| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Departments of
*
Medicine and
Microbiology and Center for Immunology, University of Minnesota Medical School, Minneapolis, MN 55455
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The molecular basis for these Ag/APC-induced alterations in T cell
clonal frequency and function in vivo is only incompletely understood.
In addition to TCR occupancy, the costimulation of
CD4+ T cells by APC-derived B7 ligands (CD80 and
CD86) through CD28 receptors appears essential for Ag-stimulated clonal
expansion (5). The production of IL-2 and signaling by
IL-2Rs during the Ag encounter may be one consequence of this
costimulation that can act to promote cell cycle progression (6, 7). However, other T cell growth factors and receptors are
probably also important, since IL-2-/- and
IL-2R
-/- T cells can proliferate in vivo
(8, 9). In contrast to CD28, another B7-binding protein
called CTLA-4 (CD152) is induced on T cells during the course of the
primary Ag challenge, and when it is ligated, this molecule acts to
limit the extent of the clonal expansion (5, 10, 11).
Furthermore, CTLA-4 signals appear necessary for the induction of
clonal anergy (12, 13, 14).
The mechanism by which CTLA-4 signaling promotes anergy induction is unknown. CTLA-4 ligation can limit the production of IL-2, the expression of IL-2Rs, and entry of T cells into cell cycle (14, 15, 16). Since the avoidance and/or reversal of T cell clonal anergy have been shown in several in vitro systems to be associated with IL-2R signaling and proliferation (17, 18, 19), the capacity of CTLA-4 to place a brake on cell cycle progression may explain its importance to anergy development. Nevertheless, a more straightforward model may be that CTLA-4 signals synergize with the TCR to directly induce a putative anergy gene product.
Given the correlation observed in vivo between effective T cell clonal expansion in the primary Ag response and the apparent avoidance of clonal anergy, we questioned whether cell cycle progression per se plays a direct role in the regulation of Ag responsiveness. Specifically, we examined the hypothesis that a biochemical signaling event that occurs when naive T cells are induced to progress through cell cycle leads to a reversal of the clonal anergy state. Such a model predicts that the anergy state is induced whenever the TCR is productively engaged, and that CTLA-4 molecules only indirectly promote the maintenance of tolerance through an inhibition of cell cycle progression. To control the rate of cell cycle progression without influencing TCR, CD28, or CTLA-4 ligation, we have made use of the T cell immunosuppressant rapamycin (a specific inhibitor of the cell cycle regulator called mammalian target of rapamycin (mTOR)3; Ref. 20), and monitored its effects with the cell division-sensitive fluorescence-tracking dye CFSE. Our results suggest that in CD4+ T cells undergoing a primary Ag challenge, an mTOR-dependent signaling pathway leading to cell cycle progression is primarily responsible for establishing the level of recall Ag responsiveness within daughter cells. Furthermore, our experiments indicate that CTLA-4 signals are not necessary for clonal anergy induction in vivo if cell cycle progression is controlled with rapamycin. Thus, CTLA-4 acts indirectly to promote the development of clonal anergy through its inhibition of cell cycle progression.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The DO11.10 TCR-transgenic mice (21) were bred to homozygosity and maintained in our animal facility, as previously described (22). CD4+ T cells in these mice are uniformly reactive to chicken OVA peptide 323–339 (OVAp)/I-Ad complexes and express a clonotypic TCR detectable with the KJ1-26 mAb (23). BALB/c mice, 6–8 wk old, were purchased from Charles River Breeding Laboratories (Wilmington, MA) through a contract with the National Cancer Institute at the National Institutes of Health (Frederick, MD). All mice were housed under specific pathogen-free conditions and used in accordance with National Institutes of Health guidelines and the University of Minnesota Institutional Animal Care and Use Committee. Mice were age and sex matched for all experiments. For some experiments, DO11.10 mice were backcrossed to BALB/c RAG-2-/- (Taconic Farms, Germantown, NY) to generate TCR-transgenic mice that lacked endogenous TCR gene rearrangements.
Abs and staining reagents
The following Abs and fluorochrome conjugates were purchased from BD PharMingen (San Diego, CA): CyChrome-labeled anti-CD4, allophycocyanin-labeled anti-CD4, PE-labeled anti-IL-2, PE-labeled irrelevant IgG2, PE-labeled anti-CD69, PE-labeled anti-CD25 mAb, biotin-labeled anti-mouse IgG1, FITC-labeled streptavidin, and CyChrome-labeled streptavidin. FITC-conjugated KJ1-26 mAb, biotin-labeled anti-rabbit IgG, and PE-labeled streptavidin were obtained from Caltag Laboratories (South San Francisco, CA). Anti-phospho c-Jun Ab (KM1) was purchased from Santa Cruz Biotechnology (Santa Cruz, CA). Anti-phospho p38 polyclonal Ab was purchased from New England Biolabs (Beverly, MA). Anti-FcR mAb 2.4G2 (American Type Culture Collection, Manassas, VA) was used as a 10% spent hybridoma cell culture supernatant. FCS, rat serum, formaldehyde, and saponin were obtained from Sigma-Aldrich (St. Louis, MO).
