Introduction

Top Abstract Introduction Materials & Methods Results Discussion References

Heme oxygenase (HO) catalyzes the first and rate-limiting step in the oxidative degradation of heme to bilirubin (1). The binding of HO to the heme molecule leads to the cleavage of a mesocarbon bond, resulting in the production of biliverdin and carbon monoxide, and the release of free iron (2). Biliverdin is subsequently reduced to bilirubin by biliverdin reductase (1, 2). Two isoforms of HO, designated HO-1 and HO-2, are the products of two distinct genes and differ in molecular weight, antigenicity, and tissue distribution (2). Although HO-2 is constitutively expressed, HO-1 is induced by heme as well as by nonheme agents including heavy metals, hormones, and tumor-promoting phorbol esters (3).

Although initial interest in HO-1 focused on its role in heme metabolism, recent studies demonstrating that HO-1 is highly induced by agents that cause oxidative stress (6- 9) have generated renewed interest in the regulation and function of HO-1. Studies both in vitro and in vivo have implicated HO-1 in the protection against cellular and tissue injury mediated by heme (10) and nonheme (13) compounds, including agents known to cause oxidative stress. Although the mechanism(s) responsible for this protection is not clearly understood, it is believed that byproducts of HO catalysis, such as iron-induced ferritin and bilirubin, may reduce levels of the prooxidant heme-containing compounds (16).

Based on observations that HO-1 induction plays an important role against oxidant-induced cellular injury, and that HO-1 expression is primarily dependent on gene transcription (2, 9, 19), our laboratory has focused on delineating the transcriptional regulation of the HO-1 gene in cultured cells in response to LPS. We recently reported that activation of the HO-1 gene is mediated by a 5' distal enhancer, SX2, located 4 kb upstream from the transcription initiation site (25). In this report we describe the extension of our studies to identify whether other potential regulatory elements in the entire 5' distal flanking sequences (11-kb region) of the HO-1 gene, including the recently identified second distal enhancer AB1, located 10 kb upstream of the transcription site (23), may also help mediate LPS-induced HO-1 gene transcription. Through deletional analysis, we found that the second distal enhancer, AB1, plays an important role in LPS-mediated HO-1 gene transcription. Mutational analysis of this second distal enhancer, and electrophoretic-mobility-shift assays, further demonstrate that the transcription factor activator protein-1 (AP-1) plays a crucial role in mediating transcriptional activation of the HO-1 gene after LPS. Further, we demonstrate that the antioxidants N-acetyl-L-cysteine (NAC) and dimethyl sulfoxide (DMSO) not only attenuate LPS-induced production of reactive oxygen intermediates (ROI), but also significantly inhibit LPS-mediated increases in HO-1 messenger RNA (mRNA) and gene transcription. These data suggest that ROI serve as important upstream mediators in the signal-transduction pathways leading to AP-1-dependent LPS-induced HO-1 gene transcription.

	Materials and Methods

Top Abstract Introduction Materials & Methods Results Discussion References

Cell Culture

RAW 264.7 murine macrophage cells obtained from the American Tissue Type Collection (ATTC, Rockville, MD) were maintained in Dulbecco's modified Eagle's medium (DMEM) (Gibco-BRL Laboratories, Grand Island, NY) supplemented with 10% fetal bovine serum (FBS) (Hyclone Laboratories, Logan, UT) and gentamicin (50 mg/ ml). Cultures were maintained at 37°C in a humidified atmosphere of 5% CO₂ and 95% air.

Chemicals

LPS (Escherichia coli serotype 055:B5) was purchased from Sigma Chemical Co. (St. Louis, MO). Treatment of cells with LPS was performed in subconfluent cultures maintained in DMEM containing 10% FBS. N-acetyl-L-cysteine (NAC) was purchased from Sigma. Dimethyl sulfoxide (DMSO) was purchased from J.T. Baker, Inc. (Philipsburg, NJ). Cells were pretreated with these agents for 1 h prior to LPS treatment.

