A New Dynamic Model of CD8?T Effector Cell Responses via
CD4?T Helper-Antigen-Presenting Cells1
Jim Xiang,2Hui Huang, and Yongqing Liu
A long-standing paradox in cellular immunology has been the conditional requirement for CD4?Th cells in priming of CD8?CTL
responses. We propose a new dynamic model of CD4?Th cells in priming of Th-dependent CD8?CTL responses. We demonstrate
that OT II CD4?T cells activated by OVA-pulsed dendritic cells (DCOVA) are Th1 phenotype. They acquire the immune synapse-
composed MHC II/OVAII peptide complexes and costimulatory molecules (CD54 and CD80) as well as the bystander MHC class
I/OVAI peptide complexes from the DCOVAby DCOVAstimulation and thus also the potential to act themselves as APCs. These
CD4?Th-APCs stimulate naive OT I CD8?T cell proliferation through signal 1 (MHC I/OVAI/TCR) and signal 2 (e.g., CD54/
LFA-1 and CD80/CD28) interactions and IL-2 help. In vivo, they stimulate CD8?T cell proliferation and differentiation into CTLs
and induce effective OVA-specific antitumor immunity. Taken together, this study demonstrates that CD4?Th cells carrying
acquired DC Ag-presenting machinery can, by themselves, efficiently stimulate CTL responses. These results have substantial
implications for research in antitumor and other aspects of immunity. The Journal of Immunology, 2005, 174: 7497–7505.
priming (1). Such help was originally thought to be mediated by
CD4?T cell IL-2 acting at short range to promote CD8?T cell
Two models of CD4?T help for CD8?CTL responses have
been proposed previously, including the passive model of three-
cell interaction (3, 4) and the dynamic model of sequential two-cell
interactions by APCs (5). The three-cell model suggested that ac-
tivated CD4?T cells and naive CD8?T cells must interact si-
multaneously with a common APC and that the CD4?Th cells
provide CD8?T cell help via expression of IL-2 (Fig. 1A). The
conundrum, however, is how a rare Ag-specific CD4?Th cell and
an equally rare Ag-specific CD8?T cell (typically 1 in 105–106T
cells) would simultaneously find the same Ag peptide-carrying
APC in an unprimed animal (6). Instead, Ridge et al. (5) have
proposed a dynamic model of two sequential interactions, in which
an APC first offers costimulatory signals to a CD4?Th cell and
then to a CD8?CTL cell (Fig. 1B). According to this model, the
APC-stimulated CD4?Th cells must first reciprocally counter-
stimulate the APCs (through CD40 (CD40L)3signaling) such that
this newly “conditioned” APC can then directly costimulate CD8?
CTLs. Support for this model comprises evidence that Ag-specific
eneration of effective CTL responses to minor histocom-
patibility or tumor Ags not associated with danger sig-
nals often requires help from CD4?Th cells via cross-
CTL responses can be induced by vaccination with either large
numbers of APCs activated in vitro through CD40 signaling or, in
either MHC class II knockout (KO) or CD4?T cell-depleted mice,
by high level activation of APCs in vivo with anti-CD40 Ab (5,
7–9). Although this model provides a possible explanation for the
conditional nature of T cell help for CTL responses, the experi-
mental conditions used in the above studies may well not accu-
rately model the physiology of Th cell-dependent immune re-
sponses in vivo. In addition, a scarcity caveat remains (10), in that
very small numbers of Ag-bearing APCs (11) must first activate
and be conditioned by the rare naive Ag-specific CD4?Th cells,
and then find and activate in turn equally rare naive Ag-specific
CD8?CTLs. In addition, this model does not explain how Th
cells’ IL-2 would be precisely targeted to Ag-specific CD8?Ag-
specific CTLs. Furthermore, the life span of an activated dendritic
cell (DC) in the T cell zone of a lymph node is ?48 h (11–13),
possibly due to CD4?T cell killing of the cognate APCs (14–15),
whereas the Ag-specific CTL response is first detected at about day
5 in the lymph nodes (11, 16). Thus, this dynamic model also does
not explain compellingly the temporal gap between Ag presenta-
tion and the acquisition of CTL effector function in vivo.
It is recognized that stimulation of T cells by APCs involves at
least two signaling events: one elicited by TCR recognition of
peptide-MHC complexes and the other by costimulatory molecule
signaling (e.g., T cell CD28/APC CD80) (17). A consequence of
such Ag-specific T cell-APC interactions is the formation of an
immunological synapse, comprising a central cluster of TCR-
MHC-peptide complexes and CD28-CD80 interactions surrounded
by rings of engaged accessory molecules (e.g., complexed LFA-
1-CD54) (18, 19). One important feature of synapse physiology is
that APC-derived surface molecules are transferred to the Th cells
during the course of their TCR internalization followed by recy-
cling (20, 21). In theory, if tumor peptide-presenting DCs do trans-
fer a complete set of Ag-presenting molecules (e.g., MHC pep-
tides, CD54, CD80) to IL-2-expressing CD4?Th cells, then these
recipient T cells might also have full capacity to in turn present
these tumor epitopes to CD8?T cells and thereby initiate produc-
tive antitumor CTL responses (Fig. 1C). In this study, we exam-
ined membrane molecule transfer from APCs to CD4?T cells by
APC stimulation. We found that CD4?T cells cannot only acquire
Research Unit, Saskatchewan Cancer Agency, Departments of Oncology, Microbi-
ology, and Immunology, College of Medicine, University of Saskatchewan, Saska-
toon, Saskatchewan, Canada
Received for publication August 30, 2004. Accepted for publication March 25, 2005.
The costs of publication of this article were defrayed in part by the payment of page
charges. This article must therefore be hereby marked advertisement in accordance
with 18 U.S.C. Section 1734 solely to indicate this fact.
1This work was supported by Research Grants MOP63259 and 67230 from the Ca-
nadian Institute of Health Research (to J.X.). Y.L. was supported by Postdoctoral
Fellowship Award from the Canadian Institute of Health Research.
2Address correspondence and reprint requests to Dr. Jim Xiang, Research Unit,
Saskatchewan Cancer Agency, Departments of Oncology, Microbiology and Immu-
nology, University of Saskatchewan, 20 Campus Drive, Saskatoon, Saskatchewan
S7N 4H4, Canada. E-mail address: firstname.lastname@example.org
3Abbreviations used in this paper: CD40L, CD40 ligand; KO, knockout; DC, den-
dritic cell; DCOVA, OVA-pulsed DC; LB27OVAII, OVAII peptide-pulsed LB27 cells;
ECD, energy-coupled dye.
The Journal of Immunology
Copyright © 2005 by The American Association of Immunologists, Inc.0022-1767/05/$02.00
the synapse-composed MHC class II and costimulatory molecules
(CD54 and CD80), but also the bystander MHC class I-peptide
complexes from APCs and these acquired molecules on CD4?T
cells are functional. The CD4?T cells carrying acquired APC
Ag-presenting machinery can thus act as CD4?Th-APCs in stim-
ulation of CTL responses in vitro and in vivo.
