Defining the Cellular Phenotype of “Ankyrin-B Syndrome” Variants
Human ANK2 Variants Associated With Clinical Phenotypes Display a Spectrum of Activities in Cardiomyocytes
Background— Mutations in the ankyrin-B gene (ANK2) cause type 4 long-QT syndrome and have been described in kindreds with other arrhythmias. The frequency of ANK2 variants in large populations and molecular mechanisms underlying the variability in the clinical phenotypes are not established. More importantly, there is no cellular explanation for the range of severity of cardiac phenotypes associated with specific ANK2 variants.
Methods and Results— We performed a comprehensive screen of ANK2 in populations (control, congenital arrhythmia, drug-induced long-QT syndrome) of different ethnicities to discover unidentified ANK2 variants. We identified 7 novel nonsynonymous ANK2 variants; 4 displayed abnormal activity in cardiomyocytes. Including the 4 new variants, 9 human ANK2 loss-of-function variants have been identified. However, the clinical phenotypes associated with these variants vary strikingly, from no obvious phenotype to manifest long-QT syndrome and sudden death, suggesting that mutants confer a spectrum of cellular phenotypes. We then characterized the relative severity of loss-of-function properties of all 9 nonsynonymous ANK2 variants identified to date in primary cardiomyocytes and identified a range of in vitro phenotypes, including wild-type, simple loss-of-function, and severe loss-of-function activity, seen with the variants causing severe human phenotypes.
Conclusions— We present the first description of differences in cellular phenotypes conferred by specific ANK2 variants. We propose that the various degrees of ankyrin-B loss of function contribute to the range of severity of cardiac dysfunction. These data identify ANK2 variants as modulators of human arrhythmias, provide the first insight into the clinical spectrum of “ankyrin-B syndrome,” and reinforce the role of ankyrin-B–dependent protein interactions in regulating cardiac electrogenesis.
Received August 3, 2006; accepted November 6, 2006.
Cardiac contraction events are tightly controlled by the coordinate activities of a large collection of ion channels and transporters. Proper physiological function of each cardiac ion channel and transporter is strictly governed by defined biophysical properties and precise expression and localization within appropriate excitable membrane domains. Over the past 10 years, many of the molecular components responsible for regulating the biophysical properties of cardiac ion channels and transporters have been identified.1–3 This information has been critical for defining a number of fatal arrhythmias based on gene variants that affect the function of ion channels (or channel subunits) that control ventricular depolarization and repolarization.1,4
Editorial p 428
Clinical Perspective p 441
Ankyrins localize ion channels and transporters at specialized membrane domains in excitable and nonexcitable cells.5,6 Dysfunction in ankyrin-B is linked with fatal arrhythmia in humans.7,8 The ANK2 A4274G variant leading to E1425G causes the type 4 form of the congenital long-QT syndrome (LQTS).9 Affected individuals display QT interval prolongation, stress- and/or exercise-induced polymorphic ventricular arrhythmia, syncope, and sudden cardiac death. Individuals heterozygous for ankyrin-B E1425G also have sinus node dysfunction (bradycardia or junctional escape rhythm) and episodes of atrial fibrillation.8,9 In primary cardiomyocytes, E1425G leads to loss of ankyrin-B function.8 Subsequent evaluation of additional probands heterozygous for ankyrin-B E1425G and identification and characterization of the phenotypes of probands with other ankyrin-B loss-of-function variants (L1622I, T1626N, R1788W, E1813K) demonstrated that the clinical phenotypes often extend beyond the typical LQTS, leading to the label “ankyrin-B syndrome”; these phenotypes include bradycardia, sinus arrhythmia, delayed conduction/conduction block, idiopathic ventricular fibrillation, and catecholaminergic polymorphic ventricular tachycardia.7 To date, no biophysical mechanism has been offered for the range of severity of cardiac phenotypes associated with different ANK2 variants.
Here, we report 4 novel ANK2 loss-of-function variants in individuals with a variety of cardiac phenotypes. Moreover, we identify for the first time a spectrum of in vitro functional defects conferred by these and previously identified variants from wild-type to severe loss-of-function activity. Interestingly, carriers of ANK2 variants that display severe loss-of-function activity in cardiomyocytes display the most severe clinical phenotypes. These data confirm ANK2 variants as modulators for human arrhythmias, provide a mechanism for the clinical spectrum of the ankyrin-B syndrome, and reinforce the role of protein interactions involving ankyrin-B in regulating normal cardiac electrogenesis.
