Molecular evidence for the antiquity of group I introns inter-rupting transfer RNA genes in cyanobacteria

Abstract

Genes interrupted by group I introns have been the perennial focus of evolutionary studies. Previous work has demonstrated the importance of lateral transfer in the evolutionary history of these auto-catalytic molecules. In this respect the group I intron interrupting the tRNA-Leu (UAA) gene in cyanobacteria and chloroplasts has attracted a great deal of scientific attention primarily because of its perceived age. Recent studies have concluded that the group I introns interrupting tRNA-fMet and tRNA-Arg (CCU) genes in cyanobacteria and proteobacteria have arisen through recent genetic exchange and suggest that the origin of the tRNA-Leu intron is also in doubt. However, direct phylogenetic evidence for these competing hypotheses has been lacking. In this study molecular systematic approaches were undertaken to examine the evolutionary history of the group I introns interrupting tRNA genes in chloroplasts, cyanobacteria, and a-proteobacteria. Highly congruent support was found for the co-evolution of the introns and the genomes in which they are inserted. The introns interrupting the tRNA-fMet and the tRNA-Leu (UAA) genes predate cyanobacteria and chloroplasts respectively while the tRNA-Arg (CCU) intron predates mitochondria. The scattered and sporadic distribution of the introns is best explained by pervasive parallel losses in the more derived lineages of cyanobacteria and a-proteobacteria (Sections 3.2-3.5). This study provides convincing phylogenetic evidence that the tRNA group I intron subfamily is ancient and this means that these introns are between 2.1 and 3.5 billion years old. This strengthens the argument for the antiquity of this class of RNA enzyme. During phylogenetic analyses of cyanobacterial taxa containing group I introns it became apparent that the controversial sister taxa relationship between the non-heterocyst forming cyanobacteria Chroococcidiopsis PCC 7203 and the heterocyst forming cyanobacteria received highly congruent support with the inclusion of additional members of the genus and through independent and combined phylogenetic analyses of rpoC1, tufA and 16S rRNA gene datasets (Section 3.1). This is important because it means that the complex baeocyte differentiation process has arisen independently at least twice in the cyanobacterial radiation, that the morphological identical genus Myxosarcina is not closely related to Chroococcidiopsis and rejects Chroococcidiopsis as the most primitive living cyanobacterium.