T and myocardial perfusion by producing vasoactive

Transcription

1 Production of Endothelium-Derived Contracting Factor Is Enhanced After Coronary Reperfusion Paul J. Pearson, MD, PhD, Pyng J. Lin, MD, and Hartzell V. Schaff, MD Section of Cardiovascular Surgery, Mayo Clinic and Mayo Foundation, Rochester, Minnesota To determine whether coronary reperfusion enhances the production of endothelium-derived contracting factor, we investigated dogs subjected to global cardiac ischemia (45 minutes) followed by reperfusion (60 minutes). Segments of reperfused and control coronary arteries were suspended in organ chambers to measure isometric force. Perfusate hypoxia caused endothelium-dependent contraction in the control and reperfused arteries. However, reperfused arteries exhibited hypoxic contraction that was significantly greater than control segments. The hypoxic contractions in both the control and reperfused arteries could be inhibited by p-monomethyl-l-arginine (L-NMMA), the blocker of endothelial cell synthe- sis of nitric oxide from L-arginine. The action of 1,-NMMA could be reversed by L-arginine but not D-arginine. Thus, after reperfusion, augmented production of endothelium-derived contracting factor occurs by an L-arginine-dependent pathway. We hypothesize that nitric oxide produced by L-arginine metabolism combines with superoxide anion to produce the peroxynitrite anion r:onoo-), which is metabolized to endothelium-derived contracting factor or induces its synthesis. Augmented production of endothelium-derived contracting factor would favor vasospasm after reperfusion. (Ann Thoruc Surg 2992;52:788-93) he coronary endothelium modulates vascular tone T and myocardial perfusion by producing vasoactive factors that act on the underlying vascular smooth muscle [I]. However, after global cardiac ischemia and reperfusion, the release of endothelium-derived relaxing factor to aggregating platelets and platelet products becomes impaired [2, 31. Impaired release of relaxing factor would promote platelet adhesion and aggregation as well as platelet-induced constriction of the coronary smooth muscle; these events would be expected to increase risk of subsequent ischemia due to vasospasm and thrombosis [4]. Indeed, impaired release of endothelium-dependent relaxing factor, in particular to aggregating platelets, appears to be an important cause of coronary vasospasm after cardiac operations. The coronary endothelium can also produce contracting factor (EDCF). When exposed to hypoxia, coronary segments with endothelium exhibit augmented contraction to agonists such as serotonin, prostaglandin Fza, and potassium ions [5, 61. Endothelium-dependent contractions caused by hypoxia have also been described in the human internal mammary artery [7]. It is possible that coronary reperfusion could enhance the production of EDCF. Increased production of EDCF after global myocardial reperfusion as occurs during cardiopulmonary bypass might lead to coronary vasospasm after cardiac operations. The present study was designed to determine whether reperfusion after cardiopulmonary bypass alters the production of EDCF. Accepted for publication Jan 15, Address reprint requests to Dr Schaff, Section of Cardiovascular Surgery, Mayo Clinic, 200 First St, SW, Rochester, MN Material and Methods Animal Preparation Heartworm-free mongrel dogs (25 to 30 kg) of either sex were anesthetized with intravenous pentobarbital sodium (30 mg/kg bolus injection; Fort Dodge Laboratories, Fort Dodge, IA), intubated with a cuffed endotracheal tube, and ventilated with 95% oxygen. The left common carotid artery was isolated and catheterized to monitor aortic blood pressure. A left lateral thoracotomy was performed in the fourth intercostal space to expose the heart, and a pericardial cradle was created. After heparinization (2 mg/kg), the right atrium was cannulated through its appendage for venous return, and the right femoral artery was isolated and cannulated for arterial inflow. Cardiopulmonary bypass was then commenced at 2.0 to 2.4 L min- * m- (32 C). When perfusion flow and pressure were stable, the ascending aorta was cross-clamped. A left ventricular