| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Center for Immunotherapy of Cancer and Infectious Diseases, University of Connecticut Health Center, Farmington, CT 06030
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Given that APCs can both prime and tolerize naive T cells, understanding the functional relationship between the APCs that induce these divergent processes will be fundamental to understanding how naive T cells distinguish self from pathogen-derived Ags. One possibility is that tolerogenic APCs (tAPCs) and immunogenic APCs (iAPCs) are functionally distinct, either arising from distinct populations or representing differentially programmed versions of the same population. Alternatively, tolerogenicity vs immunogenicity might not be determined by the innate functional properties of the APCs themselves. For example, the duration of Ag presentation might be critical; short term presentation of rapidly cleared pathogen-derived Ags would induce immunity, while overstimulation caused by chronic presentation of self-Ags would lead to clonal exhaustion in which primed T cells are driven into a tolerant state (8, 9, 10, 11). Another mechanism by which functionally equivalent APCs could induce either tolerance or priming would be for pathogens to induce the activity of a third party (i.e., non-APC) that delivers immunogenic signals directly to naive T cells that redirect them to undergo immunogenic rather than tolerogenic differentiation in response to generic APCs.
To study the functional relationship between tAPCs and iAPCs, we developed an adoptive transfer system in which naive TCR transgenic CD4 cells specific for the model Ag influenza hemagglutinin (HA) can either be primed to express effector function upon transfer into mice infected with an HA-recombinant vaccinia virus or tolerized upon transfer into mice expressing HA as a parenchymal self-Ag (1, 12). Our results indicate that tAPCs, which are both necessary and sufficient to induce tolerization, are likely to be functionally distinct from iAPCs given that their different activities cannot be explained by either their kinetics of Ag presentation or the activity of a third party.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Mice were on the B10.D2 background (except where otherwise indicated). C3-HAlow and C3-HAhigh transgenic mice both express the influenza HA gene (A/PR/8/34 Mount Sinai strain) under the control of the rat C3 (1) promoter, which directs HA expression to a variety of nonlymphoid organs. Although both transgenic founder lines express HA in the same subset of tissues, HA protein expression in C3-HAhigh mice appears to be at least 1000-fold higher than that in C3-HAlow mice (1, 12). 6.5 TCR transgenic mice express a clonotypic TCR that recognizes an I-Ed-restricted HA epitope (110SFERFEIFPKE120) (13).
BM chimeras
BM chimeras were generated as previously described
(1) with the following modifications.
Parent
F1 chimeras were generated using
F1 hosts (B10.D2 x C57BL/6J
(H-2bxd)) that either received 1000 rad ionizing
radiation in a single dose or 1300 rad in split doses of 650 rad 4
h apart. One day before allogeneic BM transplantation, hosts were
depleted of NK cells by i.p. injection of 15 µl rabbit
anti-asialo GM1 gammaglobulin (WAKO Chemicals, Richmond, VA).
Adoptive transfers
Adoptive transfers of 2.5 x 106 naive clonotypic CD4 cells were performed as previously described (1, 12) with the following modifications. For adoptive transfers using CFSE-labeled clonotypic CD4 cells, 6.5 transgenic donors on a B10.D2 background expressing the Thy1.1 congenic marker were used. Before labeling with CFSE (Molecular Probes, Eugene, OR) (12), 6.5 LN preparations were depleted of CD8 cells using anti-mouse CD8-conjugated magnetic beads (Dynal, Lake Success, NY). All adoptive transfer recipients expressed the Thy1.2 congenic marker.
Proliferation assays
Proliferation assays were performed as previously described (1, 12), with data expressed as counts per minute/clonotypic CD4 cell (mean ± SEM ).
Recombinant vaccinia
The recombinant vaccinia virus expressing HA (vacc-HA) has previously been described (14), and the recombinant vaccinia vacc-GH that expresses the CMV GH protein was provided by Drs. J. Shanley and C. Wu (University of Connecticut Health Center). Both vacc-HA and vacc-GH were constructed by recombining either the HA or GH gene, respectively, into the thymidine kinase gene of wild-type vaccinia. Vaccinia were amplified in HuTK- cells, purified over sucrose, and titrated on HuTK- cells using standard protocols (15). Inoculations were performed by i.p. injection of the indicated dose in 0.2 ml HBSS.
