Sarcomeric gene variants among Indians with hypertrophic cardiomyopathy: A scoping review

Hypertrophic cardiomyopathy (HCM) is one of the common inherited cardiomyopathies1. It manifests during adolescence or early adulthood as unexplained left ventricular hypertrophy, without indications of another cardiac or systemic diseases1,2. The HCM phenotype is characterized by diastolic ventricular dysfunction, asymmetric hypertrophy of the interventricular septum (IVS) and arrhythmias2. Currently, HCM is perceived as a disease with relatively good prognosis and a low mortality rate across all ages, when contemporary treatment options are availed3. However, HCM is a major cause of sudden and premature cardiac death in young adults, as well as among athletes4,5.

From the diagnostic point of view, there is extensive functional and familial segregation evidence to indicate that HCM is a typical monogenic disorder that follows an autosomal dominant pattern of inheritance. The major causative factors are the pathogenic genetic mutations that affect the contractile apparatus of the cardiac sarcomere, although occasionally, metabolic aetiologies may also be causative6. Pathogenic mutations in eight definitive genes namely: MYBPC3, TPM1, TNNT2, TNNI3, MYH7, ACTC1, MYL2 and MYL3 have been identified that encode proteins in the cardiac sarcomere, accounting for 40-60 per cent of HCM cases7,8. Interestingly, the genetic evidence derived from the prevalence of these mutations in large population based genotyping studies has led to a revision in the originally known prevalence of HCM.

Bick et al9 reported in 2012 that the burden of rare sarcomeric gene variants in the Framingham and Jackson Heart Study cohorts was ~1 in 200 in the general population. A recent study by Maron et al10 in 2022, reiterated that the prevalence of HCM ranged from 1:200-1:500 (0.2-0.5%) in the general population, and suggested that HCM was under recognized since only a fraction (10-20%) of the cases were clinically diagnosed. Based on the above reported range of prevalence the authors made a crude estimate for India, with currently the largest population of 1.43 billion, would be carrying a disease burden that ranged from 2.86 to 7.15 million HCM-affected individuals. The high prevalence is indicative of the benign natural history of the HCM with little impact on reproductive fitness11. Further, the prevalence of HCM is expected to be higher among older patients since the diagnostic criteria of HCM such as cardiac hypertrophy indicated by IVS thickness of ≥15 mm is an age-dependent phenotype12.

Variations in the allele frequencies of mutations in sarcomeric genes across global populations suggest the need for population-specific cardiomyopathy mutation panels for early detection13,14. Conversely, it is also argued that similar results and health equity for ethnically diverse populations may not be attainable where the reference population data used in variant interpretation are limited15. Disparities observed in the diagnostic efficiency among racio-ethnic groups point to racio-ethnic disparities in the published literature. This translates to the differences in variant classification that eventually influences the repertoire of mutations that get selected into a genetic testing panel for a particular disease16.

A consolidate report on the prevalence and type of mutations reported from Indian patients of south Asian ancestry has not been compiled to-date. There is a lack of data based on the next-generation sequencing technology, which could otherwise provide comprehensive gene coverage for HCM that is typically caused by a single heterozygous mutation in one of many definitive genes. Similarly, population-specific genome-wide association studies (GWAS) or biobank investigations done among Indian patients are also lacking. Hence, this scoping review was carried out on the literature published on Indian patients who were clinically diagnosed with HCM and carried mutations exclusively in the sarcomeric genes. The specific aims were to catalogue all the reported genetic variants and annotate the variants based on pathogenicity scores, disease propensity scores and minor allele frequency (MAF) in south Asian (SAS) and other global populations. In particular, the pathogenic or likely pathogenic variants (P/LP) were prioritized and further assessed for reported association with outcomes such as sudden cardiac death (SCD). Since severe phenotypes have been observed in carriers having more than one pathogenic mutation, the literature was searched for reports on compound and digenic or double heterozygotes. The compiled genetic information was organized into a knowledgebase for reference by clinicians and scientists involved in genetic research on HCM.

Material & Methods

Search strategy and review of literature: A scoping review of the literature was performed systematically according to a previously described five-stage methodological framework and the Preferred Reporting Items for Systematic reviews and Meta-analyses extension for Scoping Reviews (PRISMA-ScR) Checklist17,18. The primary objective of this scoping review was to identify and summarize the P/LP genetic variants reported in Indian HCM patients and to delineate those associated with SCD. Articles published in the English language on genetic screening of mutations exclusively present in the cardiac sarcomeric genes, conducted in Indian patients of SAS ancestry who were clinically diagnosed for HCM, were eligible for inclusion in the scoping review. An electronic literature search was conducted from the time of inception to August 2021 using indexed databases such as Medline, Scopus, Web of Science core collection and from Google Scholar, using key words such as “Gene”, “Genetic”, “Polymorphism”, “Mutation”, “Variation”, “Deletion”, “Sequencing”, “Cardiomyopathy”, “Cardiomyopathies”, “India”, “Indian”, “South Asia” and “South Asian”, queried either in the title or abstract of the articles. The protocol of the study was registered with Open Science Framework (https://osf.io/53gde/).

