Skip to main content
ARS Home » Research » Publications at this Location » Publication #105569


item Van Berkum, Peter

Submitted to: Nitrogen Fixation International Congress
Publication Type: Proceedings
Publication Acceptance Date: 10/3/1999
Publication Date: N/A
Citation: N/A

Interpretive Summary: Classification schemes in bacteria rely on DNA sequences because no informative morphological characters and no fossil records are available. DNA sequences are used to diagram ancestral lineages. It is common practice to use the small subunit ribosomal (16S rRNA) gene for this purpose. However, we and others have critisized this since the 16S rRNA gene probably has undergone genetic recombination. Genetic recombination leads to mosaic geneologies which do not reflect the evolutionary history of each of the species. We present evolutionary relationships of rhizobia based on two other genes and show that they differ from that based on the 16S rRNA gene. We concluded from our data that partitioning of Sinorhizobium from Rhizobium as separate genera is not warranted. Also, we present two ancestral diagrams of the soybean serotype strains. We concluded that the 129 serogroup is genetically more similar to the 110, 122, and 62 serogroups than to the 123 and 127 serogroups. Also, our results would question the separation of Bradyrhizoum liaoningense as a species separate from B. japonicum. Our results will be useful for establishing common criteria for defining species limits in bacteria and for the separation of species into distinct bacterial genera.

Technical Abstract: Bacterial classification schemes are based upon variation in phenotypic and genetic traits. Bacterial evolutionary history or phylogeny is reconstructed from sequence variation of the 16S rRNA genes, which traditionally is diagramed as a hierarchical tree. The assumptions are that each bacterial species is monophyletic, meaning that the DNA was derived from a single common ancestor, and that the hierarchy across 16S rRNA genes approximates that across the species. The separation of bacteria into separate genera is almost exclusively based upon their reconstructed phylogeny. The aim of this contribution is to address some issues that have arisen in our research which may be useful to evaluate common criteria for estimating phylogeny among rhizobia. The focus will be on our analysis of the bacterial symbionts of Medicago and Glycine.