J Mol Cell Cardiol 31, 1117–1123 (1999)

Article No. jmcc.1999.0943, available online at http://www.idealibrary.com on

Delayed Endothelial Protective Effects ofMonophosphoryl Lipid A After MyocardialIschemia and Reperfusion in RatsVincent Richard, Eric Danielou, Nathalie Kaeffer and Christian ThuillezINSERM E9920, Department of Pharmacology (VACOMED, IFRMP 23), Rouen University MedicalSchool and Rouen University Hospital, Rouen, France

(Received 15 October 1998, accepted in revised form 16 February 1999)

V. R, E. D, N. K C. T. Delayed Endothelial Protective Effects of MonophosphorylLipid A After Myocardial Ischemia and Reperfusion in Rats. Journal of Molecular and Cellular Cardiology (1999)31, 1117–1123. Monophosphoryl lipid A (MLA) induces delayed (24 h) myocardial protection in various animalmodels of ischemia/reperfusion injury, and thus mimics the second window of preconditioning against cardiacinjury. However, the potential endothelial protective effects of this drug have not been evaluated. The presentstudy was designed to assess whether MLA exerts delayed protective effects against reperfusion-induced coronaryendothelial dysfunction in rats, as well as the protective role of iNOS in this protection. Wistar rats received asingle i.v. injection of MLA (450 lg/kg) or solvent. Twenty-four hours later, they were anesthetized and subjectedto 20 min ischemia with 60 min reperfusion, in the absence or the presence of the iNOS inhibitor aminoguanidine(300 mg/kg i.p.). At the end of reperfusion, 1.5–2 mm coronary segments (average diameter 250 lm) wereremoved distal to the site of occlusion and mounted in wire myographs. Endothelium-dependent relaxations toacetylcholine were determined in arteries pre-contracted by serotonin. Ischemia/reperfusion induced a markeddecrease in the coronary responses to acetylcholine (maximal relaxations: sham 64±8%, n=8; ischemia/reperfusion: 41±9%, n=8; P<0.05). This impaired response was partially restored by MLA (55±4%, n=10;P<0.05 vs ischemia/reperfusion). The effect of MLA was not affected by aminoguanidine (57±5%, n=6). Thus,in addition to protecting myocytes, MLA induces a delayed protection against coronary endothelial dysfunction.However, in contrast to its effects on myocytes, the endothelial protective effects do not appear to involve iNOS.

1999 Academic Press

K W: Ischemia; Reperfusion; Endothelium; Preconditioning; Endotoxin; Coronary artery; Nitric oxidesynthase; Nitric oxide; Monophosphoryl lipid A.

1990; Ma et al., 1993). Given the essential roleIntroductionof the endothelium (and especially endothelium-derived NO) in the regulation of vascular tone,Vascular endothelial cells play an essential role in

the control of vascular tone, but also in the re- platelet aggregation and leukocyte adhesion, it islikely that such a persistent impairment may havegulation of smooth muscle growth, as well as of

platelet and leukocyte function. Numerous ex- important deleterious consequences on the cor-onary arterial wall. Thus, coronary endothelial cellsperimental and clinical data suggest that these

essential physiological functions of the endothelium may be considered as a major therapeutic target ofanti-ischemic treatments.are altered in various pathophysiological situations,

including myocardial ischemia followed by acute Recent experiments from our group and othershave shown that preconditioning with brief periodsor long term reperfusion (Ku, 1982; VanBenthuysen

et al., 1987; Pearson et al., 1990a, b; Tsao et al., of intermittent ischemia, in addition to limiting

Please address all correspondence to: Vincent Richard, Department of Pharmacology, Rouen University Medical School, 22 BoulevardGambetta, 76183 Rouen Cedex, France.

