Surg TodayJpn J Surg (1999) 29:1158–1163

Experimental Evaluation of the Effects of the IntraportalAdministration of Cyclic Guanosine Monophosphate onIschemia/Reperfusion in the Porcine Liver

Hideo Matsumoto, Ryuji Hirai, Tadahiro Uemura, Tetsuya Ota, Atsushi Urakami, and Nobuyoshi Shimizu

Department of Surgery II, Okayama University Medical School, 2-5-1 Shikata-cho, Okayama 700-8558, Japan

regulate the sinusoidal flow6 and to modulate variousmetabolic functions in hepatocytes.7,8

The majority of the biological functions of NO ap-pears to be mediated primarily by the activation ofsoluble guanylate cyclase (sGC) concomitant with acyclic guanosine monophosphate (cGMP) increase.9

Through this pathway, endothelium-derived NO in-duces the relaxation of vascular smooth muscle andinhibits platelet aggregation.10

Unfortunately, NO is also cytotoxic in some circum-stances. In ischemia/reperfusion, macrophages, neu-trophils, and endothelial cells may be activated toproduce both superoxide and NO, which leads to formperoxynitrite and hydroxy radicals,11 causing tissue in-jury.5,8 These reactions are the direct effects of NO,being independent of cGMP.11,12

The purpose of this study was to examine the effect ofcGMP on warm ischemia/reperfusion injury in the por-cine liver, using a membrane-permeable cGMP analog,8-bromoguanosine 39, 59 monophosphate (8-Br-cGMP).We therefore assumed that the supplementation ofcGMP instead of NO or NO donors might abate vascu-lar dysfunction by modulating the vasodilatation ofmicrocirculation without producing toxic radicals byischemia/reperfusion in the porcine liver.

Materials and Methods

Surgical Preparation

A total of 12 white pigs, weighing 18.6–28.9kg, werefasted for 24 h with water available, and anesthetizedwith intramuscular pentobarbital sodium (10mg/kg)and ketamine hydrochloride (20mg/kg). An ear veinwas then cannulated, and anesthesia was maintainedwith continuous infusion of pentobarbital sodium(1.0mg/kg per h) and ketamine hydrochloride (1.0 mg/kg per h). Respiration was mechanically controlled

Abstract: This study was done to examine the protectiveeffects of cyclic guanosine monophosphate (cGMP), a secondmessenger of nitric oxide, for ischemia/reperfusion injury ofthe liver, since it is known to induce vasodilatation and toinhibit platelet aggregation. Using an experimental model ofporcine liver ischemia, 8-bromoguanosine 3¢,5¢ monophos-phate, a cGMP analog, was continuously administered intothe portal vein before ischemia and after reperfusion 30minfor each in the cGMP group (n 5 6). Saline water was admin-istered in the same way in the control group (n 5 6). Thecardiac output (CO), mean arterial blood pressure (MAP),portal venous flow (PVF), hepatic arterial flow (HAF), he-patic tissue blood flow (HTBF), and hepatic tissue cGMPlevel were determined. Hepatic enzymes and the bile dis-charge were also assessed as indicators of hepatic function.The hepatic tissue cGMP level was significantly higher, andPVF, HTBF, and the bile discharge were significantly greaterin the cGMP group, while there were no remarkable differ-ences between the groups with CO, MAP, HAF, and hepaticenzymes. In conclusion, the continuous supplementation ofcGMP into the portal vein was found to be beneficial forpreserving both the hepatic circulation and, consequently, thehepatic function after warm ischemia of porcine liver.

Key Words: cyclic guanosine monophosphate, nitric oxide,ischemia/reperfusion injury, microcirculation, bile flow

Introduction

Nitric oxide (NO) is well known to be an endothelialrelaxing factor,1,2 and is reported to regulate the micro-circulation3 and to be related with the pathogenesis ofvarious liver diseases.4 NO is produced by all majorcells, including hepatocytes, Kupffer cells, endothelialcells, and stellate cells.5 In the liver, NO was shown to

Reprint requests to: H. Matsumoto(Received for publication on July 21, 1998; accepted on May27, 1999)

using a respirator after tracheotomy. The right carotidartery was cannulated for the measurement of meanarterial blood pressure (MAP) and the withdrawal ofblood samples. Via the right subclavian vein, a ther-modilution cardiac output (CO) catheter was insertedfor the measurement of CO and infusion of lactateRinger solution supplemented with sorbitol (10 ml/kgper h).