Genotyping and DNA Analysis
Published primers designed to amplify individual exons and flanking introns were used, and polymerase chain reactions were optimized.8 Polymerase chain reaction products were screened for variants with the SpectruMedix Reveal temperature gradient capillary electrophoresis system, and fragments containing variants were then directly resequenced. Exons 1 through 46 of ANK2 (except for brain-specific exon 38) were screened in the Coriell human variation panels, and exons 35 through 46 (excluding exon 38) were screened in the patient panel. The congenital arrhythmia syndrome panel included patients with LQTS and other entities such as idiopathic ventricular fibrillation or drug-induced LQTS. Drug-induced LQTS was defined as previously reported.10 The anonymized samples screened, obtained from the North American Human Variation Panels at Coriell, included the following panels: black (NA17149 through NA17199), white (NA17249 through NA17296), Han People of Los Angeles (NA17733 through NA17791), and Mexican American Community of Los Angeles (NA17438 through NA17467, NA17614 through NA17642). Specifically, we used AA NA17149 through NA17161, NA17163, and NA17165 through NA17199; Cauc NA17249 through NA17296 and NA17298; Mexican NA17438 through NA17446, NA17448 through NA17454, NA17456 through NA17463, NA17465 through NA17467, NA17614 through NA17419, NA17622, NA17624, NA17626, NA17629 through NA17634, NA17636 through NA17637, and NA17639 through NA17642; and HAN NA17733 through NA17747, NA17749, NA17752 through NA17759, NA17761 to NA17762, NA17764 to NA17771, NA17773 to NA17776, NA17779, NA17780, NA17782, NA17783, NA17785, NA17787, and NA17789 to NA17791. All genetic information generated in these studies has been deposited in the PharmGKB database at www.pharmgkb.org.
Animal care was in accordance with institutional guidelines. Neonatal cardiomyocytes were isolated and transfected from wild-type and ankyrin-B+/− postnatal day 1 mice (backcrossed >17 generations on pure C57/Bl6 background).8 After cDNA expression, we measured spontaneous contraction rates and/or Ca+2 diastolic/systolic peak intensity ratios as a function of time.8 For each variant, we analyzed >50 green fluorescent protein (GFP)–positive cells from >3 preparations. Relative levels of GFP–ankyrin-B expression were carefully monitored by immunoblot and immunostaining in each expression experiment to ensure that cellular phenotypes were the result of variant activity and not differences in expression levels.
Whole-cell lysates from wild-type, ankyrin-B+/−, and ankyrin-B−/− cells transfected with 220-kDa ankyrin-B–GFP variants were prepared, and 50 μg lysate was analyzed by SDS-PAGE and immunoblot.
Antibodies used for experiments include affinity-purified antibodies to ankyrin-B, GFP (monoclonal and polyclonal), InsP3 receptor (Affinity Bioreagents, Golden, Colo, and affinity-purified pan InsP3 receptor Ig), Na/Ca exchanger 1 (RDI), α-actinin (Sigma), and Na/K ATPase (Affinity Bioreagents).
Neonatal cardiomyocytes were washed with warm phosphate-buffered saline (pH 7.4) and fixed in 2% paraformaldehyde (37°C). Cells were blocked/permeabilized in PBS containing 0.075% Triton X-100 and 3% fish oil gelatin (Sigma) and incubated in primary antibody overnight at 4°C. After they were washed (phosphate-buffered saline plus 0.075% Triton X-100), cells were incubated in secondary antibody (goat anti-mouse, goat anti-rabbit Alexa 488 or 568, Molecular Probes, Carlsbad, Calif) for 8 hours at 4°C and mounted with Vectashield (Vector Laboratories, Burlingame, Calif) and No. 1 coverslips. Images were collected on a Zeiss 510 Meta confocal microscope (40 power oil 1.4 NA [Zeiss, Thornwood, NY] or 63 power oil 1.4 NA [Zeiss]; pinhole equals 1.0 Airy Disc) with Carl Zeiss Imaging software. All channels were collected on PMT3. Images were imported into Adobe Photoshop for cropping and linear contrast adjustment. Our cellular protocol is optimized to express only minimal levels of ankyrin-B–GFP to avoid any potential artifacts caused by overexpression. Note that unlike the adult cardiomyocyte, the neonatal cardiomyocyte does not display an organized transverse-tubule network. Therefore, Na/Ca exchanger 1 localization is primarily at the plasma membrane. Additionally, unlike the adult myocyte (primarily perinuclear), the InsP3 receptor in the differentiated neonatal cardiomyocyte is generally spread across the sarcoplasmic reticulum network.
