337

Molecular and Cellular Biochemistry 176: 337–347, 1997.© 1997 Kluwer Academic Publishers. Printed in the Netherlands.

Studies on hepatic injury and antioxidant enzymeactivities in rat subcellular organelles following invivo ischemia and reperfusion

Mahesh Gupta,1 Kazushige Dobashi,1 Eddie L. Greene,2 John K. Orak1

and Inderjit Singh2Departments of 1Pediatrics; 2Internal Medicine, Medical University of South Carolina, Charleston, SC 29425, USA

Abstract

The activities of rat hepatic subcellular antioxidant enzymes were studied during hepatic ischemia/reperfusion. Ischemia wasinduced for 30 min (reversible ischemia) or 60 min (irreversible ischemia). Ischemia was followed by 2 or 24 h of reperfusion.Hepatocyte peroxisomal catalase enzyme activity decreased during 60 min of ischemia and declined further during reperfusion.Peroxisomes of normal density (d = 1.225 gram/ml) were observed in control tissues. However, 60 min of ischemia also produceda second peak of catalase specific activity in subcellular fractions corresponding to newly formed low density immatureperoxisomes (d = 1.12 gram/ml). The second peak was also detectable after 30 min of ischemia followed by reperfusion for 2or 24 h. Mitochondrial and microsomal fractions responded differently. MnSOD activity in mitochondria and microsomalfractions increased significantly (p < 0.05) after 30 min of ischemia, but decreased below control values following 60 min ofischemia and remained lower during reperfusion at 2 and 24 h in both organelle fractions. Conversely, mitochondrial andmicrosomal glutathione peroxidase (GPx) activity increased significantly (p < 0.001) after 60 min of ischemia and was sustainedduring 24 h of reperfusion. In the cytosolic fraction, a significant increase in CuZnSOD activity was noted following reperfusionin animals subjected to 30 min of ischemia, but 60 min of ischemia and 24 h of reperfusion resulted in decreased CuZnSODactivity. These studies suggest that the antioxidant enzymes of various subcellular compartments respond to ischemia/reperfusionin an organelle or compartment specific manner and that the regulation of antioxidant enzyme activity in peroxisomes maydiffer from that in mitochondria and microsomes. The compartmentalized changes in hepatic antioxidant enzyme activity maybe crucial determinant of cell survival and function during ischemia/reperfusion. Finally, a progressive decline in the level ofhepatic reduced glutathione (GSH) and concomitant increase in serum glutamate pyruvate transaminase (SGPT) activity alsosuggest that greater tissue damage and impairment of intracellular antioxidant activity occur with longer ischemia periods, andduring reperfusion. (Mol Cell Biochem 176: 337–347, 1997)

Key words: antioxidant enzymes, sub-cellular organelles, liver, ischemia-reperfusion

Abbreviations: I/R – ischemia/reperfusion; Cat – catalase; GPx – glutathione peroxidase; GSH – glutathione; GR – glutathionereductase; MnSOD – manganese superoxide dismutase; CuZnSOD – copper zinc superoxide dismutase; SGPT – serum glutamatepyruvate transaminase

Present address: M.P. Gupta, Department of Pulmonary Critical Care and Medicine, Indiana University Medical Center, 1001 West 10th Street, OPW 425,Indianapolis, Indiana 46202-2879, USAAddress for offprints: I. Singh, Department of Pediatrics, Medical University of South Carolina, 171 Ashley Avenue, Charleston, SC 29425, USA

Introduction

Organ, cell and subcellular organelle injury due to ischemia/reperfusion are common causes of morbidity, and possibly

mortality. Oxidative stress and the resulting generation ofoxidants have been implicated as putative mediators of injuryin multiple diseases including hepatic ischemia [1–3] andallograft dysfunction [4], bowel ischemia [5, 6], acute renal

338

failure [7, 8], myocardial ischemia [9], hyperoxia inducedlung injury [10, 11], adverse reactions to xenobiotics [12],as well as accelerated aging [14]. The total organ responseand in part, the whole cell response to ischemia/reperfusionhas been extensively investigated in hepatic tissues, andseveral recent studies suggest the availability of antioxidantenzymes that counteract and/or alleviate the oxidant burdenobserved under these conditions [1–4, 20, 37, 38, 40, 43].However, limited information is available describing the invivo subcellular organelle antioxidant enzyme response ofliver to ischemia/reperfusion mediated oxidative stress [24,25]. Since subcellular compartments possess the machineryto generate local reactive oxygen species (ROS) [17, 35], theyare likely to be particularly vulnerable during periods ofoxidative stress. The oxidative modification of critical cellcomponents including membrane lipids, proteins and nucleicacids impairs important cellular functions [13–15]. Withoutprovisions for protection from oxidative injury, dysregu-lation, permanent cell injury and/or cell death can occur [15,18]. A well developed antioxidant defense system composedof enzymatic components (superoxide dismutases, catalase,glutathione peroxidase and glutathione reductase) andnonenzymatic components (e.g. glutathione, ascorbic acid,β carotene, α-tocopherol and urate among others) has beendemonstrated in most tissues [16, 19–23]. Antioxidantenzymes activity and mRNA transcripts have also beendemonstrated in subcellular compartments. The findingssuggest that subcellular organelles possess an intrinsic andpossibly specific antioxidant system capable of counter-balancing oxidative stress under normal or pathophysiologicconditions [24, 25]. The protection provided by enzymaticand nonenzymatic antioxidants provides reasonable evidencethat oxidants play a major role in ischemia/reperfusioninduced injury.

