www.elsevier.com/locate/yviro

Virology 337 (20

Phenotypic and kinetic analysis of effective simian–human

immunodeficiency virus-specific T cell responses in

DNA- and fowlpox virus-vaccinated macaques

Ivan Stratov, C. Jane Dale, Stephen J. Kent*

Department of Microbiology and Immunology, University of Melbourne, Parkville, Victoria 3010, Australia

Received 21 October 2004; returned to author for revision 14 April 2005; accepted 22 April 2005

Available online 23 May 2005

Abstract

Although T cell immunity is important in the control of HIV-1 infection, the characteristics of effective HIV-specific T cell responses are

unclear. We previously observed protection from virulent SHIV challenges in macaques administered priming with DNA vaccines and

boosting with recombinant fowlpox viruses expressing shared SIV Gag antigens. We therefore performed a detailed kinetic and phenotypic

study of the T cell immunity induced by these vaccines prior to and following SHIV challenge utilizing intracellular cytokine staining. Pigtail

macaques vaccinated intramuscularly with DNA/recombinant fowlpox virus exhibited a coordinated induction of first Gag-specific CD4 T

cell responses and then a week later Gag-specific CD8 T cell responses following the fowlpox virus boost. Overall, the magnitude and timing

of the peak CD8 T cell responses following challenge was significantly associated with reductions in SHIV viremia following pathogenic

challenge. After pathogenic lentiviral challenge, virus-specific effector memory T cells derived from animals controlling SHIV infection

recognized a broad array of epitopes, expressed multiple effector cytokines and rapidly recognized virus-exposed cells ex vivo. These results

shed light on some of the requirements for T cells in the control of pathogenic lentiviral infections.

D 2005 Elsevier Inc. All rights reserved.

Keywords: HIV; Vaccine; Phenotype; T cell; Kinetics; Cytokine ICS; Memory

Introduction

The development of an effective vaccine against HIV-1 is

one of humanity’s greatest challenges. Generating effective

immunity to HIV is hampered by the virus’s capacity to

evade innate, humoral and cell-mediated immunity, the

likely need to generate effective mucosal immunity and the

lack of a precise immune correlate of protection (Nabel,

2001; Stratov et al., 2004).

Although induction of broad neutralizing antibody to

HIV by vaccines would be ideal, this has not been

achievable to date. Many current HIV-1 vaccine research

efforts endeavor to primarily induce effective HIV-specific

T cell immunity. CD8+ cytotoxic T lymphocytes (CTLs)

are important in controlling HIV-1 infection of humans and

0042-6822/$ - see front matter D 2005 Elsevier Inc. All rights reserved.

doi:10.1016/j.virol.2005.04.023

* Corresponding author. Fax: +61 383443846.

E-mail address: skent@unimelb.edu.au (S.J. Kent).

SIV/SHIV infection in macaques and generally correlate

with protective immunity in macaque-SHIV studies (Bar-

ouch et al., 2000; Borrow et al., 1994; Koup et al., 1994;

Ogg et al., 1998). There are several strategies to induce

high-level T cell immunity, including priming with DNA

with or without boosting with recombinant viral/bacterial

vectors or cytokines. Promising data have been obtained in

non-human primate models, showing protection from AIDS

and �100-fold reductions in SHIV viral load following the

induction of SHIV-specific T cell immunity with DNA/

poxvirus prime/boost vaccination regimens (Amara et al.,

2001; Dale et al., 2004a). DNA prime/poxvirus boost

vaccine regimens have provided less durable protection

against SIV in macaques (Horton et al., 2002).

HIV-specific Tcell immunity involves both CD4 and CD8

T cells. The absence of functional HIV-specific T cells in

HIV-infected humans is recognized as a key defect in

enabling more efficient control of viremia (Rosenberg et

05) 222 – 234

I. Stratov et al. / Virology 337 (2005) 222–234 223

al., 1997). The safe induction of both CD4 and CD8 Tcells is

a critical goal of HIV vaccine studies. HIV-specific CD4 T

cells are likely to enhance the durability, memory levels and

generation of HIV-specific CD8 T cells (Grakoui et al., 2003;

Kalams et al., 1999). The temporal generation of vaccine-

specific CD4 T cells would ideally precede and assist the

generation of CD8 T cell responses, but this has not been

previously analyzed in HIV vaccine studies of outbred

macaques.

The significant limitations of relying only on HIV-

specific CD8 T cells in controlling HIV has become clear

in recent years. In vivo selection for mutant HIVor SIV viral

variants, which escape CD8 T cell responses, is the rule.

Although T cell escape mutants incur a fitness cost

(Friedrich et al., 2004; Leslie et al., 2004), progression to

AIDS is likely to eventually ensue (Barouch et al., 2002). A

narrowly directed CD8 T cell response is likely to be more

susceptible to T cell escape. The lack of virus-specific CD4

T cells may enhance the generation of T cell escape in HCV-

primate models (Grakoui et al., 2003).

The functional phenotype of the CTL response and its

association with outcome is of great interest (McMichael

and Rowland-Jones, 2001). The effector/memory phenotype

of the CTL response is likely to determine both the rapidity

of the CTL response upon infectious virus challenge and the

durability of T cell memory in the setting of long latency

between vaccination and virus exposure. Effector memory

CD8 + T cells (i.e., cells terminally differentiated for direct

anti-pathogen effector activity) have been defined by panels

of surface molecules and are phenotypically CD62L, CCR7

and CD28 negative, CD95 positive and express varying

levels of CD45RA (Pitcher et al., 2002). The cytokine

secretion profile of cognate CTL responses is also a

critically important functional characteristic of effective T

cells (Pantaleo and Koup, 2004). Interferon-gamma (IFNg)

production by CTLs is closely linked to virus-specific

immunity but IFNg production during HIV-1 infection may

still occur even though these cells are defective in their

ability to proliferate or lyse target cells (Appay et al., 2000).

Production of a broader array of cytokines, such as TNFa,

IL-2 and perforin, may be an important determinant of more

effective CD4 T-helper and CD8 CTL responses (McMi-

chael and Rowland-Jones, 2001).

Although most in vitro studies of HIV vaccines analyze

T cell responses to peptide epitopes or peptide pools, for

protective immunity to be optimally effective, T cell

responses should rapidly recognize processed intact virions,

preferably at mucosal surfaces. Phenotyping T cell

responses to processed virions by intracellular cytokine

staining (ICS), which is a short (generally 7 h) assay,

provides an opportunity to study rapid processing and

presentation of the virus itself.

To further understand protective immunity to HIV, we

first evaluated the kinetics of Gag-specific CD4 and CD8 T

cell responses, both before and after SHIV challenge, in

pigtail macaques vaccinated with standard IM regimens of

DNA and rFPV (Dale et al., 2004a). We subsequently

analyzed the relationship between the virologic outcome of

acute SHIV challenge with the magnitude and timing of the

Gag-specific CD8 T cell response after challenge. To

characterize the SIV/HIV-specific T cell immunity in

successfully immunized macaques further we then analyzed

the T cell immunity in individual animals for their surface

marker phenotype, the number of epitopes recognized, the

cytokine secretion profile and their ability to recognize

intact virions of the T cell responses.

Results

Vaccination and SHIV challenge studies

We studied the kinetics and phenotype of SIV/HIV in

pigtail macaques immunized with DNA prime/FPV boost

vaccine regimens involved in three separate studies (Dale et

al., 2004a; De Rose et al., 2005; Kent et al., submitted). Two

of these studies involved SHIV challenge, the first examin-

ing 30 animals receiving combinations of IM DNA and

rFPV vaccinations and a highly pathogenic intrarectal

CXCR4-utilizing SHIVmn229 challenge (Dale et al.,

2004a). The second study involving immunization of 20

macaques with combinations of IM and mucosally delivered

DNA and rFPV vaccines followed by a moderately

pathogenic CCR5-utilizing SHIVSF162P3 challenge. Animals

involved in the 3rd study were immunized IM with various

doses of HIV-1 DNA and rFPV vaccines and were studied

for the cytokine-secretion profile of T cells described below

(De Rose et al., 2005). A summary of the regimens and

outcomes is shown in Table 1.