RNA Isolation and Northern Blot Analysis

Total RNA was isolated with the STAT-60 RNAzol method, with direct lysis of cells in RNAzol lysis buffer followed by chloroform extraction (Tel-Test "B" Inc., Friendswood, TX). Northern blot analysis was performed as previously described (25). Ten micrograms of total RNA was electrophoresed in a 1% agarose gel and then transferred to a nylon membrane by capillary action. The nylon membranes were then prehybridized in hybridization buffer (1% bovine serum albumin [BSA]; 7% sodium dodecylsulfate [SDS]; 0.5 M phosphate buffer, pH 7.0; and 1.0 mM ethylenediamine tetraacetic acid [EDTA]) at 65°C for 2 h, followed by incubation in hybridization buffer containing ³²P-labeled complementary DNA (cDNA) probe at 65°C for 24 h. Nylon membranes were then washed twice in wash buffer A (0.5% BSA; 5% SDS; 40 mM phosphate buffer, pH 7.0; 1 mM EDTA) for 25 min at 65°C, followed by three washes in buffer B (1% SDS; 40 mM phosphate buffer, pH 7.0; 1.0 mM EDTA) for 15 min at 65°C. Autoradiogram signals were quantified by densitometric scanning (Personal Densitometer, Model 375; Molecular Dynamics, Sunnyvale, CA). To control for variation in either the amount of RNA among samples or loading errors, membranes were hybridized with an oligonucleotide probe corresponding to 18S ribosomal RNA (rRNA). All densitometric values obtained with HO-1 mRNA were normalized to values for 18S rRNA obtained on the same blot. Quantification of steady-state HO-1 mRNA levels in treated cells was normalized to control samples.

cDNA and Oligonucleotide Probes

A full-length rat HO-1 cDNA (pHO1) was generously provided by Dr. S. Shibahara (26). pHO1 was subcloned into pBluescript vector, and a HindIII/EcoRI digestion was performed to isolate the 0.9-kb HO-1 cDNA insert. A 24-bp oligonucleotide (5'-ACG GTA TCT GAT CGT CTT CGA ACC-3') complementary to the 18S rRNA was synthesized commercially (IDT; Coralville, IA). The 0.9-kb HO-1 cDNA insert probe was labeled with (d-³²P)-deoxycytosine triphosphate ([d-³²P]CTP), using the random primer kit from Boehringer Mannheim (Mannheim, Germany). 18S rRNA oligonucleotide was labeled with (d- ³²P)-deoxyadenosine triphosphate ([d-³²P]ATP) at the 3' end with terminal deoxynucleotidyl transferase (Bethesda Research Laboratories, Gaithersburg, MD). c-fos and c-jun cDNAs were purchased from American Type Culture Collection (Rockville, MD).

Plasmids and Transfections

The construction and characterization of pHO1CAT, the mouse HO-1 promoter (1.3 kb), linked to the reporter gene chloramphenicol acetyl transferase (CAT), has been described previously (20). The construction of pHO1CAT+ SX2, which contains the 5' distal enhancer fragment of the HO-1 gene (SX2) and the 1.3 kb promoter linked to the reporter gene CAT, has also been described previously (20). The construction of pMHO3CAT, the 5' flanking region of the HO-1 gene up to 4 kb but not containing the SX2 site, linked to the reporter gene CAT, has also been described previously (20). The construction of pMHO9CAT, which contains a large portion of the 5' flanking region of the HO-1 gene, including the two distal enhancers SX2 and AB1 linked to the reporter gene CAT, was accomplished by cloning the 11.5-kb (3.5 to 15 kb) BamHI/ BamHI fragment of MH02-1 (22) into the BamHI site upstream of the HO-1 promoter in plasmid pMHO1CAT. The construction of pMHO1CAT+SX4, the 5' flanking region of the HO-1 gene upstream to the SX2 enhancer but not containing the SX2 or AB1 site, linked to the reporter gene CAT (20), pMHO1CAT+RH2, the 5' flanking region of the HO-1 gene containing the second distal enhancer AB1 site, linked to the reporter gene CAT (23), and pMHO4BX, the 5' flanking region of the HO-1 gene upstream to the SX2 enhancer but not containing the SX2 or AB1 enhancer, linked to the reporter gene CAT (20), have all been described previously. Plasmids (10 µg) were stably cotransfected into RAW 264.7 cells with pcDNA 3-Neo (1 µg), a plasmid containing neomycin selection marker, using Lipofectin Reagent (Gibco-BRL) according to manufacturer's protocol. The cells were transfected for 24 h, after which time the plates were washed twice with serum-free medium and then incubated in DMEM containing 10% FBS, gentamicin 50 µg/ml, and G418 sulfate 100 µg/ml (Geneticin; Gibco-BRL). Approximately every 3 days, the concentration of Geneticin in the medium was increased, to a maximum dose of 800 µg/ml. The surviving colonies (neomycin resistant) on each plate were pooled to establish the sublines.

