Author + information
- Received February 26, 2014
- Revision received May 26, 2014
- Accepted June 2, 2014
- Published online August 26, 2014.
- Patrick A. Schweizer, MD∗,†∗ (, )
- Julian Schröter, BSc∗,‡,
- Sebastian Greiner, MD∗,
- Jan Haas, PhD∗,†,
- Pessah Yampolsky, PhD∗,
- Derliz Mereles, MD∗,
- Sebastian J. Buss, MD∗,
- Claudia Seyler, PhD∗,
- Claus Bruehl, PhD‡,
- Andreas Draguhn, MD‡,
- Michael Koenen, PhD∗,‡,§,
- Benjamin Meder, MD∗,†,
- Hugo A. Katus, MD∗,† and
- Dierk Thomas, MD∗,†∗∗ ()
- ∗Department of Cardiology, University of Heidelberg, Heidelberg, Germany
- †DZHK (German Centre for Cardiovascular Research), partner site Heidelberg/Mannheim, University of Heidelberg, Heidelberg, Germany
- ‡Institute for Physiology and Pathophysiology, University of Heidelberg, Heidelberg, Germany
- §Department of Molecular Neurobiology, Max-Planck-Institute for Medical Research, Heidelberg, Germany
Background Inherited arrhythmias were originally considered isolated electrical defects. There is growing evidence that ion channel dysfunction also contributes to myocardial disorders, but genetic overlap has not been reported for sinus node dysfunction (SND) and noncompaction cardiomyopathy (NCCM).
Objectives The study sought to investigate a familial electromechanical disorder characterized by SND and NCCM, and to identify the underlying genetic basis.
Methods The index family and a cohort of unrelated probands with sinus bradycardia were examined by electrocardiography, Holter recording, exercise stress test, echocardiography, and/or cardiac magnetic resonance imaging. Targeted next-generation and direct sequencing were used for candidate gene analysis and mutation scanning. Ion channels were expressed in HEK293 cells and studied using patch-clamp recordings.
Results SND and biventricular NCCM were diagnosed in multiple members of a German family. Segregation analysis suggested autosomal-dominant inheritance of the combined phenotype. When looking for potentially disease-causing gene variants with cosegregation, a novel hyperpolarization-activated cyclic nucleotide channel 4 (HCN4)-G482R mutation and a common cysteine and glycine-rich protein 3 (CSRP3)-W4R variant were identified. HCN4-G482R is located in the highly conserved channel pore domain. Mutant subunits were nonfunctional and exerted dominant-negative effects on wild-type current. CSRP3-W4R has previously been linked to dilated and hypertrophic cardiomyopathy, but was also found in healthy subjects. Moreover, different truncation (695X) and missense (P883R) HCN4 mutations segregated with a similar combined phenotype in an additional, unrelated family and a single unrelated proband respectively, which both lacked CSRP3-W4R.
Conclusions The symptom complex of SND and NCCM is associated with heritable HCN4 defects. The NCCM phenotype may be aggravated by a common CSRP3 variant in one of the families.
Noncompaction cardiomyopathy (NCCM) is characterized by excessive ventricular hypertrabeculation and complicated by heart failure, arrhythmia, and thromboembolic events (1). Mutations in genes encoding sarcomeric, cytoskeletal, Z-disk, chaperone, Notch-pathway, or nuclear membrane proteins have been related to the disorder (2–5). However, a comprehensive analysis of the genetic basis of NCCM is not available to date, and genotype-phenotype correlations are poorly understood (2).
Sinus node dysfunction (SND) is associated with syncope or bradyarrhythmic death and frequently requires cardiac pacemaker implantation. The hyperpolarization-activated cyclic nucleotide channel 4 (HCN4) is a major constituent of the pacemaker current (If) in the sinoatrial node (SAN) and contributes significantly to spontaneous diastolic depolarization (6–8). Accordingly, mutations in the HCN4 gene were associated with familial forms of SND (9–12). Abnormalities of SAN function are particularly common in heart failure, suggesting a genetic and mechanistic link between electrical and structural dysfunction (13). Indeed, the depolarizing cardiac sodium channel gene SCN5A and non–ion channel genes MYH6 and CASQ2 have recently been implicated in the pathogenesis of both SND and structural heart disease (14–16). However, genetic overlap between SND and NCCM has not been previously described.
