1040

Thromboxane B2 in Cardiac LymphEffect of Superoxide Dismutase and Catalase During

Myocardial Ischemia and Reperfusion

L.H. Michael, Z. Zhang, C.J. Hartley, R. Bolli, A.A. Taylor, and M.L. Entman

Neutrophils have been implicated in the genesis of myocardial ischemia-reperfusion injury, andtheir mechanism of action has been linked to the production of reactive oxygen species, fattyacid-derived prostanoids, and leukotrienes. In this study, we examined the potential relationbetween production of reactive oxygen species and cyclooxygenase-derived autacoids bystudying the effects of superoxide dismutase (SOD) and catalase (CAT) on the rise inthromboxane formation observed with myocardial ischemia and reperfusion. Immunoreactivethromboxane B2 was measured in cardiac lymph from conscious dogs during reperfusion aftera 60-minute occlusion in the presence of infusions of SOD alone, CAT alone, and a combinationof SOD and CAT. Reperfusion after 60 minutes of ischemia causes an immediate elevation inthromboxane B2. SOD and CAT infusion prevented this rise in thromboxane B2 as did infusionof SOD alone. The ability of SOD-CAT to suppress thromboxane B2 production dissipatedwithin 3 hours after the cessation of infusion, at which time thromboxane B2 excretionincreased. The modulation of fatty acid oxygenases by reactive oxygen species may be a veryimportant pathogenic factor in considering the origin of the protective effect of antioxidants in thesetting of myocardial ischemia-reperfusion injury. (Circulation Research 1990;66:1040-1044)

R eactive oxygen species such as superoxideanion have been implicated as importantagents involved in causing cell death in the

setting of myocardial ischemia and reperfusion.12Leukocytes may exert their membrane-damagingeffects by the production of these reactive oxygenspecies as well as by the alteration of microvascularand endothelial function. Studies have demonstratedfavorable effects of free radical scavengers on recov-ery of myocardial function and infarct reduction afterreversible regional ischemia.1,3,4 In addition, periph-eral blood polymorphonuclear leukocytes synthesizelipoxygenase and cyclooxygenase derivatives on expo-sure to chemotactic stimuli, and these autacoids areknown to induce vasoconstriction, platelet aggrega-tion, and direct tissue toxicity.5,6

Thus, the production of reactive oxygen speciesand the production of oxygenase-derived autacoidsrepresent two potential independent mechanisms by

From the Sections of Cardiovascular Sciences (L.H.M., Z.Z.,C.J.H., M.L.E.), Cardiology (R.B.), and Experimental Therapeu-tics (A.A.T.), Department of Medicine, and The DeBakey HeartCenter, Baylor College of Medicine and The Methodist Hospital,Houston, Texas.

Supported by National Institutes of Health grants HL-23161,HL-22512, and HL-42550 and a grant from the American HeartAssociation, Texas Affiliate.Address for correspondence: Lloyd H. Michael, PhD, Depart-

ment of Medicine, Cardiovascular Sciences, Baylor College ofMedicine, One Baylor Plaza, Houston, TX 77030.

Received March 21, 1989; accepted November 8, 1989.

which tissue injury can be induced by leukocyte acti-vation. These functions, however, may not be totallyindependent. The work of Hemler and Lands7 hassuggested that fatty acid oxygenases require a hydro-peroxide activator. Therefore, augmentation of theproduction of molecular oxygen species may poten-tially amplify oxygenase enzyme activity. Work byShappell et a18 demonstrated that both superoxidedismutase (SOD) and catalase (CAT) inhibit arachido-nate-induced platelet function, suggesting the converse(i.e., that scavenging of reactive oxygen species mayinhibit arachidonate-dependent function).Our past studies of conscious animals demonstrated

a striking increase in cardiac lymph immunoreactivethromboxane B2 (TXB2) concentration after ischemiaand reperfusion following a coronary artery occlu-sion.9 In view of the potential relation between molec-ular oxygen species and cyclooxygenase-derived auta-coid production,7 we chose to examine the effects ofSOD and CAT on the rise in TXB2 formationobserved with myocardial ischemia and reperfusion.The results provide in vivo evidence that SOD andCAT in combination prevent the rise in TXB2 appear-ance in cardiac lymph observed during ischemia andreperfusion. SOD alone is also effective, but CATalone is ineffective in preventing the TXB2 generation.