In the current set of studies we sought to investigate theeffect(s) of short-term (or reversible) hepatic ischemia (30min) or long-term (or irreversible) hepatic ischemia (60 min)followed by varied periods of subsequent reperfusion (for 2or 24 h) on subcellular antioxidant enzyme activity. The 2 htime point and the 24 h time point of organ reperfusion havebeen selected to study the early and late changes associatedwith hepatic ischemia/reperfusion. These studies providefurther evidence for the alterations in antioxidant enzymeactivity and suggest mechanisms responsible for the removalof ROS generated in local subcellular organelle sites duringischemia/reperfusion.

Materials and methods

Xanthine, xanthine oxidase, nitroblue tetrazolium, batho-cuproine disulfonic acid, diethylenetriamine pentaacetic acid,

reduced glutathione (GSH), oxidized glutathione (GSSG),5,5′-dithio-bis(2-nitrobenzoic acid), glutathione reductase,NADPH, cumene hydroperoxide, sucrose, pepstatin A,leupeptin, phenylmethylsulfonyl fluoride (PMSF) and anassay kit to quantitate glutamate pyruvate transaminaseactivity were purchased from Sigma Chemical Company (St.Louis, MO, USA). Aprotinin and antipain were purchasedfrom Calbiochem (CalBiochem, La Jolla, CA, USA). Nyco-denz was obtained from Accurate Chemical and ScientificCorporation (Westbury, NY, USA). Sprague-Dawley ratswere procured from Charles River Laboratories (Wilmington,MA, USA).

Preparation of ischemic animals

Adult male Sprague-Dawley rats weighing 250–300 gramswere subjected to laparotomy. Hepatic ischemia was accomp-lished by placing a clamp around the hepatic vessel pedicle.Briefly, the animals were anesthetized with xylazine (10 mg/Kg bodyweight, intraperitoneal injection) and ketamine (90mg/Kg bodyweight, intramuscular injection). The leftfemoral vein was exposed to administer heparin (0.25 ml ofsodium heparin, 1000 units/ml). A midline abdominalincision was made to expose the liver, portal vein, and hepaticartery. A rubber tourniquet was placed circumferentiallyaround the entire hepatic pedicle including the portal vein,hepatic artery, and bile duct. The tourniquet was clamped toinduce ischemia for 30 or 60 min. The animals were dividedinto six groups and stratified according to the period ofischemia or ischemia/reperfusion. The ischemia or ischemia/reperfusion time periods were designated as follows: control(no ischemia or reperfusion), I30 (30 min of ischemia), I60(60 min of ischemia), I30/R2 (30 min of ischemia followedby 2 h of reperfusion), I30/R24 (30 min of ischemia followedby 24 h of reperfusion), I60/R2 (60 min of ischemia followedby 2 h of reperfusion), and I60/R24 (60 min of ischemiafollowed by 24 h of reperfusion). Animals subjected toischemia only (I30 and I60) were immediately sacrificedfollowing the ischemic period. The ischemic liver wasexcised and prepared for subsequent isolation of the sub-cellular organelle fractions. In contrast, to evaluate theeffect(s) of reperfusion, the vascular clamp was removedfollowing the ischemic period to re-establish blood flow tothe ischemic liver. The surgical incision was closed and theanimals were revived. They were subsequently sacrificedafter 2 or 24 h of reperfusion. As above, the liver was excisedand prepared for subcellular fractionation. All animalsreceived humane care in compliance with the MedicalUniversity of South Carolina’s guidelines and the NationalResearch Council’s criteria for humane care as outlined in‘Guide for the Care and Use of Laboratory Animals’.

339

Subcellular fractionation and organelle isolation

The livers obtained from each group of rats was weighed andcut into fine pieces. Livers were freed of excessive blood bywashing with homogenizing buffer that included 0.25 Msucrose, 3 mM imidazole, 1 mM EDTA, 0.2 mM phenyl-methylsulfonylfluoride, 0.7 mg/L pepstatin A, 1 mg/Laprotinin, 1 mg/L leupeptin and 1 mg/L antipain and 0.1%ethanol at pH 7.4. The tissue was homogenized in a glassTeflon homogenizer. Aliquots were retained for analysis ofreduced glutathione content and antioxidant enzymes inwhole unfractionated homogenates. Remaining homogenateswere used for subcellular fractionation. The subcellularfractions were obtained by subjecting the homogenates todifferential centrifugation. The procedure yields heavymitochondria, light mitochondria, microsomal and cytosolicfractions and is performed according to methods describedby de Duve et al. [26]. Subcellular organelles were subse-quently purified from the light mitochondrial fraction bycontinuous isopycnic nycodenz gradient centrifugation asdescribed previously [27]. Fractions were collected from thebottom of the gradient and analyzed for the distribution ofmicrosomes, mitochondria and peroxisomes by assaying theactivity of marker enzymes. Catalase was assayed to detectperoxisomes, NADPH cytochrome C reductase for micro-somes, and similarly, cytochrome C oxidase was used as themarker enzyme to detect mitochondria. Cytosol and micro-somal fractions were prepared from post mitochondrialsupernatants by centrifugation at 110,000 g for 1 h.

