| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Departments of
*
Pathology and Laboratory Medicine,
Medicine, and
Otolaryngology, University of Pennsylvania, Philadelphia, PA 19104; and
§
Institute of Viral Preparation, Russian Academy of Medicine, Moscow, Russia
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
production in an
Ag-specific manner, and 4) dramatic tissue invasion of T cells. These
results support that CD86 plays a central role in CTL induction in
vivo, enabling non-bone marrow-derived cells to prime CTLs, a property
previously associated solely with bone marrow-derived
APCs. |
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
We recently reported that CD86 molecules play a prominent role in the Ag-specific induction of CD8+ CTL when delivered as vaccine adjuvants 8 . Coadministration of CD86 cDNA along with DNA encoding HIV-1 Ags i.m. dramatically increased Ag-specific T cell responses without a significant change to the level of the humoral response. This enhancement of CTL response was both MHC class I restricted and CD8+ T cell dependent. Similar results have been obtained by other investigators who also found that CD86, not CD80, coexpression results in the enhancement of T cell-mediated immune responses 9, 10 . Accordingly, we speculated that engineering of nonprofessional APCs, such as muscle cells, to express CD86 costimulatory molecules could empower them to prime CTL precursors. On the other hand, the enhancement effect of CD86 codelivery could also have been mediated through the direct transfection of a small number of professional APCs residing within the muscle tissue. Subsequently, these cells could have greater expression of costimulatory molecules and could in theory become more potent.
To investigate this issue, we constructed a set of bone marrow chimeric animals between normal mice and mice bearing a disrupted ß2-microglobulin (ß2m)4 gene. These bone marrow chimeras could respond and develop functional CTL responses following immunization with vaccinia virus. Next, we immunized chimeric animals with a DNA vaccine expressing HIV-1MN envelope protein (pCEnv) and plasmids encoding CD80 or CD86 genes (pCD80 or pCD86). Using this model, we observed that in vivo transfection of only pCEnv and pCD86 could enable non-bone marrow-derived cells to prime and expand CTLs. This study suggests that CD86, and not CD80, plays a central role in the generation of Ag-specific CTL responses. This result has important implications for our understanding of the generation of a primary immune response.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Four-week-old female C57BL/6J (ß2m+/+) and C57BL/6J-B2mtm1/Unc (ß2m-/-) mice were purchased from The Jackson Laboratory (Bar Harbor, ME). Both ß2m+/+ and ß2m-/- animals were used for reciprocal bone marrow transplant. The preparation of bone marrow chimeras has been previously described in detail 11 . Briefly, recipient mice (both ß2m+/+ and ß2m-/-) were depleted of NK cells with i.p. injections of 200 µg/ml of mAbs PK136 (anti-NK1.1) on day (-2) and (-1). This pretreatment prevents the rejection of bone marrow cells originating from C57BL/6J-B2mtm/Unc mice by radio-resistant NK cells in C57BL/6J mice 12 . On the day of reciprocal bone marrow transplant, recipient mice were lethally irradiated with a total of 1050 rad given in two equally divided doses 3 h apart. Donor mice were sacrificed, and bone marrow was harvested separately by flushing tibias and femurs. Bone marrow cells were depleted of mature T cells by incubation (37°C, 1 h) of cells with Low-Tox-M rabbit complement (Cedarlane Laboratories, Hornby, Ontario, Canada) following incubation (4°C, 45 min) with saturating concentration of a mixture of mAbs anti-CD4 (172.4), anti-CD8 (31M), and anti-Thy1.2 (mmt 1). Recipient mice were reconstituted with reciprocal bone marrow cells with an i.v. injection of 107 cells (0.3 ml). All animals were housed in a temperature-controlled, light-cycled facility at the University of Pennsylvania, and their care was under the guidelines of the National Institutes of Health and the University of Pennsylvania.
Immunization of mice
DNA vaccine construct encoding for the HIV-1MN envelope protein (pCEnv) was prepared as previously described 13, 14 . CD80 and CD86 expression cassettes were prepared as we described 8 . Each mouse received three i.m. injections (2 wk apart) with 50 µg of each DNA construct of interest formulated in PBS and 0.25% bupivacaine-HCl (Sigma, St. Louis, MO) as described earlier 8, 15 . Fifty micrograms of pCEnv administered in a regimen described above has been shown to induce moderate but positive immune responses in mice 8, 15 . This dosage was selected to demonstrate the enhancement of immune responses with the codelivery of costimulatory genes. We also injected animals with recombinant vaccinia virus, which express HIV-1 envelope protein (vMN462) (National Institutes of Health AIDS Research and Reference Reagent Program). Mice were injected i.v. with vMN462 (5 x 106 plaque-forming units (PFU) per mouse). Seven days later, spleens were removed and used for detection of direct CTL assay. Mice were also analyzed for indirect CTL after 4 wk of immunization with the same dose of vMN462.
