|
|
||||||||
Ann Thorac Surg 2005;79:917-923
© 2005 The Society of Thoracic Surgeons
a Department of Medicine, Nordland Hospital, Bodø and University of Tromsø, Tromsø, Norway
b Department of Immunology and Transfusion Medicine, Nordland Hospital, Bodø and University of Tromsø, Tromsø, Norway
c Department of Medical Biochemistry, Nordland Hospital, Bodø and University of Tromsø, Tromsø, Norway
d Carmeda AB, Stockholm, Sweden
e Laboratory of Protein Chemistry, Department of Pathology and Laboratory Medicine, University of Pennsylvania, Philadelphia, Pennsylvania
f Institute of Immunology, Rikshospitalet University Hospital, Oslo, Norway
Accepted for publication August 10, 2004.
* Address reprint requests to Dr Lappegård, Department of Medicine, Nordland Hospital, N-8092 Bodø, Norway (E-mail: knut.lappegard{at}nlsh.no).
| Abstract |
|---|
|
|
|---|
METHODS: Human whole blood was incubated in rotating loops of polyvinyl chloride or heparin-coated polyvinyl chloride tubing for 4 hours. Plasma concentrations of the eicosanoids leukotriene B4, prostaglandin E2 and thromboxane B2 were quantified.
RESULTS: Polyvinyl chloride induced a substantial increase in leukotriene B4, prostaglandin E2, and thromboxane B2. Inhibition of complement activation by the complement factor 3 binding peptide compstatin or blockade of the complement factor 5a receptor with a specific antagonist significantly and specifically inhibited the synthesis of leukotriene B4, whereas thromboxane B2 and prostaglandin E2 synthesis were apparently complement independent. The increase in all three mediators was significantly reduced by the heparin coating. Indomethacin abolished the increase of the cyclooxygenase products prostaglandin E2 and thromboxane B2, but had no effect on the increase of the lipoxygenase product leukotriene B4, consistent with the specificity of indomethacin for the cyclooxygenase and confirming the specificity of complement inhibition.
CONCLUSIONS: Polyvinyl chloride-induced increase in all three eicosanoids was attenuated by heparin coating, whereas complement inhibition selectively reduced the synthesis of leukotriene B4.
| Introduction |
|---|
|
|
|---|
| Dr Riesenfeld discloses that he has a financial relationship with Carmeda AB.
|
Exposure of blood to artificial surfaces in medical devices, including extracorporeal circuit systems, leads to a reaction between the blood constituents and the surface. Polyvinyl chloride (PVC) is a frequently used artificial surface material in such devices. Clinical cardiopulmonary bypass (CPB) induces a systemic inflammatory response involving activation of both leukocytes, platelets, and plasma cascade systems [1, 2]. There are several contributors to these reactions; eg, the artificial surface itself, the membrane oxygenator, the surgical trauma, and the ischemia-reperfusion reaction. The relative contribution of these factors is only partly understood. It is well known, both from in vitro and in vivo experiments, that modification of the artificial surface by endpoint attachment of heparin attenuates several of the observed inflammatory responses [35]. Similar in vivo observations have been made with inhibitors of the complement system [6, 7].
Complement activation is responsible for a number of the inflammatory reactions taking place during CPB, and the complement inhibitory properties of the heparin coating may account for many of the beneficial effects of this surface. However, in an in vitro model, we have recently shown that various leukocyte responses, including expression of surface markers and synthesis of cytokines, differ in their dependence on complement and that surface modification with covalently attached heparin attenuates both complement-dependent and complement-independent reactions [8, 9]. The model is reductionistic in its nature as it aims to restrict the factors involved to the surface only and is thus not readily comparable to a clinical CPB setting. This should be kept in mind when comparing our results with those of previous studies.