vent was then inserted through the left atrial appendage to decompress the heart during ischemia and reperfusion. After 45 minutes of normothermic arrest, the aortic cross-clamp was removed, and the heart was allowed to reperfuse. After 5 minutes of reperfusion, the heart was defibrillated, and cardiopulmonary bypass was slowly discontinued. The heart was then allowed to function for a total of 60 minutes of reperfusion, at which time the animal was exsanguinated into the cardiotomy reservoir, and the still-beating heart was excised and immersed in cold, oxygenated physiological salt solution of the following composition (mmol/l): NaCI, 118.3; KCI, 4.7; MgSO,, 1.2; KHJ O,, 1.22; CaCl,, 2.5; NaHCO,,, 25.0, and glucose, 11.1 (control solution). An additional dog was anesthetized (sodium pentobarbital, 30 mg/kg intravenous injection) and exsanguinated, and the beating by The Society of Thoracic Surgeons /91/$3.50

2 Ann Thorac Surg 1991;51:78693 PEARSON ET AL 789 heart excised and placed in cold, oxygenated control solution. Preliminary experiments demonstrated no difference in the hypoxic response between exsanguinated control dogs and dogs exposed to cardiopulmonary bypass without ischemia (data not shown). All animals received humane care in compliance with the Guide for the Care and Use of Laboratory Animals published by the National Institutes of Health (NIH publication No , revised 1985). The procedures and the handling of the animals were reviewed and approved by the Institutional Animal Care and Use Committee of the Mayo Foundation. In Vitro Experiments The left circumflex arteries of reperfused and unmanipulated control hearts were carefully dissected free and placed in control solution. Care was taken to remove as much of the connective tissue as possible. A maximum of four coronary ring segments (5 to 6 mm in length) were prepared from each artery. In some rings, the endothelium was removed by gently rubbing the intimal surface with the tip of a pair of watchmaker s forceps [8, 91. Previous histological studies have shown that this procedure removes the endothelial cells in large canine arteries and preserves the ability of the vascular smooth muscle to contract and relax [lo, 111. The rings were suspended in 25-mL organ chambers filled with control solution maintained at 37 C and aerated with 95% 0,-5% CO, (ph = 7.4) (Fig 1). Each ring was suspended by two stainless steel clips passed through the lumen. One clip was anchored to the bottom of the organ chamber; the other was connected to a strain gauge (Grass FT03; Grass Instrument Co, Quincy, MA) for the measurement of isometric force. Rings were placed at the optimal point of their length-tension relationship by progressively stretching them until the contraction to KCl(20 mmovl), imposed at each level of distention, was maximal [8, 91. In all experiments, the presence or absence of endothelium was confirmed by the response to acetylcholine movl) of rings contracted with potassium ions (20 mmovl) [lo, 121. After optimal tension was achieved, the rings were allowed to equilibrate for 45 minutes before the administration of drugs. Protocols Rings of reperfused and control arteries (with and without endothelium) were studied in parallel. Before exposure to hypoxia, rings were contracted with prostaglandin F, so that the vessel could either relax or exhibit additional contraction to hypoxia [5, 131. Previous experiments have shown that control and reperfused canine coronary arteries exhibit the same contractile response to prostaglandin F, [2]. Reperfused coronary segments also exhibit normal active basal tone as evidenced by relaxations to nitric oxide and isoproterenol [2]. Hypoxia was induced by aerating the organ bath with a mixture of 95% N,-5% CO, (ph = 7.4, oxygen tension = 30 to 40 mm Hg, 20 min). Rings were only exposed to one period of hypoxia. When used to modify the hypoxic response, drugs were added Blood vessel ring (-> Recorder JI Force transducer 95% 02-95% Np- 5% coq 5% cop water bath Fig 1. Organ chamber for the in vitro study of isolated blood vessels. to the organ chamber at least 10 minutes before the addition of the prostaglandin