| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Center for Immunology, Department of Laboratory Medicine and Pathology, University of Minnesota, Minneapolis, MN 55455
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
A novel approach for enhancing CTL responses for tumor immunotherapy was suggested by the observation that CD8 T cells are most effectively stimulated by subcellular Ag if it is presented on the surfaces of particles about the same size as cells (11). Plasma membranes or purified class I alloantigen were effective in stimulating in vitro generation of CTL responses or activating cloned CTL effector function when immobilized on 5-µm beads but ineffective if presented on smaller beads or in free form. In contrast, in vivo administration of Ag on cell-sized beads was not sufficient to stimulate detectable CTL responses. However, this form of Ag, termed large multivalent immunogen (LMI),5 was able to very substantially augment CTL responses when administered to mice that had been challenged with cells bearing the same Ag. This was initially demonstrated for responses to allogeneic tumors, using purified class I MHC protein to prepare the LMI (12). Augmentation required that the LMI have the same Ag on the surface as was present on the cells used for the challenge, and the resulting effector CTL retained specificity for the Ag.
The LMI approach was extended to syngeneic tumors using plasma membranes isolated from the tumor cell as the Ag. Administration of LMI at the same time that mice were inoculated with syngeneic tumor resulted in generation of detectable ex vivo tumor-specific cytolytic activity and a concomitant reduction in tumor growth (12, 13). When mice with established tumors were treated, it was found that administering LMI alone did not have a significant effect. However, low dose cyclophosphamide (Cy) and LMI were highly synergistic in reducing growth of established tumors and extending survival in several tumor models, including P815 mastocytoma growing as a solid s.c. tumor and fibrosarcomas in a lung metastasis model (14). A substantial fraction of P815-bearing mice treated with Cy and LMI survived long term (>150 days) with no detectable tumor and were immune to rechallenge with the tumor. Based on these studies, a small clinical trial was performed using LMI alone (without Cy) to treat stage III and IV melanoma patients, using plasma membranes derived from two in vitro melanoma lines as the Ag to prepare the LMI. The LMI treatment had no significant adverse effects, and an increased frequency of tumor-specific precursor CTL was found in PBL of the majority of treated patients postvaccination (M. S. Mitchell, J. Kan-Mitchell, P. Morrow, D. Darrah, V. Jones, and M. Mescher, manuscript in preparation). A Phase I/II trial has now been initiated to examine safety and efficacy of LMI for melanoma and renal carcinoma immunotherapy using autologus tumor as the source of membrane Ag for preparing the LMI, and comparing treatment with LMI alone and together with Cy treatment (15).
The therapeutic effects of LMI in the murine models that were studied were tumor Ag specific and depended on presentation of the Ag on beads; free Ag or Ag with CFA were not effective (12, 14). Based on the effects of LMI on allogeneic responses, where the Ag used was purified class I MHC alloantigen, it appeared likely that the LMI effects seen in the syngeneic tumor therapy models involved tumor-specific Ag being presented by class I MHC proteins on the plasma membranes isolated from the tumors, but this was not directly demonstrated. Furthermore, the mechanism by which Ag presentation on cell-sized beads was able to augment in vivo responses was not defined. The experiments described in this report demonstrate that class I/peptide Ag complexes immobilized on cell-sized beads to prepare LMI can mediate augmentation of tumor-specific CD8 T cell responses, results that have implications for the development of better defined LMI for human immunotherapy. Furthermore, the experiments have used adoptive transfer of TCR-transgenic T cells (16, 17), thus allowing identification, enumeration, and phenotypic characterization of the Ag-specific cells during the course of the responses to tumor and LMI. The results have begun to provide some insight into the mechanism by which LMI augment generation of effective tumor-specific CTL responses. An understanding of these mechanisms will be necessary for realizing the full potential of this novel approach for immunotherapy.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
OT-I (H-2b) mice (18), a gift from Dr. F. Carbone (University of Melbourne, Melbourne, Australia), express a transgenic TCR specific for an OVA-derived peptide (SIINFEKL) presented on H-2Kb. OT-I mice were bred to C57BL/6.PL (Thy-1.1 congenic) and the resulting OT-I.PL mice were used as the source of CD8+ OT-I T cells for in vitro experiments and for adoptive transfer into C57BL/6 (Thy-1.2) mice.
