Browsing by Subject "genome mining"

Antibiotic resistance is a worldwide problem and it threatens the prevention and treatment of infections caused by different pathogens. All living organisms produce natural products including ribosomal peptides with great variety. They are widely distributed in nature and they are playing more significant role in the search of new antimicrobial compounds used as therapeutical agents. Bacteria are a prolific source of peptides many of which are antimicrobial and microbial genomes are widely believed to encode new antimicrobial peptides. Genome mining has expanded the number of families of ribosomally synthesized natural products in recent years. These In silico approaches together with molecular biology and chemical analysis aim to identify novel compounds. In this study an unknown cyanobactin-like gene cluster was discovered by genome mining from genomes of cyanobacteria and also other bacteria. The aim of this work was to study the occurrence of the gene clusters in various bacterial genomes and the structures of novel peptides. The active biosynthesis of these peptides was tested by LCMS- and Q-TOF -analyses based on bioinformatic predictions. The production of the predicted peptides was also tested with stable sulphur isotope labelling. The aim was also to clone the genes needed for peptide biosynthesis into E. coli and to study antimicrobial activities of these peptides. Bioinformatic analyses suggested that the gene clusters encoded 1–8 precursor peptides together with protease. The precursor peptides had conserved leader sequence (LPxQxxPVxR) and a highly variable core sequences, often encoding an even number of cysteines. The mature peptide is eventually formed from core sequence through post-translational changes in the precursor peptide. The gene cluster was present in 38 bacterial genomes representing a diverse selection of bacterial phyla including cyanobacteria, proteobacteria, actinobacteria, bacteroidetes, firmicutes and planctomycetes. Analyses of the precursor peptide core regions suggested that the products are 8–131 amino acids in length. These peptides could be divided into two groups based on their structures: They form a selection of disulphide-bridge stabilized peptides with 2–5 disulphide-bridges as well as short cationic peptides with an ?-helical structure. Surprisingly, these types of peptides are common in eukaryotes and part of the innate immune system displaying potent antimicrobial properties but very rarely reported for bacteria. The peptides predicted from bioinformatic analysis were detected from Pseudanabaena sp. PCC 6802 using a combination of molecular biology and structural chemistry. Heterologous expression of the gene cluster from Pseudanabaena sp. PCC 6802 in E. coli confirmed that the gene cluster is active. A set of short cationic synthetic peptides with ?-helical structure predicted from Oscillatoria sp. PCC 10802, Dickeya zeae Ech1591, Vibrio nigripulchritudo SOn1, Agarivorans albus MKT 106, Roseibium sp. TrichSKD4 and Yersinia frederiksenii ATCC 33641 were shown to have potent antimicrobial activity between 0.8–100 ?g/ml. These findings prove that predicted cysteine containing peptides are produced by bacteria and some peptides from this novel family have antimicrobial activity, which might pave the way for new possible drugs derived from natural products.