Flow cytometry
Surface staining for the clonotypic TCR 6.5, CD4, and CD44 on
non-CFSE-labeled cells was performed as previously described (1, 12). Surface staining of CFSE-labeled clonotypic CD4 cells was
performed using PerCP-conjugated anti-Thy1.1 and either
PE-conjugated anti-CD25 or an isotype control (BD Pharmingen, San
Diego, CA). Intracellular cytokine staining was performed by
stimulating 1 x 107 splenocytes (prepared
from adoptive transfer recipients) with 100 µg/ml synthetic HA
peptide and 5 µg/ml brefeldin A (Sigma-Aldrich, St. Louis, MO) for
5 h at 37°C in 1 ml CTL medium (1). Cells were then
washed once in FACS buffer, fixed in PBS and 2% formaldehyde for 5 min
at 37°C, permeabilized by washing twice in FACS buffer containing
0.25% saponin, and then stained with PerCP-conjugated anti-Thy1.1
and PE-conjugated mAbs specific for IL-2, IFN-
, or an isotype
control (BD Pharmingen). All quantitative FACS data are expressed as
the mean ± SEM. To allow direct comparison of FACS data collected
from separate experiments, all samples were analyzed on the same flow
cytometer (FACScan; BD Biosciences, San Jose, CA) using identical
settings.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
In transgenic mice expressing low levels of the model parenchymal
self-Ag HA, adoptively transferred naive clonotypic TCR transgenic CD4
cells that recognize an I-Ed-restricted HA
epitope become tolerized upon encountering BM-derived APCs that have
acquired and presented parenchymally derived HA. Thus, in
parentF1 BM chimeras in which the
HA-expressing parenchyma are genetically capable of presenting the HA
epitope (H-2bxd), but APCs are not because they
express a nonrestricting haplotype (i.e., H-2b),
adoptively transferred clonotypic CD4 cells remained naive
(1). Since these initial studies used mice expressing low
levels of HA (1, 12), it might have been
possible that high level HA expression would enable parenchyma to
directly present the I-Ed-restricted epitope to
naive clonotypic CD4 cells. To test this possibility, we
generated parentF1 chimeras using both the
C3-HAlow as well as the
C3-HAhigh transgenic mice that express HA in the
same subset of parenchymal tissues, but differ in their level of HA
expression by at least 1000-fold (12). In the first
experiment (Fig. 1
A), F1
(H-2bxd) nontransgenic (NT),
C3-HAlow
andC3-HAhigh mice were lethally
irradiated with 1000 rad and reconstituted with NT BM expressing either
the HA-restricting (H-2d) or a nonrestricting
(H-2b) haplotype. Consistent with our previous
results (1, 12), clonotypic CD4 cells recovered 9 days
post-transfer from the spleen and LN of C3-HAlow
as well as C3-HAhigh chimeras reconstituted with
H-2d BM (dlow and dhigh, respectively)
expressed levels of the activation marker CD44 that were at least
3-fold higher than those of naive counterparts recovered from control
NT chimeras, indicating that the clonotypic CD4 cells had encountered
the HA epitope and lost their naive phenotypes. Furthermore, clonotypic
CD4 cells recovered from C3-HAlow chimeras
reconstituted with H-2b BM (blow) expressed
low levels of CD44, indicating that they retained a naive phenotype and
confirming that direct parenchymal presentation of the
I-Ed-restricted epitope does not occur in
chimeras expressing low levels of parenchymal HA (1).
Interestingly, clonotypic CD4 cells recovered from bhigh chimeras
exhibited a moderate increase in CD44 expression (2-fold over control
levels), possibly resulting from either direct parenchymal Ag
presentation or radioresistant host-derived APCs
(H-2bxd). We reasoned that if the latter
possibility were correct, a more severe myeloablative regimen would
eliminate this residual presentation. Consistent with this prediction,
when hosts received 1300 rad, clonotypic CD4 cells recovered from
bhigh chimeras expressed CD44 levels that were equivalently low as
those of the NT control chimeras (Fig. 1B). Thus, when the
host-derived APCs were efficiently eliminated, the parenchymal cells
expressing high levels of HA were revealed to be incapable of
presenting the class II-restricted HA epitope to naive clonotypic CD4
cells.
