Abstract
As a step toward the development of variant forms of Env with enhanced immunogenic properties, we have expressed the glycoprotein in the yeast surface display system in a form that can be subjected to random mutagenesis followed by screening for forms with enhanced binding to germline antibodies. To optimize the expression and immunogenicity of the yeast-displayed Env protein, we tested different approaches for cell wall anchoring, expression of gp120 and gp140 Env from different viral strains, the effects of introducing mutations designed to stabilize Env, and the effects of procedures for altering N-linked glycosylation of Env. We find that diverse forms of HIV envelope glycoprotein can be efficiently expressed at the yeast cell surface and that gp140 forms of Env are effectively cleaved by Kex2p, the yeast furin protease homolog. Multiple yeast-displayed gp120 and gp140 proteins are capable of binding to antibodies directed against the V3-variable loop, CD4 binding site, and gp41 membrane-proximal regions, including some antibodies whose binding is known to depend on Env conformation and N-linked glycan. Based on antibody recognition and sensitivity to glycosidases, yeast glycosylation patterns partially mimic high mannose-type N-glycosylation in mammalian cells. However, yeast-displayed Env is not recognized by some anti-Env antibodies sensitive to quaternary structure, suggesting either that the displayed protein exists in a monomeric state or that for these antibodies, yeast glycosylation in certain regions hinders recognition or access. Consistent with studies in other systems, reconstructed predicted unmutated precursors to anti-Env antibodies exhibit little affinity for the yeast-displayed envelope protein.
Highlights
Despite the success of anti-retroviral drugs in treating HIV-infected individuals and slowing the spread of infection, these drugs remain unaffordable by much of the world and are subject to side effects and development of resistance
In evaluating antibody binding to envelope glycoprotein (Env) constructs, we considered the statistical significance of the observed mean fluorescence intensity, compared to binding of the same antibody to cells expressing empty vector. (S2 Table provides a tabulation of the p values for each antibody/Env construct combination). 16 of the 19 anti-Env antibodies tested in Fig 4 exhibited binding to at least one yeast-displayed Env construct characterized by a p value of less than 0.01 and no significant binding was observed to the human negative control antibody 1069-D6
To examine effects of glycosylation on antibody binding to yeast-displayed Env, we treated yeast-displayed QH0692 gp120dsm and gp140dsm proteins with two glycosidases; endoglycosidase H, which cleaves between the N-acetylglucosamine residues of the chitobiose core of N-linked glycans, leaving one N-acetylglucosamine residue attached to the backbone asparagine [91] and Peptide:N-Glycosidase F (PNGase F), which cleaves the linkage between the backbone asparagine and the first N-acetylglucosamine residue [92]
Summary
Despite the success of anti-retroviral drugs in treating HIV-infected individuals and slowing the spread of infection, these drugs remain unaffordable by much of the world and are subject to side effects and development of resistance. Intensive efforts to develop an effective AIDS vaccine (more than 200 clinical trials [2]) have failed to yield an effective immunogen, aside from a single trial showing only a modest protective effect [3]. The potential feasibility of effective vaccination is suggested by the observation that some infected individuals develop antibodies that can neutralize a broad array of virus strains in vitro and can prevent virus transmission in animal models. All known broadly neutralizing antibody responses against HIV are targeted to the viral envelope glycoprotein (Env) that is initially translated as a 160 kDa product (gp160), cleaved by cellular furin proteases into two chains, the gp120 and gp glycoproteins. These include the CD4 receptor binding site [6, 7], certain glycans on the Env surface [8], sites on gp that are likely to be involved in membrane fusion with target cells [9], and sites on the V2 and V3 loops [10] and V1 and V2 loops [11] of gp120 that appear to depend on a particular quaternary structure of the envelope protein
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
