Introduction

Top Abstract Introduction Experimental Procedures Results Discussion References

Nitric oxide (NO) produced in biological systems mediates a diverse array of physiological functions (Ignarro et al., 1988; Furchgott and Vanhoutte, 1989; Moncada and Higgs, 1993). In the vascular system, NO regulates organ blood flow, inhibits platelet aggregation, and attenuates neutrophil adherence (Ignarro et al., 1988; Kubes et al., 1991; Moncada and Higgs, 1993). A cytoprotective effect of NO has been reported for ischemia-reperfusion injuries in various organs (Tsao et al., 1990; Konorev et al., 1995; Liu et al., 1998; Ohmori et al., 1998; Cottart et al., 1999). Impairment of endothelial NO production may contribute to the pathogenesis of ischemia-reperfusion injuries, because a decrease in NO release can trigger neutrophil adherence and exudate into the ischemic area, which exacerbates reperfusion injury.

It was recently proposed that various redox isoforms of NO have critical roles in the diverse physiological and pathophysiological events induced by NO. For example, biological S-nitrosation occurs via one-electron oxidation of NO catalyzed by heavy metal ions, such as copper and iron, and particularly by ceruloplasmin, which is a major multicopper-containing plasma protein in mammals (Inoue et al., 1999; Akaike, 2000). Sulfhydryl-containing molecules such as glutathione are particularly susceptible to nitrosation; a nucleophilic attack by NO⁺ results in S-nitrosothiol adducts (nitrosothiols; RS-NOs) (Stamler et al., 1992; Akaike, 2000). Nitrosothiols may function as endogenous NO reservoirs and preserve the antioxidant activities of NO (Ignarro et al., 1981; Stamler et al., 1992; Gaston et al., 1993; Rauhala et al., 1998).

We recently found that ₁-protease inhibitor (₁-PI) from human plasma is readily S-nitrosated under physiological conditions and that its nitrosylation is 10 times more efficient than nitrosylation of bovine serum albumin and glutathione (Miyamoto et al., 2000a,b). ₁-PI, which is the most abundant serine protease inhibitor in human plasma (30-60 µM), is known to be an important defense-oriented acute phase protein (Heidtmann and Travis, 1986). ₁-PI (mol. wt., 53,000) has no intramolecular disulfide bridge but does have a single Cys residue at position 232. In our earlier study, ₁-PI was nitrosylated with isoamylnitrite at this single SH, yielding 100% S-nitrosylated ₁-PI (S-NO-₁-PI), which exhibits discrete homogeneity. More importantly, S-NO-₁-PI has multiple biological functions, including potent antimicrobial activity and inhibition of cysteine protease (Miyamoto et al., 2000a,b). In the present report, we describe a clear protective effect of S-NO-₁-PI on hepatic ischemia-reperfusion injury in rats.

	Experimental Procedures

Top Abstract Introduction Experimental Procedures Results Discussion References

Materials. ₁-PI, provided by Chemo-Sero-Therapeutic Institute (Kumamoto, Japan), was purified from human plasma as described previously (Miyamoto et al., 2000a). S-NO-₁-PI was prepared by S-nitrosation (100%) of a single SH group of ₁-PI with isoamylnitrite and was purified to homogeneity by gel filtration column chromatography (Miyamoto et al., 2000a,b). N-Nitro-L-arginine methyl ester (L-NAME) was purchased from Sigma Chemical Co. (St. Louis, MO). S-Nitrosoglutathione (GS-NO) and S-nitroso-N-acetyl penicillamine (SNAP) were from Dojindo Laboratories (Kumamoto, Japan).

Animals. Male Wistar rats were obtained from a commercial supplier (Kyudo, Inc., Kumamoto, Japan). All animals were maintained under standard conditions and were fed water and rodent chow ad libitum. The animals weighed between 200 and 230 g.

