Brazilian Journal of Medical and Biological Research (2006) 39: 1549-1562 Myocardial proteome profile in Chagas' disease 1549 ISSN 0100-879X Proteomic inventory of myocardial proteins from patients with chronic Chagas’ cardiomyopathy P.C. Teixeira1,3,4, L.K. Iwai1,4, 1Laboratório de Imunologia, 2Divisão de Cirurgia Torácica, Instituto do Coração, A.C.K. Kuramoto1,4, Hospital das Clínicas, 3Disciplina de Imunologia Clínica e Alergia, R. Honorato2, A. Fiorelli2, Faculdade de Medicina, Universidade de São Paulo, São Paulo, SP, Brasil N. Stolf2, J. Kalil1,3,4 4Instituto de Investigação em Imunologia, Instituto do Milênio, CNPq/MCT, and E. Cunha-Neto1,3,4 São Paulo, SP, Brasil Abstract Correspondence Chronic Chagas’ disease cardiomyopathy (CCC) is an often fatal Key words E. Cunha-Neto outcome of Trypanosoma cruzi infection, with a poorer prognosis • Chagas’ disease Laboratório de Imunologia than other cardiomyopathies. CCC is refractory to heart failure treat- • Cardiomyopathy InCor-HC-FM, USP ments, and is the major indication of heart transplantation in Latin • Proteomic analysis Av. Dr. Eneas C. Aguiar, 44 America. A diffuse myocarditis, plus intense myocardial hypertrophy, • Two-dimensional Bloco II, 9º andar damage and fibrosis, in the presence of very few T. cruzi forms, are the electrophoresis 05403-000 São Paulo, SP • MALDI-ToF Brasil histopathological hallmarks of CCC. To gain a better understanding of Fax: +55-11-3069-5953 the pathophysiology of CCC, we analyzed the protein profile in the E-mail: [email protected] affected CCC myocardium. Homogenates from left ventricular myo- cardial samples of end-stage CCC hearts explanted during heart Research supported by FAPESP transplantation were subjected to two-dimensional electrophoresis (Nos. 00/14549-4 and 01/00729-3). with Coomassie blue staining; protein identification was performed P.C. Teixeira is supported by a FAPESP grant (No. 04/15322-4), and by MALDI-ToF mass spectrometry and peptide mass fingerprinting. E. Cunha-Neto and J. Kalil are The identification of selected proteins was confirmed by immunoblot- recipients of productivity awards ting. We demonstrated that 246 proteins matched in gels from two from CNPq (Nos. 302970-2004-5 and CCC patients. They corresponded to 112 distinct proteins. Along with 307541-2003-7). structural/contractile and metabolism proteins, we also identified proteins involved in apoptosis (caspase 8, caspase 2), immune system (T cell receptor ß chain, granzyme A, HLA class I) and stress pro- cesses (heat shock proteins, superoxide dismutases, and other oxida- Received December 20, 2005 Accepted August 23, 2006 tive stress proteins). Proteins involved in cell signaling and transcrip- tional factors were also identified. The identification of caspases and oxidative stress proteins suggests the occurrence of active apoptosis and significant oxidative stress in CCC myocardium. These results generated an inventory of myocardial proteins in CCC that should contribute to the generation of hypothesis-driven experiments de- signed on the basis of the classes of proteins identified here. Braz J Med Biol Res 39(12) 2006 1550 P.C. Teixeira et al. Introduction evaluation of large-scale protein profiles (13). In order to study the global protein pro- Chagas’ disease is a significant cause of file of the myocardium under the effect of morbidity and mortality in Central and South chronic inflammation in CCC heart tissue, America, affecting about 13 million people and to gain insight about the pathophysiol- (1). The disease is caused by infection with the ogy of CCC, we analyzed the myocardial intracellular protozoan parasite Trypanosoma proteome from end-stage CCC patients. cruzi. About 30% of Chagas’ disease patients develop chronic Chagas’ disease cardiomy- Patients, Material and Methods opathy (CCC), an inflammatory cardiomyop- athy that occurs decades after the initial infec- Reagents tion, and one-third progress further to a par- ticularly aggressive, life-threatening dilated Immobilized pH gradient gel buffer, tryp- cardiomyopathy. In spite of the recent ad- sin (sequencing grade), α-cyano-4-hydroxy- vances in vector control, the millions of pa- cinnamic acid and other analytical grade tients already infected deserve more attention reagents were purchased from Amersham from the scientific community. Furthermore, Biosciences (Uppsala, Sweden), with the clinical progression, length of survival and exception of ammonium bicarbonate, ana- overall prognosis are significantly worse in lytical grade acetonitrile and trifluoroacetic CCC patients compared with patients with acid (TFA) that were obtained from Merck dilated cardiomyopathy of non-inflammatory (Darmstadt, Germany). ACTH peptide 18- etiology (2,3). 