- Email: [email protected]
Beraprost sodium, prostacyclin analogue, attenuates glomerular hyperfiltration and glomerular macrophage infiltration by modulating ecNOS expression in diabetic rats
Diabetes Research and Clinical Practice 57 (2002) 149– 161 www.elsevier.com/locate/diabres
Beraprost sodium, prostacyclin analogue, attenuates glomerular hyperfiltration and glomerular macrophage infiltration by modulating ecNOS expression in diabetic rats Tetsuji Yamashita, Kenichi Shikata *, Mitsuhiro Matsuda, Shinichi Okada, Daisuke Ogawa, Hikaru Sugimoto, Jun Wada, Hirofumi Makino Department of Medicine III, Okayama Uni6ersity Medical School, 2 -5 -1, Shikata-cho, Okayama 700 -8558, Japan Received 9 January 2002; received in revised form 4 March 2002; accepted 18 March 2002
Abstract Stable prostacyclin analogue, beraprost sodium (BPS) has recently been reported to attenuate glomerular hyperfiltration in diabetic rats, however, the mechanism has been still unknown. We previously reported that overexpression of endothelial cell nitric oxide synthase (ecNOS) in afferent arterioles and glomeruli induce inappropriate dilatation of afferent arterioles and glomerular hyperfiltration through overproduction of nitric oxide in early stage of diabetic nephropathy. In this study, we tested the hypothesis that BPS ameliorates glomerular hyperfiltration through modulating ecNOS expression in diabetic nephropathy. Furthermore, we examined the effects of BPS on the expression of intercellular adhesion molecule-1 (ICAM-1) and macrophage infiltration in diabetic glomeruli, because glomerular hyperfiltration induces the expression of ICAM-1 resulting in macrophage infiltration. Male Sprague– Dawley (SD) rats were administered continuously with BPS for 4 weeks after induction of diabetes by streptozotocin. In diabetic rats, the diameters of afferent arterioles, glomerular volume, creatinine clearance and urinary excretion of albumin and NO2/NO3 were increased as compared with non-diabetic control rats. Treatment with BPS improved these changes. The expression of ecNOS was increased in afferent arterioles and glomeruli in diabetic rats and suppressed by BPS. Prostacyclin receptor was expressed along afferent arterioles. Our results suggest that BPS attenuates glomerular hyperfiltration by modulating ecNOS expression in early stage of diabetic nephropathy. Moreover, BPS may inhibit ICAM-1-dependent infiltration of macrophages in diabetic glomeruli. © 2002 Elsevier Science Ireland Ltd. All rights reserved. Keywords: Diabetic nephropathy; Hyperfiltration; Prostacyclin; Nitric oxide; ICAM-1; Macrophage
1. Introduction
* Corresponding author. Tel.: +81-86-235-7234; fax: + 8186-222-5214 E-mail address: [email protected] (K. Shikata).
Dilatation of afferent arterioles is one of the characteristic changes in the early stage of diabetic nephropathy. Inappropriate dilatation of afferent arterioles is believed to induce glomerular
0168-8227/02/$ - see front matter © 2002 Elsevier Science Ireland Ltd. All rights reserved. PII: S 0 1 6 8 - 8 2 2 7 ( 0 2 ) 0 0 0 5 4 - 2
150
T. Yamashita et al. / Diabetes Research and Clinical Practice 57 (2002) 149–161
hyperfiltration and hypertrophy followed by thickening of the glomerular basement membrane and accumulation of mesangial matrix [1,2]. We have recently reported that expression of endothelial cell nitric oxide synthase (ecNOS) isoform is up-regulated in afferent arterioles and glomeruli in the early stage of streptozotocin (STZ)-induced diabetic rat [3]. Treatment with L-NG-nitroarginine methyl ester HCL (LNAME), an inhibitor of NOS, ameliorated glomerular enlargement and glomerular hyperfiltration suggesting that overproduction of nitric oxide in afferent arterioles and glomeruli induce dilatation of afferent arterioles, glomerular hypertrophy and glomerular hyperfiltration [3]. Our results have been further supported by Veelken et al. [4]. We have also reported that overproduction of NO is involved in the mechanism of glomerular hyperfiltlation in patients with type 2 diabetes [5]. Recently, it has been reported that prostacyclin analogue, beraprost sodium (BPS), improved glomerular hyperfiltration and reduced urinary albumin excretion in STZ-induced diabetic rat [6], although the mechanism is still unknown. BPS is known to have a relaxative action on vascular smooth muscle [7], and an inhibitory action against platelet aggregation [8,9]. It was also reported that prostaglandins interact with the process of nitric oxide synthesis [10– 13]. In the present study, we tested the hypothesis that BPS ameliorates glomerular hyperfiltration through modulating ecNOS expression in the early stage of diabetic nephropathy using STZinduced diabetic rats. It is well known that macrophages infiltrate into glomeruli and interstitium in diabetic kidney [14]. Macrophage is considered to play an important role in the progression of diabetic glomerulosclerosis [14– 16]. Intercellular adhesion molecule-1 (ICAM-1) is up-regulated in diabetic glomeruli and mediates glomerular macrophage infiltration [15– 17]. We previously reported that glomerular hyperfiltration is one of the major causes of up-regulation of ICAM-1 in diabetic glomeruli [17]. In the current study, we have also examined the effects of BPS on the expres-
sion of ICAM-1 and macrophage infiltration in diabetic glomeruli.
2. Materials and methods
2.1. Animals Male Sprague–Dawley (SD) rats weighing about 120 g (4 weeks of age) were purchased from Charles River Japan (Yokohama, Japan). These rats received a standard chow and water diet.
2.2. Experimental protocol Diabetes was induced by the intravenous injection of 65mg/kg STZ (Sigma, St. Louis, MO) in citrate buffer, pH 4.5 as previously described [18]. Non-diabetic control rats were age-and sexmatched with the experimental rats and received citrate buffer alone. From 1 day after injection of STZ, one group was received continuous intraperitoneal administration of 30 mg per rat per day of BPS by an osmotic pump (ALZA Co., Palo Alto, CA, USA) (STZ-BPS group; n= 5). Another diabetic group (STZ group; n= 5) and non-diabetic control group (ND group; n= 5) were received saline by osmotic pump. BPS (TRK-100) was kindly provided by Toray Industries, Inc., Tokyo, Japan. The levels of fasting blood glucose in all diabetic rats were kept between 400 and 500 mg/dl by insulin treatment, using nearly 24 h-acting insulin, Humalin N (Shionogi, Osaka, Japan). Dose-dependent anti-platelet effect of BPS is observed at the dose of 3–300 mg/kg per day in rats [19]. Our previous studies indicated that 250–300 mg/kg per day of BPS does not affect systemic blood pressure in rats [20]. Therefore, we used 30 mg per rat per day (250 mg/kg per day) of BPS in this study. This study was carried out in accordance with the Guidelines for Animal Experiments at Okayama University Medical School, Japanese Government Animal Protection and Management Law (No. 105) and Japanese Government Notification on Feeding and Safekeeping of Animals (No. 6).
T. Yamashita et al. / Diabetes Research and Clinical Practice 57 (2002) 149–161
2.3. Metabolic data On 28 days after induction of diabetes, all rats were transferred to metabolic cages for the collection of 24 h urine samples, and blood samples were taken from the tail vein. They then had free access to water only ad libitum. The chemical analysis of creatinine in serum and urine was performed by the standard laboratory methods. Fructosamine as measured by Precimat Fructosamine (Boehringer Mannheim, Tokyo, Japan). Fasting blood glucose was measured by the glucose oxidase method. Systolic blood pressure was measured three times by the tail cuff method. Urine samples were assayed for the stable NO metabolic end products, NO2 and NO3 with a Nitrate/Nitrite Assay Kit (Cayman Chemical, Ann Arbor, MI, USA). The level of urinary albumin was measured using NEPHRAT (Exocell, Inc., Philadelphia, USA), a kit for enzyme-linked immunosorbent assay. Five rats for each group were killed under anesthesia with 5 mg/100 g body weight of sodium pentobarbital at 4 weeks after BPS or saline administration, then both kidneys were removed. A part of kidney was fixed in 95% ethanol with 1% acetic acid at 4 °C overnight [21], then dehydrated and embedded in paraffin. Tissue blocks were stored at 4 °C until use. Other tissues were snap frozen in acetone and stored at − 20 °C.
151
were observed and photographed. The maximum afferent and efferent arteriolar diameters (luminar diameter) were then measured on photographs (× 200). Using the serial sections described above, we selected 30 glomeruli for each kidney specimen. Then the maximum diameter of selected glomeruli in serial sections was measured on photograph (× 200). The diameters were calculated as the mean of the longest and shortest diameters. The glomerular volume was determined from the mean glomerular diameter, d (G), using the formula: 4y(d (G)/2)3/3 [3].