We introduced single-amino-acid changes to full-length 220-kDa ankyrin-B–GFP (C-terminal enhanced green fluorescent protein fusion) with Quickchange mutagenesis (Stratagene, Garden Grove, Calif) to produce T1404I, G1406C, R1450W, L1503V, V1516D, T1552N, and V1777M. Once the mutation was confirmed by sequencing, a small fragment containing the desired mutation was subcloned into a parental plasmid to ensure that no additional variants were introduced into the parental plasmid by the mutagenesis polymerase chain reaction protocol (≈11 kb). Immunoblots were performed on transfected HEK293 cells to ensure that variant constructs produced full-length proteins.
When appropriate, data were analyzed by use of a 2-tailed Student t test. Values of P<0.05 were considered significant. Values are expressed as mean±SD.
The authors had full access to and take full responsibility for the integrity of the data. All authors have read and agree to the manuscript as written.
Identification of ANK2 Variants in Humans With Cardiac Dysfunction
We screened exons 35 through 46 and flanking intronic regions of ANK2 in 445 individuals with congenital arrhythmia (including LQTS and other entities such as idiopathic ventricular fibrillation) or drug-induced LQTS for potential ANK2 variants. In all subjects, screening the coding regions of the disease-associated genes KCNQ1, KCNH2, KCNE1, KCNE2, and SCN5A did not yield mutations. We identified 8 variants in ANK2: 3 coding region changes that resulted in amino acid substitutions, 2 variants that did not result in amino acid substitutions, and 3 changes to intronic regions.
The transition from 4211 to 4212CG-TA (T1404I; Table 1) was identified in a 54-year-old male black proband with bradycardia, atrial fibrillation, and quinidine-induced LQTS (>630 ms). The T4547A transition (resulting in V1516D) was identified in 4 probands: 1 with drug-induced LQTS and 3 with congenital arrhythmia (Table 1). The first was a 50-year-old white woman who developed QTc prolongation (>700 ms) and multiple episodes of polymorphic ventricular tachycardia torsades de pointes after quinidine treatment for atrial fibrillation. The second was a 19-year-old white man who presented with cardiac arrest while dancing. He subsequently had episodes of polymorphic ventricular tachycardia, with a normal QTc (417 ms) and T-wave morphology while in sinus rhythm. There was a family history of sudden cardiac death. The third proband was a 27-year-old white man who presented with syncope during a soccer match and an ECG showing ventricular tachycardia at 280 bpm. He also had bradycardia, a QTc of 460 ms, and a history of palpitations after stress. He developed atrial flutter after a stress test and was successfully treated with flecainide. There was no history of sudden cardiac death. The fourth V1516D proband was a 33-year-old white man resuscitated from cardiac arrest while driving. His ECG showed a typical (type 1) pattern of Brugada syndrome.11 An ajmaline infusion during cardiac electrophysiological testing was followed by sustained ventricular tachycardia requiring cardioversion. There was a family history of sudden cardiac death.
