Systematic comparison of HIV-1 Envelope-specific IgG responses induced by different vaccination regimens: Can we steer IgG recognition towards regions of viral vulnerability

Abstract

Introduction: There is still an urgent need for a sufficiently protective HIV vaccine. Enormous genetic variability, high viral mutation rates leading to escape variants, and poor surface accessibility of the HIV-1 Env protein are major hurdles to the elicitation of a protective immune response (1–3). The HIV-1 virus is always “one step ahead of its host”, and viral variants that successfully evade the human immune response typically prevail. An HIV vaccine might be able to break this vicious cycle (4).The holy grail of HIV vaccine research remains the induction of broadly neutralizing antibodies (bNAb) with the ability to neutralize a wide range of viral variants. Despite substantial progress in this field, induction of broad cross-neutralizing antibodies by vaccination still remains challenging (1, 5). In addition, the very few people living with HIV who produce highly potent bNAb do so only after several years of exposure to HIV (6, 7). On the other hand, vaccine-induced non-neutralizing antibodies (nNAb) to the HIV-1 Env, the immune correlate of reduced risk of HIV infection in the RV144 trial, appear to be a more achievable goal.RV144, the only efficacy trial to date in which at least moderate protection against HIV acquisition was achieved (8, 9), was conducted between 2003 and 2009 in 16,402 volunteers in Thailand in the context of a HIV-1 subtype CRF01_AE dominated epidemic. A modified intention to treat analysis showed an estimated overall efficacy of 60.5% at 12 months after the first vaccination (10), which waned to 31.2% 3.5 years later (8). Binding IgG antibodies to specific linear epitopes of the HIV-1 Env variable regions 2 (V2) and 3 (V3) correlated inversely with HIV-1 infection in RV144, whereas neutralizing antibodies were not associated with a reduction in infection risk (9, 11, 12). Envelope sequence analyses of breakthrough infections confirmed the selective pressure of V2-specific antibody responses in the RV144 trial (13–15) and further studies showed a parallel decline of vaccine efficacy and the level of anti-V2 IgG responses over time (12, 16–18). During natural infection, antibodies against the highly variable V2 region are found in less than 50% of infected individuals (19, 20) while anti-V3 antibodies can be found in almost all naturally HIV-1 infected individuals and are elicited by most vaccination regimen tested so far (18, 21–27). The V3 sequence is the most conserved of all the variable Env regions (28) and important for the pathogenicity of the virus (29). The protective potential of V3-specific antibodies is further supported by the association of maternal anti-V3 nNAb with a reduced mother-to-child transmission (30, 31).Contrary to these promising findings on the role of nNAb, it has been hypothesized that their immunodominance, especially of V3-directed nNAb, is to blame for the great difficulty in inducing bNAb (32).The rationale behind this is that in the germinal centre B cells with high affinity to such immunodominant epitopes as V3, strongly activate and recruit T follicular helper cells (Tfh) and might have a selection advantage over bNAb B cell precursors with lower affinity (33). Particularly, if Tfh help is limited and nNAb and bNAb epitopes are in competition, V3-responses may outcompete the maturation of weaker-affinity binding antibodies, necessary for bNAb formation (32). Accordingly, there is an effort in vaccine development to remove or repress the highly immunogenic V3 epitope to eliminate such potential decoy effects, in hopes of inducing bNAb against the HIV-1 Env (34, 35). There are currently many ingenious vaccination strategies under investigation to guide antibody affinity maturation towards the development of bNAbs (1), a formidable challenge that will, despite initial success (36), not be achieved in the foreseeable future. Regions of putative viral vulnerability, including V2 and V3, should therefore continue to be regarded as key target regions for a protective HIV vaccine. The overall objective of this study therefore was i) to investigate how different prime boost vaccination regimen in multiple clinical vaccine trials influence the pattern of IgG Env recognition; ii) to investigate vaccine parameters influencing IgG targeting of HIV-1 Env V2 and V3 epitopes and their sequence variants; and iii) to understand whether strong induction of V3-specific IgG responses compromises the recognition of other antigenic regions. To this end, we systematically mapped HIV-1 Env IgG epitopes from multiple HIV vaccine studies (RV144, TaMoVac02, UKHVC Spoke003, X001, and RV172) to identify immunogens and their combinations for optimal induction of responses to (36) wards regions of putative viral vulnerability.