2.5. Expression of ecNOS in afferent and efferent arterioles and glomeruli Double labeling immunofluorescence studies were performed, using mAb against rat ecNOS and mAb against SM2 isoform of myosin as primary antibodies. It is known that expression of SM2 isoform of myosin is restricted in the preglomerular vessels including afferent arterioles, not in efferent arterioles [21]. The ecNOS fluorescence intensities of arteiroles and glomeruli were graded semiquantiatively according to the following scale: ‘— ’, no detectable staining; ‘+ ’, weak staining; ‘2+’, moderate staining; ‘3+’, strong staining. The intensities were graded by two separate investigators in 30
2.4. Measurement of the diameters of afferent and efferent arterioles and glomerular 6olume We cut 4 mm-thick serial sections for the immunohistochemical studies. In each kidney specimen, 30 serial sections were prepared. Immunoperoxidase staining was done, using mouse monoclonal antibody (mAb) against human a-smooth muscle actin (a-SMA) as primary antibody. Interlobular arteries, afferent and efferent arterioles were sequentially identified with a-SMA [3] (Fig. 1). In each kidney specimen, 25 afferent and efferent arterioles in arteriolar longitudinal sections were selected randomly from the superficial nephrons and evaluated. Several serial sections
Fig. 1. (a – c) Serial sections of kidney from a non-diabetic rat stained with anti-a-SMA antibody. Afferent arterioles (af) and efferent arterioles (ef) can be clearly distinguished. ILA, interlobular artery; gl, glomerulus (original magnification, ×200).
152
T. Yamashita et al. / Diabetes Research and Clinical Practice 57 (2002) 149–161
glomeruli and 25 afferent arterioles and 25 efferent arterioles per animal and expressed as the average intensity.
oped in 3,3-diaminobenzidine and hydrogen peroxide. The sections were then counterstained with Mayer’s hematoxylin.
2.6. Expression of ICAM-1 and infiltration of macrophages in glomeruli
2.9. Indirect immunofluorescence study
Indirect immunofluorescence studies were performed, as described below, using mAb against rat ICAM-1 as primary antibody. Immunoperoxidase staining was done, using mouse mAb against rat monocytes/macrophages (ED1) as primary antibody. The estimation of ICMA-1 fluorescence intensities in glomeruli was done semiquantitatively by the same method as described above. The number of ED1 positive cells (monocytes/macrophages) was counted in 30 glomeruli per rat. The average number of positive cells per glomerulus was calculated.
2.7. Localization of prostacyclin receptor (IP receptor) in arterioles and glomeruli Double labeling immunofluorescence studies were performed, using rabbit polyclonal antibody against rat prostacyclin receptor and mAb against human SM2 as primary antibodies.
2.8. Immunoperoxidase staining Immunoperoxidase staining was done by using ABC kit (Vector Lab., Burlingame, CA, USA) as described previously [16]. In brief, non-specific protein binding was blocked by incubating the cryostat sections with 10% bovine serum in Tris– buffered saline for 20 min. Non-specific staining was blocked by 15 min incubation with avidin and then biotin using the avidin– biotin blocking kit (Vector Lab.). Endogenous peroxidase activity was inhibited by incubating the sections in methanol containing 0.3% H2O2 for 20 min. Sections were first incubated for 60 min with primary antibodies at room temperature. Then the sections were incubated with secondary antibodies for 30 min at room temperature. Biotinylated horseradish peroxidase was applied for 30 min at room temperature. Peroxidase activity was devel-
Indirect immunofluorescence studies were performed on 4 mm cryostat sections fixed with acetone that were stained with a primary antibodies at 4 °C overnight and followed by a FITC-labeled goat anti-mouse or rabbit IgG for 30 min at room temperature.
2.10. Double labeling immunofluorescence study Double labeling immunofluorescence studies were performed on 4 mm cryostat sections fixed with acetone that were stained with primary antibodies (anti-rat ecNOS mAb or anti-rat IP receptor antibody) at 4 °C overnight and followed by a FITC-labeled goat anti-mouse or rabbit IgG for 30 min at room temperature. Then the sections were stained with a mouse mAb against SM2 for 30 min at room temperature and rhodamine-labeled goat anti-mouse IgG for 30 min at room temperature.
2.11. Antibodies As primary antibodies, we used mouse mAbs against human a-SMA (DAKO, Kyoto, Japan), human SM2 (Yamasa Co. Ltd., Chiba, Japan), and rat ecNOS (Transduction Laboratory, Kentucky, USA), and rabbit polyclonal antibody against rat prostacyclin receptor (IP receptor) (kindly provided by Toray Industries, Inc., Tokyo, Japan). Anti-human a-SMA antibody shows cross-reaction with rat a-SMA [3]. Polyclonal antibody against the rat IP receptor was gained by immunization with the peptide MVASGGRPDGPPSI, located at the N-terminus of the receptor protein, which was synthesized onto a lysine matrix. The immune serum was affinity purified using the lysine coupled peptide. The specificity of the polyclonal anti-IP receptor antibody was demonstrated by Western blot analysis. As secondary antibodies, biotinylated goat anti-mouse IgG, rhodamine-labeled goat anti-
T. Yamashita et al. / Diabetes Research and Clinical Practice 57 (2002) 149–161
153
Fig. 2. (a – d) Luminal diameters of afferent and efferent arterioles (a, b), the ratio of the diameters of afferent arterioles to those of efferent arterioles (c), and glomerular volumes (d). The diameters of the afferent arterioles and glomerular volumes were increased after the induction of diabetes (a, d). The diameters of the afferent arterioles and glomerular volume were decreased after treatment with BPS (a, d). , ND group; , STZ group; a, STZ-BPS group (mean 9S.E.M., *, P B0.05; **, PB0.01).
mouse IgG and fluorescein isothiocyanate (FITC)-labeled goat anti-rabbit IgG (Jackson Immunoresearch Laboratories, West Grove, PA) were used.
2.12. Statistical analysis Values are expressed as mean9S.E.M. The significance of the differences among three groups were done by two-way analysis of variance (ANOVA) followed by Scheffe’s test. Differences were considered significant when the P value was B 0.05.
3. Results
3.1. Metabolic data All diabetic rats were hyperglycaemic (Table 1). The serum fructosamine concentration in STZ group was significantly higher than in ND group, and no significant difference was observed between STZ group and STZ-BPS group. Body weight in ND group was higher than the other two groups. No significant difference in systolic
blood pressure was found in three groups. Creatinine clearance and urinary albumin excretion were higher in STZ group than in ND group. In STZ-BPS group, creatinine clearance and urinary albumin excretion were significantly decreased compared with STZ group.
3.2. Diameters of afferent and efferent arterioles and glomerular 6olume In STZ group, the afferent arterioles were dilated than in ND group. In STZ-BPS group, the diameters of afferent arterioles were significantly decreased compared with STZ group (ND group 10.89 0.2, STZ group 13.290.6, STZ-BPS group 9.59 0.3 mm) (mean9 S.E.M.) (Fig. 2a). The diameters of the efferent arterioles in ND group were larger than those in the other two groups (ND group 12.290.3, STZ group 10.69 0.5, STZ-BPS group 10.390.4 mm) (Fig. 2b). The ratio of diameter of afferent arteriole to that of efferent arteriole was significantly higher in STZ group than in ND group. In STZ-BPS group, the ratio was significantly lower than in STZ group (ND group 0.989 0.05, STZ group 1.479 0.14, STZ-BPS group 1.019 0.07) (Fig. 2c). The
Fructosamine (mmol/l) 124.09 3.7b 189.3 9 15.8 181.29 3.6
Plasma glucose (mg/dl)
148.39 13.7a 452.89 27.2 497.89 23.7
SBP (mmHg) 151.4 9 5.0 150.8 9 5.5 146.1 9 2.5
Body weight (g) 365.0 911.7a 299.2 9 17.1 281.4 910.8
591.2 9138.0a 2483.0 9 259.0 1629.3 960.0b
Ccr (ml/min)
162.7 9 42.3a 857.2 9 111.7 573.6 9 41.8b
Ccr/100 g BW (ml/min 100 per g)
0.4390.17a 2.91 91.27 0.489 0.13a
UAE (mg/day)
ND, non-diabetic; SBP, systolic blood pressre; a, PB0.05; b, PB0.01 vs. STZ; STZ, diabetic; Ccr, creatinine clearance; STZ+BPS, diabetic+BPS; Ccr/100 g BW, creatinine clearance/100 g body weight; UAE, urinary albumin excretion.
ND (n= 5) STZ (n=5) STZ+BPS (n= 5)
Table 1 Laboratory data
154 T. Yamashita et al. / Diabetes Research and Clinical Practice 57 (2002) 149–161
T. Yamashita et al. / Diabetes Research and Clinical Practice 57 (2002) 149–161
glomerular volume in STZ group was significantly larger than those in the other two groups (ND group 1.1390.06, STZ group 2.209 0.11, STZBPS group 0.98 90.04 106 mm3) (Fig. 2d).
3.3. Expression of ecNOS in afferent and efferent arterioles and glomeruli The afferent arterioles in all groups showed strong staining for SM2, while the efferent arterioles showed only faint staining (Fig. 3a– f). We could discriminate between pre- and post-glomerular arterioles clearly. In STZ group, fluorescence intensities for ecNOS in afferent arterioles and glomeruli were significantly higher than those in
155
ND group. In STZ-BPS group, fluorescence intensities for ecNOS in afferent arterioles and glomeruli were significantly suppressed compared with STZ group (afferent arterioles, ND group 1.179 0.05, STZ group 2.649 0.05, STZ-BPS group 2.109 0.08, glomerulus, ND group 0.489 0.05, STZ group 1.649 0.06, STZ-BPS group 0.809 0.08) (Fig. 3a–c, Fig. 4a–c, Fig. 5a and c). Fluorescence intensities for ecNOS in the efferent arterioles were weak (Fig. 3d–f). Semiquantitative analysis showed higher intensities in efferent arterioles in STZ-BPS group than those in the other two groups (ND group 1.629 0.07, STZ group 1.499 0.06, STZ-BPS group 2.589 0.07) (Fig. 5b). Urinary NO2/NO3 excretion was higher in STZ group than
Fig. 3. (a – f) Double labeling immunofluorescence staining of afferent (a – c) and efferent arterioles (d – f) for SM2 (red) and ecNOS (green). (a, d) Non-diabetic rat; (b, e) STZ-induced diabetic rat; (c, f) BPS-treated diabetic rat; (af) afferent arteriole; (ef) efferent arteriole (original magnification, ×200).
