Chloroplast DNA as an evolutionary marker in thePhaseolus vulgaris complex

Publisher Website
Google Scholar
Abstract
We have analyzed the changes occurring in the chloroplast DNA (cpDNA) of taxa belonging to thePhaseolus vulgaris complex to help clarify relationships among species of this complex. Two restriction maps for 11 restriction enzymes comprising the whole chloroplast genome from a wildP. vulgaris and a wildP. coccineus accession were constructed. These maps allowed us to compare a total of 330 restriction sites between the two genomes in order to identify polymorphisms, assess the type of mutations detected, and identify regions of high variability. A region, located in the large single-copy region near the borders with the inverted repeats, accounted for a large portion of the variation. Most of the mutations detected were due to restriction sites gains or losses. Variable and conserved regions were then evaluated in 30 accessions belonging to taxa of theP. vulgaris complex. Phylogenetic analyses were made using parsimony methods. Conclusions obtained from such analyses were the following: (1) there was high cpDNA variability withinP. coccineus but not inP. vulgaris. (2)P. coccineus subsp.glabellus showed a very distinct cpDNA type that strongly suggests that it actually belongs to a different but as yet undetermined section of the genus. Our cpDNA observations are supported by distinctive morphological traits and reproductive biology of this taxon. (3) InP. coccineus subsp.darwinianus (also classified asP. polyanthus), the cpDNA lineage was in disagreement with data obtained from nuclear markers and suggested a reticulated origin by hybridization betweenP. coccineus as the male parent and an ancestralP. polyanthus type, closely allied toP. vulgaris, as the seed parent. This initial cross was presumably followed by repeated backcrossing toP. coccineus. Our cpDNA studies illustrate the importance of molecular markers in elucidating phylogenetic relationships. They also indicate that accurate phylogenies will require analyses of both nuclear and cytoplasmic genomes.