Adoptive transfer and in vivo treatments of mice
Lymph node and spleen cells containing 2.5–5 x 106 KJ1-26+ CD4+ cells from DO11.10 or DO11.10 RAG-2-/- TCR-transgenic mice were prepared for adoptive transfer (4), as described previously (22), and injected into the tail vein of recipient mice. In most experiments, cells were labeled with CFSE (Molecular Probes, Eugene, OR) before transfer, using a modification of a technique previously described (24). Briefly, lymph node and spleen cells at a concentration of 1 x 107 cells/ml were incubated in 2.5 µM CFSE for 5 min at 37°C. The labeling reaction was stopped with the addition of a 1:1 mixture of Eagle’s Hanks’ amino acids medium (Biofluids, Rockville, MD) and RPMI 1640 (Life Technologies, Grand Island, NY) containing 10% FCS, 2 mM L-glutamine, 100 U/ml penicillin, 100 U/ml streptomycin, and 5 x 10-5 M 2-ME. The CFSE-labeled T cells were washed twice in PBS before adoptive transfer. OVAp323–339 was produced in our microchemical facility and dissolved in PBS (either 100 or 250 µg), and then delivered systemically to mice by i.v. injection. The hamster anti-mouse CTLA-4 mAb 4F10 (10) was purified from hybridoma cell culture supernatants using standard protein A-Sepharose chromatography techniques. Anti-CTLA-4 mAb or control hamster IgG were injected i.v. into mice at the time of Ag administration (250 µg/injection). Rapamycin was obtained from S. N. Sehgal (Wyeth-Ayerst Research, Princeton, NJ). A stock solution of 1 mg/ml rapamycin in 100% ethanol was prepared. Rapamycin was then suspended in 0.2% carboxymethylcellulose (CMC), as previously described (25), and delivered i.p. at a dose of 0.5–0.75 mg/kg per day for 4–8 days beginning on the day of T cell adoptive transfer.
Detection of surface marker expression and intracellular cytokines in DO11.10 T cells
Adoptive transfer recipients containing naive DO11.10 T cells were challenged with an i.v. injection of OVAp (100 µg) and sacrificed at various later times. To detect cell surface activation markers, lymph node and spleen cells were washed with staining buffer (PBS containing 2% FCS and 0.2% azide), and then incubated with anti-Fc mAb 2.4G2 and 10% rat serum to block FcR. Cells were then stained with the combination of CyChrome-labeled anti-CD4 mAb, biotinylated KJ1-26 mAb, FITC-labeled streptavidin, and PE-labeled anti-CD25 mAb. Alternatively, cells were stained with a mixture of allophycocyanin-labeled anti-CD4 mAb, biotinylated KJ1-26 mAb, CyChrome-labeled streptavidin, and PE-labeled anti-CD69 mAb. Intracellular IL-2 was detected essentially as previously described (8, 22). Briefly, washed and Fc-blocked lymph node or spleen cells were first incubated with allophycocyanin-labeled anti-CD4 mAb, biotinylated KJ1-26 mAb, and CyChrome-labeled streptavidin. Cells were then washed one time in PBS, fixed in 2% formaldehyde for 20 min at room temperature, permeabilized in 0.5% saponin, and incubated at room temperature for 30 min with PE-labeled anti-IL-2 mAb or PE-labeled irrelevant IgG2a. Finally, cells were washed twice in 0.5% saponin and once in staining buffer. For all experiments, more than 1000 KJ1-26+CD4+ events were collected using the FACSCalibur flow cytometer (BD Biosciences, Mountain View, CA) and analyzed using CellQuest (BD Biosciences) and FlowJo (Tree Star, San Carlos, CA) software.
