The RNA polymerase beta-subunit genes (rpoB) of 67 Streptomyces strains, representing 57 species, five Kitasatospora strains and Micromonospora echinospora KCTC 9549 were partially sequenced using a pair of rpoB PCR primers. Among the streptomycetes, 99.7-100 % similarity within the same species and 90.2-99.3 % similarity at the interspecific level were observed by analysis of the determined rpoB sequences. The topology of the phylogenetic tree based on rpoB sequences was similar to that of 16S rDNA. The five Kitasatospora strains formed a stable monophyletic clade and a sister group to the clade comprising all Streptomyces species. Although there were several discrepancies in the details, considerable agreement was found between the results of rpoB analysis and those of numerical phenetic classification. This study demonstrates that analysis of rpoB can be used as an alternative genetic method in parallel to conventional taxonomic methods, including numerical phenetic and 16S rDNA analyses, for the phylogenetic analyses of the genera Streptomyces and Kitasatospora.

Thiopeptides, with potent activity against various drug-resistant pathogens, contain a characteristic macrocyclic core consisting of multiple thiazoles, dehydroamino acids, and a 6-membered nitrogen heterocycle. Their biosynthetic pathways remain elusive, in spite of great efforts by in vivo feeding experiments. Here, cloning, sequencing, and characterization of the thiostrepton and siomycin A gene clusters unveiled a biosynthetic paradigm for the thiopeptide specific core formation, featuring ribosomally synthesized precursor peptides and conserved posttranslational modifications. The paradigm generality for thiopeptide biosynthesis was supported by genome mining and ultimate confirmation of the thiocillin I production in Bacillus cereus ATCC 14579, a strain that was previously unknown as a thiopeptide producer. These findings set the stage to accelerate the discovery of thiopeptides by prediction at the genetic level and to generate structural diversity by applying combinatorial biosynthesis methods.

The catalytic domain structure of Streptomyces sioyaensis 1,3-beta-glucanase (278 amino acids), a member of glycosyl hydrolase family 16 (GHF16), was determined to 1.5 A resolution in space group P2(1)2(1)2(1). The enzyme specifically hydrolyzes the glycosidic bond of the 1,3-beta-linked glucan substrate. The overall structure contains two antiparallel six-and seven-stranded beta-sheets stacked in a beta-sandwich jelly-roll motif similar to the fold of GHF16 1,3-1,4-beta-glucanases. The active-site cleft of the enzyme is distinct, with the closure of one end primarily caused by two protruding loop insertions and two key residues, Tyr38 and Tyr134. The current known structures of 1,3-1,4-beta-glucanases and 1,3-beta-glucanase from Nocardiopsis sp., on the other hand, have open-channel active-site clefts that can accommodate six beta-D-glucopyranosyl units. The active-site structure of 1,3-beta-glucanase was compared with those of other homologous structures in order to address the binding and enzymatic specificity for 1,3-beta-linked glucans in Streptomyces. This information could be helpful in the development of specific antifungal agents.

A combined approach, comprising PCR screening and genome mining, was used to unravel the diversity and phylogeny of genes encoding 5-aminolevulinic acid synthases (ALASs, hemA gene products) in streptomycetes-related strains. In actinomycetes, these genes were believed to be directly connected with the production of secondary metabolites carrying the C5N unit, 2-amino-3-hydroxycyclopent-2-enone, with biological activities making them attractive for future use in medicine and agriculture. Unlike "classical" primary metabolism ALAS, the C5N unit-forming cyclizing ALAS (cALAS) catalyses intramolecular cyclization of nascent 5-aminolevulinate. Specific amino acid sequence changes can be traced by comparison of "classical" ALASs against cALASs. PCR screening revealed 226 hemA gene-carrying strains from 1,500 tested, with 87% putatively encoding cALAS. Phylogenetic analysis of the hemA homologs revealed strain clustering according to putative type of metabolic product, which could be used to select producers of specific C5N compound classes. Supporting information was acquired through analysis of actinomycete genomic sequence data available in GenBank and further genetic or metabolic characterization of selected strains. Comparison of 16S rRNA taxonomic identification and BOX-PCR profiles provided evidence for numerous horizontal gene transfers of biosynthetic genes or gene clusters within actinomycete populations and even from non-actinomycete organisms. Our results underline the importance of environmental and evolutionary data in the design of efficient techniques for identification of novel producers.