No abstract

Defects in the dystrophin complex (DC) underlie several human genetic disorders, but our dissection of its function is complicated by potential redundancy of the multiple vertebrate isoforms of most DC components. We here complete our previous description of Drosophila dystrophin, and show that the fly retains all essential components of the DC, but with substantially less diversity. Seventeen known human components (three dystrophin-related proteins, two dystrobrevins, five sarcoglycans, five syntrophins, one dystroglycan and one sarcospan) appear to be reduced to eight in Drosophila (one, one, three, two, one and none, respectively). The simplicity of this system recommends it as a model for its human counterpart. ß

Background: Dystrophins and dystrobrevins are distantly related proteins with important but poorly understood roles in the function of metazoan muscular and neuronal tissues. Defects in them and their associated proteins cause a range of neuromuscular disorders. Members of this superfamily have been discovered in a relatively serendipitous way; we set out to compile a comprehensive description of dystrophin-and dystrobrevin-related sequences from available metazoan genome sequences, validated in representative organisms by RT-PCR, or acquired de novo from key species.

The 3'-untranslated region (3'UTR) of some vertebrate dystrophin genes shows an extraordinary degree and extent of conservation (better than that of many coding regions), a phenomenon that remains unexplained. We examine novel sequence and mutational data to explore the possible reasons for this. We show that loss of the human dystrophin 3'UTR is sufficient to cause Becker muscular dystrophy with pronounced reduction in dystrophin protein levels. The acquisition of dystrophin 3'UTR sequence from an amphibian and a cartilaginous fish allows us to refine previously identified functionally constrained regions which might account for the observed phenotype. These comprise (a) the open reading frame encoding the ancestral 'alternative' amphipathic C-terminal a-helix, normally removed from adult dystrophin by inclusion of a poorly conserved frameshifting penultimate exon, and (b) two highly conserved untranslated regions ('Lemaire A', 350 nucleotides and 'Lemaire D', 250 nucleotides) separated by a non-conserved 700 -2000-nucleotide spacer. We consider the possibility that the 3'UTR may represent a significant target for pathogenic mutations.