Clostridium difficile toxins and severe C. difficile infection.

The pathogenesis of colitis due to Clostridium difficile is closely linked to the elaboration of 2 large, single-unit glucosyltransferases referred to as toxin A (TcdA) and toxin B (TcdB). Clinical isolates from patients with symptomatic C. difficile infection (CDI) invariably produce both TcdA and TcdB or TcdB alone. Vaccination of experimental hamsters against both toxins prevents fatal C. difficile cecitis, and isogenic mutants of C. difficile in which both toxin genes are inactivated are rendered avirulent [1]. Debate remains about the relative importance and specific roles of these 2 toxins, but both toxins, which are very similar in structure and function, are likely involved [2]. Infection with C. difficile is usually manifest as colitis, with rare reports of bacteremia and extracolonic infections, such as hepatic or splenic abscess. Although the infection is typically limited to the colon, CDI is not infrequently severe and accompanied by systemic complications, including fever, hypotension, and shock. The epidemiology of CDI in North America has changed significantly in the last decade, with increased rates of disease throughout the United States and Quebec, Canada [3, 4]. In parallel with increased rates of disease, the relative severity of illness has also increased. During the multihospital CDI outbreak in Quebec in 2004, the directly attributable CDI mortality was 6.9%, and CDI contributed to death in another 7.5%; these rates are in contrast to an overall CDI mortality rate of 1.5% in Canada 7 years earlier [4]. While progress has been made in the management of recurrent CDI, there has been little improvement in the management of severe CDI, and most treatment and strategies are empirical or based on anecdotal observations. Rational treatments for severe CDI await a better understanding of the pathogenesis of severe disease. Steele and colleagues [5] report in this issue of the Journal their findings correlating the detection of TcdA and TcdB in serum of piglets and mice infected with C. difficile and the development of systemic disease. Bartlett and colleagues [6] first reported toxemia in clindamycintreated hamsters in 1978. Blood from all 20 hamsters in that study were positive for a cytotoxin neutralized by antitoxin to Clostridium sordellii, and the authors speculated that death in the hamsters was a result of absorption of the toxin(s) produced by C. difficile. This observation, however, has not been repeated or extended until recently. While mortality is nearly 100% in clindamycin-treated hamsters, morbidity and mortality following CDI in other animal models is variable. Steele and colleagues were able to use gnotobiotic piglets and mice to test the hypothesis that toxemia correlates with systemic disease and mortality. The authors used 2 substrains of the epidemic BI/NAP1/027 C. difficile strain responsible for the recent CDI outbreaks in Quebec and the United States [3, 4] to infect the animals, as well as a novel immunocytotoxicity (ICT) assay to detect TcdA in tissue culture. This assay, based on antibody enhancement of TcdA toxicity in cells expressing Fc-c receptor I, allows enhanced detection of TcdA, which is inherently less cytotoxic than TcdB for most cell lines. The authors found toxin by the ICT assay in approximately one-third of the serum samples from the infected piglets and mice, as well as in the pleural fluid and ascites from these animals. Rac1 glycosylation assays corroborated these findings, and neutralization of both toxins in the in vitro assays by anti-TcdA plus antiTcdB antiserum indicated the presence of TcdA and TcdB. Furthermore, 67% of serum samples (12/18) from piglets with systemic CDI and none of the serum samples (0/25) from the piglets with nonsystemic CDI were positive for toxin, and 85% of the serum samples (23/27) from mice with severe CDI and none of the serum samples (0/34) from mice with nonsystemic CDI were positive. Both of these results were highly significant. The authors were also able to prevent toxemia and systemic CDI in vivo by administering neutralizing anti-TcdA Received and accepted 29 September 2011; electronically published 5 December 2011. Correspondence: Stuart Johnson, MD, Research Service, Hines Veterans Affairs Hospital, 5000 S 5th St., Hines, IL 60141 ([email protected]). The Journal of Infectious Diseases 2012;205:353–4 Published by Oxford University Press on behalf of the Infectious Diseases Society of America 2011. DOI: 10.1093/infdis/jir752