Syntaxin-1A Binds the Nucleotide-binding Folds of Sulphonylurea Receptor 1 to Regulate the KATP Channel
2004; Elsevier BV; Volume: 279; Issue: 6 Linguagem: Inglês
10.1074/jbc.m309667200
ISSN1083-351X
AutoresEwa A. Pasyk, Youhou Kang, Xiaohang Huang, Ningren Cui, Laura Sheu, Herbert Y. Gaisano,
Tópico(s)Pancreatic function and diabetes
ResumoATP-sensitive potassium (KATP) channels in neuron and neuroendocrine cells consist of a pore-forming Kir6.2 and regulatory sulfonylurea receptor (SUR1) subunits, which are regulated by ATP and ADP. SNARE protein syntaxin 1A (Syn-1A) is known to mediate exocytic fusion, and more recently, to also bind and modulate membrane-repolarizing voltage-gated K+ channels. Here we show that Syn-1A acts as an endogenous regulator of KATP channels capable of closing these channels when cytosolic ATP concentrations were lowered. Botulinum neurotoxin C1 cleavage of endogenous Syn-1A in insulinoma HIT-T15 cells resulted in the increase in KATP currents, which could be subsequently inhibited by recombinant Syn-1A. Whereas Syn-1A binds both nucleotide-binding folds (NBF-1 and NBF-2) of SUR1, the functional inhibition of KATP channels in rat islet β-cells by Syn-1A seems to be mediated primarily by its interactions with NBF-1. These inhibitory actions of Syn-1A can be reversed by physiologic concentrations of ADP and by diazoxide. Syn-1A therefore acts to fine-tune the regulation of KATP channels during dynamic changes in cytosolic ATP and ADP concentrations. These actions of Syn-1A on KATP channels contribute to the role of Syn-1A in coordinating the sequence of ionic and exocytic events leading to secretion. ATP-sensitive potassium (KATP) channels in neuron and neuroendocrine cells consist of a pore-forming Kir6.2 and regulatory sulfonylurea receptor (SUR1) subunits, which are regulated by ATP and ADP. SNARE protein syntaxin 1A (Syn-1A) is known to mediate exocytic fusion, and more recently, to also bind and modulate membrane-repolarizing voltage-gated K+ channels. Here we show that Syn-1A acts as an endogenous regulator of KATP channels capable of closing these channels when cytosolic ATP concentrations were lowered. Botulinum neurotoxin C1 cleavage of endogenous Syn-1A in insulinoma HIT-T15 cells resulted in the increase in KATP currents, which could be subsequently inhibited by recombinant Syn-1A. Whereas Syn-1A binds both nucleotide-binding folds (NBF-1 and NBF-2) of SUR1, the functional inhibition of KATP channels in rat islet β-cells by Syn-1A seems to be mediated primarily by its interactions with NBF-1. These inhibitory actions of Syn-1A can be reversed by physiologic concentrations of ADP and by diazoxide. Syn-1A therefore acts to fine-tune the regulation of KATP channels during dynamic changes in cytosolic ATP and ADP concentrations. These actions of Syn-1A on KATP channels contribute to the role of Syn-1A in coordinating the sequence of ionic and exocytic events leading to secretion. All the excitable cell types, including neurons, muscles, and endocrine cells, are equipped with KATP channels (1Seino S. Annu. Rev. Physiol. 1999; 61: 337-362Crossref PubMed Scopus (456) Google Scholar). The KATP channel couples the intracellular metabolic state to electrical activity at the plasma membrane to regulate a number of cellular events such as hormone secretion, muscle contraction, and neuron excitability (1Seino S. Annu. Rev. Physiol. 