Materials and Methods
Tumor cells, reagents, and animals
The highly lung metastatic B16 mouse melanoma BL6-10 and OVA-trans-
fected BL6-10 (BL6-10OVA) cell lines were generated in our own labora-
tory (22). Both cell lines form numerous lung metastasis after i.v. tumor
cell (0.5 ? 106cells/mouse) injection. The mouse B cell hybridoma cell
line LB27 expressing both H-2Kband Iab, the mouse thymoma cell line
EL4 of C57BL/6 mice, and the OVA-transfected EL4 (EG7) cell line,
which is sensitive to CTL killing, were obtained from American Type
Culture Collection. Both BL6-10 and BL6-10OVAexpress similar level of
H-2Kb, but not Iab(data not shown). Both BL6-10OVAand EG7 cells ex-
pressed OVA by flow cytometric analysis, whereas BL6-10 and EL4 cells
did not (Fig. 2). T cell hybridoma cell line RF3370 expresses TCR specific
for H-2Kb-OVA peptide complexes (23). The biotin-labeled mAbs specific
for H-2Kb(AF6-88.5), Iab(AF6-120.1), CD3 (145-2C11), CD4 (GK1.5),
CD8 (53-6.7), CD11b (MAC-1), CD11c (HL3), CD25 (7D4), CD54 (3E2),
CD69 (H1.2F3), CD80 (16-10A1), and V?2V?5?TCR (MR9-4) were
obtained from BD Pharmingen. The OVAI (SIINFEKL) and OVAII
(ISQAVHAAHAEINEAGR) peptides (24, 25) are OVA tumor peptides for
H-2Kband Iab, respectively, whereas Mut1 (FEQNTAQP) peptide is an
irrelevant 3LL lung carcinoma for H-2Kb(26). These peptides were syn-
thesized by Multiple Peptide Systems. The OVA-specific TCR transgenic
OT I and OT II mice and H-2Kb, Iab, CD4, CD8, CD54, and CD80 KO
mice on a C57BL/6 background were obtained from The Jackson Labo-
ratory. Homozygous OT II/H-2Kb?/?, OT II/Iab?/?, OT II/CD54?/?, and
OT II/CD80?/?mice were generated by backcrossing the designated gene
KO mice (H-2Kb) onto the OT II background for three generations; ho-
mozygosity was confirmed by PCR according to The Jackson Laboratory’s
protocols. All mice were maintained in the animal facility at the Saskatoon
Cancer Center and treated according to animal care committee guidelines
of the University of Saskatchewan.
Preparation of DCs
Activated, mature bone marrow-derived DCs, expressing high levels of
MHC class II, CD40, CD54, and CD80, were generated from C57BL/6
mice as described previously (27). To generate OVA-pulsed DCs
(DCOVA), DCs were pulsed overnight at 37°C with 0.1 mg/ml OVA (Sig-
ma-Aldrich, then washed extensively (26).
Preparation of OT II CD4?and OT I CD8?T cells
Naive OVA-specific CD4?T and CD8?T cells were isolated from OT II
or OT I mouse spleens, respectively, and enriched by passage through
nylon wool columns. CD4?and CD8?cells were then purified by negative
selection using anti-mouse CD8 (Ly2) or CD4 (L3T4) paramagnetic beads
(Dynal Biotech) to yield populations that were ?98% CD4?/V?2V?5?or
CD8?/V?2V?5?, respectively. To generate DCOVA-activated CD4?T
cells, CD4?T cells (2 ? 105cells/ml) from OT II mice or designated
gene-deleted OT II mice were stimulated for 3 days with irradiated (4000
rad) bone marrow-derived DCOVA(1 ? 105cells/ml) in the presence of
IL-2 (10 U/ml), IL-12 (5 ng/ml), and anti-IL-4 Ab (10 ?g/ml) (R&D Sys-
tems) (28). These in vitro DCOVA-activated CD4?T cells, also referred to
herein as CD4?Th-APCs, were then isolated by Ficoll-Paque (Sigma-
Aldrich) density gradient centrifugation or further purified using CD4 mi-
crobeads (Miltenyi Biotec) in some experiments. Con A-stimulated OT II
CD4?T (Con A-OT II) cells were similarly generated by incubating
splenocytes from OT II or OT II/KO mice with Con A (1 ?g/ml) and IL-2
(10 U/ml) for 3 days, after which the CD4?T cells were purified on density
gradients. To ascertain that no DCs were in purified Th-APCs or Con A-OT
II cells, these active T cells were further purified by using CD4 microbeads
Phenotypic characterization of DCOVA-activated CD4?T cells
For the phenotypic analyses, Th-APCs were stained with Abs specific for
H-2Kb, Iab, CD3, CD4, CD8, CD11b, CD11c, CD25, CD54, CD69, CD80,
and V?2V?5?TCR (BD Pharmingen), respectively, and analyzed by flow
cytometry. For the intracellular cytokines, cells were restimulated with
4000 rad-irradiated BL27 tumor cells pulsed with OVAII peptide for 4 h
(28), and then processed using a Cytofix/CytoPerm Plus with GolgiPlug kit
Th to CD8?CTL. A, The “passive,” three-cell interac-
tion model, in which APCs simultaneously present Ag
to the Th and the CTL, but deliver costimulatory signals
only to the helper. The CD4?Th cell in turn produces
IL-2, which is required for CTL activation. B, The dy-
namic model of sequential two-cell interactions by
APCs, in which the APC offers costimulatory signals to
the CD4?Th, which reciprocally “licenses” the APC
(left side) such that it can only then directly costimulate
the CTL (right side). C, The new dynamic model of
sequential two-cell interactions, in which APCs license
CD4?Th cells to act as APCs (Th-APCs). APCs di-
rectly transfer MHC class I-Ag complexes and costimu-
latory molecules to expanding populations of IL-2-pro-
ducing Th cells, which thereby act directly as Th1-
APCs to simulate CTL activation.
Three models for the delivery of CD4?
tometry. EG7 (thick solid lines, a) and EL4 (thick dotted lines, a) and
BL6-10OVA(thick solid lines, b) and BL6-10 (thick dotted lines, b) tumor
cells were stained with the rabbit anti-OVA Ab (Sigma-Aldrich), followed
with the FITC-conjugated goat anti-rabbit IgG Ab, and then analyzed by
flow cytometry. Tumor cells stained with normal rabbit serum were used as
control populations (thin dotted lines). One representative experiment of
two is displayed.
Flow cytometric analysis of OVA expression by flow cy-
7498A NEW DYNAMIC MODEL OF TWO-CELL INTERACTIONS BY Th-APCs
(BD PharMingen), with R-PE-conjugated anti-IL-4, -perforin, and -IFN-?
Abs (R&D Systems), respectively. Culture supernatants of the restimulated
Th-APCs were analyzed for IFN-?, IL-2, and IL-4 expression using ELISA
kits (Endogen) as reported previously (26).
In vitro and in vivo membrane molecule transfer assays
In in vitro membrane transfer assay, DCOVAor DCs were incubated with
CFSE (0.5 ?M) at 37°C for 15 min and washed three times with PBS.
CFSE-labeled DCOVAor DCs were incubated with Con A-OT II cells at
37°C for 4 h, then the cell mixtures, the original DCOVAand Con A-OT II
cells, were stained with a panel of PE-energy-coupled dye (ECD) Abs
specific for H-2Kb, CD54, and CD80, respectively, and analyzed by con-
focal fluorescence microscopy. CD4?T cells in the cell mixture were also
purified by cell sorting and analyzed by flow cytometry. Con A-OT II cells
stained with biotin-labeled isotype-matched Abs and ECD-avidin (BD
Pharmingen) were used as controls.
In in vivo membrane transfer assay, naive T cells were isolated from OT
II/Iab?/?and OT II/CD80?/?mouse spleens, respectively, and enriched by
passage through nylon wool columns. The CD4?T cells (5 ? 106cells/
mouse) were further purified by negative selection using the anti-mouse
CD8 (Ly2) paramagnetic beads (Dynal Biotech) and then i.v. injected into
wild-type C57BL/6 mice. One group of mice remained untreated. One day
subsequent to the injection, another group of mice were i.v. immunized
with irradiated (4000 rad) DCOVA(0.2 ? 106cells/mouse). Three days
after the immunization, mice were sacrificed. T cells were isolated from the
spleens of these two groups of mice and enriched by passage through nylon
wool columns. The OVA-specific CD4?OT II T cells were further purified
from these T cells by positive selection using the biotin-anti-TCR Ab and
anti-biotin microbeads (Miltenyi Biotec), and then stained with FITC-anti-
Iaband FITC-anti-CD80 Abs for flow cytometric analysis, respectively.
