Aprotinin is a serine protease inhibitor derived from bovine lung, which can decrease blood loss during and after cardiac surgery . In a post hoc analysis of 816 coronary artery bypass patients from a multi-centre study , aprotinin administration was associated with a lesser incidence of stroke (3.1% vs. 0.0%; P = 0.04). A meta-analysis of placebo-controlled, randomized, double-blind studies of coronary artery bypass patients receiving high-dose aprotinin or placebo also supports the hypothesis that aprotinin is neuroprotective in this setting .
Patients undergoing cardiac surgery demonstrate a marked, generalized inflammatory response . Based on animal investigations , it is likely that non-specific inflammation exacerbates the injury associated with focal cerebral ischaemia following microgaseous or macroatheromatous cerebral embolization, as occuring during cardiopulmonary bypass (CPB). Therapies aimed at preventing this inflammatory response are neuroprotective in experimental models of cerebral ischaemia . The use of heparin-coated circuits, which decrease the bypass-induced inflammatory response, is associated with a better neurological outcome following cardiac surgery .
To date, the anti-inflammatory effects of aprotinin have been demonstrated in clinical trials  and in in vitro models of neutrophil  and endothelial cell activation . The protective effect of aprotinin in ischaemia-reperfusion injury has been demonstrated in various tissues such as the brain (piglets and rats) and myocardium (isolated rat heart) [10-12]. The neuro-protective effect associated with aprotinin administration during coronary artery bypass grafting (CABG) may be due its anti-inflammatory actions.
Neutrophil emigration from the vasculature is governed by an orderly series of contact events between neutrophils and endothelium, involving adhesion molecules of the selectin, integrin and immunoglobulin superfamilies . Hypoxia-reoxygenation promotes neutrophil-endothelial interactions by upregulating expression of a number of neutrophil and endothelial cell adhesion molecules including neutrophil CD11b, CD18  and endothelial intercellular adhesion molecule-1 (ICAM-1) . Cytokine interleukin (IL)-1β plays a central role in the endothelial response to hypoxia-reoxygenation [16,17]. The effects of aprotinin on hypoxia-reoxygenation-induced changes in neutrophil and endothelial cell adhesion molecule expression have not been studied. This is relevant to understanding its neuroprotective effects. In the present in vitro study, the effects of aprotinin on neutrophil and endothelial cell adhesion molecule expression and endothelial IL-1β supernatant concentrations in response to hypoxia-reoxygenation were investigated.
With institutional ethical approval and informed consent from each, venous blood (30 mL) samples were obtained from healthy volunteers (n = 8). Freshly discarded human umbilical cords were obtained from the delivery suite of the Erinville Hospital, Cork. Volunteers were excluded if they had undergone major surgery within the past 6 months, were taking concurrent medication, or had an infection within the previous month.
Preparation of purified populations of neutrophils
The technique used for isolation of human neutrophils has been described previously . Human neutrophils were isolated by sequential sedimentation in 6% Dextran (molecular weight 520000, Sigma, UK) in 0.9% sodium chloride for 45 min at 22°C, centrifugation in Ficoll-Paque (Pharmacia LKB Biotechnology, Piscataway, NJ, USA) at 300g for 30 min to pellet granulocytes and remaining erythrocytes, and centrifugation of the resuspended pellet over an 81% isotonic Percoll (Sigma, UK) gradient at 350g for 15 min to pellet erythrocytes. The diffuse layer at the interface containing neutrophils was then harvested, washed, resuspended in medium and counted. Cell viability was assessed using trypan blue exclusion (Merck, Darmstadt, Germany). The proportion of neutrophils in the preparation was determined using Rapi-diff II (Diagnostic Developments, Lancashire, UK) staining on cytocentrifuged samples.
