A mushroom under attack from bread mold(Phil Sandlin / AP Photo)

Some interesting things have been found in trash cans throughout history, no doubt, but perhaps few with as much potential to change the world as what Tradd Cotter discovered in his.

Returning to the lab after a weeklong vacation in 2012, Cotter, owner of Mushroom Mountain––a mushroom farm and research facility in Greenville, South Carolina––noticed that in a contaminated Petri plate he'd discarded a week earlier, two fungi were waging what he calls a “massive attack” on one another. A fungus he’d grown in the lab had spread throughout the plate to destroy another fungus that had snuck in from the air and was trying to reach a food source––in this case, agar. Intrigued by the patterns the battling molds were forming, Cotter brought the plate to a well-lit room and took a variety of high-resolution photographs. "I zoomed in and noticed that there were these droplets all over the surface of the competitor mold and nowhere on the body of the [original] fungus,” he says. “And I thought, 'Maybe it's producing some kind of weapon.’”

The “weapon” to which he’s referring is secondary metabolites––finely calibrated compounds that mushrooms and other organisms produce as defenses against dangerous microbes. What caught Cotter's attention wasn't the metabolites themselves, but that these metabolites were being created in an regulable environment. If he could “sweat this fungus out” against a variety of other molds, he realized, he might be able to produce shields for his other mushrooms from a whole range of attackers.

So he did just that: He started matching metabolite-producing fungi in his lab with the most common contaminates––wild fungi that often cost him thousands a year in ruined product. He then used the resulting metabolites to "clean" his growing media––like a natural, pre emergent fungicide. And it worked. The secondary metabolites kept pesky contaminates at bay, allowing Cotter to better culture his fungi for his business.  

Thanks to their strong antimicrobial properties, secondary metabolites have long been exploited in medicine for use against infections. Penicillin and ethanol are both products of fungal metabolites. Lovastatin and cyclosporine, too. Focused on defending his mushrooms, Cotter at first didn’t see any medicinal potential in his discovery. But after two years of experimenting, it clicked: If his mushrooms could grow tailor-made weapons against any other types of fungi, would it be possible for them to do the same against any type of bacteria, too?

*   *   *

The rise of drug-resistant bacteria is sobering. Just last week, colistin-resistant E. coli––a “superbug” resistant to the antibiotic that’s considered the last resort for combatting particularly dangerous types of infections––landed in the U.S. Soon, public health officials anticipate, infections will be harder to stop; 10 million people could die of drug-resistant superbugs annually by 2050 due to a lack of effective medicine.

It may be a long shot, but it’s conceivable that Cotter's process offers a new kind of hope. While scientists have been working on the problem of antibiotic resistance for many years—some are looking to harness the human immune system to better fight it; others are working on simply detecting the superbugs faster—his vision is to beat superbugs with medicine that actually adapts to destroy them. It’s not pharmaceuticals he has in mind; he's not planning to mass produce many different types of secondary metabolites. Rather, he believes it’s his unique style of co-culturing itself––the process of culturing two different microbes together to produce a defense entirely specific to the attacker––that may be able to create custom antibiotics that, at least in theory, could be inherently less susceptible to resistance.

His goal, in other words, is to grow mushrooms that are themselves medicine, because they could create whatever metabolites a sick person needs.

"The best situation I could describe is something everyone has gone through, like a strep throat culture,” Cotter says, imagining a scenario in which an infected patient walks into the doctor’s office, gets a throat swab, and then has the swab dropped into a specially designed module containing a fungus. That fungus would then sweat metabolites into a reservoir that would be naturally calibrated to combat the patient’s illness.

Cotter doesn’t know how the metabolites would be administered yet. A lollipop or throat spray for strep? Delivered topically for staph? His testing is still thoroughly ongoing. Should he receive the NIH grant he's applying for––a grant backed by a $1.2 billion White House Initiative to stop resistant diseases––answers could arrive rapidly. Analytical labs would go up, animal testing would begin, streptococcus lollipopus before we know it.

But Cotter’s dramatic departure from traditional pharmaceuticals sits uneasy with some scientists I spoke to about his idea. For starters, it strikes many as dangerous. In medicine, science painstakingly searches for and isolates helpful metabolites, because metabolites can be toxic, too. Cotter’s process would have to find a way to identify metabolites that are harmful to people and somehow remove or neutralize them first. “Maybe you found a fungus that produces a metabolite that kills strep. Well fungi don’t just make one metabolite,” says Nancy Keller, a University of Wisconsin microbiologist. “The fungus is likely making many other metabolites, and although one might be useful, other metabolites might be injurious.” At worst, the magic Cotter’s hoping for may just be pseudoscience.

Not everyone was so dismissive, though. Aaron Hawkins, a microbiologist and chemical engineer at the Danish biotechnology company Novozymes, described Cotter's process as a unique, "almost holistic" idea, comparing it to the difference between herbal and allopathic medicine, in that herbal medicine relies on both active and supporting compounds to heal people. It's the difference between taking vitamin C versus eating an orange; the vitamin C is only part of what makes it healthy.

Still, Hawkins concedes that “in the way that modern medicine looks at things,” Cotter’s process “is really messy.”

*   *   *

Science isolates. Science synthesizes. But in the case of rapidly adapting bacteria, science also seems to be falling behind. Significant time, money, and effort are required to get traditional antibiotics to market (if and when we discover them), and once there, with no adaptivity, these drugs often become sitting ducks for disease resistance. Penicillin had its first resistant bacterium within two years. Methicillin rapidly begot MRSA. Ditto: VRE, Salmonella, C. difficile—the list goes on. Our current drugs may be doing a better job of arming dangerous bacteria than defeating them.

Cotter's concept, early in the process though it may be, aspires to give medicine a leg up with an inherently less-susceptible approach. It's adaptive, and allows the fungus to decide—as fungi have since they first had to defend themselves against bacteria—what it would take to best defeat each bacteria in the moment.

Currently, Cotter is still testing the toxicity of these secondary metabolites while "playing matchmaker" to locate the best fungus or fungi for any given bacterium. He is also working with agricultural pathogens—botrytis (known for attacking wine grapes) and sclerotinia (known for causing white mold), for instance––which is an area where his technique first may be applied. Part of this process has been licensed out to Clemson University, where they are better equipped to manage drug-resistant bacteria. And though advisors there couldn't yet comment on specific results of the trials—due to a patent Cotter has on his technique and his collaborative agreement with the university—Clemson microbiologist Tamara McNealy did confirm the superbug MRSA is actively being tested.

According to Cotter, a paper from McNealy on the results could be published within the next few months. If the results are promising, they may quell some concerns from the scientists who, until they could test things in their own labs, will remain incredulous.

That said, there was certainly a palpable curiosity among the researchers to whom I spoke. As Keller, the Wisconsin microbiologist, notes, the basic concept of letting the microbes keep outsmarting each other at least has a long precedent of success. It’s "a nice idea,” she says, “as it's probably been going on in the soil for eons."

We want to hear what you think about this article. Submit a letter to the editor or write to letters@theatlantic.com.