Prior reports have shown that CD4+CD25+ regulatory T cells suppress naive T cell responses by inhibiting IL-2 production. In this report, using an Ag-specific TCR transgenic system, we show that naive T cells stimulated with cognate Ag in the presence of preactivated CD4+CD25+ T cells also become refractory to the mitogenic effects of IL-2. T cells stimulated in the presence of regulatory T cells up-regulated high affinity IL-2R, but failed to produce IL-2, express cyclins or c-Myc, or exit G0-G1. Exogenous IL-2 failed to break the mitotic block, demonstrating that the IL-2 production failure was not wholly responsible for the proliferation defect. This IL-2 unresponsiveness did not require the continuous presence of CD4+CD25+ regulatory T cells. The majority of responder T cells reisolated after coculture with regulatory cells failed to proliferate in response to IL-2, but were not anergic and proliferated in response to Ag. The mitotic block was also dissociated from the antiapoptotic effects of IL-2, because IL-2 still promoted the survival of T cells that had been cocultured with CD4+CD25+ T cells. IL-2-induced STAT5 phosphorylation in the cocultured responder cells was intact, implying that the effects of the regulatory cells were downstream of receptor activation. Our results therefore show that T cell activation in the presence of CD4+CD25+ regulatory T cells can induce an alternative stimulation program characterized by up-regulation of high affinity IL-2R, but a failure to produce IL-2, and uncoupling of the mitogenic and antiapoptotic effects of IL-2.
CD4+ T lymphocytes that coexpress the CD25 high affinity IL-2R include a lineage of regulatory T cells that can down-modulate CD4 and CD8 T cell function (1). This population includes <5% of CD4+ T cells and is characterized by the expression of the forkhead/winged helix transcription factor, Fox P3. Genetic deficiency of the Fox P3 for gene leads to an absence of CD4+CD25+ regulatory T cells and the spontaneous development of multiorgan autoimmunity (2, 3, 4). CD4+CD25+ regulatory T cells, therefore, play a crucial role in immune homeostasis and tolerance.
How CD4+CD25+ T cells function is poorly understood. In vitro, these cells are anergic, failing to proliferate after TCR stimulation except in the presence of IL-2 or IL-2/IL-4 (5, 6, 7). In addition to their own unresponsiveness, CD4+CD25+ T cells are able to suppress the proliferative response of bystander T lymphocytes. Suppression is Ag and MHC independent, requires stimulation of the CD4+CD25+ T cell, and appears to be cell contact or proximity dependent. A role for TGF-β in the antiproliferative effects of CD4+CD25+ T cells has been proposed; however, conflicting data on this remain unresolved (8, 9, 10, 11).
Prior studies have shown that CD4+CD25+ regulatory T cells block IL-2 production by, and thereby presumably block the proliferation of, their CD4+CD25− targets (5). In this study we extend these findings. By analyzing target T cells during and after coculture with regulatory T cells, we found that CD4+CD25+ regulatory T cells can promote a disengagement of the IL-2 signaling pathway, in which upstream signaling is uncoupled from downstream cell cycle progression, but not from the promotion of cell survival. This uncoupling is distinct from anergy, and reisolated target cells are fully responsive to Ag if restimulated in the absence of CD4+CD25+ T cells.
Materials and Methods
AND mice, transgenic for a rearranged pigeon cytochrome c peptide (PCC)3-specific-TCR, were bred >20 generations onto the B10.BR background. B10.BR mice were obtained from The Jackson Laboratory.
Media, reagents, and Abs
Cells were grown in Eagle’s-Hank’s amino acid (BioSource) supplemented with 10% heat-inactivated Premium FCS (BioWhittaker), penicillin G (100 U/ml), streptomycin (100 μg/ml), 292 μg/ml l-glutamine (Invitrogen Life Technologies), and 50 μM 2-ME (FisherBiotech). PCC (KAERADLIAYLKQATAK) was synthesized and HPLC purified by the St. Jude Children’s Research Hospital Hartwell Center for Biotechnology. FITC-conjugated or biotinylated anti-CD25 (7D4), allophycocyanin-conjugated anti-CD4 (L3T4), PE-conjugated-anti-CD69 (H1.2F3), and anti-CD16/CD32 Fc block (2.4G2) were purchased from BD Pharmingen.
