Introduction

In wild-type animals, the survival of T lymphocytes is carefully regulated, and abnormal survival of lymphocytes contributes to the development of immune deficiency, lymphoid malignancy and autoimmune diseases.1 Bcl-2 family members are important regulators of apoptosis and play an essential role in the maintenance of homeostasis in the immune system.2

In the periphery, naive CD4+ or CD8+ lymphocytes may survive for long periods, up to several months in mice, in the absence of overt exposure to antigens (Ag). Their survival depends on the engagement of T-cell receptors (TCRs) by selecting MHC proteins3,4,5 and on the availability of cytokines, especially IL-7.6,7,8 A link between T-cell survival and the Bcl-2 protein has been provided following the demonstration that IL-7 plays an essential role in blocking apoptosis during differentiation and activation of T lymphocytes. Indeed, the constitutive expression of Bcl-2 protein was demonstrated to rescue T-cell numbers and function in IL-7-receptor-deficient mice.9,10 However, the role of Bcl-2 family members in the control of naive T cells survival is still not fully understood. Mice invalidated for bcl-2 gene in lymphocytes showed a rapid decline of peripheral T cells after 2 weeks of age partly because of a shortened lifespan of mature T cells.11 In contrast, the absence of bcl-x affects the lifespan and apoptosis of developing immature lymphoid cells rather than survival of their mature counterparts.12

In response to Ags, signals transduced by engagement of TCR and costimulatory molecules drive the entry of T cells into the cell cycle, the expression of cytokine receptors and the production of cytokines. Cytokines will act in an autocrine fashion to induce further divisions of activated T cells. However once activated, T cells are short lived and only a minority of effector T cells survive to become memory T cells. Analysis of Bcl-2 and Bcl-xl proteins showed that their expression is downregulated at the peak of the T-cell response, just before they died in vivo,13,14 whereas memory cells persisting after clearance of Ag show upregulation of both Bcl-2 and Bcl-xl again.15,16,17 Among the proapoptotic Bcl-2 family members, Bim seems to play a critical role in the death of mature T cells. Indeed both resting and activated T cells, from Bim-deficient mice are resistant to cytokines withdrawal,18 and Bim deficiency prevents deletion of superantigen-activated T cells both in vitro and in vivo.14 Interestingly, Bim deficiency was demonstrated to restore normal lifespan to bcl-2−/− T cells following cytokine deprivation.19 Altogether, those studies indicate that death of T cells following antigenic stimulation or cytokine deprivation is largely dependent on Bim, and that the ratio of Bim to Bcl-2 is crucial for their survival.

In this paper we have analyzed the regulation of A1, an antiapoptotic member of the Bcl-2 family, following TCR and cytokine receptors engagement in both naive and Ag-activated T cells. A1 was originally identified as a GM-CSF-inducible gene product.20 It is expressed in hematopoietic cell lineage including T and B cells, macrophages, dendritic cells and neutrophils.20,21 A1 shares many properties with Bcl-2 and Bcl-xl. Like these proteins, A1 can heterodimerize with proapoptotic proteins such as Bax22 or Bid23 and was shown to decrease cell death when transfected into cell lines.24,25,26 In B lymphocytes, A1 expression increases when immature murine B cells transit to the stage of long-lived mature B cells.27 Moreover, it has been shown that CD40 stimulates the expression of A1 mRNA in primary murine B cells and in WEHI 231 cells, and that A1-transduced WEHI 231 cells have a significantly increased survival rate in the presence of anti-IgM.26 Expression of A1 is also regulated during T-cell development,28 with a strong expression in CD4+CD8+ thymocytes and a moderate level in both CD4−CD8− and CD4+ or CD8+ single positive thymocytes. Finally, A1 was shown to be rapidly induced by mitogens in an Rel-dependent manner in peripheral lymphocytes.25,29

We focused our interest on the role of TCR versus cytokine receptors engagement in the regulation of A1 expression. We took advantage of both DO11.10 and F5 TCR-transgenic T cells that can be stimulated in vitro and in vivo with antigenic peptides, to analyze A1 expression.

Results

T-cell activation rapidly and transiently induces A1 expression

To test how A1 expression is regulated following T-cell activation, splenocytes from C57BL/10 mice were stimulated in the presence of mitogenic monoclonal antibody (mAb) to CD3 for 3.5, 7, 24 or 48 h and RNA analyzed in an mAPO-2 multiprobe RNase protection assay (Figure 1a). A low level of mRNA coding for A1 was detected in unstimulated cells. Stimulation with mAb to CD3 rapidly induced its expression with a maximum observed at 7 h. Expression returned to the basal level at 48 h. Costimulation with mAb to CD28 did not modify the level of A1 expression or kinetic, whereas in the same experiment it increased the expression of Bcl-xl mRNA (Figure 1b).