156
T. Yamashita et al. / Diabetes Research and Clinical Practice 57 (2002) 149–161
Fig. 4. (a – c) Indirect immunofluorescence staining for ecNOS in glomerulus. (a) Non-diabetic rat; (b) STZ-induced diabetic rat; (c) BPS-treated diabetic rat (original magnification, × 200).
in ND group. There was a tendency to decrease in STZ-BPS group compared with in STZ group (ND group 0.5290.16, STZ group 2.379 0.61, STZ-BPS group 1.2490.42 mmol/day) (Fig. 5d).
3.4. Expression of ICAM-1 and infiltration of macrophages in glomeruli In STZ group, ICAM-1 expression in glomeruli was increased than in ND group. ICAM-1 expression in glomeruli was significantly decreased suppressed in STZ-BPS group compared with STZ group (ND group 0.3090.05, STZ group 1.509 0.07, STZ-BPS group 0.199 0.04) (Fig. 6a– d). In STZ group, the number of ED1 positive cells (macrophages/monocytes) in glomeruli was significantly higher than in ND group. The number of ED1 positive cells were significantly decreased in STZ-BPS group compared with STZ group (ND group 0.689 0.09, STZ group 1.9390.12, STZ-BPS group 0.6090.08) (Fig. 7a– d).
3.5. Localization of prostacyclin receptor (IP receptor) in arterioles and glomeruli In each group, IP receptor was expressed in afferent arterioles and glomeruli, while very weak staining was observed in efferent arterioles (Fig. 8).
4. Discussion Glomerular hyperfiltration occurs in the early stage of diabetes. There have been several candidates for the mediators of diabetic glomerular hyperfiltration, including nitric oxide (NO), atrial natrium peptide (ANP) and vasodilatory prostaglandins, i.e. prostaglandin E2 and prostacyclin [3,4,22–26]. Jensen et al. demonstrated that the hydraulic resistance of the isolated afferent arterioles in diabetic rats is lower than in non-diabetic rats by micropuncture study [24]. They also reported that administration of indomethacin increased the resistance of afferent arterioles suggesting that prostaglandins play a role in dilatation of afferent arterioles. Both prostaglandin E2 and prostacyclin are increased within 2 weeks after induction of diabetes, however, the increase in the prostaglandins were not observed during the 25–28 days period while GFR was significantly elevated compared with normal control [23]. On the other hand, we previously showed that expression of ecNOS in afferent arterioles and glomeruli and production of NO were increased from 2 to 4 weeks after induction of diabetes in rats by STZ followed by dilatation of afferent arteriolar diameters and glomerular enlargement [3]. The administration of
T. Yamashita et al. / Diabetes Research and Clinical Practice 57 (2002) 149–161
L-NAME, an inhibitor of NOS, attenuated hyperfiltration without change in blood glucose level and systemic blood pressure [3,4,25]. In line with these data, overproduction of NO is one of the strong inducers of glomerular hyperfiltration in the early stage of diabetes. Prostacyclin has a variety of biological activities besides potent anti-platelet and vasodialtion activities. It has been reported that stable prostacyclin analogue, BPS, improves glomerular hyperfiltration in the early stage of STZ-induced diabetic rats [6]. They described that BPS did not influence the glomerular synthesis of prostacyclin and thromboxane B2 in their experiments [6]. In the present study, BPS decreased ecNOS expression in afferent arterioles and glomeruli and diameters of afferent arterioles and glomeruli. BPS also improved glomerular hyperfiltration. Several investigators described that prostacyclin inhibits the expression of ecNOS and inducible NOS [10 –12]. Barker et al. described that prostacyclin analogue suppressed the production of NO in human saphenous vein [10]. Our results strongly suggest that BPS ameliorates glomerular hyperfiltration by modulating ecNOS expression in afferent arterioles and glomeruli.
157
Arima et al. described that indomethacin enhanced vasoconstriction of efferent arterioles by angiotensin II in micropuncture study suggesting that the prostaglandins including prostaglandin E2 and prostacyclin may modulate vascular reactivity to angiotensin II of the efferent arterioles [27]. Thus, it may be possible hypothesis that BPS directly acted on efferent arterioles resulting in dilatation. However, BPS did not dilated efferent arterioles in this study. As for the distribution of IP receptor in the kidney, in situ hybridization studies showed the expression of IP receptor mRNA in afferent arterioles and in glomeruli [28,29]. Our results from double labeling immunostaining also showed that IP receptor expressed in afferent arterioles and in glomeruli, however, the staining for IP receptor in efferent arterioles was faint. These findings suggest that BPS did not act on efferent arterioles directly in the present study. We found the higher fluorescence intensities for ecNOS in the efferent arterioles in BPS group than in non-diabetic and STZ group, whereas the efferent arterioles were not dilated. The mechanism that the expression of ecNOS in efferent arterioles was increased after administration of
Fig. 5. (a – c) Fluorescence intensities for ecNOS in afferent arterioles (a), and efferent arterioles (b), and glomeruli (c). In STZ-induced diabetic rats, the intensities for ecNOS in afferent arterioles and glomeruli were significantly higher than in −
, ND; , STZ; a, STZ-BPS non-diabetic group and BPS-treated diabetic group. (d) Urinary excretion of NO− 2 /NO3 (mean 9 S.E.M., *, P B0.05; **, PB 0.01).
158
T. Yamashita et al. / Diabetes Research and Clinical Practice 57 (2002) 149–161
Fig. 6. (a – c) Indirect immnuofluorescence staining for ICAM-1 in the glomrulus in non-diabetic rat (a), STZ-induced diabetic rat (b), BPS-treated diabetic rat (c). (d) Fluorescence intensities for ICAM-1 in glomerulus. In STZ-induced diabetic group, the intensities for ICAM-1 were significantly higher than ND group and BPS-treated diabetic group (original magnification, × 200) (mean 9S.E.M., *, P B0.05; **, PB 0.01).
BPS has remained unclear. However, it was reported that NO modulates angiotensin II-induced vasoconstriction in the afferent arterioles but not in the efferent arterioles [30,31]. It may be possible that the increase of ecNOS in efferent arterioles have little influence on the diameters of the efferent arterioles. In this study, urinary excretion of NOx was decreased by BPS. On the other hand, the expression of ecNOS was decreased in afferent arteriole and glomerulus but increased in efferent arteriole in STZ-BPS group as compared with STZ group. Urinary excretion of NOx reflects the total
amount of NO produced in whole body. It is difficult to explain the discrepancy between the increase of ecNOS expression in efferent arteriole and decrease of urinary NOx excretion because we did not examine NOS expression in tubuli, renal interstitium or other organs. However, it might be possible that BPS decreased total amount of NO production in diabetic kidney by ameliorating ecNOS expression in afferent arterioles and glomeruli despite the increase of ecNOS expression in efferent arteriole. We previously reported that ICAM-1 is up-regulated in diabetic glomeruli and promotes the
T. Yamashita et al. / Diabetes Research and Clinical Practice 57 (2002) 149–161
recruitment of mononuclear cells into glomeruli [17]. ICAM-1 is induced by shear stress on endothelial cells [32,33]. We demonstrated that glomerular hyperfiltration is one of the major causes of induction of ICAM-1 in diabetic glomeruli [17]. In this study, the expression of ICAM-1 was suppressed by BPS. One of the possible mechanisms is that BPS decreased shear stress on glomerular endothelial cells resulting in suppression of ICAM-1 expression. We previously demonstrated that BPS ameliorated glomerular ICAM-1 expression in rat glomerulonephritis [20]. It is well known that ICAM-1 is induced by inflammatory cytokines [34].
159
Prostacyclin can inhibited the activation of platelets [35–37]. Therefore, it may be possible that BPS inhibited glomerular ICAM-1 expression by preventing the release of inflammatory cytokines from macrophages and platelets. BPS also decreased the infiltration of macrophages in diabetic glomeruli. It is strongly suggested that BPS decreased glomerular macrophage infiltration by down-regulation of ICAM-1. The inhibition of glomerular macrophage infiltration may be a beneficial effect of BPS besides amelioration of glomerular hyperfiltration for the treatment of diabetic nephropathy.
Fig. 7. (a – c) Indirect immnuoperoxidase staining for ED1 (indicated by red arrow heads) in glomerulus. A non-diabetic rat (a), a STZ-induced diabetic rat (b), a BPS-treated diabetic rat (c). (d) The number of ED1 positive cells in glomeruli. In the STZ-induced diabetic group, the number of ED1 positive cells (monocytes/macrophages) was significantly higher than in the non-diabetic group and BPS-treated diabetic group (original magnification, × 200) (mean 9 S.E.M., *, PB 0.05; **, P B 0.01).