|
To determine whether APCs that indirectly present (i.e.,
cross-present) parenchymal self-Ag are sufficient to induce CD4 cell
tolerization, allogeneic chimeras were generated by reconstituting
lethally irradiating C3-HAhigh mice backcrossed
onto a H-2b background with NT
H-2d BM, so that only APCs could present the HA
epitope. As shown in Fig. 2A,
11 days post-transfer naive clonotypic CD4 cells adoptively transferred
into C3-HAhigh db chimeras developed a
tolerant phenotype (i.e., diminished proliferative potential to peptide
stimulation). Since the clonotypic CD4 cells were prepared from mice on
an H-2d background, this experiment might have
been complicated by the possibility that they were alloresponsive to
residual host-derived APCs (H-2b). However, this
did not appear to be the case, as clonotypic CD4 cells recovered from
control NT db chimeras maintained a naive CD44 expression pattern
(Fig. 2B). Additionally, 6.5 clonotypic CD4 cells develop
normally on an F1 (H-2bxd)
background (data not shown). Perhaps the skewed 6.5 transgenic TCR
repertoire precludes the development of alloresponsiveness against the
H-2b haplotype. Nonetheless, these results extend
our previous finding that APCs are not only necessary for tolerization
to parenchymal self-Ags (1), but are also sufficient.
|
Naive clonotypic CD4 cells expressing a congenic Thy1.1 marker
were labeled with the fluorescent marker CFSE and adoptively
transferred into the following Thy1.2-expressing recipients:
C3-HAhigh and NT mice that had been infected with
recombinant vaccinia virus (1 x 106 PFU)
expressing either HA (vacc-HA) or an irrelevant Ag (vacc-GH). Since
C3-HAhigh recipients express and present HA
constitutively, vaccinia-infected recipients were inoculated 1 day
earlier so that they would be presenting viral Ag at the time of the
adoptive transfer. Six days postadoptive transfer, donor
Thy1.1+ CD4 cells recovered from the spleens of
both C3-HAhigh (self-HA) and vacc-HA-infected
(viral-HA) recipients had undergone similarly robust proliferative
responses (i.e., seven or more divisions), whereas those recovered from
recipients infected with the control irrelevant vaccinia remained
undivided (Fig. 3A). While
there was a small population of cells in both the self-HA as well as
the viral-HA groups that remained undivided, they were determined to be
nonclonotypic because they did not stain with the anti-clonotypic
mAb 6.5, whereas the vast majority of divided cells were
6.5+ (data not shown). Thus,
Thy1.1+ cells that had diluted CFSE fluorescence
were clonotypic CD4 cells that had encountered HA. Although it was not
possible to resolve more than seven cell divisions in either the
self-HA or the viral-HA recipient groups because CFSE fluorescence had
been diluted to background levels, it appeared that most clonotypic CD4
cells in both groups had returned to a resting state as they were no
longer blasting; the FSC of naive cells before transfer was 365 ±
5 (n = 10), while the respective values for self-HA and
viral-HA on day 6 was 491 ± 11 (n = 5) and
429 ± 5 (n = 8) compared with 739 ± 27
(n = 6) and 806 ± 23 (n = 7) on
day 2 when the clonotypic CD4 cells are actively dividing (refer to
Fig. 6A).
|
|
A
and 6A), as we have previously observed (12),
upon termination of their respective differentiation processes they
exhibit profound functional differences (Fig. 3, B and
C). Self-HA-exposed clonotypic CD4 cells are severely
impaired in their ability to proliferate upon secondary Ag exposure in
comparison with viral-HA-primed counterparts (12).
Furthermore, as measured by intracellular staining, viral-HA-primed
clonotypic CD4 cells efficiently produce both the T cell growth and
differentiation factor IL-2 and the prototypical Th1 effector cytokine
IFN- upon in vitro restimulation. In contrast, self-HA-exposed
counterparts express these cytokines weakly (Fig. 3, B and
C). These differences in cytokine expression are manifested
in the percentages of clonotypic CD4 cells that express each cytokine
as well as the level of cytokine expression within the positively
expressing cells. When the products of these two parameters were
calculated to assess the total levels of cytokine expression (expressed
as arbitrary units), there were 6- and 28-fold differences in the
levels of IL-2 and IFN- expression, respectively, between viral-HA-
and self-HA-exposed clonotypic CD4 cells (Fig. 3C). Thus,
self-HA-exposed clonotypic CD4 cells exhibit a cytokine expression
pattern consistent with a tolerant phenotype, while that of
viral-HA-primed counterparts is consistent with an effector/memory
phenotype. The duration of Ag presentation does not determine tolerogenic vs immunogenic APC activity
To test whether differences in the duration of self-HA vs viral-HA
presentation play a role in programming clonotypic CD4 cell
tolerization vs priming, respectively, we assessed the duration of
viral-HA presentation by infecting NT mice with vacc-HA at various
times before adoptively transferring CFSE-labeled naive clonotypic CD4
cells (Fig. 4). Vacc-HA infections were
performed using either 1 x 106 PFU (as in
Fig. 3) or a higher dose (1 x 107 PFU) that
induces similar levels of effector function (i.e., ability to produce
cytokines; data not shown). Clonotypic CD4 cells transferred into NT
recipients that had been infected with either dose of vacc-HA 1 day
earlier underwent robust proliferative responses (seven or more
divisions at 6 days post-transfer). However, even at the higher
inoculation dose, the proliferative response began to diminish when the
transfer was given 4 days postinoculation and lessened even further at
6 days postinoculation, suggesting that the level of viral-HA
presentation diminishes over the course of 6 days. At the lower
inoculation dose the clonotypic CD4 cell proliferative response was
somewhat variable even at 2 days postinfection (compare days -2, 1,
and 2), suggesting that the duration of viral-HA presentation
diminishes even more quickly. Given that the level of viral-HA
presentation does not appear to remain high for the time required to
induce tolerization to self-HA, it might have been possible that the
transient nature of viral-HA presentation plays a role in conferring
immunogenicity. To test this possibility, NT adoptive transfer
recipients were subjected to multiple vacc-HA inoculations (at 1
x 106 PFU/inoculation) throughout a 6-day
experiment to maintain a high level of viral-HA presentation.
