Introduction

Chagas disease is an anthropozoonotic disease caused by the hemoflagellate protozoan Trypanosoma cruzi. It is a neglected tropical disease that affects 6–8 million people worldwide and is responsible for the death of approximately 50,000 individuals per year [1, 2]. Trypanosoma cruzi infection occurs when the infectious forms of the parasite (i.e., metacyclic trypomastigotes), after adhesion and penetration, invade different cell types (e.g., macrophages, fibroblasts, and epithelial cells), differentiate (blood trypomastigotes), and invade cells of other organs [3,4,5], such as myocardiocytes, and smooth muscle cells of the esophagus and colon [6].

The acute phase of the disease is characterized by mild and unspecific symptoms (or even no symptoms), high parasitemia, parasitism, intense inflammatory responses, and particularly rapid (6–8 weeks) evolution. In the chronic phase (lifelong), parasitemia becomes intermittent and a specific anti-T. cruzi immune response is elicited [7, 8]; approximately 70% of the infected individuals present the indeterminate or asymptomatic form of the disease, whereas 30% manifest the main clinical forms after decades, namely, digestive (megaesophagus and megacolon), cardiac, and cardiodigestive [1, 6, 9]. In chronic chagasic cardiomyopathy, severe and diffuse myocarditis favors the death of myocardiocytes and their replacement by collagen fibers. This alters the transmission of nerve impulses and results in hypertrophy of the cardiac chambers and failure of organs [10,11,12].

The control of T. cruzi infection is dependent on both innate and adaptive immune responses, which are mediated by macrophages, natural killer cells, T CD4+, T CD8+ lymphocytes, B lymphocytes, and the complement cascade [13, 14].

The complement cascade consists of more than 40 circulating plasma proteins, whose functions are to opsonize and eliminate pathogens from the bloodstream, recruit phagocytic cells to infection sites [15, 16], activate B lymphocytes, and stimulate immunoglobulin (Ig) synthesis [15, 17]. In Chagas disease, the classical complement pathway is activated when anti-T. cruzi-IgM and -IgG opsonize the blood trypomastigote forms, enabling the binding of the C1q protein and activation of the C1r and C1s serine proteases responsible for the cleavage of the C4 and C2 proteins, thereby forming C3 convertase [15, 17].

Human isotypes IgG1 and IgG3 are effective activators of the classical complement pathway [18], whereas some studies indicate that the IgG2 and IgG4 isotypes are weak activators or that they are not activators [19, 20]. Although the IgG3 antibody better binds to C1q, IgG1 is more effective in activating complement pathway-dependent cell lysis [21]. This functional difference may affect the quality of Ig activity in the immune response to T. cruzi infection. Experimental studies have demonstrated an important role of IgG1 and IgG2b in the immune response to T. cruzi [22, 23], with IgG2a being the main factor responsible for the clearance of blood trypomastigotes [24].

Indeed, the trypanocidal activity of benznidazole reduces parasitemia and parasitism in the chronic and acute through induce proinflammatory cytokines and interleukin-10 which contribute to increased IgG subclass synthesis, indicating a double effect in the control of parasite replication and tissue damage [25, 26]. However, the repercussions of this medication in individuals with the indeterminate form of Chagas disease as well as in the prevention and progression of the cardiac and digestive forms have to be elucidated [27, 28].

Therefore, the objective of this study was to evaluate the levels of total IgG and cross-specific IgG subclasses in patients with chronic Chagas disease of different clinical forms before and after 4 years of treatment with benznidazole.