1999; 61: 337-362Crossref PubMed Scopus (456) Google Scholar). The dominant KATP channels in neurons and neuroendocrine cells make up an octameric protein complex comprised of four Kir6.2 subunits, which are members of the inward rectifying K+ channel family, and four SUR1 1The abbreviations used are: SURsulfonylurea receptorGSTglutathione S-transferaseaaamino acidSNAREsoluble NSF attachment protein receptorNSFN-ethylmaleimide-sensitive factorNBFnucleotide-binding foldsCFTRcystic fibrosis transmembrane conductance regulator proteinGFPgreen fluorescent proteinEGFPenhanced GFPGlybglybenclamideHIThamster insulin-secreting tumorIPimmunoprecipitationFfaradCaVvoltage-dependent Ca2+ channelsKVvoltage-dependent K+ channels. subunits (1Seino S. Annu. Rev. Physiol. 1999; 61: 337-362Crossref PubMed Scopus (456) Google Scholar, 2Inagaki N. Gonoi T. Clement J.P. Namba N. Inazawa J. Gonzalex G. Aguilar-Bryan L. Seino S. Bryan J. Science. 1995; 270: 1166-1170Crossref PubMed Scopus (1623) Google Scholar, 3Aguilar-Bryan L. Nichols C.G. Herrera-Sosa H. Nguy K. Bryan J. Nelson D.A. Science. 1995; 268: 423-426Crossref PubMed Scopus (1286) Google Scholar). Each SUR1 protein contains two NBFs, NBF-1 and C-terminal NBF-2 (3Aguilar-Bryan L. Nichols C.G. Herrera-Sosa H. Nguy K. Bryan J. Nelson D.A. Science. 1995; 268: 423-426Crossref PubMed Scopus (1286) Google Scholar). Much work has been done to elucidate the actions of ATP and ADP on this channel complex (3Aguilar-Bryan L. Nichols C.G. Herrera-Sosa H. Nguy K. Bryan J. Nelson D.A. Science. 1995; 268: 423-426Crossref PubMed Scopus (1286) Google Scholar, 4Ashcroft F.M. Ashcroft S.J.H. Insulin: Molecular Biology to Pathology. Oxford University Press, Oxford, UK1992Google Scholar, 5Shyng S.-L. Ferrigni T. Nichols C.G. J. Gen. Physiol. 1997; 110: 643-654Crossref PubMed Scopus (246) Google Scholar, 6Ueda K. Inagaki N. Seino S. J. Biol. Chem. 1997; 272: 22983-22986Abstract Full Text Full Text PDF PubMed Scopus (151) Google Scholar, 7Matsuo M. Tanabe K. Kioka N. Amachi T. Ueda K. J. Biol. Chem. 2000; 275: 28757-28763Abstract Full Text Full Text PDF PubMed Scopus (120) Google Scholar, 8Tucker S.J. Gribble F.M. Zhao C. Trapp S. Aschroft F.M. Nature. 1997; 387: 179-183Crossref PubMed Scopus (683) Google Scholar, 9Vanoye C.G. MacGregor G.G. Dong K. Tan L. Buschmann A.S. Hall A.E. Lu M. Giebisch G. Hebert S.C. J. Biol. Chem. 2002; 277: 23260-23270Abstract Full Text Full Text PDF PubMed Scopus (47) Google Scholar), which, however, remains insufficient to fully explain the regulation of the KATP channel (1Seino S. Annu. Rev. Physiol. 1999; 61: 337-362Crossref PubMed Scopus (456) Google Scholar). Toward this, the neuroendocrine pancreatic islet β-cell has been the best studied physiologic model (1Seino S. Annu. Rev. Physiol. 1999; 61: 337-362Crossref PubMed Scopus (456) Google Scholar, 2Inagaki N. Gonoi T. Clement J.P. Namba N. Inazawa J. Gonzalex G. Aguilar-Bryan L. Seino S. Bryan J. Science. 1995; 270: 1166-1170Crossref PubMed Scopus (1623) Google Scholar, 3Aguilar-Bryan L. Nichols C.G. Herrera-Sosa H. Nguy K. Bryan J. Nelson D.A. Science. 1995; 268: 423-426Crossref PubMed Scopus (1286) Google Scholar, 4Ashcroft F.M. Ashcroft S.J.H. Insulin: Molecular Biology to Pathology. Oxford University Press, Oxford, UK1992Google Scholar). In the β-cell, glucose entry into the cell increases the ATP/ADP ratio, which closes the KATP channels (4Ashcroft F.M. Ashcroft S.J.H. Insulin: Molecular Biology to Pathology. Oxford University Press, Oxford, UK1992Google Scholar). This leads to membrane depolarization, which opens voltage-dependent Ca2+ channels (CaV), allowing Ca2+ to enter the cell and interact with Ca2+-sensitive proteins, which eventually leads the SNARE proteins to effect exocytosis (10Sudhof T.C. Nature. 