Calcineurin (Cn) is a Ca2+/calmodulin-dependent phosphatase that dephosphorylates and activates NFAT, a transcription factor essential for T cell activation. T lymphocytes predominantly express the calcineurin Aβ (CnAβ) isoform, and the deletion of the CnAβ gene results in defective T cell proliferation and IL-2 production in response to TCR stimulation. In this study, we show that CnAβ enhances the spontaneous survival of naive T cells by maintaining high levels of Bcl-2, a critical homeostatic survival factor for naive T cells. T cells obtained from CnAβ−/− mice displayed accelerated spontaneous apoptosis. The observed apoptosis of the CnAβ−/− T cells was prevented by IL-7 and IL-15, two cytokines critical for the homeostatic survival of naive T cells. Furthermore, CD4+ or CD8+ single positive CnAβ−/− thymocytes also underwent accelerated apoptosis. However, no obvious difference in the apoptosis of CD4+CD8+ double positive thymocytes was observed between CnAβ−/− and wild-type mice, suggesting a specific function of CnAβ in the survival of single positive T cells. Bcl-2 levels were found to be significantly lower in CnAβ−/− T cells. Transgenic expression of Bcl-xL restored the survival of the CnAβ−/− T cells. Thus, in addition to its role in mediating TCR signals essential for T cell activation, CnAβ is also required for the homeostatic survival of naive T cells.
Calcineurin (Cn)3 is a Ca2+/calmodulin-dependent phosphatase consisting of a catalytic and calmodulin-binding subunit A (CnA) and a Ca2+-binding regulatory subunit B (CnB) (1). In the resting T cells, enzymatic activity of CnA is kept inactive by binding to CnB. Increased intracellular Ca2+ concentration following TCR stimulation leads to the activation of calmodulin that then binds to CnA, resulting in the activation of enzymatic activity of CnA. Activated Cn dephosphorylates and activates NFAT. Dephosphorylated NFAT translocates to the nucleus and stimulates its target gene transcription, IL-2, which is required for T cell activation. The critical role of calcineurin is also demonstrated by the discovery of the immunosuppressive drugs FK506 and cyclosporin A, which are broadly used to prevent allograft rejection. FK506 and cyclosporin A, when bound to their respective binding proteins, FKBP12 and cyclophilin A, inhibit Cn activity and thus prevent T cell activation (2). Three different isoforms of CnA transcribed from distinct genes have been identified (3). T cells predominantly express CnAβ isoforms. Consistently, T cells deficient in CnAβ, but not CnAα, displayed significant defects in T cell activation (4, 5). Thus, CnAβ mediates TCR signals required for T cell activation.
In addition to T cell activation, TCR signals are also required for the homeostatic survival of the naive T cells (6, 7). Adoptive transfer of naive T cells to recipient mice with mismatched MHC type or MHC-deficient mice indicated that long-term survival of naive T cells requires the interactions of TCR and MHC-peptide complexes (8, 9). When the peptides with different affinities for TCR were tested in the adoptive transfer experiments, it was found that the low affinity peptide that results in weak interactions is required for naive T cell survival, whereas the high affinity peptide that results in strong interactions between TCR and MHC complexes is required for T cell activation (10). These results suggest that transduction of different signals through TCR is required to mediate T cell activation and homeostatic survival. Many signaling molecules including CnAβ are known to govern the TCR signaling events for T cell activation, but little is known about the signaling events that mediate the survival of naive T cells.
Anti-apoptotic Bcl-2 is the first important gene identified to regulate programmed cell death (11, 12). During T cell development, Bcl-2 is predominantly expressed in CD4+ or CD8+ single positive T cells. Bcl-xL, another critical anti-apoptotic molecule of the Bcl-2 family, is specifically up-regulated in CD4+CD8+ double positive cells (13). Bcl-2 thus primarily enhances the survival of the single positive cells, and Bcl-xL is the survival molecule for double positive cells. The essential role of Bcl-xL in the survival of double positive cells was demonstrated by the apoptosis of Bcl-xL−/− double positive T cells (13, 14). Bcl-2−/− mice have reduced peripheral T cells (14), suggesting that Bcl-2 plays a role in the homeostatic survival of naive T cells. In this study we show that CD4+ or CD8+ T cells obtained from CnAβ−/− mice displayed accelerated spontaneous apoptosis as well as significantly reduced Bcl-2 levels. Forced expression of Bcl-xL restored survival to CnAβ−/− T cells. CnAβ thus regulates the homeostatic survival of CD4+ and CD8+ single positive T cells by maintaining high levels of Bcl-2.
