Ann Thorac Surg 1999;67:652-656
© 1999 The Society of Thoracic Surgeons
Original Articles
Coronary artery surgery is associated with different forms of atherogenic lipoprotein modifications
Marjan Jahangiri, FRCSa,
Iren B. Kovacs, FRCPathb,
Colin D. Ridler, PhDb,
Gareth M. Rees, FRCSa,
Peter Gorog, FRCPathb
a Department of Cardiothoracic Surgery, William Harvey Research Institute, St. Bartholomew’s Hospital, London, England, UK
b Department of Pathopharmacology, William Harvey Research Institute, St. Bartholomew’s Hospital, London, England, UK
Accepted for publication August 3, 1998.
Address reprint requests to Dr Jahangiri, Department of Cardiac Surgery, Royal Brompton Hospital, London, SW36NP, England
 |
Abstract
|
|---|
Background. Increasing evidence shows that thrombogenicity and atherogenicity of lipoproteins are related to modifications involving oxidative, enzymatic, or physical alterations of these molecules. Findings on lipid peroxidation associated with cardiopulmonary bypass are conflicting, and the possible other forms of atherogenic lipid modification are unknown. The various forms of lipoprotein modifications including lipid peroxidation, desialylation, and leukocytic elastase activity after coronary artery bypass graft operations are examined.
Methods. In patients undergoing coronary artery bypass graft operations, plasma total lipid hydroperoxides (n = 102), plasma leukocytic elastase activity (n = 125), free radical formation (n = 30), low-density lipoprotein oxidation, and sialic acid content before operation and at 2, 24, 48, and 72 hours after cardiopulmonary bypass and 3 months after operation were measured.
Results. Preoperative plasma lipid peroxide concentration (2.2 µmol/L) increased after cardiopulmonary bypass (peak, 7.5 µmol/L; p < 0.001) and remained significantly elevated at 3 months after surgery (4.2 µmol/L; p < 0.01). There was a significant correlation between increased free radical generation and lipid peroxide levels in blood at all postoperative intervals. Low-density lipoprotein separated from plasma samples showed increased oxidation 48 hours after bypass. Sialic acid content of low-density lipoprotein was significantly reduced 48 hours after bypass. Plasma elastase activity increased significantly at all postoperative intervals.
Conclusions. Coronary artery bypass graft operation is associated with generation of sustained blood levels of modified lipoproteins. These thrombogenic and atherogenic particles may play an important role in hemostatic and arteriosclerotic complications of coronary artery bypass graft operations.
|
Introduction
|
|---|
Increasing evidence shows that atherogenicity of lipids and lipoproteins is related to modifications involving oxidative, enzymatic, or physical alterations of these molecules [1]. Modified low-density lipoproteins (LDL) are thrombogenic by activating platelets and interfering with the fibrinolytic system [2, 3]. Recent findings show that oxidized LDL are a potent stimulant of both growth and migration of arterial smooth muscle cells [4, 5], the key events of restenosis [6]. We have recently provided evidence that lipid peroxide levels in patients with coronary artery disease are significantly elevated [7].
Several groups have investigated free radical formation and lipid peroxidation during and after coronary artery bypass graft operation (CABG), but their findings are controversial. Although some studies documented an increase in plasma lipid hydroperoxides, reaching a peak at 4 to 6 hours after cardiopulmonary bypass after which it rapidly returned to the baseline level [8], others could not find evidence for such lipid peroxidation [9]. Similarly, although myocardial antioxidant level was reduced during CABG, indicating free radical generation in the ischemic heart, the antioxidant capacity of plasma during cardiopulmonary bypass did not decrease but increased [10].
Most of these contradictions are related to methodology. Difficulties in measuring short-lived free radicals encourage indirect assessment through changes in antioxidants [11]. However, plasma contains an array of antioxidants and even the assessed "total antioxidant capacity" will not detect subtle ongoing oxidative processes. The thiobarbituric acid assay, the most commonly used test for measuring lipid peroxides in plasma, lacks specificity. In particular, release of malondialdehyde from the heart during cardiopulmonary bypass renders this test unsuitable for assessing lipid peroxidation in the systemic circulation during CABG [12].