Experimental Protocol. Guidelines of the Center for Animal Resource and Development, Kumamoto University, were followed for anesthesia during the operative procedure and subsequent postoperative care. The animals were fasted for 9 h before surgery but were allowed access to water. Rats were anesthetized with ether during the operation. After the abdomen was shaved and disinfected with 70% ethanol, a complete midline incision was made. The portal vein and hepatic artery were exposed and cross-clamped for 30 min with a noncrushing microvascular clip. Saline or various compounds such as S-NO-₁-PI (6.0, 20, 100, and 200 nmol; 0.3, 1.0, 5.3, 10.6 mg/rat), native ₁-PI (6.0, 20, 100, and 200 nmol; 0.3, 1.0, 5.3, and 10.6 mg/rat), GS-NO (6.0, 20, and 100 nmol; 2.0, 6.7, and 33.6 µg/rat), SNAP (100 nmol; 22.0 mg/rat), and L-NAME (10 mg/kg) were given via the portal vein immediately after reperfusion was initiated, and then the abdomen was closed in two layers with 2-0 silk. Rats were kept under warming lamps until they awakened and became active.

Measurement of Liver Enzyme Activity in Plasma. Because blood loss caused by frequent blood sampling might have affected liver functions, animals were sacrificed by taking whole circulating blood via abdominal aorta under anesthesia at various time points after reperfusion was initiated. Plasma alanine aminotransferase (ALT), aspartate aminotransferase (AST), and lactate dehydrogenase (LDH) activities were determined by using a sequential multiple AutoAnalyzer system (Hitachi Ltd., Tokyo, Japan). Activities were expressed in international units per liter.

Hepatic Tissue Blood Flow and Blood Pressure Measurement. A laser Doppler flowmeter (Laser Flowmeter, ALF21; Advance Co. Ltd., Tokyo, Japan) was used to measure hepatic tissue blood flow before ischemia and 30 and 60 min after initiating reperfusion. In each animal, the probe of the flowmeter was inserted at the same site in the median lobe of the liver. The mean value of the blood flow at respective time points was calculated and expressed as a percentage of the preischemic initial blood flow value. For blood pressure measurement, the femoral artery was cannulated and arterial blood pressure was continuously recorded by using a pressure transducer as described previously (Yoshida et al., 1994). These measurements were performed with the animals under urethane (ethyl carbamate) anesthesia.

Identification of Neutrophil Infiltration and Apoptotic Change in the Liver. The liver was removed at various time points after ischemia-reperfusion and was cut into small tissue blocks (3 × 4 × 5 mm³) with a razor blade. The tissue blocks were embedded in Tissue-Tek OCT compound (Miles, Elkhart, IN) and were immediately frozen in dry ice-acetone. Frozen sections, 6 µm thick, were prepared with a cryostat, and cryosections were air-dried overnight. After fixation in pure acetone for 10 min at room temperature, the cryosections were stained with an antileukocyte monoclonal antibody (Serotec Inc., Raleigh, NC) and were visualized by use of an indirect immunoperoxidase method, with 3,3'-diaminobenzidine as a substrate (Doi et al., 1999). Neutrophil infiltration was also analyzed histochemically with an esterase stain, according to a method reported earlier (Molony et al., 1960).

Detection of Heme Oxygenase-1 mRNA Expression in the Liver after Ischemia-Reperfusion. Heme oxygenase-1 (HO-1) expression was analyzed by Northern blotting as described previously (Doi et al., 1999). Briefly, liver specimens obtained from animals exsanguinated at various time points after reperfusion were frozen quickly in liquid nitrogen and were stored at 80°C until RNA extraction. Total RNA was extracted from the liver after ischemia-reperfusion by using the guanidine thiocyanate lysis method with Trizol reagent (Life Technologies, Gaithersburg, MD). Each RNA sample (20 µg) underwent electrophoresis on agarose gel and was transferred to the Hybond-N⁺ nylon membrane, followed by hybridization of a DNA probe for rat HO-1. The DNA probe was radiolabeled by the random primer technique using [-³²P]dCTP and the Megaprime labeling system (Amersham Pharmacia Biotech, Buckinghamshire, UK). An HO-1 cDNA fragment of 882 base pairs was used for hybridization as reported previously (Doi et al., 1999), and a cDNA fragment for glyceraldehyde-3-phosphate dehydrogenase was used as a control for gene expression. The HO-1 mRNA signals were quantified by densitometric analysis, after normalization with glyceraldehyde-3-phosphate dehydrogenase mRNA signals, using a Macintosh computer with an image scanner (GT6500; Epson Co., Ltd., Tokyo, Japan) and the public domain IMAGE program (National Institutes of Health).

Statistical Analysis. Statistical difference was determined by the two-tailed unpaired t test. Dose-response data were evaluated by two-way ANOVA. A P value of < .05 was considered statistically significant.

