reperfusion injury in the isolated blood-free perfused rat heart

Thrombosis Research 108 (2003) 249 – 255

Regular Article

Influence of antithrombin on ischemia/reperfusion injury in the isolated blood-free perfused rat heart Josef Margreiter a,*, Markus Mittermayr a, Johannes Mair b, Angelika Hammerer-Lercher c, Jordan Kountchev c, Anton Klingler d, Wolfgang Schobersberger a a

Department of Anesthesia and Intensive Care Medicine, The Leopold-Franzens University of Innsbruck, Anichstrasse 35, A-6020 Innsbruck, Austria b Department of Internal Medicine, The Leopold-Franzens University of Innsbruck, Anichstrasse 35, A-6020 Innsbruck, Austria c Institute of Medical Chemistry and Biochemistry, The Leopold-Franzens University of Innsbruck, Fritz Preglstr.3, A-6020 Innsbruck, Austria d Department of General and Transplant Surgery, Theoretical Surgery Unit, Schoepfstr. 41, The Leopold-Franzens University of Innsbruck, A-6020 Innsbruck, Austria Received 11 July 2002; received in revised form 27 December 2002; accepted 14 January 2003

Abstract Introduction: Antithrombin (AT) is well known as an important inhibitor of the coagulation system. An interesting new hypothesis is that antithrombin exerts specific anti-inflammatory effects by stimulating the production of prostacyclin in endothelial cells. Recent studies report beneficial influence on ischemia/reperfusion injury in several organs. These effects are independent of the coagulation system. We investigated the influence of antithrombin on ischemia/reperfusion injury and prostacyclin release in the isolated rat heart. Since the perfusion of the hearts was without blood, the used model essentially describes effects of antithrombin on endothelial cells. Material and methods: Experiments were performed using the temperature-controlled and pressure-constant Langendorff apparatus. The hearts of 32 male Sprague – Dawley rats were subjected to 20 min of global ischemia followed by 30 min of reperfusion. Antithrombin was administered in three different concentrations (1, 4 and 8 U/ml) 15 min prior to global ischemia. Cardiac contractility parameters and biochemical parameters were measured. Results: Treatment with antithrombin did not increase the release of prostacyclin significantly after ischemia. Antithrombin at a concentration of 8 U/ml led to a significant increase in creatine kinase (CK; p < 0.05) and troponin I ( p < 0.05), whereas measurements of lactate dehydrogenase (LDH) revealed no significant differences between treated and untreated hearts. Conclusion: Our study shows that antithrombin did not reduce ischemia/reperfusion injury in the isolated heart, and prostacyclin is not significantly released following antithrombin treatment. High concentrations of antithrombin, however, might have a negative influence on the reperfused heart. The underlying mechanism remains unclear. D 2003 Elsevier Science Ltd. All rights reserved. Keywords: Ischemia; Reperfusion; Prostaglandins; Endothelial function; Inflammation

1. Introduction Antithrombin (AT) is an important inhibitor of serine proteases that are part of the coagulation cascade. Recently, interest has focused on the anti-inflammatory effects of AT. Administration of a high concentration of AT improved metabolic dysfunctions, organ failure and mortality in animals exposed to lipopolysaccharides [1]. Taylor et al. [2] showed that baboons infected with Escherichia coli and treated with high dose of AT had a reduced incidence of disseminated intravascular coagulation (DIC), organ failure * Corresponding author. Tel.: +43-512-504-2400; fax: +43-512-5042450. E-mail address: [email protected] (J. Margreiter).

and lower mortality. Injection of DEGR-FXa, which is a selective inhibitor of factor X, resulted in comparable anticoagulatory effects, but no improvement in mortality was observed in these animals [3]. These data suggest that the beneficial effects of AT in animal sepsis models go beyond its anticoagulatory effect. The concept of AT acting as an anti-inflammatory agent in septic patients has been demonstrated in several clinical studies. Eisele et al. [4] demonstrated a decrease in the incidence of multiple organ failure and the length of stay at the ICU in septic AT-treated patients. Long-term AT supplementation improved lung function and prevented the development of septic liver and kidney failure in patients with severe sepsis [5]. In none of these studies, however, a significant reduction in mortality was observed. Interest-

