1. Introduction
Inherited cardiomyopathies represent relatively rare but morbid diseases and may lead to severe heart failure (HF), life-threatening ventricular arrhythmias (VA), and even sudden cardiac death (SCD) at a young age [
1]. The prevalence of inherited cardiomyopathies is estimated at 1/250 to 1/500 for hypertrophic cardiomyopathy (HCM), 1/250 for dilated cardiomyopathy (DCM), and 1/5000 for arrhythmogenic cardiomyopathy (ACM) [
2,
3,
4,
5]. In the majority of cases, an autosomal dominant inheritance can be found, yet recessive (homozygous or compound heterozygous) or X-linked forms can be identified [
6,
7]. Rare pathogenic variants are identified in a large proportion of cardiomyopathy patients, including genes encoding cardiac desmosomal proteins for arrhythmogenic cardiomyopathy (ACM), genes encoding sarcomeric proteins for hypertrophic cardiomyopathy (HCM) and dilated cardiomyopathy (DCM), and genes encoding structural myocardial proteins in DCM [
4,
8,
9]. Clinical and genetic overlap between those cardiomyopathies is recognized, which may compromise an accurate diagnosis. For example, the pathogenic variant c.40_42delAGA; p.(Arg14del) in the phospholamban (
PLN) gene may manifest as ACM or DCM (OMIM 609909) [
10].
Disease onset and severity are highly variable. Even within families carrying the same genetic variant, variable disease expression and age-related incomplete penetrance are frequently observed [
1]. Therefore, additional genetic and non-genetic factors (secondary hits) are believed to contribute to disease development and progression. For example, intensive exercise might increase disease penetrance and enhance the risk for life-threatening VA in ACM patients [
11]. Similarly, the presence of multiple (likely) pathogenic variants in ACM-associated genes may increase disease severity [
12]. Nevertheless, the prediction of severe outcomes among ACM and DCM patients remains exceedingly difficult, necessitating the identification of novel and accurate markers for risk stratification.
Recent and significant focus has been on the disturbance of calcium (Ca
2+) homeostasis within cardiomyocytes with regard to cardiomyopathy etiology. Ca
2+ plays a crucial role in the excitability and contraction of the heart. In failing cardiomyocytes, disturbed Ca
2+ handling contributes to HF and life-threatening VA [
13]. Therefore, dysfunction of factors that regulate Ca
2+ homeostasis, such as the histidine-rich calcium-binding protein (HRC), may contribute to the development and progression of inherited cardiomyopathies [
14]. HRC can interact with triadin when [Ca
2+] levels increase, thereby regulating RyR2 and modulating SR Ca
2+ release. When [Ca
2+] is low in the SR, HRC binds to the sarcoplasmic/endoplasmic reticulum Ca
2+-ATPase (SERCA2a), which suppresses SERCA2a function [
14,
15]. The serine residue at position 96 is a phosphorylation site of potential importance to triadin affinity (
Figure 1). In one candidate gene study, a polymorphism in the
HRC gene, c.286T>G p.(Ser96Ala)), was found to confer a four-fold higher risk of life-threatening VA and SCD in a small cohort of 123 patients with idiopathic DCM [
16].
To follow up on this observation, the aim of our study was to examine (1) the presence of the HRC p.(Ser96Ala) polymorphism in the general population (control cohort); (2) the effect of the HRC p.(Ser96Ala) polymorphism on major cardiac events in three patient cohorts, namely carriers of the PLN p.(Arg14del) pathogenic variant (cohort 1); patients diagnosed with, or predisposed to, ACM (cohort 2); and DCM patients (United Kingdom (UK) Biobank, cohort 3). We hypothesized, based on the earlier observation, that the HRC p.(Ser96Ala) polymorphism may contribute to the risk of major disease manifestation, which could potentially improve risk stratification.
2. Results
2.1. Patient Characteristics
In cohort 1, 1005 carriers of the
PLN p.(Arg14del) variant passed quality control (QC) and were included. Due to incomplete or unavailable health record data, 157 patients were excluded, leaving 848 patients for data analysis. In cohort 2, among 1033 ACM patients of European (-American) ancestry, 882 patients with complete health record data were included for further analysis. Finally, 1031 DCM patients of European ancestry were found in cohort 3, of which 985 individuals passed QC and were used for downstream analyses (
Figure 2).
Table 1,
Table 2 and
Table 3 summarize the clinical characteristics of the specific patient cohorts
PLN p.(Arg14del), ACM, and DCM, respectively, dividing the cohorts into wild-type (TT), heterozygous (TG) or homozygous (GG) for the
HRC polymorphism. The median age of cohort 1 (
PLN p.(Arg14del) carriers) ranged between 49 (interquartile range (IQR) 32–62) years for homozygous, 51 (IQR 37–63) for the wild type (WT), and 52 (IQR 36–65) years for heterozygous
HRC carriers. No significant differences in male sex (43% for the WT; 45% for heterozygous and homozygous) or index patients (WT = 20%, heterozygous = 19%, and homozygous = 16%) were found between the three different polymorphism groups. Median follow-up was 6 (IQR 2–10) years for WT and 5 (IQR 2–9) years for heterozygous and homozygous
HRC carriers. However, no statistically significant differences were found between the three groups.
In cohort 2 (ACM,
Table 2), approximately half of the patients included were males (WT = 50%, heterozygous = 47%, and homozygous = 55%), while 45% of the WT, 48% of heterozygous, and 47% of homozygous
HRC carriers were index patients. The median follow-up was 8 [4–15] years for WT and heterozygous carriers and 10 [3–14] years for homozygous
HRC carriers. For all parameters, no significant differences between groups were found.
Finally, in cohort 3 (DCM,
Table 3), the mean age at enrollment was 61 years for all three polymorphism groups. Compared to cohorts 1 and 2, predominantly males were included in the analysis: 71% in WT, 68% for heterozygous, and 79% for homozygous carriers. The mean follow-up time of all three
HRC groups was approximately 10 years. No statistical differences were found between the groups.
2.2. HRC p.(Ser96Ala) Polymorphism in General and Cardiomyopathy Populations
The minor allele frequency (MAF) of the HRC p.(Ser96Ala) polymorphism in the control cohorts ranged from 40.3 to 42.2%, as analyzed in different databases.
