Cytotoxic T lymphocyte-associated antigen-4
We have previously shown that human IL-10-treated dendritic cells (DC) induce an antigen-specific anergy in CD4+ T lymphocytes. These anergic T cells are characterized by an inhibitedproliferation, a reduced production of IL-2, and additionally display antigen-specific suppressor activity. In this study we investigated the mechanisms underlying the anergic state and regulatory function of these T cells. We did not observe enhanced rates of programmed cell death of anergic CD4+ suppressor T cells compared to T cells stimulated with mature DC. Cell cycle analysis by DNA staining and Western blot experiments revealed an arrest of anergic CD4+ T suppressor cells in the G1 phase. High levels of the IL-2-dependent cyclin-dependent kinase (cdk) inhibitor p27Kip1 were found in anergic CD4+ suppressor T cells resulting in an inhibited activation of retinoblastoma protein and an arrest of cell cycle progression in the G1 phase. Addition of IL-2, but not blocking of the CTLA-4 pathway restored the proliferation of the suppressor T cells. In contrast, both treatments induced a down-regulation of p27Kip1 and acomplete inhibition of the antigen-specific regulatory function as demonstrated by high proliferation and enhanced IFN-γ production of co-cultured T cells. Further experiments demonstrated thatp27Kip-expressing regulatory CD4+CD25+ T cells did not contribute to induction of T cell anergy in this model. Our data show that regulatory function of anergic CD4+ suppressor T cells is associated with an arrest in the G1 phase of the cell cycle mediated by increased levels of the IL-2- and CTLA-4-dependent cdk inhibitor p27Kip1.
The induction of antigen-specific tolerance is critical for the prevention of autoimmunity and the maintenance of immune homeostasis. The ability of the immune system to distinguish between self and non-self and between innocuous and harmful foreign antigens is controlled by central and peripheral tolerance mechanisms. The process of central tolerance mediates the deletion of autoreactive T cells in the thymus 1. Induction of full tolerance requires additional mechanisms of peripheral tolerance including active suppression, deletion and the induction of anergy inT cells 2–4.
Anergy is defined as the inability of antigen-specific T cells to proliferate and produce interleukin-2 (IL-2) on rechallenge with fully competent antigen-presenting cells (APC) delivering T cell receptor and costimulatory signals 5. Proliferation of T cells is controlled by cell cycle progression. Cell cycling is a complex process that is activated by cyclins that associate with catalytically active cyclin-dependent kinases (cdk) to form active holoenzymes and is inhibited by cdk inhibitors 6. The cdk inhibitor p27Kip1 was shown to bean IL-2-dependent protein associated with cyclin D-cdk4 and cyclin E-cdk2, inhibiting the kinase activities of these complexes 7, 8. In normal cells, p27Kip1 is abundant during the quiescent state (G0 phase), but its expression is rapidly down-regulated when cells re-enter the cell cycle upon stimulation with growth factors 9.
Dendritic cells (DC) are the most potent initiators of antigen-specific T cell responses 10. However, a role for DC in the induction of peripheral tolerance is supported by several studies 11, 12. At present, the mechanisms involved in this process are not well defined. It is widely assumed, that control of the maturation of DC and/or the generation of subtypes of DC in the presence of self or foreign antigens is fundamental in the induction of peripheral tolerance. We have previously demonstrated that human DC pretreated with IL-10 during their maturation induce antigen-specific anergy in CD4+ and CD8+ T cells 13. These anergic T cells are characterized by a diminished proliferation, reduced production of IL-2 and an inhibited cytotoxicity 13, 14. Furthermore, anergic T cells display antigen-specific regulatory activities resulting in an impaired proliferation and activation of co-cultured T cells of the same specificity 15.
Here, we demonstrate that anergy induction in CD4+ suppressor T cells is associated with an enhanced expression of the IL-2-dependent inhibitor of the G1 phase p27Kip1, resulting in an inhibited activation of retinoblastoma protein (Rb) and a G1 cell cycle arrest. Addition of IL-2 but not blocking of the CTLA-4 pathway overcame the state of anergy. Notably, both treatments inhibited the suppressor function of the T cells via cell cycle progression by down-regulation of the cdk inhibitor p27Kip.
