- Email: [email protected]
Carbon monoxide mediates heme oxygenase 1 induction via Nrf2 activation in hepatoma cells
BBRC Biochemical and Biophysical Research Communications 343 (2006) 965–972 www.elsevier.com/locate/ybbrc
Carbon monoxide mediates heme oxygenase 1 induction via Nrf2 activation in hepatoma cells Bok-Soo Lee a, JungHee Heo a, Yong-Man Kim b, Sang Moo Shim a, Hyun-Ock Pae a, Young-Myeong Kim c, Hun-Taeg Chung a,* a
Medicinal Resources Research Institute, Department of Microbiology and Immunology, Wonkwang University School of Medicine, Iksan, Chonbuk 570-749, South Korea b FCB-Pharmicell Co., Ltd, Stem Cell Therapy Institute, Jungwon-gu, Sungnam-si, Kyungki-do 462-120, South Korea c Department of Molecular and Cellular Biochemistry, College of Medicine and Vascular System Research Center, Kangwon National University, Chunchon, Kangwon-do, South Korea Received 22 February 2006
Abstract Carbon monoxide (CO) and nitric oxide (NO) are two gas molecules which have cytoprotective functions against oxidative stress and inflammatory responses in many cell types. Currently, it is known that NO produced by nitric oxide synthase (NOS) induces heme oxygenase 1 (HO1) expression and CO produced by the HO1 inhibits inducible NOS expression. Here, we first show CO-mediated HO1 induction and its possible mechanism in human hepatocytes. Exposure of HepG2 cells or primary hepatocytes to CO resulted in dramatic induction of HO1 in dose- and time-dependent manner. The CO-mediated HO1 induction was abolished by MAP kinase inhibitors (MAPKs) but not affected by inhibitors of PI3 kinase or NF-jB. In addition, CO induced the nuclear translocation and accumulation of Nrf2, which suppressed by MAPKs inhibitors. Taken together, we suggest that CO induces Nrf2 activation via MAPKs signaling pathways, thereby resulting in HO1 expression in HepG2 cells. 2006 Elsevier Inc. All rights reserved. Keywords: Carbon monoxide; Tricarbonyl dichlororuthenium (II) dimmer (RuCO); Heme oxygenase 1; Nitric oxide/inducible NO synthase; Mitogenactivated protein kinases; NF-E2-related factor; Anti-oxidant response element
HO1 and its metabolic products play important regulatory roles in both physiological and pathological status [1]. HO1 is induced in many cell types and catalyzes the ratelimiting reaction, in the catabolism of heme, yielding final products in vivo: CO, bilirubin, and ferrous iron. In general, CO is considered to have pleiotropic cytoprotective activities [2–5]. CO protects cells/tissues from damages by free radicals or oxidative stress [6,7]. It also shows antiinflammatory, anti-proliferative, and anti-apoptotic effects by modulating related gene expressions or enzymatic activities [4,8,9]. HO1/CO may have a dual role in tissue pathology since high levels of HO1 are frequently detected in
*
Corresponding author. Fax: +82 63 851 5066. E-mail address: [email protected] (H.-T. Chung).
0006-291X/$ - see front matter 2006 Elsevier Inc. All rights reserved. doi:10.1016/j.bbrc.2006.03.058
various pathological states and generally in states of cellular oxidative stress [2,10]. It may not be therapeutic in all instances because there are some evidences showing that too much HO1/CO is a perpetrator of tissue injury [1,2]. Therefore, deciphering the underlying molecular mechanism that controls endogenous level of CO would be valuable to fine-tune CO production and minimize the potential detrimental situations under different pathological status, which eventually takes advantage of using CO as an effective therapeutic agent. There are many reports regarding the mechanism of HO1 expression induced by hypoxia, metal ions or inflammatory cytokines [11,12] and CO-mediated gene modulation including iNOS and several inflammatory cytokine genes [13,14]. Three major groups of mitogen-activated protein kinases (MAPKs), which are the extracellular
966
B.-S. Lee et al. / Biochemical and Biophysical Research Communications 343 (2006) 965–972
signal-regulated kinase (ERK), the jun NH2-terminal kinase (JNK), and the p38 MAP kinase (p38 MAPK), have been described for HO1 gene regulation [11,15,16]. Maine’s group added up that biliverdin reductase (BVR) functions as a transcriptional factor in HO1 gene expression [17]. The HO1 promoter contains several transcriptional regulatory elements that respond to redox-sensitive transcription factors, including NF-E2-related factor 2 (Nrf2) [18]. Nrf2 is a member of the basic leucine zipper family of transcription factors. There are some reports that Nrf2 resides in the cytoplasm bound to its inhibitor protein, Keap1, and translocates to the nucleus after stimulation and binds to a DNA sequence in target genes as the anti-oxidant response element (ARE)/stress response element (stRE) [12,19,20]. Other reports demonstrated that Nrf2 is located in nucleus together with Bach1, a negative transcription repressor, which keeps away most of Nrf2 to DNA binding that leads to basal level of HO1 transcription [21]. However, there is no report so far whether CO could directly affect HO1 gene expression. In the present study, we investigated the mechanism by which exogenous/endogenous CO induces HO1 expression in HepG2 cells. We demonstrated here that CO directly upregulates HO1 gene transcription by enhancing nuclear translocation and accumulation of Nrf2, presumably increasing Nrf2 binding to ARE site in HO1 promoter. We also showed that the activation of MAPKs is important for the CO-mediated Nrf2 nuclear translocation.
sequences of pGL3/HO1/4384-Luci and pARE-Luci were confirmed and verified the presence of the correct sequence and the absence of any other nucleotide changes by DNA sequencing. Administration of CO gas. Saturated stock solutions were prepared in buffer containing 140 mM NaCl, 5 mM KCl, and 20 mM Hepes, pH 7.3, as described [22]. CO stock solutions were freshly prepared for every experiment. After HepG2 cells were treated with various concentrations of CO gas for 6 h, HO1 expression was determined by immunoblot assay. Western blot analysis. Cell pellets were lysed in 1· sample buffer [50 mM Tris–HCl (pH 7.4), 150 mM NaCl, 0.1% SDS, 2 mM b-mercaptoethanol, 1 mM DTT, BPB, and xylene cyanol]. The cell lysates were electrophoresed on a SDS–polyacrylamide gel and proteins were transferred to polyvinylidene difluoride (PVDF) membranes. Immunoblotting was performed according to the manufacturer’s instructions and the blots were washed and developed with supersignal enhanced chemiluminescence (ECL) substrate solution (Pierce, IL). The signals were then visualized with X-ray film. Measurement of promoter activity. The cells were transiently transfected with the promoter constructs using the transfection reagent Fugene 6 (Roche; Mannheim, Germany). After harvest, cells were lysed in reporter lysis buffer (Promega; Madison, WI). Twenty microliters of cell extract was mixed with 100 lL of the luciferase assay reagent and the emitted light intensity was measured using the luminometer AutoLumat LB953 (EG and G Berthold; Bad Wildbad, Germany). Fold induction was calculated as intensity value from each experimental group divided by value from control group after normalization of transfection efficiency by b-gal assay. Immunostaining and confocal microscopy. For localization of Nrf2, the HepG2 cells were grown on Lab-Tek II chamber slides and treated as described in figure legends. Cells were fixed with ice-cold acetone for 10 min, blocked with 10% donkey serum for 30 min, and stained with antirabbit Nrf2. After washing, cells were stained with Alexa 488-conjugated donkey anti-rabbit antibody. Cells were then washed and mounted in mounting medium (Vector, Burlingame, CA). Confocal microscopy was performed using a FLUOVIEW FV1000 (Olympus, PA). Images were collected at 512 · 512 pixel resolution and overlayed with DIC image.
Materials and methods Cell culture and reagents. Human hepatocyte cell line, HepG2, was purchased from ATCC (Manassas, VA) and maintained with RPMI1640 supplemented with 10% fetal bovine serum (FBS), 1% penicillin/streptomycin. Tricarbonyl dichlororuthenium (II) dimer (RuCO), RuCl2, bilirubin, biliverdin, ferrous citrate (FeC), and sodium nitroprusside (SNP) were purchased from Sigma (St. Louis, MO). Cobalt protoporphyrin (CoPP) and tin protoporphyrin (SnPP) were from Porphyrin Products (Logan, UT). SB203508, SP600125, U0125, Bay-11, and Wortmannin, NG-monomethyl-L-arginine (NGMMA), and antibody against HO1 was obtained from Calbiochem (La Jolla, CA). Antibodies to Nrf2, b-actin, and HRP-conjugated secondary antibodies were from Santa Cruz Biotechnology (Santa Cruz, CA). Alexa 488-conjugated secondary antibodies were purchased from Molecular Probes (Carlsbad, CA). Plasmid constructions. HO1, original clone of which was a kind gift from Dr. A.M.K. Choi (University of Pittsburg, PA), was subcloned into pcDNA3 vector. Nrf2 and its dominant negative mutant were obtained from Dr. S. Kim (Catholic University). Human HO1 promoter construct was generated by PCR amplification of the target sequence and followed by cloning it into plasmid containing reporter gene. Briefly, about 4.4 kb fragment of the 5 0 -flanking region of the human HO1 gene including the transcription initiation site (spanning region of 4384 bp to +24 bp) was amplified from HeLa cell genomic DNA using PCR primers containing proper restriction enzyme sites for the cloning (5 0 -gctgagctcCAGCCTGT CACACAGCAGTTAGGC-3 0 and 5 0 -acgctcgagAGGAGGCAGGCGTT GACTGCC-3 0 ). Enzyme-digested fragment was cloned into SacI and XhoI site of the pGL3 basic vector containing the firefly luciferase cDNA (Promega, Madison, WI) to obtain pGL3HO1/4384-Luci construct. To construct pGL2ARE-Luci, double stranded oligonucleotides containing a single copy of the 41-bp pair murine GST-Ya ARE (5 0 -TAGCTTGGAAA TGACATTGCTAATGGTGACAAAGCAACTTT-3 0 ; the core sequence underlined) were cloned into the pGL2 promoter vector (Promega). All
Results Exogenous CO up-regulates HO1 induction in HepG2 cells and primary rat hepatocytes Liver is a major organ to detoxify or process drugs, infected pathogens, and many other stress signals. Since CO and/or HO1 has emerged as a major cytoprotective protein/effector molecule in liver [1,23], we wondered whether exogenous CO could affect HO1 gene expression in liver cells. First, we estimated CO-mediated HO1 expression at protein level. HepG2 cells were treated with various concentrations of RuCO for 6 h and immunoblot assay was performed using anti-HO1 antibody. The HO1 expression was gradually increased by CO in a dose-dependent manner up to 160 lM of RuCO (Fig. 1A). Since HO1 was expressed well at 40 lM RuCO, we tested time-kinetics at this concentration. As shown in Fig. 1B, HO1 expression was increased in a time-dependent manner with peak at 6 h but disappeared by 24 h after RuCO treatment. We also found that HO1 is induced by CO gas in a dose-dependent manner, which peaked at 5 lM concentration (Fig. 1C) and appeared similar pattern of time-kinetics as shown in RuCO treatment (data not shown). Based on these results, 40 lM RuCO and 6 h treatment were chosen as the experimental condition for further assays if it is not mentioned otherwise.