Inoculations were either performed every second day (days -1, 1, 3,
and 5), or daily (days -1 through 5). As shown in Fig. 5A, inoculating every second
day did not affect the ability of clonotypic CD4 cells to express IL-2
and IFN- upon in vitro restimulation (relative to single inoculation
controls). Daily inoculations did not effect IL-2 expression; however,
there was a modest reduction in both the percentage of clonotypic CD4
cells expressing IFN- (25%) and the level of IFN- expression
(30%). Nonetheless, total IFN- expression was still 12-fold greater
than that in the self-HA-tolerized group, indicating that while
extending the duration of viral Ag presentation might have reduced
effector function somewhat, it did not lead to tolerization. To ensure
that the initial vaccinia inoculations did not generate neutralizing
immunity that prevented productive infections from the later
inoculations, CFSE-labeled clonotypic CD4 cells were adoptively
transferred into NT mice that had been inoculated daily with vaccinia
that does not express HA (vacc-GH) from days -1 through 4, then given
a single inoculation of vacc-HA on day 5. If the repetitive vacc-GH
inoculations had induced neutralizing immunity (either innate or
adaptive), the final vacc-HA inoculation would not have been
productive, and the clonotypic CD4 cells not stimulated to proliferate.
To the contrary, this regimen did elicit a clonotypic CD4 cell
proliferative response equivalent to that of controls receiving a
single vacc-HA inoculation (Fig. 5B). This result infers
that repetitive vacc-HA inoculations did not induce neutralizing
immunity during the previous experiments (Fig. 5A), thus
supporting the conclusion that extending the duration of viral-HA
presentation does not induce tolerization.
|
|
, CFSE-labeled clonotypic CD4 cells were
recovered for analysis from the spleens of either
C3-HAhigh or vacc-HA-infected NT recipients only
42 h post-transfer. Although the clonotypic CD4 cells in both
groups had undergone an average of three divisions, they exhibited both
phenotypic as well as functional differences. While self-HA induced a
significant up-regulation in the expression of the high affinity IL-2R
(i.e., CD25), viral-HA induced even high levels. Additionally,
clonotypic CD4 cells that had encountered viral-HA produced much higher
levels of IFN- upon restimulation than those that had encountered
self-HA. IL-2 expression was also higher in viral-HA- relative to
self-HA-exposed clonotypic CD4 cells, although this difference was less
than that observed for IFN-. To assess whether fluctuations in the
level of viral-HA presentation might have played a role in conferring
increased responses in the viral-HA-primed group, we also included a
group that received daily vacc-HA inoculations (days -1, 0, and 1).