Studies that met the following criteria were included in the analysis:

• (i) Full text articles that had reported results of genetic testing for sarcomeric gene mutations in extended familial cases as well as unrelated sporadic HCM cases.

• (ii) Studies reporting the genetic component including sarcomeric gene mutations residing in 11 main genes implicated in HCM pathology and expressed in the heart muscle (OMIM phenotypic series, 192600) as specified in the European Society of Cardiology guidelines19. These included the MYBPC3, TPM1, TNNT2, TNNI3, MYH7, MYL2, MYL3, ACTC1, CSRP3, TTN and PLN genes.

• (iii) The population component included studies among the Indian population of SAS ancestry. The term ‘south Asia ancestry’ implied ancestral origin in a country within south Asia, where the residing population was located in an area containing the modern nations of India, Nepal, Bangladesh, Bhutan, Pakistan, Sri Lanka and the Maldives20.

• (iv) The disease component included patients diagnosed with HCM, without other metabolic or systemic diseases.

Studies were excluded if, the patients were diagnosed with other types of cardiomyopathies such as arrhythmogenic right ventricular cardiomyopathy (ARVC) or left ventricular non-compaction (LVNC), dilated cardiomyopathy (DCM); studies on animal/in vitro models; studies restricted to computational models or in silico analysis; studies that analysed mitochondrial mutations or mutations in non-sarcomeric genes as implicated in phenocopies of HCM such as Fabry disease or PRKAG2 cardiomyopathy, and reviews and meta-analysis.

Data charting process and data items: Data items from eligible articles were manually charted by three reviewers (LK, MM and VY) who screened the combined results, removed duplicates and evaluated the titles, abstracts and full texts. Missing studies were identified by screening the citations of full text and review articles. The abstracted data from eligible articles included the author details, year of publication and PMID/digital object identifier, genotyping method, number of cases/families screened, name/symbol of genes studied, Human Genome Variation Society (HGVS) nomenclature of genetic variants and chromosome locations (GRCh38), as shown in Supplementary Tables

Additional data on other characteristics such as variant type and the functional consequence of the variants were retrieved from the National Center for Biotechnology Information (NCBI) single nucleotide polymorphism database (dbSNP). The HGVS nomenclature of the reported genetic variants was appended with their corresponding reference SNP cluster ID (rsID) from the NCBI dbSNP, to enable cross-referencing to reports from other global populations. To retrieve the variants’ rsIDs, the flanking sequences from the published sequence electropherograms were mapped to the GRCh38 coordinates using BLAST-Like Alignment Tool (BLAT) for human (https://genome.ucsc.edu/cgi-bin/hgBlat).

Variant annotation and prioritization: To identify the P/LP variants, we queried the variants reported from Indian studies against the VarSome genome archive, which follows the America College of Medical Genetics (ACMG) guidelines for variant classification21. Ensembl Variant Effect Predictor (VEP) was used as the annotation tool for retrieving combined annotation-dependent depletion (CADD-Phred scores to indicate in silico predictions of pathogenicity, and to compile minor allele frequencies from the gnomAD-exomes and 1000 Genomes for SAS population22,23. CADD-Phred scores with cut-off >20 indicated ‘likely deleterious’ variants. The Rare Exome Variant Ensemble Learner (REVEL) score was used to prioritize missense variants, where a cut-off ≥0.5 indicated ‘likely disease causing’ variants24. The Genomic Evolutionary Rate Profiling (GERP) scores were used to identify constraint sites, where the evolutionary conservation of each variant was denoted by positive constraint scores25.

The VarSite disease propensity scores were used for predicting the range of effect of P/LP variants on protein 3D structures and categorized as very high (>1.5), high (>1) or low (<1) propensity26. The P/LP variants reported in Indian studies as associated with SCD were queried against global reports in ClinVar database to identify in vivo or in vitro or functional studies that included transgenic mouse models and familial segregation. We also searched for studies that reported compound or digenic/double heterozygotes amongst HCM patients. Compound heterozygotes were defined as patients who were the carriers of two (biallelic) heterozygous mutations in the same HCM associated gene. Digenic or double heterozygotes were defined as patients who were the carriers of two distinct heterozygous mutations in two different HCM associated genes.

Synthesis of results: The characteristics of the reported genetic variants were expressed as descriptive statistics using the statistical software STATA, version 16.0, (Stata Corp., TX, USA). The data summarized included the range of values for pathogenicity, conservation and disease propensity scores for the functional consequences, clinical significance and variant type within each sarcomeric gene.