Preparation of hypoxic and hypoxic-reoxygenated neutrophils
The technique used for hypoxia-reoxygenation of isolated neutrophils has been described previously . Human neutrophils were preincubated for 30 min in the absence or presence of aprotinin (1600 KIU mL−1, a clinically relevant concentration ). Hanks balanced salt solution was sponged with nitrogen gas for 30 min, and the partial pressure of oxygen (PO2) was measured to assess oxygen depletion. PO2 and pH (7.3 ± 0.005) were measured in the cells suspended in solution before and after hypoxia and reoxygenation using a blood gas analyser (Eschweiler System C 2000, Keil, Germany). Neutrophils (1 × 106 mL−1) were incubated in oxygen-depleted Hanks balanced salt solution for 30 min at 37°C and PO2 was measured at the end of the incubation time (PO2 69.9 ± 2.6 mmHg, n = 30). Neutrophils were reoxygenated by suspending the cells in normoxic solution (PO2 120.9 ± 2.6 mmHg, n = 30) after centrifugation and removal of hypoxic medium, and PO2 was again measured (119 ± 2.4 mmHg, n = 30). Viability of the cells was measured at the end of the experiment by a trypan blue exclusion test, which was (96 ± 1.3%) for normoxic cells and was unchanged after hypoxia and hypoxia-reoxygenation. Eight independent experiments were performed.
Samples were taken from supernatants at four time points to assess neutrophil CD11b and CD18 expression.
T0: Prior to hypoxic stimulus.
T1: At end of hypoxic period (60 min) but prior to reoxygenation.
T2: Immediately on reoxygenation.
T3: Following 15 min of reoxygenation.
Expression of CD11b and CD18 on neutrophils
Two hundred μL aliquots of stimulated neutrophil suspension (1 × 106 cells mL−1) were used for assessment of adhesion molecule expression. Neutrophils were stained using monoclonal antibodies for CD11b and CD18 (Serotec, Kidlington, UK). Nonspecific binding was quantified by employing the respective isotype-matched negative control, and the background signal was subsequently subtracted. Neutrophils were gated by their characteristic light scatter profile. The mean channel fluorescence (MCF) intensity of stained neutrophils was detected on the basis of a minimum number of 5000 cells collected, analysed using the fluorescence-activated cell-sorter scanner (FACScan cytofluorometer) (Becton Dickinson, CA, USA). Due to considerable variation of CD11b expression depending on the neutrophil donor, the data were expressed as percent increase of the respective basal value (before incubation with conditioned medium).
Endothelial cell cultures
Human umbilical vein endothelial cells (HUVECs) from fresh placental cords were isolated by previously described methods  and grown until confluence at 37°C in humidified 5% CO2. The growth medium consisted of complete Medium 199 supplemented with 20% fetal calf serum, penicillin (100 U mL−1), streptomycin sulphate (100 μg mL−1), fungizone (0.25 μg mL−1), heparin (16 U mL−1), endothelial cell growth supplement (75 μg mL−1), and glutamine (2 mmol L−1). In all experiments, HUVECs were used as individual isolates between passage 3 and 5. At confluence, HUVECs were detached from the culture flask by trypsinization using 0.05% trypsin/0.02% ethylenediamine tetra-acetic acid and seeded out on fibronectin-coated polycarbonate filters bearing 3.0 μm pores size in Transwell culture plate inserts (Costar, Cambridge, MA, USA). Confluent endothelial monolayers with tight cell conjunctions were formed after 30 h at 37°C in humidified 5% CO2 in culture.
Preparation of hypoxic and hypoxic-reoxygenated HUVEC
Following culture, the medium was replaced with a combination of glucose-free Krebs solution and exposed to hypoxia in a modular incubation chamber (Billups-Rothenberg) flushed with a gas mixture (1% O2-5% CO2-94% N2) to purge it of atmospheric air. HUVECs were incubated in the presence or absence of aprotinin (1600 KIU mL−1) for 2h, and exposed to hypoxia as described above for 24h. Reoxygenation was induced by exposing cells to ambient air and by suspending cells in normoxic culture medium for 24h. Supernatants were collected at the end of this period. The expression of ICAM-1 at a single time point was analysed by FACScan cytofluorometer. Three (n = 3) independent experiments were performed.
