Danon disease is an X-linked disorder with the clinical triad of cardiomyopathy, skeletal myopathy, and mental retardation. Early diagnosis of this disease remains a challenge, especially in the pediatric population. In this study, we developed a targeted panel-based next generation sequencing pipeline to identify mutations by sequencing of selected candidate genes in 136 pediatric patients with either hypertrophic cardiomyopathy (HC) or idiopathic dilated cardiomyopathy (IDC). This led to the identification of lysosome-associated membrane protein 2 ( LAMP2 ) mutations in 4 of the 64 (6%) probands with HC, including 3 novel nonsense mutations (p.Q240X, p.S250X, and p.G22X). No LAMP2 mutation was detected in the other 72 probands with IDC. All 4 probands and one additional affected family member (2 men and 3 women) had an early-onset age and presented either HC alone or combined with Wolff-Parkinson-White syndrome and skeletal myopathy. Immunofluorescence staining and Western blot analysis revealed absent LAMP2 expression in both cardiac and skeletal muscle samples of the first proband and severely decreased LAMP2 expression in the skeletal muscle samples of the second proband. In conclusion, cardiomyopathy in the patients with Danon disease may occur during early childhood and tend to be HC rather than IDC in both affected men and women. Therefore, Danon disease should be considered as one of the leading causes of unexplained ventricular hypertrophy in pediatric patients. The inclusion of LAMP2 gene in cardiomyopathy genetic screening panels may contribute to early diagnosis of Danon disease.
Danon disease (OMIM# 300257 ) is a rare X-linked dominant disease associated with the clinical triad of cardiomyopathy, skeletal myopathy, and mental retardation, which is caused by mutations in the lysosome-associated membrane protein 2 (LAMP2) gene located at Xq24. Cardiac involvement is very common and is the most important cause of death in Danon disease. Early diagnosis is crucial for improving prognosis of the disease, but it may be difficult in the pediatric population where symptoms are often nonspecific. Genetic testing is now increasingly used as a means to confirm the specific diagnosis of cardiomyopathy but remains a challenge because of the large number of causative genes involved in its pathogenesis. One possible approach that has yet to be widely implemented is the use of high-throughput DNA sequencing techniques offered by next generation sequencing (NGS). In the present study, we developed a targeted panel-based NGS pipeline to identify mutations by sequencing of selected candidate genes in a large cohort of pediatric patients with either hypertrophic cardiomyopathy (HC) or idiopathic dilated cardiomyopathy (IDC). This led to the identification of LAMP2 mutations in 4 of the 64 (6%) probands with HC, including 3 novel nonsense mutations (p.Q240X, p.S250X, and p.G22X).
Methods
This study was approved by the Medical Ethics Committee of Shanghai Children’s Medical Center and complied with the principles expressed in the Declaration of Helsinki. Written informed consent was obtained from the patient or legally responsible guardian.
During the period from 2012 to 2015, all patients with a clinical diagnosis of HC or IDC who were referred to our institution were enrolled in this study. IDC was defined as left ventricular ejection fraction <45% and left ventricular end-diastolic dimension >2 SDs above the normal mean for body surface area in the absence of known causes such as ischemic heart disease, primary valvular heart disease, or inflammatory cardiomyopathy; and HC was defined as left ventricular posterior and/or septal wall thickness >2 SDs above the normal mean for body surface area in the absence of an identifiable hemodynamic cause such as hypertension, congenital heart disease, or exposure to drugs known to cause cardiac hypertrophy. All patients were evaluated by a clinical history, physical examination, electrocardiography, echocardiography, and serum biochemical analyses. When available, cardiac magnetic resonance imaging with late gadolinium enhancement was performed.
Genomic DNA was extracted from whole blood using a QIAamp Blood DNA Mini kit (Qiagen GmbH, Hilden, Germany). Oligonucleotide-based target capture (Custom target enrichment kit, SureSelect; Agilent, Santa Clara, California) followed by NGS (Illumina HiSeq2500) on the DNA samples was used to capture variants of 62 genes implicated in the causation of cardiomyopathy ( Supplementary Table 1 ). Alignment of sequence reads to the reference human genome (Human 37.3, SNP135) was performed using the NextGENe software (SoftGenetics, State College, Pennsylvania). All single nucleotide variants and indels were saved in a VCF format and uploaded to Ingenuity Variant Analysis (Ingenuity Systems, Redwood City, California) for filtering and interpretation. A set of filters cascade was applied for variants analysis that significantly narrow down the number of variants starting from the “raw variants/vcf file” to “call quality/confidence” to “common variants” to “predicted deleterious” to “candidate causal variants.” All the variations were classified according to the recommended method of American College of Medical Genetics and Genomics. Where possible, clinically significant variants were confirmed by independent Sanger sequencing and validated by parental testing and segregation analysis. Variants classified as benign or likely benign were not pursued for validation by Sanger sequencing. NM_002294.2 was used as the reference sequence for the coding regions of the LAMP2 gene.