Materials and MethodsThe instrumentation of conscious dogs and cannu-

lating cardiac lymph ducts have been described

by guest on July 11, 2018http://circres.ahajournals.org/

Dow

nloaded from

Michael et al Effect of Free Radical Scavengers on Thromboxane B2 1041

previously.9,10 This study was based on 31 healthymongrel dogs of either sex weighing 16-24 kg. Atleast 2 days after surgery, cardiac lymph sampleswere collected every 30 minutes on ice in polypropyl-ene tubes with 0.1 ,M prostaglandin E1, 10 ,uMindomethacin, 0.5 mM theophylline, and 5 mMEDTA and then centrifuged at 8,000g for 15 minutesat 40 C. The supernatant from each tube was frozenin dry ice-acetone and stored at -70° C. Radio-immunoassay9 was performed with [1251]TXB2 andrabbit anti-TXB2 (Advanced Magnetics, Cambridge,Massachusetts).TXB2 was measured in cardiac lymph during the

following protocols: 1) 5-minute occlusion followedby reperfusion in the presence of 0.9% saline infu-sion, 2) 60-minute occlusion followed by reperfusionin the presence of 0.9% saline infusion, 3) 60-minuteocclusion followed by reperfusion in the presence ofSOD infusion, 4) 60-minute occlusion followed byreperfusion in the presence of CAT infusion, and 5)60-minute occlusion followed by reperfusion in thepresence of SOD and CAT infusion.Animals receiving drug were infused with either

0.06 mg/kg/min CU-ZN-SOD purified from bovineliver (activity, 3,300 units/mg protein; DDI Pharma-ceuticals, Mountain View, California) plus 0.06 mg/kg/min CAT purified from bovine liver (activity,11,000 units/mg protein; Sigma Chemical, St. Louis,Missouri) in saline, SOD alone, CAT alone, or anequal volume of saline alone. The infusion wasadministered through a left atrial catheter at 0.2ml/min starting 15 minutes before coronary arteryocclusion and continuing until 15 minutes after theonset of reperfusion or 75 minutes after the onset ofreperfusion (for study of effects of SOD for a longerterm). The occlusion period was 60 minutes, andthe reperfusion was continuous thereafter. Resto-ration of blood flow was confirmed by noting theimmediate reactive hyperemic response detected bythe coronary flow velocity probe.

Data AnalysisThe data were analyzed by BMDP5V, an extended

repeated measures program. Further analysis byeither paired t test or F test determined significantdifferences between the various parameters withsignificance accepted at p<0.05. Values are ex-pressed as mean±SEM.

ResultsThe multivariate analysis of repeated measures for

incomplete data was chosen by our biostatistician asthe appropriate method since our data did not havecompound symmetry and, hence, were not acceptablefor univariate analysis. The Wald tests of significanceof fixed effects and covariates were used, and a valueof p<0.02 indicated statistical significance fortreatment-time interactions. Further analyses bypaired t test compared the effect of treatment at eachtime as well as between time periods.

eooo A--

ow vo_!/~~ ~ ~-0

400+

0.200

0.160+0

c E 0.120.E E70.08040.040+

100}

90t804.704.

cNc601.a E 50

4040O1~

20+10

I><+

1

1

4 A

i fI

rT '/1'

us a-

R

pr.-occ 60@ 90' 120'

TME

FIGURE 1. Graphs showing effects of superoxide dismutaseplus catalase on ischemia-reperfusion-induced changes inthromboxane B2 (TxB2) formation (bottom panel) and TxB2concentration (top panel) in cardiac lymph and in cardiaclymph flow (middle panel). *, TxB2 in control group after a60-minute occlusion (n= 7); o, TxB2 in presence ofsuperoxidedismutase plus catalase (n= 7).

