NADPH oxidase limits lipopolysaccharide-induced lung inflammation and injury in mice through reduction-oxidation regulation of NF-κB activity

Abstract

Although reactive oxygen species (ROS) produced by NADPH oxidase are known to regulate inflammatory responses, the impact of ROS on intracellular signaling pathways is incompletely understood. In these studies, we treated wild-type (WT) and p47(phox)-deficient mice with LPS to investigate mechanisms by which NADPH oxidase regulates signaling through the NF-κB pathway. After intratracheal instillation of LPS, ROS generation was impaired in p47(phox)(-/-) mice, whereas these mice had increased neutrophilic alveolitis and greater lung injury compared with WT controls. In mice interbred with transgenic NF-κB reporters (HIV-long terminal repeat/luciferase [HLL]), we found exaggerated LPS-induced NF-κB activation and increased expression of proinflammatory cytokines in lungs of p47(phox)(-/-)/HLL mice compared with controls. Both lung macrophages and bone marrow-derived macrophages (BMDMs) isolated from p47(phox)(-/-)/HLL mice showed enhanced LPS-stimulated NF-κB activity compared with controls. Although nuclear translocation of NF-κB proteins was similar between genotypes, EMSAs under nonreducing conditions showed increased DNA binding in p47(phox)(-/-)/HLL BMDMs, suggesting that ROS production reduces NF-κB binding to DNA without affecting nuclear translocation. Increased intracellular reduced glutathione/glutathione disulfide ratio and greater nuclear redox factor 1 (Ref-1) levels were present in p47(phox)(-/-)/HLL compared with WT BMDMs, pointing to NADPH oxidase modulating intracellular redox status in macrophages. Treatment with the Ref-1-specific inhibitor E3330 or hydrogen peroxide inhibited LPS-induced NF-κB activation in p47(phox)(-/-)/HLL BMDMs but not in WT/HLL cells. Consistent with these findings, small interfering RNA against Ref-1 selectively reduced NF-κB activity in LPS-treated p47(phox)(-/-)/HLL BMDMs. Together, these results indicate that NADPH oxidase limits LPS-induced NF-κB transcriptional activity through regulation of intracellular redox state.

Figure 1

Intratracheal LPS administration causes increased lung inflammation, injury, and cytokine levels in p47^phox-/- mice compared to WT mice. A) Bronchoalveolar lavage (BAL) inflammatory cells, B) BAL neutrophils, C) BAL protein concentration, and D) lung wet to dry ratio were measured at baseline, 4 hours and 24 hours following intratracheal LPS (3 μg/g body weight), E-F) H&E stained lung sections and lung injury score harvested 24 hours after LPS treatment, G) Representative immunohistochemistry for myeloperoxidase positive cells in lung sections 24 hours after LPS treatment, H) BAL cytokines were measured at 4 hours following LPS treatment. Results are presented as means ± SEM; n=4-7 per group, *=p<0.05.

Figure 2

LPS-induced ROS production in lungs inversely correlates with NF-κB activation. A) Bioluminescence imaging of ROS production was performed in WT and p47^phox-/- mice. After a single intratracheal injection of LPS (3μg/g), bioluminescence was measured at baseline and at 4 and 24 hours using L-012 as a luminescent probe. B) ROS production expressed as photon emission over the thorax. C) NF-κB dependent luciferase activity in lung homogenates of p47^phox-/- mice crossed with a NF-κB reporter line (HLL) and controls (WT/HLL) at baseline and at 4 and 24 hours after IT LPS. D) Luciferase activity in cells from NF-κB-luciferase reporter mice obtained from BAL. E) Luciferase activity in lung macrophages isolated by adherence from NF-κB-luciferase reporter mice at 4 hours after LPS treatment. Results are presented as means ± SEM; n=3-6 per group, *=p<0.05, **=p<0.001, respectively.

Figure 3

NADPH oxidase regulates LPS-induced ROS production, NF-κB activity, and pro-inflammatory cytokine expression in bone marrow-derived macrophages (BMDMs). A) Superoxide was measured in WT/HLL and p47^phox-/- /HLL BMDMs following LPS (0.2μg/ml) treatment, shown as relative light units (RLU). Results are presented as means ± SEM; n=5 per group, **=p<0.01. B) NF-κB-dependent luciferase activity was measured in BMDMs from p47^phox-/- /HLL compared to WT/HLL mice at baseline and up to 24 hours after stimulation with LPS (0.2 μg/ml) stimulation. Each point represents the mean ± SEM; n=9 per point, **=p<0.01. C-D) CXCL1 and TNFα mRNA levels were measured at 1 hour after LPS treatment. The relative mRNA expression was presented as 2Δ [Ct (housekeeping gene)-Ct (target gene)]. Each point represents the mean ± SEM; n=3-4 per group, *=p<0.05.

Figure 4

NADPH oxidase modulates LPS-induced NF-κB binding to DNA without altering nuclear translocation. A) Western blot for RelA (p65) compared to TATA box binding protein (TBP) in nuclear fractions of BMDMs from WT/HLL and p47^phox-/- /HLL mice after LPS (0.2 μg/ml) treatment. B) Densitometry of p65 bands normalized to TBP from 3 separate experiments. C and D) EMSA and density of NF-κB bands under reducing (C) and non-reducing (D) conditions in nuclear extracts from WT/HLL and p47^phox-/- /HLL BMDMs at 1 hour after LPS treatment. Specificity of binding is shown by addition of unlabeled oligonucleotide competitor.

Figure 5

NADPH oxidase alters the redox environment in macrophages. A) Redox pair (GSH/GSSG) ratio assessed by HPLC from whole cell lysates from WT/HLL and p47^phox-/- /HLL BMDMs at baseline and after LPS treatment. Results are mean ± SEM; n=4 per group, *=p<0.05, **=p<0.01, respectively. B-C) Western blots and densitometry from nuclear protein extracts showing levels of thioredoxin 1 (Trx1) and redox factor 1 (Ref-1) normalized by TBP in untreated BMDMs from WT/HLL and p47^phox-/- /HLL mice. D) Western blot densitometry from cytoplasmic protein extracts showing levels of Ref-1 normalized to β-actin in untreated BMDMs. Results are mean ± SEM; n=3-4 per group, *=p<0.05.

Figure 6

Blockade of redox factor-1 (Ref-1) or addition of hydrogen peroxide selectively inhibits LPS-induced NF-κB activation in macrophages from p47^phox-/- mice. A) NF-κB dependent luciferase activity in BMDMs from WT/HLL mice and p47^phox-/- /HLL mice at baseline and 4 hours after treatment with LPS (0.2μg/ml) and Ref-1 specific inhibitor E3330 (25μM). B) NF-κB dependent luciferase activity at baseline and 4 hours after treatment with LPS +/- H₂O₂ (5μM). C) BMDMs were transfected with Ref-1 (or non-targeting control) siRNA, the levels of Ref-1 in whole cell lysate were determined by western blot and quantified by densitometry. D) NF-κB dependent luciferase activity with siRNA-mediated knockdown of Ref-1. Results are mean ± SEM; n=3-6 per group, *=p<0.001.