exocytosis of neurotransmitters or hormones. Vesicle fusion with the plasma membrane is initiated by the sensing of Ca2+ by the vesicle protein synaptotagmin. This is followed by, through yet unclear mechanisms, core complex formation by SNARE (soluble N-ethylmaleimide-sensitive factor attachment protein receptors) proteins, which involve another vesicle protein synaptobrevin (vesicle-SNARE) and two plasma membrane SNARE proteins SNAP-25 and syntaxin 1 (target-SNAREs) (1Gerber S.H. Sudhof T.C. Diabetes. 2002; 51 Suppl. 1: S3-S11Crossref PubMed Google Scholar, 2Rizo J. Sudhof T.C. Nat. Rev. Neurosci. 2002; 3: 641-653Crossref PubMed Scopus (435) Google Scholar, 3Rettig J. Neher E. Science. 2002; 298: 781-785Crossref PubMed Scopus (275) Google Scholar). It is proposed that core complex formation brings the two apposing membranes together and liberates the energy required to drive lipid re-orientation during fusion (1Gerber S.H. Sudhof T.C. Diabetes. 2002; 51 Suppl. 1: S3-S11Crossref PubMed Google Scholar, 2Rizo J. Sudhof T.C. Nat. Rev. Neurosci. 2002; 3: 641-653Crossref PubMed Scopus (435) Google Scholar, 3Rettig J. Neher E. Science. 2002; 298: 781-785Crossref PubMed Scopus (275) Google Scholar). SNARE proteins have been known to be tethered to various VDCC, and thus such a protein complex may provide rapid release response with SNARE proteins exposed to a high local Ca2+concentration permeating through the VDCCs, which in turn are being modulated by the SNARE proteins (4Seagar M. Takahashi M. J. Bioenerg. Biomembr. 1998; 30: 347-356Crossref PubMed Scopus (46) Google Scholar, 5Catterall W.A. Ann. N. Y. Acad. Sci. 1999; 868: 144-159Crossref PubMed Scopus (231) Google Scholar, 6Atlas D. J. Neurochem. 2001; 77: 972-985Crossref PubMed Scopus (121) Google Scholar, 7Kang Y. Huang X. Pasyk E.A. Ji J. Holz G.G. Wheeler M.B. Tsushima R.G. Gaisano H.Y. Diabetologia. 2002; 45: 231-241Crossref PubMed Scopus (54) Google Scholar, 8Ji J. Yang S.N. Huang X. Li X. Sheu L. Diamant N. Berggren P.O. Gaisano H.Y. Diabetes. 2002; 51: 1425-1436Crossref PubMed Scopus (71) Google Scholar). Because of such an intimate physical and functional coupling, the secretory vesicle-SNARE protein-Ca2+ channel complex has been termed excitosome (6Atlas D. J. Neurochem. 2001; 77: 972-985Crossref PubMed Scopus (121) Google Scholar). Therefore, besides their participation in membrane fusion, SNARE proteins appear to have a regulatory role on other components (i.e. membrane ion channels) of the exocytotic process. During increased glucose metabolism, a high intracellular ATP to ADP concentration ratio ([ATP]/[ADP]) causes inhibition of pancreatic islet औ-cell ATP-sensitive K+ channels (KATP) channels, which in turn results in depolarization and consequently insulin release (9Nichols C.G. Koster J.C. Am. J. Physiol. 2002; 283: E403-E412PubMed Google Scholar). Outward currents carried by voltage-gated K+ (Kv) channels in औ-cells are responsible for repolarization, which results in closure of VDCC and subsequent termination of exocytosis (10Hille B. Ionic Channels in Excitable Membranes. 2nd Ed. Sinauer Associates, Sunderland, MA1992: 115-139Google Scholar, 11Yellen G. Nature. 2002; 419: 35-42Crossref PubMed Scopus (536) Google Scholar, 12Smith P.A. Bokvist K. Arkhammar P. Berggren P.O. Rorsman P. J. Gen. Physiol. 