INTRODUCTION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

VASCULAR SMOOTH MUSCLE (VSM) cells migrate from the tunica media to the intima and proliferate to form a neointima (32). These are critical events in the development of atherosclerosis and in restenosis after balloon angioplasty (32). A multitude of growth factors and cytokines upregulated at the site of atherosclerotic lesions play an important role in mediating VSM cell migration and proliferation (32). Growth factors that differ in their ability to regulate VSM cells activate similar signaling pathways. For instance, platelet-derived growth factor is both mitogenic and motogenic to smooth muscle cells, whereas basic fibroblast growth factor is only mitogenic (32). The mitogen-activated protein kinase (MAPK) pathway regulates both mitogenic and motogenic responses in VSM cells. It has been suggested (27) that the duration of MAPK activation may be important in regulating different VSM cell functions.

The MAPK pathway is a cascade of serine-threonine kinases downstream of a wide variety of growth factors and cytokines (14). In vivo studies have demonstrated that extracellular signal-regulated kinase (ERK) undergoes a biphasic activation after arterial balloon-stretch injury-an early transient phase followed by a sustained phase that persists for 2 to 8 days (23, 29). Similarly, in vitro studies (20, 27) have demonstrated that growth factors and cytokines stimulate biphasic ERK activation. The two phases of ERK activation are involved in regulation of different VSM cell functions-the early phase in cell migration and the late phase in proliferation (27). Recent studies have demonstrated that interleukin-1 (IL-1) increases the production of reactive oxygen species (ROS) in VSM cells (1). ROS are important signaling molecules in regulation of VSM cell migration and proliferation (17, 30, 34). However, the role of ROS in sustained ERK activation has not been characterized. ROS participate as second messengers in the regulation of the MAPK pathway, survival kinase Akt, transcription nuclear factor-B, and Ca²⁺ signaling (for a review, see Ref. 16). In addition, ROS regulate a variety of genes including adhesion and chemotactic molecules, suggesting a possible role for ROS in the pathogenesis of atherosclerosis (for a review see Ref. 16).

Matrix metalloproteinases (MMPs) are enzymes involved in the breakdown of extracellular matrix and are suggested to play a role in the pathogenesis of vasculoproliferative diseases (8). MMP-9 is transiently up-regulated after vascular injury (2). Overexpression of MMP-9 facilitates VSM cell migration (25). A recent study (5) demonstrated that ROS stimulate MMP-9 secretion in human fetal membranes. The ROS scavenger N-acetyl-L-cysteine (NAC) inhibits MMP-9 secretion in atherosclerotic lesions isolated from hypercholesterolemic rabbits (12). A recent study (26) suggested that sustained ERK activation is involved in MMP-9 induction in epidermal cells. Therefore, we hypothesized that IL-1 stimulates MMP-9 induction via the ROS-ERK pathway in VSM cells.

To investigate whether IL-1 stimulates superoxide generation in VSM cells, we used the redox-sensitive fluorescent dye dihydroethidium. To determine the role of sustained ERK activation in mediating MMP-9 induction after exposure to IL-1, cells were treated with the ERK pathway inhibitor PD-98059 at different time intervals after stimulation with IL-1, and MMP-9 mRNA levels were measured by RT-PCR. Cells were treated with the antioxidant NAC to determine the role of ROS in IL-1-stimulated ERK activation and MMP-9 induction. We confirmed the role of ROS in IL-1-stimulated MMP-9 induction by overexpression of human manganese superoxide dismutase (MnSOD) gene in VSM cells. Our results suggest that ROS are important in causing IL-1-stimulated sustained ERK activation and that the sustained ERK activation is required for MMP-9 induction in VSM cells.

	MATERIALS AND METHODS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

Materials. Chemicals and materials were obtained from the following sources: 10% gelatin zymography precast gels, renaturing buffer, developing buffer, and Seablue molecular weight markers (Novex); human rIL-1 (R&D Systems); rabbit and mouse anti-mouse IgG-HRP conjugate and rabbit anti-ERK 1/2 antibody (Santa Cruz Biotechnology); PD-98059 and mouse anti-phospho-ERK (pERK) monoclonal antibody (Cell Signaling Technology); dihydroethidium (Molecular Probes); RNA-STAT60, (Teltest); Supersignal chemiluminescence detection kit (Pierce); Superscript 1-Step RT-PCR kit (Life Technologies); NAC and other chemicals not listed were of the highest grade available from Sigma.

