Comparison of oral toxicological properties of botulinum neurotoxin serotypes A and B

Abstract

Botulinum neurotoxins (BoNTs) are among the most potent biological toxins for humans. Of the seven known serotypes (A–G) of BoNT, serotypes A, B and E cause most of the foodborne intoxications in humans. BoNTs in nature are associated with non-toxic accessory proteins known as neurotoxin-associated proteins (NAPs), forming large complexes that have been shown to play important roles in oral toxicity. Using mouse intraperitoneal and oral models of botulism, we determined the dose response to both BoNT/B holotoxin and complex toxins, and compared the toxicities of BoNT/B and BoNT/A complexes. Although serotype A and B complexes have similar NAP composition, BoNT/B formed larger-sized complexes, and was approximately 90 times more lethal in mouse oral intoxications than BoNT/A complexes. When normalized by mean lethal dose, mice orally treated with high doses of BoNT/B complex showed a delayed time-to-death when compared with mice treated with BoNT/A complex. Furthermore, we determined the effect of various food matrices on oral toxicity of BoNT/A and BoNT/B complexes. BoNT/B complexes showed lower oral bioavailability in liquid egg matrices when compared to BoNT/A complexes. In summary, our studies revealed several factors that can either enhance or reduce the toxicity and oral bioavailability of BoNTs. Dissecting the complexities of the different BoNT serotypes and their roles in foodborne botulism will lead to a better understanding of toxin biology and aid future food risk assessments.

Introduction

The disease botulism is caused by seven known serotypes of BoNTs that are produced by Clostridium botulinum (alphabetically from BoNT/A-BoNT/G), C. butyricum (BoNT/E), C. baratii (BoNT/F), and C. argentinense (BoNT/G) (Hill et al., 2007). Most food-borne intoxications are caused by BoNT serotypes A, B, E and occasionally by serotype F (Arnon et al., 2001, Bigalke and Rummel, 2005, Scarlatos et al., 2005). BoNT is synthesized by the bacterium as a 150 kDa polypeptide (referred to here as holotoxin), which consists of a 50 kDa light chain, and a 100 kDa heavy chain domains linked by a disulfide bond. The heavy chain contains the translocation (N-terminus region of the heavy chain) and cell-binding domain (C-terminus region of the heavy chain) (Arnon et al., 2001, Simpson, 2004). The light chain contains the endopeptidase domain, which cleaves proteins associated with intracellular vesicular transport, such as SNAP25 (synaptosomal-associated protein of 25 kDa) for BoNT/A and VAMP2 (vesicle-associated membrane protein) for BoNT/B, and consequently inhibits acetylcholine release from neurons, leading to muscle paralysis.

BoNTs assemble with non-toxic neurotoxin-associated proteins (NAPs) to form large protein complexes (described here as BoNT complex), known also as “progenitor” toxins. Toxin complexes vary in sizes of 12S, 16S or 19S, also known as M, L and LL toxins of molecular sizes 300 kDa, 500 kDa, and 900 kDa, respectively (Eisele et al., 2011, Hines et al., 2005, Inoue et al., 1996, Ohishi et al., 1977). The 12S toxin is composed of BoNT and a non-toxic, non-hemaglutinin protein (NTNH); the larger complexes of 16S and 19S are formed from BoNT with NTNH and hemaglutinin proteins (HA). NAPs play a substantial role in oral poisoning by protecting the holotoxin from degradation and promoting toxin uptake by epithelial cells (Fujinaga et al., 2009, Simpson, 2004, Sugii et al., 1977). The molecular size of toxin complexes was shown to be directly proportional to its oral bioavailability; the larger the molecular size, the greater the resistance to degradation by pepsin and/or gastric secretions (Chen et al., 1998, Ohishi et al., 1977, Sugii et al., 1977).

The mechanism by which the BoNT complex traverses the intestinal barrier into the blood stream and then finds its target, the peripheral motor neurons, is not fully understood. BoNT holotoxins, specifically the heavy chain fragments, have been shown to cross the intestinal epithelium via a transcytosis mechanism. This is thought to occur at the apical side of epithelial membranes to reach the basolateral side and then into the blood stream, where the complex presumably breaks apart (Ahsan et al., 2005, Fujinaga, 2010, Fujinaga et al., 2009, Maksymowych and Simpson, 1998, Maksymowych and Simpson, 2004). Other studies have suggested a role for NAPs such as HA33 to disrupt epithelial cell membranes and thus promote toxin complex passage (Fujinaga et al., 2009, Jin et al., 2009). Thus, there appears to be at least two mechanisms for BoNT complex translocation, one NAP-independent and the other NAP-dependent.

The organization and arrangement of toxin complex genes are similar for some serotypes of BoNT. For example, the BoNT/B complex contains NAPs NTNH, HA70, HA33 and HA17, and is similar to some BoNT/A subtypes (Hines et al., 2005, Smith et al., 2007). Yet, the BoNT/B complex has been shown to be more toxic than the BoNT/A complex in oral toxicity studies (Ohishi, 1984, Sugii et al., 1977). The basis for these serotype differences has not been explained. Most studies have been performed with purified holotoxin or complex BoNTs in buffer. The effect of a complex matrix, such as food, on BoNT oral bioavailability is still unclear. Recently, toxin dose effects and oral bioavailability of BoNT/A in a few food matrices have been investigated (Cheng et al., 2008). Few published studies are available on the oral bioavailability of BoNT/B and BoNT/E, both of which are important in foodborne intoxications.

In this study, we determined the in vivo dose-response relationship between purified BoNT/B holotoxin and its toxin complex, and the effect of different food matrices on oral bioavailability. In addition, the oral bioavailability of BoNT/B complex was contrasted with that of BoNT/A complex, revealing important serotype differences in toxicity.