Role of Angiotensin II, Endothelin-1, and Nitric Oxide in HgCl2-Induced Acute Renal Failure

TOXICOLOGY AND APPLIED PHARMACOLOGY ARTICLE NO.

152, 315–326 (1998)

TO988459

Role of Angiotensin II, Endothelin-1, and Nitric Oxide in HgCl2-Induced Acute Renal Failure Hiroyuki Yanagisawa, Makoto Nodera, Yasuhiko Umemori, Yukio Shimoguchi,* and Osamu Wada Department of Hygiene and Preventive Medicine, Faculty of Medicine, Saitama Medical School 38 Morohongo, Moroyama, Iruma-gun, Saitama 350-0495; Japan; and *Teijin Bio Laboratories, Inc., Higashimatsuyama, Saitama 355, Japan Received January 13, 1998; accepted April 23, 1998

Role of Angiotensin II, Endothelin-1, and Nitric Oxide in HgCl2-Induced Acute Renal Failure. Yanagisawa, H., Nodera, M., Umemori, Y., Shimoguchi, Y., and Wada, O. (1998). Toxicol. Appl. Pharmacol. 152, 315–326. To elucidate the mechanisms underlying the development of HgCl2-induced acute renal failure (ARF), we examined the expression of endothelin (ET)-1, endothelial (e) nitric oxide synthase (NOS) and inducible (i) NOS, and a role of angiotensin II (ANG II) and tumor necrosis factor (TNF) in glomeruli and cortices from rats at 20 h after exposure of HgCl2. Prepro-ET-1 and iNOS mRNA were significantly increased in glomeruli and cortices from rats with HgCl2-induced ARF. However, eNOS mRNA was markedly decreased in glomeruli of rats with HgCl2-induced ARF. Blockade of the action of endogenous ANG II with TCV-116, an ANG II receptor type 1 antagonist, or prior administration of TNF antibody (Ab) neutralizing TNF bioactivity or aminoguanidine, an iNOS inhibitor, substantially suppressed the increase in the expression of prepro-ET-1 or iNOS mRNA seen in rats with HgCl2induced ARF. Both TCV-116 and TNF Ab had no effects on the expression of eNOS mRNA. The abundance of ET-1, iNOS, and eNOS proteins was paralleled by the magnitude of each mRNA expression. Additionally, the aggravation of blood urea nitrogen and serum Cr observed in rats with HgCl2-induced ARF were significantly ameliorated together with the alleviation of proximal tubule epithelial cell injury when the expression of prepro-ET-1 or iNOS mRNA was blunted by prior administration of TCV-116 or prior injection of TNF Ab or aminoguanidine. These observations indicate that ANG II, ET-1, and NO may play an important role in the progression of HgCl2-induced ARF through the acceleration of proximal tubule epithelial cell injury and the deterioration of glomerular hemodynamics. In HgCl2-induced ARF, the gene expression of ET-1 or iNOS is at least in part up-regulated at the transcription level by endogenous ANG II or TNF. © 1998 Academic Press

The vasoconstrictive peptides, angiotensin (ANG) II and endothelin (ET)-1, and the vasodilatory gas, nitric oxide (NO), contribute to the regulation of glomerular hemodynamics interacting in a paracrine and/or autocrine fashion (Baylis and Qiu, 1996; Brenner et al., 1980; Kon and Badr, 1991; Nicola et al., 1992). ET-1 is synthesized by enzymatic cleavage of the two immature prohormones, prepro-ET-1 and pro-ET-1 (Kon

and Badr, 1991; Marsen et al., 1994). ET-1 is considered to be a more powerful vasoconstrictor than ANG II (Kon and Badr, 1991; Marsen et al., 1994). On the other hand, the vasodilator NO is biosynthesized from the terminal guanidino nitrogen of L-arginine through the oxidative process due to NO synthase (NOS) proteins (Baylis and Qiu, 1996; Ketteler et al., 1994; Nicola et al., 1992; Raij and Baylis, 1995). Three isoforms of the enzymes, brain (b) NOS, inducible (i) NOS, and endothelial (e) NOS proteins were initially identified in the central nervous system, macrophages, and vascular endothelial cells, respectively (Baylis and Qiu, 1996; Ketteler et al., 1994; Raij and Baylis, 1995). Recently, the induction of iNOS protein by cytokines such as tumor necrosis factor (TNF) is known to relate to cell and/or tissue injury via the cytotoxic effect of the free radical, NO, produced in large quantities by this enzymes (Cross et al., 1994; Ketteler et al., 1994; Pinsky et al., 1995). Mercury chloride (HgCl2) causes acute renal failure (ARF) by severely damaging proximal tubule epithelial cells. No significant pathological findings are observed in glomeruli. The onset of HgCl2-induced ARF results from tubular obstruction and backleak due to acute tubular necrosis (ATN) (Hostetter et al., 1983; Kreisberg et al., 1983). Subsequently, these tubular effects, particularly backleak, play a fundamental role in the development of this ARF (Hostetter et al., 1983; Conger and Falk, 1986). On the other hand, an increment in the ratio of afferent arteriole resistance (RA) to efferent arteriole resistance (RE) and a decrement in net effective glomerular filtration pressure (Pnet) and glomerular capillary hydraulic pressure (Pg) are seen at 16 to 28 h after administration of HgCl2 (Wolfert et al., 1987). Bowman’s space pressure is depressed below the control level by this time (Wolfert et al., 1987). The decrement in Pnet and Pg may be caused by decreased glomerular capillary plasma flow (QA) due to the increased RA (Wolfert et al., 1987) Accordingly, changes in the circumstances at the glomerulus level causing such vascular effects may also have a role in the progression or maintenance of HgCl2-induced ARF. The present study was designed to explore if the vasoactive substances such as ANG II, ET-1, and NO may be involved in the progression or maintenance of HgCl2-induced ARF by affecting the vascular and/or tubular effects mentioned above. We, therefore, examined the relative levels of prepro-ET-1,

315

0041-008X/98 $25.00 Copyright © 1998 by Academic Press All rights of reproduction in any form reserved.

