The Active Form of Vitamin D Transcriptionally Represses Smad7 Signaling and Activates Extracellular Signal-regulated Kinase (ERK) to Inhibit the Differentiation of a Inflammatory T Helper Cell Subset and Suppress Experimental Autoimmune Encephalomyelitis*

Background: Transcriptional regulation of Smads that modulate T helper (Th) cell differentiation is not well understood. Results: Active form of vitamin D (1,25(OH)2D3) leads to VDR-RXR-Smad3-HDAC2 repressive complex on VDRE-Smad7 promoter. 1,25(OH)2D3 activates ERK. Conclusion: 1,25(OH)2D3-VDR represses Smad7 and activates ERK leading to inhibition of inflammatory T cells and EAE. Significance: TGFβ Smad and non-Smad MAPK are involved in 1,25(OH)2D3-VDR mediated inhibition of EAE. The ability of the active form of vitamin D, 1,25-dihydroxyvitamin D3 (1,25(OH)2D3), to transcriptionally modulate Smads to inhibit Th17 differentiation and experimental autoimmune encephalomyelitis (EAE) has not been adequately studied. This study reports modulation of Smad signaling by the specific binding of the VDR along with its heterodimeric partner RXR to the negative vitamin D response element on the promoter of Smad7, which leads to Smad7 gene repression. The vitamin D receptor-mediated increase in Smad3 expression partially explains the IL10 augmentation seen in Th17 cells. Furthermore, the VDR axis also modulates non-Smad signaling by activating ERK during differentiation of Th17 cells, which inhibits the Th17-specific genes il17a, il17f, il22, and il23r. In vivo EAE experiments revealed that, 1,25(OH)2D3 suppression of EAE correlates with the Smad7 expression in the spleen and lymph nodes. Furthermore, Smad7 expression also correlates well with IL17 and IFNγ expression in CNS infiltered inflammatory T cells. We also observed similar gene repression of Smad7 in in vitro differentiated Th1 cells when cultured in presence of 1,25(OH)2D3. The above canonical and non-canonical pathways in part address the ability of 1,25(OH)2D3-VDR to inhibit EAE.


The ability of the active form of vitamin D, 1,25-dihydroxyvitamin D 3 (1,25(OH) 2 D 3 ), to transcriptionally modulate Smads to inhibit Th17 differentiation and experimental autoimmune encephalomyelitis (EAE) has not been adequately studied. This study reports modulation of Smad signaling by the specific bind-
ing of the VDR along with its heterodimeric partner RXR to the negative vitamin D response element on the promoter of Smad7, which leads to Smad7 gene repression. The vitamin D receptormediated increase in Smad3 expression partially explains the IL10 augmentation seen in Th17 cells. Furthermore, the VDR axis also modulates non-Smad signaling by activating ERK during differentiation of Th17 cells, which inhibits the Th17-specific genes il17a, il17f, il22, and il23r. In vivo EAE experiments revealed that, 1,25(OH) 2 D 3 suppression of EAE correlates with the Smad7 expression in the spleen and lymph nodes. Furthermore, Smad7 expression also correlates well with IL17 and IFN␥ expression in CNS infiltered inflammatory T cells. We also observed similar gene repression of Smad7 in in vitro differentiated Th1 cells when cultured in presence of 1,25(OH) 2

D 3. The above canonical and non-canonical pathways in part address the ability of 1,25(OH) 2 D 3 -VDR to inhibit EAE.
Smad and non-Smad signaling pathways are crucial for relaying the effects of transforming growth factor ␤ (TGF-␤), a pleiotropic cytokine that modulates the growth, differentiation, apoptosis, and state of activation of diverse immune cells. In T helper cells (Th cells), 2 TGF-␤ plays a crucial role in maintain-ing the balance between inflammatory and immunosuppressive Th cell phenotypes and thereby modulates the progression of autoimmune diseases (1)(2)(3). Upon activation by TGF-␤, receptor-regulated Smads (R-Smads), Smad2 and Smad3, form a heteromeric complex with co-Smad (Smad4) and translocate into the nucleus where the complex regulates the transcription of target genes by recruiting other co-regulators (2). On the other hand, inhibitory Smads (I-Smads), Smad6 and Smad7, which are also transcriptionally induced by TGF-␤, negatively regulate TGF-␤ signaling in a feedback regulatory mechanism (4). The Smad-independent signaling pathway of TGF-␤ involves modulation of protein kinases, including extracellular-signalregulated kinases (ERK), c-Jun N-terminal kinase (JNK), p38, phosphoinositide 3 kinase (PI3K), and Rho-associated kinase (ROCK) (5)(6)(7)(8). The discrete and selective regulation of TGF-␤mediated Smad and non-Smad pathways may explain its differential requirement for the differentiation of functionally paradoxical T cell types. Smad3 has been shown to be crucial for Treg cells, whereas Smad7 has been shown to be associated with inflammatory T cell phenotype (9 -11). However, the transcriptional switches that contribute to this distinct signaling have remained elusive. Previous reports have shown that several nuclear receptors (NRs), such as the vitamin D receptor (VDR), androgen receptor, glucocorticoid receptor, estrogen receptor (ER␣), and hepatocyte nuclear factor 4␣ (HNF4␣), are able to bind to Smad proteins (12)(13)(14), which suggests the possibility that NRs participate in TGF-␤ signaling.
