Metabolomics and Genomics-Driven Genome-Mining in Streptomyces

Abstract

Bacterial natural products are important sources of novel therapeutics. However, not all secondary metabolites are expressed concurrently, increasing rediscovery of compounds and limiting identification of novel molecules. Trimethoprim (TMP) and atenolol (AT) were used as elicitors of secondary metabolism and evaluated by metabolomics experiments for two Streptomyces strains: S. sp. PBH53 and S. armeniacus ATCC 15676. TMP acted as a pathway-specific activator in S. sp. PBH53, but as a global regulator in S. armeniacus, while AT acted as a pathway-specific regulator in both strains. One peak of interest was identified from each treatment, with a total of four molecules chosen for further characterization and structural elucidation. Three of the four are likely uncharacterized molecules, with one having a potential match to salinomycin. TMP and AT prove to be effective elicitors in Streptomyces strains, enabling discovery of uncharacterized natural products. Advancements in whole genome assembly techniques have provided opportunities for natural product discovery to be based on genomic information rather than compound isolation based on biological activity. However, methods for assembling short read sequence data often results in assembly errors in polyketide synthase (PKS) pathways due to repetitive sequences in the acyl-transferase (AT) domains of the pathway. These errors exist in publicly available genomes in the NCBI database, as well as in the assembly of simulated reads of long PKS pathways. Similar errors exist in the S. sp. PBH53 genome, where there are assembly errors in the pathway for Naphthomycin A, but the strain is able to produce the molecule indicating a complete biosynthetic gene cluster (BGC) that doesn’t appear in the assembly. Evaluations of SPAdes and Biosynthetic SPAdes did not show reliable solutions to the assembly errors occurring in genomes. Future work will explore the possibilities of using literature BGC sequences as scaffolds for more accurate assemblies of genomes.