Materials and Methods
CnAβ−/− mice were originally described by Bueno et al. (4). Mice were kept in the animal facility of the Biological Resource Laboratory at the University of Illinois (Chicago, IL) following the university guidelines. Bcl-xLTg mice (where Tg is transgenic) have been previously described (15). CnAβ−/− and Bcl-xLTg mice were intercrossed to generate CnAβ−/−/Bcl-xLTg. Wild-type (WT) C57BL/6 mice were purchased from The Jackson Laboratory.
Preparation of lymphocytes
Thymus, spleens, or mesenteric lymph nodes were harvested from 8- to 12-wk-old WT, CnAβ−/−, Bcl-xLTg, and CnAβ−/−/Bcl-xLTg mice and dissociated into single-cell suspensions. RBCs were lysed by adding 5 ml of RBC lysing buffer. CD4+ or CD8+ T cells were purified from splenocytes using a CD4+ or CD8+ isolation kit (Miltenyi Biotec) according to the manufacturer’s protocol. CD4+ or CD8+ T cells were separated using an autoMACS magnetic cell sorter (Miltenyi Biotec). The purity of the T cells, determined by a flow cytometry, was >90%.
Enriched T cells (0.3 × 105 cells per well) were cultured in 96-well plates in the absence or presence of 10 ng/ml recombinant mouse IL-2, IL-7, or IL-15 (R&D Systems) and stimulated for 48 h in the presence or absence of anti-CD3 and CD28 Abs. After the 48 h of stimulation the cultures were pulsed with 1 μCi of [3H]thymidine per well for 12 h.
Purified splenic CD4+ or CD8+ T cells were cultured in the absence or presence of 10 ng/ml recombinant mouse IL-2 or IL-7 or IL-15 (R&D Systems), and stimulated as described in the proliferation assay for various time points. Cells were washed once with ice-cold annexin V binding buffer (BD Pharmingen) and cell pellets were stained with PE-conjugated annexin V and 7-aminoactinomycin D 7-AAD (BD Pharmingen) according to the manufacturer’s protocol. Analyses were performed by a flow cytometer with CELLQuest software. Viable cells were analyzed by excluding dead cells using low forward light scatter.
Western blot analysis
The protein levels of Bcl-2 and β-actin were analyzed by Western blotting. Enriched T cells were purified as described above. Enriched T cells (2 × 106) from WT or CnAβ−/− mice were lysed by 1% Nonidet P-40-containing buffer as described previously (16). Anti-Bcl-2 (BD Pharmingen) and anti-β-actin Abs were used for Western blot analysis.
A 2.6-kb Bcl-2 promoter element was cloned upstream of a luciferase gene (pGL2-Basic vector) (17). Jurkat or Jurkat-Tag cells (107/ml) were transfected by electroporation with 5 μg of the Bcl-2 luciferase reporter plasmid together with 15 μg of the constitutively active or the inhibitory calcineurin B subunit expression plasmid. Identical amounts of the corresponding parental vectors were used as control. For normalization, 100 ng of the Renilla luciferase reporter vector pTK-Renilla-LUC was used. After 36 h, cells were lysed and assayed for dual luciferase activity (Promega).
We performed Student’s t test for statistical analysis.