Figure 1
figure 1

Expression of A1, Bcl-xl and Bcl-2 mRNA in T-cell activation. Purified C57BL/10 T cells were stimulated with (a) anti-CD3 alone (10 μg/ml) or (b) in combination with anti-CD28 (2 μg/ml). At different times, mRNA were extracted from purified T cells and levels of mRNA coding for A1, Bcl-xl and Bcl-2 were measured by RNase protection assays. Amounts of mRNA were normalized using the L32 internal control. Normalized results are expressed as fold increase from values obtained at 0 h. Results shown are representative of three independent experiments. (c) Purified T cells were stimulated with PMA (2 ng/ml) or ionomycine (1 μM) or both in combination, or with anti-CD3 + anti-CD28 in the presence of MG132 (1 μM), or EDTA (1 mM) or CsA (0.1 μg/ml). After 7 h of culture, mRNA were extracted and analyzed by RNase protection assays as described before. Results are expressed as mean±S.E.M. of three independent experiments

Full size image

Multiple signaling pathways are activated downstream of TCR, and can be involved in the upregulation of A1. A1 or its human homologue Bfl-1 were first demonstrated to be transcriptionally regulated by NF-κB in lymphocytes,25,29 however, more recently A1 expression in mast cells was reported to be dependent on calcium.30 We observed that PMA alone efficiently increased A1 mRNA in T lymphocytes, whereas ionomycine alone only slightly increased A1 expression (two fold). However, ionomycine synergized with PMA for induction of A1 expression (Figure 1c). As expected, expression of A1 mRNA induced by stimulation with mAb to CD3 and CD28 is prevented by the proteasome inhibitor MG132, which inhibits NF-κB activation.31 Expression of A1 is also prevented by EDTA and CsA (Figure 1c), suggesting that both NF-κB- and calcium-dependent signals are required for A1 expression in lymphocytes.

TCR engagement by antigenic peptide in vitro induces A1 expression in naive T cells

We next studied A1 expression following antigenic stimulation, in both CD8+ and CD4+ T cells. Splenocytes from F5 TCR-transgenic mice were stimulated with NP68 peptide (10 nM) alone or in the presence of exogenous IL-2 or blocking anti-IL-2 mAb. CD8+ T cells were then purified and mRNA levels for A1, Bcl-xl and Bcl-2 were analyzed (Figure 2a). Peptide stimulation markedly induced the expression of mRNA coding for A1 with a maximum at 6 h (∼20-fold). Expression returned to a basal level at 48 h. Similar to A1, Bcl-xl mRNA was transiently upregulated by peptide stimulation alone in CD8+ T cells. Addition of exogenous IL-2 or inhibition of IL-2-dependent signals by antagonistic anti-IL-2 mAb did not modify the kinetic of A1 or Bcl-xl mRNA expression (Figure 2a). In contrast, the expression of Bcl-2 mRNA was only minimally induced by peptide alone, but was upregulated (∼ three fold), with a delayed kinetic as compared with A1 and Bcl-xl, when exogenous IL-2 is added to peptide stimulation (Figure 2a). Similar kinetics of A1 and Bcl-xl expression were observed when CD4+ T cells from D011.10 TCR-transgenic mice were stimulated with OVA peptide (Figure 2b). Reinforcing the idea that IL-2 is not required for the upregulation of A1, we observed that OVA peptide strongly upregulated A1 transcription in IL-2−/− DO11.10 mice (Figure 2b).

Figure 2
figure 2

Expression of A1, Bcl-xl and Bcl-2 mRNA in CD8+ T cells following peptide stimulation in vitro. (a) Splenocytes from F5 mice were stimulated with NP68 peptide (10 nM) alone or in the presence of exogenous IL-2 (50 ng/ml) or in the presence of the blocking anti-IL-2 mAb S4B6 (50 ng/ml). At different times mRNA were extracted from purified CD8+ T cells using the Trizol method. (b) Purified CD4+ T cells were stimulated with OVA peptide as described in Material and Methods. At the indicated times, mRNA were extracted using the Trizol method. Levels of mRNA for A1, Bcl-xl and Bcl-2 were measured by RNase protection assays. Amounts of mRNA were normalized using the L32 internal control. Normalized results are expressed as fold increase from values obtained at 0 h in (a) and values obtained at 0h with mRNA from DO11.10 mice in (b). Results shown are representative of three (a) or two (b) independent experiments

Full size image

A1 protein expression, in CD8+ purified T cells, was evaluated by Western blot. A1 protein could not be detected in unstimulated cells; however, following peptide stimulation A1 was expressed with a maximum observed between 24 and 72 h (Figure 3a). The expression of A1 was not increased in the presence of exogenous IL-2 or decreased if the endogenously produced IL-2 was neutralized with blocking anti-IL-2 mAb (Figure 3a). In parallel, the relative expression of Bcl-xl and Bcl-2 proteins was evaluated by intracellular staining and flow cytometry analysis (Figure 3b). Similar to what we observed for A1, expression of Bcl-xl was transient with a maximum at 48 h and did not require IL-2. Bcl-2 protein expression was delayed compared to that of A1 and Bcl-xl, with a maximum at 96 h, and required the presence of exogenous IL-2, confirming the mRNA expression kinetic shown in Figure 2.