160
T. Yamashita et al. / Diabetes Research and Clinical Practice 57 (2002) 149–161
Fig. 8. Double labeling immunohistochemical staining of afferent and efferent arteriole and glomerulus for SM2 (red) and prostacyclin receptor (IP receptor) (green). IP receptor was stained in afferent arteriole and in glomerulus. IP, IP receptor; af, afferent arteriole; ef, efferent arteriole; gl, glomerulus (original magnification, ×200).
In conclusion, the current results suggest that stable prostacyclin analogue, BPS improved glomerular hyperfiltration and urinary albumin excretion through modulating ecNOS expression in afferent arterioles and glomeruli in diabetic rats. Moreover, BPS ameliorated glomerular expression of ICAM-1 and infiltration of macrophages. BPS may be beneficial for the therapy in the early stage of diabetic nephropathy.
Acknowledgements This work was supported by grant from Ministry of Education, Science, Sports and Culture of Japan (C11671036, C13671116) to K. Shikata.
References [1] S.M. Mauer, M.W. Steffes, E.N. Ellis, et al., Structural – functional relationship in diabetic nephropathy, J. Clin. Invest. 74 (1984) 1143 –1155. [2] H. Makino, Y. Yamasaki, T. Haramoto, et al., Ultrastructural changes of extracellular matrices in diabetic nephropathy revealed by high resolution scanning and immunoelectron microscopy, Lab. Invest. 68 (1993) 45 – 55.
[3] H. Sugimoto, K. Shikata, M. Matsuda, H. Makino, et al., Increased expression of endotherial cell nitric oxide synthase (ecNOS) in afferent and glomerular endotherial cells is involved in glomerular hyperfiltraion of diabetic nephropathy, Diabetologia 41 (1998) 1426 – 1434. [4] R. Veelken, K.F. Hilgers, A. Hartner, A. Haas, K.P. Bohmer, R.B. Sterzel, Nitric oxide synthesis isoforms and glomerular hyperfiltration in early diabetic nephropathy, J. Am. Soc. Nephrol. 11 (2000) 71 – 79. [5] K. Hiragushi, H. Sugimoto, K. Shikata, et al., Nitric oxide system is involved in glomerular hyperfiltration in Japanese normo- and micro-albuminuric patients with type 2 diabetes, Diabetes Res. Clin. Prac. 53 (2001) 149 – 159. [6] L.N. Wang, Z. Tang, S. Ichiyu, et al., Effect of the PGI2 analog beraprost sodium on glomerular prostanoid synthesis in rats with streptozotocin-induced diabetes, Nephron 73 (1996) 637 – 643. [7] J.M. Armstrong, N. Lattimer, S. Moncada, et al., Comparison of the vasodepressor effects of prostacyclin and 6-oxoprostaglandin F1a with those of prostacyclin E2 in rats and rabbits, Br. J. Pharmacol. 62 (1987) 125 – 130. [8] B.J.R. Whittle, S. Moncada, J.R. Vane, Comparison of the effects of prostacyclin (PGI2), prostaglandin E1 and D2 on platlet aggregation in different species, Prostaglandins 16 (1978) 373 – 388. [9] S. Moncada, Biology and therapeutic potential of prostacyclin, Stroke 14 (1983) 157 – 168. [10] J.E. Barker, Y.S. Bakhle, J. Anderson, et al., Reciprocal inhibition of nitric oxide and prostacyclin synthesis in human saphenous vein, Br. J. Pharmacol. 118 (1996) 643 – 648. [11] P. Marotta, L. Sautebin, M. DiRosa, Modulation of induction of nitric oxide synthase by eicosanoids in the murine macrophage cell line J774, Br. J. Pharmacol. 107 (1992) 640 – 641. [12] F. D’Acquisto, L. Sautebin, T. Iuvone, Prostaglandin prevent inducible nitric oxide synthase protein expression by inhibiting nuclear factor-kB activation in J774 macrophages, FEBS Lett. 440 (1998) 76 – 80. [13] D. Salvemini, M.G. Currie, V. Mollace, Nitric oxide-mediated cyclooxygenase activation. A key event in the antiplatelet effects of nitrovasodilators, J. Clin. Invest. 97 (1996) 2562 – 2568. [14] T. Furuta, T. Saito, T. Ootaka, et al., The role of macrophages in diabetic glomerulosclerosis, Am. J. Kidney Dis. 21 (1993) 480 – 485. [15] K. Hirata, K. Shikata, H. Makino, et al., Increased expression of selectins in kidneys of patients with diabetic nephropathy, Diabetologia 41 (1998) 185 – 192. [16] C. Sassy-Prigent, D. Heudes, C. Mandet, et al., Early glomerular macrophage recruitment in streptozotocin-induced diabetic rats, Diabetes 49 (2000) 466 – 475. [17] H. Sugimoto, K. Shikata, H. Makino, et al., Increased expression of intercellular adhesion molecule-1 (ICAM-1) in diabetic rat glomeruli, Diabetes 46 (1997) 2075 – 2081.
T. Yamashita et al. / Diabetes Research and Clinical Practice 57 (2002) 149–161 [18] H. Sugimoto, K. Shikata, H. Makino, et al., Increased gene expression of insulin-like growth factor(IGF-1) receptor in experimental diabetic rat glomeruli, Nephron 72 (1996) 648 – 653. [19] T. Umetsu, T. Murata, Y. Tanaka, et al., Antithrombotic effect of TRK-100, a novel, stable PGI2 analogue, Jpn. J. Pharmacol. 43 (1987) 81 –90. [20] M. Kushiro, K. Shikata, H. Makino, et al., Therapeutic effects of prostacyclin analog on crescentic glomerulonephritis of rat, Kidney Int. 53 (1998) 1314 –1320. [21] K. Kimura, R. Nagai, T. Sakai, et al., Diversity and variability of smooth muscle phenotypes of renal arterioles as revealed by myosin isoform expression, Kidney Int. 48 (1995) 372 –382. [22] R. Kikkawa, M. Haneda, K. Sakamoto, et al., Antagonist for atrial natriuretic peptide receptors ameliorates glomerular hyperfiltration in diabetic rats, Biochem. Biophys. Res. Commun. 193 (1993) 700 –705. [23] P.A. Craven, M.A. Caines, F.R. DeRubertis, Sequential alterations in glomerular prostaglandin and thromboxane synthesis in diabetic rats: relationship to the hyperfiltration of early diabetes, Metabolism 36 (1987) 95 –103. [24] P.K. Jensen, K. Steven, H.H. Parving, et al., Effect of indomethacin on glomerular hemodynamics in experimental diabetes, Kidney Int. 29 (1986) 490 –495. [25] R. Komer, T.J. Allen, M.E. Cooper, Role of endothelium-derived nitric oxide in the pathogenesis of the renal hemodynanic changes of experimental diabetes, Diabetes 43 (1994) 1190 – 1197. [26] N. Bank, H.S. Aynedjian, Role of EDRF (nitric oxide) in diabetic renal hyperfiltration, Kidney Int. 43 (1993) 1306 – 1312. [27] S. Arima, Y.L. Ren, L.A. Juncos, et al., Glomerular prostaglandins modulate vascular reactivity of the downstream efferent arterioles, Kidney Int. 45 (1994) 50 –658.
161
[28] H. Oida, T. Namba, Y. Sugimoto, et al., In situ hybridization studies of prostacyclin receptor mRNA expression in various mouse organs, Br. J. Pharmacol. 116 (1995) 2828 – 2837. [29] M. Komhoff, B. Lesener, K. Nakao, et al., Localization of the prostacyclin receptor in human kidney, Kidney Int. 54 (1998) 1899 – 1908. [30] S. Ito, S. Arima, Y.L. Ren, et al., Endothelium-derived relaxing factor/nitric oxide modulates angiotensin II action in the isolated microperfused rabbit afferent but not efferent arteriole, J. Clin. Invest. 91 (1993) 2012 – 2019. [31] S. Ito, C.S. Johnson, O.A. Carretero, Modulation of angiotensin II-induced vasoconstriction by endotheliumderived relaxing factor in the isolated microperfused rabbit afferent arteriole, J. Clin. Invest. 87 (1991) 1656 – 1663. [32] T. Nagel, N. Resnick, W.J. Atkinson, et al., Shear stress selectively upregulates intercellular adhesion molecule-1 expression in cultured human vascular endothelial cells, J. Clin. Invest. 94 (1994) 885 – 891. [33] D.E. Stauton, S.D. Marlin, C. Stratowa, et al., Primary structure of ICAM-1 demonstrates interaction between members of the immunoglobulin and integrin supergene families, Cell 52 (1988) 925 – 933. [34] D.E. Staunton, S.D. Marlin, C. Stratowa, et al., Primary structure of ICAM-1 demonstrates interaction between members of the immunogloblin and integlin supergene families, Cell 52 (1988) 925 – 933. [35] M.W. Radomski, R.M.J. Palmer, S. Moncada, The antiaggregating properties of vascular endothelium: interactions between prostacyclin and nitic oxide, Br. J. Pharmacol. 92 (1987) 639 – 646. [36] S.L. Kunkel, S.W. Chensue, The role of arachidonic acid metabolism in mononuclear phagocytic cell interactions, Int. J. Dermatol. 25 (1986) 83 – 89. [37] J. Marcinkiewicz, B.M. Chain, Differential cytokine regulation by eicosanoids in T cells primed by contact sensitization with TNP, Cell Immunol. 149 (1993) 303 – 314.
