Mutant Caveolin-3 Induces Persistent Late Sodium Current and Is Associated With Long-QT Syndrome
Background— Congenital long-QT syndrome (LQTS) is a primary arrhythmogenic syndrome stemming from perturbed cardiac repolarization. LQTS, which affects ≈1 in 3000 persons, is 1 of the most common causes of autopsy-negative sudden death in the young. Since the sentinel discovery of cardiac channel gene mutations in LQTS in 1995, hundreds of mutations in 8 LQTS susceptibility genes have been identified. All 8 LQTS genotypes represent primary cardiac channel defects (ie, ion channelopathy) except LQT4, which is a functional channelopathy because of mutations in ankyrin-B. Approximately 25% of LQTS remains unexplained pathogenetically. We have pursued a “final common pathway” hypothesis to elicit novel LQTS-susceptibility genes. With the recent observation that the LQT3-associated, SCN5A-encoded cardiac sodium channel localizes in caveolae, which are known membrane microdomains whose major component in the striated muscle is caveolin-3, we hypothesized that mutations in caveolin-3 may represent a novel pathogenetic mechanism for LQTS.
Methods and Results— Using polymerase chain reaction, denaturing high-performance liquid chromatography, and direct DNA sequencing, we performed open reading frame/splice site mutational analysis on CAV3 in 905 unrelated patients referred for LQTS genetic testing. CAV3 mutations were engineered by site-directed mutagenesis and the molecular phenotype determined by transient heterologous expression into cell lines that stably express the cardiac sodium channel hNav1.5. We identified 4 novel mutations in CAV3-encoded caveolin-3 that were absent in >1000 control alleles. Electrophysiological analysis of sodium current in HEK293 cells stably expressing hNav1.5 and transiently transfected with wild-type and mutant caveolin-3 demonstrated that mutant caveolin-3 results in a 2- to 3-fold increase in late sodium current compared with wild-type caveolin-3. Our observations are similar to the increased late sodium current associated with LQT3-associated SCN5A mutations.
Conclusions— The present study reports the first CAV3 mutations in subjects with LQTS, and we provide functional data demonstrating a gain-of-function increase in late sodium current.
Received April 21, 2006; revision received August 29, 2006; accepted September 1, 2006.
Congenital long-QT syndrome (LQTS), a potentially fatal, heritable cardiac channelopathy characterized by delayed cardiac repolarization, prolonged QT interval, and lethal ventricular arrhythmias, results in significant risk of syncope, seizures, and sudden cardiac death.1,2 Genetic investigations demonstrate that mutations in α subunits of cardiac potassium and sodium channels account for ≈75% of congenital LQTS (LQT1, LQT2, and LQT3) and that ≈1% of LQTS are due to mutations involving ankyrin-B, potassium channel β subunits, or voltage-dependent L-type calcium channel α 1C subunit (LQT4, LQT5, LQT6, LQT7, and LQT8).3–9 Thus far, all known LQTS susceptibility genes encode cardiac ion channels (either α pore-forming subunits or auxiliary β subunits) except LQT4, caused by ANK2-encoded ankyrin-B. This protein, thought to participate in localization of sodium or calcium channels to the sarcolemma and mutations, results in ion channel dysfunction caused by altered localization. The genetic defect in ≈25% of patients with LQTS remains unknown. We hypothesized that mutations in an ion channel–associated protein, which disrupts ion channel function, similar to the functional mechanism of LQT4, could precipitate LQTS. We further hypothesized that CAV3, which encodes caveolin-3, the major scaffolding protein present in caveolae in heart, could be such an LQTS susceptibility gene.
Clinical Perspective p 2112
Caveolae are 50- to 100-nm omega-shaped microdomains of the plasmalemma, particularly abundant in cells of the cardiovascular system, including endothelial cells, smooth muscle cells, macrophages, cardiomyocytes, and fibroblasts.10 Caveolae are involved in vesicular trafficking and serve as a platform to organize and regulate a variety of signal transduction pathways; they also play an important part in cholesterol homeostasis.10 Caveolins are the principal proteins of caveolae. The 3 isoforms of caveolin are encoded by separate genes. Although caveolin-1 (CAV1) and caveolin-2 (CAV2) are coexpressed in most cell types, CAV3-encoded caveolin-3 is specifically expressed in cardiomyocytes and skeletal muscle. Some cardiac ion channels have been specifically localized to caveolae such as the SCN5A-encoded voltage-gated Na+ channel (hNav1.5),11 the voltage-dependent K+ channel (Kv1.5),12 the sodium-calcium exchanger,13 and the L-type Ca2+ channel.14 In the heart, a variety of other signaling molecules have been found in caveolae, including the β2-adrenergic receptor and associated proteins of the G-protein/adenylyl cyclase/protein kinase A pathway.14 Therefore, caveolae can serve both to compartmentalize and to regulate ion channel function and cell signaling factors. A critical role for caveolin-3 and caveolae in cardiac excitability is likely, but so far, no evidence has been provided for their involvement in heritable arrhythmia syndromes.