Site-directed Mutagenesis

Oligonucleotide-directed mutagenesis to generate mutant plasmids pMHO1CAT-33+AB1M16 (one AP-1 binding site mutated), pMHO1CAT-33+AB1M31 (two AP-1 binding sites mutated), and pMHO1CAT-33+AB1M45 (three AP-1 binding sites mutated) has been described previously (23). The construction of the wild-type plasmid pMHO1CAT-33+AB1 (containing the second distal enhancer AB1 linked to the minimal promoter of the HO-1 gene) has been described previously (23).

CAT Assay

Cellular protein extracts were harvested within 24 h of LPS treatment. Cells were washed twice with ice-cold phosphate buffered saline (PBS), centrifuged at 2,000 × g for 5 min, then resuspended in 50 to 200 µl of 0.25 M Tris (pH 7.5). Cells were then lysed with three cycles of freezing and thawing. Debris was removed by centrifugation for 15 min at 14,000 rpm in a microcentrifuge. Protein concentrations of the supernatants used for CAT assays were determined with the Coomassie blue dye-binding assay (Bio-Rad Protein Assay; Bio-Rad Laboratories, Hercules, CA). Fifty to 100 µg of protein were incubated at 37°C for 1 to 24 h in a reaction mixture containing excess acetyl coenzyme A (Pharmacia Biotech, Uppsala, Sweden) and 0.1 µCi [¹⁴C]chloramphenicol (Amersham, Arlington Heights, IL). Chloramphenicol was extracted in 400 µl of ethyl acetate and subjected to separation by ascending thin-layer chromatography (TLC). Chloramphenicol acetylation was obtained by directly counting [¹⁴C]chloramphenicol with a Beckman LS 6000SC scintillation counter (Beckman Instruments, Inc., Fullerton, CA). Percent acetylation was determined by dividing the acetylated counts by the sum of acetylated and nonacetylated counts separated by ascending TLC. Percent chloramphenicol acetylation was obtained over the linear range of the assay for each sample. Mock transfections showed a chloramphenicol conversion range of < 0.5%.

Cellular Nuclear Protein Extraction

Cells were scraped in cold PBS and centrifuged at 14,000 × g at 4°C for 10 min. The supernatant was discarded and the cell pellet was lysed in lysis buffer containing 10 mM 4-(2-hydroxyethyl)-l-piperazine-N'-2-ethanesulfonic acid (Hepes), pH 7.9; 1 mM EDTA; 60 mM KCl; 1 mM dithiothreitol (DTT); 0.5% NP-40; and 1 mM phenylmethyl sulfonyl fluoride (PMSF). The lysate was chilled on ice for 5 min and then centrifuged at 1,500 × g to obtain nuclei. The nuclei were washed in lysis buffer without NP-40 and centrifuged again at 1,500 × g for 5 min. The supernatant was removed and the pellet was resuspended in nuclear resuspension buffer containing 25 mM Tris, pH 7.8; 60 mM KCl; 1 mM DTT; and 1 mM PMSF. The nuclei were frozen (80°C) and thawed (37°C) three times to obtain nuclear protein. The protein was kept in nuclear resuspension buffer and stored at 80°C.

