Logo of pnasPNASInfo for AuthorsSubscriptionsAboutThis Article
Proc Natl Acad Sci U S A. 2005 Nov 1; 102(44): 16007–16012.
Published online 2005 Oct 20. doi:  10.1073/pnas.0506516102
PMCID: PMC1257750

Regulation of relB in dendritic cells by means of modulated association of vitamin D receptor and histone deacetylase 3 with the promoter


The NF-κB component RelB is essential for dendritic cell (DC) differentiation and maturation. The vitamin D receptor (VDR) is a nuclear receptor that mediates inhibition of DC maturation and transcriptional repression of relB after engagement of its ligand, 1α,25-dihydroxyvitamin D3, or related analogs (D3 analogs). Ligand-dependent relB suppression was abolished by a histone deacetylase (HDAC) inhibitor. Constitutive association of VDR with the relB promoter was demonstrated in DCs by chromatin immunoprecipitation. Promoter binding by VDR was enhanced by ligand and reduced by LPS. Association of HDAC3 and HDAC1 with the relB VDR-binding site was observed, but only HDAC3 was reciprocally modulated by D3 analog and LPS. Overexpression of HDAC3 caused relB promoter suppression, increased sensitivity to D3 analog, and resistance to LPS. Depletion of HDAC3 attenuated relB suppression by D3 analog. In vivo, D3 analog resulted in reduced RelB in DCs from VDR WT mice but not VDR knockout mice. Other NF-lation of RelB and c-Rel in control animals. We conclude that vitamin D-regulated relB transcription in DCs is controlled by chromatin remodeling by means of recruitment of complexes including HDAC3.

Keywords: NF-κB, antigen presentation, nuclear receptors, chromatin remodeling, immunity

Dendritic cells (DCs) occupy a unique role in initiating immune responses through their ability to mingle with and activate naïve T cells (1, 2). The differentiation of bone marrow-derived precursors toward DC lineages and the functional plasticity necessary for an effective spectrum of DC/T cell interactions is tightly regulated. Identification of individual gene products that are essential to normal DC differentiation and function has provided critical insights for understanding DC biology and for therapeutic modulation of DCs. The clearest example of such a gene product is the NF-κB component RelB. Expression of RelB appears late in embryogenesis and is concentrated in the thymus and secondary lymphoid organs: specifically, in interdigitating DCs (iDCs) (3). Deletion of relB resulted in the virtual absence of iDCs with accompanying deficits in cognate immunity (4, 5). Functional analysis of the remaining DCs or ex vivo derivation of relB-/- DCs revealed failure to up-regulate MHC and costimulatory proteins and deficiency in the capacity to stimulate naïve T cells, cross-present MHC I-restricted peptides, and facilitate Ig isotype switching (6). Significantly, RelB-deficient bone marrow-derived DCs (BMDCs) or BMDCs in which RelB activity is inhibited have the potential to induce antigen-specific immune tolerance in vivo (7).

RelB intracellular levels and signaling are enhanced during DC maturation and participate in up-regulation of immunostimulatory proteins (8, 9). Transcriptional regulation of relB has not been extensively studied, although it is known to be inducible by means of NF-κB response elements (10). We observed reduced intracellular levels of RelB in BMDCs generated in the presence of the active form of vitamin D3, 1α,25-dihydroxyvitamin D3 (1α,25(OH)2D3), and related analogs (D3 analogs) (11, 12). This effect depended on expression of the vitamin D receptor (VDR), which is a member of the nuclear receptor (NR) family of DNA binding proteins, and was mediated through binding of VDR/retinoic acid X receptor α (RXRα) to vitamin D response elements (VDREs) in the promoter (12). These findings added to a growing body of evidence linking 1α,25(OH)2D3/VDR to negative regulation of Th1-type immune responses and to inhibition of DC maturation (1316). The implications of this literature are best exemplified by animal models (allograft rejection, autoimmune diabetes mellitus, and experimental allergic encephalomyelitis) that are ameliorated by 1α,25(OH)2D3 or D3 analogs (1317), by evidence that D3 analog-conditioned DCs mediate immune tolerance in vivo (1316,18), and by epidemiologic studies that identify VDR genotype and vitamin D status as risk factors for autoimmunity (19, 20).