Cryostat sections of 8-μm thickness were used for routine histological and histochemical studies, including hematoxylin-eosin, Gomori modified trichrome, nicotinamide adenosine dinucleotide-tetrazolium reductase, succinate dehydrogenase, adenosine triphosphatase, cytochrome c oxidase, and periodic acid-Schiff staining. Mouse monoclonal antibodies against dystrophin (DYS1: Rod domain and DYS3: N-terminus; Novocastra, Newcastle upon Tyne, United Kingdom), dysferlin (NCL-Hamlet; Novocastra), and sarcoglycan complex (NCL-g-SARC, NCL-a-SARC, NCL-b-SARC, NCL-d-SARC; Novocastra) were used in the immunohistochemical staining. The LAMP2 antibody (H4B4; Abcam, Cambridge, United Kingdom) was used for further immunofluorescence study.
Western blot was performed on muscle homogenates as described. Briefly, muscle specimens (30 mg of wet weight for each) were homogenized in 20 volumes of lysis buffer and boiled at 100°C for 5 minutes. After centrifugation, the supernatant was subjected to SDS-PAGE, followed by transfer to nitrocellulose membrane. Mouse monoclonal antibodies against LAMP2 and glyceraldehyde-3-phosphate dehydrogenase were used as primary antibodies.
Results
A total of 136 unrelated Chinese pediatric patients clinically diagnosed with HC or IDC were enrolled in this study. Among the index cases, there were 64 patients with HC (39 men and 25 women; age at diagnosis: 2 months to 14 years), and 72 patients with IDC (44 men and 28 women; age at diagnosis: 3 months to 15 years). We performed targeted NGS on all these patients and identified 3 novel and one previously reported LAMP2 nonsense mutations in 4 of 64 probands with HC. No LAMP2 mutation was detected in the other 72 probands with IDC. The results were validated by Sanger sequencing in probands and family members. The clinical features and laboratory test results pertinent to the 4 probands and their family members are described in the following section and summarized in Table 1 .
| Patients | DD1 | DD2 | DD3 | DD4 | DD5 |
|---|---|---|---|---|---|
| Sex | Male | Female | Female | Male | Female |
| Presentation | Heart failure | None | Exertional dyspnea | Heart murmur | Heart murmur |
| Age at diagnosis (Years) | 13 | 2 | 10 | 0.25 | 2 |
| Muscle weakness | Mild | None | None | None | None |
| Mental retardation | None | None | None | None | None |
| Serum creatine kinase (U/L) (30–135) | 1336 | 142 | 147 | 309 | 138 |
| Aspartate aminotransferase (U/L) (10–42) | 205 | 61 | 80 | 258 | 65 |
| Alanine aminotransferase (U/L) (10–40) | 177 | 16 | 32 | 130 | 35 |
| Lactate dehydrogenase (U/L) (313–618) | 2248 | 1057 | 1890 | 1191 | 1020 |
| Amino-terminal pro-brain natriuretic peptide (ng/L) (0–125) | 18759 | 635 | 2855 | N/A | 548 |
| Electrocardiography abnormality | WPW, LVH, Inverted T-waves | LVH | LVH, Inverted T-waves | LVH | WPW, LVH, Inverted T-waves |
| Echocardiography | |||||
| Left ventricular outflow gradient (mm Hg) | 0 | 0 | 50.0 | 0 | 41.8 |
| Maximum left ventricular posterior wall thickness (mm) | 16.5 | 6.8 | 28.8 | 15.7 | 22.1 |
| Maximum septum thickness (mm) | 19.4 | 9.7 | 17.5 | 15.9 | 17.5 |
| Left ventricular ejection fraction (%) | 22.2 | 67.4 | 77.7 | 79.5 | 89.5 |
| Left ventricular end-diastolic dimension (mm) | 57.2 | 28.1 | 36.7 | 19.2 | 26.3 |
| Left ventricular end-systolic dimension (mm) | 51.3 | 18.0 | 20.0 | 10.5 | 11.0 |
Family 1
The proband (DD1 in Table 1 ; II: 1 in Figure 1 ) in this family was a 13-year-old man who presented with a 3-month history of short of breath and edema of lower extremities. He was born of an uneventful full-term pregnancy and apparently had normal growth and developmental milestones. On admission, he presented with mild skeletal weakness but no learning difficulties. Laboratory tests showed elevated creatine kinase and alanine aminotransferase. The aminoterminal probrain natriuretic peptide was strikingly elevated at 18,759 ng/L. His electrocardiography revealed Wolff-Parkinson-White pattern and left ventricular hypertrophy (LVH; Figure 2 ). Echocardiography revealed LVH with dilatation of the left ventricle and severe systolic dysfunction, a finding compatible with the end-stage phase of HC ( Figure 2 ). The patient died of progressive heart failure 6 months after his initial diagnosis despite maximal medical management and treatment of the condition.
