| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
,
,
,,,,,
* Department of Laboratory Medicine and Pathology,
Center for Immunology,
Cancer Center, and
Department of Pediatrics, University of Minnesota, Minneapolis, MN 55455
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Professional APCs such as dendritic cells (DCs) and macrophages express
CD91 and the CD14/Toll-like receptor 4 complex, both of which bind
mammalian hsp70 (13, 14). Another receptor, CD40, was
recently shown to bind hsp70 derived from Mycobacterium
tuberculosis (TBhsp70), but not mammalian hsp70 (15).
Receptor-mediated binding of stress proteins initiates signal cascades
in immature DCs that cause them to differentiate and migrate from the
periphery to lymph nodes (LN), express costimulatory molecules,
increase cell surface expression of MHC class I and II molecules,
produce NO, and secrete proinflammatory cytokines (4, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22). Binding of TBhsp70 by CD40 also causes human DCs to
release CC chemokines such as RANTES (15). Little is known
about how these stress protein receptors signal APC to differentiate,
except that two or more pathways exist, and MyD88/IRAK/NF-
B is
required for at least one of them (13, 19, 20, 21, 22). Regardless
of how these signals are transduced, stress proteins are potent
adjuvants that can elicit protective CTL-mediated immunity in viral and
tumor animal models (4, 17, 23, 24). Success in these
models has led to clinical trials to assess the safety and efficacy of
Ag/stress protein complexes as cancer vaccines (25).
To maximize their potential as cancer vaccines, it is crucial to understand how stress proteins affect CTLs, since the immunological destruction of tumor cells is mediated primarily by this cell type. The effects stress proteins have on CTLs have been measured directly in vitro and ex vivo in terms of CTL phenotype and cytolytic activity (26, 27, 28, 29) and indirectly in vivo by the enhanced survival of immunized animals upon tumor challenge (17, 23, 24). However, direct assessments of the quantitative and qualitative effects that stress proteins have on Ag-specific CD8+ T cells in vivo are lacking. To begin to address this, Young and colleagues (29, 30) constructed a fusion protein containing 115 aa of chicken OVA fused to TBhsp70 (OVA.TBhsp70) and found that immunization with this protein elicited a CD4+ T cell-independent, OVA-specific CTL response in mice. To quantitate this CTL response we have now used an adoptive cell transfer system to track OVA-specific donor CD8+ T cells in immunized recipients. Immunization with the OVA.TBhsp70 fusion protein resulted in the rapid and persistent activation of essentially all the transferred CD8+ T cells. This quick, relatively durable, and complete response can account for the potent adjuvant effects of TBhsp70.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Six- to 8-wk-old C57BL/6 mice were purchased from The Jackson
Laboratory (Bar Harbor, ME), Charles River (Wilmington, MA), or Harlan
Sprague Dawley (Indianapolis, IN). B6.129S2-Cd4tm1
Mak (CD4-deficient) mice were purchased from The Jackson
Laboratory. C57BL/6-Tg (Tcr
Tcr
)1100 Mjb (OT-I) mice
(31) were bred to wild-type congenic
B6.PL-Thy1a/Cy (Thy1.1) and B6.SJL-Ptprca
Pep3b/BoyJ (CD45.1) mice to generate the OT-I/PL and OT-I.SJL strains,
respectively. All mice were housed under specific pathogen-free
conditions and offered food and water ad libitum. These studies were
approved by the institute animal care and use committee at the
University of Minnesota.
EL4 is a thymoma of C57BL/6 (H-2b) origin (32). EG.7 is an EL4 clone transfected with the gene encoding chicken OVA (33). EL4 was cultured in Alloclone medium (RPMI 1640 supplemented with 10% FCS, 50 nM 2-ME, 4 mM L-glutamine, 100 U/ml of penicillin, 100 µg/ml of streptomycin, 10 mM HEPES, 100 µM nonessential amino acids, and 1 mM sodium pyruvate). EG.7 was cultured in Alloclone plus 400 µg/ml of Geneticin (Life Technologies, Gaithersburg, MD). Both cell lines were maintained at 37°C in a humidified atmosphere containing 5% CO2.
