Nuclear factor-κB binding motifs specify Toll-like receptor-induced gene repression through an inducible repressosome

Abstract

Sustained Toll-like receptor (TLR) stimulation continuously activates antimicrobial genes but paradoxically represses inflammatory genes. This phenomenon, termed TLR tolerance, is essential for preventing fatal inflammatory conditions such as sepsis, but its underlying mechanisms are unclear. We report here that NF-κB binding nucleic acids of gene promoters are tolerogenic motifs, which selectively recruit an NcoR-Hdac3-deacetylated-p50 repressosome to inflammatory genes. Genome-wide analyses of TLR4-induced genes revealed that NF-κB motifs were the only regulatory elements significantly enriched in tolerizable genes. Mutating the NF-κB motifs of tolerizable genes converted them into nontolerizable ones, whereas inserting NF-κB binding motifs into nontolerizable genes conferred the tolerance. Although NF-κB p50 was essential for assembling the repressosome, genetic disruption of the NcoR-Hdac3 interaction alone was sufficient to completely abolish TLR4 tolerance and to render mice vulnerable to sepsis. Thus, the specificity of TLR tolerance is dictated by evolutionally conserved nucleic acid motifs that bound by NF-κB and the NcoR repressosome.

Fig. 1.

NF-κB binding motifs of gene promoters specify LPS-induced tolerance. (A) Motif-based bioinformatic analysis of 508 LPS-induced genes (371 T and 137 NT genes) from the murine genome (1, 16) using the oPOSSUM program reveals specific enrichment of NF-κB binding motifs in tolerizable (red) but not nontolerizable (green) genes. Putative transcription factor-binding motifs located between 5000 bp upstream and 5000 bp downstream of the transcription start site of each gene were analyzed. Each data point represents a transcription factor-binding motif; only those that are significantly enriched in a class of genes are labeled (Upper Left quadrant). REL or NF-κB represents the binding site for the NF-κB/Rel family of transcription factors. IRF represents a binding site for the IRF family of transcription factors. (B–D) NF-κB motif insertion into the promoter constructs of nontolerizable genes converts them into tolerizable ones. The RAW246.7 macrophages were transiently transfected with the promoter reporter plasmids as indicated. They were either (i) left untreated for 36 h (naïve, N), or (ii) rested for 24 h and treated with 100 ng/mL LPS (L) for an additional 12 h (NL), or (iii) treated with 100 ng/mL LPS for 24 h and then rested in fresh media for an additional 12 h (tolerized, T), or (iv) treated with 100 ng/mL LPS for 24 h and restimulated with 100 ng/mL LPS for another 12 h (TL). The luciferase activities were quantified at the end of the culture as described in Materials and Methods. 2κB and 4κB, two and four identical NF-κB binding motifs, were inserted, respectively; 4AP1, 4IRF3, and 4SP1, four identical binding sites of AP1, IRF3, and SP1, respectively, were inserted. Numbers in parentheses of the gene construct names indicate the location where the NF-κB binding motifs were inserted. (E) Schematic illustration of the LPS stimulation time line. All luciferase assays were performed 12 h after the last LPS stimulation. i.e., at the +12 h. (F) Mutations of NF-κB binding motifs of tolerizable genes abolish tolerance. Mutations were introduced into an NF-κB binding motif (at −95) of the Il23p19 promoter region (−1,180/+110), and deletion of −218/−50 nucleotides of the Tnf promoter region (−512/+61) removed its NF-κB motifs. Wild-type (WT) and mutated (κm) promoter constructs were tested as in B above. Error bars represent SD of the mean. Experiments were repeated at least three times with similar results.

Fig. 2.

Selective binding of NF-κB, NcoR, and histone deacetylases to tolerizable genes and the lack of LPS-induced gene derepression in tolerized cells. Naïve or tolerized (T) bone-marrow–derived macrophages were stimulated with 100 ng/mL LPS for 0, 1, or 3 h as indicated. Cells were then fixed and chromatin IP was performed for the indicated genes and factors as described in Materials and Methods. Tolerance was induced by pretreating cells with 100 ng/mL LPS for 24 h. T0, T1, and T3 indicate tolerized cells restimulated with LPS for 0, 1, and 3 h, respectively. Experiments were repeated three times with similar results.

Fig. 3.

Nfkb1 null mutation prevents NcoR and histone deacetylase binding to tolerizable genes, while having no effect on nontolerizable genes. Bone-marrow–derived macrophages from WT and Nfkb1^−/− mice (n = 4) were either left untreated (0) or treated with 100 ng/mL LPS for up to 24 h as indicated. Cells were then fixed and chromatin IP was performed for the indicated genes and factors as described in Materials and Methods. Error bars depict SD of the mean. Experiments were repeated three times with similar results.

Fig. 4.

A stable TLR4-induced repressosome mediates gene-specific tolerance and prevents sepsis. (A) Visualizing the TLR4-induced repressosome by re-ChIP. Naïve (NL) and tolerized (TL) bone-marrow–derived macrophages were either left untreated (0) or treated with 100 ng/mL LPS for 1 h as indicated. Tolerization was performed by pretreating cells with 100 ng/mL LPS for 24 h before the experiment. Cells were then fixed, and ChIP was performed using anti-p50, anti-NcoR, anti-Hdac3, or control IgG (control) as indicated. Re-ChIP was then carried out using the precipitates from the first round of ChIP for the indicated genes (p19 and Tnf) and factors (p50, NcoR, and Hdac3). (B) Detecting the TLR-induced repressor complex by coimmunoprecipitation. Bone-marrow–derived macrophages were either left untreated or treated with 100 ng/mL LPS for 24 h as indicated. Nuclear extracts were prepared and co-IP was performed using anti-NcoR, anti-p50, or control IgG. Western blots were performed using antibodies for the indicated proteins for both immunoprecipitates and total nuclear preparations (lysate). (C) Reduced p50 acetylation in tolerized cells. Bone-marrow–derived macrophages from WT, NcoR DADm, and Hdac3^−/− mice were either left untreated or treated with 100 ng/mL LPS for 24 h to induce tolerance. Nuclear extracts were prepared and used for IP with anti-p50. The precipitated p50 was then subjected to electrophoresis and Western blot using an antiacetyllysine antibody (Ac-Lys) or anti-p50. (D) Lack of LPS tolerance in NcoR DADm macrophages. Bone-marrow–derived macrophages of the indicated genotypes were either (i) left untreated for 26 h (naïve, N), or (ii) rested for 24 h and treated with 100 ng/mL LPS for an additional 2 h (NL), or (iii) treated with 100 ng/mL LPS for 24 h and then rested in fresh media for an additional 2 h (tolerized, T), or (iv) treated with 100 ng/mL LPS for 24 h and restimulated with 100 ng/mL LPS for another 2 h (TL). TNFα and IL-6 mRNA levels were determined by real-time PCR. Error bars represent SD of the mean. (E) NcoR DAD mutation renders mice hypersensitive to septic shock. Sex- and age-matched WT (n = 9) and NcoR DADm mice (n = 8) were injected intraperitoneally with LPS (30 mg/kg) on day 0 and their survival rates recorded. The difference between the two groups was statistically significant (P < 0.0001). (F) Sera from the two groups of mice as treated in E were collected 36 h after the LPS injection. Concentrations of IL-6, TNFα, and IL-12p40 in the sera were determined by ELISA. Error bars depict SD of the mean. Experiments were repeated at least three times with similar results.