Cardiovascular risk factors in chronic Chagas' disease are associated

Cardiovascular risk factors in chronic Chagas’disease are associated with a di¡erent pro¢le of putative heart-pathogenic antibodies ´ Marcipar1, Stella Maris Pezzotto3, Juan Beloscar4, Oscar Pellizzon4, Cristina Diez1, Susana Gea2, Ivan 1 Alberto Marcipar & Oscar Bottasso3 1

INTEBIO, Facultad de Ciencias Bioqu´ımicas, Universidad Nacional del Litoral, Santa Fe, Argentina; 2CIBICI-CONICET, Facultad de Ciencias Qu´ımicas, ´ ´ ´ Universidad Nacional de Cordoba, Ciudad Universitaria, Cordoba, Argentina; 3Instituto de Inmunolog´ıa, Facultad de Ciencias Medicas, Universidad ´ en Cardiolog´ıa, Facultad de Ciencias Medicas, ´ Nacional de Rosario, Santa Fe, Rosario, Argentina; and 4Carrera de Especializacion Universidad Nacional de Rosario, Santa Fe, Rosario, Argentina

Correspondence: Oscar Bottasso, Instituto de Inmunolog´ıa, Facultad de Ciencias ´ Medicas, Santa Fe 3100, Rosario 2000, Argentina. Tel.: 154 341 4 804559; fax: 154 341 4 804569; e-mail: [email protected] Received 2 December 2005; revised 22 March 2006; accepted 20 April 2006. First published online 9 June 2006. DOI:10.1111/j.1574-695X.2006.00115.x Editor: Willem van Eden Keywords cardiovascular risk factors; Chagas’ disease; heart damage; crossreactive antibodies.

Abstract Given that cardiovascular risk factors (CRF), such as smoking, alcoholism and hypertension, may contribute to the development of heart lesions, chronically Trypanosoma cruzi-infected individuals were studied to explore the relationship between the presence of such CRF, cardiomyopathy and antibodies that have been proposed to play a pathogenetic role in Chagas’ disease. The targets of these antibodies were T. cruzi antigens such as cruzipain (Cz), a P ribosomal antigen (P2), and a component of myelin sheaths also present in T. cruzi (sulphatide). Individuals were classified into four groups on the basis of specific serology and presence of CRF: subjects with T. cruzi infection and CRF; those with positive serology and no CRF; seronegatives with CRF; and seronegatives without CRF, were analysed. Seronegatives or seropositives with CRF showed a greater occurrence of heart involvement (chest X-ray and/or electrocardiogram abnormalities). Seropositives with CRF displayed significantly higher levels of antisulphatide antibodies than the three remaining groups and higher levels of antibodies against Cz and P2 compared to the seropositives without CRF. Increased amounts of antiP2 and antisulphatide antibodies were also found in seropositives with marked heart involvement. The presence of CRF is associated with a different profile of antibody responses and degree of cardiac effects.

Introduction Chagas’ disease, or American trypanosomiasis, is a parasitic disease caused by the protozoan Trypanosoma cruzi. The disease is endemic from northern Mexico to Argentina, with 16–18 million infected humans, and nearly 90 million people at risk of acquiring the infection, being one the largest parasitic disease burdens in the continent (Kirchhoff, 1993). After a mostly asymptomatic initial episode, the majority of infected individuals remain free from heart disease, in the so-called ‘indeterminate’ form, whereas 25–30% eventually develop irreversible heart lesions (chronic chagasic cardiomyopathy, CCC), 15–30 years following acute infection (Andrade, 1999). The outcome of the disease and progression from indeterminate form to CCC in infected individuals is not completely understood. As the development of CCC is distant from the acquisition of infection, such an event may depend not only on the particular features of the infectious 2006 Federation of European Microbiological Societies Published by Blackwell Publishing Ltd. All rights reserved

c

process, infecting parasite strain, re-infections and host genetic background (Da´ vila et al., 1987; Montamat et al., 1996; Cruz-Robles et al., 2004; Risso et al., 2004), but also on the presence of additional factors able to promote heart damage. In this sense, it is rational to analyse whether cardiovascular risk factors (CRF) present in the general population, such as smoking, alcoholism and hypertension, may influence the degree of heart disease in CCC. An initial step to establish the pathological contribution of such factors may be to analyse whether their presence in chronically infected individuals is associated with a different pattern of cardiac involvement and potentially tissuedamage-associated markers. Despite some controversy, CCC may be the result of several pathogenic mechanisms acting concomitantly, such as immune responses to parasite antigens (Reis et al., 1997), autoimmune processes (Kalil & Cunha-Neto, 1996; Leon & Engman, 2001; Girones & Fresno, 2003), and the inflammation accompanying these FEMS Immunol Med Microbiol 48 (2006) 26–33