Materials and Methods

Patient Population

This was a prospective study carried out from September 2007 to December 2013. All individuals were from Uberaba and the region and were monitored and followed-up by the staff of Chagas Medical Clinic, Federal University of Triângulo Mineiro, Minas Gerais, Brazil. The inclusion criteria for patients were as follows: positive serology for T. cruzi in the chronic phase, aged between 18 and 65 years, signed the informed consent form, blood culture and positive polymerase chain reaction (PCR) for T. cruzi at the time of intervention [29] Between September 2007 and December 2008, 1254 cases were evaluated, 39 individuals met the inclusion criteria, but only 32 individuals agreed to undergo treatment with benznidazole for 60 days after assessing the risks and benefits. However, three individuals did not complete the treatment (one due to desistance and two due to severe allergic reactions) (See supplementary materials), six did not agree to participate in the final evaluation, eight patients were diagnosed with gastrointestinal forms, three individuals with the indeterminate form developed the cardiac form, and two individuals with the cardiac form presented worsening of the clinical status. Thus, 17 patients who completed the treatment protocol were evaluated: (1) 8 individuals with the indeterminate form: 4 males and 4 females and (2) 9 individuals with the cardiac form: 5 males and 4 females. Disease progression was determined according to the Second Brazilian Consensus on Chagas Disease [29]. The control group formed by 21 healthy individuals (12 males and 9 females) was evaluated to validate this method, to distinguish previous infection with T. cruzi and cross-reactivity (Table 1).

Table 1 Demographic and clinical data
Full size table

Experimental Protocol

Clinical history recording and physical examination were performed according to the propaedeutic rules. In the patient selection and group formation phase, the following tests were performed: cardiac autonomic function tests and tests to determine clinical forms (12-lead electrocardiogram [ECG], echocardiogram, chest radiography, esophagogram, and opaque enema), serology (indirect immunofluorescence [30], indirect hemagglutination [31], and fluorescent immunosorbent assay [32]), blood culture [33, 34], and PCR [35, 36] for the diagnosis of T. cruzi infection (with serological positivity confirmed in at least two of the three techniques mentioned), and biochemical tests [37], performed according to clinical and laboratorial routine of Chagas disease outpatient clinic at Clinical Hospital of Federal University of Triângulo Mineiro and using clinically validated commercial kits (See supplementary materials). After the biochemical evaluation and according to the criteria of the Second Brazilian Consensus on Chagas Disease [29], individuals were classified based on the presence of other comorbidities (diabetes mellitus [38], dyslipidemia [39, 40], thyroid disorders [41,42,43,44], renal failure [45], and liver function [46]) (See supplementary materials). All subjects were followed-up during and after treatment. All of them presented negative T. cruzi blood cultures after 60 days of treatment and at 48 months after the follow-up.

Preparation of Soluble Antigens of T. cruzi for Anti-T.cruzi FLISA Assay

Epimastigote forms of T. cruzi Y strain were grown in Schneider medium (Sigma-Aldrich, St. Louis, MO, USA) supplemented with 20% inactivated fetal bovine serum. At the end of the logarithmic phase, the parasites were collected, washed three times with sterile phosphate-buffered saline (PBS: 80.0 g NaCl, 11.6 g Na2HPO4, 2.0 g KH2PO4, 2.0 g KCL, q.s. to 10 L pH to 7.0; centrifugation: 3000 × g, 30 min, 4 °C), counted in a Neubauer chamber (Kasvi, São José dos Pinhais, PR, Brazil), and resuspended in 1 mL of TLCK lysis buffer (Tris–HCl pH 7.2, 1% Nonidet P-40, and 0.025% sodium azide) at a concentration of 108 live epimastigotes/mL, and placed on ice under agitation for 10 min. The suspension was centrifuged at 10,000 × g for 30 min at 4 °C, and the supernatant (soluble protein fraction) was collected and passed through a 0.22 µm pore-size filter (Millipore, Molsheim, France). Protein was quantified using the MicroLowry method (Pierce, Rockford, IL, USA), and the protein fraction was adjusted to a concentration of 4.8 mg/mL in sterile PBS and stored in aliquots at −70 °C.

