If you don't remember your password, you can reset it by entering your email address and clicking the Reset Password button. You will then receive an email that contains a secure link for resetting your password
If the address matches a valid account an email will be sent to __email__ with instructions for resetting your password
Division of Cardiology, Department of Internal Medicine, Tongji Hospital, Tongji Medical College, Huazhong University of Science and Technology, Wuhan, ChinaHubei Key Laboratory of Genetics and Molecular Mechanism of Cardiologic Disorders, Huazhong University of Science and Technology, Wuhan, China
Division of Cardiology, Department of Internal Medicine, Tongji Hospital, Tongji Medical College, Huazhong University of Science and Technology, Wuhan, ChinaHubei Key Laboratory of Genetics and Molecular Mechanism of Cardiologic Disorders, Huazhong University of Science and Technology, Wuhan, China
Division of Cardiology, Department of Internal Medicine, Tongji Hospital, Tongji Medical College, Huazhong University of Science and Technology, Wuhan, ChinaHubei Key Laboratory of Genetics and Molecular Mechanism of Cardiologic Disorders, Huazhong University of Science and Technology, Wuhan, China
Division of Cardiology, Department of Internal Medicine, Tongji Hospital, Tongji Medical College, Huazhong University of Science and Technology, Wuhan, ChinaHubei Key Laboratory of Genetics and Molecular Mechanism of Cardiologic Disorders, Huazhong University of Science and Technology, Wuhan, China
Division of Cardiology, Department of Internal Medicine, Tongji Hospital, Tongji Medical College, Huazhong University of Science and Technology, Wuhan, China
Division of Cardiology, Department of Internal Medicine, Tongji Hospital, Tongji Medical College, Huazhong University of Science and Technology, Wuhan, ChinaHubei Key Laboratory of Genetics and Molecular Mechanism of Cardiologic Disorders, Huazhong University of Science and Technology, Wuhan, China
Division of Cardiology, Department of Internal Medicine, Tongji Hospital, Tongji Medical College, Huazhong University of Science and Technology, Wuhan, ChinaHubei Key Laboratory of Genetics and Molecular Mechanism of Cardiologic Disorders, Huazhong University of Science and Technology, Wuhan, China
Division of Cardiology, Department of Internal Medicine, Tongji Hospital, Tongji Medical College, Huazhong University of Science and Technology, Wuhan, China
Division of Cardiology, Department of Internal Medicine, Tongji Hospital, Tongji Medical College, Huazhong University of Science and Technology, Wuhan, China
Division of Cardiology, Department of Internal Medicine, Tongji Hospital, Tongji Medical College, Huazhong University of Science and Technology, Wuhan, China
Division of Cardiology, Department of Internal Medicine, Tongji Hospital, Tongji Medical College, Huazhong University of Science and Technology, Wuhan, ChinaHubei Key Laboratory of Genetics and Molecular Mechanism of Cardiologic Disorders, Huazhong University of Science and Technology, Wuhan, China
Corresponding author: Dr Dao Wen Wang, Division of Cardiology, Department of Internal Medicine, Tongji Hospital, Tongji Medical College, Huazhong University of Science and Technology, Wuhan 430030, PR China. Tel. and fax: +86-27-83663280.
Division of Cardiology, Department of Internal Medicine, Tongji Hospital, Tongji Medical College, Huazhong University of Science and Technology, Wuhan, ChinaHubei Key Laboratory of Genetics and Molecular Mechanism of Cardiologic Disorders, Huazhong University of Science and Technology, Wuhan, China
The genetic basis of a considerable fraction of hypertrophic cardiomyopathy (HCM) cases remains unknown. Whether the gene encoding RNA binding motif protein 20 (RBM20) is implicated in HCM and the correlation of clinical characteristics of RBM20 heterozygotes with HCM remain unresolved. We aimed to investigate the association between RBM20 variants and HCM.
Methods
We compared rare variants in the RBM20 gene by exome sequencing in 793 patients with HCM and 414 healthy controls. Based on a case-control approach, we used optimal sequence kernel association test (SKAT-O) to explore whether RBM20 is associated with HCM. The genetic distribution of RBM20 rare variants was then compared between HCM heterozygotes and dilated cardiomyopathy (DCM) heterozygotes. Clinical features and prognosis of RBM20 heterozygotes were compared with nonheterozygotes.