1990; 95: 1041-1059Crossref PubMed Scopus (68) Google Scholar, 13Roe M.W. Worley 3rd, J.F. Mittal A.A. Kuznetsov A. DasGupta S. Mertz R.J. Witherspoon 3rd, S.M. Blair N. Lancaster M.E. McIntyre M.S. Shehee W.R. Dukes I.D. Philipson L.H. J. Biol. Chem. 1996; 271: 32241-32246Abstract Full Text Full Text PDF PubMed Scopus (106) Google Scholar). The role for these Kv channels in औ-cell excitation-secretion coupling is of clinical therapeutic importance, as blocking Kv channels with pharmacological agents can prolong depolarization and enhance Ca2+ entry, and thereby sustain insulin secretion in a glucose-dependent manner (14Henquin J.C. Meissner H.P. Preissler M. Biochim. Biophys. Acta. 1979; 587: 579-592Crossref PubMed Scopus (26) Google Scholar, 15MacDonald P.E. Sewing S. Wang J. Joseph J.W. Smukler S.R. Sakellaropoulos G. Wang J. Saleh M.C. Chan C.B. Tsushima R.G. Salapatek A.M. Wheeler M.B. J. Biol. Chem. 2002; 277: 44938-44945Abstract Full Text Full Text PDF PubMed Scopus (148) Google Scholar). Although SNARE protein-Ca2+ channel interaction has been studied in great detail, little is known about SNARE protein-Kv channel interaction. Recently, using the Xenopusoocyte expression system and coimmunoprecipitation experiments, we have shown that SNAP-25 and Syn-1A physically interact with Kv1.1 and Kv2.1 (16Ji J. Tsuk S. Salapatek A.M. Huang X. Chikvashvili D. Pasyk E.A. Kang Y. Sheu L. Tsushima R. Diamant N. Trimble W.S. Lotan I. Gaisano H.Y. J. Biol. Chem. 2002; 277: 20195-20204Abstract Full Text Full Text PDF PubMed Scopus (43) Google Scholar, 17MacDonald P.E. Wang G. Tsuk S. Dodo C. Kang Y. Tang L. Wheeler M.B. Cattral M.S. Lakey J.R. Salapatek A.M. Lotan I. Gaisano H.Y. Mol. Endocrinol. 2002; 16: 2452-2461Crossref PubMed Scopus (71) Google Scholar, 18Fili O. Michaelevski I. Bledi Y. Chikvashvili D. Singer-Lahat D. Boshwitz H. Linial M. Lotan I. J. Neurosci. 2001; 21: 1964-1974Crossref PubMed Google Scholar, 19Michaelevski I. Chikvashvili D. Tsuk S. Fili O. Lohse M.J. Singer-Lahat D. Lotan I. J. Biol. Chem. 2002; 277: 34909-34917Abstract Full Text Full Text PDF PubMed Scopus (25) Google Scholar). Functionally, SNAP-25 inhibits Kv1.1 and Kv2.1 currents, and such inhibition was mediated through binding of SNAP-25 to the Kv1.1 and Kv2.1 cytoplasmic N termini (16Ji J. Tsuk S. Salapatek A.M. Huang X. Chikvashvili D. Pasyk E.A. Kang Y. Sheu L. Tsushima R. Diamant N. Trimble W.S. Lotan I. Gaisano H.Y. J. Biol. Chem. 2002; 277: 20195-20204Abstract Full Text Full Text PDF PubMed Scopus (43) Google Scholar, 17MacDonald P.E. Wang G. Tsuk S. Dodo C. Kang Y. Tang L. Wheeler M.B. Cattral M.S. Lakey J.R. Salapatek A.M. Lotan I. Gaisano H.Y. Mol. Endocrinol. 2002; 16: 2452-2461Crossref PubMed Scopus (71) Google Scholar). We have also shown that Syn-1A has a concentration-dependent biphasic effect on Kv1.1 current amplitudes: at low concentration it enhances current without affecting surface channel expression while at high concentration it decreases current amplitude probably by reducing surface channel expression (18Fili O. Michaelevski I. Bledi Y. Chikvashvili D. Singer-Lahat D. Boshwitz H. Linial M. Lotan I. J. Neurosci. 2001; 21: 1964-1974Crossref PubMed Google Scholar). More recently, we further demonstrated that Syn-1A also binds to the cytoplasmic N terminus of Kv1.1, at the T1A domain and forms a stable complex with Gऔγ subunits (19Michaelevski I. Chikvashvili D. Tsuk S. Fili O. Lohse M.J. Singer-Lahat D. Lotan I. J. Biol. Chem. 2002; 277: 34909-34917Abstract Full Text Full Text PDF PubMed Scopus (25) Google Scholar). In this work, we surprisingly found that Syn-1A binds to the cytoplasmic C terminus of islet औ-cell Kv2.1 channel protein and modulates channel properties. Syn-1A, when overexpressed, also inhibited Kv2.1 surface expression and reduced Kv2.1 current density in heterologous HEK293 cells. Although SNARE protein interactions with Kv channels follow a similar paradigm as the Ca2+ channels, the interacting domains within and between the Kv channel families seem to be distinct in contrast to the highly conserved synprint site (cytoplasmic II-III loop) between the Ca2+ channel families (4–6,8). HEK293 cells were grown at 37 °C in 57 CO2 in Dulbecco's modified Eagle's medium supplemented with 107 fetal bovine serum (Invitrogen) and penicillin-streptomycin (100 units/ml, 100 ॖg/ml) (Invitrogen). The cells were transiently transfected with GFP and Kv2.1 with or without Syn-1A using LipofectAMINE 2000 (Invitrogen). Two days after transfection, cells were trypsinized and placed in 35-mm dishes for voltage-clamp experiments. Transfected cells were identified by visualization of the fluorescence of the co-expressed GFPs. Rat pancreatic islets were isolated by collagenase digestion as described previously (17MacDonald P.E. Wang G. Tsuk S. Dodo C. Kang Y. Tang L. Wheeler M.B. Cattral M.S. Lakey J.R. Salapatek A.M. Lotan I. Gaisano H.Y. Mol. Endocrinol. 2002; 16: 2452-2461Crossref PubMed Scopus (71) Google Scholar). Islets were dispersed to single cells by treatment with 0.0157 trypsin (Invitrogen) in Ca2+- and Mg2+-free phosphate-buffered saline. Islet cells were plated on glass coverslips in 35-mm dishes and cultured in low glucose Roswell Park Memorial Institute medium (2.5 mm glucose) supplemented with 7.57 fetal bovine serum, 0.257 HEPES (Sigma-Aldrich Canada Ltd.), and 100 units/ml penicillin G sodium, 100 ॖg/ml streptomycin sulfate (Invitrogen). Islet cells were cultured for 2 days before electrophysiological recordings. The vectors pcDNA3-Kv2.1, pcDNA3-Syntaxin 2, and pcDNA3-Syntaxin 1A were generously provided by Dr. R. Joho (University of Texas, Southwestern Medical Center, Dallas, TX) and Richard Scheller (Stanford University, Palo Alto, CA). The coding sequences corresponding to N-terminal region (1–182) of Kv 2.1 were amplified by polymerase chain reaction (PCR) and cloned into pGEX-5X-1 expression vector (Amersham Biosciences Inc.) for generation of GST fusion proteins. The primers used for PCR were 5′-ATGACGAAGCATGGCTCGC (sense) and 5′-CACCGACGAGTTGGGCT (antisense). The plasmids pGEX-4T-1-Kv2.1-C1 (the region corresponding to amino acids 412–633) and pGEX-4T-1-Kv2.1-C2 (the region corresponding to amino acids 634–853) were similarly generated. These constructs were verified by DNA sequencing. pGEX-4T-1-syntaxin-1A (wild type) is a gift from Dr. W. Trimble (The Hospital for Sick Children, Toronto, Ontario, Canada). GST fusion protein expression and purification were performed following the manufacturer's instructions. Syn-1A was obtained by cleavage of GST-Syn-1A with thrombin (Sigma). GST (as a control) and GST-Kv2.1-N or -C1 or -C2 (500 pmol of protein each) were bound to glutathione-agarose beads and incubated with thrombin-cleaved Syn-1A (500 pmol of protein) in 200 ॖl of binding buffer (25 mmHEPES pH 7.4, 50 mm NaCl, 0.17 gelatin, 0.17 Triton X-100, 0.17 bovine serum albumin, and 0.27 औ -mercaptoethanol) at 4 °C for 2 h with constant agitation. The beads were then washed two times with washing buffer containing 20 mm HEPES pH 7.4, 150 mm KOAc, 1 mm EDTA, 1 mm MgCl2, 57 glycerol, and 0.17 Triton X-100. The samples were then separated on 157 SDS-PAGE, transferred to nitrocellulose membrane (Millipore, Bedford, MA), and identified with specific primary antibody against Syn-1A (1:2000) (Sigma). HEK293 cells were voltage-clamped in the whole-cell configuration (20Hamill O.P. Marty A. Neher E. Sakmann B. Sigworth F.J. Pflugers Arch. 1981; 391: 85-100Crossref PubMed Scopus (15145) Google Scholar) using an EPC-9 amplifier and Pulse software (HEKA Electronik, Lambrecht, Germany) as we previously described (7Kang Y. Huang X. Pasyk E.A. Ji J. Holz G.G. Wheeler M.B. Tsushima R.G. Gaisano H.Y. Diabetologia. 2002; 45: 231-241Crossref PubMed Scopus (54) Google Scholar, 8Ji J. Yang S.N. Huang X. Li X. Sheu L. Diamant N. Berggren P.O. Gaisano H.Y. Diabetes. 