619 Ann Thorac Surg 46: , Dec Copyright by The Society of Thoracic Surgeons

Transcription

1 Clastogenic Factor in Ischemia-Reperfusion Injury During Open-Heart Surgery: Protective Effect of Allopurinol Ingrid Emerit, M.D., Jean-Noel Fabiani, M.D., Olivier Ponzio, M.D., Andrew Murday, M.D., Francoise Lunel, M.D., and Alain Carpentier, M.D., Ph.D. ABSTRACT The hypothesis tested was that free radicals generated following ischemia and reperfusion in cardiac operations can produce clastogenic factor that results in chromosomal aberration. Fourteen randomized patients undergoing coronary artery bypass grafting were divided into two groups. In Group 1 (7 patients), myocardial protection was achieved using a cardioplegic solution without allopurinol. In Group 2 (7 patients), 100 mg of allopurinol (xanthine oxidase inhibitor) was added to the solution. In both groups, blood samples were taken from the coronary sinus before the aorta was clamped and 20 minutes after myocardial reperfusion was achieved. The blood samples were used to study the patients chromosomes. The results were given as the percentage of chromosomal aberrations observed in 100 mitoses. There were no significant differences between the preischemic values in both groups and the postischemic values in Group 2. On the other hand, there was a significant difference between the postischemic values in Groups 1 and 2 (p < 0.01). In conclusion, reperfusion following myocardial ischemia in cardiac operations can produce clastogenic aberrations. This clastogenic activity can be reduced by adding allopurinol to the cardioplegic solution. It is well established that oxygen-derived free radicals play a crucial role in the pathogenesis of ischemiareperfusion injury [l-31. The source of these radicals, however, is still subject to debate. McCord [4] proposed that oxygen radicals are generated during reoxygenation of ischemic myocardium by the action of xanthine oxidase on hypoxanthine. Adenosine triphosphate is progressively degraded during ischemia to yield hypoxanthine, which then acts as a substrate for xanthine oxidase. This enzyme is produced during ischemia from xanthine dehydrogenase by the action of protease. On reperfusion, molecular oxygen becomes available, and all participants are present for the reaction producing 0;. Based on their observations of an allopurinolmediated reduction of reperfusion injury, a number of investigators [5-101 supported this hypothesis. Also, in From the Laboratoire de Genetique, Institut Biomedical des Cordeliers, CNRS, Universite Pans VI, and the Clinique de Chirurgie Cardiovasculaire, HBpital Broussais, Pans, France. Accepted for publication June 27, Address reprint requests to Dr. Fabiani, HBpital Broussais, 96 rue Didot, Paris, France. a large series of patients having operation at Broussais Hospital, allopurinol treatment proved to be beneficial [ll, 121. Recently, however, this mechanism has been questioned, in particular because of the virtual absence of xanthine activity in mammalian hearts except the heart of the rat. The protective effect of allopurinol is probably due to the free radical scavenging properties of this compound rather than to its role as an inhibitor of xanthine oxidase [7, 81. Not only direct action of free radicals, but also toxic effects of secondary products have been considered as the origin of reperfusion injury. Signals suggestive of carbon-centered free radical adducts formed by secondary reactions of initially generated 0; and hydroxyl radicals with membrane lipids have been recorded [3]. Hydroxyconjugated dienes were fivefold higher on reperfusion than at the end of ischemia [9]. From skeletal muscle ischemia, it also is known that peroxidation products of free fatty acids are formed from muscle phospholipids [13]. With reperfusion of the ischemic myocardium, the material formed secondary to oxyradical generation may be similar to the so-called chromosome-breakage factors or clastogenic factors released by cells under conditions of oxidative stress. These factors were described by radiobiologists, who noted that therapeutically or accidentally irradiated persons show not only an increased frequency of chromosome breaks and rearrangements in their own cells but also that the plasma of these persons contains chromosome-damaging material able to induce chromosome breakage after transfer into other cell culture systems [14]. Work in our laboratory [15, 161 demonstrated the presence of similar transferable clastogenic material in the plasma of patients with chronic inflammatory diseases such as progressive systemic sclerosis, lupus erythematosus, rheumatoid arthritis, disseminated sclerosis, and Crohn s ileocolitis. The formation and action of clastogenic factors are related to oxyradical generation. This is documented by the protective effect of superoxide dismutase and the ability to generate clastogenic factor in vitro by exposure of cells to a source of 0; [17,18]. The action of xanthine oxidase on hypoxanthine, supposed to be a source of oxyradical generation in reperfusion injury, is a widely used method to generate 0; in vitro. We [18] were able to produce clastogenic factors in vitro by exposure of human blood cultures to hypoxanthine plus xanthine oxidase, and demonstrated that lipid peroxidation products are components of this clastogenic material. It seemed of interest to search for similar clastogenic factors in patients having reperfusion following ischemia, 619 Ann Thorac Surg 46: , Dec Copyright by The Society of Thoracic Surgeons

