Background
Arrhythmogenic cardiomyopathy (AC), characterized by gradual myocardium loss and fibrofatty replacement predominately in the right ventricle [
1], is one of the primary causes for life-threatening ventricular arrhythmia and sudden cardiac death (SCD), particularly in young and athletes [
2]. The clinical presentations vary, including palpitations, syncope, symptomatic ventricular tachycardia, right heart failure and SCD. Sometimes, SCD was the only manifestation in AC patients, posting tremendous challenges to the diagnosis post mortem [
2,
3]. Diagnosis of AC, according to the guideline proposed by the international task force [
4], is mainly based on findings of electrophysiological, structural and histological features, family history and genetic testing, hence, for those SCD patients, their family screening is of utmost importance. The current treatments for AC are mostly supportive and palliative [
5], aiming at alleviation of arrhythmic and heart failure symptom and prevention of SCD, and heart transplantation is the final solution for end-stage patients. However, reversal or a complete cure of the disease requires further in-depth understanding of its etiology and pathogenesis.
Known as genetically determined cardiomyopathy, AC is mainly inherited in an autosomal dominant pattern with genetic and phenotypic heterogeneity [
6]. Genetic studies have identified mutations in 5 components of cardiac desmosomes as main etiology of AC [
6], namely Plakophilin 2 (
PKP2), Desmoplakin (
DSP), Desmoglein 2 (
DSG2), Desmocollin 2 (
DSC2), and Junction plakoglobin (
JUP). Genetic defects of above genes can be found in 40–60% of AC patients [
4]. However, the specific etiology in individual case remains largely unknown. First identified in a recessive disorder of keratoderma, woolly hair, and AC with left ventricle predominance (Carvajal syndrome) [
7],
DSP mutations are responsible for nearly 2–12% of AC patients [
8,
9]. Recent study interestingly found that the left ventricle predominance or bi-ventricle involved phenotypes were associated with
DSP non-missense mutations [
10], but the genotype-phenotype correlations remain uncertain due to small sample size and need to be further characterized in individual families as well as large sample cohorts. Recent studies also suggested mutations that impaired ion channel activities may be causal or modifier to AC [
11,
12], however, their prevalence is unsure.
In the current study, the underlying genetic defects in a 4-generation family presenting syncope, life-threatening ventricular arrhythmia and SCD were explored using next generation high-throughput sequencing platform, and a novel frame-shift variant c.832delG in DSP was identified. Cardiac magnetic resonance (CMR) further reveled the diagnosis of AC on two asymptomatic family members carrying the identical DSP variant. Through co-segregation and genotype-phenotype association analysis, and functional study on HEK293T cells, we infer that the novel frame-shift variant DSP c.832delG was associated with AC in this family.
Methods
Study subjects
The study protocol conforms to the ethical guidelines of the 1975 Declaration of Helsinki and was approved by Institutional Review Board (IRB) at the Second Affiliated Hospital, Zhejiang University School of Medicine (2016–087). Written informed consent was obtained from all participants. Ten out of total 31 family members in a 4-generation SCD family were recruited in the current study. A complete clinical information including family history, medical history, physical examination, lab test, 12-lead echocardiogram (ECG), 24-h Holter monitoring, transthoracic echocardiography and CMR were collected.
The proband was selected for next generation sequencing using a commercial capture array (Roche NimbleGen, WI, USA) covering the exons and 50 base pairs of adjacent introns of 1876 cardiovascular diseases associated genes, including inherited cardiomyopathy, arrhythmogenic diseases, congenital heart diseases, mitochondrial diseases, etc.
Genomic DNA was extracted from peripheral blood lymphocytes by standard procedures using Axygen® AxyPrep™-96 Blood Genomic DNA Kit (Axygen, NY, United States). The DNA libraries were constructed and sequenced using the Illumina 2000 platform (Illumina, CA, United States), providing an average sequencing depth of > 100-fold of targeted exons.
The screening algorithms for potential disease-causing variants were as follows. Initially, intronic and synonymous exonic variants were excluded. Secondly, matched population and in-house database minor allele frequencies (MAF) were used to rule out common variants, defined by MAF > 0.01. MAF of 3 major SNP databases were compared: ExAc (
http://exac.broadinstitute.org/), 1000 genomes (
http://www.1000genomes.org/) and ESP6500 (
http://evs.gs.washington.edu/EVS/). Thirdly, rare non- synonymous variants were examined with HGMD (
http://www.hgmd.cf.ac.uk/ac/), OMIM (
http://www.omim.org/) and ClinVar databases (
https://www.ncbi.nlm.nih.gov/clinvar/) and finally analyzed using 3 known prediction tools, namely PolyPhen-2 (
http://genetics.bwh.harvard.edu/pph2/), SIFT (
http://sift.jcvi.org/) and MutationTaster (
http://www.mutationtaster.org/), and categorized according to the recommended guidelines of the American College of Medical Genetics and Genomics (ACMG) and the Association for Molecular Pathology [
13]. Sanger sequencing was performed bidirectionally for the verification of
AKAP9 c.10714C > G,
FLNC c.7778C > G,
SYNE1 c.25954C > T and
DSP c.832delG in all participants.
