Adenovirus vector vaccination induces expansion
of memory CD4 T cells with a mucosal homing
phenotype that are readily susceptible to HIV-1
Adel Benlahrecha, Julian Harrisb, Andrea Meisera, Timos Papagatsiasa, Julia Horniga, Peter Hayesa, Andre Lieberc,
Takis Athanasopoulosb, Veronique Bachyd, Eszter Csomord, Rod Danielse, Kerry Fisherf, Frances Gotcha, Len Seymourf,
Karen Logana, Romina Barbagalloa, Linda Klavinskisd, George Dicksonb, and Steven Pattersona,1
aDepartment of Immunology, Imperial College London, Chelsea and Westminster Hospital, 369 Fulham Road, London SW10 9NH, United Kingdom;bSchool
of Biological Science, Royal Holloway University of London, Egham, Surrey TW20 0EX, United Kingdom;cDivision of Medical Genetics, Department of
Medicine, University of Washington, 1705 NE Pacific Street, Seattle, WA 98195;dPeter Gorer Department of Immunobiology, Guy’s Hospital, King’s College
London, London SE1 9RT, United Kingdom;eVirology Division, National Institute for Medical Research, The Ridgeway, Mill Hill, London NW7 1AA, United
Kingdom; andfHybrid Systems Ltd., 77 Heyford Park, Upper Heyford OX25 5HD, Oxfordshire, United Kingdom
Edited by Robert C. Gallo, University of Maryland, Baltimore, MD, and approved September 30, 2009 (received for review July 16, 2009)
In the recently halted HIV type 1 (HIV-1) vaccine STEP trial, indi-
viduals that were seropositive for adenovirus serotype 5 (Ad5)
showed increased rates of HIV-1 infection on vaccination with an
Ad5 vaccine. We propose that this was due to activation and
expansion of Ad5-specific mucosal-homing memory CD4 T cells. To
test this hypothesis, Ad5 and Ad11 antibody titers were measured
in 20 healthy volunteers. Dendritic cells (DCs) from these individ-
uals were pulsed with replication defective Ad5 or Ad11 and
co-cultured with autologous lymphocytes. Cytokine profiles, pro-
liferative capacity, mucosal migration potential, and susceptibility
to HIV infection of the adenovirus-stimulated memory CD4 T cells
were measured. Stimulation of T cells from healthy Ad5-seropos-
itive but Ad11-seronegative individuals with Ad5, or serologically
memory CD4 T cells expressing ?4?7integrins and CCR9, indicating
a mucosal-homing phenotype. CD4 T-cell proliferation and IFN-?
production in response to Ad stimulation correlated with Ad5
antibody titers. However, Ad5 serostatus did not correlate with
total cytokine production upon challenge with Ad5 or Ad11.
Expanded Ad5 and Ad11 memory CD4 T cells showed an increase
in CCR5 expression and higher susceptibility to infection by R5
tropic HIV-1. This suggests that adenoviral-based vaccination
against HIV-1 in individuals with preexisting immunity against Ad5
results in preferential expansion of HIV-susceptible activated CD4
T cells that home to mucosal tissues, increases the number of virus
targets, and leads to a higher susceptibility to HIV acquisition.
STEP trial ? vaccine ? alpha 4 beta 7
There were great expectations of the recent HIV-1 vaccine phase
risk of HIV infection. Because preexisting vector immunity was
expected to blunt the effectiveness of vaccination, volunteers were
above 200). The vaccine consisted of three first-generation repli-
cation defective Ad5 vectors expressing HIV-1 Gag, Pol, and Nef.
ineffective. More worryingly however, individuals who received the
vaccine were more susceptible to HIV-1 infection compared to the
placebo arm, this was most significant in vaccinees with high
antibody titers to the vector (2), [Robertson, http://www.hvtn.org/
fgm/1107slides/Robertsonfinal.pdf, reviewed in (3, 4)].
