Skip to main page content
Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2013 Aug 30;288(35):25362-74.
doi: 10.1074/jbc.M113.496281. Epub 2013 Jul 12.

Histone Deacetylase 7 Promotes Toll-like Receptor 4-dependent Proinflammatory Gene Expression in Macrophages

Affiliations
Free PMC article

Histone Deacetylase 7 Promotes Toll-like Receptor 4-dependent Proinflammatory Gene Expression in Macrophages

Melanie R Shakespear et al. J Biol Chem. .
Free PMC article

Abstract

Broad-spectrum inhibitors of histone deacetylases (HDACs) constrain Toll-like receptor (TLR)-inducible production of key proinflammatory mediators. Here we investigated HDAC-dependent inflammatory responses in mouse macrophages. Of the classical Hdacs, Hdac7 was expressed at elevated levels in inflammatory macrophages (thioglycollate-elicited peritoneal macrophages) as compared with bone marrow-derived macrophages and the RAW264 cell line. Overexpression of a specific, alternatively spliced isoform of Hdac7 lacking the N-terminal 22 amino acids (Hdac7-u), but not the Refseq Hdac7 (Hdac7-s), promoted LPS-inducible expression of Hdac-dependent genes (Edn1, Il-12p40, and Il-6) in RAW264 cells. A novel class IIa-selective HDAC inhibitor reduced recombinant human HDAC7 enzyme activity as well as TLR-induced production of inflammatory mediators in thioglycollate-elicited peritoneal macrophages. Both LPS and Hdac7-u up-regulated the activity of the Edn1 promoter in an HDAC-dependent fashion in RAW264 cells. A hypoxia-inducible factor (HIF) 1 binding site in this promoter was required for HDAC-dependent TLR-inducible promoter activity and for Hdac7- and HIF-1α-mediated trans-activation. Coimmunoprecipitation assays showed that both Hdac7-u and Hdac7-s interacted with HIF-1α, whereas only Hdac7-s interacted with the transcriptional repressor CtBP1. Thus, Hdac7-u positively regulates HIF-1α-dependent TLR signaling in macrophages, whereas an interaction with CtBP1 likely prevents Hdac7-s from exerting this effect. Hdac7 may represent a potential inflammatory disease target.

Keywords: Histone Deacetylase; Hypoxia-inducible Factor (HIF); Inflammation; Macrophages; Toll-like Receptor (TLR).