B.-S. Lee et al. / Biochemical and Biophysical Research Communications 343 (2006) 965–972
Fig. 1. Induction of HO1 by over-expressed HO1 or exogenous CO. (A) Dose-dependent HO1 induction by exogenous CO. Cells were treated with various concentrations of RuCO as exogenous CO donor for 6 h and the cell lysates were used for immunoblot analysis. (B) Time-dependent HO1 induction by exogenous CO. Cells were treated with 40 lM RuCO for indicated time points and analyzed to determine time-kinetics for HO1 expression. (C) Effect of CO gas on HO1 expression. Cells were treated for 6 h with various concentrations of CO gas, which were prepared as saturated form of 1 M CO in Hepes buffer before use. The cell lysates were subjected to immunoblot assay. Results are one representative of three independent experiments.
967
CO was considered as primary effector molecule [1,24], next, we analyzed the HO1 promoter activity to determine whether CO or any other heme products could be responsible for the HO1 induction at transcriptional level. As shown in Figs. 3A and B, only the cells treated with RuCO showed the induction of HO1 promoter activity and protein expression, but cells treated with other heme end products or carrier metal ruthenium did not. Biliverdin, unstable intermediate heme product, was also examined because BVR, an enzyme which mediates conversion of biliverdin to bilirubin, was known as transcription factor for HO1 expression [17]. Both promoter activity and protein expression were increased maximum about twofold in the biliverdin-treated cells compared to the control cells (Fig. 3). Although it is known that NO induces HO1 and CO inhibits iNOS expression [5,9], major concern remains because Otterbein’s group recently demonstrated that CO augmented TNF-a-mediated iNOS gene expression and NO produced by that NOS induced HO1 expression [23]. To rule out the involvement of NOS/NO in the CO-mediated HO1 expression, HO1 promoter activity and protein expression were measured after treatment of the cells with CO in the presence or absence of NGMMA, a NOS inhibitor. As shown in Fig. 5, NGMMA did not affect CO-mediated HO1 induction in both promoter activity and protein expression. Taken all together, the data clearly demonstrate that CO is the
Fig. 2. Induction of HO1 by exogenous CO in primary rat hepatocytes. Primary hepatocytes were isolated from rat livers. Purified hepatocytes were cultured overnight in the presence of dexamethasone and hepatocyte growth factors and treated with various concentrations of RuCO or CoPP for 6 h. Cell lysates were used for immunoblot with antibody against HO1. One representative result of two independent experiments is shown.
To examine whether CO could also induce HO1 expression in primary hepatocytes, rat hepatocytes were isolated from liver. Cells were then treated with RuCO for 6 h and the cell lysates were used for immunoblot analysis. CO induced HO1 expression in primary hepatocytes in a dose-dependent manner comparable to the effect of CoPP, a well-known inducer of HO1 in many cell types (Fig. 2). These results demonstrate that exogenous CO induces HO1 expression in HepG2 cells as well as in primary hepatocytes. CO specifically induces HO1 expression at the transcriptional level in HepG2 cells Since the major function of HO1 is to degrade free heme into CO, ferrous iron, and biliverdin/bilirubin in vivo and
Fig. 3. Effects of other heme products on HO1 induction. (A) HO1 promoter activity. Cells were transfected with pGL3HO1/4384-Luci constructs and treated with end products of heme by HO enzymatic activity; RuCO for CO, FeC for ferrous iron, and bilirubin as indicated for additional 6 h at 16 h post-transfection. RuCl2 was included to rule out the effect of metal carrier of RuCO. Values are means ± SD from three independent experiments. (B) HO1 protein expression. HepG2 cells were treated with heme end products as Fig. 1A for 6 h and cell lysates were subjected to immunoblot analysis using anti-HO1 antibody and b-actin. (C,D) Effect of biliverdin, intermediate metabolite of heme, on HO1 promoter activity and HO1 expression. The cells were transfected with pGL3 HO1/4384-Luci and pGK/b-gal and treated with different concentration of biliverdin for 6 h at 24 h post-transfection. Cell lysates were assayed for luciferase activity as the fold induction by normalizing the transfection efficiency and dividing values of each experiment relative to the control (C). HO1 expression was determined by immunoblot assay (D). Data shown are representative results of three independent experiments.
968
B.-S. Lee et al. / Biochemical and Biophysical Research Communications 343 (2006) 965–972
major inducer among the HO1 end products and NOS/NO is not likely to be involved in the signaling of CO-mediated HO1 induction in HepG2 cells. Signaling pathways through MAPKs are involved in CO-mediated HO1 promoter activity and protein expression Three groups of MAPKs have been described for HO1 gene regulation in many cell types [16]. To determine whether any of these pathways could be involved in the signaling, we examined phosphorylation of MAPKs in HepG2 cells treated with CO. CO induced the phosphorylation of ERK and JNK in a time-dependent manner but showed minor effect on p38 MAPK phosphorylation in HepG2 cells (Fig. 5A). To further confirm the signaling pathways, we tested HO1 induction using specific inhibitors of MAPKs. Immunoblot assays revealed that HO1 expressions were significantly blocked by SP600125, SB203508, and U0126 but not by Wortmannin, a PI3K inhibitor (Fig. 5B). Bay-11, a NF-jB inhibitor, had no effect on HO1 expression (data not shown). To investigate the role of kinases in HO1 induction at the transcriptional level, next, we performed HO1 promoter assay. In consistent with data of protein expression, HO1 promoter activities were blocked by SP600125, SB203508, and U0126 but not by Wortmannin (Fig. 6). These data indicate that all three MAPKs pathways are involved in the regulation of CO-mediated HO1 expression. Nrf2 and anti-oxidant responsive element are important for CO-mediated HO1 induction The HO1 promoter contains several transcriptional regulatory elements including the ARE/stRE that respond to redox-sensitive transcription factor, NF-E2-related factor 2 (Nrf2) [12]. To determine the Nrf2 involvement, we examined CO-mediated HO1 promoter activity using dominant negative mutant of Nrf2, Nrf2-DN. As shown in Fig. 6A (open square), induction of HO1 promoter activity by RuCO was completely inhibited in cells over-expressing Nrf2-DN mutant. Since Nrf2 is a well-known transcription activator which specifically binds to ARE site, we also measured ARE promoter activity. As expected, cells expressing Nrf2-DN completely suppressed RuCO-induced ARE promoter activity (Fig. 6A, closed square). In addition, SP600125, SB203508, and U0126 dramatically inhibited CO-induced ARE promoter activity (Fig. 6B) similar to the pattern shown in HO1 promoter activity (Fig. 5B). These data clearly support that Nrf2 is an important transcription factor responsible for CO-mediated HO1 induction. Nuclear translocation and accumulation of Nrf2 are the key step in CO-mediated HO1 expression To further investigate the role of Nrf2, we examined the localization of Nrf2 by confocal analysis. As shown in
Fig. 7A, Nrf2 was mainly localized in the cytoplasm of control cells. Nuclear translocation of Nrf2 was gradually increased by 4 h after RuCO treatment and maintained plateau thereafter up to 6 h. We also examined the localization of Nrf2 in the presence of kinases inhibitors. While SP600125, SB203508, and U0126 blocked the nuclear translocations of Nrf2, Wortmannin and Bay-11 did not (Fig. 7B and data not shown). Next, we measured the amount of the Nrf2 in cytoplasm and nucleus after fractionation of the cells followed by immunoblot analysis. Although Nrf2 was almost exclusively detected in the nuclear fractions in both RuCO-treated and non-treated cells unlike confocal data, it was dramatically increased in the cells treated with RuCO (Fig. 8). Collectively, these data support that Nrf2 nuclear translocation and accumulation is essential for the CO-mediated HO1 induction in HepG2 cells. Discussion Previous studies demonstrated that HO1 protects liver failure against inflammations, which is further supported by the fact that the tissue damage could not be prevented by exogenous CO in the HO1 knock-out mice [23,24]. Other studies demonstrated that CO and the bile pigments, biliverdin or bilirubin, when administered exogenously, exert potent cytoprotective effects through different action mechanisms in different cell types [2]. Or ferritin expression resulting from iron produced by HO1 activity might provide protective effects [25,26]. These led us to hypothesize that end products of HO1 activity might be HO1 inducers since the induction of HO1 is essential for manifestation of cytoprotective function of CO/HO1 system. To address this hypothesis, we first generated promoter construct and measured HO1 promoter activity (details in Materials and methods). When the cells were treated with end products of HO1 activity, only RuCO or CO gas significantly induced HO1 promoter activity and HO1 protein expression (Figs. 1 and 3). We paid special attention to biliverdin because it was reported recently that biliverdin reductase (BVR) acts as a transcriptional regulator for redox-related genes including HO1 [17]. In our experiment, biliverdin showed only a minor effect, as maximum twofold increase in promoter activity and protein expression, on HO1 induction (Fig. 3). The two gasotransmitters, CO and NO, share many common downstream signaling pathways and have overlapping regulatory functions [1,6]. Recent reports demonstrated that CO induces iNOS expression and NO induces HO1 expression in many cell types [5,9,27,28]. In our study, CO did not induce iNOS in HepG2 cells (data not shown). Moreover, NOS inhibitor, NGMMA, which used to remove any trace amount of NO from other NOS form, did not affect CO-mediated HO1 induction (Fig. 4). These data support that CO is the major inducer among end products of HO activity and directly mediates HO1 expression in HepG2 cells. This might explain why CO alone could not show cytoprotective
B.-S. Lee et al. / Biochemical and Biophysical Research Communications 343 (2006) 965–972
Fig. 4. Effect of NO synthase inhibitor on exogenous CO-mediated HO1 expression. HepG2 cells were transfected with pGL3 HO1/4384-Luci and pGK/b-gal. Sixteen hours later, the cells were treated for 30 min with different concentrations of NGMMA, a NOS inhibitor, as indicated and followed by RuCO for an additional 6 h. Sodium nitroprusside is included as positive control for HO1 expression. Cell lysates were analyzed for luciferase activity (A) and immunoblot assay (B). Results are representative of two independent experiments.