We therefore sought to investigate whether CAV3 may be implicated in LQTS. Furthermore, we investigated whether mutations in CAV3 could alter cardiac repolarization by modulating the cardiac SCN5A-encoded sodium channel activity because of the previous association of this channel with caveolae and the established pathogenic link between gain-of-function SCN5A mutations and LQT3. We discovered that perturbations in caveolin-3 constitute a novel pathogenic mechanism for LQTS.
Between 1998 and 2004, 905 (571 female, 63.3%) unrelated patients (364 from Baylor College of Medicine and Texas Children’s Hospital, 541 from the Mayo Clinic College of Medicine) were referred to either the Phoebe Willingham Muzzy Molecular Cardiology Laboratory at Baylor College of Medicine/Texas Children’s Hospital or Mayo Clinic’s Sudden Death Genomics Laboratory for LQTS genetic testing. All patients underwent physical examination, family history, and ECG analysis. The average age at diagnosis was 28±9 years (range, 0 to 81 years), and the resting QTc was 480±40 ms (range, 365 to 759 ms).
CAV3 Mutational Analysis
Blood was obtained after informed written consent was given for subjects referred for LQTS genetic testing. Genomic DNA was extracted from peripheral blood lymphocytes and lymphoblastoid cell lines, which were established as an alternative DNA source as previously described.15 Only a minimal number of patients had DNA extracted from lymphoblastoid cell lines for characterization, and no CAV3 mutations were identified in those samples. Using polymerase chain reaction (PCR), denaturing high-performance liquid chromatography, and direct DNA sequencing, we performed open reading frame/splice site mutational analysis on CAV3 (chromosome 3p25, 2 exons). PCR amplification was performed using coding region flanking primers designed in our laboratory with Oligo software. Primer sequences, PCR conditions, and denaturing high-performance liquid chromatography conditions are available on request.
Mutational analyses of the entire coding regions of the LQTS-associated genes—KCNQ1 (LQT1), KCNH2 (LQT2), SCN5A (LQT3), KCNE1 (LQT5), KCNE2 (LQT6), and KCNJ2 (LQT7 or ATS1)—and targeted analysis of ANK2 (LQT4) and RyR2 (CPVT1) were performed previously.8,15
All genetic variants regarded as putative LQTS-associated mutations changed a conserved residue or splice site that altered the primary amino acid structure of the encoded protein; this change must have been absent in at least 500 ethnically matched reference alleles. Control genomic DNA was obtained from the Human Genetic Cell Repository sponsored by the National Institute of General Medical Sciences and the Coriell Institute for Medical Research (Camden, NJ).
Synonymous single nucleotide polymorphisms were excluded from consideration.
CAV3 Cloning and Mutagenesis
The wild-type (WT) human CAV3 456-bp coding sequence was cloned from human cardiac cDNA by PCR using the proofreading Pfu+ Turbo DNA polymerase (Stratagene, La Jolla, Calif) to avoid improper nucleotide incorporation using the following primers: 5′-ATGATGGCAGAAGAGCACAC-3′ (sense) and 5′-TGGTGCTGCGGAAGGAGGTC-3′ (antisense). Site-directed mutagenesis was performed with the QuikChange Site-Directed Mutagenesis Kit (Stratagene) using the vector containing the CAV3-WT insert as a template. The primers used for mutagenesis are available on request; PCR and bacteria transformation were performed according to manufacturer’s instructions.
The mutated F97C-CV3 and S141R-CAV3 clones were sequenced to ensure the presence of the mutations and the absence of other substitutions introduced by the DNA polymerase.