The G5437A single nucleotide polymorphism (leading to E1813K) was identified in 2 probands with atrial fibrillation, and both developed ventricular arrhythmias during drug treatment (Table 1). The first was a 79-year-old white woman who developed a long QTc (670 ms) and torsades de pointes during procainamide treatment. The second was a 34-year-old white woman who developed a long QTc (>600 ms) and syncope during treatment with quinidine. This specific variant had been identified in a previous ANK2 screen of probands with arrhythmia phenotypes.7
Identification of ANK2 Variants in Panels of Defined Ethnicity Asymptomatic Humans
We screened all exons and flanking intronic regions of ANK2 in 190 anonymized subjects of 4 defined ethnicities (Han Chinese, black, white, and Mexican American) to define other ANK2 variants. We identified 7 subjects with 6 different ANK2 nonsynonymous single nucleotide changes (Table 2). These variants were G4216T (G1406C; 1 Mexican American), C4348T (R1450W; 1 white), C4507G (L1503V; 1 black), C4655A (T1552N; 2 blacks), G5329A (V1777M; 1 Han Chinese), and G5437A (E1813K; 1 white; Table 2). In addition, the screen identified 11 synonymous coding single nucleotide polymorphisms, 17 intronic single nucleotide polymorphisms, 1 intronic deletion, and 1 intronic insertion that are not considered further here. Therefore, ANK2 variants are present in the normal population from multiple ethnic groups.
Human ANK2 Variants Display Ankyrin-B Loss-of-Function Activity
Unlike wild-type mouse cardiomyocytes, ankyrin-B+/− and ankyrin-B−/− neonatal cardiomyocytes display reduced frequency of spontaneous contractions, abnormal spatial/temporal patterns of cytoplasmic Ca2+ release,8,12 and aberrant localization and expression of multiple cardiac ion channels and transporters, including Na/Ca exchanger 1, Na/K ATPase, and InsP3 receptor (Figure 1A and 1B).8,12 Normal neonatal cardiomyocyte properties are restored in ankyrin-B+/− and ankyrin-B−/− neonatal cardiomyocytes transfected with ankyrin-B (220-kDa ankyrin-B; Figure 1C).6–8,12–15 Previously characterized ankyrin-B variants E1425G, L1622I, T1626N, R1788W, and E1813K display normal ankyrin-B expression and localization when transfected into ankyrin-B+/− and ankyrin-B−/− neonatal cardiomyocytes.7,8 However, these variants do not restore normal contraction rates, spatial/temporal patterns of Ca2+ release, or ion channel and transporter expression and localization in ankyrin-B+/− and ankyrin-B−/− neonatal cardiomyocytes,7,8 demonstrating that they confer an “altered function” phenotype. Accordingly, we used this ankyrin-B+/− neonatal cardiomyocyte rescue assay to evaluate new human variants for altered ankyrin-B function.
Three newly identified variants, ankyrin-B G1406C, R1450W, and L1503V, displayed properties similar to wild-type ankyrin-B. Similar to ankyrin-B–GFP, GFP-linked constructs of these variants were normally expressed and properly targeted within ankyrin-B+/− neonatal cardiomyocytes (Figure 1, Table 3). Additionally, like ankyrin-B–GFP, these 3 variants rescued abnormal ankyrin-B+/− and ankyrin-B−/− phenotypes, including abnormal contraction rates and aberrant spatial/temporal patterns of cytoplasmic Ca2+ release (Figure 2). Moreover, ankyrin-B–GFP G1406C, R1450W, and L1503V variants rescued abnormal localization and expression patterns of InsP3 receptor (Figure 1), Na/Ca exchanger 1, and Na/K ATPase (not shown).
In contrast, ankyrin-B variants T1404I, V1516D, T1552N, V1777M, and E1813K were ankyrin-B loss-of-function mutations in ankyrin-B+/− neonatal cardiomyocytes. Although none of the variants affected the localization of ankyrin-B–GFP itself (Figure 1), ankyrin-B+/− neonatal cardiomyocytes transfected with ankyrin-B–GFP variants T1404I, V1516D, T1552N, V1777M, and E1813K at equivalent levels displayed abnormal contraction rates and aberrant spatial/temporal patterns of Ca2+ release (Figure 2). Additionally, neonatal cardiomyocytes transfected with these variants at similar levels displayed abnormal localization and expression of InsP3 receptor (Figure 1), Na/Ca exchanger 1, and Na/K ATPase (not shown). Specifically, in ankyrin-B+/− neonatal cardiomyocytes transfected with T1404I-GFP, V1516D-GFP, T1552N-GFP, V1777M-GFP, and E1813K-GFP, ankyrin-B–associated ion channels and transporters were clustered near the perinuclear region in a pattern resembling the cardiomyocyte Golgi apparatus (Figure 1D and 1H through 1J).15 Together, these data identify 4 new human ankyrin-B variants that display ankyrin-B loss-of-function activity in cardiomyocytes.
