IL-18 Induces Monocyte Chemotactic Protein-1 Production in Macrophages through the Phosphatidylinositol 3-Kinase/Akt and MEK/ERK1/2 Pathways¹

Innate immunity lies behind most inflammatory responses, which are triggered by macrophages and polymorphonuclear leukocytes through their innate receptors (1). Macrophages have been shown to be activated during an inflammatory response and rapidly produce multiple inflammatory molecules to induce the recruitment and activation of circulating leukocytes (2, 3, 4). IL-18 is one of the most important innate cytokines produced from macrophages in the early stages of the inflammatory immune response. IL-18 was originally identified as IFN-γ-inducing factor (5) and induces Th1-mediated immune responses in collaboration with IL-12, whereas production of Th2 cytokines by IL-18 does not require IL-12 (6, 7). It was previously reported that IL-18 induces the secretion of chemokines such as IL-8 from human PBMCs via TNF-α production (8). In addition, IL-18 has been shown to be up-regulated in human inflammatory and autoimmune diseases, including rheumatoid arthritis, type 1 diabetes, systemic lupus erythematosus, multiple sclerosis, and Crohn’s disease (9).

Monocyte chemoattractant protein 1 (MCP-1),³ also known as CCL2, was first designated monocyte chemotactic and activating factor because it stimulates chemotactic migration of human monocytes and activates them to kill tumors in vitro (10). In addition, MCP-1 is known to be a major chemoattractant for memory T cells (11) and induces degranulation and histamine release from recruited leukocytes such as NK cells, CD8⁺ T cells, and basophils at inflammatory sites (12, 13). MCP-1 has been shown to be expressed in many different inflammatory diseases, such as atherosclerosis, allergic asthma, inflammatory bowel disease, and allogenic transplant rejection, which are characterized by the infiltration of mononuclear cells (14). The expression of MCP-1 was found to be correlated with the severity of disease in multiple sclerosis and in its animal model, experimental autoimmune encephalomyelitis (15). The presence of inflammatory cells in the joints of patients with rheumatoid arthritis was found to be related to the expression of MCP-1 in the synovial fluid (16).

Both IL-18 and MCP-1 have been shown to be involved in inflammatory immune responses. However it has been unclear whether there is any relationship in the induction process between IL-18 and MCP-1. This investigation was initiated to determine whether the inflammatory cytokine, IL-18, can induce MCP-1 production in macrophages and which signaling pathways are involved. We now report that IL-18 induced the production of MCP-1 in macrophages through the PI3K/Akt and MEK/ERK1/2 pathways. Inhibition of either the PI3K/Akt or MEK/ERK1/2 signaling pathway results in partial inhibition of IL-18-induced MCP-1 production, whereas inhibition of both of these signaling pathways results in the complete inhibition of IL-18-induced MCP-1 production. On the basis of these results, we propose that IL-18 alone can promote inflammatory responses in inflammatory diseases by the induction of MCP-1 production in macrophages and subsequent recruitment and activation of circulating leukocytes at the inflammatory site.

Eight-week-old male C57BL/6J and B6.129S7-Ifng^tm1Ts/J (IFN-γ knockout) mice were purchased from Taconic Farms and The Jackson Laboratory, respectively. Animals were maintained under specific pathogen-free conditions and provided with sterile food and water at the Animal Resource Centre, Faculty of Medicine, University of Calgary.

Biotinylated anti-CD11b, anti-CD11c, anti-B220, anti-CD3e, anti-DX5, and FITC-conjugated anti-CD11b Abs were purchased from BD Pharmingen. FITC-conjugated F4/80 Ab was obtained from eBioscience, and anti-biotin microbeads were obtained from Miltenyi Biotec. Recombinant mouse IL-18, rat IgG1 isotype, anti-IL-18 Ab, and anti-mouse JE/CCL2-blocking Ab were purchased from R&D Systems. Recombinant mouse IL-12 was obtained from Peprotech (Ottawa, ON, Canada). A PI3K inhibitor, LY294002, and an MEK inhibitor, PD98059, were purchased from Calbiochem. Rabbit Ab against IκBα was purchased from Santa Cruz Biotechnology. Rabbit Abs against p65, p38 MAPK, ERKs, Akt, stress-activated protein kinase/JNK (SAPK/JNK), phosphorylated p65 (Ser⁵³⁶), phosphorylated p38 MAPK (Thr¹⁸⁰)/Tyr(¹⁸²), phosphorylated ERKs (Thr²⁰²)/Tyr²⁰⁴), phosphorylated Akt (Ser⁴⁷³)/Thr³⁰⁸ and phosphorylated SAPK/JNK (Thr¹⁸³)/Tyr¹⁸⁵) were purchased from New England Biolabs. Mouse MCP-1 and IFN-γ ELISA kits were purchased from R&D Systems. NE-PER nuclear and cytoplasmic extraction reagents kit was purchased from Pierce.

