SUBJECTS AND METHODS

Patients

We examined an Italian BDA-1 family from Tuscany with three affected members in three different generations. A pedigree of the family is shown in fig 1.

Case III.1, 33 year old male (proband)

The proband was admitted to our clinic in October 1999, because of a four month history of increasing low back pain. He was 1.60 m tall (below the 3rd centile) and 58 kg in weight. Physical examination showed an upper/lower segment ratio of 1.4 and brachydactyly. In particular, the middle finger was 9 cm and the palm was 5.2 cm in length (fig 2A). The remainder of the clinical examination was within normal limits. Clinical, laboratory, and biochemical findings excluded endocrine diseases or other heritable disorders of connective tissue, such as the chondrodysplasias. Skeletal x rays showed signs of brachydactyly of the hands (fig 2D) and feet, in accordance with the clinical diagnosis of BDA-1.

Case II.6, 65 year old mother

The proband’s mother was born in southern Italy. She was 1.42 m tall (below the 3rd centile) and 50 kg in weight. She showed marked brachydactyly, with the palm 7.5 cm and the middle finger 4 cm in length (fig 2B). X rays of the hands (fig 2E) and feet confirmed the clinical evidence of BDA-1.

Case IV.2, 10 year old daughter

The proband’s daughter was born in 1991. At birth, she was 48 cm long (25th centile) and 3820 g in weight (50th centile). She has been previously seen in a paediatric department on several occasions and she always showed normal psychophysical development. We first observed the child in November 1999 when she was 8 years old, 122 cm tall (25th centile), and 25 kg in weight (50th centile). The head circumference was 51 cm (50th centile). Her physical examination was normal, except for signs of brachydactyly, with the palm 5 cm and the middle finger 4.3 cm in length (fig 2C). X rays of the hands (fig 2F) and feet showed a less severe degree of disease than her grandmother and her father. The evaluation of the skeletal age¹² corresponded to the age of 7 years 7 months.

Linkage and mutation analysis

Peripheral blood samples were taken from available family members and standard procedures were used to isolate genomic DNA.

Linkage analysis was performed by using six markers on chromosome 2q35-q36 (D2S164, D2S2249, D2S173, D2S1242, PAX3, D2S130). The primer sequences and allele sizes of these markers were obtained from Genome Database. PCR products were run in vertical polyacrylamide gel electrophoresis (6%) and silver stained.

PCR amplification of part of the coding region of the IHH gene (part of exon 1, exon 2, and exon 3) was performed by using the published primers.¹⁰ Genomic DNA (100 ng) was denatured at 97°C for five minutes, mixed with 10 × buffer, 0.5 μmol/l primers, 200 mmol/l dNTP, 1 U Taq polymerase in a final volume of 25 μl, and cycled 35 × at 94°C for one minute, 62°C for one minute, 72°C for one minute, and finally incubated at 72°C for five minutes. DNA sequencing was performed with Big dye terminator cycle sequencing kit (Applied Biosystems) on an ABI 310 Automated Sequencer and analysed with the Genescan package software. In order to sequence the entire exon 3, we designed an internal primer (5′ gcagctgtctctacacacgtgg 3′).

RESULTS

Linkage analysis suggested segregation of the disease with a locus on chromosome 2q35-q36, since all affected subjects shared the same 131441 haplotype (fig 1). Sequence analysis of the three exons of the IHH gene showed the presence of a novel mutation in the three affected patients. The mutation was a heterozygous transition, c.298G>A, in exon 1, resulting in the substitution of asparagine for aspartic acid at residue 100 (fig 3). The presence of the mutated amino acid eliminates a BseI restriction site. By restriction endonuclease digestion we confirmed the mutation in the affected subjects and excluded its presence in the two unaffected analysed members of the family (III.2 and IV.1 ), as well as in 100 unrelated Italian subjects.