Specific Ankyrin-B Variants Display Severe Loss-of-Function Activity in Cardiomyocytes
To date, 9 different human ankyrin-B variants have been identified that demonstrate loss-of-function activity in ankyrin-B+/− and ankyrin-B−/− neonatal cardiomyocytes.7,8 However, the clinical phenotypes associated with these variants vary strikingly, from no obvious phenotype to the manifest LQTS and sudden death. Thus, different individual variants may lead to various degrees of cellular ankyrin-B loss of function that may correlate with a range of severity of cardiac dysfunction.
To determine whether there is a spectrum of in vitro severity of ankyrin-B loss-of-function variants, we expressed ankyrin-B–GFP and all published ankyrin-B loss-of-function variants in wild-type neonatal cardiomyocytes. Wild-type myocytes were used because we hypothesized that variants associated with the most severe clinical phenotypes may still display strong loss of function, even in the presence of normal wild-type ankyrin-B activity. As shown in Figure 3, all ankyrin-B variants were normally expressed and targeted in wild-type neonatal cardiomyocytes. Interestingly, ankyrin-B–GFP significantly increased neonatal cardiomyocyte contraction rates compared with nontransfected cells (Figure 4; ≈13%; n=50; P<0.05). We observed no difference in contraction rates between nontransfected neonatal cardiomyocytes and wild-type neonatal cardiomyocytes transfected with the ankyrin-B–GFP loss-of-function variants T1404I, T1552N, V1777M, and E1813K (Figure 4; n >50 cells analyzed per mutant; P=NS). Wild-type neonatal cardiomyocytes expressing ankyrin-B–GFP L1622I or T1626N displayed a small but significant (≈14%) decrease in contraction rates and small but significant differences in spatial/temporal Ca2+ release patterns compared with nontransfected cells (Figure 4, n >50 cells per mutant; P<0.05). In contrast, wild-type neonatal cardiomyocytes expressing loss-of-function variants E1425G, V1516D, and R1788W displayed pronounced reductions (>60%) in both contraction rates and cytoplasmic calcium release events compared with control neonatal cardiomyocytes (n >50 neonatal cardiomyocytes evaluated for each mutant; P<0.05). In fact, contraction rates in these cells were similar to rates in nontransfected ankyrin-B+/− or ankyrin-B−/− neonatal cardiomyocytes (Figure 4). All mutants were expressed in cardiomyocytes at equivalent levels as assessed by staining intensity as well as immunoblot (GFP Ig) on transfected cardiomyocyte lysates. Similar to results from previously identified loss-of-function mutants, the loss-of-function phenotypes observed were not due to low transfection efficiencies or abnormal expression levels of ankyrin-B mutants in transfected neonatal cardiomyocytes.7,8,12,14
We next assessed ankyrin-B–associated ion channel and transporter localization in wild-type neonatal cardiomyocytes that expressed ankyrin-B variants with loss-of-function activity (Figure 3). In control cells or cells expressing ankyrin-B–GFP, we observed normal expression and localization of InsP3 receptor, Na/Ca exchanger 1 (Figure 3), and Na/K ATPase (not shown). We also observed normal expression and localization of ankyrin-B–GFP T1404I, T1552N, L1622I, T1626N, V1777M, and E1813K (Figure 3). In contrast, wild-type neonatal cardiomyocytes expressing ankyrin-B–GFP E1425G, V1516D, and R1788W displayed abnormal localization/reduced expression of InsP3 receptor, Na/Ca exchanger 1, and Na/K ATPase. InsP3 receptor in neonatal cardiomyocytes expressing ankyrin-B–GFP E1425G, V1516D, and R1788W was localized primarily to the perinuclear region, whereas Na/Ca exchanger 1 and Na/K ATPase localization was reduced throughout the myocyte with very little perinuclear staining (Figure 3). Finally, we observed significant reductions in InsP3 receptor, Na/Ca exchanger 1, and Na/K ATPase expression in wild-type neonatal cardiomyocytes transfected with ankyrin-B E1425G, V1516D, and R1788W mutants compared with control cells or neonatal cardiomyocytes transfected with ankyrin-B mutants T1404I, T1552N, V1777M, or E1813K (Figure 5). We observed a small but significant decrease in InsP3 receptor, Na/K ATPase, and Na/Ca exchanger 1 expression in wild-type neonatal cardiomyocytes transfected with ankyrin-B mutants L1622I and T1626N (Figure 5; P<0.05; n=3). These data demonstrate a spectrum of in vitro functional defects conferred by ankyrin-B variants and indicate that some, including E1425G (identified in the original LQT4 kindred), V1516D, and R1788W, display severe ankyrin-B loss-of-function activity (Table 3).