Results
Gene-based association analysis implicated RBM20 as a susceptibility gene for developing HCM. Patients with RBM20 variants displayed a higher prevalence of sudden cardiac arrest (SCA) (6.7% vs 0.9%, P = 0.001), increased sudden cardiac death (SCD) risk factor counts and impaired left ventricle systolic function. Further survival analysis revealed that RBM20 heterozygotes had higher incidences of resuscitated cardiac arrest, recurrent nonsustained ventricular tachycardia, and malignant arrhythmias. Mendelian randomization suggested that RBM20 expression in the left ventricle was causally associated with HCM and DCM with opposite effects.
Conclusions
This study identified RBM20 as a potential causal gene of HCM. RBM20 variants are associated with increased risk for SCA in HCM.
Résumé
Contexte
Le fondement génétique d’une fraction considérable de cas de cardiomyopathie hypertrophique (CMH) demeure inconnu. On ne sait pas si le gène encodant la protéine à motif de liaison à l’ARN 20 (RBM20) joue un rôle dans la CMH et on ne connaît pas la corrélation entre les caractéristiques cliniques des patients porteurs d’une mutation hétérozygote de RBM20 et la CMH. Notre objectif est d’étudier l’association entre les variants de RBM20 et la CMH.
Méthodologie
Nous avons comparé des variants rares du gène RBM20 en procédant à un séquençage de l’exome chez 793 patients atteints de CMH et 414 témoins en bonne santé. En nous fondant sur une approche cas-témoins, nous avons utilisé un test d’association à noyau de séquence optimale (SKAT-O) pour déterminer si RBM20 est associé à la CMH. La répartition génétique de variants rares de RBM20 a ensuite été comparée entre les patients hétérozygotes atteints de CMH et les patients hétérozygotes atteints de cardiomyopathie dilatée (CMD). Les caractéristiques cliniques et le pronostic des patients porteurs d’une mutation hétérozygote de RBM20 ont été comparés à ceux des patients porteurs d’une mutation non hétérozygote.
Résultats
Selon l’analyse de l’association génétique, RBM20 serait un gène de prédisposition à la CMH. Les patients porteurs de variants de RBM20 ont affiché une prévalence plus élevée d’arrêt cardiaque soudain (ACS) (6,7 % contre 0,9 %; p = 0,001), un facteur de risque de décès cardiaque soudain (DCS) accru et une dysfonction systolique du ventricule gauche. D’autres analyses de la survie ont révélé que les patients porteurs d’une mutation hétérozygote de RBM20 affichaient une incidence plus élevée d’arrêt cardiaque nécessitant une réanimation, de tachycardie ventriculaire non soutenue récurrente et d’arythmies malignes. La randomisation mendélienne semble indiquer un lien causal entre l’expression de RBM20 dans le ventricule gauche et la CMH de même que la CMD, avec des effets opposés.
Conclusions
Cette étude révèle que RBM20 est un gène possiblement responsable de la CMH. Les variants de RBM20 sont associés à un risque accru d’ACS dans la CMH.
Hypertrophic cardiomyopathy (HCM) affects nearly 0.29% of the general adult population, contributing to most of the sudden cardiac deaths (SCDs) of adolescents and young adults.
2014 ESC Guidelines on diagnosis and management of hypertrophic cardiomyopathy: the Task Force for the Diagnosis and Management of Hypertrophic Cardiomyopathy of the European Society of Cardiology (ESC).
However, the genetic basis of the disease in the rest of the patient population remains largely unknown, which is referred to as the missing causal genes.
Previous studies have identified the gene-encoding RNA binding motif protein 20 (RBM20), which regulates alternative pre-messenger RNA splicing, as a causal gene of dilated cardiomyopathy (DCM), whereas its role in HCM is still unknown. Electrocardiograms (ECGs) indicated left ventricular (LV) hypertrophy in some family members from the study by Brauch et al.
Subsequent studies revealed that several gene targets (TNNT2, MYH7, TTN, NEXN, RYR2, and OBSCN) directly regulated by RBM20 were involved in DCM, as well as HCM.
Accordingly, we proposed that variants in RBM20 are implicated in HCM.
Thus, we compared rare variants in RBM20 by exome sequencing in a HCM cohort with healthy controls to explore the association of RBM20 with HCM. Further, clinical features and outcomes between patients with HCM carrying and not carrying RBM20 variants were compared during follow-up.