In the three patient cohorts, the MAF of the
HRC p.(Ser96Ala) polymorphism was 40.9% among
PLN p.(Arg14del) carriers (cohort 1), 43.9% for the overall ACM cohort (cohort 2), and 41.3% for the UK Biobank DCM cohort (cohort 3). In more detail, the Dutch ACM cohort (
n = 491) showed a MAF of 43.6%, while a MAF of 43.0% was found among US ACM patients (
n = 337) and 50.9% among the Swiss ACM cohort (
n = 54). Data regarding the different observed frequencies are summarized in
Table 4.
2.3. Life-Threatening Ventricular Arrhythmias and HRC Polymorphism
Among
PLN p.(Arg14del) carriers from cohort 1, life-threatening VA events were observed in 49/294 (16%) of patients with WT p.(Ser96Ala), 54/392 (14%) of heterozygotes, and 16/93 (17%) of p.(Ser96Ala) homozygotes (
Table 1). Logistic regression revealed no significant association between p.(Ser96Ala) and life-threatening VA among
PLN p.(Arg14del) pathogenic variant carriers (Odds Ratio (OR) (95% confidence interval (CI)) 0.792 (0.584–1.066),
p = 0.128;
Table 5). Moreover, in prespecified analyses of subgroups (restricting to index patients or relatives), only a significant association between the p.(Ser96Ala) polymorphism and life-threatening VA in index patients was found (OR (95% CI) = 0.554 (0.300–0.985),
p = 0.049 *), but after correction for multiple testing, this statistical result was not sustained (
Supplemental Table S1). Finally, time-to-event analyses revealed that the p.(Ser96Ala) polymorphism was not significantly associated with life-threatening VA when subjected from birth or from enrollment into the registry (
Supplemental Table S2).
Among ACM patients from cohort 2, life-threatening VA events were observed in 120/283 (42%) WT carriers, 186/375 (50%) heterozygotes, and 77/174 (44%) p.(Ser96Ala) homozygotes.
HRC status was not statistically associated with life-threatening arrhythmias in this patient group (OR (95% CI) 0.862 (0.576–1.288),
p = 0.467,
Table 6), further supporting the results observed among
PLN carriers in cohort 1.
We then assessed the risk of events by p.(Ser96Ala) status among patients with DCM using UK Biobank samples from cohort 3. Logistic regression among these DCM cases also revealed no significant association with different arrhythmic parameters, including ventricular tachycardia (VT) (OR (95% CI) 0.848 (0.524–1.171),
p = 0.210), SCD (OR (95% CI) 1.016 (0.985–1.046),
p = 0.941), or implantable cardioverter defibrillator (ICD) implantation (OR (95% CI) 0.887 (0.652–1.121),
p = 0.284,
Table 7). Time-to-event analyses similarly yielded insignificant results, with hazard ratios inconsistent with large effects (
Supplemental Table S3). In addition, there was no evidence of increased risk for all-cause mortality (HR (95% CI) 1.140 (0.943–1.378),
p = 0.172).
2.4. Heart Failure and HRC Polymorphism
Among
PLN p.(Arg14del) carriers from cohort 1, HF was observed in 42/257 (16%) WT
HRC carriers, 53/345 (15%) that were heterozygous, and 15/113 (13%) p.(Ser96Ala) homozygotes (
Table 1). No significant association was found between the
HRC polymorphism and the HF outcome using logistic regression (OR (95% CI) 0.858 (0.615–1.188),
p = 0.360;
Table 5). The prevalence and severity of progressive HF in ACM might be low given a currently debated definition of HF among these cases, particularly in the right dominant subforms of the disease [
17]. In addition, overlapping criteria for congestive HF and DCM are frequently used in the research literature (i.e., LVEF < 45%). For these two reasons, no analyses for the risk of HF outcomes were performed for cohorts 2 and 3.
2.5. Composite Endpoint in PLN p.(Arg14del) Patients
In
PLN p.(Arg14del) carriers (cohort 1), a composite endpoint was observed in 67/302 (22%) of the WT
HRC carriers, 81/404 (20%) of the heterozygous carriers, and 24/142 (17%) for homozygous
HRC carriers. No effect of the p.(Ser96Ala) polymorphism with this composite endpoint in
PLN p.(Arg14del) carriers was observed when adjusting for the prespecified covariates (OR (95% CI) 0.842 (0.649–1.089),
p = 0.193;
Table 5).
3. Discussion
In this study, we examined the role of the previously identified
HRC p.(Ser96Ala) polymorphism in relation to major cardiac events in clinical cohorts diagnosed with, or predisposed to, different forms of cardiomyopathy. Most carriers develop symptoms between their third or fifth decade of life, suggesting an incomplete penetrance. In addition, a wide range of symptoms among patients carrying the same genetic variant can be found [
18]. Therefore, improvement in proper risk stratification is needed in these patients. We established that the
HRC p.(Ser96Ala) polymorphism is frequent (40–42%) in the general population, with highly similar frequencies among the different studied patient cohorts. Importantly, we found that the p.(Ser96Ala) polymorphism did not significantly contribute to the risk of major cardiac events among
PLN p.(Arg14del) carriers, ACM patients, or DCM patients, contrasting with previous reports.
Recently, studies have started to explore the role of common genetic variants in modifying the risk of cardiomyopathies. Genome-wide association studies for DCM have identified several common genetic variants associated with disease risk [
19]. Notably, common genetic variants have been shown to modify the penetrance and expressivity of HCM in patients with rare variants [
20]. Given the marked incomplete penetrance and variable expressivity of rare variants—and lack of known pathogenic variants in many index patients with ACM and DCM—common variants may similarly contribute to expressivity in these cardiomyopathies. This is especially relevant as the prediction of major adverse events, including HF and VA, remains exceedingly challenging.
In the present study, we assessed the risk of HF and VA among a broad range of patient cohorts, namely,
PLN p.Arg14del carriers, ACM index patients and family members, and DCM patients. This approach was chosen given that the phenotypic outcomes of these entities share overlapping features, among which there is a significant risk of life-threatening VA. Classically, ACM was described as arrhythmogenic right ventricular cardiomyopathy (ARVC) with sole involvement of the right ventricle; however, biventricular and left dominant forms are now increasingly recognized [
5]. Conversely, a decline in right ventricular function is a predictor of worse outcomes in DCM patients [
21]. Carriers of
PLN p.(Arg14del) can be diagnosed with either of these cardiomyopathies, reflecting the clinical and genetic overlap between these disease entities [
10]. Finally, in both ACM and DCM, disturbance of Ca
2+ handling is known to drive the development of life-threatening arrhythmias and HF [
22,
23].
