Protective Effect of Preconditioning on the Injury Associatedto Hepatic Ischemia-Reperfusion in the Rat: Role of Nitric Oxide

and Adenosine

CARMEN PERALTA, GEORGINA HOTTER, DANIEL CLOSA, EMILI GELPI, ORIOL BULBENA, AND JOAN ROSELLO-CATAFAU

Both adenosine and NO are synthesized in vascular endo-The effects of ischemic preconditioning on rat liverthelium and are released into the surrounding vascular andintegrity, as well as the implication of nitric oxide (NO)interstitial compartments during the ischemic and reperfu-and adenosine in this process, has been evaluated. Pre-sion periods. Experimental data suggest that NO is a majorconditioning before ischemia-reperfusion prevented themediator of adenosine-induced vasodilatation.9 In addition,increases in alanine transaminase (ALT), aspartaterecent evidence suggests that adenosine stimulates, by a re-transaminase (AST), and lactate dehydrogenase (LDH)ceptor-mediated mechanism, the production of NO by endo-levels, but did not modify blood flow. Adenosine or NOthelial cells.10 On the other hand, it has also been reportedadministration previous to hepatic ischemia-reperfu-that NO could modulate the adenosine release.11sion simulated the effect of preconditioning, whereas

Ischemic preconditioning has been commonly studied ininhibition of adenosine or NO synthesis abolished thethe heart, but few studies have been performed on liver pre-protective effect of hepatic preconditioning. Neverthe-conditioning.12 In this study, the concerted roles of NO andless, inhibition of adenosine and simultaneous adminis-adenosine in the protective effect of preconditioning againsttration of NO in preconditioned animals offered similarhepatic ischemia-reperfusion–induced injury have beenresults to those found in the preconditioned group, indi-studied.cating that, in the absence of adenosine, NO is able to

maintain the preconditioning benefits. It is suggestedMATERIALS AND METHODSthat, in preconditioning, adenosine stimulates NO pro-

duction to protect against the injury associated with Surgical Procedure. Male Wistar rats (8 in each group; 250-300 gbody weight) were used. All animals (including controls) were anes-ischemia-reperfusion. (HEPATOLOGY 1997;25:934-937.)thetized with urethane (10 mg/kg intraperitoneally) and placed in asupine position on a heating pad for maintenance of body tempera-

It has been reported that repetitive, short periods of is- ture between 367C and 377C. To induce hepatic ischemia, laparotomychemia, separated by intermittent reperfusion, render the was performed, and the blood supply to the right lobe of the liverheart more tolerant to subsequent, longer ischemic episodes, was interrupted by placement of a bulldog clamp at the level of the

hepatic artery and portal vein. Reflow was initiated by removal ofand attenuate the injury observed after ischemia-reperfu-sion.1-3 This protective effect has been called ischemic precon-ditioning.

Although the mechanism of ischemic preconditioning is notyet known, potential mediators include nitric oxide (NO),adenosine, prostacyclin, or bradykinin.3

Production of NO has been shown to influence organ integ-rity in response to different challenges. In this sense, it isknown that L-arginine or NO donors provide significant pro-tection in ischemia-reperfusion.4 In addition, inhibition of NOcan induce most of the alterations elicited by this process.5,6

Adenosine is an endogenous compound produced by theaction of various enzymes on adenosine triphosphate, adeno-sine diphosphate, and adenosine monophosphate. Duringischemia, cellular consumption of adenosine triphosphateleads to accumulation of adenosine. This increases the levelof extracellular adenosine, which is believed to confer cyto-protection to the ischemic tissue.7 Although the exact mecha-nism underlying this cytoprotective effect is not completelyunderstood, some authors have reported that activation ofadenosine receptors may play an important role.8 FIG. 1. Experimental protocol. I/R: animals subjected to 90 minutes of