In this paper, we report on a symptom complex comprising SND and NCCM observed in 2 unrelated families and in an additional unrelated proband. In a candidate gene approach, we identified a novel HCN4-G482R loss-of-function mutation that segregated with all affected members in the index family. In addition, a common cysteine and glycine-rich protein 3 (CSRP3)-W4R variant that has previously been found in dilated and hypertrophic cardiomyopathy patients and in healthy subjects (17–19) cosegregated with the clinical syndrome in the family. Furthermore, multiple members of a previously reported family (11) and an unrelated proband exhibited similar combined phenotypes and carried truncation (695X) (11) and missense (P883R) HCN4 mutations, respectively, but lacked CSRP3-W4R.
Our results confirm a primary role of HCN4 channels in cardiac pacemaker function and in SND. Furthermore, the data give rise to the notion that HCN4 channels are implicated in the formation of ventricular structure and may contribute to the development of NCCM when dysfunctional.
A detailed description of the methods is provided in the Online Appendix.
Patients and clinical investigations
Patients were evaluated by clinical examination, 12-lead electrocardiogram (ECG), 24-h Holter recording, exercise test, echocardiography, and cardiac magnetic resonance imaging (CMR). Holter recordings were pre-analyzed using H-Scribe 4.0 software (Mortara, Essen, Germany) and validated by an electrophysiologist. All patients provided written informed consent prior to clinical and genetic investigations. The Ethics Committee of Heidelberg University (Germany) approved the research protocol, and the investigation conforms to the principles outlined in the Declaration of Helsinki.
Candidate gene analysis and validation in the index family
Targeted next-generation sequencing (NGS) and direct sequencing were employed in the index family to analyze genes commonly involved in cardiomyopathies and heart rhythm disorders (Online Table 1). NGS was performed on an Illumina HiSeq2000 (Illumina, San Diego, California) using the paired end 2x100bp method (Online Appendix). Variations present in the publicly available variant database “dbSNP137common” and flagged as validated by frequency, indicating an allele frequency of ≥1% in populations, were considered to have a benign effect and therefore were excluded from further analysis. Other genetic variants were retained and tested for segregation in the family by Sanger sequencing.
Comprehensive mutation scans in a sinus bradycardia cohort
We carried out comprehensive mutation scans of HCN4 using direct exon sequencing in a cohort of 86 unrelated patients diagnosed with sinus bradycardia. Study subjects were included in the cohort when resting heart rates and minimum heart rates were <60 and <40 beats/min, respectively. Probands with prolonged QT intervals or ischemic heart disease were excluded. Sequencing of the exon harboring the HCN4 mutation was then performed in relatives of each proband to determine their mutation carrier status. Probands and their families diagnosed with a combined phenotype of sinus bradycardia and myocardial noncompaction also were tested for carrying CSRP3-W4R by direct sequencing.
Site-directed mutagenesis (QuikChange II Site-Directed Mutagenesis Kit, Stratagene, La Jolla, California) was performed to introduce the G482R mutation into the human HCN4 gene (GenBank accession number NM_005477). Automated DNA sequence analysis (GATC Biotech, Konstanz, Germany) verified the mutation.
Human embryonic kidney cells (HEK293) were grown on glass cover slips (CS) (3 × 104 cells per CS) in Dulbecco's Modified Eagles's Medium supplemented with 2 mM glutamine, 10% fetal calf serum, 100 U/ml penicillin-G sodium, and 100 mg/ml streptomycin sulfate in 5% CO2 at 37°C. Cells were transfected using the calcium-phosphate method with 0.6 μg plasmid DNA per CS encoding mutant or wild-type HCN4 channel subunits. Successful transfection was visualized by coexpressed CD8 antigen (0.1 μg DNA/CS) identified by anti-CD8 antibody-coated dynabeads (Life Technologies, Carlsbad, California). In coexpression experiments, equal amounts (0.3 μg) of mutant and wild-type DNA were used (11).