Electrophoretic-mobility-shift Assay

Mobility-shift assays were performed as described by Barberis and coworkers (27) with minor modifications. JA13 (5'-GAT CCT TTT GCT GAG TCA CCC TCT GTT G-3') and its complementary sequence JA14 encompass a consensus AP-1 binding site (TGAGTCA) in the distal enhancers of the HO-1 gene (two sites in the SX2 enhancer and three sites in the AB1 enhancer). Both strands were labeled by incubating 100 ng of DNA with 30 µCi (³²P)-ATP (Amersham) and 1.0 µl polynucleotide kinase (Boehringer Mannheim) at 37°C for 45 min, followed by heat inactivation at 70°C for 10 min. The complementary oligonucleotides were annealed by incubating at 95°C for 20 min, followed by slow cooling to room temperature. DNA-binding activity was determined after incubation of 2 µg of RAW 264.7 nuclear protein extract with 20 fmol (20,000 to 50,000 cpm) of JA13/14 in reaction buffer containing 10 mM Hepes (pH 7.9); 1 mM DTT; 1 mM EDTA; 80 mM potassium chloride; 1 µg poly deoxyinosine-deoxycytosine [dIdC][dIdC]; and 4% Ficoll. After 20 min, the reaction mixture was electrophoresed on a 6% polyacrylamide gel. The gel was transferred to DE81 ion-exchange chromatography paper (Whatman, Maidstone, UK) and dried down prior to exposure to autoradiographic film. Self competitions were conducted under the same conditions, using 50- and 100-fold molar excesses of the unlabeled JA13/ 14 oligonucleotide probe. Nonspecific competitions were performed similarly, using a 50-fold excess of an unlabeled oligonucleotide probe encompassing an Sp1 transcription-factor binding site (5'-GAT CGA TCG GGG CGG GGC GAT C-3') (Stratagene, La Jolla, CA).

Determination of Intracellular Oxidants by Flow Cytometry

The fluorescent probe dichlorodihydrofluorescein diacetate (H₂-DCF; Molecular Probes, Eugene, OR) was used to determine intracellular levels of reactive oxygen species (28). Cells were adjusted to 1 × 10⁶ per ml in HBSS, and were loaded by incubation with 10 µM 2',7'-dichlorodihydrofluorescein diacetate (H₂-DCF) for 15 min at 37°C. Cell samples were then exposed to LPS (1 µg/ml) at 37°C for 20 min, and allowed to reach room temperature for analyses of the conversion of the nonfluorescent H₂-DCF to the green fluorescent dichlorofluorescein (DCF). This indicator of cellular oxidation (DCF fluorescence) was determined with a FACStar+ Flow Cytometer (Becton Dickinson Immunocytometry Systems, San Jose, CA). For NAC and DMSO experiments, cells were pretreated with NAC or DMSO for 5 min prior to LPS treatment.

Statistical Analysis

Data are expressed as the mean ± SEM. Differences in measured variables between experimental and control group were assessed with Student's t tests. Statistical calculations were performed on a Macintosh personal computer (Apple, Inc., Cupertino, CA), using the Statview II Statistical Package (Abacus Concepts, Berkeley, CA). Statistical difference was accepted at P < 0.05.

	Results

Top Abstract Introduction Materials & Methods Results Discussion References

Deletional Analyses Demonstrate that Both Distal Enhancers, SX2 and AB1, Mediate HO-1 Gene Activation

Previously, we showed that the SX2 enhancer is important in LPS-mediated HO-1 gene transcription (25). In order to identify any other LPS-responsive regulatory elements in the upstream 5' flanking sequences of the HO-1 gene, we performed deletional analysis of the entire distal upstream 11-kb region, using the reporter gene CAT. Our laboratory has recently identified a second distal enhancer, AB1, located 10 kb upstream of the transcription site, and we were particularly interested in whether this enhancer is functional in LPS-induced HO-1 gene transcription. RAW 264.7 cells were transfected with plasmids containing various fragments of the 11-kb 5' upstream region linked to the reporter CAT gene (Figure 1A). The stable transfectants were treated with LPS, and CAT activity assays were performed within 24 h. Figure 1B shows that the HO-1 gene promoter (pMHO1CAT) is not functionally important in the transcriptional activation of the HO-1 gene by LPS. However, the HO-1 gene requires either the SX2 or AB1 distal enhancer for activation in response to LPS. Increased CAT activity was observed only in those cells containing the SX2 enhancer (pMHO1CAT+SX2), the AB1 enhancer (pMHO1CAT+RH2), or both enhancers (pMHO9CAT) (Figure 1B). The pMHO3CAT and pMHO1CAT+SX4 cell lines, which did not contain either the SX2 or the AB1 enhancer, appeared to have reduced CAT activity after LPS, but this was not statistically significant. The pMHO4 CATBX cell line, which encompasses distal sequences further upstream of the SX2, not including the SX2 or AB1 enhancers, also did not respond to LPS treatment.

