A De Novo Mutation in α-Tropomyosin That Causes Hypertrophic Cardiomyopathy
Background Two missense mutations in the gene for α-tropomyosin have been described that segregate with hypertrophic cardiomyopathy in single families. To confirm that these mutations are the cause of the disease, we have investigated the origins of one of these mutations, Asp175Asn, in a third and unrelated family.
Methods and Results The presence or absence of an α-tropomyosin mutation and the haplotypes of the flanking chromosomal regions were determined for members of a family with hypertrophic cardiomyopathy. Haplotypes were constructed by use of an intragenic polymorphism and 10 flanking polymorphisms spanning a region of 35 centimorgans. The Asp175Asn missense mutation was present in the proband and his two affected offspring but not in any of the proband’s three siblings. Although both parents were deceased, the haplotypes of the four parental chromosomes could be reconstructed. One parental chromosome was transmitted to two offspring: one bearing the Asp175Asn mutation (the affected proband) and one clinically unaffected sibling who lacked the α-tropomyosin mutation. Thus, the Asp175Asn mutation must have arisen de novo.
Conclusions De novo mutations in the α-tropomyosin gene can result in hypertrophic cardiomyopathy that may appear to be sporadic but in subsequent generations gives rise to familial disease. Individuals with sporadic hypertrophic cardiomyopathy should be advised of the risk of transmission to offspring. In addition, these findings provide the strongest genetic evidence that mutations in the α-tropomyosin gene are directly responsible for hypertrophic cardiomyopathy.
Two different α-tropomyosin gene mutations have been identified in families with hypertrophic cardiomyopathy. Each of the missense mutations (Asp175Asn and Glu180Gly) affects a conserved residue and has not been found in the genomes of unaffected individuals.1 These data suggest that α-tropomyosin gene mutations are a cause of familial hypertrophic cardiomyopathy. However, some features of the mutations in the α-tropomyosin gene contrast with those of mutations in the other disease genes for hypertrophic cardiomyopathy. Unlike the β-cardiac myosin heavy chain2 and cardiac troponin T genes,1 the α-tropomyosin gene is expressed ubiquitously, yet the disease phenotype is limited to cardiac muscle. In addition, screening of the α-tropomyosin gene sequences has revealed only one further mutation in more than 120 unrelated individuals with hypertrophic cardiomyopathy; an unrelated proband (DB) was identified with the Asp175Asn mutation.3 Because of the remote possibility that these α-tropomyosin mutations are only linked polymorphisms, further evidence demonstrating a cause-effect relation with hypertrophic cardiomyopathy would be desirable.
Spontaneously arising, or de novo, germ-line mutations in a postulated disease gene are sufficiently rare that when they coincide with the new development of an inherited trait in an individual, this provides strong evidence that the mutation causes the trait.4 5 We therefore have studied the proband with the Asp175Asn mutation, along r with all available family members (family DB), to determine whether the α-tropomyosin gene mutation arose de novo. Such analyses typically require the availability of both parents of an individual with sporadic disease to demonstrate that the chromosome on which the mutation arises was previously normal and to exclude nonpaternity as an alternative explanation. Although the proband’s parents were not available for study, by constructing chromosomal haplotypes in the extended pedigree, we are able to demonstrate that the Asp175Asn mutation the proband carries did arise de novo. Demonstration of a de novo occurrence in a family with hypertrophic cardiomyopathy of a mutation previously seen to segregate with disease provides the strongest evidence that the α-tropomyosin gene is itself a disease gene for hypertrophic cardiomyopathy.
Clinical evaluation of family members was performed as described,3 6 with history, physical examination, 12-lead ECG, and two-dimensional echocardiogram. A venous blood sample was obtained for extraction of DNA after informed consent was obtained in accordance with the Brigham and Women’s Hospital Human Subjects Committee.
