INSIGHTS | PERSPECTIVES
antibodies to prevent HIV-1
By Myron S. Cohen1 and Lawrence Corey2
Advances in technology—especially single-cell antibody cloning tech- niques—have led to the isolation and characterization of antibodies from people with HIV infection that can eutralize many variants (1). These
are referred to as broadly neutralizing antibodies (bnAbs). Such antibodies can be
detected in about 25% of persons with untreated HIV-1 infection (2), reflecting a host
immune response to unremitting viral replication, generation of large numbers of viral
variants, and shifting antigen exposure ( 3).
Although bnAbs may exert some selective
pressure as they develop, they generally do
not reduce viral burden, improve health, or
slow the progression of disease ( 4). However, they offer considerable
opportunities for treatment
and prevention of HIV-1
infection in others. At this
time, hundreds of bnAbs
have been identified; those
that have attracted the most
attention are bnAbs with the
greatest breadth, neutralizing the largest number of
HIV-1 strains, including those traditionally
most neutralization resistant (1, 5); or bnAbs
that have the greatest potency, requiring the
smallest concentration to neutralize resistant strains of HIV-1 ( 5–7). A study by Xu et
al. ( 5) on page 85 of this issue and by Julg
et al. ( 7) in Science Translational Medicine
illustrate advances in the potential use of
bnAbs to prevent HIV-1 infection.
The key purpose of the initial search to
detect and understand HIV-1 bnAbs was to
identify HIV-1 envelope proteins that might
be targeted by an effective HIV-1 vaccine ( 8),
because induction of antibodies to levels as-
sociated with protection from an infectious
agent has been the most effective approach
for vaccine development ( 4). However, can-
didate HIV-1 vaccines designed to generate
bnAbs have been confounded by the diffi-
culty of inducing human germline cells that
produce such antibodies, the high degree of
somatic mutation required for the evolution
of most bnAbs, and the potential for auto-
immunity—and hence elimination—of such
antibodies early in the ontogeny of B cell
and antibody development ( 9).
As an alternative, HIV-1 bnAbs can be
manufactured and administered through
intermittent blood infusions, providing both
circulating and mucosal concentrations
of antibodies at levels that might be able
to block HIV-1 acquisition (genitourinary
and rectal mucosal epithelia are the sites
of HIV-1 acquisition) ( 4). The bnAbs isolated so far target proteins expressed on the
HIV-1 envelope ( 8): the critical CD4 binding
site (which HIV-1 uses, via
its glycoprotein gp120, to
enter cells); glycan-coated
viral loops, including the
V1-V2 glycan and V3 glycan; and a conserved viral
membrane-proximal external region (MPER) (see the
figure). bnAbs directed at
these targets reduce simian
HIV (SHIV) replication in nonhuman primates (NHPs) ( 7) and HIV-1 in humans ( 10),
and can prevent SHIV in NHPs ( 5, 7). However, after treatment of an infected NHP or
person with a single bnAb, rebound viremia
from outgrowth of low-frequency resistant
strains has been observed ( 7, 10), providing a potential barrier to the use of bnAbs
for long-term therapy of people with HIV-1
infection. Outgrowth of resistant strains occurs less frequently when a bnAb is administered prior to experimental challenge for
prevention from infection, rather than for
treatment of existing infection. In the latter experimental situation, passive administration of bnAbs can produce complete
mucosal protection even at low serum concentrations ( 4, 5, 7). However, a weakness
of the NHP models is the homogeneity of
the SHIV inoculum used, because humans
are generally exposed to a large and diverse
HIV-1 viral “swarm” (i.e., a mixture of viral
variants) ( 4).
1The Division of Infectious Diseases and the Institute for
Global Health and Infectious Diseases, The University of North
Carolina (UNC) at Chapel Hill, Chapel Hill, NC, USA. 2Vaccine
and Infectious Disease Division, Fred Hutchinson Cancer
Research Center, Seattle, WA, USA. Email: firstname.lastname@example.org.
amyloid fibrils are pathogenic, functional
fibrils have been identified as well ( 9).
The native fold, sequence, and function of
amyloid fibril-forming proteins are not obviously related. Thus, the existence of an
amyloid state cannot readily be predicted
( 8). The availability of two fibril structures
at high resolution is a first step toward a
general understanding of amyloid fibrils
and their formation. However, additional
structures, especially of different polymorphs, are needed to understand and predict protein behavior.
Gremer et al. identify hydrophobic clusters and probable salt-bridges that stabilize
both the conformation of each subunit and
the fibril along its axis. The knowledge of
those interactions enables analysis of familial mutations and helps to determine
how protective and disease-causing mutations relate to the structure. The authors
pinpoint sites of pathogenic and protective
familial mutations inside the Ab(1– 42) fibril
structure and explain their impact based on
the altered interactions. This is an excellent
example of how protein structures help to
explain the molecular basis of familial mutations, laying the foundation for the directed
tailoring of pharmaceuticals.
During AD progression, fibril accumulation expands approximately exponentially
due to the fragmentation of existing fibrils
and subsequent prion-like seeding events
( 3). Thus, tackling the emergence of fibrils is
a promising therapeutic approach that can
probably be extended to a wide range of neurodegenerative diseases, provided that fibril
accumulation is a major cause for pathogenicity, which is still under debate ( 10).
The structural studies of Ab(1– 42) ( 6)
and tau fibrils ( 7) mark substantial progress toward understanding the molecular
basis of AD. They also establish cryo-EM
as a powerful method for studying amyloid fibrils. Solving additional structures of
different polymorphs and fibrils obtained
from confined areas of human AD brains
will help to characterize distinct and common interactions within fibril structures
and most likely facilitate characterization
of disease progression. j
1. A. Kumar, A. Singh, Ekavali, Pharmacol. Rep. 67, 195
2. A. Serrano-Pozo, M. P. Frosch, E. Masliah, B. T. Hyman, Cold
Spring Harb. Perspect. Med. 1, a006189 (2011).
3. M.Jucker,L.C. Walker,Nature 501, 45(2013).
4. L. Gremer et al ., Science 358, 116 (2017).
5. M. Fändrich, J. Meinhardt, N. Grigorieff, Prion 3, 89 (2009).
6. F. Merino, S. Raunser, Angew. Chem. 129, 2890 (2017).
7. A. W. P. Fitzpatrick et al. , Nature 547, 185 (2017).
8. T. P. J. Kno wles, M. Vendruscolo, C. M. Dobson, Nat. Rev.
Mol. Cell Biol. 15, 384 (2014).
9. R. Riek, D. S. Eisenberg,Nature 539, 227 (2016).
10. A. Abbott, E. Dolgin, Nature 540, 15 (2016).
herald a new class
of synthetic drug.”
Studies show the potential of synthetic and combinations
of broadly neutralizing antibodies
46 6 OCTOBER 2017 • VOL 358 ISSUE 6359