Results

The number of studies screened at each stage of the search narrowed down to the final number of eligible articles identified for the scoping review. A total of 756 articles were retrieved from the literature survey (Medline, Scopus, Web of Science core collection and Google Scholar). After removing duplicates (n=308), the titles and abstracts were screened for 448 articles. Two hundred and twenty seven articles were excluded as their outcomes were not related to genetic screening of sarcomeric genes in HCM. During full text screening (n=171), 28 articles were identified that reported genetic screening from the year 2000 to 2020 in Indian patients consisting of sporadic or familial HCM cases27-54. The abstracted genetic information from the 28 articles, updated with reference sequence ID (rsIDs) was compiled into freely accessible HCMvar database, available at https://hcmvar.heartfailure.org.in.

Of the 28 articles, a total of 19 full text articles were selected that exclusively reported P/LP variants in sarcomeric genes, and were linked to SCD. A total of 54 P/LP variants spanning five sarcomeric genes were reported from the 19 studies, namely MYBPC3, MYH7, TNNT2, TNNI3 and TPM1. Of these, 45 different variants were identified and comprised of 41 single nucleotide variants (SNVs) (91%) and four (9%) insertion-deletion variants (indels), as shown in Table I. The maximum number of P/LP variants was reported from the MYH7, which was the most frequently studied gene (n=21; 20 SNVs, 1 indel), followed by the MYBPC3 gene (n=17; 14 SNVs, 3 indels). Representative gene and protein domains of MYH7 and MYBPC3 with the P/LP variants are shown in Figure.

Close

Figure

P/LP variants reported from Indian studies in, (A) MYBPC327,30,34,35,42,52 and (B) MYH727,28,35-38,40,45,46,53genes.

Export to PPT

The summarized pathogenicity scores and functional consequences are shown in Tables II and III. Forty variants had CADD-Phred scores of >20 classified as likely deleterious, and 27 variants had REVEL scores of >0.5 classified as likely disease causing. Fifteen out of 31 missense mutations had scores corresponding to high disease propensity. The very low minor allele frequencies of the P/LP variants in the SAS population compared to other global populations indicated that these variants were targeted by purifying natural selection. Single and digenic P/LP variants associated with reported SCD phenotype in Indian patients were compiled gene wise, along with their reports on functional studies and familial segregation.

Single variants associated with SCD phenotype: The scoping review of literature revealed several instances of P/LP variants reported in association with SCD in Indian patients in the following genes, along with evidence from functional studies and familial segregation reported from other global populations.

MYBPC3 gene: Four studies reported mutations associated with SCD in MYBPC3 gene42,47,48,52. Tanjore et al48 identified a pathogenic frameshift variant MYBPC3 p.R589Qfs*16 in a 36 yr old female proband. The proband was diagnosed with non-obstructive HCM, having an IVS thickening of 21 mm. The proband’s mother was symptomatic, with a family history of SCD reported in a maternal uncle. Waldmuller et al52 identified a pathogenic frameshift 2 basepair (bp) deletion MYBPC3 p.P453fs (rs727503203) in two families. However, the manifested clinical phenotypes were vastly different in both the families. In the first family (indicated as HH), mild symptoms and incomplete penetrance were observed with no cardiac related death. Out of five p.P453fs mutation carriers in this family, one member had IVS thickness of 24 mm, while the other four members showed late onset with mild symptoms. Contrarily, cardiac death was reported in four members of the second family (indicated as NP), three of them were SCD at 45, 52 and 55 yr of age, and were asymptomatic carriers of the p.P453fs mutation. In family NP, two symptomatic carriers showed unusually high IVS thickness values (26 and 44 mm) and LVOT gradient (45 and 30 mm), respectively. In another study, Bashyam et al42 identified a stop gained mutation MYBPC3 p.C1124* (rs727504289) in a three year old male child. The proband’s mother and an elder sibling were both asymptomatic, whereas the maternal grandmother suffered SCD. The p.C1124X variant was also associated with HCM in a cohort from Mayo clinic55. A 25 bp deletion (rs36212066) in intron 32 of the MYBPC3 gene has been cited as a population specific variant associated with HCM among south Asians, and has been reported in eight studies27,29,30,35,42,47,48,52. Dhandapany et al47 reported that at least 20 of the HCM probands who carried the 25 bp deletion variant had family members who had died of SCD, and suggested a possible association with ventricular tachyarrhythmias and increased risk for SCD.

MYH7 gene: Five studies reported mutations associated with SCD in MYH7 gene45,49,51-53. Sakthivel et al53 identified a likely pathogenic missense variant MYH7 p.R712L (rs1224554825) where six out of nine members in a three-generation familial HCM case were carriers. The segregation pattern of the mutation revealed an incomplete and age-dependent penetrance. Four carriers were symptomatic, one was asymptomatic, one was unassessed and one case of SCD was reported at 40 years of age. Waldmuller et al52 identified a likely pathogenic 3 bp inframe deletion MYH7 p.E927del (rs1381717637) in a family (indicated as MM) that was associated with complete penetrance. All five family members who were mutation carriers were reported to be affected with severe symptoms and four carriers showed an early age-of-onset (< 20 yr). Two members showed high IVS thickness values (25 and 27 mm) and two cases of SCD were reported at the age of 52 and 20 years.

