The American Journal of Surgery 185 (2003) 589 –595
Scientific paper
Differential nitric oxide synthase expression during hepatic ischemia-reperfusion Ferdinand Serracino-Inglott, M.D., M.Sc.a,b,*, Ioannis T. Virlos, M.Sc.a, Nagy A. Habib, Ch.M.a, Robin C. N. Williamson, M.A., M.D.a, Robert T. Mathie, B.Sc., Ph.D.a a
Division of Surgery, Anesthetics and Intensive Care, Imperial College School of Medicine, Hammersmith Hospital, London, United Kingdom b Department of Vascular Surgery, North Manchester General Hospital, Crumpsall, Manchester M8 5RB, United Kingdom Manuscript received October 5, 2001; revised manuscript September 1, 2002
Abstract Background: In recent years the important role of nitric oxide in hepatic ischemia-reperfusion injury has been increasingly recognised. The prevailing consensus is that reperfusion injury may be partly the result of decreased production of nitric oxide from endothelial nitric oxide synthase and excessive production of nitric oxide from the inducible isoform. We therefore undertook this study to characterize the expression of different nitric oxide synthase isoforms during hepatic reperfusion. Methods: Male Wistar rats (n ⫽ 6) were subjected to 45 minutes of partial hepatic ischemia (left lateral and median lobes) followed by 6 hours of reperfusion. Control animals (n ⫽ 6) were subjected to sham laparotomy. The expression of endothelial and inducible nitric oxide synthase was examined using immunohistochemistry and Western blotting. Liver sections were also stained with nitrotyrosine antibody, a specific marker of protein damage induced by peroxynitrite (a highly reactive free radical formed from nitric oxide). Results: Liver sections from all the control animals showed normal expression of the endothelial isoform and no expression of inducible nitric oxide synthase. Livers from all the animals subjected to hepatic ischemia showed decreased expression of endothelial nitric oxide synthase, and all but one animal from this group showed expression of the inducible isoform both in inflammatory cells and in hepatocytes. Western blotting confirmed these findings. Staining with the antinitrotyrosine antibody was also confined to five liver sections from animals subjected to hepatic ischemia. Conclusions: During the reperfusion period after hepatic ischemia, endothelial nitric oxide synthase is downregulated while inducible nitric oxide synthase is expressed in both hepatocytes and inflammatory cells. The presence of nitrotyrosine in livers subjected to hepatic ischemia-reperfusion suggests that the expression of inducible nitric oxide synthase plays an important role in mediating reperfusion injury in this model. © 2003 Excerpta Medica, Inc. All rights reserved. Keywords: Nitric oxide; Nitric oxide synthases; Hepatic reperfusion injury
A number of surgical procedures on the liver require a period of ischemia, whether by inflow control (the Pringle maneuvre) or more especially when using total vascular exclusion to deal with extensive hepatic trauma or to resect large intrahepatic lesions [1,2]. Liver ischemia can also occur during haemorrhagic shock and late sepsis [3]. Paradoxically, when the blood flow is restored, further injury to the already ischemic liver will ensue. This phenomenon is termed ischemia-reperfusion (I-R) injury and in the field of
* Corresponding author. Tel.: ⫹44-7985221909. E-mail address:
[email protected]
hepatic transplantation, it may contribute to a poorly functioning graft [4]. The insult to the liver after the onset of reperfusion may be broadly divided into two phases. In the early stages of reperfusion, endothelial cell swelling [5], vasoconstriction [6], leucocyte entrapment [7,8] and possibly intravascular haemoconcentration [9] result in failure of the microcirculation. This process prolongs the period of hypoxia, with areas of the liver remaining ischemic after the onset of reperfusion. The second phase is a result of the production of inflammatory cytokines [10] and oxygen-derived free radicls [11]. Since the identification of nitric oxide (NO) as endothe-
0002-9610/03/$ – see front matter © 2003 Excerpta Medica, Inc. All rights reserved. doi:10.1016/S0002-9610(03)00075-8
590
F. Serracino-Inglott et al. / The American Journal of Surgery 185 (2003) 589 –595