VDR is a member of the steroid receptor family and has 1,25dihydroxyvitamin D 3 (1,25(OH) 2 D 3 ), the active form of vitamin D, as its ligand. VDR has a propensity to bind to and regulate gene promoters with DR3-or ER6-type steroid response ele-ments with the consensus half-site of PuGGTCA (15). VDR has been reported to modulate Smad signaling largely by inducing Smad3 phosphorylation and activity (16,17). Smad3 in turn modulates VDR biological activity through its physical interaction and functions as a ligand-dependent co-regulator, an interaction inhibited by Smad7 (16). VDR has also been reported to modulate non-Smad MAPK signaling cascades in a cell-specific manner (18 -20). 1,25(OH) 2 D 3 is known to inhibit experimental autoimmune encephalomyelitis (EAE); however, its ability to modulate Smad and non-Smad pathways in Th cells has not been adequately addressed (21,22).
In this study we investigated the role of the 1,25(OH) 2 D 3 -VDR role in the modulation of these pathways. This study shows that the 1,25(OH) 2 D 3 -VDR-RXR heterodimer binds directly to the negative vitamin D response element (VDRE) in the promoter of Smad7 and represses its expression in Th cells. 1,25(OH) 2 D 3 also activates ERK in Th17 cells. In addition, we show that these two functional activities lead to induction of IL-10 and repression of Th17-specific genes and explains inhibition of EAE by 1,25(OH) 2 D 3 -VDR axis.
Mice and Ethics Statement-Male inbred C3He mice and female C57BL/6 mice were bred in the animal facility of the Institute of Microbial Technology (IMTECH). Mice were fed with LabDiet-5W11 (TestDiet), which is devoid of vitamin D and retinoic acid. Experiments with mice were approved by the Institutional Animal Ethics Committee of the Institute of Microbial Technology and performed according to the National Regulatory Guidelines issued by the Committee for the Purpose of Supervision of Experiments on Animals (no. 55/1999/CPCSEA), Ministry of Environment and Forest, Government of India.
Naïve T Cell Purification and Differentiation-Naïve CD4 ϩ CD62L ϩ CD44 Ϫ T lymphocytes were isolated from the lymphoid organs of 4-6-week-old mice by using the IMag TM Mouse CD4 T Lymphocyte Enrichment Set supplemented with biotin-tagged anti-mouse CD44 antibody. In vitro differentiation was performed as mentioned elsewhere (23). Cells were plated at a density of 2.5 ϫ 10 5 cells per well in a 24-well plate. The cells were stimulated with 5 g/ml concentrations of platebound anti-CD3 and 2 g/ml soluble anti-CD28 antibodies. For differentiation into specific subtypes, these naïve cells were provided with different cytokines for 3-4 days and after every 2 days cells were replaced with fresh medium along with the cytokines and 1,25(OH) 2 D 3 . For Th0, cells were given 100 units of IL-2 along with 5 g/ml concentrations each of anti-IL-4 and anti-IFN␥ antibodies. For differentiation into Tregs, cells were provided with 5 ng/ml TGF-␤ and 100 units of IL-2. For differentiation into Th17 cells, cells were provided with 2.5 ng/ml of TGF-␤ and 20 ng/ml of IL-6 along with 5 g/ml each of anti-IL-4 and anti-IFN␥ antibodies. 100 nM 1,25(OH) 2 D 3 was added to the culture at 0 h or as indicated. SIS3 (10 M) and U0126 (5 M) were added to cultures just 1 h before the T cell receptor stimulation.
Plasmids and Expression-Mouse VDR and RXR cDNA were purchased from Open Biosystems and cloned into mammalian expression vector pFLAG. For silencing adenoviral plasmid pAd/BLOCK-iT-DEST carrying short hairpin RNA (shRNA) specific for VDR or shRNA for LacZ was constructed according to the manufacturer's instructions (Invitrogen), and replication-deficient adenovirus particles were generated as described previously (24). The silencing efficiency of the constructs was monitored by RT-PCR/Western analysis.
Luciferase Assay-The mouse Smad7 promoter (Ϫ1150 to ϩ170) was cloned into pGL3 vector with Firefly luciferase, which was purchased from Promega. The luciferase assay was performed as described (24). COS-1 cells were plated in 24-well plates and grown to 70% confluence. Just before transfection media was changed to Opti-MEM and pFLAG-mVDR, pFLAG-mRXR plasmids were transfected into COS-1 cells along with the pGL3-promoter plasmid. In another experiment, pGL3promoter plasmid was transfected into EL-4 cells. In both experiments pBIND having Renilla luciferase was also transfected and used as transfection control. pGL3-Smad7 Firefly luciferase activity was normalized against Renilla luciferase, and the normalized luciferase activities (relative light units) were plotted as the average (ϮS.D.) of triplicate samples from typical experiments (SigmaPlot).
