Introduction

Top Abstract Introduction Materials and Methods Results Discussion References

Expression of cell adhesion molecules such as vascular cell adhesion molecule-1 (VCAM-1), intercellular adhesion molecule-1 (ICAM-1), and E-selectin on endothelial cells represents one of the earliest pathological changes in immune and inflammatory diseases such as atherosclerosis (Springer, 1995). Induction of these molecules by tumor necrosis factor- (TNF-) and other inflammatory cytokines is regulated at the level of gene transcription and requires binding of the transcription factor NF-B to the regulatory region within the promoters of each of these genes (Collins et al., 1995). NF-B is a ubiquitous transcription factor that induces the expression of a variety of inflammatory and immune genes (Baeuerle and Henkel, 1994). In most cells, NF-B is retained in the cytoplasm in an inactive complex with the inhibitor IB. Inflammatory agents such as cytokines cause phosphorylation of IB followed by its degradation through the ubiquitin-proteasome pathway. NF-B is then dissociated from IB and translocated to the nucleus, where it binds to promoter recognition sites (Baeuerle and Henkel, 1994).

Oxidative signals play an important role in the regulation of inflammatory gene expression in endothelial cells (Marui et al., 1993; Chen and Medford, 1999). VCAM-1 and monocyte chemoattractant protein-1 (MCP-1) gene expression by diverse inflammatory signals occurs through an oxidation reduction (redox)-sensitive mechanism via NF-B (Marui et al., 1993; Chen and Medford, 1999). Inflammatory stimuli-activated endothelial expression of VCAM-1 and MCP-1 is inhibited by antioxidants such as pyrolidine dithiocarbamate (Marui et al., 1993; Chen and Medford, 1999). The NADPH oxidase inhibitor diphenylene iodonium suppresses TNF--induced superoxide (O) generation and VCAM-1 gene expression in endothelial cells (Tummala et al., 2000). In contrast, reactive oxygen species (ROS) such as H₂O₂ stimulate MCP-1 gene expression in vascular smooth muscle cells (Chen et al., 1998). However, the role of specific ROS, such as O and H₂O₂, in the regulation of the expression of inflammatory genes remains unclear.

Rac1 is a member of the Rho family of small GTPases involved in signal transduction pathways that control proliferation, adhesion, and migration of cells during embryonic development and invasiveness of tumor cells (Matos et al., 2000). In phagocytic cells, Rac proteins are involved in the assembly of the neutrophil NADPH oxidase system and responsible for transferring electrons from NADPH to molecular oxygen with the subsequent production of O (Abo et al., 1991). Many inflammatory signals, including cytokines and lipopolysaccharide, activate Rac1 (Sulciner et al., 1996a,b). Rac1 functions as a regulator of ROS generation in a variety of nonphagocytic cells, including endothelial cells (Sulciner et al., 1996a,b). Recently, it has been reported that Rac1 is involved in the activation of NF-B (Sulciner et al., 1996a). Expression of constitutively active Rac1 (V12Rac1) results in an increase in O generation in fibroblasts and activation of NF-B in HeLa cells. Conversely, expression of dominant negative Rac1 inhibits O generation by a variety of hormone stimuli, such TNF-, IL-1, and platelet-derived growth factor in fibroblasts. Inhibition of Rac1 also suppresses cytokine-stimulated NF-B activation (Sulciner et al., 1996a,b).

Because Rac1 regulates intracellular ROS generation and NF-B activation in response to cytokines, two important signaling events in the up-regulation of cell adhesion molecules, we tested the hypothesis that Rac1 may be involved in cytokine-induced expression of cell adhesion molecules in endothelial cells. In this study, we found that Rac1 is necessary for cytokine-induced expression of cell adhesion molecules. We further demonstrate that O, but not H₂O₂, is involved in the up-regulation of VCAM-1, ICAM-1, and E-selectin gene expression stimulated by TNF-.