27

Cardiovascular risk factors in chronic Chagas’ disease

responses (Higuchi et al., 1987; Brener & Gazzinelli, 1997; Pontes-De-Carvalho et al., 2002). Several parasite structures and autoantigens seem to be involved in the pathology of Chagas’ disease. Cruzipain (Cz), the major T. cruzi cysteine proteinase, and T. cruzi ribosomal proteins (P2) are recognized in most sera from chronic patients (Levitus et al., 1991; Martinez et al., 1991; Aznar et al., 1995). In particular, antibody levels against different Cz epitopes (Duschak et al., 2001) and the P3 C-terminal region appeared to be related to the clinical status of chronically T. cruzi-infected patients (Levitus et al., 1991; Aznar et al., 1995). Furthermore, antibodies against P2 and Cz are known to cause electrocardiographic alterations in noninfected hearts of immunized mice, suggesting a possible involvement of these antibodies in experimental Chagas cardiopathy (Masuda et al., 1998; Giordanengo et al., 2000b; Lopez Bergami et al., 2001). Targets of autoimmune reactions elicited either by cross-reactive T. cruzi antigens or antigens released by host cell lysis may include self-antigens such as laminin (Szarfman et al., 1982), and/or some neural antigens such as sulphatides. These lipids are specific components of myelin sheaths of peripheral nerves (Roberts, 1991) and their occurrence has also been documented in T. cruzi epimastigotes (De Lederkremer et al., 1985; Petry et al., 1988). Antisulphatide antibodies have been demonstrated in T. cruzi-infected or uninfected individuals with cardiac disease (Avila et al., 1993) and experimentally infected rats, in which these antibodies were found to bind homologous neural structures (Feldman et al., 1999). We hypothesized that the presence of CRF during T. cruzi infection may contribute to the mechanisms underlying cardiac damage in Chagas’ disease. As most of these processes present an immunological component, exposed individuals may have an increased occurrence of cardiac lesions and presence of some potentially pathogenic antibodies, some of them against parasite antigens exhibiting cross-reactivity with self-structures. To explore this question, chronically T. cruzi-infected individuals presenting CRF were studied to analyse their degree of cardiac involvement and antibodies against antigens such as Cz, P2 (TcP2b protein), and sulphatides.

Materials and methods Subject population Participants recruited in the study were those attending at the Chagas’ Disease Service from the Hospital Provincial del Centenario. It is a reference centre located within the Department of Cardiology where chagasic patients are usually assisted. Attendees were blood donors with a possible serological diagnosis of T. cruzi infection, people with a FEMS Immunol Med Microbiol 48 (2006) 26–33

personal interest in knowing whether they have Chagas’ disease or chagasic persons derived from other care health services. Recruitment was made by consecutive sampling between June 2001 and August 2002, and a high response rate was obtained among those asked to participate in the study. All cases were born in rural areas of northern Argentina where Chagas’ disease is endemic, and had migrated during their young adulthood to Rosario, a city where autochthonous vectorial cases have never been reported. None of them were under specific treatment or had concomitant pathological disorders able to affect the end point under analysis (i.e. congenital or rheumatic cardiopathies and immunological diseases). Participants gave their informed consent and the protocol was approved by the Ethic Committee of the Facultad de Ciencias Medicas de Rosario. At the time of inclusion, potential candidates were asked to be subjected to the following protocol: (a) careful clinical examination; (b) a two-arm stratification based on the presence of CRF or not, that is smoking, alcoholism, and hypertension (T. cruzi infection constitutes no risk for this condition). These risk factors were defined according to the following criteria: (i) smoking, a 10-year history of at least 20 cigarettes per day; (ii) alcoholism, as defined by the World Health Organization (1984). Such a condition was reported to be present when the daily intake of alcohol was above 100 mg day1 for a minimum period of 10 years. (iii) Hypertension; to this end blood pressure was measured in the recumbent position on the right arm, c. 30–45 min after arrival at the centre. A mercury sphygmomanometer standing on a bedside table at the same level as the subject’s heart was used. Blood pressure was then recorded on the examination sheet and the physical examination was performed. After this, blood pressure was measured again (30 min after the first reading), using the same method. Based on World Health Organization specifications (Guidelines Subcommittee 1999), individuals whose diastolic blood pressure was 90 mmHg or higher were considered to be hypertensive. Persons carrying no CRF were nonsmokers, nonalcohol drinkers, with normal blood pressure. (c) Frontal chest Xray, and 12-lead resting electrocardiogram (ECG), were performed by the same personnel and in similar conditions and were interpreted independently by two experienced cardiologists, who were unaware of the serologic results. Routine laboratory studies and an estimation of the body mass index (weight in kilograms divided by the square of height in meters) were also included.

Serology Specific serology was performed even in subjects with previous diagnosis, by enzyme-linked immunosorbent assay (ELISA) and indirect haemagglutination (IHA). Sera yielding positive reactions to both tests were considered positive. 2006 Federation of European Microbiological Societies Published by Blackwell Publishing Ltd. All rights reserved

c

28

Seronegative subjects were those showing negative reactions to ELISA and IHA tests. Doubtful reactions were regarded as an exclusion criterion.

Classification of cardiac involvement According to an established consensus (Storino, 1994), seropositives were grouped into the following three categories: Group 1 (G1), symptomless, normal ECG and chest X-ray; and G2, no congestive heart failure, ECG showing any of the following alterations: (a) incomplete right bundle branch block or complete right bundle branch block; (b) ventricular arrhythmia or (c) a1b, chest X-ray cardiothoracic ratio o 0.55; or G3, with congestive heart failure, pathological ECG tracings and chest X-ray cardiothoracic ratio 4 0.55. Seronegative individuals were classified into three categories: G1, no cardiac affectation; G2, only ECG alterations; and G3, congestive heart failure, pathological ECG tracings and chest X-ray cardiothoracic ratio 4 0.55.