ischemia, followed by 90 minutes of reperfusion. Prec.: preconditioned group,induced by 10 minutes of ischemia and 10 minutes of reperfusion before theischemic period. I/R/NO: animals subjected to 90 minutes of ischemia treatedwith a NO donor, 5 minutes before ischemia. Prec. / NAME: preconditionedAbbreviations: AST, aspartate transaminase; ALT, alanine transaminase; LDH, lactate

dehydrogenase; NAME, N-nitro-L-arginine methyl ester; ADA, adenosine deaminase. group treated with NAME 5 minutes before preconditioning. I/R / Ado: ani-mals subjected to 90 minutes of ischemia treated with a continous infusion ofFrom the Department of Medical Bioanalysis, Institut d’Investigacions Biomediques de

Barcelona, CSIC, Barcelona, Spain. adenosine during 20 minutes before ischemia. Prec. / ADA: preconditionedgroup treated with ADA 20 minutes before preconditioning and for all precondi-Received May 30, 1996; accepted October 29, 1996.

Supported by grant no. 95/1009. Fondo de investigaciones sanitarias. tioning processes. Prec. / Ado / NAME: preconditioned group treated withNAME and a continous infusion of adenosine for 20 minutes before precondi-Address reprint requests to: Georgina Hotter, Ph.D., Department of Medical Bioanalysis,

IIBB, CSIC, C/ Jordi Girona, 18-26, 08034 Barcelona, Spain. Fax: 34-3-2045904. tioning and for all preconditioning processes. Prec. / NAME / NO: precondi-tioned group treated with a NO donor and a continuous infusion of ADA 20Copyright q 1997 by the American Association for the Study of Liver Diseases.

0270-9139/97/2504-0024$3.00/0 minutes before preconditioning and for all preconditioning processes.

934

AID Hepa 0015 / 5p1e$$$281 03-11-97 17:02:23 hpta WBS: Hepatology

HEPATOLOGY Vol. 25, No. 4, 1997 PERALTA ET AL. 935

FIG. 2. ALT, AST, and LDH levels (U/L) in the experimental groups(n Å 8). *P õ .05 vs. control; /P õ .05 vs. I/R.; and oP õ .05 vs. Prec.

the clamp.12 Blood samples were obtained 90 minutes after reperfu- spermine NONOate (10 mg/kg intravenously in phosphate-bufferedsaline [pH 7.4]) 5 minutes before ischemia.sion and processed to determine plasma aminotransferases (aspar-

tate transaminase [AST], alanine transaminase [ALT], and lactate Group 5. Preconditioning / N-nitro-L-arginine methyl ester(NAME) (Prec./ NAME): Animals subjected to 90 minutes of isch-dehydrogenase [LDH]). The studies performed were in concordance

with the European Union regulations for animal experiments. emia with previous preconditioning were treated with L-NAME (10mg/kg intravenously) 5 minutes before preconditioning.Experimental Design. In the initial series of experiments, the pro-

tective effect of ischemic preconditioning was tested in the following Group 6. Ischemia-reperfusion / adenosine (I/R / Ado): Animalssubjected to 90 minutes of ischemia were treated with a continuousgroups (Fig. 1):

Group 1. Control: animals subjected to anesthesia and laparot- infusion of adenosine dissolved in bicarbonate-buffered saline (pH7.4) (0.066 mL/min, portal vein) during 20 minutes of previous is-omy.

Group 2. Ischemia-reperfusion (I/R): animals subjected to 90 min- chemia.Group 7. Preconditioning / adenosine deaminase (ADA)(Prec./utes of right-lobe hepatic ischemia, followed by 90 minutes of reperfu-

sion. ADA): Animals subjected to 90 minutes of ischemia with previouspreconditioning were treated with a continuous infusion of ADA dis-Group 3. Preconditioning (Prec.): Before the ischemic period (as

in group 2), animals were subjected to 10 minutes of ischemia and solved in bicarbonate-buffered saline (pH 7.4) (0.066 mL/min, portalvein) 20 minutes before preconditioning and for all preconditioning10 minutes of reperfusion.

