- Email: [email protected]
Proteasomal degradation of Kir6.2 channel protein and its inhibition by a Na+ channel blocker aprindine
BBRC Biochemical and Biophysical Research Communications 331 (2005) 1001–1006 www.elsevier.com/locate/ybbrc
Proteasomal degradation of Kir6.2 channel protein and its inhibition by a Na+ channel blocker aprindine Hiroaki Tanaka a, Junichiro Miake b, Tomomi Notsu b, Kazuhiko Sonyama a, Norihito Sasaki b, Kazuhiko Iitsuka a, Masaru Kato a, Shin-ichi Taniguchi a, Osamu Igawa a, Akio Yoshida a, Chiaki Shigemasa a, Yoshiko Hoshikawa b, Yasutaka Kurata c, Akihiko Kuniyasu d, Hitoshi Nakayama d, Nobuo Inagaki e, Eiji Nanba f, Goshi Shiota g, Takayuki Morisaki h, Haruaki Ninomiya i, Masafumi Kitakaze j, Ichiro Hisatome b,* a
Division of Cardiovascular Medicine, The First Department of Internal Medicine, Tottori University Faculty of Medicine, Yonago 683-8504, Japan b Division of Regenerative Medicine and Therapeutics, Department of Gene Therapy and Regenerative Medicine, Tottori University Graduate School of Medical Science, Japan c Department of Physiology, Kanazawa Medical University, Kanazawa, Japan d Department of Medical Biochemistry, Graduate School of Medical and Pharmaceutical Sciences, Kumamoto University, Kumamoto, Japan e Department of Physiology, Akita University School of Medicine, Akita 010-8543, Japan f Division of Functional Genomics, Research Center for Bioscience and Technology, Tottori University, Yonago, Japan g Division of Molecular and Genetic Medicine, Department of Genetic Medicine and Regenerative Therapeutics, Faculty of Medicine, Tottori University, Japan h Department of Bioscience, National Cardiovascular Center Research Institute, Osaka, Japan i Department of Neurobiology, Tottori University Faculty of Medicine, Yonago, Japan j Department of Cardiology, National Cardiovascular Center, Osaka, Japan Received 25 March 2005 Available online 11 April 2005
Abstract ATP-sensitive K+ channels (KATP:SUR2A+Kir6.2) play a pivotal role in cardiac protection against ischemia and reperfusion injury. When expressed in COS cells, Kir6.2 was short-lived with a half-life time of 1.9 h. The half-life time of Kir6.2 was prolonged by proteasome inhibitors MG132, ALLN, proteasome inhibitor 1, and lactacystine, but not at all by a lysosomal inhibitor chloroquine. MG132 also increased the level of ubiquitinated Kir6.2 without affecting its localization in the endoplasmic reticulum and Golgi apparatus. In electrophysiological recordings, MG132 augmented nicorandil-activated KATP currents in COS cells expressing SUR2A and Kir6.2 as well as the same currents in neonatal rat cardiomyocytes. Like MG132, a Na+ channel blocker aprindine prolonged the half-life time of Kir6.2 and augmented KATP. Finally, both aprindine and MG132 inhibited the 20S proteasome activity in vitro. These results suggest a novel activity of aprindine to enhance KATP currents by inhibiting proteasomal degradation of Kir 6.2 channels, which may be beneficial in the setting of cardiac ischemia. Ó 2005 Elsevier Inc. All rights reserved. Keywords: Kir6.2; Proteasome; Aprindine
*
Corresponding author. Fax: +81 859 34 8099. E-mail address: [email protected] (I. Hisatome).
0006-291X/$ - see front matter Ó 2005 Elsevier Inc. All rights reserved. doi:10.1016/j.bbrc.2005.04.011
ATP-sensitive K+ channels (KATP channels) are a group of inwardly rectifying K+ channels expressed in various tissues/cell types involved in diverse physiological functions [1,2]. Cardiac KATP channels are
1002
H. Tanaka et al. / Biochemical and Biophysical Research Communications 331 (2005) 1001–1006
heteromultimers composed of at least two structurally distinct subunits. The pore-forming inwardly rectifying K+ channel core, Kir6.2, is primarily responsible for K+ permeance, whereas the regulatory subunit, also known as the sulfonylurea receptor (SUR) 2A, is implicated in ligand-dependent channel gating [3,4]. These channels have been shown to exert protective effects against ischemic and reperfusion damage. The protein levels of channel proteins expressed in individual cells are controlled by various mechanisms. While almost all previous studies focused on the transcriptional control, a few studies suggested the post-translational control of the Kir6.2 protein levels: for example, its stability could be modulated both by adenosine and PKC activity [5]. Its intracellular localization is also important, since mutations of SUR1 that impaired Kir6.2 trafficking to the cell surface caused persistent hyperinsulinemic hypoglycemia of infant [6]. It has been reported that channel proteins with relatively short half-life times, such as an epithelial Na+ channel [7], a cystic fibrosis transmembrane regulator [8], and a gap junction channel connexin 43 [9], are degraded through the ubiquitin–proteasome pathway. In hearts, Na+ channels are a target of ubiquitin–protein ligase Nedd4 [10] and a mutant Na+ channel protein M1766L undergoes rapid degradation and causes Brugada syndrome [11]. Interestingly, a class 1b Na+ channel blocker mexiletine has been reported to stabilize this mutant Na+ channel as well as wild-type channels [12]. The initial purpose of the present study was to determine whether Kir6.2 was degraded through the ubiquitin–proteasome pathway. Since we obtained pharmacological evidence for Kir6.2 degradation through this pathway, we further tested an activity of a class 1b Na+ channel blocker aprindine to modulate this degradation process.