Knowledge regarding the synthesis and release of eicosanoids in clinical CPB or in vitro models is scarce. To our knowledge, only two studies have investigated the effect of heparin coating on eicosanoids in simulated CPB [10] or in vivo [11], whereas one group has compared uncoated and heparin-coated cardiac catheters [12]. All three studies were restricted to thromboxane B2 (TXB2). The effect of complement inhibition on PVC-induced eicosanoids has not been studied. Since eicosanoids play important roles in regulation and propagation of inflammatory processes [13] and are the common target of several antiinflammatory drugs [14], it is important to obtain information on artificial surface-induced eicosanoid synthesis and how to inhibit such a reaction. Thus, the aim of the present study was to investigate the role of complement and heparin coating in the PVC-induced synthesis of eicosanoids.
| Material and Methods |
|---|
|
|
|---|
Complement Inhibitors
The cyclic hexapeptide AcF[OPdChaWR], a complement factor 5a receptor antagonist (C5aRA) [15] was synthesized as previously described [16]. Details of the synthesis, purification, mass spectrometry, and measurement of inhibitory activities of the compstatin analogue Ac-I[CVWQDWGAHRC]T-NH2 are discussed elsewhere [17]. This peptide, an inhibitor of the C3 convertase [18, 19], is 45 times more active than the parent peptide I[CVVQDWGHHRC]T-NH2. Both complement inhibitors were synthesized in the laboratory of one of the authors (JDL).
Experimental Model
The in vitro model for circulating whole blood through PVC loops has previously been described in detail [20]. The method was modified on the critical point of anticoagulation. Blood was drawn from healthy volunteers using lepirudin, a recombinant form of hirudin, instead of heparin as anticoagulant. Hirudin is a highly specific thrombin inhibitor and has no effect on the complement system. This is in contrast to heparin, which can either potentiate or attenuate complement activation, depending on the concentration used [21]. With regard to anticoagulation, lepirudin is equally effective as heparin. Fresh samples of blood were supplied with specific complement inhibitors (C5aRA 1 or 5 µmol/L, compstatin 5 or 25 µmol/L), indomethacin (2 or 10 µmol/L), or equal volumes of PBS. A volume of 750 µL blood was then transferred to segments of PVC or H-PVC tubing (length 30 cm, internal diameter 3 mm). The total volume of the tubing was 2,100 µL, and approximately 1/3 of the tubing was filled with blood, the remainder being air as to ascertain that the blood was actually circulating. Each segment was closed end-to-end and incubated by rotating slowly in an incubator at 37°C for 4 hours, based on pilot experiments designed for optimal conditions for incubation time. In previous experiments we have found that pH in the blood is stable during this period [21]. The blood was processed immediately after collection and in order to reduce delay in handling, the experimental setup did not allow for testing of all the different inhibitors in the same experiment. However, baseline values as well as blood circulated in PVC tubing without inhibitor and blood circulated in H-PVC tubing were included in all experiments. All inhibitors were tested in blood from 6 to 8 different donors. Since baseline values as well as values from PVC tubing and H-PVC tubing were determined in every experiment, each donor served as his or her own control. After incubation, the blood was centrifuged at 3,220 x g for 15 minutes at 4°C. The plasma was frozen in aliquots at 70°C and stored until all samples were analyzed in one batch.
Enzyme Immunoassays
Assays for prostaglandin E2, thromboxane B2, and leukotriene B4 were purchased from Cayman Chemical Company (Ann Arbor, MI). Activation of the terminal complement pathway was determined in an enzyme immunoassay using the monoclonal antibody aE11 specific for a C9 neoepitope in the sC5b-9 complex using a modification of an assay described in detail previously [22].
Statistics
Wilcoxon's test for paired observations was used for comparison within groups. Statistical significance was defined as p less than 0.05 (two-tailed). Baseline values for the various eicosanoids, as well as control PVC and H-PVC loops were included in every experimental setup and thus represent a larger number of experiments than each of the inhibitors tested.
| Results |
|---|
|
|
|---|
|
|
|
Effect of Complement Inhibition on PVC-Induced Synthesis of Eicosanoids
LEUKOTRIENE B4
Complement inhibition markedly suppressed the PVC-induced increase in LTB4 (Fig 1). Blocking the C3 convertase with compstatin 5 or 25 µmol/L reduced mean LTB4 from 1,232 pg/mL to 838 pg/mL (not significant) and 598 pg/mL (p < 0.02), respectively. Blocking the C5a receptor with 1 or 5 µmol/L C5a receptor antagonist reduced mean LTB4 to 482 pg/mL (p < 0.05) and 465 pg/mL (p < 0.01), respectively.
PROSTAGLANDIN E2
Complement inhibition did not affect PGE2 levels significantly, except for 5 µmol/L of the C5a receptor antagonist, which gave a reduction from 2,557 pg/mL to 1,385 pg/mL (p < 0.01) (Fig 2).
THROMBOXANE B2
Complement inhibition did not affect artificial surface-induced synthesis of thromboxane B2 (Fig 3).