316

YANAGISAWA ET AL.

eNOS, and iNOS mRNA, the expression of ET-1, eNOS, and iNOS proteins, and the levels of blood urea nitrogen (BUN) and serum creatinine (Cr) in control and HgCl2-induced ARF rats. We, moreover, examined a role of ANG II and TNF in HgCl2-induced ARF using the ANG II receptor type 1 (AT1 receptor) antagonist, TCV-116 and anti-TNF antibody (Ab) neutralizing TNF bioactivity. Concomitantly, we measured renal renin and TNF mRNA, the activity of the systemic renin–angiotensin system, (RAS) and levels of serum TNF in control and HgCl2-induced ARF rats. MATERIALS AND METHODS Chemicals and reagents. TCV-116, an AT1 receptor antagonist, was kindly provided by Takeda Chemical Industries (Osaka, Japan). Polyclonal rabbit anti-TNF-a Ab neutralizing TNF-a bioactivity was purchased from Genzyme Corp. (Cambridge, MA). Monoclonal mouse Ab recognizing the cytoplasmic antigen ED1 of macrophages and monocytes (monoclonal Ab ED1) was supplied from BMA Biomedicals Ltd. (Augst, Switzerland). Polyclonal rabbit anti-ET-1 Ab was obtained from Immuno-Biological Laboratories (Gunma, Japan). Rabbit antiserum against eNOS or iNOS protein was supplied from Affinity Bioreagents, Inc. (Neshanic Station, NJ). Aminoguanidine hemisulfate salt, an iNOS inhibitor, was purchased from Sigma (St. Louis, MO). Block Ace, a blocking reagent was obtained from Dainippon Pharmaceutical Corp., Ltd. (Osaka, Japan). RNA polymerase chain reaction (PCR) kits were purchased from Takara Biochemicals (Shiga, Japan). Experimental protocol. Female Sprague–Dawley rats (Japan Biochem. Supple. Center, Tokyo, Japan) weighing approximately 200 g were divided into three groups (Fig. 1). Group 1 rats had free access to 0.5% ethanol (diluted with tap water) alone or 0.5% ethanol containing 50 mg/liter of the AT1 receptor antagonist TCV-116 for 5 days (Kim et al., 1994; Nio et al., 1995). As a result, the animals (n 5 30) were pretreated with 7.67 6 0.12 mg/kg TCV-116 per day. Group 2 rats were ip injected normal rabbit serum or 100 mg of TNF Ab twice at 24-h intervals before euthanasia. Group 3 rats had five sc injections (once a day) of saline alone or 400 mg/kg of the iNOS inhibitor aminoguanidine dissolved in saline (Cross et al., 1994). The last injection of aminoguanidine was carried out 24 h before euthanasia. The animals of each group were then given ip saline alone or 7.5 mg/kg of HgCl2 dissolved in saline 20 h before euthanasia. The control and HgCl2-induced ARF rats prepared were used for the measurements of glomerular prepro-ET-1 and NOS mRNA, BUN, and serum Cr and for the immunohistochemical analysis of their kidneys. Blood collection and isolated glomeruli. The abdominal cavity of rats was opened under light ether anesthesia. Subsequently to blood collection for the measurements of circulating renin and ANG I converting enzyme (ACE) activities and of ANG II, TNF, BUN, and serum Cr concentrations (Girndt et al., 1995; Yanagisawa et al., 1994), both kidneys were thoroughly perfused with ice-cold Ca21 and Mg21-free Hanks’ balanced salt solution (HBSS) injected just above the bifurcation of the aorta (Yanagisawa et al., 1990, 1994). The kidneys were immediately removed, decapsulated, and dissected into cortices and medullas. Glomeruli were then obtained from the cortices by differential sieve techniques (mesh sizes 250, 150, and 75 mg) (Yanagisawa et al., 1990, 1994), washed twice, and resuspended in ice-cold Ca21 and Mg21free HBSS. The preparations were confirmed to consist of over 90% isolated glomeruli by light microscopy. Extraction of RNA. Total RNA was extracted from cortices or glomeruli with the method of guanidinium thiocyanate (Chomczynski and Sacchi, 1987). In brief, cortices or glomeruli were homogenized in 1 ml of ice-cold ISOGEN (Nippon Gene, Toyama, Japan). Total RNA from the preparations was precipitated with 500 ml of isopropanol. The RNA pellets were then washed with 1 ml of 75% ethanol, dried, and dissolved in RNase-free distilled water. The quantitative analysis of total RNA was carried out at 260 and 280 nm using a

FIG. 1. Schema of experimental protocols. ip, intraperitoneally; NRS, normal rabbit serum; TNF, tumor necrosis factor; Ab, antibody; sc, subcutaneously.

Beckman UV 640 spectrophotometer. The OD260/OD280 ratio of all RNA samples used was over 1.8. cDNA synthesis and amplification. Total RNA from cortices or glomeruli was reverse transcribed to first-strand cDNA in reverse transcription buffer consisting of 10 mM Tris–HCl, 50 mM KCl, 5 mM MgCl2, 1 mM deoxynucleotide triphosphate (dNTP) mixture, 1 U/ml RNase inhibitor, 0.25 U/ml avian myeloblastosis virus (AMV) reverse transcriptase, and 2.5 mM random primers under the particular incubation conditions (30°C for 10 min and 42°C for 30 min) controlled by a Programmed Thermal Cycler System (Astec, Tokyo, Japan). Following the incubation period, the reaction mixture was heated at 99°C for 5 min and immediately chilled to 4°C. Using established specific oligonucleotide primers shown in Table 1 (Macica et al., 1994; Morrissey et al., 1994; Pelayo et al., 1994; Rocco et al., 1992; Tracey et al., 1994; Ujiie et al., 1992; Xie et al., 1992), PCR amplification was done for renin, TNF, prepro-ET-1, iNOS, and glyceraldehyde-3phosphate dehydrogenase (GAPDH) cDNA from cortices or for prepro-ET-1, eNOS, iNOS, and GAPDH cDNA from glomeruli. Briefly, renin, TNF, preproET-1, eNOS, or iNOS cDNA was amplified with 35 cycles in the reaction mixture consisting of PCR amplification buffer (10 mM Tris–HCl, 50 mM KCl, 4 mM MgCl2, 200 mM dNTP mixture, and 0.025 U/ml Takara Taq DNA polymerase), 100 mM oligonucleotide primers (sense and antisense), and reverse-transcribed cDNA from 1 mg of total RNA under the following conditions: denaturation at 94°C for 1 min, annealing at the indicated temperature (Table 1) for 1 min, and extension at 72°C for 2 min. The housekeeping gene GAPDH cDNA was coamplified with 35 cycles under the same amplification protocol using the reaction mixture composed of PCR amplification buffer, 100 mM oligonucleotide primers (sense and antisense), and reverse transcribed cDNA from 0.5 mg of total RNA. There was a linear increase in each PCR product at least up to 40 cycles. Quantitative analysis of PCR products. PCR amplification products were separated by means of a 2.0% agarose gel with ethidium bromide (0.5 mg/ml).