Materials and Methods
Study population
The cohort comprised 793 unrelated patients with HCM recruited from 2007 to 2019, at Tongji Hospital, Wuhan, China. HCM was diagnosed as a maximal LV wall thickness ≥ 15 mm on echocardiography or cardiac magnetic resonance imaging (MRI), in the absence of abnormal loading conditions or other cardiac or systemic disease capable of producing that magnitude of hypertrophy, such as congenital heart disease, aortic stenosis, hypertension, and phenocopied conditions. More limited hypertrophy (13 to 14 mm) was diagnostic for patients with family histories of HCM.
2020 AHA/ACC guideline for the diagnosis and treatment of patients with hypertrophic cardiomyopathy: a report of the American College of Cardiology/American Heart Association Joint Committee on Clinical Practice Guidelines.
Peripheral blood samples were obtained from all participants when enrolling. All clinical variables were extracted from patient records blinded to patient genotype. For comparison, 414 ethnically matched control subjects without any cardiovascular disease were recruited from hospital and community.
Variants pathogenicity evaluation
DNA was extracted from whole blood and whole-exome sequencing (WES) was performed on Illumina platform (Diagenode, Liège, Belgium). Details were described in Supplemental Appendix S1. American College of Medical Genetics and Genomics (ACMG) recommendations were adopted to decide the pathogenicity for each variant. Specifically, only rare nonsynonymous or truncating variants distributing in HCM associated genes,
with minor allele frequency (MAF) < 0.5% in the East Asian population from public database, and labelled as deleterious in functional prediction using Meta support vector machine (SVM) and Meta logistic regression (LR), were then subjected to ACMG evaluation.
Follow-up with the recruited patients was conducted by December 2019, through face-to-face interviews and phone calls. For those who died during hospitalization, the outcome data were derived from their electronic medical records. The primary endpoint was cardiovascular death, including heart failure-related death and SCD. Other clinical outcomes of note included sudden cardiac arrest (SCA), recurrent nonsustained ventricular tachycardia (NSVT), and malignant arrhythmia. Referring to previously identified risk factors for SCD,
2014 ESC Guidelines on diagnosis and management of hypertrophic cardiomyopathy: the Task Force for the Diagnosis and Management of Hypertrophic Cardiomyopathy of the European Society of Cardiology (ESC).
previous episodes of cardiac arrest, family history of SCD, unexplained syncope, sustained and repetitive NSVT, maximal LV wall thickness ≥ 30 mm, and severe left ventricular outflow tract (LVOT) obstruction (LVOT gradient > 80 mm Hg) were taken into account.
RBM20 carriers with DCM
In an attempt to trace causes of distinct consequences associated with RBM20 variants, we compared the genetic distribution of RBM20 rare variants found in HCM with those identified in patients with DCM (n = 21), which were extracted from our own DCM cohort (n = 1041). This DCM cohort included a total of 1041 patients with DCM enrolled from Tongji Hospital between 2007 and 2018. Detailed information can be found in our previously published article.
Classification of the cardiomyopathies: a position statement from the European Society Of Cardiology Working Group on Myocardial and Pericardial Diseases.
DNA was isolated, and WES was performed, as was done for HCM. The sequencing depth and average coverage were listed in Supplemental Table S1.
RBM20 mRNA expression in patients with HCM and Mendelian randomization
To check RBM20 mRNA expression in patients with HCM, we retrieved the published RNA-seq data from GEO: GSE36961, which contained 106 human heart tissues with HCM and 39 normal samples. After normalization using log2 transformation and variance and abundance filtration, read counts were processed to differential expression analysis using R-package Limma. We further used 2-sample Mendelian randomization analyses to test whether RBM20 mRNA expression had a causal relationship with risk of HCM.
Specifically, we first extracted genome-wide association study (GWAS) summary statistics for HCM (I9_HYPERTROCARDMYOP), obstructive HCM (I9_CARDMYOHYP), and nonischemic cardiomyopathy (I9_NONISCHCARDMYOP) from FinnGen Datafreeze 4 release and nonischemic cardiomyopathy (excluding HCM) from the study by Aragam et al.,
in which HCM cases were defined by discharge diagnoses of I42.1 or I42.2, according to the International Classification of Diseases (ICD)-10 criteria. Genotype-tissue expression (GTEx v.7) LV expression results were used as the exposure
and GWAS summary statistics for cardiomyopathy as the outcome. Significant (nominal P value < 0.05) local (1 megabase up- and downstream of the transcription start site) eQTLs from GTEx v.7 LV samples for RBM20 were considered to be potential instrumental variables. The inverse-variance-weighted (IVW) method, under a multiplicative random-effects model, was used to obtain the primary causal estimates.