Tanjore et al51 detected a homozygous MYH7 p.R870H (rs36211715) mutation in an extended SAS Indian family with consanguinity in three generations, where 18 out of 47 members were carriers. The likely pathogenic missense variant was observed to segregate with the disease in 11 affected family members with a penetrance of 59 per cent that was age dependent. Proband 3.21/patient 6, who was born of a consanguineous marriage between affected parents, had a pacemaker implant and had died at 36 yr of age, indicating a poor prognosis for homozygous mutation carriers. All affected members were symptomatic, although no correlation with interventricular septal thickness was observed among mutation carriers. The p.R870H mutation in the heterozygous state was associated with mild hypertrophy, relatively better prognosis and longer life span, while the same mutation in the homozygous state was associated with an increased risk of SCD at an early age in the absence of severe hypertrophy. A study from the Oxford Medical Genetics laboratory had also reported the p.R870H among HCM patients56.

Bashyam et al49 screened a multi-generational family of 29 family members and detected 14 members who were carriers of the missense variant MYH7 p.R870H (rs36211715). The proband who was a homozygous carrier died of SCD, whereas other members manifested with asymmetrical septal hypertrophy, with significant variation in IVS reported amongst heterozygous mutation carriers. Almost all affected members exhibited clinical symptoms such as dyspnoea, angina, syncope and presyncope characteristics of HCM. Rai et al45 reported two P/LP variants in HCM probands associated with SCD. A likely pathogenic missense variant MYH7 p.M515T (rs863224900) was identified in the first proband (H10) diagnosed with asymmetrical septal hypertrophy, having IVS thickness of 19 mm and who died of SCD at the age of 20 yr48. A pathogenic missense variant MYH7 p.G716R (rs121913638) was reported in a second proband (H91), diagnosed with severe asymmetrical septal hypertrophy, having IVS thickness of 31 mm, and died of SCD at the age of 21 yr. Since both parents of the proband were non-carriers, the de novo nature of this mutation was confirmed by a paternity test using short tandem-repeat analysis. Functional studies have reported that the mutant p.G716R variant weakened affinity with actin which in turn reduced the force generated during muscle contraction57. Multiple studies have reported the association of the p.G716R variant with HCM with strong evidence of segregation58-61. The p.G716R mutation was also identified as a de novo variant in multiple unrelated patients diagnosed with HCM62.

TNNI3 gene: Two studies reported mutations associated with SCD in TNNI3 gene40,46. Rai et al46, reported two familial cases of TNNI3 p.R192H (rs104894729) pathogenic missense mutation carriers with early age-of-onset at seven and 10 years of age. The proband at the age of 14 had two syncopal attacks, was diagnosed with asymmetrical left ventricular hypertrophy having maximal IVS thickness of 17 mm, and died of SCD at 16 years of age. The proband’s sister who was reported to have had four syncopal episodes, was diagnosed to have restrictive cardiomyopathy (RCM) with normal LV wall thickness and also died of SCD at 16 years of age. Functional studies demonstrated that the mutant p.R192H variant increased Ca2+ sensitivity compared to wild-type allele, and enhanced the binding affinity of the thin filament63-65. Transgenic mouse models of this variant were also shown to develop a restrictive cardiomyopathy phenotype66,67. Multiple studies have reported the association of the p.R192H variant with early onset RCM45,68-70. Two other likely disease-causing variants at this position (p.Arg192Cys and p.Arg192Leu) have been identified in individuals with RCM and HCM71,72.

Rani et al40 observed two heterozygous missense mutations in HCM patients who had a family history of SCD. The mutations involved arginine to glutamine (R to Q) substitutions at two amino acid positions (98 and 162) in the TNNI3 gene. The exon 6, TNNI3 p.R98Q (rs747522089) likely pathogenic missense mutation was observed in a 28 year old male proband. The proband was diagnosed with severe hypertrophic obstructive cardiomyopathy (HOCM), and reported an IVS thickness of 25 mm, with a family history of three cases of SCD reported for the probands’ mother, brother and grandfather. His five year old son who was an obligate carrier of this mutation was asymptomatic at the time of reporting. The second likely pathogenic TNNI3 p.R162Q (rs397516354) mutation was identified in a proband with severe asymmetric septal hypertrophy with IVS thickness of 29 mm. This mutation was also present in nine other individuals in the family with IVS thickness >25 mm, with a reported family history of SCD. In vitro functional studies using HEK293 cells showed that the p.R162Q variants resulted in a 50 per cent reduction of the troponin I and troponin C interaction73. Multiple studies had also reported the association of the p.R162Q variant with HCM along with familial segregation68,73-77.

