Scientists at the University of Cambridge have developed a new way to modify complex drug molecules using light rather than toxic chemicals. This discovery has the potential to accelerate and improve the way drugs are designed and manufactured.
Published in natural synthesisthis study introduces what the research team calls an “anti-Friedel-Crafts” reaction. Classic Friedel-Crafts reactions use strong chemicals or metal catalysts under harsh experimental conditions. This means that reactions occur only in the early stages of drug manufacturing, followed by many additional chemical steps to produce the final drug.
Cambridge’s new approach reverses that pattern, allowing scientists to modify drug molecules at the final stage of manufacturing.
Rather than relying on heavy metal catalysts, the chemical reaction is carried out by LED lamps at ambient temperature. Once activated, it triggers a self-sustaining chaining process that forms new carbon-carbon bonds under mild conditions without the use of toxic or expensive chemicals.
In practical terms, this means that chemists can make targeted changes late in the development of new or existing drugs, rather than dismantling complex molecules and rebuilding them from scratch. It’s a process that typically takes several months.
We have discovered a new way to make precise changes to complex drug molecules, especially those that have been extremely difficult to modify. ”
David Vahey, first author and postdoctoral fellow, St John’s College, University of Cambridge
Vahey added, “Scientists sometimes spend months rebuilding large portions of molecules just to test one small change. Now, instead of doing a multi-step process on hundreds of molecules, scientists can start with a hit and make small modifications later.”
“This reaction allows scientists to make precise adjustments much later in the process under mild conditions and without relying on toxic or expensive reagents. This opens up areas of chemistry that were previously difficult to access and gives medicinal chemists a cleaner and more efficient tool to explore new drug versions.”
Fewer steps mean fewer chemicals, less energy consumption, a smaller environmental footprint, and significant time savings for chemists. This highly selective reaction allows scientists to make precise adjustments much later in the process. This is extremely important in drug development, where even small structural adjustments can have a significant impact on a drug’s effectiveness, how it behaves in the body, or the number of side effects it causes.
The Cambridge breakthrough addresses one of the most fundamental steps in that process: the formation of the carbon-carbon bonds that underpin everything from fuels to complex biomolecules.
This method is highly selective, meaning that parts of the molecule can be changed without disturbing other sensitive regions. Chemists call this “high functional group tolerance.” That makes it particularly suitable for late-stage optimization, an important part of modern medicinal chemistry, where scientists fine-tune molecules to improve a drug’s performance.
The reaction also has the potential to dramatically reduce toxic chemical waste and energy use in drug development by avoiding heavy metal catalysts and hazardous conditions and reducing the need for long synthetic sequences, which is an increasing priority as the industry seeks to reduce its environmental footprint.
Vahey is a member of Professor Erwin Reisner’s research group at the University of Cambridge. Reisner’s group is known for developing photosynthesis-inspired systems that use sunlight to convert certain types of waste, water, and the greenhouse gas carbon dioxide into useful chemicals and fuels.
Leissner, lead author of the paper and Yusuf Hameed Professor of Energy and Sustainability in the Department of Chemistry, said the importance of the latest research lies in expanding what chemists can do under real-world conditions while developing greener manufacturing methods.
“This is a new way to form fundamental carbon-carbon bonds, which is why the potential impact is so large. It also means chemists can avoid unwanted and inefficient drug modification processes.”
The research team demonstrated reactions across a wide range of drug-like molecules and showed that they can be adapted to the continuous flow systems increasingly used in industry. The collaboration with AstraZeneca helped test whether this method could meet the practical and environmental demands of large-scale drug development.
“Transitioning the chemical industry to a sustainable industry is probably one of the most difficult parts of the entire energy transition,” Reissner explained.
And that breakthrough, like some of science’s most famous discoveries, from X-rays and penicillin to Viagra and modern weight-loss drugs, was born out of laboratory setbacks.
“Failure after failure. Then out of the chaos we discovered something unexpected: a real diamond in the rough. And it was all because of a failed control experiment,” Verhey said.
As he was testing photocatalysts, he removed the photocatalyst as part of a control test and found that the reactions worked just as well, and in some cases better, without the photocatalyst.
At first, the unusual product seemed like a mistake. Rather than scrap it, the team decided to make sense of it. Reisner said the breakthrough depended not only on chemistry but also on judgment.
“Recognizing the value of unexpected events is probably one of the key characteristics of successful scientists,” he said.
“We generate huge amounts of data and are increasingly using artificial intelligence to help analyze it. We have algorithms that can predict reactivity. AI is helpful because chemists don’t have to do endless trial and error, but algorithms just follow the rules they’re given. We still need humans to look at something that looks wrong and ask whether it might actually be something new.”
In this case, it was Vahey who recognized its importance and investigated it further.
“David could have written it off as a failure of control,” Reisner said. “Instead, he stopped and thought about what he was seeing. Discovery happens in those moments when you choose to investigate rather than ignore.”
After mapping the underlying chemical structure, the team deployed a machine learning model developed in collaboration with Trinity College Dublin to predict where reactions would occur in entirely new molecules that had never been tested in the lab.
By learning patterns from established chemistry, AI can effectively simulate reactions before executing them, allowing researchers to identify the most promising candidates faster and with far less trial and error. The result is a tool that not only works in the lab, but can actively help scientists develop new drugs faster in the future.
For Vahey, this will provide researchers with an important new tool in their drug discovery and development toolbox.
“The future implications are what industry and other researchers do with it next,” he said. “For us, the lab is mostly about average or bad days. The good days are very good days.”
Reisner added, “Chemists only need one or two good days a year. Those days can come from failed experiments.”
sauce:
St John’s College, University of Cambridge
Reference magazines:
Vahey, DM; Others. (2026). Anti-Friedel-Crafts alkylation by electron donor-acceptor photoinitiation. natural synthesis. DOI: 10.1038/s44160-026-00994-w. https://www.nature.com/articles/s44160-026-00994-w.
