Unearthing Nature's Bounty

Natural product hunters have another weapon to add to their arsenal. Rooted in proteomics, the approach turned up a new molecule in a spoonful of dirt from a Louisiana backyard and may aid the search for new biochemical tools and drugs.

More than half of today's drugs are derived from natural products, and many more potentially useful entities await discovery. Traditional bioassay-guided searches have a way of finding the same molecules over and over again, so many researchers have turned to other discovery methods, such as genome mining, which is the use of genome sequences to predict the existence of unknown compounds.

In that spirit, Neil L. Kelleher and coworkers Stefanie B. Bumpus, Bradley S. Evans, Ioanna Ntai, and Paul M. Thomas of the University of Illinois, Urbana-Champaign, have found a way to use gel-based proteomics and mass spectrometry to sift through a range of bacterial strains to find the ones that might make previously undiscovered natural products (Nat. Biotechnol., DOI: 10.1038/nbt.1565).

They need no genome information in advance. Instead, they search bacterial cultures for high-molecular-weight enzyme systems that are likely to be natural product assembly lines. Finding these expressed enzymes is relatively straightforward because they are bigger than most other proteins, and they have a common phosphopantetheinyl cofactor that produces a distinct signature ion in a mass spectrum, Kelleher says.

With the enzyme systems in hand, the team next obtains DNA sequences for all their components. "Because of the work of enzymologists, we can look at the DNA sequence of this machinery and infer a lot about the natural product structure it encodes," Kelleher says. With that information, they then examine bacterial cultures with mass spectrometry and run a targeted search for putatively novel natural products.

The team calls the approach Prism (Proteomic Investigation of Secondary Metabolism). To demonstrate proof of concept, the team examined samples of soil that Louisiana native Evans scooped from his parents' backyard over a holiday break. After two years of development, the team hit pay dirt: They found one bacterial strain from the backyard samples that makes a previously unknown member of a family called the kurstakins. Thanks to Prism, which uncovers biosynthetic enzymes as well as natural products, the Illinois team is the first to figure out how kurstakins are made in nature.

Because this proteomics-based technique doesn't require advance knowledge of a microbe's entire genome sequence, it complements other genome-mining strategies for natural product discovery, all of which currently require that information, says natural products researcher Ben Shen of the University of Wisconsin, Madison.

"This is a very clever and 'outside the box' approach to small-molecule discovery," says Bradley S. Moore of the University of California, San Diego, who studies natural product biosynthesis. "The ultimate combination of this proteomics approach with genomics has a chance of forever shaping the way in which we discover natural products from microbes."