INTRODUCTION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

ACTIVATION of the transcription factor nuclear factor-B (NF-B) and associated gene induction has been suggested to be a major regulator of inflammatory and proliferative vascular responses (1, 13). In vitro studies showed the presence of members of the Rel family of proteins in human vascular smooth muscle cells (SMC) (10) and demonstrated activation of this transcription factor in these cells by proinflammatory cytokines and platelet-derived growth factor (PDGF) (23). Moreover, in situ studies revealed the presence of activated NF-B in SMC as well as macrophages and endothelial cells of atherosclerotic lesions (12).

In resting cells the NF-B dimer is held in an inactive state in the cytoplasm by an inhibitory protein of the I-B family (2, 7). Studies showed that stimulating the cell causes I-B to become phosphorylated, followed by polyubiquitination and rapid proteolytic degradation (16). These events release the NF-B dimer, unmasking the nuclear translocation and DNA binding domains and allowing the transcription factor to move to the nucleus, bind to the promoter of target genes, and induce transcription (6, 7, 16).

In most cell types NF-B can be activated by a diverse range of stimuli (3-5), suggesting that several different signaling pathways are capable of triggering the activation of this transcription factor. Candidates for this convergence of signals are the activation of the recently identified I-B kinase (IKK) and the upstream NF-B-inducing kinase (NIK). The sequence of events and second messengers resulting in the activation of IKK and NIK are, at present, poorly understood. Several studies suggested that oxidant stress might play a role in the signaling events leading to the induction of NF-B (27). This hypothesis is supported by the observation that H₂O₂ is capable of activating NF-B in some cell types (28). Furthermore, chemically unrelated antioxidants have been reported to be effective inhibitors of NF-B activation induced by a variety of stimuli (24, 25, 28, 30-33, 36, 37). On the basis of these observations, it was proposed that reactive oxygen intermediates (ROI) are important and widely used second messengers in NF-B activation (26). However, more recent studies questioned the hypothesis that oxidative stress brought on by ROI would be the universal second messenger for NF-B activation and suggested that this mechanism may be restricted to certain cell types (9, 14).

The aim of this study was to clarify these issues in human aortic SMC by determining whether interleukin-1 (IL-1)-induced activation of NF-B proceeds via the production of ROI and to explore the role of redox regulation of this transcription factor in these cells.

	MATERIALS AND METHODS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

Cell culture. SMC were prepared from human donor aorta by enzymatic digestion. Briefly, aortas from heart and heart/lung transplant donors were cleaned of endothelium and adventitia and the resulting tissue was finely minced and agitated with 0.5 mg/ml elastase and 1 mg/ml collagenase 1A at 37°C for 4 h. The samples were then strained through a 100-µm cell strainer and centrifuged at 1,500 rpm for 10 min. The supernatant was discarded, and the pelleted cells were resuspended in growing medium (DMEM containing 100 mg/ml penicillin, 100 U/ml streptomycin, 2 mM L-glutamine, and 10% heat-inactivated FCS) and transferred to a 75-cm² tissue culture flask. Cells were maintained in growing medium at 37°C in a humidified atmosphere of 5% CO₂ and, when confluent, were seeded into 24-well plates or 10-cm dishes for various experimental protocols. Cells were characterized as SMC by positive immunostaining for smooth muscle specific -actin and typical "hill and valley" morphology. Cells were used between passages 3 and 10, and experiments were repeated in cell isolates of at least four different donors.

Experimental design and procedure. In the initial phase of the study, we established the presence of I-B and NF-B proteins in resting cells. Activation of NF-B was monitored by four different methods. Phosphorylation and changes in cellular levels of I-B were assessed by Western blotting of total cell lysates. Association and dissociation of NF-B from I-B were studied by immunoprecipitating NF-B p65 from total cell lysates and by determining I-B levels coprecipitating with NF-B p65. Nuclear translocation and accumulation of NF-B p65 were investigated in nuclear extracts by Western blotting, and the DNA binding capacity of NF-B was established by electrophoretic mobility shift assay (EMSA). After initial dose-response characterization of IL-1-induced activation of NF-B, a submaximal concentration of IL-1 was used to perform time-course experiments. The potential role of oxidant stress in IL-1-induced activation of NF-B was investigated by a variety of experiments. First, we explored the influence of IL-1 on cellular redox status by measuring the fluorescence of a redox-sensitive dye, 2',7'-dichlorodihydrofluorescin (DCFH) and by monitoring intracellular sulfhydryls, a moiety that is a primary target of ROI. Second, we explored the influence of H₂O₂ on intracellular oxidant stress and the NF-B activation pathway. Finally, we established the effects of the widely used antioxidant, N-acetylcysteine (NAC), a cell-permeant sulfhydryl compound and glutathione precursor, and the metal chelators pyrrolidine dithiocarbamate (PDTC) and diethyldithiocarbamate (DETC) on IL-1-induced NF-B activation.

