Since a groundbreaking study in 2009, researchers have discovered that common gut bacteria Bacteroides fragilisby secreting toxins that damage the lining of the colon, promote the formation of colon tumors and can lead to colorectal cancer. But until now, the exact mechanism the toxin uses to attach to these cells has remained a mystery.
A multicenter team led by researchers from the Bloomberg Kimmel Cancer Immunotherapy Institute at the Johns Hopkins Kimmel Cancer Center and the Johns Hopkins University School of Medicine has identified the missing link. The study was published on April 22nd. naturethe following becomes clear: B. fragilis The toxin BFT must first bind to the host receptor claudin-4 before causing damage. This research was supported in part by the National Institutes of Health.
This is a very interesting moment because we have made several attempts over time to identify the receptor. Understanding how bacterial toxins work could open the door to new approaches for the detection and treatment of related diseases such as diarrhea, colorectal cancer, and bloodstream infections. ”
Cynthia Sears, MD, Bloomberg Kimmel Lead Author, Professor of Cancer Immunotherapy, Johns Hopkins University School of Medicine
Indeed, this discovery has already led to the development of molecular decoys that successfully block the effects of toxins in animal models, providing a potential strategy for toxin prevention. B. fragilis Colon damage.
B. fragilis It is detected in up to 20% of healthy people and has a strong ability to induce colonic inflammation and tumor formation. Previous research in Sears’ lab showed that BFT causes chronic inflammation in the intestine by disrupting E-cadherin, a protein essential for maintaining the colon’s protective barrier. their previous natural medicine A study by Sears et al. showed that the effects of BFT lead to colon tumor formation. However, BFT did not appear to bind directly to E-cadherin. Some other elusive mechanism seems to be at work.
Identifying the mechanism began with a genome-wide CRISPR screen led by Maxwell White, MD/PhD. candidate in the Sears lab and is collaborating with Matthew Waldo’s lab at Harvard Medical School. White and colleagues in Waldor’s lab identified claudin-4 as the link by systematically knocking out the gene in colonic epithelial cells. When Dr. White knocked out claudin-4, the BFT toxin could no longer bind to cells, leaving the E-cadherin target intact.
“It took some time to get the assay working and validate the approach, but once we were able to screen, claudin-4 was clearly a top hit,” White says. “It was an exciting moment.”
Sears added that the receptor’s identity was surprising because he and other researchers in the field had long suspected it was a signaling protein, such as a G-coupled protein receptor, but not claudin-4. According to a literature review, most proteases go directly to their target rather than first binding to another receptor, so the research team could not identify other toxins that function in this way.
To confirm that the toxin and receptor are physically bound, the Johns Hopkins team collaborated with structural biologists F. Xavier Gomis Roos and Ulrich Eckhardt of the Barcelona Institute of Molecular Biology. White and Barcelona’s group used biophysical analysis to demonstrate that BFT and claudin-4 form a tight one-to-one complex in vitro, providing the first physical evidence of a binding interaction.
The research then moved from test tubes to living systems through collaboration with Min Dong’s lab at Harvard Medical School. Kang Wang and colleagues in Dong’s lab used a mouse model to assess how toxins behave in the complex environment of the intestine.
The researchers tried to block the toxin from binding to colon cells by creating a decoy claudin-4, a soluble protein that displays the claudin-4 sequence. In fact, BFT bound to the decoy rather than the receptor. This decoy strategy successfully protected mice from BFT-induced damage.
“This approach can be repeated using small molecules and other biologics with better pharmacological properties,” White says. The research team is currently investigating which molecular approaches are likely to be most successful in blocking the toxin.
Researchers note that one piece of the puzzle remains. Although the team identified the receptor and demonstrated binding, the exact experimental structure of the interaction between BFT and claudin-4 has not yet been captured. Current AI modeling tools, such as AlphaFold, were unable to fully resolve interactions.
Other authors of the paper include Jason Chen, Xiaoguan Wu, Abby L. Geis, and Jessica Queen of Johns Hopkins University, and Hailong Zhang, Karthik Hulahari, and Ji Zhang of Harvard Medical School.
This research was supported by the Bloomberg Kimmel Institute for Cancer Immunotherapy, Janssen Research and Development, Cancer Research UK, the National Institutes of Health (grant numbers R01 AI042347, R01 NS080833, R01 NS117626, R01 AI170835 and R01 AI189789) and the Howard Hughes Medical Institute.
Sears receives royalties for writing and reviewing UpToDate. This arrangement will be managed by Johns Hopkins University in accordance with its conflict of interest policy.
sauce:
Reference magazines:
White, Montana; Others. (2026). Carcinogenic bacterial toxins bind to claudin-4 and cleave E-cadherin. nature. DOI: 10.1038/s41586-026-10375-0. https://www.nature.com/articles/s41586-026-10375-0
