Fas Ligand-Expressing Invariant NK T Cells Promote ...

The Journal of Immunology

Oral Nickel Tolerance: Fas Ligand-Expressing Invariant NK T Cells Promote Tolerance Induction by Eliciting Apoptotic Death of Antigen-Carrying, Effete B Cells1 Michael Nowak,2 Frank Kopp, Karin Roelofs-Haarhuis, Xianzhu Wu, and Ernst Gleichmann Whereas oral nickel administration to C57BL/6 mice (Nihigh mice) renders the animals tolerant to immunization with NiCl2 combined with H2O2 as adjuvant, as determined by ear-swelling assay, it fails to tolerize J␣18ⴚ/ⴚ mice, which lack invariant NKT (iNKT) cells. Our previous work also showed that Nihigh splenic B cells can adoptively transfer the nickel tolerance to untreated (Nilow) recipients, but not to J␣18ⴚ/ⴚ recipients. In this study, we report that oral nickel administration increased the nickel content of splenic Nihigh B cells and up-regulated their Fas expression while down-regulating expression of bcl-2 and Bcl-xL, thus giving rise to an Ag-carrying, apoptosis-prone B cell phenotype. Although oral nickel up-regulated Fas expression on B cells of both wild-type Nihigh and J␣18ⴚ/ⴚ Nihigh mice, only the former showed a reduced number of total B cells in spleen when compared with untreated, syngeneic mice, indicating that iNKT cells are involved in B cell homeostasis by eliciting apoptosis of effete B cells. Upon transfer of Nihigh B cells, an infectious spread of nickel tolerance ensues, provided the recipients are immunized with NiCl2/H2O2. As a consequence of immunization, Fas ligand-positive (FasLⴙ) iNKT cells appeared in the spleen and apparently elicited apoptosis of Nihigh B cells. The apoptotic Nihigh B cells were taken up by splenic dendritic cells, which thereby became tolerogenic for nickel-reactive Nilow T cells. In conclusion, FasLⴙ iNKT cells may act as ready-to-kill sentinels of innate immunity, but at the same time assist in tolerance induction by eliciting Fas/FasL-mediated apoptosis of effete, Ag-containing B cells. The Journal of Immunology, 2006, 176: 4581– 4589.

I

t is generally accepted that Ag presentation by immature and semimature dendritic cells (DCs)3 can induce peripheral T cell tolerance (1). The cascade of cellular and molecular reactions, however, that precedes the tolerogenic Ag presentation by DCs is less well understood. Different mechanisms have been reported to be involved here; however, it has hardly been unraveled to what extent and how they interact. Of particular interest is the contribution of apoptotic cell material to tolerance induction (2, 3). Physiologically, apoptotic cells are rapidly and inconspicuously removed by phagocytic cells, including DCs (4). In the spleen, CD8␣⫹ DCs efficiently engulf circulating apoptotic cells and present and cross-present their peptides on MHC-II and MHC-I molecules, respectively, leading to the induction of T regulatory (Treg) cells specific for apoptotic cell-associated Ags (3, 4). Defects in apoptosis can result in a loss of self tolerance. Thus, mutant mouse strains expressing nonfunctional Fas or Fas ligand (FasL) molecules develop a lymphoproliferative disease associated with systemic autoimmunity (5). Among others, the basic toler-

Institut fuer umweltmedizinische Forschung at Heinrich Heine University, Duesseldorf, Germany Received for publication November 23, 2005. Accepted for publication January 24, 2006. The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. 1 This study was financially supported by Deutsche Forschungsgemeinschaft, Bonn, through Sonderforschungsbereich 503, and by NiPERA. 2 Address correspondence and reprint requests to Dr. Michael Nowak, Institut fuer Umweltmedizinische Forschung at Heinrich Heine University gGmbH, Duesseldorf, Auf⬘m Hennekamp 50, D-40225 Duesseldorf, Germany. E-mail address: [email protected] 3 Abbreviations used in this paper: DC, dendritic cell; ACAID, anterior chamberacquired immune deviation; DNFB, 2,4-dinitrofluorobenzene; FasL, Fas ligand; i.d., intradermal; iNKT, invariant NKT; Treg, T regulatory; WT wild type.

Copyright © 2006 by The American Association of Immunologists, Inc.

ance defect of these mouse strains is also reflected by their resistance to the immune tolerance inducible by UV irradiation (6, 7). Apart from DCs and Treg cells, another cell type known to be involved in tolerance induction in different mouse models is the invariant NKT (iNKT) cell. NKT cells are characterized by the coexpression of NK1.1, an NK cell marker, and the TCR, and iNKT cells constitute a major subpopulation of NKT cells. In mice, they express the TCR encoded by the segments V␣14 and J␣18 (8). They are CD1d restricted and recognize glyco- and phospholipids, including self glycolipids (9). In response to TCR ligation, iNKT cells are able to rapidly secrete large amounts of cytokines, such as IL-4, IL-10, and/or IFN-␥ (reviewed in Ref. 10). Cytokine secretion by iNKT cells was shown to be involved in several tolerance models (11–13), including the induction of oral tolerance to nickel (14). It may be questioned though, whether the role of iNKT cells in tolerogenic immune responses can be explained solely on the basis of cytokine secretion or whether they also manipulate apoptotic mechanisms for tolerance induction. Yet another cell type reported to be involved in the induction of T cell unresponsiveness in a number of mouse models is the B cell (15–17). Notably, it was shown that the B cells from tolerized animals have to interact with iNKT cells to exert their tolerogenicity (18). This was first documented in the anterior chamberacquired immune deviation (ACAID) model, which is induced by intraocular injection of OVA. In ACAID, CD1d-expressing splenic B cells as well as eye-derived, OVA-transporting CD1d⫹F4/80⫹ macrophages are required for the activation of iNKT cells, which, in turn, are needed for the induction of OVAspecific CD8⫹ Treg cells (18). The iNKT cells in ACAID need to express IL-10 and recruit other cells via expression of chemokines (19). During the induction phase of Treg cells, splenic clusters of APCs, NKT cells, and T cells were observed, suggesting close cooperation of these cell types in spleen (20). 0022-1767/06/$02.00

4582

TOLERANCE INDUCTION BY FasL⫹ iNKT CELLS, B CELLS, AND DCs

In the mouse model of oral nickel tolerance, T cell-depleted spleen cells (termed APCs (14)), which must express CD1d, also proved tolerogenic. These APCs, consisting mainly of B cells, were able to transfer the tolerance from donor mice that were orally treated with NiCl2 (termed Nihigh throughout this work) to untreated, syngeneic recipients (termed Nilow), provided the latter possessed iNKT cells (14). Although purified splenic B cells from Nihigh donors could transfer nickel tolerance (17), the question remained whether they needed to interact with iNKT cells in this process. When the inoculum of Nihigh donor cells consisted of T cells, instead of APCs, iNKT cells were not required for the tolerance transfer. This indicated that the transferred APCs (possibly Nihigh B cells) and the recipient’s iNKT cells were involved in the induction phase of specific Treg cells and not their effector phase (14). Previously, it has been demonstrated that nickel tolerance, induced by transferred APCs from Nihigh donors, generated specific Treg cells in the recipient. Paradoxically, this infectious spread of tolerance from the cells of Nihigh donors to those of Nilow recipients required vigorous immunization of the latter, which was achieved through intradermal (i.d.) injection of NiCl2 combined with H2O2 as adjuvant (17). It is only poorly understood what functional links exist between tolerogenic B cells, iNKT cells, DCs, and Treg cells, and how these cell types interact with one another in the above-mentioned tolerance models (15–18). Therefore, in the present study, we asked whether iNKT cells are involved in the apoptotic death of Nihigh B cells, leading to the uptake of apoptotic cell fragments by DCs and, thus, induction of tolerogenic DCs and eventually nickelspecific Treg cells.

measurement after elicitation, the mice were killed by asphyxiation with CO2. Measurements were performed using a micrometer (Oditest D 1000 gauge; Kroeplin) and in a blind manner.