PVC-Induced Complement Activation
The terminal complement complex (TCC) increased significantly (p < 0.001) in the PVC tubing, confirming that uncoated PVC activates the complement system. This activation was markedly lower (p < 0.01) in H-PVC, consistent with previous data that this heparin coating efficiently attenuates complement activation. Compstatin reduced TCC generation in a dose-dependent manner, at the highest concentration tested the effect was comparable to that observed for H-PVC (p < 0.01 for both concentrations). As expected, the C5aRA had no effect on TCC formation since the antagonist blocks the receptor but does not interfere with the cleavage of C5 into C5a and C5b, or the subsequent assembly of TCC (Fig 4).
|
| Comment |
|---|
|
|
|---|
Circulation of blood in PVC tubing induces complement activation primarily through the alternative pathway [8]. Increased levels of complexes between C1-inhibitor and its substrates, eg, C1s and kallikrein, has been demonstrated during in vivo as well as simulated extracorporeal circulation [24]. We have previously demonstrated increased C1rs-C1-inhibitor complexes by the heparin-coated surface in the present model [8]. This increase was not followed by C4 activation and therefore not indicative of classic pathway activation, but rather reflected the potentiation of heparin on C1-inhibitor binding to its substrate [25]. The effect of heparin on C1-inhibitor is thus operative both for solid-phase and fluid-phase heparin. A similar effect of heparin has been described for the plasma contact activation system, where antithrombin bound to a heparinized surface was functionally accelerated by the heparin molecules and efficiently neutralized activated factor XII [26]. In general, therefore, increases in complexes between C1-inhibitor and its substrates should be interpreted with caution. To indicate that classic complement activation occurs, a combined increase in C1rs-C1-inhibitor complexes and C4 activation products should be documented.
Anticoagulation may directly affect complement, either by activation (eg, by low doses of heparin) or by inhibition, as seen with calcium binders and high doses of heparin. Thus, conflicting results from different experimental models may in part be due to the use of different anticoagulants. We have recently shown that lepirudin, in doses sufficient to achieve anticoagulation, has no effect on complement activation [21], and we suggest that lepirudin at present is the best anticoagulant available for such studies. Lepirudin has also been employed by others, both in whole blood models of inflammation [27] and in animal models of CPB [28].
It has previously been shown that heparin coating of the artificial surface markedly reduces the inflammatory response in blood [29], including complement activation [5]. However, it is largely unknown to what extent the complement inhibitory properties of the heparin coating are responsible for the antiinflammatory effects of this surface.
The dependency on complement of the different inflammatory reactions can be determined by the use of specific complement inhibitors. We have recently reported that certain inflammatory responses induced by PVC are strictly complement-dependent, some are partly dependent on complement, whereas others are totally complement-independent. For instance, formation of granulocyte-platelet conjugates, but not formation of monocyte-platelet conjugates could be attenuated by complement inhibition [8]. Furthermore, formation of IL-8 is totally complement-dependent in this model, whereas formation of monocyte chemoattractant protein 1 is only partially reduced by complement inhibition [9].
The eicosanoids are important mediators of inflammation [13]. The leukotrienes (eg, LTB4) are synthesized by the 5-lipoxygenase, whereas the cyclooxygenase is responsible for the synthesis of the prostaglandins (eg, PGE2) and thromboxanes (eg, TXB2). A connection between complement and eicosanoid production has been shown in several animal models of eicosanoid-mediated injury. In a model of rat membranous nephropathy it was shown that C5b-9 upregulated the cyclooxygenase-2 [30], and complement activation products induced synthesis of LTB4 [31]. Furthermore, in a rabbit model of dermal inflammation it was shown that C5a recruits leukocytes to the site of inflammation through an LTB4-dependent mechanism [32]. A complex interaction between complement and LTB4 is emphasized by the fact that a specific antagonist against LTB4 blocked LTB4-mediated, but not C5a-mediated neutrophil activation, as measured by release of calcium. Thus, complement-mediated leukocyte activation may be both LTB4-dependent and independent [33].