Detection of intracellular phosphoproteins in DO11.10 T cells
Phosphorylated c-Jun and p38 were measured by flow cytometry, as previously described (26). In some experiments, spleens were removed and fixed immediately by preparing a single-cell suspension in PBS containing 2% formaldehyde. Following a 20-min incubation at room temperature in the fixative, cells were washed twice in PBS and incubated for 15 min in 1% mouse serum, 1% rat serum, and 10% anti-FcR mAb 2.4G2 culture supernatant. Cells were then stained with anti-CD4 CyChrome and FITC-labeled KJ1-26 mAb in staining buffer for 15 min at room temperature. Following a wash in staining buffer, cells were permeabilized with two washes in staining buffer containing 0.5% saponin and incubated for 30 min with PE-labeled anti-IL-2 mAb, PE-labeled anti-CD69, anti-phospho c-Jun mAb KM1, or anti-phospho p38 polyclonal Ab. Anti-IL-2- and anti-CD69-stained cells were then washed sequentially with saponin buffer, PBS, and staining buffer. Anti-phospho c-jun or anti-phospho p38-stained cells were washed in saponin buffer and then incubated for 30 min with biotin-labeled anti-mouse IgG1 or biotin-labeled anti-rabbit IgG Abs, respectively, washed, and then incubated for 30 min with PE-labeled strepavidin. Cells were then washed, as described above, before analysis. The fluorescence intensities of more than 1000 KJ1-26+ CD4+ events were collected using the FACSCalibur flow cytometer and analyzed using CellQuest (BD Biosciences) and FlowJo (Tree Star) software.
Measurement of cell cycle progression in vivo
Cell cycle progression was monitored using the CFSE dye as a
marker of cell division (24). FL1 fluorescence was
measured in the KJ1-26+CD4+
T cells at various times after stimulation, and cells were found to
demonstrate a 2-fold decrease in FL1 fluorescence with each successive
round of cell division. Based on these peaks of CFSE fluorescence
intensity within the population, each T cell was assigned to a
particular cell division group d (with d = 0
to n cell divisions), and the number of T cell events
(E) observed within each cell division group
(Ed) was determined. To calculate the
average number of cell divisions within a population, the following
equation was used:
 |
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
A single i.v. bolus of OVAp323–339 was
previously shown to induce a significant clonal expansion of adoptively
transferred OVAp-specific
KJ1-26+CD4+ T cells within
3 days (4). Thereafter, the number of OVAp-reactive T
cells fell to near baseline (presumably as a consequence of Ag
clearance and apoptosis of the lymphocytes), and the remaining T cells
demonstrated a persistent hyporesponsiveness to Ag rechallenge
consistent with the induction of clonal anergy (27). To
further investigate this activation event and the tolerance elicited by
it, we studied cell cycle progression in relation to other activation
events following the i.v. administration of OVAp. Fig. 1
A shows that within 3 min of
Ag infusion, the stress-activated protein kinase (SAPK) p38 becomes
phosphorylated in virtually all of the
KJ1-26+CD4+ T cells, as
previously reported (26). In contrast, host T cells were
generally unaffected by the OVAp infusion and preparation of the cells
for flow cytometry. Over the course of the next 5 h of
stimulation, the OVA-specific T cells uniformly expressed increasing
levels of phosphorylated c-Jun, consistent with the actions of another
SAPK called c-Jun N-terminal kinase (Fig. 1B, and data not
shown). With i.v. peptide stimulation,
KJ1-26+CD4+ T cells also
demonstrated a unimodal increase in both CD25 and CD69 surface marker
expression. Consistent with our previous observations (8, 22), however, IL-2 accumulated in the cytoplasm of Ag-stimulated
T cells in a heterogenous fashion. At no time after Ag exposure were
more than about half of the splenic T cells positive for intracellular
IL-2 (Fig. 1C, and data not shown). Thus, infusion of
aqueous peptide Ag into the bloodstream resulted in a remarkably rapid
and synchronous activation of virtually all CD4+
T cells bearing an Ag-reactive TCR. Nevertheless, some of these
activated T cells failed to synthesize significant amounts of
IL-2.
|
C, and data not shown). Likewise,
OVA-specific T cells in PBS control animals demonstrated no evidence of
CFSE dye dilution. However,
KJ1-26+CD4+ T cells from
OVAp-treated animals became uniformly enlarged, and CFSE fluorescence
was reduced in a pattern, indicating that cell divisions had occurred.
Like the synthesis of the autocrine growth factor IL-2 many hours
earlier, proliferation in response to the OVAp challenge appeared to be
heterogenous, with some cells dividing only once and others dividing as
many as four times during the 3-day period. Transgenic TCR expression
did not correlate with either the level of IL-2 accumulation or degree
of CFSE dye dilution within individual cells (data not shown).
