Date Published: January 26, 2017
Publisher: Public Library of Science
Author(s): Susan Secore, Su Wang, Julie Doughtry, Jinfu Xie, Matt Miezeiewski, Richard R. Rustandi, Melanie Horton, Rachel Xoconostle, Bei Wang, Catherine Lancaster, Adam Kristopeit, Sheng-Ching Wang, Sianny Christanti, Salvatore Vitelli, Marie-Pierre Gentile, Aaron Goerke, Julie Skinner, Erica Strable, David S. Thiriot, Jean-Luc Bodmer, Jon H. Heinrichs, Yung-Fu Chang.
Clostridium difficile infections (CDI) are a leading cause of nosocomial diarrhea in the developed world. The main virulence factors of the bacterium are the large clostridial toxins (LCTs), TcdA and TcdB, which are largely responsible for the symptoms of the disease. Recent outbreaks of CDI have been associated with the emergence of hypervirulent strains, such as NAP1/BI/027, many strains of which also produce a third toxin, binary toxin (CDTa and CDTb). These hypervirulent strains have been associated with increased morbidity and higher mortality. Here we present pre-clinical data describing a novel tetravalent vaccine composed of attenuated forms of TcdA, TcdB and binary toxin components CDTa and CDTb. We demonstrate, using the Syrian golden hamster model of CDI, that the inclusion of binary toxin components CDTa and CDTb significantly improves the efficacy of the vaccine against challenge with NAP1 strains in comparison to vaccines containing only TcdA and TcdB antigens, while providing comparable efficacy against challenge with the prototypic, non-epidemic strain VPI10463. This combination vaccine elicits high neutralizing antibody titers against TcdA, TcdB and binary toxin in both hamsters and rhesus macaques. Finally we present data that binary toxin alone can act as a virulence factor in animal models. Taken together, these data strongly support the inclusion of binary toxin in a vaccine against CDI to provide enhanced protection from epidemic strains of C. difficile.
Clostridium difficile infections are the most widely recognized cause of hospital acquired infectious diarrhea . There is a critical need for a vaccine for the prevention of this disease. A recent study by the Duke Infection Outreach Network found that C. difficile has superseded Methicillin-Resistant Staphylococcus aureus (MRSA) as the most common pathogen causing healthcare associated infections in the southeastern United States . A recent bulletin from the Centers for Disease Control and Prevention (CDC) (http://www.cdc.gov/drugresistance/threat-report-2013) listed the current threat level from C. difficile as urgent. According to this CDC bulletin, there are 250,000 infections each year caused by this bacterium that require hospitalization or affect already hospitalized patients resulting in 14,000 deaths and at least $1 billion in excess medical costs each year. The organism is associated with persistent diarrhea primarily in individuals of advanced age with pre-existing co-morbidities, during prolonged hospitalization, and, most importantly, with the use of broad-spectrum antibiotics. Because the organism can form spores which are impervious to antibiotics, there is a significant risk of recurrence (about 30%).
Traditional vaccine development against CDI has focused on either toxoid or inactive fragments of TcdA and/or TcdB, the only known causative agents of CDI [19, 20, 22–24, 45]. Here we describe a novel tetravalent vaccine composed of inactivated TcdA, TcdB and binary toxin, expressed in insect cells using a baculovirus system. Initial attempts to express these proteins using several heterologous bacterial expression systems including E. coli, Pseudomonas fluorescens, and Bacillus megaterium yielded very low expression of proteins that were typically unstable and difficult to purify. Expression of these toxins in eukaryotic systems was deemed to be impractical, due to the conservation of the toxins’ targets in these systems and the perceived potential for cytopathic effects of the toxins on eukaryotic cells. However, despite the low levels of residual toxicity in 5mTcdA and 5mTcdB, we were able to stably express all four proteins at high levels in insect cells. This was likely because the residual toxicity detected in 5mTcdA and 5mTcdB was unrelated to these molecules’ endogenous glucosyltransferase activity. Our hypothesis was supported by the observation that we were unable to detect glucosyltransferase activity from 5mTcdA and 5mTcdB using a biochemical Rho-UDP glucosylation assay (data not shown). Furthermore, the cell intoxication resulting from the addition of 5mTcdA or 5mTcdB was morphologically different then that observed by the addition of native toxins (data not shown). Exposing Vero cells to the native toxins resulted in cell rounding of the still attached cells, presumably through actin depolymerization following glucosylation of cellular Rho-like proteins. Conversely, the mutant toxins did not cause cell rounding, but rather the cells detached from the plate and exhibited signs of lysis. Our findings were similar to those reported by Donald et al. , who also observed a similar cell phenotype with their inactivated toxins that was consistent with pore-induced membrane leakage, swelling and then lysis, and likely results from caspase activation . It appears that insect cells are less sensitive to this alternative toxic mechanism, allowing production of mutated LCTs.