Given this apparent functional mechanism, the p.(Ser96Ala) polymorphism could potentially influence Ca
2+ regulation and thereby contribute to VA or deterioration of Ca
2+ handling. Indeed, a marked effect was described in an initial study of 123 idiopathic DCM patients—where p.(Ser96Ala) homozygotes had an over four-fold risk of major arrhythmic events [
16]. In contrast, we were unable to confirm a role for the
HRC p.(Ser96Ala) polymorphism in stratifying the risk of major events among three larger cohorts of patients with (predisposition to) cardiomyopathy. Furthermore, directionally inconsistent results were found in our study compared to the study of Arvanitis et al., as the p.(Ser96Ala) polymorphism seemed to be protective for major cardiac events [
16]. We note that several of our cohorts included individuals with preclinical disease or genetic predisposition only, which contrasts with the established DCM patients analyzed in the work of Arvanitis et al.; however, we performed several sensitivity analyses, including index patients only, which generally showed consistent effects. In addition, the one significant association between the index and life-threatening VA, even if not significant after correction for multiple tests, was directionally inconsistent compared to the aforementioned study. Furthermore, it is possible that our study remains underpowered to detect a small effect of p.(Ser96Ala) on major events. Nevertheless, the 95% CI of our estimates—in all cohorts and analyses—excludes a large effect, especially one as large as described by Arvanitis et al.
4. Materials and Methods
4.1. Study Design and Patient Selection
In this study, individuals from four different patient and control cohorts were included:
Cohort 1 included individuals of Dutch ancestry enrolled in the Dutch ACM registry in whom the pathogenic p.(Arg14del) variant in
PLN was identified (accessed on 7 March 2023). Within this database, we selected
PLN p.(Arg14del) carriers, and this will be referred to as the
PLN registry. This registry includes data from individuals obtained from three Dutch university medical centers (Groningen, Amsterdam, and Utrecht) [
24]. The study cohort consisted of both index patients (i.e., first affected family member tested positive for the disease-associated genetic variant, mostly because of suspected inherited cardiac disease) and relatives after genetic cascade testing.
Cohort 2 included individuals enrolled in the ACM registries from the Netherlands [
24], Switzerland, and The United States (USA, data accessed on 8 March 2023). Index and relatives with a definitive ARVC diagnosis based on the 2010 modified Task Force Criteria (TFC) and relatives with a pathogenic variant in an ARVC-related gene (who do not meet TFC) or gene elusive were included.
Cohort 3 included DCM patients identified within the UK Biobank cohort. Patients with a clinical DCM diagnosis were identified using International Classification of Diseases (ICD-10 code I42.0) [
25].
Control cohort included individuals that belong to the “general population”. Population data consisted of three publicly available databases: (1) Genome Aggregation Database (gnomAD v2.1.1,
https://gnomad.broadinstitute.org/, accessed on 23 March 2023), (2) 998 individuals from The Genome of the Netherlands (GoNL, accessed on 25 May 2023) [
26], and (3) participants of the UK Biobank (accessed on 26 April 2023). The UK Biobank is a large population-based prospective study that included 500,000 UK participants between 40 and 69 years [
27]. Additionally, 31,400 individuals who were referred to the Department of Genetics of the UMC Utrecht, the Netherlands (inclusion between 2017 and 2023, data accessed on 5 May 2023) were evaluated irrespective of diagnosis. In this group, a whole exome sequencing (WES)-based genetic test was available as part of regular clinical care.
For all cohorts, only individuals with European (-American) ancestry were selected for further analysis. Written informed consent was provided by all participating patients. Furthermore, all data were fully anonymized before data could be accessed. This study was conducted according to the Declaration of Helsinki. Study protocol was approved by UMC Utrecht (Biobank number 12–387). Use of UK Biobank data was performed under application number 17,488 and was approved by the local Massachusetts General Hospital Institutional Review Board.
4.2. Genetic Data Extraction
Cohort 1 and 2: DNA samples of PLN p.(Arg14del) carriers, ACM patients, and preclinical gene variant carriers enrolled in the ACM registry were genotyped on the Illumina Global Screening Array (-24 v3.0 BeadChip, San Diego, CA, USA) at the Human Genomics Facility (HuGe-F), Erasmus Medical Center, Rotterdam, the Netherlands and at the Genetic Resources Core Facility (GRCF), at John Hopkins University, Baltimore, the USA. The p.(Ser96Ala) polymorphism (T/G) (rs3745297) in HRC was extracted on imputed data after standard sample QC (INFO score > 0.99; removal of samples was performed for those who revoked consent, had a mismatch between genetically predicted and self-reported sex, were outliers for heterozygosity or missingness, had putative sex chromosome aneuploidy, or were ancestral outliers in a Principal Component (PC) Analysis) using Plink v1.9 and v2.0. The HRC polymorphism was categorized as either WT (TT), heterozygous (TG), or homozygous (GG).
Cohort 3: UK Biobank participants underwent dense genotyping using the UK Biobank Axiom Array or the Affymetrix UK BiLEVE Axiom Array (Thermo Fisher Scientific, Waltham, MA, USA), as described by Bycroft et al. [
28]. We used the version 3 imputed data, where the p.(Ser96Ala) polymorphism attained near-perfect imputation accuracy (INFO > 0.99). Standard sample QC was performed as described before.
Control cohort: the
HRC p.(Ser96Ala) polymorphism was extracted from gnomAD (WES data used) [
29]; GoNL (whole-genome sequencing (WGS) data used) [
26]; the UK Biobank (version 3 imputed data; INFO score > 0.99; sample QC described above) [
28]; and in patients in whom WES was performed as part of regular clinical care, irrespective of diagnosis, in UMC Utrecht, Utrecht, The Netherlands between 2017 and 2023.