Screening of 3 large populations of patients with congenital arrhythmia, drug-induced arrhythmia, and anonymized “control” patients identified 8 different ANK2 variants. ANK2 variants were present in all ethnic groups tested, and 3 were detected in multiple probands. Evaluation of the activities of each variant in ankyrin-B+/− neonatal cardiomyocytes revealed that 4 of 7 new ANK2 variants display abnormal ankyrin-B activity (Figures 3 through 5⇑⇑).
A key unanswered question in the cardiac ion channel field is why all subjects with disease-associated DNA variants do not display markedly abnormal phenotypes like type 4 LQTS or the broader ankyrin-B syndrome.16,17 Thus, an important new finding here is the identification for the first time of 3 distinct functional classes of ANK2 loss-of-function variants (Figure 6 and Table 3). The first class of variants (4 variants) had negligible affect on wild-type neonatal cardiomyocyte contraction rates or channel/transporter localization/expression (Figure 6B, red boxes). The second class (2 variants) behaved as simple loss-of-activity variants while producing minor depressions in wild-type neonatal cardiomyocyte ion channel and transporter localization and expression and small depressions in neonatal cardiomyocyte spontaneous contraction rates (Figure 6B, white boxes). Finally, the third class of variants displayed severe ankyrin-B loss-of-function activity, both decreasing contraction rates and normal channel/transporter expression and localization (Figure 6B, yellow boxes). Together, these findings demonstrate that ANK2 variants produce a spectrum of in vitro defects. Additional cellular studies are necessary to determine the molecular mechanism of each human ANK2 variant. We propose that this striking functional variability in vitro likely contributes to the variability in clinical phenotypes among patients with ANK2 loss-of-function variants. In the patients identified here, other LQTS mutations were absent, suggesting that the identified variants determine the human phenotypes observed. Indeed, the 3 variants with strong loss-of-function characteristics identified here (E1425G, V1516D, R1788W) were associated with generally more severe arrhythmia. Thus, although the numbers are small, we predict that such variants represent risk factors for development of more severe arrhythmias in the face of exogenous stressors such as drug exposure, adrenergic stimulation, or even acute myocardial ischemia.
Our new findings demonstrate the importance of combining targeted mutational analysis with careful functional examination of the physiological relevance of identified gene variants. For example, our screen identified a number of variants (G1406C, R1450W, L1503V) with absolutely no functional effect on ankyrin-B activity. Moreover, new and previously identified variants with loss-of-function phenotypes can now be more carefully scrutinized for degree of dysfunction. On the basis of our new functional data, we hypothesize that variants with less severe in vitro phenotypes (eg, L1622I) are more likely to be present in unaffected populations compared with variants that have severe effects on the myocyte (E1425G). In support of these findings, a recent report found L1622I to be present in a significant percentage of a control population.18 Additionally, Sherman et al18 identified a number of variants in both control and proband populations (E1543K, E1578K, S1721T, T1726N, S1791P) that need to be functionally characterized in myocytes before speculation can be made regarding their clinical significance.
