Hypersensitivity of aryl hydrocarbon receptor-deficient mice to lipopolysaccharide-induced septic shock

Abstract

Aryl hydrocarbon receptor (AhR), a ligand-activated transcription factor, is known to mediate a wide variety of pharmacological and toxicological effects caused by polycyclic aromatic hydrocarbons. Recent studies have revealed that AhR is involved in the normal development and homeostasis of many organs. Here, we demonstrate that AhR knockout (AhR KO) mice are hypersensitive to lipopolysaccharide (LPS)-induced septic shock, mainly due to the dysfunction of their macrophages. In response to LPS, bone marrow-derived macrophages (BMDM) of AhR KO mice secreted an enhanced amount of interleukin-1beta (IL-1beta). Since the enhanced IL-1beta secretion was suppressed by supplementing Plasminogen activator inhibitor-2 (Pai-2) expression through transduction with Pai-2-expressing adenoviruses, reduced Pai-2 expression could be a cause of the increased IL-1beta secretion by AhR KO mouse BMDM. Analysis of gene expression revealed that AhR directly regulates the expression of Pai-2 through a mechanism involving NF-kappaB but not AhR nuclear translocator (Arnt), in an LPS-dependent manner. Together with the result that administration of the AhR ligand 3-methylcholanthrene partially protected mice with wild-type AhR from endotoxin-induced death, these results raise the possibility that an appropriate AhR ligand may be useful for treating patients with inflammatory disorders.

FIG. 1.

High susceptibility of AhR KO mice to LPS-induced endotoxin shock. (A) Survival of AhR WT and AhR KO mice (n = 10) after LPS challenge (20 mg/ml). (B) TNF-α, IL-6, IL-12p70, IL-1β, IL-18, and IFN-γ plasma levels 2 h after LPS challenge (20 mg/ml). Horizontal bars show the mean results. (C) Partial protection of AhR WT mice from septic shock by intraperitoneal injection of 3MC at 2 h before LPS challenge (30 mg/ml) and survival of corn oil-injected mice. AhR WT-oil, n = 29; AhR WT-3MC, n = 28; AhR KO-oil, n = 13; AhR KO-3MC, n = 13. *, P < 0.001; **, P = 0.001; ***, P < 0.005; ****, P = 0.002; NS, not significant.

FIG. 2.

LPS induces abnormal secretion of IL-1β by BMDM from AhR KO mice. (A) Survival of AhR^flox/− (AhR f/−; n = 15) and AhR^flox/−::LysM Cre (AhR f/−::cre; n = 19) mice after LPS challenge (25 mg/ml). (B) IL-1β and TNF-α levels in the culture supernatants of AhR WT and AhR KO BMDM 8 h after LPS stimulation (10 ng/ml) (n = 4). (C) Relative expression levels of IL-1β and TNF-α mRNA 4 h after LPS stimulation (10 ng/ml) of AhR WT and AhR KO BMDM. Gray and black bars show results with LPS; white bars show results for untreated cells. Error bars show standard deviations. *, P = 0.03; **, P < 0.001; NS, not significant.

FIG. 3.

Decreased Pai-2 expression is one of the causes of the increased IL-1β secretion by LPS-treated AhR KO BMDM. (A) Inhibition of IL-1β oversecretion from AhR KO BMDM by treatment with caspase-1 inhibitor (Z-YVAD-FMK) or caspase inhibitor (Z-VAD-FMK). (B) Relative expression levels of Pai-2 and Bcl-2 mRNA in AhR WT and AhR KO BMDM. (C) The effect of hPai-2 and hBcl-2 reconstitution on the LPS-induced secretion of IL-1β and TNF-α by AhR KO BMDM. BMDM from AhR KO mice were infected with the individual adenovirus (adeno) vectors and then washed and incubated for 24 h. IL-1β and TNF-α levels in the supernatants 8 h after LPS stimulation (n = 3) were determined by ELISA. (D) Assessment of hPai-2 and hBcl-2 mRNA expression in adenovirus (adeno) vector-infected BMDM by conventional RT-PCR. Error bars show standard deviations. *, P < 0.001; NS, not significant.

FIG. 4.