Proteins, Abs, and reagents
The OVA (pKS28), TBhsp70 (pKS74), and OVA.TBhsp70 (pKS76)
expression constructs were gifts from Drs. J. Richmond and R. Young
(Massachusetts Institute of Technology, Cambridge, MA) (28, 30). To express these recombinant proteins, Escherichia
coli BL21(DE3)pLysS transformants were grown at room temperature
in Luria-Bertoni medium containing 100 µg/ml of carbenicillin and 35
µg/ml of chloramphenicol to an OD600 of 
0.4.
Protein expression was then induced with 0.5 mM
isopropylthiogalactoside for 8–9 h at room temperature. Rifampicin
(150 µg/ml) was added to the pKS74 culture 30 min after
isopropylthiogalactoside. TBhsp70 was expressed as a soluble
protein. OVA.TBhsp70 and OVA were both expressed as inclusion bodies
and refolded sequentially in 5, 2, 1, and 0.5 M guanidine-HCl before
dialysis against the appropriate buffer for subsequent purification by
column chromatography. The OVA.TBhsp70 and TBhsp70 proteins were
purified by ATP affinity chromatography as previously described
(28). The recombinant OVA fragment contained an N-terminal
six-histidine tag and was purified on an
NiSO4-charged chelating column according to the
manufacturer’s instructions (Amersham Pharmacia Biotech, Arlington
Heights, IL). Column fractions containing protein were pooled and
dialyzed extensively against PBS. Protein purity was assessed by
SDS-PAGE. Protein concentrations were determined with the Coomassie
Protein Reagent Assay (Pierce, Rockford, IL). The LPS content of the
purified proteins was determined using the E-toxate assay (Sigma, St.
Louis, MO). The SIINFEKL peptide was synthesized by the Microchemical
Facility at University of Minnesota and stored at -70°C as a
200-µM stock solution in PBS.
With the exceptions of anti-CD45.1-biotin and anti-CD69-PE (eBiosciences, San Diego, CA), flow cytometric reagents were purchased from BD PharMingen (San Diego, CA). CFA, LPS, and polymyxin B were purchased from Sigma, and CFSE was obtained from Molecular Probes (Eugene, OR). The murine CD8 enrichment mixture and StemSep reagents were purchased from StemCell Technologies (Vancouver, Canada).
Adoptive transfer and immunization
The CD8+ T cell response to a defined Ag
was tracked using the adoptive transfer technique pioneered by Kearney
et al. (34) and modified by Mescher and colleagues
(35, 36, 37). In this system nonadherent LN cells from TCR
transgenic mice are transferred into congenic recipients such that the
final frequency of the donor cells in the recipients is 0.5% of the
lymphocyte total. Donor lymphocytes were prepared by homogenizing LN
from OT-I/PL or OT-I.SJL mice. Adoptively transferred OT-I/PL and
OT-I.SJL cells responded indistinguishably to challenge with Ag
(our unpublished observations), and so the choice of the Thy1.1
or CD45.1 allelic marker was usually based on donor availability.
