A New Antibiotic Found in Dirt Can Kill Drug-Resistant Bacteria

Tests in mice show that the new drug works against numerous nasty diseases and should stay viable for decades

Soil in Hand
A handful of promising drug candidates? Marnie Burkhart/Corbis

Antibiotics are trusted weapons against many types of bacterial disease, but growing resistance to the drugs is a major problem. “Pathogens are acquiring resistance faster than we can introduce new antibiotics, and this is causing a human health crisis,” says biochemist Kim Lewis of Northeastern University.

Lewis is part of a team that recently unveiled a promising antibiotic, born from a new way to tap the powers of soil microorganisms. In animal tests, teixobactin proved effective at killing off a wide variety of disease-causing bacteria—even those that have developed immunity to other drugs. The scientists' best efforts to create mutant bacteria with resistance to the drug failed, meaning teixobactin could function effectively for decades before pathogens naturally evolve resistance to it.

The 20th century's “antibiotic era” introduced a widely successful, targeted effort against disease-causing bacteria. Drugs like penicillin and streptomycin became household names, and millions of people benefited from them.

But widespread use—and misuse, such as patients not taking the drugs properly—meant that bacteria began working overtime to develop resistance to antibiotics. Now some pathogens, including some strains of tuberculosis, are resistant to all available antibiotics. Because resistance can evolve quickly, the high costs of drug development aren't seen as having long-term value, and fewer new antibiotics are reaching the market.

Part of the problem has been trouble growing the most promising candidates in the lab. Natural microbial substances from soil bacteria and fungi have been at the root of most antibiotic drug development during the past century. But only about one percent of these organisms can be grown in a lab. The rest, in staggering numbers, have remained uncultured and of limited use to medical science, until now.

Lewis and his team decided on a different approach. “Instead of trying to figure out the ideal conditions for each and every one of the millions of organisms out there in the environment, to allow them to grow in the lab, we simply grow them in their natural environment where they already have the conditions they need for growth,” he says.

To do this, the team designed a gadget that sandwiches a soil sample between two membranes, each perforated with pores that allow molecules like nutrients to diffuse through but don't allow the passage of cells. “We just use it to trick the bacteria into thinking that they are in their natural environment,” Lewis says.

The team isolated 10,000 strains of uncultured soil bacteria and prepared extracts from them that could be tested against nasty pathogenic bacteria. Teixobactin emerged as the most promising drug. Mice infected with bacteria that cause upper respiratory tract infections (including S. aureus or Streptococcus pneumoniae) were treated with teixobactin, and the drug knocked out the infections with no noticeable toxic effects.

It's likely teixobactin is effective because of the way it targets disease: The drug breaks down bacterial cell walls by attacking the lipid molecules that the cell creates organically. Many other antibiotics target the bacteria's proteins, and the genes that encode those proteins can mutate to produce different structures. That means the drug's attack isn't always effective, so some hardy bacteria may survive to eventually help build a resistant strain.

One existing antibiotic that also targets lipid cell-wall precursors, vancomycin, worked effectively for nearly 40 years before bacteria developed resistance. The new compound is considerably better protected from resistance than vancomycin, so it may have a very long run of effectiveness, the team reports today in the journal Nature.  

Today teixobactin can cure mice of infection, which is a good start, and the drug is perhaps two years away from beginning the clinical tests that could eventually lead to approval for human treatment. And promising as it may be, teixobactin represents just the tip of the iceberg, Lewis says. Who knows what may be found among the many millions of uncultured soil bacteria species?

“It's a tremendous source of new antibiotic compounds,” Lewis says. “You could imagine all kinds of compounds that could be there and could do all kinds of things. Even apart from antibiotics the compounds you get from soil microorganisms have also been used to develop anti-cancer drugs, immunosuppressants and anti-inflammatories. So really, these bacteria are very good at making antibiotics, but there are definitely many other therapeutics that they can make as well.” 

Get the latest Science stories in your inbox.