Content uploaded by Ronald S Veazey
Author content
All content in this area was uploaded by Ronald S Veazey
Content may be subject to copyright.
Macaque studies of vaccine and microbicide
combinations for preventing HIV-1 sexual transmission
Dan H. Barouch
a,b,1
, Per Johan Klasse
c
, Jason Dufour
d
, Ronald S. Veazey
d
, and John P. Moore
c,1
a
Beth Israel Deaconess Medical Center, Boston, MA 02115;
b
Ragon Institute of MGH, MIT, and Harvard, Boston, MA 02114;
c
Department of Microbiology and
Immunology, Weill Cornell Medical College, New York, NY 10021; and
d
Tulane National Primate Research Center, Tulane University School of Medicine,
Covington, LA 70433
Edited* by Robert C. Gallo, Institute of Human Virology, University of Maryland School of Medicine, Baltimore, MD, and approved April 12, 2012 (received for
review February 22, 2012)
Vaccination and the application of a vaginal microbicide have
traditionally been considered independent methods to prevent
the sexual transmission of HIV-1 to women. Both techniques can
be effective in macaque models, and limited efficacy has been ob-
served in clinical trials for each. Here, we have addressed whether
vaccines and microbicides can be used together to provide rein-
forced protection against virus challenge of rhesus macaques. In
two separate experiments, four groups of animals were vacci-
nated with a T-cell–based adenovirus (Ad) vectored vaccine aimed
at reducing postinfection viral loads and/or a partially effective
dose of a vaginal microbicide aimed at blocking infection of a
high-dose vaginal challenge with SIVmac251 or SHIV-162P3. In
the first study, the only two protected animals were in the group
that received Ad26/Ad5HVR48 vaccine vectors combined with
the fusion inhibitor T-1249 as the vaginal microbicide before SIV-
mac251 challenge. In the second study, vaccination with Ad35/
Ad26 vectors combined with the CCR5 inhibitor maraviroc as the
vaginal microbicide led to significant reductions of both acquisi-
tion of infection and postinfection viral loads following SHIV-
SF162P3 challenge. As expected, the vaccine by itself reduced viral
loads but had no acquisition effect, whereas the microbicide had
a partial acquisition effect but minimal impact on viral loads. For
both measures of protective efficacy, the vaccine–microbicide com-
bination differed more from controls than did either separate in-
tervention. Overall, the data suggest that vaccines and microbicides
are complementary techniques that may protect better when used
together than separately.
HIV type 1 (HIV-1) continues to spread globally through
sexual transmission, particularly among young women in
the developing world who have little power to insist on the use of
condoms to protect themselves (1, 2). In such circumstances,
biology-based interventions become of substantial importance
(2). Traditionally, the spread of infectious disease has been most
effectively controlled by vaccination, but this approach has had,
at best, limited success against HIV-1 in efficacy trials (3, 4). New
vaccines based on live viral vectors are under evaluation in the
macaque model of HIV-1 infection, with partial success at
blocking acquisition of infection and a more consistent ability to
reduce postinfection viral loads (5, 6). However, inducing neu-
tralizing antibodies of sufficient breadth and potency to prevent
virus transmission remains a major challenge (7).
An alternative approach to HIV-1 prevention involves the
application of antiretroviral drugs (ARVs), either systemically as
oral preexposure prophylaxis (PrEP) or locally at the mucosal
sites of entry of the virus into the body (8). In the latter method,
a gel or other formulation containing an ARV is administered
vaginally or rectally as a microbicide to prevent the initial stages
of entry of HIV-1 into the body (9). Studies with oral PrEP and
vaginal microbicides have shown partial efficacy against HIV-1
sexual transmission to humans in the CAPRISA and iPREX
trials but not in other studies (10–13).