	Materials and Methods

Top Abstract Introduction Materials and Methods Results Discussion References

Cell Culture and DNA Plasmids. Human aortic endothelial cells (HAECs) were obtained from Clonetics Corporation (San Diego, CA) and cultured in EGM-2 growth medium. Cells were used between passages 5 and 9. Human microvascular endothelial cells (HMECs) were described previously (Chen et al., 1997) and were cultured in modified MCDB 131 (Invitrogen, Carlsbad, CA) supplemented with 10% fetal bovine serum and EGM Singlequots (Clonetics Corporation). All cells were maintained at 37°C in a 5% CO₂ incubator. p288VCAM-Luc is a chimeric reporter construct containing coordinates 288 to 12 of the human VCAM-1 promoter, linked to a luciferase reporter gene. p(HIVB)₄-CAT contains four tandem copies of HIV long terminal repeat B DNA sequences linked to a chloramphenicol acetyl transferase (CAT) reporter gene (Kunsch et al., 1992). The empty expression vector pEXV and the myc epitope-tagged dominant negative (N17Rac1) expression vector were gifts from Dr. A. Hall (University College, London, UK) and have been described previously (Ridley et al., 1992).

Preparation of RNA and Northern Blot Analysis. Total cellular RNA was isolated by a single extraction with Tripure reagent (Roche Diagnostics, Indianapolis, IN) and size-fractionated using 1% agarose formaldehyde gels. RNA was transferred to nitrocellulose, and hybridizations were performed as described previously (Chen et al., 1997). The cDNAs used were human VCAM-1, ICAM-1, E-selectin, and GAPDH cDNA as described previously (Marui et al., 1993). Autoradiography was performed with a PhosphorImager 445sI (Amersham Biosciences, Inc., Sunnyvale, CA).

Determination of Cell Surface Expression of Adhesion Molecules by ELISA. HAECs were plated in 96-well plates and infected with recombinant adenovirus at indicated multiplicity of infection (MOI). Then, cells were incubated with TNF- (100 U/ml) for 16 h. Primary mouse antibodies for VCAM-1, E-selectin, and ICAM-1 were obtained from Southern Biotechnology Associates (Birmingham, AL). Cell surface expression of adhesion molecules was determined by primary binding with specific mouse antibodies, followed by secondary binding with a horseradish peroxidase-conjugated goat anti-mouse IgG antibody. Quantification was performed by determination of colorimetric conversion at OD450 nm of 3,3',5,5'-tetramethylbenzidine.

Transfection and Assay of Reporter Gene Activity. Because HAECs are relatively resistant to efficient transient transfection, we used HMECs for these experiments. HMECs were grown to 60 to 70% confluence in six-well plates and transfected with various plasmids as indicated in figure legends using SuperFect transfection reagent according to manufacturer's instructions (QIAGEN, Valencia, CA). After a 24-h recovery, HMECs were exposed to TNF- (100 U/ml) for 16 h. Then, protein extracts were prepared by rapid freeze-thaw in 0.25 M Tris, pH 8.0. For CAT activity assay, 5 µg of protein per sample was incubated with 5 µCi of ¹⁴C-labeled chloramphenicol (Amersham Biosciences, Inc.) and 5 µg of n-butyryl coenzyme A (Pharmacia, Peapack, NJ) for various times. The acetylated chloramphenicol forms were extracted using a 2:1 mixture of 2,6,10,14-tetra-methyl-pentadecane/xylenes and subjected to centrifuge (Kinston and Sheen, 1995). The organic phase was removed and counted to determine CAT activity. The pRL-TK (Renilla luciferase constitutively expressed under the control of thymidine kinase) was used to normalize transfection efficiency. All CAT activities were normalized against the Renilla luciferase activity. Firefly and Renilla luciferase activities were measured by using a luciferase reporter assay system according to the manufacturer's instructions (Promega, Madison, WI).