Single-cell, mixed lymph node and spleen suspensions were prepared, erythrocytes were lysed with Gey’s solution, and T cells were purified on nylon wool columns (Polysciences). CD4+CD25− and CD4+CD25+ T cells were isolated by staining with Fc block, allophycocyanin-anti-CD4, and FITC-anti-CD25 Abs in PBS/5% FCS for 20 min before flow cytometric sorting. Sorted cell purity ranged from 97 to 99%.
CD4+CD25+ AND T cells were expanded by culturing in complete medium with 3000-rad irradiated B10.BR splenocyte feeders, 5 μM PCC peptide, 10 ng/ml recombinant mouse IL-2 (rmIL-2; R&D Systems), and 500 U/ml rmIL-4 (R&D Systems). After 2 days, viable cells were purified by Ficoll/Hypaque centrifugation (Lymphocyte Separation Media; BioWhittaker) and cultured in growth factor-enriched medium. Cells were restimulated every 9–10 days for a maximum of three passages. Cells were analyzed at least 4 days after stimulation or as indicated.
[3H]Thymidine proliferation assays
CD4+CD25− T cells (5 × 104) from 5- to 8-wk-old AND mice were cultured for 72 h in Eagle’s-Hank’s amino acid complete medium in the absence of any exogenous cytokines in round-bottom, 96-well plates (Corning-Costar) with 2.5 × 105 3000-rad irradiated B10.BR splenocytes and 5 μM PCC peptide, with or without 5 × 104 activated and purified CD4+CD25+ AND T cells. Cultures were pulsed with [3H]thymidine (1 μCi/well) for 18–20 h.
CFSE proliferation assays
Flow cytometrically purified CD4+CD25− AND T cells were washed and resuspended at 10–50 × 106 cells/ml in 5 μM CFSE (Molecular Probes)/PBS/0.1% BSA for 8 min at 37°C. The cells were then washed three times and mixed at a 1:1 ratio with unlabeled preactivated CD4+CD25+ AND T cells or control freshly isolated CD4+CD25− AND T cells. Irradiated splenocyte feeders were added at a concentration 3–5 times that of the T cells, and the cells were stimulated with 5 μM PCC peptide. The cocultures were analyzed by flow cytometry at 24, 48, or 72 h. In some cases the CFSE+ cells were resorted and analyzed after reculturing them in different conditions as described.
Cells were stained and analyzed on a FACSCalibur (BD Biosciences) using CellQuest software (BD Biosciences). Cell cycle progression was determined by resuspending the cells at 1 × 106 cells/ml in PBS, then adding propidium iodide (PI; 0.05 mg/ml) in a 0.1% sodium citrate/0.1% Triton X-100 solution. Samples were incubated for 30 min at room temperature in the dark in the presence of 0.2 mg/ml RNase before flow cytometric analysis using ModFit LT software (Verity Software). Cell sorting was performed using a MoFlo high speed sorter (DakoCytomation). Quantitative flow cytometry to determine viable cell counts was performed by adding 2000 6-μm fluorescent TruCount beads (BD Biosciences) to culture wells. Cells were then stained with PI, and viable cell numbers, gated for PI staining and forward/side scatter (FSC/SSC), were determined by normalizing cellular events to beads counted.
Cytokine secretion analysis
Cell-free supernatants (50 μl), harvested at the designated times, were tested for IL-2 content by Bio-Plex assay according to the manufacturer’s instructions (Bio-Rad).
RT-PCR for mIL-2 and c-Myc
Total RNA was extracted using the RNeasy kit (Qiagen) from an identical number of CFSE-labeled, flow cytometrically sorted T cells, before or after coculture as described. Extracted RNA was quantified by OD, and identical quantities (50 ng to 1 μg) were reverse transcribed (Omniscript; Qiagen) and analyzed by PCR (30 cycles) with mIL-2-, c-Myc-, or β-actin-specific primers.