4.3. Clinical Outcomes
The first primary outcome was defined as the first occurrence of a life-threatening VA event, including sustained VT or ventricular fibrillation (VF), appropriate ICD therapy, and (aborted) SCD. The second primary outcome was defined as the first occurrence of an HF-related event. This included hospitalization for HF-related complaints, implantation of a left ventricular assistance device (LVAD), heart transplantation, or HF-related death. A secondary outcome was a composite of both life-threatening VA and/or HF-related events. For cohorts 1–2, clinical information from the first cardiac evaluation and follow-up visits was retrospectively extracted from the ACM and
PLN registries, as described before [
24]. Age at enrollment in the registry was defined by presenting with clinical presentation due to cardiomyopathy-related symptoms, SCD, or family screening [
24].
For the UK Biobank DCM cohort (cohort 3), participants attended an entry assessment at centers across the UK to provide baseline characteristics. Follow-up data were collected via hospital event data and death registry linkage; main outcomes (VT, SCD, and ICD implantation) were defined using ICD codes, as previously described [
25]. The latest update of health record linkage data was performed in October 2020.
4.4. Statistical Analysis
4.4.1. Registry Cohorts
Outcomes and covariates of interest were collected and presented as descriptive statistics for each registry cohort; categorical variables were analyzed using Chi-Square tests, and continuous variables were analyzed using Kruskal–Wallis tests. In each dataset, logistic regression was performed with life-threatening VA, HF, or both (composite outcome) modeled as outcome; the
HRC p.(Ser96Ala) was modeled as a predictor, adjusting for covariates of sex, age at enrollment (in the registry cohorts), and the first 12 PCs of ancestry. Logistic regression analyses were repeated within subgroups, i.e., restricting to index cases or relatives. Additionally, we performed analyses using only incident events modeled in a Cox proportional hazard regression (time-to-event). Cox models were adjusted for the same covariates as in the logistic regression analysis and right-censored at last follow-up at the cardiology outpatient clinic or death if the primary outcome had not occurred. Start of incident time was modeled in two ways, namely, (1) from birth (which effectively is true since the genetic variant is already present at birth) and (2) from enrollment in the ACM or
PLN registry. In addition to the aforementioned outcomes, mortality was also assessed in time-to-event analyses. Differences were considered statistically significant when
p < 0.05. Statistical analyses were performed with R version 4.0.3, using the “survival” and “survminer” packages [
30,
31].
4.4.2. UK Biobank DCM Cohort
In the UK Biobank DCM cohort, logistic regression was used, pooling incident and prevalent events to maximize case numbers and power. Outcomes included VA, SCD, and ICD implantation. Models were adjusted for age, sex, genotyping array, and the first 12 PCs of ancestry. Because logistic regression may not correctly model temporality and competing risks, we also performed analyses using only incident events modeled in a Cox proportional hazard regression, using function coxph() from R package "survival" [
30]. Cox models were adjusting for the same covariates and right-censored at death to account for competing risks of death. Start of incident time was modeled in two ways, namely, (1) from enrollment in UK Biobank and (2) from DCM diagnosis. In addition to the aforementioned outcomes, mortality was also assessed in time-to-event analyses. Differences were considered statistically significant when
p < 0.05.
5. Conclusions
In this study, we found that the p.(Ser96Ala) polymorphism is common in the general population and cardiomyopathy-affected patients. Furthermore, there is a lack of evidence for the role of the p.(Ser96Ala) HRC polymorphism in modifying the risk of major cardiac events among cardiomyopathy patients. Our data indicate that any possible effect is, at most, small, limiting its use as a sole predictor of severity among ACM and DCM patients. Further research is required to identify bona fide predictors for stratification of cardiomyopathy patients and their risk for life-threatening outcomes.
Limitations
Despite the unique (and rare) composition of our included cohorts, this study could be subjected to several limitations. A larger part of the
PLN p.(Arg14del) pathogenic variant carriers and ACM patients are classified as still being in the preclinical phase. Therefore, clinical events might be low compared to a cohort that includes only diagnosed patients, such as the DCM patients from the UK Biobank. In addition, lifestyle factors such as diet, environment, and exercise might influence and accelerate disease manifestation and events. However, these data were not available when we performed our analyses. In addition, some clinical, MRI, or echocardiographic parameters have not been reliably collected (e.g., medication use) or were not available. However, we have the impression that the unavailability of these data does not hamper the major conclusions and genetic observations of this study. When estimating a population MAF, most of the studies make use of a controlled setting in which cases and controls are matched. However, whether this group represents the population frequencies might not be taken into account [
32]. Therefore, extrapolation of the observed allele frequencies to the general population can be inadequate. Finally, we note that our study was largely focused on individuals of European ancestry, potentially limiting the generalizability to other ancestry groups. As illustrated in a single study on Japanese paroxysmal atrial fibrillation patients, a MAF of 26% was reported, which was similar to population data of East Asian ancestry found in gnomAD [
33].
Author Contributions
Conceptualization, S.M.v.d.V., E.v.D., J.P.v.T. and T.A.B.v.V.; methodology, S.M.v.d.V. and E.v.D.; software, E.v.D. and S.J.J.; validation, S.M.v.d.V., E.v.D. and S.J.J.; formal analysis, S.M.v.d.V., E.v.D. and S.J.J.; investigation, S.M.v.d.V. and E.v.D.; resources, V.P., R.F.E., C.A.J., C.T., B.M., H.C., A.M.S., F.D. and S.J.J.; data curation, ACM/PLN Registry, E.v.D., V.P. and S.J.J.; writing—original draft preparation, S.M.v.d.V. and E.v.D.; writing—review and editing, C.A.J., C.T., B.M., H.C A.M.S., F.D., S.J.J., J.P.v.T. and T.A.B.v.V.; visualization, S.M.v.d.V., E.v.D., K.D. and S.J.J.; supervision, P.T.E., C.R.B., J.P.v.T. and T.A.B.v.V.; project administration, J.P.v.T. and T.A.B.v.V.; funding acquisition, T.A.B.v.V., S.J.J., E.v.D., A.M.S., F.D., C.A.J., C.T., B.M. and H.C. All authors have read and agreed to the published version of the manuscript.