EL4 murine thymoma derived from the C57BL/6 mouse (H-2b) was maintained in vitro in complete RPMI medium (RPMI 1640, 10% FCS, 0.2% L-glutamine, 0.1% penicillin-streptomycin, 0.1% HEPES, 0.1% nonessential amino acids, 0.01% sodium pyruvate, 0.05% 2-ME. The EL-4 derived E.G7 tumor expressing secreted whole OVA (19) was maintained in vitro in complete RMPI medium containing 400 µg/ml G418. Both EL-4 and E.G7 were washed extensively in PBS before in vivo challenge using 4 x 106 cells/mouse injected i.p. on day 0.
Preparation of Kb/OVA complexes and LMI
H-2Kb/OVA257–264
(SIINFEKL) peptide complexes were generated using a procedure developed
by Garboczi et al. (20) but modified by D. Busch and E.
Pamer (Yale University, New Haven, CT). Briefly, cDNAs encoding a
H-2Kb heavy chain-biotinylation site-fusion
protein and human 
2-microglobulin were
expressed in Escherichia coli and purified from inclusion
bodies. Denatured Kb and
2-microglobulin proteins were then refolded in
the presence of high concentrations of
OVA257–264 peptide to form the antigenic
complex. Complexes were subsequently biotinylated by BirA ligase
(Avidity, Denver, CO).
To prepare LMI, 5-µm sulfated polystyrene latex beads (Interfacial
Dynamics, Portland, OR) were coated with streptavidin-APC (BD
PharMingen, San Diego, CA) by incubating with 1.25
µg/107 beads in PBS for 20 min at 4°C. Beads
were then washed, resuspended in PBS at 1–2 x
107 beads/ml, and incubated with biotinylated
Kb/OVA for 1–2 h at 4°C. Beads coated with
differing amounts of complex were prepared for the experiments shown in
Fig. 1
. All in vivo experiments were
performed using
Kb/OVA257–264-LMI prepared
by incubating 25 ng complex per 106
streptavidin-coated beads. After incubation with the complex, an
aliquot of beads was removed and stained with Y3-FITC (mouse
anti-Kb) mAb to determine the level of Ag
immobilization. The remaining beads were blocked with 1% normal mouse
serum 30 min at 4°C, centrifuged, and resuspended at 2 x
107 LMI/ml in PBS. LMI were sterilized by
gamma-irradiation and injected i.p. or i.v. (tail vein) at
107 beads/mouse.
|
Lymph node (LN) cells from OT-I.PL mice were isolated and adherence depleted for 1 h at 37°C. Remaining cells were enriched for CD8+ cells using Cellect columns (Cedarlane, Vancouver, Canada). CD8-enriched cells, containing >90% CD8+ cells, were plated in a flat-bottom 96-well plate (Falcon) at 1 x 105 cells/well. LMI beads were added at 2 x 105/well in a final volume of 0.2 ml. Where indicated, recombinant human IL-2 (TECIN-NIH) was added at 2.5 U/well. Cultures were incubated at 37°C for 48 h, 1 µCi [3H]thymidine per well was added for the final 6 h, and incorporation was determined.
Adoptive transfer and analysis by flow cytometry
OT-1/PL lymph node cells were adherence depleted, and the nonadherent cells were washed three times in PBS. CD8 T cells (1–3 x 106/mouse) were adoptively transferred by i.v. injection (tail vein) into Thy-1-congenic C57BL/6 sex-matched recipients, and these mice were challenged 2 or 3 days later with tumor and/or LMI. The adoptively transferred cells equilibrate in the spleen and lymph nodes of the recipients within 1 day. The number of adoptively transferred cells present in the spleen and LN of the recipients varies somewhat between experiments but is very consistent between animals in the same experiment In some experiments, OT-I cells were labeled with fluorescent dye before transfer by incubating 107 cells/ml in HBSS at 37°C for 5 min with 3 µM CFSE (Molecular Probes, Eugene, OR). Cells were then washed once in cold complete RPMI medium and three times in PBS before adoptive transfer.