TNNT2 gene: Biswas et al33 reported a mutation associated with SCD in TNNT2 gene in a case of familial HCM, where the proband and his brother were found to be carriers of the TNNT2 p.R92W (rs397516456) missense mutation. The proband had left ventricular hypertrophy with atrial fibrillation, was categorized under NYHA class III with severe symptoms and was implanted with ICD. His affected brother who was also mutation positive was categorized under NYHA class II. The proband’s affected father died of SCD, and the daughter of the proband who was a mutation positive, was operated for the atrial septal defect at five years of age. Overall, this mutation was associated with severe symptoms, moderate hypertrophy and high penetrance. Functional studies had shown that p.R92W disrupts tropomyosin binding, increased calcium sensitivity of the cardiac thin filament and impairs adult cardiac myocyte contractility78-80. Transgenic mice carrying the variant showed diastolic dysfunction and altered calcium kinetics in isolated transgenic myocytes, with severe energetic abnormalities81. Multiple studies had also reported the association of the p.R92W variant with HCM with strong familial segregation82-87.

Digenic variants associated with HCM phenotype: Searching the literature for compound and digenic or double heterozygotes among Indian patients with HCM divulged one report by Rani et al39 on P/LP digenic heterozygotes. Initially, they had identified a likely pathogenic missense variant TPM1 p.S215L (rs199476316) in two unrelated patients and their family members with asymmetrical septal HOCM. However, they observed a phenotypic variability that ranged from mild-to-severe symptoms; therefore, a second round of genetic screening in other sarcomeric genes was done in four generation pedigrees of these two patients. This revealed two additional likely pathogenic variants in the MYH7 gene.

From the first family, the proband aged 44 yr, (age-of-onset 16 yr) in addition to the TPM1 p.S215L heterozygous mutation also harboured MYH7 p.D896N (rs606231340) homozygous and MYH7 p.E525K (rs606231324) heterozygous mutations. The proband was diagnosed with asymmetric septal HOCM and categorized as NYHA class IV with a poor prognosis. Three family members aged 15, 17 and 39 yr, who were positive for these digenic mutations also presented a severe phenotype, that included episodes of syncope, dyspnoea, palpitations, IVS thickness ≥30 mm, left ventricular outflow tract (LVOT) gradient, severe mitral and trivial tricuspid regurgitation and significant LVOT obstruction. A family history of two SCDs was reported, at 35 and 19 yr of age. The MYH7 p.E525K mutation was also identified as a de novo variant in at least two unrelated individuals with infantile-onset DCM88. However, Rani et al36, reported the TPM1 p.S215L in a second family, where the proband aged 65 yr, (age-of-onset 43 yr) along with other family members harboured a single heterozygous TPM1 p.S215L missense variant but did not carry the MYH7 p.D896N or p.E525K mutations. They observed that the members of the second family presented either mild symptoms or were asymptomatic, with no family history of SCD. Experimental studies had shown that the TPM1 p.S215L variant altered TPM1 protein hydrophobicity and increased calcium sensitivity thereby affecting the contractile properties of the tropomyosin protein, and reduced its affinity for actin89,90. Multiple studies had also reported the association of the p.S215L variant with HCM with moderate segregation74,91,92.

Functional evidences and familial segregation of P/LP variants associated with SCD: Five P/LP variants had evidence from functional studies suggestive of their mechanistic role in HCM pathology. The TPM1 p.S215L, TNNT2 p.R92W and TNNI3 p. R192H missense variants were reported to increase Ca+ sensitivity, whereas the TNNI3 p.R162Q and MYH7 p.G716R missense variants were reported to decrease protein-protein interactions and binding affinity for other sarcomeric proteins such as actin74,78,81-86. Seven variants (TNNT2 p.R92W, TPM1 p.S215L, TNNI3 p.R162Q, MYH7 p.G716R, MYH7 p.R870H and MYH7 p.R712L) associated with SCD in Indian patients were reported to show familial segregation in other familial cases from other populations, along with reports in multiple unrelated patients with HCM68,73-75,78,81-86,88,91-94.

Discussion

Through this scoping review, we catalogued all the P/LP mutations in sarcomeric genes reported to-date in the Indian HCM patients of SAS ancestry. The HGSV nomenclature was appended with variant rsIDs, on which data of each variants’ clinical significance, functional consequence and pathogenicity scores were annotated. The P/LP variants consisted of missense, nonsense, frameshift or splice site altering variants, as well as insertion/deletions in coding regions, which were further assessed for reported correlation with clinical phenotypes like SCD. Sixteen singletons (private) were identified, one de novo and one digenic mutation associated with SCD among Indian patients. Most familial cases were reported to harbour single (private) mutations, whereas digenic or de novo variants were rare.