Enzyme assays

Total superoxide dismutase (SOD) and manganese super-oxide dismutase (MnSOD) activity were assayed by themethods of Spitz and Oberley [28]. The assay is based on theinhibition of nitroblue tetrazolium (NBT) dye reduction bySOD. Bathocuperoine sulfonate (0.05 mM) was used in theassay mixture to overcome tissue specific interference. Oneunit of enzyme activity is defined as the amount of enzymeprotein required for inhibiting the NBT reduction by 50% atpH 7.8 and 25°C. Glutathione peroxidase (GPx) activity wasmeasured by the glutathione reductase coupled oxidation ofNADPH using cumene hydroperoxide as substrate accordingto the method of Wendel [29]. One enzyme unit of GPxenzyme activity is defined as µmol of NADPH oxidized/hour/mg protein. Activity of glutathione reductase was estimatedusing the method of Dieter-Horn [30]. The decrease ofNADPH at 340 nm is measured as an index of conversion ofoxidized glutathione to the reduced form of glutathione. Oneunit of enzyme activity is defined as µmol of NADPHconsumed/h/mg protein. Catalase activity was measured bythe method of Baudhin et al. [31] following the rate of H

2O

2

consumption. Briefly, the reaction mixture contained 0.25 µLof 30% H

2O

2 in 1 ml of 10 mM imidazole-HCl buffer, at pH

7.2, and 50 µL of 2% triton X-100. The reaction was initiatedby the addition of 50 µL of sample enzyme protein andterminated by the addition of 1 ml of TiOSO

3. Catalase

activity was measured spectrophotometrically as a yellowcolor at 405 nm. Serum activity of glutamate pyruvatetransaminase (SGPT) was assayed using a commercial kitfrom Sigma Chemical Co, St. Louis, USA. Employing themethod of Moron et al. [32], the total thiol content wasestimated spectrophotometrically by measuring its coloredcomplex with 5,5′ dithiobis nitrobenzene (DTNB) at 412 nm.The protein content of all samples were measured by theBiorad method using Bradford reagent [33].

Statistical analysis

Each experimental group was comprised of 3–5 animals witheach sample analyzed in duplicate. Analysis of variance followedby post hoc analysis using Bonferroni’s test was performed formultigroup comparisons and the level of statistical significancewas taken at p < 0.05. Activity levels for the control values wereregarded as representing 100% of activity. Activity levels ofthe enzymes in the various experimental groups were presentedas a percentage of control.

Results

The activity of antioxidant enzymes in the various subcellularfractions from rat liver is shown in Table 1. The highestspecific catalase activity was detected in peroxisomes, whilemitochondria, microsomes and cytosol were without detect-able catalase activity. Glutathione peroxidase activity wasgreatest in the cytosolic fraction, followed in decreasing orderby mitochondria, microsomes and peroxisomes. Similarly,glutathione reductase activity levels were also highest in thecytosol. Mitochondria, peroxisomes and microsomes hadlower levels of activity. Cu/ZnSOD activity was greatest inthe cytosolic fraction although quantitatively smaller levelsof activity were also detected in peroxisomal and microsomal/lysosomal fractions. The greatest MnSOD activity waspresent in mitochondria however detectable activity was alsoobserved in the peroxisomal fraction.

The subcellular distribution of catalase in isopycnicnycodenz gradient fractions demonstrated the peak catalasespecific activity in fractions 2–5 corresponding to an averagedensity of 1.225 g/ml in all the groups, indicating a highconcentration of peroxisomes in these fractions (Fig. 1).Mitochondrial peak activity determined by high cytochromeoxidase specific activity was detected between fractions13–15. The specific activity of cytochrome reductase was

340

maximum in fractions 16–20 demonstrating the presence ofmicrosomes. These findings demonstrate that subcellularorganelles are resolvable by the gradient centrifugationtechniques employed in the current studies. In the catalasegradient profile we observed that samples obtained from liverssubjected to 60 min of ischemia demonstrated an additionalpeak of catalase specific activity in fractions 10–13 corre-sponding to a density of 1.12 mg/ml. Although the secondcatalase peak was less prominent in samples obtained fromlivers subjected to only 30 min of ischemia, in contrast thesecond catalase peak had a greater presence in samples fromlivers subjected to 30 min of ischemia followed by reperfusionfor 2 or 24 h (Fig. 1). These observations suggest that hepaticperoxisomes undergo metabolic and functional alterationsduring prolonged ischemia and/or ischemia/reperfusion.

Figure 2 demonstrates the effect of ischemia or ischemia/reperfusion on antioxidant enzyme activity in whole liverhomogenates. Catalase specific activity (Cat) decreasedprogressively during increased ischemia duration and declinedfurther during reperfusion (50 ± 8% after 60 min of ischemiaand 24 h of reperfusion, p < 0.001). In contrast, glutathioneperoxidase (GPx) activity was not significantly altered byischemia or ischemia/reperfusion. The activity of Cu/ZnSODdecreased 17% (p < 0.05) below control values following 30min of ischemic time. However, after 30 min of ischemia andreperfusion for 2 or 24 h, activity levels were increased 35%(p < 0.05) and 12% above control, respectively. Afterprolonged ischemia (60 min) and reperfusion for 2 or 24 hCu/ZnSOD activity was reduced to levels 20–25% below thatobserved for control values. Homogenate MnSOD andglutathione reductase activities were unaltered by ischemiaalone or ischemia/reperfusion.

Specific antioxidant enzyme activity was evaluated in thecytosol and several subcellular liver organelles (mito-chondria, peroxisomes, and microsomes/lysosomes). In thecytosolic fraction, results for Cu/ZnSOD activity were similarfor those obtained from whole liver homogenates (Fig 3). Cu/ZnSOD activity in cytosol was not significantly affected byischemia for 30 or 60 min. However, following reperfusionfor 24 h the activity increased by 26% (p < 0.05) in animalssubjected to 30 min of ischemia, whereas during 60 min of

ischemia followed by 24 h of reperfusion, activity decreasedby 17% (p < 0.05). Cytosolic glutathione peroxidase (GPx)activity was not significantly changed due to ischemia and/or ischemia/reperfusion. Glutathione reductase activitydecreased during 60 min of ischemia and 2 h of reperfusion,however the decline in activity was not maintained following24 h of reperfusion.

