Split-Hand/foot Malformation 3

A number sign (#) is used with this entry because it represents a contiguous gene duplication syndrome on chromosome 10q24.

Description

Split-hand/split-foot malformation is a limb malformation involving the central rays of the autopod and presenting with syndactyly, median clefts of the hands and feet, and aplasia and/or hypoplasia of the phalanges, metacarpals, and metatarsals. Some patients with SHFM3 have been found to have mental retardation, ectodermal and craniofacial findings, and orofacial clefting (Elliott and Evans, 2006).

For additional phenotypic information and a discussion of genetic heterogeneity in this disorder, see SHFM1 (183600).

Clinical Features

Roscioli et al. (2004) described a large autosomal dominant pedigree segregating split-hand/foot linked to chromosome 10q24 (SHFM3). The phenotype was highly variable, ranging from the classic ectrodactyly deformity to partial absence of the thumb and agenesis of the distal tip of the index finger. The feet were more severely affected than the hands. Two individuals had nail dysplasia, and 1 had cleft palate. Radiologic features included short metacarpals with rounded proximal heads, agenesis of the radial ray, epiphyseal coning, and an unusual supernumerary ossicle opposed to the distal phalanx of the left thumb. Roscioli et al. (2004) proposed that the key diagnostic features of SHFM3-linked ectrodactyly include nail dysplasia, differential limb severity with the feet more involved than the hands, and the absence of deafness, mental retardation, and other features characteristic of the EEC syndrome (see 129900).

In a literature review including 48 SHFM1 patients, 40 SHFM3 patients, 45 SHFM4 (605289) patients, and 20 SHFM5 (606708) patients, Elliott et al. (2005) found that preaxial involvement of the upper extremities was a significant locus discriminator, most commonly seen in patients with SHFM3 (60% of patients). Preaxial involvement occurred in approximately 40% of SHFM5, 4% of SHFM4, and 2% of SHFM1 patients. The preaxial involvement consisted of proximally placed thumbs and/or triphalangeal thumbs, preaxial polydactyly, and/or absence of the first ray. However, SHFM3 patients tended to have a classic central longitudinal deficiency of the feet without a significant preaxial component. Elliott et al. (2005) postulated that differential anterior-posterior patterning exists between the upper and lower extremities. In further analysis of the previously studied SHFM patients, Elliott and Evans (2006) identified phenotypic patterns involving mental retardation, ectodermal and craniofacial findings, and orofacial clefting associated with the mapped genetic SHFM loci.

Clinical Variability

Buttiens and Fryns (1987) described a brother and sister with symmetric severe distal limb deficiencies affecting all 4 limbs and associated with microretrognathia and microstomia. The tongue was normal. The man was moderately mentally retarded and his sister mildly retarded. The man also had myopia and a renal change, referred to as oligomeganephronia, demonstrated by renal biopsy.

Keymolen et al. (2000) reported a boy, born of nonconsanguineous parents, with a similar phenotype. Severe congenital anomalies, including distal limb defects, micrognathia, cleft palate, and nystagmus, had been noted at birth. Radiologic examination of the limbs showed normal long bones, the presence of 2 metacarpals and 1 ulnar finger with 2 phalanges bilaterally, and the presence of 1 metatarsal and 1 phalanx bilaterally. The tarsal and carpal bones were not completely ossified at that time. Abdominal ultrasound disclosed bilateral renal hypoplasia. The child developed severe myopia, slowly progressive renal failure, and chronic serous otitis media resulting in bilateral hearing loss. At 4 years of age, he had micrognathia, a small tongue, and mild maxillary hypoplasia. Limb defects were symmetric, involving all 4 limbs. Psychomotor and intellectual development were appropriate for age. At the time of report, he was a candidate for kidney transplantation.

Mapping

In a multigenerational kindred with a typical split-hand/split-foot malformation phenotype, Palmer et al. (1994) excluded linkage to the SHFM1 locus on chromosome 7q21-q22. Gurrieri et al. (1994) likewise excluded linkage to 7q22.1 in a family with normal chromosomes and a highly penetrant type of SHSF segregating as an autosomal dominant trait.

Gurrieri et al. (1996) concluded that there must be at least 1 more autosomal locus for split-hand/split-foot malformation, inasmuch as some families fail to show linkage to 10q, 7q, 6q, or 2q, where loci for this malformation had been postulated.

Nunes et al. (1995) identified a stillborn infant with ectrodactyly and hemimelia of the right lower extremity, preaxial polydactyly of the left foot, cleft palate, and mild hypertelorism, in whom an unbalanced translocation resulted in monosomy 4pter-p15.1 and trisomy 10q25.2-qter. To investigate 10q25 as a possible split-hand/split-foot locus, microsatellite markers spanning 52 cM of 10q were used for linkage analysis of a family in which linkage to the SHFM1 region on 7q had previously been excluded (Palmer et al., 1994). Nunes et al. (1995) stated that the marker D10S583 was 'fully informative' in the family, giving a maximum lod score of 4.21 at theta = 0.00. Recombination haplotypes defined the 9-cM region between D10S541 and D10S574 as inclusive for this locus for which they proposed the designation SHFM3.