Materials and Methods

Before transfer, isolated cell populations were washed twice in pyrogenfree PBS. Cell suspensions (150 ␮l), containing the indicated cell numbers, were injected i.v. into the tail vein of recipient mice. First recipient mice of Nihigh B cells (shown in Fig. 6) comprised groups of two mice each; all groups of recipient mice used for measurement of ear-swelling reactions were comprised of four to five mice. One day later, mice were challenged i.d., as described above. Ten days later, secondary reactions were elicited by i.d. injections at the ears, and 48 h thereafter their ear-swelling response was measured. Serial adoptive cell transfers with two consecutive sets of Nilow recipient mice were performed, as described (17), with modifications. Nilow mice were injected i.v. with 107 Nihigh or Nilow B cells and immunized with NiCl2/H2O2 18 h later. After another 24 h, 104 splenic DCs of these first recipients were prepared and injected i.v. into a second set of Nilow recipients. The latter then were immunized with either NiCl2/H2O2, NiCl2 alone, or DNFB, and 10 days later challenged for recall at the ears before their ear-swelling responses were determined.

Reagents and Abs NiCl2 ⫻ 6H2O (denoted as NiCl2) and 2,4-dinitrofluorobenzene (DNFB) were purchased from Sigma-Aldrich, and H2O2 from E. Merck. Abs were purchased from BD Biosciences, microbeads from Miltenyi Biotec, and Newport Green DCF from Invitrogen Life Technologies.

Mice and oral tolerance induction Specific pathogen-free C57BL/6 mice (wild-type (WT) mice) were obtained from Elevage Janvier. B6.MRL.Tnfsflpr (referred to as Fas-lpr) and B6.Smn.Fasgld (referred to as Fas-gld) mice were purchased from The Jackson Laboratory. J␣18⫺/⫺V␣14⫺/⫺ (referred to as J␣18⫺/⫺) mice were created at Chiba University (Chiba, Japan) and backcrossed nine times with C57BL/6 mice (21); they were a gift from Dr. S. Balk (Beth Israel Deaconess Medical Center, Harvard University, Boston, MA). All animals received standard rodent diet (V-1324-703; Ssniff). Female mice were used throughout; they were 6 –14 wk old at the onset of experiments. Nilow mice are animals (from the indicated strain) that received nonsupplemented tap water as drinking water. Unless mentioned otherwise, the term Nilow mice denotes WT mice. Nihigh mice (from the indicated strain) received 10 mM NiCl2 in their drinking water for tolerance induction (22). They were treated for at least 4 wk and up to a maximum of 8 wk; the treatment was stopped 1 day before they were immunized. Unless mentioned otherwise, the term Nihigh mice denotes orally treated WT mice.

Immunization

Cell separations Splenic B cells were isolated from erythrocyte-free single-cell suspensions by depleting Thy-1.2⫹, CD11b⫹, and CD11c⫹ cells using magnetic cell sorting (autoMACS; Miltenyi Biotec), thus enriching CD19⫹ B cells to a purity of ⬎95%. For separation of splenic T cells, CD19⫹, CD11b⫹, and CD11c⫹ cells were depleted magnetically, yielding a purity of ⬎90%, as determined by flow cytometry (FACSCalibur; BD Biosciences). DCs were enriched, as described by Steinman et al. (24), with modifications. Briefly, spleen cell homogenates were digested with collagenase D (1 mg/ml) and DNase I (1 ␮g/ml) for 1 h at 37°C, then washed and centrifuged in dense BSA. From the interphase cells, DCs (CD11c⫹ MHC-II⫹) were then sorted using the FACSCalibur to a purity of ⬎95%.

Cell tracing experiments Splenic B cells were washed in sterile, pyrogen-free PBS (Sigma-Aldrich) and incubated with 0.5 ␮M carboxyfluorescein diacetate (Molecular Probes) at 37°C for 10 min. After incubation, cells were washed twice in pyrogen-free PBS before being injected i.v. into Nilow mice.

Nickel loading in vitro Splenic B cells were cultured overnight in complete medium containing 75 ␮M NiCl2 and then washed twice in sterile, pyrogen-free PBS (Sigma-Aldrich).

Apoptosis induction in vitro B cell apoptosis was induced by ␥-irradiation with 6 Gy in a 137Cs source (gammacell 2000; Munksgaard).

Adoptive cell transfers

Flow cytometry For FACS analysis of B cells and DCs, respectively, FcRs were blocked by using purified anti-CD16/32 mAb. Apoptotic cells were analyzed by double staining with 7-aminoactinomycin D (Molecular Probes) and biotinlabeled annexin V, using streptavidin-allophycocyanin. Early apoptotic cells were characterized as annexin V⫹ 7-aminoactinomycin D⫺, using groups of at least four mice. Flow cytometric analyses were performed on a FACSCalibur and analyzed with CellQuest software (BD Biosciences).

Nickel content of B cells

For immunization to nickel, mice received i.d. injections into both flanks (50 ␮l each) of 10 mM NiCl2 in sterile, pyrogen-free saline containing 1% H2O2. Control mice received 10 mM NiCl2 in saline (23). In the case of DNFB, mice were painted with 0.5% DNFB (w/v, dissolved in acetone: olive oil) onto shaved flanks.

The nickel content of Nihigh and Nilow B cells was assessed by using the dye Newport Green DCF (Molecular Probes), which stains divalent metal ions, such as nickel, and can be detected by flow cytometry (25).

Challenge for recall and mouse ear-swelling test

RNA isolation and RT-PCR

Ten days after immunization, mice were challenged for recall by injecting 50 ␮l of 10 mM NiCl2 in sterile, pyrogen-free saline into the pinnae of both ears or by applying 50 ␮l of 0.2% DNFB (w/v) onto both ears. Forty-eight hours later, delayed-type hypersensitivity was determined by measuring the increment in ear thickness compared with prechallenged value. For determination of prechallenge values, mice were anesthetized with CO2; for

RNA was isolated using TRIzol (Invitrogen Life Technologies). Reversetranscriptase reactions were conducted using 1 ␮g of RNA in a total volume of 40 ␮l using mouse mammary tumor virus-reverse transcriptase (Invitrogen Life Technologies). Real-time PCR were performed using the Quantitect SYBR Green PCR Kit (Qiagen) and conducted in a LightCycler (Roche Applied Science).

The Journal of Immunology Statistical analysis Statistical significances of results were determined by one-way ANOVA, followed by the Newman-Keuls test or Student’s t test (flow cytometry data), performed with GraphPad Prism (GraphPad) or Microsoft Excel (Microsoft).