Mammalian Cell Transfection
WT CAV3 and mutant CAV3 were subcloned into the mammalian expression vector pcDNA3 (Invitrogen Corp, Carlsbad, Calif) for transient transfection in HEK293 cells stably expressing hNav1.5. The stable HEK293 cell lines expressing hNav1.5 were established as previously described.16 The WT or mutant DNA was transiently cotransfected with green fluorescent protein (GFP) at a ratio of 1:10 into the HEK293 cell line with SuperFect from Qiagen (Valencia, Calif). DNA mixture (sample DNA plus GFP) was diluted to 150 μL with OptiMEM (Gibco, Invitrogen Corp). The 150 μL DNA and OptiMEM mixture was mixed with SuperFect at a ratio of 1:5 and incubated for 10 minutes at room temperature. After 10 minutes of incubation, 1 mL MEM-complete media was added to the DNA and SuperFect mixture for 5 hours of transfection under normal growth conditions. After 5 hours of transfection, the transfected cell was washed twice with phosphate-buffered saline (PBS) and returned to normal growth conditions with 3 mL media for 24 hours before macroscopic sodium current was measured.
Macroscopic sodium current (INa) was measured at 22°C to 24°C with the standard whole-cell patch-clamp method as previously described.17 Briefly, an Axopatch 200B amplifier and pClamp8.0 software (Axon Instruments, Foster City, Calif) were used with series resistance compensation and a low-pass filter of 5 kHz. The extracellular (bath) solution contained (in mmol/L) 140 NaCl, 4 KCl, 1.8 CaCl2, 0.75 MgCl2, and 5 HEPES. The pipette solution contained (in mmol/L) 120 CsF, 20 CsCl, 2 ethyleneglycoltetraacetic acid, and 5 HEPES. Standard activation, steady-state inactivation, and recovery protocols were performed, and the results were fit to a Boltzmann equation for activation and inactivation and recovery data were fit to an exponential equation with protocol details as previously reported.17 Late INa was measured at the end of a 775-ms depolarization from −140 to 0 mV after passive leak subtraction as previously described.17 We have previously shown that this leak subtraction method is comparable to saxitoxin subtraction methods. Data are reported as mean±SD. The Kruskal-Wallis test was performed to determine statistical significance among multiple groups. Statistical significance was determined by a value of P<0.05.
WT and mutant CAV3 constructs were subcloned into a bicistronic vector (pIRESGFP, kind gift from David Johns, Johns Hopkins University) to express the desired CAV3 construct and GFP under control of a cytomegalovirus promoter.
SCN5A was subcloned into pcDNA3, a mammalian expression vector (Invitrogen). Plasmids containing cDNAs encoding WT SCN5A, WT CAV3, and mutations of CAV3 were cotransfected in the following combinations: WT SCN5A+WT CAV3, WT SCN5A+CAV3-P104L, WT SCN5A+CAV3-S141R, and WT SCN5A+CAV3-F97C in HEK293 cells using the transfection reagent FuGENE 6 (Roche Applied Science, Indianapolis, Ind), according to the manufacturer’s protocol. Further details of the immunoprecipitation procedures can be found in the online-only Data Supplement.
Immune complexes were analyzed by sodium dodecyl sulfate–polyacrylamide gel electrophoresis (4% to 15% gradient gels, Bio-Rad, Hercules, Calif) and transferred to polyvinylidene difluoride membranes. Nonspecific binding sites were blocked by immersion of membranes at 4°C for 30 minutes in PBS-Tween (0.1%, Tween-20) containing 5% (wt/vol) dried skim milk. Membranes were subsequently probed with an anti-hNav1.5 antibody (rabbit polyclonal, 1:200, directed against conserved segment in the intracellular III-IV loop, Upstate USA, Inc, Charlottesville, Va) and an anti–caveolin-3 antibody (1:10 000, BD Biosciences, San Diego, Calif).
To evaluate the localization of CAV3-encoded caveolin-3 and SCN5A-encoded hNav1.5 in human heart, 5-μm sections from frozen right myocardial ventricles were stained with a goat-raised anti–caveolin-3 antibody and a rabbit-raised anti-hNav1.5 antibody (Santa Cruz Biotechnology, Inc, Santa Cruz, Calif). Each primary antibody was diluted 1:500 in PBS and incubated with the sections for 1 hour at room temperature. After 3 washes in 1× PBS, the slides were incubated for 30 minutes at room temperature with the following secondary antibodies: a fluorescein isothiocyanate–conjugated anti-goat antibody to detect caveolin-3 and the Texas red–conjugated anti-rabbit antibody to identify hNav1.5 (Molecular Probes, Carlsbad, Calif), diluted 1:500, and then mounted as previously described.18
The authors had full access to the data and take full responsibility for their integrity. All authors have read and agree to the manuscript as written.