After laparotomy, the common bile duct was can-nulated for the measurement of bile outflow. Aftersplenectomy, a bypass tube (Anthron bypass tube VTT-4860, Toray, Tokyo, Japan) was placed between thesplenic vein and left jugular vein for portal drainage.For the measurement of the portal venous pressure(Ppv), a catheter was inserted into the portal vein viathe side branch to within 1 cm of the hilum of the liver.Blood-flow probes were placed around the commonhepatic artery and the portal vein. Two laser Dopplerflow probes were also placed on the surface of the rightand left hepatic lobes. After the completion of surgicalpreparations, the abdominal wall was closed to mini-mize the loss of body temperature and body fluid.

Next, severe hepatic ischemia was induced by cross-clamping the hepatoduodenal ligament for 60 min. 8-Br-cGMP (Sigma, St. Louis, MO, USA) sodium salt(1.0mg/kg per h) was administered into the portal vein30 min before and after ischemia for each group of sixpigs (cGMP group), while saline water was adminis-tered in the same way to the other six (control group).

Evaluation of Hemodynamics

PVF and HAF were measured with ultrasound transittime flow probes (T201 Transonic Systems, Advance,Tokyo, Japan) and the hepatic tissue blood flow (HTBF)was estimated by a laser Doppler flowmeter (ALF21D,Advance). The blood pressure was measured with pres-sure transducers (San-ei Polygraph 360 system, NEC,Tokyo, Japan). These signals were all recorded by a penrecorder. PVF, HAF, and HTBF are presented as apercentage of each value before ischemia.

The blood pressure and blood flow were measuredat preischmia, during ischemia, at 5min after the startof reperfusion, and every 30 min up to 180 min afterthe start of reperfusion. The CO was measured atpreischemia and at every 60 min up to 180min after thestart of reperfusion.

Measurement of Liver Enzymes

As the indicator of ischemic hepatic damage, asparateaminotransferase (AST) and lactate dehydrogenase(LDH) were assessed 60, 120, and 180 min after com-mencement of reperfusion, and compared with eachvalue before ischemia.

Measurement of the Discharge

As an indicator of global liver function, the bile dis-charge was examined, collected before ischemia, and at60, 120, and 180 min after the start of reperfusion, ex-pressed as a percentage of each value before ischemia.Bile was collected from the cannula at 30-min intervalsat each corresponding point.

Measurement of cGMP Level in Hepatic Tissue

Wedge liver biopsies collected before ischemia and at60, 120, and 180min after the start of reperfusion wereused to measure the cGMP level in the hepatic tissueusing a cGMP assay kit (Yamasa Soy Sauce, Tokyo,Japan).

Data Analysis

A statistical analysis was done using Fisher’s t-test andrepeated-measures ANOVA to compare the variablesbetween the two groups. The pressure data, serum lev-els of liver enzymes, and cGMP levels of hepatic tissueare expressed as mean 6 SD, and other data are ex-pressed as mean 6 SEM. When a P value was less than0.05, the difference was considered to be statisticallysignificant.

All animals received humane care in accordance withthe Principles of Laboratory Animal Care formulatedby the National Society of Medical Research, and theGuide of the National Academy of Sciences, publishedby the U.S. National Institutes of Health (NIH Publica-tion no. 80-23, revised in 1978).

Results

MAP, Ppv, and CO

The MAP of both groups did not change significantlythroughout the experiment. The CO decreased duringthe ischemia in both groups but recovered afterreperfusion in both groups, showing no significant dif-ference between the groups. On the other hand, Ppv ofthe cGMP group was generally lower than that of thecontrol group and the difference was significant at 5minafter reperfusion (Table 1).

PVF and HAF

PVF decreased to 72.5% 6 9.4% and 56.9% 6 5.5% ofeach control value after 120min of reperfusion in thecGMP group and control group, respectively. Conse-quently, there was a significant difference between thetwo groups after 120 min of reperfusion (P , 0.05 foreach time point). The PVF in the cGMP group was

H. Matsumoto et al.: cGMP Effects on Ischemia/Reperfusion in Liver 1159

significantly higher than that of the control group (P 50.044). The PVF in the control group gradually de-creased after reperfusion, whereas that in the cGMPgroup was preserved at the same level after reperfusion(Fig. 1). HAF increased after reperfusion in bothgroups, reaching 153.3% 6 18.7% and 130.3% 6 14.9%at the 180-min time point in the cGMP group andcontrol group, respectively (Fig. 2). HAF in the cGMPgroup tended to be higher than that of the control groupat each time point, but the difference was notsignificant.