Figure 3
figure 3

Expression of A1, Bcl-xl and Bcl-2 proteins in CD8+ T cells following peptide stimulation in vitro. Splenocytes from F5 mice were stimulated with NP68 peptide alone or in the presence of exogenous IL-2 or the blocking anti-IL-2 mAb S4B6. (a) At different times, CD8+ T cells were purified and A1 expression was determined by Western blot. Amounts of loaded proteins were controlled with anti-β-actin mAb. Results are representative of two independent experiments. (b) At different times following activation, splenocytes were stained for CD8, Bcl-xl and Bcl-2. Results show the expression of Bcl-xl and Bcl-2 by gated CD8+ T cells. Results are expressed as ratio of MFI as indicated in Material and Methods, and are expressed as the mean ± S.E.M. of three independent experiments

Full size image

Immunization with peptide induces A1 expression in CD8+ T cells

In order to test whether A1 expression is also regulated by peptide in vivo, F5 TCR-transgenic mice were injected once i.p. with NP68 peptide (50 nM) in PBS. Then we followed A1 mRNA and protein expression in CD8+ T cells from day 0 to 7 and compared it to Bcl-xl and Bcl-2 expression. Such immunization protocol leads, after a profound decrease, to a rapid expansion of CD8+ T cells between 24 and 72 h in spleen and inguinal lymph nodes. The absolute number of CD8+ T cells returned approximately to the basal level at day 7 (Figure 4a). Kinetic of A1 expression was analyzed at the mRNA and protein level and compared to that of Bcl-xl and Bcl-2. Total RNA from spleen and inguinal lymph nodes was extracted and A1 mRNA expression was analyzed as before by multiprobe RNase protection assay (Figure 4b). As previously observed in vitro, A1 mRNA expression was rapidly but transiently induced following peptide activation in vivo. Maximal expression of A1 mRNA (17-fold) was observed 2 h after immunization and returned to basal level within 48 h (Figure 4b). Protein expression was followed by Western blot. Maximal A1 protein was observed 18–24 h after peptide injection in both spleen (Figure 4c) and lymph nodes (not shown). Kinetic of Bcl-xl mRNA expression was similar to that of A1 with a maximum observed 2 h after immunization, whereas Bcl-2 mRNA were not affected during the first 72 h following peptide injection (Figure 4b). A reproducible increase of Bcl-xl protein expression was observed at 24 h, whereas Bcl-2 expression was poorly affected by peptide stimulation (Figure 4d). All together, our results demonstrate that in vitro and in vivo TCR engagement by antigenic peptide rapidly upregulates A1 expression at both mRNA and protein level.

Figure 4
figure 4

Regulation of A1 in CD8+ T cells following immunization with peptide. F5 mice were immunized with NP68 peptide. At different times after immunization, cells from spleen and inguinal lymph nodes were harvested, counted and stained for CD8. (a) The number of CD8+ T cells was calculated from the total number of cells from spleen and lymph nodes and the percentage of these cells, results are expressed as mean ±S.E.M. obtained from six mice. (b) At the indicated times, mRNA were extracted from purified CD8+ T cells and levels of mRNA for A1, Bcl-xl and Bcl-2 were measured by RNase protection assays. Amounts of mRNA were normalized using the L32 internal control and expressed as relative intensities. Results are expressed as mean ±S.E.M. of three independent experiments. (c) At the indicated times, CD8+ T cells were purified and A1 expression was determined by Western blot. Amounts of loaded proteins have been controlled with anti-β-actin mAb. Western blot from one experiment among two is showed. (d) Histograms for Bcl-xl and Bcl-2 expression, gated on CD8+ T cells, are shown. Results are expressed as ratio of MFI as indicated in Material and Methods and are representative of two independent experiments

Full size image

Regulation of A1 mRNA subtype in antigen-activated T cells

In mice there are four A1 genes, encoding four isoforms named A1-a, A1-b, A1-c and A1-d, of which A1-c is likely a pseudogene.21 We assess the relative proportion of A1-a, A1-b and A1-d in resting or activated CD8+ T cells by RT-PCR followed by restriction digestion as previously described (Figure 5). A1-b is the major A1 subtype expressed in resting CD8+ T cells, whereas A1-a and A1-d isoforms were very weakly expressed (Figure 5b and d). Similar pattern of expression is observed for A1-a, -b and -d, in resting CD4+ T cells as well as in B lymphocytes and thymocytes (data not shown). Following peptide stimulation, either in vitro (Figure 5b) or in vivo (Figure 5d), the three isoforms were upregulated, with a preferential increase of A1-a and A1-d.

