Identification of Clostridium difficile toxin B cardiotoxicity using a zebrafish embryo model of intoxication Elaine E. Hamm, Daniel E. Voth, and Jimmy D. Ballard* Department of Microbiology and Immunology, University of Oklahoma Health Sciences Center, Oklahoma City, OK 73104 Edited by Stanley Falkow, Stanford University, Stanford, CA, and approved July 31, 2006 (received for review June 6, 2006) Clostridium difficile toxin B (TcdB) has been studied extensively by life-threatening cases of CDAD, including cardiopulmonary using cell-free systems and tissue culture, but, like many bacterial arrest (18), acute respiratory distress syndrome (19), multiple toxins, the in vivo targets of TcdB are unknown and have been organ failure (20), renal failure (21), and liver damage (22). difficult to elucidate with traditional animal models. In the current Hence, identification of the cells targeted in vivo by TcdB is study, the transparent Danio rerio (zebrafish) embryo was used as needed to gain relevant insight into the disease-related activities a model for imaging of in vivo TcdB localization and organ-specific of this toxin and advance our understanding of CDAD. damage in real time. At 24 h after treatment, TcdB was found to Identification of systemic targets of bacterial toxins such as localize at the pericardial region, and zebrafish exhibited the first TcdB has been limited, because it is difficult to directly visualize signs of cardiovascular damage, including a 90% reduction in the impact of these proteins on major organs in real time. To systemic blood flow and a 20% reduction in heart rate. Within 72 h overcome this problem, zebrafish embryos were used herein to of exposure to TcdB, the ventricle chamber of the heart became characterize the systemic impact of TcdB in real time. Unlike deformed and was unable to contract or pump blood, and the fish other vertebrates, zebrafish embryos are transparent, and major exhibited extensive pericardial edema. In line with the observed organs can be visualized by standard light microscopy (23). Thus, defects in ventricle contraction, TcdB was found to directly disrupt zebrafish embryos provide a unique system for directly visual- coordinated contractility and rhythmicity in primary cardiomyo- izing the temporal and spatial effects of TcdB intoxication. By cytes. Furthermore, using a caspase-3 inhibitor, we were able to using the zebrafish embryo as a model, in the current work, we block TcdB-related cardiovascular damage and prevent zebrafish have found that TcdB functions as potent cardiotoxin, reducing death. These findings present an insight into the in vivo targets of blood flow and ventricle contraction. Furthermore, correspond- TcdB, as well as demonstrate the strength of the zebrafish embryo ing to TcdB’s known proapoptotic activity, a caspase-3 inhibitor as a tractable model for identification of in vivo targets of bacterial was found to alleviate the cardiotoxic effects of TcdB. These toxins and evaluation of novel candidate therapeutics. findings provide important insight into the in vivo activities of TcdB and present the zebrafish embryo as a model for deter- ͉ ͉ bacterial toxin Clostridium difficile-associated disease mining the systemic targets of bacterial toxins. large clostridial toxins Results rotein toxins are produced by bacterial pathogens during dis- Localizaton of TcdB in Zebrafish Embryos. To determine the local- Pease and have evolved different functions, ranging from pore ization of TcdB, zebrafish were treated with TcdBAlexa-546 and formation in plasma membranes to enzymatic activities that alter examined by fluorescence microscopy for sites of toxin tropism. intracellular signaling, cell cycle, apoptosis, and protein synthesis in As shown in Fig. 1, after a 24-h treatment with the toxin, targeted cells (1). Mechanisms of receptor binding, cell entry, TcdBAlexa-546 localized at the frontal ventral portion of the fish, membrane insertion, and enzymology are routinely determined by with specific foci formed within the pericardial region. Local- using a broad range of cell types in vitro, yet for many toxins, the cell ization of TcdBAlexa-546 was also observed in an anatomical types targeted during disease are unknown (2). region corresponding to the outflow chamber of the heart (see Clostridium difficile toxin B (TcdB) is an example of a bacterial arrow in Fig. 1A). Magnified views, as shown in Fig. 9 A and B, toxin studied extensively in vitro, but the in vivo activities remain which is published as supporting information on the PNAS web ϭ ͞ poorly understood (3). TcdB is a potent (LD50 200 ng kg) site, reveal intense localization near the cardiac region. In intracellular bacterial toxin; the protein enters cells by receptor- contrast, the negative control, BSAAlexa-546 did not show detect- mediated endocytosis; translocates to the cytosol; hydrolyzes able anatomical localization within the zebrafish (Fig. 9 C