Nuclear Extract Preparation, ELISA, or Gel Shift Analysis of Nuclear NF-B Binding Activity. HAEC nuclear extracts were prepared as described previously (Tummala et al., 2000). Nuclear NF-B binding activity was determined by using TransAM NF-B p65 transcriptional factor assaying kit according to the manufacturer's instruction (Active Motif, Carlsbad, CA). This is an ELISA-based quantitative assay of NF-B binding activity using antibody directed to NF-B p65 subunit. Nuclear NF-B binding activity was also determined by gel shift analysis. The oligonucleotide containing the VCAM-1 B sequence is as follows: 5'-CTGCCCTGGGTTTCCCTTGAAGGGATTTCCCTCCGCCTCTGCAACA-3'. The sequences of the NF-B consensus binding sites are underlined. The DNA binding reaction was performed at 30°C for 15 min in a volume of 20 µl, containing 5 µg of nuclear extract, 225 µg/ml bovine serum albumin, 1.0 × 10⁵ cpm of ³²P-labeled probe, 0.1 µg/ml poly(dI-dC), and 15 µl of binding buffer (12 mM HEPES, pH 7.9, 4 mM Tris, 60 mM KCl, 1 mM EDTA, 12% glycerol, 1 mM dithiothreitol, and 1 mM phenylmethylsulfonyl fluoride). After the binding reaction, the samples were subjected to electrophoresis in 1× Tris-glycine buffer using 4% native polyacrylamide gels. Autoradiography was performed with a PhosphorImager 445sI (Amersham Biosciences, Inc.).

Adenoviruses. The adenovirus encoding myc-tagged cDNA of dominant negative N17Rac1 (Ad.N17Rac1), and cDNA of human Cu/Zn superoxide dismutase (Ad.SOD) and human catalase (Ad.Catalase) were generous gifts of Toren Finkel (National Institutes of Health, Bethesda, MD) and have been described previously (Sundaresan et al., 1995; Sulciner et al., 1996b; Moldovan et al., 1999). The viruses were amplified in human embryonic kidney 293 cells and purified on double cesium gradients. Infection was carried out with the indicated MOI for indicated times, after which the infection medium was aspirated and replaced with fresh medium. The Ad.LacZ, an adenovirus encoding the Escherichia coli LacZ gene, was used as a control for adenovirus infection. The ability of infected HAECs to express N17Rac1 were assessed by Western blot analysis via the myc-epitope tag using mouse monoclonal anti-myc antibody 9E10, which recognizes the myc peptide only on fusion proteins and not in endogenous myc proteins (Santa Cruz Biotechnology, Inc., Santa Cruz, CA).

Western Blot Analysis. HAECs were lysed for 30 min on ice in 1 ml of a lysis buffer containing 0.5% Nonidet P-40, 50 mM HEPES (pH 7.3), 150 mM NaCl, 2 mM EDTA, 1 µg/ml aprotinin, 1 µg/ml leupeptin, 1 µg/ml pepstatin, 1 mM phenylmethylsulfonyl fluoride, 0.1 mM sodium orthovanadate, and 1 mM NaF. Protein samples (15 µg) were subjected to electrophoresis on 10% or 15% SDS-polyacrylamide gel electrophoresis gels and transferred to a nitrocellulose membrane. Antibody-bound protein bands are then visualized via horseradish peroxidase-dependent chemiluminescence (Amersham Biosciences, Inc.). Anti-SOD and anti-catalase antibodies were purchased from Calbiochem (San Diego, CA). Anti-Rac1 antibodies were obtained from Upstate Biotechnology (Lake Placid, NY), and anti-IB- antibodies were obtained from Santa Cruz Biotechnology, Inc.

	Results

Top Abstract Introduction Materials and Methods Results Discussion References

The Dominant Negative Mutant N17Rac1 Inhibits TNF--Induced VCAM-1, E-selectin, and ICAM-1 mRNA Accumulation in HAECs. We first characterized the expression of N17Rac1 in Ad.N17Rac1-infected HAECs. As shown in Fig. 1, HAECs were infected with either Ad.LacZ (MOI of 100) or Ad.N17Rac1 (MOI of 25, 50, and 100) for 24 h. By Western blot analysis, myc-tagged N17Rac1 protein levels were increased in a dose-dependent manner after infection with Ad.N17Rac1. HAECs that were mock infected or infected with Ad.LacZ had no myc-N17Rac1 protein expression (Fig. 1A). Infection of HAECs with Ad.N17Rac1 (MOI of 100) also induced a time-dependent increase in the expression of myc-tagged N17Rac1. There is a progressive increase in the levels of myc-tagged N17Rac1 from 6 to 48 h in HAECs infected with Ad.N17Rac1.