Total cell lysis. Cells were washed in cold PBS, 30 µl of 1% SDS lysis buffer containing 0.5 mM phenylmethylsulfonyl fluoride (PMSF), 2.5 mg/ml aprotinin, 2.5 µg/ml pepstatin A, and 2.5 µg/ml leupeptin were added to each well, and the cells were disrupted with the rubber plunger from a 1-ml syringe. Lysates from triplicate wells were combined and aspirated 10 times through a 25-gauge needle to shear DNA. The samples were assayed for protein content using the Bradford assay (11) and stored at 20°C in aliquots containing 25 µg of protein in 40 µl of 1% SDS lysis buffer.

Detection of NF-B p65-associated I-B. The whole procedure was carried out at 4°C. Cells were washed in cold PBS and incubated on ice with 50 ml/well of RIPA buffer (50 mM Tris · HCl, 150 mM NaCl, 1% Nonidet-P40, 0.5% sodium deoxycholate, 0.5 mM PMSF, 2.5 µg/ml aprotinin, 2.5 µg/ml pepstatin A, 2.5 µg/ml leupeptin, 50 mM NaF, 1 µM Na₃VO₄, 20 mM -glycerol phosphate, 10 mM sodium molybdate) for 10 min. Lysates were prepared as described in Total cell lysis, 25 µl from each sample were removed, and the proteins were analyzed by SDS-PAGE as total cell lysis preparations. The remaining samples were centrifuged at 3,000 rpm at 4°C for 15 min, and 0.5 µg of anti-NF-B p65 polyclonal antibody was added to the supernatant. After 1-h incubation at 4°C, 10 µl of protein G-agarose were added, and the samples were incubated for a further 60 min at 4°C on a rocker. The samples were then centrifuged at 2,500 rpm for 5 min at 4°C, supernatants were discarded, and the pellet was washed four times in cold RIPA buffer. The final pellet was resuspended in 150 µl of Laemmli buffer and subjected to SDS-PAGE and immune blotting with antibodies against I-B and NF-B p65.

Preparation of nuclear extracts. Cells were scraped in ice-cold PBS, centrifuged, and lysed in buffer A [10 mM HEPES (pH 7.9), 1.5 mM MgCl₂, 10 mM KCl, 0.1% Nonidet-P40, 0.5 mM dithiothreitol (DTT), 0.5 mM PMSF, 2.5 µg/ml aprotinin, 2.5 µg/ml pepstatin A, 2.5 µg/ml leupeptin]. The nuclei were pelleted by centrifugation, and the supernatant was removed, assayed for protein content, and stored as cytosolic proteins. The nuclear pellet was washed in buffer A and resuspended in 50 ml of buffer B [20 mM HEPES (pH 7.9), 25% glycerol, 1.5 mM MgCl₂, 420 mM NaCl, 0.2 mM EDTA, 0.5 mM DTT, 0.5 mM PMSF, 2.5 µg/ml aprotinin, 2.5 µg/ml pepstatin A, 2.5 µg/ml leupeptin] and incubated at 4°C for 30 min on a rocker. The nuclear debris was removed by centrifugation, and the resulting nuclear extract was assayed for protein content. The nuclear proteins were either stored in aliquots containing 10 µg of protein for Western blot analysis or diluted in buffer C [20 µM HEPES (pH 7.9), 50 mM KCl, 0.2 mM EDTA, 0.5 mM DTT, 0.5 mM PMSF] for DNA binding analysis.

Western blotting. All protein samples were resolved on 10% SDS polyacrylamide gels and transferred to Hybond C Super nitrocellulose membrane. After blocking nonspecific binding with 3% nonfat dry milk solution in PBS containing 0.05% Tween 20 (PBS-T), we incubated the membranes for 60 min at room temperature with primary antibody (rabbit polyclonal anti-I-B; 1:250), rabbit polyclonal anti-NF-B p65 (1:200), and mouse monoclonal anti-smooth muscle specific -actin (clone 1A4, 1:2,000; Sigma Chemical, Poole, UK) in blocking solution. Each membrane was then washed with PBS-T for 30 min and incubated for 30 min with the relevant horseradish peroxidase-conjugated secondary antibody. The membranes were washed again in PBS-T, and immunoreactive protein bands were visualized using an enhanced chemiluminescence (ECL) detection system. Where necessary, the antibodies were stripped from the membrane with stripping solution (62.5 mM Tris · HCl pH 6.7, 2% SDS, 6 µl/ml 2-mercaptoethanol) and probed as before. Films were scanned using a molecular dynamics laser densitometer and analyzed by the one-dimensional software package (21).