Blood Collection and Ig Level Determination by Fluorescent-Linked Immunosorbent Assay

96-well microplates were sensitized with the T. cruzi crude antigen, obtained as described above. On day 0 and after 48 months of treatment with benznidazole, 30 mL of venous blood was collected via peripheral puncture into tubes without additives and centrifuged at 700 × g for 10 min at 25 °C, and the plasma was stored in aliquots at − 70 °C. Ninety-six-well microtiter plates were sensitized with 100 μL/well T. cruzi antigens, diluted to 5 μg/mL with carbonate-bicarbonate buffer (pH 9.5), and incubated for 18 h at 4 °C. Thereafter, the plates were washed with PBS + 0.05% Tween 20 (LCG Biotecnologia, Cotia, SP, Brazil) solution and blocked with 200 μL of PBS + 10% inactivated fetal bovine serum for 24 h at 4 °C. Subsequently, the serum were diluted in PBS + 10% inactivated fetal bovine serum (Sigma-Aldrich) at a concentration of 1:50 for IgG1, IgG2, IgG3, and IgG4 and 1:250 for total IgG, and the plates were incubated for 4 h at room temperature (each IgG subclasses analysis was carried out in independent plate for each one). After incubation, the plates were washed with PBS + 0.05% Tween 20 (LCG Biotecnologia) solution and incubated for 2 h. To measure the level of total IgG, monoclonal anti-IgG antibody conjugated with fluorescein isothiocyanate (FITC) (BD Biosciences, San Jose, CA, USA) was diluted 1:1000 in PBS + 10% inactivated fetal bovine serum. To measure the level of IgG isotypes, monoclonal anti-IgG1 and anti-IgG4 conjugated with FITC and anti-IgG2 and anti-IgG3 conjugated with phycoerythrin (PE) (BD Biosciences) were added, diluted 1:100 in PBS + 1% bovine serum albumin (BSA), and incubated for 2 h at room temperature. Subsequently, the plates were washed, and the sample was resuspended in 100 µL of PBS/well. The Igs were measured using a fluorescence reader for microplates (Enspire; PerkinElmer, Waltham, MA, USA). The excitation and detection wavelengths for FITC-labeled antibodies were 488 and 512 nm, respectively, and those for PE-labeled antibodies were 488 and 570 nm, respectively. The results are expressed as relative fluorescence units from the detection wavelength.

Statistical Analyses

Data were analyzed in GraphPad Prism 7.0 (GraphPad Software, La Jolla, CA, USA). Normality was checked using the Kolmogorov–Smirnov test and a paired t test was used to compare the before and after treatment; Student t test was used to compare different groups at the same time point: first, the serum levels of total IgG and IgG subclasses were analyzed in patients with the indeterminate and cardiac forms on day 0 and at 48 months after treatment. Second, the serum levels of total IgG and IgG subclasses were analyzed in individuals with Chagas disease according to the presence (Yes) or absence (No) of worsening of the clinical form and/or alteration of the clinical form of the disease before and after 48 months of benznidazole treatment. After assessing the serum levels of total IgG and anti-T. cruzi-IgG subclasses, we verified the fraction occupied by each subclass in relation to the total IgG level using the IgG subclass/total IgG ratio. Finally, the ratio of IgG subclasses/total IgG was analyzed in patients with Chagas disease according to the presence (Yes) or absence (No) of worsening of the clinical symptoms and/or alteration of the clinical form of the disease before and after 48 months of treatment. The results were expressed as mean ± standard deviation and considered statistically significant at P < 0.05.

Results

Elevation in Anti-T. cruzi-IgG1 and Reduction in Anti-T. cruzi-IgG3 and -IgG4 in Patients with the Cardiac form of Chagas Disease 48 Months After Treatment with Benznidazole

As expected, individuals without Chagas disease showed extremely low levels of anti-T. cruzi-IgG (total and subclasses < 250) compared with patients with Chagas disease (total IgG > 2000; IgG1 > 1000; IgG2 > 400; IgG3 > 1000). Only the IgG4 level in the chagasic group (IgG4 < 200) was lower than that in the control group. As a result, differences between the chagasic and control groups were suppressed in the analysis performed here.