To analyze STAT5, p27, and cyclin expression, CFSE-labeled responder cells were flow cytometrically sorted and, where indicated, activated with 10 ng/ml rmIL-2 for 30 min. They were immediately washed three times with ice-cold PBS, and equal numbers of cells were lysed in ice-cold buffer containing 50 mM Tris (pH 8.0), 150 mM NaCl, 5 mM EDTA, 1 mM EGTA, 1 mM Na2VO4, 1% Nonidet P-40, and protease inhibitor mixture (Roche). Thirty micrograms of protein was separated on 12% SDS-PAGE and blotted, and blots were probed with anti-mouse p27kip1 (Santa Cruz Biotechnology); anti-cyclin D2, D3, or E (prepared from hybridomas, gift from Dr. M. Roussel); or anti-phospho-STAT5 (BD Bioscience), followed by anti-β-actin (BD Bioscience), or anti-STAT5 (BD Bioscience). Blots were washed, incubated with horseradish peroxide-conjugated secondary Ab, and developed using chemiluminescence as per manufacturer’s instructions (Amersham Biosciences).
Isolation and characterization of AND Tg CD4+CD25+ regulatory cells
Some 1.9 ± 1.0% of CD4+ T cells in AND TCR transgenic mice coexpressed CD25 (data not shown). Flow cytometric sorting of combined lymph node cells and splenocytes allowed us to isolate ∼1–5 × 105 AND TCR transgenic CD4+CD25+ T cells/mouse. These regulatory T cells were anergic, and stimulation with PCC peptide failed to induce their proliferation. However, they were able to suppress the Ag-induced proliferation of naive, flow cytometrically purified CD4+CD25− AND T cells (Fig. 1 A).
Studies using Ag-nonspecific CD4+CD25+ regulatory T cells have shown that they retain their suppressive function when expanded by stimulation in the presence of IL-2 and/or IL-4 (7, 12). To acquire the requisite numbers of CD4+CD25+ AND T cells to analyze their impact on target T cells, we expanded them with PCC peptide in the presence of IL-4 and IL-2. We found that Ag-specific regulatory function, as determined by suppression of bystander T cell proliferation, was preserved after even three cycles of stimulation (Fig. 1 B). This verifies that culture-expanded, Ag-specific regulatory T cells are functionally intact.
Preactivated CD4+CD25+ AND T cells block IL-2 production and G0-G1 progression
Analysis of suppression mediated by freshly isolated Ag-nonspecific CD4+CD25+ T cells has demonstrated that despite activation-induced up-regulation of high affinity IL-2R (CD25) on target T cells, curtailed IL-2 production prevents the activated T cells from proliferating (5). We similarly observed up-regulation of both CD25 and CD69 activation Ags on responding T cells after Ag stimulation in the presence of preactivated, Ag-specific regulatory T cells (Fig. 2). This up-regulation was only mildly decreased relative to that of cells stimulated in the presence of CD25− control T cells, implying that the regulatory T cells did not prevent Ag engagement and TCR stimulation. Although constitutively expressed, IL-2Rβ (CD122) may also be regulated with T cell activation. However, similar levels of CD122 were observed on T cells 2 days after stimulation in the presence or the absence of regulatory T cells (data not shown).
In contrast with control stimulated CD4+CD25− cells, peptide stimulation in the presence of CD4+CD25+ T cells led to little IL-2 accumulation in the culture medium (Fig. 3,A). This resulted from an IL-2 production defect. CFSE-labeled CD4+CD25− cells stimulated with peptide Ag in the presence of control unlabeled CD4+CD25− cells and then flow cytometrically purified demonstrated accumulation of IL-2 mRNA. In contrast, labeled CD4+CD25− cells similarly isolated after stimulation in the presence of unlabeled CD4+CD25+ T cells lacked detectable IL-2 mRNA (Fig. 3 B). Therefore, naive T cells recognize Ag in the presence of PCC-stimulated CD4+CD25+ T cells. This recognition leads to an incomplete stimulation program characterized by up-regulation of the CD25 and CD69 early activation Ags, but failure to produce IL-2 or proliferate.