Leukocytic elastase can also modify lipoproteins [13], and leukocytic elastase protein has been shown to be increased during CABG [9, 14]. Furthermore, desialylation (enzymatic reduction of sialic acid content) was also shown to cause atherogenic modification of LDL particles [15].
The aim of the present study was to assess the overall lipoprotein modification during and after CABG. Highly specific assays were used to measure generation of oxidants in blood, lipid peroxides both in plasma and in the isolated LDL fraction, plasma elastase activity rather than antigen levels, and sialic acid content of isolated LDL before and after CABG.
|
Material and methods
|
|---|
One hundred twenty-five patients undergoing elective CABG were studied. Informed consent that had ethical approval by the ethics committee was obtained. Plasma total lipid hydroperoxides (n = 102), plasma leukocytic elastase activity (n = 125), and free radical formation (n = 30) before operation and at 2, 24, 48, and 72 hours after cardiopulmonary bypass and 3 months after operation were measured. Low-density lipoprotein oxidation and sialic acid content were measured in 10 patients before operation and at 48 hours after bypass.
The median age was 59 years (range, 42 to 71 years) with a male to female ratio of 2.2:1. Anesthesia was induced using benzodiazepine, narcotic, and relaxant sequence and was maintained with further intravenous or inhalational anesthetics as required. Flow rates of 2.0 to 2.5 L · min-1 · m-2 to provide perfusion pressures of 50 to 75 mm Hg were maintained. An infusion of nitroglycerine was used to control blood pressure. Myocardial protection was provided by cold antegrade crystalloid cardioplegic solution with topical cooling and systemic hypothermia at 28°C. None of the patients received corticosteroids, aprotinin, or inotropic support in the perioperative period. The patients were extubated at a mean time of 3.5 hours (range 1 to 9 hours) after operation. Venous blood samples were obtained at 2, 24, 48, and 72 hours after cardiopulmonary bypass. A further sample was obtained at 3 months after operation. Blood samples anticoagulated with ethylendiaminetetraacetic acid (for lipid peroxide measurement) or citrate (for elastase measurement) were centrifuged at 12,500 g for 3 minutes within 15 minutes of withdrawal. Plasma samples were stored at 4°C for 48 hours until the assays were performed.
Lipid hydroperoxides
The technique of measuring lipid hydroperoxides from human plasma has been described in detail [7]. Liposorb gel was prepared freshly by adding 0.24 mL citrated saline (3.2% citrate in 0.9% NaCl) to 20 mg PHM-L-Liposorb (Calbiochem, UK) in a light-protecting, amber-colored 2-mL microcentrifuge tube. The contents were mixed and equilibrated for 5 minutes. Five hundred-microliter plasma samples were added to the freshly prepared liposorb gels, mixed, and incubated by rotation of the tubes at room temperature for 15 minutes to allow adsorption of all plasma lipids and lipoproteins to the gel. Gels were washed twice with 0.9% NaCl and centrifuged, the supernatants were discarded, and the gels were resuspended in 1.2 mL CHOD-iodide reagent (BDH-Merck, UK). The contents were incubated by rotating the tubes at room temperature for 60 minutes followed by centrifugation at 12,000 g for 5 minutes. Optical densities of clear supernatants were measured at 405 nm. Plasma lipid hydroperoxide concentrations (micromolar) were calcualted by the molar absorptivity of 2.46 · mol/L-1 · cm-1.
Isolation of LDL
Low-density lipoprotein was separated from EDTA-anticoagulated plasma by density gradient (KBr) ultracentrifugation (Beckman NVT90 single vertical spin separation).
Oxidative modification of LDL
Oxidative modification of LDL was characterized by fluorescence spectroscopy with excitation and emission set to 355 nm and 430 nm, respectively [16]. Results are expressed in arbitrary units corresponding with the intensity of fluorescence (millimeter chart recording).
