Proteomic Analysis of Plasma from Patients Undergoing Coronary Artery Bypass Grafting Reveals a Protease/Antiprotease Imbalance in Favor of the Serpin r1-Antichymotrypsin Cristina Banfi,*,†,‡ Alessandro Parolari,§ Maura Brioschi,†,‡ Simona Barcella,†,‡ Claudia Loardi,§ Chiara Centenaro,† Francesco Alamanni,§ Luciana Mussoni,‡ and Elena Tremoli†,‡ Centro Cardiologico Monzino I.R.C.C.S., Department of Cardiac Surgery, Unit for Clinical Research in Atherothrombosis, Centro Cardiologico Monzino I.R.C.C.S, University of Milan, Italy, and Department of Pharmacological Sciences, University of Milan, Italy. Received November 24, 2009
We used proteomics to identify systematic changes in the plasma proteins of patients undergoing coronary artery bypass grafting (CABG) by means of cardiopulmonary bypass surgery. It is known that, after CABG, a complex systemic inflammatory responses ensues that favors the occurrence of adverse postoperative complications frequently recognizing inflammation itself and/or thrombosis as the underlying mechanism. We found a marked and persistent postoperative increase in the levels of the serin-protease inhibitor R1-antichymotrypsin (R1-ACT) that fully maintains the inhibitory activity blunting its protease substrate cathepsin G. An intraoperative increase followed by a rapid decline in proteases activation was documented, accompanied by a substantial induction of leucine-rich-R-2glycoprotein, a protein involved in neutrophilic granulocyte differentiation. Finally, a time-dependent alteration in the expression of haptoglobin, transthyretin, clusterin, and apoE was observed. In conclusion, we showed that after CABG, a protease/antiprotease imbalance occurs with early cathepsin G activation and a more delayed increase in R1-ACT. As cathepsin G is a serpin involved both in inflammation and coagulation activation, this confirms and expands the concept of a marked dysregulation of both inflammatory and hemostatic balances occurring after CABG. The pharmacologic modulation of this imbalance may be a new therapeutic target to reduce postoperative complications. Keywords: proteomics • coronary artery bypass grafting • proteases • protease inhibitor
Introduction Coronary artery bypass grafting (CABG) is one of the most frequently performed operations worldwide; the use of cardiopulmonary bypass (CPB) has, however, been associated with deleterious effects, such as central nervous system complications, coagulopathy, and a variable degree of end-organ dysfunction, that could be related to a protracted pro-inflammatory and pro-thrombotic response occurring after surgery.1,2 Over the last 50 years, a great effort to understand the physiology and pathophysiology of the response to CPB has led to steady improvements in patient management and to the reduction in postoperative morbidity.3 Several targeted strategies, using different drugs, modifying CPB components, or even abolishing CPB itself have been tested to attenuate the systemic inflammatory and pro-thrombotic response occurring after CABG,4 but none of these was able to blunt with efficacy the whole body activation occurring after this kind of surgery.5,6 * To whom correspondence should be addressed. Cristina Banfi, PhD Centro Cardiologico Monzino IRCCS, Via Parea 4, 20138 Milano MI, Italy. Tel: +39-0258002018. Fax: +39-0258002342. E-mail:
[email protected]. † Centro Cardiologico Monzino I.R.C.C.S. ‡ Department of Pharmacological Sciences, University of Milan. § Department of Cardiac Surgery, Unit for Clinical Research in Atherothrombosis, Centro Cardiologico Monzino I.R.C.C.S, University of Milan. 10.1021/pr901079v
2010 American Chemical Society