View larger version (19K): [in this window] [in a new window]   Figure 1.   Deletional analysis of the 11-kb upstream region of the HO-1 gene in RAW 264.7 cells after LPS treatment. (A) A structural organization of the mouse HO-1 gene is shown, with Position +1 denoting the transcription initiation site. The locations of the 1.3-kb promoter and the distal enhancers SX2 (4 kb upstream of the transcription initiation site) and AB1 (10 kb upstream of the transcription initiation site) are illustrated. Solid line = encompassing regions; dotted lines = deleted regions. (B) Cells were stably transfected with the various chimeras of the 5' distal upstream region of the HO-1 gene as indicated in (A) Cells were harvested 24 h following LPS treatment (1 µg/ml) and assayed for CAT activity as described in MATERIALS AND METHODS. CAT activity data are expressed as percent acetylation, and represent the mean ± SE of three independent experiments. *P < 0.05 compared with controls. Open bars, untreated controls; enclosed bars, LPS.

Mutational Analyses of the AB1 Enhancer Show That AP-1 is Essential for Activation of the HO-1 Gene Following LPS

The pMHO1CAT+RH2 construct, which exhibited significantly increased CAT activity after addition of LPS (Figure 1B), is of strong interest to our laboratory since it contains the recently identified second distal enhancer AB1 (23). The AB1 enhancer contains three putative DNA binding sites for the transcription factor AP-1. Chimeric DNA containing either a wild-type AB1 enhancer (AB1) or mutant plasmids containing an AB1 enhancer with mutations in one AP-1 site (AB1M16), two AP-1 sites (AB1M31), or three AP-1 sites (AB1M45) linked to the CAT reporter gene were stably transfected into RAW 264.7 cells (Figure 2A). The transfectants were treated with LPS for 24 h, at which time cellular protein extracts were assayed for CAT activity. Increased CAT activity after LPS treatment was seen in those cell lines containing the wild-type AB1 enhancer (AB1) or the AB1 enhancer with the proximal mutated AP-1 binding site (AB1M16) (Figure 2B). However, mutations of either the two proximal (AB1M31) or all three (AB1M45) AP-1 binding sites abolished LPS-induced CAT activity (Figure 2B).

View larger version (14K): [in this window] [in a new window]   Figure 2.   Mutational analysis of the AB1 distal enhancer of the HO-1 gene in RAW 264.7 cells after LPS treatment. (A) Putative AP-1 elements of the AB1 distal enhancer fragment of the mouse HO-1 gene. X denotes mutated AP-1 sites. 1. AB1 (pMHO1 CAT-33+AB1) wild-type cells with intact AP-1 sites. 2. AB1M16 (pMHO1CAT-33+AB1M16) mutant cells with one mutated AP-1 site. 3. AB1M31 (pMHO1CAT-33+AB1M31) mutant cells with two mutated AP-1 sites. 4. AB1M45 (pMHO1CAT-33+AB1 M45) mutant cells with three mutated AP-1 sites. (B) Stably transfected wild-type cells (1) AB1, mutants (2) AB1M16, (3) AB1M31, or (4) AB1M45 cells were treated with LPS (1 µg/ml) for 24 h, at which time cells were harvested and assayed for CAT activity. CAT activity data are expressed as percent acetylation and represent the mean ± SE of three independent experiments. *P < 0.05 compared with controls. #P < 0.05 compared with wild-type AB1 cells treated with LPS. **P < 0.05 compared with mutant AB1M16 cells treated with LPS.

