PubMed2004
Insertions and deletions in protein-coding genes are relatively rare events compared with sequence substitutions because they are more likely to alter the tertiary structure of the protein. For this reason, insertions and deletions which are clearly homologous are considered to be stable characteristics of the proteins where they are found, and their presence and absence has been used extensively to infer large-scale evolutionary relationships and events. Recently, however, it has been shown that the pattern of highly conserved, clearly homologous insertions at positions with no other detectable homoplasy can be incongruent with the phylogeny of the genes or organisms in which they are found. One case where this has been reported is in the enolase genes of apicomplexan parasites and ciliates, which share homologous insertions in a highly conserved region of the gene with the apparently distantly related enolases of plants. Here we explore the distribution of this character in enolase genes from the third major alveolate group, the dinoflagellates, as well as two groups considered to be closely related to alveolates, haptophytes and heterokonts. With these data, all major groups of the chromalveolates are represented, and the distribution of these insertions is shown to be far more complicated than previously believed. The incongruence between this pattern, the known evolutionary relationships between the organisms, and enolase phylogeny itself cannot be explained by any single event or type of event. Instead, the distribution of enolase insertions is more likely the product of several forces that may have included lateral gene transfer, paralogy, and/or recombination. Of these, lateral gene transfer is the easiest to detect and some well-supported cases of eukaryote-to-eukaryote lateral transfer are evident from the phylogeny.