Animals were sacrificed on the indicated days after challenge, and lymph node (LN) cells (pooled inguinal, axillary, brachial, iliac, popliteal, mesenteric), spleen cells, and peritoneal exudate cells were harvested and analyzed by flow cytometry. Peritoneal exudate cells were obtained from two consecutive washes of the peritoneal cavity (PC) using 20 ml HBSS. The number of total live lymphocytes collected from each compartment was determined by flow cytometry using PKH reference beads (Sigma-Aldrich, St. Louis, MO). Cells (2–3 x 106) from transferred recipients were incubated in PBS containing 2% FCS and 0.02% NaN3. FcR block, anti-CD8-APC mAb, and anti-Thy-1.1-PE mAb were added to each sample. In addition, anti-CD44-FITC or anti-VLA-4-FITC mAbs were added to some samples. FcR block and mAbs were all from BD PharMingen. Samples were incubated for 1 h at 4°C, washed twice, and fixed in 0.4% paraformaldehyde in PBS for analysis by flow cytometry on a FACSCalibur using CellQuest software (BD Biosciences, San Jose, CA).
OT-I cells were identified as
CD8+Thy-1.1+ cells, and
this was confirmed by staining for the transgenic TCR 
-chain
V2. No background was present in the OT-I gate
if the anti-Thy-1.1 mAb was not included or if cells from normal
C57BL/6 mice were stained with the anti-CD8 and anti-Thy-1.1
mAbs. The total number of OT-I cells in a given compartment was
determined by multiplying the percent of OT-I cells by the total number
of cells recovered. For CFSE experiments, the cells were first gated on
the CD8+Thy-1.1+ population
of live lymphocytes and then analyzed for CFSE fluorescence levels.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The effects of LMI on tumor-specific CD8 T cell responses have
been examined using an adoptive transfer model that allows direct
visualization and characterization of the Ag-specific CD8 T cells
during the course of the response. The model uses CD8 T cells from OT-I
mice that have a transgenic TCR specific for chicken OVA-derived
SIINFEKL peptide (OVA257–264) presented by
H-2Kb, and E.G7 tumor, an OVA transfectant of the
EL-4 thymoma (19). E.G7 cells express 
100
Kb/SIINFEKL complexes on the cell surface
(21) and are specifically lysed by OT-I effector CTL. OVA
therefore serves as a pseudo-tumor Ag in this model. This model has
been used in several studies of tumor-specific CTL responses
(22, 23, 24, 25). The OT-I TCR was backcrossed onto the C57BL/6.PL
background which expresses the Thy1.1 allele (OT-I.PL). Thus,
SIINFEKL-specific CD8+ OT-I cells from OT-I.PL
mice can be identified by flow cytometry when adoptively transferred
into congenic Thy-1.2+ C57BL/6 recipients using
Abs specific for the Thy-1.1 marker and CD8. Greater than 97% of the
cells identified as
CD8+Thy-1.1+ also displayed
the TCR V2 receptor encoded by the OT-I
transgene (data not shown).
LMI were prepared by adsorbing streptavidin-APC to 5-µm latex
microspheres and then incubating with
biotinylated-H-2Kb/2-microglobulin/OVA257–264
peptide complexes to coat the beads with Ag. The amount of Ag on the
beads can be controlled by varying the amount of complex incubated with
a fixed number of beads, and the extent of Ag immobilization can be
determined by staining with anti-class I mAb. LMI prepared using a
given amount of
Kb/OVA257–264 Ag exhibited
a relatively narrow range of Ag density when stained with the Y3
anti-H-2Kb mAb and analyzed by flow
cytometry, and varying the amount of Ag used to coat the beads resulted
in LMI with varying surface densities of Ag (Fig. 1
A). To
confirm that Ag was present on the LMI in a form that could be
recognized by T cells and to determine the optimal Ag density for
stimulation of OT-I cells, CD8 T cells from OT-I.PL mice were incubated
with Kb/OVA257–264-LMI or
with LMI made using Kb complexed with an
irrelevant SIYRYYGL peptide (Kb/SIYRYYGL-LMI)
that is recognized by the 2C transgenic TCR (26).