There is an urgent need to understand the mechanisms under-
lying the increased HIV infection rate in Ad5-seropositive vacci-
nees to ascertain whether adenoviral or any viral-vector-based
approach may ever be suitable for HIV vaccination. Natural Ad5
infection occurs via the nasopharynx or gut and replicates in
eplication-defective recombinant Ad5 vectors have been con-
sidered as potential vaccine delivery vehicles for HIV-1 (1).
we hypothesized that vaccination of individuals immune to Ad5
with adenovirus vectors would activate and expand T cells express-
ing a mucosal homing phenotype, and these cells would migrate to
the gut mucosa, increasing the number of permissive HIV-1 target
cells and the risk of infection.
Adenovirus-Specific Cytokine Responses Do Not Correlate with Ad5
Antibody Titers. To investigate the relationship between Ad5
serostatus and Ad-specific cellular immune responses, Ad5 and
Ad11 antibody titers were measured in 20 healthy volunteers (Fig.
2A). IFN-? ELISPOT was performed on 15 of these donors (Fig.
S1A). In 73.3% (mean response of 103.7 ? 30.8 SFCs/106PBMCs)
responses against Ad5 and Ad11, respectively. For comparison,
responses against inactivated influenza were also measured. All
individuals but one showed influenza-specific IFN-? ELISPOT
Ad5 serostatus is a surrogate for Ad5- or Ad11-induced IFN-?
plotted against Ad5 antibody titers (Fig. S1B). We observed no
significant correlations between Ad5 or Ad11 IFN-? ELISPOT
responses and Ad5 antibody titers. Furthermore, there was no
correlation between responses generated against Ad5 and Ad11
(Fig. S1C). To identify which cell populations were responsive to
Ad5 and Ad11, we undertook multiparameter flow cytometric
analysis for the detection of intracellular IFN-?, IL-2, and TNF-?.
Dendritic cells (DCs) from all 20 donors were pulsed with Ad5,
Ad11, tetanus toxoid, heat-inactivated influenza, or Staphylococcus
The majority of the IFN-? response against Ad5 and Ad11 was
respectively) in comparison to CD4 T cells (means of 0.043 ?
0.001%, P ? 0.043 and 0.058 ? 0.01, P ? 0.078, respectively) (Fig.
S1D). Similarly, CD8 T cells produced significantly more TNF-? in
response to Ad5 or Ad11 (means of 0.211 ? 0.04, P ? 0.002 and
0.257 ? 0.05, P ? 0.016, respectively) (Fig. S1D). In contrast, we
detected no or very little IL-2 production in response to Ad5 and
Author contributions: A.B., T.A., V.B., E.C., R.D., K.F., F.G., L.S., K.L., R.B., L.K., G.D., and S.P.
designed research; A.B., J. Harris, A.M., T.P., J. Hornig, P.H., and G.D. performed research;
wrote the paper.
The authors declare no conflict of interest.
This article is a PNAS Direct Submission.
1To whom correspondence should be addressed. E-mail: email@example.com.
This article contains supporting information online at www.pnas.org/cgi/content/full/
November 24, 2009 ?
vol. 106 ?
Ad11 by either CD4 or CD8 cells (means of 0.028 ? 0.005, 0.044 ?
0.025 and 0.029 ? 0.008, 0.033 ? 0.008, respectively) (Fig. S1D).
Furthermore, we observed no significant correlation between Ad5
serostatus and IL-2 or TNF-? production in response to Ad5 or
Ad11, while a moderate but statistically significant correlation was
(r ? 0.547, P ? 0.012, Spearman test) (Fig. S2).
The functional quality of the responding Ad5 and Ad11-specific
CD4 T cells is depicted in Fig. 1E. No significant differences were
found between Ad5- and Ad11-responding CD4 T cells in terms of
cytokine profiles with cells predominantly producing TNF-? alone,
or in combination with IFN-? or IL-2. Cytokine-producing CD8 T
The majority of responding cells produced IFN-?, TNF-?, or both,
indicating an effector phenotype (Fig. S1E).