Cell culture. Cells were isolated from Wistar male rats as described previously (19). Animals were maintained and used in compliance with the principles set forth in the Guide for Care and Use of Laboratory Animals (National Institutes of Health). All procedures were approved by The University of Iowa Animal Care and Use Committee. Cells were grown and subcultured weekly in DMEM low-glucose medium supplemented with 10% fetal bovine serum and antibiotics (100 µg/ml of streptomycin, 100 U/ml penicillin, and 2.5 µg/ml fungizone). VSM cells were used in experiments between the passages 3 and 6.

Treatment of cells and collection of conditioned media. VSM cells were serum starved in DMEM/0.1% BSA for 48 h. Cells were stimulated with IL-1 (5 ng/ml), and the ERK pathway inhibitor PD-98059 (50 µM) was added at different time intervals after stimulation. Similarly, cells were stimulated with IL-1 in the absence or presence of NAC (10 mM), and conditioned media was collected after 24 h, centrifuged to remove cell debris, and stored in aliquots at 80°C.

Adenovirus-mediated gene transfer. Cells were infected with adenoviral vectors carrying the gene of interest (either LacZ or MnSOD). The adenoviral vectors were prepared by The Vector Core, The University of Iowa College of Medicine as described elsewhere (19, 33). The adenovirus was deleted of sequences in the E1A, E1B, and E3 regions, impairing the ability of the virus to replicate in nonpermissive cells. Confluent VSM cells were infected with 50-100 multiplicity of infection of Ad5/CMV.LacZ or Ad5/CMV.MnSOD virus in 0.1% BSA/DMEM for 3 h, and then fresh serum-free DMEM/0.1% BSA was added. Ad5/CMV.LacZ infected cells were used as control. Expression of MnSOD gene was confirmed by Western blot analysis using a rabbit polyclonal antibody.

Estimation of intracellular superoxide. The redox-sensitive probe dihydroethidium was used to detect IL-1-stimulated superoxide generation (4). Dihydroethidium is oxidized by superoxide to ethidium, which binds to nuclear DNA and gives a bright red nuclear fluorescence. VSM cells were subcultured to 60-70% confluence on circular 25-mm glass coverslips and were serum starved in 0.1% BSA/DMEM. After 24 h, the serum-starved cells were stimulated with IL-1 for 0, 30, 60, or 90 min. Dihydroethidium (5 µM) was added to each well for 60 min. Cells were then rinsed with 0.1% BSA/DMEM and mounted on a chamber. Fluorescence was detected using a confocal laser scanning microscope (Ziess). Excitation wavelength was 488 nm, and emission wavelength was 650 nm. Dihydroethidium alone did not produce any fluorescence. Images were collected and analyzed by a confocal assistant program.

Western blot analysis. Cells were serum starved for 24 h, treated with NAC (10 mM), and then stimulated with IL-1 (5 ng/ml). After the desired time, cells were rinsed with ice-cold phosphate-buffered saline to terminate the reaction, and lysed with lysis buffer as described previously (33). The lysates were centrifuged at 14,000 g for 10 min at 4°C, and the supernatants were collected. Total proteins were quantified using Bio-Rad (Bradford) reagents. Supernatant (100 µl) from each sample was mixed with 6× sample buffer and boiled for 5 min. An equal amount of total protein (15 µg) from each sample was resolved by SDS-PAGE and transferred to Immobilon-P membranes. Membranes were serially incubated with: 1) blocking buffer of 100 mmol/l NaCl, 10 mmol/l Tris · HCl (pH 7.5), 0.1% (vol/vol) Tween 20, and 5% (wt/vol) nonfat milk for 30 min; 2) mouse pERK antibody diluted (1:1,000) in blocking buffer for 1 h; and 3) anti-rabbit/mouse IgG-HRP diluted (1:2,000) in blocking buffer for 1 h. Immunoreactive bands were visualized using Supersignal chemiluminescence detection kit (Pierce) on a Bio-Rad Fluoro-S-Max Chemidoc system. To determine total ERK levels the blots were stripped using stripping buffer (10 mM 2-mercaptoethanol, 62.5 mM Tris · HCl and 2% SDS, pH 6.8) at 65°C for 30 min and reprobed with rabbit polyclonal ERK antibody (1:2,000). MnSOD protein expression was determined using rabbit polyclonal MnSOD antibody. Blots were quantified using Bio-Rad-Quantity One software.

Detection of MMP-9 activity by zymography. Gelatinase activity in conditioned-media (5 µl) collected from cell cultures was measured using zymography as described previously (19, 21). Equal amounts of conditioned media (5 µl) were subjected to electrophoresis using Novex 10% zymography gels containing 0.1% gelatin. Gels were washed with renaturing buffer (Novex) for 30 min and incubated at 37°C for 20 h in developing buffer (Novex). After 20 h, gels were stained with Coomassie blue. Gels were scanned, and bands were quantified using Bio-Rad-Quantity One software.