Methods Plasmids and expression. An expression construct pRC/Kir6.2FLAG was engineered by ligating an oligonucleotide encoding a FLAG epitope to the carboxy terminus of rat Kir6.2 cDNA. An expression construct pcDNA/SUR2A was described [3]. COS cells were maintained in DulbeccoÕs modified EagleÕs medium (GibcoBRL)/10% fetal bovine serum at 37 °C in a 5% CO2 incubator. Cells were transfected by using Lipofectamine (Gibco-BRL) according to the manufacturerÕs instructions. Forty-eight hours after transfection, cells were subjected to assays. Proteasome inhibitors or antiarrhythmic agent was applied 36 h after transfection. All drugs except for MG132 were dissolved in buffer. MG132 was dissolved in DMSO. The final concentration of DMSO in the culture or reaction medium was equal to or less than 0.01% v/v. Aprindine was kindly provided by Nippon Shering (Tokyo, Japan). Western blotting and immunoprecipitation. Cells were scraped into lysis buffer (PBS/1% NP40, 0.5% sodium deoxycholate, 0.1% SDS, 10 lg/ml aprotinin, 10 lg/ml leupeptin, 10 lg/ml pepstatin, and 1 mM phenylmethylsulfonyl fluoride), lysed by sonication, and insoluble materials were removed by centrifugation. Protein concentrations were
determined with a BCA protein assay kit (Piearce). The same amounts of protein (10 lg) were separated on SDS–PAGE and electrotransferred to a PVDF membrane. Membranes were probed with antibodies against FLAG (1:1000, Cosmo Bio), GFP (1:4000, Molecular Probes), ubiquitin (1:1000, MBL) or b-actin (1:5000, Oncogene) and were developed by using an ECL system. Immunoprecipitation was carried out in PBS/1% Triton X-100, 0.5% SDS, 0.25% sodium deoxycholate, 1 mM EDTA, and protease inhibitors for 2 h at 4 °C. Immunocomplexes were collected with protein G–agarose (Pharmacia) and bound proteins were analyzed by SDS–PAGE followed by immunoblotting. Pulse-chase analysis. Cells were pulse-labeled for 2 h in methionine-free DMEM supplemented with [35S]methionine (3.7 lCi/ml: Amersham) and then chased in DMEM supplemented with 1 mM methionine. Where indicated, proteasome inhibitors or aprindine was included both in pulse and chase media. Anti-FLAG precipitation was carried out as described above and bound proteins were analyzed by SDS–PAGE followed by autoradiography. The band intensities were quantified by using an NIH image software. The decay constant (k) was estimated by fitting first-order decay curves to the form y = e kt, by using SigmaPlot (Jandel Scientific). A halflife time was calculated using a formula t1/2 = 0.693/k [13]. Immunofluorescence. Cells were transfected with pRC/Kir6.2FLAG. Anti-FLAG staining of fixed cells was performed as described [14] by using Texas red-conjugated anti-mouse IgG as a secondary antibody and images were collected with a BioRad MRC1024 confocal microscope. Preparation of cardiomyocytes. Cardiomyocytes were isolated from neonatal rat as described [15]. In brief, cardiac cells were enzymatically dissociated and seeded in serum-free DMEM containing bovine serum albumin. After incubation for 15 min, cells that adhered to the plate (mostly fibroblasts and endothelial cells) were discarded and myocytes were further cultured in new dishes. Cells were cultured for 12 h in the presence or absence of indicated drugs. Electrophysiological recordings. Cells were co-transfected with pRC/Kir6.2-FLAG, pcDNA3/SUR2A, and pEGFP C1 (Clontech). Transfected cells were visualized by EGFP fluorescence and subjected to whole-cell voltage clamp experiments. Briefly, currents were elicited every 6 s by 500 ms step pulses from a holding potential of 60 mV to test potentials of 40 to +60 mV at 20 mV intervals. KATP channel currents were defined as the nicorandil-activated current; the current recorded in the absence of nicorandil was subtracted from that recorded in the presence of 1 mM nicorandil. Measurement of 20S proteasome activity. By using an assay kit (Calbiochem), 20S proteasome activity was assessed by fluorescence of free AMC (7-amino-4-methylcoumarin, excitation: 380 nm, emission: 460 nm) liberated from a substrate peptide (Suc-Leu-Leu-Val-TyrAMC). The reaction mixture contained 20S proteasome (500 lg/ml), the substrate peptide (10 lM), and indicated drugs in a buffer (25 mM Hepes, 0.5 mM EDTA, and 0.03% SDS, pH 7.6). The mixture was incubated at 37 °C for 1 h. AMC fluorescence liberated in the absence of drugs was taken as the basal value (100%). Because the proteasome in this reaction is activated by SDS, the fluorescence liberated in the absence of SDS was taken as the background value (0%).
Results Kir6.2 channels are ubiquitinated and degraded by the proteasome Kir6.2-FLAG expressed in COS cells were detected by anti-FLAG Western blotting of cell lysates at 40 kDa a predicted size of full-length Kir6.2-FLAG. Treatment with MG132 (50 lM for 12 h) increased the density of this band (Fig. 1A). The same treatment did
H. Tanaka et al. / Biochemical and Biophysical Research Communications 331 (2005) 1001–1006
1003
Fig. 1. Effects of MG132 on Kir6.2-FLAG in expressed COS cells. In (A–C) cells were transfected with pRC/Kir6.2-FLAG plus/minus pEGFP and treated with MG132 (50 lM) or vehicle (0.01% DMSO) for 12 h. Ten microgram proteins were loaded in each lane. (A) Effects on the protein level. Cell lysates were subjected to Western blotting with indicated antibodies. (B) Effects on protein stability. pRC/Kir6.2-FLAG-transfected cells were pulse-labeled with [35S]methionine for 2 h and chased for indicated times. Both pulse- and chase media contained 50 lM MG132 or solvent (0.01% DMSO). (C) Effects on ubiquitination. Anti-FLAG immunoprecipitates (IP) from cell lysates were subjected to immunoblotting (IB) with indicated antibody. Molecular weights are indicated on the left (kDa). (D) Immunofluorescence of Kir6.2-FLAG. Cells were transfected with Kir6.2-FLAG together with Golgi-YFP or ER-YFP. They were treated with MG132 (5 lM) or vehicle (0.01% DMSO) for 12 h, fixed, and stained with anti-FLAG. Shown are the representative images obtained with a confocal microscope. All these experiments were repeated more than 4 times with similar results.
not affect the level of co-expressed GFP (Fig. 1A), indicating similar transfection efficiencies between drugtreated and untreated cells as well as a specificity of the effect. We found complete extraction of the proteins in our lysis buffer, excluding drug-induced changes in protein solubility (data not shown). To confirm MG132-induced stabilization of Kir6.2FLAG, we determined a half-life time (t1/2) of expressed proteins by pulse-chase analyses. Kir6.2-FLAG had a t1/2 of 1.43 ± 0.4 h (n = 4), and this value was prolonged to 4.3 ± 2.4 h (n = 4) in MG132-treated cells (Fig. 1B). Other proteasome inhibitors lactacystine (25 lM), proteasome inhibitor 1 (100 lM), and ALLN (20 lM) had a similar effect, whereas a lysosome inhibitor chloroquine (2 mM) had no effect (data not shown). There were similar levels of 35S-labeled proteins between drug-treated and untreated cells at the end of the pulse period (time 0 point in Fig. 1B), excluding a significant effect of MG132 on protein synthesis. Anti-ubiquitin blotting of anti-FLAG precipitates revealed a marked increase of ubiquitinated Kir6.2-FLAG in MG132treated cells (Fig. 1C).
