Search

Abstract Background Sinorhizobium meliloti is a soil bacterium that forms nitrogen-fixing nodules on the roots of leguminous plants such as alfalfa (Medicago sativa). This species occupies different ecological niches, being present as a free-living soil bacterium and as a symbiont of plant root nodules. The genome of the type strain Rm 1021 contains one chromosome and two megaplasmids for a total genome size of 6 Mb. We applied comparative genomic hybridisation (CGH) on an oligonucleotide microarrays to estimate genetic variation at the genomic level in four natural strains, two isolated from Italian agricultural soil and two from desert soil in the Aral Sea region. Results From 4.6 to 5.7 percent of the genes showed a pattern of hybridisation concordant with deletion, nucleotide divergence or ORF duplication when compared to the type strain Rm 1021. A large number of these polymorphisms were confirmed by sequencing and Southern blot. A statistically significant fraction of these variable genes was found on the pSymA megaplasmid and grouped in clusters. These variable genes were found to be mainly transposases or genes with unknown function. Conclusion The obtained results allow to conclude that the symbiosis-required megaplasmid pSymA can be considered the major hot-spot for intra-specific differentiation in S. meliloti.

The Phenotype MicroArray (PM) system is used for testing different growth conditions in order to compare the metabolic capabilities of the reference strain Rm1021 with those of natural Sinorhizobium meliloti isolates, which are analyzed by comparative genomic hybridization (CGH). The results have shown that the analysis of phenotypic variability of Sinorhizobium meliloti populations is a vital and complementary step in the examination of genetic polymorphism of rhizobia and has helped in analyzing rhizobial evolutionary dynamics.

Sinorhizobium meliloti and Sinorhizobium medicae are two closely related species of the genus Sinorhizobium showing a similar host range, nodulating leguminous species of the genera Medicago, Melilotus and Trigonella, but their phylogenic relationship has not been elucidated yet. In this paper we report the application of three different molecular markers, (i) RFLP of nodD genes, (ii) 16S–23S rDNA intergenic gene spacer fingerprinting and (iii) amplification fragment length polymorphism to S. meliloti and S. medicae strains isolated from the Caucasian area, which is the region of origin of the host plant Medicago. The analysis of data could suggest the origin of S. medicae strains from an ancestral S. meliloti population. [Copyright &y& Elsevier]

Biological species may remain distinct because of genetic isolation or ecological adaptation, but these two aspects do not always coincide. To establish the nature of the species boundary within a local bacterial population, we characterized a sympatric population of the bacterium Rhizobium leguminosarum by genomic sequencing of 72 isolates. Although all strains have 16S rRNA typical of R. leguminosarum, they fall into five genospecies by the criterion of average nucleotide identity (ANI). Many genes, on plasmids as well as the chromosome, support this division: recombination of core genes has been largely within genospecies. Nevertheless, variation in ecological properties, including symbiotic host range and carbon-source utilization, cuts across these genospecies, so that none of these phenotypes is diagnostic of genospecies. This phenotypic variation is conferred by mobile genes. The genospecies meet the Mayr criteria for biological species in respect of their core genes, but do not correspond to coherent ecological groups, so periodic selection may not be effective in purging variation within them. The population structure is incompatible with traditional 'polyphasic taxonomy' that requires bacterial species to have both phylogenetic coherence and distinctive phenotypes. More generally, genomics has revealed that many bacterial species share adaptive modules by horizontal gene transfer, and we envisage a more consistent taxonomic framework that explicitly recognizes this. Significant phenotypes should be recognized as 'biovars' within species that are defined by core gene phylogeny.