Clinical Evaluation and Mutation Analysis
Mutational analysis of CAV3 in 905 unrelated LQTS subjects identified 6 nonsynonymous single nucleotide polymorphisms (Figure 1A) in 17 patients (1.9%). Sequencing analysis identified the single nucleotide polymorphisms as heterozygous missense mutations (Figure 1B) leading to amino acid changes in residues conserved across several species (Figure 1C).
The novel mutation F97C was observed in a 13-year-old female asthmatic patient presenting with shortness of breath and chest pain. The ECG demonstrated marked QT prolongation (QTc, 532 ms; Figure 2 and Table 1, case 5). Notably, this prolongation was present only but reproducibly on β-agonist inhaler therapy for her asthma. Her family history was unremarkable, and screening ECGs among all first-degree relatives revealed normal QTc (data not shown). Genetic analysis revealed a de novo 290T>C mutation (Figure 1B), leading to F97C, involving a highly conserved residue (Figure 1C) in the transmembrane domain (Figure 1A).
The 423C>G mutation, resulting in the missense mutation S141R, was discovered in a 16-year-old white male patient who presented with nonexertional syncope and a QTc of 480 ms (Table 1, case 6). The mutation localizes to the functional C-terminal domain of caveolin-3 and modifies a conserved residue (Figure 1A through 1C). Comprehensive open reading frame/splice site mutational analysis of all other known LQTS-associated genes was negative. Consistent with the negative family history and normal screening ECGs among first-degree relatives, genetic testing confirmed that S141R was also a spontaneous de novo mutation.
The other novel CAV3 mutations (T78M and A85T), which alter highly conserved residues, were identified in 4 patients (Figure 1 and Table 1) and, like F97C and S141R, were absent in >1000 control alleles. T78M was noted in 3 probands with a family history from small kindreds (cases 1 through 3), with 1 proband also possessing a published LQT2-associated KCNH2 mutation (A913V). The A85T mutation was elucidated as a de novo mutation in 1 case (Table 1, case 4).
In addition, we identified the G56S and C72W common polymorphisms, localized to the caveolin-3 scaffolding domain (Figure 1A).19,20 In the present study, we compared the frequency of the G56S polymorphism among healthy black versus healthy white controls using the Fisher exact test. G56S was identified in 25% of blacks and 0% of white controls (P<10−7), indicating that G56S is a relatively black-specific genetic variant. With a prevalence similar to that found in the black controls, 8 black cases had the G56S polymorphism, and 5 of these 8 black cases also had mutations in other established LQTS-causing channels. Similarly, C72W was observed in 3 white LQTS cases, 0% of white controls, and 2% of black controls.
CAV3 and hNav1.5 Localization in Human Myocardium
Although colocalization of the cardiac sodium channel and caveolin-3 at the sarcolemma of cells in myocardial tissue has been already established in rat ventricular cardiomyocytes,11 we sought to examine whether caveolin-3 and hNav1.5 also colocalize in human myocardium. Sections from human right ventricular biopsies were used for immunofluorescence labeling of caveolin-3 and hNav1.5. Figure 3 demonstrates that both caveolin-3 and hNav1.5 colocalize at the plasma membrane of human cardiomyocytes, suggesting that in human heart, caveolin-3 and hNav1.5 may be part of the same macromolecular protein complex, which substantiates the possible pathological effect of mutant caveolin-3 on the cardiac sodium channel hNav1.5. Although cardiac expression of caveolin-3 has previously been established in human subjects,21 to the best of our knowledge, this is the first report demonstrating caveolin-3 and hNav1.5 colocalization in human myocardium by immunohistochemical analysis.