Kir6.2 channels are mainly localized in the endoplasmic reticulum (ER) and Gogi apparatus Next, we examined the intracellular localization of Kir6.2-FLAG using ER-YFP and Golgi-YFP as markers for the ER and Golgi apparatus, respectively (Fig. 1D). Kir6.2-FLAG co-localized both with Golgi-YFP and ER-YFP (upper panels), indicating that Kir6.2 channels are mainly localized in the ER and Golgi apparatus. Pretreatment with MG132 (50 lM) did not affect this localization of Kir6.2, but caused an apparent increase in the immunoreactivity both in the ER and Golgi apparatus (lower panels). MG132-induced increase of reconstituted and native KATP channel currents To see whether MG132 altered the levels of cell-surface Kir6.2-FLAG, we examined nicorandil-induced KATP currents. In COS7 cells expressing both Kir6.2-FLAG and SUR2A (Fig. 2A, upper panel), nicorandil (1 mM) increased the outward currents elicited by the depolarizing
1004
H. Tanaka et al. / Biochemical and Biophysical Research Communications 331 (2005) 1001–1006
Fig. 2. Effects of MG132 on KATP channel currents in COS cells expressing both SUR2A and Kir6.2-FLAG, and neonatal rat cardiac myocytes. (A) Representative current traces from the cells transfected with pRC/Kir6.2-FLAG and SUR2A, and then treated with MG132 (50 lM) or vehicle (0.01% DMSO) for 12 h. The currents were elicited by 300 ms test potential to 0 mV from a holding potential of 60 mV before (open circles) and during perfusion (closed circles) with 1 mM nicorandil in the vehicle treated cells (upper panel) and in MG132 treated cells (lower panel). (B) The averaged amplitudes of the KATP channel currents-induced by 1 mM nicorandil (at 0 mV) in the vehicle treated cells and in MG132 treated cells (n = 8–14). (C) Representative traces of the currents induced by nicorandil from neonatal rat cardiomyocytes treated with MG132 (50 lM) or vehicle (0.01% DMSO) for 12 h. The currents were elicited by 300 ms test pulses to 0 mV from a holding potential of 60 mV before (open circles) and during perfusion (closed circles) with 1 mM nicorandil in the vehicle treated cells (upper panel) and in MG132 treated cells (lower panel). (D) The averaged KATP channel currents induced by 1 mM nicorandil (at 0 mV) in the vehicle treated neonatal cardiomyocytes and in MG132 treated cells (n = 5–10).
pulse (0 mV) with a slight outward shift of the holding current (at 60 mV). This nicorandil-induced component was considered to represent KATP currents. The current recordings were repeated by using native COS cells and those transfected with an empty vector and we found no nicorandil-induced currents, indicating that observed KATP currents totally derived from expressed Kir6.2FLAG and SUR2A. MG132 treatment (50 lM for 12 h) caused obvious increases in nicorandil-induced currents at 0 mV with an outward increase of the holding current at 60 mV (Fig. 2A, lower panel). MG132 also increased the amplitude of nicorandil-induced KATP currents at the test potential of 0 mV, (Fig. 2B). MG132 treatment (50 lM for 12 h) had similar effects in primary cultured neonatal cardiomyocytes (Fig. 2C, lower panel). There was also a significant increase in the amplitude of nicorandil-induced currents at the test potential of 0 mV (Fig. 2D).
nel proteins [12], we tested the ability of aprindine, a class 1b antiarrhythmic agent to modify Kir6.2 degradation. Western blotting revealed increased levels of Kir6.2-FLAG in cells treated for 12 h with aprindine (10 lM), and unaltered levels of co-expressed GFP (Fig. 3A). Pulse-chase analyses showed that aprindine prolonged the half-life time of Kir6.2-FLAG from 1.43 to 3.23 h (Fig. 3B). Anti-ubiquitin blotting of anti-FLAG precipitates revealed increased levels of ubiquitinated Kir6.2-FLAG in aprindine-treated cells (Fig. 3C). In COS cells expressing both Kir6.2-FLAG and SUR2A, aprindine treatment (10 lM for 12 h) increased the nicorandil-induced currents at the test potential of 0 mV (Figs. 3D and E). This effect depended on preincubation of the cells with the drug, because it caused no effect when applied immediately before the current recording (data not shown). Inhibition of 20S proteasome activity by aprindine in vitro
Aprindine stabilized Kir6.2-FLAG and increased KATP currents Because it has been reported that some class 1b antiarrhythmic agents modulate the stability of Na+ chan-
To obtain direct evidence for inhibition of the proteasome by aprindine, we tested whether aprindine inhibits in vitro activity of 20S proteasome, which acts as a catalytic core of the 26S proteasome complex. As expected,
H. Tanaka et al. / Biochemical and Biophysical Research Communications 331 (2005) 1001–1006
1005
Fig. 3. Effects of aprindine on Kir6.2-FLAG in expressed COS cells. In A, B, and C, cells were transfected with pRC/Kir6.2-FLAG plus/minus pEGFP and treated with aprindine (20 lM) or vehicle (0.01% DMSO) for 12 h. Ten microgram proteins were loaded in each lane. (A) Effects on the protein level. Cell lysates were subjected to Western blotting with indicated antibodies. (B) Effects on protein stability. pRC/Kir6.2-FLAGtransfected cells were pulse-labeled with [35S]methionine for 2 h and chased for indicated times. Both pulse- and chase media contained with or without aprindine (20 lM). (C) Effects on ubiquitination. Anti-FLAG immunoprecipitates (IP) from cell lysates were subjected to immunoblotting (IB) with indicated antibody. Molecular weights are indicated on the left (kDa). (D) Representative current traces of cells transfected with pRC/ Kir6.2-FLAG and SUR2A, and treated with or without aprindine (20 lM) for 12 h. The currents were elicited by 300 ms test potential to 0 mV from a holding potential of 60 mV before (open circles) and during perfusion (closed circles) with 1 mM nicorandil in the vehicle treated cells (upper panel) and in aprindine treated cells (lower panel). (E) The averaged KATP channel currents induced by 1 mM nicorandil in the vehicle treated cells and in aprindine treated cells (n = 14–15).
MG132 caused almost complete inhibition of the 20S proteasome activity with an IC50 value of 4.1 ± 2.4 nM and with a Hill value of 1.0 ± 0.17. Aprindine also inhibited the activity; the maximum effect of aprindine was 60% of that of MG132, and its IC50 and Hill values were 4.0 ± 0.9 lM and 1.1 ± 0.19, respectively (Fig. 4).
