Fasudil, a Rho-kinase inhibitor, reverses L-NAME exacerbated severe nephrosclerosis in spontaneously hypertensive rats

Background In this study, we tested the hypothesis that long-term Rho-kinase inhibition would reverse nitro-L- arginine methyl ester-exacerbated nephrosclerosis in spontaneously hypertensive rats and attempted to elucidate the mechanism involved.

Methods Five groups (each n U 8) were studied: untreated spontaneously hypertensive rats; nitro-L-arginine methyl ester (50 mg/l in drinking water, for 3 weeks)-treated spontaneously hypertensive rats; nitro-L-arginine methyl ester with fasudil (10 mg/kg/day)-treated spontaneously hypertensive rats; nitro-L-arginine methyl ester for 3 weeks followed by fasudil for 3 weeks-treated spontaneously hypertensive rats (same doses), and nitro-L-arginine methyl ester for 3 weeks followed by untreated for 3 weeks. We examined renal function, blood pressure, histological features, oxidative stress markers, and mRNA expression in the renal cortex.

Results Nitro-L-arginine methyl ester-treated spontaneously hypertensive rats had higher blood pressure, proteinuria, and serum creatinine and lower creatinine clearance, urinary NO3/NO2 ratio, and urinary cGMP excretion compared with control spontaneously hypertensive rats (all Ps < 0.05). Nitro-L-arginine methyl ester-treated spontaneously hypertensive rats also had increased free radical metabolites and abnormal morphological findings with increased nicotinamide adenine dinucleotide phosphate oxidase activity, phosphorylation of myosin phosphatase targeting subunit- 1, and mRNA expression of RhoA, RhoB, RhoC, collagen I and III, transforming growth factor-b, nicotinamide adenine dinucleotide phosphate subunit, endothelial nitric oxide synthase, plasminogen activator inhibitor, and intercellular adhesion molecule-1 in the renal cortex compared with control spontaneously hypertensive rats. Long-term co- treatment with fasudil slightly improved these indices, but most of them were not statistically significant. Late fasudil treatment significantly improved kidney function, morphological changes, and alterations of mRNA expression in the renal cortex, although late untreated controls did not show any improvement.

Conclusion These results suggest that Rho-kinase inhibition partly reverses hypertensive glomerulosclerosis. The renoprotective effect of the Rho-kinase inhibitor may have multiple mechanisms including inhibition of extracellular matrix production, oxidative stress, adhesion molecule production, and antifibrinolysis. J Hypertens 26:1837 –1848 Q 2008 Wolters Kluwer Health | Lippincott Williams & Wilkins.

Keywords: endothelial nitric oxide synthase, fasudil, hypertension, nephrosclerosis, nitric oxide, Rho-kinase

Introduction

Rho, a small GTP binding protein, is known to function as a molecular switch in various cellular functions, including formation of stress fibers and focal adhesions, regulation of calcium ion sensitivity in smooth muscle cells, regulation of cytokines following nuclear division, and regulation of G1 to S phase in cell cycle progression [1,2]. Among various Rho effectors, the cellular function and signal transduction of Rho-kinase have been extensively studied. After the discovery of a specific inhibitor of Rho-kinase [3], studies regarding the role of the Rho/Rho-kinase pathway in vitro and in vivo have greatly advanced. As Rho modulates the Ca2+ sensitization of vascular smooth muscle cells by inhibiting myosin phosphatase activity, the effects of Rho/Rho-kinase on the tone of blood vessels and the role of Rho in the pathogenesis of arteriosclerosis have been intensively investigated [4].