The effect of ischemia alone, and ischemia/reperfusion onmitochondrial antioxidant enzyme activity is demonstratedin Fig. 4. MnSOD activity increased by 20% (p < 0.05)following 30 min of ischemia, however prolonging ischemiatime to 60 min resulted in a 15% decrease in mitochondriaMnSOD activity. Thirty min of ischemia followed by reper-fusion (for 2 or 24 h) returned MnSOD activity to controllevels. If animals were subjected to 60 min of ischemiaenzyme activity was 13–28% lower than control regardlessof the length of the reperfusion period. Mitochondrialglutathione peroxidase activity was not altered by 30 min ofischemia alone, and similarly for 30 min of ischemia subse-quently followed by 2 or 24 h of reperfusion. Sixty min ofischemia increased mitochondrial glutathione peroxidaseactivity by 35% (p< 0.001). Enzyme activity also remainedgreater than control level when the 60 min of ischemia wasfollowed by reperfusion for 2 or 24 h. Thirty and 60 min ofischemia without subsequent reperfusion resulted in decreasedmitochondrial glutathione reductase activity by 24% and22% respectively (p < 0.05). Reperfusion, whether for 2 or24 h did not alter the activities.

In the microsomal/lysosomal fraction demonstrated in Fig.5, total SOD enzyme activity increased by 22% (statisticallyinsignificant) in livers subjected to 30 min of ischemia.When ischemia alone was continued for 60 min there wasan 18% decrease in total SOD enzyme activity (p < 0.05).Thirty min of ischemia followed by reperfusion for 2 or 24h also decreased total SOD activity by 15–18% (p < 0.05),and by 37% (p < 0.001) after 60 min of ischemia and 24 hof reperfusion. Glutathione peroxidase (GPx) activity wasnot significantly altered by 30 min of ischemia with orwithout reperfusion. However, during prolonged ischemiafor 60 min glutathione peroxidase activity was increased by82% (p < 0.05). Reperfusion for 24 h caused a 28–58%

Table 1. Subcellular distribution and activity of antioxidant enzymes in normal rat liver

Fraction Catalase* Glutathione# Glutathione# CuZnSOD* MnSOD*peroxidase reductase

Homogenate 0.61 ± 0.08 9.66 ± 0.60 2.04 ± 0.060 224 ± 22 58 ± 4.33Cytosol ND 22.50 ± 5.32 11.753 ± 1.36 1231 ± 113 NDMitochondria ND 5.637 ± 0.99 5.731 ± 1.131 ND 189 ± 20.9Microsomes ND 5.381 ± 1.68 0.411 ± 0.60 12.78 ± 1.20 NDPeroxisomes 8.83 ± 0.98 0.351 ± 0.056 1.158 ± 0.311 4.24 ± 0.77 6.56 ± 0.62

Enzyme activities for catalase, CuZnSOD, and MnSOD (indicated by *) are measured as units/mg protein (U/mg) protein. Enzyme activities for glutathioneperoxidase and glutathione reductase (indicated by #) are measured as µmoles of NADPH consumed/hour/mg protein. Values are mean ± SEM from fourdifferent animals. Each sample was analyzed in duplicate.

341

Fig. 1. Distribution of subcellular marker enzymes in various gradient fractions obtained from rat liver following ischemia for 30 or 60 min and 30 min ofischemia followed by reperfusion for 2 or 24 h. The data illustrated is representative of at least three different sets of experiments.

decline (p < 0.05) in GPx activity relative to the I60 group,but these values remain significantly greater than control.Glutathione reductase enzyme activity (GR) decreasedprogressively as the duration of ischemia increased (12.5%after 30 min and 27% (p < 0.05) following 60 min ofischemia). Short or prolonged reperfusion time (2 or 24 hrespectively) resulted in GR enzyme activities similar to

controls when the ischemic period was 30 min and actuallyincreased 14–16% above control levels when ischemiacontinued for 60 min.

Enzyme activity in peroxisomes following ischemia/reperfusion was different from that observed in mitochondriaor microsomes. As illustrated in Fig. 6, the activity ofperoxisomal CuZnSOD activity did not change significantly

342

in response to ischemia or ischemia/reperfusion. PeroxisomalMnSOD activity decreased 18% (p < 0.05) after 60 minof ischemia alone. In contrast, the activity increased after30 or 60 min of ischemia/reperfusion when the reperfusionperiod was 2 h (22–44%) (p < 0.05) or 24 h (27–30%)(p < 0.05). Peroxisomal glutathione peroxidase activitydecreased 15% after prolonged ischemia (60 min), howeverduring 24 h of subsequent reperfusion the enzyme activityincreased 28% (p < 0.05). Glutathione reductase activityin peroxisomes increased 33% and 27% respectively (p< 0.05) when the livers were subjected to 30 and 60 minof ischemia. Enzyme activity levels remained elevatedafter short or prolonged ischemia when ischemia wasfollowed by 2 or 24 h of reperfusion. In contrast, peroxi-somal catalase activity decreased after short or prolongedischemia (22% (p < 0.05) at 30 min and 33% (p < 0.001)

Fig. 2. Percent activity of Cat, GPx, GR, CuZnSOD and MnSOD in rat liver homogenate following ischemia for 30 or 60 min with and without reperfusionfor 2 or 24 h. Results are indicated as the mean value ± SEM. The control value for each enzyme is presented as 100%. Activity of the enzymes for eachexperimental group is presented as a percentage of the mean control value. An asterisk (*) indicates significant difference from the control value (p < 0.05).Each experimental group was composed of 3–5 animals with each sample analyzed in duplicate.