Furthermore, similar results were obtained using donor T cells from
DO11.10 RAG-2-/- (Fig. 1D, and data
not shown). Therefore, the failure of some of the activated T cells to
produce IL-2 and proliferate was not the result of a heterogeneity in
the level of TCR transgene expression because of endogenously
rearranged TCR -chains, or because of previous environmental Ag
recognition. CD4+ T cells that proliferate only poorly to primary Ag stimulation are unresponsive to Ag rechallenge
As described above, a number of in vitro studies have suggested
that TCR stimulation that is unaccompanied by proliferation leads to
the development of clonal anergy. To determine whether a similar
relationship exists in vivo between proliferation in the primary
stimulation and recall Ag responsiveness, we rechallenged the
KJ1-26+ T cells with Ag and measured their
capacity for IL-2 production as a function of cell cycle history using
the CFSE fluorescence as a marker. CFSE-labeled DO11.10 T cells
were adoptively transferred into normal
BALB/c mice, and then given a single
infusion of OVAp. Three and 7 days after primary challenge
(Figs. 2 and 3, respectively, as well as Table I), the animals received a second
injection of OVAp i.v., and 3 h later the cells were recovered and
examined for intracellular IL-2 accumulation. Recall IL-2 production in
the bulk KJ1-26+ T cell population was always
higher at the 7-day time point after primary Ag challenge, suggesting
that the IL-2 gene in day 3-activated T cells is relatively
refractory to stimulation (Figs. 2 and 3, and data not shown). However,
the single i.v. infusion of OVAp in PBS was not found to effectively
induce clonal anergy in the bulk population of
KJ1-26+ T cells, as measured by the accumulation
of intracellular IL-2 during a 3-h period following secondary Ag
infusion. This is in fact consistent with our previous observation that
multiple OVAp infusions are necessary to achieve a significant level of
tolerance to Ag restimulation in vivo (22). It should also
be noted that the level of tolerance in the bulk population may be
underestimated in this study because IL-2 synthesis was measured in
these Ag-experienced T cells 3 h after reexposure to Ag, whereas
naive T cell IL-2 accumulation peaks only after 5 h of stimulation
(26). When compared with KJ1-26+ T
cells primed with OVAp plus LPS, these T cells exposed to OVAp alone
consistently demonstrated a deficit in IL-2 production (data not
shown).
|
|
|
and 3). This observation of low recall
IL-2 inducibility in lymphocytes that had experienced only weak cell
cycle progression during the primary Ag stimulation indicated the
possibility of a causal relationship between the primary and secondary
responses. KJ1-26+ T cells may be an
intrinsically heterogenous population with regard to their capacity to
produce IL-2 and proliferate, and poor responsiveness in the primary
immunization may simply predict a poor capacity to respond in the
rechallenge. Alternatively, T cells stimulated to vigorously progress
through the cell cycle in the primary response may maintain or increase
their capacity to later respond on Ag rechallenge with the production
of IL-2. It should be noted that the defect in IL-2 gene
expression observed in T cells that had undergone little previous cell
cycle progression was not a consequence of some intrinsic inability of
the cells to recognize OVAp-MHC complexes. Poorly dividing T cells
underwent a size enlargement by day 3 that was indistinguishable from
the majority of the Ag responders (Fig. 2). Furthermore, CD69
expression could still be markedly up-regulated in such T cells during
Ag rechallenge (Fig. 2, and unstimulated control data not shown). Thus,
our data again suggested that TCR occupancy after i.v. peptide Ag
infusion is relatively uniform for all Ag-reactive T cells.
Nevertheless, IL-2 production and cell cycle progression are
heterogenous, with responsiveness in the primary and secondary Ag
stimulations highly correlated. Rapamycin inhibits cell cycle progression in vivo and promotes the development of clonal anergy
This observation of a direct correlation between cell cycle progression in the primary Ag stimulation and the level of IL-2 gene responsiveness in the secondary Ag challenge was reminiscent of data obtained using cloned Th1 cells in vitro. Th1 cells stimulated with Ag and normal APC lose their capacity to make IL-2, if their cell cycle progression during the incubation is prevented by the addition of neutralizing Ab to IL-2 and CD25 (17). Furthermore, anergic Th1 cells regain their capacity to produce IL-2 when driven to proliferate with exogenous growth factor (18). Therefore, we speculated that the poorly dividing OVA-reactive T cells were induced into clonal anergy in vivo as a consequence of TCR occupancy that was unaccompanied by vigorous growth factor-dependent cell cycle progression. However, it remained possible that a significant fraction of KJ1-26+ T cells is intrinsically deficient in their ability to secrete IL-2 and, as a result, they failed to proliferate well during the primary challenge. To experimentally address these two possibilities, we made use of the T cell antiproliferative agent rapamycin (20). If cell cycle progression does not in fact positively regulate recall Ag responsiveness, then a reduction of T cell proliferation by inhibiting mTOR during the primary response should not alter IL-2 production in the rechallenge.
Treatment of mice with rapamycin during the time of the primary i.v.
OVAp administration significantly inhibited cell cycle progression by
the OVA-specific T cells (Fig. 4 and
Table I). By day 3 of the OVAp-induced clonal expansion,
KJ1-26+CD4+ T cells were
reduced in frequency in the lymph nodes and spleen 61 ± 15% as a
result of the rapamycin treatment, and this result was statistically
significant (p < 0.01; n = 9).