In addition, recent advances in the study of Rho/Rho- kinase have revealed that the Rho/Rho-kinase pathway is involved in the pathogenesis of various cardiovascular diseases, such as hypertension, angina pectoris, myo- cardial infarction, and pulmonary hypertension [5]. However, few studies have investigated the role of Rho/Rho-kinase in renal disease. Recent studies revealed that specific Rho-kinase inhibitors, Y-27632 or fasudil, significantly attenuate the tubulointerstitial fibrosis in the kidney induced by unilateral ureteral obstruction [6,7]. In addition, we and other investigators have recently shown that the Rho-kinase inhibitor fasudil attenuates glomer- ulosclerosis in salt-induced hypertensive rats [8], subto- tally nephrectomized spontaneously hypertensive rats (SHR) [9], severely hypertensive rats [10], and aldoster- one-induced glomerular injury in rats [11]. More recently, we have shown that a Rho-kinase inhibitor has renopro- tective effects mediated partly via the upregulation of the endothelial nitric oxide synthase (eNOS)/nitric oxide pathway in malignant hypertensive rats [12]. As the eNOS/nitric oxide pathway is known to play an important role in the development of renal disease [13], we hypoth- esized that the renoprotective effect of the Rho-kinase inhibitor may be attenuated in the absence of the eNOS/ nitric oxide pathway. To test this hypothesis, we examined showed that the mechanism of the renoprotective effect of Rho-kinase inhibitor is mediated via many pathways including inhibition of extracellular matrix gene expres- sion, monocyte/macrophage infiltration, oxidative stress, and upregulation of eNOS gene expression. In the present study, we therefore measured the mRNA expression level of transforming growth factor (TGF-b), collagen I, collagen III, plasminogen activator inhibitor-1 (PAI-1), intercellular adhesion molecule-1 (ICAM-1), eNOS, nico- tinamide adenine dinucleotide phosphate (NADPH) sub- unit and Rho family, and NADPH oxidase activity and Rho-kinase activity to elucidate the mechanism involved. We also measured localization of superoxide generation by dihydroethidium (DHE) staining and migration of macrophage/monocytes by immunohistochemistry.

Methods

The current study was approved by our institutional animal care committee, and all of the procedures were in accordance with the National Institutes of Health Guide for the Care and Use of Laboratory Animals.

Experimental animals and design

Seventeen-week-old male SHR (Clea Japan, Tokyo, Japan, n ¼ 40) were studied. Untreated SHR (n ¼ 8) were used as a control model. L-NAME SHR (n ¼ 32) were used as a severe hypertensive rat model in this study, because this model has extensive end organ damage, including severe nephrosclerosis [14,15]. L-NAME (50 mg/l/day) was dissolved in drinking water and admi- nistered for 3 weeks. L-NAME SHR were randomly divided into the following four groups: L-NAME mono- treatment group (n ¼ 8), low-dose fasudil co-treatment group (10 mg/kg/day, n ¼ 8), late fasudil treatment group (10 mg/kg/day, n ¼ 8), and late control group. In the late fasudil treatment group, SHR were treated with L-NAME for 3 weeks, followed by a subsequent 3 weeks of treatment with fasudil (10 mg/kg/day). In the late control group, SHR were treated with L-NAME for 3 weeks, followed by a subsequent 3 weeks of no treat- ment (Fig. 1). Fasudil was administered in the drinking water. After oral administration, fasudil is metabolized to hydroxyfasudil, a major active metabolite of fasudil that specifically inhibits Rho-kinase [16]. In the preliminary study, we determined the dose of fasudil (10 mg/kg/day) that is sufficient to improve renal function, but does not change blood pressure (BP) in this model.

Urine collection

After rats were acclimatized to the metabolism cages prior to collection for 2 days, 24-h urine samples were collected from rats in metabolic cages at 20 weeks in control SHR, L-NAME treated SHR, and L-NAME-fasudil co-treated SHR. In another two groups, the late fasudil treatment and late control groups, 24-h urine samples were collected at 23 weeks (Fig. 1). Urinary protein and creatinine (Cre) levels were measured as reported previously [17]. Cre clearance (CCr) was calculated using standard formulae.

Urinary nitric oxides and cyclic GMP measurement The amounts of urinary nitric oxides (NO2 and NO3) and cyclic GMP were measured as reported previously, and the NO3/NO2 ratio was calculated [15]. Blood pressure measurement and blood sampling Systolic BP (SBP) was measured every week by the tail- cuff method, as reported previously [18]. Blood was obtained from the carotid artery, and the plasma was stored at —808C until the assay. Immediately after the heart was arrested by the injection of 2 mmol of KCl, the right kidney was removed, weighed, and cut in half. One half kidney was postfixed in 10% neutral-buffered for- malin, and other half was embedded in optimal cutting temperature (OCT) compound for measurement of superoxide (O2—) anion generation by DHE. The left kidney was removed, weighed, separated into the cortex and medulla, frozen in liquid nitrogen, and stored at —808C until the reverse transcriptase-PCR, Western blot, and NADPH oxidase analyses.

