Mutations in TNNT3 Cause Multiple Congenital Contractures: A Second Locus for Distal Arthrogryposis Type 2B

To the Editor:

We recently reported that distal arthrogryposis type 1 (DA1 [MIM 108120]) and distal arthrogryposis type 2B (DA2B [MIM 601680]), both of which are characterized by congenital contractures of the hands/wrists and feet/ankles (Bamshad et al. 1996), are caused by mutations in TNNI2 and TPM2, respectively (Sung et al. 2003). TNNI2 encodes an isoform of troponin I; this isoform and the isoforms of troponin T (TnT) and troponin C constitute the troponin complex of fast-twitch myofibers. This complex is the primary sensor of intracellular Ca⁺² ion concentration in skeletal muscle, and, consequently, it is an important regulator of muscle contraction. The troponin complex of fast-twitch myofibers exerts its effect on muscle contraction by binding to actin and β-tropomyosin, the product encoded by TPM2 (Clark et al. 2002). These findings led us to hypothesize that mutations in genes encoding other contractile-apparatus proteins specific to fast-twitch myofibers might also cause multiple congenital contractures. We now report the discovery of a mutation, in TNNT3 (the gene encoding TnT specific to fast-twitch myofibers), that causes DA2B.

We sequenced TNNT3 in 47 families with either DA2A (classical Freeman-Sheldon syndrome [MIM 193700]) or DA2B. We found a G→A missense mutation, at nucleotide position 188 in exon 9 of the TNNT3 cDNA (GenBank accession number NM_006757), that causes an arginine-to-histidine substitution at amino acid residue 63 (R63H) of TnT in a mother with DA2B and her two affected children (fig. 1). For several reasons, this mutation is probably disease causing. First, the mutation identified in the proband was also present in all affected family members (fig. 1). There is, however, a probability of 1/4 that this pattern occurred by chance. The inference that R63H causes DA2B would be strengthened by demonstrating that this mutation did not occur in the unaffected parents of I-2 (i.e., that it is a de novo mutation). However, the only living parent of I-2 is unavailable for study. Second, this change was not found in 488 chromosomes from an ethnically matched control group that we screened. Third, R63H results in the substitution of an amino acid residue that is conserved in all known isoforms of TnT (fig. 2), implying that this difference is likely to have structural and/or functional consequences. Fourth, substitution of the homologous amino acid residue in the cardiac-specific form of TnT causes cardiomyopathy (Varnava et al. 1999).

Figure  1.

Electropherogram demonstrating heterozygosity for a G→A missense mutation at nucleotide position 188 in exon 9 of TNNT3 in a family with DA2B. To confirm the presence of this mutation, we incorporated a MluI restriction site into the amplicon by mismatch PCR. The presence of the mutation eliminates this site, producing fragments of 144 bp and 110 bp in the affected mother and her two affected children (blackened symbols), whereas the unaffected father is homozygous for the 110-bp fragment.

Figure  2.

Amino acid sequences of fast-twitch TnT in human, mouse, and bird, aligned with amino acid sequences of human slow-twitch TnT and human cardiac TnT.

Because mutations in TNNI2 have been found in only ∼10% of cases of DA2B, we suspected that DA2B is a genetically heterogeneous condition (Sung et al. 2003). To date, however, linkage studies have not identified any candidate regions other than chromosome 11p15.5 (Krakowiak et al. 1997). The observation that DA2B can be caused by mutations in either TNNI2 or TNNT3 confirms that DA2B is genetically heterogeneous. Because TNNI2 and TNNT3 are located within several hundred kilobases of one another on chromosome 11p15.5, this conclusion is also consistent with the results of our prior linkage studies (Sung et al. 2003). Nevertheless, the absence of mutations in TNNI2 or TNNT3 in most cases of DA2B suggests either that regulatory regions of these genes harbor mutations or that mutations in genes yet to be identified also cause DA2B.

Although the cause of DA2B can be distinguished by direct testing of TNNT3 and TNNI2, there appear to be few, if any, ways to distinguish, on the basis of only clinical characteristics, which gene is responsible. There may, however, be sufficient phenotypic differences between DA2B and DA1 to distinguish between them. In addition to the facial features (e.g., small mouth and prominent nasolabial folds) common to DA2B but lacking in individuals with DA1, several characteristics (e.g., vertical talus and scoliosis) are more frequent in DA2B than in DA1. Additionally, the hand and foot contractures in patients with DA2B appear to be more resilient to medical intervention (e.g., occupational therapy and casting). It should be cautioned, however, that mutations have been found in too few families with both DA1 and DA2B to lend much credibility to broad generalizations about genotype-phenotype relationships.

The mechanism by which the R63H substitution in TnT in fast-twitch myofibers causes congenital contractures is unknown. Missense mutations in TNNT2—a TNNT3 paralogue, encoding a cardiac-specific form of TnT—cause ∼15% of cases of familial hypertrophic cardiomyopathy (Watkins et al. 1995). One of these mutations is an arginine-to-leucine substitution of amino acid residue 94 (R94L), which is homologous to amino acid residue 63 in fast-twitch myofiber TnT (Varnava et al. 1999). The R94L substitution perturbs tropomyosin-dependent functions of TnT, including the binding of tropomyosin to actin (Palm et al. 2001), an effect that might be due, in part, to impaired flexibility of the N-terminal tail of TnT (Hinkle and Tobacman 2003). The R63H substitution may have a similar effect on TnT in fast-twitch myofibers.

The theme that is emerging from this and our previous studies is that perturbation of the function of the contractile apparatus of skeletal muscle during fetal development can cause multiple congenital contractures in individuals with an otherwise normal neuromuscular examination. On the basis of this result, it seems plausible that polymorphisms in one or more of the genes encoding the proteins of the troponin-tropomyosin complex of fast-twitch myofibers may influence an individual’s susceptibility to isolated contractures (e.g., idiopathic clubfoot) or modify the phenotype of common myopathic disorders (e.g., Duchenne muscular dystrophy). At minimum, this report underscores the existence of a new class of genetic muscle diseases that lack many of the findings typical of a heritable myopathy.