Single-cell suspensions were adherence-depleted at 37°C for 60–90
min on Integrid plates (BD Biosciences, Franklin Lakes, NJ). The
nonadherent LN cells were collected, and 3 x
106 in 500 µl pyrogen-free PBS (BioWhittaker,
Walkersville, MD) were injected in the tail vein of each recipient
mouse. In some experiments CD8+ T cells were
enriched by depleting other LN cell types by magnetic removal following
incubation with a cocktail of lineage-specific Abs linked to magnetic
beads according to the manufacturer’s instructions (Stem Cell
Technologies). In other experiments the donor lymphocytes were labeled
before injection with CFSE by incubating 1 x
107 cells/ml in HBSS with 1 µl of 1.7 mM
CFSE/ml of cells for 10 min at 37°C. Free CFSE was quenched by an
additional brief incubation in 10% FCS in Alloclone medium. The
labeled cells were then washed in PBS, and 3 x
106 cells in 500 µl were injected into the
recipients’ tail veins. All mice in a given experiment were adoptively
transferred with OT-I cells on the same day. In all experiments the
donor OT-I cells had a naive phenotype before transfer (CD25, CD44, and
CD69 negative; our unpublished observations).
Recipients were immunized s.c. with a total of 300 µl of PBS containing OVA.TBhsp70, OVA in LPS, TBhsp70, or SIINFEKL in CFA. In some experiments all mice were immunized at the same time and were sacrificed on different days postimmunization. In other experiments all mice were sacrificed at the same time, but were immunized on different days before sacrifice.
Flow cytometric analyses
LN and spleen cell suspensions were homogenized in PBS
containing 2% FCS and 0.02% sodium azide and were incubated with
anti-Fc
IIR to block Fc receptors. OT-I/PL and OT-I.SJL cells
were then identified by flow cytometry with
anti-CD8-allophycocyanin and anti-Thy1.1-PE, or with
anti-CD8-allophycocyanin and anti-CD45.1-biotin plus
streptavidin-PerCP, respectively. For analysis of CD44 or CD69
expression levels, cells were also incubated with CD44-FITC (for
OT-I/PL cells) or with anti-CD69-PE or anti-CD44-PE (for
OT-I.SJL cells). Cells were incubated with Abs at 4°C in the dark for
30 min. Cells were then washed in PBS with 2% FCS and 0.02% sodium
azide and were either analyzed directly or fixed in 1%
paraformaldehyde in PBS. Events were acquired on a FACSCalibur (BD
Biosciences, Mountain View, CA) and were analyzed using Flow Jo
software version 3.4 (Tree Star, San Carlos, CA).
CTL assays
Cytotoxicity was assessed in a standard 4-h 51Cr release assay. LN and spleen cells from immunized recipients were used as effectors against 1 x 104 51Cr-labeled EG.7 or EL4 target cells. After 4 h cells were pelleted by centrifugation, and the amount of 51Cr released into the supernatant was measured. Spontaneous release was determined by incubating target cells in the absence of effectors. Total release was determined by incubating target cells in medium containing Triton X-100. The percent lysis was calculated as: (experimental mean - spontaneous release)/(total release - spontaneous release). Assays were performed in triplicate.
IFN- intracellular staining
Draining LN or spleen cells from immunized OT-I/PL recipients
were homogenized and incubated overnight in Alloclone medium containing
200 nM SIINFEKL. The next day 1 x 106 cells
were incubated with anti-FcIIR, followed by
anti-CD8-allophycocyanin and anti-Thy1.1-PE to identify OT-I
cells. To determine IFN- production, cells were additionally labeled
with anti-mouse IFN--FITC or FITC-labeled isotype-matched
control after treatment with the Cytofix/Cytoperm Plus
(withGolgiStop) kit according to the manufacturer’s instructions
(BD PharMingen). Events were acquired on a FACSCalibur (BD
Biosciences) and were analyzed using Flow Jo software version 3.4 (Tree
Star).
Statistical analyses
The one-tailed Student’s t test was used to determine the significance of the results.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
OVA.TBhsp70, OVA161–276, and TBhsp70 were
expressed as recombinant proteins in E. coli and were
purified to 
95% homogeneity. Because LPS has adjuvant properties
similar to those attributed to stress proteins, LPS levels in the
purified recombinant protein preparations were quantified and
functionally assessed. Each protein was endotoxin-free within the
limits of the Limulus amebocyte lysate assay (<0.125
endotoxin units/ml). As a functional control, each protein
preparation was compared with LPS for its ability to increase the
expression of the costimulatory molecules CD80 and CD86 on DCs
(38). Murine DCs were incubated in vitro with 100 ng of
LPS with or without the LPS inhibitor polymyxin B or with 100 µg of
each protein with or without polymyxin B. CD80 and CD86 expression was
only enhanced following incubation with LPS alone (our
unpublished observations). Together these results demonstrated that the
three immunogens were essentially devoid of LPS.