Beraprost sodium, prostacyclin analogue, attenuates glomerular hyperfiltration and glomerular macrophage infiltration by modulating ecNOS expression in diabetic rats Tetsuji Yamashita, Kenichi Shikata *, Mitsuhiro Matsuda, Shinichi Okada, Daisuke Ogawa, Hikaru Sugimoto, Jun Wada, Hirofumi Makino Department of Medicine III, Okayama Uni6ersity Medical School, 2 -5 -1, Shikata-cho, Okayama 700 -8558, Japan Received 9 January 2002; received in revised form 4 March 2002; accepted 18 March 2002
Abstract Stable prostacyclin analogue, beraprost sodium (BPS) has recently been reported to attenuate glomerular hyperfiltration in diabetic rats, however, the mechanism has been still unknown. We previously reported that overexpression of endothelial cell nitric oxide synthase (ecNOS) in afferent arterioles and glomeruli induce inappropriate dilatation of afferent arterioles and glomerular hyperfiltration through overproduction of nitric oxide in early stage of diabetic nephropathy. In this study, we tested the hypothesis that BPS ameliorates glomerular hyperfiltration through modulating ecNOS expression in diabetic nephropathy. Furthermore, we examined the effects of BPS on the expression of intercellular adhesion molecule-1 (ICAM-1) and macrophage infiltration in diabetic glomeruli, because glomerular hyperfiltration induces the expression of ICAM-1 resulting in macrophage infiltration. Male Sprague– Dawley (SD) rats were administered continuously with BPS for 4 weeks after induction of diabetes by streptozotocin. In diabetic rats, the diameters of afferent arterioles, glomerular volume, creatinine clearance and urinary excretion of albumin and NO2/NO3 were increased as compared with non-diabetic control rats. Treatment with BPS improved these changes. The expression of ecNOS was increased in afferent arterioles and glomeruli in diabetic rats and suppressed by BPS. Prostacyclin receptor was expressed along afferent arterioles. Our results suggest that BPS attenuates glomerular hyperfiltration by modulating ecNOS expression in early stage of diabetic nephropathy. Moreover, BPS may inhibit ICAM-1-dependent infiltration of macrophages in diabetic glomeruli. © 2002 Elsevier Science Ireland Ltd. All rights reserved. Keywords: Diabetic nephropathy; Hyperfiltration; Prostacyclin; Nitric oxide; ICAM-1; Macrophage
1. Introduction
* Corresponding author. Tel.: +81-86-235-7234; fax: + 8186-222-5214 E-mail address: [email protected] (K. Shikata).
Dilatation of afferent arterioles is one of the characteristic changes in the early stage of diabetic nephropathy. Inappropriate dilatation of afferent arterioles is believed to induce glomerular
0168-8227/02/$ - see front matter © 2002 Elsevier Science Ireland Ltd. All rights reserved. PII: S 0 1 6 8 - 8 2 2 7 ( 0 2 ) 0 0 0 5 4 - 2
150
T. Yamashita et al. / Diabetes Research and Clinical Practice 57 (2002) 149–161
hyperfiltration and hypertrophy followed by thickening of the glomerular basement membrane and accumulation of mesangial matrix [1,2]. We have recently reported that expression of endothelial cell nitric oxide synthase (ecNOS) isoform is up-regulated in afferent arterioles and glomeruli in the early stage of streptozotocin (STZ)-induced diabetic rat [3]. Treatment with L-NG-nitroarginine methyl ester HCL (LNAME), an inhibitor of NOS, ameliorated glomerular enlargement and glomerular hyperfiltration suggesting that overproduction of nitric oxide in afferent arterioles and glomeruli induce dilatation of afferent arterioles, glomerular hypertrophy and glomerular hyperfiltration [3]. Our results have been further supported by Veelken et al. [4]. We have also reported that overproduction of NO is involved in the mechanism of glomerular hyperfiltlation in patients with type 2 diabetes [5]. Recently, it has been reported that prostacyclin analogue, beraprost sodium (BPS), improved glomerular hyperfiltration and reduced urinary albumin excretion in STZ-induced diabetic rat [6], although the mechanism is still unknown. BPS is known to have a relaxative action on vascular smooth muscle [7], and an inhibitory action against platelet aggregation [8,9]. It was also reported that prostaglandins interact with the process of nitric oxide synthesis [10– 13]. In the present study, we tested the hypothesis that BPS ameliorates glomerular hyperfiltration through modulating ecNOS expression in the early stage of diabetic nephropathy using STZinduced diabetic rats. It is well known that macrophages infiltrate into glomeruli and interstitium in diabetic kidney [14]. Macrophage is considered to play an important role in the progression of diabetic glomerulosclerosis [14– 16]. Intercellular adhesion molecule-1 (ICAM-1) is up-regulated in diabetic glomeruli and mediates glomerular macrophage infiltration [15– 17]. We previously reported that glomerular hyperfiltration is one of the major causes of up-regulation of ICAM-1 in diabetic glomeruli [17]. In the current study, we have also examined the effects of BPS on the expres-
sion of ICAM-1 and macrophage infiltration in diabetic glomeruli.
2. Materials and methods
2.1. Animals Male Sprague–Dawley (SD) rats weighing about 120 g (4 weeks of age) were purchased from Charles River Japan (Yokohama, Japan). These rats received a standard chow and water diet.
2.2. Experimental protocol Diabetes was induced by the intravenous injection of 65mg/kg STZ (Sigma, St. Louis, MO) in citrate buffer, pH 4.5 as previously described [18]. Non-diabetic control rats were age-and sexmatched with the experimental rats and received citrate buffer alone. From 1 day after injection of STZ, one group was received continuous intraperitoneal administration of 30 mg per rat per day of BPS by an osmotic pump (ALZA Co., Palo Alto, CA, USA) (STZ-BPS group; n= 5). Another diabetic group (STZ group; n= 5) and non-diabetic control group (ND group; n= 5) were received saline by osmotic pump. BPS (TRK-100) was kindly provided by Toray Industries, Inc., Tokyo, Japan. The levels of fasting blood glucose in all diabetic rats were kept between 400 and 500 mg/dl by insulin treatment, using nearly 24 h-acting insulin, Humalin N (Shionogi, Osaka, Japan). Dose-dependent anti-platelet effect of BPS is observed at the dose of 3–300 mg/kg per day in rats [19]. Our previous studies indicated that 250–300 mg/kg per day of BPS does not affect systemic blood pressure in rats [20]. Therefore, we used 30 mg per rat per day (250 mg/kg per day) of BPS in this study. This study was carried out in accordance with the Guidelines for Animal Experiments at Okayama University Medical School, Japanese Government Animal Protection and Management Law (No. 105) and Japanese Government Notification on Feeding and Safekeeping of Animals (No. 6).
T. Yamashita et al. / Diabetes Research and Clinical Practice 57 (2002) 149–161
2.3. Metabolic data On 28 days after induction of diabetes, all rats were transferred to metabolic cages for the collection of 24 h urine samples, and blood samples were taken from the tail vein. They then had free access to water only ad libitum. The chemical analysis of creatinine in serum and urine was performed by the standard laboratory methods. Fructosamine as measured by Precimat Fructosamine (Boehringer Mannheim, Tokyo, Japan). Fasting blood glucose was measured by the glucose oxidase method. Systolic blood pressure was measured three times by the tail cuff method. Urine samples were assayed for the stable NO metabolic end products, NO2 and NO3 with a Nitrate/Nitrite Assay Kit (Cayman Chemical, Ann Arbor, MI, USA). The level of urinary albumin was measured using NEPHRAT (Exocell, Inc., Philadelphia, USA), a kit for enzyme-linked immunosorbent assay. Five rats for each group were killed under anesthesia with 5 mg/100 g body weight of sodium pentobarbital at 4 weeks after BPS or saline administration, then both kidneys were removed. A part of kidney was fixed in 95% ethanol with 1% acetic acid at 4 °C overnight [21], then dehydrated and embedded in paraffin. Tissue blocks were stored at 4 °C until use. Other tissues were snap frozen in acetone and stored at − 20 °C.
151
were observed and photographed. The maximum afferent and efferent arteriolar diameters (luminar diameter) were then measured on photographs (× 200). Using the serial sections described above, we selected 30 glomeruli for each kidney specimen. Then the maximum diameter of selected glomeruli in serial sections was measured on photograph (× 200). The diameters were calculated as the mean of the longest and shortest diameters. The glomerular volume was determined from the mean glomerular diameter, d (G), using the formula: 4y(d (G)/2)3/3 [3].