Peritoneal macrophages were isolated as described previously (17, 18, 19). Mice were injected with 2 ml of 4% thioglycolate i.p., and peritoneal cells were harvested 4 days later and treated with biotinylated anti-mouse CD11c, CD3e, DX5, and B220 Abs. The cells were then washed using MACS buffer, incubated with anti-biotin microbeads, and macrophages were purified by negative selection using the MACS separation system (Miltenyi Biotec). Macrophage purity was assessed by flow cytometric analysis on a FACScan flow cytometer (BD Biosciences) after CD11b and F4/80 staining. The purity of the macrophages was ≥98%. The purified macrophages (1 × 10⁶ cells) were placed on 0.8 μm polycarbonate filters (Corning Costar), floated on 1 ml of RPMI 1640 medium (Invitrogen Life Technologies) supplemented with 2% FBS, 2 mM glutamine, 5 × 10⁻⁵ M 2-ME, 100 U/ml penicillin, and 100 μg/ml streptomycin and incubated for 24 h at 37°C in 5% CO₂ in a humidified incubator. This method resulted in significant reduction of the background activation level of signaling molecules by preventing nonspecific macrophage activation on culture plates.

Purified macrophages were stimulated with various doses of IL-18 (0.0056, 0.056, 0.56, and 5.6 nM) and MCP-1 production was measured. For further studies, we chose the dose of 5.6 nM, which resulted in the highest production of MCP-1 and was used to stimulate macrophages and T cells in previous studies (20, 21).

Macrophages were treated with various concentrations of IL-12, IL-18, or a combination of IL-12 and IL-18 in the presence or absence of a PI3K inhibitor (LY294002) or a MEK inhibitor (PD98059) for 48 h. To determine whether IFN-γ or TNF-α induce the production of MCP-1, macrophages were treated with various concentrations of IFN-γ or TNF-α for 48 h. The culture supernatant was harvested and used for quantification of IFN-γ or MCP-1 production using the respective ELISA kits (R&D Systems) according to the manufacturer’s protocol. The minimum detectable amount of mouse IFN-γ is <2 pg/ml using the mouse IFN-γ Quantikine ELISA kit.

The chemotactic ability of MCP-1 was assessed using a double-chamber system (Costar Transwell with 5 μm pore polycarbonate membrane inset; Corning) as previously described (22). Mouse monocytes were collected from the peritoneum of 8-wk-old male C57BL/6 mice. Briefly, the peritoneal cavity was washed with RPMI 1640 medium, the cells were harvested, and CD11b-positive cells were then purified using the MACS separation system. Culture medium (300 μl) harvested from macrophages stimulated with IL-18 was placed in the lower chamber with or without anti-MCP-1 blocking Ab (5 μg/ml), and the freshly isolated monocytes (3 × 10⁵ cells/well) were suspended in RPMI 1640 medium containing 0.5% BSA and added to the upper chamber. After incubation in 5% CO₂ for 4 h, the total number of cells that migrated to the lower chamber was counted by FACScan using forward and side-scatter gates for monocytes.