RF3370 hybridoma cells (0.5 ? 105cells/well) were cultured with irradi-
ated (4000 rad) DCOVAor Th-APCs or Con A-OT II (1 ? 105cells/well)
for 24 h. To investigate the fate of acquired MHC class I/peptide expres-
sion, Th-APCs alone were cultured for 1, 2, and 3 days in culture medium
containing IL-2 (10 U/ml), termed Th-APC (1, 2, and 3 days), and then
harvested for stimulation of RF3370 cells, respectively. The supernatants
were harvested for measurement of IL-2 secretion using an ELISA kit
CD8?T cell proliferation assays
For in vitro CD8?T cell proliferation assay, irradiated (4000 rad) stimu-
lators, the Th-APCs, Con A-OT II cells (0.4 ? 105cells/well), DCOVA
(0.1 ? 105cells/well), and their 2-fold dilutions, were cultured with a
constant number of responders, the naive OT I or C57BL/6 (B6) CD8?T
cells (0.5 ? 105cells/well). To rule out the potent effect of endogenous
H-2Kb, Th-APCs generated from H-2Kb?/?OT II T cells were termed
Kb?/?Th-APCs and used as stimulator. In some experiments, each of a
panel of neutralizing reagents (anti-IL-2, -H-2Kbor -LFA-1 Abs, and
CTLA-4/Ig fusion protein) (each 15 ?g/ml; R&D Systems) or a mixture of
the above reagents were added to the cells, while control cells received a
mixture of isotype-matched irrelevant Abs and fusion protein. In other
experiments, the irradiated CD4?Th-APCs and naive OT I CD8?T cells
were cultured in Transwell plates (Costar), separated by 0.4-?m pore-sized
membranes. After 48 h, thymidine incorporation was determined by liquid
scintillation counting (26).
For the in vivo CD8?T cell proliferation assay, purified naive OT I
CD8?T cells were labeled with CFSE (1.5 ?M) and i.v. injected into
C57BL/6 mice (2 ? 106cells each). Twelve hours later, each mouse was
i.v. injected with 2 ? 106Th-APCs and Con A-OT II cells, respectively,
or 0.2 ? 106DCOVA. In another group, mice were injected with PBS.
Three days later, the splenic T cells from the recipients were stained with
ECD-anti-CD8 Ab (Beckman Coulter) and then analyzed by flow
For the in vitro cytotoxicity assay, the activated CD8?T cells derived from
the above 3-day coculture with irradiated (4000 rad) DCOVA, Th-APCs,
and Con A-OT II cells were purified on density gradients and termed
DCOVA/OT I, Th-APC/OT I, and Con A-OT II/OT I, respectively. These
cells as well as Th-APCs were used as effector cells, while51Cr-labeled
EG7, the control EL4 tumor cells, DCOVA, LB27, and OVAII-pulsed LB27
(LB27OVAII) tumor cells were used as target cells, respectively. Specific
killing was calculated as: 100 ? [(experimental cpm ? spontaneous cpm)/
(maximal cpm ? spontaneous cpm)] as previously described (26).
We adapted a recently reported in vivo cytotoxicity assay (29). Briefly,
C57BL/6 mice were i.v. immunized with DCOVA(0.5 ? 106cells), Th-
APCs or Con A-OT II cells (2 ? 106cells). Seven days later, mice were
boosted once. In another group, mice were injected with PBS. Naive mouse
splenocytes were incubated with either high (3.0 ?M, CFSEhigh) or low
(0.6 ?M, CFSElow) concentrations of CFSE to generate differentially la-
beled target cells. The CFSEhighcells were pulsed with OVAI, whereas the
CFSElowcells were pulsed with the irrelevant 3LL lung carcinoma H-2Kb
peptide Mut1 and served as internal controls. These peptide-pulsed target
cells were washed extensively to remove free peptide and then i.v. coin-
jected at 1:1 ratio into the above described immunized mice 3 days after the
boost. Sixteen hours after target cell delivery, the spleens were removed
and residual CFSEhighand CFSElowtarget cells remaining in the recipients’
spleens were sorted and analyzed by flow cytometry.
Wild-type C57BL/6 mice (n ? 8) were injected i.v. with 0.2 ? 106DCOVA,
2 ? 106Th-APCs, and Con A-OT II cells, respectively, and then 7 days
later they were boosted once. To study the immune mechanism, CD4 and
CD8 KO mice (n ? 8) were injected i.v. with 2 ? 106Th-APCs and then
7 days later the mice were boosted once. Three days subsequent to the
boost, the mice were i.v. given 0.5 ? 106BL6-10OVAor BL6-10 tumor
cells. The mice were sacrificed 4 wk after tumor cell injection and the lung
metastatic tumor colonies were counted in a blind fashion (22). Metastases
on freshly isolated lungs appeared as discrete black-pigmented foci that
were easily distinguishable from normal lung tissues and confirmed by
histological examination. Metastatic foci too numerous to count were as-
signed an arbitrary value of ?100.
CD4?Th-APCs acquire the synapse-composed MHC class II
and CD54 molecules and the bystander MHC class I from APCs
by APC stimulation
To explore DC membrane-derived APC machinery acquisition by
CD4?T cells, Con A-stimulated CD4?T cells from OVA-specific
TCR-transgenic OT II mice were cultured for 4 h. either alone or
with DCOVAor DC. The CD4?T cells were then sorted and ex-
amined for expression of MHC class I and II, CD54, and CD80 by
flow cytometry. The control Con A-stimulated OT II CD4?T cells
expressed some MHC class I and II, CD54, and CD80. However,
following incubation with DCOVA, these T cells displayed mod-
erately augmented levels of these molecules (Fig. 3A), suggesting
that DC molecules could have been transferred to the T cells. The
membrane transfer can be mostly blocked by addition of anti-H-
2Kband LFA-1 Abs and CTLA-4/Ig fusion protein (data not
shown), indicating that the membrane acquisition of Th-APCs
from DCOVAis mediated by TCR and costimulatory molecules. In
addition, these T cells following interaction with DCs without
OVA pulsing also displayed augmented levels of these molecules,
but to a lesser extent (data not shown), indicating that these DC
molecule transfer is mediated by both the Ag-specific and nonspe-
Since all T cells express MHC class I and CD54 and some
activated T cells also express MHC class II and CD80 molecules
(30, 31), it was necessary to confirm that the increased levels of T
cell-associated MHC class I and II, CD54, and CD80 were not due
to endogenous T cell up-regulation of these molecules. Thus, we
incubated CFSE-labeled DCOVAwith Con A-stimulated CD4?T
cells derived from OT II mice with homozygous H-2Kb, Iab,
CD54, and CD80 gene KO, respectively, then sorted the T cells
and assessed their expression of these markers. The T cells did not
express their respectively deleted gene products when cultured
alone, but did discernibly express H-2Kb, Iab, CD54, and CD80
after a 4-h incubation with DCOVA, as determined by flow cytom-
etry (Fig. 3B) or confocal fluorescence microscopy (Fig. 4). These
results indicate that, besides previously reported MHC class I
transferred onto CD8?T cells during DC-CD8?T cell interaction
and MHC class II and CD80 molecules transferred onto CD4?T
7499The Journal of Immunology
cells during DC-CD4?T cell interaction (21, 32, 33), CD4?T
cells can also acquire CD54 forming the immune synapse (18, 19)
as well as the bystander MHC class I molecules from DCs after
DC stimulation of CD4?T cells. In addition to the mechanism of
Ag-specific MHC-TCR-mediated internalization and recycling
(20, 21), the uprooting of APC molecules or APC-released vesicles
may also contribute to the above described membrane transfer,
especially the bystander MHC class I (34).
