N-linked glycosylation of HIV-1 and Hantaan virus glycoproteins. Implications for vaccination and therapy

Abstract

Enveloped viruses have a great advantage comparedto the non-enveloped viruses due to theirability to acquire not only a host membrane butalso host-specific glycosylation of their surfaceglycoproteins. These carbohydrate structures,mainly of the N-linked glycan class, have a lowimmunogenicity and this feature enables envelopedviruses to keep a low profile towards theimmune system by shielding crucial epitopesfrom antibody recognition. This phenomenonconstitutes hurdles for development of vaccinesand therapies targeting highly glycosylated virusesincluding HIV and Hantaan virus. The aimsof the present work were to identify specific Nlinkedglycans that are especially active in thisvirus-shielding process and develop DNA vaccinesand therapeutic vectors to the two virusesmentioned.An N-linked glycan attached to N306 in the V3loop of the HIV-1 gp120 is involved in protectingthe virus from antibodies. A DNA vaccinewas created, where this particular glycosylationsite was eliminated using site-directed mutagenesis.Mice were immunised intranasally withthe mutant and a wild type vaccine, and theimmune responses were analysed. The protocolincluded an analysis of animals immunised withDNA only and also those given a protein boost.The mutant DNA followed by protein boostinduced a significantly higher IgG response togp120 as measured by ELISA. Both DNA vaccinesinduced similar levels of HIV-neutralisingantibodies in serum. The IgA response was analysedboth for BAL and faeces specimens. SecretedIgA was induced by both DNA vaccinesbut protein boost was required to obtain virusneutralisingantibodies.Attempts were made to construct an adenovirusvector, with receptor specificity retargeted togp120 of the HIV-infected cell surface. To obtaina suitable affinity module for gp120 a combinatorialprotein library of 108 members, based onthe 58 amino acid residue staphylococcal proteinA (Affibody), was screened against gp120.The selected affibody (zg120) and its bivalentform had Kd- values of 100 nM and 10 nM,respectively, towards gp120. The affibody constructswere introduced to the adenovirus 5 fibregene, and the recombinant fibres bound selectivelyto gp120 in a biosensor analysis and togp160, transiently expressed in GMK cells. Thesuccessful selection of a gp120-binding affibodyligand indicates that future affibody-basedstrategies might evolve to complement antibodybasedefforts for HIV-1 therapy.The same strategy as described above for HIV-1was also applied to develop a DNA vaccine toHantaan virus glycoproteins. All six potentialglycosylation sites (N1-N6) of the two Hantaanvirus glycoproteins G1 and G2 were eliminatedand expressed as single mutations in a DNAvaccine. Mice were immunized by Gene Gunwith these constructs. One mutated DNA vaccineconstruct, N6, was found to induce equallyhigh neutralisation titres as did the wild typeDNA vaccine. However, the mutant but not thewild type DNA vaccine was able to induce partialprotection against Hantaan virus challenge.In conclusion, manipulation of glycosylationsites of viral glycoprotein genes is still open as aroad to an improved DNA vaccine capable ofeliciting high titres of broadly neutralising antibodies,but it is clear that there are still obstaclesto be overcome.