One hundred μL of stimulated endothelial cell suspension (1 × 106 cells mL−1) was stained with 10 μL of fluorescein-isothiocyanate-conjugated anti-CD54 (anti-ICAM-1) mouse anti-human mAb (Serotec, Kidlington, UK) or 10 μL of fluorescein-isothiocyanate-conjugated isotype IgG1 control mAb and incubated for 30 min at 4°C. ICAM-1 expression on endothelial cells was analysed on a FACScan cytofluorometer. The MCF intensity of stained cells was detected on the basis of a minimum number of 5000 cells collected and analysed using the software Lysis II.
IL-1β supernatant concentrations were determined in cell media using enzyme-linked immunosorbent assays (ELISA, Quantikine R&D Systems, Europe Ltd., Lancashire, UK) according to manufacturer instructions. Concentrations were estimated at the end of hypoxia-reoxygenation. The sensitivity for IL-1β was 10 pg mL−1. The inter-and intra-assay precisions for IL-1β for the range of values obtained in this study are 4.1-8.4% and 2.8-5.4%, respectively.
The Sigma Stat 2.0 for windows (SPSS, Inc., Chicago, IL, USA) software package was used for all statistical analysis. Percent intensity of fluorescence compared to time 0 was calculated for neutrophil adhesion molecule expression. Absolute values (MCF or IL-1β concentrations) were compared in endothelial studies. Data obtained were normally distributed and analysed using two-way ANOVA (neutrophil studies) and unpaired t-test (endothelial cell studies). P < 0.05 was considered significant. Data is presented as mean ± SD.
CD11b and CD18
Exposure to 60-min hypoxia increased neutrophil CD11b expression compared to normoxia (170 ± 46% vs. 91 ± 27%, P = 0.001) (percent intensity of fluorescence compared to time 0) (n = 8) (Fig. 1a). Following hypoxia (60 min) the magnitude of increase in neutrophil CD11b expression was less in those treated with aprotinin [(129 ± 40% vs. 170 ± 46%) (P = 0.04)] (percent intensity of fluorescence compared to time 0) (n = 8) (Fig. 1a). CD18 expression did not change in response to hypoxia or hypoxia-reoxygenation (Fig. 1b).
Hypoxia-reoxygenation increased HUVEC ICAM-1 expression compared to normoxia (155 ± 3.7 vs. 43 ± 21 MCF, P = 0.0003) (n = 4) (Fig. 2). ICAM-1 expression was less in aprotinin pretreatment cells compared to control (116 ± 0.7 vs. 155 ± 3.3 MCF, P = 0.001) (n = 3) (Fig. 2).
Hypoxia-reoxygenation increased IL-1β endothelial supernatant concentrations compared to normoxia (3.4 ± 0.4 vs. 2.6 ± 0.2, P = 0.02) (n = 3) (Fig. 3). IL-1β endothelial supernatant concentrations were less in aprotinin pretreatment cells compared to control (2.6 ± 0.1 vs. 3.4 ± 0.3, P = 0.01) (n = 3) (Fig. 3).
Aprotinin (1600 KIU mL−1) diminishes the increase in CD11b expression in isolated human neutrophils and ICAM-1 expression on HUVECs in response hypoxia-reoxygenation. It also diminishes the increase of endothelial IL-1β concentrations in response to hypoxia-reoxygenation.