Within the HIV-1 prevention fields, vaccines and microbicides
have been regarded as independent, and in some respects, rival
technologies. And yet there is logic to considering the two ap-
proaches as complementary. On a very simple level, two partially
effective barriers to infection might be superior to either one
alone. A more sophisticated argument is that, by reducing the
extent to which an incoming virus replicates in mucosal tissues,
a microbicide might buy time for the maturation of vaccine-in-
duced immune responses that either eliminate the virus or fur-
ther counter its expansion such that the viral load set point is
reduced (14, 15). Another possibility is that the combined use
of a microbicide that partially blocks acquisition of infection
with a T-cell–based vaccine that reduces postinfection viral loads
will result in both protective effects following virus challenge. We
sought to test the underlying hypotheses by carrying out two
experiments in the rhesus macaque vaginal challenge model,
using inhibitors of HIV-1 entry as the microbicides and adeno-
virus (Ad)-vector–based vaccines.
Results
Design of Combination Studies. Both studies were of broadly sim-
ilar design. The rhesus macaques were immunized by the intra-
muscular route with an Ad-vector–based vaccine. Eight months
after the boost immunization, a vaginally applied inhibitor was
then given as the microbicide for 30 min before vaginal simian
immunodeficiency virus (SIV) or simian-human immunodefi-
ciency virus (SHIV) challenge. For comparison, other animals
were not vaccinated or did not receive the microbicide. In both
experiments, the macaques were treated with progesterone 30
d before challenge, to ensure that most control animals would
become infected after a single exposure to virus (16). We con-
sidered other designs, particularly for the second experiment,
such as the use of multiple challenges without prior progesterone
treatment. However, our experience has been that infection of
control animals can be highly inconsistent under such conditions,
which adversely impacts how the outcome of the study can be
interpreted (17). The Ad-vector–based vaccines were primarily
T-cell–based vaccines aimed at reducing viral loads post-
infection; they would not be expected to block acquisition of
infection based on the high-dose challenge in the SIVmac251
study and the mismatched Env in the SHIV-162P3 study (6, 18).
We hypothesized that the acquisition and viral load effects of
each individual component would both occur when the two
methods were used together. Alternatively, a microbicide-medi-
ated reduction in the infectivity of the incoming virus, or in the rate
at which it expanded locally after deposition in the vagina, might
increase the protective capabilities of the vaccine.
A critical aspect of the experimental designwas to use a partially
protective microbicide concentration; too much inhibitor might
be so strongly protective as to prevent us from ascertaining any
Author contributions: D.H.B., P.J.K., and J.P.M. designed research; D.H.B., J.D., and R.S.V.
performed research; P.J.K. analyzed data; and D.H.B., P.J.K., and J.P.M. wrote the paper.
The authors declare no conflict of interest.
*This Direct Submission article had a prearranged editor.
1
To whom correspondence may be addressed. E-mail: dbarouch@bidmc.harvard.edu or
jpm2003@med.cornell.edu.
8694–8698
|
PNAS
|
May 29, 2012
|
vol. 109
|
no. 22 www.pnas.org/cgi/doi/10.1073/pnas.1203183109
impact of the vaccine, but too little might have no measurable ef-
fect. Hence in preliminary experiments we titrated each microbicide
against the challenge virus in an attempt to define a concentration
that would protect half of animals when used alone. In both
experiments, the animals were divided into four groups: Control
(i.e., no intervention), group C; vaccine only, group V; microbicide
only, group M; and vaccine + microbicide, group V+M.
Fusion-Inhibitor Microbicide and SIVmac251 Challenge. In the first
study, the vaccine was an Ad26/Ad5HVR48 vector regimen
expressing Gag-Pol-Env-Nef (18). The microbicide was the fu-
sion inhibitory peptide, T-1249, given at 200 μg/mL (40 μM) in
4 mL of an aqueous gel 30 min before high-dose vaginal chal-
lenge with SIVmac251 (19, 20). Each group contained six ani-
mals, but one animal was omitted from group V when it was
found to be retrovirus infected before challenge. All of the
animals in groups C (n= 6) and V (n= 5) became infected, and
postinfection viral loads were consistently ∼10-fold lower in
vaccinees (group V) as expected (Fig. 1). This magnitude of viral
load reduction is comparable with what was observed using the
same vaccine in other challenge formats (18). Hence any effect
of progesterone treatment on vaccine efficacy and/or local im-
munity is not substantial (21, 22).