2.1 Rates of apoptosis are not enhanced in anergic CD4+ suppressor T cells
We have previously shown that IL-10-treated DC induce a state of antigen-specific anergy in various populations of CD4+ and CD8+ suppressor T cells. Anergic T cells are characterized by a reduced proliferation and production of IL-2 13, 15, 16. To rule out the involvement of programmed cell death in our model of anergy induction, studies during primary culture (Fig. 1) and restimulation (Fig. 2) were performed to detect apoptotic T cells at various time points. Annexin V/propidium iodide (PI) staining (detection of early apoptotic T cells; Fig. 1A, C; 2A, C) and terminal deoxynucleotidyl transferase-mediated dUTP nick end-labeling (TUNEL) assays (detection of late apoptosis; Fig. 1B, D; 2B, D) were used.
Analysis of the anergic T cells during primary culture revealed slightly but not significantly reduced numbers of apoptotic cells compared to control T cells stimulated with mature DC at day 2 and 5 (Fig. 1). The enhanced rate of apoptosis of stimulated control T cells induced by DC and/or autocrine IL-2 correlates with a higher proliferative response and is paralleled by a higher frequency of cells receiving death signals (Fig. 1A, B). Analysis of early or late apoptotic signals (day 2 and 7) after restimulation showed similar rates of apoptosis in cultures of anergic (Fig. 2C, D) or control T cells (Fig. 2A, B) stimulated with mature DC. These results indicate that the impaired proliferative response during primary culture and the unresponsiveness after restimulation are not the results of apoptosis of the anergic T lymphocytes.
2.2 Lack of proliferation in anergic CD4+ suppressor T cells is due to a G1 arrest of cell cycle
Proliferation is a multistage process characterized by successive entering of various stages of the cell cycle. To assess the mechanisms involved in the down-regulation of the proliferative responses of anergic CD4+ suppressor T cells, cell cycle experiments were performed by analyzing the DNA content.
CD4+ T cells were co-cultured with mature or IL-10-treated DC for 5 days and restimulated with anti-CD3 mAb. As demonstrated in Fig. 3, 57.6 % of stimulated CD4+ T cells co-cultured with mature untreated DC had entered the G0/1 phase of the cell cycle after 48 h of restimulation. In contrast, a significantly higher percentage of the anergic CD4+ T cells (81.9 %) were in the G0/1 phase at day 2 after restimulation (Fig. 3). Similar results were obtained at various time points (day 1–3) after restimulation (data not shown). These data demonstrate that the lack of proliferation of anergic T cells is due to a G1 arrest of the cell cycle.
2.3 G1 arrest of anergic CD4+ suppressor T cells is associated with an up-regulation of the cdk inhibitor p27Kip1 and an impaired Rb hyperphosphorylation
To further characterize the G1 block of the cell cycle in anergic CD4+ suppressor T cells, we investigated the levels of various regulatory proteins during the cell cycle 24 and 48 h after restimulation (Fig. 4A). Western blot analysis of total lysate revealed a slightly enhanced expression of cdk4 and of cyclin D2/D3 in anergic CD4+ T cells co-cultured with IL-10-treated DC as compared to optimally stimulated control T cells (Fig. 4A). No differences between the amounts of cyclin E and cdk2, which are responsible for cell cycle progression in the late G1 phase, were observed (data not shown).
IL-2- or TGF-β-dependent G1 phase cdk inhibitor p27Kip1, p21Cip1/WAF1 and p15 reportedly prevent the entry in the G2/S phase of the cell cycle. We therefore analyzed the expression of p27Kip1, p21Cip1/WAF1 and p15. Levels of p27Kip1 were enhanced in anergic CD4+ suppressor T cells before restimulation (0 min) as compared to T cells stimulated with mature DC (Fig. 4A). Furthermore, anergic CD4+ suppressor T cells exhibited a defective down-regulation of the p27Kip1 inhibitor 24 h and 48 h after restimulation in contrast to activated T cells. Similar to control T cells, these activated T cells showed decreased amounts of p27Kip1 after restimulation (Fig. 4A). No differences in the expression of p15 or p21waf were observed between anergic and stimulated T cells (data not shown). These data indicate that the presence of increased p27Kip1 in anergic CD4+ suppressor T cells resulted in an aberrant ratio of p27Kip1 over the cyclin D-cdk holoenzyme, leading to blockade of its enzymatic activity. Investigations of Rb amounts 24, 48 and 72 h after restimulation revealed an inhibited expression of the hyperphosphorylated form in anergic CD4+ suppressor T cells as compared to control T cells, indicating an abrogated transcription of S phase genes (Fig. 4B).