1995; 375: 645-653Crossref PubMed Scopus (1770) Google Scholar, 11Regazzi R. Wollheim C.B. Lang J. Theler J.M. Rossetto O. Montecucco C. Sadoul K. Weller U. Palmer M. Thorens B. EMBO J. 1995; 14: 2723-2730Crossref PubMed Scopus (210) Google Scholar). Membrane repolarization effected by the opening of voltage-dependent K+ (Kv) channels then closes CaV to terminate exocytosis. This sequence of ionic and exocytic events in the β-cell parallels those of neurons (11Regazzi R. Wollheim C.B. Lang J. Theler J.M. Rossetto O. Montecucco C. Sadoul K. Weller U. Palmer M. Thorens B. EMBO J. 1995; 14: 2723-2730Crossref PubMed Scopus (210) Google Scholar). sulfonylurea receptor glutathione S-transferase amino acid soluble NSF attachment protein receptor N-ethylmaleimide-sensitive factor nucleotide-binding folds cystic fibrosis transmembrane conductance regulator protein green fluorescent protein enhanced GFP glybenclamide hamster insulin-secreting tumor immunoprecipitation farad voltage-dependent Ca2+ channels voltage-dependent K+ channels. The role of SNARE proteins in the plasticity of neuro- and neuroendocrine secretion extends beyond exocytotic fusion (10Sudhof T.C. Nature. 1995; 375: 645-653Crossref PubMed Scopus (1770) Google Scholar, 11Regazzi R. Wollheim C.B. Lang J. Theler J.M. Rossetto O. Montecucco C. Sadoul K. Weller U. Palmer M. Thorens B. EMBO J. 1995; 14: 2723-2730Crossref PubMed Scopus (210) Google Scholar) and is now known to also directly regulate not only CaV channels (12Sheng Z.-H. Rettig J. Cook T. Catterall W.A. Nature. 1996; 379: 451-454Crossref PubMed Scopus (313) Google Scholar, 13Yang S.-N. Larsson O. Branstrom R. Bertorello A.M. Leibiger B. Leibiger I.B. Moede T. Kohler M. Meister B. Berggren P.O. Proc. Natl. Acad. Sci. U. S. A. 1999; 96: 10164-10169Crossref PubMed Scopus (122) Google Scholar), but more recently, the Kv channels (14Leung Y. Kang Y.H. Gao X. Xia F. Xie H. Sheu L. Tsuk S. Lotan I. Tsushima R. Gaisano H.Y. J. Biol. Chem. 2003; 276: 17532-17538Abstract Full Text Full Text PDF Scopus (112) Google Scholar, 15Fili 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 well. Here, we have used the neuroendocrine β-cell models to examine whether SNARE proteins could regulate KATP channel closure, which would trigger the sequence of ionic events leading to exocytosis. The SURs are members of the ATP-binding cassette protein superfamily, which includes the CFTR Cl– channel (16Higgins C.F. Cell. 1995; 82: 693-696Abstract Full Text PDF PubMed Scopus (341) Google Scholar, 17Naren A.P. Quick M.W. Collawn J.F. Nelson D.J. Kirk K.L. Proc. Natl. Acad. Sci. U. S. A. 1998; 95: 10972-10977Crossref PubMed Scopus (128) Google Scholar). Since SNARE protein, Syn-1A, binds CFTR (17Naren A.P. Quick M.W. Collawn J.F. Nelson D.J. Kirk K.L. Proc. Natl. Acad. Sci. U. S. A. 1998; 95: 10972-10977Crossref PubMed Scopus (128) Google Scholar), we speculated that SUR1 might have similar interactions with SNARE proteins. Whereas Syn-1A binds the N-terminal tail of CFTR (17Naren A.P. Quick M.W. Collawn J.F. Nelson D.J. Kirk K.L. Proc. Natl. Acad. Sci. U. S. A. 1998; 95: 10972-10977Crossref PubMed Scopus (128) Google Scholar), we surprisingly found that Syn-1A binds SUR1 at its NBF-1 and NBF-2 domains. Since NBF-1 and NBF-2 are nucleotide-regulated domains, this Syn-1A binding would likely influence ATP and ADP regulation of this channel. Cell Culture and Transfection—HIT-T15 cells (P. Robertson, Seattle, WA), passages 75–85, grown in Petri dishes (RPMI 1640 medium supplemented with 20 mm glutamine, 10% fetal calf serum, 100 units/ml penicillin, 100 mg/ml streptomycin) were transfected with plasmids (in pcDNA3, Clontech) containing cDNAs of full-length Syn-1A, or with the light chain of BoNT/C1 (H. Niemann, Hanover, Germany). The cells were cotransfected with pcDNA3-GFP to identify the BoNT/C1- or Syn-1A-transfected cells (14Leung Y. Kang Y.H. Gao X. Xia F. Xie H. Sheu L. Tsuk S. Lotan I. Tsushima R. Gaisano H.Y. J. Biol. Chem. 2003; 276: 17532-17538Abstract Full Text Full Text PDF Scopus (112) Google Scholar, 18Kang H.Y. Huang X.H. Pasyk E.A. Ji J. Holz H. Wheeler M.B. Tsushima R. Gaisano H.Y. Diabetologia. 