To better define the location of the mitotic block in the suppressed cells, we performed cell cycle analysis. Flow cytometrically purified CD4+CD25− T cells were labeled with CFSE and stimulated in the presence of unlabeled regulatory T cells or control naive T cells. Twenty-four and 48 h later, we analyzed loss of CFSE as an indicator of mitosis (data not shown) and determined DNA content (Fig. 4) for cell cycle progression. At 24 h, neither control nor regulatory cell-cocultured, CFSE-labeled cells had divided, and ∼85% of cells cultured in each condition were in G0-G1. By 48 h, proliferation was detected among the control cultured cells, but not among those cultured with regulatory cells. A significant block in G0-G1 was observed, with 70% of control cells in S or G2-M compared with 23% of experimental cells. Therefore, CD4+CD25+ regulatory T cells induce an early block of cell cycle progression.
Cyclins D2 and D3 are up-regulated and mediate early cell cycle progression of resting T cells (13). IL-2 plays a critical role in progression from G0-G1, in part through the up-regulation of cyclin D expression (14). This can occur indirectly through the induction of c-Myc, a transcriptional activator, or through the direct interaction of STAT5 with the cyclin D promoter and may be enhanced by IL-2-induced PI3K activity (15, 16). Little cyclin D2 or D3 expression was seen in CFSE-labeled cells purified after activation in the presence of regulatory cells compared with control cells (Fig. 5,A). The absence of D cyclins was consistent with the G0-G1 block observed in our cell cycle analysis. This also correlated with diminished expression of cyclin E (Fig. 5,A), failure to induce c-Myc (Fig. 5,B), and elevated levels of the cyclin-dependent kinase inhibitor p27kip1 (Fig. 5 C), which is normally down-modulated in response to IL-2 or effective costimulation (17, 18). These results demonstrate that despite up-regulation of the early activation markers CD25 and CD69, T cells stimulated in the presence of CD4+CD25+ regulatory cells fail to engage the cellular machinery required for cell cycle progression.
Targeted T cells are refractory to the mitogenic effects of IL-2, but signal through IL-2R
Blockade of CD4+CD25− T cell IL-2 production may have been responsible for the antiproliferative effects of the regulatory T cells. Published data showing proliferation of cocultures of freshly isolated CD4+CD25+ T cells and targeted responder cells in the presence of IL-2 support this view. However, most of those experiments used IL-2 concentrations in which isolated CD4+CD25+ regulatory T cells proliferated vigorously, and whether CD4+CD25− cell expansion was independently inhibited was not fully established (5, 12).
To test whether IL-2 could circumvent the activity of our preactivated CD4+CD25+ T cells after Ag stimulation, we stimulated regulatory cells alone, CD4+CD25− T cells alone, or cocultures of the cells in the absence of IL-2 or in the presence of 2 or 10 ng/ml IL-2. A concentration of 10 ng/ml and, to a lesser extent, 2 ng/ml IL-2 induced the proliferation of CD4+CD25+ AND T cells, which otherwise failed to respond to Ag (Fig. 6,A). Response magnitude was significantly less than that of equal numbers of CD4+CD25− T cells. Interestingly, the addition of IL-2 to the cocultured population resulted in a net proliferative response that was identical with that observed among the CD4+CD25+ regulatory T cells alone and decreased compared with that of the CD4+CD25− cells. This suggested that CD4+CD25− T cell proliferation was suppressed even in the presence of exogenous IL-2, a finding that was confirmed in studies specifically assessing proliferation of CFSE-labeled CD4+CD25− T cells by loss of fluorescence (Fig. 6, B and C). Indeed, proliferation inhibition was nearly complete in cocultures with preactivated regulatory T cells and was unaffected by the addition of IL-2. This strong inhibition was probably essential to clearly manifest the IL-2 unresponsiveness, because proliferation by incompletely or nonsuppressed responder cells would be expected to be highly IL-2 dependent in the IL-2-deficient environment of the coculture. Thus, despite the expression of high affinity IL-2R on the cocultured target T cells, supplementation with exogenous IL-2 did not overcome target cell proliferation inhibition. Therefore, a defect in IL-2 production occurs with preactivated CD4+CD25+ regulatory T cell-mediated suppression. However, this defect is not exclusively responsible for the proliferation failure. Rather, refractoriness of the target cells to the promitotic effects of IL-2, as defined by their failure to proliferate after 72 h in the presence of IL-2, acts cooperatively to prevent naive T cell proliferation.