Adenovirus-Induced T-Cell Proliferation Correlates with Ad5 Antibody
Titers. We next addressed the ability of Ad vectors to induce T-cell
influenza, or SEB then co-cultured with autologous CFSE-stained
T cells. Unlike cytokine production, which was mainly mediated by
CD8 T cells, the majority of proliferating T cells in response to Ad5
2.901 ? 0.754%, respectively) in comparison to just 1.360 ? 0.275
and Fig. S3). No major differences were seen between the percent-
ages of expanded CD4 and CD8 T lymphocytes in response to
tetanus toxoid and influenza (Fig. 1A). We observed a strong
correlation between Ad5 and Ad11-mediated CD4 T-cell prolifer-
ation (Spearman r ? 0.8602, P ? 0.0001) (Fig. 1B), likely due to
was a trend toward a positive correlation between Ad5- or Ad11-
induced CD4 T lymphocyte proliferation and Ad5 serostatus (r ?
was statistically significant when Ad5 titers ?200 were analyzed
toxoid, influenza, or SEB was unrelated to the individuals’ Ad5
titers (Fig. 1C).
?4?7 Expression by Expanded Adenovirus-Specific Memory CD4 T
on the expression of a heterodimer comprising of ?4 and ?7
integrins that bind to MAdCAM-1 on endothelial cells in the
gastrointestinal tract (5–9). We therefore measured expression of
5 days with Ad5, Ad11, second-generation Ad5 vector, influenza,
tetanus toxoid, or SEB-pulsed DCs. Representative profiles are
a total of 14 out of 26 samples (four samples from buffy coats for
which we did not have neutralizing antibody data) that exhibited a
proliferative response against adenovirus stimulation. The majority
of the expanded Ad5-specific CD4 T cells co-expressed ?4and ?7
(means of 71.5 ? 3.6%) (Fig. S6), whereas undivided cells (CFSE
high) CD4 T cells exhibited minimal expression levels (Fig. S5).
Moreover, stimulation with Ad11 or second-generation Ad5 (n ?
4), considered to produce less adenovirus-encoded protein than
first-generation vectors, resulted in a similar expansion of ?4??7?
CD4?T cells (means of 72.1 ? 3.3 and 78.8 ? 12.2%, respectively).
newly synthesized adenovirus proteins were not augmenting stim-
ulation. Conversely, CD4 T lymphocytes that proliferated in re-
sponse to un-pulsed or SEB-stimulated DCs showed a higher
expression of ?4but not ?7(Fig. S5). Since SEB stimulation results
in proliferation of a heterogeneous population of naive and mem-
ory T cells, while responses against Ad5 or Ad11 would presumably
be mediated by memory CD4 T cells, we repeated the SEB
stimulations using purified memory CD4 T cells (n ? 2). In both
samples tested, memory CD4 T cells that proliferated against SEB
resulted in higher expression of ?4but not ?7(Fig. S7), indicating
that the lower levels of ?7expression were not skewed by naive
Memory CD4 T lymphocytes that proliferated in response to
influenza exhibited a mean percentage of cell expressing ?4?7of
Ad5 serostatus. (A) CFSE-labeled lymphocytes from 20
tetanus toxoid, influenza, or SEB-stimulated DCs. The
mean percentages ? SD of proliferating (CFSE low)
CD4 (triangles) and CD8 (squares) T cells are shown. (B)
The percentages of proliferating CD4 T cells in re-
those in response to Ad11-pulsed DCs (y axis). P and R
values were obtained using the spearman correlation
test. (C) The percentages of expanded Ad5, Ad11, tet-
anus toxoid, influenza, and SEB-specific CD4 T cells (y
axis) were plotted against the individuals’ Ad5 anti-
body titers (x axis). Continued and dotted lines repre-
sent the best fit line and 95% confidence intervals
the Spearman test.
Ad-induced T-cell expansion correlates with
Benlahrech et al. PNAS ?
November 24, 2009 ?
vol. 106 ?
no. 47 ?
58.72 ? 4.50. Although this was lower than Ad5 or Ad11-induced
?4?7expression, it was not statistically significant (Fig. S6). How-
ever, proliferating T cells against tetanus toxoid exhibited signifi-
cantly lower expression levels of ?4?7(mean percentage of 41.7 ?
