JAW VOL 1% No. I
197
laaaa.y1992'157-151
EXPERIMENTAL STUDIES
Low Doses of Superoxide Dismutase and a Stable Prostacyclin
Analogue Protect in Myocardial Ischemia and Reperfusion
XIN-LIANG MA, MD, PHD, GERALD JOHNSON III, PHD, ALLAN M . LEFER, PHD, FACC
Philadelphia, Pennsylvania
The effects of low dose her an saperoxido disnutase and low dose
taprodern, a stable analogue of prostacydln, were investigated
separately and logether in a model of myocardial Isrhemla (1 .5 h)
with rrperfuslan (4.5 h) In open chest, anesthetized cats. Tapros •
tewe (60 a)kg per min), human superoxide dismumse (0 .25 mg/kg
per h) . built ageata Wgeto., or their vehicle, were infused
intravenously to cats starting 0 .5 h after occlusion or the left
anterior descending coronary artery. Neither low dose taproslene
nor low dose human superoxide dismumse exerted any endotheltal
or myocardial protection In this model . However, the two agents
together showed a significant eadothelid and myocardial prreterthat in cats with myocardial iscbemia and reperfasioa . Compared
with cats that were untreated or received only mprostem or
human vnpaoxide dismotase, cats receiving both agents exhibited
a lower plasma creates kinose activity at every lime point
observed after reperfuslat, a reduced arm of cardiac Borealis
17 ± 2% vi. 21 t 5% am al risk, p < 0.001), lower myeloper .
oxidase activity In she lssiemle region (p < *At) and a dgotfand
preservation of vasorelavmlt responses of bill anterior desexudlng
coronary rings In erMMsNwmdepeadent visodilatas, acetyteho .
line (p < 0.001) and A-23197 (p < 0.001) . Tapraekm appears to
act oddhlvely wtm human speroxide dlaututase m Inhibit metrophe adherence aW activation and to locllvutr nperv.dde rm0•
cats, and thus reduce cellular Injury 4.5 h after reperfudua of the
Ischemte heart . Use of tits agent may allow law dtna orsaperns•
Ide ditmutase to be used mare effectively In early myoardid
Iseh®da.
U Aar Cdr CJNo>t 1992,-0-1#7_204)
Evidence is accumulating that oxygen-derived free radicals
play a critical role in endothelial cell and myocyte injury
associated with myocardial ischemia and reperfusion (1-3) .
Substances that either scavenge free radicals or inhibit their
formation have been demonstrated to protect the endothelial
cells and myocardium from ischemia-reperfusion injury (46), although negative studies (7) also have been reported .
fusion injury in a variety of animal models (I I-16) . Because
taprosiene and iuman superoxide dismutase protect against
free radical-produced endothelial and myocardial injury by
different mechanisms, these two substances have the potential to exert an additive effect in protecting against reperfo .
sion injury . fn this study, we infused taprosteree and human
superoxide dismutase together in low doses that were ineffective when used separately . This was done to avoid
Recently, high doses of human superoxide dismutase were
shown (A) to increase infarct size in the same preparation in
which low doses provided protection . Taprostene, a stable
prostacyclin analogue, has also been found to protect against
ischemia and reperfusion-induced endothelial dysfunction
and cardiac myocyte injury (9) . These effects were attributed
to an inhibitory effect on neutrophil adherence and activation, therefore decreasing the release of oxygen-derived free
radicals (10) . On the other hand, free radical scavengers,
particularly recombinant human superoxide dismutase,
which scavenges superoxide radicals, protect against reper-
15 w the Npartmrnl of PhysioWgy, Jde- Medlral College, Tromas
Inferior University, Philadelphia . Prnnsyl-app, Ibis study aces supported in
part by gescarm Grant HL 25575 from the Nadonal Hear. Lang . cad Blood
Jnstiune, Naierul hadames of Health . Bedi sda. Maryland aced supported in
pan by a grant tense Geunemhsl Gmbtt . Aarhen . Germany- Hr- leheon was
a Senior Research MI- of the Isehenia-Shock Research Institi rte . Thomas
reffersnn .University
Phaadelwa at the time of the study .
received)-7,
Marwssthsreceived February 22,1991; revised manuscnpc
1991, attepted My 1 . 1991 .
Address for r aixtx; Allan M. Lefer, FIX . Department of Physiology,
reffavm,.]Made
College, Thomas Jefferson Univeesily . 1020 Locust Stmt.
Philadelphia 14nnsytvania 19107-6799,
E1e92 by it, American College
orCa.dralogy
hypoteusive and coronary steal-inducing effects of taprosteee and potential infarct-enhancing effects of higher doses
of superoxide dismutase . The major purposes of this study
were 1) to determine whether t aprostene and human superoxide dismutase have additive effects in protecting endothelial cells and myocardium from iscbemia and reperfusion,
and 21 to study the mechanism by which combined administration of taprostene and human superoxide dismutase may
achieve cardioprotective effects. These studies, H positive,
would tend to prevent undesirable side effects of both
agents, ana therefore be a more useful ihaapeutic modality .
Methods
Animal preparation, Adult male cats (2.6 to 3 .3 kg) were
anesthetized with sodiuu, pentobarbital (30 mgdtg body
weight, intraveuouslyl . An intratracheal carmula was inserted through a midline incision, and all cats were dated on
intermittent positive-pressare ventilation (Harvard small animal respitalorl. Polyethylene catheters were inserted into
the right external jugular and the right femoral veins for
m ssaovrrvym.so
198
MA ET AL .
