Intron-dependent evolution of the nucleotide-binding domains within alcohol dehydrogenase and related enzymes

It has been suggested that the intron/exon structure of a gene corresponds to its evolutionary history. Accordingly, early in evolution DNA segments encoding short functional polypeptides may have been rearranged (exon-shuffling) to create full-length genes and RNA splicing may have been developed to remove intervening sequences (introns) in order to preserve translational reading frames. A conflicting viewpoint would be that introns were randomly inserted into previously uninterrupted genes after their initial evolutionary development. If so, the sites of introns would be unlikely to consistently reflect the domain structure of the protein. To address this question, the intron/exon structure of the gene encoding human alcohol dehydrogenase (ADH) was determined and compared to the gene structures for other ADHs and related proteins, all of which possess nucleotide-binding domains. Our results indicate that the introns in the nucleotide-binding domains of all the genes examined do indeed fall at positions which separate the short functional polypeptides (i.e. beta strands) which are believed to comprise this domain. We argue that our data is most easily explained by the hypothesis that introns were present in an ancestral nucleotide-binding domain which was later rearranged by exon-shuffling to form the various dehydrogenases and kinases which utilize such a domain.