0049-3848/03/$ - see front matter D 2003 Elsevier Science Ltd. All rights reserved. doi:10.1016/S0049-3848(03)00031-8

250

J. Margreiter et al. / Thrombosis Research 108 (2003) 249–255

ingly, Baudo et al. [6] found that AT treatment improved overall mortality in the subgroup of patients with septic shock. The main mechanism suggested for AT’s anti-inflammatory effect is the release of prostacyclin (PGI2) from endothelial cells. AT has been shown to induce the release of PGI2 in vitro [7,8] and in vivo [9] by interacting with cell surface glycosaminoglycans (GAG). PGI2 promotes vasodilatation and inhibits platelet aggregation by increasing cyclic AMP levels. In addition, it suppresses the release of TNFa [10] and inhibits neutrophil activation [11] and neutrophil adhesion to the surface receptors on endothelial cells [12]. AT (250 U/kg) reduced coagulation abnormalities and pulmonary vascular injury in endotoxin-exposed rats which was accompanied by elevated plasma levels of PGI2. These effects were completely inhibited by indomethacin [13]. Interestingly, Dschietzig et al. [14] demonstrated that AT treatment failed to increase PGI2 levels in isolated endotoxin-exposed rat lungs, but augmented release of endothelin 1. In recent studies, AT was found to have beneficial influence on ischemia/reperfusion injury of the liver and the intestine. Ozden et al. [15] showed a decrease in the histopathological injury of the reperfused intestine when pretreated with AT. A recovery of tissue blood flow in the liver after clamping the hepatoduodenal ligament was improved by AT treatment. In addition, PGI2 levels were elevated [16]. Harada et al. [17] confirmed the stimulatory influence of AT on PGI2 production in the reperfused rat liver and found that all the effects were prevented by additional administration of indomethacin. Ischemia/reperfusion injury plays an important role following cardiac transplantation, bypass surgery and myocardial infarction. AT had positive effects on ischemia/ reperfusion injury in the liver and the intestine, but was

hitherto not tested for its effects in the heart. The objective of our study was to evaluate whether AT stimulates cardial PGI2 release. In addition, we investigated possible influences of AT on ischemia/reperfusion injury in the isolated heart as measured by markers of myocardial damage (creatine kinase, troponin I, lactate dehydrogenase) and hemodynamic parameters.

2. Materials and methods The present study was approved by the National Animal Research Ethics Committee in Vienna and conforms with the Guide for the Care and Use of Laboratory Animals published by the US National Institutes of Health (NIH publication No. 85-23, revised 1985). Male Sprague – Dawley rats (275 – 320 g) were anesthetized with ketamine (75 mg/kg) and xylacin (10 mg/kg) intraperitoneally and given 1000 IU heparin intravenously. After opening the chest cavity, the hearts were quickly excised and arrested in icecold St. Thomas cardioplegic solution. Retrograde perfusion via the aorta as described by Langendorff was immediately commenced using a modified Krebs– Henseleit buffer that contained 116 mmol/l NaCl, 4.7 mmol/l KCl, 2.5 mmol/l CaCl2, 22.02 mmol/l NaHCO3, 1.25 mmol/l KH2PO4, 1.05 mmol/l MgSO4 and 10 mmol/l glucose. The perfusate was bubbled with a mixture of 95% O2 and 5% CO2, which produced a pO2 range of 600 –700 mm Hg, a pCO2 range of 38– 42 mm Hg and a pH range of 7.38 –7.42. The perfusate solution and bath temperature were maintained at 37.0 jC using a thermostatically controlled water circulating system (PC/4 JULABO, Germany). Perfusion pressure was maintained at 60 mm Hg by a transducer (P23XL Statham, Gould Electronics, USA) via a computerized feedback controlling

