Whole-Exome Sequencing Identified a Novel Variant (C.405_422+39del) in DSP Gene in an Iranian Pedigree with Familial Dilated Cardiomyopathy

Background: Dilated cardiomyopathy (DCM) is a progressive heart condition characterized by left ventricular chamber enlargement associated with systolic heart failure and prolonged action potential duration. Genetic variations in genes that encode cytoskeleton, sarcomere, and nuclear envelope proteins are responsible for 45% of cases. In our study, we focused on a pedigree with familial DCM to decipher the potential genetic cause(s) in affected members developing arrhythmia, end-stage heart failure, and sudden death. Methods: Whole-exome sequencing (WES) was exploited for a 27-year-old heart-transplanted female as the proband, and the derived data were filtered using the standard pipelines. Results: A 57-nucleotide deletion (c.405_422+39del) in the desmoplakin gene ( DSP ) (NM_004415.4) was identified as a novel pathogenic variant. Familial segregation analysis indicated that this variant is present in clinically affected members and absent in unaffected members. Conclusions: It seems that the detected variant induces intron retention, resulting in a premature stop codon in intron 3 of DSP leading to production of a truncated, nonfunctional protein. Additionally, it can trigger a nonsense-mediated mRNA decay pathway associated with inhibition of protein production. The present study results illustrated that a novel deletion in DSP can cause DCM in an Iranian family.


Introduction
Dilated cardiomyopathy (DCM) is a serious myocardial disorder in which myocardium becomes weak and stretched, so that systolic function becomes impaired because of the enlargement and insufficient contraction of ventricles, a condition that leads to heart failure, arrhythmia, and sudden death (1)(2)(3). The prevalence of DCM was reported as 1 out of 2500 in 1989, but was estimated at 1 out of 250 based on later findings (4). The disease is more common in men than in women. It is one of the most common causes of heart failure and heart transplantation worldwide (5). Approximately 45% of DCM cases have a genetic Rep etiology; however, pathogenic variations have been found in only 30-40% of patients (6)(7)(8). It is shown that genetic variations in more than 50 genes encoding constituents of the cytoskeleton, sarcomere, nuclear membrane, and mitochondrial proteins can cause DCM, indicating considerable genetic heterogeneity in genetic cases of DCM (9)(10)(11). Up to 90% of familial DCM cases follow an autosomal dominant inheritance mode, and about 10% of cases may be inherited in an autosomal recessive, X-linked recessive, or mitochondrial pattern (12). DSP (MIM 125647) encodes a desmosomal plaque component, desmoplakin, and its variations have been attributed to a variety of disorders related to heart muscle, hair, skin, and tooth tissues (13). Taking advantage of wholeexon sequencing (WES) technology and gene annotation, a novel small deletion variant (c.405_422+39del) was detected in DSP that may underlie the pathogenesis of DCM, and the results were validated by Sanger sequencing.

Case report
An Iranian family with autosomal dominant familial DCM has been referred ( Fig. 1

Subjects
Available members of the pedigree were enrolled in the present study. Two experienced cardiologists comprehensively examined ten family members. Peripheral venous blood from the proband (IV: 9) and five appropriate members with close relationships to the proband (III: 2, III: 5, IV: 5, IV: 7 and IV: 10), including two affected individuals (III: 5 and IV: 7), was collected for the genetic study. All medical records of healthcare, routine physical, and fundus examinations were collected.

Whole-exome sequencing
Genomic DNA was isolated from peripheral venous blood cells using the Exgene™ Blood SV DNA purification kit (GeneAll®, Korea). DNA concentrations were determined on a Thermo Scientific™ Nanodrop 2000. Exome sequencing was performed on the proband (IV: 9) by CeGaT GmbH (Tübingen, Germany). A paired-end DNA library was constructed, and the whole-exome capturing was performed using the Twist Human Core Exome Kit. After quality assessment, the captured DNA library was sequenced on the Illumina NovaSeq platform following the Illumina protocols (Illumina Inc., San Diego, CA, U.S.A.) with an average coverage of nearly 100×. About 97% of the targeted bases were covered more than 10 times.

Bioinformatics analysis
After the base calling and quality assessment of sequencing data, sequence reads were analyzed and aligned to the human reference genome applying the alignment tools. Burrows-Wheeler Aligner (BWA) and Assembly Based ReAligner (ABRA) tools were applied to sort sequencing alignments and mark duplicate reads, respectively. The single nucleotide polymorphisms (SNPs) and insertion-deletion mutations (indels) were called using the GATK HaplotypeCaller program, and subsequently annotated using the ANNOVAR program.

Co-segregation analysis
Segregation analysis was applied to involved family members using PCR-Sanger sequencing in accordance with the WES results. Primer pairs were designed using Gene Runner 6.0 software, and their sequences are as follows: forward: 5′-CAAGGGGAAGGTTAGCATTCAGCA-3′. reverse: 5′-CTTGGGAACATTTGTGCTGCCTTCA-3′. The PCR products then were evaluated by gel electrophoresis using 1.5% (w/v) agarose gel.