SUPEROXIDE DISMUTASE AND PROSTACYCLIN ANALOGUE IN ISC1tEMtA
infusion of additional sodium pentobarbital . drugs or their
vehicle . The right femoral artery was also cannulated and
connected to a Statham P23AC pressure transducer (Gould)
to record mean arterial blood pressure electronically . A
midstemal thoracolomy was performed and the pericardium
was opened. A 2.0 silk ligature was placed around the left
anterior descending coronary artery 8 to 10 mm from its
origin . Standard lead 11 of the scalar electrocardiogram
(ECG) was used to determine heart rate and ST segment
elevation . The ECG and mean arterial blood pressure were
continuously recorded on a Grass model 7 oscillographic
recorder. The pressure-rate index, an approximation of
myocardial oxygen demand, was calculated as the product of
mean arterial bland pressure and heart rate divided by 1,000.
Experimeafal protocol . After completion of all surgical
procedures and a 30-min stahiliration period, the eats underwent a baseline reading of heart rate and mean arterial blood
pressure; the initial blood sample was dawn ai this time.
Myocardial ischemia (defined as time 0) was produced by
lightening the previously placed reversible ligature around
the left anterior descending coronary artery to completely
occlude the vessel . After 1.5 h of ischemia . the ligature was
untied and the ischemic myocardium was reperfused for
4 .5 h, resulting in a total observation period of 6 h . Thirty
minutes after coronary occlusion, taprostette (pH 9)1Grunenthal GmbH), human superoxide dismutase 13,109 superoxide dismutase Ulmg protein . Grunenthal GmbH), the two
agents together . or their vehicle, were given as a corstaot
infusion for the romaindcr of the experiment (5 .5 h tota ) . In
previous experiments 19.15) using the same car model of
myocardial ischemia and reperfasion . we have demonstrated
that infusion of 100 np/kg per min taprostene or 5 mg/kg per
h human superoxide dismutase separately produced significant endothelial and cardiac protection . To obtain an infusion rate that produced a marginal amount of myocardial
protection when used alone, we initially employed a variety
of infusion rates of taprostene (20 to 80 ng/kg per min) or
human superoxide dismutase (0 .1 to I mg/kg per h). An
infusion rate of 60 ng/kg per min taprostene and 0 .25 mg/kg
per it human superoxide dismutase was eventually chosen .
Administration of human superoxide dismutase or iaprostene separately of a higher dose protected against myocardial
injury, but combined administration of these agents at a
lower dose did not exert full protection . The cats were
randomly separated into four groups subjected to myocardial
ischemia: group a, myocardial ischemia + vehicle (n = 6) ;
group b, myocardial ischemia + taprostene (n = 7); group c,
myocardial ischemia + human superoxide dismutase (n = 6)
and group d, myocardial ischemia + taprostene + human
superoxide dismutase (n = 6). Additional cats were subjected to a sham myocardial ischemia procedure in which all
procedures were identical to those in the cars subjected to
myocardial ischemia except that the ligature around the left
anterior descending coronary artery was not lightened . This
model of myocardial ischemia in the car has been previously
shown to have very low collateral blood flow in the necrotic
JACC V.I. W. Na . I
January 1992497-201
region (17) . At 5 h after coronary ligation . myocardial blood
flow determined with 15 pm mierospheres was found to be
5"c of control coronary flow (17).
Plasma trealine Mum (CK) Analysis. Arterial blood samples (2 ml) were drawn immediately before ligation and hourly
thereafter . The blond was collected in polyethylene tubes
containing 200 IU of heparin sodium . Samples were centrifuged at 2 .000 x g and 4'C for 20 min and the plasma was
removed for biochemical analysis. Plasma protein concentration was assayed with the biaret method (181 and plasma CK
activity was measured with the method of ltosalki (19) and
expressed as lUlmg protein x t0- '.
Myocardial Ussat analysis. At the end of the 6 h experimental period, the ligature around the left anterior descending coronary artery was retightened . Then 30 ml of 0 .5%
Evans blue dye was injected into the left atrium to stain the
area of the myocardium that was petfused by the patent
coronary arteries. The area at risk was thus determined by
negative staining . The heart was excised rapidly and placed
in warmed, oxygenated Krebs-Henseleit buffer . The left
circumflex and left anterior descending coronary arteries
were isolated and removed for the subsequent study of
coronary ring vasoaclivily. The right ventricle and great
vessels were removed and the left ventricle was sliced
parallel to the atrioventricular groove in 3-trim (hick see
.
. The unstained portion of the myocardium (that is, the
Lions
total areail risk) was separated from the Evans blue-stained
portion of the myocardium Ithat is, the area not at risk). The
area at risk was again sectioned into i-mm thick slices and
incubated in 0 .1% nitroblue letrazolium in phosphate buffer
at pH 7.4 and 37 0C for 15 min . The lelrazulium dye forms a
blue formazan complex in the presence of coenzyme and
dehydrogenases . The necrotic portion of the myocardium at
risk that did not stain was separated from the stained portion
of the myocatdium (that is. the ischemic but nonnecmtie
area). All three portions of the left ventricular myocardium
Inonischemic . ischemic nonnecrotic and ischemic necrotic)
were weighed and the results are expressed as 'be area at
risk as a percent of the total left ventricular mass ; the area of
necrotic tissue is computed as a percent of the area at risk
and the total left ventricular mass . The three portions of the
myocardium were then stored at -70'C for subsequent
myeloperoxidase assay .