Discussion
Fig. 4. Effects of MG132 and aprindine on the 20S proteasome activity in vitro. Each point represents mean ± SEM of more than 7 determinations. 20S proteasome activities were expressed as relative to the values in the absence of drugs (100%).
The present results indicate that Kir6.2-FLAG was degraded in the cytosol by proteasome. This conclusion was drawn from the following observations: (1) Kir6.2FLAG had a relatively short half-life and a part of the protein was ubiquitinated, (2) a proteasome inhibitor MG132 prolonged the half-life time and increased the steady state levels of the protein as well as ubiquitinated
1006
H. Tanaka et al. / Biochemical and Biophysical Research Communications 331 (2005) 1001–1006
form, and (3) similar effects were observed with other proteasome inhibitors lactacystine, ALLN, and PI but not with a lysosomal inhibitor chloroquine. Both immature proteins localized in the ER/Golgi apparatus and mature proteins on the cell-surface can be a target of proteasomal degradation. The main site of MG132 action appeared to be the ER/Golgi apparatus, because MG132 increased the level of proteins in these compartments, without affecting the localization pattern. This notion is in accordance with the finding that when Kir6.2 was expressed without SUR, this protein could not reach the cell-surface and was retained in the ER/Golgi [4]. In cells transfected by both Kir6.2FLAG and SUR2A, MG132 increased the amplitude of nicorandil-induced KATP currents, most likely because of an increase of cell-surface channel proteins secondary to their stabilization in the ER. The most prominent finding of the present study was that a class 1b Na+ channel blocker aprindine, by mimicking the action of MG123, stabilized Kir6.2-FLAG and increased nicorandil-induced KATP currents. The in vitro effects on 20S proteasome suggested a direct action of aprindine on the proteasome catalytic subunit. It has been shown that a class 1b antiarrhythmic agent mexiletine blocked the mutant Na+ channel as well as the wild-type channels and stabilized their channel proteins [15]. The mode of action of aprindine on Kir6.2 channels was distinct from that of mexiletine on Na+ channels, since aprindine did not inhibit Kir6.2 channel activity. Since aprindine was effective within a range of clinical concentrations, the effect on the stability of Kir6.2 channel stability may have a clinical relevance. The present findings raised a possibility that aprindine shortens the cardiac action potential duration of ischemic cardiomyocytes or nicorandil-treated cardiomyocytes via increasing KATP channel currents. This may be an adverse effect of the drug by helping generation of reentrant arrhythmias. However, the same effect of aprindine may be beneficial for ischemic or failing heart by reducing the intracellular Ca2+ overload. In conclusion, Kir6.2 channels are ubiquitinated and degraded by the proteasome. A Na+ channel blocker aprindine, by mimicking the action of MG132, could stabilize Kir6.2 channels which may pave a novel pharmacological approach to activate cardiac KATP for protecting the ischemic injury of the heart.
References [1] L. Aguilar-Bryan, J.P. Clement, G. Gonzalez, K. Kunjilwar, A. Babenko, J. Bryan, A view of SUR/Kir6.X KATP channels, Physiol. Rev. 78 (1998) 227–245. [2] S.J. Tucker, F.M. Ashcroft, A touching cse of channel regulation: ATP-sensitive K+ channel, Curr. Opin. Neurobiol. 8 (1998) 316– 320. [3] N. Inagaki, T. Gonoi, J.P. Clement, N. Namba, J. Inazawa, G. Gonzalez, L. Aguilar-Bryan, S. Seino, J. Bryan, Reconstitution of IKATP: an inward rectifier subunit plus the sulphonylurea receptor, Science 270 (1995) 1166–1170. [4] N. Zerangue, B. Schwappach, Y.N. Jan, L.Y. Jan, A new ER trafficking signal regulates the subunit stoichiometry of plasma membrane K(ATP) channels, Neuron 22 (1999) 537–548. [5] K. Hu, C.S. Huang, Y.N. Jan, L.Y. Jan, ATP-sensitive potassium channel traffic regulation by adenosine and protein kinase C, Neuron 38 (2004) 417–432. [6] E.A. Cartier, L.R. Conti, C.A. Vandenberg, S.-L. Shyng, Defective trafficking and function of KATP channels caused by a sulfonylurea receptor 1 mutation associated with persistent hyperinsulinemic hypoglycemia of infancy, PNAS 98 (2001) 2882–2887. [7] O. Staub, I. Gautschi, T. Ishikawa, K. Breitschopf, A. Ciechanover, L. Schild, D. Rotin, Regulation of stability and function of the epithelial Na+ channel (ENaC) by ubiquitination, EMBO J. 16 (1997) 6325–6336. [8] C.L. Ward, S. Omura, R.R. Kopito, Degradation of CFTR by the ubiquitin–proteasome pathway, Cell 83 (1995) 121–127. [9] M.A. Beadslee, J.G. Laing, E.C. Beyer, J.E. Sattitz, Rapid turnover of connexin 43 in the adult rat heart, Circ. Res. 83 (1998) 629–635. [10] H. Abriel, E. Kamynina, J.D. Horisberger, O. Staub, Regulation of the cardiac voltage-gated Na+ channel (H1) by the ubiquitinprotein ligase Nedd4, FEBS Lett. 466 (2000) 377–380. [11] J.M. Lupoglazoff, T. Cheav, G. Baroudi, M. Berthet, I. Denjoy, B. Cauchemez, F. Extramiana, M. Chahine, P. Guicheney, Homozygous SCN5A mutation in long-QT syndrome with functional two-to-one atrioventricular block, Circ. Res. 89 (2001) 16–21. [12] C.R. Valdivia, M.J. Ackerman, D.J. Tester, T. Wada, J. McCormack, B. Ye, J.C. Makielski, A novel SCN5A arrhythmia mutation, M1766L, with expression defect rescued by mexiletine, Cardiovasc. Res. 55 (2002) 229–232. [13] P.J. Garlick, J.C. Waterlow, R.W. Swick, Measurement of protein turnover in rat liver. Analysis of the complex curve for decay of label in a mixture of proteins, Biochem. J. 156 (1976) 657–663. [14] Y. Sugimoto, H. Ninomiya, Y. Ohsaki, K. Higaki, J.P. Davies, Y.A. Ioannou, K. Ohno, Accumulation of cholera toxin and GM1 ganglioside in the early endosome of Niemann–Pick C1deficient cells, Proc. Natl. Acad. Sci. USA 98 (2001) 12391–12396. [15] C.R. Valdivia, D.J. Tester, B.A. Rok, C.B. Porter, T.M. Munger, A. Jahangir, J.C. Makielski, M.J.A. Ackerman, A trafficking defective, Brugada syndrome-causing SCN5A mutation rescued by drugs, Cardiovasc. Res. 62 (2004) 53–62.