Based on their absence in >1000 reference alleles, the nature of amino acid substitution and conservation across species, sporadic de novo status, and the lack of any other LQTS-associated mutations, F97C and S141R represented the strongest candidate mutations for the pathogenesis of LQTS and were functionally characterized. WT CAV3 (Figure 4B) and mutant CAV3 (Figure 4C and 4D) were transfected into a stable hNav1.5-expressing HEK293 cell line, as well as an empty vector control (Figure 4A), and robust INa was present in all cases. Summary data for peak current density and kinetic parameters of activation, and inactivation also showed no differences (Table 2). For recovery parameters, the caveolin-3 mutations tended to have a smaller recovery time constant for the slow component of recovery by about half, but this did not reach statistical significance for the F97C mutant (Table 2). A small amount of INa continued to flow after the initial rapid inactivation, and inspection of the traces in Figure 4 suggests that this late INa may be increased in the mutants, especially F97C. When this was rigorously examined by prolonged depolarization and careful leak subtraction, late INa was clearly increased in the presence of each of the caveolin-3 mutations, F97C and S141R (Figure 5A). Summary data (Figure 5B) show a 4-fold increase in late INa for F97C and a 3-fold increase for S141R mutants. This degree of accentuated late INa is equal to that observed in patients established to have LQT3-associated mutations in SCN5A.22
Immunoprecipitation of hNav1.5 and Caveolin-3
Given that a previous study has suggested that hNav1.5 and CAV3 are associated on the basis of their coimmunoprecipitation from cardiac muscle lysates,11 we sought to determine whether the increase in late INa induced by the caveolin-3 mutants was conferred by a loss of the association between normal hNav1.5 and mutant caveolin-3. HEK293 cells were transfected with WT SCN5A+WT CAV3, WT SCN5A+P104L-CAV3, and WT SCN5A+S141R-CAV3, SCN5A+F97C-CAV3, and these cell lysates were solubilized in Triton X-100 and n-octyl-d-glucoside containing buffer and subjected to immunoprecipitation with anti–caveolin-3 or control mouse immunoglobulin G. Immune complexes were analyzed with Western blotting. In Figure 6, anti-hNav1.5 antibody detected an immunoreactive band for the hNav1.5 channel at 240 kDa in the lysate lane and anti–caveolin-3 immunoprecipitation lanes, demonstrating that hNav1.5 coimmunoprecipitates with WT caveolin-3 in HEK293 cells, similar to the result in cardiac muscle. In the bottom part of the blot, caveolin-3 protein was detected in the lysate and the immunoprecipitation lanes at 18 kDa. Neither protein immunoprecipitated with control mouse immunoglobulin G. Furthermore, the LQTS-associated mutant F97C-caveolin-3 and S141R-caveolin-3 and the limb girdle muscular dystrophy–associated P104L-caveolin-3 immunoprecipitated with hNav1.5, similar to WT caveolin-3. These results suggest that these disease mutations in CAV3 do not result in the loss of association between hNav1.5 and caveolin-3 as a basis for the increase in late INa. Subtle changes in the interactions between the associated proteins would not be detected by this assay, however.
The pathogenetic basis for LQTS has focused principally on ion channels. Only recently, the finding of LQTS-causing mutations in ankyrin-B (LQT4) as a nonchannel protein has led to further exploration of secondary disruptions in ion channel metabolism (expression, posttranslational modification, trafficking, and electrophysiology).
Caveolae, within the plasmalemma, are compartments involved in the scaffolding of ion channels such as hNav1.523; their activity is critical in the development of ventricular action potential. Colocalization of hNav1.5 and caveolin-3 has been previously demonstrated at the sarcolemma of rat cardiomyocytes11 but not in human myocardium.
Here, we demonstrate that both caveolin-3 and hNav1.5 colocalize at the plasma membrane of human cardiomyocytes, suggesting that caveolin-3 and hNav1.5 are part of a macromolecular complex important in cardiac function across species, thus substantiating the possible pathological effect of mutant caveolin-3 on hNav1.5.
In addition, caveolin-3 modulates the sympathetic response in the heart by colocalizing key signaling molecules of the β2-adrenergic receptor cascade, including G proteins, adenyl cyclase, and protein kinase A, which can directly phosphorylate hNav1.5.24 It is intriguing to note that the adolescent female hosting the de novo F97C–caveolin-3 mutation displayed pronounced QT prolongation only during albuterol inhaler therapy for her asthma. This may suggest a role of F97C–caveolin-3 in altered sympathetic response, with a direct modulator effect on hNav1.5 or an effect that is mediated by G protein–coupled receptors and other signaling molecules such as protein kinase A. However, the β2-adrenergic receptor stimulation also may regulate other sensitive channels, notably Cav1.2 calcium channels, which also are associated with caveolae and may be involved in the response.14,25,26
Caveolin-3 abnormalities have been implicated previously in limb girdle muscular dystrophy, rippling muscle disease, hyper-CKemia, and distal myopathy, but no cardiac manifestations resulting from these mutations have been clearly described.19 A distinct CAV3 mutation (T63S) was identified in siblings with hypertrophic cardiomyopathy, although no clear mechanism has been defined.27 However, no signs of skeletal muscular involvement or primary cardiomyopathy were present in our patients.
