INTRODUCTION

Top Abstract Introduction Materials and Methods Results Discussion References

Chagas' disease, which is caused by the protozoan parasite Trypanosoma cruzi, affects several million people in Central and South America (3). Approximately 30% of infected persons develop symptoms of the disease in their lifetime, which include cardiomyopathy, neuropathies, and dilatation of colon or esophagus (27). The finding of a T-cell-rich inflammatory mononuclear cell infiltrate and the scarcity of parasites in heart lesions questioned the direct participation of T. cruzi in chronic Chagas' cardiomyopathy (CCC) and suggested the possible involvement of autoimmunity (24), although this remains a hotly debated issue (13).

Natural infections occur via the triatomid insect vector and have been almost abolished through vector control programs. Congenital transmission and transfusion of blood from infected donors have become the major routes of transmission of Chagas' disease, and blood bank testing is now necessary in many countries. Diagnosis of T. cruzi infection often requires a combination of some of the commercially available tests (15). Traditional methods of parasite detection such as xenodiagnosis and hemocultures have low sensitivity and require long periods of time to carry out. Recently, PCR amplification of nuclear or kinetoplast DNA has been shown to be very sensitive (2, 4, 25, 31). However, PCR is not yet feasible for blood bank testing in many of the areas where Chagas' disease is endemic. At present, the best way of diagnosing an indeterminate or chronic T. cruzi infection is the serologic detection of antibodies directed against the parasite. Usually, two tests based on different methodologies are required, indirect immunofluorescence (IIF) and indirect hemagglutination (IHA), with the results confirmed by a third test, an enzyme-linked immunosorbent assay (ELISA) (15). Thus, specificity of Chagas' disease diagnosis is still a problem.

Previously, we isolated a human antigen recognized by chagasic sera, named Cha. The epitope of Cha recognized by chagasic sera was mapped to amimo acids 120 to 129 (the R3 peptide) (11). We studied whether the R3 peptide of the Cha autoantigen could be used as a marker of the disease. For this, we studied the reactivity of chagasic sera, including sera from patients at different clinical stages, against the R3 peptide in ELISA. In addition, we compared the R3 ELISA with other available tests.

	MATERIALS AND METHODS

Top Abstract Introduction Materials and Methods Results Discussion References

Synthetic peptides. Peptides R3 (MRQLDTNVERRALGEIQNV) from human Cha and S1 (STPSTPADSSAHSTPSTPV) from T. cruzi shed acute-phase antigen were synthesized on an Applied Biosystems synthesizer model 431A. Peptides were purified by high-pressure liquid chromatography and checked for accuracy by mass spectrometry.

Human sera. A total of 79 sera from patients with chronic Chagas' disease from Venezuela and Argentina were tested. Of those, 50 were from patients at different clinical stages, including chronic patients treated with antiparasite drugs (Radanil or Lampit). Sera from Argentina were obtained from the Servicio Nacional de Chagas Argentina, and sera from Venezuela were a gift from J. Sequí (Centro de Investigación Clínica, Instituto Carlos III, Madrid). Nonchagasic patients included 10 healthy individuals from an area were Chagas' disease is endemic (EHS samples) (Servicio Nacional de Chagas, Argentina), 10 individuals infected with the Leishmania parasite whose antigens cross-react with T. cruzi (kindly supplied by C. Alonso, Centro de Biología Molecular, Madrid), and 6 patients with nonchagasic cardiomyopathy with diagnosis of idiopathic dilated cardiomyopathy (IDC), a disease with similar cardiac symptoms as CCC (kindly supplied by Barbieri, Chagas Center and Regional Pathology, Santiago del Estero, Argentina).

ELISA. ELISA with total T. cruzi antigens was performed in microtiter plates covered with soluble antigens following the directions of the manufacturer (T. cruzi, Biozima-Ch, Polychaco, Argentina). The sera were diluted 1:100. The second antibody was monoclonal anti-human immunoglobulin G (IgG) labeled with horseradish peroxidase. Hydrogen peroxide-tetramethylbenzidine was used for color development, and the reaction was stopped with 2 N H₂SO₄. The developed color was measured in a microplate reader at 495 nm.

IHA. The IHA was performed essentially as described (18) using commercial kits (Polychaco), and following the instructions of the manufacturer.

IIF. T. cruzi epimastigotes were fixed with glutharaldehyde, washed, and incubated with human sera and anti-human IgG conjugated with fluorescein isothiocyanate following the method described (10).

Statistical analysis. We evaluated the accuracy of the ELISA technique against the R3 peptide in function of specificity and sensitivity. For the ELISA against the R3 peptide, a receiver operating characteristics (ROC) curve was used to choose the best cutoff. The cutoff level with the best specificity and sensitivity was an optical density at 450 nm (OD₄₅₀) of 0.122. Spearman correlation analysis, a variant of the Pearson correlation coefficient, was also used. The sign of the R coefficient indicates the sense of the association, being direct when there is a positive sign and inverse when negative. A value of 0 indicates no relationship.