lium-derived relaxing factor in 1987 [12], a number of other physiological actions have been attributed to it. These include reducing inflammatory cell [13] and platelet activity [14], and decreasing the expression of cytokines [15] and adhesion molecules [16]. Nitric oxide has also been shown to act as a free radical via its conversion to peroxynitrite [17]. All these actions may influence the pathophysiological changes occurring in I-R injury. Nitric oxide is synthesized from L-arginine by the enzyme NO synthase (NOS) in a reaction that requires the presence of oxygen and the cofactors nicotinamide adenine dinucleotide phosphate (NADPH) and tetrahydrobiopterin. Two isoforms of NOS present within the liver are thought to play an important role in the pathophysiology of hepatic I-R—the endothelial (eNOS) and inducible (iNOS) forms [18]. A third isoform— neuronal NOS—also exists. This isoform is mainly involved in neuronal signaling and it does not participate in the events involved during I-R [19]. Endothelial NOS is constitutively expressed in the sinusoidal cells and is dependent on intracellular calcium levels for its activity. The basal production of NO from eNOS is partly responsible for maintaining the normal vascular tone within the sinusoids [20]. It has been suggested that early reperfusion injury is a result of microvascular failure as a consequence of the increased production of endothelin and the decreased production of NO from eNOS. This theory is mainly based on a number of studies showing that increasing NO availability, either prior to the establishment of ischemia or in the immediate reperfusion period, markedly improves the hepatic microcirculation, attenuating reperfusion injury [21–23]. Inducible NOS, unlike the constitutive forms of NOS, is independent of intracellular calcium levels for its activity [18]. Hepatocytes and inflammatory cells, such as neutrophils and Kupffer cells, express iNOS within 6 hours after stimulation by a number of different inflammatory mediators such as tumor necrosis factor and interleukin-1 [24 –26]. Recently, Hur et al [27] demonstrated that in the liver iNOS mRNA is expressed within 5 hours of reperfusion. Since iNOS produces NO at magnitudes several orders greater than eNOS [18], and some studies have shown that NOS inhibition attenuates hepatic reperfusion injury [28], it is believed that the excessive production of NO from this isoform plays a part in inducing hepatic injury. We tested the hypothesis that, after a period of hepatic ischemia (1) eNOS is downregulated, and (2) the expression of iNOS is associated with hepatocellular injury. Using immunohistochemistry, we proceeded to investigate the expression and localisation of these two enzymes in a rat model of hepatic I-R. We set a time point of 6 hours from the onset of reperfusion to study the expression of NOS isoenzymes, in order to allow full expression of iNOS, if any, to take place [24 –25,26,27].
Materials and methods Experimental animal model Male Wistar rats weighing 250 to 300 g were randomly assigned to two experimental groups— control or I-R. All animals were anesthetized by intraperitoneal administration of 2.7 mL/kg of a mixture of one part midazolam plus one part Hypnorm plus two parts sterile water, and body temperature was maintained between 36.5°C and 37.5°C by the use of a heating pad. The carotid artery was cannulated using an intravenous cannula (20G/32 mm; Portex, Hythe, United Kingdom) to allow for continuous blood pressure monitoring (Lectromed, United Kingdom) and the infusion of normal saline at 0.5 mL/h. All animals underwent laparotomy. The left lateral and median lobes were mobilized by dividing their suspensory ligaments, and then delivered through the incision. An avascular plane was developed between the ventral surface of the liver and the portal vein and hepatic artery supplying these lobes. In animals assigned to the control group (n ⫽ 6) no further intervention was undertaken. In animals assigned to the I-R group (n ⫽ 6) an atraumatic microvascular clip was applied to occlude the portal vein and hepatic artery just distal to the branches supplying the right lateral lobe for 45 min. After 6 hours of reperfusion in the I-R group or, in the control group, a time period equivalent to the duration of ischemia and reperfusion, blood (0.5 mL) was withdrawn from the arterial cannula for the measurement of liver enzymes. The left lateral and median lobes of the liver were harvested and divided for histological paraffin wax examination, immunohistochemical staining and Western blotting as described below. All animal experimental procedures were carried out in accordance with UK government legislation. Liver transaminases: aspartate and alanine transaminase levels in plasma were measured using an Olympus AU600 Analyser (Olympus Optical, Tokyo, Japan) as markers of hepatocellular