317

HgCl2-INDUCED ACUTE RENAL FAILURE

TABLE 1 Specific Amplification Primers

Gene Renin TNF Prepro-ET-1 eNOS iNOS GAPDH

Oligonucleotide sequence Sense Antisense Sense Antisense Sense Antisense Sense Antisense Sense Antisense Sense Antisense

59-CTGTGGGTGGAATCATTGTGACAC-39 59-CGTAGTGAAAGTTGCCCTGGTAT-39 59-ATGAGCACAGAAAGCATGATC-39 59-TACAGGCTTGTCACTCGAATT-39 59-CGTTGCTCCTGCTCCTCCTTGATGG-39 59-AAGTCCCAGCCAGCATGGAGAGCG-39 59-GCAGAATTCTGTTTGGCCGAGTCCTC-39 59-AATGGATCCCTCCTGCAAGGAAAAGC-39 59-GTGTTCCACCAGGAGATGTTG-39 59-CTCCTGCCCACTGAGTTCGTC-39 59-AATGCATCCTGCACCACCAA-39 59-GTAGCCATATTCATTGTCATA-39

Annealing temperature (°C)

Product size (bp)

60

276

Pelayo et al. (1994)

58

276

Macica et al. (1994)

60

543

Ujiie et al. (1992)

60

341

Tracey et al. (1994)

60

576

60

515

Morrissey et al. (1994) Xie et al. (1992) Rocco et al. (1992) Xie et al. (1992)

Reference

Note. bp, base pair; TNF, tumor necrosis factor; ET, endothelin; eNOS, endothelial nitric oxide synthase; iNOS, inducible nitric oxide synthase; GAPDH, glyceraldehyde-3-phosphate dehydrogenase.

The gel visualized with UV light was photographed with Polaroid Type 667 positive film. The intensity of bands was measured by densitometry for quantification. The relative levels of renin, TNF, prepro-ET-1, or NOS mRNA were determined by normalizing the quantity of each cDNA to the amount of GAPDH cDNA. Immunohistochemistry. For immunohistochemical studies, commercially available monoclonal Ab ED1 and antiserum against ET-1, eNOS, or iNOS protein was used. Monoclonal Ab ED1 is an available tool that specifically detects macrophages and monocytes (Diamond et al., 1994; Dijkstra et al., 1985). Antiserum against ET-1 is a polyclonal Ab purified with affinity columns and specific for ET-1. The cross-reactivities of antiserum against ET-1, eNOS, or iNOS protein when determined by immunoassay or Western blot analysis are as follows: anti-ET-1 Ab, 1% for ET-2 and 0.4% for ET-3; anti-eNOS Ab, no cross-reactivity with bNOS and iNOS proteins; and antiiNOS Ab, no cross-reactivity with bNOS and eNOS proteins. Longitudinally sliced preparations (2–3 mm) of perfused kidneys were fixed at 4°C overnight with 4% paraformaldehyde dissolved in phosphate-buffered saline and mounted in paraffin blocks. The preparations were cut at a thickness of 4 mm and placed on microscopic slides. To remove paraffin the tissue sections were incubated for 3 min each serially in xylene (33), 99.5% ethanol (23) and 70% ethanol (13). The sections were then washed in a gently running stream of tap water for 3 min (13) and immersed in distilled water for 1 min (23). Endogenous peroxidase activity of the kidney preparations was inactivated by 0.3% hydrogen peroxide treatment. The kidney sections were then blocked for 10 min in a blocking solution (20% Block Ace); washed twice for 5 min each in TBS consisting of 50 mM Tris–HCl, pH 7.8, 0.05% Tween 20, 0.3 M NaCl and 0.02% NaN3; and incubated for 2 h with monoclonal Ab ED1 (1:10, 1:25, 1:50, and 1:100 dilution) or with antiserum against ET-1 (1:50 dilution), eNOS (1:50 dilution), or iNOS (1:100 dilution) protein in TBS containing 0.1% bovine serum albumin (BSA). Each dilution of pre-immune serum was used as negative controls. After washing with TBS three times for 5 min each, the sections were further incubated for 20 min with biotinconjugated second Ab (LSAB 2 Kit, Dako Corp.) in TBS containing 0.1% BSA. The sections were thoroughly washed with TBS and reacted with horseradish peroxidase-labeled streptavidin. Staining was then carried out in 0.3% diaminobenzidine. Following nucleus stain with hematoxylin the sections were mounted with 2 mg/ml p-phenylenediamine in 50% glycerol. Statistical analysis. All data reported represent means 6 SE. Intergroup comparisons were routinely evaluated using unpaired Student’s t-test or twoway analysis of variance (ANOVA) with Student’s t-test. Intragroup comparisons were performed by Duncan’s multiple range comparison test for the time

course of renin and TNF mRNA expression. Differences were considered significant when p , 0.05.

RESULTS

HgCl2-Induced ARF To evaluate the effect of HgCl2 on the induction of ATN, H & E stain was done using the kidney sections of salinetreated control rats and rats at 1, 3, 6, 12, and 20 h after ip administration of 7.5 mg/kg HgCl2 dissolved in saline. As shown in Figs. 2A–2C, H & E stain demonstrated proximal tubular necrosis in the kidneys of rats at 6 to 20 h after treatment of HgCl2. Glomeruli had no significant pathological changes by exposure of HgCl2. The degree of the tubular injury was augmented in rats at 20 vs 6 h after administration of HgCl2. However, there were no significant lesions in the kidneys of rats at 1 to 3 h after treatment of HgCl2 (not shown). These observations indicate that in this setting ATN develops at latest by 6 hours after exposure of HgCl2. Additionally, we explored the inhibitory effect of the AT1 receptor antagonist, TCV-116 or TNF Ab or the iNOS inhibitor aminoguanidine on the development of ATN using the kidney sections stained with H & E of HgCl2-induced ARF rats that were pretreated or not with their agents. As seen in Figs. 2D–2I, prior administration of TCV-16, TNF Ab, or aminoguanidine alleviated the degree of proximal tubule epithelial cell injury and reduced the extent of ATN. There were no significant morphologic changes in glomeruli by use of their agents. Levels of Renin and TNF mRNA To examine the effect of HgCl2 on the expression of renal renin and TNF mRNA, we evaluated their mRNA expression in renal cortices of saline-treated control rats and rats at 1, 3, 6, 12, and 20 h after ip treatment of 7.5 mg/kg HgCl2 dissolved in saline.