Table 1 Coronary flow, LVP and dp/dtmax in controls and AT-treated hearts (1, 4 and 8 U/ml) at baseline, after ischemia and at the end of observation Baseline Median Coronary flow (ml/min) Control 13.3 AT 1 U/ml 15.8 AT 4 U/ml 14.9 AT 8 U/ml 16.5 LVP (mm Hg) Control AT 1 U/ml AT 4 U/ml AT 8 U/ml

102.7 102.1 111.8 107.8

dp/dtmax (mm Hg/s) Control 2639 AT 1 U/ml 2310 AT 4 U/ml 2743 AT 8 U/ml 2612

After ischemia Min

Max

Median

End of observation Min

Max

Median

Min

Max

12.2 12.3 13.4 13.8

25.0 19.1 17.1 19.6

7.8 12.3 9.8 9.8

6.1 6.1 7.1 7.3

19.0 16.1 15.2 12.0

9.2 12.9 1.0 11.4

7.5 8.4 8.9 8.8

23.8 18.6 16.2 14.5

81.8 82.0 107.5 84.0

119.1 123.7 131.6 139.0

33.3 26.0 66.4 50.8

25.6 7.1 13.6 39.1

66.3 68.0 80.0 95.0

50.8 82.6 93.0 52.5

15.3 71.7 11.5 22.9

73.1 99.1 101.1 113.0

2224 2027 2513 1999

3052 2949 3126 3615

663 365 1152 866

284 66 165 268

1223 1215 1680 1896

1171 1748 2551 940

200 1547 128 278

1780 2228 2351 2054

J. Margreiter et al. / Thrombosis Research 108 (2003) 249–255

system that regulates the velocity of a roller pump (MC-MS CA 4/6, Ismatec, Switzerland). The hearts beat spontaneously. Left ventricular pressure (LVP) was measured isovolumetrically with a transducer (P10EZ Statham, Gould Electronics) connected to a thin catheter. A small balloon was mounted on this catheter and inserted in the left ventricle via the mitral valve. Heart rate, rate of pressure rise (dp/dtmax) and rate of pressure fall (dp/ dtmin) were taken from the left ventricular pressure curve. Coronary flow was determined by an ultrasonic flow probe at the aortic root (T108, Transonic Systems, Germany). All parameters were continuously monitored on a personal computer (Compaq, Deskpro) with special software (LabVIEW, National Instruments, USA). The hearts were allowed to equilibrate for 15 min to stabilize heart rate, LVP and coronary flow before baseline measurements were taken. Afterwards, global ischemia was induced by stopping perfusion for 20 min. Reperfusion had then taken place for 30 min. All hemodynamic parameters were recorded every 30 s during an observation period of 65 min. For statistical analysis of the hemodynamic parameters, mean values out of a period of 5 min were used. 2.1. Experimental procedure The hearts were either treated with AT added to the perfusate 15 min prior to ischemia or remained untreated

251

to serve as controls. Out of a total of 32 hearts, 6 were treated with AT at a concentration of 1 U/ml, 5 hearts at a concentration of 4 U/ml and 11 at a concentration of 8 U/ml. Ten hearts served as controls and remained untreated. The effluent was collected before ischemia, immediately after ischemia, 5, 15 and 30 min after ischemia in order to measure creatine kinase (CK), lactate dehydrogenase (LDH), troponin I and 6-keto-PGF1a, which is a stable metabolite of prostacyclin. AT was kindly provided by Aventis-Behring (Marburg, Germany). Because of side effects on the hemodynamic performance of the buffer per se containing aminoacetic acid, sodium chloride, and sodium citrate, AT was prepared in Krebs – Henseleit solution. 2.2. Biochemical measurements A column packed with ethyl-bonded silica gel was used to extract 6-keto-PGF1a (C2-ethyl, Amersham, Buckinghamshire, UK). The column was prepared by washing it with 1 ml methanol and 1 ml water. Interfering substances were washed out using 1 ml water, 1 ml ethyl alcohol and 1 ml hexane. 6-keto-PGF1a was then eluted with methyl formate. The concentration of 6-ketoPGF1a was measured using a specific ELISA (Biotrak, Amersham).