LPS Increases AP-1 Binding Activity in RAW 264.7 Cells

To further confirm that AP-1 plays an important role in LPS-induced HO-1 gene activation, we looked at AP-1 DNA binding activity after treatment with LPS in RAW 264.7 cells. Nuclear proteins were isolated at 0.5 h, 1 h, 2 h, 4 h, 8 h, and 25 h of LPS treatment, and were incubated with JA13/14, a double-stranded DNA probe that encompasses AP-1 binding sites (5'-GAT CCT TTT GCT GAG TCA CCC TCT GTT G-3') in the distal enhancers of the HO-1 gene. Electrophoretic-mobility-shift assays revealed AP-1 binding activity after 1 h of LPS treatment (Figure 3A). The specificity of AP-1 DNA binding activity was demonstrated by the ability of excess unlabeled JA13/14 oligonucleotide (50-fold or 100-fold excess) to compete with the radiolabeled AP-1 sequence (Figure 3B). An unlabeled Sp1 (50-fold excess) oligonucleotide containing an unrelated consensus sequence did not compete with radiolabeled JA13/14 probe (Figure 3B).

View larger version (53K): [in this window] [in a new window]   Figure 3.   Electrophoretic-mobility-shift assays of nuclear protein extracts for AP-1 binding activity following LPS treatment. (A) Kinetics of AP-1 DNA binding activity in RAW 264.7 cells following LPS treatment. Nuclear protein extracts were obtained from cells at the indicated times following LPS (1 µg/ml) treatment, and were analyzed for AP-1 binding activity. (B) Competition of AP-1 binding activity in RAW 264.7 cells following LPS treatment. Self competitions were performed with nuclear protein extracts obtained from cells treated with LPS (1 µg/ml) for 1 h in the presence of a 50-fold or 100-fold excess of unlabeled JA13/14. Noncompetitions were performed with nuclear protein extracts obtained from cells treated with LPS (1 µg/ ml) for 1 h in the presence of a 50-fold excess of labeled SP-1.

LPS Induces Expression of AP-1 c-fos and c-jun mRNA

Based on the observation described previously that LPS increased AP-1 DNA binding activity, we further examined whether LPS could also increase expression of c-fos and c-jun, two major members of the AP-1 transcription-factor complex. Total RNA was isolated from RAW 264.7 cells at 15 min, 30 min, 1 h, 2 h, and 4 h of LPS treatment and then investigated for c-fos and c-jun mRNA with Northern blot analyses. As shown in Figure 4, LPS treatment led to a rapid and transient increase in the steady-state levels of both c-fos and c-jun mRNA. Increased c-fos mRNA expression occurred as early as with 15 min of LPS treatment, and remained elevated for up to 30 min of LPS treatment. c-jun mRNA expression also increased after 15 min of LPS treatment, peaked at 30 min, and remained elevated for up to 1 h.

View larger version (40K): [in this window] [in a new window]   Figure 4.   Northern blot analysis of c-fos and c-jun mRNA in RAW 264.7 cells following LPS treatment. Total RNA was isolated from cells at the indicated times following LPS (1 µg/ml) treatment and analyzed c-fos and c-jun mRNA expression through Northern blot analyses. 18S rRNA hybridization is shown as a normalization control.

LPS Induces Production of ROI

It has been shown that HO-1 is induced by a variety of agents that lead to oxidative stress, including hydrogen peroxide, UV irradiation, heavy metals, and hyperoxia (6, 7, 20, 24). Since LPS is known to prime inflammatory cells to produce and secrete ROI, we hypothesized that ROI serve as upstream signaling molecules that mediate LPS-induced HO-1 gene expression and transcription. First, we examined whether LPS directly increases levels of intracellular ROI in RAW 264.7 cells. Cells were stained with H₂-DCF fluorescent probe and analyzed for oxidized DCF fluorescence through flow cytometry, after stimulation of the cells with LPS. Figure 5 shows increased DCF fluorescence in cells treated with LPS compared with unstimulated cells. A threefold increase in the number of cells staining for DCF fluorescence was observed among the LPS-treated cells (Figure 5B) as compared with control cells (Figure 5A).

View larger version (25K): [in this window] [in a new window]   Figure 5.   Determination of intracellular oxidants by DCF fluorescence in RAW 264.7 cells following LPS treatment. Single-parameter histogram plots were generated from DCF fluorescence in cells after LPS treatment. The x axis denotes the intensity of intracellular DCF fluorescence; the y axis denotes the number of cells. (A) untreated control cells. (B) LPS. (C) NAC/ LPS. (D) DMSO/LPS.

