In a finding that highlights ways to optimize the efficacy of antibody-based therapeutics and vaccines, the activity of potent HIV-1-neutralizing antibodies has been confirmed to depend on cellular binding to the antibodies' Fc regions.
Protection conferred against virus infection by licensed antiviral vaccines is mostly due to the generation of 'neutralizing' antibodies that bind to virus particles in such a way that they block their entry into a cell1. Current strategies for designing treatments and vaccines for HIV-1 focus on such antibodies, particularly on the highly potent, broadly neutralizing antibodies that are produced during the immune responses of rare infected individuals2,3 and which bind to the envelope protein (a highly variable protein) of multiple strains of the virus. However, it is becoming clear that the impact of these antibodies is not limited to direct viral neutralization through binding. Writing in Cell, Bournazos et al.4 show that in vivo activity of even the most potent broadly neutralizing anti-HIV-1 antibodies relies on antiviral functions that are triggered by binding of cellular receptors — the Fcγ receptors — to the 'tail' region of antibodies of the immunoglobulin G class, binding that was previously found5 to be crucial for sustaining the in vivo activity of less-potent neutralizing antibodies.
Despite showing promising neutralizing activity in vitro, attempts to use broadly neutralizing antibodies to control HIV-1 infections in vivo have had limited success. Up to now, only the most potent neutralizing antibodies, used at high dose and often only in combination, have yielded measurable control of HIV-1, and this effect rapidly declines with waning antibody levels and as a result of mutations that allow the virus to 'escape' antibody binding6,7,8,9. Difficulties with tissue penetration10 and reduced neutralizing activity during cell-to-cell viral transmission11 may also contribute to the lower efficacy of these antibodies in vivo and necessitate the maintenance of antibody levels orders of magnitude above those sufficient in vitro6,7,8,9.
Antibodies are notable for their capacity to link the adaptive and innate arms of the immune response: the variable domains of an antibody specifically recognize different targets, and their constant tail domain, the Fc region, stimulates an array of immune defences. The Fc region is bound by proteins of the complement system, which activate signalling pathways to induce inflammatory responses and destruction of antibody-bound cells or viruses. The Fc region of immunoglobulin G is also recognized by activating and inhibitory Fcγ receptors (FcγRs) on effector cells of the immune system, such as monocytes, macrophages, dendritic cells, neutrophils and natural killer cells. Fc binding provides these cells with essential stimulatory and regulatory signals12. FcγR-mediated effector-cell functions include antibody-dependent cellular cytotoxicity (in which antibodies bind to viral proteins expressed on the infected cell, triggering its killing by effector cells), phagocytic clearance (cellular engulfment and destruction of viruses), and the release of soluble antimicrobial and immunomodulatory factors, such as cytokines and chemokines (Fig. 1).
Recruitment of effector functions has long been a focus in the harnessing of antibody-based protection against HIV-14,5,12,13,14, but the precise contributions of direct antibody-mediated virus neutralization, activity of the complement system and FcγR-mediated effector functions remain unresolved13. Pioneering work indicated that neutralizing antibodies require FcγR functions for their in vivo activity5, whereas non-neutralizing antibodies that act solely through effector functions have shown limited in vivo activity against HIV-113,15, suggesting that a combination of both neutralization and effector-mediated activity is needed. By contrast, complement action seems dispensable, at least when neutralizing antibodies are used as a pre-exposure prophylaxis (to try to prevent infection)5. Now, Bournazos et al. expand on the previous finding that FcγR-dependent mechanisms are required5 by investigating the effect of antibody treatment on HIV-1 infection in mice in the presence or absence of Fc–FcγR interactions, either by modulating the antibodies' Fc region or by using mice lacking FcγRs.
Particularly notable is the authors' finding that, despite their superior potency, even the action of broadly neutralizing antibodies is largely dependent on interactions with activating FcγRs. When the researchers engineered the antibodies to improve the strength of Fc–FcγR binding, they observed increased viral control in vivo, highlighting the potential of antibody improvement for therapeutic use. On reflection, however, this dependence on FcγR functions might prove to be a crucial limitation on the use of engineered molecules designed to inhibit HIV entry to cells by targeting viral envelope proteins, because such molecules lack the ability to recruit immune effector functions.