Electrophoretic mobility shift assay. The DNA binding capacity of nuclear proteins was analyzed by EMSA. Double-stranded oligonucleotide containing the NF-B binding element (5'-AGTTGCAGGC, 3'-GCCCTCAACT; Pharmacia Biotech) was end-labeled using 0.74 MBq of [-³²P]ATP and T4 polynucleotide kinase (New England Biolabs) and purified through Probequant G-50 microcolumns (Pharmacia Biotech). Four micrograms of nuclear proteins were incubated with two microliters (0.0175 pmol) of ³²P-labeled oligonucleotide, two microliters of 10× buffer [100 mM Tris (pH 7.5), 200 mM NaCl, 10 mM EDTA, 50% glycerol], one microliter (2.5 µg) of poly(dI-dC), and one microliter of nonspecific oligonucleotide for 1 h. The samples were separated on a nondenaturing polyacrylamide gel and exposed to X-ray film. Specificity was determined by the addition of 100-fold excess cold oligonucleotide, 100-fold excess of cold mutant B oligonucleotide [5'-AGTTGAGGCGACTTTCCCAGGC, 3'-GCCTGGGAAAGTCGCCTCAACT (single base pair substitution in bold)], 4 µg of purified I-B protein corresponding to amino acids 1-317, or 2 µg of antibody directed against specific NF-B protein subunits.

Fluorescent measurement of intracellular oxidation. This method is based on the internalization of the nonfluorescent compound 2',7'-dichlorodihydrofluorescin diacetate (DCFH-DA) by the cells, cleavage of the diacetate group by intracellular esterases resulting in the formation of 2',7'-dichlorodihydrofluorescin (DCFH), and the oxidation of DCFH on oxidative stress generating the highly fluorescent compound dichlorofluorescein (DCF). Human aortic SMC were grown in DMEM without phenol red, containing 100 µg/ml penicillin, 100 U/ml streptomycin, 2 mM L-glutamine, and 10% heat-inactivated FCS, for 1 wk before use. When confluent, the cells were trypsinized, washed in PBS, and resuspended in DMEM without phenol red, and the cell density was adjusted to 1 × 10⁶ cells/ml. One hundred microliters of cells were removed and incubated with one microliter of ethanol, as vehicle, for 20 min at 37°C in the dark and used as unloaded control cells. Ten microliters per milliliter of 20 mM DCFH-DA in ethanol were added to the remainder of the cells. The cells were incubated at 37°C for 20 min in the dark to load the cells with the dye. Cells were then kept on ice in the dark and used within 90 min. Thirty thousand cells were made up to 1 ml with DMEM without phenol red, protected from light at all times, and incubated with the appropriate treatments at 37°C as indicated. Fluorescence was detected by flow cytometric analysis using a Coulter XL measuring log fluorescence intensity at 570 nm of at least 3,000 cells over 5 min.

Determination of intracellular sulfhydryl concentration. Intracellular acid soluble thiols were measured according to the method of Sedlak and Lindsay (29). Briefly, cells were washed in cold PBS and incubated with ice-cold 5% trichloroacetic acid for 30 min and the supernatants from duplicate wells were combined. Two hundred microliters of a 0.4 M Tris-0.02 M EDTA solution (pH 8.9) and ten microliters of 10 mM Ellman's reagent [5,5'-dithiobis(2-nitrobenzoic acid)] in methanol were added to one hundred microliters of each sample or reduced glutathione standards (0-100 µM), and the absorbance was measured at 414 nm. The protein precipitate remaining in the wells after sulfhydryl extraction was washed three times with PBS and solubilized in 50 µl of 1% SDS, and the amount of protein in each sample was determined by Bradford assay. All results are expressed as nanomoles of sulfhydryl per milligram of protein.

Materials and reagents. DMEM, FCS, L-glutamine, penicillin-streptomycin solution, trypsin-EDTA solution, EBSS, PBS (10×), NAC, PDTC, DETC, 30% vol/vol H₂O₂, DTT, PMSF, aprotinin, pepstatin A, and DCFH-DA were all obtained from Sigma Chemical. Tween 20 and 2-mercaptoethanol were from BDH. All primary antibodies unless otherwise stated were from Santa Cruz. Horseradish peroxidase-conjugated secondary antibodies were from Dako (High Wycombe, UK). Hybond C Super nitrocellulose membrane, ECL detection kit, hyperfilm, poly(dI-dC), and [-³²P]ATP were from Amersham Pharmacia Biotech (Amersham, UK). IL-1, collagenase 1A, elastase, leupeptin, and protein G agarose were all obtained from Boehringer Mannheim (Lewis, UK). All tissue culture plastics (Falcon) were from Marathon Laboratory Supplies (London, UK). Nonidet-P40 was from Millipore (Watford, UK).