HTBF

HTBF was 99.5% 6 7.3% and 71.6% 6 9.0% of eachcontrol value after 120min of reperfusion in the cGMPand control groups, respectively. There was a significantdifference between the two groups after 120min ofreperfusion (P , 0.05 for each time point) (Fig. 3).HTBF after reperfusion in the cGMP group was higherthan that of the control group, and the difference wassignificant (P 5 0.017). HTBF of the cGMP group waspreserved at the same level after reperfusion.

Liver Enzyme Levels and Bile Discharge

There was no significant difference between the twogroups regarding the serum level of liver enzymes(Table 2). The percentage change of the bile dischargewas 96.5% 6 17.1% and 51.1% 6 26.4% at 120min, and

Table 1. Hemodynamic variables

45min after 5min after 60min after 120min after 180min afterGroup Before ischemia start of ischemia reperfusion reperfusion reperfusion reperfusion

CO (l/min) Control 2.27 6 0.24 1.77 6 0.49 2.53 6 0.40 2.407 6 0.38 2.43 6 0.33 2.31 6 0.51cGMP 2.71 6 0.58 1.73 6 1.04 2.54 6 0.42 2.32 6 0.37 2.25 6 0.57 2.31 6 0.61

MAP (mmHg) Control 95.8 6 4.1 95.4 6 7.5 101.7 6 12.6 96.3 6 12.6 95.8 6 8.1 93.8 6 10.6cGMP 98.5 6 11.5 102.1 6 11.7 102.5 6 10.9 102.5 6 12.1 101.9 6 9.8 99.2 6 9.2

Ppv (mmHg) Control 8.1 6 0.7 — 11.7 6 2.8 8.7 6 1.5 8.9 6 1.9 9.2 6 2.7cGMP 7.8 6 0.8 — 8.7 6 1.1* 8.1 6 1.3 8.0 6 1.6 8.3 6 1.2

CO, cardiac output; MAP, mean arterial pressure; Ppv, portal venous pressure. Values are expressed as mean 6 SD*P , 0.05 vs control value

Fig. 1. Changes in the portal venous flow (PVF ) during is-chemia and after reperfusion. Open circles, cyclic guanosinemonophosphate (cGMP); closed circles, control. Bars expressSEM. *P , 0.05 vs control

Fig. 2. Changes in the hepatic arterial flow (HAF ) duringischemia and after reperfusion. Open circles, cGMP; closedcircles, control. Bars express SEM

Fig. 3. Changes in the hepatic tissue blood flow (HTBF ) dur-ing ischemia and after reperfusion. Open circles, cGMP;closed circles, control. Bars express SEM. *P , 0.05 vs control

1160 H. Matsumoto et al.: cGMP Effects on Ischemia/Reperfusion in Liver

96.8% 6 14.7% and 46.3% 6 8.7% at 180 minafter reperfusion in the cGMP and control groups,respectively (P , 0.05 for each) (Fig. 4). The biledischarge after reperfusion in the cGMP group dem-onstrated a significant difference after 120 min ofreperfusion.

cGMP Level of the Hepatic Tissue

Before ischemia, there was no significant difference inthe cGMP level of hepatic tissue between the twogroups (Fig. 5). The level in the cGMP group increasedremarkably after reperfusion as compared with thepreischemic level, and the difference was significant:0.057 6 0.002 vs 1.471 6 0.566, 0.690 6 0.469, and 0.4906 0.225pg/mg protein at 60min, 120 min, and 180 minafter reperfusion (P , 0.01 for each), respectively. Thelevel in the control group also increased from 0.047 60.005 pg/mg before ischemia to 0.090 6 0.031pg/mg pro-tein at 60 min after reperfusion, but the difference wasnot significant. As a result, the cGMP level of hepatictissue was significantly higher in the cGMP group thanin the control group at all points of reperfusion (P ,0.01 for each).

