JOURNAL OF VIROLOGY, Sept. 2011, p. 9646–9650
Copyright © 2011, American Society for Microbiology. All Rights Reserved.
Vol. 85, No. 18
Inhibition of HIV-1 Integration in Ex Vivo-Infected
CD4 T Cells from Elite Controllers?
Maria J. Buzon,1,2Katherine Seiss,2Robert Weiss,3Abraham L. Brass,2Eric S. Rosenberg,1
Florencia Pereyra,2,4Xu G. Yu,2and Mathias Lichterfeld1*
Infectious Disease Division, Massachusetts General Hospital, Boston, Massachusetts1; Ragon Institute, Boston, Massachusetts2;
Division of Nephrology, University of California, Davis, California3; and Infectious Disease Division, Brigham and
Women’s Hospital, Boston, Massachusetts4
Received 7 June 2011/Accepted 27 June 2011
Elite controllers spontaneously maintain undetectable levels of HIV-1 replication for reasons that remain
unclear. Here, we show that in elite controllers, direct ex vivo infection of purified CD4 T cells without prior in
vitro activation results in disproportionately low levels of integrated HIV-1 DNA relative to the quantity of
reverse transcripts, while the levels of two-long terminal repeat (2-LTR) circles were excessively elevated
relative to those of integrated HIV-1 DNA. This indicates that chromosomal HIV-1 integration is inhibited in
ex vivo-infected CD4 T cells from elite controllers. This defect in HIV-1 integration was unrelated to p21, a host
protein that can restrict early HIV-1 replication steps, and was not visible following infection of in vitro-
activated CD4 T cells from elite controllers. These data contribute to increasing evidence that intrinsic
inhibition of specific HIV-1 replication steps plays an important role in the ability of elite controllers to
maintain undetectable viral loads.
Despite encouraging findings in the recent RV144 HIV-1
vaccine study (13), correlates of immune protection against
HIV-1 remain poorly understood and continue to represent a
high-priority area of research (7). Since individuals who spon-
taneously clear HIV-1 infection do not exist, efforts to identify
effective mechanisms of immune defense have focused on elite
controllers, a group of persons who maintain undetectable
levels of viral replication in the absence of antiretroviral ther-
apy, although residual low-level viremia remains detectable in
most of these patients by ultrasensitive detection techniques
(12). Current views suggest that effective suppression of HIV-1
replication in these patients is likely to involve a synergistic
interplay between multiple innate and adaptive immune de-
fense mechanisms and may be facilitated by specific polymor-
phisms in the human HLA class I gene locus (8, 11). Strong,
highly functional HIV-1-specific CD8 T cell responses have
been described for the majority of elite controllers (2), and
these cells are able to effectively restrict HIV-1 replication, at
least in in vitro experiments (15). In addition to HIV-1-specific
T cell responses, recent studies from two separate laboratories
indicated that cell-intrinsic inhibition of HIV-1 replication
steps can also importantly contribute to HIV-1 immune de-
fense in elite controllers (6, 14). In these investigations, it was
shown that in vitro infection of CD4 T cells from elite control-
lers consistently resulted in lower levels of HIV-1 replication
than in vitro infection of CD4 T cells from progressors and
healthy volunteers. This reduced susceptibility to HIV-1 in-
volved inhibition of early viral replication steps and was asso-
ciated with a selective upregulation of p21, a host protein from
the cyclin-dependent kinase inhibitor family that can modulate
HIV-1 replication in macrophages (1), hematopoietic stem
cells (16), and CD4 T cells (6). Intrinsic inhibition of HIV-1
replication steps in elite controllers was also suggested by a
recent study in which HIV-1 DNA was quantified in direct ex
vivo assessments of purified CD4 T cells (9). These investiga-
tions demonstrated that in comparison to HIV-1 patients with
highly-active antiretroviral therapy (HAART)-mediated sup-
pression of viral replication, elite controllers had significantly
lower levels of chromosomally integrated HIV-1 DNA but
elevated levels of HIV-1 2-long terminal repeat (2-LTR) cir-
cles, an episomal HIV-1 DNA form that results from aborted
integration of HIV-1 DNA into host chromosomes. This pat-
tern closely resembles alterations in HIV-1 replication dynam-
ics observed after exposure to pharmaceutical HIV-1 integrase
inhibitors and suggests that at least under specific circum-
stances, HIV-1 integration is restricted in elite controllers.