We did not achieve our intended goal of protecting half of the
animals in group M. Here, four of the animals became system-
ically viremic with typical viral load kinetics. However, the other
two did so only after a delay of 4–6 wk. In our experience, such a
delay is highly unusual and thus likely relates to partially pro-
tective effects of the T-1249 microbicide (Discussion). Two of the
six animals in the V+M group remained uninfected throughout
the period of monitoring (Fig. 1). This outcome was not statis-
tically significant for acquisition (V+M vs. C, P= 0.23). However,
as the only two animals to remain uninfected in this experiment
were in the V+M combination, we considered that the data
pattern was sufficiently encouraging as to warrant a further ex-
ploration of the vaccine–microbicide combination concept.
CCR5-Inhibitor Microbicide and SHIV-SF162P3 Challenge. In the sec-
ond study, we changed the vaccine to the more potent Ad35/
Ad26 vector regimen expressing SIVsmE543 Gag-Pol-Env (6).
We also switched the microbicide to the CCR5 inhibitor mar-
aviroc as, unlike T-1249, this drug is used for treating HIV-1–
infected people and is currently being evaluated for preventing
transmission (23, 24). Our original intent was to continue to use
SIVmac251 as the challenge virus; accordingly, the vaccine in-
cluded an SIV Env component. However, during the period
between vaccination and challenge, we concluded that we could
not reliably identify a maraviroc dose that provided 50% pro-
tection against SIVmac251. In contrast, we obtained a smooth
dose–response curve for maraviroc-mediated protection against
SHIV-162P3 and decided to use this challenge virus instead
(24). We also considered adopting a multiple challenge protocol
without progesterone, but were unable to obtain a sufficiently
consistent rate of infection of control animals under these con-
ditions (17). Given the necessity to infect almost all of the control
animals and approximately half of those in the microbicide-only
group, we elected to proceed with a single, high-dose challenge
with SHIV-162P3. Thus, the Env component of the vaccine (SIV)
and the challenge virus (HIV-1) were mismatched.
As a result, we predicted that the vaccine would reduce viral
loads but would not block acquisition of infection following virus
challenge (6). The study therefore allowed us to assess whether
the combined use of the partially protective microbicide and the
T-cell–based vaccine would prove superior to each individual
modality by affording both protection against acquisition of in-
fection and improved virologic control. The basic predictions
were that: (i) V+M would prove superior to V alone for blocking
acquisition of infection and (ii) V+M would prove superior to M
alone for virologic control. A more tangible advantage to the
combination would be evident if V+M also protected better
against acquisition than M and curbed viral loads more than V.
Each group in this study included eight animals to increase
statistical power, although one macaque was omitted from the
combination group when it was found to be retrovirus infected
Fig. 1. Viral load profiles for animals given vaccine (V), T-1249 (M), the combination (V+M), or neither (C) before challenge with SIVmac251. The viral load
over time after challenge is shown in one diagram for each group (n= 5 for V; n= 6 for each of the other groups). Baseline values are offset from each other
for clarity.
Barouch et al. PNAS
|
May 29, 2012
|
vol. 109
|
no. 22
|
8695
MEDICAL SCIENCES
before challenge. The immunogenicity of the Ad35/Ad26 vac-
cine, as judged by IFN-γELISPOT assays, was comparable to
that observed previously in experiments involving the same
vaccine but different challenge conditions, e.g., intrarectal with
SIVmac251 (6).
After the high-dose vaginal SHIV-162P3 challenge (Fig. 2),
seven of eight control animals (group C) were infected. Similarly,
seven of eight animals in the vaccine-only group (group V) were
infected, as expected for this T-cell–based vaccine with no rele-
vant Env component. In the microbicide-only group (group M),
four of eight animals were infected. Thus, we achieved our goal
of using a microbicide dose that protected half of the animals. In
the combination group (group V+M), two of seven animals were
infected. The intervention efficacies in groups M and V+M were
43 and 67%, respectively. For V+M compared with V or C, the
infection rate was significantly lower (P= 0.035), although the
rate for group M was not significantly lower than for group C
(P= 0.14). V+M was, therefore, superior to V for blocking
acquisition of infection, in accordance with our first prediction.
