I have worked with a range of collaborators to solve the glycan and/or protein binding from a range of toxin families including cholesterol dependent cytolysins (CDCs), AB5 toxins and leukocidins.
In this broad ranging program I have worked with Lucy Shewell, Mike Jennings, James Paton and Mark Walker to solve the carbohydrate receptors of eight CDCs (1) and a shared protein target for the CDCs pneumolysin (Ply) and StreptolysinO (SLO).
The protein targeting of the CDCs was further enabled by Victor Torres (NYU) as we collaborated on the host target identification of the leukocidinAB. LukAB had high affinity binding to the I-domain of the human integrin protein CD11b (2). While investigating the affinity for LukAB mutants (3) for the I-domain of CD11b, I was also screening CDCs against another target and happened to discover by chance the high affinity binding of Ply and SLO to CD11b. The binding of SLO to CD11b is remarkably similar to that of LukAB, recognizing the same region and also having human specificity. Ply bound in a very different manner recognizing the I-domain in a region that is glycosylated in the native receptor, a glycosylation site that typically contains a terminal sialyl-Lewis X motif that is preferred by Ply (1).
An unexpected outcome from toxin analysis came with work on the toxin SubB in collaboration with the Paton's, Mike Jennings and Lucy Shewell. SubB was already known to be quite specific for the non-human produced sugar N-Glycolylneuraminic acid (Neu5Gc), however, it only recognized certain linkages and still bound to N-Acetylneuraminic acid (Neu5Ac). Neu5Gc is a well recognized cancer marker and was being investigated as a potential diagnostic by several groups. By reengineering SubB to SubB2M we developed a hugely specific and sensitive protein that can recognize a pan cancer marker (4,5,6).
Through the study of toxins I have increased the knowledge of host targets and helped to create a novel application for a bacterial toxin.