Cruzipain purification Epimastigote forms of T. cruzi Tulahuen strain were grown at 28 1C in brain heart infusion (Becton Dickinson, France) supplemented with 0.5% tryptose, 10% FCS, 200 mg mL1 hemin, 100 U mL1 penicillin, and 100 mg mL1 streptomycin. Parasites were harvested at the exponential growth phase, centrifuged at 5000 g for 10 min at 4 1C, and washed with PBS. Parasites were resuspended in 3 vol of 0.25 M sucrose, 5 mM KCl and the next irreversible protease inhibitors: 1 mM N-alpha-p-tosyl-lysine chloro methyl ketone (TLCK) and 1 mM phenylmethylsulfonylfluoride (PMSF) (Sigma) were added. The epimastigotes were disrupted by three cycles of freezing ( 20 1C) and thawing (4 1C) and the homogenates were centrifuged at 7000 g for 15 min at 4 1C. Cruzipain was purified to homogeneity according to Giordanengo et al. (2000b).

Expression and purification of recombinant TcP2b protein A T. cruzi trypomastigote cDNA clone encoding TcP2b protein was subcloned in pET 32a (Novagene) and expressed in the BL21 (DE3) Escherichia coli as described previously (Marcipar et al., 2004). The pET32a vector directs the synthesis of the foreign polypeptides in E. coli as fusion polypeptide with the 20.5 kDa tiorredoxin protein and a histidine tag to facilitate the purifications of the protein. The 33 kDa expressed fusion protein was purified using the NiNTA resin (Quiagen) according to the manufacturer’s specifications. Briefly, cultures were induced to protein expression for 3 h with 1 mM isopropyl-b-D-thiogalactopyranoside and then sonicated and centrifuged for 30 min at 2006 Federation of European Microbiological Societies Published by Blackwell Publishing Ltd. All rights reserved

c

C. Diez et al.

4500 g and 4 1C. The supernatants were passed through the NiNTA column. Then the column was washed with 50 mM NaH2PO4 (pH 8), 300 mM NaCl, 50 and 100 mM imidazol buffer and then eluted with the same buffer plus 250 mM imidazol. The purity of the recombinant protein was analysed by 12% sodium dodecyl sulphate (SDS)-polyacrylamide gel electrophoresis (Laemmli, 1970), followed by Coomasie blue staining. Protein quantification was performed with the Bradford assay (Bradford, 1976).

Antibody assays Serum levels of anti-Cz, anti-P2 or antisulphatide IgG antibodies were measured by ELISA, as previously described (Avila et al., 1993; Giordanengo et al., 2000b; Marcipar et al., 2004). In brief, microtiter plates (Costar, Cambridge, MA) were coated with 10 mg mL1 of Cz or 0.5 mg TcP2b in 0.05 M carbonate-bicarbonate pH 9.6 and incubated overnight at 4 1C. In the case of antisulphatide antibodies, microtiter plates were coated with 1.14 mg of sulphatide from bovine brain (Sigma) dissolved in absolute ethanol until complete ethanol evaporation. Plates were blocked with 5% of bovine serum albumin, and incubated with a 1 : 100 dilution of human sera. After washing, a peroxidaseconjugated goat antihuman IgG (Sigma) was added. All incubations were performed at 37 1C for 60 min. Plates were read at 490 nm in an ELISA reader (BioRad) after incubation with H2O2 and O-phenylendiamine for determining anti-Cz and at 450 nm after incubation with trimetilbenzidine in H2O2 for anti-P2 and antisulphatide antibodies. For each plate, reactive and nonreactive sera for each antigen (positive and negative controls, respectively) were assayed simultaneously. Such fractionated sera were kept at 20 1C until use. The between-plate variation coefficient was o 10%.

Statistical analysis Comparisons among groups were made in relation to age, sex, length of residence, CRF distribution, antibody levels and presence of heart involvement. Categorical variables were analysed by the w2 or Fisher’s exact test when applicable, whereas the analysis of variance and Student’s t-test were used to evaluate differences in mean values. The odds ratios (ORs) were the measures employed for comparisons of cardiac abnormalities according to serologic status and presence of CRF. Background factors were controlled by the unconditional logistic regression method. To assess the eventual effects of other variables on the means of antibody levels to each one of the three antigens, the general linear model was applied. This procedure provides a regression analysis and analysis of variance for one dependent variable adjusted by one or more factors. The level of significance was set at P o 0.05. FEMS Immunol Med Microbiol 48 (2006) 26–33

