Troponin T autoantibodies correlate with chronic cardiomyopathy in human Chagas disease


  • Daniela Ferreira Nunes,

    1. Graduate Program in Health Sciences, Center for Health Sciences, Federal University of Rio Grande do Norte, Natal, Brazil
    Search for more papers by this author
  • Paulo Marcos da Matta Guedes,

    1. Department of Microbiology and Parasitology, Center for Biosciences, Federal University of Rio Grande do Norte, Natal, Brazil
    Search for more papers by this author
  • Cléber de Mesquita Andrade,

    1. Graduate Program in Health Sciences, Center for Health Sciences, Federal University of Rio Grande do Norte, Natal, Brazil
    Search for more papers by this author
  • Antonia Cláudia Jácome da Câmara,

    1. Department of Clinical and Toxicological Analyses, Center for Health Sciences, Federal University of Rio Grande do Norte, Natal, Brazil
    Search for more papers by this author
  • Egler Chiari,

    1. Department of Parasitology, Institute of Biological Sciences, Federal University of Minas Gerais, Belo Horizonte, Brazil
    Search for more papers by this author
  • Lúcia Maria da Cunha Galvão

    Corresponding author
    • Graduate Program in Health Sciences, Center for Health Sciences, Federal University of Rio Grande do Norte, Natal, Brazil
    Search for more papers by this author

Corresponding Author Lúcia M. C. Galvão, Programa de Pós-Graduação em Ciências da Saúde, Centro de Ciências da Saúde, Universidade Federal do Rio Grande do Norte (UFRN), Rua Gal. Gustavo Cordeiro de Farias s/n, Faculdade de Farmácia, 2º andar Petrópolis, 59012-570 Natal, RN, Brazil. Tel.: +55 84 3342 9827; Fax: +55 84 3342 9824; E-mail:



To evaluate the potential involvement of anti-Trypanosoma cruzi and cardiac protein antibody (IgG total and isotypes) production and their possible association with different clinical forms of human chronic Chagas disease.


IgG total and isotypes were measured by ELISA, using epimastigote and trypomastigote forms of T. cruzi as antigens and human cardiac proteins (myosin and troponin T) in sera of patients with indeterminate (IND, n = 72), cardiac (CARD, n = 47) and digestive/cardiodigestive (DIG/CARD-DIG, n = 12) clinical forms of the disease. Samples from uninfected health individuals (CONT, n = 30) and patients with ischaemic cardiomyopathy (ISCH, n = 15) were used as controls. Autoantibody levels were correlated with parameters of cardiac function obtained by electrocardiographic, radiographic and echocardiographic examinations.


Fifty five per cent of patients were classified as IND, 35.9% as CARD and 9.1% as DIG/CARD-DIG. Greater total IgG production was observed in IND, CARD and DIG/CARD-DIG chagasic patients than in CONT and ISCH, using trypomastigote, epimastigote and cardiac antigens. Moreover, patients with CARD and DIG/CARD-DIG presented greater total IgG production (trypomastigote and epimastigote antigen) than IND, and a negative correlation was determined between total IgG and left ventricular ejection fraction (LVEF). Patients with IND and CARD presented similar higher levels of total IgG specific to troponin T and myosin than CONT and ISCH individuals. Patients with chronic Chagas disease presented a negative correlation between left ventricular ejection fraction (LVEF) and the production of anti-myosin and troponin T autoantibodies. When grouped as low and high antibody producers and compared with LVEF, we observed that high anti-troponin T (= 0.042) and myosin (P = 0.013) producers presented lower LVEF than low producers. Moreover, there was a positive correlation (r = 0.9508, P = 0.0001) between the production of troponin T and myosin autoantibodies.


These findings indicate that increased production of anti-cardiac troponin T and myosin autoantibodies probably influences the left ventricular ejection fraction and could be related to chagasic cardiomyopathy.



Evaluer l'implication potentielle de la production des anticorps anti-Trypanosoma cruzi et ceux de la protéine cardiaque (IgG totale et isotypes) et leur éventuelle association avec les différentes formes cliniques de la maladie de Chagas humaine chronique.


Les IgG et isotypes totaux ont été mesurés par ELISA en utilisant les formes épimastigotes et trypomastigotes de T. cruzi comme antigènes et des protéines cardiaques humaines (myosine et troponine T) dans le sérum des patients atteints de formes cliniques de la maladie, indéterminées (IND, n = 72), cardiaques (CARD, n = 47) et digestives/cardio-digestives (DIG/CARD-DIG, n = 12). Des échantillons provenant de personnes non infectées (CONT, n = 30) et de patients atteints de cardiomyopathie ischémique (ISCH, n = 15) ont été utilisés comme témoins. Les taux d'auto-anticorps ont été corrélés avec les paramètres de la fonction cardiaque obtenus par l'examen électrocardiographique, radiologique et échographique.