Although the infection rate was lowest in the combination group,
V+M was not, however, significantly different from M for ac-
quisition (P= 0.38).
Plasma viral loads were assessed to determine postinfection
virologic control in the infected animals. Viral loads in the four
groups differed in terms of peak viral loads (Fig. 3A;P= 0.0006,
V vs. C, P<0.01; V+M vs. M, P<0.05; and V+M vs. C, P<
0.001), and of viral loads on day 28, which was the first day on
which all infected monkeys had detectable viremia (Fig. 3B;P=
0.0001, V vs. C, P<0.01; V+M vs. M, P<0.01; and V+M vs. C,
P<0.001). The log viral load reductions measured as peaks and
on day 28 were, respectively, 1.2 and 1.1 for V; 0.70 and 0.45 for
M; and 2.4 and 2.3 for V+M. The magnitude of the viral load
reduction by the vaccine was again similar to what we have
reported in previous experiments with the Ad35/Ad26 vaccine
(6). Hence V+M was superior to M for virologic control, in
accordance with our second prediction.
Furthermore, the viral loads in the V+M group were the
lowest among the four groups in both comparisons and they
differed more from controls than did either V or M.
Discussion
In this study, we used a rhesus macaque vaginal transmission
model to test the concept that vaccines and microbicides are two
prevention methods that might work better together than in-
dividually. In both experiments, the fewest infected animals were
found in the V+M combination group, and the difference in
acquisition between the V and V+M groups was statistically
significant in the second experiment. Moreover, the difference in
viral loads between groups M and V+M was statistically signif-
icant in the second experiment. Thus, these data provide overall
support to the concept that combining vaccines and microbicides
may be superior to using either by itself. Whether a vaccine
containing a relevant Env component combined with a micro-
bicide might provide synergistic protection against acquisition of
infection remains to be determined.
We observed a trend toward augmented protection against
acquisition of infection in the V+M group, compared with group
M, in both studies (two of six vs. zero of six animals uninfected
for V+M vs. M in the first study, five of seven vs. four of eight
animals in the second). One interpretation of this trend is that
the partially protective microbicide dose slowed down the local
expansion of the incoming virus in the mucosal tissues of a subset
of the animals, and hence the effective size of the inoculum (see
below). We note that, in the first experiment, two animals in
group M (i.e., T-1249 recipients) become infected only after an
unusually long delay of 2–4 wk (Fig. 1). We have only previously
seen such a long delay in a single animal, one that was also given
a partially protective dose of T-1249 before SIVmac251 chal-
lenge. In the combination groups, we hypothesize that the
microbicide-mediated delay in virus expansion may have allowed
vaccine-elicited immune responses enough additional time to
either eliminate the infection entirely or to restrict replication
Fig. 2. Viral load profiles for animals given vaccine (V), maraviroc (M), the combination (V+M), or neither (C) before challenge with SHIV-162P3. The viral load
over time after challenge is shown in one diagram for each group (n= 7 for V+M; n= 8 for each of the other groups). Baseline values are offset from each
other for clarity.
8696
|
www.pnas.org/cgi/doi/10.1073/pnas.1203183109 Barouch et al.
and lower the set point. In the second experiment, maraviroc
may reduce the number of target cells in the vaginal mucosa that
express sufficient free CCR5 for virus entry, noting that the
abundance of CCR5
+
target cells is an important influence on
the establishment of infection (25). We think it is unlikely that
vaccine-induced antibodies are involved in these effects, partic-
ularly given the complete mismatch of the Env component of the
vaccine and challenge virus in the second study. Instead, cellular
immunity, such as CD8
+
T-cell responses, is likely responsi-
ble (6). Consistent with this possibility is our observation that
these Ad vectors, delivered by the intramuscular route to rhesus
macaques, elicit potent and durable effector memory CD8
+
T
lymphocytes at mucosal surfaces, including in the cervicovaginal
mucosa (26). We note that if a vaccine–microbicide combination
approach were used in humans, the Env components of the
vaccine strain and the incoming virus would probably also be
mismatched, but not to the same extent as in our experiment
(SIV vs. HIV-1). Accordingly, at least some additional protective
contribution from the Env component of the vaccine could
reasonably be anticipated (6).