Proteasomal degradation of Kir6.2 channel protein and its inhibition by a Na+ channel blocker aprindine Hiroaki Tanaka a, Junichiro Miake b, Tomomi Notsu b, Kazuhiko Sonyama a, Norihito Sasaki b, Kazuhiko Iitsuka a, Masaru Kato a, Shin-ichi Taniguchi a, Osamu Igawa a, Akio Yoshida a, Chiaki Shigemasa a, Yoshiko Hoshikawa b, Yasutaka Kurata c, Akihiko Kuniyasu d, Hitoshi Nakayama d, Nobuo Inagaki e, Eiji Nanba f, Goshi Shiota g, Takayuki Morisaki h, Haruaki Ninomiya i, Masafumi Kitakaze j, Ichiro Hisatome b,* a
Division of Cardiovascular Medicine, The First Department of Internal Medicine, Tottori University Faculty of Medicine, Yonago 683-8504, Japan b Division of Regenerative Medicine and Therapeutics, Department of Gene Therapy and Regenerative Medicine, Tottori University Graduate School of Medical Science, Japan c Department of Physiology, Kanazawa Medical University, Kanazawa, Japan d Department of Medical Biochemistry, Graduate School of Medical and Pharmaceutical Sciences, Kumamoto University, Kumamoto, Japan e Department of Physiology, Akita University School of Medicine, Akita 010-8543, Japan f Division of Functional Genomics, Research Center for Bioscience and Technology, Tottori University, Yonago, Japan g Division of Molecular and Genetic Medicine, Department of Genetic Medicine and Regenerative Therapeutics, Faculty of Medicine, Tottori University, Japan h Department of Bioscience, National Cardiovascular Center Research Institute, Osaka, Japan i Department of Neurobiology, Tottori University Faculty of Medicine, Yonago, Japan j Department of Cardiology, National Cardiovascular Center, Osaka, Japan Received 25 March 2005 Available online 11 April 2005
Abstract ATP-sensitive K+ channels (KATP:SUR2A+Kir6.2) play a pivotal role in cardiac protection against ischemia and reperfusion injury. When expressed in COS cells, Kir6.2 was short-lived with a half-life time of 1.9 h. The half-life time of Kir6.2 was prolonged by proteasome inhibitors MG132, ALLN, proteasome inhibitor 1, and lactacystine, but not at all by a lysosomal inhibitor chloroquine. MG132 also increased the level of ubiquitinated Kir6.2 without affecting its localization in the endoplasmic reticulum and Golgi apparatus. In electrophysiological recordings, MG132 augmented nicorandil-activated KATP currents in COS cells expressing SUR2A and Kir6.2 as well as the same currents in neonatal rat cardiomyocytes. Like MG132, a Na+ channel blocker aprindine prolonged the half-life time of Kir6.2 and augmented KATP. Finally, both aprindine and MG132 inhibited the 20S proteasome activity in vitro. These results suggest a novel activity of aprindine to enhance KATP currents by inhibiting proteasomal degradation of Kir 6.2 channels, which may be beneficial in the setting of cardiac ischemia. Ó 2005 Elsevier Inc. All rights reserved. Keywords: Kir6.2; Proteasome; Aprindine
*
Corresponding author. Fax: +81 859 34 8099. E-mail address: [email protected] (I. Hisatome).
0006-291X/$ - see front matter Ó 2005 Elsevier Inc. All rights reserved. doi:10.1016/j.bbrc.2005.04.011
ATP-sensitive K+ channels (KATP channels) are a group of inwardly rectifying K+ channels expressed in various tissues/cell types involved in diverse physiological functions [1,2]. Cardiac KATP channels are
1002
H. Tanaka et al. / Biochemical and Biophysical Research Communications 331 (2005) 1001–1006
heteromultimers composed of at least two structurally distinct subunits. The pore-forming inwardly rectifying K+ channel core, Kir6.2, is primarily responsible for K+ permeance, whereas the regulatory subunit, also known as the sulfonylurea receptor (SUR) 2A, is implicated in ligand-dependent channel gating [3,4]. These channels have been shown to exert protective effects against ischemic and reperfusion damage. The protein levels of channel proteins expressed in individual cells are controlled by various mechanisms. While almost all previous studies focused on the transcriptional control, a few studies suggested the post-translational control of the Kir6.2 protein levels: for example, its stability could be modulated both by adenosine and PKC activity [5]. Its intracellular localization is also important, since mutations of SUR1 that impaired Kir6.2 trafficking to the cell surface caused persistent hyperinsulinemic hypoglycemia of infant [6]. It has been reported that channel proteins with relatively short half-life times, such as an epithelial Na+ channel [7], a cystic fibrosis transmembrane regulator [8], and a gap junction channel connexin 43 [9], are degraded through the ubiquitin–proteasome pathway. In hearts, Na+ channels are a target of ubiquitin–protein ligase Nedd4 [10] and a mutant Na+ channel protein M1766L undergoes rapid degradation and causes Brugada syndrome [11]. Interestingly, a class 1b Na+ channel blocker mexiletine has been reported to stabilize this mutant Na+ channel as well as wild-type channels [12]. The initial purpose of the present study was to determine whether Kir6.2 was degraded through the ubiquitin–proteasome pathway. Since we obtained pharmacological evidence for Kir6.2 degradation through this pathway, we further tested an activity of a class 1b Na+ channel blocker aprindine to modulate this degradation process.