5.9) (Fig. S6). Taken together, this suggests that antigens originally
encountered through mucosal surfaces, adenovirus, and influenza,
expanded memory CD4 T cells with a mucosal homing phenotype,
while tetanus toxoid, which most individuals initially encounter as
a systemic intramuscular vaccine, expanded memory CD4 T cells
that are predominantly negative for mucosal homing markers.
Adenovirus-Induced ?4?7 Increases Correlate with Ad5 Titers. We
next investigated the relationship between increased ?4?7expres-
sion by Ad-specific memory CD4 T cells and preexisting Ad5
neutralizing antibodies. For this purpose, we determined the fold
in relation to background expression levels of unstimulated CD4 T
lymphocytes as the latter varied considerably among the donors
tested (Fig. 2B). We observed a positive correlation between the
fold increases in total CD4 T lymphocytes expressing ?4?7 in
response to Ad5 or Ad11 and Ad5 antibody titers (P ? 0.066 and
P ? 0.028, respectively) (Fig. 2C). Conversely, no correlations with
individuals with Ad5 titers ?200 showed significantly higher per-
centages of CD4 T cells expressing ?4?7in comparison to donors
0.023, and P ? 0.003, respectively) (Fig. 2D). This finding was
specific to adenoviral challenge as no differences were seen be-
tween the two groups when tetanus toxoid or influenza was used as
a stimulus (Fig. 2D).
To further address the possibility that the observed increases in
?4?7expression was an experimental artifact, we plotted ?4?7fold
increases against the percentages of proliferating cells in response
S8, there was a positive and significant correlation between Ad5-,
Ad11-, and influenza-induced proliferation and ?4?7expression,
while tetanus toxoid-mediated expansion was unrelated to ?4?7
increases, indicating that expression of these molecules was re-
stricted to Ad5, Ad11, and flu-specific memory CD4 T cells rather
than a non-specific property of proliferating cells.
CCR9 and CCR5 Upregulation in Response to Ad5 and Ad11 Challenge.
Migration to mucosal tissues is also facilitated by the chemokine
receptor CCR9 in response to its ligand CCL25 which is constitu-
tively expressed in the small intestine (10, 11). We examined the
CD4 T cells in 4 Ad5 responders by flow cytometry. There were
significantly increased levels of surface CCR9 in cells expanded by
vectors in comparison to non-proliferating T cells (Fig. 3 A and C).
CCR5 expression is upregulated by T cells with effector/memory
phenotype and Th1 cells (12, 13) and is expressed on gut mucosal
we investigated CCR5 expression on expanded adenovirus-specific
T cells when cultured with first- or second-generation Ad5, or
Ad11 correlates with Ad5 titers. (A) Ad5 and Ad11
antibody titers of 20 donors were measured as indi-
cated in Materials and Methods. (B) T lymphocytes
from donors were co-cultured with unpulsed DC, or
Ad5, Ad11, tetanus toxoid, or influenza-pulsed DCs.
response to Ad5 (blue bars), Ad11 (red bars), tetanus
toxoid (white bars), or influenza (shaded bars) were
that were cultured with unpulsed DCs. (C) The fold
increases in ?4?7expression by T cells in response to
Ad5, Ad11, tetanus toxoid or influenza were plotted
against the respective Ad5 antibody titers. R and P
values were obtained using the nonparametric Spear-
man test. (D) The fold increases in ?4?7expression in
response to Ad5, Ad11, tetanus toxoid, or influenza
shown with lines representing means ? SD whereas P
values were obtained using the Mann Whitney U test.
The expression of ?4?7in response to Ad5 and
www.pnas.org?cgi?doi?10.1073?pnas.0907898106 Benlahrech et al.
