Published online before print May 29, 2003, doi: 10.1161/01.RES.0000079028.31393.15
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
© 2003 American Heart Association, Inc.
UltraRapid Communication |
Role of the Cytosolic Chaperones Hsp70 and Hsp90 in Maturation of the Cardiac Potassium Channel hERG
From the Rammelkamp Center for Education and Research (E.F., A.T.D., L.W.), MetroHealth Campus, Case Western Reserve University, Cleveland, Ohio; and Department of Physiology and Biophysics (A.M.B.), Case Western Reserve University, Cleveland, Ohio.
Correspondence to Dr Eckhard Ficker, Rammelkamp Center, MetroHealth Medical Center, 2500 MetroHealth Drive, Cleveland, OH 44109. E-mail eficker{at}metrohealth.org
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The human ether-a-gogo–related gene (hERG) encodes the
subunit of the cardiac potassium current IKr. Several mutations in hERG produce trafficking-deficient channels that may cause hereditary long-QT syndrome and sudden cardiac death. Although hERG currents have been studied extensively, little is known about the proteins involved in maturation and trafficking of hERG. Using immunoprecipitations, we show that the cytosolic chaperones heat shock protein (Hsp) 70 and Hsp90, but not Grp94, interact with hERG wild type (WT) during maturation. The specific Hsp90 inhibitor geldanamycin prevents maturation and increases proteasomal degradation of hERG WT, while reducing hERG currents in heterologous expression systems. In ventricular myocytes, inhibition of Hsp90 also decreases IKr, whereas geldanamycin had no effect on IKs or heterologously expressed Kv2.1 and Kv1.5 currents. Both Hsp90 and Hsp70 interact directly with the core-glycosylated form of hERG WT present in the endoplasmic reticulum but not the fully glycosylated, cell-surface form. For the trafficking-deficient LQT2 mutants, hERG R752W and hERG G601S, interactions with Hsp90 and Hsp70 are increased as both mutants remained tightly associated with Hsp90 and Hsp70 in the endoplasmic reticulum. Incubation at lower temperature for R752W or with the hERG blocker astemizole for G601S dissociates channel-chaperone complexes and restores trafficking. In contrast, nonfunctional but trafficking-competent hERG G628S is released from chaperone complexes during maturation comparable to WT. We conclude that Hsp90 and Hsp70 are crucial for the maturation of hERG WT as well as the retention of trafficking-deficient LQT2 mutants. The full text of this article is available online at http://www.circresaha.org.
Key Words: human ether-a-gogo–related gene • heat shock proteins • chaperones • geldanamycin
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The human ether-a-gogo–related gene (hERG), KCNH2, encodes the
subunit of the cardiac potassium current IKr, the current controlling rapid repolarization of the cardiac action potential in human heart.1 IKr is reduced as a consequence of mutations found in the hERG gene in hereditary human long-QT syndrome type 2 (LQT2), a cardiac disorder characterized by a prolonged QT interval and often associated with life-threatening arrhythmias and sudden cardiac arrest, particularly in young patients.2 Mutations in hERG are "loss of function" mutations producing either malfunctioning channels in the cell membrane or abnormally folded and assembled channels that are retained in the endoplasmic reticulum (ER) by quality control mechanisms (eg, see references3–7).
How cells recognize and retain abnormally folded mutant hERG channels and how they move hERG wild-type (WT) channels along productive folding pathways toward their native conformation is unknown. In mammalian cells, the primary level of quality control for membrane proteins such as hERG is composed of ER-resident and cytoplasmic chaperones that recognize incompletely folded proteins and promote productive folding.8 hERG channels represent a typical potassium channel protein composed of six -helical transmembrane segments, one of which functions as a voltage sensor, and a highly selective ion conduction pathway located in the linker between transmembrane segments S5 and S6.9,10 Predicted topologies of potassium channels suggest only very small portions of the tetrameric hERG protein are exposed to the ER lumen, with large N- and C-terminal domains including the Per, Arnt, and Sim (PAS) and cyclic-nucleotide binding (cNBD) domains projecting into the cytoplasm.11,12 Cytoplasmic chaperones such as heat shock protein (Hsp) 90 and/or Hsp/c 70 are involved in the folding of the cAMP-regulated chloride channel cystic fibrosis transmembrane conductance regulator (CFTR).13–15 We hypothesized that they might be prime candidates for interaction with newly synthesized hERG channels. In the present report, we set out to identify molecular chaperones associated with the following: (1) hERG WT channels during maturation in the ER; (2) LQT2 mutant hERG channels that are retained in the ER as a result of defects in folding and assembly but are rescuable; and (3) with a nonfunctional LQT2 mutant protein that trafficks to the cell surface.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Antibodies
The polyclonal anti-hERG antibody (HERG 519) used in Western blots and for immunostaining was generated in rabbits against a GST fusion protein as described previously.6 We used mouse anti-KDEL antibody, which recognizes the ER-resident chaperones Grp78/BiP and Grp94 as a subcellular ER marker (Stressgen, SPA-827). HA-hERG and HA-ubiquitin were detected using rat anti-HA antibody 3F10 (Roche). The following antibodies were used for Western blotting: rat anti-Hsp90
/ß (Stressgen, SPA-835), mouse anti-Hsp70 (Stressgen, SPA-810), rat anti-Grp94 (Stressgen SPA-850), rabbit anti-calnexin (Stressgen, SPA-860), and mouse anti-actin (Sigma, A4700). Rabbit anti-human Kv1.5 antibody was a gift from B. Wible (Case Western Reserve University, Cleveland, Ohio). For immunoprecipitation, rabbit anti-erg1 (Alomone, APC-016), rabbit anti-hERG (Alomone, APC-062), mouse anti Hsp/Hsc70 (Santa Cruz, SC24), mouse anti-Hsp90/ß (Santa Cruz, SC 13119), and rat anti-Grp94 (Stressgen SPA-850) were used.
Cell Lines
Human embryonic kidney (HEK) 293 cells stably transfected with hERG WT (HEK/hERG WT), hERG R752W HAex, hERG G601S, and hERG G628S were maintained in DMEM, and 10% fetal bovine serum (FBS) plus penicillin/streptomycin/geneticin at 37°C, 5%CO2. The hERG G628S cell line was a gift from C. January (University of Wisconsin, Madison, Wis). Human Kv1.5 (hKv1.5) was stably expressed in mouse L cells. L/hKv1.5 cells were maintained in DMEM, 10% fetal bovine serum (FBS) plus penicillin/streptomycin/geneticin.
Drugs
The following drugs were used in hERG trafficking assays: Geldanamycin (Sigma, G3381), Doxorubicin (Sigma, D1515), Radicicol (Calbiochem, 553400), and Taxol/Paclitaxel (Sigma, T 7402).
Molecular Biology
hERG mutations were generated by overlap extension PCR, verified by sequencing, and subcloned into full-length hERG-pcDNA3.6 An HA-epitope tag of the sequence 436TEEGPPATNSEHYPYDVPDYAVTFEECGY (insertion shown in bold, HA epitope underlined) was inserted in the extracellular S1-S2 loop to generate hERG WT HAex and hERG R752W HAex. The insertion did not change the electrophysiological properties or the trafficking phenotype of these channels. Complementary DNAs for HA- and HIS6-ubiquitin were kindly provided by D. Bohmann, Rochester, Mass. Hsp70 cDNA was a gift from H. Wong, Cincinnati, Ohio.
