A 260-bp segment of the DNA that encodes 16S rRNA, corresponding to positions 44 to 337 in the Escherichia coli sequence, was amplified by the polymerase chain reaction and sequenced frofn each of 13 bacteria (rhizobia and purple phototrophs) in the alpha subdivision of the class Proteobacteria. The phylogenetic tree calculated from differences in the sequenced segment conforms well to our expectations based on other previously published data. The sequence from BTAil (a recently described phototrophic symbiont of the legume Aeschynomene) and that from the free-living phototroph Rhodopseudomonas palustris both fall within the range of variation found among strains of the soybean symbiont Bradyrhizobium japonicum. This suggests that it would be appropriate to include all of these organisms in a single genus.

No abstract

It is evident from complete genome sequencing results that lateral gene transfer and recombination are essential components in the evolutionary process of bacterial genomes. Since this has important implications for bacterial systematics, the primary objective of this study was to compare estimated evolutionary relationships among a representative set of ␣-Proteobacteria by sequencing analysis of three loci within their rrn operons. Tree topologies generated with 16S rRNA gene sequences were significantly different from corresponding trees assembled with 23S rRNA gene and internally transcribed space region sequences. Besides the incongruence in tree topologies, evidence that distinct segments along the 16S rRNA gene sequences of bacteria currently classified within the genera Bradyrhizobium, Mesorhizobium and Sinorhizobium have a reticulate evolutionary history was also obtained. Our data have important implications for bacterial taxonomy, because currently most taxonomic decisions are based on comparative 16S rRNA gene sequence analysis. Since phylogenetic placement based on 16S rRNA gene sequence divergence perhaps is questionable, we suggest that the proposals of bacterial nomenclature or changes in their taxonomy that have been made may not necessarily be warranted. Accordingly, a more conservative approach should be taken in the future, in which taxonomic decisions are based on the analysis of a wider variety of loci and comparative analytical methods are used to estimate phylogenetic relationships among the genomes under consideration.Rhizobia are nitrogen-fixing bacterial symbionts of legumes that are of economic importance in low-input sustainable agriculture, agroforestry, and land reclamation. Descriptions of phenotypic and genetic variation among rhizobia have been extensive. Currently there are six genera of rhizobia, Allorhizobium, Azorhizobium, Bradyrhizobium, Rhizobium, Mesorhizobium, and Sinorhizobium, that have been proposed (65), and there is a report of a single species of Methylobacterium (of the family Methylobacteriaceae), Methylobacterium nodulans, which forms a symbiosis with specific species of Crotolaria (59). The primary criterion by which these genera are defined is analysis of 16S rRNA gene sequence (65).Sequencing the 16S rRNA gene has profoundly affected how relationships among the bacteria are portrayed (45). The 16S rRNA gene sequence is useful for this purpose because it is slowly evolving and the gene product is both universally essential and functionally conserved. However, basing bacterial phylogeny on 16S rRNA gene sequence variation not only presupposes that evolution throughout the genome progresses at a constant rate by mutation and Darwinian selection but also assumes that the evolution of the genome and of the 16S rRNA gene is strictly hierarchical. From a practical point of view this approach also requires each genome to harbor a single copy of the 16S rRNA gene or that multiple alleles within single genomes have identical sequences.Although seemingly convenient for c...

Previously, we found that genetically diverse rhizobia nodulating Lotus corniculatus at a field site devoid of naturalized rhizobia had symbiotic DNA regions identical to those of ICMP3153, the inoculant strain used at the site (J. T. Sullivan, H. N. Patrick, W. L. Lowther, D. B. Scott, and C. W. Ronson, Proc. Natl. Acad. Sci. USA 92:8985-8989, 1995). In this study, we characterized seven nonsymbiotic rhizobial isolates from the rhizosphere of L. corniculatus. These included two from plants at the field site sampled by Sullivan et al. and five from plants at a new field plot adjacent to that site. The isolates did not nodulate Lotus species or hybridize to symbiotic gene probes but did hybridize to genomic DNA probes from Rhizobium loti. Their genetic relationships with symbiotic isolates obtained from the same sites, with inoculant strain ICMP3153, and with R. loti NZP2213 T were determined by three methods. Genetic distance estimates based on genomic DNA-DNA hybridization and multilocus enzyme electrophoresis were correlated but were not consistently reflected by 16S rRNA nucleotide sequence divergence. The nonsymbiotic isolates represented four genomic species that were related to R. loti; the diverse symbiotic isolates from the site belonged to one of these species. The inoculant strain ICMP3153 belonged to a fifth genomic species that was more closely related to Rhizobium huakuii. These results support the proposal that nonsymbiotic rhizobia persist in soils in the absence of legumes and acquire symbiotic genes from inoculant strains upon introduction of host legumes.