Flow cytometry
The generation of chimeric mice was confirmed by FACS analysis
using mAbs to the 
3 domain of H-2Db molecule as
previously described 15 . One µg/ml of mouse mAbs 28-14-8s (IgG2a
isotype), which recognized 3 domain of H-2Db molecule
(courtesy of Dr. J. Frelinger, Chapel Hill, NC), was added to PBMC
(10 x 105) isolated from individual mice. Data were
analyzed by FACScan with CELLQuest data acquisition and software
(Becton Dickinson Immunocytometry Systems, San Jose, CA).
Immunohistochemical assays on muscle cells
Immunized leg muscle was examined immunohistochemically for the
in vivo expression of CD80, CD86, and envelope proteins as previously
described 8, 15 . Mouse quadriceps muscle was inoculated with 50 µg
of pCEnv + pCD80, pCEnv + pCD86, or control vector. Seven days
following inoculation, the mice were sacrificed, and the quadriceps
muscles were removed. The fresh muscle tissue was then frozen in O.C.T.
compound (Sakura Finetek USA, Torrance, CA). Four micron frozen
sections were made using a Leica 1800 cryostat (Leica, Deerfield, IL).
The sections were placed onto ProbeOn Plus slides (Fisher Scientific,
Pittsburgh, PA). The slides were fixed in acetone and blocked with
1.5% goat serum (Vector Laboratories, Burlingame, CA). To detect the
coexpression, the slides were incubated with biotinylated--gp120 Abs
(Immuno Diagnostics, Bedford MA.) diluted 1:20 along with either
FITC-conjugated anti-CD80 or anti-CD86 Abs (PharMingen, San
Diego, CA.) diluted 1:5 at 28°C for 12 h. The slides were then
incubated with streptavidin Texas Red (NEN Life Sciences, Boston MA) at
1:400 in PBS for 30 min at room temperature. To detect the presence of
lymphocytes in muscle, the slides were stained with hematoxylin and
eosin (H&E) stain. The slides were viewed with a Nikon OPTIPHOT
fluorescing microscope (Nikon, Tokyo, JAPAN) using a x40 objective
(Nikon Fluo x40 Ph3D2). Slide photographs were obtained using a Nikon
FX35DX camera with exposure control by Nikon UFX-II and Kodak
Ektachrome 160T slide film.
Infiltration of lymphocytes in muscle was analyzed by preparing frozen muscle sections from DNA-injected animals and stained with H&E stain (Vector Labs). The slides were also stained with anti-CD4 or anti-CD8 Abs (PharMingen).
ELISA
Fifty microliters of recombinant gp120 (ImmunoDiagnostics) diluted in 0.1 M carbonate-bicarbonate buffer (pH 9.5) to 2 µg/ml concentration was adsorbed onto microtiter wells overnight at 4°C as previously described 8, 15 . The plates were washed with PBS-0.05% Tween 20 and blocked with 3% BSA in PBS with 0.05% Tween 20 for 1 h at 37°C. Mouse antisera were diluted with 0.05% Tween 20 and incubated for 1 h at 37°C, then incubated with horseradish peroxidase-conjugated goat anti-mouse IgG (Sigma). The plates were washed and developed with 3'3'5'5' TMB (Sigma) buffer solution. The plates were read on a Dynatech MR5000 plate reader with the OD at 450 nm. The Ab titer was defined as the highest dilution of serum in which the absorbency of an experimental well exceeded the mean preimmune value by at least two SDs.
Cytotoxic T lymphocyte assay
A 5-h 51Cr release CTL assay was performed as we
described previously 8, 15 . Normal and vaccinia-infected EL-4 cells
(H-2b T cell lymphoma) were analyzed by FACS for their
ability to express MHC class II molecules. As we expected, EL-4 cells
did not express MHC class II molecules in either case (data not shown).
The effectors were stimulated nonspecifically for 2 days with CTL
culture media consisting of RPMI 1640 (Life Technologies, Grand Island,
NY), 10% FCS (Life Technologies), and 10% RAT-T-STIM without Con A
(Becton Dickinson) at 5 x 106 cells per ml. The
effectors were also stimulated for 4 days by specific target cells.
