| Abstract | In previous studies protective antibodies that could facilitate bactericidal killing of Neisseria meningitidis (Nm) serogroup B strains were derived from immunisation with glycoconjugates prepared from O-deacylated lipopolysaccharide (LPS-OH) via direct reductive amination between the reducing end of the oligosaccharide molecule, created by treatment with alkaline phosphatase, and amino functionalities on the CRM197 carrier protein. These glycoconjugates proved difficult to prepare because the presence of amide linked fatty-acyl groups results in glycolipids that are relatively insoluble and aggregate. Therefore, we have examined several strategies to prepare glycoconjugates in order to identify a robust, consistently reproducible strategy that produces glycoconjugates with a high loading of LPS derived oligosaccharides. Initially we used completely deacylated LPS molecules, but lacking phosphoethanolamine (PEtn) from the core OS as the strong basic conditions required to completely deacylate the LPS would modify the PEtn residue. We utilised a squarate linker and conjugated via the reducing end of the carbohydrate antigen following removal of the glycosidic phosphate to amino groups on CRM197, however carbohydrate loading on the carrier protein was low. Glycoconjugates were then produced utilising amidases produced by Dictyostelium discoideum (Dd), which partially remove N-linked fatty acids from the lipid A region of the Nm LPS molecule, which enabled the retention of the PEtn residue. LPS-OH was treated with Dd amidase, the reducing glycosidic phosphate removed, and using a cystamine linker strategy, conjugated to the carrier protein. Carbohydrate loading was somewhat improved but still not high. Finally, we have developed a novel conjugation strategy that targets the amino functionality created by the amidase activity as the attachment point. The amino functionality on the PEtn residue of the inner core was protected via a novel blocking and unblocking strategy with t-butyl oxycarbonyl. A maleimide-thiol linker strategy, targeting lysine residues on the carrier protein did not result in high loading of the carbohydrate molecules, however when we targeted the carboxyl residues we have consistently obtained a high loading of carbohydrate antigens per CRM197, which can be controlled by variation in the amount of activated carbohydrate utilised in the conjugation reaction. |
|---|