55% des patients ont été classés comme IND, 35.9% comme CARD et 9.1% comme DIG/CARD-DIG. Des productions plus élevées d’IgG ont été observées chez les patients chagasiques IND, CARD et DIG/CARD-DIG que chez les patients CONT et ISCH, en utilisant les antigènes trypomastigotes, épimastigotes et cardiaques. De plus, les patients CARD et DIG/CARD-DIG présentaient une plus grande production d’IgG totales (antigènes trypomastigote et épimastigote) que les patients IND et une corrélation négative a été déterminée entre les IgG totales et la fraction d’éjection ventriculaire gauche (FEVG). Les patients IND et CARD présentaient des taux semblables plus élevés d'IgG totales spécifiques de la troponine T et de la myosine que les patients CONT et ISCH. Les patients atteints de la maladie de Chagas chronique présentaient une corrélation négative entre la FEVG et la production d'auto-anticorps anti-myosine et troponine T. Lorsque regroupés sous forme de producteurs faibles ou élevés d'anticorps et comparés avec la FEVG, nous avons observé que les producteurs élevés d'anti-troponine T (P = 0.042) et de myosine (P = 0.013) présentaient une FEVG inférieure à celle des producteurs faibles. En outre, il y avait une corrélation positive (r = 0.9508, P = 0.0001) entre la production d'auto-anticorps de troponine T et de myosine.


Ces résultats indiquent que la production élevée de troponine T anti-cardiaque et d'auto-anticorps de myosine influence probablement la FEVG et pourrait être liée à la cardiomyopathie chagasique.



Evaluar el posible papel de la producción de anticuerpos anti-Trypanosoma cruzi y anticuerpos frente a proteínas cardiacas (IgG total e isotipos) y su posible asociación con diferentes formas clínicas de la fase crónica de la enfermedad de Chagas en humanos.


Se midieron la IgG total e isotipos mediante ELISA, utilizando como antígenos las formas de epimastigotes y tripomastigotes de T. cruzi y proteínas cardiacas humanas (miosina y troponina T) en suero de pacientes con formas clínicas indeterminada (IND, n=72), cardiaca (CARD, n = 47) y digestiva/cardio-digestiva (DIG/CARD-DIG, n = 12) de la enfermedad. Se utilizaron como controles muestras de individuos sanos no infectados (CONT, n = 30) y pacientes con cardiomiopatía isquémica (CIS, n = 15). Los niveles de autoanticuerpos se correlacionaron con parámetros de función cardiaca obtenidos mediante pruebas electrocardiográficas, radiográficas y ecocardiográficas.


Un 55% de los pacientes fueron clasificados como IND, 35.9% como CARD y 9.1% como DIG/CARD-DIG. Se observó una mayor producción de IgG total en pacientes chagásicos IND, CARD y DIG/CARD-DIG que en pacientes CONT y CIS, utilizando antígenos de formas de tripomastigotes, epimastigotes y cardiacos. Más aún, los pacientes CARD y DIG/CARD-DIG presentaban una mayor producción de IgG total (antígeno de tripomastigote y epimastigote) que los IND, y se determinó una correlación negativa entre la IgG total y la fracción de eyección ventricular izquierda (FEVI). Los pacientes IND y CARD tenían niveles altos parecidos de IgG total específica para la troponina T y la miosina, comparados con los individuos CONT y CIS. Los pacientes con una enfermedad de Chagas crónica presentaban una correlación negativa entre la FEVI y la producción de autoanticuerpos anti-miosina y troponina T. Al agruparlos como productores de anticuerpos según el nivel de producción (alto y bajo) y compararlos con FEVI, observamos que los que producían niveles altos de anti-troponina T (= 0.042) y miosina (= 0.013) presentaban una menor FEVI que los que producían niveles bajos. Más aún, había una correlación positiva (r = 0.9508, = 0.0001) entre la producción de autoanticuerpos anti-troponina T y miosina.


Estos hallazgos indican que una mayor producción de los autoanticuerpos anti cardiacos troponina T y miosina, probablemente tienen influencia sobre la FEVI y podrían estar relacionados con la cardiomiopatía chagásica.


American trypanosomiasis is one of the most serious public health problems in the poorest endemic rural areas in Latin American, extending from the Southern United States to Southern Argentina and Chile. It is estimated that it currently affects 7.7 million people, and nearly 28 million remain at risk of acquiring the infection (Salvatella 2007). In 2008 alone, 11 000 deaths were registered (WHO 2010).