Data analysis. Data are presented as either representative figures or as means ± SE of at least four independent experiments. Flow cytometry data are expressed as median fluorescence intensity. Statistical analysis was performed using one-way analysis of variance and t-test where appropriate. All images are representative of at least four independent experiments.

	RESULTS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

IL-1-induced activation of NF-B in human aortic SMC. Western blot analysis of total protein samples from 1% SDS-lysed SMC revealed a strong I-B immunoreactive band demonstrating the presence of I-B in resting cells (Fig.1A). Exposure of the cells to IL-1 (50 U/ml) for 5 min resulted in the appearance of a slower migrating I-B band (Fig. 1A). Longer exposure (30 min) caused a significant decrease in the amount of I-B in the cells (22 ± 5% of that in resting cells compared with 117 ± 7% after 5-min IL-1 treatment, P < 0.05, n = 10). Depletion of I-B appeared to be transient, because 60-min exposure to IL-1 caused I-B to reappear in the cell (Fig. 1B), and after 90-min IL-1 treatment I-B levels were significantly higher than those after 30 min of IL-1. This effect was inhibited by cotreatment with the protein synthesis inhibitor cycloheximide (10 µM), indicating synthesis of new protein (Fig. 1B). Immunoprecipitating NF-B p65 from resting cells resulted in the coprecipitation of I-B, as demonstrated by the presence of an I-B immunoreactive protein of ~37 kDa (Fig. 1C), indicating that NF-B p65 is bound to the inhibitor protein I-B. Longer exposure (30 min) caused a decrease in the amount of I-B coprecipitating with NF-B (9.7 ± 3.8% of that in resting cells, n = 5; Fig. 1C). This event appears to coincide with the time course of I-B depletion from the cells (Fig. 1A). Separation of cytosolic and nuclear proteins revealed the presence of small amounts of NF-B p65 subunit in the nuclei of resting cells (Fig. 2A, bottom). Associated with I-B depletion from the cytoplasm (Fig. 2A, top), after 30-min treatment with IL-1 (50 U/ml), nuclear levels of p65 increased fourfold (399 ± 54% of that in resting cells, n = 9) and remained elevated for a further 90 min of IL-1 treatment (Fig. 2A, bottom). IL-1 treatment also increased the binding of nuclear proteins to the B consensus DNA as demonstrated by EMSA using ³²P-labeled double-stranded oligonucleotide containing a specific B consensus sequence (Fig. 2B). Specificity of the reaction is supported by experiments in which 100-fold excess cold competitive oligonucleotide caused the disappearance of all protein-DNA complexes whereas addition of unlabeled mutant B oligonucleotide to the binding reaction resulted in the disappearance of nonspecific binding activity (Fig. 2C, lanes 5 and 6, respectively). Furthermore, inclusion of a purified fragment of I-B protein (amino acids 1-317) in the reaction mixture inhibited the specific NF-B DNA binding activity (Fig. 2C, lane 7). The molecular identity of the DNA binding complex was analyzed by supershift assay using specific antibodies. The increased binding capacity of nuclear proteins to specific B consensus DNA after treatment with IL-1 was supershifted by both the p50 and p65 antibodies (Fig. 2C, lanes 3 and 4, respectively).

View larger version (33K): [in this window] [in a new window]   Fig. 1.   Time course of changes in cellular levels of inhibitory protein I-B and I-B binding to nuclear factor-B (NF-B) p65 in human aortic smooth muscle cells (SMC) after interleukin-1 (IL-1, 50 U/ml) stimulation. A: Western blot analysis of total cell I-B levels from human aortic SMC. Slower-migrating protein, indicated as I-Bp, corresponds to phosphorylated form of I-B. Graph depicts optical density analysis of total cell I-B (* P < 0.05 from 5-min IL-1 treatment, n = 10 preparations). B: Western blot analysis of total cell I-B over 90-min IL-1 (50 U/ml) treatment, in absence (top) and presence (bottom) of protein synthesis inhibitor cycloheximide (10 µM). C: Western blot analysis of I-B coprecipitated with NF-B p65 from human aortic SMC. Graph depicts optical density analysis of I-B immunoreactive proteins from NF-B p65 immunoprecipitations (n = 6).