Real Time Quantitative RT-PCR (qPCR)-Total RNA was isolated from in vitro differentiated T cells or mononuclear cells isolated from the mice brain by the TRIzol method, and 1 g of total RNA was reverse-transcribed with the Verso cDNA Synthesis kit (Thermo Scientific) according to manufacturer's instructions. cDNA was amplified with gene-specific primers using the DyNAmo ColorFlash SYBR Green R kit (Thermo Scientific) and 18 S mRNA as a control. The relative -fold change was calculated by using the formula 2 Ϫ⌬⌬Ct .
Western Blot-Cell lysates were prepared by treating cells with cell lysis buffer; protein levels in the samples were estimated in a Bio-Rad Protein Assay (Bio-Rad). For immunoblotting, 30 g of total protein extract was resolved by 10% SDS-PAGE and transferred to polyvinylidene difluoride membranes (Immobilon-P; Millipore). Membranes were blocked with 5% skim milk and incubated overnight with primary antibodies followed by incubation with HRP-conjugated secondary antibodies. The blots were developed with Luminata Forte Western HRP substrate (Millipore).
Flow Cytometry and Enzyme-linked Immunosorbent Assay (ELISA)-For flow cytometry all antibodies were used according to manufacturer instructions and titrated as required. Staining of cell surface markers CD4, CD44, and CD62L was performed as described previously (25). Intracellular and nuclear staining was performed for IL17A, IFN␥, and Foxp3 according to manufacturer instructions (Intracellular Fixation and Permeabilization Buffer Set, Foxp3/Transcription Factor Buffer Set; eBioscience). All samples were acquired by gating on CD4 cells. Samples were acquired on BD FACSAria and analyzed by using FlowJo software. ELISA for IL10 (BD Bioscience) was used to analyze supernatants from 3-4-day cultures.
Electrophoretic Mobility Shift Assay (EMSA)-5Ј-End labeling of annealed oligonucleotides was performed with T4 polynucleotide kinase (New England Biolabs) and [␥-32 P]ATP. pFLAG-mVDR and pFLAG-mRXR plasmids were transcribed and translated using the TNT Quick Coupled Transcription Translation System according to the manufacturer's instructions. DNA binding to recombinant proteins was assessed using minimal amounts of probe as described earlier (25). For competition experiments, a molar excess of unlabeled oligonucleotide was added to compete out the DNA binding. Gel shift immunoassays were performed with 0.4 g of anti-VDR antibody (Thermo Scientific) or 0.4 g of anti-RXR antibody (Santa Cruz Biotechnology). The oligonucleotide sequence corresponding to the site Ϫ1031 to Ϫ1046 on Smad7 promoter (5Ј-GAAGG-AATAAACTCTCTGACCCTAAGCAGG-3Ј and 5Ј-CCTGC-TTAGGGTCAGAGAGTTTATTCCTTC-3Ј) were used as a test. Mouse osteopontin oligonucleotides 5Ј-GATCCACAAG-GTTCACGAGGTTCACGTCTG-3Ј and 5Ј-CAGACGTGAA-CCTCGTGAACCTTGTGGATC-3Ј were used as positive control. Negative control sequences were also attempted to rule out nonspecific binding. Mutant oligonucleotide 5Ј-GAAGG-AATAAACTCTCTGAAAATAAGCAGG-3Ј and 5Ј-CCTGC-TTATTTTCAGAGAGTTTATTCCTTC-3Ј were also used for performing EMSA.
Experimental Autoimmune Encephalitis Generation-EAE was induced in 8-week-old C57BL/6 female mice by subcutaneous injection at 2 sites with 100 l of an emulsion containing 100 g of myelin oligodendrocyte glycoprotein (MOG) peptide (amino acids  in complete Freund's adjuvant. 200 ng of pertussis toxin was injected intraperitoneally on the day of immunization and 2 days thereafter. From day 4, mice received on alternate days either 300 ng of 1,25(OH) 2 D 3 in soybean oil or vehicle (control) intraperitoneally until the disease progression reached its peak. The mice were scored daily for clinical signs of EAE on the basis of the following scale: 0 ϭ no clinical sign, 1 ϭ loss of tail tonicity, 2 ϭ flaccid tail, 3 ϭ partial hind limb paralysis, 4 ϭ total hind limb paralysis, and 5 ϭ fore and hind limb paralysis. Spleen and lymph nodes were collected from mice, and single cell suspensions were prepared. These cells were re-stimulated with 20 g of MOG peptide before intracellular staining. Mono nuclear cells were isolated from brain by Percoll gradient (30%:70%) as mentioned in previous literature (27).
Statistics-Results are expressed as the mean Ϯ S.D. unless otherwise mentioned. SigmaPlot was used for statistical analysis. Two-tailed Student's t tests were performed to obtain p values. Statistical significance was established at p Ͻ 0.05 (*).

RESULTS
Th17 Expresses Smad7-Although previous studies have highlighted the significant role played by Smad proteins in Treg-and Th17-cell development (11,23,28,29), the related expression of Smads in these cells has not been studied. In this work we examined the relative expression of effector Smads in Treg and Th17 cells differentiated from naïve CD4 T lymphocytes isolated from mice, and real time qPCR analyses were performed to determine the expression levels of Smad2, Smad3, Smad4, and Smad7 in Treg and Th17 cells compared with those in naïve CD4 ϩ T lymphocytes cultured in the presence of IL-2 (Th0) (Fig. 1A). Compared with Treg cells, Th17 cells expressed significantly less Smad3, although the expression of Smad7 was significantly elevated in Th17 cells. The significant expression of Smad3 in Tregs and of Smad7 in Th17 suggests of a functional role of these Smad proteins in these cell types.