Ad11-pulsed autologous DCs in comparison to CFSE-high undi-
?4?7receptor, which was expressed by a greater number of ade-
novirus-specific T cells, increased CCR5 expression by CD4 T cells
was seen in response to tetanus toxoid, influenza virus, and SEB
Re-Stimulated Adenovirus-Specific CD4 T Cells Are More Permissive to
cells express higher levels of CCR5, the co-receptor required for
infection by R5 strains of HIV-1 (15, 16), we tested whether these
cells were susceptible to HIV-1 infection. DCs from buffy coats
(n ? 4) that showed a proliferative response against Ad5 were
pulsed with first- or second-generation Ad5 or Ad11 and co-
cultured with autologous lymphocytes for 3 days; they were then
infected with the R5 virus, HIV-1BAL,for an additional 4 days.
Infection was monitored by intracellular staining for p24 and by
ELISA for p24 in the supernatant. As can be seen in Fig. 4 A and
B, intracellular staining for HIV-1 p24 was significantly higher in
the proliferating CFSE-low CD4 T cells compared to undivided
with Ad5, Ad11, or second-generation Ad5-pulsed DCs.
CCR9 (A) and CCR5 (B) expression by divided CFSElow
(green histograms) or undivided CFSEhighCD3?CD4?T
lymphocytes (blue histograms) was measured by flow
cytometry. Red histograms represent the non-specific
representative of four samples for CCR9 and eight sam-
CCR5, respectively, are shown with lines representing
means ? SD.
CCR5 and CCR9 expression by expanded Ad-
specific CD4 T cells. CFSE-stained lymphocytes were co-
cultured with autologous DCs that were unpulsed or
days. Cells were then cultured in the presence of infec-
of CD3?CD8?cells for p24 gag. The gating strategy to
(A) where quadrants were based on p24 stained unin-
fected T cells subjected to the same conditions. (B) The
mean percentages of HV-1 p24 positive divided CFSElow
P values were obtained using the Mann Whitney U test.
(C) Lymphocytes from 4 Ad5-responders were either cul-
tured alone or co-cultured with unstimulated or Ad5,
days. Cells were then infected with HIV-1BALfor an addi-
culture supernatants by ELISA as indicated in Materials
and Methods. The means are given; error bars, SEM. P
values were obtained using the Mann Whitney U test.
HIV-1BALpreferentially infects expanded Ad-
Benlahrech et al.PNAS ?
November 24, 2009 ?
vol. 106 ?
no. 47 ?
adenovirus stimulated cultures (Fig. 4C). Furthermore, we ob-
served similar levels of HIV-1 p24 in culture supernatants from
Ad5- or Ad11-stimulated lymphocytes.
Recent reports showed no correlation between Ad5 serostatus and
Ad5 stimulated cytokine production suggesting that activation of
adenovirus-specific memory cells could not explain the increased
susceptibility to HIV infection in the STEP trial (17–19). We
confirmed these observations as measured by total IFN-? ELIS-
POTS and by flow cytometry and found no correlation between
Ad5 neutralizing antibody and IL-2 and TNF-? production. How-
ever, by flow cytometry, we observed a moderate but significant
correlation between IFN-? secretion by CD4 T cells and Ad5
antibody. This discrepancy may reflect the use of antigen-pulsed
DC versus PBMC stimulations.
In contrast to cytokine responses, CD4 T-cell proliferation
stimulated by Ad5 or Ad11 correlated with Ad5 antibody titers,
particularly in individuals with Ad5 neutralizing antibody titers
above 200. In an early study, Chirmule et al. reported an
association between CD4 T-cell proliferation and Ad5- neu-
tralizing antibodies in 74 donors (20). We observed a highly
significant association between CD4 T-cell expansion in re-
sponse to Ad5 and Ad11, most likely reflecting conserved CD4
T-cell epitopes across the two sero subgroups (21–23). Given
that there are 51 known human strains of adenovirus, one
might expect most individuals to have encountered at least one
strain and thus CD4 memory T cells from Ad5-seronegative
individuals may be expected to proliferate in response to
conserved epitopes on Ad5. Although we do not have data to
explain this observation it could be due to differences in
immunodominance hierarchy such that conserved epitopes
may not be the most immunodominant in all strains of
adenovirus. This would also be influenced by HLA composi-
tion. Also of relevance to this question are observations of
clear differences between Ad5-seronegative subjects (who may
be seropositive for other Ad subtypes) and donors with
preexisting Ad5 immunity with regards to NF-?B and cell
death pathways and IRF1/IRF1-binding sites as revealed by
gene microarrays. In addition, the finding of higher levels of
individuals who received the saline placebo in the STEP trial (24,
Zak et al., www.hivvaccineenterprise.org/conference/archive/2008/
specific for Ad5-seropositive individuals.