Antioxidants Attenuate LPS-mediated HO-1 Gene Expression and Transcription

To further investigate the role of ROI in the activation of the HO-1 gene, we studied the effect of antioxidants on HO-1 mRNA expression and gene transcription. RAW 264.7 cells were incubated with the antioxidants NAC (20 mM) or DMSO (2%) prior to treatment with LPS. Total RNA was isolated after 24 h of treatment with LPS and was then analyzed for HO-1 mRNA expression with Northern blot analyses. Figure 6 shows that the antioxidants NAC and DMSO significantly inhibited LPS-induced HO-1 mRNA expression. To determine whether ROI played a role in LPS-mediated HO-1 gene activation, we pretreated the wild-type pMHO1CAT-33+AB1 stable transfectants with NAC (20 mM) or DMSO (1%) for 1 h prior to treatment with LPS. Transcriptional activation of the HO-1 gene in these cells by LPS was significantly attenuated in cells treated with the antioxidants NAC and DMSO (Figure 7). Furthermore, the antioxidants not only attenuated LPS-induced HO-1 gene expression and transcription, but also blunted LPS-induced production of ROI (NAC/LPS, 42% inhibition as compared with LPS; DMSO/LPS, 40% inhibition as compared with LPS) (Figures 5C and 5D). The intensity of DCF fluorescence in cells treated with NAC or DMSO alone was similar to that in untreated control cells (data not shown). These data suggest that ROI may serve as signaling molecules in mediating LPS-induced HO-1 gene transcription.

View larger version (35K): [in this window] [in a new window]   Figure 6.   Effect of antioxidants on LPS-induced HO-1 mRNA induction in RAW 264.7 cells following LPS treatment. RAW 264.7 cells were pretreated with DMSO (2%) or NAC (20 mM) for 1 h prior to incubation with LPS (1 µg/ml). Total RNA was isolated after 24 h of LPS treatment and analyzed for HO-1 mRNA expression through Northern blot analysis. 18S rRNA is shown as a normalization control. Data shown are representative blots of two independent experiments.

View larger version (15K): [in this window] [in a new window]   Figure 7.   Effect of antioxidants on LPS-induced HO-1 gene transcription. AB1 (pMHO1CAT-33+AB1) wild-type cells were pretreated with DMSO (1%) or NAC (20 mM) for 1 h prior to incubation with LPS (1 µg/ml). Cells were harvested after 24 h of LPS treatment and assayed for CAT activity. CAT activity data are expressed as percent acetylation, and represent the mean ± SE of four or five independent experiments. *P < 0.001 compared with controls. **P < 0.001 (NAC/LPS) compared with LPS. **P < 0.05 (DMSO/LPS) compared with LPS.

	Discussion

Top Abstract Introduction Materials & Methods Results Discussion References

We have previously reported that LPS induces high levels of HO-1 expression in vivo and in vitro (25), and that HO-1 plays an important functional role in providing protection against LPS (19, 29). On the basis of these observations and the finding that HO-1 expression is regulated at the transcriptional level in response to other known inducing agents so far tested, our laboratory has focused on delineating the transcriptional regulation of HO-1 after addition of LPS. To this end, we recently reported that LPS- induced HO-1 expression depends on gene transcription, and that the induction of HO-1 by LPS requires the activation of a distal enhancer called SX2, located 4 kb upstream from the HO-1 transcription initiation site (25). In the study reported here, we extended our investigations to whether other regulatory elements within the entire 11-kb flanking region 5' to the HO-1 gene transcription site, including the recently identified second distal enhancer AB1 (23), are functionally active in mediating HO-1 gene transcription after addition of LPS. We also examined the role of ROI in the signal-transduction pathway(s) leading to LPS-mediated HO-1 gene activation.