1,25(OH) 2 D 3 Promotes the Expression of Smad3 and Inhibits the Expression of Smad7 during Differentiation of Th17 Cells-1,25(OH) 2 D 3 has been shown to inhibit Th17 cells differentiation, but the mechanism is not fully understood (30). 1, 25(OH) 2 D 3 has been reported to suppress Th17 function by restricting the expression of IL17 (31). Here we examined the possible participation of Smad proteins in 1,25(OH) 2 D 3mediated inhibition of Th17 cell differentiation. We used qPCR to investigate the effect of 1,25(OH) 2 D 3 on the expression of Smads in naïve T cells cultured in the presence of TGF-␤ and IL6 with or without 1,25(OH) 2 D 3 . Treatment with 1, 25(OH) 2 D 3 induces the expression of Smad3 while at the same time suppressing the expression of Smad7 in these cells (Fig.  1B). This treatment also induced the expression of VDR, Foxp3, and suppressed the expression of il17a (Fig. 1C). Furthermore, we evaluated the role of VDR in the 1,25(OH) 2 D 3mediated modulation of Smad3 and Smad7 expression in a VDR-knockdown background (Fig. 1D). Interestingly, 1, 25(OH) 2 D 3 modulation of these Smads was abrogated in the knockdown background of VDR. Hence, it is clear that 1,25(OH) 2 D 3 modulates Smad3 and Smad7 gene expression via VDR (Fig. 1D). A similar pattern was also observed for Smad3, p-Smad3, and Smad7 protein levels (Fig. 1E).
VDR Forms a Repressive Complex on the Smad7 Promoter-To elucidate whether VDR modulates Smad expression through a genomic mechanism, a bioinformatics search for the presence of a VDRE was performed on the proximal promoters of Smad3 and Smad7 by using NubiScan and also validated manually (32). We found a putative VDRE at position Ϫ1031 to Ϫ1046 from the transcription start site on the minus strand of Smad7 promoter; however, we could not identify any VDRE on the Smad3 promoter ( Fig. 2A) (33,34). The VDRE on Smad7 was the DR3 type with one consensus and one non-consensus half-site. This type of VDRE has been reported to assist VDR in recruiting a co-repressor and leading to gene repression. Hence, the predicted VDRE found on Smad7 could potentially be a negative response element (15,35).
A ChIP experiment was performed in 1,25(OH) 2 D 3 -treated EL4 cells to determine if VDR and its heterodimerization partner RXR could occupy the VDRE on the Smad7 promoter (Fig.  2B). The ChIP and re-ChIP results suggest that both VDR and RXR were recruited to the Smad7 promoter, and this recruitment was 1,25(OH) 2 D 3 -dependent (Fig. 2B). Re-ChIP experiments were also performed to identify the repressive partner of this complex, and HDAC2 was identified as the part of this complex (Fig. 2B). We also performed qPCR by using Smad7 promoter-specific primers to obtain a quantitative measure of  (60). B, ChIP assays were performed with EL4 cells cultured in the presence or absence of 1,25(OH) 2 D 3 . Immunoprecipitation was performed with antibodies to VDR, RXR, and HDAC2. DNA eluted after the IP was amplified using Smad7 VDRE promoter-specific primers. GAPDH promoter-specific primers were used as a control. C, VDR recruitment on the Smad7 promoter was analyzed by qPCR and represented as percent of input. D-F, EMSA was performed by incubating in vitro translated VDR and RXR proteins with end-labeled Smad7, and control osteopontin oligonucleotides and complex formation was monitored. D, additionally, antibody against VDR or RXR was added to confirm the heterodimer formation. the enrichment of VDR upon 1,25(OH) 2 D 3 treatment (Fig. 2C). A 3-fold enrichment of VDR on the promoter was observed upon 1,25(OH) 2 D 3 treatment. EMSA was performed to determine if the VDR-RXR heterodimer regulates Smad7 by directly binding to the regulatory motifs of its promoter. To monitor the DNA-protein complex formation, a radiolabeled oligonucleotide containing the sequence for VDRE was incubated with recombinant VDR and RXR proteins. The VDR heterodimer appeared to form a stable complex with the Smad7 oligonucleotide as indicated by reduced electrophoretic mobility (Fig.  2D). The presence of heterodimer was confirmed by the addition of antibodies to VDR and RXR, which reduced the intensity of the shifted bands possibly due to the interference of the antibody with the protein-DNA binding. Interference of antibody in DNA-protein complex formation has also been reported earlier (25). An osteopontin oligonucleotide was used as a control because that gene has been reported to be directly regulated by the VDR heterodimer. Furthermore, to validate the specificity of VDR-RXR binding, a competition assay was performed (Fig.  2E). In the presence of unlabeled putative oligonucleotide the binding of heterodimer to Smad7-labeled oligonucleotide was diminished in a dose-dependent manner. To verify the nucleotide specificity, mutant probe was prepared. Mutation in the consensus half-site completely abolished the binding (Fig. 2F).