Methods Plasmids and expression. An expression construct pRC/Kir6.2FLAG was engineered by ligating an oligonucleotide encoding a FLAG epitope to the carboxy terminus of rat Kir6.2 cDNA. An expression construct pcDNA/SUR2A was described [3]. COS cells were maintained in DulbeccoÕs modified EagleÕs medium (GibcoBRL)/10% fetal bovine serum at 37 °C in a 5% CO2 incubator. Cells were transfected by using Lipofectamine (Gibco-BRL) according to the manufacturerÕs instructions. Forty-eight hours after transfection, cells were subjected to assays. Proteasome inhibitors or antiarrhythmic agent was applied 36 h after transfection. All drugs except for MG132 were dissolved in buffer. MG132 was dissolved in DMSO. The final concentration of DMSO in the culture or reaction medium was equal to or less than 0.01% v/v. Aprindine was kindly provided by Nippon Shering (Tokyo, Japan). Western blotting and immunoprecipitation. Cells were scraped into lysis buffer (PBS/1% NP40, 0.5% sodium deoxycholate, 0.1% SDS, 10 lg/ml aprotinin, 10 lg/ml leupeptin, 10 lg/ml pepstatin, and 1 mM phenylmethylsulfonyl fluoride), lysed by sonication, and insoluble materials were removed by centrifugation. Protein concentrations were
determined with a BCA protein assay kit (Piearce). The same amounts of protein (10 lg) were separated on SDS–PAGE and electrotransferred to a PVDF membrane. Membranes were probed with antibodies against FLAG (1:1000, Cosmo Bio), GFP (1:4000, Molecular Probes), ubiquitin (1:1000, MBL) or b-actin (1:5000, Oncogene) and were developed by using an ECL system. Immunoprecipitation was carried out in PBS/1% Triton X-100, 0.5% SDS, 0.25% sodium deoxycholate, 1 mM EDTA, and protease inhibitors for 2 h at 4 °C. Immunocomplexes were collected with protein G–agarose (Pharmacia) and bound proteins were analyzed by SDS–PAGE followed by immunoblotting. Pulse-chase analysis. Cells were pulse-labeled for 2 h in methionine-free DMEM supplemented with [35S]methionine (3.7 lCi/ml: Amersham) and then chased in DMEM supplemented with 1 mM methionine. Where indicated, proteasome inhibitors or aprindine was included both in pulse and chase media. Anti-FLAG precipitation was carried out as described above and bound proteins were analyzed by SDS–PAGE followed by autoradiography. The band intensities were quantified by using an NIH image software. The decay constant (k) was estimated by fitting first-order decay curves to the form y = e kt, by using SigmaPlot (Jandel Scientific). A halflife time was calculated using a formula t1/2 = 0.693/k [13]. Immunofluorescence. Cells were transfected with pRC/Kir6.2FLAG. Anti-FLAG staining of fixed cells was performed as described [14] by using Texas red-conjugated anti-mouse IgG as a secondary antibody and images were collected with a BioRad MRC1024 confocal microscope. Preparation of cardiomyocytes. Cardiomyocytes were isolated from neonatal rat as described [15]. In brief, cardiac cells were enzymatically dissociated and seeded in serum-free DMEM containing bovine serum albumin. After incubation for 15 min, cells that adhered to the plate (mostly fibroblasts and endothelial cells) were discarded and myocytes were further cultured in new dishes. Cells were cultured for 12 h in the presence or absence of indicated drugs. Electrophysiological recordings. Cells were co-transfected with pRC/Kir6.2-FLAG, pcDNA3/SUR2A, and pEGFP C1 (Clontech). Transfected cells were visualized by EGFP fluorescence and subjected to whole-cell voltage clamp experiments. Briefly, currents were elicited every 6 s by 500 ms step pulses from a holding potential of 60 mV to test potentials of 40 to +60 mV at 20 mV intervals. KATP channel currents were defined as the nicorandil-activated current; the current recorded in the absence of nicorandil was subtracted from that recorded in the presence of 1 mM nicorandil. Measurement of 20S proteasome activity. By using an assay kit (Calbiochem), 20S proteasome activity was assessed by fluorescence of free AMC (7-amino-4-methylcoumarin, excitation: 380 nm, emission: 460 nm) liberated from a substrate peptide (Suc-Leu-Leu-Val-TyrAMC). The reaction mixture contained 20S proteasome (500 lg/ml), the substrate peptide (10 lM), and indicated drugs in a buffer (25 mM Hepes, 0.5 mM EDTA, and 0.03% SDS, pH 7.6). The mixture was incubated at 37 °C for 1 h. AMC fluorescence liberated in the absence of drugs was taken as the basal value (100%). Because the proteasome in this reaction is activated by SDS, the fluorescence liberated in the absence of SDS was taken as the background value (0%).
Results Kir6.2 channels are ubiquitinated and degraded by the proteasome Kir6.2-FLAG expressed in COS cells were detected by anti-FLAG Western blotting of cell lysates at 40 kDa a predicted size of full-length Kir6.2-FLAG. Treatment with MG132 (50 lM for 12 h) increased the density of this band (Fig. 1A). The same treatment did
H. Tanaka et al. / Biochemical and Biophysical Research Communications 331 (2005) 1001–1006
1003
Fig. 1. Effects of MG132 on Kir6.2-FLAG in expressed COS cells. In (A–C) cells were transfected with pRC/Kir6.2-FLAG plus/minus pEGFP and treated with MG132 (50 lM) or vehicle (0.01% DMSO) for 12 h. Ten microgram proteins were loaded in each lane. (A) Effects on the protein level. Cell lysates were subjected to Western blotting with indicated antibodies. (B) Effects on protein stability. pRC/Kir6.2-FLAG-transfected cells were pulse-labeled with [35S]methionine for 2 h and chased for indicated times. Both pulse- and chase media contained 50 lM MG132 or solvent (0.01% DMSO). (C) Effects on ubiquitination. Anti-FLAG immunoprecipitates (IP) from cell lysates were subjected to immunoblotting (IB) with indicated antibody. Molecular weights are indicated on the left (kDa). (D) Immunofluorescence of Kir6.2-FLAG. Cells were transfected with Kir6.2-FLAG together with Golgi-YFP or ER-YFP. They were treated with MG132 (5 lM) or vehicle (0.01% DMSO) for 12 h, fixed, and stained with anti-FLAG. Shown are the representative images obtained with a confocal microscope. All these experiments were repeated more than 4 times with similar results.
not affect the level of co-expressed GFP (Fig. 1A), indicating similar transfection efficiencies between drugtreated and untreated cells as well as a specificity of the effect. We found complete extraction of the proteins in our lysis buffer, excluding drug-induced changes in protein solubility (data not shown). To confirm MG132-induced stabilization of Kir6.2FLAG, we determined a half-life time (t1/2) of expressed proteins by pulse-chase analyses. Kir6.2-FLAG had a t1/2 of 1.43 ± 0.4 h (n = 4), and this value was prolonged to 4.3 ± 2.4 h (n = 4) in MG132-treated cells (Fig. 1B). Other proteasome inhibitors lactacystine (25 lM), proteasome inhibitor 1 (100 lM), and ALLN (20 lM) had a similar effect, whereas a lysosome inhibitor chloroquine (2 mM) had no effect (data not shown). There were similar levels of 35S-labeled proteins between drug-treated and untreated cells at the end of the pulse period (time 0 point in Fig. 1B), excluding a significant effect of MG132 on protein synthesis. Anti-ubiquitin blotting of anti-FLAG precipitates revealed a marked increase of ubiquitinated Kir6.2-FLAG in MG132treated cells (Fig. 1C).