Plasmids construction and site-directed mutagenesis
AICSDP-9:DSP-mEGFP was a gift from the Allen Institute for Cell Science (Addgene plasmid # 87424;
http://n2t.net/addgene:87424; RRID:Addgene_87,424) [
14]. In order to facilitate the observation following transfection of mutant plasmid, GFP were cleaved and inserted in between the promoter and
DSP gene. The frame-shift mutation was introduced into a wild-type
DSP clone using a QuikChange II XL Site-Directed Mutagenesis Kit (Stratagene, La Jolla, CA, USA). The clones were sequenced to confirm the desired mutation and to exclude any other sequence variations.
RT-PCR and real-time PCR
HEK293T cells were transfected with either blank, wild type or mutant plasmids using lipofectamine 3000 (Invitrogen, MA, USA) according to the manufacturer’s instructions. Total RNA was extracted from transfected cells using the Trizol reagent (Invitrogen, MA, USA). cDNA was synthesized using PrimeScript RT reagent Kit (Takara, Shiga, Japan). The resulting cDNA was subjected to real-time PCR using TB Green Premix Ex Taq kits (Takara, Shiga, Japan) on an Applied Biosystems 7500 Fast Real-Time PCR System (ABI, CA, USA). The primers named “N-terminal” detected the mRNA levels in the N-terminal side of the DSP mutation site, and the primers named “C-terminal” detected the mRNA levels in the C-terminal side of the DSP mutation site. GAPDH was used as an endogenous control.
The sequences of primers were listed as follows:
N-terminal-F: 5′-GCAGGATGTACTATTCTCGGC-3′,
N-terminal-R: 5′-CCTGGATGGTGTTCTGGTTCT-3′;
C-terminal-F: 5′-ACATCATTCAGGCCACGT-3′;
C-terminal-R: 5′- CCAGTTGACTCATGCGTA-3′;
GAPDH-F: 5′-CGCTCTCTGCTCCTCCTGTT-3′;
GAPDH-R: 5′-CCATGGTGTCTGAGCGATGT-3′.
Western blots
24 h after transfection, total cell extracts were lysed by RIPA lysis buffer. Nuclear and cytoplasmic extracts were separated using Nuclear and Cytoplasmic Protein Extraction Kit (Beyotime Biotechnology, Shanghai, China). Next, proteins were separated by sodium dodecyl sulfate poly-acrylamide gel electrophoresis (SDS-PAGE) and transferred to polyvinylidene fuoride (PVDF) membranes. The membranes were blocked for 1 h in a blocking solution of 5% (w/v) non-fat milk in PBS containing 0.1% (v/v) Tween-20 and incubated at 4 °C overnight with indicated primary antibodies. Primary antibodies included antibodies against JUP (1:1000, sc-8415, Santa Cruz Biotechnology, CA, USA), β-catenin (1:1000, ab6302, Abcam, Cambridge, UK), GFP (1:1000, AF1483, Beyotime Biotechnology), GAPDH (1:5000, 3683S, Cell Signaling Technology, MA, USA), Lamin B1 (1:1000, ET1606–27, HuaBio antibodies, China). Excess primary antibodies were washed off, and then the membranes were incubated with secondary antibodies conjugated with horseradish peroxidase for 1 h at room temperature. The western blot bands were visualized were visualized using the enhanced chemiluminescence western blotting detection system (Bio-Rad, CA, USA).
Immunofluorescence analysis
Cells seeded on cover slips were fixed with 4% paraformaldehyde (PFA)/PBS, permeabilized in 0.5%(v/v) Triton X-100 (Sigma-Aldrich, MO, USA) and blocked with 5% (w/v) BSA. Then the cells were incubated using the antibody mouse-anti-JUP (1:1000, sc-8415, Santa Cruz Biotechnology) overnight at 4 °C, followed by secondary antibodies anti-mouse Alexa Fluor 594 (1:200, Thermo Fisher, A-21203, CA, USA) incubation in 5% BSA in PBS for 1 h at room temperature. Finally, coverslips were mounted on microscope slides using mounting medium contained with DAPI (H-1200, Vector, CA, USA). Images were acquired using a fluorescence microscope (Leica, IL, USA). Colocalization analysis between JUP and nuclear was performed by Coloc 2 ImageJ in random high-power fields. Pearson’s correlation coefficient was used to represent the colocalization quantification, + 1 for perfect correlation, 0 for no correlation, and − 1 for perfect anti-correlation. Optical confocal microscopies of cells were obtained using Leica TCS SP8 (Leica Microsystems Inc).
Statistical analysis
Data were presented as the means ± SEM of at least three independent experiments. Student T test was performed to evaluate differences of continuous variables between two groups. One-way ANOVA was used for comparison among three groups. P values of less than 0.05 were considered statistically significant. Statistical calculations were carried out using GraphPad Prism 8.0.1.