29


Results Subject characteristics and cardiac involvement After the study protocol was completed, individuals were distributed into four groups. The group constituted by persons with T. cruzi infection and one or more CRF was named Tc1CRF1. The second group had no serologic evidence of T. cruzi infection but presented CRF (Tc CRF1). Individuals showing positive serology and lacking CRF constituted the third group (Tc1CRF ), and the fourth group consisted of those with neither serological evidence of T. cruzi infection nor CRF (Tc CRF ). Place of birth and living conditions before migration revealed no differences among groups. The subject profile is depicted in Table 1. There was a higher female/male ratio in the Tc1 group with no CRF (P o 0.02), with mean ages being close to the level of statistical significance (P o 0.06). No statistical differences in length of residence in high endemicity areas were recorded among groups. Overall numbers on the occurrence of each CRF in seropositives and seronegatives were as follows: Tc CRF1 Tobacco (To) 20, Alcoholism (Al) 7, Hypertension (Hy) 13; Tc1CRF1, To 31, Al 14, Hy 13 (ns). When the distribution of two or three CRF was compared between Tc1CRF1 and Tc CRF1 groups, there was a higher occurrence of two CRF in Tc1CRF1 group (P = 0.03, Fisher’s exact test, Table 1). Analysis of body mass index revealed no differences Table 1. Subject profile according to serology and presence of cardiovascular risk factors (CRF) Tc CRF Sex (female/male) 11/8 Agew 47 2.3 Length of residencew 25.2 4.4 CRF distribution 1 factor Alcohol (Al) Hypertension (Hy) Tobacco (To) 2 factors Al1To Al1Hy To1Hy 3 factors

Tc1CRF

Tc CRF1 Tc1CRF1

23/6 14/14 18/24 46.5 2.4 52 2.4 54.3 2.3 21 2.6 24.7 4.2 26.5 3.2 20 1 7 12 4 2 – 2 4

26 3 5 18 15z 7 3 5 1

Statistically different from Tc1CRF1 and Tc CRF1: P o 0.02. w

Mean SEM in years. Distribution of two CRF significantly different from the Tc-CRF1group: P = 0.3 (Fisher’s exact test). Tc1CRF1, persons with Trypanosoma cruzi infection and one or more CRF; Tc CRF1, persons with no serologic evidence of T. cruzi infection presenting CRF; Tc1CRF , subjects showing positive serology and lacking CRF; Tc CRF , persons with neither serological evidence of T. cruzi infection nor CRF. z

FEMS Immunol Med Microbiol 48 (2006) 26–33

among groups (mean SEM), Tc CRF 26.3 0.94, Tc CRF126.9 0.83, Tc1CRF 26.1 0.60, Tc1CRF1 27.2 0.85. Clinical biochemistry studies identified seven type-2 diabetic subjects, three in the Tc CRF group and four among Tc1CRF1 individuals. Four subjects presented lipid disorders, three among Tc CRF1 persons and one in the Tc1CRF1 group. The group of seronegative persons with cardiac affectation was composed of 14 individuals with pathological ECG tracings, one of them also showing a chest X-ray cardiothoracic ratio 4 0.55. Twenty persons had one CRF (mostly To n = 12, or Hy, n = 7), four patients presented two CRF (To and Hy or To and Al) and another four cases had the three CRF. Ten cases had a single ECG pathological tracing and four patients presented two ECG abnormalities. Types of ECG alterations were as follows, left anterior hemiblock (n = 6), right bundle branch block (n = 5), left ventricular hypertrophy (n = 4) and left bundle branch block (n = 3). We next analysed the distribution of cases with cardiac effects according to the above-described criteria. As shown in Table 2, it was clear that Tc CRF1 and Tc1CRF1 subjects had an increased occurrence of heart involvement (G2 and G3) compared with those subjects presenting no CFR (P = 0.01 and P o 0.015, respectively). In a further step, ORs were calculated. Analysis among seropositives was initially carried out by considering those with cardiopathy (G2 and G3) or without cardiopathy (G1). Compared with the Tc1CRF group, the crude OR for Tc1CRF1 was 2.63 (95% CI 0.97–7.16, P = 0.05), with the trend becoming insignificant when controlling for sex and age. Considering the unequal distribution of G2 and G3 categories among

Table 2. Distribution of subjects according to the classification of cardiac involvement Cardiopathy groups Subjects

1

2

3

Total

Tc CRF Tc CRF1 Tc1CRF Tc1CRF1w Total

18 15 14 11

1 12 14 18

0 1 1 13

19 28 29 42 118

Distribution of cardiopathy groups 1 and 2 significantly different from

Tc CRF : P = 0.01 (Fisher’s exact test). Distribution of cardiopathy groups 2 and 3 significantly different from Tc1CRF : P o 0.015 (Fisher’s exact test). For comparison purposes, the presence of cardiopathy among seronegatives was classified according to the criteria applied for the Trypanosoma cruzi-positive group. Tc1CRF1, persons with T. cruzi infection and one or more CRF; Tc CRF1, persons with no serologic evidence of T. cruzi infection presenting CRF; Tc1CRF , subjects showing positive serology and lacking CRF; Tc CRF , persons with neither serological evidence of T. cruzi infection nor CRF. w

2006 Federation of European Microbiological Societies Published by Blackwell Publishing Ltd. All rights reserved

c

30

C. Diez et al.

Table 3. Circulating levels of antibodies to sulphatide, Cz and P2 in the subject groups Subjects

N

Antisulphatide

Anti-Cz

Anti-P2

Tc CRF1 Tc CRF Tc1CRF1 Tc1CRF Overall P value

28 19 42 29

0.43 0.03 0.39 0.03 0.57 0.03 0.50 0.04 o 0.01

0.18 0.05 0.14 0.01 1.16 0.05 0.90 0.08 o 0.0001

0.20 0.04 0.15 0.01 0.81 0.04 0.64 0.06 o 0.0001

Values are expressed as mean SEM of OD units. Comparison between Tc1CRF1 and Tc1CRF , anti-Cz: P o 0.005; anti-P2: P o 0.015. Tc1CRF1, persons with Trypanosoma cruzi infection and one or more CRF; Tc CRF1, persons with no serologic evidence of T. cruzi infection presenting CRF; Tc1CRF , subjects showing positive serology and lacking CRF; Tc CRF , persons with neither serological evidence of T. cruzi infection nor CRF; Cz, cruzipain.