Detection of the Asp175Asn Mutation
Exon 5 of the α-tropomyosin gene was amplified from genomic DNA of each family member by polymerase chain reaction (PCR) as described.1 The presence or absence of the G→A transition that encodes the Asp175Asn mutation was determined in the PCR-amplified DNA of each individual by cycle sequencing.1
Analysis of Short Tandem Repeat Polymorphisms
Short tandem repeat (STR) polymorphisms were typed to identify each chromosome by definition of the alleles present at each locus (ie, to define the haplotype of each chromosome). The previously described intragenic STR polymorphism, HTMαCA,1 was used, along with 10 flanking polymorphic markers. These polymorphisms were selected for their high heterozygosity and appropriate map locations as indicated by our analyses7 and published linkage maps of chromosome 15.8 9 10 11 Alleles were typed by PCR amplification with end labeling of the forward primer, followed by denaturing PAGE, as described.1 All STR polymorphisms were run against sequencing ladders to allow precise identification of individual alleles by size.
Construction of Haplotypes
STR polymorphisms were typed at increasing genetic distances from the α-tropomyosin gene to delineate the extent of the chromosomal region inherited without recombination in family DB. Once the alleles carried by each family member at each polymorphism were identified, the alleles specific to each parental chromosome were deduced by inspection of inheritance patterns in nuclear families. The order assumed for the STR markers was based on published map data8 9 10 11 and the recombinations observed in our own data (Reference 6 and unpublished data). This order was (from centromeric to telomeric) D15S118, THBS, CYP19, D15S126, D15S209, D15S98, D15S117, HTMαCA, D15S159, D15S108, D15S125. STR loci within two clusters (D15S98-D15S117 and HTMαCA-D15S159-D15S108) were inseparable from each other, and the order used was assigned arbitrarily. Because no recombination events were seen between any of these STR loci in the four chromosomes inherited from the grandparents, the actual order is not of consequence for these analyses.
Confirmation of Paternity
The possibility that individuals with the same haplotype had inherited these chromosomes from different fathers was assessed by estimation of the probability of an identical match by chance. To minimize the unknown effects of linkage disequilibrium, probabilities were calculated only for loci separated by a minimum of 3 centimorgans (cM). Allele frequencies were taken from published data.
All available members of family DB (Fig 1A⇓) were clinically evaluated. Three members of family DB—the proband, III-2, and his two sons, IV-1 and IV-3—were considered to have hypertrophic cardiomyopathy on the basis of left ventricular hypertrophy detected by echocardiogram. Clinical findings in these three individuals are reported in the Table⇓. No other members of the family had evidence of cardiac hypertrophy on echocardiogram. Both of the proband’s parents were deceased; autopsies were not performed. Individual III-5 died of noncardiac causes in her 60s.
The aspartate-to-asparagine substitution in exon 5 of the α-tropomyosin gene (Asp175Asn, numbered according to Reference 15), which was previously identified in the proband, results from a guanine-to-adenine transition at nucleotide residue 579 (G579A). Direct sequencing of exon 5 of the α-tropomyosin gene was performed on amplified DNA from all available first-degree relatives. The G579A mutation was present in the proband and his two affected sons and was absent in all other family members (Fig 2⇓). Thus, the diagnoses of disease status based on genetic analyses were concordant with the clinical studies. Of particular importance for this study was the finding that individual III-4 (the proband’s brother) did not carry the Asp175Asn mutation.
Haplotype analysis was performed to define the extent of the chromosome 15 region surrounding the α-tropomyosin gene that was shared by each family member. Analysis of the alleles present at each STR locus in the nuclear family comprising the proband, his wife, and his children (III-2, III-3, IV-1, IV-2, and IV-3) indicates that the three affected individuals share the same haplotype: 22.214.171.124.126.96.36.199.1.3.5, designated haplotype A (Fig 1A⇑). Analysis of the alleles present in the proband’s three siblings (III-4, III-5, and III-7) indicates the other haplotypes (B, C, and D) that together identify each of the four chromosomes inherited from the grandparents (II-3 and II-4). No recombination events affect the inheritance of these chromosomes within the region identified by the 11 STR loci (approximately 35 cM,8 9 10 11 Fig 1B⇑). Only the configuration of haplotypes in Fig 1⇑ fitted the observed data; no other permutation was compatible with the inheritance patterns seen.