Discussion

NO is considered to play a protective role for tissuein the pathogenesis of various liver diseases,13,14 and inischemia/reperfusion it plays a very important role inmaintaining vascular homeostasis,4,6,15 because NOhas functions such as mediating vasodilatation,16 thuspreventing neutrophil adherence to the endothelium,17

maintaining endothelial barrier properties,18 and inhib-iting platelet aggregation.10

However, NO by itself is also known to present cyto-toxicity in some circumstances. Under oxidative stresssuch as inflammation and ischemia/reperfusion, NO andsuperoxide are generated by macrophages, neutrophils,and endothelial cells, and NO may combine rapidly withsuperoxide to form highly toxic peroxynitrite and hy-droxy radical.11,12 Peroxynitrite has been shown to be thesource of a strong oxidizing agent with properties iden-tical to those of the hydroxy radicals, which may directly

Table 2. Liver enzymes

Before 60min after 120min after 180min afterGroup ischemia reperfusion reperfusion reperfusion

AST (IU/l) Control 33.7 6 12.1 118.8 6 69.3 112.0 6 71.3 99.5 6 63.9cGMP 29.6 6 4.2 116.5 6 52.6 118.3 6 60.6 122.2 6 57.6

LDH (IU/l) Control 133.5 6 43.9 171.9 6 17.9 180.8 6 52.1 179.3 6 57.3cGMP 108.7 6 31.2 156.7 6 39.7 61.2 6 46.9 164.1 6 48.7

AST, aspartate aminotransferase; LDH, lactate dehydrogenase. Values are expressed as mean 6SD

Fig. 4. Changes in the bile discharge. Open circles, cGMP;closed circles, control. Bars express SEM. *P , 0.05 vscontrol

Fig. 5. Changes in the concentration of cGMP in hepatic tis-sue. Shaded bars, cGMP; closed bars, control. Error bars ex-press SD. *P , 0.01 vs control

H. Matsumoto et al.: cGMP Effects on Ischemia/Reperfusion in Liver 1161

cause severe cell injury. This potentially deleterious in-teraction has been reported in the reperfusion injury ofmany organs, such as the heart19 and lung.20

On the other hand, NO is known to be an intercellu-lar mediator. In cells, NO activates soluble guanylatecyclase (sGC), which results in an increase in the cGMPlevel.9 cGMP then develops smooth muscle relaxation(vasodilatation)16 and inhibits platelet aggregation.10

This pathway is not concerned with producing superox-ide and peroxynitrite. Therefore, attention has recentlybeen focused on cGMP in the transplantation of lung20

and in myocardial reperfusion injury.21 Recent findingssuggest that cGMP may play a protective role in hepato-cytes, mediating the damage induced by hypoxia andreactive oxygen species, as a messenger of atrial natri-uretic peptide (ANP).22 It has been suggested that thebeneficial actions of cGMP are not limited to vasodila-tation but also extend to the attenuation of neutrophilinfiltration.17,18 Therefore, we hypothesized that thesupplementation of cGMP instead of NO or NO donorsmight confer the beneficial vascular effects of NO,avoiding their potential toxicity by ischemia/reperfusionin the liver.

We used 8-Br-cGMP, which is a membrane-permeable cGMP analog, in this study. This analog hasa high potency as an activator and a high affinity for thecGMP-binding sites.23 The dosage of 8-Br-cGMP usedin the experiment was based on previous reports.24,25

The portal infusion of cGMP was selected to restrict thiseffect as much as possible in hepatic circulation, sincecGMP has been reported to depress myocardial con-tractility.25,26 The continuous infusion was expected tosupply a sufficient dose of cGMP to the hepatic tissue.Our results showed no significant difference in hemody-namics, including the cardiac output, between thecGMP and control groups, and the liver tissue level ofcGMP was significantly higher in the cGMP group atall time points after reperfusion. We therefore considerthat the dose (1.0mg/kg) and administration route ofcGMP (continuous infusion) may be efficient for evalu-ating the effects of cGMP using a 60-min total hepaticischemia model. However, it remains to be clarifiedas to how cGMP, which passed through the liver andmigrated into the systemic circulation, influenced anyother organs.

Huguet et al.27 reported that the interruption of he-patic blood flow in normothermia is safe for at least60 min in human beings; it was also reported that120min of continuous ischemia was submitted withoutbiochemical and/or pathologic damage in experimentalsurgery in pigs.28 However, regarding the systemic he-modynamics in our preliminary experiment, PVF andHAF were unsettled after 120min total hepatic is-chemia, most likely due to problems with the porto-venous bypass tube. To eliminate such inappropriate

influences on the results, we selected an ischemic timeof 60 min in this study.