effect in HO1 knock-out mice although exogenous CO clearly has cytoprotective effect in many studies using wild type cells. Depending on the cell types and the nature of the stimuli, HO1 induction may be mediated by different signaling pathways. These pathways include cAMP-dependent mechanisms, Ca2+/calmodulin-dependent protein kinase, NF-jB, and the PI3K/Akt pathway [17,29–31]. Thus, we addressed signaling pathways of which might be involved in the HO1 induction by CO. Using known kinases inhibitors, we found that JNK, ERK, and p38 MAPK activities
969
were closely related to CO-mediated HO1 expression in HepG2 cells (Fig. 5). While Ras and raf-1 were also involved in signaling, PI3K/Akt and NF-jB were not (Fig. 5 and data not shown). Numerous studies have shown that exogenous and endogenous CO in fact elicits cytoprotective and anti-inflammatory responses by inhibiting the production of the proinflammatory cytokine or stimulates the synthesis of the anti-inflammatory cytokine by activating one or two kinases and subsequently transcriptional factors [4,32,33]. Based on our data, we can conclude that CO-mediated HO1 expression in HepG2 cells could be achieved through unique signaling pathways, which reflect cross-talk among three MAPKs. Further detailed study might be needed to clarify this relationship. Extensive analysis of the mouse gene and to a lesser extent of the human gene has identified stRE/Nrf2 transcription factor for gene regulation in response to a diverse array of HO1 inducers [12,20]. In agreement with this notion, Nrf2 played crucial role in CO-mediated HO1 expression. When the cells were transfected with dominant negative mutant of Nrf2, induction of both HO1 and ARE promoter activities by CO was completely suppressed (Fig. 6A). Furthermore, ARE promoter activity is regulated by the same MAPKs signaling as for HO1 promoter activity (Fig. 6B). Confocal analysis showed that Nrf2 is presented in the cytoplasm in the untreated cells and translocated into the nucleus after CO treatment (Fig. 7A) in agreement with a previous report [21]. Localization data using kinases inhibitors supported that Nrf2 nuclear translocation, which presumably Nrf2 binding to ARE site in HO1 promoter is indeed a key step for HO1 gene regulation (Fig. 7B). Immunoblot data from nuclear fractionation revealed that Nrf2 are mainly detected in nuclear fraction with clear patterns of accumulation after CO treatment (Fig. 8). Previous studies demonstrated that repressor
Fig. 5. Effects of MAP kinases inhibitors on exogenous CO-mediated HO1 expression. (A) Phosphorylation of MAPKs by CO. HepG2 cells were treated with RuCO for indicated times and immunoblot assays were performed with anti-pERK and anti-ERK (top panel), anti-pJNK and anti-JNK (middle panel), and anti-pp38 and anti-p38 (bottom panel) antibodies. (B,C) Effects of MAPKs inhibitors on HO1 induction by CO. HepG2 cells were transfected with pGL3 HO1/4384-Luci and pGK/b-gal, pre-treated with indicated inhibitors for 30 min followed by RuCO for 6 h. The HepG2 cells were then pretreated with MAPKs inhibitors SP600125 (20 lM, SP), a JNK inhibitor or SB203508 (40 lM, SB), a p38 MAPK inhibitor or U0126 (20 lM, U), an ERK inhibitor or Wortmannin, a PI3 kinase inhibitor (1 lM, W) for 30 min and then treated with or without 40 lM RuCO for 6 h. Cell lysates were assayed for promoter activity (B) by luciferase assay and protein expression (C) by immunoblot. Results are one representative of three independent experiments.
970
B.-S. Lee et al. / Biochemical and Biophysical Research Communications 343 (2006) 965–972
Fig. 6. Involvement of Nrf2 and ARE site in exogenous CO-induced HO1 promoter activity. (A) Effect of Nrf2-DN on HO1/ARE promoter activity induced by CO. HepG2 cells were transfected with pGL3 HO1/4384-Luci or pGL3 ARE-Luci with or without Nrf2-DN, dominant negative form of Nrf2, and pGK/b-gal. After 16 h, cells were treated with 40 lM RuCO for 6 h. The cells were harvested at 36 h post-transfection and the cell lysates were assayed for luciferase activity as the fold induction after normalized it to b-gal and vector control (opened square, pHO1/4384-Luci; closed square, pARE-Luci). Values are means ± SD from three independent experiments. (B) Effects of MAPKs inhibitors on ARE promoter activity induced by CO. HepG2 cells were transfected with pGL3 ARE-Luci and pGK/b-gal. After overnight, the cells were pre-treated with indicated inhibitors for 30 min as described in Fig. 5B and incubated for another 6 h in the presence or absence of RuCO. Cell lysates were then assayed for luciferase activity and the fold induction was determined relative to the control group after normalization of transfection efficiency by b-gal assay. One representative of three independent experiments is shown.
proteins, Keap1 and Bach1, regulate Nrf2 transcriptional activity [12,19,34]. Unlike most studies, however, Bach1 was localized in the cytoplasm in control cells and translocated into the nucleus after CO treatment although it
Fig. 8. Nuclear accumulation of Nrf2 in RuCO-treated HepG2 cells. HepG2 cells were treated with 80 lM RuCO for 6 h. The nuclei were then fractionated from the cytosol using hypotonic buffer as described in Materials and methods. The cell lysates from both fractions were subjected to immunoblot analysis for the detection of Nrf2. One representative result of three independent experiments is shown.
revealed different time-kinetics from Nrf2 (data not shown), which was also shown in a previous report [21]. Presumably, a balance of Nrf2 versus Bach1 inside the nucleus along with function of Keap1 might influence upor down-regulation of Nrf2/ARE-mediated HO1 expression. The detailed study for the mechanism by which CO regulates HO1 gene expression is underway. In summary, we demonstrated for the first time that exogenous CO directly induces HO1 expression in HepG2 cells. CO activates MAPKs pathways, which lead to Nrf2 nuclear translocation and accumulation, resulting in positive feed-forward regulation of HO1 gene expression. These results might provide helpful insight that local administration of CO itself without HO1 gene delivery can be a therapeutic means for certain hepatic injury.
Fig. 7. Localization of Nrf2 in RuCO-treated HepG2 cells. (A) Time-kinetics of Nrf2 nuclear translocation. HepG2 cells were prepared in 8-chamber slide a day before and treated with 80 lM RuCO for indicated times. The cells were fixed in acetone/ethanol and air-dried. After blocking with donkey serum for 30 min, the cells were stained with rabbit anti-Nrf2 antibody for 1 h followed by Alexa488-conjugate donkey anti-rabbit antibody. After three times washing with PBS, the cells were mounted with mounting media and the images were collected using FLUOVIEW FV1000 (OLYMPUS). (B) Effect of kinases inhibitors on CO-induced Nrf2 nuclear translocation. The cells were prepared in 8-chamber slides, treated with MAPK inhibitors or Wortmannin, a PI3K inhibitor for 30 min, and followed by RuCO for another 6 h. As described in (A) cells were stained with anti-Nrf2 antibodies and then with secondary antibodies. All images were overlaid on DIC image for cell morphology. One representative image set of three independent experiments is shown.