Figures

FIGURE 1.
FIGURE 1.
Hdac7 expression is elevated in inflammatory macrophages. A, quantitative PCR primers detecting the classical Hdacs were used to quantify mRNA levels relative to Hprt in BMMs (black bars), TEPMs (white bars), and RAW264 cells (gray bars). Data (mean ± S.E. of five independent cell preparations) are shown relative to BMMs for each gene. B, protein lysates prepared in 2% SDS from TEPMs, BMMs, and RAW264 cells were separated by SDS-PAGE and probed for Hdac7, Hdac4, Hdac1, and Gapdh. C, quantification of Hdac7 protein levels relative to Gapdh in TEPMs, BMMs, and RAW264 cells (n = 5, p < 0.001). D, primers that detect the extra exon in Hdac7-u were used to quantitate expression of Hdac7-u relative to Hprt in TEPMs, BMMs, and RAW264 cells. Data show the mean ± S.E. for five independent cell preparations. ANOVA with Tukey's test was used to compare all samples. **, p < 0.01).
FIGURE 2.
FIGURE 2.
Overexpression of Hdac7-u, but not Hdac7-s, in RAW264 cells amplifies the TLR4-inducible expression of a subset of inflammatory genes. Independent pools of RAW264 cells stably transfected with either empty vector (n = 4), Hdac7-u (n = 3), or Hdac7-s (n = 3) were treated with LPS (100 ng/ml) for 4 h. Total Hdac7 mRNA levels were determined in the different pools (A), as was LPS-regulated gene expression for Edn1 (B), IL-12p40 (C), IL-6 (D), IL-1β (E), iNOS (F), Ccl7 (G), and Tnfα (H). Data show the mean ± S.E. of fold induction in response to LPS across the independent pools of stable cell lines. ANOVA with Tukey's test was used. ***, p < 0.001.
FIGURE 3.
FIGURE 3.
Hdac7-dependent amplification of TLR4-inducible gene expression and cytokine release in macrophages. Time course of LPS-inducible Edn1 (A), Il-12p40 (B), Il-6 (C), and iNOS (D) mRNA expression in RAW264 cells overexpressing empty vector (RAW-pEF6, solid line) or Hdac7-u (RAW-Hdac7-u, dotted line). Data (mean ± S.D. of technical triplicates) are representative of two independent experiments. Equal numbers of RAW-pEF6 (open bars) and RAW-Hdac7-u (filled bars) cells were stimulated with LPS for 12 or 24 h, and culture supernatants were analyzed for IL-12p40 (E), IL-6 (F), and TNFα (G). Data (relative to RAW-Hdac7-u at 24 h LPS) are combined from three independent experiments (mean ± S.E.) (Student's t test and one-sample Student's t test for 12- and 24-h data, respectively. *, p < 0.05; **, p < 0.01; ***, p < 0.001.
FIGURE 4.
FIGURE 4.
A class IIa HDAC inhibitor inhibits TLR-inducible inflammatory mediator production from primary mouse macrophages. A, inhibition of recombinant hHDAC7 enzyme activity with compound 6. M, molar. B, TEPMs were treated with HDAC inhibitor (shown in micromolar) or vehicle control (Con) for 4 h. Protein lysates in 2% SDS were analyzed by immunoblotting to detect acetylated tubulin (acTub), acetylated histone H3 (acH3), and Gapdh as a loading control. Data are representative of three independent experiments. C–F, TEPMs were treated with LPS (100 ng/ml), and the indicated concentration (shown in micromolar) of compound 6 (c6), TSA, or appropriate vehicle (DMSO (D) for c6 and EtOH (Et) for TSA) for 8 h. Levels of secreted ET-1 (C), IL-12p40 (D), IL-6 (E), and TNFα (F) in culture supernatants were determined by ELISA. Data (mean ± S.E.) are combined from four independent experiments and are displayed relative to the LPS + DMSO-treated sample. ANOVA with Dunnett's multiple comparison test was used to compare the c6- and TSA-treated samples to the relevant vehicle control. *, p < 0.05; **, p < 0.01; ***, p < 0.001.
FIGURE 5.
FIGURE 5.
Hdac7 activates the Edn1 promoter in an Hdac-dependent fashion in mouse macrophages. A, RAW264 cells were transiently transfected with an Edn1 promoter construct driving luciferase, the empty vector pGL2B, or the LPS-responsive positive control pGL2C (Con). After 20 h, cells were treated with LPS (100 ng/ml) or LPS + TSA (500 nm) for 8 h. Luciferase activity is shown relative to the control. Data (mean ± S.E., ANOVA and Tukey-Kramer test) are combined from three independent experiments. *, p < 0.05; ***, p < 0.001. B, RAW264 cells were transfected with Edn1 promoter alone or with Edn1 plus Hdac7-u or Hdac7-s. After 20 h, cells were treated with LPS for 8 h, after which luciferase activity was analyzed. Data (mean ± S.E. for three independent experiments) are shown relative to the unstimulated control. *, p < 0.05, Student's t test. C, RAW264 cells were transfected with Edn1 promoter alone (control), Edn1 plus Hdac7-u, or Edn1 plus the N-terminal region of Hdac7-u, Hdac7 (N-term, amino acids 23–504). Luciferase activity was measured after 8-h stimulation with LPS. Data (mean + range of duplicate transfections within the experiment) are displayed relative to the Edn1 promoter alone and are representative of three independent experiments. D, RAW264 cells were transfected with Edn1 plus empty vector (open bars) or Edn1 plus Hdac7-u (filled bars) and treated with EtOH (vehicle control), LPS, TSA, or LPS + TSA for 8 h. Luciferase activity was measured and is shown relative to the vehicle control (mean ± S.E. for three independent experiments). E, experiments were performed as for D, except that a concentration range of compound 6 (in micromolar) was examined. Data (mean ± S.E. for three independent experiments) are shown relative to the LPS-treated Edn1 promoter plus a Hdac7-u sample. ANOVA with Dunnett's multiple comparison was used to compare LPS alone to LPS + compound 6 