Fig. 3. Percent activity of GR, GPx and CuZnSOD in rat liver cytosol fractions following ischemia for 30 or 60 min with and without reperfusion for 2 or24 h. Results are indicated as the mean value ± SEM. The control value for each enzyme is presented as 100%. Activity of the enzymes for each experimentalgroup is presented as a percentage of the mean control value. An asterisk (*) indicates significant difference from the control value (p < 0.05). Eachexperimental group was composed of 3–5 animals with each sample analyzed in duplicate.

after 60 min) and remained significantly lower than controllevels during reperfusion whether for 2 or 24 h.

The level of reduced glutathione (GSH) in liver homoge-nates decreased by 20% (p < 0.05) following 60 min ofischemia, and further declined commensurate with the lengthof increasing reperfusion time (34% at 2 h (p < 0.05) and 53%after 24 h (p < 0.05) as demonstrated in Fig. 7. On the otherhand the levels of GSH also decreased in a group with 30 minof ischemia followed by 2 h of reperfusion, however it wasnormalized by 24 h of reperfusion. Serum levels of glutamatepyruvate transaminase activity (SGPT) were measured as ageneral marker of global liver injury during ischemialreperfusion (Fig. 8). Enzyme activity levels increased onlymoderately during ischemia, but markedly during 2 h ofreperfusion suggesting that hepatocellular injury occursduring ischemia, but is greatly augmented by reperfusion.

343

Discussion

Because of variability in the response of tissues to oxidativestress when exogenous antioxidants are used, employing thisapproach as the major therapeutic option to attenuate injury intissues subjected to ischemia/reperfusion has met with limitedsuccess. Moreover, optimal concentrations of exogenousantioxidants and their targeted delivery required to replacedepleted subcellular antioxidant reserves remains ill defined.

Fig. 4. Percent activity of GR, GPx, and MnSOD in rat liver mitochondrial subcellular fractions following ischemia for 30 or 60 min with and withoutreperfusion for 2 or 24 h. Results are indicated as the mean value ± S.E.M. The control value for each enzyme is presented as 100%. Activity of the enzymesfor each experimental group is presented as a percentage of the mean control value. An asterisk (*) indicates significant difference from the control value (p< 0.05). Each experimental group was composed of 3–5 animals with each sample analyzed in duplicate.

Fig. 5. Percent activity of GR, GPx, and SOD in the rat liver mitochondria/lysosomal fraction following ischemia for 30 or 60 min with and withoutreperfusion for 2 or 24 h. Results are indicated as the mean value ± SEM. The control value for each enzyme is presented as 100%. Activity of the enzymesfor each experimental group is presented as a percentage of the mean control value. An asterisk (*) indicates significant difference from the control value (p< 0.05). Each experimental group was composed of 3–5 animals with each sample analyzed in duplicate.

Antioxidants in the cell perform several important func-tions [36]. These functions include limiting lipid peroxi-dation, decreasing sulfhydryl dependent enzyme damage (i.e.preventing the decline in covalent binding) in critical cellproteins, limiting destruction of nucleotide coenzymes andDNA damage, and attenuating the alteration of secondary ortertiary structures of proteins that eventuate in enhancedsusceptibility to degradation by cellular proteases. Theimplications from studies finding relatively minor protection

344

against injury during ischemia/reperfusion suggest that thedifficulty may largely lie in the inability of exogenouslysupplied antioxidants and antioxidant enzymes to reachcritical cellular targets in subcellular organelles. Endoge-nously antioxidants and antioxidant enzymes generated insufficient numbers are possibly better positioned to protectcellular structural and functional integrity during ischemia/

Fig. 6. Percent activity of Cat, GR, GPx, CuZnSOD and MnSOD in the rat liver peroxisomal fraction following ischemia for 30 or 60 min with and withoutreperfusion for 2 or 24 h. Results are indicated as the mean value ± SEM. The control value for each enzyme is presented as 100%. Activity of the enzymesfor each experimental group is presented as a percentage of the mean control value. An asterisk (*) indicates significantly different from the control value (p< 0.05). Each experimental group was composed of 3–5 animals with each sample analyzed in duplicate.

Fig. 7. Effect of short or prolonged ischemia alone and ischemia/reperfusion on the level of reduced glutathione (GSH) in whole liverhomogenates. Results are indicated as the mean value ± SEM. The controlvalue for each enzyme is presented as 100%. Activity of the enzymes foreach experimental group is presented as a percentage of the mean controlvalue. An asterisk (*) indicates significantly different from the control value(p < 0.05). Each experimental group was composed of 3–5 animals witheach sample analyzed in duplicate.

reperfusion. Additionally, the generation of endogenousantioxidant enzymes may be specifically tailored for particularsubcellular organelles [24, 25, 36]. For example, it hasrecently been reported that several endogenous antioxidantenzyme proteins are synthesized and targeted specially fortranslocation to peroxisomes by SKL (amino terminal serine-lysine-leucine residues) topogenic signals [17]. Similartranslocation signals may also exists for other antioxidantenzyme proteins targeted to other organelles. Since endoge-nous antioxidant enzymes can exist in several compart-mentalized organelle specific sites, they provide a built-in setof cellular defense mechanisms that limit the adverse effect(s)of oxidants.