Gene-based burden test and statistical analysis
We chose optimal sequence kernel association test (SKAT-O),
which was a linear combination of unidirectional and variance-component tests, to evaluate the association between RBM20 and risk of HCM. Unpaired Student's t-tests were used to compare continuous variables; χ2 or Fisher’s exact tests were used to compare noncontinuous variables expressed as proportions. Survival curves were constructed according to the Kaplan-Meier method, and comparisons were performed using the log-rank test. Cox proportional hazard models were used to assess the effects of multiple clinical features on the risk of outcome events. All reported probabilities were 2-sided and considered significant when < 0.05.
Results
Mutational profile of RBM20 and gene-level analysis
We compared rare variants in the RBM20 gene (Supplemental Table S2) by exome sequencing in 793 patients with HCM and 414 healthy controls. After filtration, 14 rare deleterious variants were identified, including 12 missense variants, 1 splicing site, and 1 truncation variant. Of these variants, 13 were detected in 19 (2.40%) independent HCM cases, whereas only 3 variants were identified in 3 (0.72%) control subjects (Fig. 1). Detailed information of the cases carrying RBM20 deleterious variants is listed in Supplemental Table S3.
Figure 1Distribution of rare deleterious variants in RBM20. Structural and functional domains of RBM20 showing variant sites identified in patients with HCM (up) and healthy controls (down). The node counts represent total number of heterozygotes. HCM, hypertrophic cardiomyopathy; RBM20, RNA binding motif protein 20.
The genetic relatedness between pairs of individuals was checked with proportion identity by descent (IBD) (Supplemental Fig. S1). The homogenous population structure between the cases and control subjects was revealed by the principal component analysis (Supplemental Fig. S2). To avoid the contribution of additional causal variants in other HCM-associated genes to the 19 affected heterozygotes of RBM20 variants, we discarded subjects with pathogenic or likely pathogenic variants in genes associated with HCM (Supplemental Table S4) before the burden test. A total of 68 variants (Supplemental Table S5) in 92 patients were evaluated as pathogenic or likely pathogenic according to ACMG recommendations and discarded. To evaluate the association between RBM20 and HCM, gene-based burden tests was performed and revealed an enrichment of RBM20 heterozygotes in the HCM population (19 of 701 vs 3 of 414, PSKAT-O = 0.0332).
Suspicious pedigree analysis
To further determine the specific association between RBM20 and HCM, we screened all heterozygotes with RBM20 variants for a family history of HCM. As Figure 2A shows, 2 family members of pedigree A were diagnosed as HCM, carrying a same variant (NM_001134363.3:c.2109G > T) in RBM20. In addition, other rare deleterious variants in HCM associated genes were not observed in proband (Ⅱ-5). Further family screening revealed another 3 heterozygotes, whereas none of them fulfilled the diagnostic criteria for HCM. Meanwhile,Ⅰ-2 presented with a mild hypertrophy (11 mm) in interventricular septum.
Figure 2Pedigree of index family with hereditary hypertrophic cardiomyopathy (HCM). Pedigree structures for kindred number 2738 and number 1260 are shown. Square = male; circle = female; solid = HCM; open = unaffected; grey = clinical status unknown; slash through the symbol = deceased. Parallel diagonal lines denote subjects with hypertrophy but not fulfilling diagnostic criteria (A), left ventricular diastolic dysfunction and left atrial enlargement (B), respectively. Arrow points to the proband. Those carrying the variants are identified by the + and those who do not by the – signs.
As for pedigree B, the proband was diagnosed with HCM at the age of 35 and accepted a heart transplant 3 years later for rapid progression of LV dilation and a failing heart. The son of the proband died suddenly at the age of 22 and was diagnosed with HCM before death. WES revealed that the proband carried both RBM20 (NM_001134363.3:c.3545G > A) and TNNT2 (NM_001276345.1:c.305G > A) variants. LV diastolic dysfunction and left atrial enlargement were also observed in all members carrying only the RBM20 variant except forⅠ-2 (clinical status unknown).