Preparation of specific targets for all CTL experiments was done
by infecting EL-4 cells with vMN462 as previously described 8, 15, 16 . As a nonspecific control for vaccinia virus and DNA immunization
experiments, the uninfected EL-4 cells and EL-4 cells infected with WR
vaccinia virus (National Institutes of Health AIDS Research and
Reference Reagent Program) were used, respectively. A standard chromium
release assay was performed, in which target cells were labeled with
100 µCi/ml Na251CrO4 for 2 h
and used to incubate with the effector cells for 5 h at 37°C.
CTL lysis was determined at E:T ratios ranging from 50:1 to 12.5:1.
Supernatants were harvested and counted on an LKB Clini gamma
counter. Percentage specific lysis was determined from the following
formula: 100 x [(experimental release - spontaneous
release)/(maximum release - spontaneous release)].
Maximum release was determined by lysis of target cells in 10% Triton X-100 containing medium. An assay was not considered valid if the value for the "spontaneous release" counts are in excess of 20% of the "maximum release." To calculate specific lysis of targets, the percentage lysis of nonspecific (WR-infected) targets was subtracted from the percentage lysis of specific (vMN462-infected) targets. The direct CTL assay was performed as described above except without in vitro stimulation of effector cells (neither specific or nonspecific) 8, 15 .
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Expression of ß2m is required for the cell
surface expression of MHC class I molecules, which play an important
role in the generation of protective cytotoxic immune responses against
infectious pathogens 17, 18 . These molecules present short peptide
fragments derived from foreign Ags synthesized in the cytosol to
CD8+ cytotoxic lymphocytes. We utilized
C57BL/6J-B2mtm/Unc mice, homozygous for the
ß2m knockout gene (ß2m-/-)
along with normal C57BL/6J (ß2m+/+) animals
for the generation of reciprocal bone marrow chimeras of the same
haplotype (H-2b) (Fig. 1
).
|
3
domain of H-2Db molecule 19 . Chimerization of these
animals was completed at 3 mo. The resulting chimeric mice displayed a
differential expression of MHC class I molecules on the surface of
muscle cells and APCs (Fig. 2). The
ß2m-/- 
ß2m+/+
mice possessed bone marrow-derived APCs (donor) without MHC class I
molecule expression and muscle cells (recipient) with MHC class I
molecule expression. In contrast, ß2m+/+
ß2m-/- chimeric mice possessed MHC class
I-positive bone marrow-derived APCs and MHC class I-negative muscle
cells. Several reports have suggested that a low level of -chains
(below the level of FACS sensitivity) could be in fact expressed on the
surface of ß2m-/- cells and that these
-chains could in turn bind free ß2m and cognately
present foreign peptides to CD8+ T cells 17, 20 . On the
other hand, ß2m+/+
ß2m-/- chimeric mice, which do not have MHC
Class I+ thymic epithelial cells, may not generate
functional CD8+ CTLs during T cell differentiation 21 .
Even positive selection of these lymphocytes by MHC Class
I+ bone marrow cells has been demonstrated 22 . In fact,
it has been shown that both ß2m-/-
ß2m+/+ and ß2m+/+
ß2m-/- chimeras generated significant
numbers of CD4 and CD8 cells 22 . However, we first decided to
investigate the ability of the reciprocal chimeric mice to generate
anti-viral CD8+ CTL immune responses. We immunized
normal C57BL/6 (ß2m+/+) and
ß2m-/- knockout mice as well as the
ß2m-/- ß2m+/+
and ß2m+/+
ß2m-/- chimeric mice with recombinant
vaccinia virus (vMN462). Subsequently, we analyzed the anti-viral
CD8+ CTL immune response generated in these animals. As
shown in Fig. 3, we observed that both
C57BL/6 and ß2m+/+
ß2m-/- mice generated primary (direct) and
secondary (indirect) CD8+ CTL responses. CTL responses
in C57BL/6 mice were more potent than in
ß2m+/+ ß2m-/-
chimeras, and these results agree with earlier observations 22 . In
contrast, we did not observe anti-viral CD8+ CTL
responses in ß2m-/- knockout or
ß2m-/- ß2m+/+
chimeric mice. Therefore, these results demonstrate that the chimeras
between normal and ß2m-knockout mice provide a clean
model to examine the role of muscle cells in MHC class I-restricted T
cell responses following DNA immunization.