F2a. Drugs The following drugs were used: acetylcholine chloride (Sigma Chemical Co, St. Louis, MO), KCl (Sigma), D-arginine (Calbiochem, San Diego, CA), L-arginine (Calbiochem), fl-monomethyl-l-arginine acetate (Calbiochem), and prostaglandin F,, (Sigma). The concentrations are expressed as final molar concentration in the organ chamber. Data Analysis The results are expressed as mean f standard error of the mean. In all experiments, n refers to the number of animals from which rings were taken. In rings contracted with prostaglandin F, responses are expressed as percent change from the contracted levels. Statistical evaluation of the data was performed by Student s t test for either paired or unpaired observations. Values were considered to be statistically significant when p was smaller than Results Control and reperfused coronary segments with and without endothelium exhibited comparable contractile response to prostaglandin F,, (2 x lof6 molll): 6.05 f

3 790 PEARSON ET AL Ann Thorac Surg 1991;51: lium-dependent responses to hypoxia. Dogs were sub- 95% N2-5% co2 95% N2-5% Cop Fig 2. Isometric tension recording of the effect of global myocardial ischemia and reperfusion on endothe- Wlth endothellurn Wlthout endothellurn 7-b- jetted to global myocardial ischemia (45 minutes) followed by reperfusion (60 minutes). Segments from Control LCX control and reperfused arteries were suspended in orgun chambers and contracted with prostaglandin F,, I 4 1 I (PGF,,,,pha)(2 x rnmo/ll). When the contractile by response changing to PGF,, to a 95% was N2-5%C0, stable, hypoxia gas was mix. induced (LCX = Repertused LCX /eft circumflex coronary artery.) - 1 &4g& L J 2 L - J 4 4 PGF2alpha, 2x10-6 M - 5 mln 0.82 and g for control and reperfused arteries with endothelium, and 6.47 k 0.83 and 6.70 * 0.83 g for control and reperfused arteries without endothelium (n = 6). In contracted control and reperfused coronary segments, hypoxia caused an initial decrease in tension in segments with and without endothelium (hypoxic inhibition) (Figs 2, 3); this transient decrease in tension was comparable in the control and reperfused arteries. Hypoxia then induced contraction in control and reperfused segments with endothelium (hypoxic potentiation) (Figs 2, 3). The hypoxic contraction of the reperfused coronary segments was significantly greater than that of control coronary segments (168.0% k 20.1% versus 117.7% * 17.2% of initial prehypoxic tension, respectively; n = 6; p < 0.05). Reoxygenation produced comparable transient relaxations in control and reperfused segments with and without endothelium (posthypoxic inhibition). Coronary segments then gradually returned to their original prehypoxia tension (posthypoxic recovery). Hypoxic inhibition Hypoxic contraction Fig 3. Effect of hypoxia (95% N,-5% COz) on tension in contracted (prostaglandin FZa, 2 x M) control and reperfused canine coronary segments with endothelium. Data are shown as mean 2 standard error of the mean (n = 6) and expressed as percent of the initial contraction to prostaglandin. (Hypoxic inhibition = initial transient dilation exhibited by the coronary segment when exposed to hypoxia; hypoxic contraction = subsequent endothelium-dependent constriction exhibited by the coronary artery segment when exposed to hypoxia; asterisk denotes significant difference from control segments with endothelium [p < ) * I Pretreatment of control and reperfused coronary segments with fl-monomethyl-l-arginine (L-NMMA) mol/l) inhibited the hypoxic contractions in the coronary segments with endothelium but had no effect on segments without endothelium (n = 5; p < 0.05) (Figs 4, 5). The effect of L-NMMA could be blocked with L-arginine (low4 movl) but not by D-arginine mol/l) (Figs 4,5). L-arginine and D-arginine had no effect on the hypoxic response of control and reperfused segments without endothelium. Comment The major finding of this study is that early after global ischemia and reperfusion, hypoxia induces augmented endothelium-dependent contraction in the epicardial coronary artery. In addition, we determined that hypoxic contraction is mediated by an L-arginine-dependent pathway. Release