To meet the growing need for improved antibiotics, scientists a few years ago began to develop a drug-screening method based on arrays of genetically modified bacteria. That method, they now report, can be used to rapidly identify compounds that kill bacteria by inhibiting targets inside the bacterial cell.

In the method, researchers engineer bacteria to produce very low levels of an essential protein. Just barely alive, these bacteria are placed on an array and systematically exposed to a library of compounds. Compounds that kill more than one bacterium are strong drug candidates.

Because the array contains bacteria with related genetic flaws, the method reveals compounds that inhibit several targets. These will be more robust against antibacterial resistance than drugs that inhibit a single target protein.

"This method is a way to screen lots of compounds and look for very specific anti-microbial activity and lowered potential for acquired bacterial resistance," says David L. Pompliano, of GlaxoSmithKline in Collegeville, Pennsylvania, who led the study. The research, which included testing the strategy on E. coli bacteria, was done at DuPont Pharmaceuticals.

The most time-consuming part of the process is to work out the growth conditions for the bacterial strains. Some protein targets are expressed at levels far greater than what the bacterium needs to survive, and the goal is to get the expression down to the minimum level required for life.

"The beauty of this strategy is that you don't need to know the function of the target protein to do the study," says Pompliano. "So you can go to bacterial genomes and pick out a protein target found in many bacteria."

The DuPont team that developed the method is no longer together, but they published a description of the strategy in this month's Nature Biotechnology. Just hours after the study was posted on the Internet, Pompliano received emails from researchers in India and elsewhere expressing interest in the method.