Results In WT Nihigh mice, B cells are numerically decreased and prone to apoptosis Continuous oral treatment of C57BL/6 mice with 10 mM NiCl2 induces profound immune tolerance to nickel without signs of overt toxicity (22). This notwithstanding, we noted a significant reduction in spleen weight of WT Nihigh mice compared with WT Nilow mice. However, no reduction in spleen weight was detectable in J␣18⫺/⫺ Nihigh mice lacking iNKT cells (Fig. 1A, left panel). Flow cytometric analysis of splenocytes showed that the reduction in spleen weight observed in WT Nihigh mice was due to a significant decrease in the average number of CD19⫹ B cells, whereas no such decrease was observed in J␣18⫺/⫺ Nihigh mice

4583 (Fig. 1A, right panel). T cells and DCs were not affected in WT Nihigh mice and J␣18⫺/⫺ Nihigh mice (data not shown). As revealed by analysis of Fas expression on B cells, Nilow J␣18⫺/⫺ mice exhibited an increased fraction of Fas⫹ splenic B cells (Fig. 1B, right panel) compared with Nilow WT mice (Fig. 1B, left panel). Additionally, in both WT and J␣18⫺/⫺ mice, we observed an increased frequency of Fas⫹ B cells upon oral administration of NiCl2, and, again this frequency was higher in J␣18⫺/⫺ mice (Fig. 1B, right panel) than in WT mice (Fig. 1B, left panel). These observations prompted us to further investigate the susceptibility to apoptosis induction in splenic B cells of Nihigh mice. Indeed, compared with the results from WT Nilow mice, WT Nihigh B cells showed an increased mRNA expression of the proapoptotic surface protein Fas (Fig. 1C, left panel), concomitantly with a strong reduction in mRNA expression of the antiapoptotic proteins bcl-2 and Bcl-xL (Fig. 1C, center and right panels). When comparing the uptake of nickel ions in B cells from Nilow and Nihigh mice, indeed, Nihigh B cells contained about three times as many nickel ions as Nilow B cells (Fig. 1D). This conforms to recent results to be reported in another study, in which the nickel concentration in spleen was measured by atomic absorption spectroscopy (X. Wu, K. Roelofs-Haarhuis, J. Zhang, M. Nowak, L. Layland, M. Fang, E. Jermann, and E. Gleichmann, submitted for publication). Apparently, the proapoptotic alterations seen in Nihigh B cells were a direct consequence of the increased nickel concentrations of these cells. Fas-defective mice are resistant to oral tolerance induction toward nickel To analyze whether Fas or FasL molecules were, in fact, required for the induction of oral nickel tolerance, WT and Fas-defective Fas-lpr mice were subjected to the oral tolerance treatment or remained untreated. Mice were then immunized with either NiCl2 alone or NiCl2/H2O2, challenged with NiCl2 for elicitation, and tested for nickel hypersensitivity vs tolerance. As expected, immunization of WT Nilow mice with NiCl2 alone induced background ear swelling, whereas immunization with NiCl2/H2O2 induced nickel hypersensitivity (Fig. 2, bar 1 vs bar 2). Immunization of Fas-lpr Nilow mice with NiCl2 alone, however, induced a marked ear-swelling response, comparable to that induced by NiCl2/H2O2 (Fig. 2, bar 3 vs bar 4). This discrepancy between WT Nilow and Fas-lpr Nilow mice may be due to differences in the expression of costimulatory molecules CD80 and

FIGURE 1. Preapoptotic B cells are increased in both WT Nihigh mice and J␣18⫺/⫺ Nihigh mice, but in the latter this is not associated with a decrease in B cell numbers. A, Spleen weight (left panel) and percentages of splenic B cells (right panel) from WT Nihigh and J␣18⫺/⫺ Nihigh mice and the respective Nilow mice. B, Fas protein expression by splenic B cells of WT Nilow and WT Nihigh mice (left panel) and J␣18⫺/⫺ Nihigh mice and J␣18⫺/⫺ Nilow mice (right panel); FACS analysis of gated splenic B cells (CD19⫹) from Nilow (filled line) and Nihigh (solid line) mice stained for Fas. C, mRNA expression of Fas, bcl-2, and Bcl-xL by purified CD19⫹ splenic B cells from Nilow and Nihigh mice. D, Newport green fluorescence of B220⫹ spleen cells from either Nilow or Nihigh mice. Results shown represent one of three experiments yielding comparable results. In this and the following figures, asterisks denote statistical significances (ⴱ, p ⬍ 0.05; ⴱⴱ, p ⬍ 0.01; ⴱⴱⴱ, p ⬍ 0.001); White indicates the critical experimental groups; and bar numbers mentioned in the text refer to the bars read from top to bottom.

FIGURE 2. Fas-defective mice are resistant to oral tolerance induction toward nickel. Ear-swelling responses of Fas-lpr mice and WT controls having received either nonsupplemented drinking water or 10 mM NiCl2 in the drinking water for a period of 4 wk, as indicated. Mice were then immunized in the flanks with either NiCl2 or NiCl2/H2O2, as described in Materials and Methods. Ten days later, the mice were challenged for recall by i.d. injection of 10 mM NiCl2 into the pinnae of the ears. Ear thickness was measured 48 h thereafter. Results shown represent one of two experiments yielding comparable results.

4584


CD86 on DCs of draining lymph nodes. In WT Nilow mice, immunization with NiCl2/H2O2 up-regulates the expression of these molecules, whereas immunization with NiCl2 alone fails to do so (17). Because Fas-lpr mice possess an a priori stimulated immune system with increased expression of CD80 and CD86 (26), they promptly responded with hypersensitivity to immunization with NiCl2 alone. Nickel hypersensitivity was significantly reduced in WT Nihigh mice when compared with WT Nilow mice (Fig. 2, bar 5 vs bar 2), confirming previous results (22). Notably, however, Fas-lpr Nihigh mice failed to show a reduced response to nickel (Fig. 2, bar 7). Comparable results were obtained using FasLdefective Fas-gld mice (data not shown). Apparently, the induction of oral nickel tolerance involves Fas/FasL-mediated apoptotic cell death. Nihigh B cells need to express Fas to acquire tolerogenicity The tolerance of Nihigh mice can be adoptively transferred not only through T cells, but also by B cells (17). Because induction of nickel tolerance requires intact Fas and FasL molecules (Fig. 2, and results not shown), we asked whether the B cells of Nihigh mice need to express functional Fas to be effectively tolerogenic upon transfer to Nilow recipients. Prospective WT and Fas-lpr donor mice were orally treated with NiCl2 or remained untreated. Splenic B cells from these donors were then transferred into Nilow recipients, which were subsequently immunized with NiCl2 alone or NiCl2/H2O2, rechallenged with NiCl2, and tested for induction of nickel tolerance. As shown before (17), the transfer of B cells from WT Nihigh donors, but not Nilow donors, conferred nickel tolerance to Nilow recipients (Fig. 3, bar 4 vs bar 3). In contrast, transfer of B cells from Fas-lpr Nihigh donors failed to do so (Fig. 3, bar 6), demonstrating that Nihigh B cells require the expression of functional Fas to induce tolerance in Nilow recipients. The preapoptotic state of Nihigh B cells is enhanced by NiCl2/ H2O2 immunization Previously, our laboratory reported that upon adoptive transfer of Nihigh B cells to Nilow recipients, immunization of the latter with NiCl2/H2O2 was required for the infectious spread of nickel tolerance to ensue, whereas injection of NiCl2 alone was ineffective (17). Therefore, we asked whether NiCl2/H2O2 immunization could further enhance the susceptibility of Nihigh B cells to apoptosis and, thus, render them tolerogenic. For this purpose, we assessed mRNA expression of the proapoptotic protein bax in Nilow and Nihigh splenic B cells after immunization with either NiCl2/H2O2 or NiCl2 alone (Fig. 4). In the absence of immuniza-

FIGURE 3. Fas expression by Nihigh B cells is required for their tolerogenicity. Ear-swelling responses of B cell recipients of Fas-lpr and WT mice after immunization and challenge. Prospective Fas-lpr and WT donor mice were orally treated with NiCl2 (Nihigh) or left untreated (Nilow). Splenic B cells, were then adoptively transferred into Nilow recipients (103 cells/mouse). After immunization and challenge, their ear-swelling responses were determined. Results shown represent one of three experiments yielding comparable results.