of endothelium-derived relaxing factor (EDRF) in the coronary artery, particularly the release induced by aggregating platelets, is impaired early after global myocardial ischemia and reperfusion during cardiopulmonary bypass [2], and early and late after regional myocardial ischemia [8, 91. Because EDRF induces vasodilation [12], inhibits platelet adhesion [14] and aggregation [15, 161, and promotes platelet disaggregation [16], such an impairment would put the vessel at risk for ischemic events due to thrombosis or vasospasm. The present study suggests that if an ischemic event were to occur after cardiopulmonary bypass, tissue hypoxia would increase production of EDCF, thereby inducing or exacerbating coronary vasospasm. Indeed, such a sequence of events could lead to cardiovascular collapse at a time when the heart is exquisitely sensitive to ischemia because of preexisting myocardial disease or intraoperative injury. The identity of the EDCF released in response to hypoxia has been elusive. The factor can diffuse [5], but has not been quantified in bioassay [17]. Previous studies have excluded catecholamines, serotonin, histamine, adenine nucleotides, and cyclooxygenase products as the contracting factor(s) induced by hypoxia (5, 13, 181. In addition, the rapid onset and reversal of the contraction, coupled with the fact that hypoxic contraction can be

4 Ann Thorac Surg 1991;51:78&93 PEARSON ET AL 791 With endothelium Without endot helium presence of: I I I! Fig 4. lsometric tension recordings of the effect of NG-monomethyl-L-arginine (L-NMMA) on endothelium-dependent hypoxic contraction in reperfused canine coronay segments. L-NMMA (I 0-5 rnolll) was added to the organ bath 10 minutes before contraction with prostaglandin F,, fpgf, nlyhn). When L-arginine molll) and D-arginine molll) were used, they were added to the organ bath 10 minutes before the addition of L-NMMA. f i, J j D-arginine I I 29 j" + - L-NMMA I I J L PGF2alpha PGF2alpha 5 min blocked by calcium-channel blockers, make the vasoconstrictive peptide endothelin an unlikely candidate [19]. The present study demonstrates that endotheliumdependent hypoxic contraction can be inhibited by L-NMMA, a blocker of nitric oxide synthesis [20]. Nitric oxide has been identified as the active component of EDRF [21, 221, which is produced from L-arginine metabolism [23]. Martin [24] postulated that endotheliumdependent contractions to hypoxia might be due to a "down-regulation" of basally released EDRF. A basal level of EDRF is constantly being produced by many blood vessels [25]; impairment in EDRF production would remove the tonic relaxant effect on the vascular smooth muscle and, in effect, cause contraction. This hypothesis is supported by studies that demonstrate that hypoxia impairs the stimulated release of EDRF [ll, 261. With hypoxia, however, such a mechanism is implausible because inhibitors of EDRF do not modify endotheliumdependent contractions [5]. Although nitric oxide relaxes vascular smooth muscle, it also can combine with superoxide anion (0;) to form the peroxynitrite anion (ONOOJ [27]. ONOO- is very toxic and unstable, and it can decompose into highly reactive agents [28]. We hypothesize that 0,' produced in the hypoxic milieu of the endothelial cell [29] combines with nitric oxide to form ONOO'. ONOO- is then metabolized to EDCF or induces its synthesis (Fig 6). This hypothesis is supported by the finding that hypoxic contractions are augmented in arteries after reperfusion injury; such tissue is rich in oxygen-derived free radicals [30]. As free radicals also scavenge EDRF [31, 321 and inhibit endotheliumdependent vasodilation [33], free radical production by loo[ 80 Control Reperf used a= untreated.n=6 In the presence of: Fig 5. Effect of hypoxia on tension in contracted control and reperfused arteries with endotheliunz. Data are shown as mean? standard error of the mean and exvressed as vercent of = L-NMMA, 10-5,n=5 the initial contraction to prostaglandin F2-. fl-nmma = Nc-monomethyl-L-arginine.) = L-NMMA + L-arginine, 10-4 M,