injury. Hematoxylin and eosin staining: liver samples from each animal were fixed in 10% neutral buffered formalin, dehydrated by passage through graded ethanol series, cleared in xylene and embedded in paraffin blocks. The blocks were 4-m sectioned and stained with hematoxylin and eosin according to standard protocols. Sections were evaluated by light microscopic examination. Hepatic immunohistochemistry Liver samples were fixed in 1% paraformaldehyde in phosphate buffered saline (PBS) for 4 hours and then transferred into a storage buffer, PBS/sucrose buffer at 4°C until the cryostat blocks were prepared. Tissue sections (4 m) were cut on a cryostat. The sections were dewaxed in xylene and then rehydrated by passage through a graded alcohol series: 2 minutes each in absolute alcohol, 90% alcohol, 70% alcohol to de-ionised water. Any mercury pigment that
F. Serracino-Inglott et al. / The American Journal of Surgery 185 (2003) 589 –595
could have formed was removed with iodine. The liver sections were then immersed in a solution of 0.3% hydrogen peroxide in PBS for 30 minutes to block any endogenous peroxidase activity. After rinsing in PBS, the slides were incubated with normal goat serum in a dilution of 1 in 30 to block background staining. Excess blocking antibody was drained onto a paper tissue. The liver sections were then incubated overnight at 4°C with the primary antibodies. Anti-iNOS polyclonal produced in rabbit, and anti-eNOS monoclonal and antinitrotyrosine monoclonal both produced in mouse were obtained from Affiniti Research Products Ltd (Exeter, United Kingdom). The primary antibodies were diluted in PBS containing 0.05% bovine serum albumin and 0.01% sodium azide to a dilution of 1 in 200. The next morning, the sections were rinsed three times in PBS for 5 minutes each time and then incubated for 30 minutes with the secondary antibody. This consisted of diluted biotinylated antirabbit or antimouse immunoglobulin G (both obtained from Vector Laboratories, United Kingdom), depending on in which species the primary antibody was raised. The sections were once again rinsed three times in PBS for 5 minutes each time prior to incubation for 30 minutes with the avidin-biotin complex reagent (Vectastain ABC reagent; Vector Laboratories, United Kingdom). After another three rinses in PBS, the sections were incubated with the 3,3'-diaminobenzidine substrate (DAB). Then 25 mg of DAB kept frozen in aqueous solution was dissolved in 100 mL of PBS and added to 100 L of 100 volume (30%) hydrogen peroxide prior to incubation of the sections. Since the development time is rather short (1 to 5 minutes), the reaction was controlled by examining the sections under the microscope, taking care to prevent contamination of the microscope by DAB. The sections were thoroughly washed with PBS followed by tap water and counter stained lightly with hematoxylin prior to being dehydrated in methanol and mounted in Pertex. In some sections, the primary antibody was omitted and these slides were used as negative controls. The presence of positive staining with the specific antibody was indicated by the development of a reddish-brown stain in the section, as a consequence of using DAB. The presence or absence of staining and its cellular location was noted for each type of primary antibody in every section, in each group. Western blotting Proteins were extracted from liver samples of rats in groups 1 and 2, which were immediately snap frozen in liquid nitrogen and stored at ⫺80°C until required. The protein extracts were made by homogenization of the samples on ice, using a ratio of 1 mL of lysis buffer (50 mM TRIS base and 50 mM NaCl, pH 7.4 with 1% SDS) for each 300 g powdered sample of rat liver. The protease inhibitors leupeptin 1 /mL, chymostatin 10 g/mL, bestatin 40 /mL, pepstatin A 1 /mL, and N-'␣-p-tosyl-L-lysine chloromethyl ketone 50 g/mL were added to inhibit protein
591