Discussion
In the current study, through targeted next generation sequencing platform covering a board rang of inherited cardiovascular disease genes, a novel frame-shift variant
DSP c.832delG is identified in a large SCD family. CMR unveils the typical manifestations of myocardium thinning, fatty replacement and severely impaired heart function, particular in the right heart of the variant carriers, fulfilling the international Task Force criteria for the diagnosis of AC [
4]. Functional study on HEK293t cells reveals truncation of
DSP protein, down-regulation of
JUP and up-regulation of β-catenin expression in nuclear, but not cytoplasm upon transfection of plasmids with
DSP c.832delG.
Desmoplakin, a member of the plakin family, anchors other desmosome components to intermediate filaments as to maintain the integrity of desmosome structure [
16]. SCD is reported to be more prevalent in
DSP defect patients, especially truncations [
17], when compared with other desmosome defects [
9]. In our AC family, 4 family members present with SCD/aborted SCD as first clinical manifestation, and the VT/VF survivor carries
DSP c.832delG truncation, consistent with previous findings. It has been proposed that
DSP missense mutation exert a negative dominant effect whereas non-missense mutation exert haploinsufficiency [
18], leading to phenotypic discrepancy.
DSP missense mutation presents with more severe phenotype than non-missense mutation [
19], such as earlier disease onset and more prevalence of lethal arrhythmia. However, this correlation is inconsistently reported in clinical studies. Up to date, the largest AC cohort with
DSP mutation recruiting 27 patients suggests that non-missense mutations is only associated with left-dominant forms [
10]. In the current study, despite normal TTE, CMR exam sensitively detects that 2 of our
DSP c.832delG carriers present mild to moderate left ventricle involvement, nevertheless, right ventricular impairment is dominant, suggesting phenotype is possibly mutation-dependent. Apparently, larger sample of AC cohort with various types of
DSP mutation will be needed to further explore the genotype-phenotype correlation.
The canonical Wnt/β-catenin signaling is considered to play a central role in the pathogenesis of AC with
DSP defects [
20]. Non-specific heterozygous
DSP-deficient mice demonstrate substantial adiposity and fibrosis in the ventricular myocardium, recapturing the human AC phenotype [
21]. Nuclear translocation of the desmosomal protein plakoglobin (JUP) and suppression of Wnt/β-catenin signaling pathway activity are found to be the underlying mechanism [
21]. However, cardiac-restricted
DSP-deficient mice develop a biventricular form of AC and no significant changes in JUP or β-catenin expression were detected [
22], indicating that mechanisms other than Wnt pathway are responsible. In addition, silencing in HL-1 cells result in decreased expression and redistribution of the
Nav1.5 protein and reduced sodium current [
23], indicating an orchestra of canonical and non-canonical pathways synergically modulated the disease pathogenesis. Hence, immortal lymphoblastoid cell lines from the
DSP c.832delG carriers and non-carriers in this family are established as to investigate the molecular pathogenesis. However, in our study no obvious
DSP expression is detected by either western-blot or flow cytometry (data not shown), hindering the utilization of this cell line in downstream study. Therefore, plasmid carrying
DSP c.832delG is constructed and transfected into HEK293T cells. Upregulation of
JUP and downregulation of β-catenin in the nuclear suggest canonical Wnt/β-catenin signaling pathway is likely to play a central role in the development of AC phenotype as previously reported [
21]. However, HEK293T cells are unable to simulate the character of cardiomyocyte, hindering further studies on non-canonical pathways and cardiac phenotype.
Various cell models have been established to explore the potential effect of mutations [
24]. Buccal mucosa cells from AC patients exhibit redistribution of desmosomes and gap junction protein, similar to those observed in heart [
25]. However, in-depth phenotypic and mechanistic studies are not possible due to its distinct cellular features from cardiomyocytes. Patients-specific induced pluripotent stem cells (iPSc) derived cardiomyocytes contain the unique mutations and complete genetic background [
26], thus providing us an ideal model to investigate the precise etiology and molecular mechanism. Moreover, the combination of iPSc and latest genome editing technology, such as CRISPR/Cas9, has been succeeded in correcting LQT causal mutations and reversing phenotype [
27,
28], promoting it as a promising approach towards precision medicine, and thereby should be introduced in our future study.
Limitations
In the current study, only HEK293T, a non-cardiac cell line, is utilized. Though human non-myocardial cell lines have been used as a cell model for investigating adhesive junction functions in AC [
29], the effects of mutant
DSP may differ in HEK293T cells from cardiomyocytes. Furthermore, non-cardiac cells are unable to reproduce the phenotype observed in human disease. Human iPSCs derived cardiomyocytes contain the unique genetic background of the patients and features of cardiac cells, hence they are robust tools to perform future studies and explore the mechanistic pathways. Transgenic animals, especially murine genetic knock-ins, are the most powerful and convincing models to investigate human inherited diseases, and also also be considered in the future studies.
Publisher’s Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.