Table 2 Troponin, 6-keto-PGF1a, creatine kinase and lactate dehydrogenase in control and AT-treated group (1, 4 and 8 U/ml) at baseline, after ischemia and at the end of observation Baseline Median 6-keto-PGF1a (pg/ml) Control 125.2 AT 1 U/ml 177.6 AT 4 U/ml 221.6 AT 8 U/ml 105.6

After ischemia

End of observation

Min

Max

Median

Min

Max

Median

Min

Max

29.0 117.9 195.6 44.0

248.6 327.4 364.4 319.2

697.1 1535.0 1460.0 1627.0

343.2 842.3 1033 702.1

1757.0 2183.0 3642.0 2696.0

83.9 137.5 111.3 99.5

27.8 80.2 109.6 51.4

221.4 217.3 204.1 170.3

Troponin I (ng/ml) Control AT 1 U/ml AT 4 U/ml AT 8 U/ml

0.0 0.0 0.1 0.1*

0.0 0.0 0.0 0.0

0.1 0.1 0.1 0.4

0.1 0.0 0.2 0.7*

0.1 0.0 0.0 0.1

0.2 0.3 0.3 2.5

0.4 0.1* 0.5 1.0*

0.1 0.0 0.3 0.5

0.8 0.4 0.7 1.9

CK (U/l) Control AT 1 U/ml AT 4 U/ml AT 8 U/ml

7.8 9.5 9.0 2.0*

7.2 5.3 6.8 0.1

23.6 16.2 12.0 6.5

15.2 32.6 34.0 73.6*

0.2 11.1 19.0 57.7

60.6 48.0 43.6 144.2

11.7 68.1* 73.9 102.1*

0.7 12.4 1.9 40.9

70.4 112.8 116.6 196.7

LDH (U/l) Control AT 1 U/ml AT 4 U/ml AT 8 U/ml

5.4 4.4 1.6 4.9

0.5 2.6 0.0 0.6

76.6 6.9 11.6 12.8

15.9 4.7 10.4 22.7

1.5 3.2 4.2 9.9

50.3 7.4 15.0 33.8

4.5 4.8 7.8 11.4

1.0 0.5 0.9 4.1

14.5 13.7 20.7 31.6

* Differences between control and corresponding group, p < 0.05.

252

J. Margreiter et al. / Thrombosis Research 108 (2003) 249–255

3. Results Table 1 shows baseline values before ischemia (15 min after AT administration), values after ischemia and at the end of observation. There were no significant differences between controls and the AT-treated hearts regarding the hemodynamic parameters (coronary flow, LVP and dp/ dtmax). Concentrations of the biochemical parameters (6-ketoPGF1a, troponin I, CK and LDH) for the control group and the AT-treated groups are illustrated in Table 2. When compared to untreated controls, the release of 6-keto-PGF1a was not influenced by treatment with antithrombin at any concentration. The detailed time course of 6-keto-PGF1a is depicted in Fig. 1. In the 8 U/ml AT group, the concentrations of troponin I and CK remained significantly higher after ischemia than in the control group ( p < 0.05), while under baseline conditions, perfusion of the rat heart with 8 U/ml AT resulted in higher concentrations of troponin I and lower concentrations of CK ( p < 0.05). Thirty minutes after ischemia (end of observation), troponin I levels were lower and CK measure-

Fig. 1. 6-keto-PGF1a release in controls and in AT-treated group (1, 4 and 8 U/ml). Median, percentiles and extreme values (minimum and maximum) are given in a boxplot.

