|
|
| |||||||||||||||||||||||||||||||||||||||||||||||
Background Unexplained left ventricular hypertrophy often prompts the diagnosis of hypertrophic cardiomyopathy, a sarcomere-protein gene disorder. Because mutations in the gene for AMP-activated protein kinase
Methods Genetic analyses performed in 75 consecutive unrelated patients with hypertrophic cardiomyopathy detected 40 sarcomere-protein mutations. In the remaining 35 patients, PRKAG2, lysosome-associated membrane protein 2 (LAMP2),
Results Gene defects causing Fabry's disease (GLA) and Pompe's disease (GAA) were not found, but two LAMP2 and one PRKAG2 mutations were identified in probands with prominent hypertrophy and electrophysiological abnormalities. These results prompted the study of two additional, independent series of patients. Genetic analyses of 20 subjects with massive hypertrophy (left ventricular wall thickness,
Conclusions LAMP2 mutations typically cause multisystem glycogen-storage disease (Danon's disease) but can also present as a primary cardiomyopathy. The glycogen-storage cardiomyopathy produced by LAMP2 or PRKAG2 mutations resembles hypertrophic cardiomyopathy but is distinguished by electrophysiological abnormalities, particularly ventricular preexcitation.
2 (PRKAG2) cause an accumulation of cardiac glycogen and left ventricular hypertrophy that mimics hypertrophic cardiomyopathy, we hypothesized that hypertrophic cardiomyopathy might also be clinically misdiagnosed in patients with other mutations in genes regulating glycogen metabolism. 
-galactosidase (GLA), and acid -1,4-glucosidase (GAA) genes were studied. 
30 mm) but without electrophysiological abnormalities revealed mutations in neither LAMP2 nor PRKAG2. Genetic analyses of 24 subjects with increased left ventricular wall thickness and electrocardiograms suggesting ventricular preexcitation revealed four LAMP2 and seven PRKAG2 mutations. Clinical features associated with defects in LAMP2 included male sex, severe hypertrophy, early onset (at 8 to 17 years of age), ventricular preexcitation, and asymptomatic elevations of two serum proteins.
30 mm) is increasingly recognized as an important risk factor for sudden death.3,4
Sarcomere-protein gene mutations cause familial or sporadic hypertrophic cardiomyopathy and 15 percent of the cases of elderly-onset hypertrophic cardiomyopathy.5 To date, more than 200 mutations in 10 different genes are known.6 Molecular studies of patients with clinical features of hypertrophic cardiomyopathy but without sarcomere-protein gene defects have led to the identification of other genetic causes of cardiac hypertrophy, including mutations in PRKAG2,7,8,9 the regulatory subunit of AMP-activated protein kinase. PRKAG2 mutations cause myocyte hypertrophy by stimulating glycogen-filled vacuoles but cause neither myocyte disarray nor interstitial fibrosis, which typically occur with defects of sarcomere-protein genes.9,10
Pathologic vacuoles containing glycogen or intermediary metabolites also occur in Pompe's disease (a recessively inherited lysosomal acid -1, 4-glucosidase [GAA] deficiency), Danon's disease (an X-linked lysosome-associated membrane protein [LAMP2] deficiency), and Fabry's disease (an X-linked lysosomal hydrolase a-galactosidase A [GLA] deficiency).11,12,13,14,15,16 These multisystem disorders cause neuromuscular deficits, abnormal liver and kidney function, and abnormalities of the central nervous system as well as cardiac hypertrophy. Although some, atypical, patients with Fabry's disease have mild systemic manifestations and, predominantly, cardiac disease,11,17 the pleiotropic manifestations of Pompe's disease and Danon's disease rarely prompt the consideration of these disorders in the differential diagnosis of unexplained left ventricular hypertrophy.
