Immunity

Previews

What mAbs Tell Us about Shapes:Multiple Roads Lead to Rome

Galit Alter1 and Margaret E. Ackerman2,*1Ragon Institute of MGH, MIT, and Harvard, 149 13th Street, Charlestown, MA 02139, USA2Thayer School of Engineering, Dartmouth College, 14 Engineering Drive, Hanover, NH 03755, USA*Correspondence: margaret.e.ackerman@dartmouth.eduhttp://dx.doi.org/10.1016/j.immuni.2013.01.003

In this issue of Immunity, Liao et al. (2013) utilize monoclonal antibodies isolated from RV144 vaccinees togain key insights into the structural vulnerabilities of the V1-V2 loops of HIV gp120 protein and how theymightbe associated with vaccine efficacy.

The RV144 trial, which consisted of canar-

ypox viral vector (ALVAC) as the prime

and clade B and E gp120 (AIDSVAX B/E)

as the boost and which was conducted

in a low-risk Thai population, demon-

strated modest efficacy (31.2%) (Rerks-

Ngarm et al., 2009). However, it remains

a source of controversy largely because

the vaccine did not induce the broadly

neutralizing antibodies or strong T cell

responses thought to be necessary for

protection. Instead, initial studies gener-

ated the hypothesis that nonneutralizing

antibodies might have provided pro-

tection. Moreover, correlates analysis

(Haynes et al., 2012a) pointed to an asso-

ciation between antibodies against the

V1-V2 loops of gp120 and decreased

risk of infection, further supported by viral

sieve analyses, in which vaccination

altered the likelihood of infection with

specific viral variants within the V2 loop

(Rolland et al., 2012).

Liao et al. demonstrate that vaccine-

induced antibodies can recognize the

V1-V2 loop in dramatically different

conformations (Figure 1), providing data

that pertain to three widely debated ques-

tions in HIV-vaccine research: What can

we learn from RV144? What can we

learn about vaccine-mediated protection

from studies of monoclonal antibodies

(mAbs)? Do vaccines targeting variable

regions of HIV, such as the V2 loop, hold

promise?

Despite conventional wisdom that vari-

able regions make poor vaccine targets,

there is a substantial body of literature

describing the critical nature of the V2

loop (Zolla-Pazner and Cardozo, 2010).

Importantly, transmitted viral variants

exhibit shorter V1-V2 loops containing

fewer N-glycosylation sites (Sagar et al.,

8 Immunity 38, January 24, 2013 ª2013 Elsev

2006), reflecting an advantage of short

loops in early infection and subsequent

elongation and glycosylation driven by

viral evasion of intense immune pressure

mediated by autologous antibodies

(Rong et al., 2009; Sagar et al., 2006).

Furthermore, V2 antibodies display immu-

nologic cross-reactivity, suggesting that

the V2 region contains conserved three-

dimensional elements (Zolla-Pazner and

Cardozo, 2010).

To probe the characteristics of RV144-

vaccine-induced antibodies, Liao et al.

cloned and crystallized V2-specific

mAbs and linked these mAbs to popula-

tion analyses to determine the profile

and the predominance of particular

patterns of reactivity among vaccinees.

Together, these studies offer critical

insights into loop-recognition patterns

associated with vaccine protection but

also raise questions related to the mecha-

nism by which vaccines might elicit anti-

bodies that target the most vulnerable,

common, or relevant conformations of

flexible and sequence-variable loops.

Despite the sequence variability in the

V2 loop, V2 antibodies can be broadly

cross-reactive and can recognize V2

epitopes in different conformations. As

salient examples, CH58 and CH59,

cloned by Liao et al., bind to linear V2

peptides and therebymark a class of anti-

bodies that react primarily with peptides,

distinct from previously defined confor-

mation-specific 697D-like mAbs and the

glycopeptide, quaternary-specific PG9-

like mAbs. According to their crystal

structures, CH58, CH59, and PG9 recog-

nize overlapping V2 epitopes in remark-

ably different conformations, ranging

from helical to beta strand, begging the

question as to whether the protective

ier Inc.

breadth of these V2-specific antibodies

might relate to the ability of V2 to adopt

these different conformations in the con-

text of different envelopes and whether

particular V2-specific antibodies might

lock the V2 loop into conformations that

inhibit or mark the virus more effectively.

Interestingly, CH58 and CH59 interact

differentially with prime and boost enve-

lopes despite identical V1-V2 sequences

between positions 165 and 184, further

supporting the notion that conformation

plays a prominent role in determining

V2-loop immunogenicity, antigenicity,

and possibly sensitivity. Alternatively,

differential recognition of prime and boost

envelopes might also be due to altered

presentation. Conformation and flexibility

differences due to in vivo production and

cell-surface-tethered exposure might

have allowed the prime antigen to adopt

specific shapes, permitting the induction

of CH58- and CH59-like antibodies.

Investigating whether similar antibodies

are inducible by protein vaccination

would help define whether V2 antibodies

are preferentially elicited in different con-

texts. Because sequence alone does not

provide a clear indication of envelope

vulnerabilities, Liao et al. evoke the possi-

bility that understanding V2 shape and

conformational plasticity might provide

critical information for the rational design

of promising immunogens.