We then examined whether naive T cells can also acquire DC
Ag-presenting machinery in culture. Naive OT II CD4?T cells
were first purified by using a nylon column to remove DCs and B
cells and anti-CD8 paramagnetic beads (Dynal Biotech) to remove
CD8?T cells and then incubated for 3 days with irradiated
DCOVA. The activated OT II CD4?T cells were then purified by
using Ficoll-Paque density gradient centrifugation and CD4 mi-
crobeads (Miltenyi Biotec) and then analyzed by flow cytometry.
These T cells, which proliferated in response to DCOVAstimula-
tion, expressed cell surface CD4, CD25, and CD69 and intracel-
lular perforin and IFN-?, but not IL-4 (Fig. 3C); they also secreted
IFN-? (?2 ng/ml per 106cells/24 h) and IL-2 (?2.5 ng/ml per 106
cells/24 h), but not IL-4, in culture. These data indicate that these
OVA-TCR-transgenic CD4?T cells were Th1. In addition, there
was no CD11b?/11c?DC population existing in these purified
CD4?T cells (Fig. 3C). This is because that any survival of irra-
diated DCOVAcells and the potential small amount of contamina-
tion of spleen DCs or B cells within the original naive OT II CD4?
T cell preparation, which might have picked up OVA peptides
from irradiated DCOVAin the culture, would be eliminated by the
cells. A, CFSE-labeled DCOVAwere incubated with Con A-stimulated
CD4?T cells from OT II mice. T cells with (thick solid lines) and without
(thick dotted lines) incubation of DCOVAwere stained with Abs and ana-
lyzed for expression of H-2Kb, Iab, CD54, and CD80 by flow cytometry,
respectively. B, CFSE-labeled DCOVAwere incubated with Con A-stimu-
lated CD4?T cells from H-2Kb, Iab, CD54, and CD80 gene KO OT II
mice, respectively. T cells with (thick solid lines) and without (thick dotted
lines) incubation of DCOVAwere stained with Abs and analyzed for ex-
pression of the above molecules, respectively. T cells with incubation of
DCOVAwere also stained with isotype-matched Abs and used as control
populations (thin dotted lines). C, DCOVA-activated CD4?T cells (Th-
APCs) from OT II mice were stained with a panel of Abs (thick solid lines)
and analyzed by flow cytometry. The control CD4?T cells (thin dotted
lines) were only stained with isotype-matched Abs. D, DCOVA-activated
CD4?T cells (Th-APCs) from H-2Kb, Iab, CD54, and CD80 gene KO OT
II mice, respectively, were stained with a panel of Abs (thick solid lines).
The control CD4?T cells (thin dotted lines) were only stained with iso-
type-matched Abs. One representative experiment of two in the different
experiments above is shown.
Transfer of DC membrane molecules to active CD4?T
microscopy. CFSE-labeled DCOVAwere incubated with Con A-stimulated
CD4?T cells from H-2Kb(A), CD54 (B), and CD80 gene KO OT II (C)
mice, stained with fluorochrome-labeled Abs, and analyzed by confocal
fluorescence microscopy. Images include DCs (larger cells) alone, T
(smaller) cells alone, or a mixture of DCs and T cells (a) under differential
interference contrast, with a cell surface stain consisting of ECD (red) Ab
for either H-2Kb, CD54, or CD80 (b), with cytoplasmic CFSE stain (green,
c), and with both stains (d). Our data confirm that 1) DCOVA(larger cells),
but not gene-deleted T cells (smaller cells), express H-2Kb, CD54, and
CD80 molecules (arrows) and 2) during coculture of DCOVAwith T cells,
the T cells acquire H-2Kb, CD54, and CD80 molecules (arrowheads). One
representative experiment of two is shown.
Membrane acquisition analysis by confocal fluorescence
7500 A NEW DYNAMIC MODEL OF TWO-CELL INTERACTIONS BY Th-APCs
killing activity of these activated Th1 cells expressing perforin (see
Fig. 7B) (35, 36). In addition to the common H-Kbexpression,
these Th cells also expressed Iab, CD54, and CD80 molecules and
here too they did so whether they were derived from wild-type or
homozygous H-2Kb?/?, Iab?/?, CD54?/?, or CD80?/?KO mice
(Fig. 3D). Thus, we clearly demonstrate that naive CD4?T cells
can also acquire MHC class II and costimulatory molecules (CD54
and CD80) composing the immune synapse as well as the by-
stander MHC class I from DCs by in vitro DC stimulation.
To further confirm the membrane acquisition in vivo, wild-type
C57BL/6 mice were first injected with purified CD4?OT II/
Iab?/?and OT II/CD80?/?T cells and then immunized with
DCOVA. Three days after the immunization, mice were sacrificed.
CD4?OT II T cells were purified from these immunized mouse
spleens and then stained with FITC-anti-Iaband FITC-anti-CD80
Abs for flow cytometric analysis, respectively. As shown in Fig. 5,
CD4?OT II/Iab?/?and OT II/CD80?/?T cells derived from mice
immunized with DCOVAbecame slightly Iaband CD80 positive,
rified from OT II/Iab?/?and OT II/CD80?/?mice were transferred into
wild-type C57BL/6 mice, respectively. The first group of mice was un-
treated and the second group of mice was immunized with DCOVA. The
CD4?OT II/Iab?/?and OT II/CD80?/?T cells were then purified from the
first (thick dotted lines) and the second group (solid lines) of mice and then
stained with the FITC-anti-Iaband FITC-anti-CD80 Abs and the FITC-
conjugated isotype-matched Abs (thin dotted lines) for flow cytometric
analysis, respectively. One representative experiment of three is shown.
In in vivo membrane transfer assay. The CD4?T cells pu-
OT I CD8?T cells. A, MHC class I presentation of
OVA to RF3370 hybridoma by Th-APCs. The amount
of IL-2 secretions of stimulated RF3370 cells in ex-
amining wells was subtracted by the amounts of IL-2
in wells containing DCOVA, Th-APC, and Con A-OT
II alone, respectively. ?, p ? 0.01 (Student’s t test) vs
cohorts of Con A-OT II. B, In vitro CD8?T cell pro-
liferation assay. Varying numbers of irradiated Th-
APCs, Kb?/?Th-APCs, Con A-OT II, and DCOVA
cells were cocultured with naive OT I or B6 CD8?T
cells. After 3 days, the proliferative responses of the
CD8?T cells were determined by [3H]thymidine up-
take assays. C, Th-APCs were cultured with OT I
CD8?T cells either separated in Transwells or not (all
other bars). In the latter cultures, the impact on OT I
CD8?T cell proliferation of adding each of the neu-
tralizing reagents, all neutralizing reagents together
(mixed reagents), or all control Abs and fusion pro-
teins (control reagents) was assessed. In one set of
wells, supernatants from cultured Th-APCs (superna-
tants) were added to the CD8?T cells in place of the
Th-APCs themselves. ?, p ? 0.01 (Student’s t test) vs
cohorts of Th-APC. D, In vivo CD8?T cell prolifer-
ation assay. CFSE-labeled OT I CD8?T cells were
i.v. injected into C57BL/6 mice. Twelve hours later,
each mouse was i.v. given Th-APCs or Con A-OT II
cells or DCOVAor PBS, then 3 days later the numbers
of division cycles of the CFSE-labeled CD8?T cells
in the recipient spleens were determined by flow cy-
tometry. One representative experiment of three in the
above experiments is shown.
CD4?T-APCs stimulate RF3370 and
7501The Journal of Immunology
respectively, whereas these T cells derived from mice without im-
munization remained Iaband CD80 negative, indicating that CD4?