Isolated coronary ring studks . Both left circumflex and
left anterior descending coronary artery segments removed
were placed into warmed Krebs-Henseleit buffer consisting
of (in mM) : sodium chloride 118; potassium chloride 4.7. ;
calcium chloride-211,O 2 .54 ; potassium biphosphate .119;
magnesium sulfate-711420 1 .19; sodium bicarbonate 12.5 ; and
glucose 10 . Isolated coronary vessels were cleaned and cut
into rings of 2 to 3 mm in length . The rings were then
mounted on stainless steel hooks, suspended in tissue baths
and subse,aently canaceled to [T-03 farce displacement
transducers (Grass Instrument) to record changes in tension
on a Grass model 7 osc'lographic recorder . The baths were
filled with 20 ml of Krebs-Henseleit buffer and aerated at
JACC Vol . I9. No. I
Janaaq 1992 :197-301
199
MA ET AL.
SUPEROXIDE DISMUTASE AND PROSTACYCLIN ANALOGUE IN LSCHEMIA
37°C with a gas mixture of 95% oxygen and 5% carbon
dioxide. Coronary rings were initially stretched to give a
preload of 0.5 g of force, and allowed to equilibrate for I h .
In a previous study, we have observed that preloads of wI g
caused injury to the endothelium and thus could not be used
(15). During this period, the Krebs-Henseleit buffer in the
tissue baths was replaced every 20 min . After equilibration,
the rings were exposed to 10 ngiml U-46619 (Upjohn), a
thromboxane A2 mimetic used to generate about 0 .5 g of
developed force. Once a stable contraction was obtained,
0 .1, I, 10 and 100 nM acetylcholine was added to the bath .
After the response stabilized, the rings were washed and
allowed to equilibrate to baseline once again . The procedure
+vas repeated with Ad31a7 (0 .001 . 0.01 . 0.1 and I µM1 and
. Sodium
then with sodium nitrite (0.1 . 1, 10 and 100 µM)
nitrite was prepared by dissotvin4 the compound in 0 .1 N
hydrochloric acid and titrating it to pH 2 . Titrating distilled
water to pH 2 and adding aliquots to buffer in the bath did
not produce any vasorelaxation .
Determination oftissuentyeloperoxidaseactivity . Myeenrdial myeloperoxidase activity occurring virtually exclusively
in neutrophils . was determined with the method of Bradley
et al . (20) end Mullane et al . (21) . The myocardium was
homogenized in 0.5% hexadecyltrimelhyl ammonium bromide (HTAB) (Sigma Chemical) and dissolved in 50 nLM
potassium phosphate buffer (pH 6) using a Potylron (PCU-2)
homogenizer . Homogenates were centrifuged at 12.5(0 x g
and 2° C for 30 min. The supernatants were then collected
and reacted with 0 .167 mg/ml of o-dianisidine dihydrochto .
ride (Sigma Chemical) and 0.01105% hydrogen peroxide in
50 mM phosphate buffer (pH 6) . The change in absorbance
was measured spectropholometrically at 460 not . One unit of
myeloperoxidase is defined as that quantity of enzyme
hydrolyzing I µm01 of peroxide/min at 25'C .
Statistical analysis . All values in the text, table and figures are presented as mean values t SEM of n independent
experiments. All data were subjected to analysis of variarr;;
followed by the Bonferroni correction for multiple comparisons by post-hoc r test . Probabilities 3.05 were considered
to be statistically significant.
Results
lkmodynamie and ECG changes. In preliminary studies,
we observed that there were no significant changes in any of
the hemodynamic or biochemical variables observed during
the 6-h observation period in cats with sham myocardial
ischemin. infusion of taprostene (60 nglkg per min), human
superoxide dismutase (0.25 mgrkg per h), or their combination had no effect on any of the variables observed in these
cats (Table 1). Before occlusion of the left anterior descending coronary artery, heart rate, mean arterial blood pressure
and pressure-mte index values were quite similar in the four
groups of cats with myocardial ischemia. After occlusion,
mean arterial blood pressure dropped sharply and heart rate
decreased, leading to a sharp decline in pressure-rate index .
30
25
20
15
t
G
R
10
5
0
1
2
3
4
5
6
TWO (Smel)
Figure] . Pressure-rate index (mean arterial blood pressnmIMABPI
times heart rate IHR1) expressed as mm Hg x (beats/min/1
.000
sampled hourly during 6 h of myocardial ischemia followed by
reperfusion . All values are mean values } SEM for six or seven
cats . O---- O - myocardial ischemia + vehicle ; 4-O = myocardial ischemia + taprostene alone;1s-A = myocardial iscbemia +
human supcroxide dismutate ; A-A = myocardial Ischemia +
combined treatment with taprmrene and human snperonlde dismutasc.
There was no significant difference among the four groups in
mean arterial blood pressure, heart rate and pressure-rate
index either during the ischemic period or after reperfusion .
Therefore, any cardioprotective effect observed in any of the
treated groups could not be attributed to a reduction in
myocardial oxygen demand (Fig . 1) .
No ST segment elevation was observed in any group of
cats studied before coronary occlusion . However, soon after
occlusion, the ST segment became significantly elevated,
reaching a peak at 20 to 40 min after occlusion . Furthermore,
ST segment elevations were not trigtiiicantly different among
any of the four gronps studied, indicating that there were no
significant ,iirterences in the severity of the ischemia pro .
aced as a result of the occlusion (Fig . 2) . In cats in the sham
Figero L ST segment elevation after occlusion of the left anterior
descending coronary artery . All values are mean values ' SEM for
six or xvcn cats. Soon after coronary occlusion, ST segment
elevated significantly . No differences were observed among groups .
Symbols as in Figure ( .
a3D
0.25
0.20
0.13
0.10
0.05
-0.05
-0 .10
0
20 40
it,
21, 3h
r
4h
5h
ah
2%
MA FT AL.