Membrane currents were recorded 1 to 2 days after transfection under voltage-clamp conditions (whole-cell configuration) at room temperature (21°C to 23°C) as published (11). Functional properties of HCN4 channels were investigated employing different voltage protocols as reported (11,20).
Statistics and data analysis
Electrophysiological data were recorded and analyzed offline using Signal software (Version 4.05; CED, Cambridge, England). GraphPad Instat (Version 3.06; La Jolla, California) was used for statistical analyses. All results are provided as mean ± SEM. Data were tested for normality (Kolmogorov-Smirnov test), and a 2-tailed unpaired Student t test, followed by a Welch correction to account for unequal variances was used for statistical comparisons. A Mann-Whitney test was performed for comparison of current densities (molecular electrophysiology) that did not follow Gaussian distributions. A p value <0.05 was considered statistically significant.
Clinical characteristics of index family
The 23-year-old index patient of family A (IV.1) (Fig. 1A) presented with syncope resulting in a car accident. He complained dizziness, fatigue, palpitations, and was aware of bradycardia since childhood. Resting 12-lead electrocardiogram showed sinus bradycardia, first-degree atrioventricular block, and T-wave inversions in the inferior leads (Fig. 2A). The QTc interval was normal (Table 1). Holter recording demonstrated minimum and average heart rates of 21 and 34 beats/min, respectively (Figs. 2B and 2C, Table 1). Exercise testing revealed impaired chronotropic capacity, intermittent ectopic atrial rhythms, and ventricular bigeminy (Figs. 2D and 2E). In addition to the electrical phenotype, transthoracic echocardiography exhibited biventricular hypertrabeculation that met diagnostic criteria of left ventricular noncompaction, according to Jenni or Chin (21), and mitral valve prolapse (MVP) (Figs. 3A and 3D, Online Video 1). Pronounced intracavitary trabeculations and perfused intertrabecular recesses were apparent in right and left ventricular walls (Figs. 3A and 3D, Online Videos 1 and 2), and basal left ventricular longitudinal strain was attenuated (Online Table 2). CMR confirmed that noncompacted segments mainly involved apical, lateral, and inferior walls of the left ventricle (Figs. 4A and 4B). Left ventricular ejection fraction (LVEF) was preserved despite significant noncompaction (Table 1). The patient received a dual chamber pacemaker. During 6-month follow-up he did not report any recurrence of symptoms. Pacemaker interrogation revealed that atrial rates were below 60 beats/min during 96.8% of the follow up period, and sinus node recovery time was prolonged (1,900 ms; corrected sinus node recovery time, 650 ms) (Figs. 2F and 2G).
Family analysis (Fig. 1A) identified the patient’s sister (IV.2; 16 years of age) (Figs. 2H and 2I, Figs. 3B and 3E, Online Videos 3 and 4, Fig. 4C, Table 1, Online Table 2) and mother (III.2; 45 years of age) (Figs. 3C and 3F, Online Videos 5 and 6, Table 1, Online Table 2), as affected by the symptom complex comprising SND, NCCM, and MVP. Of note, no extracardiac anomalies were apparent, and none of the patients actively practiced or had a history of endurance sports that might explain a vagotone-mediated sinus bradycardia. Advanced functional echocardiographic analyses revealed impaired apical rotation (22) in all affected family members (Online Fig. 1). Patient III.2 showed impaired LVEF and symptomatic heart failure (New York Heart Association functional class II to III), while patient IV.2 exhibited preserved LVEF similar to the index patient (Table 1). No overlap with other cardiomyopathies was observed. Clinically affected patients did not receive any medication at the time of evaluation. The father (III.3) of the index patient exhibited impaired LVEF due to myocardial infarction 5 years prior and was treated according to current guidelines. Clinical evaluation showed normal sinus node function and no signs of myocardial noncompaction (Table 1, Online Table 2, Fig. 3, Online Video 7). Furthermore, the grandmother (II.2), granduncle (II.3), and great-grandfather (I.1) on the maternal side of the index patient had a history of sinus bradycardia and congestive heart failure, and they had received pacemakers implanted between 40 to 50 years of age. Patients I.1 and II.2 died from stroke at 73 and 54 years of age, respectively, while patient II.3 experienced sudden death at 50 years of age. As these probands had passed away at the time of family assessment, they were not available for further clinical or genetic analysis.