Kir6.2 channels are mainly localized in the endoplasmic reticulum (ER) and Gogi apparatus Next, we examined the intracellular localization of Kir6.2-FLAG using ER-YFP and Golgi-YFP as markers for the ER and Golgi apparatus, respectively (Fig. 1D). Kir6.2-FLAG co-localized both with Golgi-YFP and ER-YFP (upper panels), indicating that Kir6.2 channels are mainly localized in the ER and Golgi apparatus. Pretreatment with MG132 (50 lM) did not affect this localization of Kir6.2, but caused an apparent increase in the immunoreactivity both in the ER and Golgi apparatus (lower panels). MG132-induced increase of reconstituted and native KATP channel currents To see whether MG132 altered the levels of cell-surface Kir6.2-FLAG, we examined nicorandil-induced KATP currents. In COS7 cells expressing both Kir6.2-FLAG and SUR2A (Fig. 2A, upper panel), nicorandil (1 mM) increased the outward currents elicited by the depolarizing
1004
H. Tanaka et al. / Biochemical and Biophysical Research Communications 331 (2005) 1001–1006
Fig. 2. Effects of MG132 on KATP channel currents in COS cells expressing both SUR2A and Kir6.2-FLAG, and neonatal rat cardiac myocytes. (A) Representative current traces from the cells transfected with pRC/Kir6.2-FLAG and SUR2A, and then treated with MG132 (50 lM) or vehicle (0.01% DMSO) for 12 h. The currents were elicited by 300 ms test potential to 0 mV from a holding potential of 60 mV before (open circles) and during perfusion (closed circles) with 1 mM nicorandil in the vehicle treated cells (upper panel) and in MG132 treated cells (lower panel). (B) The averaged amplitudes of the KATP channel currents-induced by 1 mM nicorandil (at 0 mV) in the vehicle treated cells and in MG132 treated cells (n = 8–14). (C) Representative traces of the currents induced by nicorandil from neonatal rat cardiomyocytes treated with MG132 (50 lM) or vehicle (0.01% DMSO) for 12 h. The currents were elicited by 300 ms test pulses to 0 mV from a holding potential of 60 mV before (open circles) and during perfusion (closed circles) with 1 mM nicorandil in the vehicle treated cells (upper panel) and in MG132 treated cells (lower panel). (D) The averaged KATP channel currents induced by 1 mM nicorandil (at 0 mV) in the vehicle treated neonatal cardiomyocytes and in MG132 treated cells (n = 5–10).
pulse (0 mV) with a slight outward shift of the holding current (at 60 mV). This nicorandil-induced component was considered to represent KATP currents. The current recordings were repeated by using native COS cells and those transfected with an empty vector and we found no nicorandil-induced currents, indicating that observed KATP currents totally derived from expressed Kir6.2FLAG and SUR2A. MG132 treatment (50 lM for 12 h) caused obvious increases in nicorandil-induced currents at 0 mV with an outward increase of the holding current at 60 mV (Fig. 2A, lower panel). MG132 also increased the amplitude of nicorandil-induced KATP currents at the test potential of 0 mV, (Fig. 2B). MG132 treatment (50 lM for 12 h) had similar effects in primary cultured neonatal cardiomyocytes (Fig. 2C, lower panel). There was also a significant increase in the amplitude of nicorandil-induced currents at the test potential of 0 mV (Fig. 2D).
nel proteins [12], we tested the ability of aprindine, a class 1b antiarrhythmic agent to modify Kir6.2 degradation. Western blotting revealed increased levels of Kir6.2-FLAG in cells treated for 12 h with aprindine (10 lM), and unaltered levels of co-expressed GFP (Fig. 3A). Pulse-chase analyses showed that aprindine prolonged the half-life time of Kir6.2-FLAG from 1.43 to 3.23 h (Fig. 3B). Anti-ubiquitin blotting of anti-FLAG precipitates revealed increased levels of ubiquitinated Kir6.2-FLAG in aprindine-treated cells (Fig. 3C). In COS cells expressing both Kir6.2-FLAG and SUR2A, aprindine treatment (10 lM for 12 h) increased the nicorandil-induced currents at the test potential of 0 mV (Figs. 3D and E). This effect depended on preincubation of the cells with the drug, because it caused no effect when applied immediately before the current recording (data not shown). Inhibition of 20S proteasome activity by aprindine in vitro
Aprindine stabilized Kir6.2-FLAG and increased KATP currents Because it has been reported that some class 1b antiarrhythmic agents modulate the stability of Na+ chan-
To obtain direct evidence for inhibition of the proteasome by aprindine, we tested whether aprindine inhibits in vitro activity of 20S proteasome, which acts as a catalytic core of the 26S proteasome complex. As expected,
H. Tanaka et al. / Biochemical and Biophysical Research Communications 331 (2005) 1001–1006
1005
Fig. 3. Effects of aprindine on Kir6.2-FLAG in expressed COS cells. In A, B, and C, cells were transfected with pRC/Kir6.2-FLAG plus/minus pEGFP and treated with aprindine (20 lM) or vehicle (0.01% DMSO) for 12 h. Ten microgram proteins were loaded in each lane. (A) Effects on the protein level. Cell lysates were subjected to Western blotting with indicated antibodies. (B) Effects on protein stability. pRC/Kir6.2-FLAGtransfected cells were pulse-labeled with [35S]methionine for 2 h and chased for indicated times. Both pulse- and chase media contained with or without aprindine (20 lM). (C) Effects on ubiquitination. Anti-FLAG immunoprecipitates (IP) from cell lysates were subjected to immunoblotting (IB) with indicated antibody. Molecular weights are indicated on the left (kDa). (D) Representative current traces of cells transfected with pRC/ Kir6.2-FLAG and SUR2A, and treated with or without aprindine (20 lM) for 12 h. The currents were elicited by 300 ms test potential to 0 mV from a holding potential of 60 mV before (open circles) and during perfusion (closed circles) with 1 mM nicorandil in the vehicle treated cells (upper panel) and in aprindine treated cells (lower panel). (E) The averaged KATP channel currents induced by 1 mM nicorandil in the vehicle treated cells and in aprindine treated cells (n = 14–15).
MG132 caused almost complete inhibition of the 20S proteasome activity with an IC50 value of 4.1 ± 2.4 nM and with a Hill value of 1.0 ± 0.17. Aprindine also inhibited the activity; the maximum effect of aprindine was 60% of that of MG132, and its IC50 and Hill values were 4.0 ± 0.9 lM and 1.1 ± 0.19, respectively (Fig. 4).
Discussion
Fig. 4. Effects of MG132 and aprindine on the 20S proteasome activity in vitro. Each point represents mean ± SEM of more than 7 determinations. 20S proteasome activities were expressed as relative to the values in the absence of drugs (100%).