2.4 Addition of IL-2 restores the impaired proliferation of anergic suppressor T cells
IL-2 secreted by T lymphocytes is known to be the most important cytokine for activation and proliferation of T cells. Therefore, we assessed the influence of IL-2 on the state of anergy in several experimental settings of restimulation. In this two-step anergy assay, allogeneic CD4+ T cells were co-cultured with untreated or IL-10-treated DC. After the first co-culture, T cells were rescued, cultured for 36 h in the presence of IL-2 (2 U/ml) and subsequently restimulated with untreated, fully mature DC or immobilized anti-CD3 mAb. Additionally, in some experiments IL-2 was added at high concentrations (100 U/ml).
As described, anergic CD4+ suppressor T cells stimulated with IL-10-treated DC in primary culture showed markedly diminished proliferation after primary culture and after restimulation in contrast to control T cells (Fig. 5A). More importantly, addition of IL-2 to anergic CD4+ suppressor T cells induced a restored proliferation of T cells (Fig. 5A). In contrast, blocking of the CTLA-4 molecule, known to be highly expressed on anergic CD4+ suppressor T cells and involved in negative regulatory processes, did not influence the state of anergy, independent of the presence or absence of DC (Fig. 5A) 15.
2.5 IL-2 and inhibition of the CTLA-4 pathway abrogate the regulatory properties of the anergic T cells
As previously shown, anergic T cells induced by co-culture with IL-10-treated human DC display antigen-specific suppressor activity and are characterized by a high expression of the CTLA-4-molecule 15. Furthermore, previous studies showed that cell-to-cell contacts and the presence of DC are mandatory for the regulatory activity of suppressor T cells 15. To assess the influence of IL-2 and the CTLA-4 signaling on the suppressive effects of anergic T cells, we analyzed the effects of alloantigen-specific anergic T cells on the proliferation of syngeneic activated control T cells previously stimulated with mature DC in the absence or presence of IL-2 or blocking antibodies against CTLA-4 (Fig. 5B). Anergic CD4+ suppressor T cells were pretreated with IL-2 or anti-CTLA-4 mAb and subsequently co-cultured with an equal number of syngeneic activated control T cells and restimulated with mature DC from the same donor as used during the induction of anergy. As described, the proliferation of activated control T cells co-cultured with anergic CD4+ suppressor T cells was markedly inhibited compared to co-culture experiments with T lymphocytes stimulated with mature DC during the primary culture 15. In contrast, in the presence of IL-2 or anti-CTLA-4 mAb an unrestricted T cell response was observed, demonstrating that IL-2- and CTLA-4-dependent pathways are involved in the regulation of the suppressor activity of the anergic CD4+ suppressor T cells (Fig. 5B).
Additional experiments revealed that the proliferation of co-cultured T cells after addition of IL-2 or anti-CTLA-4 was paralleled by an enhanced activation and induction of a Th1 response as demonstrated by an increased production of IFN-γ but not of Th2 cytokines such as IL-4 (Fig. 5C). These results indicate that the addition of IL-2 or the blocking of the CTLA-4 molecule induced altered pathways of the cell cycle leading to an inhibition of the regulatory function of T cells.
2.6 Inhibition of the suppressor activity is associated with an IL-2- and CTLA-4-dependent down-regulation of the G1 inhibitor p27Kip1
To explore the mechanisms involved in the interaction between cell cycle regulation and regulatory function of the anergic CD4+ suppressor T cells, the effects of IL-2 and of blocking CTLA-4 on the levels of p27Kip1 were studied by Western blot analysis. Suppressor T cells were restimulated with DC and in some experiments treated with IL-2 or anti-CTLA-4 mAb. Subsequently, cell lysates were prepared and Western blot analysis was performed. As shown in Fig. 6, addition of IL-2 (A) or inhibition of the CTLA-4 pathway (B) by blocking antibodies in the presence of DC induced a down-regulation of the G1 cell cycle inhibitor p27Kip1 in anergic CD4+ suppressor T cells after 24 and 48 h, indicating an important role of p27Kip1 and the G1 cell cycle arrest for the regulatory function of the T cells. No alteration of the expression of the cdk inhibitors p21Cip1/WAF1 or p15 was observed after treatment with IL-2 or anti-CTLA-4 mAb (data not shown).