The interaction of VDR with chromosomal DNA is necessary for positive transcriptional responses, but NR association with additional proteins (coactivators, corepressors, and histone-modifying enzymes) is also crucial (21, 22). Although transcriptional suppression by 1α25(OH)2D3/VDR may be mediated by competitive displacement of positive regulatory factors, active promoter suppression by means of interaction of DNA-bound VDR with adaptor proteins such as NR corepressor and SMRT (silencing mediator for retinoid and thyroid hormone receptors) that subsequently recruit histone deacetylases (HDACs) has also been demonstrated (21, 22). In this case, HDAC activity results in a tight DNA/histone conformation that inhibits transcription. This mechanism has been most frequently characterized as a ligand-independent process involving VDR or other NRs including estrogen receptor, thyroxine receptor, or glucocorticoid receptor (22). Examples of ligand-dependent NR inhibitory complexes have also been identified. Most notably, Kitagawa et al. (23) have characterized a complex (WINAC) incorporating VDR and the William's syndrome transcription factor that, in the presence of 1α25(OH)2D3, mediates both histone acetyltransferase-associated transcriptional up-regulation and HDAC-associated transcriptional down-regulation of separate promoters.

We present here experimental evidence of a ligand-dependent, HDAC-containing complex associated with relB promoter-bound VDR in DCs. Complementary results were obtained in a DC-derived cell line, in BMDCs, and in DCs extracted from secondary lymphoid organs. Evidence is presented that a specific HDAC (HDAC3) is involved in negative regulation of relB and that a DC maturational stimulus (LPS) results in shedding of VDR/HDAC3 from the relB promoter. To our knowledge, the results represent one of the first discrete examples of the importance of chromatin remodeling in regulating DC function and of the potential for harnessing this process to enhance DC-based immunotherapy.

Materials and Methods

Cell Culture, Experimental Animals, and Reagents. Additional details are provided in Supporting Materials and Methods, which is published as supporting information on the PNAS web site. Murine BMDCs were prepared as described in ref. 13. The DC-like cell line D2SC1 was cultured in Iscove's modified Dulbecco's medium/5% FCS. C57BL/6, VDR knockout (KO), and WT mice (24) were maintained in a specific pathogen-free facility. 1α,25(OH)2-16-ene-23-yne-26,27-hexafluoro-19-nor-D3 was provided by Milan Uskokovic (BioXell, Nutley, NJ). Reagents used were as follows: polyclonal antibodies against RelB; RelA, c-Rel, VDR, HDAC3, HDAC1, and SMRT (Santa Cruz Biotechnology); anti-β-actin, trichostatin A (TSA), and LPS (Sigma-Aldrich); and horseradish peroxidase-protein A (Amersham Pharmacia Biotech). Mouse VDR and HDAC3 expression constructs and luciferase reporter plasmids of mouse and human relB promoters with WT and mutant VDREs are described in refs. 12 and 25. The pSHAG-HDAC3 short-hairpin RNA (shRNA) vector and a control vector with the HDAC3-specific sequence scrambled have been reported previously (25), and additional details are provided in Fig. 5, which is published as supporting information on the PNAS web site.

Chromatin Immunoprecipitation (ChIP). For BMDCs and D2SC1 cells, the culture additions carried out before ChIP assays were 10-10 M D3 analog or 50 ng/ml LPS. ChIP assays were carried out as described by Kuo and Allis (26). Precipitated DNA was subjected to PCR using primers flanking the mouse relB promoter VDRE and control primers flanking a proximal promoter region (additional details provided in Fig. 6, which is published as supporting information on the PNAS web site).

Luciferase Reporter Assays. D2SC1 cells (5 × 106) were transfected in six-well plates with 1 μg of reporter and 10 ng of pRL-TK Renilla luciferase plasmids (Promega) using FuGENE 6 (Roche Applied Science, Indianapolis). Cotransfections for individual experiments were as follows: 0.5 μg of mouse VDR in pcDNA3.1 or empty pcDNA3.1, 0.5 μg of HDAC3 in pDEP4F or empty pDEP4F, and 25 ng of pSHAG-HDAC3 or scrambled pSHAG-HDAC3. Medium was replaced 10 h later with medium containing one or more of the following: D analog (10-12 to 10-8 3M), LPS (50 ng/ml) and TSA (10 ng/ml), or control medium containing equivalent volumes of vehicle (absolute ethanol). After a further 24 h, luciferase activity was assayed with the Dual Luciferase Assay Kit (Promega).

Immunoblotting. BMDCs and D2SC1 culture additions (D3 analog or vehicle and/or LPS) were applied for 24 h. Cell lysis and immunoblotting were carried out as described in ref. 11.

i.p. Treatment of Mice and Purification of Murine DCs. D3 analog. Groups of two to three VDR WT and VDR KO mice were injected i.p. with 50 ng of D3 analog in 250 μl of PBS or with PBS containing vehicle (absolute ethanol) 12 h before being killed. In some experiments, VDR WT mice also received 50 ng of TSA i.p. in PBS or PBS containing vehicle (absolute ethanol).