FIGURE 4. Immunization with NiCl2/H2O2 enhances the preapoptotic state of Nihigh B cells. Expression of bax mRNA by splenic B cells of Nilow and Nihigh mice was determined at 24 h after indicated immunizations. Expression levels were calculated as relative expression levels compared with nonimmunized Nilow B cells. Results shown represent one of three experiments yielding comparable results.

tion, Nilow and Nihigh B cells expressed comparable amounts of bax mRNA. In contrast, Nihigh B cells, but not Nilow B cells, showed a 20-fold up-regulation of bax mRNA 24 h after immunization with NiCl2/H2O2, the injection of NiCl2 alone being ineffective. Anatomic site of B cell apoptosis following i.d. immunization As previously reported (17), adoptive transfer of splenic APCs and B cells, respectively, from Nihigh donors elicits the infectious spread of tolerance to nickel-reactive T cells in Nilow recipients. Paradoxically, to be set into motion, this tolerance spread required posttransfer immunization of recipients with NiCl2 and the adjuvant H2O2 (17). Because immunization with NiCl2/H2O2 up-regulated mRNA expression of the proapoptotic protein bax in Nihigh B cells (Fig. 4), we examined whether posttransfer recipient immunization with NiCl2/H2O2 would drive Nihigh donor B cells into apoptosis. Applying annexin V expression as readout parameter, indeed, there was a significant increase of Nihigh B donor cells undergoing apoptosis in the recipients immunized with NiCl2/ H2O2 (Fig. 5, B and D, right panel) compared with the fraction of Nilow donor B cells dying from apoptosis under these conditions (Fig. 5, A and C, right panel). Moreover, in general, the percentage of apoptotic donor B cells in the recipient spleen (Fig. 5, C and D) was higher than in draining lymph nodes (Fig. 5, A and B). Two conclusions can be drawn from the results shown in Figs. 4 and 5. First, only Nihigh B cells up-regulated the proapoptotic protein bax in response to immunization with NiCl2/H2O2, and second, significantly more B cells from Nihigh than from Nilow donors underwent apoptosis in recipient mice after posttransfer NiCl2/H2O2 immunization. DCs from recipients of Nihigh B cells can induce nickel tolerance To test whether adoptive transfer of Nihigh B cells, followed by the apoptosis-enhancing recipient immunization with NiCl2/H2O2, could generate DCs that were tolerogenic for nickel, we performed a serial adoptive cell transfer experiment, in which we sequentially used two sets of Nilow recipient mice. Adoptive transfer of 107 Nihigh, but not Nilow B cells, followed by recipient immunization, generated DCs tolerogenic for nickel, but not DNFB, as determined by DC transfer from the first to the second set of Nilow recipients (Fig. 6). For the following two reasons we can rule out the possibility that the tolerogenic effect induced by the second set of transferred cells (i.e., 104 DCs) was due to a contamination within DCs by residual B cells of the first transfer. First, as shown in Fig. 5, the majority of Nihigh B cells undergo apoptotic death within 24 h after the posttransfer recipient immunization. Second, a spleen from a Nilow recipient mouse contains ⬃108 cells (data not shown), and of these, 46.7% are B cells (Fig. 1A). In the unlikely event that donor Nihigh B cells from the primary donor

The Journal of Immunology

4585

FIGURE 6. Serial adoptive cell transfer: splenic DCs prepared from recipients of Nihigh B cells induce tolerance upon transfer into a second set of Nilow recipients. Splenic B cells from Nilow or Nihigh primary donors were isolated and injected (107 cells/mouse) into a first set of Nilow recipients. Twenty-four hours later, the first recipients were immunized with NiCl2/H2O2. After another 24 h, splenic DCs from first recipients were isolated and injected into a second set of Nilow recipients (104 cells/mouse). Following indicated immunizations and elicitation, the ear-swelling responses of the second recipients were determined. Results shown represent one of three experiments yielding comparable results. FIGURE 5. Immunization of recipient mice induces apoptosis of adoptively transferred Nihigh B cells. Spleen cells from Nilow (nonfilled curves, A and C) and Nihigh donor mice (shaded curves, B and D), labeled with CFSE, were adoptively transferred into Nilow recipients (107 cells/mouse). Twenty-four hours after transfer, the recipients remained either nonimmunized (left panel), or were immunized by i.d. injection of either NiCl2 alone (center panel) or NiCl2/H2O2 (right panel). Numbers indicate percentages of annexin V⫹ CFSE⫹CD19⫹ cells (⫾ SD) in spleens (A and B) and draining axillary lymph nodes (C and D) 24 h after indicated immunizations. Data shown represent one of three experiments yielding comparable results.

would remain intact and replace recipient B cells in the spleen, the 107 B cells that were initially transferred from the primary donors would still only account for 21.4% of total splenic B cells in the first set of recipient mice. Even in the improbable case that all of the 5% non-DCs contaminating the 104 cells used in the second transfer consisted of intact Nihigh B cells, they were still fewer than 103 cells, and thus would have fallen below the number of Nihigh B cells required for transfer nickel tolerance (17). Appearance of FasL-expressing NKT cells in the draining lymph nodes and spleen is required for tolerance transfer by Nihigh B cells Previously, our group demonstrated that the presence of iNKT cells in recipients and recipient immunization with NiCl2/H2O2 were required for the tolerance transfer by T cell-depleted spleen cells from Nihigh donors (14). In view of the pivotal role of Fasmediated apoptosis in the tolerance transfer by B cells (Fig. 2, bar 7), we tested whether, apart from their IL-4 and IL-10 production (14), iNKT cells might contribute to nickel tolerance by apoptosis induction of donor B cells, using their expression of FasL. Because recipient injection with NiCl2/H2O2 is necessary for the spread of tolerance from Nihigh donor cells to Nilow recipient cells (17), we asked whether this immunization up-regulates FasL on recipient iNKT cells, which would enable them to elicit apoptosis in the strongly Fas⫹ Nihigh B cells shown above (Fig. 1, B and C, left panels). For this purpose, we determined both the total number of NKT cells and the number of FasL⫹ NKT cells in the draining axillary and inguinal lymph node and spleen of Nilow mice following i.d. injection at the flanks of either NiCl2/H2O2, NiCl2 alone, or saline. Under physiological conditions, i.e., in the absence of any injection, the numbers of NKT cells in both lymph nodes and spleen