Western Blot Analysis
In experiments with Hsp70, HEK293 cells were transiently transfected with 1 µg hERG WT cDNA and increasing amounts of Hsp70 cDNA using Lipofectamine/Plus (Invitrogen). In transient hERG-Hsp70 transfections, total cDNA concentration was kept constant at 2 µg using pcDNA3 vector cDNA. Cells were harvested 2 days after transfection for Western blotting. In ubiquitination studies, stable HEK/hERG WT cells were transiently transfected with 1.6 µg of either HA-ubiquitin or HIS-ubiquitin cDNA using Fugene (Roche). Cells were harvested 2 days after transfection. Drugs were diluted in pre-warmed culture medium and added to stably transfected HEK/hERG or L/hKv1.5 cells for 12 to 16 hours (overnight) before Western blot analysis. N-glycosidase F treatments (NEB P0704S) were performed as recommended by the manufacturer. Cells were solubilized for 1 hour at 4°C in lysis buffer (50 mmol/L Tris-HCl [pH7.5], 150 mmol/L NaCl, 1 mmol/L EDTA, 1% Triton X-100), containing protease inhibitor mix (Complete, Roche Biochemicals). Protein concentrations were determined by the BCA method (Pierce). Proteins were separated on 7.5% SDS polyacrylamide gels and transferred to polyvinylidene diflouride membranes. Membranes were blocked overnight with 5% nonfat dry milk in phosphate-buffered saline (PBS) plus 0.1% Tween and immunoblotted with suitable primary antibodies followed by horseradish peroxidase–conjugated secondary antibody (1:3000; 1 hour at RT; Amersham Pharmacia Biotech). ECL Plus (Amersham Pharmacia Biotech) was used to develop Western blots with chemiluminescence being captured directly on a Storm PhosphorImager (Molecular Dynamics). Image densities were quantified off-line using ImageQuant 5.0 (Molecular Dynamics) by integrating pixel densities of individual protein bands (core- and fully glycosylated forms of hERG/hKv1.5; Hsp70 protein band) from equal areas after background subtraction. Image densities were normalized to densities measured in control samples.
Metabolic Labeling and Immunoprecipitation
One to 2 days after plating of stable cell lines (HEK hERG WT, hERG R752W HAex, hERG G601S, and hERG G628S: cell lines were selected for comparable expression levels of hERG protein using Western blotting), cells were starved in methionine/cysteine-free medium for 30 minutes and then pulse-labeled for 30 or 60 minutes in 100 to 150 µCi/mL [35S]-methionine/cysteine (EXPRESS [35S] Protein Labeling Mix, Dupont NEN) containing medium. To cross-link, cells were washed once with phosphate-buffered saline (PBS) and then incubated for 5 minutes at room temperature with 1 mL PBS containing 2 mmol/L dithiobis(succinimidyl propionate) (DSP, Pierce). The cross-linker was quenched by addition of 10 mmol/L glycine. Cells were harvested in 1 mL cold PBS, pelleted, and lysed in 0.1% NP40, 20 mmol/L MoO4, 50 mmol/L Tris pH 7.5, 150 mmol/L NaCl, and COMPLETE protease inhibitor (Roche). Insoluble material was pelleted and supernatants were collected and assayed for protein concentration using the BCA method (Pierce). Immunoprecipitation reactions containing 300 to 375 µg of total lysate protein were performed in a final volume of 300 to 350 µL. Reactions were incubated overnight at 4°C. Immunocomplexes were collected with protein G Dynabeads (Dynal) for 2 hours at 4°C. Samples were washed three times with 500 µL lysis buffer, resuspended in 30 µL 5% ß-mercaptoethanol/SDS sample buffer, and boiled 5 minutes to reverse cross-linking and to release the immunoprecipitated protein from the magnetic beads. Eluted samples were analyzed by SDS-PAGE and bands visualized by imaging with the Storm PhosphorImager system (Molecular Dynamics). To quantify hERG/chaperone interactions at specific time points, image densities of core-glycosylated hERG protein bands were determined on autoradiograms after immunoprecipitation with anti-hERG, anti-Hsp90, and anti-Hsp/c70 antibodies by integrating pixel densities from equal areas after background subtraction. Image densities of hERG protein bands isolated with anti-chaperone antibodies were normalized to image densities of hERG protein bands isolated with anti-hERG antibody to define association quotients that were used to assess time-dependent changes in hERG-chaperone interactions independently from small variations in expression levels of the different hERG cell lines used. Data are shown as mean±SE of at least two to three independent experiments. In pulse-chase experiments that were performed without chemical cross-linking, fully and core-glycosylated hERG protein bands were quantified as described. Image densities were determined after different chase periods and normalized to the image density of the freshly synthesized, core-glycosylated hERG protein band isolated immediately after radiolabeling at t=0. Pulse-chase data represent measurements from at least three independent experiments and are shown as mean±SE. After normalization, the time course of protein degradation as well as generation of fully glycosylated hERG protein can be compared directly in the different hERG cell lines used.
Immunocytochemistry
COS-7 cells were grown on polylysine-coated glass coverslips and transfected with hERG WT HAex using Fugene. For immunostaining of cell surface epitopes, live cells were blocked in 5% goat serum/PBS for 30 minutes at 4°C and incubated for 2 hours at 4°C with rat anti-HA antibody (3F10, Roche, 1:100). After incubation with primary antibody, cells were washed with PBS and fixed in ice-cold 4% formaldehyde/PBS for 30 minutes. After fixation, cells were re-blocked in 5% goat serum/PBS (30 minutes at room temperature, RT) and permeabilized with 0.1% Triton X-100. Permeabilized cells were then incubated with rabbit anti-hERG 519 (1:100) for 2 hours at RT, washed in PBS, re-blocked, and labeled with FITC-conjugated anti-rat (1:100) and TRITC-conjugated anti-rabbit (1:250) secondary antibodies. After a final wash in PBS, coverslips were mounted with Vectashield and examined using an epifluorescence microscope equipped with appropriate filter sets. For confocal analysis, COS-7 cells were transfected with hERG WT. Cells were fixed in 4% formaldehyde/PBS for 15 minutes at RT. After fixation, cells were washed in PBS, permeabilized for 10 minutes in 0.1% saponin and blocked for 30 minutes in 5% goat serum/PBS/0.1% saponin (GS/PBS/SP). For labeling, cells were incubated overnight at 4°C with rabbit anti-hERG (Alomone, 1:200, in GS/PBS/SP) and mouse anti-KDEL (1:200, in GS/PBS/SP). After several washes in PBS cells were re-blocked for 30 minutes and labeled for 1 hour at RT with goat anti-rabbit FITC (1:100) and goat anti-mouse RED X (1:100)–conjugated secondary antibodies (in GS/PBS/SP). After a final wash in PBS, coverslips were mounted and analyzed using a Leica confocal laser scan microscope.
Guinea Pig Ventricular Myocytes: Isolation and Short-Term Culture
Single ventricular myocytes were isolated from adult guinea pigs using a method described by Zakharov and Harvey, 1997.16 Briefly, guinea pigs were anesthetized by injection of pentobarbital. Hearts were quickly removed and perfused via the aorta with a physiological salt solution (PSS) containing (in mmol/L) NaCl 140, KCl 5.4, MgCl2 2.5, CaCl2 1.5, glucose 11, and HEPES 5.5 (pH 7.4). After 5 minutes, perfusate was switched to a nominally calcium-free PSS with collagenase (Roche, 0.5 mg/mL) being added after an additional 5 minutes. After 20 to 35 minutes of digestion, hearts were perfused with a high K+ solution containing (in mmol/L) potassium glutamate 110, KH2PO4 10, KCl 25, MgSO4 2, taurine 20, creatine 5, EGTA 0.5, glucose 20, and HEPES 5 (pH 7.4). Ventricles were minced in high K+ solution, and single myocytes were obtained by filtering through an 115-µm nylon mesh. Myocytes were collected in a low-speed spin and plated on E-C-L–coated coverslips (Upstate) at low density in culture medium consisting of MEM buffered with 25mmol/L HEPES to pH 7.3 and supplemented with 5% fetal bovine serum, 5% NU serum (BD), ITS (BD, 2.5 µg/mL insulin, 2.5 µg/mL transferrin, 2.5 ng/mL selenious acid), and 50 µg/mL gentamycin.17 Cultures were incubated in 5%CO2 at 37°C.