OT-I T cells proliferated in response to
Kb/OVA257–264-LMI provided
that IL-2 was also added to the cultures, and responses increased with
increasing density of Ag on the beads (Fig. 1B). Only weak
proliferation occurred in the absence of exogenous IL-2, consistent
with the lack of costimulatory ligands on the LMI. The OT-I response to
LMI was Ag specific; no response by the OT-I cells was obtained using
Kb/SIYRYYGL-LMI in the absence of presence of
IL-2 (Fig. 1B). Reciprocal results were obtained when the
responding cell population was from a 2C mouse (data not shown). Thus,
CD8+ OT-I cells can respond in an Ag-specific
manner to the
Kb/OVA257–264 presented on
LMI. Based on these in vitro results, the in vivo experiments described
below were performed with LMI made using 25 ng
Kb/OVA257–264 per
106 beads, a level of Ag that stimulates a very
weak proliferative response in the absence of IL-2 and a plateau
response in the presence of IL-2 (Fig. 1).
LMI-mediated reduction of E.G7 tumor growth
Having established the efficacy of
Kb/OVA257–264-LMI in
vitro, experiments were conducted to determine whether these LMI could
mediate tumor growth reduction in vivo in the manner previously seen
using LMI coated with tumor cell plasma membranes (12).
Mice received adoptively transferred OT-I cells i.v. on day -2 and
were then injected with tumor (i.p.) and LMI (i.v.) on day 0. Fourteen
days later, peritoneal exudate cells were collected, and the number of
tumor cells was determined by flow cytometry. Mice that received just
tumor had 250 x 106 tumor cells in the
PC at this time, and the tumor burden was not significantly different
in mice that had received OT-I by adoptive transfer (Fig. 2). In contrast, mice that received OT-I
and Kb/OVA257–264-LMI had
a substantially reduced tumor load. Previous studies examining LMI
effects on syngeneic tumor growth used LMI made using tumor cell plasma
membranes, and thus displaying numerous proteins in addition to class I
Ag, and did not address whether class I/peptide Ag was necessary or
sufficient to mediate growth reduction. The results in Fig. 2, and additional experiments (data not shown), demonstrate that LMI bearing
the appropriate class I Ag in the absence of other cell surface
proteins is sufficient to mediate growth reduction of a syngeneic tumor
that expresses the same class I Ag.
|
In previous work examining the effects of LMI bearing purified
class I alloantigen, it was found that administration of LMI could
substantially augment ex vivo cytolytic activity generated in response
to allogeneic tumor but elicited no detectable response when
administered to mice that were not also challenged with the tumor
(12). The ability to identify and characterize adoptively
transferred OT-I cells provides a much more sensitive and direct means
of determining the effects of LMI on the Ag-specific T cells in vivo.
When adoptive transfer recipients of OT-I cells were challenged by i.v.
injection of
Kb/OVA257–264-LMI, there
was a smallbut significant increase in the number of OT-I cells
in the spleen and LN on days 2 through 4 in comparison with the numbers
in mice that did not receive LMI (Fig. 3, A and B).
|
C). This effect was specific to the OT-I cells,
because no significant differences were found in total CD8 T cell
numbers in the PC in response to LMI (Fig. 3D).