OT II T cells acquire Iaband CD80 molecules by in vivo DCOVA
Th-APCs stimulate CD8?T cell proliferation in vitro and in
If the CD4?T cell-acquired H-2Kbwere in the form of OVAI
peptide complexes and the DC costimulatory molecules were func-
tional, then these IL-2-secreting T cells could potentially act as
direct APCs (termed CD4?Th-APCs) for CD8?T cell stimula-
tion. To examine the functionality of these putative Th-APC cells,
we initially assessed their ability to stimulate IL-2 secretion of T
cell hybridoma RF3370. As shown in Fig. 6A, RF3370 cells alone
did not secret IL-2. However, Th-APCs significantly stimulated
RF3370 to secret IL-2 (95 pg/ml) as did DCOVA(220 pg/ml),
indicating that Th-APCs expressed functional H-2Kb-OVAI pep-
tide complexes. If the acquired MHC I-OVAI peptide complexes
are of any functional significance in vivo, they would need to be
stably expressed. The rate of their decay was assessed by culturing
these Th-APCs after MHC class I acquisition for varying time
periods. As shown in Fig. 6A, the ability to stimulate IL-2 secretion
of RF3370 cells did decay over time. However, readily detectable
MHC class I/peptide expression was still observed as much as 3
days after in vitro culture.
To further confirm it, we then assessed their ability to induce
proliferation of naive OT I CD8?T cells in vitro. The positive
control DCOVAcells that were previously demonstrated to possess
a highly activated phenotype (27) strongly induced OT I cell pro-
liferation (Fig. 6B). DCOVA-activated CD4?Th-APCs, which
were purified by Ficoll-Paque density gradient centrifugation and
using CD4 microbeads, did indeed stimulate proliferation of OT I
CD8?T cells, but to a lesser extent due to 1) fewer costimulatory
molecules and 2) lacking the third signal, DC-secreted IL-12 (37),
compared with DCOVA. However, they did not stimulate responses
of the control naive C57BL/6 (B6) mouse CD8?T cells, nor did
Con A-stimulated OT II CD4?T (Con A-OT II) cells (secreting
IFN-? (?4.0 ng/ml per 106cells/24 h) and IL-2 (?3.3 ng/ml per
106cells/24 h), but lacking self-IL-4 and acquired H-2Kb-OVA
peptide complexes) stimulate OT I CD8?T cell proliferation. In
addition, Kb?/?Th-APCs derived from the H-2Kb?/?OT II KO
mice (Fig. 3D) showed similar CD8?T cell stimulatory activity as
Th-APCs derived from the wild-type OT II mice (Fig. 6B), indi-
cating that the activation of CD8?OT I T cells is mediated via the
acquired H-2Kb-OVA peptide complexes, but not the endogenous
H-2Kbof Th-APCs. In separate experiments, we demonstrated that
CD8?T cell stimulatory activity of the Th-APCs was contact de-
pendent since Transwells blocked CD8?T cell proliferation (Fig.
6C). Furthermore, adding anti-MHC class I or anti-LFA-1 Abs or
CTLA-4/Ig fusion protein could significantly inhibit the OT I
CD8?T cell proliferative response in the cocultures by 38, 50, and
58%, respectively, whereas anti-IL-2 Ab had less effect (19% in-
hibition; p ? 0.01). Simultaneous addition of all blocking reagents
reduced the proliferative response by 92% (p ? 0.01). Taken to-
gether, these data indicate that this response is critically dependent
on H-2Kb/OVAI/TCR specificity and is greatly affected by non-
specific costimulatory CD54-LFA-1 and CD80-CD28 interactions
between the CD4?Th-APCs and CD8?T cells. That this prolif-
erative effect was not simply an in vitro artifact was confirmed by
demonstrating that these Th1-APCs can also stimulate prolifera-
tive responses in vivo. We adoptively transferred CFSE-labeled
naive OT I CD8?T cells into mice that were also given Th-APCs,
Con A-OT II cells, DCOVA, or PBS. The labeled CD8?T cells did
not show any division in mice treated with PBS (data not shown).
However, the labeled CD8?T cells underwent some cycles of cell
division in the mice given either Th-APCs or DCOVA, but did not
respond in the animals given Con A-OT II cells (Fig. 6D).
Th-APCs stimulate CD8?T cell differentiation into CTL
effectors in vitro and in vivo
As a critical test of the functionality of these purified CD4?Th-
APCs, we tested their ability to induce the differentiation of naive
OT I CD8?T cells into CTL effectors, as determined using in vitro
51Cr release assays with EG7 tumor cells expressing an OVA
transgene. The Th-APC-activated OT I CD8?T (Th-APC/OT I)
cells displayed substantial cytotoxic activity (33% specific killing;
E:T ratio, 12) against an OVA-expressing EG7 cell line as did the
DCOVA-activated OT I CD8?T (DCOVA/OT I) cells (46% killing;
E:T ratio, 12), but not against its parental EL4 tumor cells (Fig.
7A), indicating that the killing activity of these CTLs is OVA
tumor specific. In addition, these CD4?Th-APCs expressing per-
forin (Fig. 3C) displayed killing activities for DCOVA and
LB27OVAIIcells with Iab/OVAII expression (Fig. 7B). However,
they themselves did not show any killing activity to LB27 and EG7
(Fig. 7B) or BL6-10OVAcells (data not shown) without Iab/OVAII
expression. As with the proliferation assays, the in vitro CD8?
CTL induction capacity of CD4?Th-APCs can also be translated
into an induction of effector CTL function in vivo. We adoptively
activity in vitro and in vivo. In vitro cytotoxicity assay. A, Three types of
activated CD8?T cells (DCOVA/OT I, Th-APC/OT I, and Con A-OT II/OT
I) were used as effector (E) cells, whereas51Cr-labeled EG7 or control EL4
tumor cells used as target (T) cells. B, Th-APCs were used as effector cells,
whereas51Cr-labeled EG7, DCs, DCOVA, LB27, and EG7OVAIIcells used
as target cells. The data are presented as the percent specific target cell lysis
in51Cr release assay. Each point represents the mean of triplicate cultures.
C, In vivo cytotoxicity assay. C57BL/6 splenocytes differentially labeled to
CFSEhighand CFSElowwere pulsed with OVAI and Mut1 peptide, respec-
tively. These splenocytes were then i.v. injected at a ratio of 1:1 into mice
immunized with DCOVA, Th-APCs, or Con A-OT II cells or PBS. Sixteen
hours later, the CFSEhighor CFSElowcells remaining in the spleens were
determined by flow cytometry. The value in each panel represents the
percentage of CFSEhighcells vs CFSElowcells remaining in the spleens.
CD4?T-APC induce the development of Ag-specific CTL
7502 A NEW DYNAMIC MODEL OF TWO-CELL INTERACTIONS BY Th-APCs
transferred OVAI peptide-pulsed splenocytes that had been
strongly labeled with CFSE (CFSEhigh), as well as the control pep-
tide Mut1-pulsed splenocytes that had been weakly labeled with
CFSE (CFSElow), into recipient mice that had been vaccinated
with these purified Th-APCs, DCOVA, Con A-OT II cells, or PBS.
We assessed the disappearance of the labeled cells from the mice
by flow cytometric analysis and found that the CFSElow(irrelevant
Mut1 peptide-pulsed) cells were unaffected by the vaccination pro-
tocol. In addition, we found that no substantial loss (1%) of the
CFSEhigh(OVAI peptide-pulsed) cells from the PBS-immunized
mice (data not shown). However, there was substantial loss of the
CFSEhigh(OVAI peptide-pulsed) cells from the Th-APC-immu-
nized (86%) or DCOVA-vaccinated (97%) mice, but not from the
Con A-OT II cell-vaccinated (2%) mice (Fig. 7C). These data in-
dicate that CD4?Th-APCs carrying H-2Kb-OVAI complexes and
DC costimulatory molecules can stimulate the development of
OVA-specific CTL effector cells in vivo.
