Whole-exome sequencing reveals a likely pathogenic LMNA variant causing hypertrophic cardiomyopathy

Abstract

Objective

We studied the clinical and molecular features of a family with hypertrophic cardiomyopathy (HCM).

Background

A very heterogeneous disease affecting the heart muscle, HCM is mostly caused by variants in the proteins of sarcomeres. The detection of HCM pathogenic variants can affect the handling of patients and their families.

Methods

Whole-exome sequencing (WES) was performed to assess the genetic cause(s) of HCM in a consanguineous Iranian family.

Results

Missense likely pathogenic variant c.1279C>T (p.Arg427Cys) within exon 7 of the LMNA gene (NM_170707) was found. The segregations were confirmed by polymerase chain reaction–based Sanger sequencing.

Conclusions

Variant c.1279C>T (p.Arg427Cys) in the LMNA gene seemed to have been the cause of HCM in the family. A few LMNA gene variants related to HCM phenotypes have been recognized so far. Identifying HCM genetic basis confers significant opportunities to understand how the disease can develop and, by extension, how this progression can be arrested. Our study supports WES effectiveness for first-tier variant screening of HCM in a clinical setting.

Introduction

Hypertrophic cardiomyopathy (HCM) is the most common genetic heart disease, estimated at 1:500 in the general population.^1,2 It is defined by asymmetric left ventricular hypertrophy, diastolic heart failure, cardiac fibrosis, cardiomyocyte disarray, and sudden cardiac death.^3–5 Hypertrophic cardiomyopathy is the primary cause of sudden cardiac death in competitive athletes and young adults.⁶ It is an autosomal dominant Mendelian disease; however, some cases are defined by de novo variants and, less frequently, autosomal recessive inheritance.⁷ Hypertrophic cardiomyopathy is generally considered a genetically heterogeneous disease with variable phenotypic manifestations and incomplete penetrance, posing diagnostic and prognostic challenges. The first chromosomal location was mapped in 1989,⁸ since which time variants in numerous genes causing HCM have been reported. Various genome-wide association investigations have identified many sarcomeric gene variants related to HCM phenotypes in multiple large affected families, including MYH7, MYL3, MYL2, MYBPC3, TNNI3, TNNT2, ACTC1, TTN, and TPM1.^9–15 Other genes such as ACTN2, CSRP3, MYOZ2, TNNC1, NEXN, PLN, and TTR play a role; nevertheless, they do not definitively cause the disease.¹⁶ For pathogenic variants in other genes, evidence is either insufficient or nonexistent.^17–19 Approximately 20% to 30% of patients with HCM have sarcomere gene variants and are considered or assumed to be pathogenic.^20–23 It is, however, worthy of note that sarcomeric genes have been gradually expanded to nonsarcomeric ones.^18,24 Indeed, isolated and sporadic HCM cases with probands who have no family history of HCM and do not carry known HCM variants may account for up to 40% of all patients with this heart disease.²⁵ Consequently, sarcomere dysfunction seems to be an essential component but not the needed initiating event in the pathogenesis of HCM. The etiology of HCM seems multifactorial rather than strictly genetic,^17,21 as families and patients with the same genetic variants are mostly found with a wide range of disease manifestations.^17,26

The LMNA gene (OMIM *150330) encodes the lamin A/C protein, an integral component of the nuclear membrane, that can maintain nuclear stability. It also has been reported to be involved in the structural integrity of the entire cell through interactions with the cytoskeleton, nuclear lamina, and extracellular matrix.^27,28 There is a link between LMNA variants and dilated cardiomyopathy, with LMNA variants causing one-third of heritable patients to have dilated cardiomyopathy.²⁹ Nonetheless, associations between LMNA variants and HCM have also been previously reported, suggesting the crucial effect of LMNA on the development of the disease. An incomplete penetrance pattern is found in families with heterozygous LMNA variants, which implies that several sporadic cases reported in earlier investigations could be a consequence of incomplete penetrance.³⁰

