hypertension has been associated with arterial hypertrophy and an increase in extracellular matrix, especially in collagen content (9, 19,39). Although these alterations are partly related to elevated arterial pressure, other factors have been found to stimulate vascular remodeling during the development of hypertension. Recently, several groups of investigators (4, 31, 34) have reported that hypertension causes an increased vascular production of radical oxygen species such as superoxide anion (O${}_2^{-}$) in various models of hypertension. The induction of radical oxygen species is known to induce activation of the redox-sensitive transcription factor nuclear factor (NF)-κB (13), which is the most important transcription regulatory element in the intercellular adhesion molecule-1 (ICAM-1) promotor system (36). ICAM-1 modulates inflammatory cell adhesion to the vascular endothelium, and its expression is upregulated by cytokines in vascular endothelial cells and other cell types in vitro (2, 5, 7) as well as in atherosclerotic lesions in vivo (24, 26). Thus upregulation of ICAM-1 is suggested to play an important role in the pathogenesis of vascular remodeling.

3-Hydroxy-3-methylglutaryl (HMG) CoA reductase inhibitors (statins) have been widely used to reduce cardiovascular risks (28, 29,32). The major action of statins has generally been attributed to the well-documented low-density lipoprotein cholesterol-lowering properties of these drugs (20a). Although it is well known that lipid-rich plaques are more prone to rupture, the mechanisms by which statins reduce coronary events are not completely understood. Recently, there has been a suggestion that statins may exert effects separate from their cholesterol-lowering action, including promotion of endothelial nitric oxide (NO) synthesis in humans (35) and other animals (18, 37), but those effects are not completely understood.

The purpose of this study was to determine the effects of the HMG CoA reductase inhibitor fluvastatin on O${}_2^{-}$ production and ICAM-1 expression in a rat model with vascular remodeling induced by pressure overload. In addition, we also examined whether this effect of fluvastatin was attributable to either its inhibition of arterial hypertension or lowering of serum cholesterol under these conditions.

METHODS

The experiments in the present study were reviewed and approved by the Committee on Ethics of Animal Experiments of Tanabe Seiyaku and conducted according to the Guidelines for Animal Experiments of Tanabe Seiyaku and the Law (No.105) and Notification (No. 6) of the Japanese Government.

Animal experiments.

Nine-week-old male Sprague-Dawley rats (Charles River Japan; Tokyo, Japan) were maintained on standard rat chow and tap water ad libitum. Pressure overload was produced by constriction of the abdominal aorta, as described previously (21). Briefly, under pentobarbital sodium anesthesia, the abdominal aorta was constricted at a point proximal to the right renal artery with a piece of cotton thread and a blunted 20-gauge needle (external diameter, 0.9 mm), which was pulled out later. Sham-operated rats underwent similar surgical procedures except for the narrowing of the abdominal aorta. Rats were then randomly divided into the following four groups: sham-operated rats (sham group, n = 5); operated rats (Band group,n = 5); operated rats receiving 0.03 mg/ml fluvastatin (∼3 mg · kg⁻¹ · day⁻¹; B+FV3 group, n = 5); and operated rats receiving 0.1 mg/ml fluvastatin (∼10 mg · kg⁻¹ · day⁻¹; B+FV10 group, n = 5). Fluvastatin (Tanabe Seiyaku; Saitama, Japan) was given in the drinking water (∼30–40 ml per rat per day for 2 wk).

Two weeks after aortic banding, the rats were anesthetized, and a catheter was inserted into the right carotid artery. The blood pressure and heart rate were monitored with the carotid catheter connected to a preamplifier (AP-621G; Nihon Kohden; Tokyo, Japan) and a linerecorder (WR- 3310, Graphtech; Tokyo, Japan).

Measurement of serum lipids and lipid peroxides.

Total cholesterol and triglycerides in serum were determined using commercially available kits (Wako Pure Chemical; Osaka, Japan). The lipid peroxides in serum were determined as thiobarbituric acid-reactive substances (TBARS) using a commercially available kit (Wako Pure Chemical), and the results are given in nanomoles of malondialdehyde equivalents per milliliter.