OVA.TBhsp70 is more immunogenic than SIINFEKL/CFA
To assess the quantitative and qualitative aspects of the CTL
response to OVA.TBhsp70, adoptively transferred OT-I cells in immunized
recipients were tracked by flow cytometry. OT-I cells respond to
OVA.TBhsp70 because it contains SIINFEKL (OVA residues 257–264), the
peptide for which the OT-I TCR is specific (31). The
adjuvant effects of TBhsp70 relative to CFA were assessed by comparing
the responses of adoptively transferred OT-I cells in mice immunized
with OVA.TBhsp70 vs SIINFEKL/CFA. The phenotypes and total numbers of
OT-I cells were determined 3 and 17 days following immunization with 10
µg (0.1 nmol) of fusion protein or 50 µg (50 nmol) of SIINFEKL in
CFA, amounts shown to stimulate optimal proliferation of OT-I cells
(our unpublished observations). Significant numbers of OT-I
cells were detected in the draining LN (p <
0.01) and spleens (p < 0.05) of mice 3 days
after immunization with either OVA.TBhsp70 or SIINFEKL/CFA, but not
with PBS or TBhsp70 alone (Fig. 1
). OT-I
cells in the former two groups also expressed high levels of CD44,
indicating that they had undergone Ag-specific activation (our
unpublished observations). Compared with PBS controls, immunization
with OVA.TBhsp70 yielded 74- and 10-fold increases in OT-I
cell numbers in LN and spleen, respectively, whileSIINFEKL/CFA
caused 126- and 18-fold increases. The responses to OVA.TBhsp70
and SIINFEKL/CFA were not significantly different. In contrast, only
mice immunized with OVA.TBhsp70 had significant
(p < 0.01) numbers of OT-I cells 17 days later
(21- and 11-fold increases in the draining LN and spleen, respectively;
Fig. 1). Because OVA.TBhsp70 induces a comparable proliferative
response to 500-fold more peptide plus CFA by day 3 and induces a
greater response by day 17, we conclude that TBhsp70 is a superior
adjuvant to CFA.
|
Responses to OVA.TBhsp70 were also compared with those elicited by OVA161–276 plus LPS, where comparable processing and presentation of the antigenic epitope are required and where neither adjuvant provides an Ag depot, as is the case for mineral oil in CFA. C57BL/6 mice were adoptively transferred with OT-I.SJL LN cells and then immunized with 10 µg (0.1 nmol) of OVA.TBhsp70 or 1.5 µg (0.1 nmol) of OVA161–276 mixed with 10 µg of LPS (OVA/LPS) on days 3, 6, 12, 17, and 21 before analysis.
OT-I cell numbers peaked in the draining LN 3 days postimmunization in
both groups, but the frequency in the OVA.TBhsp70 group was about 6
times greater (p < 0.01) than that in the
OVA/LPS controls (Fig. 2). A comparison
with Fig. 1
reveals that the OT-I response to SIINFEKL/CFA is 10 times
greater than the OVA/LPS response. This could result from the 500-fold
molar excess of injected SIINFEKL vs OVA, more efficient loading of MHC
class I molecules with peptide (i.e., processing is not required for
SIINFEKL, while it is for OVA), or additional adjuvants in CFA vs
LPS.
|
) or OVA/LPS (Fig. 2).