Figure 5
figure 5

Analysis of A1 mRNA subtypes in CD8+-activated T cells. RNA from in vitro (a and b) or in vivo (c and d) NP68-activated CD8+ T cells were subjected to Rnase protection assays (a and c) or to RT-PCR followed by BglII+NsiI digestion (a and d). The bands corresponding to A1-b (743 bp), A1-d (602 bp) and A1-a (471 bp) are quantified as described in Materials and Methods. Expression of each subtype before activation at the indicated time is normalized using HPRT as internal control, and expressed as relative intensities. Results shown are representative of two independent experiments

Full size image

Cytokines regulate Bcl-2 but not A1 and Bcl-xl expression in naive CD8+ T cells

As previously mentioned, both signals mediated by TCR or cytokine receptors play an important role in T-cell survival. In agreement with recent reports,8,32, culturing-purified naive T cells in vitro resulted in a rapid cell death that can be prevented by cytokines such as IL-4 and IL-7. In contrast, IL-2 did not inhibit cell death, whereas IL-15 only delayed occurrence of cell death (Figure 6a). The expression of A1 or Bcl-xl protein was not detected in unstimulated naive CD8+ purified T cells (Figure 3). Those cells however expressed a very low level of Bcl-2 (Figure 6c and Figure 3b). None of the cytokines tested induced expression of A1 or Bcl-xl mRNA or proteins (Figure 6b and data not shown). In contrast, Bcl-2 mRNA expression was increased in the presence of IL-4, IL-7 or IL-15 (Figure 6b). As shown in Figure 6c, Bcl-2 protein expression became barely detectable after 48h of culture in medium alone or supplemented with IL-2, whereas the addition of IL-4, IL-7 or IL-15 increased its expression. Of note, the upregulation of Bcl-2 in the presence of IL-15 was very transient, as compared to what was observed in the presence of IL-4 or IL-7.

Figure 6
figure 6

Regulation of A1 and Bcl-2 expression in CD8+ naive T cells by cytokines. CD8+ T cells purified from spleens of F5 mice were cultured in medium alone or in the presence of cytokines: IL-2 (50 ng/ml), IL-4 (10 ng/ml), IL-7 (12.5 ng/ml) or IL-15 (20 ng/ml). (a) Percentages of dead cells were measured by annexin V staining and flow cytometry at 0, 24, 48 and 72 h. (b) Total mRNA was extracted from purified CD8+ T cells after 12 h in culture. The levels of mRNA coding for A1 and Bcl-2 were measured by RNase protection assays. Amounts of mRNA were normalized using the L32 internal control and expressed as relative intensities. (c) At the indicated times, splenocytes were stained for CD8 and Bcl-2. Control IgG3 were used as a control for Bcl-2 staining and results are expressed as ratio of MFI. Results show the expression of Bcl-2 by gated CD8+ T cells. Results shown in (a) are representative of three independent experiments, whereas results shown in (b and c) are expressed as mean ±S.E.M. of three independent experiments

Full size image

TCR but not cytokine receptors engagement regulates A1 expression in activated and posteffector CD8+ T cells

Survival of activated T cells is also dependent on the presence of growth factors.33,34 When F5 CD8+ T cells were cultured in the presence of NP68 peptide plus IL-2 for 6 days, then starved in fresh culture medium, 90% of activated T cells died within 72 h (Figure 7a). Cell death was associated with a rapid decrease of Bcl-2 mRNA and protein expression (Figure 7b). Addition of cytokine such as IL-2, IL-4, IL-7 or IL-15 prevented apoptosis (Figure 7a), an effect which is associated with the maintenance of Bcl-2 mRNA and protein expression (Figure 7b). After 6 days of activation, A1 mRNA was expressed at low levels in CD8+-activated T cells (Figure 7b and c). None of the cytokines tested (IL-2, IL-4, IL-7 and IL-15) induced expression of A1 mRNA (Figure 7b) or protein (Figure 7c and data not shown). However, restimulation of activated CD8+ T cells in the presence of NP68 peptide strongly upregulated A1 mRNA expression and prevented downregulation of A1 protein (Figure 7c). Of note, peptide stimulation downregulated Bcl-2 mRNA in activated CD8+ T cells (Figure 7c). Interestingly, similar results were obtained when posteffector CD8+ T cells were used. Those cells are CD8+ T cells purified from F5 TCR-transgenic mice 7 days after peptide immunization at a stage where they become resting, they have lost their ex vivo cytolytic activity but they have already acquired some memory cell characteristics, such as an increased capacity to produce γIFN in response to TCR engagement.35 As shown in Figure 8, cytokines (IL-4, IL-7 or IL-15) that allowed their survival (data not shown) increased Bcl-2 but not A1 mRNA expression, whereas TCR engagement highly upregulated A1 but not Bcl-2 mRNA. Of note, similar results were obtained using memory CD8+ T cells purified from mice 1 month after peptide immunization (data not shown).