Additionally we cloned the 1320 bp of promoter region that is upstream to transcription start site of Smad7 in pGL3-basic vector, a Firefly luciferase reporter plasmid (Fig. 3A). First, a luciferase assay in COS-1 cells was performed, which is an excellent system for reconstitution experiments by ectopically expressing pFLAG-mVDR and pFLAG-mRXR along with pGL3-Smad7 gene-promoter reporter. Compared with the activity in control plasmids, the basal activity of this Smad7 promoter was reduced in COS-1 cells when plasmids containing the sequences for VDR and RXR were co-transfected. The promoter activity was further significantly reduced by the addition of 1,25(OH) 2 D 3 (Fig. 3A). We also observed a 1,25(OH) 2 D 3 dose-dependent (10 -100 nM) suppression of Smad7 promoter activity in a luciferase reporter assay performed in EL4 cells that endogenously express VDR and RXR and so were transfected with pGL3-Smad7 gene-promoter plasmid only (Fig. 3B). Using EL4 cells here is of more physiological relevance and is in keeping with the fact that EL4 cells express ROR␥t constitutively (36,37).
Smad3 is known to interact with VDR and promotes its function (16). To determine the possible role of Smad3 in 1,25(OH) 2 D 3 -VDR-mediated suppression of Smad7, we have performed a pGL3-Smad7 gene-promoter reporter assay in which EL4 cells were cultured in the presence of 1,25(OH) 2 D 3 alone or together with the SIS3 (23,38). The presence of SIS3 relieved the repression of 1,25(OH) 2 D 3 to an extent (Fig. 3C). Later we conducted ChIP and re-ChIP experiments in EL4 cells with VDR and p-Smad3 antibodies. The results suggests that Smad3 interact with the VDR and forms a complex that is recruited on the VDRE of the Smad7 promoter (Fig. 3D). We additionally performed a ChIP experiment to explain how Smad3 inhibits the transcription of Smad7 in the presence of 1,25(OH) 2 D 3 -VDR although it induces the expression of Smad7 in the presence of TGF␤. In this ChIP assay we used three sets of primers. Set 1 specifically amplifies Smad7 VDRE region, set 2 amplifies to Smad7 SBE, and set 3 spans the BamHI site, which lies in between the VDRE and SBE (Fig. 3E). In this site-specific ChIP experiment, EL4 cells were cultured in the presence of TGF␤ and IL6 or TGF␤ and IL6 along with 1,25(OH) 2 D 3 . After fragmentation of chromatin by sonication, we further digested it with BamHI enzyme and then performed immunoprecipitation with the combinations of p-Smad3 and VDR antibodies. Results clearly suggesting that in the presence of 1,25(OH) 2 D 3 p-Smad3 was recruited to VDRE only, whereas SBE is unoccupied. BamHI digestion was verified by using set 3 primers. qPCR was performed to investigate the effect of SIS3, on 1,25(OH) 2 D 3 -mediated Smad7 repression in EL4 cells (Fig.  3F) and in differentiating Th17 cells (Fig. 3G), which further suggests the role of Smad3 in VDR-mediated repression of Smad7 expression. We also confirmed the formation of same repressive complex in differentiating Th17 cells in the presence of 1,25(OH) 2 D 3 by ChIP and re-ChIP assays (Fig. 3H). 2 D 3 has been implicated in the modulation of MAPKs (ERK, p38, and JNK) by a nongenomic mechanism with rapid or late response, and it has been understood to be a cell-specific phenomenon (18 -20). Cell membrane and cytoplasmic localization of VDR has been shown to be vital for relaying the non-genomic functions of 1,25(OH) 2 D 3 in osteoblast, keratinocytes, and cancer cells (18,20,39,40). Although rapid responses of 1,25(OH) 2 D 3 were mainly executed by membrane VDR, it is cytoplasmic VDR that carries out late responses (18,40). To investigate the localization of VDR in T cells, we performed VDR surface staining on total CD4 T cells but failed to find any significant VDR surface expression; however, significant amounts of VDR were present intracellularly (data not shown). The role of MAPKs in Treg and Th17 cell development has been elucidated recently. Even though the role of ERK in Th17 cell differentiation is equivocal, its activation has been shown to promote Treg-cell function; p38 has been shown to be crucial for Th17, and JNK plays an intermediate role in both cell types (23,41). However, the role of 1,25(OH) 2 D 3 -VDR in the regulation of MAPKs in these cells is not known.

1,25(OH) 2 D 3 Modulates Non-Smad MAPK Pathways in Differentiating Th17 Cells-1,25(OH)
To determine if 1,25(OH) 2 D 3 modulates MAPKs during differentiation of Th17 cells, we cultured naïve T cells in Th17 conditions in the presence of 1,25(OH) 2 D 3 or the vehicle (control) and analyzed ERK, p38, and JNK activation by Western blot analysis (Fig. 4, A-C). 1,25(OH) 2 D 3 induces ERK activation during differentiation of Th17 cells; however, a short treatment of 30 min at the terminal stage of Th17 differentiation did not show any significant activation, which further confirms the absence of surface VDR (Fig. 4A). 1,25(OH) 2 D 3 treatment of differentiating Th17 cells slightly inhibited the activation of stress-activated protein kinases p38, although it did not show any significant effect on JNK activation (Fig. 4, B and C). 1,25(OH) 2 D 3 dose-dependent activation of ERK was also observed (Fig. 4D). Quantitative representation of blots has also been shown.