Natural transmission occurs through the deposition of faeces containing trypomastigote forms of the parasite on injured skin or intact mucosa by blood feeding triatomine bugs. Upon infection, an acute phase initiates that is usually asymptomatic and spontaneously resolves in about 90% of infected individuals even without treatment. The chronic phase occurs 4–8 weeks after the acute phase and is clinically divided into indeterminate, cardiac, digestive or cardiodigestive forms. The indeterminate form appears early in the chronic phase and is a latency period with reactive serology and/or demonstration of the parasite in the blood and the absence of symptomology. The majority of individuals remain free from clinical alterations in the indeterminate form; however, after a period 10–30 years, the other clinical forms begin to show compromise of the heart, the oesophagus and the colon, or two or more of these organs (Prata 2001). In fact, the most common clinical consequence of T. cruzi infection is chronic chagasic cardiomyopathy characterised by severe myocarditis, T cell-rich lymphomononuclear infiltrate, interstitial fibrosis and cardiomyocyte hypertrophy that can lead to dilated cardiomyopathy, end-stage heart failure and death (Chagas & Villela 1922; Laranja et al. 1956).

The reason why most individuals present no transition from the indeterminate form, with no signs and symptoms of disease, while a few develop severe illness with pathological cardiac involvement remains an intriguing question. Two alternative hypotheses have been proposed to try and explain the pathogenesis of Chagas heart disease: persistence of the parasite and autoimmunity theories. The parasite persistence theory affirms that tissue parasitism is an obligatory prerequisite for the development of chagasic cardiomyopathy, substantiated by studies showing a preferential tissue distribution of T. cruzi strains in humans and experimental animals since the beginning of the past century (Andrade et al. 1985; Vago et al. 1996). Further evidence to support this theory arises through methods such as immunohistochemistry (Higuchi et al. 1993) and polymerase chain reaction (Jones et al. 1993; Vago et al. 1996; Añez et al. 1999; Lages-Silva et al. 2001), which have shown a strict correlation between the presence of parasite and tissue lesions, highlighting the important role of the parasite in the pathogenesis of the disease. However, the relative lack of parasites in the myocardium during chronic Chagas disease led to the proposal of numerous theories suggesting autoimmune involvement, with the participation of autoantibodies or autoreactive T lymphocytes derived from molecular mimicry between parasite antigens and host, bystander activation induced by environmental antigens derived proinflammatory or cryptic epitopes with altered processing and the presentation of self-antigens (Cossio et al. 1974; Ribeiro-dos-Santos & Hudson 1981; Kierszenbaum 1986; Rizzo et al. 1989; Gironès et al. 2001; Leon et al. 2001; Abel et al. 2005; Iwai et al. 2005).

The presence of antibodies directed against human antigens appeared to be associated with dilated cardiomyopathy, including the production of the autoantibody anti-sarcolemmal proteins, myosin and troponin T (Rizzo et al. 1989; Tibbetts et al. 1994; Cunha-Neto et al. 1995; Leon et al. 2001; Basquiera et al. 2003; Saravia et al. 2011). These proteins are not present on the cell surface; they are predominantly expressed in the cytoplasm and are exposed to cardiac autoimmune reaction only under physiological conditions, releasing these proteins into the bloodstream after cardiac injury (Jahns et al. 2008). For this reason, the release of the cardiac-specific autoantibodies against both proteins in circulation has the potential to assess cardiomyocyte function and cardiac remodelling, and could be considered a tool that indicates autoimmune-mediated myocardial injury (Kaya et al. 2010). Thus, the present study aimed to evaluate the potential involvement of troponin T and myosin autoantibodies (IgG and isotypes) with the development of chronic clinical forms of Chagas disease.

Material and methods

Study population

A total of 141 participants from the rural zones of 11 municipalities of the State of Rio Grande do Norte, north-eastern Brazil, were initially selected for the study. Ten individuals dropped out and did not complete the study; therefore, they were not included in the clinical analysis. The inclusion criteria were positive epidemiology and serology for Tcruzi infection. The serological methods used were based on two reactions with different principles (Chagatest® recombinant ELISA and HAI, and indirect immunofluorescence assay), in accordance with recommendations of the World Health Organization and Brazilian Consensus of Chagas Disease (Brasil 2005). The exclusion criteria were as follows: diabetes, the presence of a cardiac pacemaker and non-chagasic cardiomyopathy (e.g. ischaemic, hypertensive or valvular).