We sequenced eight sarcomere-protein genes in 75 unrelated patients with hypertrophic cardiomyopathy in whom echocardiography showed unexplained left ventricular hypertrophy. Subsequent analyses of PRKAG2, LAMP2, GAA, and GLA in samples of patients who did not have a sarcomere-protein gene mutation revealed previously unidentified LAMP2 and PRKAG2 mutations. The clinical manifestations associated with these mutations prompted studies of two additional patient series: one involved subjects with massive hypertrophy, and one involved those with left ventricular hypertrophy plus electrophysiological defects.
Methods
Clinical Evaluations
Studies were performed in accordance with institutional guidelines for human research. The research protocol was reviewed and approved by the institutional review boards at the participating institutions, and written informed consent was obtained from all research subjects. Three independent series of patients were studied: one involving 75 consecutive subjects 12 to 75 years of age who had hypertrophic cardiomyopathy as diagnosed on the basis of echocardiograms showing unexplained left ventricular hypertrophy (wall thickness, 13 mm)3,18; one involving 20 subjects (9 to 58 years of age) with massive left ventricular hypertrophy (wall thickness, 30 mm) of unknown cause; and one involving 24 patients (8 to 42 years of age) with hypertrophic cardiomyopathy in whom electrocardiograms suggested the presence of ventricular preexcitation (a short PR interval, delta wave, or both).
Study subjects were from North America, South America, and Europe and identified themselves as white (86 percent), black (7 percent), or Hispanic (7 percent). Medical records, clinical evaluations, electrocardiograms, and echocardiograms were reviewed. Clinical studies that were performed before enrollment at the discretion of the referring cardiologist were included when available. After completion of genetic studies, cardiac evaluations were performed of family members carrying a mutation. Patients with LAMP2 mutations also underwent noninvasive neurologic and musculoskeletal evaluations and serum chemistry analyses. When available, pathological specimens were examined. All values are reported as means ±SD.
Genetic Studies
The genes encoding cardiac 
-myosin heavy chain, cardiac myosin-binding protein C, cardiac troponin T, cardiac troponin I, cardiac actin, essential myosin light chain, regulatory myosin light chain, -tropomyosin, and PRKAG2 were sequenced from genomic DNA as described previously.9,18 Exons 1 through 8, 9a, and 9b of LAMP2, exons 2 through 20 of GAA, and exons 1 through 7 of GLA were amplified with the use of the polymerase chain reaction (PCR) and sequenced and compared with GenBank accession numbers AC002476
[GenBank]
, NT_024915, and AL035422
[GenBank]
with the use of primers available on the Internet (at http://genetics.med.harvard.edu/~seidman/). Sequence variants were confirmed by restriction-enzyme digestion. Variants that segregated with clinical status in family members and that were absent from 180 normal subjects who were matched with the subjects with hypertrophic cardiomyopathy for race or ethnic background (self-reported) were considered disease-causing mutations18,19 and were denoted by standard nomenclature.20 LAMP2 alleles were distinguished by single-nucleotide polymorphisms 156 A/T and 927 C/T, numbered according to complementary DNA (cDNA) (GenBank accession number NM_013995
[GenBank]
).
RNA was extracted with the use of Trizol (Invitrogen). We performed reverse transcription (RT) using a kit (One-Step RT-PCR, Qiagen) with primers available on the Internet (at http://genetics.med.harvard.edu/~seidman/).
Protein Analyses
Western blot analyses were performed in immunoprecipitation assay buffer with the use of 30 to 40 µg of protein lysates from lymphocytes or fibroblasts, as described previously,21 with polyclonal LAMP2 and glyceraldehyde-3-phosphate dehydrogenase (GAPDH) antibodies (Santa Cruz Biotechnology).
Histopathological Analyses
Specimens were examined after staining with hematoxylin and eosin and LAMP2 immunohistochemistry. Electron-microscopical examinations of embedded tissue were performed after paraffin removal with the use of previously described procedures.9
Electrocardiography
Standard 12-lead electrocardiographic recordings were examined for ventricular preexcitation and left ventricular voltage (reported as the maximal S wave in V1 or V2 + maximal R wave in V5 or V6 [SV1 or SV2 + RV5 or RV6] or the maximal R or S deflection in any lead22).
