Snapshot: Broadly Neutralizing Antibodies Devin Sok, Brian Moldt, and Dennis R

SnapShot: Broadly Neutralizing Antibodies Devin Sok, Brian Moldt, and Dennis R. Burton The Scripps Research Institute, La Jolla, CA 92037, USA

> Immunodominant conserved epitope: predominantly broadly neutralizing antibody response

> Immunodominant variable epitope: predominantly strain-speciﬁc antibody response IMMUNODOMINANCE

Measles hemagglutinin Measles bnAbs Variable regions Durably elicited by live attenuated virus, less durably by inactivated virus Conserved bnAb epitopes: Receptor-binding site 1) Ep1 (res. 380–400) 2) Ep2 (res. 190–200) > 3) Ep3 (res. 571–579) Despite the extensive variability of measles virus, nAb responses to vaccination are broad because Conserved epitopes the conserved epitopes are immunoprominent. targeted by bnAbs May relate to overlap of the receptor-binding site and the bnAb epitopes. IMMUNODOMINANT CONSERVED EPITOPE

HIV Env HIV-1 bnAbs Variable loop regions targeted by strain- Isolated only from a subset of chronically infected speci c nAbs individuals after several years of infection, have not been elicited by vaccination to date Conserved epitopes Conserved bnAb epitopes: targeted by bnAbs 1) CD4-binding site (CD4bs) > 2) Conserved region of V3/V4 loops-glycan 3) Conserved region of V2 loop-glycan Shielding glycans 4) Membrane-proximal external region (MPER) Variable loops are immunodominant strain-speci c nAb epitopes. Many nonneutralizing Abs are elicited to Env subunits (gp120 and gp41) in con gurations not found on the functional Env spike.

Inﬂuenza hemagglutinin Inﬂuenza bnAbs Variable HA1 head region targeted by strain- Isolated from infected and vaccinated individuals speci c nAbs Conserved bnAb epitopes: 1) HA2 stem region Conserved epitopes in 2) HA1 sialic acid receptor-binding site HA2 stem region and > HA1 sialic acid receptor- bnAbs show varying degrees of coverage: binding site individual subtypes, group 1 viruses, group 2 viruses, inuenza A and B viruses. Immunodominant nAb epitopes are highly variable regions of the HA1 head region.

IMMUNODOMINANT VARIABLE EPITOPES IMMUNODOMINANT VARIABLE Shielding glycans

Malaria (Plasmodium falciparum)

DBL1 Variable epitopes Malaria bnAbs CIDR1 Not yet identi ed but of potentially great value DBL2 pfEMP1 is a possible candidate for a blood-phase DBL3 ? vaccine. It is expressed on the surface of infected pfEMP1 erythrocytes and is highly variable, but conserved DBL4 Potential sites are being identi ed. CSP, MSP, and AMA 1 conserved are other potential protein targets. DBL5 epitopes TM

728 Cell 155, October 24, 2013 ©2013 Elsevier Inc. DOI http://dx.doi.org/10.1016/j.cell.2013.10.009 See online version for legend and references. SnapShot: Broadly Neutralizing Antibodies Devin Sok, Brian Moldt, and Dennis R. Burton The Scripps Research Institute, La Jolla, CA 92037, USA

Neutralization is defined as the in vitro ability of an antibody to inhibit the entry of a virus into a target cell in the absence of other effector proteins or cells. The presence of neutralizing antibodies in the serum of a vaccinated individual has been shown to be one of the best correlates for the efficacy of most licensed vaccines (Plotkin, 2010). Some viruses show considerable sequence variation in their surface proteins, which are targeted by neutralizing antibodies to the extent that immunization elicits strain-specific neutralizing antibodies. These antibodies will only be able to provide protection against a subset of circulating viruses. However, an ideal vaccine should induce antibodies that are able to neutralize the majority or even all circulating viruses; such antibodies are termed broadly neutralizing antibodies (bnAbs). Though they vary in characteristics and features, bnAbs universally function by targeting epitopes that are highly conserved and exposed on the surface proteins of the variable virus. Conserved epitopes have typically been targeted on variants of the same viral species but recently have been described on several different species of paraxomyxovirus (Corti et al., 2013).

Immunodominance/Immunoprominence/Immunoquiescence The immunogenicity of the conserved, exposed epitopes on the surface of the virus will determine the extent to which bnAbs will be induced through vaccination or infection (Burton, 2002). Viruses that are highly variable but nevertheless express immunodominant or immunoprominent conserved epitopes are expected to elicit bnAb responses. In contrast, highly variable viruses with immunodominant variable epitopes and relatively immunoquiescent conserved epitopes will more likely elicit strain-specific neutralizing antibody responses and weak or no bnAb responses.

Viruses with Immunodominant/Immunoprominent Conserved Epitopes Measles virus is an example of a virus whose surface glycoprotein, measles hemagglutinin, shows considerable sequence variability but for which a highly successful vaccine has been developed (Bellini and Rota, 2011). It appears that an immunoprominent conserved epitope overlaps with its receptor-binding site and induces a robust neutralizing antibody response that protects against all known viral genotypes, with some monoclonal antibodies isolated recently able to neutralize a viral strain from 1954 (Tahara et al., 2013). Although not typically described as such, the protective antibodies induced by vaccination are therefore broadly neutralizing.