|
|
We had previously reported that i.m. injection of mice
with plasmids encoding for CD80 and CD86 costimulatory molecules
resulted in expression of CD80 and CD86 molecules in muscle with
similar transfection efficiencies 8 . We further investigated whether
the codelivery of two expression constructs (one encoding for HIV-1
envelope protein and one encoding for a costimulatory molecule) results
in coexpression of these proteins in the same cell. We coimmunized
ß2m+/+ mice with a DNA vaccine expressing
HIV-1MN envelope protein (pCEnv) and plasmids encoding CD80
or CD86 genes (pCD80 or pCD86) or control plasmid (pCDNA3). We
immunohistochemically examined the expression of CD80, CD86, and
envelope proteins in the injected leg muscles (Fig. 4). We observed that coimmunization with
pCEnv + pCD80 or pCEnv + pCD86 resulted in coexpression of
these proteins in muscle cells. We also observed that the coexpression
levels of envelope and CD80 or envelope and CD86 from the mice injected
with pCEnv + pCD80 and pCEnv + pCD86, respectively, were similar. In
contrast, control legs did not show expression of these
proteins.
|
We next analyzed both humoral and cellular immune responses in
chimeric and control mice immunized with plasmids encoding viral Ag
and costimulatory molecules. Humoral immune responses in sera
collected from experimental mice before and after immunization were
analyzed by ELISA. As shown in Table I,
HIV-1 envelope-specific humoral responses were generated in both types
of chimeras. Humoral immune responses of
ß2m-/- mice were similar to those of the
chimeric mice (data not shown). These results demonstrate that
Ag-specific humoral immune responses could be generated in the
ß2m knockout mice after plasmid DNA immunization and
agree with results previously reported in this model system following
protein immunization 23 . Furthermore, these results indicate that the
coimmunization of reciprocal chimeras with either pCD80 or pCD86 had
little effect on the specific Ab endpoint titer induced by pCEnv
immunizations, as we had previously observed in normal BALB/c mice 8 .
|
ß2m-/-
chimeric mice immunized with control plasmid, pCEnv, pCEnv + pCD80 or
pCEnv + pCD86 (Fig. 5). However,
ß2m-/- ß2m+/+
mice coimmunized with pCEnv + pCD86, but not pCEnv or pCEnv + pCD80,
resulted in a high level of envelope-specific CTL (37% at E:T ratio of
50:1). To further examine the potency of CTL induction, we analyzed the
ability to induce direct, unstimulated CTL responses in
ß2m-/- ß2m+/+
chimeric mice (Fig. 6) 8 . The mice
immunized with pCEnv did not induce specific killing. In contrast,
specific lysis of 23% was observed from the pCEnv + pCD86 immunization
group at an E:T ratio of 50:1 and titered out to 11% at the 12.5:1 E:T
ratio. Since the bone marrow-derived cells in
ß2m-/- ß2m+/+
mice could not generate substantial anti-viral CD8+
CTLs (Fig. 3
), these results suggest that coexpression of HIV envelope
and CD86 molecules on nonhemopoietic cells can enable them to prime
anti-HIV-1-specific CTL responses. In addition, the results from
the immunologically normal ß2m+/+
ß2m-/- chimeras (Fig. 3) indicate that the
enhancement effect of CD86 molecules on CTL expansion is not observed
through conversion of the small number of APCs, but rather is more
prevalent on the nonhemopoietic cells.
|
|
and IL-4
cytokines are produced not only by CD4+ but also by
CD8+ T cells 24 . IFN- is intricately involved in the
regulation of T cell-mediated cytotoxic immune responses 25, 26 while
IL-4 plays a dominant role in B cell-mediated immune responses 27 .
Therefore, in addition to our CTL analysis, we collected supernatant
from the effector cells stimulated in vitro for CTL assay and tested
them for the release of IFN- and IL-4. As shown in Fig. 7, the level of IFN- release
corresponded with the level of CTL response seen in Fig. 5. In fact,
the level of IFN- released from ß2m-/-
ß2m+/+ mice immunized with pCEnv + pCD86
(45 ng/ml) was at least three times those of the other groups. On
the other hand, the levels of IL-4 released from all groups were
similar. Therefore, IFN- release data supports that CD86 expression
on nonhemopoietic cells could prime cytokine induction primarily in
the context of MHC class I expression, supporting direct TCR coligation
by nonprofessional APCs.
|
To further clarify the ability of non-bone marrow cells
transfected with CD86 to directly drive T cells, we looked for direct
evidence of T cell ligation to transfected muscle cells in vivo. We
observed much more infiltration of lymphocytes into the muscle of
ß2m+/+ mice immunized with pCEnv + pCD86 than
in the muscle of control or pCEnv + pCD80-immunized
ß2m+/+ mice at 7 days postimmunization (Fig. 8). These numerous infiltrating
lymphocytes at the site of Ag and CD86 expression seem likely to attack
the presenting muscle cells. We stained the slides
immunohistochemically for T cells and observed that the infiltrating T
cells included both CD4+ and CD8+ T cells (Fig. 9). The lymphocyte infiltration in the
immunized muscle was observed to clear within 1 mo, correlating to the
duration of Ag expression following cDNA expression (data not shown).