In-situ detection of renal superoxide anion

Superoxide (O2—) anion generation in renal tissue was evaluated by using the fluorescent dye DHE as pre- viously reported [19]. The renal tissue embedded in OCT was quickly frozen and was cut by cryostat (5 mm, cross section). Sections were incubated with DHE 1 mmol/l in a light-protected humidified chamber at 378C for 5 min. After washing with PBS, images were acquired with identical acquisition parameters using a fluorescence microscope (Olympus AX80, Tokyo, Japan). DHE specifically reacts with intracellular O2— and is converted to the red fluorescent compound ethidium, which then binds irreversibly to double-stranded DNA and appears as punctuate nuclear staining. Fluorescence intensity was quantified using Mac SCOPE software (Mitani Shoji Co. Ltd., Fukui, Japan) and recorded as average gray scale intensity.

Plasma oxidative stress marker measurement Biological antioxidant potential (BAP-1) and reactive oxygen metabolites (d-ROM) were measured using a commercially available kit (H&D, s.r.l., Parma, Italy). Nicotinamide adenine dinucleotide phosphate oxidase activity NADPH oxidase-derived superoxide anion generation was measured using lucigenin-enhanced chemilumines- cence, as described previously [19,20]. In brief, renal tissues were placed in chilled phosphate-buffered saline containing protease inhibitor and homogenized on ice. Protein concentration of homogenates was measured using the Bradford protein assay kit (Bio-Rad Laboratories, Hercules, California, USA). After centrifuging, the super- natant was transferred into a test tube containing lucigenin [final concentration 5 mmol/l in Krebs-(4-(2-hydroxy- ethyl)-1-piperazineethanesulfonic acid (HEPES) buffer].

Chemiluminescence was then recorded every 15 s for 10 min with a luminescence reader (BLR-301; Aloka, Tokyo, Japan), and the readings in the last 5 min were averaged. After measurement of background lucigenin chemiluminescence, NADPH was added to a final con- centration of 100 mmol/l. Thereafter, chemiluminescence was recorded for another 10 min, and the readings in the last 5 min were averaged. The differences between the values obtained before and after adding the NADPH were calculated.

Renal morphology

The right kidney was excised and immersed in neutral- ized formalin as described above for histological exam- ination. Histological examination was performed by two observers in a blinded fashion. For semiquantitative evaluation, the glomerular injury score (GIS) and arter- iolar injury score (AIS) were determined as reported previously [14]. The area of fibrotic lesions in the cortical interstitium (fibrosis area) was determined on sections stained with Masson’s trichrome method to stain the collagen fibers (stained blue), using a computer-aided manipulator program as described previously [14].

Immunohistochemical analysis

Immunohistochemical analysis using an antibody against ED-1 was performed as reported previously [12].Quantification of mRNA by reverse transcriptase-PCR All of the procedures used for mRNA extraction, cDNA synthesis, and PCR were described in detail in previous studies [21]. The numbers of PCR cycles for the 14 genes examined were as follows: TGF-b, 28; collagen I, 29; collagen III, 28; eNOS, 30; PAI-1, 28; ICAM-1, 31; RhoA, 25; RhoB, 27; RhoC, 26; p22phox, 33; gp91phox, 31; p47phox, 33; p40phox, 33; and glyceraldehyde 3- phosphate dehydrogenase (GAPDH), 22. Using each of these numbers of PCR cycles, reverse transcriptase-PCR was performed in the linear range of the reaction. Quanti- fication of each species of mRNA was performed as reported previously.

Western blot analysis

We used immunoblotting with antibodies against phospho- myosin phosphatase targeting subunit-1 (MYPT1) (at Thr-696; 1 : 2000 dilution; Upstate Biochemistry, Lake Placid, New York, USA) to evaluate Rho-kinase activity [6,11]. To check for equal loading, membranes were reprobed with an antibody against b-actin (1 : 10 000 dilution; Sigma Chemical, St. Louis, Missouri, USA). Data are expressed as the relative differences between untreated SHR and other groups after normalization to b-actin expression.

