The Journal of Immunology
Selective Expansion of Foxp3-Positive Regulatory T Cells and Immunosuppression by Suppressors of Cytokine Signaling 3-Deficient Dendritic Cells1 Yumiko Matsumura,* Takashi Kobayashi,* Kenji Ichiyama,* Ryoko Yoshida,* Masayuki Hashimoto,* Tomohito Takimoto,* Kentaro Tanaka,* Takatoshi Chinen,* Takashi Shichita,* Tony Wyss-Coray,‡ Katsuaki Sato,† and Akihiko Yoshimura2* Dendritic cells (DCs) induce immunity and immunological tolerance as APCs. It has been shown that DCs secreting IL-10 induce IL-10ⴙ Tr1-type regulatory T (Treg) cells, whereas Foxp3-positive Treg cells are expanded from naive CD4ⴙ T cells by coculturing with mature DCs. However, the regulatory mechanism of expansion of Foxp3ⴙ Treg cells by DCs has not been clarified. In this study, we demonstrated that suppressors of cytokine signaling (SOCS)-3-deficient DCs have a strong potential as Foxp3ⴙ T cell-inducing tolerogenic DCs. SOCS3ⴚ/ⴚ DCs expressed lower levels of class II MHC, CD40, CD86, and IL-12 than wild-type (WT)-DCs both in vitro and in vivo, and showed constitutive activation of STAT3. Foxp3ⴚ effector T cells were predominantly expanded by the priming with WT-DCs, whereas Foxp3ⴙ Treg cells were selectively expanded by SOCS3ⴚ/ⴚ DCs. Adoptive transfer of SOCS3ⴚ/ⴚ DCs reduced the severity of experimental autoimmune encephalomyelitis. Foxp3ⴙ T cell expansion was blocked by anti-TGF-␤ Ab, and SOCS3ⴚ/ⴚ DCs produced higher levels of TGF-␤ than WT-DCs, suggesting that TGF-␤ plays an essential role in the expansion of Foxp3ⴙ Treg cells. These results indicate an important role of SOCS3 in determining on immunity or tolerance by DCs. The Journal of Immunology, 2007, 179: 2170 –2179.
endritic cells (DCs)3 include a heterogeneous family of professional APCs involved in the initiation of immunity and in immunological tolerance. Specifically, peripheral tolerance can be achieved and maintained by promoting regulatory T (Treg) cell responses and/or T cell anergy or deletion (1, 2). DC maturation is paralleled by up-regulation of MHC class II and costimulatory CD80/CD86 molecules and by the production of IL-12. In contrast, it has been proposed that DCs with an immature phenotype induce the anergic/suppressive T cells (3). Thus, the tolerogenic potential of DCs may be correlated with the absence of activation of DCs by proinflammatory signals. However, in vivo experiments have suggested that induction of T cell toler-
*Division of Molecular and Cellular Immunology, Medical Institute of Bioregulation, Kyushu University, Fukuoka, Japan; †Laboratory for Dendritic Cell Immunobiology, Research Center for Allergy and Immunology, RIKEN Yokohama Institute, Tsurumiku, Yokohama, Japan; and ‡Department of Neurology and Neurological Sciences, Stanford University School of Medicine, Stanford, CA 94305 Received for publication October 16, 2006. Accepted for publication June 12, 2007. 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 work was supported by special grants-in-aid from the Ministry of Education, Science, Technology, Sports and Culture of Japan (to A.Y. and K.T.), the Yamanouchi Foundation for Research on Metabolic Disorders (to A.Y.), the Takeda Science Foundation (to K.T.), the Kato Memorial Foundation (to K.T.), the Kanae Foundation for the Promotion of Medical Science (to K.T.), and the Uehara Memorial Foundation (to T.C.). 2 Address correspondence and reprint requests to Dr. Akihiko Yoshimura, Division of Molecular and Cellular Immunology, Medical Institute of Bioregulation, Kyushu University, 3-1-1 Maidashi, Higashi-ku, Fukuoka 812-8582, Japan. E-mail address: [email protected]
3 Abbreviations used in this paper: DC, dendritic cell; BMDC, bone marrow-derived DC; EAE, experimental autoimmune encephalomyelitis; KLH, keyhole limpet hemocyanin; LN, lymph node; LysM, lysozyme M; MOG, myelin oligodendrocyte glycoprotein; SEAP, secreted alkaline phosphatase; SOCS, suppressor of cytokine signaling; Tr1, IL-10-producing regulatory T; Treg, T regulatory; WT, wild type.
Copyright © 2007 by The American Association of Immunologists, Inc. 0022-1767/07/$2.00 www.jimmunol.org
ance is mediated by a subtype of specialized DCs (4). DC residence in a tolerizing milieu, e.g., in mucosal or immune privileged sites, affects the DC capacity for priming Treg cells. DCs isolated from Peyer’s patches, lungs, or the anterior chamber of the eye display a mature phenotype, secrete IL-10, but not IL-12, and drive the development of IL-10-producing regulatory T (Tr1) cells. Furthermore, DCs treated with intestinal peptides, such as vasoactive intestinal peptide and pituitary adenylate cyclase-activating polypeptide, have also been shown to induce Tr1 (5). DCs also enhance Ag-specific, TGF-␤1-producing Treg cells (6). Additionally, mature DCs expand CD4⫹CD25⫹ Treg cells (Foxp3⫹ Treg cells) with retained suppressive activity and that are capable of inhibiting diabetes development in NOD mice (7). These studies highlight the fact that mature DCs might be critical for the restimulation and/or expansion of functional Treg cells. Foxp3⫹ Treg cells have been shown to be expanded or induced by TCR stimulation in the presence of TGF-␤1 (8). However, little is known about the mechanism of induction and/or expansion of Foxp3⫹ T cells by DCs. STATs are cytoplasmic transcription factors that are key mediators of cytokine and growth factor signaling pathways. Recently, one of the members of the STAT family, STAT3, has emerged as a negative regulator of inflammatory responses. This is mainly because STAT3 is an essential transcription factor that transmits the signal of IL-10 (9). IL-10-treated DCs (IL-10-DCs) have been shown to be tolerogenic and to induce Tr1-type Treg cells (10). In contrast, STAT3-deficient DCs induce hyperactivation of T cells (11). Thus, the activation status of STAT3 in DCs could be an important regulator of immune regulation. Suppressor of cytokine signaling (SOCS) proteins regulate the strength of cytokine signals. Among these, SOCS3 is strongly induced by a variety of cytokines and other stimulators, including IL-6, G-CSF, erythropoietin, leptin, and LPS. Because binding of the Src homology 2 domain of SOCS3 is relatively specific to
The Journal of Immunology receptors for STAT3-activating cytokines, such as IL-6, G-CSF, and leptin, the effect of SOCS3 has been shown to be relatively restricted to STAT3 (12). However, SOCS3 does not inhibit IL-10 signaling, because the Src homology 2 domain of SOCS3 does not bind to the IL-10R (13). In this study, we investigate the role of SOCS3 in DCs by using SOCS3-deficient bone marrow-derived DCs (BMDCs). We found that constitutive activation of STAT3 and an immature phenotype were maintained in SOCS3⫺/⫺ DCs after LPS treatment. SOCS3⫺/⫺ DCs suppressed the development of experimental autoimmune encephalomyelitis (EAE). In addition, Foxp3⫹ TGF␤1⫹ Treg cells, but not Foxp3⫺ effector T cells, were expanded by SOCS3⫺/⫺ DCs. From the effect of anti-TGF-␤ Ab and the levels of TGF-␤ in DCs, we propose that SOCS3 negatively regulates Treg expansion by modulating TGF-␤ secretion from DCs. Our study suggests that SOCS3 in DCs is an important determinant for immunity or tolerance.
