Mining genomic islands for novel secondary metabolites gene clusters – MiGenIs

The main goal of our project is to develop and validate a new genome mining method for the discovery of new SMs and their biosynthetic gene clusters. The approach we want to explore differs fundamentally from already established genome mining methods based on the search of homologs of known secondary metabolite biosynthetic genes. Our method relies on the detection of genomic islands in genomes of closely related species. Indeed, species- or strain- specific genomic islands have been observed to be enriched in genes of specialized metabolism. Thus, identifying genomic islands in closely related strain genomes may lead to the identification of secondary metabolism gene clusters that fail to be detected by classical homology-based genome mining approaches. We will apply this approach to exhaustively explore the metabolic capacities of three phylogenetically closely related Streptomyces strains (Streptomyces ambofaciens ATCC23877, DSM40697 and M1013). This approach to identify candidate SM biosynthetic gene clusters will be combined with various strategies aiming at expressing genes that are silent in laboratory growth conditions (changes in culture conditions, mutations in global regulators controlling the specialized metabolism) and with the construction of mutants deleted for the candidate SM biosynthetic gene clusters. Thus it will be possible to establish links between antibacterial activities detected in some culture conditions and candidate clusters. The clusters directing the biosynthesis of antibacterial compounds will be studied, the compounds purified and characterized and the the biosynthetic pathways of novel compounds elucidated.

- A bioinformatics tool for the research and analysis of synteny breaks in genome sequences has been developed. This tool first identifies blocks of synteny in the genomes of two strains, i.e. regions in which homologous genes are in the same relative positions in the genome. It then delineates areas where this synteny is interupted and that may correspond to genomic islands (regions containing specific genes, present in only one of the two genomes). These genomic islands are analyzed. A visualization interface offers different ways to explore the data generated.

- Genomic islands present in S. ambofaciens ATCC23877 were detected after comparing the genome of S. ambofaciens ATCC23877 with the genomes of Streptomyces coelicolor, Streptomyces ambofaciens DSM 40697 or Streptomyces sp. M1013.

- The comparison of the regions in which the genomic islands are present in several strains of Streptomyces has revealed the existence of conserved hot spots in which different genomic islands are found in different strains.

- A genomic island that contains a gene cluster that directs the biosynthesis of several specialized metabolites has been identified and characterized. This cluster is not detected by conventional programs for identifying specialized metabolism genes. This validates our approach and shows that it allows finding specialized metabolism gene clusters not detected by the usual approaches.

- The chemical characterization of one of these specialized metabolites was carried out.

Genomic island have been deleted in the genome of S. ambofaciens ATCC 23877. Some of these islands have also been cloned and will be introduced into other Streptomyces species where they will be heterologously expressed.

Development of a new bioinformatic approach to study the specialized metabolome. In addition to the usual genome mining approach based on sequence similarity, we will develop a new method to identify potential specialized metabolite gene clusters for which no prediction can be drawn from homology-based sequence analyses. Our strategy will be applied to the study of S. ambofaciens specialized metabolism but could find applications with other organisms.
New antibacterial compounds. This project is likely to result in the characterization of new antibacterial agents. Our preliminary results indicate that a derivative of S. ambofaciens ATCC23877 possesses at least two new distinct antibacterial activities. The compounds responsible for these activities and their biosynthetic gene clusters will be identified and characterized.
New biosynthetic pathways and clusters. The characterization of the biosynthetic pathways and clusters for new compounds will provide new tools (enzymes) and raw material (molecules) that can find numerous applications in the fields of combinatorial biosynthesis and green chemistry.
Better understanding of the specialized metabolism. This project will lead to a better understanding of the specialized metabolism at the strain level: regulatory networks, cross-talks, synergies and competitions occurring between various specialized metabolite pathways. It will also give an overview of the intraspecific diversity of the specialized metabolism and possibly of its evolution. Finally, the data generated could facilitate further studies aiming at establishing specialized metabolism regulation networks in Streptomyces or at determining the ecological functions of specialized metabolites.

Complete genome sequence of Streptomyces ambofaciens ATCC 23877, the spiramycin producer. Thibessard A, Haas D, Gerbaud C, Aigle B, Lautru S, Pernodet JL, Leblond P. J Biotechnol. 2015 Nov 20;214:117-8.
doi: 10.1016/j.jbiotec.2015.09.020.

Complete Genome Sequence of Streptomyces ambofaciens DSM 40697, a Paradigm for Genome Plasticity Studies. Thibessard A, Leblond P. Genome Announc. 2016;4(3). pii: e00470-16.
doi: 10.1128/genomeA.00470-16.

With this project, we aim at adding a new tool/component to the secondary metabolite genome mining toolbox/kit. The approach we want to explore differs radically from already established methods. Indeed, all genome mining methods developed so far rely on the identification of secondary metabolite biosynthetic candidate genes and clusters, through in-silico sequence similarity-based analyses of genome sequences. Although very useful and having led to the discovery of many natural products (see for example, Nikolouli and Mossialos, 2012), these methods can only conduct to the discovery of products synthetized by enzymes with sequences similar to enzyme sequences already reported in the literature. Thus, products whose biosyntheses involve uncharacterized families of enzymes escape these approaches.
To identify natural product biosynthetic gene clusters undetected by sequence similarity-based genome analyses, our method relies on the detection of genomic islands in genomes of closely related species. Indeed, it has been observed repeatedly that most secondary metabolite gene clusters reside in genomic islands (see for example, Penn et al., 2009). In addition, secondary metabolomes are often strain-specific and even closely related species can possess very different sets of secondary metabolite gene clusters (for example S. ambofaciens and S. coelicolor, see below). Therefore, identifying genomic islands in closely related species constitute a good starting point for the localization of natural product gene clusters. The next step consists in establishing a link between the identified islands and natural products isolated by metabolomics (OSMAC approach, LC-MS analyses) or by their biological activity. Newly discovered biosynthetic genes can then be characterized using traditional functional analysis (involving gene deletion and characterization of biosynthetic intermediates).
As a proof of concept, this new genome mining approach will be applied to three Streptomyces ambofaciens strains: S. ambofaciens ATCC23877, DSM40697 and M1013 for which complete (S. ambofaciens ATCC23877) or draft (S. ambofaciens DSM40697 and M1013) genome sequences are available in the laboratories involved in this project. Initially, the project will focus on the genome islands of S. ambofaciens ATCC23877, as we have isolated several antibacterial activities and one metabolite that could not be linked to any of the secondary metabolite gene clusters detected by traditional in-silico sequence similarity searches. It will then be broadened to the genome islands identified in the two other S. ambofaciens strains.