IL-2R-induced phosphorylation of STAT5 plays a critical upstream role that is necessary for T cell proliferation (19). Phosphorylation of STAT5 by JAKs permits its homodimerization, converting it into an active form. To determine whether IL-2R in CD4+CD25+, T cell-treated, IL-2-unresponsive T cells was itself wholly uncoupled from signaling in response to IL-2, we analyzed STAT-5 phosphorylation. CFSE+ cells were flow cytometrically purified from cocultures with Ag, APC, and either unlabeled CD4+CD25+ or control CD4+CD25− T cells. CFSE+ T cells isolated from cocultures with CD4+CD25+ regulatory cells did not show detectable basal STAT5 phosphorylation by Western analysis (Fig. 7). In contrast, cells isolated from control cultures showed low levels of phosphorylation, presumably resulting from the IL-2 accumulating in the medium during culture (Fig. 3 A). To determine whether IL-2R could activate STAT5 in the regulatory cell-cocultured population, we stimulated flow cytometrically isolated CFSE-labeled T cells with IL-2. Effective phosphorylation of STAT5 was observed among cells that had been cocultured with regulatory T cells. This demonstrates that IL-2R is indeed functional in the regulatory cell-cocultured population. It signals in response to IL-2, as evidenced by the phosphorylation of STAT5. However, STAT5 activation is dissociated from cell cycle progression.
CD4+CD25+ T cell persistence is not required for persistent refractoriness
We were next interested in whether persistence of the regulatory T cells was required to maintain the suppressed and IL-2 refractory state of the responder cells. To test for this, naive CD4+CD25− T cells were CFSE labeled, cocultured with unlabeled CD4+CD25+ T cells or control CD4+CD25− T cells, and stimulated with Ag. Twenty-four or 48 h later, at which time full T cell stimulation should normally be induced (20, 21), CFSE-labeled cells were rescued from regulatory cells by flow cytometric sorting. When the sorted CFSE+ cells were returned to culture, the cells failed to proliferate (Fig. 8 A and data not shown). Indeed, by 72 h after sorting, few cocultured cells remained, implying that they had undergone apoptosis. This demonstrates that stimulation of responder cells in the presence of CD4+CD25+ T cells leads to an incomplete activation program and not a transient proliferative block dependent on continued regulatory T cell presence.
The cocultured responder cells uniformly up-regulated high affinity IL-2R, but failed to produce or respond to IL-2. If the removal of regulatory T cells alleviated the block in responder cell IL-2 signaling, supplementation of the reisolated responder cells with exogenous IL-2 should relieve their proliferation suppression. To test for this, we added exogenous IL-2 to the culture medium. In the control cultured population, this resulted in mildly increased proliferation, defined by the proportion of cells in each proliferation peak, and seemed to result in enhanced viability, determined by flow cytometric cell counts (Fig. 8,A). However, even with added IL-2, the majority of responder cells isolated from cocultures with regulatory cells failed to divide, whereas a minority of cells showed limited cell cycling. Indeed, after 72 h of reculture in IL-2, >60% of viable CFSE+ cells had failed to divide. This demonstrates that despite the presence of high affinity IL-2R (Fig. 2) capable of signaling through STAT5 (Fig. 7), exogenous IL-2 could not relieve the suppression of the majority of responder T cells after regulatory T cells had been removed.
