The world desperately needs new medicines to treat deadly fungal infections.
According to a report from the World Health Organization, only four antifungals were approved by the US Food and Drug Administration between 2014 and 2024. Meanwhile, pathogenic fungi are becoming increasingly resistant to legacy treatments, and climate change is accelerating the problem.
“There’s real problems in the future if we don’t get new antifungal treatments,” says Jason Micklefield, a chemical biologist at Imperial College London.
The polyene natural product amphotericin B is a powerful antifungal agent that has been in use for decades, but it can damage the kidneys. Micklefield and his team sought a way to make a polyene that would be less harmful to humans while still deadly to fungi—and hopefully easy to produce on scale. So they turned to nature’s tools for making molecules.
The researchers found two new enzymes capable of modifying polyenes and used them to make a series of new natural product derivatives with improved properties (Nature 2026, DOI: 10.1038/s41586-026-10834-8).
Polyenes with multiple sugar groups on their backbones tend to be more soluble and more potent than those without, so the researchers first searched a database of polyene-producing bacteria and related microbes for glycosyltransferase enzymes.
The team identified one capable of attaching the sugar L-digitoxose to several different polyenes. The researchers were even able to insert the enzyme’s gene cluster into a different strain of bacteria, a first step toward a fermentation platform. To modify a carboxyl group associated with higher toxicity, the team found an amidotransferase enzyme and expanded its scope to accept larger polyenes and a range of amines.
The researchers used these two enzymes to modify nystatin, a polyene used to treat candida infections, to make a series of novel compounds. They tested each molecule against a panel of 10 of the nastiest pathogenic fungi and found that the new molecules were often more potent than the parent compound. In some cases—notably, when tested against a drug-resistant strain of Aspergillus fumigatus—they slightly outperformed amphotericin B.
The group chose one of the series, Nys34, for additional testing and found that it significantly beat back A. fumigatus in mice and showed lower toxicity in human cells than amphotericin B. Micklefield says one of the team’s next steps is to try to untangle the molecule’s mechanism of action, which is different from that of amphotericin B.
“The world needs a better, safer polyene, and we need tons of it,” says Martin Burke, a chemist at the University of Illinois Urbana-Champaign who also works on polyene antifungals but was not involved in this work. Burke says the new work paves the way for more-efficient production of these important medicines.
The modified molecules aren’t dramatically more potent than their predecessors, and their toxicity needs more study, but this paper is still “a very exciting step,” Burke says. “It’s really about synthesis and about showing this possibility of fermenting better molecules.”
Micklefield says he has a wealth of ideas for what he’d like to do next. He has filed a patent and is currently seeking industry partnerships to continue developing Nys34 as a drug candidate. He and his team are also looking into engineering the enzymes to access a wider range of molecules.
The ultimate goal, Burke says, is to have a “palette of enzymes” for fermenting bespoke polyene derivatives in a scalable fashion.
