Contractural Arachnodactyly, Congenital

A number sign (#) is used with this entry because of evidence that congenital contractural arachnodactyly (CCA) is caused by heterozygous mutation in the gene encoding fibrillin-2 (FBN2; 612570) on chromosome 5q23.

For a phenotypic description and a discussion of genetic heterogeneity of distal arthrogryposis, see DA1 (108120).

Congenital contractural arachnodactyly is a rare, autosomal dominant connective tissue disorder characterized by contractures, arachnodactyly, scoliosis, and crumpled ears (Hecht and Beals, 1972). It shares overlapping features with Marfan syndrome (154700), which is caused by mutation in the gene encoding fibrillin-1 (FBN1; 134797).

Beals and Hecht (1971) described father and 2 sons affected in 1 kindred and father, daughter and son (by different mothers) affected in a second kindred. They proposed that the disorder be called 'contractural arachnodactyly' and further suggested that the patient reported by Marfan (1896) had this disorder rather than the Marfan syndrome (154700) as presently delineated (Hecht and Beals, 1972). They found several other reports, apparently of the same disorder. Beyer et al. (1965) probably described the same condition in a mother and 4 children and some of the reports of combined Marfan syndrome and arthrogryposis multiplex congenita may be further examples (e.g., Reeve et al., 1960; Kingsley-Pillers, 1946). Epstein et al. (1968) described father and son with a connective tissue disorder with some features suggesting the Marfan syndrome and some suggesting osteogenesis imperfecta. Severe kyphoscoliosis, generalized osteopenia, flexion contractures of the fingers and abnormally shaped ears were among the characteristics. Abnormally shaped ('crumpled') ears have been emphasized by other students of CCA. According to Mirise and Shear (1979), the ocular and cardiovascular complications of the Marfan syndrome do not occur in contractural arachnodactyly (Mirise and Shear, 1979). Hence, the correct diagnosis has prognostic significance.

Bass et al. (1981) described CCA and Marfan syndrome in the same family. CCA was, however, the predominant finding in 4 generations of the family; the father of the propositus had keratoconus in addition to CCA. Pyeritz (1986) described several patients who had joint contractures and ear changes in the pinna seemingly characteristic of CCA but severe aortic changes typical of the Marfan syndrome. Bawle and Quigg (1992) described a black male with 'crumpled ear' deformity, scoliosis, and arachnodactyly, who had dilatation of the aortic root and ectopia lentis. They pointed to the patient of Reeve et al. (1960), an infant described as having 'Marfan syndrome and arthrogryposis,' who showed ectopia lentis in the right eye and in whom autopsy showed dilatation of the ascending aorta.

Anderson et al. (1984) reported a kindred in which many members of 3 generations showed features consistent with CCA. Of the 7 affected persons they examined, 6 had mitral valve prolapse. Family members without CCA did not have mitral valve prolapse. Although enlargement of the aortic root was not found, a 9-year-old girl was said to have 'an aortic diameter at the upper limits of normal.' Gruber et al. (1978) described severe mitral regurgitation in a premature infant with CCA. Reviewing 4 new families and 29 reported ones, Ramos Arroyo et al. (1985) stated that no ocular problems and no aortic problems have been encountered but that congenital heart defects have occurred 'in 14.7%.' Mitral regurgitation is a well-established feature of CCA; involvement of the aorta remains to be documented. Involvement of the eyes is also unclear. Huggon et al. (1990) described an infant girl with CCA complicated by mitral regurgitation. Although slit-lamp biomicroscopy showed no evidence of lens subluxation, the infant had iridodonesis apparently caused by anterior megalophthalmos. Langenskiold (1985) reported a case he followed for 37 years.

Currarino and Friedman (1986) described 2 unrelated infants with severe CCA, both of whom died in the first year of life. Cole and Hughes (1992) described an infant with presumed CCA who also had deficiency in the right lower limb. Viljoen et al. (1991) observed 8 affected persons in a family of Asiatic Indian descent. No linkage could be demonstrated with type I collagen probes (120150, 120160).

Viljoen (1994) published a review that included at least 40 families with more than 120 affected members with CCA.

Comparison of CCA and Marfan Syndrome

Zhang et al. (1994) showed differences of expression of fibrillin-1 and fibrillin-2 in human ear cartilage. They noted that this may account for the fact that abnormally shaped (i.e., crumpled) auricular helices are a hallmark of CCA. Most persons with Marfan syndrome do not have abnormally shaped ears, although some with neonatal Marfan syndrome may have crumpled ears (Godfrey et al., 1995). Similarities between neonatal Marfan syndrome and severe lethal CCA include arachnodactyly, joint contractures, and some facial characteristics. Importantly, although both have severe cardiovascular abnormalities that lead to very early death, the specific cardiac changes are quite different. Wang et al. (1996) tabulated the differences between the 2 syndromes with valvular insufficiency and aortic root dilatation in neonatal Marfan syndrome and structural defects in severe lethal CCA; scoliosis and vertebral anomalies predominantly in CCA; and duodenal atresia, esophageal atresia, and intestinal malrotation only in CCA. Zhang et al. (1995) suggested that expression of fibrillin-2 directs the assembly of elastic fibers during early embryogenesis, whereas fibrillin-1 provides the major structural (i.e., load bearing) function of the microfibrils.

