Portal de Periódicos da CAPES

H2O2 and Tumor Necrosis Factor-α Activate Intercellular Adhesion Molecule 1 (ICAM-1) Gene Transcription through Distinct cis-Regulatory Elements within the ICAM-1 Promoter

1995; Elsevier BV; Volume: 270; Issue: 32 Linguagem: Inglês

10.1074/jbc.270.32.18966

1083-351X

Kenneth A. Roebuck, Arshad Rahman, Venkatesh Lakshminarayanan, K. Janakidevi, Asrar B. Malik,

Inflammatory mediators and NSAID effects

We investigated the mechanisms by which H2O2 increases intercellular adhesion molecule 1 (ICAM-1; CD54) expression in endothelial cells. The H2O2-induced increase in ICAM-1 mRNA was inhibited by actinomycin D, by the antioxidant N-acetylcysteine, and by 3-aminobenzamide (which blocks oxidant-induced AP-1 activity), but not by pyrrolidine dithiocarbamate (which blocks oxidant-induced NF-κB activity). Nuclear run-on and transient transfections of ICAM-1 promoter constructs indicated that H2O2 stimulated ICAM-1 gene transcription by activation of a distinct region of the ICAM-1 promoter. The H2O2-responsive element was localized to sequences between −981 and −769 (relative to the start codon). Located within this region are two 16-base pair repeats, each containing binding sites for the transcription factors AP-1 and Ets. A similar composite AP-1/Ets element isolated from the macrophage scavenger receptor gene conferred H2O2 responsiveness to a minimal promoter. Mutation of the 16-base pair repeats within the ICAM-1 promoter prevented H2O2-induced DNA binding activity, and their deletion abrogated the H2O2-induced transcriptional activity. In contrast, TNFα induced ICAM-1 transcription via activation of promoter sequences between −393 and −176, a region with C/EBP and NF-κB binding sites. The results indicate that H2O2 activates ICAM-1 transcription through AP-1/Ets elements within the ICAM-1 promoter, which are distinct from NF-κB-mediated ICAM-1 expression induced by TNFα. We investigated the mechanisms by which H2O2 increases intercellular adhesion molecule 1 (ICAM-1; CD54) expression in endothelial cells. The H2O2-induced increase in ICAM-1 mRNA was inhibited by actinomycin D, by the antioxidant N-acetylcysteine, and by 3-aminobenzamide (which blocks oxidant-induced AP-1 activity), but not by pyrrolidine dithiocarbamate (which blocks oxidant-induced NF-κB activity). Nuclear run-on and transient transfections of ICAM-1 promoter constructs indicated that H2O2 stimulated ICAM-1 gene transcription by activation of a distinct region of the ICAM-1 promoter. The H2O2-responsive element was localized to sequences between −981 and −769 (relative to the start codon). Located within this region are two 16-base pair repeats, each containing binding sites for the transcription factors AP-1 and Ets. A similar composite AP-1/Ets element isolated from the macrophage scavenger receptor gene conferred H2O2 responsiveness to a minimal promoter. Mutation of the 16-base pair repeats within the ICAM-1 promoter prevented H2O2-induced DNA binding activity, and their deletion abrogated the H2O2-induced transcriptional activity. In contrast, TNFα induced ICAM-1 transcription via activation of promoter sequences between −393 and −176, a region with C/EBP and NF-κB binding sites. The results indicate that H2O2 activates ICAM-1 transcription through AP-1/Ets elements within the ICAM-1 promoter, which are distinct from NF-κB-mediated ICAM-1 expression induced by TNFα. Adhesion of circulating polymorphonuclear leukocytes (PMN) 1The abbreviations used are: PMNpolymorphonuclear leukocytesICAM-1intercellular adhesion molecule 1LUCluciferaseHUVEChuman umbilical vein endothelial cellTNFαtumor necrosis factor-αN-Cys(Ac)N-acetylcysteinePDTCpyrrolidine dithiocarbamate3-AB3-aminobenzamideAP-1activator protein-1NF-κBnuclear factor κBAREanti-oxidant response elementMSRmacrophage scavenger receptorbpbase pair(s)kbkilobase(s)DMEMDulbecco's modified Eagle's mediumMOPS4-morpholinepropanesulfonic acidPMSFphenylmethylsulfonyl fluoridePBSphosphate-buffered salineDTTdithiothreitolGSTglutathione S-transferase. 