lysis. After centrifugation at 100,000g for 1 hour, the supernatant was collected and its protein concentration estimated, prior to storing it at ⫺80°C until Western blotting was performed. An equal concentration of protein was used for each animal. The protein was combined with an equal volume of loading buffer (0.5M TRIS-HCL—pH 6.8, 20% glycerol, 10% SDS, 0.1M DL-dithiothrietol and 0.05% bromophenol blue) and denatured by boiling for 10 minutes. The proteins in the mixture were separated according to their molecular weight by electrophoresis through a 7.5% SDS-polyacrylamide gel, after which they were transferred to a 0.45 m nitrocellulose membrane (Schleicher, Dassel, Germany). The membranes were then immersed in a blocking solution containing 5% nonfat dry milk before being incubated with the NOS antibodies (monoclonal anti-eNOS or polyclonal anti-iNOS) in TRIS– buffer saline with 0.15% Tween-20 (TBS-T). After 2 hours incubation at room temperature, the membranes were washed (once for 15 minutes and three times for 5 minutes) in TBS-T. They were then incubated for 1 hour with either goat antimouse (for the anti-eNOS antibody) or goat antirabbit (for the anti-iNOS antibody) antisera conjugated with horseradish peroxidase in a 1:13000 dilution with TBS-T. After repeat washes, the proteins were detected using an ECL-chemiluminescence detection kit (Amersham, Little Chalfont, United Kingdom). Statistical analysis Continuous data are expressed as means (SEM) and were statistically evaluated using a two-sample t test. Fisher’s exact test was used for the analysis of categorized outcome (ie, routine histology and immunohistochemistry).
Results Experimental animal model In all animals within the I-R group, the left lateral and median lobes assumed an obvious ischemic appearance after inflow occlusion, when compared with the other lobes that were still being perfused. This appearance was not associated with mesenteric congestion since portosystemic flow was maintained via the intact circulation through the right lateral and caudate lobes. Mean arterial blood pressure was not affected by either the onset of ischemia or reperfusion, remaining above 100 mm Hg throughout the procedure in all animals. Liver transaminases: aspartate and alanine transaminase plasma levels were markedly higher in the I-R group (AST: 2,159 IU/L (177); ALT: 3,292 IU/L (369)) when compared with the control group (AST: 169 IU/L (15); ALT: 139 IU/L (20)) at the end of the experiment (P ⬍0.0001 for each enzyme). Hematoxylin and eosin staining: all animals in the I-R
592
F. Serracino-Inglott et al. / The American Journal of Surgery 185 (2003) 589 –595
Hepatic immunohistochemistry Expression of eNOS: in all liver samples from the control group, staining with the anti-eNOS antibody revealed pronounced staining of all the endothelial cells, ie, those lining both the blood vessels and the sinusoids (Fig. 1C). In all specimens from the I-R group, staining of sinusoidal endothelial cells in the centrilobular region was absent (P ⫽ 0022), while staining of endothelial cells lining blood vessels and sinusoids in the periphery of the lobule was much less intense than in samples from control animals (Fig. 1D). Expression of iNOS: none of the livers from the control group stained with anti-iNOS antibody (Fig. 1E) while five of six livers from the I-R group showed positive staining (P ⫽ 0125). This staining was localised to hepatocytes in the centrilobular region (Fig. 1F) and inflammatory cells throughout the liver sections. Staining was more pronounced in injured hepatocytes. Nitrotyrosine expression: there was clear positive staining (Fig. 1H) of injured hepatocytes with the antinitrotyrosine antibody in five liver sections of the I-R group, compared with no staining (Fig. 1G) in any section from the control group (P ⫽ 0125). Perivenular staining was also present in the I-R group. Aberrations: there was one section in the I-R group which failed to stain for nitrotyrosine, and another section from the same animal failed to stain for iNOS. Histological examination of these sections also showed less evidence of cellular damage. The different response in this animal could possibly be due to aberrant hepatic blood supply, which was missed at laparotomy, resulting in decreased ischemic insult. Fig. 1. (A) Section from the control group showing normal liver histology, and (B) a section from the intervention group demonstrating the extent of hepatic injury after ischemic-reperfusion ([I-R] original magnification ⫻200). (C) Section from the control group showing normal expression of endothelial nitric oxide synthase (eNOS) in the sinusoidal endothelial cells, and (D) and its absence in sections from the I-R group (original magnification ⫻400). (E) Absence of staining with anti-inducible nitric oxide synthase (iNOS) antibody in a section from the control group, and (F) its presence in the centrilobular region of a section from the I-R group (original magnification ⫻400). (G) Section from the control group showing absence of staining with antinitrotyrosine antibody, as opposed to (H) a section from the I-R group demonstrating staining of hepatocytes close to the central vein (original magnification ⫻200). CV ⫽ central vein.