Statistical analysis

All data are expressed as mean SD. Statistical compari- sons among the five groups were carried out by analysis of variance (ANOVA). If appropriate, the data were com- pared with the use of the Bonferroni post-hoc test for multiple comparisons. P less than 0.05 was considered to indicate statistical significance. Statistical analysis was performed using STATVIEW, version 5 (Abacus Con- cepts, Berkeley, California, USA), or STATA, version 8 (STATA Corp., College Station, Texas, USA).

Results

Physiological profiles

The changes in systolic BP as measured by the tail-cuff method in the five groups are presented in Fig. 1. The systolic BPs in 17-week-old SHR of the five groups were similar. L-NAME treatment significantly increased the systolic BP in 18-week-old SHR, and the systolic BPin this group graduallyincreased thereafter and reached a maximum at 20 weeks. Long-term low-dose fasudil co- treatment did not affect the systolic BP of L-NAME- treated SHR. After 3 weeks of L-NAME treatment, the systolic BP in the late control group decreased, although it still remained higher than that in untreated SHR at 20 weeks. Late fasudil treatment did not change BP, either.

The physiological profiles of the five experimental groups are summarized in Table 1. Body weight was significantly lower in L-NAME SHR than in untreated SHR. Long- term low-dose fasudil co-treatment did not change the body weight of L-NAME-treated SHR. Late fasudil treatment significantly ameliorated body weight loss. L-NAME SHR and the low-dose fasudil co-treatment group did not show significantly different kidney weight/ body weight or left ventricular weight (LVW)/body weight compared with those in untreated SHRs. However, late fasudil treatment significantly decreased kidney weight/body weight and significantly increased LVW/body weight.

Renal parameters

Renal parameters are shown in Table 2. The CCr was significantly decreased in L-NAME SHR, while serum urea nitrogen, serum Cre, and urinary protein excretion were significantly increased as compared with those in untreated SHR. Late fasudil treatment significantly re- duced serum urea nitrogen, serum Cre, and urinary protein excretion and increased CCr in L-NAME SHR. Low-dose fasudil co-treatment did not change these parameters, except for blood urea nitrogen (BUN), significantly.

Oxidative stress and biochemical markers

Oxidative stress and biochemical marker levels are also shown in Table 2. L-NAME SHR had significantly decreased BAP-1 and increased d-ROM compared with untreated SHR. There were no significant differences in these parameters between the late fasudil treatment group and the untreated SHR group, whereas there were significant differences in these parameters between the untreated SHR group and L-NAME and fasudil-treated SHR or late control groups. L-NAME SHR showed significantly decreased urinary NO3 and cGMP excretion and NO3/NO2 ratio compared with untreated SHR (Table 2). Late fasudil treatment significantly increased these parameters. There was no difference in urinary NO2 excretion among the five groups (Table 2).

Renal pathological findings

The morphological appearance of the glomeruli was almost normal in SHR (Fig. 2a, f, k). Histological exam- ination of the kidneys in L-NAME SHR revealed mild- to-moderate glomerulosclerosis, marked thickening of the vascular wall, perivascular fibrosis with marked inflammatory cell infiltration, and tubulointerstitial fibro- sis with inflammatory cell infiltration and luminal protein cast formation (Fig. 2b, g, and l). Late fasudil treatment attenuated these glomerular, vascular, and interstitial changes (Fig. 2d, i, and n). The GIS was markedly higher in L-NAME SHR than in SHR. Late fasudil treatment significantly attenuated these changes (Fig. 2p). The AIS and fibrosis area were also markedly higher in L-NAME SHR than in SHR. Late fasudil treatment significantly attenuated these changes (Fig. 2q, r). However, there were still significant differences in these variables between SHR and late fasudil-treated L-NAME SHR (Fig. 2n and o). Fasudil co-treatment did not improve these changes significantly.

Immunohistochemical analysis

ED-1-positive cells (monocytes/macrophages) in the cortical interstitium and in the glomerulus are shown in Fig. 3a– e. The number of ED-1-positive cells in the interstitium was significantly increased in L-NAME SHR (Fig. 3b). Late fasudil treatment significantly decreased the number of ED-1-positive cells (Fig. 3d). However, there were still significant differences between SHR and late fasudil-treated L-NAME SHR (Fig. 3k). Fasudil co-treatment did not improve the number of ED- 1-positive cells significantly.