Figure 7
figure 7

Regulation of A1 and Bcl-2 in CD8+-activated T cells. Splenocytes from F5 mice were stimulated with NP68 peptide (10 nM) and exogenous IL-2 (50 ng/ml) for 6 days as described in Materials and methods. Dead cells were removed and viable cells were cultured in medium alone or in the presence of IL-2 (50 ng/ml), IL-4 (10 ng/ml), IL-7 (12.5 ng/ml) or IL-15 (20 ng/ml). (a) Percentages of dead cells were measured by annexin V staining and flow cytometry at 24, 48 and 72 h, and are expressed as mean±S.E.M. of duplicate from four individual experiments. (b) Total mRNA was extracted after 8 h of culture. The levels of mRNA coding for A1 and Bcl-2 were measured by RNase protection assays. Amounts of mRNA were normalized using the L32 internal control and expressed as mean±S.E.M. of three individual experiments. Normalized values are expressed as fold increase from values obtained after 8 h in medium alone (left). After 24 h of culture, cells were stained for CD8 and Bcl-2. Results show the expression of Bcl-2 by gated CD8+ T cells. Results are expressed as ratio of MFI±S.E.M. of four independent experiments (right). (c) Splenocytes from F5 mice were stimulated with NP68 peptide (10 nM) or exogenous IL-2 (50 ng/ml) for 6 days, and viable cells were cultured in medium alone or in the presence of IL-2 (50 ng/ml) or NP68 peptide (10 nM). The levels of mRNA coding for A1 and Bcl-2 were measured by RNase protection assays as in a (left). Lysates were made from CD8+ T cells after 24 h in culture and expression of A1 was determined by Western blot. Amounts of loaded proteins were controlled with anti-β-actin mAb (right). Results are representative of two independent experiments

Full size image
Figure 8
figure 8

Regulation of A1 and Bcl-2 in CD8+ posteffector T cells. F5-mice were injected twice at 24 h interval with NP68 peptide (50 nM). At 7 days after the first injection, mice were killed and spleen CD8 T cells were purified and cultured for 2 h in the presence of medium, IL-4 (10 ng/ml), IL-7 (12.5 ng/ml), IL-15 (20 ng/ml) (a and b) or NP68 peptide (10 nM) (a and c). Then, mRNA were extracted and levels of mRNA coding for A1 and Bcl-2 were analyzed by RNase protection assays. Results are expressed as mean ±SEM of two (for cytokine stimulation) and three (peptide stimulation) independent experiments

Full size image

Discussion

Survival of peripheral T lymphocytes is a tightly regulated process involving signals through TCR and/or cytokine receptors as well as competition with other lymphocytes for environmental factors. Regulation of T-cell survival contributes to preserve T-cell homeostasis in normal individuals. Activation by Ag, which results in a transient expansion of Ag-specific clones, is followed by a contraction phase, so that the total number of peripheral cells remains in homeostatic equilibrium. Bcl-2 family members through their pro- or antiapoptotic properties play an essential role in the control of T-cell survival.

We have studied A1 expression following mitogenic or antigenic stimulation of T cells. A1 was previously demonstrated to be developmentally regulated in T cells, with a high expression restricted to double positive thymocytes.28 We report here that A1 protein is rapidly and transiently upregulated following TCR engagement in mature peripheral T cells both in vitro and in vivo. Kinetic of A1 expression is very similar to that of Bcl-xl, with a maximal expression of A1 and Bcl-xl proteins observed 24–48 h after stimulation. However, in contrast to what was found for Bcl-xl, we did not observe any effect of CD28 costimulation on A1 expression. Bcl-2 protein was also increased following peptide stimulation but with a delayed time course. Moreover, whereas TCR stimulation alone is sufficient for expression of A1 or Bcl-xl, addition of exogenous IL-2 is required for the late Bcl-2 expression. Previous studies identified A1 and its human homologue bfl-1 as two genes regulated by Rel/NF-κB transcription factor. More recently, it has been demonstrated that calcium mobilization induced A1 expression in mast cells, suggesting that other transcription factors may control A1 transcription. The observation that CsA strongly inhibits A1 expression in T cells would suggest that transcription factors from the NF-AT family36 may also be involved in the regulation of A1 transcription in lymphocytes. The present data obtained with murine T cells are likely to apply to human T cells. Indeed, parallel experiments performed with enriched T-cell suspensions of human PBL stimulated with immobilized mAb to CD3 showed a comparable increase of Bfl-1, the human homologue of A1 (data not shown). As far as we know this is the first demonstration, both in vitro and in vivo, that A1 expression is regulated during Ag-mediated activation of peripheral mature T cells. Moreover, our results confirm previous data demonstrating that Bcl-xl and Bcl-2 are independently regulated following activation of naive T cells.37

