Gene Sequence Based Clustering Assists in Dereplication of Pseudoalteromonas luteoviolacea Strains with Identical Inhibitory Activity and Antibiotic Production

Some microbial species are chemically homogenous, and the same secondary metabolites are found in all strains. In contrast, we previously found that five strains of P. luteoviolacea were closely related by 16S rRNA gene sequence but produced two different antibiotic profiles. The purpose of the present study was to determine whether such bioactivity differences could be linked to genotypes allowing methods from phylogenetic analysis to aid in selection of strains for biodiscovery. Thirteen P. luteoviolacea strains divided into three chemotypes based on production of known antibiotics and four antibacterial profiles based on inhibition assays against Vibrio anguillarum and Staphylococcus aureus. To determine whether chemotype and inhibition profile are reflected by phylogenetic clustering we sequenced 16S rRNA, gyrB and recA genes. Clustering based on 16S rRNA gene sequences alone showed little correlation to chemotypes and inhibition profiles, while clustering based on concatenated 16S rRNA, gyrB, and recA gene sequences resulted in three clusters, two of which uniformly consisted of strains of identical chemotype and inhibition profile. A major time sink in natural products discovery is the effort spent rediscovering known compounds, and this study indicates that phylogeny clustering of bioactive species has the potential to be a useful dereplication tool in biodiscovery efforts.