Vanderbilt Institute of Chemical Biology



Discovery at the VICB







Defining a Route to Intoxication


By: Carol A. Rouzer, VICB Communications
Published:  June 9, 2015



The discovery of a cellular receptor for the C. difficile toxin TcdB opens the door to new approaches to therapy for pseudomembranous colitis.


A major adverse effect of the use of broad-spectrum antibiotics is the elimination of normal bacteria in the gastrointestinal tract, leading to the overgrowth of pathogenic species. Clostridium difficile is the bacterium most likely to thrive under these conditions. It produces diarrhea in infected patients, and in severe cases, a condition known as pseudomembranous colitis. As morbidity and mortality associated with C. difficile infection continue to rise, new therapeutic approaches are desperately needed. To address this need, Vanderbilt Institute of Chemical Biology member Borden Lacy and her laboratory are investigating the mechanism by which TcdB, one of two major C. difficile toxins, associates with and enters its target cell. Now they report the discovery of a protein that serves as a receptor for the toxin [M. E. LaFrance, et al. (2015) Proc. Natl. Acad. Sci. U.S.A., 112, 7073].


The C. difficile toxins, TcdA and TcdB, are proteins that induce the fluid secretion, influx of immune cells, and tissue damage seen in the colons of infected patients. Both proteins comprise four domains: a glucosyltransferase domain (GTD), a delivery/pore-forming domain, an autoprotease domain, and a combined repetitive oligopeptide (CROPS) domain (Figure 1A). The toxin is taken up by a target cell through the process of receptor-mediated endocytosis, which delivers it to the acidic environment within an endosome. The acid induces a conformational change in the protein that allows the delivery/pore forming domain to create a pore in the endosomal membrane. Through this pore, the GTD gains access to the cell cytoplasm. Binding of inositol hexakisphosphate from the host cell to the toxin activates the TcdB autoprotease domain, which cleaves off the GTD, allowing it to seek out its target (Figure 1B).




Figure 1.
Structure of the gene for TcdB, and the mechanism of its entry into cells. (A) Diagrams of the genes for TcdA and TcdB outlining the major domains. (B) Entry of TcdA or TcdB (shown as four colored spheres corresponding to the domains in part A), occurs through receptor-mediated endocytosis (1), which traps the toxin in an endocytic vesicle (2). Acidificaiton of the vesicle leads to a structural change, allowing the delivery domain to form a pore in the vesicle membrane (3). Activation of the autoprotease domain by inositol tetrakis-phosphate (insP6) leads to cleavage of the glucosyl-transferase domain, which can then glucosylate members of the Rho family of small GTPases. Reprinted R. N. Pruitt & D. B. Lacy (2012) Front. Cell. Infect. Microbiol., 2, 1, doi: 10.3389/fcimb.2012.00028 under the Creative Commons Attribution Non Commercial License. Copyright 2012 Pruitt and Lacy.


TcdB’s toxicity results from the ability of its GTD to transfer glucose from UDP-glucose to a specific threonine residuce of Rho, Rac1, and Cdc42, which are all members of the Rho family of small GTPases. These proteins play a key role in the function of cytoskeletal actin, and their glucosylation by the GTD of TcdB results in cytoskeletal dysfunction, and eventually, death by apoptosis. At high concentrations of the toxin, formation of reactive oxygen species within the cell leads to its death by a necrotic pathway.


Although researchers generally agree that the entry of TcdB into cells is through interaction with a receptor, the exact mechanism by which this occurs is not completely clear. A body of data suggests a role for TcdB’s CROPS domain; however, this domain is not absolutely necessary for intoxication. The condroitin sulfate proteoglycan 4 (CSPG4) protein appears to play a role in TcdB-mediated apoptosis in some cell types (HeLa and HT29), but it is not involved in necrotic events. These findings led the Lacy group to hypothesize that an as yet unknown receptor is important in TcdB intoxication, and they set out to test this hypothesis. To maximize the likelihood of finding a TcdB receptor, the investigators first used a retroviral gene-trap vector to induce random mutations in Caco-2 human colorectal adenocarcinoma cells. Incorporation of the vector into a region of cellular DNA that is actively transcribed both inactivates the cellular gene and drives the expression of a vector-encoded gene that confers neomycin resistance to the cell. Thus, cells in which a gene has been inactivated can be easily identified by their ability to survive when cultured in the presence of neomycin. The investigators selected these cells and subjected them to culture in the presence of TcdB. Their goal was to identify cells carrying gene mutations that convey resistance to the toxin. This approach yielded 61 clones of cells, each carrying an inactivated gene that allowed them to survive in the presence of TcdB. The researchers were interested to discover that two of the clones carried mutations in the gene for Polio Receptor-Like 3 (PVRL3, also called Nectin-3). PVRL3 is a membrane protein notable for its three extracellular immunoglobulin domains. It is a member of a family of proteins that also includes PVRL1, PVRL2, and PVR. The family plays a role in adherens junction formation between cells in many tissues, and a number of the family members have been shown to play a role in the entry of some viruses into cells.