The increase in OT-I cell numbers in the spleen and LN in response to
LMI could result from altered migration of the cells or from
proliferation and clonal expansion. OT-I cells with light-scattering
properties consistent with blastogenesis were present at these sites in
LMI-treated mice beginning on day 1 (data not shown), suggesting that
some proliferation was occurring. This possibility was directly
assessed by labeling the OT-I cells with the fluorescent dye CFSE
before adoptive transfer (27). CFSE enters cells and
stably resides in the cytosol. When cells divide, the dye is diluted
equally among the daughter cells, thereby resulting in a 2-fold
decrease in mean fluorescence intensity at each cell division. By day 3
after administration of
Kb/OVA257–264-LMI, a
fraction of the OT-I cells in the LN had diluted their CFSE, with some
cells having undergone up to eight rounds of division (Fig. 4A). The CFSE profile (Fig. 4A) shows the progeny cells and therefore does not directly
reveal the fraction of the starting cells that have divided. Thus, for
example, a single original cell that has undergone 4 divisions will be
represented by 16 events in the peak with 4-fold diluted CFSE. Analysis
of the number of events in each peak shown in Fig. 4A, and
calculation of the number of starting cells from which these events
derived showed that 75% of OT-I cells remained undivided. Of the cells
that did divide, more than one-half underwent a single cell division,
and 40% underwent two to four cell divisions. In contrast,
administration of Kb/SIYRYYGL-LMI that are not
recognized by OT-I cells in vitro (Fig. 1) resulted in no
proliferation, as evidenced by a single bright peak of CFSE
fluorescence (Fig. 4B). Similar results were obtained when
OT-I cells in the spleen were examined (data not shown).
|
). In addition, almost all OT-I cells
found in the PC of LMI-treated mice displayed an activated
CD44highVLA-4high phenotype
(data not shown). The LMI-mediated changes in CD44 and VLA-4 expression
levels were specific to the OT-I cells; no changes in expression of
these markers were detected on the
CD8+Thy-1.1- C57BL/6 host
T cells (Fig. 5). Thus, although LMI administration in the absence of
tumor challenge causes modest clonal expansion of the Ag-specific CD8 T
cells (Fig. 3), with only 25% of the cells undergoing some cell
division (Fig. 4), it appears that the majority of the cells (75 to
80%) have interacted with Ag to result in up-regulation of CD44
and VLA-4 (Fig. 5) and that trafficking is altered to result in
increased migration to peripheral sites (Fig. 3C).
|
The rapid changes in surface marker expression and migration of
OT-I cells in response to LMI suggested that these effects might
contribute to more rapid and extensive increase in the number of OT-I
cells at the site of tumor growth, thereby contributing to reduced
tumor load in the PC. We therefore examined the responses of adoptively
transferred OT-I cells in mice challenged by i.p. injection of E.G7
tumor and either left untreated or treated with
Kb/OVA257–264-LMI or
control LMI (beads coated with streptavidin but no Ag) at the time of
tumor challenge by either i.p. or i.v. injection. In marked contrast to
the clonal expansion of OT-I in the spleen and LN caused by LMI in the
absence of tumor (Fig. 3), no significant differences in the numbers of
OT-I cells at these sites were found when tumor was present (Fig. 6A). The numbers of OT-I cells
in spleens and LN of the
Kb/OVA257–264-LMI-treated
mice on day 4 were the same as those in the tumor-bearing mice that
received no LMI or control LMI, and in mice that were adoptively
transferred but received neither tumor nor LMI (Fig. 6A).
The fact the the numbers of OT-I cells do not increase at these sites
is not due to a limitation of the maximum number of OT-I cells that can
be present, because challenge of adoptively transferred mice with
peptide Ag and adjuvant results in >2 x
106 OT-I cells at these sites (data not
shown).
|
B). These
numbers were significantly further increased in mice that had received
Kb/OVA257–264-LMI either
i.p. or i.v. but were unaffected by treatment with control LMI. Again,
the effect of the
Kb/OVA257–264-LMI was
specific to the OT-I cells; the number of
CD8+Thy-1.1- host T cells
showed some increase in the PC when tumor was present, but this was
unaffected by LMI treatment (data not shown).