Viruses with Immunoquiescent Conserved Epitopes Although viruses in this category typically elicit weak bnAb responses through vaccination, some individuals do develop strong responses, particularly through natural infection over longer time periods (Corti and Lanzavecchia, 2013). These individuals can give rise to bnAbs that, in turn, can help to define the conserved neutralizing epitopes on viral surface proteins, thereby opening avenues to rational vaccine design for highly antigenically variable viruses (Burton et al., 2012). HIV The HIV envelope spike (Env) is the sole target for neutralizing antibodies and contains several conserved epitopes that can be targeted by bnAbs (Kwong and Mascola, 2012). On the functional Env spike, both variable neutralizing epitopes and broadly neutralizing epitopes are exposed. The latter, however, tend to have low immunogenicity. This is either because they have strict steric restrictions on antibody recognition (the CD4-binding site, the membrane proximal external region) or because they are composed, at least in part, of glycans (variable loop/glycan epitopes). Generally, the very high density of N-linked glycans on HIV Env is a major barrier to antibody recognition and elicitation. The bnAbs that have been identified appear to overcome these barriers by incorporating unusual features, including very high levels of somatic hypermutation (up to 44% [amino acid]), high frequency of deletions and insertions (up to 9 amino acids in length), and long CDR3s (up to 34 amino acids in length) (Corti and Lanzavecchia, 2013). Also of note, the native trimer is metastable and readily dissociates into gp120 and gp41 subunits, revealing strain-specific neutralizing epitopes that tend to be immunoprominent. Influenza Virus The target for neutralizing antibodies is the surface protein hemagglutinin, which contains a head (HA1) and a stem (largely HA2) region (Ekiert and Wilson, 2012). A seasonal vaccine against the predicted dominant strain of influenza is developed each year, but the high level of sequence diversity between strains has proven challenging for the design of a universal vaccine, which ideally would be given only once in a lifetime. The variability of the hemagglutinin protein is mainly found in the head region, and neutralizing antibodies to this immunodominant region are typically strain specific. The stem region is highly conserved compared to the head region, and most of the known broadly bnAbs recognize this region, although a subset are capable of recognizing the highly conserved sialic-acid-binding site in the head region. The frequency of developing bnAbs that target the stem is rare perhaps because of the tight packing of hemagglutinin proteins on the virus surface, which may reduce the accessibility of antibodies to broadly neutralizing epitopes in this region (Corti and Lanzavecchia, 2013). Compared to HIV-1 bnAbs, influenza bnAbs have shorter CDR3 lengths (up to 26 amino acids in length) and lower mutation frequency (up to 18% [amino acid]) and rarely have insertions or deletions. Malaria (Plasmodium falciparum) Although a parasite and not a virus, malaria does show great antigenic variability that make it worthy of consideration here, although bnAbs have not yet been identified. Malarial diversity is represented by the number of possible antigens present through the course of the lifecycle and the genetic variability of each antigen (Thera and Plowe, 2012). Though a number of different approaches are being considered, vaccines can be generated for specific stages of the lifecycle. In the pre-erythrocytic phase, the circumsporozoite protein (CSP) is a relatively conserved target but has been found to be poorly immunogenic. In the blood stage, MSP-1, AMA1, and PfEMP1 have been considered as potential candidates. These latter targets are much more antigenically diverse, but recent studies are beginning to identify conserved regions that could serve as targets for bnAbs (Thera and Plowe, 2012).

Acknowledgments

The authors acknowledge the financial support of the National Institute of Allergy and Infectious Diseases; the Center for HIV/AIDS Vaccine Immunology and Immunogen Discov- ery; the International AIDS Vaccine Initiative; the Ragon Institute of MGH, MIT, and Harvard; and the Lundbeck Foundation.

References

Bellini, W.J., and Rota, P.A. (2011). Virus Res. 162, 72–79.

Burton, D.R. (2002). Nat. Rev. Immunol. 2, 706–713.

Burton, D.R., Poignard, P., Stanfield, R.L., and Wilson, I.A. (2012). Science 337, 183–186.

Corti, D., and Lanzavecchia, A. (2013). Annu. Rev. Immunol. 31, 705–742.

Corti, D., Bianchi, S., Vanzetta, F., Minola, A., Perez, L., Agatic, G., Guarino, B., Silacci, C., Marcandalli, J., Marsland, B.J., et al. (2013). Nature 501, 439–443.

Ekiert, D.C., and Wilson, I.A. (2012). Curr. Opin. Virol. 2, 134–141.

Kwong, P.D., and Mascola, J.R. (2012). Immunity 37, 412–425.

Plotkin, S.A. (2010). Clin. Vaccine Immunol. 17, 1055–1065.

Tahara, M., Ito, Y., Brindley, M.A., Ma, X., He, J., Xu, S., Fukuhara, H., Sakai, K., Komase, K., Rota, P.A., et al. (2013). J. Virol. 87, 666–675.

Thera, M.A., and Plowe, C.V. (2012). Annu. Rev. Med. 63, 345–357.