1,25(OH) 2 D 3 Abrogates the Expression of Smad7 and Induces
Activation of ERK in CD4 Cells from EAE Mice-Smad7 expression has been associated with inflammatory phenotype and Smad7Tg mice (over-expressing Smad7) and has been reported for elevated expression of Th17-specific transcription factors (42). Additionally, Smad7 induction has been observed in multiple sclerosis patients (42). ERK inhibition has also been shown to have deleterious effects in inflammatory diseases (10). As we observed the suppression of Smad7 and ERK activation by 1,25(OH) 2 D 3 -VDR in in vitro differentiated Th17 cells, we extended our results to an in vivo setting. We generated EAE in C57BL/6 mice by using the MOG peptide. All mice were fed with diets devoid in vitamin D and retinoic acid from 1 week before of the start of experiment and maintained throughout the experiment. At the first clinical score after immunization with pertussis toxin, the mice were divided into two groups. On alternate days one group was injected intraperitoneally with FIGURE 3. VDR suppresses Smad7 promoter activity and Smad3 also participates in this repression by forming a complex with VDR. A, pictorial representation of Smad7 promoter having VDRE cloned in pGL3 upstream of Firefly luciferase (upper panel). Shown are luciferase assays in COS-1 cells cotransfected with pFLAG-mVDR, pFLAG-mRXR, and pGL3-Smad7 promoter plasmids, cultured in the absence or presence of 100 nM 1,25(OH) 2 D 3 (lower panel). B, shown is a luciferase assay in EL4 cells transiently transfected with pGL3-Smad7 promoter plasmid, cultured in the absence or presence of 10, 50, 100 nM 1,25(OH) 2 D 3 . C, shown is a luciferase assay in EL4 cells transiently transfected with pGL3-Smad7 promoter plasmid, cultured in the absence or presence of 1,25(OH) 2 D 3 (100 nM) and SIS3 (10 M). The percentage of luciferase activity was plotted by taking the control as 100. D, ChIP assays were performed with EL4 cells cultured in the absence or presence of 1,25(OH) 2 D 3 . The immunoprecipitation was performed with antibodies to VDR and p-Smad3. DNA was eluted after immunoprecipitation and amplified using Smad7 VDRE promoter-specific primers. GAPDH promoter-specific primers were used as the control. E, ChIP experiment was performed in EL4 cells treated with TGF␤ and IL6 in the presence and absence of 1,25(OH) 2 D 3. Chromatin was site specifically digested at the BamHI site, which spans VDRE and SBE. ChIP/re-ChIP was performed with antibodies specific for VDR, RXR␣, p-Smad3, and HDAC2. DNA was eluted and amplified using Smad7 VDRE-and Smad7 SBE-specific primers. F, qPCR analysis of Smad7 mRNA in EL4 cells cultured with 1,25(OH) 2 D 3 in the presence or absence of SIS3 (10 M). G, qPCR analysis of Smad7 mRNA in Th17 cells cultured with 1,25(OH) 2 D 3 in the presence or absence of SIS3 (10 M). H, ChIP assays were performed on Th17 cells cultured in the absence or presence of 1,25(OH) 2 D 3 . ChIP/re-ChIP was performed with antibodies to VDR RXR␣, p-Smad3, and HDAC2. Results are represented as the % of input (A, B, C, E, F, G, and H). the vehicle, and the experimental group was injected with 1, 25(OH) 2 D 3 . It is evident from the subsequent clinical scores, which are statistically significant, that 1,25(OH) 2 D 3 reduced the progression of the disease (Fig. 5A).
Total CD4 cells isolated from a single-cell suspension prepared from the spleen and lymph nodes of unimmunized, EAE, and EAE mice treated with 1,25(OH) 2 D 3 were subjected to Western blot analysis with antibodies for Smad7, Smad3, and p-Smad3. In accordance with in vitro differentiation data, EAE mice treated with 1,25(OH) 2 D 3 showed significantly decreased Smad7 and increased Smad3 expression, whereas EAE mice expressed abundant Smad7 and minimal Smad3 expression (Fig. 5B). We also performed a ChIP assay in CD4 cells isolated from untreated and EAE and EAE mice treated with 1, 25(OH) 2 D 3 to investigate the VDR/Smad3/HDAC2 complex on the Smad7 promoter that we have observed in in vitro differentiated Th17 cells. Similar repressive complex was observed (Fig. 5C). Western blot analysis for ERK and phospho ERK (p-ERK) was performed in CD4-positive T cells obtained from EAE and mice treated with 1,25(OH) 2 D 3. Significant phosphorylation was observed in CD4 T cells treated with 1,25(OH) 2 D 3 (Fig. 5, C and D). Additionally, we performed intracellular staining for IL17, Foxp3, in single-cell suspensions from spleen and lymph nodes from mice after re-stimulation with MOG peptide (Fig. 6, A and B). Total CD4 cells from EAE mice expressed significantly more IL17, which was repressed in the mice in the 1,25(OH) 2 D 3 -treated EAE group (Fig. 6A). Unlike IL17, EAE mice expressed less Foxp3, which was significantly elevated in the 1,25(OH) 2 D 3 -treated EAE group (Fig.  6B). Smad7 and IL17 expression was analyzed in the brain of EAE mice, which suggest that the elevated Smad7 expression correlates with the disease severity and Th17 infiltration in central nervous system (Fig. 6, C and D). However, in EAE mice treated with 1,25(OH) 2 D 3 this increase in Smad7 and IL17 was abrogated, and their expression correlated with the decrease in disease severity. These results are in keeping with the reported role of Smad7 in central nervous system (CNS) infiltration by inflammatory T cells (42).