The combined effects of the vaccine and the microbicide on
acquisition and viral loads are compatible with our emerging
understanding of the early events in mucosal SIV/SHIV infection
(15). In a simple model, the crucial event is the productive in-
fection of the first target cell, with successive generations of
progeny virus then irrevocably colonizing the rest of the organism.
However, postexposure prophylaxis with a reverse-transcriptase
inhibitor can be started as late as 24–48 h after inoculation and
still completely prevent SIV infection (27), and the earliest foci of
infection appear during the first few days after intravaginal in-
oculation (28). A more sophisticated view of mucosal infection is
that the events preceding systemic breakout involve a branching
process (29); a limited number of replication foci are first estab-
lished but they can be extinguished in a critical labile phase before
they spread the virus to increasingly distant lymphoid tissue sites
(15). Sometimes, the initial foci may be eradicated by CD8
+
T
lymphocytes within the mucosa, leading to a failed infection at
the level of the whole organism (30). This beneficial outcome may
be more probable when the inoculum size is low or is lowered by
an intervention (e.g., a topically or orally delivered ARV). We
therefore suggest that, in our experiments, the partially active
microbicide doses reduced the infectivity of the inoculum to this
labile zone where the outcome of infection could go either way
(systemic breakout or extinction), perhaps stochastically. The
vaccine-induced immune responses may therefore contribute to
reducing acquisition when the number of infectious foci is close
to the critical threshold level.
In the only previous study of combining a vaccine (DNA/
rAd5) and a microbicide (a zinc-finger inhibitor) in macaques,
there were modest differences in the rate of infection and viral
loads between the combination and the no-intervention control
group (31). However, as neither the vaccine nor the microbicide
caused detectable reductions in viral loads or acquisition, re-
spectively, the outcome was inconclusive as to the benefitof
the combination; the Kaplan–Meier curves for the vaccine and
combination groups were largely superimposed, as were those
for the microbicide and control groups (31). Any interpretation
is further complicated by two escalations of the challenge virus
dose [eventually to 3,000 TCID
50
(50% tissue culture infectious
dose)] to try to overcome inconsistent infections of the control
animals (17, 31).
In summary, our data support the concept that vaccines and
microbicides might be useful when used in combination for
protection against sexual HIV-1 transmission to women by the
vaginal route. By extrapolation, the same benefits might apply to
rectal microbicides for men and women, although this supposi-
tion will need to be confirmed experimentally given that trans-
mission conditions may be tissue dependent. Further research in
these areas is important for the HIV-1 prevention field, because
if microbicides do become licensed for clinical use they would
likely be used in the context of HIV-1 vaccine efficacy trials.
Although we used inhibitors of virus-cell entry as the micro-
bicides, blockers of other stages in the HIV-1 replication cycle
might have similar effects and thus warrant further investigation
(9, 10). Any beneficial effect of a microbicide on vaccine efficacy
would, of course, only come into play when the product is ac-
tually used but does not fully prevent HIV-1 transmission.
Microbicide failures due to nonuse require different solutions
such as sustained release devices (vaginal rings) or long-lasting
gels (32–34). Whether these coitally independent delivery
methods, and also orally delivered ARVs, can be combined with
vaccination for reinforced protection remains to be determined.
A conceptually related combination approach would be to use
a vaccine designed to induce mucosal antibodies of appropriate
quality and quantity to block acquisition of infection together
with a T-cell vaccine that limits replication postinfection. All of
these hypotheses could, in principle, be tested in the macaque
model using various challenge formats. The availability of larger
numbers of animals would increase the statistical power of tests
of vaccine–microbicide combinations to reduce acquisition,
particularly when the microbicide itself has, by design, a partial
effect by itself, as was the case in our second experiment.