2.5. Expression of ecNOS in afferent and efferent arterioles and glomeruli Double labeling immunofluorescence studies were performed, using mAb against rat ecNOS and mAb against SM2 isoform of myosin as primary antibodies. It is known that expression of SM2 isoform of myosin is restricted in the preglomerular vessels including afferent arterioles, not in efferent arterioles [21]. The ecNOS fluorescence intensities of arteiroles and glomeruli were graded semiquantiatively according to the following scale: ‘— ’, no detectable staining; ‘+ ’, weak staining; ‘2+’, moderate staining; ‘3+’, strong staining. The intensities were graded by two separate investigators in 30
2.4. Measurement of the diameters of afferent and efferent arterioles and glomerular 6olume We cut 4 mm-thick serial sections for the immunohistochemical studies. In each kidney specimen, 30 serial sections were prepared. Immunoperoxidase staining was done, using mouse monoclonal antibody (mAb) against human a-smooth muscle actin (a-SMA) as primary antibody. Interlobular arteries, afferent and efferent arterioles were sequentially identified with a-SMA [3] (Fig. 1). In each kidney specimen, 25 afferent and efferent arterioles in arteriolar longitudinal sections were selected randomly from the superficial nephrons and evaluated. Several serial sections
Fig. 1. (a – c) Serial sections of kidney from a non-diabetic rat stained with anti-a-SMA antibody. Afferent arterioles (af) and efferent arterioles (ef) can be clearly distinguished. ILA, interlobular artery; gl, glomerulus (original magnification, ×200).
152
T. Yamashita et al. / Diabetes Research and Clinical Practice 57 (2002) 149–161
glomeruli and 25 afferent arterioles and 25 efferent arterioles per animal and expressed as the average intensity.
oped in 3,3-diaminobenzidine and hydrogen peroxide. The sections were then counterstained with Mayer’s hematoxylin.
2.6. Expression of ICAM-1 and infiltration of macrophages in glomeruli
2.9. Indirect immunofluorescence study
Indirect immunofluorescence studies were performed, as described below, using mAb against rat ICAM-1 as primary antibody. Immunoperoxidase staining was done, using mouse mAb against rat monocytes/macrophages (ED1) as primary antibody. The estimation of ICMA-1 fluorescence intensities in glomeruli was done semiquantitatively by the same method as described above. The number of ED1 positive cells (monocytes/macrophages) was counted in 30 glomeruli per rat. The average number of positive cells per glomerulus was calculated.
2.7. Localization of prostacyclin receptor (IP receptor) in arterioles and glomeruli Double labeling immunofluorescence studies were performed, using rabbit polyclonal antibody against rat prostacyclin receptor and mAb against human SM2 as primary antibodies.
2.8. Immunoperoxidase staining Immunoperoxidase staining was done by using ABC kit (Vector Lab., Burlingame, CA, USA) as described previously [16]. In brief, non-specific protein binding was blocked by incubating the cryostat sections with 10% bovine serum in Tris– buffered saline for 20 min. Non-specific staining was blocked by 15 min incubation with avidin and then biotin using the avidin– biotin blocking kit (Vector Lab.). Endogenous peroxidase activity was inhibited by incubating the sections in methanol containing 0.3% H2O2 for 20 min. Sections were first incubated for 60 min with primary antibodies at room temperature. Then the sections were incubated with secondary antibodies for 30 min at room temperature. Biotinylated horseradish peroxidase was applied for 30 min at room temperature. Peroxidase activity was devel-
Indirect immunofluorescence studies were performed on 4 mm cryostat sections fixed with acetone that were stained with a primary antibodies at 4 °C overnight and followed by a FITC-labeled goat anti-mouse or rabbit IgG for 30 min at room temperature.
2.10. Double labeling immunofluorescence study Double labeling immunofluorescence studies were performed on 4 mm cryostat sections fixed with acetone that were stained with primary antibodies (anti-rat ecNOS mAb or anti-rat IP receptor antibody) at 4 °C overnight and followed by a FITC-labeled goat anti-mouse or rabbit IgG for 30 min at room temperature. Then the sections were stained with a mouse mAb against SM2 for 30 min at room temperature and rhodamine-labeled goat anti-mouse IgG for 30 min at room temperature.
2.11. Antibodies As primary antibodies, we used mouse mAbs against human a-SMA (DAKO, Kyoto, Japan), human SM2 (Yamasa Co. Ltd., Chiba, Japan), and rat ecNOS (Transduction Laboratory, Kentucky, USA), and rabbit polyclonal antibody against rat prostacyclin receptor (IP receptor) (kindly provided by Toray Industries, Inc., Tokyo, Japan). Anti-human a-SMA antibody shows cross-reaction with rat a-SMA [3]. Polyclonal antibody against the rat IP receptor was gained by immunization with the peptide MVASGGRPDGPPSI, located at the N-terminus of the receptor protein, which was synthesized onto a lysine matrix. The immune serum was affinity purified using the lysine coupled peptide. The specificity of the polyclonal anti-IP receptor antibody was demonstrated by Western blot analysis. As secondary antibodies, biotinylated goat anti-mouse IgG, rhodamine-labeled goat anti-
T. Yamashita et al. / Diabetes Research and Clinical Practice 57 (2002) 149–161
153
Fig. 2. (a – d) Luminal diameters of afferent and efferent arterioles (a, b), the ratio of the diameters of afferent arterioles to those of efferent arterioles (c), and glomerular volumes (d). The diameters of the afferent arterioles and glomerular volumes were increased after the induction of diabetes (a, d). The diameters of the afferent arterioles and glomerular volume were decreased after treatment with BPS (a, d). , ND group; , STZ group; a, STZ-BPS group (mean 9S.E.M., *, P B0.05; **, PB0.01).
mouse IgG and fluorescein isothiocyanate (FITC)-labeled goat anti-rabbit IgG (Jackson Immunoresearch Laboratories, West Grove, PA) were used.
2.12. Statistical analysis Values are expressed as mean9S.E.M. The significance of the differences among three groups were done by two-way analysis of variance (ANOVA) followed by Scheffe’s test. Differences were considered significant when the P value was B 0.05.
3. Results
3.1. Metabolic data All diabetic rats were hyperglycaemic (Table 1). The serum fructosamine concentration in STZ group was significantly higher than in ND group, and no significant difference was observed between STZ group and STZ-BPS group. Body weight in ND group was higher than the other two groups. No significant difference in systolic
blood pressure was found in three groups. Creatinine clearance and urinary albumin excretion were higher in STZ group than in ND group. In STZ-BPS group, creatinine clearance and urinary albumin excretion were significantly decreased compared with STZ group.
3.2. Diameters of afferent and efferent arterioles and glomerular 6olume In STZ group, the afferent arterioles were dilated than in ND group. In STZ-BPS group, the diameters of afferent arterioles were significantly decreased compared with STZ group (ND group 10.89 0.2, STZ group 13.290.6, STZ-BPS group 9.59 0.3 mm) (mean9 S.E.M.) (Fig. 2a). The diameters of the efferent arterioles in ND group were larger than those in the other two groups (ND group 12.290.3, STZ group 10.69 0.5, STZ-BPS group 10.390.4 mm) (Fig. 2b). The ratio of diameter of afferent arteriole to that of efferent arteriole was significantly higher in STZ group than in ND group. In STZ-BPS group, the ratio was significantly lower than in STZ group (ND group 0.989 0.05, STZ group 1.479 0.14, STZ-BPS group 1.019 0.07) (Fig. 2c). The
Fructosamine (mmol/l) 124.09 3.7b 189.3 9 15.8 181.29 3.6
Plasma glucose (mg/dl)
148.39 13.7a 452.89 27.2 497.89 23.7
SBP (mmHg) 151.4 9 5.0 150.8 9 5.5 146.1 9 2.5
Body weight (g) 365.0 911.7a 299.2 9 17.1 281.4 910.8
591.2 9138.0a 2483.0 9 259.0 1629.3 960.0b
Ccr (ml/min)
162.7 9 42.3a 857.2 9 111.7 573.6 9 41.8b
Ccr/100 g BW (ml/min 100 per g)
0.4390.17a 2.91 91.27 0.489 0.13a
UAE (mg/day)
ND, non-diabetic; SBP, systolic blood pressre; a, PB0.05; b, PB0.01 vs. STZ; STZ, diabetic; Ccr, creatinine clearance; STZ+BPS, diabetic+BPS; Ccr/100 g BW, creatinine clearance/100 g body weight; UAE, urinary albumin excretion.
ND (n= 5) STZ (n=5) STZ+BPS (n= 5)
Table 1 Laboratory data
154 T. Yamashita et al. / Diabetes Research and Clinical Practice 57 (2002) 149–161
T. Yamashita et al. / Diabetes Research and Clinical Practice 57 (2002) 149–161
glomerular volume in STZ group was significantly larger than those in the other two groups (ND group 1.1390.06, STZ group 2.209 0.11, STZBPS group 0.98 90.04 106 mm3) (Fig. 2d).
3.3. Expression of ecNOS in afferent and efferent arterioles and glomeruli The afferent arterioles in all groups showed strong staining for SM2, while the efferent arterioles showed only faint staining (Fig. 3a– f). We could discriminate between pre- and post-glomerular arterioles clearly. In STZ group, fluorescence intensities for ecNOS in afferent arterioles and glomeruli were significantly higher than those in
155
ND group. In STZ-BPS group, fluorescence intensities for ecNOS in afferent arterioles and glomeruli were significantly suppressed compared with STZ group (afferent arterioles, ND group 1.179 0.05, STZ group 2.649 0.05, STZ-BPS group 2.109 0.08, glomerulus, ND group 0.489 0.05, STZ group 1.649 0.06, STZ-BPS group 0.809 0.08) (Fig. 3a–c, Fig. 4a–c, Fig. 5a and c). Fluorescence intensities for ecNOS in the efferent arterioles were weak (Fig. 3d–f). Semiquantitative analysis showed higher intensities in efferent arterioles in STZ-BPS group than those in the other two groups (ND group 1.629 0.07, STZ group 1.499 0.06, STZ-BPS group 2.589 0.07) (Fig. 5b). Urinary NO2/NO3 excretion was higher in STZ group than
Fig. 3. (a – f) Double labeling immunofluorescence staining of afferent (a – c) and efferent arterioles (d – f) for SM2 (red) and ecNOS (green). (a, d) Non-diabetic rat; (b, e) STZ-induced diabetic rat; (c, f) BPS-treated diabetic rat; (af) afferent arteriole; (ef) efferent arteriole (original magnification, ×200).