Results
To reveal the disease-causing genetic variant(s), the genomic DNA obtained from the proband was investigated by WES. Then, the detected variant (Table 1) was validated by PCR-gel electrophoresis and Sanger sequencing. As expected, the DNA bands represented two different fragment sizes in heterozygous status of the deletion in the proband (IV: 9), IV: 7, IV: 10, and III: 5 ( Fig.  2A). Moreover, we found the heterozygous deletion disrupting the pattern of base arrangement in the sequencing chromatograms of them (Fig. 2B). The gel electrophoresis findings of III: 2 and IV: 5 individuals displayed a single band in expected size, and their sequencing chromatograms had an intact base arrangement.  Considering appearance of the pedigree and inheritance pattern, we assumed that the causative variant(s) might be inherited in heterozygous status; subsequently, several filtering steps were applied to prioritize the variants in terms of: (1) having minor allele frequency less than 0.01 in the 1000 Genomes Project (http://www.internationalgenome.org/), ExAC (http://exac.broadinstitute.org/) and gnomAD (http://gnomad.broadinstitute.org/) databases; (2) occurrence in the coding regions or essential/canonical splicing sites; (3) having potential damaging or deleterious functional effects that were predicted by multiple lines of in silico computational analysis; (4) occurrence in a candidate gene related to inherited DCM. The filtering procedure was carried out with assuming the homozygous status of inheritance to be on the safe side, but this did not lead to any candidate variant in the family.

Discussion
Dilated cardiomyopathy is the most common non-ischemic heart condition described by the remodeling and contractile impairment of the left ventricle in absence of other disorders affecting its performance (14). Dilated cardiomyopathy onset typically occurs during adulthood but can occur any time from infancy to adolescence. Dilated cardiomyopathy has a variable phenotypic expressivity among affected patients. Clinical severity may also range from asymptomatic to mild to acute heart failure and sudden cardiac death (15). To date, more than 50 genes related to DCM have been reported (16). However, the molecular basis of a significant number of cardiomyopathies is still uncertain. The next-generation sequencing (NGS) technologies are powerful procedures that can fulfill this gap, so they have significantly accelerated the detection of disease-causing genetic changes. Among NGS techniques, WES has emerged as a strong and cost-effective method to detect the causative variant(s) in diseases with locus heterogeneity (17,18). DSP comprises 24 exons and encodes desmoplakin, an essential protein with a key role in desmosomal adhesions (19,20). Pathogenic variants in DSP have been associated with skin, hair, tooth, and heart disorders in humans. The affected members of this study had wooly hairs, but they were not investigated for possible skin and tooth anomalies. However, no defects of these types were found in the patients' records. Until March 2020, the ClinVar database lists 256 truncating DSP variants, of which 232 (90.6%) are classified as pathogenic or likely pathogenic. The variants with non-pathogenic classification have been localized in the region encoding the protein's C terminus. Since this gene is dosage sensitive, truncating variants (especially N terminus variants) can lead to cardiac disease by a haploinsufficiency mechanism (ClinGen haploinsufficiency Score: 1) (21,22). The identified variant (c.405_422+39del) deletes 57 nucleotides from the genomic DNA, of which 18 nucleotides are located in exon 3, and 39 nucleotides remain in intron 3 (Fig. 3A). This variation can cause disease with several scenarios such as p.C135_R141delinsW, exon skipping, using an exonic cryptic splicing donor site, and using an intronic cryptic splicing donor site. However, it seems that the most probable scenario is intron retention. Thus, the identified variant leads to intron 3 retention by deleting a normal splicing donor site (Fig. 3B). Subsequently, this variant leads to a premature stop codon, and is assumed to induce nonsensemediated mRNA decay. It was shown in a recent study that the truncating DSP variant demonstrates relatively low penetrance by age 40 (23). In the present study, all family members carrying the variant expressed cardiac involvement, except for proband's younger sister (IV: 10) who is apparently healthy due to her young age, although she is expected to express DCM in the future.
Multiple lines of evidence support the pathogenicity of the variant detected in the present study: a) it is a null variant; a deletion encompassing a canonical splice site (donor GT in 5′ end of intron 3) affecting DSP mRNA splicing, which is a reported mechanism of disease associated with DCM and several other phenotypes related to incomplete integrity of heart muscle, hair, skin, and tooth tissues; b) this variant is absent in public and local allele frequency databases such as 1000 Genomes Project, gnomAD database, and Iranome. Additionally, no publication has introduced this deletion as a causative variant, suggesting it is a novel variation; c) this small deletion was not detected in other family members who were not diagnosed with DCM; and d) it is predicted as a pathogenic variant (vs. benign predictions) by in silico computational prediction tools.
In conclusion, combining WES with filtering tools, we introduce a novel DSP deletion variant (c.405_422+39del) that may be a causative variant in a heart-transplanted case of DCM and other affected family members. The present study may help to further understand the relationship between DSP and DCM and introduce a novel marker for future genetic diagnosis and counseling of families with DCM.