Author + information
- Received August 8, 1996
- Revision received October 31, 1996
- Accepted November 12, 1996
- Published online March 1, 1997.
- Johanna C Moolman, MSA,
- Valerie A Corfield, PhDA,* (, )
- Berthold Posen, MDB,
- Kholiswa Ngumbela, BSc(Hons)A,
- Christine Seidman, MDC,
- Paul A Brink, MDB and
- Hugh Watkins, MD, PhDCD
- ↵*Dr. Valerie A. Corfield, University of Stellenbosch/Medical Research Council Centre for Molecular and Cellular Biology, P.O. Box 19063, Tygerberg 7505, Republic of South Africa.
Objectives. This study was designed to verify initial observations of the clinical and prognostic features of hypertrophic cardiomyopathy caused by cardiac troponin T gene mutations.
Background. The most common cause of sudden cardiac death in the young is hypertrophic cardiomyopathy, which is usually familial. Mutations causing familial hypertrophic cardiomyopathy have been identified in a number of contractile protein genes, raising the possibility of genetic screening for subjects at risk. A previous report suggested that mutations in the cardiac troponin T gene were notable because they were associated with a particularly poor prognosis but only mild hypertrophy. Given the variability of some genotype:phenotype correlations, further analysis of cardiac troponin T mutations has been a priority.
Methods. Deoxyribonucleic acid from subjects with hypertrophic cardiomyopathy was screened for cardiac troponin T mutations using a ribonuclease protection assay. Polymerase chain reaction-based detection of a novel mutation was used to genotype members of two affected pedigrees. Gene carriers were examined by echocardiography and electrocardiology, and a family history was obtained.
Results. A novel cardiac troponin T gene mutation, arginine 92 tryptophan, was identified in 19 of 48 members of two affected pedigrees. The clinical phenotype was characterized by minimal hypertrophy (mean [±SD] maximal ventricular wall thickness 11.3 ± 5.4 mm) and low disease penetrance by clinical criteria (40% by echocardiography) but a high incidence of sudden cardiac death (mean age 17 ± 9 years).
Conclusions. These data support the observation that apparently diverse cardiac troponin T gene mutations produce a consistent disease phenotype. Because this is one of poor prognosis, despite deceptively mild or undetectable hypertrophy, genotyping at this locus may be particularly informative in patient management and counseling.
(J Am Coll Cardiol 1997;29:549–55)
The annual incidence of sudden death is reported to be between 19 and 159/100,000 male subjects, representing up to 50% of all deaths in industrialized countries (); it is believed that sudden cardiac death comprises a major component of this figure. This form of death brings particular sorrow, especially when it occurs unexpectedly in healthy young subjects. One of the commonest causes of sudden cardiac death in the adolescent to 35-year old age group is hypertrophic cardiomyopathy ([2, 3]). The majority of hypertrophic cardiomyopathy is familial, in which case relatives of the deceased are also at risk of sudden cardiac death (). Frequently, the questions asked are whether the death was avoidable and whether it is possible to identify other high risk family members. The clinical heterogeneity and incomplete penetrance that are hallmarks of familial hypertrophic cardiomyopathy have confounded answers to these queries (). There is no consistent correlation of reported risk indicators, such as substantial left ventricular hypertrophy, recurrent syncope or nonsustained ventricular tachycardia on Holter monitoring, and the occurrence of sudden cardiac death ([6–8]).