0022–2828/99/051117+07 $30.00/0 1999 Academic Press

V. Richard et al.1118

infarct size, also protected coronary endothelial cells animals were intubated with a small metal cannulaand mechanically ventilated with room air sup-against acute (Richard et al., 1994) or chronic

(Kaeffer et al., 1996) reperfusion injury. Recently, plemented with low flow oxygen, using a smallrodent ventilator, at a rate of 60 cycles/min andwe were also able to demonstrate for the first time

the existence of delayed endothelial protection 24 h a tidal volume of 1 ml/100 g body weight. Bodytemperature was maintained at 37°C using a ther-after pre-conditioning (Kaeffer et al., 1997). These

findings extend to endothelial cells the concept mostated heating blanket connected to a rectalthermometer. A left thoracotomy was performed,of delayed protection, already described for other

aspects of ischemia/reperfusion injury (Kuzuya et and the heart was exposed. A 6.0 polypropylenesuture was passed around the proximal left cor-al., 1993; Marber et al., 1993; Sun et al., 1995;

Richard et al., 1996). Because of these numerous onary artery, in order to induce myocardial isch-emia. Rats were subjected to 20 min of coronarybeneficial effects, it is clear that the search for

pharmacologic agents which mimic such delayed occlusion followed by 1 h reperfusion. Sham op-erated animals were subjected to the same protocol,protective effects may lead to the development of

new, potent anti-ischemic interventions. However, except that the snare was not tied.so far, very few compounds have been showed toinduce delayed protection, similar to that of pre-conditioning. In vitro vascular studies

Monophosphoryl lipid A (MLA) induces delayedmyocardial protection in various animal models of Isolated coronary arteries were studied as described

previously (Richard et al., 1994). At the end of theischemia/reperfusion injury (Yao et al., 1995;Baxter et al., 1996; Yoshida et al., 1996; Zhao reperfusion period, the heart was removed and

immediately placed in cold, oxygenated physio-et al., 1996, 1997; Elliott, 1998). However, thepotential endothelial protective effects of this drug logical saline (control solution) of the following

composition (m): NaCl 118.3, KCl 4.7, CaCl2 2.5,have not been evaluated. Thus, the present studywas designed to assess whether MLA exerts delayed MgSO4 1.2, EDTA 0.02 and glucose 11.1. The left

coronary artery was then carefully dissected freeprotective effects against reperfusion-induced cor-onary endothelial dysfunction. under a microscope. One segment of the left cor-

onary artery (length 1.5–2 mm; internal diameterSeveral mechanisms have been advanced to ex-plain the beneficial effects of MLA at the level of 250–300 lm) was taken distal to the site of oc-

clusion and mounted in a small vessel myographmyocytes. However, recent experiments dem-onstrated that the infarct size limiting of MLA could for isometric tension recording (JP Trading, Aarhus,

Denmark). Care was taken during the dissectionbe abolished by the inhibitor of inducible nitricoxide synthase (iNOS) aminoguanidine (Zhao et al., procedure to avoid damage to the endothelium.

After equilibration, segments were exposed to in-1997), suggesting that induction of NOS con-tributes to the myocardial protective effect of MLA. creasing concentrations of serotonin (10−9–10−5 M),

which is not associated in this rat coronary pre-Inducible nitric oxide synthase also contributes tothe infarct size limiting effect of late preconditioning paration with the release of endothelium-derived

relaxing factors, and thus only induces direct(Bolli et al., 1997). Thus, another goal of ourstudy was to assess whether any MLA associated smooth muscle contraction. Vessels were then

washed, and concentration-response curves toendothelial protective effects also involve iNOS.acetylcholine (10−9–10−5 M) or the NO donor so-dium nitroprusside (10−9–10−5 M) were studied ineach ring after pre-contraction by serotonin(10−5 M).Materials and Methods

In vivo induction of myocardial ischemiaExperimental protocol

Experiments were performed in male Wistar rats,weighing between 300 and 400 g. Twenty-four Experiments were performed in four groups of rats.

