For nearly two decades, the scientific community has grappled with a deeply unsettling biological puzzle: how a common bacterium residing quietly in the human gut manages to deliver a destructive toxin capable of damaging the colon and driving the formation of tumors. While researchers have long known that certain strains of this microorganism unleash a potent compound that erodes the intestinal lining and provokes chronic inflammation, the precise mechanism of the initial attack remained frustratingly elusive. Specifically, scientists could not explain how the toxin first attaches to intestinal cells to begin its destructive campaign.
Now, an international team of researchers has successfully uncovered that missing step, revealing a surprising cellular vulnerability that could pave the way for entirely new strategies to protect the human colon. The breakthrough, which bridges a critical gap in our understanding of microbiology and oncology, was achieved by a collaborative team led by scientists at the Johns Hopkins Kimmel Cancer Center, its Bloomberg~Kimmel Institute for Cancer Immunotherapy, and the Johns Hopkins University School of Medicine. Their findings, which shed light on decades-old questions regarding microbial pathogenesis, were published in the prestigious scientific journal Nature.
At the heart of the discovery is the identification of a specific human protein known as claudin-4, which serves as the essential docking point, or receptor, for the bacterial toxin. Even more promising from a therapeutic standpoint, the research team has already leveraged this discovery to design and test a molecular decoy engineered to intercept the toxin before it can ever make contact with vulnerable colon cells. In rigorous laboratory experiments, this defensive strategy successfully protected mice from the toxin’s harmful tissue damage, offering a compelling glimpse into a potential future approach for preventing the bacterial complications associated with colorectal cancer. The research received partial financial support from the National Institutes of Health, underscoring its broad public health significance.
"We’ve made several attempts over time to identify the receptor, so this is an exciting moment," says senior author Cynthia Sears, M.D., the Bloomberg~Kimmel Professor of Cancer Immunotherapy and professor of medicine at Johns Hopkins. "Understanding how bacterial toxins work can open doors to new approaches for detection and therapy for associated diseases, including diarrhea, colorectal cancer and bloodstream infections."
How a Common Gut Bacterium Can Promote Colon Cancer
The investigation centers on Bacteroides fragilis, a remarkably common bacterium that can be found peacefully colonizing the intestines of up to 20 percent of healthy human beings. Under normal physiological conditions, the presence of this bacterium is entirely benign and does not automatically mean an individual will develop cancer or experience chronic gastrointestinal distress. However, a specific subset of these bacteria produces a powerful and destructive compound known as B. fragilis toxin, or BFT. When these toxigenic strains proliferate or interact unfavorably with the host’s immune system, the toxin they release is capable of disrupting the delicate protective lining of the colon, triggering localized inflammation, and encouraging abnormal tumor growth over time.
The dangerous nature of B. fragilis toxin has been under intense scientific scrutiny for many years. In a landmark 2009 study published in Nature Medicine, Dr. Sears and her esteemed colleagues demonstrated for the first time that BFT could directly promote the development and progression of colon tumors in experimental models. Their pioneering work helped establish a vital clinical and biological connection between toxin-producing gut bacteria, long-term chronic inflammation, and the onset of colorectal cancer, a disease that continues to claim hundreds of thousands of lives globally each year.
Earlier experiments conducted by various research groups also revealed a key piece of the puzzle: BFT specifically targets and damages a human protein called E-cadherin. This critical cellular adhesion molecule acts like molecular mortar, helping to hold neighboring epithelial cells tightly together and maintain the structural integrity of the colon’s protective barrier. When E-cadherin is broken apart by the enzymatic action of the toxin, the barrier becomes severely compromised. This breakdown allows harmful substances, antigens, and other opportunistic bacteria to leak through, deeply interfering with the surrounding tissue and perpetuating an inflammatory cycle.
Despite these established insights, one critical and glaring question remained entirely unanswered. Although researchers knew that BFT ultimately broke down E-cadherin, careful biochemical analysis revealed that the toxin did not actually appear to attach directly to that specific protein. Scientists long suspected that another intermediate molecule must be helping the toxin dock and reach its primary target, but repeated, independent attempts by multiple laboratories to identify this elusive docking molecule had consistently failed.