2 620 The Annals of Thoracic Surgery Vol46 No 6 December 1988 Table 1. Composition of Cardioplegic Solutions Cold Crystalloid Cardioplegia Medium 1 L of water + NaCl 8.59 gm KCl gm + CaCI gm +THAM gm + Allopurinol 0.1 gm Cardioplegia ampule (20 n-4 + KC gm MgCI gm + Procaine hydrochloride gm Warm Blood Cardioplegia Medium 500 ml of water + NaCl 4.3 gm i- KCl 0.14 gm + CaC gm +THAM gm + Allopurinol 0.1 gm 500 ml of autologous oxygenated blood Cardioplegia ampule (20 ml) + KCl gm Mga gm + Procaine hydrochloride gm Glutamate 26 mmol THAM2 = thromethamine. and to compare patients having operation with the protection of allopurinol with patients having operation without this protection. Material and Methods This study comprises 14 randomized patients having coronary artery bypass operation. The patients were divided into two equal groups. Group 1 patients underwent operation in the same manner as Group 2 except that they received no allopurinol. Group 2 patients received 100 mg of allopurinol in 1 liter of cold (4 C) cardioplegic solution injected into the ascending aorta immediately following cross-clamping of the aorta (injection pressure, 80 mm Hg). After bypass grafting was performed (average, 2.3 grafts per patient) and before the aorta was unclamped, a second or reperfusion cardioplegic solution was injected into the bypass grafts. This was a warm (37 C) blood cardioplegic solution also containing 100 mg of allopurinol. The composition of both cardioplegic solutions is given in Table 1. Cross-clamping time (between injections of cold and warm cardioplegia) ranged from 38 to 67 minutes (mean, 50 k 10 minutes). Neuroleptanalgesic regimens were used in all patients. Standard monitoring techniques included continuous ECG monitoring as well as arterial, central venous, and Swan-Ganz catheters inserted preoperatively. Continuous recording of blood pressure and perfusion pressure was undertaken throughout the intervention. Measurement of the cardiac index was performed before and 15 minutes following bypass grafting using the thermodilution technique. All patients received nitroglycerin intravenously at a rate of 2 to 3 kg per kilogram per minute. Operative mortality due to myocardial infarction as evidenced by ECG and enzyme studies, postoperative arrhythmias, and the need for inotropic support or mechanical support (intraaortic balloon pump) were recorded and analyzed in both groups. Cardiac output measurements were assessed in all patients. One blood sample was taken from the coronary sinus E; I COLD CARDlOPLEGlA WARM CARDlOPLEGlR 4-c 37. c.ischemir SAMPLE A (cross-clamp time : 50 nn) REPERFUSION Fig I. Operative procedure and timing for blood samples from the coronary sinus. Group I received 1 liter of Cardioplegia and Group 2, 1 liter of cardioplegia plus allopurinol. (ECC = extracorporeal circulation; CABG = coronary artery bypass grafting.) at the beginning of the operation before the aorta was clamped (sample A). A second blood sample was taken 20 minutes after release of the aortic clamp (sample B). Therefore samples A and B correspond to the phases of preischemia and postischemia, respectively (Fig 1). In 6 patients, an additional sample was taken immediately after reflow. All samples were from the coronary sinus blood. The blood samples were transferred immediately to the laboratory. The origin of the samples, that is, whether from treated or untreated patients, remained unknown. Part of each sample was used for the study of the patient s chromosomes. The blood cultures were set up according to conventional cytogenetic procedures. Each culture flask contained 5 ml of tissue culture medium 199 (Flow Lab., Paris, France), 1.5 ml of heatinactivated fetal calf serum, and 0.5 ml of whole blood. The cell cultures were stimulated to divide with phytohemagglutinin P (Difco Lab., Detroit, MI). After 68 to 72 hours of incubation at 37 C, the mitoses were arrested in metaphase by the addition of colchicine. Slides were prepared for chromosome analysis, and 50 mitoses were studied for each sample. The results were shown as the number of aberrations observed in 100 mitoses.