1The abbreviations used are: PMNpolymorphonuclear leukocytesICAM-1intercellular adhesion molecule 1LUCluciferaseHUVEChuman umbilical vein endothelial cellTNFαtumor necrosis factor-αN-Cys(Ac)N-acetylcysteinePDTCpyrrolidine dithiocarbamate3-AB3-aminobenzamideAP-1activator protein-1NF-κBnuclear factor κBAREanti-oxidant response elementMSRmacrophage scavenger receptorbpbase pair(s)kbkilobase(s)DMEMDulbecco's modified Eagle's mediumMOPS4-morpholinepropanesulfonic acidPMSFphenylmethylsulfonyl fluoridePBSphosphate-buffered salineDTTdithiothreitolGSTglutathione S-transferase. to the vascular endothelium is a critical step in the inflammatory response (31Nourshargh S. Williams T.J. Warren J.B. The Endothelium: An Introduction to Current Research. Wiley-Liss, New York1990: 171-186Google Scholar). PMN adhesion to the endothelium occurs during reperfusion of tissues when reactive oxygen intermediates such as H2O2 are generated (13Hernandez L.A. Grisham M.B. Twohig B. Arfors K.E. Harlan J.M. Granger D.N. Am. J. Physiol. 1987; 253: H699-H703PubMed Google Scholar). The adhesion event is mediated by molecules present or expressed on the surface of endothelial cells and PMN (21Lo S.K. Van Seventer G.A. Levin S.M. Wright S.D. J. Immunol. 1989; 169: 1779-1793Google Scholar). Endothelial cells express intercellular adhesion molecule 1 (ICAM-1; CD54), a counter-receptor for CD11/CD18 integrin (10Dustin M.L. Rothlein R. Bahn A.K. Dinarello C.A. Springer T.A. J. Immunol. 1988; 137: 245-254Google Scholar) that promotes adhesion and transendothelial migration of PMN (39Smith C.W. Marlin S.D. Rothlein R. Toman C. Anderson D.C. J. Clin. Invest. 1989; 83: 2008-2017Crossref PubMed Scopus (939) Google Scholar). Studies using monoclonal antibodies show that increased cell surface ICAM-1 expression is required for migration of PMN to sites of inflammation and PMN-mediated endothelial injury associated with reperfusion (16Kukielka G.L. Hawkins H.K. Michael L. Manning A.M. Youker K. Lane C. Entman M.L. Smith C.W. Anderson D.C. J. Clin. Invest. 1993; 92: 1504-1516Crossref PubMed Scopus (214) Google Scholar). ICAM-1 gene expression is induced by tumor necrosis factor-α (TNFα), interferon γ, and interleukin-1β (26Myers C.L. Wertheimer S.J. Schembri-King J. Parks T.P. Wallace R.W. Am. J. Physiol. 1992; 262: C365-C373Crossref PubMed Google Scholar; 26Myers C.L. Wertheimer S.J. Schembri-King J. Parks T.P. Wallace R.W. Am. J. Physiol. 1992; 262: C365-C373Crossref PubMed Google Scholar; 23Look D.C. Pelletier M.R. Holtzman M.J. J. Biol. Chem. 1994; 269: 8952-8958Abstract Full Text PDF PubMed Google Scholar). polymorphonuclear leukocytes intercellular adhesion molecule 1 luciferase human umbilical vein endothelial cell tumor necrosis factor-α N-acetylcysteine pyrrolidine dithiocarbamate 3-aminobenzamide activator protein-1 nuclear factor κB anti-oxidant response element macrophage scavenger receptor base pair(s) kilobase(s) Dulbecco's modified Eagle's medium 4-morpholinepropanesulfonic acid phenylmethylsulfonyl fluoride phosphate-buffered saline dithiothreitol glutathione S-transferase. polymorphonuclear leukocytes intercellular adhesion molecule 1 luciferase human umbilical vein endothelial cell tumor necrosis factor-α N-acetylcysteine pyrrolidine dithiocarbamate 3-aminobenzamide activator protein-1 nuclear factor κB anti-oxidant response element macrophage scavenger receptor base pair(s) kilobase(s) Dulbecco's modified Eagle's medium 4-morpholinepropanesulfonic acid phenylmethylsulfonyl fluoride phosphate-buffered saline dithiothreitol glutathione S-transferase. Recent studies show that the reactive oxidant, H2O2, also promotes ICAM-1 expression in endothelial cells and ICAM-1-dependent adhesion of PMN (22Lo S.K. Janakidevi K. Lai L. Malik A.B. Am. J. Physiol. 