To be most effective antioxidants must scavenge oxidantspromptly and efficiently at local sites of generation. Ourcurrent studies assessed the temporal response and quanti-tated the activity levels of antioxidant enzymes in hepaticsubcellular organelles during hepatic ischemia and ischemia/reperfusion. Hepatic injury measured as elevated SGPTactivity, increases with increasing ischemic time and wasmade worse by subsequent reperfusion periods for as shortas 2 h. The findings are consistent with biochemical andpathologic findings reported in several recent studies thatsimilarly demonstrate hepatic injury during ischemia/reper-fusion [1–4]. The moderate decline in liver homogenate GSHlevels during ischemia, and further decline during reperfusionare consistent with similar observations reported by Stein etal. [38]. The findings suggest that the injury results fromgreater oxidative stress under these conditions. The declinein endogenous GSH might be attributable to several possi-bilities: (1) continuous consumption of GSH when used as acofactor by antioxidant enzymes like glutathione peroxidase,

345

(2) direct interactions with oxidants as a scavenger [32, 40],and (3) decreased regeneration of GSH from oxidizedglutathione.

In the whole homogenate fraction, the total endogenousantioxidant enzyme capacity appeared to be decreased by thelonger ischemia period and prolonged reperfusion. Theseresults are consistent with significant hepatic injury and aconcomitant decrease in liver glutathione content. Althoughtotal homogenate glutathione reductase activity recoveredafter 24 h of reperfusion to levels not different from control,it was not enough to significantly limit cell injury. Peroxi-somal glutathione reductase activity was significantlyincreased by ischemia for 30 min (33%) and 60 min (27%)and remained significantly higher than control followingreperfusion for 2 or 24 h of reperfusion in animals subjectedto 30 min of ischemia. It is important to note that theseelevations in glutathione reductase activity are not maintainedafter 24 h of reperfusion in animals subjected to prolongedischemia (60 min). Glutathione reductase activity in thecytosol and mitochondria was lowered by ischemia andreperfusion. Our studies offer the first report of measurableglutathione reductase activity in hepatic peroxisomes.Although increased, endogenous glutathione reductase alonedid not protect the liver from injury during ischemia/reper-fusion. A possible explanation is that control levels ormoderate increases of glutathione reductase activity aresuboptimal to regenerate and maintain total intracellular GSHduring severe or prolonged oxidative stress. Additionally,GSSG (oxidized glutathione) levels accumulate and mayactually diffuse from the cell further depleting the cellularpool of glutathione [32]. GSH depleted rats subjected tohepatic ischemia/reperfusion are reported to have evidence

of increased lipid peroxidation and greater hepatic injurywhen intracellular glutathione levels are suboptimal for thedegree of oxidative stress. Also antioxidants are likely to bemost effective when working in concert.

Peroxisomal catalase activity also declined progressivelyas ischemic time periods were prolonged and declined furtherfollowing prolonged reperfusion. These findings are inagreement with observations made by Marubayashi et al.[41]. Barnard et al. [20] employing an isolated perfused ratliver preparation subjected to ischemia for 2 h followed by 1h of reperfusion failed to detect any significant change inhepatic catalase activity, but did report a decline in seleniumand nonselenium containing glutathione peroxidase activity[20]. The authors did demonstrate an increase in glutathionereductase activity after two h of ischemia that was sustainedduring reperfusion, findings consistent with the increasedglutathione reductase activity we also observed. Our measure-ment of catalase activity are in contrast to the findingsreported by Barnard et al. [20]. The current data suggest adecline in total catalase activity in whole tissue homogenateand peroxisomes. Although the activity of glutathionereductase was increased in the current study during reper-fusion, its ability to protect peroxisomes against oxidativedamage may be possibly limited by the relative overalldepletion of an important cofactor, intracellular reducedglutathione (GSH). The studies suggest that ischemia/reperfusion induced alterations in antioxidant enzymeactivities occur in subcellular organelles of the liver andmoreover, that the specific subcellular organelle response tooxidative stress may be varied. The observed inconsistenciesare possibly due to differences in the study models employedand the ischemia/reperfusion times.

Fig. 8. Effect of short or prolonged ischemia alone and ischemia/reperfusion on the level of rat serum glutamate pyruvate transaminase activity (SGPT).Results are indicated as the mean value ± SEM. An asterisk (*) indicates significantly different from the control value (p < 0.05). Each experimental groupwas composed of 3–5 animals with each sample analyzed in duplicate.

346

The current studies also demonstrate the presence of bothCu/ZnSOD and GPx activity in peroxisomes, complimentingearlier investigations from this laboratory confirming thepresence of Cu/ZnSOD activity in peroxisomes [24, 25]. Thestimulation of MnSOD activity concomitant with a declinein peroxisomal catalase activity during ischemia/reperfusionincreased the susceptibility of peroxisomes to oxidativestress by possibly increasing the accumulation of H

2O

2

(hydrogen peroxide) and OH· (hydroxyl radical) in thisorganelle. A slight increase in GPx activity and glutathionereductase activity under these conditions may not be sufficientto provide enough protection as total GSH is significantlydepleted. Although peroxisomes appear to be equipped witha well developed local antioxidant enzyme defense systemthey remain sensitive to the damage caused by oxidativestress [35, 37, 42]. This is suggested by the presence of asecond peroxisomal peak corresponding to a lower densityof 1.12 g/ml (fractions 10–13) in organs subjected toprolonged ischemia/reperfusion. Similar observationssuggesting the presence of a second peroxisomal peak werereported in a previous study from this laboratory [37]. Thenew peroxisomal peaks and newly formed hepatic peroxi-some populations will require further characterization.Besides showing physical alterations, peroxisomes seem toundergo changes in several other metabolic functions. Ourearlier studies indicated that renal ischemia/reperfusioncaused impairment in lignoceric acid oxidation, which inturn may affect several other metabolic functions [37, 42].