2002; 51: 1425-1436Crossref PubMed Scopus (71) Google Scholar). Recording pipettes were pulled from 1.5-mm borosilicate glass capillary tubes (World Precision Instruments, Inc., Sarasota, FL) using a programmable micropipette puller (Sutter Instrument). Pipettes were then fire-polished and tip resistances ranged from 1.5–3 MΩ (for HEK cells) or 2.5–4 MΩ (for औ-cells) when filled with intracellular solution, containing (in mm): 140 KCl, 1 MgCl2, 1 EGTA, 10 HEPES, and 5 MgATP (pH 7.25 adjusted with KOH). Bath solution contained (mm): 140 NaCl, 4 KCl, 1 MgCl2, 2 CaCl2, 10 HEPES (pH 7.3 adjusted with NaOH). After a whole-cell configuration was established, membrane potential was held at −70 mV and outward currents were triggered with depolarizing voltage pulses (+70 mV, 250 ms). Steady-state outward currents was determined as the mean current in the final 95–997 of the 250-ms pulse. All experiments were performed at room temperature (22–24 °C). Data for voltage dependence of activation and steady-state inactivation were fit by the Boltzmann equation:I/Imax = 1/{1 + exp[(V − V12)/k]}, where V12 is the half-maximal activation potential (for voltage dependence of activation) or the half-maximal inactivation potential (for steady-state inactivation), and k the slope factor. Results are presented as means ± S.E. Unpaired Student'st test was employed, and p < 0.05 was considered statistically significant. Laser confocal immunofluorescence microscopy was performed as described previously (17MacDonald P.E. Wang G. Tsuk S. Dodo C. Kang Y. Tang L. Wheeler M.B. Cattral M.S. Lakey J.R. Salapatek A.M. Lotan I. Gaisano H.Y. Mol. Endocrinol. 2002; 16: 2452-2461Crossref PubMed Scopus (71) Google Scholar). Transfected HEK cells were fixed with 1007 methanol on 3-aminopropyltriethoxysilane-treated glass slides. The slides were then incubated at 4 °C overnight with primary antibodies, including mouse monoclonal anti-Kv2.1 (1:100) (Upstate Biotechnology Inc., Lake Placid, NY), rabbit anti-syntaxin-1 (Calbiochem, San Diego, CA) (1:100) and rabbit anti-syntaxin-2 (1:200, kindly provided by Dr. V. Olkkonen, National Public Health Institute, Helsinki, Finland). The slides were rinsed four times with phosphate-buffered saline containing 0.17 saponin and treated with secondary antibodies (FITC sheep anti-mouse IgG 1:500 or Texas Red-labeled goat anti-rabbit IgG 1:250) for 1 h. Next, they were incubated with 0.17 p-phenylenediamine (ICN, Cleveland, OH) in glycerol and examined using a laser scanning confocal imaging system (LSM-410; Carl Zeiss, Thornwood, NY). FITC signal was visualized by excitation at a wavelength of 488 nm and emitted fluorescence was measured through a 515- to 540-nm bandpass filter. Texas Red signal was visualized by an excitation wavelength of 568 nm and emitted fluorescence was detected through a 590-nm long-pass filter. Two days after transfection HEK293 cells were washed and harvested in phosphate-buffered saline. The cells were further washed with borate buffer (154 mmNaCl, 7.2 mm KCl, 1.8 mm CaCl2, 10 mm boric acid, pH 9.0) and then incubated in 5 ml of Sulfo-NHS-SS-Biotin (Pierce Biotechnology Inc.) (0.5 mg/ml) in borate buffer at 4 °C for 30 min. After washing three times with ice-cold quenching buffer (192 mm glycine, 25 mm Tris, pH 8.3), cells were solubilized on ice in 500 ॖl of immunoprecipitation buffer (17 deoxycholic acid, 17 Triton X-100, 0.17 SDS, 150 mm NaCl, 1 mm EDTA, 10 mm Tris-Cl, pH 7.5) containing a mixture of protease inhibitors (Roche Applied Science). The cell lysate was centrifuged for 20 min at 16,000 × g, and the supernatant was retained. 