were relatively low (Fig. 7, B and D). However, 24 h after injection of either NiCl2/H2O2, NiCl2 alone, or saline, the absolute number of NKT cells in both the draining lymph node and spleen was increased. By 48 h, the numbers of NKT cells had declined to the baseline level seen in untreated mice or below (Fig. 7, B and D). Similarly, in the absence of any injection, the numbers of NKT cells expressing FasL on their surface were relatively low (Fig. 7, C and E), but at 24 h after injection of either NiCl2 H2O2, NiCl2 alone, or saline, again, the number of FasL⫹ NKT cells in the draining lymph node and, even more pronounced, the spleen was increased (Fig. 7, C and E). Again, this effect was irrespective of the type of substance injected. In both the draining lymph nodes and spleen, the appearance of FasL⫹ NKT cells peaked on day 1 after injection and returned to preinjection values by day 2 (Fig. 7, C and E). In contrast to the draining axillary and inguinal lymph nodes, no changes in NKT cellularity were observed in the distal popliteal lymph nodes (Fig. 7F). The increase in and the kinetics of the number of FasL⫹ NKT cells (Fig. 7, C and E) showed a very similar pattern to that of the total number of NKT cells in the draining lymph nodes and spleen (Fig. 7, B and D). This suggests that the increase in the number of FasL⫹ NKT cells was due to an influx of NKT cells into these lymphoid organs, rather than upregulation of FasL on resident NKT cells. iNKT cells contribute to tolerance induction by expression of FasL Being aware that iNKT cells are required for the tolerance transfer by Nihigh B cells (14), we analyzed whether iNKT cells need to express FasL to mediate the tolerance transfer. For this purpose, iNKT-deficient J␣18⫺/⫺ mice, instead of Nilow mice, were used as recipients and injected with either Nilow or Nihigh B cells, as indicated (Fig. 8A). Two groups receiving Nihigh B cells were coinjected with spleen cells from either untreated Nilow or untreated, FasL-defective Fas-gld mice, serving as a source of iNKT cells (Fig. 8A, bars 3 and 4). Confirming and extending the results previously obtained after the transfer of Nihigh APCs (i.e., T celldepleted spleen cells) to J␣18⫺/⫺ recipients (14), in this study we show that in the absence of iNKT cells, the Nihigh B cells failed to transfer nickel tolerance (Fig. 8A, bar 2), but were able to do so when the recipients were reconstituted by cotransfer of Nilow (i.e., WT) spleen cells as a source of iNKT cells (Fig. 8A, bar 3). However, when the iNKT cell-reconstituting splenocytes for cotransfer were provided by FasL-defective Fas-gld mice, the ear-swelling

4586


FIGURE 8. In J␣18⫺/⫺ recipients, substitution of FasL-expressing NKT cells is required for successful tolerance transfer by Nihigh B cells, unless the latter were rendered apoptotic before transfer. A, Nihigh or Nilow B cells alone, or together with spleen cells from either Nilow WT or Fas-gld mice, were adoptively transferred into J␣18⫺/⫺ recipient mice. After NiCl2/H2O2 immunization and elicitation with NiCl2, their ear-swelling responses were determined. B, Splenic B cells from Nilow and Nihigh mice, respectively, were isolated, and aliquot of Nihigh B cells was irradiated with 6 Gy. B cells were then adoptively transferred into J␣18⫺/⫺ recipient mice, with or without cotransfer of Nilow B cells, as indicated. After immunization and elicitation, their ear-swelling responses were determined. Results shown represent one of three experiments yielding comparable results.

response of J␣18⫺/⫺ recipients of Nihigh B cells failed to be reduced (Fig. 8A, bar 4). Hence, iNKT cells were involved in the tolerance transfer through their expression of FasL, which can elicit apoptotic death of Fas⫹ Nihigh B cells. Apoptosis induction in Nihigh B cells before transfer replaces iNKT cells for tolerance induction

FIGURE 7. Following i.d. injection, FasL-expressing NKT cells appear in draining lymph nodes and spleen. A, Staining of splenocytes for NKT cells and FasL⫹ NKT cells using mAbs against ␣␤TCR, NK1.1, and FasL. Spleen cells from an untreated Nilow mouse (left panel) and a Nilow mouse immunized with NiCl2/H2O2 (center panel) were gated for NKT cells. The histogram (right panel) shows FasL expression profiles of gated NKT cells from the untreated (shaded curve) and the immunized Nilow mouse (open curve). B–F, Data shown in B, D, and F are absolute numbers of ␣␤TCR⫹ NK1.1⫹ cells in B, draining axillary and inguinal lymph nodes; D, spleen; and F, nondraining popliteal lymph nodes at indicated times after injections. Data shown in C and E are absolute numbers of FasL-expressing ␣␤TCR⫹ NK1.1⫹ cells in C, axillary and inguinal lymph nodes, and E, spleen at indicated times after injections. Results represent one of three experiments yielding comparable results.

Next, we tested whether the apoptosis-inducing function of FasL⫹ iNKT cells could be replaced by inducing apoptosis of Nihigh B cells before transfer. Therefore, Nihigh B cells were ␥-irradiated in vitro and transferred into iNKT-deficient J␣18⫺/⫺ mice, which then were immunized, rechallenged, and tested for tolerance. Transfer of nonirradiated Nihigh B cells into iNKT-deficient J␣18⫺/⫺ mice failed to mediate the tolerance. This failure could be corrected by cotransfer of Nilow spleen cells as a source of iNKT cells (Fig. 8B, bars 2 and 3). However, even in the absence of iNKT cells, Nihigh B cells were able to transfer the tolerance when they were irradiated and, hence, rendered apoptotic before transfer (Fig. 8B, bar 4).

Discussion Interaction of Fas with its natural ligand FasL contributes to the induction and maintenance of immune tolerance (3, 6, 7). However, it was not resolved by these studies whether Fas/FasL-mediated apoptosis operated in the induction or effector phase of Treg cells and which cell types required expression of which molecules for tolerance to be established. In the present study, we show that Fas⫹ B cells loaded with Ag have superb tolerogenic potential and that the FasL⫹ cell type that elicits their apoptotic death is the iNKT cell.