The identification of familial hypertrophic cardiomyopathy-causing mutations in the genes encoding sarcomeric contractile proteins, namely the beta-myosin heavy chain (), cardiac troponin T (), alpha-tropomyosin () and cardiac myosin-binding protein C ([11, 12]), offers another approach to defining risk prognosticators. Several genotype:phenotype evaluations of specific beta-myosin heavy chain gene mutations and the associated spectrum of clinical features, including the risk of sudden cardiac death, have been reported in familial hypertrophic cardiomyopathy ([13, 14]). The presence of symptomatic gene carriers (incomplete penetrance), which may exceed 25% in some families, has been a noticeable feature of the studies (). As a result of these assessments, mutations have been categorized as malignant or benign (). However, the correlations have not proved invariable; reports of discrepancies in beta-myosin heavy chain gene-caused familial hypertrophic cardiomyopathy genotype:phenotype correlations exist ([13, 15]). In addition, the value of genetic screening may also be limited in beta-myosin heavy chain gene-caused familial hypertrophic cardiomyopathy because cardiac hypertrophy is generally clinically apparent in subjects with malignant mutations and because the large size of the gene hampers mutation detections ().
Recent findings () have focused attention on the role of cardiac troponin T gene mutations, which may account for at least 15% of cases of familial hypertrophic cardiomyopathy. The eight different cardiac troponin T gene mutations described fall into three classes, namely, missense, in-frame deletion and splice donor site mutations. The only reported study of genotype:phenotype correlations to date () showed that the clinical features associated with all these cardiac troponin T gene mutations were remarkably similar despite the distinct types of gene defects that would be expected to produce different classes of dysfunctional product. The average maximal ventricular wall thickness associated with cardiac troponin T gene mutations was considerably less than that recorded in beta-myosin heavy chain gene-caused familial hypertrophic cardiomyopathy, with an associated reduction of disease penetrance. However, the number of disease-related and sudden cardiac deaths attributable to cardiac troponin T gene mutations was high, similar to that seen with the most malignant beta-myosin heavy chain gene mutations.
If frequent sudden cardiac death, in the setting of mild to undetectable hypertrophy and the presence of asymptomatic gene carriers, is characteristic of cardiac troponin T gene-caused familial hypertrophic cardiomyopathy, then mutation detection in this gene may be of particular clinical relevance. Therefore, further genotype:phenotype studies in large families in different geographical areas are needed to evaluate the role of other cardiac troponin T gene mutations in the development of familial hypertrophic cardiomyopathy and to assess their prognostic implications.
In the present study, we describe genotype:phenotype correlations of familial hypertrophic cardiomyopathy resulting from a novel cardiac troponin T gene, arginine 92 tryptophan mutation, in two South African families of mixed racial descent, revealing the insidious nature of defects in this gene.
1.1 Patient study group and clinical evaluation.
The study was performed according to institutional ethics committee approval. Informed consent was obtained from subjects or from parents of minors participating in the study. Pedigree 0, reported in a previous study as pedigree 100 (), and pedigree 9, described in the present study, both of mixed racial descent, were established from familial hypertrophic cardiomyopathy-affected probands in whom no beta-myosin heavy chain gene mutations had been detected. Members of the two pedigrees were identified by the prefix 0 or 9, followed by generation and individual numbers. Family members were traced, and the history of disease-related and sudden cardiac deaths was established. All available members of pedigree 0 had been examined clinically during the earlier study. Among newly traced members of pedigree 0 and pedigree 9, only those genotypically positive for the arginine codon 92 tryptophan mutation were examined clinically.
Clinical examination was performed as previously described (). Briefly, echocardiographic diagnosis of familial hypertrophic cardiomyopathy was made in the presence of a maximal ventricular wall thickness ≥13 mm. In children, diagnosis was made with reference to age-adjusted tables. Electrocardiographic diagnosis of familial hypertrophic cardiomyopathy was based on the presence of left ventricular hypertrophy or abnormal Q waves. Other echocardiographic and electrocardiographic (ECG) changes previously described in familial hypertrophic cardiomyopathy were noted for each patient.
1.2 Source of DNA and DNA extraction.
After informed consent was obtained, peripheral blood for genotypic analysis was collected from all probands and family members entered into the study, and DNA was extracted as previously described ().
1.3 Mutation detection.
Linkage analysis excluded a beta-myosin heavy chain gene mutation as the cause of familial hypertrophic cardiomyopathy in pedigree 0 (); support for linkage was obtained with markers linked to the cardiac troponin T gene locus (data not shown). Consequently, cardiac troponin T messenger RNAs derived from two affected members of pedigree 0 were screened by ribonuclease protection assays, as previously described (). Abnormal cleavage products (indicative of sequence differences) were sequenced according to standard protocols.