Notably, single homozygous mutations in MYH7 gene (p.R870H and p.D896N) were reported to be associated with increased risk of SCD and severe phenotypes compared to single heterozygous mutations36,49,51. The Sarcomeric Human Cardiomyopathy Registry (SHaRe) had also gathered data implicating P/LP mutations in sarcomeric genes as predictors of adverse prognosis; associated with a 2 fold increased risk of a combined endpoint (that included atrial fibrillation, arrhythmias, heart failure and all-cause death); a 6 fold increased risk of severe LV systolic dysfunction; and a 4 fold increased likelihood for a heart transplantation or mechanical cardiac support93. Evidence from functional studies on in vitro and transgenic mice models indicated that these variants could alter inter-domain binding affinity and calcium sensitivity leading to sarcomere disarray, reduced force generation and hypertrophy.

We could identify one report on digenic mutations associated with HCM, where three P/LP mutations located in two different genes were discovered in one multi-generation family36. In this family the carriers of the TPM1 p.S215L (rs199476316) heterozygous, MYH7 p.D896N (rs606231340) homozygous and MYH7 p.E525K (rs606231324) heterozygous mutations, were associated with disease severity, penetrance and SCD. Among patients with HCM, the presence of double or compound sarcomere mutations was reportedly associated with an enhanced risk of SCD, even in the absence of massive left-ventricular hypertrophy or an indication of a positive family history of SCD, which are regarded as conventional risk factors94. It is suggested that multiple gene mutation in HCM families can result in more sever clinical phenotype because of a double dose effect73. Ingles et al73 reported that 43 per cent of affected individuals from multiple mutation families experienced an SCD event, compared to 18 per cent from single mutation families, along with a significant increase in septal wall thickness in patients with compound mutations (mean±SD, 30.7±3.1 vs. 24.4±7.4 mm; P<0.05). Moreover, double heterozygotes have special implication in the process of genetic counselling. It has been reported that amongst double heterozygotes, the risk of inheriting at least one disease causing mutation increases to 75 per cent, as opposed to the risk of 50 per cent for single heterozygotes typically seen in autosomal dominant disorders like HCM73.

Although there are several experimental evidence that imply a pathogenic role for the MYBPC3 25 bp deletion, its association with HCM phenotype is not unequivocal47,52,95,96. As per ACMG guidelines, the MYBPC3 25 bp deletion has been classified as a benign variant21. This classification is based on the following clinical evidence rules: (i) the genome aggregation database (gnomAD) exomes SAS allele frequency for the deletion variant is 0.0321, which is greater than the threshold frequency of 0.00529 derived from the 2782 clinically reported variants in MYBPC3 gene; (ii) the deletion variant is observed in healthy adults, with gnomAD genomes homozygous allele count at 3, which is >2 for autosomal dominant or recessive gene disorders and; (iii) the position of the deletion variant is not conserved (phyloP = 2.3). A recent meta-analysis also implied the possibility that the selection method of controls and linkage disequilibrium with other pathogenic variants could be plausible factors that may conceal the benign nature of the MYBPC3 25 bp deletion variant14.

Validated sarcomeric genes: The genetic aetiology of HCM is dominated by variations in eight sarcomeric genes. The validation of these eight core sarcomeric genes (MYH7, MYBPC3, TNNT2, TPM1, MYL2, MYL3, TNNI3 and ACTC1) was reported by Walsh et al56. The reassessment of variants previously reported as pathogenic was based on detailed HCM family linkage studies (Maximum LOD score >3), supported by hundreds of functional and segregation data and by the calculation of significant excess of rare variation in each gene in HCM case cohorts over ExAC reference population. Nearly 40 per cent of all HCM cases, and especially, majority of large familial HCM cases are caused by the two most common causal genes encoding β-myosin heavy chain (MYH7) and myosin-binding protein C (MYBPC3)97.

Majority of variants identified in sarcomeric genes are single nucleotide substitutions that translate to a mutant protein, which when incorporated into the sarcomere exerts a poison peptide effect98. Such variants, known as dominant-negative mutations when incorporated into wild-type protein subunits have been speculated to reduce the activity of the co-expressed wild-type protein, leading to increased stabilization of active states, rigidification of protein complexes and increased protein stability99. For instance, studies supporting the dominant negative effect of MYH7 missense mutations showed that although the mutated protein was incorporated into the sarcomere without disrupting the packing of myofibrils, the mutated protein disturbed the normal mechanism of force generation and sarcomere function100.

The MYBPC3 gene is an exception, in which majority of the mutations are either deletions or insertions that cause a frameshift in the codon sequence, which in turn translates to a truncated protein with reduced or no function98. Expression analysis studies using immunoblotting and RT-PCR analysis in myectomy samples from HCM patients showed that in tissues containing either truncation or missense MYBPC3 mutations had a lower level of full-length protein, which supported the mechanism of haploinsufficiency to cause the disease101. Moreover, truncated MyBP-C peptides were not detected in whole muscle homogenates, which argued against any dominant negative effect.