50 ॖl of immobilized streptavidin resin (Pierce) (507 slurry in phosphate-buffered saline containing 2 mmNaN3) was added to the supernatant, which was then incubated overnight at 4 °C with gentle rocking. Samples were centrifuged for 2 min at 8,000 × g, and the resin was washed five times with immunoprecipitaton buffer. The protein was eluted from the resin by the addition of SDS-PAGE sample buffer containing 57 2-mercaptoethanol and incubation at 65 °C for 5 min. The samples were analyzed for Kv2.1 expression by Western blotting using anti-Kv2.1 (1:1000, Alomone Labs, Jerusalem, Israel). Integrated density of the bands was determined using a commercial software (Scion Image Beta 4.02; Scion Corporation, Frederick, MD). We had previously shown that Kv2.1 channels account for ∼607 of the outward current in rat islet औ-cells (15MacDonald P.E. Sewing S. Wang J. Joseph J.W. Smukler S.R. Sakellaropoulos G. Wang J. Saleh M.C. Chan C.B. Tsushima R.G. Salapatek A.M. Wheeler M.B. J. Biol. Chem. 2002; 277: 44938-44945Abstract Full Text Full Text PDF PubMed Scopus (148) Google Scholar, 17MacDonald P.E. Wang G. Tsuk S. Dodo C. Kang Y. Tang L. Wheeler M.B. Cattral M.S. Lakey J.R. Salapatek A.M. Lotan I. Gaisano H.Y. Mol. Endocrinol. 2002; 16: 2452-2461Crossref PubMed Scopus (71) Google Scholar, 21MacDonald P.E. Ha X.F. Wang J. Smukler S.R. Sun A.M. Gaisano H.Y. Salapatek A.M. Backx P.H. Wheeler M.B. Mol. Endocrinol. 2001; 15: 1423-1435Crossref PubMed Scopus (163) Google Scholar), making this cell an excellent model to examine the effects of Syn-1A on this channel. Fig.1A shows the current-voltage relationships of outward K+ currents in rat islet औ-cells after an 8-min dialysis with various fusion proteins. For clarity, the results are presented as bar graphs at two positive voltages, which triggered outward currents (+10 and +60 mV) (Fig. 1, B andC). GST alone was used as a control that by itself did not cause any significant effect on औ-cell outward K+currents (data not shown). GST-Syn-1A (1 ॖm) caused a 26 and 227 reduction (p < 0.05) in current density at +10 and +60 mV, respectively. We then examined whether this is mediated via the cytoplasmic N terminus (amino acids 1–182) as we had reported with Kv1.1 (16Ji J. Tsuk S. Salapatek A.M. Huang X. Chikvashvili D. Pasyk E.A. Kang Y. Sheu L. Tsushima R. Diamant N. Trimble W.S. Lotan I. Gaisano H.Y. J. Biol. Chem. 2002; 277: 20195-20204Abstract Full Text Full Text PDF PubMed Scopus (43) Google Scholar) or with the cytoplasmic C terminus (amino acids 411–853). Since the cytoplasmic C terminus of Kv2.1 is quite large and difficult to generate the recombinant protein, we generated two smaller sections of this protein, C1 (412–633) and C2 (634–853). Surprisingly, we found that co-dialysis with C1 and/or C2 prevented Syn-1A from inhibiting the Kv2.1 currents, with C2 being more effective than C1, whereas the cytoplasmic Kv2.1 N terminus had no effect on Syn-1A actions. These data suggest that Syn-1A inhibited औ-cell Kv2.1 currents by interacting with the Kv2.1 C terminus. We next investigated whether the interactions between Syn-1A and the Kv2.1 C1 and C2 domains are direct by direct protein binding studies and functional studies using a heterologous expression model system (i.e. HEK293), which has little if any endogenous expression of these proteins (22Jarvis S.E. Zamponi G.W. J. Neurosci. 2001; 21: 2939-2948Crossref PubMed Google Scholar). Binding assay and electrophysiological data from our previous reports (16Ji J. Tsuk S. Salapatek A.M. Huang X. Chikvashvili D. Pasyk E.A. Kang Y. Sheu L. Tsushima R. Diamant N. Trimble W.S. Lotan I. Gaisano H.Y. J. Biol. Chem. 2002; 277: 20195-20204Abstract Full Text Full Text PDF PubMed Scopus (43) Google Scholar, 17MacDonald P.E. Wang G. Tsuk S. Dodo C. Kang Y. Tang L. Wheeler M.B. Cattral M.S. Lakey J.R. Salapatek A.M. Lotan I. Gaisano H.Y. Mol. Endocrinol. 