Materials and Methods Mice Conditional targeting of SOCS3 using the loxP system and crossing with lysozyme M (LysM)-Cre mice have been described (13, 14). To delete the SOCS3 gene in hemopoietic stem cells, Tie2-Cre mice were crossed with SOCS3fl/fl mice (14). Age-matched SOCS3fl/fl (SOCS3⫹/⫹) and LysM-Cre:SOCS3fl/fl or Tie2-Cre:SOCS3fl/fl (SOCS3⫺/⫺) mice were used for analysis. All experiments were approved by the Animal Ethics Committee of Kyusyu University.
DC preparation BMDCs were prepared from bone marrow suspensions obtained from the femurs and tibias, as described (15). Bone marrow cells were cultured in 20 ng/ml murine GM-CSF (PeproTech) or culture supernatant from J558L cells transfected with the murine GM-CSF gene. On day 8, LPS (SigmaAldrich) was added at 100 ng/ml for 16 h. IL-10-treated DCs (IL-10-DCs) were generated from bone marrow cells cultured with murine GM-CSF (20 ng/ml) for 8 days and murine IL-10 (20 ng/ml; PeproTech) for 2 days, and then stimulated with LPS. Floxed Socs3 gene deletion was evaluated by PCR using genomic DNAs obtained from BMDCs, as described (13, 14).
Preparation of CD4⫹ T cells and allogeneic MLR Spleen and lymph node (LN) cells of BALB/c or OT-II transgenic mice were incubated with anti-CD4-coated MACS magnetic beads (Miltenyi Biotec) and were positively selected. The purity of CD4⫹ T cells was ⬎95%, as determined by flow cytometry. CD4⫹ T cells (2 ⫻ 105) from BALB/c mice (H-2d) were cocultured with BMDCs (1 ⫻ 104) from SOCS3⫹/⫹ or SOCS3⫺/⫺ mice (H-2b) in RPMI 1640 medium with 10% FCS and antibiotics for 3– 4 days. Expanded T cells (4 ⫻ 105) were restimulated with indicated concentrations of plate-bound anti-CD3 Ab (0.1 ml/well), and then cell proliferation was assessed after 72 h of culture by a [3H]thymidine incorporation assay. Cytokine concentration in the supernatants was measured by ELISA. For Ab-blocking experiments, antiTGF-␤ mAb (1D11), anti-IFN-␥ mAb (R4.6A2), and anti-IL-2 mAb (JES6-1A12) were added at 10 g/ml during MLR.
Suppression assay To isolate naturally occurring Treg cells (CD4⫹CD25⫹ T cells), CD4⫹ T cells were positively selected with anti-CD25-coated MACS beads (purity ⬎98%). CD4⫹ T cells or CD4⫹CD25⫹ T cells were expanded by BMDCs for 4 days, as described above, and then CD25⫹ cells were collected by MACS beads (Miltenyi Biotec) and used as suppressor cells. To isolate responder CD4⫹CD25⫺ T cells, splenocytes were labeled with biotin-conjugated anti-CD8 (53-6.7), anti-CD11b (M1/70), anti-B220 (RA3-6B2), anti-DX-5, anti-Ter119, and anti-CD25 (PC61) mAbs (eBioscience); incubated with streptavidin magnetic beads; and loaded onto MACS separation columns. Indicated number of responder CD4⫹CD25⫺ T cells and suppressor CD25⫹ T cells expanded by DCs was cultured with irradiated whole spleen cells (1 ⫻ 105) with 1.0 g/ml anti-CD3 Ab. The number of responder cells was fixed (1 ⫻ 105) and that of suppressors was varied. [3H]Thymidine was added for the last 16 h of a 72-h assay.
2171 Flow cytometric analysis Cells were stained with FITC-, PE-, biotin-, and allophycocyanin-conjugated anti-CD86 (GL1), and I-Ab (AF6-120.1) from BD Pharmingen, and CD40 (1C10), CD80 (16-10A1), CD4 (RM4-5), and CD25 (PC61) from eBioscience. Biotinylated Ab staining was followed by streptavidin-PerCP Cy5.5 (BD Pharmingen). For anti-mouse allophycocyanin-Foxp3 (FJK16s) (eBioscience) intracellular staining, cells were labeled with anti-CD4 and anti-CD25 Abs, fixed, and permeabilized, according to the manufacturer’s protocol.
Western blotting, ELISA, and NO measurement Western blotting for detection of phosphorylated STATs was performed, as described (16). IL-4, IL-6, IL-10, and IFN-␥ were detected in culture supernatants with OptEIA ELISA sets (BD Biosciences), and IL-17 was detected with DuoSet ELISA Development Systems (R&D Systems), as per the manufacturer’s instructions. NO was measured as the accumulation of nitrite in the incubation medium, as described (16).
EAE induction and treatment by DCs Myelin oligodendrocyte glycoprotein (MOG) peptide 35–55 (MEVGW YRSPFSRVVHLYRNGK) (BEX) was used to induce EAE in C57BL/6 mice. Briefly, mice were injected s.c. with 200 g of MOG peptide in 100 l of PBS emulsified in 100 l of CFA that was further enriched with 5 mg/ml Mycobacterium tuberculosis (H37Ra; Difco/BD Pharmingen). In addition, 500 ng of pertussis toxin (Calbiochem) was injected i.p. on days 0 and 2. Paralysis was evaluated according to the following scores: 0 ⫽ no signs, 1 ⫽ full tail, 2 ⫽ hind limbs, 3 ⫽ complete back, 4 ⫽ fore limbs, and 5 ⫽ dead. Three hours after stimulation and peptide pulse, DCs were injected (1 ⫻ 106 cells) i.v. three times on days 7, 5, and 3 before EAE induction (day 0) (17). DC-treated and control mice were immunized with MOG peptide, according to the protocol for EAE induction. After indicated days, splenic CD4⫹ T cells were isolated and cultured (2 ⫻ 105 cells/well) with plate-bound anti-CD3 Ab in a 96-well plate.
