By any measure, the diabetes drug Ozempic has been a blockbuster, racking up billions of dollars in annual sales. In the United States alone, pharmacies fill millions of prescriptions for Ozempic and related drugs, which have become popular for their weight-loss effects, every month.

But in the beginning, before the celebrity endorsements and the think pieces and the global supply crunch, there was just a strange, venomous lizard with a flair for intermittent fasting. The Gila monster, which is native to the deserts of North America, can survive on just a few meals a year, thanks to a digestion-slowing hormone in its venom.

The discovery of this hormone paved the way for Ozempic, making the Gila monster an enormously profitable gift to modern medicine. And last summer, one particular Gila monster, a former pet named Pebbles, needed medicine in return.

Pebbles, a resident at the Creature Conservancy, a wildlife education organization in Ann Arbor, Mich., had been infected with a parasite called Cryptosporidium. Hard to kill, the parasite colonizes the digestive tract and is typically a death sentence for reptiles.

A veterinarian had recommended that Pebbles be humanely euthanized. But the Creature Conservancy wasn’t ready to accept that fate for Pebbles, who had at least another decade of life potentially ahead of her.

“If he can fix her,” he said Steve Marsh, founder of the Creature Conservancy, in July. He nodded toward a sharply dressed man who was cradling Pebbles in his gloved hands: Tim Cernak, a pharmaceutical chemist at the University of Michigan.

A few minutes later, a veterinarian inserted a tube into the lizard’s throat, collecting liquid from her stomach. Later, Cernak would study this sample in his lab, hoping to isolate the parasite and find a drug that could vanquish it.

Pebbles was not the patient Cernak had in mind when he began his career. Until 2018, he had worked at pharmaceutical giant Merck, developing drugs for people with cancer, HIV, diabetes and other conditions. Along the way, he had helped develop cutting-edge approaches, involving robots and artificial intelligence, to speed up the invention of new human drugs.

A few years ago, however, Cernak decided that he wanted to use those tools to make medicines for ailing plants and animals, forging a new field he called “conservation chemistry.”

Conservation chemistry

Cernak, a native Canadian, had grown up catching leopard frogs and crayfish at his grandparents’ lakeside cottage in Quebec. When he left his job at Merck, he wanted to use his chemical expertise for a greater environmental good.

“Pharma’s rad,” said Cernak, who has a youthful face and the upbeat energy to match. “But I wanted to apply my talents to a love of nature.”

After COVID-19 hit, he began to think more about the existential, population-level risks posed by disease. A fungal disease called chytridiomycosis, or chytrid, was driving frogs to extinction. Bald eagles and elephant seals were succumbing to bird flu. And sea turtles were washing ashore with a contagious form of cancer.

Treatment options left much to be desired. There were human medicines that could help some sick frogs and sea turtles, but they could cause severe side effects and were suitable for only certain animal patients.

Robo drug discovery

Drug development is a famously failure-prone pursuit. Cernak has ratcheted up the degree of difficulty by focusing on pathogens and patients that are poorly understood. “People haven’t chosen to look inside of the Gila monster too much,” he said.

But that’s exactly what his team was doing one day in July, in a chemistry lab on the University of Michigan’s campus.

There are many species of Cryptosporidium, which can infect a wide range of mammals, birds and reptiles. All of them are understudied, Cernak said, but the reptile pathogens are a particular mystery.

And when the scientists put a sample from Pebbles under the microscope, they were startled by what they saw: a single-celled parasite wrapped in a thick, jellylike coating, an unexpected extra layer of cellular protection. “It’s like this citadel,” Cernak said.

Soon, they would begin looking for a drug to breach this line of defense. On another lab bench, however, the hunt for a better cure for chytrid, the amphibian fungal disease, was already underway. A modular robot in a transparent box glided back and forth, preparing to dispense minute doses of antifungal drugs into a grid of shallow wells.

The short-term goal was to identify an existing drug that would be more effective and less toxic than one common chytrid treatment.

The team identified a compound, which Cernak declined to disclose, that proved promising in both the cells and in the African dwarf frogs that were paddling around in a nearby biology lab. The next step is to test it in additional species.

Into the wild

Over the longer term, Cernak is trying to engineer a new drug specifically for chytrid. He is using several AI-powered tools — including Google’s Nobel-winning software AlphaFold — to visualize the 3D structure of a critical fungal protein and design a drug that binds to it . (He is using the same basic approach in his search for new drugs for bird flu and sea turtle cancer.)

A drug that is sufficiently precise, binding only to that chytrid protein, could potentially be dispensed directly into frog ponds.

It’s a vision he shares with some reluctance. “I just really don’t want to appear to be the mad chemist who’s sprinkling chemicals all over,” he said. “Particularly just because I think chemists are so easy to villainize in conservation. And so we’re really trying to find the right way to get it to our patient with having the absolute minimal impact on the ecosystem.”

There are likely to be regulatory and economic hurdles, too: Who, exactly, is going to pay for a precision cancer drug for sea turtles? And although novel drugs can help buy time for critically endangered species, protecting biodiversity will require addressing larger problems, including climate change and habitat loss.

Pills for Pebbles

Cernak is cleareyed about these challenges.

He has been waiting for more than a year for the federal permits he needs to receive sea turtle tissue. And he has not been able to attract funding for the Pebbles project, even after approaching a drug company profiting from Gila-monster-inspired drugs. “We have kept the work alive on fumes,” he said.

Without funding, he could not culture Pebbles’ parasite in the lab, which he needed to do before he could unleash the robotic drug testers. So, his team used AI-based tools to scour the scientific literature on Cryptosporidium, looking for potential treatments safe enough to try without extensive laboratory trials. “My biggest fear was that I would overdose and poison Pebbles,” Cernak said.

He homed in on a drug that can cure Cryptosporidium infections in livestock but has yielded mixed results in reptiles. In December and January, Pebbles gobbled down seven doses of the drug, which had been secreted inside dead mice and quail.

So far, she has shown no ill effects. But because her digestive system moves so slowly, it could be months before the scientists know whether the drug worked. Cernak is bracing himself for the possibility that it won’t.

And then what? “We’ll try again,” he said.