Consistent with this, the mean number of KJ1-26+
cell divisions observed after the OVAp infusion (based on CFSE dye
dilution) was decreased after rapamycin treatment. In contrast,
rapamycin only modestly inhibited T cell size enlargement, suggesting
that TCR signaling was still effective even in the presence of the drug
(Fig. 4B). In addition to its effects on cell cycle
progression during the primary OVAp response, rapamycin treatment
clearly also led to an overall decrease in recall IL-2 production in
the Ag-primed KJ1-26+ T cells (Fig. 4A
and Table I). Importantly, even in the presence of rapamycin,
intracellular IL-2 accumulation following Ag rechallenge remained
directly proportional to the number of cell divisions achieved during
the primary response. Thus, Ag stimulation in the presence of rapamycin
(and unaccompanied by a vigorous cell cycle progression) induced even
the most highly responsive T cells to become more anergic.
|
A). This loss of Ag responsiveness was
not the result of rapamycin directly inhibiting intracellular IL-2
accumulation, as rapamycin had no effect on signaling to the
IL-2 gene in naive animals responding to OVAp (Table I).
Rather, the hyporesponsiveness observed in animals treated with
rapamycin depended on an infusion of OVAp 3 to 7 days earlier. Finally,
there was no evidence that rapamycin was specifically toxic to
Ag-experienced cells. OVAp/rapamycin-pretreated
KJ1-26+ T cells remained fully competent to
express the activation marker CD69 during the Ag rechallenge (Fig. 4C, and control unstimulated cell data not shown). Thus,
this experiment indicates that a vigorous cell cycle progression is
required during the primary Ag response to sustain a high level of
responsiveness for a later Ag challenge. CTLA-4 signaling promotes clonal anergy induction through its inhibitory effects on cell cycle progression
The CTLA-4 inhibitory receptor acts to limit T cell responsiveness
in vivo (5). This may result in part from its capacity to
compete effectively with CD28 for costimulatory B7 molecules; CTLA-4
signaling may also directly inhibit cell cycle progression (16, 28). Interestingly, neutralizing mAb against CTLA-4 has also
been shown to prevent the induction of clonal anergy in vivo
(12). Furthermore, CTLA-4-/- T
cells are resistant to tolerance induction (14). This
could suggest that signals from the CTLA-4 molecule are directly
responsible for inducing the anergy state. An alternative hypothesis is
that CTLA-4 signals affect the level of recall Ag responsiveness by
controlling the intensity of cell cycle progression during the primary
immune response. Donor DO11.10 OVA-specific T cells were found to
undergo more rounds of cell division when challenged with OVAp in the
presence of the anti-CTLA-4 mAb 4F10 (Fig. 5 and Table I). T cell clonal expansion
in the Ag-stimulated 4F10-treated animals was similarly increased.
Therefore, the CTLA-4 molecule was indeed acting as a brake on cell
cycle progression in our model system, and the 4F10 mAb was an
effective antagonist of its function. Consistent with Perez et al.
(12), treatment of animals with the anti-CTLA-4 mAb
during Ag stimulation also led to an enhanced recall Ag responsiveness
at the level of the IL-2 gene in the OVAp-reactive
KJ1-26+CD4+ T cell
population (Table I).
|
). The typical
T cell (mode of six cell divisions) found in anti-CTLA4 mAb-treated
mice on day 7 demonstrated the same level of recall Ag responsiveness
as a T cell from a control animal that had divided to a similar extent
during the primary Ag challenge. Likewise, the more rare T cell that
had undergone only four rounds of cell division in 4F10-treated mice
produced the same amount of IL-2 on rechallenge as similarly divided T
cells from animals treated with OVAp alone. This result suggested that
CTLA-4 signals promote clonal anergy induction by limiting cell cycle
progression. In the presence of the 4F10 mAb, recall Ag responsiveness
is improved, but only in those T cells that have undergone additional
rounds of cell division.
|
Is the relationship that exists between enhanced Ag-stimulated
cell cycle progression and the avoidance of clonal anergy in
anti-CTLA-4 mAb-treated animals coincidental or causal? Once again,
we examined the effect of rapamycin on the response. Rapamycin was
found to effectively block cell cycle progression induced by OVAp
treatment even in the presence of 4F10 mAb (Table I and Fig. 6). Clonal
expansion measured at day 7, however, was not as dramatically affected
by the addition of rapamycin as would be expected based on cell
division history alone (Table I). This perhaps suggests that the
inhibition of the mTOR signaling molecule with rapamycin promotes T
cell survival during the clonal expansion. Despite this enhanced
survival of T cells responding to the OVAp in the presence of
anti-CTLA-4 mAb and rapamycin, the Ag responsiveness of the
survivors was significantly reduced. In fact, T cells from OVAp plus
4F10/rapamycin-treated animals accumulated less intracellular IL-2 on
Ag rechallenge than did T cells from animals treated with OVAp alone.