RT-PCR The cells were lysed in TRIzol reagent (Invitrogen Life Technologies) for RNA preparation. RT-PCR was performed with a standard procedure. The expression level of G3PDH was first evaluated as an internal control. The primer sequences and PCR cycles were as follows: IFN-␥, 5⬘-gca tcg ttt tgg gtt ctc ttg gct gtt act gc-3⬘ and 5⬘-ctc ctt ttt cgc ttc cct gtt tta gct g-3⬘ (30 cycles); IL-10, 5⬘-tac ctg gta gaa gtg atg cc-3⬘ and 5⬘-cat cat gta tgc ttc tat gc-3⬘ (30 cycles); and TGF-␤1, 5⬘-taa tgg tgg acc gca aca acg c-3⬘ and 5⬘-gac gga ata cag ggc ttt cg-3⬘ (30 cycles); IL-17, 5⬘-cag cag cga tca tcc ctc aaa g-3⬘ and 5⬘-cag gac cag gat ctc ttg ctg-3⬘ (30 cycles); Foxp3, 5⬘-cag ctg cct aca gtg ccc cta g-3⬘ and 5⬘-cat ttg cca gca gtg ggt ag-3⬘ (32 cycles); CTLA-4, 5⬘-tgg act ccg gag gta caa ag-3⬘ and 5⬘-cag tcc ttg gat ggt gag gt-3⬘ (30 cycles); G3PDH, 5⬘-acc aca gtc cat gcc atc ac-3⬘ and 5⬘-tcc acc acc ctg ttg ctg ta-3⬘ (28 cycles).
Transfer of Ag-pulsed DCs into mice DCs from SOCS3⫹/⫹ and SOCS3⫺/⫺ mice were incubated in culture medium with 50 g/ml keyhole limpet hemocyanin (KLH) for 16 h, resuspended in PBS (5 ⫻ 106 cells of DCs in 40 l), and then administered into the hind footpads. The draining LNs were removed and teased into a cell suspension on day 5. The lymphocytes were cultured with or without Ag (20 g/ml KLH) at 5 ⫻ 105 cells in 96-well plates for 96 h, and cytokine levels in the culture supernatant were determined by ELISA (15). For adoptive transfer into OT-II mice, DCs from SOCS3⫺/⫺ and SOCS3⫹/⫹ mice were cultured in the presence of 1 g/ml OVA peptide 323–339 (ISQAVHAAHAEINEAGR) for 16 h, then 4 ⫻ 106 DCs were injected i.v. into OTII transgenic mice. After 7 days, splenic CD4⫹ T cells were analyzed by Foxp3 intracellular staining.
TGF-␤ bioassay MFB-F11 cells (18) were seeded at 1– 4 ⫻ 104 cells/well in 96-well plates. After overnight incubation, cells were washed twice with PBS and incubated in 50 l of serum-free DMEM supplemented with penicillin/streptomycin for 2 h before test samples were added in 50 l vol. Ten-l aliquots of the culture supernatants were collected after 24-h incubation with MEB-F11 cells. Secreted alkaline phosphatase (SEAP) activity was measured using Reporter Assay Kit SEAP (Toyoba), according to the manufacturer’s instructions, and measured with a Lumat LB 9507 tube luminometer (EG & G Berthold).
LOSS OF SOCS3 INDUCES TOLEROGENIC DC amined the effect of SOCS3 gene disruption on STAT3 activation. STAT3 was constitutively activated in SOCS3⫺/⫺ DCs and further hyperactivated in response to LPS (Fig. 1C). These data were consistent with the previous observations that SOCS3 negatively regulates STAT3. Effect of SOCS3 deletion on DC maturation
FIGURE 1. SOCS3 gene deletion in BMDCs from LysM-Cre and Tie2-Cre:SOCS3fl/fl mice. A, Schematic diagrams of the floxed Socs3 locus in the SOCS3 flox allele, and the deleted Socs3 locus (SOCS3del). The positions of PCR primers A and B are also shown. B, PCR analysis of genomic DNA from BMDCs. The PCR product of the SOCS3 flox allele is 2000 bp. A band of 250 bp indicates Cre-mediated deletion of the SOCS3 gene in LysM-Cre and Tie2-Cre:SOCS3fl/fl-derived BMDCs. Deletion efficiency is almost 100%. C, Effects of LPS on STAT3 activation in SOCS3⫺/⫺ DCs. DCs from SOCS3⫹/⫹ mice and SOCS3⫺/⫺ mice were cultured with LPS for 24 h. Whole cell extracts were immunoblotted with the indicated Abs. Data represent one of three similar experiments.
LPS injection and in vivo DC and T cell analysis LPS (10 g/head) derived from Escherichia coli O55:B5 (Sigma-Aldrich) was i.p. injected into SOCS3⫺/⫺ (LysM-Cre:SOCS3fl/fl or Tie2-Cre: SOCS3fl/fl) and SOCS3⫹/⫹ mice. Sixteen hours later, spleens and mesenteric LNs were dissected. Single-cell suspensions were prepared and stained for flow cytometric analysis with anti-CD86, I-Ab, CD40, and CD11c Abs. Permeabilized T cells were stained with anti-CD4, anti-CD25, and anti-Foxp3 Abs.