We observed that both PVF and HTBF weresignificantly higher in the cGMP group than in the con-trol group. cGMP was thus thought to reduce the portalvascular resistance against the enhanced production ofendothelin,3,6 namely, the contractility of stellate cellswas controlled29 and platelet aggregation was inhib-ited.10 Moreover, the discharge of bile, which was anindicator of liver function, was observed to increaseafter the start of reperfusion in the cGMP group. Theimprovement of HTBF might contribute to the increaseof the bile discharge. A similar study also reported thesupplementation of cGMP to increase the bile flow andbiliary concentration.30

NO suppression of the release of liver enzymes bycGMP was observed in this study. This was most likelydue to the effect that the ischemic time of 60min and theobservation period of the reperfusion of 180min mightbe too short to evaluate the effect of cGMP on the liverenzyme levels. This question should be elsewhere usingother ischemic models.

In conclusion, an intraportal supplementation of8-Br-cGMP ameliorated PVF and HTBF after warmischemia/reperfusion of the porcine liver, with hemody-namic stability. The bile discharge, as an indicator ofliver function, was increased by the cGMP supplemen-tation. We propose that cGMP is beneficial for pre-serving hepatic circulation after surgery involving liverischemia, such as that performed during an extendedhepatectomy and/or liver transplantation.

References

1. Palmer RMJ, Ferrige AG, Moncade S (1987) Nitric oxide releaseaccounts for the biological activity of endothelium-derived relax-ing factor. Nature 327:524–526

2. Ignarro LJ, Buga GM, Wood KS, Byrns RE, Chaudhuri G (1987)Endothelium-derived relaxing factor produced and released fromartery and vein is nitric oxide. Proc Natl Acad Sci USA 84:9265–9269

3. Zhang JX, Pegoli W, Clements MG (1994) Endothelin-1 inducesdirect constriction of hepatic sinusoids. Am J Physiol 266:G624–632

4. Pannen BHJ, Bauer M, Noldge-Schomburg GFE, Zhang JX,Robotham JL, Clemens MG, Geiger KK (1997) Regulationof hepatic blood flow during resuscitation from hemorrhagicshock: role of NO and endothelins. Am J Physiol 272:H2736–H2745

5. Rocky DC, Chung JJ (1996) Regulation of inducible nitric oxidesynthase in hepatic sinusolidal endothelial cells. Am J Physiol271:G260–G267

6. Mittal MK, Gupta TK, Lee FY, Sieber CC, Groszmann RJ (1994)Nitric oxide modulates hepatic vascular tone in normal rat liver.Am J Phisiol 267:G416–G422

7. Curran RD, Ferrari FK, Kispert PH, Stadler J, Stuehr DJ,Simmons RL, Billiar TR (1991) Nitric oxide and nitric oxide-generating compounds inhibit hepatocyte protein synthesis.FASEB J 5:2085–2092

1162 H. Matsumoto et al.: cGMP Effects on Ischemia/Reperfusion in Liver

8. Horton RA, Ceppi ED, Knowles RG, Titheradge MA (1994)Inhibition of hepatic gluconeogenesis by nitric oxide: a compari-son with endotoxic shock. Biochem J 299:735–739

9. Schmidt HHHW, Lohmann SM, Walter U (1993) The nitric oxideand cGMP signal trnsduction system: regulation and mechanismof action. Biochim Biophys Acta 1178:153–175

10. Partrick DA, Moore EE, Offner PJ, Barnett CC, Barkin M,Silliman C (1997) Nitric oxide attenuates platelet-activating factorpriming for elastase in human neutrophils via a cyclic guanosinemonophosphate-dependent pathway. Surgery 122:196–203

11. Hogg N, Darley-Usmar VM, Wilson MT, Moncade S (1992) Pro-duction of hydroxyl radicals from the simultaneous generation ofsuperoxide and nitric oxide. Biochem J 281:419–424

12. Beckman JS, Beckman TW, Chin J, Marshall PA, Freeman BA(1990) Apparent hydroxy radical production by peroxynitrite:implications for endothelial injury from nitric oxide and superoxide. Proc Natl Acad Sci USA 87:1620–1624

13. Nishida J, McCusky RS, McDonnell D, Fox ES (1994) Protectiverole of NO in hepatic microcirculatory dysfunction duringendotoxemia. Am J Physiol 267:G1135–G1141