We show that memory CD4 T cells expanded by adenovirus
stimulation exhibited elevated expression of the gut-homing mol-
ecules ?4?7and CCR9 with expression of ?4?7being significantly
higher in individuals with Ad5 antibody titers greater than 200.
There is no commercially available antibody that recognizes this
the two subunits. However, the ?7 integrin is only known to
with T-cell trafficking or retention in mucosal sites (8, 25), and thus
our approach should identify mucosal homing cells.
In the current investigation, we also show that Ad5 or Ad11
re-stimulated memory CD4 T cells are more susceptible to HIV-1
infection than unstimulated cells. HIV-1 preferentially replicates in
activated T cells (26), and therefore, CD4 T cells activated in vitro
to any antigen could be infected by HIV. However, our key
observation with respect to the STEP trial is that Ad5-activated
memory cells would home to mucosal tissue, the site of HIV
transmission. In a recent study by Fauci’s group (27), it has been
shown that the ?4?7receptor binds to HIV-1 gp120 and may serve
as a co-receptor for HIV-1 uptake. Thus, it is plausible that the
increased expression of ?4?7by re-challenged Ad5 memory CD4 T
cells may have contributed toward their increased susceptibility to
Other hypotheses have been put forward to explain the STEP
trial outcome, including a report from Perreau et al. demonstrating
in in vitro studies that Ad5 immune complexes activate DCs that in
turn would promote Ad5-specific CD4 T-cell activation and a
higher susceptibility to HIV-1 infection (28). However, vaccine
recipients who were Ad5-seronegative at baseline would have
developed anti Ad5 neutralizing antibodies after the first immuni-
zation. Since the majority of volunteers in the STEP study received
all three immunizations, all vaccinees would have become Ad5-
seropositive regardless of their baseline serostatus and shown
increased rates of HIV infection.
In the Merck vaccine trial individuals with neutralizing Ad5
antibody titers of less than 18 showed no increase in infection
rates compared with the placebo group (2, Robertson, http://
www.hvtn.org/fgm/1107slides/Robertsonfinal.pdf). Given that in
the vaccine group these individuals would have been expected to
become strongly Ad5-positive after three intramuscular vacci-
nations, why did they not show increase risk of infection?
Naturally acquired adenovirus replicates in epithelial cells in
mucosal tissue and would induce CD4 T cells that home to the
site of infection. On re-exposure to adenovirus, expansion and
activation of a mucosal homing CD4 memory population would
occur, increasing in the number of potential target cells at the
site of infection and increasing the incidence of infection. The
vaccine was given by intramuscular injection and in individuals
with very low or no immunity to adenovirus, this would lead to
the generation of activated CD4 cells that are susceptible to
HIV-1 infection but which do not home to mucosal tissue and
thus should not increase the risk of infection. Thus a testable
prediction of this hypothesis is that in vitro stimulation with
adenovirus vectors of PBMCs from vaccinated individuals who
had no preexisting neutralizing antibodies for Ad5 would cause
expansion of CD4 T cells expressing low levels of mucosal
homing markers. Interestingly in vaccines, Ad5-specific memory
CD4 T cells were found to exhibit lower levels of ?4?7 in
individuals seronegative for Ad5 at baseline compared to those
that were Ad5-seropositive before vaccination (18). Immuno-
logical analysis of the STEP trial showed that the percentage of
activated adenovirus-specific CD4 T cells in the blood was lower
in the group with preexisting immunity to Ad5 (19; Robertson,
CROI 2008, http://www.retroconference.org/2008/data/files/
retro2008?frameset.html) and may reflect recruitment of Ad5-
specific mucosal CD4 T cells to mucosal tissues, as suggested by
this study. Antigen-induced CD4 T-cell activation is transient
and fades as the antigen disappears; however, there was in-
creased susceptibility to infection in the Ad-seropositive group
for several months. This may reflect known persistence of
adenovirus, which is reported to last for several years, thus
causing more prolonged stimulation (29). Additionally, Tatsis et
al. demonstrated that replication-incompetent Ad5 vectors, such
Finally, our data also suggests that because there are conserved
problem may not be overcome by vectors derived from rarer
serotypes. As proposed above, analysis of the mucosal homing
phenotype of Ad5 vaccinated, Ad5-seronegative and -seropositive
individuals may provide further relevant data. Our findings suggest
a cautious approach in the development of adenovirus virus vector
vaccines for HIV.
