Opposing Roles for the Cyclin-Dependent Kinase Inhibitor p27kip1 in the Control of CD4+ T Cell Proliferation and Effector Function
http://www.100md.com
免疫学杂志 2005年第6期
Abstract
Cell division drives T cell clonal expansion and differentiation, and is the result of concerted signaling from Ag, costimulatory, and growth factor receptors. How these mitogenic signals are coupled to the cell cycle machinery in primary T cells is not clear. We have focused on the role of p27kip1, a major cyclin-dependent kinase binding protein expressed by CD4+ T cells. Our studies using p27kip1 gene dosage demonstrate that early after activation, p27kip1 acts to promote, rather than inhibit, G1 to S phase progression within the first division cycle. However, throughout subsequent cell divisions p27kip1 behaves as a negative regulator, directly establishing the threshold amount of growth factor signaling required to support continued cell division. During this phase, signals from CD28 and IL-2R cooperate with the TCR to "tune" this threshold by inducing the degradation of p27kip1 protein, and we show that agents that block these pathways require elevated p27kip1 levels for their full antiproliferative activity. Finally, we show that p27kip1 opposes the development of CD4+ T cell effector function, and is required for the full development of anergy in response to a tolerizing stimulus. Our results suggest that p27kip1 plays a complex and important role in the regulation of cell division and effector function in primary CD4+ T cells.
Introduction
The role of cell division in a T cell response is 2-fold. This process not only drives clonal expansion, thereby increasing the frequency of T cells capable of recognizing a given Ag, but also drives effector T cell differentiation at the single cell level. Those T cells that undergo more rounds of division during a response are more likely to produce effector cytokines upon re-encounter with Ag (1, 2, 3, 4).
Signals from Ag, costimulatory, and growth factor receptors cooperate to induce cell cycle progression in quiescent T cells. An important step in linking these extracellular signals to the cell cycle machinery is the regulated and ordered synthesis of cyclins and their interaction with cyclin-dependent kinases (CDK)3 (5). The activity of these cyclin-CDK enzymes is further regulated by CDK inhibitors, proteins that bind to cyclin-CDK complexes and generally act to oppose cell cycle progression (6). One such protein capable of binding to multiple cyclin-CDK pairs is p27kip1. The importance of this CDK inhibitor in development is demonstrated by mice genetically deficient for p27kip1, which exhibit multiorgan hyperplasia, gigantism, and increased incidence of tumors (7, 8, 9).
In mature T lymphocytes, like in other tissues, p27kip1 is highly expressed in quiescent cells, but is down-regulated upon mitogenic stimulation (10, 11). If proliferation is inhibited during T cell activation, such as during immunosuppression or tolerance induction, p27kip1 levels instead remain elevated (4, 10, 12, 13). These observations have led to the notion that p27kip1 is an important negative regulator of G1 to S phase progression in T lymphocytes. Indeed, studies using forced overexpression have shown that excess p27kip1 can impair T cell proliferation in response to mitogenic stimuli, confirming its role as a negative regulator of the cell cycle when expressed at appropriately high levels (12, 14). However, studies using genetic deletion of p27kip1 have been unable to clearly establish whether this molecule has a significant influence over T lymphocyte proliferation and differentiation under physiologic circumstances. p27kip1-deficient mouse strains exhibit a striking increase in thymic cellularity, as well as a significant augmentation of splenic cellularity. However, when peripheral T lymphocyte function was examined, some studies found increased proliferative responses of p27kip1-deficient T cells in response to TCR ligation and/or IL-2R signaling (7, 15, 16), while other studies found no difference in proliferation between p27kip1-deficient and T cells (8, 9, 17). Likewise, attempts to establish genetically whether immunosuppressive drugs such as rapamycin act directly through p27kip1 have led to similar opposing conclusions, with some studies suggesting that p27kip1 is required for immunosuppression (18), while other studies found no role for this CDK inhibitor during immunosuppression (8, 9).
Cell cycle progression is required for efficient T cell differentiation and the avoidance of anergy (4). Multiple correlative studies have established an association between elevated p27kip1 levels and T cell anergy, but again, the role of p27kip1 in the induction or maintenance of this state has remained controversial. A forced overexpression approach suggested that p27kip1 could induce lasting tolerance in T cells through suppression of IL-2 gene expression (12), while separate studies using p27kip1-deficient T lymphocytes concluded that this CDK inhibitor was not required for anergy induction in the models tested (19). These contrasting results imply that the role of p27kip1 in a T cell response may be more complex than previously appreciated.
One potential problem with forced overexpression and full genetic deletion is that these situations do not faithfully recapitulate the more subtle and dynamic posttranslational mechanisms by which p27kip1 is controlled physiologically. We have used T cells that either contain wild-type levels of p27kip1, completely lack p27kip1, or contain significantly reduced levels of p27kip1 to study the extent to which signals from the TCR, CD28, and IL-2R are coupled to the cell cycle through this molecule. Our careful analysis of these T cell responses reveals that p27kip1 functions physiologically within a 2-fold range of protein levels. Low levels of p27kip1 can actually promote cell cycle progression by activated CD4+ T cells, while a higher concentration of p27kip1 restricts cell division, opposes effector function, and promotes anergy. Our results imply that p27kip1 acts as a "rheostat" that influences T cell differentiation by regulating cell cycle progression.
Materials and Methods
Mice
Wild-type C57BL/6 mice and C57BL/6 mice homozygous for deletion of p27kip1 (7) were obtained from The Jackson Laboratory and bred under specific pathogen-free conditions. Progeny were screened for the wild-type and null alleles by PCR as previously described (7). As previously observed, p27kip1+/– and p27kip1–/– mice in our colony contained 50–90% more peripheral lymphocytes as compared with age-matched wild-type mice (7), however, the ratio of B cells to T cells, CD4+ to CD8+ T cells, and naive (CD44lowCD62Lhigh) vs memory (CD44highCD62Llow) T cells did not differ significantly from wild-type littermates (data not shown).
Cell culture and reagents
Pooled spleen and lymph node cell suspensions were prepared using 100-μm nytex fabric, and CFSE (Molecular Probes) labeling was performed as described previously (20). Cells were cultured at 2 x 106/ml in medium (RPMI 1640 supplemented with 10% FBS, L-glutamine, penicillin, streptomycin, and 2-ME) with soluble anti-CD3 (145-2C11), anti-CD28 (37.51, 1 μg/ml), CTLA-4Ig (5 μg/ml), control human IgG (5 μg/ml), rapamycin (Calbiochem; 10 ng/ml in DMSO), or rmIL-2 (Roche) as indicated in the figure legends. In this system, accessory cells provide "signal 1" via FcR-mediated cross-linking of TCR-bound anti-CD3 Ab, and also provide the costimulatory signals required for a physiologic proliferative response. For anergy assays, primed CD4+ cells were rested for 24 h in medium, purified, and restimulated for 24–48 h with anti-CD3 and anti-CD28, at concentrations indicated in figure legends, immobilized on either flat-bottom 96-well plates or on latex beads (Interfacial Dynamics). DNA synthesis was measured by the addition of BrdU (Sigma-Aldrich) to the medium at a final concentration of 10 μM 6 h before harvest.
Flow cytometry and CD4+ T cell purification
Lymphocytes (0.5–1 x 106) from primed cultures were stained with PerCP-conjugated anti-CD4 Ab (BD Pharmingen) and PE-conjugated anti-Thy1.2 Ab (BD Pharmingen), and cell division was analyzed by CFSE dilution. The absolute number of cell divisions accumulated within the CD4+ T cell subset was calculated as previously described (3). BrdU incorporation by individual cells was detected using a Phoenix Red-conjugated anti-BrdU Ab (Phoenix Flow Systems) after fixation and permeabilization (Fix & Perm; Caltag Laboratories). DNA content was measured by the addition of 7-aminoactinomycin D (25 μg/ml) to the fixed and permeabilized cells for 30 min. The above fluorescence parameters were assessed using a FACSCalibur flow cytometer (Becton Dickinson). Where indicated, CD4+ T cells were purified from primed cultures by negative selection using Bio-Mag beads (Qiagen) with Abs against I-A/E, CD19, CD16/32, CD11b, and CD8 (BD Pharmingen).
Immunoblot analysis
Total cultures containing both CD4+ and CD8+ T cells were harvested, lysed, and subjected to SDS-PAGE (0.5 x 106 cell equivalents per lane). The resolved proteins were transferred to nitrocellulose, blocked with a commercial blocking reagent (Roche), and probed with an Ab against p27kip1 (Santa Cruz Biotechnologies). The blots were stripped and reprobed with an Ab against actin (Santa Cruz Biotechnologies). Immunoreactive protein was visualized using HRP-conjugated secondary Ab (Jackson ImmunoResearch Laboratories) and a chemiluminescent substrate (Pierce) on a Bio-Rad Chemidoc system. Densitometric quantification was performed using ImageJ image analysis software (Version 1.31k, National Institutes of Health, http://rsb.info.nih.gov/ij).
Measurement of IL-2 secretion
IL-2 secreted into the supernatant was measured by fluorescence-linked immunosorbent assay (FLISA) or by ELISA. IL-2 FLISA was performed by coating 5-μm latex beads (107/ml; Interfacial Dynamics) with IL-2 capture Ab (2 μg/ml; BD Pharmingen) in PBS for 2 h at 37°C. Beads were washed with PBS and blocked for 30 min at 37°C culture medium. IL-2 capture beads (300,000 per test) were incubated with 100 μl of culture supernatant or titrated IL-2 standards (R&D Systems) for 2 h at 37°C, then washed in culture medium. Bead-bound IL-2 was detected by adding PE-conjugated anti-IL-2 (0.1 μg/test; BD Pharmingen) for 30 min at 4°C, washing in PBS, and measuring fluorescence using a FACSCalibur flow cytometer (Becton Dickinson). IL-2 ELISA was performed by coating a NUNC Maxisorp 96-well plate with capture Ab overnight at 4°C. Wells were washed with PBS/0.05% Tween 20 and blocked for 1 h at room temperature, then incubated with 100 μl of culture supernatant and dilutions of IL-2 standard for 2 h at room temperature (eBiosciences). Wells were washed and detection Ab was added at for 1 h at room temperature. Following additional washes the plate was incubated with Avidin-HRP for 30 min at room temperature. IL-2 was detected by addition of tetramethylbenzidine substrate solution for 15 min, followed by neutralization with 1 M H3PO4. Absorbance was then measured at 450 nm on a Bio-Rad microplate reader.
Results
p27kip1promotes early G1 to S phase progression in activated CD4+ T cells
To experimentally modulate p27kip1 protein levels during CD4+ T cell activation, we used a gene dosage approach in which T cells from wild-type (p27kip1+/+) mice represent normal levels of p27kip1 protein, while T cells from mice with a homozygous deletion of the entire p27kip1 coding region (7) (p27kip1–/–) completely lack p27kip1 protein. In addition, we took advantage of the fact that T cells from mice hemizygous for the p27kip1 gene (p27kip1+/–) contain approximately half the wild-type amount of p27kip1 protein (Ref.8 and see Fig. 2). We subjected these T cells to polyclonal TCR ligation in the presence of physiologic costimulatory signals in vitro, and measured the kinetics of DNA synthesis and mitosis over a 4-day period (Fig. 1, A and B). In wild-type cultures, the majority of CD4+ T cells had begun to synthesize DNA by 36 h after stimulation, and most of these cells were undergoing the first cell division (Fig. 1B and data not shown). By 48 h, the majority of the CD4+ T cells had progressed into a second S phase, and a significant portion was passing through mitosis for a second time (Fig. 1, A and B). If p27kip1 were acting primarily as a CDK inhibitor during this early phase of the response, p27kip1–/– CD4+ T cells should exhibit enhanced DNA synthesis and cell division. In contrast, we observed a 10–20% decrease in the frequency of p27kip1–/– CD4+ T cells entering S phase and mitosis at these early time points compared with p27kip1+/+ CD4+ T cells (Fig. 1, A and B). These results demonstrate that G1 to S phase progression during the first one to two mitoses is delayed in the complete absence of p27kip1.
FIGURE 2. p27kip1 protein dynamics in activated lymphocytes from wild-type and p27kip1–hemizygous mice. Spleen and LN cells from p27kip1+/+ or p27kip1+/– mice were stimulated in vitro with anti-CD3 (1 μg/ml) and either control Ig (5 μg/ml) or CTLA-4 Ig (5 μg/ml). A, Cultures were harvested at the indicated time points, the live cells were purified over Ficoll, and whole cell extracts were subjected to immunoblot analysis for p27kip1 (0.5 x 106 cell equivalents per lane). B, All blots were stripped and reprobed for actin as a loading control, and actin and p27kip1 signals were quantified by densitometry. p27kip1 protein levels are shown normalized against actin protein levels. The results are representative of two separate experiments.