In cloning the fibrillin gene (FBN1; 134797), located on chromosome 15 and mutant in the Marfan syndrome, Lee et al. (1991) isolated a cDNA for a second fibrillin locus, fibrillin-2 (612570). Study of 2 families with CCA demonstrated linkage between this locus and that phenotype; maximal combined lod score = 4.5 at theta = 0.00. For demonstrating linkage, a VNDR (variable number dinucleotide repeat) related to the FBN2 gene was used as the marker. The FBN2 gene was mapped to 5q23-q31 by in situ hybridization. The linkage between a candidate gene and a disease locus, although highly suggestive, does not constitute final proof of causal relationship. In this case, however, linkage between 2 structurally related genes (FBN1 and FBN2) and 2 phenotypically related disorders (Marfan syndrome and CCA, respectively) gave strong support to the causal association. Identification of mutations in the FBN2 gene in cases of CCA was required for final proof.

Putnam and Milewicz (1995) and Wang et al. (1995) identified point mutations in the FBN2 gene in cases of CCA. A mutation in a calcium-binding EGF-like motif (612570.0001) was found by the first authors and a mutation in a TGF-binding protein-like motif (612570.0002) by the second group.

In the father of 2 sibs affected with CCA, Putnam et al. (1997) demonstrated somatic mosaicism for an FBN2 mutation. The 2 sisters had been reported by Delemarre-van de Waal et al. (1980). The proband had arachnodactyly, contractures, crumpled ears, a highly arched palate, and mild retrognathia evident at birth. From the age of 4 years she had progressive thoracolumbar scoliosis, which required surgical correction at the age of 13 years. At 18 years of age, she was 173 cm tall, had crumpled ears, a preauricular tag on the right side, a low posterior hairline, slight bilateral ptosis, and mild retrognathia. She also had striae on her thighs, mild pectus carinatum, arachnodactyly, and contractures of elbows, knees, and fingers. Echocardiogram and ophthalmologic examinations were normal. The sister showed similar findings at birth. Like the sister she had normal mental development but delayed motor development. At 21 years of age, she was 176 cm tall and had crumpled ears, slight midthoracic scoliosis, striae on the upper thighs, arachnodactyly, and contractures. Echocardiogram and ophthalmologic examinations were normal. Both parents were unaffected. Putnam et al. (1997) noted that the mutation (612570.0005), resulting in abnormal splicing of the exon, was an unusual alteration that disrupted the invariant A in a putative branch point sequence found in the upstream intron. Exon 29 was deleted in the mutant allele. Analysis of FBN2 transcript levels by use of fibroblasts from one of the affected sibs indicated that the allele inherited from the mother, which did not contain the exon splicing mutation, was reduced in expression. This difference in FBN2 allele expression levels was also observed in CCA cell strains with previously characterized mutations, which showed greater expression of the mutated alleles (Putnam et al., 1995). These data expanded the spectrum of mutations that cause CCA. Putnam et al. (1997) suggested that the effects of mutations on fibrillin-2 are similar to those observed in fibrillin-1 and Marfan syndrome.

Park et al. (1998) identified FBN2 mutations in 6 of 12 unrelated CCA patient cell strains. All of the identified mutations were clustered in a limited region of the gene, a region corresponding to that in FBN1 where mutations produce the severe, congenital form of Marfan syndrome, so-called neonatal Marfan syndrome. Furthermore, 3 of the identified mutations occurred in the FBN2 locations exactly corresponding to FBN1 mutations that had been reported in cases of neonatal Marfan syndrome. These mutations indicate that this central region of both fibrillins plays a critical role in human embryogenesis. The limited region of FBN2 that can be mutated to cause CCA may also help explain the rarity of CCA compared to Marfan syndrome.

Belleh et al. (2000) reported 2 additional FBN2 mutations in CCA: C1141F in exon 26 (612570.0008) and C1252W in exon 29 (612570.0009). As in previous cases, mutations clustered in the region of fibrillin-2 homologous to the so-called neonatal Marfan syndrome region of fibrillin-1 (FBN1; 134797) (Kainulainen et al., 1994).

Gupta et al. (2002) noted that all of the identified CCA mutations in FBN2 cluster in a limited region similar to that where severe Marfan syndrome mutations cluster in FBN1, specifically between exons 23 and 34. Gupta et al. (2002) screened exons 22 through 36 of FBN2 for mutations in 13 patients with classic CCA by single-stranded conformation polymorphism analysis followed by direct sequencing. They successfully identified 10 novel mutations in this critical region of FBN2 in these patients, indicating a mutation detection rate of 75% in this region. None of these identified FBN2 mutations alter amino acids in the calcium-binding consensus sequence in the EGF-like domains, whereas many of the FBN1 mutations alter the consensus sequence. Gupta et al. (2002) reviewed the 21 known CAA mutations in the FBN2 gene, along with available clinical information on the probands. They found that 3 of the 21 patients had dilatation of the aortic root. All 3 were young, and the degree of dilatation appeared to have been borderline in all. However, because of the lack of knowledge of the natural history of aortic involvement in CCA, Gupta et al. (2002) recommended that all CCA patients have an echocardiogram. They cited Su et al. (2000) as indicating that approximately 15% of CCA patients have congenital heart defects. Their review did not support this conclusion, instead suggesting that congenital heart defects are only an occasional finding in these patients.

In a revised and extended classification scheme of the distal arthrogryposes, Bamshad et al. (1996) referred to this disorder as distal arthrogryposis type 9 (DA9).

Chaudhry et al. (2001) analyzed the classic mouse mutant 'Shaker-with-syndactylism' (sy) using a positional candidate approach. The authors demonstrated that several loss-of-function mutations, each located outside the 'neonatal region' of Fbn2, caused syndactyly in mice, rather than CCA as in man. The deafness in these animals is caused by mutations in the contiguous Na-K-2Cl cotransporter gene Slc12a2 (600840) (Dixon et al., 1999).