*Zak DE, AIDS Vaccine 2008, October 18, 2008, Cape Town, South Africa.
www.pnas.org?cgi?doi?10.1073?pnas.0907898106 Benlahrech et al.
Materials and Methods Download full-text
A detailed description of materials and methods is given in the SI Materials and
Methods but is briefly as follows:
Vectors and Virus Neutralizing Antibody. First-generation E1- and E3-deleted
and pTP-deleted Ad5 vectors all carrying the gene for green fluorescent protein
(GFP) were propagated in packaging cell lines and purified by CsCl gradient
in 20 healthy volunteers by adding vectors [1,000 virus particles (vp)/cell] treated
after 48 h.
Proliferation of Ad-Specific Memory T Cells. Monocytes were isolated from the
blood of healthy volunteers with CD14 immunomagnetic beads and used to
2,500 vp/cell of first-generation Ad5 or second-generation Ad5 or Ad11 vectors
and co-cultured with carboxyfluorescein succinimidyl ester (CFSE)-labeled autol-
ogous lymphocytes for 5 days. At the end of culture, cells were labeled with
fluorescently labeled antibodies to CD3, CD4, CD8, CD11c, CD195 (CCR5), CD49d
(?4), ?7, and CCR9 and analyzed by flow cytometry. Responses to other antigens
cytes cells with Ad vector-pulsed DCs for 3 days, infecting with HIV-1Bal, and
fixed, stained for HIV p24, and analyzed by flow cytometry. Supernatant was
analyzed for p24 by ELISA.
Cytokine Analysis. DCs were pulsed for 3 h with Ad vectors or other antigens
described above, matured with LPS for 3 h, and co-cultured with autologous
gating strategy indicated in Fig. S10. Antigen-pulsed, DC-stimulated cells were
also analyzed for IFN-? secretion by ELISPOT.
Bergin for their intellectual input and Dr. Mario Roederer for providing Simula-
tion Program with Integrated Circuit Emphasis (SPICE) and Pestle software. This
support from the Bill and Melinda Gates Foundation and The Stephens Trust,
Chelsea and Westminster Hospital.
replication-incompetent adenovirus vectors. Annu Rev Med 55:355–372.
2. Buchbinder SP, et al. (2008) Efficacy assessment of a cell-mediated immunity HIV-1
vaccine (the Step Study): A double-blind, randomized, placebo-controlled, test-of-
concept trial. Lancet 372:1881–1893.
4. Sekaly RP (2008) The failed HIV Merck vaccine study: A step back or a launching point
for future vaccine development? J Exp Med 205:7–12.
5. Berlin C, et al. (1993) Alpha 4 beta 7 integrin mediates lymphocyte binding to the
mucosal vascular addressin MAdCAM-1. Cell 74:185–195.
6. Hamann A, Andrew DP, Jablonski-Westrich D, Holzmann B, Butcher EC (1994) Role of
?4-integrins in lymphocyte homing to mucosal tissues in vivo. J Immunol 152:3282–
with an antibody to a gut-homing integrin ?4?7. Gastroenterology 111:1373–1380.
8. Johansson-Lindbom B, Agace WW (2007) Generation of gut-homing T cells and their
localization to the small intestinal mucosa. Immunol Rev 215:226–242.