318

YANAGISAWA ET AL.

FIG. 2. Photographs of the kidney sections stained with hematoxylin-eosin. The photographs A–C present the kidney sections which show the development of ARF after exposure of HgCl2. (A) Saline-treated control. (B) ARF at 6 hours after exposure of HgCl2. (C) ARF at 20 hours after exposure of HgCl2. Microscopic magnification for A–C: 2003. The photographs D–I present the kidney sections of ARF at 20 hours after exposure of HgCl2 (see Fig. 1). (D) 0.5% ethanol-pretreated ARF as a control. (E) TCV-116-pretreated ARF. (F) NRS-pretreated ARF as a control. (G) TNF Ab-pretreated ARF. (H) saline-pretreated ARF as a control. (I) aminoguanidine-pretreated ARF. Microscopic magnification for D–I: 1003. ARF, acute renal failure; NRS, normal rabbit serum; TNF, tumor necrosis factor; Ab, antibody.

Table 2 shows data on the time course of renin and TNF mRNA expression (Fig. 3) in renal cortices from rats (n 5 4). Renin and TNF mRNA were significantly increased by 40 and 70%, respectively, above the basal value at 12 h after admin-

istration of HgCl2. Subsequently, the significant increase in their mRNA was at least observed up to 20 h after treatment of HgCl2. No significant amounts of renin and TNF mRNA were detected in glomeruli before and after exposure of HgCl2.

319

HgCl2-INDUCED ACUTE RENAL FAILURE

FIG. 2—Continued

Renin and TNF mRNA may originate in juxtaglomerular cells of afferent arterioles (Hackenthal et al., 1987) and proximal tubule cells (Ong and Fine, 1994), respectively. Renin and ACE Activities and Levels of ANG II and TNF Table 3 shows data on the activities of plasma renin and serum ACE and the levels of plasma ANG II and serum TNF

in saline-treated control rats (n 5 5 or 8) and rats with HgCl2-induced ARF (n 5 5 or 8). ARF rats had significantly enhanced activities of circulating renin and ACE and markedly increased levels of plasma ANG II when compared to control rats. This indicates an increase in the activity of the systemic RAS in the HgCl2-induced ARF setting. Similarly, circulating levels of TNF were significantly greater in ARF rats than in control rats.

TABLE 2 Time Course of Renin and Tumor Necrosis Factor mRNA Expression in Renal Cortices from Rats that Were Given HgCl2 ip Saline

HgCl2

mRNA

Basal

1h

3h

6h

12 h

20 h

Renin TNF

2.48 6 0.10 5.72 6 0.50

2.54 6 0.09 5.71 6 0.47

2.46 6 0.06 5.21 6 0.24

2.78 6 0.05 6.98 6 0.58

3.46* 6 0.38 9.71* 6 0.91

3.10* 6 0.08 7.91* 6 1.01

Note. Time course was examined for renin and TNF mRNA expression in renal cortices using saline-treated control rats and rats at 1, 3, 6, 12, and 20 h after ip treatment of 7.5 mg/kg HgCl2 dissolved in saline. Data reported represent means (arbitrary units) 6 SE of values obtained from four separate cortical preparations. Intragroup comparisons were done using Duncan’s multiple range comparison test. *p , 0.05 compared to each basal value. TNF, tumor necrosis factor.

320

YANAGISAWA ET AL.

only the activity of iNOS protein (Cross et al., 1994) but also the expression of iNOS mRNA. Prior administration of this drug markedly decreased iNOS mRNA in glomeruli and cortices of control rats and rats with ARF. In particular, the enhanced expression of iNOS mRNA seen in glomeruli of ARF rats was dramatically restored to levels comparable to that of vehicle-pretreated control rats. Immunohistochemical Study

FIG. 3. Renin, TNF, and GAPDH mRNA expression in renal cortices from saline-treated control rats (C) and rats at 12 h after exposure of HgCl2 (H) by means of RT-PCR techniques. TNF, tumor necrosis factor; GAPDH, glyceraldehyde-3-phosphate dehydrogenase; RT-PCR, polymerase chain reaction coupled to reverse transcription.

Levels of Prepro-ET-1 and NOS mRNA Table 4 shows data on the relative levels of prepro-ET-1, eNOS, and iNOS mRNA in glomeruli (Fig. 4) and cortices from saline-treated control rats (n 5 5) and rats with HgCl2induced ARF (n 5 5) that were pretreated or not with the AT1 receptor antagonist TCV-116 or TNF Ab. Levels of prepro-ET-1 mRNA were significantly greater in ARF rats than in control rats. Pretreatment of rats with TCV116 significantly decreased prepro-ET-1 mRNA in ARF rats. The mRNA, however, was still increased substantially compared to control rats. On the other hand, TNF Ab had no effects on the expression of prepro-ET-1 mRNA in both groups of rats. Levels of eNOS mRNA were significantly reduced in ARF rats relative to control rats. There were no significant effects of TCV-116 and TNF Ab on the expression of eNOS mRNA in both groups of rats. Levels of iNOS mRNA were markedly increased in ARF vs control rats. Interestingly, iNOS mRNA was clearly detected in glomeruli of control rats while iNOS protein is known to be an inducible enzyme. The expression of iNOS mRNA in cortices of control rats may be derived from its glomerular iNOS mRNA expression. TCV-116 did not affect the expresion of iNOS mRNA in glomeruli and cortices from both groups of rats. By contrast, the increased iNOS mRNA observed in ARF rats was significantly decreased by prior administration of TNF Ab, but it was still augmented substantially compared to control rats. Table 5 shows the effect of the iNOS inhibitor aminoguanidine on the expression of iNOS mRNA in glomeruli and cortices from saline-treated control rats (n 5 5) and rats with HgCl2-induced ARF (n 5 5). Aminoguanidine affected not