Materials and Methods
Challenge Viruses. The R5 virus SHIV-162P3, derived from the HIV-1 SF162
primary isolate as described elsewhere (35) and propagated in phytohe-
magglutin in (PHA)-activated rhesus macaque peripheral blood mono-
nuclear cells (PBMC), was obtained through the National Institutes of Health
(NIH) AIDS Research and Reference Reagent Program, Division of AIDS,
National Institute of Allergy and Infectious Diseases, NIH (cat. no. 6526;
contributors: Janet Harouse, Cecilia Cheng-Mayer, and Ranajit Pal). The
SIVmac251 stock we used has been described previously (18).
Fig. 3. Peak and set-point viral loads for animals given vaccine (V), maraviroc (M), the combination (V+M), or neither (C) before challenge with SHIV-162P3.
Measurements of viral loads are depicted on the yaxis for each of the four groups as indicated on the category axis. Means are marked by horizontal bars ±
SEM for (A) the peak viral loads and (B) the viral loads on day 28, the first day on which all of the animals that became infected were detectably viremic.
Barouch et al. PNAS
|
May 29, 2012
|
vol. 109
|
no. 22
|
8697
MEDICAL SCIENCES
Microbicides. T-1249 was a gift from Michael Greenberg and Dani Bolognesi
who, at the time, worked for Trimeris (Durham, NC) (19, 20). Maraviroc
was obtained via John Pottage (ViiV Healthcare, Research Triangle Park, NC)
(23, 24).
Vaccine Delivery. Ad26, Ad35, and Ad5HVR48 vectors expressing SIV Gag/Pol/
Env antigens were produced as previously described (36, 37). Indian-origin
rhesus monkeys were immunized twice by the intramuscular route with 10
10
viral particles per vector in Ad26/Ad5HVR48 or Ad35/Ad26 heterologous
prime-boost regimens (6). Vaccine immunogenicity was confirmed by IFN-γ
ELISPOT assays. Viral challenges were performed at least 6 mo following the
final vaccination.
Microbicide Delivery and Virus Challenge. A single intramuscular injection
of Depo-Provera (progesterone) was given to female Indian rhesus macaques
30 d before challenge, to synchronize the menstrual cycle, thin the vaginal
epithelium, and facilitate virus transmission (16). Virus challenge and
microbicide delivery protocols are more fully described elsewhere (20, 24,
38). On the day of challenge, 4 mL of the microbicide formulated in a
hydroxyethylcellulose (HEC) gel, or a placebo HEC gel, were applied atrau-
matically to the vagina, 30 min before SHIV-162P3 or SIVmac251 was added
in a 1-mL volume containing 500 TCID
50
or 4 ×10
7
RNA copies/mL, re-
spectively. Infection status was determined by measuring plasma viral load
at 7, 14, 21, 28, 42, 56, and 70 d postchallenge, using a commercially avail-
able branched DNA (bDNA) assay with a sensitivity limit of 125 RNA copies/
mL (Siemens). All protocols were approved by the institutional animal care
and use committees. The animals were housed in accordance with the
American Association for Accreditation of Laboratory Animal Care stand-
ards. All 55 animals described in Figs. 1 and 2 were experimentally naïve at
the start of the studies and were negative for antibodies against SIV and
type D retrovirus.
Statistical Analysis. Acquisition frequencies were analyzed by Fisher’s exact
test. Because the interventions reduced acquisitions in the hypothesized
direction, and any marginal increase in acquisition over that for controls
(possible only in the second experiment) would have been entirely attrib-
utable to random effects, the Pvalues given are for one-tail comparisons.