Cytokines drive both the survival and the homeostatic expansion of lymphocytes, two mechanisms that contribute to the maintenance of T-cell populations. The increased survival induced by cytokines is associated with increased Bcl-2 expression level. Indeed, IL-7 was shown to increase survival of naive T cells in vitro or in vivo,8,38 an effect associated with the capacity of IL-7 to promote upregulation of Bcl-2 protein.8 Similarly, activated T cells that are relatively short lived can be rescued from cell death in vitro, by the addition of cytokines such as IL-2, IL-4, IL-7 or IL-15 but also IFN-α.33,34,39 We confirmed that most cytokines that signal through the IL-2 receptor common γ-chain promote the survival of naive and activated T cells. Delayed cell death is associated with the capacity of those cytokines to induce Bcl-2 expression in naive CD8+ T cells or to prevent the downregulation of Bcl-2 and Bcl-xl (data not shown) in activated T cells. However, only TCR engagement is able to induce A1 mRNA and to maintain A1 protein expression.

Altogether, our results clearly demonstrate that signals that upregulate A1 expression in T cells differ from those acting on Bcl-xl and Bcl-2. Indeed, in both CD4+ and CD8+ T cells, the expression of A1 is driven only by engagement of TCR with specific Ag, but not by engagement of cytokine receptors.

The physiological relevance of A1 upregulation in T cells remains to be clearly defined. Our results suggest that A1 may be a key element of TCR-mediated survival in lymphocytes. Interestingly, the recent study by Gonzalez et al.40 indicates that transgenic overexpression of A1-a in T cells via the lck distal promoter reduces apoptosis of both resting and activated T cells in vitro. Moreover, unlike Bcl-2, A1-a overexpression does not inhibit S-phase entry of activated T cells, and therefore results in a more efficient accumulation of T cells after a 5-day culture period in vitro. However, whether upregulation of A1 following antigenic stimulation is essential for T cells to survive as an activated T cell and to differentiate in effector and memory cells in vivo remains to be demonstrated. This might be the case for B cells. Indeed, for peripheral B cells there is a good correlation between A1 expression following BCR engagement and survival.41 Of interest also is the demonstration that memory T cells express a higher level of A1,17 as well as Bcl-xl and Bcl-215,16,17 than naive T cells. And at least for CD44high CD8+ memory T cells, higher expression of those antiapoptotic genes has been correlated with their increased resistance to apoptosis.17 However, it is not clear in those memory T cells whether high expression of A1 or Bcl-xl reflects crossreaction of TCR with environmental Ag or cytokine-induced upregulation.

The murine genome contains four genes coding for A1 homologues, A1-a, -b, -c and -d.21 The coding regions among A1 genes are highly conserved (>96%) at the nucleotide and amino-acid sequence levels. With the exception of A1-c, which may be a pseudogene or a gene encoding an aberrant protein because of a single base-pair insertion, they all seem to code for functional proteins.21 The three functional isoforms of A1 cannot be discriminated by RNase protection assay or Western blot analysis; however, it can be done by restriction digest on RT-PCR products.21 Contrary to neutrophils in which A1-a, -b and -d subtype mRNA are expressed at a comparable level,21 we observed that the three A1 subtypes are differently expressed in lymphocytes. The pattern of A1 subtype mRNA in lymphocytes is similar to the one previously described in peritoneal exudate macrophages,42 with a strong expression of A1-b, intermediate expression of A1-d and very low expression of A1-a in unstimulated cells. Following antigenic stimulation, the differential increase of A1-a, -b and -d isoforms in both lymphocytes (Figure 5) and peritoneal exudate macrophages42 following stimulation suggest some differences in the regulatory sequences of these three genes. So far, all the functional studies demonstrating the antiapoptotic activity of mouse A1 have been carried out with A1-a.26,29 Deletion of A1-a gene has been performed and demonstrate an essential role of A1-a for neutrophil43 and mast cell survival,30 but no enhancement of T-cell apoptosis was observed in A1-a−/− mice.43 Such observation might be explained by the very low expression of A1-a, compared to that of other subtypes in T cells. Figure 5 Nevertheless, certainly because of the high homology of the three A1 proteins and their expected redundant function, mice overexpressing A1-a in their T cells showed an increased cellularity of lymphoid organs.40 Therefore, combined inhibition of the three A1 proteins by either gene targeting or RNA interference should give us some information on the real contribution of A1 to T-cell survival.

