Fungal diseases pose an escalating global health problem, driven by antimicrobial resistance (AMR) and a critical shortage of new antifungal drugs in development. Unlike bacterial infections where there remains a pipeline of treatment options, the development of new antifungals is more problematic. Fungal cells are more closely related to human cells, so developing new drugs that are selective for killing fungal vs human cells (e.g. non-toxic) is very challenging.
A recent World Health Organisation (WHO) report emphasises the urgent need for novel drug treatments to combat the escalating threat posed by fungal diseases. Fungal infections already kill more people than malaria, and mortality rates are rising. Moreover, the spread of fungal diseases is likely to increase with climate change, as pathogens adapt to better tolerate human body temperature. Without immediate intervention, Candida auris, Aspergillus fumigatus, and other emerging superbugs that are resistant to the limited number of current drugs, could cause more severe outbreaks of disease, posing a significant risk to countless lives and overwhelming healthcare systems globally. Fungal diseases have already devastated other forms of animal life. If the next human pandemic emerges as a fungal pathogen, we would be poorly equipped to respond, and the consequences could be catastrophic.
Current fungal treatments rely on polyenes, an important class of antifungal drugs. However, the principal member of this class, amphotericin, is associated with substantial toxicity, highlighting the urgent need for safer antifungals. Modifications to amphotericin to improve its safety profile rely on low yielding multi-step chemical syntheses, which would preclude affordability in underdeveloped parts of the world where deaths from fungal infections are highest, such as the global south.
Our research focuses on the discovery of next generation polyene antifungals and the use of an enzyme driven approach to improve existing polyene drugs. We developed a bioinformatics discovery pipeline that revealed a wealth of biosynthetic diversity in soil bacterial species capable of producing novel polyene antifungals. Following isolation from fermentation extracts, the new molecules were structurally elucidated using advanced NMR spectroscopy, leading to the discovery a new family of polyene antifungals which we named kasufungins, and previously unreported congeners of mycoheptin (Figure 1).
A notable feature of these molecules was the presence of a rare second sugar motif, which led us to investigate the glycosyltransferases (GTs) responsible for this unusual modification. We identified and functionally characterised three novel GTs KfuSV, MycS3, and NysSV that catalyse the installation of a second sugar to polyenes. Among these, KfuSV exhibited exceptional catalytic promiscuity, accepting 12 of the 20 sugars tested and tolerating various polyene scaffolds (Figure 2). We next sought to exploit this glycosylation machinery in a living host. We expressed the novel sugar biosynthesis machinery in the amphotericin producer that otherwise lacks this second sugar biosynthetic gene cluster. The engineered strain allowed us to produce a diglycosylated amphotericin derivative at scale through fermentation.
The formation of amide groups from the carboxylic acid present in many polyenes has also been reported to improve safety profiles. In light of this, we also utilised the amidotransferase PcsA to enzymatically replace the detrimental carboxyl group with groups such as amides, hydroxamates and acyl hydrazines. This step had previously been undertaken using chemical coupling reagents and organic solvents.
By harnessing these biological catalysts, we created a library of dual-modified polyene analogues. When tested against the WHO’s top ten priority fungal pathogens, our polyene analogues revealed striking improvements in activity and human cell toxicity, with several derivatives outperforming amphotericin, the current gold standard therapy. One of our derivatives, Nys34, emerged as a lead candidate as it combines enhanced antifungal potency with a favourable safety profile. In a clinically relevant Aspergillus lung infection mouse model, Nys34 effectively cleared infection while showing no observable signs of toxicity. Surprisingly, we found that Nys34 exerts its effects via a different mechanism to amphotericin. By killing fungi through an alternative mechanism, Nys34 offers a compelling opportunity to develop new therapies against pathogens that have evolved resistance to amphotericin.
Our platform can produce improved polyenes by clean, scalable, and high-yielding enzymatic methods, which can reduce cost and make treatments more widely available.
This research has now been published in Nature
https://doi.org/10.1038/s41586-026-10834-8
Saadia N. Mirza, Alberto Carella, Joseph W. Thompson, Deepanjan Panda, Anna R. I. McDonald, Matthew D. Crossley, Wei Li Thong, Katherine J. Robins, Sarah. A. Shepherd, Matthew J. Cliff, Clara Valero, Andrew Thom, Michael Bromley and Jason Micklefield
https://twitter.com/Micklefield_Lab
@micklefield-lab.bsky.social on Bluesky