DISCUSSION

In this paper, we describe a family with clinical and radiological findings of a mild form of BDA-1. Affected subjects showed broad hands, with proportionate shortening of all the digits and the most severely shortened bones represented by the middle phalanges. Moreover, some of the middle phalanges were absent in patients III.1 and II.6. Persistent epiphyseal nucleus of the third metacarpus and absence or hypoplasia of the styloid ulnaris were also seen, as in previous reports.^2,13 By contrast, the involvement of the first metacarpal in patients III.1 and IV.2 and of the third metacarpal in patient II.6 appears to be uncommon.^14,15 The feet showed a similar pattern of abnormalities, with an absent and/or rudimentary middle phalanx and proximal phalanx of the big toe. In contrast with other cases,¹³ the bone age of patient IV.2 was not delayed.

A previous report by Yang et al⁹ mapped the locus for BDA-1 to a 8.1 cM interval on chromosome 2q35-q36 in two large unrelated families from China.⁹ Consistent with this observation, the same authors recently reported that BDA-1 is the result of mutations in the IHH gene on chromosome 2q35-q36.¹⁰IHH is a secreted signalling protein member of the mammalian hedgehog family, which is implicated in cell-cell interaction.^16,17 Its mouse orthologue, Ihh, is expressed in gut and cartilage and coordinates proliferation and differentiation of chondrocytes during endochondral bone development.^17–19 To date, there have been no other reports confirming the association of BDA-1 with mutations in the IHH gene. By contrast, a recent study in a Canadian kindred linked BDA-1 to a novel locus on chromosome 5p13.2-13.3,¹¹ indicating that the disease may be genetically heterogeneous. Importantly, some phenotypic differences existed between the Canadian and the Chinese BDA-1 families. While the former showed a mild form of the disease, the latter had some features that were apparently beyond Fitch’s description,² such as severely shortened distal phalanges, shortened third metacarpals, and shortened proximal phalanges of the fifth digit.⁹ These differences were not surprising given that different loci have been implicated. Thus, different disease causing genes may be responsible for the variable phenotypes often described for BDA-1. A similar feature has been also proposed for other subtypes of brachydactyly, such as brachydactyly type B.²⁰ In this paper, we confirm the association of BDA-1 with mutations in the IHH gene in a three generation family showing variable clinical features. Our kindred had clinical and radiological characteristics different from both the Canadian and the Chinese families, indicating a relatively moderate form of BDA-1. In particular the middle phalanges and the proximal first phalanges were mainly involved, while the distal phalanges were of normal size. The IHH mutation we found in our family was different from those previously described.¹⁰ However, it falls in the same residue (aspartic acid 100) that is substituted in the D100E missense mutation found in one of the three pedigrees analysed by Gao et al.¹⁰ This amino acid is highly conserved across all vertebrates and intervertebrates studied so far and it is predicted to lie on the surface of the protein and to interact with the Patched receptor.¹⁰ Therefore, D100N mutation may cause aberrant signalling by interfering with IHH binding to its receptor or promoting its binding to other receptors.

Short stature is often a component of the brachydactylies even if less frequent in type A1 and more frequent in type C.^2,15 Interestingly, all the affected subjects showing an Asp100 mutation in the study of Gao et al¹⁰ were shorter than unaffected subjects in the same family, while BDA-1 subjects with different IHH mutations were of normal height. In the reported BDA-1 family showing a similar mutation on Asp100, patients III.1 and II.6 were short in stature, while patient IV.2 had normal growth. Thus, it is unlikely that short stature is phenotypically linked to a particular IHH mutation, such as that involving Asp100.

Finally, a decrease in the severity of phenotypes from the first to the third generation was observed in our BDA-1 family. However, since BDA-1 is known to exhibit considerable variation in severity both between and within affected families, the phenotypic differences observed in this Italian family may also be the result of variable expressivity occurring by chance in different generations. A detailed phenotypic description of the large families described by Gao et al¹⁰ would have been useful to define this issue.

In summary, results from the present study confirm that BDA-1 is a phenotypically heterogeneous disease resulting from mutation in the IHH gene and that the same IHH mutation may exhibit considerable variation in severity both between and within affected families.