The Journal of Immunology Our results indicate that nickel ions exert two different functions in the model of oral nickel tolerance: apart from generating specific neo-Ag, they exert proapoptotic effects and, hence, open up a pathway to tolerance induction. Therefore, three different explanations, which are not mutually exclusive, could account for the tolerogenicity of Nihigh B cells. First, due to the higher nickel content of Nihigh B cells, the number of nickel-induced neo-Ags present in these cells probably exceeded that in Nilow B cells. This would correlate to findings that, in general, higher Ag doses induce more profound immune tolerance than lower doses (27). Second, nickel ions per se could be tolerogenic because they induce proapoptotic events (28). Consistent with this, we observed an eminent apoptotic susceptibility of Nihigh B cells, as manifested by down-regulation of bcl-2 and Bcl-xL and up-regulation of Fas and bax. The enhanced Fas expression by Nihigh B cells was shown to be due to excess of nickel ions, and not to lack of iNKT cells that delete Fas⫹ cells. Consequently, upon transfer and recipient immunization, Nihigh B cells rapidly underwent apoptosis, which at last put their tolerogenic potential into effect. In contrast, Nilow B cells were less susceptible to apoptosis and, consequently, failed to be tolerogenic. Third, the tolerogenicity of Nihigh B cells might also be accounted for by their lower CD40 expression compared with that of Nilow B cells (17). It is known that ligation of CD40 may up-regulate mRNA expression of the antiapoptotic protein Bcl-xL (29), and such an up-regulation was not seen in Nihigh B cells. What could be the mechanism of the enhanced susceptibility of Nihigh B cells to apoptosis? Nickel ions decrease the cellular threshold for apoptosis induction by inducing DNA strand breaks (28), which can activate the DNA-damage sensor and transcription factor p53 (reviewed in Ref. 30). p53 activation may up-regulate transcription of the proapoptotic proteins bax and Fas (31), and down-regulate Bcl-xL (32). In addition to nickel ions, reactive oxygen species, such as H2O2, also can activate p53 and, hence, may further decrease the threshold for apoptosis induction in Nihigh B cells, as injection of NiCl2/H2O2, but not NiCl2 alone, led to dramatic up-regulation of bax mRNA expression in Nihigh B cells. Further enhancement of the preapoptotic state of Nihigh B cells might explain the paradox that posttransfer recipient immunization with NiCl2/H2O2 was needed for the infectious tolerance spread from 103 Nihigh donor B cells to Nilow recipient T cells (17). Clearly, Nihigh B cells induce tolerance upon adoptive transfer and immunization of Nilow recipient mice. However, we did not investigate whether oral nickel tolerance can or cannot be induced in B cell-depleted animals. Therefore, we cannot state with certainty that B cells also are required for induction of oral tolerance, that is, in the absence of cell transfer and immunization. Yet, this possibility is likely because WT Nihigh mice, which possess powerful nickel-specific Treg cells and maintain persistent tolerance (22), were found to contain a decreased number of splenic B cells. Conversely, J␣18⫺/⫺ Nihigh mice, which are resistant to oral nickel tolerance (14), possessed a normal number of splenic B cells despite increased Fas expression. Supposing that the B cell pool in WT Nihigh mice was subjected to enhanced cellular turnover, the increased number of Fas⫹, nickel-containing Nihigh B cells arising during the 4 wk of oral nickel treatment would generate an increased number of neo-Ag-containing apoptotic bodies for continuous tolerance induction. Presumably, the increased nickel load of WT Nihigh B cells just precipitated apoptotic B cell death and, thus, enhanced tolerance induction to the self Ags and nickel-induced quasi-self Ags that were carried by these cells. This view is further supported by comparing the immunological status of WT Nihigh mice with that of the apoptosis-defective Fas-lpr and FasL-gld strains. Whereas the latter two strains show increased numbers of

4587 splenic B cells associated with loss of immune tolerance toward self (33, 34), Nihigh mice exhibited just the opposite phenomena, namely a decreased number of splenic B cells and profound immune tolerance toward the quasi-self Ags induced by nickel. Apparently, by exerting proapoptotic effects on B cells, high concentrations of nickel ions put forth an excessive stress on a physiological pathway to tolerance. Although the infectious tolerance spread after adoptive cell transfer in our model was elicited by i.d. injection of NiCl2/H2O2 into flanks (17), we found that apoptosis of transferred Nihigh B cells primarily took place in spleen, an organ that has been demonstrated to be an important site for tolerance induction in a variety of other models (3, 35, 36); within 1 day, the NiCl2/H2O2 injection into flanks triggered maximal apoptotic death of Nihigh donor B cells in recipient spleen. In time-lapse fashion, the small number of Nihigh B cells transferred thus became able to exert a tolerogenic effect comparable to that of the much higher number of effete B cells in Nihigh mice themselves, which did not require NiCl2/H2O2 injection for tolerance induction after the 4 wk of oral nickel administration. Together, these findings suggest that the spleen is an important organ for transfer of tolerogenic signals also during the induction of oral nickel tolerance and, moreover, that the iNKT cells, known to be required for nickel tolerance induction (14), might use their FasL molecules to induce apoptosis of Fas⫹ Nihigh B cells. Indeed, in cotransfer experiments with Nihigh B cells (Fig. 8A), we found that the FasL-defective splenocytes of Fas-gld mice, unlike those of Ni low mice, were unable to compensate for the iNKT cell deficiency of J␣18⫺/⫺ recipients of tolerogenic Nihigh B cells. Admittedly, we cannot rule out the possibility that iNKT transferred from Fas-gld mice have other defects besides their FasL mutation. This is unlikely, however, in view of our finding that Nihigh B cells, which were driven into apoptosis by ␥-irradiation ex vivo, proved able to induce tolerance in the absence of iNKT cells (Fig. 8B), thus bypassing their function. NKT cells are known to exert effects through cytokine secretion (13) or cytotoxic mechanisms, including FasL-Fas interaction (37). With regard to tolerance induction, the implication in most studies was that iNKT cells operate solely through cytokine secretion. Only two studies suggested that they might promote tolerance by other mechanisms, namely ␣-GalCer-induced cytotoxicity (38) or elicitation of apoptotic death of activated T cells (39). That B cells interact with iNKT cells in the process of tolerance induction was first realized in the ACAID model (40), in which Ag-transporting F4/80⫹ APCs and splenic marginal zone B cells cooperate in a CD1d-dependent manner with iNKT cells (20), the latter then acting by secretion of IL-10 (19). Likewise, for successful transfer of nickel tolerance, Nihigh B cells must express CD1d on their surface (14), suggesting that they activate iNKT cells before these elicit B cell apoptosis. The results of the present study, obtained from adoptive transfer experiments with Nihigh B cells, are first to indicate that iNKT cells must induce FasL-mediated apoptosis of preapoptotic B cells to make B cell tolerogenicity effective. Therefore, the lack of iNKT cells in J␣18⫺/⫺ mice, which failed to become tolerant upon transfer of Nihigh B cells, could be bypassed if Nihigh B cells were rendered apoptotic (i.e., ␥-irradiated) before transfer, indicating that elicitation of apoptosis of Nihigh B cells is the main tolerogenic function of iNKT cells in our model. This conclusion does not necessarily contradict the previous report from our laboratory (14) that production of IL-4 and IL-10 by iNKT cells is required for the tolerance transfer by nonirradiated Nihigh APCs (i.e., T cell-depleted spleen cells). Our observations that ␥-irradiated, apoptotic Nihigh B cells are able to induce tolerance in the absence of FasL-expressing (Fig. 8) and IL-4- and

4588

TOLERANCE INDUCTION BY FasL⫹ iNKT CELLS, B CELLS, AND DCs ing, effete B cells conditioned DCs such that they conveyed tolerogenic signals to nickel-reactive T cells (Fig. 9).

Acknowledgments Ingo Uthe and Sabine Dierkes are acknowledged for technical assistance. We thank our colleagues Charlotte Esser, Irmgard Foerster, Helga Gleichmann, Marc Majora, and Laura Layland for critically reading the manuscript.

Disclosures The authors have no financial conflict of interest. FIGURE 9. Schematic view of the interactions of Ag-laden preapoptotic B cells, iNKT cells, and DCs for the induction of specific T regulatorsuppressor cells, based on the results of the present study and references (3, 14, 17, 41) (see text for details).