Th-APCs induce OVA-specific antitumor immunity in vivo
In addition, Th-APCs can also stimulate OVA-specific CTL-me-
diated antitumor immunity in vivo. We injected these purified Th-
APCs i.v. into mice, followed by i.v. challenge with OVA-express-
ing BL6-10OVAor OVA-negative BL6-10 tumor cells. All mice
immunized with Con A-OT II cells (i.e., cells lacking acquired
H-2Kb-OVAI complexes and costimulatory molecules) as well as
the control mice (eight of eight) without any immunization had
large numbers (?100) of lung metastatic tumor colonies 4 wk after
tumor cell challenge (Expt. 1 in Table I and Fig. 8). In addition, all
mice (eight of eight) immunized with naive OT II T cells also died
of lung metastasis (data not shown). However, all mice (eight of
eight) immunized with Th-APCs had no lung tumor metastasis.
DCOVAimmunization was equally effective in inducing antitumor
immunity. The specificity of the protection was confirmed with the
observation that Th-APCs did not protect against BL6-10 tumors
that did not express OVA, with all mice having large numbers
(?100) of lung metastatic tumor colonies after tumor cell chal-
lenge. To study the immune mechanism, CD4 and CD8 KO mice
were used for immunization of Th-APCs. As shown in experiment
2 in Table I, all of the CD4 KO mice (eight of eight) were still
protected from BL6-10OVAtumor challenge, indicating that acti-
vation of CD8?CTL response by Th-APCs is independent on the
host CD4?T cells. However, all CD8 KO mice (eight of eight) had
numerous lung tumor metastases, indicating that the Th-APCs-
driven antitumor immunity is mediated by CD8?CTLs. The Th-
APC-induced CD8?CTL response is more likely through direct
interaction between Th-APCs and CD8?CTLs rather than cross-
presentation of the host DCs picking up OVA peptides released
from Th-APCs, because the former is CD4?T cell independent,
whereas the latter is CD4?T cell dependent.
A long-standing paradox in cellular immunology has been the con-
ditional requirement for CD4?Th cells in priming of CD8?CTL
responses. CTL responses to noninflammatory stimuli (e.g., MHC
class I alloantigen Qa-1, the male HY Ag) are CD4?T cell de-
pendent (2, 38, 39). Our data clearly demonstrate the critical helper
requirement for CTL induction, as have two other recent reports.
Wang and Livingstone (40) showed that the primary CD8?T cell
responses to Ags presented in vivo by peptide-pulsed DCs are also
Table I. Vaccination with CD4?Th-APC protects against lung tumor metastases in mice
ImmunizationTumor Cell Challenge
Median Number of Lung
Con A-OT II cells
Th-APCs (B6 mice)
Th-APCs (CD4 KO)
Th-APCs (CD8 KO)
aIn experiment 1, C57BL/6 mice (n ? 8) were immunized with DCOVA, Th-APCs, and Con A-OT II cells or PBS. Following
the immunization, each mouse was challenged i.v. with OVA transgene-expressing (BL6-10OVA) or wild-type BL6-10 tumor
cells. The mice were sacrificed 4 wk after tumor cell challenge and the numbers of lung metastatic tumor colonies were counted.
One representative experiment of two is shown.
bIn experiment 2, wild-type C57BL/6 (B6) and CD4 or CD8 KO mice (n ? 8) were immunized with Th-APCs. Following
the immunization, each mouse was challenged i.v. with OVA transgene-expressing (BL6-10OVA) tumor cells. The mice were
sacrificed 4 wk after tumor cell challenge and the numbers of lung metastatic tumor colonies were counted. One representative
experiment of two is shown.
with Th-APCs. Pulmonary metastases were formed in different groups of
immunized mice by i.v. injection of 0.5 ? 106BL6-10OVAor BL6-10
tumor cells. Four weeks later, mouse lungs were removed. The extent of
lung metastasis in six different groups of mice described in experiment 1 in
Table I is displayed.
Immune protection of lung metastasis in mice immunized
7503 The Journal of Immunology
dependent on help from CD4?T cells. More important, Behrens et
al. (41) have demonstrated that coinjection of Ag-presenting DC-
activated, but not naive, CD4?OT II T cells induces CTL re-
sponses against islet ? cell OVA Ag and leads to diabetes in rat
insulin promoter OVAhightransgenic mice. They also found that
activated CD4?OT II T cells provide CD40-mediated help to
CD8?T cell responses without these T cells necessarily seeing Ag
on the same APC (41). In contrast, some have suggested that
CD4?T cell help is only essential for memory CTL responses
(29). Thus, the generation of effectors from naive CD8?T cells is
reported to be helper independent in mice immunized with irradi-
ated embryonic cells expressing an adenovirus type 5 E1A trans-
gene (42). Having said that, it is highly relevant that such adeno-
viral challenge would also introduce potent inflammatory signals
into the sensitizing microenvironment (leading to high level DC
maturation) (43), to say nothing of the potential for help from NK
cells (44). In addition, the E1A adenoviral Ag features multiple
CD8?T cells epitopes (45) and therefore also a greater base of
Ag-specific CD8?T cell precursors from which to draw (46). A
strong and direct activation of DCs (47) would explain the previ-
ous demonstrations that induction of some antiviral CTL responses
is CD4?Th cell independent.
T cell-to-T cell (T-T) Ag presentation, dependent upon activated
CD4?T cells first acquiring MHC class II and CD80 molecules
from APCs and then stimulating other CD4?T cells, is increas-
ingly attracting attention (32, 33). However, the roles such T-APCs
may play in vivo have been as yet ill-defined and the results of the
relevant in vitro studies disparate, in part because multiple exper-
imental systems have been used. For example, CD4?T-APCs can
induce IL-2 production and proliferative responses among naive
responder T cells (48, 49), which is consistent with our results in
this study. However, these T-APCs have also been shown to in-
duce apoptosis in activated CD4?T cells or anergization of CD4?
T cell lines (33, 50–52). In this study, we found that in vivo trans-
fer of CD4?Th1-APCs expressing high levels of IFN-? and IL-2,
which were generated by incubation of OT II CD4?T cells with
DCOVAin the presence of IL-12 and anti-IL-4 Ab, were able to
stimulate OVA-specific CTL responses. Interesting, we also found
that in vivo transfer of CD4?Th2-APCs expressing high levels of
IL-4 and IL-10, which were generated by incubation of OT II
CD4?T cells with DCOVAin the presence of IL-4 and anti-IFN-?
Ab, were able to induce OVA-specific immune suppression (data
not shown). In other reports, however, in vivo transfer of CD4?
Th1-APCs derived from IL-2-dependent transformed T cell lines
has been reported to induce immunosuppressive, but not immuno-
stimulatory effects in the context of autoimmune responses (52,
53). In these studies, the T-APCs used were derived from rather
uncharacterized Con A-stimulated allogeneic or Ag-pulsed CD4?
T cell lines. Therefore, it is difficult to assess the extent to which
they are representative of T-APCs as they would be generated in
vivo. In addition, these studies have addressed only the activation
of CD4?T cell responses.
In this study, we have shown that CD4?T cells can acquire
synapse-composed MHC class II, CD54, and CD80 molecules
from APCs by APC stimulation. In addition, for the first time, we
have shown that CD4?T cells can also acquire the bystander
MHC class I-OVAI peptide complexes which are critical mole-
cules in stimulation of OVA-specific CTL responses. Furthermore,
we have provided a complete line of evidence that compellingly
substantiates the practical aspects of CD4?T cells acting as APCs
for effective CD8?T cell responses in vitro and in vivo. A model
of CD4?T cell help for CTL induction that takes these observa-
tions into account would address multiple important aspects of this
paradigm in cellular immunology. A central caveat in models of
CD4?T cell help for CTL responses is that of scarcity, or how rare
Ag peptide-carrying DCs, Ag-specific CD4?, and Ag-specific
CD8?T cells manage to encounter each other with enough effi-
ciency to ensure that we expeditiously and appropriately respond
to all Ags/pathogens (i.e., to maintain the integrity of the organ-
ism). It is counterintuitive that a function as critical as this not be
optimized in some way. The model wherein APCs that are them-
selves licensed by Th cells to directly activate CD8?T cells (Fig.