However, following ex vivo infection of CD4 T cells from elite
controllers, using spinoculation protocols, no evidence for cell-
intrinsic inhibition of HIV-1 integration in CD4 T cells from
elite controllers was found (9). Yet, intrinsic restriction of
HIV-1 replication may not be visible after infection of CD4 T
cells by spinoculation (14).
To overcome this possible limitation, we performed a de-
tailed investigation of HIV-1 reverse transcription and integra-
tion in directly ex vivo-isolated CD4 T cells that were infected
without spinoculation or prior to in vitro activation. For this
purpose, CD4 T cells from elite controllers (HIV-1 viral load,
?50/74 copies/ml; CD4 T cell count, 618/?l [363 to 1,001/?l])
recruited from the International HIV Controllers Study (www
.hivcontrollers.org) and reference cohorts of HIV-1-negative
volunteers and untreated HIV-1 progressors (viral load, 98,000
copies/ml [7,560 to 449,000 copies/ml]; CD4 T cell count,
488/?l [199 to 1,000/?l]) were ex vivo purified by negative
immunomagnetic selection (purity, ?90%). Afterwards, cells
* Corresponding author. Mailing address: Infectious Disease Divi-
sion, Massachusetts General Hospital, Boston, MA 02114. Phone:
(617) 724-4076. Fax: (617) 726-5611. E-mail: mlichterfeld@partners
?Published ahead of print on 6 July 2011.
were infected with a yellow fluorescence protein (YFP)-encod-
ing vesicular stomatitis virus G protein (VSV-G) pseudotyped
HIV-1 virus (3) (50% tissue culture infective dose [TCID50] of
5,000) that infects cells independently of coreceptor-mediated
entry processes and causes only a single round of infection,
thus allowing for detailed assessments of individual early
HIV-1 replication steps (4). After two washes, cells were plated
at a concentration of 5 ? 105cells/ml in 24-well round-bottom
plates in RPMI medium supplemented with 10% fetal calf
serum (FCS) but without the addition of exogenous interleu-
kin-2 (IL-2). Forty-eight hours after infection, cell lysates were
collected and subjected to quantification of HIV-1 late reverse
transcripts (LRT) and 2-LTR circles; chromosomally inte-
grated HIV-1 was detected in cell lysates collected 96 h after
infection using PCR protocols described in our previous work
Following infection of ex vivo-isolated CD4 T cells with
YFP-encoding VSV pseudotyped HIV-1, the proportions of
YFP-positive CD4 T cells were 0.58% (0.17 to 1.4%) in elite
controllers, 1.4% (0.77 to 2.14%) in progressors, and 1.79%
(1.07 to 2.58%) in HIV-1-negative persons. As summarized in
Fig. 1, infection of CD4 T cells resulted in significantly lower
levels of LRT and integrated HIV-1 DNA in CD4 T cells from
elite controllers than in cells from HIV-1-negative persons or
progressors, as described in our earlier findings (6). In con-
trast, no significant differences were found between 2-LTR
quantities from elite controllers and those from the two refer-
ence cohorts. These data resulted in significantly reduced ra-
tios of LRT to 2-LTR and integrated DNA to 2-LTR in elite
controllers in comparison to those for HIV-1-negative persons
or progressors, indicating that relative to LRT and integrated
HIV-1 DNA quantities, 2-LTR circles were disproportionately
elevated in elite controllers. Moreover, ratios of integrated
DNA to LRT were also significantly lower in elite controllers
than in HIV-1-negative persons or progressors, consistent with
a disproportionate decrease of integrated HIV-1 DNA relative
to LRT levels in elite controllers. Overall, this altered pattern
of early HIV-1 replication products strongly suggests a defect
at the level of HIV-1 integration in ex vivo-infected CD4 T cells
from elite controllers.