B.-S. Lee et al. / Biochemical and Biophysical Research Communications 343 (2006) 965–972
Acknowledgments We gratefully thank Drs. D. Sohn and E. Park for providing primary rat hepatocytes. This study was supported by the Korean Research Fundation Grant (KRF-2004005-E00038). References [1] L. Wu, R. Wang, Carbon monoxide: endogenous production, physiological functions, and pharmacological applications, Pharmacol. Rev. 57 (2005) 585–630. [2] M. Bauer, I. Bauer, Heme oxygenase-1: redox regulation and role in the hepatic response to oxidative stress, Antioxid. Redox Signal 4 (2002) 749–758. [3] F. Amersi, X.D. Shen, D. Anselmo, J. Melinek, S. Iyer, D.J. Southard, M. Katori, H.D. Volk, R.W. Busuttil, R. Buelow, J.W. Kupiec-Weglinski, Ex vivo exposure to carbon monoxide prevents hepatic ischemia/reperfusion injury through p38 MAP kinase pathway, Hepatology 35 (2002) 815–823. [4] S.W. Ryter, L.E. Otterbein, D. Morse, A.M. Choi, Heme oxygenase/ carbon monoxide signaling pathways: regulation and functional significance, Mol. Cell Biochem. 234-235 (2002) 249–263. [5] B.M. Choi, H.O. Pae, Y.M. Kim, H.T. Chung, Nitric oxide-mediated cytoprotection of hepatocytes from glucose deprivation-induced cytotoxicity: involvement of heme oxygenase-1, Hepatology 37 (2003) 810–823. [6] J.K. Sarady, B.S. Zuckerbraun, M. Bilban, O. Wagner, A. Usheva, F. Liu, E. Ifedigbo, R. Zamora, A.M. Choi, L.E. Otterbein, Carbon monoxide protection against endotoxic shock involves reciprocal effects on iNOS in the lung and liver, FASEB J. 18 (2004) 854–856. [7] S. Mazzola, M. Forni, M. Albertini, M.L. Bacci, A. Zannoni, F. Gentilini, M. Lavitrano, F.H. Bach, L.E. Otterbein, M.G. Clement, Carbon monoxide pretreatment prevents respiratory derangement and ameliorates hyperacute endotoxic shock in pigs, FASEB J. 19 (2005) 2045–2047. [8] R. Song, R.S. Mahidhara, Z. Zhou, R.A. Hoffman, D.W. Seol, R.A. Flavell, T.R. Billiar, L.E. Otterbein, A.M. Choi, Carbon monoxide inhibits T lymphocyte proliferation via caspase-dependent pathway, J. Immunol. 172 (2004) 1220–1226. [9] H.O. Pae, B.M. Choi, G.S. Oh, M.S. Lee, D.G. Ryu, H.Y. Rhew, Y.M. Kim, H.T. Chung, Roles of heme oxygenase-1 in the antiproliferative and antiapoptotic effects of nitric oxide on Jurkat T cells, Mol. Pharmacol. 66 (2004) 122–128. [10] L.E. Otterbein, J.K. Kolls, L.L. Mantell, J.L. Cook, J. Alam, A.M. Choi, Exogenous administration of heme oxygenase-1 by gene transfer provides protection against hyperoxia-induced lung injury, J. Clin. Invest. 103 (1999) 1047–1054. [11] P.J. Lee, B.H. Jiang, B.Y. Chin, N.V. Iyer, J. Alam, G.L. Semenza, A.M. Choi, Hypoxia-inducible factor-1 mediates transcriptional activation of the heme oxygenase-1 gene in response to hypoxia, J. Biol. Chem. 272 (1997) 5375–5381. [12] P. Gong, B. Hu, D. Stewart, M. Ellerbe, Y.G. Figueroa, V. Blank, B.S. Beckman, J. Alam, Cobalt induces heme oxygenase-1 expression by a hypoxia-inducible factor-independent mechanism in Chinese hamster ovary cells: regulation by Nrf2 and MafG transcription factors, J. Biol. Chem. 276 (2001) 27018–27025. [13] A. Nakao, J.S. Neto, S. Kanno, D.B. Stolz, K. Kimizuka, F. Liu, F.H. Bach, T.R. Billiar, A.M. Choi, L.E. Otterbein, N. Murase, Protection against ischemia/reperfusion injury in cardiac and renal transplantation with carbon monoxide, biliverdin and both, Am. J. Transplant 5 (2005) 282–291. [14] R. Foresti, R. Motterlini, The heme oxygenase pathway and its interaction with nitric oxide in the control of cellular homeostasis, Free Radic Res. 31 (1999) 459–475.
971
[15] S.W. Ryter, S. Xi, C.L. Hartsfield, A.M. Choi, Mitogen activated protein kinase (MAPK) pathway regulates heme oxygenase-1 gene expression by hypoxia in vascular cells, Antioxid. Redox Signal 4 (2002) 587–592. [16] T. Kietzmann, A. Samoylenko, S. Immenschuh, Transcriptional regulation of heme oxygenase-1 gene expression by MAP kinases of the JNK and p38 pathways in primary cultures of rat hepatocytes, J. Biol. Chem. 278 (2003) 17927–17936. [17] A. Kravets, Z. Hu, T. Miralem, M.D. Torno, M.D. Maines, Biliverdin reductase, a novel regulator for induction of activating transcription factor-2 and heme oxygenase-1, J. Biol. Chem. 279 (2004) 19916–19923. [18] N.M. Inamdar, Y.I. Ahn, J. Alam, The heme-responsive element of the mouse heme oxygenase-1 gene is an extended AP-1 binding site that resembles the recognition sequences for MAF and NF-E2 transcription factors, Biochem. Biophys. Res. Commun. 221 (1996) 570–576. [19] M. McMahon, K. Itoh, M. Yamamoto, J.D. Hayes, Keap1dependent proteasomal degradation of transcription factor Nrf2 contributes to the negative regulation of antioxidant response element-driven gene expression, J. Biol. Chem. 278 (2003) 21592– 21600. [20] J. Alam, J.L. Cook, Transcriptional regulation of the heme oxygenase-1 gene via the stress response element pathway, Curr. Pharm. Des. 9 (2003) 2499–2511. [21] S. Dhakshinamoorthy, A.K. Jain, D.A. Bloom, A.K. Jaiswal, Bach1 competes with Nrf2 leading to negative regulation of the antioxidant response element (ARE)-mediated NAD(P)H:quinone oxidoreductase 1 gene expression and induction in response to antioxidants, J. Biol. Chem. 280 (2005) 16891–16900. [22] B.E. VanUffelen, B.M. de Koster, J. VanSteveninck, J.G. Elferink, Carbon monoxide enhances human neutrophil migration in a cyclic GMP-dependent way, Biochem. Biophys. Res. Commun. 226 (1996) 21–26. [23] B.S. Zuckerbraun, T.R. Billiar, S.L. Otterbein, P.K. Kim, F. Liu, A.M. Choi, F.H. Bach, L.E. Otterbein, Carbon monoxide protects against liver failure through nitric oxide-induced heme oxygenase 1, J. Exp. Med. 198 (2003) 1707–1716. [24] L.E. Otterbein, M.P. Soares, K. Yamashita, F.H. Bach, Heme oxygenase-1: unleashing the protective properties of heme, Trends Immunol. 24 (2003) 449–455. [25] L.E. Otterbein, A.M. Choi, Heme oxygenase: colors of defense against cellular stress, Am. J. Physiol. Lung Cell Mol. Physiol. 279 (2000) L1029–L1037. [26] G. Balla, H.S. Jacob, J. Balla, M. Rosenberg, K. Nath, F. Apple, J.W. Eaton, G.M. Vercellotti, Ferritin: a cytoprotective antioxidant strategem of endothelium, J. Biol. Chem. 267 (1992) 18148– 18153. [27] W. Durante, M.H. Kroll, N. Christodoulides, K.J. Peyton, A.I. Schafer, Nitric oxide induces heme oxygenase-1 gene expression and carbon monoxide production in vascular smooth muscle cells, Circ. Res. 80 (1997) 557–564. [28] R. Foresti, M. Hoque, S. Bains, C.J. Green, R. Motterlini, Haem and nitric oxide: synergism in the modulation of the endothelial haem oxygenase-1 pathway, Biochem. J. 372 (2003) 381–390. [29] M. Salinas, J. Wang, M. Rosade Sagarra, D. Martin, A.I. Rojo, J. Martin-Perez, P.R. Ortiz de Montellano, A. Cuadrado, Protein kinase Akt/PKB phosphorylates heme oxygenase-1 in vitro and in vivo, FEBS Lett. 578 (2004) 90–94. [30] W. Durante, N. Christodoulides, K. Cheng, K.J. Peyton, R.K. Sunahara, A.I. Schafer, cAMP induces heme oxygenase-1 gene expression and carbon monoxide production in vascular smooth muscle, Am. J. Physiol. 273 (1997) H317–H323. [31] S. Brouard, P.O. Berberat, E. Tobiasch, M.P. Seldon, F.H. Bach, M.P. Soares, Heme oxygenase-1-derived carbon monoxide requires the activation of transcription factor NF-kappa B to protect endothelial cells from tumor necrosis factor-alpha-mediated apoptosis, J. Biol. Chem. 277 (2002) 17950–17961.
972
B.-S. Lee et al. / Biochemical and Biophysical Research Communications 343 (2006) 965–972
[32] L.E. Otterbein, F.H. Bach, J. Alam, M. Soares, H. Tao Lu, M. Wysk, R.J. Davis, R.A. Flavell, A.M. Choi, Carbon monoxide has antiinflammatory effects involving the mitogen-activated protein kinase pathway, Nat. Med. 6 (2000) 422–428. [33] D. Morse, S.E. Pischke, Z. Zhou, R.J. Davis, R.A. Flavell, T. Loop, S.L. Otterbein, L.E. Otterbein, A.M. Choi, Suppression of
inflammatory cytokine production by carbon monoxide involves the JNK pathway and AP-1, J. Biol. Chem. 278 (2003) 36993– 36998. [34] A. Kobayashi, T. Ohta, M. Yamamoto, Unique function of the Nrf2Keap1 pathway in the inducible expression of antioxidant and detoxifying enzymes, Methods Enzymol. 378 (2004) 273–286.