As previously described, coronary occlusions last-ing 60 minutes followed by reperfusion result inelevation of TXB2 excretion into cardiac lymph (Fig-ures 1 and 2). Elevated TXB2 excretion was notobvious until the onset of reperfusion (data notshown) but rose to significant levels (p<0.01) imme-diately at the onset of reperfusion. Peak TXB excre-tion rate was reached, and significant increases weremaintained within the first 2 hours of reperfusion.The increase in TXB2 release rate was paralleled byboth an increase in cardiac lymph flow (p<0.01) andTXB2 concentration (p<0.05) in cardiac lymph, bothof which persisted throughout the 2-hour period ofreperfusion (Figure 1). Coronary blood flow studiesdemonstrated identical reactive hyperemia responsesin all groups; coronary flow returned to normal in10-15 minutes (data not shown).

zuu 0~~~ I

U.v 4-

4

n xan.

by guest on July 11, 2018http://circres.ahajournals.org/

Dow

nloaded from

1042 Circulation Research Vol 66, No 4, April 1990

35C

cr 250' 200

x,

z 50I5l100

<E 50 1,I-

-50+430 60 90 120

REPERFUSION (MIN)

Considerable data exists within the literature sug-gesting that endothelial injury in the constrictedcoronary artery can occur consequent to coronaryartery instrumentation and constriction"1; therefore,we performed an experiment in which coronaryocclusion was maintained for only 5 minutes but inwhich the remainder of the protocol was identicalwith that used for longer occlusions. No elevation inTXB2 concentration occurred consequent to a 5-minute coronary occlusion in any of the animalstested (Figures 2 and 3). Thus, it seems unlikely thatcoronary instrumentation per se is resulting in endo-thelial damage sufficient to activate blood elementsto excrete cyclooxygenase products.

Infusion of SOD-CAT prevented a significant risein TXB2 excretion rate observed in the untreatedanimal (p>0.1; Figures 1-3). SOD-CAT treatmentalso prevented the increase in cardiac lymph flow andTXB2 concentration in cardiac lymph (Figure 1).Independent effects of SOD and CAT infusion illus-trated that SOD alone reduced TXB2 generation sothat no significant rise in TXB2 formation could beobserved when compared with baseline (p>0.1)(Figure 3). At 90 minutes of reperfusion, the values

= 60 MIN N=7

M SOD+CAT N=7

5S MIN N=4

FIGURE 2. Bar graph showing percent changein thromboxane B2 (TXB2) concentration duringreperfusion after a 60-minute coronary occlusion.w, TXB2 in control; *, TXB2 in presence ofsuperoxide dismutase plus catalase (SOD+CAT);E, TXB2 during reperfusion after a 5-minuteocclusion; N, number of animals. Error barsrepresent SEM.

150

were not significantly different from those observedwith SOD and CAT (p>0.1) combined and weresignificantly lower (p<0.05) than those in theuntreated control.SOD infusions (in the presence or absence of

CAT) are uniformly effective in reducing TXB2 for-mation during the first 90 minutes after reperfusion.After 90 minutes, there was considerable variability.In a number of animals, partial or complete return ofTXB2 concentration to levels observed in untreatedanimals with ischemia and reperfusion was observed(Figures 1-3), and cardiac lymph flow and TXB2concentrations also increased (Figure 1). Becausethe original protocol called for administration of thepotential therapeutic agents to cease 15 minutes afterthe onset of reperfusion, it was reasoned that poten-tially effective agents may have been cleared from thecirculation within 2 hours. Therefore, we examinedthe effects of an SOD infusion that began 15 minutesbefore the onset of ischemia and persisted for 75minutes after the onset of reperfusion. Cessation ofthis prolonged SOD infusion resulted in a suppres-sion of TXB2 for as long as 3 hours, after which therewas a trend toward increased excretion (Figure 4).

C. 60 MIN (N-7)

M SOD + CAT (N-7)

LW SOD (N-3)

FIGURE 3. Thromboxane B2(7XB2) production before occlu-sion and 60, 90, and 120 minutes

0 CAT (N-4) after total coronary occlusion inall experimental groups. n, TXB2

W S MUN (Nm4) in control group after a 60-

minute occlusion; *, TXB2 in thepresence ofsuperoxide dismutase(SOD) plus catalase (CAT); 2,TXB2 with SOD alone; E, TXB2with CAT alone; X, 5-minuteocclusion with no drug, followedby reperfusion; N, number ofani-mals. Drugs were given for a totalof 90 minutes including the first15 minutes of reperfusion. Errorbars represent SEM.