LPS. VDR WT mice were injected i.p. with 100 μg of LPS 2 h before being killed. CD11c positive and negative fractions from lymphoid organs were separated by using anti-CD11c-coated microbeads and the MiniMACS cell sorting system (Miltenyi Biotec, Auburn, CA). Separated cells were used immediately for ChIP assays or immunoblotting.

Statistical Analysis. Experiments were carried out a minimum of three times with consistent results. For reporter assays, triplicate samples for each condition were prepared, and final results were expressed as mean ± SD. Experimental results were compared by two-tailed, unpaired Student's t tests with significance at P < 0.05.


Modulated Association of VDR and HDAC3 with the VDRE-Containing Region of the relB Promoter in DCs. The role of chromatin remodeling in VDR-mediated negative transcriptional regulation of relB in DCs was initially investigated by reporter assay in D2SC1 cells using a mouse relB promoter-driven luciferase construct. Fig. 1A illustrates the effect of the HDAC inhibitor TSA on relB promoter suppression by D3 analog and/or VDR overexpression. In this, and in multiple repeat experiments, a low concentration of TSA (10 ng/ml) resulted in abolition of ligand-dependent promoter suppression as well as attenuation of ligand-independent suppression.

Fig. 1.
Results of experiments demonstrating the role of histone deacetylation in relB transcriptional suppression and the modulated association of VDR and HDAC3 with the relB promoter. (A) Luciferase reporter assays carried out in D2SC1 cells 24 h after transfection ...

ChIP was next used to identify candidate proteins associated with the VDRE-containing region of the relB promoter. Reasoning that signaling events associated with up-regulation of RelB expression may result in disassociation of negative regulatory elements, ChIP assays were carried out from murine BMDCs with and without LPS exposure. Fig. 1B demonstrates the results from one experiment in which protein/DNA complexes were precipitated with antibodies against VDR itself, HDAC1, HDAC3, and the corepressor SMRT. Although the appropriate band was amplified from all four precipitates from unstimulated BMDCs (suggesting that each protein is constitutively associated with the promoter), there was specific attenuation (confirmed in repeated experiments) of the band from anti-VDR and anti-HDAC3 precipitates after LPS stimulation. Based on this finding and on recent reports that identified HDAC3 as a nonredundant component of multiple negative regulatory nuclear complexes (25, 27), subsequent experiments focused on examining the dynamics and functional significance of HDAC3 in VDR-mediated regulation of relB. In this regard, Fig. 1C illustrates results of ChIP assays carried out in VDR WT and KO BMDCs as well as in the D2SC1 cell line in the presence or absence of D3 analog and LPS. As shown, band intensity in anti-VDR and anti-HDAC3 precipitates from VDR WT BMDCs and D2SC1 cells was increased in D3 analog-treated samples compared with vehicle-treated samples and was decreased in samples from LPS-stimulated cells. Although a baseline level of RXRα and HDAC3 association was observed in VDR KO BMDCs, no modulation occurred with either D3 analog or LPS. Immunoblots for RelB, VDR, RXRα, and HDAC3 using cell lysates from VDR WT and KO BMDCs under vehicle-treated, D3 analog-treated, and LPS-stimulated conditions ruled out the relatively trivial explanation that changes of the band intensity identified in the ChIP assays were caused by reduced intracellular levels of the VDR and HDAC3 proteins (Fig. 7, which is published as supporting information on the PNAS web site). Intracellular relocation of HDAC3 and VDR was also ruled out by immunofluorescence microscopy of VDR WT BMDCs (data not shown). We conclude that VDR and HDAC3 are specifically recruited to and shed from the VDRE-containing region of the relB promoter in conjunction with exogenous negative (D3 analog) and positive (LPS) transcriptional stimuli.

HDAC3 Regulates relB Promoter Activity and Sensitivity to VDR Ligand. The capacity for HDAC3 to contribute to VDR-mediated negative transcriptional regulation of the mouse relB promoter was examined by reporter assay. As shown in Fig. 2A, overexpression of HDAC3 resulted in substantial inhibition of WT promoter activity that was comparable with and additive to that caused by VDR overexpression. These inhibitory effects were all but eliminated when a reporter construct in which the VDRE sequence had been mutated to attenuate VDR/RXRα binding (VDRE-Mut) was used. An identical result was obtained with human WT and VDRE-Mut relB promoter constructs (Fig. 8, which is published as supporting information on the PNAS web site). Overexpression of HDAC3 was also shown to be associated with proportionately greater suppression of the mouse relB promoter by 10-8 and 10-10 M D3 analog as well as with suppression by D analog concentrations (10-12 and 10-14 M) that had no effect on promoter activity in untransfected cells (Fig. 2B). In additional experiments, combined overexpression of VDR and HDAC3 prevented LPS-induced increase of relB WT but not VDRE-Mut promoter activity (Fig. 9, which is published as supporting information on the PNAS web site). Taken together, these experiments suggest a potent functional role for HDAC3 in mediating ligand-independent as well as ligand-dependent VDR-mediated suppression of relB in DCs. Furthermore, results with mutant constructs clearly localize this effect to the previously identified VDRE.