The principal mechanism that leads to enhanced contractility and impaired relaxation (diastolic dysfunction) in HCM involves pathogenic variants that affect mRNA, protein, sarcomere (disarray), myocyte (hypertrophy), myocardium (altered transcriptomics and expression of trophic and mitotic factors), histology (fibrosis) and clinical phenotypes (cardiac arrhythmias and heart failure), in a series of sequential events97. Mutations in the MYH7 and MYBPC3 genes that act either through the dominant-negative effect or haploinsufficiency when enriched in the myosin interacting-heads motif (IHM), can cause a conformational shift in the myosin heads from the super-relaxed state (SRX) to a strong bound state of myosin heads with the actin molecules. This reduction in the proportion of SRX state with a low ATPase activity and the subsequent increase in the proportion of strong bound state with high ATPase activity, results in increased energy consumption, hypercontractility and impaired relaxation, leading to HCM102.

The first genetic locus for familial HCM was mapped to chromosome 14q1 near microsatellite D14S26 (LOD score=9.3) using genetic linkage analysis in a large HCM kindred103. Subsequent sequence analysis and statistical analyses of the co-segregation of genotype and phenotype revealed an Arg403Glu missense variant in exon 13 of MYH7 gene in all affected individuals104. Henceforth, more than 2000 variants rare pathogenic variants have been identified in HCM families and individual patients105.

Certain characteristics have been observed for sarcomeric mutations. Most of these variants have been described as private mutations, implying that they are unique to one or a few families. Similarly, a high degree of genetic heterogeneity in these mutations has been observed in HCM which has been attributed to several factors. SCD in the young and pathogenic de novo mutations with early-onset comorbidities have been suggested to promote the gradual loss of HCM mutations from populations and evoke negative selection. Since mutations are random events in the genome, larger-sized sequences of sarcomeric genes (~23 kb) are prone to a higher chance of acquiring more mutations106. Several population-specific founder mutations have also been reported in sarcomeric genes along with haplotype sharing among mutation carriers.

Metabolic genes: Apart from the sarcomeric gene mutations, a small proportion of the genetic aetiology of HCM is accounted for by mutations in metabolic genes. Nearly two per cent of patients often misdiagnosed as having HCM may have metabolic disorders or mitochondrial cytopathies that produce cardiac phenocopies of HCM102. Walsh et al107, found five genes that were robustly associated with metabolic conditions that mimic and initially present as HCM (LAMP2, GAA, GLA, PRKAG2 and TTR) . These genes are known to activate different pathogenic mechanisms to elicit hypertrophy as well as additional clinical phenotypes that do not occur in HCM104.

A considerable proportion of individuals with HCM have also been reported to have comorbid diabetes mellitus (~10%)108. A recent study by Dhandapany et al109 identified novel ultra-rare mutations in the adiponectin receptor 1 (ADIPOR1) gene that were found to increase the risk for HCM in patients who had coexisting diabetes mellitus. They provided supporting experimental evidence that the p38/mammalian target of rapamycin (mTOR) and/or extracellular signal-regulated kinase pathways, implicated in dysregulated glucose and lipid metabolism, could promote cardiac hypertrophy in the presence ADIPOR1 variants.

Childhood cardiomyopathies: HCM variants rarely cause overt manifestations early in childhood, yet they carry a substantial burden of disease due to the high risk of morbidity and mortality102,110. Majority of childhood cardiomyopathies are caused by infections, toxin exposure, tachyarrhythmias, neuromuscular disorders, mitochondrial defects, metabolic disorders involving glycogen and fatty acid metabolism, lysosomal storage defects and syndromic conditions among infants of mothers with diabetes mellitus111. According to incidence rates of cardiomyopathy subtypes in large population-based studies, DCM accounted for majority of the cases among children and adolescents (~50%)111. HCM accounted for 35 to 50 per cent of cases, and this incidence was thrice higher in children who were less than one year of age. Among DCM cases, 10 to 25 per cent of cases were due to acute myocarditis111. Restrictive cardiomyopathies (RCMs) and left ventricular myocardial non-compaction (LVNC) accounted for <5 and ~5 per cent of cases, respectively 111.

Non-sarcomeric genes: Despite high thoroughput sequencing technologies and the now widely available population genomic data, the genetic causes in the other half of the individuals with HCM who lack sarcomere gene mutations, still remain elusive. In an important study that re-evaluated non-sarcomeric genes, Walsh et al107, sequenced 31 non-sarcomeric HCM genes, in a large prospective HCM cohort (n=804). They used the exome aggregation consortium (ExAC, n=60706) and a large cohort of clinically genotyped HCM patients (n=6179), as the reference population. Interestingly, they found strong evidence for HCM causation for only three genes, namely CSRP3 (Z-disc), FHL1 (desmosome) and PLN (calcium signalling). No significant excess of rare (MAF <1:10000 in ExAC) protein-altering variants was observed in HCM cases over controls among the tested genes. Their study on non-sarcomeric mutations, described as the ‘other side of the coin’ had rigorously re-evaluated many non-sarcomeric gene associations and concluded that majority of non-sarcomeric genes that were previously implicated in HCM was not associated with the HCM phenotype112.