1B) (5) offers the advantage that a single licensed APC can contact
multiple CD8?T cells and thereby expand the activation signal.
However, a very limited number of DCs arriving in lymph nodes
would interact with many CD4?T cells, and our evidence dem-
onstrates that they both induce marked proliferative responses
among the naive Ag-specific CD4?T cell population and also
bestow on them of these progeny Th-APC functionality. In turn,
each new Th-APC can interact with and activate naive CD8?CTL
precursor cells, such that they also undergo expansion. The gain in
this system is thereby dramatically increased even before the
newly activated CTL precursors begin to proliferate. Other aspects
of this new model also fit in well with the practical and theoretical
constraints of Th-cell-dependent CTL responses in the host. Ex-
perimental evidence clearly shows that provision of IL-2 dramat-
ically augments the efficiency of precursor CTL expansion (2–4).
We have shown that Th-APCs produce IL-2, and our data explain
quite simply how CD4?Th cells’ IL-2 would be efficiently and
precisely targeted to Ag-specific CD8?T cells. It also addresses
the requirement for cognate CD4?T cell help for CD8?CTL
precursors (3, 4, 54), with the APCs in this case being by definition
a cognate Th cell.
Taken together, this study clearly delineates the role CD4?Th-
APCs can play in stimulation of CD8?CTL responses. It also
provides a solid experimental foundation for each of the tenants of
a new dynamic model of sequential two-cell interactions by CD4?
Th-APCs in Th-cell-dependent CTL immune responses. Not only
are Th-APC effective inducers of Ag-specific CTL activity in vitro,
but also they efficiently induce protective antitumor immunity in
vivo, thereby confirming their physiological relevance. Although
we have addressed multiple parameters of this new model in the
context of Th cell-dependent CTL responses, in principle its con-
ditions could be equally well met in regulatory T cell-dependent
tolerance induction. From our perspective, we are most interested
in its impact on the study of antitumor immunity, cancer vaccine
development, and other immune disorders (e.g. autoimmunity).
We thank J. Gordon for useful comments in preparation of this manuscript
and Y. Wei and M. Boyd for help in confocal fluorescence microscopy and
flow cytometry, respectively.
The authors have no financial conflict of interest.
1. Pardoll, D. 1998. Cancer vaccines. Nat. Med. 4(Suppl. 5): 525–531.
2. Keene, J. A., and J. Forman. 1982. Helper activity is required for the in vivo
generation of cytotoxic T lymphocytes. J. Exp. Med. 155: 768–782.
3. Mitchison, N. A., and C. O’Malley. 1987. Three-cell-type clusters of T cells with
antigen-presenting cells best explain the epitope linkage and noncognate require-
ments of the in vivo cytolytic response. Eur. J. Immunol. 17: 1579–1583.
4. Bennett, S. R., F. R. Carbone, F. Karamalis, J. F. Miller, and W. R. Heath. 1997.
Induction of a CD8?cytotoxic T lymphocyte response by cross-priming requires
cognate CD4?T cell help. J. Exp. Med. 186: 65–70.
5. Ridge, J. P., F. Di Rosa, and P. Matzinger. 1998. A conditioned dendritic cell can
be a temporal bridge between a CD4?T-helper and a T-killer cell. Nature 393:
6. Bousso, P., and E. Robey. 2003. Dynamics of CD8?T cell priming by dendritic
cells in intact lymph nodes. Nat. Immunol. 4: 579–585.
7504A NEW DYNAMIC MODEL OF TWO-CELL INTERACTIONS BY Th-APCs
7. Bennett, S. R., F. R. Carbone, F. Karamalis, R. A. Flavell, J. F. Miller, and
W. R. Heath. 1998. Help for cytotoxic-T-cell responses is mediated by CD40
signalling. Nature 393: 478–480.
8. Schoenberger, S. P., R. E. Toes, E. I. van der Voort, R. Offringa, and C. J. Melief.
1998. T-cell help for cytotoxic T lymphocytes is mediated by CD40-CD40L
interactions. Nature 393: 480–483.
9. Yang, Y., and J. M. Wilson. 1996. CD40 ligand-dependent T cell activation:
requirement of B7-CD28 signaling through CD40. Science 273: 1862–1864.
10. Bretscher, P. A. 1999. A two-step, two-signal model for the primary activation of
precursor helper T cells. Proc. Natl. Acad. Sci. USA 96: 185–190.
11. Ingulli, E., A. Mondino, A. Khoruts, and M. K. Jenkins. 1997. In vivo detection
of dendritic cell antigen presentation to CD4?T cells. J. Exp. Med. 185:
12. Ruedl, C., P. Koebel, M. Bachmann, M. Hess, and K. Karjalainen. 2000. Ana-
tomical origin of dendritic cells determines their life span in peripheral lymph
nodes. J. Immunol. 165: 4910–4916.
13. Norbury, C. C., D. Malide, J. S. Gibbs, J. R. Bennink, and J. W. Yewdell. 2002.
Visualizing priming of virus-specific CD8?T cells by infected dendritic cells in
vivo. Nat. Immunol. 3: 265–271.
14. Tite, J. P. 1990. Evidence of a role for TNF-? in cytolysis by CD4?, class II
MHC-restricted cytotoxic T cells. Immunology 71: 208–212.
15. Rathmell, J. C., M. P. Cooke, W. Y. Ho, J. Grein, S. E. Townsend, M. M. Davis,
and C. C. Goodnow. 1995. CD95 (Fas)-dependent elimination of self-reactive B
cells upon interaction with CD4?T cells. Nature 376: 181–184.
16. Lanzavecchia, A., and F. Sallusto. 2000. Dynamics of T lymphocyte responses:
intermediates, effectors, and memory cells. Science 290: 92–97.
17. Lenschow, D. J., T. L. Walunas, and J. A. Bluestone. 1996. CD28/B7 system of
T cell costimulation. Annu. Rev. Immunol. 14: 233–258.
18. Grakoui, A., S. K. Bromley, C. Sumen, M. M. Davis, A. S. Shaw, P. M. Allen,
and M. L. Dustin. 1999. The immunological synapse: a molecular machine con-
trolling T cell activation. Science 285: 221–227.
19. Viola, A., S. Schroeder, Y. Sakakibara, and A. Lanzavecchia, 1999. T lympho-
cyte costimulation mediated by reorganization of membrane microdomains. Sci-
ence 283: 680–682.
20. Hwang, I., J. F. Huang, H. Kishimoto, A. Brunmark, P. A. Peterson,
M. R. Jackson, C. D. Surh, Z. Cai, and J. Sprent. 2000. T cells can use either T
cell receptor or CD28 receptors to absorb and internalize cell surface molecules
derived from antigen-presenting cells. J. Exp. Med. 191: 1137–1148.
21. Huang, J. F., Y. Yang, H. Sepulveda, W. Shi, I. Hwang, P. A. Peterson,
M. R. Jackson, J. Sprent, and Z. Cai. 1999. TCR-mediated internalization of
peptide-MHC complexes acquired by T cells. Science 286: 952–954.
22. Kimura, A. K., and J. H. Xiang. 1986. High levels of Met-72 antigen expression:
correlation with metastatic activity of B16 melanoma tumor cell variants. J. Natl.
Cancer Inst. 76: 1247–1254.
23. Mitchell, D., S. Nair, and E. Gilboa. 1998. Dendritic cell/macrophage precursors
capture exogenous antigen for MHC class I presentation by dendritic cells. Eur.
J. Immunol. 28: 1923–1933.
24. Li, M., G. M. Davey, R. M. Sutherland, K. Christian, A. M. Lew, C. Hirst,
F. R. Carbone, and R. H. William. 2001. Cell-associated ovalbumin is cross-
presented much more efficiently than soluble ovalbumin in vivo. J. Immunol. 166:
25. Slingluff, C. 1996. Tumor antigens and tumor vaccines: peptides as immunogens.
Semin. Surg. Oncol. 12: 446–453.
