Caltech researchers breed new genes to make natural products in bacteria

PASADENA—Using a new process of "sex in the test tube," a California Institute of Technology research group has been able to mate genes from different organisms and breed new genetic pathways in bacteria. These bacteria make an array of natural products that are naturally found in much more complex organisms.

The natural products, which are carotenoids similar to the pigment that gives carrots their color, are made by many different plants and microbes, but are totally foreign to the E. coli bacteria the researchers used. The new results, reported in the July issue of the journal Nature Biotechnology, show that the carotenoid-producing genes from different parent organisms can be shuffled together to create many-colored E. coli. Many of the carotenoids made in the bacteria are not even made by the organisms from which the parent genes came.

One of the reddish products, torulene, is not produced by any known bacteria, although it is found in certain red yeasts. "With molecular breeding, the experimenter can train the molecules and organisms to make new things that may not even be found in nature, but are valuable to us," says Frances Arnold, professor of chemical engineering and biochemistry at Caltech and coauthor of the new study.

Conceptually similar to dog breeding, the process generates progeny that are selected by researchers on the basis of attractive features. In this study, former Caltech researcher Claudia Schmidt-Dannert (now on the faculty at the University of Minnesota) and Caltech postdoctoral researcher Daisuke Umeno selected the new bacteria by their color.

This process of directed evolution, which Arnold has been instrumental in developing, is capable of creating new biological molecules and even new organisms with new or vastly improved characteristics. Unlike evolution in nature, where mutations are selected by "survival of the fittest," directed evolution, like breeding, allows scientists to dictate the characteristics of the molecules selected in each generation.

"We are now able to create natural products that usually have to come at great cost from esoteric sources simply by breeding ordinary genes in ordinary laboratory organisms," says Schmidt-Dannert.

The researchers believe that this method will be widely useful for making complex and expensive natural molecules such as antibiotics, dyes, and flavors. "Imagine being able to produce in simple bacteria many of the compounds that come from all over nature," says Arnold.

And, according to the authors, an even more irresistible target of directed evolution is finding bacteria that make biological molecules not yet found in nature.