Total SOD activity was also detected in the microsomal/lysosomal fraction. The enzyme activity detected in themicrosomal fraction may be partly attributable to minorcontamination of the microsomal fraction with lysosomes andperoxisomes. Further studies are necessary to demonstrate thedirect presence and specific activity of MnSOD in microsomes.

In addition to directly causing injury, oxidants can stimu-late cells to secrete substances that increase adhesion mole-cules and recruit inflammatory cells (e.g. macrophages,lymphocytes, neutrophils) to sites of injury, thereby indirectlyexacerbating cellular damage [39, 43]. Newly recruitedinflammatory cells can then generate cytokines like IL-1β andTNF further initiating a cascade of events that can attenuatethe ability of the affected organ to recover from ischemia/reperfusion [44]. Although we did not directly evaluate the roleof inflammatory cells in our system it is possible that thesemechanisms might explain some of the loss of subcellularorganelle function during hepatic ischemia/reperfusion.

In conclusion, the differences in activity of the variousantioxidant enzymes found in hepatic subcellular organellessubjected to oxidative stress suggest that these organelles mayhave different susceptibilities to ischemia/reperfusion inducedinjury. The current study offers new insights into the charac-terization of hepatic subcellular antioxidant enzyme activitiesduring ischemia/reperfusion. These studies support previous

work originating from this laboratory and others suggestingthat cytosolic, mitochondrial, and peroxisomal antioxidantenzymes may be regulated by different mechanisms. Inaddition, antioxidant enzymes in various subcellular fractionscan respond to oxidative stress independently according tolocal organelle conditions. This may ultimately aid in thedevelopment of specific endogenous antioxidant therapies inclinical situations.

Acknowledgements

This work was supported by grants from Dialysis ClinicsIncorporated and NS-22576 from the National Institutes ofHealth.

References

1. Younes M, Strubelt O: The involvement of reactive oxygen species inhypoxic injury to rat liver. Res Commun Chem Pathol Pharmacol 59:369–381, 1988

2. Liu PT, Symons AM, Howarth JA, Boulter PS, Darke DV: Studies insurgical trauma: oxidative stress in ischemia-reperfusion of rat liver.Clin Sci 86(4): 453–60, 1994

3. Jaeschke H: Reactive oxygen species and ischemia/reperfusion injuryof the liver. Chem Biol Interact 79: 115–136, 1991

4. Goode HF, Webster NR, Howdle PD, Leek JP, Lodge JP, Sadek SA,Walker BE: Reperfusion injury, antioxidants and hemodynamic duringorthotopic liver transplantation. Hepatology 19(2): 354–359, 1994

5. Schoenberg MH, Berger HG: Oxygen radicals in intestinal ischemiaand reperfusion. Chem Biol Interact 76: 141–161, 1990

6. Granger DN, Rutili G, McCord JM: Superoxide radicals in felineintestinal ischemia. Gastroenterology 81: 22–29, 1981

7. Canavese C, Stra P, Vercellone A: The case of oxygen free radicals inpathogenesis of ischemic acute renal failure. Nephron 49(1), 9–15,1988

8. Paller MS, Hoidal JR, Ferris TF: Oxygen free radicals in ischemicacute renal failure in rat. J Clin Invest 74: 1156–1164, 1984

9. Ferrari R, Ceconi C, Currello S, Guarnieri C, Caldarera CM, VisioliO: Oxygen mediated myocardial damage during ischemia andreperfusion: Role of cellular damage against oxygen toxicity. J MolCell Cardiol 17: 937–945, 1985

10. Koyama I, Toung TJ, Rogers MC, Gurtner GH, Traystman RJ: O2

radicals mediate reperfusion lung injury in ischemic O2 ventilated

canine pulmonary lobe. J Appl Physiol 63(1): 111–115, 198711. Kennedy TP, Rao NV, Hopkins C, Pennington L, Tolley E, Hoidal

JR: Role of reactive oxygen species in reperfusion injury of the rabbitlung. J Clin Invest 83(4): 1326–1335, 1989

12. Comporti M: Lipid peroxidation and cellular damage in toxic liverinjury. Lab Invest 53(6): 599–623, 1985

13. Wiseman H, Halllwell B: Damage to DNA by reactive oxygen andnitrogen species: role in inflammatory disease and progression tocancer. Biochem J 313: 17–29, 1996

14. Davies KJA, Desnigore ME: Protein damage and degradation byoxygen radicals: modifications of secondary and tertiary structure. JBiol Chem 262: 9908–9913, 1987

15. Sies H: Biochemistry of oxidative stress. Angewandte Chemie Int 25:1058–1071, 1986

347

16. Chen H, Pellett LJ, Anderson HJ, Tappel AL: Protection by vitamin E,selenium and β carotene against oxidative damage in rat liver slicesand homogenates. Free Radic Biol Med 14(5): 509–517, 1993

17. Singh I: Mammalian Peroxisomes: antioxidant enzymes and oxidativestress. In: KJA Davies, F Ursini (eds). The Oxygen Paradox. CLUEPUniversity Press, Padova, Italy, 1995, pp 209–223

18. Mathews WR, Guido DM, Fischer MA, Jaeschke H: Lipid peroxidationas molecular mechanism of liver cell damage during reperfusion injuryafter ischemia. Free Radic Biol Med 16(6): 763–770, 1994