9. Wagner N, et al. (1996) Critical role for beta7 integrins in formation of the gut-
associated lymphoid tissue. Nature 382:366–370.
10. Kunkel EJ, et al. (2000) Lymphocyte CC chemokine receptor 9 and epithelial thymus-
expressed chemokine (TECK) expression distinguish the small intestinal immune com-
partment: Epithelial expression of tissue-specific chemokines as an organizing princi-
ple in regional immunity. J Exp Med 192:761–768.
11. Zabel BA, et al. (1999) Human G protein-coupled receptor GPR-9–6/CC chemokine
receptor 9 is selectively expressed on intestinal homing T lymphocytes, mucosal lym-
phocytes, and thymocytes and is required for thymus-expressed chemokine-mediated
chemotaxis. J Exp Med 190:1241–1256.
12. Bleul CC, Wu L, Hoxie JA, Springer TA, Mackay CR (1997) The HIV coreceptors CXCR4
and CCR5 are differentially expressed and regulated on human T lymphocytes. Proc
Natl Acad Sci USA 94:1925–1930.
13. Sallusto F, et al. (1999) Switch in chemokine receptor expression upon TCR stimulation
reveals novel homing potential for recently activated T cells. Eur J Immunol 29:2037–
14. Agace WW, et al. (2000) Human intestinal lamina propria and intraepithelial lympho-
cytes express receptors specific for chemokines induced by inflammation. Eur J Immu-
15. Berger EA, Murphy PM, Farber JM (1999) Chemokine receptors as HIV-1 coreceptors:
Roles in viral entry, tropism, and disease. Annu Rev Immunol 17:657–700.
16. Wyatt R, Sodroski J (1998) The HIV-1 envelope glycoproteins: Fusogens, antigens, and
immunogens. Science 280:1884–1888.
do not induce measurable vector-specific T cells in human trials. J Virol 83:6318–6322.
18. Hutnick NA, et al. (2009) Baseline Ad5 serostatus does not predict Ad5 HIV vaccine-
induced expansion of adenovirus-specific CD4(?) T cells. Nat Med 15:876–878.
HIV-1 vaccine candidate in humans. NatMed 15:873–875.
20. Chirmule N, et al. (1999) Immune responses to adenovirus and adeno-associated virus
in humans. Gene Ther 6:1574–1583.
21. DiPaolo N, et al. (2006) Evaluation of adenovirus vectors containing serotype 35 fibers
for vaccination. Mol Ther 13:756–765.
22. Olive M, Eisenlohr L, Flomenberg N, Hsu S, Flomenberg P (2002) The adenovirus capsid
protein hexon contains a highly conserved human CD4? T-cell epitope. Hum Gene
23. Onion D, et al. (2007) The CD4? T-cell response to adenovirus is focused against
conserved residues within the hexon protein. J Gen Virol 88:2417–2425.
24. McElrath MJ, et al. (2008) HIV-1 vaccine-induced immunity in the test-of-concept Step
Study: A case-cohort analysis. Lancet 372:1894–1905.
25. Shaw SK, Brenner MB (1995) The beta 7 integrins in mucosal homing and retention.
Semin Immunol 7:335–342.
26. Stevenson M, Stanwick TL, Dempsey MP, Lamonica CA (1990) HIV-1 replication is
controlled at the level of T cell activation and proviral integration. EMBO J 9:1551–
27. Arthos J, et al. (2008) HIV-1 envelope protein binds to and signals through integrin
?4?7, the gut mucosal homing receptor for peripheral T cells. Nat Immunol 9:301–309.
immune complexes creates an improved environment for replication of HIV in T cells.
J Exp Med 205:2717–2725.
C adenovirus DNA in human mucosal lymphocytes. J Virol 76:10608–10616.
30. Tatsis N, et al. (2007) Adenoviral vectors persist in vivo and maintain activated CD8? T
cells: Implications for their use as vaccines. Blood 110:1916–1923.
Benlahrech et al.PNAS ?
November 24, 2009 ?
vol. 106 ?
no. 47 ?