The Journal of Immunology
A Steady State of CD4+T Cell Memory Maturation and
Activation Is Established during Primary Subtype C
Pholo Maenetje,* Catherine Riou,* Joseph P. Casazza,†David Ambrozak,†
Brenna Hill,†Glenda Gray,‡Richard A. Koup,†Guy de Bruyn,‡and Clive M. Gray*
The functional integrity of CD4+T cells is crucial for well-orchestrated immunity and control of HIV-1 infection, but their selective
depletion during infection creates a paradox for understanding a protective response. We used multiparameter flow cytometry to
measure activation, memory maturation, and multiple functions of total and Ag-specific CD4+T cells in 14 HIV-1– and CMV-
coinfected individuals at 3 and 12 mo post HIV-1 infection. Primary HIV-1 infection was characterized by elevated levels of CD38,
HLA-DR, and Ki67 in total memory and Gag-specific CD4+and CD8+T cells. In both HIV-infected and 15 uninfected controls, the
frequency of activated cells was uniformly distributed among early differentiated (ED; CD45RO+CD27+), late differentiated
(CD45RO+CD272), and fully differentiated effector (CD45RO2CD272) memory CD4+T cells. In HIV-1–infected individuals,
activated CD4+T cells significantly correlated with viremia at 3 mo postinfection (r = 0.79, p = 0.0007) and also harbored more gag
provirus DNA copies than nonactivated cells (p = 0.04). Moreover, Gag-specific ED CD4+T cells inversely associated with plasma
viral load (r = 20.87, p , 0.0001). Overall, we show that low copy numbers of gag provirus and plasma RNA copies associated
with low CD4 activation as well as accumulation of ED HIV-specific CD4+memory. Significant positive correlations between 3 and
12 mo activation and memory events highlighted that a steady state of CD4+T cell activation and memory maturation was
established during primary infection and that these cells were unlikely to be involved in influencing the course of viremia in the
first 12 mo of HIV-1 infection.The Journal of Immunology, 2010, 184: 4926–4935.
tivation in HIV-1 infection may be either a direct consequence of
HIV Ag load or a consequence of exposure to other pathogens,
such as bacteria translocating from the gut (4–6), or from endemic
coinfections (7). Whether this occurs to the same degree in acute
and primary infection is unclear. Nevertheless, the resulting per-
sistent activation of the immune system is accompanied by loss of
peripheral CD4+T cells (2, 4, 8, 9) and a skewing of CD8+T cell
differentiation to a more mature memory phenotype that would
lead to accumulation of effector cells and premature terminal
differentiation (10, 11). For example, persistent exposure to HIV-1
drives the differentiation of central memory HIV-specific IL-2–
IV-1 infection is characterized by generalized immune
activation (1, 2), and the hyperactivation of T cells may
accelerate HIV disease progression (3, 4). Immune ac-
producing CD4+T cells to an IFN-g–producing effector memory
phenotype, and this latter phenotype has been associated with
higher levels of HIV viremia (12–14). It is thus important to un-
derstand the effect of immune activation on the maturation and
functionality of T cell memory and specifically within the CD4
Until recently, there has been little focus on unraveling the
relationship between activation and maturation of HIV-specific
CD4+T cell memory with viral control. HIV-specific CD4+T cells
have been shown to play an important role in maintaining func-
tional HIV-specific CD8+T cell responses (15, 16) and control of
viremia during chronic HIV infection (17–19). The maintenance
and preservation of HIV-specific CD4+T cells endowed with the
ability to produce multiple cytokines in individuals also coincides
with apparent protective immunity against HIV (20–22). However,
individuals with progressive HIV disease still exhibit significant
numbers of cytokine-producing HIV-specific CD4+T cells (23,
24), implying that the causative link between HIV-specific CD4+
T cell responses and viral control remain to be resolved.
To explore the association of CD4+T cells with in vivo viral
replication, we examined the association among markers of
memory maturation, activation, and polyfunction in total and HIV-
specific CD4+T cells in a prospective cohort of recently HIV-
infected individuals in South Africa. We show that profiles of
CD4+T cell memory maturation and activation reach an estab-
lished steady state early after HIV-1 infection and are unlikely to
be related to control of viremia.
Materials and Methods
Primary HIV-infection cohort. HIV-1–infected individuals were recruited
to a longitudinal cohort. All study participants were enrolled from an HIV-
negative cohort and tested prospectively for HIVinfection every 3 mo. The
*AIDS Research Unit, National Institute for Communicable Diseases;‡Perinatal HIV
Research Unit, University of the Witwatersrand, Johannesburg, South Africa; and
†Immunology Laboratory, Vaccine Research Center, National Institute of Allergy
and Infectious Disease, National Institutes of Health, Bethesda, MD 20892
Received for publication November 30, 2009. Accepted for publication February 27,
This work was supported in part by the National Institute of Allergy and Infectious
Diseases, National Institutes of Health, U.S. Department of Health and Human Serv-
ices Grant AI070079 (to G.d.B.), and a South African AIDS Vaccine Initiative grant
(to C.M.G.). P.M. was supported by a Columbia University-Southern Africa Fogarty
AIDS International Training Fellowship and C.R. was supported by the Canadian
African Prevention Trials network.
Address correspondence and reprint requests to Prof. Clive M. Gray, National In-
stitute for Communicable Diseases, Private Bag X4, Sandringham 2131, Johannes-
burg, Gauteng, South Africa. E-mail address: email@example.com
The online version of this article contains supplemental material.
Abbreviations used in this paper: ED, early differentiated; F, female; FD, fully
differentiated; Int, intermediate; IQR, interquartile range; LD, late differentiated;
M, male; NR, no response; NS, no sample; PID, participant identification number;
pVL, plasma viral load.
time postinfection was estimated as the midpoint of the last Ab-negative
and the first Ab-positive ELISA test prior to enrollment. None of the study
participants received antiretroviral therapy during the first 12 mo of in-
fection. All participants provided written informed consent for participa-
tion in this study. An additional cohort of 15 HIV-negative individuals
were used as control subjects and have been described elsewhere (11). The
clinical protocols were approved by the Human Research Ethics Com-
mittee (Medical) of the University of the Witwatersrand (M050832 and
M070249, respectively; Johannesburg, South Africa).
Plasma viral load and absolute CD4+T cell counts
Plasma HIV-1 RNA levels were quantified using the COBAS AMPLICOR
HIV-1 monitor test version 1.5 (Roche Diagnostic Systems, Somerville,
NJ). Absolute blood CD4+and CD8+T cell counts were measured using an
FACSCalibur flow cytometer (BD Biosciences, San Jose, CA) and ex-
pressed as cells/mm3.
Synthetic subtype C peptides
A panel of 66 overlapping peptides corresponding to the consensus sybtype
C sequence were made into a single pool covering the complete region of
Gag and resuspended in DMSO (Sigma-Aldrich, St. Louis, MO) as pre-
viously described (25). A final peptide concentration of 2 mg/ml/peptide
was used with ,1% DMSO concentration. A set of 138 peptides (15-mers)
overlapping by 11-aa residues corresponding to human CMV pp65 were
obtained from the National Institutes of Health AIDS Research and Ref-
erence Reagent Programs (Bethesda, MD). All prepared peptides were
stored at 280˚C prior to use.
PBMCs were isolated by standard Ficoll-Hypaque density gradient cen-
heat-activated FBS (Invitrogen, Paisely, U.K.) plus 10% DMSO, and stored
in liquid nitrogen until needed. Thawed PBMCs were washed twice with
RPMI 1640 supplemented with 10% heat-inactivated FBS, 100 U/ml
penicillin G, 100 mg/ml streptomycin sulfate, and 1.7 mM sodium gluta-
mate (R10). The cells were then rested in R10 at 37˚C and 5% CO2for 2 h
in the presence of 10 U/ml DNase I (Roche Diagnostic Systems) prior to
use in intracellular cytokine staining assays.