Our in vitro models of hypoxia-reoxygenation resulted in increased expression of neutrophil CD11b and endothelial cell ICAM-1 adhesion molecules. This is consistent with the findings of Scannell and colleagues (neutrophil CD11b)  and Mataki and colleagues (endothelial cell ICAM-1) . Contrary to the reports of Scannell and colleagues  our model of neutrophil hypoxia-reoxygenation did not result in increased neutrophil CD18 expression. This may due to relative hypoxia (PO2 69.9 ± 2.6 mmHg) in this study compared to that of Scannell and colleagues . In this study, aprotinin diminished the increase in neutrophil and endothelial cell adhesion molecule expression in response to in vitro hypoxia-reoxygenation. Aprotinin also diminished the increase in endothelial IL-1β concentrations in response to hypoxia-reoxygenation. Endothelial cell response to hypoxia-reoxygenation is mediated in part by IL-1β . The ability of IL-Ra to decrease ischaemic cerebral injury in vivo is probably in part, the result of both attenuated cerebral vascular endothelial activation via IL-1β produced by perivascular astrocytes and decreased neutrophil infiltration into the brain. Previously, therapies that targeted endothelial IL-1R1 or production/release of IL-1β by the glial cell compartment have been suggested as strategies to target post-ischaemic brain inflammation and secondary brain injury .
The aetiology of the inflammatory response associated with CPB surgery has been traced to the stress of surgery and contact activation of platelets and neutrophils (neutrophil) within the bypass circuit. Both of these stimuli result in an increase in plasma cytokine concentrations such as tumour necrosis factor-α, IL-1, IL-6 and IL-8 [23-25]. Inflammatory cytokines in turn cause endothelial cell activation and expression of adhesion molecules involved in recruitment of neutrophils to sites of inflammation or tissue injury. Acute inflammatory reaction associated with ischaemia-reperfusion contributes to the development of secondary brain injury [26,27]. Early in ischaemia and reperfusion molecular adhesive events and cytokine production occur and underlie the transition from ischaemic to inflammatory injury. The subsequent recruitment of neutrophils to the ischaemic zone may lead to reocclusion of microvessels . The same neutrophils also produce proteolytic enzymes, oxygen-free radicals, and other molecular effects, which, in addition to direct neuronal damage, may injure cerebrovascular endothelium .
Pruefer and colleagues using intravital microscopy technique have previously shown that aprotinin, in clinically relevant doses, inhibits neutrophil-endothelial cell interactions in the microvasculature during acute inflammatory events . This conclusion was based on the following observations:
• Aprotinin-inhibited thrombin-induced neutrophil-endothelium interaction in vivo.
• Systemic administration of aprotinin-inhibited neutrophil-endothelium interactions elicited by ischaemia-reperfusion.
• Aprotinin-attenuated cell-surface expression of P-selectin, an adhesion molecule that is important in the regulation of cell-to-cell interaction.
Experimental studies have shown that aprotinin, in addition to its antiproteolytic and membrane stabilizing properties, decreases the release of lysosomal enzymes and increases intracellular adenine nucleotides . Preservation of neutrophil ATP stores, though not confirmed in this study, may be a mechanism of diminishing increased neutrophil adhesion molecule expression associated with hypoxia. Hypoxia produces sustained rises in neutrophil intracellular Ca2+ and Ca2+-ATPase activity . Pretreatment of microsomes of pulmonary smooth muscle by oxidant increases protease activity, Ca2+-ATPase activity and ATP-dependent Ca2+ uptake which is diminished by aprotinin . Neutrophil adhesion molecule regulation is calcium dependent. Thus aprotinin may prevent increased neutrophil adhesion molecule expression in response to hypoxia by inhibition of a neutrophil protease. While aprotinin decreases reperfusion injury by suppressing bradykinin , it can also inhibit the production of superoxides and peroxides, which originate from human neutrophils [34,35].
Extrapolation of in vitro studies to complex clinical scenarios needs to be done with extreme caution. In this study we focused on a limited set of adhesion molecules and cytokines. Study limitations also included absence of determination of the mechanism of effect. Incubation of ischaemia-reperfusion media with isolated neutrophils can lead to artefactual upregulation of adhesion molecule expression from neutrophil isolation procedures . However, addition of ischaemia-reperfusion media results in a significant and stepwise increase in markers of neutrophil activation in both isolated neutrophils and whole blood as reported by Barry and colleagues . We chose to use a model similar to Scannell and colleagues  for neutrophil studies to allow comparison of results. Using aprotinin pretreatment in whole blood instead of treating isolated neutrophils may be more representative of plasma concentrations of aprotinin used as we do not know the proportion of aprotinin that is absorbed by erythrocytes preventing its action on neutrophils. Achievement of only relative hypoxia (PO2 69.9 ± 2.6 mmHg) in the neutrophil study arm represents a study limitation.