Gurrieri et al. (1996) tested a panel of 8 multiply affected families with 17 marker loci located in the 10q24-q25 region. Maximum lod scores of 3.73, 4.33, and 4.33 at a recombination fraction of zero were obtained for the loci D10S198, PAX2 (167409), and D10S1239, respectively. They defined a 19-cM critical region by haplotype analysis and noted that several genes with a potential role in limb morphogenesis were located in this region. The SHFM3 locus is in a region syntenic to mouse chromosome 19 where the mouse dactylaplasia (Dac) gene was mapped by Johnson et al. (1995). Dactylaplasia resembles SHSF type 1 (183600). The expression of Dac is regulated by a modifier locus (mdac) as demonstrated by the existence of permissive and nonpermissive strains of mice. The mdac locus maps to mouse chromosome 13 in an area syntenic to human chromosome 5q (Johnson et al., 1995).

Raas-Rothschild et al. (1996) mapped SHFM3 to 10q25 in 2 large kindreds of French ancestry. Two recombinant events reduced the critical region to a 9-cM interval encompassing several candidate genes.

By 2-point linkage analysis in a large Turkish family with isolated SHFM, Ozen et al. (1999) mapped the disorder to 10q24 and narrowed the SHFM3 region from 9 cM to an approximately 2-cM critical interval between genetic markers D10S1147 and D10S1240. In several instances they found a more severe phenotype in offspring of a mildly affected parent, suggesting anticipation. Data from this family, combined with those from 6 other pedigrees mapped to 10q24, demonstrated biased transmission of SHFM3 alleles from affected fathers to offspring. The degree of this segregation distortion was obvious in male offspring and was possibly of the same magnitude for female offspring.

In a large autosomal dominant pedigree linked to SHFM3, Roscioli et al. (2004) found a meiotic recombination event that enabled narrowing of the critical interval to between D10S1265 and D10S222, with the minimal critical region being between D10S1240 and D10S1267.

Molecular Genetics

De Mollerat et al. (2003) conducted mutation analysis of the dactylin gene (FBXW4; 608071) in 7 patients with split-hand/foot malformation (SHFM) and found no point mutations. However, Southern blot, pulsed field gel electrophoresis, and dosage analyses demonstrated a complex rearrangement associated with a 0.5-Mb tandem duplication in all the patients. The distal and proximal breakpoints were within 80- and 130-kb regions, respectively; the smallest duplicated region common to all patients was 444 kb and contained a disrupted extra copy of the dactylin gene from exon 9 to exon 6, as well as the entire LBX1 (604255), BTRC (603482), and POLL (606343) genes.

Because structural alterations of the gene encoding dactylin, a constituent of the ubiquitination pathway, lead to reduced levels of transcript in the mouse mutant Dac (see ANIMAL MODEL), which resembles human SHFM, Basel et al. (2003) studied several individuals with SHFM and reported a significant decrease of dactylin transcript. This observation supported a central role for dactylin in the pathogenesis of SHFM3.

Kano et al. (2005) screened 28 Japanese families with nonsyndromic SHFM for tandem genomic duplication of chromosome 10q24 by Southern blot and sequence analysis of the dactylin gene. No mutations were found in coding regions and flanking intronic sequences of the dactylin gene, but 2 families had genomic rearrangements. One was a 511.6-kb tandem duplication, and the other was a 447.3-kb duplication containing the LBX1, BTRC, POLL, and DPCD genes and a disrupted extra copy of the dactylin gene. Kano et al. (2005) noted that the smaller duplicated region was almost identical to that described by de Mollerat et al. (2003).

Everman et al. (2006) used PFGE to analyze the SHFM3 locus in 33 sporadic patients and 11 families with SHFM and identified 8 rearrangements in 7 nonsyndromic cases, 4 of which were familial, and in 1 sporadic syndromic case. The authors noted that all cases linked to the SHFM3 locus studied to date have chromosome rearrangements, and suggested that such rearrangements are the major, if not only, type of genetic abnormality at this locus causing SHFM.

Lyle et al. (2006) used FISH and quantitative PCR to narrow the SHFM3 locus to a minimal 325-kb duplication containing only the BTRC and POLL genes. Expression analysis of 13 candidate genes within and flanking the duplicated region showed that BTRC and SUFU (607035), present in 3 copies and 2 copies, respectively, were overexpressed in SHFM3 patients compared to controls. Lyle et al. (2006) suggested that SHFM3 may be caused by overexpression of BTRC and SUFU, both of which are involved in beta-catenin signaling.