for either the Edn1 promoter or the Edn1 promoter + Hdac7-u groups. *, p < 0.05; **, p < 0.01; ***, p < 0.001. F, RAW264 cells were transiently transfected with the Edn1 promoter construct plus class IIa Hdac expression constructs or an empty vector (control). After 20 h, transfected cells were treated for 8 h with LPS (filled bars) or left untreated (open bars), after which cell lysates were immunoblotted (IB) for the V5 tag of the ectopically expressed Hdacs. Data are representative of two independent experiments. G, experiments were performed as above, except that luciferase activity was monitored. Pooled data from five independent experiments (mean ± S.E.) are shown relative to the Edn1 promoter alone (Con), and ANOVA with Dunnett's multiple comparison test was used to compare the Hdac expression constructs to the relevant control (control - LPS or control + LPS). **, p < 0.01.
FIGURE 6.
FIGURE 6.
Amplification of TLR4 responses by Hdac7 involves HIF-1α. A, schematic diagram of the HIF-1 binding site in the Edn1 promoter and the three nucleotide residues mutated to create the Edn1-ΔHIF promoter construct (37). Luc, luciferase. B, RAW264 cells were transiently transfected with the Edn1 (wild-type) or Edn1-ΔHIF promoter constructs, with or without an Hdac7-u expression construct and treated with LPS for 8 h. Data (relative to the Edn1 promoter alone) are the mean + range of duplicate transfections and are representative of two independent experiments. C, RAW264 cells were transfected with Edn1 or Edn1-ΔHIF promoter constructs with or without an HIF-1α expression construct and were treated with LPS for 8 h. Promoter activity was assessed by luciferase assay. Data (mean ± S.E.) are combined from three independent experiments and are shown relative to the Edn1 promoter untreated control. ANOVA with Dunnett's multiple comparison test was used. *, p < 0.05; **, p < 0.01). D, the Edn1 promoter construct was transfected into RAW264 cells with either an HIF-1α expression construct or empty vector. pGL-2B was also included as a negative control. Cells were treated with EtOH (vehicle control), LPS (100 ng/ml), TSA (500 nm), or LPS + TSA. Data (average of duplicate transfections + range) are representative of two independent experiments and are displayed relative to the Edn1 promoter alone.
FIGURE 7.
FIGURE 7.
LPS-inducible HIF-1α expression in macrophages requires HDAC activity. A, RAW264 cells stably expressing hHIF-1α-V5 were treated with LPS or LPS + TSA for 1, 2, or 4 h. hHIF-1α was detected by Western blot analysis using an anti-v5 antibody, and the activity of TSA was confirmed by monitoring acetylated histone H3 (ac-H3). Gapdh levels are shown as a loading control. Data are representative of three independent experiments. veh, vehicle. B, RAW-HIF-1α-V5 cells were treated as in A, and mRNA levels of ectopically expressed HIF-1α were determined by quantitative PCR. Data (mean ± S.E.) are combined from three independent experiments and are displayed as expression relative to untreated control cells. ANOVA with Bonferroni's multiple comparison test was used. *, p < 0.05. C, RAW264 cells stably expressing hHIF-1α-V5 were treated with LPS (100 ng/ml), LPS + DMSO, LPS + compound 6 (c6, 100 μm), and LPS + TSA (0.1 μm) or were left untreated (Unstim.) for 2 h. HIF-1α-protein levels in whole cell lysates were assessed by immunoblotting. Data are representative of three independent experiments. Con, control.
FIGURE 8.
FIGURE 8.
Hdac7 and HIF-1α interact and synergize. A, RAW264 cells were transfected with the Edn1 promoter construct alone (control), the Edn1 promoter construct plus 1 μg (suboptimal) of HIF-1α expression construct, the Edn1 promoter construct plus 2 μg (suboptimal) of Hdac7-u expression construct, or the Edn1 promoter construct plus HIF-1α and Hdac7-u. Cells were treated with LPS (filled bars) for 8 h or were left untreated (open bars) before analysis of luciferase activity. Data (mean + range of duplicate transfections) are representative of two independent experiments and are displayed relative to the Edn1 promoter alone (control). B, both Hdac7-u and Hdac7-s interact with HIF-1α. Coimmunoprecipitation (IP) experiments were performed in HEK293 cells using Hdac-FLAG expression constructs as bait. Immunoprecipitated HIF-1α was detected by anti-V5 immunoblotting (IB). Data are representative of three independent experiments. C, HEK293 cells were cotransfected with CtBP1-FLAG and either V5 empty vector (EV) or V5-tagged Hdac7-u, Hdac7-s, Hdac7-C-term (Cterm), or Fam96a (irrelevant control protein). Immunoprecipitation was performed with an anti-V5 antibody, and immunoprecipitated CtBP1-FLAG was detected with an anti-FLAG antibody. Data are representative of two independent experiments.
FIGURE 9.
FIGURE 9.
Proposed model of Hdac7-u involvement in TLR4 responses. LPS signaling up-regulates HIF-1α mRNA and protein expression in macrophages. The early response is dependent upon HDAC activity (but is independent of class IIa Hdacs), whereas the later response is HDAC-independent. Both Hdac7-u and Hdac7-s can interact with HIF-1α, but an interaction between CtBP1 and Hdac7-s prevents this isoform from promoting HIF-1α-dependent transcriptional responses. In contrast, Hdac7-u promotes HIF-1α-dependent expression of Edn1 as well as coregulated TLR4 target genes.

Similar articles

See all similar articles

Cited by 28 articles

See all "Cited by" articles

Publication types

MeSH terms

Feedback