	Results

Top Abstract Introduction Experimental Procedures Results Discussion References

Ischemia-Reperfusion Liver Injury Assessed by Measurement of Liver Enzymes in Plasma. Liver injury in this model was evaluated by measuring the extracellular release of the liver enzymes AST, ALT, and LDH in the plasma. These enzymes increased to a maximum at 3 h after reperfusion and then decreased gradually during 24 h (Fig. 1). When S-NO-₁-PI was administered just after initiating reperfusion, the increased levels of both ALT and AST were markedly reduced at almost all time points except for 24 h after reperfusion (Fig. 1A). S-NO-₁-PI treatment also reduced the elevated LDH levels in the plasma at 3 h after reperfusion (Fig. 1B). The elevated liver enzyme activities were lowered by S-NO-₁-PI in a dose-dependent manner, as assessed at 3 h after reperfusion (Fig. 2, A and B). Native ₁-PI slightly reduced the levels of all these enzymes (Fig. 2, C and D). However, significant changes in ALT and AST levels were not observed after treatment with the same dose of GS-NO and SNAP (Fig. 3, A and B). L-NAME administration tended to exacerbate the ischemia-reperfusion injury (Fig. 3, A and B). A similar trend of exacerbation of liver damage as assessed by LDH levels was observed for groups treated with either GS-NO or SNAP (Fig. 3B).

View larger version (17K): [in this window] [in a new window]   Fig. 1.   Time profile of changes in serum levels of AST (A), ALT (A), and LDH (B) after hepatic ischemia-reperfusion in rats. Ischemia was induced by occluding both the portal vein and the hepatic artery for 30 min followed by reperfusion. Vehicle (saline) and S-NO-1-PI (0.1 µmol; 5 mg/rat) were administered via the portal vein immediately after reperfusion was initiated. Data are means ± S.E.M. (n = 5 at each time point). *P < .05, **P < .01, versus the vehicle-treated group.

View larger version (37K): [in this window] [in a new window]   Fig. 2.   Effect of various doses of S-NO-1-PI (A and B) and native 1-PI (C and D) on hepatic ischemia-reperfusion injury in rats. S-NO-1-PI was given to the animals in a manner similar to the protocol described in Fig. 1. The plasma levels of ALT and AST (A and C) and LDH (B and D) were measured at 3 h after reperfusion was initiated. Each experimental group consisted of five animals. Data are means ± S.E.M. *P < .05, **P < .01, versus the vehicle-treated group.

View larger version (30K): [in this window] [in a new window]   Fig. 3.   Effect of various nitrosothiols (S-NO-1-PI, GS-NO, SNAP), 1-PI, and NOS inhibitor (L-NAME) on hepatic ischemia-reperfusion injury in rats. A and B, each compound was administered in the same manner as described in Fig. 1, and the plasma levels of liver enzymes (A, ALT and AST; B, LDH) were determined at 3 h after reperfusion was initiated. Each experimental group consisted of five animals. Data represent means ± S.E.M. *P < .05, **P < .01, versus the vehicle-treated group (A and B). C, synergistic effect of S-NO-1-PI on the ischemia-reperfusion injury. The experimental design was almost similar to those in A and B, except that GS-NO and 1-PI were administered simultaneously at equimolar doses. Data are means ± S.E.M. (n = 5). *P < .01 by two-way ANOVA, versus the GS-NO-treated group.

Hepatic Tissue Blood Flow. Hepatic blood flow before and after ischemia-reperfusion was measured by using a laser Doppler flowmeter (Fig. 4). Blood flow decreased immediately after ischemia and did not change during reperfusion; it remained significantly lower in the vehicle-treated group at 30 and 60 min after reperfusion was initiated compared with that before ischemia. In contrast, with S-NO-₁-PI, the impaired hepatic blood flow almost completely recovered after ischemia-reperfusion: recovery was 95.2% at 30 min and 100.8% at 60 min after reperfusion. GS-NO and SNAP at the same dose did not affect the reduction of hepatic blood flow induced by ischemia-reperfusion. Inhibition of NO production by L-NAME treatment caused a greater decrease in hepatic blood flow after ischemia-reperfusion.

View larger version (20K): [in this window] [in a new window]   Fig. 4.   Time profile of change in hepatic blood flow before and after ischemia-reperfusion. Blood flow values are expressed as a percentage of the value measured before ischemia was started. Drugs were administered as described in Fig. 1. Some of the data in this figure has been adapted from Ikebe et al. (2000). Data are means ± S.E.M. (n = 4). *P < .05, versus the vehicle-treated group.