UDP-glucose; and transfers the liberated sugar to a reactive and D). threonine in the effector-binding loops of the small GTPases To further demonstrate specificity of TcdB localization, com- Rho, Rac, and Cdc42 (4–7). As a result, cultured cells treated petition experiments were performed by using the putative with TcdB exhibit changes in cell morphology and undergo receptor-binding domain (RBD) of TcdB. As shown in Fig. 1B, apoptosis, eventually leading to the death of the cell (8–10). cotreatment with a 30-fold molar excess of the TcdB RBD TcdB intoxicates numerous cell types in vitro, including fibro- reduced the detectable levels of labeled toxin to that observed blasts, neuronal cells, epithelial cells, endothelial cells, lympho- cytes, and hepatocytes (4, 11–15), yet whether any of these cell types are targeted during C. difficile-associated disease (CDAD) Author contributions: E.E.H. and J.D.B. designed research; E.E.H. and D.E.V. performed is unknown. research; E.E.H. and J.D.B. analyzed data; and E.E.H. and J.D.B. wrote the paper. In addition to TcdB, C. difficile also produces toxin A (TcdA), The authors declare no conflict of interest. which is known to function as an enterotoxin, causing gastroin- This paper was submitted directly (Track II) to the PNAS office. testinal damage (16, 17). Previous studies have shown that TcdB Abbreviations: TcdB, Clostridium difficile toxin B; CDAD, Clostridium difficile-associated is effective only when the intestinal mucosa is damaged (17), disease; PA, protective antigen; LFn, anthrax toxin lethal factor; RBD, receptor-binding suggesting that the intestinal effects of TcdA facilitate the entry domain; RCm, rat cardiomyocyte(s). of TcdB into the bloodstream. The TcdA-mediated release of *To whom correspondence should be addressed. E-mail: [email protected]. TcdB could explain the systemic complications observed in © 2006 by The National Academy of Sciences of the USA 14176–14181 ͉ PNAS ͉ September 19, 2006 ͉ vol. 103 ͉ no. 38 www.pnas.org͞cgi͞doi͞10.1073͞pnas.0604725103 Downloaded by guest on September 25, 2021 Fig. 2. RBC perfusion rate and vein integrity of TcdB-treated zebrafish. RBC perfusion rate was used as a measurement of blood flow and calculated as the average number of RBC per 30 s in the intersegmental veins (n ϭ 27), with error bars representing standard error. Zebrafish treated with TcdB (solid bars) had a reduced RBC rate compared with the heat-inactivated TcdB control (hatched bars). RBC per 30 s (P Ͻ 0.0001; see Fig. 2 and Movies 1 and 2, which are published as supporting information on the PNAS web site). Reduced blood flow was observed in the caudal and interseg- mental veins and appeared to occur in the absence of detectable damage to the vascular endothelium. To confirm this, fli1::EGFP zebrafish, which express GFP in endothelial cells, were used to assess vein integrity after treatment with TcdB. As shown in Fig. 10B, which is published as supporting information on the PNAS web site, despite a loss in blood flow, the veins of toxin-treated zebrafish appeared to be intact. TcdB-treated zebrafish were examined for cardiac damage, as a possible explanation for the reduction in blood flow. Between 24 and 48 h after treatment, there was a decrease in ventricle Fig. 1. TcdB localization observed in zebrafish. Fluorescently labeled TcdB chamber contractility and a loss in heart looping (see Fig. 3; see was used to follow localization of the toxin to specific anatomical regions. (A) also Fig. 11 and Movies 3 and 4, which are published as Zebrafish treated with 37 nM TcdBAlexa-546 for 24 h. Arrows indicate toxin supporting information on the PNAS web site). As shown in Fig. MICROBIOLOGY accumulation around the pericardial sac as well as distinct foci on the yolk sac 3, in control fish, the ventricle exhibited a dynamic change in size Alexa-546 and upper cranial region. (B) Zebrafish treated with 37 nM TcdB and of 20% during contraction and expansion (see Movie 3). How- 370 nM TcdB (RBD). (C) Brightfield image of zebrafish treated with 37 nM TcdBAlexa-546 and 370 nM TcdB (RBD). Arrows 1, 2, and 3 denote the heart, yolk ever, treatment with TcdB substantially reduced the change in sac, and eye of zebrafish, respectively. ventricle size during beating (see Movie 4), indicating the heart was unable to contract and expand in a normal fashion. At Ϸ48 h after treatment, both the atrium and ventricle were deformed in the BSA control. Collectively, these observations suggested (Fig. 4). By 7 days after treatment, 100% of TcdB-treated fish TcdB exhibits specific tissue tropism in the zebrafish, with the exhibited pericardial edema, with 70% developing whole-body toxin primarily localizing to the yolk-sac, pericardial, and cardiac edema (Fig. 5). TcdB-treated fish survived between 7 and 10 regions of the zebrafish. days after initial exposure to TcdB, but by 11 days after treatment, 100% of the fish succumbed to the effects of the toxin.