Analysis of mRNA by RT-PCR. Cells were stimulated with IL-1 and treated with PD-98059 at different time intervals. Total RNA was collected 12 h after stimulation using RNA-STAT60 (Teltest) and was quantified and stored at 80°C for future use. One-step RT-PCR was set up using MMP-9 primers 5'-CTT AGA TCA TTC TTC AGT GCC-3' (sense) and 5'-GAT CCA CCT TCT GAG ACT TCA-3' (antisense) (9). GAPDH primers 5'-ATT TGG CCG TAT TGG CCG CCT-3' (sense) and 5'-ACA GCC TTG GCA GCA CCA GTG G-3' (antisense) were used as internal controls to normalize for the variations in RNA loading, and 35 cycles were performed for each reaction. The RT-PCR products (0.6 kb for GAPDH and 0.7 kb for MMP-9) were resolved on 1.5% agarose gels, and the ethidium bromide-stained products were visualized and quantified using Bio-Rad Fluoro-S-Max Chemidoc system. MMP-9 induction was expressed as MMP-9/GAPDH product density.

Data analysis. Western blots, RT-PCR, and zymogram gels were scanned, and relative intensity of bands was determined by densitometry. Statistical analysis was carried out by Student's t-test by use of a commercially available program (Statview; Cricket Software). Differences were considered significant at P < 0.05. The results are presented as means ± SE; n = number of separate experiments.

	RESULTS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

NAC treatment and MnSOD gene transfer inhibit IL-1-stimulated superoxide generation in VSM cells. To determine the effects of NAC and MnSOD gene transfer on IL-1-stimulated superoxide production, cells were treated with the redox-sensitive dye dihydroethidium for 60 min, and images were captured by confocal microscopy. Serum-starved control cells did not show any superoxide fluorescence (Fig. 1), whereas stimulation of cells with IL-1 (5 ng/ml) markedly increased superoxide generation (red nuclear fluorescence) in the cells at 90 min (Fig. 1). To confirm that the nuclear fluorescence is due to superoxide generation, we treated VSM cells with the ROS scavenger NAC or infected cells with adenovirus expressing the human MnSOD gene to scavenge superoxide radical. Treatment of cells with NAC and MnSOD overexpression inhibited the IL-1-stimulated increase in superoxide (Fig. 1). These results demonstrate that IL-1 stimulates superoxide generation in VSM cells and that NAC treatment or MnSOD overexpression can significantly inhibit this response.

View larger version (27K): [in this window] [in a new window]   Fig. 1.   Interleukin-1 (IL-1) stimulates superoxide production. Vascular smooth muscle (VSM) cells grown on 25-mm coverslips were either infected (with Ad5/CMV.LacZ or Ad5/CMV.MnSOD) or were serum starved for 24 h. Cells were left untreated or stimulated with IL-1 (5 ng/ml) in the absence or presence of NAC (10 mM). Dihydroethidium (5 µM) was added to each well, and confocal microscope images were captured 90 min after stimulation with IL-1. Excitation and emission wavelengths were 488 and 630 nm, respectively. A: representative images presented from one of 3 experiments showed similar results. B: densitometric analysis of superoxide levels (mean ± SE, n = 3) is shown. IL-1 stimulation significantly (P < 0.05) increased superoxide levels as evidenced by an increase in red nuclear fluorescence due to breakdown of dihydroethidium to ethidium compared with unstimulated cells. *Treatment of cells with NAC or overexpression of MnSOD gene significantly (P < 0.05) inhibited IL-1-stimulated superoxide production. NAC, N-acetyl-L-cysteine; MnSOD, manganese superoxide dismutase.

NAC inhibits IL-1-stimulated sustained ERK activation. To determine whether IL-1 stimulation of VSM cells results in biphasic ERK activation, serum-starved cells were treated with IL-1 and cell lysates were collected after various time intervals. ERK activation was measured as an increase in pERK levels using a pERK-specific antibody. Stimulation of VSM cells with IL-1 (5 ng/ml) resulted in a transient increase in pERK levels at 15 min that decreased to near baseline levels by approximately 30-60 min after stimulation (Fig. 2A). A second wave of increase in pERK levels appeared around 2-3 h and was sustained for up to 8 h (Fig. 2A).