Rats from group 1 were subjected to sham surgery,hours after administration of MLA or solvent (seebelow), rats were anesthetized with sodium pen- and rats from groups 2–4 were subjected to isch-

emia/reperfusion. Rats from group 2 were pre-tobarbital intraperitoneally (40 mg.kg−1). A singlei.p. injection was sufficient to maintain anesthesia treated i.v. with 1 ml/kg solvent alone (ethanol

10%, propylene glycol 40% in water for injection),throughout the ischemia/reperfusion protocol. The

Endothelial Protection by Monophosphoryl Lipid A 1119

while rats from groups 3 and 4 were treated withMLA (450 lg/kg i.v.) in the same volume of solvent24 h before induction of ischemia. The dose of MLAwas selected on the basis of previous rat studies inwhich the use of such a high dose was necessaryto induce anti-ischemic effects in rats (Tosaki et al.,1998). Rats from group 4 received aminoguanidine(300 mg/kg) administered i.p. (1 h) before ischemia.

Data analysis

All results are expressed as mean±SEM. In in vitroexperiments, n refers to the number of animalsfrom which the arteries were taken. Contraction Figure 1 Contractile responses induced by increasing

concentrations of serotonin in coronary arteries isolatedto serotonin are normalized to vessel length andfrom sham rats (n=7) or rats subjected to ischemia andexpressed in mN/mm. Contractions are expressed asreperfusion, 24 h after administration of vehicle (I/R,percentage of the maximal response. Concentration- n=9), or MLA. In rats pre-treated by MLA, ischemia/

dependent relaxations to acetylcholine and SIN-1 reperfusion was performed either in the absence (MLA+are expressed as a percentage of the contractions I/R, n=10), or the presence of aminoguanidine (MLA+

I/R+AMG, n=7). Values are mean±SEM.to serotonin. In addition, the negative logarithm ofeither the IC50 (in case of relaxations) or EC50 (incase of contractions) was calculated from con-

Figure 2, while the EC50 are shown in Table 1. Incentration-response curves after adjustment to aall groups, SIN-1 induced concentration-dependentsigmoidal curve by a curve-fitting software (Origin,relaxations that reached 100% at high con-MicroCal Software, Inc., Northampton, MA, USA)centration. I/R or treatment by MLA did not affectand the mean±SEM of these values are presented.the responses to the NO donor.Contractile or relaxing responses were compared

by one-way ANOVA followed, when ANOVA wassignificant, by a Tukey test for multiple com-

Relaxing responses to acetylcholineparisons. A value of P<0.05 was considered stat-istically significant.

In coronary arteries isolated from sham operatedanimals, acetycholine induced concentration-de-pendent relaxations that reached 64±8% at theResultshighest dose (Fig. 3).

Normalized vessels diameters and contractile responsesto serotonin Effects of ischemia and reperfusion without MLA

(vehicle group)The normalized internal diameters were

Compared with sham operated animals, the re-284±13 lm in the sham group, 249±14 lm insponse to acetylcholine was markedly reduced inthe vehicle group, 255±12 lm in the MLA grouparteries taken from animals subjected to ischemiaand 270±15 lm in the MLA+AMG group. Thereand reperfusion (maximal response: 41±9%;were no significant differences between the groups.P<0.05).The responses to increasing concentrations of

serotonin are shown in Figure 1. No significantdifferences at each concentration of serotonin were Effects of ischemia and reperfusion after treatment byobserved between the treatment groups. EC50 also MLAdid not differ significantly (Table 1).

The impaired response to acetylcholine after I/Rwas significantly improved by MLA administered24 h before ischemia (maximal response: vehicle,Relaxing responses to SIN-141±9%, MLA 55±4%; P<0.05).