Polymerase chain reaction-based detection of the arginine 92 tryptophan mutation, which resulted in the loss of an MspI restriction enzyme site in exon 9 of the cardiac troponin T gene, was used to identify genotypically positive relatives in pedigrees 0 and 9. Exons 8 and 9, with the intervening intron, were polymerase chain reaction amplified using primers 214F and 393R (using the numbering system of Mesnard et al. []) to generate a 498 base pair product for digestion with MspI (), according to previously described methods ().
2.1 Mutation screening and detection.
Linkage analysis in pedigree 0 indicated that familial hypertrophic cardiomyopathy in this family was linked to the region of the cardiac troponin T gene. Ribonuclease protection assays of cardiac troponin T complementary DNA acid identified a sequence variant in exon 9 of two affected subjects from pedigree 0. Nucleotide sequencing revealed a previously undescribed cytosine to thymine transition at nucleotide 286, the first base of codon 92. This modification results in the substitution of arginine by tryptophan at this position in the protein, fortuitously resulting in the loss of an MspI site and thus allowing polymerase chain reaction-based restriction enzyme mutation detection. The presence of three bands in the digested product of polymerase chain reaction-amplified exon 9 of an affected subject would indicate the presence of both the normal and mutated allele.
Subsequently, the MspI-based detection assay also identified heterozygous loss of the restriction enzyme recognition sequence in exon 9 of the proband of pedigree 9. Nucleotide sequencing revealed the same cytosine to thymine transition in the first base of codon 92 that had been detected in the proband of pedigree 0. This mutation was not detected in 100 unrelated, unaffected control subjects.
2.2 Disease penetrance and clinical features of the arginine 92 tryptophan mutation.
Nineteen of the 48 genotyped members of the two pedigrees carried the arginine 92 tryptophan mutation (Fig. 1), and 18 of these were examined clinically. In addition, subjects 0.II.7 and 0.II.14 were obligate gene carriers by virtue of having transmitted the mutation to some of their children; individual 0.IV.2 was genotypically positive, but had not been examined clinically. The standard echocardiographic diagnostic criterion of a ventricular wall thickness ≥13 mm was met in only six subjects, shown in bold in Table 1. The average maximal wall thickness, calculated for subjects ≥16 years in both pedigrees, was 11.3 ± 5.4 mm (mean ± SD, n = 14), within the normal range. Electrocardiographic abnormalities characteristic of familial hypertrophic cardiomyopathy were also seen in four of the six persons with overt hypertrophy. In the remaining two, subject 0.III.15 exhibited normal ECG results, whereas the abnormalities present in subject 9.III.1 were not specific for hypertrophy. An additional six family members who displayed no echocardiographic evidence of familial hypertrophic cardiomyopathy possessed characteristic ECG abnormalities (Table 1). Another six subjects were genotypically positive but showed no evidence of hypertrophy and had either normal ECG patterns or nonspecific abnormalities (Table 1). Three of the gene carriers with normal clinical studies were >16 years old, namely subjects 0.II.18, 0.III.20 and 9.III.5, who were 47, 27 and 28 years old, respectively.
The disease penetrance in subjects >16 years old calculated for the combined pedigrees was 40% by echocardiography when the standard diagnostic criterion of ventricular wall thickness was applied and 66% when characteristic ECG features were applied. The penetrance was 80% on the basis of both echocardiographic and ECG abnormalities characteristic of the disease.