IL-10-producing (14), iNKT cells might suggest that these cytokines act upstream of apoptosis induction. This view is supported by the fact that iNKT cell-derived IL-4 in an autocrine fashion up-regulates the FasL expression on the producing cells (41). Hence, the tolerance-promoting activity of iNKT cells may be due to an autocrine action of IL-4 that promotes FasL-mediated cytotoxicity, which is bypassed by rendering Nihigh B cells apoptotic by ␥-irradiation. There is no information indicating a tolerogenic activity of IL-10 upstream of B cell apoptosis. Interestingly, Gao et al. (42) showed that apoptosis of murine spleen cells prompts the cells to synthesize and secrete IL-10, regardless whether apoptosis was induced by Fas-FasL interaction or ␥-irradiation. Supposing that this mechanism operates in ␥-irradiated Nihigh B cells as well, it could account for their tolerogenicity in J␣18⫺/⫺ recipients. The present study is the first to show that iNKT cells use expression of FasL for elicitation of apoptotic cell death leading to tolerance induction. This characterizes iNKT cells as cells to assist in maintaining tolerance toward the self Ags carried by apoptotic cells, in our case the nickel-induced quasi-self Ags carried by Nihigh B cells. Furthermore, our finding of an increased fraction of Fas⫹ B cells in the spleens of untreated (Nilow) J␣18⫺/⫺ mice suggests that a physiological role of iNKT cells is the elicitation of FasL-mediated apoptosis in effete B cells. Apparently, this is a continuous process in WT Nilow mice, which requires neither oral nickel nor immunization with NiCl2/H2O2. Following the uptake of apoptotic bodies, DCs present and cross-present phagocytosed Ags to T cells. This process, which has begun to be elucidated for autologous cell types other than B cells (4), is known to result in T cell tolerance, rather than autoimmunity. The uptake by DCs of apoptotic B cells that contain nickelinduced neo-Ags follows this tolerogenic pathway. Indeed, we observed that DCs incorporated apoptotic bodies of Nihigh B cells (data not shown). The functional consequence of this uptake was the induction of tolerogenic DCs in recipients of Nihigh B cells, as demonstrated by sequential adoptive cell transfers. Those DCs apparently presented antigenic material carried by the apoptotic cells, in our case the nickel ions present in Nihigh B cells, in tolerogenic fashion to T cells of the second set of Nilow recipients. This finding is consistent with data reported by Ferguson et al. (3). They transferred DCs from recipients of haptenated, apoptotic splenocytes to a second set of naive recipients and found that the DCs had transduced the tolerogenic signal so that eventually CD8⫹ Treg cells were induced. The existence of an iNKT cell-DC axis for tolerance induction proposed in the present work is supported by a recent paper of Chen et al. (43); albeit these authors did not invoke apoptosis of Ag-carrying cells as the initial tolerogenic signal. In summary, the present study demonstrated that iNKT cell-mediated apoptosis of Ag-carry-

References 1. Lutz, M., and G. Schuler. 2002. Immature, semi-mature and fully mature dendritic cells: which signals induce tolerance or immunity? Trends Immunol. 23: 445– 449. 2. Sauter, B., M. L. Albert, L. Francisco, M. Larsson, S. Somersan, and N. Bhardwaj. 2000. Consequences of cell death: exposure to necrotic tumor cells, but not primary tissue cells or apoptotic cells, induces the maturation of immunostimulatory dendritic cells. J. Exp. Med. 191: 423– 434. 3. Ferguson, T. A., J. Herndon, B. Elzey, T. S. Griffith, S. Schoenberger, and D. R. Green. 2002. Uptake of apoptotic antigen-coupled cells by lymphoid dendritic cells and cross-priming of CD8⫹ T cells produce active immune unresponsiveness. J. Immunol. 168: 5589 –5595. 4. Liu, K., T. Iyoda, M. Saternus, Y. Kimura, K. Inaba, and R. M. Steinman. 2002. Immune tolerance after delivery of dying cells to dendritic cells in situ. J. Exp. Med. 196: 1091–1097. 5. Kanauchi, H., F. Furukawa, and S. Imamura. 1991. Characterization of cutaneous infiltrates in MRL/lpr mice monitored from onset to the full development of lupus erythematosus-like skin lesions. J. Invest. Dermatol. 96: 478 – 483. 6. Hill, L. L., V. K. Shreedhar, M. L. Kripke, and L. B. Owen-Schaub. 1999. A critical role for Fas ligand in the active suppression of systemic immune responses by ultraviolet radiation. J. Exp. Med. 189: 1285–1294. 7. Schwarz, A., S. Grabbe, K. Grosse-Heitmeyer, B. Roters, H. Riemann, T. A. Luger, G. Trinchieri, and T. Schwarz. 1998. Ultraviolet light-induced immune tolerance is mediated via the Fas/Fas-ligand system. J. Immunol. 160: 4262– 4270. 8. Taniguchi, M., H. Koseki, T. Tokuhisa, K. Masuda, H. Sato, E. Kondo, T. Kawano, J. Cui, A. Perkes, S. Koyasu, and Y. Makino. 1996. Essential requirement of an invariant V␣14 T cell antigen receptor expression in the development of natural killer T cells. Proc. Natl. Acad. Sci. USA 93: 11025–11028. 9. Zhou, D., J. Mattner, C. Cantu III, N. Schrantz, N. Yin, Y. Gao, Y. Sagiv, K. Hudspeth, Y. P. Wu, T. Yamashita, et al. 2004. Lysosomal glycosphingolipid recognition by NKT cells. Science 306: 1786 –1789. 10. Hammond, K. J., and M. Kronenberg. 2003. Natural killer T cells: natural or unnatural regulators of autoimmunity? Curr. Opin. Immunol. 15: 683– 689. 11. Hammond, K. J., L. D. Poulton, L. J. Palmisano, P. A. Silveira, D. I. Godfrey, and A. G. Baxter. 1998. ␣␤-T cell receptor (TCR)⫹CD4⫺CD8⫺ (NKT) thymocytes prevent insulin-dependent diabetes mellitus in nonobese diabetic (NOD)/Lt mice by the influence of interleukin (IL)-4 and/or IL-10. J. Exp. Med. 187: 1047–1056. 12. Seino, K. I., K. Fukao, K. Muramoto, K. Yanagisawa, Y. Takada, S. Kakuta, Y. Iwakura, L. Van Kaer, K. Takeda, T. Nakayama, et al. 2001. Requirement for natural killer T (NKT) cells in the induction of allograft tolerance. Proc. Natl. Acad. Sci. USA 98: 2577–2581. 13. Singh, A. K., M. T. Wilson, S. Hong, D. Olivares-Villagomez, C. Du, A. K. Stanic, S. Joyce, S. Sriram, Y. Koezuka, and L. Van Kaer. 2001. Natural killer T cell activation protects mice against experimental autoimmune encephalomyelitis. J. Exp. Med. 194: 1801–1811. 14. Roelofs-Haarhuis, K., X. Wu, and E. Gleichmann. 2004. Oral tolerance to nickel requires CD4⫹ invariant NKT cells for the infectious spread of tolerance and the induction of specific regulatory T cells. J. Immunol. 173: 1043–1050. 15. Fillatreau, S., C. H. Sweenie, M. J. McGeachy, D. Gray, and S. M. Anderton. 2002. B cells regulate autoimmunity by provision of IL-10. Nat. Immunol. 3: 944 –950. 16. D’Orazio, T. J., and J. Y. Niederkorn. 1998. Splenic B cells are required for tolerogenic antigen presentation in the induction of anterior chamber-associated immune deviation (ACAID). Immunology 95: 47–55. 17. Roelofs-Haarhuis, K., X. Wu, M. Nowak, M. Fang, S. Artik, and E. Gleichmann. 2003. Infectious nickel tolerance: a reciprocal interplay of tolerogenic APCs and T suppressor cells that is driven by immunization. J. Immunol. 171: 2863–2872. 18. Sonoda, K. H., and J. Stein-Streilein. 2002. CD1d on antigen-transporting APC and splenic marginal zone B cells promotes NKT cell-dependent tolerance. Eur. J. Immunol. 32: 848 – 857. 19. Sonoda, K. H., D. E. Faunce, M. Taniguchi, M. Exley, S. Balk, and J. Stein-Streilein. 2001. NK T cell-derived IL-10 is essential for the differentiation of antigen-specific T regulatory cells in systemic tolerance. J. Immunol. 166: 42–50. 20. Faunce, D. E., K. H. Sonoda, and J. Stein-Streilein. 2001. MIP-2 recruits NKT cells to the spleen during tolerance induction. J. Immunol. 166: 313–321. 21. Cui, J., T. Shin, T. Kawano, H. Sato, E. Kondo, I. Toura, Y. Kaneko, H. Koseki, M. Kanno, and M. Taniguchi. 1997. Requirement for V␣14 NKT cells in IL-12mediated rejection of tumors. Science 278: 1623–1626.