In a second series of experiments, the role of NO and adenosine processes.Group 8. Preconditioning / adenosine / NAME (Prec. / Ado /in the preconditioning process was evaluated in the following groups:

Group 4. Ischemia-reperfusion / NO donor (I/R/NO): Animals NAME) . Animals subjected to 90 minutes of ischemia with previouspreconditioning were treated with L-NAME (10 mg/kg intravenously)subjected to 90 minutes of ischemia were treated with the NO-donor

AID Hepa 0015 / 5p1e$$$281 03-11-97 17:02:23 hpta WBS: Hepatology

936 PERALTA ET AL. HEPATOLOGY April 1997

and a continuous infusion of adenosine dissolved in bicarbonate-buffered saline (pH 7.4) (0.066 mL/min, portal vein) for 20 minutesbefore preconditioning and for all preconditioning processes.

Group 9. Preconditioning / ADA / NO donor (Prec./ADA/NO):Animals subjected to 90 minutes of ischemia with previous precondi-tioning were treated with the NO donor, spermine NONOate (10mg/kg intravenously in phosphate-buffered saline [pH 7.4]), and acontinuous infusion of ADA dissolved in bicarbonate-buffered saline(pH 7.4) (0.066 mL/min, portal vein) 20 minutes before precondi-tioning and for all preconditioning processes.

Enzymatic Determinations. Evaluation of hepatic injury was per-formed by determinations of AST, ALT, and LDH in blood plasmausing a commercial kit from Boehringer Mannheim (Munich, Ger-many).

Hepatic Blood Perfusion Measurement. Hepatic microcirculationwas analyzed using a laser-Doppler blood flowmeter (model LD5000,Transonic Systems Inc., Ithaca, NY). A fiberoptic probe was posi-tioned against the surface of the right lobe of the liver for monitoringhepatic capillary blood flow. The laser-Doppler flowmeter providescontinuous measurement of relative changes in microcirculatoryblood flow in various tissues.13

Statistical Analysis. Data are expressed as means { SEM andwere statistically evaluated by ANOVA, followed by the Student-Newman-Keuls test. An associated probability of Põ .05 was consid-

FIG. 3. Profiles of blood perfusion measured during ischemia and reperfu-ered to be significant.sion in groups 2 (I/R) (s) and 3 (Prec.) (h). Values are expressed as percentagesof the initial blood flow. Differences between these groups are not significant.RESULTS

As shown in Fig. 2, significant increases in AST, ALT, andLDH levels were observed in the group subjected to ischemia ences in blood perfusion as a consequence of preconditioningand reperfusion (I/R) with respect to the control group. When (Fig. 3). Several theories have been suggested to explain theischemia was preceded by 10 minutes of ischemia and reper- phenomenon. The possibility that the ischemic tissue pro-fusion periods, the increases in AST and LDH were prevented duces substances against injury has been proposed.15 Poten-and those in ALT were attenuated (Prec.). tial mediators of the protective effects of ischemic precondi-

Administration of adenosine (I/R / Ado) or NO donors (I/ tioning include NO and adenosine. Vegh et al.16 observedR / NO) resulted in the same effects on ALT, AST, and that, in dog hearts, the marked antiarrhythmic effects of isch-LDH as above. However, infusion of ADA (Prec. / ADA) or emic preconditioning are abolished by an inhibitor of NOinhibition of NO synthesis (Prec. / NAME) prevented the synthesis (l-NAME). Nevertheless, there are conflicting databeneficial effects of preconditioning, except in the case of ALT in the literature. Thus, other authors have reported that l-in the group (Prec. / ADA). NAME does not affect ischemic preconditioning in the iso-