This, especially in patients with severe organ dysfunction or limited functional reserve, may become clinically significant, translating into major postoperative morbidity or even in mortality. Although a balanced and controlled inflammatory response is potentially beneficial as it aids host defenses against infection and facilitates wound healing, loss of control may herald the onset of a systemic inflammatory response and single or multiple organ dysfunction.7 However, not enough is known about the complex interactions of pro- and anti-inflammatory components, the significance of alterations in their magnitude or time course, or their relationship with the clinical sequelae of CPB. These paradigms are based on results of a variety of studies assessing the response of single proteins or genes to CABG. However, as the complex nature of the inflammatory/activation network includes a multitude of redundant but also antagonistic pathways, the changes in single proteins or genes cannot adequately describe the functional state of the inflammatory system. In an attempt to overcome this shortcoming, we adopted a proteomic approach that simultaneously assesses multiple plasma proteins to detect differential protein expression profiles during CABG. Journal of Proteome Research 2010, 9, 2347–2357 2347 Published on Web 03/19/2010
research articles
Banfi et al. a
Table 1. Clinical Characteristics of the Study Population Age (yrs.) Male gender BSA (m2) Hypertension Hypercholesterolemia Previous myocardial infarction Serum creatinine (mg/dL) Ejection fraction (%) No. of diseased vessels No. of distal anastomoses CPB time (min) Aortic cross-clamp time (min)
60 ( 4.5 5/5 (100%) 2.04 ( 0.04 5/5 (100%) 4/5 (80%) 2/5 (40%) 1.10 ( 0.08 60 ( 6.1 2.8 ( 0.2 3.6 ( 0.4 116 ( 7 77 ( 2.5
a Mean values ( SEM. BSA: body surface area; CPB, cardiopulmonary bypass.
Methods Patients. In accordance with a protocol approved by the Institutional Review Board of the Centro Cardiologico Monzino, we enrolled five first-time, isolated, low-risk (EuroSCORE 0.3 and a left ventricular end-diastolic pressure of 80 years, renal or liver disease, intake of drugs affecting platelet function, coagulation or fibrinolysis within the ten days preceding surgery, Q-wave myocardial infarction in the previous six weeks, and unstable angina; the intra- and postoperative exclusion criteria were excessive bleeding (>1000 mL/24 h), postoperative bleeding or re-exploration because of bleeding, or perioperative myocardial infarction, stroke or renal failure requiring dialysis. Conventional therapy was allowed on the basis of clinical judgment. All of the patients were hospitalised until the eighth postoperative day and attended a follow-up visit (physical examination, electrocardiogram, and blood collection) on the 30th postoperative day.1 Thirty-five consecutive patients with the same characteristics were also enrolled. Time of blood sampling and measurements are described in the Results section. Surgical Procedure. The patients received thiopentone 3-5 mg/kg and fentanyl 1 µg/kg to induce anesthesia, which was maintained by means of sufentanil boluses of up to 4-5 µg/kg associated with a continuous propofol infusion of 3 mg/kg/ hour. After orotracheal intubation, they were ventilated with oxygen and air (fraction of inspired oxygen: 50%), keeping the partial arterial pressure of carbon dioxide (Paco2) between 35 and 38 mmHg. Rectal and cervical esophageal probes were used to monitor temperature, and acid-base equilibrium was maintained by means of the alpha-stat method. After internal mammary takedown, they received systemic heparin (300 IU/kg bovine lung heparin), and anticoagulation was assessed on the basis of Celite-activated clotting time with a trigger level for additional heparin set at 440 s every 30 min during surgery. Upon the completion of the distal and proximal coronary anastomoses, heparin was antagonized with protamine sulfate at a 1:1 ratio (3 mg/kg), the protamine dose was based on the total heparin administered during surgery. None of the patients received aprotinin during or after surgery, and no patient cell-saver was used. 2348
Journal of Proteome Research • Vol. 9, No. 5, 2010