Fig. 2.
Results of experiments demonstrating that overexpression of HDAC3 augments VDR and D3 analog-mediated relB suppression. (A) Luciferase reporter assays in D2SC1 cells 24 h after transfection with reporter constructs containing the WT murine relB promoter ...

Depletion of HDAC3 Attenuates VDR Ligand-Mediated Inhibition of relB. To address the concern that the findings obtained by overexpression represented artifactual results, experiments were carried out in which HDAC3 expression was specifically suppressed by using a previously validated shRNA construct (ref. 25 and Fig. 5). In four identical experiments, the proportionate suppression by D3 analog of mouse relB promoter activity was compared in D2SC1 cells cotransfected with HDAC3 shRNA or with an ineffective (“scrambled”) shRNA construct. The result of each experiment (summarized in Fig. 3 as percentage of the vehicle-treated control for each condition) demonstrated a lesser degree of VDR ligand-mediated relB suppression in the presence of the HDAC3 shRNA, consistent with a distinct role for HDAC3 in this process.

Fig. 3.
Results of four individual luciferase reporter assay experiments (Expt. 1–4) carried out in D2SC1 cells 24 h after transfection with WT murine relB promoter construct. Cell aliquots were cotransfected with shRNA against murine HDAC3 (HDAC3 shRNA) ...

Modulation of VDR and HDAC3 Association with the relB Promoter Occurs in DCs in Vivo and Corresponds with Intracellular RelB. DCs purified from lymphoid organs of VDR WT mice were directly subjected to ChIP assay. In separate experiments, the protein/DNA interactions of interest were compared for DCs from mice receiving PBS or LPS 2 h before killing and for mice receiving vehicle or D3 analog 12 h before killing (Fig. 4A). As had been observed in BMDCs and D2SC1 cells, VDR and HDAC3 association with the relB promoter was reduced after LPS administration and enhanced after administration of D3 analog (these results were consistently reproducible). Immunoblots of total cell lysates from purified DCs of D3 analog and vehicle-treated mice confirmed that intracellular RelB, although not RelA or c-Rel, was substantially reduced by D3 analog (Fig. 4B). As anticipated, down-regulation of RelB did not occur in DCs purified from D3 analog-treated VDR-deficient animals (Fig. 4C). Finally, the significance of HDAC activity for D3 analog-mediated suppression of RelB in VDR WT mice was tested by concomitant administration of D3 analog and TSA (with appropriate vehicle controls for each agent) followed by immunoblotting of separated CD11c+ve and CD11c-ve cells (Fig. 4D). Strikingly, TSA was associated with a moderate increase of RelB in cells from vehicle-treated animals and with elimination of RelB suppression in cells from D3 analog-treated animals. It is of note that, although basal levels of RelB were substantially lower in CD11c-ve cell fractions, similar patterns of RelB modulation were observed (Fig. 4D, two right columns). Immunoblotting for RelA revealed no modulation after TSA administration. In contrast, the HDAC inhibitor induced a moderate up-regulation of c-Rel levels in CD11c+ve and CD11c-ve cells. Overall, this series of experiments was interpreted as confirmation that ligand-dependent negative regulation of relB occurs in DCs in vivo and that this process (i) involves modulation of VDR and HDAC3 association with the VDRE-containing region of the promoter, (ii) results in substantial and specific reduction in intracellular RelB, and (iii) depends on HDAC activity.

Fig. 4.
Results of experiments demonstrating that modulated VDR and HDAC3 binding to the relB promoter occurs in DCs in vivo and is associated with D3 analog-mediated alterations in intracellular RelB. (A) ChIP assays of DCs from VDR WT mice 2 h after treatment ...


The results reported here indicate that a HDAC-based chromatin-remodeling mechanism participates in regulating RelB levels in DCs. Specifically, the data suggest the following. (i) There is constitutive binding of VDR and HDAC3 to the relB promoter that is decreased by the promaturational stimulus LPS. (ii) A VDR ligand enhances HDAC activity through increased retention of VDR and HDAC3 on the promoter. This model is supported by experiments that demonstrate that modulations observed for RelB were not observed for other NF-κB components, were absent in DCs from VDR-deficient animals, were augmented by overexpression of VDR and HDAC3, and were dependent on previously defined VDREs (12). The results are consistent with recognized immunomodulatory properties of the vitamin D system (1318). More broadly, the study reveals the potential importance of chromatin remodeling in regulating DC differentiation and function.