The roles of ankyrins for cell physiology continue to be elucidated, but it is clear that key functions include the cellular organization of a range of other proteins.19 Thus, our new data reinforce the role of protein interactions involving ankyrin-B in regulating normal cardiac electrogenesis and raise new questions regarding ankyrin-B regulation in vivo. To date, all ankyrin-B variants associated with abnormal cardiac function in humans are localized in or near the ankyrin-B regulatory domain. Although ankyrin-B ANK repeats are responsible for direct interactions with membrane-bound ion channels and transporters, multiple independent studies strongly support a critical role for the C-terminal region of ankyrin-B for normal activity in vivo by regulating both intermolecular and intramolecular interactions.12,14,19–22
In summary, the finding that ANK2 variants with altered function are common across human populations, including those with arrhythmias, clearly identifies this gene and those with related functions as candidate modulators of the electrophysiological response of the cardiac myocyte to exogenous stressors such as exercise and drug exposure.23
Predicting clinical phenotype on the basis of genotype is difficult for most autosomal-dominant diseases. Genotype/phenotype predictions are further confounded in the case of ankyrin-B syndrome because congenital LQTS and associated arrhythmias display low penetrance17,24 and because ANK2 variants associated with severe arrhythmia and sudden cardiac death are rare.8 Moreover, in view of the latency of onset of phenotypes observed in probands with ANK2 variants, it also is possible that a number of asymptomatic/control ANK2 variant carriers will display cardiac symptoms in the future. Although these issues are not unique to ANK2-associated arrhythmia, we acknowledge that additional probands and family members harboring ANK2 variants with cardiac phenotypes are critical for further establishing ANK2 genotype-phenotype correlations. As with other disease phenotypes, genetic modifiers and environmental factors likely play a key role in the ultimate cardiac phenotype of ANK2 carriers.23 Our new data provide the first insight into the severity of cellular phenotype for each identified ANK2 loss-of-function variant. Although determining the specific molecular mechanism underlying each individual variant is clearly beyond the scope of a single study, our new results suggesting that specific ANK2 variants lead to severe cardiomyocyte phenotypes clearly prioritize which ANK2 variants should be evaluated quickly and more critically in future experiments. Finally, although our present experiments have focused on the role of ankyrin-B function in ventricular cardiomyocytes, cardiac phenotypes associated with the ankyrin-B syndrome extend beyond the ventricular cardiomyocyte (sinoatrial node, atria). The role of ankyrin-B for targeting specific ion channels and transporters to T-tubule and sarcoplasmic reticulum membranes in ventricular cardiomyocytes is well established.19 In contrast, cellular roles for ankyrin-B polypeptide function in other excitable cardiac cell types are currently unknown but likely based on phenotypes observed in mice with reduced ankyrin-B expression and human probands harboring ANK2 variants.7–9,18 Therefore, correlations between the ventricular cardiomyocyte phenotypes observed in this study and nonventricular cardiac phenotypes in human ANK2 probands are less clear. Undoubtedly, uncovering the roles for ankyrin polypeptides in other excitable cardiac cells presents an exciting future focus for cardiac ankyrin research.
Sources of Funding
Dr Mohler is supported by R01HL084583 and R01HL083422. This work also was supported in part by HL65962, the Pharmacogenomics of Arrhythmia Therapy U01 Site of the Pharmacogenetics Research Network (Dr Roden), the Direction de la Recherche Clinique des Hôpitaux de Paris (PHRC AOR04070), and the GIS-Institut des Maladies Rares (P040111; Dr Guicheney and Dr Denjoy). Dr Escande and Dr Roden are part of the Transatlantic Network of Excellence on Sudden Cardiac Death supported by the Leducq Foundation.
Nerbonne JM, Kass RS. Molecular physiology of cardiac repolarization. Physiol Rev. 2005; 85: 1205–1253.
Bennett V, Baines AJ. Spectrin and ankyrin-based pathways: metazoan inventions for integrating cells into tissues. Physiol Rev. 2001; 81: 1353–1392.
Mohler PJ, Gramolini AO, Bennett V. Ankyrins. J Cell Sci. 2002; 115: 1565–1566.
Mohler PJ, Splawski I, Napolitano C, Bottelli G, Sharpe L, Timothy K, Priori SG, Keating MT, Bennett V. A cardiac arrhythmia syndrome caused by loss of ankyrin-B function. Proc Natl Acad Sci U S A. 2004; 101: 9137–9142.
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.
Yang P, Kanki H, Drolet B, Yang T, Wei J, Viswanathan PC, Hohnloser SH, Shimizu W, Schwartz PJ, Stanton M, Murray KT, Norris K, George AL Jr, Roden DM. Allelic variants in long-QT disease genes in patients with drug-associated torsades de pointes. Circulation. 2002; 105: 1943–1948.