Patch-Clamp Recordings
HEK293/hERG WT and L/hKv1.5 cells were recorded using patch pipettes filled with (in mmol/L) K-aspartate, 100, KCl, 20, MgCl2, 2, CaCl2, 1, EGTA, 10, and HEPES 10 (pH7.2). The extracellular solution had the following composition (in mmol/L) NaCl, 140, KCl, 5, MgCl2, 1, CaCl2, 1.8, HEPES 10, and glucose 10 (pH 7.4). For electrophysiological experiments with Hsp70, HEK/hERG cells were transiently transfected (Fugene, Roche) with Hsp70 cDNA together with 0.25 µg EGFP to allow for identification of successfully transfected cells. Total cDNA concentration was kept constant at 1.25 µg using pcDNA3 vector cDNA. Recordings were performed 2 days after transfection. To study the effects of taxol, doxorubicin, geldanamycin, and radicicol on hERG or Kv1.5 current densities, drugs were diluted in prewarmed culture medium and added to stably transfected HEK/hERG or L/hKv1.5 cells for 12 to 16 hours (overnight) before recording. Ventricular myocytes isolated from guinea pigs and cultured overnight were recorded using the following intracellular solution (in mmol/L): K-gluconate 119, KCl 15, MgCl2 3.75, EGTA 5, HEPES 5, K-ATP 4, phosphocreatine 14, Tris-GTP 0.3, and 50U/mL creatine phosphokinase (pH 7.2). The extracellular solution had the following composition (in mmol/L): NaCl 132, KCl 4, MgCl2 1.2, CaCl2 1.8, and HEPES 5 (pH 7.4). Glucose was used to adjust the osmolarity to 300 mosm. L-type Ca2+ current was blocked with 1 µmol/L nisoldipine. E4031 (gift from Eisai, Tokyo, Japan) was used to isolate IKr current in myocytes. To analyze changes in current densities, membrane capacitances were measured using the analogue compensation circuit of an Axon 200B patch-clamp amplifier. Pclamp software (Axon Instruments) was used for generation of voltage-clamp protocols and data acquisition. No leak subtraction was applied. All current recordings were performed at room temperature (20°C to 22°C). Whenever possible, data are presented as mean±SE of n experiments.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Inhibition of Hsp90 by Geldanamycin Prevents Maturation of hERG
HERG channels are synthesized in two forms: (1) an immature core-glycosylated protein (cg) of about 135 kDa that is localized to the ER, and (2) a fully glycosylated (fg) mature protein of about 155 kDa, which is transported to the cell surface.4 Geldanamycin (GA) is a specific inhibitor of the cytosolic chaperone Hsp90 and its ER-restricted homologue Grp9418,19 that occupies the ATP binding site and interrupts the ATPase cycle of these chaperones.20 Incubation with GA produced a time-dependent decrease in the amount of mature, fully glycosylated hERG protein under steady-state conditions (Figure 1A). To further validate our biochemical results, we studied the functional effects of Hsp90 inhibition on hERG WT currents stably expressed in HEK293 cells. Whereas acute application of GA had no effect (data not shown), hERG currents were reduced when WT cells were cultured overnight in the presence of 1 µg/mL (1.8 µmol/L) GA (Figure 1B).
|
Because GA is known to induce expression of endogenous chaperones via activation of the heat shock transcription factor HSF1,21,22 we assessed variations in steady-state levels of Hsp90, Hsp/c70, calnexin, and Grp94 as potential sources for the observed drug effect on hERG maturation (Figure 1A). The most prominent change was seen in the Hsp70 family chaperones Hsc70 and Hsp70, which is in line with previous reports that steady-state levels of stress-inducible Hsp70 can be dramatically increased during the cellular response to GA.21 Therefore, we overexpressed Hsp70 protein by transient transfection and investigated effects on hERG maturation (Figure 2). In these experiments, changes in Hsp70 levels did not inhibit maturation of hERG WT protein (Figure 2A). In contrast, steady-state levels of hERG protein and current densities in HEK/hERG cells were slightly elevated with increasing Hsp70 levels as determined by quantitative analysis of Western blots (Figures 2B and 2C) and electrophysiological current recordings (Figure 2D).
|
The proapoptotic potential of benzoquinone ansamycins such as GA may provide another explanation for the inhibition of channel maturation. To control for proapoptotic effects, we tested two other chemotherapeutic agents, taxol and doxorubicin.23–25 While taxol is believed to induce apoptosis mainly via the stabilization of microtubules causing mitotic arrest, the anthracycline doxorubicin is considered a DNA-damaging agent that creates oxidative protein damage by generating free-oxygen radicals. However, neither compound interfered significantly with hERG maturation measured by Western blotting or patch-clamp recording (Figure 3). With 10 µmol/L doxorubicin, we observed some cytotoxicity in biochemical experiments as indicated by a decrease in the fully glycosylated form of hERG (Figure 3C). However, this decrease was mirrored by a similar decrease in the core-glycosylated form of hERG, suggesting cytotoxic cell death (data not shown). In line with this interpretation, current density was not reduced by 10 µmol/L doxorubicin as only viable cells are accessible for patch-clamp recordings. In contrast to our results with taxol and doxorubicin, treatment with GA reduced the expression of fully glycosylated hERG protein as shown by Western blotting (Figure 3C). Together with changes in protein expression, we measured a significant reduction in tail current amplitudes on incubation with 1 and 10 µmol/L GA from 187.8±21.2 (n=31) to 92.1±12.7 pA/pF (n=19) and 82.2±14.0 pA/pF (n=12), respectively (Figure 3D).
|
To provide further evidence for a specific inhibition of Hsp90/Grp94-hERG complexes, we tested the structurally unrelated macrocyclic antibiotic radicicol.26,27 Radicicol (RAD) is known to bind to and block the same N-terminal nucleotide binding pocket of Hsp90/Grp94 as GA.20 RAD inhibited hERG maturation on Western blots (Figure 3B). After incubation with 1, 10, and 30 µmol/L RAD, image densities of fully glycosylated hERG protein bands were decreased (Figure 3C). In electrophysiological experiments, 10 and 30 µmol/L RAD reduced tail current amplitudes significantly from 187.8±21.2 to 36.0±6.4 pA/pF (n=8) and 12.4±1.9 pA/pF (n=7), respectively (Figure 3D). We did not detect any changes in current kinetics or voltage dependence with either GA or RAD (data not shown). Based on these experiments, we exclude the possibilities that GA affects hERG maturation via induction of apoptotic cell death or oxidative protein damage.28 Taken together, our steady-state analysis shows that GA interacts specifically with Hsp90 and/or possibly Grp94 to disrupt maturation of hERG.
Specificity of Hsp90 for hERG Maturation
In unstressed eukaryotic cells, Hsp90 is thought to function as a selective chaperone for a specialized subset of proteins that have difficulties reaching their native state.29 Thus, the question arises as to the selectivity of Hsp90 for different potassium channels. To address this question, we chose Kv1.5 for the following reasons: Kv1.5 is synthesized as a core-glycosylated protein (cg) of about 68 kDa, which matures to a fully glycosylated (fg) protein of about 75 kDa, most likely representing the cell surface form of the channel (Figure 4A2). In addition, the half-life of the Kv1.5 channel protein has been determined to be 4 hours.30 Thus, we can exclude the possibility that protein levels may not change in the presence of GA simply because the target protein is extremely stable. When L/hKv1.5 cells were treated under steady-state conditions with GA we were not able to detect any change in the expression of fully glycosylated hKv1.5 protein (Figures 4A1 and 4B). To validate our biochemical results, we studied the functional effects of GA on hKv1.5 current expression. Current densities were not significantly altered on incubation with GA. Under control conditions, we measured 342.3±51.9 pA/pF (n=15); in the presence of 1 and 10 µmol/L GA, current densities were 307.5±32.9 (n=13) and 367.4±41.8 pA/pF (n=17). Similarly, Kv2.1 current densities were not altered on incubation with GA when assayed with electrophysiological recordings (data not shown).
|
Interaction With Hsp90 Is Crucial for Maturation of hERG During Biosynthesis, and Long-Lasting Inhibition of Hsp90 Promotes Ubiquitination
To study the kinetic effects of GA on the conversion of newly synthesized ER-resident hERG protein into the fully glycosylated cell surface form, we performed pulse-chase experiments (Figure 5A). Measured immediately after radiolabeling (t=0 hour), similar quantities of ER-resident core-glycosylated hERG were synthesized under control conditions and in the presence of GA. In control cells within the first 4 hours, about 50% of the initially synthesized 135-kDa form was converted into the mature fully glycosylated 155-kDa channel protein, which decayed with a half-life of about 11 hours (Figure 5B). In contrast, cells treated with GA produced only small amounts of fully glycosylated mature protein. Under these conditions, the main signal corresponded to the ER-resident 135-kDa protein, which decayed with a half-life of about 4 hours. This turnover rate was not different from the time-dependent decay observed for the 135-kDa form synthesized under control conditions. Block of hERG maturation by GA seemed to result in neither the accumulation of additional core-glycosylated protein in the ER nor in an accelerated turnover of ER-resident hERG as was described for CFTR.14 Rather, it appears that hERG protein was directed very efficiently toward pathways leading to proteasomal degradation. This hypothesis was tested directly by coexpression of hERG WT and ubiquitin tagged with either N-terminal hemagglutinin (HA) or poly-histidine (His6, negative control) epitopes.31 Transfected ubiquitin/hERG WT cells were cultured in the absence or presence of GA, and lysates immunoprecipitated with anti-hERG antibody. Immunoprecipitates were then blotted in duplicate with either anti-HERG or anti-HA antibodies. In these experiments, GA-treated WT cells showed a marked increase in multiubiquitinated hERG protein seen as a dark smear reaching from about 135 kDa to the top of the gel (Figure 5D). We conclude that, after inhibition of Hsp90 by GA, multiubiquitination and subsequent degradation are increased, and suggest that hERG protein is primarily consumed by the proteasome.