Based on our previous studies (Dale et al., 2004b; Kent et

al., 1998), we anticipated the most successful vaccination

regimens would be those that used IM priming with two or

three DNA vaccinations followed by one or more rFPV

boosts. This prediction proved true (Table 1). In trial 1, the six

animals vaccinated IM utilizing two DNA priming and with

non-cytokine expressing rFPV boosting (group 1) induced

peak SIVgag-specific CD8+ T lymphocytes capable of

expressing IFNg ranging from 0.3% to 4.7% of all CD8+ T

cells, early after the rFPV boost, as previously reported (Dale

et al., 2004a). After DNA vaccine priming alone, CD8 and

CD4 T cell responses were low (<0.3%) or undetectable.

Kinetics of Gag-specific T cell responses in blood following

prime/boost vaccination

We first studied the kinetic relationship of the SIVgag-

specific CD4 and CD8 Tcell responses in detail early after the

rFPV boost vaccine in fresh peripheral blood samples from

the six animals in group 1 of trial 1 vaccinated with our most

effective, DNA/rFPV prime/boost regimen. Five of the six

vaccinated animals demonstrated a remarkably coordinated

induction of Gag-specific T cell immune responses (see Fig.

Table 1

Summary of three separate macaque DNA/fowlpox virus vaccine trials

Trial Group Vaccination

regimenaRoute of vaccine

(DNA/FPV)

n Animal ID numbers Peak Gag-specific

CD8/CD4 T cell

responsesb

Challenge virus Reduction

in acute

SHIV VLc

Reference

1 1 2 DNA/FPV IM/IMd 6 17105, 4247, 4277,

4290, 4295, 4296

2.4/0.97 SHIVmn229

IR

1.48 Dale et al.

(2004a)

2 1 DNA/FPV IM/IM 6 4236, 4237, 4262,

4379, 4380, 4382

0.04/0.48 0.84

3 3 DNA IM/IM 6 4241, 4297, 4299,

4301, 4386, 4523

0.08/0.02 1.53

4 2 DNA/FPV-IFNg IM/IM 6 4243, 4251, 4663,

4664, 4667, 4668

0.07/0.4 0.88

5 Controls N/A 6 1253, 1281, 4194,

4265, 4529, H3

0/0 N/A

2 1 3DNA/2FPV IM/IM 4 4245, 4246, 4254,

4381

2.33/0.53 SHIVSF162P3

IVag

0.64 Kent et al.

(submitted for

publication)

2 3DNA/2FPV IM/IN 4 4240, 4292, 4648,

4666

0.06/0.11 0.88

3 3DNA/2FPV IM/IR 4 3790, 4293, 4532,

4635

0.02/0.04 0.86

4 3DNA/2FPV IN/IN 4 4238, 4253, 4258,

4259

0.1/0.07 0.26

5 Controls N/A 8 4250, 4658, 9E, 18E,

19E, 21E, 23E, 24E

0/0 N/A

3 3 DNA/FPV IM/IM 12e 200B, 1D52, 710D,

1302, 6820, 4C05,

2976, 3560, 0C18,

5D2F, 505D, 207F

0.34/1.17 Nil N/A De Rose

et al.

(2005)

a Refers to number of DNA primes and FPV boost vaccines. All vaccines were administered at 3–4 weekly intervals.b Mean % T cell response in group.c Mean reduction in peak SHIV plasma RNA (log10 copies/ml plasma) in group compared to control group.d IM = intramuscular; IR = intrarectal; IN = intranasal; IVag = intravaginal.e All animals received the same HIV-1 subtype AE DNA and rFPV vaccines at varying doses.

I. Stratov et al. / Virology 337 (2005) 222–234224

1A) with an early CD4 T cell response 1 week after the rFPV

boost (mean 1.0%; range 0.2–2.1%) followed by a larger

(week 2) CD8 Tcell response (mean 2.4%; range 0.9–4.7%).

To confirm this kinetic relationship of the virus-specific

CD4 and CD8 Tcell response, we again studied the temporal

association of SIV Gag-specific T cell responses in the four

animals vaccinated IMwith the sameDNAand rFPVvaccines

in trial 2. The pattern of an initial Gag-specific CD4 T cell

response followed by a CD8 T cell response was repeated

(Fig. 1B). In trial 2, a second boost with rFPV resulted in only

a modest increase in SIVgag-specific CD8 T cells, which did

not reach the level obtained from the initial boosting, likely

due to immunity to the rFPV vector (Dale et al., 2004b).

Kinetics of Gag-specific T cell responses following SHIV

challenge

We then studied the kinetics of recall Gag-specific T cell

responses following SHIV challenge. Following pathogenic

challenge with CXCR4-utilising SHIVmn229, the six animals

in group 1 of trial 1 vaccinated IM twice with DNA and

boosted with rFPV preserved peripheral CD4 T cells and

had lower peak and set point plasma SIV viral loads than

control animals (Dale et al., 2004a). A large anamnestic SIV

Gag-specific CD8 T cell response (mean of 11.7%) was

present in these animals. This response peaked 3 weeks after

challenge and 1 week after the peak of the plasma SHIV

RNA (mean 6.67 log10 copies/mL; Fig. 1C). We also

studied the kinetics of T cell immunity in the animals

vaccinated IM with DNA and rFPV in trial 1 following the

SHIVsf162p3 challenge. A similar pattern was seen in the

four animals in group 1 of trial 2 challenged with CCR5-

utilising SHIVSF162P3. Peak plasma SIV viral load was 5.50

log10 copies/mL at week 3, closely followed by peak Gag-

specific CD8 T cell responses of 7.5% at week 3.5 (data not

shown). Taken together, these kinetic data are consistent

with previous data on the role of CD8 T cells in control of

acute infection (Borrow et al., 1994; Koup et al., 1994).

Post-challenge SIV Gag-specific CD4 T cell responses were

much lower than the CD8 T cell responses and there was no

temporal relationship to the peak viral load or SHIV-specific

CD8 T cell responses (Fig. 1C).

Relationship of CD8 T cell immunity after challenge with

virologic outcome

The temporal relationship between the Gag-specific CD8

T cell response and peak viral load in group 1 animals in

Fig. 1. Kinetics of mean SIV Gag-specific CD4 (closed circles) and CD8 T cell responses in macaque blood samples (closed diamonds). (A) Following rFPV

boosting, an early (week 1) Gag-specific mean (TSE) CD4 T cell response is detected, followed by a larger and later (week 2) mean Gag-specific CD8 T cell

response detected in the six monkeys in group 1 from trial 1 vaccinated with the standard IM DNA and rFPV regimen. (B) The same temporal pattern of mean

SIV Gag-specific CD4 and CD8 T cell immunity is repeated in the four animals from group 1 in trial 2 vaccinated the same IM DNA and rFPV vaccines as for

trial 1. Note: animals in trial 2 were given a second rFPV boost 4 weeks after the initial rFPV boost. (C) Following SHIVmn229 challenge in trial 1, mean SIV

Gag-specific CD8 T cell responses peak 1 week after peak SHIV plasma viral load (open squares) in the six animals from group 1 vaccinated with the standard

IM DNA and rFPV regimen.

I. Stratov et al. / Virology 337 (2005) 222–234 225

both SHIV challenge trials 1 and 2 suggested that there may

be a correlation between either the timing or strength of T

cell immunity and outcome of challenge. We therefore

analyzed the timing and magnitude of the peak CD8 T cell

response in comparison to the peak level of acute SHIV

viremia. To broaden and improve power for these analyses

we studied all 50 animals (including controls) challenged

with either SHIVmn229 or SHIVSF162P3 across both trials 1

and 2.