3 621 Emerit et ak Clastogenic Factor and Reperfusion Untreated Patients IA IB I1 IIA IIB Mean Patients with Allopurinol IA IB I1 IIA IIB Mean The majority of the blood was used for isolation of clastogenic material according to the procedure developed in our laboratory for clastogenic factor preparations from plasma of irradiated persons or patients with chronic inflammatary diseases [15, 161. The plasma was filtered through Diaflo ultrafiltration filters (YM lo), that retain substances with a molecular weight of more than 10,000 daltons. The ultrafiltrates were then concentrated ten times by filtration through a second filter (YM 2) with a cutoff at 1,000 daltons. The concentrates were tested for their chromosome-damaging (clastogenic) effect in a test culture set up with whole blood from normal volunteers (i.e., laboratory personnel). Aliquots of 0.5 ml were added to each culture flask containing 5 ml of tissue culture medium 199, 1.5 ml of fetal calf serum, and 0.5 ml of whole blood, as described for the cultures of patients blood. Test cultures were set up in duplicate. The ultrafiltrates from samples A and B were studied simultaneously on chromosomes from the same blood donor. Results The frequency of structural chromosomal aberrations was in the normal range in all 14 patients before operation (sample A) and corresponded to the frequency observed in the cultures of the healthy donors used as the test system for the ultrafiltrates, as shown in columns IA and 11 in Table 2. The concentrated ultrafiltrates prepared from these samples did not induce chromosomal damage in the test culture system (Table 2, column IIA). An increased aberration rate was noted for 9 of the 14 patients in the cultures set up with sample B, the samples collected after 20 minutes of reperfusion. The same samples also induced chromosomal damage after addition to the test cultures (Table 2, column IIB). After identification of the origin of the blood samples, it was concluded that 7 of the 9 patients with chromosomal damage belonged to the group of untreated pa- tients (Group l) (Fig 2). The 5 patients without chromosomal damage had received allopurinol (Group 2). AS noted in Table 1, the mean aberration rate in the 7 untreated patients increased from 6 to 24% in the patients blood cultures and from 6 to 24% in the test cultures to which the concentrates of patients plasma had been added. In 6 of the 7 patients, an additional blood sample had been taken immediately after the aorta was unclamped. The incidence of chromosomal aberration was not significantly increased (8% versus 6% before the operation). Also the concentrated ultrafiltrates of these samples were only slightly clastogenic (increase in the aberration rate in the test culture system from 6 to 12%). This shows that a certain delay was necessary for the formation of the chromosome-damaging material. Among the 7 patients who received allopurinol, 2 showed increased chromosomal damage (18% aberration in each) and induced an aberration rate of 12% and 24% in the test cultures. In the other 5 patients, no increase in aberrations was observed either in the patients or in the donors chromosomes (Fig 3). These results are summarized in Figure 4. The statistical method used to compare the results was the t test for paired values (Statview [Abacus Concept, Inc] for MacIntosh 11). There were no significant differences between the preischemic values of chromosomal aberrations in Groups 1 and 2 and the postischemic values in Group 2. However, the difference was significant in GrouF 2 (p < 0.01). Although the results show significant differences between the two groups in terms of clastogenic factors, the clinical postoperative course was uneventful in all 14 patients. There were no deaths or complications. No patient had evidence of myocardial infarction or cardiac failure. Comparison of the cardiac indices recorded preoperatively and postoperatively in both groups did not indicate any significant differences. The types of aberrations observed are shown in Table 3. In the patients blood cultures as well as in the control