To date, the mouse HO-1 gene is the best characterized among the HO-1 genes of different species (20). The first distal enhancer, SX2, is a 268-bp fragment located approximately 4 kb upstream from the transcription initiation site (22). The second distal enhancer, AB1, is a recently identified 161-bp fragment located 10 kb upstream from the HO-1 transcription start site (23). The current study strongly suggests that LPS-induced HO-1 gene transcription is primarily dependent on 5' distal enhancer elements, and not on the promoter of the HO-1 gene. Our observation that the mouse HO-1 gene promoter (Figure 1B) does not function in response to LPS is similar to observations reported for other inducers including cadmium, heme, and 12-O-tetradecanoylphorbol-13-acetate (20, 22). Interestingly, however, the promoters of the rat and human HO-1 genes have been shown to be functionally active in response to various inducers such as heat shock and UV irradiation (30). Further, the HO-1 gene requires either the SX2 or AB1 distal enhancer for activation in response to LPS (Figure 1B), in that cell lines not having either of these two distal enhancers were unresponsive to LPS. This dependency on distal enhancers alone for transcriptional activation of HO-1 by LPS is different when compared with other known inducers of HO-1 gene activation. For example, distal enhancers alone are not sufficient for HO-1 gene induction in response to hyperoxia. Hyperoxia-induced HO-1 gene transcription requires both the distal enhancer SX2 and the promoter of the HO-1 gene (24). Further, the complexity of the transcriptional regulation of the HO-1 gene is highlighted by our recent observations that neither the promoter nor distal enhancers are required for hypoxia-induced HO-1 gene activation, in that hypoxia-inducible factor-1 (HIF-1), which mediates hypoxia-induced HO-1 gene transcription, binds to HIF-1 binding sites proximal to the AB1 distal enhancer (33). Interestingly, we have also observed increased CAT activity after interleukin-6 (IL-6) treatment in pMHO1CAT+SX4 transfectants that do not contain either of the distal enhancers (unpublished data).

The first distal enhancer, SX2, contains two AP-1 binding sites and two CCAAT-enhancer-binding-protein (C/EBP) binding sites. Each of these binding sites is necessary for full transcriptional activity in response to cadmium, although the AP-1-binding elements appear to be more important (21). Induction of increased transcriptional activity of the SX2 enhancer by LPS, however, depends only on AP-1 activation and binding (25). The second distal enhancer, AB1, contains three putative AP-1-binding sites (23). In this report we demonstrate that this second distal enhancer mediates LPS-induced HO-1 gene transcription through its AP-1 DNA-binding sites. On the basis of our mutational analyses (Figure 2B), we conclude that the two distal AP-1 binding sites in the AB1 enhancer are sufficient for LPS-induced HO-1 gene transcription. We have further demonstrated a role for AP-1 in LPS-mediated HO-1 gene induction through electrophoretic-mobility-shift assays showing increased AP-1 DNA-binding activity in cells after LPS treatment. It is interesting that LPS also induces expression of c-fos and c-jun mRNA (Figure 4). Our data support findings by Kaminska and coworkers (34), who demonstrated that LPS induced both c-Fos and c-Jun protein expression in human monocytes, with kinetics similar to those we observed in our study. Interestingly, unlike the mouse HO-1 gene, the human HO-1 promoter contains a nuclear factor-kappa B (NF-B)-responsive element (35). Kurata and associates recently reported that activation of the human HO-1 gene by LPS is mediated by NF-B rather than by AP-1 (36).

There is evidence that LPS causes cellular and tissue damage, presumably through the production of ROI (37, 38). LPS has also been shown to increase lung-tissue lipid peroxidation in unanesthetized sheep (39), and antioxidants can protect against LPS-induced lung injury in sheep (40, 41) and rodents (42). Using flow cytometric analyses of DCF-stained cells (Figure 5), we provide evidence that LPS causes increased production of ROI in RAW 264.7 macrophages. We also demonstrate that the antioxidants NAC and DMSO attenuate LPS-induced production of ROI. In addition, NAC and DMSO inhibit LPS-induced HO-1 gene expression and transcription (Figures 6 and 7) at the same concentrations required to attenuate LPS-induced ROI production (Figures 5C and 5D). Our data are consistent with those of Rizzardini and associates (43) and Kurata and colleagues (36), who have reported that NAC can reduce HO-1 accumulation in the livers of LPS-treated rats and in mouse M1 cells, respectively.

In summary, we have shown that LPS activates HO-1 gene transcription by inducing the production of ROI, which serve as early signaling molecules triggering signal-transduction pathway(s) that lead to the activation of AP-1-dependent HO-1 gene transcription. Future studies will focus on identifying the trans-acting factors responsible for LPS-induced HO-1 gene transcription, and on delineating the HO-1-mediated mechanism(s) responsible for protection against LPS, using transgenic approaches.