1,25(OH) 2 D 3 Induces IL10 Expression via Smad3 and Represses Th17
Genes via pERK-IL10 has been reported to be vital for VDR-mediated suppression of EAE, and 1,25(OH) 2 D 3 has been shown to induce IL10 production in Tregs and even in Th17 cells (30,43). We also observed that 1,25(OH) 2 D 3 induces the expression of IL10 in Th17 cells (Fig. 7A). In addition, to direct gene repression of the Th17-signature cytokine il17a, 1,25(OH) 2 D 3 has been shown to negatively regulate the expression of il23r, il17f, il22, and other genes that are crucial for the maintenance of the Th17 phenotype (44). To understand the mechanism behind negative regulation, il23r, il17f, il22 promoters were searched for a potential negative VDRE, but we failed to identify any VDRE on these promoters.
At this point we investigated if either of the identified arms of the 1,25(OH) 2 D 3 pathway (Smads and p-ERK) plays a role in the regulation of these genes. Differentiating Th17 cells were treated with SIS3 or U0126 (a non-Smad ERK pathway inhibitor) along with 1,25(OH) 2 D 3 (38,45). Results from qPCR clearly depict that the ERK inhibition relieved the 1,25(OH) 2 D 3 -mediated suppression of different Th17-related genes (Fig. 7B), and Smad inhibition repressed the 1,25(OH) 2 D 3 -mediated induction of IL-10 to some extent (Fig. 7B). These results suggest the role of p-ERK and Smad3 in 1,25(OH) 2 D 3 -VDR mediated modulation of Th17 genes and IL10.

1,25(OH) 2 D 3 Also Inhibits Smad7 Expression in Th1
Cells-Th1 cells are other inflammatory T cells that are also known to play a causative role in progression of EAE. In keeping with the previous reports, 1,25(OH) 2 D 3 suppressed Th1-specific cytokine IFN␥ in EAE mice (Fig. 8, A and B) (46). Th1 cells also express Smad7, and its expression strongly correlates with Th1 master transcription factor T-bet (42). Furthermore, Smad7 knock-out mice are known to display blunted Th1 proliferation and responses (42). As Smad7 is crucial for Th1 cell function, we looked for the expression pattern of Smad7 in Th1 cells when cultured in the presence of 1,25(OH) 2 D 3 . Results clearly suggested that 1,25(OH) 2 D 3 inhibits the expression of Smad7 in Th1 cells also. , total CD4 cells from spleen and lymph nodes were isolated, and the expression of Smad3, p-Smad3, and Smad7 was analyzed by immunoblot analysis. Quantified data were shown above the each panel. C, ChIP experiment performed in CD4 cells isolated from untreated and EAE and EAE mice treated with 1,25(OH) 2 D 3 . ChIP/re-ChIP was performed with antibodies specific for VDR, RXR␣, p-Smad3, and HDAC2. D, immunoblot analysis of p-ERK and ERK from total CD4 cells isolated from spleen and lymph nodes of unimmunized and EAE and EAE mice treated with 1,25(OH) 2 D 3. shown is a quantitative representation of blots by normalizing p-ERK protein level over ERK protein level. Data are representative of three independent experiments with at least five mice for each group. In A and C, data represent the mean Ϯ S.D. (error bars) of three independent experiments each performed in triplicate. *, p Ͻ 0.05, compared with EAE group or as indicated. B and D are representative of three independent experiments with similar results.
This report elucidates the crucial role of 1,25(OH) 2 D 3 -VDR in the regulation of Th17 prototype genes and EAE progression by modulating Smad and MAPK (non-Smad) pathways that are downstream of TGF␤ (Fig. 9). To the best of our knowledge this is the first report of 1,25(OH) 2 D 3 -VDR-mediated suppression of Smad7 by virtue of VDR-Smad3-HDAC2-repressive complex formation on its promoter, which would allow permissiveness to TGF␤ signaling to induce Foxp3. This study also highlights the role of ERK in 1,25(OH) 2 D 3 -mediated suppression of Th17 prototype genes and EAE.