Intriguingly, in experiments using HIV particles that can complete only one round of infection, such that newly infected cells do not express HIV envelope proteins and hence lack binding sites for antibodies, Bournazos et al. still saw FcγR-dependent antibody-mediated viral control. This suggests that classical antibody-dependent cellular cytotoxicity can be ruled out as a sole driving force, and that effector-cell release of soluble antiviral factors and the phagocytic clearance of viral particles may be decisive.
Nevertheless, a central question that remains is, why would a potent neutralizing antibody need to rely on fast, FcγR-mediated removal of antibody-bound viral particles? There are several possible explanations. Neutralization can be reversible if the antibody binding has a high off-rate16,17,18, and if an antibody fails to irreversibly neutralize the virus, it may be crucial to eliminate these viruses rapidly before they regain infectivity. Tissue penetration may be another limiting factor10: lower antibody doses at certain sites may mean that not all envelope proteins on the virus are immediately bound and neutralized.
How many envelope proteins HIV carries, how many of these are needed to infect (and in turn must be neutralized), and how many antibody molecules are needed to trigger irreversible neutralization of envelope proteins will all also contribute to the rate of virus inactivation19. Alongside these stoichiometric requirements, the kinetics of neutralization will be steered by the on-rate of antibody binding. Considering all these factors, enhanced clearance of antibody-bound viral particles may become relevant in situations in which the virus is not fully neutralized, either because the threshold of neutralizing-antibody binding required for inactivation has not been reached, or because the specific neutralizing antibody fails to irreversibly block the virus. In support of the idea that quantity of antibody, and hence antibody occupancy, plays a part, Bournazos and colleagues observed that eliciting FcγR effector functions had an additive effect on viral control at lower, but not higher, antibody doses.
Further work will be needed to precisely quantify the impact of phagocytic clearance, and to define if only one or a combination of FcγR-mediated functions is needed to achieve in vivo control of HIV-1 by neutralizing antibodies. Several technical challenges may arise in attempts to investigate this. HIV-1 infection is commonly monitored by quantifying levels of viral RNA, but this approach does not assess the infectivity of the viral particles present, and so both neutralized and non-neutralized virions will be counted. Thus, short-term experiments in which the effects of neutralizing antibodies are assessed solely on the basis of a reduction in RNA levels will not be able to quantify the contribution of Fc-mediated effects. Long-term monitoring of infections, as performed by Bournazos et al.4 and in previous work5, in combination with measurements of circulating infectious virus and infected cells, will be key to quantifying the influence of FcγR-dependent mechanisms.
Koff, W. C. et al. Science 340, 1232910 (2013).
Burton, D. R. et al. Cell Host Microbe 12, 396–407 (2012).
Mascola, J. R. & Montefiori, D. C. Annu. Rev. Immunol. 28, 413–444 (2010).
Bournazos, S. et al. Cell 158, 1243–1253 (2014).
Hessell, A. J. et al. Nature 449, 101–104 (2007).
Barouch, D. H. et al. Nature 503, 224–228 (2013).
Klein, F. et al. Nature 492, 118–122 (2012).
Trkola, A. et al. Nature Med. 11, 615–622 (2005).
Poignard, P. et al. Immunity 10, 431–438 (1999).
Ko, S.-Y. et al. Nature http://dx.doi.org/10.1038/nature13612 (2014).
Abela, I. A. et al. PLoS Pathog. 8, e1002634 (2012).
Ackerman, M. E., Dugast, A. S. & Alter, G. Annu. Rev. Med. 63, 113–130 (2012).
Ackerman, M. E. & Alter, G. Curr. HIV Res. 11, 365–377 (2013).
Huber, M. et al. PLoS Med. 3, e441 (2006).
Burton, D. R. et al. Proc. Natl Acad. Sci. USA 108, 11181–11186 (2011).
Platt, E. J., Gomes, M. M. & Kabat, D. Proc. Natl Acad. Sci. USA 109, 7829–7834 (2012).
Ruprecht, C. R. et al. J. Exp. Med. 208, 439–454 (2011).
Yasmeen, A. et al. Retrovirology 11, 41 (2014).
Magnus, C., Rusert, P., Bonhoeffer, S., Trkola, A. & Regoes, R. R. J. Virol. 83, 1523–1531 (2009).
About this article
PLOS Pathogens (2019)
HIV-1 resistance to neutralizing antibodies: Determination of antibody concentrations leading to escape mutant evolution
Virus Research (2016)
Capacity of Broadly Neutralizing Antibodies to Inhibit HIV-1 Cell-Cell Transmission Is Strain- and Epitope-Dependent
PLOS Pathogens (2015)