Results
Seventy-five unrelated patients with hypertrophic cardiomyopathy (30 female and 45 male patients, 12 to 75 years of age at diagnosis) were prospectively enrolled for genetic analyses of sarcomere-protein mutations. Maximal left ventricular wall thickness ranged from 13 to 60 mm and in four persons exceeded 30 mm. In addition to typical electrocardiographic manifestations of hypertrophic cardiomyopathy,3,18 three persons had short PR intervals and three others had ventricular preexcitation patterns. Forty sarcomere-protein gene mutations were identified in these 75 patients.6,23 In the remaining 35 patients (27 of whom were male and 8 female), PRKAG2, LAMP2, GLA, and GAA sequences were analyzed. No mutations were identified in GLA or GAA sequences.
A previously undetected missense mutation — tyrosine to histidine at codon 487 — of PRKAG2 in Proband IP was associated with moderate hypertrophy (left ventricular wall thickness, 13 mm) and an extremely short PR interval (0.09 msec). LAMP2 mutations in which the sequence GTGA was deleted from the splice-donor site of intron 6 (IVS6+1_4delGTGA) and in which there was an A-to-G change in the splice-acceptor site of intron 6 (IVS6 –2A
G) in Probands CZ and FI were associated with severe hypertrophy (left ventricular wall thicknesses, 29 mm and 60 mm, respectively) and unusual electrocardiograms, with short PR intervals, short delta waves, or both, and extreme voltage, suggestive of ventricular preexcitation (Figure 1). These clinical findings prompted the genetic studies of two additional patient series. LAMP2 and PRKAG2 sequences were determined in 20 patients with a left ventricular wall thickness of 30 mm or more; however, no mutations were identified. LAMP2 and PRKAG2 sequences were also determined in 24 probands with increased left ventricular wall thickness and electrocardiograms suggesting ventricular preexcitation. Seven PRKAG2 mutations9 and four LAMP2 mutations were identified, which corresponded to a mutation-detection rate of 46 percent in this group.
|
Six new LAMP2 mutations were detected — in Probands CZ, IP, LS, MFE, NR, and SS — that are predicted to alter substantially the lysosome-associated membrane protein, a 410-amino-acid molecule with a small cytoplasmic tail involved in receptor-mediated lysosomal uptake, and a large internal lysosome domain composed of highly glycosylated residues. One nonsense mutation (in Proband NR) signaled premature termination at amino acid 109. Five other mutations altered splice signals; the consequences of these alterations were assessed in the LAMP2 RNAs isolated from lymphocytes.
Mutation IVS1+1GT (in Proband MFE) altered the intron 1 splice donor site; RNA maturation occurred by a cryptic splice site that excised 21 amino acids after the initiation codon. Mutation IVS1–2AG (in Proband SS) altered the intron 1 splice-acceptor site; RNA deleted exon 2 residues and produced a frameshift mutation. Mutation IVS6+1_4delGTGA (in Proband CZ) altered the intron 6 splice-donor site and excised 41 codons. No mutant RNA was detected from mutation IVS6– 2AG (in Proband FI) that altered the splice-acceptor site of intron 6, perhaps indicating that this defect triggered missense-mediated decay. Mutation 928GA (in Proband LS) substituted isoleucine for valine (at residue 310), affected RNA processing, and hence produced a frameshift.
Expression of the mutant LAMP2 protein was assessed by Western blotting of protein extracts probed with antibodies to LAMP2 (Figure 2) and antibodies to GAPDH (data not shown). Protein extracts from the lymphocytes of Probands MFE and CZ (mutations IVS1+1GT and IVS6+1_4delGTGA, respectively) contained a nearly full-length LAMP2 protein (100 kD), whereas protein extract from the lymphocytes and fibroblasts of Proband SS (mutation IVS1–2AG) did not react with LAMP2 antibodies.
|
Five of six probands with LAMP2 mutations were male. One proband had a family history of heart disease. None had mental retardation or overt neurologic or musculoskeletal deficits. Two male probands had histories of attention-deficit disorder and mild behavioral problems; both were taking psychoactive medications.