Measurement of vascular O${}_2^{-}$ production.

O${}_2^{-}$ production levels were measured by the lucigenin-enhanced chemiluminescence (ECL) technique as previously described (34). Ring segments (5 mm) of the aorta were placed in modified Krebs-Hepes buffer containing lucigenin (0.25 mM), and chemiluminescence measured with a scintillation counter (Luminescence Reader BLR 301, Aloka; Tokyo, Japan). The counts were recorded after the intracellular superoxide scavenger Tiron (4,5-dihydroxy-1,3-benzenedisulfonic acid) was added to the vial. In all experiments, >90% of the signals from the aortic rings were scavenged by Tiron. The specific chemiluminescence signal was expressed as counts per minute minus the average background counts.

Hydroethidine, an oxidative fluorescent dye, was used to evaluate levels of O${}_2^{-}$ in situ as previously described (31). Unfixed frozen ring segments were cut into 30-μm-thick slices and mounted on slides. Hydroethidine (2 μM, Molecular Probes; Eugene, OR) was applied to each tissue section, and tissues were incubated at 37°C for 30 min. The tissue sections were then visualized with a Leica TCS NT confocal microscope (Leica; Munich, Germany), with fluorescence detected with a 585-nm long-pass filter.

Western blot analysis.

Thoracic aortas from four rats in each group were homogenized in 50 mM of Tris buffer (pH 7.4) containing 1% SDS and 10 mM EDTA. The homogenates were centrifuged, and the supernatants were removed. Their supernatant protein concentrations were determined by a bicinchoninic acid protein assay (Pierce; Rockford, IL). Protein (20 μg per lane) was loaded and electrophoresed through a 7.5% SDS-polyacrylamide gel. Proteins were transferred to a nitrocellulose membrane and incubated with a 1:2,000 dilution of the anti-rat ICAM-1 antibody (Seikagaku; Tokyo, Japan) and peroxidase-conjugated goat anti-mouse IgG (Amersham; Arlington Heights, IL). Bound antibody was visualized with the ECL system (Amersham).

Collagen morphometry.

Transverse aortic sections were cut into 5-μm-thick slices and stained with collagen-specific picrosirius red (0.1% Sirius Red F3BA in aqueous picric acid) for estimation of the thickening of the thoracic aorta wall and perivascular fibrosis, as previously described (11, 30).

Immunohistochemistry.

Thoracic aortas from five rats in each group were immediately embedded in optimum cutting tissue compound (Miles; Elkhart, IN) and frozen. The 5-μm-thick slices were fixed in acetone and then incubated with 0.3% H₂O₂ to quench endogenous peroxidase. The sections were preincubated with 10% horse serum to reduce nonspecific binding and incubated for 2 h at room temperature with anti-rat ICAM-1 monoclonal antibody (1 μg/ml, Seikagaku) or with anti-human endothelial NO synthase (eNOS) monoclonal antibody (1 μg/ml, Transduction Laboratories; Lexington, KY). The slides were washed and incubated with biotinylated horse anti-mouse IgG (Vector Laboratories). After avidin-biotin amplification, the samples were visualized with 3′,3′-diaminobenzidine and counterstained with hematoxylin.

Statistical analysis.

Data are expressed as means ± SE. Differences among groups were analyzed by one-way ANOVA, followed by a Tukey-Kramer test for multiple comparisons. A level of P < 0.05 was considered statistically significant.

RESULTS

Body weights, hemodynamic parameters, serum lipids, and lipid peroxides.

Body weights and heart rates did not significantly differ among groups. After 2 wk of treatment, the mean arterial pressure showed a rise in the Band, B+FV3, and B+FV10 groups (Table1). Total cholesterol, triglycerides, and TBARS did not significantly change among groups (Table 1).

O${}_2^{-}$ production.

When measured by lucigenin chemiluminescence, production of O${}_2^{-}$ in the aortic segments was higher in the Band group than in the sham group (Fig.1A). Treatment with fluvastatin dose dependently inhibited the increase in O${}_2^{-}$ production induced by pressure overload.