Purified macrophages were treated with IL-18 for 6 h and harvested. RT-PCR was performed as described previously (23). Briefly, total RNA was isolated from the macrophages using Trizol buffer (Invitrogen Life Technologies) and 0.5 μg of the isolated RNA was used to synthesize cDNA using Superscript II reverse transcriptase and oligo(dT)_12–18. PCR was then performed using specific primers for murine MCP-1, TNF-α, IL-12, IFN-γ, IL-18, and hypoxanthine phosphoribosyltransferase (HPRT): MCP-1, sense: 5′-AGAGAGCCAGACGGGAGGAA-3′, anti-sense: 5′-GTCACACTGGTCACTCCTAC-3′; TNF-α, sense: 5′-CAAAAGATGGGGGGCTTCCAGAAC-3′, anti-sense: 5′-AGTTAGCAAATCGGCTGACGGTGTG; IFN-γ, sense: 5′-AGCTCTGAGACAATGAACGC-3′, anti-sense: 5′-GGACAATCTCTTCCCCACCC-3′; IL-12p40, sense: 5′-AAACAGTGAACCTCACCTGTGACAC-3′, anti-sense: 5′-TTCATCTGCAAGTTCTTGGGCG-3′; IL-18, sense: 5′-ACTGTACAACCGCAGTAATACGG-3′, anti-sense: 5′AGTGAACATTACAGATTTATCCC-3′; HPRT, sense: 5′-GTAATGATCAGTCAACGGGGGAC-3′, anti-sense: 5′-CCAGCAAGCTTGCAACCTTAACCA-3′. The PCR conditions were optimized for each set of primers. The reaction was initially denatured at 95°C for 5 min, then subjected to the denaturation cycles at 95°C for 1 min, annealing at 58°C (68°C for TNF-α) for 1 min, and extension at 72°C for 1 min (MCP-1, 30; TNF-α, 32; IFN-γ, IL-12p40, IL-18, 40; HPRT, 33 cycles) before final extension at 72°C for 5 min. The products were then subjected to electrophoresis in a 1% agarose gel and detected by ethidium bromide staining.

To check for the nuclear translocation of NF-κB p65 in IL-18-stimulated macrophages, macrophages were treated with IL-18 or LPS as a control and nuclear extracts were prepared using the NE-PER nuclear and cytoplasmic extraction reagents (Pierce) as described in the manufacturer’s protocol. Nuclear extracts were subjected to 10% SDS-PAGE and Western blots were performed using specific Abs against p65 proteins.

To check the activation of kinase molecules, including Akt, ERK1/2, JNK, and p38 MAPK, in IL-18-stimulated macrophages, macrophages were treated with IL-18 or LPS as a control and harvested after various incubation times. Whole cell lysates were prepared as described previously (24, 25, 26, 27) and subjected to 10% SDS-PAGE. Western blots were performed using specific Abs against the phosphorylated form of Akt, ERK1/2, JNK, or p38 MAPK. To confirm that the same amount of cellular protein was loaded in each lane, the primary-Ab/secondary-Ab complex was removed by incubating the blot in mild Ab stripping solution (Chemicon) for 15 min at 37°C. The blots were then subjected to autoradiography to confirm that the Ab signal was removed. After this procedure, the blots were reprobed with the specific Abs against total Akt, ERK1/2, JNK, or p38 MAPK. The bands were detected by a chemiluminescence detection kit.

Macrophages were stimulated with IL-18 in the presence or absence of a PI3K inhibitor, LY294002, and/or a MEK inhibitor, PD98059, for 48 h, and the stimulated macrophages were harvested. The macrophages were then fixed in 50% ethanol, washed in PBS, and stained with PI (10 μg/ml final concentration) in the presence of 50 μg/ml RNase A for 20 min at room temperature. Cell cycle profile was analyzed by FACScan (BD Biosciences).

The statistical significance between groups was analyzed by Student’s t test. A level of p < 0.05 was considered to be significant. Data are expressed as means ± SD.

To determine the effect of IL-18 on the expression of MCP-1, purified peritoneal macrophages from C57BL/6J mice were treated with different concentrations of IL-18 (0.0056, 0.056, 0.56, or 5.6 nM) for 6 h. RNA was isolated from the macrophages, and the expression of MCP-1 was examined by RT-PCR. We found that the expression of MCP-1 mRNA increased in a dose-dependent manner (Fig. 1,A). Similarly, the secretion of MCP-1 into the medium also increased depending on the concentration of IL-18 added to the macrophage culture (Fig. 1,B). To confirm that MCP-1 production is due to IL-18, we stimulated macrophages with IL-18 in the presence of IL-18 blocking Ab or control rat IgG1 and examined the production of MCP-1 by ELISA. We found that the addition of IL-18 blocking Ab inhibited MCP-1 production by IL-18 in a dose-dependent manner, and 50 μg/ml anti-IL18 Ab almost completely inhibited IL-18-induced MCP-1 production (Fig. 1 C), indicating that IL-18 specifically induces MCP-1 production in macrophages.

