Despite enormous scientific efforts, development of an effective vaccine for HIV remains a significant challenge. Efforts to date have largely overlooked a key vulnerability of the virus with respect to its surface glycosylation. With this application we propose to take the first step in exploiting the vulnerability of the virus with respect to its glycosylation, which is an approach that is distinct from the current and earlier attempts to identify monoclonal antibodies and to identify and elicit broadly neutralizing polyclonal sera. The virus carries on a continuous battle with the host immune system, including the humoral antibody response. The sole target of neutralizing antibodies is the viral Env protein. One part of the viruses'defense of this protein is encoding a large number of glycosylation sites to occlude regions of Env surface from recognition by antibodies. The sites for glycosylation in the conserved domains of Env can vary in their presence, in some cases missing in 40% of isolates. We hypothesize that antibodies that can prevent the acquisition of infection can be targeted to surface structures of Env where a carbohydrate attachment site is missing. We have grouped the carbohydrates and surface structures into six CONEs - Carbohydrate-Occluded Neutralization Epitopes. In a survey of a large number of transmitted viruses we have documented that 70% of these viruses are missing at least two of these conserved carbohydrates, which indicates that these viruses will be susceptible to a combinatorial approach of neutralization directed at the six CONEs. Thus, we hypothesize that it is possible is to develop immunogens for each CONE that can be used to generate antibodies that bind specifically to each of the CONEs in the context full-length Env. Given that these six CONEs belong to surface regions of Env that are relatively conserved but with the possibility of exposure of at least 1-2 CONEs in most isolates, our strategy has the potential when used combinatorially to have a broad neutralization effect. We will first design soluble protein scaffolds that display the individual CONEs (Aim 1) by identifying low molecular weight protein scaffolds that display a surface patch that is similar to the given CONE and redesigning the scaffolds to display the CONEs. We will then screen polyclonal human sera to identify naturally occurring antibodies that bind to specific CONEs. which will provide evidence for the selective pressure that maintains the high level of prevalence of the glycosylation sites in the conserved regions (Aim 2). We will then test sera that react to the CONE mimetics for their ability to neutralize pseudotyped viruses based on the presence or absence of specific carbohydrate attachment sites at the specific CONE (Aim 2). We will then immunize rabbits to raise polyclonal antibodies to CONE mimetics and then test their sera for neutralization properties (Aim 3). Determining the breadth of antibody activity by testing the sera in a broader panel of Env proteins using pseudotyped viruses (Aim 4) will inform us of potential broadly neutralizing antibody. We will examine the relationship between the presence of a glycosylation site and its impact on viral fitness (Aim 5).
We propose to design immunogenic proteins that mimic specific surfaces of HIV envelope that will help in developing neutralizing antibodies against the virus. Significantly, we are targeting surfaces of the HIV envelope that are especially vulnerable due to the occasional loss of glycosylation due to mutation, as is seen in clinical isolates. Our proposed research combines computational, scaffold-based protein engineering with HIV antibody development. Thus, this proposal has direct relevance in HIV vaccine development and the methodology can be generalized to engineer immunogens against other viruses.
|Zhu, Cheng; Mowrey, David D; Dokholyan, Nikolay V (2017) Computational Protein Design Through Grafting and Stabilization. Methods Mol Biol 1529:227-241|
|Redler, Rachel L; Das, Jhuma; Diaz, Juan R et al. (2016) Protein Destabilization as a Common Factor in Diverse Inherited Disorders. J Mol Evol 82:11-6|
|Dokholyan, Nikolay V (2016) Controlling Allosteric Networks in Proteins. Chem Rev 116:6463-87|
|Politi, Regina; Convertino, Marino; Popov, Konstantin et al. (2016) Docking and Scoring with Target-Specific Pose Classifier Succeeds in Native-Like Pose Identification But Not Binding Affinity Prediction in the CSAR 2014 Benchmark Exercise. J Chem Inf Model 56:1032-41|
|Kimura, Sachie; Caldarini, Martina; Broglia, Ricardo A et al. (2014) The maturation of HIV-1 protease precursor studied by discrete molecular dynamics. Proteins 82:633-9|
|Ping, Li-Hua; Joseph, Sarah B; Anderson, Jeffrey A et al. (2013) Comparison of viral Env proteins from acute and chronic infections with subtype C human immunodeficiency virus type 1 identifies differences in glycosylation and CCR5 utilization and suggests a new strategy for immunogen design. J Virol 87:7218-33|