SUPFRIIXIDE DISMUTASE AND PROSTACYCLIN ANALOGUE IN ISCHEMIA
an
IF .
0
1
2
4
3
Time (Hours)
5
6
Fugue 3. Plasma creatine kinase (CK) activity exprwscd as imernatianal . Units (W)/mg protein x 10-1 "-asured at v:c-c;
dtirin g 1 .5 h of ischemia and after 4 .5 h of reperfuston . All values are
mean values 2: SFM of five to seven cats . Plasma CK activity rises
slightly following ischemia and increases significantly after repedsSion. ap < 0.05, "p < 0.01 versus myocardial bohemia + vehicle ;
0---0 = myocardial ischemia + human superoxide dismutase ;
other symbols as in Figure I
.
myocardial ischemia group, the ST segment was 0 ± 0 mV at
all times.
Myneirdial prmccdhn by human superoxde dismatsae ad
taprmtex. Plasma (K activity . Figure 3 summarizes the
changes in plasma CK activity in all four groups of cats with
myocardial ischemia. In preliminary studies in control sham
myocardial ischemia cats, the plasma CK activity increased
only slightly throughout the experiment, reaching a final
value of 10 ± 3 IUhng protein X 10 - ' end administration of
human superoxide dismutasc, taproslenc or their combination did not change the plasma CK activity (Table I) .
However, large increases in plasma CK activity were oh .
served in untreated cats subjected to myocardial ischemia
and reperfusion. Neither the myocardial ischemia + taprosteno group nor the myocardial ischemia + human superoxide
dismutase groups showed significant ;otection when compared with the untreated myocardial ischemia group, because ah cats given either agent alone exhibited markedly
elevated CK activity associated with myocardial ischemia
and reperfusion . There was no significant difference in
plasma CK activity at any time point observed between
taprostene- or human superoxide dismutase-treated groups
and the untreated group . In contrast, the myocardial inch .
emia + combined treatment group experienced a markedly
JACC Vol . 19. No. I
January IM :1m-204
lower plasma CK activity . The plasma CK activities of the
myocardial ischemia + combined treatment group were
significantly different from those of the myocardial ischemia
+ vehicle group et each time point after 2 h . Thus . combined
treatment with low doses of taprostede and human superoxide dismutase effectively inhibited plasma accumulation of
CK activity, an index of myocardial cellular injury .
Percent necrotic tissoe. Another more direct index of
myocardial injury induced by ischemia and reperfusion is
anatomic estimation of necrtic tissue . The wet weights of
the areas of myocardium subjected to ischemia (area at riskl
expressed as a percent of the total left ventricular weights
were not significantly different among any groups, providing
additional evidence that the severity of the ischemic insult
was comparable in all myocardial ischemia groups. i n contrast, the weights of the necrotic myocardial tissue expremed either as a percent or the area at risk or as the total
left ventricular weight were not equivalent among these
groups. A relatively large percent of necrotic tissue was
observed in the untreated myocardial ischemia group- Treatmew with either low dune taprostene or human superoxide
dismutase alone only slightly decreased she percent of necrotic myocardial tissue. There were no statistically significant differences among these three groups . However. when
low doses of taprostene and human supemxide dismurase
were infused together, the percent of necrotic myocatdrdi
tissue expressed either as area at risk or as total left
ventricular mass was significantly lower than in any of the
other myocardial ischemia groups (Fig . 4). indicating a
significant cardioprmective elyect of the combined therapy .
Area at risk was 0 in all cats with sham myocardial ischerain.
In four vehicle-treated cats, the area at risk was randomly
divided into two portions and incubated wilt mianisure
tetrazolium solution seauralsiv in the presence and absence
of 2v pgrnd human superoxide dismubne and 10 nglml
taprossene . The results indicated that addition of human
superoxide dismutase and taprastene in vitro did not have
any effect on nitrobl ue tetrazolium staining .
Effect of lemon sapervaide dinmem+e and esprowae on
ekaagcs in myocardial myekperwtfdsm sdiviy. One of the
major mechanisms thought to be responsible for myocardial
rererfusion injury is infiltration of neutrophils into the reperfused ischemic region . Myeloperoxidase activity is specific
for nemrophils, and generally accepted as a marker for
Table 1. HSects of Human Superoxide Maturate and Taprostene on tlemedynamic Variables and Creatine Kinase Activity in Cats With
Sham Myocardial Ischemia
HR Ibealstotn)
Ini'ul
Final
Sl.mtL
Sham tAr v h3ll) v 7np
10114
194 2 14
NIAOP trio, Hat
Initial
Fiat!
176`13
119±10
182 1
_ 12
116 ± 9
lot 11
105±_11
PRI
IHRvMAuP11Agp)
Initial
Final
23 m4
22±3
1923
19±2
CK
(IUlmgpdeinxlp - r)
Initial
Final
1.x±0.6
10-3
1.6±0.6
11 , 3
Values an man values 2 SKM for tour to six c ats. C K creative tines,; HR - heart vale ; MOD = hutnan supemxide dismutase 10 .25 nrt/kt per hl;
MAOP = mean arterial bloom pressure : PRI = =re-ram index; Tap = up-t- (m nykg per mil) .
JACC V .1 19, No. I
CM LAD Cauaarp
40.0 I
ACh
A231e7
N .NOa
L
3010
20.0
20.0
PA + Vehicle
15.0
6 10.0
6 .0
$ 010 5.
M
King*
rt,anmt
e 35.0
a
201
MA ET AL.