2002; 45: 231-241Crossref PubMed Scopus (55) Google Scholar). HIT-T15 cells have a >90% probability of picking up multiple plasmids (14Leung Y. Kang Y.H. Gao X. Xia F. Xie H. Sheu L. Tsuk S. Lotan I. Tsushima R. Gaisano H.Y. J. Biol. Chem. 2003; 276: 17532-17538Abstract Full Text Full Text PDF Scopus (112) Google Scholar, 18Kang H.Y. Huang X.H. Pasyk E.A. Ji J. Holz H. Wheeler M.B. Tsushima R. Gaisano H.Y. Diabetologia. 2002; 45: 231-241Crossref PubMed Scopus (55) Google Scholar). Transfection efficiency at 48 h was ∼30–40% as determined by visualization of the co-expressed GFP, which serves to identify the transfected cells for electrophysiological studies. Islet β-Cell Isolation—Pancreatic islets were isolated from male Sprague-Dawley rats by collagenase digestion and dispersed into single cells by treatment with 0.015% trypsin in Ca2+- and Mg2+-free phosphate-buffered saline as described previously (14Leung Y. Kang Y.H. Gao X. Xia F. Xie H. Sheu L. Tsuk S. Lotan I. Tsushima R. Gaisano H.Y. J. Biol. Chem. 2003; 276: 17532-17538Abstract Full Text Full Text PDF Scopus (112) Google Scholar). Islet cells were plated on glass coverslips in 35-mm dishes and cultured in 2.8 mm glucose (with 7.5% fetal calf serum, 0.25% sodium, 100 μg/ml streptomycin). For intracellular dialysis experiments, islet cells were cultured for 1–2 days before electrophysiological recordings. Electrophysiology—Single islet β-cells and HIT-T15 cells were studied with the standard whole-cell patch voltage clamp technique as reported previously (14Leung Y. Kang Y.H. Gao X. Xia F. Xie H. Sheu L. Tsuk S. Lotan I. Tsushima R. Gaisano H.Y. J. Biol. Chem. 2003; 276: 17532-17538Abstract Full Text Full Text PDF Scopus (112) Google Scholar, 18Kang H.Y. Huang X.H. Pasyk E.A. Ji J. Holz H. Wheeler M.B. Tsushima R. Gaisano H.Y. Diabetologia. 2002; 45: 231-241Crossref PubMed Scopus (55) Google Scholar). Thin-walled (1.5 mm) borosilicate glass tubes were pulled with a two-stage Narishige (Tokyo, Japan) micropipette puller and heat-polished. The typical tip resistance was 2–4 megaohms. The external solution contained 140 mm NaCl, 4 mm KCl, 1 mm MgCl2, 2 mm CaCl2, 10 mm HEPES, pH 7.3, and as indicated, glybenclamide or diazoxide. The internal patch pipette solution contained 140 mm KCl, 1 mm MgCl2, 10 mm HEPES, 1 mm EGTA, and the indicated amounts of MgATP and MgADP; pH 7.3. The indicated recombinant proteins were added to the intracellular solution for dialysis into β-cells via the patch pipette. All electrophysiological experiments were done at room temperature (22–24 °C). Currents were elicited with 250-ms voltage steps of 20 mV from –140 mV to –20 mV from a holding potential of –50 mV using an EPC-9 amplifier and Pulse software (HEKA Electronik, Lambrecht, Germany). Data were presented as mean ± S.E. Data were compared by Student's t test. Recombinant GST Fusion Proteins—The coding sequences corresponding to NBF-1 (aa 696–893) and NBF-2 (aa 1358–1544) regions of SUR1 (3Aguilar-Bryan L. Nichols C.G. Herrera-Sosa H. Nguy K. Bryan J. Nelson D.A. Science. 1995; 268: 423-426Crossref PubMed Scopus (1286) Google Scholar), Syn-1A (aa 1–266, without transmembrane domain), and the cytoplasmic N (aa 1–70) and C terminus (aa 263–390) of Kir6.2 (2Inagaki N. Gonoi T. Clement J.P. Namba N. Inazawa J. Gonzalex G. Aguilar-Bryan L. Seino S. Bryan J. Science. 