View larger version (43K): [in this window] [in a new window]   Fig. 2.   Time-dependent activation of NF-B by IL-1 (50 U/ml). A: cytosolic I-B levels and nuclear NF-B p65 content determined by Western blot analysis. B: electrophoretic mobility shift assay (EMSA) demonstrating increased binding activity of nuclear proteins to specific B consensus DNA in response to IL-1. C: analysis of specificity and molecular identity of DNA-protein complexes induced by IL-1 treatment. Lane 1, nuclear proteins from untreated human aortic SMC; lanes 2-7, nuclear proteins from cells treated with IL-1 (50 U/ml) for 30 min; lane 2, IL-1 treatment alone; lane 3, addition of anti-p50 antibody in binding reaction, lane 4, addition of p65 antibody in binding reaction; lane 5, addition of 100-fold excess cold B oligonucleotide; lane 6, 100-fold excess cold mutant oligonucleotide; lane 7 contains purified I-B protein corresponding to amino acid sequence 1-317.

Influence of IL-1 and H₂O₂ on intracellular oxidant stress using flow cytometry. Figure 3, A and B, demonstrates that treatment of DCFH-loaded human aortic SMC with the oxidant H₂O₂ (0.1 mM) for 5 min caused an increase in the fluorescence signal from 11.2 ± 2.3 arbitrary units (AU) in control cells to 58.3 ± 8.5 AU in H₂O₂-treated cells (P < 0.05, n = 4), indicating increased intracellular oxidant stress. Incubating the cells for longer periods of time caused the fluorescence intensity to increase further, from 34.3 ± 8.5 AU in control cells to 157.98 ± 36.8 AU after 30-min H₂O₂ treatment (P < 0.05, n = 4). Higher concentrations of H₂O₂ (10 mM) caused an even greater increase in fluorescence (265.9 ± 60.5 AU) after 30 min, which was significantly reduced by pretreatment with NAC (94.2 ± 27.7 AU, P < 0.05, n = 4). In contrast, the intensity of the fluorescence signal generated by IL-1 treatment did not differ from that in control treated cells (14.3 ± 2.6 AU after 5-min IL-1 treatment compared with 11.2 ± 2.3 AU in time-matched control cells, n = 4; Fig. 3, A and C). Thirty- and sixty-minute IL-1 treatment also failed to generate a fluorescent signal above that of the time-matched control cells, indicating the absence of a significant oxidant stress in IL-1-treated cells.

View larger version (40K): [in this window] [in a new window]   Fig. 3.   Influence of IL-1 and H2O2 on intracellular oxidant stress using flow cytometry. A: representative histograms showing fluorescence signal generated by treating human aortic SMC loaded with redox-sensitive dye 2',7'-dichlorodihydrofluorescein (DCFH) with H2O2 (0.1 mM) or IL-1 (50 U/ml) for 5, 30, and 60 min. Each diagram shows a histogram from treated cells and a histogram from time-matched control cells. Graphs summarize changes in fluorescence intensity of DCFH-loaded cells treated with H2O2 (0.1 mM, B) and IL-1 (50 U/ml, C) after 5, 30, and 60 min. * P < 0.05 from time-matched control (n = 4 preparations).

Effect of IL-1, H₂O₂, and antioxidants on level of intracellular sulfhydryls. The concentration of intracellular acid-soluble thiols in unstimulated control human aortic SMC was 14.2 ± 0.7 nmol/mg protein and remained unaltered (16.5 ± 1.7 nmol/mg protein, n = 7; Fig. 4) in the presence of IL-1 (50 U/ml) for 30 min, a condition that fully activated NF-B. Incubating the cells with 10 mM H₂O₂ for 30 min significantly reduced intracellular sulfhydryl concentration to 6.9 ± 1.0 nmol/mg protein (n = 7, P < 0.05). NAC, the cell-permeant antioxidant and glutathione precursor, increased intracellular sulfhydryls to 49.5 ± 6.3 nmol/mg protein (n = 7, P < 0.05; Fig. 4). NAC also prevented the H₂O₂-induced depletion of these low-molecular-weight sulfhydryls (n = 7, P < 0.05), maintaining intracellular sulfhydryl concentration similar to that of control cells (18.0 ± 1.9 nmol/mg protein). Neither 20 mM PDTC nor 10 mM DETC had any effect on the levels of intracellular acid-soluble thiols in unstimulated cells (15.49 ± 1.3 and 15.5 ± 0.8 nmol/mg protein) or in cells exposed to IL-1 or H₂O₂.

