Fas/Fas Ligand Interactions Promote Activation-Induced Cell Death of NK T Lymphocytes

Animals and reagents

Wild-type, Fas-deficient C57BL/6-lpr/lpr and Fas ligand (FasL)-deficient C57BL/6-gld/gld mice were bred in our own facilities and used at the age of 6–8 wk, before the onset of lymphadenopathy in the mutant strain (17). In some experiments 7-mo-old mice were used. RPMI 1640 (Life Technologies, Grand Island, NY) supplemented with 10% heat-inactivated FCS (TechGen, Les Ulis, France), 100 IU/ml penicillin, 100 μg/ml streptomycin, 10 mM HEPES buffer (all from Life Technologies), and 5 × 10⁻⁵ M 2-ME was used as the culture medium. Murine IL-2, IL-4, IL-12, IL-18, and IFN-γ were purchased from R&D Systems (Abingdon, U.K.). Human rIL-7 (sp. act., 8.8 × 10⁶ U/mg) was provided by Sanofi (Labege, France). Anti-IL-4 mAbs (11B11 and BVD6-24G2.3), anti-IFN-γ mAbs (AN18 and R46A2), and anti-CD3 mAb (145-2C11) were purified in our laboratory. The BVD6-24G2.3 clone was obtained from DNAX (Palo Alto, CA). mAbs against CD8 (53.67), CD24 (J11d), B220 (RA3-6B2), and Mac1 (M1/70) used for cell depletion were purified in our laboratory. CD4-PE (YTS 191.1), PE- or FITC-conjugated CD8 (YTS 169.4), CD3-FITC (500-A2), TCRαβ-FITC (H57-597), and streptavidin-PE (SAV-PE) were purchased from Caltag (Le Perray en Yvelines, France). Biotinylated anti-NK1.1 (PK136) or anti-CD69 (H1.2F3), anti-CD122-FITC (TM-β1), anti-TCRαβ-APC (H57-597), annexin V-FITC or -PE, anti-Fas-PE (Jo2), SAV-Cy-Chrome, and blocking NA/LE anti-FasL (clone Kay-10) were obtained from PharMingen (San Diego, CA), and anti-rat Ig-coated magnetic beads were purchased from Dynal (Compiegne, France). The irreversible, large spectrum caspase inhibitor benzyl-oxy-carbonyl-Val-Ala-Asp (zVAD)-fmk, was purchased from Bachem (Voisins-le-Bretonneux, France).

In vivo treatment, purification of NKT lymphocytes, stimulation, and apoptosis assay

Wild-type and C57BL/6-lpr/lpr mice received a single i.v. injection of 2 μg of α-GalCer (Kirin Brewery Co., Gunma, Japan) (18) dissolved in PBS containing 0.025% polysolvate 20 or vehicle alone and were sacrificed 2 or 18 h later. Freshly isolated splenocytes were enriched for CD4⁺ and CD4⁻CD8⁻ T cells by immunomagnetic depletion of CD8⁺, Mac1⁺ and B220⁺ cells. Enriched CD8⁻CD24⁻ NKT thymocytes were obtained as previously described (14). At least 90% of TCRαβ⁺ splenocytes or thymocytes were obtained after depletion. For in vitro experiments, the enriched lymphocyte population was further labeled with anti-TCRαβ and anti-NK1.1 mAbs, and TCRαβ⁺NK1.1⁺ NKT lymphocytes were sorted on a FACS Vantage cell sorter (Becton Dickinson, Mountain View, CA). Purity was >99% upon reanalysis.

Sorted lymphocytes were then stimulated at a concentration of 5 × 10⁵/ml with IL-12 (10 ng/ml) plus IL-18 (100 ng/ml) in the presence or the absence of IL-7 (40 ng/ml). In some experiments, zVAD-fmk or blocking anti-FasL mAb was added at a concentration of 50 μM or 10 μg/ml, respectively. After 1, 2, and 3 days of incubation, cells were washed in PBS and stained with annexin V-FITC and propidium iodide (PI) according to the manufacturer’s instructions.

In another series of experiments, NKT lymphocytes were incubated with 1 μM 5-(and 6-)-carboxyfluorescein diacetate, succinimidyl ester (CFSE; Molecular Probes, Leiden, The Netherlands) at 37°C for 5 min. Labeled cells were washed and then stimulated with IL-12 (10 ng/ml) plus IL-18 (100 ng/ml). After different periods of incubation, they were washed in PBS, stained with annexin V and PI, and analyzed for apoptotic cells. Supernatants were harvested in all experiments and stored at −80°C until IL-4 and IFN-γ assays as previously described (7, 10).