T. cruzi-infected persons, Tc1CRF and Tc1CRF1 were also compared according to these categories. As stated above, the G3 category was more likely to be present in those with CRF [crude OR 10.11 (95% CI 1.11–232.2), P o 0.015], as was when controlling for age and sex (P o 0.05). For comparison purposes, seronegatives were analysed in the same way. Individuals from the Tc CRF1 group had an increased chance of showing cardiac involvement (G2 and G3 categories) with respect to the Tc CRF group [crude OR 15.6 (95% CI 1.82–133.40), P o 0.012]. The trend remained significant after age- and sex-adjustment (P o 0.045).

Antibody results We further compared antibody levels in the four subject groups. Data from Table 3 indicate that Tc1CRF1 subjects presented the highest levels of antisulphatide antibodies, being statistically different with respect to the three remaining groups (P o 0.01). Comparisons on the levels of antisulphatide antibodies from Tc CRF1, Tc CRF and Tc1CRF groups, showed no statistical differences. Presence of anti-Cz and P2 antibodies was only detected in the infected individuals (Table 3). Serum IgG antibody levels to Cz or P2 from Tc1CRF1 persons were higher than those recorded in the Tc1CRF group (Cz, P o 0.005; P2, P o 0.015). Results from multivariate analysis controlling for age and sex did not deviate from crude comparisons. Seropositive individuals presenting one CRF (To, Al or Hy) were analysed to see whether any of the risk factors was associated with increased levels of the studied antibodies. Comparisons among these three groups of seropositives revealed no significant differences as to the levels of antisulphatide, anti-Cz or anti-P2 antibodies. In the next step antibody levels were compared according to the presence of cardiac involvement. As seen in Table 4, 2006 Federation of European Microbiological Societies Published by Blackwell Publishing Ltd. All rights reserved

c

Table 4. Analysis of anti-P2, anti-Cz and antisulphatide antibodies among seropositives and seronegatives according to cardiac involvement OD units Classification Seropositives G1 G2 G3 Seronegatives G1 G2

n

Anti-P2

Anti-Cz

Sulphatide

25 32 14

0.70 0.05 0.71 0.06 0.81 0.06w

1.05 0.07 1.03 0.075 1.13 0.09

0.44 0.03 0.54 0.04 0.73 0.07z

33 13

0.37 0.02z 0.51 0.05‰

Mean SEM. w

Difference with the G1 seropositive group: P o 0.04. Difference with the G1 and G2 seropositive groups: P o 0.001. ‰ Statistically different from the G1 seronegative group: P o 0.005. z Statistically different from the G1 seropositive group: P o 0.02. For comparison purposes, the presence of cardiopathy among seronegatives was classified according to the criteria applied for the Trypanosoma cruzi-positive group. G1, symptomless, normal ECG and chest X-ray; G2, no congestive heart failure with ECG alterations; G3, congestive heart failure and pathological ECG tracings; Cz, cruzipain. z

seropositive persons falling in G3 had increased levels of anti-P2 antibodies, when compared with G1 individuals (P o 0.04). Applying the same kind of criteria for analysis of anti-Cz antibodies showed no differences in relation to the different CCC groups (Table 4). We also analysed antisulphatide antibodies according to the same criteria applied for classifying CCC, but separately depending on the presence of T. cruzi positive serology or not. In the case of seropositive individuals, levels of antisulphatide antibodies were significantly higher in G3 individuals with respect to the two other groups (P o 0.001, Table 4). Comparisons among seronegative individuals revealed that higher concentrations of antisulphatide antibodies were more prevalent in cases with heart involvement (G1 vs. G2: P o 0.005, Table 4). The patient belonging to the G3 classification yielded a value of 0.71 OD units. For a better analysis, individuals from G1 and G2 categories were compared according to the presence of T. cruzi infection, or not. As shown in Table 4, levels of antisulphatide antibodies in G1 seropositives were significantly above the values seen in the G1 seronegative individuals (P o 0.02). No differences were found when comparing G2 seropositives with G2 seronegative individuals. Analysis of the relationship between levels of circulating antibodies and cardiopathy severity groups was also adjusted for sex, age and presence CRF. In the case of antisulphatide antibodies, differences continued to be significant upon adjustment for CRF, age and sex in either seropositives or seronegatives. The same was true for anti-P2 antibodies, whereas adjusted analysis of anti-Cz antibodies remained as insignificant as in crude comparisons. FEMS Immunol Med Microbiol 48 (2006) 26–33