Modifier genes: Mutations in modifier genes along with environmental factors are postulated to account for interindividual variability in subjects sharing identical causal mutations, and a few individual modifier genes for HCM have been reported105,113,114. Based on linkage and segregation analysis in a very large family with HCM (100 members), integrin alpha 8 (ITGA8) and caspase-associated recruitment domain-containing protein (CARP) were identified as biologically plausible candidate genes, which were implicated in cardiac fibrosis and apoptosis113. A variation of the number of tandem repeats (VNTR) in the myocyte enhancer factor 2C (MEF2C) gene, a transcriptional regulator of several cardiac genes, was implicated as a modifier gene since it was associated with modulating left ventricular outflow tract obstruction (LVOTO) in HCM patients, but was not associated with the risk of HCM114. In a genome-wide association study (GWAS) on 363 patients from the UK Biobank, two novel modifier genes, lysine-specific methyltransferase 2c (KMT2C) and par3 family cell polarity regulator, beta (PARD3B) were found to be associated with ICD-10 codes for HCM diagnosis105. A novel association between the myosin-binding protein H (MYBPH) gene and hypertrophy traits was identified in HCM patients carrying the MYH7 p.A797T mutation that suggested that variation in MYBPH can modulate the severity of hypertrophy in HCM76.

Missing causal genes: Finally, coming to the case of missing causal genes, there may be several challenges to detect and to unambiguously assign pathogenicity to these genes. A continuing debate is the deterministic vs. probabilistic approach to understanding the genetic basis of HCM97. The deterministic approach to genetic studies of HCM is based on the conventional understanding that the presence of a single pathogenic variant or mutation is necessary and sufficient to cause HCM. This approach is currently followed for genetic diagnostics in HCM and here pathogenic variants are detected using whole-exome sequencing (WES) or candidate-gene panels. On the other hand, the probabilistic approach considers the HCM phenotype as a complex polygenic trait influenced by multiple pathogenic variants that may exhibit a gradient of effect sizes97. This approach which is research oriented and hypothesis free leaves more room for speculation and discovery. However, it requires mutation detection using whole genome sequencing (WGS), which is by far more time and computation intensive. Even so, it is plausible that mutations in the missing causal genes may have modest effect sizes, exhibit incomplete penetrance, or be de novo, in which case obliterates any opportunity to conduct co-segregation and linkage analysis. The application of genetic testing in cardiomyopathies as standard practice, the development of more accurate long-read WGS approaches, the generation of data from functional studies for reclassification of variants of uncertain significance (VUS), and the integration of cardiomyopathy genetic databases across diverse populations could bring more clarity and to a certain extent resolve the enigma of missing causal genes in HCM.

Limitations: Since this scoping review was retrospective in nature, it is likely that data on age-dependent phenotypes such as IVS thickness that may have manifested with the progression of the disease might have been missed, especially among younger patients who were asymptomatic at the time of reporting. Most studies included in this scoping review were hypothesis-driven replications of previously reported candidate genes for HCM from other populations. In addition, majority of reported variants were detected through exon screening of isolated candidate genes using a laborious technique like polymerase chain reaction based single strand conformation polymorphism. Here, a major drawback of screening and reporting mutations from isolated candidate genes is that the failure to identify a pathogenic mutation in one sarcomeric gene does not rule out the possibility that the patient may be carrying a causal mutation in a different sarcomeric gene. Since, HCM is essentially a monogenic disorder which can affect one of many definitive genes, multi-gene panels or whole-exome sequencing would have efficiently and quickly identified rare single mutations, and especially the digenic mutations that may have been missed while using single-gene screening techniques. The critical appraisal of the individual articles used as sources of evidence was not done, since this was a scoping review conducted to determine the scope and spectrum of sarcomeric gene mutations reported among Indian HCM patients.

Conclusions

The genetic evaluation of HCM and its post-testing phase is complex. This complexity results from allelic heterogeneity that arises from involvement of various genes and different rare mutations within those genes. From this scoping review, we observed that single variants reported from the Indian studies were associated with varying levels of disease severity. Notably, carriers of single homozygous, de novo and digenic mutations were observed to be associated with severe phenotypes compared to carriers of single heterozygous mutations. Exploring the spectrum of reported genetic mutations amongst patients can be useful in creating genetic mutation databases for HCM. The open accessibility of in silico data on variant pathogenicity and the availability of allele frequencies of genetic variants from major global populations that also include south Asia will further facilitate the discovery of population-specific diagnostic and prognostic biomarkers for HCM.