2002; 16: 2452-2461Crossref PubMed Scopus (71) Google Scholar) suggest that SNAP-25 inhibits Kv1.1 and Kv2.1 current by binding to the cytoplasmic N terminus. We have also shown that Syn-1A binds to the N terminus of Kv1.1 (19Michaelevski I. Chikvashvili D. Tsuk S. Fili O. Lohse M.J. Singer-Lahat D. Lotan I. J. Biol. Chem. 2002; 277: 34909-34917Abstract Full Text Full Text PDF PubMed Scopus (25) Google Scholar). We therefore examined if Syn-1A bound to the cytoplasmic N or C terminus of Kv2.1 by performing the binding assay with recombinant proteins (Fig. 2). Syn-1A (Fig. 2, top panel) bound very strongly with C1 and less so with C2. Syn-1A bound only weakly with the N terminus. As a negative control, GST did not bind to Syn-1A at all. Fig. 2, bottom panel, shows a Ponseau S staining of the blot, which demonstrates the equal protein loading of the C1 and C2 proteins, whereas the N-terminal protein and GST were loaded somewhat more but nonetheless showed little and no binding to Syn-1A, respectively. Immunostaining of this blot with anti-GST antibodies confirmed the presence of these proteins (data not shown). As shown in Fig. 3, dialyzing Syn-1A-GST fusion protein through the recording pipette into Kv2.1-transfected cell caused a reduction (14.4 ± 5.07; p < 0.05) of Kv2.1 current after 6–8 min. This reduction could be abolished by co-dialysis with C1 or C2. Again, similar to the results with rat islet औ-cell Kv2.1 channels, C2 was more effective than C1 in blocking the effects of the dialyzed GST-Syn-1A. However, Kv2.1 N terminus was completely ineffective in preventing such reduction. These data indicate that Syn-1A inhibited Kv2.1 currents by a direct interaction with the cytoplasmic C terminus. Dialysis of the Syn-1A might either interact with a cytosolic protein and may not be specifically targeted to the plasma membrane where the Kv2.1 channel is situated. This could be circumvented by overexpressing Syn-1A, which would be appropriately targeted to the plasma membrane compartment (7Kang Y. Huang X. Pasyk E.A. Ji J. Holz G.G. Wheeler M.B. Tsushima R.G. Gaisano H.Y. Diabetologia. 2002; 45: 231-241Crossref PubMed Scopus (54) Google Scholar). Furthermore, Syn-1A has been shown to not only affect Kv1.1 channel function but also affected surface expression of Kv.1.1 in Xenopus oocytes (18Fili O. Michaelevski I. Bledi Y. Chikvashvili D. Singer-Lahat D. Boshwitz H. Linial M. Lotan I. J. Neurosci. 2001; 21: 1964-1974Crossref PubMed Google Scholar). As shown in Fig. 4A, expression of Syn-1A drastically reduced Kv2.1 current density (0.21 ± 0.05 nA/pF compared with Kv2.1 alone 0.99 ± 0.18 nA/pF; p < 0.05). The reduction in Kv2.1 current density in the presence of syntaxin-2 (Syn-2) was small and insignificant. To investigate whether Syn-1A reduced Kv2.1 current density by inhibiting the trafficking of Kv2.1 protein to the plasma membrane, we performed confocal immunofluorescence microscopy (Fig. 4B). When Kv2.1 was expressed alone, there was bright and clear fluorescence at the cell periphery, suggesting that the majority of the Kv2.1 channel protein was present at the plasma membrane (Fig. 4B,left panel). With Syn-1A co-expression, the Kv2.1 plasma membrane fluorescence was dimmer and diffuse (note patches of fluorescence beneath the cell periphery) (Fig. 4B,middle panel), indicating that the overexpressed Syn-1A (inset) inhibited Kv2.1 from surfacing to the plasma membrane, and a substantial proportion of Kv2.1 was retained in the cytoplasm. Consistent with the current density data shown in