group had evidence of moderate hepatic injury, with the microscopic features of centrilobular hepatocyte necrosis, trabecular derangement, and a polymorphonuclear cell infiltrate (Fig. 1A and B). In the preserved areas of the liver, sinusoids were congested and had decreased diameter. Liver samples from the control group showed no evidence of hepatic injury.
Western blotting In control animals, Western blotting of liver homogenates revealed an intense band at 135 kDa which stained with the anti-eNOS antibody (Fig. 2A) and a very faint band at 130 kDa which stained with the anti-iNOS antibody (Fig. 2B). In the I-R group, Western Blotting of liver homogenates revealed a much less intense band at 135 kDa when compared with that of liver homogenates from the control group (Fig. 2A). This was consistent with the immunnohistochemistry findings, and confirmed that a lesser concentration of eNOS protein is present in the livers of those animals subjected to hepatic I-R when compared with rats who only underwent sham laparotomy. Conversely, the 130 kDa band representing iNOS protein was much more intense in liver extracts from the I-R group when compared with extracts from the control group (Fig. 2B). This was also in keeping with the above immunohistochemistry findings.
F. Serracino-Inglott et al. / The American Journal of Surgery 185 (2003) 589 –595
Fig. 2. Western blots of crude liver homogenates using anti-endothelial nitric oxide synthase (eNOS) antibody (A) or anti-inducible nitric oxide synthase (iNOS) antibody (B) from a rat subjected to sham laparotomy (control) and from a rat subjected to 45 minutes of ischemia followed by 6 hours of reperfusion (I-R). The 135 kDa band (A) is less intense in the liver from the I-R group, while the 130 kDa band (B) is much more intense in the liver from the I-R animal.
Comments Decreased production of NO in the early reperfusion period has been largely attributed to a decrease in the concentration of oxygen and co-factors such as NADPH required for eNOS activity [29] and a decrease in L-arginine as a result of increased release of arginase [30]. The current study has shown that eNOS expression is downregulated by 6 hours from the onset of reperfusion, adding an additional explanation for the decrease in NO production during this period. There is little doubt that this contributes to hepatic injury, since the administration of L-arginine or NO donors prior to ischemia or during the early reperfusion period attenuate hepatic reperfusion injury [21,22]. Nitric oxide donors are able to release NO in the absence of NOS enzymatic activity. In animal models, NO donors have been found to be much more effective than L-arginine in attenuating hepatic reperfusion injury [22]. Since Larginine is converted to NO by NOS, the decreased expression of eNOS demonstrated in this study could possibly explain that report. Increasing the dose of L-arginine has no effect on NO production within the immediate reperfusion period [22]. This could also be explained by the finding of decreased eNOS expression. The scanty eNOS present during early reperfusion will have all its active sites bound by a low concentration of L-arginine. Increasing this concentration will have no effect since there will be no free binding sites as a result of too little eNOS enzyme for the extra
593