Correlation of RBM20 sequence architecture and phenotype
These findings raise our curiosity for regions of the RBM20 transcript enriched for variants associated with HCM and DCM, respectively. So we further compared the frequencies of variants lying within each functional domain and exon between the HCM and DCM population.Supplemental Figure S3 showed a similar distribution between the 2 groups, both enriched in exon 2, exon 9, and exon 13, with a hot spot in p.R1182H.
Simultaneously, the 21 DCM subjects were shown to have thicker interventricular septum (IVS) compared with those not harbouring variants in RBM20 (10.4 mm vs 9.6mm, P = 0.029). Furthermore, of these RBM20 variant heterozygotes, 5 were observed with increases in left ventricular end-diastolic diameter (LVEDD) accompanied by decreases in IVS thickness during follow-up (Supplemental Fig. S4), especially for subjects cp735 and cp65, whose IVS images became relatively normal.
Baseline clinical characteristics of RBM20 cases
To better understand the clinical characteristics of RBM20 heterozygotes, patients with HCM carrying rare variants (MAF < 0.5%) in RBM20 were compared with nonheterozygotes. Detailed features are listed in Table 1. Notably, there was a higher prevalence of SCA and resuscitation (6.7% vs 0.9%, P = 0.001) in patients with RBM20 variants compared with those without RBM20 variants. A severe LVOT obstruction (LVOT gradient [G] > 80 mm Hg) was also more common in RBM20 variant heterozygotes (23.8% vs 9.9%, P = 0.004). These heterozygotes presented with a decrease in LVEF (54.88 vs 58.92, P = 0.047), suggesting impaired LV systolic function. In addition, patients with RBM20 were more likely to have 2 or more identified SCD risk factors (11.1% vs 3.3%, P = 0.008). Taken together, these results suggest a strong association between RBM20 variants and risk of SCD.
Table 1Baseline clinical features of HCM patients carrying variants in RBM20
We next compared these 45 heterozygotes with 27 patients carrying variants recognized as pathogenic or likely pathogenic in MYH7 and MYBPC3. From Supplemental Table S6, we can see that RBM20 heterozygotes exhibited less severity in LV hypertrophy, accompanied by significantly increased LVEDD and apparent reduction in left ventricular ejection fraction (LVEF), compared with MYH7 carriers. Similar results were observed in the comparison with 33 patients with HCM and MYBPC3 mutation (Supplemental Table S7). Overall, adverse cardiac remodelling seemed more common in patients with RBM20 variants.
Genotype-phenotype correlation for clinical outcomes
We further checked the prognosis of these RBM20 heterozygotes. A total of 775 (97.7%) participants accepted final evaluation. Mean follow-up time was 32.78 ± 27.58 months. As survival analysis in Figure 3 shows, RBM20 patients had a higher but not significant likelihood of experiencing SCD (hazard ratio [HR] 3.65, P = 0.074). In addition, it seemed that patients with RBM20 variants had a higher incidence of resuscitated cardiac arrest (HR 7.52, P < 0.001), recurrent NSVT (HR 4.45, P = 0.012) and malignant arrhythmias (HR 2.70, P = 0.019) (including SCD, SCA, recurrent NSVT, and recurrent syncope). In accordance with baseline data, these findings provided evidence supporting our assumption that RBM20 variants may influence prognosis of patients with HCM by contributing to malignant arrhythmias. However, event-free survival did not differ between the 21 RBM20 heterozygotes and nonheterozygotes in patients with DCM (Supplemental Fig. S5).
Figure 3Clinical outcomes in RBM20+ vs RBM20– HCM. (A) Kaplan-Meier curve illustrating survival free of SCD. (B) Survival free of resuscitated cardiac arrest. (C) Survival free of recurrent NSVT. (D) Survival free of malignant arrhythmias, including SCD, resuscitated cardiac arrest, recurrent NSVT, and recurrent episode of syncope. The probability values are calculated with the log-rank test. NVST, nonsustained ventricular tachycardia; RBM20, RNA binding motif protein 20; SCD, sudden cardiac death.