|
Influence of Geldanamycin on Subcellular Localization of hERG
Our biochemical data indicate that, on incubation with GA, hERG protein is primarily present as an ER-resident core-glycosylated form. To verify our biochemical results, we determined the subcellular localization of hERG protein in the presence of GA using immunostaining of transiently expressed hERG channels in COS-7 cells (Figure 6). HERG channels at the cell surface were visualized directly by expressing a channel protein extracellularly tagged with an HA epitope inserted in the S1-S2 linker of hERG (hERG S1HAS2). Insertion of this epitope tag affected neither the electrophysiology nor the trafficking of hERG WT channels (data not shown). Figure 6A1 shows surface fluorescence in control cells stained prior to fixation and permeabilization with rat anti-HA antibody. The intracellular distribution of hERG was determined in the same cells after permeabilization and staining with a polyclonal rabbit anti-hERG antibody directed against the intracellular C-terminus of the protein. After permeabilization, immunofluorescence was primarily detected in a perinuclear compartment, consistent with the ER localization of core-glycosylated channel protein (Figure 6A2). After treatment with GA, surface labeling was no longer observed (Figure 6A3), whereas perinuclear staining was preserved (Figure 6A4). To match the observed perinuclear staining with a well-defined subcellular compartment, we performed double labeling experiments in permeabilized cells using anti-hERG and anti-KDEL-antibodies. Anti-KDEL antiserum recognizes a KDEL retention signal expressed on several ER-resident chaperones, including Grp78/BiP and Grp94. An ER marker was chosen because cells exposed to GA produced mainly the core-glycosylated form of hERG, which had been previously assigned to the ER.4 Using confocal analysis, we found that in control cells, a large fraction of hERG protein colocalized with the ER marker KDEL (Figures 6B1 through 6B3). However, labeling was not confined to the ER but was also found in peripheral compartments such as the cell surface. Upon incubation with GA, the subcellular staining pattern changed dramatically, now showing exclusive ER staining as indicated by overlapping staining patterns produced with anti-hERG and anti-KDEL antibodies (Figures 6B4 to 6B6).
|
Newly Synthesized hERG WT Channels Form Complexes With Hsp/c70 and Hsp90
To test directly whether GA prevents maturation of hERG WT channels by inhibiting chaperone-channel interactions, we performed coimmunoprecipitation experiments using an HEK/hERG WT cell line and radiolabeling to determine association of hERG protein with Hsp90 and Grp94, two chaperone targets for GA,18,19 as well as Hsp/c70.32 Immediately after radiolabeling, hERG WT could be immunoprecipitated in its core-glycosylated 135-kDa form using anti-hERG antibody. When the formation of hERG-Hsp/c70 chaperone complexes was assessed using anti-Hsp/c70 antibody we found in addition to Hsp/c70 protein, a second protein band that migrated with the exact mobility of newly synthesized, core-glycosylated hERG protein (Figure 7A). In marked contrast, hERG-Hsp90 complexes could only be isolated together with core-glycosylated hERG protein after chemical cross-linking. Formation of hERG-Hsp90 complexes could be inhibited by blockade of the functionally important ATPase activity of Hsp90 using GA during the labeling period.20 However, even with a chemical cross-linker, we were not able to isolate hERG protein in association with the GA-sensitive chaperone Grp94, which represents an ER-resident Hsp90 homologue (Figure 7A). Thus, we conclude that GA prevents maturation of hERG channels into the functional fully glycosylated cell surface form by inhibiting a highly dynamic interaction between hERG and Hsp90, as judged from the requirement for chemical cross-linking to stabilize this interaction. In these experiments, GA targeted hERG/Hsp90 complexes with high specificity and did not interfere with the formation of hERG/Hsp/c70 chaperone complexes or the pull-down of hERG protein with either anti-hERG antibody or Hsp90 protein with anti-Hsp90 antibody (Figure 7B). Furthermore, the isolation of hERG/Hsp/c70 complexes could be reduced in the presence of an ATP regenerating system during immunoprecipitations as described for many other proteins that interact specifically with Hsp/c70 (see online Figure S1, in the online data supplement at http://www.circresaha.org).15 As expected for chaperones that are thought to facilitate the folding of ER-resident precursor substrates into mature native protein conformations, both Hsp70 and Hsp90 interacted exclusively with newly synthesized core-glycosylated hERG protein but not with fully glycosylated mature cell surface protein synthesized during a chase of 6 hours (Figure 7C).
|
Interactions Between Chaperones and LQT2 Mutant hERG Channels Are Altered
To determine whether Hsp/c70 and Hsp90 bind to WT channels and LQT2 mutant channels differentially, we analyzed the interaction of three different LQT2 mutations, hERG R752W,6 hERG G601S,33,34 and hERG G628S,3,4 with these chaperones.
First, we examined the temperature-sensitive, trafficking-deficient LQT2 mutation hERG R752W (Figure 8A). hERG R752W is in the cytoplasmic, C-terminal cNBD domain of the channel protein. At physiological temperature (37°C), hERG R752W is retained in the ER and no mature, fully glycosylated protein is generated in pulse-chase experiments (Figures 8A and 9
A). When immunoprecipitations with anti-Hsp/c70 or Hsp90 antisera were performed immediately after radiolabeling (t=0 hour), core-glycosylated R752W mutant protein could be isolated together with either Hsp90 or Hsp/c70. Conversely, in immunoprecipitations with anti-hERG antibody, we were able to detect complexes formed by core-glycosylated R752W protein with both Hsp90 and Hsp/c70 as major constituents together with several smaller not yet identified protein bands (Figure 8A). To determine the stability of chaperone-channel complexes over time, additional immunoprecipitations were performed after chase periods of 6 and 16 hours. In these experiments, we found that HERG R752W formed complexes with both Hsp90 and Hsp/c70 during the entire chase period analyzed (Figure 8A). To determine the abundance of channel-chaperone complexes and time-dependent changes in their composition more precisely, we first quantified the amount of hERG R752W protein that could be immunoprecipitated with either Hsp/c70 or Hsp90 antibody using a PhosphorImager. In a second step, these image densities were normalized to the amount of hERG protein immunoprecipitated with anti-hERG antibody. This normalization procedure was necessary because hERG R752W protein decayed along a complex time course with a half-life of about 6 hours during pulse-chase experiments (Figure 9A). When compared with newly synthesized WT channels (at t=0 hour), hERG R752W-Hsp90 complexes were about 2.6-fold more abundant, reflecting most likely discrete intermediates of the folding pathway and subtle folding defects of the mutant. Moreover, chaperone association remained constant over time for R752W, whereas both Hsp90 and Hsp/c70 interacted only with newly synthesized hERG WT protein (t=0 hour) but not with mature fully glycosylated WT channels (t=6 hours; Figure 9B). Chaperone interactions similar to the one described for hERG R752W analyzed at 37°C were also detected for the LQT2 mutant hERG A561V, which is strictly retained in the ER and rapidly degraded in the proteasome (see online Figure S2).5,7
|
|
Given the differences in chaperone association observed between trafficking-competent WT channels and trafficking-deficient R752W channels at 37°C, we asked how restoration of trafficking in R752W by incubation at low temperature (26°C) affected the formation and stability of R752W channel-chaperone complexes. As described previously6 at 26°C, we detected in immunoprecipitations with anti-hERG antibody not only core-glycosylated but also fully glycosylated hERG protein generated with slow kinetics during chase periods of 16 and 24 hours (Figure 8B). At 26°C, generation of fully glycosylated hERG protein was not only slowed down for R752W but also for hERG WT channels. In pulse-chase experiments, synthesis of fully glycosylated WT protein peaked after a chase period of about 16 hours, which is considerably slower than 5 hours measured for WT protein at 37°C. Fully glycosylated hERG R752W protein was generated with the same kinetics as WT protein in pulse-chase experiments performed at 26°C (Figure 9C). However, we noted a major difference in the stability of core-glycosylated hERG WT and R752W protein forms. Although the amount of core-glycosylated WT protein was stable for about 10 hours after synthesis, hERG R752W was being degraded immediately after labeling. The fast degradation of core-glycosylated hERG R752W might help to explain why little R752W channel protein was ultimately converted into the fully glycosylated cell surface form at 26°C (Figure 9C). Given the small amount of fully glycosylated R752W protein generated and the large pool of core-glycosylated ER-resident R752W protein left even after chase periods of 30 hours, it is not surprising that immunoprecipitations with anti-Hsp/c70 and anti-Hsp90 isolated channel-chaperone complexes in the entire time period analyzed (0, 16, and 24 hours; Figure 8B). However, when the abundance of R752W chaperone complexes formed at 26°C was quantified over time, we found that association of R752W with Hsp/c70 was reduced in parallel with the generation of fully glycosylated R752W protein by about 1.6-fold (from 0.9 to 0.58), and with Hsp90 by about 1.4-fold (from 0.5 to 0.35; Figure 9D). At the same time, R752W-Hsp90 complexes were 1.6-fold more abundant at low incubation temperature than corresponding WT complexes (Figure 9D). Thus, temperature-dependent induction of channel folding and trafficking in R752W was accompanied by a time-dependent reduction in chaperone association, which is, however, less pronounced than the changes observed for WT channels at 26°C. These quantitative differences may result mainly from the small amount of fully glycosylated R752W protein ultimately generated at 26°C and indicate that the majority of R752W protein remained in misfolded, trafficking-deficient conformations even at 26°C.