L-arginine molecules. This situation will change by 6 hours from the onset of reperfusion, by which time iNOS would have been expressed and will provide binding sites for the extra L-arginine. A minority of researchers believes that NO levels are actually increased, instead of decreased, during the early reperfusion period. This theory is based on the finding that the administration of a NOS inhibitor in the pig has no effect on portal flow or pressure under normal conditions [31]. This suggests that basal NO production is negligible and therefore an I-R insult cannot lower it any further. Since, on the other hand, the administration of NOS inhibitors during hepatic I-R significantly aggravates the injury [32,33], NO production during this period must be significant. If this were to be the case, one would expect the expression of eNOS to be increased or, at the least, remain unchanged after hepatic ischemia. The decreased expression of eNOS demonstrated in the current study refutes the above hypothesis. Since eNOS expression is known to vary in response to changes in shear flow [34], the demonstrated decrease in eNOS expression may be secondary to absence of flow within the sinusoids during the ischemic period. Expression of protein kinase C isoforms in sinusoidal endothelial cells during the ischemic period [35] may also play a part, since there is evidence that protein kinase C activation downregulates eNOS expression [36]. Inducible NOS expression in hepatic I-R is the result of activation of nuclear factor-B, a higher eukaryotic transcriptional factor, secondary to oxidative stress [27]. It is assumed that in hepatic I-R, iNOS expression is not limited to inflammatory cells such as Kupffer cells but also occurs within hepatocytes. This assumption is based on studies that have shown hepatocyte cell cultures to produce NO in response to cytokines [24 –25,26]. However, since whole liver homogenates have been used to study iNOS mRNA expression [27] and iNOS activity [37] in animal models of hepatic I-R, there still was no proof that iNOS expression occurs within hepatocytes during in vivo hepatic I-R. Although the current study also used whole liver homogenates for Western Blot analysis of iNOS, it has revealed intense staining with anti-iNOS antibody within hepatocytes using immunohistochemistry. This confirmed the hypothesis that in vivo iNOS is also expressed in hepatocytes during the later reflow period. It was also noted that iNOS staining was most marked in the centrilobular areas. It has been noted before that these areas are most vulnerable to reperfusion injury because they form the most distant part of the hepatic lobule. Peroxynitrite, a highly reactive free radical, is formed by the very rapid combination of NO with superoxide anion. Under normal conditions, this rection is prevented by the constant removal of NO by hemoglobin [17]. However, during the later stages of reperfusion, NO production is vastly increased as a result of the expression of iNOS [27]. The excessive production of NO from iNOS is therefore
594
F. Serracino-Inglott et al. / The American Journal of Surgery 185 (2003) 589 –595
thought to contribute towards hepatic injury, via the formation of peroxynitrite [29,38,39]. Nitrotyrosine is a product of peroxynitrite mediated protein damage which differs from products resulting from modifications mediated by free radicals other than peroxynitrite [38,39]. The present data, showing positive staining with the antinitrotyrosine antibody in only those animals subjected to hepatic I-R, are strongly suggestive that the expression of iNOS contributes toward mediating hepatic injury. In summary, this study has shown that, during the reperfusion period after hepatic ischemia, eNOS within sinusoidal endothelial cells is downregulated while iNOS is expressed in hepatocytes. We have also shown that expression of iNOS is associated with the formation of nitrotyrosine, a marker of peroxynitrite-induced injury. We therefore conclude that downregulation of eNOS and the expression of iNOS may play an important role in mediating hepatic reperfusion injury. These findings should prove useful in devising strategies to attenuate hepatic I-R injury in the clinical setting.