The present results indicate that Kir6.2-FLAG was degraded in the cytosol by proteasome. This conclusion was drawn from the following observations: (1) Kir6.2FLAG had a relatively short half-life and a part of the protein was ubiquitinated, (2) a proteasome inhibitor MG132 prolonged the half-life time and increased the steady state levels of the protein as well as ubiquitinated
1006
H. Tanaka et al. / Biochemical and Biophysical Research Communications 331 (2005) 1001–1006
form, and (3) similar effects were observed with other proteasome inhibitors lactacystine, ALLN, and PI but not with a lysosomal inhibitor chloroquine. Both immature proteins localized in the ER/Golgi apparatus and mature proteins on the cell-surface can be a target of proteasomal degradation. The main site of MG132 action appeared to be the ER/Golgi apparatus, because MG132 increased the level of proteins in these compartments, without affecting the localization pattern. This notion is in accordance with the finding that when Kir6.2 was expressed without SUR, this protein could not reach the cell-surface and was retained in the ER/Golgi [4]. In cells transfected by both Kir6.2FLAG and SUR2A, MG132 increased the amplitude of nicorandil-induced KATP currents, most likely because of an increase of cell-surface channel proteins secondary to their stabilization in the ER. The most prominent finding of the present study was that a class 1b Na+ channel blocker aprindine, by mimicking the action of MG123, stabilized Kir6.2-FLAG and increased nicorandil-induced KATP currents. The in vitro effects on 20S proteasome suggested a direct action of aprindine on the proteasome catalytic subunit. It has been shown that a class 1b antiarrhythmic agent mexiletine blocked the mutant Na+ channel as well as the wild-type channels and stabilized their channel proteins [15]. The mode of action of aprindine on Kir6.2 channels was distinct from that of mexiletine on Na+ channels, since aprindine did not inhibit Kir6.2 channel activity. Since aprindine was effective within a range of clinical concentrations, the effect on the stability of Kir6.2 channel stability may have a clinical relevance. The present findings raised a possibility that aprindine shortens the cardiac action potential duration of ischemic cardiomyocytes or nicorandil-treated cardiomyocytes via increasing KATP channel currents. This may be an adverse effect of the drug by helping generation of reentrant arrhythmias. However, the same effect of aprindine may be beneficial for ischemic or failing heart by reducing the intracellular Ca2+ overload. In conclusion, Kir6.2 channels are ubiquitinated and degraded by the proteasome. A Na+ channel blocker aprindine, by mimicking the action of MG132, could stabilize Kir6.2 channels which may pave a novel pharmacological approach to activate cardiac KATP for protecting the ischemic injury of the heart.
References [1] L. Aguilar-Bryan, J.P. Clement, G. Gonzalez, K. Kunjilwar, A. Babenko, J. Bryan, A view of SUR/Kir6.X KATP channels, Physiol. Rev. 78 (1998) 227–245. [2] S.J. Tucker, F.M. Ashcroft, A touching cse of channel regulation: ATP-sensitive K+ channel, Curr. Opin. Neurobiol. 8 (1998) 316– 320. [3] N. Inagaki, T. Gonoi, J.P. Clement, N. Namba, J. Inazawa, G. Gonzalez, L. Aguilar-Bryan, S. Seino, J. Bryan, Reconstitution of IKATP: an inward rectifier subunit plus the sulphonylurea receptor, Science 270 (1995) 1166–1170. [4] N. Zerangue, B. Schwappach, Y.N. Jan, L.Y. Jan, A new ER trafficking signal regulates the subunit stoichiometry of plasma membrane K(ATP) channels, Neuron 22 (1999) 537–548. [5] K. Hu, C.S. Huang, Y.N. Jan, L.Y. Jan, ATP-sensitive potassium channel traffic regulation by adenosine and protein kinase C, Neuron 38 (2004) 417–432. [6] E.A. Cartier, L.R. Conti, C.A. Vandenberg, S.-L. Shyng, Defective trafficking and function of KATP channels caused by a sulfonylurea receptor 1 mutation associated with persistent hyperinsulinemic hypoglycemia of infancy, PNAS 98 (2001) 2882–2887. [7] O. Staub, I. Gautschi, T. Ishikawa, K. Breitschopf, A. Ciechanover, L. Schild, D. Rotin, Regulation of stability and function of the epithelial Na+ channel (ENaC) by ubiquitination, EMBO J. 16 (1997) 6325–6336. [8] C.L. Ward, S. Omura, R.R. Kopito, Degradation of CFTR by the ubiquitin–proteasome pathway, Cell 83 (1995) 121–127. [9] M.A. Beadslee, J.G. Laing, E.C. Beyer, J.E. Sattitz, Rapid turnover of connexin 43 in the adult rat heart, Circ. Res. 83 (1998) 629–635. [10] H. Abriel, E. Kamynina, J.D. Horisberger, O. Staub, Regulation of the cardiac voltage-gated Na+ channel (H1) by the ubiquitinprotein ligase Nedd4, FEBS Lett. 466 (2000) 377–380. [11] J.M. Lupoglazoff, T. Cheav, G. Baroudi, M. Berthet, I. Denjoy, B. Cauchemez, F. Extramiana, M. Chahine, P. Guicheney, Homozygous SCN5A mutation in long-QT syndrome with functional two-to-one atrioventricular block, Circ. Res. 89 (2001) 16–21. [12] C.R. Valdivia, M.J. Ackerman, D.J. Tester, T. Wada, J. McCormack, B. Ye, J.C. Makielski, A novel SCN5A arrhythmia mutation, M1766L, with expression defect rescued by mexiletine, Cardiovasc. Res. 55 (2002) 229–232. [13] P.J. Garlick, J.C. Waterlow, R.W. Swick, Measurement of protein turnover in rat liver. Analysis of the complex curve for decay of label in a mixture of proteins, Biochem. J. 156 (1976) 657–663. [14] Y. Sugimoto, H. Ninomiya, Y. Ohsaki, K. Higaki, J.P. Davies, Y.A. Ioannou, K. Ohno, Accumulation of cholera toxin and GM1 ganglioside in the early endosome of Niemann–Pick C1deficient cells, Proc. Natl. Acad. Sci. USA 98 (2001) 12391–12396. [15] C.R. Valdivia, D.J. Tester, B.A. Rok, C.B. Porter, T.M. Munger, A. Jahangir, J.C. Makielski, M.J.A. Ackerman, A trafficking defective, Brugada syndrome-causing SCN5A mutation rescued by drugs, Cardiovasc. Res. 62 (2004) 53–62.