Prior studies have shown that p21, a host protein from the
cyclin-dependent kinase inhibitor family that is highly upregu-
lated in CD4 T cells from elite controllers, can inhibit HIV-1
integration in hematopoietic stem cells (16) and may also be
involved in restriction of early HIV-1 replication steps in mac-
rophages (1) and CD4 T cells (6). To analyze whether p21 is
FIG. 1. Inhibition of HIV-1 integration in purified, ex vivo-infected CD4 T cells from elite controllers. CD4 T cells from elite controllers,
HIV-1-negative persons, or HIV-1 progressors were ex vivo infected with HIV-1 without prior in vitro activation. (A) Summary of the quantity of
LRTs, 2-LTR circles, and integrated HIV-1 DNA in the three study cohorts. RT, reverse transcripts. (B) Ratios of the indicated HIV-1 DNA forms
from the three study cohorts. Autologous HIV-1 DNA levels in CD4 T cells without exogenous HIV-1 infection were subtracted from corre-
sponding HIV-1 DNA quantities after ex vivo infection. Significance was tested using Mann-Whitney U tests. Data are presented as box-and-
whisker plots, indicating the median, interquartile ranges, and minimum and maximum values.
VOL. 85, 2011 NOTES9647
involved in the observed inhibition of HIV-1 integration in ex
vivo-infected CD4 T cells from elite controllers, we performed
HIV-1 infection experiments with ex vivo-isolated CD4 T cells
in the presence of a small molecule inhibitor of p21 (#15;
concentration of 2 ?M) that selectively eliminates p21 through
proteasomal degradation (10); control cells were treated with
the carrier dimethyl sulfoxide (DMSO) only. In line with prior
work (6), the addition of the p21 inhibitor had no effect on
LTR, 2-LTR circles, and integrated HIV-1 DNA in CD4 T
cells from HIV-1-negative persons, likely as a result of low-
baseline p21 expression in these individuals (Fig. 2). However,
following inhibition of p21 in CD4 T cells from elite control-
lers, LRT and 2-LTR circles significantly increased and
reached levels similar to those in HIV-1-negative persons; this
suggests that p21 inhibition can overcome intrinsic restriction
at the level of HIV-1 reverse transcription in elite controllers.
In contrast, integrated HIV-1 DNA only weakly increased af-
ter p21 inhibition in CD4 T cells from elite controllers and
remained significantly lower than in control cells from HIV-1-
negative persons. Moreover, inhibition of p21 resulted in lower
ratios of integrated DNA to 2-LTR and integrated DNA to
LRT than in control cells, indicating that the disproportionate
decrease of integrated HIV-1 DNA relative to LRT and
2-LTR circles described above cannot be corrected by the
silencing of p21 (Fig. 2). Overall, this suggests that the intrinsic
inhibition of HIV-1 integration in ex vivo-infected CD4 T cells
from elite controllers is unrelated to p21.
We subsequently investigated whether an inhibition of
HIV-1 integration is also detectable in CD4 T cells from elite
controllers that were infected after in vitro activation. Our
previous work has shown that following in vitro activation, CD4
T cells from elite controllers were significantly less susceptible
to HIV-1 infection (6); this appeared to be related to blockages
at the level of HIV-1 reverse transcription and mRNA tran-
scription, while a possible inhibition of viral integration in
these cells remained unclear. To investigate this further, we
activated CD4 T cells from elite controllers and HIV-1-nega-
tive persons using CD3/CD8-bispecific antibodies (0.5 ?g/ml)
and IL-2 (50 IU/ml) as described before. After 5 days, CD4 T
cell populations without contaminating CD8 T cells (?0.1%)
were infected with the VSV-G pseudotyped HIV-1 virus
(TCID50of 1,000) and plated at a concentration of 5 ? 105
cells/ml in RPMI medium supplemented with 10% FCS and
IL-2. Cell lysates collected after 18 h were used for quantifi-
cation of LRTs and 2-LTR circles, while samples obtained
after 48 h were used for assessments of integrated HIV-1
DNA. Overall, we observed that LRT and integrated DNA
levels were significantly lower in elite controllers than in HIV-
1-negative persons, as demonstrated previously (6). The levels
of 2-LTR circles showed a similar pattern and were also re-
FIG. 2. Inhibition of HIV-1 integration in ex vivo-infected CD4 T cells from elite controllers is unrelated to p21. CD4 T cells from elite
controllers or HIV-1-negative persons were ex vivo-infected with HIV-1 without prior in vitro activation. Experiments were performed in the
presence of a small molecule inhibitor of p21 (striped bars) or the carrier DMSO as a control (solid bars). (A) Quantitation of levels of LRT,
2-LTR, and integrated HIV-1 DNA in the two study cohorts. (B) Corresponding ratios of indicated HIV-1 DNA forms. Significance was tested
by Mann-Whitney U tests or paired Wilcoxon tests, as appropriate. Data are presented as box-and-whisker plots, indicating the median,
interquartile ranges, and minimum and maximum values.