View larger version (30K): [in this window] [in a new window]   Fig. 1.   Characterization of myc-tagged N17Rac1 expression after infection with Ad.N17Rac1. A, HAECs were infected with Ad.LacZ (MOI of 100) or Ad.N17Rac1 (MOI of 25, 50, and 100) for 24 h. B, HAECs were infected with Ad.N17Rac1 (MOI of 100) for 6, 24, and 48 h. Whole cell lysates were harvested and analyzed by immunoblotting with an antibody to myc.

View larger version (38K): [in this window] [in a new window]   Fig. 2.   Dominant negative N17Rac1 inhibits TNF--induced VCAM-1, E-selectin, and ICAM-1 mRNA accumulation. A, HAECs were infected with Ad.LacZ (MOI of 100; lanes 1 and 2) or Ad.N17Rac1 (MOI of 100; lanes 3 and 4) for 24 h and then exposed to TNF- (100 U/ml) for 4 h (lanes 2 and 4). B, untreated HAECs (lanes 1 and 2) or HAECs infected with Ad.LacZ (MOI of 100; lanes 3 and 4), Ad.N17Rac1 (MOI of 100; lanes 5 and 6) for 48 h were exposed to TNF- (100 U/ml) for 4 h (lanes 2, 4, and 6). C, HAECs were infected with Ad.LacZ (MOI of 100; lanes 1 and 2) or Ad.N17Rac1 (MOI of 25, 50, and 100; lanes 3, 4, 5, and 6) for 48 h and then exposed to TNF- (100 U/ml) for 4 h (lanes 2 and 4, 5 and 6). Total RNA was isolated and VCAM-1, E-selectin, and ICAM-1 mRNA levels were determined by Northern analysis. Two independent experiments showed similar results.

The Dominant Negative Ad.N17Rac1 Inhibits TNF--Induced Cell Surface Expression of Adhesion Molecules. To determine whether inhibition of endothelial cell adhesion molecule expression by Ad.N17Rac1 occurs at the cell surface level, we infected HAECs with Ad.LacZ or Ad.N17Rac1 for 24 h at indicated MOI, and treated HAECs with TNF- for 16 h. As shown in Fig. 3A, Ad.N17Rac1 at MOI of 100 and 50 inhibited TNF--induced VCAM-1 (top) and E-selectin (middle) cell surface expression in a dose-dependent manner. In contrast, a 24-h infection with Ad.N17Rac1 did not inhibit TNF--induced ICAM-1 protein expression (bottom). However, when HAECs were infected with Ad.N17Rac1 for 48 h, TNF--induced ICAM-1 protein expression was also suppressed (Fig. 3B, bottom). These data suggest that Rac1 is involved in TNF--induced VCAM-1, E-selectin, and ICAM-1 gene expression. These data also demonstrate that ICAM-1 gene up-regulation is relatively refractory to Rac1 inhibition compared with VCAM-1 and E-selectin.

View larger version (33K): [in this window] [in a new window]   Fig. 3.   Dominant negative N17Rac1 inhibits TNF--induced cell surface expression of VCAM-1, E-selectin, and ICAM-1. A, Untreated HAECs (lanes 1 and 2) and HAECs infected with Ad.LacZ (MOI of 100; lanes 1 and 2) or Ad.N17Rac1 (MOI of 25, 50, and 100) for 48 h were exposed to TNF- (100 U/ml) for 16 h. B, HAECs were infected with Ad.LacZ (MOI of 100) or Ad.N17Rac1 (MOI of 100) for 48 h and then exposed to TNF- (100 U/ml) for 16 h (lanes 2 and 4). Cell surface expression of adhesion molecules was determined by ELISA, as described under Materials and Methods. Values represent mean ± S.D., n = 4. *, P < 0.05 compared with TNF- alone-treated group.