Results Hyperactivation of STAT3 in SOCS3-deficient DCs SOCS3-deficient mice die during embryonic development as a result of placental deficiency. Thus, to delete the SOCS3 gene in BMDCs, SOCS3-flox/flox mice were crossed with either Tie2-Cre mice or LysM-Cre mice (Fig. 1A) (13, 14). Deletion of the SOCS3 gene was confirmed in BMDCs from both Tie2-Cre:SOCS3fl/fl mice and LysM-Cre:SOCS3fl/fl mice, in which the SOCS3 gene has been deleted in all hemopoietic lineages and monocytes/neutrophils, respectively. The SOCS3 gene was efficiently deleted in BMDCs from these mice (Fig. 1B). Because similar data were obtained in BMDCs from Tie2-Cre:SOCS3fl/fl mice and LysM-Cre:SOCS3fl/fl mice, and we used only BMDCs as DCs for all experiments, data of BMDCs from Tie2-Cre:SOCS3fl/fl mice are shown and designated as SOCS3⫺/⫺ DCs throughout the text, except for notification. Loss of the SOCS3 protein was confirmed by Western blotting in DCs stimulated with LPS. SOCS3 has been shown to be an important negative regulator for STAT3 (12). Therefore, we ex-
We then examined LPS-induced maturation of DCs. LPS up-regulates class II MHC (I-Ab), CD40, and costimulators (CD80 and CD86), and induces inflammatory cytokines, such as IFN-␥, IL-12, and IL-6 in wild-type (WT) (SOCS3⫹/⫹) DCs (Fig. 2, A and B). However, up-regulation of these surface molecules and inflammatory cytokines was severely repressed in SOCS3⫺/⫺ DCs (Fig. 2, A and B). In contrast, secretion of IL-10 and production of NO were enhanced in SOCS3⫺/⫺ DCs compared with SOCS3⫹/⫹ DCs. All these immature phenotypes were observed to be similar to those in tolerogenic IL-10-treated DCs (IL-10-DCs) (10, 19) (data not shown). To confirm that SOCS3-deficeint DCs were resistant to LPSinduced maturation in vivo, we i.p. injected LPS into SOCS3⫺/⫺ mice, and I-Ab⫹ CD11c⫹ DC fractions were analyzed. As shown in Fig. 2C, the levels of CD80 and CD40 were not significantly different between WT and SOCS3⫺/⫺ splenic and LN DCs without LPS administration. As shown previously (20), injection of LPS strongly up-regulated CD86 and CD40 expression on splenic DCs in WT mice, whereas their up-regulation was partly repressed in SOCS3⫺/⫺ splenic DCs. In LN DCs, CD86 up-regulation was also impaired in SOCS3⫺/⫺ LN DCs (Fig. 2C). CD40 was not significantly up-regulated by LPS administration in both WT and SOCS3⫺/⫺ LN DCs (data not shown). Thus, these data suggest that SOCS3-deficient DCs were resistant to LPS-induced maturation not only in vitro, but also in vivo. Low proliferation capacity of T cells expanded by SOCS3⫺/⫺ DCs It has been shown that IL-10-DCs induce an Ag-specific anergy and suppressor activity in CD4⫹ T cells (4, 5, 10). Thus, we examined the nature of T cells expanded by SOCS3⫺/⫺ DCs in vitro. Allogeneic CD4⫹ T cells were incubated with WT or SOCS3⫺/⫺ DCs for 4 days, and then the same number of T cells was restimulated with anti-CD3 Ab. As shown in Fig. 3A, anti-CD3 Abmediated proliferation of T cells expanded by SOCS3⫺/⫺ DCs was severely inhibited compared with after coculture with WT-DCs, suggesting that T cells expanded by SOCS3⫺/⫺ DCs were anergic. As shown in Fig. 3B, not only IFN-␥, but also IL-10 productions were extremely low in T cells cocultured with SOCS3⫺/⫺ DCs. These findings were confirmed by RT-PCR analysis (Fig. 3C). Interestingly, TGF-␤1 levels as assessed by RT-PCR were high in T cells expanded by SOCS3⫺/⫺ DCs (Fig. 3C). Tr1-type T cells induced by tolerogenic DCs have been shown to express high levels of IL-10 (10, 19). Thus, T cells expanded by SOCS3⫺/⫺ DCs appeared to be different from Tr1 cells. Preferential expansion of Foxp3-positive T cells by SOCS3⫺/⫺ DCs Then we investigated the expression of Foxp3, which is essential for Treg development. Approximately 5– 8% of naive CD4⫹ T cells were CD25⫹ Foxp3⫹ (data not shown). It has been demonstrated that Foxp3⫹ Treg cells are expanded by mature DCs in vitro (6). As shown in Fig. 3D, ⬃50 –70% of CD25⫹ T cells became Foxp3 positive after priming with SOCS3⫺/⫺ DCs, whereas Foxp3-positive CD25⫹ T cells were ⬃20 –30% after priming with WT-DCs. Very high percentage of Foxp3⫹ T cells after coculture with SOCS3⫺/⫺ DCs is mostly due to lower expansion of CD25⫹
The Journal of Immunology
FIGURE 2. Expression of DC markers and cytokines in SOCS3⫺/⫺ DCs in response to LPS. A, Flow cytometric analysis of CD40, CD86, CD80, and MHC class II expression on DCs from SOCS3⫺/⫺ mice and littermates cultured in the presence of LPS for 24 h. Data shown are representative flow cytometric histograms from three independent experiments. B, IFN-␥, IL-12, IL-6, and IL-10 levels in the culture supernatant of 1 ⫻ 106 DCs after 24-h stimulation with LPS were measured with ELISA. NO concentration was determined by Griess reagent. Mean ⫹ SD of triplicate samples is shown (ⴱ, p ⬍ 0.0005; ⴱⴱ, p ⬍ 0.001; Student’s t test, SOCS3⫺/⫺ DC vs SOCS3⫹/⫹ DC). C, In vivo effects of LPS on splenic and LN DCs. LPS (10 g/head) was injected i.p. into SOCS3⫺/⫺ and SOCS3⫹/⫹ mice. Sixteen hours later, spleen and LN cells were analyzed. Histograms of CD40 or CD86 in I-Ab⫹ CD11c⫹ cells are shown. Dark lines and light gray histograms show with and without LPS administration, respectively. Data represent one of the two independent mice.
Foxp3⫺ effector T cells by SOCS3⫺/⫺ DCs (Fig. 3D). IL-10-DCs also hardly induced expansion of CD25⫹ Foxp3⫺ effector T cells; however, CD25⫹ Foxp3⫹ T cells was not also expanded
by IL-10-DCs. Thus, SOCS3⫺/⫺ DC was apparently different from IL-10-DC, and preferentially promoted expansion of Foxp3⫹ T cells.
LOSS OF SOCS3 INDUCES TOLEROGENIC DC
FIGURE 3. CD4⫹ T cells primed by SOCS3-deficient BMDCs. Allogeneic BALB/c CD4⫹ T cells (2 ⫻ 105) from BALB/c mice were cocultured with SOCS3⫹/⫹ and SOCS3⫺/⫺ BMDCs (1 ⫻ 104) for 4 days. Then the same number (4 ⫻ 105) of T cells was restimulated with plate-bound anti-CD3 Ab for 72 h. Proliferation (ⴱ, p ⬍ 0.005; Student’s t test) (A) and cytokine secretion (ⴱ, p ⬍ 0.001; ⴱⴱ, p ⬍ 0.01; ⴱⴱⴱ, p ⬍ 0.05; Student’s t test) (B) were measured by [3H]thymidine incorporation and ELISA, respectively. Error bars, SD of triplicates. C, Cytokines and Foxp3 levels in restimulated T cells were measured by RT-PCR. One microgram of total RNA from T cells was used as template. Data shown are representative from two independent experiments. D, Predominant expansion of Foxp3⫹ Treg cells primed by SOCS3⫺/⫺ DCs. To generate IL-10-DCs, WT-BMDCs were cultured in the presence of IL-10 for 48 h, then stimulated with LPS. CD4⫹ T cells from BALB/c mice were cocultured with SOCS3⫹/⫹, SOCS3⫺/⫺ DCs, or IL-10-DCs for 4 days, then stained with anti-CD4, anti-CD25, and anti-Foxp3 Abs. CD4⫹-gated cell populations are shown in upper panels. Foxp3-expression profiles gated on CD4⫹CD25⫹ cells or CD4⫹ are shown below. One representative experiment of four is shown.