The strength of the peak peripheral blood CD8 T cell

post-challenge was indeed significantly inversely correlated

with the peak SHIV viral load across the 50 animals infected

in both trials 1 and 2 with the SHIV challenge viruses (Fig.

2A). We then also assessed whether the timing of the post-

challenge peak CD8 immunity affected the peak level of

SHIV viremia post-challenge. Animals with a very early

(week 1) or late (�4 weeks) peak peripheral blood CD8 T

cell response following SHIV challenge were less likely to

Fig. 2. Post-challenge correlation between peak SHIV plasma viral load and gag-specific CD8 T cell IFNg responses in PBMC. All 50 control and

vaccinated macaques receiving the various DNA and rFPV regimens in both trial 1 and trial 2 (see Table 1) were included in these analyses. (A) The peak

viral load is inversely correlated with the peak IFNg response (P = 0.003; R-squared 0.18). (B) The peak viral load correlates with the timing of the peak

IFNg response (P = 0.014, t test).

I. Stratov et al. / Virology 337 (2005) 222–234226

control the peak SHIV viremia, with a mean 0.88 log10copies/mL higher level of plasma SHIV RNA compared to

animals with peak SIV Gag-specific CD8 T cell responses

that peaked at weeks 2–3 (Fig. 2B).

Phenotype of Gag-specific T cells

To further characterize effective T cell immunity to

SIV/HIV in macaques, we then performed additional

experiments on the phenotype, breadth, cytokine secretion

profile and ability of T cell responses to recognition of

whole virions. For these analyses, we studied some of the

animals vaccinated with the DNA/FPV vaccine regimens

(group 1 of trials 1 and 2) associated with high levels of

peripheral T cell immunity and good outcomes from SHIV

challenge.

To first determine the phenotype of the effector CD8 T

cells to SIV Gag assisting in the control of SHIV viremia,

we studied the surface expression of a series of molecules

associated with an effector/memory phenotype in an ICS

assay on PBMC from animals in group 1 of trial 1. The

h-chain of CD8 is specific to CD3+8+ T lymphocytes.

CD8hhi T cells are more characteristic of naı̈ve cells, with

increased expression in cord blood and larger Vh repertoire

or they express markers of central memory (CCR7, CD62L

and CD28), while CD8hlo T cells display an effector

memory phenotype with increased levels of activation and

cytotoxic molecules and express low levels of CD62L and

CCR7 but high levels of surface CD95 (Konno et al., 2002;

Schmitz et al., 1998). The vast majority of the responding

CD8 T cells were confined to the CD8hlo-expressing cells

(Fig. 3A). These effector SIVgag-specific CD8 T cells were

phenotypically CD45RA negative and CD95 positive

(Fig. 3B). To study CD62L expression, a marker of effector

T cells (Kaech et al., 2002), PBMC were sorted for CD62L

expression (high or low) prior to ICS assay to obviate

CD62L down regulation during the 7 h in vitro incubation

for the ICS assay. CD8 T cells producing IFNg in response

Fig. 4. Differential cytokine production by SIV/HIV-specific T cells. (A)

Representative flow cytometric data from animal 4295 (group 1, trial 1) 18

weeks post-SHIV challenge: CD8 Gag-specific peripheral blood T cells

expressing IFNg also produced TNFa. (B) Time course of HIV Gag-

specific T cell responses post-rFPV boosting (and prior to challenge) of a

representative macaque in trial 3; IL-2 production from peripheral blood T

cells paralleled the expression of IFNg and TNFa in both CD8- and CD4-

positive T cells.

Fig. 3. Phenotype of SIV Gag-specific T cell responses in peripheral blood

by intracellular cytokine staining. Whole blood was stimulated with

overlapping Gag peptides and lymphocytes gated by forward and side-

scatter criteria and their phenotype analyzed by four-color flow cytometry

using combinations of CD3-PE, CD8h-PE, CD45RA-PE, CD62L-PE,

CD8a-PerCP, CD95-FITC and IFNg-APC. For all analyses, macaques from

group 1 of trial 1 were studied, with representative plots of one animal from

at least four animals studied shown for panels A and B and two animals for

panel C. (A) Five weeks following pathogenic SHIV challenge, a large Gag-

specific CD3+CD8a+ IFNg immune response (left plot, using CD3-PE,

CD8a-PerCP, CD4-FITC and IFNg-APC) is induced in animals previously

vaccinated with DNA followed by rFPV boosting. When the Gag-specific

cells (gated directly from lymphocytes, using CD8h-PE, CD4-FITC and

IFNg-APC) were analyzed for CD8h expression (right panel), the majority

of the Gag-specific cells expressing IFNgwere contained within the CD8hlopopulation. (B) Two weeks post-rFPV boost, CD8a+lymphocytes were

studied for CD45RA and CD95 expression—Gag-specific lymphocytes are

CD45RA negative (right plot, using CD45RA-PE, CD8a-PerCP, CD4-FITC

and IFNg-APC) and CD95 positive (left plot, using CD95-FITC, CD3-PE,

CD8a-PerCP and IFNg-APC). (C) Twenty-nine weeks after SHIV

challenge, cells were initially sorted for CD62L expression prior to ICS

assay and then CD8a expressing T lymphocytes studied for Gag-specific

IFNg expression (using CD3-PE, CD8a-PerCP, CD4-FITC and IFNg-

APC)—Effector cells are CD62Llo (right plot).

I. Stratov et al. / Virology 337 (2005) 222–234 227

to SIV Gag following SHIV challenge were confined to the

sorted cells that were CD62L negative (0.34% Gag-specific

CD8 T cells cf. 0.05% for CD62Lhi; Fig. 3C).

Differential cytokine secretion of Gag-specific T cells

IFNg is only one of several effector cytokines likely to be

important in the control of lentivirus infections, with

evidence that TNFa and IL-2 are potentially important

mediators of effective control of these infections (McKay et

al., 2002; Pantaleo and Koup, 2004). We therefore studied

the differential cytokine profile of Gag-specific T cells in

DNA/rFPV-vaccinated macaques (Fig. 4).

In 9 of 11 blood samples studied in animals in group 1 of

trials 1 and 2, DNA/rFPV vaccine-induced Gag-specific

CD4 and CD8 T cells simultaneously produced both IFNg

and TNFa (Fig. 4A). We then also studied the kinetics of the

simultaneous production of IL-2, IFNg and TNFa after the

I. Stratov et al. / Virology 337 (2005) 222–234228

rFPV in two animals vaccinated with HIV-1 expressing

DNA and rFPV vaccines in trial 3. IL-2 was also produced

by Gag-specific T cells although we did not co-stain for all

three cytokines in the same tube. Nevertheless, the

production of all three cytokines by Gag-specific CD4 and

CD8 T cells followed the same temporal pattern after rFPV

boosting when studied kinetically in macaques vaccinated

with HIV-1 DNA and rFPV vaccines in trial 3, although the

production of IL-2 was shorter-lived (Fig. 4B). A similar

pattern was observed for both animals studied and this was

duplicated even when individual epitopes were tested, as

opposed to a pool of peptides (data not shown).

There were, however, interesting exceptions to this

general observation that IFNg secretion from Gag-specific

T cells paralleled that of other cytokines. In 2 of 11

animals studied, Gag-specific CD8 T cells primarily

liberated TNFa and not IFNg (Fig. 5). After the rFPV

boosting, DNA-primed animals 4245 and 4381 (from trial

2, group 1, respectively) demonstrated minimal or no

(0.02% and 0.03%, respectively) CD8 + T cells expressing

IFNg in response to SIV Gag but did demonstrate TNFa

responses of 0.27% and 0.42%, respectively. In these two

animals, the Gag-specific CD4 T cell responses secreted

both IFNg and TNFa. Even after SHIV challenge,

macaque 4245, which controlled SHIVSF162P3 plasma

viremia to <3.2log10 copies/mL by 11 weeks after

challenge, failed to mount Gag-specific CD8+ T cells

expressing IFNg, although Gag-specific CD4 T cells

expressing IFNg were again detected. Although anecdotal,

this observation suggests that secretion of TNFa by SHIV-

specific T cells may, at least in some animals, facilitate

control of viremia.