The Dominant Negative Ad.N17Rac1 Does Not Inhibit TNF--Induced NF-B Translocation and IB Degradation. The genes for VCAM-1, ICAM-1, and E-selectin contain promoter elements with recognition sites for the transcription factor NF-B (Collins et al., 1995). To determine whether N17Rac1 inhibits VCAM-1, E-selectin, and ICAM-1 gene expression through inhibition of NF-B activation, we examined the effect of inhibition of Rac1 on TNF--induced NF-B nuclear translocation through two approaches: nuclear NF-B binding activity and IB degradation. HMECs were infected with Ad.N17Rac1 (MOI of 100) or Ad.LacZ (MOI of 100) for 48 h and treated for 1 h with TNF- (100 U/ml). Nuclear extracts were used to perform NF-B binding analysis using TransAM NF-B p65 transcriptional factor assaying kit. As expected, TNF- induced nuclear NF-B binding activity in HMECs (Fig. 4A, lane 2). Infection with Ad.LacZ had no effect on basal or TNF--induced nuclear NF-B binding activity (lanes 3 and 4). Infection with Ad.N17Rac1 did not suppress TNF--induced NF-B nuclear binding activity (Fig. 4A, lane 6). Similarly, infection with dominant negative N17Rac1 did not inhibit TNF--induced NF-B nuclear binding activity in HAECs by gel mobility shift analysis (data not shown). These data suggest that Rac1 is not involved in TNF- induced NF-B nuclear translocation.

View larger version (21K): [in this window] [in a new window]   Fig. 4.   Dominant negative N17Rac1 does not inhibit TNF--induced nuclear NF-B binding activity and IB- degradation. A, untreated HMECs (lanes 1 and 2) or HMECs infected with Ad.LacZ (MOI of 100; lanes 3 and 4), Ad.N17Rac1 (MOI of 100; lanes 5 and 6) for 48 h, were exposed to TNF- (100 U/ml) for 1 h (lanes 2, 4, and 6). Nuclear extracts were isolated and nuclear NF-B binding activity was determined by using TransAM NF-B p65 transcriptional factor assaying kit. Values are mean ± S.D., n = 3. *, P < 0.05 compared with control group. B, HAECs infected with Ad.LacZ (MOI of 100; lanes 1 and 2), Ad.N17Rac1 (MOI of 100; lanes 3 and 4) for 48 h were exposed to TNF- (100 U/ml) for 1 h (lanes 2 and 4). Whole cell lysates were analyzed by immunoblotting with antibodies to Rac1 or IB-.

The Dominant Negative N17Rac1 Suppresses TNF--Induced Transactivation of the NF-B-Driven HIV-B Promoter and 288 VCAM-1 Promoter. To explore whether the nuclear-translocated NF-B in the presence of N17Rac1 is able to transactivate NF-B-driven promoter, we transiently transfected HMECs with p(HIVB)₄-CAT, a construct consisting of four HIV B-binding sites (Kunsch et al., 1992). HMECs were treated with TNF- (100 U/ml) for 16 h. As expected, TNF- induced more than 8-fold increase in p(HIVB)₄-CAT promoter activity in the presence of empty vector (Fig. 5A). Expression of dominant negative N17Rac1 inhibited TNF--induced transactivation of p(HIVB)₄-CAT promoter activity (Fig. 5A). These data suggest that the dominant negative Rac1 suppresses TNF--induced transactivation of NF-B, but not nuclear translocation of NF-B in endothelial cells.

View larger version (16K): [in this window] [in a new window]   Fig. 5.   Dominant negative N17Rac1 inhibits TNF--induced transactivation of the NF-B-driven HIV(B)4 promoter and VCAM-1 promoter in HMECs. A, HMECs cultured in six-well plates were transfected with 1 µg of pHIV(B)4-CAT plus 1 µg of pEXV-N17Rac1, or empty vector pEXV. These cells were also transfected with 0.5 µg of pRL-TK for normalization of transfection efficiency. After a 24-h recovery, cells were exposed to TNF- for 16 h, cell extracts were harvested, and 5 µg of protein was used for CAT assay. B, HMECs cultured in six-well plates were transfected with 1 µg of p288VCAM-Luc plus 1 µg of pEXV-N17Rac1, or empty vector pEXV. These cells were also transfected with 0.1 µg of pRL-TK for normalization of transfection efficiency. After a 24-h recovery, cells were exposed to TNF- for 16 h, cell extracts were harvested and luciferase assays were performed. Values are mean ± S.D., n = 4. *, P < 0.05 compared with TNF- alone group.