Animals exhibited no clear phenotypic effects of this invasion
compared with nonvaccinated animals (data not shown). Examination
of muscle sections at later time points demonstrated a normal muscle
phenotype without lymphocyte invasion. It is interesting that even
during the early phase of lymphocyte infiltration, the mice behaved
normally. These results suggest that muscle cells that are engineered
to express viral Ag along with MHC class I and CD86, but not CD80,
molecules could effectively attract lymphocytes and directly interact
with them. These data clearly distinguish that attraction, per se, is
not the function of CD80.
|
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Recently, three groups (including ours) independently reported that coadministration of CD86 cDNA along with DNA Ags i.m. dramatically increased Ag-specific CTL responses 8, 9, 10 . In contrast, this enhancement of CTL responses was not observed with CD80 coexpression 8, 9, 10 . More recently, CD80 has been reported to enhance CTL responses when it was coexpressed with plasmid DNA encoding a peptide minigene 42 . This result is different from three other recent papers suggesting that free epitopes rather than natural Ags can behave uniquely. While further work will be required to investigate the differences, the similarities of the results from different groups are instructive.
To investigate whether the bone marrow-derived professional APCs or
muscle cells were responsible for the enhancement of CTL priming
following CD86 coadministration, we developed a set of bone marrow
chimeric animals using ß2m knockout mice. The
ß2m chimeric animals represent an appropriate model to
conduct our studies. These chimeric animals provide a straightforward
way to study CTL responses, which are restricted by MHC class I
molecules of the bone marrow-derived cells or the muscle cells. The
ß2m-/- ß2m+/+
mice possessed bone marrow-derived APCs (donor) without MHC class I
molecule expression and muscle cells (recipient) with MHC class I
molecule expression. In contrast, ß2m+/+
ß2m-/- chimeric mice possessed MHC class
I-positive bone marrow-derived APCs and MHC class I-negative muscle
cells (Fig. 1). ß2m, peptide, and -chain (heavy chain)
are all necessary for effective transport and expression of functional
MHC class I molecules. Therefore, the cells that do not express
ß2m do not express MHC class I molecules on the cell
surface 43 . Many studies have confirmed that the -chains of MHC
class I are not detectable by FACS analysis in
ß2m-/- animals 20, 23, 43, 44, 45 . As
expected, our FACS analysis failed to detect a significant presence of
MHC class I -chains on the PBMC of
ß2m-/- or
ß2m-/- ß2m+/+
chimeric mice (Fig. 2).
There was a concern that ß2m+/+
ß2m-/- chimeric mice may not possess
functional CD8+ T cells without MHC Class I+
thymic epithelial cells to educate them 21 . It has been reported,
however, that in ß2m+/+
ß2m-/- chimeric mice, which have
ß2m-/- thymic epithelial cells and
ß2m+/+ hemopoietic cells, CD8+ T
cells are educated by MHC class I+ bone marrow cells 22, 23 . In an elegant set of experiments, Bix and Raulet demonstrated that
ß2m+/+ hemopoietic cells grafted into the
ß2m-/- host direct positive selection of
functional CD8+ T cells 22 . They also found that this
positive selection of CD8+ T cells was provided by the bone
marrow cells and not by the potentially reconstituted thymic epithelial
MHC class I molecules (ß2m-/- thymic
epithelial cells reconstituted by donor ß2m) 22 .
Earlier, the rate of positive selection of CD8+ T cells by
ß2m+/+ hemopoietic cells in
ß2m+/+ ß2m-/-
chimeric mice was observed to be less than that of normal mice (about
1/6 of that in normal mice) 22 . Nevertheless, these chimeric mice
possessed functional CD8+ T cells that mounted strong CD8
and MHC class I-restricted CTL responses 22 . The responses were
comparable to those observed in normal mice 22 . We also analyzed the
immunocompetency of our ß2m+/+
ß2m-/- chimeras and observed that these
mice produced functional CD8+ CTL response against vaccinia
virus, even though it was at a lower level than that of
ß2m+/+ mice (Fig. 3).