4 622 The Annals of Thoracic Surgery Vol46 No 6 December 1988 % I PATIENTS Fig 2. Chromosome analysis in untreated patients. 7 u) a w m U PATlENlS Fig 3. Chromosome analysis in allopurinol-treated patients. donor cultures exposed to the ultrafiltrates from patients plasma, chromatid breaks represented about 50% of the total aberrations observed. Breaks of both chromatids, acentric fragments, and structurally abnormal chromosomes (e.g., dicentric and translocation) represented about 20% of the total. This means that chromatid-type aberrations were predominant. If gaps are excluded from the total, the increase in chromosomal aberrations is reduced by around 25%. Compared with donor controls, this represents a threefold and still highly sigruficant elevation. Since no diagnostic roentgenograms and no drugs were used during the operation, the chromosomal damage observed cannot be explained by well-known clastogenic agents. We conclude that in agreement with our working hypothesis, clasto- genic factors were generated as a consequence of the oxidase stress situation created on reperfusion. The protective effect of allopurinol is explained either by the inhibition of a xanthine oxidase reaction or, as already mentioned, by the free radical scavenging properties of this substance. Comment Clastogenic factor preparation followed the procedure developed for clastogenic factor in irradiated persons, patients with chronic inflammatory diseases, or clastogenic factor induced in vitro by a xanthine oxidase reaction. This is an indication that the molecular weight of this material is less than 10,000 daltons. Ultrafiitrates from plasma that went through filters with a cutoff at

5 623 Emerit et al: Clastogenic Factor and Reperfusion p < 0.01 T CONTROLGROUP ALLOPURINOL 1 PRE ISCHEMIA 2 POST ISCHEMIA Fig 4. Chromosomal aberrations before and after ischemia in the two groups. (NS = not significant.) Table 3. Types of Chromosomal Aberrations Seen Cultures of Cultures of Donors Exposed Aberration Patients' Blood (%) to CF (%) Chromatid gaps Chromatid breaks Isochromatid gaps Isochromatid breaks Acentric fragments Dicentrics and translocations CF = clastogenic factor. 10,OOO daltons were concentrated by a second passage through a filter with a cutoff at 1,OOO daltons. However, since certain molecules may be retained by specific binding to the filter membrane, the molecular weight of clastogenic factors may well be less than 1,OOO daltons. Free radicals such as 0; or OH cannot be considered components of clastogenic factors, since they would not be transferable into other culture systems. More longlived species formed through free radical reactions may be direct deoxyribonucleic acid-(dna) damaging agents or may exert their clastogenic effect through free radicals. The protective effect of free radical scavenging agents in the test cultures exposed to clastogenic factors is an indication of this. Furthermore, as suggested by preliminary results, clastogenic factors stimulate 0; production by neutrophils. This observation is interesting with respect to preactivation of neutrophils invading the infarct zone. Clastogenic factor generated in vitro by a source of 0; contains lipid peroxidation products as indicated by the thiobarbituric acid assay and the presence of conjugated dienes [18]. Formation of clastogenic factor in hypoxanthine plus xanthine oxidase-treated cultures needs about 15 hours to reach detectable levels. The clastogenic factor observed in patients undergoing ischemia and reperfusion seems to appear also after a delay with respect to the burst of free radicals. No clastogenic factor activity was detectable immediately after the aorta was unclamped. The activity was detectable after 20 minutes, a much shorter time than in the in vitro system. However, the latter is composed of 2.5 million cells exposed only in 6 ml of culture medium to hypoxanthine and xanthine oxidase at a final concentration of 7 &nl and 10 ~g/ml, respectively [MI. We assume that the long delay in this culture system is due to the small amounts of clastogenic factors formed, which have to accumulate to reach detectable levels. This hypothesis has to be confirmed by similarities in the nature of the components of clastogenic factor in the in vitro and in vivo systems. The biochemical identification of these components is presently under study. Clastogenic factor formation is interesting not only for the understanding of the mechanisms involved in reperfusion injury but also for a quantitative assessment of this phenomenon. In this small series, we did not observe a correlation between clastogenic factor formation and clinical course. The sensitivity of the test needs a larger number of patients for correlation with hemodynamic variables. Nevertheless, it will be possible to use the clastogenic factor as a test system for the efficacy of drugs in the prevention of reperfusion injury. Induction of chromosomal mutations by clastogenic factors is accompanied by a risk for carcinogenesis and