To ascertain the relationship between NO and adenosine, lated perfused rat.17

NO synthesis was inhibited in a group of animals subjected In the case of the liver, our work reveals that administra-to preconditioning and pretreated with adenosine (Prec. / tion of an NO donor before ischemia-reperfusion can preventNAME / Ado). Furthermore, the effects of adenosine were equally the increases in AST and LDH as observed for theabolished with infusion of ADA in another group of animals preconditioned group. Accordingly, inhibition of NO synthe-subjected to preconditioning and pretreated with an NO do- sis abolishes the beneficial effects of preconditioning (Fig. 2).nor (Prec. / ADA / NO). Results show that inhibition of NO As with NO, there are controversial data with respect toabolished the preconditioning effect on liver enzyme levels the role of adenosine in this process. In this regard, adminis-despite adenosine administration, whereas ADA infusion tration of adenosine before ischemia-reperfusion in rabbitplus NO administration failed to abolish the beneficial effect hearts conferred some degree of cytoprotection. In addition,of preconditioning. some authors have reported that administration of an adeno-

With respect to changes in blood flow, Fig. 3 shows the sine receptor antagonist blocks the beneficial effect of is-profiles of blood perfusion measured during ischemia and chemic preconditioning.18 Recent investigations, however,reperfusion in groups 2 (I/R) and 3 (Prec.). Differences be- have shown that postischemic dysfunction in the rat heart istween these groups were not significant. not attenuated by adenosine agonists or exogenous adeno-

sine.19

DISCUSSION We have found that preadministration of adenosine in he-patic ischemia-reperfusion prevents or attenuates the in-Ischemic preconditioning could be defined as a protective

mechanism toward ischemia-reperfusion injury, and it con- creases in ALT, AST, and LDH in the same manner as ob-served in the preconditioned group; however, infusion of ADAsists of previous short periods of ischemia followed by reper-

fusion. Figure 2 shows the ALT, AST, and LDH activity mea- abolishes the beneficial effects of preconditioning (Fig. 2).The mechanism by which adenosine reduces postischemicsured in plasma after hepatic ischemia-reperfusion. All three

parameters showed increased values in the I/R group. In con- injury was not determined in the present study, althoughinduction of NO synthesis is likely to be part of it. Further-trast, when preconditioning was performed before ischemia,

no increases in AST or LDH release were detected, whereas more, the effect of adenosine in inducing NO release by endo-thelial cells is well documented.9,20 Adenosine stimulates thethe increase observed in ALT was attenuated. This fact shows

the protective effect of preconditioning on preventing hepatic release of NO from endothelial cells via adenosine receptoractivation.10 Thus, simultaneous administration of adenosineischemia-reperfusion damage.

The exact mechanism underlying ischemic preconditioning and inhibition of NO synthesis before preconditioning (Prec./NAME/Ado) leads to results comparable with those foundis unknown; however, experimental evidence suggests that

it is unlikely that the protective actions of ischemic precondi- in the I/R group, indicating that, in the absence of NO, adeno-sine was unable to prevent hepatic lesion by the precondi-tioning can be attributed to altered blood flow.14 Along this

line, our results indicate that there are no significant differ- tioning effect. Furthermore, administration of NO donors

AID Hepa 0015 / 5p1e$$$281 03-11-97 17:02:23 hpta WBS: Hepatology

HEPATOLOGY Vol. 25, No. 4, 1997 PERALTA ET AL. 937

10. Li JM, Fenton A, Cutler BS, Dobson JM. Adenosine enhances nitric oxideplus ADA in preconditioned animals (Prec. / ADA / NO)production by vascular cells. Am J Physiol 1995;269:C519-C523.provides results comparable with those found in the precondi- 11. Fischer H, Prast H, Philippu A. Adenosine release in the ventral striatum

tioned group, indicating that, in the absence of adenosine, of the rat is modulated by endogenous nitric oxide. Eur J Pharmacol 1995;275:R5-R6.NO is still able to simulate the preconditioning.