In the LN the number of OT-I cells in OVA.TBhsp70-immunized mice
took longer to return to basal levels (21 days) than did the number of
cells in the OVA/LPS group (6 days; Fig. 2). In the spleen the number
of OT-I cells had not returned to basal levels 21–35 days
postimmunization with OVA.TBhsp70 (Fig. 2 and our unpublished
observations). This was not simply due to an increase in total cell
numbers, because only the percentages of OT-I cells were high, while
the numbers of total splenocytes remained comparable (our
unpublished observations). Moreover, OT-I cell frequencies in the
spleen displayed a reproducible periodicity in which two comparable
peaks occurred 6 and 21–28 days postimmunization with OVA.TBhsp70
(Fig. 2 and our unpublished observations). We conclude that
TBhsp70 is a more potent adjuvant than LPS in terms of both the
magnitude and the duration of its proliferative effect on
CD8+ T cells. CD4+ T cells are not required for the OT-I response to OVA.TBhsp70
C57BL/6.CD4 -/- gene-targeted recipients were adoptively transferred with enriched (>97%) CD8+ OT-I cells to determine whether CD4+ T cells were required for the OT-I response to OVA.TBhsp70. OT-I cell proliferation and differentiation in OVA.TBhsp70-immunized CD4-deficient recipients were indistinguishable from those in wild-type recipients (our unpublished observations). Our results support the conclusion of Huang et al. (29) that CD4+ T cells are not required for the TBhsp70 adjuvant effect on CTL responses.
OVA.TBhsp70 rapidly activates the majority of OT-I cells
The persistence of splenic OT-I cells induced by OVA.TBhsp70 could
be residual from the original proliferative burst or result from the
continuous division of a subset of cells in this initial burst. To
distinguish between these possibilities, the kinetics of the OT-I
proliferative and activation responses to OVA.TBhsp70 were compared
with those to OVA/LPS. TBhsp70 alone was not examined as OT-I cells
were previously shown to be unresponsive to it (Fig. 1 and our
unpublished observations). OT-I.SJL cells were labeled with CFSE before
adoptive transfer and immunization with either 0.1 nmol OVA.TBhsp70 or
OVA in LPS. Draining LN and spleen cells were harvested on days 1–5
after immunization, and OT-I cells were analyzed for proliferation, as
measured by loss of CFSE intensity (39, 40), and for
activation, as measured by a transient increase in CD69 expression.
Fig. 3A shows that OVA.TBhsp70
activates more OT-I cells in a shorter time than OVA/LPS. The frequency
of CD69+ OT-I cells in the draining LN of mice in
the OVA/LPS group peaked at 32% 1 day after immunization, but in
OVA.TBhsp70-immunized mice it continued to rise until it peaked at 72%
on day 2. OVA.TBhsp70 immunization activated more than twice as many
OT-I cells by day 2 as OVA/LPS immunization (72 vs 30%, respectively).
Compared with the draining LN, there were fewer
CD69+ splenic OT-I cells in both groups.
|
B illustrates the frequencies of OT-I cells that had
proliferated at the indicated times. Consistent with the CD69 staining,
almost 3 times as many OT-I cells in the draining LN had divided at
least once by day 2 in mice immunized with OVA.TBhsp70 vs OVA/LPS (59
vs 20%, respectively). By day 3 only 6% of OT-I cells in the draining
LN of OVA.TBhsp70-immunized mice had not divided compared with 28%
in the OVA/LPS group.