2.7 CD4+CD25+ T cells do not contribute to the anergic state and the suppressor function of T cells co-cultured with IL-10-treated DC
Human CD4+CD25+ T cells are characterized as natural regulatory T cells with anergic properties inhibiting activated T cells in an antigen-unspecific fashion 17, 18. The following experiments were performed to analyze the function of this regulatory T cell population in our system of T cell anergy induced by IL-10-treated DC.
As shown in this report, the cell cycle arrest of anergic T cells in our model was mediated by an inhibited degradation of the cdk inhibitor p27Kip1. CD4+CD25+ T cells showed a similar high expression of p27Kip1 after primary culture (0 h) and restimulation (24 h) as compared to anergic suppressor T cells (Fig. 7A). The dramatically enhanced expression of p27Kip1 in anergic suppressor T cells after primary culture (0 h) and restimulation (24 h) was not affected by depletion of CD4+CD25+ T cells (Fig. 7B). These results suggest that the mechanism of T cell anergy induced by IL-10-treated DC is not mediated by CD4+CD25+ T cells.
As described previously, we observed that both T cell populations, CD4+CD25+ T cells and anergic T cells induced by IL-10-teated DC, showed an inhibited proliferation after primary stimulation and restimulation with DC as compared to control CD4+ T cells (Fig. 7C) 13, 17, 18. Notably, depletion of CD4+CD25+ T cells did not alter the reduced proliferation of suppressor T cells during primary culture and after restimulation (Fig. 7C). More importantly, co-culture experiments of suppressor T cells and activated control T cells demonstrated a markedly diminished T cell proliferation independently of the absence or presence of CD4+CD25+ T cells, indicating that these natural regulatory T cells are not involved in both the anergy state and regulatory activity of the anergic suppressor T cells induced by IL-10-treated DC (Fig. 7C).
In the present study we analyzed the molecular mechanisms underlying the induction of anergy and the suppressor function of T cells induced by co-culture with IL-10-treated DC. During the process of central and peripheral tolerance, the negative selection of autoreactive T cells induced by apoptotic pathways is one of the most important immunological mechanisms 19. Anergic T cells induced by IL-10-treated DC did not show an enhanced ratio of apoptotic death during primary culture or after restimulation as compared to optimally stimulated control cells. Therefore, the reduced proliferation observed after priming of CD4+ T cells with IL-10-treated DC may be due to a block in proliferation and not to a deletion of reactive T cells.
Blockade of cell cycle-specific kinases by regulatory proteins has been reported to play an essential role in the induction and maintenance of anergy 20. Analysis of DNA content showed an arrest in the G1 phase of anergic CD4+ suppressor T cells induced by IL-10-treated DC. Our findings correlate with the work of Gilbert and Weigle, who were the first to demonstrate that anergy induction in a Th1 clone mediated by chemically fixed APC was a result of G1a cell cycle blockade 21.
The mammalian cell cycle is regulated by cdk 22. Binding of negative regulatory proteins (cdk inhibitors) to the cdk-cyclin complexes leads to the inhibition of kinase activity and to blockade of cell cycle progression 6. Importantly, in our model anergic CD4+ T cells primed with IL-10-treated DC showed a markedly enhanced expression of p27Kip1 levels. These results are in agreement with findings showing an up-regulated amount of p27Kip1 in anergic T cells using various methods of tolerance induction. Induction of anergic T cells in the presence of the n-butyrated G1 blocker, in the absence of costimulation or by addition of IL-10 and TGF-β resulted in an increased expression of p27Kip1, indicating that the defective down-regulation of p27Kip1 is an essential mechanism of cell cycle arrest of anergic T cells 20, 23, 24.