The range of symptoms varied, with presyncope or syncope expressed in three subjects, dyspnea in seven, angina in three and palpitations in three (data not shown). Sudden cardiac death at a young age was frequent in pedigree 0, occurring in six subjects (Fig. 1) and possibly two other young subjects whose deaths cannot be directly attributable to the disease. In subjects 0.II.5 and 9.III.7, the cause of death was a tumor and a motor vehicle accident, respectively. Three subjects died of a cerebral vascular accident, at ages ranging from 38 to 62 years. Of these, 0.II.7 was an obligate heterozygote, 0.II.20 was unlikely to be a gene carrier because he had a number of unaffected children, and 0.II.2, who had marked concentric hypertrophy and long-standing hypertension, was genotyped after death and did not carry the mutation. In pedigree 9, one sudden cardiac death at age 28 years, with characteristic pathology of familial hypertrophic cardiomyopathy found on post mortem, was reported. The mean age of sudden cardiac death in the two families studied was 17 ± 9 years.
Survival analysis in the two families highlighted the strikingly high incidence of sudden cardiac death in adolescents and young adults (Fig. 2). However, the cumulative mortality rate did not reach 50% in these kindreds (see Discussion). The prognosis for the arginine 92 tryptophan mutation appeared particularly poor for male subjects. Although 14 of the 18 genotypically positive family members were female, all of the sudden cardiac deaths occurred in young male subjects. The cumulative mortality rate for male subjects only was 64% by 28 years.
The clinical features associated with familial hypertrophic cardiomyopathy in the two pedigrees are summarized in Table 2.
3.1 Identification of novel cardiac troponin T gene mutation.
A novel cardiac troponin T gene, arginine 92 tryptophan mutation, was identified and shown to cause familial hypertrophic cardiomyopathy in two pedigrees. The mutation was present in all clinically affected relatives. In addition, a number of clinically unaffected relatives with the mutation were identified, confirming that incomplete penetrance is a feature of familial hypertrophic cardiomyopathy caused by cardiac troponin T gene mutations. The arginine 92 tryptophan mutation was not present in 100 unaffected unrelated subjects of mixed racial ancestry. Previous reports () have described an arginine 92 glutamine mutation in the same cardiac troponin T codon, which invariably codes for a basic residue in vertebrate species. Evidence that the arginine 92 glutamine mutation must have arisen on at least three independent occasions () and the fact that the two mutations occur at a mutationally fragile cytosine:guanine doublet within codon 92, suggest that this is a mutation hotspot (). Identification of regions of the cardiac troponin T gene sequence prone to disease-causing mutation may simplify mutation screening in unrelated families.
3.2 Clinical and prognostic features of cardiac troponin T gene mutations.
Clinical and prognostic evaluations were based on combined data for the two pedigrees because they share the same arginine 92 tryptophan mutation. The latter was characterized by mild, often undetectable, hypertrophy; the mean maximal wall thickness of affected adults was in the normal range (11.3 ± 5.4 mm). This was reflected in the low penetrance of only 40% when based on standard echocardiographic definition and 66% by electrocardiography. However, despite the apparent mildness of symptoms and hypertrophy, the prognosis associated with this mutation was extremely poor, particularly for adolescent and young adult male subjects. The cumulative mortality rate was 34% among both genders, but 64% for male subjects, at the age of 28 years. The combined cumulative mortality rate did not reach 50% in these families. Two explanations may be proposed: 1) The family members studied were of relatively young age, providing little data regarding later life; 2) there were no late deaths due to progressive heart failure in this group. This observation is consistent with data from other cardiac troponin T gene mutations studied, where the survival curves also suggested a plateau after the third decade, and where later deaths from heart failure were rare (). These data confirm earlier observations () that the clinical features of cardiac troponin T gene mutations may be distinguished from those associated with beta-myosin heavy chain gene defects, where malignant mutations are generally associated with clinically obvious disease. In addition, data from this study highlight the possibility that cardiac troponin T gene mutations are less prone to lead to congestive heart failure than are mutations in the beta-myosin heavy chain gene. It is tempting to speculate about the relation between this observation and the noted differences in the severity of hypertrophy.