Our findings demonstrate for the first time that caveolin-3 directly modifies hNav1.5 kinetics and subserves a novel pathogenic substrate for LQTS. We identified 6 nonsynonymous genetic variants in 905 unrelated patients referred for LQTS genetic testing, including the 2 common polymorphisms G56S and C72W.19
The 4 novel LQTS-associated mutations were absent in >1000 reference alleles and altered highly conserved residues predicted to affect the protein secondary structure using in silico models (data not shown), suggesting perturbation of normal caveolin-3 function.
On the basis of the aforementioned characteristics and the lack of any other LQTS-associated mutations, F97C and S141R represented the most compelling molecular candidates for LQTS pathogenesis. Electrophysiological analysis of F97C and S141R in a cell line with stable expression of hNav1.5 showed dramatic increases in late INa compared with hNav1.5 alone or WT CAV3. Interestingly, expression of WT CAV3 compared with empty plasmid in the HEK293 cells did not alter INa in any measurable way, suggesting that WT CAV3 does not modulate hNav1.5 function; however, the related protein flotillin-1 endogenously expressed in HEK293 cells may serve a role comparable to that of caveolin-3 in maintaining basal hNav1.5 function, as suggested previously for HCN4 channels expressed in this model system.28 Nevertheless, the mutant F97C-CAV3 and S141R-CAV3 clearly demonstrated a gain-of-function effect on late INa which is quantitatively similar to that reported in LQT3,22 suggesting that this may be 1 substrate for the prolonged QT interval and associated arrhythmias.
The molecular mechanisms linking these CAV3 mutations to altered hNav1.5 function are not fully defined, however. The immunoprecipitation data suggest that CAV3 mutations do not alter the overall association between caveolin-3 and cardiac sodium channels, so more subtle molecular interactions or channel regulation may be altered. For none of the previously identified modulators of late INa, including heart failure, ischemia, and cytoskeletal disruption, have the molecular mechanisms been defined.29 Thus, for the CAV3 mutations, significant future research is necessary to identify the precise mechanisms directly responsible for the increase in late INa.
Although we did not study the electrophysiological properties of the T78M-CAV3 mutation, it is important to note that both cases 1 and 2 presented with sinus bradycardia (Table 1). In 2003, Veldkamp and colleagues30 investigated in an elegant study the role of the 1795insD hNav1.5 mutant on sinoatrial node electrical activity. They demonstrated that 1795insD presented a persistent inward current and a negative shift in voltage dependence of inactivation, which causes a reduction in sinus rate, leading to bradycardia. This suggests that T78M could modulate the INa mimicking the 1795insD hNav1.5 mutant, consistent with the clinical features characterizing the LQT3 phenotype.30
Caveolin-3 is the second non–ion channel protein (after ankyrin-B) implicated in the pathogenesis of congenital LQTS. Consistent with the final common pathway hypothesis,31 genes encoding cardiac channel–interacting proteins, which secondarily disrupt ion channel function, may confer genetic susceptibility for LQTS. Caveolin-3 is one of a potentially large group of such proteins. In addition, hNav1.5 appears to be disrupted in a variety of arrhythmia-associated disorders, including LQTS and Brugada syndrome,32 both possible substrates of sudden infant death syndrome33 and atrioventricular block.34
In this report, although we evaluated a large cohort of patients with LQTS, we identified primarily de novo mutations in CAV3. The few small kindreds with T78M-CAV3 declined further genotyping. Thus, we did not have access to large multigenerational families with complete or nearly complete penetrance of the clinical LQTS phenotype, which could have provided cosegregation data further supporting the causality of CAV3 mutations in LQTS. In addition, the clinical data sets were not complete for all patients in the cohort, as is often the case for sudden death cases in young patients. Therefore, further analysis of certain mutations is needed to completely define the clinical phenotypes.
In addition, although mutations in the coding sequence and exon/intron boundaries35 of the candidate genes are the most obvious target for a first screening, defects other than in the coding sequence have indeed been discovered and associated with primary arrhythmias. In particular, a large DNA duplication in the gene coding for KCNH236 and a common sodium channel promoter haplotype associated with conduction defects among Asians37 have been described.