The high level of allelic and locus heterogeneity as a characteristic of HCM can necessitate sequence analysis of the entire coding area of multiple genes, which is a costly and time-consuming process through conventional methods.^31,32 Next-generation sequencing is a high-throughput method by comparison with classical sequencing methods, and not only does it allow a rapid analysis of many genes in a more affordable manner^32–34 but also it provides a unique opportunity for the early detection of at-risk individuals before clinical diagnosis. The identification of the HCM genetic basis creates significant opportunities for differentiating genetic sarcomeric HCM from phenocopies such as athlete’s heart, hypertensive heart disease, and metabolic or storage disorders.¹⁴

Here, we describe a consanguineous Iranian family with HCM of unknown cause. Via whole-exome sequencing (WES) analysis on the proband’s genome, we found missense variant c.1279C>T in LMNA (NM_170707), leading to a cysteine-for-arginine substitution at position 427 (p.Arg427Cys), which segregated with the disease in the family.

Materials and Methods

Ethics Approval and Consent to Participate

This study was performed following the principles of the Declaration of Helsinki and approved by the ethics committee of Rajaie Cardiovascular Medical and Research Center (IR.RHC.REC.1400.098). The subjects gave informed written consent for participation in the study and the publication of this report.

Family Recruitment and Clinical Characteristics

Three generations of an Iranian family were recruited for this study (FIGURE 1A). Eight members of the pedigree, consisting of 2 affected (FIGURE 1A: II-4 and III-1) and 6 healthy individuals (FIGURE 1A: II-1, II-2, I II-3, I II-5, II-6, and III-2) who consented to study participation, were evaluated. The clinical and demographic information of the studied pedigree, including family history and clinical symptoms, was collected from the database at Rajaie Cardiovascular Medical and Research Center.

Echocardiography and cardiovascular magnetic resonance showed no evidence of cardiomyopathy in all the unaffected members. The proband (III-1) was an 8-year-old boy who suffered from heart murmurs and was referred for genetic testing at this age. He also experienced clinical symptoms such as breathlessness, fatigue, and chest tightness following exercise. The proband (III-1) had been diagnosed with extreme cardiac wall thickening and cardiomyopathy. Elevated levels of lactic acid and pyruvate were reported in blood biochemistry evaluation. According to the available reports, the proband (III-1) did not suffer from Pompe disease (MIM #232300) nor did he have a pathogenic variant in the molecular genetic evaluation of the GAA gene.

The proband (FIGURE 1A: III-1) had a healthy brother (FIGURE 1A: III-2, 2 years old). The proband’s father (FIGURE 1A: II-3) and mother (FIGURE 1A: II-4) were clinically unaffected with no symptoms. Evidence of diabetes and high blood cholesterol levels (FIGURE 1A: I-1), as well as dysentery and intestinal infections (FIGURE 1A: II-7), was reported in the family. Other clinical information was unavailable.

Echocardiography

The first diagnostic echocardiography (GE Vivid S60 cardiovascular ultrasound machine, cardiac sector probe 3Sc-RS [1.3-4.5 MHz]) was performed on the proband (III-1) at the age of 4 years and was repeated every 6 to 12 months. Sequential segmental analysis was performed on cardiovascular anatomy in different image orientations using 2D images, color Doppler, continuous-wave Doppler, and M-mode.

Cardiovascular Magnetic Resonance

Cardiac magnetic resonance (1.5T MAGNETOM, Avanto, Siemens Healthcare) was performed on the proband (III-1) at the age of 4 years for diagnosis and 4 years afterward for follow-up. The standard protocol was performed, including steady-state free precession cine imaging in the long- (4-, 2-, and 3-chamber) and short-axis cine views and the right ventricular outflow view. After the administration of 0.15 mol/kg of gadoterate meglumine (gadolinium-DOTA, Dotarem, Guerbet), early and late gadolinium enhancement images of the short-axis stack and the 3 long-axis views were obtained in magnitude and phase-sensitive inversion recovery reconstructions.