Dihydroethidium staining

Intracellular O2— levels in the renal cortex were measured using DHE and fluorescence microscopy. Representative results of DHE staining in five groups are shown in Fig. 3f– j. The average of fluorescence intensity values is shown in Fig. 3l. L-NAME SHR had higher DHE staining than control SHR. Late fasudil treatment sig- nificantly decreased DHE staining compared with L- NAME SHR, whereas L-NAME and fasudil co-treatment or late control did not change DHE staining significantly.

Gene expression levels of transforming growth factor-b, collagen I, III, intercellular adhesion molecule-1, plasminogen activator inhibitor-1, and endothelial nitric oxide synthase in the renal cortex

The mRNA levels of TGF-b, collagen I, collagen III, ICAM-1, PAI-1, and eNOS relative to the GAPDH mRNA level in the renal cortex were all increased in L-NAME SHR as compared with control SHR (Fig. 4a– f). Late fasudil treatment reduced the mRNA levels of TGF-b, collagen I, collagen III, ICAM-1, PAI-1, and eNOS compared with L-NAME SHR (Fig. 4a–f). In contrast, L-NAME and fasudil-co-treated SHR or late control SHR did not change the mRNA levels of TGF-b, collagen I, collagen III, ICAM-1, or PAI-1 significantly (Fig. 4a– e). Late control SHR had lower mRNA levels of eNOS compared with L-NAME SHR (Fig. 4f).

Gene expression levels of nicotinamide adenine dinucleotide phosphate subunits and nicotinamide adenine dinucleotide phosphate oxidase activity in the renal cortex

As shown in Fig. 5a– d, the mRNA expression levels of p22phox, p40phox, p47phox, and gp91phox in the renal cortex were higher in L-NAME SHR than in SHR. Late fasudil treatment significantly decreased the mRNA expression levels of p22phox, p40phox, p47phox, and gp91phox. Fasudil co-treatment only decreased the increased mRNA level of p22phox compared with L- NAME SHR. NADPH oxidase activity in renal cortical tissue was also increased in L-NAME SHR compared with control SHR (Fig. 5e). Late fasudil and L-NAME and fasudil co-treatment significantly decreased it, how- ever; there was still a difference between these groups and control SHR group.

Gene expression levels of RhoA, RhoB, and RhoC in the renal cortex

The mRNA levels of RhoA, RhoB, and RhoC relative to the GAPDH mRNA level in the renal cortex were significantly increased in untreated L-NAME SHR as compared with SHR (Fig. 6a– c). Late fasudil treatment significantly decreased the mRNA levels of RhoA and RhoC relative to the GAPDH mRNA level. However, there was no significant difference in the mRNA levels of RhoB among the four groups (Fig. 6b). Western blot analysis revealed that the level of MYPT1 phosphoryl- ation in renal cortical tissues was increased in L-NAME SHR compared with control SHR, indicating that Rho- kinase was activated in kidneys of L-NAME SHR. Late fasudil treatment significantly decreased the level of MYPT1 phosphorylation compared with L-NAME SHR (Fig. 6d).

Discussion

The current study showed that 3 weeks of L-NAME administration to SHR induced severe hypertension and renal injury characterized by marked proteinuria, elevation of Cre and BUN, reduced CCr, glomerulo- sclerosis, interstitial fibrosis, and renal vascular arterio- sclerosis, which is in agreement with previous studies [14,15]. Our study also provided evidence for the first time that renal injury in L-NAME SHR is associated with increases in the gene expression of several Rho-kinase system was inhibited by L-NAME. These results suggest that Rho-kinase inhibition reverses hypertensive nephro- sclerosis in rats and that the renoprotective effects of Rho-kinase inhibitor may be mediated by multiple mech- anisms, including the inhibition of extracellular matrix production, oxidative stress, adhesion molecule pro- duction, and antifibrinolysis.