Fig. 5. Primary production of TNFa and not IFNg by SIV Gag-specific CD8

lymphocytes, obtained 3 weeks after the rFPV boost, were analyzed for Gag-s

responses noted in the upper left quadrants reflect CD4+ T cells. Both macaques

Breadth of SHIV-specific CD4 and CD8 T cell responses

Macaques receiving DNA/rFPV vaccination (primarily

expressing shared SIV Gag/Pol antigens) are able to develop

effective CD8 and CD4 T cell responses following SHIV

challenge virus (Dale et al., 2004a). Responses were

detected against multiple SIV/HIV proteins in many

animals, including peptide pools derived from HIV-1 Env,

SIV Pol and Nef as well as SIV Gag, although responses

directed against Gag were the strongest, presumably

reflecting better priming of Gag-specific T cells by the

Gag-expressing vaccines (data not shown).

Sufficient data were generated in many animals to study,

in detail, the breadth of epitopes recognized within the Gag

protein. Utilizing truncated sets of overlapping 15mer

peptides, at least 19 different CD8 and 18 different CD4

epitopes were detected (Fernandez et al., 2005). An example

of epitope mapping in a macaque with a broad and robust

Gag-specific T cell response 4277 (group 1, trial 1) is shown

in Fig. 6. This macaque recognized a minimum of seven

epitopes (three CD8 and four CD4 epitopes) on initial

screening with pools of five peptides. Of the seven epitopes

identified (including the peptide 41 epitope which had been

detected in other macaques), six of these T cell responses

were narrowed to single 15mer peptides or to a pair of

overlapping 15mer peptides. We have recently restricted one

of the peptide epitopes (peptide 41) to a novel pigtail

macaque MHC class I molecule ManeA*10 (Smith et al.,

2005) and work on MHC restriction of other epitopes is

ongoing.

Overall, the breadth of the T cell response across IM

vaccinated macaques in group 1 of trials 1 and 2 averaged

T cells in 2 of 11 macaques studied. Gated CD3/CD8a peripheral blood

pecific cytokine expression by ICS. The percentage of non-CD8a T cell

were from trial 2, group 1.

Fig. 6. Broad SIV Gag-specific T cell responses in PBMC following DNA/

rFPV vaccination and SHIV challenge. At week 14 following SHIV

challenge, animal 4277 (group 1, trial 1) was studied for T cell immunity to

pools of five 15mer SIV Gag peptides, overlapping by 11 amino acids. The

percentage of CD8 or CD4 T cells expressing IFNg in response to peptide

pools or individual peptides are shown. Many of these T cell responses were

mapped to single 15mer peptides or two overlapping 15mer peptides in

subsequent weeks (arrows); confirmed positive results are in bold. Specific

amino acid sequences of the epitopes were as follows: Peptide 1/2 overlap

NSVLSGKKADE; Peptide 4/5 overlap EKIRLRPNGKK; Peptide 14/15

overlap CQKILSVLAPL; Peptide 48/49 overlap VGDHQAAMQII; Pep-

tide 78 QTDAAVKNWMTQLL. A number of likely additional CD4

epitopes (underlined) were also present but responses not narrowed to

single or adjacent 15mers.

I. Stratov et al. / Virology 337 (2005) 222–234 229

3.3 Gag epitopes per animal in peripheral blood. In contrast,

only 2 of the 10 controls across both trials 1 and 2 had a

sufficient Gag-specific T cell response after challenge to

map T cell epitopes, and in both of these animals

recognition of only a single 15mer peptide each was

detected.

Recognition of whole SIV by CD8 and CD4 T cells

We then studied the Gag-specific CD4 and CD8 T cell

immune responses induced by DNA/rFPV vaccination for

their ability to recognize SIV/HIV in its native conforma-

tion, rather than just as individual 15mer peptides. Using

conformationally intact whole inactivated SIV, we inves-

tigated the time required in vitro for T cells in whole blood

to respond to processed whole SIV virions. The time taken

to process and present SIV was evaluated by adding BFA,

which blocks intracellular processing of whole antigen for

presentation by MHC class I to CD8 T cells, at varying time

intervals after exposure of antigen. A macaque, 4290 from

group 1 in trial 1, which had a favorable outcome from

SHIV challenge with persistent high levels of both CD8 and

CD4 Gag-specific T cell immunity was studied for these

experiments.

Within 30 min antigen-presenting cells in peripheral

blood processed and began to present epitopes of the

whole virus on their surface for recognition by circulating

CD8 T cells (Fig. 7A). CD8 T cell expression of

intracellular IFNg in response to whole inactivated SIV

peaked with a 2-h processing time prior to the addition of

BFA. The CD8 T cell responses to the SIV Gag peptide

pool were 10-fold greater than that to whole SIV,

presumably reflecting the higher molarity of peptide

epitopes available. Unlike responses to whole SIV, CD8

T cell responses to the Gag peptide pool are comparatively

unaffected by addition of BFA at time zero. This suggests

that the responses to the whole inactivated SIV particle are

not due to contaminating free Gag peptide fragments. The

CD4 T cell responses to the whole inactivated SIV

parallels, and is nearly equivalent to, the CD4 T cell

response to free Gag peptides, indicating that alternative

processing of class II antigens from the whole inactivated

SIV for CD4 recognition does not require such extensive

intracellular processing (Fig. 7B).

Discussion

Protection from disease, at least in the short term, is

observed following pathogenic SHIV challenge of mac-

aques using prime/boost vaccine regimens (Dale et al.,

2004a; Robinson et al., 1999; Shiver et al., 2002).

Several previous studies have either measured SHIV-

specific T cell immune responses by IFNg ELISpot,

which is unable to readily phenotype effector cells, or

focus primarily on measuring the magnitude of the CD8

T cell response to a single immunodominant peptide by

MHC class I tetramer analysis (Barouch et al., 2000).

Vaccines based on single viral epitopes have been

relatively inefficient in protecting macaques against SIV

challenge (Yasutomi et al., 1995) and there is increasing

evidence for CTL escape from narrowly directed CD8 T

cell immunity (O’Connor et al., 2002). Further defining

the phenotype and kinetics of protective CD4 and CD8 T

cell immune responses may assist rational improvement

of vaccine design.

The systematic study of SIV Gag-specific CD4 and CD8

T cells by ICS following rFPV boost in standard IM

regimens of DNA and rFPV vaccinations in 10 macaques

across two trials revealed an interesting pattern. We

observed a coordinated induction of first SIV Gag-specific

CD4 T cell immune responses, followed by larger peaks of

CD8 T cell immune responses. Whether this relationship

between CD4 and CD8 T cell immune responses also occurs

with other viral proteins remains to be determined and likely

to depend in part on the efficiency of vaccination, epitopes

Fig. 7. Short interval of processing of whole inactivated SIV for presentation to Gag-specific T cells. A time course of IFNg production by T cells in PBMC

obtained from macaque 4290 (trial 1, group 1) 37 weeks post-challenge to either SIV Gag 15mer peptide pool or whole inactivated SIV. (A) CD8+ responses to

whole inactivated SIV increase over time; peak responses occur with a 2- to 4-h time delay prior to the addition of BFA. (Note log scale). (B) CD4 T cell

responses to free peptide and whole inactivated SIV parallel each other, reflecting different cellular processing.

I. Stratov et al. / Virology 337 (2005) 222–234230

encoded by each protein and MHC haplotype of the host.