5 792 PEARSON ET AL Ann Thorac Surg 1991;51: Fig 6. Authors' hypothesis of the mechanism of endothelium-dependent hypoxic contraction. L-arginine is utilized for endothelial cell production of nitric oxide (NO). During hypoxia, superoxide anion (0;) formed in the hypoxic milieu of the endothelial cell combines with NO to form the peroxynitrite anion (ON003. ONOO- is then metabolized to endothelium-derived contracting factor or induces its synthesis. The production of endothelium-derived contracting factor is enhanced after reperfusion injury secondary to an abundance of free radicals that can be utilized for ONOO- production. (ADP = adenosine diphosphate; ATP = adenosine triphosphate.) Hypoxia ischemia- Nitric oxide Metabolites I ++ RELAXATION 1 - the reperfused endothelium would greatly increase vascular tone. Coronary vasospasm appears to be an important but unrecognized cause of cardiovascular collapse after cardiac operations. Indeed, after global cardiac ischemia and reperfusion, coronary arteries with endothelium lose the ability to break the direct vasoconstrictive effect of aggregating platelets on the vascular smooth muscle [2]. Intracoronary platelet deposition that occurs after cardiopulmonary bypass [34] would favor increased vascular tone in the coronary circulation and might lead to myocardial ischemia. As the present study demonstrates, local tissue hypoxia caused by myocardial ischemia could augment production of EDCF and favor maintenance or exacerbation of coronary vasospasm. It is also possible that systemic hypoxemia could induce vasoconstriction in the reperfused coronary circulation after cardiopulmonary bypass. Indeed, hypoxemia can occur after cardiac operations for a variety of reasons including hypoventilation, intrapulmonary shunting of blood, and during states of low cardiac output. The present study demonstrates that if hypoxemia were to occur, potentiated endothelium-dependent contraction of the reperfused coronary artery could decrease coronary blood flow and impair perfusion to the myocardium. Coupled with the direct depressed effect of hypoxemia on myocyte function, such a situation could rapidly lead to cardiovascular collapse. References 1. Vanhoutte PM. The endothelium-modulator of vascular smooth-muscle tone. N Engl J Med 1988;319: Pearson PJ, Lin PJ, Schaff HV. Global myocardial ischemia and reperfusion impairs endothelium-dependent relaxations to aggregating platelets in the canine coronary artery: a possible etiology of vasospasm following cardiopulmonary bypass. J Thorac Cardiovasc Surg (in press). 3. Hashimoto K, Pearson PJ, Schaff HV, Cartier R. Endothelial cell dysfunction after ischemic arrest and reperfusion: a possible mechanism of myocardial injury during reflow. J Thorac Cardiovasc Surg (in press). 4. Vanhoutte PM, Shimokawa H. Endothelium-derived relaxing factor and coronary vasospasm. Circulation 1989;80: Rubanyi GM, Vanhoutte PM. Hypoxia releases a vasoconstrictor substance from the canine vascular endothelium. J Physiol 1985;364: Iqbal A, Vanhoutte PM. Flunarizine inhibits endotheliumdependent hypoxic facilitation in canine coronary arteries through an action on vascular smooth muscle. Br J Pharmacol 1988;95: Lin PJ, Pearson PJ, Schaff HV. Endothelium-dependent contractions to hypoxia in the human internal mammary artery. Surg Forum 1990;41: Pearson PJ, Schaff HV, Vanhoutte PM. Acute impairment of endothelium-dependent relaxations to aggregating platelets following reperfusion injury in canine coronary arteries. Circ Res 1990;67: Pearson PJ, Schaff HV, Vanhoutte I'M. Long-term impairment of endothelium-dependent relaxations to aggregating platelets after reperfusion injury in canine coronary arteries. Circulation 1990;81: DeMey JG, Vanhoutte PM. Role of the intima in cholinergic and purinergic relaxation of isolated femoral arteries. J Physiol 1981;316: DeMey JG, Vanhoutte PM. Heterogeneous behavior of the canine arterial and venous wall: importance of the endothelium. Circ Res 1982;51: Furchgott RF, Zawadski JV. The obligatory role of endothelial cells in the relaxation of arterial smooth muscle by acetylcholine. Nature 1980;288: DeMey JG, Vanhoutte PM. Anoxia and endotheliumdependent reactivity of the canine femoral artery. J Physiol 1983;335: Radomski MW, Palmer RMJ, Moncada S. The role of nitric oxide and cyclic GMP in platelet adhesion to vascular endothelium. Biochem Biophys Res Commun 1987;148: Furlong B, Henderson AH, Lewis MJ, Smith JA. Endothelium-