DISCUSSION
TGF-␤ signaling is essential for the differentiation of Treg and Th17 cells and is orchestrated by both Smad and non-Smad pathways. This study reports that 1,25(OH) 2 D 3 -VDR functions as a regulatory switch by virtue of two distinct modulatory pathways: the Smad pathway and ERK activation. In the Smad pathway 1,25(OH) 2 D 3 -VDR is involved in direct gene repression of Smad7 and enhanced Smad3 expression, which partly contribute to the secretion of the Treg-specific cytokine IL10 (Fig. 7, A and B) (47, 48). 1,25(OH) 2 D 3 -VDR also induces ERK activation, which inhibits Th17-related genes involved in the differentiation of Th17 cells (Fig. 7B) (47)(48)(49)(50). The observed increase in Smad3 expression and activation may also be attributed to ERK activation as reported earlier (51). All of these observations suggest a key regulatory role for 1,25(OH) 2 D 3 -VDR axis.
Smads have been implicated in Treg and Th17 cellular events; Smad2 knock-out mice have a decreased proportion of Th17. whereas Smad3 knock-out mice have reduced levels of Tregs (11,28). Smad2 and Smad3 play crucial roles in Th17 differentiation and maintenance, which involve competition for interaction with ROR␥t. Smad2 interaction promotes and Smad3 interaction inhibits the function of ROR␥t (11,28). Smad3 expression and activation is one of the key elements responsible for Foxp3 expression in Tregs (52,53). Retinoic acid inhibits Th17 cell differentiation by inducing the expression and phosphorylation of Smad3 and also induces Foxp3 expression (54, 55). Tregs do not express Smad7, and it has been suggested that T cells lacking Smad7 expression would allow permissiveness to TGF␤-Smad3-mediated signaling and may function as Tregs (56 -58). Other than the ability of VDR to activate and interact with Smad3, the mechanistic evidence of the 1,25(OH) 2 D 3 -VDR axis in modulation of Smads and to regulate in vitro differentiation of Th17 cells has remained elusive (16).
In this study we report the expression patterns of these Smad proteins along with those of co-Smad (Smad4) and inhibitory Smad (Smad7) in in vitro differentiated Tregs and Th17 cells. 1,25(OH) 2 D 3 induced the expression of Smad3, inhibited the expression of Smad7, and had no effect on the expression of Smad2 or Smad4 (Fig. 1B). The 1,25(OH) 2 D 3 -VDR-HDAC2 inhibitory complex was recruited at the negative VDRE of the Smad7 promoter and reduced its expression (Figs. 2, 3, and 5). Interestingly Smad3, whose expression appeared to be up-regulated by 1,25(OH) 2 D 3 -VDR, was identified as a partner of this complex (Figs. 3, D-H and 5C). Our results suggest that inhibition of Smad3 by its specific inhibitor SIS3 relieved the 1,25(OH) 2 D 3 -VDR-mediated inhibition of Smad7 to some extent (Figs. 3C, F, and G). Smad3 has been shown to induce the expression of Smad7 by binding to its promoter, in response to TGF␤ (59,60). This study highlighted the role of Smad3 as a negative modulator of Smad7 expression in the presence of 1,25(OH) 2 D 3 -VDR (Figs. 3 and 5). This is suggestive of a stimuli-specific role of Smad3 on Smad7 expression. An increase in Smad3 expression had been attributed to Smad7 repression along with ERK activation (47,48,51). An increase in functional Smad3 competes out Smad2 to physically interact with ROR␥t and contributes to inhibition of ROR␥t transcriptional activity, whereas it cooperates with NFAT for the expression of Foxp3 (11,28,52).
Th17 cells play a key role in induction of autoimmune diseases; hence, they are postulated to be pathogenic (62, 63).

1,25(OH) 2 D 3 -VDR Represses Smad7 and Inhibits EAE
Tregs are immunosuppressive in nature and thereby inhibit development of autoimmune diseases (1). Most autoimmune diseases are characterized as having increased levels of Th17 cells and decreased levels of Treg cells in the CNS (1, 64, 65). 1,25(OH) 2 D 3 inhibits the progression of rheumatoid arthritis and multiple sclerosis in both humans and mice (21,66). To validate our in vitro findings, we studied the effect of 1,25(OH) 2 D 3 in EAE-induced mice. Expectedly, CD4 cells from EAE mice treated with 1,25(OH) 2 D 3 showed increased activation of ERK (Fig. 5, C and D). Our results also suggested that CD4 cells from EAE mice expressed more Smad7 and IL17A, all of which were significantly reduced in mice treated with 1,25(OH) 2 D 3 (Figs. 5B and 6A). Smad7 and IL17 expression was also observed in the brains of EAE mice or EAE mice treated with 1,25(OH) 2 D 3 , which suggests that Smad7 expression correlates with the disease severity and Th17 infiltration in the CNS (Fig. 7, A and B).
CCR6, CXCR3, and Smad7 have been implicated in the infiltration of CD4 cells (Th1/Th17/Treg) in CNS. A T cell-specific deletion of Smad7 in mice decreased the frequency of CD3positive mononuclear cells in CNS. Smad7 expression has been clearly shown to be up-regulated in CD4 cells in active EAE (42). Transgenic mice overexpressing Smad7 in T cells developed an enhanced course of encephalitis accompanied by elevated infiltration of CD4 inflammatory cells in CNS (42). Mice lacking CCR6 were highly resistant to EAE, and a blockade of