Flow cytometric analysis

Cells were stained in PBS containing 2% FCS and 0.01 M sodium azide and were incubated for 30 min with appropriate dilutions of various mAbs coupled to biotin, PE, APC, or fluorescein. For biotinylated mAbs, SAV-PE or SAV-Cy-Chrome was used as a second-step reagent. At least 10⁴ live lymphoid cells were acquired in each run and analyzed on a FACScalibur (Becton Dickinson) cytometer using CellQuest software.

Detection of FasL mRNA by RT-PCR

Crude RNA was extracted from sorted TCRαβ⁺NK1.1⁺ NKT lymphocytes after 18 h of stimulation with IL-18 and IL-12 using TRIzol reagent (Life Technologies, Cergy-Pontoise, France), according to the manufacturer’s instructions. The semiquantitative RT-PCR technique used was based on the comparison between FasL mRNA levels and those of the transcripts encoding the ubiquitous housekeeping gene β₂-microglobulin as described previously (7). The following primers (synthesized by Bioprobe, Montreuil, France) were used: FasL 5′, CTA CCA CCF CCA TCA CAA CC; FasL 3′, CAA CCT CTT CTC CTC CAT TA; β₂-microglobulin 5′, TGA CCG GCT TGT ATG CTA TC; and β₂-microglobulin 3′, CAG TGT GAG CCA GGA TAT AG.

Statistics

Data were expressed as the mean ± SD, and differences between means were evaluated using Student’s t test.

Activation of NKT lymphocytes stimulated with IL-12 plus IL-18 in vitro

We have recently demonstrated that in the absence of TCR engagement, NKT cells produce IFN-γ and become cytotoxic upon stimulation with IL-12 plus IL-18, while either factor alone has no significant effect (10). In the present study we addressed the question of whether the acquisition of these functional capacities was accompanied by other features of cellular activation. As shown in Fig. 1⇓A, we found that a 24-h incubation of FACS-sorted, NKT cells with IL-12 plus IL-18 resulted in a significant up-regulation of the activation marker CD69. This effect coincided with another manifestation of cell activation, namely an increase in the proportion of blast cells, as judged by light scatter characteristics (Fig. 1⇓B). NKT cell activation was accompanied by increased proliferation assessed by the fluorescent dye CFSE, which is a means of quantifying cell divisions by flow cytometry (19). Fig. 2⇓A shows that 36% of the NKT population had divided after 3 days of culture in IL-12 plus IL-18. Yet, the number of cells effectively recovered at this time point was surprisingly low, amounting merely to about 20% of the cells initially plated. This result could only be explained by the disappearance of NKT cells once they had been activated by IL-12 plus IL-18.

• Download figure
• Open in new tab
• Download powerpoint

FIGURE 1.

IL-18 plus IL-12 activate NKT lymphocytes. Sorted TCRαβ⁺NK1.1⁺ thymocytes from wild-type (C57BL/6) mice were cultured (2.5 × 10⁵ cells/ml) with IL-18 (100 ng/ml) plus IL-12 (10 ng/ml) or medium. A, Twenty-four hours later, CD69 expression on stimulated (solid line) and control cells (broken line) was analyzed. B, The percentage of blasts among live TCRαβ⁺NK1.1⁺ cells stimulated with IL-18 plus IL-12 (○) or medium (▵) was determined daily for 3 days, according to forward light scatter characteristics. Similar results were observed in three separate experiments.

• Download figure
• Open in new tab
• Download powerpoint

FIGURE 2.

IL-18 plus IL-12 mediate activation-induced cell death of NKT lymphocytes. A, Sorted TCRαβ⁺NK1.1⁺ cells from the thymus of wild-type mice were labeled with CFSE and stimulated (2.5 × 10⁵ cells/ml) with (solid line) or without (broken line) IL-18 plus IL-12. Three days later, cells were analyzed for CFSE fluorescence. The percentage of viable cells (PI negative) that had divided at least once in response to IL-18 plus IL-12, based on CFSE staining, is represented in the first histogram. These cells were gated and analyzed for annexin V labeling, as shown in the second histogram. B, Sorted TCRαβ⁺NK1.1⁺ lymphocytes from wild-type mice were cultured in medium alone or together with IL-7 (40 ng/ml), IL-18 (100 ng/ml) plus IL-12 (10 ng/ml), or all three cytokines together. The percentage of dead cells was determined 3 days later by PI staining. Similar results were observed in three separate experiments.

Activation-induced cell death of NKT lymphocytes in response to IL-12 plus IL-18

To test our hypothesis, we analyzed the viability of proliferating NKT lymphocytes gated from the population that had completed division. Using the apoptosis assay, based on the binding of annexin V that occurs early in programmed cell death after the externalization of phosphatidylserine, (20), we found that nearly 40% of divided NKT cells were about to die (Fig. 2⇑A).