Arnt is not required for LPS-induced enhancement of Pai-2 expression. (A) Relative Pai-2 and AhR mRNA expression levels in AhR WT and AhR KO PEMs 4 h after treatment with (black or gray bars) or without (white bars) LPS (10 ng/ml). (B) Immunoblot analysis of Pai-2 and AhR expression in AhR WT and KO PEMs after a 16-h incubation with LPS (10 ng/ml). (C) Immunoblot analysis of Arnt in Arnt^flox/flox and Arnt^flox/flox::LysM Cre PEMs. (D) Relative Pai-2 mRNA expression levels 4 h after incubation of Arnt^flox/flox (black bar) and Arnt^flox/flox::LysM Cre (hatched bar) PEMs with LPS (10 ng/ml). (E) Left, relative expression levels of Mmp-8 and Nqo1 mRNA in Arnt^flox/flox (black bar) and Arnt^flox/flox::LysM Cre (hatched bar) PEMs treated with DMSO (white bars) or 3MC (black or hatched bar) (1 μM). Right, relative expression levels of Mmp-8 and Nqo1 mRNA in AhR WT (black bar) and AhR KO (gray bar) PEMs treated with DMSO (white bars) or 3MC (black or gray bar) (1 μM). (F) Survival of Arnt^flox/− (Arnt f/−; n = 7) and Arnt^flox/−::LysM Cre (Arnt f/−::cre; n = 12) mice after LPS challenge (25 mg/ml). Error bars show standard deviations. IB, immunoblot; +, present; −, absent; α, anti.

FIG. 5.

Recruitment of transcription factors necessary for LPS-induced Pai-2 expression. (A) LPS-induced luciferase expression from the Pai-2 (−2.7 kb) and Pai-2 (−0.8 kb) reporter genes. RAW 264.7 cells were transfected with each reporter gene, with and without pcDNA3-AhR (0 ng, 50 ng, 100 ng) and/or pcDNA3-p65 (1 ng). Values represent the means, normalized to Renilla luciferase activity (used as an internal control), ± standard deviations of the results of three independent experiments. The activities shown by the fourth and seventh pairs of bars were used as standards for normalizing the relative activities of the other conditions. (B) AhR WT and AhR KO PEMs were left untreated or were treated with LPS for 1 h. Cytoplasmic (Cyto) and nuclear (Nuc) extracts were immunoblotted with antibodies against AhR, p65, tubulin, and lamin. (C) Co-IP of AhR and p65. Whole-cell extracts from AhR WT PEMs were coimmunoprecipitated with anti-AhR antibody. Co-IPs and Western blotting were performed as described in Materials and Methods. (D) Relative expression levels of Cox-2 mRNA in AhR WT and AhR KO PEMs after 4 h of treatment with or without LPS (10 ng/ml). Bars are as labeled in panel A. Error bars show standard deviations. (E) Top, transcription factor binding sites in the Pai-2 and Cox-2 genes. Bottom, results of ChIP analyses of the Pai-2 and Cox-2 promoters. ChIP analyses were performed using antibodies to p65, AhR, and PolII in LPS-induced AhR WT and AhR KO PEMs. ChIP analyses and real-time PCRs were performed as described in Materials and Methods. Error bars show standard deviations. +, present; −, absent; α, anti; IgG, immunoglobulin G.

FIG. 6.

Nuclear localization, activation, and DNA binding domains of AhR are required for AhR-dependent Pai-2 expression. (A) Immunoblot analysis of full-length AhR or mutants in LacZ or AhR transformant ANA-1 cells. Paired lanes labeled 1 and 2 show results from experiments using two independent transformants. (B) Relative expression levels of Pai-2 mRNA in ANA-1 cells transfected with LacZ or full-length AhR or mutants. Bars show quantification of the results in the 12 lanes in panel A; error bars show standard deviations. *, P < 0.001; NS, not significant. (C) Interaction of p65 and AhR mutants. Co-IP of p65 and full-length AhR or mutants expressed in 293T cells, using anti-p65 antibody. AhR FL (full-length) comprises amino acids 1 to 805, AhR ΔC comprises amino acids 1 to 544, and AhR CA comprises amino acids 1 to 276 and 419 to 805; in AhR Y9F, Y9 was mutated to F; and in AhR NLSm 37R, 38H, and 39R were mutated to A, G, and S, respectively. IB, immunoblot; α, anti; +, present.