The improvement of the response to acetylcholineThe relaxing responses induced by increasing con-centrations of the NO donor SIN-1 are shown in by MLA was not modified by the administration of

V. Richard et al.1120

Table 1 Normalized internal diameters, EC50 for serotonin and IC50 for SIN-1 in isolated coronary arteries from thefour groups. Numbers in brackets represent the number of animals from which the arteries were studied

Sham (7) Vehicle+I/R (9) MLA+I/R (10) MLA+I/R+AMG (7)

ID (lm) 284±13 249±14 255±12 270±15EC50 serotonin (lm) 0.65±0.19 0.47±0.09 0.40±0.10 0.38±0.09IC50 SIN-1 (lm) 0.31±0.07 0.32±0.09 0.33±0.07 0.19±0.05

Figure 2 Relaxing responses induced by increasing con- Figure 3 Relaxing responses induced by increasing con-centrations of the NO donor SIN-1 in coronary arteries centrations of acetylcholine in coronary arteries isolatedisolated from sham rats (n=7) or rats subjected to isch- from sham rats (n=7) or rats subjected to ischemia andemia and reperfusion, 24 h after administration of vehicle reperfusion, 24 h after administration of vehicle (I/R,(I/R, n=9), or MLA. In rats pre-treated by MLA, ischemia/ n=9), or MLA. In rats pre-treated by MLA, ischemia/reperfusion was performed either in the absence (MLA+ reperfusion was performed either in the absence (MLA+I/R, n=10), or the presence of aminoguanidine (MLA+ I/R, n=10), or the presence of aminoguanidine (MLA+I/R+AMG, n=7). Arteries were pre-contracted by sero- I/R+AMG, n=7). Arteries were pre-contracted bytonin. serotonin. Values are mean±SEM. ∗: P<0.05 v sham;

†: P<0.05 v I/R.

the iNOS inhibitor aminoguanidine 1 h before I/R(maximal response, MLA, 55±4%; MLA+AMG,

affect the responsiveness of the coronary smooth57±5; NS).muscle either in terms of contraction or relaxation.

Endothelial-dependent relaxations to acetyl-choline were markedly reduced after ischemia andDiscussionreperfusion, in agreement with previous results byus (Richard et al., 1994; Kaeffer et al., 1996, 1997)The major findings of our study, performed in a ratand others (Ku, 1982; VanBenthuysen et al., 1987;model of myocardial ischemia and reperfusion, werePearson et al., 1990a, b; Tsao et al., 1990; Ma etthat ischemia and reperfusion markedly decreasedal., 1993). The mechanisms of such dysfunctionthe response to acetylcholine and that this im-appear to involve expression of oxygen-derived freepairment was partly reversed by MLA (450 lg/kg),radicals (Mehta et al., 1989; Tsao et al., 1990; Grossgiven 24 h before prolonged ischemia. The delayedet al., 1992; Kaeffer et al., 1997) and adhesion ofprotective effect of MLA against reperfusion-inducedneutrophils to endothelial cells (Ma et al., 1995)endothelial injury was not affected by the selectiveupon myocardial reperfusion (Ma et al., 1992;iNOS inhibitor aminoguanidine.Weyrich et al., 1993).In agreement with our previous results, we con-

In this setting, we found that MLA, given 24 hfirmed that ischemia and reperfusion did not affectbefore ischemia, significantly increased the vaso-the responses to serotonin (which does not inducedilatory response to acetylcholine post-ischemia-endothelium-dependent relaxations in rat coronaryreperfusion. Thus, these experiments demonstratearteries) or to the endothelium-independent vaso-

dilator, NO donor SIN-1, suggesting that it did not for the first time that, in addition to protecting

Endothelial Protection by Monophosphoryl Lipid A 1121

myocardial cells, MLA also protects coronary endo- effect of MLA, suggesting that prevention of endo-thelial dysfunction does not involve inducible iNOS.thelial cells against ischemia and reperfusion injury.