Genetic analysis in the index family
Candidate analysis of genes commonly involved in cardiomyopathies and cardiac arrhythmias (Online Table 1) was carried out in the index family using targeted next-generation and direct Sanger sequencing. We found 593 variants in the index patient (Fig. 1A). No truncating variant was identified. Filtering out noncoding, synonymous, and benign variants present in the dbSNP137 common database yielded 9 nonsynonymous variants that were validated and tested in the family for segregation by Sanger sequencing. Disease relevant segregation compatible with an autosomal recessive trait was not observed (e.g., no variant was homozygously shared by the affected individuals). However, pedigree analysis (Fig. 1A) suggested autosomal dominant inheritance. Six heterozygous variants were detected in the index patient and shared by other family members (Online Table 3). Of those, 2 variants (HCN4-G482R and CSRP3-W4R) were detected in each family member showing the combined phenotype (III.2, VI.1, and VI.2) but not in the nonaffected father (III.3) (Fig. 1A). The novel HCN4 variant was heterozygously carried by all affected family members and absent from 566 unrelated controls. It results in replacement of a glycine residue with arginine and is located in the highly conserved Glycine-Tyrosine-Glycine motif of the channel pore (G482R) (Figs. 5A to 5D). Furthermore, patient III.2 was homozygous for the common CSRP3-W4R variant (19), while her son (VI.1) and daughter (VI.2) were heterozygous.
Prevalence of myocardial noncompaction in a sinus bradycardia cohort
We investigated a cohort of 86 unrelated probands with sinus bradycardia by comprehensive echocardiography and/or CMR and discovered 3 individuals that exhibited myocardial noncompaction. Interestingly, we found mutations in the exon sequence of HCN4 in all 3 probands affected by the combined phenotype within this cohort. In addition to the index patient of family A, who was heterozygous for HCN4-G482R and CSRP4-W4R (Fig. 1A), a patient and multiple members of his family carried the previously reported HCN4-695X mutation (11) (Fig. 1B, Online Fig. 2). Furthermore, in another unrelated patient, we detected a heterozygous HCN4-P883R mutation (Fig. 1C, Online Fig. 3) that was absent from 566 unrelated controls. However, the CSRP3-W4R variant was not present in any of these 2 latter probands and their families, respectively.
Patients carrying HCN4-695X and HCN4-P883R exhibit combined phenotypes
We observed biventricular hypertrabeculation in multiple members of the previously reported family affected by the HCN4-695X mutation (11) (Fig. 1B, Online Fig. 2). Although LVEF was preserved and no overt structural abnormalities were initially noted (11), all family members carrying HCN4-695X that were examined by multi-modal imaging exhibited biventricular hypertrabeculation (Online Fig. 2) and MVP. The patient that heterozygously carried the HCN4-P883R mutation (Fig. 1C, Online Fig. 3) presented with sinus bradycardia (35 to 40 beats/min) and paroxysmal atrial fibrillation (bradycardia-tachycardia syndrome) and required pacemaker implantation. Echocardiography/CMR revealed myocardial noncompaction with preserved LVEF resembling the combined phenotype of the aforementioned HCN4 mutation carriers.