The observation that some HIV-specific

antibodies exhibit limited germline diver-

sity, implying that only a restricted set of

suitable naive variable gene segments

exist, has motivated the design of vaccine

regimens that drive B cell evolution along

certain lineages (Haynes et al., 2012b). In-

deed, previous studies of conformation-

dependent V2-specific mAbs suggested

Figure 1. Multiple Roads to Recognition ofthe V2 Loop on the HIV gp120 ProteinRemarkably diverse conformations of V2 loop(indicated by colored shapes) can be recognizedby antibodies. Understanding whether the mostvulnerable, common, or relevant conformations ofsequence-variable loops represent themost prom-ising immunogensmight be an important avenue ofinvestigation in rational vaccine design.

Immunity

Previews

that they arise predominantly from the

VH1-69 family (Gorny et al., 2012).

However, perhaps unsurprisingly given

their recognition of dramatically different

V2-peptide conformations, CH58 and

CH59 descended from different VH fami-

lies. Given the diversity of germlines

used and the divergent structures

assumed by V2 peptides, these data are

consistent with the notion that there are

many paths to V2 recognition (Figure 1).

Moreover, the ability of the CH58 and

CH59 unmutated ancestors (UAs) to

bind to the same envelopes from two

different clades, in conjunction with the

remarkably disparate reactivity of UAs of

other V2 antibodies (PG9 and 697D),

again supports the argument that in the

case of V2-reactive antibodies, limited

germlines do not appear to dominate,

and understanding structural features of

V2 variants might leverage the germline

repertoire more effectively.

Therefore, despite the controversy

associated with RV144, the substantial

data implicating the importance of the

variable loops in transmission, their

involvement in immune evasion, their tar-

geting by highly cross-reactive antibodies

that reduce infection (Karasavvas et al.,

2012), and their ability to mediate bio-

logic functions including neutralization

and antibody-dependent cell-mediated

cytotoxicity (Gorny et al., 2012), provide

a strong impetus for defining the critical

nature of this viral determinant in antiviral

immunity. Consequently, studies such as

that of Liao et al., using vaccine-induced

mAbs to map structural features of this

vulnerable variable loop, become more

powerful and relevant to vaccination

when paired with broader population

analyses in that they provide a measure

of confidence that eliciting similar anti-

bodies in different subjects is feasible.

Continued study of V2 antibodies and

others with restricted but definitively

protective activity might lead to the

understanding of the structural features

that prompt strain-specific immunity.

Such studies could indicate whether

these protective antibodies are reactive

with conformations shared by most

strains or with immunologically-defined

subtypes of HIV-1, requiring a finite

number of vaccine constructs, analogous

to polio and influenza vaccines. Accord-

ingly, Liao et al., offer hope, but they

also point to significant hurdles encoun-

Immunity

tered in targeting a sequence- and struc-

turally-variable domain. The biggest

impact of this and similar studies might

be that sequence diversity and limited

crystal structures do not prevent a region

from being a valuable target for a protec-

tive vaccine.

REFERENCES

Gorny,M.K., Pan, R.,Williams, C.,Wang, X.H., Vol-sky, B., O’Neal, T., Spurrier, B., Sampson, J.M., Li,L., Seaman, M.S., et al. (2012). Virology 427,198–207.

Haynes, B.F., Gilbert, P.B., McElrath, M.J., Zolla-Pazner, S., Tomaras, G.D., Alam, S.M., Evans,D.T., Montefiori, D.C., Karnasuta, C., Sutthent,R., et al. (2012a). N. Engl. J. Med. 366, 1275–1286.

Haynes, B.F., Kelsoe, G., Harrison, S.C., and Kep-ler, T.B. (2012b). Biotechnol. 30, 423–433.

Karasavvas, N., Billings, E., Rao, M., Williams, C.,Zolla-Pazner, S., Bailer, R.T., Koup, R.A., Mad-note, S., Arworn, D., Shen, X., et al.; MOPH TAVEGCollaboration. (2012). AIDS Res. Hum. Retrovi-ruses 28, 1444–1457.

Liao, H.-X., Bonsignori, M., Alam, S.M., McLellan,J.S., Tomaras, G.D., Moody, M.A., Kozink, D.M.,Hwang, K.-K., Chen, X., Tsao, C.-Y., et al. (2013).Immunity 38, this issue, 176–186.

Rerks-Ngarm, S., Pitisuttithum, P., Nitayaphan, S.,Kaewkungwal, J., Chiu, J., Paris, R., Premsri, N.,Namwat, C., de Souza, M., Adams, E., et al.;MOPH-TAVEG Investigators. (2009). N. Engl. J.Med. 361, 2209–2220.

Rolland, M., Edlefsen, P.T., Larsen, B.B., Tovana-butra, S., Sanders-Buell, E., Hertz, T., deCamp,A.C., Carrico, C., Menis, S., Magaret, C.A., et al.(2012). Nature 490, 417–420.

Rong, R., Li, B., Lynch, R.M., Haaland, R.E.,Murphy, M.K., Mulenga, J., Allen, S.A., Pinter, A.,Shaw, G.M., Hunter, E., et al. (2009). PLoS Pathog.5, e1000594.

Sagar, M., Wu, X., Lee, S., and Overbaugh, J.(2006). J. Virol. 80, 9586–9598.

Zolla-Pazner, S., and Cardozo, T. (2010). Nat. Rev.Immunol. 10, 527–535.

38, January 24, 2013 ª2013 Elsevier Inc. 9