Aortas of banded rats had increased O${}_2^{-}$ levels, as measured by hydroethidine red fluorescence, compared with sham-operated rats (Fig. 2A). The increase in red fluorescence was observed in endothelial cells, media, and adventitia. The red fluorescence was reduced by the treatment with fluvastatin (Fig. 2A).

ICAM-1 expression.

Aortic ICAM-1 protein levels were significantly increased in the Band group (Fig. 1B). Treatment with fluvastatin dose dependently inhibited the increase in ICAM-1 induced by pressure overload (Fig.1B).

Immunoreactivity for ICAM-1 was only weakly present in the endothelial cells of the sham group (Fig. 2B, a). In the Band group, ICAM-1 immunoreactivity was intensely present in endothelial cells and adventitia (Fig. 2B, b). Treatment with fluvastatin markedly reduced the ICAM-1 immunoreactivity seen in the Band group (Fig. 2B, c and d).

eNOS immunohistochemistry.

Immunostaining for eNOS showed that the enzyme was present in the endothelium of all aortas (Fig. 3). No difference was observed between the sham and Band groups (Fig. 3,a and b). In contrast, treatment with fluvastatin clearly increased the eNOS immunoreactivity (Fig. 3, c andd).

Aortic remodeling.

Micrographs of the aortas obtained from the sham, Band, B+FV3, and B+FV10 groups are shown in Fig.4A. The wall-to-lumen ratios and perivascular fibrosis in the aortas were significantly greater in the Band group than in the sham group. Treatment with fluvastatin slightly reduced the wall-to-lumen ratios and significantly prevented perivascular fibrosis in the aortas (Fig. 4B).

DISCUSSION

This in vivo study demonstrates, for the first time, that HMG CoA reductase inhibition by fluvastatin prevents O${}_2^{-}$production and ICAM-1 expression in a rat model with vascular remodeling induced by pressure overload. In addition, fluvastatin exerted no influence on the arterial hypertension and serum lipids. The present findings clearly point toward an action of fluvastatin other than its cholesterol-lowering effect.

The increase in vascular O${}_2^{-}$ production detected by lucigenin-ECL was confirmed by in situ hydroethidine staining. Oxidation of hydroethidine to ethidium, as detected by this fluorescent technique, has been shown to be specific for O${}_2^{-}$(3). In diseased blood vessels, O${}_2^{-}$ may be overproduced by the endothelium (25), smooth muscle cells (17, 27), adventitial fibroblasts (38), or inflammatory cells that migrated to the vessel (6). Interestingly, our observation that fluorescence was particularly increased not only in endothelial and adventitial cells but also in vascular smooth muscle cells of the banded rats compared with the paired images of the sham-operated rats was consistent with the above reports. However, the mechanisms and enzyme systems responsible for increased O${}_2^{-}$ production in these various cell layers remain unidentified and may vary among the cell types.

It has been shown in rats that the increase in vascular O${}_2^{-}$ production occurs in various models of hypertension (4, 31, 34). Although treatment with fluvastatin showed a tendency to reduce the increased blood pressure, the mean blood pressure of the B+FV10 group was significantly higher than that of the sham group. However, fluvastatin inhibited vascular O${}_2^{-}$ production to a level comparable with that in the sham-operated rats. Therefore, the effects of fluvastatin cannot be explained by its antihypertensive action.