Cell stimulation and intracellular staining
Measurement of T cell activation. Thawed PBMCs were washed and
resuspendedat 2 3106cells/ml withR10 andstimulatedfor 6 h at 37˚Cand
in the presence of 1 mg/ml aCD28 and aCD49d costimulatory Abs (BD
Biosciences) and 10 mg/ml brefeldin A (Sigma-Aldrich, St. Louis, MO). A
negative control containing PBMCs and costimulatory Abs from the same
subject, but without the peptide mix, was also included for each assay. Fol-
lowing stimulation, cells were washed with PBS and surface stained with
violet reactive dye (Vivid; Molecular Probes, Eugene, OR) and a mixture of
mAbs containing HLA-DR Alexa 680, CD14 Pacific blue, CD19 Pacific
blue, CD57 QD565, CD8 QD655 (all conjugated under standard protocols),
CD27 PE-Cy5, and CD45RO Texas Red-PE (Beckman Coulter, Fullerton,
CA) for 20 min in the dark at 4˚C. The cells were then washed with PBS
containing 1% FBS and 0.1% sodium azide and permeabilized according to
the manufacturer’s instructions using a Cytofix/Cytoperm buffer kit (BD
Biosciences) and stained intracellularly with IFN-g and IL-2 PE (BD Bio-
sciences), CD3 APC-Cy7, CD38 APC, Ki67 FITC (BD Pharmingen, San
Diego, CA), and CD4 PE-Cy5.5 (Caltag Laboratories, Burlingame, CA).
After labeling, cells were washed and fixed in PBS containing 1% parafo-
maldehyde (Sigma-Aldrich) and stored at 4˚C prior to flow cytometry ac-
quisition within 24 h.
Detection of T cell polyfunction. Under the same conditions as explained
above, thawed PBMCs were stimulated for 6 h with or without HIV-Gag C
peptides (2 mg/ml), but in the presence of CD107 Alexa 680 (conjugated
under standard protocols) and 0.7 mg/ml monensin plus 1 mg/ml anti-
CD28, anti-CD49d, and 10 mg/ml brefeldin A. After washing, cells were
stained with a panel consisting of CD14 Pacific blue, CD19 Pacific blue,
CD57 QD656, CD8 QD655, CD27 PE-Cy5, CD45RO Texas Red, and the
viability violet reactive dye. Following incubation, cells were washed and
permeabilized using the Cytofix/Cytoperm kit (BD Biosciences) and then
stained intracellularly with CD3 APC-Cy7, IFN-g FITC, IL-2 APC, MIP-
1b PE, TNF-a PE-Cy7 (BD Pharmingen), and CD4 PE-Cy5.5. After la-
beling, cells were washed and fixed in PBS containing 1% para-
formaldehyde (Sigma-Aldrich) and stored at 4˚C prior to flow cytometry
acquisition within 24 h.
Flow cytometry analysis
Approximately 500,000–1,000,000 events were collected per sample on an
LSRII flow cytometer (BD Biosciences). Electronic compensation was
conducted with Ab capture beads (BD Biosciences) stained separately with
individual mAbs used in test samples. Data was analyzed with FlowJo
version 8.8.6 (Tree Star, Ashland, OR). Dead cells (Vivid+), monocytes
(CD14+), and B cells (CD19+) were removed from the analysis. Cells were
then gated on singlets, live CD3+, CD4+, CD8+, and memory cells, and then
on combinations of maturation and activation markers. A positive single
cytokine response was defined as .0.06% of memory CD4+T cell re-
sponses after background subtraction. This is consistent with other HIV-
specific CD4+T cell studies (26). For the Boolean gating analysis to detect
multiple cytokine responses, values .0.01% and twice the background
were considered as positive after background subtraction. A threshold of
0.01% has been previously applied for the analysis of CD4+T cells pro-
ducing multiple cytokines (21, 27).
Cell sorting was performed using the modification of the method de-
scribed by Douek et al. (28). HIV Gag-specific, activated, and nonactivated
total memory CD4+T cells were sorted using an FACSAria cell sorter (BD
Biosciences) at 70 Ib/in2. Activated cells were defined as cells expressing
CD38, Ki67, or HLA-DR, whereas nonactivated cells did not express any of
these markers. At least 40 million PBMCs were used for sorting in each
experiment, and sorted populations were consistently .98% pure. The in-
strument setup was performed according to the manufacturer’s instructions.
The level of HIV infection of these cells was then determined using real-
time PCR to quantify the amount of HIV-gag DNA per cell.
Quantitative real-time PCR
conical tubes, the supernatant was removed, and they were frozen at 220˚C
prior to use. Cells were then lysed in 25–100 ml 10 mM Tris buffer con-
input DNA for the quantification of HIV gag-DNA using the 59 nuclease
(TaqMan) assay with an ABI 7500 system (Applied Biosystems, Foster
City, CA) (28, 29). HIV gag-DNA degenerate primers and probes were
designed in conserved regions of subtype C gag genes found in the Los
Alamos HIV sequence database (www.hiv.lanl.gov/). Thegag C degenerate
primersand sequence were:gag-forward:
TAGCAGGA-39, gag-reverse: 59-GGYCCTTGTYTTATGTCCAA-39, and
BHQ1 (Inqaba Biotec, Pretoria, South Africa). For determining the cell
number per reaction, quantitative real-time PCR was performed simulta-
neously for albumin copy numbers using primers and probe sequences
previously described (28). Absolute quantitation of gag C and human al-
bumin copy numbers were performed using DNA standards and standard
curves generated from 10-fold serial dilutions starting at 106copies. Du-
plicate reactions were run and template copies calculated using ABI 7500
software (Applied Biosystems).
Statistical analysis and graphical presentation were performed using
expressed as median values and analyzed by the use of nonparametric sta-
tistics. Statistical significance was determined using the Mann-Whitney
U test, Wilcoxon paired t test, or Kruskal-Wallis ANOVA using Dunn’s test
for multiple comparisons. All tests were two-tailed, and a value of p , 0.05
was considered statistically significant. The relationship among the pro-
portions of memory subpopulations, immune activation with absolute CD4
Cohort characteristics and responses to Gag and CMV peptide
result (see Materials and Methods). Table I shows clinical char-
acteristics of the participants, the majority of whom were women,
stratified by change in viral load between 3 and 12 mo post-
infection. Two participants (PHR006 and PHR009) were lost to
follow-up, and there was no 12 mo viral load (Table I, N). The
median reduction of plasma viremia in the group over the first
9 mo of follow-up was 20.27 log10RNA copies/ml, and median
rate of absolute CD4 cell loss was 215 cells/mo (Table I). Viral
loads at baseline ranged from 2.6–5.88 log10 RNA copies/ml,
The Journal of Immunology4927
providing a variance of 3.28 log10to correlate with cell meas-
urements. One individual (PHR0012) showed an increase in vi-
remia in the first year (Table I, 4); six individuals showed
a change of 60.5 log10RNA copies/ml and were considered as
having reached a set point (Table I, d), and six individuals showed
reduced RNA copies/ml below 0.5 log10change between 3 and 12
mo (Table I, s). In terms of cellular responses, two patients
showed no CD4 response to Gag peptide pools (PHR009 and
PHR011), although one of these showed a positive response to
CMV. Fifteen HIV-uninfected individuals were used as control
subjects, and they all responded to CMV peptide pools; none re-
sponded to Gag peptides. Table I also shows the median CD4
counts and percent of Ag-specific responses to CMV and Gag in
both CD4+and CD8+T cells.