1993; 264: L406-L412Crossref PubMed Google Scholar; 5Bradely J.R. Johnson D.R. Pober J.S. Am. J. Pathol. 1993; 142: 1598-1607PubMed Google Scholar; 36Sellak H. Franzini E. Hakim J. Pasquier C. Blood. 1994; 83: 2669-2677Crossref PubMed Google Scholar). H2O2 was recently reported to also increase ICAM-1 expression on keratinocytes (15Ikeda M. Schroeder K.K. Mosher L.B. Woods C.W. Akeson A.L. J. Invest. Dermatol. 1994; 103: 791-796Crossref PubMed Scopus (33) Google Scholar). In human umbilical vein endothelial cells (HUVEC), we found that oxidant-induced ICAM-1 expression was associated with increased ICAM-1 mRNA levels occurring 1 h after H2O2 exposure (22Lo S.K. Janakidevi K. Lai L. Malik A.B. Am. J. Physiol. 1993; 264: L406-L412Crossref PubMed Google Scholar). H2O2 activates transcription factors, AP-1 and NF-κB, in a mouse osteoblastic cell line (30Nose K. Shibanuma M. Kikuchi K. Kageyama H. Sakiyama S. Kuroke T. Eur. J. Biochem. 1991; 201: 99-106Crossref PubMed Scopus (269) Google Scholar) and in HeLa and Jurkat cells (25Meyer M. Schreck R. Baeuerle P.A. EMBO J. 1993; 12: 2005-2015Crossref PubMed Scopus (1257) Google Scholar). The ICAM-1 gene contains a number of AP-1-like and NF-κB-like binding sites within its promoter region (41Voraberger G. Schafer R. Stratowa C. J. Immunol. 1991; 147: 2777-2786PubMed Google Scholar). Taken together, these observations suggest that the activation of these transcription factors by H2O2 may be a mechanism of endothelial ICAM-1 gene expression. In this study, we examined the basis of H2O2-induced ICAM-1 expression in endothelial cells. We showed that H2O2 activated ICAM-1 gene transcription via a 212-base pair (bp) promoter region between 981 and 769 bp upstream of the coding sequences. This region contained two 16-bp repeats which are binding sites for the transcription factors AP-1 and Ets. AP-1/Ets composite elements were shown to be sufficient to mediate H2O2-induced transcription. Although the AP-1/Ets elements also responded to TNFα, the TNFα-induced ICAM-1 expression was mediated by promoter sequences between 393 and 176 bp upstream of the gene, containing binding sites for C/EBP and NF-κB. Therefore, H2O2 and TNFα activate ICAM-1 gene transcription in endothelial cells through distinct cis-regulatory elements within the ICAM-1 promoter. The results identify a novel oxidant response element and indicate that mediator-specific regulation of ICAM-1 expression involves the interaction of multiple factors with the ICAM-1 promoter. Diethylpyrocarbonate, DMEM, heparin, HEPES, 3.0% H2O2, MOPS, PMSF, spermidine, spermine, pyrrolidine dithiocarbamate (PDTC), and N-acetylcysteine (N-Cys(Ac)) were purchased from Sigma. We purchased 3-aminobenzamide (3-AB) from Pfaltz and Bauer (Stamford, CT). Guanidine thiocyanate, restriction enzymes, random primer labeling kit, QuikHybTM hybridization mix, and Duralose-UVTM nitrocellulose membranes were purchased from Stratagene. Human ICAM-1 cDNA was provided by Dr. T. Springer, Harvard Medical School, Boston, MA. Plasmid containing the cDNA for rRNA was provided by Dr. M. L. Brown, Boehringer-Ingelheim Pharmaceuticals, Ridgefield, CT. Plasmid EL1-BS, containing partial human E-selectin cDNA, was provided by Dr. L. Osborn, Biogen, Cambridge, MA. Agarose, actinomycin D, LipofectAMINE, and RPMI were purchased from Life Technologies, Inc. Riboprobe Gemini System II and RNase-free DNase were purchased from Promega Biotech. Fetal bovine serum was obtained from Hyclone Laboratories. [α-32P]dCTP (3,000 Ci/mmol), [γ-32P]ATP (3,000 