FIGURE 1. Effect of p27kip1 on DNA synthesis and cell division at the single cell level in populations of activated CD4+ T cells. A, CFSE-labeled spleen and lymph node (LN) cells from p27kip1+/+, p27kip1+/–, or p27kip1–/– mice were stimulated in vitro with soluble anti-CD3 (0.25 μg/ml). Cultures were subjected to 6-h pulses of BrdU at 30, 42, 66, or 90 h, and harvested at 36, 48, 72, or 96 h, respectively. Density plots are gated on CD4+ T cells. Horizontal lines indicate the channel below which 98% of non-BrdU-pulsed cells fluoresce, and vertical lines indicate discrete cell divisions as determined by CFSE fluorescence (with the zero division on the far right). B, The mean ± SEM BrdU-positive CD4+ cells in each culture from A is depicted in the left panel, and means are also inset in each plot in A. The right panel depicts the mean BrdU-positive CD4+ cells from the p27kip1+/– and p27kip1–/– normalized to the means of corresponding p27kip1+/+ cultures. C, In a separate experiment, CFSE-labeled spleen and LN cells of each genotype were stimulated with the indicated doses of anti-CD3 Ab, and the total number of cell divisions accumulated within the CD4+ T cell subset by 72 h was calculated from the CFSE profiles. D, Depicts the same experiment as in A, only that cultures were stimulated in the presence of CTLA-4 Ig (5 μg/ml). E, Depicts the raw and normalized mean ± SEM BrdU-positive CD4+ cells in each culture from D. F, Depicts the same experiment as in C, only that cultures were stimulated in the presence of CTLA-4 Ig (5 μg/ml). These data are representative of at least three separate experiments.
Interestingly, CD4+ T cells from mice with only one functional copy of the p27kip1 gene behaved differently than CD4+ T cells from either p27kip1–/– or p27kip1+/+ mice. At 36 h postactivation, p27kip1+/– CD4+ T cells exhibited a similar delay in S phase entry as experienced by the p27kip1–/– cells. However, at 48 h, p27kip1+/– cultures showed a 40% increase in the frequency of CD4+ T cells synthesizing DNA compared with p27kip1–/– cultures, and a 20% increase compared with p27kip1+/+ cultures (Fig. 1, A and B). These results, together with the decrease in early cell cycle progression by CD4+ T cells that completely lack p27kip1, suggest that p27kip1 can actually promote cell cycle progression at this stage of the response.
p27kip1 restricts CD4+ T cell division late in the proliferative response
In these in vitro stimulation cultures, the rate of DNA synthesis by CD4+ T cells dropped precipitously after 48 h, such that roughly 15% of the wild-type cells were in S phase at 72 h, and <10% were still synthesizing DNA at 96 h (Fig. 1, A and B). The rate of DNA synthesis also dropped over time in cultures of p27kip1–/– CD4+ T cells, however, nearly twice as many p27kip1–/– cells were still actively cycling at 72 and 96 h as compared with p27kip1+/+ cells (Fig. 1, A and B). p27kip1+/– CD4+ T cells showed an intermediate phenotype, with frequencies of cycling cells similar to wild-type at 72 h, but increased by 30% compared with wild-type at 96 h. The increased rate of late DNA synthesis led to as much as a 30% increase in the degree of clonal expansion by both p27kip1–/– and p27kip1+/– CD4+ T cells at lower doses of anti-CD3 stimulation (Fig. 1C). These data show that p27kip1 functions late in the CD4+ T cell response to restrict the number of actively dividing cells, and thereby functions to oppose clonal expansion.
p27kip1 governs the requirement for CD28 costimulation
T cells stimulated in the absence of CD28 costimulation exhibit elevated levels of p27kip1 (4, 12), produce very little IL-2 (21), and fail to divide and expand efficiently (20). If p27kip1 plays an important role in restricting cell cycle progression under these conditions, then elimination or reduction of p27kip1 expression should result in CD28-independent T cell proliferation. To test this, we subjected T cells from p27kip1+/+, p27kip1+/–, or p27kip1–/– mice to anti-CD3 stimulation in the presence of CTLA-4 Ig. Consistent with previous studies (20, 22), blockade of CD28/B7 interactions using CTLA-4 Ig resulted in an abortive proliferative response by wild-type CD4+ T cells (Fig. 1, D and E). At 36 h, the number of CD4+ T cells entering S phase and progressing through the first mitosis in CTLA-4 Ig-treated cultures was comparable to that in CD28-costimulated cultures (Fig. 1E). However, the frequency of CD4+ T cells synthesizing DNA in CTLA-4 Ig-treated cultures decreased rapidly by 48 h to less than half of that in untreated cultures, and fell below 10% by 72 h (Fig. 1, D and E). In the absence of CD28 costimulation, p27kip1–/– CD4+ T cells exhibited a 20% proliferative advantage over wild-type cells at 48 h, which rose to a 75% advantage by 72 h (Fig. 1, D and E). Similarly, p27kip1+/– CD4+ T cells exhibited an early 80% increase in the number of cells in S phase at 48 h, and this increased rate of DNA synthesis was maintained throughout the rest of the response (Fig. 1, D and E). In fact, the rate of DNA synthesis at 72 and 96 h by both p27kip1–/– and p27kip1+/– CD4+ T cells was considerably less affected by CD28-costimulatory blockade than wild-type CD4+ T cells. Blockade of CD28/B7 interactions resulted in a 3- to 4-fold decrease in clonal expansion by wild-type CD4+ T cells (Fig. 1, C and F), however, CTLA-4 Ig treatment resulted in only a 25% decrease in the proliferation of p27kip1–/– CD4+ T cells (Fig. 1, C and F). p27kip1-hemizygous CD4+ T cells also exhibited a lesser but significant increase in cumulative cell division under these conditions (Fig. 1F). These data demonstrate that CD28 costimulation is not strictly required for the maintenance of clonal expansion by CD4+ T cells in the absence of p27kip1, and suggest that wild-type levels p27kip1 participate in determining the CD28-costimulatory requirement for a CD4+ T cell response.
A 2-fold reduction in the level of p27kip1 protein is sufficient to promote normal cell cycle progression by CD4+ T cells
We were next interested in why p27kip1+/– CD4+ T cells showed enhanced proliferative responses, which in some cases equaled that of the p27kip1–/– cells. This could be explained in part by the enhanced rate of early S phase entry exhibited by these cells, however, the hemizygous cells also showed enhanced proliferation late in the response, when p27kip1 acts as a negative regulator of cell cycle progression. Another explanation may be that these cells maintain low levels of p27kip1 under conditions where wild-type cells contain high levels of this CDK inhibitor. To address this, we compared the kinetics of p27kip1 protein expression in p27kip1+/– and p27kip1+/+ T cells. Resting lymphocytes from wild-type mice express a relatively high level of p27kip1 (Fig. 2, A and B). Immediately following productive activation, the amount of p27kip1 protein drops slightly, but quickly rebounds and this high level is maintained throughout the first 48 h of the response. Subsequently, p27kip1 expression decreases to roughly half the starting level by 72 h (Fig. 2, A and B). When CD28 costimulation is inhibited during stimulation, wild-type cells down-regulate p27kip1 slightly, but by 36 h, p27kip1 begins to accumulate until at 96 h the CTLA-4 Ig-treated cells express 3-fold more than untreated cells (Fig. 2, A and B). Resting lymphocytes from p27kip1+/– mice contain approximately half the amount of p27kip1 protein as lymphocytes from p27kip1+/+ animals, and this level is maintained after activation and throughout the duration of the proliferative response (Fig. 2, A and B). Remarkably, this low level of p27kip1 expression is even maintained when CD28 costimulation is blocked during activation (Fig. 2, A and B). Under these conditions, the hemizygous cells express 3- to 5-fold less p27kip1 than wild-type cells. These results show that the 2-fold reduction in p27kip1 protein that normally occurs after productive T cell activation represents a highly significant functional difference, leaving the cells with a high enough level of p27kip1 to promote S phase progression during the first cell division, but not high enough to restrict clonal expansion later in the response.
p27kip1 establishes a threshold for mitogenic signaling through the IL-2R
One possible mechanism for the enhanced CD4+ T cell proliferation in the absence of CD28 costimulation could be that p27kip1 deficiency leads to deregulated IL-2 production. Forced overexpression of p27kip1 has been shown to inhibit IL-2 transcription in T cell lines (12, 19), and therefore, genetic elimination of p27kip1 may be expected to augment IL-2 production. However, we found that naive p27kip1–/– or p27kip1+/– CD4+ T cells did not secrete greater amounts of IL-2 during primary stimulation as compared with wild-type cells (Fig. 3A). Furthermore, blockade of CD28 costimulation with CTLA-4 Ig resulted in a 100-fold reduction in the amount of IL-2 available in the cultures of both wild-type and p27kip1-mutant T cells (Fig. 3A). These data suggest that physiologic levels of p27kip1 do not directly influence IL-2 expression, in agreement with previous studies (19), and that the enhanced proliferative capacity of CD4+ T cells that lack or express reduced levels of p27kip1 is not a result of increased IL-2 secretion.
FIGURE 3. Effect of p27kip1 on T cell IL-2 production and IL-2 responsiveness. A, Spleen and LN cells from p27kip1+/+ () and p27kip1–/– () mice were stimulated in vitro with anti-CD3 (1 μg/ml) in the presence of anti-CD28 (0.5 μg/ml) or CTLA-4 Ig (10 μg/ml). IL-2 secreted into the supernatants during the first 24 h was measured by FLISA (left panel). In a separate experiment, IL-2 secretion by p27kip1+/+ () and p27kip1+/– () cells was also compared (right panel). IL-2 secretion at 48 and 72 h showed similar trends between p27kip1+/+, p27kip1–/–, and p27kip1+/– cells (data not shown). The data shown are representative of two experiments for each genotype. B, Lymphocytes from p27kip1+/+, p27kip1–/–, and p27kip1+/– mice were cultured as in A, and the expression of CD25, CD122, and CD132 on the CD4+ T cell subset was assessed by flow cytometry. The 48-h timepoint is shown. C, CFSE-labeled spleen and LN cells from p27kip1+/+ (filled symbols) and p27kip1–/– (open symbols) mice were primed in vitro with anti-CD3 (1 μg/ml) and anti-CD28 (0.5 μg/ml), for 24 h. Cultures were washed to remove the TCR/CD28 stimulus, and the cells were rested for overnight. IL-2 was added at the indicated concentrations, and cell division was quantified 72 h later. D, Spleen and LN cells from p27kip1+/+ and p27kip1+/– mice were primed and rested as in C, and IL-2 was added at the indicated concentrations. After 48 h, the cultures were harvested, live cells were purified over Ficoll, and whole cell extracts were subjected to immunoblot analysis for p27kip1 (0.5 x 106 cell equivalents per lane). All data are representative of two separate experiments.
We also tested whether the lack or reduced levels of p27kip1 protein results in enhanced expression of a high affinity IL-2R by measuring the surface expression of CD25 (IL-2R -chain), CD122 (IL-2R -chain), and CD132 (IL-2R -chain) on p27kip1+/+, p27kip1+/–, and p27kip1–/– CD4+ T cells. In the presence of CD28 costimulation, all receptor chains were up-regulated to a similar degree on CD4+ T cells of all three genotypes (Fig. 3B, upper panels). Blockade of CD28 costimulation during TCR stimulation hampered the induction of these receptor subunits, particularly CD25 and CD122, and affected CD4+ T cells of all three genotypes to a similar degree (Fig. 3B, lower panels). These data suggest that the enhanced proliferative capacity of CD4+ T cells that lack or express reduced levels of p27kip1 is not a result of increased IL-2R expression.
Therefore, we hypothesized that the enhanced proliferation of p27kip1–/– CD4+ T cells under limiting IL-2 conditions may be due to an increased sensitivity to the mitogenic signals provided by this growth factor. To test this possibility, p27kip1+/+ and p27kip1–/– T cells were primed for 24 h with agonistic Abs against CD3 and CD28, and stimulation was subsequently stopped by washing in fresh medium. The cells were then exposed to various fixed concentrations of IL-2 and cultured for an additional 72 h. At high concentrations of IL-2 (5–20 ng/ml), wild-type and p27kip1-deficient CD4+ T cells exhibited comparable proliferation (Fig. 3B), consistent with the comparable response of these two cell populations to high-dose polyclonal stimulation (Fig. 1C). However, at lower concentrations of IL-2, such as those found in cultures of T cells stimulated in the absence of CD28 costimulation (<1 ng/ml; Fig. 3A), p27kip1-deficient CD4+ T cells accumulated significantly more cell divisions throughout the culture period than wild-type cells (Fig. 3B). Primed CD4+ T cells from p27kip1+/– mice also showed enhanced sensitivity to IL-2 signaling, as these cells were able to efficiently down-regulate p27kip1 and proliferate (data not shown) at doses of IL-2 that did not induce degradation in wild-type cells (Fig. 3C). These results show that p27kip1 regulates not only the kinetics of cell cycle entry by CD4+ T cells in response to IL-2 (15), but also the cumulative number of cell divisions and the degree of clonal expansion achieved.