When this recall IL-2 production was examined as a function of cell
division history, Ag-experienced KJ1-26+ T cells
from anti-CTLA-4 mAb and rapamycin-treated animals maintained the
strict relationship between cell division history and IL-2 inducibility
that has been observed throughout these experiments. Thus, signaling by
the CTLA-4 molecule during primary Ag challenge does not directly
affect recall IL-2 responsiveness. Rather, CTLA-4 signals promote
anergy induction by limiting Ag-stimulated and mTOR-dependent cell
cycle progression. Furthermore, rapamycin can promote the development
of clonal anergy even in the absence of CTLA-4 signals because of its
capacity to directly block cell cycle progression.
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The rapid activation of the SAPK was found to be closely paralleled by a profound change in T cell phenotype. Specifically, surface expression of the CD69 and CD25 molecules was uniformly and highly induced upon exposure to the soluble Ag, and all of the T cells enlarged in size. Although the functional importance of these activation markers has not been formally tested in this study, one can speculate that a rapid increase in CD25 surface molecule expression on all of the Ag-specific T cells would contribute to the development of competence to proliferate in the presence of IL-2. Therefore, cell cycle progression in response to this TCR occupancy may be limited more by growth factor production (or availability) than by growth factor receptor expression. We have in fact observed that only a fraction of the Ag-stimulated KJ1-26+ T cell population is capable of producing high levels of IL-2 during the primary Ag challenge. Furthermore, this heterogeneity in the production of IL-2 is accompanied by a highly variable pattern of cell cycle progression, with some cells dividing as often as seven times during the course of the response, and others in the same population dividing only once.
Our experiments cannot determine whether those T cells accumulating the highest levels of intracellular IL-2 early in the primary response are also the ones that undergo the highest number of cell divisions. However, when taken together with several previous in vitro studies, this would seem to be a logical extension. Weaver and colleagues (29) examined IL-2 gene responsiveness using DO11.10 double-transgenic T cells that also carry a green fluorescence protein (GFP)-based reporter gene containing DNA sequences from the 5' IL-2 gene promoter/enhancer. Coordinate expression of both GFP protein and IL-2 was induced in a heterogenous pattern by in vitro challenge with syngeneic APC and OVAp. Interestingly, GFP+ cells sorted at the end of the experiment were found to be competent to produce IL-2 on rechallenge, whereas GFP- T cells were eventually found to be anergic to further stimulation. Wells et al. (30) have shown that, like the in vivo activation model presented in this study, recall IL-2 production following in vitro anti-CD3 mAb-induced proliferation is heterogenous and enriched in those T cells that have previously undergone the highest number of cell divisions. Therefore, the results of these experiments predict that in vivo those T cells induced to secrete the highest amounts of IL-2 during the primary challenge undergo the most vigorous cell cycle progression and retain the highest level of IL-2 gene inducibility upon rechallenge.
The contrast between the unimodal increase in forward scatter, CD69, and CD25, and the variable extent to which the naive cells could be induced to produce IL-2 and proliferate during the primary challenge supports the notion that growth factor production and cell cycle progression are more stringently regulated in naive CD4+ T cells than other phenotypic changes. The nature of this regulation, however, remains uncertain. It is possible that within a population of naive Ag-specific T cells, some cells will simply lack the capacity to secrete IL-2 and proliferate in response to effective TCR ligation. Alternatively, the activation of naive Ag-specific T cells may not be as uniform as the SAPK activation and surface marker expression data would lead one to believe. Such activation markers may have a relatively low threshold for TCR serial engagements, while IL-2 production and cell cycle progression may require more intense or prolonged TCR signaling events to reach threshold in naive T cells (31, 32). It should be noted that simply increasing the dose of OVAp in the primary Ag infusion does not eliminate the heterogeneity in IL-2 production observed (data not shown). The hypothesis we favor is that costimulatory ligands such as CD80 and/or CD86 are available in excess to only a fraction of the T cells during the primary stimulation. Although these costimulatory ligands can be detected in the secondary lymphoid tissues of these mice at the time of the Ag challenge, the levels remain low relative to CD80 and CD86 expression that is induced with adjuvants such as LPS (33). Consistent with this hypothesis, the CD28 antagonist CTLA-4Ig, while having no effect on SAPK activation in naive CD4+ T cells (26), is a potent inhibitor of IL-2 production and clonal expansion in vivo (5, 8). Likewise, T cells from CD28-/- mice demonstrate defective IL-2 production and proliferation, and yet are capable of normal increases in forward scatter following Ag stimulation (8). Regardless of the underlying nature of this heterogeneity in IL-2 production and cell cycle progression, our data strongly indicate that after a single bolus i.v. infusion of aqueous OVAp, each of the OVA-reactive T cells experiences a stimulation of its TCR.