Carbon monoxide mediates heme oxygenase 1 induction via Nrf2 activation in hepatoma cells Bok-Soo Lee a, JungHee Heo a, Yong-Man Kim b, Sang Moo Shim a, Hyun-Ock Pae a, Young-Myeong Kim c, Hun-Taeg Chung a,* a
Medicinal Resources Research Institute, Department of Microbiology and Immunology, Wonkwang University School of Medicine, Iksan, Chonbuk 570-749, South Korea b FCB-Pharmicell Co., Ltd, Stem Cell Therapy Institute, Jungwon-gu, Sungnam-si, Kyungki-do 462-120, South Korea c Department of Molecular and Cellular Biochemistry, College of Medicine and Vascular System Research Center, Kangwon National University, Chunchon, Kangwon-do, South Korea Received 22 February 2006
Abstract Carbon monoxide (CO) and nitric oxide (NO) are two gas molecules which have cytoprotective functions against oxidative stress and inflammatory responses in many cell types. Currently, it is known that NO produced by nitric oxide synthase (NOS) induces heme oxygenase 1 (HO1) expression and CO produced by the HO1 inhibits inducible NOS expression. Here, we first show CO-mediated HO1 induction and its possible mechanism in human hepatocytes. Exposure of HepG2 cells or primary hepatocytes to CO resulted in dramatic induction of HO1 in dose- and time-dependent manner. The CO-mediated HO1 induction was abolished by MAP kinase inhibitors (MAPKs) but not affected by inhibitors of PI3 kinase or NF-jB. In addition, CO induced the nuclear translocation and accumulation of Nrf2, which suppressed by MAPKs inhibitors. Taken together, we suggest that CO induces Nrf2 activation via MAPKs signaling pathways, thereby resulting in HO1 expression in HepG2 cells. 2006 Elsevier Inc. All rights reserved. Keywords: Carbon monoxide; Tricarbonyl dichlororuthenium (II) dimmer (RuCO); Heme oxygenase 1; Nitric oxide/inducible NO synthase; Mitogenactivated protein kinases; NF-E2-related factor; Anti-oxidant response element
HO1 and its metabolic products play important regulatory roles in both physiological and pathological status [1]. HO1 is induced in many cell types and catalyzes the ratelimiting reaction, in the catabolism of heme, yielding final products in vivo: CO, bilirubin, and ferrous iron. In general, CO is considered to have pleiotropic cytoprotective activities [2–5]. CO protects cells/tissues from damages by free radicals or oxidative stress [6,7]. It also shows antiinflammatory, anti-proliferative, and anti-apoptotic effects by modulating related gene expressions or enzymatic activities [4,8,9]. HO1/CO may have a dual role in tissue pathology since high levels of HO1 are frequently detected in
*
Corresponding author. Fax: +82 63 851 5066. E-mail address: [email protected] (H.-T. Chung).
0006-291X/$ - see front matter 2006 Elsevier Inc. All rights reserved. doi:10.1016/j.bbrc.2006.03.058
various pathological states and generally in states of cellular oxidative stress [2,10]. It may not be therapeutic in all instances because there are some evidences showing that too much HO1/CO is a perpetrator of tissue injury [1,2]. Therefore, deciphering the underlying molecular mechanism that controls endogenous level of CO would be valuable to fine-tune CO production and minimize the potential detrimental situations under different pathological status, which eventually takes advantage of using CO as an effective therapeutic agent. There are many reports regarding the mechanism of HO1 expression induced by hypoxia, metal ions or inflammatory cytokines [11,12] and CO-mediated gene modulation including iNOS and several inflammatory cytokine genes [13,14]. Three major groups of mitogen-activated protein kinases (MAPKs), which are the extracellular
966
B.-S. Lee et al. / Biochemical and Biophysical Research Communications 343 (2006) 965–972
signal-regulated kinase (ERK), the jun NH2-terminal kinase (JNK), and the p38 MAP kinase (p38 MAPK), have been described for HO1 gene regulation [11,15,16]. Maine’s group added up that biliverdin reductase (BVR) functions as a transcriptional factor in HO1 gene expression [17]. The HO1 promoter contains several transcriptional regulatory elements that respond to redox-sensitive transcription factors, including NF-E2-related factor 2 (Nrf2) [18]. Nrf2 is a member of the basic leucine zipper family of transcription factors. There are some reports that Nrf2 resides in the cytoplasm bound to its inhibitor protein, Keap1, and translocates to the nucleus after stimulation and binds to a DNA sequence in target genes as the anti-oxidant response element (ARE)/stress response element (stRE) [12,19,20]. Other reports demonstrated that Nrf2 is located in nucleus together with Bach1, a negative transcription repressor, which keeps away most of Nrf2 to DNA binding that leads to basal level of HO1 transcription [21]. However, there is no report so far whether CO could directly affect HO1 gene expression. In the present study, we investigated the mechanism by which exogenous/endogenous CO induces HO1 expression in HepG2 cells. We demonstrated here that CO directly upregulates HO1 gene transcription by enhancing nuclear translocation and accumulation of Nrf2, presumably increasing Nrf2 binding to ARE site in HO1 promoter. We also showed that the activation of MAPKs is important for the CO-mediated Nrf2 nuclear translocation.
sequences of pGL3/HO1/4384-Luci and pARE-Luci were confirmed and verified the presence of the correct sequence and the absence of any other nucleotide changes by DNA sequencing. Administration of CO gas. Saturated stock solutions were prepared in buffer containing 140 mM NaCl, 5 mM KCl, and 20 mM Hepes, pH 7.3, as described [22]. CO stock solutions were freshly prepared for every experiment. After HepG2 cells were treated with various concentrations of CO gas for 6 h, HO1 expression was determined by immunoblot assay. Western blot analysis. Cell pellets were lysed in 1· sample buffer [50 mM Tris–HCl (pH 7.4), 150 mM NaCl, 0.1% SDS, 2 mM b-mercaptoethanol, 1 mM DTT, BPB, and xylene cyanol]. The cell lysates were electrophoresed on a SDS–polyacrylamide gel and proteins were transferred to polyvinylidene difluoride (PVDF) membranes. Immunoblotting was performed according to the manufacturer’s instructions and the blots were washed and developed with supersignal enhanced chemiluminescence (ECL) substrate solution (Pierce, IL). The signals were then visualized with X-ray film. Measurement of promoter activity. The cells were transiently transfected with the promoter constructs using the transfection reagent Fugene 6 (Roche; Mannheim, Germany). After harvest, cells were lysed in reporter lysis buffer (Promega; Madison, WI). Twenty microliters of cell extract was mixed with 100 lL of the luciferase assay reagent and the emitted light intensity was measured using the luminometer AutoLumat LB953 (EG and G Berthold; Bad Wildbad, Germany). Fold induction was calculated as intensity value from each experimental group divided by value from control group after normalization of transfection efficiency by b-gal assay. Immunostaining and confocal microscopy. For localization of Nrf2, the HepG2 cells were grown on Lab-Tek II chamber slides and treated as described in figure legends. Cells were fixed with ice-cold acetone for 10 min, blocked with 10% donkey serum for 30 min, and stained with antirabbit Nrf2. After washing, cells were stained with Alexa 488-conjugated donkey anti-rabbit antibody. Cells were then washed and mounted in mounting medium (Vector, Burlingame, CA). Confocal microscopy was performed using a FLUOVIEW FV1000 (Olympus, PA). Images were collected at 512 · 512 pixel resolution and overlayed with DIC image.
Materials and methods Cell culture and reagents. Human hepatocyte cell line, HepG2, was purchased from ATCC (Manassas, VA) and maintained with RPMI1640 supplemented with 10% fetal bovine serum (FBS), 1% penicillin/streptomycin. Tricarbonyl dichlororuthenium (II) dimer (RuCO), RuCl2, bilirubin, biliverdin, ferrous citrate (FeC), and sodium nitroprusside (SNP) were purchased from Sigma (St. Louis, MO). Cobalt protoporphyrin (CoPP) and tin protoporphyrin (SnPP) were from Porphyrin Products (Logan, UT). SB203508, SP600125, U0125, Bay-11, and Wortmannin, NG-monomethyl-L-arginine (NGMMA), and antibody against HO1 was obtained from Calbiochem (La Jolla, CA). Antibodies to Nrf2, b-actin, and HRP-conjugated secondary antibodies were from Santa Cruz Biotechnology (Santa Cruz, CA). Alexa 488-conjugated secondary antibodies were purchased from Molecular Probes (Carlsbad, CA). Plasmid constructions. HO1, original clone of which was a kind gift from Dr. A.M.K. Choi (University of Pittsburg, PA), was subcloned into pcDNA3 vector. Nrf2 and its dominant negative mutant were obtained from Dr. S. Kim (Catholic University). Human HO1 promoter construct was generated by PCR amplification of the target sequence and followed by cloning it into plasmid containing reporter gene. Briefly, about 4.4 kb fragment of the 5 0 -flanking region of the human HO1 gene including the transcription initiation site (spanning region of 4384 bp to +24 bp) was amplified from HeLa cell genomic DNA using PCR primers containing proper restriction enzyme sites for the cloning (5 0 -gctgagctcCAGCCTGT CACACAGCAGTTAGGC-3 0 and 5 0 -acgctcgagAGGAGGCAGGCGTT GACTGCC-3 0 ). Enzyme-digested fragment was cloned into SacI and XhoI site of the pGL3 basic vector containing the firefly luciferase cDNA (Promega, Madison, WI) to obtain pGL3HO1/4384-Luci construct. To construct pGL2ARE-Luci, double stranded oligonucleotides containing a single copy of the 41-bp pair murine GST-Ya ARE (5 0 -TAGCTTGGAAA TGACATTGCTAATGGTGACAAAGCAACTTT-3 0 ; the core sequence underlined) were cloned into the pGL2 promoter vector (Promega). All
Results Exogenous CO up-regulates HO1 induction in HepG2 cells and primary rat hepatocytes Liver is a major organ to detoxify or process drugs, infected pathogens, and many other stress signals. Since CO and/or HO1 has emerged as a major cytoprotective protein/effector molecule in liver [1,23], we wondered whether exogenous CO could affect HO1 gene expression in liver cells. First, we estimated CO-mediated HO1 expression at protein level. HepG2 cells were treated with various concentrations of RuCO for 6 h and immunoblot assay was performed using anti-HO1 antibody. The HO1 expression was gradually increased by CO in a dose-dependent manner up to 160 lM of RuCO (Fig. 1A). Since HO1 was expressed well at 40 lM RuCO, we tested time-kinetics at this concentration. As shown in Fig. 1B, HO1 expression was increased in a time-dependent manner with peak at 6 h but disappeared by 24 h after RuCO treatment. We also found that HO1 is induced by CO gas in a dose-dependent manner, which peaked at 5 lM concentration (Fig. 1C) and appeared similar pattern of time-kinetics as shown in RuCO treatment (data not shown). Based on these results, 40 lM RuCO and 6 h treatment were chosen as the experimental condition for further assays if it is not mentioned otherwise.
