Scientists at MIT encouraged bacteria to produce a novel antibiotic by pitting them against a microbial enemy. The newly discovered compound can kill H. pylori, bacteria linked to stomach ulcers. The approach could provide a new way to discover novel antibiotics and shed light on how and when bacteria churn out these toxic compounds.

"The lab is a tame place if you're a bacterium: you don't have to fight for a crystal of sugar," says Philip Lessard, a molecular biologist at MIT who collaborated on the work. "So maybe we're not seeing them spitting out chemical-warfare compounds like they would normally."

Antibacterial resistance--when bacteria become invincible to a particular drug--is becoming a major crisis in American hospitals. According to the Centers for Disease Control and Prevention, approximately two million Americans acquire infections while in hospitals every year, 90,000 of which are fatal. About 70 percent of those infections are resistant to at least one type of antibiotic.

Scientists across the globe are looking for ways to make new antibiotics. Some projects involve melding existing drugs into potent new molecules, while other approaches focus on designing new drugs that target specific mechanisms of microbial resistance. But recent sequencing studies suggest that bacteria possess an untapped well of novel antibiotics that they don't produce under normal lab conditions, thereby remaining hidden to scientists for decades.

Scientists working in Anthony Sinskey's lab at MIT sequenced the genome of a strain of soil-dwelling bacteria known as Rhodococcus fascians. They were surprised to find that this organism, not known for its antibiotic-producing powers, harbored a number of genes involved in the metabolism of antibiotic-like compounds. (In the wild, bacteria produce antibiotics as a survival mechanism, to clear themselves a niche in the crowded microbial world.)

While Rhodococcus seemed genetically capable of producing the compounds, the organisms did not do so in the lab--until, that is, they were grown alongside another type of bacteria, called Streptomyces, which are among the most prolific antibiotic producers in the microbial world. Microbiologist Kazuhiko Kurosawa and his colleagues published their discovery last month in the Journal of the American Chemical Society.

The novel compound, dubbed rhodostreptomycin, belongs to a class of antibiotics known as aminoglycosides, which include neomycin, used in many first-aid creams, and streptomycin, a tuberculosis drug. While it's unclear if the drug would be appropriate for clinical use, early tests show that it can kill H. pylori, bacteria linked to stomach ulcers, and it can survive highly acidic environments like that of the stomach. The molecule also appears to contain a novel structural component, which could provide a jumping-off point for chemists keen to design new drugs. "This opens a new domain in the chemical-diversity space," says Lessard.