There are three possible mechanisms by which statins inhibit vascular O${}_2^{-}$ production in vivo. First, the observed effects of fluvastatin may be due to a lowering of NADPH oxidase activity through inhibition of the mevalonate step. In support of this concept, an increase in O${}_2^{-}$ production occurs during the development of hypertension, and enhanced NADPH oxidase activity has been described in the aortas of rats with angiotensin II-induced hypertension (27). Because the renin-angiotensin system is rapidly activated after aortic banding (1, 20), it is conceivable that an angiotensin II-coupled mechanism may be responsible for the increase in O${}_2^{-}$ production observed after aortic banding. Recently, several reports (15, 16) have shown that some of the direct effects of statins on the vascular wall are mediated by inhibition of isoprenoid but not cholesterol synthesis. Indeed, Wagner et al. (37) reported that NADPH oxidase activity might also be activated by isoprenylated GTP-binding protein and that a mevalonate-sensitive inhibition of phorbol ester-stimulated O${}_2^{-}$ production by atorvastatin was observed in isolated rat aortic segments. Second, recent reports have shown that a reduction in NO synthesis increases endothelial intracellular oxidative stress (14, 23) and that statins might directly upregulate eNOS activity and increase NO production (15, 18, 35, 38). Immunohistochemistry demonstrated increased eNOS expression in the endothelium of the aorta in the fluvastatin-treated rats. Therefore, we considered the possibility that both the scavenging of O${}_2^{-}$ production and the decreased expression of ICAM-1 by fluvastatin were due to an increase in NO production. In addition, the aorta from banded rats did not have decreases in the expression of eNOS. This observation is consistent with the previous report in the aorta from 2-wk aortic-banded rats (4). However, an increase in the local O${}_2^{-}$ production may lead to a decrease in NO availability by the chemical reaction of NO with O${}_2^{-}$, suggesting that the formation of peroxynitrite may contribute to the development of vascular remodeling seen in this model. Third, the mevalonate pathway may not participate in the effect of fluvastatin. When the inhibition of active oxygen species production was chemically observed with the xanthine and Fenton system, fluvastatin (10 μM) showed scavenging activity for O${}_2^{-}$ and hydroxyl radical (40). Furthermore, Kanno et al. (10) reported that pravastatin (0.5 mM) suppressed O${}_2^{-}$ production in neutrophils stimulated by chemokines through inhibition of tyrosine phosphorylation and that the inhibition was not reversed by mevalonic acid. Thus it is possible that the effect of fluvastatin at a high dose may be independent of inhibition of the mevalonate pathway. Further studies are needed to prove such possibilities.

We observed that ICAM-1 expression was associated both temporally and spatially with O${}_2^{-}$ expression in this model. Oxygen radical species may act as signal transduction messengers for several important transcription factors, such as NF-κB (13). NF-κB is the most important transcription regulatory element in the ICAM-1 promotor system (36). In addition, antioxidants inhibit the expression of ICAM-1 in vitro and in vivo (2,22). Thus it is likely that the increase in vascular O${}_2^{-}$ production greatly upregulated ICAM-1 expression through activation of the redox-sensitive transcriptional factor in this model.

Upregulation of ICAM-1 expression is known to cause inflammatory infiltration into the lesions. Infiltration by inflammatory cells, mainly macrophages, has already been observed in perivascular areas and in the intima in different models of hypertensive rats (6,12). This inflammatory infiltration could lead to fibrosis via the production of profibrotic cytokines such as transforming growth factor-β (8). Thus we examined the effect of fluvastatin on ICAM-1 expression and found that fluvastatin significantly decreased the ICAM-1 expression as well as the development of perivascular fibrosis. Therefore, we interpret these findings to suggest that fluvastatin exerted an improvement in the structural changes at least by decreasing ICAM-1 expression and activity in this model.

It certainly remains to be determined whether the present findings seen in the aorta-banded rats can indeed be extrapolated to the situation of patients with hypertension. However, this effective dose of fluvastatin is much closer to the clinically relevant dose; the lower dose of fluvastatin (∼3 mg/kg) used was three to six times higher than the expected clinical dose (0.5–1 mg/kg). Thus it is possible that beneficial effects separate from its cholesterol-lowering action may be observed in patients receiving fluvastatin.

In conclusion, our present observations suggest that the beneficial effects of fluvastatin could involve reduction of both oxidative stress and cell adhesion molecule expression and an increase in eNOS expression independent of its lipid-lowering effect. Thus these biological effects of fluvastatin as well as its lipid-lowering effect may reduce the risk of atherosclerosis progression in the vasculature. The antiatherosclerotic effects of fluvastatin could also be explained by this mechanism.

FOOTNOTES

• Address for reprint requests and other correspondence: M. Katoh, Discovery Research Laboratories, Tanabe Seiyaku, , 2-2-50, Kawagishi, Toda-shi, Saitama 335-8505, Japan (E-mail:[email protected]co.jp).

• The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.