
Full text loading...
The ribosomal AT-L30 proteins from 81 species of the genus Streptomyces as listed by Williams et al. in Bergey’s Manual of Systematic Bacteriology were analyzed. My results provided further evidence that the genus Streptomyces is well circumscribed. On the basis of levels of AT-L30 N-terminal amino acid sequence homology, the strains were classified into four groups (groups I to IV) and a nongrouped category, whose members contained amino acid sequences characteristic of each species. A phylogenetic tree constructed on the basis of the levels of similarity of the amino acid sequences revealed the existence of six clusters within the genus. The first cluster contains the members of groups I and II together with several other species; the second cluster contains the members of groups III and IV and several other species; the third cluster contains Streptomyces ramulosus and Streptomyces ochraceiscleroticus; the fourth cluster contains only Streptomyces rimosus; the fifth cluster contains Streptomyces aurantiacus and Streptomyces tubercidicus; and the sixth cluster contains Streptomyces albus and Streptomyces sulphureus. Considerable agreement between the results of the AT-L30 analyses and the results of numerical phenetic classification was found, although there were numerous disagreements in details. For example, four groups (groups I to IV) defined by the AT-L30 analysis data did not correlate with the aggregate groups defined by numerical classification. In general, but not always, the species classified in a particular cluster in the numerical classification system had the same or similar AT-L30 terminal amino acid sequences. The AT-L30 analysis data were more consistent with the 16S rRNA analysis data than with the numerical classification data, indicating that there was a good correlation between the four groups defined by AT-L30 analysis data and the aggregate groups defined by 16S rRNA analysis data. I stress that discrepancies between results of phenetic analyses and results of phylogenetic analyses should be taxonomically significant and can be resolved by other taxonomic approaches, such as DNA relatedness analysis.