Several reports have suggested that a low level of -chains (below
the level of FACS sensitivity) could be expressed on the surface of
ß2m-/- cells and that these -chains
could in turn bind free ß2m and cognately present foreign
peptides to CD8+ T cells 17, 20 . These studies, however,
were performed in vitro, and the activity detected was significantly
less than that induced by wildtype ß2m-expressing cells.
Although in vitro ß2m-independent presentation of
endogenous self H-2Db peptides by MHC class I restricted
-chains has been reported, such effects were not observed in in vivo
experiments 22, 23, 46 . More importantly, we did not observe
antiviral CTL response after inoculation of
ß2m-/- ß2m+/+
animals with vMN462 (Fig. 3). Therefore, even if a low level of the
MHC-class I molecules was expressed on the bone marrow-derived APCs of
these mice, these molecules were functionally ineffective.
One should also note that MHC-class I alloantigen-specific
CD8+ CTL responses in ß2m-deficient mice has
been reported 47, 48, 49 . It has been shown that both
ß2m+/+ and ß2m-/-
allospecific CD8+ CTL could lyse
ß2m+/+ and ß2m-/-
target cells, but the lysis of ß2m-/-
targets was 50 times less than that of ß2m+/+
targets 47 . Furthermore, the frequency of alloantigen-specific
CD8+ CTL precursors has been estimated to be 20–200 times
higher than that of precursors against viral Ags 50 . To our
knowledge, no studies have been reported on the induction of
CD8+ CTL in ß2m-/- and
ß2m-/- ß2m+/+
mice during viral infection 23, 51 . Accordingly, our
ß2m-/- or
ß2m-/- ß2m+/+
animals immunized with vMN462 or pCEnv did not generate Ag-specific
CD8+ CTLs. Although many investigators using the
ß2m knockout mice have reported MHC class II-restricted
CD4+ CTL in their experiments 23, 51 , such analysis was
irrelevant in our experiment using EL4 cells as target cells since they
do not express MHC class II molecules. These results established the
use of ß2m+/+
ß2m-/- or
ß2m-/- ß2m+/+
chimeric animals as an appropriate model to examine the
CD8+ antiviral CTL responses restricted by the MHC class I
molecules of the APCs or muscle cells.
Since ß2m-/- professional APCs could not be
involved in the induction of CD8+ CTL in
ß2m-/-ß2m+/+
mice, the high level of CTL responses generated by immunization of
these animals with pCEnv + pCD86 supports our hypothesis that
nonhemopoietic cells such as muscle cells can be engineered to function
as APCs. In contrast, pCEnv + pCD80 immunization did not result in high
levels of CTL responses (Figs. 5 and 6). These results were consistent
with previous results obtained in normal mice, and they again support
the critical role of CD86 expression presentation of Ag. However, these
studies shed little light on the function of CD80 in this process. CD80
and CD86 have limited homology (28%), but both have been shown to
efficiently costimulate proliferation of T cells and to induce cytokine
production in vitro 2, 52, 53 . Although more work may be needed to
fully elucidate the differential roles of CD80 and CD86, we speculate
that CD86 may be able to activate CD8+ T cells better than
CD80 because of a higher affinity for CD28, which is involved in T cell
activation, and we could speculate that CD80 could have higher avidity
for CTLA-4, which is involved in suppression of T cells 54, 55, 56, 57 .
We have previously reported that the dosage of pCEnv selected for
immunization (50 µg) induces a low level of CTL responses (5–10%
specific lysis) in ß2m+/+ normal mice 8 . We
utilized this dosage in the current experiment to detect the full CTL
enhancement effect of CD86 coimmunization in the reciprocal chimeric
mice. As shown in Fig. 5, we observed that immunization of
ß2m+/+ ß2m-/-
chimeras (which had normal professional APCs) with pCEnv resulted in a
low level of CTL responses, similar to the level observed in normal
mice. On the other hand, we did not observe the enhancement of CTL
responses in ß2m+/+
ß2m-/- chimeric animals immunized with
pCEnv+pCD86. There could be several possible explanations for this
finding. First, enhancement of antiviral CTL responses with CD86
coadministration may be restricted to the expression of CD86 in
non-APCs, such as muscle cells. This hypothesis is supported by the
fact that most professional APCs, such as the dendritic cells and
macrophages, already express CD86 constitutively, and the expression is
further increased upon activation 5, 6, 7 . Thus, transfection of these
professional APCs with CD86 would not enhance their ability to activate
and expand T cells. Moreover, although several studies have identified
the importance of directly transfecting APCs, such as the dendritic
cells and macrophages, via DNA immunization, the efficiency of such
direct in vivo transfection has been low 40, 41 . On the other hand,
i.m. injection of DNA results in a predominant level of transfection in
muscle cells. Therefore, potentiating these nonprofessional APCs to
coexpress CD86 may be crucial for the priming and enhancement of CTL
expansion. In the case of the ß2m+/+
ß2m-/- mice, whose muscle cells (and other
nonhemopoietic cells) lacked functional MHC class I molecules, such
CD86-mediated CTL expansion would not be possible. One of the other
possible mechanisms of transferring Ag or peptides from peripheral
cells to the professional APCs is thought to be a "hand-off"
mechanism 38, 39, 58, 59 between transfected non-APC protein factory
and the APC. If the "hand-off" hypothesis is correct, then the CTL
experiment results from the ß2m+/+
ß2m-/- chimera immunization suggest that
such a mechanism would be MHC class I restricted. Finally, it is
possible that the animals that responded to vaccinia virus could not
generate anti-Env CTLs due to differences in quantity of
CD8+ T cell precursors specific to the Env protein vs the
large number of vaccinia virus Ags.