Our data suggest that T cells experience greater mitogenic signaling from a given amount of IL-2 in the absence of p27kip1 than when p27kip1 is present. If this were the case, p27kip1-deficient cells should be less sensitive to proximal inhibitors of IL-2R-mediated signal transduction. To test this hypothesis, wild-type, p27kip1-deficient, or p27kip1-hemizygous CD4+ T cells were stimulated in the presence of rapamycin, an immunosuppressive drug that inhibits T cell proliferation by reportedly blocking IL-2-mediated down-regulation of p27kip1 (10). Rapamycin reduced the degree of clonal expansion by wild-type CD4+ T cells in response to TCR/CD28 by >2-fold (Fig. 4A), and this was associated with a 3- to 6-fold increase in p27kip1 protein levels compared with control-treated cells (Fig. 4, B and C). Rapamycin likewise strongly inhibited IL-2-induced degradation of p27kip1 in primed p27kip1+/+ cells (Fig. 4D). However, cell division by p27kip1-deficient CD4+ T cells was reduced by only 25% in the presence of rapamycin, with these cultures accumulating nearly twice as many cell divisions than cultures of wild-type T cells (Fig. 4A). The proliferation of p27kip1+/– CD4+ T cells was also significantly resistant to rapamycin (Fig. 4A), and this corresponded to a 3- to 6-fold reduction in the amount of p27kip1 detected throughout the response as compared with rapamycin-treated wild-type cells (Fig. 4, B and C). Rapamycin was also much less effective at blocking IL-2-induced p27kip1 degradation in p27kip1+/– cells (Fig. 4D).
FIGURE 4. Effect of p27kip1 on rapamycin sensitivity. A, CFSE-labeled spleen and LN cells from p27kip1+/+ (filled symbols), p27kip1+/– (gray symbols), or p27kip1–/– (open symbols) mice were stimulated with the indicated doses of anti-CD3 Ab in the presence of rapamycin (10 ng/ml) or vehicle (DMSO). Cells were cultured for 72 h, and the total number of cell divisions accumulated within the CD4+ T cell subset of each culture was calculated from the CFSE profiles. B and C, Spleen and LN cells from p27kip1+/+ or p27kip1+/– mice were stimulated and treated with rapamycin as in A, and whole cell extracts were subjected to immunoblot analysis for p27kip1 and actin as described in Fig. 2. D, Spleen and LN cells from p27kip1+/+ and p27kip1+/– mice were primed and rested as in Fig. 3C, and IL-2 was added at the indicated concentrations in the presence or absence of rapamycin. After 48 h, whole-cell extracts were subjected to immunoblot analysis for p27kip1. All data are representative of two experiments.
These data show that elimination or reduction of p27kip1 protein renders CD4+ T cells less susceptible to rapamycin, and formally establish p27kip1 as a downstream mediator of the antiproliferative effect of this agent. However, these data also specifically demonstrate a rapamycin-sensitive component of the proliferative response in p27kip1-deficient CD4+ T cells (Fig. 4A), suggesting that a significant degree of rapamycin-mediated inhibition occurs independently of p27kip1. Our data, together with the results of other models in which p27kip1 levels have been modulated (14, 15, 23), suggest that p27kip1 acts in CD4+ T cells as an intracellular sensor of available growth factor that helps to set the threshold for mitogenic signaling and cell cycle progression.
Primed CD4+ T cells exhibit enhanced effector function in the absence of p27kip1
Cell division promotes differentiation in CD4+ T cell populations (1, 2, 3). Because p27kip1-deficient CD4+ T cells tend to proliferate more than wild-type cells during the primary response, we hypothesized that these cells might exhibit enhanced effector function upon restimulation. To test this, we primed wild-type or p27kip1–/– lymphocytes with anti-CD3 for 72 h as in Fig. 1, and allowed the cultures to come to rest in medium for 24 h. CD4+ T cells were then restimulated through TCR/CD28, and IL-2 secretion was measured at 24 h. p27kip1-deficient CD4+ T cells produced 4-fold more IL-2 than CD4+ cells upon restimulation (Fig. 5A, first and third columns), and showed a slightly increased frequency of cells entering S phase 24 h later (Fig. 5B, first and third panels in the middle row). These results suggest that p27kip1-deficient CD4+ T cells are better able to respond when re-encountering Ag.
FIGURE 5. Effect of p27kip1 on CD4+ T cell effector function. A, Spleen and LN cells from p27kip1+/+ () and p27kip1–/– () mice were stimulated in vitro with anti-CD3 (1 μg/ml) in the presence of control Ig (10 μg/ml) or CTLA-4 Ig (10 μg/ml) for 72 h and rested for 24 h in culture medium. CD4+ T cells were purified from each culture by negative selection and an equal number were restimulated with immobilized anti-CD3 (5 μg/ml) and anti-CD28 (5 μg/ml). Supernatants were collected after 24 h and IL-2 secretion was measured by FLISA. B, p27kip1+/+ or p27kip1–/– CD4+ T cells cultured and purified as in A were restimulated with medium alone, immobilized anti-CD3 (2 μg/ml), or IL-2 (20 U/ml) for 48 h. BrdU was added to the cultures 6 h before harvest and DNA synthesis (BrdU) and DNA content (7-aminoactinomycin D (7-AAD)) were assessed by flow cytometry. C, CD4+ T cells from wild-type or hemizygous (p27kip1+/–) mice were cultured, purified and restimulated as in A, and proliferation was assessed by BrdU incorporation as in B. D, Additionally, purified CD4+ T cells from cultures as in C were assessed for p27kip1 content by immunoblot analysis (0.5 x 106 cell equivalents per lane). All data shown are representative of two to three separate experiments.
T cell proliferation also opposes anergy induction (4, 24, 25), and T cell anergy has been associated in several models with elevated levels of p27kip1 (4, 12, 24, 26). These results have led to the hypothesis that p27kip1 may directly participate in the induction or maintenance of anergy in T cells. To test this, we primed wild-type or p27kip1–/– lymphocytes with anti-CD3 as above, but blocked B7/CD28 interactions with CTLA-4 Ig to provide a tolerizing stimulus. Consistent with previous studies (27, 28), wild-type CD4+ T cells subjected to this tolerizing stimulus secreted 3- to 4-fold less IL-2 when restimulated through TCR/CD28 compared with cells primed in the presence of CD28 costimulation (Fig. 5A,
). The amount of IL-2 produced by CTLA-4 Ig-treated p27kip1–/– CD4+ T cells was reduced only 2-fold compared with fully primed p27kip1–/– cells (Fig. 5A,
). Remarkably, this amount of IL-2 was >7-fold the levels produced by anergic wild-type cells, and >2-fold the levels produced by wild-type cells primed in the presence of costimulation (Fig. 5A,
vs
). To test whether the enhanced production of IL-2 by the restimulated p27kip1–/– T cells leads to increased proliferation during the secondary culture, we stimulated wild-type or p27kip1–/– cells in the presence or absence of CD28 costimulation as in Fig. 5A. Purified CD4+ T cells from these cultures were restimulated with medium alone, immobilized anti-CD3, or IL-2 for 48 h, and DNA synthesis by individual cells was measured. The frequency of p27kip1+/+ cells proliferating in response to TCR ligation was reduced by 50% if costimulation was blocked during the primary stimulus (Fig. 5B, compare first and third middle panels), consistent with their anergic phenotype of reduced IL-2 production (Fig. 5A). Conversely, the proliferation of p27kip1–/– T cells was not adversely affected by costimulatory blockade (Fig. 5B, compare second and fourth middle panels), resulting in a 2-fold increase in the frequency of proliferating cells as compared with p27kip1+/+ CD4+ T cells stimulated under the same conditions (Fig. 5B, compare third and fourth middle panels). As expected, the response of all these cells to IL-2 was intact, with 30% or more of the cells within each population having synthesized DNA during the period assessed (Fig. 5B, lower panels). These data show that p27kip1–/– CD4+ T cells exhibit significantly enhanced effector function even when primed in the absence of CD28 costimulation, and suggest that p27kip1 is required for the full development of anergy in this model.
CD4+ T cells from p27kip1+/– mice also exhibit a proliferative advantage in the primary response in our model, therefore, we predicted that these cells might likewise exhibit a reduced susceptibility to anergy induction. To test this, we stimulated wild-type or p27kip1+/– cells in the presence or absence of CD28 costimulation as in Fig. 5A. Purified CD4+ T cells from these cultures were restimulated with medium alone, immobilized anti-CD3, or IL-2 for 48 h, and DNA synthesis by individual cells was measured. Similar to p27kip1–/– cells, CD3/28-primed CD4+ T cells from p27kip1+/– mice showed a slight increase in proliferation upon restimulation, as compared with wild-type cells (Fig. 5C, first and third panels in the middle row). However, unlike p27kip1–/– cells, p27kip1+/– CD4+ T cells from CD3/CTLA-4Ig-primed cultures exhibited a strong proliferative defect that was similar in magnitude to that of anergized wild-type CD4+ T cells (Fig. 5C, second and fourth panels in the middle row). This suggests that p27kip1+/– CD4+ T cells are as susceptible to anergy induction as cells with two functional copies of the p27kip1 gene. Because p27kip1+/– CD4+ T cells are still able to induce p27kip1 protein expression, we hypothesized that the induction of p27kip1 during restimulation might enforce the anergic state in these cells. Consistent with the experiments depicted in Fig. 2, resting p27kip1+/– CD4+ T cells contained less p27kip1 protein than wild-type cells after primary stimulation (Fig. 5D, lanes 1–4). However, upon restimulation, anergized p27kip1+/– cells failed to down-regulate p27kip1 as compared with effector cells (Fig. 5D, lanes 5–8), and instead, p27kip1 protein accumulated in these cells to levels comparable to that expressed by wild-type anergic T cells (Fig. 5D, lanes 7 and 8). These data may explain why T cells with one functional copy of the p27kip1 gene are still susceptible to anergy, and further support the idea that p27kip1 protein levels are an important factor that determines the responsiveness of CD4+ T cells to mitogenic stimulation.
Discussion
p27kip1 was first identified biochemically in cell lines as an inhibitor of G1 and S phase cyclin-CDK complexes, particularly cyclin D-CDK4, cyclin E-CDK2, and cyclin A-CDK2 complexes (29, 30). Subsequent genetic studies have confirmed a general role for p27kip1 in controlling cell proliferation throughout the development of multiple tissues, including the thymus and the mature T cell pool (7, 8, 9). The contribution of p27kip1 to peripheral T cell immune responses, however, has been less clear. Down-modulation of p27kip1 is associated with T cell activation, and while this process has generally been considered to be a crucial event for T cell proliferation (10, 14, 15, 31), some studies have suggested that peripheral T cell responses proceed normally in the absence of p27kip1 (8, 9, 17, 19) Earlier studies of p27kip1 expression and function in T lymphocytes were, for the most part, limited to very short time frames and did not explore the role of p27kip1 in the context of different T cell stimuli. Therefore, we have used both a full and a partial loss-of-function approach to carefully examine the role of p27kip1 in regulating the kinetics of proliferation and the development of effector function in populations of mature T lymphocytes. Our studies bring together previously disparate findings by revealing that p27kip1 can serve dual, opposing roles in T cell cycle progression. We establish that p27kip1 functions physiologically within a 2-fold range of protein concentration to first promote, not inhibit, G1 to S phase progression during the initial phase of T cell proliferation. Following the peak of the response, p27kip1 then functions as a negative regulator of T cell clonal expansion by promoting exit from the cell cycle. Finally, we show that the net effect of p27kip1 is to oppose T cell effector function, and this negative regulatory role is required for the full effect of tolerizing stimuli and immunosuppressive therapies.
How can p27kip1 act as both a positive and a negative regulator of T cell cycle progression? The mechanism by which p27kip1 inhibits cell cycle progression is fairly clear. p27kip1 can bind to and inhibit the kinase activity of cyclin D-, E-, and A-containing CDK complexes (32). It is important, however, to note that p27kip1 does not play an obligatory role in regulating the maximal proliferative "output" in response to mitogenic stimuli in our system (i.e., fully p27kip1–deficient T cells do not proliferate indefinitely and uncontrollably). The role of p27kip1 is subtle when T cells receive an optimal combination of stimuli, but becomes apparent when mitogenic signals are limiting. Our data suggest that p27kip1 functions in T cells to set the threshold amount of growth factor required to initiate G1 to S phase progression. For instance, our IL-2 dose response and rapamycin data demonstrate that p27kip1–/– and p27kip1+/– CD4+ T cells are roughly 10-fold more sensitive to growth factor receptor signaling (Figs. 3 and 4). The simplest interpretation of these results is that p27kip1+/– cells begin with an amount of p27kip1 protein that approximates the lowest possible steady-state level achievable in response to normal mitogenic signals. This level can be reduced further (i.e., approximating a complete absence of p27kip1 as in p27kip1–/– T cells), but in our system this is only achieved by the addition of exogenous IL-2. Once this growth factor signaling threshold has been reached, other downstream factors take over to limit the proliferative response.