Figures 5A–5F show the results of immunohistochemical studies on the expression of ET-1 (Figs. 5A and 5B), eNOS (Figs. 5C and 5D), and iNOS (Figs. 5E and 5F) proteins in glomeruli of the kidneys from saline-treated control rats and rats with HgCl2-induced ARF. The expression of ET-1 and iNOS protein was observed in glomerular epithelial cells and, to a lesser extent, in glomerular mesangial cells of ARF rats. Immunoreactive eNOS protein existed in the glomerular tuft of control rats. The immunohistochemical observation of ET-1, iNOS, or eNOS protein was essentially consistent with that of Hattori et al. (1997), Buttery et al. (1994) or Bosse et al. (1994). Both ET-1 and iNOS protein were markedly enhanced in ARF rats relative to control rats. By contrast, eNOS protein was remarkably reduced in ARF rats compared to control rats. Immunodetectable amounts of ET-1 and iNOS protein were not present in control rats. Inversely, immunoidentifiable amounts of eNOS protein did not exist in ARF rats. Figures 5G–5J present the results of immunohistochemical studies on the expression of ET-1 (Figs. 5G and 5H) and iNOS (Figs. 5I and 5J) protein in cortices of the kidneys from saline-treated control rats and rats with HgCl2-induced ARF. Immunodetectable quantity of ET-1 and iNOS protein was present in damaged proximal tubule epithelial cells of ARF rats. As with glomeruli, these were substantially increased in ARF rats relative to control rats. Both ET-1 and iNOS protein were not detectable in control rats. Additionally, ED1-positive cells were not detected in both groups of rats, although low to high concentrations of monoclonal Ab ED1 were utilized.

TABLE 3 Circulating Renin and Angiotensin I Converting Enzyme Activities and Circulating Levels of Angiotensin II and Tumor Necrosis Factor in Control Rats and Rats with HgCl2-Induced Acute Renal Failure

Model

Renin (ng/ml/h) (n 5 8)

ACE (nmol/ml/min) (n 5 8)

ANG II (pg/ml) (n 5 8)

TNF (pg/ml) (n 5 5)

CTL ARF

21.3 6 3.7 59.3 6 5.0*

24.6 6 0.9 34.6 6 1.6*

11.3 6 3.0 63.6 6 15.7*

1.3 6 0.4 15.1 6 1.9*

Note. Values are means 6 SE from five or eight rats. Intergroup comparisons were done using unpaired Student’s t-test. *p , 0.05 compared to each control value. ACE, angiotensin I converting enzyme; ANG II, angiotensin II; TNF, tumor necrosis factor; CTL, control; ARF, acute renal failure.

321

HgCl2-INDUCED ACUTE RENAL FAILURE

TABLE 4 Relative Levels of Preproendothelin-1, Endothelial Nitric Oxide Synthase, and Inducible Nitric Oxide Synthase mRNA in Glomeruli and Cortices from Control Rats and Rats with HgCl2-Induced Acute Renal Failure that Were Pretreated or Not with TCV-116 or Tumor Necrosis Factor Antibody 0.5% Ethanol

TCV-116

NRS

TNF Ab

Segment

mRNA

CTL

ARF

CTL

ARF

CTL

ARF

CTL

ARF

Glomerulus

Prepro-ET-1 eNOS iNOS Prepro-ET-1 iNOS

4.48 6 0.35 4.75 6 0.71 1.73 6 0.41 2.97 6 0.32 0.39 6 0.08

9.85* 6 0.90 2.58* 6 0.54 17.17* 6 1.42 13.41* 6 1.28 28.35* 6 4.81

4.45 6 0.24 3.95 6 0.48 1.50 6 0.29 3.26 6 0.45 0.46 6 0.14

6.74*,† 6 0.78 2.67* 6 0.12 21.26* 6 1.22 7.38*,† 6 0.69 32.46* 6 3.75

4.57 6 0.92 4.27 6 0.08 1.83 6 0.77 2.76 6 0.41 0.33 6 0.06

9.87* 6 1.81 2.38* 6 0.60 18.88* 6 2.96 14.93* 6 1.86 34.72* 6 5.14

4.39 6 0.95 3.68 6 0.29 1.58 6 0.38 3.32 6 0.57 0.42 6 0.11

9.25* 6 1.83 1.97* 6 0.33 8.55*,† 6 3.24 13.08* 6 1.77 12.51*,† 6 2.36

Cortex

Note. Data reported represent means (arbitrary units) 6 SE of values obtained from five separate prepartions. Intergroup comparisons were done using two-way ANOVA with Student’s t-test. *p , 0.05 compared to each control value. †p , 0.05 compared to each vehicle-pretreated value. NRS, normal rabbit serum; TNF Ab, tumor necrosis factor antibody; CTL, control; ARF, acute renal failure; Prepro-ET-1, preproendothelin-1; eNOS, endothelial nitric oxide synthase; iNOS, inducible nitric oxide synthase.

Levels of BUN and Serum Cr Table 6 shows data on the levels of BUN and serum Cr in saline-treated control rats (n 5 10) and rats with HgCl2induced ARF (n 5 10) that were pretreated or not with the AT1 receptor antagonist TCV-116 or TNF Ab or the iNOS inhibitor aminoguanidine. Levels of BUN and serum Cr were markedly increased in ARF rats compared to control rats. Pretreatment of rats with TCV-116, TNF Ab, or aminoguanidine significantly decreased the increased levels of BUN and serum Cr seen in ARF rats. Compared to TNF Ab-pretreated ARF rats, levels of BUN and serum Cr were further decreased in aminoguanidine-pretreated ARF rats. This may be due to nearly complete inhibition of the enhanced expression of iNOS mRNA by aminoguanidine (Table 5). In addition, there was a parallel relationship between the degree of inhibition of proximal tubule epithelial cell injury

(Figs. 2G and 2I) and the decreased levels of BUN and serum Cr by pretreatment of TNF Ab or aminoguanidine in ARF rats. DISCUSSION

The present study demonstrated increased levels of preproET-1 and iNOS mRNA in cortices of rats with HgCl2-induced ARF compared to control rats. Also, immunohistochemical techniques revealed increased expression of ET-1 and iNOS protein in proximal tubule epithelial cells of HgCl2-induced ARF vs control rats. There was a parallel relationship between the expression of their mRNA and protein in control and HgCl2-induced ARF rats. Blockade of the biological action of endogenous ANG II and inhibition of an increase in preproET-1 mRNA with the AT1 receptor antagonist TCV-116 mitigated the degree of proximal tubule epithelial cell injury together with a decrease in elevated levels of BUN and serum Cr in rats with HgCl2-induced ARF. Similarly, neutralization

TABLE 5 Relative Levels of Inducible Nitric Oxide Synthase mRNA in Glomeruli and Cortices from Control Rats and Rats with HgCl2Induced Acute Renal Failure that Were Pretreated or Not with Aminoguanidine Saline Segment

CTL

Glomerulus 1.90 6 0.56 Cortex 0.31 6 0.04

FIG. 4. Prepro-ET-1, eNOS, iNOS, and GAPDH mRNA expression in glomeruli from saline-treated control (C) and HgCl2-induced ARF (H) rats by means of RT-PCR techniques. ET, endothelin; eNOS, endothelial nitric oxide synthase; iNOS, inducible nitric oxide synthase; GAPDH, glyceraldehyde-3phosphate dehydrogenase; RT-PCR, polymerase chain reaction coupled to reverse transcription; ARF, acute renal failure.