In accordance with these findings, an important percentage of dead NKT lymphocytes was detected after 3 days of stimulation with IL-12 plus IL-18, similar to that for cells cultured in medium alone (Fig. 2⇑B). In contrast, only about 10% of cells died in the presence of IL-7 (Fig. 2⇑B). This survival effect might be mediated at least in part through up-regulation of the anti-apoptotic molecule Bcl-2 (21, 22). Yet, even though IL-7 saved NKT cells from spontaneous cell death, it did not prevent apoptosis induced by IL-18 plus IL-12 (Fig. 2⇑B). This was also true when IL-2 or IL-4 was used instead of IL-7, since neither factor could restore NKT cell survival (data not shown).

Implication of Fas/FasL interactions in NKT cell apoptosis induced by IL-12 plus IL-18

We have previously demonstrated that the cytotoxic functions acquired by NKT cells stimulated by IL-18 plus IL-12 involve the Fas pathway (10). It could therefore be argued that this mechanism was also responsible for the death of the effector cells themselves. Consistent with this view, we observed that IL-12 plus IL-18 induced FasL transcription (Fig. 3⇓A) and up-regulated the surface expression of Fas (Fig. 3⇓B), which is spontaneously displayed by NKT cells.

• Download figure
• Open in new tab
• Download powerpoint

FIGURE 3.

IL-18 plus IL-12 up-regulate both FasL and Fas expression. A, RNA was extracted from sorted TCRαβ⁺NK1.1⁺ thymocytes from wild-type mice after an 18-h incubation with medium or IL-18 (100 ng/ml) plus IL-12 (10 ng/ml). RT-PCR was then performed, and 2-fold diluted RNA from these cells was analyzed for FasL mRNA expression. β₂-Microglobulin (β₂-m) mRNA expression served as an internal control. B, In parallel, Fas expression by TCRαβ⁺NK1.1⁺ cells stimulated with IL-18 plus IL-12 (open histogram) or cultured in medium (closed histogram) was analyzed by flow cytometry. Similar results were obtained in at least two experiments.

Considering the importance of caspases in the signal transduction pathway leading to apoptosis (23), we evaluated the effect of the irreversible inhibitor zVAD-fmk, which blocks most of these proteases, on the viability of NKT cells treated with IL-12 plus IL-18. Fig. 4⇓ shows that the inhibition of the caspase cascade did indeed strikingly reduce cell death. To test whether NKT cell apoptosis was Fas dependent, we stimulated NKT cells with IL-12 plus IL-18 in the presence of blocking anti-FasL mAb. Fig. 4⇓ clearly shows that anti-FasL mAb inhibited NKT cell apoptosis. Taken together, our findings support the implication of the Fas death pathway in the apoptosis of activated NKT cells.

• Download figure
• Open in new tab
• Download powerpoint

FIGURE 4.

IL-18 plus IL-12-induced NKT cell apoptosis is Fas dependent and is blocked by zVAD-fmk. Sorted TCRαβ⁺NK1.1⁺ thymocytes from wild-type (C57BL/6) or C57BL/6-lpr/lpr mice were cultured (2.5 × 10⁵ cells/ml) with IL-18 (100 ng/ml) plus IL-12 (10 ng/ml) in the presence or the absence of zVAD-fmk (50 μM) or anti-FasL mAb (10 μg/ml). IL-7 (40 ng/ml) was added at the onset of all cultures. Three days later, cells were harvested, dead cells were determined by PI staining, and their percentage was calculated in relation to cells cultured in IL-7, which were considered 100% viable. Data represent the mean ± SD from three separate experiments. ∗, p < 0.05; ∗∗, p < 0.01; ∗∗∗, p < 0.001.

To confirm this view, we analyzed the effect of IL-12 plus IL-18 on NKT lymphocytes from Fas-deficient, C57BL/6-lpr/lpr mice. We found that Fas deficiency markedly diminished apoptosis of NKT cells exposed to IL-12 plus IL-18 (Fig. 4⇑). Similar results were obtained with FasL-deficient C57BL/6-gld/gld mice (data not shown), proving the involvement of both Fas and FasL in this activation-induced cell death. Increased survival of Fas-deficient NKT cells could not be explained by a poor responsiveness to IL-12 plus IL-18, since they were similarly activated in terms of IFN-γ production (up to 700 ng/2.5 × 10⁵ NKT cells), up-regulation of CD69 expression (∼2-fold after 18 h of stimulation), and blast transformation (data not shown). Despite the drastically reduced cell death among NKT lymphocytes lacking functional Fas receptors, a significant percentage remained apoptotic and was rescued by the caspase inhibitor zVAD. These findings suggest that a second caspase-dependent, but Fas-independent, pathway was involved in the death of NKT cells stimulated with IL-12 plus IL-18.

