Heme oxygenase-1 couples activation of mitochondrial biogenesis to anti-inflammatory cytokine expression

Abstract

The induction of heme oxygenase-1 (HO-1; Hmox1) by inflammation, for instance in sepsis, is associated both with an anti-inflammatory response and with mitochondrial biogenesis. Here, we tested the idea that HO-1, acting through the Nfe2l2 (Nrf2) transcription factor, links anti-inflammatory cytokine expression to activation of mitochondrial biogenesis. HO-1 induction after LPS stimulated anti-inflammatory IL-10 and IL-1 receptor antagonist (IL-1Ra) expression in mouse liver, human HepG2 cells, and mouse J774.1 macrophages but blunted tumor necrosis factor-α expression. This was accompanied by nuclear Nfe2l2 accumulation and led us to identify abundant Nfe2l2 and other mitochondrial biogenesis transcription factor binding sites in the promoter regions of IL10 and IL1Ra compared with pro-inflammatory genes regulated by NF-κΒ. Mechanistically, HO-1, through its CO product, enabled these transcription factors to bind the core IL10 and IL1Ra promoters, which for IL10 included Nfe2l2, nuclear respiratory factor (NRF)-2 (Gabpa), and MEF2, and for IL1Ra, included NRF-1 and MEF2. In cells, Hmox1 or Nfe2l2 RNA silencing prevented IL-10 and IL-1Ra up-regulation, and HO-1 induction failed post-LPS in Nfe2l2-silenced cells and post-sepsis in Nfe2l2(-/-) mice. Nfe2l2(-/-) mice compared with WT mice, showed more liver damage, higher mortality, and ineffective CO rescue in sepsis. Nfe2l2(-/-) mice in sepsis also generated higher hepatic TNF-α mRNA levels, lower NRF-1 and PGC-1α mRNA levels, and no enhancement of anti-inflammatory Il10, Socs3, or bcl-x(L) gene expression. These findings disclose a highly structured transcriptional network that couples mitochondrial biogenesis to counter-inflammation with major implications for immune suppression in sepsis.

Figures

FIGURE 1.

Increases in IL-10 and HO-1 mRNA and protein expression after LPS and CO exposures. A, increases in IL-10 mRNA levels in C57BL/6 mouse liver at 24 h after a 1-h exposure to CO at 250 or 500 ppm. B, increases in IL-10 mRNA levels in HepG2 cells 24 h after a 1-h exposure to DCM/CO (50 μm) or 24 h after LPS (10 ng); the combination of CO + LPS have an additive effect. Error bars indicate means ± S.E. of duplicate samples in triplicate experiments (*, p < 0.05 versus control; †, p < 0.05 versus control and LPS). C, Western blots of C57BL/6 mouse liver protein indicate dose increases in HO-1 and IL-10 levels at 24 h after 1-h exposures to 250 or 500 ppm CO. D, time course of hepatic HO-1 and IL-10 protein induction after 250 ppm CO. E, hepatic HO-1 and IL-10 induction after E. coli peritonitis in C57BL/6 mice. HO-1 is strongly up-regulated by 24 h; IL-10 is constitutive and increases at 24 h. TNF-α is shown for comparison.

FIGURE 2.

Hmox1 RNA silencing in HepG2 and J774.1 cells blocks IL-10 expression and counter-inflammatory activity. A, HepG2 cells treated with DCM/CO or LPS for 6 h show HO-1 and IL-10 up-regulation at 24 h. CO + LPS promotes HO-1 and IL-10 expression. Hmox1 silencing blocks the effect of CO + LPS on IL-10 expression, and incubation with IL-10 increases HO-1 protein levels. B, HepG2 cells stimulated with DCM/CO, LPS, or both show TFAM and NRF-1 up-regulation at 24 h. Hmox1 RNA silencing blocks TFAM and NRF-1 induction by CO + LPS. C, LPS-stimulated TNF-α and nitric oxide synthase-2 (NOS2) induction in HepG2 cells is inhibited by IL-10 (10 ng/ml), but suppression of TNF-α and nitric oxide synthase-2 by IL-10 is inhibited by siHmox1. D, murine J774.1 macrophages show HO-1 and IL-10 induction after DCM/CO and after LPS challenge. IL-10 induces HO-1 within 6 h and to full effect by 24 h. Hmox1 silencing blocks the DCM/CO effect on IL-10.