WES and Bioinformatics Analysis

WES was performed on the proband’s genomic DNA. Fragmented DNA was captured using an Agilent SureSelect Exome Capture kit (Agilent). Thereafter, paired-end sequencing of Exon-Enriched Libraries was carried out on the Illumina HiSeq 4000 (Macrogen). The alignment of the sequencing reads to the human genome reference (hg19/NCBI) was done using the BWA (v07.17) tool,³⁵ with 98.8% quality and 99% region coverage. Variant calling was performed on the mapping result file (binary alignment map) by applying the GATK (v4.1.4.1) tool. Duplicates were marked and removed with the aid of SAMtools (in the GATK package)³⁶; then, SNP/INDEL calling was done. The annotated variants were prioritized and filtered by considering the minor allele frequency (>0.05) of 1000 Genomes Project and ExAc databases.³⁷ Bioinformatics analysis was performed to evaluate the pathogenicity score of the candidate variants. Prediction software tools, namely MutationTaster, SIFT, PolyPhen-2, PROVEAN, CADD, and GERP++, were applied. The variant predicted as pathogenic in at least 3 tools was considered for subsequent validation and segregation analysis.

Variant Validation

The putative pathogenic variant was verified by polymerase chain reaction and Sanger sequencing. Specific forward (5'CTTGTGATGTTCAGAGCTGGCT3') and reverse (5'TGTGGAGGAGATATACAGGCTCAC3') primers were designed using Primer3 (v.04.0) (http://bioinfo.ut.ee/primer3-0.4.0/). DNA extraction was performed using our in-house method of salting out. The polymerase chain reaction test was carried out on a SimpliAmp Thermal Cycler (Thermo Fisher Scientific) with 1.5 mmol/L of MgCl₂, 10 pmol/L of primers, 200 ng of DNA, 200 mmol/L of dNTP, and 1 U of Taq DNA polymerase (Amplicon). The incubation schedule was 95°C for 5 minutes, followed by 32 amplification cycles (30 seconds at 95°C, 30 seconds at 63°C, and 30 seconds at 72°C), and the product was sequenced on an ABI Sequencer 3500XL PE (Applied Biosystems) in our center.

Homology Modeling and Docking

Lamin A/C and Cellular Signaling

Lamin A/C can bind Smad antagonists, namely MAN1 (inner nuclear membrane protein Man1) and PP2A (protein phosphatase 2A activator). Lamin A/C-PP2A and lamin A/C-MAN1 complexes reduce the Smad pathway through dephosphorylation and sequestering, respectively, and subsequently inhibit transforming growth factor-beta (TGF-β) signaling.³⁸ TGF-β regulates differentiation, proliferation, and apoptosis in many cell types.³⁹ Blocking TGF-β signaling diminishes the proliferation and expression of downstream myocyte targets and, thus, leads to less hypertrophy.⁴⁰ The results of a previous study indicated that lamin variants increased TGF-β activity via the Smad-dependent pathway in mice.⁴¹ Another investigation on mice with HCM showed that TGF-β activated the canonical Smad-dependent pathway.⁴²

Docking of LMNA with MAN1 Proteins

With the aid of the RCSB Protein Data Bank (PDB) (https://www.rcsb.org/), human lamin A/C (PDB: 7CRG) and MAN1 (2CH0) were downloaded. Mutant lamin A/C (R427C) was created with the SWISS-MODEL Homology-Modeling Server (https://swissmodel.expasy.org/).^43–47 The structures of the proteins were corrected with ViewerLite (v.1.5.1). Briefly, polar hydrogens were added, and water molecules and ligands were deleted. Energy minimization was performed using the YASARA Energy Minimization Server (http://www.yasara.org/minimizationserver.htm).⁴⁸ The 3D structures of the compounds were imported as an SCE file into the YASARA View to deliver low-energy structures of the compounds and then saved in PDB file format. The protein-protein docking of the modeled normal lamin A/C and mutant lamin A/C (R427C) with the MAN1 protein (PDB: 2CH0) was performed using the HADDOCK Web Server (https://wenmr.science.uu.nl/haddock2.4/),^49,50 which is very efficient in protein-protein docking. Data visualization was accomplished using PyMOL (v.2.5.2),⁵¹ and the interactions of all the compounds across the interface were checked with the DIMPLOT program, implemented in LigPlus+ (v.2.2.4).⁵²