View larger version (26K): [in this window] [in a new window]   Fig. 4.   Effect of IL-1 (50 U/ml) or H2O2 (10 mM) for 30 min on intracellular acid-soluble thiols in presence and absence of 20 mM N-acetylcysteine (NAC), 20 mM pyrrolidine dithiocarbamate (PDTC), and 10 mM diethyldithiocarbamate (DETC). * P < 0.05 from control untreated cells; + P < 0.05 from control H2O2-treated cells (n = 7 preparations).

Effects of antioxidants and metal chelators on IL-1-induced activation of NF-B. Figure 5 summarizes the influence of NAC and dithiocarbamates on the kinetics of NF-B activation on treatment with IL-1. Western blotting of total cell lysates indicates that NAC had no significant effect on the kinetics of phosphorylation and depletion of I-B after IL-1 treatment (Fig. 5A). After 30 min of IL-1 treatment, I-B levels in NAC-treated cells were 20 ± 7% of those in resting cells compared with 26 ± 6% in control cells (n = 7, not significant). In contrast, PDTC (500 mM) had a partial inhibitory effect on IL-1-induced NF-B activation by delaying I-B depletion, resulting in 72 ± 2% of that in resting cells still remaining after 30-min IL-1 treatment compared with 26 ± 6% in control treated cells. However, IL-1-induced depletion of I-B was not prevented by this concentration of PDTC; therefore, 20 mM PDTC was used in an attempt to completely abolish NF-B activation in response to IL-1. The appearance of the slower-migrating I-B species in response to IL-1 treatment was greatly attenuated by 20 mM PDTC (Fig. 5A), and the I-B protein levels remained preserved for longer periods of time. In PDTC-treated cells, I-B levels after 30 min of IL-1 treatment were 73 ± 20% of those in resting cells compared with 26 ± 6% in cells treated with IL-1 alone (P < 0.05, n = 6). Nevertheless, I-B became depleted in the presence of high concentrations of PDTC after 45-60 min of IL-1 exposure. The possible cytotoxic effect of this concentration (20 mM) of PDTC was investigated using neutral red cytotoxicity assay and by monitoring morphological changes by phase-contrast microscopy. After 4-h treatment with 20 mM PDTC, cell viability was comparable to that of control treated cells (101 ± 7% of control values). Furthermore, the morphology of PDTC-treated cells was not different from that of control cells. Twenty millimolar PDTC also preserved the NF-B-I-B complex in the presence of IL-1 for up to 45 min (data not shown). Similar results were obtained with DETC, for which I-B levels were 42 ± 11% those in resting cells after 30 min of IL-1 treatment. Separation of nuclear and cytosolic proteins demonstrated that PDTC and DETC significantly inhibited nuclear accumulation of NF-B p65 after 30 min of IL-1 treatment (183 ± 11 and 170 ± 49% of that in resting cells, respectively, compared with 341 ± 35% in cells treated with IL-1 alone, P < 0.05, n = 6; Fig. 5B, bottom). The decrease in nuclear NF-B is reflected in the decreased binding of nuclear proteins to NF-B DNA (Fig. 5C). NAC, which had no effect on I-B depletion, also had no effect on nuclear NF-B p65.

View larger version (25K): [in this window] [in a new window]   Fig. 5.   Effect of antioxidants and metal chelators on IL-1-induced activation of NF-B. A: Western blot analysis of time course of I-B depletion induced by IL-1 (50 U/ml) after 30-min pretreatment with medium alone, 20 mM NAC, 20 mM PDTC, or 10 mM DETC. Slower-migrating protein, indicated as I-Bp, corresponds to phosphorylated form of I-B. B: Western blot analysis of nuclear (top) and cytosolic (bottom) proteins from same cells after 30-min treatment with IL-1 (50 U/ml). C: EMSA showing effects of NAC, PDTC, and DETC on NF-B DNA binding of nuclear proteins induced by IL-1 (50 U/ml).

View larger version (32K): [in this window] [in a new window]   Fig. 6.   Effect of antioxidants on I-B resynthesis after treatment with IL-1 (50 U/ml). A: representative Western blots showing total cell I-B. B: optical density analysis of I-B Western blots after 30-min (I-B depletion) and 90-min (I-B resynthesis) treatment with IL-1. * P < 0.05 from appropriate control cells treated with IL-1 for indicated time in absence of NAC and dithiocarbamates (n = 6-7 preparations).