Interestingly, this group did exhibit a decrease in the level of CD25
expression relative to the group that received a single vacc-HA
inoculation on day -1; however, CD25 expression was still greater than
that in the self-HA group. Similar to the day 6 analysis (Fig. 5),
repetitive vacc-HA inoculations did not alter IL-2 expression, but did
reduce total IFN- expression by 
50%. Nonetheless, total IFN-
expression was still 23-fold greater than that in the self-HA group
(Fig. 6, A and B). Thus, it appeared that
distinct differentiation programs had been established following only
brief encounters with self-HA vs viral-HA. Irrelevant vaccinia infection does not induce priming to self-Ag
To determine whether non-antigenic immunogenic signals associated
with vaccinia infection can elicit priming to self-HA,
C3-HAhigh mice were infected with the irrelevant
vaccinia vacc-GH (1 x 106 PFU) 1 day before
receiving an adoptive transfer of naive CFSE-labeled clonotypic CD4
cells, and the clonotypic CD4 cell response was analyzed 42 h
post-transfer (Fig. 6). Although the inherent immunogenicity of vacc-GH
is comparable to that of vacc-HA, as evidenced by their abilities to
prime equivalent vaccinia-specific responses from the endogenous T cell
repertoire (data not shown), vacc-GH infection did not elicit an
immunogenic clonotypic CD4 cell response to self-HA: there were only
marginal increases in CD25 expression and the ability to produce
cytokines. At 6 days post-transfer, clonotypic CD4 cells exposed to
self-HA plus vacc-GH also exhibited weak cytokine responses; total
cytokine expression was 2,984 ± 2,073 (IL-2) and 3,889 ±
1,120 (IFN-) arbitrary units (n = 3) compared with
12,089 ± 1,919 (IL-2) and 41,141 ± 4,984 (IFN-)
arbitrary units for vacc-HA-primed cells (n = 7; data
taken from Fig. 3). Additionally, infection of either
C3-HAlow or C3-HAhigh
transgenic recipients with 1 x 107 PFU of
wild-type vaccinia (which is at least 10-fold more virulent than
recombinant vaccinia (16)) also failed to prevent
tolerization (data not shown).
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
F1
C3-HAhigh chimeras reconstituted with NT BM
expressing a nonrestricting haplotype for the HA epitope
(H-2b), adoptively transferred naive clonotypic
CD4 cells remained in a naive state, indicating that indirect
presentation is the exclusive pathway regardless of the level of
parenchymal HA expression. Interestingly, the clonotypic CD4
cells were only able to retain their naive phenotype in
bC3-HAhigh chimeras that had received high
dose irradiation (1300 rad), while in
bC3-HAlow chimeras the clonotypic CD4 cells
remained naive in hosts that had received either 1000 or 1300 rad.
Thus, radioresistant host-derived APCs (H-2bxd)
that can engage naive clonotypic CD4 cells are present in
bC3-HAhigh, but not in
bC3-HAlow, chimeras that have received low
dose irradiation. This observation suggests that when APCs present
greater numbers of MHC-self-peptide complexes, fewer APCs are required
to induce tolerization.
While the above-mentioned experiments demonstrate that HA-expressing
parenchyma in C3-HA transgenics cannot initiate CD4 cell tolerization,
they did not rule out a role for these parenchyma during later stages
of tolerization. When naive CD4 cells receive a tolerizing stimulus in
the form of a bolus injection of soluble peptide, they undergo an
initial expansion phase in secondary lymphoid tissues, followed by the
migration of a significant fraction into nonlymphoid tissues such as
lung, etc. Subsequently, all the CD4 cells develop a tolerant phenotype
(19). Since the C3-HA transgenic mice express HA in the
lung (1, 12), it might have been possible that after
having encountered APCs indirectly presenting parenchymally derived HA
in the spleen or LN, the clonotypic CD4 cells migrated into the lung
(and possibly other HA-expressing tissues) where parenchymal HA
presentation provides a second tolerogenic signal. In db
C3-HAhigh chimeras (in which only APCs were
capable of presenting the HA epitope) naive clonotypic CD4 cells became
fully tolerant. While this result does not necessarily infer that CD4
cells undergoing tolerization cannot interact with parenchyma
expressing cognate self-Ag, if this interaction does take place, it is
not essential for tolerization. Thus, APCs that indirectly present
parenchymal self-Ags are not only necessary for CD4 cell tolerization,
but they are also sufficient.
Both our current as well as previous studies (1) have
indicated that parenchymal self-HA is indirectly presented by tAPCs;
however, we do not know the pathway by which vaccinia-derived HA is
presented to induce priming. Although priming is initiated by APCs
presenting vaccinia-derived HA (20), it is unknown whether
APCs indirectly present HA acquired from vaccinia-infected parenchyma,
if vaccinia-infected APCs present endogenously synthesized HA, or if
both pathways operate. Nonetheless, given that APCs perform the duel
function of inducing both T cell tolerance and immunity, it is
important to understand the parameters that confer these different
functions. As a first step in understanding the relationship between
tAPCs and iAPCs, we simply asked whether they differed functionally.