6 Ann Thorac Surg 1991;51: PEARSON ET AL 793 derived relaxing factor inhibits in vitro platelet aggregation. Br J Pharrnacol1987;90: Radomski MW, Palmer RMJ, Moncada S. Comparative pharmacology of endothelium-derived relaxing factor, nitric oxide and prostacyclin in platelets. Br J Pharmacol 1987;92: Vanhoutte PM, Auch-Schwelk W, Boulanger C, et al. Does endothelin-1 mediate endothelium-dependent contractions during anoxia? J Cardiovasc Pharmacol 1989;13(SuppI 5): s Katusic ZS, Vanhoutte PM. Anoxic contractions in isolated canine cerebral arteries: contribution of endothelium-derived factors, metabolites of arachidonic acid, and calcium entry. J Cardiovasc Pharmacol 1986;8(Suppl8):S Vanhoutte PM, Auch-Schwelk W, Boulanger C, et al. Does endothelin-1 mediate endothelium-dependent contractions during anoxia? J Cardiovasc Pharm 1989;13(Suppl5):S Rees DD, Palmer RMJ, Hodson HF, Moncada S. A specific inhibitor of nitric oxide formation from L-arginine attenuates endothelium-dependent relaxation. Br J Pharmacol 1989;96: Palmer RMJ, Ferrige AG, Moncada S. Nitric oxide release accounts for the biological activity of endothelium-derived relaxing factor. Nature 1987; Ignarro LJ, Buga GM, Wood KS, Byrns RE, Chaudhuri G. Endothelium-derived relaxing factor produced and released from artery and vein is nitric oxide. Proc Natl Acad Sci USA 1987;84: Palmer RMJ, Rees DD, Ashton DS, Moncada S. L-arginine is the physiological precursor for the formation of nitric oxide in endothelium-dependent relaxation. Biochem Biophys Res Commun 1988;153: Martin W. Basal release of endothelium-derived relaxing factor. In: Vanhoutte PM, ed. Relaxing and contracting factors. Clifton, NJ: Humana Press, 1988: Rees DD, Palmer RMJ, Moncada S. Role of endotheliumderived nitric oxide in the regulation of blood pressure. Proc Natl Acad Sci USA 1989;86: DeMey J, Vanhoutte PM. Interactions between Na+, K+ exchanges and the direct inhibitory effect of acetylcholine on canine femoral arteries. Circ Res 1980;46: Blough NV, Zafiriou OC. Reaction of superoxide with nitric oxide to form peroxynitrite in alkaline aqueous solution. Inorgan Chem 1985;24: Beckman JS, Beckman TW, Chen J, Marshall PA, Freeman BA. Apparent hydroxyl radical production by peroxynitrite: implications for endothelial injury from nitric oxide and superoxide. Proc Natl Acad Sci USA 1990;87:162& Ratych RE, Chuknyiska RS, Bulkley GB. The primary localization of free radical generation after anoxia/reoxygenation in isolated endothelial cells. Surgery 1987;102: Zweier JL, Kuppusamy P, Lutty GA. Measurement of endothelial cell free radical generation: evidence for a central mechanism of free radical injury in postischemic tissue. Proc Natl Acad Sci USA 1988;85: Gryglewski RJ, Palmer RMJ, Moncada S. Superoxide anions play a role in the breakdown of endothelium-derived relaxing factor. Nature 1986;320: Rubanyi GM, Vanhoutte PM. Superoxide anions and hyperoxia inactivate endothelium-derived relaxing factor. Am J Physiol 1986;250(Heart Circ Physiol 19):H Stewart DJ, Pohl U, Bassenge E. Free radicals inhibit endothelium-dependent dilation in the coronary resistance bed. Am J Physiol 1988;255(Heart Circ Physiol 24):H Feinberg H, Rosenbaum DS, Levitsky S, Silverman NA, Kohler J, LeBreton G. Platelet deposition after surgically induced myocardial ischemia. J Thorac Cardiovasc Surg 1982;84: Notice From the American Board of Thoracic Surgery The part 1 (written) examination will be held at the Hyatt- Regency, Dallas Fort Worth Airport, Dallas TX, on February A candidate applying for admission to the certifying examination must fulfill all the requirements of the board 16, The closing date for registration is August 1, in force at the time the application is received. To be admissible for the part I1 (oral) examination, a Please address all communications to the American candidate must have successfully completed the part I Board of Thoracic Surgery, One Rotary Center, Suite 803, (written) examination. Evanston, IL