z0

150

125

._

cm

100

75

50

T

25

0PREOCCLUSION 60 90 120

REPERFUSION (MIN)

by guest on July 11, 2018http://circres.ahajournals.org/

Dow

nloaded from

Michael et al Effect of Free Radical Scavengers on Thromboxane B2 1043

200 T

150}

100 +504

OE

T T f T11---- T

11~~~~~~c

T-- -G 010 60 90 120 150 180 210 240 270 300

REPERFUSION (MIN)

FIGURE 4. Thromboxane B2 (7XB2) concentrations in car-diac lymph at times shown during reperfusion after a 60-minute coronary occlusion with etended superoxide dismu-tase treatment for the first 75 minutes of reperfusion. N,number of animals. Error bars represent SEM.

DiscussionA large series of studies has been done to demon-

strate that free radical scavengers exert favorableeffects on a variety of acute ischemic models includ-ing myocardial infarction1,3 and stunned myocardi-um.4 The original work in Lucchesi's laboratory (seeReference 1 for summary) suggested that SOD plusCAT was effective in preventing the additional injuryoccurring during reperfusion after a coronary occlu-sion. Subsequently, it has become clear that SOD isthe major operant in the reduction of infarct size andthat its protective effect only requires its presenceduring the time of reperfusion.3 It is important toemphasize that the effect of SOD plus CAT or SODalone is highly dependent on the protocol used in thestudies. Measurement of infarct size after longerperiods of occlusion or after longer periods of reper-fusion shows variable effects with regard to theusefulness of free radical scavengers. It is likely thatfree radical scavengers are required for some signif-icant period of time during reperfusion to ultimatelyreduce the size of a myocardial infarction and thatshorter periods of treatment may only delay but notdecrease infarct size.' At any rate, the concentrationused in this study was identical with that whichminimized infarct sizel,3 and markedly improved therecovery of myocardial function after a 15-minutereversible coronary occlusion (i.e., reduced stunnedmyocardium).4

In previous studies from our laboratory,9 weobserved marked elevations in TXB2 concentrationseen in cardiac lymph subsequent to a coronaryocclusion and demonstrated that this could beblocked by either ibuprofen or prostacyclin. Sincethromboxane formation is both accompanied by12and may be potentiated by7 the generation of oxygenfree radicals, this study was designed to examine theeffects of a free radical scavenger with known salu-tory effects on myocardial ischemia on the generationof thromboxane.

Molecular oxygen species are potentially a majorfactor in the genesis of myocardial ischemia-reperfu-sion injury.'-3 Several pathways may be responsiblefor the generation of oxygen free radicals at the timeof reperfusion including 1) production during lipoxy-genase or cyclooxygenase activation,12 2) release ofsuperoxide anions by activated neutrophils migratinginto ischemically damaged tissue,13,14 3) accumula-tion of hypoxanthine and xanthine during myocardialischemia catalyzed to uric acid by xanthine oxidase,15and 4) oxidation of tissue catecholamines.16 In ourexperiments, an administration of SOD and CAT 15minutes before a 60-minute occlusion that continuedfor 15 minutes into reperfusion resulted in decreas-ing concentration of cardiac lymph TXB2 concentra-tion. Since SOD and CAT act extracellularly, it islikely that the molecular oxygen species affected werein the extracellular compartment and, therefore,were, at least in part, of leukocyte origin."'15We suggest that TXB2 formation begins during

ischemia and relates, in part, to the length of isch-emia since coronary occlusion alone (in short dura-tion) does not seem to result in increases in TXB2formation. The initial reperfusion collection periodfrequently demonstrated equal or higher TXB2 excre-tion rates than future reperfusion periods. This sug-gests that, at least in part, the initial TXB2 has beengenerated during the ischemic period and appears incardiac lymph at the onset of reperfusion as previouslysuggested by our laboratory'0 on examination of cre-atine phosphokinase appearance in cardiac lymph.The fact that the TXB2 excretion remains elevatedand, in some cases, continues to rise suggests that anongoing process during reperfusion continues to pro-mote TXB2 excretion.