References [1] Huguet C, Addario-Chieco P, Gavelli A, et al. Technique of hepatic vascular exclusion for extensive liver resection. Am J Surg 1992;163: 602–5. [2] Delva E, Camus Y, Nordlinger B, et al. Vascular occlusion for liver resections. Operative management and tolerance to ischemia: 142 cases. Ann Surg 1989;209:211–18. [3] McCuskey RS, Urbaschek R, Urbaschek B. The microcirculation during endotoxaemia. Cardiovasc Res 1996;32:752– 63. [4] Henderson JM. Liver transplantation and rejection: an overview. Hepato-Gastroenterol 1999;46:1482– 4. [5] Vollmar B, Glasz J, Leiderer R, et al. Hepatic microcirculatory perfusion failure is a determinant for liver dysfunction in warm ischemia-reperfusion. Am J Pathol 1994;145:1421–3. [6] Marzi I, Takei Y, Ru¨ ker M, et al. Endothelin-1 is involved in hepatic sinusoidal vasoconstriction after ischemia and reperfusion. Transpl Int 1994;7:S503– 6. [7] Jaeschke H, Farhood A, Smith CW. Neutrophils contribute to ischemia/reperfusion injury in rat liver in vivo. FASEB J 1990;4:3355–9. [8] Yadav SS, Howell DN, Gao W, et al. L-selectin and ICAM-1 mediate reperfusion injury and neutrophil adhesion in the warm ischemic mouse liver. Am J Physiol 1998;275:G1341–52. [9] Menger MD, Sack FU, Barker JH, et al. Quantitative analysis of microcirculatory disorders after prolonged ischemia in skeletal muscle: therapeutic effects of prophylactic isovolaemic haemodilution. Res Exp Med 1988;188:151– 65. [10] Coletti LM, Kunkel SL, Walz A, et al. The role of cytokine networks in the local liver injury following hepatic ischemia/reperfusion in the rat. Hepatology 1996;23:506 –14. [11] Shirasugi N, Wakabayashi G, Shimazu M, et al. Up-regulation of oxygen derived free radicals by interleukin-1 in hepatic ischemia/ reperfusion injury. Transplantation 1997;64:1398 – 403. [12] Palmer RMJ, Ferrige AG, Moncada S. Nitric oxide release accounts for the biological activity of endothelium-derived relaxing factor. Nature 1987;327:524 – 6. [13] Clancy RM, Leszczynska-Piziak J, Abramson SB. Nitric oxide, an endothelial cell relaxation factor, inhibits neutrophil superoxide production via a direct action on the NADPH oxidase. J Clin Invest 1992;90:1116 –21.
[14] Salvemini D, Radziszewski W, Korbut R, Vane J. The use of oxyhaemoglobin to explore the events underlying the inhibition of platelet aggregation induced by NO or NO-donors. Br J Pharmacol 1990; 101:991–5. [15] De Caterina R, Libby P, Peng H-B, et al. Nitric oxide decreases cytokine-induced endothelial activation. Nitric oxide selectively reduces endothelial expression of adhesion molecules and pro-inflammatory cytokines. J Clin Invest 1995;96:60 – 8. [16] Liu P, Xu B, Hock CE, et al. Nitric oxide modulates P-selectin and ICAM-1 mRNA expression and hemodynamic alterations in hepatic I/R. Am J Physiol 1998;275:H2191– 8. [17] Pryor WA, Squadrito GL. The chemistry of peroxynitrite: a product from the reaction of nitric oxide with superoxide. Am J Physiol 1995;268:L699 –722. [18] Wang Y, Marsden PA. Nitric oxide synthases: gene structure and regulation. Adv Pharmacol 1995;34:71–9. [19] Mathie RT. The hepatic haemodynamic effects of nitric oxide. In: Mathie RT, Griffith TM, editors. The haemodynamic effects of nitric oxide. 