Funding
This research was funded by a grant from the Netherlands Cardio Vascular Research Initiative (CVON): the Dutch Heart Foundation, the Dutch Federation of University Medical Centers, the Netherlands Organization for Health Research and Development and the Royal Netherlands Academy of Sciences (CVON-eDETECT 2015-12, PREDICT2 2018-30 and DoubleDose 2020B005), and the Leducq CUREPLaN. Further financial support was obtained from the Netherlands Heart Institute in conjunction with the PLN patient foundation (to TABvV). This work was also supported by a grant from the Dutch Heart Foundation (grant No. 03-007-2022-0035), an Amsterdam UMC Doctoral Fellowship to S.J.J., and a CVON-PREDICT2 Young Talent Program Grant to E.v.D. The Zurich ARVC Program is supported by the Georg und Bertha Schwyzer-Winiker Foundation, Baugarten Foundation, USZ Foundation (Dr. Wild Grant), Swiss Heart Foundation grant No. FF17019 and FF21073 to A.M.S., and Swiss National Science Foundation grant No. 160327 to F.D. The Johns Hopkins ARVD/C Program is supported by the Leonie-Wild Foundation; the Leyla Erkan Family Fund for ARVD Research; The Hugh Calkins, Marvin H. Weiner, and Jacqueline J. Bernstein Cardiac Arrhythmia Center; the Dr. Francis P. Chiramonte Private Foundation; the Dr. Satish, Rupal, and Robin Shah ARVD Fund at Johns Hopkins; the Bogle Foundation; the Campanella family; the Patrick J. Harrison Family; the Peter French Memorial Foundation; and the Wilmerding Endowments.
Institutional Review Board Statement
This study was conducted in accordance with the Declaration of Helsinki and approved by the Institutional Review Board (or Ethics Committee) of University Medical Center Utrecht (protocol code #12-387), Zurich Hospital (KEK-ZH-Nr. PB 2016-02109) and by the local Massachusetts General Hospital Institutional Review Board (protocol code #17488).
Informed Consent Statement
Informed consent was obtained from all subjects involved in the study. Written informed consent has been obtained from the patients to publish this paper.
Data Availability Statement
The data presented in this study are available on request from the corresponding author.
Acknowledgments
We would like to thank all the patients, relatives, and biobank participants, without whom this work was not possible. Collaborators on the ACM/PLN Registry: Steven A. Muller, Mimount Bourfiss, Tom E. Verstraelen, Myrthe Y.C. van der Heide, Remco de Brouwer, Anneline S.J.M. te Riele, Rudolf A. de Boer, Paul A. van der Zwaag, Arjan C. Houweling, and Moniek G.P.J. Cox.
Conflicts of Interest
A.M.S. received educational grants through his institution from Abbott, Bayer Healthcare, Biosense Webster, Biotronik, Boston Scientific, BMS/Pfizer, and Medtronic and speaker/advisory board/consulting fees from Bayer Healthcare, Biotronik, Daiichi-Sankyo, Medtronic, Novartis, Pfizer, and Stride Bio Inc. Calkins is a consultant for Medtronic Inc., Biosense Webster, Pfizer, StrideBio, Rocket, and Abbott. Murray is a consultant for MyGeneCounsel. James has consulted for Pfizer, Inc., Lexeo, Inc., and StrideBio, Inc. (uncompensated). Tichnell is a consultant for StrideBio Inc. C.A.J. receives research funding from Lexeo Therapeutics and Stride Bio Inc. H.C. receives research support from Tenaya Inc. and Boston Scientific Corp. C.T. receives salary support from these grants.
References
- McKenna, W.J.; Judge, D.P. Epidemiology of the inherited cardiomyopathies. Nat. Rev. Cardiol. 2021, 18, 22–36. [Google Scholar] [CrossRef] [PubMed]
- Semsarian, C.; Ingles, J.; Maron, M.S.; Maron, B.J. New perspectives on the prevalence of hypertrophic cardiomyopathy. J. Am. Coll. Cardiol. 2015, 65, 1249–1254. [Google Scholar] [CrossRef] [PubMed]
- Maron, B.J.; Gardin, J.M.; Flack, J.M.; Gidding, S.S.; Kurosaki, T.T.; Bild, D.E. Prevalence of hypertrophic cardiomyopathy in a general population of young adults. Echocardiographic analysis of 4111 subjects in the CARDIA Study. Coronary Artery Risk Development in (Young) Adults. Circulation 1995, 92, 785–789. [Google Scholar] [CrossRef] [PubMed]
- Hershberger, R.E.; Hedges, D.J.; Morales, A. Dilated cardiomyopathy: The complexity of a diverse genetic architecture. Nat. Rev. Cardiol. 2013, 10, 531–547. [Google Scholar] [CrossRef] [PubMed]
- Corrado, D.; Basso, C.; Judge, D.P. Arrhythmogenic Cardiomyopathy. Circ. Res. 2017, 121, 784–802. [Google Scholar] [CrossRef] [PubMed]
- Brodehl, A.; Meshkov, A.; Myasnikov, R.; Kiseleva, A.; Kulikova, O.; Klauke, B.; Sotnikova, E.; Stanasiuk, C.; Divashuk, M.; Pohl, G.M.; et al. Hemi- and Homozygous Loss-of-Function Mutations in DSG2 (Desmoglein-2) Cause Recessive Arrhythmogenic Cardiomyopathy with an Early Onset. Int. J. Mol. Sci. 2021, 22, 3786. [Google Scholar] [CrossRef] [PubMed]
- Lester, G.; Femia, G.; Ayer, J.; Puranik, R. A case report: X-linked dystrophin gene mutation causing severe isolated dilated cardiomyopathy. Eur. Heart J. Case Rep. 2019, 3, ytz055. [Google Scholar] [CrossRef]
- Gerull, B.; Brodehl, A. Insights into genetics and pathophysiology of arrhythmogenic cardiomyopathy. Curr. Heart Fail. Rep. 2021, 18, 378–390. [Google Scholar] [CrossRef]
- Lopes, L.R.; Rahman, M.S.; Elliott, P.M. A systematic review and meta-analysis of genotype-phenotype associations in patients with hypertrophic cardiomyopathy caused by sarcomeric protein mutations. Heart 2013, 99, 1800–1811. [Google Scholar] [CrossRef]
- Vafiadaki, E.; Glijnis, P.C.; Doevendans, P.A.; Kranias, E.G.; Sanoudou, D. Phospholamban R14del disease: The past, the present and the future. Front. Cardiovasc. Med. 2023, 10, 1162205. [Google Scholar] [CrossRef]