Defining memory maturation of HIV- and CMV-specific CD4+
T cells at 3 mo post HIV infection
We first wished to quantify the frequency of total and Ag-specific
after short-term stimulation of isolated PBMCs with CMV and
subtype C-based Gag peptide pools. Using the differentiation
markers CD45ROandCD27,wewereable todiscriminate fourand
five CD4+and CD8+T cell populations, respectively. Fig. 1A
shows representative plots of naive CD45RO2CD27+, early dif-
ferentiated (ED) memory CD45RO+CD27+, intermediate (Int)
memory CD45RO2CD27dim, late differentiated (LD) memory
CD45RO+CD272, and fully differentiated (FD) effector memory
CD45RO2CD272cells. Int memory was a unique population
within CD8+T cells, which we have previously shown to be
distinct from naive and effector cells according to levels of CD127
and CD57 (11). We purposely employed a conservative gating
strategy (30–32) to avoid misclassifying cells bearing dim ex-
pression of CD27 or CD45RO in the ED or LD compartment.
Our first level of analysis assessed the proportions of total CD4
and CD8 memory populations. Fig. 1B compares the proportions
of ED-, LD-, and FD-memory CD4 and ED-, LD-, Int-, and FD-
memory CD8 populations between HIV-infected and HIV-
uninfected controls, in which no differences were identified. We
examined Ag-specific cells using a combined IL-2/IFN-g readout
and identified distinct differences in maturation profiles between
HIV- and CMV-specific cells in both CD4+and CD8+T cells. For
the CD4 compartment (Fig. 1C), there was a significantly higher
proportion of ED-memory and lower proportions of LD-memory
Gag-specific cells relative to CMV-specific cells within the same
individuals (p = 0.0013 and p = 0.0006, respectively), which has
been shown previously (33). We noted from our parallel poly-
functional panel that the majority cytokine response was IFN-g,
which was 5- and 16-fold greater than IL-2 expression at 3 and
12 mo postinfection, respectively, in all subsets (Supplemental
Fig. 1). The proportions of ED or LD CMV-specific memory pop-
ulations at 3 mo showed no difference between HIV-infected and
between Gag- and CMV-specific cells were reflected in the Int-
and LD-memory populations (p = 0.00014 and p = 0.0008, re-
spectively), in which there was a higher proportion of the Int-
was comparable between HIV-infected and HIV-uninfected con-
trols. The differing numbers of absolute CD4 counts, which may
affect the difference in CMV- or Gag-specific CD4+memory sub-
sets, were taken into account, and there were no significant differ-
ences in memory subsets between individuals with CD4 counts
below or above 500 cells/ml (not shown). Also taken into consid-
eration were those individuals who went on to control viremia
(Fig. 1B, 1C, Table I, open symbols) and those who had reached
a viral set point (Table I, closed symbols). Overall, these data show
that during primary HIV infection, regardless of the course of vi-
remia, Gag-specific CD4+and CD8+T cells possessed a pre-
dominantly ED-memory maturation status compared with CMV-
To understand the profile of activated T cells during primary HIV
Ag-specific CD4+and CD8+T cells using a combination of Ki67,
HLA-DR, and CD38 markers. To ensure that short-term
Table I.Clinical characteristics of the study subjects stratified by viral load differences between 3 and 12 mo postinfection
PIDAge (y) Sex3 mo 12 mo3 mo 12 moCMVGAG CMVGAG
220 6 10
25 6 2
22.0 6 11
215 6 13
219 6 36
219 6 13
223 6 9
23 6 11
25 6 6
224 6 27
214 6 13
219 6 11
10 6 13
–3.58–5.07 3.88–4.55 20.81–0.12 0.13–0.26 0.14–0.21 0.22–0.37 0.1–0.72
38 64% F1118
F, female; IQR, interquartile range; M, male; NR, no response; NS, no sample; PID, participant identification number; pVL, plasma viral load.
4928 ACTIVATED AND MEMORY CD4 CELLS DURING HIV-1 INFECTION
stimulation did not result in increased expression of the activation
markers used, our preliminary experiments showed that staphy-
lococcal enterotoxin B stimulus had no effect on upregulating
expression of CD38, HLA-DR, and Ki67 on Ag-specific memory
CD4+T cells when compared with no staphylococcal enterotoxin
B stimulus. We also showed that the presence of brefeldin A
during the 6-h stimulation did not limit the expression of HLA-DR
and/or CD38 (data not shown). Fig. 2A shows representative ex-
pression plots of the three markers, in which distinct populations
of activated cells could be discerned. Fig. 2B confirms that total
memory CD4+and CD8+T cells from HIV-infected individuals
were significantly more activated than T cells from HIV-un-
infected controls. On Gag-specific CD4+T cells, there was ele-
vated CD38 and Ki67 expression levels compared with CMV-
specific cells within HIV-infected individuals (p = 0.0001 and p =
0.0002, respectively). In turn, CMV-specific CD4+T cells from
HIV-infected individuals were more activated than CMV-specific
cells from HIV-uninfected controls (CD38: p = 0.03; Ki67: p =
0.01). There was no difference for HLA-DR expression. In the
CD8 compartment, regardless of Ag specificity, cells were char-
acterized by higher expression of HLA-DR and Ki67 in HIV-in-
fected subjects when compared with HIV-negative controls (Fig.
2C). By comparing the CD4 and CD8 compartments, our data
show that: 1) total memory CD8+T cells were significantly more
activated than total memory CD4+T cells in HIV-infected in-
dividuals and HIV-uninfected controls for expression of CD38 and
HLA-DR (p , 0.0005 and p , 0.0001, respectively; data not
shown); and 2) Ki67 expression within Gag-specific CD4+T cells
was significantly (p , 0.0005) higher than in Gag-specific CD8+
T cells, indicating that HIV-specific CD4+T cells have a higher
turnover than CD8+T cells during primary HIV-1 infection. The
expression of activation markers on total and Gag-specific CD4+
T cells were unrelated to differences in the absolute number of
CD4+T cells (data not shown). Additionally, the differences in
activation status were unrelated to whether HIV-infected in-
dividuals went onto control initial viral load or had already
reached viral set point (Fig. 2B, 2C, open and closed symbols). In
summary, these data showed that Ag-specific CD8+T cells during
primary HIV infection were highly activated regardless of being
HIV-specific or CMV-specific and that Gag-specific CD4+T cells
were characterized by high surface expression of CD38 and Ki67
expression levels as compared with CMV-specific CD4+T cells.
Distribution of activation markers within CD4 memory subsets
To identify whether increased CD4 activation was preferentially
distributed within a specific memory subset and to understand the
relationship between activation and memory maturation, we
employed Boolean gating to associate memory maturation
CD8+memory T cell at 3 mo postinfection. PBMCs
from HIV-uninfected and HIV-infected individuals
(3 mo postinfection) were stimulated with Gag and/or
CMV peptide pools. The frequency of memory sub-
sets was quantified from IFN-g and/or IL-2–
producing T cells (Gag- and CMV-specific) and total
memory T cells (Total).Usingdifferentiation markers
different memory cell subsets: naive (CD45RO2
CD27+), ED (CD45RO+CD27+), Int (CD45RO2
CD27dim), LD (CD45RO+CD272), and FD effector
memory cells (CD45RO2CD272). A, Representative
LD, Int, and FD) in total and Ag-specific CD4+and
CD8+T cells. B, Distribution of CD4+and CD8+
T cell total memory subsets (ED, LD, Int, and FD)
in HIV-infected (n = 14) and uninfected controls (n =
15, n). C, Comparison of memory maturation profiles
of CMV- and Gag-specific T cells in HIV-infected
and -uninfected controls. Open circles (s) represent
individuals who showed an HIV viral load decline
of .0.5 log10RNA copies/ml between 3 and 12 mo
postinfection;closedcircles (d) representindividuals
who showed a viral load change within 60.5log10
RNA copies/ml between 3 and 12 mo. Open triangle
(4) represents the one participant who showed a viral
load increase .0.5 log10RNA copies/ml; open square
(N) represents two individuals whose viral evolution
could not be determined due to missing viral load data
at 12 mo. Statistical comparisons where determined
by a Mann-Whitney U nonparametric t test.