To better understand the role of PVRL3 in TcdB-mediated intoxication, the Lacy lab carried out further experiments on the E19 clone, one of the two carrying a mutation in the PVRL3 gene. They found that transfection of E19 cells with an expression vector carrying PVRL3 resulted in a return to TcdB sensitivity. Similarly, transduction of wild-type Caco-2 cells with shRNAs targeting PVRL3 resulted in TcdB resistance. Similar knockdown of PVRL3 expression by shRNA or CRISPR/Cas9-mediated approaches in HeLa cells led to resistance to short-term TcdB exposure, but these cells were not resistant to long-term toxin exposure as were the shRNA-treated Caco-2 cells. The investigators attributed this difference to the fact that HeLa cells express CSPG4, whereas Caco-2 cells do not. CSPG4 could be mediating intoxication in the HeLa cells that lack PVRL3 in the case of long-term TcdB exposure.


The cell culture results supported the hypothesis that PVRL3 is involved in TcdB-mediated cytotoxicity. To see if TcdB directly interacts with PVRL3, the Lacy lab investigators expressed the extracellular domain of PVRL3 with a His-tag and incubated it together with full-length TcdB, the TcdB CROPS domain, or TcdB lacking the CROPS domain (amino acids 1-1834). The TcdB proteins had been labeled with biotin, allowing the investigators to capture those proteins, and any others associated with them, using streptavidin beads. Analysis of the captured proteins revealed that the PVRL3 extracellular domain binds directly to full-length TcdB, and TcdB 1-1834, but not to the TcdB CROPS domain. These results support the hypothesis that PVRL3 may act as a cellular receptor for TcdB, but if so, its binding site does not involve the CROPS domain.


To further test the hypothesis that PVRL3 is a receptor for TcdB, the researchers preincubated Caco-2 cells with an antibody against the extracellular domain of PVRL3, and then exposed the cells to TcdB. They found that the antibody pretreatment protected the cells from TcdB intoxication. Similarly preincubation of TcdB with soluble PVRL3 ectodomain prior to adding it to Caco-2 cells markedly reduced its toxicity. This did not occur upon preincubation of TcdB with the ectodomains of PVRL1 or PVRL2, suggesting that the interaction with TcdB is specific for PVRL3.


Although Caco-2 cells provide a reasonably good model for colonic epithelium, the Lacy lab investigators wanted to see if they could find evidence that PVRL3 mediates the interaction of TcdB with intact human colonic mucosa. So they obtained explants of healthy colonic tissue and used immunohistochemistry to demonstrate the presence of PVRL3, but not CSPG4 on the surface of the epithelial cells. When they treated the explants with TcdB, they observed an overlap in the localization of the toxin with that of PVRL3. Furthermore, they observed a colocalization of TcdB and PVRL3 in colonic tissue obtained from a patient suffering from pseudomembranous colitis (Figure 2).



Figure 2.
Immunohistochemistry to detect TcdB (Toxin) and PVRL3 in TcdB treated healthy colonic mucosa (A) or in the mucosa from a patient suffering from pseudomembranous colits (B). TcdB is stained in green and PVRL3 is stained in red. Yellow indicates co-localization of the two proteins. Reproduced with permission from M. E. LaFrance, et al. (2015) Proc. Natl. Acad. Sci. U.S.A., 112, 7073. Copyright 2015, M. E. LaFrance, et al.




The data strongly support the conclusion that PVRL3 acts as a receptor for TcdB in colonic epithelial cells. The investigators note that when PVR binds to PVRL3, the protein complex is internalized by clathrin-mediated endocytosis, and they hypothesize that a similar fate might apply to complexes of TcdB with PVRL3. Although interaction with PVRL3 with TcdB does not involve the TcdB CROPS domain, they point out that the sum of all the current data suggests that TcdB uses a dual-receptor mechanism to interact with a target cell. According to this mechanism, TcdB first interacts with cell surface oligosaccharides through its CROPS domain, an interaction that facilitates its subsequent specific binding to PVRL3. If the Lacy lab is correct that an interaction between TcdB and PVRL3 is critical to cell intoxication, then therapeutics designed to disrupt the TcdB-PVRL3 interaction could provide a novel approach to the prevention and/or treatment of C. difficile-mediated diarrhea and pseudomembranous colitis.






The Vanderbilt Institute of Chemical Biology, 896 Preston Building, Nashville, TN 37232-6304, phone 866.303 VICB (8422), fax 615 936 3884
Vanderbilt University is committed to principles of equal opportunity and affirmative action. Copyright © 2014 by Vanderbilt University Medical Center