Results

Echocardiography Findings

The first echocardiography on the proband (III-1) at the age of 4 years in 2016 revealed asymmetric septal hypertrophy with reversed septal curvature (maximum thickness = 18 mm), mild dynamic outflow obstruction, mild mitral regurgitation due to the systolic anterior motion of the mitral valve, and supernormal left ventricular systolic function (ejection fraction = 75%). Additionally, mitral inflow signals revealed abnormal diastolic function, whereas right ventricular function was normal. The last echocardiographic study, in 2020, illustrated a maximum thickness of 28 mm for the interventricular septum. Moreover, there was no significant change in the function of the left and right ventricles and the severity of left ventricular outflow obstruction and mitral regurgitation (FIGURE 2).

Cardiovascular Magnetic Resonance Findings

In 2016, cardiovascular magnetic resonance on the proband (III-1) at the age of 4 years revealed left ventricular asymmetric hypertrophy in the basal to mid-septal wall with a maximum thickness of 18 mm compared with a thickness of 3 mm in the lateral wall. The modality also demonstrated systolic anterior motion of the mitral valve, accelerated flow in the left ventricular outflow tract (Supplementary Videos 1 and 2), supernormal left ventricular systolic function (ejection fraction= 75%), and a right ventricular ejection fraction of 65%. Late gadolinium enhancement images showed no evidence of replacement fibrosis in the myocardium. Based on the clinical and imaging findings, the cardiomyopathy phenotype was defined as HCM.

The subsequent cardiovascular magnetic resonance study, performed 4 years later at follow-up, showed a significant increase in left ventricular wall thickness to a maximum of 28 mm at the mid-septal portion in comparison with the 2016 study (FIGURE 3A and 3B), the systolic anterior motion of the mitral valve, and accelerated flow in the left ventricular outflow tract. Additionally, there was no significant change in left and right ventricular ejection fractions. The late gadolinium enhancement images yielded no evidence of replacement fibrosis (FIGURE 3C and 3D).

Molecular Analysis

The known C>T pathogenic heterozygous variant at nucleotide position c.1279 (transcript name: NM170707) in LMNA exon 7 was identified through WES. This variant resulted in arginine replacement at the 427th amino acid position with cysteine (p.Arg427Cys). This variant was verified in the proband by Sanger sequencing (FIGURE 1A: III-1), whereas it was not found in the proband’s father (FIGURE 1A, II-3) or brother (FIGURE 1A: III-2). Although the mother (FIGURE 1A, II-4) of the proband (III-1) did not have clinical symptoms, she had the same pathogenic variant (LMNA c.1279C>T [p.Arg427Cys]) and the magnetic resonance imaging (MRI) result indicated that she also had HCM (Supplementary File 3). The other pedigree members studied, comprising the healthy individuals (II-1, II-2, II-5, and II-6), lacked the variant (FIGURE 1B). Previous research has shown that amino acid Arg427 is located in the LTD domain of the LMNA protein (FIGURE 1C), and amino acid Arg427 is greatly conserved in several species (FIGURE 1C). According to the American College of Medical Genetics and Genomics 2015,⁵³ c.1279C>T is interpreted as a likely pathogenic variant (ie, criteria: PM1, PM2, PP2, PP3, and PP4). The missense substitution variant was considered disease causing by MutationTaster, SIFT, PolyPhen-2, PROVEAN, CADD (Phred = 27.6), and GERP++ (5.7).