However, such a temporal relationship should be desirable

in maintaining memory levels of CD8 T cell responses,

given evidence that CD4 T cell help is critical in the

expansion of CD8 + CTLs upon re-encounter with antigen

(Janssen et al., 2003). Janssen et al. demonstrated that mice

depleted of CD4 T cells prior to secondary encounter with

antigen can still mount CD8 + CTL responses provided

CD4 help was available during the initial encounter with

antigen during immunization. This is of critical importance

given that acute HIV infection is tropic for CD4 T cells

rendering them dysfunctional (Douek et al., 2002).

The kinetics and magnitude of the anamnestic Gag-

specific T cell response following SHIV challenge was also

intriguing. For these studies we examined the relationship of

peak CD8 T cell responses after SHIV challenge with peak

acute SHIV viremia levels in 50 control and vaccinated

animals across two trials. The timing (weeks 2–3) and

magnitude of the peak Gag-specific CD8 T cell response

following challenge were significantly associated with

improved control of SHIV viremia across the two SHIV

challenge studies performed. Optimal generation and

expansion of virus-specific CD8 T cell responses following

encounter with pathogenic lentiviruses is likely to be an

important determinant of the outcome of infection. Interest-

ingly, there was no clear relationship between the recall of

SHIV-specific CD4 T cell responses and control of peak

viremia, possibly because these cells are readily infected,

and likely rendered dysfunctional, by primate lentiviruses

(Douek et al., 2002).

The phenotype and cytokine of Gag-specific CD8 T cells

following challenge was further probed by studying

individual macaques vaccinated with DNA and rFPV

regimens associated with successful outcomes of SHIV

challenge. As expected, the phenotype was that of an

effector/memory T cell, having the surface cell phenotype of

CD8hlo, CD45RA negative, CD95hi and CD62L negative.

The proportion of SIVgag-specific cells in the CD8hlofraction was approximately twofold higher than that in the

CD3/CD8a fraction. Additional surface markers such as

CCR7 would have been useful to confirm this phenotype

but were not available for pigtail macaque cells, although

CD8hlo cells are typically CCR7 negative (Konno et al.,

2002). Generally, all three cytokines studied (IFNg, TNFa

I. Stratov et al. / Virology 337 (2005) 222–234 231

and IL-2) were expressed by Gag-specific T cells, both after

vaccination and after challenge. Expression of IL-2 was

more short lived in both CD4 and CD8 T cells after the

rFPV boost compared to IFNg and TNFa expression,

consistent with their effector/memory phenotype (Konno et

al., 2002). IL-2 expression is likely to drive expansion of the

virus-specific T cells and may wane as vaccine-expressed

antigen levels decline. More prominent IL-2 expression

from virus-specific T cells is a feature of central memory

cells (Konno et al., 2002; Sallusto et al., 1999) and may

result in more durable levels of T cell immunity.

There were exceptions observed to the general rule that

multiple cytokines are co-expressed by Gag-specific T cells.

Two of 11 animals studied predominantly expressed TNFa

and not IFNg, and one of these animals achieved full control

of viral replication. Although requiring confirmation in

larger studies, this observation suggests that TNFa expres-

sion may be sufficient for effective immunity and perhaps

an improved surrogate marker for effective T cell immunity.

However, we consistently noted mildly higher levels of

background TNFa expression from T cells in unstimulated

blood (data not shown), compared to IFNg expression,

which limits the sensitivity of TNFa expression in studying

antigen-specific cells.

A large breadth of SIV-specific CD4 and CD8 T cell

responses was also observed following SHIV challenge of

IM DNA/rFPV-vaccinated macaques. A broad T cell

response is likely to be more important than a large

narrowly focussed one. Rapid recognition of whole virions

by both CD4 and CD8 T cell responses prior to release

of progeny viruses should be critical in effectively con-

trolling viremia (Yang et al., 1996, 1997). We studied T

cell recognition of whole processed virions. SIV-specific

CD8 T cells induced by DNA/rFPV vaccination begin to

secrete IFNg within 30 min, indicating rapid processing of

whole inactivated SIV virions within antigen-presenting

cells. SIV-specific CD4 T cells do not require processing

of SIV through the endoplasmic reticulum and can express

IFNg when BFA is added simultaneously with the

antigen. This rapid recognition of virus exposed cells by

SIV-specific CD8 and CD4 T cells is likely to be an

important factor in the control of virus replication. Future

studies could address the comparative rapidity of recog-

nition of virus-infected cells by specific T cell responses

to more accurately define the requirements for protective

immunity.

In summary, this study of DNA/rFPV-vaccinated mac-

aques found a coordinated induction of broad CD4 and CD8

T cell responses. Recall of Gag-specific CD8 T cell immune

responses correlated with reductions in SHIV viremia

following pathogenic challenge. The broad T cell responses

induced were of an effector-memory phenotype, capable of

expressing multiple effector cytokines and rapidly recog-

nized virus-exposed cells. Taken together, these results

illustrate some of the potentially complex requirements of

effective T cell immunity to HIV-1.

Materials and methods

Animal experimentation

Studies were performed on juvenile pigtail macaques

(Macaca nemestrina). All experimentation was approved by

relevant ethics committees under previously described

conditions (Dale et al., 2002, 2004a).

Vaccine protocols

Three separate pigtail macaque vaccine trials studying

DNA/rFPV prime/boost vaccination were undertaken using

various protocols as previously described (Dale et al.,

2004a; Kent et al., 1998) (see Table 1). Briefly, trial 1

involved two intramuscular DNA vaccine priming and

boosting with a single intramuscular rFPV vaccine (both

encoding shared SIVmac239 Gag/Pol proteins), followed by

rectal challenge with highly pathogenic SHIVmn229 which

results in reductions in SHIV viremia and retention of CD4

T cells (Dale et al., 2002, 2004a; Kent et al., 2002). Trial 2

involved three DNA priming vaccinations then two boosting

vaccinations with rFPV (again encoding shared SIV

proteins), followed by vaginal challenge with moderately

pathogenic SHIVSF162P3 (Harouse et al., 1999, 2001), which

also resulted in reduction in SHIV viral load in comparison

to control unvaccinated animals (S. Kent et al., manuscript

submitted). Trial 3 involved three intramuscular DNA

priming vaccinations and a single intramuscular rFPV

boosting, both encoding HIV-1 subtype AE proteins, with-

out viral challenge (De Rose et al., 2005).

DNA vaccines

DNA vaccine, pHIS-SHIV-B, encoded full-length unmu-

tated SIVmac239 Gag and Pol, HIV-1NL43, Tat, Rev and Vpu

and the 5V one third of HIV-1AD8 Env (Dale et al., 2004a).

Genes were inserted into vector pHIS-64 (Dr. Heather

Davis, Coley Pharmaceuticals) behind the human cytome-

galovirus immediate–early promoter. Plasmid vector pHIS-

64 has kanamycin resistance, bovine growth hormone

polyA termination signal and 14 primate-optimized CpG

immunostimulatory sequences. Empty vector plasmid DNA,

pHIS, served as a control vaccine. Plasmid pHIS-SHIV-B

(1mg/dose IM) for immunization was prepared by QIAGEN

(Hilden, Germany), control DNA vaccine pHIS was

prepared using the EndoFree Plasmid Giga kit (QIAGEN).

Recombinant fowlpox virus vaccines

Construction of the FPV vaccines has been described

(Boyle et al., 2004; Dale et al., 2004a). Briefly, FPVgag/pol,

expressing SIV Gag/Pol, was constructed by inserting the

promoter-SIV gag/pol PCR amplicon into pKG10aSIVgag-

pol for insertion into FPV-M3 at the F6,7,9 site. PCR

primers used the FPV early/late promoter and an early

I. Stratov et al. / Virology 337 (2005) 222–234232

transcription terminator for insertion into fowlpox virus

vector FPV-M3. HIV-193TH254 env (mutated to remove the

middle third) was amplified from plasmid vector

pCH34AEenv(m) and inserted into a separate FPV-M3

vector at the REV site. Recombinants were selected on the

basis of co-expression of the Escherichia coli gpt gene and

plaque purified on the basis of co-expression of the E. coli

beta-galactosidase gene. FPV vaccines were prepared in

saline.