Effect of prooxidant conditions on NF-B activation. To assess the influence of prooxidant conditions on activation of NF-B human aortic SMC were treated with a range of H₂O₂ concentrations. Incubating the cells with increasing concentrations of H₂O₂ (1 µM-10 mM) for 30 min did not induce any changes in I-B (Fig. 7A). The protein synthesis inhibitor cycloheximide was present throughout to prevent resynthesis of I-B and to eliminate potential masking of I-B depletion. Furthermore, no evidence of nuclear accumulation of NF-B p65 or binding of nuclear proteins to the B consensus DNA was observed when cells were challenged with increasing concentrations of H₂O₂ (Fig. 7B). On the basis of the lack of effect of these concentrations of H₂O₂ on NF-B activation within 30 min, we investigated the effects of a high concentration of H₂O₂ (10 mM) on I-B over a time course of 1 h in the presence of cycloheximide to prevent I-B resynthesis. Figure 7C, top, shows that there was no evidence of I-B phosphorylation or degradation throughout the 60-min time course. Immunoprecipitating NF-B p65 from the same H₂O₂-treated cells and immunoblotting the resulting protein samples to determine the amount of I-B coprecipitated revealed that the p65 subunit of NF-B remained complexed to I-B for the duration of H₂O₂ exposure (Fig 7C, middle). More than 60 min in the presence of 10 mM H₂O₂ was cytotoxic to the cells, as determined by neutral red cytotoxicity assay and trypan blue exclusion test; therefore, 1 mM H₂O₂ was used to study an extended time course. However, up to 4 h in the presence of H₂O_2, I-B protein levels in whole cell preparations remained unchanged (data not shown).

View larger version (33K): [in this window] [in a new window]   Fig. 7.   Effect of H2O2 on NF-B activation in human aortic SMC. A: optical density analysis of Western blots showing cytosolic I-B levels after 30-min treatment with increasing concentrations of H2O2. B: nuclear proteins from cells treated with increasing concentrations of H2O2 for 30 min were analyzed for DNA binding activity by EMSA using 32P-labeled double-stranded oligonucleotide containing a specific B consensus sequence. C: Western blot analysis showing effects of H2O2 (10 mM) for increasing lengths of time on I-B. Top: levels of I-B in total cell preparations. Middle: I-B coprecipitated with NF-B p65 from same total cell preparation. Bottom: staining of NF-B p65 immunoreactive proteins from this immunoprecipitation demonstrating even loading of lanes.

	DISCUSSION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

In this study we have found members of the NF-B family of proteins in resting human aortic SMC and observed activation of this transcription factor by inflammatory cytokines. This activation involves phosphorylation of the main inhibitory protein I-B followed by dissociation of NF-B p65 from I-B and degradation of I-B. The transcription factor then moves to the nucleus, where it binds to the B element in the enhancer/promoter region of target genes. From the results obtained by EMSA and supershift assays, the main NF-B species in this cell type appears to consist of p65 and p50 subunits of NF-B. The activation pathway does not seem to be different from the pathway described in other systems (15, 17), suggesting the widespread and perhaps ubiquitous nature of this reaction.

In this study the transient depletion of I-B is followed by the resynthesis and gradual reappearance of new I-B protein in the cytoplasm. This appears to be associated with a slow decrease in the quantity of nuclear p65 and the DNA binding capacity of nuclear proteins. This observation supports the hypothesis that the newly synthesized inhibitor may have the capacity to regulate NF-B activation by assisting the NF-B p65 subunit back to the cytoplasm and into its original inactive state (6).

Because a diverse range of stimuli have the capacity to activate NF-B, it has been suggested that the signals from all these agents converge at one common point in the pathway of activation. In HeLa and Jurkat T cells, H₂O₂ is able to activate NF-B, and chemically unrelated antioxidants were shown to inhibit the activation of NF-B induced by a variety of agents (27, 28). For these reasons, ROI have become the popular choice for the intracellular messenger common to all stimuli. However, this model of ROI as the universal messenger has been brought into question by several studies. NF-B could not be activated in a murine thymoma line, EL4.NOB, or the human epithelial carcinoma cell line KB by H₂O₂ even after 4 h, and the activation by IL-1 in these cell lines could not be inhibited by the antioxidant NAC (14). Another study has also shown that NF-B activation involving ROI is selective to only some cell lines. In the human ovarian carcinoma cell line OVCAR-3, H₂O₂ was able to activate NF-B but only after a long exposure time and I-B depletion was only partial. Although IL-1 in this same cell line caused a rapid activation of the transcription factor, it did not cause an increase in intracellular H₂O₂ (9). Other transformed epithelial cell lines (SK-OVA-3 and MCF7-A/Z) have also been studied, and the IL-1-mediated NF-B activation could not be inhibited by either NAC or PDTC (9). Hence, the controversy over the involvement of ROI in this apparently complex sequence of events leading to the activation of NF-B still requires attention.