We have previously reported that mouse and human relB promoters contain VDREs, which are necessary for transcriptional suppression (12). Although VDR-mediated negative regulation of immune-related genes (IFN-γ, CD40, IL-2, and granulocyte–macrophage colony-stimulating factor) had been carefully investigated by other groups (2831), the observations for relB were more consistent with direct NR-based promoter remodeling than these existing reports. This contention is supported by ChIP experiments that establish that VDR and RXRα are constitutively bound to the relB promoter of BMDCs, D2SC1 cells, and DCs from lymphoid organs. In each case, a VDR ligand enhanced the association of VDR with the promoter. Interestingly, although the relB VDRE sequences only supported binding of VDR/RXRα heterodimers in gel shift assays (12), ChIP results indicate that binding of VDR and RXRα are not codependent, as evidenced by persistent RXRα binding in VDR KO BMDCs and by the lack of RXRα modulation in the presence of D3 analog or LPS. This observation may suggest the existence of one or more additional binding partners for RXRα on the relB promoter: a form of promiscuity among NR heterodimers for which there is precedence (32). Significantly, the observed VDR-independent binding of RXRα is devoid of negative regulatory activity after D3 analog exposure and is not associated with HDAC3 modulation.

The abolition of D3 analog-mediated relB suppression by TSA provides strong evidence for ligand-enhanced assembly of a chromatin remodeling complex. The validity of the in vitro finding is well borne out by subsequent experiments in TSA-treated animals. Multiple experimental results implicate HDAC3 as having a specific role in relB suppression that is closely linked with VDR engagement and shedding. ChIP assays demonstrated close correlations between VDR and HDAC3 band intensity after D3 analog and LPS exposure, whereas there was no modulation of HDAC3 in VDR KO BMDCs. Overexpression of HDAC3 caused potent suppression of relB promoter activity that was additive to the effect of VDR overexpression, occurred in the absence of VDR ligand, and was abolished by mutation of the VDRE. Furthermore, overexpression of HDAC3 resulted in heightened sensitivity of the relB promoter to D3 analog concentration. Finally, combined overexpression of VDR and HDAC3 rendered the relB promoter virtually unresponsive to LPS. These observations also imply that the addition of ligand serves to enhance formation of the VDR/HDAC3-containing complex without being essential for activation of HDAC function and that the suppressive capacity of the complex is limited by availability of VDR and HDAC3. This latter point is further supported by the effect of HDAC3 suppression to attenuate D3 analog-mediated relB suppression.

A number of recent studies identify HDAC3 as a functional constituent of distinct negative regulatory complexes. Zhang et al. (33) found HDAC3 to be both necessary and sufficient for repression of the growth-differentiation factor 11 (gdf11) gene in fibroblasts (33). Recruitment of HDAC3-containing complexes to repressed promoter DNA by unliganded thyroxine receptor was observed by Li et al. (34) in Xenopus oocytes and by Ishizuka and Lazar (35) in 293 cells. Tamoxifen-bound estrogen receptor-α was shown by Liu and Bagchi (36) to coordinately recruit distinct HDAC3 and HDAC1-containing complexes to repress estrogen-responsive promoters in breast cancer cells (36). Finally, Baek et al. (37) demonstrated disassociation of a HDAC3-containing complex from NF-κB-regulated gene promoters after IL-1β stimulation, a mechanism analogous to LPS-induced disassociation of VDR/HDAC3 from the relB promoter. Although the experiments reported here demonstrate the potential for HDAC3 to mediate RelB down-regulation in DCs and indicate that HDAC3 is necessary for at least part of this activity, it should be acknowledged that other HDACs may participate. We find, for example, that association of HDAC1 with the same promoter region is demonstrable by ChIP but is not altered by LPS (data not shown) or by D3 analog. Regarding the role of known corepressor proteins in VDR/HDAC3-mediated relB suppression, we have observed no demonstrable association of NR corepressor with the VDRE-containing region of the relB promoter in DCs (data not shown). In contrast, SMRT association was readily demonstrated by ChIP but was not modulated by D3 analog or LPS and is thus unlikely to be an essential scaffold for VDR-mediated recruitment of HDAC3. We anticipate that additional ligand-inducible binding partners for VDR and HDAC3 will be identified and may reveal previously unrecognized features of DC gene regulation.