SUPEKOXrDP MSMUTASE AND PA,0sTACYaJN ANALOGUE IN ISCHEMIA
January 1992 :197-•A,
F
J
}
raenaot~
MI + TaprgStm .
war m rah
w.c,aue
F-
..r .oral
0
L~6VJxl~_
Nm,oue
_
}
T.OW
Flgme,L Tissue wet weight of area at 0sk as a percent of the total
left ventricular wet weight . and of necrotic tissue as a percent of the
area, at risk and of the total left ventricle. Bar helxhts show mean
MI + hOOD
(La. c . ..)
values ; hrneketr indicate _ SEW for sit or seven samples . C7 =
myocardial ischemia + vehicle, M - myocardial ish. .-
}
taprostene; ® - myocardial irchem,a r human superonde dis
motase ; M = myocardial ischemia
}
}
combined treatment with
t
luptostene and human superoxide dismmase .
La
+ Tatrnstnna
t
l,
+ hSOD
Iem. 4 . .I
neutruphil accumulation in the heap . Figure 5 illustrates
myeloperoxidase activities in heart tissue samples from the
foot myocardial ischemia groups. In the nonischemic nonrepetfused myocardium (that is, area not at riskl, myeloperoxidase activity was very low in allgroups and there were no
significant differences among any of the groups . However,
marked increases in myeloperox±daac activity were observed in the a,
t risR and the necrotic area in the
myoeardisd ischemia+ vehicle group . When taprostene or
human superoxide dismulase was infused alone, cardiac
myeloperoxidase activity remained markedly elevated- In
contrast, the combined treatment group exhibited markedly
lower cardiac myeloperuxidase activities in both the area at
risk and the necrotic area, indicating that the combination of
low dose taprostene with low dose human superoxide dismutase significantly inhibits neutrophil infiltration into the
ischemic myocardium . Cardiac myeloperoxidase activity
Fienre 5. Myeloperoxidase activity in the area at risk, the ea+'ntic
area, and me area not at risk in Ul100 mg issue wet weight .
Conventions and symbols as in Figure 4.
11w~1
v
v
m
i
Nanonck.a
w.a Nat a xw,
}
}
Figure6. Representativereexdingofendetheliumdependentvasodilators . acelylehoiine (ACh) and A231B7, and endotheliumindependent vamdilator, sodium nitrate (NaN0 2}mdueed relax-
ation 4 precomracted (U-AM,19) iseiumic and repefilsed left
anterior descending (LAD) coronary artery rings . The arrows indicate the addition of U-x6619 ; the dw indicate the addition of
acetylcho:ine or sodum nitrite. Aeeryteholae nod A21187 induced
almost no relaxation to rind from cans in the myocardial iadwnia
(Ml) + vehicle pomp
. Neither taproatrne nor human nuperaxidc
dismutase IhSDD) administered alone improved the response of
coronary rings to acetylcholine. However, the acetykholine and
A21187 responses of rings isolated from a cat with myocardial
ischemia given combined treatment with taprmlene and human
superoxide dismuaase were rigridcatdly prermrd.
in cats with sham myocardial ischemia was 0 .04 ±
0 .02 VIS00 mg tissue.
Endothelial rotation df hattw open aide dkmtdnte and
taprmtem. Endothelial dysfunction or it jury has been demonstrated to be one of the earliest -ehaws observed after
rcpetfusion of ischemie myocardial tissue . We tested enda,thelial integrity by measuring vasoactivhy of corortary artery
rings after addition of the endothelium-dependent vasodilators, acetyleholine and A-23157, and the endotheliumindependent vasodilator, acidified sodium nitrite .
Figure 6 illustrates typical recordings of iseb(aic left
anterior descending coronary rings obtained from the differtmt groups . The response of coronary rings obtained from
the myocardial ischemia + vehicle group to the endothelium .
dependent vasodilators, aceeyleholine and A-23157, was
almost totally abolished at 1 .5 h after reperfusion, and
neither laprostene nor human superoxide dismutase infusion
alone preserved the vasorelaxant responses to either acetylcholine or A-23157. However, these coronary rings relaxed
202
MA ET AL.
SIIPFROXIVE DISMUTASF ANn PWO
rACYCLIN ANALOGVE IN ISCHEMIA
too tb
ea
70
60
50
40
30
29
10
0
am m
Axatar
Nanoz
Fleec 7. Summary of responses of ischemia-reperfused left anterior descending coronary artery sings 1 . 100 nM acetylch.lmc
(ACh), I µM A-23197 and 110 aM sodium nitrate INaNO,I . Hue
keights show means ; broduac indicate ±SEM for 9 or f 0rings. CI
= myocardial ischemia - vehicle ; M = myocardial ischemia +
taprostene; t to = myocardial isclimaia + human superoxide dismutase: - .- myVCarlial ischemia + laproslem + human super .
oxide dismutase .
fully when the endothelium-independent vasodilator sodium
nitrite was added, indicating that the responsiveness of the
coronary vascular smooth muscle to direct vasodilators
remained normal . In contrast, rings obtained from the combined taprostene- and human superoxide dismutase-treated
group significantly relaxed in response to both endatheliumdependent vasodilators (acetylcholine and A-23187) and to
the endothelium-independent vasodilator (sodium nitrite) .
Figure 7 summarizes the vasorelaxant responses to acetylcholine, A-23187 and sodium nitrite in isolated cat left
anterior descending ec=-onary artery rings . Clearly, the response of ischemic rings to the endothelium-dependent
vasudilatars was significantly preserved by combined
taprostene and human superoxide dismutase treatment . Vasorelaxant responses to acetylcholine and A-23167 were 96
± 6% and 97 ± 5%, respectively, in left anterior descending
coronary artery rings isolated from cats with sham myocardial ischemia . Moreover, the rlonischemic left circumflex
rings from all groups showed equal and complete relaxation
in response to acetylcholiine, A-23187 and sodium nitrite .