To examine whether SOCS3⫺/⫺ DCs can promote stronger expansion of Foxp3⫹ T cells with Ag-specific manner than WT-DCs in vitro and in vivo, we used T cells from OTII-TCR transgenic mice. Naive T cells from TCR transgenic mice have been shown to express low levels of Foxp3 (7). Syngeneic BMDCs were pulsed with OVA and cocultured with CD4⫹ T cells from OTII mice. Similar to allogeneic responses, proliferation of T cells restimulated with anti-CD3 Ab was severely repressed by coculture with SOCS3⫺/⫺ DCs (Fig. 4A). In addition, expression of TGF-␤1 and Foxp3 was high in T cells after coculture with Ag-loaded SOCS3⫺/⫺ DCs (Fig. 4B). In vivo administration of WT and SOCS3⫺/⫺ DCs resulted in an enhancement of Foxp3⫹ T cell fractions in splenic CD4⫹ T cells (Fig. 4C). SOCS3⫺/⫺ DCs induced expansion of Foxp3⫹ T cells more strongly than WT DCs in vivo (Fig. 4C). These data suggest that SOCS3⫺/⫺ DC has a higher potential for expansion of Foxp3⫹ Treg cells compared with Foxp3⫺ effector T cells both in vitro and in vivo.
To confirm higher Foxp3⫹ T cell expansion in vivo, we compared splenic and LN T cells from WT and SOCS3⫺/⫺ mice (Fig. 4D). Thymic CD4⫹CD25⫹ T cell population was not different between WT and SOCS3⫺/⫺ mice (5.64% in Tie2-Cre:SOCS3fl/fl mice and 4.98% LysM-Cre:SOCS3fl/fl mice, whereas 5.33% in WT mice). However, the fractions of CD25⫹ Foxp3⫹ T cells in the spleen and LN of SOCS3⫺/⫺ mice were higher than those of WT mice. Especially in LN, CD25⫹ Foxp3⫹ T cells in SOCS3⫺/⫺ mice were 1.5 times more than those in WT mice (15.0 vs 10.9%). Similar results were obtained for both Tie2-Cre:SOCS3fl/fl and LysM-Cre:SOCS3fl/fl mice. Thus, our data suggest that Treg cells were expanded more efficiently at the periphery in SOCS3⫺/⫺ mice than in WT mice. Recently, increased number of Treg cells has been observed in septic spleens (21). Therefore, we next examined in vivo effect of LPS administration on Treg expansion in WT and SOCS3⫺/⫺ mice. Consistent with previous report, the faction of CD25⫹ Foxp3⫹ T cells was increased from 8.14 to 11.1% in the spleen and
The Journal of Immunology
FIGURE 4. Ag-specific Foxp3⫹ T cell expansion by SOCS3⫺/⫺ DCs. A, OVA-specific CD4⫹ T cells (2 ⫻ 105) were purified from OTII-TCR transgenic mice and cocultured with DCs (1 ⫻ 104) pulsed with OVA for 3 days. Expanded T cells (4 ⫻ 105) were restimulated with plate-bound anti-CD3 Ab. Proliferation (A) and Treg marker expression (B) were determined by [3H]thymidine incorporation and RT-PCR, respectively. Data presented as the mean values of triplicate cultures ⫹ SD (ⴱ, p ⬍ 0.01; ⴱⴱ, p ⬍ 0.05; Student’s t test). Similar results were obtained in two independent experiments. C, In vivo expansion of Foxp3⫹ T cells by Ag-loaded DCs. OVA peptide-loaded 4 ⫻ 106 DCs were injected i.v. into OTII transgenic mice. After 7 days, splenic CD4⫹ T cells were analyzed by Foxp3 intracellular staining and antiCD25 staining. D, The population of Foxp3 Treg cells in the spleen and mesenteric LNs in WT and SOCS3⫺/⫺ mice without or with LPS (10 g/head) administration. Sixteen hours later, spleen and LN cells were stained with anti-CD4, anti-CD25, and anti-Foxp3 Abs. One representative experiment of two is shown.
10.9 to 17.7% in the LN of WT mice. We noticed more CD25⫹ Foxp3⫹ T cells were present in SOCS3⫺/⫺ mice (12.8% in the spleen and 19.6% in the LN) compared with WT mice. These data further support our hypothesis that SOCS3-deficient DCs have a stronger potential to expand Foxp3⫹ T cells not only in vitro, but also in vivo. SOCS3⫺/⫺ BMDC suppresses T cell activation in vivo Next, to assess the initiation of T cell responses in vivo, SOCS3⫹/⫹ and SOCS3⫺/⫺ DCs were pulsed with KLH and then injected into each footpad of the same mouse. Swelling of the popliteal LNs was observed on the side of the SOCS3⫹/⫹ DCinjected footpad, but little LN swelling was observed on the side of the SOCS3⫺/⫺ DC-injected footpad 5 days after injection (Fig. 5A). KLH-induced IFN-␥ production from LN cells was lower in SOCS3⫺/⫺ DC-injected mice than WT-DC-injected
mice (Fig. 5B), suggesting that SOCS3⫺/⫺ DC is less immunogenic than WT-DC. Then we investigated the in vivo immunnosuppressive effect of SOCS3⫺/⫺ DCs on experimental autoimmune encephalomyelitis (EAE). MOG-peptide-pulsed DCs were i.v. injected before immunization of the mice. Control mice receiving SOCS3⫹/⫹ DCs (WT-DCs) as well as untreated mice exhibited characteristic signs of EAE starting on day 8. In contrast, mice receiving SOCS3⫺/⫺ DCs developed significantly less severe EAE, indicating that SOCS3⫺/⫺ DC is immunosuppressive in vivo (Fig. 5C). Then the nature of CD4⫹ T cells from mice with EAE was examined. The CD4⫹ T cells were isolated from the spleen of mice on day 15 and stimulated with anti-CD3 Ab. Proliferation and IFN-␥ production of restimulated T cells from SOCS3⫺/⫺ DCtreated mice were severely reduced (Fig. 5D). However, levels of IL-17, which is important for EAE development (22), were not
LOSS OF SOCS3 INDUCES TOLEROGENIC DC
FIGURE 5. Tolerogenic nature of SOCS3-deficient DCs. A and B, SOCS3⫺/⫺ DCs and WT-DCs (5 ⫻ 106) were cultured in the presence or absence of KLH for 16 h, and then injected into the footpad of C57BL/6 mice. A, The draining LNs were photographed on day 5 after injection of DCs. B, LN cells (1 ⫻ 106) were stimulated with KLH for 96 h, and the levels of IFN-␥ in the culture supernatant were measured. One representative of two independent experiments is shown (ⴱ, p ⬍ 0.05; Student’s t test). C, EAE was induced in C57BL/6 mice (n ⫽ 8 –12 for each group) with MOG peptide and CFA, and DCs (1 ⫻ 106) were administrated 7, 5, and 3 days before peptide/CFA immunization. Asterisks indicate significant differences (p ⬍ 0.05) compared with SOCS3⫹/⫹ DCs using Mann-Whitney U test. D, Splenic CD4⫹ T cells (2 ⫻ 105 cells) from DC-transferred mice 15 days after primary immunization were stimulated with plate-bound anti-CD3 Ab for 3 days. The proliferation and cytokine production of CD4⫹ T cells were determined (ⴱ, p ⬍ 0.05; Student’s t test).