3.3 Genotype:phenotype correlations in hypertrophic cardiomyopathy.
Initial genotype:phenotype correlations of beta-myosin heavy chain gene-caused familial hypertrophic cardiomyopathy suggested that variations resulting in a change in charge of the substituted amino acid were associated with a poor survival index ([14, 21]). However, the effects of various mutations on prognosis are probably more complex ([22, 23]). In particular, it has not proved possible to predict the effect of a mutation by extrapolation from phenotypes of apparently similar mutations. For example, a cytosine:guanine mutation hotspot responsible for familial hypertrophic cardiomyopathy has also been described () in the beta-myosin heavy chain gene, where the arginine encoded by codon 403 may be substituted by either tryptophan or glutamine. In both mutations, a basic residue is replaced by an uncharged one. However, whereas the beta-myosin heavy chain gene arginine 403 glutamine mutation has a malignant phenotype, the beta-myosin heavy chain gene arginine 403 tryptophan mutation was associated with mild hypertrophy and no sudden cardiac death in two unrelated families ([13, 25]). Therefore, we believe that independent clinical data, of the type presented here, were necessary to confirm that there is a characteristic phenotype among cardiac troponin T gene mutations.
3.4 Screening for cardiac troponin T gene mutations.
Evidence is accumulating that in familial hypertrophic cardiomyopathy, a “monogenic” disorder, the development of the disease is influenced not only by a major gene and its specific mutations, but by other contributary factors ([13, 23, 26]). However, in the case of familial hypertrophic cardiomyopathy caused by cardiac troponin T gene mutations, it appears that this major gene is strongly predictive of the course of the disease. The consistent association of a poor prognosis with deceptively mild clinical features and reduced penetrance emphasizes the relevance of screening for cardiac troponin T gene mutations. This process will be facilitated by the relatively small coding sequence of the cardiac troponin T gene () and the presence of the codon 92 mutation hotspot. We believe that screening for cardiac troponin T gene mutations will be of value in families with sudden cardiac death and familial hypertrophic cardiomyopathy. Moreover, such screening may be appropriate for first-degree relatives of subjects with sudden cardiac death where a diagnosis of familial hypertrophic cardiomyopathy was suspected, for example on histologic grounds, even in the absence of hypertrophy. Detection of disease-causing mutations will identify subjects at risk of sudden cardiac death that will assist in designing and testing treatment strategies and in patient management and counseling.
☆ Financial support for this study was received from The University of Stellenbosch, Stellenbosch; The South African Medical Research Council, Tygerberg; The Harry and Doris Crossley Fund, Stellenbosch; and the British Heart Foundation, London, England, United Kingdom.
- Received August 8, 1996.
- Revision received October 31, 1996.
- Accepted November 12, 1996.
- The American College of Cardiology
- Goldstein S,
- Bayés-de-Luna A,
- Soldevila JG
- Maron BJ,
- Epstein SE,
- Roberts WC
- McKenna WJ,
- Watkins H
- ↵Maron BJ, Bonow RO, Cannon RO, Leon MB, Epstein SE. Hypertrophic cardiomyopathy: interrelations of clinical manifestations, pathophysiology, and therapy. N Engl J Med 1987;316:780–9, 844–52.
- Maron BJ,
- Roberts WC,
- Epstein SE
- Newman H,
- Sugrue D,
- Oakley CM,
- Goodwin JF,
- McKenna WJ
- Schwartz K,
- Carrier L,
- Guicheney P,
- Komadja M
- Posen BM,
- Moolman JC,
- Corfield VA,
- Brink PA
- Roberts R
- Moolman JC,
- Ferreira A,
- Hirschfield W,
- Hunter J,
- Corfield VA,
- Brink PA
- Corfield VA,
- Moolman JC,
- Martell R,
- Brink PA
- Rayment I,
- Holden HM,
- Sellers JR,
- Fananapazir L,
- Epstein N
- Watkins H,
- Seidman JG,
- Seidman CE
- Moolman JC,
- Brink PA,
- Corfield VA
- Marian AJ,
- Mares A,
- Kelly DP,
- et al.