Recently, another group has reported on the potential of converting
muscle cells to APCs. Iwasaki et al. utilized chimeric mice generated
with H-2bxd recipients and H-2b or
H-2d donors and reported on injection of wildtype
nucleoprotein immunogen with CD86 60 . Earlier they had
reported significant enhancement using plasmid encoding for a mutant
form of nucleoprotein and CD86 in normal animals 10 . In their
subsequent chimera experiments, however, they did not observe any
significant enhancement with the addition of CD86 60 . No positive
effects were observed in their chimeric animals; therefore, it was
difficult to draw extensive conclusions in this study due to a lack of
CD86 effect. In our experiments, pCEnv vaccination of
ß2m-/- ß2m+/+
chimeric mice did not generate antiviral CTL responses (less than 6%
of specific lysis), but immunization with pCEnv and pCD86 induced high
anti-HIV CTL activity before (up to 23%) and after (up to 37%) in
vitro stimulation (Figs. 5 and 6). The positive results in these
experiments allow us to draw additional insight into the role of CD86
and the function of bone marrow-derived APCs in CTL induction.
Moreover, our results do not conflict with published results indicating
the bone marrow-derived APCs as the presenter of Ag from DNA
immunization 38, 39, 40, 41, 60 . Rather, we show that nonhemopoietic cells,
such as muscle cells, can be engineered to present DNA-encoded Ag by
potentiating them to express the key costimulatory molecule CD86. This
molecule is normally the sole providence of bone marrow-derived cells.
This interpretation, which is the most direct and simplest, is further
supported by the immunohistochemistry results presented in Fig. 8. We
observed more infiltration of lymphocytes into the muscle of mice
immunized with pCEnv + pCD86 than in the muscle of pCEnv +
pCD80-immunized mice. These infiltrating cells included both
CD4+ and CD8+ T cells (Fig. 9). It appears that
CD86 expressed in the absence of other cytokines or costimulatory
molecules can directly prime T cells that can then destroy the
presenting target. Macrophages and dendritic cells would in this sense
be analogous to the CD86-displaying muscle cells and must be capable of
activating CTLs without suffering their wrath. Protection from
inadvertent destruction during the activation process must be mediated
by other cell surface molecules, which might include other members of
the B7 family, which would be expected to be coordinately regulated.
Furthermore, as shown by the higher level of lymphocytic infiltration,
transfection of muscle cells with CD86 may enhance their ability to
chemoattract T cells. This is especially interesting since it has been
recently reported that CD8+T cells may expand Ag-specific
responses in vivo through the elaboration of specific chemokines at the
peripheral site of infection during the effector stage of the immune
response 61 . Such peripheral restimulation may be crucial in the
expansion of T cell responses and bears further investigation.
Our findings clarify that one function of the CD86 molecule on APCs itself is, in the context of MHC class I molecules, to prime and expand T cells. The demonstration that this function segregates with CD86 itself even on non-bone marrow-derived APCs could represent an important step in the pursuit of rationally designed vaccines and immune therapies through the control of MHC class I restriction.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
2 Both authors contributed equally to this work.
3 Address correspondence and reprint requests to Dr. Michael G. Agadjanyan, 505 B-Stellar Chance Building, 422 Curie Boulevard, Philadelphia, PA 19104. E-mail address:
4 Abbreviations used in this paper: ß2m, ß2-microglobulin; H&E, hematoxylin and eosin; pCEnv, plasmid encoding HIV-1MN envelope protein; pCD80, plasmid encoding CD80 gene; pCD86, plasmid encoding CD86 gene; PCDNA3, control plasmid; WR, wild-type vaccinia virus; vMN462, recombinant vaccinia virus-expressing HIV-1 envelope protein.