Sialic acid content of LDL
Sialic acid was released from LDL (1.0 mg protein/mL) by incubating the lipoprotein with neuraminidase (Clostridium perfingens, attached to beaded agarose, Sigma, 0.05 U/mL) at 37°C for 60 minutes. Contents were then added to freshly prepared PHM-L-Liposorb gels, incubated for 15 minutes at room temperature by rotation, and centrifuged at 12,000 g for 5 minutes. Centrifugation sediments both gel-bound LDL and insoluble neuraminidase, thus leaving the released free sialic acid in the supernatant, which was then incubated in sealed ampules with 3,5-diaminobenzoic acid at 90°C for 12 hours. Fluorescence of the product formed by sialic acid was measured at 436 nm excitation and 525 nm emission maximums [17].
Whole blood chemiluminescence
Luminol-amplified total oxidant activity of diluted whole blood samples was measured in a luminometer (model 1251, LKB-Wallace, Turku, Finland), as described earlier [18]. The peak spontaneous chemiluminescence (mV) reflects the generation of reactive oxygen species by activated neutrophils.
Elastase activity
Plasma elastase-type activity was measured by modification of described chromogenic assay conditions [19]. Plasma sample (0.2 mL) was incubated with a Tris buffer (1.8 mL, 0.1 mol/L, pH 8.5) at 40°C for 5 minutes before substrate (N-succinyl-Ala-Ala-Ala-p-nitroanilide, Sigma UK, dissolved in dimethyl sulfoxide, 1 mmol/L final concentration) was added to the mixture, and incubated for a further 15 minutes. The initial absorbency at 410 nm was determined against distilled water, then the mixture replaced into the original test tube and incubated for a further 24 hours at 40°C. From the changes of absorbency during 24 hours, elastase activity was expressed as units per liter. From the molar extinction coefficient of p-nitroanilide (8.8 cm2 · µmol), one unit was defined as the amount of enzyme activity that converts 1 µmole substrate per minute at 40°C. Specificity of the assay was tested by using N-methoxysuccinyl-Ala-Ala-Pro-Val chloromethyl ketone (MAAPV, 50 µmol/L, Sigma), a specific inhibitor of human neutrophil elastase. Because of significant hemodilution during cardiopulmonary bypass, both lipid peroxide and elastase measurements were corrected for hematocrit changes.
Statistical analysis
Data are expressed as the mean ± SEM. Individual means were compared using the unpaired Student’s t test. Values of p < 0.05 were considered significant. Correlation was calculated by the Pearson’s test.
|
Results
|
|---|
Plasma lipid peroxide concentrations before and after cardiopulmonary bypass are shown in Figure 1. The highest increase in plasma lipid hydroperoxide concentration occured shortly after bypass (at 2 hours), then it reached a second maximum at 48 hours. Nevertheless, even 72 hours after bypass the increase in plasma lipid hydroperoxides was highly significant. Patients tested at 3 months after CABG still had significantly raised plasma lipid hydroperoxide levels.
Whole blood spontaneous chemiluminescence reflects production of reactive oxygen species mainly by activated neutrophils. Spontaneous chemiluminescence of whole blood greatly increased shortly after bypass operation and remained raised up to 3 days. There was a significant positive correlation between chemiluminescence and plasma lipid hydroperoxide levels (Fig 2).
View larger version (26K):
[in this window]
[in a new window]
|
Fig 2. Spontaneous chemiluminescence of whole blood after coronary artery bypass graft operation. Correlation coefficients were calculated between the corresponding data in Figures 1 and 2.
|
|
The measured plasma elastase-type activity was strongly inhibited by MAAPV, a specific inhibitor of human neutrophil elastase (93% ± 2% inhibition, mean ± SEM, n = 5). Elastase activities in plasma before and after CABG are shown in Figure 3. Elastase activity increased only slightly at 2 hours after the cardiopulmonary bypass, but a significant increase occurred later, reaching a maximum at 72 hours after bypass.