On the other hand, Powell et al. reported that during the induction, maintenance and rechallenge phases of anergy p27Kip1 levels did not correlate with the anergic phenotype 25. In their study, the expression of p27Kip1 was down-regulated by IL-2, but the amount of IL-2 required to produce this effect was far lower than required to prevent the induction of anergy. Additionally, T cell lines from p27Kip1 knockout mice were anergized as well as T cells from mice heterozygous for p27Kip1. In contrast to the data mentioned above and our results, the data from Powell et al. serve to disassociate the ability of IL-2 to down-regulate p27Kip1 and its function to prevent or to reverse anergy. The discrepancies between these results might be due to different populations of T cells and methods of anergy induction as well as to the fact that many groups used only high doses of IL-2 in their experiments to reverse the anergic state of the T cells.
Because the G1 blockade could potentially be caused by alterations of other G1 cell cycle regulatory components, we also assessed the amounts of G1-specific cyclins and cdk. Although in our system anergic T cells synthesized G1-specific cyclins and cdk at equivalent or higher levels as the primed cells, the presence of increased p27Kip1 resulted in an aberrant ratio of p27Kip1 over the cyclin D-cdk holoenzyme, leading to blockade of its enzymatic activity and reduced levels of hyperphosphorylated Rb.
Tolerance in vivo and its in-vitro counterpart anergy are defined as the inability of T cells to produce IL-2 and to expand after stimulation with APC. Although anergic T cells fail to proliferate after restimulation, the cells retain their ability to proliferate in response to IL-2. Presumably, IL-2 can alter the balance between cdk inhibitors and cyclin-cdk complexes.For example, IL-2 has been shown to facilitate the ubiquitation and proteasome-dependent degradation of cdk inhibitors such as p27Kip126. The experiments described in this study suggest that anergy induction is associated with a potent G1 arrest that is mediated, in large parts, by the induction of the cdk inhibitor p27Kip1. This blockade was overcome by exogenous IL-2, leading to down-regulation of p27Kip1, cell cycle progression and, more importantly, to an inhibition of the suppressor activity of the T cells.
Evidence exists that CTLA-4 can act as a negative regulator of T cell activation by inhibiting T cell proliferation via regulation of cell cycle progression 27. The interaction between p27Kip1 and CTLA-4 during tolerance induction was analyzed in vivo using transgenic mice lacking CTLA-4 28. CTLA-4–/– T cells are resistant to tolerance induction as demonstrated by their proliferative responses and progression in the cell cycle, mediated by down-regulation of p27Kip1 and hyperphosphorylation of Rb. These results are in accordance with our observation that anergic CD4+ suppressor T cells are characterized by increased levels of p27Kip1 and hypoposphorylated Rb resulting in an impaired proliferation and also, as previously shown, by increased intracellular and surface levels of CTLA-4 15. Blocking of the CTLA-4 molecule inhibited the suppressor activity of the T cells by down-regulating the levels of the cdk inhibitor p27Kip1, indicating that CTLA-4 signaling is involved in the regulatory function of the anergic T cells via regulation of the cell cycle. In contrast to the addition of IL-2, blocking of the CTLA-4 pathway did not reverse the state of anergy. Therefore, the two different ways of T cell activation and cell cycle progression induced by different treatments both resulted in a down-regulation of the cdk inhibitor p27Kip1.
To the best of our knowledge, this is the first report demonstrating an interaction between cell cycle mechanisms and the suppressor function of anergic T cells. Anergy is defined as the inability of antigen-specific T cells to proliferate and produce IL-2 on rechallenge with fully competent APC delivering T cell receptor and costimulatory signals 5. However, anergic T cells are viable cells with the potency to be activated, as was shown by others and us that anergic CD4+ T cells after stimulation can exert regulatory effects as suppressor cells in vivo and in vitro15, 29–31. We performed additional experiments analyzing various surface antigens on T cells to characterize the anergic suppressor T cells in more detail. Antigens known to be expressed by regulatory T cells (e.g. CD103, GITR) were not found on anergic suppressor T cells (data not shown) with the exception of a high expression of CTLA-4 as described previously 15. Depletion of all CTLA-4+ T cells completely abolishes both the anergic and suppressor activity of the T cells, suggesting that all anergic T cells exhibit suppressor activity (data not shown). Furthermore, addition of IL-2, known to overcome the state of anergy, prevents the regulatory function of the suppressor T cells supporting the concept that not only a particular subpopulation of the T cells exhibits suppressor activity.