31


Discussion The development of CCC in humans is a two-stage process in which a susceptible person exposed to the parasite first becomes infected and after an interval of years or decades may later develop the heart abnormalities. As the development of cardiomyopathy is distant from the primary infection, there may be different cofactors for the establishment and severity of CCC. In line with this assumption, our findings indicate that presence of CRF is associated with a different profile of antibody responses and degree of cardiac involvement. Within this setting, it was found that presence of CRF and severe CCC coexisted with higher levels of anti-P2 antibodies. These antibodies were found to cause electrocardiographic alterations in noninfected hearts and may be involved in pathology by virtue of the crossreactivity between T. cruzi ribosomal proteins and the b-adrenergic receptor and the muscarinic receptor (Masuda et al., 1998; Lopez Bergami et al., 2001). Other authors also reported that anti-P2 reactivity could serve to differentiate asymptomatic from symptomatic individuals (Aznar et al., 1995). Our results provide new evidence suggesting a further increase of anti-P2 antibody levels, when CRF are also present. The levels of antisulphatide antibodies were slightly but significantly increased in seropositive individuals, showing no statistical differences according to the existence of CRF, although cases with advanced CCC had even more increased concentrations. Notably, among seronegatives, those with heart damage also displayed the highest levels of reactivity. It suggests that additional noninfectious stress capable of promoting heart damage might also drive antisulphatide antibody production, suggesting that this type of antibodies is not specific to cardiac Chagas’ disease. Without being mutually exclusive, some form of facilitated antisulphatide antibody production related to the presence of T. cruzi infection seems to exist, since G1 seropositive individuals displayed higher levels of these antibodies compared to their noninfected counterparts. Increased antibody reactivity against sulphatide during T. cruzi infection may rely not only on the molecular mimicry (Petry et al., 1988), but also on the reported release of oligodendrocyte-associated glycolipids, occurring after the destruction of autonomic ganglia by T. cruzi infection, or even the expression of parasitic galactocerebroside on nerve cell membranes (Towbin et al., 1987). In theory, antisulphatide antibodies may be potentially harmful, since injection of sera from galactocerebroside-immunized rabbits into sciatic nerves of normal rabbits was followed by demyelination (Saida et al., 1979). On the other hand, Kohriyama et al. (1988) reported an important degree of peripheral nerve dysfunction following the sensitization of rabbits with sulphatide, this phenomenon also coexisting with an increased synthesis of antisulphocerebroside antibodies. FEMS Immunol Med Microbiol 48 (2006) 26–33

We also analysed the occurrence of anti-Cz antibodies because it has been reported that Cz is a major T. cruzi antigen recognized by most sera from chronic patients (Sharfstein et al., 1986; Martinez et al., 1991). On the other hand, deposits of Cz have been localized in myocardial tissue from Chagas’ heart disease patients, suggesting a pathogenic role for this glycoprotein (Morrot et al., 1997). Although some experimental evidence shows that the immunization of BALB/c mice with enzymatically inactive Cz led to a tissue damage-accompanying autoreactive response against skeletal and heart muscles (Giordanengo et al., 2000a; Duschak et al., 2001), in this study no relationship was found between the levels of anti-Cz antibodies and degree of cardiac involvement. It is important to note that we evaluated the total reactivity to purified Cz and not to putative Cz epitopes. Recent studies suggest that Cz might contribute to human cardiac pathogenesis, since the humoral immune response against different Cz epitopes seemed to be related with the severity of heart disease (Giordanengo et al., 2000b). Beyond these facts, our study revealed that anti-Cz antibodies were increased in the presence of CRF, as were the anti-P2 immunoglobulins. In trying to interpret the present results within the context of CCC development, the following scenario may be envisaged. Parasite infection elicits an immune response which, by virtue of true molecular mimicry, the release of cryptic host antigens or appearance of neoantigens during the accompanying myocardial inflammation, may serve to initiate a potentially dangerous heart response. Nevertheless, it is possible that exposures to contributory events, adding an extra level of complexity, have to occur to damage the cardiac tissue further for the emergence of CCC. That is, reactivation of infection and/or self-sustaining noninfectious pathologic processes, i.e. the potentially heart-damage facilitating CRF reported here, that working in conjunction may exacerbate the basic pathogenic mechanisms, which seem to be implicated in tissue injury. That is, the anti-T. cruzi immune response (Reis et al., 1997), autoimmunity reactions (Kalil & Cunha-Neto, 1996; Leon & Engman, 2001; Girones & Fresno, 2003) and the associated immune-mediated inflammation (Higuchi et al., 1987; Brener & Gazzinelli, 1997; Pontes-DeCarvalho et al., 2002). Such an intricate interplay may determine a varied pattern of tissue damage-associated humoral abnormalities, as our results seem to indicate. Data from the literature indicate that the occurrence of these risk factors can be associated with cardiomyopathy in the general population (Hartz et al., 1984; Piano & Schwertz, 1994; Novo et al., 1997), although in a former study we have found no evidence of an increased occurrence of cardiac involvement in T. cruzi-infected individuals presenting these CRF (Berra et al., 1998). Nevertheless, it should be mentioned that this study was carried out in younger individuals and the analysis only included abnormal ECG tracings. 2006 Federation of European Microbiological Societies Published by Blackwell Publishing Ltd. All rights reserved

c

32

Although in cross-sectional studies predictors and outcome variables are measured at the same time, our findings may favour the view of present CRF as playing a contributory role in the development of CCC. Such an assumption is based on the fact that cardiopathy cases were rather incident than prevalent, exposures may have been present before the establishment of myocardial damage and taking into account that CCC is a nonremitting disease. Nevertheless, the possibility that such an exposure–event relationship was overrepresented among participants of the study cannot be fully disregarded.