Aminoguanidine ARF

CTL

ARF

19.10* 6 3.37 26.98* 6 4.33

0.54† 6 0.05 Not detected

1.90*,† 6 0.51 0.76† 6 0.29

Note. Data reported represent means (arbitrary units) 6 SE of values obtained from five separate preparations. The limit of detection of iNOS mRNA was 0.11 when the quantity of iNOS cDNA was normalized by the amount of glyceraldehyde-3-phosphate dehydrogenase cDNA. Intergroup comparisons were done using two-way ANOVA with Student’s t-test. *p , 0.05 compared to each control value. †p , 0.05 compared to each vehiclepretreated value. CTL, control; ARF, acute renal failure.

322

YANAGISAWA ET AL.

FIG. 5. Photographs of immunohistochemical studies for ET-1 (A,B,G,H) eNOS (C,D) and iNOS (E,F,I,J) proteins. Glomeruli from saline-treated control (A,C,E) and HgCl2-induced ARF (B,D,F) rats are presented. Similarly, cortices from saline-treated control (G,I) and HgCl2-induced ARF (H,J) rats are shown. Microscopic magnification: 4003. ET, endothelin; eNOS, endothelial nitric oxide synthase; iNOS, inducible nitric oxide synthase.

of endogenous TNF activity with TNF Ab or prior administration of the iNOS inhibitor aminoguanidine decreased the extent of proximal tubule epithelial cell injury and levels of BUN and serum Cr in rats with HgCl2-induced ARF. There was a parallel relationship between the magnitude of these effects and the degree of inhibition of iNOS mRNA expression

by TNF Ab or aminoguanidine (aminoguanidine . TNF Ab). Indeed, the expression of renin and TNF mRNA in cortices and levels of the systemic RAS and circulating TNF were significantly augmented after administration of HgCl2. Taken together, these observations indicate that either endogenous ANG II or ET-1, or both, and iNOS may be involved in the

HgCl2-INDUCED ACUTE RENAL FAILURE

323

FIG. 5—Continued

development of ATN through the acceleration of proximal tubule epithelial cell injury, predominantly causing a further decline in single nephron (SN) GFR. The biochemical mechanisms by which increased expression of iNOS mRNA and protein leads to the progression of proximal tubule epithelial cell injury may be due to the cytotoxic effect of large amounts of the free radical NO biosynthesized by this mRNA and protein (Cross et al., 1994; Ketteler et al.,

1994; Pinsky et al., 1995). However, the mechanisms responsible for the development of the epithelial cell injury caused by ANG II and/or ET-1 remain unclear. It may be worth exploring if a factor such as ischemia resulting from the vasoconstrictor(s) is involved in the mechanisms. On the other hand, there was a rise in prepro-ET-1 mRNA and ET-1 and a fall in eNOS mRNA and protein in glomeruli of rats with HgCl2-induced ARF relative to control rats. Sup-

324

189.4* 6 6.8 4.31* 6 0.09 17.9 6 1.2 0.43 6 0.04 BUN (mg/dl) Cr (mg/dl)

Note. Data reported represent means 6 SE of values obtained from 10 rats. Intergroup comparisons were done using two-way ANOVA with Student’s t-test. *p , 0.05 compared to each control value. †p , 0.05 compared to each vehicle-pretreated value. ¶p , 0.05 when comparing each aminoguanidine-pretreated value to the TNF Ab-pretreated value. NRS, normal rabbit serum; TNF Ab, tumor necrosis factor antibody; CTL, control; ARF, acute renal failure; BUN, blood urea nitrogen; Cr, creatinine.

141.5*,†,¶ 6 1.9 3.08*,†,¶ 6 0.09 18.8 6 1.2 0.40 6 0.01 20.3 6 1.9 190.5* 6 10.8 0.46 6 0.04 4.55* 6 0.13 156.8*,† 6 2.8 3.55*,† 6 0.11 18.2 6 1.1 0.41 6 0.01 194.8* 6 5.3 4.42* 6 0.10 16.9 6 1.8 0.48 6 0.05 152.2*,† 6 5.4 3.11*,† 6 0.09

CTL ARF CTL ARF CTL ARF CTL

19.7 6 1.1 0.52 6 0.02

ARF CTL ARF ARF

CTL

Saline TNF Ab NRS TCV-116 0.5% Ethanol

TABLE 6 Levels of Blood Urea Nitrogen and Serum Creatinine in Control Rats and Rats with HgCl2-Induced Acute Renal Failure that Were Pretreated or Not with TCV-116, Tumor Necrosis Factor Antibody, or Aminoguanidine

Aminoguanidine

YANAGISAWA ET AL.