Fig.4A, overexpression of Syn-2 (inset) did not cause significant inhibition of surfacing of Kv2.1 (Fig. 4B,right panel). To quantitatively determine the amount of reduction of plasma membrane surfacing of Kv2.1 caused by the overexpression of Syn-1A and Syn-2, we performed the following study. Transfected HEK293 cells were biotinylated so that plasma membrane proteins can be separated from the rest of the cells using the streptavidin resin. Fig. 4C(upper panel) shows that the levels of plasma membrane Kv2.1 proteins pulled down by the streptavidin resin was reduced by 497 with the Syn-1A co-expression, but only by 227 with the Syn-2 expression. The Kv2.1 proteins in the total lysates obtained just prior to the treatment with streptavidin resin did not change, indicating that both syntaxins did not have any significant effect on total protein synthesis of Kv2.1 (lower panel). A number of reports have already shown that SNARE proteins profoundly affected Kv channel electrophysiological properties, such as activation and inactivation kinetics (16Ji J. Tsuk S. Salapatek A.M. Huang X. Chikvashvili D. Pasyk E.A. Kang Y. Sheu L. Tsushima R. Diamant N. Trimble W.S. Lotan I. Gaisano H.Y. J. Biol. Chem. 2002; 277: 20195-20204Abstract Full Text Full Text PDF PubMed Scopus (43) Google Scholar, 18Fili O. Michaelevski I. Bledi Y. Chikvashvili D. Singer-Lahat D. Boshwitz H. Linial M. Lotan I. J. Neurosci. 2001; 21: 1964-1974Crossref PubMed Google Scholar, 19Michaelevski I. Chikvashvili D. Tsuk S. Fili O. Lohse M.J. Singer-Lahat D. Lotan I. J. Biol. Chem. 2002; 277: 34909-34917Abstract Full Text Full Text PDF PubMed Scopus (25) Google Scholar). We next explored whether Syn-1A targeted to the plasma membrane by co-expression would modulate the electrophysiological properties of Kv2.1 channel current. Only those cells expressing currents greater than 4 nA were selected for analysis because HEK293 cells express endogenous outward K+ currents as high as 0.4 nA (data not shown). Kv2.1 had a fairly rapid activation rate, with a τ of 5.6 ± 0.6 ms (Fig.5A). The overexpressed Syn-1A significantly (p < 0.05) slowed down the activation rate (τ = 8.5 ± 0.7 ms) while Syn-2 had no effect. Kv2.1 exhibited a very slow inactivation rate (Fig. 5B), which was neither affected by Syn-1A nor Syn-2 co-expression. Since Syn-1A slows down Kv2.1 activation, we then examined whether Syn-1A would affect the voltage dependence of activation of Kv2.1. To study this, instantaneous activation curves were obtained using the protocol in which voltage steps from −50 to +70 mV in 10 mV increments were followed by a −40 mV step to trigger tail currents. Normalized peak tail currents are then plotted against the various voltage steps and fit by the Boltzmann equation (Fig.6A). Syn-1A did not significantly alter the voltage dependence of activation of Kv2.1. We performed the steady-state inactivation experiments to determine channel availability for activation as a function of membrane potential. A dual-pulse protocol was used in which a test pulse step of +70 mV was preceded by a long pre-pulse (12 s) of different potentials. The test pulse currents are normalized to the largest test pulse current and plotted against the pre-pulse voltages. The curves are best fit by the Boltzmann equation (Fig. 6B). Kv2.1 currents have a half-maximal inactivation potential (V12) of −29.9 ± 2.5 mV. The left shift of the inactivation curve (V12 value of −34.7 ± 1.3 mV) caused by Syn-1A was slight and statistically insignificant. However, Syn-1A significantly decreased the slope factor from 6.2 ± 0.63 to 4.4 ± 0.25 (p < 0.05), indicating that Syn-1A increased the sensitivity of voltage-dependent inactivation with
Referência(s)