To determine the biological activity of MCP-1 produced by IL-18 stimulation, we examined the migration of monocytes in response to MCP-1. Culture medium from IL-18-stimulated macrophages was applied to the lower chamber of a transwell with or without anti-MCP-1 blocking Ab, purified monocytes were added to the upper chamber, and the number of monocytes that migrated to the lower chamber was counted. We found that the number of monocytes that migrated in response toward medium from IL-18-stimulated macrophages was significantly higher than the number that migrated toward medium from unstimulated macrophages. Addition of anti-MCP-1 blocking Ab resulted in the decrease in the number of migrating monocytes to control levels (Fig. 1 D). These results suggest that IL-18 stimulates macrophages to produce biologically active MCP-1.

As IL-18 is known to be an IFN-γ-inducing factor (5) and the secretion of MCP-1 is up-regulated by IFN-γ (28), we first examined whether IFN-γ can induce MCP-1 production in macrophages. We stimulated macrophages with various concentrations of IFN-γ (0.013, 0.13, 1.3, and 6.4 nM) and measured MCP-1 production. Consistent with a previous report (29), we found that MCP-1 production was induced in a dose-dependent manner (Fig. 2,A). To determine whether MCP-1 secretion from macrophages is mediated by autocrine IFN-γ, we first examined the production of IFN-γ by ELISA in culture supernatants of macrophages treated with different doses of IL-18. We found that IFN-γ was not induced in IL-18-stimulated macrophages, regardless of the concentration of IL-18, and IL-12 induced a small amount of IFN-γ, whereas treatment with a combination of IL-12 and IL-18 significantly induced IFN-γ production (Fig. 2,B), as reported previously (20). These results suggest that IFN-γ may not contribute to the IL-18-induced production of MCP-1 in macrophages. To confirm that IFN-γ is not involved in IL-18-induced MCP-1 secretion, we examined MCP-1 production by ELISA in the culture supernatant of IL-18-treated macrophages from IFN-γ knockout mice. We found that there was no significant difference in the amount of MCP-1 secreted by macrophages stimulated with IL-18 between wild-type and IFN-γ knockout mice (Fig. 2,C). In addition, we examined the expression of other inflammatory cytokines including IL-12p40 and IL-18, as well as IFN-γ, by RT-PCR in IL-18-stimulated macrophages and did not detect expression of these cytokines (Fig. 2 D). Taken together, these results indicate that the production of MCP-1 by IL-18 stimulation is not mediated by autocrine IFN-γ, IL-12, or IL-18.

It was also reported that IL-18-stimulated MCP-1 secretion from human PBMC was mediated by TNF-α production (8). Therefore, we examined whether TNF-α can induce MCP-1 production in murine macrophages. We stimulated macrophages with various concentrations of TNF-α (0.029, 0.29, 2.9, and 5.9 nM) and measured MCP-1 production. We found that only a small amount of MCP-1 was produced, and production peaked at 2.9 nM of TNF-α. (Fig. 3,A). When we examined the expression of TNF-α by RT-PCR in macrophages stimulated with various concentrations of IL-18, the expression of TNF-α was not detected at any of the concentrations of IL-18 that resulted in MCP-1 production (Fig. 3 B), and no TNF-α production was detected in the culture supernatant by ELISA (data not shown). Taken together, these results suggest that IL-18-induced MCP-1 production in macrophages is not mediated by autocrine TNF-α production, but may be mediated by direct IL-18R signaling.

It is known that the IL-18 receptor is a member of the IL-1R family, which share a signaling cascade of sequential recruitment of myeloid differentiation 88 (MyD88) and IL-1R-associated kinase (IRAK), followed by activation of IκB kinase (IKK), degradation of IκBα, and release of NF-κB p65 to translocate into the nucleus (6). Therefore, we examined whether the IKK/NF-κB pathway is activated by IL-18 in macrophages. We stimulated macrophages with IL-18, prepared cell extracts, and examined the amount of IκBα protein by Western blot. We did not find any detectable change in the amount of IκBα in macrophages stimulated with IL-18 compared with unstimulated macrophages (Fig. 4,A), indicating that IκBα is not degraded. We also found that the amount of NF-κB p65 in the nuclear extracts of IL-18-stimulated macrophages was not different from unstimulated macrophages (Fig. 4,B), indicating that NF-κB p65 is not translocated into the nucleus. We then determined whether NF-κB was activated by examining phosphorylated NF-κB p65 (Ser⁵³⁶) by Western blot. No phosphorylated NF-κB p65 was detected in the cytoplasm (Fig. 4,C). In contrast, stimulation of macrophages with LPS, a known stimulant of NF-κB (30), clearly induced IκBα degradation, nuclear translocation, and phosphorylation of Ser⁵³⁶ on the NF-κB molecule (Fig. 4, A–C). All of these results consistently show that the IKK/NF-κB signaling pathway is not activated by IL-18 in macrophages.