In conclusion, these data support the concept that the serine protease inhibitor aprotinin inhibits cell-surface expression of adhesion molecules (i.e. CD11b). This may be a key mechanism, besides the inhibition of complement activation, by which aprotinin could further inhibit neutrophil-endothelial interaction under inflammatory and ischaemia-reperfusion states. Therefore, perioperative administration of aprotinin may limit neutrophil and endothelial dysfunction in patients undergoing cardiac surgery resulting in a clinically important neuroprotective effect.
1. Alderman EL, Levy JH, Rich JB, et al.
Analyses of coronary graft patency after aprotinin
use: results from the International Multicenter Aprotinin
Graft Patency Experience (IMAGE) trial. J Thorac Cardiovasc Surg
2. Murkin JM. Attenuation of neurologic injury during cardiac surgery. Ann Thorac Surg
3. Paparella D, Yau TM, Young E. Cardiopulmonary bypass induced inflammation: pathophysiology and treatment. An update. Eur J Cardiothorac Surg
4. del Zoppo GJ, Schmid-Schonbein GW, Mori E, Copeland Br, Chang CM. Polymorphonuclear leukocytes occlude capillaries following middle cerebral artery occlusion and reperfusion in baboons. Stroke
5. Mori E, del Zoppo GJ, Chambers D, et al.
Inhibition of polymorphonuclear leukocyte adherence suppresses no-reflow after focal cerebral ischemia in baboons. Stroke
6. Jansen P, Baufreton C, Besnerais P, et al.
Heparin coated circuits and aprotinin
prime for coronary artery bypass grafting. Ann Thorac Surg
7. Lord RA, Roath OS, Thompson JF, Chant AD, Francis JL. Effect of aprotinin
function after major vascular surgery. Br J Surg
8. Wachtfogel YT, Kucich U, Hack CE, et al. Aprotinin
inhibits the contact, neutrophil
, and platelet activation systems during simulated extracorporeal perfusion. J Thorac Cardiovasc Surg
9. Asimakopoulos G, Lidington EA, Mason J, O'Haskard D, Taylor KM, Landis RC. Effect of aprotinin
on endothelial activation. J Thorac Cardiovasc Surg
10. Aoki M, Jonas RA, Nomura F, et al.
Effects of aprotinin
on acute recovery of cerebral metabolism in piglets after hypothermic circulatory arrest. Ann Thorac Surg
11. Kamiya T, Katayama Y, Kashiwagi F, Terashi A. The role of bradykinin in mediating ischaemic brain edema in rats. Stroke
12. Gurevitch J, Barak J, Hochhauser E, Paz Y, Yakirevich V. Aprotinin
improves myocardial recovery after ischemia and reperfusion. Effects of the drug on isolated rat hearts. J Thorac Cardiovac Surg
13. Carlos TM, Harlan JM. Leukocyte-endothelial cell adhesion molecules. Blood
14. Scannell G, Waxman K, Vazira ND, et al.
Hypoxia-induced alterations of neutrophil
membrane receptors. J Surg Res
15. Mataki H, Inagaki T, Yokoyama M, Maeda S. ICAM-1 expression and cellular injury in cultured endothelial cells under hypoxia/reoxygenation. Kobe J Med Sci
16. Clark ET, Desai TR, Hynes KL, Gewertz BL. Endothelial cell response to hypoxia-reoxygenation is mediated by IL-1. J Surg Res
17. Shreeniwas R, Koga S, Karakurum M, et al.
Hypoxia-mediated induction of endothelial cell interleukin-1A. An autocrine mechanism promoting expression of leukocyte adhesion molecules on vessel surface. J Clin Invest
