DNA repair puzzle finally solved

This is part of a collection: The Crick Magazine | Issue 3

After 20 years, a combination of three advanced technologies has revealed elusive cancer-linked DNA repair proteins. 

Structure of the protein RAD51, shown rotating against a white background.

Three-dimensional cryo-electron microscopy structure of RAD51 and the RAD51 paralogs. Credit: Luke Greenhough

A cell’s ability to repair its damaged DNA is one of the most fundamental processes in biology, helping protect us from diseases like cancer. Over the years, researchers have unravelled many of cells’ DNA repair mechanisms, ultimately leading to precision cancer treatments. 

But for more than two decades, a particular set of repair proteins, known as the RAD51 paralogs, has sat at the edge of our understanding. Mutations in these genes are linked to breast and ovarian cancer, as well as Fanconi anaemia – a rare and life-limiting, cancer-inducing disorder.

“The RAD51 paralogs operate in the same biological pathway as well-known DNA repair genes BRCA1 and BRCA2,” says Stephen West, head of the Crick’s DNA Recombination and Repair Laboratory. “As a result, some cancers caused by mutations in these genes can be treated with drugs called PARP inhibitors, because they all rely on a shared defence system.”

Yet despite their importance, how the RAD51 paralogs functioned has remained a mystery. “It was a real gap in the field,” says Stephen. “It was impossible to investigate their biochemistry.”

The problem was not a lack of curiosity. It was a lack of tools. “The proteins were so difficult to work with that we’d reached a point where hardly anyone was trying to understand what they did,” he says. By the early 2000s Stephen and his team discovered that the RAD51 paralogs interact to form protein complexes, but further progress stalled. These complexes had turned out to be notoriously difficult to isolate in the lab, often clumping together or degrading before they could be studied. Without a stable, soluble sample or the technology to see it, the proteins remained effectively intractable to the scientists.

What changed was not a single discovery, but a quiet revolution in scientific technology. For Luke Greenhough, a researcher in Stephen’s lab, the breakthrough came by combining three particular technologies: AlphaFold3 software, cryoelectron microscopy (cryo-EM) and single-molecule imaging.

AlphaFold3 is an advanced AI-based system that predicts proteins’ 3D structures from their amino acid sequences. Cryo-EM, on the other hand, is a powerful way to look at the 3D structures of proteins directly.

The team ran an initial AlphaFold3 prediction, and this helped shape the project’s direction, pointing them to explore a structure they might not otherwise have pursued. It was the clue the team needed to be able to start investigating how the paralogs worked together using cryo-EM, which had also undergone a series of advances.

“Cryo-EM technology underwent a ‘resolution revolution’ in the 2010s, and has now advanced to the point where we can study clusters of proteins in astonishing, high molecular detail,” explains Luke. “When we combine that with single-molecule imaging – which allows us to watch individual proteins interact with DNA in real time – we get to do something that had long been out of reach: see the machinery in action.”

The results, published across two papers in 2023 and 2026, reveal that the RAD51 paralogs form two distinct but closely related complexes, each made up of four subunits – correcting earlier assumptions that had underestimated their complexity. More importantly, the researchers uncovered how these complexes behave: assembling and capping DNA repair filaments, playing a crucial role in fixing damaged genetic material.

The implications extend far beyond structural biology. With their close connection to the BRCA genes, studying mutations in the RAD51 paralogs helps researchers understand how they can weaken DNA repair and lead to cancer. “By mapping the precise structural effects of these mutations, our work helps explain why they lead to disease,” Luke says.

“This has been a 20-year puzzle,” adds Stephen, “and really, it’s only down to technological developments in the last few years – and Luke’s determination to take on this challenge – that we’ve uncovered this missing piece of information.”

Sign up for our newsletters

Join our mailing lists to receive updates about our latest research and to hear about our free public events and exhibitions.  If you would like to find out more about how we manage your personal information please see our privacy policy.