In conclusion, we demonstrate here that induction of A1 expression is restricted to TCR engagement, and therefore its expression can be clearly dissociated from that of Bcl-xl and Bcl-2 in peripheral T cells. The possibility that A1 plays a role in TCR-mediated survival and in the control of T-cell homeostasis is currently under investigation.

Materials and Methods

Mice and immunizations

In this study, we used wild-type and IL-2-deficient mice bearing a transgenic DO11. 10 TCR, specific for OVA323–329 + I-Ad on a Balb/c background; F5 mice, transgenic for a TCR recognizing the (366–374) peptide (nucleoprotein, NP68) derived from the influenza virus NP in the context of the MHC class I molecule H2 Db on a C57BL/10 background and C57BL/10 mice. All these mice were bred and maintained in mouse facilities under conventional conditions and used at an age of 5–8 weeks. For immunization, F5 TCR-transgenic mice were injected in the peritoneal cavity with 50 nM of the ANT/60/68 influenza virus nucleoprotein peptide: Ala-Ser-Asn-Glu-Asn-Met-Asp-Ala-Met (NP68-(366–374)) (Neo-systems laboratoire, Strasbourg, France). Control mice were either not treated or injected with PBS alone.

Cytokines and antibodies

Human rIL-2 was obtained from Chiron (France, Suresnes). Murine rIL-4, and human rIL-7 and rIL-15 were purchased from TEBU (Le Perray en Yvelines, France). Peridinin chlorophyll-a protein (PerCP)-conjugated anti-mouse CD8 (53-6.7), PE-anti-murine Bcl-2 (3F11), IgG1 control isotype, anti-mouse IL-2 mAb (S4B6) and anti-mouse CD28 (CD28.2) were from BD Pharmingen (Le Pont de Claix, France). FITC-conjugated anti-human Bcl-xl (7B2.5) and IgG3 control isotype were obtained from Cliniscience (Montrouge, France). Anti-mouse β-actin mAb (AC-15), PMA, ionomycine and EDTA were purchased from Sigma (St. Quentin Fallavier, France). MG132 was purchased from Calbiochem (La Jolla, CA, USA), and CsA was kindly provided by Novartis corporation. Anti-mouse CD3 (2C11) was prepared in the laboratory.

Preparation of anti-peptide antiserum

A peptide corresponding to residues 1–18 of the human bfl-1 sequence was synthesized (Neo-systems laboratoire). Rabbits were injected s.c. with peptide (100 μg)+CFA and boosted every 15 days for 12 weeks with peptide+CFA. Bleeds were screened initially by ELISA against peptide and by Western blot analysis of cell lysates from A1 overexpressing cells. Serum IgG fraction was then purified on a protein G sepharose (Amersham, Saclay, France) column according to the manufacturer's instructions.

Purification of T cells, CD8+ and CD4+ lymphocytes

Splenic T cells were purified by magnetic beads using negative selection strategy. Briefly, cells were incubated for 30 min at 4°C with a mixture of culture supernatants containing the following rat mAbs anti-GR-1 (RB6.8), anti-Mac-1 (M1/70.15) and anti-I-Ab (M5/114.15.2). Cells were then washed three times with a medium and incubated for 30 min at 4°C with goat anti-rat IgG (H+L)-coupled magnetic beads (BioMag, PerSeptive Diagnostics, GB) at a ratio of 20 beads per cell. Positive cells were removed by application to a magnet and centrifugation on a layer of PANCOLL (Dutcher). Splenic CD8+ T cells were purified following the same protocol, with the rat mAbs anti-CD4 (GK1.5), anti-GR-1 (RB6.8), anti-Mac-1 (M1/70.15) and anti-I-Ab (M5/114.15.2). Cell population purified by this method contained 92–98% of T cells or CD8+ T lymphocytes, as assessed by flow cytometry. CD4+ cells were purified by negative selection with immuno-columns for CD4 T cells according to the manufacturers' instructions (Cederlane/Canada). Purity of CD4 cell preparations was typically greater than 93% as determined by flow cytometry.