1st ed. London: Imperial College Press, 1999, p 22–51. [20] Shah V, Haddad FG, Garcia-Cardena G, et al. Liver sinusoidal endothelial cells are responsible for nitric oxide modulation of resistance in the hepatic sinusoids. J Clin Invest 1997;100:2923–30. [21] Shimamura T, Zhu Y, Zhang S, Jin MB, et al. Protective role of nitric oxide in ischemia and reperfusion injury of the liver. J Am Coll Surg 1999;188:43–52. [22] Ohmori H, Dhar DK, Nakashima Y, et al. Beneficial effects of FK409, a novel nitric oxide donor, on reperfusion injury of rat liver. Transplantation 1998;66:579 – 85. [23] Dhar DK, Yamanoi A, Ohmori H, et al. Modulation of endothelin and nitric oxide: a rational approach to improve canine hepatic microcirculation. Hepatology 1998;28:782– 8. [24] Kitade H, Sakitani K, Inoue K, et al. Interleukin 1 markedly stimulates nitric oxide formation in the absence of other cytokines or lipopolysaccharides in primary cultured rat hepatocytes but not in Kupffer cells. Hepatology 1996;23:797– 802. [25] Geller DA, de Vera ME, Russell DA, et al. A central role for IL-1 in the in vitro and in vivo regulation of hepatic inducible nitric oxide synthase. J Immunol 1995;155:4890 – 8. [26] Shiratori Y, Ohmura K, Hikiba Y, et al. Hepatocyte nitric oxide production is induced by Kupffer cells. Dig Dis Sci 1998;43:1737– 45. [27] Hur GM, Ryu YS, Yun HY, et al. Hepatic ischemia/reperfusion in rats induces iNOS gene transcription by activation of NF-B. Biochem Biophys Res Commun 1999;261:917–22. [28] Isobe M, Katsuramaki T, Hirata K, et al. Beneficial effects of inducible nitric oxide synthase inhibitor on reperfusion injury in the pig liver. Transplantation 1999;68:803–13. [29] Stewart AG, Barker JE, Hickey MJ. Nitric oxide in ischemia-reperfusion injury. In: Grace PA, Mathie RT, editors. Ischemia-reperfusion injury. 1st ed. Oxford: Blackwell Science, 1999, p 180 –95. [30] Shiraishi M, Hiroyasu S, Nagahama M, et al. Role of exogenous L-arginine in hepatic ischemia-reperfusion injury. J Surg Res 1997; 69:429 –34. [31] Ayuse T, Brienza N, Revelly JP, et al. Role of nitric oxide in porcine liver circulation under normal and endotoxemic conditions. J Appl Physiol 1995;78:1319 –29. [32] Koeppel TA, Thies JC, Schemmer P, et al. Inhibition of nitric oxide synthesis in ischemia/reperfusion of the rat liver is followed by impairment of hepatic microvascular blood flow. J Hepatol 1997;27: 163–9. [33] Cottart CH, Do L, Blanc MC, et al. Hepatoprotective effect of endogenous nitric oxide during ischemia-reperfusion in the rat. Hepatology 1999;29:809 –13. [34] Griffith TM. Shear stress and nitric oxide release: physiological integration of cellular mechanisms, physical forces and flow regulation. In: Mathie RT, Griffith TM, editors. The haemodynamic effects
F. Serracino-Inglott et al. / The American Journal of Surgery 185 (2003) 589 –595 of nitric oxide. 1st ed. London: Imperial College Press, 1999, p 22–51. [35] Davies MG, Huynh TTT, Hagen P-O. Endothelial physiology. In: Grace PA, Mathie RT, editors. Ischemia-reperfusion injury. 1st ed. Oxford: Blackwell Science, 1999, p 157–79. [36] Ohara Y, Sayegh HS, Yamin JJ, Harrison DG. Regulation of endothelial constitutive nitric oxide synthase by protein kinase C. Hypertension 1995;25:415–20.
595
[37] Peralta C, Closa D, Hotter G, et al. Liver ischemic preconditioning is mediated by the inhibitory action of nitric oxide on endothelin. Biochem Biophys Res Commun 1996;229:264 –70. [38] Ischiropoulos H, Al-Mehdi AB. Peroxynitrite-mediated oxidative protein modifications. FEBS Lett 1995;364:279 – 82. [39] Van Der Vliet A, Eiserich JP, O’Neil CA, et al. Tyrosine modification by reactive nitrogen species. A closer look. Arch Biochem Biophys 1995;319:314 –19.