CnAβ−/− T cells undergo accelerated spontaneous apoptosis
Previous studies including ours have shown that protein kinase C θ (PKC-θ) regulates the Ca2+/NFAT pathway (18, 19, 20) as well as T cell survival (16, 21, 22); we thus intended to determine whether Cn plays a role in PKC-θ-regulated T cell survival. Mice deficient in catalytic subunit Aβ (CnAβ−/−) were used in this study because, similar to PKC-θ, CnAβ also mediates TCR signals required for T cell activation (4). We first examined T cell proliferation in response to TCR stimulation using a [3H]thymidine incorporation assay (Fig. 1,a). Consistent with previous results (4), CnAβ−/− T cells displayed significant defects in CD3/CD28 stimulation-induced proliferation, and such defects could not be prevented by exogenous IL-2. This result confirmed a critical role of CnAβ in T cell activation. To determine whether cell death contributed to the significantly reduced [3H]thymidine incorporation by CnAβ−/− T cells, we performed a apoptosis assay using annexin V and 7-aminoactinomycin D as described previously (16, 23, 24). We first analyzed apoptosis of the CD4+ T cells obtained from WT and CnAβ−/− mice (Fig. 1,b). CD4+ T cells were left in medium alone or treated with IL-2 for 36 h. Apoptotic cells were then detected with a flow cytometer. Approximately 30% of the WT CD4+ T cells were found dead after 36 h in medium. Compared with the WT T cells, CnAβ−/− T cells displayed greatly increased apoptosis as indicated by the fact that ∼60% of the CnAβ−/− T cell underwent apoptosis. IL-2 treatment could not prevent accelerated spontaneous apoptosis of the CnAβ−/− cells. These results suggest that CnAβ is a critical signaling molecule required for maintaining the spontaneous survival of the naive CD4+ T cells. TCRs deliver signals required for stimulating proliferation and enhancing survival (25, 26). Such survival signals ensure the completion of the T cell activation process essential for differentiating naive T cells to effectors that mediate actual immune responses (27). We therefore determined whether stimulation of TCR can prevent the apoptosis of the CnAβ−/− T cells (Fig. 1,b). Compared with the unstimulated CD4+ T cells (30% apoptotic cells), CD3/CD28 stimulation did enhance the survival of the WT CD4+ T cells (20% apoptotic cells). However, the survival of the CnAβ−/− T cells was not significantly changed by CD3/CD28 stimulation even in the presence of IL-2, indicating that CnAβ mediates the TCR signals required for enhancing T cell survival. Similar results were obtained from analyzing the apoptosis of CD8+ T cells (Fig. 1 c). Altogether, CnAβ mediates the critical survival signals for CD4+ and CD8+ T cells.
The preceding experiments used defective CnAβ−/− T cells; we next determined the effects of blocking the Cn pathway on the survival of the WT T cells. FK506, a specific inhibitor for Cn, was used to treat the purified CD4+ and CD8+ T cells (Fig. 1 d). Indeed, inhibition of Cn by FK506 greatly increased spontaneous apoptosis both in CD4+ and in CD8+ T cells from ∼30% to 60%, confirming the critical role of Cn in the survival of naive T cells.
IL-7 and IL-15 prevent the apoptosis of the CnAβ−/− T cells
IL-7 and IL-15 have been shown to be critical for the homeostatic survival of the naive T cells (28, 29). We therefore determined whether these two cytokines can prevent spontaneous apoptosis of the CnAβ−/− CD4+ T cells (Fig. 2,a). Consistent with their roles in the homeostatic survival of T cells, IL-7 and IL-15 enhanced the survival of WT T cells. IL-7 treatment inhibited apoptosis of the CnAβ−/− CD4+ T cells from ∼60 to ∼40%. IL-15 treatment further inhibited apoptosis of the CnAβ−/− CD4+ T cells to ∼30%, which is equivalent to that observed in WT CD4+ T cells. These results suggest that IL-7 and IL-15 can compensate for CnAβ-mediated survival. Furthermore, because IL-7 and IL-15 can enhance T cell survival in the absence of CnAβ, IL-7- and IL-15-mediated survival is independent of CnAβ. CD4+ T cells were also stimulated with CD3/CD28 in the presence or absence of IL-7 or IL-15. Consistent with the previous results, TCR stimulation could not prevent apoptosis of the CnAβ−/− T cells. IL-7 and IL-15 treatment, although inhibit spontaneous apoptosis, had no significant effects on apoptosis of the CnAβ−/− CD4+ T cells induced by TCR stimulation. This result suggests that the spontaneous apoptosis does not likely contribute to the apoptosis induced by TCR crosslinking. The role of CnAβ in the spontaneous survival and the TCR stimulation-mediated survival signals are likely two separate events. Similar results were also observed from analyzing the apoptosis of CD8+ T cells (Fig. 2 b). Thus, CnAβ, together with cytokines such as IL-7 and IL-15, is critical for maintaining the survival of naive T cells.