There were no significant differences in response to any of
the vasodilators studied among the groups of left circumflex
coronary artery rings . Thus, nonischemic coronary artery
rings relaxed completely in response to all vasodilators,
indicating normal endothelium and vascular smooth muscle .
Discussion
Rule of free radicals in reperfusivn injury . The early
restoration of myocardial blood flow after myocardial ischemia is essential to retard the progression of myocardial cell
death and to permit the functional recovery of reversibly
injured myocardium. However. many studies also point to a
potentially detrimental effect of reperfusion on endothelial
JACC Vol . 19. No . I
J.- 'y 1992]97-iM
integrity and myocardial function (12 .22) . Although the
cause of this eeperfusion-induced enhancement of cardiac
injury is apparently multifactorial. a mounting body of
evidence now indicates that oxygenaledved free radicals are
important mediators f ischemia •repelfusion-induced endu .
thelial and myocardial injury (2,15,23,24) .
A wide variety of cells . organdies and enzymes may be
involved in the free radical formation activated by isrhemia
and eepenfusion . These include neutrophils, xanthine oxi .
dose . eyclooxygenase and lipoxygenase . amooxidation of
catecholamioes . mpechondria and the sarcoplasmic reliculum (10,25,26)- Am^_ng these, xanthine oxidase localized in
endothelial cells (27-29) and nicotine adenine dinucleotide
phosphate, reduced form (N"ADPH) found in neutrophils
41,31 he"e been demonstrated y .obe important sources Affree
rdd ;cals in the reperfused ischemic heart . Free radicals
produced by sunshine oxidase in endothelial cells could act
as a powerful cherne ltraclant for nemrophils . Neutrophils
would then adhere to endothelial cells and generate larger
amounts of free radicals resultingin endothelial and myocardial injury . This sequence has been termed an "endothelial
cell trigger" and "nemrophil amplifier" mechanism by BulkIcy (30) .
Oxygen-derived free radicals may, exert diverse biochemical effects on both intracellular and extraecllnlar sites, and
there is extensive evidence that cardiac structure and function can be altered by these effects (75). It has been demonstrated that free radicals can initiate lipid peroxidation, alter
membrane permeability to ions and inactivate' ndotheliumderived relaxing factor (16), which is a potential protector of
endothelial and cardiac muscle cells in ischemia and reperfusion (3I ). Further evidence for the role of free radicals as
a major nediator of reperfusion injury was derived from the
studies in which free radical scavengers limited ultimate
infarct size (I3) and preserved the endothelium-dependent
relaxation of the reperfused ischemic coronary artery
(15.23) . lafusim of higher doses of human snperoxide t1smutase can scavenge the superoxide radicals produced from
xanthine oxidase, thereby decreasing neutrophil recruitment
(32). Human superoxide dismutase can also scavenge the
superoxide produced by neutrophils or any other cells and
therefore can protect endothelial cells and myocardium from
reperfusion injury directly.
Rile of seutrophifs in reperfusion injury. On the other
hand, substances that inhibit oeutrophil adherence and acts
valion, like monoclonal antibodies against CDI IICDIA adherence glycoprotein of neutrophils (6) and prostacyclin
(33), have been shown to be effective in protecting against
endothelial and myocardial injury associated with ischemia
and reperfusion. Taprostene is a synthalic, chemically stable
analogue of prostacyclin that has been specifically designed
to enhance its cytopratective actions while minimizing on .
wanted homodynamic effects (34) . In a previous study (9),
we observed them infusion of taprostene at a rate of 100 nglkg
per min exerted significant endothelial and cardioprotective
effects in a cat model of myocardial ischemia and reperfu-
1ACC Vot . 19. No . I
lmumry 199119n-210
MA ET AL.
SUPEROXIDE DISMVTsE ANn PROSTACOCLIN ANAIAGGE IN ISCHEMIA
sion. There area variety of possible mechanisms by which
taprostene could afford cardiopretection, Because infusion
of taprostene at a rate of IOU nglkg per min does not induce
any significant changes in mean arterial blood pressure or
presume-rate index, it is unlikely that taprostene produces a
significant coronary steal . However, taprostene significantly
attenuated myeloperoxidase activity, a tracker of neutrophil
infiltration, bath in the area at risk and in the necrotic area,
implying that prevention of neutrophil adherence and acti .
vation is a primary mechanism by which taprostene provides
its endothelial and cardioprotection,
Although a higher dose of human superoxide dismutase
exerts significant cardioprotective effects in some experiments, the relatively short pharmacologic half-life (6 to
10 min) of this agent represents a drawback to its study in
experimental protocols extending for several days after
reperfusion . Moreover, a recent clinical investigation 7)
indicated that administration of humeos superoxide dismutase had no protective effect against myocardial injury in
humans . Neutrophil chemoattraction to the ischemically
injured myocardium and the local formation of neutrophilderived free radicals continue to exert a cytotoxie influence
on the viable myocardial cells in the reperfused tissue.
Therefore, it would be expected that using substances that
inhibit neutrophil adherence and activation together with
human superoxide dismutase should have additive effects in
protecting endothelial cells and myocardium from reperfu .
sion injury .
