The standard treatment for a deadly amphibian infection is often lethal to the amphibians it is meant to save. A former Merck chemist argues the fix is to design the drug for the non human patient from the start.
The antifungal that is supposed to save frogs from a devastating skin infection is, in practice, often the thing that kills them. Itraconazole, the standard of care for chytridiomycosis, the fungal disease that has wiped out amphibian populations worldwide, is itself frequently lethal to the amphibians it is meant to treat. That is not an edge case. It is the default.
Tim Cernak, an associate professor of chemistry at the University of Michigan, thinks the reason is structural. For most of pharmaceutical history, drugs have been designed for human patients and then applied off-label to animals whose biology is just similar enough to make that work, until it doesn't. Cernak spent nearly two decades at Merck developing precision therapies for cancer, HIV, and diabetes that targeted disease while sparing healthy cells, work that a MIT Technology Review profile describes as the foundation for his next project. The project is to take that precision seriously for the patients who never had a chemist designing for them in the first place: the frog, the Gila monster, the bald eagle, the hemlock tree, the loggerhead turtle. His framing, in the profile, is that "the patient was always meant to be a frog in the first place, from the beginning to the end."
Cernak is calling this proposed discipline "conservation chemistry." It is a name, not yet a field, and the MIT Technology Review feature frames it as a corrective to a record that already includes some sharp pharmaceutical failures in non-human populations. DDT, diclofenac in vultures, and the routine practice of dosing animals with formulations designed for humans have all functioned, on non-human biology, like the indiscriminate cancer drugs Cernak spent his industry career trying to replace. Conservation chemistry, as he sketches it, would treat the non-human patient as the design target from the start and accept the cost of doing the chemistry again for them.
The Michigan lab is set up to do that work. AlphaFold, an AI protein-structure tool, helps the team visualize what a drug target looks like in a non-human species, including cases where the human target and the non-human target diverge enough to make a human-designed molecule the wrong shape. Robots in the lab run roughly 1,500 reactions a day, the throughput needed to test compounds against the actual biology of a frog or a Gila monster, rather than refining a molecule built for a human patient. The caseload, as MIT Technology Review reports, includes a Gila monster with a parasite and bald eagles with avian flu. Each of those is a different patient, in the sense Cernak is using the word: a body with its own biology, and a drug problem that has to be solved against that biology, not borrowed from one solved against ours.
The case for doing this is partly about the species in the room and partly about the failures of the alternatives. Itraconazole, dosed to a frog with chytrid, can complete the work the fungus started. Human anti-cancer regimens, scaled to a loggerhead with a contagious tumor, hit the tumor and the turtle. A painkiller formulated for cattle, given to a vulture that scavenges the carcass, can collapse the bird's kidneys. The pattern is the same: a drug designed for one body, used in another, with effects on healthy cells that the original chemistry never had to defend against. Cernak's argument is that conservation chemistry would build the defense in, by designing the molecule for the actual patient.
What is not yet settled is whether the discipline builds. "Conservation chemistry" is Cernak's coinage, and the article presents it as a proposal. The infrastructure that would let a second lab do this work, the training pathway, the regulatory and veterinary pathways, and the funding lines that treat wildlife drug design as a category of its own, are not described in the profile. The feature is part of MIT Technology Review's "Job titles of the future" series, which frames wildlife and ecosystem drug design as a possible emerging role, not a present profession.
For now, the constraint is concrete. The standard of care for a frog with a deadly skin infection is a drug that can kill the frog. A chemist who spent twenty years making cancer drugs more selective thinks he can do better for the patient that was never designed for in the first place. The open question is whether the rest of the discipline shows up to help.