CnAβ is required for the survival of the CD4+ or CD8+ single positive but not the CD4+CD8+ double positive thymocytes
The thymus is the location for T cell differentiation and maturation. Immature CD4+CD8+ thymocytes are subject to positive and negative selection in the thymus. Only the cells that meet the selection criteria are allowed to mature into CD4+ or CD8+ single positive cells and migrate out of the thymus. We thus also determined whether CnAβ provides survival signals for thymocytes. Spontaneous apoptosis assays were performed using thymocytes obtained from both WT and CnAβ−/− mice (Fig. 3) as described previously (23, 24). No obvious differences were detected in the apoptosis of CD4+CD8+ thymocytes between WT and CnAβ−/− T cells. However, increased apoptosis was observed in CD4+ (∼40%) and CD8+ (∼65%) single positive T cells deficient in CnAβ when compared with the corresponding WT CD4+ (∼20%) and CD8+ (∼40%) T cells. This result suggests that CnAβ is required for enhancing the survival of CD4+ and CD8+ single positive, but not the CD4+CD8+ double positive, thymocytes.
CnAβ−/− T cells express greatly reduced levels of anti-apoptotic protein Bcl-2
Bcl-2 is a critical survival molecule for naive T cells, because Bcl-2−/− T cells are susceptible to apoptosis and transgenic expression of Bcl-2 protects T cells from apoptosis (30, 31). We thus detected protein levels of Bcl-2 in T cells obtained from spleens using Western blot analysis (Fig. 4 a). Indeed, CnAβ−/− T cells had greatly reduced levels of Bcl-2 compared with that of the WT T cells. CnAβ is thus required to maintain high levels of Bcl-2 in naive T cells.
The 2 promoter contains multiple consensus NFAT-binding sites, and the mutation of two critical NFAT-binding sites reduces the Bcl-2 promoter activity in cardiac myocytes (32). We thus decided to determine whether Cn can regulate Bcl-2 promoter activity in T cells. We obtained a Bcl-2-luciferase reporter containing a 2.6-kb Bcl-2 promoter region described in a previous report (17). The Bcl-2 reporter was transfected into Jurkat cells by electroporation alone or together with an expression plasmid encoding the inhibitory B subunit (CnB) or the constitutively active Cn (Active-Cn) (Fig. 4 b). CnB inhibited ∼50% of the Bcl-2 reporter activity. Constitutively active Cn drastically stimulated Bcl-2 reporter activity. This result suggests that Cn can directly stimulate Bcl-2 promoter activity.
Transgenic expression of Bcl-xL restored the survival of CnAβ−/− T cells
To determine whether the reduced Bcl-2 observed in CnAβ−/− mice is responsible for the accelerated apoptosis of the CnAβ−/− T cells, CnAβ−/− mice were crossed with Bcl-xLTg mice (Bcl-xLTg). Bcl-xL is another anti-apoptotic member of the Bcl-2 family. Similar to Bcl-2, forced expression of Bcl-xL protects cells from apoptosis (15). Overexpression of Bcl-xL in Bcl-xLTg mice is achieved by using a Lck proximal promoter that targets a transgene specifically to T cell compartments. Apoptosis was compared between T cells obtained from the spleens of WT, CnAβ−/−, Bcl-xLTg, and CnAβ−/−/Bcl-xLTg mice (Fig. 5, a and b). Consistent with the results observed in Fig. 1, CnAβ−/− CD4+ T cells underwent increased apoptosis compared with that of the WT CD4+ T cells. However, forced expression of Bcl-xL restored the survival of the CnAβ−/− CD4+ T cells to levels similar to those observed in the WT CD4+ T cells. Similar results were also observed in CD8+ T cells (Fig. 5,b). These results suggest that CnAβ-regulated T cell survival is dependent on the Bcl-2 survival factor. We then counted the number of CD4+ (Fig. 5,c) and CD8+ (Fig. 5 d) T cells in the spleens of different genotypes of mice. The number of both CD4+ and CD8+ T cells in CnAβ−/− mice were greatly reduced compared with that of the WT mice. Surprisingly, Bcl-xLTg, which restored survival to CnAβ−/− T cells, only slightly increased the number of CD4+ and CD8+ cells, but far from the levels observed in the WT mice.