In the peripheral immune system, exposure of CD4+
T cells to aqueous peptide Ag in the absence of infection or adjuvant
leads to an abortive clonal expansion of the Ag responders
(4). The end result is death for the majority of the
expanded T cells and reduced Ag responsiveness within the bulk of the
survivors. As a consequence, T cells with that particular Ag
specificity are left at only low frequency and in a defective state in
which Ag-dependent autocrine growth is ineffective (27).
Such a mechanism may normally be important to limit further T cell
aggression directed against peripheral tissues that persistently
express peptide Ag (34). In vitro examinations of T cell
activation events have lent insight into the nature of this Ag-induced
tolerance. Intense, continuous stimulation of Th1-like cloned
CD4+ T cells in the absence of APC using
immobilized anti-TCR or anti-CD3 mAbs leads to a suboptimal
proliferative response and increased rate of apoptosis
(35). In addition, surviving T cells demonstrate a
long-lived unresponsiveness to further Ag stimulation characterized by
defective IL-2 production and proliferation. This unresponsiveness can
last indefinitely; however, exposure to rIL-2 induces cell cycle
progression in the anergic cells and causes a reversal of the tolerant
state (18). The notion that IL-2Rs and downstream
biochemical signaling cascades play an important role in regulating the
level of Ag responsiveness is also supported by the observation that
stimulation of cloned CD4+ Th1 cells with Ag plus
normal APC in the presence of neutralizing IL-2 and IL-2R mAbs leads to
clonal anergy (17). Furthermore, ligation of the IL-2R
common 
-chain has been shown to prevent the development of clonal
anergy (19). Taken together, the results suggest that TCR
stimulation unaccompanied by IL-2 production or IL-2R engagement
promotes clonal anergy induction, whereas TCR occupancy that leads to
IL-2R signaling promotes continued Ag responsiveness.
This regulation of clonal anergy by growth factors and their receptors
indicates that cell cycle progression may itself be the important
cellular event that controls Ag responsiveness. The role of cell cycle
progression in the regulation of Ag responsiveness has previously been
examined in vitro. The antiproliferative cytokine TGF-
1 has been
shown to cause the development of clonal anergy in naive
CD4+ T cells undergoing a primary response to Ag
and APC (36). Likewise, Gilbert and Weigle
(37) have demonstrated that the G1a
blocker n-butyrate, a histone deacetylase inhibitor, can
cause the development of clonal anergy in Th1 cells when present during
stimulation with Ag and normal APC. In contrast, drugs that blocked Th1
cell cycle progression in G0,
G1b, or S-G2 phases did not
reduce their Ag responsiveness. Powell et al. (38) later
showed that Th1 cells responding to Ag and APC would develop an anergic
phenotype when treated with the G1 phase
inhibitor rapamycin, but not the S phase inhibitor hydroxyurea.
Rapamycin forms a molecular complex with the FK506-binding protein in T cells (39), and inhibits the activity of the serine/threonine protein kinase called mTOR (20, 40). One function of mTOR activity during G1 phase is thought to be the down-regulation of the cyclin-dependent kinase-negative regulator p27Kip1. High constitutive levels of p27Kip1 prevent G1 to S phase progression by binding to and inhibiting the activity of cyclin E-cyclin-dependent kinase 2 complexes. Rapamycin prevents the degradation of p27Kip1 that normally occurs in response to growth factor receptor signals (41). This maintenance of high p27Kip1 levels in rapamycin-treated T cells may be responsible for both the block in cell cycle progression observed, as well as the development of Ag unresponsiveness. Boussiotis et al. (42) have in fact shown in vitro that p27Kip1 levels rise in T cells during the induction of clonal anergy, and the accumulation of this repressor protein can lead to a defect in IL-2 production as a result of its capacity to bind the IL-2 gene transcriptional coactivator JAB1 (43). Whether increased p27Kip1 levels can account for all of the defects in signaling to the IL-2 gene that have previously been reported in anergic T cells remains to be seen. Regardless, the relationship between cell cycle progression and the elimination of one IL-2 gene repressor factor is a paradigm worthy of further investigation.