Our study additionally demonstrates that the suppressor activity of anergic T cells induced by IL-10-treated DC is associated with a G1 cell cycle arrest mediated by an inhibited degradation of the IL-2-dependent cdk inhibitor p27Kip1. Addition of IL-2 and blocking CTLA-4 signaling abrogated the regulatory function of suppressor T cells as demonstrated by an unrestricted T cell response and an enhanced production of IFN-γ of co-cultured activated T cells. Similar results were observed in experimental settings using mouse or human regulatory CD4+CD25+ T cells. Stimulation with IL-2 or anti-CD28 antibody did not only break the anergic state of these T cells but also simultaneously abrogated their suppressor activity 17, 18, 32. This natural regulatory T cell population, representing only 5%–10% of CD4+ T cells, inhibits the activation of T cells in an antigen-unspecific fashion mediated by cell-to-cell contacts. In contrast, IL-10-treated DC induce anergic regulatory T cells in both CD4+ and CD8+ T cell populations which are characterized by antigen-specific suppressor activity 15. As shown by our experiments, CD4+CD25+ T cells are not involved in our model of anergy, because depletion of CD4+CD25+ T cells did not alter the anergic state, the suppressor function or the expression of the cdk inhibitor p27Kip1 in T cells induced by IL-10-treated-DC.
Collectively, our results reveal p27Kip1 as an important factor for tolerance regulation and suppressor activities in T cells rendered anergic by stimulation with IL-10-treated DC. Our results suggest that the mechanisms of anergy induction and maintenance and the regulatory function of T cells may be regulated at the level of cell cycle progression involving IL-2, CTLA-4 signaling and the cdk inhibitor p27Kip1.
4 Materials and methods
4.1 Generation of dendritic cells
DC were generated as described previously 33. Briefly, PBMC were isolated from buffy coats using Ficoll gradients and suspended in culture dishes for 45 min. Non-adherent cells were rinsed off the plates and remaining cells were cultured in 3 ml X-VIVO 15 (BioWhittaker, MD) supplemented with 800 U/ml GM-CSF (Leukomax 300; Sandoz, Nürnberg, Germany), 1,000 U/ml IL-4 (DNAX, Palo Alto, CA) and 1% autologous plasma. At day 7, nonadherent cells were rinsed off the plates, resuspended in X-VIVO 15 supplemented with 800 U/ml GM-CSF and 1,000 U/ml IL-4, and additionally stimulated with 10 ng/ml IL-1β, 10 ng/ml TNF-α, 1,000 U/ml IL-6 (Strathmann Biotech, Hannover, Germany) and 1 μg/ml PGE2 (Minprostin; Upjohn, Erlangen, Germany). Two days prior to the end of the culture, 40 ng/ml IL-10 (DNAX) was added to the culture.
4.2 Isolation of T cell subpopulations and proliferation assays
CD4+ T cells were prepared from buffy coats using positive or negatively selected CD4 MACS® MultiSort beads (MACS systems; Miltenyi, Bergisch Gladbach, Germany) according to standard protocols (purity >95%). After detaching, CD4+ T cells were washed once in PBS plus 0.5% HSA plus 3 mM EDTA, stained with anti-CD25 beads (3 μl per 107 cells; Miltenyi Biotec) and positively selected according to the manufacturer's instructions (purity >90%). DC were prepared as described above, and 5×105 DC were co-cultured with 5×106 T cells per well in 3 ml X-VIVO 20 (BioWhittaker) supplemented with 1% autologous plasma of DC and 2 U/ml IL-2 (Chiron GmbH, Ratingen, Germany) in six-well plates. After 5 days T cells were separated by CD4+ microbeads (Miltenyi) and rested for 2 days in culture medium containing 2 U/ml IL-2. Subsequently, the T cells were restimulated in six-well plates coated with 1 μg/ml anti-CD3 mAb (OKT3, ATCC, CRL 8001) or mature DC, generated from the same donor as used for the primary culture or left unstimulated.