Ci/mmol), and [α-32P]UTP (3,000 Ci/mmol) were purchased from DuPont NEN. The antisense oligomer ISIS 1570 (5′-TGGGAGCCATAGCGAGGCTGA-3′) to the 5′ end of the ICAM-1 cDNA and a nonsense oligomer (5′-AGTCGGAGCGATACCGAGGGT-3′) were generous gifts from Sterling-Winthrop Drug Co., Rensselaer, NY. Oligonucleotides were purchased from Integrated DNA Technologies Inc. (Coralville, IA). ICAM-1 luciferase reporter gene plasmids were gifts from Dr. C. Stratowa, Vienna, Austria. Human umbilical vein endothelial cells (HUVEC) at the first passage were purchased from Clonetics Corp. (San Diego, CA). HUVEC were grown on fibronectin-coated flasks or plates in RPMI medium containing 10-20% fetal calf serum, 6.5 μg/ml endothelial-derived growth factor from bovine neural tissue, and 75 μg/ml heparin. All experiments used cells under the eighth passage. EAhy926 cells, a hybrid cell line of HUVEC and A549 cell line (derived from human lung epithelial type II cells), was provided by Dr. Edgell (University of North Carolina, Chapel Hill) and cultured as described (11Edgell C.-J. McDonald C.C. Graham J.B. Proc. Natl. Acad. Sci. U. S. A. 1983; 80: 3734-3737Crossref PubMed Scopus (1330) Google Scholar). EAhy926 cells retain endothelial cell morphology and express the endothelial cell-specific marker human factor VIII-related antigen (11Edgell C.-J. McDonald C.C. Graham J.B. Proc. Natl. Acad. Sci. U. S. A. 1983; 80: 3734-3737Crossref PubMed Scopus (1330) Google Scholar). EAhy926 cells were maintained in DMEM-high glucose, in 5% CO2, 10% fetal calf serum, and passaged by removal in trypsin-EDTA buffer (0.14 M NaCl, 2.68 mM KCl, 0.42 mM NaH2PO4, 0.012 M NaHCO3, 0.01 M dextrose, 0.05% trypsin, 0.53 mM EDTA). Confluent cells were washed twice with serum-free DMEM (without phenol red) containing 20 mM HEPES and incubated for 2 h before treatment with the agents described below. The experiments using the inhibitors (PDTC, N-Cys(Ac), or 3-AB) required a 1-h preincubation period in serum-free medium with each inhibitor, and treatment was continued during the 1-h H2O2 exposure period. The ICAM-1 LUC reporter plasmid and its 5′ deletion derivatives have been described previously (41Voraberger G. Schafer R. Stratowa C. J. Immunol. 1991; 147: 2777-2786PubMed Google Scholar). The full-length ICAM-1 promoter construct contains approximately 1.4 kb of ICAM-1 5′-flanking DNA linked to the firefly luciferase (LUC) gene. Transfection into cells showed that this ICAM-1 construct was responsive to phorbol 12-myristate 13-acetate and TNFα (41Voraberger G. Schafer R. Stratowa C. J. Immunol. 1991; 147: 2777-2786PubMed Google Scholar). The macrophage scavenger receptor constructs containing three copies of the AP-1/Ets element or mutations of the element linked to luciferase have been described previously (45Wu H. Moulton K. Horval A. Parik S. Glass C.K. Mol. Cell. Biol. 1994; 14: 2129-2139Crossref PubMed Google Scholar). Cells were plated 24 h prior to transfection at 5 × 105 cells per 6-cm plate. The cells were refed with fresh medium containing 10% fetal calf serum 4 h before lipofection (24Malone R.W. Felgner P.L. Verma I.M. Proc. Natl. Acad. Sci. U. S. A. 1989; 86: 6077-6081Crossref PubMed Scopus (489) Google Scholar) with LipofectAMINE as described by Life Technologies, Inc. Transfection of 100-mm plates at 80% confluency typically contained 8 μg of reporter plasmid (ICAM-1 LUC) and 2 μg of β-gal expression plasmid DNA. The cells were transfected for 5 to 14 h. After a recovery period, the cells were divided into five 35-mm plates. At 24 h before treatment with H2O2 (concentration range 100-400 μM), TNFα (100 units/ml) or phorbol 12-myristate 13-acetate (50 ng/ml), the cells were incubated in medium containing 0.5% fetal calf serum. The cell extract was prepared and assayed