Natural products, genetically encoded small molecules that appear to have evolved to interact with specific cellular targets and evade cellular defense mechanisms, have been the inspiration for the majority of FDA-approved drugs. They have also served as the starting point for the development of many small-molecule probes that are used to interrogate complex biological systems. One key insight from the large-scale sequencing of bacterial genomic DNA is that traditional, pure culture-based strategies have only provided access to a small fraction of the biosynthetic diversity encoded in bacterial genomes. Developing methods to access these previously inaccessible natural product biosynthetic pathways in bacterial genomes should increase the diversity of small molecules available to test as probes of biological processes and therapeutic agents. My research is focused on the discovery and characterization of natural products encoded by gene clusters found in the genomes of both uncultured bacteria and bacterial pathogens.
Natural Products from Uncultured Bacteria
Many lines of evidence indicate that conventional, pure culture-based approaches have cultured only a tiny minority of soil microbes. Soil microbes that have not yet been cultured outnumber their cultured counterparts by two to three orders of magnitude. A single gram of soil is estimated to contain more than 10,000 bacterial species, greater than 99 percent of which have likely never been cultured in the laboratory. This uncultured majority no doubt produces secondary metabolites that could serve as molecular probes of biological processes and future therapeutic agents. Although there is currently no systematic way to culture this large collection of unstudied microorganisms, it is possible to isolate microbial DNA directly from environmental samples (environmental DNA, eDNA). The heterologous expression of natural product biosynthetic gene clusters captured on large fragments of eDNA cloned into easily cultured bacteria should provide access to many of the small molecules produced by these previously uncultured bacteria.
The immense size of environmental microbiomes has made it difficult to clone natural product biosynthetic gene clusters directly from environmental samples. We have optimized eDNA library construction protocols so that we can readily generate libraries with more than 10 million clones from DNA isolated directly from an environmental sample. Each eDNA megalibrary contains in excess of 400 billion base pairs of genomic DNA, the equivalent of more than 100,000 unique 4-megabase bacterial genomes. These megalibraries are being constructed from soils that have been collected from "biodiversity hot spots" around the world, including spots in Central America, Africa, and the United States. This geographically and ecologically diverse collection should provide us access to a large collection of novel natural product biosynthetic gene clusters present in the environment.
I have worked extensively on the development of methods for accessing new natural products from large eDNA libraries. Using culture-independent strategies, we have identified novel natural products, new biosynthetic enzymes, and new examples of bacterial signaling systems. We are expanding these culture-independent strategies to provide access to a more diverse collection of the molecules that are encoded in genomes of uncultured bacteria. Although capturing the genetic diversity present in environmental samples is now a relatively straightforward process, screening large libraries to identify clones that might yield new natural products remains a challenge.
We are using both expression-dependent and expression-independent screening to identify eDNA clones that produce new biologically active natural products. In expression dependent screening, we shuttle large eDNA libraries into a diverse collection of easily cultured model bacterial hosts. We then examine these libraries in high-throughput phenotypic assays to identity those clones that produce clone-specific secondary metabolites.
In expression-independent screening, clones containing genes known to be involved in the biosynthesis of natural product substructures commonly seen in bioactive compounds are recovered from large eDNA clone libraries. We then test each clone in a variety of bacterial hosts for the ability to confer the production of novel small molecules to the host. By examining the biosynthetic capacity of naturally occurring bacterial populations, instead of just the small fraction of bacteria that are easily cultured in the laboratory, we hope to increase the likelihood of finding compounds with higher potency, new biological activities, and novel means of circumventing resistance mechanisms.
Cryptic Pathways from Bacterial Pathogens
Small molecules play important roles in the establishment and propagation of bacterial infections. Cryptic small-molecule biosynthetic gene clusters that do not appear to encode the biosynthesis of any known metabolites are routinely found in sequenced bacterial genomes. This group of pathways, like the pathways we are studying from uncultured bacteria, have not been previously available for study using traditional microbiological methods. In bacterial pathogens, these cryptic pathways represent the pool of pathways from which additional signaling systems and toxins will be found. My research group is using the same heterologous expression strategies that we use to study pathways cloned from eDNA samples to study cryptic pathways found in the genomes of sequenced bacterial pathogens.
The development of robust DNA-based methods to access new natural product biosynthetic pathways from both cultured and uncultured bacterial genomes should increase the number and diversity of natural products that are available to test as probes of biological processes. This, in turn, should shed light on how best to use this large collection of previously inaccessible metabolites for the benefit of human health.
Grants from the National Institutes of Health provided partial support for these projects.
As of May 30, 2012