How can p27kip1, a CDK inhibitor, function to promote cell cycle progression? D-type cyclins are the first induced upon T cell activation, and cyclin D-CDK4/6 complexes are required for progression from G0 to G1 (33, 34). Cyclin D-containing CDK, unlike cyclin E-, A- and B-containing complexes, are subject to an additional level of regulation by the inhibitor of CDK4 family of CDK inhibitors (35, 36, 37, 38). These members inhibit only cyclin D-type CDK, and block the assembly of cyclin D-CDK4/6 complexes by binding to free CDK4 and -6 directly (39, 40, 41). Conversely, p27kip1 cannot efficiently bind to free cyclins or CDK, but rather bind stably in a trimolecular complex (30, 42, 43). Consistent with this, studies in non-T cells have demonstrated that the accumulation of cyclin D-CDK4/6 complexes is impaired in the absence of p27kip1 (42, 44, 45). In this way, p27kip1 actually promotes the assembly and stability of cyclin D-CDK4/6 complexes, and protects CDK4/6 from regulation by inhibitor of CDK4 family members (32). Although p27kip1 levels remain high, the assembled cyclin-CDK complexes are inhibited. However, once growth factor signaling reaches sufficient strength, p27kip1 is targeted for degradation, and the already-formed cyclin D-CDK4/6 complexes are freed to phosphorylate downstream targets and promote cell cycle progression. Residual p27kip1 protein then acts to oppose the activity of cyclin E- and -A-CDK complexes. This scenario explains how these two opposing biochemical functions of p27kip1 can operate concurrently during T cell activation, and how the apparent biological effect of p27kip1 can switch from positive to negative later in the response when cyclin E- and A-containing CDK complexes are predominantly expressed and active. As we observe in our current study, a 2-fold reduction in initial p27kip1 levels (i.e., as in p27kip1+/– cells) is sufficient to significantly enhance cell cycle progression, potentially by influencing this balance between p27kip1-mediated cyclin-CDK assembly vs assembly and inhibition. Additional biochemical studies will be required to address this hypothesis.
An additional intriguing finding from our studies is that CD4+ T cells that lack p27kip1 exhibit enhanced effector function during restimulation. T cell effector function and anergy avoidance are linked to cell division (1, 2, 3, 4), therefore derepression of the proliferative response might be predicted to result in enhanced T cell effector function. Therefore, these current studies provide important support for the role of cell cycle progression in driving T cell differentiation. Significantly, the enhanced effector function of p27kip1–/– CD4+ T cells was still observed even under conditions that would induce anergy in normal T cells. Although the loss of p27kip1 in CD4+ T cells does not render either primary cell division or the recall response completely insensitive to CD28 costimulation, the fact that p27kip1–/– CD4+ T cells produce more IL-2 and proliferate more following a tolerizing stimulus than wild-type T cells do under full stimulation leads us to conclude that T cells are refractory to anergy induction in the absence of p27kip1.
Our results demonstrate that, in the absence of p27kip1, CD4+ T cells are less susceptible to the effects of CTLA-4 Ig, and formally establish p27kip1 as a downstream mediator of the antiproliferative effect of CD28-costimulatory blockade. An extension of this finding is that p27kip1 participates in setting the CD28-costimulatory requirement for T cell clonal expansion under normal circumstances, an idea that was previously suggested by correlative studies (46, 47). Under physiologic conditions, p27kip1 does not govern the requirement for CD28 costimulation by influencing the amount of IL-2 produced during a primary response (Fig. 3A). Instead, our data using rapamycin, and previous studies by Powell et al. (19) using IL-2-deficient T cells, suggest that the link between CD28 and p27kip1 is indirect and controlled instead at the level of IL-2 responsiveness. The amount of costimulation needed is a direct reflection of the amount of IL-2 a cell needs to produce to overcome p27kip1-mediated inhibition of proliferation.
The net effect of p27kip1, through its capacity to inhibit cell cycle progression, is apparently to oppose effector differentiation and to promote the induction of anergy in CD4+ T cells. This role of p27kip1 is not limited to CD4+ T cells, as we have also observed that p27kip1–/– CD8+ T cells are hyperproliferative and a higher frequency of CD8+ cells differentiate into IFN--producing effector cells as compared with wild-type CD8+ cells (our unpublished observations). The importance of this CDK inhibitor in the regulation of peripheral T cell responses is further supported by our preliminary studies in a cardiac allograft model, which have demonstrated that p27kip1–/– mice are significantly resistant to tolerance induction in vivo by costimulatory blockade as compared with wild-type mice (our unpublished observations). By acting as a sensor of available growth factor and determining the sensitivity of CD4+ T cells to tolerizing signals, p27kip1 is critical in determining the extent of clonal expansion and whether the outcome of encounter with Ag is tolerance or immunity.
Disclosures
The authors have no financial conflict of interest.
Acknowledgments
We thank Drs. J. Powell, V. Boussiotis, and L. Turka for helpful discussions.
Footnotes
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.
1 This work was supported by National Institutes of Health (NIH) Grants K01-DK02771 and R01-AI054643 (to A.D.W.). A.D.W. is a member of the Joseph Stokes, Jr. Research Institute and the Biesecker Pediatric Liver Disease Center at The Children’s Hospital of Philadelphia, and E.A.R. is supported by NIH Training Grant T32-AR007442-18.
2 Address correspondence and reprint requests to Dr. Andrew D. Wells, 916F Abramson Research Center, 3615 Civic Center Boulevard, Philadelphia, PA, 19104. E-mail address: adwells{at}mail.med.upenn.edu
3 Abbreviations used in this paper: CDK, cyclin-dependent kinase; FLISA, fluorescence-linked immunosorbent assay; LN, lymph node.
Received for publication July 28, 2004. Accepted for publication December 22, 2004.
References
Bird, J. J., D. R. Brown, A. C. Mullen, N. H. Moskowitz, M. A. Mahowald, J. R. Sider, T. F. Gajewski, C. R. Wang, S. L. Reiner. 1998. Helper T cell differentiation is controlled by the cell cycle. Immunity 9:229.
Gett, A. V., P. D. Hodgkin. 1998. Cell division regulates the T cell cytokine repertoire, revealing a mechanism underlying immune class regulation. Proc. Natl. Acad. Sci. USA 95:9488
Gudmundsdottir, H., A. D. Wells, L. A. Turka. 1999. Dynamics and requirements of T cell clonal expansion in vivo at the single-cell level: effector function is linked to proliferative capacity. J. Immunol. 162:5212.
Wells, A. D., M. C. Walsh, D. Sankaran, L. A. Turka. 2000. T cell effector function and anergy avoidance are quantitatively linked to cell division. J. Immunol. 165:2432.
Morgan, D. O.. 1995. Principles of CDK regulation. Nature 374:131.[Medline]
Vidal, A., A. Koff. 2000. Cell-cycle inhibitors: three families united by a common cause. Gene 247:1.
Fero, M. L., M. Rivkin, M. Tasch, P. Porter, C. E. Carow, E. Firpo, K. Polyak, L. H. Tsai, V. Broudy, R. M. Perlmutter, et al 1996. A syndrome of multiorgan hyperplasia with features of gigantism, tumorigenesis, and female sterility in p27kip1-deficient mice. Cell 85:733.
Kiyokawa, H., R. D. Kineman, K. O. Manova-Todorova, V. C. Soares, E. S. Hoffman, M. Ono, D. Khanam, A. C. Hayday, L. A. Frohman, A. Koff. 1996. Enhanced growth of mice lacking the cyclin-dependent kinase inhibitor function of p27kip1. Cell 85:721
Kwon, T. K., M. A. Buchholz, P. Ponsalle, F. J. Chrest, A. A. Nordin. 1997. The regulation of p27Kip1 expression following the polyclonal activation of murine G0 T cells. J. Immunol. 158:5642.
Boussiotis, V. A., G. J. Freeman, P. A. Taylor, A. Berezovskaya, I. Grass, B. R. Blazar, L. M. Nadler. 2000. p27kip1 functions as an anergy factor inhibiting interleukin 2 transcription and clonal expansion of alloreactive human and mouse helper T lymphocytes. Nat. Med. 6:290.
Jackson, S. K., A. DeLoose, K. M. Gilbert. 2001. Induction of anergy in Th1 cells associated with increased levels of cyclin-dependent kinase inhibitors p21Cip1 and p27Kip1. J. Immunol. 166:952.
Tsukiyama, T., N. Ishida, M. Shirane, Y. A. Minamishima, S. Hatakeyama, M. Kitagawa, K. Nakayama. 2001. Down-regulation of p27Kip1 expression is required for development and function of T cells. J. Immunol. 166:304.
Zhang, S., V. A. Lawless, M. H. Kaplan. 2000. Cytokine-stimulated T lymphocyte proliferation is regulated by p27Kip1. J. Immunol. 165:6270.
Huleatt, J. W., J. Cresswell, K. Bottomly, I. N. Crispe. 2003. P27kip1 regulates the cell cycle arrest and survival of activated T lymphocytes in response to interleukin-2 withdrawal. Immunology 108:493.
Kovalev, G. I., D. S. Franklin, V. M. Coffield, Y. Xiong, L. Su. 2001. An important role of CDK inhibitor p18INK4c in modulating antigen receptor-mediated T cell proliferation. J. Immunol. 167:3285.
Luo, Y., S. O. Marx, H. Kiyokawa, A. Koff, J. Massague, A. R. Marks. 1996. Rapamycin resistance tied to defective regulation of p27Kip1. Mol. Cell. Biol. 16:6744
Powell, J. D., D. Bruniquel, R. H. Schwartz. 2001. TCR engagement in the absence of cell cycle progression leads to T cell anergy independent of p27kip1. Eur. J. Immunol. 31:3737
Wells, A. D., H. Gudmundsdottir, L. A. Turka. 1997. Following the fate of individual T cells throughout activation and clonal expansion: signals from T cell receptor and CD28 differentially regulate the induction and duration of a proliferative response. J. Clin. Invest. 100:3173.
Harding, F. A., J. G. McArthur, J. A. Gross, D. H. Raulet, J. P. Allison. 1992. CD28-mediated signalling co-stimulates murine T cells and prevents induction of anergy in T-cell clones. Nature 356:607.
Sperling, A. I., J. A. Auger, B. D. Ehst, I. C. Rulifson, C. B. Thompson, J. A. Bluestone. 1996. CD28/B7 interactions deliver a unique signal to naive T cells that regulates cell survival but not early proliferation. J. Immunol. 157:3909.
Mohapatra, S., D. Agrawal, W. J. Pledger. 2001. p27Kip1 regulates T cell proliferation. J. Biol. Chem. 276:21976.[Abstract/Free Full Text]
Wells, A. D., M. C. Walsh, J. A. Bluestone, L. A. Turka. 2001. Signaling through CD28 and CTLA-4 controls two distinct forms of T cell anergy. J. Clin. Invest. 108:895
DeSilva, D. R., K. B. Urdahl, M. K. Jenkins. 1991. Clonal anergy is induced in vitro by T cell receptor occupancy in the absence of proliferation. J. Immunol. 147:3261.
Gilbert, K. M., W. O. Weigle. 1993. Th1 cell anergy and blockade in G1 phase of the cell cycle. J. Immunol. 151:1245.
Wells, A. D., X. C. Li, Y. Li, M. C. Walsh, X. X. Zheng, Z. Wu, G. Nunez, A. Tang, M. Sayegh, W. W. Hancock, et al 1999. Requirement for T-cell apoptosis in the induction of peripheral transplantation tolerance. Nat. Med. 5:1303.
Wells, A. D., Q. H. Liu, B. Hondowicz, J. Zhang, L. A. Turka, B. D. Freedman. 2003. Regulation of T cell activation and tolerance by phospholipase c-1-dependent integrin avidity modulation. J. Immunol. 170:4127.
Toyoshima, H., T. Hunter. 1994. p27, a novel inhibitor of G1 cyclin-Cdk protein kinase activity, is related to p21. Cell 78:67.
Polyak, K., M. H. Lee, H. Erdjument-Bromage, A. Koff, J. M. Roberts, P. Tempst, J. Massague. 1994. Cloning of p27Kip1, a cyclin-dependent kinase inhibitor and a potential mediator of extracellular antimitogenic signals. Cell 78:59.
LaBaer, J., M. D. Garrett, L. F. Stevenson, J. M. Slingerland, C. Sandhu, H. S. Chou, A. Fattaey, E. Harlow. 1997. New functional activities for the p21 family of CDK inhibitors. Genes Dev. 11:847.
Cheng, M., P. Olivier, J. A. Diehl, M. Fero, M. F. Roussel, J. M. Roberts, C. J. Sherr. 1999. The p21Cip1 and p27kip1 CDK "inhibitors" are essential activators of cyclin D-dependent kinases in murine fibroblasts. EMBO J. 18:1571.
Appleman, L. J., A. Berezovskaya, I. Grass, V. A. Boussiotis. 2000. CD28 costimulation mediates T cell expansion via IL-2-independent and IL-2-dependent regulation of cell cycle progression. J. Immunol. 164:144.