Haplotype A, which identifies the chromosome carrying the Asp175Asn mutation in the three genetically and clinically affected individuals, was also inherited by individual III-4 (Fig 1A⇑). Individual III-4 is an unaffected male who does not have the Asp175Asn mutation. One possible explanation for the differences at the Asp175 residue between individuals III-2 and III-4 is that they have different fathers who by chance have transmitted an identical haplotype. The probability that different fathers might transmit identical alleles at each marker was estimated from the published frequencies of the specific alleles. Because particular combinations of alleles at closely linked loci might be coinherited, we used data from polymorphisms spaced across the region and excluded data from loci tightly linked to each other (Fig 1B⇑; see the “Methods” section). Based on the frequencies of alleles at eight loci spaced across 35 cM, the chance that individuals III-2 and III-4 would inherit the identical haplotype if they had different fathers is approximately 1 in 1 million. Further, 10 very highly polymorphic STR markers from other chromosomes were typed and did not reveal evidence of nonpaternity (data not shown). We conclude that nonpaternity does not explain the discordant finding of a shared haplotype (A) in two brothers but an α-tropomyosin mutation in only one. Because haplotypes A and B together identify one parent (either individual II-3 or II-4) and haplotype B is present in a relative on the paternal side, we can further conclude that the chromosome with haplotype A was inherited from the proband’s father, individual II-3.
We have demonstrated by haplotype analysis that individual II-3 passed a normal copy of the α-tropomyosin gene to an unaffected child (III-4) and, on the same segment of chromosome 15, a mutated copy to another son (III-2), who developed hypertrophic cardiomyopathy. This observation demonstrates that a de novo α-tropomyosin gene mutation, Asp175Asn, occurred in individual III-2, who has hypertrophic cardiomyopathy (Fig 1A⇑). This finding shows conclusively that this mutation alone is sufficient to cause the disease and excludes the possibility that another abnormality in this or a linked gene is responsible.
The Asp175Asn mutation in the α-tropomyosin gene of individual III-2 was present in three different tissues: cardiac myocytes, lymphocytes, and gametes. This suggests that the mutation arose either in development of the gamete (ie, during spermatogenesis in the father, II-3) or very early in embryonic development. We cannot determine whether the proband’s father (individual II-3) exhibited germ-line mosaicism for this mutation; individual II-3 is deceased, and only the proband and his unaffected brother inherited the relevant copy of the α-tropomyosin gene (haplotype A, Fig 1⇑). Individual III-2 did transmit the mutation to his two sons (IV-1 and IV-3). Three examples of de novo germ-line mutations in the β-cardiac myosin heavy chain gene have also been reported in individuals with apparent sporadic hypertrophic cardiomyopathy.16 17 The finding of a de novo mutation in the germ line of proband III-3 again raises concerns for individuals diagnosed with sporadic hypertrophic cardiomyopathy. These individuals should be advised that they may transmit the disease to their offspring.
Because mutations in many genes can cause hypertrophic cardiomyopathy (β-cardiac myosin heavy chain,2 cardiac troponin T,1 α-tropomyosin, and an unknown gene on chromosome 1118 ), linkage analysis is commonly the first step in the genetic investigation of a newly ascertained family. In families of sufficient size, this is an efficient way to target the search for a mutation to the appropriate disease gene. However, in a pedigree with an unrecognized instance of new mutation, linkage data would suggest crossovers, leading to a false exclusion of the disease gene. Recognition of the possibility of a de novo mutation will allow correct interpretation of the linkage data in a kindred such as family DB.