Therefore, although we focused on the coding sequence of CAV3 in this study, we believe that further investigation, including genomic rearrangements and noncoding sequence variants, may play a significant role in the orphan form of LQTS.
Finally, although we provide strong evidence for a mechanistic link between CAV3 mutations and LQTS mediated by an increase in late INa, we anticipate that modulation of other ion channels or signaling pathways regulating ionic currents also may contribute to the pathophysiology, given the localization of multiple ion channels and signaling proteins to caveolae in ventricular cardiomyocytes. Despite these limitations, we have identified novel mutations in CAV3 that increase the late sodium current and are specifically associated with LQTS.
We are grateful to Joel Chang, W. Buck Kyle, and Eileen Merkle for technical assistant. We wish to express our gratitude to the study participants and to the physicians who referred patients to these research programs.
Sources of Funding
Dr Vatta is funded by the National Heart, Lung, and Blood Institute (NHLBI R21HL077706 and R21HL078807). Dr Ackerman is supported by the Dr Scholl Foundation, the CJ Foundation for SIDS, an Established Investigator Award from the AHA, and RO1 HD42569 from the National Institutes of Health (NIH). Dr Ye is supported by the Scientist Development Grant (SDG-0435030N) from the AHA. Dr Makielski is supported by NIH RO1 HL71092, and Dr Kamp is supported by NIH P01 HL47053. Dr Towbin is funded by the Texas Children’s Foundation Chair in Pediatric Molecular Cardiology Research and NIH U01-HL65652.
Zareba W, Moss AJ, Schwartz PJ, Vincent GM, Robinson JL. Priori SG, Benhorin J, Locati EH, Towbin JA, Keating MT, Lehmann MH, Hall WJ. Influence of genotype on the clinical course of the long-QT syndrome: International Long-QT Syndrome Registry Research Group. N Engl J Med. 1998; 339: 960–965.
Mohler PJ, Schott JJ, Gramolini AO, Dilly KW, Guatimosim S, duBell WH, Song LS, Haurogne K, Kyndt F, Ali ME, Rogers TB, Lederer WJ, Escande D, Le Marec H, Bennett V. Ankyrin-B mutation causes type 4 long-QT cardiac arrhythmia and sudden cardiac death. Nature. 2003; 421: 634–639.
Splawski I, Timothy KW, Sharpe LM, Decher N, Kumar P, Bloise R, Napolitano C, Schwartz PJ, Joseph RM, Condouris K, Tager-Flusberg H, Priori SG, Sanguinetti MC, Keating MT. Ca(V)1.2 calcium channel dysfunction causes a multisystem disorder including arrhythmia and autism. Cell. 2004; 119: 19–31.
Yarbrough TL, Lu T, Lee HC, Shibata EF. Localization of cardiac sodium channels in caveolin-rich membrane domains: regulation of sodium current amplitude. Circ Res. 2002; 90: 443–449.
Martens JR, Sakamoto N, Sullivan SA, Grobaski TD, Tamkun MM. Isoform-specific localization of voltage-gated K+ channels to distinct lipid raft populations: targeting of Kv1.5 to caveolae. J Biol Chem. 2001; 276: 8409–8414.
Balijepalli RC, Foell JD, Hall DD, Hell JW, Kamp TJ. Localization of cardiac L-type Ca(2+) channels to a caveolar macromolecular signaling complex is required for beta(2)-adrenergic regulation. Proc Natl Acad Sci U S A. 2006; 103: 7500–7505.
Vatta M, Dumaine R, Varghese G, Richard TA, Shimizu W, Aihara N, Nademanee K, Brugada R, Brugada J, Veerakul G, Li H, Bowles NE, Brugada P, Antzelevitch C, Towbin JA. Genetic and biophysical basis of sudden unexplained nocturnal death syndrome (SUNDS), a disease allelic to Brugada syndrome. Hum Mol Genet. 2002; 11: 337–345.
Makielski JC, Ye B, Valdivia CR, Pagel MD, Pu J, Tester DJ, Ackerman MJ. A ubiquitous splice variant and a common polymorphism affect heterologous expression of recombinant human SCN5A heart sodium channels. Circ Res. 2003; 93: 821–828.
Nagatomo T, Fan Z, Ye B, Tonkovich GS, January CT, Kyle JW, Makielski JC. Temperature dependence of early and late currents in human cardiac wild-type and long Q-T DeltaKPQ Na+ channels: mutations in the caveolin-3 gene: when are they pathogenic? Am J Physiol. 1998; 275 (pt 2): H2016–H2024.