Mohler PJ, Gramolini AO, Bennett V. The Ankyrin-B C-terminal domain determines activity of ankyrin-B/G chimeras in rescue of abnormal inositol 1,4,5-trisphosphate and ryanodine receptor distribution in ankyrin-B (-/-) neonatal cardiomyocytes. J Biol Chem. 2002; 277: 10599–10607.
Mohler PJ, Davis JQ, Davis LH, Hoffman JA, Michaely P, Bennett V. Inositol 1,4,5-trisphosphate receptor localization and stability in neonatal cardiomyocytes requires interaction with ankyrin-B. J Biol Chem. 2004; 279: 12980–12987.
Mohler PJ, Hoffman JA, Davis JQ, Abdi KM, Kim CR, Jones SK, Davis LH, Roberts KF, Bennett V. Isoform specificity among ankyrins: an amphipathic alpha-helix in the divergent regulatory domain of ankyrin-B interacts with the molecular co-chaperone Hdj1/Hsp40. J Biol Chem. 2004; 279: 25798–25804.
Mohler PJ, Yoon W, Bennett V. Ankyrin-B targets beta2-spectrin to an intracellular compartment in neonatal cardiomyocytes. J Biol Chem. 2004; 279: 40185–40193.
Moss AJ, Zareba W, Kaufman ES, Gartman E, Peterson DR, Benhorin J, Towbin JA, Keating MT, Priori SG, Schwartz PJ, Vincent GM, Robinson JL, Andrews ML, Feng C, Hall WJ, Medina A, Zhang L, Wang Z. Increased risk of arrhythmic events in long-QT syndrome with mutations in the pore region of the human ether-a-go-go-related gene potassium channel. Circulation. 2002; 105: 794–799.
Priori SG, Napolitano C, Schwartz PJ. Low penetrance in the long-QT syndrome: clinical impact. Circulation. 1999; 99: 529–533.
Mohler PJ, Davis JQ, Bennett V. Ankyrin-B coordinates the Na/K ATPase, Na/Ca exchanger, and InsP(3) receptor in a cardiac T-tubule/SR microdomain. PLoS Biol. 2005; 3: 2158–2167.
Abdi KM, Mohler PJ, Davis JQ, Bennett V. Isoform specificity of ankyrin-B: a site in the divergent C-terminal domain is required for intramolecular association. J Biol Chem. 2006; 281: 5741–5749.
Hall TG, Bennett V. Regulatory domains of erythrocyte ankyrin. J Biol Chem. 1987; 262: 10537–10545.
Davis LH, Davis JQ, Bennett V. Ankyrin regulation: an alternatively spliced segment of the regulatory domain functions as an intramolecular modulator. J Biol Chem. 1992; 267: 18966–18972.
Mutations in more than a dozen genes have now been implicated as causes of primary congenital arrhythmia syndromes like the long-QT syndrome, Brugada syndrome, or catecholaminergic ventricular tachycardia. Most of these genes encode ion channels or their function-modifying β-subunits. In 2003, linkage analysis in a very large kindred established that a mutation in a non–ion channel gene, ANK2, causes the rare type 4 long-QT syndrome, whereas follow-up studies suggested that the phenotype can be more complicated, including sinus bradycardia and atrial fibrillation. Here, screening ANK2 in normal populations and patients with diverse arrhythmias, including sudden death and drug-associated long-QT syndrome, yielded a surprisingly high number of variants. The key question raised by such an analysis is whether and how identified variants affect arrhythmia susceptibility. In this case, studies in myocytes from wild-type and ANK2 heterozygous mice established that the in vitro functional defect—abnormal cardiomyocyte contraction rates likely caused by misprocessing of key calcium control proteins—conferred by specific human variants varies from none to severe suppression of normal ankyrin-related physiology. This finding reinforces the notion that variants in many genes contributing to normal electrogenesis may modify arrhythmia susceptibility and thus further blurs the distinction between congenital and acquired arrhythmias. With the advent of increasingly robust high-throughput genotyping, identification of gene variants will become ever more common, so studies such as this that approach the problem of translating those findings to the bedside will become an increasingly important component of contemporary genomics.
Guest Editor for this article was Douglas P. Zipes, MD.