Cellular phenotype of HCN4-G482R channel subunits
Functional effects associated with the HCN4-G482R mutation were studied in HEK293 cells expressing HCN4 channels using the whole-cell patch clamp technique (Fig. 6A). Mutant HCN4-G482R subunits did not produce relevant hyperpolarization-activated inward currents, reflecting loss-of-function (Fig. 6A). To mimic the heteromeric configuration present in affected patients, cells were cotransfected with equal amounts of plasmid DNA encoding for wild-type and mutant HCN4 subunits. I-V relationships showed lower current densities of heteromeric compared to wild-type channels (Fig. 6B). Quantification during single voltage pulses to –120 mV (4 s) yielded wild-type HCN4-mediated current densities of –110.8 ± 15.8 pA/pF (n = 10) compared to –38.4 ± 8.8 pA/pF (n = 32) in the heteromeric configuration (p < 0.01) (Fig. 6C, Online Table 4). Activation and deactivation properties and reversal potential of HCN4 wild-type currents were not significantly affected by coexpression of G482R (Figs. 6D to 6F, Online Table 4). To investigate whether altered membrane trafficking could underlie current reduction, we analyzed HEK cells that expressed wild-type and mutant channels in heteromeric and homomeric configuration, using mutant constructs FLAG-tagged at the N-terminal region. Confocal microscopy showed similar surface expression of HCN4 wild-type and mutant subunits indicating equivalent trafficking (Online Fig. 4). Overlapping immunoreactivity suggested equal distribution and colocalization of wild-type and mutant FLAG-HCN4-G482R subunits.
We describe a previously unrecognized electromechanical overlap syndrome of SND and NCCM with autosomal-dominant inheritance in 2 German families and a single proband. Our approach highlights the use of NGS to uncover the underlying genetic basis of combined cardiovascular disorders. A novel HCN4-G482R mutation and a common CSRP3-W4R variant (19) were identified and found to cosegregate in the index family. Moreover, the previously reported HCN4-695X mutation cosegregated in the second family (11) and a HCN4-P883R mutation was identified in a single proband—both were clinically characterized by the combined phenotype. While HCN4 dysfunction represents an established mechanism of SND (9–12), we for the first time have linked mutations in HCN4 to a structural cardiac defect.
Molecular electrophysiology underlying HCN4-G482R associated SND
The replacement of glycine by arginine (G482R) within the highly conserved GYG motif of the channel pore (Figs. 5C and 5D) suggests a major impact on ion permeation. Accordingly, patch-clamp recordings revealed that homozygous HCN4-G482R channels were nonfunctional (Figs. 6A and 6C). Heteromeric channels composed of mutant and wild-type HCN4 subunits exhibited ∼65% current reduction compared to wild-type channels (Fig. 6C), indicating a dominant-negative effect as primary mechanism of If current reduction in heterozygous patients. At the molecular level, current decrease may be caused by altered functional properties of the channel and/or by impaired cell surface expression. HCN4-G482R subunits did not affect surface expression of the channels (Online Fig. 4). In addition, there were no changes in activation or deactivation parameters (Figs. 6D and 6E), confirming defective ion permeation as underlying mechanism.
Carriers of HCN4-G482R exhibited sinus bradycardia in agreement with previous reports on HCN4 dysfunction (9,11). While slow heart rates associated with HCN4 mutations reported were mostly asymptomatic and benign, HCN4-G482R carriers were severely affected and showed pronounced rate decrease leading to syncope and requiring pacemaker implantation in the index patient. The particularly severe bradycardia phenotype may be explained by dominant-negative reduction of If current induced by mutant HCN4-G482R subunits in the SAN.
The mechanistic significance of HCN4 dysfunction in NCCM
In addition to its established function in cardiac pacemaking, HCN4 was recently identified as a primary marker for cardiac progenitors of the first heart field, significantly involved in the early embryonic heart development, forming main parts of the cardiac muscle, and the conduction system (23,24). During later development, HCN4 is down-regulated in the working myocardium, and abundant expression is restricted to the SAN and the conduction system (7). Based on these findings, we hypothesize that functional HCN4 loss interferes with molecular mechanisms required during cardiac development, resulting in NCCM (Central Illustration). Myocardial compaction occurs at 5 to 8 weeks of gestation, and impairment or arrest of this process gives rise to hypertrabeculation (25). In this regard, recent data suggested that Notch-pathway disturbance causes noncompaction with congenital heart disease while sarcomere, cytoskeletal, and Z-disk mutations provoke a more myocardial disease-only phenotype (26). In support of a congenital origin both juvenile HCN4-G482R mutant-carriers (patients IV.1 and IV.2) exhibited marked hypertrabeculation and additional MVP early in life. Hence, a potential implication of HCN4 in signaling pathways involved in ventricular wall maturation and compaction (e.g., Notch, Neuregulin, Ephrin, or Bone morphogenic protein) (26) is of particular interest and requires future investigation.