significantly different in restimulated T cells. Therefore, suppression of EAE by SOCS3⫺/⫺ DCs was mainly due to reduced T cell activation. These data suggest a tolerogenic nature of SOCS3⫺/⫺ DCs in vivo. T cells expanded by SOCS3⫺/⫺ DCs have suppressor activity Then to examine a tolerogenic nature of SOCS3⫺/⫺ DCs, suppression assay was conducted using T cells expanded by DCs. Expanded T cells primed with WT-DCs or SOCS3⫺/⫺ DCs were cocultured with freshly isolated CD4⫹CD25⫺ BALB/c responder T cells and ␥-irradiated spleen cells as APCs. Proliferation of T cells (including not only responder T cells, but also DC-expanded T cells) in response to anti-CD3 Ab was measured (Fig. 6A). [3H]Thymidine incorporation was strongly enhanced when responder T cells were cocultured with T cells expanded by WT-DC because a majority of WT-DC-expanded T cells were Foxp3⫺ effector T cells (Fig. 6A, center). In contrast, T cells expanded by SOCS3⫺/⫺ DC marginally proliferated and suppressed proliferation of responder T cells (Fig. 6A, right). These data confirmed that SOCS3⫺/⫺ DCs predominantly expand Foxp3⫹ T cells and poorly expand Foxp3⫺ effector T cells (see Fig. 3D). Then, to compare the suppression activity, Treg cells were expanded from naive CD4⫹CD25⫹ T cells by coculturing with WT or SOCS3⫺/⫺ DCs. WT-DCs and SOCS3⫺/⫺ DCs showed similar expansion of CD4⫹CD25⫹ T cells (data not shown), and expanded Treg cells possessed similar suppression activity (Fig. 6B). Thus, Foxp3⫹ T cells expanded by SOCS3⫺/⫺ DCs can be taken as Treg cells, although we could not compare suppressor activity of
CD4⫹CD25⫹ T cells expanded by SOCS3⫺/⫺ DCs and naturally occurring Treg cells because former fraction contained 30 – 40% Foxp3⫺ effector T cells. Because SOCS3⫺/⫺ DCs predominantly expanded Foxp3⫹ Treg cells from CD4⫹ T cells, whereas WTDCs expanded Foxp3⫺ effector T cells, only CD25⫹ T cells expanded by SOCS3⫺/⫺ DCs could show suppression activity in vitro (Fig. 6A). This implies that SOCS3 is an important factor for determining immunity or tolerance in DC. Higher production of TGF-␤1 is important for higher Foxp3⫹ Treg expansion by SOCS3⫺/⫺ DCs TGF-␤1 and IL-2 have been implicated in the induction of Foxp3⫹ Treg cells by DCs in vitro (23–25). IL-2 has been shown to be secreted from DCs (26). To define the molecular basis for Foxp3⫹ Treg induction by SOCS3⫺/⫺ DCs, we examined the effect of Abs against cytokines. Anti-IFN-␥ Ab showed little effect on Foxp3⫹ Treg expansion by DCs (Fig. 7A). Anti-IL-2 Ab showed profound effect on Foxp3⫺ and Foxp3⫹ T cell expansion by WT-DCs or SOCS3⫺/⫺ DCs, suggesting an important role of IL-2 in effector T cell and Treg expansion (Fig. 7A). However, we could not find any differences in the IL-2 production between WT and SOCS3⫺/⫺ DCs (Fig. 7B). Thus, IL-2 from DCs may not be able to account for the differences between WT and SOCS3⫺/⫺ DCs. However, anti-TGF-␤1 Ab strongly enhanced Foxp3-negative effector T cell population and reduced Foxp3⫹ T cell population when cocultured with WT-DCs (Fig. 7A). Furthermore, antiTGF-␤1 Ab clearly suppressed Foxp3⫹ T cell expansion by coculture with SOCS3⫺/⫺ DCs (Fig. 7A). We have shown that
The Journal of Immunology
FIGURE 6. Suppressor activity of T cells primed with SOCS3-deficient DCs. A, CD25⫹ T cells (suppressors) were isolated from CD4⫹ T cells expanded by SOCS3⫹/⫹ or SOCS3⫺/⫺ DCs. These cells were further cocultured with CD4⫹CD25⫺ cells (3 ⫻ 104) from BALB/c mice (responders) and APCs in the presence of soluble anti-CD3 Ab (1 g/ml) for 72 h. Proliferation was assessed by [3H]thymidine uptake. In the left three columns, proliferation of suppressors and responders alone (3 ⫻ 104 cells) is shown. The numbers of responder and APCs were fixed, and the ratio of suppressor/responder is shown on the right of the panel. Similar results were obtained in two independent experiments. B, Suppressor activity by DC-expanded CD4⫹CD25⫹ T cells. CD4⫹CD25⫹ T cells were isolated from BALB/c mice and expanded by coculturing with SOCS3⫹/⫹ or SOCS3⫺/⫺ DCs. These cells were further cocultured with CD4⫹CD25⫺ cells (1 ⫻ 105) from BALB/c mice (responders) and allogeneic APCs from C57BL/6 for 72 h. In the left three columns, proliferation of suppressors and responders alone (1 ⫻ 105 cells) is shown as in the left columns (ⴱ, p ⬍ 0.01; Student’s t test, vs responder).
TGF-␤1 levels were higher in SOCS3-deficient T cells and hepatocytes than in normal cells (27). Therefore, we measured production of biologically active TGF-␤ by using a reporter cell line, MFB-F11 (18). MFB-F11 cells were stably transfected with a re-
porter plasmid consisting of TGF-␤-responsive Smad-binding elements coupled to a SEAP reporter gene. This cell line can detect a biologically active form of TGF-␤ (all three forms of TGF-␤) with extremely high sensitivity (detectable as little as 1 pg/ml
FIGURE 7. Essential role of TGF-␤1 in selective Foxp3⫹ T cell expansion by SOCS3⫺/⫺ DCs. A, Effect of anti-cytokine Abs. Allogeneic CD4⫹ T cells were stimulated with DCs in the presence of 10 g/ml indicated Abs for 4 days and analyzed by FACS. B, Production of IL-2 and TGF-␤1 from LPS-stimulated DCs for 24 h. IL-2 and TGF-␤1 levels were determined by ELISA and a bioassay on MFB-F11 cells, respectively. One representative experiment of three is shown (ⴱ, p ⬍ 0.05; Student’s t test). C, Total RNA (1 g) of DCs treated with LPS for 24 h was extracted, and the mRNA levels of indicated genes were determined by RT-PCR. IL-10 treatment was done 48 h before LPS stimulation (IL-10-DC). One representative experiment of three is shown.