Antigens

Antigens studied for T cell responses were primarily

15mer peptides overlapping by 11 amino acids spanning

SIVmac239 Gag, Pol and Nef, HIV-1MN Env, all kindly

obtained from the NIH AIDS Research and Reference

Reagent Repository. For trial 3 we obtained homologous

subtype AE HIV-19TH253 Gag 15mer peptides (Auspep,

Parkville, Australia). Peptides were solubilized in pure

DMSO at high concentrations (1 mg peptide/10–40 ALDMSO) and peptides for each antigen pooled. 1 Ag/mL of

each 15mer peptide within the pools was used for the ICS

assay. To assess immune responses to intact virions we used

whole aldrithiol-2 inactivated SIVmne (Arthur et al., 1998;

Dale et al., 2004a) (10 Ag/mL, kindly provided by Dr J

Lifson, National Cancer Institute, Frederick, MD) and the

relevant microvesicle control preparation derived from the

same cell line used to grow the virus.

Intracellular cytokine staining

Intracellular cytokine staining (ICS) assay was per-

formed as previously described (Dale et al., 2004a; Maecker

et al., 2001). Briefly, in the presence of co-stimulatory

molecules CD28 and CD49d (BD Sciences), whole blood or

mononuclear cell preparations were incubated with pools of

15mer peptides or whole inactivated SIV. Brefeldin A

(BFA; Sigma; final concentration 10 Ag/mL) was added to

prevent protein export from the endoplasmic reticulum.

Incubation times were generally terminated after 7 h, the

last 5 h in the presence of BFA, although this varied in some

experiments. Cells were then stained with surface antibodies

including anti-CD4 FITC and PerCP, anti-CD3 PE, anti-

CD8 PerCP and APC, anti-CD45RA PE, anti-CD95 FITC,

anti-CD62L PE (BD Biosciences, Pharmingen, San Diego,

CA) or CD8h PE (Immunotech, Marseille, France), then

lysed and permeabilized as per standard protocols. Intra-

cellular staining was then performed with anti-IFNg APC,

anti-IL-2 APC and/or anti-TNF-a FITC (BD) and fixed

prior to acquisition via FACSort (BD) and analyzed using

cell quest software (BD).

Determining SHIV-specific T cells by ICS

Gating on T lymphocytes was performed using forward

and side scatter criteria and combined this with surface

expression of CD3, based on the nadir of a CD3 histogram.

Two-dimensional dot plots (x-axis: CD8; y-axis: IFNg)

were then formed and responses calculated using quadrant

lines. The vertical quadrant line (defining CD3/CD8+ cells)

was placed along the margin where the density of clearly

CD8+ cells started to reduce, so as to be confident that all

cells included in the quadrant were truly CD8+, although

very similar percentages were obtained when the vertical

quadrant line was placed at the nadir of a CD8+ cell

histogram (not shown). The horizontal quadrant line

(defining IFNg-producing cells) was placed within approx-

imately 10 channel values above the margin where the

density of CD8+ cells had clearly reduced. Analyses

generally were performed after collecting approximately

100,000 lymphocytes. This produced a final CD3/CD8+

count of approximately 20,000 events; counts of less than

1000 CD3/CD8 (or CD3/CD4) events were excluded from

analysis.

Validation of true positive antigen-specific responses

Validation of the ICS assay in M. nemestrina was

performed by studying background levels of IFNg

production in the first 101 negative control samples; one

(1%) sample was excluded from our statistical validation

of the cut-off for a true positive. This sample had a

background of 1.88% (i.e., 48-fold above the mean

background level) and was 5.7-fold larger than the next

largest background level of 0.33%. From these 100

samples used for final statistical analysis, 81% of negative

control samples had a background �0.05%, the mean

background being 0.04% (range 0.00–0.33%; standard

deviation 0.05%). Ninety-six (96%) of the samples had a

background level within 2 standard deviations of the mean.

We determined that a true positive was an antigen-specific

result more than two standard deviations above the

background level (i.e., approximately 0.10% or threefold

above background). To be confident of a true positive in

samples with backgrounds greater than 0.05%, we deter-

mined that IFNg production in antigen-specific samples

needed to be >0.10% above background and threefold

above background.

Background ICS results were consistent throughout the

studies in naı̈ve, vaccinated and SHIV-infected macaques;

that is, 80–90% of samples had background levels <0.05%

and only 1–2% of samples had very high backgrounds

(>10-fold above average background).

Cell sorting to determine CD62L status

Sorting for cells for the expression of CD62 ligand

(CD62L) was performed on a MoFlo flow cytometer (Dako

cytomation, Fort Collins, USA). Briefly, fresh PBMC were

stained with CD62L-PE (at 4 -C for 30 min) and CD62L-

positive and -negative cells sorted into separate tubes before

proceeding to Gag antigen stimulation.

I. Stratov et al. / Virology 337 (2005) 222–234 233

Acknowledgments

We thank Rob De Rose, Socheata Chea, Caroline

Fernandez, Richard Sydenham, Alistair Ramsay, Charani

Ranasinghe, Scott Thomson, Ian Ramshaw, Barbara Cou-

par, David Boyle, Matthew Law and all other members of

the Australian Thai HIV Vaccine Consortium for helpful

assistance and advice. Supported by the NIH, a division of

the US Department of Health and Human Services, award

N01-AI-05395 and Australian National Health and Medical

Research Council grants 299907, 251653 and 251654.

References

Amara, R.R., Villinger, F., Altman, J.D., Lydy, S.L., O’Neil, S.P., Staprans,

S.I., Montefiori, D.C., Xu, Y., Herndon, J.G., Wyatt, L.S., Candido,

M.A., Kozyr, N.L., Earl, P.L., Smith, J.M., Ma, H.L., Grimm, B.D.,

Hulsey, M.L., Miller, J., McClure, H.M., McNicholl, J.M., Moss, B.,

Robinson, H.L., 2001. Control of a mucosal challenge and prevention

of AIDS by a multiprotein DNA/MVA vaccine. Science 292 (5514),

69–74.

Appay, V., Nixon, D.F., Donahoe, S.M., Gillespie, G.M., Dong, T., King,

A., Ogg, G.S., Spiegel, H.M., Conlon, C., Spina, C.A., Havlir, D.V.,

Richman, D.D., Waters, A., Easterbrook, P., McMichael, A.J.,

Rowland-Jones, S.L., 2000. HIV-specific CD8 (+) T cells produce

antiviral cytokines but are impaired in cytolytic function. J. Exp.

Med. 192 (1), 63–75.

Arthur, L.O., Bess Jr., J.W., Chertova, E.N., Rossio, J.L., Esser, M.T.,

Benveniste, R.E., Henderson, L.E., Lifson, J.D., 1998. Chemical

inactivation of retroviral infectivity by targeting nucleocapsid protein

zinc fingers: a candidate SIV vaccine. AIDS Res. Hum. Retroviruses 14

(Suppl. 3), S311–S319.

Barouch, D.H., Santra, S., Schmitz, J.E., Kuroda, M.J., Fu, T.M., Wagner,

W., Bilska, M., Craiu, A., Zheng, X.X., Krivulka, G.R., Beaudry, K.,

Lifton, M.A., Nickerson, C.E., Trigona, W.L., Punt, K., Freed, D.C.,

Guan, L., Dubey, S., Casimiro, D., Simon, A., Davies, M.E., Chastain,

M., Strom, T.B., Gelman, R.S., Montefiori, D.C., Lewis, M.G., Emini,

E.A., Shiver, J.W., Letvin, N.L., 2000. Control of viremia and

prevention of clinical AIDS in rhesus monkeys by cytokine-augmented

DNA vaccination. Science 290 (5491), 486–492.