Based on the clinical observations among our study patients, it appears surprising that no structural cardiac abnormalities were reported on HCN4 knockout mice (27). This finding may be explained by death of homozygous animals at early embryonic stages, most likely due to failure of rate initiation (27). Heterozygous animals, however, were reported indistinguishable from wild-type littermates, while detailed post-natal structural investigations have not been performed (27). Furthermore, an absence of noncompaction phenotypes in mouse models is a previously described phenomenon (22). Luedde et al. (22) reported an NCCM-related human troponin T mutation that lacked an obvious noncompaction phenotype in their transgenic mouse model. It was speculated whether differences in species might render mice more resistant to NCCM phenotype (22). Consistent with this notion, data from other genetic cardiomyopathy models in rodents observed specific phenotypes only in certain strains (28), indicating the importance of a human-specific or at least distinctive predisposing genetic background.
Potential role of CSRP3-W4R in the index family
Hetero- (patients VI.1 and VI.2) or homozygous (patient III.2) cosegregation of CSRP3-W4R with the clinical phenotype in the index family suggests that this variant also may contribute to the structural phenotype. However, CSRP3-W4R was reported prevalent in up to 1% of control probands (19,29) and has never been linked to NCCM despite its high allele frequency. Functional characterization of CSRP3-W4R using a knock-in mouse model (29) revealed that it contributes to hypertrophic cardiomyopathy and heart failure in mice. In humans, it was previously detected in dilated cardiomyopathy and hypertrophic cardiomyopathy patients but often failed to clearly segregate with either disease in families (17). Hence, it was suggested that CSRP3-W4R does not represent a disease-causing mutation in humans, but serves as phenotypic modifier in cardiomyopathies (18). It is reasonable to speculate in this context that the disposition to HCM associated with CSRP-W4R could aggravate the NCCM phenotype on the basis of an underlying HCN4-G482R mutation. Supporting this notion, patient III.2, who exhibited the most severe NCCM phenotype, is homozygous for CSRP3-W4R.
Ventricular hypertrabeculation in SND patients
Whether HCN4-G482R alone or together with CSRP3-W4R underlies patients’ hypertrabeculation cannot be clarified within the index family, as all affected members alive carry both mutations (Fig. 1A). To address this issue, we carried out multimodal imaging and performed mutation scans of HCN4 in a cohort of 86 unrelated patients diagnosed for sinus bradycardia. We identified 2 additional unrelated patients that carried HCN4 mutations and exhibited the combined phenotype: The index patient of the previously reported HCN4-695X mutation family (11) and a single proband that carried a heterozygous HCN4-P883R mutation. Of note, CSRP3-W4R was not present in these patients. Based on these findings, we re-evaluated the cardiac morphology of family members, carrying the HCN4-695X mutation and of their unaffected relatives (11) (Fig. 1B, Online Fig. 2). Although LVEF was preserved and no overt structural abnormalities were initially noted (11), all HCN4-695X mutant-carriers examined by comprehensive echocardiography and/or CMR exhibited biventricular hypertrabeculation (Online Fig. 2). These changes were less severe than in patients carrying HCN4-G482R, which might be explained by the fact that 695X compromises cAMP responsiveness of the HCN4 channel but not ion permeation itself and/or by the absence of CSRP3-W4R in carriers of the HCN4-695X mutation. Furthermore, the proband heterozygously carrying HCN4-P883R exhibited myocardial noncompaction and was affected by SND requiring pacemaker implantation (Online Fig. 3). This case provides additional, independent evidence that validates the association of the proposed disease entity with mutations in HCN4. Based on these data, it appears unlikely that CSRP3-W4R present only in the index family serves as primary cause of the NCCM phenotype. Thus, our study strongly supports the notion that HCN4 mutations are critically involved in the development of myocardial noncompaction.