A nonpulsatile roller pump, hollow-fiber oxygenator with integrated heat exchanger, arterial filter, open cardiotomy reservoir, and polyvinyl tubing system were used in all cases. Each operation was performed under tepid hypothermia and hemodilution. Blood flow during surgery was kept at 2.4 L/min/ m2, and hematocrit at 18-25%. Myocardial protection was achieved by administering cold multidose blood cardioplegia infused through the aortic root and coronary sinus. Blood Sampling. Blood samples were collected six times in tubes containing citrate as anticoagulant (0.129 mmol/L): on the day before the induction of anesthesia (T0), five minutes after protamine administration (T1), and 1 (T2), 4 (T3), 8 (T4) and 30 days (T5) after surgery. Plasma was immediately prepared by means of centrifugation at 1500× g for 20 min at 4 °C, divided into aliquots, and frozen at -80 °C until assayed. Sample Preparation and 2-DE. To eliminate two of the most abundant plasma proteins,8 albumin and immunoglobulins, the samples were pretreated using a ProteoPrep Blue Albumin Depletion Kit (Sigma, Milan, Italy), after which 200 µg of plasma proteins (measured by means of the Bradford method) were diluted with a buffer to yield final concentrations of 7 mol/L urea, 2 mol/L thiourea, 0.2% w/v Sodium Dodecyl Sulfate (SDS), 4% w/v CHAPS, 2% v/v carrier ampholytes, pH 3-10, 20 mmol/L Tris, 55 mmol/L dithiothreitol, and bromophenol blue. IPG ready strips, 7 cm, pH 4-7 linear gradient (Biorad, Milan, Italy), were actively rehydrated at 50 V for 24 h and, after focusing, were first equilibrated for 15 min using a solution containing 50 mmol/L Tris-HCl, 6 mol/L urea, 30% v/v glycerol, 2% w/v SDS and 2% w/v dithiothreitol, and then with the same buffer containing 4.5% w/v iodoacetamide instead of dithiothreitol. The focused proteins were then separated according to size on 7-17% polyacrylamide gradient gels and stained with Coomassie Colloidal Blue G-250. Briefly, the gels were fixed with a fixing solution containing 40% v/v methanol and 10% v/v acetic acid, and stained overnight with a solution containing 0.12% w/v Coomassie Blue G-250, 8% w/v (NH4)2SO4, 1.6% v/v phosphoric acid, and 20% v/v methanol. They were then destained with a solution containing 25% v/v methanol. All of the images were scanned using a GS-800 densitomer (Biorad, Milan, Italy) before being analyzed by means of Progenesis SameSpot software (Nonlinear Dynamics, v 2.1). Mass Spectrometry. The protein spots of interest were manually excised from the gels and underwent in-gel digestion with trypsin as previously described.9 The samples were analyzed by means of LC-ESI-MS/MS, with the spectra being recorded by a Q-TOF spectrometer (Micromass, Manchester, UK) connected to a Micromass CapLC capillary chromatograph. The samples were dissolved in 0.1% formic acid, and aliquots were injected onto a 300 µm × 5 mm PepMap C18 column (LC Packings, Amsterdam, The Netherlands) and eluted with an acetonitrile, 0.1% formic acid gradient. The capillary voltage was set to 3500 V. A survey scan over the m/z range of 400-1300 was used to identify protonated peptides with charge states of 2, 3 or 4, which were automatically selected for datadependent MS/MS analysis and fragmented by collision with argon. The resulting product ion spectra were transformed onto a singly charged m/z axis using a maximum entropy method (MaxEnt3, Waters, Manchester, UK), and the proteins were identified by correlating the uninterpreted spectra with entries in Swiss-Prot/TrEMBL using ProteinLynx Global Server (Version 1.1, Waters, Manchester, UK). The database was created by merging the FASTA format files of Swiss-Prot/TrEMBL and their associated splice variants, without applying any taxonomic or
research articles
Coronary Artery Bypass Grafting Proteomic Analysis protein mass and pI constraints. Methionine oxidation was considered a variable modification, one missed cleavage per peptide was allowed, and the initial mass tolerance window was set to 100 ppm. In parallel, the spectra were also searched against the National Center for Biotechnology Information nonredundant (NCBI nr) database using Mascot (Matrix Science). Valid identification required two or more peptides independently matching the same protein sequence, a significant peptide score (typically >55, p < 0.05), and the manual confirmation of agreement between the spectra and peptide sequence. In addition, Mascot