- Martínez-Solé, J.; Sabater-Molina, M.; Braza-Boïls, A.; Santos-Mateo, J.J.; Molina, P.; Martínez-Dolz, L.; Gimeno, J.R.; Zorio, E. Corrigendum: Facts and Gaps in Exercise Influence on Arrhythmogenic Cardiomyopathy: New Insights from a Meta-Analysis Approach. Front. Cardiovasc. Med. 2021, 8, 816280. [Google Scholar] [CrossRef] [PubMed]
- Nagyova, E.; Hoorntje, E.T.; Te Rijdt, W.P.; Bosman, L.P.; Syrris, P.; Protonotarios, A.; Elliott, P.M.; Tsatsopoulou, A.; Mestroni, L.; Taylor, M.R.G.; et al. A Systematic Analysis of the Clinical Outcome Associated with Multiple Reclassified Desmosomal Gene Variants in Arrhythmogenic Right Ventricular Cardiomyopathy Patients. J. Cardiovasc. Transl. Res. 2023. online ahead of print. [Google Scholar] [CrossRef] [PubMed]
- Piacentino, V.; Weber, C.R.; Chen, X.; Weisser-Thomas, J.; Margulies, K.B.; Bers, D.M.; Houser, S.R. Cellular basis of abnormal calcium transients of failing human ventricular myocytes. Circ. Res. 2003, 92, 651–658. [Google Scholar] [CrossRef] [PubMed]
- Arvanitis, D.A.; Vafiadaki, E.; Johnson, D.M.; Kranias, E.G.; Sanoudou, D. The Histidine-Rich Calcium Binding Protein in Regulation of Cardiac Rhythmicity. Front. Physiol. 2018, 9, 1379. [Google Scholar] [CrossRef] [PubMed]
- Arvanitis, D.A.; Vafiadaki, E.; Sanoudou, D.; Kranias, E.G. Histidine-rich calcium binding protein: The new regulator of sarcoplasmic reticulum calcium cycling. J. Mol. Cell. Cardiol. 2011, 50, 43–49. [Google Scholar] [CrossRef] [PubMed]
- Arvanitis, D.A.; Sanoudou, D.; Kolokathis, F.; Vafiadaki, E.; Papalouka, V.; Kontrogianni-Konstantopoulos, A.; Theodorakis, G.N.; Paraskevaidis, I.A.; Adamopoulos, S.; Dorn, G.W.; et al. The Ser96Ala variant in histidine-rich calcium-binding protein is associated with life-threatening ventricular arrhythmias in idiopathic dilated cardiomyopathy. Eur. Heart J. 2008, 29, 2514–2525. [Google Scholar] [CrossRef] [PubMed]
- Chen, S.; Chen, L.; Duru, F.; Hu, S. Heart Failure in Patients with Arrhythmogenic Cardiomyopathy. J. Clin. Med. 2021, 10, 4782. [Google Scholar] [CrossRef] [PubMed]
- Kingdom, R.; Wright, C.F. Incomplete penetrance and variable expressivity: From clinical studies to population cohorts. Front. Genet. 2022, 13, 920390. [Google Scholar] [CrossRef]
- Villard, E.; Perret, C.; Gary, F.; Proust, C.; Dilanian, G.; Hengstenberg, C.; Ruppert, V.; Arbustini, E.; Wichter, T.; Germain, M.; et al. A genome-wide association study identifies two loci associated with heart failure due to dilated cardiomyopathy. Eur. Heart J. 2011, 32, 1065–1076. [Google Scholar] [CrossRef]
- Biddinger, K.J.; Jurgens, S.J.; Maamari, D.; Gaziano, L.; Choi, S.H.; Morrill, V.N.; Halford, J.L.; Khera, A.V.; Lubitz, S.A.; Ellinor, P.T.; et al. Rare and common genetic variation underlying the risk of hypertrophic cardiomyopathy in a national biobank. JAMA Cardiol. 2022, 7, 715–722. [Google Scholar] [CrossRef]
- La Vecchia, L.; Varotto, L.; Zanolla, L.; Spadaro, G.L.; Fontanelli, A. Right ventricular function predicts transplant-free survival in idiopathic dilated cardiomyopathy. J. Cardiovasc. Med. 2006, 7, 706–710. [Google Scholar] [CrossRef] [PubMed]
- Moccia, F.; Lodola, F.; Stadiotti, I.; Pilato, C.A.; Bellin, M.; Carugo, S.; Pompilio, G.; Sommariva, E.; Maione, A.S. Calcium as a key player in arrhythmogenic cardiomyopathy: Adhesion disorder or intracellular alteration? Int. J. Mol. Sci. 2019, 20, 3986. [Google Scholar] [CrossRef] [PubMed]
- Liu, L.; Zhou, K.; Liu, X.; Hua, Y.; Wang, H.; Li, Y. The interplay between cardiac dyads and mitochondria regulated the calcium handling in cardiomyocytes. Front. Physiol. 2022, 13, 1013817. [Google Scholar] [CrossRef]
- Bosman, L.P.; Verstraelen, T.E.; van Lint, F.H.M.; Cox, M.G.P.J.; Groeneweg, J.A.; Mast, T.P.; van der Zwaag, P.A.; Volders, P.G.A.; Evertz, R.; Wong, L.; et al. The Netherlands Arrhythmogenic Cardiomyopathy Registry: Design and status update. Neth. Heart J. 2019, 27, 480–486. [Google Scholar] [CrossRef] [PubMed]
- Jurgens, S.J.; Choi, S.H.; Morrill, V.N.; Chaffin, M.; Pirruccello, J.P.; Halford, J.L.; Weng, L.-C.; Nauffal, V.; Roselli, C.; Hall, A.W.; et al. Analysis of rare genetic variation underlying cardiometabolic diseases and traits among 200,000 individuals in the UK Biobank. Nat. Genet. 2022, 54, 240–250. [Google Scholar] [CrossRef] [PubMed]
- Genome of the Netherlands Consortium. Whole-genome sequence variation, population structure and demographic history of the Dutch population. Nat. Genet. 2014, 46, 818–825. [Google Scholar] [CrossRef] [PubMed]
- Sudlow, C.; Gallacher, J.; Allen, N.; Beral, V.; Burton, P.; Danesh, J.; Downey, P.; Elliott, P.; Green, J.; Landray, M.; et al. UK Biobank: An open access resource for identifying the causes of a wide range of complex diseases of middle and old age. PLoS Med. 2015, 12, e1001779. [Google Scholar] [CrossRef] [PubMed]
- Bycroft, C.; Freeman, C.; Petkova, D.; Band, G.; Elliott, L.T.; Sharp, K.; Motyer, A.; Vukcevic, D.; Delaneau, O.; O’Connell, J.; et al. The UK Biobank resource with deep phenotyping and genomic data. Nature 2018, 562, 203–209. [Google Scholar] [CrossRef]
- Karczewski, K.J.; Francioli, L.C.; Tiao, G.; Cummings, B.B.; Alföldi, J.; Wang, Q.; Collins, R.L.; Laricchia, K.M.; Ganna, A.; Birnbaum, D.P.; et al. The mutational constraint spectrum quantified from variation in 141,456 humans. Nature 2020, 581, 434–443. [Google Scholar] [CrossRef]
- Therneau, T.M. Package for Survival Analysis in R. Available online: https://CRAN.R-project.org/package=survival (accessed on 31 May 2023).