1995; 270: 1166-1170Crossref PubMed Scopus (1623) Google Scholar) were amplified by PCR and cloned into pGEX 5X-1 expression vector (Amersham Biosciences) for generation of GST fusion proteins. GST fusion protein expression and purification were performed following the manufacturer's instructions (14Leung Y. Kang Y.H. Gao X. Xia F. Xie H. Sheu L. Tsuk S. Lotan I. Tsushima R. Gaisano H.Y. J. Biol. Chem. 2003; 276: 17532-17538Abstract Full Text Full Text PDF Scopus (112) Google Scholar). Before elution of GST fusion proteins from glutathione-Sepharose beads, Syn-1A protein was obtained by cleavage of GST-Syn-1A with thrombin (Sigma). In Vitro Binding Assay—In vitro binding assays were performed as described previously (14Leung Y. Kang Y.H. Gao X. Xia F. Xie H. Sheu L. Tsuk S. Lotan I. Tsushima R. Gaisano H.Y. J. Biol. Chem. 2003; 276: 17532-17538Abstract Full Text Full Text PDF Scopus (112) Google Scholar) with some modifications. Briefly, the GST (control), GST-NBF-1 or GST-NBF-2, or GST-Syn-1A were bound to glutathione-agarose beads and incubated with thrombin-cleaved Syn-1A, GST-NBF-1, or GST-NBF-2 (500 pmol) in 200 μl of binding buffer (25 mm HEPES, (pH 7.4)), 50 mm NaCl, 0.1% gelatin, 0.1% Triton X-100, 0.1% bovine serum albumin, 0.2% β-mercaptoethanol) or with solubilized rat brain synaptic membranes (25 μg of protein) in 100 μl of immunoprecipitation assay buffer (Tris-HCl, pH 7.5, 150 mm NaCl, 1% Triton X-100, 1 mm benzamidine, 1 μg/ml pepstatin A, 5 μg/ml aprotinin, 1 μg/ml leupeptin, and 1 mm phenylmethylsulfonyl fluoride) at 4 °C for 2 h (or as indicated) with constant agitation. The beads were then washed twice with washing buffer (20 mm HEPES (pH 7.4), 150 mm KOAC, 1 mm EDTA, 1 mm MgCl2, 5% glycerol, 0.1% Triton X-100). The bound proteins were separated on 15% SDS-PAGE and identified by specific primary antibody against the SUR1 C-terminal 13 amino acids (1:1000, a gift from J. Ferrer, Barcelona, Spain) (19Hernandez-Sanchez C. Ito Y. Ferrer J. Reitman M. LeRoith D. J. Biol. Chem. 1999; 274: 18261-18279Abstract Full Text Full Text PDF PubMed Scopus (35) Google Scholar) or Syn-1A (1:200, Sigma). Immunoprecipitation Assays—Protein A-sephorose beads (40 μl; 50% slurry, Sigma) were washed with saline (pH 7.4) and incubated with 3 μg of rabbit anti-Syn-1A antibody in 200 μl of immunoprecipitation (IP) buffer (25 mm HEPES, pH 7.4, 100 mm KCl, 2 mm EDTA, 2% Triton X-100, 20 μm NaF, 1 mm phenylmethylsulfonyl fluoride, 1 μg/ml leupeptin, 10 μg/ml aprotinin) at 4 °C for 1 h followed by washing twice with IP buffer. 200 μl of IP buffer and 500 pmol of recombinant Syn-1A were then added and incubated at 4 °C for 1 h followed by another wash. The immunoprecipitated Syn-1A was incubated with GST (negative control), GST-NBF-1, or GST-NBF-2 (500 pmol) in 200 μl of IP buffer at 4 °C for 2 h, and then washed three times. The co-precipitated proteins were separated on 15% PAGE, and GST-NBF proteins were identified on the gel by Coomassie Blue staining. Confocal Microscopy—As described previously (18Kang H.Y. Huang X.H. Pasyk E.A. Ji J. Holz H. Wheeler M.B. Tsushima R. Gaisano H.Y. Diabetologia. 2002; 45: 231-241Crossref PubMed Scopus (55) Google Scholar), HIT-T15 cells were plated on polylysine-coated coverslips, transfected as described above, and then fixed in 2% formaldehyde for 0.5 h at room temperature followed by blocking with 5% normal goat serum (0.1% saponin, 0.5 h, room temperature) and then immunolabeled with primary antibodies (1:50, overnight, 4 °C) against Syn-1 (Sigma) or SNAP-25 (Sternberger Monoclonal, Lutherville, MD). After rinsing with 0.1% saponin/phosphate-buffered saline, the coverslips were incubated with the appropriate rhodamine-labeled secondary antiserum for 1 h, mounted on slides in a fading retarder (0.1% p-phenylenediamine in glycerol), and then examined by a laser scanning confocal imaging system (LSM510, Carl Zeiss, Oberkochen, Germany). Transfected cells were identified by visualization of the co-expressed EGFP. Syntaxin-1A Inhibits KATP Channel Opening—Using the whole-cell patch clamp technique (14Leung Y. Kang Y.H. Gao X. Xia F. Xie H. Sheu L. Tsuk S. Lotan I. Tsushima R. Gaisano H.Y. J. Biol. Chem. 2003; 276: 17532-17538Abstract Full Text Full Text PDF Scopus (112) Google Scholar, 18Kang H.Y. Huang X.H. Pasyk E.A. Ji J. Holz H. Wheeler M.B. Tsushima R. Gaisano H.Y. Diabetologia. 