Because the trafficking defect of hERG R752W was corrected by low incubation temperature only for a small channel population, we decided to analyze a second LQT2 retention mutant, hERG G601S. This temperature-sensitive, hypomorphic LQT2 mutant channel was selected for two reasons: (1) hERG G601S trafficking can be restored at physiological temperature (37°C) using astemizole,34 and (2) the folding defect of hERG G601S can be corrected with high efficiency. The mutation in hERG G601S resides in the outer mouth of the channel protein and represents a mild phenotype such that some functional channels are generated even at 37°C. Pharmacological rescue of hERG G601S can be initiated by astemizole binding to the methanesulfonanilide binding site, which has been localized to the central cavity of the hERG channel protein.34 Given the location of this drug-binding site, we assume in our rescue experiments that astemizole will not interfere directly with Hsp/c70 or Hsp90 binding to cytoplasmic N- or C-terminal protein domains.
At 37°C, hERG G601S was mainly isolated in its core-glycosylated form when immunoprecipitated with anti-hERG antibody (Figure 10A), and only a small fraction of the newly synthesized protein was converted into mature, fully glycosylated channels in line with the original description of G601S as hypomorphic, ie, generating small currents even at 37°C (Figure 10B). After incubation with 5 µmol/L astemizole, however, trafficking of hERG G601 was restored and newly synthesized G601S protein was converted with high efficiency into its mature fully glycosylated form (Figures 10A and 10B; and online Figure S3). Resumption of trafficking on incubation with astemizole was also reflected in immunoprecipitations using Hsp/c70 and Hsp90 antibodies as predicted from our experiments with WT. Under control conditions, core-glycosylated hERG G601S protein could be isolated together with either Hsp/c70 or Hsp90 and remained associated with these chaperones over time (Figure 10C). After restoration of trafficking with astemizole, however, channel-chaperone complexes dissociated completely during maturation of channel proteins. It is interesting to point out that the degree of chaperone association was lower for hERG G601S (association coefficients between 0.1 and 0.4) than for R752W (between 0.5 and 1.1) at 37°C. This observation may be related to the mild mutant phenotype presented by hERG G601S with its high propensity to fold into a native conformation even at 37°C. Notwithstanding this low degree of overall chaperone association, we found that immediately after synthesis, hERG G601S-Hsp/c70 complexes were about 1.8-fold more abundant (t=0 hour; 0.43 versus 0.24; Figure 10C) under control conditions than in the presence of astemizole. Our failure to detect similar differences for Hsp90 was surprising given our results with hERG R752W. However, hERG G601S-Hsp90 complexes were present only in small quantities so that differences between control and astemizole-treated cells might have been difficult to detect. To ascertain that despite its low association Hsp90 still participated in productive folding of hERG G601S, we studied the effects of Hsp90 inhibition by GA in the absence and presence of astemizole (Figure 10D). In both cases, we found that GA inhibited the maturation of hERG G601S as indicated by a decrease in the small amount of fully glycosylated hERG G601S present under control conditions and by the complete reversal of pharmacological rescue induced by astemizole.
|
Our experiments with hERG R752W and hERG G601S demonstrated that both of these trafficking-deficient LQT2 mutants remained associated with cytosolic Hsp/c70 and Hsp90 chaperones while in the ER. More important, in both instances, restoration of channel trafficking and function was tightly coupled to the dissociation of channel-chaperone complexes. To explore whether intact trafficking even when associated with a severe mutant phenotype may be accompanied by WT-like chaperone interactions, we analyzed the ER handling and processing of a nonfunctional but trafficking-competent LQT2 mutant channel, hERG G628S.3,4 hERG G628S resides in the selectivity filter of the channel protein and generates nonconducting channels that reach the cell surface as determined by their glycosylation pattern. In immunoprecipitations with anti-hERG antibody, we found that hERG G628S was initially synthesized as a core-glycosylated protein that then matured into the fully glycosylated cell surface form at physiological temperature as reported previously (Figure 11). With Hsp/c70 and Hsp90 antibodies, we isolated hERG G628S in complex with each of these chaperones immediately after synthesis (t=0 hour) but not after a chase period of 6 hours during which hERG G628S matured into the fully glycosylated cell surface form (t=6 hours, Figure 11). Thus, nonfunctional hERG G628S mutant channels are able to escape initially formed channel-chaperone complexes as described for WT channels. hERG G628S channels leave the ER and reach the cell surface most likely by adopting protein conformations so close to the native WT conformation that they can no longer be recognized and eliminated by the cellular quality control machinery.
|
Inhibition of Hsp90 Reduces Amplitudes of Native IKr Currents Expressed in Guinea Pig Myocytes
We complemented our analysis with experiments in ventricular cardiomyocytes to determine the effects of Hsp90 inhibition on native cardiac IKr channels. Cultured cardiomyocytes were treated for 24 hours with increasing concentrations of GA and analyzed using patch-clamp recordings. Currents were elicited during 650ms depolarizations from a holding potential of -40 mV. E4031-sensitive tail currents were isolated upon return from 60 mV (Figure 12A). GA reduced tail current densities in guinea pig cardiomyocytes from 0.72±0.09 to 0.32±0.04 pA/pF and 0.19±0.03 pA/pF upon exposure to 1 or 2 µg/mL (1.8 or 3.6 µmol/L) GA, respectively (Figure 12C). Thus, 1 µg/mL (1.8µmol/L) GA applied for 24 hours reduced both native IKr currents and heterologously expressed hERG currents by about 50%. To gain further insight in the specificity of the described effects, we analyzed possible changes in cardiac IKs currents upon exposure to GA. IKs currents were elicited in the presence of 5 µmol/L E4031 with 2.5-second step depolarizations (Figure 12B). Under these recording conditions, slow tail currents represent a nearly quantitative measure of IKs. Tail current amplitudes were quantified in myocytes exposed to increasing concentrations of GA and were not decreased by drug exposure (Figure 12D). These data indicate that processing of IKs channels is insensitive to GA and that there is specificity in the interaction of Hsp90 with native cardiac potassium channels.
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
In the present study, we have shown for the first time that folding and maturation of the cardiac potassium channel hERG/IKr are controlled by the cytosolic chaperone Hsp90. Our evidence is 3-fold: first, we used the specific Hsp90 binding compounds GA and radicicol to pharmacologically inhibit hERG channel maturation. Second, we showed the formation of a transient chaperone complex between immature ER-resident hERG protein and Hsp90 but not Grp94, another member of the Hsp90 family expressed in the lumen of the ER. Third, the Hsp90-binding compound GA prevented formation of Hsp90-hERG complexes but did not prevent formation of Hsp/c70 complexes with newly synthesized hERG channels.
