Leveraging Microbial Genomes and Genomic Context for Chemical Discovery

The genomic era has dramatically changed how we discover and investigate microbial biochemistry. In particular, the exponential expansion in the number of sequenced microbial genomes provides investigators with a vast wealth of sequence data to exploit for the discovery of biochemical functions and mechanisms, as well as novel enzymes and metabolites. In contrast to early biochemical work, which was largely characterized by "forward" approaches that proceed from biomass to enzyme to gene, the availability of genome sequences enables the discovery of new microbial metabolic activities, enzymes, and metabolites by "reverse" approaches that originate with genetic information or by approaches that incorporate features of both forward and reverse methodologies. In the genomic era, the canonical organization of microbial genomes into gene clusters presents a singular opportunity for the utilization of genomic data. Specifically, genomic context (information gleaned from the genes surrounding a gene of interest in the chromosome) is a powerful tool for chemical discovery in microbial systems because of the functional and/or physiological relationship that usually exists between genes found within a gene cluster. This means that the investigator can use this inferred link to generate hypotheses about the functions of individual genes in the cluster or even the function of the entire cluster itself. Here, we discuss how analysis of genomic context in combination with a mechanistic understanding of enzymes can facilitate numerous facets of microbial biochemical research including the identification of biosynthetic gene clusters, the discovery of important and novel enzymes, the elucidation of natural product structures, and the identification of new metabolic pathways. We highlight work from our laboratory using genomic context to discover and study biosynthetic pathways that produce natural products, including the cylindrocyclophanes, nitrogen-nitrogen bond-containing metabolites, and the gut microbial genotoxin colibactin. Although use of genomic context is most commonly associated with studies of natural product biosynthesis, we also show that it can be applied to the study of primary metabolism. We illustrate this with examples from our work studying the members of the glycyl radical enzyme superfamily involved in choline and 4-hydroxyproline degradation in the human gut. Looking forward, we envision increased opportunities to use such information, with the combination of biochemical knowledge and computational tools poised to fuel a new revolution in our ability to connect genes and their biochemical functions. In particular, we note a need for methods that computationally formalize the functional association between genes when such associations are not obvious from manual gene annotations. Such tools will drastically augment the feasibility and scope of gene cluster analysis and accelerate the discovery of new microbial enzymes, metabolites, and metabolic processes.