The anti-T. cruzi total IgG level in patients with the indeterminate form (P = 0.012) and cardiac form (P = 0.007) significantly reduced at 48 months after treatment. A comparison of the IgG level between the clinical forms before (P = 0.45) and after treatment (P > 0.99) revealed no significant difference (Fig. 1a). Similar levels of anti-T. cruzi-total IgG were recorded before and after treatment. Patients with the indeterminate and cardiac forms presented significantly different anti-T. cruzi-IgG1 levels; patients with the indeterminate form presented a higher IgG1 level (P = 0.009). After treatment, both groups presented similar anti-T. cruzi-IgG1 levels (P = 0.33), although patients with the indeterminate form showed a reduction in the IgG1 level (P = 0.002) and those with the cardiac form showed an increase in the IgG1 level (P = 0.002) at 48 months (Fig. 1b). There was no difference in the anti-T. cruzi-IgG2 level before and after treatment (Fig. 1c).

Fig. 1
figure 1

Levels of serum anti-T.cruzi immunoglobulins among chagasic individuals with indeterminate and cardiac form. FLISA fluorescence intensity for serum total IgG (a), IgG1 (b), IgG2 (c), IgG3 (d) and IgG4 (e) anti-T.cruzi antibodies before and after treatment. The symbols represent mean and the vertical lines the standard deviation. *p < 0.05, paired t test or Student t test. Black *—comparison between Indeterminate and Cardiac; Red and blue *—comparison between before and after treatment

Full size image

Before treatment, patients with the indeterminate and cardiac forms presented similar anti-T. cruzi-IgG3 level (P = 0.45). However, at 48 months after benznidazole treatment, patients with the cardiac form (P = 0.07) showed a significant reduction in the IgG3 level, whereas patients with the indeterminate form (P = 0.03) showed a significant increase (Fig. 1d).

Patients with the cardiac form showed a significant reduction in the anti-T. cruzi-IgG4 level (P = 0.007) after treatment, whereas those with the indeterminate form (P = 0.11) showed no changes in the periods evaluated (P = 0.11), despite an increase in the mean level of anti-T. cruzi-IgG4. Before treatment, patients with the cardiac and indeterminate forms presented significant differences in the anti-T. cruzi-IgG4 level (P = 0.001; Fig. 1e); however, at both time points, patients with Chagas disease presented lower anti-T. cruzi-IgG4 level than healthy individuals without Chagas disease.

Anti-T. cruzi-IgG1 and -IgG3 Levels are Affected by Benznidazole Treatment in Individuals with Worsening and/or Alteration of the Clinical Form

The total anti-T. cruzi-IgG level showed a significant reduction independent of clinical worsening (“No” group, P = 0.03; “Yes” group, P = 0.002). A comparison before (P = 0.62) and after (P = 0.69) treatment between the groups revealed no significant difference in the IgG levels (Fig. 2a). The anti-T. cruzi IgG1 level was significantly higher and the anti-T. cruzi-IgG3 level was significantly lower in patients presenting clinical forms and/or alteration of the clinical form at 48 months after benznidazole treatment (group Yes, P = 0.3 for IgG1 and P = 0.03 for IgG3; Fig. 2b, d). There was no significant difference in the IgG2 and IgG4 levels between the groups (Fig. 2c, e).

Fig. 2
figure 2

Levels of serum anti-T.cruzi immunoglobulins among chagasic individuals with worsening and/or alteration of the clinical form. FLISA fluorescence intensity for serum total IgG (a), IgG1 (b), IgG2 (c), IgG3 (d) and IgG4 (e) anti-T.cruzi antibodies before and after treatment. The symbols represent mean and the vertical lines the standard deviation. *p < 0.05, paired t test or Student t test. Black *—comparison between Indeterminate and Cardiac; Red and blue *—comparison between before and after treatment

Full size image

Total anti-T. cruzi-IgG4/-IgG ratio increased in patients with the indeterminate forms, whereas the IgG3/IgG1 ratio decreased in patients with the cardiac form at 48 months after the treatment with benznidazole