Differentiation profiles of CD4+and
The Journal of Immunology4929
phenotype with permutations of activation markers. Fig. 3A shows
proportions of triple-, double-, and single-positive and triple-nega-
tive marker combinations of CD38, HLA-DR, and Ki67 at the sin-
gle-cell level within total, ED-, LD-, andFD-memory CD4+T cells.
The activation profiles of total CD4+T cells were equal across each
memory subset in either HIV-uninfected controls or HIV-infected
individuals (Fig. 3A). However, when comparing activation profiles
between HIV-uninfected controls and -infected individuals, there
were significantly larger proportions of activated CD4+T cells
infected individuals, regardless of the stage of memory maturation.
Commensurate with the higher activation status, there were signif-
icantly less triple-negative cells (CD382HLADR2Ki672) in HIV-
infected individuals when compared with uninfected controls.
Collectively, these data show that activated CD4+T cells were: 1)
uniformly distributed across ED-, LD-, and FD-memory CD4+
T cells; and 2) more populous in HIV-infected individuals.
Similar observations were made for Ag-specific memory pop-
ulations, where no significant differences were observed within the
proportions of triple-positive (CD38+HLADR+Ki67+) or triple-
negative (CD382HLADR2Ki672) cells across each memory CD4
subset for either CMVor Gag specificities. However, Gag-specific
CD4+T cells were characterized by higher levels of activation
relative to CMV-specific cells in HIV-infected individuals, as
shown by the significant accumulation of activated cells that ex-
press two or three of the activation markers. The population of
highly activated cells was equally expanded across ED-, LD-, and
FD-memory populations (Fig. 3B), in which no significant dif-
ferences were found in activation profiles among ED, LD, or FD
CD4+T cells. This is collectively shown in Fig. 3C for ED- or LD-
memory cells, in which there was no significant correlation be-
tween memory maturation and triple-activated CD4+T cells.
Reciprocal associations between CD4+T cell activation and
memory maturation with HIV-1 viral load
To examine the relationship between CD4+T cell activation status
and memory maturation with viral load, we correlated activation
and memory profiles with viremia. For activation profiles, Fig. 4A
shows a significant positive correlation between the frequency of
CD38+HLADR+Ki67+total and Gag-specific CD4+T memory
and CD8+memory T cells at 3 mo post-
infection. Multiparameter flow cytometry
was used to determine the activation pro-
file in HIV-infected and -uninfected con-
trols based on the surface expression of
CD38 and HLA-DR and intracytoplasmic
memory cells. A, Representative dot plots
DR, and Ki67 in total and Ag-specific
CD4+and CD8+memory T cells. Com-
HIV-infected (n = 14) and HIV-uninfected
controls (n = 15, n) (B) and in Ag-specific
CD4+and CD8+T cells (C). Open circles
(s) represent individuals who showed an
HIV viral load decline of .0.5 log10RNA
copies/ml between 3 and 12 mo post-
infection; closed circles (d) represent in-
dividuals who showed a viral load change
3 and 12 mo. The open triangle (4) rep-
resents the one participant who showed
a viral load increase .0.5 log10 RNA
copies/ml; open squares (N) represent two
individuals whose viral evolution could
data at 12 mo. Statistical comparisons
were determined by either Mann-Whitney
U or Wilcoxon nonparametric t tests.
4930ACTIVATED AND MEMORY CD4 CELLS DURING HIV-1 INFECTION
cells with viremia at 3 mo postinfection (r = 0.79, p = 0.0007; and
r = 0.58, p = 0.035, respectively). As expected, when correlating
the triple-negative CD4+T cells (nonactivated cells), there were
significant inverse correlations for both total and Gag-specific
CD4+T cells (p = 0.0018, r = 20.75; and p = 0.012, r = 20.66,
respectively; data not shown). We also were able to show signif-
icant positive associations when using either double or single
activation marker expression (Supplemental Fig. 2). We wished to
understand if there was any grouping of individuals within the
correlations who were able to control HIV within the first 12 mo
of infection. It was evident (Fig. 4A) that there was a uniform
spread of highly activated total and Gag-specific CD4+memory
cells regardless of who was subsequently able to spontaneously
reduce viremia. HIV, like other lentivirus, can infect both dividing
and nondividing cells but requires T cell activation signals (34).
To directly test the susceptibility of total memory-activated CD4+
T cells to in vivo HIVinfection, we sorted populations of activated
(defined by the expression of any of the three activation markers,
CD38, Ki67, or HLA-DR, and CD45RO+) and nonactivated
(CD382HLADR2Ki672CD45RO+) memory CD4+T cells and
quantified the number of gag proviral DNA copies/cell in the
sorted populations. Fig. 4B shows that activated cells possessed
significantly higher quantities of gag proviral copies when com-
pared with sorted nonactivated cell fractions. Although found at
very low frequency of CD4+T cells (0.15 gag copies/cell maxi-
mum), these data directly show that activated total memory CD4+
T cells are preferred targets for in vivo HIV infection and that
activated memory CD4+T cells support ongoing viral replication.
For memory maturation, Fig. 4C shows a significant negative
correlation between the frequency of Gag-specific ED-memory
p , 0.0001) and a positive correlation with LD-memory cells (r =
0.85, p = 0.0002; not shown). Similarly, accounting for those in-
dividuals who were able to subsequently reduce viral loads in the
first 12 mo of infection (Fig. 4A, 4C, open symbols), there was
a uniform spread of total and Gag-specific ED- and LD-memory
CD4+T cells, suggesting that frequencies of these cell populations
were not determining the trajectory of viremia. Of note, the pro-
portion of total memory CD4+ED or LD T cells did not associate
with viral load.
Multifunctional profile of Gag-specific ED- and LD-memory
CD4+T cells at 3 mo postinfection
To explore the possibility that CD4+T cell multifunctionality may
have played a role in viral control rather than memory maturation,
we compared ED- and LD-memory Gag-specific CD4+T cells at
tivation profile of total (A) and Ag-specific cells (B) in ED-, LD-, and FD-memory CD4+T cell subsets in HIV-infected (n = 14) and HIV-uninfected (n = 15)
controls. The proportion of activated cells in each subset is represented as pie charts in which red corresponds to the frequency of cells expressing all three
CD38, HLADR, and Ki67 markers, orange corresponds to the frequency of cells expressing two of the three markers (i.e., CD38+HLADR+Ki672, CD382
HLADR+Ki67+, and CD38+HLADR2Ki67+), yellow represents the frequency of cells expressing at least one of the activation markers (CD38+HLADR2
Ki672, CD38-HLADR+Ki672and CD382HLADR2Ki67+), and green corresponds to the frequency of cells not expressing any of the markers (CD382
HLADR2Ki672, triple negative). Statistical comparisons were performed in SPICE using permutation analysis of the pie distributions on proportions of
single- double, triple-positive, and triple-negative cells. The p values are shown for comparisons between HIV2and HIV+individuals (A) and Gag-specific
and CMV-specific cells in HIV+individuals (B). C, Correlations between the proportion of Gag-specific triple-positive cells (CD38+HLADR+Ki67+) and the
proportion of Gag-specific ED (top panel) or Gag-specific LD (bottom panel).
