Role of envelope compactness and glycosylation in HIV-1 resistance to neutralising antibody responses

Doctoral Thesis


Permanent link to this Item
Journal Title
Link to Journal
Journal ISSN
Volume Title

University of Cape Town

Understanding the mechanisms used by HIV-1 to evade antibody neutralisation may contribute to the design of a high-coverage vaccine. This thesis explores the mechanism used by a Tier 3 virus leading to its high antibody neutralisation resistance phenotype. This thesis also describes how the glycans at the base of the V3 loop contribute to (i) breadth and potency in a cohort of unselected HIV-1-infected individuals and (ii) the selective pressures resulting from the V3/glycans shielding the virus from neutralisation and the glycans themselves being targets of broad antibody responses. HIV-1 isolates that are highly resistant to broadly neutralising antibodies could limit the efficacy of an antibody-based vaccine. For this reason, it is important to understand the mechanisms behind high HIV-1 resistance to neutralising antibodies. Chapter 2 and Chapter 3 of this thesis describe virus 253-11, a highly neutralisation resistant virus, which is particularly resistant to commonly-elicited, anti-membrane proximal external region (MPER) antibodies in sera. To further understand its resistance, mutations in the MPER were introduced that are known to delay fusion following CD4-binding and thus increase the time the virus spends in the open conformation. Interestingly, we found that these mutations affect the 253-11 Envelope (Env) spike before CD4-binding by destabilising the closed trimer structure. From these data, we hypothesized that the neutralisation resistance of 253-11 was due to an unusually tight, compact pre-fusion Env trimer that resists transient changes to the open conformation. The open conformation frequently exposes narrowly-neutralising antibody epitopes. Because the unliganded 253-11 Env presumably transitions infrequently into the open conformation, it would be able to evade these responses. 253-11 was sensitive to most but not all of the most potent broadly neutralising antibodies (bnAbs) tested, most likely because those broadly neutralising antibodies can access their epitopes in the pre-fusion Env conformation. To gain further information about the structure of the 253-11 Env, we designed a recombinant 253-11 SOSIP trimer and found it to be stable and predominantly adopt a closed conformation. The crystal structure of the SOSIP trimer revealed structural elements likely responsible for 253-11 Env compactness including the inward disposition of the heptad repeat helices and gp120 protomers towards the trimer axis. Taken together, the data from Chapter 2 and Chapter 3 highlight an underappreciated Env compactness mechanism of HIV-1 resistance to neutralising antibodies and these data may be useful in HIV-1 immunogen design research. Previous candidate HIV vaccines have failed to induce wide-coverage neutralising antibodies capable of substantially protecting vaccinees. A key approach in HIV immunogen development has been to define and model epitopes recognised by anti-HIV bnAbs. Candidate immunogen models identified by bnAbs include the V3/glycans, the V2/apex and the MPER epitopes. Autoreactivity and polyreactivity of anti-V3/glycan and anti-MPER antibodies are thought to pose both direct and indirect barriers to achieving neutralisation breadth. Chapter 4 of this thesis explored which of these bnAb epitopes were associated with breadth and potency in a South African cohort of chronically HIV-infected individuals. The study found that antibodies targeting the V3/glycans were associated with breadth and potency. In contrast, antibodies to the V2/apex were not associated with neutralisation breadth/potency. This suggests that auto/polyreactivity are not critical factors in the development of breadth and potency and that the V3/glycans should remain a high-priority vaccine candidate. Since targeting the V3/glycans was associated with breadth and potency in this cohort, the study continued to look at this epitope to investigate the role of these glycans in neutralisation resistance of Tier 2 viruses. The HIV-1 Env is surrounded by glycans that often prevent antibody neutralisation, leading to the term the "glycan shield", however some bnAbs have evolved to recognise these carbohydrates. Chapter 4 of this thesis describes how the N-linked glycan at position N301 is critical for maintaining neutralisation resistance of one subtype C virus (Du156.12), but not for another subtype-matched virus (CAP45.2.00.G3). Thus, the loss of the N301 glycan may have a substantial antibody-related fitness cost for some viruses but not others. The N301 glycan, as well as glycans at positions 332 and 334, are the primary targets of the anti-V3/glycan class of neutralising antibodies, which may select for loss of the targeted glycan. The evidence presented in Chapter 4 suggests that in some viruses, loss of the N301 glycan may result in evasion of anti-V3/glycan antibody responses while maintaining overall neutralisation resistance. This phenomenon may impair efficacy of passively-infused anti-V3/glycan bnAbs or a therapeutic vaccine.