Responder cells that had been suppressed by regulatory T cells may have developed anergy after Ag encounter, thereby limiting their ability to respond even after removal of the regulatory population, as has been observed in other systems (22). This, however, was not the case. Rescued CFSE-labeled cells restimulated with Ag exhibited proliferative responses similar to those of control treated cells (Fig. 8 B). These results demonstrate that T cell stimulation in the presence of Ag-stimulated CD4+CD25+ T cells induces a partial stimulation program, leading to up-regulation of IL-2R, but not IL-2 production. Even after only 24-h coculture with regulatory T cells, the majority of T cells are not responsive to the mitogenic effects of IL-2. Implicitly, the IL-2-signaling pathway is unengaged in these cells and therefore is unable to induce cell cycling. Restimulation with Ag presumably re-engages this pathway, allowing normal proliferation.
IL-2 promotes survival of regulatory T cell-treated responder cells
Our data implied that responder T cells removed from Ag and regulatory T cells died when recultured in medium, but not when recultured in IL-2. Indeed, although an equivalent proportion of each sample was analyzed by flow cytometry in Fig. 8 A, by 72 h after reisolation, few T cells from cocultures were detected in the absence of IL-2, whereas significant numbers were detected in its presence. To better evaluate the influence of IL-2 on cell survival, we quantitatively analyzed the persistence of CFSE-labeled responder cells similarly isolated and cultured. Freshly isolated, CFSE-labeled CD4+CD25− AND T cells were Ag-stimulated in the presence of preactivated CD4+CD25+ T cells. Twenty-four hours later, CFSE+ cells were flow cytometrically isolated, and equal numbers of cells (∼5 × 104) were added per well of a 96-well plate. Zero or 10 ng/ml rmIL-2 was added. Twenty-four, 48, or 72 h later, 2000 6-μm fluorescent latex beads were added to each well, and the samples were stained with PI to assess viability and analyzed by flow cytometry. Absolute cell counts in the wells could be quantified by normalizing the number of cells counted with the latex bead count.
Quantitative analysis of recultured cells showed a significantly greater loss of viable cells in cultures lacking IL-2 than in those including it. In the example shown in Fig. 9,A, 24 h after reculture there was 1.7 times more FSCbright PI− cells in the culture with IL-2 than in that lacking it. By 72 h there were 12 times more cells. The loss of viable cells in the absence of IL-2 corresponded to an increase in the number of FSCbright/moderate cells that were PI+, implying increased cell death in the absence of IL-2. Fig. 9,B shows that this increased viability was consistent across multiple samples, with a mean 39-fold difference in viable CFSE+ cell counts at 72 h. The increased cell numbers in samples recultured with IL-2 did not result from cell proliferation. As shown in Fig. 8,A, limited proliferation of the reisolated IL-2-cultured cells occurred (Fig. 9 C). However, enumeration of undivided (generation 0) viable CFSE+ cells at 72 h showed a mean 26-fold increase in cell number in cultures receiving IL-2 compared with cultures without it. These results demonstrate that the cocultured T cells are responsive to the antiapoptotic effects of IL-2. They also imply that CD4+CD25+ regulatory T cells promote a dissociation between IL-2-induced mitogenesis, which is largely blocked, and cell survival, which is preserved.
Expansion of T cells after effective TCR engagement is critical for development of the adaptive immune response. One characteristic of CD4+CD25+ regulatory T cells is the suppression of T cell proliferation, defined in vitro in coculture assays (1). In these studies we extend, using an Ag-specific system, previous results demonstrating that CD4+CD25+ regulatory T cells block the induction of IL-2 in stimulated T cells. Not surprisingly considering the absence of IL-2, we demonstrate that regulatory T cells induce an early block in CD4 T cell cycling. This is associated with a failure to progress from G0-G1, a failure to up-regulate cyclins or c-Myc, and increased p27kip1. More surprisingly, we now show that although responder T cell IL-2R remains functional, with its engagement leading to STAT5 phosphorylation and enhanced survival, most regulatory T cell-cocultured responder cells are unresponsive to the mitogenic effects of IL-2. Hence, CD4+CD25+ T cells can promote at least two stimulation defects, a failure to produce IL-2 and a failure to proliferate in response to this cytokine.