2002; 45: 231-241Crossref PubMed Scopus (55) Google Scholar), we first identified the KATP channels in the insulinoma HIT-T15 cells by using glybenclamide (Glyb), the sulfonylurea KATP channel inhibitor. Glyb (0.1 μm) greatly inhibited the HIT-T15 KATP currents (Fig. 1, A and B) recorded at low ATP (0.3 mm) concentration (n = 3). Dialysis of GST-Syn-1A (1 μm) into the HIT-T15 cells also greatly inhibited the KATP currents (Fig. 1, A and B; n = 10). As a control, we used HIT-T15 cells dialyzed with GST (1 μm), which had no effect on KATP currents when compared with the pipette solution (data not shown). Fig. 1C shows the time course of GST-Syn-1A inhibitory effect on the KATP currents under the low ATP (0.3 mm) concentration. Just after the formation of the whole-cell configuration (t = 0 min), the KATP currents remained low for ∼2 min (asterisk) for both control and GST-Syn-1A-treated cells. After 2 min, control currents increased steadily as the cell interior equilibrated with the low ATP pipette solution so that the KATP channels would open. At t = 10 min, the control KATP current amplitude exhibited a vigorous increase to 169.0 ± 15.9 pA (n = 5), whereas the current amplitude of the cells dialyzed with GST-Syn-1A showed only a small increase to 65.4 ± 9.7 pA (n = 5). Fig. 1D shows the summary bar graph in which the KATP currents were normalized to cell capacitance (pA/pF) to eliminate the variations in cell size. Glyb (0.1 μm) inhibited KATP current amplitude by ∼85% (n = 3), whereas GST-Syn-1A (1 μm) reduced KATP currents by ∼65% (n = 5) of the control current (61.9 ± 8.8 pA/pF; n = 5). To determine whether the effect of Syn-1A is at the plasma membrane or cytosolic level, we examined HIT-T15 cells overexpressing the full-length Syn-1A (aa 1–288) (20Bennett M.K. Calakos N. Scheller R.H. Science. 1992; 257: 255-259Crossref PubMed Scopus (1077) Google Scholar), which included the transmembrane domain, thereby specifically targeting Syn-1A to the plasma membrane (18Kang H.Y. Huang X.H. Pasyk E.A. Ji J. Holz H. Wheeler M.B. Tsushima R. Gaisano H.Y. Diabetologia. 2002; 45: 231-241Crossref PubMed Scopus (55) Google Scholar). This Syn-1A overexpression (>3-fold by Western blot) exhibited very similar inhibition of the KATP whole-cell currents (data not shown) as the dialysis of cytosolic Syn-1A (aa 1–266) shown in Fig. 1. These results indicate that the active Syn-1A domain that inhibits this KATP channel involves only the cytoplasmic domain (aa 1–266) and not the transmembrane domain (aa 267–288) of Syn-1A. We have postulated that the closed state of this KATP channel is attributed not only to cytosolic ATP levels but possibly also to the levels of endogenous Syn-1A. To examine whether Syn-1A acts as an endogenous inhibitor of the KATP channel, we expressed botulinum neurotoxin C1 (BoNT/C1) in HIT-T15 cells, which would specifically cleave the endogenous Syn-1A (21Schiavo G. Shone C.C. Bennett M.K. Scheller R.H. Monteccuco C. J. Biol. Chem. 1995; 270: 10566-10570Abstract Full Text Full Text PDF PubMed Scopus (241) Google Scholar). Since HIT-T15 cells very reliably pick up multiple plasmids (>90%), we co-expressed GFP to identify the BoNT/C1-expressing cells (18Kang H.Y. Huang X.H. Pasyk E.A. Ji J. Holz H. Wheeler M.B. Tsushima R. Gaisano H.Y. Diabetologia. 