Genome Mining Approaches to Bacterial Natural Product Discovery

When the first whole genome sequences of the model streptomycete Streptomyces coelicolor A3(2) and the avermectin producer Streptomyces avermitilis were obtained and analyzed,1,2 there was a big surprise that the genomes of these well-studied strains did not only code for biosynthetic pathways of their “known” antibiotics, such as actinorhodin (S. coelicolor) or avermectin (S. avermitilis), but contained many other gene clusters with genes coding for enzymes commonly found in natural product biosynthetic pathways. With an increasing number of other whole genome sequences of antibiotics/natural products-producing bacteria, it became evident in the following years that the genetic potential, that is, the abundancy of so-called secondary metabolite biosynthetic gene clusters (BGCs) is a lot higher than the number of compounds known from these organisms. Thus, the approach of “genome mining,” which tries to identify such BGCs in genomic data and uses these data to identify the corresponding chemical molecules, has become a widely used complementary approach to the classical—often bioactivity-guided—screening for novel natural products. In this article, we provide an overview on the background of the currently used genome mining technologies, its impact on academic and industrial natural products research and an outlook on how this growing multidisciplinary field may develop in the future.