B.-S. Lee et al. / Biochemical and Biophysical Research Communications 343 (2006) 965–972
Fig. 1. Induction of HO1 by over-expressed HO1 or exogenous CO. (A) Dose-dependent HO1 induction by exogenous CO. Cells were treated with various concentrations of RuCO as exogenous CO donor for 6 h and the cell lysates were used for immunoblot analysis. (B) Time-dependent HO1 induction by exogenous CO. Cells were treated with 40 lM RuCO for indicated time points and analyzed to determine time-kinetics for HO1 expression. (C) Effect of CO gas on HO1 expression. Cells were treated for 6 h with various concentrations of CO gas, which were prepared as saturated form of 1 M CO in Hepes buffer before use. The cell lysates were subjected to immunoblot assay. Results are one representative of three independent experiments.
967
CO was considered as primary effector molecule [1,24], next, we analyzed the HO1 promoter activity to determine whether CO or any other heme products could be responsible for the HO1 induction at transcriptional level. As shown in Figs. 3A and B, only the cells treated with RuCO showed the induction of HO1 promoter activity and protein expression, but cells treated with other heme end products or carrier metal ruthenium did not. Biliverdin, unstable intermediate heme product, was also examined because BVR, an enzyme which mediates conversion of biliverdin to bilirubin, was known as transcription factor for HO1 expression [17]. Both promoter activity and protein expression were increased maximum about twofold in the biliverdin-treated cells compared to the control cells (Fig. 3). Although it is known that NO induces HO1 and CO inhibits iNOS expression [5,9], major concern remains because Otterbein’s group recently demonstrated that CO augmented TNF-a-mediated iNOS gene expression and NO produced by that NOS induced HO1 expression [23]. To rule out the involvement of NOS/NO in the CO-mediated HO1 expression, HO1 promoter activity and protein expression were measured after treatment of the cells with CO in the presence or absence of NGMMA, a NOS inhibitor. As shown in Fig. 5, NGMMA did not affect CO-mediated HO1 induction in both promoter activity and protein expression. Taken all together, the data clearly demonstrate that CO is the
Fig. 2. Induction of HO1 by exogenous CO in primary rat hepatocytes. Primary hepatocytes were isolated from rat livers. Purified hepatocytes were cultured overnight in the presence of dexamethasone and hepatocyte growth factors and treated with various concentrations of RuCO or CoPP for 6 h. Cell lysates were used for immunoblot with antibody against HO1. One representative result of two independent experiments is shown.
To examine whether CO could also induce HO1 expression in primary hepatocytes, rat hepatocytes were isolated from liver. Cells were then treated with RuCO for 6 h and the cell lysates were used for immunoblot analysis. CO induced HO1 expression in primary hepatocytes in a dose-dependent manner comparable to the effect of CoPP, a well-known inducer of HO1 in many cell types (Fig. 2). These results demonstrate that exogenous CO induces HO1 expression in HepG2 cells as well as in primary hepatocytes. CO specifically induces HO1 expression at the transcriptional level in HepG2 cells Since the major function of HO1 is to degrade free heme into CO, ferrous iron, and biliverdin/bilirubin in vivo and
Fig. 3. Effects of other heme products on HO1 induction. (A) HO1 promoter activity. Cells were transfected with pGL3HO1/4384-Luci constructs and treated with end products of heme by HO enzymatic activity; RuCO for CO, FeC for ferrous iron, and bilirubin as indicated for additional 6 h at 16 h post-transfection. RuCl2 was included to rule out the effect of metal carrier of RuCO. Values are means ± SD from three independent experiments. (B) HO1 protein expression. HepG2 cells were treated with heme end products as Fig. 1A for 6 h and cell lysates were subjected to immunoblot analysis using anti-HO1 antibody and b-actin. (C,D) Effect of biliverdin, intermediate metabolite of heme, on HO1 promoter activity and HO1 expression. The cells were transfected with pGL3 HO1/4384-Luci and pGK/b-gal and treated with different concentration of biliverdin for 6 h at 24 h post-transfection. Cell lysates were assayed for luciferase activity as the fold induction by normalizing the transfection efficiency and dividing values of each experiment relative to the control (C). HO1 expression was determined by immunoblot assay (D). Data shown are representative results of three independent experiments.
968
B.-S. Lee et al. / Biochemical and Biophysical Research Communications 343 (2006) 965–972
major inducer among the HO1 end products and NOS/NO is not likely to be involved in the signaling of CO-mediated HO1 induction in HepG2 cells. Signaling pathways through MAPKs are involved in CO-mediated HO1 promoter activity and protein expression Three groups of MAPKs have been described for HO1 gene regulation in many cell types [16]. To determine whether any of these pathways could be involved in the signaling, we examined phosphorylation of MAPKs in HepG2 cells treated with CO. CO induced the phosphorylation of ERK and JNK in a time-dependent manner but showed minor effect on p38 MAPK phosphorylation in HepG2 cells (Fig. 5A). To further confirm the signaling pathways, we tested HO1 induction using specific inhibitors of MAPKs. Immunoblot assays revealed that HO1 expressions were significantly blocked by SP600125, SB203508, and U0126 but not by Wortmannin, a PI3K inhibitor (Fig. 5B). Bay-11, a NF-jB inhibitor, had no effect on HO1 expression (data not shown). To investigate the role of kinases in HO1 induction at the transcriptional level, next, we performed HO1 promoter assay. In consistent with data of protein expression, HO1 promoter activities were blocked by SP600125, SB203508, and U0126 but not by Wortmannin (Fig. 6). These data indicate that all three MAPKs pathways are involved in the regulation of CO-mediated HO1 expression. Nrf2 and anti-oxidant responsive element are important for CO-mediated HO1 induction The HO1 promoter contains several transcriptional regulatory elements including the ARE/stRE that respond to redox-sensitive transcription factor, NF-E2-related factor 2 (Nrf2) [12]. To determine the Nrf2 involvement, we examined CO-mediated HO1 promoter activity using dominant negative mutant of Nrf2, Nrf2-DN. As shown in Fig. 6A (open square), induction of HO1 promoter activity by RuCO was completely inhibited in cells over-expressing Nrf2-DN mutant. Since Nrf2 is a well-known transcription activator which specifically binds to ARE site, we also measured ARE promoter activity. As expected, cells expressing Nrf2-DN completely suppressed RuCO-induced ARE promoter activity (Fig. 6A, closed square). In addition, SP600125, SB203508, and U0126 dramatically inhibited CO-induced ARE promoter activity (Fig. 6B) similar to the pattern shown in HO1 promoter activity (Fig. 5B). These data clearly support that Nrf2 is an important transcription factor responsible for CO-mediated HO1 induction. Nuclear translocation and accumulation of Nrf2 are the key step in CO-mediated HO1 expression To further investigate the role of Nrf2, we examined the localization of Nrf2 by confocal analysis. As shown in
Fig. 7A, Nrf2 was mainly localized in the cytoplasm of control cells. Nuclear translocation of Nrf2 was gradually increased by 4 h after RuCO treatment and maintained plateau thereafter up to 6 h. We also examined the localization of Nrf2 in the presence of kinases inhibitors. While SP600125, SB203508, and U0126 blocked the nuclear translocations of Nrf2, Wortmannin and Bay-11 did not (Fig. 7B and data not shown). Next, we measured the amount of the Nrf2 in cytoplasm and nucleus after fractionation of the cells followed by immunoblot analysis. Although Nrf2 was almost exclusively detected in the nuclear fractions in both RuCO-treated and non-treated cells unlike confocal data, it was dramatically increased in the cells treated with RuCO (Fig. 8). Collectively, these data support that Nrf2 nuclear translocation and accumulation is essential for the CO-mediated HO1 induction in HepG2 cells. Discussion Previous studies demonstrated that HO1 protects liver failure against inflammations, which is further supported by the fact that the tissue damage could not be prevented by exogenous CO in the HO1 knock-out mice [23,24]. Other studies demonstrated that CO and the bile pigments, biliverdin or bilirubin, when administered exogenously, exert potent cytoprotective effects through different action mechanisms in different cell types [2]. Or ferritin expression resulting from iron produced by HO1 activity might provide protective effects [25,26]. These led us to hypothesize that end products of HO1 activity might be HO1 inducers since the induction of HO1 is essential for manifestation of cytoprotective function of CO/HO1 system. To address this hypothesis, we first generated promoter construct and measured HO1 promoter activity (details in Materials and methods). When the cells were treated with end products of HO1 activity, only RuCO or CO gas significantly induced HO1 promoter activity and HO1 protein expression (Figs. 1 and 3). We paid special attention to biliverdin because it was reported recently that biliverdin reductase (BVR) acts as a transcriptional regulator for redox-related genes including HO1 [17]. In our experiment, biliverdin showed only a minor effect, as maximum twofold increase in promoter activity and protein expression, on HO1 induction (Fig. 3). The two gasotransmitters, CO and NO, share many common downstream signaling pathways and have overlapping regulatory functions [1,6]. Recent reports demonstrated that CO induces iNOS expression and NO induces HO1 expression in many cell types [5,9,27,28]. In our study, CO did not induce iNOS in HepG2 cells (data not shown). Moreover, NOS inhibitor, NGMMA, which used to remove any trace amount of NO from other NOS form, did not affect CO-mediated HO1 induction (Fig. 4). These data support that CO is the major inducer among end products of HO activity and directly mediates HO1 expression in HepG2 cells. This might explain why CO alone could not show cytoprotective
B.-S. Lee et al. / Biochemical and Biophysical Research Communications 343 (2006) 965–972
Fig. 4. Effect of NO synthase inhibitor on exogenous CO-mediated HO1 expression. HepG2 cells were transfected with pGL3 HO1/4384-Luci and pGK/b-gal. Sixteen hours later, the cells were treated for 30 min with different concentrations of NGMMA, a NOS inhibitor, as indicated and followed by RuCO for an additional 6 h. Sodium nitroprusside is included as positive control for HO1 expression. Cell lysates were analyzed for luciferase activity (A) and immunoblot assay (B). Results are representative of two independent experiments.