The role of complement in PVC-induced eicosanoid production has not previously been investigated. In the present study we demonstrate that PVC-induced LTB4 synthesis to a large extent is complement-dependent since it was dose dependently reduced by the complement inhibitors. However, LTB4 was not completely complement-dependent since it was more efficiently reduced by the heparin coating than by complement inhibition. In contrast to LTB4, the production of PGE2 and TXB2 was virtually complement-independent, although the highest concentration of the C5a receptor antagonist reduced the levels of PGE2 to a certain degree but markedly less than the effect of complement inhibition on LTB4 production. However, the results obtained with indomethacin confirm that inhibition of PGE2 and TXB2 production in our model could be achieved, further emphasizing that the effect of complement inhibition on LTB4 production indicates substantial differences in modes of stimulation for the various eicosanoids. The LTB4 generated in our model is most likely a product of leukocytes [34, 35], whereas TXB2 and PGE2 originate from platelet [36] and monocyte [37] arachidonic acid. Consistently, we have previously shown that release of thrombospondin, a major constituent of platelet alpha granules, is unaffected by complement inhibition in this model [8]. Furthermore, in a previous study [9] we showed that synthesis of IL-8 in this model is totally C5a dependent. The present results show that blocking the C5aR markedly reduces PVC-induced LTB4 synthesis. It has previously been shown that IL-8 can stimulate human neutrophil 5-lipoxygenase and increase formation of LTB4 [38]. Collectively, these data suggest that there is complex interaction between complement, chemokines, and eicosanoids in the inflammatory reaction induced by PVC.
The heparin coating essentially eliminated both the PVC-induced formation of LTB4, as well as PVC-induced complement activation, as judged by the formation of TCC. However, the fact that the same effect was observed for release of PGE2 and TXB2 shows that the heparin coating prevents eicosanoid synthesis through a mechanism that is independent of complement. The beneficial properties of the heparin coating as compared to uncoated PVC have previously been documented for several factors, including complement activation [5], synthesis of chemokines [9], leukocyte CD11b-expression [8], and neutrophil and platelet degranulation [8]. In the present study we clearly show that this also applies to synthesis of LTB4, PGE2, and TXB2. Thus, the heparin coating makes the artificial surface highly biocompatible, and this may account for the beneficial results seen when used in clinical cardiopulmonary bypass [39]. The platelet-compatible properties of the heparin coating may decrease the risks of microemboli in clinical settings, which is in line with previous observations that the use of heparin-coated surfaces can inhibit deterioration of mental function after coronary bypass [40]. There are not many clinical trials with complement inhibition in cardiopulmonary bypass, but in one study a monoclonal antibody against C5 reduced complications after coronary artery bypass graft [6].
In conclusion, the PVC surface induced a marked increase in the concentrations of LTB4, PGE2, and TXB2. The increase was almost completely inhibited by heparin coating of the surface, whereas complement inhibition specifically reduced LTB4 formation. Thus, the attenuating effects of the heparin coating on eicosanoid synthesis can only partly be explained by its complement inhibitory properties. As the model focuses on the interaction between blood and the artificial surface, and as a clinical situation involving CPB also contains numerous other inflammatory stimuli including the wound, direct extrapolations of our results to a clinical setting are not warranted.
| Acknowledgments |
|---|
|
|
|---|
| References |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
M. Saethre, M. K. J. Schneider, J. D. Lambris, P. Magotti, G. Haraldsen, J. D. Seebach, and T. E. Mollnes Cytokine Secretion Depends on Gal{alpha}(1,3)Gal Expression in a Pig-to-Human Whole Blood Model J. Immunol., May 1, 2008; 180(9): 6346 - 6353. [Abstract] [Full Text] [PDF] |
||||
![]() |
K. Ritis, M. Doumas, D. Mastellos, A. Micheli, S. Giaglis, P. Magotti, S. Rafail, G. Kartalis, P. Sideras, and J. D. Lambris A Novel C5a Receptor-Tissue Factor Cross-Talk in Neutrophils Links Innate Immunity to Coagulation Pathways J. Immunol., October 1, 2006; 177(7): 4794 - 4802. [Abstract] [Full Text] [PDF] |
||||
![]() |
D. J. Kozik and J. S. Tweddell Characterizing the Inflammatory Response to Cardiopulmonary Bypass in Children Ann. Thorac. Surg., June 1, 2006; 81(6): S2347 - S2354. [Abstract] [Full Text] [PDF] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| ANN THORAC SURG | ASIAN CARDIOVASC THORAC ANN | EUR J CARDIOTHORAC SURG |
| J THORAC CARDIOVASC SURG | ICVTS | ALL CTSNet JOURNALS |