26. Zhang, W., Z. Chen, F. Li, H. Kamencic, B. Juurlink, J. R. Gordon, and J. Xiang.
2003. Tumour necrosis factor-? (TNF-?) transgene-expressing dendritic cells
(DCs) undergo augmented cellular maturation and induce more robust T-cell
activation and anti-tumour immunity than DCs generated in recombinant TNF-?.
Immunology 108: 177–188.
27. Liu, Y., H. Huang, Z. Chen, L. Zong, and J. Xiang. 2003. Dendritic cells engi-
neered to express the Flt3L stimulate type 1 immune response, and induce en-
hanced cytotoxic T and natural killer cell cytotoxcities and antitumor immunity.
J. Gene Med. 5: 668–680.
28. Nishimura, T., K. Iwakabe, M. Sekimot, Y. Ohmi, T. Yahata, M. Nakui, T. Sato,
S. Habu, H. Tashiro, M. Sato, and A. Ohta. 1999. Distinct role of antigen-specific
T helper type 1 (Th1) and Th2 cells in tumor eradication in vivo. J. Exp. Med.
29. Bourgeois, C., B. Rocha, and C. Tanchot. 2002. A role for CD40 expression on
CD8?T cells in the generation of CD8?T cell memory. Science 297:
30. Azuma, M., H. Yssel, J. H. Phillips, H. Spits, and L. L. Lanier. 1993. Functional
expression of B7/BB1 on activated T lymphocytes. J. Exp. Med. 177: 845–850.
31. Sansom, D. M., and N. D. Hall. 1993. B7/BB1, the ligand for CD28, is expressed
on repeatedly activated human T cells in vitro. Eur. J. Immunol. 23: 295–298.
32. Tatari-Calderone, Z., R. T. Semnani, T. B. Nutman, J. Schlom, and H. Sabzevari.
2002. Acquisition of CD80 by human T cells at early stages of activation: func-
tional involvement of CD80 acquisition in T cell to T cell interaction. J. Immunol.
33. Tsang, J. Y., J. G. Chai, and R. Lechler. 2003. Antigen presentation by mouse
CD4?T cells involving acquired MHC class II:peptide complexes: another
mechanism to limit clonal expansion. Blood 101: 2704–2710.
34. Hudrisier, D., and P. Bongrand. 2002. Intercellular transfer of antigen-presenting
cell determinants onto T cells: molecular mechanisms and biological significance.
FASEB J. 16: 477–486.
35. Huang, H., F. Li, J. Gordon, and J. Xiang. 2002. Synergistic enhancement of
antitumor immunity with adoptively transferred tumor-specific CD4 and CD8 T
cells and intratumoral lymphotactin transgene expression. Cancer Res. 62:
36. Li, Y., N. Wang, H. Li, K. King, R. Bassi, H. Sun, A. Santiago, A. Hooper,
P. Bohlen, and D. Hicklin. 2002. Active immunization against the vascular en-
dothelial growth factor receptor flk1 inhibits tumor angiogenesis and metastasis.
J. Exp. Med. 195: 1575–1584.
37. Curtsinger, J., C. Schmidt, A. Mondino, D. Lins, R. Kedl, M. Jenkins, and
M. Mescher. 1999. Inflammatory cytokines provide a third signal for activation
of naı ¨ve CD4?and CD8?T cells. J. Immunol. 162: 3256–3262.
38. Rosenberg, A. S., T. Mizuochi, S. O. Sharrow, A. Singer. 1987. Phenotype,
specificity, and function of T cell subsets and T cell interactions involved in skin
allograft rejection. J. Exp. Med. 165: 1296–1315.
39. Rees, M. A., A.S. Rosenberg, T. I. Munitz, and A. Singer. 1990. In vivo induction
of antigen- specific transplantation tolerance to Qa1a by exposure to alloantigen
in the absence of T-cell help. Proc. Natl. Acad. Sci. USA 87: 2765–2769.
40. Wang, J. C., and A. M. Livingstone. 2003. Cutting edge: CD4?T cell help can
be essential for primary CD8?T cell responses in vivo. J. Immunol. 171:
41. Behrens, G. M., M. Li, G. M. Davey, J. Allison, R. A. Flavell, F. R. Carbone, and
W. R Heath. 2004. Helper requirements for generation of effector CTL to islet ?
cell antigens. J. Immunol. 172: 5420–5426.
42. Janssen, E. M., E. E. Lemmens, T. Wolfe, U. Christen, M. G. von Herrath, and
S. P. Schoenberger. 2003. CD4?T cells are required for secondary expansion and
memory in CD8?T lymphocytes. Nature 421: 852–856.
43. Lyakh, L. A., G. K. Koski, H. A. Young, S. E. Spence, P. A. Cohen, and
N. R. Rice. 2002. Adenovirus type 5 vectors induce dendritic cell differentiation
in human CD14?monocytes cultured under serum-free conditions. Blood 99:
44. Mailliard, R. B., Y. I. Son, R. Redlinger, P. T. Coates, A. Giermasz, P. A. Morel.
W. J. Storkus, and P. Kalinski. 2003. Dendritic cells mediate NK cell help for Th1
and CTL responses: two-signal requirement for the induction of NK cell helper
function. J. Immunol. 171: 2366–2373.
45. Wang, B., C. C. Norbury, R. Greenwood, J. R. Bennink, J. W. Yewdell, and
J. A. Frelinger. 2001. Multiple paths for activation of naive CD8? T cells:
CD4?-independent help. J. Immunol. 167: 1283–1289.
46. Mintern, J. D., G. M. Davey, G. T. Belz, F. R. Carbone, and W. R. Heath. 2002.
Cutting edge: precursor frequency affects the helper dependence of cytotoxic T
cells. J. Immunol. 168: 977–980.
47. Hou, S., X. Y. Mo, L. Hyland, and P. C. Doherty. 1995. Host response to Sendai
virus in mice lacking class II major histocompatibility complex glycoproteins.
J. Virol. 69: 1429–1434.
48. Mauri, D., T. Wyss-Coray, H. Gallati, and W. J. Pichler. 1995. Antigen-present-
ing T cells induce the development of cytotoxic CD4?T cells. I. Involvement of
the CD80-CD28 adhesion molecules. J. Immunol. 155: 118–127.
49. Taams, L. S., W. van Eden, and M. H. Wauben. 1999. Antigen presentation by
T cells versus professional antigen-presenting cells (APC): differential conse-
quences for T cell activation and subsequent T cell-APC interactions. Eur. J. Im-
munol. 29: 1543–1550.
50. Nisini, R., A. Fattorossi, C. Ferlini, and R. D’Amelio. 1996. One cause for the
apparent inability of human T cell clones to function as professional superanti-
gen-presenting cells is autoactivation. Eur. J. Immunol. 26: 797–803.
51. Satyaraj, E., S. Rath, and V. Bal. 1994. Induction of tolerance in freshly isolated
alloreactive CD4?T cells by activated T cell stimulators. Eur. J. Immunol. 24:
52. Mannie, M. D., S. K. Rendall, P. Y. Arnold, J. P. Nardella, and G. A. White.
1996. Anergy- associated T cell antigen presentation: a mechanism of infectious
tolerance in experimental autoimmune encephalomyelitis. J. Immunol. 157:
53. Patel, D. M., P. Y. Arnold, G. A. White, J. P. Nardella, and M. D. Mannie. 1999.
Class II MHC/peptide complexes are released from APC and are acquired by T
cell responders during specific antigen recognition. J. Immunol. 163: 5201–10.
54. Ossendorp, F., E. Mengede, M. Camps, R. Filius, and C. J. Melief. 1998. Specific
T helper cell requirement for optimal induction of cytotoxic T lymphocytes
against major histocompatibility complex class II negative tumors. J. Exp. Med.
7505The Journal of Immunology