searches of all spectra were performed against a randomized version of the NCBI database using the same parameters as in the main search to obtain a Decoy score. Quantitative Analysis of Alpha-1-antichymotrypsin (r1ACT). The plasma concentrations of R1-ACT were determined by means of rocket immunoelectrophoresis on the basis of a quantitative estimate of the proteins undergoing electrophoresis in 1% agarose gel containing monospecific polyclonal rabbit antibody against R1-ACT (Dako Denmark A/S, Glostrup, Denmark) at a concentration of 0.93 µg per square cm gel area.10 The gels were pressed between filter papers and dried before being stained with Coomassie Blue, and R1-ACT was quantified by measuring the distance between the tip of the rocket-shaped immunoprecipitates and the application well. Standard curves were generated using serial dilutions of the Human Serum Protein Calibrator (Dako Denmark A/S, Glostrup, Denmark), which was run in parallel to the samples on every gel. The interand intra-assay percentage coefficients of variation (% CV) were 7.8% and 5.7%. Cathepsin G Activity. Cathepsin G activity was measured as previously described11 using N-Succinyl-L-alanyl-L-alanylL-prolyl-L-phenylalanine 4-nitroanilide (Sigma Aldrich, Milan, Italy) as a specific substrate in 25 mmol/L N-(2-hydroxyethyl)piperazine-N′-2-ethanesulfonic acid (HEPES, Sigma Aldrich, Milan, Italy), with 0.1% w/v polyoxyethylene (23) lauryl ether (Brij-35, Sigma Aldrich, Milan, Italy) as a buffer solution (pH 7.4). Dilution series (125, 62.5, 31.3, 15.7, and 7.8 U/L) of purified human neutrophil cathepsin G (Sigma Aldrich, Milan, Italy) were used as positive controls; and the substrate and buffer as negative controls. The samples or controls were applied to a MaxiSorp ELISA-plate (NUNC, Roskilde, Denmark), and a mixture of 10 µL N-Succinyl-L-alanyl-L-alanyl-L-prolylL-phenylalanine 4-nitroanilide (26 mmol/L in dimethyl sulfoxide) and 80 µL 25 mmol/L HEPES with 0.1% Brij-35 was added. The reaction was performed in humidified air at 37 °C. Absorbance at 405 nm was measured at 10 min intervals using a Microplate Reader (Mithras LB 940, Berthold Technologies GmbH & Co. KG, Bad Wildbad, Germany) with the correction of background values. r1-ACT Activity Assay. The ability of plasma R1-ACT to inhibit cathepsin G was evaluated by incubating plasma containing a known concentration of R1-ACT as measured by means of rocket immunoelectrophoresis with a stoichiometric amount of cathepsin G as previously described.12 Gel Electrophoresis and Immunoblotting Analysis. Immunoblotting was performed as previously described13 using the mouse monoclonal anti leucine-rich alpha-2-glycoprotein 1 (2E3) antibody (Novus Biologicals, Littleton, CO). Immunoenzymatic Assays. The ApoE4/Pan-ApoE ELISA kit came from Medical & Biological Laboratories Co. Ltd. (Japan); the human haptoglobin ELISA kit from Immunology Consultants Laboratory, Inc. (Newberg, OR); the prealbumin/tran-
sthyretin ELISA kit from Immundiagnostik AG (Bensheim, Germany); and the human clusterin ELISA kit from BioVendor (Modrice, Czech Republic). Active human neutrophil elastase (HNE) was measured using the InnoZyme HNE Immunocapture Activity Assay (Calbiochem, Darmstadt, Germany). Statistical Analysis. Each sample was analyzed by 2-DE in triplicate in order to evaluate gel reproducibility and improve the reliability of the qualitative and quantitative changes in protein expression measured by means of electrophoresis. Progenesis SameSpot software (v 2.1, NonLinear Dynamics) was used for gel alignment, spot detection, spot quantification, and normalization for total spot volume in each gel, and the data were statistically analyzed using the incorporated statistical package. The cutoff level for a differentially expressed protein was defined as at least a 1.5-fold increase or decrease in spot intensity. Statistically significant between-group differences for each protein were computed using analysis of variance (ANOVA) followed by Tukey’s post hoc test to allow both multiple group and individual group-to-group comparison; a p value of