Acknowledgements This work was partly supported by grants from Consejo Nacional de Investigaciones Cient´ıficas y Técnicas (CON´ Cientifica y ICET) and Agencia Nacional de Promocion ´ Tecnologica (ANPCYT). We wish to thank to Wiener Laboratories for the reagents employed in T. cruzi-specific serology. S.G. is Research Career Investigator from CONICET. O.B. and S.M.P. are researchers from the Consejo de Investigaciones Universidad Nacional de Rosario.

References Andrade Z (1999) Immunopathology of Chagas’ disease. Mem Inst Oswaldo Cruz 94: 71–80. Avila JL, Rojas M & Carrasco H (1993) Elevated levels of antibodies against sulphatide are present in all chronic chagasic and non-chagasic dilatatory cardiomyopathy sera. Clin Exp Immunol 92: 460–465. Aznar C, Lopez-Bergami P, Brandariz S, et al. (1995) Prevalence of anti-R-13 antibodies in human Trypanosoma cruzi infection. FEMS Immunol Med Microbiol 12: 231–238. Berra H, Carnevali F, Revelli S, Moreno H, Pezzotto SM, Morini JC & Bottasso OA (1998) Electrocardiographic alterations in chronically Trypanosoma cruzi-infected persons exposed to cardiovascular risk factors. Arch Med Res 29: 241–246. Bradford MM (1976) A rapid and sensitive method for the quantitation of microgram quantities of protein utilizing the principle of protein-dye binding. Anal Biochem 72: 248–254. Brener Z & Gazzinelli RT (1997) Immunological control of Trypanosoma cruzi infection and pathogenesis of Chagas’ disease. Int Arch Allergy Immunol 114: 103–110. ´ Cruz-Robles D, Reyes PA, Monteon-Padilla VM, Ortiz-Muñiz AR ´ G (2004) MHC class I and class II genes in & Vargas-Alarcon Mexican patients with Chagas disease. Hum Immunol 65: 60–65. Dávila H, Beloscar JS, Bottasso OA & Morini JC (1987) Alteraciones electrocardiográficas en individuos infectados con Tripanosoma cruzi con distintos tiempos de residencia en a´ reas de alta endemicidad. Medicina (Buenos Aires) 47: 154–158.


c

C. Diez et al.

De Lederkremer R, Zingales B, Confalonieri A, Couto A, Martin N & Colli W (1985) In vivo incorporation of palmitic acid and galactose in glycolipids of Trypanosoma cruzi epimastigotes. Biochem Int 10: 78–88. Duschak VG, Riarte A, Segura EL & Laucella SA (2001) Humoral immune response to cruzipain and cardiac dysfunction in chronic Chagas disease. Immunol Lett 78: 135–142. Feldman S, Garc´ıa G, Svetaz MJ, Avila JL, Revelli S, Bottasso OA & Marcipar A (1999) Evidence that antisulphatide autoantibodies from experimentally Trypanosoma cruziinfected rats bind to homologous neural tissue. Parasitol Res 85: 446–451. Giordanengo L, Fretes R, Diaz H, Cano R, Bacile A, Vottero-Cima E & Gea S (2000a) Cruzipain induces autoimmune response against skeletal muscle and tissue damage in mice. Muscle Nerve 23: 1407–1413. Giordanengo L, Maldonado C, Rivarola HW, Iosa D, Girones N, Fresno M & Gea S (2000b) Induction of antibodies reactive to cardiac myosin and development of heart alterations in cruzipain-immunized mice and their offspring. Eur J Immunol 30: 3181–3189. Girones N & Fresno M (2003) Etiology of Chagas disease myocarditis:autoimmunity, parasite persistence, or both? Trends Parasitol 19: 19–22. Guidelines Subcommittee (1999) World Health Organization – International Society of Hypertension. Guidelines for the management of hypertension. J Hypertens 17: 151–183. Hartz AJ, Anderson AJ, Brooks HL, Manley JC, Parent GT & Barboriak J (1984) The association of smoking with cardiomyopathy. N Engl J Med 311: 1201–1206. Higuchi ML, De Morais F, Pereira-Barreto AC, Lopes EA, Stolf N, Bellotti G & Pileggi F (1987) The role of active myocarditis in the development of heart failure in chronic Chagas disease:a study based on endomyocardial biopsies. Clin Cardiol 10: 665–670. Kalil J & Cunha-Neto E (1996) Autoimmunity in Chagas disease cardiomyopathy:fulfilling the criteria at last? Parasitol Today 12: 396–399. Kirchhoff LV (1993) American trypanosomiasis (Chagas’ disease). A tropical disease now in the United States. N Engl J Med 329: 639–644. Kohriyama T, Ariga T & Yu RK (1988) Preparation and characterization of antibodies against sulfated glucuronic acid-containing glycosphingolipid. J Neurochem 51: 869–877. Laemmli UK (1970) Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature 227: 680–685. Leon JS & Engman DM (2001) Autoimmunity in Chagas heart disease. Int J Parasitol 31: 555–561. Levitus G, Hontebeyre-Joskowicz M, Van Regenmortel M & Levin MJ (1991) Humoral autoimmune response to ribosomal P proteins in chronic Chagas’ heart disease. Clin Exp Immunol 85: 413–417. Lopez Bergami P, Scaglione J & Levin MJ (2001) Antibodies against the carboxylterminal end of the Trypanosoma