Cell culture

Purified T cells were cultured at 1 × 106/ml in RPMI medium (Sigma) supplemented with 10% FSC, 5 × 10−5 M β2-mercaptoethanol, 2 mM L-glutamine and antibiotics (100 U/ml penicillin and 100 μg/ml streptomycin). Cells were activated with anti-CD3 mAb coated to microtiter plates in the absence or presence of anti-CD28 mAb (2 μg/ml), or activated in the presence of PMA (2 ng/ml), ionomycine (1 μM) or both. For activation assays of CD8+ T cells, splenocytes from F5 mice (1 × 106/ml) were incubated in the presence of NP68 peptide (10 nM) plus rIL-2 (100 ng/ml). For growth factor withdrawal experiments, splenocytes from F5 mice were activated with NP68 peptide (10 nM) plus rIL-2 (100 ng/ml) in the presence of irradiated (30 Gy) C57BL/10 spleen cells (2 × 106/ml) for 6 days. Then, dead cells were removed by centrifugation on a layer of PANCOLL (Dutcher, Brumath, France), and viable cells washed twice and resuspended in medium alone or supplemented with rIL-2 (50 ng/ml), rIL-4 (10 ng/ml), rIL-7 (12.5 ng/ml) or rIL-15 (20 ng/ml). For experiments with CD4+ cells, cells were cultured at 0.5 × 106/ml in the presence of irradiated APC (0.5 × 105/ml) enriched for dendritic cells44 in ex vivo medium (Biowhitaker, MA, USA), supplemented with 4.8 mM L-glutamine, 1 mM sodium pyruvate, 1 × non essential amino acids, 5 × 10−5 M β2-mercaptoethanol, 400 U/ml penicillin, 0.24 mg/ml streptomycin sulfate and 5% heats inactivated FCS.

RNA preparation and Rnase protection assays

Cells were lysed in TRIZOL Reagent (Life Technologies, Cergy Pontoise, France). Total RNA was prepared following the manufacturer's instructions. Between 2 and 4 μg of total RNA was used in the RiboQuant Multi-Probe RNase Protection Assay System (Pharmingen). The assays were carried out following the manufacturer's instructions. The bands of 32P-labeled protected probes were visualized and quantified using a PhosphoImager and ImagQuant software (both from Molecular Dynamics, Sunnyvale, CA, USA). The length of their respective fragments was used to identify the transcripts. The L32-band intensities were used as internal standards.

A1mRNA subtype diagnostic

To distinguish A1-a, -b and -d subtypes, cDNA was generated from RNA sample (1 μg) using Superscript preamplification (Life Technologies) with an oligo-dT primer. PCR was performed on 10% of the cDNA product using either A1 primers previously described by Hatakeyama et al.21 (forward primer=AATTCCAACAGCCTCCAGATATG; reverse primer=GAAACAAAATATCTGCAACTCTGG) or HPRT primers (forward primer=GTAATGATCAGTCAACGGGGGAC; reverse primer=CCAGCAAGCTTGCAACCTTAACCA). Then PCR product, obtained with A1-specific primers, was digested with a combination of NSiI and BGlII (Ozyme, St. Quentin Fallavier, France) using the manufacturer's buffer. The relative proportion of the specific bands representing A1-a, -b and -d were quantified in an 8% acrylamide gel by using ImagQuant software. Results are expressed as relative intensities using HPRT signal as internal control.

Flow cytometry

Bcl-2 and Bcl-xl protein expression was investigated by three-color immunofluorescence after permeabilization of the cell membrane with Cytofix/Cytoperm™ kits (Pharmingen). Briefly, cells (1 × 106) were first labeled with anti-CD8 mAb, washed once in phosphate-buffered saline (PBS) containing BSA (10 g/l) and azide 0.2%, and resuspended in Cytofix/Cytoperm™ solution. After 20 min of incubation at room temperature and one wash with 2 ml of Perm/wash buffer, cells were incubated with anti-Bcl-xl, anti-Bcl-2 or their control isotype diluted in the Perm/wash buffer, and incubated for 20 min. After washes, cells were resuspended in PBS and analyzed by FACS with the CellQuest software. Results are expressed as MFI ratio according to the following formula: MFI ratio =MFI (Bcl-2 or Bcl-xl)/MFI (control isotype).

Western blot analysis

Whole-cell lysates were prepared by lysing 1 × 106 cells in 50 μl of Laemmli buffer. Proteins were separated by SDS-PAGE, transferred on to nitrocellulose membrane (Schleicher & Schuell, Ecquevilly, France) and analyzed by Western blot. Blots were probed with either anti-A1/Bfl-1 Ab (1/100) or anti-β-actin mAb and bound antibodies were detected with HRP-conjugated goat anti-rabbit Ig or HRP-conjugated sheep anti-mouse Ig (Amersham). An enhanced chemiluminescence system (Amersham) was used for detection. The bands corresponding to A1 or actin proteins were quantified after scanning using the Image Master software (Pharmingen).

Measurement of apoptosis

At the indicated time cell death was evaluated by annexin V staining. Cells were resuspended in binding buffer and incubated with FITC-conjugated annexin V (Bender MedSystems, Austria) for 5 min and analyzed by FACS with the CellQuest software.