One asymptomatic proband came to medical attention because of an abnormal electrocardiogram. The other five probands presented with cardiac symptoms typically seen in hypertrophic cardiomyopathy, including chest pain, palpitations, syncope, and, in one, cardiac arrest. The onset of symptoms occurred between the ages of 8 and 15 years, younger than average for patients with mutations of the sarcomere-protein gene or PRKAG2 gene (33±17 years and 31±15 years, respectively) (Table 1).
|
10 mm) was found without pulmonary disease in three probands. At the time of initial clinical presentation, all probands had normal left ventricular function and ejection fractions of 60 percent or more. Twelve-lead electrocardiograms were strikingly abnormal in all probands. Left ventricular voltage was markedly increased and significantly greater (P<0.001) than in patients with sarcomere-protein gene mutations or PRKAG2 mutations (Table 1). In five probands, electrocardiograms showed ventricular preexcitation patterns with short PR intervals and delta waves (Figure 1B). Electrophysiological studies in three persons showed accessory atrioventricular connections; two had supraventricular arrhythmias, atrial fibrillation, or both, that required radiofrequency ablation.
The identification of LAMP2 mutations prompted analyses of serum chemistry. Creatine kinase and alanine aminotransferase levels were elevated by a factor of two or more in four of the six probands, and organ-specific enzyme isoforms indicated cardiac, musculoskeletal, and liver involvement. Serum levels of these enzymes were normal in the only female proband and in one male proband with a mosaic LAMP2 mutation.
Cardiac function progressively deteriorated during a six-year period in Family Member CZ III-1,25 and he died at the age of 22 years while awaiting heart transplantation. A pathological study of his heart (Figure 2A through 2D) showed marked cardiomegaly (weight, 1266 g) and diffuse hypertrophy. Histopathological examination showed myocyte hypertrophy and prominent interstitial fibrosis. Enlarged cardiomyocytes had extensive sarcoplasmic vacuolation with a spiderweb-like appearance; some had large, polymorphic, periodic acid–Schiff–positive perinuclear inclusions. LAMP2 antibodies reacted with the inclusions but lacked the lysosomal perinuclear granular pattern found in normal myocardium. Electron microscopy showed that some sarcoplasmic vacuoles were empty, without recognizable membranes, whereas other vacuoles contained inclusions consisting of amorphous, osmophilic, and focally granular material of variable density. Partially degraded vacuolar contents have been observed in specimens from persons with Danon's disease.12,15
LAMP2 Mutations in Family Members
In Families CZ and MFE, the X-linked LAMP2 mutation was maternally transmitted; one additional man (Family Member CZ III-2) and seven women also carried these LAMP2 mutations (Figure 1). Cardiac disease occurred only in Family Member CZ III-2. When he was 16 years old, his electrocardiogram showed prominent voltage and ventricular preexcitation. His echocardiogram from the same period was normal except for asynchronous contraction, which was attributed to preexcitation. Reevaluation at the age of 22 (after genetic diagnosis) demonstrated substantial hypertrophy and reduced function. He remains asymptomatic.
Although none of seven surviving female family members (14 to 46 years of age) with LAMP2 mutations had cardiac symptoms or abnormal cardiac studies, one woman (Family Member CZ I-2) died from congestive heart failure at the age of 44. She probably carried the LAMP2 mutation, given that both of her daughters are genetically affected and that her husband does not have cardiac disease (Family Member CZ-1).