We next examined the effects of Bcl-xLTg on the survival of CnAβ−/− thymocytes (Fig. 5, c and d). Similar to peripheral T cells, apoptosis of the CnAβ−/− CD4+ (Fig. 5,c) and CD8+ (Fig. 5,d) thymocytes was inhibited by Bcl-xLTg. We also counted the number of each subset of thymocytes (Fig. 5, e–h). Compared with the WT mice, the total thymocyte number was reduced in CnAβ−/− mice (Fig. 5,e), which was also reflected by the reduced CD4+CD8+ double positive thymocytes in CnAβ−/− mice (Fig. 5,f). The most significant reduction was observed in the number of CD4+ (Fig. 5,f) and CD8+ (Fig. 5 h) single positive cells in CnAβ−/− mice. Consistent with the observation in the number of peripheral T cells, Bcl-xLTg, although it could slightly increase the number of single positive T cells, could not restore the number of CD4+ and CD8+ thymocytes in CnAβ−/− mice to the WT levels. The fact that Bcl-xLTg restored the survival, but not the number, of single positive T cells suggests that CnAβ regulates additional functions such as the proliferation of the T cells in addition to regulating survival.
Using CnAβ−/− mice, we showed that deletion of the CnAβ gene leads to an increased apoptosis of naive CD4+ and CD8+ T cells likely due to the reduced survival molecule Bcl-2. Apoptosis of the CnAβ−/− T cells could be prevented by IL-7, IL-15, or the transgenic expression of Bcl-xL. The survival of naive T cells depends on the signals generated from the continuous interactions between TCR and MHC complexes and from cytokines such as IL-7 and IL-15 (33, 34, 35). It is likely that CnAβ mediates the TCR signals essential for the homeostatic survival of naive T cells.
After emigrating from the thymus, the naive T cell can survive an extended period of time only in the presence of continuous interaction between TCR and MHC complexes (33, 34, 35). By adoptive transfer experiments, Takeda et al. (36) have shown that the long-term survival of CD4+ T cells requires MHC II, because the number of CD4 T cells adoptively transferred to MHC II-deficient mice is gradually reduced since the number of CD4 T cells transferred to WT mice. Similarly, survival of the naive CD8+ T cells also depends on the presence of MHC I binding to peptide Ags (8). Furthermore, conditional ablation of the TCR α-chain gene in the mature T cell stage leads to a shorter lifespan of naive T cells (37), suggesting that TCR-mediated signals are essential for naive T cell survival. Therefore, signals resulting from the interaction between MHC and TCR maintain the survival of naive T cells. In addition to TCR signals, cytokines such as IL-7 and IL-15 are also required for the survival of naive T cells. Naive T cells survive for only a short period of time when adoptively transferred to IL-7-deficient mice (28, 38). Elevating IL-7 levels either by expressing a transgene or injecting rIL-7 results in a significantly increased T cell population (39, 40). IL-15 seems to have a similar function as IL-7 in promoting T cell survival, likely due to the fact that they share a common receptor subunit, γc (41). Although it is known that both the TCR and cytokines are required for the survival of naive T cells in vivo, little is known about the molecules mediating the survival signals. Our results demonstrated that CnAβ mediates the critical survival signals for naive T cells, because lack of CnAβ leads to a significantly increased spontaneous apoptosis. CnAβ−/− T cells had increased levels of phosphorylated NFAT, indicating the lack of activation of NFAT (4). It is thus likely that CnAβ regulates T cell survival via activating NFAT. As both TCR- and cytokine-mediated signals are required for the naive T cell survival, it is not clear whether CnAβ mediates the survival signals for TCR or cytokines or both. Our data indicated that IL-7 or IL-15 is able to enhance the survival of CnAβ−/− mice, suggesting that IL-7 or IL-15 can deliver survival signals independently of CnAβ. It is unlikely that activation-induced cell death (AICD) is responsible for the apoptosis of the CnAβ−/− T cells, as CnAβ is required for T cell activation (4). Consistently, we did not find more activated T cells in CnAβ−/− mice by examining T cell activation markers such as CD25. Furthermore, we have shown that CnAβ−/− T cell apoptosis induced by TCR stimulation cannot be prevented by IL-7 or IL-15 (Fig. 2). However, the spontaneous apoptosis we focused on can be inhibited by IL-7 or IL-15. In addition, we have shown previously that up-regulation of FasL also requires activation of the calcineurin/NFAT pathway (16). Therefore, calcineurin is critical for both T cell activation and up-regulation of FasL that is critical for AICD. It is thus unlikely that the observed apoptosis of the CnAβ−/− T cells results from AICD.
