As scientists continue their exploration of the thousands of organisms that live inside our body, a new study is showing just how much one “good” bacteria can be exploited to its host’s advantage. Scientists were able to engineer this bacteria to find and attack a harmful pathogen in mice and worms, preventing and halting infections in the process.

The thousands of bacteria and fungi that live inside us are called the microbiome, and the “good” bacteria and yeast are often referred to as a probiotic. Because mice and worms are different from people — and their microbiomes are accordingly different — it’s too early to know if this lab-made bacteria will work in humans. But the study, published today in the journal Nature Communications, is a good example of how we might be able to reprogram the body’s microorganisms to precisely attack pathogens. The hope is that these genetically engineered probiotics could one day help us prevent infections, or even provide an alternative to antibiotics, which can lead to resistance in bacteria.

“From the animal studies it appears viable,” study co-author John March, a professor in the Department of Biological and Environmental Engineering at Cornell University, writes in an email to The Verge. But “since we can’t test this in people yet, it’s tough to know how it will perform.”

The probiotic used in the study is called Escherichia coli Nissle 1917. You might know E. coli for causing diarrhea, UTIs, and pneumonia, but most strains are actually harmless, and E. coli Nissle 1917 is one of them. It was first isolated from the poop of one lucky World War I soldier who, unlike the others in his trench, wasn’t affected by an outbreak of heavy, bloody diarrhea, caused by bacterial dysentery. The probiotic strain has been shown to have beneficial effects on some gut disorders, and today, it’s often taken as the supplement Mutaflor.

Researchers had previously reprogrammed a strain of E. coli to detect and kill the pathogen Pseudomonas aeruginosa. This bacteria is widely known for causing serious infections in hospital patients and people with a weakened immune system. In that study, however, the pathogen was killed in the lab, not inside an animal, so it was unclear if the system would actually work in a living being.

In today’s study, the researchers engineered the probiotic E. coli Nissle 1917 to attack the same pathogen, P. aeruginosa, in two animal models. The microorganism was engineered to destabilize the sticky, mat-like colonies that allow the P. aeruginosa bacteria to cooperate with each other, called biofilms. These biofilms make it hard for the immune system or antibiotics to fight P. aeruginosa infections. The probiotic was basically reprogrammed to detect the pathogen, break open its protective barrier, and kill it, says March. In roundworms and mice, the engineered E. coli Nissle 1917 was able to thwart the activity of P. aeruginosa during gut infection. The probiotic was especially efficient in preventing the onset of an infection rather than fighting a preestablished one, the study says.

In mouse intestines, for example, the E. coli strain stuck around for up to three weeks. That suggests that a single dose of the engineered antibiotic could provide protection from P. aeruginosa infections for several weeks. So in the future, it could be used to prevent infections in at-risk patients, says lead author Matthew Chang at the National University of Singapore. Or it could be even used as an alternative to antibiotics. Inevitably, of course, bacteria adapt to our antibiotics, becoming resistant — something that’s been seen since the very first drug, penicillin. No new classes of antibiotics have been introduced since the 1980s, so we’re starting to run out of the lifesaving drugs that most modern medicine relies on. That makes today’s finding more than a curiosity: anything that can help keep antibiotics in reserve might help slow the resistance crisis.

“Resistance to antibiotics is something that will always exist, I expect,” says March. But this organism may prove more effective than a standard antibiotic, since it’s attacking along several lines of resistance, he says.

For now, however, the lab-made probiotic was only tested in animals, so it’s unclear whether it would work in people. And a host of things could go wrong when the engineered microorganism is tried in humans. “One can imagine that the host won't accept the probiotic, although this is pretty rare,” say March. “The host could have an allergic reaction much in the same way that could happen with any drug.”

So there’s still lots of work ahead — but today’s findings suggest the enormous potential hidden in our microbiome.