Using array comparative genomic hybridization (CGH), Dimitrov et al. (2010) detected a chromosome 10q24 microduplication in a brother and sister with severe distal limb deficiencies, microretrognathia, and mental retardation, originally described by Buttiens and Fryns (1987). Analysis of 4 additional patients, including a boy with a similar phenotype involving limb deficiencies and micrognathia who was previously reported by Keymolen et al. (2000), revealed a duplication in the same region of chromosome 10q24 in all of them. Five of the affected individuals carried a duplication of at least 500 kb, encompassing 5 genes (LBX1, BTRC, POLL, DPCD, and FBXW4); the sixth patient, who had classic SHFM of the upper and lower limbs and renal hypoplasia, had a larger, 630-kb duplication that encompassed all genes from TLX1 (186770) to FGF8 (600483). Dimitrov et al. (2010) stated that these findings extend the clinical spectrum of SHFM3.

In a cohort of 56 families with SHFM, Klopocki et al. (2012) identified 17p13.3 duplications (SHFLD3; 612576) in 17 (30%) of the families, 10q24 duplications in 11 (20%), and TP63 (603273) mutations (SHFM4; 605289) in 5 (9%).

Animal Model

Sidow et al. (1999) noted that early outgrowth of the vertebrate embryonic limb requires signaling by the apical ectodermal ridge (AER) to the progress zone (PZ), which in response proliferates and lays down the pattern of the presumptive limb in a proximal to distal progression. Signals from the PZ maintain the AER until the anlagen for the distal phalanges have been formed. The semidominant mouse mutant dactylaplasia (Dac) disrupts the maintenance of the AER, leading to truncation of distal structures of the developing footplate, or autopod. Adult Dac homozygotes thus lack hands and feet except for malformed single digits, whereas heterozygotes lack phalanges of the 3 middle digits. Dac resembles the human split-hand/foot malformation. The area of mapping of SHFM3 on 10q24 shows conservation of synteny with the Dac region on mouse chromosome 19; thus, SHFM3 may be due to disruption of the human homolog of Dac (608071).

Because Dac mice exhibit the ectrodactyly phenotype only on certain genetic backgrounds, Chai (1981) postulated that the manifestation of the mutant gene Dac is controlled by another locus, which was termed Mdac. The Dac mutation is expressed as an autosomal semidominant trait only if the Mdac allele is homozygous. This hypothesis was verified by the identification of the Dac mutation (Sidow et al., 1999) and the mapping of Mdac to mouse chromosome 13 (Johnson et al., 1995).

The mouse dactylaplasia phenotype depends on 2 loci. The first locus, Dac, is an insertional mutation around the dactylin gene that is inherited as a semidominant trait. Two independent, spontaneous Dac alleles, Dac(1J) and Dac(2J), have been identified. The second locus, Mdac, is a modifier on chromosome 13 that is polymorphic in inbred strains. The dominant Mdac allele suppresses the dactylaplasia phenotype in Dac mice, whereas the recessive Mdac allele does not. Thus, dactylaplasia is observed only in mice homozygous for the recessive Mdac allele. Kano et al. (2007) showed that the Dac insertion is an LTR retrotransposon caused by the type D mouse endogenous provirus element (MusD), which contains intact ORFs for the viral gag, pro, and pol genes. In Dac(2J), MusD is inserted in intron 5 of the dactylin gene and no dactylin transcripts are expressed. In Dac(1J), MusD is inserted in the upstream region of dactylin and neither the size nor the amount of dactylin transcripts are affected. Kano et al. (2007) refined the Mdac locus to a 9.4-Mb region on chromosome 13. Dac mutants homozygous for the recessive Mdac allele had unmethylated CpGs and active chromatin at the LTR of the MusD insertion, as well ectopic expression of MusD elements. In contrast, Dac mutants carrying the dominant Mdac allele had heavily methylated CpGs and inactive chromatin at the LTR of the insertion, correlating with inhibition of the dactylaplasia phenotype. Ectopic expression of MusD was not observed in mice carrying the dominant Mdac allele. Kano et al. (2007) concluded that ectopic expression of the MusD insertion, rather than lack of dactylin expression, correlates with the dactylaplasia phenotype. They proposed the Mdac acts a defensive factor to protect the host genome from pathogenic MusD insertions.

Nomenclature

Palmer et al. (1994) proposed that the locus at chromosome 7q21-q22 be designated SHSF1 and that the former nomenclature, which employed the designation SHFD1 (split-hand/split-foot deformity), be changed to reflect the fact that this developmental disorder is not a deformity but a malformation. Accordingly, they designated 2 other forms of split-hand/foot malformation as SHSF2 (SHFM3) and SHSF3 (SHFM2; 313350). The Human Genome Nomenclature Committee determined in 1994 that split-hand/foot malformation should be symbolized SHFM.