for luciferase activity using Promega Biotec assay systems and β-galactosidase activity using the Tropix (Bedford, MA) assay system. Protein content was determined using a Bio-Rad protein determination kit. Mean luciferase activity per μg of protein extract was normalized to the β-galactosidase activity (which in control experiments was not affected by H2O2). All solutions used for RNA analysis were treated with diethylpyrocarbonate (0.1%) and sterilized or prepared in sterile diethylpyrocarbonate-treated water. Glassware was baked at 240°C for a minimum of 4 h to remove traces of RNase. Total RNA was isolated according to the procedure of 8Chomczynski P. Sacchi N. Anal. Biochem. 1987; 162: 156-159Crossref PubMed Scopus (62787) Google Scholar. Medium was removed, and the endothelial cell layer was rinsed with ice-cold, Ca2+ -and Mg2+-free phosphate-buffered saline (PBS), and lysed in acid guanidine thiocyanate. The lysate was drawn through a 26-gauge needle and extracted with acid phenol/chloroform (5:1). After a 30-60-min incubation on ice, the mixture was centrifuged for 30 min at 12,000 × g. The aqueous phase was collected and RNA was precipitated with equal volume of ice-cold isopropyl alcohol. After allowing the RNA to precipitate for 1 h at −70°C, RNA was pelleted by centrifugation for 30 min at 12,000 × g. The RNA pellet was washed twice with 75% ethanol, briefly dried, and dissolved in 0.5% SDS in diethylpyrocarbonate-treated water. Quantification and purity of RNA were assessed by A260/A280 absorption, and RNA samples with ratios above 1.9 were used for further analysis. The RNA samples (20 μg/lane) were subjected to gel electrophoresis in denaturing 1% formaldehyde-agarose gels and transferred overnight in 20 × SSC (3 M sodium chloride, 0.3 M sodium citrate, pH 7.0) to Duralose-UVTM nitrocellulose membranes. The membranes were baked for 2 h in vacuo at 80°C to fix the RNA. Blots were prehybridized for 30 min at 68°C in QuikHybTM solution and hybridized for 2 h at 68°C with random-primed 32P-labeled probes. After hybridization, the blots were washed twice for 15 min each at room temperature in 2 × SSC with 0.1% SDS followed by 2 washes for 30 min each at 60°C in 0.1 × SSC with 0.1% SDS. The washed blots were exposed to Hybond film (Amersham) for 12 to 48 h at −70°C using an intensifying screen. The signal intensities were quantified by scanning the autoradiograms with the Beckman R112 densitometer. All blots were hybridized with 32P-labeled probes of ICAM-1 cDNA (0.96-kb SalI to PstI fragment) and glyceraldehyde-3-phosphate dehydrogenase (1.1 kb PstI fragment). Glyceraldehyde-3-phosphate dehydrogenase was used as an internal control for RNA loading and normalized by densitometry of the ICAM-1 signal. For treatment of HUVEC with antisense and nonsense oligonucleotides, the cells were rinsed as described above with serum-free DMEM and incubated for 4 h in serum-free DMEM medium with addition of 5 μg/ml Lipofectin and an oligonucleotide at concentrations of 50 and 100 nM. After incubation, the medium was removed, fresh medium containing Lipofectin and the particular oligonucleotide was added, and the cells were treated for 1 h with 100 μM H2O2. RNA was isolated and processed for Northern analysis. Nuclei were isolated from HUVEC (3-5 × 107) according to the procedure described by 9Clayton D.F. Darnell Jr., J.E. Mol. Cell. Biol. 1983; 3: 1552-1561Crossref PubMed Scopus (225) Google Scholar. Cells were washed twice with cold PBS, harvested by scraping, and centrifuged for 10 min at 1,000 rpm. The cell pellet was washed once with ice-cold PBS and once with RSB buffer (10 mM Tris-HCl, pH 7.4, 10 mM NaCl, and 3.0 mM MgCl2). The cells were suspended in 10 ml of RSB and incubated on ice for 10 min. The cell pellets were collected by