This study does not directly speak to the origin ofthe thromboxane production; this autacoid is pro-duced by both neutrophils and platelets, and each ofthese blood elements is thought to be pertinent to thepathophysiology of myocardial ischemia.'7 However,despite the fact that coronary blood flow returns tonormal within 10-15 minutes, increased TXB2 gen-eration is associated with increased coronary lymphflow during the 2 hours of reperfusion. This findingsuggests the presence of a pathophysiological factorthat alters microvascular permeability associatedwith the 1-hour ischemia-reperfusion protocol. Inview of the postulated pathophysiological role ofneutrophils in alterations of the microvascular endo-thelium in ischemic damage, the possibility that aneutrophilic origin of some of the observed TXB2appears likely. It would be tempting to speculate thatincreased neutrophil activation results in microvas-cular damage and also results in increased TXB2production. These results are compatible with thesuggestions of others that oxygen free radical gener-ation is an important part of neutrophil-induceddysfunction and damage in tissuel'13,15 and in thegeneration of lipid-derived autacoids.7

In this study, SOD is the active agent in reducingTXB2 formation in this ischemia-reperfusion proto-

z0

00~0-

mx

LL.0

0zI:0UE

by guest on July 11, 2018http://circres.ahajournals.org/

Dow

nloaded from

1044 Circulation Research Vol 66, No 4, April 1990

col. It is important to point out that the absence of aneffect of CAT does not necessarily belie its functionas an oxidative scavenger; it is possible that theendogenous CATs and peroxidases present withinthe cells in question and within the circulation aresufficient to exert an optimal effect.

In our experiments, TXB2 and cardiac lymph flowbegin to rise within 3 hours after discontinuing theSOD infusion; this occurrence further suggests thatongoing SOD treatment may be necessary for pro-tection for a greater length of time. The infusion timerequired for ultimate protection from a 1-hour isch-emic injury is not determined by this study. Recentwork by Simpson et a118 has suggested that periods upto 72 hours are required for ultimate prevention ofneutrophil-derived damage.

Therefore, in addition to the direct toxic effect ofperoxidative tissue injury, oxidative stress may alsocatalytically stimulate the production of vasoactiveautacoids that may induce vasoconstriction andplatelet aggregation.17,19 Together with the microvas-cular effects of leukocyte adherence and aggre-gation,20,21 these effects may induce severe endothe-lial damage and microvascular dysfunction associatedwith occlusion-reperfusion injury.

AcknowledgmentsThe authors gratefully acknowledge the expert

technical assistance of Peggy Jackson and GaryLiedtke and the manuscript typing and preparationby Ms. Concepcion Mata.

References1. Simpson P, Fantone J, Lucchesi B: Oxygen radicals and tissue

injury, in Proceedings of the Upjohn Symposium, Augusta,Michigan, April 27-29, 1987. Bethesda, Md, Federation of theAmerican Society for Experimental Biology, 1988, pp 63-77

2. Mullane K, Salmon J, Kraemer R: Leukocyte-derived metab-olites of arachidonic acid in ischemia-induced myocardialinjury. Fed Proc 1987;46:2422-2433

3. Ambrosio G, Weisfeldt M, Jacobus W, Flaherty J: Evidencefor a reversible oxygen radical-mediated component of reper-fusion injury: Reduction by recombinant human superoxidedismutase administered at the time of reflow. Circulation1987;1:282-291

4. Myers M, Bolli R, Lekich R, Hartley C, Roberts R: Enhance-ment of recovery of myocardial function by oxygen free radicalscavengers after reversible regional ischemia. Circulation 1985;72:915-921

5. Handin RI, Karagin R, Boxer GJ: Enhancement of plateletfunction by superoxide anion. J Clin Invest 1977;59:959-965

6. Clark RA, Klebanoff S: Neutrophil-platelet interaction medi-ated by myeloperoxidase and hydrogen peroxide. J Immunol1980;124:399-405

7. Hemler M, Lands W: Evidence for a peroxide-initiated freeradical mechanism of prostaglandin biosynthesis. J Biol Chem1980;255:6253-6261

8. Shappell S, Hughes H, Smith C, Moake J, Mitchell J: Antiox-idants may protect against ischemia-reflow injury by modulat-ing reactive oxygen effects on cyclooxygenase/lipoxygenaseproducts (abstract). Pharmacologist 1987;29:A212