While functionally distinct APCs might induce either tolerance or
immunity depending on qualitative differences in the signals that they
deliver to cognate naive T cells, it is also possible that functionally
equivalent APCs induce either tolerance or immunity depending upon
extrinsic parameters. While we do not know the relative levels of
self-HA vs viral-HA epitope presentation in our system (although they
induce similar kinetics of naive clonotypic CD4 cell proliferation), it
is unlikely that this parameter determines immunogenicity vs
tolerogenicity. Thus, when naive clonotypic CD4 cells are transferred
into C3-HA transgenic mice that differ in their level of HA expression
by at least 1000-fold, the time course of the responses differ, but the
final outcome (i.e., tolerization) does not (12) (data not
shown). A similar relationship has been observed for CD8 cell
tolerization (21). Along similar lines, varying the dose
of vacc-HA used to prime clonotypic CD4 cells in our system did not
alter functional differentiation (12) (data not shown).
Similarly for CD8 cell responses, priming to Listeria
monocytogenes Ags does not differ qualitatively over a range of
infection doses (22). Although high dose lymphocytic
choriomeningitis virus infection leads to the phenomenon of high
zone tolerance in which cognate CD8 cells undergo accelerated priming,
followed by complete exhaustion of function over a 2-wk period
(10, 23), high zone tolerance appears to be qualitatively
different from the tolerization observed in our system that develops
toward self-Ag, in that the former seems to represent a strong
immunogenic response that cannot sustain itself, while the latter
response is programmed to be tolerogenic from the outset (refer to Fig. 6). In addition to the above-mentioned empirical evidence, it seems
logical that the immune system is programmed to be nonresponsive to
parenchymal self-Ags regardless of their expression levels.
Functionally equivalent APCs could potentially induce either
tolerization or immunity through differential kinetics of Ag
presentation. Thus, naive T cells encountering APCs presenting either
cognate self or pathogen-derived epitopes would initially undergo
priming; however, due to the constitutive expression and presentation
of self-Ags, self-reactive T cells would become overstimulated and
driven into a tolerant state. In the case of pathogen-derived Ags, as
the pathogen is cleared, the level of APC presented Ag would diminish
below the threshold required to induce tolerization. If this clonal
exhaustion model were correct, two predictions could be made. The first
is that at early stages of tolerization and priming, cognate CD4 cells
should exhibit similar responses and should only begin to differ at
later stages when the level of pathogen-derived Ag presentation
declines. Second, extending the duration of pathogen-derived Ag
presentation for the time required to induce tolerization toward
self-Ag should result in tolerization. Contrary to the first
prediction, clear differences were observed between CD4 cells
undergoing tolerization vs priming after only 42 h. Despite having
both undergone an average of three divisions (indicating that these two
responses develop at similar rates) clonotypic CD4 cells that had
encountered viral-HA expressed higher levels of the high affinity IL-2R
(CD25) and were also able to produce higher levels of both IFN- and
IL-2 upon in vitro restimulation than counterparts exposed to self-HA.
Comparable results have been observed between naive CD8 cells
encountering cognate self vs viral Ag (24). The decreased
IL-2R expression and ability to produce IL-2 in CD4 cells undergoing an
early stage of tolerization (relative to primed counterparts) are
consistent with previous studies suggesting that suboptimal IL-2
signaling plays a role in this process (25, 26). Further
investigation, however, will be required to determine whether
suboptimal IL-2 signaling in our system actually contributes to or is
simply a result of tolerization. Nonetheless, the observation that CD4
cells undergoing tolerization vs priming exhibit different responses
soon after initial Ag encounter suggests that naive CD4 cells become
programmed to undergo either tolerogenic or immunogenic differentiation
long before the potential effects of chronic vs transient Ag
presentation come into play. When the duration of viral-HA presentation
was extended for the 6 days required to induce tolerization toward
self-HA, clonotypic CD4 cells were not tolerized, although they did
exhibit a slight reduction in their level of effector function, raising
the possible that extending viral-HA presentation over much longer
periods could eventually lead to tolerization. In fact, chronic
lymphocytic choriomeningitis virus infection does result in
cognate CD4 cell tolerization, but only after 6 wk (11).
Along similar lines, CTL primed by admixture of soluble peptide and a
CD40 agonist become tolerized if peptide presentation persists for
several weeks (27). Taken together, the processes of T
cell priming to pathogen-derived Ags and tolerization to parenchymal
self-Ags appear to be programmed soon after initial Ag encounter, and
while the duration of Ag presentation does not appear play a role in
establishing these basic programs, it might exert an effect over the
long term.