centrifugation, when the cells were sufficiently swollen as monitored by microscopy. Cell pellets were resuspended in 5 ml of RSB and homogenized with a dounce type B homogenizer. The homogenate was treated briefly with 0.1% Triton X-100 to remove cytoplasmic tags. The nuclear pellet was collected by centrifugation for 10 min at 1500 rpm at 4°C. The nuclear pellet was resuspended in 210 μl of freezing buffer (50 mM Tris-HCl, pH 8.0, 5 mM MgCl2, 0.5 mM dithiothreitol, and 40% glycerol), flash frozen in liquid nitrogen, and stored at −70°C. The nuclei were incubated for 30 min at 30°C in 0.3 ml of the assay mixture (25 mM Tris-HCl, pH 8.0, 1.25 mM concentration each of ATP, CTP, and GTP, 12.5 mM MgCl2, 325 mM KCl, and 250 μCi of [α-32P]UTP). RNase-free DNase (20 μl of 2 μg/ml) was added and incubated for an additional 15 min at 30°C. The run-on reaction was terminated by the addition of 36 μl of 10 × SET buffer (10% SDS, 100 mM Tris-HCl, pH 7.5, and 50 mM EDTA). Proteinase K (100 μg) was added and incubated for 45 min at 37°C, and the reaction mixture was extracted once with buffer-saturated phenol/chloroform (1:1) and once with chloroform/isoamyl alcohol (24:1). The aqueous phase was collected, ammonium sulfate (final concentration of 2.3 M) was added, and the RNA was precipitated with an equal volume of isopropyl alcohol. After 1 h at −70°C, the RNA was pelleted and washed twice with 75% ethanol. The pellet was dissolved in 100 μl of TE buffer (10 mM Tris-HCl, pH 7.4, and 1 mM EDTA) and passed through a Sephadex G-50 column to remove any unincorporated nucleotides. Filters were prepared for hybridization by application of denatured plasmids (5 μg/slot) using a slot blot apparatus. Plasmids containing cDNAs for ICAM-1, E-selectin, ribosomal RNA (rRNA), and glyceraldehyde-3-phosphate dehydrogenase were used for the experiments. Baked filters were hybridized with the RNA in the run-on assay as described for Northern analysis, and autoradiograms were developed. HUVEC were prepared for nuclear extracts as described by 38Shapiro D.J. Sharp P.A. Wahli W.W. Keller M.J. DNA (NY). 1988; 7: 47-55Crossref PubMed Scopus (478) Google Scholar. Briefly, the cells were treated with H2O2, TNFα (100 units/ml) or medium for 1 h prior to harvesting. The cells were washed twice with ice-cold PBS and collected by centrifugation (Sorvall RT6000) for 5 min at 2,000 rpm. The cell pellet was resuspended in 5 volumes of hypotonic buffer (10 mM HEPES, pH 7.9, 0.75 mM spermidine, 0.15 mM spermine, 0.1 mM EDTA, 0.1 mM EGTA, 1.0 mM DTT, 10 mM KCl, 0.5 mM PMSF) and incubated on ice for 10 min to allow the cells to swell. Cells were collected by centrifugation (Sorvall RT6000) for 7.5 min at 3,000 rpm in the cold. The cell pellet was resuspended in twice the original volume of ice-cold hypotonic buffer. Cells were homogenized with 30 strokes of a Wheaton dounce glass homogenizer (pestle B), followed by the addition of one-tenth volume of restore buffer (2.2 M sucrose, 10 mM HEPES pH 7.9, 0.75 mM spermidine, 0.15 mM spermine, 0.1 mM EDTA, 0.1 mM EGTA, 1.0 mM DTT, 10 mM KCl, and 0.5 mM PMSF). Nuclei were collected by centrifugation (Sorvall RC2-B) at 10,000 rpm in a Sorvall HB4 rotor in the cold for 3.5 min. The nuclei pellet was resuspended in 3 ml of nuclei lysis buffer (20 mM HEPES, pH 7.9, 0.42 M NaCl, 0.75 mM spermidine, 0.15 spermine, 0.2 mM EDTA, 0.2 mM EGTA, 2.0 mM DTT, 25% glycerol, and 0.5 mM PMSF). Nuclear debris was removed by ultracentrifugation (Beckman L8-M ultracentrifuge) at 40,000 rpm in a Beckman Ti80 fixed angle rotor for 90 min at 1°C, and 0.33 g of finely powdered ammonium sulfate was added to each milliliter of the collected supernatant and mixed gently by rocking in the cold for 60 to 90 min until the ammonium sulfate was completely dissolved. The