18. Asimakopoulos G, Thompson R, Nourshargh S, et al.
An anti-inflammatory property of aprotinin
detected at the level of leukocyte extravasation. J Thorac Cardiovasc Surg
19. Sethi S, Singh MP, Dikshit M. Nitric oxide mediated augmentation of polmorphonuclear free radical generation following hypoxia-reoxygenation. Blood
20. Wellicome SM, Thornhill MH, Pitzalis C, et al.
A monoclonal antibody that detects a novel antigen on endothelial cells that is induced by tumor necrosis factor, IL-1, or lipopolysaccharide. J Immunol
21. Wang X, Feuerstein GZ. Induced expression of adhesion molecules following focal brain ischemia. J Neurotrauma
22. Zhang W, Smith C, Howlett C, Stanimirovic D. Inflammatory activation of human brain endothelial cells by hypoxic astrocytes in vitro
is mediated by IL-1β. J Cereb Blood Flow Metab
23. Lahat N, Zlotnick AY, Shtiller R, Bar I, Merin G. Serum levels of IL-1, IL-6 and tumor necrosis factor in patients undergoing coronary artery bypass grafts or cholecystectomy. Clin Exp Immunol
24. Butler J, Parker D, Pillai R, Westaby S, Shale DJ, Rocker GM. Effects of cardiopulmonary bypass on systemic release of neutrophil
elastase and tumor necrosis factor. J Thorac Cardiovasc Surg
25. Fujiwara T, Seo N, Murayama T, Hirata S, Kawahito K, Kawakami M. Transient rise in serum cytokines during coronary artery bypass graft surgery. Eur Cytokine Netw
26. Cole DJ, Patel PM, Schell RM, Drummond JC, Osborne TN. Brain ecosanoid levels during temporal focal cerebral ischaemia in rats: a microdialysis study. J Neurosurg Anesthesiol
27. Kontos MA. Oxygen radicals in cerebral ischaemia. In: Ginsberg MD, Dietrich WD, eds. Cerebovascular Diseases: 16th Princeton Conference.
New York: Raven Press, 1989: 365-372.
28. Ames A, Wright L, Masayoshi K, Thurston J, Majino G. Cerebral ischemia: The no-reflow phenomenon. Am J Pathol
29. Jean W, Spellman SR, Nussbaum ES, Low WC. Reperfusion injury after focal ischaemia: the role of inflammation and the therapeutic horizon. Neurosurgery
30. Pruefer D, Makowski J, Dahm M, et al. Aprotinin
inhibits leukocyte-endothelial cell interactions after hemorrhage and reperfusion. Ann Thorac Surg
31. Sunamori M, Sultan I, Suzuki A. Effect of aprotinin
to improve myocardial viability in myocardial preservation followed by reperfusion. Ann Thorac Surg
32. Scannell G. Leukocyte responses to hypoxic/ischemic conditions. New Horiz
33. Chakraborti T, Ghosh SK, Michael JR, Chakraborti S. Role of an aprotinin
-sensitive protease in the activation of Ca(2+)-ATPase by superoxide radical (O−2
) in microsomes of pulmonary vascular smooth muscle. Biochem J
34. Nagahiro I, White T, Yano M, et al.
Recombinant Kunitz protease inhibitor ameliorates reperfusion injury in rat lung transplantation. Ann Thorac Surg
35. Tamura K, Manabe T, Imanishi K, et al.
Effects of synthetic protease inhibitors on superoxide (O−2
), hydrogen peroxide (H2
) and hydroxyl radical production by human polymorphonuclear leukocytes. Hepato-gastroenterology
36. Fearon DT, Collins LA. Increased expression of C3b receptors on polymorphonuclear leukocytes induced by chemotactic factors and by purification procedures. J Immunol
37. Barry MC, Wang JH, Kelly CJ, et al.
Plasma factors augment neutrophil
and endothelial cell activation during aortic surgery. Eur J Vasc Endovasc Surg