pression of the action of endogenous ANG II with the AT1 receptor antagonist TCV-116 reduced increased levels of BUN and serum Cr together with a decrease in enhanced expression of glomerular prepro-ET-1 mRNA in HgCl2-induced ARF rats. The vasoconstrictors ANG II and ET-1 decrease QA and SNGFR through an increment in glomerular arteriole resistance and a decrement in the ultrafiltration coefficient (Kf) due to mesangial cell contraction (Brenner et al., 1980; Kon and Badr, 1991; Marsen et al., 1994). Inversely, the vasodilator NO, derived from the constitutive NOS protein eNOS, augments QA and SNGFR by causing a decrement in glomerular arteriole resistance and an increment in Kf due to mesangial cell relaxation (Deng and Baylis, 1993; Nicola et al., 1992; Raij and Baylis, 1995). Collectively, these findings suggest that, in addition to ATN, the glomerulus itself, although it is not a primary factor (Hostetter et al., 1983; Kreisberg et al., 1983; Wolfert et al., 1987), also plays an important role in a reduction in SNGFR via a decrement in QA and Kf due to the imbalance between the vasoconstrictors ANG II and ET-1 and the vasodilator NO derived from eNOS. As with renal cortices, glomeruli also had markedly greater expression of iNOS mRNA and protein in rats with HgCl2induced ARF than in control rats. Concomitant inhibition of iNOS mRNA expressed in glomeruli and cortices of rats with HgCl2-induced ARF with TNF Ab or the iNOS inhibitor aminoguanidine blunted the development of proximal tubule epithelial cell damage with no significant morphological changes in glomeruli and ameliorated the deterioration of BUN and serum Cr. Accordingly, it is suggested that enhanced expression of iNOS mRNA and protein seen in glomeruli of rats with HgCl2-induced ARF participates in proximal tubule epithelial cell injury rather than glomerular cell damage through the resultant cytotoxic effect and rather than antagonizes decreased expression of glomerular eNOS mRNA and protein, probably causing a reduction in SNGFR in the setting of this ARF. In addition to the above descriptions, the results of the AT1 receptor antagonist TCV-116 or TNF Ab pretreatment, furthermore, indicate that the gene expression of prepro-ET-1 or iNOS is at least in part up-regulated at the transcription level by endogenous ANG II or TNF in glomeruli and cortices of rats with HgCl2-induced ARF. By contrast, both TCV-116 and TNF Ab had no effects on the expression of eNOS mRNA in glomeruli of control and HgCl2-induced ARF rats, indicating that the gene expression of eNOS is not controlled by ANG II and TNF. It is reported, however, that HgCl2 causes reversible damage of renal arterioles (Kreisberg et al., 1983). This suggests a primary decrease in eNOS mRNA and protein secondary to glomerular arteriole injury in the HgCl2-induced ARF model. We demonstrated immunodetectable amounts of ET-1 and iNOS protein in glomerular epithelial and mesangial cells and proximal tubule epithelial cells of rats with HgCl2-induced ARF and immunoidentifiable amounts of eNOS protein in the glomerular tuft of control rats. Immunohistochemical studies also found no significant invasion of ED1-positive cells in the kidney sections from rats with HgCl2-induced ARF, indicating

HgCl2-INDUCED ACUTE RENAL FAILURE

no substantial involvement of infiltrating macrophages and/or monocytes in the setting of this ARF because the monoclonal Ab ED1 used recognizes both macrophages and monocytes (Diamond et al., 1994; Dijkstra et al., 1985). Thus, preproET-1 and iNOS mRNA may be enhanced or induced in glomerular epithelial and mesangial cells and proximal tubule epithelial cells of the HgCl2-induced ARF setting. Glomerular eNOS mRNA may be constitutively expressed in glomerular endothelial cells of the control setting. Of interest is that iNOS mRNA is slightly but constitutively expressed in glomeruli of control rats. This observation is consistent with that of Mohaupt et al. (1994) and Morrissey et al. (1994). The physiological role of iNOS mRNA expression is not evident in glomeruli of the normal kidney. Several reports, however, suggest that NO derived from constitutively expressed iNOS protein plays a role in maintaining and regulating renal medullary blood flow (Brezis et al., 1991; Mohaupt et al., 1994; Morrissey et al., 1994). Based on this, it is possible that constitutively expressed iNOS mRNA in glomeruli of the normal kidney partially takes part in driving glomerular hemodynamics. The basal production levels of cytokines such as TNF, although we have demonstrated the presence of circulating TNF in control rats, may be sufficient to induce iNOS constitutively (Morrissey et al., 1994). In vivo pretreatment of rats with the iNOS inhibitor aminoguanidine decreased the expression of iNOS mRNA without affecting the abundance of the internal standard GAPDH mRNA in glomeruli and cortices of the control and HgCl2induced ARF setting, suggesting the transcriptional control of aminoguanidine for the iNOS gene expression. Presumably, aminoguanidine blocks iNOS protein-derived NO production via the mechanism to decrease iNOS abundance in addition to the action as an inhibitor of iNOS activity. Further studies, however, remain to elucidate the mechanism responsible for the inhibitory effect of aminoguanidine on the expression of iNOS mRNA. In summary, novel aspects of this study are the findings that ANG II, ET-1, and NO participate in the development of HgCl2-induced ARF. These vasoactive/cytotoxic substances may lead to the further progression of HgCl2-induced ARF by causing the facilitation of proximal tubule epithelial cell damage which is a central factor for the development of this ARF and the deterioration of glomerular hemodynamics. ACKNOWLEDGMENTS This work was supported by Grants (05454217, 06404027, and 07457094) for Scientific Research from the Ministry of Education, Science, Sports, and Culture of Japan.

REFERENCES Baylis, C., and Qiu, C. (1996). Importance of nitric oxide in the control of renal hemodynamics. Kidney Int. 49, 1727–1731. Bosse, H-M., Moser, D., and Bachmann, S. (1994). Constitutive isoforms of