Although the number of OT-I cells did not increase in the spleens of
tumor-bearing mice in response to
Kb/OVA257–264-LMI, the
OT-I cells at this site had responded to the LMI as evidenced by the
finding that the majority of the cells had up-regulated CD44 expression
on day 4 (Fig. 7A). In
contrast, the majority of the OT-I cells in the spleens of
tumor-bearing mice treated with control LMI had not up-regulated CD44
(Fig. 7B). Similar results were obtained when OT-I cells in
the LN of these mice were examined (data not shown). Spleen and LN OT-I
cells from
Kb/OVA257–264-LMI-treated
mice also showed increased VLA-4 expression on day 4, whereas those
from mice treated with control LMI did not (data not shown). As
expected, almost all of the OT-I cells in the PC of tumor-bearing mice
express high CD44 levels whether treated with
Kb/OVA257–264-LMI or
control LMI (Fig. 7, C and D).
|
A). In contrast, OT-I cell
numbers were significantly increased in the PC on
Kb/OVA257–264-LMI
administration to mice challenged with E.G7, but not in the PC of mice
challenged with EL-4, whereas control LMI had no effect in either case
(Fig. 8B). Thus, LMI-mediated expansion of the Ag-specific
OT-I cells at the site of tumor growth occurs only when the tumor bears
that specific Ag.
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
).
Given the large number of class I-restricted tumor-specific peptide
epitopes that have been identified for human tumors
(1, 2, 3, 4), and the ability to prepare HLA-class I/peptide
complexes in relatively large quantities (8), this raises
the possibility of preparing well-defined LMI for use in clinical
trials. Potential advantages of this approach include the ability to
prepare LMI having higher Ag densities than can be achieved using tumor
membranes, in which the amount of tumor-specific Ag is not known and
cannot be controlled, and the ability to treat patients from whom
sufficient autologous tumor material cannot be obtained to prepare
LMI. In previous studies, the effects of LMI treatment on CD8 T cell responses could only be indirectly assessed by measuring ex vivo cytolytic activity (12). In the experiments described here, the use of adoptively transferred TCR-transgenic CD8 T cells specific for the Ag on the LMI made it possible to more directly examine the mechanism by which LMI augment responses. OT-I mice express a transgenic TCR specific for OVA257–264 peptide bound to H-2Kb class I protein (18) and can specifically lyse E.G7 tumor, the EL-4 thymoma transfected with OVA. Thus, OVA serves as a pseudo-tumor Ag in this model, allowing the OT-I T cell response to the tumor to be monitored during the course of the response. By adoptively transferring OT-I CD8 T cells into normal recipients in small numbers and then challenging with tumor, the entire course of the response can be followed with respect to the locations, activation status, and clonal expansion of the tumor-specific CD8 T cells, and several studies have used this model (22, 23, 24, 25).
Ag-bearing LMI are uniquely effective in mediating reduction of
syngeneic tumor growth; the same tumor membrane Ag administered
alone or in CFA has little or no effect (14). It was
therefore somewhat surprising to find that administration of
Kb/OVA257–264-LMI caused
relatively little clonal expansion of OT-I T cells when tumor was not
present (Fig. 3). Only 3- to 5-fold expansion occurred in the
spleen or LN on days 2 through 4, and 25% of the OT-I cells divided
as measured by CFSE dye dilution (Fig. 4). This is in marked contrast
to immunization of adoptive transfer recipients with peptide in CFA or
along with LPS, where the Ag-specific cells expand 30- to 60-fold by
day 3 and all of the cells undergo multiple rounds of division (Ref.
28 and J. Goldberg and M. F. Mescher, unpublished
results). Although division and clonal expansion in response to LMI
were limited, the majority of the OT-I cells up-regulated CD44
expression, and about one-half also had increased VLA-4 expression
(Fig. 5). Thus, it appears that most of the OT-I cells had recognized
Ag and responded by altering expression of some surface receptors but
that this did not lead to proliferation of most of the cells.