View larger version (41K): [in this window] [in a new window]   Fig. 2.   NAC inhibits IL-1-stimulated sustained extracellular signal-regulated kinase (ERK) activation. Serum-starved VSM cells were left untreated or stimulated with IL-1 (5 ng/ml) in the absence or presence of NAC (10 mM), and cell lysates were collected at different time intervals and analyzed by Western blotting for pERK. Densitometric analysis of phospho-ERK (pERK) bands is presented as mean ± SE (n = 4). A: a representative Western blot demonstrates IL-1-stimulated increase in pERK levels over a period of time. IL-1 stimulation markedly increased pERK levels at 15 min and between 2 and 8 h compared with unstimulated cells. B: graphical representation of the effect of NAC on IL-1-stimulated pERK increase is shown. Treatment of cells with NAC (10 mM) had no effect on IL-1-stimulated increase in pERK levels at 15 min. *However, NAC treatment significantly (P < 0.05) decreased pERK levels at 8 h compared with IL-1-stimulated control (Con) cells.

IL-1-stimulated sustained ERK activation is critical for MMP-9 induction. To evaluate the role of sustained ERK activation in IL-1-stimulated MMP-9 induction, we treated VSM cells with the ERK-pathway inhibitor PD-98059 (50 µM) at different time intervals (0 through 10 h) after IL-1 stimulation. Total RNA was collected 12 h after IL-1 stimulation and analyzed for MMP-9 mRNA levels using RT-PCR. PD-98059 treatment inhibited IL-1-stimulated MMP-9 induction. Furthermore, addition of PD-98059 for up to 4 h after IL-1 stimulation significantly (P < 0.05) inhibited MMP-9 induction (Fig. 3). PD-98059 almost completely inhibited ERK activation (data not shown). Because PD-98059 was effective in inhibiting MMP-9 induction even when added 2-4 h after IL-1 stimulation, we conclude that sustained ERK activation is required for MMP-9 induction.

View larger version (33K): [in this window] [in a new window]   Fig. 3.   IL-1-stimulated sustained ERK activation is required for MMP-9 induction. Serum-starved VSM cells were left untreated or stimulated with IL-1, and PD-98059 was added at different time intervals after stimulation. Total RNA was collected and analyzed for MMP-9 RNA. Data are presented as means ± SE (n = 3). A: a representative RT-PCR from one of 3 experiments shows similar results. B: graphic representation of the summary data is shown. IL-1 stimulation resulted in matrix metalloproteinase-9 (MMP-9) induction. *Treatment of cells with PD-98059 until 4 h after IL- stimulation resulted in significant (P < 0.05) inhibition of MMP-9 induction.

NAC treatment and MnSOD overexpression inhibit IL-1-stimulated MMP-9 induction. Our earlier results demonstrated that sustained ERK activation is important for MMP-9 induction and that ROS are important for sustained ERK activation (Figs. 2 and 3). Therefore, we investigated the role of ROS in IL-1-stimulated MMP-9 induction. VSM cells were stimulated with IL-1 in the absence or presence of NAC as described above, and conditioned media were collected after 24 h for gelatin zymography. IL-1 stimulation resulted in MMP-9 induction. Treatment of cells with NAC significantly (P < 0.05) inhibited the IL-1-stimulated MMP-9 induction (Fig. 4). Quantitative analysis of zymograms showed a 50% decrease in IL-1-stimulated MMP-9 activity in NAC-treated cells compared with that of untreated cells (Fig. 4). Treatment of cells with NAC alone did not affect cell viability at 24 h (data not shown). To verify the role of superoxide in MMP-9 induction, we overexpressed MnSOD in VSM cells. Overexpression of MnSOD significantly inhibited IL-1-stimulated MMP-9 induction (Fig. 4). These results suggest that superoxide plays an important role in IL-1-stimulated MMP-9 induction in VSM cells.

View larger version (28K): [in this window] [in a new window]   Fig. 4.   NAC treatment and MnSOD overexpression inhibits IL-1-stimulated MMP-9 induction. Serum-starved VSM cells were either infected (with Ad5/CMV.LacZ or Ad5/CMV.MnSOD) or were serum starved for 48 h. Thereafter, cells were stimulated with IL-1 (5 ng/ml) in the absence or presence of NAC (10 mM), and conditioned media were analyzed for MMP-9 activity by gelatin zymography. A: representative zymograms from 4 similar experiments are shown. B: summary data of MMP-9 activity is presented as means ± SE (n = 4). IL-1 stimulated MMP-9 induction. NAC treatment or MnSOD overexpression resulted in a significant (*P < 0.05) inhibition of IL-1-stimulated MMP-9 induction.