Beneficial effect of combined human superoxide dismutase
and taproslene treatment me reperfusistn injury. We obtained
several lines of evidence that the combination of taerostene
and human superoxide dismutase exerts a cardioprotective
elect in ischemia and reperfusion . First, plasma CK activities, indicative of the severity of ischemic injury to the
myocardium, were significantly lower in cats treated with
taprostene and human superoxide dismutase together than in
cats receiving taprostene or human superxide dismutase
alone or their vehicle. Second, although analysis of myocardial tissue clearly indicated that all four ischemic groups
were exposed to a comparable degree of myocardial ;eopardy as a result of left anterior descending coronary artery
occlusion, a relatively large proportion of the area at risk
became necrotic in untreated cats with myocardial ischemia
and in cats given taprostene or human superoxide dismutase
alone- Infusion of taprostene add human superoxide dismutase together prevented much of the ischemic tissue from
becoming necrotic, whether calculated as a percent of the
area at risk or as a percent of the total left ventricular mass.
Third, the coronary ring data indicate that ischemia and
reperfusion produced endothelial dysfunction characterized
by a reduced vasorelaxing response to the endotheliumdependent dilators, acetylcholine and A-23187 . In contrast,
coronary rings from cats with myocardial ischemia and
combination treatment exhibited a significantly greater vastnelaxant response to acetylcholine and A-23187, indicating
that endothelial function as exemplified by producing and
203
releasing endothelium-derived relaxing factor was preserved
by combined treatment with taprostene and human superoxide dismutase. Similar findings occur in the isehemic reperfused isolated rat heart, indicating that the microvascular
endothelium responds similarly to that of large coronary
arteries (29) .
Neither human superoxide dismutase alone nor taprostene alone in the low dose used in this study sufficiently
inhibited neutrophil adherence or infiltration. However,
when these two substances were given together, a significantly decreased myeloperoxidase activity was observed, in
both the area at risk and the necrotic area . These results
suggest that administration of low dose human superoxide
dismutase can only partly scavenge superoxide radicals,
which either act as a powerful chemotactic factor or allow
the expression of chemotactic factors for neutrophil adherence and activation . However, administration of low dose
taprostene can partly inhibit neutrophil adherence, When the
two substances were used together, an enhanced effect was
obtained. Therefore, neutrophil adherence and infiltration
were prevented and the endothelial and myocardial injury
that was presumably cr" .rrbuted to by activated neutrophils
was significantly decreased.
Conclusions. In summary, occlusion of the left anterior
descending coronary artery in cats for 90 min followed by
4 .5 h of reperfusion resulted in significant endothelial dysfunction and myocardial injury . The doses of human superoxide dismutase and taprostene employed in this study were
lower than those that produced significant endothelial and
myocardial protection when the agents were administered
separately in this model of myocardial ischemia and reperfusion. However, when these agents were combined, highly
significant beneficial effects occurred, indicating an enhanced effect of these two substances 6 h after the onset of
myocardial ischemia. Although these results do not indicate
that such a protective effect Occurs at 49 h after ischemia,
when the infarction process is complete, it represents a
significant early anti-ischemic effect . Further work on a48-h
model of ischemia and reperfusion would be helpful in
assessing the clinical usefulness of these findings .
Jahamxa Sehoeider of Grwoorlml GmbH for [1K generous supply
of raproslene and human sJpNOaide dismutase used in this eatrerinem .
We ,rre„k
References
Leeches HR. My-dal mulanda, repmrfosion, mod tree radical injury .
Am J Cardiol 1990 tsuppl n:t4t-211.
2. Mehra 1L . Nichols WW . IMnnelly WH, a al . Pretnelion by supemxide
dismnxae from myacard 1 dy>fuoesan and etteuccuen of vasadilalw
reserve alto. epnxmry occlusion and reperfusion in dog . Cm, item
1989b5:12R}95.
3. WeiringRS,WeanR,RmgD.Quamimliveaspectsofeheprudu^liorlof
supmranidm radint phagocymsing human grsnulacyten . 1 Lab Can Med
1975b5:245-52.
4. Ambrosia G . &axed LC . Hutchios GM. Weisman HF, Weisfetdi ML.
Redumion in esperinumel infarct slre by «combinane hunmll sulreraxlde
1.
204
MA ET AL .
SUPEROXIDE DISMUTASE AND PROSTACYCLIN ANALOGUE IN ISCHEMIA
dismurase : insights imo parhopbysiology of repsefusiao injury . Circulation 1986 ;74:1424-33 .
5. Johnson G III, Tsad PS, Lefer AM . Synergism betweso superonide
dlsmutme and sodium nionte in cardioprouelian following iscbema and
reperfiao. Am Heart J 1990 ;119:530-7 .
6. Simpson PJ, Todd RF III, Mickelsen JK, e1 al . Sustained limimtioa of
myacmdid r<Irrfusion i jury by a murwcJan .) antibody ubac alleys
leukanyte fundinn . Circulation 1990$1 :226-37 .
7. Engler R, Gilpin E. Can slpemeide dismulae sire nyacardinl infarct
sin1 Ciselaiion 1989 ;790137-42.
R. Or- HA, MaCned JM . The rardiopraluctive eieet of Mn-supeaoaide
dismulavc in loot x high doses in the poslischemic ouied rtnhil bean .
Pree Radical Bill Men 19912,9477-8 .
9. In
C Dl, Fudan L£, Aoki N, Lefer AM . Eaddheliom and myocardial protecting actions of laprovene, a stable prosucyclin ando90e.
after acute myaardinl iselrmu and re perfusion in cats Circ Roe 1990 :
66 :1762-70.
10. I0Qhen : BR, K'ems SW, Fanlone JC .110 role of the r ootroplul andfrcc
radicals in iuhemlc m3lwardid kljue5' J Mod Cell Canliol 195931134131 .