IL-18 is also known to induce the activation of several signaling molecules such as PI3K/Akt in synovial fibroblasts (31, 32) and MAPKs and JNK in a human NK cell line (21, 33). Therefore, we determined whether the PI3K/Akt pathway is activated by IL-18 in macrophages. Stimulation of macrophages with IL-18 increased the phosphorylation of Akt. Treatment with the PI3K inhibitor, LY294002, resulted in inhibition of the phosphorylation (Fig. 5,A), indicating that IL-18 induces the activation of PI3K/Akt pathway in macrophages. We then examined the activation of p38 MAPK, ERK1/2, and JNK in IL-18-stimulated macrophages and found that only MEK/ERK1/2 pathway was activated, evidenced by the increase in phosphorylated ERK1/2 after IL-18 treatment (Fig. 5, B–D). Phosphorylation of ERK1/2 was blocked by treatment with the MEK inhibitor, PD98059 (Fig. 5,D). Stimulation of macrophages with LPS clearly induced the activation of both p38 MAPK and JNK kinases, but neither phosphorylated p38 MAPK nor phosphorylated JNK was detected in IL-18-stimulated macrophages (Fig. 5, B and C), indicating that IL-18 does not activate these signaling pathways in murine macrophages.

Because we found that IL-18 induces the activation of PI3K/Akt and MEK/ERK1/2 pathways, we determined whether these signaling pathways are involved in IL-18-induced MCP-1 production. We treated macrophages with the PI3K inhibitor, LY294002, or the MEK inhibitor, PD98059, or both during stimulation with IL-18 and examined the secretion of MCP-1 into the culture supernatant by ELISA. Inhibition of the PI3K/Akt pathway with LY294002 resulted in 51.6% decrease of MCP-1 secretion, and inhibition of MEK/ERK1/2 with PD98059 resulted in 63.7% decrease of MCP-1 secretion compared with macrophages stimulated with IL-18 without inhibitor treatment. Treatment with both inhibitors completely eliminated the secretion of MCP-1 to the level of macrophages without IL-18 stimulation (Fig. 6,A). To confirm these results, we examined the expression of MCP-1 mRNA by RT-PCR in macrophages stimulated with IL-18 with or without the inhibitors. Consistent with the ELISA results, the expression of MCP-1 mRNA was partially inhibited by treatment with either LY294002 or PD98059 and completely inhibited by a combination of the two (Fig. 6 B).

To determine whether the inhibition of MCP-1 secretion from macrophages by the inhibitors was due to any toxic effects, we examined apoptotic cell death of macrophages by PI staining and FACS analysis. We found no apoptotic cell death in macrophages treated with inhibitors or their vehicle, DMSO (Fig. 6 C). These results suggest that the inhibition of IL-18-induced MCP-1 secretion by these inhibitors is not due to any toxic effects, but to specific inhibition of the respective signaling pathways.

A pleiotropic cytokine, IL-18, has been shown to be involved in the regulation of both innate and adaptive immunity. IL-18 was found to activate macrophages and other immune cells to secrete proinflammatory cytokines and chemokines (2). Chemokines play a pivotal role in the mediation of inflammation. One chemokine, MCP-1, was previously shown to accelerate the pathogen-specific T cell immune response at the inflammatory site (11, 34, 35). IL-18 also contributes to inflammatory response by synergy with other inflammatory cytokines, particularly IL-12 (6, 7). Although both IL-18 and MCP-1 are involved in inflammatory immune responses, the relationship between the cytokine, IL-18, and the chemokine, MCP-1, is unclear.