Appleman, L. J., A. A. van Puijenbroek, K. M. Shu, L. M. Nadler, V. A. Boussiotis. 2002. CD28 costimulation mediates down-regulation of p27kip1 and cell cycle progression by activation of the PI3K/PKB signaling pathway in primary human T cells. J. Immunol. 168:2729.(Emily A. Rowell, Matthew )
Cell division drives T cell clonal expansion and differentiation, and is the result of concerted signaling from Ag, costimulatory, and growth factor receptors. How these mitogenic signals are coupled to the cell cycle machinery in primary T cells is not clear. We have focused on the role of p27kip1, a major cyclin-dependent kinase binding protein expressed by CD4+ T cells. Our studies using p27kip1 gene dosage demonstrate that early after activation, p27kip1 acts to promote, rather than inhibit, G1 to S phase progression within the first division cycle. However, throughout subsequent cell divisions p27kip1 behaves as a negative regulator, directly establishing the threshold amount of growth factor signaling required to support continued cell division. During this phase, signals from CD28 and IL-2R cooperate with the TCR to "tune" this threshold by inducing the degradation of p27kip1 protein, and we show that agents that block these pathways require elevated p27kip1 levels for their full antiproliferative activity. Finally, we show that p27kip1 opposes the development of CD4+ T cell effector function, and is required for the full development of anergy in response to a tolerizing stimulus. Our results suggest that p27kip1 plays a complex and important role in the regulation of cell division and effector function in primary CD4+ T cells.
Introduction
The role of cell division in a T cell response is 2-fold. This process not only drives clonal expansion, thereby increasing the frequency of T cells capable of recognizing a given Ag, but also drives effector T cell differentiation at the single cell level. Those T cells that undergo more rounds of division during a response are more likely to produce effector cytokines upon re-encounter with Ag (1, 2, 3, 4).
Signals from Ag, costimulatory, and growth factor receptors cooperate to induce cell cycle progression in quiescent T cells. An important step in linking these extracellular signals to the cell cycle machinery is the regulated and ordered synthesis of cyclins and their interaction with cyclin-dependent kinases (CDK)3 (5). The activity of these cyclin-CDK enzymes is further regulated by CDK inhibitors, proteins that bind to cyclin-CDK complexes and generally act to oppose cell cycle progression (6). One such protein capable of binding to multiple cyclin-CDK pairs is p27kip1. The importance of this CDK inhibitor in development is demonstrated by mice genetically deficient for p27kip1, which exhibit multiorgan hyperplasia, gigantism, and increased incidence of tumors (7, 8, 9).
In mature T lymphocytes, like in other tissues, p27kip1 is highly expressed in quiescent cells, but is down-regulated upon mitogenic stimulation (10, 11). If proliferation is inhibited during T cell activation, such as during immunosuppression or tolerance induction, p27kip1 levels instead remain elevated (4, 10, 12, 13). These observations have led to the notion that p27kip1 is an important negative regulator of G1 to S phase progression in T lymphocytes. Indeed, studies using forced overexpression have shown that excess p27kip1 can impair T cell proliferation in response to mitogenic stimuli, confirming its role as a negative regulator of the cell cycle when expressed at appropriately high levels (12, 14). However, studies using genetic deletion of p27kip1 have been unable to clearly establish whether this molecule has a significant influence over T lymphocyte proliferation and differentiation under physiologic circumstances. p27kip1-deficient mouse strains exhibit a striking increase in thymic cellularity, as well as a significant augmentation of splenic cellularity. However, when peripheral T lymphocyte function was examined, some studies found increased proliferative responses of p27kip1-deficient T cells in response to TCR ligation and/or IL-2R signaling (7, 15, 16), while other studies found no difference in proliferation between p27kip1-deficient and T cells (8, 9, 17). Likewise, attempts to establish genetically whether immunosuppressive drugs such as rapamycin act directly through p27kip1 have led to similar opposing conclusions, with some studies suggesting that p27kip1 is required for immunosuppression (18), while other studies found no role for this CDK inhibitor during immunosuppression (8, 9).
Cell cycle progression is required for efficient T cell differentiation and the avoidance of anergy (4). Multiple correlative studies have established an association between elevated p27kip1 levels and T cell anergy, but again, the role of p27kip1 in the induction or maintenance of this state has remained controversial. A forced overexpression approach suggested that p27kip1 could induce lasting tolerance in T cells through suppression of IL-2 gene expression (12), while separate studies using p27kip1-deficient T lymphocytes concluded that this CDK inhibitor was not required for anergy induction in the models tested (19). These contrasting results imply that the role of p27kip1 in a T cell response may be more complex than previously appreciated.
One potential problem with forced overexpression and full genetic deletion is that these situations do not faithfully recapitulate the more subtle and dynamic posttranslational mechanisms by which p27kip1 is controlled physiologically. We have used T cells that either contain wild-type levels of p27kip1, completely lack p27kip1, or contain significantly reduced levels of p27kip1 to study the extent to which signals from the TCR, CD28, and IL-2R are coupled to the cell cycle through this molecule. Our careful analysis of these T cell responses reveals that p27kip1 functions physiologically within a 2-fold range of protein levels. Low levels of p27kip1 can actually promote cell cycle progression by activated CD4+ T cells, while a higher concentration of p27kip1 restricts cell division, opposes effector function, and promotes anergy. Our results imply that p27kip1 acts as a "rheostat" that influences T cell differentiation by regulating cell cycle progression.
Materials and Methods
Mice
Wild-type C57BL/6 mice and C57BL/6 mice homozygous for deletion of p27kip1 (7) were obtained from The Jackson Laboratory and bred under specific pathogen-free conditions. Progeny were screened for the wild-type and null alleles by PCR as previously described (7). As previously observed, p27kip1+/– and p27kip1–/– mice in our colony contained 50–90% more peripheral lymphocytes as compared with age-matched wild-type mice (7), however, the ratio of B cells to T cells, CD4+ to CD8+ T cells, and naive (CD44lowCD62Lhigh) vs memory (CD44highCD62Llow) T cells did not differ significantly from wild-type littermates (data not shown).
Cell culture and reagents
Pooled spleen and lymph node cell suspensions were prepared using 100-μm nytex fabric, and CFSE (Molecular Probes) labeling was performed as described previously (20). Cells were cultured at 2 x 106/ml in medium (RPMI 1640 supplemented with 10% FBS, L-glutamine, penicillin, streptomycin, and 2-ME) with soluble anti-CD3 (145-2C11), anti-CD28 (37.51, 1 μg/ml), CTLA-4Ig (5 μg/ml), control human IgG (5 μg/ml), rapamycin (Calbiochem; 10 ng/ml in DMSO), or rmIL-2 (Roche) as indicated in the figure legends. In this system, accessory cells provide "signal 1" via FcR-mediated cross-linking of TCR-bound anti-CD3 Ab, and also provide the costimulatory signals required for a physiologic proliferative response. For anergy assays, primed CD4+ cells were rested for 24 h in medium, purified, and restimulated for 24–48 h with anti-CD3 and anti-CD28, at concentrations indicated in figure legends, immobilized on either flat-bottom 96-well plates or on latex beads (Interfacial Dynamics). DNA synthesis was measured by the addition of BrdU (Sigma-Aldrich) to the medium at a final concentration of 10 μM 6 h before harvest.
Flow cytometry and CD4+ T cell purification
Lymphocytes (0.5–1 x 106) from primed cultures were stained with PerCP-conjugated anti-CD4 Ab (BD Pharmingen) and PE-conjugated anti-Thy1.2 Ab (BD Pharmingen), and cell division was analyzed by CFSE dilution. The absolute number of cell divisions accumulated within the CD4+ T cell subset was calculated as previously described (3). BrdU incorporation by individual cells was detected using a Phoenix Red-conjugated anti-BrdU Ab (Phoenix Flow Systems) after fixation and permeabilization (Fix & Perm; Caltag Laboratories). DNA content was measured by the addition of 7-aminoactinomycin D (25 μg/ml) to the fixed and permeabilized cells for 30 min. The above fluorescence parameters were assessed using a FACSCalibur flow cytometer (Becton Dickinson). Where indicated, CD4+ T cells were purified from primed cultures by negative selection using Bio-Mag beads (Qiagen) with Abs against I-A/E, CD19, CD16/32, CD11b, and CD8 (BD Pharmingen).
Immunoblot analysis
Total cultures containing both CD4+ and CD8+ T cells were harvested, lysed, and subjected to SDS-PAGE (0.5 x 106 cell equivalents per lane). The resolved proteins were transferred to nitrocellulose, blocked with a commercial blocking reagent (Roche), and probed with an Ab against p27kip1 (Santa Cruz Biotechnologies). The blots were stripped and reprobed with an Ab against actin (Santa Cruz Biotechnologies). Immunoreactive protein was visualized using HRP-conjugated secondary Ab (Jackson ImmunoResearch Laboratories) and a chemiluminescent substrate (Pierce) on a Bio-Rad Chemidoc system. Densitometric quantification was performed using ImageJ image analysis software (Version 1.31k, National Institutes of Health, http://rsb.info.nih.gov/ij).
Measurement of IL-2 secretion
IL-2 secreted into the supernatant was measured by fluorescence-linked immunosorbent assay (FLISA) or by ELISA. IL-2 FLISA was performed by coating 5-μm latex beads (107/ml; Interfacial Dynamics) with IL-2 capture Ab (2 μg/ml; BD Pharmingen) in PBS for 2 h at 37°C. Beads were washed with PBS and blocked for 30 min at 37°C culture medium. IL-2 capture beads (300,000 per test) were incubated with 100 μl of culture supernatant or titrated IL-2 standards (R&D Systems) for 2 h at 37°C, then washed in culture medium. Bead-bound IL-2 was detected by adding PE-conjugated anti-IL-2 (0.1 μg/test; BD Pharmingen) for 30 min at 4°C, washing in PBS, and measuring fluorescence using a FACSCalibur flow cytometer (Becton Dickinson). IL-2 ELISA was performed by coating a NUNC Maxisorp 96-well plate with capture Ab overnight at 4°C. Wells were washed with PBS/0.05% Tween 20 and blocked for 1 h at room temperature, then incubated with 100 μl of culture supernatant and dilutions of IL-2 standard for 2 h at room temperature (eBiosciences). Wells were washed and detection Ab was added at for 1 h at room temperature. Following additional washes the plate was incubated with Avidin-HRP for 30 min at room temperature. IL-2 was detected by addition of tetramethylbenzidine substrate solution for 15 min, followed by neutralization with 1 M H3PO4. Absorbance was then measured at 450 nm on a Bio-Rad microplate reader.
Results
p27kip1promotes early G1 to S phase progression in activated CD4+ T cells
To experimentally modulate p27kip1 protein levels during CD4+ T cell activation, we used a gene dosage approach in which T cells from wild-type (p27kip1+/+) mice represent normal levels of p27kip1 protein, while T cells from mice with a homozygous deletion of the entire p27kip1 coding region (7) (p27kip1–/–) completely lack p27kip1 protein. In addition, we took advantage of the fact that T cells from mice hemizygous for the p27kip1 gene (p27kip1+/–) contain approximately half the wild-type amount of p27kip1 protein (Ref.8 and see Fig. 2). We subjected these T cells to polyclonal TCR ligation in the presence of physiologic costimulatory signals in vitro, and measured the kinetics of DNA synthesis and mitosis over a 4-day period (Fig. 1, A and B). In wild-type cultures, the majority of CD4+ T cells had begun to synthesize DNA by 36 h after stimulation, and most of these cells were undergoing the first cell division (Fig. 1B and data not shown). By 48 h, the majority of the CD4+ T cells had progressed into a second S phase, and a significant portion was passing through mitosis for a second time (Fig. 1, A and B). If p27kip1 were acting primarily as a CDK inhibitor during this early phase of the response, p27kip1–/– CD4+ T cells should exhibit enhanced DNA synthesis and cell division. In contrast, we observed a 10–20% decrease in the frequency of p27kip1–/– CD4+ T cells entering S phase and mitosis at these early time points compared with p27kip1+/+ CD4+ T cells (Fig. 1, A and B). These results demonstrate that G1 to S phase progression during the first one to two mitoses is delayed in the complete absence of p27kip1.
FIGURE 2. p27kip1 protein dynamics in activated lymphocytes from wild-type and p27kip1–hemizygous mice. Spleen and LN cells from p27kip1+/+ or p27kip1+/– mice were stimulated in vitro with anti-CD3 (1 μg/ml) and either control Ig (5 μg/ml) or CTLA-4 Ig (5 μg/ml). A, Cultures were harvested at the indicated time points, the live cells were purified over Ficoll, and whole cell extracts were subjected to immunoblot analysis for p27kip1 (0.5 x 106 cell equivalents per lane). B, All blots were stripped and reprobed for actin as a loading control, and actin and p27kip1 signals were quantified by densitometry. p27kip1 protein levels are shown normalized against actin protein levels. The results are representative of two separate experiments.