325

nitric oxide synthase (NOS) in the kidney localization in various mammalian species including man. J. Am. Soc. Nephrol. 5(3), 574(99P). Brenner, B. M., Badr, K. F., Schor, N., and Ichikawa, I. (1980). Hormonal influences and glomerular filtration. Miner. Electrolyte Metab. 4, 49 –56. Brezis, M., Heyman, S. N., Dinour, D., Epstein, F. H., and Rosen, S. (1991). Role of nitric oxide in renal medullary oxygenation. J. Clin. Invest. 88, 390 –395. Buttery, L. D., Evans, T. J., Springall, D. R., Carpenter, A., Cohen, J., and Polak, J. M. (1994). Immunohistochemical localization of inducible nitric oxide synthase in endotoxin-treated rats. Lab. Invest. 71, 755–764. Chomczynski, P., and Sacchi, N. (1987). Single-step method of RNA isolation by acid guanidinium thiocyanate-phenol-chloroform extraction. Anal. Biochem. 162, 156 –159. Conger, J. D., and Falk, S. A. (1986). Glomerular and tubular dynamics in mercuric chloride-induced acute renal failure. J. Lab. Clin. Med. 107, 281–289. Cross, A. H., Misko, T. P., Lin, R. F., Hickey, W. F., Trotter, J. L., and Tilton, R. G. (1994). Aminoguanidine, an inhibitor of inducible nitric oxide synthase, ameliorates experimental autoimmune encephalomyelitis in SJL mice. J. Clin. Invest. 93, 2684 –2690. Deng, A., and Baylis, C. (1993). Locally produced EDRF controls preglomerular resistance and ultrafiltration coefficient. Am. J. Physiol. 264, F212– F215. Diamond, J. R., Kees-Folts, D., Ding, G., Frye, J. E., and Restrepo, N. C. (1994). Macrophages, monocyte chemoattractant peptide-1, and TGF-b1 in experimental hydronephrosis. Am. J. Physiol. 266, F926 –F933. Dijkstra, C. D., Dopp, E. A., Joling, P., and Kraal, G. (1985). The heterogeneity of mononuclear phagocytes in lymphoid organs: Distinct macrophage subpopulations in the rat recognized by monoclonal antibodies ED1, ED2, and ED3. Immunology 54, 587–599. Girndt, M., Ko¨hler, H., Schiedhelm-Weick, E., Schlaak, J. F., Meyer Zum Buschenfelde, K. H., and Fleischer, B. (1995). Production of interleukin-6, tumor necrosis factor a and interleukin-10 in vitro correlates with the clinical immune defect in chronic hemodialysis patients. Kidney Int. 47, 559 –565. Hackenthal, E., Metz, R., Buhrle, C. P., and Taugner, R. (1987). Intrarenal and intracellular distribution of renin and angiotensin. Kidney Int. 31(Suppl. 20), S4 –S17. Hattori, T., Fujitsuka, N., Kuroki, A., and Shindo, S. (1997). Sairei-to may inhibit the synthesis of endothelin-1 in nephritic glomeruli. Jpn. J. Nephrol. 39, 121–128. Hostetter, T. H., Wilkes, B. M., and Brenner, B. M. (1983). Renal circulatory and nephron function in experimental acute renal failure. In Acute Renal Failure (B. M. Brenner and J. M. Lazarus, Eds.), pp. 99 –115. W. B. Saunders Company, Philadelphia. Ketteler, M., Border, W. A., and Noble, N. A. (1994). Cytokines and L-arginine in renal injury and repair. Am. J. Physiol. 267, F197–F207. Kim, S., Ohta, K., Hamaguchi, A., Omura, T., Yukimura, T., Miura, K., Inada, Y., Wada, T., Ishimura, Y., Chatani, F., and Iwao, H. (1994). Contribution of renal angiotensin II type I receptor to gene expression in hypertensioninduced renal injury. Kidney Int. 46, 1346 –1358. Kon, V., and Badr, K. F. (1991). Biological actions and pathophysiologic significance of endothelin in the kidney. Kidney Int. 40, 1–12. Kreisberg, J. I., Matthys, E., and Venkatachalam, M. A. (1983). Morphologic factors in acute renal failure. In Acute Renal Failure (B. M. Brenner and J. M. Lazarus, Eds.), pp. 21– 46. W. B. Saunders Company, Philadelphia. Macica, C. M., Escalante, B. A., Conners, M. S., and Ferreri, N. R. (1994). TNF production by the medullary thick ascending limb of Henle’s loop. Kidney Int. 46, 113–121. Marsen, T. A., Schramek, H., and Dunn, M. J. (1994). Renal action of endothelin: Linking cellular signaling pathways to kidney disease. Kidney Int. 45, 336 –344.

326

YANAGISAWA ET AL.

Mohaupt, M. G., Elzie, J. L., Ahn, K. Y., Clapp, W. L., Wilcox, C. S., and Kone, B. C. (1994). Differential expression and induction of mRNAs encoding two inducible nitric oxide synthases in rat kidney. Kidney Int. 46, 653– 665. Morrissey, J., McCracken, R., Kaneto, H., Vehaskari, M., Montani, D., and Klahr, S. (1994). Location of an inducible nitric oxide synthase mRNA in the normal kidney. Kidney Int. 45, 998 –1005. Nicola, L. D., Blantz, R. C., and Gabbai, F. B. (1992). Nitric oxide and angiotensin II. J. Clin. Invest. 89, 1248 –1256. Nio, Y., Matsubara, H., Murasawa, S., Kanasaki, M., and Inada, M. (1995). Regulation of gene transcription of angiotensin II receptor subtypes in myocardial infarction. J. Clin. Invest. 95, 46 –54. Ong, A. C. M., and Fine, L. G. (1994). Tubular-derived growth factors and cytokines in the pathogenesis of tubulointerstitial fibrosis: Implications for human renal disease progression. Am. J. Kidney Dis. 23, 205–209. Pelayo, J. C., Mobilia, M. A., Tjio, S., Singh, R., Nakamoto, J. M., and Van Dop, C. (1994). A method for isolation of rat renal microvessels and mRNA localization. Am. J. Physiol. 267, F497–F503. Pinsky, D. J., Cai, B., Yang, X., Rodriguez, C., Sciacca, R. R., and Cannon, P. J. (1995). The lethal effects of cytokine-induced nitric oxide on cardiac myocytes are blocked by nitric oxide synthase antagonism or transforming growth factor b. J. Clin. Invest. 95, 677– 685. Raij, L., and Baylis, C. (1995). Glomerular action of nitric oxide. Kidney Int. 48, 20 –32.

Rocco, M., Neilson, E., Hoyer, J., and Ziyadeh, F. (1992). Attenuated expression of epithelial cell adhesion molecules in murine polycystic kidney disease. Am. J. Physiol. 262, F679 –F686. Tracey, W. R., Pollock, J. S., Murad, F., Nakane, M., and Forstermann, U. (1994). Identification of an endothelial-like type III NO synthase in LLCPK1 kidney epithelial cells. Am. J. Physiol. 266, C22–C28. Ujiie, K., Terada, Y., Nonoguchi, H., Shinohara, M., Tomita, K., and Marumo, F. (1992). Messenger RNA expression and synthesis of endothelin-1 along rat nephron segments. J. Clin. Invest. 90, 1043–1048. Wolfert, A. I., Laveri, L. A., Reilly, K. M., Oken, K. R., and Oken, D. E. (1987). Glomerular hemodynamics in mercury-induced acute renal failure. Kidney Int. 32, 246 –255. Xie, Q., Cho, H., Calaycay, J., Mumford, R., Swiderek, K., Lee, T., Ding, A., Troso, T., and Nathan, C. (1992). Cloning and characterization of inducible nitric oxide synthase from mouse macrophages. Science (Washington, D.C.) 256, 225–228. Yanagisawa, H., Morrissey, J., Kurihara, N., Wada, O., and Klahr, S. (1994). Effects of dietary protein on glomerular eicosanoid production in rats with bilateral ureteral obstruction. Proc. Soc. Exp. Biol. Med. 207, 234 –241. Yanagisawa, H., Morrissey, J., Morrison, A. R., Purkerson, M. L., and Klahr, S. (1990). Role of ANG II in eicosanoid production by isolated glomeruli from rats with bilateral ureteral obstruction. Am. J. Physiol. 258, F85–F93.