Activation profiles of the different CD4+T cells memory subsets at 3 mo postinfection. Boolean gating analysis was used to assess the ac-
The Journal of Immunology4931
3 mo postinfection for different combinations of CD107, IFN-g,
IL-2, MIP-1b, and TNF-a. It has been shown for CD8+T cells
that the same combination of multifunctionality is related to viral
control (35, 36), and thus, we wished to explore whether such an
association existed with multifunctional CD4+T cells. Fig. 4D
shows that at 3 mo postinfection there was an equal multifunc-
tional profile (cells producing two or more cytokines simulta-
neously) between ED- and LD-memory Gag-specific CD4+
T cells. Taken together, these data show that ED- and LD-memory
Gag-specific CD4+T cells possess the same multifunctional pro-
file, and the significant association of ED-memory with low vi-
remia is independent and distinct from the activation and
multifunctional nature of CD4+T cell profiles.
Steady state of CD4+T cell activation and memory maturation
To identify whether the status of CD4 activation or memory
maturation at3mopostinfection mayhavereachedasteady stateor
set point (2), we correlated the frequency of activated and ED-
memory CD4+T cells at 3 and 12 mo postinfection. We defined
a steady state as the frequency of cells remaining within 20%
variation between two time points post HIV infection. To identify
whether activation and memory maturation had reached such
a steady state or may have had a role in determining the course of
viremia, we grouped participants into those having a viral load
decline of .0.5 log10RNA copies/ml and those who fell within
a 60.5 log10variation (Table I). Fig. 5A, 5B show significant
positive correlations between 3 and 12 mo measurements of triple
CD38+HLADR+Ki67+(r = 0.84, p = 0.003) and ED-memory (r =
0.94, p = 0.0003) CD4+T cells. These data suggest that the ac-
tivation and ED-memory status of CD4+T cells made at 3 mo
postinfection had reached a steady state early during primary in-
fection for the duration of the study period and that these cells are
unlikely to determine the course of viremia. When we performed
a similar analysis looking at the polyfunctional profile of Gag-
specific CD4+T cells (i.e., cells producing two or more cytokines
simultaneously), we found that there was no significant associa-
tion between measurements made at 3 and 12 mo, although there
was a negative trend to less functionality at 12 mo. This trend
disappeared when we made more stringent criteria of cells able to
produce three to four cytokines per cell (data not shown). In Fig.
5C, when accounting for those individuals who were able to
subsequently reduce viral loads (open symbols), there was a uni-
form spread of polyfunctional CD4+
T cells from these
between the proportion of CD38+HLADR+Ki67+in total memory and Gag-specific CD4+T cells with viral load at 3 mo. B, Differences in gag DNA copies/
103CD4+T cells between nonactivated (CD382HLADR2Ki672) and activated (cells expressing at last one of the activation Ags) total memory CD4+
T cells at 3 mo postinfection (n = 8). Each symbol represents the average of three measurements performed independently. Statistical significance was
determined by Wilcoxon signed-rank t test. C, Correlations between the proportion of ED-memory cells in total memory and Gag-specific CD4+T cells
with viral load at 3 mo. D, Comparison of proportion of multifunctional Gag-specific CD4+T cells between ED- and LD-memory cells. PBMCs from HIV-
infected individuals (n = 14) at 3 mo postinfection were stimulated with Gag peptide pools for 6 h and labeled with Abs against CD107, IFN-g, IL-2, MIP-
1b, TNF-a, CD3, CD4, CD8, CD45RO, and CD27. Using Boolean gating, the distribution of cells presenting any combination of functional profiles was
determined for ED- and LD-memory cells. Cells expressing two or more cytokines simultaneously were considered as multifunctional. Open circles (s)
represent individuals who showed an HIV viral load decline of .0.5 log10RNA copies/ml between 3 and 12 mo postinfection; closed circles (d) represent
individuals who showed a viral load change within 60.5 log10RNA copies/ml between 3 and 12 mo. The open triangle (4) represents the one participant
who showed a viral load increase .0.5 log10RNA copies/ml, and open squares (N) represent two individuals whose viral evolution could not be determined
due to missing viral load data at 12 mo. Statistical associations were performed by a two-tailed nonparametric Spearman rank correlation.
Correlations between memory CD4+T cell activation profiles and memory subsets with viral load at 3 mo postinfection. A, Correlations
4932ACTIVATED AND MEMORY CD4 CELLS DURING HIV-1 INFECTION
individuals, suggesting that polyfunctionality was independent
from the course of viremia.
The challenge of seeking what may constitute an anti-HIV pro-
tective function in the CD4+T cell compartment is that these cells
undergo early activation and depletion during infection, which is
considered a clinical hallmark of immunopathogenesis and im-
munosuppression. We studied the behavior of CD4 cells during
primary HIV-1 infection in antiretroviral-naive individuals to ex-
amine the balance between CD4+memory maturation and acti-
vation and the association among activation, memory status, and
viremia. Our central question was whether memory or activation
status of HIV-specific CD4+T cells had any impact on viral
control or the course of viremia in the first year of infection. We
provide evidence that ED HIV-specific memory CD4+T cells are
more populous in individuals with low viremia and that both the
activation and memory status reach a steady state during primary
We have previously observed that ED central memory CD8+
T cells correlated significantly with low viral set point (11), and
the presence of these cells during primary infection appeared to
provide the individual with some degree of immune advantage.
We also showed that the activation status was associated with the
stage of CD8 memory maturation. In our current work, we wished
to ask whether a population of ED CD4+T cells would also as-
sociate with viral control and whether activation events may be the
driving force behind memory differentiation in the CD4 com-
partment and with the inability to control viral load. It is clear
from studies in Sooty Mangabeys that attenuated immune acti-
vation most probably protects the natural host of SIV from pro-
gression to AIDS (37) and is probably related to no or little
microbial translocation that would lead to systemic hyperimmune
activation (6, 38) and increased viral replication (39). In humans,
the extensive immune activation during chronic infection is
thought to be the driver of pathogenesis, in which there is either
microbial translocation of bacterial DNA and LPS into the lym-
phatic circulation (40, 41) or coinfections with multiple endemic
pathogens (7), resulting in nonspecific hyperactivation of T cells
(42, 43). Whether this occurs during primary or acute infection
and whether high levels of viral replication during the initial
stages of HIV infection can result in Ag-specific immune activa-
tion is not clear. One fundamental question is whether immune
activation during primary HIV infection causes an imbalance in
CD4+T cell memory lineage.