Received for publication October 8, 1998. Accepted for publication December 16, 1998.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
-stimulated monocytes: a potential mechanism for amplification of T cell activation through the CD28 pathway. Cell. Immunol. 137:429.[Medline]
by activated human CD4+ and CD8+ T cell clones. J. Immunol. 141:849.[Abstract]
-chain to a cross-reactive CD8+ CTL clone. J. Immunol. 153:4070.[Abstract]
ß T-cell receptor cytotoxic T-cells with major histocompatibility complex (MHC) class I antigen-directed cytotoxicity in ß2-microglobulin, MHC class I deficient mice. Proc. Natl. Acad. Sci. USA 90:2837.This article has been cited by other articles:
|
|
 |
|
  S. Radhakrishnan, L. T. Nguyen, B. Ciric, D. Flies, V. P. V. Keulen, K. Tamada, L. Chen, M. Rodriguez, and L. R. Pease Immunotherapeutic Potential of B7-DC (PD-L2) Cross-Linking Antibody In Conferring Antitumor Immunity Cancer Res., July 15, 2004; 64(14): 4965 - 4972. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. Giri, K. E. Ugen, and D. B. Weiner DNA Vaccines against Human Immunodeficiency Virus Type 1 in the Past Decade Clin. Microbiol. Rev., April 1, 2004; 17(2): 370 - 389. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. G. Agadjanyan, M. A. Chattergoon, M. J. Holterman, B. Monzavi-Karbassi, J. J. Kim, T. Dentchev, D. Wilson, V. Ayyavoo, L. J. Montaner, T. Kieber-Emmons, et al. Costimulatory Molecule Immune Enhancement in a Plasmid Vaccine Model Is Regulated in Part Through the Ig Constant-Like Domain of CD80/86 J. Immunol., October 15, 2003; 171(8): 4311 - 4319. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
V. V. Parekh, D. V. R. Prasad, P. P. Banerjee, B. N. Joshi, A. Kumar, and G. C. Mishra B Cells Activated by Lipopolysaccharide, But Not By Anti-Ig and Anti-CD40 Antibody, Induce Anergy in CD8+ T Cells: Role of TGF-{beta}1 J. Immunol., June 15, 2003; 170(12): 5897 - 5911. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. Vasu, A. Wang, S. R. Gorla, S. Kaithamana, B. S. Prabhakar, and M. J. Holterman CD80 and CD86 C domains play an important role in receptor binding and co-stimulatory properties Int. Immunol., February 1, 2003; 15(2): 167 - 175. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
X. Shen, S. B. J. Wong, C. B. Buck, J. Zhang, and R. F. Siliciano Direct Priming and Cross-Priming Contribute Differentially to the Induction of CD8+ CTL Following Exposure to Vaccinia Virus Via Different Routes J. Immunol., October 15, 2002; 169(8): 4222 - 4229. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. Mercier, H. Gahery-Segard, M. Monteil, R. Lengagne, J.-G. Guillet, M. Eloit, and C. Denesvre Distinct Roles of Adenovirus Vector-Transduced Dendritic Cells, Myoblasts, and Endothelial Cells in Mediating an Immune Response against a Transgene Product J. Virol., February 22, 2002; 76(6): 2899 - 2911. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
V. Ayyavoo, K. Muthumani, S. Kudchodkar, D. Zhang, P. Ramanathan, N. S. Dayes, J. J. Kim, J.-I. Sin, L. J. Montaner, and D. B. Weiner HIV-1 viral protein R compromises cellular immune function in vivo Int. Immunol., January 1, 2002; 14(1): 13 - 22. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
K. Chan, D. J. Lee, A. Schubert, C. M. Tang, B. Crain, S. P. Schoenberger, and M. Corr The Roles of MHC Class II, CD40, and B7 Costimulation in CTL Induction by Plasmid DNA J. Immunol., March 1, 2001; 166(5): 3061 - 3066. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 D. J. Shedlock and D. B. Weiner DNA vaccination: antigen presentation and the induction of immunity J. Leukoc. Biol., December 1, 2000; 68(6): 793 - 806. [Abstract] [Full Text]
|
|
|
|
 |
|
 T. Maeda, M. Towatari, H. Kosugi, and H. Saito Up-regulation of costimulatory/adhesion molecules by histone deacetylase inhibitors in acute myeloid leukemia cells Blood, December 1, 2000; 96(12): 3847 - 3856. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