For determination of troponin I, a specific ELISA (Access Troponin I, Beckman, Palo Alto, USA) was used. Creatine kinase was measured photometrically (Granutest 25, Merck, Darmstadt, Germany). Lactate dehydrogenase was also measured photometrically (LDH, Boehringer, Mannheim, Germany). 2.3. Statistics For hemodynamic parameters, the mean value of a period of 5 min was used; for determination of biochemical markers, five samples were taken (baseline, after ischemia, 5 min after ischemia, 15 min after ischemia and 30 min after ischemia). Because our data are not normally distributed, nonparametrical statistical tests were used. Differences within each group were compared using the Wilcoxon test for paired observations; for the comparisons between the AT-treated groups and the control group, a Wilcoxon two-sample test was applied. A p-value below 5% was considered significant. All parameters are described by median and range, and graphics are depicted as boxplots.

Fig. 2. Troponin I release in controls and in AT-treated group (1, 4 and 8 U/ ml). Median, percentiles and extreme values (minimum and maximum) are given in a boxplot. *= p < 0.05.

J. Margreiter et al. / Thrombosis Research 108 (2003) 249–255

Fig. 3. Creatine kinase release in controls and in AT-treated group (1, 4 and 8 U/ml). Median, percentiles and extreme values (minimum and maximum) are given in a boxplot. *= p < 0.05.

ments were higher in hearts, treated with AT at a concentration of 1 U/ml ( p < 0.05), than in untreated controls. The time course of troponin I is shown in Fig. 2, while that of creatine kinase is shown in Fig. 3. Measurement of LDH revealed no significant differences between treated and untreated hearts.

4. Discussion In this study, the administration of AT showed no positive effect on ischemia/reperfusion injury in the isolated rat heart. Not even at high concentrations of antithrombin a significant release of PGI2, but an increased release of the tissue injury markers troponin I and creatine kinase was observed. With regard to the anti-inflammatory properties of AT, binding of antithrombin to thrombin is suggested as a possible mechanism, as thrombin promotes enhanced expression of p-selectin and the platelet activating factor [18], and therefore causes adhesion of leukocytes to the vessel wall. AT has also been shown to have thrombin-

253

independent effects like inhibition of IL-6 [19] and tissue factor [20] release and the reduction of inflammatory cell adhesion and migration [21]. Another pathway by which AT acts as an anti-inflammatory agent could be the release of PGI2. In the present study, administration of AT at all concentrations induced an increase in PGI2 immediately after ischemia (n.s.), but during reperfusion, PGI2 decreased to the same low concentrations as in untreated controls. Yamauchi et al. [8] showed that AT releases prostacyclin in bovine aortal endothelial cells after 2-h incubation. The additional administration of heparin blocked this effect, and the authors thus suggested that the binding of AT to heparinlike glycosaminoglycans on the surface of endothelial cells promotes the release of prostacyclin. Horie et al. [7] confirmed these results on human umbilical vein endothelial cells (HUVEC). It is still not clear, however, whether the binding of AT to the surface GAG is the only mechanism responsible for the release of PGI2. Uchiba et al. [9] demonstrated these effects for the first time in rats and found a dose-dependent increase of plasma-PGI2 30 min after administration of AT. This increase in PGI2 began with the administration of 100 U/kg AT and reached its maximum at 250 U/kg AT [22]. The fact that a very high concentration of AT is necessary to produce significant PGI2 release was recently shown by Schobersberger et al. Only 25 U/ml AT induced a significant increase in PGI2 in the culture supernatant of human dermal microvascular cells (HDMEC) [23]. It is not clear why such a high dose of AT is necessary. A possible explanation is the reduced affinity of human AT to heparin-like GAGs in rats [9]. The binding of AT to thrombin is also discussed as another mechanism. The latter is very unlikely in our study because we used a blood-free solution for perfusion of the heart. Hence, the isolated heart model used in this experiment allows the effects of AT on the vessel wall to be observed without interferences from other blood components. Uchiba et al. [13] demonstrated that infusion of 250 U/kg AT in septic rats led to a significant increase in PGI2. The endotoxin-induced pulmonary vascular injury was markedly reduced. These effects of AT were blocked by indomethacin, which is known to inhibit the formation of prostacyclin. Following ischemia, an enhanced release of PGI2 in the isolated heart can be observed in the reperfusion period, starting immediately after ischemia with values dropping very quickly to pre-ischemic levels [24]. In our study, PGI2 followed the same time course, and administration of AT increased the concentration of PGI2 only immediately after ischemia. In other studies where AT promoted elevated PGI2 levels after ischemia, peak values were observed from 1 to 3 h after reperfusion [16,17]. Since this time course was different in our study, it is unlikely that this effect is based on the same mechanism. Whether elevated levels of PGI2 are beneficial for the reperfused heart is the subject of controversial discussion. Inhibition of PGI2 synthesis by