Infiltration of Neutrophils and Apoptotic Change in the Liver after Ischemia-Reperfusion Immunostaining with an antineutrophil antibody revealed a large number of neutrophils in the exudate in the sinusoidal areas of the liver after ischemia-reperfusion: 122.1 ± 24.7/mm² and 114.3 ± 6.9/mm² in vehicle-treated and native ₁-PI-treated rats, respectively, at 3 h after reperfusion (Fig. 5). Neutrophil infiltration was also identified by an esterase stain, which revealed a similar distribution of neutrophils in the liver (data not shown). The number of neutrophils increased in a time-dependent manner after reperfusion was initiated, with a peak at 12 h after reperfusion. S-NO-₁-PI treatment remarkably reduced neutrophil accumulation in the liver: to 53.3 ± 7.8/mm² at 3 h after reperfusion and throughout the observation period up to 24 h after ischemia-reperfusion.

View larger version (109K): [in this window] [in a new window]   Fig. 5.   Immunohistochemistry for neutrophil infiltration after ischemia-reperfusion. S-NO-1-PI and vehicle (saline) were given to the animals as described in Fig. 1, and liver specimens were obtained at 3 h (A-C) or at indicated time points (D) after reperfusion. Anti-rat neutrophil antibody was used for immunostaining. A, control (vehicle treatment); B and C, treatment with 0.1 µmol (5 mg) of S-NO-1-PI and native 1-PI, respectively. Magnification: 41×. D, time profile of neutrophil accumulation in the liver of rats treated with vehicle or S-NO-1-PI; the inset shows the effects of vehicle, S-NO-1-PI, and 1-PI on neutrophil accumulation at 3 h after reperfusion. *P < .05, **P < .01, versus the normal value (n = 5).

View larger version (97K): [in this window] [in a new window]   Fig. 6.   Effect of S-NO-1-PI on apoptotic change in the liver induced by ischemia-reperfusion. Animals with ischemia-reperfusion were treated with vehicle, 1-PI, or S-NO-1-PI in the same manner as described in Fig. 1. At 3 h after reperfusion was initiated, apoptotic change was determined by the TUNEL method with an apoptotic detection kit. A, control (vehicle treatment); B and C, treated with 0.1 µmol (5 mg)/rat of S-NO-1-PI and 1-PI, respectively. Magnification: 41×. D, time profile of apoptotic change in the liver with or without S-NO-1-PI (0.1 µmol/rat) treatment; the inset shows the effects of vehicle, 1-PI, and S-NO-1-PI on apoptosis in the liver measured at 3 h after reperfusion. *P < .05, versus normal value (n = 5).

View larger version (181K): [in this window] [in a new window]   Fig. 7.   Localization of neutrophil infiltration and apoptotic cells in the liver after ischemia-reperfusion. The liver section was obtained at 3 h after reperfusion; it was first immunostained with anti-neutrophil antibody, followed by in situ apoptosis detection as described in Figs. 6 and 7. Arrows and arrowheads indicate apoptotic cells and neutrophils, respectively. Magnification: 297×.

HO-1 Induction in the Liver after Ischemia-Reperfusion. HO-1 mRNA transcript was induced as early as 1 h after perfusion and peaked sharply at 3 h. It was sustained at a moderately elevated level for 24 h after reperfusion (Fig. 8A). S-NO-₁-PI administration significantly potentiated HO-1 mRNA induction obtained at 3 h after initiating reperfusion, but native ₁-PI had no effect (Fig. 8B). The induction of HO-1 by S-NO-₁-PI was also confirmed by elevation of enzyme activity determined as described previously (Doi et al., 1999) (data not shown). In contrast, GS-NO treatment did not affect HO-1 induction in the liver, at least at the same dose as that of S-NO-₁-PI.

View larger version (29K): [in this window] [in a new window]   Fig. 8.   Effect of S-NO-1-PI on HO-1 induction in the liver after ischemia-reperfusion. A, time profile of HO-1 mRNA induction after ischemia-reperfusion. HO-1 mRNA signals were quantified by densitometric analysis of Northern blotting data. B, Northern blotting of HO-1 mRNA expressed in the liver at 3 h after reperfusion was initiated, after various treatments. The HO-1 mRNA signals obtained in B were quantitated by densitometric analysis. Data are means ± S.E.M. (n = 5). **P < .01, versus normal level in A; versus vehicle- and 1-PI-treated groups in B.