Barouch, D.H., Kunstman, J., Kuroda, M.J., Schmitz, J.E., Santra, S.,

Peyerl, F.W., Krivulka, G.R., Beaudry, K., Lifton, M.A., Gorgone,

D.A., Montefiori, D.C., Lewis, M.G., Wolinsky, S.M., Letvin, N.L.,

2002. Eventual AIDS vaccine failure in a rhesus monkey by viral

escape from cytotoxic T lymphocytes. Nature 415 (6869), 335–339.

Borrow, P., Lewicki, H., Hahn, B.H., Shaw, G.M., Oldstone, M.B., 1994.

Virus-specific CD8+ cytotoxic T lymphocyte activity associated with

control of viremia in primary human immunodeficiency virus type 1

infection. J. Virol. 68 (9), 6103–6110.

Boyle, D.B., Anderson, M.A., Amos, R., Voysey, R., Coupar, B.E., 2004.

Construction of recombinant fowlpox viruses carrying multiple vaccine

antigens and immunomodulatory molecules. BioTechniques 37 (1),

104–6, 108–11.

Dale, C.J., Liu, X.S., De Rose, R., Purcell, D.F., Anderson, J., Xu, Y.,

Leggatt, G.R., Frazer, I.H., Kent, S.J., 2002. Chimeric human papilloma

virus-simian/human immunodeficiency virus virus-like-particle vac-

cines: immunogenicity and protective efficacy in macaques. Virology

301 (1), 176–187.

Dale, C.J., De Rose, R., Stratov, I., Chea, S., Montefiori, D.C.,

Thomson, S., Ramshaw, I.A., Coupar, B.E., Boyle, D.B., Law, M.,

Kent, S.J., 2004a. Efficacy of DNA and fowlpox virus priming/-

boosting vaccines for simian/human immunodeficiency virus. J. Virol.

78 (24), 13819–13828.

Dale, C.J., De Rose, R., Wilson, K., Croom, H., Thomson, S.A., Coupar,

B.E., Ramsay, A.J., Purcell, D.F., Ffrench, R., Law, M., Emery, S.,

Cooper, D.A., Ramshaw, I.A., Boyle, D.B., Kent, S.J., 2004b.

Evaluation in macaques of HIV-1 DNA vaccines containing primate

CpG motifs and fowlpox virus vaccines co-expressing IFNg or IL-12.

Vaccine 23, 188–197.

De Rose, R., Chea, S., Dale, C.J., Reece, J., Fernandez, C.S., Wilson, K.M.,

Thomson, S., Ramshaw, I.A., Coupar, B.E., Boyle, D.B., Sullivan,

M.T., Kent, S.J., 2005. Subtype AE HIV-1 DNA and recombinant

fowlpox virus vaccines encoding five shared HIV-1 genes: safety and T

cell immunogenicity in macaques. Vaccine 23 (16), 1949–1956.

Douek, D.C., Brenchley, J.M., Betts, M.R., Ambrozak, D.R., Hill, B.J.,

Okamoto, Y., Casazza, J.P., Kuruppu, J., Kunstman, K., Wolinsky, S.,

Grossman, Z., Dybul, M., Oxenius, A., Price, D.A., Connors, M., Koup,

R.A., 2002. HIV preferentially infects HIV-specific CD4+ T cells.

Nature 417 (6884), 95–98.

Fernandez, C.S., Stratov, I., De Rose, R., Walsh, K., Dale, C.J., Smith,

M.Z., Agy, M.B., Hu, S.L., Krebs, K., Watkins, D.I., O’Connor, D.H.,

Davenport, M.P., Kent, S.J., 2005. Rapid viral escape at an immunodo-

minant simian-human immunodeficiency virus cytotoxic T-lymphocyte

epitope exacts a dramatic fitness cost. J. Virol. 79 (9), 5721–5731.

Friedrich, T.C., Dodds, E.J., Yant, L.J., Vojnov, L., Rudersdorf, R., Cullen,

C., Evans, D.T., Desrosiers, R.C., Mothe, B.R., Sidney, J., Sette, A.,

Kunstman, K., Wolinsky, S., Piatak, M., Lifson, J., Hughes, A.L.,

Wilson, N., O’Connor, D.H., Watkins, D.I., 2004. Reversion of CTL

escape-variant immunodeficiency viruses in vivo. Nat. Med. 10 (3),

275–281.

Grakoui, A., Shoukry, N.H., Woollard, D.J., Han, J.H., Hanson, H.L.,

Ghrayeb, J., Murthy, K.K., Rice, C.M., Walker, C.M., 2003. HCV

persistence and immune evasion in the absence of memory T cell help.

Science 302 (5645), 659–662.

Harouse, J.M., Gettie, A., Tan, R.C., Blanchard, J., Cheng-Mayer, C., 1999.

Distinct pathogenic sequela in rhesus macaques infected with CCR5 or

CXCR4 utilizing SHIVs. Science 284 (5415), 816–819.

Harouse, J.M., Gettie, A., Eshetu, T., Tan, R.C., Bohm, R., Blanchard, J.,

Baskin, G., Cheng-Mayer, C., 2001. Mucosal transmission and

induction of simian AIDS by CCR5-specific simian/human immuno-

deficiency virus SHIV (SF162P3). J. Virol. 75 (4), 1990–1995.

Horton, H., Vogel, T.U., Carter, D.K., Vielhuber, K., Fuller, D.H., Shipley,

T., Fuller, J.T., Kunstman, K.J., Sutter, G., Montefiori, D.C., Erfle, V.,

Desrosiers, R.C., Wilson, N., Picker, L.J., Wolinsky, S.M., Wang, C.,

Allison, D.B., Watkins, D.I., 2002. Immunization of rhesus macaques

with a DNA prime/modified vaccinia virus Ankara boost regimen

induces broad simian immunodeficiency virus (SIV)-specific T-cell

responses and reduces initial viral replication but does not prevent

disease progression following challenge with pathogenic SIVmac239.

J. Virol. 76 (14), 7187–7202.

Janssen, E.M., Lemmens, E.E., Wolfe, T., Christen, U., von Herrath, M.G.,

Schoenberger, S.P., 2003. CD4+ T cells are required for secondary

expansion and memory in CD8+ T lymphocytes. Nature 421 (6925),

852–856.

Kaech, S.M., Wherry, E.J., Ahmed, R., 2002. Effector and memory T-cell

differentiation: implications for vaccine development. Nat. Rev.,

Immunol. 2 (4), 251–262.

Kalams, S.A., Buchbinder, S.P., Rosenberg, E.S., Billingsley, J.M., Colbert,

D.S., Jones, N.G., Shea, A.K., Trocha, A.K., Walker, B.D., 1999.

Association between virus-specific cytotoxic T-lymphocyte and helper

responses in human immunodeficiency virus type 1 infection. J. Virol.

73 (8), 6715–6720.

Kent, S.J., Zhao, A., Best, S.J., Chandler, J.D., Boyle, D.B., Ramshaw, I.A.,

1998. Enhanced T cell immunogenicity and protective efficacy of a

human immunodeficiency virus type 1 vaccine regimen consisting of

consecutive priming with DNA and boosting with recombinant fowlpox

virus. J. Virol. 72 (12), 10180–10188.

Kent, S.J., Dale, C.J., Preiss, S., Purcell, D.F., 2002. Evidence of

recombination between 3V and 5V LTRs in macaques inoculated with

SIV DNA. AIDS Res. Hum. Retroviruses 18 (3), 227–230.

I. Stratov et al. / Virology 337 (2005) 222–234234

Konno, A., Okada, K., Mizuno, K., Nishida, M., Nagaoki, S., Toma, T.,

Uehara, T., Ohta, K., Kasahara, Y., Seki, H., Yachie, A., Koizumi, S.,

2002. CD8alpha alpha memory effector T cells descend directly from

clonally expanded CD8alpha+beta high TCRalpha beta T cells in vivo.

