ILLUSTRATION: JANET IWASA/UNIVERSITY OF UTAH130 14 JANUARY 2022 • VOL 375 ISSUE 6577 science.org SCIENCE
A
n HIV infection remains madden-
ingly difficult to cure because the
virus is so good at hiding out. Yes,
antiretroviral (ARV) drugs can con-
trol an infection, but HIV integrates
its genes into human chromosomes,
evading both drugs and the immune sys-
tem. Now, a research group studying a
handful of HIV-infected people who have
been on treatment for at least 9 years has
made a tantalizing discovery: The inte-
grated HIV genomes, or proviruses, that
persist the longest are increasingly con-
fined to largely inactive stretches of host
DNA, perhaps stymieing production of
new viruses—and opening new avenues for
cure research.
In earlier work the team had found the
same “blocked and locked” integration
phenomenon in rare HIV-infected indi-
viduals who remain healthy for decades
without treatment. Together, the findings
raise the possibility of “a peaceful coexis-
tence between HIV and humans,” proposes
Mathias Lichterfeld, an infectious disease
clinician at Brigham and Women’s Hospi-
tal who led the new analysis.
The group believes long-treated, in-
fected people with this signature integra-
tion “landscape” could consider stopping
treatment to see whether their immune
systems can check any remaining virus
production, a strategy they hope to test in
a volunteer soon.
The work, published in Cell this week,
“provides a road map to a cure” for HIV
infections, says Steven Deeks, an HIV cli-
nician at the University of California, San
Francisco, who was not part of the study
but has collaborated with Lichterfeld and
his colleagues. Others are more circum-
spect. “It’s good to be cautious until this
is done on hundreds of [HIV-infected
people],” says Mary Kearney, an HIV re-
searcher at the National Cancer Institute.
“But this is a great start.”
Over the past 15 years, researchers have
tried to drain the reservoirs of HIV hid-
ing in the chromosomes of infected people
with drugs meant to prod the proviruses
into producing new viruses. The white
blood cells that harbor the active pro-
viruses then either self-destruct or become
easy prey for other immune warriors.
But these “shock-and-kill” strategies have
made little headway.The new strategy builds on a landmark
study in which Lichterfeld, Xu Yu of the
Ragon Institute of MGH, MIT and Har-
vard, and other scientists examined “elite
controllers,” the tiny group of untreated
HIV-infected people—less than 0.5%—who
live with the virus for decades and suffer
no obvious harm from it. In these rare
cases, they reported in 2020 in Nature, the
proviruses tended to cluster in regions of
chromosomes that lack genes or harbor
largely inactive genes that code for a type
of protein known as a zinc finger (ZNF)—
which, intriguingly, evolved to repress
ancient retroviruses. In both kinds of re-
gions, the DNA is more tightly packed than
elsewhere in the human genome, making
proviruses less accessible to factors that
drive transcription.
In the 15 December 2021 issue of
Science Translational Medicine (STM),
Yu and Lichterfeld—who are married—
followed up on that finding by proposing
how these unusual, virus-hostile land-
scapes might evolve in elite controllers.
(Deeks was a co-author.) Researchers have
long thought latent proviruses can form
large, effectively invisible cellular reser-
voirs. But the STM study suggests HIV in-
tegrated into regions with active genes is
never entirely invisible. Instead infected
cells produce a trickle of new HIV and, as
a result, get eliminated.
“The reservoir has always been de-
scribed as transcriptionally silent and
stable, but we actually find it’s tran-
scriptionally active to a large extent,”
Lichterfeld says. The team concluded that
elite controllers have special immune re-
sponses and other murky mechanisms that
speed the elimination of this active reser-
voir, resulting in what they described as a
“skeleton reservoir” dominated by blocked
and locked proviruses.
In the latest work, Lichterfeld’s and Yu’s
team examined 1270 proviruses detected in
the blood of six people at different points
during prolonged HIV treatment. The group
found that in three of the people, intact HIV
genes gradually accumulated in human
gene deserts and the quiescent ZNF genes.
“It’s a little bit like a chess game: There are
only a few spots left where the king can still
go,” Lichterfeld says. The result: an integra-
tion landscape increasingly similar to the
one seen in the elite controllers.
Independently, Lillian Cohn, an immuno-
logist at the Fred Hutchinson Cancer Re-
search Center, made a similar finding:By Jon Cohen
BIOMEDICINE
Mapping where HIV hides
suggests cure strategy
Long-term antiretrovirals may corner viral genomes
in inactive regions of DNA
In HIV infection, a DNA copy of its genome (dark blue
in illustration) integrates with human DNA (teal). In
people on antiretrovirals for years, these “proviruses”
can become trapped in quiescent stretches.NEWS | IN DEPTH