11 . Auki N. Bitlerman M . Buzinski ME . LglOrAM . Cardioprdeclivc notion
oflsumaa superoeide disneutase in true repedustoa models o1 myacm dial
ischemia in the m1 .14 J Phannacol 1981;95:735-40 .
: a double-edged
12. Bkamrwald E, Wooer RA . Myocardial "perfusion
nwad7I Cfin IaviM 19913;76:1713-9 .
Jolly
SR
.
BajEf
MB
.
Abrams
GD . Lucahesi Bk. Camirr
13,
. Kmu WI
myocardial repeRleslon injury, its reduction by the combined adrninisualion ofsapemaide dismlnase and canlase. On Re, 1994 :54:277- in.
M . Nalund V, Haggnrark SAW- .G . Markhand ST. Rein S. Limimlion
of myocardial infarction size by superaslde dismolase a on adjunct to
icrerfuslon after diferenl dumrlon of coronary occlusion in the peg . Can
Roe 1990;66:1294-301 .
15 . Tsar P5, Aoki N, Lefer IH, Johnson G ill, Lefer AM . Tim course of
endolleliel dysfunction and myocardial injury during myocardial Wheat,
a and repedosion in the cal . Circulation 1990 :80:1402-12.
16. Rubanyl GM, Vanhoaue PM . Oeygeademai free radicals, endothelial
and responsiveness of vascular smoolh muscle . Am I Physiol 1957;250 :
H815-71 .
17. Schrdr K. Smith EF 111, 14cr AM, The cal as as in vivo u-^site! f
myocardial nichemia and iniaion. Progr Plurniecol 1955;6:31-91.
18. Ooead AG, Saudi CT, Scold MM . Oelerminauior, of serum protein
by means of lbe himel mnhod . I Bid Chem 1949;117:751-66.
19. Rasalki SB . An improved prousdmr for serum cmalim phosphokinase
determination. J Lab Ch Men 1%7169 :696-705.
20. BIeJley PP. Priebal DA, Chnstensen RD, Ruthslcir GR . Measurement of
IACC Val 19, N. I
Jan.. y 1972:197-214
cmaneous mdammmron: emnmllon annprrephll eomenl with nn enayme
maker
.I la .eel Dermalal 191279 :MK-9 .
21 . Mullane KM . Kraemer K, Smith B . Myeloperoeidase aclivily as an
-mum of oeutrophl indllralion imo isuhenuc myagrrdi :,m. J Prnr
nlaool Melh 1985 :14r1S1-61.
22 . Klnner RA. Elks SG . Lange R. OeaawaM E. SLUdies of experirsnlal
coronary artery roperfaninr Dens on iofara size, myooardinl rmetinn,
bmchcmislry. ullnslnrelurr, and microcheulalwn damage . Cirulaion
1905,5&suppl 1:1.1-18.
23. Iersaa DL . Moist JL, Ni, hill, WW . Curanary rcpsrfasion In dogs
iuhibis oeooIhclium-0epundrm relaxation: role of supermAe radkals
Free Radical Bid Mcd ,Nra i .afrfi0.
24. MaiCand SL. Rule of low elects d 91 In reperfuswo damage .
1 Mol Ca13 Curdled 1945 2t1hnppl 111x11-23-11-30 .
25. Wino SW, Lucchnei BR. N e .dieals and iocheric linear injury.
Trends Phasmacol5wncea 0 190JI :161-6.
26. 2weier IL. Knppusamy P . I ally GA. 14--m- of urcblhrid all
hoe radical generation : evWenco fa a aoand medunism of free radical
injury in poslischemic tissueo Prat Nail Acad 5ci 191&65 :4016-30.
27. McCaicha NJ, Schwappacb JR . EaquisL EG. et a1. KSnmine oclda5edeaived H .O . aontliblges to re perfusion injury of isvlulnk skelelal
morale. Am I Physiol 1990105&H1414-9.
28. Vokora 5, Hildrcrh J, Kuhada F, clot. Immunahtnw;6lldly locelination
of nanlhiue oaidase is the mirros^escularerdaheera cells o porcine and
human olgoet Iabslrl . Coo Shock i990Ot d2,
29. Tsao PS . Lefa AM . That course and mechansm sf endahelal dysNncin iadmrd ioohr .w. arA hypovnpern sod nt heads . A. J Physiol
1990159:111660-6.
30. BulkkyOB.Mnlimunufsplanchnicmganinjury:ooerviewarWpernpccturn . In : Maastan A, Bulkky GM . FWdi.-G en RG. lLgluad UH . eds .
Splanchrec Luhemia and Organ Failure. Land- Edward AmnW, 1999 :
163-91 .
31 . Idnsan U Ill.'isno PS. Malloy D, L ofcr AM . CMdmpml elave eftents of
acidified sodium nilnle in myocardial ischenua with reperfusion . I litmeal Fop That 1999;252 :35-41 .
32 . Warren JS . YabrelKR, Mandel DM . Johnson X.J. Word PA. Bole urO,
in neorrophil rxruilmlent into sites of dermal and mdnumary vauulilis,
Fret Kadkai die! Med 1990:5 :163-72
33. Ronlsen IL . Hook BG. Riga VH. SoMrk MA. 5wanson DP, Lacchest
BR. This: 0Bea of iioprufen on xcomolorion of 111-indium labeled
paelels rind leukocytes in eIferin nasal myocardial infarction . CRCUtalion 1982 :66:I1K'z-11 .
14. Lefer AM. Dorms I1 . Taprosrene (CG'4203). Corsiovac Drug R-v
1989 :139-51 .