Following enrolment, individuals with confirmatory positive serology results were submitted to a complete clinical evaluation, including electrocardiogram (ECG) mapping and chest X-ray, 2D-echocardiogram (ECHO) and in cases involving cardiac alterations, a 24-h Holter examination. All selected individuals were then classified according to clinical form as follows: those with ECG mapping and radiologic imaging presenting no sign of heart or gastrointestinal disease, as indeterminate (IND, n = 72); those exclusively presenting cardiac alterations, as cardiac (CARD, n = 47); and those suffering from megacolon or megacolon combined with cardiomyopathy, as digestive/cadiodigestive (DIG/CARD-DIG, n = 12). Samples from uninfected healthy individuals (CONT, n = 30) and ischaemic patients (ISCH, n = 15) were used as controls. Informed consent for this study was obtained from the participants and was approved by the Research Ethics Committee of the State University of Rio Grande do Norte (UERN) under protocol number No. 027.2011. All the experiments described here were performed according to human experimental guidelines of the Brazilian Ministry of Health and the Declaration of Helsinki.

Imaging techniques

As described previously, clinical screening was conducted in all the selected individuals. Patients were submitted to a comprehensive echocardiography with colour flow mapping performed in standard views (VIVID e, GE Healthcare, USA). The echocardiographic techniques and calculations of different cardiac dimension and volumes were evaluated in accordance with the recommendations of the American Society of Echocardiography (Lang et al. 2005). Left ventricular ejection fraction (LVEF) was calculated according to the modified Simpson's rule (biplane method). Twenty four-hour Holter monitoring was performed only on patients with chronic chagasic cardiomyopathy (CCC) using a three-channel portable recording system (Cardiolight®, Cardios, São Paulo, Brazil). The recording was analysed on a Holter DMI-Cardios 8300 system, using a semiautomatic technique. Patients were encouraged to continue their normal activities during the recording period. The examinations were conducted and analysed by one experienced observer who was blind to the serological profile, antibody measurements and clinical data of the participants.

Measurement of antibodies against human cardiac proteins and parasite antigens by ELISA

Peripheral blood samples were collected from participants by venipuncture, and the serum samples were stored at −20 °C prior to antibody quantification, which was performed by enzyme-linked immunosorbent assay (ELISA) for specific IgG antibodies and isotypes (IgG1, IgG2, IgG3 and IgG4), according to Voller et al. (1976). Trypomastigote (5.0 μg/ml) and epimastigote (7.5 μg/ml) forms of Tcruzi Y strain were used as antigens, prepared using alkaline extract of parasite growth in acellular medium Liver Tryptose infusion (LIT) and human cardiac myosin heavy chain proteins (0.062 μg/ml) (H00004621-Q01, Novus Biologicals, Littleton, CO, USA) and troponin T (0.062 μg/ml) (NBC1-28765, Novus Biologicals). All antigens were measured by the method proposed by Lowry et al. (1951), followed by serial titration to determine the minimum concentration of antigen for ELISA plate sensitisation. Microassay plates were coated overnight at 4 °C with all 100 μl of antigens preparation in coating buffer (15 mm Na2CO3 and 34 mm NaHCO3, pH 9.6). Blocking was performed with phosphate-buffered saline (PBS) with 1% foetal bovine serum (FBS) for 45 min at 37 °C, followed by two washes with PBS plus 0.05% Tween-20. One hundred microlitres of sera per well were diluted to 1:40 and 1:80 and incubated for 45 min at 37 °C. The wells were then washed four times with PBS plus 0.05% Tween-20 and incubated with 100 μl of horseradish peroxidase conjugate (Sigma Chemical Company, St. Louis, USA) anti-total IgG or IgG1, IgG2, IgG3 and IgG4 isotypes for 45 min at 37 °C and then rinsed three times with PBS plus 0.05% Tween-20. Following incubation for 20 min at room temperature with 100 μl of tetramethylbenzidine (TMB, Kirkegaard & Perry Laboratories, Maryland, USA), the reaction was stopped by the addition of 32 μl of 5N sulphuric acid. The plates were read in a spectrophotometer using a 450 ηm filter (Microplate Reader Mindray, model MR-96A, Shenzhen, China). The cut-off was determined using the mean absorbance of non-infected individuals plus two standard deviations. Each serum sample was assayed in duplicate and each plate had negative and positive controls.

Statistical analysis

Data are reported as mean ± standard deviation (SD). Comparisons between groups were performed using analysis of variance (anova) followed by the Tukey post test. To determine correlation between variables with non-normal distribution, the Spearman rank correlations test was used. Regression analysis was used to compare antibody levels and the parameters of the clinical forms. In all cases, differences were considered significant when P < 0.05. Our analyses were performed using the spss 20.0 (Chicago, IL, USA) and prism 5.0 (GraphPad, San Diego, CA, USA) statistical programs.