One interpretation of the suppressive effects of CD4+CD25+ T cells is that they promote a transient, contact-dependent block in proliferation despite otherwise normal T cell activation. Naive T cells require only as little as 8-h exposure to Ag to develop sensitivity to the mitogenic effects of IL-2 (20, 21). Because removal of the responder cells from regulatory cells and Ag after even 24–48 h of stimulation failed to alleviate the proliferative block, the effects of the regulatory cells do not seem to require continuous contact. Mitotic inhibition of the responder cells persists after removal of the regulatory cells even if IL-2 is added to their culture medium. This suggests that, instead, a partial stimulation program is induced in the responder cells, resulting in CD25 expression, but neither IL-2 expression nor sensitivity to the mitogenic effects of IL-2. Anergy does not develop, because antigenic restimulation in the absence of regulatory cells is able to initiate cellular proliferation.
Our results support the idea that effective T cell stimulation not only requires the coordinate up-regulation of a growth factor receptor, IL-2R, and its stimulus, IL-2, but must independently enable downstream components of the IL-2-signaling pathway. CD4+CD25+ regulatory T cells may act to either induce mediators that turn off these downstream signaling pathways or block stimuli required for their activation. Additional studies will be needed to define the responsible biochemical machinery. Nevertheless, recent data may provide useful insights and suggest that the uncoupling of IL-2 signaling from cellular proliferation observed in this study with CD4+CD25+ T cells may form a common mechanism regulating T cell activation. Thus, T cells that fail to cycle after stimulation were found to be similarly refractory to the mitogenic effects of IL-2 (23, 24). Likewise, stimulation of T cells with TGF-β blocks IL-2-mediated proliferation through direct effects on cell cycle genes (25, 26). As in this study, the STAT5 signaling pathway remains intact; however, downstream elements regulated by STAT5, including cyclins and c-Myc, are not induced. Interestingly, TGF-β has been implicated in the suppressive effects of CD4+CD25+ regulatory T cells. However, because we were not able to block regulatory T cell function using TGF-β-specific Abs in our system (data not shown), we cannot currently attribute our results to the expression of this cytokine.
Recent findings suggest an important role for ryanodine receptors (RyR) in the Ca2+-dependent signal transduction required for TCR-mediated T cell activation and IL-2-driven T cell proliferation. In one study, blockade of RyR signaling inhibited proliferation and IL-2 synthesis. Interestingly, similar to the results of our study, this blockade was not relieved by the addition of exogenous IL-2, despite up-regulation of the high affinity IL-2R (27). This may be associated with differential activation of transcription factors in response to Ca2+ signals of differing amplitude and duration occurring with RyR blockade (low, sustained intracellular Ca2+ plateau for NFAT activation, and transient initial intracellular Ca2+ peak for NF-κB activation (28)). Also, transient tyrosine phosphorylation of RyR by the tyrosine kinase p59fyn may regulate Ca2+ mobilization via RyR, suggesting a potential physiologic mechanism for modulating RyR activity (29). Considering the similarity of RyR blockade and regulatory T cell activity, analysis of this as well as other TCR signaling pathways in the targets of CD4+CD25+ T cells may provide insight into how regulatory T cells subvert Ag-induced expansion.
A more complete dissection of IL-2 signal transduction will also be important. Our analyses were limited to STAT5 induction by IL-2 and demonstrate that IL-2 was able to signal through the IL-2R in suppressed T cells. However, the completeness of that signal will need to be assessed. Signals from the IL-2R diverge into multiple signaling pathways (19, 30, 31, 32), and these separate pathways may be independently regulated and play distinct roles in governing the effects of CD4+CD25+ regulatory T cells. STAT5 is critical for T cell proliferation, but regulation of signaling through Shc or other pathways that act cooperatively with STAT5 may be important to the effects of the regulatory T cells on T cell proliferation.