Two of three known α-tropomyosin gene mutations involve independent occurrences of the same nucleotide alteration. While this may reflect an increased tendency to mutation at this particular residue (a G→A transition at a CpG dinucleotide), an alternative explanation is that the Asp 175 residue is of specific importance to α-tropomyosin function in the heart. Analysis of the functional consequences of this mutation may explain the cardiac-specific phenotype. The demonstration that the Asp175Asn mutation has occurred independently in two families with hypertrophic cardiomyopathy is analogous to the finding of recurrent identical mutation in the β-cardiac myosin heavy chain and cardiac troponin T genes.3 19 Thus, the majority of disease-causing mutations in individuals and families with hypertrophic cardiomyopathy have arisen de novo rather than from a common ancestor. This is consistent with the negative selective pressure expected for a condition associated with premature mortality.
This work was supported by grants from the British Heart Foundation (Dr Watkins is the recipient of a British Heart Foundation clinical scientist fellowship), Telethon-Italia grant No. 380 and Consiglio Nazionale delle Richerche (Dr Coviello), Telethon-Italia grant No. 600 (Dr Spirito), the NIH, the Howard Hughes Medical Foundation, and Bristol-Myers Squibb Co (Drs J.G. Seidman and C.E. Seidman). We are grateful to family members for participating in these studies.
- Received January 3, 1995.
- Accepted January 13, 1995.
- Copyright © 1995 by American Heart Association
Thierfelder L, MacRae C, Watkins H, Tomfordhe J, Williams M, McKenna WJ, Bohm K, Noeske G, Schlepper M, Bowcock A, Vosberg H-P, Seidman JG, Seidman CE. A familial hypertrophic cardiomyopathy locus maps to chromosome 15q2. Proc Natl Acad Sci U S A. 1993;90:6270-6274.
Bowcock AM, Barnes RI, Weiss LE, Tomfohrde J, Dracopoli ND, Hudson T, Sarfarazi M, Tsipouras P, Jenkins T, Vergnaud G. A comprehensive genetic linkage map of the human genome: chromosome 15. Science. 1992;258:67-86.
Polymeropoulos MH, Xiao H, Rath DS, Merril CR. Dinucleotide repeat polymorphism at the human thrombospondin gene (THBS1). Nucleic Acids Res. 1991;18:7469.
Polymeropoulos MH, Xiao H, Rath DS, Merril CR. Tetranucleotide repeat polymorphism at the human aromatase cytochrome P-450 gene (CYP19). Nucleic Acids Res. 1991;19:195.
Macleod AR, Gooding C. Human hTM alpha gene: expression in muscle and nonmuscle tissue. Mol Cell Biol. 1988;8:433-440.
Watkins H, Thierfelder L, Hwang D-S, McKenna WJ, Seidman JG, Seidman CE. Sporadic hypertrophic cardiomyopathy due to de novo myosin mutations. J Clin Invest. 1992;90:1666-1671.
Greve G, Bachinski L, Friedman DL, Czernuzewicz G, Anan R, Towbin J, Seidman CE, Roberts R. A de novo mutation of the β-myosin heavy chain gene occurring with germ-line mosaicism induces HCM in subsequent generations and is expressed in the myocardial β-myosin heavy chain protein. Hum Mol Genet. 1995;3:2073-2075.
Carrier L, Hengstenberg C, Beckmann JS, Guicheny P, Dufour C, Bercovici J, Dausse E, Berebbi-Bertrand I, Wisnewsky C, Pulvenis D, Fetler L, Vignal A, Weissenbach J, Hillaire D, Feingold J, Bouhour J-B, Hagege A, Desnos M, Isnard R, Duborg O, Komajda M, Schwartz K. Mapping of a novel gene for familial hypertrophic cardiomyopathy to chromosome 11. Nat Genet. 1993;4:311-313.