Initially, the distribution of several clinical forms of Chagas disease was investigated, together with the demographic characterisation of these distinct groups. We observed that of 131 participants who completed the study protocol, 72 (55.0%) were classified as presenting the IND clinical form, 47 (35.9%) as presenting the CARD form and 12 (9.1%) presenting the DIG/CARD-DIG forms. The mean age of the entire population was 49 years old [standard deviation (SD) = 13 years; range, 23–78 years old]. For patients with IND Chagas disease, the mean age was 48 years old (SD = 12 years; range, 23–75 years old). Among the patients with CARD, the mean age was 51 years old (SD = 13 years; range, 51–78 years old). The mean age for patients with DIG/CARD-DIG was 56 years old (SD = 15 years; range, 55–77 years old). Regarding sex, the female rate was 40/72 (55.6%), 16/47 (34.0%) and 7/12 (58.3%), respectively. No significant differences were determined in the patients regarding age or sex.

The next step was to compare electrocardiographic, radiographic and echocardiographic parameters in the patients with Chagas disease among the different clinical forms. Comparison of the radiographic parameters revealed that both the CARD (P = 0.0001) and DIG/CARD-DIG groups presented a higher cardiothoracic index (P = 0.005) than patients with IND. Echocardiographic parameter analyses showed significant differences between the IND and CARD groups. Patients with CARD presented the highest left atrium diameter (P = 0.040), left end-diastolic (P = 0.025) and systolic dimension (P = 0.005), and lowest left ventricular ejection fraction (P = 0.0001) (Table 1). Neither of these differences were observed between the CARD and DIG/CARD-DIG groups (data not shown).

Table 1. Electrocardiographic, radiographic and echocardiographic parameters of patients with indeterminate (n = 72), cardiac (n = 47) and digestive/cardiodigestive (n = 12) clinical forms of Chagas disease
VariablesClinical formsCorrelation between clinical forms
MeanSDMeanSDMeanSDMean differenceP-valueMean differenceP-value
  1. LV, left ventricular; IND, indeterminate; CARD, cardiac; DIG/CARD-DIG, digestive/cardiodigestive; ND, no difference; SD, standard deviation.

  2. There was no significant mean difference between the cardiac and digestive/cardiodigestive groups. Continuous variables are expressed as mean ± standard deviation.

  3. a

    The mean difference is significant at a level of 0.05.

Cardiothoracic index (CTi)0.4390.0370.4760.0490.4830.064−0.037a0.0001−0.044a0.005
Left atrium diameter (mm)32.8064.99835.3626.36733.8335.167−2.556a0.040−1.028ND
LV end-diastolic dimension (mm)47.5694.47547.5694.47548.3334.539−2.729a0.025−0.764ND
LV end-systolic dimension (mm)29.3064.11133.0648.83030.2505.101−3.758a0.005−0.944ND
LV ejection fraction (%)67.2226.26957.68114.80662.08312.7179.541a0.00015.139ND

Next, we investigated total IgG production in patients with Chagas disease using parasite antigens to verify whether these findings were similar using human cardiac antigens. The results showed higher production of total IgG observed in sera from patients with Chagas disease with the IND (P = 0.0001) and CARD clinical forms (P = 0.0001) for all antigens and with the DIG/CARD-DIG form [P = 0.0001 (trypomastigote, epimastigote and troponin T antigens), P = 0.002 (myosin antigen)] compared with the CONT group. Additionally, the ISCH group showed production of anti-trypomastigote, epimastigote, myosin and troponin T antibodies below the cut-off level (Figure 1a–d). Moreover, patients with CARD showed high levels of specific anti-Tcruzi total IgG using epimastigote (P = 0.0027) and trypomastigote (P = 0.0018) antigens compared with patients with IND (Figure 1a and b). Surprisingly, the CARD and IND groups showed similarly high levels of autoantibodies against myosin (P = 0.0001) and troponin T (P = 0.0001) proteins (Figure 1c and d, respectively).

Figure 1.

Concentration of total IgG antibodies anti-trypomastigote (a), epimastigote (b), myosin (c) and troponin T (d) in the sera of patients with IND (n = 72), CARD (n = 47) and DIG/CARD-DIG (n = 12) Chagas disease, ISCH (n = 15) and the CONT group (non-infected individuals). Individual results for each participant are presented as mean ± SEM. Each data point represents the mean absorbance of duplicate wells. Statistical differences for comparisons between the control group and patients with Chagas disease are shown for each graph, P ≤ 0.05 (Spearman). CONT, control group; ISCH, ischaemic cardiomyopathy; IND, indeterminate; CARD, cardiac; DIG/CARD-DIG, digestive and cardiodigestive.