12. Lloris JM, Cejalvo D, Toledo-Pereyra LH, Calvo MA, Suzuki S. Precondi-Overall, these results suggest that the mechanism leadingtioning: effect upon lesion modulation in warm liver ischemia. Transplantto preconditioning in the ischemic liver involves the release Proc 1993;25:3303-3304.

of adenosine, which, in turn, induces the generation of NO. 13. Seino Y, Ohki K, Nakamura T, Tsukamoto H, Takano T, Aramaki T,Okumura H, et al. Pathophysiological characteristics of cutaneous micro-

REFERENCES circulation in patients with liver cirrhosis: relationship to cardiovascularhemodynamics and plasma neurohormonal factors. Microvasc Res 1993;1. Meldrum D, Mitchell MB, Banerjee A, Harken AH. Cardiac precondi-46:206-215.tioning. Arch Surg 1993;128:1208-1211.

14. Cohen MV, Liu GS, Downey JM. Preconditioning causes improved wall2. Liang-Xiong Fu, Kirkeboen KA, Liang QM, Sjorgren KG, Hjalmarson A,motion as well as smaller infarct after transient coronary occlusion inIlebekk A. Free radical scavenging enzymes and G protein mediated recep-rabbit. Circulation 1991;84:341-349.tor signalling systems in ischaemically preconditioned porcine myocar-

15. Parrat JR. Protection of the heart by ischaemic preconditioning: mecha-dium. Cardiovasc Res 1993;27:612-616.nisms and possibilities for pharmacological exploitation. Trends Pharma-3. Parrat JR. Protection of the heart by ischaemic preconditioning: mecha-col Sci 1994;15:19-25.nisms and possibilities for pharmacological exploitation. Tips Reviews

16. Vegh A, Szekeres L, Parrat JR. Preconditioning of the ischemic myocar-1994;15:19-25.dium: involvement of the l-arginine nitric oxide patway. Br J Pharmacol4. Villarreal D, Grisham MB, Granger DN. Nitric oxide donors improve gut1992;107:648-652.function after prolonged hypothermic ischemia. Transplantation 1995;59:

17. Lu HR, Remeysen P, DeClerck F. Does the antiarrhytmic effect of ischemic685-689.preconditioning in rats involve the L-arginine nitric oxide pathway? J5. Kubes P, Suzuki M, Granger DN. Nitric oxide: an endogenous modulatorCardiovasc Pharmacol 1995;25:524-530.of leukotriene adhesion. Proc Natl Acad Sci U S A 1991;88:4651-4657.

18. Van Winkle DM, Downey JM, Thornton JD, Davis RF. Ischemic precondi-6. Kubes P. Nitric oxide modulates epithelial permeability in feline smalltioning on myocardium: effect of adenosine. In: Maruyama, et al. Recentintestine. Am J Physiol 1992;262:G1138-G1144.Advances in Coronary Circulation. Tokyo: Springer-Verlag, 1993:223-234.7. Jenkins DP, Batxer GF, Yellon DM. The pathophysiology of ischaemic

19. Vuorinen K, Ylitalo K. Peuhkurinen K, Raatikaien P, Ala-Rami, Hassinenpreconditioning. Pharmacol Res 1995;31:219-223.IE. Mechanism of ischemic preconditioning in rat myocardium. Circulation8. Ramkumar V, Nie Z, Rybak LP, Maggiwar SB. Adenosine, antioxidant1995;91:2810-2818.enzymes and cytoprotection. Trends Pharmacol Sci 1995;16:283-285.

20. Smits P, Williams SB, Lipson DE, Banitt P, Rongen GA, Creager MA.9. McKie LD, Bass BL, Dunkin BJ, Harmon JW. Nitric oxide mediates theEndothelial release of nitric oxide contributes to the vasodilator effect ofblood flow response to intravenous adenosine in the rabbit. Circ Shock

1994;43:103-106. adenosine in humans. Circulation 1995;92:2135-2141.

AID Hepa 0015 / 5p1e$$$281 03-11-97 17:02:23 hpta WBS: Hepatology