In this study, we did not carry out experiments on the pathogenetic role of HCN4 during heart development and the modifying contribution of the CSRP3 variant. Such studies will be necessary to elucidate the mechanistic implication of HCN4 defects in ventricular malformation and to validate its significance for NCCM.
We report for the first time that HCN4 channel dysfunction in addition to its known association with SND may play a distinct role in the development of structural cardiac abnormalities. The novel HCN4 loss-of-function mutation G482R was identified and segregated with a combined disease phenotype in family members with SND and NCCM. The structural phenotype might be modified by the common CSRP3-W4R variant in the index family. Different HCN4 mutations (695X and P883R) segregated with a similar combined phenotype in an unrelated family and in a single unrelated proband and further support a disease-relevant association with HCN4.
COMPETENCY IN MEDICAL KNOWLEDGE: Mutations in the HCN4 gene are linked to a symptom complex of SND and myocardial noncompaction.
TRANSLATIONAL OUTLOOK: Future studies are required to determine the clinical utility of genetic testing and long-term cardiac rhythm monitoring in patients with myocardial noncompaction.
For expanded Methods and References sections as well as supplemental tables, figures, and videos, please see the online version of this article.
This work was supported in part by grants from the Max-Planck-Society (TANDEM project to Drs. Schweizer and Koenen), the German Cardiac Society (research scholarship to Dr. Yampolsky), the Medical Faculty of the University of Heidelberg (Young Investigator Award to Dr. Yampolsky), the German Cardiac Society and the Hengstberger Foundation (Klaus-Georg and Sigrid Hengstberger Scholarship to Dr. Thomas), and the Joachim Siebeneicher Foundation (to Dr. Thomas). All authors have reported that they have no relationships relevant to the content of this paper to disclose.
- Abbreviations and Acronyms
- cardiac magnetic resonance imaging
- cover slips
- cysteine and glycine-rich protein 3
- hyperpolarization-activated cyclic nucleotide channel
- left ventricular ejection fraction
- mitral valve prolapse
- noncompaction cardiomyopathy
- next-generation sequencing
- sinoatrial node
- sinus node dysfunction
- Received February 26, 2014.
- Revision received May 26, 2014.
- Accepted June 2, 2014.
- American College of Cardiology Foundation
- Oechslin E.N.,
- Attenhofer Jost C.H.,
- Rojas J.R.,
- et al.
- Oechslin E.,
- Jenni R.
- Vatta M.,
- Mohapatra B.,
- Jimenez S.,
- et al.
- Baruscotti M.,
- Bucchi A.,
- Viscomi C.,
- et al.
- Schweizer P.A.,
- Duhme N.,
- Thomas D.,
- et al.
- Duhme N.,
- Schweizer P.A.,
- Thomas D.,
- et al.
- Zicha S.,
- Fernández-Velasco M.,
- Lonardo G.,
- et al.
- Butters T.D.,
- Aslanidi O.V.,
- Inada S.,
- et al.
- Glukhov A.V.,
- Kalyanasundaram A.,
- Lou Q.,
- et al.
- Ludwig A.,
- Zong X.,
- Stieber J.,
- et al.
- Luedde M.,
- Ehlermann P.,
- Weichenhan D.,
- et al.
- Liang X.,
- Wang G.,
- Lin L.,
- et al.
- Stieber J.,
- Herrmann S.,
- Feil S.,
- et al.
- Knöll R.,
- Kostin S.,
- Klede S.,
- et al.
- Sizarov A.,
- Devalla H.D.,
- Anderson R.H.,
- Passier R.,
- Christoffels V.M.,
- Moorman A.F.