Leading Edge
Previews
Teaching a Clone to Walk, One Step at a Time
James E. Crowe, Jr.1,*
(^1) Departments of Pediatrics, Pathology, Microbiology, and Immunology and Vanderbilt Vaccine Center, Vanderbilt University Medical Center,
Nashville, TN 37232, USA
*Correspondence:[email protected]
http://dx.doi.org/10.1016/j.cell.2016.08.
Development of broad HIV neutralizing antibodies during infection requires prolonged exposure
to viral variants. A series of three studies published in this issue ofCelland a related paper inIm-
munityreport new HIV structure-based design efforts focused on sequential boosting regimens
with antigens of increasing maturation that point the way forward for an effective HIV vaccine
strategy.
The long road to development of HIV
broadly neutralizing antibodies (bNAbs)
in the setting of natural HIV infection
typically takes years and results from
continuous and recurrent exposure to a
never-ending swarm of viral quasispe-
cies. One candidate vaccine strategy to
mimic this process is to identify the
most antigenically distinct of prominent
nodal points in the phylogeny of evolving
sequences of HIV envelope (Env) pro-
teins and to express and immunize
sequentially with a limited series of
increasingly mature Env immunogens in
order to engage and evolve B cell clones
that express highly mutated bNAbs.
A series of three studies published in
this issue ofCell(Briney et al., 2016;
Escolano et al., 2016; Tian et al., 2016)
and a related paper inImmunity(Stei-
chen et al., 2016) show that the
challenging sequential combination of
germline targeting, followed by a second
step of mutating low-affinity clones for-
ward toward the bNAb path, may be
possible through careful and detailed
iterative design and testing cycles.
The organizing principle is that one
must jumpstart the process of bNAb
development with ‘‘germline targeting’’
primary immunization, then push the evo-
lution of the initiated clones by stepwise
coaxing using intermediate heteroge-
neous boosts that are structurally and
antigenically different enough to re-immu-
nize and induce somatic hypermutation of
antibody genes. But where to find the
intermediate antigens that do the job?
The solutions offered here required the
use of a full panoply of fashion-forward
tools, including computational design us-
ing structural supercomputing methods,
deep sequencing of antibody and virus
genes, libraries for expression and display
of very diverse antigens, and other high
content methods.
Schief and colleagues (Briney et al.,
2016 ) present progress on a long-term
project to develop a multicomponent HIV
vaccine that uses a stepwise process of
inducing increasingly mutated human an-
tibodies to the HIV CD4 binding site. The
investigators combine sequential use of
a previously defined germline targeting
antigen at the beginning of the scheme
and finish with native-like HIV Env trimers.
The innovation here is that they are able to
add the design and testing of intermediate
step boosting antigens, discovered prin-
cipally by yeast display methods. In
related work inImmunity(Steichen et al.,
2016 ), they use mammalian display stra-
tegies to design constructs intended to
induce antibodies to another major
neutralizing determinant on the HIV Env
V3-loop and surrounding glycans, using
design based on the predicted germline
version of a broadly neutralizing antibody
to that region termed PGT121. These in-
vestigators show that the designed tri-
mers can activate B cells displaying the
inferred-germline B cells for PGT
when presented as multimers on lipo-
somes and that they stimulate responses
in mice with the inferred ancestor genes
for PGT121 knocked in.
Nussenzweig and colleagues (Escolano
et al., 2016) make the next leap forward,
applying the designs developed byStei-
chen et al., 2016to test sequential immu-
nization strategies in vivo. They start with
an antigen designed to engage germline
antibody expressing B cells and then
combine it with a sequence of less modi-
fied antigens that were triaged for their
immunogenicity in sequence, based on
response in ELISA. They find that se-
quences of highly modified followed by
less modified immunogens are able
to achieve induction of heterologous
neutralizing responses, even to some
viruses that are more difficult to inhibit
(tier 2 viruses). Importantly, the neutral-
izing antibodies generated by sequential
immunization present somatic mutations
with patterns that resemble PGT
mature clones. In other words, the
sequential immunization studies did not
only work, but they worked in the intended
manner inducing somatic mutations char-
acteristic of real broadly neutralizing
antibodies. This research shows a sur-
prisingly high level of control over manip-
ulation of antibody evolution in a directed
fashion.
These papers also address another
central challenge in the HIV design field,
and that is how to test large numbers of
experimental HIV vaccine candidates to
see if they can initiate or sustain B cell
clonal lineage development in an HIV-
naive human immune receptor repertoire.
Ideal animal models for testing human
B cell lineage development following HIV
vaccination are lacking. Small animals
such as rodents do not replicate virus
or make comparable immune responses
to those of humans, and rabbits, guinea
pigs, and nonhuman primates use dif-
ferent B cell genetics or receptor diversifi-
cation mechanisms. These conventional
animal models are not appropriate pre-
clinical models to test the feasibility of
‘‘germline targeting’’ and intermediate
antigen-based strategies for human
immunogenicity.
1360 Cell 166 , September 8, 2016ª2016 Elsevier Inc.