156
T. Yamashita et al. / Diabetes Research and Clinical Practice 57 (2002) 149–161
Fig. 4. (a – c) Indirect immunofluorescence staining for ecNOS in glomerulus. (a) Non-diabetic rat; (b) STZ-induced diabetic rat; (c) BPS-treated diabetic rat (original magnification, × 200).
in ND group. There was a tendency to decrease in STZ-BPS group compared with in STZ group (ND group 0.5290.16, STZ group 2.379 0.61, STZ-BPS group 1.2490.42 mmol/day) (Fig. 5d).
3.4. Expression of ICAM-1 and infiltration of macrophages in glomeruli In STZ group, ICAM-1 expression in glomeruli was increased than in ND group. ICAM-1 expression in glomeruli was significantly decreased suppressed in STZ-BPS group compared with STZ group (ND group 0.3090.05, STZ group 1.509 0.07, STZ-BPS group 0.199 0.04) (Fig. 6a– d). In STZ group, the number of ED1 positive cells (macrophages/monocytes) in glomeruli was significantly higher than in ND group. The number of ED1 positive cells were significantly decreased in STZ-BPS group compared with STZ group (ND group 0.689 0.09, STZ group 1.9390.12, STZ-BPS group 0.6090.08) (Fig. 7a– d).
3.5. Localization of prostacyclin receptor (IP receptor) in arterioles and glomeruli In each group, IP receptor was expressed in afferent arterioles and glomeruli, while very weak staining was observed in efferent arterioles (Fig. 8).
4. Discussion Glomerular hyperfiltration occurs in the early stage of diabetes. There have been several candidates for the mediators of diabetic glomerular hyperfiltration, including nitric oxide (NO), atrial natrium peptide (ANP) and vasodilatory prostaglandins, i.e. prostaglandin E2 and prostacyclin [3,4,22–26]. Jensen et al. demonstrated that the hydraulic resistance of the isolated afferent arterioles in diabetic rats is lower than in non-diabetic rats by micropuncture study [24]. They also reported that administration of indomethacin increased the resistance of afferent arterioles suggesting that prostaglandins play a role in dilatation of afferent arterioles. Both prostaglandin E2 and prostacyclin are increased within 2 weeks after induction of diabetes, however, the increase in the prostaglandins were not observed during the 25–28 days period while GFR was significantly elevated compared with normal control [23]. On the other hand, we previously showed that expression of ecNOS in afferent arterioles and glomeruli and production of NO were increased from 2 to 4 weeks after induction of diabetes in rats by STZ followed by dilatation of afferent arteriolar diameters and glomerular enlargement [3]. The administration of
T. Yamashita et al. / Diabetes Research and Clinical Practice 57 (2002) 149–161
L-NAME, an inhibitor of NOS, attenuated hyperfiltration without change in blood glucose level and systemic blood pressure [3,4,25]. In line with these data, overproduction of NO is one of the strong inducers of glomerular hyperfiltration in the early stage of diabetes. Prostacyclin has a variety of biological activities besides potent anti-platelet and vasodialtion activities. It has been reported that stable prostacyclin analogue, BPS, improves glomerular hyperfiltration in the early stage of STZ-induced diabetic rats [6]. They described that BPS did not influence the glomerular synthesis of prostacyclin and thromboxane B2 in their experiments [6]. In the present study, BPS decreased ecNOS expression in afferent arterioles and glomeruli and diameters of afferent arterioles and glomeruli. BPS also improved glomerular hyperfiltration. Several investigators described that prostacyclin inhibits the expression of ecNOS and inducible NOS [10 –12]. Barker et al. described that prostacyclin analogue suppressed the production of NO in human saphenous vein [10]. Our results strongly suggest that BPS ameliorates glomerular hyperfiltration by modulating ecNOS expression in afferent arterioles and glomeruli.
157
Arima et al. described that indomethacin enhanced vasoconstriction of efferent arterioles by angiotensin II in micropuncture study suggesting that the prostaglandins including prostaglandin E2 and prostacyclin may modulate vascular reactivity to angiotensin II of the efferent arterioles [27]. Thus, it may be possible hypothesis that BPS directly acted on efferent arterioles resulting in dilatation. However, BPS did not dilated efferent arterioles in this study. As for the distribution of IP receptor in the kidney, in situ hybridization studies showed the expression of IP receptor mRNA in afferent arterioles and in glomeruli [28,29]. Our results from double labeling immunostaining also showed that IP receptor expressed in afferent arterioles and in glomeruli, however, the staining for IP receptor in efferent arterioles was faint. These findings suggest that BPS did not act on efferent arterioles directly in the present study. We found the higher fluorescence intensities for ecNOS in the efferent arterioles in BPS group than in non-diabetic and STZ group, whereas the efferent arterioles were not dilated. The mechanism that the expression of ecNOS in efferent arterioles was increased after administration of
Fig. 5. (a – c) Fluorescence intensities for ecNOS in afferent arterioles (a), and efferent arterioles (b), and glomeruli (c). In STZ-induced diabetic rats, the intensities for ecNOS in afferent arterioles and glomeruli were significantly higher than in −
, ND; , STZ; a, STZ-BPS non-diabetic group and BPS-treated diabetic group. (d) Urinary excretion of NO− 2 /NO3 (mean 9 S.E.M., *, P B0.05; **, PB 0.01).
158
T. Yamashita et al. / Diabetes Research and Clinical Practice 57 (2002) 149–161
Fig. 6. (a – c) Indirect immnuofluorescence staining for ICAM-1 in the glomrulus in non-diabetic rat (a), STZ-induced diabetic rat (b), BPS-treated diabetic rat (c). (d) Fluorescence intensities for ICAM-1 in glomerulus. In STZ-induced diabetic group, the intensities for ICAM-1 were significantly higher than ND group and BPS-treated diabetic group (original magnification, × 200) (mean 9S.E.M., *, P B0.05; **, PB 0.01).
BPS has remained unclear. However, it was reported that NO modulates angiotensin II-induced vasoconstriction in the afferent arterioles but not in the efferent arterioles [30,31]. It may be possible that the increase of ecNOS in efferent arterioles have little influence on the diameters of the efferent arterioles. In this study, urinary excretion of NOx was decreased by BPS. On the other hand, the expression of ecNOS was decreased in afferent arteriole and glomerulus but increased in efferent arteriole in STZ-BPS group as compared with STZ group. Urinary excretion of NOx reflects the total
amount of NO produced in whole body. It is difficult to explain the discrepancy between the increase of ecNOS expression in efferent arteriole and decrease of urinary NOx excretion because we did not examine NOS expression in tubuli, renal interstitium or other organs. However, it might be possible that BPS decreased total amount of NO production in diabetic kidney by ameliorating ecNOS expression in afferent arterioles and glomeruli despite the increase of ecNOS expression in efferent arteriole. We previously reported that ICAM-1 is up-regulated in diabetic glomeruli and promotes the
T. Yamashita et al. / Diabetes Research and Clinical Practice 57 (2002) 149–161
recruitment of mononuclear cells into glomeruli [17]. ICAM-1 is induced by shear stress on endothelial cells [32,33]. We demonstrated that glomerular hyperfiltration is one of the major causes of induction of ICAM-1 in diabetic glomeruli [17]. In this study, the expression of ICAM-1 was suppressed by BPS. One of the possible mechanisms is that BPS decreased shear stress on glomerular endothelial cells resulting in suppression of ICAM-1 expression. We previously demonstrated that BPS ameliorated glomerular ICAM-1 expression in rat glomerulonephritis [20]. It is well known that ICAM-1 is induced by inflammatory cytokines [34].
159
Prostacyclin can inhibited the activation of platelets [35–37]. Therefore, it may be possible that BPS inhibited glomerular ICAM-1 expression by preventing the release of inflammatory cytokines from macrophages and platelets. BPS also decreased the infiltration of macrophages in diabetic glomeruli. It is strongly suggested that BPS decreased glomerular macrophage infiltration by down-regulation of ICAM-1. The inhibition of glomerular macrophage infiltration may be a beneficial effect of BPS besides amelioration of glomerular hyperfiltration for the treatment of diabetic nephropathy.
Fig. 7. (a – c) Indirect immnuoperoxidase staining for ED1 (indicated by red arrow heads) in glomerulus. A non-diabetic rat (a), a STZ-induced diabetic rat (b), a BPS-treated diabetic rat (c). (d) The number of ED1 positive cells in glomeruli. In the STZ-induced diabetic group, the number of ED1 positive cells (monocytes/macrophages) was significantly higher than in the non-diabetic group and BPS-treated diabetic group (original magnification, × 200) (mean 9 S.E.M., *, PB 0.05; **, P B 0.01).