FIGURE 1. Effect of p27kip1 on DNA synthesis and cell division at the single cell level in populations of activated CD4+ T cells. A, CFSE-labeled spleen and lymph node (LN) cells from p27kip1+/+, p27kip1+/–, or p27kip1–/– mice were stimulated in vitro with soluble anti-CD3 (0.25 μg/ml). Cultures were subjected to 6-h pulses of BrdU at 30, 42, 66, or 90 h, and harvested at 36, 48, 72, or 96 h, respectively. Density plots are gated on CD4+ T cells. Horizontal lines indicate the channel below which 98% of non-BrdU-pulsed cells fluoresce, and vertical lines indicate discrete cell divisions as determined by CFSE fluorescence (with the zero division on the far right). B, The mean ± SEM BrdU-positive CD4+ cells in each culture from A is depicted in the left panel, and means are also inset in each plot in A. The right panel depicts the mean BrdU-positive CD4+ cells from the p27kip1+/– and p27kip1–/– normalized to the means of corresponding p27kip1+/+ cultures. C, In a separate experiment, CFSE-labeled spleen and LN cells of each genotype were stimulated with the indicated doses of anti-CD3 Ab, and the total number of cell divisions accumulated within the CD4+ T cell subset by 72 h was calculated from the CFSE profiles. D, Depicts the same experiment as in A, only that cultures were stimulated in the presence of CTLA-4 Ig (5 μg/ml). E, Depicts the raw and normalized mean ± SEM BrdU-positive CD4+ cells in each culture from D. F, Depicts the same experiment as in C, only that cultures were stimulated in the presence of CTLA-4 Ig (5 μg/ml). These data are representative of at least three separate experiments.
Interestingly, CD4+ T cells from mice with only one functional copy of the p27kip1 gene behaved differently than CD4+ T cells from either p27kip1–/– or p27kip1+/+ mice. At 36 h postactivation, p27kip1+/– CD4+ T cells exhibited a similar delay in S phase entry as experienced by the p27kip1–/– cells. However, at 48 h, p27kip1+/– cultures showed a 40% increase in the frequency of CD4+ T cells synthesizing DNA compared with p27kip1–/– cultures, and a 20% increase compared with p27kip1+/+ cultures (Fig. 1, A and B). These results, together with the decrease in early cell cycle progression by CD4+ T cells that completely lack p27kip1, suggest that p27kip1 can actually promote cell cycle progression at this stage of the response.
p27kip1 restricts CD4+ T cell division late in the proliferative response
In these in vitro stimulation cultures, the rate of DNA synthesis by CD4+ T cells dropped precipitously after 48 h, such that roughly 15% of the wild-type cells were in S phase at 72 h, and <10% were still synthesizing DNA at 96 h (Fig. 1, A and B). The rate of DNA synthesis also dropped over time in cultures of p27kip1–/– CD4+ T cells, however, nearly twice as many p27kip1–/– cells were still actively cycling at 72 and 96 h as compared with p27kip1+/+ cells (Fig. 1, A and B). p27kip1+/– CD4+ T cells showed an intermediate phenotype, with frequencies of cycling cells similar to wild-type at 72 h, but increased by 30% compared with wild-type at 96 h. The increased rate of late DNA synthesis led to as much as a 30% increase in the degree of clonal expansion by both p27kip1–/– and p27kip1+/– CD4+ T cells at lower doses of anti-CD3 stimulation (Fig. 1C). These data show that p27kip1 functions late in the CD4+ T cell response to restrict the number of actively dividing cells, and thereby functions to oppose clonal expansion.
p27kip1 governs the requirement for CD28 costimulation
T cells stimulated in the absence of CD28 costimulation exhibit elevated levels of p27kip1 (4, 12), produce very little IL-2 (21), and fail to divide and expand efficiently (20). If p27kip1 plays an important role in restricting cell cycle progression under these conditions, then elimination or reduction of p27kip1 expression should result in CD28-independent T cell proliferation. To test this, we subjected T cells from p27kip1+/+, p27kip1+/–, or p27kip1–/– mice to anti-CD3 stimulation in the presence of CTLA-4 Ig. Consistent with previous studies (20, 22), blockade of CD28/B7 interactions using CTLA-4 Ig resulted in an abortive proliferative response by wild-type CD4+ T cells (Fig. 1, D and E). At 36 h, the number of CD4+ T cells entering S phase and progressing through the first mitosis in CTLA-4 Ig-treated cultures was comparable to that in CD28-costimulated cultures (Fig. 1E). However, the frequency of CD4+ T cells synthesizing DNA in CTLA-4 Ig-treated cultures decreased rapidly by 48 h to less than half of that in untreated cultures, and fell below 10% by 72 h (Fig. 1, D and E). In the absence of CD28 costimulation, p27kip1–/– CD4+ T cells exhibited a 20% proliferative advantage over wild-type cells at 48 h, which rose to a 75% advantage by 72 h (Fig. 1, D and E). Similarly, p27kip1+/– CD4+ T cells exhibited an early 80% increase in the number of cells in S phase at 48 h, and this increased rate of DNA synthesis was maintained throughout the rest of the response (Fig. 1, D and E). In fact, the rate of DNA synthesis at 72 and 96 h by both p27kip1–/– and p27kip1+/– CD4+ T cells was considerably less affected by CD28-costimulatory blockade than wild-type CD4+ T cells. Blockade of CD28/B7 interactions resulted in a 3- to 4-fold decrease in clonal expansion by wild-type CD4+ T cells (Fig. 1, C and F), however, CTLA-4 Ig treatment resulted in only a 25% decrease in the proliferation of p27kip1–/– CD4+ T cells (Fig. 1, C and F). p27kip1-hemizygous CD4+ T cells also exhibited a lesser but significant increase in cumulative cell division under these conditions (Fig. 1F). These data demonstrate that CD28 costimulation is not strictly required for the maintenance of clonal expansion by CD4+ T cells in the absence of p27kip1, and suggest that wild-type levels p27kip1 participate in determining the CD28-costimulatory requirement for a CD4+ T cell response.
A 2-fold reduction in the level of p27kip1 protein is sufficient to promote normal cell cycle progression by CD4+ T cells
We were next interested in why p27kip1+/– CD4+ T cells showed enhanced proliferative responses, which in some cases equaled that of the p27kip1–/– cells. This could be explained in part by the enhanced rate of early S phase entry exhibited by these cells, however, the hemizygous cells also showed enhanced proliferation late in the response, when p27kip1 acts as a negative regulator of cell cycle progression. Another explanation may be that these cells maintain low levels of p27kip1 under conditions where wild-type cells contain high levels of this CDK inhibitor. To address this, we compared the kinetics of p27kip1 protein expression in p27kip1+/– and p27kip1+/+ T cells. Resting lymphocytes from wild-type mice express a relatively high level of p27kip1 (Fig. 2, A and B). Immediately following productive activation, the amount of p27kip1 protein drops slightly, but quickly rebounds and this high level is maintained throughout the first 48 h of the response. Subsequently, p27kip1 expression decreases to roughly half the starting level by 72 h (Fig. 2, A and B). When CD28 costimulation is inhibited during stimulation, wild-type cells down-regulate p27kip1 slightly, but by 36 h, p27kip1 begins to accumulate until at 96 h the CTLA-4 Ig-treated cells express 3-fold more than untreated cells (Fig. 2, A and B). Resting lymphocytes from p27kip1+/– mice contain approximately half the amount of p27kip1 protein as lymphocytes from p27kip1+/+ animals, and this level is maintained after activation and throughout the duration of the proliferative response (Fig. 2, A and B). Remarkably, this low level of p27kip1 expression is even maintained when CD28 costimulation is blocked during activation (Fig. 2, A and B). Under these conditions, the hemizygous cells express 3- to 5-fold less p27kip1 than wild-type cells. These results show that the 2-fold reduction in p27kip1 protein that normally occurs after productive T cell activation represents a highly significant functional difference, leaving the cells with a high enough level of p27kip1 to promote S phase progression during the first cell division, but not high enough to restrict clonal expansion later in the response.
p27kip1 establishes a threshold for mitogenic signaling through the IL-2R
One possible mechanism for the enhanced CD4+ T cell proliferation in the absence of CD28 costimulation could be that p27kip1 deficiency leads to deregulated IL-2 production. Forced overexpression of p27kip1 has been shown to inhibit IL-2 transcription in T cell lines (12, 19), and therefore, genetic elimination of p27kip1 may be expected to augment IL-2 production. However, we found that naive p27kip1–/– or p27kip1+/– CD4+ T cells did not secrete greater amounts of IL-2 during primary stimulation as compared with wild-type cells (Fig. 3A). Furthermore, blockade of CD28 costimulation with CTLA-4 Ig resulted in a 100-fold reduction in the amount of IL-2 available in the cultures of both wild-type and p27kip1-mutant T cells (Fig. 3A). These data suggest that physiologic levels of p27kip1 do not directly influence IL-2 expression, in agreement with previous studies (19), and that the enhanced proliferative capacity of CD4+ T cells that lack or express reduced levels of p27kip1 is not a result of increased IL-2 secretion.
FIGURE 3. Effect of p27kip1 on T cell IL-2 production and IL-2 responsiveness. A, Spleen and LN cells from p27kip1+/+ () and p27kip1–/– () mice were stimulated in vitro with anti-CD3 (1 μg/ml) in the presence of anti-CD28 (0.5 μg/ml) or CTLA-4 Ig (10 μg/ml). IL-2 secreted into the supernatants during the first 24 h was measured by FLISA (left panel). In a separate experiment, IL-2 secretion by p27kip1+/+ () and p27kip1+/– () cells was also compared (right panel). IL-2 secretion at 48 and 72 h showed similar trends between p27kip1+/+, p27kip1–/–, and p27kip1+/– cells (data not shown). The data shown are representative of two experiments for each genotype. B, Lymphocytes from p27kip1+/+, p27kip1–/–, and p27kip1+/– mice were cultured as in A, and the expression of CD25, CD122, and CD132 on the CD4+ T cell subset was assessed by flow cytometry. The 48-h timepoint is shown. C, CFSE-labeled spleen and LN cells from p27kip1+/+ (filled symbols) and p27kip1–/– (open symbols) mice were primed in vitro with anti-CD3 (1 μg/ml) and anti-CD28 (0.5 μg/ml), for 24 h. Cultures were washed to remove the TCR/CD28 stimulus, and the cells were rested for overnight. IL-2 was added at the indicated concentrations, and cell division was quantified 72 h later. D, Spleen and LN cells from p27kip1+/+ and p27kip1+/– mice were primed and rested as in C, and IL-2 was added at the indicated concentrations. After 48 h, the cultures were harvested, live cells were purified over Ficoll, and whole cell extracts were subjected to immunoblot analysis for p27kip1 (0.5 x 106 cell equivalents per lane). All data are representative of two separate experiments.
We also tested whether the lack or reduced levels of p27kip1 protein results in enhanced expression of a high affinity IL-2R by measuring the surface expression of CD25 (IL-2R -chain), CD122 (IL-2R -chain), and CD132 (IL-2R -chain) on p27kip1+/+, p27kip1+/–, and p27kip1–/– CD4+ T cells. In the presence of CD28 costimulation, all receptor chains were up-regulated to a similar degree on CD4+ T cells of all three genotypes (Fig. 3B, upper panels). Blockade of CD28 costimulation during TCR stimulation hampered the induction of these receptor subunits, particularly CD25 and CD122, and affected CD4+ T cells of all three genotypes to a similar degree (Fig. 3B, lower panels). These data suggest that the enhanced proliferative capacity of CD4+ T cells that lack or express reduced levels of p27kip1 is not a result of increased IL-2R expression.
Therefore, we hypothesized that the enhanced proliferation of p27kip1–/– CD4+ T cells under limiting IL-2 conditions may be due to an increased sensitivity to the mitogenic signals provided by this growth factor. To test this possibility, p27kip1+/+ and p27kip1–/– T cells were primed for 24 h with agonistic Abs against CD3 and CD28, and stimulation was subsequently stopped by washing in fresh medium. The cells were then exposed to various fixed concentrations of IL-2 and cultured for an additional 72 h. At high concentrations of IL-2 (5–20 ng/ml), wild-type and p27kip1-deficient CD4+ T cells exhibited comparable proliferation (Fig. 3B), consistent with the comparable response of these two cell populations to high-dose polyclonal stimulation (Fig. 1C). However, at lower concentrations of IL-2, such as those found in cultures of T cells stimulated in the absence of CD28 costimulation (<1 ng/ml; Fig. 3A), p27kip1-deficient CD4+ T cells accumulated significantly more cell divisions throughout the culture period than wild-type cells (Fig. 3B). Primed CD4+ T cells from p27kip1+/– mice also showed enhanced sensitivity to IL-2 signaling, as these cells were able to efficiently down-regulate p27kip1 and proliferate (data not shown) at doses of IL-2 that did not induce degradation in wild-type cells (Fig. 3C). These results show that p27kip1 regulates not only the kinetics of cell cycle entry by CD4+ T cells in response to IL-2 (15), but also the cumulative number of cell divisions and the degree of clonal expansion achieved.