9648 NOTES J. VIROL.
duced in elite controllers compared to those in HIV-1-negative
persons (Fig. 3). Ratios of LRT to 2-LTR, integrated HIV-1
DNA to 2-LTR, and LRT to integrated HIV-1 DNA were not
significantly different between the two study cohorts, suggest-
ing that reduced levels of integrated HIV-1 DNA in ex vivo-
activated CD4 T cells from elite controllers represent a con-
sequence of reduced HIV-1 reverse transcripts and not an
independent restriction at the level of HIV-1 integration.
In this study, we analyzed early HIV-1 replication steps in
purified CD4 T cells that were infected directly ex vivo without
prior activation or spinoculation. We showed that exogenous
HIV-1 infection of directly ex vivo-isolated CD4 T cells from
elite controllers leads to disproportionate reductions in the
levels of integrated HIV-1 DNA relative to those of LRT and
2-LTR circles; moreover, the levels of 2-LTR circles were
disproportionately increased relative to those of LRT and in-
tegrated HIV-1 DNA. This specific pattern is consistent with a
block at the level of chromosomal HIV-1 integration and cor-
responds well to the recent description of increases in 2-LTR
quantities relative to that of chromosomally integrated HIV-1
DNA in ex vivo-isolated CD4 T cells from elite controllers (9).
In combination, these studies strongly suggest that at least
under specific circumstances, the efficacy of HIV-1 integration
can be markedly reduced in CD4 T cells from elite controllers
and warrant further studies to identify molecular mechanisms
that contribute to such a block. Notably, chromosomal inte-
gration of HIV-1 DNA depends on a number of different host
proteins, and alterations in the expression or function of such
proteins may lead to conditions that only insufficiently support
HIV-1 integration in elite controllers. Moreover, it is possible
that specific molecular inhibitors that block host proteins re-
quired for effective HIV-1 integration and in this way exert an
indirect effect on chromosomal HIV-1 integration are available
in CD4 T cells from elite controllers. Importantly, activation
levels of CD4 T cells from our elite controller cohort were
slightly elevated in comparison to those of HIV-1-negative
persons, as determined by surface expression levels of
HLA-DR in direct ex vivo assessments (data not shown); this
indicates that defective HIV-1 integration in CD4 T cells from
these patients cannot be attributed simply to reduced activa-
tion of CD4 T cells in elite controllers. Overall, the studies
presented here contribute to increasing evidence that intrinsic
restriction of HIV-1 replication plays an important role in the
ability of elite controllers to maintain undetectable viral loads
and may stimulate future mechanistic studies to identify cell-
intrinsic inhibitors of chromosomal HIV-1 integration in CD4
T cells from elite controllers.
This work was supported by the U.S. National Institutes of Health
(AI093203 to M.L. and AI078799 and AI089339 to X.G.Y.). M.L. and
X.G.Y. are both recipients of the Doris Duke Clinical Scientist De-
velopment Award. M.J.B. is supported by a Fellowship Award from
FIG. 3. Uncompromised HIV-1 integration in CD4 T cells from elite controllers infected after in vitro activation. CD4 T cells from elite
controllers or HIV-1-negative persons were activated with CD3/CD8-bispecific antibodies and IL-2 for 5 days before being infected with HIV-1.