Expression of SOD Inhibits TNF--Induced VCAM-1, E-selectin, and ICAM-1 mRNA and Cell Surface Protein Expression in HAECs. Activation of Rac1 by cytokine is associated with increased production of O in a variety of cells, including endothelial cells (Sulciner et al., 1996a,b). Oxidant signals play a role in the regulation of endothelial cell adhesion molecules such as VCAM-1 (Marui et al., 1993). We used an adenovirus expressing SOD to determine the role of O in the regulation of vascular adhesion molecule expression. SOD is responsible for converting O to H₂O₂. HAECs were infected with Ad.SOD (MOI of 100) or Ad.LacZ (MOI of 100) for 24 h. Then, HAECs were treated with TNF- (100 U/ml) for 4 h. By Western blot analysis, infection of HAECs with Ad.SOD increased intracellular Cu/Zn SOD protein levels (Fig. 6A). By Northern analysis, TNF--induced increases in VCAM-1, E-selectin, and ICAM-1 mRNA levels were inhibited by infection with Ad.SOD in HAECs (Fig. 6B).

View larger version (30K): [in this window] [in a new window]   Fig. 6.   SOD suppresses TNF--induced VCAM-1, E-selectin, and ICAM-1 mRNA accumulation. A, whole cell lysates from HAECs infected with Ad.SOD at MOI of 100 for 24 h were analyzed by immunoblotting with an antibody to SOD. B, HAECs were infected with Ad.LacZ (MOI of 100; lanes 1 and 2) or Ad.SOD (MOI of 100; lane 3 and 4) for 24 h and then exposed to TNF- (100 U/ml) for 4 h (lanes 2 and 4). Total RNA was isolated and VCAM-1, E-selectin, and ICAM-1 mRNA levels were determined by Northern analysis. Two independent experiments showed similar results.

Expression of Catalase Only Partially Inhibits TNF--Induced E-Selectin Expression, but Has No Effect on VCAM-1 and ICAM-1 Gene Expression in HAECs. O generated in cells can be converted spontaneously or enzymatically into H₂O₂. To determine the role of H₂O₂ in the regulation of vascular adhesion molecule expression, we used an adenovirus expressing catalase, which can convert H₂O₂ into H₂O. HAECs were infected with Ad.Catalase (MOI of 100) or Ad.LacZ (MOI of 100) for 24 h, followed by treatment with TNF- (100 U/ml) for 4 h. Infection of HAECs with Ad.Catalase for 24 h results in a 4-fold increase in intracellular catalase protein levels by Western blot analysis (Fig. 7A). By Northern blot analysis, infection with Ad.Catalase only partially inhibited TNF--induced E-selectin mRNA levels by approximately 50% (Fig. 7B, 1). Expression of catalase had no effect on TNF--induced VCAM-1 and ICAM-1 gene expression.

View larger version (31K): [in this window] [in a new window]   Fig. 7.   Catalase only partially inhibits TNF--induced E-selectin mRNA accumulation, but has no effect on VCAM-1 or ICAN-1 gene up-regulation. A, Whole cell lysates from HAECs infected with Ad.Catalase at MOI of 100 for 24 h were analyzed by immunoblotting with an antibody to catalase. B, HAECs were infected with Ad.LacZ (MOI of 100; lanes 1 and 2) or Ad.Catalase (MOI of 100; lanes 3 and 4) for 24 h and then exposed to TNF- (100 U/ml) for 4 h (lanes 2 and 4). Total RNA was isolated and VCAM-1, E-selectin, and ICAM-1 mRNA levels were determined by Northern analysis. Two independent experiments showed similar results.

View larger version (17K): [in this window] [in a new window]   Fig. 8.   Effects of Ad.SOD or Ad.Catalase on TNF--induced cell surface expression of VCAM-1, E-selectin, and ICAM-1. HAECs infected with Ad.LacZ (MOI of 100) or Ad.SOD (MOI of 100), or Ad.Catalase (MOI of 100) for 24 h were exposed to TNF- (100 U/ml) for 16 h. Cell surface expression of VCAM-1 (top), E-selectin (middle), or ICAM-1 (bottom) was determined by ELISA, as described under Materials and Methods. Values represent mean ± S.D., n = 4. *, P < 0.05 compared with TNF--treated cells infected with Ad.LacZ.