effect in HO1 knock-out mice although exogenous CO clearly has cytoprotective effect in many studies using wild type cells. Depending on the cell types and the nature of the stimuli, HO1 induction may be mediated by different signaling pathways. These pathways include cAMP-dependent mechanisms, Ca2+/calmodulin-dependent protein kinase, NF-jB, and the PI3K/Akt pathway [17,29–31]. Thus, we addressed signaling pathways of which might be involved in the HO1 induction by CO. Using known kinases inhibitors, we found that JNK, ERK, and p38 MAPK activities
969
were closely related to CO-mediated HO1 expression in HepG2 cells (Fig. 5). While Ras and raf-1 were also involved in signaling, PI3K/Akt and NF-jB were not (Fig. 5 and data not shown). Numerous studies have shown that exogenous and endogenous CO in fact elicits cytoprotective and anti-inflammatory responses by inhibiting the production of the proinflammatory cytokine or stimulates the synthesis of the anti-inflammatory cytokine by activating one or two kinases and subsequently transcriptional factors [4,32,33]. Based on our data, we can conclude that CO-mediated HO1 expression in HepG2 cells could be achieved through unique signaling pathways, which reflect cross-talk among three MAPKs. Further detailed study might be needed to clarify this relationship. Extensive analysis of the mouse gene and to a lesser extent of the human gene has identified stRE/Nrf2 transcription factor for gene regulation in response to a diverse array of HO1 inducers [12,20]. In agreement with this notion, Nrf2 played crucial role in CO-mediated HO1 expression. When the cells were transfected with dominant negative mutant of Nrf2, induction of both HO1 and ARE promoter activities by CO was completely suppressed (Fig. 6A). Furthermore, ARE promoter activity is regulated by the same MAPKs signaling as for HO1 promoter activity (Fig. 6B). Confocal analysis showed that Nrf2 is presented in the cytoplasm in the untreated cells and translocated into the nucleus after CO treatment (Fig. 7A) in agreement with a previous report [21]. Localization data using kinases inhibitors supported that Nrf2 nuclear translocation, which presumably Nrf2 binding to ARE site in HO1 promoter is indeed a key step for HO1 gene regulation (Fig. 7B). Immunoblot data from nuclear fractionation revealed that Nrf2 are mainly detected in nuclear fraction with clear patterns of accumulation after CO treatment (Fig. 8). Previous studies demonstrated that repressor
Fig. 5. Effects of MAP kinases inhibitors on exogenous CO-mediated HO1 expression. (A) Phosphorylation of MAPKs by CO. HepG2 cells were treated with RuCO for indicated times and immunoblot assays were performed with anti-pERK and anti-ERK (top panel), anti-pJNK and anti-JNK (middle panel), and anti-pp38 and anti-p38 (bottom panel) antibodies. (B,C) Effects of MAPKs inhibitors on HO1 induction by CO. HepG2 cells were transfected with pGL3 HO1/4384-Luci and pGK/b-gal, pre-treated with indicated inhibitors for 30 min followed by RuCO for 6 h. The HepG2 cells were then pretreated with MAPKs inhibitors SP600125 (20 lM, SP), a JNK inhibitor or SB203508 (40 lM, SB), a p38 MAPK inhibitor or U0126 (20 lM, U), an ERK inhibitor or Wortmannin, a PI3 kinase inhibitor (1 lM, W) for 30 min and then treated with or without 40 lM RuCO for 6 h. Cell lysates were assayed for promoter activity (B) by luciferase assay and protein expression (C) by immunoblot. Results are one representative of three independent experiments.
970
B.-S. Lee et al. / Biochemical and Biophysical Research Communications 343 (2006) 965–972
Fig. 6. Involvement of Nrf2 and ARE site in exogenous CO-induced HO1 promoter activity. (A) Effect of Nrf2-DN on HO1/ARE promoter activity induced by CO. HepG2 cells were transfected with pGL3 HO1/4384-Luci or pGL3 ARE-Luci with or without Nrf2-DN, dominant negative form of Nrf2, and pGK/b-gal. After 16 h, cells were treated with 40 lM RuCO for 6 h. The cells were harvested at 36 h post-transfection and the cell lysates were assayed for luciferase activity as the fold induction after normalized it to b-gal and vector control (opened square, pHO1/4384-Luci; closed square, pARE-Luci). Values are means ± SD from three independent experiments. (B) Effects of MAPKs inhibitors on ARE promoter activity induced by CO. HepG2 cells were transfected with pGL3 ARE-Luci and pGK/b-gal. After overnight, the cells were pre-treated with indicated inhibitors for 30 min as described in Fig. 5B and incubated for another 6 h in the presence or absence of RuCO. Cell lysates were then assayed for luciferase activity and the fold induction was determined relative to the control group after normalization of transfection efficiency by b-gal assay. One representative of three independent experiments is shown.
proteins, Keap1 and Bach1, regulate Nrf2 transcriptional activity [12,19,34]. Unlike most studies, however, Bach1 was localized in the cytoplasm in control cells and translocated into the nucleus after CO treatment although it
Fig. 8. Nuclear accumulation of Nrf2 in RuCO-treated HepG2 cells. HepG2 cells were treated with 80 lM RuCO for 6 h. The nuclei were then fractionated from the cytosol using hypotonic buffer as described in Materials and methods. The cell lysates from both fractions were subjected to immunoblot analysis for the detection of Nrf2. One representative result of three independent experiments is shown.
revealed different time-kinetics from Nrf2 (data not shown), which was also shown in a previous report [21]. Presumably, a balance of Nrf2 versus Bach1 inside the nucleus along with function of Keap1 might influence upor down-regulation of Nrf2/ARE-mediated HO1 expression. The detailed study for the mechanism by which CO regulates HO1 gene expression is underway. In summary, we demonstrated for the first time that exogenous CO directly induces HO1 expression in HepG2 cells. CO activates MAPKs pathways, which lead to Nrf2 nuclear translocation and accumulation, resulting in positive feed-forward regulation of HO1 gene expression. These results might provide helpful insight that local administration of CO itself without HO1 gene delivery can be a therapeutic means for certain hepatic injury.
Fig. 7. Localization of Nrf2 in RuCO-treated HepG2 cells. (A) Time-kinetics of Nrf2 nuclear translocation. HepG2 cells were prepared in 8-chamber slide a day before and treated with 80 lM RuCO for indicated times. The cells were fixed in acetone/ethanol and air-dried. After blocking with donkey serum for 30 min, the cells were stained with rabbit anti-Nrf2 antibody for 1 h followed by Alexa488-conjugate donkey anti-rabbit antibody. After three times washing with PBS, the cells were mounted with mounting media and the images were collected using FLUOVIEW FV1000 (OLYMPUS). (B) Effect of kinases inhibitors on CO-induced Nrf2 nuclear translocation. The cells were prepared in 8-chamber slides, treated with MAPK inhibitors or Wortmannin, a PI3K inhibitor for 30 min, and followed by RuCO for another 6 h. As described in (A) cells were stained with anti-Nrf2 antibodies and then with secondary antibodies. All images were overlaid on DIC image for cell morphology. One representative image set of three independent experiments is shown.