19. Reilly PM, Schiller HJ, Bulkley GB: Pharmacological approach totissue injury mediated by free radicals and other reactive oxygenmetabolises. Am J Surgery 161: 488–503, 1991

20. Barnard ML, Snyder SJ, Engerson TD, Turrens JF: Antioxidantenzymes of ischemic and postischemic liver and ischemic kidneys inrats. Free Radic Biol Med 15: 227–232, 1993

21. Meister A, Anderson ME: Glutathione. Annul Rev Biochem 52: 711–760, 1983

22. White AC, Thannickal VJ, Fanburg BL: Glutathione deficiency inhuman disease. J Nutr Biochem 5: 218–226, 1994

23. Witting LA: Vitamin E and lipid peroxidation in free radical initiatedreactions. In: WA Pryor (ed). Free Radicals in Biology, Vol. IV, NYAcademic Press, 1989, pp 295–319

24. Dhaunsi G, Gulati S, Singh AK, Orak JK, Asayama K, Singh I:Demonstration of Cu-Zn superoxide dismutase in rat liver peroxisomes:Biochemical and immunochemical evidence. J Biological Chem 267:6870–6873, 1992

25. Singh AK, Dhaunsi G, Gupta MP, Orak JK, Asayama K, Singh I:Demonstration of glutathione peroxidase in rat liver peroxisomes andits intraorganellar distribution. Arch Biochem Biophys 315(2): 331–338, 1994

26. de Duve C, Pressman BC, Gianetto R, Wattinox R, Applemans F:Tissue fractionation Studies: distribution pattern of enzymes in ratliver tissue. Biochem J 60: 604–617, 1955

27. Lazo O, Contreras M, Singh I: Effect of ciprofibrate on the activationand oxidation of very long chain fatty acids. Mol Cell Biochem 100:159–167, 1991

28. Spitz DR, Oberley LW: An assay for superoxide dismutase inmammalian tissue homogenates. Analytical Biochemistry 179: 8–18,1989

29. Wendel A: Glutathione peroxidase. Methods in Enzymol 77: 325–333, 1981

30. Dieter Horn H: Glutathione reductase assay. In: HU Bergmeyer (ed).Methods of Enzymatic Analysis. Academic Press, New York, 1963, p875

31. Baudhuin P, Beaufay Y, Rehman LY, Sellinger OZ, Wautiaux R,

Jacques P, de Duve C: Tissue fraction studies: Intracellular distributionof monoamine oxidase, aspartate aminotransferase, alanine amino-transferase, d-aminoacid oxidase and catalase in rat liver tissue.Biochem J 92: 179–184, 1964

32. Moron MS, Depierre JW, Mannervik B: Levels of glutathione,glutathione reductase and glutathione S transferase activities in ratlung and liver. Biochem Biophys Acta 582: 67–78, 1979

33. Bradford M: A rapid and sensitive method for the quantitation ofmicrogram quantities of protein utilizing the principle of protein dyebinding. Analytical Biochemistry 72: 248–254, 1979

34. Takayama FT, Egashira T, Kudo Y, Yamamaka Y: Effect of anti freeradical intervention on phosphatidylcholine hydroperoxide in plasmaafter ischemia reperfusion in the liver of rats. Biochem Pharmacol46(10): 1749–1757, 1993

35. Elliot BM, Dodd NJF, Elcombe CR: Increased hydroxyl radicalproduction in liver peroxisomal fractions from rats treated withperoxisomal proliferators. Carcinogenesis 7: 795–799, 1986

36. Yu BP: Cellular Defenses against damage from reactive oxygenspecies. Physiological Reviews. 74(1): 139–162, 1994

37. Gulati S, Singh AK, Irazu C, Orak J, Rajagopalan PR, Fitts CT, SinghI: Ischemia reperfusion injury: Biochemical alterations in peroxisomesof rat kidney. Arch Biochem Biophys 295: 90–100, 1992

38. Stein HJ, Oosthuizen MMJ, Hinder RA, Lamprechts H: Effect ofverapamil on hepatic ischemia and reperfusion injury. Am J Surgery165: 96–100, 1993

39. Jaeschke H, Farhood A, Bautista AP, Spolarics Z, Spitzer JJ:Complement activates Kupfer cells and neutrophils during reperfusionafter hepatic ischemia. Am J Physiol 264(4, Part 1): G801–G809, 1993

40. Lenz MC, Michael LH, Smith CV, Hughes H, Shappell SB, TaylorAA, Entman ML, Mitchell JR: Glutathione disulphide formation andlipid peroxidation during cardiac ischemia and reflow in the dog invivo. Biochem Biophys Res Commun 164: 722–727, 1989

41. Marubayashi S, Dohi K, Sumimoto K, Oku J, Ochi K, Kawasaki T:Changes in activity of oxygen radical scavengers and in levels ofendogenous antioxidants during hepatic ischemia and subsequentreperfusion. Transplant Proc 21: 1317–1318, 1989

42. Gulati S, Ainol L, Orak J, Singh AK, Singh I: Alterations ofperoxisomal function in ischemia - reperfusion injury of rat kidney.Biochim Biophys Acta 1182: 291–298, 1993

43. Granger DN: Role of xanthine oxidase and granulocytes in ischemiaand reperfusion injury. Am J Physiol 255: H1269–H1275, 1988

44. Scales WE, Campbell DA, Green ME, Remick DG: Hepatic ischemia/reperfusion injury: importance of oxidant/tumor necrosis factorinteractions. Am J Physiol 267: (Gastrointest. Liver Physiol 6 part 1):G1122–G1127, 1994