The results of in vivo experiments provide important validation of data obtained in cultured cells. Most notably, down-regulation of RelB in CD11c+ve cells occurred within 24 h of a single dose of D3 analog to VDR WT mice. Other NF-κB proteins (RelA and c-Rel) were unaffected. In agreement with in vitro assays, this in vivo suppression was shown to depend on HDAC activity. Functional characterization of such RelB-depleted DCs and development of strategies for augmentation of HDAC3/VDR interactions in vivo will determine whether this mechanism can be exploited to promote immune tolerance. Of additional interest is the finding that inhibition of HDAC in control animals was observed to result in increased RelB and, to a lesser degree, c-Rel expression in DCs and non-DCs. This finding indicates a basal level of hypoacetylation of these genes that is amenable to pharmacological manipulation and merits additional study as a means to enhance DC-mediated presentation in vivo.

The apparent potency of VDR-mediated RelB suppression observed in these experiments contrasts somewhat with the lack of a broad dysregulation of cellular immunity in VDR-deficient animals (38). Furthermore, a number of our experimental results argue against a nonredundant role for VDR in the suppression of basal relB activity in DCs and DC precursors. For example, we have not observed RelB overexpression in VDR KO BMDCs compared with VDR WT BMDCs, and engraftment of DC lineages occurs similarly after transfer of either VDR WT or KO hematopoietic stem cells into irradiated mice. In addition, we do not find that VDRE-mutant relB promoter constructs are consistently more active or less responsive to LPS stimulation in D2SC1 cells compared with the WT constructs (Fig. 7 and X.D and M.D.G., unpublished data). As previously discussed, the persistence of RXRα and HDAC3 association with the relB promoter in VDR KO DCs may indicate some level of redundancy for basal suppression of relB in DCs. In addition, other non-VDR-binding regulatory elements within the promoter likely suffice for appropriate control of relB transcription during embryonic development and DC differentiation. Nonetheless, there is evidence that VDR deficiency is associated with abnormalities of DC/T cell dynamics and that localized production of 1α,25(OH)2D3 is involved in regulating induced immunological responses. s.c. lymph nodes of VDR KO mice are enlarged compared with WT littermates and contain a greater proportion of MHC IIhi/CD40hi DCs (13). Splenocytes from VDR KO mice have diminished capacity to generate Th1-type T cell responses (39). Enlarged lymph nodes have also been observed in mice lacking 25-hydroxyvitamin D3-1α-hydroxylase (1α-OHase), the enzyme responsible for generation of 1α,25(OH)2D3 (40). Chronic inflammatory conditions including tuberculosis, sarcoidosis, and Crohn's disease are associated with high concentrations of locally generated 1α,25(OH)2D3 that exert inhibitory effects on T cell responses (41). Hewison and colleagues (41, 42) have shown that 1α-OHase is inducible in DCs and macrophages and suggest that the resulting production of 1α,25(OH)2D3 serves as a mechanism whereby antigen-presenting cell maturation at sites of inflammation is limited to avoid autoimmunity.

Based on these published observations and on the data presented here, we propose that the induced expression of RelB in DCs during inflammatory processes is negatively regulated in a paracrine fashion by localized generation of 1α,25(OH)2D3. The mechanism underlying this regulatory activity involves ligand-induced recruitment of HDAC3 and is mediated through VDR-binding sites on the promoter. Further elucidation of this pathway as well as broad investigation of chromatin remodeling events in DCs should provide important insights into the molecular mechanisms underlying the regulation of cellular immunity.

Supplementary Material

Supporting Information:


We acknowledge the help and advice of Dr. David McKean, Mr. Michael Bell, and Mrs. Catherine Huntoon. This work was supported by National Institutes of Health (NIH) Grants R01 DK59505 (to M.D.G.); R01 DK25409, R01 DK58546, and R01 DK65830 (to R.K.); and R01 AR48147 and R01 AR50074 (to J.J.W.). This work was also supported by NIH Training Grants DK07013 (to X.D.) and CA09138 (to T.M.S.).


Author contributions: X.D., J.J.W., R.K., and M.D.G. designed research; X.D., T.M.S., and L.A.B. performed research; W.L., T.M.S., J.J.W., and R.K. contributed new reagents/analytic tools; X.D., R.K., and M.D.G. analyzed data; and X.D. and M.D.G. wrote the paper.

Conflict of interest statement: No conflicts declared.

This paper was submitted directly (Track II) to the PNAS office.