Because GA is able to lock Hsp90 to substrates such as eNOS with which it interacts tightly,35 the inhibition of Hsp90/hERG complex formation by GA indicated that the interaction may be either indirect or, alternatively, that complex formation is highly dynamic. We found no additional major protein other than Hsp70 in complex with Hsp90/hERG, and therefore favor the existence of highly dynamic Hsp90/substrate interactions as proposed for Hsp90 clients such as HSF1, tyrosine kinases, p53 mutants, and the CFTR ion channel.14,18,22,36,37 GA-mediated inhibition of Hsp90 most often induces increased turnover and proteasomal degradation of folding substrates38 although a few proteins such as ApoB48 or HSF1 are stabilized by GA.22,39 As described for steroid receptors and a large number of tyrosine kinases,38 hERG channels were increasingly ubiquitinated upon exposure to GA. In CFTR, another polytopic membrane protein that forms stable complexes with Hsp/c70 and more transient complexes with Hsp90, GA increased the turnover of immature, ER-resident protein thereby curtailing export to the cell surface.14 In contrast, turnover of immature forms of hERG was not affected as if hERG proteins that could not interact properly with Hsp90 were diverted directly into the proteasomal degradation pathway.
Unlike Hsp/c70, which acts by holding most newly synthesized proteins in a folding competent state,40 Hsp90 is required for the folding of a small subset of proteins that are believed to have difficulties reaching their native conformations.29 HERG is structurally different from other potassium channel proteins in that it contains two complex protein folds on its cytoplasmic termini: a PAS domain12 on its N-terminus and a cNBD domain on its C-terminus,11 both of which may constitute possible targets for Hsp90 binding. Several other potassium channels that apparently lack such specialized domains have been investigated and were not affected by GA. Thus, voltage-gated potassium channel do not constitute a general target for Hsp90 as shown clearly in our experiments with Kv1.5, Kv2.1, and the native IKs channel, which is assembled from KCNQ1 and KCNE1 ß subunits.41
Hsp90/hERG complex formation is not only important for the productive folding of hERG WT channels but also for the handling of LQT2 mutant channels that are retained in the ER. For two different LQT2 mutations, hERG R752W and hERG G601S, we demonstrated prolonged association with Hsp90 and Hsp/c70 in the ER. In addition, association with chaperones was increased, suggesting that these mutants are arrested as distinct folding intermediates that are detected specifically by cytosolic chaperones in their sustained attempts to re-fold these proteins into their native conformation. Both of these observations closely mirrored results obtained with CFTR
508,15 a trafficking-deficient mutant identified in most cases of cystic fibrosis. In rescue experiments with hERG R752W and G601S, we showed that initiation of productive folding by lowering the incubation temperature or by incubation with pharmacological chaperones led ultimately to the dissociation of channel-chaperone complexes, followed by successful channel export to the cell surface. However, it is not clear at present whether association with cytosolic chaperones can directly cause ER retention, or how dissociation of channel-chaperone complexes and initiation of forward transport are linked. Is the enhanced chaperone decoration only an indicator of protein misfolding that is not at all related to the failure of mutant proteins to leave the ER or is it possible that chaperones play a more active role in that they may shield export signals present in cytosolic domains of ion channel proteins,42 thereby preventing proper interaction with proteins in ER exit sites? Whatever the precise mechanism of ER retention, rescue and export of LQT2 mutants were linked to a channel conformation that was no longer tied up in chaperone complexes. Similarly, hERG G628S, a nonfunctional LQT2 mutant with intact trafficking showed a transient WT-like chaperone association pattern and demonstrated clearly that mutations are not necessarily retained unless they induce a conformational change recognized by cellular quality control mechanisms. Given the mutation of a strictly conserved glycine residue in the selectivity filter of the G628S channel protein,9 it is somewhat surprising that this amino acid exchange was not accompanied by a detectable alteration in protein conformation. However, our data are unequivocal in that this protein behaves like WT channels in its interaction with both Hsp90 and Hsp/c70 chaperones.
We consistently detected several additional protein bands in autoradiograms of immunoprecipitated hERG mutant proteins, which suggest that a large multi-chaperone complex with Hsp90 and Hsp/c70 at its core may be ultimately responsible for the productive folding of hERG. Although it is possible that some of these protein bands were nonspecifically immunoprecipitated, we favor the hypothesis that hERG is folded by a large chaperone machine that is built around Hsp/c70 and Hsp90 and functions in concert with smaller co-chaperones such as p23, FKBP51/52, Hip, Hop, or Hdj-1/2 as reported for steroid receptors and many other proteins.43–50 Future experiments will be necessary to determine which co-chaperones participate in the folding of hERG.
Our results in heterologous expression systems were closely mirrored in ventricular cardiomyocytes exposed to GA and analyzed for changes in the expression of the native delayed rectifier current IKr. We found that functional expression of IKr was decreased in a dose-dependent manner by GA, as expected if IKr channels depend on Hsp90 during biogenesis. Two independent lines of evidence strengthen this proposition. First, we tested for possible nonspecific side effects by analysis of the other major repolarizing cardiac potassium current, IKs, and found this current was insensitive to GA. Second, amplitude reductions observed in native IKr currents were comparable to reductions of heterologously expressed hERG currents upon exposure to equivalent amounts of GA.
Ansamycin antibiotics that block Hsp90 function, such as GA and radicicol, are used to target oncogenic kinases for degradation and are being tested in preclinical as well as clinical trials against various cancers.51 With the addition of hERG to the list of Hsp90 substrates, our results imply that cardiotoxic risk may be associated with the therapeutic use of these and other small molecules that target the cytosolic chaperone Hsp90.52,53 Pharmacological data indicate that concentrations above 0.1 µg/mL GA are necessary to achieve antineoplastic effects.54 This is well within the concentration range that inhibits hERG maturation. Our data raise the possibility that pharmacological inhibition of Hsp90 may produce acquired long-QT syndrome by a novel mechanism where hERG channels are not acutely blocked,55 but reduced in their expression level due to interference with channel folding and maturation.
|
Acknowledgments |
|---|
This work was supported by NIH grants HL36930, HL61642, HL55404, and DK54178 to A.M.B.
|
Footnotes |
|---|
Original received May 7, 2002; resubmission received March 14, 2003; revised resubmission received May 15, 2003; accepted May 16, 2003.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
1. Sanguinetti MC, Jiang C, Curran ME, Keating MT. A mechanistic link between an inherited and an acquired cardiac arrhythmia: HERG encodes the IKr potassium channel. Cell. 1995; 81: 299–307.[CrossRef][Medline] [Order article via Infotrieve]
2. Keating MT, Sanguinetti MC. Molecular and cellular mechanisms of cardiac arrhythmias. Cell. 2001; 104: 569–580.[CrossRef][Medline] [Order article via Infotrieve]
3. Sanguinetti MC, Curran ME, Spector PS, Keating MT. Spectrum of HERG K+-channel dysfunction in an inherited cardiac arrhythmia. Proc Natl Acad Sci U S A. 1996; 93: 2208–2212.
4. Zhou Z, Gong Q, Epstein ML, January CT. HERG channel dysfunction in human long QT syndrome. J Biol Chem. 1998; 273: 21061–21066.
5. Kagan A, Yu Z, Fishman GI, McDonald TV. The dominant negative LQT2 mutation A561V reduces wild-type HERG expression. J Biol Chem. 2000; 275: 11241–11248.
6. Ficker E, Thomas D, Viswanathan PC, Dennis AT, Priori SG, Napolitano C, Memmi M, Wible BA, Kaufman ES, Iyengar S, Schwartz PJ, Rudy Y, Brown AM. Novel characteristics of a misprocessed mutant HERG channel linked to hereditary long QT syndrome. Am J Physiol. 2000; 279: H1748–H1756.
7. Ficker E, Dennis AT, Obejero-Paz CA, Castaldo P, Taglialatela M, Brown AM. Retention in the endoplasmic reticulum as a mechanism of dominant-negative current suppression in human long QT syndrome. J Mol Cell Cardiol. 2000; 32: 2327–2337.[CrossRef][Medline] [Order article via Infotrieve]
8. Ellgard L, Molinari M, Helenius A. Setting the standards: quality control in the secretory pathway. Science. 1999; 286: 1882–1888.