Isotypes analyses of specific-Tcruzi antibodies and autoantibodies against troponin T and myosin production demonstrated different production profiles. Trypomastigote preparations showed that sera of patients with Chagas disease presented higher levels of T. cruzi-specific IgG1, IgG2 and IgG3 antibodies than CONT. Interestingly, patients with CARD presented higher levels of IgG1 (P = 0.0005) and IgG2 (P = 0.0094) antibodies than patients with IND (Figure 2a). Epimastigote antigen analyses detected enhanced levels of IgG1 and IgG3 isotypes in patients with Chagas disease compared with the CONT group. Moreover, higher levels of IgG3 were observed in the CARD group compared with the IND group (P = 0.0021) (Figure 2b). When analysing the anti-myosin and troponin T autoantibodies, we observed production of IgG2, IgG3 and IgG4 among the chagasic groups. Moreover, higher levels of IgG2, IgG3 and IgG4 autoantibodies against myosin and troponin T were observed in patients with CARD compared with the IND group (Figure 2c and d).

Figure 2.

Concentration of specific IgG isotypes (IgG1, 2, 3 and 4) antibodies anti-trypomastigote (a), epimastigote (b), myosin (c) and troponin T (d) in the sera of IND (n = 72), CARD (n = 47) and DIG/CARD-DIG (n = 12) individuals with chronic Chagas disease and the control group (non-infected individuals). Individual results for each participant are presented as mean ± SEM. Each data point represents the mean absorbance of duplicate wells. Statistical differences for comparisons between the control group and patients with Chagas disease are shown for each graph, P ≤ 0.05 (Spearman). CONT, control group; IND, indeterminate; CARD, cardiac; DIG/CARD-DIG, digestive and cardiodigestive.

As no initial difference in total IgG production was verified between patients with IND and CARD using myosin and troponin T antigens, we investigated whether radiographic and echocardiographic parameters correlated with the levels of antibodies in the sera of patients with Chagas disease. Analysis of our results revealed that during the chronic phase of the disease, total IgG anti-myosin (P = 0.0403, r = 0.1742) and total IgG anti-troponin T (P = 0.0421, r = 0.1727) and IgG1 (P = 0.0435, r = 0.1715) levels were directly associated with left ventricular end-systolic dimension (Figure 3 a–c). Likewise, a negative correlation between total IgG anti-trypomastigote (P = 0.0210, r = 0.1956), myosin (P = 0.0097, r = 0.2193) and troponin T (P = 0.0276, r = 0.1855), and IgG1 anti-troponin T (P = 0.048, r = 0.262) with left ventricular ejection fraction was determined for patients with Chagas disease (Figure 3 d–g).

Figure 3.

Correlations between the total immunoglobulin G (IgGt and isotypes) levels anti-trypomastigote, myosin (a, e) and troponin T (b–c, e–f) with clinical parameters of cardiac function in patients with Chagas disease. Left ventricular systolic diameter (a–c) and left ventricular ejection fraction (d–g). Individual results are shown for each patient, and the line represents the linear regression for each comparison. Spearman's rank correlation test considered significant when P-value was P ≤ 0.05.

In an attempt to clarify whether these autoreactive antibodies were involved in the pathogenesis of Chagas disease, we divided the patients with IND, CARD and DIG/CARD-DIG into two groups, low (below cut-off) and high (above cut-off) autoantibody producers, and compared the electrocardiographic, radiographic and echocardiographic parameters of these groups. Greater left ventricular end-systolic diameter (P = 0.0240), left atrial diameter (P = 0.0183) and cardiothoracic index (P = 0.0250) were verified in patients with CARD with high production of anti-epimastigote antibodies than in those with low antibody production (Figure 4 a–c). Likewise, high anti-epimastigote (P = 0.0419), myosin (P = 0.013) and troponin T (P = 0.042) producers presented lower LVEF compared with low antibody producers (Figure 4 d–f).

Figure 4.

Comparison of clinical parameters of cardiac function in high (+) and low antibody producing patients with Chagas disease for (−) anti-epimastigote (a–d), myosin (e) and troponin T (f) antigens with indeterminate-IND; cardiac-CARD and digestive/cardiodigestive-DIG/CARD-DIG clinical forms. Mann–Whitney non-parametric t-test was used and considered significant when P-value was P ≤ 0.05.