	Discussion

Top Abstract Introduction Experimental Procedures Results Discussion References

The data presented here demonstrate that S-NO-₁-PI exerts potent protective effects against hepatic ischemia-reperfusion injury as evidenced by decreased levels of plasma liver enzymes, improvement of impaired of hepatic blood flow, inhibition of neutrophil infiltration and apoptosis, and enhanced induction of HO-1 in the liver after ischemia-reperfusion.

Local excessive formation of NO has pro-inflammatory effects such as edema formation due to enhanced vascular permeability, inflammatory cell infiltration, and cytotoxicity (possibly through its conversion to peroxynitrite and other reactive nitrogen oxides) (Beckman and Koppenol, 1996; Rubbo et al., 1996; Akaike and Maeda, 2000). However, NO also has completely opposite physiological functions, i.e., anti-inflammatory activities, such as inhibition of platelet aggregation and of neutrophil adherence to endothelium, anti-apoptotic effects, and inhibition of lipid peroxidation (Kubes et al., 1991; De Caterina et al., 1995; Rubbo et al., 1996; Ogura et al., 1997; Mannick et al., 1999). A number of reports show a beneficial effect of NO on hepatic microcirculation and liver injury as evidenced by cytoprotective effects of NO donors and detrimental consequences of NOS inhibition (Liu et al., 1998; Ohmori et al., 1998; Cottart et al., 1999). The present experiments support these earlier observations, because the novel NO donor S-NO-₁-PI had a remarkable ameliorating effect on such injury and the NOS inhibitor L-NAME exacerbated the injury.

Typical biological actions of NO include vasodilatation (vascular smooth muscle relaxation) and inhibition of platelet aggregation, which are mediated partly through elevation of intracellular cGMP produced by soluble guanylate cyclase activated by NO (Rapoport and Mura, 1983; Furchgott and Vanhoutte, 1989; Radomski et al., 1990; Moncada and Higgs, 1993; Ignarro et al., 1988). Some part of the endothelium-dependent vasodilation and inhibition of platelet aggregation is reported to be cGMP-independent (Bolotina et al., 1994; Pawloski et al., 1998). In addition, NO has inhibitory effects on endothelium-neutrophil interaction, partly via NF-B-mediated down-regulation of the expression of adhesion molecules on both endothelium and neutrophils (Kubes et al., 1991; Gauthier et al., 1994; De Caterina et al., 1995; Liu et al., 1998) and via suppression of cytokine production by inflammatory cells (De Caterina et al., 1995). Also, NO causes hepatic sinusoidal dilatation and improves liver microcirculation by altering the morphofunctional activity of fat-storing (Ito) cells (Kawada et al., 1993). In this context, the potent therapeutic effect of S-NO-₁-PI on hepatic ischemia-reperfusion injury might be produced by NO supplied to vascular systems to prevent neutrophil-mediated endothelial damage and sustain microcirculation in the liver. In fact, only a small amount of NO, as low as a nanomolar concentration, was produced in our ischemia-reperfusion model, as identified in an ex vivo perfusion system of the liver with ischemia-reperfusion injury (data not shown). However, given the profound effect of S-NO-₁-PI on the microcirculation in the liver, the reduced neutrophil accumulation may be a consequence of the improved microcirculation and not a direct effect of NO.

It is well documented that the inflammatory response involving neutrophil infiltration is generally elicited during reperfusion of various ischemic organs and that neutrophils are the major effector cells contributing to tissue injury in ischemia-reperfusion (Jordan et al., 1999; Lentsch et al., 1999). These neutrophils can induce reperfusion injury of endothelial cells and hepatocytes by releasing a variety of cytotoxic substances, including proteases such as neutrophil elastase and matrix metalloproteases, cytokines, leukotrienes, cationic proteins, and reactive oxygen and nitrogen species, all of which can cause tissue damage (Beckman and Koppenol, 1996; Rubbo et al., 1996; Liu et al., 1998; Jordan et al., 1999; Lentsch et al., 1999; Akaike and Maeda, 2000).