The kinetics of OT-I proliferation in the spleen differed from those in
the draining LN. Cells that had divided did not appear in the spleen
until day 3, coincident with the peak of OT-I expansion in the draining
LN. The relatively low percentage of CD69+ OT-I
cells in the spleen suggested that they were activated elsewhere, as
did their expression of the activation marker CD44 (Fig. 4). Immunization with either OVA.TBhsp70
or OVA/LPS led to increased CD44 expression on OT-I cells. However,
OVA.TBhsp70 immunization led to a higher percentage of
CD44high OT-I cells than did OVA/LPS. This was
sustained throughout the course of the 21-day experiment, while CD44
levels dropped after day 6 in the OVA/LPS group.
|
). These data suggest that OT-I cells in the former group remain
cytolytic effectors, while in the latter group they convert to a memory
population (41, 42). OT-I cells present 17 days after OVA.TBhsp70 immunization are functional CTLs
To determine whether immunization with OVA.TBhsp70 resulted in the
development of CTL effectors, day 17 splenic and draining LN OT-I cells
were tested for IFN- production and cytolytic activity
(35, 36, 37). To measure IFN- levels, cell suspensions were
pulsed overnight with SIINFEKL and then examined for intracellular
IFN- production (Fig. 5). OT-I cells
from both the draining LN and spleens of OVA.TBhsp70-immunized mice
were capable of making IFN- on day 17 (81 and 96%, respectively).
This compares to 61 and 81% of IFN-+ OT-I
cells from the draining LN and spleens of OVA/LPS-immunized mice,
respectively.
|
) and not the EL4 targets
(our unpublished observations). The overwhelming majority of
these effectors were OT-I cells; the others were endogenously derived.
These data indicate that functional CTLs persist at least 17 days after
immunization with OVA.TBhsp70.
|
One explanation for the persistence of effector CTLs is that naive OT-I cells are still being activated by Ag 2 wk postimmunization. This is unlikely because OVA.TBhsp70 activates almost all (94%) detectable OT-I cells within 3 days, and the day 17 OT-I cells do not express CD69 and are not blasting (based on their low forward and side light scatters; our unpublished observations). To formally rule out the possibility that OVA.TBhsp70 formed an Ag depot that lasted for 2 wk, CFSE-labeled naive OT-I cells were transferred into mice that had been immunized 15 days previously but had not been adoptively transferred before immunization. Draining LN and spleens were harvested from these mice 1, 2, and 3 days after transferring the CFSE-labeled OT-I cells (days 16, 17, and 18 postimmunization), and OT-I cells were analyzed for their CFSE intensity. Newly transferred cells had not detectably proliferated at any time point (our unpublished observations), suggesting that Ag was not available to activate naive CD8+ T cells 15 days postimmunization. Therefore, the pool of activated OT-I cells present 2–3 wk after immunization with OVA.TBhsp70 apparently does not arise from the continuous recruitment of naive OT-I cells, but, rather, reflects the persistence of cells activated within the first few days following immunization.
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
upon Ag rechallenge in
vitro. Taken together, our data demonstrate that TBhsp70 is both
qualitatively and quantitatively superior to CFA and LPS in eliciting
Ag-specific CTL cell responses in vivo. We propose that these effects
account for the success of TBhsp70 as a vaccine in animal models
(4, 17, 23, 24) and speculate that other stress proteins
act similarly. Analogous to mammalian gp96 and hsp70, the adjuvanticity of TBhsp70 is most likely due to its ability to activate DCs (8, 10, 11, 12, 13, 21). Because activated DCs can deliver all the signals (Ag, costimulation, and IL-12) required to fully stimulate naive CD8+ T cells (39), it is not surprising that the development of CTL responses is critically dependent on this professional APC (43). Mammalian hsp70 induces DC maturation through receptors such as the CD14/Toll-like receptor 4 complex and CD91 (13, 14). To our knowledge, it is not yet known whether TBhsp70 binds these two receptors. However, TBhsp70 specifically binds CD40 and through it induces human monocyte-derived DCs to produce CC chemokines (15). TBhsp70 binding to CD40 also causes the human myelomonocytic cell line KG1 to differentiate into cells with DC features (15). Finally, TBhsp70 can activate immature bovine DCs through an undefined receptor (44). Together these data strongly suggest that mammalian and TB hsp70 are potent adjuvants because they both activate DCs. Although they clearly can signal through different receptors, it remains to be seen whether they share signaling pathways.