Abbreviations: DC, dendritic cell; BMDC, bone marrow-derived DC; 1α25,(OH)2D3, 1α,25-dihydroxyvitamin D3; VDR, vitamin D receptor; NR, nuclear receptor; RXRα, retinoic acid X receptor α; VDRE, vitamin D response element; HDAC, histone deacetylase; ChIP, chromatin immunoprecipitation; TSA, trichostatin A; SMRT, silencing mediator for retinoid and thyroid hormone receptors; KO, knockout; shRNA, short-hairpin RNA.


1. Mellman, I. & Steinman, R. M. (2001) Cell 106, 255-258. [PubMed]
2. Banchereau, J. & Steinman, R. M. (1998) Nature 392, 245-252. [PubMed]
3. Carrasco, D., Ryseck, R. P. & Bravo, R. (1993) Development (Cambridge, U.K.) 118, 1221-1231. [PubMed]
4. Weih, F., Carrasco, D., Durham, S. K., Barton, D. S., Rizzo, C. A., Ryseck, R. P., Lira, S. A. & Bravo, R. (1995) Cell 80, 331-340. [PubMed]
5. Burkly, L., Hession, C., Ogata, L., Reilly, C., Marconi, L. A., Olson, D., Tizard, R., Cate, R. & Lo, D. (1995) Nature 373, 531-536. [PubMed]
6. Zanetti, M., Castiglioni, P., Scoenberger, S. & Gerloni, M. (2003) Ann. N.Y. Acad. Sci. 987, 249-257. [PubMed]
7. Martin, E., O'Sullivan, B., Low, P. & Thomas, R. (2003) Immunity 18, 155-167. [PubMed]
8. Hofer, S., Rescigno, M., Granucci, F., Citterio, S., Francolini, M. & Ricciardi-Castagnoli, P. (2001) Microbes Infect. 3, 259-265. [PubMed]
9. Clark, G. J., Gunningham, S., Troy, A., Vuckovic, S. & Hart, D. N. (1999) Immunology 98, 189-196. [PMC free article] [PubMed]
10. Solan, N. J., Miyoshi, H., Carmona, E. M., Bren, G. D. & Paya, C. V. (2002) J. Biol. Chem. 277, 1405-1418. [PubMed]
11. Xing, N., Maldonado, M. L., Bachman, L. A., McKean, D. J., Kumar, R. & Griffin, M. D. (2002) Biochem. Biophys. Res. Commun. 297, 645-652. [PubMed]
12. Dong, X., Craig, T., Xing, N., Bachman, L. A., Paya, C. V., Weih, F., McKean, D. J., Kumar, R. & Griffin, M. D. (2003) J. Biol. Chem. 278, 49378-49385. [PubMed]
13. Griffin, M. D., Lutz, W., Phan, V. A., Bachman, L. A., McKean, D. J. & Kumar, R. (2001) Proc. Natl. Acad. Sci. USA 98, 6800-6805. [PMC free article] [PubMed]
14. Griffin, M. D., Xing, N. & Kumar, R. (2003) Annu. Rev. Nutr. 23, 117-145. [PubMed]
15. Mathieu, C. & Adorini, L. (2002) Trends Mol. Med. 8, 174-179. [PubMed]
16. Deluca, H. F. & Cantorna, M. T. (2001) FASEB J. 15, 2579-2585. [PubMed]
17. Cantorna, M. T., Hayes, C. E. & DeLuca, H. F. (1996) Proc. Natl. Acad. Sci. USA 93, 7861-7864. [PMC free article] [PubMed]
18. Gregori, S., Casorati, M., Amuchastegui, S., Smiroldo, S., Davalli, A. M. & Adorini, L. (2001) J. Immunol. 167, 1945-1953. [PubMed]
19. Hypponen, E., Laara, E., Reunanen, A., Jarvelin, M.-R. & Virtanen, S. M. (2001) Lancet 358, 1500-1503. [PubMed]
20. Holick, M. F. (2004) Am. J. Clin. Nutr. 79, 362-371. [PubMed]
21. Belandia, B. & Parker, M. G. (2003) Cell 114, 277-280. [PubMed]
22. Glass, C. K. & Rosenfeld, M. G. (2000) Genes Dev. 14, 121-141. [PubMed]
23. Kitagawa, H., Fujiki, R., Yoshimura, K., Mezaki, Y., Uematsu, Y., Matsui, D., Ogawa, S., Unno, K., Okubo, M. & Tokita, A. (2003) Cell 113, 905-917. [PubMed]
24. Li, Y. C., Pirro, A. E., Amling, M., Delling, G., Baron, R., Bronson, R. & Demay, M. B. (1997) Proc. Natl. Acad. Sci. USA 94, 9831-9835. [PMC free article] [PubMed]
25. Schroeder, T. M., Kahler, R. A., Li, X. & Westendorf, J. J. (2004) J. Biol. Chem. 279, 41998-42007. [PubMed]
26. Kuo, M.-H. & Allis, C. D. (1999) Methods 19, 425-433. [PubMed]
27. de Ruijter, A. J., van Gennip, A. H., Caron, H. N., Kemp, S. & van Kuilenburg, A. B. (2003) Biochem. J. 370, 737-749. [PMC free article] [PubMed]