Blood 100 (12), 4090–4097.

Koup, R.A., Safrit, J.T., Cao, Y., Andrews, C.A., McLeod, G., Borkowsky,

W., Farthing, C., Ho, D.D., 1994. Temporal association of cellular

immune responses with the initial control of viremia in primary human

immunodeficiency virus type 1 syndrome. J. Virol. 68 (7), 4650–4655.

Leslie, A.J., Pfafferott, K.J., Chetty, P., Draenert, R., Addo, M.M., Feeney,

M., Tang, Y., Holmes, E.C., Allen, T., Prado, J.G., Altfeld, M., Brander,

C., Dixon, C., Ramduth, D., Jeena, P., Thomas, S.A., St. John, A.,

Roach, T.A., Kupfer, B., Luzzi, G., Edwards, A., Taylor, G., Lyall, H.,

Tudor-Williams, G., Novelli, V., Martinez-Picado, J., Kiepiela, P.,

Walker, B.D., Goulder, P.J., 2004. HIV evolution: CTL escape mutation

and reversion after transmission. Nat. Med. 10 (3), 282–289.

Maecker, H.T., Dunn, H.S., Suni, M.A., Khatamzas, E., Pitcher, C.J.,

Bunde, T., Persaud, N., Trigona, W., Fu, T.M., Sinclair, E., Bredt, B.M.,

McCune, J.M., Maino, V.C., Kern, F., Picker, L.J., 2001. Use of

overlapping peptide mixtures as antigens for cytokine flow cytometry.

J. Immunol. Methods 255 (1–2), 27–40.

McKay, P.F., Schmitz, J.E., Barouch, D.H., Kuroda, M.J., Lifton, M.A.,

Nickerson, C.E., Gorgone, D.A., Letvin, N.L., 2002. Vaccine protection

against functional CTL abnormalities in simian human immunodefi-

ciency virus-infected rhesus monkeys. J. Immunol. 168 (1), 332–337.

McMichael, A.J., Rowland-Jones, S.L., 2001. Cellular immune responses

to HIV. Nature 410 (6831), 980–987.

Nabel, G.J., 2001. Challenges and opportunities for development of an

AIDS vaccine. Nature 410 (6831), 1002–1007.

O’Connor, D.H., Allen, T.M., Watkins, D.I., 2002. Cytotoxic T lymphocyte

escape monitoring in simian immunodeficiency virus vaccine challenge

studies. DNA Cell Biol. 21 (9), 659–664.

Ogg, G.S., Jin, X., Bonhoeffer, S., Dunbar, P.R., Nowak, M.A., Monard, S.,

Segal, J.P., Cao, Y., Rowland-Jones, S.L., Cerundolo, V., Hurley, A.,

Markowitz, M., Ho, D.D., Nixon, D.F., McMichael, A.J., 1998.

Quantitation of HIV-1-specific cytotoxic T lymphocytes and plasma

load of viral RNA. Science 279 (5359), 2103–2106.

Pantaleo, G., Koup, R.A., 2004. Correlates of immune protection in HIV-1

infection: what we know, what we don’t know, what we should know.

Nat. Med. 10 (8), 806–810.

Pitcher, C.J., Hagen, S.I., Walker, J.M., Lum, R., Mitchell, B.L., Maino,

V.C., Axthelm, M.K., Picker, L.J., 2002. Development and homeostasis

of T cell memory in rhesus macaque. J. Immunol. 168 (1), 29–43.

Robinson, H.L., Montefiori, D.C., Johnson, R.P., Manson, K.H., Kalish,

M.L., Lifson, J.D., Rizvi, T.A., Lu, S., Hu, S.L., Mazzara, G.P.,

Panicali, D.L., Herndon, J.G., Glickman, R., Candido, M.A., Lydy,

S.L., Wyand, M.S., McClure, H.M., 1999. Neutralizing antibody-

independent containment of immunodeficiency virus challenges by

DNA priming and recombinant pox virus booster immunizations. Nat.

Med. 5 (5), 526–534.

Rosenberg, E.S., Billingsley, J.M., Caliendo, A.M., Boswell, S.L., Sax,

P.E., Kalams, S.A., Walker, B.D., 1997. Vigorous HIV-1-specific CD4+

T cell responses associated with control of viremia. Science 278 (5342),

1447–1450.

Sallusto, F., Lenig, D., Forster, R., Lipp, M., Lanzavecchia, A., 1999. Two

subsets of memory T lymphocytes with distinct homing potentials and

effector functions. Nature 401 (6754), 708–712.

Schmitz, J.E., Forman, M.A., Lifton, M.A., Concepcion, O., Reimann Jr.,

K.A., Crumpacker, C.S., Daley, J.F., Gelman, R.S., Letvin, N.L., 1998.

Expression of the CD8alpha beta-heterodimer on CD8(+) T lympho-

cytes in peripheral blood lymphocytes of human immunodeficiency

virus� and human immunodeficiency virus+ individuals. Blood 92 (1),

198–206.

Shiver, J.W., Fu, T.M., Chen, L., Casimiro, D.R., Davies, M.E., Evans,

R.K., Zhang, Z.Q., Simon, A.J., Trigona, W.L., Dubey, S.A., Huang,

L., Harris, V.A., Long, R.S., Liang, X., Handt, L., Schleif, W.A., Zhu,

L., Freed, D.C., Persaud, N.V., Guan, L., Punt, K.S., Tang, A., Chen,

M., Wilson, K.A., Collins, K.B., Heidecker, G.J., Fernandez, V.R.,

Perry, H.C., Joyce, J.G., Grimm, K.M., Cook, J.C., Keller, P.M.,

Kresock, D.S., Mach, H., Troutman, R.D., Isopi, L.A., Williams,

D.M., Xu, Z., Bohannon, K.E., Volkin, D.B., Montefiori, D.C., Miura,

A., Krivulka, G.R., Lifton, M.A., Kuroda, M.J., Schmitz, J.E., Letvin,

N.L., Caulfield, M.J., Bett, A.J., Youil, R., Kaslow, D.C., Emini, E.A.,

2002. Replication-incompetent adenoviral vaccine vector elicits

effective anti-immunodeficiency-virus immunity. Nature 415 (6869),

331–335.

Smith, M.Z., Dale, C.J., De Rose, R., Stratov, I., Fernandez, C.S., Brooks,

A.G., Weinfurter, J., Krebs, K., Riek, C., Watkins, D.I., O’Connor,

D.H., Kent, S.J., 2005. Analysis of pigtail macaque major histocompat-

ibility complex class I molecules presenting immunodominant simian

immunodeficiency virus epitopes. J. Virol. 79 (2), 684–695.

Stratov, I., DeRose, R., Purcell, D.F., Kent, S.J., 2004. Vaccines and vaccine

strategies against HIV. Curr. Drug Targets 5 (1), 71–88.

Yang, O.O., Kalams, S.A., Rosenzweig, M., Trocha, A., Jones, N., Koziel,

M., Walker, B.D., Johnson, R.P., 1996. Efficient lysis of human

immunodeficiency virus type 1-infected cells by cytotoxic T lympho-

cytes. J. Virol. 70 (9), 5799–5806.

Yang, O.O., Tran, A.C., Kalams, S.A., Johnson, R.P., Roberts, M.R.,

Walker, B.D., 1997. Lysis of HIV-1-infected cells and inhibition of viral

replication by universal receptor T cells. Proc. Natl. Acad. Sci. U.S.A.

94 (21), 11478–11483.

Yasutomi, Y., Koenig, S., Woods, R.M., Madsen, J., Wassef, N.M., Alving,

C.R., Klein, H.J., Nolan, T.E., Boots, L.J., Kessler, J.A., et al., 1995. A

vaccine-elicited, single viral epitope-specific cytotoxic T lymphocyte

response does not protect against intravenous, cell-free simian immu-

nodeficiency virus challenge. J. Virol. 69 (4), 2279–2284.