Docking studies were performed between the normal and mutant forms of lamin A/C and MAN1. Among the clusters resulting from the docking experiments, the first one, which had the least root mean square deviation of the structure, was better than the others. The top complex structures for the normal and mutant forms of lamin A/C in interaction with MAN1, reductions in the HADDOCK score, changes in electrostatic energy, and the buried surface area for the interaction between the normal and mutant forms of lamin A/C with MAN1 are depicted in FIGURE 4. The HADDOCK score for the interaction between normal lamin A/C and MAN1 was −69.4 ± 8.1, whereas it was significantly reduced for the interaction between mutant lamin A/C and MAN1 (the HADDOCK score = −80.8 ± 12.5). Electrostatic energy reduction was significant in the interaction between normal lamin A/C and MAN1 (electrostatic energy = −94.8 ± 16.0) and between mutant lamin A/C and MAN1 (electrostatic energy =−251.8 ± 92.1). The buried surface area exhibited a more pronounced change between these 2 conditions (normal lamin A/C and MAN1 = 1452.7 ± 38.2 and mutant lamin A/C and MAN1 = 1265.2 ± 146.8). Changes in the position of MAN1 in interaction with mutant lamin A/C in comparison with normal lamin A/C are shown in FIGURE 5.

Discussion

The genetic heterogeneity of HCM is highly diverse. Because of this marked allelic heterogeneity and the high frequency of novel individual variants,^54,55 limited data are available on genotype-phenotype relationships.^55–57 Although HCM is known as a disease of the sarcomere, variable penetrance in families who have the same genetic variants can challenge the notion of a monogenic origin for HCM and instead imply a multifactorial cause. Several patients and families with HCM are left without a definite molecular genetic diagnosis.⁵⁸ Further, large-scale genome sequencing investigations have offered little or no evidence of a relationship between HCM and several genes already reported as its cause, suggesting a sarcomere-independent mechanism for cardiomyocyte hypertrophy.⁵⁹ Cardiac imaging techniques such as cardiac MRI and echocardiography are used to diagnose HCM, which mostly presents as asymmetrical septal hypertrophy.^5,60 Accordingly, we performed exhaustive clinical and genetic evaluations, including cardiovascular magnetic resonance, echocardiography, and WES, on our proband (III-1) and succeeded in establishing a diagnosis of HCM via these workups. The proband (FIGURE 1A: III-1) had a wall thickness of 28 mm, whereas a clinical diagnosis of HCM is based on a maximum wall thickness of ≥15 mm.

Assessments of genotype-phenotype associations have confirmed that essentially there is compatibility between any wall thickness and the existence of an HCM mutant gene; however, a high prevalence of sudden death is observed in patients with thicker walls.^61,62 According to previous research, genetic testing is suitable for any level of hypertrophy.⁶³

Positioned on chromosome 1q22, LMNA encodes 2 main lamin A/C isoforms as major nuclear protein components in mammals and acts as a meshwork structure. Recently, certain disease phenotypes called “laminopathies” have been associated with variants in LMNA, ranging from accelerated aging diseases to striated muscle diseases such as cardiomyopathy and muscular dystrophy.^64,65 In general, the LMNA gene has exhibited a high level of conservation throughout evolution. In this study, we investigated LMNA protein interactions. Our protein-protein interaction analysis suggested that variant p.Arg427Cys could diminish hydrogen bands and hydrophobic surfaces in lamin A/C and MAN1, thereby lessening the binding affinity of Smad to the complex given that MAN1 is an indirect inactivator of Smad in the TGF-β pathway. Hence, this variant can result in increased TGF-β activity in myocytes, followed by myocyte hypertrophy.