Before treatment, patients with the indeterminate form presented a higher IgG1/total IgG ratio than those with the cardiac form (P = 0.003), and this difference was not observed at 48 months after treatment (P = 0.94). After treatment, the IgG1/total IgG ratio in patients with the cardiac form and the IgG3/total IgG ratio in patients with the indeterminate form were significantly higher than those before treatment (cardiac form, P = 0.033; indeterminate form, P = 0.034). In addition, patients with the indeterminate form presented a lower IgG4/total IgG ratio before treatment than patients with the cardiac form (P = 0.003), and this ratio significantly increased after 48 months (P = 0.002; Fig. 3a).

Fig. 3
figure 3

Ratio between serum levels of subclasses of IgG and total IgG between chagasic individuals with indeterminate and cardiac form or with worsening and / or alteration of the clinical form. Ratio between FLISA fluorescence intensity for serum IgG1 / total IgG, total IgG2 / IgG, total IgG3 / IgG and total IgG4 / IgG (a and c). Ratio between the levels of subclasses of IgG3 and IgG1 serum anti-T.cruzi (b and d). The symbols represent mean and the vertical lines the standard deviation. *p < 0.05, paired t test or Student t test. Black *—comparison between Indeterminate and Cardiac; Red and blue *—comparison between before and after treatment

Full size image

Based on the dynamics of IgG1 and IgG3 between patients with the cardiac and indeterminate forms and changes in the pattern of these Igs after treatment, we proceeded to calculate the ratio between IgG3 and IgG1, where values > 1 represented an increase for IgG3 and values < 1 represented an increase for IgG1. Before treatment, patients with the cardiac and indeterminate forms presented a significant difference in IgG3/IgG1 (P = 0.012). After treatment, the ratio of IgG3/IgG1 in patients with the cardiac form (P = 0.007) showed a significant reduction, whereas those with the indeterminate form (P = 0.037 showed a significant increase in this ratio (Fig. 3b).

Ratio of IgG1/total IgG iincreased in individuals with an alteration in the cclinical form of the disease, whereas the ratio of IgG3/total IgG decreased in the group with worsening of clinical symptoms and/or alteration in the clinical form of the disease at 48 months after treatment with benznidazole

The ratios of IgG1/total IgG, IgG2/total IgG, and IgG3/total IgG showed no significant difference between these time points (Fig. 3c). The IgG4/total IgG ratio significantly increased after benznidazole treatment in patients who did not present clinical worsening (P = 0.034). In addition to the differential pattern of the IgG3/IgG1 ratio between patients with the same clinical form after treatment and those who presented clinical worsening, no significant difference was observed (Fig. 3d).

Discussion

The involvement of IgG has been demonstrated in experimental animal models [24, 47,48,49,50,51,52,53] and human [54,55,56,57] with Chagas disease. Infection with T. cruzi triggers the activation of polyclonal B cells at the beginning of the infection and persists during the chronic phase of the disease [16, 53, 58]. Two weeks after infection, hypergammaglobulinemia was observed, with the IgGl, IgG2a, and IgG2b isotypes prevalent in the specific humoral immune response [47, 59]. IgG2a and IgG2b favor the elimination of trypomastigote forms by polymorphonuclear phagocytic cells and macrophages, which express receptors for the Fc portion of IgG [53, 60, 61]. However, there was a higher IgG1 avidity [62] during infection.

In murine models, IgG2a is induced by the secretion of IFN-γ (Th1), whereas IgG1 is induced by the secretion of IL-4 (Th2); these Igs with TGF-β induce IgG2b [63,64,65]. In humans, different cytokines alter the secretion of antibodies, with overlapping effects. For example, IL-4 and IL-13 are involved in the production of IgG1 and IgG4, IL-10 is involved in the synthesis of IgG1 and IgG3, and IL-21 and IL-27 are involved in the induction of IgG1 and IgG3 [66,67,68,69,70,71]. Proinflammatory cytokines (e.g., IL-2, IL-6, and IFN-γ) mainly increase the production of IgG subclasses [26, 72, 73].