	RESULTS

Top Abstract Introduction Materials and Methods Results Discussion References

R3 ELISA of chronic chagasic sera. A total of 105 serum samples were analyzed. Those included sera from 79 persons in the chronic stage of Chagas' disease. Figure 1 shows that the majority of chagasic sera, 73 of 79 (92.4%), had IgG against R3 above the cutoff level (OD₄₅₀ of 0.122), with 58 of 79 sera (73%) having ELISA OD values above the cutoff (greater than 1). The specificity of the R3 peptide was analyzed with serum samples from nonchagasic patients, including EHS, and patients with other parasitic diseases such as leishmaniasis or unrelated cardiomyopathy such as IDC. All nonchagasic sera showed OD values below the cutoff level (Fig. 1). The statistical analysis showed that the assay was 92.4% sensitive and 100% specific for chagasic sera, while it showed no sensitivity for the nonchagasic controls (leishmaniasis, IDC, and EHS sera) (Table 1).

View larger version (18K): [in this window] [in a new window]   FIG. 1.   Reactivity of sera from chronic chagasic patients against the R3 peptide. The reactivity of 79 sera with positive serology for Chagas' was tested in ELISA with the R3 peptide. Ten sera from individuals infected with Leishmania infantum, 6 sera from patients suffering (IDC, and samples from 10 healthy individuals living in an area were Chagas' disease is endemic (EHS) were used as controls. Reactivity is expressed as OD at 450 nm. The horizontal dashed line represents the cutoff value.

Correlation of anti-R3 IgG antibody levels with Chagas' disease progression. Next, we analyzed the titers of anti-R3 antibodies in sera from patients in different stages of clinical disease. The ELISA reactivity against R3 peptide was compared in serum samples from patients in different stages of Chagas' disease (11 patients with and 33 patients without symptoms) and 10 serum samples of EHS. Although it was not statistically significant, we observed an increase in the mean anti-R3 titer (from 0.8 to 1.1) with progression of disease, since antibody titer was higher in symptomatic than in asymptomatic chagasic patients (Fig. 2A). We included in the analysis an ELISA based on the S1 peptide, which corresponds to the C-terminal repeats of the shed acute-phase antigen of T. cruzi, a highly immunogenic T. cruzi antigen. The reasons to include it were twofold: shed acute-phase antigen has been considered one of the most immunogenic antigens and the best one for the acute phase, and to compare the R3 peptide ELISA with another peptide-based ELISA (not based on parasite extracts) against which IgG prevalence in samples from chronic patients has been reported (23). Noteworthy, the titers of anti-S1 antibodies were much lower than anti-R3 titers in all patients.

View larger version (9K): [in this window] [in a new window]   FIG. 2.   Correlation of anti-R3 IgG antibodies with symptomatology and treatment. (A) Correlation of IgG antibody levels against R3 and S1 peptides with Chagas' disease progression. Sera were grouped as control (EHS), asymptomatic, and symptomatic and plotted against the mean OD490 ± standard deviation (SD) values obtained with the R3 (solid circles) and S1 (solid triangles) peptide ELISA. The number of individual sera tested in each group is indicated. Sera were assayed at 1:100 dilution. (B) Correlation of IgG antibody levels against R3 and S1 peptides with the treatment of asymptomatic chagasic patients. The results obtained with sera from untreated and treated chagasic patients were plotted against the mean OD490 ± SD values obtained with the R3 (solid circles) and S1 (solid triangles) peptide ELISA. The number of individuals in each group is indicated.

Relationship of anti-R3 IgG antibody levels to treatment of chagasic patients. The titers of antibodies against the R3 peptide and S1 peptide were also tested in a group of 19 asymptomatic chagasic patients treated with nifurtimox (Lampit) or benznidazol (Radanil) and compared to 30 sera from untreated patients of the same asymptomatic group. A statistically significant decrease in the mean reactivity against R3 was observed in patients treated with either Lampit or benznidazol (P < 0.01). The analysis of the S1 reactivity, although lower in magnitude, also showed a statistically significant decrease in the OD values after treatment (P < 0.01), (Fig. 2B).

Comparison of R3 peptide ELISA with other serological tests. The results obtained with ELISA of the R3 or S1 peptides were compared with the results of other commercially available diagnostic tests, including IFA, IHA, and ELISA with crude parasite extract, and statistical analyses were performed using the Spearman test. Correlation between the R3 ELISA and IIF (Fig. 3A) and IHA (Fig. 3B) titers as well as with conventional ELISA (Fig. 3C) showed indexes of 0.61, 0.71, and 0.64, respectively (P < 0,001). Although the S1 ELISA also showed a direct correlation with IIF, IHA, and ELISA, the indexes were smaller than with the R3 ELISA (0.39, 0.5, and 0.35, respectively, with P < 0.01 to 0.001).