In this study, we detected candidate variant LMNA c.1279C>T (p.Arg427Cys) in a family with HCM using next-generation sequencing. We found that the proband (III-1) carried a heterozygous c.1279C>T (p.Arg427Cys) variant in the LMNA gene associated with HCM. Interestingly, the pathogenic variant LMNA c.1279C>T (p.Arg427Cys) was also identified in the proband’s (III-1) mother (II-4), who showed no clinical symptoms; however, MRI indicated she also had HCM. No family history of HCM was reported in the proband’s (III-1) family, and DNA was not available for further genetic assessment of the remaining family members. Surprisingly, although the majority of missense variants in the LMNA gene are particularly linked to dilated cardiomyopathy, there are few reports of LMNA variants leading to phenotypes consistent with HCM. Furthermore, the relationship between LMNA gene variants and extremely variable clinical manifestations is complicated and not sufficiently elucidated. The 2 novel LMNA variants, namely c.1772G>T (p.Cys591Phe) and c.1930C>T (p.Arg 644Cys) located at exon 11, show pleiotropic effects, with variant carriers causing several cardiomyopathies such as HCM and dilated cardiomyopathy.^66,67 These findings imply external factors as possible pathogenic co-drivers for the development of the disease.⁵⁹ Caux et al⁶⁸ reported a heterozygous variant in exon 2: a CGG-for-CTG transversion at LMNA codon 133 (R133L), resulting in an arginine-for-leucine substitution in a patient with insulin-resistant diabetes, HCM with aortic valve involvement, and generalized lipoatrophy. Francisco et al⁶⁹ described a 64-year-old woman with HCM harboring the variant c.1718C>T (p.Ser573Leu) in LMNA exon 11. The patient suffered from severe symptomatic ventricular hypertrophy and left ventricular outflow tract obstruction with severe dyslipidemia, diabetes, and obesity.

The clinical utility of genetic testing is limited by challenges in identifying variants of unknown significance, the absence of consistent genotype-phenotype correlations, and inadequate knowledge regarding all genes involved in HCM.²³ On the other hand, genetic testing for HCM provides essential clinical data for family screening and disease outcomes, albeit the yield is variable.⁷⁰ In a retrospective assessment of 285 cases with the pathogenic variants of sarcomere protein genes causing HCM with no diagnostic criteria for HCM, nearly 50% developed HCM during 15 years of follow-up.⁷¹ Therefore, it is essential to identify cases with the pathogenic variants of HCM and provide proper follow-up testing, even for asymptomatic cases.

The existing literature contains a dearth of data on the spectrum and clinical characteristics of the causal genetic variants of HCM in Iranians. To the best of our knowledge, this investigation is the first report regarding LMNA pathogenesis as causative of HCM in Iranians. We recommend WES as the most cost-effective first-line genetic assay to identify the causal genetic variants of HCM or the diseases misdiagnosed as HCM.

Conclusions

Although it has been demonstrated that HCM is primarily a disease affecting the sarcomere, its genetic basis is highly varied, and the association between a single variant and a particular typical phenotype has yet to be elucidated. Furthermore, sarcomere variants are associated with other diseases affecting cardiac function and structure. Nonsarcomeric gene variants have also been linked with HCM. Genetic assessment is appropriate for precision medicine in HCM and is essential for optimal treatment strategies, genetic counseling, and clinical management of the carriers of LMNA variants in families with HCM. In this study, we used WES and found a heterozygous variant in the LMNA gene related to the HCM phenotype, thereby demonstrating the utility of this approach for the timely and precise detection of the genetic causes of such diseases.

Abbreviations

Abbreviations
 
• HCM

  hypertrophic cardiomyopathy

 
• WES

  whole-exome sequencing

 
• MRI

  magnetic resonance imaging

Acknowledgments

Our sincere gratitude goes to the family members who participated in this study. This research was funded by Rajaie Cardiovascular Medical and Research Center, Tehran, Iran.

Conflict of Interest Disclosure

The authors have nothing to declare.

Data Availability

The datasets generated or analyzed during the current study are available in the ClinVar repository [https://www.ncbi.nlm.nih.gov/clinvar/variation/200943/]. The submission ID of the variant in ClinVar is as follows: LMNA (NM_170707.4): c.1279C>T (p.Arg427Cys): VCV000200943.18.

REFERENCES