Studies indicate that in individuals with different clinical forms of Chagas disease, there may be a predominance of different IgG subclasses, with an increase in the levels of IgG1 and IgG3 [18, 74]. In this study, treatment with benznidazole favored the reduction in IgG in all groups, and it was significantly associated with worsening of the clinical symptoms. Individuals with the cardiac form presented an increased level of IgG1 and a decreased level of IgG3 and IgG4, acting in an antagonistic manner to those in the indeterminate form. In addition, worsening of the clinical symptoms favored an increase in the IgG1 level and a reduction in the IgG3 level.

A study in individuals treated or not treated with benznidazole for 13 years highlighted a reduction in the total anti-T. cruzi-IgG level after treatment. However, in the study, only the clinical form diagnosed before treatment was considered, without considering any relationship between the level of total IgG and the change in the clinical form over the years [75]. Similarly, children with indeterminate form evaluated for 4 years after treatment with benznidazole showed a significant reduction in the level of anti-T. cruzi-IgG [76, 77].

Studies have reported that patients with chronic indeterminate form of Chagas disease or with different degrees of chagasic cardiomyopathy presented a prevalence of IgG1 and IgG3, low levels of IgG2, and very low levels of IgG4 [18, 74]. This demonstrated that the targets of these antibodies vary among different clinical forms and that IgG subclasses are strongly associated with the anti-T. cruzi response [59, 78, 79]. These findings corroborate the data obtained in this study.

The relationship between the intensity of serological reactivity of different IgG subclasses and the morbidity of Chagas disease has been extensively discussed. Morgan et al. (1996) and Verçosa et al. (2007) reported increased levels of IgG2 in individuals with the cardiac form, in addition to increased IgG1 and IgG2 levels in patients with Chagas disease regardless of the clinical form [80, 81], whereas analyses using different antigenic preparations revealed the predominance of IgG1 and IgG3 [82]. However, similar levels of IgG1, IgG2, IgG3, and IgG4 were observed in individuals with the indeterminate and cardiac forms [18, 74].

Benznidazole treatment did not change the ratio of IgG2/total IgG in the individuals evaluated in the present study. However, in patients with the cardiac form of the disease, an increase in the total IgG1/IgG ratio was observed after treatment, and it was similar to the ratio in patients with the indeterminate form. In addition, patients with the indeterminate form presented an increase in the IgG3/total IgG ratio, with no variation in patients with the cardiac form. It should be noted that the same pattern was observed in the IgG3/total IgG ratio in individuals who maintained their initial clinical form. Therefore, the increase in the IgG3/total IgG ratio can be associated with less cardiac damage.

IgG1 and IgG3 seem to be the main Igs associated with anti-T. cruzi response. Other studies have highlighted the importance of IgG3 in individuals with Chagas disease, without highlighting the difference among the clinical forms of the disease [83, 84]. Although further studies are needed, our results suggest that monitoring IgG3 levels may indicate a possible prognosis to the cardiac form or worsening of the already established clinical form.

Conclusions

The results of this study show that patients with chronic Chagas disease after treatment with benznidazole presented reduced levels of total IgG. Increased levels of IgG1 and decreased levels of IgG3 were observed in patients with the cardiac form of Chagas disease; the worsening of the clinical forms favored an increase in the IgG1 level and a reduction in the IgG3 level relative the level of total IgG. The reduction in the IgG3/IgG1 was strongly associated with the cardiac form. Therefore, these findings seem to demonstrate that the monitoring of IgG1 and IgG3 levels in individuals with the cardiac form of Chagas disease can indicate disease progression and may be a useful biomarker of poor prognosis. However, it is recognized that the relatively low number of chagasic individuals included here reinforces that more studies are necessary to better elucidate their immunopathological significance in Chagas disease.