Renal fibrosis is characterized by excessive synthesis and accumulation of extracellular matrix components, includ- ing collagen. Accumulating evidence indicates that TGF- b plays an important role in the pathogenesis of glomer- ulosclerosis, arteriolar sclerosis, and tubulointerstitial fibrosis [6,8,15]. We previously demonstrated that the level of TGF-b mRNA in renal tissue was significantly increased in L-NAME SHR compared with control SHR [15], which is consistent with the current results. These data suggest a possible contribution of the TGF-b/ collagen pathway in nephrosclerosis in L-NAME SHR. Of interest, recent studies revealed that Rho-kinase plays a role in mediating renal fibrosis through a TGF- b-dependent mechanism. Nagatoya et al. [6] reported that another specific Rho-kinase inhibitor, Y-27632, suppressed interstitial fibrosis and the augmentation of TGF-b and collagen gene expression in the mouse kidney with uni- lateral ureteral obstruction. In addition, we also reported that the TGF-b/collagen pathway was activated in the kidney in salt-induced hypertensive rats and that long- term administration of fasudil attenuated the augmenta- tion of the TGF-b/collagen pathway and improved the renal injury in this model [8]. Furthermore, recent in-vitro studies demonstrated that Rho-kinase is an essential factor for connective tissue growth factor accumulation in human renal fibroblasts [22] and for mechanical stretch- induced TGF-b synthesis in hepatic stellate cells [23]. In the current study, late fasudil treatment significantly attenuated the nephrosclerosis and activation of TGF- b, collagen I, and collagen III gene expression in the renal cortex in L-NAME SHR. Thus, one of the possible mechanisms of the renoprotective effect of fasudil might be suppression of TGF-b mRNA expression and sub- sequent collagen mRNA expression.

In addition, increased mRNA expression of PAI-1 in the renal cortex was also observed in L-NAME SHR. Plasmin and matrix metalloproteinases (MMPs) play
an important role in extracellular matrix degradation, and PAI-1 plays a critical role in extracellular matrix accumulation by interfering with the generation of plasmin and the activation of MMP by blocking plasmin [24]. The PAI-1 gene is known to be activated by TGF-b [25]. In the present study, the mRNA expres- sion of PAI-1 was increased in L-NAME SHR, and late fasudil treatment significantly decreased its expression. These results suggest that PAI-1 is involved in the pathogenesis of hypertensive nephrosclerosis and provide evidence that fasudil decreases the PAI-1 gene expression level partly via a TGF-b cascade. Collectively, one of the possible mechanisms of the renoprotective effect of fasudil may be the suppression of TGF-b and the subsequent collagen and PAI-1 pathways.

Cellular reactive oxygen species are increased in hyper- tension and have been shown to contribute to the patho- genesis of chronic vascular complications, including nephrosclerosis [26]. Oxidative stress is known to enhance the TGF-b/collagen and TGF-b/PAI-1 cascades [27,28]. In the present study, the expression of NADPH oxidase subunits, such as p47phox, p40phox, p22phox, and gp91phox, and NADPH oxidase activity were increased in the renal cortex in L-NAME SHR compared with SHR. A recent study showed that p47phox and p67phox expression was increased in the kidneys of SHR, with prominent expression in the glomerulus, renal vasculature, the thin limb of the loop of Henle, macula densa cells, distal convoluted tubules, and collecting
ducts [29]. The gene expression of NADPH oxidase subunits (p40phox, p47phox, and p67phox) was also increased in the renal cortex in malignant hypertensive rats. In addition, angiotensin II stimulates NADPH oxidase activity and the expression of p22phox and p67phox in vascular smooth muscle cells, and its acti- vation is mediated via the Rho/Rho-kinase pathway [16]. Furthermore, hypertension causes oxidative stress in the blood vessels and produces superoxide by NADPH oxidase in endothelial and vascular smooth muscle cells [29]. Plasma renin activity is reported to be markedly increased in L-NAME SHR [30]. Thus, increased BP per se or the activated renin– angiotensin system or both may stimulate the expression of NADPH oxidase sub- units in L-NAME SHR. Fasudil may attenuate oxidative stress by inhibiting intracellular signaling cascades, inde- pendently of blood-pressure-lowering mechanisms.