	Discussion

Top Abstract Introduction Materials and Methods Results Discussion References

Regulation of endothelial cell adhesion molecule expression, particularly VCAM-1, is mediated by redox-coupled signaling mechanisms involved in the activation of the transcription factor NF-B (Marui et al., 1993). However, the signaling pathway and the specific reactive oxygen species that regulate the expression of these proteins are not well understood. In the present study, we demonstrated that dominant negative N17Rac1 suppresses TNF--induced VCAM-1 and E-selectin gene expression, and to a lesser degree ICAM-1 gene expression. We further demonstrated that dominant negative N17Rac1 inhibits TNF--induced activation of NF-B-dependent transcriptional activity but has no effect on TNF--induced NF-B nuclear translocation. Furthermore, inhibition of generation of O, a downstream signal of Rac1, by SOD suppresses TNF--induced VCAM-1, E-selectin, and ICAM-1 gene expression. Our results suggest that Rac1 and O play critical roles in TNF--mediated expression of endothelial cell adhesion molecules.

Rac proteins are involved in the assembly of NADPH oxidase and O generation (Bokoch, 1994). NADPH oxidase is the major source of O in endothelial cells (Mohazzab et al., 1994) and plays an essential role in TNF--induced O generation (Li et al., 2002). Rac1 is involved in the activation of NADPH oxidase in endothelial cells (Abid et al., 2001) and is required for O generation in response to inflammatory stimuli in a variety of cell types, including endothelial cells (Yeh et al., 1999; Deshpande et al., 2000; Ozaki et al., 2000). Using an NADPH oxidase inhibitor, diphenylene iodonium, we previously reported that flavin binding proteins such as NADPH oxidase are required for TNF--induced O generation and the activation of VCAM-1 and ICAM-1 gene expression (Tummala et al., 2000). Consistent with our early results, our finding that SOD, but not catalase, inhibits TNF--induced endothelial cell adhesion molecule expression provides the first direct evidence that O, but not H₂O₂, is involved in the redox-sensitive regulation of inflammatory gene expression. A mechanism by which O may mediate endothelial cell adhesion molecule expression may occur through its downstream reactive species peroxynitrite, generated from interaction of O with nitric oxide (Tarpey and Fridovich, 2001). Recent studies indicate that peroxynitrite may function as an intracellular signal for the production of IL-8 (Zouki et al., 2001). Exogenous peroxynitrite has been shown to stimulate NF-B activation in endothelial cells (Cooke and Davidge, 2002) and monocytes (Matata and Galinanes, 2002). Peroxynitrite also activates expression of inducible nitric-oxide synthase and IL-6 (Cooke and Davidge, 2002; Matata and Galinanes, 2002).

In the present study, we found that the inhibition of Rac1 results in differential inhibitory effects on the induction of VCAM-1, E-selectin, and ICAM-1 by TNF-. Among three adhesion molecules studied, VCAM-1 and E-selectin were more profoundly inhibited by dominant negative N17Rac1 compared with ICAM-1. We showed that ICAM-1 is more resistant to the Rac1 inhibition and can only be suppressed by a higher Ad.N17Rac1 dose. Similar differential inhibitory effects have been observed with antioxidant treatment. We demonstrated previously that the treatment of endothelial cells with thiol antioxidants such as pyrolidine dithiocarbamate or N-acetylcysteine selectively inhibit TNF--induced VCAM-1 and to a lesser degree E-selectin gene expression, but have no effects on ICAM-1 gene expression (Marui et al., 1993). Relative differences in the inhibitory effects of N17Rac1 may be explained by the fact that different combinations of transcriptional factors are required for the activation of E-selectin, VCAM-1, and ICAM-1 promoters (Collins et al., 1995). Although all these endothelial cell adhesion genes have NF-B binding sites, other transcription factors have also been shown to contribute to the regulation of expression of these genes. In the case of ICAM-1, Ets-, signal transducer and activator of transcription-, and interferon--response element-dependent transcriptional mechanisms are required for ICAM-1 gene up-regulation (Duff et al., 1997; Audette et al., 2001; Roy et al., 2001).