FIGURE 3.

HO-1/CO and LPS-induced sIL1Ra gene expression. A, C57BL/6 mouse hepatic sIL-1Ra mRNA and protein levels measured by quantitative RT-PCR and by Western analysis 24 h after 1 h of CO (250 or 500 ppm). B, HepG2 cell sIL-1Ra mRNA and protein levels after DCM/CO, LPS, or both. RNA silencing of Hmox1 or Nfe2l2 blocks up-regulation of mRNA and protein. C, in J774.1 macrophages, sIL-1Ra mRNA and protein expression increase after LPS and DCM/CO or both. IL-1Ra mRNA and protein levels are decreased by siHmox1 or by siNfe2l2. *, p < 0.05 versus control values for n = 3 studies.

FIGURE 4.

Nuclear MEF2, Nfe2l2, and NRF-2 (Gabpa) protein levels in mouse liver and HepG2 and J774.1 cells. These nuclear protein levels were checked in mouse liver after CO and in both cell lines after DCM/CO exposure. A, 1 h of CO breathing in mice increased hepatic MEF2, Nfe2l2, and NRF-2 nuclear protein relative to TATA-binding protein (TBP). B, in HepG2 cells, DCM/CO and LPS each increased nuclear MEF2, Nfe2l2, and NRF-2 protein levels, and the DCM/CO + LPS combination enhanced the effects. Hmox1 RNA silencing decreased nuclear MEF2, Nfe2l2 and NRF-2 protein levels, whereas Nfe2l2 silencing sharply decreased nuclear MEF2 and modestly decreased nuclear NRF-2. C, changes in J774.1 cell nuclear protein levels after DCM/CO and LPS and after siHmox1 or siNfe2l2 were similar to those for Hep2G cells for all three proteins.

FIGURE 5.

Nuclear Nfe2l2 and MEF2 transcription factor binding activity after HO-1 induction in cells. Binding activity in HepG2 cell nuclear extracts for Nfe2l2 (A) and MEF2 (B) after DCM/CO or LPS. Both exposures activate these transcription factors, but DCM/CO is more pronounced. The ability of DCM/CO to activate Nfe2l2 is partially blocked by p38 MAPK inhibition (SB20) and fully blocked by Akt inhibition (LY29), whereas either inhibitor blocks MEF2 activation by DCM/CO. The histograms represent means ± S.E. of duplicates in three experiments. *, p < 0.05 compared with control cells. C, sequential ChIP assay (ChIP-ReChIP) was performed on HepG2 cell genomic DNA with anti-RNA polymerase II (POL II) or with control IgG (not shown) and probed for interactions with Nfe2l2, MEF2, and NRF-2 at the human IL10 promoter. DCM/CO (and to a lesser extent LPS) recruits RNA polymerase II to MEF2, Nfe2l2, and NRF-2 binding sites on the IL10 promoter. D, ChIP assay of HepG2 cell nuclei pre- and post-DCM/CO (100 μm) or LPS (15 ng/ml) treatments. PCR was conducted with primer sets for the human IL10 or the IL1Ra promoter regions of interest. LPS and DCM/CO induced MEF2, Nfe2l2, and NRF-2 recruitment to IL10 promoter and MEF2 and NRF-1 to the IL1Ra promoter. Input lanes show PCR product derived from chromatin before immunoprecipitation (IP) to verify equal loading. GAPDH is a positive control. E, ChIP assay of the Il10 promoter using mouse liver genomic DNA and antibodies for MEF2, Nfe2l2, and NRF-2α (Gabpa). The Il10 promoter DNA did not co-precipitate with control IgG antibody. Nfe2l2 binding was not detectable at the Il10 promoter at basal conditions or after LPS but was detectable after CO exposure. MEF2 and Gabpa binding were weakly detected at the basal state but strongly recruited to the Il10 promoter after CO exposure.

FIGURE 6.