We found that IL-1, at concentrations that activate NF-B in human aortic SMC, did not cause oxidant stress sufficient to oxidize DCFH and produce a fluorescent signal or deplete intracellular reduced sulfhydryls. The lack of effect of IL-1 on oxidation of the internalized dye DCFH and intracellular sulfhydryl content cannot be caused by specific cell preparations, because the study was performed in cells obtained from several different samples of human donor aorta. However, the nature of these assays may not detect a subtler intracellular oxidant stress in the microenvironment of the NF-B complex in response to cytokine exposure. It is interesting to note that Marumo et al. (23) also failed to detect increased superoxide anion release from a human aortic smooth muscle cell line in response to IL-1 exposure.

Another line of evidence against the role of intracellular oxidant stress in this study is the failure of NAC to inhibit IL-1-induced activation of NF-B. Intracellular thiols represent one of the primary targets of oxidant stress and the principal defense system to prevent oxidation of macromolecules. Inability of NAC to augment antioxidant capacity can be ruled out, because intracellular low-molecular-weight sulfhydryls increased significantly on NAC treatment and NAC partially prevented sulfhydryl depletion and oxidation of DCFH in response to H₂O₂. Taken together, these data suggest that augmentation of cellular antioxidant capacity is not sufficient to prevent IL-1-induced NF-B activation in human aortic SMC.

In the present study, dithiocarbamates appear to partially protect against IL-1-induced NF-B activation, delaying I-B depletion and nuclear accumulation and DNA binding of NF-B. The significance of the inhibition of NF-B activation by dithiocarbamates on endogenous gene induction can be determined by analysis of I-B resynthesis. I-B has been documented to possess a critical NF-B promoter element, and it has been suggested to be regulated by NF-B activation. It is therefore expected that I-B resynthesis after depletion reflects endogenous NF-B-driven gene induction (19). Consistent with this assumption, PDTC and DETC significantly attenuated I-B resynthesis after IL-1 treatment, whereas NAC remained ineffective in this measure of NF-B-driven gene induction.

Dithiocarbamates are widely used compounds whose metal-chelating properties have been exploited for the treatment of metal poisoning (34) and acquired immunodeficiency syndrome (20, 35). It has been suggested that metal chelation might reduce oxidant stress by preventing formation of toxic free radicals such as hydroxyl radicals from H₂O₂ via the Fenton reaction. However, the ultimate effect of these compounds on cellular redox state may be more complex. It has been shown, for example, that DETC can produce a net prooxidant condition by increasing superoxide anion production via inhibition of superoxide dismutase in vascular endothelial cells (22). Because PDTC and DETC had no influence on intracellular sulfhydryl in either the presence or absence of H₂O₂, we conclude that the partial inhibitory effects of dithiocarbamates on NF-B activation might relate to their metal-chelating properties, which may be independent of their effects on the cellular redox state. It is possible that the activity of the IL-1-activated signaling mechanism or activation of the I-B kinase requires heavy metals, a mechanism that could be attenuated by dithiocarbamates.

This study has shown that under the conditions used H₂O₂ is not capable of activating NF-B in human aortic SMC. Since H₂O₂ caused oxidation of DCFH and depletion of intracellular thiols, it appears that H₂O₂ produced a significant intracellular oxidant stress but remained ineffective in the activation NF-B. One study suggested that H₂O₂ caused I-B phosphorylation at tyrosine residues rather than serine phosphorylation, causing the dissociation of NF-B from I-B, releasing NF-B without the degradation of the inhibitor (18). However, this is unlikely to be the case in the present study because throughout the time course of H₂O₂ exposure I-B coprecipitated with p65, indicating that the NF-B dimer was still held inactive in the cytoplasm by the inhibitor I-B. Also, there was no detectable binding of nuclear proteins to the B consensus DNA after incubation with increasing concentration of H₂O₂ for 30 min, the time frame in which IL-1-induced activation of NF-B was maximal. These data suggest that oxidant stress, in its own right, is not able to activate NF-B and is not responsible for IL-1-induced activation in our model. In a recent study, Marumo et al. (23) found evidence for oxidant stress-induced activation of NF-B in a human smooth muscle cell line. They reported that PDGF caused a time-dependent release of superoxide anion and a small (2-fold) increase in nuclear NF-B. However, this response was only partially blocked by superoxide anion scavengers, suggesting that other signal transduction pathways activated by PDGF could have been involved (23). Moreover, IL-1 produced activation of NF-B in the absence of superoxide production in that study.

In conclusion, this study suggests that heavy metals, but not intracellular ROI, are likely involved in IL-1-induced activation of the transcription factor NF-B in human aortic SMC, a mechanism that appears to be a primary event in gene activation associated with clinically relevant pathologies, such as progression of atherosclerosis and systemic inflammatory responses (5, 8, 13).