The present study shows that SOD is more efficient in inhibition of ICAM-1 gene expression than dominant negative N17Rac1 (Fig. 6 versus Fig. 2). These data suggest that in addition to Rac1-mediated O generation, other cellular sources such as mitochondria (Thannickal and Fanburg, 2000) may also generate O and contribute to ICAM-1 gene up-regulation. Inhibition of Rac1 can only suppress NADPH oxidase-mediated O generation but has no effects on mitochondria-mediated O generation. In contrast, SOD can scavenge O generated from all cellular sources and may therefore be more efficient in suppressing ICAM-1 gene expression.

Two separate signaling events are involved in the NF-B activation pathway. First, NF-B dimers, kept in the cytoplasm through interaction with inhibitory proteins IB, become activated by phosphorylation and degradation of IB, resulting in the subsequent release and nuclear translocation of NF-B. This pathway depends on the IB kinase, which is essential for inducible IB phosphorylation and degradation (Zandi et al., 1997). The second pathway regulates the transactivating potential of the p65 subunit of NF-B once it is bound to its consensus sequence (Bergmann et al., 1998; Jefferies and O'Neill, 2000). Several studies demonstrated that upon stimulation with either TNF- or IL-1, the p65 subunit of NF-B becomes phosphorylated on multiple serine sites, thus potentially acting to enhance NF-B p65 transactivating potential (Wang and Baldwin, 1998; Wang et al., 2000). The kinase directly responsible for phosphorylation of NF-B p65 has yet to be identified, although casein kinase II has been shown to phosphorylate transactivation region found in the COOH-terminal domain of p65 (Wang et al., 2000). Calmodulin-dependent protein kinase IV has been reported to stimulate NF-B transactivation via phosphorylation of the p65 subunit (Jang et al., 2001).

Rac1 plays an important role in the regulation of NF-B activation evoked by environmental stresses and proinflammatory cytokines (Sulciner et al., 1996a). In the present study, we demonstrated that dominant negative N17Rac1 inhibits TNF--induced transactivation of NF-B-driven promoter in HMECs but has no effect on TNF--induced nuclear NF-B translocation in both HAECs and HMECs. These data are consistent with other studies. Jefferies and coworkers reported that inhibition of Rac1 suppressed IL-1-driven p65-mediated NF-B transactivation but had no effect on IL-1-induced nuclear NF-B translocation or IB degradation in a thymoma cell line and in porcine aortic endothelial cells (Jefferies and O'Neill, 2000). Inhibition of NF-B transactivation by the suppression of phosphatidylinositol 3-kinase (PI3K) has also been reported. Sizemore and coworkers reported that inhibition of PI3K suppressed NF-B-dependent gene expression but had no effect on the IL-1-stimulated degradation IB-, or NF-B nuclear translocation and DNA binding. In contrast, PI3K inhibitors blocked the IL-1-stimulated phosphorylation and transactivation of NF-B p65 (Sizemore et al., 1999). However, other studies have found that Rac1 mediates NF-B activation through regulation of NF-B nuclear translocation in HeLa cells (Sulciner et al., 1996a). Sanlioglu et al. (2001) reported that dominant negative N17Rac1 inhibits lipopolysaccharide-induced nuclear NF-B binding activity. Similarly, inhibition of Rac1 suppressed TNF--induced nuclear NF-B binding activity in human umbilical vein endothelial cells (Deshpande et al., 2000). These conflicting observations on the mechanism of Rac1 in NF-B activation illustrate the complex systems of interaction between Rac1 and NF-B pathways as well as cell type differences in this interaction. Nevertheless, we provide evidence that Rac1 is not involved in TNF--induced IB- degradation and NF-B nuclear translocation. Instead, Rac1 plays a role in the modulation of the ability of the NF-B to transactivate gene expression in endothelial cells.

Our findings suggest that the activation of Rac1-dependent pathways and O generation are required for the up-regulation of endothelial adhesion molecules stimulated by TNF-. Cumulatively, our results may provide the molecular link between Rac1, NADPH oxidase, and O in the signaling pathway that mediates TNF--induced expression of endothelial cell adhesion molecules, suggesting that Rac1-dependent signaling pathways may serve as important pharmacological targets for the treatment of inflammatory diseases.