When we looked at memory, our data showed that HIV-specific
CD4 memory had a predominantly ED phenotype during primary
infection relative to CMV-specific CD4+T cell subsets. Memory
phenotypes of Ag-experienced CD4+T cells is related to Ag ex-
posure and persistence (44), and the relative ED Gag-specific
CD4+T cells is likely related to time of HIV versus CMV expo-
sure. This notion may be supported by recent data showing that
Gag-specific CD4+T cells were more mature than CMV-specific
cells in chronic infection (45). In fact, the wide range of HIV loads
at 3 mo postinfection allowed us to associate ED- and LD-memory
CD4+T cells with viral replication. By so doing, we identified an
imbalance toward ED HIV-specific CD4+memory T cells with
low viremia, with LD CD4+memory T cells associating with high
viremia. These data would appear consistent with previous reports
suggesting that high numbers of less differentiated HIV-specific
CD4+T cells would favor a better clinical outcome (46, 47). When
we looked at the activation status of cells, there were strong
positive correlations between both activated HIV-specific and total
memory CD4+T cells with viremia, which appeared to be in-
dependent of memory lineage. ED-memory cells were as equally
activated as LD-memory cells, and there was a poor association
when memory maturation was correlated with activation, leading
us to conclude that activation events were not driving differenti-
ation of CD4 memory. It remains to be determined if similar
observations hold true for the CD8 compartment (48). Upon
sorting activated CD4+T cells, we were able to show that these
cells were more susceptible to in vivo HIV infection, although we
were unable to determine which memory subset was preferentially
infected, due to lack of material. Prior studies have shown that
memory CD4+T cell subsets (49) and CD4+CD572cells (29) are
more preferentially infected by HIVin vivo and support the notion
that ED-activated cells may be susceptible targets. Collectively,
these data confirm that CD4 activation events are directly pro-
portional to viral load and possibly infectivity. However, it re-
mains to be resolved whether activated CD4+T cells are harbors
of viral pools or whether higher levels of viral replication are
causing CD4 activation, of which either or both scenarios would
result in the significant correlations we observed.
Jointly, our observations represent an apparent paradox, as-
suming that memory maturation in the CD4 compartment is
thought tobelinked withcell activation. Theconundrum is thatlow
viremia, and possibly viral control, is associated with the main-
tenance of Gag-specific ED CD4+memory T cells, of which
memory CD4+T cells at 3 and 12 mo postinfection. A, Correlation be-
tween the proportions of triple-activated (CD38+HLADR+Ki67+) total
memory CD4+T cells at 3 and 12 mo postinfection. B, Correlation be-
tween the proportions of ED-memory CD4+T cells at 3 and 12 mo
postinfection. C, Correlation between the proportions of Gag-specific
CD4+T cells producing at least two cytokines at 3 and 12 mo post-
infection. Open circles (s) represent individuals who showed an HIV viral
load decline of .0.5 log10RNA copies/ml between 3 and 12 mo post-
infection; closed circles (d) represent individuals who showed a viral load
change within 60.5 log10RNA copies/ml between 3 and 12 mo. The open
triangle (4) represents the one participant who showed a viral load in-
crease .0.5 log10RNA copies/ml. Statistical associations were performed
by a two-tailed nonparametric Spearman rank correlation.
Relationship between activation and proportions of ED total
The Journal of Immunology 4933
almost half are activated and likely to be susceptible to infection.
There are four possible scenarios that could explain these ob-
servations: 1) there is preferential infection and depletion of ac-
tivated ED- or LD-memory populations of CD4+T cells, giving an
apparent equal distribution of activation markers across cell sub-
sets; 2) ED CD4 memory T cells, even in an activated state, are
more resistant to HIV infection; 3) there is no causative link be-
tween ED CD4 memory T cells and control of viral replication;
and 4) activation and memory differentiation are independent
events. Although there is evidence to show that ED-memory CD4+
T cells have a higher survival potential (50, 51), it is also likely
that these cells may be preferentially infected. Whatever the
scenario, it is unlikely that activation events per se push CD4
We propose, from our data, that the inverse association between
HIV-specific ED CD4+memory T cells and viral load is a re-
flection of Ag load and not a determining factor. This was sup-
ported when we found strong associations between 3 and 12 mo
activation and memory maturation phenotypes, suggesting that
levels of both activation and memory status were more a reflection
of pre-existing and established events prior to the analysis and
unlikely to be determining levels of viral replication. This ap-
peared to be independent of the course of viral loads in which, in
some individuals, there was spontaneous control of viremia de-
spite possessing populations of highly activated CD4+T cells. The
simplest interpretation from our data is that the dynamics of CD4+
T cell activation and memory maturation are determined by Ag
load, and the course of viremia over time is unrelated to these
events. Whatever the mechanisms, it was clear from our data that
activation and memory status within the individuals studied had
reached a steady state at some point during primary infection.
memory cells and which may partly account for viral control, we
assessed a five-functional profile that has been associated with viral
control when applied to CD8+T cells (35). As with the activation
status of cells, we found that the polyfunctional nature of HIV-
and the proportions of CD4+T cells possessing polyfunctional
characteristics did not strongly associate between 3 and 12 mo.
Although the lack of temporal association was most likely due to
loss ofCD4functionovertime,itwas possiblethatthetoolsusedto
dissect differences between ED- and LD-memory were not suffi-
identity of CD4 function may not be as simple as translating those
used to assess CD8 function. Casazza et al. (45) have shown, using
a similar phenotype panel, that the multifunctional nature of CMV-
specific CD4+T cells increased with memory maturation, and that,
was performed in chronic infection, and it is possible, judging from
our results, that the multifunctional nature of HIV-specific CD4+
T cells diminishes during primary infection, regardless of memory
In conclusion, our data show that low viral load associated with
both low activation levels and maintenance of ED HIV-specific
CD4+T cells. On closer examination, there was a steady state of
CD4 activation and memory maturation profiles, regardless of
viral load changes over time, suggesting that neither activation nor
memory status was influencing the course of viremia in the first
year of HIV infection in this cohort.
We thankthe participantsof the study and the clinical and laboratorystaff at
the Perinatal HIV Research Unit for expert patient care and handling of
specimens. We also thank Dr. Mario Roederer (National Institutes of
Health) for the generous gift of Q-dot labeled reagents as well as the SPICE
programusedto analyzemulticytokineprofiles (whichis freely availableon
The authors have no financial conflicts of interest.
1. Hazenberg, M. D., S. A. Otto, B. H. van Benthem, M. T. Roos, R. A. Coutinho,
J. M. Lange, D. Hamann, M. Prins, and F. Miedema. 2003. Persistent immune
activation in HIV-1 infection is associated with progression to AIDS. AIDS 17:
2. Deeks, S. G., C. M. Kitchen, L. Liu, H. Guo, R. Gascon, A. B. Narva ´ez, P. Hunt,
J. N. Martin, J. O. Kahn, J. Levy, et al. 2004. Immune activation set point during
early HIV infection predicts subsequent CD4+ T-cell changes independent of
viral load. Blood 104: 942–947.
3. Choudhary, S. K., N. Vrisekoop, C. A. Jansen, S. A. Otto, H. Schuitemaker,
F. Miedema, and D. Camerini. 2007. Low immune activation despite high levels
of pathogenic human immunodeficiency virus type 1 results in long-term
asymptomatic disease. J. Virol. 81: 8838–8842.
4. Hunt, P. W., J. Brenchley, E. Sinclair, J. M. McCune, M. Roland, K. Page-Shafer,
P. Hsue, B. Emu, M. Krone, H. Lampiris, et al. 2008. Relationship between
T cell activation and CD4+ T cell count in HIV-seropositive individuals with
undetectable plasma HIV RNA levels in the absence of therapy. J. Infect. Dis.
5. Doisne, J. M., A. Urrutia, C. Lacabaratz-Porret, C. Goujard, L. Meyer,
M. L. Chaix, M. Sinet, and A. Venet. 2004. CD8+ T cells specific for EBV,
cytomegalovirus, and influenza virus are activated during primary HIV infection.
J. Immunol. 173: 2410–2418.
6. Brenchley, J. M., D. A. Price, T. W. Schacker, T. E. Asher, G. Silvestri, S. Rao,
Z. Kazzaz, E. Bornstein, O. Lambotte, D. Altmann, et al. 2006. Microbial
translocation is a cause of systemic immune activation in chronic HIV infection.