LOSS OF SOCS3 INDUCES TOLEROGENIC DC
active TGF-␤1). As shown in Fig. 7B, WT-DCs secreted TGF-␤ at below detectable levels by MFB-F11. In contrast, SOCS3⫺/⫺ DCs secreted biologically active TGF-␤ at ⬃10 pg/ml, and the secretion was slightly enhanced by LPS treatment. Up-regulation of TGF-␤1 in SOCS3⫺/⫺ DCs was confirmed by RT-PCR (Fig. 7C). TGF-␤1 levels in IL-10-DC were as low as those in WT-DCs (Fig. 7C). These data suggest that higher expression of TGF-␤ in SOCS3⫺/⫺ DCs is one of an important mechanism for enhanced Foxp3⫹ T cell expansion by SOCS3⫺/⫺ DCs.
enhanced SOCS3 expression in DCs blocked the IL-12 and IL-23 responses, and that SOCS3-transduced DCs expressed a low level of MHC class II and CD86 on their surface, producing a high level of IL-10, but low levels of IL-12. Thus, SOCS3-overexpressing DCs resemble IL-10-DCs (30). It is still not very clear how SOCS3 in DCs regulates the induction of Foxp3⫹ Treg and Th2. Our data and these works indicate that the regulation of intracellular signaling pathways is extremely important for the decision of Th cell fates.
DCs have an important role in the control of the adaptive immune response. They simultaneously induce not only Ag-specific effector T cells, but also Treg cells (24). However, little is known how DCs balance induction of these functionally opposite T cells. In this study, we found that SOCS3 in BMDCs plays a critical role in the balance of Foxp3⫹ Treg cells and effector T cells. Although the mechanism is still not clear at present, predominant Foxp3⫹ Treg expansion by SOCS3⫺/⫺ DCs is striking. Enhanced Foxp3⫹ Treg expansion was observed not only in vitro, but also in vivo (Fig. 4). We described in this study that SOCS3-deficient DCs were tolerogenic probably because of selective expansion of Treg cells. TGF-␤1 has been shown to strongly enhance Foxp3⫹ Treg expansion from naive CD4⫹ T cell (8, 23). Our Ab-blocking experiments (Fig. 7A) confirmed that Foxp3⫹ Treg expansion by SOCS3⫺/⫺ DC is dependent on TGF-␤1. We showed that TGF-␤1 production was suppressed by expression of dominant-negative STAT3 and enhanced by STAT3c (27, 28). Furthermore, the TGF-␤1 promoter contains two potential STAT3 binding sites, and STAT3 enhanced TGF-␤1 promoter activity (28). These data suggest that STAT3 positively and SOCS3 negatively regulates the production of TGF-␤1 from DCs. However, the levels of TGF-␤1 in the culture supernatant of SOCS3⫺/⫺ DCs were as low as 10 pg/ml. This level was too low to induce Foxp3⫹ T cells from CD4⫹CD25⫺ naive T cells by anti-TCR stimulation in vitro (data not shown). Thus, we suspect that local TGF-␤1 activation at cellcell contact sites or cell surface-bound active TGF-␤1 is important for the induction of Foxp3⫹ T cells. In addition to higher TGF-␤1, other phenotypes of SOCS3⫺/⫺ DCs may be also necessary to selectively expand Foxp3⫹ T cells. We suspect that lower levels of class II MHC and costimulators in SOCS3⫺/⫺ DCs are also important for selective Foxp3⫹ Treg expansion, because effector T cells require higher amounts of antiTCR stimulation for proliferation than Treg cells (Y. Matsumura, unpublished data). However, because IL-10-DCs also have an immature phenotype, but did not strongly induce Foxp3⫹ Treg cells (Fig. 3D), the immature nature of DCs is not sufficient to explain the ability of SOCS3⫺/⫺ DCs. IL-10-DC did not express high levels of TGF-␤1 (Fig. 7C). Probably, both the immature phenotype of DCs and the high levels of TGF-␤1 are necessary for the predominant expansion of Foxp3⫹ Treg cells by SOCS3⫺/⫺ DCs. Effector T cell expansion was reduced because of lower levels of MHC and higher levels of TGF-␤1 in SOCS3⫺/⫺ DCs, whereas the expansion of Treg cells may not be so strongly affected by these. This idea is consistent with a recent paper showing a predominant induction of Foxp3⫹ cells by immature splenic DCs in the presence of small amount of TGF-␤1 (23). Recently, Li et al. (29) reported that DCs overexpressing SOCS3 exhibit a tolerogenic phenotype that directs Th2 differentiation and suppresses EAE. This situation resembles the phenotypes of T cells overexpressing or lacking SOCS3. We have shown that forced expression of SOCS3 in T cells promotes Th2 differentiation, whereas deletion of the SOCS3 gene in T cells induces TGF-␤1-secreting Th3 cells (28). Li et al. (29) demonstrated that
We thank M. Ohtsu, Y. Honda, and T. Yoshioka for technical assistance; Y. Nishi for manuscript preparation; and Dr. Y. Fukui and Dr. T. Hanada for discussion and comments.
Disclosures The authors have no financial conflict of interest.
References 1. Moser, M., and K. M. Murphy. 2000. Dendritic cell regulation of TH1-TH2 development. Nat. Immunol. 1: 199 –205. 2. Steinman, R. M., S. Turley, I. Mellman, and K. Inaba. 2000. The induction of tolerance by dendritic cells that have captured apoptotic cells. J. Exp. Med. 191: 411– 416. 3. Jonuleit, H., E. Schmitt, G. Schuler, J. Knop, and A. H. Enk. 2000. Induction of interleukin 10-producing, nonproliferating CD4⫹ T cells with regulatory properties by repetitive stimulation with allogeneic immature human dendritic cells. J. Exp. Med. 192: 1213–1222. 4. Akbari, O., R. H. DeKruyff, and D. T. Umetsu. 2001. Pulmonary dendritic cells producing IL-10 mediate tolerance induced by respiratory exposure to antigen. Nat. Immunol. 2: 725–731. 5. Gonzalez-Rey, E., A. Chorny, A. Fernandez-Martin, D. Ganea, and M. Delgado. 2006. Vasoactive intestinal peptide generates human tolerogenic dendritic cells that induce CD4 and CD8 regulatory T cells. Blood 107: 3632–3638. 6. Dhodapkar, M. V., R. M. Steinman, J. Krasovsky, C. Munz, and N. Bhardwaj. 2001. Antigen-specific inhibition of effector T cell function in humans after injection of immature dendritic cells. J. Exp. Med. 193: 233–238. 