To identify potential clinical markers of the risk of manifesting the CARD, DIG, CARD-DIG forms during the chronic phase of Chagas disease, we determined the percentage of patients who presented high and low antibody production for each antigen. Eighty-two (62.6%) patients with Chagas disease showed high levels (above cut-off) of total IgG anti-trypomastigotes, while antibody production was reduced in 49 (37.4%) patients (below cut-off). We verified that among patients showing high anti-trypomastigote production, 41.5% (34/82) presented the IND clinical form, 45.1% (37/82) presented CARD and 13.4% (11/82) presented DIG/CARD-DIG (Figure 5a). Among patients showing high autoantibody production against troponin T, 47.5% (37/78) presented IND, 39.7% (31/78) presented CARD and 12.8% (10/78) presented DIG/CARD-DIG (Figure 5b). Similarly, among patients showing high anti-myosin autoantibody production, 50.0% (37/74) presented IND, 37.8% (28/74) presented CARD and 12.2% (9/74) presented DIG/CARD-DIG (Figure 5c). The majority of patients with the IND clinical form showed low production of anti-troponin T and myosin autoantibodies (Figure 5b and 5c).

Figure 5.

Distribution of anti-trypomastigote (a), troponin T (b) and myosin (c) antibodies in high (+) and low (−) producing patients with Chagas disease, according to clinical form of Chagas disease. IND, indeterminate; CARD, cardiac; DIG/CARD-DIG, digestive/cardiodigestive.

Subsequently, we determined whether high anti-trypomastigote producing patients were those that showed high production of anti-epimastigote, troponin T and myosin antibodies. A positive correlation was observed between anti-trypomastigote and epimastigote (r = 0.6988; P = 0.0001), anti-myosin (r = 0.6233; P = 0.0001) and anti-troponin T antibody production (r = 0.6179; P = 0.0001). Surprisingly, a positive correlation of 95% (r = 0.9508; P = 0.0001) was observed between the production of anti-troponin T and myosin (Figure 6), indicating that almost all patients generate autoantibodies against both antigens.

Figure 6.

Correlation of total immunoglobulin G (IgGt) levels anti-trypomastigote with anti-epimastigote (a), myosin (b), troponin T (c) and correlation of anti-troponin T with anti-myosin autoantibodies in patients with Chagas disease with different clinical forms. Mann–Whitney non-parametric t-test was used and considered significant when P-value was P ≤ 0.05.


In this study, our efforts were directed to evaluating possible associations among the autoantibodies levels in the sera of patients with different clinical forms of chronic Chagas disease. The results verified that specific anti-Tcruzi antibodies and autoantibodies against myosin and troponin T are frequently detected in patients with chronic Chagas disease. In addition, there is a correlation between patients with the cardiac clinical form, the production of anti-troponin T/myosin autoantibodies and diminished left ventricular ejection fraction (LVEF), which is an important indicative of systolic dysfunction. These findings suggest a key role for autoantibody production in unleashing the pathological process observed in Chagas disease.

Initially, total IgG production was measured and a highly consistent immunoglobulin response was verified in patients with Chagas disease compared with the CONT and ISCH non-infected groups. Furthermore, we observed a marked increase in antibodies production in patients with CARD compared with those presenting the IND clinical form. Supporting these findings, data in the literature indicate a different serological profile in ELISA using trypomastigote and epimastigote antigens in both the acute and chronic phases of Chagas disease (Umezawa & Silveira 1999). Low titres of IgG anti-T. cruzi were reported in indeterminate cases, while patients with chronic Chagas disease presented a high humoral immune response (Monteón-Padilla et al. 1999, 2001). Apparently, specific anti-T. cruzi antibodies could contribute to the pathophysiology of this disease.

Our results concerning IgG isotypes production using trypomastigote and epimastigote forms are in agreement with data that demonstrated high levels of IgG1 and IgG3, followed by lower levels of IgG2 and IgG4, in patients with chronic Chagas disease using epimastigote preparations, with no significant differences between the clinical forms (Cerban et al. 1993; D’Ávila et al. 2009). Other researchers using the same antigenic preparation have demonstrated high levels of IgG1 and IgG2, followed by IgG3 and IgG4, and a tendency for an association between elevated levels of IgG2 and the presence of cardiomegaly (Morgan et al. 1996; Cordeiro et al. 2001; Hernández-Becerril et al. 2001). These contradictory results taken together suggest a mixed Th1/Th2 immune response with cardiac alterations associated with high Th1 immune response, while patients with the indeterminate clinical form are divided evenly between Th1 and Th2 (Reis et al. 1997; Bahia-Oliveira et al. 1998; Abel et al. 2001; Gomes et al. 2003). Of note, IgG isotype production is controlled by different profiles of cytokines: pro-inflammatory cytokines, such as IL-12, IFN-γ and TNF-α (Th1 profile), will induce IgG1 and IgG3 production, while anti-inflammatory cytokines, IL-4 and IL-10 (Th2 profile), will promote IgG2 production (Briere et al. 1994; Kawano et al. 1994). Based on this, it is possible that preferential induction of different subsets of T helper cells by a specific antigen or host genetic background may be crucial to the pattern of antibody response and, subsequently, to the immunopathology of Chagas disease.