In this regard, the protective effect of MLA resembles Yet, the results of our study do not preclude a roleof eNOS in the protection, although this hypothesisthat previously described with ischemic pre-

conditioning, which induces a similar delayed is difficult to test in our experimental conditions,in which in vivo blockade of eNOS leads to a marked(24 h) improvement of endothelial function after

ischemia and reperfusion (Kaeffer et al., 1997). inhibition of endothelium-dependent relaxations invitro in all experimental groups. Finally, it mustThus, to the best of our knowledge, this is the first

description of a substance capable of phar- also be noted that aminoguanidine is not fullyspecific to iNOS and may affect other enzymaticmacologically mimicking the second window of

preconditioning against endothelial injury. systems. For example, aminoguanidine has beenshown to also inhibit catalase (Ou and Wolff, 1993).Several mechanisms could contribute to the endo-

thelial protective effect of MLA. Indeed, as mentioned Thus, direct proof that iNOS is not involved in theendothelial effects of MLA would require the use ofabove, there is evidence that adhesion of neutrophils

do contribute to reperfusion injury of endothelial cells. more specific inhibitors of this enzyme.Given the fact that oxidant stress is a majorAlthough we have not assessed neutrophil adhesion

in the present study, there is evidence that MLA contributor of endothelial injury during reperfusion,another possibility is that MLA protects endothelialindeed limits myocardial neutrophil infiltration after

ischemia/reperfusion (Yao et al., 1993; Zhao et al., cells through an increased expression and/or ac-tivity of antioxidant enzymes. Again, several ex-1997). Thus, possibly, the endothelial protective ef-

fects of MLA may be due at least in part to its ability periments have suggested that the protective effectof delayed pre-conditioning may be mediated byto reduce neutrophil adhesion.

Based on the known possible mechanisms of pre- increased antioxidant defenses (Yamashita et al.,1994; Zhou et al., 1996). Indeed, MLA induces aconditioning, another possibility is that some of the

effects of MLA could be due to the expression of heat delayed (24 h) increase in the antioxidant enzymecatalase (Yao et al., 1993).shock proteins (HSP), which might be important

mediators of pre-conditioning (Marber et al., 1993, In conclusion, our experiments demonstrate thatMLA induces a marked delayed protective effect1995). Indeed, cultured rat myocytes have been

shown to express HSP72 24 h after treatment by against coronary endothelial injury induced by isch-emia and reperfusion in rats. In view of the knownMLA (Nayeem et al., 1997). However, in rabbit

models of ischemia and reperfusion, pre-treatment role of the endothelium in the regulation of cor-onary tone, as well as on platelet and leukocytewith MLA at a cardioprotective dose did not increase

the expression of the inducible 70 heat shock pro- function, it is likely that impairment of endothelialfunction after reperfusion could favour smooth muscletein (Baxter et al., 1996; Yoshida et al., 1996).

Thus, whether HSPs contribute to the protective contraction as well as platelet and leukocytes ad-hesion, thus increasing the risk of subsequent va-effects of MLA is still unclear. Moreover, although

we have shown that heat stress increases NO— sospasm, thrombosis, and atherosclerosis. In thiscontext, the endothelial protection afforded by MLAdependent relaxations in isolated rat arteries

(Richard et al., 1997) and prevents coronary endo- could be associated with a long term protection ofthe coronary vascular wall against these adversethelial dysfunction after ischemia and reperfusion

(Kaeffer et al., 1995), whether HSP expression in consequences of reperfusion, which would add tothe beneficial effect of this compound on ischemicitself may preserve endothelial cell function in this

setting is still unknown. myocytes.Recent experiments demonstrated that the infarct

size limiting effect of MLA could be abolished byAcknowledgementsaminoguanidine (Zhao et al., 1997). In addition,

iNOS mRNA was induced by MLA in both the ratThis work was supported in part by RIBI Im-(Tosaki et al., 1998) and the pig (Tetsuya et al.,munochem. We thank Dr Elliott and Dr Weber for1997), with the cardioprotective effect in the ratsupplying MLA.blocked by L-NAME. This suggests that induction

of NOS contributes to the myocardial protectiveeffect of MLA. iNOS also contributes to the infarct Referencessize limiting effect of late pre-conditioning (Bolli etal., 1997). However, in our experiments, amino- B GF, G RW, W MJ, K M, H

