Many scientists searching the seas for drug candidates hope to find novel antibiotics. Meanwhile, terrestrial soil, once a fertile source, hasn’t produced a new antibiotic in decades. But an innovative method for cultivating soil bacteria could have drug developers digging in the dirt once more.

Last January, researchers at Boston’s Northeastern University announced they had discovered a new type of antibiotic, known as teixobactin, using a novel tool called the iChip. Normally, scientists use a laboratory Petri dish to culture bacteria they have isolated from soil. But only about 1% of bacteria found in soil will grow that way. To get around that obstacle, biochemist Kim Lewis and biologist Salva Epstein decided to provide a more natural environment. Epstein’s lab developed the iChip, which is about the size of a flash drive and has 384 tiny wells covered by a semi-permeable membrane. After bacteria cells isolated from a soil sample are placed in each well, the iChip is buried in the very ground the soil came from.

Using this method, Lewis says it’s possible to cultivate about half of the bacteria in soil. He and Epstein have used the iChip to discover 25 new antimicrobials with strong antibiotic activity under the auspices of their new company, NovoBiotic Pharmaceuticals. The most promising antimicrobial so far has been teixobactin: In lab and animal testing, it kills a wide variety of pathogens, including multi-drug resistant Staphylococcus aureus (MRSA). Even more intriguing is that it appears infectious bugs can’t figure out how to thwart teixobactin. That’s normally the big problem with antibiotics—the more they’re used, the more pathogens evolve to defend against them. “This one apparently evolved to be largely free of resistance,” says Lewis.

Teixobactin was derived from bacteria isolated from soil dug in a grassy field in Maine. Lewis says that soil in extreme and exotic environments such as deserts or rain forests “probably” holds even more diverse and potent microbes. But he notes that the 10,000 bacteria strains his group has identified using the iChip include microorganisms dug up in the backyard of a colleague who lives in a Boston suburb.

A resistance-proof antibiotic, and a new method for finding more such drugs, would be a huge advance, but enthusiasm from other scientists is measured. “This is a wonderful finding,” says Stuart Levy, director of the Center for Adaptation Genetics and Drug Resistance at Tufts University School of Medicine in Boston, and president of the Alliance for the Prudent Use of Antibiotics. “But other than pure alcohol, I’ve never seen any drug that hasn’t elicited resistance.” NovoBiotic Pharmaceuticals is refining teixobactin’s chemistry for further testing, and has also identified a compound that shows potential as a cancer drug.