Increased fluorescence of plasma LDL at excitation and emission maximums of 355 nm and 430 nm reflects formation of crosslinks and development of fluorophores owing to oxidation of lipids in the lipoprotein molecule. Fluorescence of LDL separated from plasma samples 48 hours after operation increased significantly compared with fluorescence before the operation (Fig 4). In 7 of 10 subjects fluorescence increased considerably, in 1 subject the increase was slight, and in 2 subjects fluorescence after operation did not change or slightly decreased.
View larger version (22K):
[in this window]
[in a new window]
|
Fig 4. Fluorescence of plasma LDL before and after coronary artery bypass graft operation. (Individual data connected.)
|
|
Sialic acid content of LDL isolated from plasma decreased significantly (p < 0.01) at 48 hours after bypass operation compared with levels before the operation. In 7 of 10 subjects, the reduction in sialic acid content of LDL was considerable (Fig 5).
View larger version (19K):
[in this window]
[in a new window]
|
Fig 5. Sialic acid content of plasma LDL before and after coronary artery bypass graft operation. (Individual data connected.)
|
|
|
Comment
|
|---|
Our present findings show a greatly increased production of reactive oxygen species in blood, a highly significant increase in plasma concentration of both lipid hydroperoxides and elastase-type activity after CABG operation. It is assumed that both the increased elastase activity [13] and desialylation [15] bring about nonoxidative modification of plasma lipoproteins, particularly LDL, with increased atherogenicity.
Intravascular activation of neutrophils, monocytes, and platelets during CABG is well documented [20]. Activation of these cells is probably initiated by contact of blood with the large artificial surfaces of the cardiopulmonary bypass circuit. Activated neutrophils and monocytes release reactive oxygen species, which cause peroxidation of plasma lipoproteins. This study confirms that leukocyte activation occurs during cardiopulmonary bypass by detecting a highly increased whole blood chemiluminescence. Besides generating reactive oxygen metabolites, neutrophil and monocyte activation result in the release of granular contents, including the proteolytic enzyme elastase, into the serum. In contrast to peroxidation of LDL lipids during oxidation, elastase causes fragmentation of the apoprotein apoB-100. By proteolysis and the subsequent degradation of the apoprotein, elastase can modify LDL, which is then taken up by macrophages and causes foam cell formation, similar to oxidized LDL [13]. Under physiologic conditions, an effective control mechanism prevents the released elastase from exerting undesirable effects, ie, destruction of normal tissues. The most important natural protection against neutrophil elastase is the 
1-protease inhibitor, which is normally present in plasma in large quantities. Under physiologic conditions, elastase protein concentration in plasma always represents circulating elastase complexed with 1-protease inhibitor. However, conditions created during cardiopulmonary bypass can make this protective mechanism less efficient. Reactive oxygen species generated from leukocytes during cardiopulmonary bypass can locally inactivate 1-protease inhibitor, thus promoting the released elastase activity [21]. Furthermore, heparin, which is required in high plasma concentrations to provide adequate anticoagulation during bypass, strongly decreases the rate of inhibition of neutrophil elastase by 1-protease inhibitor [22]. Elastase activity measurement from plasma samples probably underestimates the real increase in activity in the circulation. It has been shown that stimulated human leukocytes express persistently active leukocyte elastase on their surface membranes [23]. Such cell-surface bound human leukocyte elastase is catalytically active, yet remarkably resistant to inhibition by naturally occurring protease inhibitors. Leukocyte-attached elastase activity has not been assessed in this study.
An important issue is the atherogenic potential of the measured plasma (modified) lipoprotein levels. The in vitro atherogenic effect of minimally modified LDL (stimulation of monocyte–endothelial interaction, proliferation of vascular smooth muscle cells) were demonstrated at hydroperoxide concentrations of 3 to 6 µmol/L [7]. Thus, plasma lipid hydroperoxide levels after CABG should be considered atherogenic and thrombogenic with possible pathologic significance. There is also evidence that desialylation of LDL of such magnitude as observed in this study is highly atherogenic [15] and thrombogenic [3].