9. Doyle DA, Morais Cabral J, Pfuetzner RA, Kuo A, Gulbis JM, Cohen SL, Chait BT, MacKinnon R. The structure of the potassium channel: molecular basis of K+ conduction and selectivity. Science. 1998; 280: 69–77.
10. Tseng G. IKr: the hERG channel. J Mol Cell Cardiol. 2001; 33: 835–849.[CrossRef][Medline] [Order article via Infotrieve]
11. Warmke JW, Ganetzky B. A family of potassium channel genes related to eag in Drosophila and mammals. Proc Natl Acad Sci U S A. 1994; 91: 3438–3442.
12. Morais-Cabral JH, Lee A, Cohen SL, Chait BT, Li M, MacKinnon R. Crystal structure and functional analysis of the HERG potassium channel N terminus: a eukaryotic PAS domain. Cell. 1998; 95: 649–655.[CrossRef][Medline] [Order article via Infotrieve]
13. Yang Y, Janich S, Cohn A, Wilson JM. The common variant of cystic fibrosis transmembrane conductance regulator is recognized by hsp70 and degraded in a pre-Golgi nonlysosomal compartment. Proc Natl Acad Sci U S A. 1993; 90: 9480–9484.
14. Loo MA, Jensen TJ, Cui L, Hou Y-X, Chang X-B, Riordan JR. Perturbation of hsp90 interaction with nascent CFTR prevents its maturation and accelerates its degradation by the proteasome. EMBO J. 1998; 17: 6879–6887.[CrossRef][Medline] [Order article via Infotrieve]
15. Meacham GC, Lu Z, King S, Sorscher E, Tousson A, Cyr DM. The Hdj-2/Hsc70 chaperone pair facilitates early steps in CFTR biogenesis. EMBO J. 1999; 18: 1492–1505.[CrossRef][Medline] [Order article via Infotrieve]
16. Zakharov SI, Harvey RD. Rebound stimulation of the cAMP-regulated Cl- current by acetylcholine in guinea-pig ventricular myocytes. J Physiol. 1997; 499: 105–120.
17. Schackow TE, Decker RS, Ten Eick RE. Electrophysiology of adult cat ventricular myocytes: changes during primary culture. Am J Physiol. 1995; 268: C1002–C1017.[Medline] [Order article via Infotrieve]
18. Whitesell L, Mimnaugh EG, De Costa B, Myers CE, Neckers LM. Inhibition of heat shock protein hsp90-pp60v-src heteroprotein complex formation by benzoquinone ansamycins: essential role for stress proteins in oncogenic transformation. Proc Natl Acad Sci U S A. 1994; 91: 8324–8328.
19. Chavany C, Mimnaugh E, Miller P, Bitton R, Nguyen P, Trepel J, Whitesell L, Schnur R, Moyer JD, Neckers L. p185erbB2 binds to Grp94 in vivo. J Biol Chem. 1996; 271: 4974–4977.
20. Roe SM, Prodromou C, O’Brien R, Ladbury JE, Piper PW, Pearl LH. Structural basis for the inhibition of the hsp90 molecular chaperone by the antitumor antibiotics radicicol and geldanamycin. J Med Chem. 1999; 42: 260–266.[CrossRef][Medline] [Order article via Infotrieve]
21. Kim HR, Kang HS, Kim HD. Geldanamycin induces heat shock protein expression through activation of HSF1 in K562 erythroleukemic cells. IUBMB Life. 1999; 48: 429–433.[Medline] [Order article via Infotrieve]
22. Zou J, Guo Y, Guettouche T, Smith DF, Voellmy R. Repression of heat shock transcription factor HSF1 activation by hsp90 (hsp90 complex) that forms a stress-sensitive complex with HSF1. Cell. 1998; 94: 471–480.[CrossRef][Medline] [Order article via Infotrieve]
23. Wang LG, Liu XM, Kreis W, Budman DR. The effect of antimicrotubule agents on signal transduction pathways of apoptosis: a review. Cancer Chemother Pharmacol. 1999; 44: 355–361.[CrossRef][Medline] [Order article via Infotrieve]
24. Wang T-H, Wang H-S, Soong YK. Paclitaxel-induced cell death. Cancer. 2000; 88: 2619–2628.[CrossRef][Medline] [Order article via Infotrieve]
25. Pai VB, Nahata MC. Cardiotoxicity of chemotherapeutic agents: incidence, treatment and prevention. Drug Saf. 2000; 22: 263–302.[CrossRef][Medline] [Order article via Infotrieve]
26. Sharma SV, Agatsuma T, Nakano H. Targeting of the protein chaperone, Hsp90, by the transformation suppressing agent, radicicol. Oncogene. 1998; 16: 2639–2645.[CrossRef][Medline] [Order article via Infotrieve]
27. Schulte TW, Akinaga S, Murakata T, Agatsuma T, Sugimoto S, Nakano H, Lee YS, Simen BB, Argon Y, Felts S, Toft DO, Neckers LM, Sharma SV. Interaction of radicicol with members of the heat shock protein 90 family of molecular chaperones. Mol Endocrinol. 1999; 13: 1435–1448.
28. Zou J, Salminen WF, Roberts SM, Voellmy R. Correlation between glutathione oxidation and trimerization of heat shock factor 1, an early step in stress induction of the Hsp response. Cell Stress Chaperones. 1998; 3: 130–141.[CrossRef][Medline] [Order article via Infotrieve]
29. Nathan DF, Harju Voss M, Lindquist S. In vivo functions of the Saccharomyces cerevisiae Hsp90 chaperone. Proc Natl Acad Sci U S A. 1997; 94: 12949–12956.
30. Takimoto K, Fomina AF, Gealy R, Trimmer JS, Levitan ES. Dexamethasone rapidly induces Kv1.5 K+ channel gene transcription and expression in clonal pituitary cells. Neuron. 1993; 11: 359–369.[CrossRef][Medline] [Order article via Infotrieve]
31. Treier M, Staszewski LM, Bohmann D. Ubiquitin-dependent c-Jun degradation in vivo is mediated by the 
domain. Cell. 1994; 78: 787–798.[CrossRef][Medline]
[Order article via Infotrieve]
32. Bukau B, Horwich AL. The Hsp70 and Hsp60 chaperone machines. Cell. 1998; 92: 351–366.[CrossRef][Medline] [Order article via Infotrieve]
33. Furutani M, Trudeau MC, Hagiwara N, Seki A, Gong Q, Zhou Z, Imamura S, Nagashima H, Kasanuki H, Takao A, Momma K, January CT, Robertson GA, Matsuoka R. Novel mechanism associated with an inherited cardiac arrhythmia. Circulation. 1999; 99: 2290–2294.
34. Ficker E, Obejero-Paz C, Zhao S, Brown AM. The binding site for channel blockers that rescue misprocessed human long QT syndrome type 2 ether-a-gogo–related gene (HERG) mutations. J Biol Chem. 2002; 277: 4989–4998.
35. Garcia-Cardena G, Fan R, Shah V, Sorrentino R, Cirino G, Papapetropoulos A, Sessa WC. Dynamic activation of endothelial nitric oxide synthase by Hsp90. Nature. 1998; 392: 821–824.[CrossRef][Medline] [Order article via Infotrieve]
36. Xu W, Mimnaugh E, Rosser MFN, Nicchitta C, Marcu M, Yarden Y, Neckers L. Sensitivity of mature erbB2 to geldanamycin is conferred by its kinase domain and is mediated by the chaperone protein hsp90. J Biol Chem. 2001; 276: 3702–3708.
37. Whitesell L, Sutphin PD, Pulcini EJ, Martinez JD, Cook PH. The physical association of multiple molecular chaperone proteins with mutant p53 is altered by geldanamycin, an hsp90-binding agent. Mol Cell Biol. 1998; 18: 1517–1524.
38. Pratt WB. The role of the hsp90-based chaperone system in signal transduction by nuclear receptors and receptors signaling via MAP kinase. Annu Rev Pharmacol Toxicol. 1997; 37: 297–326.[CrossRef][Medline] [Order article via Infotrieve]
39. Gusarova V, Caplan AJ, Brodsky JL, Fisher EA. Apoprotein B degradation is promoted by the molecular chaperones hsp90 and hsp70. J Biol Chem. 2001; 276: 24891–24900.
40. Hartl FU, Hayer-Hartl M. Molecular chaperones in the cytosol: from nascent chain to folded protein. Science. 2002; 295: 1852–1858.