The phylogenetic relationships among Rhizobium species that nodulate Phaseolus vulgaris (common bean) were determined by directly sequencing the amplified 16s ribosomal DNA genes of these organisms. The bean strains formed four separate clusters. One cluster was composed of Rhizobium leguminosarum bv. trifolii, R. leguminosarum bv. viciae, and R. leguminosarum bv. phaseoli. Two other clusters comprised Rhizobium etli and Rhizobium tropici, and the fourth cluster contained a single bean-nodulating strain. Data for species identification were obtained from DNA-DNA reassociation experiments. The levels of DNA relatedness among strains belonging to the three biovars of R. leguminosarum ranged from 58 to 67%. The levels of DNA relatedness between R. leguminosarum bv. phaseoli and R. etli and R. tropici ranged from 43 to 45% and 13 to 16%, respectively. The levels of DNA relatedness between the strain belonging to the fourth cluster and strains of the other three Rhizobium species that nodulate beans were less than 10%.Phaseolus vulgaris L. (common bean) is an agriculturally important legume crop which benefits from a symbiosis with bacteria belonging to the genus Rhizobium. The rhizobia that infect host legumes, such as peas, clovers, and common beans, have been placed in a single species, Rhizobium leguminosarum. This species has been subdivided into three biovars largely on the basis of specificity for host plant infection and nodulation (12). Jordan (12) noted that of the three biovars of R. leguminosarum, R. leguminosarum bv. phaseoli was more distinct than the other two. The reasons for this became apparent when the heterogeneity of strains classified as members of R. leguminosarum bv. phaseoli was described (3). Bean isolates obtained from English fields are only weakly polymorphic (26). In contrast, wide phenotypic and genotypic variations have been reported among strains originating in the Americas (16, 17, 19, 20).Bean strains that originated in Mexico and South America were identified as members of a heterogeneous complex of strongly differentiated phylogenetic lineages, and the data indicated that several species should be recognized (19). One of the deep lineages was subsequently classified as Rhizobium tropici (17). On the basis of 16s ribosomal DNA (rDNA) gene sequences, Willems and Collins (25) showed that the phylogenetic position of R. tropici was distinct from that of R. leguminosarum. However, their analysis was with reference to a clover strain, strain ATCC 14480, because the 16s rDNA sequences of R. leguminosarum bv. phaseoli had not been determined and were not available.Until recently, all bean strains other than R. tropici were classified as R. leguminosai-um. However, there are a number of differences between the nucleotide sequences of a partial analysis of the 16s rDNA genes of bean strains of American origin and the sequence of R. leguminosarum bv. viciae type strain ATCC 10004. This finding led to the suggestion that the bean strains of American origin should be referred to as Rhizobium ...

Besides nodulating the original trap host, the isolates formed nitrogen-fixing symbioses with Phaseolus vulgaris. Only half of the isolates nodulated alfalfa (Medicago sativa), but these did not form nitrogen-fixing symbioses. Rhizobium tropici also formed nitrogen-fixing symbioses with Medicago ruthenica. A total of 56 distinctive multilocus electrophoretic types (ETs) were identified among 94 of the 106 isolates which were analysed for variation in electrophoretic mobility of 12 enzyme loci. One isolate (USDA 1920) possessed a unique El, while the ETs of the other isolates formed two weakly divergent subgroups approximately equal in size. It was concluded from small subunit rRNA gene sequences of eight isolates of Medicago ruthenica that they belonged to the genus Rhizobium and not to the genus Sinorhizobium which is more commonly associated with Medicago. Genomic similarity, determined from DNA hybridization analysis, between USDA 1920 and the strain representing the remaining isolates (USDA 1844) was lower than 20%. Based upon these observations it was concluded that at least three genomic species of rhizobia form nitrogen-f ixing symbioses with Medicago ruthenica. One of these genomic species is R. tropici, another is represented by the single isolate USDA 1920 and the name Rhizobium mongolense is proposed for the third genomic species represented by USDA 1844.

No abstract

Phenotypic and DNA sequence comparisons are presented for eight Rhizobium isolates that were cultured from field-grown alfalfa (Medicago sativa L.) in Oregon. These isolates were previously shown to nodulate both alfalfa and common bean (Phaseolus vulgaris (L.) Savi.). The objective of the present study was to determine their phylogenetic relationships to the normal symbionts of these plants, Rhizobium meliloti and Rhizobium leguminosarum biovar phaseoli, respectively. Phenotypically, the Oregon isolates more nearly resemble strains from P. vulgaris than those from M. sativa. For example, even though nitrogen fixation levels were low with both host species, the symbiotic efficiency of a representative Rhizobium isolate (Or 191) with common bean was twice that observed with alfalfa. Comparative sequencing of a 260-bp segment of the 16S rRNA gene