We recently verified that S-NO-₁-PI has potent serine protease inhibitory activity similar to that of native ₁-PI: the inhibitory action of ₁-PI against porcine pancreatic trypsin and pancreatic and neutrophil elastase was not affected by S-nitrosation (Miyamoto et al., 2000a,b). Although the protective effect of native ₁-PI on ischemia-reperfusion injury of the rat liver was not as great as that of S-NO-₁-PI, it may be that the neutrophil elastase inhibitory activity of S-NO-₁-PI contributed in an additive fashion to its potent cytoprotective effect. It is also of potential interest that S-NO-₁-PI acquired anti-thiol protease activity after S-nitrosation (Miyamoto et al., 2000a). Of particular importance is the anti-apoptotic activity of various nitrosothiol compounds through their inhibition of caspases that are apoptosis-inducing intracellular thiol proteases (Ogura et al., 1997; Mannick et al., 1999). In fact, our current study revealed that S-NO-₁-PI significantly suppressed apoptotic change in the liver induced by ischemia-reperfusion. Therefore, S-NO-₁-PI may function not only as a simple NO (nitroso) donor but also as a protease inhibitor with a broad inhibitory spectrum.

Although the pro-oxidant activity of NO is now well recognized, particularly in view of cytotoxic actions of NO-related reactive nitrogen species such as peroxynitrite (Beckman and Koppenol, 1996; Rubbo et al., 1996; Akaike and Maeda, 2000), an apparently contradictory but important function of NO is its antioxidant potential, observed, for example, in lipid peroxidation (Rubbo et al., 1996) and iron-overload-induced neurotoxicity (Rauhala et al., 1998). Nitrosothiol formation may be critically involved in such antioxidant and cytoprotective actions of NO (Konorev et al., 1995; Rauhala et al., 1998; Akaike, 2000): NO is protected from reaction with superoxide after thiol-adduct formation of NO, and thus production of the potent cytotoxic peroxynitrite is attenuated. It is thought that S-NO-₁-PI could act as a nitroso donor rather than a pure NO donor after in vivo administration. The cytoprotective effect of S-NO-₁-PI obtained in the present study, therefore, might also be attributable to antioxidant activity of nitrosothiols introduced by S-NO-₁-PI.

Another important finding is that administration of S-NO-₁-PI potentiated induction of HO-1 in the liver after ischemia-reperfusion; its effect was greater that that of vehicle and native ₁-PI. Lancaster's group reported that pretreatment of rat hepatocytes with a low dose of NO donor conferred resistance to oxidative damage of the cell, possibly through induction of HO-1 (Kim et al., 1995). We confirmed a similar cytoprotective function of HO-1 in our recent in vivo study using a solid tumor model (rat hepatoma cells, AH136B) (Doi et al., 1999). HO-1, which catalyzes the conversion of heme to biliverdin and CO, has been shown to be constitutively expressed in the liver and spleen (Maines, 1997). Furthermore, HO-1 is readily induced by heme compounds, heavy metals, UV irradiation, and various oxidative stresses (Kim et al., 1995; Maines, 1997). One possible mechanism of the protective effect of HO-1 induction is thought to be the antioxidant action of bilirubin (Doré et al., 1999), which is generated by reduction of biliverdin, a product of the enzymatic reaction of HO-1 using heme as a substrate. In addition, ferritin, whose expression was triggered by iron release during heme breakdown catalyzed by HO-1, has been proposed to protect against oxidative stress by sequestering iron (Kim et al., 1995). On the basis of these findings, the beneficial effect of S-NO-₁-PI on ischemia-reperfusion injury of the liver seems to result from its direct antioxidant activity and indirectly induced antioxidant levels produced by HO-1 in the liver after ischemia-reperfusion injury.

In conclusion, the present results suggest that S-NO-₁-PI has a potent cytoprotective effect on hepatic ischemia-reperfusion injury, possibly by maintaining tissue blood flow, suppressing neutrophil-induced liver damage, and inducing HO-1. The beneficial effect of S-NO-₁-PI was found to be far superior to that of other nitrosothiols such as GS-NO, SNAP, and S-NO-albumin (data not shown), and thus the multiple biological activities of S-NO-₁-PI (protease inhibitory and nitrosothiol-mediated actions) may contribute to its potent cytoprotective activity. Another advantage of using S-NO-₁-PI as an NO donor is that S-NO-₁-PI can be readily formed from the endogenous ₁-PI that exists at a high level in human plasma, so that clinical application of this compound should be more feasible than that of other synthetic NO donors.