No LAMP2 mutations were found in the family members of four probands. LAMP2 gene sequences were normal in the mothers of three affected male probands (SS, LS, and NR), and mitochondrial DNA polymorphisms confirmed biologic maternity (data not shown). Genetic studies of Proband LS demonstrated mosaicism: both mutant and wild-type LAMP2 sequences were identified despite a normal XY karyotype (not shown). Single nucleotide polymorphisms indicated that he had inherited the X chromosome and normal LAMP2 gene from his unaffected mother. One female proband (FI) with a normal karyotype also carried a LAMP2 mutation. Haplotype analyses (not shown) demonstrated that she had inherited one X chromosome from each genetically unaffected parent. We conclude that these LAMP2 mutations (in four probands: SS, LS, NR, and FI) arose spontaneously.
Discussion
Defects in the enzymes involved in the metabolism of muscle glycogen typically cause systemic disease26 and often involve the heart (Figure 3). Our study demonstrates that cardiac disease can be the initial and predominant manifestation of defects in human glycogen metabolism. Three of 75 persons in whom hypertrophic cardiomyopathy was diagnosed by echocardiography had cardiac-glycogen–storage disorders caused by LAMP2 or PRKAG2 mutations. These gene defects, like sarcomere-gene mutations, were associated with prominent left ventricular hypertrophy, but in addition, electrophysiological abnormalities were present.
|
Previously reported LAMP2 mutations caused a variety of manifestations that are characteristic of Danon's disease.12,13,14 No probands in our series had clinically important neurologic disease, although psychological issues recognized in two young male subjects were attributed to attention-deficit disorder and an adolescent response to cardiac disease. None had overt muscle weakness, wasting, or myopathic symptoms; all had exercise restrictions because of the diagnosis of hypertrophic cardiomyopathy.
Gene dosage probably accounts for the different clinical consequences of X-linked LAMP2 mutations in men as compared with women, although unusual cardiac diseases were found in two female carriers of LAMP2 mutations. Perhaps X-inactivation sufficiently extinguished normal LAMP2 gene expression to contribute to or cause cardiomyopathy in female Proband FI and adult-onset heart failure in Family Members CZ I-2 and LO I-2 (Figure 1A). Gene dosage also contributed to clinical expression in men with identical LAMP2 mutations. During these studies, a male proband (LO in Figure 1A) with classical Danon's disease (mental retardation and musculoskeletal weakness, with protean findings on muscle biopsy) was referred for genetic analyses. Proband LO and his mother were found to have the same LAMP2 mutation (928GA) as Proband LS, although these families are genetically unrelated (data not shown). Remarkably, Proband LO was hemizygous for the mutation, whereas mosaicism in Proband LS caused expression of both normal and mutant LAMP2 alleles. We presume that some normal LAMP2 protein in Proband LS accounted for the predominance of cardiac disease in comparison with multisystem Danon's disease in Proband LO.
The partial function of mutant LAMP2 proteins may also account for the cardiac form of Danon's disease, as compared with systemic Danon's disease. The musculoskeletal pathology of Danon's disease13,14 indicates a complete absence of LAMP2 immunoreactivity, whereas we found stable LAMP2 RNA and immunoreactive LAMP2 protein in lymphocytes from Probands CZ and MFE (Figure 2E). These mutant proteins may function sufficiently to limit disease in some, but not all, tissues.
Inclusion of LAMP2 and PRKAG2 mutations in the differential diagnosis of unexplained left ventricular hypertrophy is important for patient care. These mutations increase the risk of arrhythmias, as shown by preexcitation patterns on electrocardiograms, by accessory pathways on electrophysiological evaluation, and by patients' histories of supraventricular tachyarrhythmias, syncopal episodes, and sudden death. The mechanism for ventricular preexcitation is incompletely understood; however, a mouse model of one human PRKAG2 mutation shows disruption of the anulus fibrosus by glycogen-filled myocytes, thereby allowing atrioventricular activation that bypasses the atrioventricular node.10,27 Although LAMP2 mutations accumulate glycogen in lysosomes12 and PRKAG2 mutations accumulate glycogen throughout the myocyte,10 it is likely that there is a common mechanism for ventricular preexcitation in both glycogen-storage cardiomyopathies. We suggest that patients with unexplained left ventricular hypertrophy and preexcitation patterns on electrocardiograms undergo clinical and genetic evaluation for glycogen storage disease (Figure 4).