DCs present peptides to both MHC class II-restricted CD4+ Th cells and MHC class I-restricted naive CD8+ CTLs. CD4+ T cells participate in CTL responses by supplying IL-2 and activating DCs (45). However, CD4+ T cells are not required to generate all CTL responses. Udono et al. (6) originally demonstrated that gp96-induced CTL responses were CD4+ T cell independent. This was later extended by Huang et al. (29) when they similarly showed that CD4+ T cells were not required for the CD8+ T cell response to TBhsp70. In the presence of peptide-pulsed DCs, Wang et al. (46) found that CD8+ T cells can provide their own help in the form of endogenous IL-2 if they are present at sufficiently high precursor frequencies. Following this reasoning, we speculate that the proliferative burst induced by TBhsp70 yields enough OVA-specific CD8+ T cells to produce "self-help." This would account for the CD4+ T cell independence of the OVA.TBhsp70-mediated OT-I response.
If TBhsp70 indirectly induces CD8+ T cell effectors to produce endogenous IL-2, then this cytokine could continue to drive proliferation, and this could explain the persistence of CTL effectors 3 wk postimmunization when Ag can no longer activate naive CD8+ T cells. In a model of CTL-mediated tumor immunity, Shrikant and Mescher (36) found that lytic effectors failed to remain at the tumor site and control tumor cell growth in the absence of exogenous IL-2. They suggested that effective CD8-targeted immunotherapy may be critically dependent on sustaining the response and reactivating cells that have left the site of tumor growth and become nonresponsive. Schoenberger, Ahmed, and Pamer (47, 48, 49, 50) have shown that once the CD8+ T cell developmental pathway is set into motion by antigenic stimulation under optimal conditions, it proceeds without any further need for Ag. That is, Ag-activated CD8+ T cells go through at least eight cell divisions, become cytolytic, and eventually revert to long-lived memory cells without further Ag stimulation. Bevan and Fink (51) suggest that this optimal early "hit" with APCs requires high Ag density and strong costimulation. We speculate that OVA.TBhsp70 stimulates DCs to provide both high Ag density and strong costimulation to naive OT-I cells, and that this optimal encounter causes the OT-I cells to differentiate into relatively long-lived (i.e., 3 wk) effectors that persist in the absence of Ag. Our data showing that naive OT-I cells are not activated if transferred 15 days postimmunization support this.
These data also eliminate Ag restimulation as a cause of the increase in splenic OT-I cells seen 21–28 days after immunization with OVA.TBhsp70. Therefore, this rebound probably results from the migration of CD44high OT-I cells from non-lymphoid tissues (52). We speculate that a proportion of OVA.TBhsp70-activated OT-I cells become effector memory T cells that migrate from lymphoid to non-lymphoid tissues 1–2 wk postimmunization, but return in 3–4 wk (52, 53). Because the frequencies of splenic OT-I cells on days 6 and 21–28 postimmunization are equivalently high, the return of these putative effector memory OT-I cells is apparently synchronous. The reasons for this are not yet clear.