McNally EM, de Sa Moreira E, Duggan DJ, Bonnemann CG, Lisanti MP, Lidov HG, Vainzof M, Passos-Bueno MR, Hoffman EP, Zatz M, Kunkel LM. Caveolin-3 in muscular dystrophy. Hum Mol Genet. 1998; 7: 871–877.
Cagliani R, Bresolin N, Prelle A, Gallanti A, Fortunato F, Sironi M, Ciscato P, Fagiolari G, Bonato S, Galbiati S, Corti S, Lamperti C, Moggio M, Comi GP. A CAV3 microdeletion differentially affects skeletal muscle and myocardium. Neurology. 2003; 61: 1513–1519.
Feron O, Kelly RA. Gaining respectability: membrane-delimited, caveolar-restricted activation of ion channels. Circ Res. 2002; 90: 369–370.
Rybin VO, Xu X, Lisanti MP, Steinberg SF. Differential targeting of beta-adrenergic receptor subtypes and adenylyl cyclase to cardiomyocyte caveolae: a mechanism to functionally regulate the cAMP signaling pathway. J Biol Chem. 2000; 275: 41447–41457.
Davare MA, Avdonin V, Hall DD, Peden EM, Burette A, Weinberg RJ, Horne MC, Hoshi T, Hell JW. A beta2 adrenergic receptor signaling complex assembled with the Ca2+ channel Cav1.2. Science. 2001; 293: 98–101.
Hayashi T, Arimura T, Ueda K, Shibata H, Hohda S, Takahashi M, Hori H, Koga Y, Oka N, Imaizumi T, Yasunami M, Kimura A. Identification and functional analysis of a caveolin-3 mutation associated with familial hypertrophic cardiomyopathy. Biochem Biophys Res Commun. 2004; 313: 178–184.
Barbuti A, Gravante B, Riolfo M, Milanesi R, Terragni B, DiFrancesco D. Localization of pacemaker channels in lipid rafts regulates channel kinetics. Circ Res. 2004; 94: 1325–1331.
Veldkamp MW, Wilders R, Baartscheer A, Zegers JG, Bezzina CR, Wilde AA. Contribution of sodium channel mutations to bradycardia and sinus node dysfunction in LQT3 families. Circ Res. 2003; 92: 976–983.
Bezzina CR, Shimizu W, Yang P, Koopmann TT, Tanck MW, Miyamoto Y, Kamakura S, Roden DM, Wilde AA. Common sodium channel promoter haplotype in Asian subjects underlies variability in cardiac conduction. Circulation. 2006; 113: 338–344.
Long-QT syndrome (LQTS) is a potentially lethal, heritable arrhythmia syndrome affecting ≈1 in 3000 persons. Since the sentinel discovery of cardiac channel mutations as the pathogenic basis for LQTS in 1995, LQTS has been viewed as a “cardiac channelopathy.” To date, 8 LQTS-susceptibility genes have been discovered; 7 of these genes encode critical ion channel subunits. The other gene (ANK2) encodes an adapter protein, ankyrin-B. Mutations in these 8 genes account for ≈75% of LQTS. The search continues for novel pathogenic mechanisms. In the present study, we provide the first evidence that the CAV3-encoded scaffolding protein caveolin-3 constitutes the ninth LQTS-susceptibility gene. Caveolin-3 is expressed specifically in striated muscle and localizes to specialized membrane microdomains (caveolae), little caves in which certain ion channels reside together with specific cellular signaling molecules. Importantly, the LQTS-associated mutations in caveolin-3 alter the properties of the caveolar-localized cardiac sodium channel (Nav1.5) and confer a gain-of-function phenotype on the structurally intact sodium channel. Our findings are likely to expand the perspective of the genetic underpinnings of the cardiac channelopathies. Although scaffolding proteins play critical roles in modulating ion channel function, the concept that inherited arrhythmia syndromes like LQTS may originate from defects in scaffolding or adaptor proteins rather than the channel itself opens the field to an entire class of candidate genes to consider for the remnant of LQTS (25%) that awaits a pathogenic explanation. In addition, these observations suggest that treatments aimed at scaffolding proteins rather than specific ion channels may represent an alternative antiarrhythmic strategy.
↵*The first 2 authors contributed equally to this work.
The online-only Data Supplement is available with this article at http://circ.ahajournals.org/cgi/content/full/CIRCULATIONAHA.106.635268/DC1.