precipitated nuclear protein was collected by ultracentrifugation (Beckman L8-M) at 30,000 rpm in a Beckman Ti80 fixed angle rotor for 20 min at 1°C. Nuclear protein pellets were resuspended in 200 μl of nuclear dialysis buffer (20 mM HEPES, pH 7.9, 20% glycerol, 100 mM KCl, 0.2 mM EDTA, 2.0 mM DTT, and 0.1 mM PMSF) and dialyzed twice for 90 min against 200 ml of NDB in the cold. Nuclear extracts were cleared of insoluble material by microcentrifugation for 10 min, and 30 μl of nuclear protein extracts were aliquoted and stored in a liquid nitrogen freezer until use. Protein concentrations were determined by the Bio-Rad assay kit. The electrophoretic mobility shift assay was performed as described by 33Roebuck K.A. Brenner D.A. Kagnoff M.F. J. Clin. Invest. 1993; 92: 1336-1348Crossref PubMed Scopus (44) Google Scholar. Nuclear extracts prepared from HUVEC by the method of 38Shapiro D.J. Sharp P.A. Wahli W.W. Keller M.J. DNA (NY). 1988; 7: 47-55Crossref PubMed Scopus (478) Google Scholar were incubated with 50,000 cpm (~0.1 to 0.5 ng) of various 32P-end-labeled double-stranded synthetic deoxyoligonucleotide probes for 30 min at 25°C in a 20-μl reaction volume containing 12% glycerol, 12 mM HEPES-NaOH (pH 7.9), 60 mM KCl, 5 mM MgCl2, 4 mM Tris-HCl (pH 7.9), 0.6 mM EDTA (pH 7.9), 0.6 mM DTT, and 1 μg of poly(dI)•(dC). DNA probes were end-labeled with [γ-32P]ATP (3,000 μCi/mmol) and T4 polynucleotide kinase. The labeled DNA probe was purified on push columns (Stratagene). Protein-DNA complexes were resolved in 5% native polyacrylamide gels pre-electrophoresed for 30-60 min at room temperature in 0.25 × TBE buffer (22.5 mM Tris borate and 0.5 mM EDTA, pH 8.3). Dried gels were exposed overnight to x-ray film with an intensifying screen at −70°C. Oligonucleotides used for the gel shift analysis were as follows: ICAM-1 AP-1/Ets, 5′-GCTGCTGCCTCAGTTTCCC-3′; ICAM-1 NF-κB, 5′-GCCCGGGGAGGATTCCTGGGCCCC-3′; ICAM-1 TRE, 5′-GACCGTGATTCAAGCTTA-3′; ICAM-1 AP-1 motif, 5′-TGGCCAGTGACTCGCAGCCCCAGC-3′; AP-1 m/Ets, 5′-GCTGCgtaagacGTTTCCCAGC-3′; AP-1/Ets-m, 5′-GCTGCTGCCTCAGTcagtCAGC-3′. Sequence motifs within the oligonucleotide are underlined, the mutations are in lowercase, and the relative positions of the sequence motifs are shown in Figure 7:, Figure 8:. The NF-κB oligonucleotide corresponds to the element upstream of the AP-3 site and downstream of the AP-1/Ets repeats. We have previously reported that exposure of human umbilical vein endothelial cells (HUVEC) to 100 μM H2O2 for 1 h resulted in maximal accumulation of steady-state ICAM-1 message, which could be detected as early as 30 min after oxidant exposure (22Lo S.K. Janakidevi K. Lai L. Malik A.B. Am. J. Physiol. 1993; 264: L406-L412Crossref PubMed Google Scholar). To determine whether the H2O2-induced increase in ICAM-1 mRNA was the result of increased de novo synthesis of the message or decreased rate of message degradation, we examined the effects of actinomycin D, a RNA synthesis inhibitor. We carried out two experiments: (i) actinomycin D was added to the cells at the same time as H2O2 exposure (Fig. 1) and (ii) cells were first pretreated with H2O2 for 1 h to maximize the expression of ICAM-1 message followed by treatment with actinomycin D to block new mRNA synthesis (Fig. 2). Treatment with actinomycin D at the time of H2O2 exposure (50 μM or 100 μM) abrogated ICAM-1 message induction (Fig. 1; compare lanes 2 and 3 with 4 and 5). To examine the effect of H2O2 on mRNA stability, endothelial cells were first exposed to 100 μM H2O2 for 1 h to maximize ICAM-1 expression, and this was followed by treatment with actinomycin D. Total RNA was isolated at 0.5, 1, 1.5, and 2 h after actinomycin D, and steady-state levels of ICAM-1 mRNA were analyzed by Northern blotting (Fig. 2). The H2O2-induced