- Kassambara, A. Survminer: Drawing Survival Curves Using “Ggplot2”. Available online: https://CRAN.R-project.org/package=survminer (accessed on 31 May 2023).
- Cross, D.S.; Ivacic, L.C.; Stefanski, E.L.; McCarty, C.A. Population based allele frequencies of disease associated polymorphisms in the Personalized Medicine Research Project. BMC Genet. 2010, 11, 51. [Google Scholar] [CrossRef]
- Amioka, M.; Nakano, Y.; Ochi, H.; Onohara, Y.; Sairaku, A.; Tokuyama, T.; Motoda, C.; Matsumura, H.; Tomomori, S.; Hironobe, N.; et al. Ser96Ala genetic variant of the human histidine-rich calcium-binding protein is a genetic predictor of recurrence after catheter ablation in patients with paroxysmal atrial fibrillation. PLoS ONE 2019, 14, e0213208. [Google Scholar] [CrossRef]
Figure 1.
Schematic overview of the interactions of HRC protein in the SR. At low [Ca2+] in the SR, HRC protein interacts with SERCA2a. When [Ca2+] rises, HRC is phosphorylated by Fam20C at Ser96 position, which induces interaction with triadin to modulate RyR2 function. When the serine residue is altered for an alanine residue (Ala96Ala variant), phosphorylation is not performed, and therefore affinity for SERCA2a and triadin remains unaffected. HRC; histidine-rich calcium-binding protein, SR; sarcoplasmic reticulum, Ca2+; calcium, SERCA; sarcoplasmic/endoplasmic reticulum Ca2+-ATPase, PLN; phospholamban, RyR2; ryanodine receptor 2, Ser; serine, Ala; alanine, Fam20C; family with sequence similarity 20C.
Figure 1.
Schematic overview of the interactions of HRC protein in the SR. At low [Ca2+] in the SR, HRC protein interacts with SERCA2a. When [Ca2+] rises, HRC is phosphorylated by Fam20C at Ser96 position, which induces interaction with triadin to modulate RyR2 function. When the serine residue is altered for an alanine residue (Ala96Ala variant), phosphorylation is not performed, and therefore affinity for SERCA2a and triadin remains unaffected. HRC; histidine-rich calcium-binding protein, SR; sarcoplasmic reticulum, Ca2+; calcium, SERCA; sarcoplasmic/endoplasmic reticulum Ca2+-ATPase, PLN; phospholamban, RyR2; ryanodine receptor 2, Ser; serine, Ala; alanine, Fam20C; family with sequence similarity 20C.
Figure 2.
Flowchart inclusion of carriers of the PLN p.(Arg14del) pathogenic variant (cohort 1); patients diagnosed with, or predisposed to, ACM (cohort 2); and DCM patients (cohort 3). PLN; phospholamban, ACM; arrhythmogenic cardiomyopathy, DCM; dilated cardiomyopathy, UK; United Kingdom, QC: quality control.
Figure 2.
Flowchart inclusion of carriers of the PLN p.(Arg14del) pathogenic variant (cohort 1); patients diagnosed with, or predisposed to, ACM (cohort 2); and DCM patients (cohort 3). PLN; phospholamban, ACM; arrhythmogenic cardiomyopathy, DCM; dilated cardiomyopathy, UK; United Kingdom, QC: quality control.
Table 1.
Patient characteristics of the 848 PLN p.(Arg14del) carriers included in this study. Patients were divided into wild type for HRC p.(Ser96Ser) variant carriers (N = 302), heterozygous for HRC p.(Ser96Ala) (N = 404), and homozygous carriers p.(Ala96Ala, N = 142). Data are depicted as median (interquartile range) or n/N (%).
Table 1.
Patient characteristics of the 848 PLN p.(Arg14del) carriers included in this study. Patients were divided into wild type for HRC p.(Ser96Ser) variant carriers (N = 302), heterozygous for HRC p.(Ser96Ala) (N = 404), and homozygous carriers p.(Ala96Ala, N = 142). Data are depicted as median (interquartile range) or n/N (%).
PLN p.(Arg14del) N = 848 | Wild Type (TT) (N = 302, 36%) | Heterozygous (TG) (N = 404, 48%) | Homozygous (GG) (N = 142, 16%) |
---|
Age (years) | 51 [37–63] | 52 [36–65] | 49 [32–62] |
Male sex | 131/302 (43%) | 182/404 (45%) | 64/142 (45%) |
Index patient | 61/302 (20%) | 78/404 (19%) | 23/142 (16%) |
Diagnosis | | | |
DCM | 42/195 (22%) | 61/257 (24%) | 21/100 (21%) |
ACM | 21/184 (11%) | 27/254 (11%) | 6/96 (6%) |
ICD implantation | 90/267 (34%) | 115/356 (32%) | 43/120 (36%) |
Continuous rhythm monitoring | | | |
≥500 PVCs | 61/190 (32%) | 93/250 (37%) | 28/88 (32%) |
Imaging | | | |
LVEF (%) | 51 [41–56] | 52 [41–57] | 50 [43–55] |
Life-threatening arrhythmias (MVA) | 49/294 (16%) | 54/392 (14%) | 16/139 (17%) |
(Aborted) SCD | 4 (1%) | 10 (3%) | 2 (1%) |
Sustained VT | 36 (12%) | 43 (11%) | 15 (11%) |
Appropriate ICD shock | 15 (5%) | 20 (5%) | 5 (4%) |
HF-related events | 42/257 (16%) | 53/345 (15%) | 15/113 (13%) |
Hospitalization for HF | 37 (14%) | 37 (11%) | 13 (11%) |
HTx or LVAD | 18 (7%) | 28 (8%) | 7 (6%) |
HF death | 18 (7%) | 17 (4%) | 3 (3%) |
Composite of MVA and/or HF events | 67/302 (22%) | 81/404 (20%) | 24/142 (17%) |
Follow-up clinical evaluation (years) | 6 [2–10] | 5 [2–9] | 5 [2–9] |
Table 2.