We next checked whether RBM20 differently expressed between HCM cases and controls. As Supplemental Figure S6 shows, RBM20 mRNA expression in the LV of HCM cases was significantly higher than normal controls (fold change 1.77, P = 2.2e-11). In subsequent Mendelian randomization, 49 eQTLs were retained as genetic instruments after linkage disequilibrium pruning (r2 < 0.2, with a window size of 5000 kb), and genetically higher RBM20 mRNA expression was found to be associated with higher HCM susceptibility (Supplemental Table S8). Sensitivity analysis using MR-Egger intercept suggested the absence of directional pleiotropy (P = 0.887). No heterogeneity of effects or outlier was detected by the leave-1-out analysis (Supplemental Fig. S7) or Mendelian Randomization Pleiotropy Residual Sum and Outlier (MR-PRESSO) global test (P = 0.428).
Interestingly, we also found that decreased RBM20 expression was associated with increased risk for nonischemic cardiomyopathy using GWAS results from FinnGen study and the study by Aragam et al., in which nonischemic cardiomyopathy was defined as LV dysfunction and absence of coronary artery disease and HCM.
We conducted an exome sequencing study of the RBM20 gene in 793 sporadic HCM cases and 414 healthy controls. By using gene-based association tests, we found RBM20, which maps to the chromosome 10q25.2 region, to be a gene associated with HCM. We identified a dozen rare, damaging variants present in RBM20, with putative roles in contributing to protein dysfunction. Pedigree cosegregation and hypertrophy observation in patients with DCM provided further support for the conclusion. Further baseline and survival analysis provided evidence that RBM20 plays a pro-arrhythmic role and confers an increased risk for SCA in HCM. Mendelian randomization, using published GWAS statistics, supported that RBM20 expression in LV contributed to risk of HCM and DCM with opposite directions of effect.
Previous studies have mapped and identified causal or associated genes through robust cosegregation and linkage analyses in large HCM families.
Such conventional approaches, albeit convincing and reliable, are restricted to identifying variants that have a large effect size. For variants potentially causing HCM with incomplete or low penetrance, it seems unlikely that they would cosegregate with the phenotype perfectly.
Hence, we performed gene-based association analysis on HCM cases and control samples, which enables more powerful statistics to assess the role of rare variants. Taking advantage of SKAT-O, a gene burden test that has the highest mean power across simulated datasets,
we will be able to evaluate the gene–disease association at genome-wide level. Such a genome-wide statistical study using WES data is underway and will be published separately.
Targeting RBM20 here, we first examined variants in known HCM genes and found only approximately 12% of patients demonstrate variants classified as pathogenic or believed to be disease causing. This seemed much lower than the 60% in early studies
2014 ESC Guidelines on diagnosis and management of hypertrophic cardiomyopathy: the Task Force for the Diagnosis and Management of Hypertrophic Cardiomyopathy of the European Society of Cardiology (ESC).
It was largely ascribed to the evaluation standard for pathogenic variant we adopted—namely, ACMG recommendations—which was much stricter compared with the self-defined standards in early studies. Most novel missense variants found in our study were classified as “variant of unknown significance” and filtered out in the absence of cosegregation and experimental evidence. In addition, given that family history was associated with increased mutation-positive likelihood,
the sporadic and nonfamilial nature of our cohort may also give rise to this low positive rate.
Although most literature has described the nature of RBM20 in DCM, here we focused on the relationship between RBM20 and HCM. Our current findings, although preliminary, indicate that patients carrying RBM20 variants are enriched in HCM cases, and together suggest that the role of RBM20 in HCM may be underestimated. Following cosegregation and linkage of RBM20 variants with phenotype in pedigree A supported the idea that RBM20 variants contribute to HCM pathogenesis. Family members Ⅲ-2 (age 18) and Ⅲ-3 (age 15) who are positive for the variant seemed unaffected because of their young ages, suggest that variants in RBM20 may exhibit incomplete or low penetrance. Although pedigree B was ambiguous, owing to the presence of additional variant in TNNT2, the role of RBM20 variants in HCM pathogenesis cannot be excluded. LV diastolic dysfunction and left atrial enlargement, which may be early phenotypic consequences of HCM independent of LV hypertrophy,
were found in all members carrying only the RBM20 variant, except forⅠ-2 (clinical status unknown). Also, we noticed thatⅠ-1 (age 67)—who was distinct from the proband—and Ⅲ-4, was asymptomatic in the past years, leading us to speculate that the variant in RBM20 may acted in concert with the variant in TNNT2 in modifying phenotype.