254

J. Margreiter et al. / Thrombosis Research 108 (2003) 249–255

nonsteroidal antirheumatics improves contractility in the reperfusion period [25]. Low concentrations of PGI2 suppress the recovery of contractility after ischemia, probably caused by calcium accumulation [26]. Verapamil prevented these negative effects of PGI2 [27]. Administration of indomethacin resulted in severe disturbances of myocardial function [28,29] and administration of PGI2 in improved contractility after ischemia [30]. Thus, it is not clear whether PGI2 is beneficial for the reperfused heart. 4.1. Antithrombin and ischemia/reperfusion (I/R) In our study, AT had no beneficial influences on ischemia/reperfusion injury of the isolated rat heart. Administration of antithrombin had no effect on hemodynamic performance of the isolated heart, whereas treatment with 8 U/ml caused an augmented release of troponin I and creatine kinase compared to controls ( p < 0.05). In contrast to our findings, other groups observed positive effects of AT on ischemia/reperfusion injury. Pretreatment with AT before ischemia reduced neutrophil rolling and adhesion to preischemic levels during reperfusion in a feline mesentery. Interestingly, AT posttreatment also significantly reduced neutrophil rolling and adhesion during reperfusion. The binding of AT to thrombin is suggested as a mechanism for this effect [31]. AT has also been shown to diminish reperfusion injury in the ischemic intestine of rats. Pretreatment with AT reduced histopathological damage of the intestine and accumulation of neutrophils in the mucosa [15]. Harada et al. [17] described positive influences of AT on ischemia/reperfusion injury of the rat liver and an increased release of prostacyclin. While indomethacin inhibited all these effects, iloprost showed the same effects as AT. DEGR-FXa, a selective inhibitor of thrombin, did not influence ischemia/reperfusion injury. The reason for the increased tissue injury at high-dose administration of AT in our model is not clear. The AT concentration was chosen high in order to release PGI2 (1 U/ml is equivalent to about 100% activity in human blood). Such high concentrations of AT are not used in clinical practice. Therefore, it seems unlikely that AT treatment used to normalize plasma levels causes cardiac problems in patients. In addition, the administration of AT in a high dose significantly improved survival in sepsis models of animal, where no serious side effect like heart failure was reported. Because the isolated heart is perfused with a blood-free solution, it is difficult to assess functional consequences of the ischemia/reperfusion injury, which is mainly mediated by activated neutrophils. A connection between increased PGI2 release by AT and reduced ischemia/reperfusion injury was not observed because PGI2 levels were not corresponding to the tissue injury markers troponin I and creatine kinase. In conclusion, our study shows for the first time that AT does not increase prostacyclin release in the isolated bloodfree perfused rat heart after ischemia. AT at physiological concentrations has no influence on ischemia/reperfusion

injury. At very high concentration, it worsens tissue injury in the isolated rat heart. The mechanism remains unclear.