Scientists Identify the Missing Link Using CRISPR
To finally solve this enduring biological mystery, the research team turned to CRISPR, a revolutionary and powerful gene-editing technology that allows modern scientists to selectively and systematically disable specific genes within a genome to investigate their precise functions.
Maxwell White, an M.D./Ph.D. candidate working intensively within Dr. Sears’s laboratory, spearheaded a comprehensive, genome-wide CRISPR screening effort. This complex undertaking was performed in close collaboration with the laboratory of Dr. Matthew Waldor at Harvard Medical School. Together, the research team systematically switched off thousands of individual genes within human colon epithelial cells—the specialized cells that form the inner lining of the human colon—to determine which specific genetic instructions were absolutely required for the bacterial toxin to successfully attach and cause cellular damage.
As the screening data was processed, one gene stood out dramatically above all others. It was the gene responsible for producing claudin-4, a well-known structural protein that helps neighboring cells form what are known as tight junctions. These tight junctions act as impermeable seals between cells, limiting the uncontrolled movement of fluids and substances across the intestinal tissue and helping to maintain the overall barrier function of the gut.
When the researchers intentionally disabled the specific gene responsible for producing claudin-4, the results were definitive: BFT could no longer bind to the colon cells. Without that crucial initial attachment point on the cell surface, the toxin was rendered entirely incapable of executing its subsequent attacks on E-cadherin.
The finding definitively revealed that claudin-4 acts as the long-sought receptor for the bacterial toxin. In effect, the toxin must first latch onto this specific structural protein on the cell membrane before it can carry out any of its subsequent damaging biological activity.
"It took a while to get the assay working and validate the approach, but once we were able to do the screen, claudin-4 was a clear, resounding top hit," reflects White. "That was an exciting moment."
An Unexpected Discovery Reveals a New Bacterial Attack Strategy
The exact identity of the newly discovered receptor came as a significant surprise to the veteran research team.
Prior to the genome-wide screen, Dr. Sears and other microbial pathogenesis experts had fully expected BFT to interact with a protein directly involved in active cellular signaling pathways, such as a G protein-coupled receptor. These specialized receptors typically act as cellular communication switches, allowing cells to sense and respond dynamically to chemical signals from their surrounding microenvironment.
Claudin-4, however, serves a profoundly different physiological purpose within human tissues. Rather than functioning primarily as a signaling receptor that relays external messages into the cell interior, claudin-4 is a structural component dedicated to maintaining the tight physical connections and seals between neighboring epithelial cells.
This unexpected finding also exposed an unusual and previously unappreciated mechanism for how a bacterial toxin operates at the molecular level. BFT belongs to a broad class of enzymes known as proteases, which function by cutting and cleaving other proteins. The vast majority of known proteases act directly and immediately upon their ultimate targets. In stark contrast, BFT appears to require an entirely separate structural receptor—claudin-4—to anchor itself to the cell before it can reach, position itself against, and damage its ultimate functional target, E-cadherin.
When the research team cross-referenced their findings with existing scientific literature, they could not identify any other known bacterial toxin that utilizes this exact two-step molecular mechanism to breach human cellular defenses.
To firmly establish that BFT and claudin-4 genuinely and directly bind to one another in a physical embrace, the Johns Hopkins researchers forged a crucial international collaboration with structural biologists F. Xavier Gomis-Rüth and Ulrich Eckhard at the Molecular Biology Institute of Barcelona.
Utilizing advanced biophysical techniques capable of measuring precise molecular interactions in real time, White and the Barcelona research team demonstrated unequivocally that BFT and claudin-4 form a tightly bound biochemical complex in laboratory assays. Each individual toxin molecule paired precisely with one receptor molecule, providing direct, undeniable physical evidence of the connection that had eluded science for nearly two decades.
A Clever Molecular Decoy Blocks the Dangerous Gut Toxin
Uncovering the identity of the cellular receptor did far more than simply solve a long-standing academic mystery regarding bacterial mechanics. It also immediately provided the scientists with a brilliant conceptual framework for stopping the toxin in its tracks.
The logic was straightforward: if BFT strictly requires claudin-4 on the surface of colon cells to attach and initiate its destructive cascade, perhaps a synthetic or engineered substitute could be introduced to lure the toxin away from its intended biological target.