Preclinical studies using animal models support key features identified in our cohort of HCM patients with RBM20 variants,
Beyond the enrichment of conventional SCD risk factors in the RBM20 group, we further observed a more prevalent onset of resuscitated cardiac arrest, recurrent NSVT, and malignant arrhythmias. These findings, to some extent, highlight the need for heightened attention to—and concern for—patients with HCM carrying pathogenic or likely pathogenic variants in RBM20.
We noted that patients with HCM and DCM share several identical variants, which remind us of the possibility that RBM20 variant heterozygotes may manifest with LV dilation and hypertrophy at the same time. As expected, thicker IVS was observed in patients with DCM harbouring RBM20 variants. Meanwhile, dynamic observation revealed that a part of these heterozygotes experienced obvious wall thinning following LV dilation, which indicated the possibility of regarding them as DCM simply, and increased the complexity of diagnosis.
Previous studies reported that the ratio of RBM20 to other splicing factor in LV was inversely related with the titin size,
Experimentally increasing the compliance of titin through RNA binding motif-20 (RBM20) inhibition improves diastolic function in a mouse model of heart failure with preserved ejection fraction.
On the other side, RBM20 expression upregulation through all-trans retinoic acid reverted contractility defects and ameliorated DCM phenotypes in vitro.
These findings led us to test the hypothesis that RBM20 expression is causally associated with cardiomyopathy. Our results identified such potential causality between RBM20 expression and cardiomyopathy susceptibility genetically. It can therefore be summarized that a balance exists in LV RBM20 expression, up- or downregulation of which will give rise to opposite phenotypes: namely, HCM and DCM.
Study limitations
Our findings suggesting that RBM20 is associated with HCM are preliminary in nature. The study was only conducted on persons of East Asian descent, which may limit the generalizability of the conclusions to other population groups. The rare deleterious variants in this study were selected using prediction tools, carrying a risk of false positives. The pathogenicity of these variants was not confirmed by cosegregation and linkage analysis, owing to the lack of family history. In addition, with a limited sample size because of the nature of HCM and insufficient follow-up, the evidence may not be as strong as we expected, and caution must be applied in interpreting these results. Finally, all patients were recruited from a single centre. The relatedness analysis was performed with WES data, which may be limited in finding hidden relatedness compared with GWAS data.
Conclusions
This study identified RBM20 as a potential causal gene of HCM via gene-based association analysis in a large-scale population. Further survival analysis on the current cohort suggested that RBM20 is associated with an increased risk for SCA in HCM. These results will help complete the picture of candidate genes implicated in HCM and draw attention to the unique clinical characteristics and risk of SCA for RBM20 variant heterozygotes. Management strategies should include adequate surveillance for early detection of life-threatening arrhythmia. Further work is required to determine the causal role of RBM20 in HCM.
Funding Sources
This work was supported by projects National Key R&D Program of China number 2017YFC0909400; Nature Science Foundation of China numbers 91439203, 91839302, and 81700413; Shanghai Municipal Science and Technology Major Project number 2017SHZDZX01; and the fundamental Research Funds for the Central Universities, HUST number 2016JCTD117. The funders had no role in the design of the study and collection, analysis, decision to publish, interpretation of data, or preparation of the manuscript.
Disclosures
The authors have no conflicts of interest to disclose.
Acknowledgements
The authors acknowledge the patients and their families for their participation and support for this study. Also, we want to acknowledge the participants and investigators of FinnGen study.
2014 ESC Guidelines on diagnosis and management of hypertrophic cardiomyopathy: the Task Force for the Diagnosis and Management of Hypertrophic Cardiomyopathy of the European Society of Cardiology (ESC).
2020 AHA/ACC guideline for the diagnosis and treatment of patients with hypertrophic cardiomyopathy: a report of the American College of Cardiology/American Heart Association Joint Committee on Clinical Practice Guidelines.
Classification of the cardiomyopathies: a position statement from the European Society Of Cardiology Working Group on Myocardial and Pericardial Diseases.
Experimentally increasing the compliance of titin through RNA binding motif-20 (RBM20) inhibition improves diastolic function in a mouse model of heart failure with preserved ejection fraction.
00413-X&id=gr1.jpg){kind=link}
00413-X&id=gr2.jpg){kind=link}
00413-X&id=gr3.jpg){kind=link}