RJ, Y DM, 1996. Myocardial protection afterguanidine did not affect the endothelial protective

V. Richard et al.1122

monophosphoryl lipid A: studies of delayed anti-isch- adult rat cardiac myocytes with induction of the 72-kDaemic properties in rabbit heart. Br J Pharmacol 117: heat shock protein: a cellular model of pharmacologic1685–1692. preconditioning. J Mol Cell Cardiol 29: 2305–2310.

B R, M S, T XL, T H, Q Y, O P, W SP, 1993. Aminoguanidine: a drug proposedG Y, Z Q, J AK, 1997. The protective effect for prophylaxis in diabetes inhibits catalase and gen-of late preconditioning against myocardial stunning in erates hydrogen peroxide in vitro. Biochem Pharmacolconscious rabbits is mediated by nitric oxide synthase. 46: 1139–1144.Evidence that nitric oxide acts both as a trigger and P PJ, S HV, VH PM, 1990a. Acuteas a mediator of the late phase of ischemic pre- impairment of endothelium-dependent relaxations toconditioning. Circ Res 81: 1094–1107. aggregating platelets following reperfusion injury in

E GT, 1998. Monophosphoryl lipid A induces de- canine coronary arteries. Circ Res 67: 385–393.layed preconditioning against cardiac ischemia-re- P PJ, S HV, VH PM, 1990b. Long-perfusion injury. J Mol Cell Cardiol 30: 3–17. term impairment of endothelium-dependent re-

G GJ, O’R ST, P LR, 1992. Myocardial and laxations to aggregating platelets after reperfusion in-endothelial dysfunction after multiple, brief coronary jury in canine coronary arteries. Circulation 81:occlusions: role of oxygen radicals. Am J Physiol 263: 1921–1927.H1703–H1709. R V, H JP, A C, T C, 1997.

K N, R V, H JP, T C, 1996. Prior heat stress increases NO-dependent relaxationsPreconditioning prevents chronic reperfusion— in isolated rat arteries. Circulation 96: I–349 (Abstract).induced coronary endothelial dysfunction in rats. Am R V, K N, T C, 1996. Delayed pro-J Physiol 271: H842–H849. tection of the ischemic heart—from pathophysiology

K N, R V, T C, 1995. Heat stress to therapeutic applications. Fundam Clin Pharmacol 10:protects coronary against coronary endothelial dys- 409–415.function after myocardial ischemia and reperfusion in R V, K N, T C, T C, 1994. Isch-rats. Circulation 92: I–653 (Abstract). emic preconditioning protects against coronary endo-

K N, R V, T C, 1997. Delayed cor- thelial dysfunction induced by ischemia andonary endothelial protection 24 hours after pre- reperfusion. Circulation 89: 1254–1261.conditioning. Role of free radicals. Circulation 96: S JZ, T XL, K AA, P SW, Q Y, B2311–2316. R, 1995. Late preconditioning against myocardial

K DD,1982. Coronary vascular reactivity after acute stunning. An endogenous protective mechanism thatmyocardial infarction. Science 218: 576–578. confers resistance to postischemic dysfunction 24 h

K T, H S, Y N, F H, O H, after brief ischemia in conscious pigs. J Clin Invest 95:H M, K T, T M, 1993. Delayed effects of 388–403.sublethal ischemia on the acquisition of tolerance to T Y, E RM, R JA, F JE, Bischemia. Circ Res 72: 1293–1299. M, E GT, M N, D DK, 1997. InductionM XL, L DJ, L AM, R R, 1992. Coronary

of iNOS expression by monophosphoryl lipid A: aendothelial and cardiac protective effects of a mono-pharmacological approach of ischemic preconditioningclonal antibody to intercellular adhesion molecule-1of swine hearts undergoing open heart surgery. Cir-in myocardial ischemia and reperfusion. Circulation 86:culation 96: I–620.937–946.