Fas-dependent activation-induced cell death in vivo, after treatment with the specific ligand of NKT lymphocytes, α-GalCer

The results obtained to date raised the question of whether activation-induced cell death of NKT lymphocytes occurred exclusively in response to TCR-independent stimulation or whether TCR ligation induced the same effect. To address this issue, we used the cognate Ag α-GalCer, whose function as a specific inducer of NKT cell activation has been established both in vitro and in vivo (10, 24). Within 18 h after a single injection of α-GalCer, we observed that NKT cells disappeared from both the spleen and the thymus, as shown in Fig. 5⇓A. The loss of this population occurred after its functional activation in terms of IL-4 and IFN-γ production, which was already detected 2 h postinjection, when the frequency of NKT cells in the two organs was not yet diminished (data not shown).

• Download figure
• Open in new tab
• Download powerpoint

FIGURE 5.

In vivo treatment with α-GalCer induces NKT cell depletion. Wild-type mice were injected once with α-GalCer (2 μg/mouse) and were killed 18 h later. A, The percentage of TCRαβ^intNK1.1⁺ cells among TCRαβ⁺-enriched lymphocytes, obtained as described in Materials and Methods, is represented in each quadrant. B, Annexin V staining was analyzed among gated TCRαβ^intNK1.1⁺PI⁻ splenocytes from wild-type mice injected with vehicle or α-GalCer. Data represent a typical experiment of three.

The disappearance of NKT cells after α-GalCer injection could be the consequence of NK1.1 down-modulation rather than depletion. This conclusion was not consistent with the fact that the number of TCRαβ^intCD122⁺ cells, which have been characterized as NKT lymphocytes (1, 2), diminished similarly after in vivo treatment with α-GalCer (0.8% TCRαβ^intCD122⁺ cells among TCRαβ⁺ splenocytes of mice having received α-GalCer vs 3.2% in vehicle controls). The proof that NKT cells were actually undergoing apoptosis after injection of α-GalCer was provided by their binding of annexin V, which is an early event of programmed cell death, while a much lower proportion of NKT cells was labeled in vehicle controls (Fig. 5⇑B).

Interestingly, in vivo treatment with α-GalCer not only affects peripheral NKT cells, but also depletes this population from the thymus, which might have important implications for its steady state and selection. A previous in vivo study has shown that administration of anti-CD3 mAb, which activates both T and NKT cells, results in the disappearance of the latter population from the spleen, but not from the thymus (25). The lack of effect in this organ is probably due to the inability of anti-CD3 mAb to stimulate thymic NKT cells (6), conversely to α-GalCer, which can activate this thymic subset in vivo (our unpublished observations).

We investigated the involvement of the Fas pathway in NKT cell depletion in vivo, again taking advantage of the Fas deficiency in C57BL/6-lpr/lpr mice. The absence of functional Fas did not prevent α-GalCer-induced activation in terms of CD69 up-regulation and IL-4 production in response to in vivo treatment (data not shown). Yet, as shown in Fig. 6⇓, it rendered NKT cells insensitive to activation-induced cell death, implicating Fas/FasL interactions in the disappearance of ligand-activated NKT cells.

• Download figure
• Open in new tab
• Download powerpoint

FIGURE 6.

Fas/FasL interactions are implicated in the apoptosis of NKT cells following α-GalCer treatment. Wild-type (C57BL/6) or C57BL/6-lpr/lpr mice received a single injection of α-GalCer (2 μg/mouse) and were killed 18 h later. The percentage of TCRαβ^intNK1.1⁺ cells among enriched splenocytes from both strains treated with vehicle or α-GalCer is depicted. Data represent the mean ± SD from three different experiments. ∗, p < 0.05.

We analyzed whether NKT cells accumulate with age in Fas-deficient mice. Seven-month-old C57BL/6-lpr/lpr mice presented a slight increase in the percentage as well as the number of TCRαβ⁺NK1.1⁺ splenocytes compared with controls (16.2 ± 4.3 × 10⁵ vs 9.3 ± 2.7 × 10⁵ cells for C57BL/6-lpr/lpr and wild-type mice, respectively). In vivo, this slight accumulation of NKT cells in old C57BL/6-lpr/lpr mice might be explained by the implication of other death receptors (23) in the apoptosis of these lymphocytes. It might also be argued that a marked accumulation of NKT cells cannot be observed in Fas-deficient mice because they are kept in pathogen-free conditions where exogenous activation of NKT cells is unlikely.

In conclusion, our data provide evidence for a new Fas-dependent mechanism controlling the life span of activated NKT cells in response to the cognate Ag α-GalCer as well as to TCR-independent (IL-12 plus IL-18) stimulation. A strict surveillance of these autoreactive effector cells seems requisite considering their functional capacities, which ensure a prompt riposte during the early stages of the immune response but may become harmful thereafter, causing damage to the organism itself.