The Journal of Immunology 22. Artik, S., K. Haarhuis, X. Wu, J. Begerow, and E. Gleichmann. 2001. Tolerance to nickel: oral nickel administration induces a high frequency of anergic T cells with persistent suppressor activity. J. Immunol. 167: 6794 – 6803. 23. Artik, S., C. von Vulte´e, E. Gleichmann, T. Schwarz, and P. Griem. 1999. Nickel allergy in mice: enhanced sensitization capacity of nickel at higher oxidation states. J. Immunol. 163: 1143–1152. 24. Steinman, R. M., G. Kaplan, M. D. Witmer, and Z. A. Cohn. 1979. Identification of a novel cell type in peripheral lymphoid organs of mice. V. Purification of spleen dendritic cells, new surface markers, and maintenance in vitro. J. Exp. Med. 149: 1–16. 25. Thierse, H. J., C. Moulon, Y. Allespach, B. Zimmermann, A. Doetze, S. Kuppig, D. Wild, F. Herberg, and H. U. Weltzien. 2004. Metal-protein complex-mediated transport and delivery of Ni2⫹ to TCR/MHC contact sites in nickel-specific human T cell activation. J. Immunol. 172: 1926 –1934. 26. Weintraub, J. P., and P. L. Cohen. 1999. Ectopic expression of B7-1 (CD80) on T lymphocytes in autoimmune lpr and gld mice. Clin. Immunol. 91: 302–309. 27. Ferber, I., G. Schonrich, J. Schenkel, A. L. Mellor, G. J. Hammerling, and B. Arnold. 1994. Levels of peripheral T cell tolerance induced by different doses of tolerogen. Science 263: 674 – 676. 28. Danadevi, K., R. Rozati, B. B. Saleha, and P. Grover. 2004. In vivo genotoxic effect of nickel chloride in mice leukocytes using comet assay. Food Chem. Toxicol. 42: 751–757. 29. Fang, W., K. A. Nath, M. F. Mackey, R. J. Noelle, D. L. Mueller, and T. W. Behrens. 1997. CD40 inhibits B cell apoptosis by up-regulating Bcl-xL expression and blocking oxidant accumulation. Am. J. Physiol. 272: C950 –C956. 30. Bates, S., and K. H. Vousden. 1999. Mechanisms of p53-mediated apoptosis. Cell. Mol. Life Sci. 55: 28 –37. 31. Owen-Schaub, L. B., W. Zhang, J. C. Cusack, L. S. Angelo, S. M. Santee, T. Fujiwara, J. A. Roth, A. B. Deisseroth, W. W. Zhang, and E. Kruzel. 1995. Wild-type human p53 and a temperature-sensitive mutant induce Fas/APO-1 expression. Mol. Cell. Biol. 15: 3032–3040. 32. Miyashita, T., S. Krajewski, M. Krajewska, H. G. Wang, H. K. Lin, D. A. Liebermann, B. Hoffman, and J. C. Reed. 1994. Tumor suppressor p53 is a regulator of bcl-2 and bax gene expression in vitro and in vivo. Oncogene 9: 1799 –1805. 33. Watanabe-Fukunaga, R., C. I. Brannan, N. G. Copeland, N. A. Jenkins, and S. Nagata. 1992. Lymphoproliferation disorder in mice explained by defects in Fas antigen that mediates apoptosis. Nature 356: 314 –317.

4589 34. Weintraub, J. P., R. A. Eisenberg, and P. L. Cohen. 1997. Up-regulation of Fas and the costimulatory molecules B7-1 and B7-2 on peripheral lymphocytes in autoimmune B6/gld mice. J. Immunol. 159: 4117– 4126. 35. Sy, M. S., S. D. Miller, H. B. Kowach, and H. N. Claman. 1977. A splenic requirement for the generation of suppressor T cells. J. Immunol. 119: 2095–2099. 36. Faunce, D. E., and J. Stein-Streilein. 2002. NKT cell-derived RANTES recruits APCs and CD8⫹ T cells to the spleen during the generation of regulatory T cells in tolerance. J. Immunol. 169: 31–38. 37. Wigginton, J. M., J. K. Lee, T. A. Wiltrout, W. G. Alvord, J. A. Hixon, J. Subleski, T. C. Back, and R. H. Wiltrout. 2002. Synergistic engagement of an ineffective endogenous anti-tumor immune response and induction of IFN-␥ and Fas-ligand-dependent tumor eradication by combined administration of IL-18 and IL-2. J. Immunol. 169: 4467– 4474. 38. Lisbonne, M., P. Hachem, M. B. Tonanny, J. M. Fourneau, S. Sidobre, M. Kronenberg, P. Van Endert, M. Dy, E. Schneider, and M. C. Leite-de-Moraes. 2004. In vivo activation of invariant V␣14 natural killer T cells by ␣-galactosylceramide sequentially induces Fas-dependent and -independent cytotoxicity. Eur. J. Immunol. 34: 1381–1388. 39. Margenthaler, J. A., K. Landeros, M. Kataoka, and M. W. Flye. 2002. CD1dependent natural killer (NK1.1⫹) T cells are required for oral and portal venous tolerance induction. J. Surg. Res. 104: 29 –35. 40. Sonoda, K. H., M. Exley, S. Snapper, S. P. Balk, and J. Stein-Streilein. 1999. CD1-reactive natural killer T cells are required for development of systemic tolerance through an immune-privileged site. J. Exp. Med. 190: 1215–1226. 41. Kaneko, Y., M. Harada, T. Kawano, M. Yamashita, Y. Shibata, F. Gejyo, T. Nakayama, and M. Taniguchi. 2000. Augmentation of V␣14 NKT cell-mediated cytotoxicity by interleukin 4 in an autocrine mechanism resulting in the development of concanavalin A-induced hepatitis. J. Exp. Med. 191: 105–114. 42. Gao, Y., J. M. Herndon, H. Zhang, T. S. Griffith, and T. A. Ferguson. 1998. Antiinflammatory effects of CD95 ligand (FasL)-induced apoptosis. J. Exp. Med. 188: 887– 896. 43. Chen, Y. G., C. M. Choisy-Rossi, T. M. Holl, H. D. Chapman, G. S. Besra, S. A. Porcelli, D. J. Shaffer, D. Roopenian, S. B. Wilson, and D. V. Serreze. 2005. Activated NKT cells inhibit autoimmune diabetes through tolerogenic recruitment of dendritic cells to pancreatic lymph nodes. J. Immunol. 174: 1196 –1204.