T A, M N, E GT, B IE, EM XL, T PS, L AM, 1995. Antibody to CD-18RM, D DK, 1998. Preconditioning of rat heart withexerts endothelial and cardiac protective effects inmonophosphoryl lipid A: a role for nitric oxide. J Pharmmyocardial ischemia and reperfusion. J Clin Invest 88:Exp Ther 285: 1274–1279.1237–1243.

T PS, A N, L DJ, J III G, L AM,M XL, W AS, L DJ, L AM, 1993. Dimin-1990. Time course of myocardial injury and endo-ished basal nitric oxide release after myocardial isch-thelial dysfunction during myocardial ischemia andemia and reperfusion promotes neutrophil adherencereperfusion in the cat. Circulation 82: 1402–1412.to coronary endothelium. Circ Res 72: 403–412.

VB KM, MM IF, H LD, 1987.M MS, L DS, W JM, Y DM,Reperfusion after coronary occlusion in dogs impairs1993. Cardiac stress protein elevation 24 hours afterendothelium-dependent relaxation to acetylcholinebrief ischemia or heat stress is associated with re-and augments contractile reactivity in vitro. J Clin Investsistance to myocardial infarction. Circulation 88: 1264–79: 265–274.1272.

W AS, M XY, L DJ, A KH, L AM,M MS, M R, C S-H, S R, Y DM,1993. In vivo neutralization of P-selectin protects felineD WH, 1995. Overexpression of the rat in-heart and endothelium in myocardial ischemia andducible 70-kD heat stress protein in a transgenic mousereperfusion injury. J Clin Invest 91: 2620–2629.increases the resistance of the heart to ischemic injury.

Y N, N M, H S, K T, HJ Clin Invest 95: 1446–1456.M, T N, K T, T M. Induction ofM JL, N WW, D WH, L DL,manganese superoxide dismutase in rat cardiac myo-T L, T R M, S TGP, 1989. Pro-cytes increases tolerance to hypoxia 24 hours aftertection by superoxide dismutase from myocardial dys-preconditioning, 1994. J Clin Invest 94: 2193–2199.function and attenuation of vasodilator reserve after

Y Z, A JA, P GM, G GJ, 1993.coronary occlusion and reperfusion in dog. Circ ResCardioprotective effects of monophosphoryl lipid A, a65: 1283–1295.novel endotoxin analogue, in the dog. Cardiovasc ResN MA, E GT, S MR, H-H SL,

K RC, 1997. Monophosphoryl lipid A protects 27: 832–838.

Endothelial Protection by Monophosphoryl Lipid A 1123

Y Z, E GT, G GJ, 1995. Monophosphoryl lipid following protracted ischemia/reperfusion using mon-ophosphoryl lipid A (MLA). J Mol Cell Cardiol 28:A preserves myocardial contractile function following

multiple, brief periods of coronary occlusion in dogs. 197–208.Z L, W PA, S JR, C ML, EPharmacology 51: 152–159.

Y KI, M MM, S JB, D M, G, 1997. Role of inducible nitric oxide synthase inpharmacological “preconditioning” with mono-B NL, Q YZ, E GT, K RC,

1996. Monophosphoryl lipid A induces pharmacologic phosphoryl lipid A. J Mol Cell Cardiol 29: 1567–1576.Z X, Z X, A M, 1996. Direct evidence that“preconditioning” in rabbit hearts without con-

comitant expression of 70-kDa heat shock protein. Mol initial oxidative stress triggered by preconditioningcontributes to second window of protection by endo-Cell Biochem 156: 1–8.

Z L, K CC, H SR, E GT, 1996. Pre- genous antioxidant enzyme in myocytes. Circulation93: 1177–1184.servation of global cardiac function in the rabbit