Although a link between perioperative generation of modified LDL and early (thrombotic) graft occlusion is likely, it is difficult to appreciate how the modified lipoproteins generated perioperatively could contribute to restenosis, which occurs several months or years after operation. However, the possibility that perioperative changes set off a chain of events resulting in accelerated arteriosclerosis cannot be excluded. The time course of events after experimental arterial injury with a balloon catheter clearly shows a definite link between events taking place shortly after the vascular injury and restenosis that occurs much later [24]. Activation, migration, and proliferation of medial smooth muscle cells are critical events in the subsequent restenosis [6]. Immediately after arterial injury, multiple factors lead to the activation of medial smooth muscle cells, and proliferation of these cells is evident 24 hours after balloon injury. We have recently shown that by nitric oxide synthesis, intact endothelium exerts a powerful inhibitory effect on migration of smooth muscle cells. Oxidized LDL in a concentration comparable to that measured in the plasma samples in this study completely suspended the inhibitory effect of endothelium and greatly increased migration of smooth muscle cells from the media into the intima [7]. Despite the important early events, plaque development, lumen narrowing, and occlusion are delayed by arterial remodeling, ie, the compensatory enlargement of the lumen. It is not unreasonable therefore to speculate that by stimulating smooth muscle proliferation and migration, perioperatively generated modified LDL particles may contribute not just to the acute but also to late occlusion after CABG.
In conclusion, plasma lipids and lipoproteins undergo different forms of modification during and after CABG operation. It is suggested that the lasting presence of these highly atherogenic particles in the circulation may play an important role in both early and late reocclusion. The lack of a negative correlation between the oxidative and antioxidative capacity of plasma may be partly owing to the differing efficacy of antioxidant molecules. Furthermore, it may be that oxidative damage to LDL does not occur purely in the plasma and that the intima has a comparably overwhelming antioxidant shield.
|
Acknowledgments
|
|---|
We thank Stephen J. Edmondson, FRCS from St. Bartholomew’s Hospital, and Patrick G. Magee, FRCS and John E. Wright, FRCS from the London Chest Hospital for allowing us to include their patients in this study. Marjan Jahangiri, FRCS was supported by the Joint Research Board of St. Bartholomew’s Hospital and the British Heart Foundation.
|
References
|
|---|
-
Witztum J.L. The oxidation hypothesis of atherosclerosis. Lancet 1994;344:793-795.[Medline]
-
Chen L.Y., Mehta J.L. Oxidized LDL decreases L-arginine uptake and nitric oxide synthase protein expression in human platelets. Relevance of the effect of oxidized LDL on platelet function. Circulation 1996;93:1740-1746.[Abstract/Free Full Text]
-
Pekelharing H.L.M., Kleinveld H.A., Duif P.F., Bouma B.N., van Rijn H.J.M. Modification of Lp(a) by oxidation or desialylation influences its ability to compete with plasminogen for binding to the extracellular matrix. Blood Coagul Fibrinolysis 1996;7:641-649.[Medline]
-
Bjorkerud B., Bjorkerud S. Contrary effects of lightly and strongly oxidized LDL with potent promotion of growth versus apoptosis on arterial smooth muscle cells, macrophages, and fibroblasts. Arterioscler Thromb Vasc Biol 1996;16:416-424.[Abstract/Free Full Text]
-
Gorog P. Modified LDL is a potent stimulus for smooth muscle cell outgrowth from rat aortic explant in vitro. Atherosclerosis 1997;129:1-7.[Medline]
-
Casscells W. Migration of smooth muscle and endothelial cells: critical events in restenosis. Circulation 1992;86:723-729.[Free Full Text]
-