Another mechanism by which functionally equivalent APCs could induce either tolerization or priming would be for pathogens to induce the activity of a third-party cell (i.e., non-APC) that produces an immunogenic factor(s) that acts directly on T cells to redirect them to undergo priming rather than tolerization (tolerance being the default response). Precedence that immunogenic signals can alter naive T cell responses in an APC-independent manner comes from studies demonstrating that IL-1 can enhance the proliferation of naive CD4 cells stimulated by immobilized cognate MHC-peptide complexes (28). Since this third-party model predicts that Ag and the immunogenic signal(s) do not need to be linked, we tested it by providing the immunogenic signals in trans by infecting C3-HAhigh transgenics with a vaccinia that does not express HA. Since this vaccinia expresses all the immunogenic elements required for clonotypic CD4 cell priming (minus HA), the activity of the Ag nonspecific third-party should have been activated, and naive clonotypic CD4 cells encountering APCs presenting self-HA would have been primed rather than tolerized. Tolerization was not converted to priming, however, indicating that self-HA plus irrelevant vaccinia does not equal vacc-HA, a result inconsistent with the third-party model. This result was actually somewhat surprising given that numerous immunogenic agents that are known to enhance APC function can elicit T cell priming to otherwise tolerogenic Ags. These agents include LPS (29, 30, 31), poly(IC) (27, 30), IL-1 (32), IL-12 (33), Flt3 ligand (34), CD40 agonists (35, 36, 37, 38, 39), as well as Ag-unrelated pathogens (30). Nonetheless, our current result is consistent with a previous study that found Ag-irrelevant vaccinia to be comparatively ineffective in eliciting T cell priming to a self-Ag relative to a variety of infectious and noninfectious immunogenic agents (i.e., Ag-irrelevant vesicular stomatitis virus and L. monocytogenes, LPS, and poly(IC)) (30) and suggests that the inability of vaccinia to elicit priming to self-Ag may underlie a distinct mechanism by which its confers immunogenicity upon linked Ags. Thus, while our data do not preclude the possibility that the third-party model operates for other pathogens, this model cannot explain tolerogenicity vs immunogenicity in our system.
In attempting to ascertain the critical parameter(s) that determines tolerogenic vs immunogenic APC function, our studies have argued against models that use functionally similar APCs, thus suggesting that qualitative differences in the signals delivered by APCs are the primary determinants. These differential signals might be purely nonantigenic in the form of costimulatory ligands and/or cytokines. Alternatively, as proposed by Mueller et al. (40), the ability of APCs to prime or tolerize might depend on the ratio of their ability to induce TCR ligation, which is presumed to be tolerogenic, vs their ability to deliver counteracting costimulatory signals.
The inference that tAPCs and iAPCs are functionally distinct does not necessarily imply that they derive from distinct cell lineages. While it might be possible that distinct lineages are inherently tolerogenic or immunogenic (e.g., macrophages might be tolerogenic (41, 42, 43) in contrast to dendritic cells (DCs), whose immunogenic properties are well documented (4), or different subpopulations of DCs might have different activities (44, 45)), the same APC might develop either tolerogenic or immunogenic activity depending upon the environment in which it acquires Ag. Thus, Ag acquisition in the absence of inflammatory signals (as would be the case for self-Ags) would activate a default tolerogenic activity, while APCs exposed to inflammatory signals (i.e., cytokines, LPS, etc.) during Ag acquisition would become immunogenic (46, 47). The possibility that DCs perform this duel function has been supported by two elegant studies demonstrating that Ig-mediated targeting of exogenous Ag to DCs in vivo results in tolerization of cognate T cells, while admixture of inflammatory signals induces immunity (32, 39). Thus, while it is not clear whether the same or different APC populations present parenchymal self and pathogen-derived Ags to elicit T cell tolerization and immunity, respectively, it appears that these divergent processes are induced by functionally distinct APCs.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
2 Address correspondence and reprint requests to Dr. Adam J. Adler, Center for Immunotherapy of Cancer and Infectious Diseases, University of Connecticut Health Center, Farmington, CT 06030-1601. E-mail address: aadler{at}up.uchc.edu
3 Abbreviations used in this paper: BM, bone marrow; DC, dendritic cell; HA, hemagglutinin; iAPC, immunogenic APC; NT, nontransgenic; tAPC, tolerogenic APC.