The resurgence in cancer vaccine development over the last decade has resulted from identifying tumor-associated peptides and enhancing their presentation to T cells (50, 54). Kaech et al. (50) noted that effective vaccines must induce as large an effector T cell population as possible, since this burst size determines the number of subsequent memory T cells that form. By tracking a specific CD8+ T cell response to an Ag/stress protein complex in vivo, we have quantified how profoundly a stress protein stimulates CD8+ T cell proliferation and differentiation into a relatively long-lived effector population. We have also shown qualitatively how this adjuvant effect differs from the effects of LPS and CFA. Our results may have clinical relevance, as stress protein/Ag complexes are being evaluated as cancer therapeutics. Janetzki et al. (25) isolated gp96/peptide complexes from a variety of solid tumors (e.g., breast, colon, and pancreatic carcinomas), and in six of 12 patients tested, immunization led to CD8+ T cell responses specific for the autologous tumor. NK activity was also elicited in eight patients. No adverse toxicities or autoimmune responses occurred. Another recent report showed that hsp70/peptide complexes isolated from melanomas specifically activated melanoma-specific human CD8+ T cells in vitro by cross-priming APCs (55). Together these studies suggest that stress protein/peptide complexes hold promise for cancer vaccines, and our results help to reveal why.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
2 Address correspondence and reprint requests to Dr. Christopher A. Pennell, University of Minnesota Cancer Center, MMC 86, 420 Delaware Street SE, Minneapolis, MN 55455. E-mail address: penne001{at}umn.edu
3 Abbreviations used in this paper: hsp70, heat shock protein 70; DC, dendritic cell; gp96, glucose-regulated protein 96; LN, lymph node; TBhsp70, Mycobacterium tuberculosis heat shock protein 70.
Received for publication July 9, 2002. Accepted for publication September 16, 2002.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
B pathway. Int. Immunol. 12:1539. production following antigenic challenge in vitro. Int. Immunol. 11:699.This article has been cited by other articles:
|
|
 |
|
  B. Valentinis, A. Capobianco, F. Esposito, A. Bianchi, P. Rovere-Querini, A. A. Manfredi, and C. Traversari Human recombinant heat shock protein 70 affects the maturation pathways of dendritic cells in vitro and has an in vivo adjuvant activity J. Leukoc. Biol., July 1, 2008; 84(1): 199 - 206. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 O. Boyman, C. Ramsey, D. M. Kim, J. Sprent, and C. D. Surh IL-7/Anti-IL-7 mAb Complexes Restore T Cell Development and Induce Homeostatic T Cell Expansion without Lymphopenia J. Immunol., June 1, 2008; 180(11): 7265 - 7275. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 H. Liu, B. H. Wu, G. J. Rowse, and P. C. R. Emtage Induction of CD4-Independent E7-Specific CD8+ Memory Response by Heat Shock Fusion Protein Clin. Vaccine Immunol., August 1, 2007; 14(8): 1013 - 1023. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 Z. Ye and Y.-H. Gan Flagellin Contamination of Recombinant Heat Shock Protein 70 Is Responsible for Its Activity on T Cells J. Biol. Chem., February 16, 2007; 282(7): 4479 - 4484. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
Y. Bulut, K. S. Michelsen, L. Hayrapetian, Y. Naiki, R. Spallek, M. Singh, and M. Arditi Mycobacterium Tuberculosis Heat Shock Proteins Use Diverse Toll-like Receptor Pathways to Activate Pro-inflammatory Signals J. Biol. Chem., June 3, 2005; 280(22): 20961 - 20967. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. A. R. Tobian, C. V. Harding, and D. H. Canaday Mycobacterium tuberculosis Heat Shock Fusion Protein Enhances Class I MHC Cross-Processing and -Presentation by B Lymphocytes J. Immunol., May 1, 2005; 174(9): 5209 - 5214. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. A. R. Tobian, D. H. Canaday, W. H. Boom, and C. V. Harding Bacterial Heat Shock Proteins Promote CD91-Dependent Class I MHC Cross-Presentation of Chaperoned Peptide to CD8+ T Cells by Cytosolic Mechanisms in Dendritic Cells versus Vacuolar Mechanisms in Macrophages J. Immunol., May 1, 2004; 172(9): 5277 - 5286. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 U. K. Rapp and S. H. Kaufmann DNA vaccination with gp96-peptide fusion proteins induces protection against an intracellular bacterial pathogen Int. Immunol., April 1, 2004; 16(4): 597 - 605. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
) or OVA/LPS (
).
The gate defines the CD44high population. B,
Percentages of CD44high OT-I cells from draining LN
(top) or spleens (bottom). The data shown
are the mean of three mice per group.