It is generally thought that initial T cell clonal expansion in
response to Ag occurs in lymphoid organs, with the activated cells then
migrating to peripheral sites. However, when OT-I adoptive transfer
recipients are challenged by i.p. injection of E.G7 cells, increased
numbers of OT-I cells were observed in the PC beginning about day 4
postchallenge, and no increases could be detected in draining LN or
spleen (Ref. 24 and Fig. 6). When OT-I adoptive transfer
recipients were challenged by i.p. injection of E.G7, administration of
Kb/OVA257–264-LMI resulted
in substantially increased numbers of OT-I cells in the PC by day 4, 3-
to 10-fold more than in untreated mice (Figs. 6 and 8). A significant
fraction (>20%) of the OT-I cells in the PC at this time exhibit high
forward and side scatter characteristic of blasts, indicating that some
of the OT-I cells are proliferating at this site (data not shown). This
increase was specific in that it required that the LMI bear the
Kb/OVA257–264 Ag (Fig. 6)
and occurred only if the tumor growing in the PC expressed OVA
(Fig. 8).
The effects of
Kb/OVA257–264-LMI
administration on OT-I cells in the spleen and LN differed markedly
depending on whether or not tumor was growing in the PC. The clonal
expansion induced by LMI in LN and spleen in the absence of tumor (Fig. 3) was absent when tumor was present (Fig. 6A). Although
expanded numbers of OT-I cells were not detected in LN and spleen of
tumor-bearing mice, the cells at these sites had recognized and made
some response to the
Kb/OVA257–264-LMI as
indicated by the fact that they had up-regulated CD44 (Fig. 7) and
VLA-4 surface expression (data not shown).
Together these results suggest that administration of Ag-bearing LMI
may augment tumor-specific CD8 T cell responses to syngeneic tumor by
altering trafficking of the T cells so that they reach the site of
growing tumor more rapidly and in greater numbers, where they may
further expand and acquire effector function. It is not surprising that
the majority of Ag-specific cells do not proliferate in response to the
LMI (Figs. 3 and 4), because costimulatory ligands are not present and
the LMI may not induce production of inflammatory cytokines needed to
support an effective CD8 T cell response (28, 29, 30). The
majority of cells do recognize the LMI Ag, however, leading to altered
expression of surface receptors (Figs. 5 and 7). VLA-4 is particularly
interesting in this regard, as this integrin has been shown to be
up-regulated in response to priming with Ag (31, 32) and
has been implicated in having a role in directing migration of T
lymphocytes to peripheral sites (31, 33). Thus, cells in
the spleen and LN that have recognized LMI Ag and altered their
expression of VLA-4, and perhaps other receptors that regulate
trafficking, may migrate rapidly to peripheral sites and be further
stimulated to undergo proliferation and development of effector
function if they re-encounter Ag in an inflammatory environment. More
efficient migration to the PC after recognition of LMI Ag is
demonstrated by the fact that very few OT-I cells can be recovered from
the PC of untreated mice, whereas they are present at this site in
readily detectable numbers in mice treated with
Kb/OVA257–264-LMI (Fig. 3C).
Ag-specific CD8 T cells may directly recognize and respond to Ag
displayed on the surface of the LMI in vivo. This can clearly occur in
vitro (Fig. 1), and the ability of alloantigen-bearing LMI to augment
CTL responses to allogeneic tumor (12) suggests that it
can also occur in vivo, because it is unlikely that the native class I
alloantigen on the LMI would be taken up and represented in intact form
by host APC. However, in the case of LMI-bearing Ag to syngeneic tumor,
cross-priming may also occur (34, 35, 36, 37), with the host APC
taking up the LMI and presenting tumor-specific peptides (or OVA).
