Mining natural products: Scientists unlock new molecules by looking at genome

COVENTRY, U.K.—Scientists at the University of Warwick, led by Dr. Gregory Challis, may be unlocking the door to new drugs and therapeutic targets by mining the genome of microorganisms for hidden metabolic keys. They published their work in the October issue of Nature Chemical Biology. Applied widely, genome mining should provide access to new natural products with molecular diversity unmatched by synthetic chemical techniques.

According to Challis, the traditional approach to natural product research involves screening extracts from various sources using bioassays for a predetermined effect. This system is not particularly efficient as more often than not, scientists end up finding a known natural product from a new source.

Genome mining, however, allows scientists to circumvent some of this uncertainty by either identifying new enzymes or metabolic pathways and determining whether the natural products involved are novel, or starting with a known metabolite and searching for the enzymes involved in its biosynthesis.

"The first process is particularly useful for the discovery of new natural products as potential therapeutics, while the second process can be extremely useful for identifying potential therapeutic targets," he explains. "For example, if a particular pathogen is known to produce a metabolite important for establishing infection, its genome can be mined for enzymes involved in its biosynthesis. They can then be exploited as targets for inhibitor development. Such inhibitors may become important new drug leads."

Challis recently used genome mining to identify a novel nonribosomal peptide synthetase from Streptomyces coelicolor, an organism that has provided the pharmaceutical industry a variety of antibiotic compounds. Using structural, modeling, and genetic manipulation techniques, he found he could accurately predict the enzyme's substrate selectivity and manipulate the metabolic pathways in which the enzyme was involved.

Challis's efforts come at a time, however, when many pharmaceutical companies have ceased to search for new natural products as lead compounds because of the difficulties associated with the traditional approaches to identifying them.

"When [Drug] Discovery programs are run with tight deadlines, there is sometimes a reluctance to allow for the time lag between initial identification of activity and eventual isolation of sufficient amounts of material to conduct preclinical safety and efficacy testing," said Dr. Karen Bush, antimicrobial agent scientist at Johnson & Johnson, in a 2004 Nature Reviews Drug Discovery article.

Challis questions the wisdom of this decision.

"The structural diversity and complexity of natural products cannot be matched by chemically synthesised compound libraries," he says. "Binding to a biomolecular target with high affinity and selectivity requires structural complexity. Recent developments in analytical chemistry technologies together with new approaches--like the genome mining approach--should take a lot of the pain out of discovering novel natural products."

Genome mining efforts like Challis's will also likely play an increasingly important role as government-sponsored microbial genome projects, like the Integrated Microbial Genome system of the DOE Joint Genomes Institute, gain momentum.