The αlevel was set to P= 0.05 for all tests. Measurements of viral loads
among infected animals (peak and set points) were compared overall for the
four groups by one-way ANOVA, followed by Tukey’s posttest for individual
group pairs, with Kramer’s extension for unequal group sizes. Although the
group sizes were too small to test for Gaussian distribution or homosce-
dasticity, we chose the parametric test because the high frequency of pro-
tection by V+M left so few infected animals that a nonparametric approach
would have been too blunt to make some important comparisons. In Results,
the Pvalue for the four-group comparison is given first, followed by any
significant outcome of Tukey’s posttest. Thus, out of the six pairwise com-
binations of the four groups, three results by Tukey’s test turned out to be
nonsignificant and are not listed. ANOVA and Tukey tests were performed
in Prism (GraphPad).
ACKNOWLEDGMENTS. We thank Dani Bolognesi and Michael Greenberg for
the provision of T-1249 and Kelsi Rasmussen, Megan Gardner, Meagan
Watkins, and Thomas Ketas for technical support. This work was funded by
a supplement to National Institutes of Health (NIH) Cooperative Agreement
U19 AI76982 and by NIH Grants AI078526 and AI066924, the Ragon Institute,
and the Bill and Melinda Gates Foundation.
1. Quinn TC, Overbaugh J (2005) HIV/AIDS in women: An expanding epidemic. Science
308:1582–1583.
2. Padian NS, et al. (2011) HIV prevention transformed: The new prevention research
agenda. Lancet 378:269–278.
3. Kim JH, Rerks-Ngarm S, Excler JL, Michael NL (2010) HIV vaccines: Lessons learned and
the way forward. Curr Opin HIV AIDS 5:428–434.
4. Barouch DH, Korber B (2010) HIV-1 vaccine development after STEP. Annu Rev Med
61:153–167.
5. Hansen SG, et al. (2011) Profound early control of highly pathogenic SIV by an ef-
fector memory T-cell vaccine. Nature 473:523–527.
6. Barouch DH, et al. (2012) Vaccine protection against acquisition of neutralization-
resistant SIV challenges in rhesus monkeys. Nature 482:89–93.
7. Walker LM, Burton DR (2010) Rational antibody-based HIV-1 vaccine design: Current
approaches and future directions. Curr Opin Immunol 22:358–366.
8. Wainberg MA (2011) AIDS: Drugs that prevent HIV infection. Nature 469:306–307.
9. Minces LR, McGowan I (2010) Advances in the development of microbicides for the
prevention of HIV infection. Curr Infect Dis Rep 12:56–62.
10. Abdool Karim Q, et al.; CAPRISA 004 Trial Group (2010) Effectiveness and safety of
tenofovir gel, an antiretroviral microbicide, for the prevention of HIV infection in
women. Science 329:1168–1174.
11. Grant RM, et al.; iPrEx Study Team (2010) Preexposure chemoprophylaxis for HIV
prevention in men who have sex with men. N Engl J Med 363:2587–2599.
12. Celum C, Baeten JM (2012) Tenofovir-based pre-exposure prophylaxis for HIV pre-
vention: Evolving evidence. Curr Opin Infect Dis 25:51–57.
13. van der Straten A, van Damme L, Haberer JE, Bangsberg DR (2012) How well does
PREP work? Unraveling the divergent results of PrEP trials for HIV prevention. AIDS
26:F13-F19.
14. Haase AT (2005) Perils at mucosal front lines for HIV and SIV and their hosts. Nat Rev
Immunol 5:783–792.
15. Haase AT (2010) Targeting early infection to prevent HIV-1 mucosal transmission.
Nature 464:217–223.
16. Marx PA, et al. (1996) Progesterone implants enhance SIV vaginal transmission and
early virus load. Nat Med 2:1084–1089.
17. Veazey RS, Shattock RJ, Klasse PJ, Moore JP (2012) Animal models for microbicide
studies. Curr HIV Res 10:79–87.
18. Liu J, et al. (2009) Immune control of an SIV challenge by a T-cell-based vaccine in
rhesus monkeys. Nature 457:87–91.
19. Eron JJ, et al. (2004) Short-term safety and antiretroviral activity of T-1249, a second-
generation fusion inhibitor of HIV. J Infect Dis 189:1075–1083.