In the present study, we measured the serum concen- tration of hydroperoxides, a noninvasive marker of oxidative stress. The activation of NADPH oxidase and the production of radicals play an important role in the production of hydroperoxides [31]. In the present study, we found that the serum concentration of hydro- peroxides was increased in L-NAME SHR. We also measured the biological antioxidant potential and found that this marker was reduced in L-NAME SHR. After late fasudil treatment, there were no differences in these parameters between control SHR and late fasudil-treated SHR. Thus, systemic oxidative stress is also increased and antioxidant potential is decreased in L-NAME SHR, and long-term fasudil treatment may improve them.

Macrophage/monocyte infiltration is one of the key mechanisms of the progression of nephrosclerosis [32]. The current study showed that the infiltration of ED-1- positive cells was increased around glomeruli and vascu- lature and tubulointerstitial spaces in the kidneys of L-NAME SHR, which is consistent with previous studies [14,15], and macrophage/monocyte infiltration was significantly decreased by long-term fasudil treatment. Rho/Rho-kinase is involved in many aspects of cell moti- lity, ranging from smooth muscle contraction to cell migration [33,34]. Previous studies have shown that Rho-kinase inhibitor reduces macrophage/monocyte infil- tration and improves interstitial fibrosis in different renal disease models [6–12]. In addition, an in-vitro study showed that neutrophil chemotaxis is significantly inhib- ited by fasudil [35]. These results suggest the importance of Rho-kinase in macrophage/monocyte infiltration. Another possibility is that Rho-kinase inhibitor attenuated macrophage/monocyte infiltration by reducing ICAM-1 gene expression. Oxidative stress leads to activation of transcription factors such as nuclear factor (NF)-kB, which in turn activate genes that trigger inflammation, including ICAM-1 [36]. As fasudil inhibits the generation of reactive oxygen species [16], fasudil might decrease the gene expression of ICAM-1 via the inhibition of ROS. In fact, in the present study, long-term fasudil treatment signifi- cantly reduced the levels of mRNA expression of ICAM-1 and NADPH subunit, NADPH oxidase, DHE staining, and the level of d-ROM in association with the decrease in macrophage/monocyte infiltration. Thus, fasudil might inhibit macrophage/monocyte infiltration via direct and indirect mechanisms.

In the present study, co-administration of fasudil and L-NAME did not ameliorate renal injury; however, late fasudil treatment partly reversed the nephrosclerosis induced by L-NAME. A previous study showed that inhibition of Rho-kinase upregulates eNOS via activation of the PI3-kinase/Akt pathway [37]. The eNOS/nitric oxide system plays an important role in the development of renal impairment [38], and therefore one of the possible mechanisms of the renoprotetive effect of Rho-kinase inhibitor is upregulation of the eNOS/nitric oxide system. Indeed, we very recently showed that fasudil upregulated eNOS and ameliorated renal injury in malignant hypertensive rats [12]. In the nephrosclero- sis in L-NAME SHR, infiltration of inflammatory cells plays a central role in the pathogenesis of the renal injury. Infiltration of inflammatory cells also plays a pivotal role in the pathogenesis in ischemia/reperfusion models. Wolfrum et al. [37] recently reported that administration of fasudil decreased leukocyte recruitment and adhesion to the mesenteric endothelium after ischemia/reperfu- sion injury in wild-type mice. Similarly, treatment with fasudil decreased the myocardial infarct size in rats subjected to transient coronary artery occlusion [37].

However, these beneficial effects of Rho-kinase inhibitor were completely abolished in eNOS (—/—) mice [37]. These results suggest that the renoprotective effect of fasudil is mainly mediated by inhibition of the infiltration of inflammatory cells by upregulation of the eNOS/nitric oxide system, and that the renoprotective effect of Rho- kinase inhibitor may be mediated by multiple mechan- isms that occur downstream of the eNOS/nitric oxide/ cGMP system.

In conclusion, our study showed that activation of the Rho/Rho-kinase pathway is related to the pathophysiol- ogy of L-NAME SHR, and that long-term fasudil treat- ment partly reversed the nephrosclerosis in these animals but did not prevent it. The mechanism of the renopro- tective effect of fasudil may be the result of a combi- nation of factors, including a reduction in the TGF-b/ collagen cascade, improved control of inflammation, a reduction in oxidative stress, and upregulation of eNOS gene expression,β-Nicotinamide and that multiple mechanisms of the renoprotective effect of Rho-kinase inhibitor may occur downstream of eNOS.