References [1] Emerson Jr TE, Fournel MA, Redens TB, Taylor Jr FB. Efficacy of antithrombin III supplementation in animal models of fulminant Escherichia coli endotoxemia or bacteremia. Am J Med 1989;87(3B): 27S – 33S. [2] Taylor Jr FB, Emerson Jr TE, Jordan R, Chang AK, Blick KE. Antithrombin-III prevents the lethal effects of Escherichia coli infusion in baboons. Circ Shock 1988;26(3):227 – 35. [3] Taylor Jr FB, Chang AC, Peer GT, et al. DEGR-factor Xa blocks disseminated intravascular coagulation initiated by Escherichia coli without preventing shock or organ damage. Blood 1991;78(2):364 – 8. [4] Eisele B, Lamy M, Thijs LG, et al. Antithrombin III in patients with severe sepsis. A randomized, placebo-controlled, double-blind multicenter trial plus a meta-analysis on all randomized, placebo-controlled, double-blind trials with antithrombin III in severe sepsis [see comments]. Intensive Care Med 1998;24(7):663 – 72. [5] Inthorn D, Hoffmann JN, Hartl WH, Muhlbayer D, Jochum M. Antithrombin III supplementation in severe sepsis: beneficial effects on organ dysfunction. Shock 1997;8(5):328 – 34. [6] Baudo F, Caimi TM, de Cataldo F, et al. Antithrombin III (ATIII) replacement therapy in patients with sepsis and/or postsurgical complications: a controlled double-blind, randomized, multicenter study [see comments]. Intensive Care Med 1998;24(4):336 – 42. [7] Horie S, Ishii H, Kazama M. Heparin-like glycosaminoglycan is a receptor for antithrombin III-dependent but not for thrombin-dependent prostacyclin production in human endothelial cells. Thromb Res 1990;59(6):895 – 904. [8] Yamauchi T, Umeda F, Inoguchi T, Nawata H. Antithrombin III stimulates prostacyclin production by cultured aortic endothelial cells. Biochem Biophys Res Commun 1989;163(3):1404 – 11. [9] Uchiba M, Okajima K, Murakami K, Okabe H, Takatsuki K. Effects of antithrombin III (AT III) and Trp49-modified AT III on plasma level of 6-keto-PGF1 alpha in rats. Thromb Res 1995;80(3):201 – 8. [10] Eisenhut T, Sinha B, Grottrup-Wolfers E, et al. Prostacyclin analogs suppress the synthesis of tumor necrosis factor-alpha in LPS-stimulated human peripheral blood mononuclear cells. Immunopharmacology 1993;26(3):259 – 64. [11] Kainoh M, Imai R, Umetsu T, Hattori M, Nishio S. Prostacyclin and beraprost sodium as suppressors of activated rat polymorphonuclear leukocytes. Biochem Pharmacol 1990;39(3):477 – 84. [12] Riva CM, Morganroth ML, Ljungman AG, et al. Iloprost inhibits neutrophil-induced lung injury and neutrophil adherence to endothelial monolayers. Am J Respir Cell Mol Biol 1990;3(4):301 – 9. [13] Uchiba M, Okajima K, Murakami K, Okabe H, Takatsuki K. Attenuation of endotoxin-induced pulmonary vascular injury by antithrombin III. Am J Physiol 1996;270(6 Pt 1):L921 – 30. [14] Dschietzig T, Alexiou K, Laule M, et al. Stimulation of pulmonary big endothelin-1 and endothelin-1 by antithrombin III: a rationale for combined application of antithrombin III and endothelin antagonists in sepsis-related acute respiratory distress syndrome? Crit Care Med 2000;28(7):2445 – 9. [15] Ozden A, Tetik C, Bilgihan A, et al. Antithrombin III prevents 60 min warm intestinal ischemia reperfusion injury in rats. Res Exp Med (Berl) 1999;198(5):237 – 46. [16] Okano K, Kokudo Y, Okajima K, et al. Protective effects of antithrombin III supplementation on warm ischemia and reperfusion injury in rat liver. World J Surg 1996;20(8):1069 – 75. [17] Harada N, Okajima K, Kushimoto S, Isobe H, Tanaka K. Antithrombin reduces ischemia/reperfusion injury of rat liver by increasing the hepatic level of prostacyclin. Blood 1999;93(1):157 – 64. [18] Lorant DE, Patel KD, McIntyre TM, et al. Coexpression of GMP-140