41. Sanguinetti MC, Curran ME, Zou A, Shen J, Spector PS, Atkinson DL, Keating MT. Coassembly of KvLQT1 and minK (IsK) proteins to form cardiac IKs potassium channel. Nature. 1996; 384: 80–83.[CrossRef][Medline] [Order article via Infotrieve]
42. Ma D, Zerangue N, Lin Y-F, Collins A, Yu M, Jan YN, Jan LY. Role of ER export signals in controlling surface potassium channel numbers. Science. 2001; 291: 316–319.
43. Smith DF, Whitesell L, Nair SC, Chen S, Prapapanich V, Rimerman RA. Progesterone receptor structure and function altered by geldanamycin, an hsp90-binding agent. Mol Cell Biol. 1995; 15: 6804–6812.[Abstract]
44. Bose S, Weikl T, Bugl H, Buchner J. Chaperone function of Hsp90-associated proteins. Science. 1996; 274: 1715–1717.
45. Freeman BC, Toft DO, Morimoto RI. Molecular chaperone machines: chaperone activities of the cyclophilin Cyp-40 and the steroid aporeceptor-associated protein p23. Science. 1996; 274: 1718–1720.
46. Frydman J, Hohfeld J. Chaperones get in touch: the Hip-Hop connection. Trends Biochem Sci. 1997; 22: 87–92.[CrossRef][Medline] [Order article via Infotrieve]
47. Segnitz B, Gehring U. The function of steroid hormone receptors is inhibited by the hsp90-specific compound geldanamycin. J Biol Chem. 1997; 272: 18694–18701.
48. Obermann WMJ, Sondermann H, Russo AA, Pavletich NP, Hartl FU. In vivo function of hsp90 is dependent on ATP binding and ATP hydrolysis. J Cell Biol. 1998; 143: 901–910.
49. Fink AL. Chaperone-mediated protein folding. Physiol Rev. 1999; 79: 425–449.
50. Young JC, Moarefi I, Hartl FU. Hsp90: a specialized but essential protein-folding tool. J Cell Biol. 2001; 154: 267–273.
51. Neckers L. Hsp90 inhibitors as novel cancer chemotherapeutic agents. Trends Mol Med. 2002; 8: S55–S61.[CrossRef][Medline] [Order article via Infotrieve]
52. Smith DF, Whitesell L, Katsanis E. Molecular chaperones: biology and prospects for pharmacological intervention. Pharmacol Rev. 1998; 50: 493–513.
53. Neckers L, Schulte TW, Mimnaugh E. Geldanamycin as a potential anti-cancer agent: its molecular target and biochemical activity. Invest New Drugs. 1999; 17: 361–373.[CrossRef][Medline] [Order article via Infotrieve]
54. Supko JG, Hickman RL, Grever MR, Malspeis L. Preclinical pharmacologic evaluation of geldanamycin as an antitumor agent. Cancer Chemother Pharmacol. 1995; 36: 305–315.[Medline] [Order article via Infotrieve]
55. Vandenberg JI, Walker BD, Campbell TJ. HERG K+ channels: friend and foe. Trends Pharmacol Sci. 2001; 22: 240–246.[CrossRef][Medline] [Order article via Infotrieve]
This article has been cited by other articles:
 |
|
 |
|
  B. J.J.M. Brundel, L. Ke, A.-J. Dijkhuis, X. Qi, A. Shiroshita-Takeshita, S. Nattel, R. H. Henning, and H. H. Kampinga Heat shock proteins as molecular targets for intervention in atrial fibrillation Cardiovasc Res, June 1, 2008; 78(3): 422 - 428. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. Chandarlapaty, A. Sawai, Q. Ye, A. Scott, M. Silinski, K. Huang, P. Fadden, J. Partdrige, S. Hall, P. Steed, et al. SNX2112, a Synthetic Heat Shock Protein 90 Inhibitor, Has Potent Antitumor Activity against HER Kinase Dependent Cancers Clin. Cancer Res., January 1, 2008; 14(1): 240 - 248. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 V. E. Walker, R. Atanasiu, H. Lam, and A. Shrier Co-chaperone FKBP38 Promotes HERG Trafficking J. Biol. Chem., August 10, 2007; 282(32): 23509 - 23516. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 L. Zhang, K. Foster, Q. Li, and J. R. Martens S-acylation regulates Kv1.5 channel surface expression Am J Physiol Cell Physiol, July 1, 2007; 293(1): C152 - C161. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. Liu, J. Xia, K.-H. Cho, D. E. Clapham, and D. Ren CatSperbeta, a Novel Transmembrane Protein in the CatSper Channel Complex J. Biol. Chem., June 29, 2007; 282(26): 18945 - 18952. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
D. B. Solit, S. P. Ivy, C. Kopil, R. Sikorski, M. J. Morris, S. F. Slovin, W. K. Kelly, A. DeLaCruz, T. Curley, G. Heller, et al. Phase I Trial of 17-Allylamino-17-Demethoxygeldanamycin in Patients with Advanced Cancer Clin. Cancer Res., March 15, 2007; 13(6): 1775 - 1782. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 L. Wang, B. A. Wible, X. Wan, and E. Ficker Cardiac Glycosides as Novel Inhibitors of Human Ether-a-go-go-Related Gene Channel Trafficking J. Pharmacol. Exp. Ther., February 1, 2007; 320(2): 525 - 534. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 F. Van Goor, K. S. Straley, D. Cao, J. Gonzalez, S. Hadida, A. Hazlewood, J. Joubran, T. Knapp, L. R. Makings, M. Miller, et al. Rescue of {Delta}F508-CFTR trafficking and gating in human cystic fibrosis airway primary cultures by small molecules Am J Physiol Lung Cell Mol Physiol, June 1, 2006; 290(6): L1117 - L1130. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 Y. A. Kuryshev, L. Wang, B. A. Wible, X. Wan, and E. Ficker Antimony-Based Antileishmanial Compounds Prolong the Cardiac Action Potential by an Increase in Cardiac Calcium Currents Mol. Pharmacol., April 1, 2006; 69(4): 1216 - 1225. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. Hinzpeter, J. Lipecka, F. Brouillard, M. Baudoin-Legros, M. Dadlez, A. Edelman, and J. Fritsch Association between Hsp90 and the ClC-2 chloride channel upregulates channel function Am J Physiol Cell Physiol, January 1, 2006; 290(1): C45 - C56. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 H. Abriel and O. Staub Ubiquitylation of Ion Channels Physiology, December 1, 2005; 20(6): 398 - 407. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
D. J. Milan and C. A. MacRae Animal models for arrhythmias Cardiovasc Res, August 15, 2005; 67(3): 426 - 437. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 K. Mandal, E. Torsney, J. Poloniecki, A. J. Camm, Q. Xu, and M. Jahangiri Association of High Intracellular, But Not Serum, Heat Shock Protein 70 With Postoperative Atrial Fibrillation Ann. Thorac. Surg., March 1, 2005; 79(3): 865 - 871. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
Y. A. Kuryshev, E. Ficker, L. Wang, P. Hawryluk, A. T. Dennis, B. A. Wible, A. M. Brown, J. Kang, X.-L. Chen, K. Sawamura, et al. Pentamidine-Induced Long QT Syndrome and Block of hERG Trafficking J. Pharmacol. Exp. Ther., January 1, 2005; 312(1): 316 - 323. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C. Bellocq, R. Wilders, J.-J. Schott, B. Louerat-Oriou, P. Boisseau, H. Le Marec, D. Escande, and I. Baro A Common Antitussive Drug, Clobutinol, Precipitates the Long QT Syndrome 2 Mol. Pharmacol., November 1, 2004; 66(5): 1093 - 1102. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
D. K. Marsee, A. Venkateswaran, H. Tao, D. Vadysirisack, Z. Zhang, D. D. Vandre, and S. M. Jhiang Inhibition of Heat Shock Protein 90, a Novel RET/PTC1-associated Protein, Increases Radioiodide Accumulation in Thyroid Cells J. Biol. Chem., October 15, 2004; 279(42): 43990 - 43997. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
L. Gouas, C. Bellocq, M. Berthet, F. Potet, S. Demolombe, A. Forhan, R. Lescasse, F. Simon, B. Balkau, I. Denjoy, et al. New KCNQ1 mutations leading to haploinsufficiency in a general population: Defective trafficking of a KvLQT1 mutant Cardiovasc Res, July 1, 2004; 63(1): 60 - 68. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||