Bcl-2 is the ultimate downstream antiapoptotic molecule regulating the survival of naive T cells. Bcl-2−/− T cells are susceptible to apoptosis (15). However, transgenic expression of either Bcl-2 or Bcl-xL enhances the survival of T cells (15). Our data support the idea that CnAβ is a critical signaling molecule regulating Bcl-2 expression in T cells, because the deletion of CnAβ resulted in significantly reduced Bcl-2 levels. Furthermore, blocking the Cn pathway inhibits Bcl-2 reporter activity, suggesting that Cn can stimulate Bcl-2 promoter activity. Thus, CnAβ up-regulates Bcl-2 levels at least in part via transcriptional stimulation of the Bcl-2 gene. However, using fibroblast cells (BHK-21), Shibasaki et al. (42) have shown that the forced expression of an active Cn is sufficient to induce apoptosis. This result, in contrast to ours, suggests that Cn is a proapoptotic molecule. It is possible that Cn plays different roles in different types of cells, as we used T cells instead of fibroblasts in the experiments. However, persistently high levels of intracellular Ca2+ concentration and calcineurin activation can also induce T cell apoptosis or anergy (43, 44). It seems that appropriate Cn-mediated signals are required for T cell survival. But overwhelming Cn signals can induce apoptosis. Macian et al. (44) have reported that activation of NFAT, a downstream effector of Cn, induces very different target genes, dependent on the availability of its partner AP-1. In the absence of AP-1, Ca2+/Cn signaling induces genes associated with anergy. However, in the presence of AP-1 anergy genes are mostly not induced. Therefore, by integrating other signals, the activation of Cn may regulate different or even opposite functions. The MHC-TCR interactions required for the naive T cell survival likely lead to the activation of other signaling pathways in addition to CnAβ. These signaling pathways, together with Cn, ensure the survival of naive T cells by up-regulating Bcl-2.
Cn is a critical signaling molecule required for T cell activation-induced proliferation. The critical role of Cn in T cell activation is demonstrated by its pharmacological inhibitors, cyclosporin A and FK506, which are widely used for preventing allograft rejection (2). We demonstrated in this article that CnAβ is required for the survival of naive T cells. Although our data show that forced expression of Bcl-xL in CnAβ−/− T cells can restore the survival, it failed to completely restore the peripheral T cell number to the WT mice level. It is possible that the reduced T cell number results from both apoptosis and reduced proliferation in the absence of CnAβ. Thus, inhibition of apoptosis with Bcl-xL can partially restore the cell number in CnAβ−/− mice. Because calcineurin is critical for T cell survival, cyclosporin A and FK506 may inhibit T cell-mediated immune responses not only by preventing T cell activation but also by inducing the apoptosis of naive T cells.
We thank Dr. Linda Boxer for providing Bcl-2 reporter, Dr. Anjana Rao for expression plasmids encoding calcineurin B and constitutively active Cn, Dr. Marisa-Luisa Allegre for sharing Bcl-xL transgenic mice, and Dr. Mark Dizik for critically reading the manuscript and helpful discussion.
The authors have no financial conflict of interest.
The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
This work was supported by National Institutes of Health R01-AI053147.
Abbreviations used in this paper: Cn, calcineurin; AICD, activation-induced cell death; CnA, calcineurin subunit A (catalytic and calmodulin binding); CnB, calcineurin subunit B (Ca2+ binding regulatory); PKC-θ, protein kinase C θ; Tg, transgenic; WT, wild type.