We now speculate that Ag recognition in vivo is always accompanied by a TCR-mediated signal to induce clonal anergy. This model requires that, like the expression of CD69 and CD25, the induction of clonal anergy is a relatively low-threshold event. Nevertheless, only some of the T cells will develop the capacity to produce autocrine growth factor when they encounter Ag in a B7-rich microenvironment, and it is these cells that efficiently progress through cell cycle and retain the capacity to respond vigorously to a secondary Ag challenge. Biochemical events occurring within the G1 phase of the primary response as a consequence of this growth factor signaling (e.g., activation of mTOR and the degradation of p27Kip1) not only promote further cell cycle progression, but also restore the Ag responsiveness of the T cells by eliminating a molecular repressor of IL-2 gene transcription. Although the CTLA-4 molecule is an important modulator of this process as a result of its capacity to limit cell cycle progression (12, 14) (data shown in this study), we do not imagine that it mediates those signals from the TCR that directly induce the putative clonal anergy repressor molecule (44). The ability of CTLA-4-/- T cells to develop clonal anergy in vitro is against a causal role for CTLA-4 in the induction of Ag unresponsiveness (45). Furthermore, the capacity of rapamycin to promote the induction of Ag unresponsiveness even in the absence of CTLA-4 signals provides strong evidence for a direct role of cell cycle progression in the avoidance or reversal of clonal anergy.
Results of our study seemingly differ from those of Adler et al. (34), who varied the concentration of a transgenic self Ag and observed that T cells would become tolerant independent of the rate of cell division induced by the Ag. In fact, these two studies are complementary and point out the importance of two signals in the regulation of recall Ag responsiveness. Ag unresponsiveness appears to develop as a function of the ratio of anergy-inducing TCR signals (signal 1) to cell cycle progression (signal 2). Accordingly, chronic intense TCR ligation in high self-Ag-expressing mice can eventually dominate over signals driving cell cycle progression, and clonal anergy develops. For the case of low self-Ag-expressing mice, the anergy-inducing TCR signals will be weaker; nevertheless, tolerance is again induced because these signals are accompanied by only a low rate of cell cycle progression. In the experiments described in this work, OVAp-induced TCR signals are demonstrated to be uniform across the population of responding T cells. Therefore, variations in cell cycle progression had the opportunity to independently affect the level of recall Ag responsiveness within individual T cells.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
2 Address correspondence and reprint requests to Dr. Daniel L. Mueller, Center for Immunology, University of Minnesota Medical School, 312 Church Street SE, Room 6-120, Basic Science and Biomedical Engineering Building, Mayo Mail Code 334, Minneapolis, MN 55345. E-mail address: dmuell{at}lenti.med.umn.edu
3 Abbreviations used in this paper: mTOR, mammalian target of rapamycin; CMC, carboxymethylcellulose; GFP, green fluorescence protein; OVAp, chicken OVA peptide 323–339; SAPK, stress-activated protein kinase.
Received for publication August 1, 2001. Accepted for publication September 14, 2001.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
)-deficient T cells expressing a transgenic antigen receptor. J. Immunol. 158:3738.[Abstract]
c chain of the IL-2 receptor. Science 266:1039.1 induces antigen-specific unresponsiveness in naive T cells. Immunol. Invest. 26:459.[Medline]
This article has been cited by other articles:
|
|
 |
|
  J. Kang, S. J. Huddleston, J. M. Fraser, and A. Khoruts De novo induction of antigen-specific CD4+CD25+Foxp3+ regulatory T cells in vivo following systemic antigen administration accompanied by blockade of mTOR J. Leukoc. Biol., May 1, 2008; 83(5): 1230 - 1239. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 Y. Zheng, S. L. Collins, M. A. Lutz, A. N. Allen, T. P. Kole, P. E. Zarek, and J. D. Powell A Role for Mammalian Target of Rapamycin in Regulating T Cell Activation versus Anergy J. Immunol., February 15, 2007; 178(4): 2163 - 2170. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
T. L. Vanasek, S. L. Nandiwada, M. K. Jenkins, and D. L. Mueller CD25+Foxp3+ Regulatory T Cells Facilitate CD4+ T Cell Clonal Anergy Induction during the Recovery from Lymphopenia J. Immunol., May 15, 2006; 176(10): 5880 - 5889. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. Allen, Y. Zheng, L. Gardner, M. Safford, M. R. Horton, and J. D. Powell The Novel Cyclophilin Binding Compound, Sanglifehrin A, Disassociates G1 Cell Cycle Arrest from Tolerance Induction J. Immunol., April 15, 2004; 172(8): 4797 - 4803. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. L. Bonnevier and D. L. Mueller Cutting Edge: B7/CD28 Interactions Regulate Cell Cycle Progression Independent of the Strength of TCR Signaling J. Immunol., December 15, 2002; 169(12): 6659 - 6663. [Abstract] [Full Text] [PDF]
|
|
) or 100 µg of OVAp (
) in PBS.
The * each plot represents the mode number of cell divisions observed
in the KJ1-26+CD4+ T cell population as a
result of the OVAp pretreatment. Error bars represent the range in
responses seen in duplicate animals in an individual experiment. These
duplicate animals were representative of 14 individual mice.
and 
) and KJ1-26+ (
