Patterns of Nucleotide Substitution in Angiosperm cpDNA trnL (UAA)–trnF (GAA) Regions

Open Access
Publisher Website
Google Scholar
Abstract
Patterns of substitution in chloroplast encoded trnL-F regions were compared between species of Actaea (Ranunculales), Digitalis (Scrophulariales), Drosera (Caryophyllales), Panicoideae (Poales), the small chromosome species clade of Pelargonium (Geraniales), each representing a different order of flowering plants, and Huperzia (Lycopodiales). In total, the study included 265 taxa, each with >900-bp sequences, totaling 0.24 Mb. Both pairwise and phylogeny-based comparisons were used to assess nucleotide substitution patterns. In all six groups, we found that transition/transversion ratios, as estimated by maximum likelihood on most-parsimonious trees, ranged between 0.8 and 1.0 for ingroups. These values occurred both at low sequence divergences, where substitutional saturation, i.e., multiple substitutions having occurred at the same (homologous) nucleotide position, was not expected, and at higher levels of divergence. This suggests that the angiosperm trnL-F regions evolve in a pattern different from that generally observed for nuclear and animal mtDNA (transition/transversion ratio ≥ 2). Transition/transversion ratios in the intron and the spacer region differed in all alignments compared, yet base compositions between the regions were highly similar in all six groups. A↔T and G↔C transversions were significantly less frequent than the other four substitution types. This correlates with results from studies on fidelity mechanisms in DNA replication that predict A↔T and G↔C transversions to be least likely to occur. It therefore strengthens confidence in the link between mutation bias at the polymerase level and the actual fixation of substitutions as recorded on evolutionary trees and, concomitantly, in the neutrality of nucleotide substitutions as phylogenetic markers.