Our data suggest that T cells experience greater mitogenic signaling from a given amount of IL-2 in the absence of p27kip1 than when p27kip1 is present. If this were the case, p27kip1-deficient cells should be less sensitive to proximal inhibitors of IL-2R-mediated signal transduction. To test this hypothesis, wild-type, p27kip1-deficient, or p27kip1-hemizygous CD4+ T cells were stimulated in the presence of rapamycin, an immunosuppressive drug that inhibits T cell proliferation by reportedly blocking IL-2-mediated down-regulation of p27kip1 (10). Rapamycin reduced the degree of clonal expansion by wild-type CD4+ T cells in response to TCR/CD28 by >2-fold (Fig. 4A), and this was associated with a 3- to 6-fold increase in p27kip1 protein levels compared with control-treated cells (Fig. 4, B and C). Rapamycin likewise strongly inhibited IL-2-induced degradation of p27kip1 in primed p27kip1+/+ cells (Fig. 4D). However, cell division by p27kip1-deficient CD4+ T cells was reduced by only 25% in the presence of rapamycin, with these cultures accumulating nearly twice as many cell divisions than cultures of wild-type T cells (Fig. 4A). The proliferation of p27kip1+/– CD4+ T cells was also significantly resistant to rapamycin (Fig. 4A), and this corresponded to a 3- to 6-fold reduction in the amount of p27kip1 detected throughout the response as compared with rapamycin-treated wild-type cells (Fig. 4, B and C). Rapamycin was also much less effective at blocking IL-2-induced p27kip1 degradation in p27kip1+/– cells (Fig. 4D).
FIGURE 4. Effect of p27kip1 on rapamycin sensitivity. A, CFSE-labeled spleen and LN cells from p27kip1+/+ (filled symbols), p27kip1+/– (gray symbols), or p27kip1–/– (open symbols) mice were stimulated with the indicated doses of anti-CD3 Ab in the presence of rapamycin (10 ng/ml) or vehicle (DMSO). Cells were cultured for 72 h, and the total number of cell divisions accumulated within the CD4+ T cell subset of each culture was calculated from the CFSE profiles. B and C, Spleen and LN cells from p27kip1+/+ or p27kip1+/– mice were stimulated and treated with rapamycin as in A, and whole cell extracts were subjected to immunoblot analysis for p27kip1 and actin as described in Fig. 2. D, Spleen and LN cells from p27kip1+/+ and p27kip1+/– mice were primed and rested as in Fig. 3C, and IL-2 was added at the indicated concentrations in the presence or absence of rapamycin. After 48 h, whole-cell extracts were subjected to immunoblot analysis for p27kip1. All data are representative of two experiments.
These data show that elimination or reduction of p27kip1 protein renders CD4+ T cells less susceptible to rapamycin, and formally establish p27kip1 as a downstream mediator of the antiproliferative effect of this agent. However, these data also specifically demonstrate a rapamycin-sensitive component of the proliferative response in p27kip1-deficient CD4+ T cells (Fig. 4A), suggesting that a significant degree of rapamycin-mediated inhibition occurs independently of p27kip1. Our data, together with the results of other models in which p27kip1 levels have been modulated (14, 15, 23), suggest that p27kip1 acts in CD4+ T cells as an intracellular sensor of available growth factor that helps to set the threshold for mitogenic signaling and cell cycle progression.
Primed CD4+ T cells exhibit enhanced effector function in the absence of p27kip1
Cell division promotes differentiation in CD4+ T cell populations (1, 2, 3). Because p27kip1-deficient CD4+ T cells tend to proliferate more than wild-type cells during the primary response, we hypothesized that these cells might exhibit enhanced effector function upon restimulation. To test this, we primed wild-type or p27kip1–/– lymphocytes with anti-CD3 for 72 h as in Fig. 1, and allowed the cultures to come to rest in medium for 24 h. CD4+ T cells were then restimulated through TCR/CD28, and IL-2 secretion was measured at 24 h. p27kip1-deficient CD4+ T cells produced 4-fold more IL-2 than CD4+ cells upon restimulation (Fig. 5A, first and third columns), and showed a slightly increased frequency of cells entering S phase 24 h later (Fig. 5B, first and third panels in the middle row). These results suggest that p27kip1-deficient CD4+ T cells are better able to respond when re-encountering Ag.
FIGURE 5. Effect of p27kip1 on CD4+ T cell effector function. A, Spleen and LN cells from p27kip1+/+ () and p27kip1–/– () mice were stimulated in vitro with anti-CD3 (1 μg/ml) in the presence of control Ig (10 μg/ml) or CTLA-4 Ig (10 μg/ml) for 72 h and rested for 24 h in culture medium. CD4+ T cells were purified from each culture by negative selection and an equal number were restimulated with immobilized anti-CD3 (5 μg/ml) and anti-CD28 (5 μg/ml). Supernatants were collected after 24 h and IL-2 secretion was measured by FLISA. B, p27kip1+/+ or p27kip1–/– CD4+ T cells cultured and purified as in A were restimulated with medium alone, immobilized anti-CD3 (2 μg/ml), or IL-2 (20 U/ml) for 48 h. BrdU was added to the cultures 6 h before harvest and DNA synthesis (BrdU) and DNA content (7-aminoactinomycin D (7-AAD)) were assessed by flow cytometry. C, CD4+ T cells from wild-type or hemizygous (p27kip1+/–) mice were cultured, purified and restimulated as in A, and proliferation was assessed by BrdU incorporation as in B. D, Additionally, purified CD4+ T cells from cultures as in C were assessed for p27kip1 content by immunoblot analysis (0.5 x 106 cell equivalents per lane). All data shown are representative of two to three separate experiments.
T cell proliferation also opposes anergy induction (4, 24, 25), and T cell anergy has been associated in several models with elevated levels of p27kip1 (4, 12, 24, 26). These results have led to the hypothesis that p27kip1 may directly participate in the induction or maintenance of anergy in T cells. To test this, we primed wild-type or p27kip1–/– lymphocytes with anti-CD3 as above, but blocked B7/CD28 interactions with CTLA-4 Ig to provide a tolerizing stimulus. Consistent with previous studies (27, 28), wild-type CD4+ T cells subjected to this tolerizing stimulus secreted 3- to 4-fold less IL-2 when restimulated through TCR/CD28 compared with cells primed in the presence of CD28 costimulation (Fig. 5A,
). The amount of IL-2 produced by CTLA-4 Ig-treated p27kip1–/– CD4+ T cells was reduced only 2-fold compared with fully primed p27kip1–/– cells (Fig. 5A,
). Remarkably, this amount of IL-2 was >7-fold the levels produced by anergic wild-type cells, and >2-fold the levels produced by wild-type cells primed in the presence of costimulation (Fig. 5A,
vs
). To test whether the enhanced production of IL-2 by the restimulated p27kip1–/– T cells leads to increased proliferation during the secondary culture, we stimulated wild-type or p27kip1–/– cells in the presence or absence of CD28 costimulation as in Fig. 5A. Purified CD4+ T cells from these cultures were restimulated with medium alone, immobilized anti-CD3, or IL-2 for 48 h, and DNA synthesis by individual cells was measured. The frequency of p27kip1+/+ cells proliferating in response to TCR ligation was reduced by 50% if costimulation was blocked during the primary stimulus (Fig. 5B, compare first and third middle panels), consistent with their anergic phenotype of reduced IL-2 production (Fig. 5A). Conversely, the proliferation of p27kip1–/– T cells was not adversely affected by costimulatory blockade (Fig. 5B, compare second and fourth middle panels), resulting in a 2-fold increase in the frequency of proliferating cells as compared with p27kip1+/+ CD4+ T cells stimulated under the same conditions (Fig. 5B, compare third and fourth middle panels). As expected, the response of all these cells to IL-2 was intact, with 30% or more of the cells within each population having synthesized DNA during the period assessed (Fig. 5B, lower panels). These data show that p27kip1–/– CD4+ T cells exhibit significantly enhanced effector function even when primed in the absence of CD28 costimulation, and suggest that p27kip1 is required for the full development of anergy in this model.
CD4+ T cells from p27kip1+/– mice also exhibit a proliferative advantage in the primary response in our model, therefore, we predicted that these cells might likewise exhibit a reduced susceptibility to anergy induction. To test this, we stimulated wild-type or p27kip1+/– cells in the presence or absence of CD28 costimulation as in Fig. 5A. Purified CD4+ T cells from these cultures were restimulated with medium alone, immobilized anti-CD3, or IL-2 for 48 h, and DNA synthesis by individual cells was measured. Similar to p27kip1–/– cells, CD3/28-primed CD4+ T cells from p27kip1+/– mice showed a slight increase in proliferation upon restimulation, as compared with wild-type cells (Fig. 5C, first and third panels in the middle row). However, unlike p27kip1–/– cells, p27kip1+/– CD4+ T cells from CD3/CTLA-4Ig-primed cultures exhibited a strong proliferative defect that was similar in magnitude to that of anergized wild-type CD4+ T cells (Fig. 5C, second and fourth panels in the middle row). This suggests that p27kip1+/– CD4+ T cells are as susceptible to anergy induction as cells with two functional copies of the p27kip1 gene. Because p27kip1+/– CD4+ T cells are still able to induce p27kip1 protein expression, we hypothesized that the induction of p27kip1 during restimulation might enforce the anergic state in these cells. Consistent with the experiments depicted in Fig. 2, resting p27kip1+/– CD4+ T cells contained less p27kip1 protein than wild-type cells after primary stimulation (Fig. 5D, lanes 1–4). However, upon restimulation, anergized p27kip1+/– cells failed to down-regulate p27kip1 as compared with effector cells (Fig. 5D, lanes 5–8), and instead, p27kip1 protein accumulated in these cells to levels comparable to that expressed by wild-type anergic T cells (Fig. 5D, lanes 7 and 8). These data may explain why T cells with one functional copy of the p27kip1 gene are still susceptible to anergy, and further support the idea that p27kip1 protein levels are an important factor that determines the responsiveness of CD4+ T cells to mitogenic stimulation.
Discussion
p27kip1 was first identified biochemically in cell lines as an inhibitor of G1 and S phase cyclin-CDK complexes, particularly cyclin D-CDK4, cyclin E-CDK2, and cyclin A-CDK2 complexes (29, 30). Subsequent genetic studies have confirmed a general role for p27kip1 in controlling cell proliferation throughout the development of multiple tissues, including the thymus and the mature T cell pool (7, 8, 9). The contribution of p27kip1 to peripheral T cell immune responses, however, has been less clear. Down-modulation of p27kip1 is associated with T cell activation, and while this process has generally been considered to be a crucial event for T cell proliferation (10, 14, 15, 31), some studies have suggested that peripheral T cell responses proceed normally in the absence of p27kip1 (8, 9, 17, 19) Earlier studies of p27kip1 expression and function in T lymphocytes were, for the most part, limited to very short time frames and did not explore the role of p27kip1 in the context of different T cell stimuli. Therefore, we have used both a full and a partial loss-of-function approach to carefully examine the role of p27kip1 in regulating the kinetics of proliferation and the development of effector function in populations of mature T lymphocytes. Our studies bring together previously disparate findings by revealing that p27kip1 can serve dual, opposing roles in T cell cycle progression. We establish that p27kip1 functions physiologically within a 2-fold range of protein concentration to first promote, not inhibit, G1 to S phase progression during the initial phase of T cell proliferation. Following the peak of the response, p27kip1 then functions as a negative regulator of T cell clonal expansion by promoting exit from the cell cycle. Finally, we show that the net effect of p27kip1 is to oppose T cell effector function, and this negative regulatory role is required for the full effect of tolerizing stimuli and immunosuppressive therapies.
How can p27kip1 act as both a positive and a negative regulator of T cell cycle progression? The mechanism by which p27kip1 inhibits cell cycle progression is fairly clear. p27kip1 can bind to and inhibit the kinase activity of cyclin D-, E-, and A-containing CDK complexes (32). It is important, however, to note that p27kip1 does not play an obligatory role in regulating the maximal proliferative "output" in response to mitogenic stimuli in our system (i.e., fully p27kip1–deficient T cells do not proliferate indefinitely and uncontrollably). The role of p27kip1 is subtle when T cells receive an optimal combination of stimuli, but becomes apparent when mitogenic signals are limiting. Our data suggest that p27kip1 functions in T cells to set the threshold amount of growth factor required to initiate G1 to S phase progression. For instance, our IL-2 dose response and rapamycin data demonstrate that p27kip1–/– and p27kip1+/– CD4+ T cells are roughly 10-fold more sensitive to growth factor receptor signaling (Figs. 3 and 4). The simplest interpretation of these results is that p27kip1+/– cells begin with an amount of p27kip1 protein that approximates the lowest possible steady-state level achievable in response to normal mitogenic signals. This level can be reduced further (i.e., approximating a complete absence of p27kip1 as in p27kip1–/– T cells), but in our system this is only achieved by the addition of exogenous IL-2. Once this growth factor signaling threshold has been reached, other downstream factors take over to limit the proliferative response.