7. Tarbell, K. V., S. Yamazaki, K. Olson, P. Toy, and R. M. Steinman. 2004. CD25⫹ CD4⫹ T cells, expanded with dendritic cells presenting a single autoantigenic peptide, suppress autoimmune diabetes. J. Exp. Med. 199: 1467–1477. 8. Chen, W., W. Jin, N. Hardegen, K. J. Lei, L. Li, N. Marinos, G. McGrady, and S. M. Wahl. 2003. Conversion of peripheral CD4⫹CD25⫺ naive T cells to CD4⫹CD25⫹ regulatory T cells by TGF-␤ induction of transcription factor Foxp3. J. Exp. Med. 198: 1875–1886. 9. Takeda, K., B. E. Clausen, T. Kaisho, T. Tsujimura, N. Terada, I. Forster, and S. Akira. 1999. Enhanced Th1 activity and development of chronic enterocolitis in mice devoid of Stat3 in macrophages and neutrophils. Immunity 10: 39 – 49. 10. Steinbrink, K., E. Graulich, S. Kubsch, J. Knop, and A. H. Enk. 2002. CD4⫹ and CD8⫹ anergic T cells induced by interleukin-10-treated human dendritic cells display antigen-specific suppressor activity. Blood 99: 2468 –2476. 11. Cheng, F., H. W. Wang, A. Cuenca, M. Huang, T. Ghansah, J. Brayer, W. G. Kerr, K. Takeda, S. Akira, S. P. Schoenberger, et al. 2003. A critical role for Stat3 signaling in immune tolerance. Immunity 19: 425– 436. 12. Yoshimura, A. 2006. Signal transduction of inflammatory cytokines and tumor development. Cancer Sci. 97: 439 – 447. 13. Yasukawa, H., M. Ohishi, H. Mori, M. Murakami, T. Chinen, D. Aki, T. Hanada, K. Takeda, S. Akira, M. Hoshijima, et al. 2003. IL-6 induces an antiinflammatory response in the absence of SOCS3 in macrophages. Nat. Immunol. 4: 551–556. 14. Kimura, A., I. Kinjyo, Y. Matsumura, H. Mori, R. Mashima, M. Harada, K. R. Chien, H. Yasukawa, and A. Yoshimura. 2004. SOCS3 is a physiological negative regulator for granulopoiesis and granulocyte colony-stimulating factor receptor signaling. J. Biol. Chem. 279: 6905– 6910. 15. Hanada, T., K. Tanaka, Y. Matsumura, M. Yamauchi, H. Nishinakamura, H. Aburatani, R. Mashima, M. Kubo, T. Kobayashi, and A. Yoshimura. 2005. Induction of hyper Th1 cell-type immune responses by dendritic cells lacking the suppressor of cytokine signaling-1 gene. J. Immunol. 174: 4325– 4332. 16. Aki, D., R. Mashima, K. Saeki, Y. Minoda, M. Yamauchi, and A. Yoshimura. 2005. Modulation of TLR signalling by the C-terminal Src kinase (Csk) in macrophages. Genes Cells 10: 357–368. 17. Kojo, S., K. Seino, M. Harada, H. Watarai, H. Wakao, T. Uchida, T. Nakayama, and M. Taniguchi. 2005. Induction of regulatory properties in dendritic cells by V␣14 NKT cells. J. Immunol. 175: 3648 –3655. 18. Tesseur, I., K. Zou, E. Berber, H. Zhang, and T. Wyss-Coray. 2006. Highly sensitive and specific bioassay for measuring bioactive TGF-␤. BMC Cell Biol. 7: 15. 19. Roncarolo, M. G., S. Gregori, M. Battaglia, R. Bacchetta, K. Fleischhauer, and M. K. Levings. 2006. Interleukin-10-secreting type 1 regulatory T cells in rodents and humans. Immunol. Rev. 212: 28 –50.
The Journal of Immunology 20. De Smedt, T., B. Pajak, E. Muraille, L. Lespagnard, E. Heinen, P. De Baetselier, J. Urbain, O. Leo, and M. Moser. 1996. Regulation of dendritic cell numbers and maturation by lipopolysaccharide in vivo. J. Exp. Med. 184: 1413. 21. Scumpia, P. O., M. J. Delano, K. M. Kelly, K. A. O’Malley, P. A. Efron, P. F. McAuliffe, T. Brusko, R. Ungaro, T. Barker, J. L. Wynn, et al. 2006. Increased natural CD4⫹CD25⫹ regulatory T cells and their suppressor activity do not contribute to mortality in murine polymicrobial sepsis. J. Immunol. 177: 7943–7949. 22. Chen, Y., C. L. Langrish, B. McKenzie, B. Joyce-Shaikh, J. S. Stumhofer, T. McClanahan, W. Blumenschein, T. Churakovsa, J. Low, L. Presta, et al. 2006. Anti-IL-23 therapy inhibits multiple inflammatory pathways and ameliorates autoimmune encephalomyelitis. J. Clin. Invest. 116: 1317–1326. 23. Luo, X., K. V. Tarbell, H. Yang, K. Pothoven, S. L. Bailey, R. Ding, R. M. Steinman, and M. Suthanthiran. 2007. Dendritic cells with TGF-␤1 differentiate naive CD4⫹CD25⫺ T cells into islet-protective Foxp3⫹ regulatory T cells. Proc. Natl. Acad. Sci. USA 104: 2821–2826. 24. Yamazaki, S., M. Patel, A. Harper, A. Bonito, H. Fukuyama, M. Pack, K. V. Tarbell, M. Talmor, J. V. Ravetch, K. Inaba, and R. M. Steinman. 2006. Effective expansion of alloantigen-specific Foxp3⫹ CD25⫹ CD4⫹ regulatory T cells by dendritic cells during the mixed leukocyte reaction. Proc. Natl. Acad. Sci. USA 103: 2758 –2763. 25. Banerjee, D. K., M. V. Dhodapkar, E. Matayeva, R. M. Steinman, and K. M. Dhodapkar. 2006. Expansion of FOXP3high regulatory T cells by human
dendritic cells (DCs) in vitro and after injection of cytokine-matured DCs in myeloma patients. Blood 108: 2655–2661. Granucci, F., C. Vizzardelli, N. Pavelka, S. Feau, M. Persico, E. Virzi, M. Rescigno, G. Moro, and P. Ricciardi-Castagnoli. 2001. Inducible IL-2 production by dendritic cells revealed by global gene expression analysis. Nat. Immunol. 2: 882– 888. Ogata, H., T. Chinen, T. Yoshida, I. Kinjyo, G. Takaesu, H. Shiraishi, M. Iida, T. Kobayashi, and A. Yoshimura. 2006. Loss of SOCS3 in the liver promotes fibrosis by enhancing STAT3-mediated TGF-␤1 production. Oncogene 25: 2520 –2530. Kinjyo, I., H. Inoue, S. Hamano, S. Fukuyama, T. Yoshimura, K. Koga, H. Takaki, K. Himeno, G. Takaesu, T. Kobayashi, and A. Yoshimura. 2006. Loss of SOCS3 in T helper cells resulted in reduced immune responses and hyperproduction of interleukin 10 and transforming growth factor-␤1. J. Exp. Med. 203: 1021–1031. Li, Y., N. Chu, A. Rostami, and G. X. Zhang. 2006. Dendritic cells transduced with SOCS-3 exhibit a tolerogenic/DC2 phenotype that directs type 2 Th cell differentiation in vitro and in vivo. J. Immunol. 177: 1679 –1688. Seki, Y., H. Inoue, N. Nagata, K. Hayashi, S. Fukuyama, K. Matsumoto, O. Komine, S. Hamano, K. Himeno, K. Inagaki-Ohara, et al. 2003. SOCS-3 regulates onset and maintenance of TH2-mediated allergic responses. Nat. Med. 9: 1047–1054.