Nat. Med. 12: 1365–1371.
7. Eggena, M. P., B. Barugahare, M. Okello, S. Mutyala, N. Jones, Y. Ma, C. Kityo,
P. Mugyenyi, and H. Cao. 2005. T cell activation in HIV-seropositive Ugandans:
differential associations with viral load, CD4+ T cell depletion, and coinfection.
J. Infect. Dis. 191: 694–701.
8. Sousa, A. E., J. Carneiro, M. Meier-Schellersheim, Z. Grossman, and
R. M. Victorino. 2002. CD4 T cell depletion is linked directly to immune ac-
tivation in the pathogenesis of HIV-1 and HIV-2 but only indirectly to the viral
load. J. Immunol. 169: 3400–3406.
9. Vajpayee, M., S. Kaushik, V. Sreenivas, K. Mojumdar, S. Mendiratta, and
N. K. Chauhan. 2009. Role of immune activation in CD4+ T-cell depletion in
HIV-1 infected Indian patients. Eur. J. Clin. Microbiol. Infect. Dis. 28: 69–73.
10. Papagno, L., C. A. Spina, A. Marchant, M. Salio, N. Rufer, S. Little, T. Dong,
G. Chesney, A. Waters, P. Easterbrook, et al. 2004. Immune activation and CD8+
T-cell differentiation towards senescence in HIV-1 infection. PLoS Biol. 2: E20.
11. Burgers, W. A., C. Riou, M. Mlotshwa, P. Maenetje, D. de Assis Rosa,
J. Brenchley, K. Mlisana, D. C. Douek, R. Koup, M. Roederer, et al; and the
CAPRISA 002 Acute Infection Study Team. 2009. Association of HIV-specific
and total CD8+ T memory phenotypes in subtype C HIV-1 infection with viral
set point. J. Immunol. 182: 4751–4761.
12. Younes, S. A., B. Yassine-Diab, A. R. Dumont, M. R. Boulassel, Z. Grossman,
J. P. Routy, and R. P. Sekaly. 2003. HIV-1 viremia prevents the establishment of
interleukin 2-producing HIV-specific memory CD4+ T cells endowed with
proliferative capacity. J. Exp. Med. 198: 1909–1922.
13. Tilton, J. C., M. R. Luskin, A. J. Johnson, M. Manion, C. W. Hallahan,
J. A. Metcalf, M. McLaughlin, R. T. Davey, Jr., and M. Connors. 2007. Changes
in paracrine interleukin-2 requirement, CCR7 expression, frequency, and cyto-
kine secretion of human immunodeficiency virus-specific CD4+ T cells are
a consequence of antigen load. J. Virol. 81: 2713–2725.
14. Palmer, B. E., E. Boritz, and C. C. Wilson. 2004. Effects of sustained HIV-1
plasma viremia on HIV-1 Gag-specific CD4+ T cell maturation and function. J.
Immunol. 172: 3337–3347.
15. Kumaraguru, U., K. Banerjee, and B. T. Rouse. 2005. In vivo rescue of defective
memory CD8+ T cells by cognate helper T cells. J. Leukoc. Biol. 78: 879–887.
16. Ramsburg, E. A., J. M. Publicover, D. Coppock, and J. K. Rose. 2007. Re-
quirement for CD4 T cell help in maintenance of memory CD8 T cell responses
is epitope dependent. J. Immunol. 178: 6350–6358.
17. Rosenberg, E. S., J. M. Billingsley, A. M. Caliendo, S. L. Boswell, P. E. Sax,
S. A. Kalams, and B. D. Walker. 1997. Vigorous HIV-1-specific CD4+ T cell
responses associated with control of viremia. Science 278: 1447–1450.
18. Boaz, M. J., A. Waters, S. Murad, P. J. Easterbrook, and A. Vyakarnam. 2002.
Presence of HIV-1 Gag-specific IFN-gamma+IL-2+ and CD28+IL-2+ CD4
T cell responses is associated with nonprogression in HIV-1 infection. J. Im-
munol. 169: 6376–6385.
19. Boritz, E., B. E. Palmer, and C. C. Wilson. 2004. Human immunodeficiency
virus type 1 (HIV-1)-specific CD4+ T cells that proliferate in vitro detected in
samples from most viremic subjects and inversely associated with plasma HIV-1
levels. J. Virol. 78: 12638–12646.
20. Potter, S. J., C. Lacabaratz, O. Lambotte, S. Perez-Patrigeon, B. Vingert,
M. Sinet, J. H. Colle, A. Urrutia, D. Scott-Algara, F. Boufassa, et al. 2007.
4934ACTIVATED AND MEMORY CD4 CELLS DURING HIV-1 INFECTION
Preserved central memory and activated effector memory CD4+ T-cell subsets in Download full-text
human immunodeficiency virus controllers: an ANRS EP36 study. J. Virol. 81:
21. Kannanganat, S., C. Ibegbu, L. Chennareddi, H. L. Robinson, and R. R. Amara.
2007. Multiple-cytokine-producing antiviral CD4 T cells are functionally su-
perior to single-cytokine-producing cells. J. Virol. 81: 8468–8476.
22. Pereyra, F., M. M. Addo, D. E. Kaufmann, Y. Liu, T. Miura, A. Rathod, B. Baker,
A. Trocha, R. Rosenberg, E. Mackey, et al. 2008. Genetic and immunologic
heterogeneity among persons who control HIV infection in the absence of
therapy. J. Infect. Dis. 197: 563–571.
23. Pitcher, C. J., C. Quittner, D. M. Peterson, M. Connors, R. A. Koup, V. C. Maino,
and L. J. Picker. 1999. HIV-1-specific CD4+ T cells are detectable in most in-
dividuals with active HIV-1 infection, but decline with prolonged viral sup-
pression. Nat. Med. 5: 518–525.
24. Jansen, C. A., I. M. De Cuyper, B. Hooibrink, A. K. van der Bij, D. van Baarle,
and F. Miedema. 2006. Prognostic value of HIV-1 Gag-specific CD4+ T-cell
responses for progression to AIDS analyzed in a prospective cohort study. Blood
25. Gray, C. M., M. Mlotshwa, C. Riou, T. Mathebula, D. de Assis Rosa,
T. Mashishi, C. Seoighe, N. Ngandu, F. van Loggerenberg, L. Morris, et al;
CAPRISA 002 Acute Infection Study Team. 2009. Human immunodeficiency
virus-specific gamma interferon enzyme-linked immunospot assay responses
targeting specific regions of the proteome during primary subtype C infection are
poor predictors of the course of viremia and set point. J. Virol. 83: 470–478.
26. Foxall, R. B., C. S. Cortesa ˜o, A. S. Albuquerque, R. S. Soares, R. M. Victorino,
and A. E. Sousa. 2008. Gag-specific CD4+ T-cell frequency is inversely corre-
lated with proviral load and directly correlated with immune activation in in-
fection with human immunodeficiency virus type 2 (HIV-2) but not HIV-1.
J. Virol. 82: 9795–9799.
27. Yamamoto, T., N. Iwamoto, H. Yamamoto, T. Tsukamoto, T. Kuwano,
A. Takeda, M. Kawada, Y. Tsunetsugu-Yokota, and T. Matano. 2009. Poly-
functional CD4+ T-cell induction in neutralizing antibody-triggered control of
simian immunodeficiency virus infection. J. Virol. 83: 5514–5524.
28. Douek, D. C., J. M. Brenchley, M. R. Betts, D. R. Ambrozak, B. J. Hill,
Y. Okamoto, J. P. Casazza, J. Kuruppu, K. Kunstman, S. Wolinsky, et al. 2002.
HIV preferentially infects HIV-specific CD4+ T cells. Nature 417: 95–98.