To test this ambitious hypothesis, the team engineered a specialized molecular decoy. This created construct consisted of a soluble protein carrying specific amino acid sequences derived from claudin-4, meticulously designed to act as bait that would attract and bind BFT without allowing the toxin to access or attack the colon’s protective cellular lining.
Moving from the test tube to living biological systems, the researchers partnered with the laboratory of Min Dong at Harvard Medical School, including researcher Kang Wang and colleagues, to investigate the therapeutic efficacy of this approach in animal models. These in vivo experiments allowed the scientists to observe how the bacterial toxin behaves within the vastly more complex physiological environment of a living mammalian intestine.
The experimental results were exceptionally encouraging. Instead of binding to the natural claudin-4 receptors located on the surface of the colon epithelial cells, the circulating BFT was successfully intercepted by the soluble decoy proteins. By absorbing and neutralizing the toxin before it could engage host tissue, the decoy strategy successfully and consistently protected the mice from BFT-induced tissue damage and barrier degradation.
This powerful finding strongly suggests that future therapeutic interventions might focus on intercepting bacterial toxins at their very first step of contact, rather than attempting the much more difficult task of repairing extensive tissue damage after the pathological process has already occurred.
"This approach could be iterated upon with small molecules or other biologics that have better pharmacological properties," notes White. Emphasizing the forward-looking nature of the work, the research team is actively exploring which specific molecular engineering approaches will prove most successful for translating these findings into clinically viable toxin-blocking treatments.
In pharmacological terms, small molecules are compact chemical compounds that can frequently be developed into oral medications, while biologics represent sophisticated treatments derived from or involving complex biological materials, such as engineered proteins. Future generations of this decoy strategy could potentially be optimized to circulate more effectively within the human body, evade immune clearance, or be administered with greater clinical convenience.
What This Discovery Could Mean for Colon Cancer Prevention
Despite the profound nature of the breakthrough, the researchers emphasize that important scientific questions remain to be answered before these laboratory discoveries can be directly translated into routine clinical care for human patients.
While scientists have definitively established that BFT binds to claudin-4 and that blocking this specific molecular interaction can protect experimental animals from toxin-induced tissue degradation, they have not yet captured the exact, high-resolution three-dimensional structure showing precisely how the two molecules fit together at an atomic level. Interestingly, even highly sophisticated artificial intelligence modeling tools, including advanced iterations of AlphaFold, have thus far been unable to fully resolve the intricate structural details of this specific protein-toxin interaction. Achieving a precise structural map of the complex could ultimately assist researchers in designing even more potent, drug-like compounds capable of blocking the binding site with extreme specificity.
Furthermore, additional preclinical experiments will be strictly required to determine whether this decoy-based approach can safely, effectively, and durably prevent toxin-related pathology in human populations. While the mouse experiments successfully demonstrated robust protection against acute tissue damage, those specific studies were not designed to establish whether the strategy could directly prevent the long-term development of colorectal cancer over extended periods.
Nevertheless, the discovery successfully resolves a major scientific question that has persisted unabated since the link between BFT and colon tumor formation was first firmly established in 2009. By illuminating the toxin’s previously mysterious point of attachment, researchers have unveiled a promising pathway to interrupt a dangerous biological process right at its inception.
The broader implications of the work extend well beyond oncology. The findings could eventually support entirely new diagnostic and therapeutic strategies for managing a wide array of human conditions intimately associated with bacterial toxins, ranging from severe intestinal diarrhea to persistent bloodstream infections.
The study was supported by contributions from numerous researchers. Additional scientists who contributed to the published work include Jason Chen, Shaoguang Wu, Abby L. Geis, and Jessica Queen of Johns Hopkins, alongside Hailong Zhang, Karthik Hullahalli, and Jie Zhang of Harvard Medical School.
Financial backing for the research was provided by the Bloomberg~Kimmel Institute for Cancer Immunotherapy, Janssen Research and Development, Cancer Research UK, the National Institutes of Health under specific grant numbers R01 AI042347, R01 NS080833, R01 NS117626, R01 AI170835, and R01 AI189789, and the Howard Hughes Medical Institute.