33


cruzi ribosomal P proteins are pathogenic. FASEB J 15: 2602–2612. Marcipar I, Olivares ML, Robles L, Dekanty A, Marcipar A & Silber A (2004) The diagnostic performance of recombinant Trypanosoma cruzi ribosomal P2b protein is influenced by its expression system. Protein Expres Purif 34: 1–7. Martinez J, Campetella O, Frasch ACC & Cazzulo JJ (1991) The major cysteine proteinase (cruzipain) from Trypanosoma cruzi is antigenic in human infections. Infect Immun 59: 4275–4277. Masuda MO, Levin M, De Oliveira SF, Dos Santos Costa PC, Bergami PL, Dos Santos Almeida NA, Pedrosa RC, Ferrari I, Hoebeke J & Campos de Carvalho AC (1998) Functionally active cardiac antibodies in chronic Chagas’ disease are specifically blocked by Trypanosoma cruzi antigens. FASEB J 12: 1551–1558. Montamat E, D’oro G, Gallerano R, Sosa R & Blanco A (1996) Characterization of Trypanosoma cruzi population by zymodemes:correlation with clinical picture. Am J Trop Med Hyg 55: 625–628. Morrot A, Strickland DK, Higuchi ML, Reis M, Pedrosa R & Scharfstein J (1997) Human T cell responses against the major cysteine proteinase (cruzipain) of Trypanosoma cruzi: role of the multifunctional alpha 2-macroglobulin receptor in antigen presentation by monocytes. Int Immunol 9: 825–834. Novo S, Barbagallo M, Abrignani MG, Nardi E, Di Maria GU, Longo B, Mistretta A & Strano A (1997) Increased prevalence of cardiac arrhythmias and transient episodes of myocardial ischemia in hypertensives with left ventricular hypertrophy but without clinical history of coronary heart disease. Am J Hypertens 10: 843–851. Petry K, Nudelman E, Eisen H & Hakomori SI (1988) Sulfated lipids represent common antigens on the surface of Trypanosoma cruzi and mammalian tissues. Mol Biochem Parasitol 30: 113–121. Piano MR & Schwertz DW (1994) Alcoholic heart disease: a review. Heart Lung 23: 3–17. Pontes-De-Carvalho L, Santana CC, Soares MB, Oliveira GG, Cunha-Neto E & Ribeiro-Dos-Santos R (2002) Experimental chronic Chagas’ disease myocarditis is an autoimmune disease


preventable by induction of immunological tolerance to myocardial antigens. J Autoimmunity 18: 131–138. Reis MM, Higuchi ML, Benvenuti LA, Aiello VD, Gutierrez PS, Bellotti G & Pileggi F (1997) An in situ quantitative immunohistochemical study of cytokines and IL-2R1in chronic human chagasic myocarditis:correlation with the presence of myocardial Trypanosoma cruzi antigens. Clin Immunol Immunopathol 83: 165–172. Risso MG, Garbarino GB, Mocetti E, Campetella O, Gonzalez Cappa SM, Buscaglia CA & Leguizamon MS (2004) Differential expression of a virulence factor, the trans-sialidase, by the main Trypanosoma cruzi phylogenetic lineages. J Infect Dis 189: 2250–2259. Roberts DD (1991) Sulfatide-binding proteins. Chem Phys Lipids 42: 173–183. Saida K, Saida K, Brown MJ & Silberberg DH (1979) In vivo demyelination induced by intraneural injection of antigalactocerebroside serum. A morphological study. Am J Pathol 95: 99–116. Sharfstein J, Schechter M, Senna M, Peralta JM, MendoncaPreviato L & Miles MA (1986) Trypanosoma cruzi:characterization and isolation of a 57/51,000 m.w. surface glycoprotein (GP57/51) expressed by epimastigotes and bloodstream trypomastigotes. J Immunol 137: 1336–1341. ´ Storino R (1994) Chagas cronico. Enfermedad de Chagas (Storino R & Miles J, eds), pp. 247–265. Doyma Argentina, Buenos Aires, Argentina. Szarfman A, Terranova VP, Rennard SI, Foidart JM, Fatima Lima M, Scheinman JI & Martin GR (1982) Antibodies to laminin in Chagas’ disease. J Exp Med 155: 1161–1171. Towbin H, Rosenfelder G, Wieslander J, Avila JL, Rojas M, Szarfman A, Esser K, Nowack H & Timpl R (1987) Circulating antibodies to mouse laminin in Chagas disease, American cutaneous leishmaniasis, and normal individuals recognize terminal galactosyl (alpha 1–3) galactose epitopes. J Exp Med 166: 419–432. World Health Organization (1984) Cardiomiopat´ıas. Serie de Informes Técnicos, 697, pp. 47–55. World Health Organization, Geneva, Switzerland.


c