Scientists at the Johns Hopkins Kimmel Cancer Center and collaborating institutions have identified the protein that lets a common gut bacterium's toxin latch onto colon cells, answering a question that had gone unresolved for roughly twenty years. The receptor is claudin-4, a protein that helps seal the spaces between neighboring cells in the intestinal lining. The work was published in Nature, according to Science Daily, and received partial funding from the National Institutes of Health.
The bacterium at the center of the research, Bacteroides fragilis, lives in the intestines of as many as one in five healthy people. Its presence alone does not mean cancer will develop. Certain strains, however, release a potent toxin called B. fragilis toxin, or BFT, that can break down the colon's protective barrier, drive inflammation and encourage tumors to form.
Why it matters
The finding fills a gap that had blocked a fuller understanding of how a toxin-producing gut microbe contributes to colorectal cancer. Science Daily reports that the team has already turned the discovery into a prototype countermeasure: a molecular decoy that draws the toxin away before it reaches colon cells. In mice, the approach prevented the toxin's damaging effects, pointing toward a possible preventive strategy rather than a treatment for existing disease.
The work also opens a wider door. Cynthia Sears, the study's senior author and a Bloomberg~Kimmel Professor of Cancer Immunotherapy at Johns Hopkins, said in remarks reported by Science Daily that pinning down how bacterial toxins operate could lead to new ways of detecting and treating related conditions, including diarrhea, colorectal cancer and bloodstream infections. She described the moment as exciting after repeated earlier attempts to find the receptor.
A CRISPR screen turns up a clear winner
The toxin's destructive behavior was already well documented. A 2009 study in Nature Medicine, led by Sears, showed that BFT could promote colon tumors, helping establish the link between toxin-producing gut bacteria, persistent inflammation and colorectal cancer. Later experiments showed that BFT damages E-cadherin, a protein that binds adjacent cells together and keeps the colon's barrier intact. Once E-cadherin is disrupted, the barrier weakens and harmful material may reach surrounding tissue.
What remained unexplained was the first step. BFT could break E-cadherin, but it did not seem to attach to that protein directly. Researchers suspected an intermediary molecule was involved, yet searches for it had come up empty.
The team approached the problem with CRISPR gene editing. Maxwell White, an M.D./Ph.D. candidate in Sears' laboratory, ran a genome-wide screen in colon epithelial cells — the cells lining the inner surface of the colon — working with Matthew Waldor's lab at Harvard Medical School. The researchers switched off genes one by one to see which ones the toxin needed in order to bind and cause damage.
One gene stood out sharply. It codes for claudin-4, which helps neighboring cells form tight junctions, the seals that restrict what passes between cells and support the intestinal barrier. When the gene was disabled, BFT could no longer attach to colon cells, and without that foothold it could not go on to attack E-cadherin.
White said in comments reported by Science Daily that getting the assay to work and validating the method took time, but that once the screen ran, claudin-4 emerged as an unmistakable top result.
An unexpected mechanism, confirmed in the lab
The identity of the receptor surprised the researchers. Sears and others had expected BFT to pair with a signaling protein such as a G protein-coupled receptor, which works something like a communications switch letting cells respond to their surroundings. Claudin-4 plays a different role: its main job is structural, maintaining the physical links between cells.
The result also describes an unusual route for a bacterial toxin. BFT is a protease, an enzyme that cuts other proteins. Most proteases act on their targets directly, but BFT appears to need a separate receptor before it can reach E-cadherin. Science Daily reports that a review of prior literature turned up no other toxin known to work this way.
To confirm that BFT and claudin-4 genuinely bind, the Johns Hopkins group worked with structural biologists F. Xavier Gomis-Rüth and Ulrich Eckhard at the Molecular Biology Institute of Barcelona. Using methods that measure molecular interactions, White and the Barcelona researchers showed the two form a tightly bound complex in laboratory experiments, with each toxin molecule paired to one receptor molecule — direct physical evidence of the connection.
Identifying the receptor also suggested a way to block it. If BFT needs claudin-4 to gain a hold on colon cells, a stand-in molecule might lure the toxin away from its real target. The team built such a decoy and tested it in mice, where it protected the animals from the toxin's harmful effects.
The next step will be determining whether the decoy strategy holds up beyond mouse experiments and whether it can be developed into something usable for people at elevated risk of bacterial damage in the colon.