2002; 45: 231-241Crossref PubMed Scopus (55) Google Scholar). Confocal microscopy showed that the membrane Syn-1A signal in the BoNT/C1-expressing cells (GFP-containing, indicated by arrows) was greatly reduced (Fig. 1E, lower panels) when compared with the endogenous Syn-1A levels of the adjacent cells that did not express BoNT/C1. As the BoNT/C1-expressing cells did not show an increase in the cytoslic Syn-1A signal, this suggests that the cleaved Syn-1A fragments likely undergo cytosolic proteolysis since the Syn-1A monoclonal antibody (Sigma) was generated against the full-length protein and would have recognized either the membrane-bound or the cytosolic fragments of Syn-1A. Since BoNT/C1 could also cleave SNAP-25 in some cell types (22Foran P. Lawrence G.W. Shone C.C. Foster K.A. Dolly J.O. Biochemistry. 1996; 35: 2630-2636Crossref PubMed Scopus (235) Google Scholar), we also labeled these cells with anti-SNAP-25 antibody (Fig. 1E, upper panels) but saw no difference in the membrane SNAP-25 signals between the BoNT/C1-expressing cells (GFP-containing, indicated by arrows) and the cells that did not express GFP. This result indicates specific proteolysis of endogenous Syn-1A by BoNT/C1. Next, we examined the whole-cell KATP currents of these BoNT/C1-expressing cells (Fig. 1, A and B), which was ∼155% (n = 5) of the control cells (Fig. 1D), indicating that the KATP currents were indeed being inhibited by the endogenous Syn-1A. Dialysis of GST-Syn-1A (1 μm) into these BoNT/C1-expressing cells (Fig. 1A, BoNT/C1+Syn-1A) reduced the KATP current by ∼70% (n = 8) as compared with the BoNT/C1-expressing cells, a reduction similar to the one observed with GST-Syn-1A in the control cells. The cytosolic ATP concentration in these cells was lowered to 0.3 mm ATP to reduce ATP blockade of the KATP channel. The observed Syn-1A inhibition of the KATP channels may be due to an increase in the sensitivity to ATP-mediated inhibition, or it may be independent of ATP inhibition. More studies would be required to distinguish these possibilities. Syntaxin 1A Binds SUR1 at its NBF-1 and NBF-2—Syn-1A actions on these KATP channels could be mediated by binding to the SUR1 or Kir6.2 subunits. GST-Syn-1A bound to agarose beads pulled down SUR1 from solubilized rat brain synaptic membranes, which was identified with a specific antibody generated against the SUR1 C terminus (19Hernandez-Sanchez C. Ito Y. Ferrer J. Reitman M. LeRoith D. J. Biol. Chem. 1999; 274: 18261-18279Abstract Full Text Full Text PDF PubMed Scopus (35) Google Scholar) (Fig. 2A). The major cytoplasmic domains of SUR1 are NBF-1 (aa 696–893) and C-terminal NBF-2 (aa 1358–1544) (3Aguilar-Bryan L. Nichols C.G. Herrera-Sosa H. Nguy K. Bryan J. Nelson D.A. Science. 1995; 268: 423-426Crossref PubMed Scopus (1286) Google Scholar). Fig. 2B shows that GST-NBF-1 and GST-NBF-2 bound to agarose beads pulled down native Syn-1A from rat brain synaptic membranes. To negate the possibility of a coprecipitation of unknown ternary proteins present in the brain, we demonstrated that GST-NBF-1 and GST-NBF-2 pulled down recombinant Syn-1A but not control GST or recombinant SNAP-25 (data not shown). Fig. 2C shows the reciprocal study of immunoprecipitation of Syn-1A (GST cleaved off by thrombin), which pulled down GST-NBF-1 (lane 3) and GST-NBF-2 (lane 5) and not GST (lane 1). The coprecipitated NBFs were identified by Coomassie Blue staining as the only proteins eluting at the same sizes as the control proteins (lanes 4 and 6). Fig. 2D examines the time course of binding (4 °C), showing similar rapid binding kinetics of both NBFs to Syn-1A, in which Syn-1A binding was detectable at 2 min (first time point, ∼20% of maximal binding), reaching maximal binding at 1–2 h. Half time (t½) was ∼15.2 and ∼9.3 min for NBF-1 and NBF-2, respectively. Fig. 2E examines (i) the concentration dependence of GST-NBF-1 and (ii) GST-NBF-2 binding to Syn-1A (500 pmol, 4 °C, 2 h) at saturating concentrations. NBF-1 and NBF-2 binds Syn-1A at an ∼1:1 ratio, with IC50 of 141.9 ± 12.5 and 93.4 ± 6.6 pmol, respectively. Syn-1A could inhibit the KATP channels by acting directly on the cytoplasmic N or C termini of the Kir6.2 subunit on which the ATP acts (8Tucker S.J. Gribble F.M. Zhao C. Trapp S. Aschroft F.M. Nature. 