39 and (Ile7)-angiotensin III were obtained CCC heart lesions show histopathologi- from Sigma (St. Louis, MO, USA). All stock cal findings consistent with inflammation solutions were prepared with deionized wa- and a myocardial remodeling process: T cell/ ter using a Milli-Q Academic System (Milli- macrophage-rich myocarditis, hypertrophy, pore Co., Billerica, MA, USA). and fibrosis with heart fiber damage (4). The local cytokine production profile is consist- Sample preparation ent with a T1-type response, with interferon- gamma (γ)-induced chemokines (5-12). Gene The protocol was approved by the Insti- expression profiling of CCC myocardial tis- tutional Review Board of the University of sues showed that 15% of the known genes São Paulo School of Medicine and written selectively up-regulated in CCC are IFN-γ- informed consent was obtained from the inducible (12). Exposure of cardiomyocytes patients. to IFN-γ can up-regulate the expression of Myocardial left ventricular free wall heart atrial natriuretic factor (12), suggesting that samples were obtained from two end-stage IFN-γ may directly modulate gene expres- heart failure CCC patients with a diagnosis sion in myocardial cells. These results are established by seropositivity to T. cruzi in at consistent with a possible role of IFN-γ- least two of three tests, i.e., ELISA, indirect induced chronic inflammation in modulat- immunofluorescence and indirect hemag- ing myocardial gene expression. glutination, and by positive epidemiology. However, gene expression profiling may Both patients were female and were 42 and not be faithfully reflected at the protein level. 62 years old, with an ejection fraction <40%. Advances in two-dimensional (2-D) electro- The hearts were explanted on the occasion phoresis, mass spectrometry, and bioinfor- of heart transplantation at the Heart Institute matics, along with progress in genomic se- - InCor, University of São Paulo School of quence analysis now make possible a direct Medicine, São Paulo, SP, Brazil. Samples Braz J Med Biol Res 39(12) 2006 Myocardial proteome profile in Chagas' disease 1551 were dissected, frozen in liquid nitrogen and on the basis of total volume of all spots in stored at -80ºC. The tissue (100 mg) was each gel. homogenized in 1% SDS (w/v) and 0.5 mM TLCK (1 mL), submitted to three sonication Protein digestion cycles and centrifuged at 12,000 g for 15 min at 4ºC (14). The samples were passed All spots visualized in each gel were through Millipore Centricon YM-3 filters excised, subjected to tryptic digestion and (molecular mass cutoff at 3000 Da) and the processed in a robotic workstation (Ettan protein content of the supernatant solution Spot Handling Workstation, Amersham Bio- of each extract was quantified by the bicin- sciences). Briefly, spots (1.4 mm in diam- choninic acid method (15). eter) were excised from the gel and then washed in 50 mM ammonium bicarbonate Two-dimensional gel electrophoresis (NH4HCO3) in 50% acetonitrile (ACN) until complete destaining and SDS removal and Isoelectric focusing was carried out on then dried. Dehydrated gel plugs were rehy- 24-cm long immobilized pH gradient gel drated with 30 µL of a solution containing strips, with a pH range of 3.0 to 10.0 (Im 0.3 µg trypsin in 20 mM NH4HCO3. Diges- mobiline DryStrip gels, Amersham Bio- tion was performed at 37oC for 6 h. Digested sciences). The strips were rehydrated in 8 M peptides were extracted by the addition of 50 urea, 0.5% CHAPS, 0.5% IPG buffer, 0.2% µL 50% ACN/0.5% TFA at room tempera- DTT, and traces of bromophenol blue con- ture for 1 h. The procedure was repeated and taining 1 mg of cardiac protein extract for 12 the supernatants were combined. Samples h at 20ºC. Isoelectric focusing was performed were concentrated and resuspended in 5 µL using the Ettan IPGphor Isoelectric Focus- 5% ACN/2.5% TFA. ing System (Amersham Biosciences) at 20ºC at 50 mA/strip in an increasing voltage gra- Protein identification dient (500 V for 1 h, 1000 V for 1 h, and 8000 V for 8 h, accumulating a total of Tryptic fragments were analyzed with 64,000 Vh), according to manufacturer in- the Ettan MALDI-ToF Pro mass spectrom- structions. Before the second dimension elec- eter (Amersham Biosciences). For peptide trophoresis, the strips were equilibrated twice mass fingerprinting (PMF) analysis, a small for 15 min in a buffer containing 50 mM quantity of each processed sample (0.5 µL) Tris, pH 8.8, 6 M urea, 30% glycerol (v/v), corresponding to a spot was applied on a 2% SDS (w/v), 0.002% bromophenol blue metal slide mixed with the same quantity of (w/v), and in the presence of 10 mg/mL DTT matrix (α-cyano-4-hydroxycinnamic acid) in the first step and 25 mg/mL iodoacet- and taken to the mass spectrometer for anal- amide in the second step. The second dimen- ysis. To ensure maximal accuracy, trypsin sion was carried out on a vertical DaltSix autolysis peptides (842.51 and 2211.10 Da) system (Amersham Biosciences) at 20ºC on were used for internal mass calibration and 12.5% polyacrylamide gels (24 x 18 cm) for (Ile7)-angiotensin III (897.53 Da) and ACTH 5 h at 100 W. Coomassie blue-stained gels peptide 18-39 (2465.20 Da) were used for were digitized with an ImageScanner and external calibration of the instrument.