29. Brenchley, J. M., B. J. Hill, D. R. Ambrozak, D. A. Price, F. J. Guenaga,
J. P. Casazza, J. Kuruppu, J. Yazdani, S. A. Migueles, M. Connors, et al. 2004.
T-cell subsets that harbor human immunodeficiency virus (HIV) in vivo: im-
plications for HIV pathogenesis. J. Virol. 78: 1160–1168.
30. Song, K., R. L. Rabin, B. J. Hill, S. C. De Rosa, S. P. Perfetto, H. H. Zhang,
J. F. Foley, J. S. Reiner, J. Liu, J. J. Mattapallil, et al. 2005. Characterization of
subsets of CD4+ memory T cells reveals early branched pathways of T cell
differentiation in humans. Proc. Natl. Acad. Sci. U.S.A. 102: 7916–7921.
31. Chattopadhyay, P. K., J. Yu, and M. Roederer. 2006. Live-cell assay to detect an-
tigen-specific CD4+ T-cell responses by CD154 expression. Nat. Protoc. 1: 1–6.
32. Petrovas, C., B. Chaon, D. R. Ambrozak, D. A. Price, J. J. Melenhorst, B. J. Hill,
C. Geldmacher, J. P. Casazza, P. K. Chattopadhyay, M. Roederer, et al. 2009.
Differential association of programmed death-1 and CD57 with ex vivo survival
of CD8+ T cells in HIV infection. J. Immunol. 183: 1120–1132.
33. Yue, F. Y., C. M. Kovacs, R. C. Dimayuga, P. Parks, and M. A. Ostrowski. 2004.
HIV-1-specific memory CD4+ T cells are phenotypically less mature than cy-
tomegalovirus-specific memory CD4+ T cells. J. Immunol. 172: 2476–2486.
34. Oswald-Richter, K., S. M. Grill, M. Leelawong, and D. Unutmaz. 2004. HIV
infection of primary human T cells is determined by tunable thresholds of T cell
activation. Eur. J. Immunol. 34: 1705–1714.
35. Betts, M. R., M. C. Nason, S. M. West, S. C. De Rosa, S. A. Migueles,
J. Abraham, M. M. Lederman, J. M. Benito, P. A. Goepfert, M. Connors, et al.
2006. HIV nonprogressors preferentially maintain highly functional HIV-specific
CD8+ T cells. Blood 107: 4781–4789.
36. Almeida, J. R., D. A. Price, L. Papagno, Z. A. Arkoub, D. Sauce, E. Bornstein,
T. E. Asher, A. Samri, A. Schnuriger, I. Theodorou, et al. 2007. Superior control
of HIV-1 replication by CD8+ T cells is reflected by their avidity, poly-
functionality, and clonal turnover. J. Exp. Med. 204: 2473–2485.
37. Sodora, D. L., J. S. Allan, C. Apetrei, J. M. Brenchley, D. C. Douek, J. G. Else,
J. D. Estes, B. H. Hahn, V. M. Hirsch, A. Kaur, et al. 2009. Toward an AIDS
vaccine: lessons from natural simian immunodeficiency virus infections of Af-
rican nonhuman primate hosts. Nat. Med. 15: 861–865.
38. Gordon, S. N., N. R. Klatt, S. E. Bosinger, J. M. Brenchley, J. M. Milush,
J. C. Engram, R. M. Dunham, M. Paiardini, S. Klucking, A. Danesh, et al. 2007.
Severe depletion of mucosal CD4+ T cells in AIDS-free simian immunodefi-
ciency virus-infected sooty mangabeys. J. Immunol. 179: 3026–3034.
39. Thibault, S., M. Imbeault, M. R. Tardif, and M. J. Tremblay. 2009. TLR5
stimulation is sufficient to trigger reactivation of latent HIV-1 provirus in
T lymphoid cells and activate virus gene expression in central memory CD4+
T cells. Virology 389: 20–25.
40. Jiang, W., M. M. Lederman, P. Hunt, S. F. Sieg, K. Haley, B. Rodriguez,
A. Landay, J. Martin, E. Sinclair, A. I. Asher, et al. 2009. Plasma levels of
bacterial DNA correlate with immune activation and the magnitude of immune
restoration in persons with antiretroviral-treated HIV infection. J. Infect. Dis.
41. Brenchley, J. M., and D. C. Douek. 2008. HIV infection and the gastrointestinal
immune system. Mucosal Immunol. 1: 23–30.
42. Boasso, A., and G. M. Shearer. 2008. Chronic innate immune activation as
a cause of HIV-1 immunopathogenesis. Clin. Immunol. 126: 235–242.
43. Brenchley, J. M., and D. C. Douek. 2008. The mucosal barrier and immune
activation in HIV pathogenesis. Curr. Opin. HIV AIDS 3: 356–361.
44. Harari, A., F. Vallelian, and G. Pantaleo. 2004. Phenotypic heterogeneity of
antigen-specific CD4 T cells under different conditions of antigen persistence
and antigen load. Eur. J. Immunol. 34: 3525–3533.
45. Casazza, J. P., J. M. Brenchley, B. J. Hill, R. Ayana, D. Ambrozak, M. Roederer,
D. C. Douek, M. R. Betts, and R. A. Koup. 2009. Autocrine production of beta-
chemokines protects CMV-Specific CD4 T cells from HIV infection. PLoS
Pathog. 5: e1000646.
46. Emu, B., E. Sinclair, D. Favre, W. J. Moretto, P. Hsue, R. Hoh, J. N. Martin,
D. F. Nixon, J. M. McCune, and S. G. Deeks. 2005. Phenotypic, functional, and
kinetic parameters associated with apparent T-cell control of human immuno-
deficiency virus replication in individuals with and without antiretroviral treat-
ment. J. Virol. 79: 14169–14178.
47. Barbour, J. D., L. C. Ndhlovu, Q. Xuan Tan, T. Ho, L. Epling, B. M. Bredt,
J. A. Levy, F. M. Hecht, and E. Sinclair. 2009. High CD8+ T cell activation
marks a less differentiated HIV-1 specific CD8+ T cell response that is not al-
tered by suppression of viral replication. PLoS One 4: e4408.
48. Catalfamo, M., M. Di Mascio, Z. Hu, S. Srinivasula, V. Thaker, J. Adelsberger,
A. Rupert, M. Baseler, Y. Tagaya, G. Roby, et al. 2008. HIV infection-associated
immune activation occurs by two distinct pathways that differentially affect CD4
and CD8 T cells. Proc. Natl. Acad. Sci. U.S.A. 105: 19851–19856.
49. Dai, J., L. M. Agosto, C. Baytop, J. J. Yu, M. J. Pace, M. K. Liszewski, and
U. O’Doherty. 2009. Human immunodeficiency virus integrates directly into
naive resting CD4+ T cells but enters naive cells less efficiently than memory
cells. J. Virol. 83: 4528–4537.
50. Caserta, S., and R. Zamoyska. 2007. Memories are made of this: synergy of
T cell receptor and cytokine signals in CD4(+) central memory cell survival.
Trends Immunol. 28: 245–248.
51. van Grevenynghe, J., R. Halwani, N. Chomont, P. Ancuta, Y. Peretz, A. Tanel,
F. A. Procopio, Y. shi, E. A. Said, E. K. Haddad, and R. P. Sekaly. 2008.
Lymph node architecture collapse and consequent modulation of FOXO3a
pathway on memory T- and B-cells during HIV infection. Semin. Immunol. 20:
The Journal of Immunology 4935