160
T. Yamashita et al. / Diabetes Research and Clinical Practice 57 (2002) 149–161
Fig. 8. Double labeling immunohistochemical staining of afferent and efferent arteriole and glomerulus for SM2 (red) and prostacyclin receptor (IP receptor) (green). IP receptor was stained in afferent arteriole and in glomerulus. IP, IP receptor; af, afferent arteriole; ef, efferent arteriole; gl, glomerulus (original magnification, ×200).
In conclusion, the current results suggest that stable prostacyclin analogue, BPS improved glomerular hyperfiltration and urinary albumin excretion through modulating ecNOS expression in afferent arterioles and glomeruli in diabetic rats. Moreover, BPS ameliorated glomerular expression of ICAM-1 and infiltration of macrophages. BPS may be beneficial for the therapy in the early stage of diabetic nephropathy.
Acknowledgements This work was supported by grant from Ministry of Education, Science, Sports and Culture of Japan (C11671036, C13671116) to K. Shikata.
References [1] S.M. Mauer, M.W. Steffes, E.N. Ellis, et al., Structural – functional relationship in diabetic nephropathy, J. Clin. Invest. 74 (1984) 1143 –1155. [2] H. Makino, Y. Yamasaki, T. Haramoto, et al., Ultrastructural changes of extracellular matrices in diabetic nephropathy revealed by high resolution scanning and immunoelectron microscopy, Lab. Invest. 68 (1993) 45 – 55.
[3] H. Sugimoto, K. Shikata, M. Matsuda, H. Makino, et al., Increased expression of endotherial cell nitric oxide synthase (ecNOS) in afferent and glomerular endotherial cells is involved in glomerular hyperfiltraion of diabetic nephropathy, Diabetologia 41 (1998) 1426 – 1434. [4] R. Veelken, K.F. Hilgers, A. Hartner, A. Haas, K.P. Bohmer, R.B. Sterzel, Nitric oxide synthesis isoforms and glomerular hyperfiltration in early diabetic nephropathy, J. Am. Soc. Nephrol. 11 (2000) 71 – 79. [5] K. Hiragushi, H. Sugimoto, K. Shikata, et al., Nitric oxide system is involved in glomerular hyperfiltration in Japanese normo- and micro-albuminuric patients with type 2 diabetes, Diabetes Res. Clin. Prac. 53 (2001) 149 – 159. [6] L.N. Wang, Z. Tang, S. Ichiyu, et al., Effect of the PGI2 analog beraprost sodium on glomerular prostanoid synthesis in rats with streptozotocin-induced diabetes, Nephron 73 (1996) 637 – 643. [7] J.M. Armstrong, N. Lattimer, S. Moncada, et al., Comparison of the vasodepressor effects of prostacyclin and 6-oxoprostaglandin F1a with those of prostacyclin E2 in rats and rabbits, Br. J. Pharmacol. 62 (1987) 125 – 130. [8] B.J.R. Whittle, S. Moncada, J.R. Vane, Comparison of the effects of prostacyclin (PGI2), prostaglandin E1 and D2 on platlet aggregation in different species, Prostaglandins 16 (1978) 373 – 388. [9] S. Moncada, Biology and therapeutic potential of prostacyclin, Stroke 14 (1983) 157 – 168. [10] J.E. Barker, Y.S. Bakhle, J. Anderson, et al., Reciprocal inhibition of nitric oxide and prostacyclin synthesis in human saphenous vein, Br. J. Pharmacol. 118 (1996) 643 – 648. [11] P. Marotta, L. Sautebin, M. DiRosa, Modulation of induction of nitric oxide synthase by eicosanoids in the murine macrophage cell line J774, Br. J. Pharmacol. 107 (1992) 640 – 641. [12] F. D’Acquisto, L. Sautebin, T. Iuvone, Prostaglandin prevent inducible nitric oxide synthase protein expression by inhibiting nuclear factor-kB activation in J774 macrophages, FEBS Lett. 440 (1998) 76 – 80. [13] D. Salvemini, M.G. Currie, V. Mollace, Nitric oxide-mediated cyclooxygenase activation. A key event in the antiplatelet effects of nitrovasodilators, J. Clin. Invest. 97 (1996) 2562 – 2568. [14] T. Furuta, T. Saito, T. Ootaka, et al., The role of macrophages in diabetic glomerulosclerosis, Am. J. Kidney Dis. 21 (1993) 480 – 485. [15] K. Hirata, K. Shikata, H. Makino, et al., Increased expression of selectins in kidneys of patients with diabetic nephropathy, Diabetologia 41 (1998) 185 – 192. [16] C. Sassy-Prigent, D. Heudes, C. Mandet, et al., Early glomerular macrophage recruitment in streptozotocin-induced diabetic rats, Diabetes 49 (2000) 466 – 475. [17] H. Sugimoto, K. Shikata, H. Makino, et al., Increased expression of intercellular adhesion molecule-1 (ICAM-1) in diabetic rat glomeruli, Diabetes 46 (1997) 2075 – 2081.
T. Yamashita et al. / Diabetes Research and Clinical Practice 57 (2002) 149–161 [18] H. Sugimoto, K. Shikata, H. Makino, et al., Increased gene expression of insulin-like growth factor(IGF-1) receptor in experimental diabetic rat glomeruli, Nephron 72 (1996) 648 – 653. [19] T. Umetsu, T. Murata, Y. Tanaka, et al., Antithrombotic effect of TRK-100, a novel, stable PGI2 analogue, Jpn. J. Pharmacol. 43 (1987) 81 –90. [20] M. Kushiro, K. Shikata, H. Makino, et al., Therapeutic effects of prostacyclin analog on crescentic glomerulonephritis of rat, Kidney Int. 53 (1998) 1314 –1320. [21] K. Kimura, R. Nagai, T. Sakai, et al., Diversity and variability of smooth muscle phenotypes of renal arterioles as revealed by myosin isoform expression, Kidney Int. 48 (1995) 372 –382. [22] R. Kikkawa, M. Haneda, K. Sakamoto, et al., Antagonist for atrial natriuretic peptide receptors ameliorates glomerular hyperfiltration in diabetic rats, Biochem. Biophys. Res. Commun. 193 (1993) 700 –705. [23] P.A. Craven, M.A. Caines, F.R. DeRubertis, Sequential alterations in glomerular prostaglandin and thromboxane synthesis in diabetic rats: relationship to the hyperfiltration of early diabetes, Metabolism 36 (1987) 95 –103. [24] P.K. Jensen, K. Steven, H.H. Parving, et al., Effect of indomethacin on glomerular hemodynamics in experimental diabetes, Kidney Int. 29 (1986) 490 –495. [25] R. Komer, T.J. Allen, M.E. Cooper, Role of endothelium-derived nitric oxide in the pathogenesis of the renal hemodynanic changes of experimental diabetes, Diabetes 43 (1994) 1190 – 1197. [26] N. Bank, H.S. Aynedjian, Role of EDRF (nitric oxide) in diabetic renal hyperfiltration, Kidney Int. 43 (1993) 1306 – 1312. [27] S. Arima, Y.L. Ren, L.A. Juncos, et al., Glomerular prostaglandins modulate vascular reactivity of the downstream efferent arterioles, Kidney Int. 45 (1994) 50 –658.
161
[28] H. Oida, T. Namba, Y. Sugimoto, et al., In situ hybridization studies of prostacyclin receptor mRNA expression in various mouse organs, Br. J. Pharmacol. 116 (1995) 2828 – 2837. [29] M. Komhoff, B. Lesener, K. Nakao, et al., Localization of the prostacyclin receptor in human kidney, Kidney Int. 54 (1998) 1899 – 1908. [30] S. Ito, S. Arima, Y.L. Ren, et al., Endothelium-derived relaxing factor/nitric oxide modulates angiotensin II action in the isolated microperfused rabbit afferent but not efferent arteriole, J. Clin. Invest. 91 (1993) 2012 – 2019. [31] S. Ito, C.S. Johnson, O.A. Carretero, Modulation of angiotensin II-induced vasoconstriction by endotheliumderived relaxing factor in the isolated microperfused rabbit afferent arteriole, J. Clin. Invest. 87 (1991) 1656 – 1663. [32] T. Nagel, N. Resnick, W.J. Atkinson, et al., Shear stress selectively upregulates intercellular adhesion molecule-1 expression in cultured human vascular endothelial cells, J. Clin. Invest. 94 (1994) 885 – 891. [33] D.E. Stauton, S.D. Marlin, C. Stratowa, et al., Primary structure of ICAM-1 demonstrates interaction between members of the immunoglobulin and integrin supergene families, Cell 52 (1988) 925 – 933. [34] D.E. Staunton, S.D. Marlin, C. Stratowa, et al., Primary structure of ICAM-1 demonstrates interaction between members of the immunogloblin and integlin supergene families, Cell 52 (1988) 925 – 933. [35] M.W. Radomski, R.M.J. Palmer, S. Moncada, The antiaggregating properties of vascular endothelium: interactions between prostacyclin and nitic oxide, Br. J. Pharmacol. 92 (1987) 639 – 646. [36] S.L. Kunkel, S.W. Chensue, The role of arachidonic acid metabolism in mononuclear phagocytic cell interactions, Int. J. Dermatol. 25 (1986) 83 – 89. [37] J. Marcinkiewicz, B.M. Chain, Differential cytokine regulation by eicosanoids in T cells primed by contact sensitization with TNP, Cell Immunol. 149 (1993) 303 – 314.