Patient characteristics of the 882 ACM patients included in this study. Patients were divided into wild type for HRC p.(Ser96Ser) variant carriers (N = 288), heterozygous for HRC (p.(Ser96Ala), N = 413), and homozygous carriers p.(Ala96Ala), N = 181). Data are depicted as median (interquartile range) or n/N (%).
Table 2.
Patient characteristics of the 882 ACM patients included in this study. Patients were divided into wild type for HRC p.(Ser96Ser) variant carriers (N = 288), heterozygous for HRC (p.(Ser96Ala), N = 413), and homozygous carriers p.(Ala96Ala), N = 181). Data are depicted as median (interquartile range) or n/N (%).
ACM (N = 882) | Wild Type (TT) (N = 288, 33%) | Heterozygous (TG) (N = 413, 47%) | Homozygous (GG) (N = 181, 20%) |
---|
Age (years) | 47 [33–60] | 47 [33–57] | 48 [33–58] |
Male sex | 144/288 (50%) | 196/413 (47%) | 99/181 (55%) |
Index patient | 129/287 (45%) | 195/403 (48%) | 87/184 (47%) |
Diagnosis | | | |
ACM | 178/200 (89%) | 245/272 (90%) | 108/117 (92%) |
Life-threatening arrhythmias (MVA) | 120/283 (42%) | 186/375 (50%) | 77/174 (44%) |
HF-related events | 19/142 (13%) | 14/171 (8%) | 13/92 (14%) |
HTx or LVAD | 11 (8%) | 15 (9%) | 5 (5%) |
Follow-up clinical evaluation (years) | 8 [4–15] | 8 [4–15] | 10 [3–14] |
Table 3.
Patient characteristics of the 985 DCM patients included in this study. Patients were divided into wild type for HRC p.(Ser96Ser) variant carriers (N = 316), heterozygous for HRC (p.(Ser96Ala), N = 524), and homozygous carriers (p.Ala96Ala, N = 145). Data are depicted as mean ± standard deviation (SD) or n (%).
Table 3.
Patient characteristics of the 985 DCM patients included in this study. Patients were divided into wild type for HRC p.(Ser96Ser) variant carriers (N = 316), heterozygous for HRC (p.(Ser96Ala), N = 524), and homozygous carriers (p.Ala96Ala, N = 145). Data are depicted as mean ± standard deviation (SD) or n (%).
DCM (N = 985) | Wild Type (TT) (N = 316, 32%) | Heterozygous (TG) (N = 524, 53%) | Homozygous (GG) (N = 145, 15%) |
---|
Enrollment age in years | 60.6 ± 7.1 | 60.5 ± 6.8 | 60.5 ± 6.6 |
Male sex | 225 (71%) | 357 (68%) | 114 (79%) |
VT | 62 (20%) | 80 (15%) | 24 (17%) |
(Aborted) SCD | 18 (6%) | 30 (6%) | 8 (6%) |
VT or SCD | 72 (23%) | 91 (17%) | 27 (19%) |
ICD implantation | 90 (28%) | 139 (27%) | 34 (23%) |
Mortality | 72 (23%) | 131 (25%) | 45 (31%) |
Biobank follow-up time in years | 10.2 ± 2.3 | 10.3 ± 2.6 | 9.8 ± 3.2 |
Table 4.
Minor allele frequencies of the HRC polymorphism in general and registry cohorts.
Table 4.
Minor allele frequencies of the HRC polymorphism in general and registry cohorts.
| Frequencies HRC Polymorphism |
---|
General Population | |
gnomAD, European (non-Finish) | 41.7% |
GoNL | 40.5% |
WES, UMC Utrecht | 40.3% |
UK Biobank | 42.2% |
PLN Registry | 40.9% |
ACM Registry | 43.9% |
Dutch | 43.6% |
USA | 43.0% |
Swiss | 50.9% |
DCM Cohort, UK Biobank | |
British | 41.3% |
Table 5.
Logistic regression analyses in PLN p.(Arg14del) carriers were modeled with HRC p.(Ser96Ala) as predictor. Analyses were adjusted for sex, age at enrollment, and principal components.
Table 5.
Logistic regression analyses in PLN p.(Arg14del) carriers were modeled with HRC p.(Ser96Ala) as predictor. Analyses were adjusted for sex, age at enrollment, and principal components.
PLN p.(Arg14del) | Odds Ratio | 95% CI | p-Value |
---|
Life-threatening VA event | 0.792 | 0.584–1.066 | 0.128 |
HF event | 0.858 | 0.615–1.188 | 0.360 |
Composite | 0.842 | 0.649–1.089 | 0.193 |
Table 6.
Logistic regression analysis in ACM patients was modeled with HRC p.(Ser96Ala) as predictor. Analyses were adjusted for sex, age at enrollment, and principal components.
Table 6.
Logistic regression analysis in ACM patients was modeled with HRC p.(Ser96Ala) as predictor. Analyses were adjusted for sex, age at enrollment, and principal components.
ACM | Odds Ratio | 95% CI | p-Value |
---|
Life-threatening VA event | 0.862 | 0.576–1.288 | 0.467 |
Table 7.
Logistic regression analyses in DCM patients were modeled with HRC p.(Ser96Ala) as predictor. Analyses were adjusted for sex, age at enrollment, genotype array, and principal components.
Table 7.
Logistic regression analyses in DCM patients were modeled with HRC p.(Ser96Ala) as predictor. Analyses were adjusted for sex, age at enrollment, genotype array, and principal components.
DCM | Odds Ratio | 95% CI | p-Value |
---|
VT | 0.848 | 0.524–1.171 | 0.210 |
SCD | 1.016 | 0.985–1.046 | 0.941 |
ICDimplantation | 0.887 | 0.652–1.121 | 0.284 |
| Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content. |
© 2023 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https://creativecommons.org/licenses/by/4.0/).