4.3 Co-culture experiments
Anergic CD4+ suppressor T cells and activated allogeneic CD4+ effector T cells were cultured as described above. To assess the suppressor activity of anergic T cells in the alloantigen-specific system, 1×105 anergic CD4+ and 1×105 syngeneic activated T cells restimulated with 2×104 DC (generated from the same donor as used in the primary culture) were used. Proliferation was measured 48–72 h later by thymidine incorporation. To overcome the anergy or suppressor activity, anergic T cells were preincubated with 100 U/ml IL-2 100 U/ml (Proleukin; Chiron, Emeryville, CA). For blocking experiments, anergic T cells were treated with antibodies against CTLA-4 (Fab fragments; Alexis Biochemicals, San Diego, CA; PharMingen, Hamburg, Germany) (10 μg/ml) prior to restimulation or co-culture experiments.
4.4 Annexin V binding assay
Detection of early stages of apoptosis was carried out using a phosphatidylserine detection kit (Immuno Quality Products, Groningen, Netherlands) and analyzed by flow cytometry (FACScan, Becton Dickinson, Heidelberg, Germany). Annexin V+PI– cells were determined as apoptotic cells.
4.5 TUNEL assay
TUNEL was performed using MEBSTAIN Apoptosis kit II (Immunotech, Marseille, France). Labeled cells were visualized by avidin-FITC staining and flow cytometric analysis.
4.6 Cell cycle analysis
Cell cycle analysis was performed by DNA staining with PI. Cells were washed twice in cold PBS and then fixed in 70% ethanol at 4°C for at least 1 h. The samples were rehydrated in cold PBS, treated with 50 μg/ml RNase A and stained with 50 μg/ml PI. DNA content was measured using a FACScan. G1 (2N), S (2N to 4N) and G2 fraction (4N) were determined with MultiCycle Cell Cycle Analysis Software (Phoenix Flow Systems, San Diego, CA).
Cell lysates were prepared using fresh lysis buffer containing 50 mM Tris (Roth, Karlsruhe, Germany), 150 mM NaCl (Roth), 5 mM EDTA, 1,5 mM MgCl2, 1% Triton X (Sigma, Deisenhofen, Germany), 5% glycerol (Roth) supplemented with 1 mM Pefabloc SC (Roche Diagnostics, Mannheim, Germany), 10 μg/ml aprotinin (Sigma), 10 μg/ml leupeptin (Boehringer, Mannheim, Germany), 5 mM benzamidine (Sigma), 1 mM activated Na3VO4 (Sigma), 10 mM NaF (Merck, Darmstadt, Germany) and 10 mM Na2HPO4 (Sigma). Cells (105) were washed twice in ice-cold PBS and subsequently resuspended in 50 μl lysis buffer. Incubation was performed for 30 min on ice and mixed gently. Insoluble fragments were pelleted by centrifugation for 30 min at 10,000×g.
Proteins were separated by SDS-polyacrylamide gel electrophoresis (12%; analysis of Rb: 8%) and blotted onto polyvinylidene difluoride membranes (Millipore, Bedford, MA) using a semi-dry system. Membranes were adsorbed with blocking reagent (Roti-Block; Roth) for 1 h at room temperature and subsequently probed with mouse mAb against cyclin D2, D3 and E, cdk2, cdk4, p27Kip1, p21Cip1/WAF1, p15, pRb (Clone G-3–245) (Signal Transduction Laboratories/BD, Heidelberg, Germany) and rabbit polyclonal antibody against actin (Sigma) overnight at 4°C. Immunodetection was performed by incubation with horseradish peroxidase-conjugated antibodies (goat anti-mouse, Dianova, Hamburg, Germany; and goat anti-rabbit, Santa Cruz, CA). Proteins were visualized by enhanced chemiluminescence (ECL) (Amersham Pharmacia, Freiburg, Germany) using Hyperfilm ECL. For reprobing, membranes were incubated in stripping-buffer (62.5 mM Tris, 2% SDS, 100 mM β-mercaptoethanol) for 45 min at 50°C. Equal protein loading was controlled by reprobing with anti-actin antibody.
This work was supported by a grant of the DFG (SFB548/B6) and of the University of Mainz (MAIFOR) to K. S. The authors wish to thank Drs. E. von Stebut and M. Maurer for critically reading the manuscript.