Published online 9 April 2008 | Nature | doi:10.1038/news.2008.742


Bacteria designed to search out pesticides

Biological switch triggers E. coli to swim towards chemical.

Researchers have hacked into the navigation system of the bacterium Escherichia coli, causing it to hunt down a widely used herbicide called atrazine. The finding could help improve efforts to clean up the environment using biological tricks.

Escherichia coli has receptor proteins on its cell surface that can identify chemicals of interest, enabling the bacterium to follow a chemical along its concentration gradient to its source. The recognition information is passed along the cell, eventually triggering its whip-like tail, or flagellum, to rotate either one way to move forward or the other way to tumble randomly.

Engineered E. coli bacteria follow an S shape delineated by the molecule theophylline.J. Gallivan

But this chain of events can be intercepted by tailoring the bacterium’s RNA. To do this, the researchers used a strain of E. coli that lacked the gene needed to move cells forward. Using these modified cells, the group engineered a segment of RNA, or riboswitch, containing the gene. In the presence of atrazine, the switch was turned on, allowing the bacterium to move toward higher concentrations of the chemical1.

"The cool thing here is that we can get cells to respond to things they wouldn’t ordinarily care about," says Justin Gallivan of Emory University in Atlanta, Georgia. He reported his latest results on 6 April at a meeting of the American Chemical Society in New Orleans, Louisiana.

Bait and switch

Some bacteria can metabolize atrazine, creating another chemical by-product as waste. The team members showed that they could create a riboswitch that recognized atrazine but not the by-product. By incorporating genes from atrazine-eating bacteria, it should be fairly straightforward to engineer a bacterium that can seek out and destroy the pesticide, Gallivan says.

But there may be some limitations. Surface bacterial receptors can distinguish between very slight differences in chemical concentrations in their environment. But because a chemical must enter a cell in order to be caught up by a riboswitch, the bacteria are less sensitive to slight differences in concentration than they would be if surface receptors were sensitive to the chemical.

What's more, using riboswitches to control movement may be fine on gel in a Petri dish, but too slow to work in a liquid environment. " E. coli swim fast, so they would need to respond quickly," says Mark Goulian, a biophysicist at the University of Pennsylvania in Philadelphia.

Easy selection

Still, the biggest advantage to using riboswitches may be the speed with which they can be identified. By contrast, re-engineering existing receptor proteins to become sensitive to a different molecule is time-intensive, depends on unpredictable molecular folding, and sometimes succeeds only in making a receptor sensitive to a wider range of molecules, Gallivan says.

But the number of possible riboswitch sequences is fairly small, and Gallivan's team can sort through all of them in a single experiment. If there is a riboswitch that fits a small molecule of interest, Gallivan says, "you're reasonably certain you'll find it".


The best riboswitches can then be selected by looking for E. coli that move the most. “The really cool thing about what he’s doing is that he’s able to get these switches quickly using this method,” says Goulian.

Ultimately, making bacteria move or stop may not be the only use of riboswitches. Gallivan says he hopes to use multiple riboswitches to trigger genes in a specific order. This would enable E. coli to follow a sequence of complex tasks — such as sticking to or 'picking up' a small sphere in one area and releasing it elsewhere. 

  • References

    1. Topp, S. and Gallivan, J.P. J. Am. Chem. Soc. 129, 6807-6811 (2007).
Commenting is now closed.