Received for publication January 22, 2002. Accepted for publication April 1, 2002.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
This article has been cited by other articles:
|
|
 |
|
  H.-R. Yen, T. J. Harris, S. Wada, J. F. Grosso, D. Getnet, M. V. Goldberg, K.-L. Liang, T. C. Bruno, K. J. Pyle, S.-L. Chan, et al. Tc17 CD8 T Cells: Functional Plasticity and Subset Diversity J. Immunol., December 1, 2009; 183(11): 7161 - 7168. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M.-C. St. Rose, H. Z. Qui, S. Bandyopadhyay, M. A. Mihalyo, A. T. Hagymasi, R. B. Clark, and A. J. Adler The E3 Ubiquitin Ligase Cbl-b Regulates Expansion but Not Functional Activity of Self-Reactive CD4 T Cells J. Immunol., October 15, 2009; 183(8): 4975 - 4983. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. Bandyopadhyay, M. Long, H. Z. Qui, A. T. Hagymasi, A. M. Slaiby, M. A. Mihalyo, H. L. Aguila, R. S. Mittler, A. T. Vella, and A. J. Adler Self-Antigen Prevents CD8 T Cell Effector Differentiation by CD134 and CD137 Dual Costimulation J. Immunol., December 1, 2008; 181(11): 7728 - 7737. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. Long, A. M. Slaiby, S. Wu, A. T. Hagymasi, M. A. Mihalyo, S. Bandyopadhyay, A. T. Vella, and A. J. Adler Histone Acetylation at the Ifng Promoter in Tolerized CD4 Cells Is Associated with Increased IFN-{gamma} Expression during Subsequent Immunization to the Same Antigen J. Immunol., November 1, 2007; 179(9): 5669 - 5677. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. T. Hagymasi, A. M. Slaiby, M. A. Mihalyo, H. Z. Qui, D. J. Zammit, L. Lefrancois, and A. J. Adler Steady State Dendritic Cells Present Parenchymal Self-Antigen and Contribute to, but Are Not Essential for, Tolerization of Naive and Th1 Effector CD4 Cells J. Immunol., August 1, 2007; 179(3): 1524 - 1531. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. Long and A. J. Adler Cutting Edge: Paracrine, but Not Autocrine, IL-2 Signaling Is Sustained during Early Antiviral CD4 T Cell Response J. Immunol., October 1, 2006; 177(7): 4257 - 4261. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. Long, A. M. Slaiby, A. T. Hagymasi, M. A. Mihalyo, A. C. Lichtler, S. L. Reiner, and A. J. Adler T-bet Down-Modulation in Tolerized Th1 Effector CD4 Cells Confers a TCR-Distal Signaling Defect That Selectively Impairs IFN-{gamma} Expression J. Immunol., January 15, 2006; 176(2): 1036 - 1045. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
L. Steinaa, P. B. Rasmussen, A. M. Wegener, L. Sonderbye, D. R. Leach, J. Rygaard, S. Mouritsen, and A. M. Gautam Linked Foreign T-Cell Help Activates Self-Reactive CTL and Inhibits Tumor Growth J. Immunol., July 1, 2005; 175(1): 329 - 334. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. A. Hurchla, J. R. Sedy, M. Gavrielli, C. G. Drake, T. L. Murphy, and K. M. Murphy B and T Lymphocyte Attenuator Exhibits Structural and Expression Polymorphisms and Is Highly Induced in Anergic CD4+ T Cells J. Immunol., March 15, 2005; 174(6): 3377 - 3385. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. K. Lathrop, C. A. Huddleston, P. A. Dullforce, M. J. Montfort, A. D. Weinberg, and D. C. Parker A Signal through OX40 (CD134) Allows Anergic, Autoreactive T Cells to Acquire Effector Cell Functions J. Immunol., June 1, 2004; 172(11): 6735 - 6743. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. D. H. Doody, J. T. Kovalchin, M. A. Mihalyo, A. T. Hagymasi, C. G. Drake, and A. J. Adler Glycoprotein 96 Can Chaperone Both MHC Class I- and Class II-Restricted Epitopes for In Vivo Presentation, but Selectively Primes CD8+ T Cell Effector Function J. Immunol., May 15, 2004; 172(10): 6087 - 6092. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. A. Mihalyo, A. D. H. Doody, J. P. McAleer, E. C. Nowak, M. Long, Y. Yang, and A. J. Adler In Vivo Cyclophosphamide and IL-2 Treatment Impedes Self-Antigen-Induced Effector CD4 Cell Tolerization: Implications for Adoptive Immunotherapy J. Immunol., May 1, 2004; 172(9): 5338 - 5345. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C.-T. Huang, D. L. Huso, Z. Lu, T. Wang, G. Zhou, E. P. Kennedy, C. G. Drake, D. J. Morgan, L. A. Sherman, A. D. Higgins, et al. CD4+ T Cells Pass Through an Effector Phase During the Process of In Vivo Tolerance Induction J. Immunol., April 15, 2003; 170(8): 3945 - 3953. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. D. Higgins, M. A. Mihalyo, and A. J. Adler Effector CD4 Cells Are Tolerized Upon Exposure to Parenchymal Self-Antigen J. Immunol., October 1, 2002; 169(7): 3622 - 3629. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