(A) Summary of the quantities of late reverse transcripts, 2-LTR circles, and integrated HIV-1 DNA in the two study cohorts. (B) Ratios of the
indicated HIV-1 forms in the two study cohorts. Significance was tested using Mann-Whitney U tests. Data are presented as box-and-whisker plots,
indicating the median, interquartile ranges, and minimum and maximum values.
VOL. 85, 2011 NOTES9649
the European Molecular Biology Laboratory (EMBL). R.W. is sup-
ported by the NIH (grants CA135401 and DK082690) and by the
Medical Service of the U.S. Department of Veterans Affairs.
1. Bergamaschi, A., et al. 2009. The CDK inhibitor p21Cip1/WAF1is induced by
Fc?R activation and restricts the replication of human immunodeficiency
virus type 1 and related primate lentiviruses in human macrophages. J. Virol.
2. Betts, M. R., et al. 2006. HIV nonprogressors preferentially maintain highly
functional HIV-specific CD8? T cells. Blood 107:4781–4789.
3. Brass, A. L., et al. 2008. Identification of host proteins required for HIV
infection through a functional genomic screen. Science 319:921–926.
4. Butler, S. L., M. S. Hansen, and F. D. Bushman. 2001. A quantitative assay
for HIV DNA integration in vivo. Nat. Med. 7:631–634.
5. Buzo ´n, M. J., et al. 2010. HIV-1 replication and immune dynamics are
affected by raltegravir intensification of HAART-suppressed subjects. Nat.
6. Chen, H., et al. 2011. CD4? T cells from elite controllers resist HIV-1
infection by selective upregulation of p21. J. Clin. Invest. 121:1549–1560.
7. Council of the Global HIV Vaccine Enterprise. 2010. The 2010 scientific
strategic plan of the Global HIV Vaccine Enterprise. Nat. Med. 16:981–989.
8. Fellay, J., et al. 2007. A whole-genome association study of major determi-
nants for host control of HIV-1. Science 317:944–947.
9. Graf, E. H., et al. 2011. Elite suppressors harbor low levels of integrated HIV
DNA and high levels of 2-LTR circular HIV DNA compared to HIV?
patients on and off HAART. PLoS Pathog. 7:e1001300.
10. Park, S. H., X. Wang, R. Liu, K. S. Lam, and R. H. Weiss. 2008. High
throughput screening of a small molecule one-bead-one-compound combi-
natorial library to identify attenuators of p21 as chemotherapy sensitizers.
Cancer Biol. Ther. 7:2015–2022.
11. Pereyra, F., et al. 2010. The major genetic determinants of HIV-1 control
affect HLA class I peptide presentation. Science 330:1551–1557.
12. Pereyra, F., et al. 2009. Persistent low-level viremia in HIV-1 elite controllers
and relationship to immunologic parameters. J. Infect. Dis. 200:984–990.
13. Rerks-Ngarm, S., et al. 2009. Vaccination with ALVAC and AIDSVAX to
prevent HIV-1 infection in Thailand. N. Engl. J. Med. 361:2209–2220.
14. Sa ´ez-Cirio ´n, A., et al. 3 June 2011, posting date. Restriction of HIV-1
replication in macrophages and CD4? T cells from HIV controllers. Blood
[Epub ahead of print.] doi:10.1182/blood-2010-12-327106.
15. Sa ´ez-Cirio ´n, A., et al. 2007. HIV controllers exhibit potent CD8 T cell
capacity to suppress HIV infection ex vivo and peculiar cytotoxic T lympho-
cyte activation phenotype. Proc. Natl. Acad. Sci. U. S. A. 104:6776–6781.
16. Zhang, J., D. T. Scadden, and C. S. Crumpacker. 2007. Primitive hemato-
poietic cells resist HIV-1 infection via p21. J. Clin. Invest. 117:473–481.