1997; 387: 179-183Crossref PubMed Scopus (683) Google Scholar, 9Vanoye C.G. MacGregor G.G. Dong K. Tan L. Buschmann A.S. Hall A.E. Lu M. Giebisch G. Hebert S.C. J. Biol. Chem. 2002; 277: 23260-23270Abstract Full Text Full Text PDF PubMed Scopus (47) Google Scholar). This is not the case since neither recombinant cytoplasmic N- (aa 1–70) nor C-terminal (aa 263–390) domains of Kir6.2 bound to agarose beads pulled down recombinant Syn-1A (data not shown). Furthermore, dialysis of these Kir6.2 domain proteins into rat islet β-cells did not influence GST-Syn-1A effects on KATP activity (data not shown). Since SNAP-25 partners with Syn-1A to bind and regulate Ca2+ channels (23Wiser O. Bennet M.K. Atlas D. EMBO J. 1996; 15: 4100-4110Crossref PubMed Scopus (235) Google Scholar), SNAP-25 might also directly modulate KATP channels. However, GST-SNAP-25 (1 μm) dialyzed into HIT-T15 cells (and rat islet β-cells) had no effect on KATP activity (data not shown), and neither GST-NBF-1 nor GST-NBF-2 pulled down recombinant SNAP-25 (data not shown). Syntaxin-1A Inhibition of the KATP Channel Is Mediated by Its Binding to NBF-1—The binding of Syn-1A to NBF-1 and NBF-2 of SUR1 suggests that Syn-1A inhibition of the HIT-T15 KATP channels may be mediated by its interactions with either one of the NBFs. Therefore, we examined the functional interactions of Syn-1A with each NBF using rat islet β-cells, which would be a better physiologic model. The rat islet β-cell KATP channels (Fig. 3, A–C) behaved very similarly as the HIT-T15 cell KATP channels (Fig. 1, A–D). Specifically, under low ATP (0.3 mm ATP) concentrations, the external application of Glyb (0.1 μm) and dialysis of GST-Syn-1A (1 μm) into islet β-cell inhibited the KATP currents (Fig. 3A). The time course of GST-Syn-1A inhibition of rat islet β-cell KATP channels (Fig. 3B) was in fact very similar in pattern to those observed in the HIT cell study (Fig. 1C). When normalized to cell capacitance (pA/pF; Fig. 3C), Glyb (0.1 μm) and GST-Syn-1A (1 μm) inhibited rat islet β-cell KATP control currents (65.4 ± 9.7 pA/pF; n = 12) by 85% (9.7 ± 1.7 pA/pF, n = 3) and 61% (25.6 ± 5.7 pA/pF, n = 10), respectively. These values are remarkably similar to the HIT cell studies (Fig. 1). Next we examined the effects of GST-NBF-1 and GST-NBF-2 on the inhibitory actions of Syn-1A (Fig. 3, A and C). GST-NBFs would be expected to bind the endogenous Syn-1A, which would either prevent the formation of or disrupt the complex already formed by endogenous Syn-1A with the β-cell SUR1, both of which would increase KATP currents. Since GST-NBF-1 (1 μm) and GST-NBF-2 (1 μm) alone only slightly but not significantly increased β-cell KATP currents to 77.8 ± 14.6 pA/pF (n = 6) and 72.6 ± 16.3 pA/pF (n = 8), respectively, this supports the latter possibility, that the pre-existing endogenous Syn-1A may have formed stable complexes with the NBFs of many of the endogenous SUR1 proteins, and these complexes may be resistant to disruption by the exogenous GST-NBFs. BoNT/C1 (Fig. 1) would disrupt such complexes upon cleavage of the endogenous Syn-1A. The exogenous NBFs could still pull down Syn-1A from the rat brain (as in Fig. 2A) since Syn-1A is more abundant and generally distributed in the brain than SUR1, and hence, an excess of brain Syn-1A would be available to bind the exogenous NBF proteins. More importantly, when dialyzed together with GST-Syn-1A (1 μm), GST-NBF-1 (NBF-1+Syn-1A) completely blocked the in
Referência(s)