Kovacs I.B., Jahangiri M., Rees G.M., Gorog P. Elevated plasma lipid hydroperoxide levels in patients with coronary artery disease. Am Heart J 1997;134:572-576.[Medline]
-
Davies S.W., Duffy J.P., Wickens D.G., et al. Time-course of free radical activity during coronary artery operations with cardiopulmonary bypass. J Thorac Cardiovasc Surg 1993;105:979-987.[Abstract]
-
Hind C.R.K., Griffin J.F., Pack S. Effect of cardiopulmonary bypass on circulating concentrations of leucocyte elastase and free radical activity. Cardiovasc Res 1988;22:37-41.[Medline]
-
Toivonen H.J., Ahotupa M. Free radical reaction products and antioxidant capacity in arterial plasma during coronary artery bypass grafting. J Thorac Cardiovasc Surg 1994;108:140-147.[Abstract/Free Full Text]
-
Barsacchi R., Pelosi G., Maffei S., et al. Myocardial vitamin E is consumed during cardiopulmonary bypass: indirect evidence of free radical generation in human ischaemic heart. Int J Cardiol 1992;37:339-343.[Medline]
-
Lazzarino G., Raatikainen P., Nuutinen M., et al. Myocardial release of malondialdehyde and purine compounds during coronary bypass surgery. Circulation 1994;90:291-297.[Abstract/Free Full Text]
-
Polacek D., Byrne R.E., Scanu A.M. Modification of low density lipoproteins by polymorphonuclear cell elastase leads to enhanced uptake by human monocyte-derived macrophages via the low density lipoprotein receptor pathway. J Lipid Res 1988;29:797-808.[Abstract]
-
Butler J., Pillai R., Rocker G.M., Westaby M.S., Parker D., Shale D.J. Effect of cardiopulmonary bypass on systemic release of neutrophil elastase and tumor necrosis factor. J Thorac Cardiovasc Surg 1993;105:25-30.[Abstract]
-
Orekhov A.N., Tertov V.V., Sobenin I.A., et al. Sialic acid content of human low density lipoproteins affects their interaction with cell receptors and intracellular lipid accumulation. J Lipid Res 1992;33:805-817.[Abstract]
-
Koller E., Quehenberger O., Jörgens G., Wolbeis O.S., Esterbauer H. Investigation of human low density lipoprotein by three-dimensional fluorescence spectroscopy. FEBS Lett 1986;198:229-234.[Medline]
-
Gorog P., Pearson J.D. Surface determinants of low density lipoprotein uptake by endothelial cells. Atherosclerosis 1984;53:21-29.[Medline]
-
Kovacs I.B., Meyrick-Thomas R.H., Mackay A.R., Rustin M.H.A., Kirby J.D.T. Increased chemiluminescence of polymorphonuclear leucocytes from patients with progressive systemic sclerosis. Clin Sci 1986;70:257-261.[Medline]
-
Franck C., Byrjalsen I. A simple and sensitive chromogen assay for the measurement of elastase-type enzyme activity in human plasma. Biol Chem Hoppe-Seyler 1988;369:677-682.[Medline]
-
Dreyer W.J., Michael L.H., Millman E.E., Berens K.L. Neutrophil activation and adhesion molecule expression in a canine model of open heart surgery with cardiopulmonary bypass. Cardiovasc Res 1995;29:775-781.[Medline]
-
Mohsenin V. Lipid peroxidation and antielastase activity in the lung under oxidant stress: role of antioxidant defenses. J Appl Physiol 1991;70:1456-1462.[Abstract/Free Full Text]
-
Frommherz K.J. Heparin decreases the rate of inhibition of neutrophil elastase by alpha-1-Pl. Biol Chem Hoppe Seyler 1992;373:503-508.[Medline]
-
Owen C.A., Campbell M.A., Sannes P.L., Boukedes S.S., Campbell S.J. Cell surface-bound elastase and cathepsin G on human neutrophils: novel, non-oxidative mechanism by which neutrophils focus and preserve activity of serine proteinases. J Cell Biol 1995;131:775-789.[Abstract/Free Full Text]
-
Bauters C., Meurice T., Hamon M., McFadden E., Lablanche J.M., Bertrand M.E. Mechanisms and prevention of restenosis: from experimental models to clinical practice. Cardiovasc Res 1996;31:835-846.[Medline]
This article has been cited by other articles:
|
|
|
|
 
J. T Christenson and J. T Christenson
Preoperative Cholesterol and Thrombotic Complications After Coronary Bypass
Asian Cardiovasc Thorac Ann,
December 1, 2000;
8(4):
315 - 321.
[Abstract]
[Full Text]
[PDF]
|
|
|
Top
Abstract
Introduction