In this study, we first examined whether IL-18 alone can induce MCP-1 production in macrophages and found that IL-18 activated macrophages to produce biologically active MCP-1 in a dose-dependent manner, evidenced by monocyte chemotaxis. The production of MCP-1 by IL-18 was not mediated by autocrine production of IFN-γ, IL-12, IL-18, or TNF-α. This result is different from previous reports (8) in which IL-18-induced secretion of MCP-1 in human PBMCs was mediated by TNF-α production. This difference may be due to differences in the cell types and species (human and animals) used. IL-18 was also previously shown to activate distinct signaling molecules in different cellular subsets (6, 21, 33, 36). Nevertheless, in our experimental conditions, IL-18 alone clearly stimulated the production of MCP-1 in macrophages.

Second, we investigated the signaling pathways involved in the production of MCP-1 in macrophages stimulated with IL-18. IL-18R engagement has been shown to induce the activation of several signaling pathways such as IKK/NF-κB, PI3K/Akt, MAPK (p38, p42, and p44) and JNK in several different cell types (37). IL-18R is a member of the IL-1R family, which shares a signaling cascade of sequential recruitment of MyD88 and IRAK, followed by activation of NF-κB. The activation of NF-κB was found to be required for IL-18-induced IFN-γ expression in a human myelomonocyte cell line (38). In addition, it was shown that NF-κB was activated in murine Th1 cells by IL-18 (39). Furthermore, IRAK-deficient mice showed impairment of IL-18-induced NF-κB activation, and IL-18-mediated NK and Th1 responses were also impaired (21, 40). On the basis of these previous reports, we assumed that NF-κB may be activated by stimulation with IL-18. We examined this possibility in macrophages. In contrast to our expectations, we found no activation of NF-κB after stimulation with IL-18. Neither the degradation of IκB nor the nuclear translocation of NF-κB p65 was observed in IL-18-stimulated macrophages. Our data clearly indicate that IL-18 does not induce activation of NF-κB in macrophages. This result is supported by a recent report which showed that NF-κB was not activated by IL-18 in human epithelial cells, but was activated by IL-1β (41). Therefore, the activation of NF-κB by IL-18 appears to depend on the immune environment and cell type.

The PI3K/Akt pathway was shown to be involved in IL-18-induced expression of cell adhesion molecules such as VCAM-1 and ICAM-1 in synovial fibroblasts (31, 32) and in IL-18-induced cardiomyocyte hypertrophy (42). However, it is not known whether IL-18 can induce the activation of the PI3K/Akt signaling pathway in macrophages. We examined this possibility. We found that IL-18 clearly induced the activation of the PI3K/Akt signaling pathway in macrophages. A recent study suggested that MAPKs play an important role in IL-18 signaling. IL-18-induced activation of p38 MAPK in an epithelial cell line and activation of p38MAPK and ERK1/2 in a human NK cell line was found (33). We also examined whether p38 MAPK, JNK, or MEK/ERK1/2 signaling pathways can be activated in macrophages by IL-18. We found that only the MEK/ERK1/2 signaling pathway was activated in IL-18-stimulated macrophages, but not p38 MAPK or JNK.

Third, we determined whether PI3K/Akt and/or MEK/ERK1/2 signaling pathways are involved in IL-18-induced MCP-1 production. We found that inhibition of either the PI3K/Akt or MEK/ERK1/2 pathway resulted in the attenuation of MCP-1 gene expression in IL-18-stimulated macrophages. However, the inhibition of both PI3K/Akt and MEK/ERK1/2 pathways resulted in the complete abolishment of MCP-1 gene expression. These results clearly indicate that PI3K/Akt and MEK/ERK1/2 signaling pathways are involved in IL-18-induced MCP-1 gene expression and subsequently MCP-1 production in murine macrophages.

In conclusion, we have shown for the first time that IL-18 alone can induce MCP-1 production in macrophages and this production was independent of IL-12 and was not mediated by autocrine IFN-γ or TNF-α. IL-18 did not activate IKK/NF-κB pathways, but activated PI3K/Akt and MEK/ERK1/2 signaling pathways, contributing to the production of the chemokine MCP-1, which plays an important role in the recruitment of leukocytes at the inflammatory site and subsequently the development of adaptive immunity. We suggest that IL-18 itself, in the absence of IL-12, may initiate the inflammatory responses in autoimmune, inflammatory, and infectious diseases.

We gratefully acknowledge the editorial assistance of Dr. A. L. Kyle and the technical assistance of R. Clark and L. Robertson for animal care and FACS analysis, respectively.

The authors have no financial conflict of interest.