In this proband, we identified a hemizygous c.718 C > T mutation in exon 5 of the LAMP2 gene, resulting in a premature stop codon (p.Q240X). His mother (I: 1 in Figure 1 ) carried the same mutation but had no overt clinical symptoms. She also presented with a normal electrocardiography and echocardiography. The 2-year-old sister (DD2 in Table 1 , II: 2 in Figure 1 ) of the proband was found to be heterozygous for the same mutation. She did not exhibit skeletal myopathy or mental retardation. Her serum creatine kinase was slightly elevated. Electrocardiography revealed LVH without ventricular preexcitation. Echocardiogram demonstrated mild LVH, but there was no evidence of left ventricular outflow tract obstruction.
Family 2
The proband (DD3 in Table 1 ) was a 10-year-old woman, presenting a 4-month history of exertional dyspnea and easy fatigability. She was born uneventfully, had normal growth, and did not exhibit skeletal myopathy or mental retardation. Serum creatine kinase was slightly elevated. The aminoterminal probrain natriuretic peptide was elevated at 2,855 ng/L. Her electrocardiography showed markedly increased voltages with a deeply inverted T-wave in both standard lead and left precordial lead, but no ventricular preexcitation was observed. Cardiac ultrasound demonstrated severe LVH and left ventricular outflow tract obstruction, with a peak instantaneous gradient of 50 mm Hg. Her cardiac magnetic resonance imaging also showed LVH with patchy late gadolinium enhancement ( Figure 3 ).
In this proband, we identified a novel heterozygous c.749 C > G mutation in exon 6 of the LAMP2 gene, resulting in a premature stop codon (p.S250X). This mutation was not found in her healthy parents, indicating that the mutation had arisen de novo.
Family 3
The proband (DD4 in Table 1 ) was a 3-month-old man who was diagnosed with HC after the detection of a heart murmur. He was born of an uneventful full-term pregnancy and had normal growth. In addition, there was no documented muscle weakness in the limbs or mental retardation. Blood tests revealed elevated creatine kinase and alanine aminotransferase. His electrocardiography revealed LVH without ventricular preexcitation. Echocardiography exhibited severe LVH but no left ventricular outflow tract obstruction at the time of initial diagnosis. At present, he is 17-month-old and remains asymptomatic with normal developmental milestones; however, his echocardiogram demonstrated left ventricular outflow tract obstruction with a peak instantaneous gradient of 66 mm Hg.
In this proband, we identified a reported hemizygous c.467 T > G mutation in exon 4 of the LAMP2 gene, resulting in a premature stop codon (p.L156X). Neither of his healthy parents carried this mutation. Therefore, it is considered a de novo mutation.
Family 4
The proband (DD5 in Table 1 ) was a 2-year-old woman who was diagnosed with HC after the detection of a heart murmur. She was born of an uneventful full-term pregnancy and had normal growth. A neurological examination revealed normal muscle strength and normal deep tendon reflexes. No mental retardation was observed. Serum creatine kinase was slightly elevated. Her electrocardiography revealed LVH and Wolff-Parkinson-White pattern. Echocardiogram demonstrated severe LVH and left ventricular outflow tract obstruction, with a peak instantaneous gradient of 42 mm Hg. At present, she is 3-year-old and remains asymptomatic.
In this proband, we identified a novel heterozygous c.64 G > T variant in exon 1 of the LAMP2 gene, resulting in a premature stop codon (p.G22X). This mutation was not found in her healthy parents, indicating that the mutation has arisen spontaneously.
Muscle biopsies were performed in the biceps brachii and heart within 2 hours after death for patient DD1. The skeletal muscle biopsy showed mild variation in fiber size and a few cytoplasmic vacuoles with positive reactions to antisarcolemmal protein staining. The cardiac biopsy of the patient showed hypertrophic cardiomyocytes, severe interstitial fibrosis with small cytoplasmic vacuoles in several fibers. Immunohistochemical staining with dystrophin and sarcoglycan antibodies disclosed increased immunoactivity on the rim of the small vacuoles. Immunofluorescence staining ( Figure 4 ) and Western blot analysis ( Figure 5 ) revealed a complete absence of LAMP2 expression in both cardiac and skeletal muscle samples.