20. Veazey RS, et al. (2008) Tropism-independent protection of macaques against vaginal
transmission of three SHIVs by the HIV-1 fusion inhibitor T-1249. Proc Natl Acad Sci
USA 105:10531–10536.
21. Genescà M, McChesney MB, Miller CJ (2010) Depo-provera treatment does not ab-
rogate protection from intravenous SIV challenge in female macaques immunized
with an attenuated AIDS virus. PLoS ONE 5:e9814.
22. Sanders-Beer B, et al. (2010) Depo-Provera does not alter disease progression in
SIVmac-infected female Chinese rhesus macaques. AIDS Res Hum Retroviruses 26:
433–443.
23. MacArthur RD, Novak RM (2008) Reviews of anti-infective agents: Maraviroc: The first
of a new class of antiretroviral agents. Clin Infect Dis 47:236–241.
24. Veazey RS, et al. (2010) Protection of rhesus macaques from vaginal infection by
vaginally delivered maraviroc, an inhibitor of HIV-1 entry via the CCR5 co-receptor.
J Infect Dis 202:739–744.
25. Pandrea I, et al. (2012) Mucosal SIV transmission in African Green Monkeys: Suscep-
tibility to infection is proportional to target cell availability at mucosal sites. J Virol.
26. Li H, et al. (2011) Durable mucosal simian immunodeficiency virus-specific effector
memory T lymphocyte responses elicited by recombinant adenovirus vectors in rhesus
monkeys. J Virol 85:11007–11015.
27. Tsai CC, et al. (1998) Effectiveness of postinoculation (R)-9-(2-phosphonylmethox-
ypropyl) adenine treatment for prevention of persistent simian immunodeficiency
virus SIVmne infection depends critically on timing of initiation and duration of
treatment. J Virol 72:4265–4273.
28. Miller CJ, et al. (2005) Propagation and dissemination of infection after vaginal
transmission of simian immunodeficiency virus. J Virol 79:9217–9227.
29. Wick D, Self SG (2000) Early HIV infection in vivo: Branching-process model for
studying timing of immune responses and drug therapy. Math Biosci 165:115–134.
30. Blancou P, et al. (2001) The infiltration kinetics of simian immunodeficiency virus-
specific T cells drawn to sites of high antigenic stimulation determines local in vivo
viral escape. Proc Natl Acad Sci USA 98:13237–13242.
31. Cheng-Mayer C, et al. (2011) Delay of SHIV infection and control of viral replication in
vaccinated macaques challenged in the presence of a topical microbicide. AIDS 25:
1833–1841.
32. Malcolm RK, Edwards KL, Kiser P, Romano J, Smith TJ (2010) Advances in microbicide
vaginal rings. Antiviral Res 88(Suppl 1):S30–S39.
33. Forbes CJ, et al. (2011) Non-aqueous silicone elastomer gels as a vaginal microbicide
delivery system for the HIV-1 entry inhibitor maraviroc. J Control Release 156:
161–169.
34. Malcolm RK, et al. (2012) Sustained release of the CCR5 inhibitors CMPD167 and
maraviroc from vaginal rings in rhesus macaques. Antimicrob Agents Chemother 56:
2251–2258.
35. Harouse JM, et al. (2001) Mucosal transmission and induction of simian AIDS by CCR5-
specific simian/human immunodeficiency virus SHIV(SF162P3). J Virol 75:1990–1995.
36. Abbink P, et al. (2007) Comparative seroprevalence and immunogenicity of six rare
serotype recombinant adenovirus vaccine vectors from subgroups B and D. J Virol 81:
4654–4663.
37. Roberts DM, et al. (2006) Hexon-chimaeric adenovirus serotype 5 vectors circumvent
pre-existing anti-vector immunity. Nature 441:239–243.
38. Veazey RS, et al. (2005) Protection of macaques from vaginal SHIV challenge by
vaginally delivered inhibitors of virus-cell fusion. Nature 438:99–102.
8698
|
www.pnas.org/cgi/doi/10.1073/pnas.1203183109 Barouch et al.