High levels of total IgG anti-myosin and troponin were observed in both patients with IND and CARD, predominantly the IgG2 and IgG3 isotypes. In fact, myosin represents the most important cardiac autoantigen in Chagas disease and many researchers have demonstrated autoimmunity specific against myosin induced in humans and T. cruzi infected experimental models (Rizzo et al. 1989; Tibbetts et al. 1994; Cunha-Neto et al. 1996; Leon et al. 2001). However, studies using the same approach with troponin T as antigens in Chagas disease remain scarce. Previous data evaluating serum troponin T concentrations appear to have little value as an early marker of myocardial lesions in patients with positive serology for Chagas disease, as only one patient with advanced chagasic cardiomyopathy presented elevated troponin T concentration using this highly sensitive test (Basquiera et al. 2003). Another study demonstrated autoantibodies against troponin T in patients with Chagas disease that presented the cardiac and indeterminate clinical forms of the disease; however, no difference was observed between the groups, and the study did not investigate anti-troponin IgG isotype production (Saravia et al. 2011). In contrast, our data showed high levels of anti-troponin T autoantibodies in patients with the cardiac clinical form compared with indeterminate patients. Several studies have shown that a global systolic left ventricular dysfunction is the strongest predictor of morbidity and mortality in Chagas disease (Bestetti et al. 1994; Carrasco et al. 1994; Mady et al. 1994). Despite similar levels of autoantibody total IgG anti-myosin and troponin T production, there is an association between these autoantibodies and the echocardiographic parameters obtained from patients with Chagas disease. Given that LVEF is an important parameter of left ventricular dysfunction, our study showed that cardiac patient producers of anti-troponin T and myosin autoantibodies showed lower LVEF than non-producers. An increase in the production of these autoantibodies could be related to major heart damage and unfavourable left ventricular systolic function in chagasic cardiomyopathy. In cardiomyopathies of unknown origin (idiopathic), but not ischaemic, the presence of autoantibodies (troponin I, β-adrenoceptors) and its correlation with heart failure have been described; about 70% of patients with idiopathic cardiomyopathies may present autoantibodies against cardiac antigens (Lappé et al. 2011; Brisinda et al. 2012). This phenomenon could contribute to autoantibodies generation in patients with Chagas disease, principally in low anti-trypomastigotes antibody producers.

Most of the patients who showed high production of anti-trypomastigote antibodies also produced the autoantibodies anti-troponin T and myosin. These findings allow us to infer that the autoimmune mechanism involved in chagasic pathogenesis could be molecular mimicry, in which the immune response directed to Tcruzi ‘cross-reacts’ with a self-protein sharing the target epitope in the human host. Although autoantibodies against myosin and troponin T were detected in approximately half of the indeterminate group of patients, it is probable that other factors are involved in the pathogenesis of chronic Chagas disease. However, long-term follow up of indeterminate patients producing autoantibodies patients against cardiac proteins could determine these autoantibodies as clinical markers if patients begin to present the cardiac or digestive clinical forms.

In conclusion, our findings suggest that the production of autoantibodies against troponin T and myosin could be associated with the development of distinct clinical forms of Chagas disease. These data could provide valuable support for future research and guide alternative therapeutic procedures. In addition, they could aid in establishing markers for the clinical progression of this infection/disease of long-term evolution. Early diagnosis and specific treatment can improvement the life quality of these patients, who are also exposed to other parasitic diseases in developing countries.


This work was supported by research grants from the Conselho Nacional de Desenvolvimento Científico e Tecnológico, Fundação de Amparo à Pesquisa do Estado de Rio Grande do Norte and research fellowships from the CNPq. The authors are grateful to the participants involved in this study, to the field staff for their hard work and to the Secretariat of State for Public Health of Rio Grande do Norte, represented by the health authorities and agents of the Municipal Secretaries of the west mesoregion for their indispensable support for the field activities during the development of this survey. Finally, the authors would also like to express their thanks to Philip S.P. Badiz for his critical reading and revision of the manuscript.