|
Although clinical evaluations may help to distinguish these disorders, genetic analyses can definitively establish the cause of unexplained left ventricular hypertrophy (Figure 4). This information is critical for determining the appropriate strategies of treatment and for defining genetic risk in family members. Applying the major advances in DNA sequencing to medicine has made gene-based diagnosis not only feasible, but a clinical reality.
Supported by the Howard Hughes Medical Institute and by the National Heart, Lung, and Blood Institute, National Institutes of Health.
We are indebted to Ms. Barbara A. McDonough, R.N., Ms. Susan A. Casey, R.N., Barbara A. Mostella, R.N., Dr. Brian W. Gross, and Dr. Eloisa Arbustini for their invaluable assistance in collecting clinical material and to Ms. Catherine M. Duffy, Ms. Susanne Bartlett, Mr. Howard Mulhern, and Mr. James Edwards for their technical assistance.
Source Information
From the Division of Cardiology, Brigham and Women's Hospital (M.A., C.E.S.), and the Department of Genetics, Harvard Medical School and Howard Hughes Medical Institute (M.A., J.M.G., C.E.S., J.G.S.) — all in Boston; the Hypertrophic Cardiomyopathy Center, Minneapolis Heart Institute Foundation, Minneapolis (B.J.M.); the Department of Pediatrics, Division of Pediatric Cardiology, University of Alabama at Birmingham, Birmingham (W.H.J.); the Children's Heart Program of South Carolina, Medical University of South Carolina, Charleston (J.P.S.); the Department of Pathology, Children's Hospital and Harvard Medical School, Boston (A.R.P.-A.); the Department of Cardiology, Galleria de Genova, Genova, Italy (P.S.); the Children's Heart Clinic, Minneapolis (G.B.W.); and the Division of Pediatric Cardiology, Duke University Medical Center, Durham, N.C. (R.J.K.).
Address reprint requests to Dr. J.G. Seidman at the Department of Genetics, NRB Rm. 256, Harvard Medical School, 77 Ave. Louis Pasteur, Boston, MA 02115, or at seidman{at}genetics.med.harvard.edu.
References
| |||||||||||||||||||||||||||||||||||||||||||||||
Related Letters:
Glycogen Storage Diseases and Cardiomyopathy
Roos J. C.P., Cox T. M., Arad M., Maron B. J., Seidman J.G.
Extract |
Full Text |
PDF
N Engl J Med 2005;
352:2553, Jun 16, 2005.
Correspondence
This article has been cited by other articles:
HOME | SUBSCRIBE | SEARCH | CURRENT ISSUE | PAST ISSUES | COLLECTIONS | PRIVACY | TERMS OF USE | HELP | beta.nejm.org Comments and questions? Please contact us. The New England Journal of Medicine is owned, published, and copyrighted © 2009 Massachusetts Medical Society. All rights reserved. |
Previous
denotes deletion, AA amino acids, FS frameshift, Ter termination, and SCD sudden death from cardiac causes. An asterisk indicates that the serum creatine kinase and alanine aminotransferase levels in the subject were more than 1.5 times the normal level. A dagger indicates that both mutant and normal LAMP2 sequences were present in Subject LS II-2. An electrocardiogram from the proband in Family LS (Subject II-2) (Panel B) demonstrates a short PR interval, delta waves, and prominent voltage (R waves >50 mm in precordial leads V4 and V5).The parasternal, long-axis view of an echocardiogram from the proband in Family SS (Subject II-2) (Panel C) shows marked hypertrophy (maximal left ventricular wall thickness, >35 mm) involving the interventricular septum and posterior left ventricular wall.