J. Margreiter et al. / Thrombosis Research 108 (2003) 249–255

[19]

[20]

[21]

[22]

[23]

[24]

[25]

and PAF by endothelium stimulated by histamine or thrombin: a juxtacrine system for adhesion and activation of neutrophils. J Cell Biol 1991;115(1):223 – 34. Inthorn D, Hoffmann JN, Hartl WH, Muhlbayer D, Jochum M. Effect of antithrombin III supplementation on inflammatory response in patients with severe sepsis. Shock 1998;10(2):90 – 6. Souter PJ, Thomas S, Hubbard AR, et al. Antithrombin inhibits lipopolysaccharide-induced tissue factor and interleukin-6 production by mononuclear cells, human umbilical vein endothelial cells, and whole blood. Crit Care Med 2001;29(1):134 – 9. Dunzendorfer S, Kaneider N, Rabensteiner A, et al. Cell-surface heparan sulfate proteoglycan-mediated regulation of human neutrophil migration by the serpin antithrombin III. Blood 2001;97(4):1079 – 85. Uchiba M, Okajima K, Murakami K. Effects of various doses of antithrombin III on endotoxin-induced endothelial cell injury and coagulation abnormalities in rats. Thromb Res 1998;89(5):233 – 41. Schobersberger W, Lercher A, Kountchev J, et al. High antithrombin III concentrations release prostacyclin in human dermal microvascular cells (HDMEC). Intensive Care Med 2000;26(Suppl 3):301. Engels W, Van Bilsen M, De Groot MJ, et al. Ischemia and reperfusion induced formation of eicosanoids in isolated rat hearts. Am J Physiol 1990;258(6 Pt 2):H1865 – 71. Karmazyn M. Contribution of prostaglandins to reperfusion-in-

[26]

[27]

[28]

[29]

[30] [31]

255

duced ventricular failure in isolated rat hearts. Am J Physiol 1986; 251(1 Pt 2):H133 – 40. Karmazyn M, Tani M, Neely JR. Effect of prostaglandins I2 (prostacyclin) and F2 alpha on function, energy metabolism, and calcium uptake in ischaemic/reperfused hearts. Cardiovasc Res 1993;27(3): 396 – 402. Karmazyn M, Neely JR. Inhibition of post-ischemic ventricular recovery by low concentrations of prostacyclin in isolated working rat hearts: dependency on concentration, ischemia duration, calcium and relationship to myocardial energy metabolism. J Mol Cell Cardiol 1989;21(3):335 – 46. Schror K, Woditsch I. Endogenous prostacyclin preserves myocardial function and endothelium-derived nitric oxide formation in myocardial ischemia. Agents Actions Suppl 1992;37(3):312 – 9. Woditsch I, Schror K. Prostacyclin rather than endogenous nitric oxide is a tissue protective factor in myocardial ischemia. Am J Physiol 1992;263(5 Pt 2):H1390 – 6. Schror K, Hohlfeld T. Inotropic actions of eicosanoids [editorial]. Basic Res Cardiol 1992;87(1):2 – 11. Ostrovsky L, Woodman RC, Payne D, Teoh D, Kubes P. Antithrombin III prevents and rapidly reverses leukocyte recruitment in ischemia/reperfusion. Circulation 1997;96(7):2302 – 10.