28. D'Ambrosio, D., Cippitelli, M., Cocciolo, M. G., Mazzeo, D., Lucia, P. D., Lang, R., Sinigaglia, F. & Panina-Bordignon, P. (1998) J. Clin. Invest. 101, 252-262. [PMC free article] [PubMed]
29. Cippitelli, M. & Santoni, A. (1998) Eur. J. Immunol. 28, 3017-3030. [PubMed]
30. Alroy, I., Towers, T. & Freedman, L. (1995) Mol. Cell. Biol. 15, 5789-5799. [PMC free article] [PubMed]
31. Towers, T. L., Staeva, T. P. & Freedman, L. P. (1999) Mol. Cell. Biol. 19, 4191-4199. [PMC free article] [PubMed]
32. Pascussi, J. M., Gerbal-Chaloin, S., Drocourt, L., Maurel, P. & Vilarem, M. J. (2003) Biochim. Biophys. Acta 1619, 243-253. [PubMed]
33. Zhang, X., Wharton, W., Yuan, Z., Tsai, S.-C., Olashaw, N. & Seto, E. (2004) Mol. Cell. Biol. 24, 5106-5118. [PMC free article] [PubMed]
34. Li, J., Lin, Q., Wang, W., Wade, P. & Wong, J. (2002) Genes Dev. 16, 687-692. [PMC free article] [PubMed]
35. Ishizuka, T. & Lazar, M. A. (2003) Mol. Cell. Biol. 23, 5122-5131. [PMC free article] [PubMed]
36. Liu, X.-F. & Bagchi, M. K. (2004) J. Biol. Chem. 279, 15050-15058. [PubMed]
37. Baek, S. H., Ohgi, K. A., Rose, D. W., Koo, E. H., Glass, C. K. & Rosenfeld, M. G. (2002) Cell 110, 55-67. [PubMed]
38. Mathieu, C., Van Etten, E., Gysemans, C., Decallonne, B., Kato, S., Laureys, J., Depovere, J., Valckx, D., Verstuyf, A. & Bouillon, R. (2001) J. Bone Miner. Res. 16, 2057-2065. [PubMed]
39. O'Kelly, J., Hisatake, J., Hisatake, Y., Bishop, J., Norman, A. & Koeffler, H. P. (2002) J. Clin. Invest. 109, 1091-1099. [PMC free article] [PubMed]
40. Panda, D. K., Miao, D., Tremblay, M. L., Sirois, J., Farookhi, R., Hendy, G. N. & Glotzman, D. (2001) Proc. Natl. Acad. Sci. USA 98, 7498-7503. [PMC free article] [PubMed]
41. Hewison, M., Zehnder, D., Chakraverty, R. & Adams, J. S. (2004) Mol. Cell. Endocrinol. 215, 31-38. [PubMed]
42. Hewison, M., Freeman, L., Hughes, S. V., Evans, K. N., Bland, R., Eliopoulos, A. G., Kilby, M. D., Moss, P. A. H. & Chakraverty, R. (2003) J. Immunol. 170, 5382-5390. [PubMed]

Articles from Proceedings of the National Academy of Sciences of the United States of America are provided here courtesy of National Academy of Sciences
PubReader format: click here to try


Save items

Related citations in PubMed

See reviews...See all...


  • Compound
    PubChem chemical compound records that cite the current articles. These references are taken from those provided on submitted PubChem chemical substance records. Multiple substance records may contribute to the PubChem compound record.
  • Gene
    Gene records that cite the current articles. Citations in Gene are added manually by NCBI or imported from outside public resources.
  • GEO Profiles
    GEO Profiles
    Gene Expression Omnibus (GEO) Profiles of molecular abundance data. The current articles are references on the Gene record associated with the GEO profile.
  • HomoloGene
    HomoloGene clusters of homologous genes and sequences that cite the current articles. These are references on the Gene and sequence records in the HomoloGene entry.
  • MedGen
    Related information in MedGen
  • PubMed
    PubMed citations for these articles
  • Substance
    PubChem chemical substance records that cite the current articles. These references are taken from those provided on submitted PubChem chemical substance records.

Recent Activity

Your browsing activity is empty.

Activity recording is turned off.

Turn recording back on

See more...