14. Nanji AA, Greenberg SS, Tahan SR, Fogt F, Loscalzo J,Sadrzadeh SMH, Xie J, Stamler JS (1995) Nitric oxide productionin experimental alcoholic liver disease in the rat: role in protectionfrom injury. Gastroentrology 109:899–907

15. Ayuse T, Brienza N, Revelly JP, Boitnott JK, Robotham JL(1995) Role of nitric oxide in porcine liver circulation undernormal and endotoxemic conditions. Am J Physiol 78:1319–1329

16. Feelisch M, Noack EA (1987) Correlation between nitric oxideformation during degradation of organic nitrates and activation ofguanylate cyclase. Eur J Pharmacol 139:19–23

17. Kubes P, Suzuki M, Granger D (1991) Nitric oxide: an endog-enous modulator of leukocyte adhesion. Proc Natl Acad Sci USA88:4651–55

18. Kubes P, Granger D (1992) Nitric oxide modulates microvascularpermeability. Am J Physiol 262:H611–H615

19. Patel VC, Yellon DM, Singh KJ, Neild GH, Woolfson RG (1993)Inhibition of nitric oxide limits infarct size in the in situ rabbitheart. Biochem Biophys Res Commun 194:234–238

20. Pinsky DJ, Naka Y, Chowdhury NC, Liao H, Oz MC, MichlerRE, Kubaszewski E, Malinski T, Stern DM (1994) The nitricoxide/cyclic GMP pathway in organ transplantation: critical role

in successful lung preservation. Proc Natl Acad Sci USA91:12086–12090

21. Pinsky DJ, Oz MC, Koga S, Taha Z, Broekman MJ, Marcus AJ,Liao H, Naka Y, Brett J, Cannon PJ, Nowygrod R, Malinski T,Stern DM (1994) Cardiac preservation is enhanced in heterotopicrat transplant model by supplementing the nitric oxide pathway.J Clin Invest 93:2291–2297

22. Pella R (1991) The protective effect of atrial natriuretic peptide(ANP) on cells damaged by oxygen radicals is mediated throughelevated cGMP-levels, reduction of calcium-inflow and probablyG-proteins. Biochem Biophys Res Commun 174:549–555

23. Corbin JD, Ogreid D, Miller JP, Suva RH, Jastorff B, DoskelandSO (1985) Studies of cGMP analog specificity and function ofthe two intrasubunit binding sites of cGMP-dependent proteinkinase. J Biol Chem 261:1208–1214

24. Suematsu M, Satoshi K, Sano T, Goda N, Shinoda Y, Ishimaru Y(1994) Carbon monoxide as an endogenous modulator of hepaticvascular reperfusion. Biochem Biophys Res Commun 205:1333–1337

25. Lawson CA, Smerling AJ, Naka Y, Burkhoff D, Dickstein ML,Stern DM, Pinsky DJ (1995) Selective reduction of PVR by inhi-bition of cGMP analogue in a porcine model of pulmonary hyper-tension. Am J Physiol 268:H2056–H2062

26. Flesch M, Kilter H, Cremers B, Lenz O, Sudkamp M, Kuhn-Regnier F, Bohm M (1997) Acute effects of nitric oxide and cyclicGMP on human myocardial contractility. J Pharmacol Exp Ther281:1340–1349

27. Huguet C, Gavelli A, Chieco PA, Bona S, Harb J, Joseph JM,Jobard J, Gamamlia M, Lasserre M (1992) Liver ischemia forhepatic resection: where is the limit? Surgery 111:251–259

28. Nordlinger B, Douvin D, Javaudin L, Block P, Aranda A,Boschat M, Huguet C (1980) An experimental study of survivalafter two hours of normothermic hepatic ischemia. Surg GynecolObstet 150:859–864

29. Kawada N, Tran-Thi TA, Klein H, Decker K (1993) The contrac-tion of hepatic stellate (Ito) cells stimulated with vasoactive sub-stances. Eur J Biochem 213:815–823

30. Trauner M, Nathanson MH, Mennone A, Rydberg SH, Boyer JL(1997) Nitric oxide donors stimulate bile flow and glutathionedisulfide exertion independent of cyclic guanosine 39, 59-cyclicmonophosphate in the isolated perfused rat liver. Hepatology25:263–269

H. Matsumoto et al.: cGMP Effects on Ischemia/Reperfusion in Liver 1163