B.-S. Lee et al. / Biochemical and Biophysical Research Communications 343 (2006) 965–972
Acknowledgments We gratefully thank Drs. D. Sohn and E. Park for providing primary rat hepatocytes. This study was supported by the Korean Research Fundation Grant (KRF-2004005-E00038). References [1] L. Wu, R. Wang, Carbon monoxide: endogenous production, physiological functions, and pharmacological applications, Pharmacol. Rev. 57 (2005) 585–630. [2] M. Bauer, I. Bauer, Heme oxygenase-1: redox regulation and role in the hepatic response to oxidative stress, Antioxid. Redox Signal 4 (2002) 749–758. [3] F. Amersi, X.D. Shen, D. Anselmo, J. Melinek, S. Iyer, D.J. Southard, M. Katori, H.D. Volk, R.W. Busuttil, R. Buelow, J.W. Kupiec-Weglinski, Ex vivo exposure to carbon monoxide prevents hepatic ischemia/reperfusion injury through p38 MAP kinase pathway, Hepatology 35 (2002) 815–823. [4] S.W. Ryter, L.E. Otterbein, D. Morse, A.M. Choi, Heme oxygenase/ carbon monoxide signaling pathways: regulation and functional significance, Mol. Cell Biochem. 234-235 (2002) 249–263. [5] B.M. Choi, H.O. Pae, Y.M. Kim, H.T. Chung, Nitric oxide-mediated cytoprotection of hepatocytes from glucose deprivation-induced cytotoxicity: involvement of heme oxygenase-1, Hepatology 37 (2003) 810–823. [6] J.K. Sarady, B.S. Zuckerbraun, M. Bilban, O. Wagner, A. Usheva, F. Liu, E. Ifedigbo, R. Zamora, A.M. Choi, L.E. Otterbein, Carbon monoxide protection against endotoxic shock involves reciprocal effects on iNOS in the lung and liver, FASEB J. 18 (2004) 854–856. [7] S. Mazzola, M. Forni, M. Albertini, M.L. Bacci, A. Zannoni, F. Gentilini, M. Lavitrano, F.H. Bach, L.E. Otterbein, M.G. Clement, Carbon monoxide pretreatment prevents respiratory derangement and ameliorates hyperacute endotoxic shock in pigs, FASEB J. 19 (2005) 2045–2047. [8] R. Song, R.S. Mahidhara, Z. Zhou, R.A. Hoffman, D.W. Seol, R.A. Flavell, T.R. Billiar, L.E. Otterbein, A.M. Choi, Carbon monoxide inhibits T lymphocyte proliferation via caspase-dependent pathway, J. Immunol. 172 (2004) 1220–1226. [9] H.O. Pae, B.M. Choi, G.S. Oh, M.S. Lee, D.G. Ryu, H.Y. Rhew, Y.M. Kim, H.T. Chung, Roles of heme oxygenase-1 in the antiproliferative and antiapoptotic effects of nitric oxide on Jurkat T cells, Mol. Pharmacol. 66 (2004) 122–128. [10] L.E. Otterbein, J.K. Kolls, L.L. Mantell, J.L. Cook, J. Alam, A.M. Choi, Exogenous administration of heme oxygenase-1 by gene transfer provides protection against hyperoxia-induced lung injury, J. Clin. Invest. 103 (1999) 1047–1054. [11] P.J. Lee, B.H. Jiang, B.Y. Chin, N.V. Iyer, J. Alam, G.L. Semenza, A.M. Choi, Hypoxia-inducible factor-1 mediates transcriptional activation of the heme oxygenase-1 gene in response to hypoxia, J. Biol. Chem. 272 (1997) 5375–5381. [12] P. Gong, B. Hu, D. Stewart, M. Ellerbe, Y.G. Figueroa, V. Blank, B.S. Beckman, J. Alam, Cobalt induces heme oxygenase-1 expression by a hypoxia-inducible factor-independent mechanism in Chinese hamster ovary cells: regulation by Nrf2 and MafG transcription factors, J. Biol. Chem. 276 (2001) 27018–27025. [13] A. Nakao, J.S. Neto, S. Kanno, D.B. Stolz, K. Kimizuka, F. Liu, F.H. Bach, T.R. Billiar, A.M. Choi, L.E. Otterbein, N. Murase, Protection against ischemia/reperfusion injury in cardiac and renal transplantation with carbon monoxide, biliverdin and both, Am. J. Transplant 5 (2005) 282–291. [14] R. Foresti, R. Motterlini, The heme oxygenase pathway and its interaction with nitric oxide in the control of cellular homeostasis, Free Radic Res. 31 (1999) 459–475.
971
[15] S.W. Ryter, S. Xi, C.L. Hartsfield, A.M. Choi, Mitogen activated protein kinase (MAPK) pathway regulates heme oxygenase-1 gene expression by hypoxia in vascular cells, Antioxid. Redox Signal 4 (2002) 587–592. [16] T. Kietzmann, A. Samoylenko, S. Immenschuh, Transcriptional regulation of heme oxygenase-1 gene expression by MAP kinases of the JNK and p38 pathways in primary cultures of rat hepatocytes, J. Biol. Chem. 278 (2003) 17927–17936. [17] A. Kravets, Z. Hu, T. Miralem, M.D. Torno, M.D. Maines, Biliverdin reductase, a novel regulator for induction of activating transcription factor-2 and heme oxygenase-1, J. Biol. Chem. 279 (2004) 19916–19923. [18] N.M. Inamdar, Y.I. Ahn, J. Alam, The heme-responsive element of the mouse heme oxygenase-1 gene is an extended AP-1 binding site that resembles the recognition sequences for MAF and NF-E2 transcription factors, Biochem. Biophys. Res. Commun. 221 (1996) 570–576. [19] M. McMahon, K. Itoh, M. Yamamoto, J.D. Hayes, Keap1dependent proteasomal degradation of transcription factor Nrf2 contributes to the negative regulation of antioxidant response element-driven gene expression, J. Biol. Chem. 278 (2003) 21592– 21600. [20] J. Alam, J.L. Cook, Transcriptional regulation of the heme oxygenase-1 gene via the stress response element pathway, Curr. Pharm. Des. 9 (2003) 2499–2511. [21] S. Dhakshinamoorthy, A.K. Jain, D.A. Bloom, A.K. Jaiswal, Bach1 competes with Nrf2 leading to negative regulation of the antioxidant response element (ARE)-mediated NAD(P)H:quinone oxidoreductase 1 gene expression and induction in response to antioxidants, J. Biol. Chem. 280 (2005) 16891–16900. [22] B.E. VanUffelen, B.M. de Koster, J. VanSteveninck, J.G. Elferink, Carbon monoxide enhances human neutrophil migration in a cyclic GMP-dependent way, Biochem. Biophys. Res. Commun. 226 (1996) 21–26. [23] B.S. Zuckerbraun, T.R. Billiar, S.L. Otterbein, P.K. Kim, F. Liu, A.M. Choi, F.H. Bach, L.E. Otterbein, Carbon monoxide protects against liver failure through nitric oxide-induced heme oxygenase 1, J. Exp. Med. 198 (2003) 1707–1716. [24] L.E. Otterbein, M.P. Soares, K. Yamashita, F.H. Bach, Heme oxygenase-1: unleashing the protective properties of heme, Trends Immunol. 24 (2003) 449–455. [25] L.E. Otterbein, A.M. Choi, Heme oxygenase: colors of defense against cellular stress, Am. J. Physiol. Lung Cell Mol. Physiol. 279 (2000) L1029–L1037. [26] G. Balla, H.S. Jacob, J. Balla, M. Rosenberg, K. Nath, F. Apple, J.W. Eaton, G.M. Vercellotti, Ferritin: a cytoprotective antioxidant strategem of endothelium, J. Biol. Chem. 267 (1992) 18148– 18153. [27] W. Durante, M.H. Kroll, N. Christodoulides, K.J. Peyton, A.I. Schafer, Nitric oxide induces heme oxygenase-1 gene expression and carbon monoxide production in vascular smooth muscle cells, Circ. Res. 80 (1997) 557–564. [28] R. Foresti, M. Hoque, S. Bains, C.J. Green, R. Motterlini, Haem and nitric oxide: synergism in the modulation of the endothelial haem oxygenase-1 pathway, Biochem. J. 372 (2003) 381–390. [29] M. Salinas, J. Wang, M. Rosade Sagarra, D. Martin, A.I. Rojo, J. Martin-Perez, P.R. Ortiz de Montellano, A. Cuadrado, Protein kinase Akt/PKB phosphorylates heme oxygenase-1 in vitro and in vivo, FEBS Lett. 578 (2004) 90–94. [30] W. Durante, N. Christodoulides, K. Cheng, K.J. Peyton, R.K. Sunahara, A.I. Schafer, cAMP induces heme oxygenase-1 gene expression and carbon monoxide production in vascular smooth muscle, Am. J. Physiol. 273 (1997) H317–H323. [31] S. Brouard, P.O. Berberat, E. Tobiasch, M.P. Seldon, F.H. Bach, M.P. Soares, Heme oxygenase-1-derived carbon monoxide requires the activation of transcription factor NF-kappa B to protect endothelial cells from tumor necrosis factor-alpha-mediated apoptosis, J. Biol. Chem. 277 (2002) 17950–17961.
972
B.-S. Lee et al. / Biochemical and Biophysical Research Communications 343 (2006) 965–972
[32] L.E. Otterbein, F.H. Bach, J. Alam, M. Soares, H. Tao Lu, M. Wysk, R.J. Davis, R.A. Flavell, A.M. Choi, Carbon monoxide has antiinflammatory effects involving the mitogen-activated protein kinase pathway, Nat. Med. 6 (2000) 422–428. [33] D. Morse, S.E. Pischke, Z. Zhou, R.J. Davis, R.A. Flavell, T. Loop, S.L. Otterbein, L.E. Otterbein, A.M. Choi, Suppression of
inflammatory cytokine production by carbon monoxide involves the JNK pathway and AP-1, J. Biol. Chem. 278 (2003) 36993– 36998. [34] A. Kobayashi, T. Ohta, M. Yamamoto, Unique function of the Nrf2Keap1 pathway in the inducible expression of antioxidant and detoxifying enzymes, Methods Enzymol. 378 (2004) 273–286.