The Impact of AB5 Toxin Binding on Ganglioside Mimicry in Campylobacter jejuni Lipooligosaccharides

Abstract

Campylobacter jejuni is an important pathogen in the developed world, causing more cases of acute gastroenteritis than any other bacteria. C. jejuni is also the leading cause of paralysis in the world, causing Guillain‐Barré syndrome in humans by mimicry of nerve gangliosides with its surface exposed lipooligosaccharides (LOS). In some cases this mimicry elicits a breakdown in immune tolerance and results in self‐reactive antibodies which damage the nerve cells. Interestingly, the commonly mimicked GM1 ganglioside is also the main receptor for cholera toxin (CT) from Vibrio cholerae as well as heat‐labile enterotoxin (LT) from Escherichia coli. We have shown that these two toxins bind C. jejuni strains that mimic GM1 ganglioside and clear bacterial growth. The binding and clearing phenotypes were observed closely using transmission and scanning electron microscopy. C. jejuni 11168 shows natural variation in its ability to mimic GM1 gangliosides. Due to phase‐variation in cgtB encoding the galactosyltransferase, this strain exists as a mixture cells with and without GM1 mimics. It was unclear why the strain retained this variation, given that mimicry of host gangliosides lends a significant fitness advantage by camouflaging the bacterium from host immune defenses. Therefore, LOS was isolated from 11168 exposed to CT and LT toxin and compared to the unexposed by silver staining and far Western blotting. These experiments revealed the loss of a band from toxin exposed LOS and a marked decrease in the ability to bind toxin. Furthermore, when genomic DNA was isolated from these two groups and the cgtB gene was sequenced, the variation that existed in the unexposed group was no longer observed in the presence of the toxins. This led to the hypothesis that C. jejuni 11168 may retain variation in the cgtB gene in order to avoid negative effects upon contact with these toxins in the host or the environment. In addition, when we used a strain of E. coli engineered to display the GM1 mimic on its surface, no clearance was observed suggesting that this interaction is pathogen specific. Our study identifies a new interaction between C. jejuni and the AB5 toxins that has not been previously described and provides a possible new relationship within the intestinal microbiome.Support or Funding InformationFunding provided by the Natural Sciences and Engineering Research Council of Canada, Alberta Innovates Technology Futures and Alberta Innovates Health Solutions.