6 624 The Annals of Thoracic Surgery Vol46 No 6 December 1988 teratogenesis. This is well known from radiationinduced and chemically induced DNA damage. It will be necessary to examine whether clastogenic factor persists in the blood of our patients as is the case for irradiated patients or patients with chronic inflammatory diseases (15,161. Whatever the outcome of these studies, the theoretical risk is slight compared with the benefit expected from the surgical procedure. In conclusion, ischemia and reperfusion in the human heart can induce the generation of a clastogenic factor able to produce chromosome breakage. The addition of allopurinol to the cardioplegic solution inhibited this phenomenon. The clastogenesis seems to be a marker of free radical production in this model. References 1. Chambers DE, Parks DA, Patterson G, et al: Role of oxygen-derived radicals in myocardial ischemia. Fed Prod , Granger DN, Hwlwarth ME, Parks DA Ischemiareperfusion injury: role of oxygen-derived free radicals. Acta Physiol Scand m47, Zweier JL, Tanerty JT, Weisfeldt ML: Direct measurement of free radical generation following reperfusion of ischemic myocardium. Proc Natl Acad sd USA , McCord JM. Oxygen-derived free radicals in postischemic tissue injury. N En@ J Med , Peterson DA, Asinger RW, Elsperger KJ, et ak Reactive oxygen species may cause myocardial reperfusion injury. Biachem Biophys Res Commun , Chambers DE, Parks DA, Patterson G, et ak Xanthine oxidase as source of free radical damage in myocardial ischemia. J Mol Cell Cardi ~ Godin DV, Bhimji S: Effects of allopurinol on myocardial ischemic injury induced by coronary artery ligation and reperfusion. Biochem Pharmacol36:2101, Das DK, Engelman RM, Clement R, et ak Role of xanthine oxidase inhibitor as free radical scavenger: a novel mechanism of action of allopurinol and oxypurinol in myocardial salvage. Biochem Biophys Res Commun 14fk314, Romaschin AD, Wilson GJ, Mickle DAG: Hydroxyconjugated dienes, products of free radical attack on polyunsaturated phospholipids, in reversibly and irreversibly ischemic myocardium. Presented at the Symposium on Free Radicals in Biology and Medicine: Ischemia-Reperfusion Injury, Point Clear, AL, Mar 17-19, Fabiani JN The no-reflow phenomenon following early reperfusion of myocardial infarction and its prevention by various drugs. Heart Bull (La Haye), 7134, Fabiani JN, Perier P, Chelly J, et al: Blood versus crystalloid Cardioplegia: clinical evaluation. In Engelman RM (ed): Handbook of Cardioplegia. Mt Kisco, NY, Futura, 1981, p Perier P, Fabiani IN, Bocher M, et ak Protection myocardique pendant l arret cardiaque ischemique: etude heme dynamique des effets de l allopurinol dans une solution cardioplbgique. Arch Mal Coeur 73:713, Lindsay TF, Labbe R, Walker PM The role of oxygen radicals in reperfusion injury of skeletal muscle. Presented at the Symposium on Free Radicals in Biology and Medicine: Ischemia-Reperfusion Injury, Point Clear, AL, Mar 17-19, Faguet GB, Reichard SM, Welter DA Radiation-induced dastogenic plasma factors. Cancer Genet Cytogenet 1273, Emerit I: Chromosome breakage factors: origin and possible significance. Prog Mutat Res 461, Emerit I Properties and action mechanisms of clastogenic factors. In Pick E (ed): Lymphokines. New York, Academic, 1983, pp 8, Emerit I, Cerutti P Tumor promoter phorbol-myristateacetate induces a clastogenic factor in human lymphocytes. Proc Natl Acad sd USA 79:7509, Emerit I, Khan SH, Cerutti P Treatment of lymphocyte cultures with a hypoxanthine-xanthine oxidase system induces the formation of transferable dastogenic material. J Free Radicals Biol Med 1:51, 1985