Further work will be needed to determine the in vivo fate of LMI and
whether direct recognition, cross-priming, or both are involved in the
LMI-mediated augmentation.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
2 Current address: Biology Department, Macalester College, 1600 Grand Avenue, St. Paul, MN 55105.
3 Current address: Department of Immunology, Roswell Park Cancer Institute, Elm and Carlton Streets, Buffalo, NY 14623.
4 Address correspondence and reprint requests to Dr. Matthew F. Mescher, Box 334 Mayo, 420 Delaware Street S.E., Minneapolis, MN 55455. E-mail address: mesch001{at}tc.umn.edu
5 Abbreviations used in this paper: LMI, large multivalent immunogen; LN, lymph node; Cy, cyclophosphamide.
Received for publication June 13, 2002. Accepted for publication October 28, 2002.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
in response to HLA-A2.1-binding melanoma and viral peptide antigens. J. Immunol. Methods 191:131.[Medline]
with an antigenic Tax peptide from human T lymphotropic virus type 1 and the class I MHC molecule HLA-A2. J. Immunol. 157:5403.[Abstract]
4 integrin directs virus-activated CD8+ T cells to sites of infection. J. Immunol. 154:5293.[Abstract]
4 integrin. J. Immunol. 150:1172.[Abstract]
This article has been cited by other articles:
|
|
 |
|
  S. Caserta, P. Alessi, J. Guarnerio, V. Basso, and A. Mondino Synthetic CD4+ T Cell-Targeted Antigen-Presenting Cells Elicit Protective Antitumor Responses Cancer Res., April 15, 2008; 68(8): 3010 - 3018. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. Schutz, M. Fleck, A. Mackensen, A. Zoso, D. Halbritter, J. P. Schneck, and M. Oelke Killer artificial antigen-presenting cells: a novel strategy to delete specific T cells Blood, April 1, 2008; 111(7): 3546 - 3552. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. Tao, B. H. Segal, C. Eppolito, Q. Li, C. G. Dennis, R. Youn, and P. A. Shrikant Aspergillus fumigatus extract differentially regulates antigen-specific CD4+ and CD8+ T cell responses to promote host immunity J. Leukoc. Biol., September 1, 2006; 80(3): 529 - 537. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 P. S. Heeger, P. N. Lalli, F. Lin, A. Valujskikh, J. Liu, N. Muqim, Y. Xu, and M. E. Medof Decay-accelerating factor modulates induction of T cell immunity J. Exp. Med., May 16, 2005; 201(10): 1523 - 1530. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. Moroz, C. Eppolito, Q. Li, J. Tao, C. H. Clegg, and P. A. Shrikant IL-21 Enhances and Sustains CD8+ T Cell Responses to Achieve Durable Tumor Immunity: Comparative Evaluation of IL-2, IL-15, and IL-21 J. Immunol., July 15, 2004; 173(2): 900 - 909. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. S. Mitchell, J. Kan-Mitchell, P. R. Morrow, D. Darrah, V. E. Jones, and M. F. Mescher Phase I Trial of Large Multivalent Immunogen Derived from Melanoma Lysates in Patients with Disseminated Melanoma Clin. Cancer Res., January 1, 2004; 10(1): 76 - 83. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
). Pooled LN
(axillary, brachial, inguinal, mesenteric, iliac), spleen, and
peritoneal exudate were harvested on days 1–4 and analyzed by flow
cytometry for the presence of OT-I cells
(CD8+Thy-1.1+) as described in Materials
and Methods. The number of OT-I cells in LN (A),
spleen (B), and PC (C) are shown.
D, The numbers of host-derived CD8 T cells
(CD8+Thy-1.1-) in the PC were also determined.
Values are the mean ± SD for three mice per group, and the
experiment is representative of three independent experiments.
). Pooled LN (A and C) and spleen
(B and D) cells were harvested on days 1–4 and
analyzed by flow cytometry. CD44 (A and B) and
VLA-4 (C and D) expression were determined for
OT-I cells (gating on
CD8+Thy-1.1+). Also shown
are expression levels on the host CD8 T cells in mice that had received
Kb/OVA257–264-LMI (
).
Values are means ± SD of two to four mice per group. Results are
representative of three independent separate experiments.
). Bars represent the
mean ± SD of two to four mice per group. Results are
representative of three separate experiments.