View larger version (12K): [in this window] [in a new window]   FIG. 3.   Correlation of IgG antibodies against R3 and S1 peptides by ELISA with other serological tests (IIF, IHA, and parasite ELISA). Individual sera were grouped based on the results (titer inversa) of each serological test and plotted against the mean of OD490 ± SD values obtained with the R3 (solid circles) and S1 (solid triangles) peptide ELISA. The number of serum samples in each group is indicated. (A) Correlation with IIF. A value of 0 is negative and indicates titers <1:32. (B) Correlation with IHA. A value of 0 is negative (titer < 1:16). (C) Correlation with a commercial ELISA using crude T. cruzi extract at 1:100 serum dilution.

	DISCUSSION

Top Abstract Introduction Materials and Methods Results Discussion References

Despite programs of insect control, diagnosis of Chagas' diseases in blood bank screening is necessary nowadays in order to avoid spreading the infection. Currently, the detection of positive samples for T. cruzi infection involves a combination of different tests. A recurrent problem is the presence of false-positives due to cross-reactions with sera from patients with other infectious diseases, such as leishmaniasis.

Here, we have tested the potential of the R3 peptide of Cha, a recently described autoantigen (11), as a marker of Chagas' disease. The presence of specific autoantibodies is used as marker of the disease and of disease progression in several pathologies of autoimmune etiology (7, 8, 29). Here, we report the first description of the usefulness of such a test for the diagnosis of Chagas' disease. We performed a statistical analysis of the reactivity against the R3 peptide of Cha and found that it is a good marker of the disease. The R3 peptide ELISA showed 92.4% sensitivity and 100% specificity for chronic chagasic sera in comparison with sera from individuals infected with Leishmania and patients suffering IDC and EHS, sera which were below the cutoff value. Although many autoantigens have been described during T. cruzi infection, the percentage of recognition among chagasic sera was lower (6, 9) than for the R3 peptide. The R3 peptide was not recognized by sera from IDC patients, a disease clinically similar to CCC. Very interestingly, this result is contrary to results obtained with other autoantigens which are recognized by both chagasic and IDC sera (5) and indicates the high specificity of the R3 peptide as a marker of Chagas' disease.

Currently employed serological tests for Chagas' disease include some commercially available enzyme immunoassays, based on T. cruzi extracts, as well as IIF and IHA assays. These assays are still widely used but often lead to false-positives or -negatives due to subjective interpretation. However, those tests gave between 7.5% and 17.5% positive results for sera from Leishmania-infected patients as well as false-positives and -negatives (15). Many enzyme immunoassays using either crude antigen or recombinant T. cruzi proteins have been recently described with the aim to circumvent this problem (12, 14, 16, 17, 19, 20, 22, 28). Those tests have specificities ranging from 93 to 100% and sensitivities of 97 to 100%, although few of these are really employed in the field. The test described here has similar sensitivity and specificity. Moreover, statistically significant correlations of recognition of the R3 by chagasic sera with commercially available tests (IIF, IHA, and ELISA based on the parasite) were found. Our results indicated that all sera positive for T. cruzi infection, usually diagnosed by IHA and IIF, are recognized by the anti-R3 test. Anti-R3 antibody titer tended to increase with disease progression (mean OD increased from 0.8 to 1.1) although the differences were not statistically significant because of the spreading of the OD values (from approximately 0.2 to 1.5). We observed an inverse correlation between nifurtimox/Lampit treatment and anti-R3 antibody titers in asymptomatic patients. This inverse correlation was observed as well with IIF, IHA, and ELISA based on the parasite antigen. Thus, the IgG titers against R3 of positive sera showed a statistically significant decrease in treated patients.

The R3 peptide ELISA showed a stronger titer and correlation with other tests than an ELISA against the S1 peptide, which contains the highly immunogenic C-terminal repeats of shed acute-phase antigen of T. cruzi and is considered a major antigen (23). This result indicated that a stronger reactivity against R3 of Cha than against S1 of shed acute-phase antigen develops during infection. On the other hand, T. cruzi is a polymorphic parasite, and different strains may circulate in different areas. Our results gave similar results when the sera were divided according to their origin (Venezuela and Argentina) (not shown). Another problem that some serologic tests based on T. cruzi antigens present is derived from the cross-reactivity of anti-T. cruzi antibodies with other endemic parasites such as Leishmania (1) and Trypanosoma rangeli (12, 21, 26, 30). R3 is not recognized by sera from Leishmania-infected patients. Although we have not specifically tested reactivity of R3 with sera from T. rangeli-infected individuals, we think this is unlikely by the nature of the test, since these patients have no autoantibodies.

In summary, our results indicate that the detection of antibodies against R3 peptide could be used as a confirmation marker of Chagas' disease. Due to the direct correlation with symptoms and inverse correlation with treatment, it can be used to help monitor the clinical status of patients.