9. Michael LH, Hunt JR, Lewis RM, Entman ML: Myocardialischemia: Platelet and thromboxane concentration in cardiaclymph and the effects of ibuprofen and prostacyclin. Circ Res1986;59:49-55

10. Michael L, Hunt J, Weibaecher D, Perryman M, Roberts R,Lewis R, Entman M: Creatine kinase and phosphorylase incardiac lymph: Coronary occlusion and reperfusion. Am JPhysiol 1985;248:H350-H357

11. Schmitz J, Apprill P, Buja L, Willerson J, Campbell W:Vascular prostaglandin and thromboxane production in acanine model of myocardial ischemia. Circ Res 1985;57:223-231

12. Siedlik RH, Marnett Li: Oxidizing radical generated by pros-taglandin H synthase. Methods Enzymol 1985;105:412-416

13. Fantone JC, Ward PA: Role of oxygen-derived free radicalsand metabolites in leukocyte-dependent inflammatory reac-tion. Am J Pathol 1982;107:397-418

14. Higgs G, Bunting S, Moncada S, Vane J: Polymorphonuclearleukocytes produce thromboxane A2-like activity duringphagocytosis. Prostaglandins 1976;12:749-757

15. Roy RS, McCord JM: Superoxide and ischemia: Conversion ofxanthine dehydrogenase to xanthine oxidase, in GreenwaldRA, Cohen G (eds): Oxy-radicals and Their Scavenger Systems.New York, Elsevier Science Publishing Co, Inc, 1983, pp12-145

16. Wheatley AM, Thandroyen FT, Opie LH: Catecholamine-induced myocardial cell damage: Catecholamines or adreno-chrome. J Mol Cell Cardiol 1985;17:349-359

17. Moncada S, Vane J: Arachidonic acid metabolites and theinteractions between platelets and blood vessels. N Engl JMed1984;300:1141-1147

18. Simpson PJ, Fanton JC, Mickelson JK, Gallagher KP, Luc-chesi BR: Identification of a time window for therapy toreduce experimental canine myocardial injury: Suppression ofneutrophil activation during 72 hours of reperfusion. Circ Res1988;63:1070-1079

19. Lands W: Biochemical and cellular actions of membranelipids. Am Rev Respir Dis 1987;136:200-204

20. Engler R, Dahlgren M, Morris D, Peterson M, Schmid-Schonbein G: Role of leukocytes in response to acute myocar-dial ischemia and reflow in dogs. Am J Physiol 1986;251:H314-H322

21. Engler R, Covell JW: Granulocytes cause reperfusion ventric-ular dysfunction after 15-minute ischemia in the dog. Circ Res1987;61:20-28

KEY WORDS * myocardial ischemia * myocardial reperfusion* cardiac lymph * thromboxane B2 * free radical scavengers

by guest on July 11, 2018http://circres.ahajournals.org/

Dow

nloaded from

L H Michael, Z Zhang, C J Hartley, R Bolli, A A Taylor and M L Entmanmyocardial ischemia and reperfusion.

Thromboxane B2 in cardiac lymph. Effect of superoxide dismutase and catalase during

Print ISSN: 0009-7330. Online ISSN: 1524-4571 Copyright © 1990 American Heart Association, Inc. All rights reserved.is published by the American Heart Association, 7272 Greenville Avenue, Dallas, TX 75231Circulation Research

doi: 10.1161/01.RES.66.4.10401990;66:1040-1044Circ Res. 

http://circres.ahajournals.org/content/66/4/1040World Wide Web at:

The online version of this article, along with updated information and services, is located on the

  http://circres.ahajournals.org//subscriptions/

is online at: Circulation Research Information about subscribing to Subscriptions: 

http://www.lww.com/reprints Information about reprints can be found online at: Reprints:

  document. Permissions and Rights Question and Answer about this process is available in the

located, click Request Permissions in the middle column of the Web page under Services. Further informationEditorial Office. Once the online version of the published article for which permission is being requested is

can be obtained via RightsLink, a service of the Copyright Clearance Center, not theCirculation Researchin Requests for permissions to reproduce figures, tables, or portions of articles originally publishedPermissions:

by guest on July 11, 2018http://circres.ahajournals.org/

Dow

nloaded from