MEF2 regulation and nuclear translocation by Nfe2l2 in HepG2 cells after LPS or DCM/CO exposure. A, Western blot showing that Nfe2l2, but not scrambled siRNA decreases Nfe2l2 by >80% at 72 h. B, Western blot showing increased NRF-1 protein after DCM/CO and that NRF1 silencing leads to loss of MEF2 by 24 h. Tubulin is a reference protein. C, confocal microscopy in HepG2 cells treated with DCM/CO (100 μm) for 1 h or LPS (15 ng/ml) for 2 h, washed, and incubated for 4 h in fresh media. Some cells were transfected with Nfe2l2 siRNA. Cells were fixed, incubated with primary anti-HO-1 and anti-MEF2, and developed with secondary green Alexa Fluor 488 (HO-1) or red Alexa Fluor 595 (MEF2) antibodies. In control cells, HO-1 was restricted to cytoplasm, and no nuclear MEF2 was detected (panels A–C). DCM/CO (1-h exposure) caused robust nuclear MEF2 translocation by 2 h (panels D–F). After 2 h of LPS, nuclear MEF2 levels were also increased (panels G–I). Cells after siNfe2l2 still showed cytoplasmic HO-1 after DCM/CO, but MEF2 did not accumulate in the nuclei (panels J–L). Nuclei are stained with DAPI (blue), and images for MEF2 were taken at ×600. Scr, scrambled.

FIGURE 7.

Nuclear and cytoplasmic Nfe2l2 protein in HepG2 cells. A, DCM/CO and LPS increase nuclear and cytoplasmic Nfe2l2 protein levels within 4 h. TBP, TATA binding protein. B, the MG132 proteasome inhibitor further increased Nfe2l2 nuclear protein at 4 h, whereas inhibition of transcription with α-amanitin leads to a dramatic loss of Nfe2l2 nuclear protein. In C and D, HepG2 cells were treated with DCM/CO to increase Nfe2l2 nuclear translocation and IL10 and NQO1 transcription. MG132 clearly increases mRNA levels for IL-10 (C) and NQO1 (D) in response to DCM/CO, whereas α-amanitin blocks mRNA expression for both genes. *, p < 0.05 versus control; †, p < 0.05 versus control and DCM/CO.

FIGURE 8.

E. coli sepsis in wild-type and Nfe2l2−/− mice. A, increased early mortality in Nfe2l2−/− mice after peritoneal implantation of 108 cfu E. coli-containing fibrin clots compared with C57BL/6 (WT) mice (p < 0.05 by Chi square). CO therapy (250 ppm for 1 h/day on days 1–3) rescues WT but not Nefel2−/− mice (red lines). B, after 107 cfu E. coli, hepatic Hmox1 mRNA levels increase in WT but not Nfe2l2−/− mice 6 and 24 h after infection. For activation of mitochondrial biogenesis, NRF-1 mRNA is shown in C, and PGC-1α co-activator mRNA is shown in D. Compared with WT mice, Nfe2l2−/− mice also show significantly greater increases in TNF-α mRNA levels (E) and no responses in IL-10 (F), Bcl-xL (G), or Socs3 (H) mRNA over 24 h post-infection. **, p < 0.05 versus 0 h control values and versus null strain; *, p < 0.05 versus 0 h control values (n = 3–4 mice per group at each time).

FIGURE 9.

Working diagram of the HO-1/CO system and IL10 and IL1Ra activation during mitochondrial biogenesis. The two-cell model indicates the transcriptional integration of IL10 and IL1Ra gene expression with mitochondrial biogenesis through a redox-regulated cycle involving HO-1 and Nfe2l2. Nfe2l2 is involved both in Hmox1 activation and in the activation of NRF1 and IL10. NRF-1 activates MEF2 and IL1Ra, and MEF2 activates IL10. HO-1 activates NRF-2 (Gabpa) by an unknown mechanism that also contributes to IL10 activation. The transcription factors involved in mitochondrial biogenesis and anti-inflammation are highlighted in red. Activation of counter-inflammatory genes blocks the LPS/ Toll-like receptor 4 (TLR4)-dependent early-phase proinflammatory response mediated by TNF-α and IL-1β.