How can p27kip1, a CDK inhibitor, function to promote cell cycle progression? D-type cyclins are the first induced upon T cell activation, and cyclin D-CDK4/6 complexes are required for progression from G0 to G1 (33, 34). Cyclin D-containing CDK, unlike cyclin E-, A- and B-containing complexes, are subject to an additional level of regulation by the inhibitor of CDK4 family of CDK inhibitors (35, 36, 37, 38). These members inhibit only cyclin D-type CDK, and block the assembly of cyclin D-CDK4/6 complexes by binding to free CDK4 and -6 directly (39, 40, 41). Conversely, p27kip1 cannot efficiently bind to free cyclins or CDK, but rather bind stably in a trimolecular complex (30, 42, 43). Consistent with this, studies in non-T cells have demonstrated that the accumulation of cyclin D-CDK4/6 complexes is impaired in the absence of p27kip1 (42, 44, 45). In this way, p27kip1 actually promotes the assembly and stability of cyclin D-CDK4/6 complexes, and protects CDK4/6 from regulation by inhibitor of CDK4 family members (32). Although p27kip1 levels remain high, the assembled cyclin-CDK complexes are inhibited. However, once growth factor signaling reaches sufficient strength, p27kip1 is targeted for degradation, and the already-formed cyclin D-CDK4/6 complexes are freed to phosphorylate downstream targets and promote cell cycle progression. Residual p27kip1 protein then acts to oppose the activity of cyclin E- and -A-CDK complexes. This scenario explains how these two opposing biochemical functions of p27kip1 can operate concurrently during T cell activation, and how the apparent biological effect of p27kip1 can switch from positive to negative later in the response when cyclin E- and A-containing CDK complexes are predominantly expressed and active. As we observe in our current study, a 2-fold reduction in initial p27kip1 levels (i.e., as in p27kip1+/– cells) is sufficient to significantly enhance cell cycle progression, potentially by influencing this balance between p27kip1-mediated cyclin-CDK assembly vs assembly and inhibition. Additional biochemical studies will be required to address this hypothesis.
An additional intriguing finding from our studies is that CD4+ T cells that lack p27kip1 exhibit enhanced effector function during restimulation. T cell effector function and anergy avoidance are linked to cell division (1, 2, 3, 4), therefore derepression of the proliferative response might be predicted to result in enhanced T cell effector function. Therefore, these current studies provide important support for the role of cell cycle progression in driving T cell differentiation. Significantly, the enhanced effector function of p27kip1–/– CD4+ T cells was still observed even under conditions that would induce anergy in normal T cells. Although the loss of p27kip1 in CD4+ T cells does not render either primary cell division or the recall response completely insensitive to CD28 costimulation, the fact that p27kip1–/– CD4+ T cells produce more IL-2 and proliferate more following a tolerizing stimulus than wild-type T cells do under full stimulation leads us to conclude that T cells are refractory to anergy induction in the absence of p27kip1.
Our results demonstrate that, in the absence of p27kip1, CD4+ T cells are less susceptible to the effects of CTLA-4 Ig, and formally establish p27kip1 as a downstream mediator of the antiproliferative effect of CD28-costimulatory blockade. An extension of this finding is that p27kip1 participates in setting the CD28-costimulatory requirement for T cell clonal expansion under normal circumstances, an idea that was previously suggested by correlative studies (46, 47). Under physiologic conditions, p27kip1 does not govern the requirement for CD28 costimulation by influencing the amount of IL-2 produced during a primary response (Fig. 3A). Instead, our data using rapamycin, and previous studies by Powell et al. (19) using IL-2-deficient T cells, suggest that the link between CD28 and p27kip1 is indirect and controlled instead at the level of IL-2 responsiveness. The amount of costimulation needed is a direct reflection of the amount of IL-2 a cell needs to produce to overcome p27kip1-mediated inhibition of proliferation.
The net effect of p27kip1, through its capacity to inhibit cell cycle progression, is apparently to oppose effector differentiation and to promote the induction of anergy in CD4+ T cells. This role of p27kip1 is not limited to CD4+ T cells, as we have also observed that p27kip1–/– CD8+ T cells are hyperproliferative and a higher frequency of CD8+ cells differentiate into IFN--producing effector cells as compared with wild-type CD8+ cells (our unpublished observations). The importance of this CDK inhibitor in the regulation of peripheral T cell responses is further supported by our preliminary studies in a cardiac allograft model, which have demonstrated that p27kip1–/– mice are significantly resistant to tolerance induction in vivo by costimulatory blockade as compared with wild-type mice (our unpublished observations). By acting as a sensor of available growth factor and determining the sensitivity of CD4+ T cells to tolerizing signals, p27kip1 is critical in determining the extent of clonal expansion and whether the outcome of encounter with Ag is tolerance or immunity.
Disclosures
The authors have no financial conflict of interest.
Acknowledgments
We thank Drs. J. Powell, V. Boussiotis, and L. Turka for helpful discussions.
Footnotes
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.
1 This work was supported by National Institutes of Health (NIH) Grants K01-DK02771 and R01-AI054643 (to A.D.W.). A.D.W. is a member of the Joseph Stokes, Jr. Research Institute and the Biesecker Pediatric Liver Disease Center at The Children’s Hospital of Philadelphia, and E.A.R. is supported by NIH Training Grant T32-AR007442-18.
2 Address correspondence and reprint requests to Dr. Andrew D. Wells, 916F Abramson Research Center, 3615 Civic Center Boulevard, Philadelphia, PA, 19104. E-mail address: adwells{at}mail.med.upenn.edu
3 Abbreviations used in this paper: CDK, cyclin-dependent kinase; FLISA, fluorescence-linked immunosorbent assay; LN, lymph node.
Received for publication July 28, 2004. Accepted for publication December 22, 2004.
References
Bird, J. J., D. R. Brown, A. C. Mullen, N. H. Moskowitz, M. A. Mahowald, J. R. Sider, T. F. Gajewski, C. R. Wang, S. L. Reiner. 1998. Helper T cell differentiation is controlled by the cell cycle. Immunity 9:229.
Gett, A. V., P. D. Hodgkin. 1998. Cell division regulates the T cell cytokine repertoire, revealing a mechanism underlying immune class regulation. Proc. Natl. Acad. Sci. USA 95:9488
Gudmundsdottir, H., A. D. Wells, L. A. Turka. 1999. Dynamics and requirements of T cell clonal expansion in vivo at the single-cell level: effector function is linked to proliferative capacity. J. Immunol. 162:5212.
Wells, A. D., M. C. Walsh, D. Sankaran, L. A. Turka. 2000. T cell effector function and anergy avoidance are quantitatively linked to cell division. J. Immunol. 165:2432.
Morgan, D. O.. 1995. Principles of CDK regulation. Nature 374:131.[Medline]
Vidal, A., A. Koff. 2000. Cell-cycle inhibitors: three families united by a common cause. Gene 247:1.
Fero, M. L., M. Rivkin, M. Tasch, P. Porter, C. E. Carow, E. Firpo, K. Polyak, L. H. Tsai, V. Broudy, R. M. Perlmutter, et al 1996. A syndrome of multiorgan hyperplasia with features of gigantism, tumorigenesis, and female sterility in p27kip1-deficient mice. Cell 85:733.
Kiyokawa, H., R. D. Kineman, K. O. Manova-Todorova, V. C. Soares, E. S. Hoffman, M. Ono, D. Khanam, A. C. Hayday, L. A. Frohman, A. Koff. 1996. Enhanced growth of mice lacking the cyclin-dependent kinase inhibitor function of p27kip1. Cell 85:721
Kwon, T. K., M. A. Buchholz, P. Ponsalle, F. J. Chrest, A. A. Nordin. 1997. The regulation of p27Kip1 expression following the polyclonal activation of murine G0 T cells. J. Immunol. 158:5642.
Boussiotis, V. A., G. J. Freeman, P. A. Taylor, A. Berezovskaya, I. Grass, B. R. Blazar, L. M. Nadler. 2000. p27kip1 functions as an anergy factor inhibiting interleukin 2 transcription and clonal expansion of alloreactive human and mouse helper T lymphocytes. Nat. Med. 6:290.
Jackson, S. K., A. DeLoose, K. M. Gilbert. 2001. Induction of anergy in Th1 cells associated with increased levels of cyclin-dependent kinase inhibitors p21Cip1 and p27Kip1. J. Immunol. 166:952.
Tsukiyama, T., N. Ishida, M. Shirane, Y. A. Minamishima, S. Hatakeyama, M. Kitagawa, K. Nakayama. 2001. Down-regulation of p27Kip1 expression is required for development and function of T cells. J. Immunol. 166:304.
Zhang, S., V. A. Lawless, M. H. Kaplan. 2000. Cytokine-stimulated T lymphocyte proliferation is regulated by p27Kip1. J. Immunol. 165:6270.
Huleatt, J. W., J. Cresswell, K. Bottomly, I. N. Crispe. 2003. P27kip1 regulates the cell cycle arrest and survival of activated T lymphocytes in response to interleukin-2 withdrawal. Immunology 108:493.
Kovalev, G. I., D. S. Franklin, V. M. Coffield, Y. Xiong, L. Su. 2001. An important role of CDK inhibitor p18INK4c in modulating antigen receptor-mediated T cell proliferation. J. Immunol. 167:3285.
Luo, Y., S. O. Marx, H. Kiyokawa, A. Koff, J. Massague, A. R. Marks. 1996. Rapamycin resistance tied to defective regulation of p27Kip1. Mol. Cell. Biol. 16:6744
Powell, J. D., D. Bruniquel, R. H. Schwartz. 2001. TCR engagement in the absence of cell cycle progression leads to T cell anergy independent of p27kip1. Eur. J. Immunol. 31:3737
Wells, A. D., H. Gudmundsdottir, L. A. Turka. 1997. Following the fate of individual T cells throughout activation and clonal expansion: signals from T cell receptor and CD28 differentially regulate the induction and duration of a proliferative response. J. Clin. Invest. 100:3173.
Harding, F. A., J. G. McArthur, J. A. Gross, D. H. Raulet, J. P. Allison. 1992. CD28-mediated signalling co-stimulates murine T cells and prevents induction of anergy in T-cell clones. Nature 356:607.
Sperling, A. I., J. A. Auger, B. D. Ehst, I. C. Rulifson, C. B. Thompson, J. A. Bluestone. 1996. CD28/B7 interactions deliver a unique signal to naive T cells that regulates cell survival but not early proliferation. J. Immunol. 157:3909.
Mohapatra, S., D. Agrawal, W. J. Pledger. 2001. p27Kip1 regulates T cell proliferation. J. Biol. Chem. 276:21976.[Abstract/Free Full Text]
Wells, A. D., M. C. Walsh, J. A. Bluestone, L. A. Turka. 2001. Signaling through CD28 and CTLA-4 controls two distinct forms of T cell anergy. J. Clin. Invest. 108:895
DeSilva, D. R., K. B. Urdahl, M. K. Jenkins. 1991. Clonal anergy is induced in vitro by T cell receptor occupancy in the absence of proliferation. J. Immunol. 147:3261.
Gilbert, K. M., W. O. Weigle. 1993. Th1 cell anergy and blockade in G1 phase of the cell cycle. J. Immunol. 151:1245.
Wells, A. D., X. C. Li, Y. Li, M. C. Walsh, X. X. Zheng, Z. Wu, G. Nunez, A. Tang, M. Sayegh, W. W. Hancock, et al 1999. Requirement for T-cell apoptosis in the induction of peripheral transplantation tolerance. Nat. Med. 5:1303.
Wells, A. D., Q. H. Liu, B. Hondowicz, J. Zhang, L. A. Turka, B. D. Freedman. 2003. Regulation of T cell activation and tolerance by phospholipase c-1-dependent integrin avidity modulation. J. Immunol. 170:4127.
Toyoshima, H., T. Hunter. 1994. p27, a novel inhibitor of G1 cyclin-Cdk protein kinase activity, is related to p21. Cell 78:67.
Polyak, K., M. H. Lee, H. Erdjument-Bromage, A. Koff, J. M. Roberts, P. Tempst, J. Massague. 1994. Cloning of p27Kip1, a cyclin-dependent kinase inhibitor and a potential mediator of extracellular antimitogenic signals. Cell 78:59.
LaBaer, J., M. D. Garrett, L. F. Stevenson, J. M. Slingerland, C. Sandhu, H. S. Chou, A. Fattaey, E. Harlow. 1997. New functional activities for the p21 family of CDK inhibitors. Genes Dev. 11:847.
Cheng, M., P. Olivier, J. A. Diehl, M. Fero, M. F. Roussel, J. M. Roberts, C. J. Sherr. 1999. The p21Cip1 and p27kip1 CDK "inhibitors" are essential activators of cyclin D-dependent kinases in murine fibroblasts. EMBO J. 18:1571.
Appleman, L. J., A. Berezovskaya, I. Grass, V. A. Boussiotis. 2000. CD28 costimulation mediates T cell expansion via IL-2-independent and IL-2-dependent regulation of cell cycle progression. J. Immunol. 164:144.
Appleman, L. J., A. A. van Puijenbroek, K. M. Shu, L. M. Nadler, V. A. Boussiotis. 2002. CD28 costimulation mediates down-regulation of p27kip1 and cell cycle progression by activation of the PI3K/PKB signaling pathway in primary human T cells. J. Immunol. 168:2729.(Emily A. Rowell, Matthew )