Vitamin D receptor-mediated suppression of RelB in antigen presenting cells: a paradigm for ligand-augmented negative transcriptional regulation

Abstract

The immunological effects of vitamin D receptor (VDR) ligands include inhibition of dendritic cell (DC) maturation, suppression of T-helper type 1 (Th1) T-cell responses and facilitation of antigen-specific immune tolerance in vivo. While studying the molecular profile of DCs cultured in the presence of 1alpha,25(OH)D3 or synthetic D3 analogs we observed that expression of the NF-kappaB family member RelB, which plays an essential role in DC differentiation and maturation, is selectively suppressed by VDR ligands. Further in vitro and in vivo studies of VDR-mediated RelB suppression indicated that the mechanism for this effect involves direct binding of VDR/RXR alpha to a defined region of the relB promoter and assembly of a negative regulatory complex containing HDAC3, HDAC1, SMRT and, most likely, other factors. Interestingly, promoter engagement by VDR and HDAC3, but not the other identified components, is enhanced by addition of a VDR ligand and inhibited by a pro-maturational stimulus (LPS) that results in RelB upregulation. Promoter assays in a panel of cell lines suggest that the VDR ligand-dependent component of relB suppression may occur selectively in antigen presenting cells. Cell type-specific, ligand-enhanced negative transcriptional regulation represents a potentially novel paradigm for VDR-controlled genes. In this report we review the experimental data to support such a mechanism for relB regulation in DCs and present a model for the process.

Figure 1

Chromatin immunoprecipitation (ChIP) assay results are shown for day 7 bone marrow-derived DCs (BMDCs) from untreated VDR wild-type (WT) and VDR knockout (KO) mice and from VDR WT BMDCs treated overnight with 50 ng/ml LPS. Amplification of a 178 bp chromosomal DNA fragment incorporating the mouse relB promoter VDRE are shown for total genomic DNA and for protein/DNA complexes precipitated with Rabbit IgG (Control IgG) or with polyclonal rabbit antibodies against VDR (Anti-VDR) and RXRα (Anti-RXRα).

Figure 2

A. Results are shown for an experiment in which the effect of D₃ analog treatment on transcriptional activity of a mouse osteopontin (mOPN) promoter construct was compared, by luciferase reporter assay, with that of a mouse relB promoter construct (mRelB) in an osteoblast cell line (ROS17/2.8) known to express VDR. Aliquots of transfected cells were treated with vehicle (Control) or with 10⁻¹⁰ M D₃ analog. The results are expressed as mean ± SD luciferase activity from triplicate samples for each condition. † = p < 0.05 for D₃ analog-treated vs. control-treated. B. The position and sequences are shown for the wild type VDRE within the mouse relB promoter and the mOPN VDRE generated within the relB promoter by site-directed mutagenesis. C. The effect of 10⁻¹⁰ M D₃ analog treatment, VDR over-expression, or both on transcriptional activity of the wild-type mouse relB promoter was compared, by luciferase reporter assay in D2SC1 cells, with that of a mouse relB promoter construct in which the VDRE was mutated to the sequence of the mOPN promoter VDRE. The result is expressed as mean ± SD luciferase activity from triplicate samples for each condition. † = P < 0.05 compared to untreated control; ‡ = p < 0.05 compared to D₃ analog alone and VDR alone.

Figure 3

A. The results of luciferase reporter assays in a panel of cell lines are shown. The effect of 10⁻¹⁰ M D₃ analog treatment, VDR over-expression (VDR), or both (D3 Analog + VDR) on promoter activity of the wild-type mouse relB promoter was compared in the following cell lines: (i) D2SC1 (mouse DC), (ii) HEK293 (human embryonic kidney), (iii) ROS17/2.8 (rat osteoblast), (iv) SW480 (human colonic carcinoma), and (v) RAW (mouse macrophage). The results for each cell line are expressed as the % of untreated (control) cells for luciferase activity (mean ± SD of triplicate samples for each condition). † = P < 0.05 compared to untreated control; ‡ = p < 0.05 compared to D₃ analog alone and VDR alone. B. Results of a Western blot for VDR and β-actin are shown for D2SC1 and ROS17.2.8 (ROS) cells following 24 hour culture in the absence (Ctrl) or presence (D3) of 10⁻⁷ M 1α,25(OH)₂D₃. Cultures were carried out using medium supplemented by charcoal adsorbed fetal calf serum to remove the potential influence of residual 1α,25(OH)₂D₃ in control conditions.

Figure 4

A. Schematic representations of a negative regulatory complex centered on the identified VDRE with the mouse relB promoter are shown for steady state (immature) DCs and DCs subject to a maturational stimulus. The components that have been shown experimentally to be constitutively or variably present by ChIP assay are VDR, RXRα, SMRT, HDAC1 and HDAC3. B. Schematic representations of two models for relB transcriptional suppression in DCs exposed to exogenous VDR ligand (1α,25(OH)₂D₃ or D₃ analog).

Figure 4

A. Schematic representations of a negative regulatory complex centered on the identified VDRE with the mouse relB promoter are shown for steady state (immature) DCs and DCs subject to a maturational stimulus. The components that have been shown experimentally to be constitutively or variably present by ChIP assay are VDR, RXRα, SMRT, HDAC1 and HDAC3. B. Schematic representations of two models for relB transcriptional suppression in DCs exposed to exogenous VDR ligand (1α,25(OH)₂D₃ or D₃ analog).