mRNA level returned to baseline level at 0.5 h (lane 3) and remained at this level up to 2 h (lanes 4-6). Both actinomycin D experiments indicated that H2O2 increased the synthesis of ICAM-1 mRNA. Fig. 3 compares the transcription rate of ICAM-1 with that of E-selectin, ribosomal RNA (rRNA), and glyceraldehyde-3-phosphate dehydrogenase as determined by nuclear run-on analysis. The transcription rates of E-selectin, rRNA, and glyceraldehyde-3-phosphate dehydrogenase were unaffected by H2O2 over the 2-h time course. In contrast, H2O2 increased the rate of ICAM-1 transcription at 1 h, and the rate remained high at 2 h, a finding that correlates with the H2O2 induction of ICAM-1 message (22Lo S.K. Janakidevi K. Lai L. Malik A.B. Am. J. Physiol. 1993; 264: L406-L412Crossref PubMed Google Scholar). These results indicate that H2O2 activates ICAM-1 gene transcription. We transfected a complementary ICAM-1 oligonucleotide that targets the 5′ end of the ICAM-1 mRNA (Fig. 4), to determine whether H2O2-induced ICAM-1 mRNA expression was sensitive to antisense deoxyoligonucleotides. The transfected HUVEC were exposed for 1 h with 100 μM H2O2 and analyzed by Northern blot for ICAM-1 mRNA expression. Lipofection with the antisense oligonucleotide produced a concentration-dependent reduction in the ICAM-1 message (Fig. 4, lanes 3 and 5). Neither control (nonsense oligonucleotide used as a negative control in lanes 4 and 6 or lipofection alone in lane 1) affected ICAM-1 mRNA expression. These data indicated that antisense oligonucleotides targeted to the ICAM-1 mRNA prevented the H2O2-induced ICAM-1 transcription. We used three agents to investigate possible mechanisms underlying the induction of ICAM-1. These agents were selected to study the DNA binding proteins AP-1 (Jun/Fos) and NF-κB, transcription factors known to be modulated by redox mechanisms (1Abate C. Patel L. Rauscher III, F.J. Curran T. Science. 1990; 249: 1157-1161Crossref PubMed Scopus (1367) Google Scholar; 25Meyer M. Schreck R. Baeuerle P.A. EMBO J. 1993; 12: 2005-2015Crossref PubMed Scopus (1257) Google Scholar). We used 3-aminobenzamide (3-AB), an inhibitor of poly(ADP-ribosyl)ation, to target AP-1 since 3-AB inhibited oxidant-induced c-fos expression and AP-1 binding activity (2Amstad P.A. Krupitza G. Cerutti P.A. Cancer Res. 1992; 52: 3952-3960PubMed Google Scholar). The anti-oxidant pyrrolidine dithiocarbamate (PDTC) was used to target NF-κB since PDTC inhibited oxidant-induced NF-κB activity without affecting AP-1 binding activity (35Schreck R. Meier B. Mannel D.N. Droge W. Baeuerle P.A. J. Exp. Med. 1992; 175: 1181-1194Crossref PubMed Scopus (1436) Google Scholar). N-Acetylcysteine (N-Cys(Ac)), a general antioxidant and precursor of glutathione (40Toledano M.M. Leonard W.J. Proc. Natl. Acad. Sci. U. S. A. 1991; 88: 4328-4332Crossref PubMed Scopus (566) Google Scholar; 1Abate C. Patel L. Rauscher III, F.J. Curran T. Science. 1990; 249: 1157-1161Crossref PubMed Scopus (1367) Google Scholar), was used to alter the redox state of cells. As shown in Fig. 5, 3-AB abrogated the induction of ICAM-1 message (lane 3), whereas PDTC had no effect (lane 4) suggesting a role for AP-1 in the induction of endothelial ICAM-1 transcription by H2O2. Pretreatment of endothelial cells for 1 h with N-Cys(Ac) also prevented the H2O2-induced mRNA expression (lane 5). We prepared nuclear protein extracts from endothelial cells treated with H2O2 for 1 h and examined DNA binding by electrophoretic mobility shift assay to study the effects of H2O2 on AP-1 and NF-κB binding activities (Fig. 6). H2O2 stimulated DNA binding activity on AP-1-like binding sites of the ICAM-1 promoter (lanes 1-9), but this was not the case with NF-κB binding activity (lanes 10-12) consistent with the inhibitor studies above. I