Importantly, IL-2 strongly induces protein kinase B (PKB) signaling. PKB is a potent stimulator of antiapoptotic pathways and a critical regulator of T cell longevity (33, 34, 35). This may result from phosphorylation of Bad and modulation of the expression of apoptotic regulators, such as Bim, Fas ligand, Bcl-2, and Bcl-xL. IL-2-induced activation of PKB may therefore account for the antiapoptotic effects of IL-2 observed in our study. However, PKB can also promote cell cycle progression by regulating cell cycle inhibitors and promoters, such as p27kip1 and E2F (36, 37). Considering that the suppressed T cells fail to proliferate in response to IL-2 despite phosphorylation of STAT-5, it would seem that the block of T cell proliferation is downstream of both PKB and STAT5. Additional biochemical dissection of this pathway will therefore be important in establishing the mechanism of regulatory T cell action in our system.
Of interest, our data seemingly conflict with those of one prior study demonstrating the induction of anergy in CD4+CD25− T cells stimulated in the presence of CD4+CD25+ regulatory cells with latex beads coated with anti-CD3 and limiting doses of anti-CD28 (22). In those experiments, T cells rescued from the regulatory cells were unable to respond to stimulation except in the presence of added IL-2. In our Ag-specific system, we failed to observe anergy induction. The reason for this difference is unclear, but may reflect differences in the systems used for analysis, such as the use of preactivated regulatory cells, full vs minimal costimulation, the presence of APCs, or the stimulus administered. In one recent report, CD4+CD25+ T cells that were preactivated had distinct properties from those that were not, with preactivated, but not freshly isolated, regulatory cells able to suppress Th2 cytokine production and proliferation of established Th2 cells (38). This difference supports the idea that the consequences of a regulatory cell’s interaction with a target T cell and potentially the mechanism(s) of regulatory T cell action are modified by stimulation conditions.
The use of preactivated CD4+CD25+ T cells may likewise explain other differences between our system and previous results using freshly isolated regulatory T cells. Freshly isolated CD4+CD25+ T cells proliferate as vigorously as naive T cells when stimulated in the presence of a high dose of IL-2 (5). In contrast, we observed less proliferation among stimulated preactivated CD4+CD25+ T cells than naive T cells (Fig. 6 A). A diminished proliferative response when recently activated T cells are restimulated has been well characterized with CD4+ T cells, as initially documented by Wilde and Fitch (39). This is enhanced by culture in high doses of IL-2, which we use to expand our regulatory cells. We suspect that a similar reduced responsiveness upon restimulation is occurring with the preactivated CD4+CD25+ T cells studied in this study. Indeed, we also observed a diminished proliferative response when TCR Tg regulatory cells specific for collagen II-peptide/IAq are restimulated after initial stimulation with Ag and high dose IL-2, demonstrating that this effect is not specific to the AND TCR Tg system (H. Inaba and T. Geiger, unpublished observations).
In summary we demonstrate that CD4+CD25+ regulatory T cells are able to uncouple IL-2 signaling from cellular proliferation. Antigenic signaling in the presence of regulatory cells leads to an altered cellular activation program, characterized by up-regulation of high affinity IL-2R, but a failure to produce IL-2, and only partial engagement of the IL-2 signaling pathway. These results provide new evidence for how CD4+CD25+ regulatory T cells are able to influence adaptive immune responses.
We thank Martine Roussel for assistance with cell cycle and Western analyses; Richard Cross, Dick Ashmun, and Jennifer Hoffrage for assistance with flow cytometric sorting; and Janet Gatewood for assistance with cytokine analyses.
The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
This work was supported by National Institutes of Health Grants R21AI49872 and R01AI056153 (to T.L.G.) and by the American Lebanese Syrian Associated Charities/St. Jude Children’s Research Hospital (to all authors).
Abbreviations used in this paper: PCC, pigeon cytochrome c peptide; FSC, forward scatter; SSC, side scatter; PI, propidium iodide; PKB, protein kinase B; rmIL-2, recombinant mouse IL-2; RyR, ryanodine receptor.