	DISCUSSION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

In the present study, we have demonstrated that 1) IL-1 stimulates superoxide production in VSM cells, 2) superoxide is required for IL-1-stimulated sustained ERK activation in VSM cells, and 3) sustained ERK activation is required for IL-1-stimulated MMP-9 induction. Our results provide a molecular basis for cytokine-stimulated MMP-9 induction and suggest ROS-ERK signaling as a therapeutic target to inhibit IL-1-stimulated MMP-9 induction.

ROS are important to the pathogenesis of vasculoproliferative diseases (18, 24, 28). ROS play a signaling role in growth factor-stimulated VSM cell migration and proliferation (17, 30, 34). Our results demonstrate that IL-1 increases superoxide generation in VSM cells. Treatment of cells with the ROS scavenger NAC or human MnSOD gene expression significantly (P < 0.05) decrease superoxide levels, verifying that IL-1 stimulation results in superoxide generation in VSM cells. IL-1 is increased at the site of atherosclerotic lesions and in injured arteries (7, 11). Therefore, it functions as one of the main stimulators of ROS generation at the site of vascular injury and contributes to the development of restenosis by inducing cell migration and proliferation.

In vitro studies have demonstrated that growth factors and cytokines such as IL-1 result in biphasic ERK activation-an early transient phase lasting for 5-30 min followed by a late sustained phase that lasts for up to 24 h (20, 27). Similarly, in vivo studies have also shown a biphasic ERK activation in balloon-injured arteries (23, 29). Although ROS are known to be involved in ERK activation (1, 35), the role of ROS in the various phases of cytokine-stimulated ERK activation had not been defined previously. In the present study we have demonstrated that IL-1 stimulates biphasic ERK activation. The sustained ERK activation coincides with the increase in ROS levels suggesting a role for ROS in mediating ERK activation. Treatment of cells with the ROS scavenger NAC had no effect on the early transient phase of ERK activation but significantly (P < 0.05) inhibited sustained (8 h) ERK activation. These data suggest that the initial transitory phase of ERK activation is perhaps mediated by IL-1 receptor-mediated signaling, whereas the sustained phase is mediated by superoxide. Other mechanisms such as IL-1-induced autocrine factors may also contribute to sustained ERK activation after stimulation with IL-1. These results are significant in light of a previous study that demonstrated that the early phase of ERK activation is important for VSM cell migration, whereas the sustained ERK activation is more critical for cell proliferation (27). Our results extend the understanding of the mechanisms involved in sustained ERK activation.

Development of atherosclerotic lesions and restenosis after arterial injury requires the breakdown of extracellular matrix (8). MMP-9 is an inducible enzyme that increases in the blood vessel wall at the site of atherosclerotic lesions and after balloon injury (2, 13). Studies in other cell types have suggested a role for ERK activation in MMP-9 induction (26, 31, 36). Our data demonstrate that treatment of VSM cells with the ERK-pathway inhibitor PD-98059 resulted in the inhibition of MMP-9 induction. Treatment of VSM cells with PD-98059 even up to 4 h after IL-1 stimulation significantly inhibited MMP-9 induction. These data indicate that IL-1-stimulated sustained ERK activation is required for MMP-9 induction.

We have observed that ROS are required for sustained ERK activation and that sustained ERK activation in turn is required for MMP-9 induction. These findings suggest that ROS-ERK signaling pathways are necessary for IL-1-stimulated MMP-9 induction. Our results demonstrate that inhibiting ROS generation by NAC treatment or MnSOD overexpression significantly inhibited IL-1-stimulated MMP-9 induction. Previous studies have also demonstrated that increased production of ROS leads to MMP-9 induction in human fetal membranes (5). In addition, we have previously demonstrated that nitric oxide inhibits IL-1-stimulated MMP-9 induction (19). Nitric oxide reacts with superoxide with a high affinity (6). Our recent observations demonstrate that nitric oxide inhibits IL-1-stimulated sustained ERK activation and superoxide production in VSM cells (19a). Taken together with the findings in our present study, these results suggest that nitric oxide may attenuate IL-1-stimulated MMP-9 induction by inhibiting ROS-ERK signaling. It is possible that nitric oxide may either attenuate the activity of ROS generating enzymes or inhibit signaling intermediates involved in the activation of these enzymes.

In conclusion, we have presented evidence that ROS mediate IL-1-stimulated sustained ERK activation, which is required for MMP-9 induction in VSM cells. Our results provide a molecular basis for some of the beneficial effects of antioxidant therapy observed in human and experimental models of atherosclerosis and restenosis and suggest SOD gene therapy as a useful therapeutic approach to inhibit MMP-9 induction at the site of vascular injury (15, 22). Further studies are required to investigate the mechanisms involved in the IL-1-stimulated increase in ROS and its downstream regulation of other signaling pathways in VSM cells.